Nano WO3-Catalyzed One-Pot Process for Mild Oxidative Depolymerization of Lignin and its Model Compounds

Article abstract of DOI:10.1002/cctc.202100670

Despite challenges related to the robust and irregular structure of lignin, the valorization of this aromatic biopolymer has aroused great interest. However, the current methods exhibit problems such as harsh reaction conditions, complicated operation, and difficult recovery of catalyst. Herein we present a one-pot process for the mild oxidative depolymerization of lignin and lignin model compounds catalyzed by nano WO₃, along with tert-butyl hydrogen peroxide (TBHP) as the oxidant and NaOH as the additive, which exhibits advantages of both homogeneous and heterogeneous catalysis. Under the optimized condition, it yielded 80.4 wt % of liquid oil from organosolv lignin with 7.6 wt % of vanillic acid as the main monomer product, accounting for 91.6 wt % monomeric selectivity. Mechanism studies on the model substrate suggest that the reaction proceeds via an oxidation of C_α?OH to C=O followed by C?O bond cleavage to afford phenol and ketone products which may undergo further oxidation to produce aromatic carboxylic acids. We have developed an operationally simple procedure for mild fragmentation of lignin and lignin model compounds with excellent yields, which provides the potential to expand the existing lignin usage from energy source to value-added commodity chemicals.

Full text of DOI:10.1002/cctc.202100670

Full Papers
doi.org/10.1002/cctc.202100670
ChemCatChem
Nano WO3-Catalyzed One-Pot Process for Mild Oxidative
Depolymerization of Lignin and its Model Compounds
Jing Liang,*[a] Meng-Xiao Wang,[a] Yun-Peng Zhao,*[a] Wei-Wei Yan,[a] Xing-Gang Si,[a]
Guo Yu,[a] Jing-Pei Cao,[a] and Xian-Yong Wei[a]
Despite challenges related to the robust and irregular structure
of lignin, the valorization of this aromatic biopolymer has
aroused great interest. However, the current methods exhibit
problems such as harsh reaction conditions, complicated
operation, and difficult recovery of catalyst. Herein we present a
one-pot process for the mild oxidative depolymerization of
lignin and lignin model compounds catalyzed by nano WO3,
along with tert-butyl hydrogen peroxide (TBHP) as the oxidant
and NaOH as the additive, which exhibits advantages of both
homogeneous and heterogeneous catalysis. Under the opti-
mized condition, it yielded 80.4 wt% of liquid oil from organo-
solv lignin with 7.6 wt% of vanillic acid as the main monomer
product, accounting for 91.6 wt% monomeric selectivity. Mech-
anism studies on the model substrate suggest that the reaction
proceeds via an oxidation of CαÀ OH to C=O followed by CÀ O
bond cleavage to afford phenol and ketone products which
may undergo further oxidation to produce aromatic carboxylic
acids. We have developed an operationally simple procedure
for mild fragmentation of lignin and lignin model compounds
with excellent yields, which provides the potential to expand
the existing lignin usage from energy source to value-added
commodity chemicals.
1. Introduction
69.2 to 55.9 kcal/mol, facilitating the cleavage of the CÀ O bond
and further the depolymerization of lignin. Although the
current two-step process, selective oxidation of CαÀ OH followed
by reductive fragmentation[18–20] or subsequent oxidative crack-
ing of β-O-4[21] has been proved efficient, it is still highly desired
to develop one-step or one-pot oxidation process for the
easiness and simplicity of the procedure.
With the depletion of fossil resources, the search for green and
sustainable alternatives has attracted much interest
increasingly.[1–3] Non-edible lignocellulose, produced by forestry
and agricultural activities worldwide and composed of 40–50%
cellulose, 16–33% hemicellulose, and 15–30% lignin, could be
regarded as an attractive candidate.[4,5] In the past decades,
significant advances have been achieved regarding to the
utilization of cellulose and hemicellulose, while the conversion
of lignin encountered difficulties owing to its robust and
complexed three-dimensional amorphous structure. Lignin is
the potential source of aromatic compounds,[6] where the most
common structural pattern is the alkyl-aryl ether bond (β-O-4)
which makes up about 50% of all linkages and thus its cleavage
is the key to depolymerize lignin.[7–10]
A variety of distinct depolymerization methods aiming to
break the ether bond have been extensively exploited, includ-
ing electrocatalysis,[11] pyrolysis,[12,13] oxidation,[14] and
hydrogenolysis.[15] Among them, the oxidation method is
considered to be one of the most effective methods due to its
mild conditions and sufficient selectivity for aromatic com-
pounds such as phenols, aldehydes, ketones, and aromatic
acids.[16,17] It has been identified that the oxidation of CαÀ OH to
C=O reduces the bond dissociation energy of CβÀ OÀ Ar from
Compared with homogeneous catalysts, such as Co,[22]
Cu,[23,24] and V[25] salts used in the oxidation of lignin and its
model compounds, eco-friendly heterogeneous catalysts are
promising for their robust and recyclable characteristics. A great
number of heterogeneous catalysts have been successfully
applied in the depolymerization of lignin including Pd/CeO2,[26]
Co-based nanoparticles supported on nitrogen-doped
carbon,[27] CsPMoVOÀ Co,[28] and CuÀ Mn/σ-Al2O3,[29] etc. However,
usually the degradation of lignin proceeds under harsh
conditions, such as high temperature and long reaction time,
necessitating the development of an efficient and mild method
for oxidative depolymerization of lignin over heterogeneous
catalyst.
