United States
Environmental Protection
Agency
Robert S. Kerr Environmental
Research Laboratory
Ada, OK 74820
Research and Development
EPA/600/S2-90/040 Sept. 1990
&EPA Project Summary
Abiotic Reductive
Dechlorination of Carbon
Tetrachlorideand
Hexachloroethane by
Environmental Reductants
Martin Reinhard, Gary P. Curtis, and Michelle R. Kriegman
The transformation rates of
hexachloroethane (HCA) and carbon
tetrachloride (CTET) have been
measured in model systems
representing the ground water
environment and in slurries of
fractionated Borden aquifer material.
This report summarizes research
conducted to identify the
environmental factors which affect
the abiotic (chemical) transformation
rates of HCA and CTET in systems
consisting of minerals and Borden
aquifer material under both aerobic
and anaerobic conditions The
mineral systems studied consisted of
both homogeneous solutions
containing soluble environmental
reductants and heterogeneous
systems containing well
characterized solids representative of
mineral phases in aquifers. The
following soluble reductants were
used: bisulfide (HS), L-cysteine, and
ferrous iron complexes. Reaction
rates were compared with those
measured in heterogeneous systems
containing biotite, vermiculite, pyrite,
marcasite, or 13X zeolite and one of
the dissolved reductants.
Heterogeneous reaction rates were at
least an order of magnitude faster
than the homogeneous rates. The
reductive dechlorinat ion of
hexachloroethane (HCA) to form
tetrachloroethene (PCE) has been
studied in the presence of aquifer
material excavated from the aerobic
sandy Borden aquifer. Studies with
magnetically separated fractions of
the aquifer material indicated that the
fraction consisting of quartz,
feldspars, and carbonates accounted
for most of the reactivity. Studies
with the acid and base pretreated
quartz, feldspar, and carbonates
fraction suggested that the electron
donor was associated with organic
matter in the aquifer solids. The
addition of 20 mg/l humic acids to 0.5
mN ferrous sulfate or 0.5 mM sodium
sulfide increased the rate of the
reduction reaction by a factor of 15
and 7, respectively. These results
suggest that the abiotic reduction
reactions may be closely coupled to
microbially produced reductants
such as ferrous iron and sulfide.
This Project Summary was
developed by EPA's Robert S. Kerr
Environmental Research Laboratory,
Ada, OK to announce key findings of
the research project that is fully
documented in a separate report of
the same title (see Project Report
ordering information at back).
Introduction
Reductive transformations are
important processes that influence the
fate and mobility of certain organic
pollutants in aqueous environments. In
particular, in sediment-water systems,
halogenated aliphatic compounds may be
reductively dehalogenated to form
products that may be more mobile in the
environment and of either greater or
lesser environmental concern. The
electron donors that have been
speculated to participate in these
-------
reduction reactions in sediment-water
systems include microorganisms,
reduced iron complexes such as iron
porphyrins, reduced iron minerals and
hydroquinone functional groups in humic
acid. However, the electron donor in
these reduction reactions has not been
identified.
The report is focused on reactions with
carbon tetrachloride (CTET) and
hexachloroethane (HCA), two relatively
reactive substrates. HCA was selected as
a model compound to investigate the
properties of the electron donor of an
aquifer material with respect to the
reductive dechlorination because it reacts
quantitatively to perchloroethene (PCE).
CTET transformation was studied in
model systems because it is a common
environmental contaminant and is
susceptible to a wider range of
transformations. The purpose of this work
was to further characterize the abiotic
reduction of chlorinated hydrocarbons
and to investigate the nature of the
electron donor in natural sediment
systems. The 74-125 nm size fraction of
Borden sand was used as the natural
aquifer material. In addition, the reactivity
of possible electron donors was studied
in heterogeneous model systems
consisting of zeolites, sheet silicates, and
iron sulfides.
The reductive transformation of HCA
can be described by the following
reaction:
26
24
Two electrons per mole of HCA are
required for this half reaction. CTE:T can
undergo hydrogenolysis to form
chloroform according to the reaction:
-CCL+-H+ +
2 4 2
e~=-CHCL+-Cr (2)
2 3 2 ( '
where the proton is transferred from
proton donors such as aqueous H20 or
surface coordinated H2O. In aquatic
environments, the electron donors
required to balance such half reactions
typically involve the elements carbon,
nitrogen, oxygen, sulfur, iron or
manganese. For Borden aquifer material,
which was used in this study, it was
assumed that the electron donor was
associated with the solid aquifer material.
