Transition metal catalyzed oxidation of Alcell lignin, soda lignin, and lignin model compounds in ionic liquids

Article abstract of DOI:10.1039/c001389g

Lignin is a component of lignocellulosic biomass from which important aromatic compounds can potentially be obtained. In the present work, Alcell and soda lignin were dissolved in the ionic liquid 1-ethyl-3-methylimidazolium diethylphosphate (EMIM DEP) and subsequently oxidized using several transition metal catalysts and molecular oxygen under mild conditions. CoCl ₂·6H₂O in EMIM DEP proved particularly effective for the oxidation. The catalyst rapidly oxidized benzyl and other alcohol functionalities in lignin, but left phenolic functionality and 5-5′, β-O-4 and phenylcoumaran linkages intact, as determined by analysis of various lignin model compounds and ATR-IR spectroscopy. The catalyst system oxidized the alcohol functionality contained in cinnamyl alcohol to form cinnamaldehyde or cinnamic acid or disrupted the double bond to form benzoic acid or an epoxide. The benzyl functionality in veratryl alcohol, a simple non-phenolic lignin model compound, was selectively oxidized to form veratraldehyde at a maximum turnover frequency of 1440 h^-1, compared to 10-15 h^-1 reported for earlier systems. Phenolic functional groups contained in guaiacol, syringol, and vanillyl alcohol remained intact, although the benzyl alcohol group in the latter was oxidized to form vanillin. Incorporation of strongly bound tetradentate ligands to the catalyst yielded reduced activity relative to those derived from simple metal salts in EMIM DEP. The influence of reaction conditions, such as temperature, oxygen pressure and NaOH loading, were also investigated. The system represents a potential method in a biorefinery scheme to increase the oxygen functionality in lignin prior to depolymerization or additional functionalization of already depolymerized lignin. The Royal Society of Chemistry 2010.

Full text of DOI:10.1039/c001389g

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/greenchem | Green Chemistry
Transition metal catalyzed oxidation of Alcell lignin, soda lignin, and lignin
model compounds in ionic liquids
Joseph Zakzeski, Anna L. Jongerius and Bert M. Weckhuysen*
Received 21st January 2010, Accepted 23rd March 2010
First published as an Advance Article on the web 24th May 2010
DOI: 10.1039/c001389g
Lignin is a component of lignocellulosic biomass from which important aromatic compounds can
potentially be obtained. In the present work, Alcell and soda lignin were dissolved in the ionic
liquid 1-ethyl-3-methylimidazolium diethylphosphate (EMIM DEP) and subsequently oxidized
using several transition metal catalysts and molecular oxygen under mild conditions. CoCl2·6H2O
in EMIM DEP proved particularly effective for the oxidation. The catalyst rapidly oxidized benzyl
and other alcohol functionalities in lignin, but left phenolic functionality and 5–5¢, b-O-4 and
phenylcoumaran linkages intact, as determined by analysis of various lignin model compounds
and ATR-IR spectroscopy. The catalyst system oxidized the alcohol functionality contained in
cinnamyl alcohol to form cinnamaldehyde or cinnamic acid or disrupted the double bond to form
benzoic acid or an epoxide. The benzyl functionality in veratryl alcohol, a simple non-phenolic
lignin model compound, was selectively oxidized to form veratraldehyde at a maximum turnover
frequency of 1440 h-1, compared to 10–15 h-1 reported for earlier systems. Phenolic functional
groups contained in guaiacol, syringol, and vanillyl alcohol remained intact, although the benzyl
alcohol group in the latter was oxidized to form vanillin. Incorporation of strongly bound
tetradentate ligands to the catalyst yielded reduced activity relative to those derived from simple
metal salts in EMIM DEP. The influence of reaction conditions, such as temperature, oxygen
pressure and NaOH loading, were also investigated. The system represents a potential method in a
biorefinery scheme to increase the oxygen functionality in lignin prior to depolymerization or
additional functionalization of already depolymerized lignin.
Dissolution of wood and other lignocellulosic material is
1. Introduction
a critical step for the valorization of biomass.1 In fact, the
Lignocellulosic biomass is a renewable resource that has the
insolubility of wood in common solvents has severely inhibited
potential to serve as a feedstock for the production of fu-
the development of processes for wood and lignocellulosic
els, chemicals, and energy.1 This biomass consists of three
biomass valorization.1 Although several strategies involving
major components – cellulose, hemicelluloses, and lignin –
solvents such as DMSO6 or CO2-expanded organic solvents7
each with unique physical, chemical, and structural properties
have been employed to dissolve lignocellulosic material, ionic
that offer specific opportunities to produce a multitude of
liquids in particular have recently become popular solvents
valuable chemicals.2 In particular, lignin is a three-dimensional
for the dissolution of biomass, and several types are com-
amorphous polymer consisting of methoxylated phenylpropane
mercially available.8 Some of the most common types are
structures from which high-value aromatic molecules can poten-
EMIM (1-ethyl-3-methylimidazolium) or BMIM (1-butyl-3-
tially be obtained.3 In plants, it constitutes up to 30% by weight
methylimidazolium) with various anionic counterions.8 The
and 40% by energy of lignocellulosic biomass.4 It fills the spaces
physical properties of these ionic liquids are highly dependent
between the cellulose and hemicellulose, and it acts as a resin
on the cation–anion pair. Because of their good thermal stability
to confer strength and rigidity to the plant.5 This same rigidity
and negligible vapor pressure, ionic liquids are regarded as
and strength, however, complicates the valorization of lignin
“green solvents” that could potentially replace volatile organic
and the other components of lignocellulosic biomass by making
solvents.9,10 Importantly, several recent studies have demon-
it resistant to chemical transformation. With the depletion of
strated the effectiveness of ionic liquids for the dissolution of
fossil fuels as a source of fuels, chemicals, and energy, the need
wood and lignocellulosic biomass.10–15 Vitz and coworkers, for
to develop processes to efficiently valorize lignin and the other
example, noted that 1-ethyl-3-methylimidazolium diethyl phos-
components of biomass to renewable energy and chemicals is
phate (EMIM DEP) in particular was an excellent solvent for the
becoming increasingly important.
dissolution of cellulose.16 Zakrzewska and coworkers recently
published a review of the solubility of carbohydrates, including
lignin, in ionic liquids, noting that the highest solubility of
lignin is obtained in 1,3-dimethylimidazolium methanesulfonate
(MMIM CH3SO3).17
Utrecht University, Sorbonnelaan 16, Utrecht, The Netherlands.
E-mail: b.m.weckhuysen@uu.nl
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1225–1236 | 1225
©