WO3, a semiconductor transition metal oxide, has been one
of the research foci for solar-energy-driven photoelectrochem-
ical conversion used alone or in combination with IO4À .[30]
However, only little attention has been paid to the application
of WO3 or other tungsten-based catalysts in the depolymeriza-
tion of lignin. Recently, Mai et al.[31] reported that enzymatic
hydrolysis lignin (EHL) in supercritical ethanol over WO3/γ-Al2O3
catalyst could be converted to aliphatic and aromatic com-
[a] Prof. J. Liang, M.-X. Wang, Prof. Y.-P. Zhao, W.-W. Yan, X.-G. Si, G. Yu,
Prof. J.-P. Cao, Prof. X.-Y. Wei
Key Laboratory of Coal Processing and
Efficient Utilization Ministry of Education
China University of Mining & Technology
Xuzhou 221116, Jiangsu (P. R. China)
E-mail: liangjing@cumt.edu.cn
°
pounds with a total yield of 36.34 wt% at 320 C for 8 h, and
the overall selectivity of alkylphenols reached 67.5%. Li et al.[32]
reported a direct catalytic conversion of raw woody biomass
°
into two groups of chemicals over Ni-W2C/AC at 235 C and
zhaoyp@cumt.edu.cn
6 MPa H2 for 4 h. Cellulose and hemicellulose, were converted
to ethylene glycol and other diols with a total yield of up to
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/cctc.202100670
ChemCatChem 2021, 13, 1–11
1
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202100670
ChemCatChem
75.6% (based on the amount of cellulose & hemicellulose),
while the lignin component was converted selectively into
monophenols with a yield of 46.5% (based on lignin). Guo
et al.[33] further proved organosolv lignins fractionated from
softwood and hardwood could be effectively converted to
liquid oil with the overall yields varying from 47.3 to 55.5 wt%,
WM
WL
Conversion ð%Þ ¼ ð1 À
Þ � 100 %
(1)
(2)
WI
Yield ð%Þ ¼
� 100 %
WF
where WL denotes the original mass of lignin model compound, WM
means the mass of unreacted lignin model compound, WI is the
actual yield of a given product, and WF is the theoretical yield of
the given product.
affording 7.2–8.9% monophenols in the oils over W2C/AC with
[34]
°
an initial H2 pressure of 0.69 MPa at 250 C for 2 h. Ji et al.
reported the deconstruction of organosolv lignin over tung-
sten-based bimetallic catalysts (MÀ W/AC, M = Ru, Pt and Pd) in
°
hexane under 0.7 MPa H2 at 260 C for 10 h, resulting in 42.7%
General Procedure for Oxidative Depolymerization of
Organosolv Lignin
to 66.7 wt% of aromatic liquid oil. Note that all the reports on
WO3 or W2C were related to the hydrogenolysis or hydro-
genation of lignin, and so far, no works on WO3-catalyzed
oxidative depolymerization of lignin and its model compounds
have been reported.
Organosolv lignin isolated from pinewood powder (see Section 8.1
in the SI) was subjected to the modified one-pot reaction
conditions. Organosolv lignin (0.25 g), nano WO3 (0.058 g,
0.25 mmol), 70% TBHP (1024 μL, 7.4 mmol) and 24% NaOH (1.6 mL,
9.6 mmol) were put in a 25 mL three-necked flask followed by the
addition of 1.6 mL of 1, 4-dioxane and 90 μL of ethylene glycol,
Herein, we present a one-pot process for mild oxidative
depolymerization of lignin and its model compounds. With tert-
butyl hydrogen peroxide (TBHP) as the oxidant and 24% NaOH
as the additive, commercial nano WO3 can efficiently catalyze
the cracking of lignin model compounds into phenols, ketones
and aromatic acids in good to excellent yields. When applied to
organosolv lignin, the catalytic process yielded 80.4 wt% of
liquid oil at 93.3 wt% conversion (based on lignin) with
7.6 wt% of vanillic acid, accounting for as high as 91.6% of
selectivity in the lignin-derived monomers. Moreover, the
catalyst could be recycled for 5 times without significant loss of
activity. Mechanism studies show that the CαÀ OH group in the
β-O-4 models could be selectively oxidized to the correspond-
ing ketone which then weakens the neighboring CβÀ OÀ Ar
bond, thus facilitating its cleavage to obtain phenols and
ketones. The latter may undergo further oxidization to produce
aromatic carboxylic acids.
°
then stirred for 4 h at 100 C. The mixture was cooled to room
temperature and acidified to pH 2~3 with 6 N HCl to precipitate
unreacted lignin, which was separated by filtration, then dried and
weighed (WB). After standing for a period of time, WO3 precipitated
slowly from the filtrate, then was removed by simple filtration. The
remaining filtrate was neutralized, extracted with ethyl acetate for
three times. The combined organic layer was concentrated under
vacuum to remove the solvent, then dried to afford the liquid oil
and weighed (WD). The oil was rediluted in 2 mL of ethyl acetate,
and 1 μL of which was injected in GC/MS for analysis with 3, 4, 5-
trimethoxybenzaldehyde as external standard to determine the
monomeric yields. Also, the liquid oil was submitted to QPEOTMS
analysis and the aqueous layer to a matrix-assisted laser desorp-
tion/ionization time of flight MS (MALDI-TOF MS) analysis.