Soluble reduced forms of iron,
manganese, sulfur or nitrogen were not
detected in ground water from the
Borden site. The dissolved organic
carbon content was less than 0.7mg/L
Calculations based on published
reduction potentials or thermodynamic
data showed that HCA and CTET should
be spontaneously reduced by a range of
environmental reductants including
ferrous iron and manganese (II) minerals,
humic and fulvic acids, and hydrogen
sulfide. A table of redox potentials and
their references are contained in the
report. Some potentially important
sources of ferrous iron in an aerobic
aquifer include iron-bearing silicates such
as biotite, hornblende, and clay minerals
or iron oxides such as ilmenite or
magnetite. In humic and fulvic acids, the
nature of the reducing site is poorly
understood due to the structural
complexity and heterogeneity of these
materials. It has been proposed that the
electron transfer site involves functional
groups such as hydroquinones or the
semiquinone radicals where the
semiquinone radicals are the
intermediates in the reversible oxidation
and reduction with quinone groups
present in the humiC acid.
Transformation studies with natural
reference minerals, such as biotite,
vermiculite, pyrite, and marcasite were
conducted to see if ferrous iron in
minerals could act as an electron donor
to reduce the HCA and CTET. Because it
would be difficult to distinguish surface
transformation and mineral dissolution
with subsequent transformation, systems
with a synthetic mineral, 13X zeolite,
were studied. 13X zeolite is an
aluminosilicate with a three-dimensional
cagelike pore structure which can readily
exchange divalent cations, such as
ferrous iron Its hydrophilic pores allow
the exchange of iron while sorbing
hydrophobic haloaliphatics only weakly.
This system was used as a model for
ferrous iron bearing silicates, such as
biotite, which contain Fe2+ within an
aluminosilicate lattice. Zeolite
experiments with ferrous iron and thiol
group containing reductants allowed
investigation of surface catalyzed
transformations of haloaliphatics.
Experimental Methods
The work with Borden sand focused on
a single size fraction with a nominal
diameter range of 74 - 125 urn (U.S.
Mesh 120-200). In previous studies this
fraction was observed to react rapidly
relative to other size fractions. The 74 -
125 nm fraction was further separated
nondestructively with a magnetic mineral
separator. Magnetic mineral separation
separates minerals on the basis of their
magnetic susceptibility which correlates
with the ferrous iron content of the
mineral. Experimental results with these
fractions were compared with results
obtained with reference minerals. The
most reactive fraction obtained by
magnetic separation was treated with
acid and base to investigate the
properties of the electron donor with
respect to treatment with acid and base.
Finally, the influence of added reducing
agents on the transformation rate of HCA
was investigated.
All transformation experiments were
conducted in flame-sealed glass ampules
to prevent losses of the haloaliphatics
and entrance of oxygen to the
experimental systems over the time scale
of the experiments (weeks to months). To
minimize oxidation in the autoclave, the
sediment or mineral-filled ampules were
sterilized under a nitrogen atmosphere.
The procedure for autoclaving under
nitrogen is explained in the report. After
autoclaving, ampules were placed in a
glove box with a 90%N2/10%H2
atmosphere and were filled with a
synthetic ground water or Tris buffer that
was cold-sterilized with 0.2 nm
polysulfone membrane filter. In the case
of humic acids, an aliquot of the pH 3
humic acid extract was neutralized to pH
6.0 with NaOH and diluted with a 2.5 mM
pH 5.8 phthalate buffer to give a final
concentration of 20 and 40 mg/l of
dissolved organic carbon (DOC). The
solution was filter sterilized with a 0.2 nm
nylon filter and then directly added to
autoclaved ampules.
Sealed ampules were held in the dark
in 50°C (±0.1°C) water bath for the
duration of the experiment. Samples were
regularly removed from the bath,
manually mixed and returned to the bath
within five minutes. At each sampling
time, duplicate ampules for each
experimental condition were removed
from the constant temperature bath,
centrifuged, extracted and analyzed for
CTET or HCA and their products using a
gas chromatograph equipped with an
electron capture detector.
