Aerobic oxidation of β-1 lignin model compounds with copper and oxovanadium catalysts

Article abstract of DOI:10.1021/cs400636k

The reactivity of homogeneous oxovanadium and copper catalysts toward aerobic oxidation of phenolic and nonphenolic β-1 lignin model compounds has been investigated. Aerobic oxidation of diastereomeric, nonphenolic β-1 lignin models (1T, 1E) using the six-coordinate vanadium complex (HQ) ₂V^V(O)(OⁱPr) (HQ = 8-oxyquinolinate) as a precatalyst in pyridine afforded ketone (3) and dehydrated ketone (4) derived from oxidation of the secondary alcohol. In contrast, using CuOTf/2,6-lutidine/ TEMPO (OTf = trifluoromethanesulfonate, TEMPO = 2,2,6,6-tetramethyl-piperidin-1- yl-oxyl) in toluene for the same reaction afforded 3,5-dimethoxybenzaldehyde (5) and 4-methoxybenzaldehyde (6) as major products resulting from C _α-C_β bond cleavage. Reactions of the corresponding phenolic lignin model compounds (2T, 2E) with 10 mol % CuOTf/2,6-lutidine/TEMPO gave ketone (9) as the major product, whereas 10 mol % (HQ)₂V^V(O)(OⁱPr) or a stoichiometric amount of CuOTf/2,6-lutidine/TEMPO yielded 2,6-dimethoxybenzoquinone (10) as the major product, arising from cleavage of the C_aryl-C_α bond. Different selectivity was observed in the oxidation of 2T, 2E using the five-coordinate complex (dipic)V^V(O)(OⁱPr) (dipic = dipicolinate), with α,β-unsaturated aldehyde (14) as the major product (formed from oxidation of the primary alcohol and dehydration). The key differences in chemoselectivity between the vanadium and copper catalysts in the oxidations of these phenolic and nonphenolic β-1 lignin models are discussed.

Full text of DOI:10.1021/cs400636k

Research Article
pubs.acs.org/acscatalysis
Aerobic Oxidation of β‑1 Lignin Model Compounds with Copper and
Oxovanadium Catalysts
Baburam Sedai,† Christian Díaz-Urrutia,† R. Tom Baker,*,† Ruilian Wu,‡ L. A. “Pete” Silks,‡
and Susan K. Hanson*,§
†Department of Chemistry and Centre for Catalysis Research and Innovation, University of Ottawa, Ottawa, ON K1N 6N5 Canada
§
‡Biosciences Division and Chemistry Division, Los Alamos National Laboratory, Los Alamos, New Mexico 87545, United States
S
* Supporting Information
ABSTRACT: The reactivity of homogeneous oxovanadium and copper catalysts toward aerobic
oxidation of phenolic and nonphenolic β-1 lignin model compounds has been investigated.
Aerobic oxidation of diastereomeric, nonphenolic β-1 lignin models (1T, 1E) using the six-
coordinate vanadium complex (HQ)2VV(O)(OiPr) (HQ = 8-oxyquinolinate) as a precatalyst in
pyridine afforded ketone (3) and dehydrated ketone (4) derived from oxidation of the secondary
alcohol. In contrast, using CuOTf/2,6-lutidine/TEMPO (OTf = trifluoromethanesulfonate,
TEMPO = 2,2,6,6-tetramethyl-piperidin-1-yl-oxyl) in toluene for the same reaction afforded 3,5-
dimethoxybenzaldehyde (5) and 4-methoxybenzaldehyde (6) as major products resulting from
Cα−Cβ bond cleavage. Reactions of the corresponding phenolic lignin model compounds (2T, 2E) with 10 mol % CuOTf/2,6-
lutidine/TEMPO gave ketone (9) as the major product, whereas 10 mol % (HQ)2VV(O)(OiPr) or a stoichiometric amount of
CuOTf/2,6-lutidine/TEMPO yielded 2,6-dimethoxybenzoquinone (10) as the major product, arising from cleavage of the Caryl
−
Cα bond. Different selectivity was observed in the oxidation of 2T, 2E using the five-coordinate complex (dipic)VV(O)(OiPr)
(dipic = dipicolinate), with α,β-unsaturated aldehyde (14) as the major product (formed from oxidation of the primary alcohol
and dehydration). The key differences in chemoselectivity between the vanadium and copper catalysts in the oxidations of these
phenolic and nonphenolic β-1 lignin models are discussed.
KEYWORDS: aerobic oxidation, lignin model compounds, vanadium, copper, C−C bond cleavage
INTRODUCTION
■
In recent years, there has been considerable interest in
developing new catalytic processes to derive chemicals and
fuels from non-food-based lignocellulosic biomass, an attractive
renewable carbon feedstock.1 Lignin is a randomized polymer
consisting of methoxylated phenoxypropyl units and a major
biomass constituent (15−30% by weight).2 As a result of its
enzymatic synthesis through radical condensation, lignin has a
diverse structure where monomeric units are connected
through several different types of linkages, including β-O-4,
5−5′, β-5, 4-O-5, and β-1 (Figure 1). Because of the complex
nature and insolubility of lignin, its presence has been identified
as a major challenge in the production of biofuels and chemicals
from lignocellulose.3
Over the last few decades, there has been particular interest
in utilizing inexpensive earth-abundant metal catalysts and air
or dioxygen for selective oxidation of lignin to value-added
aromatic chemicals.4−20 CuO oxidation has been used
extensively to characterize lignin components,21 with carboxylic
acids typically obtained as products. Aldehyde and ketone
products are generally more valuable and could serve as
versatile intermediates for further functionalization chemistry.
Although impressive strides have been reported for the catalytic
wet aerobic oxidation process using heterogeneous catalysts6
and, recently, using ionic liquids as both reaction and separation
media,19 detailed studies of complex lignin models provide the
Figure 1. Most prevalent chemical linkages characterized in lignin
from spruce trees. Adapted with permission from Zakzeski, J.;
Bruijnincx, P. C. A.; Jongerius, A. L.; Weckhuysen, B. M. Chem. Rev.
2010, 110, 3552−3599. Copyright 2010 American Chemical Society.
opportunity to better understand the molecular origins of
oxidation selectivity. As a result, a number of catalysts have
Received: August 1, 2013
Revised: October 18, 2013
Published: November 22, 2013
© 2013 American Chemical Society
3111
dx.doi.org/10.1021/cs400636k | ACS Catal. 2013, 3, 3111−3122

ACS Catalysis
Research Article
Figure 2. Ancillary ligands on oxovanadium complexes generate disparate oxidation selectivity of complex β-O-4 lignin models. The asterisk (*)
symbol indicates a 13C label in the lignin model compound.
Scheme 1. Synthesis of Lignin Model Compounds 1T and 1E
been applied to the oxidation of complex β-O-4 models that
contain electron-rich aromatic functionalities, both primary and
secondary alcohols, as well as a phenoxy linkage. Recent work
from our groups8 and that of Toste9 has demonstrated that
ancillary ligands in oxovanadium complexes can dictate the
selectivity for C−O vs C−C bond cleavage in both phenolic
and nonphenolic β-O-4 models. In the Toste work with salen-
type ligands,9a C−O bond cleavage is favored, and the
importance of aryloxy radicals in reoxidizing the V center was
noted (Figure 2). In contrast, coaddition of base with 8-
oxyquinolinate or dipicolinate−oxovanadium complexes led to
secondary alcohol oxidation via a proposed 2e− base-assisted
dehydrogenation mechanism.8a,d,f The latter catalysts also effect
C−C bond cleavage of the ketone intermediates, affording
phenol and carboxylic acid products. Major selectivity differ-
ences were found in reactions with phenolic lignin models with
the 8-oxyquinolinate precatalysts, again favoring C−C (over
C−O) bond cleavage (Figure 2).8d,22
The β-1 lignin linkage (Figure 1) can be a significant
component in wood lignins (i.e., ∼7−9%)4 and differs from the
β-O-4 linkage by replacement of the O-bound phenoxy unit
with a C-bound aryl group. The relative abundance of the β-1
linkage in lignin is somewhat uncertain and may depend on the
method of pretreatment.24 Degradation studies (acidolysis and
thioacidolysis) indicate that β-1 structures are prevalent,
particularly in hardwood lignins.25 However, NMR studies of
wood have revealed only minor amounts of these structures.26
This discrepancy may reflect an uneven distribution of the β-1
linkage or that such structures form upon treatment with acid.27
Herein, we compare the CuOTf/2,6-lutidine/TEMPO catalyst
system with oxovanadium complexes supported by dipicolinate
and 8-oxyquinolinate ligands for the aerobic oxidation of
nonphenolic (1T, threo; 1E, erythro) and phenolic (2T, 2E) β-1
lignin models. The catalysts exhibit major differences in
chemoselectivity. Trends are discussed and compared with
those observed previously using analogous β-O-4 lignin models.
We recently compared the activity and selectivity of CuCl/
TEMPO and dipicolinate vanadium catalysts for the aerobic
oxidation of lignin models.8c In particular, the CuCl/TEMPO
catalytic system in pyridine was found to effect direct C−C
bond cleavage in the β-O-4 lignin models, and subsequently, a
more effective copper catalytic system based on CuOTf/2,6-
lutidine/TEMPO was developed.8g Although this Cu catalyst
system offers selectivity similar to that of peroxidase enzymes
without the need for hydrogen peroxide, the oxidation
mechanism is likely more complicated,8g,23 and activity is
hampered by competing substrate formylation and inhibition
by the phenol products.8g
RESULTS AND DISCUSSION
■
Synthesis of Phenolic and Nonphenolic β-1 Lignin
Model Compounds. The phenolic β-1 lignin model
compounds threo-1-(4-hydroxy-3,5-dimethoxyphenyl)-2-(4-me-
thoxyphenyl)-propane-1,3-diol (2T), and erythro-1-(4-hydroxy-
3,5-dimethoxyphenyl)-2-(4-methoxyphenyl)-propane-1,3-diol
(2E) were prepared according to a literature procedure.28
Nonphenolic lignin model compounds 1T and 1E were
synthesized using an analogous procedure, as shown in Scheme
1 (See the Experimental Section for details). The threo and
erythro isomers were purified by silica gel chromatography, and
3112
dx.doi.org/10.1021/cs400636k | ACS Catal. 2013, 3, 3111−3122