The conversion and yields are calculated as follows [Eqs. (3)–(6)]:
�
�
WB
WA
Conversionð%Þ ¼ 1 À
� 100 %
(3)
(4)
(5)
(6)
Experimental Section
P
Wc
Yieldmonomerð%Þ ¼
WA
� 100 %
General Procedure for Oxidative Depolymerization of Lignin
Model Compounds
WD
WA
Yieldliquid oilð%Þ ¼
� 100 %
Lignin model compound (1a–1h) (0.5 mmol), nano WO3 (0.02 g,
0.1 mmol), 70% TBHP (540 μL, 3.9 mmol) and 24% NaOH (0.80 mL,
4.8 mmol) were placed in a 25 mL three-necked flask and stirred for
WC
P
Selectivitymonomerð%Þ ¼
� 100 %
°
a certain time at 80 C. Thin layer chromatography (TLC) was
WC
employed to monitor the progress of the reaction. Upon comple-
tion, the reaction mixture was cooled to room temperature and
acidified to pH 2~3 with ice-cold 6 N HCl to precipitate WO3, which
was filtered, washed three times with water and ethanol, and dried
where WA is the original mass of lignin, WB is the residual mass of
lignin, WC is the mass of a monomer, WD is the mass of liquid oil
and �WC is the total mass of all the monomers.
°
under vacuum at 80 C for 10 h for recycle use. The filtrate was
neutralized, extracted with ethyl acetate for three times, and then
the combined organic phase was washed with saturated brine,
dried over anhydrous MgSO4, and finally concentrated under
reduced pressure. The resulting oil mixture was then rediluted in
2 mL of ethyl acetate, and 1 μL of which was injected in the gas
chromatography-mass spectrometry (GC-MS) for analysis with 3, 4,
5-trimethoxybenzaldehyde as internal standard (see Section 3 in
the SI).
Catalyst Recycle Experiments
After each filtration (see section 6 in SI for details), the catalyst was
washed with water and ethanol for three times respectively, then
°
dried under vacuum at 80 C for 10 h. The recovered catalyst was
reused for the subsequent catalytic oxidation of model compound
1a to study the reusability of the catalyst.
The conversion and yields of model compounds are defined as
[Eqs. (1) and (2)]:
ChemCatChem 2021, 13, 1–11
www.chemcatchem.org
2
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202100670
ChemCatChem
2. Results and Discussion
2.2. Optimization of Model Oxidative Cleavage
2-(2-methoxyphenoxy)-1-(4-methoxyphenyl) ethanol (1a),
2.1. Catalyst Characterization
a
compound containing β-Ο-4 linkage, was selected as the lignin
model compound to optimize the reaction conditions, including
catalysts, oxidants, and alkaline content, etc. As shown in
Table 1, a series of tungsten-based catalysts, such as WO3,
H3O40PW12 ·xH2O, Na2WO4 ·2H2O and H4[Si(W3O10)4]·xH2O, were
The X-ray diffraction (XRD) patterns in Figure 1a shows that
nano WO3 is a monoclinic phase coupled with diffraction peaks
°
°
°
°
at 2θ values of 23.1 , 23.7 , 24.4 and 53.6 corresponding to
the (001), (020), (200) and (022) facets (JCPDS-752072),
respectively. Transmission electron micrograph (TEM) and
scanning electron microscope (SEM) (Figure S1) show that nano
WO3 is a sheet structure, and the particle size is in the range of
50–100 nm. The XPS spectra proves the elemental composition
of the catalyst and the valence state of the elements. Figure 1b
indicates the presence of W and O elements in nano WO3, and
further as shown in Figure 1c, the presence of W4f7/2 and W4f5/2
doublet at 36.0 eV and 38.2 eV in the W4f spectra is indicative
of W6+ in monoclinic WO3. Figure 1d shows the O1s peak of
WÀ OÀ W at 530.8 eV in WO3.
°
selected to investigate the catalytic performance at 80 C with
TBHP as the oxidant and NaOH as the additive. Among them,
WO3 gave as high as 98.6% conversion (entry 1) and compound
1a was efficiently cleaved into a phenol (3a, 17.2 wt%), a
ketone (4a, 15.7 wt%), and an aromatic carboxylic acid (5a,
80.4 wt%), along with undetectable amounts of the oxidized
ketone 2a and the dehydrated intermediate 6a in some cases.
Other tungsten-based catalysts showed much lower activity
than WO3 (entries 9–11). We tried to replace WO3 with nano
WO3 and found that 20 mol% nano WO3 (entry 2) provided the
similar conversion under the same conditions (Figure 2).
Figure 1. (a) XRD patterns of original and recycled nano WO3. (b) XPS survey spectra of original and recycled nano WO3. (c) W4f core level XPS spectra. (d) O1s
core level XPS spectra.
ChemCatChem 2021, 13, 1–11
www.chemcatchem.org
3
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202100670
ChemCatChem
Table 1. The oxidative cleavage of β-O-4 model compound 1a.[a]
Entry
Catalyst
Conversion [%]
Product yields [%][b]
3a
4a
5a
6a
1
WO3
WO3
WO3
WO3
–
WO3
WO3
WO3
98.6
98.0
<1.0
<1.0
41.1
–
17.2
16.9
trace
trace
8.3
15.7
14.5
trace
trace
38.5
80.4
81.8
trace
trace
7.0
2[c]
3[c,d]
4[c,e]
5[f]
6[c,g]
7[c,h]
8[i]
9
10
2.0
<2.0
98.9
59.9
5.1
13.2
10.9
2.3
13.2
47.4
11.3
15.7
80.6
4.8
trace
H3O40PW12 ·xH2O
Na2WO4 ·2H2O
H4[Si(W3O10)4]·xH2O
11
20.3
6.1
2.0
°
[a] Reaction conditions:1a (0.137 g, 0.5 mmol), catalyst (0.2 mmol), 70% TBHP (3.9 mmol), 24% NaOH (4.8 mmol), 80 C and 2 h, [b] Yield was determined by
GC-MS with 3, 4, 5-trimethoxybenzaldehyde as internal standard, [c] Nano WO3 (0.1 mmol), [d] Oxygen instead of TBHP, [e] H2O2 instead of TBHP, [f] Without
catalyst, [g] Without TBHP, [h] Without NaOH, [i] 1, 4-dioxane/water (1:1, v/v) as the solvent.
