Microbial biodegradation and metabolite toxicity of three pyridinium-based cation ionic liquids

Article abstract of DOI:10.1039/b919154b

Merging the disciplines of green chemistry, ecotoxicology and ecology to develop environmentally-friendly industrial chemicals represents a significant collaborative challenge. This challenge can be met by extending already-informative standard toxicity and biodegradability assays to include further information about the potential persistence and biotransformation of pollutants in the environment. Development of ionic liquids (ILs) provides an ideal and proactive test system to determine several levels of environmental impact using academically interesting and industrially relevant green chemical prototypes. In this study, we investigated the biodegradability of three ILs, 1-butyl-3-methylpyridinium bromide, 1-hexyl-3-methylpyridinium bromide and 1-octyl-3-methylpyridinium bromide, by activated sludge microbial communities. We determined that all three ILs could be fully mineralized, but that only the octyl-substituted cation could be classified as readily biodegradable . We directly examined biodegradation products of the ILs using reverse-phase high performance liquid chromatography/mass spectrometry and MS/MS methods, and identified several unique preliminary degradation products. Finally, we determined that IL-biodegradation products were less toxic than the initial compound to a standard aquatic test organism, Daphnia magna, suggesting that biodegradation in an aquatic environment would decrease toxicity hazards associated with the initial compound. This study provides further information about pyridinium IL-biodegradation and guidelines to structure future IL design and research.

Full text of DOI:10.1039/b919154b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/greenchem | Green Chemistry
Microbial biodegradation and metabolite toxicity of three pyridinium-based
cation ionic liquids
Kathryn M. Docherty,*a Michelle V. Joyce,b Konrad J. Kulackic and Charles F. Kulpad
Received 15th September 2009, Accepted 21st January 2010
First published as an Advance Article on the web 25th February 2010
DOI: 10.1039/b919154b
Merging the disciplines of green chemistry, ecotoxicology and ecology to develop
environmentally-friendly industrial chemicals represents a significant collaborative challenge. This
challenge can be met by extending already-informative standard toxicity and biodegradability
assays to include further information about the potential persistence and biotransformation of
pollutants in the environment. Development of ionic liquids (ILs) provides an ideal and proactive
test system to determine several levels of environmental impact using academically interesting and
industrially relevant green chemical prototypes. In this study, we investigated the biodegradability
of three ILs, 1-butyl-3-methylpyridinium bromide, 1-hexyl-3-methylpyridinium bromide and
1-octyl-3-methylpyridinium bromide, by activated sludge microbial communities. We determined
that all three ILs could be fully mineralized, but that only the octyl-substituted cation could be
classified as “readily biodegradable”. We directly examined biodegradation products of the ILs
using reverse-phase high performance liquid chromatography/mass spectrometry and MS/MS
methods, and identified several unique preliminary degradation products. Finally, we determined
that IL-biodegradation products were less toxic than the initial compound to a standard aquatic
test organism, Daphnia magna, suggesting that biodegradation in an aquatic environment would
decrease toxicity hazards associated with the initial compound. This study provides further
information about pyridinium IL-biodegradation and guidelines to structure future IL design and
research.
break down into innocuous products that do not persist in the
Introduction
environment.2
It is generally acknowledged that many of our current tech-
Extending the bridge between the fields of green chemi-
nologies are unsustainable because their operation is incom-
cal engineering and (eco)toxicology has called for both an
patible with the long-term well-being of organisms and the
admirable amount of collaboration between investigators of
environment.1 Green chemistry, as a discipline, strives to surpass
both disciplines,1,3,4 as well as a marked reassessment of the
the present quality of life achieved by chemical advances, while
effectiveness and ecosystem-applicability of traditional eco-
using products and processes that are benign to human health
toxicological studies.5–8 General consensus is that in order to
and the environment.2 Specifically, this means using starting
perform a thorough ecological risk assessment (and therefore,
materials and processes that have practical chemical properties
fully classify a chemical as “green”), it is important to extend
(e.g., low explosivity, flammability, volatility and corrosivity),
as well as low acute and chronic toxicity, carcinogenicity
our toxicological analyses beyond the traditional single-species
experiments. While this means extending studies to include mul-
and mutagenicity, and do not persist or bioaccumulate in
tiple trophic levels, abiotic interactions and general “chemical
the environment. The field of green chemistry operates under
stress ecology”,5 it also suggests that testing and subsequent
“Twelve Principles”, two of which inevitably and inseparably
design for nontoxic biodegradation is a crucial part of proactive
link it to the field of ecotoxicology: Principle 4 – design chemicals
green chemical design.
for reduced toxicity while maintaining efficacy and function;
While it may be impossible to predict the effects a chemical
and Principle 10 – design chemicals for degradation which will
may have upon an entire ecosystem, it is important to address
how biodegradation of a chemical compound may affect the
accuracy of risk assessment model predictions based on stan-
aCenter for Ecology and Evolutionary Biology, 5289 University of
dard toxicity tests and chemical life cycle assessments. At least
Oregon, 307 Pacific Hall, Eugene, OR, 97405, USA.
partial chemical transformation (if not full degradation) of a
chemical compound is likely once a chemical is released into
a wastewater treatment plant or directly into the environment
itself. Whether through a variety of possible microbial bioactiv-
ities, photodegradation and other abiotic interactions, or chem-
ical reactions in the tissues of exposed organisms themselves,
chemical structures of pollutants are bound to change.9 These
E-mail: kathrynd@uoregon.edu; Fax: 541-346-2364; Tel: 541-346-5279
bDepartment of Chemistry and Biochemistry, University of Notre Dame,
Notre Dame, IN, 46556, USA. E-mail: mjoyce@nd.edu
cEcology, Evolution & Marine Biology, University of California, Santa
Barbara, Santa Barbara, CA, 93106-9610, USA.
E-mail: kulacki@lifesci.ucsb.edu
dDepartment of Biological Sciences, University of Notre Dame, Notre
Dame, IN, 46556, USA. E-mail: kulpa.1@nd.edu
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 701–712 | 701
©

View Article Online
changes often result in a structure that is less hazardous than
the original pollutant and that can easily be converted into
nontoxic biomass and CO2.9,10 However, sometimes chemicals
are degraded into structures that are more toxic, persistent
or transportable, and therefore more harmful, than the initial
compound (e.g., polybrominated diphenyl ethers (PBDEs) are
converted to more-toxic debromination products;11 polycyclic
aromatic hydrocarbons (PAHs) are oxidized to epoxides which
can covalently bind to DNA and result in mutations9,12).
Therefore, it is imperative to investigate structures and toxicities
of possible biodegradation products, as well as the impact those
products may have beyond the parent chemical, particularly
when designing new chemicals that have a wide range of
applicability and inevitable environmental release.
Room-temperature ionic liquids (ILs) provide an ideal test
system for proactive risk-assessment investigation. Ionic liquids
are a class of salts that are liquid at room temperature,
and have become a major focus of proactive collaborative
research between the fields of green chemical engineering and
(eco)toxicology because of their desirable chemical properties,
broad industrial applicability, and potential for academic study.
