Metal-free and mild photo-thermal synergism in ionic liquids for lignin Cα-Cβbond cleavage to provide aldehydes

Article abstract of DOI:10.1039/d1gc00784j

Direct cleavage of the C-C bond in lignin linkages is a promising route to afford value-added aromatics, which, however, usually involves metal-based catalysts and harsh conditions. Here, a photo-thermal strategy is reported to deconstruct the Cα-Cβ bond in lignin β-O-4 and β-1 interlinkages in the ionic liquid [BMim][NTf2]. It was found that the synergism of UV light and heating could trigger the Norrish type I reaction by inducing the cleavage of the Cα-Cβ bond and afford aromatic aldehydes through a free radical route in modest yields. Meanwhile, [BMim][NTf2] could interact with lignin moieties and stabilize the intermediates, which significantly contribute to the Cα-OH protonation and accelerate lignin interunit linkage breakage. In this unique route, the Cα-Cβ bond could be cleaved at 50 °C under UV light irradiation without the presence of a metal. Further comparison with photochemical or thermochemical alone strategies demonstrates that the photo-thermal route offers a significant improvement in the Cα-Cβ bond cleavage reactivity. Alkali lignin could also be degraded into aromatic monomers in [BMim][NTf2] using the protocol. Given the necessity of biomass upgradation, this work may provide a green and unique technology to boost the inert C-C bond cleavage upon photo-thermal synergism under metal-free and mild conditions that may underpin future advances in photo-thermal transformation and renewable energy production.

Full text of DOI:10.1039/d1gc00784j

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
Metal-free and mild photo-thermal synergism in
ionic liquids for lignin Cα–Cβ bond cleavage to
provide aldehydes†
Cite this: Green Chem., 2021, 23,
5524
Ying Kang,a,b Xiaoqian Yao,a Yongqing Yang,a,b Junli Xu,a Jiayu Xin,a,b,c
a,b,c
Qing Zhou,a,b,c Minjie Li,d Xingmei Lu
*
a,b,c and Suojiang Zhang
*
Direct cleavage of the C–C bond in lignin linkages is a promising route to afford value-added aromatics,
which, however, usually involves metal-based catalysts and harsh conditions. Here, a photo-thermal strat-
egy is reported to deconstruct the Cα–Cβ bond in lignin β-O-4 and β-1 interlinkages in the ionic liquid
[BMim][NTf2]. It was found that the synergism of UV light and heating could trigger the Norrish type I
reaction by inducing the cleavage of the Cα–Cβ bond and afford aromatic aldehydes through a free
radical route in modest yields. Meanwhile, [BMim][NTf2] could interact with lignin moieties and stabilize
the intermediates, which significantly contribute to the Cα–OH protonation and accelerate lignin interunit
linkage breakage. In this unique route, the Cα–Cβ bond could be cleaved at 50 °C under UV light
irradiation without the presence of a metal. Further comparison with photochemical or thermochemical
alone strategies demonstrates that the photo-thermal route offers a significant improvement in the Cα–Cβ
bond cleavage reactivity. Alkali lignin could also be degraded into aromatic monomers in [BMim][NTf2]
using the protocol. Given the necessity of biomass upgradation, this work may provide a green and
unique technology to boost the inert C–C bond cleavage upon photo-thermal synergism under metal-
free and mild conditions that may underpin future advances in photo-thermal transformation and renew-
able energy production.
Received 2nd March 2021,
Accepted 27th May 2021
DOI: 10.1039/d1gc00784j
rsc.li/greenchem
the homolytic bond dissociation enthalpy of C–C bonds
(69–123 kcal mol−1) is theoretically higher than that of C–O
Introduction
Efficient utilization of sustainable lignin resources, owing to bonds (52–80 kcal mol−1) and additional C–C bonds are gener-
its functionalized aromatic rich feature, is currently of tremen- ated during some delignification processes,11 which makes the
dous interest due to the dwindling fossil-based feedstocks and disruption of the C–C bonds, especially the β-O-4 linkages due
the ever-increasing requirement of aromatics.1,2 Structurally, to its abundance (percentage: 45–60%),12 draw high levels of
plenty of C–C and C–O bonds are present in lignin to form attention for native lignin upgradation.
various interunit linkages, such as the β-O-4, β-1, β–β or 5–5
Commonly, the Cα–Cβ bond in lignin β-O-4 linkages is more
types, via radical coupling of phenylpropanoid units.3–5 prone to cleavage after the oxidation of Cα–OH into the CαvO
Cleaving the C–C/C–O bonds is a pivotal issue with a view to group.13 Based on the above finding, a two-step β-O-4 fragment
transforming lignin into high value-added aromatics.6–10 Yet, methodology has been widely explored involving the following
two main pathways: (1) the oxidation of the Cα–OH group to
CαvO in β-O-4-alcohols and (2) the downstream degradation
aBeijing Key Laboratory of Ionic Liquids Clean Process, CAS Key Laboratory of Green
of β-O-4-ketones bearing active Cβ–H and Cα–Cβ bonds.8,14–19
Process and Engineering, State Key Laboratory of Multiphase Complex Systems,
To further simplify the operation, direct Cα–Cβ bond fragmen-
Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190,
tation in β-O-4-alcohols in one pot has been extensively investi-
China. E-mail: xmlu@ipe.ac.cn, sjzhang@ipe.ac.cn
gated and it produced encouraging results.20,21 Nevertheless,
strong oxidants and high temperature/pressure are commonly
exploited in these existing strategies, demonstrating the
difficulty in directly cleaving the inert Cα–Cβ bond. Hence,
exploiting greener technologies for Cα–Cβ bond cleavage under
mild conditions is a great necessity but is still challenging.
Meanwhile, the cleavage of the Cα–Cβ bond in other linkages
bSchool of Chemistry and Chemical Engineering, University of Chinese Academy of
Sciences, Beijing 100190, China
cInnovation Academy for Green Manufacture, Chinese Academy of Sciences, Beijing
100190, China
dCollege of Quality & Safety Engineering, China Jiliang University, Hangzhou
310018, China
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
d1gc00784j
5524 | Green Chem., 2021, 23, 5524–5534
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
like β-1 type (percentage: 1–15%)11 should also be taken into
account.
