Promoting Lignin Depolymerization and Restraining the Condensation via an Oxidation-Hydrogenation Strategy

Article abstract of DOI:10.1021/acscatal.7b00148

For lignin valorization, simultaneously achieving the efficient cleavage of ether bonds and restraining the condensation of the formed fragments represents a challenge thus far. Herein, we report a two-step oxidation-hydrogenation strategy to achieve this goal. In the oxidation step, the O₂/NaNO₂/DDQ/NHPI system selectively oxidizes C_αH-OH to C_α=O within the β-O-4 structure. In the subsequent hydrogenation step, the α-O-4 and the preoxidized β-O-4 structures are further hydrogenated over a NiMo sulfide catalyst, leading to the cleavage of C_β-OPh and C_α-OPh bonds. Besides the transformation of lignin model compounds, the yield of phenolic monomers from birch wood is up to 32% by using this two-step strategy. The preoxidation of C_αH-OH to C_α=O not only weakens the C_β-OPh ether bond but also avoids the condensation reactions caused by the presence of C_α⁺ from dehydroxylation of C_αH-OH. Furthermore, the NiMo sulfide prefers to catalyze the hydrogenative cleavage of the C_β-OPh bond connecting with a C_α=O rather than catalyze the hydrogenation of C_α=O back to the original C_αH-OH, which further ensures and utilizes the advantages of preoxidation.

Full text of DOI:10.1021/acscatal.7b00148

Subscriber access provided by Fudan University
Article
Promoting Lignin Depolymerization and Restraining the
Condensation via an Oxidation-Hydrogenation Strategy
Chaofeng Zhang, Hongji Li, Jianmin Lu, Xiaochen Zhang,
Katherine E MacArthur, Marc Heggen, and Feng Wang
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.7b00148 • Publication Date (Web): 06 Apr 2017
Downloaded from http://pubs.acs.org on April 6, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 36
ACS Catalysis
1
2
3
4
5
Promoting Lignin Depolymerization and Restraining the
6
7
Condensation via an OxidationꢀHydrogenation Strategy
8
9
Chaofeng Zhang,a Hongji Li,a,b Jianmin Lu,a Xiaochen Zhang,a Katherine E.
MacArthur,c Marc Heggenc and Feng Wanga*
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
State Key Laboratory of Catalysis, Dalian National Laboratory for Clean Energy,
Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023
(China)
b University of Chinese Academy of Sciences, Beijing 100049 (China)
c
Ernst Ruska Centre for Microscopy and Spectroscopy with Electrons and Peter
Grünberg Institute, Forschungszentrum Juelich GmbH, Juelich 52425 (Germany)
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 36
1
2
3
4
5
ABSTRACT:
6
7
8
9
For lignin valorization, simultaneously achieving the efficient cleavage of ether bonds
and restraining the condensation of the formed fragments represents a challenge thus
far. Herein, we report a twoꢀstep oxidationꢀhydrogenation strategy to achieve this goal.
In the oxidation step, the O2/NaNO2/DDQ/NHPI system selectively oxidizes CαH–OH
to Cα=O within βꢀOꢀ4 structure. In the subsequent hydrogenation step, the αꢀOꢀ4 and
the preꢀoxidized βꢀOꢀ4 structures are further hydrogenated over a NiMo sulfide
catalyst, leading to the cleavage of Cβ–OPh and Cα–OPh bonds. Besides the
transformation of lignin model compounds, the yield of phenolic monomers from
birch wood is up to 32% by using this twoꢀstep strategy. The preꢀoxidation of CαH–
OH to Cα=O not only weakens the Cβ–OPh ether bond but also avoids the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
+
condensation reactions caused by the presence of Cα from dehydroxylation of CαH–
OH. Furthermore, the NiMo sulfide prefers to catalyze the hydrogenative cleavage of
the Cβ–OPh bond connecting with a Cα=O rather than catalyze the hydrogenation of
Cα=O back to original CαH–OH, which further ensures and utilizes the advantages of
preꢀoxidation.
KEYWORDS:
Lignin, oxidation−hydrogenation, phenolic monomer, NaNO2/DDQ/NHPI, NiMo
sulfide, lignin fragments condensation
ACS Paragon Plus Environment

Page 3 of 36
ACS Catalysis
1
2
3
4
5
GRAPHIC ABSTRACT:
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 36
1
2
3
4
5
1. INTRODUCTION
6
7
8
9
The increasing awareness of dwindling fossil resources and the everꢀincreasing need
for more aromatic chemical feedstock has led to extensive research into efficient
methods to meet this demand. In numerous alternative solutions, lignin has received a
great amount of attention, owing to its potential to produce functionalized aromatic
fine chemicals.1 The key issues for the lignin transformation lie in the development of
highly selective and active catalysts to effectively cleave the ubiquitous ether bonds,
whilst leaving the aromatic benzene rings unconverted.2 The condensation of
generated carbocation intermediates3 and aldehyde products4 should also be avoided.
It still remains a big challenge for converting lignin models and extracts, albeit
original lignin.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In recent years, a lot of works have been directed towards the depolymerization
of native lignin residing in wood. Based on the analysis of lignin conversion, products
distributions and lignin structure change, containing condensation and rearrangement,
this approach termed “ligninꢀfirst" has provide much useful guidance in lignin
valorization.5 Besides the “ligninꢀfirst" approach, another approach termed
“bottomꢀup" focuses on the mechanism research of the transformation of lignin
models and then applies the methods in the lignin conversion. Although a vast
difference exists between studies on model compounds and studies on technical/native
lignin because of the complex structure of lignin, the “bottomꢀup" approach has
presented many effective catalytic systems and strategies in the technical/native lignin
conversion.6 Nevertheless, each approach has its own advantages and limitations.
ACS Paragon Plus Environment