Figure 2. Total-ion chromatogram of fragmented lignin model compound 1a. aReaction conditions: 1a (0.5 mmol), nano WO3 (0.1 mmol), 70% TBHP
°
(3.9 mmol), 24% NaOH (4.8 mmol), 80 C and 2 h. GC-MS spectrum was recorded with 3, 4, 5 trimethoxybenzaldehyde as internal standard (IS).
Control experiments revealed that in the absence of
catalysts, the conversion was only 41.1% (entry 5), while
without TBHP or NaOH, virtually no products were detected
(entries 6and 7), suggesting that an oxidant or a base was
indispensable for the oxidative cleavage of 1a. As TBHP can
react with NaOH and then transfer an oxygen atom to nano
WO3 to form a metal peroxide species,[35] the real oxidant,
thereby facilitating the oxidative cleavage of model compound
1a. However, H2O2 and O2, the most common green oxidants,
were found inefficient for the reaction (entries 3and 4).
Alkali was another indispensable component for this
reaction. NaOH worked well and we didn’t try any other alkali
because NaOH is cheap and commercially available. Unexpect-
edly, upon the addition of NaOH, nano WO3 gradually dissolved
in it and acted like a homogeneous catalyst, thereby promoting
the reaction. But when the mixture was acidified with HCl, nano
Figure 3. The depolymerization of organosolv lignin to aromatic monomers. aReaction conditions: organosolv lignin (25 mg), nano WO3 (0.25 mmol), 70%
°
TBHP (7.4 mmol), 24% NaOH (9.6 mmol), 1, 4-dioxane (1.6 mL), ethylene glycol (90 μL), 100 C and 4 h.
ChemCatChem 2021, 13, 1–11
www.chemcatchem.org
4
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202100670
ChemCatChem
Figure 4. (a) MALDI-TOF MS spectra of organosolv lignin (b) QPEOTMS spectrum of the liquid oil resulted from lignin oxidative depolymerizationa. aReaction
conditions: organosolv lignin (25 mg), nano WO3 (0.25 mmol), 70% TBHP (7.4 mmol), 24% NaOH (9.6 mmol), 1, 4-dioxane (1.6 mL), ethylene glycol (90 μL),
°
100 C and 4 h.
WO3 slowly precipitated from the mixture, indicating the
possibility to be recycled by filtration. Hence, nano WO3 has the
advantages of both homogeneous and heterogeneous cata-
lysts. Besides, as shown in Tables S2 and S3, not only the
amount of alkali, but also the concentration of alkali had a
significant influence on the reaction. As the concentration of
NaOH ranged from 3% to 18%, the conversion of 1a increased
rapidly from 7% to 89.4% and the yields of guaiacol (3a), p-
methoxy acetophenone (4a) and p-methoxy benzoic acid (5a)
followed the similar upward trend. When the concentration of
NaOH reached 24%, 98.0% conversion was achieved accom-
panied by improved yields of 3a (16.9%), 4a (14.5%) and 5a
(81.8%). Further increase of NaOH concentration to 27%
resulted in a slight increase in conversion (99.2%), but a
significant decrease in the total yields of the products, implying
higher alkaline content may lead to side reactions. Therefore,
24% NaOH was employed for all the following experiments.
Moreover, we carried out reactions with different molar ratios
of NaOH to TBHP and found that the ratio of 1:1.2
(3.9 mmol:4.8 mmol) was optimal for this reaction (Table S3).
The influences of temperature and reaction time were also
2.3. Substrate Scope of Oxidative Cleavage
With the optimized reaction conditions in hand, we extended
the scope of substrates to check the versatility of the developed
protocol. Lignin model compounds with zero to three methoxy
groups commonly found in natural lignin structure were
synthesized and used in the mild oxidative cleavage process. As
expected, the optimized conditions efficiently cleaved the CÀ O
bond, providing the desired phenols, ketones or carboxylic
acids in good to excellent yields in all cases, but whether the
main product was a ketone, or an acid seemed to be related to
the number of methoxy groups present on the model
compounds. As shown in Table 2, For convenience, the phenyl
ring adjacent to benzyl-OH was labelled A and the other ring
was labelled B. In the absence of methoxy groups, 1d required
3 h to complete the conversion with benzoic acid as the main
product (entry 4). As for 1b and 1c, only one methoxy group
present either on ring A or on ring B, the reaction time was
shortened to 2 h and the main products were the correspond-
ing ketones (entries 2and 3), whereas for 1a, 1e and 1h,
bearing two methoxy groups in their structures, aromatic
carboxylic acids were the main products (entries 1, 5and 8). In
the case of 1f and 1g, with three methoxy groups in all, the
main products were the expected aromatic acids, but the
reaction time was extended to 6 h to afford complete
conversion (entries 6and 7). Since oxidation is a process of
losing electrons, the more electron-donating methoxy groups
the substrate contains, the more favorable for the oxidation,
while too many methoxy groups made the reactive center less
accessible for the oxidant owing to steric hindrance, thus
°
investigated and found that 80 C and reaction time of 2 h were
appropriate (Table S5 and Table S6). Since water is a green and
cheap solvent, no further attempt on solvents was performed.