Ionic liquids combine a bulky organic cation, which can be
substituted with multiple functional groups, with a small,
often halogenated, anion. Given the numerous combinations of
cations and anions, ILs can be generated to serve a wide variety
of applications. For example, imidazolium and pyridinium-
cation based ILs have been academically studied as electrolytes,13
in acylation, Friedel–Crafts, Diels–Alder, Heck and redox
reactions e.g. 14–19 as mobile phase additives for reverse-phase high-
performance liquid chromatography (RP-HPLC),20 as well as
many other applications. Several industrial applications involv-
ing ILs are in the early or pre-commercial stages. BASF has
researched a commercial extractive-distillation and chlorination
processes using ILs.e.g. 21,22 Degussa (Germany) is interested in ILs
for a variety of processes, including antifoaming and dispersing
agents and emulsifiers.21 The French Petroleum Institute has
commercialized the use of ILs as solvents involved in butane
dimerization to octane.22
Interestingly, the reverse trend is true for biodegradability –
the longer the substituted alkyl chain, the more easily the
compound can be broken down by activated sludge microbial
communities.29–31 Several studies have shown that imidazolium-
31–33
and quaternary ammonium-based34 ILs tend to resist
biodegradation, thus reducing their potential environmental
safety. Recent studies have shown that ILs containing 1-
alkyl-3-methylpyridinium cations can be classified as “readily
biodegradable” by standard tests.29,31,35,36,37 Previous studies
suggest that the type of ring substitution (i.e. different alkyl
chain lengths29,31 or alkyl vs. ester group35,37) can serve as a
useful metric for determining IL biodegradability from chemical
structure, while the number of alkyl substitutions around the
ring may be less informative.37,38 Additionally, the identity of the
anion is also strongly related to pyridinium biodegradability (i.e.
pyridiniums with octylsulfate anions are more biodegradable
than those with halide anions.37) While past studies indicate
that pyridinium cations might be “greener” structures for
future IL-design, few studies have examined pyridinium-IL
biodegradation products31,36 or the potential toxicity of these
products.
In this study, we re-examine the biodegradability of three py-
ridinium-IL compounds: 1-butyl-3-methylpyridinium bromide
(bmpyrBr), 1-hexyl-3-methylpyridinium bromide (hmpyrBr)
and 1-octyl-3-methylpyridinium bromide (ompyrBr), using a
standard biodegradability assay. The structures for these ILs
are presented in Fig. 1. Our results agree that bmpyrBr can
be fully biodegraded (as previously shown by Pham et al.36).
We also show, as per our previous results, that hmpyrBr
Given this broad range of applicability and industrial interest,
many studies have examined the toxicity of ILs prior to their
release into the environment. Tests that investigate toxicity
to aquatic organisms are of particular interest because ILs
are generally water-soluble, and therefore aquatic ecosystems
are likely to receive and transport ILs that are released into
the environment. A chemical life cycle assessment of the IL
1-butyl-3-methylimidazolium tetrafluoroborate has shown that
it is potentially more efficient and recyclable than traditional
solvents.22 However, when released into the environment, this
class of IL could have a significant impact on both aquatic
and terrestrial environments, water quality, and human health.22
As recently reviewed by Kulacki et al.4 and emphasized by
Stolte et al.,23 ILs (regardless of the cation identity) follow the
general trend of increasing toxicity to aquatic test organisms
directly correlated with increases in hydrophobicity, which is
strongly related to the length of the substituted alkyl chain.
While unrelated to mutagenicity,24 changes in alkyl chain length
are also related to sub-lethal fitness and food web interaction
patterns such as aquatic organism grazing and periphyton
production25–27 and filter feeding.28
Fig.
1 Chemical structures for 3 test compounds: 1-butyl-3-
methylpyridinium bromide, 1-hexyl-3-methylpyridinium bromide and
1-octyl-3-methylpyridinium bromide.
702 | Green Chem., 2010, 12, 701–712
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
and ompyrBr can be fully biodegraded.29 However, of the
three ILs tested, only ompyrBr can be classified as “readily
biodegradable” according to standard test methods. To further
aid in prediction of environmental impact and fate of ILs, we
examine the degradation products of each of these ILs using
direct chemical analyses throughout the biodegradation process.
We indicate that unique metabolic products were formed during
the degradation of all three ILs, suggesting that several metabolic
degradation pathways may be possible for pyridinium-cation
ILs. Finally, we determine that the biodegradation products of
these three ILs are significantly less toxic to an aquatic test
organism after cleavage of the pyridinium ring is achieved. These
data provide us with further evidence that pyridinium-based
ILs may be rendered environmentally benign to aquatic systems
through biodegradation by microorganisms. This study is meant
to extend standard biodegradability and toxicity tests to aid in
green chemical design and waste management decision-making.
Results and discussion
Recent interest in combining green chemical design with
proactive environmental risk assessment has identified many
deficiencies in simply using standard toxicity and biodegrad-
ability tests to determine whether a chemical is a potential
environmental hazard. While useful starting points, standard
tests alone provide little information about the transformation
and fate of a chemical once it is released into the environment.
In this study, we have extended standard biodegradability
tests to include direct chemical analyses of biodegradation
products as well as an assessment of the toxicity of those
products.
Biodegradability by activated sludge microbial communities
Fig. 2 Whole-community biodegradation experiment DOC concentra-
tions (mg C L-1) (A) and absorbance at 265 nm (B) through the 41-day
incubation period. Error bars represent 1 standard deviation (n = 3).
Biodegradation, as measured by DOC concentration and ab-
sorbance at 265 nm, was analyzed until DOC concentrations
reached an average of <5 mg C L-1 for the three replicates,
indicating full mineralization of the IL (Fig. 2A and 2B). All
three ILs exhibited a significant DOC concentration decrease
throughout the incubation time period (rmANOVA p < 0.0039)
as well as a decrease in maximum absorbance at 265 nm
(rmANOVA p < 0.031). For bmpyrBr and hmpyrBr, full degra-
dation took 41 days, while for ompyrBr it took only 26 days.
There were no significant decreases in DOC concentrations of
the abiotic controls, indicating that chemical degradation was
due to the presence of the microbial activated sludge community
(not shown). To determine when microbial biodegradation of
the ILs began, we compared experimental sludge-inoculated
treatments to the abiotic controls through time. For bmpyrBr,
we observed the first significant decrease in DOC concentration
on day 28 (p = 0.003) and the first significant decrease in
maximum absorbance at 265 nm on day 29 (p = 0.013). In the
hmpyrBr tests, these decreases were observed on day 34 for DOC
concentration (p = 0.044) and day 35 for absorbance at 265 nm
(p = 0.059, marginally significant). Finally, for the ompyrBr test,
significant decreases occurred on day 18 for DOC concentration
(p = 0.051, marginally significant) and on day 21 for absorbance
at 265 nm (p = 0.011). There is agreement between the first day
of significant decrease in DOC concentration and the first day
of significant decrease in maximum absorbance at 265 nm in
all three tests (Fig. 2A and 2B). This indicates that the limiting
step to full biodegradation with all three ILs is cleavage of the
pyridinium ring. Once the ring has been opened, mineralization
occurs rapidly within a 10-day window.