Page 5 of 36
ACS Catalysis
1
2
3
4
5
6
Hence, both domains should be complementary in founding efficient methods in
lignin conversion.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Representative structure of a fragment of lignin (A) and the lignin model
compounds (B) used in this study.
The βꢀOꢀ4 linkage (Figure 1) accounts for 40ꢀ60% of interunit linkages in lignin.
It is the primary target structure to be broken during lignin depolymerization.1a In
addition, αꢀOꢀ4 and 4ꢀOꢀ5 linkages account for about 10% and 5% of interunit
linkages. Methods including hydrogenation,2, 7 transfer hydrogenation,8 reduction,9
oxidation,10 acidolysis,4, 11 alkaline hydrolysis,12 alcoholysis13 and redoxꢀneutral14
have been used in the transformation of these lignin linkages.
Recently, a twoꢀstep strategy has been explored in the transformation of lignin
and model compounds. Given the fact that the bond dissociation enthalpy (BDE) of
Cβ–OPh bond of βꢀOꢀ4ꢀketones is about 40~50 kJ·mol−1 lower than Cβ–OPh bond of
βꢀOꢀ4ꢀalcohols,15 and the βꢀOꢀ4ꢀketones contain active Cβ–H and Cα–Cβ bonds for
further transformation, the CαH–OH of βꢀOꢀ4ꢀalcohols is usually preꢀconverted to
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 36
1
2
3
4
5
6
7
8
9
Cα=O by using wellꢀestablished benzyl alcohol oxidation methods. In the second step
of converting βꢀOꢀ4ꢀketones, uncatalyzed routes or reduction methods are usually
used primarily to avoid the backꢀconversion to βꢀOꢀ4ꢀalcohols. Stahl and
coꢀworkers16 found O2/AcNHꢀTEMPO/HCl/HNO3 system can realize chemoselective
aerobic oxidation of secondary benzylic alcohols within lignin model compounds to
ketones. In the subsequent transformation of preꢀoxidized βꢀOꢀ4 linkages, they
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
16
achieved the selective Cα–Cβ bond cleavage with KOH/H2O2 and Cβ–OPh bond
cleavage6a in presence of formic acid. Stephenson group17 identified that the
photocatalyst [Ir(ppy)2(dtbbpy)]PF6 can achieve the cleavage of Cβ−OPh bond within
βꢀOꢀ4ꢀketones, and the generated fragment can be reduced by HCOOH. Westwood
and coꢀworkers6c have reported that the DDQ/tBuONO/O2 system could promote the
chemoselective oxidation of CαH–OH to Cα=O within βꢀOꢀ4 linkages, and the
preꢀoxidized βꢀOꢀ4 linkages could be cleaved by Zn/NH4Cl. Besides the first
oxidation or dehydrogenation of CαH–OH to Cα=O, Barta and coꢀworkers have
reported a series of strategies combining acidolysis with etherification, hydrogenation,
or decarbonylation reactions.4a
We hold the same viewpoint that the twoꢀstep methodology is suitable for
dissociating interlinking lignin units, but we think the known methods have problems
of focusing too much on the first step but understating the following ether conversion.
And the reported methods mainly focus on the transformation of βꢀOꢀ4 linkage. To
improve the efficiency of lignin conversion, the transformation of other ether linkages
should be also taken into consideration.
ACS Paragon Plus Environment

Page 7 of 36
ACS Catalysis
1
2
3
4
5
6
7
8
9
Homogeneous catalysts with wellꢀdefined structures have been used in catalytic
hydrogenation and transfer hydrogenation cleavage of different C–O–C ether bonds
and C–C bonds. However, in converting original lignin, the homogeneous systems
face all or partial issues as listed in the follows. (1) A contradiction exists between the
specificity of homogeneous systems and the complex structure of lignin. (2) The
competitive coordination to metal active center between ligands and
oxygenꢀcontaining groups of lignin. (3) Depolymerizing the stable 3D structure in
solid original lignin into lowꢀmoleculeꢀweight fragments for further transformation is
still a challenge for homogeneous systems operated under mild condition. Although
some nonꢀnoble heterogeneous catalysts do not have such problems and thus have
drawn much attention in lignin transformation, the harsh reaction conditions make it
difficult to control the reaction selectivity and therefore the obvious transformation of
aromatic rings usually occurs along with the cleavage of target ether bonds.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Another key issue is the possible condensation of lignin fragments at the
existence of acid sites or H+ species during lignin pretreatment and/or
transformation at high temperature. As shown in Figure 2, three main
condensation processes may happen during the transformation of lignin βꢀOꢀ4
linkages. (
A
) The aromatic aldehydes generated from the acidolysis of βꢀOꢀ4
+
structure with a CαH–OH causes the condensation.4 (
B
) The Cα of βꢀOꢀ4
carbocation attacks the adjacent aromatic rings with G or H type and generates
a benzofuran structure after the formation of new C–C bond.4b, 11a
(C) The
generated βꢀOꢀ4 carbocation intermediate condensates with another G or S type
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 36
1
2
3
4
of aromatic ring.3 Based on the analysis of these processes, the generation of
5
6
7
βꢀOꢀ4 carbocation (1 in Figure 2) is the precondition for the condensation, and
8
9
is usually generated from CαH–OH of βꢀOꢀ4 linkage catalyzed by protonic
acid,4, 11a acid sites on the heterogeneous catalysts,3b or the acid solvents.11b, 11c
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Lignin
CHO
Lignin fragment condensation
caused by the carbocation
(A-C)
O
H+ / H2O
OH
- H+
(OMe)
Condensation
- H2CO
OMe
(MeO)
A
O
Lignin
-
4
(Ref 4 )
OMe
Lignin
O
O
Lignin
2
3
Lignin
Lignin
O
HO
HO
OH
OH
B
MeO
OMe
OMe
O
O
MeO
H+
(Ref 4b,11a)
H
- H+
(MeO)
(MeO)
OH
O
O
O
O
- H2O
(MeO)
Lignin
(MeO)
Lignin
H/G
Lignin
Lignin
Lignin
1
5
6
Lignin
G/S
Lignin
Lignin
G*/S*
O
Lignin
Lignin O
OMe
Lignin
O
OMe
Lignin
O
(MeO)
HO
OMe
MeO
O
MeO
O
- H+
O
(MeO)
Lignin
(MeO)
O
O
(MeO)
OMe
(MeO)
OH
(MeO)
C
OH
OH
O
HO
HO
7
Lignin
(Ref 3 )
8
9
Lignin
Lignin
(This work, D and E )
HO
HO
O2/NaNO2/DDQ/NHPI
Weak acid site OH
D
HO
+
Further
transformation
O
Lignin
NiMo sulfide
H2
O
Lignin
HO
(Ref 7k )
X
O
(MeO)
(MeO)
F
Lignin
O
OMe
12
11
NiMo sulfide
H2
Lignin
O
OMe
O
Lignin
O
Further
transformation
(MeO)
E
HO
Lignin
Phenolic
monomers
O
OMe
+
Lignin
10
Cleavage of -O-4, -O-4
and benzyl ether linkages
(MeO)
Lignin
O
OMe
13
3
Selective oxidation-hydrogenation strategy
Figure 2. The possible condensation processes (AꢀC) during the transformation of
lignin βꢀOꢀ4 linkages and our selective oxidationꢀhydrogenation strategy (D and E).
Condensation caused by the generation of aromatic aldehydes (
Condensation in the same βꢀOꢀ4 linkage with a carbocation intermediate
11a). Condensation caused by the condensation of βꢀOꢀ4 carbocation intermediate
and G/S type of aromatic ring ( , Ref 3). Hydrogenolysis of βꢀOꢀ4 alcohol over the
NiMo sulfide via an ArCHδ+CH(CH2OH)OAr intermediate (
, Ref 7k). The H, G and
A, Ref 4).
1 (B, Ref 4b,
1
C
F
S stand for the
p
ꢀhydroxyphenyl (H)ꢀ, guaiacyl (G)ꢀ and syringyl (S)ꢀtype of lignin
ACS Paragon Plus Environment