Under the optimized reaction conditions, compound 1a could
be almost completely converted into guaiacol, p-methoxy
acetophenone and p-methoxy benzoic acid.
ChemCatChem 2021, 13, 1–11
www.chemcatchem.org
5
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202100670
ChemCatChem
Figure 5. Partial HSQC spectra of organosolv lignin (a) before and (b) after oxidative depolymerizationa. aReaction conditions: organosolv lignin (25 mg), nano
°
WO3 (0.25 mmol), 70% TBHP (7.4 mmol), 24% NaOH (9.6 mmol), 1, 4-dioxane (1.6 mL), ethylene glycol (90 μL), 100 C and 4 h.
decreasing the rate of reaction. As the production of acids was
a consequence of the continued oxidation of ketones, long
reaction time was beneficial for the production of carboxylic
acids.
Note that for model compounds with methoxy groups on
phenyl ring B, the yields of phenols were much lower (entries 1,
3, 5and 7), and even for compound 1f, with two methoxy
groups on ring B, no corresponding phenol was observed
(entry 6). The presence of methoxy groups imparts phenyl ring
ChemCatChem 2021, 13, 1–11
www.chemcatchem.org
6
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202100670
ChemCatChem
Table 2. Nano WO3-catalyzed oxidative cleavage of different lignin model compounds.[a]
Entry
Substrate
Time [h]
Conversion [%]
Product yields [%][b]
1
2
98.0
2
3
2
2
99.0
96.0
4
3
97.0
5
6
8
6
99.0
98.3
7
8
6
2
99.0
99.0
°
[a] Reaction conditions: 1a–1h (0.5 mmol), nano WO3 (0.1 mmol), 70% TBHP (3.9 mmol), 24% NaOH (4.8 mmol), 80 C, [b] Yield was determined by GC-MS
with 3, 4, 5-trimethoxybenzaldehyde as internal standard.
B higher electron density which makes the resulting phenols
susceptible to excessive oxidation or polymerization[36] as
identified in Wang’s work.[21] Given the fact that β-O-4 model
compounds containing γ-OH are difficult to be cleaved,[37–39] the
catalytic fragmentation of 1e faces greater challenges. Fortu-
nately, our proposed catalytic system could almost completely
break compound 1e into the desired products within 8 h under
the mild condition, suggesting that the catalytic system has the
potential to depolymerize natural lignin.
upon the addition of NaOH, it readily reacted with TBHP and
abstracted the hydrogen atom to produce a tertiary butyl
peroxy anion, which then transferred the oxygen atom to the
tungsten atom in the center of WO3 to generate a three-
membered ring peroxide A, the real oxidative species. The
evidence for its formation came from the typical IR absorption
peaks at 495, 640, and 950 cmÀ 1.[40,41] The three-membered ring
peroxide A then reacted with compound 1a and converted the
CαÀ OH to ketone form 2a, significantly reducing the bond
energy of the adjacent CÀ O bond[42–45] which was prone to
suffer a breakage to give guaiacol and p-methoxy acetophe-
none. The resulting acetophenone may undergo further
oxidative cleavage to give benzoic acids. In order to prove our
idea, compound 2a was synthesized as the substrate and
submitted to the optimized condition, giving rise to guaiacol
and p-methoxy benzoic acid as 1a did, which indicated 2a
2.4. Plausible Mechanism Discussion
On the basis of the above experiments, the possible mechanism
is proposed as shown in Scheme 1. It was observed that no
oxidation occurred in the absence of NaOH or TBHP. However,
ChemCatChem 2021, 13, 1–11
www.chemcatchem.org
7
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202100670
ChemCatChem
Scheme 1. Possible reaction mechanism of cleaving model compound 1a.
maybe the oxidative intermediate arising from 1a. When the
same procedure was applied to p-methoxy acetophenone, p-
methoxy benzoic acids was achieved, identifying the generation
of benzoic acids via continued oxidation of the corresponding
ketones.
through multiple experiments was subjected to five successive
run of optimized experiments. In each cycle, nano WO3 was
recovered by filtration and washed with water and ethanol for
three times. Then, the recovered nano WO3 was reused directly
°
after drying under vacuum at 80 C for 10 h. To our delight,
To gain insight into the low yield of guaiacol 3a, we also
observed the behavior of guaiacol under the same conditions.
The amount of guaiacol dropped off with time but no products
were detected by GC-MS, accounting for the disproportionately
lower yields of guaiacol and other phenols bearing methoxy
groups. MALDI-TOF MS analysis (Figure S3) reveals the absence
of guaiacol (m/z 124.14) and the presence of some unknown
compounds in the products corresponding to m/z 332.3, m/z
340.7, m/z 430.3 and m/z 509.2 respectively, indicating polymer-
ization may occur under the alkaline medium (Figure 2).
The Cα -H transfer process was studied using deuterium
labeled 1a’ as the substrate and toluene as the solvent to avoid
H/D exchange in an alkaline medium (see section 5.3 in SI for
details). The results demonstrated that the deuterium atom on
Cα didn’t transfer to the methyl group of the ketone product,
suggesting that the hydrogen atoms removed from the alcohol
group did not participate in the subsequent cleavage of the
ether CÀ O bond. Further, the kinetic isotope effect (KIE)
experiments identified that the dehydrogenation of 1a to 2a is
the rate-limiting step (section 5.4 in SI).
there was no obvious loss of activity after five cycles with
conversion of 97.9%, 97.8%, 97.4%, 97.3% and 97.0%,
respectively (Figure S7). As shown in XRD and XPS (Figure 1),
the recovered catalyst (the red line) shows peak absorptions at
the same positions but with decreased intensities when
compared with original nano WO3 (the dark line), indicating no
significant structural change occurred to the recycled catalyst
under the optimized reaction conditions. The SEM image of the
recycled nano WO3 (Figure S1c) reveals the decrease of the
particle size as well as the concomitant agglomeration, and the
latter may be considered the main reason for the decline in the
intensities in XRD and XPS. The decrease of the particle size was
beneficial for the catalytic activity of nano WO3, whereas
agglomeration showed the adverse effect and counteracted the
influence of the former. Overall, the catalytic performance of
recycled nano WO3 was not so much affected and after five
runs it still exhibited a similar activity as the original nano WO3.
2.6. Large Scale Reaction
GC/MS analysis revealed 6a, the dehydrated product of 1a,
formed in negligible amount in some cases. As we thought
dehydration was just a trivial side reaction under the optimized
conditions, no further study on 6a was performed.
As described in section 7 in SI, the synthetic utility of the
developed protocol was further demonstrated by a gram-scale
reaction with 1.37 g of 1a as the substrate and other reagents
remained the same proportion. To our delight, full conversion
was observed within 6 h, resulting in 17.2% guaiacol, 12.1% p-
methoxy acetophenone and 83.4% p-methoxy benzoic acid,
almost in the same yields as the microscale reaction.
2.5. Recyclability of Catalyst
Experiments were also conducted to examine the reusability of
nano WO3. Nano WO3 readily dissolved in the alkaline medium,
when acidified to pH 2~3 it precipitated from the mixture and
the heterogeneous nature made it easily removed from the
aqueous phase by filtration. The recyclable catalyst collected
ChemCatChem 2021, 13, 1–11
www.chemcatchem.org
8
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202100670
ChemCatChem
2.7. Catalytic Oxidation of Lignin
3. Conclusion
The results obtained with the model compounds encouraged
us to explore the possibility of applying the methodology to
the conversion of authentic lignin feedstock. However, the
organosolv lignin derived from pinewood powder could not be
depolymerized at all under the optimized conditions, this being
ascribed to the poor solubility of organosolv lignin in water.
Therefore, organic solvents miscible with water such as
tetrahydrofuran (THF), acetone, ethanol and 1, 4-dioxane, were
added to the reaction mixture as the co-solvent, respectively.
Among them, a 1, 4-dioxane/water (1:1, v/v) combination could
dissolve the organosolv lignin and provide a homogeneous
mixture which was then submitted to the same procedure but
still no signals were observed by GC-MS, indicating no reaction
occurred or products obtained could not be detected by GC-
MS. Based on the results of model compounds, phenolic
products may undergo further oxidation or polymerization
under the conditions and the other main products, aromatic
ketones and/or intermediates may undergo aldol condensation
under higher alkaline content. Although related precedents
found the use of ethylene glycol beneficial in reductive lignin
depolymerization or under acidolysis conditions,[46] we at-
tempted to add ethylene glycol to our alkaline and oxidative
medium in order to suppress the recondensation reactions. To
Herein, we report a mild one-pot oxidative strategy for the
depolymerization of lignin model compounds into phenols,
ketones, and aromatic acids, catalyzed by nano WO3 in
°
combination of TBHP and NaOH at 80 C. Nano WO3 has dual
features of both homogeneous and heterogeneous catalysts,
indicating the potential of full conversion and the possibility for
catalyst recycle. Upon the addition of NaOH, nano WO3 became
homogeneous and readily reacted with TBHP to produce the
real oxidative species, three-membered ring peroxide, which
was highly effective in oxidizing benzyl-OH to C=O and further
fragmentated β-O-4 bonds in lignin model compounds to give
the corresponding aromatic ketones or acids in good to
excellent yields. After reaction, the mixture was acidified to
precipitate WO3, which could be recycled 5 times without
significant loss of activity. When the modified protocol was
applied to the depolymerization of lignin, it offered 93.3 wt%
conversion and 80.4 wt% of liquid oil with 7.6 wt% of vanillic
acid, accounting for 91.6% selectivity in the lignin-derived
monomers. With this, the mild catalytic system with sufficient
activity for lignin depolymerization are promising for the
production of useful aromatics.
our delight, after the addition of 90 μL of ethylene glycol, the Acknowledgements
organosolv lignin could be effectively decomposed into mono-
°
mers and oligomers in 1, 4-dioxane/water (1:1, v/v) at 100 C
within 4 h, and only traces of residual lignin remained, high-
lighting the efficiency of our catalytic oxidative methodology.
Under the optimized conditions, the catalytic process gave
93.3 wt% conversion for lignin and 80.4 wt% of liquid oil[47]
with three main monomers identified as 1-(4-hydroxy-3-meth-
oxyphenyl) ethanone (M1, 0.3 wt%), vanillic acid (M2, 7.6 wt%),
This work was supported by the Fundamental Research Funds for
the Central Universities (China University of Mining and Technol-
ogy, 2019XKQYMS49) and a Project Funded by the Priority
Academic Program Development of Jiangsu Higher Education
Institutions.
and 3-hydroxy-4-methoxybenzoic acid (M3, 0.4 wt%) by GC-MS Conflict of Interest
(Figure 3 and Figure S9), where the selectivity for vanillic acid
accounted for 91.6% in the lignin-derived monomers. The three
monomers contain typical G-type unit, indicating that the
pinewood lignin comprises mainly of G units. QPEOTMS
characterization of the liquid oil (Figure 4b) showed that the
obtained products are mostly in the range of 100–600 μ,
indicating fragmentation towards smaller oligomers and mono-
mers. Furthermore, the clean profile suggests the breakage
mainly occurred to some specific sites in organosolv lignin,
thereby indicating the selectivity of this method.
To gain insight into the cleavage mechanism for real lignin,
2D-HSQC-NMR spectra of the side chain region for the organo-
solv lignin before and after reaction were compared (Figure 5,
for complete HSQC see Figure S8). It was found that cross
signals of methoxy groups and β-O-4 aryl ether bond were the
dominant linkages in pinewood lignin. Besides, β-5 linkages
(resinol, structure B) and β-β linkages (phenylcoumaran,
structure C) were observed. After the oxidation, most of the A,
B, C linkages and methoxy group disappeared in the side-chain
region of the lignin residue (Figure 5, panel b vs a), suggesting
that the catalytic system could effectively cleave the three
major linkages of A, B, and C as well as demethoxylation.
The authors declare no conflict of interest.
Keywords: β-O-4 Cleavage · Lignin depolymerization · Nano
WO3 · Oxidative depolymerization · One-pot
[1] A. K. Mohanty, S. Vivekanandhan, J-M. Pin, M. Misra, Science 2018, 362,
536–542.
[2] T. T. Li, N. Ji, Z. C. Jia, X. Y. Diao, Z. J. Wang, Q. L. Liu, C. F. Song, X. B. Lu,
ChemCatChem 2020, 12, 5288–5302.
[3] B. M. Matsagar, Z. Y. Wang, C. Sakdaronnarong, S. S. Chen, D. C. W.
Tsang, K. C. W. Wu, ChemCatChem 2019, 11, 4604–4616.
[4] J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius, B. M. Weckhuysen, Chem.
Rev. 2010, 110, 3552–3599.
[5] X. Chen, H. Y. Yang, Z. Y. Zhong, N. Yan, Green Chem. 2017, 19, 2783–
2792.
[6] S. Gillet, L. Petitjean, M. Aguedo, C. H. Lam, C. Blecker, P. T. Anastas,
Bioresour. Technol. 2017, 233, 216–226.
[7] M. D. Kärkäs, B. S. Matsuura, T. M. Monos, G. Magallanes, C. R. J.
Stephenson, Org. Biomol. Chem. 2016, 14, 1853–1914.
[8] S. H. Li, S. Q. Liu, J. C. Colmenares, Y. J. Xu, Green Chem. 2016, 18, 594–
607.
[9] G. Chatel, R. D. Rogers, ACS Sustainable Chem. Eng. 2014, 2, 322–339.
[10] P. J. Deuss, K. Barta, Coord. Chem. Rev. 2016, 306, 510–532.
ChemCatChem 2021, 13, 1–11
www.chemcatchem.org
9
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cctc.202100670
ChemCatChem
[11] I. Bosque, G. Magallanes, M. Rigoulet, M. D. Kärkäs, C. R. J. Stephenson,
ACS Cent. Sci. 2017, 3, 621–628.
[12] J-Y. Kim, J. Moon, J. H. Lee, X. J. Jin, J. W. Choi, Fuel 2020, 279, 118484.
[13] K. Soongprasit, V. Sricharoenchaikul, D. Atong, Energy Rep. 2020, 6, S6,
151–167.
[14] J. Kang, S. Irmak, M. Wilkins, Renewable Energy 2019, 135, 951–962.
[15] J. Guo, Y. L. Ma, J. Y. Yu, Y. J. Gao, N. X. Ma, X. Y. Wu, BMC Chem. 2019,
13, 36
[16] S. G. Santos, A. P. Marques, D. L. D. Lima, D. V. Evtuguin, V. I. Esteves,
Ind. Eng. Chem. Res. 2011, 50, 291–298.
[17] V. E. Tarabanko, N. Tarabanko, Int. J. Mol. Sci. 2017, 18, 2421.
[18] G. Magallanes, M. D. Kärkäs, I. Bosque, S. Lee, S. Maldonado, C. R. J.
Stephenson, ACS Catal. 2019, 9, 2252–2260.
[19] N. C. Luo, M. Wang, H. J. Li, J. Zhang, H. F. Liu, F. Wang, ACS Catal. 2016,
6, 7716–7721.
[20] C. F. Zhang, H. J. Li, J. M. Lu, X. C. Zhang, K. E. MacArthur, M. Heggen, F.
Wang, ACS Catal. 2017, 7, 3419–3429.
[21] M. Wang, J. M. Lu, X. C. Zhang, L. H. Li, H. J. Li, N. C. Luo, F. Wang, ACS
Catal. 2016, 6, 6086–6090.
[32] C. Z. Li, M. Y. Zheng, A. Q. Wang, T. Zhang, Energy Environ. Sci. 2012, 5,
6383–6390.
[33] H. W. Guo, B. Zhang, Z. J. Qi, C. Z. Li, J. W. Ji, T. Dai, A. Q. Wang, T. Zhang,
ChemSusChem 2017, 10, 523–532.