View Online
Once dissolved in the ionic liquid, lignocellulosic material
is susceptible to a wide range of transformations including
acetylation,18 acid hydrolysis,19,20 acylation,21 reduction,22 and
enzymatic oxidations using horseradish peroxidase.23 Although
there are several examples of catalysts capable of using molecular
oxygen for the oxidation of lignin (or lignin-related model
compounds), such as heterogeneous24–26 and homogeneous27–32
transition metal catalysts (Co(porphyrin),33 Co(salen)34–39 and
polyoxometalate40–49), to the best of our knowledge there are
currently no examples of catalytic oxidative lignin valorization
using molecular oxygen with ionic liquids as the solvent.1
Since molecular oxygen is a cheap and abundant oxidant, the
development of such a process would be attractive for future
utilization in lignin valorization processes. In addition, there
are several properties of ionic liquids that could make them an
attractive medium for lignin oxidative valorization. They have
high gas solubility50,51 and the ability to dissolve lignocellulosic
materials including wood and the products derived therefrom.
There is often a beneficial effect of pyridine and imidazole
on catalytic performance, particularly for Co(salen) and Co
porphyrin catalysts,52,53 and imidazole impurities are often in
high abundance in imidazolium-based ionic liquids. Moreover,
ionic liquids are an environmentally-friendly solvent choice for
the production of renewable chemicals. The challenge, therefore,
is to identify an ionic liquid and catalyst pair capable of
selectively valorizing lignin, or components derived from its
depolymerization, using molecular oxygen as the oxidant.
Motivated by these impetuses, the objective of this work
was to investigate the performance of Co(salen), Co(porphyrin)
and similar catalysts for lignin oxidation in ionic liquids using
molecular oxygen. Inspired by the effective dissolution of
cellulose by EMIM DEP,16 we extended this ionic liquid to the
dissolution of lignin and performed the catalytic oxidation of
dissolved lignin in this medium. In order to better understand
the chemical transformations that occur during lignin oxidation,
we investigated the reactivity of the catalytic system towards
several lignin model compounds that exhibit the salient linkages
and functionalities contained in lignin. We also analyzed the
influence of reaction conditions on catalytic performance and
attempted to determine the important components in the ionic
liquid–catalyst system.
Aldrich, 99%) or CoCl2·6H2O (0.0024 g, Acros Organics), along
with 2 mmol NaOH (0.080 g, Aldrich, 97%). The solution was
stirred overnight in order to completely dissolve the catalyst.
The designated substrate, typically 4.15 mmol veratryl alcohol
(0.698 g, Aldrich, 96%), was dissolved in the reaction solution,
which was then transferred to the autoclave and sealed. If
the reaction involved solid substrates or lignin samples, these
substrates were also dissolved in the ionic liquid by stirring
overnight to obtain a homogeneous solution. In some cases, the
impurities contained in the substrates consisted of the oxidized
product (i.e. veratraldehyde in the case of veratryl alcohol),
and these impurities were quantified and catalytic performance
calculations were adjusted accordingly. Before conducting a
reaction, the autoclave was purged four times with oxygen. The
autoclave was then charged with oxygen (0.5 MPa, Hoekloos,
99.5%) and heated to 353 K over about 5 min. After a designated
reaction time, the autoclave was cooled to 333 K, vented, and
a 1.0 g sample was taken. The autoclave was again purged with
oxygen, recharged with 0.5 MPa O2, and re-heated to 353 K.
Between reactions, the autoclave was thoroughly washed with
water and ethanol, rinsed with acetone or ethanol, and dried.
Products and residual substrate were separated from the
ionic liquid by five sequential extractions using 1.35 g ethyl
acetate. Prior to conducting reactions, extraction efficiencies
of the potential reaction components (i.e. veratryl alcohol,
veratraldehyde, vanillin, vanillyl alcohol, etc.) using ethyl acetate
or diethyl ether were determined. Both ethyl acetate and diethyl
ether efficiently extracted the substrate and products from the
ionic liquids, but extraction efficiencies were generally higher
when ethyl acetate was used, compared to diethyl ether. Two
distinct liquid phases formed when both of these solvents were
combined with EMIM DEP, and the top ethyl acetate phase
was decanted and analyzed. Higher mixing was observed with
many of the other ionic liquids, such as BMIM PF6, requiring
centrifugation prior to decantation.
Product quantification was conducted using a Shimadzu GC-
2010A gas chromatography unit equipped with a WCOT fused
silica CP-WAX 57CB column. Products were identified using a
Shimadzu GCMS-QP2010 unit or a LCMS-2010EV unit with
acetonitrile and 10 mM ammonium acetate in water as solvents,
and compared with pure standards when available.
2.2 Lignin model compound synthesis
2. Experimental
The 5–5¢ model compound, 3,3¢-dimethoxy-5,5¢-dimethyl-
1,1¢-biphenyl-2,2¢-diol, was synthesized as reported in the
literature.54,55 The product was repeatedly purified using a
silica gel column using hexanes and ethyl acetate (1 : 3) as
the solvent, and then crystallized. The purity was checked
using 1H NMR. The b-O-4 compound 1-hydroxy-1-(4-hydroxy-
3-methoxyphenyl)-2-(2,6-dimethoxyphenoxy)ethane was syn-
thesized as reported in the literature for the synthe-
sis of 1,3-dihydroxy-1-(3,5-dimethoxy-4-hydroxyphenyl)-2-(2,6-
dimethoxyphenoxy)propane, except that the paraformaldehyde
was not added to the 1-(4-benzoyloxy-2,5-dimethoxyphenyl)-
2-(2,6-dimethoxyphenoxy)ethanone during the synthesis.56 The
compound was similarly purified using a silica gel column
with hexanes and ethyl acetate (1 : 3) as the solvent, repeatedly
crystallized, and the purity checked using 1H NMR.
2.1 Typical reactions
AlcellTM Organosolv lignin was obtained from Repap Technolo-
gies Inc., Valley Forge, PA, USA. This lignin was extracted from
mixed hardwoods (maple, birch, and poplar) by an organosolv
process using aqueous ethanol. Soda lignin was extracted from
grass and was obtained from the Energy Research Centre of The
Netherlands (ECN).
Reactions were conducted in a 40 mL Parr stainless steel
autoclave. The temperature was monitored using a thermo-
couple, and stirring was conducted using a magnetic driver
equipped with an impellor at 750 rpm. Reactions were conducted
as follows (unless noted otherwise): into a 10.00 g sample of
1-ethyl-3-methylimidazolium diethyl phosphate (EMIM DEP,
Io-li-tec, >98%) was dissolved 10 mmol Co(salen) (0.0032 g,
1226 | Green Chem., 2010, 12, 1225–1236
This journal is
The Royal Society of Chemistry 2010
©