Page 9 of 36
ACS Catalysis
1
2
3
4
aromatic ring.
5
6
7
8
9
To reduce or avoid condensation, guiding the transformation of βꢀOꢀ4
carbocation or avoiding the generation of βꢀOꢀ4 carbocation are both crucial.
Barta and coꢀworkers4a have reported a triflic acidꢀcatalyzed method, which is
based on the rapid transformation of inꢀsitu generated substituted
phenylacetaldehydes, to achieve the depolymerization of lignin βꢀOꢀ4 linkages
with low condensation products. Luterbacher and coꢀworkers3a achieved the
goal by etherifying the CαH–OH and blocking the electronꢀrich positions at the
positions ortho or para to methoxyl in G or S type of aromatic ring with
formaldehyde. Furthermore, inspired by the twoꢀstep methods containing first
conversion of CαH–OH to Cα=O in βꢀOꢀ4 model compounds, the combination
of selective oxidation and subsequent heterogeneous hydrogenation also should
be an effective strategy.6b For the advantages of this oxidationꢀhydrogenation
strategy, the first selective transformation of CαH–OH to Cα=O can not only
promote the cleavage of Cβ–OPh in the hydrogenation process but also prevent
the generation of carbocation from CαH–OH.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Although transformation of lignin into aromatic compounds via a twoꢀstep
strategy, including a first oxidation step to preꢀoxidize lignin followed by a
subsequent depolymerization is not entirely new, the heterogeneous hydrogenolysis
of preꢀoxidized lignin structure using H2 as hydrogen donor still faces some
difficulties. An appropriate heterogeneous hydrogenation catalyst should have
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 36
1
2
3
4
5
6
7
8
9
the following features. (1) The catalyst prefers to catalyze the cleavage of
βꢀOꢀ4 ether bonds rather than the hydrogenation of aromatic rings and carbonyl
groups. (2) The catalyst should have low activity to generate carbocation from
substrates containing a CαH–OH structure. (3) Besides the βꢀOꢀ4 linkage, the
other ether linkages can be also transformed. (4) For the practical lignin
conversion, a nonꢀnoble metal catalyst is favorable.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Following the above criteria, we herein used a biomimetic organocatalyst
system O2/NaNO2/DDQ/NHPI to oxidize lignin CαH–OH to Cα=O within βꢀOꢀ4
linkages at the first step. In the subsequent hydrogenation of Cβ–OPh ether
bonds, we fortunately found the asꢀprepared NiMo sulfide catalyst can achieve
the selectively hydrogenative cleavage of Cβ–OPh ether bonds with high yield
of aromatic rings. Interestingly, αꢀOꢀ4 linkage and benzyl ether linkage, which
are far less studied in previous works, can also be transformed in the
hydrogenation process. Furthermore, we also did some studies towards the
identification and rationalization of heterogeneous catalysts for the
hydrogenolysis of preꢀoxidized lignin model structure.
ACS Paragon Plus Environment