[34] J. W. Ji, H. W. Guo, C. Z. Li, Z. J. Qi, B. Zhang, T. Dai, M. Jiang, C. Q. Ren,
A. Q. Wang, T. Zhang, ChemCatChem 2018, 10, 415–421.
[35] T. M. A. Shaikh, A. Sudalai, Eur. J. Org. Chem. 2008, 4877–4880.
[36] H. F. Liu, H. J. Li, J. M. Lu, S. Zeng, M. Wang, N. C. Luo, S. T. Xu, F. Wang,
ACS Catal. 2018, 8, 4761–4771.
[37] J. M. Nichols, L. M. Bishop, R. G. Bergman, J. A. Ellman, Chem. Informa-
tionsdienst 2011, 132, 16725.
[38] A. Wu, B. O. Patrick, E. Chung, B. R. James, Dalton Trans. 2012, 41,
11093–11106.
[39] W. Huo, W. Z. Li, M. J. Zhang, W. Fan, H. M. Chang, H. Jameel, Catal. Lett.
2014, 144, 1159–1163.
[40] R. Vijayalakshmi, M. Jayachandran, C. Sanjeeviraja, Curr. Appl. Phys.
2003, 3, 171–175.
[41] Y. Lu, H. B. Yin, H. X. Wu, H. Liu, T. S. Jiang, Y. J. Wada, Catal. Commun.
2006, 7, 832–838.
[42] P. C. A. Bruijnincx, B. M. Weckhuysen, Nat. Chem. 2014, 6, 1035–1036.
[43] Z. Strassberger, A. H. Alberts, M. J. Louwerse, S. Tanase, G. Rothenberg,
Green Chem. 2013, 15, 768–774.
[22] T. Rinesch, C. Bolm, ACS Omega. 2018, 3, 8386–8392.
[23] H. F. Liu, M. Wang, H. J. Li, N. C. Luo, S. T. Xu, F. Wang, J. Catal. 2017,
346, 170–179.
[24] M. Wang, J. M. Lu, L. H. Li, H. J. Li, H. F. Liu, F. Wang, J. Catal. 2017, 348,
160–167.
[44] M. D. Kärkäs, ChemSusChem 2017, 10, 2111–2115.
[45] P. J. Deuss, M. Scott, F. Tran, N. J. Westwood, J. G. de Vries, K. Barta, J.
Am. Chem. Soc. 2015, 137, 7456–7467.
[25] S. Son, F. D. Toste, Angew. Chem. Int. Ed. 2010, 49, 3791–3794; Angew.
[46] C. W. Lahive, P. J. Deuss, C. S. Lancefield, Z. H. Sun, D. B. Cordes, C. M.
Young, F. Tran, A. M. Z. Slawin, J. G. de Vries, P. C. J. Kamer, N. J.
Westwood, K. Barta, J. Am. Chem. Soc. 2016, 138, 8900–8911.
[47] Y. X. Liu, C. Z. Li, W. Miao, W. J. Tang, D. Xue, C. Q. Li, B. Zhang, J. L. Xiao,
A. Q. Wang, T. Zhang, C. Wang, ACS Catal. 2019, 9, 4441–4447.
Chem. 2010, 122, 3879–3882.
[26] W. P. Deng, H. X. Zhang, X. J. Wu, R. S. Li, Q. H. Zhang, Y. Wang, Green
Chem. 2015, 17, 5009–5018.
[27] H. H. Luo, L. Y. Wang, G. S. Li, S. S. Shang, Y. Lv, J. Y. Niu, S. Gao, ACS
Sustainable Chem. Eng. 2018, 6, 14188–14196.
[28] W. B. Xu, X. Y. Li, J. Y. Shi, Int. J. Biol. Macromol. 2019, 135, 171–179.
[29] O. Y. Abdelaziz, S. Meier, J. Prothmann, C. Turner, A. Riisager, C. P.
Hulteberg, Top. Catal. 2019, 62, 639–649.
[30] H. Kim, H-Y. Yoo, S. Hong, S. Lee, S. Lee, B-S. Park, H. Park, C. Lee, J. Lee,
Appl. Catal. B. 2015, 162, 515–523.
[31] F. H. Mai, Z. Wen, Y. F. Bai, Z. W. Ma, K. Cui, K. Wu, F. Yan, H. Chen, Y. D.
Li, Ind. Eng. Chem. Res. 2019, 58, 10255–10263.
Manuscript received: May 9, 2021
Revised manuscript received: June 22, 2021
Accepted manuscript online: June 22, 2021
Version of record online: ■■■, ■■■■
ChemCatChem 2021, 13, 1–11
www.chemcatchem.org
10
© 2021 Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPERS
Catalytic oxidative depolymeriza-
tion: Herein we described a one-pot
process for the mild oxidative depo-
lymerization of lignin and lignin
model compounds catalyzed by
commercial nano WO3, along with
tert-butyl hydrogen peroxide (TBHP)
as the oxidant and NaOH as the
additive. The catalytic system
exhibits advantages of both homo-
geneous and heterogeneous
catalysts.
Prof. J. Liang*, M.-X. Wang, Prof. Y.-P.
Zhao*, W.-W. Yan, X.-G. Si, G. Yu,
Prof. J.-P. Cao, Prof. X.-Y. Wei
1 – 11
Nano WO3-Catalyzed One-Pot
Process for Mild Oxidative Depoly-
merization of Lignin and its Model
Compounds