View Online
2.3 Spectroscopy
UV-visible spectroscopy was conducted using a Varian Cary
50 Conc UV-visible spectrometer equipped with an immersion
probe. During a measurement, 0.1 g of sample was placed in
a vial and diluted with 1.9 g ethanol. The immersion probe
with a 1 mm path length was placed in the solution, and
the UV-visible spectrum was recorded with pure ethanol as
the background. Infrared spectroscopy was conducted at room
temperature using a Perkin Elmer System 2000 FT-IR equipped
with a Pike Technologies MIRacleTM single-reflection horizontal
ATR accessory.
3. Results and discussion
Fig. 2 Difference in ATR-IR spectra between the soda lignin after
3.1 Lignin oxidation
various reaction times and the lignin solution prior to reaction: ( ) t =
The oxidation of two types of lignin, Alcell lignin and soda
lignin, was investigated using a CoCl2·6H2O catalyst dissolved
in EMIM DEP. As depicted in Fig. 1, which is a schematic repre-
sentation of hardwood lignin, the lignin polymer contains several
alcohol functional groups that are susceptible to oxidations and
also several types of linkages that, if successfully cleaved would
potentially result in low molecular weight aromatic products.
Both the Alcell lignin and soda lignin readily dissolved in
EMIM DEP to yield dark brown solutions. Samples were taken
each hour, and a small portion of each sample was saved
for spectroscopic analysis. Analysis of the reaction solution
was conducted by first performing ethyl acetate extractions
in an attempt to isolate any possible low molecular weight,
monomeric products formed during the reaction; however, no
monomeric products were detected by GC-MS. The lignin
reaction samples were then analyzed by ATR-IR in an attempt
to observe evidence of lignin oxidation. The results are depicted
in Fig. 2, which displays the difference in the ATR-IR spectra
between the lignin solution after various reaction times and
the lignin solution prior to reaction. Two new peaks appeared
near 3255 cm-1 and 1654 cm-1 that increased in intensity
with increasing reaction time. These peaks correspond to the
region of the infrared spectrum associated with alcohol and
1 h; ( ) t = 2 h; ( ) t = 3 h; ( ) t = 4 h.
aldehyde stretches, respectively, which suggests that the dissolved
lignin was selectively oxidized by the catalyst. The absence of
monomeric products, however, suggests either that the linkages
in the lignin remained intact or were insufficiently disrupted to
yield aromatic products of sufficiently low molecular weight for
GC-MS detection.
3.2 Lignin model compounds mimicking the linkages of lignin
In order to gain increased insight into the transformations
occurring within the lignin samples by the catalyst and to bypass
the analytical challenges associated with the lignin polymer,
additional analysis was conducted by the reaction of several
lignin model compounds that mimic the salient features of
lignin. Lignin contains several linkages, the most common of
which include b-O-4, 5–5¢, and b-5, throughout its structure
with the relative content depending on the plant material
and pretreatment method.1 Several model compounds were
therefore synthesized and used to investigate the reactivity of
these linkages to the CoCl2·6H2O in EMIM DEP. Schematic
representations of the model compounds used to analyze the
b-O-4, 5–5¢, and b-5 linkages are depicted in Fig. 3. 2,3-
Dihydrobenzofuran (compound 1, Fig. 3) represents the sim-
plest model compound that resembles phenylcoumaran linkages.
No reaction was observed after 4 h with this compound, and
the starting material was completely recovered. These results
indicate that phenylcoumaran linkages are expected to remain
intact during the oxidation of lignin.
Analysis of the 5–5¢ carbon–carbon bond was conducted us-
ing the compound 3,3¢-dimethoxy-5,5¢-dimethyl-1,1¢-biphenyl-
2,2¢-diol (compound 2, Fig. 3). After 4 h at 353 K, neither
4-hydroxy-3-methoxytoluene nor similar monomeric products
were detected, indicating that the carbon–carbon bond linking
the dimers remained intact. Interestingly, two products in ap-
proximately equal proportion were detected that corresponded
to approximately 50% conversion of the starting material after
4 h. GC-MS analysis of the products revealed masses of 288 and
302 daltons, which possibly corresponded to the oxidation of
the methyl groups on the dimer to the corresponding aldehyde.
Most likely, the reaction proceeds through the oxidation of the
methyl group to form the alcohol, which is then rapidly oxidized
Fig. 1 Schematic representation of hardwood lignin.1
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1225–1236 | 1227
©

View Online
Fig. 4 Lignin model compounds used to mimic the propyl side-chain
(4), the phenolic (5, 6, 7) and nonphenolic (8) functional groups of lignin.
Fig. 3 Lignin model compounds used to mimic the (1) phenylcoumaran
b-5, (2) 5–5¢, and (3) b-O-4 linkages of the lignin samples.
to the aldehyde. Similarly to the other phenolic compounds, no
quinones were detected as products.
Cinnamyl alcohol (compound 4, Fig. 4) exhibits a propyl
side-chain containing a carbon–carbon double bond, which
is occasionally observed in lignin streams based on the pre-
treatment method, and a Cg –OH functionality. The reaction
of this substrate was therefore conducted to gain insight into
the reactivity of the catalytic system towards the propyl side
chain. The results of the cinnamyl alcohol oxidation using the
catalyst system are depicted in Fig. 5. Initially, cinnamyl alcohol
was oxidized to cinnamaldehyde leaving the carbon–carbon
double bond intact. As the reaction proceeded, several other
products were observed including benzoic acid and cinnamic
acid. Epoxidation of the double bond was also observed. After
two hours, the yield of cinnamaldehyde began to decrease,
whereas the quantity of the other products continued to
increase. The quantity of benzoic acid and cinnamic acid formed
corresponded to the decrease in the cinnamaldehyde yield,
The b-O-4 linkage is the most common found in lignin,
and the lignin model compound 1-hydroxy-1-(4-hydroxy-3-
methoxy-phenyl)-2-(2,6-dimethoxyphenoxy)ethane (compound
3, Fig. 3) was used to model these linkages. As with the other
linkages of lignin, no evidence of syringol, apocynol or related
compounds expected from the rupture of the b-O-4 bond were
detected by GC-MS after reaction. Analysis of the same solution
by LC-MS revealed the oxidation of approximately 85% of the
alcohol functionality in the compound to the corresponding
ketone, as indicated in Scheme 1. No other products were
detected.
Scheme 1 Oxidation of the b-O-4 lignin model compound.
3.3 Lignin model compounds mimicking the functional groups
of lignin
In addition to the various linkages, both phenolic and non-
phenolic structures are present in lignin, as well as variations on
the propyl-side chain depending on the pretreatment method in
which the lignin was prepared.1 In order to determine the effect
of the catalyst system on the substituent groups in lignin, the
catalyst system was applied to several lignin model compounds,
schematic representations of which are depicted in Fig. 4.
Fig. 5 CoCl2·6H2O-catalyzed cinnamyl alcohol oxidation: mmol of
substrate and products versus time. Reaction conditions: 5.00 g EMIM
DEP, 10 mmol CoCl2·6H2O (0.0024 g), 1 mmol NaOH (0.04 g),
21.7 mmol cinnamyl alcohol (2.912 g), 0.5 MPa O2 (continuous), T =
353 K.
1228 | Green Chem., 2010, 12, 1225–1236
This journal is
The Royal Society of Chemistry 2010
©

