United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
EPA-600/2-80-027
May 1980
Research and Development
Attenuation of
Water-Soluble
Polychlorinated
Biphenyls by Earth
Materials
-------
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 PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-80-027
May 1980
ATTENUATION OF WATER-SOLUBLE POLYCHLORINATED
BIPHENYLS BY EARTH MATERIALS
by
R. A. Griffin
Illinois State Geological Survey
University of Illinois
Urbana, Illinois 61801
and
E. S. K. Chian
School of Civil Engineering
Georgia Institute of Technology
Atlanta, Georgia 30332
Grant No. R-804684-01
Project Officer
Richard A. Carnes
Solid and Hazardous Naste Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
U S. Environmental Protection Agenc*
Region 5, Library (PL42J)
77 West Jackson Boulevard, 121& WW
Chicago, II 60604-3590
-------
DISCLAIMER
THIS report has been reviewed by the Municipal Environmental Research
atory, U.S. Environmental Protection Agency, and approved for publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
11
-------
FOREWORD
The Environmental Protection Agency was created because of increasing
concern by the public and government about the dangers of pollution to the
health and welfare of the American people. Noxious air, foul water, and
spoiled land are tragic testimony to the deterioration of our natural
environment. The complexity of that environment and the interplay between
its components require a concentrated and integrated attack on the problem.
Research and development— that necessary first step in paving a
lem— involves defining the problem, measuring its impact, and starphtng for
solutions. The Municipal Environmental Research Laboratory develops new
and improved technology and systems to preserve and treat public drinHing
water supplies, to minimize the adverse economic, social, health, and
aesthetic effects of polution and to prevent, treat, and manap waste water
and solid and hazardous waste pollutant discharges from municipal and) com^
munity sources. This publication is one of the products of tfiat research,
it is a vital communications link between the researcher and th§ community.
Results are reported from laboratory investigation of the capacity of
earth materials to attenuate polychlorinated biphenyls (RGBs) from waters.
These results are applicable to the design of treatment systems, earth
material liners for sanitary landfills, and to the land disposal of
pal and hazardous wastes.
Francis T. Mayo, Director
Municipal Environmental Research
Laboratory
m
-------
ABSTRACT
The aqueous solubility, adsorption, mobility, microbial degradation, and
volatility of polychlorinated biphenyls (PCBs) were studied under laboratory
conditions. The dissolution of Aroclor 1242 in water required five months to
reach equilibrium. Generally, the water-soluble fractions of the PCB fluids
were richer in the lower chlorinated isomers than in the original mixture of
isomers in the fluid. The solubilities of Aroclor 1016, 1221, 1242, and 1254
were 906 ppb, 3516 ppb, 703 ppb, and ^70 ppb, respectively.
A simple linear relation described the adsorption of water-soluble PCBs
by five earth materials and their low-temperature ashes. An adsorption con-
stant (K) unique to each adsorbent was obtained. The adsorption was strongly
correlated to the total organic carbon (TOC) content and surface area of the
earth materials. TOC was the dominant of these two earth material properties
by a ratio greater than three to one.
The mobilities of several PCBs in silica-gel, silica sand, seven soils,
and a coal char were measured with several leaching solutions using the soil
thin-layer chromatography technique. The PCBs remained immobile when leached
with water or landfill leachate, but were intensely mobile when leached with
organic solvents. The mobilities were strongly related to the solubilities of
the PCBs in the leaching solvent. The mobilities of PCBs in soils leached with
carbon tetrachloride were highly correlated to the soil TOC content.
The degradation of water-soluble Aroclor 1242 by mixed cultures of soil
microorganisms occurred in a short period of time. The lower chlorinated iso-
mers were degraded more easily than the higher chlorinated isomers. The rates
of degradation ranged from the monochloro isomers, which degraded 100 percent
within 6 hours, to the tetrachloro isomers, which averaged 42 percent degrada-
tion after 15 days. The predominant organisms found in the mixed cultures were
Alkaligenes odoranSj Alkaligenes deni-trificans3 and an unidentified bacterium.
The volatility of PCBs from water was studied in the presence of soluble
humic acid and suspended soil. Adsorption of PCBs by the soil and humic acid
reduced the total amounts of PCBs volatilized. The volatilization of PCBs
from pure water agreed with theoretical predictions that the half-life of
Aroclor 1242 stripped from water would be about 6 hours. Higher chlorinated
PCB isomers were found to be less soluble in water, preferentially adsorbed
by soil materials, less mobile in soil, less degradable by microorganisms, and
less volatile from water than lower chlorinated isomers. Thus, higher chlori-
nated isomers would be less mobile and more persistent in the environment than
lower chlorinated isomers.
This report was submitted in fulfillment of Grant No. R-804684-01 by the
Illinois State Geological Survey and the Civil Engineering Department, Univer-
sity of Illinois, under partial sponsorship of the U.S. Environmental Protec-
tion Agency. Work was completed in October, 1978.
iv
-------
CONTENTS
Disclaimer .11
Foreword ni
Abstract iv
Figures V1
Tables vii
Acknowledgments viii
1. Introduction 1
Chemical structure 1
Manufacture and industrial uses 1
Environmental effects 3
Disposal and soil attenuation 4
Microbial degradation 5
Analytical methodology 8
Concluding remarks 10
2. Conclusions H
3. Recommendations . 14
4. Materials and methods T6
PCB materials and extraction procedures 16
Earth materials 19
Adsorption studies 21
PCB mobility studies 22
Microbial degradation 24
PCB volatility measurements 26
Analytical methodology 28
5. Solubility of PCBs in water 36
Results and discussion 36
6. Adsorption of water soluble PCBs by soil materials
and coal chars 48
Results and discussion 48
7. Mobility of PCBs and Dicamba in soil materials:
Determination by soil thin-layer chromatography 58
Results and discussion 58
8. Degradation of PCBs by mixed microbial cultures 63
Results and discussion 63
9. Effect of humic acid and soil on volatilization
of PCBs from water 82
Results and discussion 82
References. 87
Appendix 92
-------
FIGURES
Number Page
1. Biphenyl structure 1
2. Modified Erlenmeyer flask for extraction of PCBs from aqueous sample 16
3. GC chromatogram of internal standard (tribromobenzene) in hexane 17
4. Chromatogram of internal standard and four PCB isomers 18
5. Flow diagram of experimental protocols used to study PCB volatilization from
aqueous solutions 26
6. Purging system using 1000-mL Erlenmeyer flask 28
7. Water-soluble Aroclor 1242 using temperature-programmed column 29
8. Water-soluble Aroclor 1242 on second capillary column 31
9. Water-soluble Aroclor 1242 on third capillary column 32
10. Reconstructed chromatogram of Aroclor 1016 obtained from mass-spectrometer 33
11. Teflon-plugged reaction vessel used in perchlorination reaction 34
12. Schematic block diagram of the perchlorination procedure f 35
13. Comparison of hexane-soluble Aroclor 1242 and used capacitor fluid 37
14. Solubility of Aroclor 1242 and used capacitor fluid 1242 in water as a
function of time 40
15. GC chromatograms of water-soluble Aroclors and used capacitor fluid 43
16. PCB adsorption by earth materials 49
17. PCB adsorption by low-temperature ashes of earth materials 50
18. Capillary column analysis of Aroclor 1242 remaining in solution 51
19. Percentage of isomers in water-soluble Aroclor 1242 remaining in solution.
after adsorption by char 57
20. PCB adsorption constants as function of total organic carbon content 57
21. PCB adsorption constant as function of surface area 57
22. Three-variable regression analysis of PCB adsorption constant as function
of total organic carbon content and surface area . 57
23. Retention of PCB in soils leached with carbon tetrachloride 60
24. Water-soluble Aroclor 1242 after 15 days of degradation v 65
25. Original water-soluble Aroclor 1242 solution prior to degradation 66
26. Comparison of 5-day blank and 5-day degradation of Aroclor 1242 using S culture . . .69
27. Results of 5-day blank of Aroclor 1242 using the N culture 70
28. Results of 5-day degradation of Aroclor 1242 using the N culture 70
29. Degradation at timed intervals of selected PCB isomers 73
30. Degradation at timed intervals of selected PCB isomers 74
31. Relative distribution of PCB isomers in soil and sludge extracts with
respect to retention time 79
32. Comparison of attenuation of volatilized PCBs by dissolved humic acid addition ... 83
33. Comparison of attenuation of volatilized PCBs by soil addition 85
VI
-------
TABLES
Tables Page
1. Characteristics of earth materials used in adsorption studies 20
2. Chemical characteristics of earth materials used in adsorption studies 20
3. Characteristics of soil materials used in PCB mobility studies 21
4. Gas chromatographic conditions used for volatilization study 27
5. Relative response of a SSfH electron capture detector to tribromobenzene
and PCB isomers 36
6. Quantitative composition and percentage of contribution of individual isomers
in Aroclors and capacitor fluid '. 38
7. Percentage of PCB isomers in water-soluble Aroclors 1242 **!
8. Comparison of aqueous solubility of PCBs in Aroclor 1242 to pure isomers ^*
9. Quantitative comparison of water-soluble isomers in Aroclor 1242 with respect
to time ^2
10. Summary of isomer composition of water-soluble Aroclors and capacitor fluid **5
11. Summary of solubility of PCBs and capacitor fluid in water . . . 46
12. Quantitative distribution of PCBs after adsorption by several adsorbents 53
13. Mobility-of Aroclor 1242 and 1254 and Dicamba in soil materials leached with
various solvents "
14. Mobility of Aroclor 1242, Aroclor 1254, and Dicamba on silica-gel TLC plates
using various leaching solvents 61
15. Results of linear regression analysis of retention of Aroclor 1242 and Aroclor
1254 by soils 61
16. Results after 15 and 45 days of degradation using culture grown in acetic acid ... 63
17. Comparison of mixed cultures grown on biphenyl only and those grown on
acetic acid 64
18. Comparison of sonified extracts and pH 2 extracts 68
19. Comparison of degradation of water-soluble Aroclor 1242 by three mixed
cultures of soil microorganisms 71
20. Results of kinetic study of degradation of water-soluble Aroclor 1242 by the
N culture 72
21. Rate constants and reaction orders for selected PCB isomers 76
22. Cometabolization of PCBs with sodium acetate by mixed culture microorganisms .... 78
23. Relative amounts of PCB volatilized from water containing humic acid or soil
compared to control 82
24. Relative attenuation of volatilization of PCBs by soil as compared with
attenuation due to adsorption from solution by same soil 84
vii
-------
ACKNOWLEDGMENTS
In addition to the authors listed on the title page, the following were
involved in writing parts of this report: A. K. Au, R. R. Clark, H. Meng,
M. C. Lee, and M. L. Miller. Their names are also listed in a footnote in
the portion of the Results and Discussion Section to which they contributed1.
The authors gratefully acknowledge the U.S. Environmental Protection
Agency, Municipal Environmental Research Lab., Solid and Hazardous Waste
Research Div., Cincinnati, Ohio, for partial support of this study under
Grant R-804684-01, Mr. Richard A. Carnes, Project Officer; Monsanto Chemical
Co., St. Louis, Missouri, for supplying the Aroclors; Illinois Power Co.,
Decatur, Illinois, for supplying the used capacitor; American Colloid Co.,
Skokie, Illinois, for supplying the montmorillonite; the Ottawa Silica Co.,
Ottawa, Illinois, for supplying the quartz sand; T. M. Hinsley, for supplying
the sludge-soil samples; and C. N. Goddard, for supplying the Hudson River
sediment samples used in this study.
The authors also wish to thank Dr. J. H. Kim, Dr. N. F. Shimp,
Dr. R. R. Ruch, Dr. J. Thomas, Jr., Dr. R. R. Frost, Dr. L. R. Follmer,
Dr. J. J. Hassett, Dr. W. A. White, F. L. Fiene, J. E. Johnson, A. D. Debus,
Dr. K. E. Smith, J. M. Masters, L. Kohlenberger, J. D. Steele,
L. R. Henderson, J. F. Ashby, Dr. F. B. DeWalle, R. A. Cahill,
Dr. J. K. Frost, L. R. Camp, R. A. Keogh, and J. K. Kuhn for assistance
with parts of this research.
viii
-------
SECTION 1
INTRODUCTION
Polychlorinated biphenyls (PCBs) is
mixtures of chlorinated organic compounds
much discussion recently because of their
persist and accumulate in the environment
air, water, soil, sediments, fish, birds,
all over the world. PCBs are not acutely
have been observed. Therefore, PCBs have
hazard to human health as well as to the
EHRC, 1976).
CHEMICAL STRUCTURE
a generic term applied to certain
PCBs have become the subject of
effect on the environment. They
and have been found in samples of
and mammals (including humans)
toxic, but several chronic effects
been considered a significant
environment (Gustafson, 1970, 1972;
PCBs belong to a class of aromatic chlorinated organic compounds that
comprise complex heterogeneous mixtures of very closely related isomers.
The basic biphenyl structure is shown in Figure 1. PCBs are made by sub-
stituting chlorine atoms for one or more of the hydrogen atoms at the
numbered positions of the biphenyl structure. Possible chlorine substitu-
ted biphenyl isomers number 209, but commercial preparations contain less
than the total possible; about 103 have been identified in various samples
(Widmark, 1967; 1968). The different
commercial preparations may vary from
batch to batch with regard to the
specific composition of the mixture
of chlorinated biphenyl isomers (EHRC,
1976; Lloyd et al., 1976).
MANUFACTURE AND INDUSTRIAL USES
PCBs have been manufactured in
the United States since 1929 by the
Monsanto Chemical Company; an esti-
mated 800 million pounds have been
produced since that time. In 1970,
the year of peak production, more
than 85 million pounds of PCBs were
produced in the United States alone,
57 percent of which was in the form Fi9urei. BIPHENYL STRUCTURE: Positions 2 toe and 2-
n-F flvr^lnv" 19/19 (UE\I ^Q79^ A to 6'indicate ten possible positions for chlorine
OT HrOCIOr \mt. ^tltW, iy/(i;. An substitution. Different amounts of chlorine substi-
estimated one-half million pounds tion form the various PCBS.
-------
per year of PCBs were imported into the United States from foreign manufac-
turers (EHRC, 1976).
The sole United States manufacturer of PCBs was the Monsanto Chemical
Company located near East St. Louis, Illinois. Monsanto has marketed its
PCB products in the United States under the trade name AROCLOR. A four-
digit numbering system was used to identify various Aroclors. The first
two numbers, 12, specified polychlorinated biphenyls, and the last two
numbers referred to the approximate percentage of chlorine in the mixture.
For example, Aroclor 1242 means a PCB mixture with 42 percent chlorine by
weight (EHRC, 1976; Nisbet and Sarofim, 1972). A product called Aroclor
1016 came into use after 1971 which did not follow the above numbering
sequence. This product contained approximately 41 percent chlorine and was
an attempt by Monsanto to produce an Aroclor that could be used in the same
applications as Aroclor 1242; it contained fewer of the higher chlorinated
isomers and was thus more degradable and less persistent in the environment
(Tucker, Litschgi, and Mees, 1975). Aroclors can also be classified by the
average number of chlorine atoms per molecule. Thus, Aroclor 1232, 1242,
1248, 1254, 1260, and 1262 correspond to di-, tri-, tetra-, penta-, hexa-,
and hepta-chlorobiphenyls, respectively. Other trade names for PCB products
produced by foreign manufacturers include: KANECHLOR and SANTOTHERM (Japan),
PHENOCLOR and PYRALENE (France), FENCLOR (Italy), CLOPHEN (Germany), and
SOVOL (Russia) (EHRC, 1976; Lloyd et al., 1976).
PCBs come in both solid and liquid forms and have properties appropriate
to a wide range of industrial applications. In resin form they were used as
protective coatings, plasticizers and extenders, sealers in waterproofing
compounds and putty, asphaltic materials, printing inks, and synthetic adhe-
sives. In liquid form they were used as dielectrics, hydraulic fluids, ther-
mostats, cutting oils, extreme pressure lubricants, grinding fluids, and heat
transfer media. As solids they were used to impregnate carbon resistors, as
sealers, and as impregnating agents for electrical apparatus. Mith few
exceptions, other than environmental concerns, the unique properties of PCBs
qualify them as the best chemical known to make capacitors, transformers,
hydraulics, gas turbines, and vacuum pumps work efficiently and safely
(EHRC, 1976).
In 1971, because of environmental concerns, Monsanto voluntarily stopped
production of Aroclor 1260 and restricted the sale of other PCBs to only
"closed" systems. Closed systems include PCB-containing insulating fluids
used in electrical transformers and capacitors; these two applications account
for essentially all the current use of PCBs in the United States (ANSC, 1976).
On October 5, 1976, Monsanto announced that it would cease to manufacture
and distribute PCBs by October 31, 1977. A timetable set by the United
States Environmental Protection Agency (EPA) called for gradual phasing out
of PCB manufacturing by January 1, 1979, and a ban on all PCB processing or
distribution in commerce by July 1, 1979 (ES&T, 1977). These steps should
reduce the introduction of PCBs into the environment in the future, but
millions of pounds of PCBs still exist in transformers and capacitors still
in service. The environmentally safe disposal of these fluids will continue
to be of concern for more than a decade.
-------
ENVIRONMENTAL EFFECTS
PCBs were largely ignored as environmental contaminants until 1966 when
Jensen (1966) reported PCB contamination of fish, eagles, and humans. PCBs
still did not attract much concern as hazardous chemicals until the incidence
of contaminated cooking oil in Japan (Risebrough et al., 1968) and of con-
taminated chicken feed in the United States (Duke, Lowe, and Wilson, 1970).
Laboratory studies with animals have shown that PCBs can cause enlargement
of the liver, induction of hepatic microsomal enzymes, reproductive failures,
gastric disorders, skin lesions, and tumors in birds and mammals (Nelson
et al., 1972).
The best documented case of toxic effects of PCBs in humans is the 1968
"Yusho" incident in Japan. About 2,000 Japanese people experienced lesions
of the skin, facial swelling, and neurological disorders. The individuals
with the highest dosages had nausea, lassitude, anorexia, impotence, and
hematuria. The severity of the symptoms were directly related to the
quantities of PCBs ingested. Placental transfer of PCBs had occurred in
pregnant women and had adversly affected the fetuses. The toxic effects
noted were similar to those reported in animal studies (Nelson et al. , 1972;
EHRC, 1976).
Several studies with animals have shown that fatty tissue accumulates
about 10 times more PCBs than liver and more than 100 times more than blood,
heart, kidney, and brain (Burse et al.,1974; Grant, Phillips, and Villeneuve,
1971; and Curley et al., 1971). A study with refuse workers found that 81
percent had detectable levels of PCBs in their blood (Hammer et al., 1972).
PCBs have found their way into the marine and terrestrial environment.
Risebrough et al. (1968) noted that PCBs, along with the chlorinated hydro-
carbon pesticides, have become the most widespread and ubiquous synthetic
pollutants in the global ecosystem. They also noted that PCBs induced hepa-
tic enzymes that degrade estradial, which leads to aberrations in calcium
metabolism, such as weak, thin egg shells.
Because PCBs are fat soluble and poorly metabolized, they have entered
the food chain and accumulated in the adipose tissues of fish, birds, and
mammals, including humans (Jensen et al., 1969; Hammond, 1972; Solly and
Shanks, 1974; Harvey and Steinhauer, 1975). The biological magnification of
PCBs that have entered the food chain seems to occur by a factor of 10 to
100 at each step. Fish and aquatic organisms accumulate PCBs by a factor
of TO4 over concentrations in the ambient waters. Predators at the top of
the food chain can accumulate PCBs by a factor of 107 over the concentra-
tions found in the ambient waters (EHRC, 1976). The major problem with
regard to PCB intake by man is that he resides at the top of most food
chains. Man can therefore amass substantial amounts of PCBs although only
trace levels are present in the waters of lakes and streams.
A two-year study of Lake Michigan showed that fish with the highest fat
content contained the highest levels of PCBs (Schacht, 1974). Other studies
have shown that salmon taken from Lake Michigan in 1974 contained PCBs in
-------
excess of the 5 ppm FDA tolerance level. Lake trout were reported to contain
PCBs as high as 43.8 ppm, and carp contained levels as high as 51.6 ppm
(EHRC, 1976).
The highest concentrations of PCBs are found in fish that come from the
Hudson River in New York. Fish taken from below the General Electric capac-
itor manufacturing plant contained PCBs in the range of 17 to 78 ppm; one
sample of rock bass contained 350 ppm (Ahmed, 1976). The bed and bank
sediments in a 40-mile reach of the river were estimated to contain more
than 400,000 pounds of PCBs, and concentrations as high as 5,600 ppm were
found (Maslansky et al., 1978).
Old electrical equipment has been discarded by many companies and the
PCBs have discharged into the atmosphere and waterways. PCBs are somewhat
volatile, and the migration of PCBs through air is considered to be one of
the basic mechanisms by which the ubiquitous presence of PCBs in nature
occurs (Persson, 1971). An average fallout rate of 40 to 80 yg/m2/year of
PCBs over continental North America was reported by the Panel on Hazardous
Trace Substances (Nelson et al., 1972).
DISPOSAL AND SOIL ATTENUATION
One problem of disposal involves the high costs and fees for transport-
ing PCB wastes to regional incinerators or approved landfills compared to
simply discarding the wastes. The costs at disposal sites that specially
treat or incinerate wastes may be as high as a dollar per gallon or more.
The costs for land burial are generally between 5 to 10 cents per gallon
(Troise and Kahn, 1978).
Incineration is considered the safest method for disposal of PCB wastes,
but this method is extremely costly and has some operating difficulties.
PCBs do not burn readily; the operating conditions of the incinerator must
be carefully controlled to prevent the reentry of PCBs into the environment
in the stack gases (Duvall and Rubey, 1977; Moon, Leighton, and Huebner,
1976).
Nisbet and Sarofim (.1972) estimated the rate of PCB disposal in dumps
and landfills in North America in 1970 to be 18,000 tons per year. In view
of this, land disposal of PCBs and related materials has concerned many
people (Jordan, 1977; Henderson, 1978) because surprisingly little is known
about the mobility of PCBs in soil systems.
The hydrophobic characteristic of PCBs make them easily adsorbed from
aqueous solution onto available surfaces; the amount of PCBs being adsorbed
depends on the nature of the surface (Haque, Schmedding, and Freed, 1974;
Lawrence and Tosine, 1976). Haque and Schmedding (1976) studied the adsorp-
tion characteristics of three selected PCB isomers by several adsorbents
and provided evidence that adsorption increases as the number of chlorine
atoms on the isomer increases.
Iwata, Westlake, and Gunther (1973) studied the persistence of PCBs in
six California soils. Preferential disappearance of some gas chromatograph
-------
(GC) peaks was evident and recovery of Aroclor 1254 was related to the
organic-matter content of the soil. In a separate study of PCB uptake by
carrots, 97 percent of the residue was found in the peel and very little
translocation into the plant tissue. Again, the lesser chlorinated biphenyls
were slowly dissipated and the more highly chlorinated biphenyls were not
appreciably affected (Iwata, Gunther, and Westlake, 1974).
Tucker, Litschgi, and Mees (1975) percolated water through soil columns
containing PCBs. In the worst case, less than 0.05 percent of the added
Aroclor 1016 was leached; they concluded that PCBs are not readily leached
from soils. Similarly, Scharpenseel, Theng, and Stephan (1978) found that
very little leaching of PCBs occurred and that the PCBs recovered from the
soil were associated with the soil organic matter.
Briggs (1973) reported that adsorption of unionized organic compounds
by soils was related to the organic-matter content of the soil and to the
octanol/water partition coefficients of the compound. On the basis of the
octanol/water partition coefficients for PCBs, he predicted that they would
be immobile in soils.
Lidgett and Vodden (1970) analyzed waters around a sanitary landfill
for PCBs and found the contamination levels to be below their detection
limits of 4 ppb. Similarly, Robertson and Li (1976) used GC/Mass Spectrom-
et'ry techniques and failed to detect PCBs in ground water. Hesse (1971)
reported that runoff from landfills was only a minor source of PCB contami-
nation to the environment. More recently, Moon, Leighton, and Huebner (1976)
reported that levels of PCBs in ground water in the vicinity of 11 sanitary
landfills were below detection (<1 ppb), but that low levels of PCBs were
found in waters from monitoring wells at several industrial PCB disposal
sites and lagoons. They concluded from analyses of water and splitspoon
soil samples that PCBs were present in most leachates from land disposal
sites and that PCBs have a strong affinity for soil. Gresshoff, Mahanty, and
Gortner (1977) have also reported that PCBs have a high affinity for soils.