Results from the OECD DOC-Die Away test indicate that
ompyrBr is the only one of the three ILs tested that falls
within the parameters of “readily biodegradable” as outlined in
the OECD protocol. “Readily biodegradable” compounds are
defined as those that exhibit a loss of 10% DOC, increasing to a
loss of at least 60% DOC within a 10-day window, in a total 28-
day long incubation period.39 In the ompyrBr biodegradation
test, a 10% decrease in DOC concentration compared to the
control began on day 18, and within 10 days, the concentration
decreased by 60% (Fig. 2A). Additionally, DOC decreased by
92% by the final day of the experiment (day 33). While bmpyrBr
and hmpyrBr DOC concentrations both decreased after 41 days
(88% and 95%, respectively), these ILs cannot be classified as
“readily biodegradable” because they did not biodegrade within
a 28-day period.
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 701–712 | 703
©

View Article Online
These results confirm prior analyses that show a correla-
tion between the length of the substituted alkyl chain on
the pyridinium ring and whole-compound biodegradability.29–31
Previous results from our group have shown that bmpyrBr
could not be mineralized after 43 days using similar methods
and microbial communities obtained from the same wastewater
treatment plant that was used in this study.29 1H-NMR results
from the final day indicated that the bmpyr cation was still intact,
suggesting resistance to degradation.29 The results here indicate
that bmpyrBr can be fully biodegraded, though it still cannot be
considered “readily biodegradable” by OECD standards. This
is contrary to the results found by Pham et al.36 Since similar
concentrations were used in the current study as in our previous
examination of bmpyrBr biodegradability, it is unlikely that
bmpyrBr toxicity significantly impacted the active microbial
consortium, as suggested by Pham et al.36 Instead, it is more
likely that a particular activated sludge sample may vary in mi-
crobial community composition, both by wastewater treatment
plant as well as seasonally within the same plant. This may
impact both the efficiency and mechanism of biodegradation.
Spatial and temporal variation in wastewater treatment plants
has been shown to affect degradation capacity.40,41 Additionally,
very specific changes in microbial community composition
may alter degradation efficiency or even ability.42 Incorporating
several treatment plant samples into standard biodegradability
assays may provide useful information about what communities
are necessary to degrade a particular chemical of interest.
Proactively assessing communities capable of biodegradation for
newly engineered chemicals using molecular microbial ecology
techniques, as well as determining whether a chemical is
degraded by a cultivatable isolate or a microbial consortium,
could provide valuable means for pollution prevention as well
as parameters to models predicting environmental fate.
Fig. 3 Concentration of IL (mg L-1) determined by HPLC-MS, of
one replicate of each of the three pyridinium ILs, and the sum of
the detectable pyridinium-based biodegradation products for each time
point during the 41-day whole-community biodegradation experiment.
Note that incubation time (x-axis) is different for ompyrBr.
As suggested by the DOC concentration and absorbance
data (Fig. 2A and 2B), once the pyridinium ring is opened,
biodegradation occurs quite rapidly, so the products after ring
cleavage are likely aliphatic and very labile to microorganisms.
This result is emphasized by comparison of the peak intensity
for each initial structure to the summed peak intensities of the
metabolite structures of an IL using RP-HPLC/MS through
time (Fig. 3). The base peak (or that which has the highest
intensity in a spectrum) in each MS spectrum in this study always
corresponds to the intact IL structure, with the exception of
the day 41 bmpyr degradation data point. These results suggest
that the rate-limiting step for 1-alkyl-3-methylpyridinium IL
degradation involves the first microbial action upon the chemical
structure and that few recalcitrant metabolites are produced
(Fig. 3). Most metabolites do not contain an aromatic ring or
are incorporated into microbial biomass or released as CO2 or
another gas, and therefore were not detectable by the methods
used in this study.
Biodegradability using HPLC/mass spectrometry
Metabolite structure assessed by RP-HPLC/MS and MS/MS
To investigate IL-metabolite structures, we removed and con-
centrated samples on days 1, 10, 17, 24, 29, 35 and 41 from the
bmpyr and hmpyr experimental treatments and on days 1, 10,
17, 21, 24, 29 and 33 from the ompyr treatment for further
examination using HPLC/MS. Different chemical structures
can be separated based on their hydrophobicities on a reverse-
phase HPLC column and then by their mass-to-charge ratios
(m/z value) in the mass spectrometer. Peaks with different
retention times corresponding to the same m/z value most likely
indicate structural-isomer variations. For each IL, the intact
molecule, M, consists of the pyridinium cation and bromide
anion. Thus, bmpyrBr is signified by MB, hmpyrBr by MH and
ompyrBr by MO. However, dissolution in water allowed us to
monitor the cation without its counter anion. Thus, the detected
intact IL base peaks were m/z 150 for [MB - Br]+ at 5.5 min
for bmpyr, m/z 178 for [MH - Br]+ at 8.3 min for hmpyr, and
m/z 206 for [MO - Br]+ at 10.3 min for ompyr. The original
RP-HPLC/MS chromatograms (not shown) provide estimate
of the concentration of initial test compound to metabolite
compound concentration (Fig. 3). It should be noted that
different acetonitrile–aqueous gradients on the LC column were
used to optimize each IL separation, making retention times
among the three ILs not comparable.
As in previous studies31,36 only biodegradation products that
contained a pyridinium ring could be detected above the
sample background using our LC/MS method, therefore full
biodegradation pathway elucidation was not possible. This is due
to a combination of factors, including a high amount of other
microbe-associated carbon molecules present in the samples,
as well as low concentrations of a diverse array of possible
biodegradation products compared to those products with the
ring intact. These detection limitations suggest that further
analysis of IL biodegradation pathways will need to include
radio- and/or stable isotope-labeled ILs, as well as higher
initial concentrations, to enable separation and detection above
background. Unfortunately, it is currently cost-prohibitive to
perform labeled-isotope experiments, since most ILs are synthe-
sized in academic labs.
