Efficient cobalt-catalyzed oxidative conversion of lignin models to benzoquinones

Article abstract of DOI:10.1021/ol401065r

Phenolic lignin model monomers and dimers representing the primary substructural units of lignin were successfully oxidized to benzoquinones in high yield with molecular oxygen using new Co-Schiff base catalysts bearing a bulky heterocyclic nitrogen base as a substituent. This is the first example of a catalytic system able to convert both S and G lignin model phenols in high yield, a process necessary for effective use of lignin as a chemical feedstock.

Full text of DOI:10.1021/ol401065r

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Efficient Cobalt-Catalyzed Oxidative
Conversion of Lignin Models to
Benzoquinones
Berenger Biannic and Joseph J. Bozell*
Center for Renewable Carbon, Center for the Catalytic Conversion of Biomass (C3Bio),
University of Tennessee, Knoxville, Tennessee 37917, United States
jbozell@utk.edu
Received April 17, 2013
ABSTRACT
Phenolic lignin model monomers and dimers representing the primary substructural units of lignin were successfully oxidized to benzoquinones
in high yield with molecular oxygen using new Co-Schiff base catalysts bearing a bulky heterocyclic nitrogen base as a substituent. This is the first
example of a catalytic system able to convert both S and G lignin model phenols in high yield, a process necessary for effective use of lignin as a
chemical feedstock.
The recognition that biomass is a viable raw material
for chemicals and fuels1 has brought increased recent
attention to lignin.2 Lignin comprises as much as 25 wt %
of lignocellulosic biomass, making it the second most
abundant source of renewable carbon in nature, and a
potentially rich source of biobased aromatics. However,
ligninconversionpresentsa significant challenge aslignin’s
biosynthesis from syringyl (S), guaiacyl (G), and p-hydroxy-
phenyl (H) subunits 1, 2, and 3 (Figure 1)3 leads to
structural heterogeneity that frustrates its utility as a
chemical feedstock. Recent reports describe systems
that deconstruct lignin models, using reductive processes
catalyzed by Ru,4 Ni,5 or Pd,6 or both oxidative and
nonoxidative processes catalyzed by Co7 or V.8 These
studies focus on the cleavage of β-aryl ethers, representa-
tive of lignin’s β-O-4 linkage, which can account for
50À65% of the interunit bonding present in native lignin.9
Native lignin, however, does not retain its original
substructural profile when it is isolated from a lignocellu-
losic matrix. Biorefining processes designed for the pre-
treatment or fractionation of biomass into its individual
components (lignin, cellulose, and hemicellulose)10 intro-
duce significant changes to the native lignin structure. For
example, solvent fractionation, acid hydrolysis, or cata-
lyzed steam explosion of biomass can nearly eliminate
β-aryl ethers.11
Moreover, as these units are cleaved, the number of free
(i.e., nonetherified) phenolic ÀOH groups present in both
(1) (a) Bozell, J. J.; Petersen, G. R. Green Chem. 2010, 12, 539–554.
(b) Bozell, J. J. Clean-Soil Air Water 2008, 36, 641–647.
(2) (a) Zakzeski, J.; Bruijnincx, P. C. A.; Jongerius, A. L.; Weckhuysen,
B. M. Chem. Rev. 2010, 110, 3552–3599. (b) Sannigrahi, P.; Pu, Y. Q.;
Ragauskas, A. Curr. Opin. Environ. Sus. 2010, 2, 383–393.
(3) Boerjan, W.; Ralph, J.; Baucher, M. Ann. Rev. Plant Biol. 2003,
54, 519–546.
(4) (a) Wu, A.; Patrick, B. O.; Chung, E.; James, B. R. Dalton Trans.
2012, 41, 11093–11106. (b) Nichols, J. M.; Bishop, L. M.; Bergman,
R. G.; Ellman, J. A. J. Am. Chem. Soc. 2010, 132, 12554–12555.
(5) (a) Sergeev, A. G.; Hartwig, J. F. Science 2011, 332, 439–443. (b)
He, J. Y.; Zhao, C.; Lercher, J. A. J. Am. Chem. Soc. 2012, 134, 20768–
20775.
(7) Zakzeski, J.; Bruijnincx, P. C. A.; Weckhuysen, B. M. Green
Chem. 2011, 13, 671–680.
(8) (a) Hanson, S. K.; Wu, R.; Silks, L. A. P. Angew. Chem., Int. Ed.
2012, 51, 3410–3413. (b) Son, S.; Toste, F. D. Angew. Chem., Int. Ed.
2010, 49, 3791–3794.
(9) Lin, S. Y.; Dence, C. W. Methods in Lignin Chemistry; Springer-
Verlag: Berlin, 1992.
(10) Mosier, N.; Wyman, C.; Dale, B.; Elander, R.; Lee, Y. Y.;
Holtzapple, M.; Ladisch, M. Bioresour. Technol. 2005, 96, 673–686.
(11) (a) Bozell, J. J.; O’Lenick, C. J.; Warwick, S. J. Agric. Food. Chem.
2011, 59, 9232–9242. (b) Li, J. B.; Gellerstedt, G.; Toven, K. Bioresour.
Technol. 2009, 100, 2556–2561. (c) Samuel, R.; Pu, Y. Q.; Raman, B.;
Ragauskas, A. J. Appl. Biochem. Biotechnol. 2010, 162, 62–74.
(6) Parsell, T. H.; Owen, B. C.; Klein, I.; Jarrell, T. M.; Marcum,
C. L.; Haupert, L. J.; Amundson, L. M.; Kenttamaa, H. I.; Ribeiro, F.;
Miller, J. T.; Abu-Omar, M. M. Chem. Sci. 2013, 4, 806–813.
r
10.1021/ol401065r
XXXX American Chemical Society

