Catalytic oxidation of para-substituted phenols with cobalt-Schiff base complexes/O2 - Selective conversion of syringyl and guaiacyl lignin models to benzoquinones

Article abstract of DOI:10.1016/j.tetlet.2012.02.093

Models of guaiacyl (G) and syringyl (S) subunits in lignin have been catalytically oxidized to their corresponding p-quinones in the presence of molecular oxygen. The oxidation of syringyl-like phenols readily occurred with 5-coordinate cobalt catalysts on which one of the ligands is a monodentate pyridine or imidazole base that coordinates axially to the metal. Formation of p-quinones with this system depends on the coordination of the axial base to the metal as influenced by its pK_a and its size. The yield of p-quinones from guaiacyl models was markedly improved by the addition of a sterically hindered aliphatic nitrogen base that does not coordinate to the catalyst. A mechanism involving deprotonation of the phenol substrate by the bulky base is proposed.

Full text of DOI:10.1016/j.tetlet.2012.02.093

Tetrahedron Letters 53 (2012) 2380–2383
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Catalytic oxidation of para-substituted phenols with cobalt–Schiff base
complexes/O2—selective conversion of syringyl and guaiacyl lignin
models to benzoquinones
⇑
Diana Cedeno, Joseph J. Bozell
Center for Renewable Carbon, Center for the Catalytic Conversion of Biomass (C3Bio), University of Tennessee, Knoxville, TN 37917, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Models of guaiacyl (G) and syringyl (S) subunits in lignin have been catalytically oxidized to their corre-
sponding p-quinones in the presence of molecular oxygen. The oxidation of syringyl-like phenols readily
occurred with 5-coordinate cobalt catalysts on which one of the ligands is a monodentate pyridine or
imidazole base that coordinates axially to the metal. Formation of p-quinones with this system depends
on the coordination of the axial base to the metal as influenced by its pKa and its size. The yield of
p-quinones from guaiacyl models was markedly improved by the addition of a sterically hindered
aliphatic nitrogen base that does not coordinate to the catalyst. A mechanism involving deprotonation
of the phenol substrate by the bulky base is proposed.
Received 18 January 2012
Revised 17 February 2012
Accepted 21 February 2012
Available online 27 February 2012
Keywords:
Phenol oxidation
Cobalt–Schiff Base
Lignin
Ó 2012 Elsevier Ltd. All rights reserved.
Biomass
Introduction
electron-rich G units in lignin, such as 3-methoxy-4-hydroxybenzyl
alcohol (4 or vanillyl alcohol), proceeded in much lower yield.
Cobalt–Schiff base complexes have been extensively used to
catalyze oxygen activation in the oxidation of phenols.1–4 However,
the use of these complexes for the oxidation of p-substituted phe-
nols as models of catalytic conversion of lignin within the biorefin-
ery has not been widely studied. Lignin, which comprises nearly
25% of terrestrial biomass, remains one of the most underused
renewable carbon sources in the biosphere because of its structural
heterogeneity and lack of processes able to convert it into a single
material in high yield. Lignin’s heterogeneity results from free rad-
ical polymerization of monomeric, p-substituted guaiacyl (G),
p-hydroxyphenyl (H), and syringyl (S) units (Fig 1). Biosynthesis
of lignin from these monolignols results in a wide variety of sub-
structural units connected through ether and carbon–carbon
linkages.5
To expand the utility of this process to a wider range of lignin’s
substructural units we have examined the reactivity of the
Figure 1. The primary monomeric units in lignin.
As part of our program to develop a synthetic methodology for
the conversion of renewable carbon sources into biobased chemicals,
we reported that O2 (50–60 psi) in the presence of 5–10% Co(salen)
and pyridine or Co(N-Me salpr) (1 and 2, Fig. 2) in MeOH at room
temperature converted syringyl alcohol (3, 3,5-dimethoxy-4-
hydroxybenzyl alcohol) and several other p-substituted phenols that
model lignin’s S units into 2,6-dimethoxybenzoquinone (DMBQ) in
high yield.2 However, oxidation of compounds modeling the less
⇑
Corresponding author. Tel.: +1 865 974 5991; fax: +1 865 946 2085.