View Online
suggesting that these products are also formed as a result of a
series of reactions, whereby high substrate concentrations favor
their formation. Interestingly, formation of the cinnamaldehyde
was initially favored over the formation of the epoxide; however,
as the quantity of substrate began to diminish, cinnamaldehyde
formation decreased significantly while the epoxide formation
continued unabated. These species are likely formed by a
different mechanism, in which the aldehyde rate is strongly
related to substrate concentration, whereas the epoxide is largely
independent.
The model compounds guaiacol (3-methoxyphenol) (5),
syringol (2,6-dimethoxyphenol) (6) and vanillyl alcohol (4-
hydroxy-3-methoxybenzyl alcohol) (7) were chosen as rep-
resentative phenolic lignin model compounds. Guaiacol and
syringol contain common functionalities observed in softwood
and hardwood lignin, respectively, without the presence of
other complicating functionalities. Without the extra methoxy
functionality serving as a protecting group, these compounds
are susceptible to oxidation to form benzoquinones. These sub-
strates, however, failed to react in the presence of CoCl2·6H2O
in EMIM DEP, and the original substrate was completely
recovered. As indicated in Fig. 6, the alcohol functionality of
vanillyl alcohol was successfully oxidized to form vanillin. As
in the case of guaiacol and syringol, the phenol component
of this compound remained intact; no quinones were detected.
No vanillic acid was detected; vanillin was the only product
observed.
by metalloporphrins and other catalysts allows facile compar-
ison of catalyst activity and selectivity, which are important
for further catalyst development. We therefore investigated the
influence of reaction conditions using veratryl alcohol with the
expectation that the results would extend to the other substrates
as well.
The CoCl2·6H2O-catalyzed veratryl alcohol oxidation was
conducted in EMIM DEP under constant oxygen pressure
in order to further analyze the ability of the catalyst to
oxidize benzyl alcohol functionality in lignin. The results of
the oxidation are given in Fig. 7. The CoCl2·6H2O catalyst
precursor rapidly oxidized veratryl alcohol to veratraldehyde
with a maximum turnover frequency of 1440 h-1. The highest
veratraldehyde yield was attained at 4 h, which corresponded
to 3345 total catalytic turnovers. Similar results were obtained
when CoCl2·6H2O was substituted with Co(salen), although
slightly reduced maximum turnover frequencies were obtained
(1300 h-1). The formation of veratrylic acid was detected and
coincided with the disappearance of veratraldehyde after 4 h.
Little acid was detected at short reaction times when the
veratraldehyde concentrations were low. These results, and the
lack of acid at low substrate concentrations, as further discussed
below, suggest that the acid forms by a series of reactions,
whereby veratryl alcohol is first oxidized to veratraldehyde
followed by over-oxidation to veratrylic acid.
Fig. 7 Semi-batch CoCl2·6H2O-catalyzed veratryl alcohol oxidation
with high concentrations of veratryl alcohol: mmol of substrate and
products versus time. Reaction conditions: 41.5 mmol veratryl alcohol
(6.98 g), 0.5 MPa O2 (continuous), T = 353 K.
Fig. 6 CoCl2·6H2O-catalyzed vanillyl alcohol oxidation: mmol of
substrate and products versus time: (
) vanillyl alcohol; (–ꢀ–)
vanillin. Reaction conditions: 5.00 g EMIM DEP, 10 mmol CoCl2·6H2O
(0.0024 g), 1 mmol NaOH (0.04 g), 21.7 mmol vanillyl alcohol (3.189 g),
0.5 MPa O2 (continuous), T = 353 K.
Based on the high activity still observed at low oxygen pressure
(see below), a semi-batch reaction at a lower oxygen pressure
was conducted to determine the effect on the acid formation.
The results are depicted in Fig. 8. The oxidation of veratryl
alcohol still proceeded readily at low oxygen pressures, although
the catalyst turnover frequency was slightly lower (790 h-1).
Importantly, significantly less acid was formed even as the
concentration of veratraldehyde increased. These results indicate
that the susceptibility of the catalytic system to over-oxidize
veratryl alcohol can be tuned based on the oxygen pressure,
effectively increasing selectivity to veratraldehyde with little loss
of activity.
Veratryl alcohol (3,4-dimethoxybenzyl alcohol) (8), is a
secondary metabolite synthesized from the degradation of lignin
by the white rot fungus.57,58 It has been widely used as a non-
phenolic lignin model compound for the analysis of processes for
catalytic lignin oxidation.52,53,57,59 The functional groups present
in this compound and ease of analysis make it a suitable
compound for the investigation of catalytic oxidation capability.
Thus, several catalytic systems have been reported for the
oxidation of veratryl alcohol as a lignin model compound.1 The
extensive use of veratryl alcohol to test for oxidation capability
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1225–1236 | 1229
©

View Online
precedence provides a convenient point of comparison in order
to elucidate the behavior of the Co catalysts in the unique
environment provided by the EMIM DEP.
3.4.1 Reactions in other ionic liquids. The imidazolium
ionic liquids have several distinctive properties that could poten-
tially result in the higher activity observed in the EMIM DEP, as
determined by the oxidation rate of the model compounds indi-
cated above, relative to other solvents. Ionic liquids are known to
exhibit high gas solubilities50,51 especially relative to water. The
increased oxidation rates may result because of the increased
oxidation potential caused by the higher oxygen content in the
ionic liquid. This possibility was explored and is discussed below.
Next, the ionic liquids themselves are typically only 98% pure,
and the remaining impurities, which are in large excess relative
to the catalyst, could influence the catalytic activity. A third
possibility for the higher activity originates from the interaction
of the anionic component of the ionic liquid with the catalyst.
In order to probe the role of the ionic liquid, and the impurities
contained therein, on catalytic performance, the oxidation of
veratryl alcohol was performed in the ionic liquids EMIM
CH3SO3 and BMIM PF6, in addition to EMIM DEP. The results
of the veratryl alcohol oxidation using Co(salen) in these ionic
liquids are depicted in Fig. 9. The oxidation using water as a
solvent was included for comparison purposes, in which about
24% of the veratryl alcohol was oxidized in 18 h. In each case, the
only product detected was veratraldehyde. Analysis of catalytic
performance in these exemplary ionic liquids yields valuable
insight regarding the role of the impurities contained in the ionic
liquids and anionic content as possible reasons for the increased
activity. Firstly, the identities of the impurities in each ionic
liquid were determined by ethyl acetate extraction and analysis
by GC-MS. In all cases, the presence of methylimidazole was de-
tected as the primary impurity in addition to residual unreacted
anionic components (i.e. PO4(C2H5)3 in the case of DEP). Axial
ligands, such as imidazole or pyridine, are known to enhance the
activity of Co(salen)60 and Co(porphyrin) complexes;66 however,
if the imidazole impurity was solely responsible for the enhanced
activity, similarly high activity would occur for the EMIM DEP,
EMIM CH3SO3 and BMIM PF6 ionic liquids because they
contain similar impurities. As seen in Fig. 8, vastly different
activity was observed in these ionic liquids; thus, the presence
Fig. 8 Semi-batch CoCl2·6H2O-catalyzed veratryl alcohol oxidation
with high concentrations of veratryl alcohol at lower oxygen pres-
sures: mmol substrate and products versus time. Reaction conditions:
41.5 mmol veratryl alcohol (6.98 g), 0.1 MPa O2 (continuous), T =
353 K.
The performance of this catalytic system for veratryl alcohol
oxidation contrasts with biomimetic metalloporphyrin and
metallosalen complexes, which have also been extensively used to
study veratryl alcohol oxidation. The oxidation rate was signifi-
cantly higher (two orders of magnitude) in the present system rel-
ative to both of the previously reported catalysts, Co(salen) and
Co(porphyrin), which are both known to reversibly bind molec-
ular oxygen and subsequently oxidize veratryl alcohol. Turnover
frequencies for the former are typically between 10–15 h-1
with total turnover numbers of 300,36–38,60 and although 100%
veratryl alcohol conversion was achieved, the total turnover
numbers for Co(porphyrin) catalysts were only ~33 in total
in 12 h.33 Different product distributions were also obtained.
With iron or manganese porphyrin catalysts (which use H2O2
as the oxidant61), and Co(porphyrins) (which use molecular
oxygen33,62), quinones were detected as reaction products33,63
in addition to veratraldehyde and veratrylic acid.52,53,63,64 Only
veratraldehyde and veratrylic acid were obtained as products
with CoCl2·6H2O or Co(salen) in EMIM DEP.
3.4 Catalysts, ligands, and other reaction conditions
The accumulated evidence indicated above suggests that
CoCl2·6H2O in EMIM DEP acts as a mild but efficient oxidation
catalyst capable of increasing the oxygen functionality in lignin.
As indicated above, Co(salen) complexes are also active in
EMIM DEP for catalytic oxidation. The effectiveness of this
and similar complexes as oxidation catalysts have been studied
in organic35 and aqueous solutions, and the effects of reaction
conditions, including temperature,60 the presence of NaOH,37
and oxygen pressure,60 have been studied in detail. The reversible
reaction of Co(salen) with oxygen gives an equilibrium mixture
of Co-superoxo complex and the dimeric peroxo complex with
the relative proportion strongly depending on the nature of the
starting complex and the reaction conditions.34,65 Moreover, the
ability of the Co complex to bind oxygen is strongly affected by
the coordination environment around the cobalt. Poor oxygen
binding occurs with 4-coordinate complexes but strong binding
occurs with 5-coordinate complexes.34 This extensive literature
Fig. 9 Conversion of veratryl alcohol to veratraldehyde and Co(salen)
turnovers versus time using various ionic liquids as solvents: (–ꢀ–) H2O;
(
) EMIM DEP; (
) EMIM CH3SO3; (
) BMIM PF6.
1230 | Green Chem., 2010, 12, 1225–1236
This journal is
The Royal Society of Chemistry 2010
©