Page 11 of 36
ACS Catalysis
1
2
3
4
5
2. EXPERIMENTAL SECTION
6
7
2.1 Materials and lignin model compounds synthesis
8
9
All chemicals are of analytical grade and used without further purification.
Dibenzyl ether, benzyloxybenzene and diphenyl ether are purchased from
Aladdin Chemicals Co., Ltd.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The βꢀOꢀ4 lignin model compounds including 2ꢀphenoxyꢀ1ꢀphenylethanone
(βꢀOꢀ4ꢀK),
other
βꢀOꢀ4ꢀketones
without
CγH–OH,18
2ꢀphenoxyꢀ1ꢀphenylethanol (βꢀOꢀ4ꢀA), other βꢀOꢀAꢀalcohols without CγH–
OH,18 (2ꢀphenoxyethyl)benzene (βꢀOꢀ4ꢀH),7e (2ꢀphenoxyꢀvinyl)benzene
(βꢀOꢀ4ꢀO),11a
(1ꢀmethoxyꢀ2ꢀphenoxyethyl)
benzene
(βꢀOꢀ4ꢀE)8a,
2ꢀmethylꢀ2ꢀphenoxyꢀ1ꢀphenylpropanꢀ1ꢀone (βꢀOꢀ4ꢀKꢀ2βCH3),8e polymeric
βꢀOꢀ4 lignin model (βꢀOꢀ4ꢀAꢀPolymer and βꢀOꢀ4ꢀKꢀPolymer),14e
βꢀOꢀ4ꢀketones and βꢀOꢀAꢀalcohols with CγH–OH structure13a are synthesized
according to the previous reports. The detailed synthesis procedures of βꢀOꢀ4
compounds are provided in the Supporting Information.
2.2 Catalysts preparation
The NiMo sulfide catalyst was prepared according to the previously
reported procedure.7k The (NH4)2MoS4 (5.2 g, 20 mmol) was dispersed in a
nickel chloride solution (200 mL, 20 mmol). After stirring at 100 °C for 6 h, the
solid was collected by filtration, washed with distilled water and ethanol, dried
at 60 °C under vacuum, followed by calcination in a horizontal furnace at
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 36
1
2
3
4
1
1
450 °C for 6 h in Ar flow (50 mL·min− ) at a 5 °C·min− heating rate from
5
6
7
25 °C.
8
9
The MoS2 was obtained by treating the asꢀprepared (NH4)2MoS4 in a
horizontal furnace at 450 °C for 6 h in an Ar flow. NiS was synthesized
according to the previous work.19 Ni/C (10 w%) was prepared by the
impregnation method with Ni(NO3)2 as the precursor, which was further
reduced by hydrogen at 400 °C for 2 h.7e The Pd/C (5 wt%) was prepared by a
impregnation method with PdCl2 as the precursor, which was preꢀreduced at
300 °C in H2 flow for 3 h before using.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.3 Characterizations
Xꢀray diffraction (XRD) characterizations were conducted on a Rigaku
D/Max 3400 powder diffraction system with Cu Kα radiation (λ=1.542 Å). The
electron microscopy analysis was carried out on the FEI Titan ChemiSTEM at
the Ernst Ruska Center Jülich, Germany, equipped with the high solidꢀangle
four quadrant SUPERꢀX energy dispersive Xꢀray (EDX) detector and operating
at 200 kV.
2.4 The density functional theory calculations
We have performed density functional theory (DFT) calculations using the
Vienna ab initio simulation package (VASP),20 which is a periodic planewave
density functional theory implementation. The PAW method,21 a frozenꢀcore
allꢀelectron method using the exact shape of the valence wave functions instead
of pseudowave functions, was chosen to describe the electron−ion interactions.
ACS Paragon Plus Environment