Results
Mineral Studies
Table 1 summarizes the overall rate
constants (k'obs) and calculated half-lives
for the disappearance of HCA in
homogeneous solution with and without 4
mM HS"; and in heterogeneous systems
with biotite and vermiculite, with and
-------
without HS" added. The experiments
were conducted at 50°C and pH 7.5 - 8.
In the absence of HS", HCA slowly
reacted with biotite and vermiculite to
form PCE. In the presence of HS" the
transformation rate was greatly increased
(Figure 1). Similarly, Table 1 also
includes k'obs at 50°C for CTET
disappearance in homogeneous solution
with 4 mM HS" and in heterogeneous
systems with HS and biotite or
vermiculite. Due to the slow rate of
transformation of HCA at 50°C in water
and in mineral systems with biotite and
vermiculite (no bisulfide), the analogous
rates for CTET were not measured. In
experiments conducted with pyrite and
marcasite, CTET was transformed very
rapidly, indicating that clean iron sulfide
surfaces are very reactive toward
haloaliphatic compounds. These
reactions support a heterogeneous
mechanism because iron sulficles are
extremely insoluble (with the pKso=18.1
for FeS). In bisulfide (Table 1) or ferrous
iron (Table 2) solutions, CTET reacted
much slower than in the pyrite or
marcasite systems, but faster than in
aqueous solution containing no reductant.
In all cases, only 10-15% of the CTET
reacted to form chloroform. These data
clearly demonstrate that bisulfide in
conjunction with solid surfaces may
significantly increase the transformation
rate of HCA and CTET.
To investigate the effect of an
aluminosilicate structure with and without
Fe2 + , studies with 13X zeolite were
conducted. In Table 2, disappearance
rates of CTET and H", cysteine, or ferrous
iron in homogeneous solution were
compared to the same systems with 13X
zeolite added. The iron concentration
varied from 10-4M to 10-3M. In the
homogeneous case where the total iron
concentration was 1 mM, an iron
oxyhydroxide precipitate formed. In the
presence of the zeolite or when [Fe2 + ]aq
= 10'4M, no precipitation was observed.
As shown in Table 2, the system with the
iron oxyhydroxide precipitate reacted
much faster than the zeolite/iron systems
or the 0.1 mM Fe(ll) system.
Interestingly, when an iron oxyhydroxide
precipitate was present, the reaction with
CTET was very fast (a half-life of 2 days).
These studies suggest that a ferrous iron
precipitate can act as an electron donor.
However, the 13X zeolite structure
apparently inhibited electron transfer
from F e 2 + sorbed within the
aluminosilicate structure to the organic
substrate.
To test the reactivity of thio-
compounds in the presence of mineral
surfaces, experiments were conducted
with bisulfide or cysteine and 13X zeolite.
The reaction rate in cysteine/CTET
systems was much faster in the presence
of zeolite than in homogeneous solution.
For the reactions of CTET with bisulfide
and 13X zeolite, the presence of zeolite
greatly enhanced the reaction rate, as
observed with cysteine. The
transformation rate with cysteine was
faster than HS" (even though the initial
sulfide concentration was slightly higher
than the cysteine concentration), possibly
due to differences in the complexation of
cysteine and HS" with zeolite. These
experiments suggest that mineral
surfaces can act as catalysts for electron
transfer reactions between haloaliphatics
and environmental reductants (such as
bisulfide), even if the mineral surfaces
themselves are not redox active.
Aquifer Material Studies
Bore/en Sand Fractions
The fractions isolated by the Frantz
magnetic separator are listed in Table 3
along with the extractable ferrous iron,
the total extractable manganese and the
organic carbon fraction (foc). The results
show that the mineral fraction consisting
of quartz, feldspars and carbonates
accounted for the majority of
transformation (data not shown). It is also
clear from Table 3 that this fraction
accounts for neariy 75% w/w of the
unfractionated parent sample and also
contains the majority of the extractable
ferrous iron, total manganese and organic
carbon. Therefore, the reactivity is in the
fraction containing the highest
concentration of the potential redox active
elements but it is not possible to suggest
which element plays the most significant
role.