View Online
of these impurities can be eliminated as a potential cause of
the higher activity observed in EMIM DEP. Comparison of
the results in which the cationic portion of the ionic liquid
remained constant whereas the anionic component was altered,
i.e. with EMIM DEP and EMIM CH3SO3, indicates instead
that the choice of ionic liquid anion plays a significant role in the
performance of the Co(salen) catalyst. With EMIM DEP, a high
oxidation rate was observed. whereas with EMIM CH3SO3 the
rate of veratryl alcohol oxidation was only slightly higher than
that observed in water. In addition, the hexafluorophosphate
component of BMIM PF6 is generally accepted as a weakly-
coordinating anion8 and is therefore not expected to interact
strongly with the catalyst. The catalytic activity in this ionic
liquid, however, was lower even than in water, as indicated
in Fig. 8. The accumulated evidence suggests that interaction
of the diethyl phosphate component of the ionic liquid with
the transition metal cation is an important factor for the high
activity observed in EMIM DEP.
3.4.3 Effect of reaction temperature. The preceding exam-
ples indicate that Co(salen) dissolved in EMIM DEP catalyzes
oxidation reactions at faster rates using lower oxygen pressures
than the same catalyst dissolved in other solvents. In order to
determine if the Co(salen)–EMIM DEP system was also capable
of performing oxidations at reduced and elevated temperatures,
the influence of temperature on veratryl alcohol oxidation
was investigated. The results are depicted in Fig. 11. At and
below 323 K, no veratryl alcohol oxidation was observed. In
accordance with previous reports of Co(salen) performance
in water, temperatures exceeding 343 K are required for high
catalytic activity.60 Increases in catalytic activity were observed
as the temperature was increased up to 383 K, when nearly 100%
of the veratryl alcohol was consumed after 3 h. Increasing the
temperature to 413 K had a deleterious effect on the catalytic
system. An orange-brown solid formed, the color of the ionic
liquid changed to dark red, and the viscosity of the solution
was increased. Not all of the products formed as a result of the
veratryl alcohol oxidation were recovered, thus indicating that a
portion of the solid observed formed as a result of the reaction
of the veratryl alcohol, possibly as a polymerization reaction.
The optimal range for the selective veratryl alcohol oxidation
therefore fell between 353 and 383 K. These results suggest that
elevated temperatures are still required for the formation of the
catalytically active species in accordance with previous reports
of Co(salen) in other solvents.
3.4.2 Influence of oxygen pressure. As indicated above,
another plausible explanation for the high oxidation rates in
ionic liquids originated with the high gas solubility relative to
other solvents, such as water. In order to explore this possibility,
the influence of oxygen pressure on catalytic performance was
investigated. Fig. 10 indicates the influence of oxygen pressure
on veratryl alcohol conversion and catalyst turnovers. Without
molecular oxygen present, no veratryl alcohol oxidation was ob-
served. As indicated in Fig. 9, the oxidation reaction proceeded
at relatively low oxygen pressures, where nearly 80% of the
veratryl alcohol was oxidized to veratraldehyde with 0.125 MPa
O2. Further increases in the oxygen pressure resulted in slight but
monotonic increases in catalytic activity. These results suggest a
high oxidation potential in the ionic liquid even at low oxygen
pressures, which allows the lignin oxidation to occur under very
mild conditions. The high oxygen solubility in the EMIM DEP
relative to other solvents is thus an important contributing factor
for the high activity observed. Moreover, the ability to use low
gas pressure is attractive for use in a biorefinery because of
decreased capital costs and increased process safety associated
with lower pressures.
Fig. 11 Conversion and catalytic turnovers versus temperature. Reac-
tion conditions: 5.00 g EMIM DEP, 5 mmol Co(salen) (0.0016 g), 1 mmol
NaOH (0.04 g), 2.08 mmol veratryl alcohol (0.349 g), 0.5 MPa O2, t =
3 h.
3.4.4 Influence of NaOH. NaOH is an important compo-
nent of the reaction system for Co(salen) catalyzed oxidations.37
In aqueous solutions, a hydroxide anion acts as an axial
ligand that facilitates coordination with molecular oxygen.37
Preliminary experiments of Co(salen) in EMIM DEP indicate
that NaOH is also important; however, the role of NaOH in
the EMIM DEP was not immediately clear. The importance of
NaOH for the Co(salen)-catalyzed veratryl alcohol oxidation
in EMIM DEP was therefore investigated to determine the
role of -OH and to determine whether it was catalytic or if
it was consumed during the course of the reaction. Firstly,
reactions in which high substrate quantities were dissolved in
the ionic liquids were conducted. Table 1 depicts the conversion
of veratryl alcohol to veratryl aldehyde using high initial
Fig. 10 Conversion of veratryl alcohol to veratraldehyde and catalyst
turnover numbers versus oxygen pressure. Reaction conditions: 5.00 g
EMIM DEP, 5 mmol Co(salen) (0.0016 g), 1 mmol NaOH (0.04 g),
2.08 mmol veratryl alcohol (0.349 g), T = 353 K, t = 3 h. At 0.0 MPa
PO , N2 was used to purge the autoclave rather than O2.
2
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1225–1236 | 1231
©