E-mail address: jbozell@utk.edu (J.J. Bozell).
Figure 2. Complexes and ligands used in this study.
0040-4039/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2012.02.093

D. Cedeno, J. J. Bozell / Tetrahedron Letters 53 (2012) 2380–2383
2381
Scheme 1. Mechanism for p-quinone formation from p-substituted phenols catalyzed by 1.
catalytically active complex in the presence of added aromatic or
Results and discussion
aliphatic nitrogen bases. We now report that the yield of quinone
from the oxidation of several lignin models catalyzed by 1 or 2 is
strongly affected by the addition of a series of structurally diverse
aromatic nitrogen-containing ligands and that oxidation of vanillyl
alcohol is markedly improved in the presence of sterically hindered
aliphatic nitrogen bases.
Oxidation of syringyl unit models
The ability of Co(salen) derivatives to reversibly bind oxygen
and form superoxo complexes in solution is well established.6–9
The oxygen binding ability is strongly enhanced by the addition
Table 1
Co(salen) catalyzed oxidation of syringyl alcohol (3)
a
Grimmett, M. R. Adv. Heterocycl. Chem. 1980, 27, 242.
Isolated yeild; av. for 3 runs.
Brown, R. S., Clewly, R. G. J. Org. Chem. 1987, 52, 1216.
After 24 h
b
c
d
e
Nozaki, Y.; Gurd, F. R. N.; Chen, R. F.; Edsall, J. T. J. Am. Chem. Soc. 1956, 79, 2123.

2382
D. Cedeno, J. J. Bozell / Tetrahedron Letters 53 (2012) 2380–2383
of a monodentate nitrogen-containing base as 1 by itself binds oxy-
gen poorly.10 The Co(salen)/ligand/O2 adduct has been effectively
used for synthesizing benzoquinones from phenolics unsubstituted
in the position para to the hydroxyl group (Scheme 1).11–14
In these reactions, the added ligand coordinates to the Co–Schiff
base complex (I), which subsequently coordinates a molecule of
dioxygen. The resulting Co(salen) superoxo adduct (II) abstracts a
hydrogen from the phenolic substrate to generate a phenoxy radi-
cal which reacts with a second molecule of II and via rearrange-
ment of the resulting intermediate III, forms a p-quinone. We
investigated whether modifying the added ligand would improve
the ability of complex II to abstract a hydrogen atom from the phe-
nol substrate, leading to higher yields of quinone. Table 1 summa-
rizes the effect of a series of imidazole and pyridine bases on the
yield of quinone from the oxidation of 3 catalyzed by 1.
When using a 10:1 excess (ligand:catalyst, L:C) of several imid-
azole ligands, the yield of quinone15 correlates with the pKa of the
imidazole’s conjugate acid, plateauing with 1,2-dimethylimidaz-
ole. These results show that as the basicity and corresponding do-
nor ability of the imidazole ligand increases, the dioxygen affinity
of the Co complex and its ability to abstract a hydrogen atom from
the starting phenol also increases. Correspondingly, we observe
that the yield of DMBQ decreases when using 1-acetyl-2-methyl-
imidazole16 as the axial ligand. The poorer donor ability of this li-
gand destabilizes the formation of the dioxygen complex and
reduces the yield of p-quinone.17 The low DMBQ yield observed
with excess Im and 1-MeIm likely results from preferential forma-
tion of an inactive bis-imidazole complex that cannot bind oxygen
to form the catalytically active Co–superoxo complex.18
When the L:C ratio is reduced to 1:1, the yield of DMBQ increases
significantly using 1-MeIm as a ligand. However, we also observe a
slight reduction in DMBQ yields with 2-MeIm, 1,2-diMeIm, or 2,4-
diMeIm, suggesting that the correlation of yield with ligand donor
ability is tempered by the size of the added axial ligand. X-ray anal-
ysis of several Co–Schiff base complexes reveals that binding to an
axial base distorts the square pyramidal geometry about the metal
by pushing the Co out of plane. This distortion forces the Schiff base
ligand to adopt an ‘umbrella’ shape, with the extent of the distortion
being greater for bulkier ligands.19 For the smaller ligands in this
study (Im, 1-MeIm), this distortion is insufficient to preclude forma-
tion of a bis-imidazole complex at high L:C ratios, but lower L:C ra-
tios allow a shift of the equilibrium between the different Co
complexes in solution away from bis-imidazole adducts. With the
larger ligands (2-MeIm, 1,2-diMeIm or 2,4-diMeIm), the methyl
groups flanking the coordinating nitrogen induce a greater distor-
tion in the salen ligand upon binding. As a result, binding of a second
bulky ligand to the metal center at either L:C ratio is inhibited. How-
ever, employing lower L:C ratios with the larger ligands also affects
the equilibrium between the Co complexes in solution, in this case,
reducing the relative amount of Co–superoxo complex present, and
thus, slightly reducing the observed DMBQ yield.