The fractions that consisted primarily of
the iron bearing silicates and oxides
isolated from the aquifer material,
showed only a small capability to
transform the HCA to PCE with the
possible exception of the fraction
containing predominantly ilmenite. This
low relative reactivity can be attributed to
the small amount of each of the individual
fractions used and possibly due to a low
specific surface area.
Acid and Base Extracted
Fractions
The sample containing the quartz,
carbonate, and feldspars was further
investigated by conducting transformation
studies after pretreating the minerals with
either strong acid to dissolve the
carbonate minerals or strong base to
extract the humic acids. Each of the
treated fractions was analyzed for
extractable Fe2 + and for the foc. As
expected, the extractable iron was
considerably depleted in the two samples
pretreated with acid while the base
pretreatment decreased the foc of the
fractions. The foc was reduced slightly by
the base extraction and was considerably
increased if the base extraction was
preceded by the acid pretreatment.
Presumably the organic matter that was
incorporated in carbonaceous rock
fragments became soluble after removal
of the carbonaceous rock matrix.
The capability of the quartz, carbonate
and feldspar fraction to reduce HCA to
PCE was tested after carbonate
dissolution with 5N HCI, humic acid
extraction with 1N NaOH, and humic acid
reprecipitation after carbonate dissolution.
The data indicate that the electron donor
is insoluble in acid and soluble in base
which corresponds closely with the
solubility behavior of humic acids with
respect to pH. The amount of PCE
formed was the greatest in the fraction
that contained the largest foc and was the
least in the sample with the least foc.
Conversely, the amount of PCE formed
was inversely related to the extractable
iron concentrations. Therefore, these
results suggest the participation of humic
acids in the reduction of HCA.
Humic Acid Systems
The results of the transformation
studies in the presence of humic acids
are summarized in Table 4. In all cases,
the total mass of HCA that disappeared
could be accounted for as PCE with the
exception of the samples that contained
the humics plus the sulfide where
approximately 10 percent of the added
haloaliphatic could not be accounted for
after two half-lives. For samples
contaming HS", the data used in the
regression were limited to samples from
the first half-life of the reaction where a
mass balance was observed.
The observed rate constants are
summarized in Table 4. No
disappearance of HCA nor formation of
PCE was observed when the humic acids
extracted from the aquifer materials were
investigated in the absence of an added
reductant. This experiment was repeated
with double the initial concentration of the
humic acids and with Suwannee River
humic acids obtained from the IHSS. In
all three cases no transformation of HCA
was detected after 10 days as indicated
by the rate constants reported in Table 4
-------
Table 1 Observed Rate Constants for HCA and CTET
Disappearance Temp = 50° C, pH = 7.7,
[Tris(hydroxymethyl)aminomethane (Tris buffer)] =0.01 M,
[Na2S*9H2O]=4mM, [CTET]initial = 1 pm, [HCAJinitial =
Experiment Name
(days-1)
IT I2 (days)3
HCA
HCA, Biotiteb
HCA, Vermiculiteb
HCA, HS
HCA, Biotite, HS
HCA, Vermiculite, HS
CTET
CTET, HS
CTET, Biotite, HS
CTET, Vermiculite, HS
CTET, Pyr/tec
CTET, Marcasited
0.0019
0.0036
0.0121
0.0463
1.0614
1.531
0.0018
0.0102
0.1541
0.2409
1.586
0.8185
290-520
150-270
41-98
9.4-37
0.49-0.70
0.38-0.57
380
69-1600
4.2-4.9
2.7-3.0
0.17-0.71
0.69-1.0
395% Confidence Interval around k'obs.
bSolids concentration for biotite and vermiculite experiments was
38.5 g/L.
cSolids concentration for pyrite experiment was 18.1 g/L.
dSolids concentration for marcasite experiment was 17.3 g/L.
o
§
O
.1 •"
.01-
.001
•a-
•F
A
-a— Exp. HS
--O-- Exp. HBS
-A - Exp. HVS
Time (Days)
Figure 1. HCA transformation in the presence of HS-, biotite (38.5 g/L) and vermiculite
(38.5 g/L) at 50°C where H = HCA, S=bisulfide, B=biotite, V=vermiculite.
that are statistically indistinguishable from
zero.