View Online
Table 1 Batch conversion of veratryl alcohol to veratryl aldehyde with high substrate concentrations (41.5 mmol, 6.98 g). Amount of O2 estimated
using ideal gas law: P = 0.6 MPa, V = 0.04 L, T = 298 K. Oxygen recharged to autoclave between samples
Starting
O2/mmol
Veratryl alcohol
consumed/mmol
Total aldehyde/
NaOH ratio
Aldehyde/
O2 ratio
Aldehyde/
O ratio
Entry
Time/h
NaOH/mmol
Conversion (%)
1
2
3
4
3
6
9
18
9.8
9.8
9.8
9.8
2.0
2.0
2.0
2.0
41.7
72.2
82.5
86.9
17.3
12.7
4.3
8.7
1.8
1.3
0.4
0.2
0.9
0.6
0.2
0.1
15.0
17.2
18.1
1.8
substrate concentrations. After three hours reaction (Entry 1),
the reactor pressure dropped to nearly zero, and analysis of the
reaction solution indicated that 41.7% of the veratryl alcohol
was oxidized to veratraldehyde, corresponding to 17.3 mmol
of substrate consumed. The ratio of the total quantity of
aldehyde produced to the quantity of NaOH charged to the
autoclave indicated an 8.7-fold excess of veratraldehyde to
NaOH. The stoichiometry suggests that NaOH is not consumed
during the course of the reaction, but instead merely facilitates
the oxidation. In addition, veratryl alcohol oxidation did not
occur in the absence of molecular oxygen even with NaOH
present (see above), further indicating that molecular oxygen
is the ultimate oxidant rather than the oxygen contained in
-
the OH anion. Furthermore, an estimate of the quantity of
oxygen initially charged to the autoclave using the ideal gas law
indicated an approximately 2 : 1 molecular ratio of aldehyde was
Fig. 12 Conversion of veratryl alcohol and catalyst turnovers versus
time with added NaOH. Reaction conditions: 5.00 g EMIM DEP, 5 mmol
Co(salen) (0.0016 g), 2.08 mmol veratryl alcohol (0.349 g), 0.5 MPa O2,
T = 353 K, t = 3 h.
produced to oxygen consumed. For other Co(salen) oxidations,
36,37,39
both H2O2
and H2O38 have been reported as reaction
by-products. These results suggest, by stoichiometry, that the
ultimate by-product of the oxidation conducted by Co(salen) in
EMIM DEP is water rather than hydrogen peroxide, and also
that both atoms of the oxygen molecule are ultimately consumed
in the reaction. The conversion of veratryl alcohol increased with
increasing reaction time, as indicated by Entries 2–4, ultimately
culminating in 86.9% conversion after 18 h.
has a distinctly different coordination environment relative to
the Co(salen) complex. In the ionic liquid, CoCl2·6H2O adopts
a tetrahedral geometry with distinct d-to-d transitions in the
visible region, giving the complex a green-blue color, whereas the
higher-energy d-to-d transitions for the square-planar Co(salen)
complex give it a yellow color in solution. In other solvents, the
different ligand environment causes vastly different reactivity
between the complexes. As noted earlier, when dissolved in water,
Co(salen) oxidizes veratryl alcohol, whereas no oxidation was
observed when CoCl2·6H2O was used as the catalyst. Never-
theless, in EMIM DEP, CoCl2·6H2O oxidized veratryl alcohol
more rapidly than the Co(salen) complex. In order to determine
if this difference in coordination environment would similarly
During Co(salen)-catalyzed reactions in water, -OH was
found to bond axially to the Co complex, which served to
activate it.37,60 In the presence of -OH anions, other components
such as pyridine, which normally increases the activity of the
complex, were unnecessary.60 Likewise, NaOH was an important
component of the reaction system for Co(salen) in EMIM DEP.
As indicated in Fig. 12, after 3 h no veratryl alcohol oxidation
was observed in the absence of NaOH, and the catalytic activity
increased as the NaOH concentration increased, approaching
nearly complete veratryl alcohol conversion at the highest
NaOH concentrations. As indicated above, the -OH anions were
not consumed as a stoichiometric reactant relative to veratryl
alcohol, suggesting that NaOH plays a role in activating the
catalyst instead. Several of the metal complexes investigated
in this study (discussed in detail below) have a vastly different
coordination environment than Co(salen) and therefore could
potentially behave significantly differently relative to Co(salen).
For example, with the exception of the Co(porphyrin) complex
and Jacobsen’s catalyst, none of the other complexes analyzed
for oxidative capability have a square-planar geometry. They
-
influence the role of OH, reactions catalyzed by CoCl2·6H2O
in the absence of NaOH were conducted, and the results are
depicted in Fig. 13. As in the case of the Co(salen), the oxidation
of veratryl alcohol was not observed initially. Remarkably, after
an approximately 5 h induction period, CoCl2·6H2O began to
oxidize the veratryl alcohol, and the rate soon approached the
rate observed with the catalyst in the presence of NaOH. The
solution color at the conclusion of the reaction, as determined by
UV-visible spectroscopy, strongly resembled the solution after
the reaction in which NaOH was present. When the experiment
was repeated using Co(salen) as the catalyst, a similar result
was obtained. After the same induction period, the oxidation
of veratryl alcohol proceeded. As indicated in Fig. 14, the
change in activity coincided with a change in the color of the
reaction solution from green, in the case of CoCl2·6H2O, to
yellow and finally orange. The onset of activity was marked
by the appearance of a strong absorbance near 345 nm, which
indicated a change in the coordination sphere of the catalyst.
-
therefore lack the axial position through which to bond OH
such that molecular oxygen would bond trans to the -OH while
leaving the other ligands in the complex in their original posi-
tions. One of the most active catalyst precursors, CoCl2·6H2O,
1232 | Green Chem., 2010, 12, 1225–1236
This journal is
The Royal Society of Chemistry 2010
©

View Online
and catalyst turnovers at various reaction times in EMIM DEP
using a wide range of other transition metals. The transition
metal salts tested included: a) cobalt porphyrin, Co(salen), and
simpler Co salts; b) Mn-based Jacobsen’s catalyst, which is
usuallyusedinconjunctionwithH2O2 or other peroxidesthatare
thought to generate high-valent oxomanganese species,67,68 and
simpler Mn salts; c) Fe salts; d) Cr salts; e) Ni salts; and finally f)
Cu salts. The activity of Co(salen) is included in all graphs to aid
in comparison of relative activities. In all cases, the only product
observed was veratraldehyde. Several of the simpler Co salts,
including CoCl2·6H2O, Co(acac)3, Co(acac)2, Co(OAc)2·4H2O,
and Co(NO3)2·6H2O, provided slightly higher activity than
the Co(salen) complex and significantly outperformed the Co
porphyrin catalyst. These results were surprising since the latter
compounds are known to engage in the selective oxidation of
veratryl alcohol to veratraldehyde.33 Nearly 90% conversion of
the veratryl alcohol was achieved after 3 h, and in many cases
the reaction was driven nearly to completion at longer reac-
tion times with these catalysts. In contrast, only approximately
80% and 10% of the starting veratryl alcohol was oxidized to
veratraldehyde with the Co(salen) and Co porphyrin complex,
respectively; the remainder was recovered as starting material.
A similar occurrence was observed with the Mn-catalyzed
reactions, as indicated by Fig. 15b. Veratryl alcohol oxidation
was observed when catalyzed by simpler Mn salts (namely
Mn(NO3)2 and Mn(OAc)2), but Jacobsen’s catalyst (a Mn–
Schiff base epoxidation catalyst that resembles the Co(salen)
structure) was ineffective for veratryl alcohol oxidation. In
general, the compounds with tetradentate ligands exhibited
reduced activity, suggesting that these strongly-binding ligands
block a coordination site on the catalyst center or, alternatively,
are resistant to transformation of the metal center to its active
form.
Fig. 13 Catalyst turnovers versus time for the oxidation of veratryl
alcohol in the absence of NaOH. (–ꢀ–) CoCl2·6H2O; ( ) Co(salen).
Fig. 14 Absorbance at 345 nm versus time for CoCl2·6H2O-catalyzed
veratryl alcohol oxidation in the absence of NaOH. The inset graph
depicts the full spectrum at various reaction times.
Veratryl alcohol oxidation was also observed with complexes
of Fe, Cr, and Ni, as indicated in Fig. 15c–e.
Relatively high activity was observed with Cr(acac)3, which
was less effective than Co, yet conversion of veratryl alcohol
approached 100% at longer times. Activity was also observed
with complexes of Fe and Ni, with approximately 50% of the
starting material oxidized to veratraldehyde with the Fe catalysts
and approximately 30% oxidized with Ni. In contrast, Cu was
an ineffective oxidation catalyst; there was no indication of any
reaction, and only the starting material was recovered.
These results indicate that several transition metal catalysts
are capable of oxidizing veratryl alcohol to veratraldehyde
when EMIM DEP is used as the solvent. The general order
of activity for the transition metal cations investigated was
Co > Cr > Fe > Ni > Mn ꢀ Cu, although some variations
existed depending on the starting complex. The EMIM DEP
proved an important solvent for the use of these transition
metals as oxidation catalysts. All of the salts investigated readily
dissolved in the ionic liquid after stirring overnight, whereas
several of the salts failed to dissolve in other solvents, such
as water. The reactivity of these salts was also significantly
different in EMIM DEP relative to other solvents such as
water. For example, although the Co(salen) complex is capable
of performing veratryl alcohol oxidation in water, as indicated
in Fig. 9 and previous reports,38,60 no oxidation was observed
when Co(OAc)2·4H2O and CoCl2·6H2O were used in water, yet
Both the Co(salen) and CoCl2·6H2O reverted to the same
color, and the UV-visible spectrum of the final solution was
indistinguishable with CoCl2·6H2O- and Co(salen)-catalyzed
reactions in the presence of NaOH at shorter reaction times.
The similar rate observed in the presence of NaOH, and in its
absence after the induction period, indicates that NaOH does
not assist the reaction by deprotonating the substrate, making
the substrate more reactive. If, in fact, NaOH was necessary
to deprotonate the substrate, a significantly lower rate would
be expected even after the induction period in the absence of
NaOH. Taken together with the other observations of NaOH
indicated above, these results suggest instead that the presence of
-OH accelerates the formation of the active catalytic complex in
the EMIM DEP solution rather than serving as an actual ligand
on the Co complex itself. After a suitable induction period, the
Co catalyst still reverts to the same active form even in the
absence of NaOH, as indicated by the UV-visible spectrum of
the complex after reaction.
3.4.5 Other transition metal catalysts. Although the most
active, Co catalysts were not the only transition metals capable
of acting as an oxidation catalyst in the EMIM DEP. Fig. 15
depicts the conversion of veratryl alcohol to veratryl aldehyde
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1225–1236 | 1233
©