Oxidation of guaiacyl unit models
Although S subunit models were easily converted into their cor-
responding p-quinones with 1 or 2, the same approach with vanillyl
alcohol, a model of the G subunit in lignin gave the corresponding 2-
methoxybenzoquinone (MMBQ) in only 21% yield. G models are
more difficult to oxidize than S models (Fig 1) because formation
of the key phenoxy radical (Scheme 1, V) is slower.22,23 Attempts
to improve these yields have had little success.24 However, we find
that the yield of MMBQ from vanillyl alcohol is significantly im-
proved by adding a sterically hindered base such as N,N-diisopropyl-
ethylamine (DIPEA), diisopropylamine (DIPA), or triethylamine
(TEA) to oxidations catalyzed by 1 or 2 (Table 2).
These bases, each with a pKa ꢀ10–11 lead to yields of MMBQ of
52–55% in the presence of 10% of 2. Reactions with 10% Co(salen)
yielded similar results in the presence of DIPEA and DIPA. We ob-
serve a limit to the basicity of the hindered amine as DBU and DABCO
(pKa = 13.28 and 8.19, respectively) either led to decomposition of
the starting phenol, or did not improve the MMBQ yield. These re-
sults are consistent with previous findings using Na2CO3 as the
base.4 Importantly, the addition of a sterically hindered base does
not reduce the yield of DMBQ from S models. Oxidation of 3 using
catalytic amounts of Co(salen) and equimolar amounts of diisopro-
pylethyl amine (DIPEA) give a 67% yield of DMBQ while oxidation
of syringaldehyde (3,5-dimethoxy-4-hydroxybenzaldehyde) in the
presence of Co(salen) and triethylamine (TEA) gives a 73% yield of
DMBQ.
In order to assess the involvement of the sterically hindered
base in the presumed mechanism (Scheme 1), we studied the
interaction between DIPEA and Co(salen) using NMR and UV–vis
spectroscopy (Supplementary data). The 1H NMR spectrum of para-
magnetic Co(salen) in CD2Cl2 exhibits chemical shift values similar
to those reported for the same complex in other non-coordinating
25,26
solvents such as CDCl3
and DMSO-d6.27 However, upon addi-
tion of pyridine, the 1H signals of Co(salen) undergo a dramatic
shift, suggesting strong interaction or coordination with the added
ligand. In contrast, addition of DIPEA does not affect the position of
the Co(salen) peaks. UV–vis spectra also reflect significant changes
only when pyridine is progressively titrated into a sample of a fixed
concentration of Co(salen) in CH2Cl2. Alternatively, titration with
DIPEA leads to no changes in the spectrum.
Table 2
Oxidation of 4 in the presence of sterically hindered bases. Reaction conditions:
10 mol % of hindered base and Co(N-Me salpr)
yield (%)b
a
Substrate
Hindered base
Conjugate acid pKa
None
DABCO
–
8.19
21
21
55
51
52
0c
DIPEA
DIPA
TEA
10.98
10.76
10.60
13.28
DBU
In contrast to substituted imidazoles, we find a poorer correla-
tion of oxidation yields and pKa when substituted pyridines are
added as ligands (Table 1). With the exception of 4-dimethylami-
nopyridine (DMAP), which undergoes competitive formation of
an unreactive binuclear dioxygen complex in solution,20 addition
of 4-methylpyridine (4-MePy), 4-cyanopyridine (4-CNPy), or pyri-
dine (Pyr), all lead to >60% yield of DMBQ regardless of the ligand’s
pKa. As with reactions in the imidazole series, poorer donors (e.g.,
4-CNpy) lead to lower DMBQ yields. Since the pyridine series con-
tains only p-substituted materials, effects from steric interaction of
each ligand with the Co center would be approximately the same,
and thus, low L:C ratios did not increase the yield of DMBQ. The
lower donor ability of substituted pyridines when compared to
imidazoles21 apparently minimizes catalyst deactivation via the
formation of bis-pyridine adducts.