When equal molar amounts of the
various reductants were added to humic
acid solutions containing 20 mg/l DOC,
the conversion of HCA to PCE depended
strongly on the reductant added. As a
control, the transformation of HCA to PCE
by the reductants was studied without
humic acid added. The reductants were
added in excess quantities relative to the
initial HCA concentration, and the
-------
Table 2. Observed Rate Constants for Disappearance of CTET
(Temp=50°C, [Tris]=0.01 M, [Fe(ll)]total =0.1 or 1 mM, [L-
cysteinel = 1 mM, [Na2S»9H2O] =2-4 mM, lCTETJinitia,= 1 pM
Experiment Name
(days-1)
t1/2 (days)
CTET, 0.1 mM Fe(!l)
CTET, Zeol., O.imM Fe(ll)
CTET, 1mM Fe(ll)(ppt)b
CTET, Zeol., 1mM Fe(ll)c
CTET, Cysteme
CTET, Zeolite, Cysteme
CTET, 4 mM HS
CTET, Zeolite, 2 mM HS
0.0066
0.0026
0.3023
0.0014
0.0108
0.1375
0.0052
0.0389
69-216
81 -a*
2.0-2.6
250-ooa
45-114
4.1-5.2
69-1600
13-26
aUpper limit cf the half-life is infinity because the 95% confidence interval
around k'obs encompasses zero.
blron oxyhydroxide precipitate.
cSolids concentration for zeolite experiments was 9.5 g/L.
Table 3. Characteristics of Fractions Isolated from the 74-125 iim Size Fraction of the Borden Aquifer Material
Mineralogy
Quaru, Feldspars & Carbonates
Ferrodolomite
Hornblende
llmenite
Garnet
Magnetite
Calculated Total
Unfractionated Materials
Mass, %
74.6
1.5
15.3
4.7
3.3
0.3
100.
Extractable
Fe(ll) (fig/g)
184± 19
1808 ±50
64 + 17
79 ±13
62 ±5
61 ±7
180
203 ±15
Extractable Mn (T)
(VQlQ)
35 ±2
210±32
7 ±5
n.d.a
n.d.
12±6
30.3
34 ±5
foe
(% gC.'g)
0.013 ±0.001
0.073 ±0.032
0.012 ±0.001
0.011 ±0.001
0.006 ±0.005
0.020 ±0.007
0.014
0.015 ±0.002
aNot detected.
reactivity of the reductants increased with
the trend: Fe2+ < hydroxylamine
hydrochloride < HS". For hydroxylamine
hydrochloride, the addition of humic acid
had an insignificant effect on the
transformation rate, however. In contrast,
the humic-Fe2+ system reacted 15 times
faster than with Fe2+ alone and humic
acids increased the rate in the presence
of HS' by a factor of seven which
corresponds to half lives of 78 and 25
hours, respectively, as indicated in Table
4. The direct comparison between
humics with either Fe2 + or HS' is not
possible due to the different pH in these
two experiments but clearly the rate was
higher in both cases in the presence of
the humic acids. [Experiments with humic
acid plus 0.5 mM FeCI3 indicated that
only a small amount of reaction occurred
relative to the case of ferrous iron.
Although the ferric hydroxide precipitated
under the conditions of the experiments
the results nevertheless suggest that the
iron must be added in the reduced form
in order to form a reactive humic acid.
The kobs of the reaction between a 1
m M solution of reduced 2,6-
anthrahydroquinone-disulfonic acid
(AHQDS) and HCA at pH 6.8 is also listed
in Table 4. Apparently, the reduced
quinone groups of AHQDS, which are
assumed to model the significant redox
properties of humic acids, quantitatively
reduce HCA to PCE.