View Online
Fig. 15 Conversion of veratryl alcohol to veratraldehyde and catalyst turnovers versus time using 10 mmol of various transition metal catalysts.
CoTK(4-MP) P = 5,10,15,20-tetrakis(4-methoxyphenyl)-21H,23H-porphyrin cobalt(II). Jacobsen’s catalyst = (R,R)-(-)-N,N¢-bis(3,5-di-tert-butyl-
salicylidene)-1,2-cyclohexane-diaminomanganese(III) chloride.
these salts proved excellent precursors for highly active oxidation
catalysts when dissolved in EMIM DEP.
during the lignin oxidation provide insight regarding the lignin
oxidation. The inability of the catalyst to disrupt the b-O-4, 5–5¢,
and other linkages present in the model compounds suggests
that these linkages remained intact during the lignin oxidation.
These results indicate the reason that the low molecular weight
monomeric compounds were not detected as a result of the ox-
idation. Similarly, phenolic functional groups remained intact,
4. Conclusions
Taken together, the results of the lignin model compound
analysis coupled with the spectroscopic evidence obtained
1234 | Green Chem., 2010, 12, 1225–1236
This journal is
The Royal Society of Chemistry 2010
©

View Online
Scheme 2 Possible structural changes in lignin samples as a result of oxidation.
but benzyl and other alcohol functionalities were susceptible Gosselink of Wageningen UR Food & Biobased Research and
to oxidation to form aldehydes or, with high oxygen pressures
or substrate loadings, to over-oxidation, forming acids. These
oxidations proceed at high rates under mild conditions. A
schematic representation of the possible transformations in the
lignin that resulted are given in Scheme 2. The transformations
indicated in the scheme represent several of the anticipated
changes based on the results of the lignin model compounds and
spectroscopic analysis. The ether and carbon–carbon linkages
remain intact, but several of the alcohol functional groups in
the lignin are oxidized to the corresponding aldehyde.
to Wouter Huijgen of ECN for supplying the Alcell and soda
lignin, respectively.
References
1 J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius and B. M.
Weckhuysen, Chem. Rev., 2010, 10.1021/cr900354u.
2 C. Hill, Wood modification. Chemical, Thermal, and Other Processes,
John Wiley & Sons, Chichester, 2006.
3 F. S. Chakar and A. J. Ragauskas, Ind. Crops Prod., 2004, 20, 131–
141.
4 R. D. Perlack, L. L. Wright, A. F. Turhollow, R. L. Graham,
B. J. Stokes and D. C. Erbach, Biomass as Feedstock for a Bioenergy
and Bioproducts Industry: The Technical Feasibility of a Billion-Ton
Annual Supply, U.S. Dept. of Energy and U.S. Dept. of Agriculture,
Oak Ridge National Laboratory, Oak Ride, Tennessee, 2005, pp.
1–59.
5 S. K. Ritter, Chem. Eng. News, 2008, 86, 10–17.
6 M. Fasching, P. Schro¨der, R. P. Wollboldt, H. K. Weber and H. Sixta,
Holzforschung, 2008, 62, 15–23.
7 C. Eckert, C. Liotta, A. Ragauskas, J. Hallett, C. Kitchens, E. Hill
and L. Draucker, Green Chem., 2007, 9, 545–548.
8 I. Krossing and I. Raabe, Angew. Chem., Int. Ed., 2004, 43, 2066–
2090.
9 R. A. Sheldon, R. M. Lau, M. J. Sorgeddrager and F. van Rantwijk,
Green Chem., 2002, 4, 147–151.
10 S. H. Lee, T. V. Doherty, R. J. Linhardt and J. S. Dordick, Biotechnol.
Bioeng., 2009, 102, 1368–1376.
11 D. A. Fort, R. C. Remsing, R. P. Swatloski, P. Moyna, M. Guillermo
and R. D. Rogers, Green Chem., 2007, 9, 63–69.
12 M. Zavrel, D. Bross, M. Funke, J. Bu¨chs and A. C. Spiess, Bioresour.
Technol., 2009, 100, 2580–2587.
13 N. Sun, M. Rahman, Y. Qin, M. L. Maxim, H. Rodr´ıquez and R. D.
Rogers, Green Chem., 2009, 11, 646–655.
The EMIM DEP and Co catalyst combination proved
suitable for lignin and lignin model compound oxidation using
molecular oxygen. Properties of the ionic liquid, including
high oxygen solubility, allowed the oxidation to proceed under
mild conditions, and the large abundance of diethyl phosphate
anion caused simple metal salts to exhibit remarkably high
activity. Catalyst precursors with strongly binding ligands, such
as tetradentate porphryin or salen ligands, tended to exhibit
reduced activity relative to simple metal salts, which suggests
that these strongly-binding ligands inhibit the transformation of
the transition metal to the active form, which is a process aided
-
by the presence of OH anions. Nevertheless, the ionic liquid
EMIM DEP provided a suitable media for the dissolution and
subsequent oxidation of lignin, or molecules derived therefrom,
using a variety of transition metal catalysts. This catalyst system
could potentially be used as a lignin pretreatment method
in a biorefinery to increase the functionality and hydrophilic
character of the lignin, prior to a subsequent step designed
to rupture the lignin linkages to yield highly-functionalized
monomers. Alternatively, given the high oxidation rate relative
to similar catalysts in common aqueous or organic solvents,
it could be used as a secondary step to functionalize the
compounds formed after the pretreatment step that forms the
lignin monomers.
14 Y. Pu, N. Jiang and A. J. Ragauskas, J. Wood Chem. Technol., 2007,
27, 23–33.
15 S. S. Y. Tan, D. R. MacFarlane, J. Upfal, L. A. Edye, W. O. S. Doherty,
A. F. Patti, J. M. Pringle and J. L. Scott, Green Chem., 2009, 11, 339–
345.
16 J. Vitz, T. Erdmenger, C. Haensch and U. S. Schubert, Green Chem.,
2009, 11, 417–424.
17 M. E. Zakrzewska, E. Bogel-Lukasik and R. Bogel-Lukasik, Energy
Fuels, 2010, 24, 737–745.
18 I. Kilpelaeinen, H. Xie, A. King, M. Granstrom, S. Heikkinen and
D. S. Argyropoulos, J. Agric. Food Chem., 2007, 55, 9142–9148.
19 C. Sievers, M. B. Valenzuela-Olarte, T. Marzialetti, I. Musin, P. K.
Agrawal and C. W. Jones, Ind. Eng. Chem. Res., 2009, 48, 1277–1286.
20 C. Li, Q. Wang and Z. K. Zhao, Green Chem., 2008, 10, 177–182.
21 H. Xie, A. King, I. Kilpelainen, M. Granstrom and D. S.
Argyropoulos, Biomacromolecules, 2007, 8, 3740–3748.
Acknowledgements
JZ gratefully thanks the National Science Foundation Interna-
tional Research Fellowship Program for support of this research
under Award No. 0856754. Thanks are also extended to Richard
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1225–1236 | 1235
©