Despite these detection limitations, the LC/MS method we
employed did yield peaks other than the IL cation that were
detected in the MS spectra of the ILs at different time points
(shown as Potential Metabolite Structures in Fig. 4, 5 and 6). We
used MS/MS (also referred to as tandem MS) to further assess
these peaks which were detected by MS and to determine if they
corresponded to structures relating to the theoretical starting
704 | Green Chem., 2010, 12, 701–712
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
points of biodegradation. This method utilizes collision-induced
dissociation (CID) to fragment a selected m/z value, (the
parent ion, in this case, a particular IL biodegradation product)
using argon gas. Through CID the parent ion is fragmented
into smaller ions, or daughter ions, that are analyzed to gain
further insight into the structure of the parent ion from which
they originate. In most cases, the pyridinium ring was highly
resistant to fragmentation by argon gas, even at collision energies
exceeding 100 kV, which is higher than those used in most
MS/MS protocols. This is due to the fact that these pyridinium
structures are both incredibly small and stable for traditional
fragmentation. Additionally, due to the low concentration of
the degradation products in some samples, particularly in the
latter time points, the low signal-to-noise ratio of the daughter
ions in the MS/MS spectrum against the background of the
sample also deterred identification. Despite these limitations we
were able to obtain some information about the structure of the
initial pyridinium-based degradation products. Here we address
each set of metabolites related to each of the three ILs separately:
We were unable to conclusively identify the [MB - Br]
metabolite corresponding to parent ion m/z 164 due to a
low signal/noise (S/N) of the daughter ions in the MS/MS
spectrum (Fig. 4, Panel C). Based on the m/z value determined
by MS alone, we suggest that this structure corresponds to a
combination of the two variants seen with parent ions m/z 148
(Fig. 4, Panel B) and m/z 166 (Fig. 4, Panel D), where the
butyl chain is reduced by a mass of 2 hydrogen atoms, and the
pyridinium ring is substituted with one hydroxyl group (Fig. 4,
Panel C).
Initial compound hmpyr [MH - Br-], m/z 178 (Fig. 5). The
MS spectra (not shown) of [MH - Br-], on Days 29 and 35
indicated three major degradation products: m/z 176 (reten-
tion times 4.17 and 7.40 min), m/z 192 (retention times 2.72 and
10.4 min) and m/z 194 (retention times 4.3 and 5.88 min). The
sum of these products for each time point is shown in black
as “m/z = other” in Fig. 3. The MS/MS spectrum for the
hmpyr parent ion m/z 178 (Fig. 4, Panel A) yielded the following
daughter ions: m/z 94.2 corresponding to 1-methylpyridine; m/z
85 indicating a hexyl group; m/z 57 from a butyl group; m/z 43
indicating a propyl group.
Initial compound bmpyr [MB - Br-], m/z 150 (Fig. 4). The
MS spectrum (not shown) of [MB - Br] on Day 41 indicated
three major degradation products: m/z 148 (retention time
5.08 min), m/z 166 (retention times 4.03, 9.99 and 15.09 min)
and m/z 164 (retention time 7.55 min). The sum of these
products for each time point is shown in black as “m/z =
other” in Fig. 3. In the case of bmpyr, as well as hmpyr and
ompyr, MS/MSexperiments were performed on the degradation
products (Panels B, C and D of Fig. 4, 5 and 6, respectively), as
well as on the intact IL (Panel A of Fig. 4, 5 and 6 respectively),
to aid in structural elucidation. The MS/MS spectrum for the
bmpyr parent ion m/z 150 (Fig. 4, Panel A) yielded the following
daughter ions: m/z 94.2 corresponding to 1-methylpyridine; m/z
57 corresponding to a butyl group (the alkyl side chain of the
intact IL); m/z 43 indicating a propyl group (partial degradation
of the side chain), and m/z 29 indicating an ethyl group (further
degradation of the side chain)
The MS/MS spectrum for the parent ion m/z 148 (Fig. 4,
Panel B) indicates that the metabolite structure is [MB - Br -
2H]+, where two hydrogen atoms have been removed from the
butyl group to reduce the total mass. This assignment is due to
the fact that the 1-methylpyridine group is intact (daughter ion
m/z 94 is present), but the butyl group has been reduced by a
mass of two atoms to yield a daughter ion m/z 55. We were
unable to identify daughter ion m/z 132 in this spectrum, but
suggest this structure may correspond to 1,3-dimethylpyridine or
a structure where the 1-butyl group has been partially degraded
either by rearrangement or removal of three carbon atoms.
The MS/MS spectrum of the parent ion m/z 166 (Fig. 4,
Panel D) suggests that the metabolite structure is a 1-butyl-3-
methylpyridine molecule with a single hydroxyl group substi-
tuted at one of the ring-carbon positions, though we are not
able to determine the position of the hydroxyl group. It should
be noted that in our MS/MS figures, a ᭢ symbol indicates
the potential position of a single hydroxyl group, as opposed
to the presence of several hydroxyl groups. Daughter ion m/z
110 indicates a hydroxylated 3-methylpyridine structure and
daughter ion m/z 57 indicates an intact butyl group.
The MS/MS spectrum of the biodegradation product parent
ion m/z 176 (Fig. 5, Panel B), indicates that the metabolite
structure contains a partially degraded side chain. While we
can not definitively identify the structure, daughter ion m/z 94
indicates that 3-methylpyridine remains intact and the presence
of the daughter ion m/z 132 suggests side-chain degradation by
three carbon atoms and side chain rearrangement.
The MS/MS spectrum of the biodegradation product parent
ion m/z 192 (Fig. 5, Panel C) suggests the identification
of the metabolite structure as 1-hexyl-3-methylpyridine with
a hydroxyl group substituted on the hexyl side chain. The
presence of both intact 3-methylpyridine (daughter ion m/z
94) and a re-arranged hydroxyl-substituted hexyl side chain
(daughter ion m/z 99.1) are suggestive of this metabolite
structure.
Finally, the MS/MS spectrum of the parent ion m/z 194
(Fig. 5, Panel D) indicates that this metabolite is 1-hexyl-3-
methylpyridine plus a single hydroxyl group substituted on a ring
position. The presence of daughter ions in this spectrum with
m/z 43 and m/z 85 indicate that the hexyl chain is intact and
unsubstituted. The presence of daughter ion m/z 110 indicates
a 3-methylpyridine structure that has been substituted with a
hydroxyl group on the ring, though the exact position cannot
be identified. We were unable to identify the daughter ion m/z
153, but suggest that this may correspond again to 1-hexyl-3-
methylpyridine in which rearrangement or degradation of the
hexyl side chain has occurred.
It should be noted that each of the three major degradation
products of hmpyr were found in two peaks in the LC/MS
total ion current trace eluting at different times. In each case,
the two peaks of the same m/z value but different retention
time indicate structural isomers of the compound. Therefore the
unsaturation and hydroxylation occurred at different positions
on the compound to yield metabolite structures with different
hydrophobicities.