Discussion
Reduction by Minerals
In homogeneous solution there was an
absence of significant reaction between
the haloaliphatics and HS", suggesting
that the reaction with HS" is
heterogeneous Although the exact
mechanism is not known, the following
three processes may be considered:
2HS + 2Mf
1) HS" can regenerate Fe(ll) sites (which
may react with a sorbed haloaliphatic
substrate), as shown in the reaction:
Jmineral "~
* HSSH +
Jmineral*
where HSSH is a polysulfide which can
become further oxidized to sulfate and
M + is a cation removed from the solution
to maintain a charge balance;
2) HS" can transfer electrons to the
haloatiphatic via a surface reaction which
does not include ferrous iron according to
the reaction:
2[HS-]adsorbed + HCA ->
[HSSH]adsorbed + PCE+2Cl-;
and 3) HS" can react with ferrous iron
which has dissolved from the minerals to
form an iron sulfide precipitate. This
precipitate can act as an electron donor
-------
Table 4. Observed Rate Constants for the Transformation of HCA to PCE in Humic-Reductant Systems
System* w p/-/ % Recovered kobs(h-i)x 10'3
Humic Acids
Humic Acids (40 mg/l DOC)
IHSS Suwannee River
Humic Acids
NH2OH*HCI
Fe2 +
HS~
Humic Acids + NH2OH»HCL
Humic Acids + Fe2 +
Humic Acids + FeCI3
Humic Acids + HS~
2, 6-Anthrahydroqumone
disulfonate
7
6
8
6
7
7
5
7
10
10
6
6.0
5.9
5.6
5.4
5.8
8.7
5.4
5.8
5.6
8.7
6.7
100 ± 5
700 ± 1
102 ± 2
102 ± 6
98+4
93 ± 7
103 ± 4
95 ± 7
97 ± 4
90 ± 5
101 ± 7
-0.0005 ± 0.0008
0.0005 ± 0.0006
0.10 ± 0.05
1.7 ± 0.1
0.59 ± 0.05
3.74 ± 0.7
1.84 ± 0.03
8.92 ± 1.4
0.36 ± 0.05
27.0 ± 6.0
0.27 ± 0.02
0.01
0.002
0.52
0.99
0.97
0.85
0.99
0.90
0.88
0.76
0.97
aHumic acids used were extracted from the Borden aquifer material and were present at 20 mg/L except as noted. All
reductants and FeCI3 were present at 0.5 mM.
bN = the number of samples used in the regression.
as was shown for ferric iron. From the
results of the sheet silicate experiments
described above, the presence of biotite
or vermiculite, with or without bisulfide,
increases the degradation rate of HCA
and CTET. However, the data are
insufficient to show that the reactions are
truly heterogeneous because it is
conceivable that the minerals are
dissolving and the solutes formed are
reacting with the haloaliphatics in
solution.
The zeolite experiments were
conducted to investigate if the sheet
silicate reactions could be heterogeneous
and to determine which environmental
reductants are reactive in heterogeneous
systems. As indicated by the experiment
with the iron precipitate, ferrous iron can
act as an electron donor to transform
CTET. However, in the zeolite systems,
ferrous iron was not reactive. Because of
the hydrophilicity of the zeolite pores, it is
likely that the zeolite is excluding CTET
but sorbing Fe2+ within the zeolite
structure, thereby inhibiting the electron
transfer between Fe2+ and CTET.
Assuming the zeolites are an adequate
model of the sheet silicates, ferrous
iron/zeolite systems may have been
reactive over longer-time scales as was
evident in the biotite and vermiculite
systems in the absence of bisulfide.
Since the thiol groups, bisulfide and
cysteine, were both an order of
magnitude more reactive in the presence
of the zeolite surface than in solution, we
can conclude that the heterogeneous
transformation of CTET is more
favorable. Consequently, it is possible
that adsorbed bisullide could be the
electron donor in the sheet silicate
systems described above (process 2).
The reactivity of CTET and iron
sulfides was studied because sulfide
minerals are very common in reducing
environments. Both pyrite and marcasite
were much more reactive than either
ferrous iron or sulfide in homogeneous
solution. Due to the low solubility of iron
sulfides, the) disappearance of CTET
could not be accounted for in a
homogeneous reaction. These results
suggest that in the abovementioned
sulflde/sheet silicate systems, a reactive
iron sulfide precipitate might be forming
and reacting with CTET (process 3).
In summary, the results of the mineral
systems show that heterogeneous
transformation of haloaliphatics is much
more favorable than homogeneous
transformation. Once the temperature
dependence of the rates is determined,
the data can be extrapolated to
environmentally relevant temperatures. It
is likely that these rates will be significant
compared with time scales of ground
water transport.