Figure 1. Primary monomeric subunits of lignin.
dissolved lignin and fragments bearing residual β-O-4
linkages increases dramatically. Aromatic units in native
lignin contain 7À13% phenolic ÀOH depending on the
source,12 but as lignin transitions from the growing plant
to a biorefinery feedstock, its dissolution and removal
from the matrix can increase this amount to over 70%.13
Even residual lignin in woody biomass after fractionation
exhibits as much as 25% free phenols.14 Accordingly,
catalyst development focusing on selective transformation
of substituted phenols would more accurately model the
structure of lignin as an isolated source of renewable carbon.
Co-Schiff base complexes catalyze the aerobic oxidation
of phenols under mild conditions via formation of an
intermediate Co-superoxo complex that abstracts a phe-
nolic hydrogen and affords a reactive phenoxy radical
(Figure 2).15 Using Co(salen) (4) and related complexes,
we demonstrated one of the first examples of selective
catalytic oxidation of para-substituted phenolic lignin
models and its utility in the synthesis of benzoquinones.16
We now report the synthesis of a new series of unsym-
metrical Co-Schiff base oxidation catalysts17 that exploits
phenolic functionality in both G and S lignin models to
afford methoxybenzoquinone (MBQ, 5) and 2,6-di-
methoxybenzoquinone (DMBQ, 6) as primary oxidation
products. Phenolic β-O-4 model dimers are converted to
quinones not via β-aryl ether cleavage, but through a
unique cleavage of the CÀC bond between the aromatic
ring and the R-carbon of the lignin model side chain to
afford the quinone products and fragment 7. This is the
first example of a catalytic system able to convert both S
and G lignin model phenols in high yield, a process
necessary for effective lignin conversion.
Figure 2. Co-Schiff base catalyzed cleavage of lignin models.
form an intermediate phenoxy radical.18 However, the
yield of MBQ was increased to 51% upon addition of a
stoichiometric amount of a sterically hindered aliphatic
nitrogen base. The additional base likely deprotonates the
phenol substrate to give a more easily oxidized phenoxide
anion.16b Based on these results, we designed a new family
of Co-Schiff base catalysts 8 and 9 incorporating a steri-
cally hindered base within their structure (Table 1).
Catalyst 8a symmetrically substituted with t-Bu groups
provides DMBQ with yields and rates comparable to those
for oxidations catalyzed by 4 (entries 2À4), but catalyst 8b
bearing symmetric bulky piperazine groups on the Co
ligand afforded lower selectivity and much higher levels
of syringaldehyde 12 (entry 5). In contrast, several unsym-
metric catalysts 9 bearing a single functionalized substitu-
ent provided 6 in good yield. Using catalyst 9a under
conditions optimized for 4 gave a 67% yield of 6. However,
9a also exhibits significantly higher reactivity than 4, as
reducing the catalyst level to 5% and the reaction time to
5 h gave nearly the same yield of 6 (entries 6 and 7). Further
reduction in catalyst level and reaction time led to poorer
DMBQ yields (entry 8). Interestingly, oxidation with
catalysts 9 does not require the addition of an external
axial ligand for Co, and indeed, the addition of pyridine to
the oxidation led to low yields of 6 and recovery of much of
the starting material (entry 9). N-Methyl catalyst 9b ex-
hibited a lower yield and reactivity (entry 10). Conversely,
N-benzyl catalyst 9c showed the highest reactivity, afford-
ing 6 in 74% yield after only 1 h (entry 11), and comparable
yields of 6 after 16 h using aslittle as 2% catalyst (entry 12).
Because oxidations with complex 4 generally require the
presence of anaxial pyridine ligand, 9f was synthesized and
oxidized 10 to 6 in moderate yields but at a lower rate
(entry 14), possibly reflecting an effect similar to that of
adding pyridine to the oxidation catalyzed by 9a.
Previous studies from our laboratory showed that
10 mol % Co(salen) in the presence of an axial ligand
(pyridine or imidazole) and molecular O2 (50À60 psi)
converted several S models to 6 in high yield.16a Although
yields for S oxidation ranged above 70%, oxidation of the
less reactive G models afforded e20% 5 under the same
conditions because of the reduced ability of G models to
(12) Lai, Y. Z.; Guo, X. P. Wood Sci. Technol. 1991, 25, 467–472.
(13) Robert, D. R.; Bardet, M.; Gellerstedt, G.; Lindfors, E. L.
J. Wood Chem. Technol. 1984, 4, 239–263.
(14) Gellerstedt, G.; Lindfors, E. Svensk Papperstidn. 1984, 87,
R115–R118.
(15) Zombeck, A.; Drago, R. S.; Corden, B. B.; Gaul, J. H. J. Am.
Chem. Soc. 1981, 103, 7580–7585.
Upon identifying 9c as a lead catalyst, several mono-
meric and dimeric lignin models representing G and S
(16) (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.
(18) Mulder, P.; Saastad, O. W.; Griller, D. J. Am. Chem. Soc. 1988,
110, 4090–4092.
(17) Kleij, A. W. Eur. J. Inorg. Chem. 2009, 193–205.
B
Org. Lett., Vol. XX, No. XX, XXXX