a
Calculated using Advanced Chemistry Development (ACD/Labs) Software
V11.02. pKa of the most basic site at temperature = 25 oC.
b
Isolated yield reported.
A complex polymeric mixture was isolated with complete consumption of
c
starting material.
Scheme 2. Proposed mechanism for the oxidation of guaiacyl models in the
presence of a hindered base.

D. Cedeno, J. J. Bozell / Tetrahedron Letters 53 (2012) 2380–2383
2383
Furthermore, the success of 2 in these reactions mitigates
against binding of the hindered base, as such binding would likely
preclude addition of O2 by blocking the available coordination site.
These findings suggest that the bulky aliphatic bases used in this
study do not bind to the catalyst and thus their role must involve
other mechanistic pathways (Scheme 2).
Research Center funded by the U.S. Department of Energy, Office of
Science, Office of Basic Energy Sciences under Award Number DE-
SC0000997. We gratefully acknowledge UT professors Dr. Nicole
Labbé and Dr. Douglas G. Hayes for UV–vis instrumentation.
Supplementary data
Our proposed mechanism involves deprotonation of the sub-
strate by the sterically hindered base to form a more readily oxi-
dized phenolate anion.28 The resulting anionic Co intermediate is
protonated and subsequently decomposed to hydroperoxy radical
and III. The phenoxy radical undergoes conversion to MMBQ, form-
aldehyde, and Co(N-Me salpr)-OH as shown in Scheme 1.
Supplementary data (experimental procedures for oxidation of
substrates, UV–vis and 1H NMR spectra for Co complexes) associ-
ated with this article can be found, in the online version, at
doi:10.1016/j.tetlet.2012.02.093.
Addition of hindered bases to the oxidations of other G models
such as isoeugenol, eugenol, vanillin, and vanillin acetal29 gave
only low yields (0–10%) of MMBQ. Currently the reason for this
behavior is not yet clear, but may be related to the release of form-
aldehyde from a benzyl-type substrate rather than the release of
other small molecules, for example, CO from vanillin (Scheme 1).
Ample and fairly recent EPR evidence suggests that phenoxy radi-
cals can also coordinate to cobalt–Schiff base complexes.30,31 Our
studies with 2 suggest that at least initially, coordination of the
substrate is not necessary for the reaction. Further investigation
into this alternative mechanism and computational evaluation of
the intermediate Co complexes is currently underway.
References and notes
1. Caldin, E. F.; Dagnall, S. P.; Mak, M. K. S.; Brooke, D. N. Faraday Discuss. 1982, 74,
215.
2. Bozell, J. J.; Hames, B. R.; Dimmel, D. R. J. Org. Chem. 1995, 60, 2398.
3. Kervinen, K.; Korpi, H.; Mesu, J. G.; Soulimani, F.; Repo, T.; Rieger, B.; Leskela,
M.; Weckhuysen, B. M. Eur. J. Inorg. Chem. 2005, 2591.
4. Zombeck, A.; Drago, R. S.; Corden, B. B.; Gaul, J. H. J. Am. Chem. Soc. 1981, 103,
7580.
5. Bunzel, M.; Ralph, J. J. Agric. Food Chem. 2006, 54, 8352.
6. Niederhoffer, E. C.; Timmons, J. H.; Martell, A. E. Chem. Rev. 1984, 84, 137.
7. Cozzi, P. G. Chem. Soc. Rev. 2004, 33, 410.
8. Jones, R. D.; Summerville, D. A.; Basolo, F. Chem. Rev. 1979, 79, 139.
9. Martell, A. E. Oxygen complexes and oxygen activation by transition metals;
Plenum: New York, 1988.