Influence of Reductants on
Humic Acid
The humic acid extracted from the
aquifer solids exhibited the capability to
reduce HCA only after a reductant was
added to the system, and even then, the
reactivity depended on the reductant
added. HS" was the most effective in
forming a reactive humic acid followed by
Fe2 + while the hydroxy lamine
hydrochloride was ineffective as indicated
in Table 4. It is plausible to assume that
the Fe2 + or HS" reduces a reactive site in
the humic acid which in turn reduces
HCA more rapidly than the original
reductant. Due to the structural
complexity and heterogeneity of the
humic acids, there may be several
different types of reactive sites in humic
acids that could be capable of reducing a
halocarbon.
The observation that Fe2 + and the HS"
both increased the rate of reduction of
HCA, but NH2OH-HCI did not, is
consistent with the different reactions
between these three reagents and the
quinone groups in humic acid. Fe2+ in
aqueous solution rapidly reduces 1,4-
benzoquinone to hydroquinone. This
reduction of quinones to form
hydroquinones by Fe2+ has been used
to quantify the number of quinone groups
in humic acid. Sulfide also reduces
quinone groups in simple laboratory
systems, and it is reasonable to assume
that HS" in aqueous solution will also
reduce quinone groups in humic acid. In
contrast, hydroxylamine reacts with
carbonyl groups by nucleophilic addition
to form the corresponding oxime. This
reaction has been used to determine the
total carbonyl content, which includes
quinone groups plus the unconjugated
carbonyl groups, of humic acids.
From this evidence, it is inferred that
the hydroquinone groups in humic acids
could play a central role in the reduction
of HCA in these systems. This inference
is supported by our observation that HCA
is reduced in model systems by AHQDS.
If the hydroquinone groups in humic acid
are oxidized back to the original quinone
group by HCA, then the humic acids
would be acting as a catalyst for the
reduction of HCA by Fe2 + or HS". The
quinone/hydroquinone couple is
-------
reversibly oxidized and reduced in
coenzyme Q which participates in
electron transport system in many
biological systems and there is also
evidence that humic acids can be
reversibly oxidized and reduced as well.
The results from this work suggest that
humic acids may be playing the unusual
role in these systems as both a sorbent
and a reactant. The results in Table 4
clearly indicate that humic acids increase
the rate of reductive dehalogenation of
HCA by Fe2+ and HS . In addition,
organic matter associated with the aquifer
solids, which includes humic acids, is
widely accepted as being the dominant
sorbent for hydrophobic organic
compounds in aqueous solution.
Therefore, the rate of disappearance of
HCA in aquatic systems containing humic
acids is likely to be enhanced because
the humic acids (1) concentrate the HCA
in sorbed state and (2) accelerate the
reduction of HCA by Fe2 + and HS'.
The observation that both Fe2 + and
HS" were both reactive, particularly in the
presence of humic acids, is relevant to
environmental systems since both Fe2 +
and HS" are often microbially produced in
anoxic environments and therefore the
reduction of HCA could be directly
coupled to the microbial production of
reductants such as Fe2+ and HS". Unlike
typical biotransformations, where it is
often assumed that sorbed compounds
are unavailable for transformation,
transformations involving humics may be
favored by the concentrating effect of
sorption to organic matter. In ground
water environments with organic carbon
concentrations as low 0.03%, as for the
Borden aquifer, 80% of the HCA would
be sorbed to the solid. Since the rates of
sorption and desorption have been
reported to be very slow for compounds
similar to HCA, slow transformation
reactions in the sorbed state may
become significant relative to
microbiological transformations. This
coupled sorption/reduction reaction is
probably most significant for the highly
halogenated compounds since these
compounds sorb more strongly and are
expected to be reduced more rapidly.
Summary and Conclusions
Environmental factors which affect the
abiotic fate of chlorinated hydrocarbon
compounds under simulated ground
water conditions have been studied.
Specifically, the capability of fractionated
Borden aquifer material, humic acid, and
commercially available reference
minerals to promote the transformation of
HCA and CTET with and without
environmental reductants present was
investigated. HS", L-cysteine and Fe2 +
were used as environmental reductants.
Reference minerals studied included 13X
zeolite, sheet silicates (biotite and
vermiculite) and iron sulfides (pyrite and
marcasite).