Leis et al. (1978) studied PCB migration in ground water from 12 dredge
disposal sites in the upper Hudson River Valley of New York. They concluded
that the velocity of the front of PCB advance was about 2 orders of magnitude
slower than the velocity of the ground water flow. They further concluded
that inputs of PCB that were due to migration through ground water were
negligible in comparison to other inputs of PCBs.
The information presently available indicates that PCBs have a strong
affinity for soil and that the nature of the surface, the organic matter
content, and the chlorine content and/or hydrophobicity of the individual
PCB isomers are factors affecting adsorption. Quantitative data on the
adsorption capacities and factors affecting PCB adsorption by earth materials
are needed to assess the impact of soil attenuation mechanisms for restrict-
ing PCB mobility in the environment.
MICROBIAL DEGRADATION
The above studies suggest that the interaction of PCBs with soil part-
icles is an important attenuation mechanism. Another potentially important
-------
mechanism is the degradation of PCBs by microorganisms.
Biphenyl Degradation
Various species of microorganisms that can degrade biphenyl have been
reported. Lunt and Evans (1970) were the first to report the microbial
degradation of biphenyl. The microorganisms used were gram-negative bac-
teria found in soil. They reported that biphenyl was degraded to 2,3-
dihydroxybiphenyl. This compound was then degraded to a a-hydroxy-3-phenyl-
muconic semi aldehyde and then to phenylpyruvate.
Catelani et al. (1971, 1973) reported that Pseudomonas put-ida could
degrade biphenyl. The biphenyl was degraded to 2,3-dihydroxybiphenyl and
further degraded to benzoic acid and 4-hydroxy-2-oxovaleric acid.
Gibson et al. (1973) reported a Beijerinckia species that could degrade
biphenyl to 2,3-dihydroxybiphenyl. They also reported that the intermediate
in this step, cis-2,3-dihydroxy-l-phenylcyclohexa-4,6-diene was formed by
bacteria, whereas the trans isomers were formed in mammals.
PCB Degradation by Pure Cultures
Wallnofer et al. (1973) isolated a soil fungus, Rhizopus japonicas,
which converts 4-chlorobiphenyl to 4-chloro-4'hydroxybiphenyl.
Ahmed and Focht (1973a, 1973b) reported the degradation of several PCB
isomers by two species of AchromobaoteT . They isolated the two species from
sewage, the first by biphenyl enrichment and the second by 4-chlorobiphenyl
enrichment. They reported that the products produced by these two species
were different; this suggests different metabolic pathways for degradation.
They showed that degradation of the unsubstituted aromatic ring was preferred.
This, and the observation that no dechlorination occurred, lead to a build-
up in chlorobenzoic acids. Recently, Omori and Alexander (1978) reported
that certain alkanes can be dehalogenated. If dehalogenation, along with
degradation, can occur, then PCBs can be broken down into small nonchlori-
nated compounds. Kaiser and Wong (1974) reported the isolation of some
microorganisms that degraded PCBs into metabolites that contained neither
chlorine nor oxygen. This is quite unusual because the PCBs are degraded
aerobically and therefore should be oxidized and should contain oxygen.
Baxter et al. (1975) measured PCB degradation by a species of Noaard-ia
and a species of Pseudomonas. With the Nooardia species they reported 88
percent and 95 percent degradation for 52 and 100 days, respectively. With
the Pseudomonas spectes they reported 76 and 85 percent for the same periods.
Both studies were'done with Aroclor 1242. They also did work with single
PCB isomers and in these studies found that PCB isomers with up to 6 chlo-
rines could be degraded to some degree and that the pattern of chlorine
substitution determines the ease of degradation. They also found that the
two species would degrade some isomers differently; for example, the Nocardia
could not degrade 4,4'-dichlorobiphenyl in 121 days, whereas the Pseudomonas
species degraded it about 50 percent in 15 days. They also reported a form
of cometabolization, in that some isomers degraded more slowly when present
-------
as the only carbon substrate, but degraded more rapidly when in a mixture
of isomers or when biphenyl was present. An example of this was 2,5,4'-
trichlorobiphenyl. When alone this isomer degraded only 15 percent in 73
days; however, with biphenyl the degradation was 60 percent in 73 days.
Furukawa and Matsumura (1976) isolated an Alkali-genes species from a
lake sediment by enrichment with biphenyl. The PCBs were degraded in two
steps. The first step was the formation of a yellow product, identified
as a chlorinated derivative of 2-hydroxy-6-oxo-6-phenylhexa-2,4-dienoic acid.
This compound was then broken down into chlorinated benzoic acids. In add-
ition, the isomers with 4'-chlorine produced longer lasting yellow inter-
mediates. PCB isomers with chlorines on only one of the rings were degraded
more easily than those with chlorines on both rings. They also found that
lower chlorinated isomers were more easily degraded and that isomers with
up to five chlorines could be degraded.
A later study by Furukawa, Tonomura, and Kamibayashi (1978) with Alka-
ligenes and Aoine-bobaoter used 31 different PCB isomers and showed similar
results. They found that isomers with 3,6 or 2,2' chlorines were poorly
degraded. They also reported less degradation as the number of chlorines
per molecule increased. The two species showed about the same amount of
degradation of each isomer except for 2,4,6-trichlorobiphenyl. Several
trichlorobiphenyl isomers could be degraded almost completely in less than
three hours by these microorganisms.
Sayler, Shou, and Colwell (1977) isolated a species of Pseudomonas
from sea water that degraded PCBs. They obtained between 9 and 39 percent
degradation of Aroclor 1254 in 22 days, the percentage depending on the
starting concentration. In 60 days they found between 63 and 84 percent
degradation, again the percentage depending on the starting concentration.
At 22 days, the lowest degradation was at the lowest concentration, 10 yg/L:
at 60 days the lowest degradation was at the highest concentration, 1000yg/L
Ballschmitter, Unglert, and Neul (1977) proposed several schemes for the
microbial breakdown of PCBs into various metabolites, including chlorinated
benzoic acids.
It has been known for some time that PCBs could be degraded by mammals
and birds. Sundstrom, Hutzinger, and Safe (1976) presented a review of the
metabolism of PCBs in animals. Hutzinger, Safe, and Zitko (1974) also
presented a review of PCBs in both animals and microorganisms. Sayler,
Shou, and Colwell (1977) noted that the degradation of PCBs in animals could
actually be caused by the action of the intestinal microbiota.
PCB Degradation in Sewage
Many PCBs end up in sewers, either through spills or dumpings, and then
go into activated sludge systems. Tucker, Litschgi, and Mees (1975) worked
with a continuously fed activated sludge unit. At a feed level of 1 mg/48
hrs., they reported 100 percent degradation of biphenyl, 81 percent degrada-
tion of Aroclor 1221, 33 percent degradation of Aroclor 1016, 26 percent
degradation of Aroclor 1242, and 15 percent degradation of Aroclor 1254.
With Aroclor 1221, only the 3-, 4-, and 5-chlorine isomers were not degraded
-------
Mihashi et al. (1975) reported 50 percent PCB degradation in activated sludge
and that the degree of degradation decreased as chlorine substitution
increased.
Herbst et al. (1977) studied the fate of two radiolabeled isomers of
PCB, 2,5,4'-trichlorobiphenyl and 2,4,6,2',4'-pentachlorobiphenyl. They
determined that the two isomers were poorly degraded by the activated sludge.
In addition, most of the PCBs ended up in the activated sludge and not in the
supernatant. Jordan (1977) reported on the case of Williams, Indiana;
activated sludge containing up to 300 ppm PCBs was used as fertilizer and
harmed both the people and the wildlife of the area.
Inhibition by PCBs
PCBs also inhibit the growth of microorganisms that have not been
acclimated to PCBs. Bourquin and Cassidy (1975) reported that a sizeable
proportion of estuarine bacteria are sensitive to PCBs. They pointed out
that if PCBs were present in large amounts, the normal microbial heter-
otrophic activity could be disrupted.
Murado, Tejedor, and Baluja (1976) reported that Aroclor 1254 would
inhibit the formation of mycelium of AspergiZlus flavus . They also found
that no degradation of PCBs occurs in these cultures. Blakemore and Carey
(1978) reported the inhibiting of growth of two marine bacteria by low
concentrations of Aroclor 1254. They found that the inhibition was dose
dependent and that respiration was not inhibited. Blakemore (1978) also
reported the inhibition of nucleic acid synthesis in a marine bacteria and
subsequent inhibition of growth in these bacteria.
It is clear from the above studies that PCBs are somewhat degradable and
may come under attack by soil microorganisms. If a mixed culture of soil
microorganisms could be enriched in organisms that could degrade PCBs at a
significant rate, it would then be feasible to use biological processes to
treat PCBs; or PCB wastes could be inoculated with the culture before land
disposal, assuming that the organisms would be competitive in the particular
environment.
ANALYTICAL METHODOLOGY
Quantitative Analysis
The analysis of PCBs has been a matter of estimation; attempts at
quantisation of PCBs have most often been done by measuring the heights of
major gas-liquid chromatographic .peaks, Koeman et al. (1969), semiquanti-
tatfvely measured only one of the major peaks, Skrentny, Hemken, and Dorough
1971) used two, and Zitko (1971) used three quantification peaks. Reynolds
(1971) applied Koemanls method and took an average of two or more major
peaks in the mixture as the standard. Hansen et al. (1971) averaged the
heights of five major peaks. Likewise, PCBs also have been quantified by
summing the areas of all the individual peaks obtained by gas-liquid chroma-
tographic analyses. Jensen et al. (1969) quantitatively estimated the con-
centration of PCBs by the total peak areas. Armour and Burke (1970) measured
-------
PCBs in terms of Aroclor 1254 by using the total area of the peaks.
Risebrough, Reiche, and Alcott (1969) multiplied a factor derived from meas-
urements of standard PCB solutions by a microcoulometric detector to the
areas measured by electron capture detector (ECD). Beezhold and Stout (1973)
used a mixed standard to quantify PCBs.
All the gas-liquid chromatography used for the above studies employed
an electron capture detector on the chromatograph. Unfortunately, the
electron capture detector does not respond quantitatively the same for all
PCB isomers, as noted by Rote and Murphy (1971), Cook (1972), Wiencke and
Roach (1972), and Beezhold and Stout (1973). A procedure designed to over-
come this problem was to perchlorinate the PCB sample (Berg, Diosady, and
Rees, 1972; Armour, 1973; Berg, Rees, and Ali, 1978; Burkard and Armstrong
(1978). In the perchlorination procedure, the PCBs were reacted with antimony
pentachloride (SbCl5) to form decachlorobiphenyl (DCB). The advantage of the
perchlorination procedure is that all the chlorinated biphenyl is detected
as one peak by the GC, thus giving improved quantitation and speed of analy-
sis. In addition, the ECD is more sensitive to the higher chlorinated com-
pound. Studies by Trotter and Young (1975), however, showed problems in the
perchlorination reaction with SbCl5 that were due to trace amounts of bromine
in SbCl5 in the form of SbBrCV Because the bromine has a higher affinity
for biphenyl than the chlorine, bromonanochlorinated biphenyl (BNCB) is
formed during perchlorination. The BNCB peak would show up in a GC trace
after the DCB peak. Further studies are needed to improve the quantitation
of PCBs by this procedure.
Improvements in PCB analysis can be made by the use of open tubular
glass capillary columns which have liquid phases custom-coated on the walls
of the column. The apparatus and dynamic-coating procedures of Grob (1975)
and Grob, Grob, and Grob (1977) can also be used.
Generally, liquid phases for capillary gas chromatography can be divided
into apolar, medium polar, and polar phases. Polar phase columns were notor-
iously short-lived. Apolar liquid phases (OV-17, OV-101, SE-30, etc.) were
widely accepted as the most efficient coatings because of their longer life;
these generally are the ones available commercially. Because of the aroma-
tic characteristics of PCBs, a polar phase is needed to achieve maximum sep-
aration efficiency. The merit of the glass-surface pretreatment methods
given by Grob is that polar phases such as Emulphor ON-870 can be coated on
the glass and used for PCB analysis with increased column life. In addition,
the column can easily be regenerated. When a column has deteriorated to an
unusable quality, a cleaning solution can be passed through the column to
wash out the residual liquid, and the column can then be recoated. In the
case of silicone liquid phases, such as OV-17, OV-101, and SE-30, the regen-
eration step is not possible.
Mobi1i ty Measurements
The technique of determining pesticide mobility in soils by soil thin-
layer chromatography, or soil TLC, was introduced by Helling and Turner
(1968). Since the introduction of the technique, the mobility of a large
number of pesticides and radionuclides in a variety of soils has been tested
-------
(Helling 1971a, 1971b, 1971c; Reeves, Francis, and Duguid, 1977). Soil
TLC is a laboratory technique that uses soil as the adsorbent phase and
developing solvent such as water, leachate, and organic solvent in a TLC
Syu-?™;. The system 1s restively simple and yields quantitative data on the
mobilities of chemical constituents in soils that appear to correlate well
with trends noted in the literature (Helling and Turner, 1968; Briggs, 1973-
Reeves et al., 1977).
CONCLUDING REMARKS
An exact value for the solubility of PCBs in water is difficult to
determine because PCBs are mixtures of substituted isomeric biphenyls that
have different numbers of chlorine atoms. The structure of the PCB molecule
was shown in figure 1 to illustrate the various possible combinations of
chlorine substitution. Of the 209 possible isomers, about 103 usually occur
in commercial PCB mixtures (Widmark, 1978). PCBs are hydrophobic and are
only slightly soluble in water. Isomers with lower chlorine substitution
have been shown to be preferentially dissolved in water (Zitko, 1971; Hoover,
(1971). Another important factor in the quantitative analysis of PCB solu-
bility is that PCBs dissolve in water very slowly. Therefore, there must
be a time-dependent study before the absolute value of the water solubility
of PCBs can be obtained (Haque and Schmedding, 1976). PCBs are also subject
to volatilization (Zitko, 1970) and photodecomposition (Ruzo, Zabik and
Shiretz, 1972; Safe and Hutzinger, 1971; and Hutzinger, Safe, and Zitko,
(1972). In view of the above difficulties, care must be taken when designing
experimental protocols and interpreting data.
The information presently available regarding soil attenuation mecha-
nisms indicates that PCBs have a strong affinity for soil and are somewhat
degradable by microorganisms. There is no evidence that ground water around
sites containing relatively low levels of PCBs, such as sanitary landfills,
has become contaminated with PCBs; however, surface and ground waters around
some industrial disposal sites and around lagoons: containing relatively
large quantities of PCBs have become contaminated by leaching PCBs. The
mechanisms of transport of PCBs in the biosphere and the mechanisms of
attenuation in soil are essentially unknown. Data on the factors affecting
PCB attenuation by earth materials and the microbial degradation of PCBs
would provide a useful basis for determining waste treatment methods, for
predicting PCB migration under landfills, and for selecting and designing
future disposal sites.
The purposes of the present project were (a) to conduct a literature
review of pertinent information on the attenuation of PCBs in soil materials
and their behavior in the environment; (b) to measure the adsorption capa-
city of selected earth materials for pure PCBs and PCB wastes; (c) to
evaluate quantitatively the effects of biological degradation, volatiliza-
tion, time, organic carbon content, and adsorbent structure on attenuation
of PCBs; (d) to use this data to develop a mathematical model that will
allow prediction of PCB adsorption and mobility; (e) to measure quantita-
tively the aqueous solubility of PCBs; and (f) to further develop analytical
procedures that will allow quantitative measurement of PCBs contained in
aqueous solutions.
10
-------
SECTION 2
CONCLUSIONS
Studies of the aqueous solubility, adsorption, mobility, microbial
degradation, and volatility of PCBs have led to the following conclusions:
1. The rate of dissolution of Aroclor 1242 and a used capacitor fluid
(Aroclor 1242 impregnated) in water were essentially identical and
required 5 months under the experimental conditions employed in this
study (the rate of dissolution of PCBs probably depends on the stirring
rate).
2. The solubility of individual PCB isomers was greatly reduced in the
Aroclor mixture where other PCB isomers were present than when compared
to their individual solubilities.
3. The water-soluble fraction of PCBs was richer in the low-chlorinated
isomers than that in the original PCB fluids.
4. Used capacitor fluids can be assumed to be the same as Aroclor 1242 in
terms of the pattern of isomer distribution in the original fluid and
the water-soluble fractions, in solubility, and presumably in its
behavior in the environment. Age and the "burn out" process do not
appear to alter appreciably the Aroclor 1242.
5. Some particular isomers in Aroclor 1242 were relatively soluble com-
pared to their counterparts with similar degrees of chlorination.^
They were identified as 2-mono-; 2,4'-di-~; 2,5,2'-tri-; 2,3,2'-tri-;
2,5,4'-tr>; 2,5,2' ,5'-tetra~; 2,4,2' ,5'-tetra-; 2,4,2',4'-tetra-;
2,3,2'5'-tetras and 2,4,3',4'-tetra-chlorobiphenyl.
6. The solubilities of Aroclor 1016, 1221, 1242, 1254 and capacitor fluid
at room temperature were 906 ppb, 3516 ppb, 703 ppb, ^70 ppb, and 698
ppb, respectively.
7. A simple linear relation described the total adsorption of PCBs from
aqueous solution by earth materials and yielded an adsorption constant
(K) unique to each adsorbent.
8. There were no significant differences between adsorption of water-
soluble Aroclor 1242 and a water-soluble used capacitor fluid
(Aroclor 1242 impregnated) by any of the five earth materials.
11
-------
9. Adsorption of PCBs by the five earth materials; medium-temperature coal
char (me), high-temperature coal char (ETC), Catlin soil (cs), mont-
morillonite clay (MO), and Ottawa silica sand (os) followed the series
MTC > ETC > CS > MC > OS.
10. Higher chlorinated isomers were preferentially adsorbed over lower
chlorinated isomers.
11. Very highly significant (.001 level) linear correlations were found
between K and total organic carbon (TOC)\ K and C02-surface area (SA)\
and KS TOO, and C02-A4. Poor correlations were found between K and
\\z-SA; and K and ethylene glycol-5.4. It was concluded that PCB mole-
cules were unable to penetrate the interlayer region of montmorillonite1
to be adsorbed.
12. TOO and C02-S4 were the most important properties of earth materials
controlling PCB adsorption; TOO was the dominant property by a factor
greater than three.
13. Aroclor 1242, Aroclor 1254, and capacitor fluid remained immobile in
soils when leached with water or leachate from the Du Page landfill
on soil thin-layer chromatography (TLC) plates, but were highly mobile
when leached with carbon tetrachloride. Dicamba showed the reverse
trend.
14. The mobilities of the PCBs and Dicamba in thin-layers of soil mat-
erials and silica-gel were highly related to their solubilities in
the solvent with which the TLC plates were being leached.
15. The higher chlorinated Aroclor 1254 was less mobile in soils leached
with carbon tetrachloride than the lower chlorinated Aroclor 1242.
16. There was a highly significant (.001 level) correlation between the
mobility of PCBs in the soil and the soil TOO content.
17. The degradation of water-soluble Aroclor 1242 by mixed cultures of
soil microorganisms occurred to a large extent in a short period of
time. Degradation rates varied from monochloro isomers that were
degraded 100 percent within 6 hours to tetrachloro isomers that
averaged 42 percent degradation after 15 days.
18. In general, lower chlorinated isomers were more easily degraded than
higher chlorinated isomers.. Position of chlorine substitution
affected the rate of PCB degradation.
19. The predominant organisms found in the mixed cultures were Alkaligenes
odorans} Alkaligenes denitrifieans, and an unidentified bacterium.
20. Volatilization of PCBs from aqueous solution was reduced by the
presence of soil and the reduction was due to preferential sorption
onto the soil particulates.
12
-------
21. Substantial binding of PCBs by soluble humic substances occurred,.
especially for the higher chlorinated isomers, but such binding was
less effective than that by soils in reducing volatilization from
aqueous solution.
22. Volatilization of PCBs from a landfill leachate can be expected to
be reduced substantially because of the high content of particulates
and high concentration of humic substances. Volatilization favored
the lower chlorinated isomers.
23. Higher chlorinated PCB isomers were less soluble in water, preferen-
tially adsorbed by soil materials, less mobile in soil, less degrad-
able by microorganisms, and less volatile from water than lower
chlorinated isomers.
13
-------
SECTION 3
RECOMMENDATIONS
The results and conclusions formulated from this study generally dealt
specifically with solubility and attenuation of PCBs from pure aqueous
solutions. The presence of soluble salts and organic materials in water, as
frequently occurs in landfills and the environment, could drastically alter
some predictions of PCB attenuation.
Soluble salts reduce the aqueous solubility of hydrophobic organic com-
pounds such as PCBs. The magnitude of the "salting out" effect and its
influence on attenuation of PCBs by earth materials should be determined.
Likewise, the effect of organic solvents and waste streams of various cate-
gories needs to be examined.
The effects of organic solvents on increasing the mobility of PCBs in
soils were dramatically illustrated during this study; however, the effects
of organic solvents on PCB adsorption remain unknown. These effects may be
of particular significance where montmorillonite clays are involved. The
organic solvent may expand the clay layers and allow access to the interlayer
surfaces; this should substantially increase the amounts of PCBs that are
adsorbed as compared to their adsorption from aqueous solution, where PCBs
were apparently restricted to adsorption on the external surfaces only. The
interrelationship of surface area and organic carbon content of soils needs
further elucidation to improve predictions of PCB migration through soils.
With regard to microbial degradation, it is recommended that studies
be carried out on the effects of soil on the degradation process. Substan-
tial and rapid degradation was observed in aqueous solutions, however,
selective adsorption of PCBs by the soil and competition from native soil
organisms may drastically alter the degradation. It is anticipated that
PCB degradation rates in soil will be substantially lower than those observed
in solution cultures.
The studies of volatility of PCBs were limited in scope. More work is
needed on the effects of soil properties and conditions on volatilization
before accurate predictions could be made of the flux of PCBs in soils due
to volatilization.
Thus, these multiple interactions of the properties of soil and the
chemistry of wastes must be considered when making predictions of the envi-
ronmental impact of PCBs. A major task still before us is the utilization
of quantitative data on the chemistry of soil to make predictions of PCB
14
-------
migration through soil materials. It is therefore recommended that further
studies involve a cooperative effort between soil chemists, ground-water
hydrologists, and modelers to implement by computer the prediction process
and to identify possible gaps in knowledge that may still bar the successful
prediction of long-term PCB migration through soil materials.
15
-------
SECTION 4
MATERIALS AND METHODS
Teflon stopcock
(large)
PCB MATERIALS AND EXTRACTION PROCEDURES
The PCBs Aroclor 1016, 1221, 1242, and 1254 were obtained from the Mon-
santo Chemical Company and were used without further purification. The
capacitor fluid was drained directly
from a burned-out 50-KVA capacitor
manufactured in 1966 and originally
impregnated with Aroclor 1242. This
capacitor was supplied by Illinois
Power Company and was scheduled for
disposal in a landfill. We believe
that this fluid is representative of
the type of PCB wastes that are nor-
mally disposed of in landfills.
A 20-mL volume of the individual
PCB fluids was placed in a 5-gallon
glass carboy, which had been cleaned
and baked at 450° C overnight. The
carboy was filled with water that
had been passed through a mixed bed
exchange resin and an activated car-
bon bed and was finally distilled in
glass. The carboy was then sealed
with an aluminum-wrapped rubber stop-
per and was wrapped with aluminum
foil and masking tape to protect it
from exposure to light. The solution
in the carboy was agitated gently with
a magnetic stirring bar coated with
Teflon. Gentle agitation was used to
avoid breaking the PCB fluids into a
fine suspension. Water samples were
collected from the carboys at regular
intervals, and 100 mL aliquots of the
PCB-water solution were filtered
through glass wool plugs tightly
packed into the stems of glass fun-
nels. The first IpO-mL aliquot was Figure2 Modifjed Er|enmeyerflaskforextractjonof pcBs
used to saturate the PCB adsorption from aqueous sample
16
-------
capacity of the glass wool and was discarded. Subsequent aliquots were
then used in the studies. The filtrates were then extracted with three
5-mL portions of either carbon tetrachloride or hexane (distilled in glass
grade, Burdick and Jackson Laboratories, Inc., Muskegon, MI) in a specially
designed extractor, which is shown in figure 2. Samples to be analyzed
by the perchlorination'procedure were extracted with carbon tetrachloride.
The samples to be analyzed by capillary column techniques were extracted with
hexane. The extracts were passed through a florisil column containing
approximately 75 percent florisil (Floridin Co., Tallahassee, FL) and 25
percent Na2S(V This was done to remove any water or polar materials that
may have been co-extracted from the samples (Reinert, 1970; Curley et al.,
1975). (The application of florisil, silica gel, and alumina to clean up
other contaminants from PCBs has been used successfully in many studies
[Reynolds, 1969; Snyder and Reinert, 1971; Berg, Diosady, and Reese, 1972;
and Ahnoff and Josefsson, 1975].)
The column was 5 mm I.D. x 50 mm long and was first flushed with sol-
vent. The sample was then passed through the column and collected in a 10
ml volumetric flask. The column was further eluted with at least one bed
volume of solvent which was subsequently added to the flask until exactly
2.49
I
IS
Figure 3. A GC chromatogram of internal standard (tribromobenzene) in hexane
17
-------
IS
3 8
18
24
Figure 4. A chromatogram of internal standard (IS) and four different PCB isomers;4-mono- (peak 3); 2,4'-di-(peak 8); 2,5,4'-
tri- (peak 18); and 2,4,2',4'-tetra-chlorobiphenyl (peak 24)
10 mL had been collected. The hexane filtrates from the florisil columns
were then spiked with tribromobenzene as an internal standard for quanti-
tative analysis. The possible interferences from the solvents and internal
standard were checked. Figure 3 is a chromatogram of a hexane solution con-
taining a high concentration of internal standard. The peaks in the chro-
matogram illustrate some of the impurities present in the reagents that do
not normally show up on chromatograms of samples or a standard (Fig. 4).
Figure 4 is a chromatogram of the internal standard and four PCB isomers.
The major impurity, shown in Figure 3 as a peak at retention time 2.49 min.,
elutes before any PCB peaks and therefore does not interfere with the
analyses.
Aqueous samples from adsorption studies with earth materials were
extracted with 20 ml of either carbon tetrachloride or hexane in three
18
-------
aliquots of 10, 5, and 5 mL each. The first extract was then placed in a
10 ml Kuderna-Danish (K-D) evaporator (Kontes Glass Co., Vineyard, NO) and
was concentrated to approximately 1 mL in a 100°C water bath by using NaCl
crystals as boiling chips. The second extract was added to the K-D and the
concentration process was repeated. The third extract was combined with the
condensed sample to give a final volume of about 6 ml. The 6 ml of condensed
sample was then passed through a florisil column as described above and was
analyzed for PCBs.
PCB Materials Used in Mobility Study
The PCBs and Aroclors 1242 and 1254 were obtained from the Monsanto
Chemical Company (St. Louis, MO) and were used without further purification.
The 14C labeled compounds were obtained from New England Nuclear Corporation
(Boston, MA). Gas chromatographic (GC) traces of the llfC labeled compounds
were identical to those of the Aroclor 1242 and 1254, therefore it was
assumed that there were no significant differences in the respective com-
pounds and that the 11+C labeled PCBs and Aroclors would behave similarly in
studies of mobility. The specific activities of Aroclor 1242 and 1254
were 0.119 m Ci/mg and 0.096 m Ci/mg, respectively.
The used capacitor fluid described previously was also used in this
study.
EARTH MATERIALS
Adsorbents
The adsorbents used in the PCB adsorption study were Ottawa sand (OS),
montmorillonite clay (MC), Catlin silt loam soil (CS), Illinois No. 6 coal
charred at 1800°F (HTC), and the same coal charred at 1200°F (MTC). The
two chars were soaked and rinsed with distilled-deionized water and then
oven dried at 110°C overnight. The coal chars are impure forms of activated
charcoal and were selected for study because of their potentially high
adsorption capacity for PCBs and their possible use as liners for disposal
sites accepting organic wastes. The sand, clay, and soil represented the
major soil materials and offered a range of chemical and physical character-
istics. These are presented in Tables 1 and 2. The chemical and physical
characterization was performed as described by Griffin and Shimp (1978).
The organic matter content of the adsorbents was varied by using the
low-temperature ashing (LTA) technique for bituminous coals described by
Gluskoter (1965) and Kuhn, Fiene, and Harvey 0978), An aliquot of each
sample was ashed except for Catlin soil for which three aliquots were ashed,
each for different time intervals. The changes in organic matter content
and surface area that are due to ashing are shown in Table 1.
Surface areas were measured by three methods, each selected to provide
different information about the surface characteristics of the adsorbents.
The BET gas adsorption method was used as described by Thomas and Frost
(1971). Both N2 and C02 were used as adsorbates. The third method was
ethylene glycol adsorption as described by Bower and Gschwend (1952).
19
-------
TABLE 1. Characteristics of earth materials used in adsorption studies.
Surface area
Adsorbent
PH
Ottawa silica sand (OS) 5.2
Montmorillonite clay (MC) 7.0
Montmorillonite clay (LTA)
Catlin silt loam AP
Catlin, 6 hr. LTA
Catlin, 1 2 hr. LTA
(CS) 7.1
Catlin, 336 hr. LTA
Coal char (1200°F)
Coal char (1200°F)
Coal char (1800°F)
Coal char (1800°F)
Constituent
AI203
CaO
Cl
Fe203
P20s
K2O
Si02
TiO2
MgO
Na20
S
Cl
V205
MnO
Na2O
Sn
Cu
Co
Ni
Be
Cr
Mo
Zr
Pb
Zn
F
(MTC) 7.3
(LTA)
(HTC) 7.0
(LTA)
TABLE 2.
High temp.
char
3.77
0.59
0.01
3.04
< 0.01
0.41
12.93
0.25
< 0.01
0.15
2.89
(ppm)
30
86
-
<50
117
< 2.4
19
1.6
26
6.8
60
< 2
36
125
CEC N,
(meq/100g) (m /g)
0.0 0.5
85.0 23.2
23.9
18.1 10.1
12.6
11.4
16.3
9.6 1.6
2.1
1 .7 3.4
10.0
Ethylene
CO2 glycol
(m2/g) (m /g)
0.4 <1
20.1 885
20.2
26.5 68
25.4
24.5
23.8
253 55
214
44 68
120
Sand Silt
100 0
0 0
1 1 .6 60.9
Organic
Clay carbon
0 ^.01
100 0.93
0.13
27.2 4.73
4.37
3.64
1.84
74.04
64.00
76.62
32.14
Chemical characteristics of earth materials used in adsorption studies.
Med. temp.
char
2.93
0.68
0.04
2.52
0.01
0.29
9.36
0.19
< 0.01
0.12
3.28
(ppm)
_
31
54
-
<50
45
< 2.4
8.6
1.3
24
6.3
22
< 2
23
95
Catlin
Ap soil
9.63
0.98
-
4.07
0.09
1.90
69.50
0.74
0.01
0.68
0.06
(ppm)
72
64
543
-
<50
45
9.6
22
3.1
62
4.0
237
16
72
160
Montmorillonite
clay
13.32
5.05
.
2.73
0.05
0.34
59.98
0.23
3.48
0.02
(ppm)
19
30
695
< 50
43
< 2.4
6.8
1.4
27
< 3.4
98
8.2
19
435
Ottawa
silica sand
0.19
0.07
.
0.09
< 0.01
0.01
97.38
0.01
< 0.01
_
0.01
(ppm)
38
< 9
< 2
27
< 50
4.3
< 2.4
< 1
< 1
< 1
< 3.4
80
< 2
< 1.6
< 10
20
-------
Soil Materials
Soil materials representing a wide range in characteristics were
selected for use in the mobility study. These included a pure silica sand,
seven soils, and medium-temperature (1200°F) coal char. The materials and
some of their chemical and physical characteristics are listed in Table 3.
All the soil materials were ground in a mortar with pestle and screened
(28 mesh) before use.
ADSORPTION STUDIES
All glassware was cleaned with tap water, rinsed with distilled-
deionized water, and baked in an oven at 450°C for at least 8 hours. Weighed
amounts of adsorbent were placed into 125 mL brown serium bottles. An
aliquot of PCB-saturated water was taken from the 5-gallon carboy described
earlier and filtered through glass-wool plugs that were packed tightly into
the stems of glass funnels. A 60-mL volume of filtrate was then added to
the reaction bottle. The bottle was covered with a sheet of baked aluminum
foil and sealed by a teflon-coated septum and a crimp-cap (Wheaton Labora-
tory Products, Millville, NJ, and Pierce Chemical Co., Rockford, IL). The
TABLE 3. Characteristics of soil materials used in PCB mobility studies.
Surface area
Material
Ava
Silty clay loam
Bloomfield
Loamy sand
Catlin
Silt loam
Catlin
Loam
Cisne
Silt loam
Coal char
(1200°F)
Drummer
Silty clay loam
Flanagan
Silty clay loam
Ottawa
Silica sand
CEC
pH (meq/100g)
4.5 13.1
5.7 0.8
7.1 18.1
8.7 4.7
5.9 8.0
7.3 9.6
6.7 29.4
6.6 23.3
5.2 0.0
N2 C02
(m2/g) (m2/g)
28.3 28.9
1.7 2.2
10.4 26.5
1 1 .5 1 1 .2
6.1 13.0
1 .60 253
22.1 29.1
12.6 32.8
0.5 0.4
Ethylene Organic
glycol Sand Silt Clay carbon
(m2/g) (%) (%) (%) (%)
55 2 69.6 28.4 1.18
2.0 82 10 8 0.21
68 11.6 60.9 27.2 4.73
17 35.9 43.5 18.5 0.57
23 13.8 70.8 14.8 1.30
55 ... 74.04
103 17.9 49.5 31.6 2.17
93 5.4 65.2 29.4 2.62
<1 100 0 0 <0.01
21
-------
brown bottles protected the PCB solutions from photodecomposition during
the adsorption process and the crimp seals prevented volatilization losses
during the reaction period. The bottles were shaken overnight in a water
bath at 25 ± 0.5°C.
The bottles were removed from the shaker and placed directly into a
Model JS-7.5 rotor and centrifuged at a constant temperature of 25 ± 1°C
in a Beckman Model J-21B centrifuge for 5 minutes at 6000 rptn. They were
then removed from the centrifuge head, the seals were broken, and 40 ml of
the clear supernatants were pipetted for PCB analysis.
The effect of reaction-time on PCB adsorption was studied by using the
methods described above. Two series of samples were prepared, each contain-
ing water saturated with Aroclor 1242 and a fixed amount of montmorillonite
clay. The two dosages of clay were 167 mg/L and 1667 mg/L. The bottles
were removed from the shaker after various time intervals and analyzed for
their PCB content. The results showed that a shaking time of 8 hours was
sufficient to reach equilibrium conditions. Therefore, all samples were
shaken overnight to ensure equilibration and for convenience of analysis.
Blanks were carried through all experiments to determine the background
levels of contamination of the adsorbents with PCBs and to evaluate the
adsorption of PCBs onto the surfaces of the glassware.
The amount of adsorption was determined from the difference between the
initial concentration and the equilibrium concentration multiplied by the
volume of solution. A blank was subtracted and the amount adsorbed by each
sample was computed on a unit basis by dividing by the dry weight of the
adsorbent.
PCB MOBILITY STUDIES
The mobilities of Aroclor 1242, Aroclor 1254, and Dicamba were studied
using the soil TLC technique described by Helling (1971a). Dicamba, a pesti-
cide of known high mobility, was used as an internal standard. The adsorb-
ent was slurried with water until moderately fluid, and was applied with a
spreader to a clean glass plate (20 cm x 20 cm) that had been washed with
ethanol and acetone. The adsorbent was spread to a thickness of 0.5 mm
and then air dried. A horizontal line was scribed with a stainless steel
spatula at 12 cm above the base to stop solvent movement; vertical lines were
scribed 2 cm apart to separate the various treatments. The radioactive
compounds were spotted on 2 cm from the base and leached 10 cm with a devel-
oping solvent. The activity of the lkC labeled PCB that was spotted was
22,000 d.p.m. (84 ng for 1242, 104 ng for 1254). The plate was immersed in
0.5 cm of solvent in a closed glass chamber and was removed when the wetting
front reached the horizontal line. Leaching was thus ascending chromato-
graphy. The developed plate was then removed and air dried. A piece of
8 x 10 inch medical x-ray film was placed in direct contact with the devel-
oped plate for a period of 2 weeks. The resulting autoradiograph indicated
the relative movement of the compound, which was measured as the frontal
Rf of the spot or streak. The Rf value is defined as the ratio of the dis-
tance the compound moved relative to the distance the solvent moved. The
Rf value is a quantitative indication of the front of PCB movement and a
22
-------
reproducible index of mobility.
The soil TLC plates were developed with three leaching solvents;
distilled-deionized water, Du Page leachate, and carbon tetrachloride.
Distilled water was further purified by passage through a Milli-Q (Millipore
Corp.) reagent grade (> 10 megohm resistance) water system. The organic
solvents were glass-distilled from Burdick and Jackson (Muskegon, MI) and
the leachate was collected from the Du Page County, Illinois sanitary land-
fill (well MM-63). The site description and well location were described
by Hughes, Landon, and Farvolden (1971) and the chemical characterization
of the Du Page leachate by Griffin and Shimp (1978). The three leaching
solvents were each extracted with hexane and analyzed for trace contamina-
tion of PCBs by using standard GC techniques and were found to be free
from contamination.
The mobilities of the two Aroclors and Dicamba were also measured on
silica-gel (GF-254, type 60, Brinkmann Instr., NY)TLC plates leached with
a variety of solvents. The solvents used were deionized H20; Du Page leach-
ate; carbon tetrachloride; benzene; acetone; 80:20 water:acetone; methanol;
15:85 water:methanol; and 9:91 water:methanol.
To confirm the data obtained from autoradiographs, zonal extractions
were done on several lanes spotted with PCB. A given lane was divided into
12 equal segments starting from 1.5 cm below the origin to 10.5 cm above
the origin. The soil in each segment was carefully transferred to a clean
graduated centrifuge tube to which 0.5 ml of 1:1 C2H5OH:H20 was added; 4 ml
of nanograde hexane was used as the extracting solvent. The ethanol mixed
with water at the soil surface and provided a more potar phase than the
hexane whereby the PCB in the soil could be more effectively extracted into
the hexane layer. The soil suspensions were sealed with aluminum foil,
mixed with a vortex mixer for a few seconds, and left to stand for an hour.
This procedure was repeated three times to allow complete extraction of
PCB from the soil, and a recovery of 99.7 ± 1.5 percent of PCB was obtained.
The soil suspension was then centrifuged at 2500 rpm for 5 minutes. The
final volume of hexane was recorded. One ml of the radioactive PCB-hexane
solution was pipetted into a scintillation vial containing 10 ml Dioxane
scintillation fluid (7 g PPO, 0.05 g POPOP; 120 g Naphthalene in 1 L
Dioxane), and the activity was counted for 10 minutes in a Packard Tri-Carb
Liquid Scintillation Spectrometer and the results corrected for quench. In
the case of nonradioactive PCBs (capicitor fluid), the hexane extract was
analyzed for PCB by standard GC techniques. The conditions for the gas
chromatographic analysis were as follows:
Instrument: Varian 2100 series Sc3H electron capture detector
Column: 6 ft glass column 2.5% OV-210 and 1.5% OV-17 on 60/80
Supelcoport G.C. Bondx
Injection port temperature: 220°C
Column temperature: 200°C
Detector temperature 250°C
Carrier gas: N2, flow 15-20 mL/min
Electrometer: range 10~lc amp/mv, attenuation 16
23
-------
MICROBIAL DEGRADATION
Isolation of Mixed Cultures
Three mixed cultures of microorganisms were isolated by biphenyl
enrichment from three different soil and sediments. To isolate the mixed
cultures, 5 g of soil or sediment and 0.1 g of biphenyl were placed in 100
mL of the following mineral media (Gray and Thornton, 1928):
KN03 1.0 g/L
MgS04 0.2 g/L
CaCl2 0.1 g/L
FeCl3 0.02 g/L
NaCl 0.1 g/L
K2HP04 1.0 g/L
The distilled water used to make up this media was saturated with Aroclor
1242.
The mixture was placed in a glass-stoppered 250 mL Erlenmeyer flask
that was wrapped completely with aluminum foil to keep out light and shaken
at room temperature (26°C). After 1 week, 5 mL of solution was placed in a
new flask containing 0.1 g of biphenyl and 100 mL of fresh media. This pro-
cedure was continued until the solution turned yellow. The yellow product,
the biphenyl metabolite described by Furukawa and Matsumura (1976), indi-
cated that biphenyl was being degraded. The solution was then transferred
every 3 days to fresh media.
Three different mixed cultures were isolated in this manner. The first,
the C-line, was isolated from dried Catlin soil. The second, the S-line,
was isolated from soil taken from an experimental farm plot at El wood,
Illinois. Activated sludge containing PCBs from a municipal waste treatment
plant at Stickney, Illinois, was applied to this plot for 7 years. The
third culture, the N-line, was isolated from Hudson River sediment from
the Fort Miller disposal site in New York (Leis et al., 1978).
ment contained about 2 ppm PCBs.
PCB Degradation Studies
The sedi-
Originally the PCB degradation study was run on the C-line for 15, 30,
and 45 days. Later studies were run on all three cultures for 5, 10, and
15 days. For these experiments, 70 mL of a 3-day old culture were added
to 1400 mL of mineral media made with Aroclor 1242 saturated water. The
solution was stirred rapidly. Immediately, six 100-mL aliquots were removed
and placed in 250-mL Erlenmeyer flasks, with ground glass stoppers, to
which 10 drops of concentrated HC1 had been added. The pH of each of these
control samples was about 2. Six more 100-mL aliquots were placed in
Erlenmeyer flasks as the test samples. All flasks were completely covered
with aluminum foil and shaken at 27°C ± 2°C. Two 80-mL aliquots of the
original solution were taken, acidified, and extracted as 0-day, or 0-hour,
blanks.
24
-------
After a set period of time (depending upon the experiment) 2 control and
2 test flasks were removed from the shaker. Next, 10 drops of concentrated
HC1 were added to the test flask to stop degradation and to maintain the
same pH for extraction for both the control and test samples. Eighty ml
of solution from each flask were extracted with 10 ml of hexane. The solu-
tions were extracted in a modified Erlenmeyer flask as shown in Figure 2.
The solution with hexane was stirred for 10 minutes; then the flask was
inverted and the layers were allowed to separate. The water layer was
drained off into a beaker, and the hexane was drained into a 25-mL brown
Wheaton bottle (Pierce, Rockford, IL) with a crimp top. The water was put
back into the extractor and 5 ml of hexane were used to rinse out the beaker.
The hexane was then poured into the extractor and the mixture was_stirred
for 5 minutes. The hexane was removed as before, and 5 ml of additional
hexane were used to extract the solution. The Wheaton bottle containing
the 20 ml of hexane was capped with an aluminum foil covering over a teflon
cap and was stored until analysis. This extraction procedure was carried
out for all samples.
Soil and Sludge Extractions
The soil and sludge from the experimental farm plot at Elwood, Illinois,
were extracted to determine their PCB content. The sludge was a composite
of sludge that had been applied to the soil about 3 months earlier. The
activated sludge had been heated and anaerobically digested at a municipal
waste treatment plant at Stickney, Illinois. The soil was a composite of
soil from 0 to 6 inches deep and was from a plot that had received a total
of 225 tons of dry solids per acre during the previous 7 years. This amounts
to about eight 1-inch liquid applications per year. Applications of sludge
were from June through August. The soil was tilled from 5 to 6 inches deep.
The sample soil was dried at 103°C and stored at room temperature prior
to extraction. The sludge was kept refrigerated.
To extract the soil, 50 g of finely ground soil were placed in a Soxhlet
extractor. About 150 ml of hexane distilled in glass were added to a 250-mL
round bottom flask, and the soil was extracted for 48 hours. The hexane-
cycle time was about 5 minutes. After extraction was complete, the extract
was treated to remove interfering organics by adding baked florisil (450°C
overnight) in increments to the round bottom flask until the solution turned
colorless. The hexane was then evaporated to approximately 0.5 ml in a
micro K-D evaporator and injected on the capillary column for PCB analysis.
About 150 ml of sludge were added to a modified Erlenmeyer extractor
(Figure 2) along with 50 ml of hexane. This solution was stirred for about_
5 minutes. The resulting solution was centrifuged (Sorvall table top centri-
fuge, Ivan Sorvall, Inc. Norwalk, CT) at 1,000 rpm for 5 minutes to break up
the sludge-hexane emulsion. The sludge was then extracted again with 50 ml
of hexane. Again the solution was centrifuged to break the emulsion. The
resulting extract, colored deep yellow, was cleaned by the addition of flor-
isil, as with the soil. The hexane then was evaporated to approximately 0.5
ml in a micro K-D evaporator and injected on the capillary column.
25
-------
Som'fication Study
To determine if the extraction procedure used to extract the PCBs from
the test solutions could extract all the PCBs that were in the microorganisms,
a 2-day degradation sample was sonified to disintegrate the microorganisms
and then was extracted. This sonified sample extract was compared to an
extract done as described previously.
The sonifier used was a Bronson Sonic Power (Danbury, CT) sonifier,
model number S75 with an exponential horn tip. The instrument was operated
at the 7 level at 6 to 7 amps for 30 minutes.
Prior to sonification, the tip and the 500-mL stainless steel beaker
were rinsed with methanol and hexane. After they dried, 240 ml of a 2-day
old S-line culture were placed in the beaker. The beaker was placed in an
ice bath and allowed to cool for 15 minutes. Then the sample was sonified
for 30 minutes. Finally, two 100-rnL aliquots were acidified with 10 drops of
concentrated HC1 each and were extracted by the usual procedure. In addition
to the sonified sample, two 100-mL aliquots of 2-day S-line culture were
acidified and extracted in the normal procedure and a 2-day blank was run
and extracted.
PCB VOLATILITY MEASUREMENTS
Experiments were set up to compare the changes in the volatility of
PCBs between aqueous solution of PCBs and PCB solutions containing different
amounts of soluble humic substances. The effect of different amounts of
soil in suspension with PCB-saturated water on the volatilization of PCBs
was also studied.
Humic acid standards were achieved by dissolving 100 mg and 1,000 mg of
commercial humic acid (Aldrich Chemical Co., Milwaukee, WI) in 1 liter of
distilled water and by adjusting the pH to 12 with NaOH pellets. The soil
used in the study was the Ap horizon of an agricultural soil, Cat!in silt
loam. The properties of the soil are listed in Tables 1 and 2.
50 mL PCB—saturated water
Mixed with 50 mL
pure water
Mixed with 50 mL water
containing 100 ppm or
1000 ppm humic acid,
equilibrated overnight
Purged with IM2 at room temperature, flow
rate 200 mL/min for 20 minutes, trap with
15 mL hexane in 3 stages
Mixed with 50 mL pure
water, add 0.1 g, 0.5 g
or 0.64 g of Catlin
soil, equilibrated
overnight
Hexane evaporated and analyzed by GC-FID
Figure 5. Flow diagram of experimental protocols used to study PCB volatilization from aqueous solutions
26
-------
TABLE 4. Gas chromatographic conditions used during analyses of RGBs for volatilization study.
Column: 183cm by 2 mm in diameter glass column packed with 3 percent OV-17 on 100/120-mesh Gas Chrom Q
Carrier: He,at 40 mL/min; H2:3Q mL/min;and air:250 mL/min
Injection port: 300°C
Detector: 300°C
Temperature programmed from 170°Cto 290 Cat 8 /min
Attenuator: 16 x 10 amps/mv —
The effects of humic acid and soil on volatilization of the PCBs were
investigated using the experimental protocols shown in Figure 5. Triplicates
of each of the samples described in Figure 5 were analyzed to obtain repre-
sentative results.