Initial compound ompyr [MO - Br-], m/z 206 (Fig. 6). The
MS spectra (not shown) of [MO - Br-], on Days 17 and 29
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 701–712 | 705
©

View Article Online
Fig. 4 MS/MS results for initial bmpyr ([MB - Br]+) parent ion m/z 150 (Panel A), and biodegradation metabolite parent ions m/z 148 (Panel
B), m/z 164 (Panel C) and m/z 166 (Panel D). Each MS/MS spectrum represents a scan of a single replicate on the indicated experiment days and
daughter ion fractionation products are identified above each spectrum to indicate how the parent ion was identified. Parent ions for the initial bmpyr
cation and the three potential metabolite structures are presented to the left of each corresponding spectrum. ᭢ symbolizes the potential positions
for a single hydroxyl substitution around the ring, not the presence of 4 hydroxyl groups.
indicated three major degradation products: m/z 204 (retention
time 7.68 min), m/z 222 (retention time 2.60 min) and m/z 224
(retention time 1.86 min) The sum of these products over time
is shown in black as “m/z = other” in Fig. 3. The MS/MS
spectrum of the initial [MO - Br] parent ion m/z 206 (Fig. 6,
Panel A) yielded daughter ion m/z 120 corresponding to the
intact ompyr cation with the loss of six carbon atoms from the
side chain; m/z 107 presumably from the ompyr cation with
the loss of seven carbon atoms from the side chain; m/z 94.2
indicating 3-methylpyridine; m/z 71 corresponding to a pentyl
706 | Green Chem., 2010, 12, 701–712
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Fig. 5 MS/MS results for initial hmpyr ([MH - Br]+) parent ion m/z 178 (Panel A), and biodegradation metabolite parent ions m/z 176 (Panel
B), m/z 192 (Panel C) and m/z 194 (Panel D). Each MS/MS spectrum represents a scan of a single replicate on the indicated experiment days and
daughter ion fractionation products are identified above each spectrum to indicate how the parent ion was identified. Parent ions for the initial hmpyr
cation and the three potential metabolite structures are presented to the left of each corresponding spectrum. ᭢ symbolizes the potential positions
for a single hydroxyl substitution around the ring, not the presence of 4 hydroxyl groups.
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 701–712 | 707
©

View Article Online
Fig. 6 MS/MS results for initial ompyr ([MO - Br]+) parent ion m/z 206 (Panel A), and biodegradation metabolite parent ions m/z 204 (Panel
B), m/z 222 (Panel C) and m/z 224 (Panel D). Each MS/MS spectrum represents a scan of a single replicate on the indicated experiment days and
daughter ion fractionation products are identified above each spectrum to indicate how the parent ion was identified. Parent ions for the initial ompyr
cation and the three potential metabolite structures are presented to the left of each corresponding spectrum. ᭢ symbolizes the potential positions
for a single hydroxyl substitution around the ring, not the presence of 4 hydroxyl groups.
708 | Green Chem., 2010, 12, 701–712
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
group; m/z 57 from a butyl group; and m/z 43 from a propyl
group.
molecular microbial community analysis and isolation could
yield a proactive means for waste removal.
We were unable to conclusively identify metabolite biodegra-
dation product for the parent ion m/z 204 from the MS/MS
spectrum, but the structure likely corresponds to the ompyr
cation with the loss of two hydrogen atoms from the octyl group
(Fig. 6, Panel B).
Toxicity of biodegradation products
While standard biodegradability assays provide useful infor-
mation concerning the potential activity or persistence of a
chemical compound in the environment, and chemical analyses
throughout a biodegradation study provide greater insight to
the mechanism of catabolism, they do not provide information
about how metabolites of the parent pollutant may affect
relevant target organisms. Degradation pathways that result
in the creation of a compound that is more toxic, capable of
bioaccumulation or transport should be carefully avoided in
order to fulfill “green chemistry” expectations. For example, mi-
crobial biomethylation of mercury creates an organic form that
is substantially more lipophilic and toxic to higher organisms.43
Metabolites of halogenated phenolic compounds containing
an O-methylation have a significantly higher bioconcentration
potential, and are therefore more of an environmental hazard.44
As mentioned previously, PBDEs can be microbially converted
into more toxic debromination products.11 In light of such
cases, it becomes a crucial part of proactive ecotoxicology and
green chemistry to understand more about the mechanisms of
biodegradability, so that metabolites can be assessed as well as
their parent compounds.
In our study, we analyzed toxicity of the biodegradation
products over the course of 23 days (for bmpyrBr and hmpyrBr)
and 39 days (for ompyrBr) using the standard test organism,
Daphnia magna (Fig. 7). These tests were performed with mixed
microbial cultures enriched from sub-samples taken from the
final day of the initial biodegradability experiment instead of
with activated sludge communities. While we measured DOC
concentrations throughout the experiment, in this case, these
measurements were meant only as a guideline for comparison
of chemical degradation to toxicity, and not to be used as an
OECD-type assessment for ready biodegradability. It should be
noted that the timeframe for biodegradation is very different
with the enriched cultures than in the original activated sludge
community experiment (Fig. 2A vs. Fig. 7). For example,
bmpyrBr and hmpyrBr degraded fully in 23 days, whereas
ompyrBr took 39 days to degrade. This is most likely due to
the fact that we did not control for cell density, but rather,
transferred enrichments based on volume. While ompyrBr
typically degrades more quickly than bmpyrBr and hmpyrBr,
cells present in the ompyrBr enrichment culture may have been
beyond log-growth phase when the experiment was begun, while
bmpyrBr and hmpyrBr cells may have been mid-log phase,
accelerating their degradation rates.
The MS/MS spectrum of the parent ion m/z 222 suggests that
the metabolite structure is the ompyr cation substituted with one
hydroxyl group on the side chain as well as a loss of two hydrogen
atoms from the side chain (Fig. 6, Panel C). Evidence for this
structure is provided by the presence of daughter ion m/z 94
from intact 3-methylpyridine, as well as daughter ion m/z 69 as
the pentyl with the loss of two hydrogen atoms, and daughter
ion m/z 55 as a butyl with the loss of two hydrogen atoms.
Similarly, the MS/MS spectrum of the parent ion m/z 224
suggests that the metabolite structure is the ompyr cation
substituted with one hydroxyl group to the octyl side chain
(Fig. 6, Panel D), though the exact arrangement cannot be
determined. Daughter ions m/z 113 and m/z 94 indicate that the
entire octyl chain and 3-methylpyridine structure, respectively,
are intact.
Summary of metabolite structure. Overall our work yields
insight into the biodegradation of the three ILs bmpyr, hmpyr,
and ompyr through MS and MS/MS analyses. Our work implies
that the degradation of bmpyr with its shorter side chain
involves unsaturation of the butyl side chain and hydroxylation
of the aromatic ring, as opposed to hydroxylation of the side
chain as determined by Pham et al.36 This suggests that there
are different possible degradation pathways for bmpyr that
need to be considered. Our experiments for hmpyr also show
unsaturation of the side chain and suggest that degradation
by hydroxylation occurs separately on the side chain and on
the aromatic ring. Our ompyr results again show degradation
by unsaturation of the side chain, as well as hydroxylation of
the side chain. While the work of Stotle et al.31 indicates that
hydroxylation occurs on the side chain of the IL, this assignment
is proposed from MS data alone. Our MS/MS experiments yield
daughter ions that are generated solely from the isolated parent
ion and strongly suggest hydroxylation to the ring in the case of
bmpyr and hmpyr.