Table 1. Catalyst Screening for the Oxidation of Syringyl
Alcohol 10 into DMBQ 6 and Syringaldehyde 12
Table 2. Co-Catalyzed Oxidation of Lignin Models
catalyst
(mol %)
additive
(mol %)c
t
6
(%)a
12
10
entry
(h)
(%)
(%)
1b
2
À
À
16
16
16
16
16
16
5
0
7
93
4(10)
pyr (100)
82
50
79
39
67
61
34
11
54
74
69
65
75
46
<5
5
0
3
8a (10)
8a (10)
8b (5)
9a (10)
9a (5)
9a (2)
9a (10)
9b (5)
9c (5)
9c (2)
9d (5)
9e (5)
9f (5)
DBZP (15)
28
4
pyr (100)
10
32
20
25
31
29
0
traces
6
5
À
6
À
traces
traces
35
7
À
8b
9b
10
11
12
13
14
15
À
3
pyr (100)
16
16
1
60
À
À
À
À
À
À
40
19
30
22
19
0
0
16
2
0
traces
0
16
16
44
a Yields are given for isolated material unless otherwise specified.
b Yields determined by 1H NMR analysis. c Mol % based on substrate;
pyr = pyridine, DBZP = dibenzylpiperazine.
a Yields are given for isolated material. b 10 mol % of catalyst used.
c Yields relative to 2 equiv of product formed. d 14% 2,5-DMBQ
isolated.19 e 30% aldehydes.20
subunits and bearing a phenolic OH were treated under the
optimized conditions (Table 2). Oxidation of substituted
benzyl alcohol 14 proceeded smoothly, and DMBQ was
obtained in high yields using 5 mol % catalyst, although
increased substitution at the benzylic position slowed the
reaction (entry 1). Importantly, the increased effectiveness
of 9c as a catalyst was demonstrated by the oxidation of
vanillyl alcohol 15 to MBQ 5 in high yield and selectivity
(entry 2) and the oxidation of the more substituted 16 to
MBQ in moderate yield (entry 3).
G models have been subject to extensive investigation
due to either their slow oxidation to 5 by Co-Schiff base
catalysts16b,21 or their dimerization via a radical shift to the
ortho22 or para23 position of the intermediate phenoxy
radical. The effect of the pendant base on the oxidation
is under investigation, but we suspect that its proximity to
the catalyst center may promote rapid oxidation of an
intermediate phenolate anion to the corresponding phe-
noxy radical.
(19) Prolonged reaction in MeOH converts MBQ 5 to 31:
(20) Inseparable mixture of aldehydes 32a and 32b:
(21) (a) Pratt, D. V.; Ruan, F. Q.; Hopkins, P. B. J. Org. Chem. 1987,
52, 5053–5055. (b) Saa, J. M.; Llobera, A.; Garciaraso, A.; Costa, A.;
Deya, P. M. J. Org. Chem. 1988, 53, 4263–4273.
(22) Kadla, J. F.; Kubo, S. Macromolecules 2003, 36, 7803–7811.
(23) Singh, S. N.; Bhattacharjee, M.; Mahanti, M. K. Bull. Chem.
Soc. Jpn. 1983, 56, 1855–1856.
Org. Lett., Vol. XX, No. XX, XXXX
C