PCBs dissolved in water were purged at room temperature by a stream of
clean nitrogen, and the purged sample was trapped in a three-stage liquid
trap filled with 5 ml of hexane in each trap (Fig. 6). The nitrogen flow
rate was 200 mL/min. After 20 minutes, the hexane was pooled and evaporated
in a Kuderna-Danish (KD) evaporator to 0.4 ml, then was further evaporated
to dryness in a 1-mL vial by a stream of nitrogen. Five yL of hexane was
then pipetted into the vial, and 1 yL of this concentrate was injected onto
a gas-chromatographic column (183 cm by 2 mm I.D. glass column packed with
3 percent OV-17 on Gas Chrom Q) 100/120 mesh and was detected by a flame
ionization detector (FID) on a Hewlett Packard 5830A gas chromatograph
equipped with an automatic integrating recorder. The gas chromatographic
conditions are given in Table 4.
The use of an electron-capture detector (ECD) was investigated earlier.
The gas-purging process described was found to carry water vapor into the
hexane; the concentrate when injected resulted in an erratic response from
the ECD. An effort to dry the hexane layer by using salts was found to
interfere with quantisation of the PCBs. Although ECD is highly sensitive
to halogenated compounds and thus would normally be used for PCB analyses,
FID was used in these experiments. The sensitivity of the electrometer of
FID was 16 x 10~12 amps/mv, which was approaching its sensitivity limit
of 1 x 10~12 amps/mv.
The GC calibration was accomplished by repeating injections of 1 yL of
500 ppm Aroclor 1242 in hexane onto the GC column and evaluating the result-
ing chromatograms. The area integration of the chromatograms then provided
a reference for quantitation of the amounts of PCBs contained in the samples
from the hexane traps.
The trapping efficiency of the volatilized PCBs in the hexane traps
(Fig. 6) was evaluated by mixing 15 mL of hexane with 1 yL of 500 ppm Aro-
clor 1242. It was then equally distributed in the three liquid traps and
27
-------
stripped by using the identical purging
gas-flow rate and time. The trapping
efficiency was found to be 71 percent.
The purging efficiency was evalu-
ated by adding 1 ml of acetone solu-
tion containing 0.5 yg of Aroclor 1242
to a purging flask containing 100 ml of
water (Fig. 6). This solution was
purged, and the contents of the hexane
traps were analyzed for PCBs. The
resulting chromatogram was then used
to evaluate the purging efficiency.
ANALYTICAL METHODOLOGY
Gas-Liquid Chromatography
Instrumental Parameters
Apparatus included a Hewlett
Packard (HP) model 5830A (Avondale,
PA) gas chromatograph equipped with
a linearized 63Ni electron capture
detector and an automatic integrator.
A 183 cm x 2 mm I.D. glass column
packed with 3 percent OV-17 on 100/120
mesh Gas Chrom. Q (Analabs Industries,
North Haven, CT) was used for the
analysis of samples using the per-
chlorination procedure. Helium served
as a carrier gas at a flow rate of
60 mL/min under operating temperatures
for injector, column, and detector at
300°C, 230°C, and 300°C respectively.
100 ml saturated
PCB solution
Figure 6. Purging system using 1000-mL Erienmeyer flask
Samples from hexane extraction were analyzed in a 40-m wall-coated open
tubular (WCOT) glass capillary column. The liquid phase, Emulphor ON-870
(Applied Science Lab., State College, PA), was custom-coated on the glass
capillary columns using the techniques of Grob (1975) and Grob, Grob, and
Grob (1977). The Emulphor ON-870 gave better resolution of the various
isomers than other commercially available capillary-column coating materials.
Helium and 5-percent methane in argon were the carrier and make-up gases,
respectively, the linear velocity of the carrier gas being 45 cm/sec. The
samples were split at a ratio varying from 1:90 to 1:150. The operating
temperature for injector, column,- and detector were 190°C, 175°C, and 320°C.
All of the results were integrated and recorded on an automatic integrator.
Capillary Column Analysis
The capillary columns would sometimes become unstable with use and
therefore had to be recoated several times. Varying degrees of success
28
-------
ro
to
Figure 7. Water- soluble Aroclor 1 242 using a temperature-programmed column
-------
were obtained for each recoating. As a consequence, the resolution, reten-
tion times, and column temperatures varied slightly with different columns.
For the early work on the microbial degradation using the C-line, the
carrier gas was N2 with a linear velocity of 25 cm/sec. The column was
programmed from 120°C to 220°C at 20°/min. Figure 7 shows a chromatogram
of water soluble Aroclor 1242 using the above column. The number of theo-
retical plates (w 7) was calculated as follows (Kaiser, 1976): N -,=
I r> j-m jP&u. Is V&Q t,
5.54 (-g—=£--) where tm = unretained retention time (solvent), and 2>10 and bQ
are the 1/2 height widths at K = 10 on a plot of 1/2 height widths vs. K.
(K.=^-^ where tr = retention time of a certain peak). For a series of
similar compounds, such as PCB isomers, N - = 35,000 plates.
For the 5-, 10-, and 15-day microbial degradation studies using the C-,
N-, and S-lines the conditions were as follows: the carrier gas was He with
a linear velocity of 45 cm/sec. The column temperature was 192°C and the
split ratio was 1:150. ®Teal = 53,000 plates. Figure 8 shows a chromato-
gram of water-soluble Aroclor 1242 using this column.
For the microbial degradation kinetic study using the N-line, the con-
ditions were as follows: the carrier gas was N2 with a linear velocity of
45 cm/sec. The column temperature was 173°C and the split ratio was 1:90.
Nreal = 33,000 plates. Figure 9 shows a chromatogram of water-soluble
Aroclor 1242 using this column, which was also used in the water solubility
and adsorption studies.
Prior to PCB analysis using the column described above, a mass spectrum
of Aroclor 1016 was run using an Emulphor-coated WCOT capillary column.
Figure 10 shows the reconstructed chromatogram. The bottom line is for mass
number 188 or monochlorobiphenyl; the next .line, for mass number 222 or di-
chlorobiphenyl; the third line, for trichlorobiphenyl; and the second to the
top line, for tetrachlorobipheny1. The top line on the figure registers the
total reconstructed chromatogram.
Quantification and Identification
Samples analyzed using the capillary column were spiked with tribromo-
benzene as an internal standard. The quantitation measurements were carried
out by obtaining the electron capture detector response factor for both the
internal standard and four representative PCB isomers. The four PCB isomers
used were 4-mono;. 2, 4'-di-; 2, 5, 4'-tri-; and 2, 4, 2', 4'-tetra-
chl orobi phenyl . The concentration of stock solution of PCB isomers were
1,502 ppm, 99.98 ppm, 60.04 ppm, and 50.17 ppm, respectively. The concen-
tration of the stock solution of internal standard, tribromobenzene, was
24,900 ppm. A combination standard solution was prepared from these stock
30
-------
3 3 3+23 333 444344
Figure 8. Water-soluble Aroclor 1242 on second capillary column. Number under peak indicates the number of chlorines on the
isomer in each peak. Top number is the peak number. Peak number 1 is the internal standard.
solutions by adding TOO yL of each to 4 ml of hexane, which were subsequently
condensed to approximately 0.5 ml in a micro Kuderna-Danish (K-D) concen-
trator. Figure 4 shows a representative chromatogram of internal standard
and the four PCB isomers.
To identify each individual isomer, two approaches were adapted.
(1) Seven isomers (which included the four isomers used for quantitation and
the three selected isomers of known structure for additional retention-time
data) were injected into the GC. By comparing the retention time of each
isomer with the results from the mass spectrometer, the seven particular
31
-------
1 12 2 22 3 3332 133 3 333444 4 34444
2*3
4 44 4 4 4344 34 44
Figure 9. Water-soluble Aroclor 1242 on third capillary column
isomers could be identified. (2) By matching the relative position of the
other isomers to the isomer distribution pattern reported by Sissons and
Welti (1971) and Webb and McCall (1972), the major water-soluble PCB isomers
were identified.
Perchlorination Procedure for Analysis of PCB
The perch!orination of PCBs to decachlorobiphenyl (Armour, 1973) was
carried out in 10 mm x 150 mm hydrolysis tubes (Pierce Chemical Co., Rock-
ford, IL), which had been modified by adding a 15 mm x 150 mm test tube to
increase the volume (Fig. 11). Also, the cap of the reaction tube was modi-
fied so that the teflon insert could be easily removed for cleaning.
Thorough cleaning of the insert between uses was necessary in order to pre-
vent contamination of succeeding runs.
The teflon inserts were cleaned by boiling in a chromic acid solution
for two hours (the chromic acid solution was prepared by mixing 10 to 15 mg
of K2Cr207 in 15 mL of water and adding sulfuric acid until the resulting
32
-------
co
GO
18
24
32
30 31 33 34
Figure 10. A reconstructed chromatogram of Aroclor 1016obtained from mass-spectrometer
-------
red mass dissolved). The inserts were removed from the chromic acid solution
and rinsed successively with tap water, distilled water, methanol, and carbon
tetrachloride.
An aliquot of a known volume of a solution of carbon tetrachloride con-
taining PCBs was placed in a reaction vessel as described above. Antimony
pentachloride (0.5 to 1.0 ml) was added to the solution. Sodium chloride
crystals were added as a boiling aid. The reaction vessels were capped se-
curely to prevent losses of PCBs by volatilization. The reaction was carried
out at 200°C ± 5°C for 16 hours. Preliminary studies had shown that 16 hours
at 200°C were sufficient for complete perchlorination of the sample.
The reaction vessel was allowed to cool and the caps were loosened.
Then 4 to 5 drops of 6N HC1 were added slowly (due to the violent reaction).
To quench the reaction and to prevent precipitation of antimony oxychloride
(Berg, Diosady, and Rees, 1972), 1 ml of HC1 was added. To ensure that the
HC1 was well mixed, the caps were closed tightly and the vessels were shaken
for about 10 seconds. The caps were then opened and 1.0 ml of Mi rex (per-
chloropentacyclodecane, Alfred Bader Library of Rare Chemicals, a Division of
Aldrich, Inc., Milwaukee, WI) was added (at a concentration of 500 ppb in a
solution with benzene), followed by an additional 3 mL of benzene. The caps
were again closed tightly and the vessels shaken vigorously for 1 minute. To
remove any impurities and/or moisture (Armour, 1973), the upper benzene layer
was removed with a long-nosed transfer pipette to test tubes containing
Na2S04 and NaHC03. The benzene solu-
tion was transfered from the test
tube to a K-D evaporator, and NaCl
crystals were added as a boiling aid.
The sample was then condensed to 0.4
ml in a 100°C ± 0.5°C water bath.
About 4 ml of benzene were added and
condensation to 0.4 mL was repeated
a second time. The addition of 4 mL
of benzene and condensation to 0.4 mL
were repeated for a third and final
time. Then 4 mL of hexane were added
to aid in the GC analysis. A block
diagram of the procedure is shown in
Figure 12. Analysis of standard Aro-
clor 1242 solutions yielded a factor
of 0.737 for conversion of decachloro-
biphenyl (DCB) concentrations to Aro-
clor 1242 concentrations. Therefore,
all DCB concentrations obtained from
the perchlorinated samples were mul-
tiplied by 0.737 for conversion to
Aroclor 1242 concentration. The GC
instrumental parameters given pre-
viously were used for the analysis.
Figure 11.
Teflon- plugged reaction vessel used in per-
chlorination reaction
34
-------
Sample Extraction
Aqueous
Quantitatively extract
with CCI-4.
Pass CCI4 through
florisil column.
Elute with 2-4
bed volumes CCU.
Evaporate CCU
in K-D to ~-0.4 mL.
Fill to known
volume quantitatively.
Add aliquot of sample
quantitatively to reaction
vessel.
Add 0.5 to 1.0mLSbCI5.
Heat sample for
16to20hrsat200°C.
Cool. Add ~4mL
6N HCI. Mix.
Add quantitatively 1.00 mL
500 ppb Mirex in benzene.
1
Add ~4 mL benzene
and shake.
Transfer benzene to
tube containing NaHCO
and Na2SO4. Mix.
KD benzene to 0.5 mL
in boiling-water bath.
±
Repeat condensation twice
withbenzene(4 mL-»0.5 mL)
i
Adjust volume in
hexane to fit G.C.
conditions.
Figure 12. Schematic block diagram of the perchlorination procedure for the analysis of PCB from aqueous and solid samples.
35
-------
SECTION 5
SOLUBILITY OF PCBs IN WATER1
RESULTS AND DISCUSSION
Quantitatlon and Identification
A PCB mixture contains a variety of isomers. Sissons and Welti (1971)
have reported that for Aroclor 1242, 1254, and 1260, the possible number of
isomers were 45, 69, and 78. Because it was not possible to obtain all the
individual isomers for quantisation and identification of the PCBs analyzed
in this study, only four isomers were used for quantisation. An assumption
was made that all of the mono-chlorobiphenyl isomers gave the same electron
capture detector response and likewise, the di-, tri-, and tetra-
chlorobiphenyls. Table 5 lists the relative response of an electron capture
detector (63Ni) to the internal standard and several PCB isomers. Although
the responses of the electron capture detector to each isomer are not exactly
the same, they are in the same range. The isomers with five or more chlorine
substitutions were quantified by the response of the electron capture detec-
tor to tetrachlorobiphenyl (2, 4, 2', 4'-).
To confirm the above
assumption for quantita-
tive analysis, two stand-
ard solutions were checked.
A standard containing 40.6
mg of Aroclor 1242 in 10
mL of hexane solution was
prepared. A 10-yL volume
of solution and 100 yL of
internal standard (tribro-
mobenzene) were injected
into 4 mL of hexane,
and then evaporated in a
micro K-D condenser to
approximately 0.5 mL.
Three yL of the concen-
trate were injected into
Authors: H. C. Lee,
E.S.K. Chian, and
R.A. Griffin
TABLES. The relative response of a ^Ni electron
capture detector to tribromobenzene and
several PCB isomers.
Chemical
Relative response (/Jg/area)
Internal standard
(tribromobenzene)
5.3234 x 10 '
3.3772 x 10 :
2.2853x10
1.9866x 10
1.0879x10
9.6283 x 10'
9.0012 x 10'
36
-------
IS
10 18
Aroclor 1242
35
Capacitor Fluid
B
Figure 13. A comparison of hexane-soluble Aroclor 1242 and used capacitor fluid: (A) Aroclor 1242, (B) used capacitor fluid
37
-------
TABLE 6. The quantitative composition and percentage of contribution of the
individual water-soluble isomers in Aroclors and capacitor fluid.
Peak
no.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
Total
No. of
Cl
1
1
1
2
2
2
2
2
3
3
3
3
3
2
2&3
3
3
3
3
4
4
4
3&4
4
4
4
4
4
4
4
3
4
4
3
4
5
5
5
5
5
5
5
5
6
Identi-
fication
2
3
4
2,2'
2.4
2.3'
2,4'
2,6,2'
2,5,2'
2,4,2'
3,4'
4,4';2.3,2'
2,4,3'
2,4,4'
2,5,4'
3,4,2'
2,5,2',5'
2,4,2',5'
2,4,2',4'
2,3,2',5'
2,5,3',4'
2,4,3',4'
AR 1016a
(W L"1) (%)
84.78
13.31
12.74
19.71
18.80
25.94
38.62
171.78
10.29
87.03
2.72
5.78
4.77
49.38
33.52
16.62
110.76
5.43
35.87
5.48
6.95
31.83
27.34
9.18
2.39
19.77
8.16
5.26
14.91
5.27,
3.03
2.12
1.85
14.16
905.55
9.36
1.47
1.47
2.18
2.08
2.86
4.26
18.97
1.14
9.61
0.30
0.64
0.53
5.45
3.70
1.84
12.23
0.60
3.96
0.61
0.77
3.51
3.02
1.01
0.26
2.18
0.90
0.58
1.65
0.58
0.33
0.23
0.20
1.56
AR 1221a
(Ha LM) (%)
2572.28
45.75
622.64
13.48
23.50
25.57
33.70
109.59
5.26
1.27
1.87
0.71
4.13
9.07
1.94
1.71
4.73
0.74
1.66
0.64
0.78
1.32
1.08
3.61
0.37
1.48
0.45
3516.33
73.15
1.30
17.71
0.38
0.67
1.50
0.96
3.12
0.15
0.04
0.05
0.02
0.12
0.26
0.06
0.05
0.13
0.02
0.05
0.02
0.02
0.04
0.03
0.10
0.01
0.04
0.01
AR 1242b
(/Ug L"1) (%)
121.59
15.07
9.06
21.18
23.76
30.60
138.93
8.33
61.35
2.15
0.48
4.50
2.98
35.30
27.47
9.87
65.07
2.59
22.25
3.59
4.09
19.51
15.88
3.37
1.09
12.83
3.13
2.50
7.21
4.70
2.70
2.63
0.82
9.79
0.99
2.47
2.84
702.67
17.30
2.14
1.29
3.01
3.38
4.35
19.77
1.19
8.73
0.31
0.07
0.64
0.42
5.02
3.91
1.40
9.26
0.37
3.17
0.51
0.58
2.78
2.26
0.48
0.16
1.83
0.45
0.36
1.03
0.67
0.39
0.38
0.12
1.39
0.14
0.35
0.40
AR 1254b
(AO L ') (%)
0.37
1.09
0.28
4.57
2.87
0.23
1.09
2.85
2.33
14.25
0.55
6.46
3.07
2.67
0.29
1.78
2.24
1.72
6.20
0.47
1.48
4.46
0.97
0.80
5.19
1.56
69.84
0.53
1.56
0.40
6.54
4.11
0.33
1.56
4.08
3.34
20.40
0.79
9.25
4.40
3.82
0.42
2.55
3.21
2.46
8.88
0.67
2.12
6.39
1.39
1.15
7.43
2.23
Capacitor fluid3
-------
the gas-liquid chromatograph. A second standard solution containing capaci-
tor fluid was prepared by adding 28 mg of capacitor fluid to 10 ml of hexane.
The same procedure as with the Aroclor 1242 standard was repeated. The
results of the analysis showed that 42.77 mg of Aroclor 1242 were quanti-
fied as compared to 40.60 mg of Aroclor 1242 actually present in the sample;
the percentage of error was 5.34 percent. Versus 28 mg of capacitor fluid
added to the sample, 30.57 mg of capacitor fluid were determined; the per-
centage of error was 9.18 percent. In both cases, the error was less than
10 percent. The results show that, considering the errors normally asso-
ciated with PCB analyses, the quantitative method developed in this study
gave satisfactory results. By using an internal standard and four PCB
isomers, the determinations of concentrations of individual PCB isomers in
a mixture of isomers were relatively accurate.
GC chromatograms of both hexane-soluble Aroclor 1242 and capacitor
fluid are shown in Figure 13; both have the same isomer distribution pattern.
This indicates that the Aroclor 1242 that was originally impregnated into
the capacitor had not undergone any significant changes during 10 years of
use or during the "burn out" of the capacitor.
The identification of the water-soluble fractions of several Aroclors
and capacitor fluid are shown in Table 6. Webb and McCall (1972) have also
reported data on isomer distributions; peak 3 (4-mono-chlorobiphenyl) was
not reported by them for Aroclor 1242; and peak 10 (2,5,2'-tri-chlorobiphenyl),
peak 18 (2,5,4'-tri-chlorobiphenyl), and peak 19 (3,4,2'-tri-chlorobiphenyl),
were not reported for Aroclor 1254. In addition, peak 22 and peak 23 (a com-
bination of a tri-chlorobiphenyl and 2,4,2', 5'-tetrarchlorobiphenyl) were
reported by Webb and McCall (1972) but were not found in Aroclor 1254 in
this study. The explanation for this apparent discrepancy seems to be in
the higher resolution of the Emulphor-coated capillary column used in this
study, differences in quantisation techniques, and/or possibly differences
in batches of Aroclors (EHRC, 1976; Lloyd et al., 1976).
The Time Dependence of PCB Dissolution in Water
One of the major difficulties in determining the aqueous solubility of
Aroclors is the length of time necessary to reach a state of equilibrium
between the Aroclors and water. Figure 14 shows that about 5 months were
required to reach a saturation point for both Aroclor 1242 and capacitor
fluid. Haque, Schmedding, and Freed (1974) reported that the equilibration
of Aroclor 1254 in water occurred in about 2 months. During the initial
solubilization of about one week (Table 7), the percentages of tri- and
tetra-chlorobiphenyl were high (about 61 percent and 32 percent, respect-
ively); however, after the solution approached equilibrium, the percentages
changed. The values for mono-, di-, tri-, and tetra-chlorobiphenyl became
20 percent, 35 percent, 31 percent, and 15 percent, respectively. That is,
the composition of water-soluble Aroclor 1242 changed during the solubili-
zation process and was also no longer the same as that of the original
Aroclor 1242 fluid. Higher portions of low-chlorinated isomers predominated
in the water-soluble Aroclor 1242.
39
-------
A comparison of the solubility of PCB isomers in Aroclor 1242 to the
individual isomers (Table 8) clearly indicates that the solubility of indi-
vidual isomers was lower in the Aroclor mixture than when measured alone.
It also appeared that the presence of other chlorinated isomers impaired
the dissolution of the lower-chlorinated isomers in the mixture, but, on
the contrary, the dissolution of the higher chlorinated isomers did not seem
to be affected as greatly by the presence of lower chlorinated isomers.
Table 9 shows quantitative values of the water solubility of the indi-
vidual isomers with respect to time. The isomers represented by peaks 8, 10,
17, 18, 20, and 24 dissolved rapidly in the water. The results shown for
Aroclor 1242 also apply to capacitor fluid.
Solubility of Aroclors and Capacitor Fluid
Representative GC chromatograms of the water-soluble PCBs, that is,
Aroclor 1016, 1221, 1242, and 1254, and capacitor fluid used in this study,
are shown in Figure 15. The quantitative composition and percentages of
contribution of the individual isomers (Table 6) show that in Aroclor 1016
peak numbers 1, 8, 18, and 20 were the major peaks of mono-, di-, tri-,
and tetra-chlorobiphenyl. They represented 76, 56, 35, and 21 percent of
the individual peaks, respectively. In Aroclor 1221 the four major peaks
1,000
-i 1 1 1 1 1 1 1 r
0 20 40 60 80 100 120 140 160 180 200 220 240
Number of days
Figure 14. The solubility of Aroclor (AR) 1242 and used capacitor fluid (CF) 1242 in water as a function of time
40
-------
TABLE 7. The percentage of PCB isomers in water-soluble Aroclor 1242.
Number of days
Number of Cl
1
2
3
4
7
(%)
a
6
61
32
20
a
29
58
13
30
a
44
47
9
50
a
45
45
11
149
20
34
31
17
251
20
35
31
15
Hexane-soluble
a
12
46
42
aNo data available because of poor resolution between IS and monochlorinated biphenyl.
of mono-, di-, tri-, and tetra-chlorobiphenyl were represented by peak
numbers 1, 8, 10, and 27, and the corresponding percentages were 79, 45,
22, and 34 percent. In Aroclor 1242, peak numbers 1, 8, 18, and 20 were
the major peaks of mono-, di-, tri-, and tetra-chlorobiphenyl. The per-
centages of these peaks were 89, 57, 30, and 22 percent, respectively. In
Aroclor 1254 the major peaks of tri-, tetra-, and penta-chlorobiphenyls were
represented by peaks numbered 34, 35, and 45; the respective percentages
were 57, 55, and 27 percent. In capacitor fluid, the four major peaks
were represented by peak numbers 1, 8, 18, and 20 for mono-, di-, tri-, and
tetra-chlorobiphenyls, respectively; the respective percentages were 88, 54,
31, and 22 percent.
The overall compositions of mono-, di-, tri-, tetra-, penta-, and
hepta-chlorobiphenyl isomers in the water-soluble fraction of these Aroclors
and capacitor fluid are listed
in Table 10. In comparing the
composition of the water-
soluble Aroclors to the orig-
inal Aroclors, such as
reported by Thurston (1971)
and measured in this study,
the water-soluble PCBs were
found to be richer in the
lower-chlorinated isomers
than were the original
PCBs. Aroclor 1254 is
reported to contain isom-
ers from tetra- to hepta-
chlorobiphenyl in the
original fluid. The GC
traces of hexane-soluble
Aroclor 1254 (for example,
Stalling and Huckins, 1971,
and Sissons and Welti, 1971)
TABLE 8. A comparison of the aqueous solubility of some PCB
isomers contained in Aroclor 1242 to the pure
individual isomers.
Solubility
Isomer
2,2'
2,5,2'
2,5,2',5'
In mixture
(AOL'1)
15.07
21.18
138.93
61.35
22.25
Individual
(PS L"1)
4003
900a
637b
248b
26.5b
aReported by Hoover (1971).
Reported by Haque and Schmedding (1975).