In the studies involving reference
minerals, the rate of disappearance of
CTET and HCA was first order with
respect to substrate concentration. The
addition of sodium sulfide or Lcystelne to
the experimental systems made the
reactions much more reactive. The sheet
silicate surfaces increased the
transformation rate of CTET and HCA by
approximately an order of magnitude.
The iron sulfides, pyrite and marcasite,
increased the reaction rate with CTET by
two orders of magnitude above the rate
with bisulfide or ferrous iron in solution.
Although reaction rates were only
measured at 50°C, the rates were fast
enough such that when they are
extrapolated to environmentally relevant
temperatures, the transformation rates
will still be appreciable, relative to ground
water residence times.
For studying the reactivity of natural
aquifer material, HCA was used as a test
oxidant and magnetically isolated mineral
fractions of Borden aquifer material. The
fraction containing the majority of the
organic matter and extractable iron and
manganese was the most reactive
fraction. Other fractions containing iron
bearing minerals such as ilmenite, biotite,
hornblende and magnetite were also
reactive although to a lower degree.
Humic acid extracted from the sediment
was capable of reducing HCA only after a
reductant was added and even then the
reactivity depended on the reductant
added. Specifically, addition of sulfide or
ferrous iron increased the reactivity of the
humic acids whereas hydroxylamine had
no significant accelerating effect.
Based on experiments with AHQDS as
a model compound, it is proposed that
hydroquinone functions in the humic acid
are responsible for at least part of the
observed transformation of HCA. These
groups are apparently reduced by Fe2 +
and HS" and are, in turn, apparently
oxidized back to the quinones by the
HCA. Humic acids in soil systems are
suggested to play the role of a catalyst
as they alternate between the oxidized
and reduced states. The ultimate electron
donor in the Borden sediment remains
poorly understood, however. It is
hypothesized that Fe2+ present in the
aquifer is responsible for the reduction of
the humic acids which then reduce the
HCA. The finding that Fe2* and HS" both
increase the rate of the reductive
dechlorination of HCA mediated by humic
acids represents a direct link between
abiotic reactions and microbially
produced reductants.
Recommendations
This report shows that humic acid and
mineral surfaces can promote the
transformation of HCA and CTET, two
perchlorinated alkanes. Although the
study has been limited to these two
substrates and to only a few reducing
agents, it seems reasonable to propose
that the transformation rate of a broad
range of substrates may be accelerated
in these systems and that other
reductants, natural or synthetic, may be
effective too. Decontamination of soils
contaminated with chlorinated
hydrocarbon compounds using chemical
reductants could be feasible if the
reductants can be brought in contact with
the soil and if their activity within the soil
matrix could be maintained. However, the
transformation reactions must be
understood at the mechanistic level and
the effect of environmental variables on
rates and product formation must be
understood before efficient treatment
schemes can be devised.
Chemical treatment may have several
advantages over biological alternatives:
(1) biological processes are generally
more difficult to control than chemical
ones; (2) it may be applicable at high
concentrations which are toxic to
microorganisms; (3) it may not depend
on desorption of the contaminant which
tends to be slow for hydrophobic
compounds and which tends to be a rate
limiting factor for bioremediation.
Accordingly, the recommendations are to:
(1) develop detailed mechanistic models
for mineral surface promoted
transformations of chlorinated
hydrocarbon compounds;
(2) develop empirical rate laws for
reductive dechlorination reactions
with humic acids representative of
variable redox states; and
(3) study the effect of natural solid
matrices on the transformation rates.
Clearly, in most natural settings, a
complex interlay between chemical and
biological phenomena may occur. Once a
detailed picture of the chemical
processes is obtained, such phenomena
may be addressed.
-------
Martin Reinhard, Gary P. Curtis, and Michelle R. Kriegman, Stanford University,
Stanford, CA 94305-4020.
Stephen R. Hutchins is the EPA Project Officer (see below).
The complete report, entitled "Abiotic Reductive Dechlorination of Carbon
Tetrachloride and Hexachloroethane by Environmental Reductants,"
(Order No. PB 90-261 553/AS; Cost: $17.00, subject to change) will be
available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Robert S. Kerr Environmental Research Laboratory
U.S. Environmental Protection Agency
Ada, OK 74820
United States Center for Environmental Research
Environmental Protection Information
Agency Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S2-90/040
------- |