ACS Catalysis
Research Article
Figure 3. Aerobic oxidation of nonphenolic lignin model compound 1T using the vanadium precatalyst (HQ)2VV(O)(OiPr) in pyr-d5 affords
predominantly products of C−H bond cleavage.
a
Table 1. Comparison of the Vanadium and Copper-Catalyzed Oxidation of the Non-Phenolic β-1 Lignin Model Compound 1
a
b
Ketone (C−H bond cleavage) products are highlighted in blue, and C−C bond cleavage products are highlighted in green. Methanol (2 %) was
also formed.
Figure 4. Aerobic oxidation of nonphenolic lignin model compound 1E using the vanadium precatalyst (HQ)2VV(O)(OiPr) (10 mol %) in toluene
with added NEt3 (15 mol %) affords predominantly ketone 3.
1
compounds 1T, 1E, 2T, and 2E were characterized by H and
13C NMR spectroscopy.
Figure S1). The identity of the products was confirmed by
carrying out the reaction of 1T on a larger scale; ketone 3 and
dehydrated ketone 4 were isolated and characterized by 1H and
13C NMR spectroscopy and high-resolution mass spectrometry.
In the oxidation of 1T and 1E, the dominant pathway of the
vanadium-catalyzed reaction is the oxidation of the secondary
benzylic alcohol, affording ketone products (∼90%). Products
arising from C−C bond cleavage were formed in lower yields
(<10%). This selectivity for oxidation of the secondary benzylic
alcohol parallels that observed in the oxidation of the analogous
nonphenolic β-O-4 lignin model compound by the vanadium
complex (HQ)2VV(O)(OiPr), which afforded primarily ketone
products (Figure 2). Previous kinetic and computational studies
of the oxyquinolinate vanadium system suggest that oxidation
proceeds through a two-electron, base-assisted pathway.8f The
observed selectivity for oxidation at the secondary benzylic
alcohol of 1T is consistent with such a mechanism.
Oxidation Catalysis of Nonphenolic and Phenolic β-1
Lignin Model Compounds. Nonphenolic β-1 Lignin
Models. To identify potential selectivity differences that could
arise between β-O-4 and β-1 lignin model compounds, we
tested the six-coordinate vanadium complex (HQ)2VV(O)-
(OiPr) as a precatalyst for aerobic oxidation of the nonphenolic
β-1 lignin models 1T and 1E in pyridine solvent (HQ = 8-
oxyquinolinate). When 1T was heated with 10 mol % of the
precatalyst under air at 100 °C (pyr-d5, 48 h), complete
conversion of the starting material was observed, as judged by
1
integration of the H NMR resonances against an internal
standard (Figure 3). The major products of this reaction were
the ketone 3 (34%) and dehydrated ketone 4 (57%).29
Additional minor products included 3,5-dimethoxybenzalde-
hyde, 4-methoxybenzaldehyde, 3,5-dimethoxybenzoic acid, and
4-methoxybenzoic acid (5−8).
The erythro isomer (1E) was oxidized more slowly by the
vanadium catalyst with only 76% conversion observed after
48 h, affording a similar product distribution with somewhat
lower yields of the ketone 3 (29%) and dehydrated ketone 4
(38%) products (100 °C, pyr-d5, Supporting Information
To assess the potential influence of the solvent on the
reaction selectivity,30 we also carried out the aerobic oxidation
of the nonphenolic β-1 lignin model compounds 1T and 1E
using DMSO as the solvent and the vanadium complex
(HQ)2VV(O)(OiPr). When the nonphenolic lignin model 1T
was heated with 10 mol % of precatalyst under air at 100 °C
3113
dx.doi.org/10.1021/cs400636k | ACS Catal. 2013, 3, 3111−3122

ACS Catalysis
Research Article
Figure 5. Aerobic oxidation of nonphenolic lignin model compound 1T using the copper catalyst system CuOTf/TEMPO/2,6-lutidine in toluene
solvent affords predominantly products of C−C bond cleavage.
Figure 6. Aerobic oxidation of the phenolic lignin model compound 2T using the vanadium catalyst (HQ)2VV(O)(OiPr) in pyr-d5 affords a mixture
of C−H and C−C bond cleavage products.
(DMSO-d6, 48 h), 91% conversion of the starting material was
observed after 48 h. The major product of this reaction was the
ketone 3 (66%). Additional minor products included the
dehydrated ketone 4 (6%), 3,5-dimethoxybenzaldehyde (5), 4-
methoxybenzaldehyde (6), 3,5-dimethoxybenzoic acid (7), 4-
methoxybenzoic acid (8), and methanol (Table 1).
Toluene was also an effective solvent for the oxidation of the
β-1 lignin model compounds. Using (HQ)2VV(O)(OiPr) (10
mol %) as a precatalyst in combination with the base promoter
NEt3 (15 mol %), complete conversion of 1E was observed
after heating in toluene for 48 h (100 °C). The major product
of this reaction was the ketone 3 (∼72%), with minor amounts
of 3,5-dimethoxybenzaldehyde (∼7%) and 4-methoxybenzalde-
hyde (∼2%) also formed (Figure 4). Overall, changing the
solvent from pyridine to toluene did not have a dramatic impact
on the reaction selectivity in the vanadium-catalyzed oxidation
of 1T and 1E; preferential oxidation at the secondary benzylic
alcohol position was observed in both cases.
3,5-dimethoxybenzoic acid (7) and 4-methoxybenzoic acid (8)
were also formed in very low yields (∼1%). There was no
significant difference in reactivity of the two diastereomers 1T
and 1E with the copper catalyst; comparable yields were
observed in both cases. The last entry in Table 1 indicates that
better catalyst performance is obtained using bulkier 2,6-
lutidine (vs pyridine), as noted previously for the β-O-4 lignin
models.8g Control experiments were also conducted for 1T and
1E under identical conditions but without added copper (with
TEMPO and excess 2,6-lutidine under O2 in toluene at 100
1
°C). Examination of the H NMR spectrum revealed that no
reaction had occurred.
The copper-catalyzed aerobic oxidation of 1T differs
markedly in selectivity from the vanadium-catalyzed reaction
(Table 1). For the copper catalyst, aldehyde products arising
from Cα−Cβ bond cleavage in 1T dominate, and no ketone
products were observed. Here, the selectivity of the copper-
catalyzed oxidation of the β-1 lignin model 1T resembles that
observed for β-O-4 lignin model compounds in previously
reported copper-catalyzed reactions. Prior work in our
laboratories found that aerobic oxidation of a β-O-4 lignin
model compound using CuCl/TEMPO afforded primarily
products of C−C bond cleavage.8c,g A similar observation was
noted by Stahl and co-workers, who tested a combination of
CuOTf/bpy/TEMPO/N-methylimidazole for the aerobic
oxidation of a β-O-4 lignin model compound.20 Veratryl
aldehyde was observed as a major product, which was attributed
to initial oxidation of the primary alcohol, followed by a retro-
aldol reaction to break the Cα−Cβ bond.31 Alternatively, the
C−C bond cleavage reaction could proceed by an initial one-
electron oxidation, forming a radical cation that undergoes
subsequent C−C bond fission. In support of this idea, Langan
and co-workers have studied the oxidation of both β-O-4 and
A previous comparison of copper and vanadium catalysts for
the aerobic oxidation of β-O-4 lignin model compounds
revealed that vanadium and copper catalysts exhibited
dramatically different selectivities.8c To further evaluate
selectivity differences between the vanadium and copper
catalysts, the aerobic oxidation of the β-1 lignin models was
also carried out using a copper-based catalytic system. When
compound 1T was heated with a combination of CuOTf (10
mol %) and TEMPO (10 mol %) in toluene with added 2,6-
lutidine (10 equiv with respect to substrate) under O2 (1 atm)
at 100 °C, complete conversion of the starting material was
1
observed within 48 h (Figure 5). Examination of the H NMR
spectrum of the reaction mixture revealed the formation of 3,5-
dimethoxybenzaldehyde (5, 81%) and 4-methoxybenzaldehyde
(6, 69%) as the major products (Figure 5). The carboxylic acids
3114
dx.doi.org/10.1021/cs400636k | ACS Catal. 2013, 3, 3111−3122