41
-------
show some small peaks in the position of trichlorobiphenyl, as has also been
confirmed in this study. The trichlorobiphenyls are peaks 10, 18, and 19,
shown in Figure 15e. Because trichlorobiphenyls are more water soluble
than the more highly chlorinated biphenyls, detection of trace amounts in
water is reasonable. The hexa- and hepta-chlorobiphenyls are enriched in
the Aroclor 1254 fluid and do not dissolve readily in water. The low solu-
bility of these highly chlorinated isomers in water is apparently the reason
for this situation.
TABLE 9.
A quantitative comparison of water-soluble isomers in Aroclor 1242 with respect to time
Concentration
Peak
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Total
No.
of Cl
1
1
1
2
2
2
2
2
3
3
3
3
3
2
2&3
3
3
3
3
4
4
4
3&4
4
4
4
4
4
4
4
3
4
4
3
4
5
5
5
5
5
Identi- 7 days 20 days
fication (;Ug L"1) (jUg L."1)
2
3
4
2,2'
2,3'
2,4' 2.90 27.30
2,6,2'
2,5,2' 5.52 22.97
2,4,2'
-
3,4'
4,4';2,3,2' - 8.83
2,4,3'
2,4,4' 2.55
2,5,4' 20.22 35.15
3,4,2'
2,5,2',5' 1.58 6.67
2,4,2',5'
2,4,2',4' 11.12 7.39
2,3,2',5'
2,5,3',4'
2,4,3',4' 1 .96
0.71
46.56 108.31
30 days
1.00
3.18
9.10
71.18
1.40
41.05
-
16.25
4.08
43.28
8.94
0.55
8.93
208.94
50 days
1.69
1.78
2.93
14.07
78.45
1.72
41.20
1.71
17.19
4.63
48.20
8.78
0.64
0.49
2.11
9.12
0.69
0.74
1.42
0.59
2.18
240.33
149 days
124.60
7.50
5.83
20.80
23.24
19.63
141.85
8.13
62.48
10.00
26.15
29.83
9.83
63.76
2.59
24.05
2.66
3.78
17.95
19.85
4.15
1.11
12.75
4.10
9.15
4.53
4.10
5.43
0.73
13.83
2.53
4.79
671.86
251 days
121.59
15.07
9.06
21.18
23.76
30.60
138.93
8.33
61.35
2.15
0.48
4.50
2.98
35.30
27.47
9.87
65.07
2.59
22.25
3.59
4.09
19.51
15.88
3.37
1.09
12.83
3.13
2.50
7.21
4.70
2.70
2.63
0.82
11.62
0.99
2.47
2.84
702.67
(-) Indicates peaks too small to quantify.
42
-------
(a) Aroclor 1016
IS 8 10 IS 1820
IS I 35678
J
II Ifl
10
(b) Aroclor 1221
35
Figure 15. GC chromatograms of water-soluble Aroclors and
used capacitor fluid.
43
-------
(c) Aroclor 1242
(d) Capacitor Fluid
(e) Aroclor 1254
8 , 34I37.38 _40 42 43
51 52 53
Figure 15. Continued
44
-------
The comparison of Aroclor 1242 and capacitor fluid as shown in Figure
15c and d indicated that the distribution of isomers was identical, but
actually the composition was slightly different (Tables 6 and 10). The
concentrations of mono-, and di-chlorobiphenyl were slightly higher in
capacitor fluid; however, each batch of Aroclor may differ slightly in isomer
composition. On this basis, the Aroclor 1242 and capacitor fluid were con-
sidered to be essentially identical. Aroclor 1016 has a composition similar
to Aroclor 1242 except for isomers with 5 or more chlorines. The distribu-
tion of isomers in water-soluble Aroclor 1016 and Aroclor 1242 indicated that
peaks 37 to 40 (pentachlorobiphenyls) were absent in the chromatogram of
Aroclor 1016 (Figure 15a and c). This observation was consistent with the
original compositions of Aroclor 1016 and 1242. Aroclor 1221 (21 percent
chlorine content) and Aroclor 1254 (54 percent chlorine content) were dif-
ferent in their isomer distribution patterns in water, as shown in Figure
15b and e. The lower-chlorinated isomers were predominant in Aroclor 1221,
whereas higher-chlorinated isomers were predominant in Aroclor 1254.
The solubility of the PCBs and capacitor fluid in water is summarized
in Table 11. It should be noted that the data of MacKay and Wolkoff (1973)
were calculated values and that the solubility of Aroclor 1016 reported by
Tucker, Litschgi, and Mees (1975) was estimated from the solubility of Aro-
clor 1242 on the assumption that Aroclor 1016 contained more of the lower-
chlorinated isomers than did Aroclor 1242. The aqueous solubility of Aro-
clors 1016, 1242, and 1254 measured in this study was higher than those
reported in the literature. The discrepancy in the solubility data for
Aroclor 1254 may be attributed to the quantisation method employed in this
study. The response of the electron capture detector to tetrachlorobiphenyl
was used to estimate the values of the water-soluble fractions of Aroclor
1254, which consisted of a high proportion of tetra- and penta-chlorobiphenyls.
The response factor used for the penta-isomers may have yielded a high value
for the overall solubility of Aroclor 1254.
TABLE 10. Summary of isomer composition of water-soluble Aroclors and capacitor fluid.
Number of Cl
1
2
3
4
5
6
7
AR 1221
(%)
92
7
1
a
0
0
0
AR 1016
12
34
35
19
0
0
0
AR 1 242
19
35
31
14
1
0
0
Capacitor fluid
18
32
36
13
1
0
0
AR 1254
0
0
20
56
24
-
-
(-) indicates trace amount.
45
-------
TABLE 11. A summary of the solubility of PCBs and capacitor fluid in water.
AR1016 AR1221 AR 1242 AR 1248 AR 1254 AR 1260
(jUgL"1) (mgL"1) (& L1) (fjg L'1) (VsL1) fa L1}
906 3.5 703 ~70
420 ± 80 340 ± 60
45 ±10
220-250 200
~56
240 54 12 2.7
200 100 50 ~25
43
41
5.0 mg L"1 0.3-3 mg L'1
in fresh in fresh
water water
3.8 mgL"1 0.3-1. 5 mg L"1
in salt in salt
water water
Capacitor
fluid Temperature
(/JQ L"1) (°C)
698 room
<23±2°C)
room
(25±1°C)
room
room
room
25°C
20°C
26°C
4°C
room
Authors
This study
(1978)
Paris et al.
(1978)
Lawrence &
Tosine (1976)
Tucker et al.
(1975)
Haque et al.
(1974)
Mackay &
Wolkoff (1973)
Nisbet &
Sarofin (1972)
Nelson (1972)
Zitko (1970,1971)
-------
The solubility of Aroclor 1016 and 1242 reported in this study also
appears to be higher than that reported by other investigators. This appears
to be due to differences in experimental and quantisation methods and pos-
sibly is due to differences in "batches" of Aroclors. A possible explana-
tion for the lower values reported by earlier workers (Nisbet and Sarofin,
1972; MacKay and Wolkoff, 1973; and Paris, Steen, and Baughman, 1978) may be
a result of the lack of information on the periods of time needed to reach
equilibrium in water. This study shows that 5 months were necessary for
Aroclor 1242 to reach equilibrium in aqueous solution, and Haque, Schmedding,
and Freed (1974) have shown that 2 months were necessary for Aroclor 1254.
Whereas, Paris, Steen, and Baughman (1978) reported that saturation of Aro-
clor 1016 and 1242 in water occurred in less than 1 week. The solubility
of Aroclor 1221 obtained in this study was close to that reported by Zitko
(1970). Because the solubilities of capacitor fluid have not been reported
previously, no comparison can be made. Therefore, caution should be used
when interpreting solubility data reported for Aroclors; the effect of the
time-dependence of the solubilization should be considered.
47
-------
SECTION 6
ADSORPTION OF WATER SOLUBLE PCBs BY SOIL
MATERIALS AND COAL CHARS1
RESULTS AND DISCUSSION
The data for PCB adsorption by the various earth materials used in this
study are shown in Figure 16 as the amount of PCB adsorbed in relation to the
equilibrium PCB concentration. Similar results for the low-temperature ashed
samples are shown in Figure 17.
Several results are evident from the data presented in Figures 16 and
17. The first result was that, over the concentration range studied, PCB
adsorption by all the earth materials could be described by the linear
relationship:
x/m = KC, (1)
where x/m = the amount of PCB adsorbed per unit weight of adsorbent (yg/g),
K = an adsorption constant OnL/g),
and c = the equilibrium PCB concentration (yg/mL).
The PCB adsorption constant (K) for each adsorbent was obtained from the
slope of the line and is shown in Figure 16 and 17. The Freundlich adsorp-
tion constant (KF) can be obtained from the foil owing relation:
x/m =
Where.£^(yg/g) and n (mL/g) are constants and the x/m and C are as defined
in equation (1). It is evident that the PCB adsorption data reported
here are a special case of the Freundlich equation where n = 1. For the
case n = 1, KF and K are identical, thus Kp values reported in the litera-
ture for other compounds in units of yg/g can be compared directly with the
K values reported here.
Another view of the PCB adsorption constant (K) is as a partition coef-
ficient. The parameter K is the ratio of the solid phase PCB (x/m} to the
solution phase PCB (c}. That is, the constant K is the partition coefficient
between adsorbed and solution phases of PCB in the experimental system used
in this study. In either view, a simple linear relation (equation [1])
described the adsorption of PCB from aqueous solution and yielded an adsorp-
1 Authors: M. C. Lee, R. A. Griffin, M. L. Miller, and E.S.K. Chian
48
-------
400 -
300-
-a
<
200-
100-
0.1
0.2 0.3 0.4
PCB Concentration (ppm)
0.5
0.6
ISGS 1979
Figure 16. PCB adsorption by earth materials. Solid symbols represent data for capacitor fluid; open symbols represent data
for Aroclor 1242;Xs in open symbols indicate analysis using capillary column.
49
-------
Amount PCB Adsorbed (jug/g)
en
O
-------
78 10
13
B
ISGS 1979
Figure 18. Capillary column analysis of Aroclor 1242 remaining in solution: (A) after adsorption on reaction bottle (blank);
and (B) after adsorption by medium-temperature char
51
-------
tion constant (K) unique to each adsorbent. Scharpenseel, Theng, and Stephen
(1978) also found a linear relation for the adsorption of 2, 3, 4-
trichlorobiphenyl from n-hexane solution by montmorillonite and humic acid.
A second result shown in Figure 16 is that there were no significant
differences between adsorption of Aroclor 1242 and a used capacitor fluid
by any of the five adsorbents. This result is in agreement with results
reported previously for the solubilities of fresh Aroclor 1242 and used
Aroclor 1242 capacitor fluid (Lee et al., 1978). No significant differences
in the aqueous solubility or the distribution of isomers were found; there-
fore, no differences in adsorption would be expected. This result is signi-
ficant because it shows that laboratory studies carried out with fresh Aro-
clor 1242 can be extrapolated to the field, where aged fluids that have
gone through a "burn out" process are found, with a greater degree of con-
fidence than previously was possible.
A third result illustrated in Figures 16 and 17 is the wide differences
in adsorption of PCBs by the various earth materials. Adsorption for the
five earth materials followed the series
MTC > HTC > CS > MC > OS.
Low-temperature ashing reduced the amounts of PCBs adsorbed by all samples.
For the series of Catlin soil (CS) samples, the PCB adsorption decreased as
the ashing time increased; this suggests a relationship between organic car-
bon content and adsorption of PCBs.
Capillary Column Analysis of Adsorption
A comparison of the adsorption of isomers in the water-soluble Aroclor
1242 solution by several different adsorbents was investigated by applica-
tion of capillary column GC analysis. Figure 18 gives representative gas
chromatograms of water-soluble Aroclor 1242 remaining in solution after
adsorption by the reaction bottle (blank) and by medium-temperature char.
The quantitative composition and percentage of individual PCB isomers
remaining in solution after adsorption by several adsorbents is shown in
Table 12. The total adsorption found by the summation of the individual
isomers is plotted in Figure 16. The figure illustrates the close accord
between the amounts of PCBs adsorbed in the two experiments. Other results
in Table 12 indicate that 31 percent of the water-soluble Aroclor 1242 was
adsorbed on the glass surface of the Wheaton reaction bottles. This result
is in agreement with that of Gresshoff, Mahanty, and Gortner (1977) who
also reported PCB adsorption on glassware. Zitko (1971) reported that water-
soluble PCBs were strongly adsorbed by plastic surfaces such as polethylene
sheets and tubes. Eichelberger (1971) also observed that 60 percent of
Aroclors, ranging from 1232 to 1268, in water samples were adsorbed on the
walls of the container or the silt contained in the river water. The results
of this study also confirm the strong affinity of water-soluble PCBs for
the surfaces of reaction and storage vessels. The glass surface preferen-
tially adsorbed peaks 1, 4, 31, and 33 as shown in Table 12. This preferen-
tial adsorption of individual isomers also depends on the characteristics
of the adsorbent. With high temperature char, peaks 3, 14, 31, and 34
52
-------
TABLE 1 2. Quantitative distribution of PCB isomers after adsorption by several adsorbents as analyzed by
capillary column GC analysis.
Peak
no.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Total
No. of
Cl
1
1
1
2
2
2
2
2
3
3
3
3
3
2
2&3
3
3
3
3
4
4
4
3&4
4
4
4
4
4
4
4
3
4
4
3
4
5
5
5
5
5
Identi-
fication
2-
3-
4-
2,2'-
2,4-
2,3'-
2,4'-
2,6,2'-
2,5,2'-
2,4,2'-
3,4'-
4,4'-&
2,3,2'-
2,4,3'-
2,4,4'-
2,5,4'-
3,4,2'-
2,5,2',5'-
2,4,2',5'-
2,4,2',4'-
2,3,2',5'-
2,5,3',4'-
2,4,3',4'-
AR1242
(ppb)
121.59
15.07
9.06
21.18
23.76
30.60
138.93
8.33
61.35
2.15
0.48
4.50
2.98
35.30
27.47
9.87
65.07
2.59
22.25
3.59
4.09
19.51
15.88
3.37
1.09
12.83
3.13
2.50
7.21
4.70
2.70
2.63
0.82
9.79
0.99
2.47
2.84
702.67
Blank
(ppb) % remaining
47.67
9.03
3.76
18.04
20.62
24.85
105.58
6.70
49.58
1.38
4.00
2.37
27.32
17.94
8.47
50.19
2.28
18.32
2.76
3.72
14.41
9.26
1.99
0.99
7.21
2.28
2.06
5.45
2.12
1.97
0.73
0.64
7.09
0.59
1.76
1.92
485.05
39.21
59.92
41.50
85.17
86.78
81.21
76.00
80.43
80.81
64.19
88.89
79.53
77.39
65.31
85.82
77.13
88.03
82.34
76.88
90.95
73.86
58.31
53.05
90.83
56.20
72.84
82.40
75.59
45.11
72.96
27.76
78.05
72.42
59.60
72.26
67.61
69.03
HTC (0.1987g)
(ppb) % remaining
21.47
1.14
1.49
8.30
5.25
7.25
31.63
3.16
15.67
0.21
1.03
0.16
5.78
3.47
1.05
6.82
0.55
2.32
0.25
0.58
1.47
0.21
0.21
0.14
0.89
0.25
0.27
0.51
0
0.08
0
0.19
121.80
45.04
12.62
39.63
46.01
25.46
29.18
29.96
47.16
31.61
15.22
25.75
6.75
21.16
19.34
12.40
13.59
24.12
12.66
9.06
15.59
10.20
2.27
10.55
14.14
12.34
10.96
13.11
9.36
0
4.06
0
2.68
25.11
MTC (0.2081 g)
Q
(ppb) % remaining
5.82
0.13
0.42
4.94
2.15
3.56
13.78
2.39
8.86
0.05
0.50
0.14
3.38
2.16
0.55
3.19
0.28
1.22
0.14
0.40
0.64
0.90
0.15
0.08
0.67
0.21
0.15
0.32
0
0.06
0
0.22
0.06
0.04
61.05
12.21
1.44
11.17
27.38
10.43
14.33
13.05
35.67
17.87
3.62
12.50
5.91
12.37
12.04
6.49
6.36
12.28
6.66
5.07
10.75
4.44
9.72
7.54
8.08
9.29
9.21
7.28
5.87
0
3.05
0
3.10
3.41
2.08
12.59
% remaining compared with AR1242.
% remaining compared with blank.
were the predominant mono-, di-, tri-, and tetra-chlorobiphenyl peaks being
adsorbed. The percentages remaining, compared with the blank (glass surface),
were 13, 7, 0, and 0, respectively. With medium temperature char, peaks 3,
14, 31, and 34 were the predominant mono-, di-, tri-, and tetra-chlorobiphenyl
peaks being adsorbed. They represent 1, 6, 0, and 0 percent remaining, re-
spectively. On the other hand, no preferential adsorption of particular
isomers was apparent by either Catlin soil or sand, but both preferentially
adsorbed higher chlorinated isomers as a class over lower chlorinated isomers.
53
-------
TABLE 12. Continued.
Peak
no.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
Total
No. of
Cl
1
1
1
2
2
2
2
2
3
3
3
3
3
2
2&3
3
3
3
3
4
4
4
3&4
4
4
4
4
4
4
4
3
4
4
3
4
5
5
5
5
5
Identi-
fication
2-
3-
4-
2,2'-
2,4'
2,3'-
2,4'-
2,6,2'-
2,5,2'-
2,4,2'-
3,4'-
4,4'-&
2,3,2'-
2,4,3'-
2,4,4'-
2,5,4'-
3,4,2'-
2,5,2',5'-
2,4,2',5'-
2,4,2',4'-
2,3.2',5'-
2,5,3',4'-
2,4,3',4'-
HTC (0.0492g)
(ppb) %
15.83
4.36
1.13
9.45
11.36
15.55
66.77
4.43
32.09
0.41
2.37
0.92
15.13
5.24
4.34
25.99
1.34
8.61
0.95
1.57
5.63
4.98
1.12
0.37
2.96
0.44
0.94
2.16
0.28
0.81
0.27
2.53
0.13
0.61
0.58
254.8
remaining
33.21
48.28
30.05
52.38
55.09
62.58
63.24
66.12
64.72
29.71
59.25
38.82
55.38
29.21
51.24
51.78
58.77
47.00
34.42
42.20
39.07
53.74
56.28
37.37
41.05
41.23
45.63
39.63
13.21
41.12
42.19
35.68
22.03
34.66
30.21
52.53
MTC (0.051 1g)
(ppb) %
45.17
1.87
2.94
10.77
6.95
8.45
34.99
3.16
15.34
0.38
1.01
0.24
5.94
6.40
1.36
7.88
0.48
3.11
0.44
0.75
2.23
1.59
0.20
0.11
1.21
0.34
0.29
0.71
0.07
0.13
0.22
0.05
0.61
0.05
165.44
remaining
94.76
20.71
78.19
59.70
33.71
34.00
33.14
47.16
30.94
27.54
25.25
10.13
21.74
35.67
16.06
15.70
21.05
16.98
15.94
20.16
15.48
17.17
10.05
11.11
16.78
14.91
14.08
13.03
3.30
6.60
30.14
7.81
8.60
2.60
34.11
CS (0.0546g)
(ppb) %
45.88
6.88
3.38
13.90
14.72
17.03
73.24
4.71
31.85
0.91
2.46
1.36
16.70
9.63
4.84
28.86
1.27
10.11
1.42
1.84
7.67
5.14
1.24
0.46
3.08
0.96
1.01
2.63
0.83
1.11
0.31
3.29
0.39
0.69
0.74
320.54
remaining
46.25
76.19
89.89
77.05
71.39
68.53
69.37
70.30
64.24
65.94
61.50
57.38
61.13
53.68
57.14
57.50
55.70
55.19
51.45
49.46
53.23
55.51
62.31
46.46
42.72
42.11
49.03
48.26
39.15
56.35
48.44
46.40
66.10
39.20
38.54
66.08
Sand (0.0521 a)
(ppb) %
85.31
12.06
6.09
18.85
19.70
21.96
34.44
6.14
41.79
1.59
3.40
2.17
23.17
17.66
7.15
41.69
1.86
15.40
2.49
3.07
12.67
7.70
1.68
0.76
5.88
1.86
1.66
4.48
2.53
1.63
1.45
0.55
5.93
0.57
1.45
1.66
478.45
remaining
100
100
100
100
95.54
88.37
89.45
91.64
84.29
100
85.0
91.56
84.81
58.44
84.42
83.06
81.58
84.06
90.22
82.53
87.93
83.15
84.42
76.77
81.55
81.58
80.58
82.20
100
82.74
100
85.94
83.64
96.61
82.39
86.46
98.64
|*% remaining compared with AR1242.
% remaining compared with blank.
In general, the higher chlorinated isomers were more adsorbable than
the lower chlorinated isomers. Haque and Schmedding (1976) indicated the
extent of adsorption for the surfaces they studied followed the sequence
hexachloro- > tetrachloro- > dichloro-biphenyl for the isomers chosen. This
preferential adsorption also depends on the position of the chlorine atom
on the biphenyl ring; some particular isomers appeared to be more adsorbable
than others in this study. They were peaks 1, 3, 4, 14, 31, and 33(Table 12),
54
-------
Peaks 1, 3, and 14 were specifically identified as 2- and 4- monochloro-
biphenyl, and peak 14 as 3, 4'-dichlorobiphenyl. Peaks 4, 31, and 33 were
identified only as a di-, tri-, and tetra-chlorobiphenyl, respectively.
The adsorption of each class of isomers is illustrated by the adsorption
data for medium-temperature char in Figure 19. In this figure, the average
adsorption for all isomers such as monochloro-, and dichloro- have been con-
sidered together. It is clear that the higher chlorinated isomers were
preferentially adsorbed.
Earth Material Properties Affecting Adsorption
The relation between total organic carbon (TOC) content and PCB adsorp-
tion was investigated. The PCB adsorption constant (K) plotted as a function
of TOC is shown in Figure 20, A very highly significant (.001 level) corre-
lation was found with a linear regression-relation of
K = 255 + 18.5 TOC
r2= 0,87.
Thus, the PCB adsorption constant (K) can be estimated from a knowledge of
the TOC content of the earth material.
The relation between surface area (SA) of the adsorbents and PCB adsorp-
tion was also investigated. Surface area was measured using three adsorbates,
nitrogen, carbon dioxide, and ethylene glycol. As is evident from the data
in Table 1, each method can yield large differences in surface area depend-
ing on the surface characteristics of the particular adsorbent. Nitrogen
adsorption gives a measure of the external surface area and does not pene-
trate pores smaller than about 5 angstroms in diameter. Carbon dioxide
also measured external surface area, but is able to penetrate pores smaller
than about 5 angstroms in diameter. The difference between these two measure-
ments is particularly evident for the medium-temperature char sample, which
has a nitrogen area of 1.6 m2/g and a carbon dioxide area of 253 m2/g.
This large difference is interpreted as a measure of the large proportion of
micropores (< 5 A) contained in the medium-temperature char. This can be
contrasted to the soil materials, which exhibited relatively small differences
in surface area between the nitrogen and carbon dioxide measurements.
The third method of measurement was ethylene glycol adsorption. Ethyl-
ene glycol penetrates the interlayer region of swelling clays such as mont-
morillonite and gives a measure of the total surface area in these types of
samples. The difference between the external area measured by nitrogen and
the area measured by ethylene glycol gives an estimate of the "internal" or
"interlayer" surface area. It is evident that the interlayer surface area
is very high for the montmorillonite clay (Table 1).
The PCB adsorption constant (K) plotted as a function of CQ2-surface
(SA) is shown in Figure 21. A highly significant (.001 level) correlation
was found with a linear regression relation of
55
-------
K = 230 + 6.64 SA
r2 = 0.82.
Very poor correlations were obtained using nitrogen and ethylene glycol
surface areas. This is interpreted to indicate that PCB molecules were
not able to penetrate into the interlayer surface regions of montmorillonite
clay to be adsorbed. Thus, the PCB adsorption constant (K) could be esti-
mated from CC-2-SA values. This result may be fortuitous since the C02-SA
values parallel the TOC values. More studies are needed to confirm this
result.
A three-variable regression analysis of the PCB adsorption constant U),
TOC, and C02--SA were investigated and the results are shown in Figure 22. A
highly significant (.001 level) correlation was obtained with a linear regres
sion relation of
K = 188 + 3.36 SA + 11.4 TOC
T2 = 0.95.
The magnitude of the coefficients for SA and TOC indicate that TOC is
the dominant property by a factor greater than three. The best estimates
of K were obtained by a knowledge of both SA and TOC; however, if only one
soil property must be chosen to estimate K, the TOC should be the property
of choice. Care should be taken when applying this equation since K may
be over-estimated for soils with low TOC content.
56
-------
monochlorobiphenyls
•/\ dichlorobiphenyls
trichlorobiphenyls
tetrachlorobiphenyls
pentachlorobiphenyls
2,000
0.10
Amount of Char (g)
0.20
ISGS1979
Figure 19. The percentage of isomers in water-soluble Aroclor
1242 remaining in solution after adsorption by
medium-temperature char
§ 1,000 -
Q.
O
CO
a
20 30 40 50 60
Total Organic Carbon (%)
70
ISGS1979
Figure 20. PCB adsorption constant (K) as a function of
total organic carbon content (TOO
o
O
2,000
1,000
K = 230 + 6.64S-4
r2 = 0.82
0 40 80 120 160 200 240 280
CO2—Surface Area (m /g).
ISGS1979
Figure 21. PCB adsorption constant (K) as a function of
surface area (SA)
2,000
I 1,000
K= 188 + 3.36S/4 + 11.4rOC
r2 = 0.95
1,000
IS
experimental
2,000
ISGS1979
Figure 22. Three-variable regression analysis of the PCB
adsorption constant (K) as a function of total
organic carbon content (TOO and surface area
(SA);/fcomputed as a function of
^experimental
57
-------
SECTION 7
MOBILITY OF PCBs AND DICAMBA IN SOIL MATERIALS:
DETERMINATION BY SOIL THIN-LAYER CHROMATOGRAPHY1
RESULTS AND DISCUSSION
Rf values obtained from autoradiography were compared with those
obtained by zonal extraction. The agreements were excellent. Autoradiog-
raphy was especially satisfactory for soil TLC studies, as it provides a
qualitative picture of movement (for example, diffusion, tailing) while
allowing measurement of frontal Rf. Zonal extraction gave a more quantita-
tive picture of PCB movement. Also detected by the zonal extraction pro-
cedure are tailing, the origin spot, and the frontal movement of the com-
pound; however, definition of the concentration profile is limited by the
length of the soil segment chosen and by the extraction and analytical
efficiencies. In summary, the two methods gave essentially identical pic-
tures of PCB movement on TLC plates.
The mobilities of Aroclor 1242, Aroclor 1254, and Dicamba in several
earth materials expressed as frontal Rf values are summarized in Table 13.
The data for capacitor fluid were identical to Aroclor 1242 and are not
shown in Table 13. The data show that under the conditions tested, Aroclor
1242 and Aroclor 1254 (and the capacitor fluid) stayed immobile in these
soil materials when leached with water or Du Page leachate, but were highly
mobile when leached with carbon tetrachloride. Dicamba showed the reverse
trend, being highly mobile in water and in Du Page leachate and quite
stationary when leached with carbon tetrachloride.
A closer look at the structure of Dicamba and PCB will help explain
the mobilities observed. Dicamba is 3,6-dichloro-o-anisic acid. Because
Dicamba has two polar groups (-COOH and -OCH3), hydrogen bonding can occur
between the water molecules and the carboxyl and methoxy groups of Dicamba;
this increases the solubility of Dicamba in polar solvents like water and
leachate. Solubility of Dicamba in water is 4,500 ppm (Herbicide Handbook,
1974).
PCBs are nonpolar in nature and are only very slightly soluble in polar
solvents like water. Solubility of Aroclor 1242 in water has been deter-
mined to be 200 ppb (Tucker, Litschgi, and Mees, 1975), and Aroclor 1254,
1 Authors: R. A. Griffin, A. K. Au, and E.S.K. Chian
58
-------
TABLE 13. Mobility of Aroclor 1242, Aroclor 1254, and Dicamba in several soil materials leached with various solvents as measured
by soil TLC.
Rf values
Ava
Silty clay loam
Bloomfield
Loamy sand
Catlin
Silt loam
Catlin
Loam
Cisne
Silt loam
Coal char
(1200°F)
Drummer
Silt loam
Flanagan
Silt loam
Ottawa
Silica sand
Aroclor
1242
.02
.03
.02
.02
.03
.03
.03
.02
.03
H2O DuPage leachate
Aroclor Aroclor Aroclor Aroclor
1254 Dicamba 1242 1254 Dicamba 1242
.02 1.00 .02 .02 1.00 1.00
.03 1.00 -
.02 .85 .04 .04 .90 1.00
.02 1.00 .03 .03 1.00 1.00
.02 1.00 .03 .02 .99 1.00
.03 .79 .04 .04 .80 1.00
.03 0.99 - - - 1.00
.02 1.00 .06 .05 1.00 1.00
.03 1.00 .03 .03 1.00 1.00
CCI4
Aroclor
1254
.96
'
1.00
1.00
1.00
1.00
1.00
1.00
1.00
Dicamba
.02
"
.02
.03
.05
.03
.03
.02
.02
56 ppb (Haque, Schmedding, and Freed, 1974); however, PCBs are quite soluble
in organic solvents like acetone, methanol, benzene, or carbon tetrachloride.
Mobilities of Aroclor 1242, Aroclor 1254 and Dicamba were tested in silica
gel leached with acetone, methanol, benzene, carbon tetrachloride, and
mixtures of water-acetone and water-methanol. The results are given in
Table 14. Consistent with the soil TLC data obtained by leaching with
carbon tetrachloride, Rf values of 1.00 were obtained using the organic
solvents. The data also indicated that relatively small amounts of water
(9 percent) in the methanol-water mixture significantly reduced the mobil-
ity of the PCBs compared to pure methanol. Dicamba mobility followed the
inverse trend. It was quite clear that mobilities of PCBs and Dicamba in
soil materials and silica-gel were highly related to the solubility of PCBs
and Dicamba in the solvent with which the TLC plates were being leached.
The above finding has great significance in the disposal of PCB wastes.
To prevent potential migration of PCBs in a landfill, PCB wastes and organic
59
-------
solvents should not be disposed of in
the same landfill location, and the
PCBs should not be allowed to come in
contact with leaching organic solvents.
The mobility of PCBs in carbon
tetrachloride can be correlated with
characteristics of the earth materials
such as clay content, surface area, PH,
cation exchange capacity, and organic
matter. The zonal extraction technique
yields a quantitative measure of the
quantity of PCB that moves to the top
of each soil TLC plate with the solvent.
Although PCBs have R* values of 1.00
for all the soil materials leached with
carbon tetrachloride (Table 13), the
percentage of PCBs that were retained
against leaching on the lower part of
the TLC plates differs for each of the
soil materials. The mobility measured
in this manner was correlated with soil
material properties (Table 15).
roc i
ISGS 1979
Figure 23.
Retention of PCB in soils leached with carbon
tetrachloride as a function of total organic
carbon (TOO content of the soils
The relation between the total
organic carbon content of the soils
and retention of PCB by soils leached with carbon tetrachloride was investi-
gated, and the results are plotted in Figure 23. The plots indicate there
is a tendency toward higher percentages of retention of Aroclor 1254 com-
pared to Aroclor 1242. This hypothesis was tested using the paired t-test
statistic (Texas Instruments, 1975) for pairs of data from the same soil.
The result was a highly significant (.01 level) difference between the
mobility of Aroclors 1242 and 1254 in carbon tetrachloride solution, Aroclor
1254 being less mobile. This result is in agreement with Tucker, Litschgi,
and Mees (1975), Haque and Schmedding (1976), and Lee et al. (1979) who
found the higher-chlorinated PCB isomers had a higher affinity for soil
materials than the lower-chlorinated isomers.
The results of linear regression analysis of the data shown in Figure
23 is tabulated in Table 15. The data yielded a highly significant (.001
level) correlation between retention of PCBs by soil and the TOC content.
The coefficients of determination (r2) were 0.87 and 0.84 for Aroclors 1242
and 1254, respectively. These results are also congruent with those of
several other workers who have noted a relation between PCB retention and
the organic matter in soils (Briggs, 1973; Iwata, Westlake, and Gunther,
1973; Tucker, Litschgi, and Mees, 1975; Scharpenseel, Theng, and Stephan,
1978; Lee et al., 1979).
Also included in Table 15 are the results of linear regression analysis
of the effect of surface area on mobility of PCBs. The results indicate no
significant correlation with nitrogen surface area values, but significant
correlations were obtained using carbon dioxide and ethylene glycol surface
60
-------
TABLE 14. Mobility of Aroclor 1242,Aroclor 1254, and Dicamba on silica-gel TUG plates using various leaching solvents.
R- values
Leaching solvent Aroclor 1242 Aroclor 1254 Dicamba
Deionized H2O 0.15 0.15 0.95
Du Page leachate 0.15 0.15 0.80
80% H20-20% acetone 0.09 0.06 0.30
Acetone 1.00 1.00 0.88
15% H20-85% methanol 0.79 0.79 0.71
9% H20-91%methanol 0.80 0.83 0.70
Methanol 1.00 1.00
Benzene 0.99 0.95 0.09
Carbon tetrachloride 1.00 1.00 0.03
TABLE 15. Results of linear regression analysis of retention of Aroclor 1242 and Aroclor 1254 by soils against leaching with
carbon tetrachloride (R) versus the soil properties of total organic carbon (TOG) content and surface area measured
by adsorption of nitrogen (N2-SA), carbon dioxide (C02-SA), and ethylene glycol (EG-SA) expressed as values of
the coefficient of determination (r ).
r values
Aroclor 1242 Aroclor 1254
R vs. TOG 0.873 0.84a
R vs. N2-SA 0.03NS 0.09NS
R vs. CO2-SA 0.54b 0.54b
R vs. EG-SA 0.74a 0.43C
R vs. TOG, CO2-SA 0.90a 0.89a
R vs. TOG, EG-SA 0.93a 0.87a
Not significant
Significant at 0.001 level
Significant at 0.05 level
Significant at 0.1 level
61
-------
area values. Highly significant correlations were obtained by including both
surface area and TOC content in regressions with the mobility of the two
Aroclors (Table 15).
The lack of strong correlations and the variability in the results
using the surface area data precludes a conclusion as to the ability of PCB
molecules in carbon tetrachloride solution to penetrate into the interlayer
region of expandable clays or as to the significance of the surface area of
soils in retarding PCB migration. There is, however, a clear relation
between the TOC content of the soils and their ability to retard PCB migra-
tion.
62
-------
SECTION 8
DEGRADATION OF PCBs BY MIXED MICROBIAL CULTURES1
RESULTS AND DISCUSSION
Preliminary Work
The first mixed culture to be grown, the C-line, yielded the yellow-
colored intermediate described by Furukawa and Matsumara (1976) in about 5
weeks; the S-line took about 2 weeks and the N-line, less than 1 week to
turn the solutions yellow. Because the S-line had some previous exposure
TABLE 16. Results after 15 and 45 days of degradation using culture grown in acetic acid.
Percentage of removal of peak
when compared to blanks
Peak number
2
4
5
6,7
8
9
11
13
15
16
17
18
22
24
25
26
27
28
32
34
37
41
44,45
Number of Cl
1
2
2
2
2
2
2
3
3
3
2 and 3
mixture
3
3
3
4
4
4
3
4
4
4
4
4
1 5 days
100
29
14
100
100
62
37
0
0
0
0
22
29
10
0
-
0
100
0
0
31
0
0
45 days
100
28
23
100
100
62
37
0
0
0
0
23
56
13
0
-
0
85
0
0
34
0
0
Retention time
in Fig. 25
(min)
10.43
15.30
15.97
16.96
17.19
18.96
19.74
22.04
23.55
24.56
25.83
26.12
27.40
28.29
29.45
29.90
30.65
30.96
31.77
34.34
35.37
38.35
40.96
Total degradation
43
44
1 Authors: R. R. Clark, E.S.K. Chian,and R. A. Griffin
63
-------
to PCBs before isolation and the N-line had much more exposure, it appears
that the presence of PCBs in a particular environment enhances the selection
of microorganisms capable of degrading PCBs. In view of this, the mixed
cultures were grown on biphenyl for more than 7 months to ensure that the
cultures were highly selective toward biphenyl-degrading organisms before
the PCB degradation experiments were conducted.
The first PCB degradation studies were run for 15, 30, and 45 days.
Figure 24 shows a chromatogram of the 15-day extract. Figure 25 shows the
chromatogram of the 15-day blank. Table 16 lists the peaks found in the
PCB samples. Occasionally peaks cannot be resolved, therefore these are
combined in the tables. As can be seen, all of the monochloro and most of
the dichlorobiphenyls are easily degraded. It also appears that most of
the degradation occurred prior to 15 days.
The results, however, were suspect because the mixed culture was grown
in acetic acid for 3 days prior to being added to the PCB solution under
study to eliminate the biphenyl in the solution. This procedure could have
made the mixed culture less able to degrade the PCBs because the selective
pressure toward a biphenyl-degrading culture was eased.
Table 17 lists the results of a 7-day study of PCB degradation using
both a mixed culture grown only on biphenyl and one grown on acetic acid for
TABLE 1 7. Comparison of mixed cultures grown on biphenyl only and those grown on acetic acid.
Peak number
Number of Cl
Percentage of removal after 7 days when compared to blanks
Biphenyl only Acetic acid grown
2
3
4
5
6,7
8
11
13
15
16
17
18
22
24
25
27
28
32
33
34
1
1
2
2
2
2
2
3
3
3
2 + 3
3
3
3
4
4
3
4
4
4
100
100
100
-81
100
100
100
0
19
0
66
100
100
93
33
0
100
0
0
0
100
100
7
13
100
100
20
0
0
0
0
0
15
0
0
0
68
0
0
0
Total degradation
82
30
64
-------
en
C71
Figure 24. Water-soluble Aroclor 1242 after 15 days of degradation
-------
cn
Figure 25. Original water-soluble Aroclor 1242 solution prior to degradation
-------
3 days prior to the experiment. The culture grown only on biphenyl did much
better than the culture grown on acetic acid for 3 days. These results indi-
cate that the data in Table 16 may not be quantitatively correct, but are
probably qualitatively correct in that most of the degradation occurred in
less than 15 days.
The data in Table 17 show that all the mono- and dichloro-biphenyl
isomers (except peak number 15) were easily degraded in 7 days. Also, many
of the trichlorobiphenyl isomers were degraded, but some were resistant to
degradation. Almost all of the tetrachlorobiphenyl isomers were not degraded
in 7 days.
Sonification Study
To determine if the extraction procedure would extract all the PCBs in
solution, a 2-day degradation sample of the S-line culture was run and soni-
fied before extraction. In this way, the PCBs that are inside the micro-
organisms would be extracted. If the pH 2 extraction did not extract all
of the PCBs inside the microorganisms, then the sonified extract would show
higher amounts of PCB isomers and/or more isomer peaks because of isomers
that could not be extracted with the pH 2 extraction. Table 18 shows the
results of the sonification study. The sonified extract showed a greater
amount of PCB removal than did the pH 2 extract. This difference may not
be significant since the error in the analysis may be more than the dif-
ference between the two samples. Volatilization of PCBs due to localized
heating of the solution near the sonifier tip is another possible explana-
tion. In addition, all peaks that were missing in the pH 2 extract were
missing in the sonified extract. Because of this, it was felt that the pH
2 extraction procedure adequately represented the degradation of PCBs that
was occurring in the solutions.
Comparison of the Three Mixed Cultures
Degradation samples were run for 5, 10, and 15 days using Aroclor 1242
saturated solution inoculated with 3-day old N-, C-, or S-line cultures.
Figure 26 shows a comparison of a 5-day blank and 5-day degradation using
the S culture. Figure 27 shows a 5-day blank, and Figure 28, a 5-day degrad-
ation, both using the N culture. Note that the analyses were performed using
a different capillary column. The C culture was also analyzed on the same
capillary column as the N culture, therefore the chromatograms resembled
the N-culture chromatograms. As can be seen from these figures, the S-
culture degradation may not be directly comparable to the C- and N-culture
degradation because of differences in the capillary column. Table 19 com-
pares the amount of degradation for each culture at 5 and 15 days. Some
peaks were not reproducibly present, so no percentage of degradation is
reported for them. Because of analytical difficulties, for those isomers
yielding very small peaks, degradation of less than 20 percent was not
considered significant. For most of the peaks the degradation was the
same for 5 as for 15'days; this observation indicated that most of the
degradation occurred in less than 5 days.
67
-------
TABLE 18. Comparison of sonified extracts and pH 2 extracts.
Peak number
Number of Cl
Percentage of degradation
pH 2 extract Sonified extract
2
3
4
5
6,7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28,29,30
31
32
33
34
35
36,37
38
39
40
41
42
43
44
45
1
1
2
2
2
2
2
3
3
3
3
2
2 + 3
3
3
3
3
3
3
4
4
4
4
3
4
4
4
4
4
4
4
4
4
4
3
4
4
4
4
4
100
100
54
45
100
95
100
0
8
29
28
100
27
56
13
74
66
100
0
16
.
100
0
100
0
0
17
0
0
0
0
0
2
38
5
5
.
.
.
-
100
100
75
68
100
99
100
34
39
45
51
100
48
70
0
82
75
100
17
35
100
19
100
0
7
80
10
10
11
13
21
4
49
65
2
-
Most of the isomers were degraded to the same degree with all three
mixed cultures; however, some were degraded to different extents. Differ-
ences in degradation can be due to different mechanisms or pathways of
degradation (Ahmed and Focht, 197.3a; Baxter et al., 1975; Furukawa,
Tonomura, and Kamibayashi, 1978). The usual patterns reported in other
literature—monochlorobiphenyIs are easily degraded, dichlorobiphenyls
are slightly less degradational, and tetrachlorobiphenyls are not readily
degraded—are seen in the data reported here. In this study, however, the
degradation occurred more rapidly than previously reported (Ahmed and Focht,
1973a; Baxter et al., 1975; Furukawa and Matsumura, 1976; Furukawa,
Tonomura, and Kamibayashi, 1978; Sundstrom, Hutzinger, and Safe, 1976;
Tucker, Saeger, and Hicks, 1975; Mihashi et al., 1975).
68
-------
5-day blank
5-day degradation
Figure 26. Comparison of 5-day blank and 5-day degradation of Aroclor 1242 using the S culture
Kinetic Study
A degradation study was run using the N culture for 3, 6, 12, 24, and
48 hours. Two samples and two blanks were taken at each time increment.
Table 20 shows the results of this study. As with previous studies, the
lower chlorinated isomers were degraded most rapidly. Many of the isomers
showed increasing amounts of degradation with each time increment. Six of
the peaks are plotted in Figures 29 and 30. The rates of degradation are
somewhat different for different isomers; even isomers with the same number
of chlorines per molecule have different rates of degradation. This dif-
ference has also been reported by Baxter et alI., (1975). This finding
would indicate that the position of substitution has a bearing on the
degradational properties of a particular isomer. Even with the use of a
capillary column, each particular peak may not be a pure isomer; a peak
69
-------
-9
10
16
'
13
4
19
18
u
Figure 27. Results of 5-day blank of Aroclor 1242 using the N culture
10
IS
15
13
16
I7I8
22
Figure 28. Results of 5-day degradation of Aroclor 1242 using the N culture
70
4.445
-------
TABLE 19. Comparison of degradation of water-soluble Aroclor 1242 by the three mixed cultures of soil microorganisms.
Peak number
2
3
4
5
6,7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Number of Cl
1
1
2
2
2
2
2
3
3
3
3
2
2 + 3
3
3
3
3
3
3
4
4
4
4
3
4
4
4
4
4
4
4
4
4
4
3
4
4
3
4
4
4
4
4
C
(%)
100
100
64.8
57.3
98.7
-
97.2
12.6
19.8
38.5
48.4
100
41.9
73,2
7.4
77.8
51.3
100
2.9
21.9
-
100
7.2
100
99
8.5
25.2
-
9.7
8.1
0.7
0
0
19.4
3.7
0
20.3
25
0
0
0
0
17
5-day degradation
S
(%)
100
100
100
93
100
100
99
88
67
-
90
-
71
99
100
98
96
-
-
72
-
99
72
100
90
42
-
•,
38
34
39
-
38
44
81
15
30
-
-
-
•
N
(%)
100
100
30.9
98.6
100
99.3
99.7
0
98.7
45.7
15.2
75.2
39.3
86.9
22.0
100
93.9
100
0
99.2
-
60.9
0
100
100
20.1
-
0
27.3
55.6
55.6
0
37.0
-
17.5
37.9
37.9
0
47.9
-
-
21.3
21.3
C
(%)
100
100
71.5
61.7
100
99.5
97.9
3.8
17.3
21.4
34.9
100
42.8
77.0
0
80.6
53.9
100
0
18.7
0
100
0
100
0
0
0
0
31.8
0
0
0
0
22.5
0
0
49.4
-
0
-
-
-
76.1
1 5-day degradation
S
(%)
100
100
100
99
100
100
99
89
71
-
89
-
78
99
100
100
98
-
-
76
-
99
73
100
90
56
-
-
43
49
67
-
51
45
85
0
43
-
-
-
-
N
(%)
100
100
25.3
100
100
100
99.9
0
100
54.8
0
78.2
45.8
96.8
0
100
98.4
100
0
100
0
70.3
0
100
100
13.0
21.6
0
0
80.2
62.8
0
38.2
0
0
100
0
0
62
-
-
-
-
could be a mixture of isomers. If it were a mixture, the isomers may have
different degradation rates.
In order to calculate the amount of PCBs left at any particular time,
the reaction order, n, and the rate constant, K, must be determined. The
basic rate equation is
71
-------
_ _
dt ~
where c is the concentration of a compound, || is the change in concentra-
tion with respect to time, t and K and n are as defined above. The equation
can be rearranged to
f---
TABLE 20. Results of the kinetic study of the degradation of water-soluble Aroclor 1242 by the N culture.
Percentage of degradation
Peak number
2
6-1
,7
10
11
12
13
14
15
1 C
16
17,18
19,20
21,22
23
24
25
26
27
28,29,30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Number of Cl
1
1
2
2
2
2
2
3
3
3
3
2
2 + 3
3
3
3
3
4
4
4
3
4
4
4
4
4
4
4
4
3
4
4
3
4
4
4
4
4
3hrs.
(%)
92.1
100
0
12.5
72.1
35.5
38.2
0
8.4
41.6
0
100
20.6
17.3
27.1
19.5
0
34.2
18.4
32.2
0
0
0
0
0
0
0
0
24.7
0
0
,
14.3
„
.
0
6hrs.
(%)
100
100
9.8
10.9
93.3
52.9
64.1
3.6
0
18.5
0
100
12.9
9.8
31.9
25.3
0
40.1
0
56.1
0
0
0
0
0
0
0
0
0
0
0
_
0
.
12hrs.
(%)
100
100
21.8
20.8
100
76.7
80
0
5.6
13.7
0
100
15.0
20.1
53.4
40.4
6.0
56.4
0
84.2
. 0
0
62.4
0
0
0
0
0
0
0
0
0
24 hrs.
(%)
100
100
19.7
35.9
100
87.9
88.5
0
20.9
14.3
0
100
15.8
26.9
64.2
49.8
14.4
0
92.9
15.0
0
0
0
0
0
0
0
0
0
0
48 hrs.
(%)
100
100
36.6
69.6
100
90.3
95.5
2.6
55.9
23.1
5.8
100
22.9
45.5
80.0
59.8
37.2
71.5
o
95.9
51.8
0
17.1
0
o
0
0
0
6.2
0
0
0
n
72
-------
Percentage
Remaining
12
24
48
Time in hours
Figure 29. Degradation at timed intervals of selected PCB isomers
73
ISGS 1978
-------
Percentage
remaining
3 6
Time in hours
Figure 30. Degradation at tinned intervals of selected PCB isomers
I SOS 1979
74
-------
By integrating equation 3, we get equation 4
(4)
l-n l-n
Rearranging this
_ l-n r l-n
°t ~ °o
or
C l~n = (n-l) Kt + C^~n ; (6)
the final equation is then
Ct = [(n-l) Kt + CQ ^ ^ - (7)
To determine the reaction order for the reduction of sample peaks
5, 8, 9, (17-18), (19-20), and 26 the following methods were used:
Since the basic rate equation is
_d£ = -Kd1 (2)
at
then
dt
By plotting inFgf) vs. in C, the slope (n) and the intercept (in K) can be
determined (|| was determined from the percentage remaining from Table 20;
C was taken as the concentration midway between the two points used to
determine a particular slope
Table 21 lists the K and n values for peaks number 5, 8, 9, (17-18),
(19-20), and 26. The rate constant, K, was calculated using the percent-
age of PCB remaining at each time; therefore the units of K are (percentage
remaining) ^~n'per hour. C equals 100 percent at time zero. The
75
-------
constants in Table 21, when used in equation 7, yield answers that are within
5 percent of the experimental data.
Several things should be pointed out with regard to these numbers.
First, the differential method used here to determine n and K is subject
to errors because of the difficulties in determining the accurate values
for -%j-- Tnis is especially applicable here because only 5 data points
were used to calculate the line for determining n and K. However, the
correlation coefficient (r) was very high (> 0.95) for all but two of the
peaks. These two peaks, (17-18) and 5, had r values of 0.867 and 0.64
respectively. Also the closeness of the calculated numbers to the experi-
mental data gives support to the calculated n and K values.
Second, the n value for peak (19-20) is very high. One would expect
n values of 2 or less. The high n value may be due to the fact that this
peak and (17-18) are a combination of two or more different PCB isomers.
If the two or more isomers degrade at rapidly different rates, then the n
value for the combined peak may be somewhat unusual. The n values, although
they may not be a true reaction order for a particular isomer, do allow an
empirical prediction using equation 7.
Cometabolization
PCBs are rarely found in isolation in the environment. There are
usually more easily oxidizable chemicals such as acetic acid and amino
acids available for the microorganisms to utilize. When microorganisms
are utilizing one compound as an energy source, cometabolization of another
compound may occur (Baxter et al., 1975; Horvath, 1972). Horvath (1972)
described cometabolization as "the process in which a microorganism oxidized
a substance without being able to utilize the energy derived from this oxida-
tion to support growth."
TABLE 21. Rate constants and reaction orders for selected PCB isomers.
Peak number
5
8
9
17,18
19,20
26
Number of Cl
2
2
2
3
3
3
n
0.347
2.211
1.747
2.04
3.72
1.651
K (percentage remaining) /hr
3.44 x 10"1
9.283 x 10"4
7.24 x 10~3
6.1 9 x 10~4
2.92 x 10~7
1.04 x 10"2
76
-------
Cometabolization of PCBs in the presence of sodium acetate was studied.
Table 22 lists the results after 5 and 15 days of cometabolization and PCB
degradation. Both cultures degraded PCBs to a large degree, but the cometab-
olization samples produced greatly enhanced degradation of the higher chlor-
inated PCB isomers and other hard-to-be-degraded isomers.
One possible reason for enhanced degradation of PCBs in the cometabo-
lization samples could be due to the increased mass of microorganisms present.
When the microorganisms are grown on PCBs alone, very little additional
biomass is produced because of the limited amount of carbon source present
in the solution. When acetic acid, or any easily degradable substrate, is
present the microorganisms can grow on the acetic acid. While they are
using the acetic acid they may oxidize the PCBs slightly. Because the
total mass of microorganisms present has increased, this slight oxidization
by each microorganism adds up to a more complete degradation for the entire
system as a whole. This type of cometabolization will only work for cul-
tures of microorganisms that are acclimated or selected to PCB degradation.
Another possible explanation is that the PCBs adsorbed onto the addi-_
tional biomass in the cometabolization samples and were not extracted. This
is a possibility because PCBs adsorb easily onto nonpolar surfaces such as
bacterial cell walls. The sonification study showed that the PCBs could
be extracted from the small amounts of biomass present in the samples con-
taining only PCBs, but did not rule out the possibility of poor extraction
in the presence of large amounts of biomass.
Soil and Sludge Extraction
Chromatograms of the soil and sludge extracts were compared to chromat-
ograms of Aroclor 1242 and 1254. By this comparison the peaks in the
extracts that were caused by PCB isomers were determined. Figure 15 shows
a plot of relative area of the identified PCB peaks versus chromatograph
retention time. If the PCBs were being degraded in the soil after applica-
tion in the sludge, then there should be a buildup of the higher chlorinated
isomers (those with longer retention times). In addition, there should be
fewer of the lower chlorinated isomers in the soil. Each point on the graph
represents a peak determined to be a PCB isomer. The number over the point
is the percentage of degradation of that peak in the cometabolization study.
The degradation from the cometabolization study was used because it would
more closely represent the environmental situation where other organics are
present.
As illustrated in Figure 31, the sludge has more low chlorine isomers
than the soil, and the soil has relatively more high chlorine isomers than
the sludge. This seems to indicate that there is degradation of some low
chlorine isomers and an accumulation of the nondegradational higher chlorine
isomers. The cometabolization degradation results agree in part with this
conclusion. The two peaks at 25 and 28 minutes are not degraded and they
build up in the soil, whereas the peak at 26 minutes is degraded and does
not build up to the same extent. This pattern, however, is not obtained for
the peaks at 15.5 and 23 minutes. This result may not be significant
77
-------
TABLE 22. Co-metabolization of PCBs with sodium acetate by mixed culture microorganisms.
Peak number
2
3
4
5
6,7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Number of Cl
1
1
2
2
2
2
2
3
3
3
3
2
2 + 3
3
3
3
3
3
3
4
4
4
4
3
4
4
4
4
4
4
4
4
4
4
3
4
4
3
4
4
4
4
4
5 day
PCB
(%)
100
100
54.5
99
100
100
100
0
96.8
96.8
8.2
100
32.9
96.8
0
100
96.7
100
9.1
95
95
24.6
0
100
100
8.0
8.0
0
4.6
50.8
50.8
0
48.1
0
0
100
1.1
0
52.2
0
0
0
0
Percentage of
Cometab.
(%)
100
100
86.0
99
100
100
100
5.0
100
100
38.9
100
90.4
100
0
100
100
100
0
97.4
97.4
100
0
100
100
62.2
100
11.0
19.1
100
100
0
100
15.4
15.4
100
61.7
0
100
0
0
0
50
degradation
PCB
(%)
100
100
45.5
100
100
100
100
6.4
100
98.6
10.4
100
58.9
100
0
100
100
100
0
90
100
57.4
0
100
100
35.2
35.2
0
0
78.1
78.1
0
64.3
13.6
13.6
100
38.7
20.6
75.2
0
0
0
0
15 day
Cometab.
(%)
100
100
78.2
100
100
100
100
9.9
100
100
68.9
100
95.4
100
0
100
100
100
0
90
100
100
0
100
100
91.2
100
0
77.6
100
100
22.1
100
35.9
36.9
100
100
0
100
0
0
0
50
78
-------
10
5.0-
4.5-
4.0-
3.5-
Percentage
relative
area
3.0-
2.5-
2.0-
1.5-
1.0-
0.5-
15
20 25 30
Retention time (min)
35
40
45
50
Figure 31. Relative distribution of PCB isomers in the soil and sludge extracts with respect to retention time. The number over the points are the percentage degrada-
tion of that peak in the cometabolization study (table 22).
-------
because different cultures have been shown in this study to degrade the vari-
ous PCB isomers differently.
Several things could lead to error in this interpretation. First, the
soil was dried in an oven, which may have volatilized some of the lower
chlorine isomers. Second, the data are not quantitative in the sense that
one cannot relate the total amounts of a PCB isomer in the sludge and how
much sludge was applied to the amount of that PCB isomer found in the soil.
Other Observations
It has been noted that PCBs adsorb to glass surfaces (Gresshoff,
Mahanty, and Gortner, 1977). In the blanks that were run in this study,
however, no adsorption of PCBs was noted. One possible explanation of this
is that when microorganisms are present in the water, the PCBs adsorb pre-
ferentially onto them rather than on the glass surface.
Several other points must be clarified when interpreting the results
of this study. First, a mixed culture of microorganisms was used in con-
trast to the pure cultures of PCB-degrading microbes used by most other
researchers (Lunt and Evans, 1970; Catelani et al., 1971, 1973; Gibson et
al., 1973; Wallnofer et al., 1973; Ahmed and Focht, 1973a, 1973b; Baxter
et al., 1975; and Furukawa, Tonomura, and Kamibayashi, 1978). Therefore,
the more rapid degradation of PCBs found in this study compared.to that
previously reported may be partly attributed to the use of the mixed culture.
In addition to the PCB-degrading microbes in such a culture, many other
species of microorganisms may be present that could speed the degradation
of the PCBs by removing potentially inhibitory intermediates. This mixed-
culture approach is closer to the conditions that exist in the environment
than a pure culture.
Second, only water-soluble PCBs were utilized. Water-soluble and total
PCBs have different relative proportions of PCB isomers (Haque, Schmedding,
and Freed, 1974). Therefore, the degradation of water-soluble PCBs is less
than would be achieved for total PCBs. Because most potential migration
of PCBs would be in an aqueous environment, the degradation of the water-
soluble isomers are the most environmentally significant.
The PCB degradation studies all showed the same general trends of other
research (Ahmed and Focht, 1973a, 1973b; Baxter et al., 1975; Furukawa,
Tonomura, and Kamibayashi, 1978; Tucker, Saeger, and Hicks, 1975; Mihashi
et al., 1975); namely, the lower chlorinated isomers, in general, were more
easily degraded than higher chlorinated ones, and different isomers with
the same number of chlorines degraded at different rates. But perhaps most
significant in this study is that PCBs can be degraded to a large extent in
a short period of time by mixed cultures of microorganisms. In addition,
if a PCB-degrading culture is used, the microbes will cometabolize PCBs
while using another compound as a source of energy.
Because two of the mixed cultures were enriched from soil or sediment
that contained PCBs, and because it appears that there were already PCB-
degrading microbes present, PCBs are expected to be degraded to some extent
80
-------
in the environment. Koeman et al. (1969) pointed out that most PCB resid-
uals in animals yield GC chromatographs similar to Aroclor 1260, a rather
highly chlorinated series of PCB isomers. This has also been noted by other
researchers (Smith, K. E., Illinois Natural History Survey, personal com-
munication) since then. Because Aroclor 1242, a relatively lower chlorinated
series of isomers, accounted for the majority of PCBs produced and marketed
(HEW, 1972), it would appear that the lower chlorinated isomers are degraded
in the environment to some extent.
The isolation from soil or sediment of microorganisms that can degrade
PCBs readily suggests the possibility of their future inoculation into
wastes containing PCBs before disposal, or biological digestion prior to
disposal.
81
-------
SECTION 9
EFFECT OF HUMIC ACID AND SOIL ON
VOLATILIZATION OF PCBs FROM WATER1
RESULTS AND DISCUSSION
PCBs are known to have low solubility in water (MacKay and Wolkoff,
1973) and to have a pronounced tendency to adsorb on particulate matter
(Griffin et al., 1978). The aromatic character of PCBs is believed to cause
them to be adsorbed onto humic substances by hydrophobic interactions.
Though the structure of humic acid is still under extensive investigations,
a major portion of humic acid is speculated to be constructed by cross!inked
phenolics and thus to have a distinct aromatic character.
The PCB volatility measurements given in table 23 show that water-
soluble humic acid attenuated the volatility of PCBs from aqueous solution.
The attenuation does not seem to be very effective, however, because as
much as 500 mg/L of humic acid were needed to reduce the volatility to 73
percent of that of the control, that is, from pure water as shown by the
chromatograms in Figure 32.
Recent evidence (Reuter and Havlicek, 1978) has indicated that humic
acids are composed predominantly of an aliphatic substructure (about 70
percent), whereas the aromatic moiety was actually present in the structure
as only about 20 to 30 percent of the total composition. This new struc-
tural information about humic acid explained the experimental observation
that relatively few PCBs were adsorbed by dissolved humic acids. Because
the binding of PCBs to humic acid in aqueous solution is relatively weak,
the PCBs in solution and those bound to humic acid actually formed a dynamic
TABLE 23. Relative amount of PCB volatilized from water containing humic acid or soil compared to control.
Control 50 mg/L humic acid 500 mg/L humic acid
100% 87% 73%
1000 mg/L (47.30)a 5000 mg/L 6400 mg/L
Control Catlin soil (236.5r (302.7)
100% 86% 50% 21%
The numbers in parentheses indicate mg/L of total organic C content.
1 Authors: Hsi Meng, E.S.K. Chian, and R. A. Griffin
82
-------
00
CO
Q.
O
h-
B
ISGS 1978
Figure 32. Comparison of attenuation of volatilized PCBs by dissolved humic acid addition. (A) Control. (B) 50 mg/L dissolved humic acid added. (C) 500 mg/L
dissolved humic acid added.
-------
equilibrium. Therefore, the binding of PCBs by soluble humic substances
does not substantially reduce the concentration of free PCBs. The volatili-
zation of PCBs is thus not effectively influenced by the presence of soluble
humic acids at a pH of 12; however, this may not be the case when the humic
acid is present as colloids at a lower pH. Adsorption of dissolved PCBs
on the particulate and nonionized molecules, such as humic substances, is
expected to be far more effective than that on the ionized molecules in
true solution; this observation is based on the results of this study of
PCB adsorption onto Cat!in soil and soluble humic acid.
Attenuation of the volatility of water-soluble PCBs by soil adsorption
resulted in a greater reduction of the volatile fractions of PCBs from the
aqueous solution (Table 23) than was the case for the soluble humic acid.
This analysis was based on the total organic content in the solution (see
the numbers in parentheses in Table 23). Therefore, the binding of PCBs
to solids does seem to effectively reduce the amount of PCB volatilized as
shown by the results of the Catlin soil addition (Table 23). Thus the
adsorption of PCBs on the particulate matter was more effective than that
on the soluble humic matter as shown by the chromatograms in Figure 33.
Table 24 compares our earlier soil adsorption study (Griffin et al., 1978)
to the soil volatility attenuation study presented here.
As indicated in Table 24, the results of volatilization and adsorption
are parallel in that for each the attenuation was more efficient by adding
a larger amount of soil. This correlation further implies that attenuation
of PCB volatilization is directly related to binding of the PCBs to the
soil solids by an adsorption mechanism.
The conditions of this experimental study were somewhat different than
would be expected under field conditions; however, a water-logged soil could
result in a condition that resembles those conditions. Also a landfill
site producing leachate may closely resemble the conditions of our experi-
ments with humic acid, because an appreciable amount of soluble humic sub-
stances has been found in landfill leachates (Chian, 1977).
Change in Composition of the Volatile PCBs
The percentage of lower-chlorinated PCBs in the samples was evaluated.
The early-eluting larger peaks in the chromatograms (Figs. 32 and 33)
represented the more volatile and lower-chlorinated PCBs. The area under
TABLE 24. The relative attenuation of volatilization of PCBs by soil as compared with the attenuation due to adsorption
from solution by the same soil.
1 000 mg/L Catlin soil 5000 mg/L 6400 mg/L
Adsorption attenuation 85 55 42
Volatilization attenuation 86 50 21
84
-------
00
(Jl
ISGS1978
Figure 33. Comparison of attenuation of volatilized PCBs by soil addition (see figure 32 for the control). (A) 1000 mg/L Catlin soil added. (B) 5000 mg/L Catlin soil
added. (C) 6400 mg/L Catlin soil added.
-------
these GC peaks was summed up and compared with the total area of all the GC
peaks in the chromatogram. The results of direct injection of PCB standards
dissolved in hexane containing 500 ppm Aroclor 1242 showed 52 percent lower-
chlorinated PCBs, whereas purging of water saturated with PCBs yielded 80
percent lower-chlorinated PCBs in the traps. By adding 50 mg/L of humic acid,
82 percent of the lower-chlorinated PCBs were found in the traps. Addition
of 500 mg/L of humic acid yielded 87 percent of the lower-chlorinated PCB
isomers in the trap. By the same token, the percentages of lower-chlorinated
PCBs found in the traps after soil additions were 80, 79, 74 and 73 percent,
respectively, with 0, 1000, 5000, and 6400 mg/L soil additions.
The higher-chlorinated PCBs were adsorbed more effectively by the
soluble humic acids; this can be explained by the strong hydrophobic bonding
of the PCBs to the humic substances. The affinity of higher-chlorinated
PCBs to the humic substances is reflected by an increased volatile portion
of the lower-chlorinated isomers; however, the soil addition experiments
showed a small decrease and a leveling off of the amount of PCBs in the traps
in the lower-chlorinated form. This is speculated to be due to limited
penetration of the smaller lower-chlorinated isomers into pores of the inter-
layer and edge regions of clay minerals contained in the Catlin soil, and
thus a slightly more efficient retention against volatilization.
The most visible change in isomer distribution comes from comparing
the hexane-soluble PCBs and water saturated with PCBs. This can be explained
by the higher volatility of the lower-chlorinated isomers and their higher
solubility in water.
Comparison with Theoretical Volatilization Process
MacKay and Wolkoff (1973) predicated the half-life of Aroclor 1242
volatilized from water to be 6 hours. Therefore, the quantity volatilized
in 20 minutes would correspond to 3 percent of the Aroclor added to the
purging flask used in this study. They pointed out that the high activity
coefficient of a sparingly soluble compound in water favors a high rate of
evaporation. The experimental conditions of this study agreed with the basic
assumptions outlined by MacKay and Wolkoff (1973). When 0.5 yg of PCBs
were introduced into water followed by purging, the amount of PCBs volatilized
was found to be 0.5 percent. The trapping efficiency was found to be 71 per-
cent as evaluated by concentrating a known amount of PCB in hexane and
following the same analytical protocol. Thus, the actual evaporation
efficiency was 0.7 percent. Since 1 vl of the 5 yL solution was intro-
duced onto the GC column, 0.7 percent times 5 gives the total evaporation
efficiency of 3.5 percent. This agrees well with the value of 3 percent
volatilized from pure water based on the theoretical calculation of MacKay
and Wolkoff's (1973) model. Similar calculations for the other experiments
allow for comparisons. For example, volatilization of PCBs from solutions
containing 500 mg/L humic acid was reduced to 2.6 percent, and volatiliza-
tion was reduced to 0.74 percent from suspensions containing 6400 mg/L soil.
86
-------
REFERENCES
Ahmed, A. K., 1976, PCBs in the environment: Environment, v. 18, p. 6.
Ahmed, M., and D. D. Focht, 1973a, Degradation of polychlorinated biphenyls by two species
of Achromobacter: Canadian Journal of Microbiology, v. 19, p. 47-52.
Ahmed, M., and D. D. Focht, 1973b, Oxidation of polychlorinated biphenyls by Achromobacter PCB:
Bulletin of Environmental Contamination and Toxicology, v. 10, p. 70-72.
Ahnoff, M., and B. Josefsson, 1975, Clean up procedures for PCB analysis on river water
extracts: Bulletin of Environmental Contamination and Toxicology, v. 13, no. 2, p. Ib9-lbb.
ANSC, 1976, American National Standards Committee, Askarel Committee responds to PCB peril:
American National Standards Institute Reporter, v. 10, p. 3.
Armour, J. A., 1973, Quantitative pe^chlorinated of polychlorinated biphenyls as a method
for confirmatory residue measurement and identification: Journal of the Association of
Official Analytical Chemists, v. 56, no. 4, p. 987-993.
Armour, J A., and J. A. Burke, 1970, Method for separating polychlorinated biphenyls from
DDT and its analogs: Journal of the Association of Official Analytical Chemists, v. 53,
no. 4, p. 761-768.
Ballschmiter, K., C. H. Unglert, and H. J. Neul, 1977, Abbau von chlorierten aromaten:
Mikrobiologischer abbau der polychlorierten biphenyle (PCB): Chemosphere, v. 6, p. 51.
Baxter, R. A., P. E. Gilbert, R. A. Lidgett, J. H. Mainprize, and H. A. Vodden, 1975, The
degradation of polychlorinated biphenyls by microorganisms: Science of the Total
Environment, v. 4, p. 53. ,
Beezhold, F. L., and V. F. Stout, 1973, The use and effect of mixed standards on the quantitation
of polychlorinated biphenyls: Bulletin of Environmefltal Contamination and Toxicology,
v. 10, no. 1, p. 10-15.
Berg, 0. W.,'P. L. Diosady, and A. V. Rees, 1972, Column chromatographic separation of poly-
chlorinated biphenyls from chlorinated hydrocarbon pesticides and their subsequent gas
chromatographic quantition in terms of derivatives: Bulletin of Environmental Contamina-
tion and Toxicology, v. 7, no. 6, p. 338-347.
Berg, 0. W., A. V. Rees, and M. S. Ali, 1978, Simplified techniques for the perchlorination
of polychlorinated biphenyls in environmental samples: Abstracts from the 29th Annual
Pittsburgh Conference, March 2, 1978, Cleveland, Ohio, p. 579.
Blakemore, R. P., and A. E. Carey, 1978, Effects of polychlorinated biphenyls on growth and
respiration of heterotrophic marine bacteria: Applied Environmental Microbiology, v. 35,
p. 323.
Blakemore, R. P., 1978, Effects of polychlorinated biphenyls on macro-molecular synthesis by
a heterotrophic marine bacterium: Applied Environmental Microbiology, v. 35, p. 329.
Bourquin, A. W., and S. Cassidy, 1975, Effect of polychlorinated biphenyl formulations on the
growth of estuarine bacteria: Applied Microbiology, v. 29, p. 125.
Bower, C. A., and F. B. Gschwend, 1952, Ethylene glycol retention by soils as a measure of
surface area and interlaying swelling: Soil Science Society of America, Proceedings,
v. 16, p. 342-345.
Briggs, G. G., 1973, A simple relationship between soil adsorption of organic chemicals and
their octenol/water partition coefficients: Proceedings 7th British Insecticide and
Fungicide Conference, British Crop Protection Council, London WIN6DT, p. 83-86.
Burkard, L. P., and D. E. Armstrong, 1978, A modified perchlorination procedure for the
determination of polychlorinated biphenyls: Technical Completion Report, Project
No. OWRT A-067-Wis. Water Resources Center, University of Wisconsin, Madison, Wisconsin.
Burse, V. W., R. D. Kimbrough, E. C. Villanueva, R. W. Jennings, R. E. Linder, and G. W. Sovocoal,
1974, Polychlorinated biphenyls storage, distribution, and recovery: Liver Morphology
After Prolonged Dietary Ingestion: Archives of Environmental Health, v. 29, p. 301.
Catelani, D., A. Colombi, C. Sorlini, and V. Treccani, 1973, 2-hydroxy-6-oxo-6-phenylhexa-2,
4-dienoate: The meta-clevage produce from 2, 3-dihydroxybiphenyl by Pseudomonas put-ida:
Journal of Biochemistry, v. 134, p. 1063-1066.
Catelani, D., C." Sorlini, and'V7"freccan'i'," 1971, The metabolism of biphenyl by Pseudomonas
put-ida: Journal of Biochemistry, v. 134, p. 1063-1066.
Catelani, D., C. Sorlini, and V. Treccani, 1971, The metabolism of biphenyl by Pseudomonas
put-ida: Experientia, v. 27, p. 1173-1174.
Chian, E.S.K., 1977, Stability of organic matter in landfill leachates: Water Research,
v. 11, p. 225-232.
Cook, J. W., 1972, Some chemical aspects of polychlorinated biphenyls (PCBs): Environmental
Health Perspective, v. 1:3.
87
-------
Curley, A., V. W. Burse, M. E. Grim, R. W. Jennings, and R. E. Linder, 1971, Polychlorinated
biphenyls: Distribution and storage in body fluids and tissues of Sherman rats-
Environmental Research, v. 4, p. 481.
Curley,_A., V. W. Burse, R. W. Jennings, E. C. Villanueva, and R. D. Kimbrough, 1975,
Evidence of tetrachlorodibenzofuran (TCDF) in Aroclor 1254 and the urine of rats following
dietary exposure to Aroclor 1254: Bulletin of Environmental Contamination and Toxicology,
V. IT") P « I JO lOO.
Duke,J W., J. J. Lowe, and A. J. Wilson, Jr., 1970, A polychlorinated biphenyl (Aroclor 1254)
in the water, sediment, and biota of Escambi Bay, Florida: Bulletin of Environmental
Contamination and Toxicology, v. 5, no. 2, p. 171.
Duvall D. S., and W. A. Rubey, 1977, Laboratory evaluation of high-temperature destruction
of polychlonnated biphenyls and related compounds: U.S. Environmental Protection Agency
Report 600/2-77/228, Cincinnati, OH 45268.