View Online
22 J. B. Binder, M. J. Gray, J. F. White, Z. C. Zhang and J. E. Holladay,
Biomass Bioenergy, 2009, 33, 1122–1130.
23 S. Sgalla, G. Fabrizi, S. Cacchi, A. Macone, A. Bonamore and A.
Boffi, J. Mol. Catal. B: Enzym., 2007, 44, 144–148.
24 F. G. Sales, L. C. A. Maranha˜o, N. M. Lima Filho and C. A. M.
Abreu, Chem. Eng. Sci., 2007, 62, 5386–5391.
25 S. Bhargava, H. Jani, J. Tardio, D. Akolekar and M. Hoang, Ind. Eng.
Chem. Res., 2007, 46, 8652–8656.
26 J. Zhang, H. Deng and L. Lin, Molecules, 2009, 14, 2747–2757.
27 R. DiCosimo and H. C. Szabo, J. Org. Chem., 1988, 53, 1673–1679.
28 G. A. A. Labat and A. R. Goncalves, Appl. Biochem. Biotechnol.,
2008, 148, 151–161.
29 A. R. Gonc¸alves and U. Schuchardt, Appl. Biochem. Biotechnol.,
1999, 77, 127–132.
30 W. Partenheimer, Adv. Synth. Catal., 2009, 351, 456–466.
31 S. Hwang, Y.-W. Lee, C.-H. Lee and I.-S. Ahn, J. Polym. Sci., Part
A: Polym. Chem., 2008, 46, 6009–6015.
32 V. E. Tarabanko, N. A. Fomova, B. N. Kuznetsov, N. M. Ivanchenko
and A. V. Kudryashev, React. Kinet. Catal. Lett., 1995, 55, 161–170.
3 3 W. Z h u a n d W. T. Fo r d , J. Mol. Catal., 1993, 78, 367–378.
34 J. J. Bozell, B. R. Hames and D. R. Dimmel, J. Org. Chem., 1995, 60,
2398–2404.
44 T. Yokoyama, H.-m. Chang, R. S. Reiner, R. H. Atalla, I. A.
Weinstock and J. F. Kadla, Holzforschung, 2004, 58, 116–121.
45 T. Voitl and P. R. von Rohr, ChemSusChem, 2008, 1, 763–769.
46 A. Gaspar, D. V. Evtuguin and C. P. Neto, Holzforschung, 2004, 58,
640–649.
47 H. Deng, L. Lin, Y. Sun, C. Pang, J. Zhuang, P. Ouyang, Z. Li and
S. Liu, Catal. Lett., 2008, 126, 106–111.
48 H. Deng, L. Lin, Y. Sun, C. Pang, J. Zhuang, P. Ouyang, J. Li and S.
Liu, Energy Fuels, 2009, 23, 19–24.
49 B. N. Kuznetsov, V. E. Taraban’ko and S. A. Kuznetsova, Kinet.
Catal., 2008, 49, 517–526.
50 J. Palgunadi, J. E. Kang, D. Q. Nguyen, J. H. Kim, B. K. Min, S. D.
Lee, H. Kim and H. S. Kim, Thermochim. Acta, 2009, 494, 94–98.
51 A. Finotello, J. E. Bara, D. Camper and R. D. Noble, Ind. Eng. Chem.
Res., 2008, 47, 3453–3459.
52 G. Labat and B. Meunier, J. Org. Chem., 1989, 54, 5008–5011.
53 F. Cui and D. Dolphin, Bioorg. Med. Chem., 1995, 3, 471–477.
54 F. A. Marques, F. Simonelli, A. R. M. Oliveira, G. L. Gohr and P. C.
Leal, Tetrahedron Lett., 1998, 39, 943–946.
55 M. Albrecht and M. Schneider, Synthesis, 2000, 1557–1560.
56 S. Kawai, K. Okita, K. Sugishita, A. Tanaka and H. Ohashi, J. Wood
Sci., 1999, 45, 440–443.
57 K. Lee, K. Pi and K.-M. Lee, World J. Microbiol. Biotechnol., 2009,
25, 1691–1694.
35 R. S. Drago, B. B. Corden and C. W. Barnes, J. Am. Chem. Soc.,
1986, 108, 2453–2454.
36 V. Sippola, O. Krause and T. Vuorinen, J. Wood Chem. Technol.,
2004, 24, 323–340.
37 K. Kervinen, M. Allmendinger, M. Leskela¨, T. Repo and B. Rieger,
Phys. Chem. Chem. Phys., 2003, 5, 4450–4454.
38 K. Kervinen, H. Korpi, J. G. Mesu, F. Soulimani, T. Repo, B. Rieger,
M. Leskela¨ and B. M. Weckhuysen, Eur. J. Inorg. Chem., 2005, 2591–
2599.
39 C. Canevali, M. Orlandi, L. Pardi, B. Rindone, R. Scotti, J. Sipila
and F. Morazzoni, J. Chem. Soc., Dalton Trans., 2002, 3007–3014.
40 J. A. F. Gamelas, A. R. Gaspar, D. V. Evtuguin and C. Pascoal Neto,
Appl. Catal., A, 2005, 295, 134–141.
58 J.-T. Liu, X.-R. Huang and P.-J. Gao, Chin. J. Chem., 2007, 25, 1627.
59 K. Gonza´lez Arzola, M. C. Are´valo and M. A. Falco´n, Electrochim.
Acta, 2009, 54, 2621–2629.
60 K. Kervinen, H. Korpi, M. Leskela¨ and T. Repo, J. Mol. Catal. A:
Chem., 2003, 203, 9–19.
61 M. Tien and T. K. Kirk, Science, 1983, 221, 661–663.
62 P. Zucca, F. Sollai, A. Garau, A. Rescigno and E. Sanjust, J. Mol.
Catal. A: Chem., 2009, 306, 89–96.
63 A. Kumar, N. Jain and S. M. S. Chauhan, Synlett, 2007, 411–414.
64 P. Zucca, G. Mocci, A. Rescigno and E. Sanjust, J. Mol. Catal. A:
Chem., 2007, 278, 220–227.
41 A. Gaspar, D. V. Evtuguin and C. Pascoal Neto, Appl. Catal., A,
2003, 239, 157–168.
42 A. R. Gaspar, J. A. F. Gamelas, D. V. Evtuguin and C. P. Neto, Chem.
Eng. Commun., 2009, 196, 801–811.
43 Y. S. Kim, H.-m. Chang and J. F. Kadla, J. Wood Chem. Technol.,
2007, 27, 225–241.
65 K. C. Gupta, A. K. Sutar and C.-C. Lin, Coord. Chem. Rev., 2009,
253, 1926–1946.
66 M. Tsuda and H. Kasai, Surf. Sci., 2007, 601, 5200–5206.
67 T. Kurahashi and H. Fujii, Inorg. Chem., 2008, 47, 7556–7567.
68 C. M. M. Santos, A. M. S. Silva, J. A. S. Cavaleiro, A. Le´vai and T.
Patonay, Eur. J. Org. Chem., 2007, 2877–2887.
1236 | Green Chem., 2010, 12, 1225–1236
This journal is
The Royal Society of Chemistry 2010
©