ACS Catalysis
Research Article
Figure 7. Comparison of the oxidation of phenolic lignin model compound 2T using catalytic (10 mol %) vs stoichiometric CuOTf/TEMPO.
β‑1 lignin model compounds by several one-electron oxidants;
products of Cα−Cβ bond fission were observed.7 The
manganese peroxidase enzyme, which catalyzes the natural
degradation of lignin, is also proposed to react by a one-
electron oxidation pathway.32
aldehyde products. The identity of ketone 9 was confirmed by
isolation from a larger scale reaction mixture; compound 9 was
1
characterized by H and 13C NMR spectroscopy and HRMS
analysis. In control experiments conducted under identical
conditions with no added catalyst, no reaction of 2T was
observed.
Phenolic β-1 Lignin Models. Prior studies of the aerobic
oxidation of β-O-4 lignin model compounds suggested that
important selectivity differences could occur between phenolic
and nonphenolic lignin model compounds.8d Consequently, we
tested (HQ)2VV(O)(OiPr) as a precatalyst for the aerobic
oxidation of the phenolic β-1 lignin model compounds 2T and
2E. When compound 2T was heated with 10 mol % of the
precatalyst under air in pyridine at 100 °C, complete
consumption of the starting material was observed (Figure
6). The major products of this reaction were the ketone 9
(23%) and 2,6-dimethoxybenzoquinone (10, 60%). In addition,
multiple minor products formed, including 4-hydroxy-3,5-
dimethoxybenzoic acid (13), 4-hydroxy-3,5-dimethoxybenzal-
dehyde (12), 4-methoxybenzoic acid (8), 4-methoxybenzalde-
hyde (6), and several unidentified aldehyde species (1−12%
yields, see Figure 6). In the catalytic aerobic oxidation of the
erythro isomer 2E using (HQ)2VV(O)(OiPr), complete
conversion of the starting material was also observed, affording
the ketone 9 (7%), 2,6-dimethoxybenzoquinone (10, 43%), and
minor amounts of the corresponding carboxylic acid and
In the reactions of the phenolic lignin model compounds 2E
and 2T, an incomplete mass balance was obtained; however, no
additional major products were evident in the 1H NMR
spectrum of the reaction mixture which could account for all of
the missing material, and the reaction mixture was a dark brown
color with some solid material formed. An incomplete mass
balance in the oxidation of lignin model compounds has been
noted in previous reports and has been attributed to
polymerization or other reactions of radical intermediates.33
To evaluate this idea, the aerobic oxidation of 2,6-
dimethoxyphenol was carried out using the vanadium complex
(HQ)2VV(O)(OiPr) (5 mol %) as a precatalyst with added base
(NEt3, 10 mol %). After 24 h of heating at 80 °C, more than
98% of the 2,6-dimethoxyphenol had reacted, affording a dark
brown solid. No soluble organic products were detected in the
1H NMR spectrum of an aliquot of the reaction mixture
(CDCl3), consistent with polymerization or other reactions to
afford intractable solid products.33
3115
dx.doi.org/10.1021/cs400636k | ACS Catal. 2013, 3, 3111−3122

ACS Catalysis
Research Article
Figure 8. Oxidation of phenolic lignin model compound 2E using catalytic (10 mol %) (dipic)VV(O)(OiPr) yields an aldehyde derived from
oxidation of the primary alcohol.
a
Table 2. Comparison of the Vanadium and Copper-Catalyzed Oxidation of the Phenolic β-1 Lignin Model Compound 2
a
The ketone product is highlighted in blue, C−C bond cleavage products are highlighted in green, and the primary alcohol oxidation product 14 is
highlighted in pink.
For the vanadium-catalyzed oxidations, the incorporation of a
phenolic group into the substrate had a significant impact on
the reaction selectivity. Although the nonphenolic lignin model
compound 1T was oxidized at the secondary alcohol position
to give ketone products, the phenolic model compound 2T
underwent Caryl−Cα bond cleavage to generate the 2,6-
dimethoxybenzoquinone product. A similar trend was
previously noted in the oxidation of phenolic and nonphenolic
β-O-4 lignin model compounds using the vanadium catalyst
(HQ)2VV(O)(OiPr).8d The formation of quinone products
upon oxidation of phenolic β-O-4, β-1, and β-5 model
compounds has also been observed by Crestini and co-
workers,11 and Bozell et al.,18 who reported several cobalt-
catalyzed reactions. In the cobalt systems, the Caryl−Cα bond
cleavage was proposed to proceed through a radical pathway
involving a cobalt−superoxo intermediate.34,35
We were interested in evaluating the selectivity of the
copper-catalyzed oxidation of the phenolic β-1 lignin model
compound, but previous work in our lab suggested that the
copper catalyst is inhibited by phenolic functionalities.8c,g
Consequently, the oxidation of the phenolic β-1 lignin model
compound 2T was tested with both catalytic copper (10 mol
%) and stoichiometric copper (Figure 7). With the catalytic
copper load, 2T was heated with CuOTf/TEMPO (10 mol %)
in toluene with added 2,6-lutidine (10 equiv with respect to
substrate) under oxygen in toluene at 100 °C. Complete
consumption of the starting material 2T was observed within
18 h, affording ketone 9 (78%), 2,6-dimethoxybenzoquinone
(10, 8%), 4-methoxybenzaldehyde (6), acrylaldehyde (13),
syringaldehyde (11), and 4-methoxybenzoic acid (8) as
products (1−5% yields, Figure 7). Formation of ketone 9 is
suggestive of a mechanism change as observed previously with
β-O-4 models at low Cu loadings.8g
The reaction selectivity changed when a stoichiometric
amount of CuOTf/TEMPO was used for the oxidation.
Complete consumption of 2T occurred, affording 2,6-
dimethoxybenzoquinone (10, 82%) and 4-methoxybenzalde-
hyde (6, 25%) as the major products. The acrylaldehyde (13,
9%), syringaldehyde (11, 6%), and 4-methoxybenzoic acid (8)
were formed as minor products (Figure 7). The products were
1
identified and quantified by GC/MS and H NMR spectros-
copy. The observed selectivity for C−C bond cleavage in 2T
with stoichiometric copper is consistent with the general
tendency of the copper systems to catalyze C−C bond cleavage.
Given the range of selectivities observed in the aerobic
oxidation of the β-1 lignin model compounds, we tested the 5-
coordinate vanadium complex (dipic)VV(O)(OiPr) as a
precatalyst for the aerobic oxidation of the phenolic β-1 lignin
model 2E. On heating 2E with 10 mol % of the precatalyst
under air (100 °C, 48 h, pyr-d5), complete conversion of the
starting material was observed (1H NMR spectroscopy). The
formation of a new aldehyde product 14 (44%) was detected
(Figure 8). Compound 14 was identified by carrying out a
reaction on a larger scale, isolating the product, and
characterizing it by 1H and 13C NMR spectroscopy and
HRMS. Other minor products of the reaction included the
ketone 9, 4-hydroxy-3,5-dimethoxybenzoic acid (12), 4-
hydroxy-3,5-dimethoxybenzaldehyde (11), 4-methoxybenzoic
acid (8), and 4-methoxybenzaldehyde (6). The formation of
aldehyde 14 was unanticipated; compound 14 was not
observed as a product with the vanadium precatalyst
(HQ)2VV(O)(OiPr), and its mechanism of formation is unclear
at this time. Complex 14 apparently results from oxidation of
the primary alcohol and dehydration of the secondary alcohol
in 2E. The product distributions from each of the vanadium
and copper catalyzed oxidations of the phenolic β-1 model
compound are shown in Table 2.
3116
dx.doi.org/10.1021/cs400636k | ACS Catal. 2013, 3, 3111−3122

ACS Catalysis
Research Article
Figure 9. Aerobic oxidation of a 1:1 mixture of β-1 (1E) and β-O-4 (15) lignin model compounds using the vanadium precatalyst
(HQ)2VV(O)(OiPr).
Figure 10. Aerobic oxidation of a 1:1 mixture of β-1 (1E) and β-O-4 (15) lignin model compounds using CuOTf/TEMPO (20 mol %) with added
2,6-lutidine (10 equiv).
Competitive Oxidation of β-O-4 and β-1 Lignin Model
Compounds. Internal competition experiments were carried
out to assess the relative reactivities of the vanadium and
copper catalysts with β-O-4 and β-1 lignin model compounds.
A 1:1 mixture of 1E and the β-O-4 lignin model compound 15
(Figure 9) was heated under air with the vanadium precatalyst
(HQ)2VV(O)(OiPr) (10 mol %) in pyr-d5 solvent. After 24 h,
∼50% conversion of 15 and ∼43% conversion of 1E were
observed. The products of this reaction were a mixture of
ketone 16 (∼27%, derived from 15) and ketone 3 (∼16%,
derived from 1E). 3,5-Dimethoxybenzaldehyde (5, 5%) and 4-
methoxybenzaldehyde (6, 1%) were also observed as minor
products. The predominant selectivity of the vanadium catalyst
for oxidation of the secondary benzylic alcohol was maintained,
and the β-O-4 lignin model compound reacted only slightly
faster than the β-1.
The oxidation of a 1:1 mixture of β-O-4 and β-1 lignin model
compounds was also carried out using the copper catalyst
system (Figure 10). Model compounds 15 and 1E were heated
in toluene under O2 with a combination of CuOTf (20 mol %),
TEMPO (20 mol %), and 2,6-lutidine (10 equivalents). After
24 h, ∼80% of the β-1 lignin model 1E had reacted, and only
∼56% of the β-O-4 lignin model compound 15 had reacted.
The major products of this reaction consisted of 3,5-
dimethoxybenzaldehyde (5, 75%), 4-methoxybenzaldehyde (6,
32%), and ketone 16 (8%). These results suggest that the β-1
lignin model reacts somewhat faster than the β-O-4 lignin
model with the copper catalyst and confirm that the copper-
catalyzed oxidation reactions proceed with selectivity for C−C
bond cleavage for both substrates.
SUMMARY AND CONCLUSIONS
■
Major selectivity differences are observed in the copper- and
vanadium-catalyzed aerobic oxidation reactions. Although both
metals are effective catalysts for the aerobic oxidation of the
nonphenolic β-1 lignin model compound 1, the vanadium
precatalyst (HQ)2VV(O)(OiPr) affords primarily ketone
products, whereas the copper catalytic system (CuOTf/
TEMPO) generates products of C−C bond cleavage. These
selectivity trends parallel the reactivity previously reported for
the β-O-4 lignin model compounds, likely reflecting different
mechanisms of oxidation between vanadium and copper
catalysts. For the copper catalysts, the C−C bond cleavage
reaction may proceed through a pathway involving either a one-
electron oxidation or a selective oxidation of the primary
alcohol position, followed by a retro-aldol reaction. The
products observed in the vanadium catalyzed reaction are
more consistent with a two-electron oxidation pathway, in line
with previous studies of the reactivity of (HQ)2VV(O)(OiPr).
The reactivity of both catalysts changed dramatically upon
incorporation of a phenolic group into the β-1 lignin model
compound, and a different type of C−C (Caryl−Cα) bond
3117
dx.doi.org/10.1021/cs400636k | ACS Catal. 2013, 3, 3111−3122

ACS Catalysis
Research Article
cleavage was observed upon oxidation of the phenolic β-1
model compound 2. Although the copper catalyst was inhibited
by the phenolic functionality, oxidation of 2 with stoichiometric
CuOTf/TEMPO afforded 2,6-dimethoxybenzoquinone and 4-
methoxybenzaldehyde, coproducts arising from breaking the
Caryl−Cα bond of 2. The vanadium complex (HQ)2VV(O)-
(OiPr) also yielded products of Caryl−Cα bond cleavage, 2,6-
dimethoxybenzoquinone and 4-methoxybenzoic acid. The
incorporation of the phenolic group had a detrimental impact
on the selectivity for the formation of monomeric products,
with lower product yields and some formation of solids
detected with both copper and vanadium catalysts.35
The diversity of pathways available to copper and vanadium
catalysts highlights the challenges associated with controlling
reaction selectivity in the aerobic oxidation of lignin, where
multiple different linkages and functional groups are present.
These results suggest that for the development of catalytic
reactions, testing catalyst reactivity with more than one type of
linkage model may be useful in trying to fully evaluate the
selectivity. Given the complexity of structures present within
lignin as well as the multiple reaction modes available to earth-
abundant metal catalysts, reaction selectivity could be otherwise
difficult to predict. Despite these challenges, the different
selectivities exhibited by the copper and vanadium catalysts
provide a promising indication that it may be possible to
influence or control selectivity in the aerobic oxidation of lignin.
phenyl)propanoate. The filtrate was concentrated and purified
on silica gel with 25% ethyl acetate in hexanes to provide 1.034
g (23.9%) of the erythro isomer (2R,3S)-ethyl 3-(3,5-
dimethoxyphenyl)-3-hydroxy-2-(4-methoxyphenyl)propanoate
and 2.38 g (55.0%) of a mixture of both isomers.
(2S,3S)-Ethyl 3-(3,5-dimethoxyphenyl)-3-hydroxy-2-
(4-methoxyphenyl)propanoate. 1H NMR (300 MHz,
CDCl3) δ 7.26 (d, 2H, J = 8.7 Hz, aryl), 6.86 (d, 2H, J = 8.7
Hz, aryl), 6.45 (d, 2H, J = 2.2 Hz, aryl), 6.36 (t, 1H, J = 2.2 Hz,
aryl), 5.19 (d, 1H, J = 7.2 Hz, ArCHOH), 4.04 (m, 2H,
OCH2CH3), 3.80 (s, 3H, −OCH3), 3.74 (s, 6H, −OCH3), 1.10
(t, 3H, J = 7.2 Hz, OCH2CH3). 13C{1H} NMR (75 MHz,
CDCl3) δ 173.0, 160.7, 159.5, 143.4, 130.5, 126.8, 114.2, 104.7,
100.4, 75.2, 61.2, 58.8, 55.5, 55.4, 14.2.
(2R,3S)-Ethyl 3-(3,5-dimethoxyphenyl)-3-hydroxy-2-
(4-methoxyphenyl)propanoate. 1H NMR (300 MHz,
CDCl3) δ 7.05 (d, 2H, J = 8.7 Hz, aryl), 6.74 (d, 2H, J = 8.7
Hz, aryl), 6.26 (m, 3H, aryl), 5.07 (dd, 1H, J = 9.2, 4.2 Hz,
ArCHOH), 4.19 (m, 2H, OCH2CH3), 3.78 (d, 1H, J = 9.1 Hz),
3.75 (s, 3H, −OCH3), 3.67 (s, 6H, −OCH3), 1.23 (t, 3H, J =
7.1 Hz, OCH2CH3). 13C{1H} NMR (75 MHz, CDCl3) δ 173.9,
160.6, 159.1, 143.5, 129.8, 127.6, 114.0, 104.6, 100.2, 61.4, 59.0,
55.5, 55.4, 19.6, 14.3.
EXPERIMENTAL SECTION
General Considerations. Unless specified otherwise, all
reactions were carried out in the presence of air. H and
■
1
13C{1H} NMR spectra were obtained at room temperature on a
Bruker AV400 MHz spectrometer, with chemical shifts (δ)
referenced to the residual solvent signal. GC/MS analysis was
performed using a Hewlett-Packard 6890 GC system equipped
with a Hewlett-Packard 5973 mass selective detector.
Deuterated solvents were purchased from Cambridge Isotope
Laboratories and dried with molecular sieves. (CuOTf)2·
toluene, 2,6-lutidine, and TEMPO were purchased from
Sigma Aldrich. The phenolic β-1 lignin model compounds
threo-1-(4-hydroxy-3,5-dimethoxyphenyl)-2-(4-methoxyphen-
yl)-propane-1,3-diol (2T), and erythro-1-(4-hydroxy-3,5-dime-
thoxyphenyl)-2-(4-methoxyphenyl)-propane-1,3-diol (2E)
were synthesized according to a literature procedure.28 Oxygen
was purchased from Linde Canada.
Threo-1-(3,5-Dimethoxyphenyl)-2-(4-methoxyphen-
yl)-propane-1,3-diol (1T). To a solution of (2S,3S)-ethyl 3-
(3,5-dimethoxyphenyl)-3-hydroxy-2-(4-methoxyphenyl)-
propanoate (2.14 g, 5.95 mmol) and triethylamine (1.20 g, 1.66
mL, 11.9 mmol) in tetrahydrofuran (27.0 mL) was added
chlorotrimethylsilane (0.97 g, 1.1 mL, 8.9 mmol) dropwise via a
syringe at 0 °C, followed by addition of 4-(N,N-
dimethylamino)pyridine (10 mg). The mixture was allowed
to warm to room temperature and stirred for 15 min and then
cooled using an ice bath. Lithium aluminum hydride (0.452 g,
11.9 mmol) was added portionwise to the reaction mixture, and
then it was allowed to warm to room temperature and stirred
for 30 min, at which time TLC analysis revealed that the
reaction was complete. The mixture was cooled using an ice
bath and then quenched by sequential addition of 450 μL of
H2O, 450 μL of 15% aqueous KOH, and 3 × 450 μL of H2O.
The precipitate was filtered and washed thoroughly with diethyl
ether. The solvent was removed under reduced pressure, and
the residue was purified on silica gel with 45% ethyl acetate in
hexanes to provide 1.202 g (63%) of the title product as a
colorless syrup. 1H NMR (400 MHz, CDCl3) δ 7.13 (d, 2H, J =
8.4 Hz, aryl), 6.85 (d, 2H, J = 8.4 Hz, aryl), 6.38 (d, 2H, J = 2.4
Hz, aryl), 6.36 (t, 1H, J = 2.4 Hz, aryl), 4.88 (d, 1H, J = 6.8 Hz,
ArCHOH), 3.79 (s, 3H, −OCH3), 3.77−3.70 (m, 2H,
CH2OH), 3.73 (s, 6H, −OCH3), 3.07 (q, 1H, J = 7.2 Hz,
CHAr). 13C{1H} NMR (100 MHz, CDCl3): 160.8, 159.0,
144.8, 130.4, 130.3, 114.2, 104.6, 100.0, 75.9, 64.3, 55.5, 55.4,
54.8. ESI/MS: Calculated m/z [1T + Na] = 341.1, found 341.4.
Synthesis of Nonphenolic β-1 Lignin Model Com-
pounds (1T and 1E). To a solution of diisopropylamine (1.85
mL, 13.2 mmol) in THF (18.0 mL) was added n-butyllithium
(8.25 mL, 13.2 mmol, 1.6 M in hexanes) dropwise via syringe
pump at 0 °C. The mixture was stirred at this temperature for
20 min and then cooled to −78 °C. A solution of ethyl 2-(4-
methoxyphenyl)acetate (2.33 g, 12.0 mmol) in THF (12.0 mL)
was added dropwise via a cannula, followed by addition of a
solution of 3,5-dimethoxybenzaldehyde (1.99 g, 12.0 mmol) in
THF (6.0 mL) via a cannula. The mixture was stirred at this
temperature for 1 h, and then saturated ammonium chloride
solution was added to quench the reaction. The mixture was
warmed to room temperature and extracted with diethyl ether;
the combined organic extracts were dried over sodium sulfate
and concentrated under reduced pressure. Diethyl ether was
then added, and the mixture was stirred at room temperature
overnight. The solids were filtered and washed with diethyl
ether to provide 0.191 g (4.4%) of the threo isomer (2S,3S)-
ethyl 3-(3,5-dimethoxyphenyl)-3-hydroxy-2-(4-methoxy-
3118
dx.doi.org/10.1021/cs400636k | ACS Catal. 2013, 3, 3111−3122

ACS Catalysis
Research Article
Erythro-1-(3,5-Dimethoxyphenyl)-2-(4-methoxyphen-
yl)-propane-1,3-diol (1E). To a solution of (2R,3S)-ethyl 3-
(3,5-dimethoxyphenyl)-3-hydroxy-2-(4-methoxyphenyl)-
propanoate (3.024 g, 8.39 mmol) and triethylamine (1.70 g,
2.34 mL, 16.8 mmol) in tetrahydrofuran (48 mL) was added
chlorotrimethylsilane (1.37 g, 1.60 mL, 12.6 mmol) dropwise
via a syringe at 0 °C, followed by addition of 4-(N,N-
dimethylamino)pyridine (10 mg). The mixture was allowed to
warm to room temperature, stirred for 15 min, and then cooled
using an ice bath. Lithium aluminum hydride (0.637 g, 16.8
mmol) was added portionwise to the reaction mixture, and it
was allowed to warm to room temperature and stirred for 30
min, at which time TLC analysis revealed that the reaction was
complete. The mixture was cooled using an ice bath and then
quenched by sequential addition of 640 μL of H2O, 640 μL of
15% aqueous KOH, and 3 × 640 μL of H2O. The precipitate
was filtered and washed thoroughly with diethyl ether. The
solvent was removed from the filtrate under reduced pressure,
and the residue was purified on silica gel with 45% ethyl acetate
in hexanes to provide 2.091 g (78%) of the title product as a
colorless syrup. 1H NMR (400 MHz, CDCl3) δ 7.00 (d, 2H, J =
8.4 Hz, aryl), 6.76 (d, 2H, J = 8.4 Hz, aryl), 6.32 (d, 2H, J = 2.0
Hz, aryl), 6.29 (t, 1H, J = 2.0 Hz, aryl), 4.95 (d, 1H, J = 8.4 Hz,
ArCHOH), 4.17 (dd, 1H, J = 10.8 Hz, J = 7.2 Hz, CH2OH),
3.96 (dd, 1H, J = 10.8 Hz, J = 4.8 Hz, CH2OH), 3.75 (s, 3H,
−OCH3), 3.70 (s, 6H, −OCH3), 3.11 (m, 1H, CHAr). 13C{1H}
NMR (100 MHz, CDCl3): 160.8, 158.7, 145.6, 131.4, 129.7,
114.1, 104.7, 99.9, 66.5, 63.2, 55.5, 55.4, 54.0. Calculated m/z
[1E + Na] = 341.1, found 341.4.
The ketone (3) and dehydrated ketone (4) were isolated by
prep-scale TLC, eluting with 7:3 hexanes:ethylacetate.
Ketone (3). 1H NMR (400 MHz, CDCl3) δ 7.18 (d, 2H, J =
8.8 Hz, aryl), 7.08 (d, 2H, J = 2.0 Hz, aryl), 6.85 (d, 2H, J = 8.8
Hz, aryl), 6.58 (t, 1H, J = 2.0 Hz, aryl), 4.70−4.66 (m, 1H,
ArCH), 4.25−4.21 (m, 1H, CH2OH), 3.86−3.79 (m, 1H,
CH2OH), 3.77 (s, 6H, −OCH3), 3.76 (s, 3H, −OCH3).
13C{1H} NMR (100 MHz, CDCl3): 200.8, 160.9, 159.3, 138.4,
129.6, 114.9, 107.1, 105.8, 65.3, 55.9, 55.7, 55.4, 42.9. HRMS
(EI): m/z calcd for C18H20O5 [M + H]+: 317.1389; found:
317.1387.
1
Dehydrated Ketone (4). H NMR (400 MHz, CDCl3) δ
7.36 (d, 2H, J = 8.8 Hz, aryl), 7.06 (d, 2H, J = 2.4 Hz, aryl),
6.88 (d, 2H, J = 8.8 Hz, aryl), 6.65 (t, 1H, J = 2.4 Hz, aryl), 5.96
(s, 1H, CCHH), 5.36 (s, 1H, CCHH), 3.82 (s, 3H,
−OCH3), 3.81 (s, 6H, −OCH3). 13C{1H} NMR (100 MHz,
CDCl3): 197.7, 160.9, 160.0, 147.8, 139.3, 128.5, 119.1, 114.3,
108.0, 105.9, 105.2, 55.8, 55.5. HRMS (EI): m/z calcd for
C18H18O4 [M + H]+: 299.1283; found: 299.1277.
Catalytic Oxidation of Nonphenolic β-1 Lignin Model
1T Using 10 mol % (HQ)2V(O)(OiPr) in DMSO-d6.
Compound 1T (20 mg, 0.063 mmol) was oxidized following
the general procedure for the vanadium-catalyzed reactions
using DMSO-d6 (1 mL) as the solvent. After 48 h, integration
of the 1H NMR spectrum against the internal standard revealed
91% conversion of the starting material, affording the ketone
(3, 66%), the dehydrated ketone (4, 6%), 4-methoxybenzoic
acid (8, 11%), 3,5-dimethoxybenzoic acid (7, 4%), 4-
methoxybenzaldehyde (6, 3%), 3,5-dimethoxybenzaldehyde
(5, 3%), and methanol (2%).
General Procedure for the Vanadium Catalyzed
Oxidation of β-1 Lignin Model Compounds. In an NMR
tube, the lignin model compound (∼20 mg, ∼0.06 mmol) was
dissolved in pyr-d5 (1 mL) containing dimethylsulfone added as
an internal standard. An initial 1H NMR spectrum was
recorded, and then the solution was added to a 50 mL
round-bottom flask containing (HQ)2V(O)(OiPr) (∼2.5 mg,
∼0.006 mmol). The flask was equipped with a stir bar and an
air condenser and then heated with stirring under air at 100 °C
for 48 h. The reaction mixture was cooled to room temperature
and then transferred into an NMR tube. The extent of
conversion and yields of the products were determined by
integration of the 1H NMR spectrum against the internal
standard.
Catalytic Oxidation of Nonphenolic β-1 Lignin Model
1E Using 10 mol % (HQ)2V(O)(OiPr) in DMSO-d6. This
reaction was carried out following the general procedure for the
vanadium-catalyzed aerobic oxidations. After 48 h, examination
of the 1H NMR spectrum against the internal standard revealed
42% of 1E reacted, affording the ketone (3, 27%), the
dehydrated ketone (4, <1%), 4-methoxybenzoic acid (8, 10%),
3,5-dimethoxybenzoic acid (7, 3%), 4-methoxybenzaldehyde
(6, 4%), 3,5-dimethoxybenzaldehyde (5, 8%), and methanol
(2%).
Catalytic Oxidation of Nonphenolic β-1 Lignin Model
1E Using 10 mol % (HQ)2V(O)(OiPr) in Toluene with
Added NEt3 (15 mol %). This reaction was carried out using
the general procedure for the vanadium-catalyzed oxidations,
except using toluene as the solvent and triethylamine (15 mol
%) as a base promoter. In addition, the reaction was heated
under air at 100 °C using a water condenser. After 48 h,
Catalytic Oxidation of Nonphenolic β-1 Lignin Model
1T Using 10 mol % (HQ)2V(O)(OiPr) in Pyr-d5. Compound
1T (20 mg, 0.063 mmol) was oxidized following the general
procedure described above for the vanadium-catalyzed
1
examination of the H NMR spectrum revealed that all of 1E
1
reactions. After 48 h, integration of the H NMR spectrum
had been consumed, affording the ketone 3 (72%), 3,5-
dimethoxybenzaldehyde (5, 7%), and 4-methoxybenzaldehyde
(6, 2%).
against the internal standard revealed that all of 1T had reacted,
affording the ketone (3, 34%), the dehydrated ketone (4, 57%),
4-methoxybenzoic acid (8, 4%), 3,5-dimethoxybenzoic acid (7,
3%), 4-methoxybenzaldehyde (6, 3%), and 3,5-dimethoxyben-
zaldehyde (5, 3%).
Catalytic Oxidation of Nonphenolic β-1 Lignin Model
1E Using 10 mol % (HQ)2V(O)(OiPr) in Pyr-d5. This
reaction was carried out using the general procedure for the
vanadium-catalyzed oxidations. After 48 h, examination of the
1H NMR spectrum revealed that 76% conversion had occurred,
affording the ketone (3, 29%), dehydrated ketone (4, 38%), 4-
methoxybenzoic acid (8, 6%), 3,5-dimethoxybenzoic acid (7,
8%), 4-methoxybenzaldehyde (6, 4%), and 3,5-dimethoxyben-
zaldehyde (5, 4%).
General Procedure for the Aerobic Oxidation of β-1
Lignin Model Compounds Using CuOTf/TEMPO (10 mol
%) with 2,6-Lutidine (10 equiv). In an NMR tube, the β-1
lignin model compound (∼12 mg, ∼0.038 mmol) was
dissolved in CDCl3 (1 mL) containing dimethylsulfone (∼1
mg, ∼0.01 mmol) as an internal standard. An initial spectrum
was recorded, and then the solvent was removed under vacuum.
The solid material was redissolved in toluene (5 mL) and then
transferred to a thick-walled 50 mL Schlenk tube equipped with
Teflon stopcock containing [(CuOTf)2·toluene] (∼2 mg,
∼0.004 mmol), TEMPO (0.6 mg, 0.004 mmol), and 2,6-
lutidine (∼0.044 mL, ∼0.38 mmol) under air. Oxygen was
3119
dx.doi.org/10.1021/cs400636k | ACS Catal. 2013, 3, 3111−3122

ACS Catalysis
Research Article
bubbled into the reaction mixture for 3 min, and the reactor
was sealed. The reaction mixture was heated at 100 °C with
constant stirring. After 18 h, the reaction was cooled to room
temperature, oxygen was bubbled again through the solution
for 2 min, the reactor was sealed, and the reaction mixture was
heated at 100 °C. After 48 h, the reaction was cooled to room
temperature. An aliquot was taken from the reaction mixture,
the solvent was removed under weak vacuum (∼100 Torr), and
CDCl3 was added to the residue. The mixture was then filtered
through glass wool, and the extent of conversion and yields of
the products were determined by 1H NMR spectroscopy
(integration of the spectrum against the internal standard).
Where indicated, the identity of the products was also verified
by GC/MS analysis.
Catalytic Oxidation of Phenolic β-1 Lignin Model 2T
Using 10 mol % (HQ)2V(O)(OiPr) in Pyr-d5. This reaction
was carried out using the general procedure for the vanadium-
catalyzed oxidations. After 24 h of heating at 100 °C, the
starting material 2T was completely consumed, and a mixture
of products was detected, including ketone (9, 23%), 2,6-
dimethoxybenzoquinone (10, 60%), 4-methoxybenzoic acid (8,
12%), 4-hydroxy-3,5-dimethoxybenzoic acid (12, 10%), 4-
methoxybenzaldehyde (6, 1%), 4-hydroxy-3,5-dimethoxyben-
zaldehyde (11, 1%), and several minor unidentified aldehyde
products (∼2−6%).
Catalytic Oxidation of Phenolic β-1 Lignin Model 2E
Using 10 mol % (HQ)2V(O)(OiPr) in Pyr-d5. This reaction
was carried out using the general procedure for the vanadium-
catalyzed aerobic oxidations. After 24 h, a mixture of reaction
products was formed, including ketone (9, 7%), 2,6-
dimethoxybenzoquinone (10, 43%), 4-methoxybenzoic acid
(8, 14%), 4-hydroxy-3,5-dimethoxybenzoic acid (12, 7%), 4-
methoxybenzaldehyde (6, 6%), 4-hydroxy-3,5-dimethoxyben-
zaldehyde (11, 3%), and several minor unidentified aldehyde
products (1−6%).
Catalytic Oxidation of Nonphenolic β-1 Lignin Model
1T Using 10 mol % CuOTf/TEMPO with 2,6-Lutidine (10
equiv) in Toluene. Compound 1T was oxidized following the
general procedure for the copper-catalyzed aerobic oxidation
1
reactions. After 48 h, integration of the H NMR spectrum
against the internal standard revealed that all of 1T had reacted,
with the products consisting of 3,5-dimethoxybenzaldehyde (5,
81%), 4-methoxybenzaldehyde (6, 69%), 3,5-dimethoxybenzoic
acid (7, 1%), and 4-methoxybenzoic acid (8, 1%). The products
The ketone (9) was isolated by chromatography on silica gel,
eluting first with 7:3 hexanes/ethylacetate, then 4:6 hexanes/
ethyl acetate.
1
were identified and quantified by H NMR spectroscopy and
1
Ketone (9). H NMR (400 MHz, CDCl3) δ 7.24 (s, 2H,
GC/MS. Yields are expressed as a percentage of the theoretical
aryl), 7.18 (d, 2H, J = 8.4 Hz, aryl), 6.85 (d, 2H, J = 8.4 Hz,
aryl), 4.68 (dd, 1H, J = 8.8 Hz, J = 4.8 Hz, ArCH), 4.24 (1H,
dd, J = 11.2 Hz, J = 8.4 Hz, CH2OH), 3.88−3.82 (m, 1H,
CH2OH), 3.86 (s, 6H, −OCH3), 3.76 (s, 3H, −OCH3).
13C{1H} NMR (100 MHz, CDCl3): 198.7, 159.2, 146.8, 140.0,
129.5, 127.8, 114.9, 106.6, 105.2, 65.4, 56.5, 55.6, 55.4. HRMS
(EI): m/z calcd for C18H20O6 [M + H]+: 333.1338; found:
333.1352
maximum based on the initial amount of substrate.
Catalytic Oxidation of Nonphenolic β-1 Lignin Model
1E Using 10 mol % CuOTf/TEMPO with 2,6-Lutidine (10
equiv) in Toluene. This reaction was carried out according to
the general procedure for the copper-catalyzed aerobic
1
oxidations. After 48 h, integration of the H NMR spectrum
against the internal standard revealed that all of 1E had been
consumed, with the products consisting of 3,5-dimethoxyben-
zaldehyde (5, 78%), 4-methoxybenzaldehyde (6, 60%), 3,5-
dimethoxybenzoic acid (7, 1%), and 4-methoxybenzoic acid (8,
1%). The products were identified and quantified by H NMR
and GC/MS.
Control Reaction: Attempted Oxidation of 2E with No
Catalyst. In an NMR tube, 2E (19.5 mg, 0.058 mmol) was
dissolved in pyr-d5 (0.8 mL) containing dimethylsulfone as an
1
1
internal standard. An initial H NMR spectrum was recorded,
Attempted Oxidation of Nonphenolic β-1 Lignin
Model 1T Using 10 mol % TEMPO with 2,6-Lutidine
(10 equiv) in Toluene (control experiment with no
copper catalyst). This reaction was carried out using the
general procedure for the copper complex-catalyzed aerobic
oxidations, except that no copper catalyst was added. After 48
and then the reaction mixture was heated under air with stirring
in a 25 mL round-bottom flask at 80 °C for 4 days. The
reaction mixture was then cooled to room temperature.
Examination of the 1H NMR spectrum revealed that no
reaction occurred.
Catalytic Oxidation of Phenolic β-1 Lignin Model 2T
Using 10 mol % (HQ)2V(O)(OiPr) in DMSO-d6. This
reaction was carried out using the general procedure for the
vanadium-catalyzed aerobic oxidations. After 48 h, examination
of the 1H NMR spectrum against the internal standard revealed
all 2T reacted, affording 2,6-dimethoxybenzoquinone (10,
44%), 4-methoxybenzoic acid (8, 35%), methanol (16%), 4-
methoxybenzaldehyde (6, 2%), 4-hydroxy-3,5-dimethoxyben-
zoic acid (12, <1%), 4-hydroxy-3,5-dimethoxybenzaldehyde
(11, <1%), and other unidentified aldehyde products (3%).
Catalytic Oxidation of Phenolic β-1 Lignin Model 2T
Using 10 mol % CuOTf/TEMPO with 2,6-Lutidine (10
equiv) in Toluene. This reaction was carried out using the
general procedure for the copper-catalyzed aerobic oxidations.
After 18 h of reaction time, integration of the 1H NMR
spectrum against the internal standard revealed that all of 2T
had reacted, with the products consisting of ketone (9, 78%), 2-
(4-methoxyphenyl)acrylaldehyde (13, 5%), 2,6-dimethoxyben-
zoquinone (10, 8%), 4-hydroxy-3,5-dimethoxybenzaldehyde
(11, 2%), 4-methoxybenzaldehyde (6, 4%), and 4-methox-
1
h, the reaction mixture was examined by H NMR spectros-
copy. No reaction had occurred.
Attempted Oxidation of Nonphenolic β-1 Lignin
Model 1E Using 10 mol % TEMPO with 2,6-Lutidine
(10 equiv) in Toluene (control experiment with no
copper catalyst). This control experiment was carried out
using the same general procedure for the copper complex-
catalyzed aerobic oxidations, except that no copper catalyst was
added. After 48 h, an aliquot of the reaction mixture was
examined by 1H NMR spectroscopy. No reaction had occurred.
Catalytic Oxidation of Nonphenolic β-1 Lignin Model
1E Using 10 mol % CuOTf/TEMPO in Pyridine. This
reaction was carried out using the general procedure for the
copper-catalyzed aerobic oxidations, except pyridine was used
instead of toluene and 2,6-lutidine. After 48 h of reaction time,
integration of the 1H NMR spectrum against the internal
standard revealed that 60% of 1E had reacted, with the
products consisting of 3,5-dimethoxybenzaldehyde (5, 48%), 4-
methoxybenzaldehyde (6, 15%), and ketone 3 (3%).
3120
dx.doi.org/10.1021/cs400636k | ACS Catal. 2013, 3, 3111−3122

ACS Catalysis
Research Article
ybenzoic acid (8, 1%). The products were identified and
quantified by both H NMR and GC/MS.
dimethoxybenzaldehyde (11, 5%), 4-methoxybenzoic acid (8,
9%), and 4-methoxybenzaldehyde (6, 3%).
1
The aldehyde (14) was isolated by prep-scale TLC eluting
with 4:6 hexanes/ethyl acetate.
Stoichiometric Oxidation of Phenolic β-1 Lignin
Model 2T Using CuOTf/TEMPO with 2,6-Lutidine (10
equiv) in Toluene. This reaction was carried out using the
same general procedure for the copper-catalyzed reactions,
except that a stoichiometric amount of CuOTf/TEMPO was
added. After 18 h, integration of the 1H NMR spectrum against
the internal standard revealed that all of 2T had reacted, with
the products consisting of 2,6-dimethoxybenzoquinone (10,
82%), 2-(4-methoxyphenyl)acrylaldehyde (13, 9%), 4-hydroxy-
3,5-dimethoxybenzaldehyde (11, 6%), 4-methoxybenzaldehyde
(6, 25%), and 4-methoxybenzoic acid (8, 1%). The products
were identified and quantified by 1H NMR and GC/MS. Yields
are expressed as a percentage of the theoretical maximum based
on the initial amount of substrate.
1
14: H NMR (400 MHz, CDCl3) δ 9.72 (s, 1H, HCO),
7.26 (s, 1H, HCC), 7.17 (d, 2H, J = 8.4 Hz, aryl), 6.99 (d,
2H, J = 8.4 Hz, aryl), 6.54 (s, 2H, aryl), 3.83 (s, 3H, −OCH3),
3.66 (s, 6H, −OCH3). 13C{1H} NMR (100 MHz, CDCl3):
194.2, 159.8, 150.6, 146.9, 140.0, 137.2, 131.1, 126.2, 125.7,
114.7, 108.3, 56.2, 55.6. HRMS (EI): m/z calcd for C18H18O5
[M + H]+: 315.1232; found: 315.1223.
Catalytic Oxidation of Phenolic β-1 Lignin Model 2E
Using 10 mol % CuOTf/TEMPO with 2,6-Lutidine (10
equiv) in Toluene. This reaction was carried out using the
same general procedure for the copper-catalyzed aerobic
1
oxidations. After 18 h, integration of the H NMR spectrum
Control Reaction: Attempted Oxidation of 2T with No
Catalyst. In an NMR tube, 2T (19.5 mg, 0.058 mmol) was
dissolved in pyridine-d5 (0.8 mL) containing dimethylsulfone
added as an internal standard. The solution was transferred to a
25 mL round-bottom flask and then heated at 80 °C under air
with stirring for 4 days. The reaction mixture was cooled to
against the internal standard revealed that all of 2E had reacted,
with the products consisting of ketone (9, 82%), 2-(4-
methoxyphenyl)acrylaldehyde (13, 3%), 2,6-dimethoxybenzo-
quinone (10, 7%), 4-hydroxy-3,5-dimethoxybenzaldehyde (11,
1%), 4-methoxybenzaldehyde (6, 5%), and 4-methoxybenzoic
acid (8, 1%). The products were identified and quantified by
1H NMR and GC/MS.
1
room temperature and then examined by H NMR spectros-
copy. Integration against the internal standard revealed that no
reaction had occurred.
Stoichiometric Oxidation of Phenolic β-1 Lignin
Model 2E using CuOTf/TEMPO with 2,6-Lutidine (10
equiv) in Toluene. This reaction was carried out using the
same general procedure as for the copper-catalyzed aerobic
oxidations, except that a stoichiometric amount of CuOTf/
Catalytic Oxidation of a 1:1 Mixture of Nonphenolic
β-1 Lignin Model 1E and β-O-4 Model 15 Using 10 mol
% (HQ)2V(O)(OiPr) in Pyr-d5. A mixture containing a 1:1 ratio
of 1E and 15 was oxidized following the general procedure for
the vanadium-catalyzed aerobic oxidation reactions. After 24 h,
integration of the 1H NMR spectrum against the internal
standard revealed that 43% of 1E and 50% of 15 had reacted,
with the products consisting of ketone 16 (27%, derived from
the β-O-4 lignin model), ketone 3 (16%, derived from the β-1
lignin model), 3,5-dimethoxybenzaldehyde (5, 5%), and 4-
methoxybenzaldehyde (6, 1%).
1
TEMPO was added. After 18 h, integration of the H NMR
spectrum against the internal standard revealed that all of the
2E had reacted, with the products consisting of 2,6-
dimethoxybenzoquinone (10, 68%), 2-(4-methoxyphenyl)-
acrylaldehyde (13, 7%), 4-hydroxy-3,5-dimethoxybenzaldehyde
(11, 3%), 4-methoxybenzaldehyde (6, 14%), and 4-methox-
ybenzoic acid (8, 2%). The products were identified and
1
quantified by H NMR and GC/MS.
Catalytic Oxidation of a 1:1 Mixture of Nonphenolic
β-1 Model 1E and β-O-4 Model 15 Using 20 mol %
CuOTf/TEMPO with 2,6-Lutidine (10 equiv) in Toluene. A
mixture containing a 1:1 ratio of 1E and 15 was oxidized
following the general procedure for the copper-catalyzed
Attempted Oxidation of Phenolic β-1 Lignin Model 2T
Using TEMPO with 2,6-Lutidine (10 equiv) in Toluene
(control experiment with no copper catalyst). This
control experiment was carried out using the general procedure
for the copper-catalyzed aerobic oxidations, except that no
copper catalyst was added. After 18 h, the examination of the
1H NMR spectrum revealed that no reaction had occurred.
Oxidation of Phenolic β-1 Lignin Model 2E Using
TEMPO with 2,6-Lutidine (10 equiv) in Toluene (control
experiment with no added copper catalyst). This control
experiment was carried out according to the general procedure
for the copper-catalyzed aerobic oxidations, except that no
copper catalyst was added. After 18 h, the examination of the
1H NMR spectrum revealed that no reaction had occurred.
Catalytic Oxidation of Phenolic β-1 Lignin Model 2E
Using (dipic)VV(O)(OiPr). This reaction was carried out
following the general procedure for the vanadium-catalyzed
aerobic oxidation reactions. After 48 h, complete conversion of
the starting material was observed. The products of the reaction
consisted of the aldehyde 14 (44%), the ketone 9 (7%), 4-
hydroxy-3,5-dimethoxybenzoic acid (12, 9%), 4-hydroxy-3,5-
1
aerobic oxidation reactions. After 24 h, integration of the H
NMR spectrum against the internal standard revealed that 80%
of 1E and 56% of 15 had reacted, with the products consisting
of 3,5-dimethoxybenzaldehyde (5, 75%), 4-methoxybenzalde-
hyde (6, 32%), and ketone 16 (8%, derived from the β-O-4
lignin model).
ASSOCIATED CONTENT
* Supporting Information
■
S
Additional experimental details and NMR spectra of reaction
products. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: rbaker@uottawa.ca.
*E-mail: skhanson@lanl.gov.
3121
dx.doi.org/10.1021/cs400636k | ACS Catal. 2013, 3, 3111−3122

ACS Catalysis
Research Article
(12) Badamali, S. K.; Luque, R.; Clark, J. H.; Breeden, S. W. Catal.
Commun. 2011, 12, 993−995.
Notes
The authors declare no competing financial interest.
(13) Wang, B.; Long, Z. Appl. Mech. Mater. 2011, 80−81, 350−354.
(14) Rodrigues Pinto, P. C.; Borges da Silva, E. A.; Rodrigues, A. E.
Ind. Eng. Chem. Res. 2011, 52, 741−748.
ACKNOWLEDGMENTS
■
(15) Rovio, S.; Kallioinen, A.; Tamminen, T.; Hakola, M.; Leskela,
M.; Siika-Aho, M. BioResources 2012, 7, 756−776.
(16) Das, L.; Kolar, P.; Sharma-Shivappa, R. Biofuels 2012, 3, 155−
166.
(17) Maa, P.; Fu, S.; Zhai, H.; Law, K.; Daneault, C. Bioresour.
Technol. 2012, 118, 607−610.
(18) (a) Bozell, J. J.; Hames, B. R.; Dimmel, D. R. J. Org. Chem. 1995,
60, 2398−2404. (b) Cedeno, D.; Bozell, J. J. Tetrahedron Lett. 2012,
53, 2380−2383. (c) Biannic, B.; Bozell, J. J. Org. Lett. 2013, 15, 2730−
2733.
This work was supported in part by the NSF under the CCI
Center for Enabling New Technologies through Catalysis
(CENTC) Phase II Renewal, CHE-1205189. R.T.B. thanks
NSERC for support through the Lignoworks strategic research
network and uOttawa, Canada Foundation for Innovation, and
Ontario Ministry of Economic Development and Innovation
for essential infrastructure. S.K.H. thanks Los Alamos National
Laboratory LDRD (20100160ER) for funding.
(19) Liu, S.; Shi, Z.; Li, L.; Yu, S.; Xie, C.; Song, Z. RSC Adv. 2013, 3,
REFERENCES
5789−5793.
■
(20) Rahimi, A.; Azarpira, A.; Kim, H.; Ralph, J.; Stahl, S. S. J. Am.
Chem. Soc. 2013, 135, 6415−6418.
(1) (a) Corma, A.; Iborra, S.; Velty, A. Chem. Rev. 2007, 107, 2411−
2502. (b) Dodds, D. R.; Gross, R. A. Science 2007, 318, 1250−1251.
(c) Centi, G., van Santen, R. A., Eds. Catalysis for Renewables. From
Feedstocks to Energy Production.; Wiley-VCH: Weinheim, 2008.
(d) Gallezot, P. ChemSusChem 2008, 1, 734−737. (e) Dhepe, P. L.;
Fukuoka, A. ChemSusChem 2008, 1, 969−975. (f) Lin, Y.-C.; Huber,
G. W. Energy Environ. Sci. 2009, 2, 68−80. (g) Alonso, D. M.; Bond, J.
Q.; Dumesic, J. A. Green Chem. 2010, 12, 1493−1513. (h) Rinaldi, R.;
(21) (a) Hedges, J. I.; Ertel, J. R. Anal. Chem. 1982, 54, 174−178.
(b) Goni, M. A.; Hedges, J. I. Geochem. Cosmochim. Acta 1992, 56,
4025−4043.
(22) Using a Co-salen catalyst, Biannic and Bozell recently converted
both aromatic nuclei of this phenolic lignin model to benzoquinone
products; see ref 18c.
(23) Hoover, J. M.; Ryland, B. L.; Stahl, S. S. J. Am. Chem. Soc. 2013,
Schuth, F. ChemSusChem 2009, 2, 1096−1107. (i) Domínguez de
̈
135, 2357−2367.
María, P. ChemSusChem 2011, 4, 327−329. (j) Gasser, C. A.;
(24) Sannigrahi, P.; Ragauskas, A. J.; Miller, S. J. Bioenerg. Res. 2008,
1, 205−214.
Hommes, G.; Schaffer, A.; Corvini, P. F.-X. Appl. Microbiol. Biotechnol.
̈
2012, 95, 1115−1134.
(25) Ede, R. M.; Ralph, J.; Torr, K. M.; Dawson, B. S. W.
Holzforschung 1996, 50, 161−164.
(2) (a) Nimz, H. Angew. Chem., Int. Ed. 1974, 13, 313−312.
(b) Ralph, J.; Lundquist, K.; Brunow, G.; Lu, F.; Kim, H.; Schatz, P. F.;
Marita, J. M.; Hatfield, R. D.; Ralph, S. A.; Christensen, J. H.; Boerjan,
W. Phytochem. Rev. 2004, 3, 29−60. (c) Lundquist, K.; Lundgren, R.
Acta Chem. Scand. 1972, 26, 2005−2023. (d) Calvo-Flores, F. G.;
Dobado, J. A. ChemSusChem 2009, 2, 1227−1235.
(26) Setala, H.; Pajunen, A.; Rummakko, P.; Sipila, J.; Brunow, G. J.
̈
̈
Chem. Soc., Perkin Trans. 1999, 461−464.
(27) Brunow, G. Methods to Reveal the Structure of Lignin; Wiley-
VCH, Weinheim, 2005.
(28) Reddy, G. V. B.; Sridhar, M.; Gold, M. H. Eur. J. Biochem. 2003,
270, 284−292.
(29) Yields are expressed as a percentage of the theoretical maximum
based on the initial amount of substrate.
(3) Sannigrahi, P.; Pu, Y. Q.; Ragauskas, A. Curr. Opin. Environ. Sus.
2010, 2, 383−393.
(4) Adler, E. Wood Sci. Technol. 1977, 11, 169.
(5) (a) Zakzeski, J.; Bruijnincx, P. C. A.; Jongerius, A. L.;
Weckhuysen, B. M. Chem. Rev. 2010, 110, 3552−3599 and references
therein. (b) Zakzeski, J.; Bruijnincx, P. C. A.; Weckhuysen, B. M. Green
Chem. 2011, 13, 671−680.
(6) Deng, H.; Lin, L.; Liu, S. Energy Fuels 2010, 24, 4797−4802 and
references therein.
(7) (a) Cho, D. W.; Parthasarathi, R.; Pimentel, A. S.; Maestas, G. D.;
Park, H. J.; Yoon, U. C.; Dunaway-Mariano, D.; Gnanakaran, S.;
Langan, P.; Mariano, P. S. J. Org. Chem. 2010, 75, 6549−6562.
(b) Cho, D. W.; Latham, J. A.; Park, H. J.; Yoon, U. C.; Langan, P.;
Dunaway-Mariano, D.; Mariano, P. S. J. Org. Chem. 2011, 76, 2840−
2852.
(8) (a) Hanson, S. K.; Baker, R. T.; Gordon, J. C.; Scott, B. L.;
Thorn, D. L. Inorg. Chem. 2010, 49, 5611−5618. (b) Hanson, S. K.;
Baker, R. T.; Gordon, J. C.; Scott, B. L.; Silks, L. A.; Thorn, D. L. J.
Am. Chem. Soc. 2010, 132, 17804−17816. (c) Sedai, B.; Díaz -Urrutia,
C.; Baker, R. T.; Wu, R.; Silks, L .A.; Hanson, S. K. ACS Catal. 2011, 1,
794−804. (d) Hanson, S. K.; Ruilian, W.; Silks, L. A. Angew. Chem., Int.
Ed. 2012, 51, 3410−3413. (e) Zhang, G.; Scott, B. L.; Wu, R.; Silks, L.
A.; Hanson, S. K. Inorg. Chem. 2012, 51, 7354−7361. (f) Wigington, B.
N.; Drummond, M. L.; Cundari, T. R.; Thorn, D. L.; Hanson, S. K.;
Scott, S. L. Chem.Eur. J. 2012, 18, 14981−14988. (g) Sedai, B.;
Baker, R. T. 2013, submitted for publication.
(30) In previous work (refs 8a−c, e) we showed that oxovanadium
complex-catalyzed aerobic oxidation of simple lignin models gave
different selectivities in pyridine and DMSO solvents.
(31) Tarabanko, V. E.; Fomova, N. A.; Kuznetsov, B. N.; Ivanchenko,
N. M.; Kudryashev, A. V. React. Kinet. Catal. Lett. 1995, 55, 161−170.
(32) Forrester, I. T.; Grabski, A. C.; Burgess, R. R.; Leatham, G. F.
Biochem. Biophys. Res. Commun. 1988, 157, 992−999.
(33) (a) Pardini, V. L.; Smith, C. Z.; Utley, J. H. P.; Vargas, R. R.;
Viertler, H. J. Org. Chem. 1991, 56, 7305−7313. (b) Pardini, V. L.;
Vargas, R. R.; Viertler, H. Tetrahedron 1992, 48, 7221−7228.
(34) Zombeck, A.; Drago, R. S.; Corden, B. B.; Gaul, J. H. J. Am.
Chem. Soc. 1981, 103, 7580−7585.
(35) Higher yields of quinone products (up to 86%) have been
obtained from syringyl β-O-4 models using Co-salen complexes in the
presence of a bulky amine that can be employed at room temperature
(see ref 18b).
(9) (a) Son, S.; Toste, F. D. Angew. Chem. 2010, 122, 3879−7982.
(b) Chan, J. M. W.; Bauer, S.; Sorek, H.; Sreekumar, S.; Wang, K.;
Toste, F. D. ACS Catal. 2013, 3, 1369−1377.
(10) Eisenstadt, M. A.; Bogolitsyn, K. G. Russ. J. Bioorg. Chem. 2010,
36, 802−815.
(11) (a) Crestini, C.; Crucianelli, M.; Orlandi, M.; Saladino, R. Catal.
Today 2010, 156, 8−22. (b) Lange, H.; Decina, S.; Crestini, C. Eur.
Polym. J. 2013, 49, 1151−1173.
3122
dx.doi.org/10.1021/cs400636k | ACS Catal. 2013, 3, 3111−3122