EHRC, 1976, Polychlorinated biphenyls: Health effects and recommendations: Environmental
Health Resource Center, Illinois Institute for Environmental Duality, IIEQ Document No
76/05, Chicago, Illinois.
Eichelberger, J., 1971, Persistence of Aroclors in surface waters: PCB Newsletter, no. 3,
ES&T, 1977, Outlook, legislation: The toxic substances control act is effective this month-
Environmental Science and Technology, v. 11, p. 28.
Furukawa, K., and F. Matsumura, 1976, Microbial Metabolism of polychlorinated biphenyls-
Studies on the relative degradability of polychlorinated biphenyl components by
AlkaUgenes Sp.: Journal of Agricultural and Food Chemistry, v. 24, p. 251.
Furukawa, K., K. Tonomura, and A. Kamibayashi, 1978, Effect of chlorine substitution on the
biodegradability of polychlorinated biphenyls: Applied Environmental Microbiology, v. 35,
Gibson, D. J., R. L. Roberts, M. C. Wells, and V. M. Kobal, 1973, Oxidation of biphenyl by
n * B*tJ«™°*t« species: Biochemical and Biophysical Research Communications, v. 50,' p. 211.
uuskoter H J., 1965, Electronic low temperature ashing of bituminous coal: Fuel, v. 44,
Grant D L. W E.J. Phillips, and D. C. Villeneuve, 1971, Metabolism of a polychlorinated
biphenyl (Aroclor 1254) mixture in the rat: Bulletin of Environmental Contamination and
Toxicology, v. 6, no. 2, p. 102.
Gray P.H.H and H. G. Thornton, 1928, Soil bacteria that decompose certain aromatic compounds:
Zentralblatt Bacteriologie Parasitenkunde, Abteilung II, v 73 p 74
Gresshoff P. M., H. K. Mahanty, and E. Gortner, 1977, Fate of polychlorinated biphenyl
lArocIor 1242) in an experimental study and its significance to the natural environment:
Bulletin of Environmental Contamination and Toxicology, v. 17, no. 6, p. 686
briffin R. A.,_R. Clark, M. Lee, and E.S.K. Chian, 1978, Disposal and removal of polychlorinated
biphenyls in soil, -in David Schultz [ed.], Land disposal of hazardous waste- U S
Environmental Protection Agency, Cincinnati, OH 45268, EPA-600/9-78-016, p 169-181
briffin, R. A., and N. F. Shimp, 1978, Attenuation of pollutants in municipal landfill leachate
by clay minerals: U.S. Environmental Protection Agency Publication EPA-600/2-78-157
Cincinnati, OH 45258, 147 p.
Grob, K., 1975, Glass capillary columns in gas chromatography, a tool and a technique-
Chromatographia, v. 8, p. 423.
Grob, K., G. Grob, and K. Grob, Jr., 1977, New improved carbonate deposition technique for
coating wall coated open tubular columns: Chromatographia, v. 10, p. 181.
Gustafson, C. G., 1970, PCBs—prevalent and persistent: Environmental Science and Technology
v. 4, no. 10, p. 814-819.
Gustafson, C. G., [ed.], 1972, Proceedings of Symposium on PCBs—still prevalent—Still
persistent: 164th American Chemical Society Meeting, New York, NY, September, 1972.
Hammer, D. I., J. F. Finklea, L. E. Priester, J. E. Keil, S. H. Sandifer and K. Bridbord, 1972,
Polychlorinated biphenyl in the plasma and hair of refuse workers: Environmental Health
Perspective, v. 1, p. 83.
Hammond, A. L., 1972, Chemical pollutioli: Polychlorinated biphenyls1: Science, v. 175, p. 155-156.
Hansen, D. J., P. R. Parrish, J. I. Lowe, A. J. Wilson, Jr., and P. D. Wilson, 1971, Chronic
toxicity, uptake, and retention of Aroclor 1254 in two estuarine fishes: Bulletin of
Environmental Contamination and Toxicology, v. 6, no. 2, p. 113-]19.
Haque, R., and D. Schmedding, 1975, A method of measuring the water solubility of hydrophobic
chemicals: Solubility of five polychlorinated biphenyls: Bulletin of Environmental
Contamination and Toxicology, v. 14, no. 1, p. 13-18.
Haque, R., D. W. Schmedding, and V. H. Freed, 1974, Aqueous solubility, adsorption, and vapor
behavior of polychlorinated biphenyl Aroclor 1254: Environmental Science and Technology,
v. 8, p. 139-142.
88
-------
Haque, R., and 0. Schmedding, 1976, Studies on the adsorption of selected polychlorinated
biphenyl isomers on several surfaces: Journal of Environmental Science and Health,
Harvey!1 G.Y, and w! G. Steinhauer, 1975, Biochemistry of PCB and DDT in the North Atlantic:
Proceedings of the 2nd International Conference on Environmental Biogeochemistry, Hamilton,
Canada, p. 203-225. , .. ... ,
Helling C S , and B. C. Turner, 1968, Pesticide mobility: Determination by soil thin-layer
HelliS:Ta19:a?^ia!CpesCticide2moKility1n3s0ils I. Parameters of thin-layer chromatography:
Soil Science Society of America Proceedings, v. 35, p. 732-737.
Helling,, C. S., 1971b, Pesticide mobility in soils II. Applications of soil thin-layer
chromatography: Soil Science Society of America Proceedings, v. Jb,p. /J/-/4J.
Helling, C. S , 1971c, Pesticide mobility in soils III. Influence of soil properties: Soil
Science Society of America Proceedings, v. 35, p. 743-747.
Henderson, H., 1978, The dilemma of Wilsonville: Illinois Times, v. 4, p. 4-7.
Herbicide Handbook, 1974, Weed Society of America, 3rd Ed., 425 Illinois Building, 113 N. Neil
St., Champaign, IL 61820. ,, ^
Herbst, E., I. Scheunert, W. Klein, and F. Korte, 1977, Fate of PCBs—14C in sewage treatment-
Laboratory experiments with activated sludge: Chemosphere, v. 6, p. 725.
Hesse J L , 1971, Summary of the Michigan Water Resources Commission PCB monitoring efforts:
Water Quality Control Division, Water Resources Commission, Bureau of Water Management,
Lansing, Michigan. . .. . . ,
HEW, Department of Health, Education, and Welfare, 1972, PCBs and the Environment: National
Technical Information Service, COM-72-10419.
Hoover, T. B., July, 1971, Water solubility of PCB isomers: PCB Newsletter, no. 3, p. 4-b.
Horvath, R. S., 1972, Microbiol co-metabolism and the degradation of organic compounds in
nature: Bacteriological Reviews, v. 36, p. 146.
Hughes, G. M., R. A. London, and R. N. Farvolden, 1971, Hydrogeology of solid waste disposal
sites in northeastern Illinois: U.S. Environmental Protection Agency, Solid Waste Manage-
ment Series, Report SW-12d., 154 p. ,Drae\
Hutzinger, 0., S. Safe, and V. Zitko, 1972, Photochemical degradation of chlorobiphenyls (PCBs).
Environmental Health Perspectives, April 15-20.
Hutzinger, D., S. Safe, and V. Zitko, 1974, The Chemistry of PCBs: CRC Press, Cleveland, Ohio.
Iwata, Y., F. A. Gunther, and W. E. Westlake, 1974, Uptake of a PCB (Aroclor 1254) from soil
by carrots under field conditions: Bulletin of Environmental Contamination and Toxicology,
v. 11, p. 523-528. . ^ , un . ^ ,
Iwata Y , W E. Westlake, and F. A. Gunther, 1973, Varying persistence of polychlonnated
biphenyls in six California soils under laboratory conditions: Bulletin of Environmental
Contamination and Toxicology, v. 9, no. 4, p. 204-211.
Jensen, S., 1966, Report of a new chemical hazard: New Scientist, v. 32, p. bl^.
Jensen, S. A., A. G. Johnels, M. Olsson, and G. Otter!ind, 1969, DDT and PCB in marine animals
from Swedish waters: Nature, v. 224, p. 247-250.
Jordan, D., 1977, The town dilemma: Environment, v. 19, p. 6-15.
Kaiser, K.L.E., and T. S. Wong, 1974, Bacterial degradation of polychlorinated biphenyls.
I. Identification of some metabolic products from AroclorK 1242: Bulletin of Environ-
mental Contamination and Toxicology, v. 11, p. 291.
Kaiser, R., 1976, Zur richtigen Messung und Bewertung von Gutekennzahlen in der Chromatographie:
Die reale Trennstutenzahl, die Tr inzahl, die Dosiergutekennzahl. Teil 1: Gas-
Chromatographie: Chromatographia, v. 9, p. 337.
Koeman, J. H., M. C. Ten Noever De Braun, and R. H. De Vos, 1969, Chlorinated Biphenyls in
fish, mussels and birds from the River Rhine and the Netherlands coastal area: Nature,
v. 221, p. 1126-1128.
Kuhn, J. K., F. Fiene, and R. Harvey, 1978, Geochemical evaluation and characterization of a
Pittsburgh No. 8 and a Rosebud Seam Coal: Report for DOE Research Contract EY-77-X-21-2155,
Morgantown Energy Technology Center, NTIS No. METC/CR-78/8.
Lawrence, J., and H. M. Tosine, 1976, Adsorption of polychlorinated biphenyls from aqueous
solutions and sewage: Environmental Science and Technology, v. 10, no. 4, p. 381-383.
Leis, W. M., W. F. Beers, J. M. Davidson, and G. D. Knowles, 1978, Migration of PCBs by ground-
water transport—A case study of twelve landfills and dredge disposal sites in the upper
Hudson Valley, New York: Conference of Applied Research and Practice on Municipal and
Industrial Waste, Box 5571, Madison, WI 53705, Sept. 10-13, 1978, p. 539-548.
Lee, M. C., R. A. Griffin, M. L. Miller, and E.S.K. Chian, 1979, Adsorption of water soluble
Aroclor 1242 and used capacitor fluid by soil materials and coal chars: Journal of
Environmental Science and Health, (In press).
89
-------
Lidgett, R. A., and H. A. Vodden, 1970, PCB—The environmental problem: PCB Conference,
Wenner-Gren Center, National Environmental Protection Board Stockholm, September 29, 1970,
p. 88-96.
Lloyd, J. W., R. M. Moore, B. S. Woolf, and H. P. Stein, 1976, Polychlorinated biphenyls:
Journal of Microbiology, v. 18, p. 109.
Lunt, 0., and W. C. Evans, 1970, The microbial metabolism of biphenyl: Journal of Biochemistry,
v. 118, 54 p.
MacKay, D., and A. W. Wolkoff, 1973, Rate of evaporation of low solubility contaminants from
water bodies to atmosphere: Environmental Science and Technology, v. 7, p. 611-614.
Maslansky, S. P., R. F. Thomas, R. F. Bonner, Jr., and P. L. Busch, 1978, Selection and design
criteria of secure landfill facilities for the containment of PCB contaminated river
debris: Conference on Applied Research and Practice on Municipal and Industrial Waste,
Box 5571, Madison, WI 53705, September 10-13, 1978, p. 635-644.
Mihashi, K., T. Hashinaya, M. Dazai, J. I to, and T. Misono, 1975, Biodegradability of poly-
chlorobiphenyl: Kogyo Gijustsuiu Biseibntsu Kogyo Gijustsu Kenkyusho Kenkyn Hokoku,
v. 46, p. 65.
Moon, D. K., I. W. Leighton, and D. A. Huebner, 1976, New England PCB waste management study,
New Bedford: Solid Waste Program, Air and Hazardous Materials Division, Region I,
U.S. - EPA, J.F.K. Federal Building, Boston, MA 02203.
Murado, M. A., C. Tejedor, and G. Baluja, 1976, Interactions between polychlorinated biphenyls
(PCBs) and soil microfungi. Effects of Aroclor 1254 and other PCBs on AspergUlus flavus
cultures: Bulletin of Environmental Contamination and Toxicology, v. 15, p. 768.
Nelson, N., P. B. Hammond, I.C.T. Nisbet, A. F. Sarofim, and W. H. Drury, 1972, Polychlorinated
biphenyls: Environmental impact: Environmental Research, v. 5, p. 249-362.
Nisbet, I.C.T.,' and A. F. Sarofim, 1972, Rates and routes of transport of PCBs in the environment:
Environmental Health Perspectives; no. 1, 21-38.
Omori, T., and M. Alexander, 1978, Bacterial dehalogenation of halogenated alkanes and fatty
acids: Applied Environmental Microbiology, v. 35, p. 867.
Paris, D. F., W. C. Steen, and G. L. Baughman, 1978, Role of physico-chemical properties of
Aroclors 1016 and 1242 in determining their fate and transport in aquatic environments:
Chemosphere, v. 4, p. 319-325.
Peakall, D. B., and J. L. Lincer, 1970, Polychlorinated biphenyls: Another long-life wide-
spread chemical in the environment: Bioscience, v. 20, p. 958-964.
Persson, B., 1971, Uptake of chlorinated hydrocarbons by Whitethroats Sylvia eommunis. aathl.
in areas sprayed with DDT: Ornis Scandinavica, v. 2, p. 127-135.
Reeves, M., C. W. Francis, J. 0. Duguid, 1977, Quantitative analysis of soil chromatography.
I. Water and radionuclide transport: Environmental Sciences Division Publication no. 1105,
Oak Ridge National Laboratory, Oak Ridge, TN 37830, 170 p.
Reinert, R. E., 1970, Pesticide concentrations in Great Lake fish: Pesticides Monitoring
Journal, v. 3, p. 233.
Reuter, J. H., and S. Havlicek, 1978, Private communications: October, Georgia Institute
of Technology.
Reynolds, L. M., 1969, Polychlorobiphenyls (PCBs) and their interference with pesticide
residue analysis: Bulletin of Environmental Contamination and Toxicology, v. 4, no. 3,
p. 128-143.
Reynolds, L. M., 1971, Pesticide residue analysis in the presence of polychlorobiphenyls
(PCBs): Residue Reviews, v. 34, p. 27-57.
Risebrough, R. W., P. Reiche, and H. S. Olcott, 1969, Current progress in the determination
of the polychlorinated biphenyls: Bulletin of Environmental Contamination and Toxicology,
v. 4, no. 4, p. 192-210.
Risebrough, R. W., P. Reiche, D. B. Peakall, S. G. Herman, and M. N. Kirven, 1968, Polychlor-
inated biphenyls in the global ecosystem: .Nature, v. 220, p. 1098.
Robertson, J. M., and E.C.C. Li, 1976, Organic leachate threatens ground water quality:
Water and Sewage Works, February 1976, p. 58-59.
Rote, J. W., and P. G. Murphy, 1971, A method for quantisation of polychlorinated biphenyl
(PCB) isomers: Bulletin of Environmental Contamination and Toxicology, v. 6, no. 4,
p. 377-384.
Ruzo, L. 0., M. J. Zabik, and R. D. Schiretz, 1972, Polychlorinated biphenyls: Photolysis
of 3, 4, 3', 4'-tetrachlorobiphenyl and 4, 4'-dichlorobiphenyl in solution: Bulletin
of Environmental Contamination Toxicology, v. 8, no. 4, p. 217-218.
Safe, S., and 0. Hutzinger, 1971, Polychlorinated biphenyls: Photolysis of 2, 4, 6, 2', 4', 6'-
hexachlorobiphenyl: Nature, v. 232, p. 641-642.
Sayler, G. S., M. Shou, and R. R. Colwell,.1977, Growth of an estuarine Pseudomonas sp. on
polychlorinated biphenyls: Microbial Ecology, v. 3, p. 241-245.
90
-------
Schact, R. A., 1974, Pesticides in the Illinois water of Lake Michigan: Ecological Research
Series, U. S. Environmental Protection Agency Publication EPA-660/3-74-002, Washington, D.C.
Scharpenseel, H. W., B.K.G. Theng, and S. Stephan, 1978, Polychlorinated biphenyls (^C)
in soils: Adsorption, infiltration, translocation, and decomposition, in Environmental
biochemistry and geomicrobiology, The terrestrial environment: Ann Arbor
Science Publishers, Inc., Ann Arbor, MI 48106, v. 2, p. 619-637.
Sissons, D., and D. Welti, 1971, Structural identification of polychlorinated biphenyls in
commercial mixtures by gas-liquid chromatography, nuclear magnetic resonance, and mass
spectrometry: Journal of Chromatography, v. 60, p. 16-32.
Skrentny, R. F., R. W. Hemken, and H. W. Dorongh, 1971, Silo sealants as a source of poly-
chlorobiphenyl (PBC) contamination of animal feed: Bulletin of Environmental Contamination
and Toxicology, v. 6, no. 5, p. 409-416.
Snyder, D., and R. Reinert, 1971, Rapid separation of polychlorinated biphenyls from DDT and
its analogues on silica gel: Bulletin of Environmental Contamination and Toxicology, v. 6,
no. 5, p. 358-390.
Solly, S.R.B., and V.'Shanks, 1974, Polychlorinated biphenyls and organochlorine pesticides
in human fat in New Zealand: New Auckland Journal Science, v. 7, p. 535-544.
Stalling, D. L., and J. N. Huckins, 1971, Gas-liquid chromatography-mass spectrometry character-
ization of polychlorinated biphenyls (Aroclors) and 36Cl-labeling of Aroclors 1248 and
1254: Journal of the Association of Official Analytical Chemists, v. 54, no. 4, p. 801-807.
Sundstrom, G., D. Hutzinger, and S. Safe, 1976, The metabolism of chlorobiphenyls: Chemosphere,
v. 5, p. 267.
Texas Instruments, 1975, Program manual STI statistics library, Model SR-52 calculator: Texas
Instruments, Inc., P.O. B.ox 22283, Dallas, TX 75222, 107 p.
Thomas, J., Jr., and R. R. Frost, 1971, Versatile apparatus for studying reactions involving
gas adsorption or evolution: Illinois Academy of Science Transactions, v. 64, p. 248-253.
Thurston, A., 1971, Quantitative analysis of PCBs: PCB Newsletter, July, no. 3, p. 4.
Troise, F., and V. M. Kahn, 1978, High cost of legal dumping leads some to organized crime:
The Ground Water Newsletter, Jan. 31, 1978, Water'Information Center, Inc., Huntington,
NY 11743, v. 7, no. 2.
Trotter, W. J., and S.J.V. Young, 1975, Limitation of the use of antimony pentachloride for
perchlorination of polychlorinated biphenyls: Journal of the Association of Official
Analytical Chemists, v. 58, p. 466-468.
Tucker, E. S., W. J. Litschgi, and W. M. Mees, 1975, Migration of polychlorinated biphenyls
in soil induced by percolating water: Bulletin of Environmental Contamination and
Toxicology, v. 13, p. 86-93.
Tucker, E. S., V. M. Saeger, and 0. Hicks, 1975, Activated sludge primary biodegradation of
polychlorinated biphenyls: Bulletin of Environmental Contamination and Toxicology,
v. 14, p. 705-713.
Wallnofer, P. R., G. Engelhardt, S. Safe, and 0. Hutzinger, 1973, Microbial hydroxylation of
4-chlorobiphenyl and 4-4'-dichlorobiphenyl: Chemosphere, v. 2, p. 69-76.
Webb, R. G., and A. C. McCall, 1972, Identities of polychlorinated biphenyl isomers in Aroclors:
Journal of the Association of Official Analytical Chemists, July, v. 55, no. 4, p. 746-752.
Widmark, G., 1967, Possible interference by chlorinated biphenyls: Journal of Association of
Official Analytical Chemists, v. 50, p. 1069.
Widmark, G., 1968, The OECD study of analysis of PCB: Report from the Institute of Analytical
Chemistry, Stockholm, Sweden.
Wiencke, W., and J. Roach, 1972, Response to GLC detectors to the chlorine content in Aroclors,
PCB Newsletter, March, no. 4, p. 9-10.
Zitko, V., 1970, Polychlorinated biphenyls (PCB) solubilized in water by monionic surfactants
for study of toxicity to aquatic animals: Bulletin of Environmental Contamination and
Toxicology, v. 5, no. 3, p. 279-285.
Zitko, V., 1971, Polychlorinated biphenyls and organochlorine pesticides in some freshwater
and marine fishes: Bulletin of Environmental Contamination and Toxicology, v. 6, no. 5,
p. 464-470.
Zitko, V., 0. Hutzinger, and S. Safe, 1971, Retention times and electron-capture detector
responses of some individual chlorobiphenyls: Bulletin of Environmental Contamination
and Toxicology, v. 6, no. 2, p. 160-163.
91
-------
APPENDIX
LIST OF PUBLICATIONS
Clark, R. R., E.S.K. Chian, and R. A. Griffin, 1979, Degradation of poly-
chlorinated biphenyls by mixed microbial cultures: Applied and Environ-
mental Microbiology, v. 37, p. 680-685.
Griffin, R. A., A. K. Au, and E.S.K. Chian, 1979, Mobility of polychlorinated
biphenyls and dicamba in soil materials: Determination by soil thin-
layer chromatography: Proceedings 8th National Conference—Municipal
and Industrial Sludge, March 19-21, 1979, Miami, FL. Copyright by
Information Transfer, Inc., Rockville, MD.
Griffin, R. A., and E.S.K. Chian, in press, Attenuation of water-soluble
polychlorinated biphenyls by earth materials: Illinois State Geological
Survey Environmental Geology Note 86, 98 p.
Griffin, R. A., Robert Clark, Michael Lee, and Edward Chian, 1978, Disposal
and removal of polychlorinated biphenyls in soil, in David Schultz [ed.],
Land disposal of hazardous waste: Publication EPA-600/9-78-016,
Cincinnati, OH 45268, p. 169-181.
Griffin, R. A., F. B. DeWalle, E.S.K. Chian, J. H. Kim, and A. K. Au, 1977,
Attenuation of PCBs by soil materials and char wastes, in
S. K. Banerji [ed.], Management of gas and leachate in landfills:
Publication EPA-600/9-77-026, U.S. Environmental Protection Agency,
Cincinnati, OH 45268, p. 208-217.
Lee, M. C., E.S.K. Chian, and R. A. Griffin, in press, Solubility of poly-
chlorinated biphenyls and capacitor fluid in water: Water Research.
Lee, M. C., R. A. Griffin, M. L. Miller, and E.S.K. Chian, in press, Adsorp-
tion of water soluble polychlorinated biphenyl Aroclor 1242-and used
capacitor fluid by soil materials and coal chars: Journal of Environ-
mental Science and Health.
Meng, Hsi, E.S.K. Chian, and R. A. Griffin, (submitted), Effect of humic acid
and soil on volatilization of polychlorinated biphenyls from water:
Water Research.
92
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-80-027
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Attenuation of Water-Soluble Polychlorinated Biphenyls
by Earth Materials
5. REPORT DATE .
May 1980 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
R. A. Griffin and E. S. K. Chian
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Illinois State Geological Survey
Urbana, Illinois 61801
10. PROGRAM ELEMENT NO.
C73D1C, SOS #1, Task 17
11. CONTRACT/GRANT NO.
R-804684-01-0
12, SPONSORIWG.AGENCY NAMEANDADDRESS . . , r. Au
Municipal Environmental Research Laboratory --Cm.,OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Richard A. Carnes, Project Officer (513/684-7871)
16. ABSTRACT
The aqueous solubility, adsorption, mobility, microbial degradation, and
volatility of polychlorinated biphenyls (PCBs) were studied under laboratory con-
ditions. The dissolution of Aroclor 1242 in water required five months to reach
equilibrium. Generally, the water-soluble fractions of the PCB fluids were richer
in the lower chlorinated isomers than in the original mixture of isomers in the fluid.
The solubilities of Aroclor 1016, 1221, 1242, and 1254 were 906 ppb, 703 ppb, and
=70 ppb, respectively.
A simple linear relation described the adsorption of water-soluble PCBs by five
earth materials and their low-temperature ashes. An adsorption constant (K) unique
to each adsorbent was obtained. The adsorption was strongly correlated to the total
organic carbon (TOC) content and surface area of the earth materials. TOC was the
dominant of these two earth material properties by a ratio greater than three to
one.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Biodeterioration
Mobility
Solubility
Hazardous Wastes
68C
13. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
101
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
93
A U.S. GOVERNMENT PRINTING OFFICE: 1980-657-146/5700
------- |