Several phenolic lignin dimers modeling both S and G
subunits linked through a β-aryl ether bond are also
converted to the corresponding benzoquinones. Com-
pound 17 was converted to 2 equiv of 6 in excellent yield
(entry 4). Gratifyingly, compounds 18, 19, and 20 incor-
porating both S and G subunits and representing residual
potential β-O-4 linkages remaining after lignin isolation
were also oxidized to high yields of 6. More moderate
yields of 5 were observed for these dimers as a result of
possible rapid dimerization of the intermediate phenoxy
radical to biphenols or diphenoquinones.20,21 Models 18
and 19 gave 5 and 6 in similar yields although longer
reaction times and higher catalyst loadings were required
to accommodate the increased level of benzylic substitution.
Only traces of ketone 12 or 13 aredetectedinthesereactions.
We suggest that the mechanism in Figure 3 may be
operative. Based on mechanistic studies for simpler com-
plexes, we propose that starting Co-Schiff base/ligand
complex 21 is converted to superoxo complex 22 upon
reaction with O2.24 Reaction of 22with a lignin model (e.g.,
19) first generates phenoxy radical 23,25 which is trans-
formed to intermediate 24 through addition of a second
equivalent of 22. Compound 24 undergoes cleavage of the
R-aryl carbon to form 6 and concomitant formation of
enol 25, Co intermediate 27, and phenoxy radical 288b
which can dimerize to 29a20 or 29b.21 Intermediate 28 can
also be trapped as 3026 through radical coupling with 22 in
the para position of the phenoxy radical, followed by elim-
ination of water and regeneration of the catalyst to give 5.
Preliminary experiments examining the direct oxidative
cleavage of lignin have been carried out. Lignin isolated
by organosolv fractionation of tulip poplar under condit-
ions11a that retain (by NMR) a significant amount of
nonphenolic β-O-4 units, and thus a lower amount of free
phenolic OH, was treated with catalyst 9c and O2 for 72 h
at rt in a MeOH/DMSO mixture.
Figure 3. Proposed mechanism for the formation of 5and 6from 19.
Scheme 1. Direct Oxidative Cleavage of Poplar Lignin
The product mixture revealed the presence of DMBQ 6
as well as benzaldehydes 12 and 13 (Scheme 1). The low
absolute yields of monomeric aromatic compounds from
this lignin sample likely result from the relatively low
amount of free ÀOH groups. Given the strong impact of
lignin isolation conditions on the concentration of pheno-
lic OH groups,13 we expect that lignins isolated under more
severe conditions will lead to higher levels of aromatics.
In conclusion, we report the oxidative cleavage of
monomeric and dimeric lignin models catalyzed by a new
family of Co-Schiff base catalysts bearing bulky cyclic
nitrogen bases as substituents. Catalyst 9c efficiently con-
verts S and G subunit models in good yield and also gives
access to benzoquinones from lignin model dimers. The
promising results from this work suggest that ongoing
catalyst design will lead to selective oxidation systems able
to navigate multiple substructural units present in lignin.
The impact of lignin structure and isolation methods on
oxidation yields and the design of new catalysts are under
active investigation.
Acknowledgment. This work was supported as part of
the Center for Direct Catalytic Conversion of Biomass to
Biofuels (C3Bio), an Energy Frontier Research Center
funded by the U.S. Department of Energy, Office of
Science, Office of Basic Energy Sciences under Award
Number DESC0000997.
Supporting Information Available. Experimental pro-
cedures and spectroscopy analysis for all new com-
pounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
(24) Cozzi, P. G. Chem. Soc. Rev. 2004, 33, 410–421.
(25) Hwang, S.; Lee, C. H.; Ahn, I. S. J. Ind. Eng. Chem. 2008, 14,
487–492.
(26) Gaudemer, A.; Nguyenvanduong, K.; Shahkarami, N.; Achi,
S. S.; Frostinrio, M.; Pujol, D. Tetrahedron 1985, 41, 4095–4106.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX