Isolation of functionalized phenolic monomers through selective oxidation and CO bond cleavage of the β-O-4 linkages in Lignin

Article abstract of DOI:10.1002/anie.201409408

Functionalized phenolic monomers have been generated and isolated from an organosolv lignin through a two-step depolymerization process. Chemoselective catalytic oxidation of β-O-4 linkages promoted by the DDQ/tBuONO/ O₂ system was achieved in model compounds, including polymeric models and in real lignin. The oxidized β-O-4 linkages were then cleaved on reaction with zinc. Compared to many existing methods, this protocol, which can be achieved in one pot, is highly selective, giving rise to a simple mixture of products that can be readily purified to give pure compounds. The functionality present in these products makes them potentially valuable building blocks.

Full text of DOI:10.1002/anie.201409408

.
Angewandte
Communications
DOI: 10.1002/anie.201409408
Renewable Chemicals
Isolation of Functionalized Phenolic Monomers through Selective
ꢀ
Oxidation and C O Bond Cleavage of the b-O-4 Linkages in Lignin**
Christopher S. Lancefield, O. Stephen Ojo, Fanny Tran, and Nicholas J. Westwood*
Abstract: Functionalized phenolic monomers have been
generated and isolated from an organosolv lignin through
a two-step depolymerization process. Chemoselective catalytic
oxidation of b-O-4 linkages promoted by the DDQ/tBuONO/
O2 system was achieved in model compounds, including
polymeric models and in real lignin. The oxidized b-O-4
linkages were then cleaved on reaction with zinc. Compared to
many existing methods, this protocol, which can be achieved in
one pot, is highly selective, giving rise to a simple mixture of
products that can be readily purified to give pure compounds.
The functionality present in these products makes them
potentially valuable building blocks.
T
echnologies to produce chemicals and fuels from renewable
bioresources are the focus of intense studies.[1] This is due to
the falling reserves and rising costs of fossil fuels. In this
context, lignocellulosic biomass, which contains cellulose,
hemicellulose, and lignin, is considered an important
resource.[2] While technologies for the production of ethanol
derived from the cellulosic component are already in
commercial operation,[3] processes for the valorization of
lignin are limited.[4] This constitutes a significant drawback for
a biorefinery where the valorization of all the biomass is
necessary for economic viability in the absence of subsidies.[4]
In addition, despite lignin being a major by-product of the
pulp and paper industry,[5] only 1–2% is used to deliver
commercial products with most being burnt as a low value
fuel. Lignin valorization has therefore become an important
challenge that has yet to be solved. The major hurdle to the
production of biorenewable chemicals from lignin is the
selective depolymerization of this recalcitrant material to
useable monomers. It is for this reason that new chemical
procedures are needed.
Figure 1. A structural representation of a typical hardwood lignin
showing the most common lignin linkages. The b-O-4 linkage is the
most abundant and is known to be present in both threo and erythro
forms in lignin. The relative stereochemistry of the b–b and b–5
linkages is shown.
chemical treatment has been suggested as a method of
achieving the selective depolymerization of lignin.[7] Here we
report a novel and practical method that achieves puts this
strategy into practice.
Our initial studies focused on the use of 2,3-dichloro-5,6-
dicyano-1,4-benzoquinone (DDQ) as a benzylic oxidant for
lignin, follow on from our recently reported use of this
reagent to modify the b–b linkage in kraft lignin.[8] DDQ
exhibits excellent selectivity in oxidizing benzylic and allylic
alcohols and has been used catalytically in the presence of
molecular oxygen and a suitable co-oxidant.[9]
In almost all lignins the b-O-4 unit is the most abundant
structural unit, with smaller amounts of b–b, b–5, and other
minor linkages (Figure 1).[6] Recently, the selective oxidation
of the b-O-4 linkage in lignin combined with a second
Encouragingly, the reaction of the b-O-4 model linkage
(1a) with a stoichiometric amount of DDQ in dichloro-
methane at room temperature led to the formation of ketone
2a in excellent yield, with complete selectivity for the
oxidation of the secondary benzyl alcohol over the primary
alcohol (Scheme 1).
Before extending our studies to catalytic procedures, we
were mindful of the fact that while lignin model compounds,
such as 1a, are soluble in a broad range of solvents that are
[*] C. S. Lancefield, Dr. O. S. Ojo, Dr. F. Tran, Prof. N. J. Westwood
School of Chemistry and Biomedical Sciences Research Complex,
University of St Andrews and EaStCHEM
North Haugh, St Andrews, Fife (UK)
E-mail: njw3@st-andrews.ac.uk
[**] This research was funded by the EPSRC grants EP/J018139/1 and
EP/K00445X/1 and through the International Training Network
SuBiCat. We would also like to thank Professors Paul Kamer, Bob
Tooze, and Derek Stewart for extensive discussions on lignin
processing and the EPSRC National Mass Spectrometry Service
Centre, Swansea, for mass spectra.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201409408.
Scheme 1. Selective benzylic oxidation of b-O-4 model compound 1a.
DDQ=2,3-dichloro-5,6-dicyano-1,4-benzoquinone.
258
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 258 –262

Angewandte
Chemie
Table 2: Selective catalytic benzylic oxidation of a lignin b-O-4 model
polymer.
commonly used in catalytic DDQ chemistry, such as chloro-
benzene,[9a] toluene,[9b] and acetic acid,[9d] lignin is not. For this
reason it was necessary to identify a reaction solvent that was
both suitable for catalytic DDQ reactions and effective at
solubilizing lignin. A literature survey identified one promis-
ing solvent, 2-methoxyethanol.[9a,10] The catalytic aerobic
oxidation of lignin models 1a–e in 2-methoxyethanol was
therefore attempted (Table 1). The models were chosen based
Entry
DDQ
[mol%]
tBuONO
[mol%]
Solvent
Conversion [%][b]
system[a]
total
G
S
1
2
3
4
5
6
5
10
5
10
15
20
10
10
10
10
15
20
A
A
B
B
B
B
50
74
69
91
97
99
64
83
81
100
100
100
35
64
57
82
94
98
Table 1: Selective catalytic benzylic oxidation of lignin b-O-4 models.
[a] A=2-methoxyethanol, B=2-methoxyethanol/1,2-dimethoxyethane
(2:3). [b] Conversions were determined by comparison of the integrals of
the aromatic cross-peaks characteristic of oxidized and unoxidized
structures in the 2D HSQC NMR. Error analysis indicated the standard
deviation for these measurements was less than 0.05. See the Support-
ing Information for more information.
as expected, a more challenging substrate to oxidize. As
a result, the reaction conditions were optimized further.
Increasing the DDQ loading to 10 mol% (Table 2, entry 2)
and addition of 1,2-dimethoxyethane as a nonprotic cosolvent
markedly improved conversions (Table 2, entries 3 and 4). In
order to achieve almost complete oxidation, the loading of
DDQ and tBuONO had to be increased to 20 mol% (Table 2,
entry 6). The beneficial effect of the cosolvent in this reaction
may in part be explained by an increase in the effective
lifetime of DDQ under these conditions. It is interesting to
note that in this polymer, G units were oxidized more readily
than S units, consistent with the reaction proceeding through
a benzylic cation.
Having developed a working catalytic oxidation method,
our investigation turned to the selective degradation of the
oxidized b-O-4 structure. Zinc is a versatile reductant, in
particular, it is an excellent reagent for the conversion of
acyloin derivatives to ketones.[15] If an analogous reaction
could be carried out on oxidized b-O-4 linkages in lignin,
depolymerization could be achieved to produce phenolic
monomers (Scheme 2).
on their resemblance to linkages found in both hard and
softwood lignins representing structures derived from con-
iferyl and sinapyl alcohol coupling and cross-coupling reac-
tions (1a–d).[11] The DDQ/tBuONO/O2 catalytic system with
2-methoxyethanol as the reaction solvent showed excellent
selectivity and reactivity in the oxidation of the benzylic
alcohol in 1a–d. The expected ketones were isolated in good
to excellent yields in all cases (2a–d). The catalytic system
also showed excellent reactivity toward veratryl alcohol (1e)
giving 2e in a comparable yield to that recently reported using
aerobic 4-AcNH-TEMPO oxidation conditions.[7a]
Unlike 1a–d, lignin is a polymeric substrate that possesses
both a secondary and tertiary[12] structure in solution, which
both affect linkage reactivity. Therefore, a more relevant
polymeric b-O-4 model was investigated. Despite the clear
advantages of using model lignin polymers, they are rarely
used, with only one reported example, to the best of our
knowledge.[13] Challenges inherent in their preparation are
probably the cause of their limited use. However, by
modification of a reported route,[14] we were able to prepare
a b-O-4-containing polymer in just three steps from vanillin
and syringaldehyde (see the Supporting Information for
details). This polymer, which contained equal proportions of
S and G units to mimic a b-O-4-rich hardwood lignin, was
used in the DDQ-catalyzed oxidation with reaction conver-
sion being assessed by semiquantitative 2D-HSQC NMR
experiments (see Figures S1–S3 for analysis and Scheme S1
for synthesis of authentic compounds to aid NMR assign-
ment). The polymer was initially treated under identical
conditions to those used for the monomer studies (Table 2,
entry 1). While the catalytic oxidation still proceeded, the
conversion was lower than observed with substrates 1a–
e (Table 1). This observation suggested that the polymer was,
When the oxidized model compounds 2a–d were treated
with an excess amount of zinc in the presence of NH4Cl, the
Scheme 2. Strategy for the production of phenolic monomers from
lignin.
Angew. Chem. Int. Ed. 2015, 54, 258 –262
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
259

.
Angewandte
Communications
ꢀ
Table 3: C O bond cleavage of oxidized b-O-4 model compounds 2a–d
and 4.
Substrate
Product
Yield[a] [%]
88 (92)[b]
84
81
90
27
23
[a] Yields of isolated products. [b] Yield of the isolated product from
a one-pot reaction.[16]
corresponding ketones (5 or 6) were isolated in good yields
(Table 3). This reaction also required no prior purification of
the substrate 2 to achieve high yields of 5 or 6. For example,
when run as a one-pot oxidation/cleavage protocol, isolated
compound 5 was obtained in 92% yield from 1a. When this
procedure was applied to the oxidized b-O-4 polymer model
4, the two phenolic monomers 7 and 8 were isolated in
approximately equal, reasonable yields, indicating high levels
ꢀ
of C O bond cleavage.
Next, we turned our focus to the use of real lignin as
a substrate. A dioxasolv process was used to extract lignin
from Birch sawdust (Betula pendula). This involved a mild-
acid-catalyzed organosolv process using 0.2n HCl in 90%
dioxane. The isolated lignin was rich in syringyl units and
contained a high proportion of b-O-4 linkages, smaller
amounts of the b–b linkage, and only just detectable amounts
of the b–5 linkage (Figure 2A). Oxidation under the opti-
mized catalytic DDQ conditions described in Table 2 pro-
ceeded smoothly to give an oxidized lignin in which the
appearance of the desired alpha ketone b-O-4 structure could
be readily identified in the 2D HSQC NMR spectrum (Fig-
ure 2B). No special drying of our lignin or solvents was
required prior to this transformation.
Figure 2. Partial 2D HSQC NMR spectra of dioxasolv Birch lignin in
[D6]DMSO before (A) and after (B) treatment with our catalytic
oxidation system (10 wt% DDQ, 4.4 wt% tBuONO). The relative
volume integrals of characteristic peaks are given. Cross-peaks for the
b-5 (B) and oxidized b-O-4 (A’) structures in native birch lignin have
been magnified to make them visible. Standard deviations for the
integral values; native lignin=0.06, n=39, oxidized lignin=0.10,
n=15 (see the Supporting Information for more details). For issues
relating to the processing of the b–b linkage, see Ref. [8] and for
structural assignment of A’ and A’’, see Figure S5.
Application of the cleavage method using zinc to the
oxidized lignin followed by extraction and purification gave
isolated compound 8 in 5 wt% yield, together with minor
amounts of 7 and 9[16] (Scheme 3). The catalytic oxidation and
260
www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 258 –262

Angewandte
Chemie
produced as a mixture with other phenolic monomers through
bimetallic hydrogenolysis of organosolv lignin in water.[17a]
Finally, 8 was converted to the monolignol sinapyl alcohol
16, which is known to possess anti-inflammatory and anti-
nociceptive activity,[24] and which is a potential substrate for
a wide range of chemical transformations that could add
complexity and hence value.[25]
In summary, we report the depolymerization of birch
lignin leading to the isolation of a pure phenolic monomer 8
as the major product in a 5 wt% yield. To the best of our
knowledge, this is the first time that compounds such as 8 (and
7) have been isolated as the major products in the depolyme-
rization of lignin. Our one-pot method involves an initial
aerobic catalytic oxidation of the b-O-4 linkages in lignin
followed by zinc-mediated cleavage of the C-OAryl bond.