These results suggest that fundamentally different biodegra-
dation pathways occur depending on the length of the alkyl
chain. These different pathways are inherently linked to the
rate of biodegradation, where ompyrBr can be readily degraded
through initial hydroxylation of the long octyl chain, but
bmpyrBr resists biodegradation because initial hydroxylation
occurs on the recalcitrant pyridinium ring instead of the short
butyl chain. Our results suggest that different pathways for
biodegradation may be applicable depending on the substituted
alkyl chain length on the pyridinium ring. Additionally, the
differences in our results from those previously published suggest
that biodegradation products may exist in different forms,
or that degradation may occur at different rates, dependent
upon the specific activated sludge microbial community present
in the test inoculate. These results provide further evidence that
environmental risk assessment of a chemical should take into
account several sources of biodegradability information and that
We did not see a significant difference in toxicity from the
positive control in the bmpyrBr test until Day 23 (Tukey’s p =
0.007). There was a significant decrease in toxicity through
time in the bmpyrBr experimental samples (rmANOVA p =
0.019, between subjects). We observed a decrease in toxicity
of hmpyrBr on Day 11 (Tukey’s p = 0.053) and an overall
decrease intoxicity throughtime (rmANOVA p < 0.001, between
subjects). Finally, there was a significant decrease in toxicity of
ompyrBr on Day 39 (Tukey’s p < 0.001) and a decrease in toxicity
through time (rmANOVA p < 0.001, between subjects).
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 701–712 | 709
©

View Article Online
degradation of 1-alkyl-3-methylpyridinium bromide ILs results
in non-toxic metabolites, further study is necessary to determine
whether metabolites of the more biodegradable ester substituted
pyridiniums or ILs with octylsulfate anions are equally non-
toxic.37 Additionally, biotransformation of ILs by plants and
animals is still a largely unexamined area of research. While some
studies have examined the toxicity of ILs to the cytochrome P450
complex,46 no studies thus far have shown whether eukaryotic
cell conversion of ILs might result in more or less harmful
metabolites. A classic example of such an adverse consequence
of biotransformation is the conversion of the drug Prontosil by
animal cells into a sulfanilamide (antibiotic) compound which
competitively blocks the incorporation of 4-aminobenzoate into
the vitamin folic acid, thus creating a toxic effect.9 Few studies to
date have sought to examine multiple trophic-level ecotoxicology
and biotransformation of ILs, the full effects of which might be
masked by standard single-species toxicity tests (see ref. 26–28
for recent exceptions). Additionally, further tests investigating
the biodegradability of ILs by natural microbial communities
may provide greater insight as to the potential persistence of ILs
in the environment.
Experimental
Test compounds
The three pyridinium test compounds used in this study, 1-butyl-
3-methylpyridinium bromide (bmpyrBr), 1-hexyl-3-methyl-
pyridinium bromide (hmpyrBr) and 1-octyl-3-methylpyridinium
bromide (ompyrBr) (Fig. 1) were synthesized by Dr JaNeille
Dixon, University of Notre Dame, Department of Chemical
Engineering, using standard procedures.47–49
Fig. 7 Number of live Daphnia magna neonates after 24 h exposure
to biodegradation products (black bars), positive IL control (white
bars), negative control with no-IL (gray bars) and DOC concentration
in mg L-1 (lines) through the experimental time period. Error bars
represent 1 standard deviation (n = 3). Positive control treatments
killed all D. magna in the hmpyrBr and ompyrBr tests, so all values are
zero.
Biodegradability by activated sludge microbial communities
We used a modified OECD Guideline for Testing of Chemicals
standard dissolved organic carbon (DOC) Die-Away Test to
assess the biodegradability of three pyridinium ILs: bmpyrBr,
hmpyrBr and ompyrBr.39 An activated sludge microbial com-
munity sample collected from the South Bend Wastewater
Treatment Facility (South Bend, IN, USA) and pre-conditioned
by aerating at room temperature for five days. At the end of
the aeration period, the total suspended solid concentration
(TSS) of the sludge inoculate was determined to be 4.14 g L-1.
We prepared three replicate 1-L bottles with sterile mineral
medium containing: 0.085 g L-1 KH2PO4, 0.2175 g L-1 K2HPO4,
0.334 g L-1 Na2HPO4·2H2O, 5 mg L-1 NH4Cl, 36.4 mg L-1
CaCl2·2H2O, 22.5 mg L-1 MgSO4·7H2O, 0.25 mg FeCl3·6H2O,
as per the OECD guidelines.39 Each of the three ILs was added
to the media at a concentration of 30 mg C L-1 for each test.
We inoculated the bottles with 7.25 mL of activated sludge
sample and diluted each experimental replicate to 1000 mL,
to yield a final concentration of 30 mg suspended solids per liter.
We inoculated three replicate 1-L bottles containing mineral
medium with no-IL as biotic controls. Additionally, we created
one replicate bottle for each IL treatment with mineral medium
treatment and no bacterial inoculate, to control for abiotic
degradation of the ILs.
In all three tests, there was a strong inverse relationship
between DOC concentration and toxicity to D. magna (RB2 MP
=
0.862, R2HMP = 0.900, RO2 MP = 0.894). The media with negative
controls (no-IL) were consistently nontoxic to D. magna,
suggesting that the toxic effect is due only to the presence of
the IL. At the test concentrations, the hmpyrBr and ompyrBr
positive controls were very toxic and above the previously-
determined IC-50 values for D. magna.45 Thus, the positive
controls for hmpyrBr and ompyrBr tests killed all neonates
within the 24 h test period (Fig. 7). In the bmpyrBr tests, fewer
neonates died in the positive controls because of the relatively
low toxicity of this IL to D. magna.45
Results of this study suggest that pyridinium-cation ILs can be
fully biodegraded and that the metabolites of biodegradation are
significantly less toxic to a common aquatic organism. Further,
we show that pyridinium-ILs appear to degrade via several
pathways, potentially depending upon the microbial community
acting upon them. Therefore, pyridinium-IL degradation may be
more ubiquitous than other potentially more-persistent ILs with
imidazolium or quaternary ammonium cations.31,32,34 Further
study is necessary to determine which organisms and enzymes
may promote each degradation pathway. Also, while microbial
All 15 treatment and control flasks were shaken aerobically
at room temperature during a 41-day incubation period. We
710 | Green Chem., 2010, 12, 701–712
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
removed 10 mL of sample 23 times (approximately 3 times
per week) from the bottles for DOC concentration analysis.
The samples were syringe filtered (0.22 mm pore size) and
acidified using 100 mL of 2 N HCl. We measured nonpurgeable
DOC concentrations within 48 h of collecting the samples by
combustion on a Shimadzu TOC5000 with an autosampler. We
continued to monitor DOC concentrations until they reached a
level below 5 mg L-1, which occurred on Day 26 for ompyrBr and
on Day 41 for hmpyrBr and bmpyrBr. After each test, DOC data
were analyzed using repeated measures ANOVA (rmANOVA),
and then one-way ANOVA with a Tukey’s pairwise comparison
test at significant time points.
argon as the collision gas and collision energies of 45 kV or
100 kV, and using the same parameters as above except: RF lens
◦
0.4 V, source temperature 125 C and desolvation temperature
300 ◦C.