In this respect, the photochemical process emerges as a
promising alternative for chemoselective cleavage of lignin ali-
phatic linkages.22,23 Particularly, upon the excitation of a
photocatalyst by light irradiation, cleavage of the C–C bond in
lignin could be achieved by its direct homolytic cleavage
through a free radical route.11 Encouraging results could be
achieved, while the photochemical methods face limited
photon efficiency issues. Fortunately, studies on photo-
thermal strategies, which couple thermal and photonic stimuli
to drive chemical transformations, have proven that they have
higher efficiency and lower energy consumption due to the
formed photo-thermal synergism.24,25 In this process, the
thermal effect could promote effective collisions among reac-
tant molecules and render them in excited states, and then the
energy for the excited molecules to overcome the activation
barrier could be easily obtained by light irradiation, indicating
that the formed photo-thermal synergism in a reaction could
contribute to lowering of the energy barrier and promotion of
active-site generation.26,27 Considerable reduction in apparent
activation energy and operating temperature has been reported
in a photo-thermal regime due to the photo-thermal syner-
gism.28 Whereas, metal-based catalysts are commonly used in
the photo-thermal route to gain photo and thermal energy for
excitation, triggering the subsequent active species generation
like free radicals and chemical bond transition, which may
lead to the depletion of limited metal resources and environ-
mental issues. Interestingly, lignin is a good light absorber
and could absorb UV light energy to yield active species due to
the chromophores in its structure, which is a relatively time-
consuming process in air.29–31 In this case, the combination of
thermal and photonic stimuli may offer a potential green
pathway, in which the photo-thermal synergism excites lignin
substrates to yield free radicals upon the absorption of
photons and thermal energy by lignin rather than metal-based
catalysts, and then lower the activation barrier to induce the
homolytic cleavage of the inert C–C bond cleavage even at rela-
tively low temperature with energy saving and high efficiency.
As reported in previous studies, ionic liquids (ILs) could
also accelerate the inert chemical bond cleavage in lignin
β-O-4-ketones by the thermal or photo alone strategy, in which
the interactions between ILs and lignin linkages could lower
the chemical bond energy and activate them, and thus induce
the formation of active substances like free radicals and the
subsequent chemical bond scission.29,32–34 Hence, in a more
powerful photo-thermal strategy, the use of IL may further
accelerate lignin interunit linkage breakage, including β-O-4-
alcohols and β-1 linkages, by interacting with lignin.
Additionally, due to the nature of ILs, the ions in ILs could
coordinate with the reaction intermediates and stabilize
them,35 which could contribute to reduction of the activation
Scheme 1 Photo-thermal Cα–Cβ bond cleavage of lignin linkages in
this work.
opportunity to accelerate lignin interunit linkage breakage by
the judicious implementation of the activation of ILs, which,
with the assistance of photo-thermal synergism, may realize
one-pot metal-free lignin C–C bond cleavage in a green, simple
but effective approach.
On the basis of the above consideration, we herein reported
a metal-free photo-thermal chemical strategy for the Cα–Cβ
bond cleavage of lignin under mild conditions in the ionic
liquid ([BMim][NTf2]) system in one pot. Using lignin model
compounds, we confirmed a significant improvement of the
synergistic effect of UV light and heating on the cleavage of the
Cα–Cβ bond in β-O-4 and β-1 linkages by testing thermal/photo
alone and photo-thermal technologies. Meanwhile, the pro-
motion effect of [BMim][NTf2] on the chemical bond cleavage
was verified by DFT calculations, where the interactions of IL
ions with lignin model compounds and the intermediates
were investigated and analyzed in depth. A free radical
mediated mechanism of the Cα–Cβ bond cleavage upon the
initiation of ILs and the Norrish type I reaction triggered by
photo-thermal synergism were revisited and clarified based on
the free radical scavenging experiment, tracking the product
distribution with time, chemical calculations and so on.
Furthermore, native lignin conversion under photochemical,
thermochemical and photo-thermal conditions was also tested
to verify the synergism of UV light and heating. The structural
variation of native lignin during the degradation process was
carefully investigated. Owing to the photo-thermal synergism
and the promotion of ILs, the photo-thermal chemical strategy
shows a high activity in the one-pot conversion of lignin by the
cleavage of inert chemical bonds (Scheme 1). Additionally, this
report will also be of considerable interest to explore photo-
thermal chemical routes for biomass upgradation, in which
the cleavage of C–C bonds is a key issue.
Results and discussion
Photo-thermal degradation of β-O-4 lignin model compounds
barrier and promotion of the chemical bond disruption. In 2-Phenoxy-1-phenylethanol (1), a common β-O-4 lignin model
consequence, considering all the abovementioned advantages compound, was initially studied to explore the photo-thermal
of ILs and their other attractive features like high stability, des- technology for the Cα–Cβ bond cleavage. It was found that in
ignability and negligible vapor pressure,36 we foresaw an the IL system, compound 1 was degraded into benzaldehyde,
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 5524–5534 | 5525

View Article Online
Paper
Green Chemistry
phenylacetaldehyde and phenol as the main products under
photo-thermal conditions (Table 1). As benzaldehyde is the
dominant product derived from the Cα–Cβ bond cleavage of 1,
we evaluate the efficiency of the photo-thermal reaction for the
Cα–Cβ bond cleavage by the yield of benzaldehyde. The photo-
thermal reaction activity is highly dependent on the compo-
sition of ILs. The absence of anion [NTf2]− resulted in low con-
version of 1 (Table 1, entries 1 and 2) as well as low yield of
mononuclear aromatic products. Upon changing anion of ILs
with [NTf2]−, not only the conversion of compound 1 signifi-
cantly increased but also the product benzaldehyde was
detected. Apparently, ILs with anion [NTf2]− favor the conver-
sion of compound 1 under the photo-thermal conditions and
lead to the cleavage of the Cα–Cβ bond. The influence of cation
was also investigated by varying the cation structures, while
the anion was maintained with [NTf2]− (Table 1, entries 3–7).
Different cation structures could result in different reaction
efficiencies, although the difference in efficiency was not as
obvious as that of anions. The cation [BMim]+ exerted the
highest efficiency among all the cations used, which allowed
95.1% conversion and 18.0%, 14.0% and 50.4% yields of benz-
aldehyde, phenylacetaldehyde and phenol (Table 1, entry 3).
From the above results, we can deduce that the IL composed
of the [BMim]+ cations and [NTf2]− anions provided a promis-
ing reaction system for the Cα–Cβ bond cleavage of compound
1 in a green and mild approach.
Varying the species of the Brønsted acid also has great
influence on the reaction activity. As shown in Fig. 1a, several
Brønsted acids referred to as inorganic acid, organic acid and
the Brønsted acid ionic liquids were used to explore their
effects on the photo-thermal reaction. On the whole, low acid
strength led to low reaction efficiency. For example,
[PrHSO3Mim][H2PO4] is with the lowest acid strength among
all the Brønsted acid ionic liquids used,29 exerted the lowest
efficiency by affording 11.8% conversion and lower than 5%
product yields. Upon increasing the acid strength of the
Brønsted acid, higher conversion and product yields were
obtained. When H2SO4 was applied as the acid catalyst, nearly
complete conversion of 1 was achieved with 20.1% yield of
Fig. 1 Influence of the Brønsted acid species and H2SO4 amount under
the following conditions: 1 5 mg, [BMim][NTf2] 250 mg, temperature
60 °C, UV light (100 mW cm−2), reaction time 20 min, air atmosphere, c
[H+] (0.2 mmol mL−1) (a); varying amounts of H2SO4 (b).
benzaldehyde, which is the highest among all the Brønsted
acids used. Hence, H2SO4 was used for the following reactions.
Increasing the H2SO4 amount also benefitted the generation of
benzaldehyde. As shown in Fig. 1b, almost complete conver-
sion was achieved in all cases with H2SO4, while the product
Table 1 Influence of IL’s structure on the photo-thermal degradation of 1
Entry
IL
Conv. (%)
Phenol (%)
Benzaldehyde (%)
Phenylacetaldehyde (%)
1
2
3
4
5
6
7
[BMim][OAc]
35.6
<1
95.1
96.5
98.1
98.6
99.8
n.d.
n.d.
50.4
43.3
49.6
22.8
27.4
n.d.
n.d.
18.0
16.1
15.9
7.9
n.d.
n.d.
14.0
13.0
11.0
7.0
[BMim][SCN]
[BMim][NTf2]
[EMim][NTf2]
[BzMim][NTf2]
[CPMim][NTf2]
[HOEtMim][NTf2]
4.6
5.5
n.d. = not detected. Reaction conditions: 1 5 mg, IL 250 mg, [PrHSO3Mim][OTf] 8 wt%, temperature 60 °C, UV light (100 mW cm−2), reaction
time 20 min, atmosphere.
5526 | Green Chem., 2021, 23, 5524–5534
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
distribution was sensitive to the Brønsted acid amount added. cleavage. In addition, the product distribution was also
The results show that the generation of benzaldehyde observed at low temperature (10 °C) with UV light irradiation
increased with H2SO4 addition and afforded 20.4% yield under of various light intensities (50 mW cm−2 and 100 mW cm−2),
the conditions of 1 wt% H2SO4. While the benzaldehyde yield these conditions afforded no benzaldehyde, meaning that heat
exhibits a decreasing trend with H2SO4 over 1 wt%, which also has a non-negligible influence on the Cα–Cβ bond clea-
might be related to the occurrence of side reactions due to vage of 1. Increasing the temperature contributed to the gene-
excess acid. In sharp comparison, the photo-thermal reaction ration of benzaldehyde, this afforded the highest yield of benz-
without the Brønsted acid shows no conversion of compound aldehyde at 50 °C. However, continuously increasing the temp-
1, which means that the reaction occurs only in the presence erature above 50 °C leads to a decline in the benzaldehyde
of the Brønsted acid, implying the important effect of H+.
yield, which might be due to the side-reaction like product
The deconstructive strategy for the Cα–Cβ bonds in model 1 repolymerization at higher temperature.37 Given the above
may be dependent on UV-light irradiation and heat. For a experiment results, we can conclude that both UV light and
better understanding of the degradation process, we also con- heat are crucial for the Cα–Cβ bond cleavage of 1 to afford
sidered diverse illumination with various temperatures to benzaldehyde in the photo-thermal degradation, which indi-
further explore the contributions of UV light irradiation and cates a synergism between these two factors. As phenylacetal-
heat on the chemical bond cleavage of compound 1. As shown dehyde is the product obtained from the Cβ–O bond cleavage
in Fig. 2a, the production of benzaldehyde increases with an of compound 1, the generation of which was also analysed at
increase in light intensity from 0 to 100 mW cm−2, the yield of various temperatures with UV light or in the dark, as shown in
benzaldehyde reached 21.1% when the light intensity was Fig. 2b. Interestingly, similar yields of phenylacetaldehyde
100 mW cm−2 at 50 °C. For comparison, in the case of dark (∼50%) were obtained at different temperatures (10 to 60 °C)
(0 mW cm−2), no benzaldehyde was detected at various temp- when there was no UV light irradiation. In comparison, the
eratures ranging from 10 to 60 °C, implying that UV light synergism of UV light and heat lead to a significant decline in
irradiation is crucial for the Cα–Cβ bond cleavage of 1 to afford phenylacetaldehyde generation. More details of degradation of
benzaldehyde. We then increased the temperature to as high 1undert UV light/heating alone conditions are presented in
as 220 °C and when the reaction proceeded in the dark Fig. S1.† Given the fact that no phenylacetaldehyde was
(Table S1†), no benzaldehyde was detected, further confirming obtained without H2SO4 (Fig. 1b), it could be concluded that
the important effect of UV light irradiation on the Cα–Cβ bond the H+ is mainly responsible for the Cβ–O bond cleavage, while
the Cα–Cβ bond cleavage of compound 1 is primarily depen-
dent on the synergism of UV light and heat.
To ensure the critical role of the synergism of UV light and
heat, photo-, thermo- and photo-thermo degradation were con-
ducted and the results were compared, as shown in Fig. 2c.
For the production of benzaldehyde, continuous UV light
irradiation without heating afforded no benzaldehyde after
5–35 minutes of the reaction. We further tested the reaction at
50 °C without and with UV-light irradiation by conducting the
degradation in the dark for a few minutes and then continuing
it under UV light for 20 minutes. In the first stage of the reac-
tion in the dark (off), benzaldehyde was not detected in all
cases. When UV light illuminated the reaction system (on) in
the subsequent stage, the yield of benzaldehyde showed a sig-
nificant increase from 0% to around 20%. That is, it is the
synergism of UV light and heat that is responsible for the Cα–
Cβ bond cleavage to produce benzaldehyde, and the light-
enhancement process is reproducible. The generation of phe-
nylacetaldehyde was also observed, as shown in Fig. 2c, which
presented an obvious contrary trend in comparison to benz-
aldehyde. A large amount of phenylacetaldehyde was gener-
ated under only heating (50 °C, 0 mW cm−2) or receiving UV
light irradiation (5 °C, 100 mW cm−2), respectively. Once the
Fig. 2 Production of benzaldehyde (a) and phenylacetaldehyde (b) reaction was conducted under the combined effect of light and
heat at the same time (50 °C, 100 mW cm−2), the output of the
phenylacetaldehyde, which was generated in the first stage
under the following conditions: 1 5 mg, [BMim][NTf2] 250 mg, H2SO4
1 wt%, temperature 10–60 °C, UV light (power density: 0–100 mW
cm−2), reaction time 20 min, air atmosphere; benzaldehyde and pheny-
(dark) is significantly reduced, and a certain amount of benz-
lacetaldehyde production under continuous illumination without
aldehyde is produced. These results indicate that the conver-
sion of phenylacetaldehyde occurs under the synergism of
heating and at 50 °C with without UV light irradiation (power density:
100 mW cm−2) (c).
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 5524–5534 | 5527

View Article Online
Paper
Green Chemistry
the presence of the Brønsted acid is well known,38,39 including
the formation of an intermediate with the enol ether structure
catalyzed by H+, which is then hydrolyzed to fragment the Cα–
O bond. The cleavage of the Cα–Cβ bond is closely related to
the synergism of light and heat as discussed above. As many
photochemical reactions in previous research studies were
involved in the free radical mechanism, we envisioned that the
generation of benzaldehyde from compound
1 proceeds
through a free radical mechanism. To verify this conjecture, we
added the radical scavenger 2,2,6,6-tetramethyl-piperidinyloxy
(TEMPO) (1.4 wt%) to the reaction system and analyzed the
product distribution. As shown in Table S1,† only phenylace-
taldehyde was detected in the products after 20 min of reac-
tion, and no benzaldehyde was generated. This phenomenon
indicated that the formation of benzaldehyde is based on a
free radical process.
Fig. 3 Reuse of [BMim][NTf2] under the following conditions: 1 5 mg,
[BMim][NTf2] 1000 mg, H2SO4 1 wt%, temperature 50 °C, UV light
(100 mW cm−2), reaction time 20 min, air atmosphere.
Further, electron paramagnetic resonance (EPR) was per-
formed to detect and identify the free radicals generated in the
photo-thermal reaction. The spectra of the 1/[BMim][NTf2]/
H2SO4(1 wt%) system was recorded with 200 mM DMPO at
50 °C under 20 mW cm−2 UV light irradiation with time (by
using the Xenotest apparatus equipped with an X-band EPR
spectrometer) or in the dark are shown in Fig. 4. It was evident
that free radicals were generated during the photo-thermal
reaction of 1. Through determining the hyperfine splitting
constants (A factors), two typical radicals could be labelled.
Scheme 2 Photo-thermal degradation of lignin model compound 2.
H
light and heat, which might be relevant to the generation of
benzaldehyde.
The spin adduct (AN = 1.33–1.39 mT, Aβ = 0.97 mT) could be
•
assigned to the generation of OOR.40 Meanwhile, the appear-
β
The cyclability of the ionic liquid system was investigated as
shown in Fig. 3. The results demonstrate that the ionic liquid
maintains high activity for the degradation of compound 1
after being circulated 5 times. In addition, we performed and
compared the 1H and 13C NMR spectra of the original
[BMim][NTf2] and that after five cycles, which presented
almost the same signals for the IL, as shown in Fig. S2.† The
above results indicated the stability of [BMim][NTf2] under UV
light and heat conditions.
Another β-O-4 lignin model compound, 2-(2-methoxyphe-
noxy)-1-phenylethanol (2), was also employed to evaluate the
efficiency of the photo-thermal method for the Cα–Cβ bond
cleavage in [BMim][NTf2]. As shown in Scheme 2, under UV
light and heating conditions for only 10 min (other conditions
were similar to those presented in Fig. 3), 97.6% of compound
2 could be converted and afford benzaldehyde with a yield of
13.1%, which is derived from the Cα–Cβ bond cleavage of com-
pound 2. This phenomenon further demonstrates the applica-
bility of the photo-thermal method in [BMim][NTf2] to frag-
ment lignin β-O-4 linkages by disrupting the Cα–Cβ bond
through a green and mild pathway.
ance of the A factors of AN = 1.43 mT and AH = 2.14 mT could
•
be reasonably assigned to the R.41 In contrast, no signals for
free radicals were observed when the reaction system was kept
in the dark, indicating that the formation of free radicals was
triggered by UV light irradiation. Furthermore, the radical
signal intensity varied with time, which was the strongest at
the beginning of the reaction, and then gradually decreased
with respect to the depletion of the reactant. For comparison,
the EPR detection of [BMim][NTf2] was conducted under the
same photo-thermal conditions, while almost no signal
Mechanism
It is very interesting to investigate the reaction pathway for the
photo-thermal degradation of compound 1 to products. As
aforementioned, the reaction led to the Cα–Cβ bond and Cα–O
bond cleavage. The reaction profile for the Cα–O bond cleavage
in the lignin β-O-4 model compound involving hydrolysis in Fig. 4 EPR spectra of 1-ILs photo-thermal system at various times.
5528 | Green Chem., 2021, 23, 5524–5534
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
appeared, as shown in Fig. S3,† indicating that no free radical ions locate around 1. Different initial configurations were opti-
was generated from [BMim][NTf2] under UV light irradiation. mized, and the conformers with the lowest energy were
To get more insight into the reaction mechanism, we obtained, as shown in Fig. 6. The NPA (natural population ana-
tracked the photo-thermal reaction with time at 50 °C under lysis) charge distributions by NBO calculations of the ‘O’ atom
UV-light irradiation with 100 mW cm−2 light intensity, as in the –OH group of 1 structures are depicted in Table 2. For
shown in Fig. 5. In the initial stage of the reaction, phenylace- the same ‘O’ atom of 1, the NPA charge values and bond
taldehyde was firstly generated while the output of benz- lengths exhibited the same increasing trends. Obviously, the
aldehyde was low. As the reaction proceeded, the yield of phe- charge values change from −0.732 to −0.742 for 1 with [NTf2]−,
nylacetaldehyde decreased significantly, accompanied by the to −0.769 for 1 with [BMim]+ and to −0.755 for 1 with
generation of a certain amount of benzaldehyde. Continuously [BMim][NTf2], indicating that the electrons are concentrated
increasing the reaction time over 20 minutes led to the for the ‘O’ atom. The increased electron density around the ‘O’
decrease of all products, which might be relevant to the side- atom is favorable for the protonation of the –OH group in the
reaction under the photo-thermal conditions. Meanwhile, the presence of H+. Meanwhile, the interaction between 1 and
photo-thermal degradation of 1 under N2 and O2 was also con- [BMim][NTf2] also leads to changes in the bond lengths. The
ducted to investigate the effect of the atmosphere on the reac- Cα–O bond length becomes longer increasing from 1.426 Å to
tion, as shown in Table S2.† 18.0% yield of benzaldehyde was 1.433 Å for the interaction of 1 with [NTf2]−, to 1.449 Å for that
afforded under an O2 atmosphere, which, however, showed a with [BMim]+ and to 1.445 Å for that with [BMim][NTf2].
significant decline under a N2 atmosphere, indicating the Noteworthily, it seems that cation [BMim]+ led to greater
necessity of O2 for the generation of benzaldehyde from the changes in charge and bond lengths by interacting with 1,
Cα–Cβ bond cleavage of 1.
while the cation could directly interact with the –OH group
Given the above experimental results, we speculated that it and cause steric hindrance, preventing the H+ from attacking
was phenylacetaldehyde that might be converted to benz- the –OH group. Therefore, it is the anion [NTf2]− which plays a
aldehyde under the synergism of UV light and heat. In order to major role in the protonation of the –OH group in 1. The com-
prove this speculation, phenylacetaldehyde was used as the bined effect of cation [BMim]+ and anion [NTf2]− makes the
substrate to perform the photo-thermal reaction while the charge density of the ‘O’ atom in the –OH group greater and
other conditions were maintained the same, as shown in the Cα–O bond length longer.
Fig. S4.† In this case, benzaldehyde was observed, supporting
Furthermore, the formation of the carbocation intermediate
the above-indicated conversion of phenylacetaldehyde to benz- is commonly involved during dehydration, which exhibits the
aldehyde, which might be related to the well-known Norrish highest energy along the reaction pathway.43 Hence, the stabi-
Type I reaction for the oxidation of an aldehyde involving the lity of the carbocation intermediate in [BMim][NTf2] is of great
formation of free radicals.42
importance for the hydrolysis of β-O-4-alcohol 1. Based on the
Meanwhile, the IL [BMim][NTf2] could also significantly above consideration, the interactions of IL with the carbo-
trigger the chemical bond cleavage of 1 according to the cation intermediate that might form during hydrolysis of 1
experimental result of IL’s structure evaluation. Hence, we
studied the interaction between [BMim][NTf2] and 1 by DFT
calculations and was anticipated to accelerate the β-O-4 bond
cleavage of 1. The interaction structures of 1 with cation
[BMim]+, anion [NTf2]− and ionic liquid [BMim][NTf2] were
analyzed and calculated. There are possible positions in which
Fig. 6 Optimized structures of [BMim][NTf2] (a), 1 (b), 1 with [NTf2]− (c),
1 with [BMim]+ (d), and 1 with [BMim][NTf2] (e) by B3LYP/6-311 + G(d,p).
Table 2 NPA charge distributions and the changes of bond length (Å)
for different structures
1 with
1 with
1 with
Structure
1
[NTf2]− [BMim]+ [BMim][NTf2]
Fig. 5 Study of product distribution with time under the following con-
ditions: 1 5 mg, [BMim][NTf2] 250 mg, H2SO4 1 wt%, temperature 50 °C,
UV light (100 mW cm−2), air atmosphere.
NPA charge of O
Bond length of C–O
bond (Å)
−0.732 −0.742 −0.769
1.426 1.433 1.449
−0.755
1.445
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 5524–5534 | 5529

View Article Online
Paper
Green Chemistry
were also selected to conduct geometry optimizations and radical which then transformed to benzaldehyde and carbonic
energy calculations. There is no stable structure for the carbo- oxide after hydrogen abstraction by the latter.
cation interacting with cation [BMim]+, because both of them
To verify the formation of carbonic oxide, FT-IR analysis of
are with a positive charge. The possible positions in which the reaction system with and without 1 was conducted as
anion [NTf2]− and [BMim][NTf2] were located around the car- shown in Fig. S5.† Compared to the spectrum without 1, a new
bocation intermediate were analyzed and the conformers with peak at 2017 cm−2 typically for CO appeared in the system with
the lowest energy were obtained and are shown in Fig. 7. As 1, indicating the generation of CO during the photo-thermal
expected, the interaction of the carbocation intermediate with reaction.
anion [NTf2]−/[BMim][NTf2] could cause significant changes in
the interaction energies which is 60.2 kcal mol−1 for the carbo-
cation with [NTf2]−, and 17.3 kcal mol−1 for [BMim][NTf2].
Apparently, the results indicated the strong interaction of the
carbocation with the ionic liquids (especially the anion
[NTf2]−), demonstrating that the carbocation becomes more
stable in the IL [BMim][NTf2], which, as a consequence, is ben-
eficial for the hydrolysis of 1 and the subsequent β-O-4 bond
cleavage.
Photo-thermal degradation of β-1 lignin model compounds
To further explore the application of the photo-thermal system
for the Cα–Cβ bond cleavage of lignin moieties, the lignin
model compound with the β-1 linkage, bibenzyl (3), was tested
for the cleavage of the Cα–Cβ bond under photo-thermal con-
ditions (Table 3). Corresponding aromatics, benzaldehyde and
phenol, were formed as the dominant products during the
reaction through the Cα–Cβ bond fragment of 3, the yields of
which gradually increased with the progress of the reaction,
along with an increase in the conversion. While no products
were detected only under heating (50 °C, 0 mW cm−2) or UV
light irradiation (5 °C, 100 mW cm−2), further signifying the
effect of non-negligible photo-thermal synergism on the Cα–Cβ
bond cleavage. Additionally, by comparing the constituents of
the substrate and the products, we noticed that additional
oxygen atoms appeared in the products, confirming the par-
ticipation of oxygen in the degradation of 3. Although the
modest yields of products were obtained, these cases still indi-
cate that the photo-thermal system could lead to the degra-
dation of lignin β-1 linkages and produced aromatic aldehydes
and phenols from the inert Cα–Cβ bond cleavage through a
green and mild pathway.
Hence, taking into account all the aforementioned results,
we proposed a photo-thermal driven Norrish Type I reaction
mechanism for the Cα–Cβ bond cleavage of compound 1 in
tandem with the hydrolysis accelerated by [BMim][NTf2], as
shown in Scheme 3. Upon the interaction of [BMim][NTf2]
with 1, the Cα hydroxyl group was initially protonated by the
H+ from the Brønsted acid and continued to dehydrate to
harvest the benzyl cation intermediate (a), which was stabil-
ized in [BMim][NTf2]. Then, a underwent deprotonation and
converted to the enol ether substructure (b), which was hydro-
lyzed under the attack of H2O to produce phenol and phenyla-
cetaldehyde. Afterward, phenylacetaldehyde underwent the
Norrish Type I reaction upon heating and under UV light
•
irradiation, affording a benzyl radical and COH radical.42,44
Subsequently, O2 attacked the benzyl radical to yield a peroxy
Photo-thermal degradation of alkali lignin
To get a better understanding of the photo-thermal degra-
dation of alkali lignin in ionic liquids, GC-MS analysis was per-
formed to evaluate the reaction at the structural level. As listed
in Table 4, a broad variety of monomers and dimers could be
Table 3 Photo-thermal degradation of compound 3
Fig. 7 Optimized structures of the carbocation intermediate (a) and its
interaction with [NTf2]− (b) and with [BMim][NTf2] (c) by B3LYP/6-311 +
G(d,p), and the corresponding changes of interaction energy (E).
Entry
Time (min)
Conv. (%)
Benzaldehyde (%)
Phenol (%)
1a
2b
3
4
5
20
20
10
20
30
40
50
<1
<1
<1
7.6
32.2
38.4
51.7
n.d.
n.d.
n.d.
10.0
13.5
9.4
n.d.
n.d.
n.d.
15.8
20.7
22.3
23.9
6
7
7.6
n.d. = not detected. a UV light (100 mW cm−2) at 5 °C. b In the dark at
50 °C. Reaction conditions: 3 5 mg, [BMim][NTf2] 250 mg, H2SO4
1 wt%, temperature 50 °C, UV light (100 mW cm−2), reaction time
20 min, atmosphere.
Scheme 3 Possible mechanism of photo-thermal degradation of 1.
5530 | Green Chem., 2021, 23, 5524–5534
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
Table 4 Percentages of liquid extracts obtained from the photo-thermal degradation of alkali lignin
Molecular
formula
Unit
type
Yield/wt
%
Yielda/wt
%
Yieldb/wt
%
Selectc/
%
Entry Name
1
2
3
4
5
6
7
8
Phenol, 4-(sec-butyl)
Phenol, 4-(tert-butyl)
Phenol, 3-(tert-butyl)
C10H14
C10H14
C10H14
C11H16
O
O
O
O
H
H
H
B
B
H
B
S
0.010
0.011
0.053
0.016
0.212
0.269
0.013
0.058
0.024
0.010
0.082
0.349
0.182
n.d.
n.d.
n.d.
n.d.
0.041
0.077
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
0.033
0.062
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
0.06
0.07
0.32
0.10
1.28
1.62
0.08
0.35
0.14
0.06
0.49
2.10
1.09
Benzene, 1-ethoxy-2-isopropyl
Benzene, 1-methoxy-4-vinyl
Phenol, 5-methyl-2-propenyl
Benzaldehyde, 4-isopropyl
Phenol, 2,6-dimethoxy-4-acetyl
Phenol, 2-methoxy-methyl acrylate
Dibutyl phthalate
Benzene, 1-(1-methoxy-4-methyl-3-amylene)
1,3-Bis(p-methoxyphenyl)-propylene
2,4-Bis(4-methoxyphenyl)-3,5-dimethyl-
tetrahydrofuran
C9H10
C10H12
C10H12
O
O
O
C10H12O4
C11H12O4
C16H22O4
C13H18O
C17H18O2
C20H24O3
9
G
B
B
10
11
12
13
14
1-Benzyl methyl ether-2-acetophenone-ethane
C18H20O2
0.077
n.d.
n.d.
0.47
a UV light (100 mW cm−2) at 5 °C. b In the dark at 50 °C. Reaction conditions: alkali lignin 100 mg, [BMim][NTf2] 1000 mg, H2SO4 1 wt%, temp-
erature 50 °C, UV light (100 mW cm−2), reaction time 40 min, atmosphere. c Selectivity in photo-thermal oil products.
separated and identified as the degradation products after the
alkali lignin is treated under UV light irradiation and heating
conditions in [BMim][NTf2]. The observation of carbonyl func-
tionality on the Cα of the products is a sign of Cα–Cβ cleavage,
suggesting lignin side-chain oxidation through the photo-
thermal degradation method. The H, G and S type products, as
well as other benzene-based aromatics, which are derived from
the lignin corresponding internal structure were also detected
in the extracted liquid oil, further demonstrating the chemical
bond cleavage among alkali lignin subunits. In sharp compari-
son, when we conducted the degradation by just applying UV
light irradiation (5 °C, 100 mW cm−2) or heating (50 °C, in the
dark), only a small amount of aromatics were obtained. The
Fig. 8 FT-IR spectra of alkali lignin and residue lignin. Reaction con-
phenomenon signified that the acquisition of most products
ditions were the same as those in Table 4.
in lignin photo-thermal degradation is dependent on the
synergism of UV light and heat, which could cause the C–C or
C–O bond cleavage in lignin interunit linkages and release aro-
matic products through a green and mild pathway.
fied the photo-thermal degradation process of alkali lignin as
The output of all monomers and dimers detected by GC-MS presented in Fig. 8. It could be concluded from the spectra
accounts for 8.22 wt% of the total oily product through that both original and regenerated samples bear the character-
Table 4, demonstrating that there are more products present istic absorption peaks of lignin. The absorption bands at 1612,
in the oil which could not be identified yet. For a more 1515, 1463 and 1426 cm−1 demonstrate similar core aromatic
detailed understanding of the oil composition, ESI analysis structures34 (Table S3†). While the differences between the two
was conducted. As shown in Fig. S6,† not only the signals in FT-IR spectra indicate that the chemical structure changes
the range of m/z = 0–400 related to the products with low mole- after photo-thermal treatment exerted on alkali lignin in the
cular weights were detected, but also those in the m/z = [BMim][NTf2] system. The signal observed at 1031 cm−1 is
400–1200 range appeared as anticipated, implying that the oil attributed to the vibration between C–H and C–O in the plane
products contain an abundance of compounds with higher of the G unit,45 which almost disappeared in the spectrum of
molecular weights like trimers and tetramers.
the regenerated lignin, thus indicating a decrease of G unit
The chemical structure comparison of original and regener- during the photo-thermal degradation. Meanwhile, the signal
ated lignin was performed by FT-IR analysis, which also veri- at 1123 cm−1 corresponds to the aromatic C–H in-plane defor-
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 5524–5534 | 5531

View Article Online
Paper
Green Chemistry
Fig. 9 HSQC NMR spectra [ppm] of alkali lignin and residue lignin. Reaction conditions were the same as those of Table 4.
mation of syringyl type,46 while the intensity of the signal is regenerated lignin sample, implying that these linkages,
lower for the regenerated lignin compared to the spectrum of including the β-O-4, β-5 and β–β bonds, diminished after
the original one, suggesting the removal of part of the S unit degradation of lignin by the photo-thermal method in
from alkali lignin under photo-thermal conditions in this [BMim][NTf2]. Meanwhile, the signals belonging to the H
work. Additionally, the peak at 1155 cm−1 represents the C–O lignin unit of C2,6–H2,6 at δC/δH 128.5/7.0 ppm and C3,5–H3,5 at
stretching in ester groups, which is not observed in the spec- δC/δH 116.1/6.8 ppm, and G lignin unit of C2–H2 at δC/δH 110.1/
trum of the regenerated lignin and revealed the cleavage of the 7.2 ppm and C5,6–H5,6 at δC/δH 115.1/6.7 ppm along with the S
C–O bond upon photo-thermal degradation of alkali lignin.45 lignin unit for C2,6–H2,6 at δC/δH 105.4/6.5 ppm appeared in
The differences mentioned above revealed that alkali lignin the spectrum of original lignin, whereas most signals men-
underwent chemical structure changes upon photo-thermal tioned above depleted and only the peak for C5,6–H5,6 in the G
treatment, which might be involved in the fragmentation of unit was observed in regenerated lignin, indicating the
lignin interunit linkages under UV light irradiation and removal of H, S or G units from lignin after photo-thermal
heating, thus yielding various types of aromatics, as those dis- treatment. The sharp contrast in the HSQC spectra between
played in Table 4.
the original and regenerated lignin samples further elucidated
HSQC analysis was also conducted to obtain structural that the synergism between UV light and heat contributes to
information of lignin samples in detail, according to which lignin degradation by fragmenting lignin interunit linkages.
the efficiency of the photo-thermal method in [BMim][NTf2] to
degrade lignin was further evaluated. As shown in Fig. 9, the
signals appearing in δC/δH 50–95/2.5–6 ppm and δC/δH
100–130/6–7.25 ppm are related to lignin aliphatic and aro-
Conclusions
matic regions, respectively.21 All lignin samples showed strong In summary, we have demonstrated that the synergism of UV
signals of aryl methoxy groups at δC/δH 56.3/3.8 ppm. While light and heating can trigger the degradation of lignin,
many differences could be observed between the spectra of especially the Cα–Cβ bond cleavage in lignin β-O-4 and β-1
original and regenerated lignin, revealing the changes in the interlinkages in the ionic liquid [BMim][NTf2] under metal-
lignin structure before and after photo-thermal degradation. free and mild conditions. Compared to the photo or thermal
The HSQC spectrum of original lignin contained the cross alone route, the photo-thermal strategy has plenty of promis-
signals which corresponded to Cα–Hα in A (δC/δH 72.3/4.9 ppm) ing features, such as environmental compatibility, energy
and B (δC/δH 87.8/5.5 ppm), and Cβ–Hβ in A (δC/δH 86.5/ saving and easy operation. Although switching from photo/
4.2 ppm) and B (δC/δH 54.7/3.6 ppm), and Cγ–Hγ in A (δC/δH thermal route to the photo-thermal strategy is a non-tra-
60.5/3.6 ppm), B (δC/δH 62.1/3.8 ppm) and C (δC/δH 70.3/ ditional approach, much attention has been paid to photo-
3.5 ppm). In contrast, these signals almost disappeared in the thermal reactions to pursue superior performance. As pre-
5532 | Green Chem., 2021, 23, 5524–5534
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
sented in this work, the degradation of lignin model com-
pounds and alkali lignin can take place in the ionic liquid
Conflicts of interest
[BMim][NTf2] by the photo-thermal synergism to aid the clea- There are no conflicts to declare.
vage of inert chemical bonds like the C–C bond under mild
conditions (50 °C, atmosphere) without any metal-based cata-
lysts present, but the product yield is rather low. Further
improvements such as learning how to minimize side reac-
Acknowledgements
tions and how to suppress the recondensation of products are This work was financially supported by the National Key R&D
needed to realize more effective and greener biomass valoriza- Program of China (No. 2021YFE0190800), the Science Fund for
tion. Our results do point out that photo-thermal methods Creative Research Groups of the National Natural Science
have a unique synergism of light and heat for inducing inert Foundation of China (No. 21921005), the Strategic Priority
chemical bond cleavage, providing a green technology for dis- Research Program of the Chinese Academy of Science (No.
rupting this kind of inert bond under mild conditions and XDA21060300 and XDA24030504), the National Natural
thus demonstrating their potential to be applied as powerful Science Foundation of China (No. 21878292, 21878314 and
technologies for upgradation of macromolecular compounds. 21878302), the K. C. Wong Education Foundation (No.
Moreover, the unique characteristics of photo-thermal techno- GJTD-2018-04), the Joint fund of Innovation Academy for
logies will also open a new avenue for investigating eco- Green Manufacture (No. IAGM2020C12 and IAGM2020C21),
friendly and energy saving approaches for practical macromol- the
China
Postdoctoral
Science
Foundation
(No.
ecular substance conversion.
2019M660797), and the Chinese Academy of Sciences
President’s International Fellowship Initiative (Grant No.
122111WGZJTPYJY20180050).
Experimental
Photo-thermal degradation of lignin model compounds and
alkali lignin
References
A certain amount of reactants (lignin model compounds/alkali
lignin, ionic liquids, the Brønsted acid) were mixed in a quartz
reaction vial under stirring at 1000 rpm, which was exposed to
UV light irradiation/or kept in the dark at a desired tempera-
ture in the atmosphere/N2/O2. After the reaction ended, the
reactants were cooled to room temperature and mixed with 3 ×
2 mL methyl tertiary butyl ether (MTBE) to extract the
products.
1 X. Wu, S. Xie, C. Liu, C. Zhou, J. Lin, J. Kang, Q. Zhang,
Z. Wang and Y. Wang, ACS Catal., 2019, 9, 8443–8451.
2 S. Li, Z.-J. Li, H. Yu, M. R. Sytu, Y. Wang, D. Beeri,
W. Zheng, B. D. Sherman, C. G. Yoo and G. Leem, ACS
Energy Lett., 2019, 5, 777–784.
3 M. Rafiee, M. Alherech, S. D. Karlen and S. S. Stahl, J. Am.
Chem. Soc., 2019, 141, 15266–15276.
4 W. Lan, J. B. de Bueren and J. S. Luterbacher, Angew.
Chem., Int. Ed., 2019, 58, 2649–2654.
Calculation of conversion and product yield
5 L. D. Tolesa, B. S. Gupta, A. H. Tiwikrama, Y.-C. Wu and
M.-J. Lee, J. Clean. Prod., 2020, 258, 1–10.
Conversion ð%Þ ¼ ðm0 ꢀ mLigninÞ ꢁ 100=m0
Yield ð%Þ ¼ mproducts ꢁ 100=m0
ð1Þ
ð2Þ
6 C. E. J. J. Vriamont, T. Chen, C. Romain, P. Corbett,
P. Manageracharath, J. Peet, C. M. Conifer, J. P. Hallett and
G. J. P. Britovsek, ACS Catal., 2019, 9, 2345–2354.
7 H.-R. Tian, Y.-W. Liu, Z. Zhang, S.-M. Liu, T.-Y. Dang,
X.-H. Li, X.-W. Sun, Y. Lu and S.-X. Liu, Green Chem., 2020,
22, 248–255.
8 C. Zhang, H. Li, J. Lu, X. Zhang, K. E. MacArthur,
M. Heggen and F. Wang, ACS Catal., 2017, 7, 3419–3429.
9 H. Zhou, H. Wang, F. A. Perras, P. Naik, M. Pruski,
A. D. Sadow and I. I. Slowing, Green Chem., 2020, 22, 4676–
4682.
where m0 represents the initial mass of lignin model com-
pounds/or alkali lignin used for the degradation, mLignin rep-
resents the mass of lignin model compounds/or alkali lignin
at sampling time, and mproducts is the mass of the obtained
products.
Reuse of [BMim][NTf2] in experiments
After the products were extracted by MTBE, the remaining reac-
tion mixture was washed 3 times with 5 mL of water and dried
in a vacuum at 70 °C for 24 h for next use.
10 W. Wang, M. Wang, X. Li, L. Cai, S. Q. Shi, C. Duan and
Y. Ni, ACS Sustainable Chem. Eng., 2019, 8, 38–43.
11 H. Liu, H. Li, N. Luo and F. Wang, ACS Catal., 2019, 10,
632–643.
Regeneration of alkali lignin
After the extraction of products by MTBE, the remaining reac-
tion mixture was mixed with 50 mL of ethanol/water solution 12 G. Magallanes, M. D. Kärkäs, I. Bosque, S. Lee,
(volume ratio of 1 : 1) and centrifuged. The above steps were
repeated twice to obtain the alkali lignin residue, which was
S. Maldonado and C. R. J. Stephenson, ACS Catal., 2019, 9,
2252–2260.
subsequently dried in a vacuum at 70 °C for 24 h for further 13 X. Shen, Y. Xin, H. Liu and B. Han, ChemSusChem, 2020,
characterization.
13, 4367–4381.
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 5524–5534 | 5533

View Article Online
Paper
Green Chemistry
14 M. Wang, J. Lu, X. Zhang, L. Li, H. Li, N. Luo and F. Wang, 30 D. N.-S. Hon and W. C. Feist, Wood Fiber Sci., 1992, 24,
ACS Catal., 2016, 6, 6086–6090. 448–455.
15 X. Yu, Z. Wei, Z. Lu, H. Pei and H. Wang, Bioresour. 31 A. Temiz, U. C. Yildiz, I. Aydin, M. Eikenes, G. Alfredsen
Technol., 2019, 291, 1–10.
and G. Çolakoglu, Appl. Surf. Sci., 2005, 250, 35–42.
32 Y. Yang, H. Fan, Q. Meng, Z. Zhang, G. Yang and B. Han,
Chem. Commun., 2017, 53, 8850–8853.
33 Y. Zhang, H. He, K. Dong, M. Fan and S. Zhang, RSC Adv.,
2017, 7, 12670–12681.
34 Y. Kang, Y. Yang, X. Yao, Y. Liu, X. Ji, J. Xin, J. Xu, H. Dong,
D. Yan, H. He and X. Lu, ChemSusChem, 2020, 13, 5945–
5953.
16 C. Zhang and F. Wang, Acc. Chem. Res., 2020, 53, 470–484.
17 A. Rahimi, A. Azarpira, H. Kim, J. Ralph and S. S. Stahl,
J. Am. Chem. Soc., 2013, 135, 6415–6418.
18 G. Han, T. Yan, W. Zhang, Y. C. Zhang, D. Y. Lee, Z. Cao
and Y. Sun, ACS Catal., 2019, 9, 11341–11349.
19 N. Luo, M. Wang, H. Li, J. Zhang, H. Liu and F. Wang, ACS
Catal., 2016, 6, 7716–7721.
20 S. Mukundan, L. Atanda and J. Beltramini, Sustain. Energy 35 B. J. Cox, S. Jia, Z. C. Zhang and J. G. Ekerdt, Polym. Degrad.
Fuels, 2019, 3, 1317–1328. Stab., 2011, 96, 426–431.
21 M. Davaritouchaee, W. C. Hiscox, E. Terrell, R. J. Mancini 36 M. Mu, J. Cheng, C. Dai, N. Liu, Z. Lei, Y. Ding and J. Lu,
and S. Chen, Green Chem., 2020, 22, 1182–1197. Green Energy Environ., 2019, 4, 190–197.
22 E. Ota, H. Wang, N. L. Frye and R. R. Knowles, J. Am. Chem. 37 S. K. Singh and P. L. Dhepe, Green Chem., 2016, 18, 4098–
Soc., 2019, 141, 1457–1462. 4108.
23 Z. Xiang, W. Han, J. Deng, W. Zhu, Y. Zhang and H. Wang, 38 A. Kaiho, M. Kogo, R. Sakai, K. Saito and T. Watanabe,
ChemSusChem, 2020, 4199–4213. Green Chem., 2015, 17, 2780–2783.
24 P. Christopher, H. Xin, A. Marimuthu and S. Linic, Nat. 39 T. Yokoyama, J. Wood Chem. Technol., 2014, 35, 27–42.
Mater., 2012, 11, 1044–1050.
40 A. TKÁC and R. STOVÍCEK, Chem. Pap., 1993, 47, 362–
25 V. Nair, M. J. Muñoz-Batista, M. Fernández-García,
368.
R. Luque and J. C. Colmenares, ChemSusChem, 2019, 12, 41 H. Chen, Y. Wang, P. Cao and Y. Liu, Food Chem., 2017,
2098–2116. 234, 439–444.
26 L. Gao, X. Cui, Z. Wang, C. D. Sewell, Z. Li, S. Liang, 42 F. Gugumus, Polym. Degrad. Stab., 1999, 65, 259–269.
M. Zhang, J. Li, Y. Hu and Z. Lin, Proc. Natl. Acad. 43 G. F. De Gregorio, C. C. Weber, J. Gräsvik, T. Welton,
Sci. U. S. A., 2021, 118, 1–10.
A. Brandt and J. P. Hallett, Green Chem., 2016, 18, 5456–
27 X. Guo, C. Hao, G. Jin, H. Y. Zhu and X. Y. Guo, Angew.
Chem., Int. Ed., 2014, 53, 1973–1977.
5465.
44 P. P. Manning, J. Am. Chem. Soc., 1957, 79, 5151–
28 L. Kang, X. Y. Liu, A. Wang, L. Li, Y. Ren, X. Li, X. Pan,
5153.
Y. Li, X. Zong, H. Liu, A. I. Frenkel and T. Zhang, Angew. 45 X. Tan, Q. Zhang, W. Wang, X. Zhuang, Y. Deng and
Chem., Int. Ed., 2020, 59, 12909–12916. Z. Yuan, Fuel, 2019, 249, 334–340.
29 Y. Kang, X. Lu, G. Zhang, X. Yao, J. Xin, S. Yang, Y. Yang, 46 C. K. Nitsos, P. A. Lazaridis, A. Mach-Aigner, K. A. Matis
J. Xu, M. Feng and S. Zhang, ChemSusChem, 2019, 12,
4005–4013.
and K. S. Triantafyllidis, ChemSusChem, 2019, 12, 1179–
1195.
5534 | Green Chem., 2021, 23, 5524–5534
This journal is © The Royal Society of Chemistry 2021