The method reported here was developed through the use of
simple model compounds and more complex b-O-4 model
polymers. This strategy has allowed us to observe, for the first
time, a clear difference in reactivity between simple and
polymeric models, which is of potential importance in the
development of future methods for lignin valorization. We
also demonstrate that 8, a compound not readily obtainable
from lignin by other methods, is a useful starting point for
subsequent synthetic sequences aimed at delivering valuable
products from this recalcitrant, but highly abundant renew-
able biopolymer.
Scheme 3. Depolymerization of lignin to phenolic monomers.
DME=1,2-dimethoxyethane.
depolymerization of lignin could also be conducted in one pot
with no impact on the yield of 8. A key advantage of our lignin
depolymerization method is the retention of functional
groups in the major phenolic product 8, as these groups are
frequently lost in other lignin depolymerization process-
es.[13,17] Future studies will focus on integrating methods to
recover and recycle both the organic solvents used and the
solubilized zinc salts in the aqueous waste stream. The
environmental and cost implications of using a halogenated
organic catalyst, such as DDQ, will also be assessed.
Optimizing these and other factors may enable this promising
initial reaction sequence to be developed into a large-scale
process.
To assess whether 8 is of potential relevance to the
valorization of lignin, several chemical transformations start-
ing from 8 were assessed (Scheme 4).[18] Products from these
reactions include potentially polymerizable monomer 10,[19]
from which propiophenone 9 and syringylpropane 11[20] could
be accessed, dihydrosinapyl alcohol 12 and b-amino acid
precursor 14a.[21–23] A recent report has shown that 12 can be
Received: September 23, 2014
Published online: November 5, 2014
Keywords: biomass · depolymerization · lignin · oxidation ·
.
renewable chemicals
[1] R. A. Sheldon, Green Chem. 2014, 16, 950 – 963.
[2] a) Y. H. P. Zhang, J. Ind. Microbiol. Biotechnol. 2008, 35, 367 –
375; b) M. FitzPatrick, P. Champagne, M. F. Cunningham, R. A.
Whitney, Bioresour. Technol. 2010, 101, 8915 – 8922.
[3] J. Larsen, M. Ø. Haven, L. Thirup, Biomass Bioenergy 2012, 46,
36 – 45.
[4] J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius, B. M. Weck-
huysen, Chem. Rev. 2010, 110, 3552 – 3599.
[5] F. T. ꢀhman, H. Tomani, P. Axegard, (LignoBoost AB), US
8486224 B2, 2006.
[6] J. Ralph, L. L. Landucci in Lignin and Lignans: Advances in
Chemistry (Eds.: C. Heitner, D. Dimmel, J. Schmidt), CRC, Boca
Raton, 2010, pp. 137 – 243.
[7] a) A. Rahimi, A. Azarpira, H. Kim, J. Ralph, S. S. Stahl, J. Am.
Chem. Soc. 2013, 135, 6415 – 6418; b) J. D. Nguyen, B. S.
Matsuura, C. R. J. Stephenson, J. Am. Chem. Soc. 2014, 136,
1218 – 1221.
Scheme 4. Synthetic conversion of phenolic monomer 8 to various
chemical compounds. Reaction conditions: a) PPh3, DDQ, NBu4I,
DBU, CH2Cl2, RT, 4 h, 78%; b) Zn dust, AcOH, RT, 12 h, 63% for 9;
c) NaBH3CN, BF3·OEt2, RT, 12 h, 89% for 11, 82% for 12; d) TBS-Cl,
DMAP, imidazole, DMF, RT, 1 h, 94%; e) (R)-tert-butyl sulfinamide,
Ti(OEt)4, PhMe, 1008C, 12 h, 60%; f) l-selectride, THF, ꢀ208C!RT,
4 h, 80%; g) TBAF, THF, 3ꢀ M.S., RT, 2 h, 63%; h) H2NNH2·H2O,
1008C, 12 h, 82%; i) I2, Et2O, 08C, 30 min, 51%. DBU=1,8-diaza-
bicyclo[5.4.0]undec-7-ene; DMAP=4-dimethylaminopyridine;
DMF=N,N-dimethylformamide; M.S.=molecular sieves; TBAF=
tetra-n-butylammonium fluoride; TBS=tert-butyldimethylsilyl.
[8] F. Tran, C. S. Lancefield, P. Kamer, T. Lebl, N. Westwood, Green
Chem. 2014 in Press, DOI: 10.1039/c4gc01012d.
[9] a) Z. Shen, L. Sheng, X. Zhang, W. Mo, B. Hu, N. Sun, X. Hu,
Tetrahedron Lett. 2013, 54, 1579 – 1583; b) W. Zhang, H. Ma, L.
Zhou, H. Miao, J. Xu, Chin. J. Catal. 2009, 30, 86 – 88; c) C. C.
Cosner, P. J. Cabrera, K. M. Byrd, A. M. A. Thomas, P. Helquist,
Org. Lett. 2011, 13, 2071 – 2073; d) L. Wang, J. Li, H. Yang, Y. Lv,
S. Gao, J. Org. Chem. 2012, 77, 790 – 794; e) Z. Shen, J. Dai, J.
Xiong, X. He, W. Mo, B. Hu, N. Sun, X. Hu, Adv. Synth. Catal.
2011, 353, 3031 – 3038.
[10] See Figure S4 in the Supporting Information.
Angew. Chem. Int. Ed. 2015, 54, 258 –262
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
261

.
Angewandte
Communications
[11] Model compounds were used as a mixture of both threo and
erythro diastereomers, as both are found in natural lignins.
[12] a) J.-L. Faulon, P. G. Hatcher, Energy Fuels 1994, 8, 402 – 407;
b) L. Petridis, R. Schulz, J. C. Smith, J. Am. Chem. Soc. 2011, 133,
20277 – 20287.
[13] T. H. Parsell, B. C. Owen, I. Klein, T. M. Jarrell, C. L. Marcum,
L. J. Haupert, L. M. Amundson, H. I. Kenttamaa, F. Ribeiro,
J. T. Miller, M. M. Abu-Omar, Chem. Sci. 2013, 4, 806 – 813.
[14] R. Katahira, H. Kamitakahara, T. Takano, F. Nakatsubo, J. Wood
Sci. 2006, 52, 255 – 260.
[19] a) N. Iranpoor, H. Firouzabadi, G. Aghapour, A. R. Vaezzadeh,
Tetrahedron 2002, 58, 8689 – 8693; b) M. Kouyama, K. Ohmura,
US 20130109826, 2013.
[20] a) A. Srikrishna, J. A. Sattigeri, R. Viswajanani, C. V. Yelamag-
gad, Synlett 1995, 93 – 94; b) S. K. Tipparaju, S. P. Muench, E. J.
Mui, S. N. Ruzheinikov, J. Z. Lu, S. L. Hutson, M. J. Kirisits, S. T.
Prigge, C. W. Roberts, F. L. Henriquez, A. P. Kozikowski, D. W.
Rice, R. L. McLeod, J. Med. Chem. 2010, 53, 6287 – 6300.
[21] Several attempts to remove the TBS and sulfoxide groups led to
decomposition.
[15] Z. Yao, D. Ye, H. Liu, K. Chen, H. Jiang, Synth. Commun. 2007,
37, 149 – 156.
[22] G. Borg, D. A. Cogan, J. A. Ellman, Tetrahedron Lett. 1999, 40,
6709 – 6712.
[16] See the Supporting Information for a discussion on the yields of
monomers obtained from the polymer and lignin and Scheme S2
for a proposed route to the formation of 9.
[17] a) J. Zhang, J. Teo, X. Chen, H. Asakura, T. Tanaka, K.
Teramura, N. Yan, ACS Catal. 2014, 4, 1574 – 1583; b) Q. Song,
F. Wang, J. Cai, Y. Wang, J. Zhang, W. Yu, J. Xu, Energy Environ.
Sci. 2013, 6, 994 – 1007.
[23] For a report on the excellent diastereoselectivity (anti) observed
with l-selectride, see; a) J. T. Colyer, N. G. Andersen, J. S.
Tedrow, T. S. Soukup, M. M. Faul, J. Org. Chem. 2006, 71,
6859 – 6862; b) L. R. Reddy, S. G. Das, Y. Liu, M. Prashad, J.
Org. Chem. 2010, 75, 2236 – 2246.
[24] J. Choi, K.-M. Shin, H.-J. Park, H.-J. Jung, H. J. Kim, Y. S. Lee, J.-
H. Rew, K.-T. Lee, Planta Med. 2004, 70, 1027 – 1032.
[25] T. H. West, D. S. B. Daniels, A. M. Z. Slawin, A. D. Smith, J. Am.
Chem. Soc. 2014, 136, 4476 – 4479.
[18] We chose phenolic monomer 8 because it was obtained in
a higher weight percent than other phenolic monomers 7 and 9.
262
www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 258 –262