Toxicity of biodegradation products
We created enrichment cultures using 100 mL sub-samples from
the final day of the above biodegradation experiment (i.e. Day 40
for bmpyrBr and hmpyrBr and Day 33 for ompyrBr) inoculated
into 100 mL of fresh IL media, for each respective IL. These
mixed cultures were transferred into fresh IL media once every
2 weeks for 6 weeks to maintain an active microbial community,
capable of degrading each respective IL. After the third transfer,
we inoculated 100 mL of each of the enrichment cultures into 1 L
flasks containing 900 mL of fresh, sterile IL media. We did not
control for biomass in the inoculations, so different cell numbers
may have been inoculated into each set of IL media. Throughout
the experimental period, we removed 10 mL sub-samples for
DOC analysis as described above to monitor biodegradation
progress and 50 mL sub-samples for toxicity analysis. Sampling
dates were dependent upon the [DOC] response. For bmpyrBr
and hmpyrBr, we removed sub-samples for [DOC] and toxicity
analyses on Days 1, 7, 11, 15, 17 and 23. For ompyrBr,
we removed sub-samples on Days 1, 7, 11, 15, 17, 23, 30
and 39.
The 50 mL sub-samples used for toxicity testing were filtered
through 0.22 mm pore size vacuum filters and then diluted
to 100 mL with aerated groundwater in 250 mL acid-washed
beakers. Additionally, 3 replicate 50 mL controls of fully
concentrated (30 mg C L-1) IL media (positive control) and no-
IL media (negative control) were diluted in aerated groundwater
as well. We performed 24 h standard acute toxicity tests at each
sampling date during the IL experiment until concentrations
reached <5 mg C L-1, using Daphnia magna, a standard
aquatic toxicity test organism. Stock animals were obtained
from Carolina Biological Supply (Burlington, NC, USA) and
cultured in the lab for several months before use. Cultures were
fed Spirulina Dry Powder Microalga (Petaluma, CA, USA)
resuspended in groundwater until separation for the toxicity
test. For each toxicity test, we used an amended version of
the standard D. magna acute toxicity test method developed
by Sprague51 and employed by Bernot et al.25 We used four D.
magna neonates (age <24 h) born from parthenogenic females
grown in batch cultures for each experimental unit. The number
of living and dead neonates was noted after 24 h exposure, where
neonates observed as motionless and without a heartbeat were
considered to be dead. The median lethal concentrations (LC-
50) of all three ILs to D. magna have previously determined to
be 13.3 mg L-1 bmpyrBr, 1.0 mg L-1 hmpyrBr and 0.7 mg L-1
ompyrBr.45
Biodegradation product analysis by reverse-phase HPLC/mass
spectrometry
Sub-samples were removed throughout the incubation period for
reverse-phase high-performance liquid chromatography/mass
spectrometry (RP-HPLC/MS) analysis on Days 1, 10, 17, 24,
29, 35 and 41 (approximately once per week) for the bmpyrBr
and hmpyrBr tests and on Days 1, 10, 17, 21, 24, 29 and 33
for the ompyrBr test. Sub-samples (10 mL) were syringe-filtered
(0.22-mm pore size) and we measured the total absorbance at
265 nm, corresponding to maximum absorbance of the pyri-
dinium ring, using a Spectronic Genesys 2 spectrophotometer.
We then freeze-dried one replicate from each IL treatment for
24 h and rehydrated each in 1 mL of sterile deionized water
to yield a 10-fold concentration increase to aid in detecting
low concentration products. Methods from Stepnowski et al.50
were used as a starting point for analysis, but optimized for our
particular experiment. Samples were analyzed using an Alliance
2690 HPLC system (Waters, Milford, MA) equipped with a 996
photoiodide array detector and interfaced with a Micromass
Quattro LC triple quadrupole mass spectrometer (Micromass,
Manchester, UK) with an electrospray ionization source oper-
ated in the positive-ion mode. This technique enabled both UV
and mass spectrometric detection. Absorbance spectra in the
HPLC were monitored from 210 to 320 nm, and mass spectra
were acquired over the mass range 20 to 600 at a scan rate of
1.5 s. A 10–20 mL sample was injected onto a 4.6 ¥ 150 mm, 3 mm
particle size Atlantis dC18 analytical column (Waters) with a
4.6 ¥ 20 mm 3 mm particle size Atlantis dC18 guard column. The
mobile phase consisted of (A) 0.5% formic acid in acetonitrile
and (B) 20 mM aqueous ammonium acetate. The initial mobile
phase for bmpyrBr-based samples was 90% A/10% B (v/v)
with a 15 min linear gradient to 40% B (v/v). For hmpyrBr-
based samples we used an 80% A/20% B to 60% A/40% B
15-minute linear gradient. For ompyrBr-based samples, we
used a 70% A/30% B to 50% A/50% B 15-minute linear
gradient. For each method there was an additional 5 min
step following the gradient to re-equilibrate the column with
the initial mobile phase. The HPLC eluent was split 50 : 50
prior to entering the electrospray ion source of the mass
spectrometer. Mass spectrometer scans were conducted in the
electrospray positive-ion mode with the following conditions:
capillary voltage 3.2 kV, cone voltage 20 V, extraction voltage,
4 V, RF lens 0.2 V, source temperature 150 ◦C and desolvation
temperature 325 ◦C. We conducted MS/MS experiments using
Positive, negative and control toxicity data were analyzed
using repeated measures analysis of variance (rmANOVA). At
time points where significant differences occurred, we used
a one-way ANOVA and Tukey’s HSD pairwise comparison
tests. Additionally, we examined regressions between DOC
concentration and number of live neonates through time to
compare biodegradability to toxicity.
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 701–712 | 711
©

View Article Online
20 W. Z. Zhang, L. J. He, Y. L. Gu, X. Liu and S. X. Jiang, Anal. Lett.,
2003, 36, 827–838.
Acknowledgements
21 M. Freemantle, Chem. Eng. News, 2005, 83, 33–38.
22 Y. Zhang, B. R. Bakshi and E. S. Demessie, Environ. Sci. Technol.,
2008, 42, 1724–1730.
23 S. Stolte, J. Arning, U. Bottin-Weber, A. Muller, W. R. Pitner, U.
Welz-Biermann, B. Jastorff and J. Ranke, Green Chem., 2007, 9, 760–
767.
We thank Joan Brennecke and JaNeille Dixon for providing
ILs, David Costello, Brendan Bohannan and Jessica Bryant for
assistance in manuscript preparation, Ken Smdzinski and the
City of South Bend, Indiana Wastewater Treatment Facility
for assistance in activated sludge sample collection. Thanks
to Mike Brueseke for his assistance with Daphnia toxicity
tests. We also thank Dr Bill Boggess, Dennis Birdsell and the
University of Notre Dame Center for Environmental Science
and Technology (CEST) for their assistance with HPLC-MS
and MS/MS analysis. Funding was provided by the National
Science Foundation Graduate Research Fellowship Program,
National Oceanic and Atmospheric Administration (Grant
#NO04OAR4600076, #NA05OAR4601153), the U.S. Depart-
ment of Education Graduate Assistance in Areas of National
Need (GAANN) Fellowship Program (P200A010448) and the
State of Indiana 21st Century Science and Technology Fund
(Grant #909010455).
24 K. M. Docherty, S. Z. Hebbeler and C. F. Kulpa, Green Chem., 2006,
8, 560–567.
25 R. J. Bernot, M. A. Brueseke, M. A. Evans-White and G. A.
Lamberti, Environ. Toxicol. Chem., 2005, 24, 87–92.
26 R. J. Bernot, E. E. Kennedy and G. A. Lamberti, Environ. Toxicol.
Chem., 2005, 24, 1759–1765.
27 M. A. Evans-White and G. A. Lamberti, Environ. Toxicol. Chem.,
2009, 28, 418–426.
28 D. M. Costello, L. M. Brown and G. A. Lamberti, Green Chem.,
2009, 11, 548–553.
29 K. M. Docherty, J. K. Dixon and C. F. Kulpa, Biodegradation, 2007,
18, 481–493.
30 M. Stasiewicz, E. Mulkiewicz, R. Tomczak-Wandzel, J. Kumirska,
E. M. Siedlecka, M. Golebiowski, J. Gajdus, M. Czerwicka and P.
Stepnowski, Ecotoxicol. Environ. Saf., 2008, 71, 157–165.
31 S. Stolte, S. Abdulkarim, J. Arning, A. K. Blomeyer-Nienstedt, U.
Bottin-Weber, M. Matzke, J. Ranke, B. Jastorff and J. Thoming, Green
Chem., 2008, 10, 214–224.
32 A. Romero, A. Santos, J. Tojo and A. Rodriguez, J. Hazard. Mater.,
2008, 151, 268–273.
33 J. R. Harjani, J. Farrell, M. T. Garcia, R. D. Singer and P. J.
Scammells, Green Chem., 2009, 11, 821–829.
34 H. Sutterlin, R. Alexy, A. Coker and K. Kummerer, Chemosphere,
2008, 72, 479–484.
References
1 T. Collins, Green Chem., 2003, 5, G51–G52.
2 P. T. Anastas and L. M. Warner, Green Chemistry: Theory and
Practice, Oxford University Press, New York, 1998.
3 M. H. Wake, BioScience, 2008, 58, 349–353.
4 K. J. Kulacki, D. T. Chaloner, D. M. Costello, K. M. Docherty,
J. H. Larson, R. J. Bernot, M. A. Brueseke, C. F. Kulpa and G. A.
Lamberti, Chim. Oggi, 2007, 25, 32–36.
35 J. R. Harjani, R. D. Singer, M. T. Garcia and P. J. Scammells, Green
Chem., 2008, 10, 436–438.
36 T. P. T. Pham, C. W. Cho, C. O. Jeon, Y. J. Chung, M. W. Lee and
Y. S. Yun, Environ. Sci. Technol., 2009, 43, 516–521.
37 J. R. Harjani, R. D. Singer, M. T. Garcia and P. J. Scammells, Green
Chem., 2009, 11, 83–90.
38 P. Stepnowski and P. Storoniak, Environ. Sci. Pollut. Res., 2005, 12,
199–204.
39 OECD Guidelines for Testing Chemicals 301 A–F, Organisation for
Economic Co-operation and Development, Paris, France, 1992.
40 C. J. van der Gast, A. S. Whiteley and I. P. Thompson, Environ.
Microbiol., 2004, 6, 254–263.
41 C. J. van der Gast, D. Ager and A. K. Lilley, Environ. Microbiol.,
2008, 10, 1411–1418.
42 M. Manefield, R. I. Griffiths, M. B. Leigh, R. Fisher and A. S.
Whiteley, Environ. Microbiol., 2005, 7, 715–722.
43 S. Jensen and A. Jernelov, Nature, 1969, 223, 753–754.
44 A. S. Allard, M. Remberger and A. H. Neilson, Appl. Environ.
Microbiol., 1987, 53, 839–845.
45 D. J. Couling, R. J. Bernot, K. M. Docherty, J. K. Dixon and E. J.
Maginn, Green Chem., 2006, 8, 82–90.
46 For example: K. L. Tee, D. Roccatano, S. Stolte, J. Arning, J. Bernd
and U. Schwaneberg, Green Chem., 2008, 10, 117–123.
47 P. Bonhote, A. P. Dias, N. Papageorgiou, K. Kalyanasundaram and
M. Gratzel, Inorg. Chem., 1996, 35, 1168–1178.
5 P. J. Van den Brink, Environ. Sci. Technol., 2008, 42, 8999–9004.
6 K. D. Kimball and S. A. Levin, BioScience, 1985, 35, 165–171.
7 J. Cairns, BioScience, 1986, 36, 670–672.
8 J. Cairns, Hydrobiologia, 1983, 100, 47–57.
9 A. H. Neilson, Organic Chemicals in the Aquatic Environment:
Distribution, Persistence and Toxicity, CRC Press, Boca Raton, FL,
1994.
10 L. P. Wackett and C. D. Hershberger, Biocatalysis and Biodegrada-
tion: Microbial Transformation of Organic Compounds, ASM Press,
Washington, D.C., 2001.
11 J. Z. He, K. R. Robrock and L. Alvarez-Cohen, Environ. Sci. Technol.,
2006, 40, 4429–4434.
12 Bioremediation: Applied Microbial Solutions for Real-World Envi-
ronmental Cleanup, ed. R. M. Atlas and J. Philip, ASM Press,
Washington, D.C., 2005.
13 Y. Song, X. F. Zhu, X. L. Wang and M. J. Wang, J. Power Sources,
2006, 157, 610–615.
14 K. Lundell, T. Kurki, M. Lindroos and L. T. Kanerva, Adv. Synth.
Catal., 2005, 347, 1110–1118.
15 J. L. Bravo, I. Lopez, P. Cintas, G. Silvero and M. J. Arevalo, Ultrason.
Sonochem., 2006, 13, 408–414.
16 K. S. Yeung, M. E. Farkas, Z. L. Qiu and Z. Yang, Tetrahedron Lett.,
2002, 43, 5793–5795.
17 S. V. Malhotra and Y. Xiao, Aust. J. Chem., 2006, 59, 468–472.
18 W. Pei, J. Mo and J. L. Xiao, J. Organomet. Chem., 2005, 690, 3546–
3551.
48 C. M. Gordon, J. D. Holbrey, A. R. Kennedy and K. R. Seddon,
J. Mater. Chem., 1998, 8, 2627–2636.
49 L. Cammarata, S. G. Kazarian, P. A. Salter and T. Welton, Phys.
Chem. Chem. Phys., 2001, 3, 5192–5200.
50 P. Stepnowski, Aust. J. Chem., 2005, 58, 170–173.
51 J. B. Sprague, Water Res., 1969, 3, 793–821.
19 D. Behar, P. Neta and C. Schultheisz, J. Phys. Chem. A, 2002, 106,
3139–3147.
712 | Green Chem., 2010, 12, 701–712
This journal is
The Royal Society of Chemistry 2010
©