10. Chen, D.; Martell, A. E.; Sun, Y. Z. Inorg. Chem. 1989, 28, 2647.
11. Vogt, L. H.; Wirth, J. G.; Finkbein, H. L. J. Org. Chem. 1969, 34, 273.
12. van Dort, H. M.; Geursen, H. J. Recl. Trav. Chim. Pays-Bas 1967, 86, 520.
13. Tomaja, D. L.; Vogt, L. H.; Wirth, J. G. J. Org. Chem. 1970, 35, 2029.
14. Sheldon, R. A.; Kochi, J. K. Metal catalyzed oxidations of organic compounds;
Academic Press: New York, 1981.
15. These oxidations primarily lead to the formation of DMBQ from 3 and MMBQ
from 4. The other isolated product is unreacted starting material. During the
oxidation of 3 we also observed incomplete oxidation represented by the
formation of syringaldehyde (detected by 1H NMR) in 620% yield.
16. Harper, J. L.; Smith, R. A. J.; Bedford, J. J.; Leader, J. P. Tetrahedron 1997, 53, 8211.
17. Carter, M. J.; Rillema, D. P.; Basolo, F. J. Am. Chem. Soc. 1974, 96, 392.
18. Wu, L.-B.; Hu, Z.-Q.; Lai, G.-Q. Chin. J. Struct. Chem. 2006, 25, 567.
19. Kennedy, B. J.; Fallon, G. D.; Gatehouse, B.; Murray, K. S. Inorg. Chem. 1984, 23,
580.
Conclusions
The effect of a series of aromatic and sterically hindered ali-
phatic bases on the yield of oxidation of G and S lignin models in
the presence of catalytic amounts of cobalt–Schiff base complexes
is influenced by both electronic and steric effects. S models give a
higher yield of p-quinone with imidazole ligands of higher pKa at
high L:C ratios, reflecting the donor ability of the ligand. However,
at lower L:C ratios, steric effects emerge for imidazole ligands bear-
ing methyl groups adjacent to the ligating nitrogen. Addition of a
bulky aliphatic base increases the yield of p-quinone when vanillyl
alcohol (a G model) is used as the substrate in the presence of 1 or
2, but importantly, does not decrease the yield of quinone forma-
tion from S models. We propose a plausible mechanism in which
the sterically hindered base does not coordinate to the catalyst
but rather deprotonates the substrate in order to ease its oxidation.
Future work will be directed toward the applications of our oxida-
tion conditions to natural lignin and EPR identification of the inter-
mediates involved in the reaction.
20. Chen, D.; Martell, A. E. Inorg. Chem. 1987, 26, 1026.
21. Sundberg, R. J.; Martin, R. B. Chem. Rev. 1974, 74, 471.
22. Mulder, P.; Saastad, O. W.; Griller, D. J. Am. Chem. Soc. 1988, 110, 4090.
23. Howard, J. A.; Ingold, K. U. Can. J. Chem. 1963, 41, 2800.
24. Badamali, S. K.; Luque, R.; Clark, J. H.; Breeden, S. W. Catal. Commun. 2009, 10,
1010.
25. Cheng, C. P.; Hwang, Y. S.; Hwang, R. S. J. Chin. Chem. Soc. 1990, 37, 457.
26. Migita, K.; Iwaizumi, M.; Isobe, T. J. Am. Chem. Soc. 1975, 97, 4228.
27. Srivanavit, C.; Brown, D. G. J. Am. Chem. Soc. 1978, 100, 5777.
28. Rappoport, Z. The chemistry of phenols; Wiley: Hoboken, 2003. Vol. II.
29. Dodo, K.; Aoyama, A.; Noguchi-Yachide, T.; Makishima, M.; Miyachi, H.;
Hashimoto, Y. Bioorg. Med. Chem. 2008, 16, 4272.
30. Arora, H.; Philouze, C.; Jarjayes, O.; Thomas, F. Dalton Trans. 2010, 39, 10088.
31. Bolzacchini, E.; Canevali, C.; Morazzoni, F.; Orlandi, M.; Rindone, B.; Scotti, R. J.
Chem. Society-Dalton Trans. 1997, 4695.
Acknowledgments
This work was supported as part of the Center for Direct Cata-
lytic Conversion of Biomass to Biofuels (C3Bio), an Energy Frontier