Page 13 of 36
ACS Catalysis
1
2
3
4
5
6
7
8
9
The exchange correlation energy has been calculated within the generalized
gradient approximation (GGA) using the PBE functional formulation.22 An
energy cutoff for plane waves of 400 eV was employed throughout this study.
To include dispersion interactions, we used the DFTꢀD3 methodology.23 For
more detailed procedures, please refer to our previous work.24
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.5 Procedure for catalytic reactions of model compounds
In a typical selective oxidation experiment, substrate (0.15 mmol), NaNO2
(6 mg), DDQ (8 mg), NHPI (8 mg), CH3CN (2.5 mL) and a magnetic stirrer
were placed into a highꢀpressured reactor (15 mL). After flushing with O2, the
reactor was charged with 0.5 MPa of O2 and placed into a preheated red copper
mantle at the desired temperature with magnetic stirring.
In a typical hydrogenation experiment, substrate (0.2 mmol), catalyst (20
mg) and methanol (2.0 mL) were placed into a highꢀpressured reactor (15 mL).
After flushing with Ar for five times, the reactor was charged with 1.0 MPa of
H2 and placed into a preheated red copper mantle at the desired temperature
with magnetic stirring.
After reaction, the reactor was quenched to ambient temperature using
cooling water. The nꢀdodecane as the standard substance in ethanol solution
was added in the reaction mixture. After the filtration with a Teflon filter
membrane, the organic products are analyzed by GCꢀFID (Agilent 7890 A) and
GC/MS (GC: Agilent 7890 A, MS: Agilent 5975 C).
2.6 Procedure for birch powders transformation
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 36
1
2
3
4
5
6
7
8
9
Birch powder was preꢀtreated with methylbenzene in a Soxhlet extractor for
24 h to remove the soluble substance. The lignin content in birch powders was
determined by the TAPPI standard method to be 19 w%.13a In the preꢀoxidation
of birch powder, birch powder 100 mg, acetonitrile 5.0 mL, NaNO2 8 mg, DDQ
12 mg, NHPI 12 mg and a magnetic stirrer were placed into a highꢀpressured
reactor (15 mL). After flushing with O2 for five times, the reactor was charged
with 1.0 MPa of O2 and placed into a preheated red copper mantle at 90 °C for
4 h. The preꢀoxidative birch powder was collected by filtration, washed with
CH3CN, dried at 60 °C under vacuum. In the hydrogenation process, birch
powder or preꢀoxidative birch powder 50 mg, methanol 5.0 mL, NiMo sulfide
50 mg and a magnetic stirrer were placed into a highꢀpressured reactor (15 mL).
After flushing with H2 for five times, the reactor was charged with 1.0 MPa of
H2 and placed into a preheated red copper mantle 200 °C for 4 h. After the
reaction, the reactor was quenched to ambient temperature using cooling water.
The naphthaline as the standard substance in ethanol solution was added in the
reaction mixture, after the filtration with a Teflon filter membrane, the organic
products are analyzed by GCꢀFID (Agilent 7890 A) and GC/MS (GC: Agilent
7890 A, MS: Agilent 5975 C).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 15 of 36
ACS Catalysis
1
2
3
4
5
3. RESULTS AND DISCUSSION
6
7
3.1
The
difference
between
direct
hydrogenation
and
8
9
oxidationꢀhydrogenation of βꢀOꢀ4 lignin model compounds
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The goal of this study is to build an efficient strategy using maneuverable
and nonꢀnoble hydrogenation catalysts to achieve the selective cleavage of
lignin
ether
bonds.
2ꢀPhenoxyꢀ1ꢀphenylethanol
(βꢀOꢀ4ꢀA),
2ꢀphenoxyꢀ1ꢀphenylethanone (βꢀOꢀ4ꢀK) and their derivatives with or without
CγHꢀOH structure were used as βꢀOꢀ4 linkage model compounds. In addition,
benzyloxybenzene represents αꢀOꢀ4 linkage, and dibenzyl ether represents the
normal benzylic ether linkage.
In the direct hydrogenation of βꢀOꢀ4ꢀA (Chart 1A), Pd/C offered complete
conversion (Entry 1), but the selectivity to products with Cβ–OPh bond
cleavage was only 11%. Most of the products were generated by the
hydrogenolysis of CαH–OH (67%) and hydrogenation of aromatic rings (22%).
Our previous work7e found that in comparison with Pd/C, Ni/C catalyst with
mild hydrogenation activity could not only promote the cleavage of Cβ–OPh
bond but also increase the yield of aromatic products. But Ni/C catalyst has a
drawback that the Ni particles easily get oxidized when being handled in air or
during reaction, which will lead to worse conversion of βꢀOꢀ4ꢀA (Chart 1A,
Entry 2).
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 36
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(A)
The hydrogenation of
β
-O-4-A
a
1. Pd/C
2. Ni/C
3. NiMo
sulfide
b
4. MoS
2
c
d
5. NiS
P1 ~ P7
P8 ~P11
P12~P16
The hydrogenation of β-O-4-K
(B)
1. Pd/C
2. Ni/C
e
f
g
3. NiMo
sulfide
4. MoS
2
h
5. NiS
0
10 20 30 40 50 60 70 80 90 100
Yield (%)
Chart 1. The hydrogenative conversion of βꢀOꢀ4ꢀA (A) and βꢀOꢀ4ꢀK (B) over
various catalysts. Reaction conditions: substrate 0.2 mmol, catalyst 20 mg,
methanol 2.0 mL, 1.0 MPa of H2, 180 °C, 4 h (βꢀOꢀ4ꢀA) or 6 h (βꢀOꢀ4ꢀK).
Products distributions: (a) P8 67%; (b) P8 1%, P9 43%; (c) P8 35%, P9 11%; (d)
ACS Paragon Plus Environment

Page 17 of 36
ACS Catalysis
1
2
3
4
5
6
P9 82%; (e) P3 33%, P6 31%; (f) P11 26%; (g) P3 41%, P6 50%; (h) P8 18%, P9
3%, P11 7%.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In searching for a nonꢀnoble metal and stable catalyst, we found Moꢀbased
catalyst with moderate hydrogenation activity was a suitable candidate.25 Moꢀbased
catalysts preferred to cleave the C–O bond rather than hydrogenate the aromatic
rings.26 Particularly, MoS2 offered a 97% conversion without aromatic rings being
hydrogenated (Chart 1A, Entry 4). But products with Cβ–OPh bond cleavage were
only 51 %. MoS2 could also catalyze CαH–OH hydrogenolysis to CαH2 (βꢀOꢀ4ꢀH,
35%) and CαH–OH etherification to CαH–OCH3 (βꢀOꢀ4ꢀE, 11%). As shown in
Scheme S1, the two byꢀproducts can restrain the cleavage of Cβ–OPh bond. In
addition, we modified MoS2 catalyst with Ni species, and the obtained NiMo sulfide
catalyst offered a > 99% conversion of βꢀOꢀ4ꢀA (Chart 1A, Entry 3) and slightly
increased the Cβ–OPh cleaved products to 56%. However, the CαH–OH bond
hydrogenolysis to CαH2 was suppressed (1%), while the CαH–OH etherification was
increased to 43%. To further reduce the etherification products, nꢀdodecane,
1,4ꢀdioxane, acetonitrile or Nꢀmethyl pyrrolidone was used as solvent, but low
conversions and high ratio of βꢀOꢀ4ꢀH were obtained (Scheme S2).
To achieve the selective cleavage of lignin ether bonds, one feasible strategy is to
firstly weaken the Cβ–OPh bond and then employ a catalytic system targeting on the
selective cleavage of target Cβ–OPh bond during the subsequent conversion. The
preꢀoxidization of CαH–OH to Cα=O can weaken the Cβ–OPh bond.6a, 16ꢀ17 So, as
discussed in the introduction, the key issue of oxidationꢀhydrogenation strategy lies in
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 36
1
2
3
4
5
6
7
8
9
the selectivity of the subsequent hydrogenation process. The most important feature of
catalyst is that the catalyst prefers to catalyze the cleavage of βꢀOꢀ4 ether bonds rather
than achieve the hydrogenation of aromatic rings and generated Cα=O back to CαH–
OH.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Then, βꢀOꢀ4ꢀK with Cα=O was used as the substrate to screen such a
hydrogenation catalyst. Except NiS offered a low conversion, Pd/C, Ni/C, MoS2 and
NiMo sulfide all gave > 99% conversions and increase in the yields of products with
Cβ–OPh bond cleavage, but their product distributions were quite different (Chart 1B).
The products with hydrogenated aromatic rings increased to 52% over Pd/C (Chart 1B,
Entry 1), and Pd/C could also catalyze the hydrogenation of Cα=O to βꢀOꢀ4ꢀA and
βꢀOꢀ4ꢀH (9%). Ni/C could catalyze the hydrogenation of βꢀOꢀ4ꢀK to βꢀOꢀ4ꢀA (26%).
When the hydrogenation of βꢀOꢀ4ꢀK was catalyzed by MoS2, the main products were
target aromatic products with Cβ–OPh bond cleavage (71%), but a part of βꢀOꢀ4ꢀK
was still transformed to βꢀOꢀ4ꢀA (7%), βꢀOꢀ4ꢀH (18%) and βꢀOꢀ4ꢀE (3%). We then
found that the NiMo sulfide catalyst was an appropriate catalyst for the hydrogenative
cleavage of Cβ–OPh bond within βꢀOꢀ4ꢀK. The main products were acetophenone and
phenol generated from Cβ–OPh bond cleavage. This confirmed the feasibility of
selective oxidationꢀhydrogenation strategy in the degradation of lignin and its model
compounds. We then focused on the establishment of catalytic system and research
about hydrogenation mechanism and factors on the reaction selectivity (vide infra).
3.2 Selective oxidation of lignin model compounds over the NaNO2/DDQ/NHPI
organocatalyst system
ACS Paragon Plus Environment

Page 19 of 36
ACS Catalysis
1
2
3
4
5
6
7
8
9
For the selective oxidation of targeted benzyl alcohols, referring to our previous
works about organocatalyst systems containing the combination of various quinones
and Nꢀhydroxyphthalimide (NHPI),27 we found that NaNO2/DDQ/NHPI
organocatalyst system could efficiently catalyze the selective oxidation of βꢀOꢀ4ꢀA to
βꢀOꢀ4ꢀK with O2. A 99% conversion and 95% selectivity to βꢀOꢀ4ꢀK was obtained in
2 h at 80 °C (Table 1, Entry 1). In the selective oxidation of βꢀOꢀ4ꢀA, each
composition of the NaNO2/DDQ/NHPI organocatalyst system was necessary (Table
S1, Entry 1ꢀ7). When 18O2 was used as the oxidant, no 18O labelled βꢀOꢀ4ꢀK was
checked (Table S1, Entry 9), which indicated an oxidative dehydrogenation process.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Proposed catalytic cycle for the oxidation of βꢀOꢀ4ꢀA to βꢀOꢀ4ꢀK
As reported in previous researches, the NOx generated from the NaNO2
decomposition can catalyze the hydroquinone oxidation to quinone with O2,27b and the
quinone can abstract H atom from NHPI and promote the generation of PINO
(phthalimideꢀNꢀoxyl) radical, which is confirmed by the UVꢀVis spectroscopy.27c
Herein, we propose the mechanism of the βꢀOꢀ4ꢀA oxidation (Scheme 1): (1) DDQ
abstracts a hydrogen atom from NHPI to generate a highly reactive PINO free radical
and hydroquinone (2HꢀDDQ); (2) PINO radicals obtain hydrogen atoms from
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 36
1
2
3
4
5
6
7
8
9
βꢀOꢀ4ꢀA to generate NHPI and βꢀOꢀ4ꢀK; (3) NO molecules generated from NaNO2
decomposition in the reaction media catalyze the oxidation of hydroquinone
(2HꢀDDQ) to quinone (DDQ) with O2.27d
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The O2/NaNO2/DDQ/NHPI catalytic system also showed excellent selectivity
and reactivity in the oxidation of the other benzyl alcohols (Table 1, 1bꢀg). The
expected βꢀOꢀ4ꢀketones were obtained in good yields (2bꢀg). The catalytic system
also showed excellent selectivity toward the oxidation of benzylic alcohol than the
hydroxymethyl group at Cγ (2e, 2g). While the intermediate or βꢀOꢀ4ꢀketone product
could lose a H2O to generate a Cβ=Cγ bond during the oxidation process (2eꢀ2, 2gꢀ2).
The O2/NaNO2/DDQ/NHPI catalytic system shows low activity in the selective
oxidation of αꢀOꢀ4 model compound.
Table 1. The selective oxidation of Cα–OH in lignin model compounds a
ACS Paragon Plus Environment

Page 21 of 36
ACS Catalysis
1
2
3
4
a
Reaction conditions: substrate 0.15 mmol, NaNO2 6 mg, DDQ 8 mg, NHPI 8 mg,
5
6
7
CH3CN 2.5 mL, O2 0.5 MPa, 80 °C, 2 h.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.3 Selectively hydrogenative cleavage of ether bonds in lignin model compounds
over the NiMo sulfide
Then, we checked the selective hydrogenation of βꢀOꢀ4ꢀketones and other ether
linkages over the NiMo sulfide catalyst. As shown in Table 2, > 97% conversions
were obtained in the hydrogenation of different βꢀOꢀ4ꢀketones, and no products with
hydrogenated aromatic ring was detected. NiMo sulfide selectively catalyzed the
hydrogenative cleavage of Cβ–OPh bond. And the ether bond of methoxyl group
connected with aromatic ring is unchanged. Given the fact that the βꢀOꢀ4 compound
with a CαH2 is difficult to be transformed over the NiMo sulfide (Scheme S1), the
products of ethylbenzene and its analogues should generate from the hydrogenation of
products like acetophenone or the fragments of βꢀOꢀ4ꢀketones. Furthermore, the
NiMo sulfide could also catalyze the hydrogenative cleavage of ether bonds of
dibenzyl ether and αꢀOꢀ4 model compound (benzyloxybenzene). No product was
generated in the hydrogenation of diphenyl ether.
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 36
1
2
3
4
Table 2. The selective hydrogenation of lignin model compounds a
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 23 of 36
ACS Catalysis
1
2
3
4
5
6
a Reaction conditions: substrate 0.2 mmol, NiMo sulfide 20 mg, methanol 2.0 mL, H2
1.0 MPa, 180 °C, 6 h.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Although the preꢀoxidation of CαH–OH to Cα=O can weaken the Cβ–OPh bond
and prevent the generation of βꢀOꢀ4 carbocation from dehydroxylation at Cα, given
the possibility that the Cα=O group of βꢀOꢀ4ꢀketones can be hydrogenated to CαH–OH
in the following hydrogenation (Figure 2,
D and E), the condensation problem may
arise again. However, the asꢀsynthesized NiMo sulfide prefers to catalyze the
cleavage of Cβ–OPh rather than hydrogenate the Cα=O group of βꢀOꢀ4ꢀketones (Table
2). Furthermore, as shown in our previous work,7k the weak and medium strong acid
sites (Figure S12) of NiMo sulfide cannot promote the generation of βꢀOꢀ4
+
carbocation with the active Cα structure for the condensation processes. As
experimental evidences, the acid sites of heterogeneous NiMo sulfide cannot
efficiently catalyze the acidolysis of βꢀOꢀ4ꢀA and βꢀOꢀ4ꢀO under Ar atmosphere
(Scheme S3). However, the NiMo sulfide can catalyze the further transformation of
βꢀOꢀ4ꢀA via an absorbed ArCHδ+CH2OAr intermediate under H2 atmosphere (Figure
2, F
).7k But the part of etherification products at Cα without cleavage of Cβ–OPh bond
(Chart 1A, Entry 3) restricts the direct hydrogenation of βꢀOꢀ4ꢀalcohols over the
NiMo sulfide. So, the twoꢀstep oxidationꢀhydrogenation is an efficient strategy, and
the NiMo sulfide is an appropriate nonꢀnoble metal hydrogenation catalyst to realize
this strategy.
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 36
1
2
3
4
5
6
7
8
9
3.4 Mechanism of βꢀOꢀ4ꢀK conversion over the NiMo sulfide
Next, we propose our understanding about the high selectivity of the
hydrogenation process over NiMo sulfide.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NiMo sulfide
(102)
(111)
∇
♦
(101)
*
∇
(100)
(110)
∇
(-220)
∇
∇
*
*
*
*
*
*
*
2
MoS
2
MoS
(002)
*
∇
NiS
Ni0.7Mo
♦
3
S
4
(100)
(101)
(110)
(112)
(103)
(105)
(114)
10
20
30
40
50
60
70
2 Theta (o)
Figure 3. XRD patterns of the NiMo sulfide and MoS2 catalysts.
As shown in Figure 3, MoS2 and NiS phase peaks exist in the XRD pattern of the
NiMo sulfide catalyst. Besides these peaks, there are also peaks attributed to
Ni0.7Mo3S4, which is one of the Ni doped heterometallic cubaneꢀtype Mo clusters.
Element mapping by EDX (Figure S13cꢀf) show that the Mo and S species are
uniformly distributed, and the Ni species appears aggregated to about 10 nm in some
zones, which corresponds to the fact that a NiS phase and Ni0.7Mo3S4 phase exist in
the XRD pattern of the NiMo sulfide catalyst.
ACS Paragon Plus Environment

Page 25 of 36
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. The routes for the βꢀOꢀ4ꢀK conversion over the NiMo sulfide catalyst
Given the intricate structure of heterogeneous NiMo sulfide catalyst (Figure 3)
and complex degradation process of simple βꢀOꢀ4ꢀK,8a, 8d, 8e, 14d, 14e, 18, 24, 28 it is an
efficient method to describe the structureꢀactivity relationship and factors on the
system selectivity by analyzing the transformation of the substrate.7k Regarding
acetophenone and phenol as the target products, herein, we propose the potential
routes of βꢀOꢀ4ꢀK transformation with detailed process of bonds cleavage (Scheme 2).
(Route 1) βꢀOꢀ4ꢀK adsorbs on the heterogeneous catalyst and transforms to activated
βꢀOꢀ4ꢀK with an activated Cα=O group, then cleavage of Cβ–OPh ether bond occurs
•
and the generated intermediates (PhCOCH2 and PhO•) are hydrogenated to PhCOCH3
8d
and PhOH.8a,
(Route 2) βꢀOꢀ4ꢀK loses an Hβ and transforms to adsorbing
intermediates (PhCOCH•• and PhO•) after the cleavage of Cβ–OPh bond, which are
14e, 24
subsequently hydrogenated to PhCOCH3 and PhOH.8e,
(Route 3ꢀ1) The
generation of adsorbing PhCOCH•OPh intermediate is promoted by a ketoꢀenol
tautomerism of βꢀOꢀ4ꢀK to PhC(OH)=CHOPh,24 then PhCOCH•OPh intermediate
transforms to PhCOCH3 and PhOH. (Route 3ꢀ2) The enol intermediate can also
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 36
1
2
3
4
5
6
7
8
9
transform to adsorbing PhC(OH)=CH• and PhO• species via the cleavage of C–OPh
bond, which are hydrogenated to PhCOCH3 and PhOH. (Route 3ꢀ3) The absorbing
enol intermediate (PhC(OH)=CHOPh) reacts with a H• species, and the generated
adsorbing PhC•(OH)CH2OPh transforms to PhC(OH)=CH2 and PhO• via a
dearyloxylation process.7k, 14a, 14c The PhC(OH)=CH2 transforms to PhCOCH3 via a
ketoꢀenol tautomerism, and PhO• is hydrogenated to PhOH.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To check the cleavage order of the corresponding bonds in βꢀOꢀ4ꢀK, the
deuterium labelling experiment and shielding experiment with methyl substituent
were carried out.8e Given the catalysis of MoS2 based catalysts in H/D exchange
reaction,29 the H or D atoms connecting with Cβ atom within βꢀOꢀ4ꢀK and
acetophenone could exchange with the solvent H species. Detecting the reaction route
by kinetic isotope effect8e and amount of D atoms in the product acetophenone13a were
not entirely accurate methods in this catalytic system (Scheme S5). Furthermore,
based on the comparison of these routes, the preꢀactivation of Cβ–H bond was a key
step for the conversion of βꢀOꢀ4ꢀK in Route 2 and Route 3, while the Route 1
involved the direct hydrogenative cleavage the Cβ–OPh bond. We then used
βꢀOꢀ4ꢀKꢀ2βCH3 as the substrate to check the effect of Cβ–H activation on the
cleavage of Cβ–OPh bond (Scheme 3). The NiMo sulfide could catalyze the selective
cleavage of βꢀOꢀ4ꢀKꢀ2βCH3 to 2ꢀmethylꢀ1ꢀphenylpropanꢀ1ꢀone and PhOH, and the
conversion of Cβ–OPh bond reached to 98%. This result indicated that Route 1 with
direct cleavage of Cβ–OPh bond was possibly the main route of βꢀOꢀ4ꢀK
transformation over the NiMo sulfide.
ACS Paragon Plus Environment

Page 27 of 36
ACS Catalysis
1
2
3
4
O
O
O
OH
+
+
+
5
6
7
β-O-4-K-2βCH3
Products GC yield:
8
88%
6%
3%
83%
Conv. 98%
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 3. The cleavage of Cβ–OPh bond in βꢀOꢀ4ꢀKꢀ2βCH3 over the NiMo
sulfide catalyst. Reaction condition: substrate 0.2 mmol, NiMo sulfide 20 mg,
methanol 2.0 mL, H2 1.0 MPa, 180 °C, 6 h.
The reaction route of βꢀOꢀ4ꢀK transformation over the NiMo sulfide via Route 1
is different from the process over the Pd surface via Route 3ꢀ1,14e, 24, 28a but it is
similar to the homogeneous Ru8d or V14a catalysts with single active site. We then
checked the DFT adsorption models of βꢀOꢀ4ꢀK over different heterogeneous
catalysts. As shown in Figure 4A and 4B, two aromatic rings of βꢀOꢀ4ꢀK tend to
adsorb on the Pd(111) surface and Ni(111) surface with a “faceꢀface” model. The
active Pd site has the ability and opportunity to active the Cβ–H bond to transform
βꢀOꢀ4ꢀK to PhCOCH•OPh for the following transformation with a low activation
energy.14e, 24 At the same time, based on the adsorption model of βꢀOꢀ4ꢀK over Pd(111)
surface and Ni(111) surface, the metal sites around the βꢀOꢀ4ꢀK can also catalyze the
hydrogenation of aromatic rings and carbonyl group (Chart 1B, Entries 1 and 2).
When MoS2 with 2D layer structure is used as catalyst, the βꢀOꢀ4ꢀK tends to adsorb at
the edge Mo sites of MoS2. Given that hydrogenative Mo sites are in a linear
arrangement (Figure 4C), besides the Mo sites working as active site to cleave Cβ–
OPh bond, there are few Mo sites work as the side reaction centers for the
hydrogenation of aromatic rings, which lead to the high yield of aromatic rings over
the MoS2. However, the Mo sites around the Cα=O group of adsorbed βꢀOꢀ4ꢀK can
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 36
1
2
3
4
5
6
also work as active sites for the hydrogenation of Cα=O, which leads to the high yield
of βꢀOꢀ4ꢀH (Chart 1B, Entry 4).
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. The DFT adsorption models of βꢀOꢀ4ꢀK on Ni(111), Pd(111) and MoS2 (Mo
edge) (AꢀC), annular darkꢀfield scanning transmission electron microscopy image of
NiMo sulfide (D). (E, F) show the element maps of Ni and Mo from energy dispersive
Xꢀray analysis.
Compared with the catalytic performance of MoS2, introducing NiSx species in
the NiMo sulfide catalyst can obviously increase the selectivity to the hydrogenative
cleavage of Cβ–OPh bond rather than the hydrogenation of Cα=O. As shown in Figure
3, MoS2, NiS and Ni0.7Mo3S4 phases exist in the NiMo sulfide. When NiS is used as
catalyst, the βꢀOꢀ4ꢀK conversion is only 1.5% (Chart 1B, Entry 5). The low activity
indicates that the NiSx species might just be the catalytic promoter to the NiMo
sulfide catalyst. Given that Ni0.7Mo3S4 has the similar cubaneꢀtype Mo clusters
ACS Paragon Plus Environment

Page 29 of 36
ACS Catalysis
1
2
3
4
5
6
7
8
9
structure with MoS2, if its edge Mo atoms are partly replaced by the low active Ni2+,
the potential side reaction sites of the hydrogenation of Cα=O can be reduced, and the
yield of aromatic products generated from the cleavage of Cβ–OPh bond can increase
(Chart 1B, Entry 3). As shown in Figure 4DꢀF, the EDX images show that Ni and Mo
atoms are equally distributed at some zones of NiMo sulfide. So, we propose that the
hydrogenative Mo sites in NiMo sulfide are not in a good linear arrangement like
MoS2 but interrupted by the Ni2+.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To sum up, because of the characteristic of hydrogenative Mo sites in NiMo
sulfide, the adsorption model of βꢀOꢀ4ꢀK and modest hydrogenative ability of NiMo
sulfide adjusted by the NiSx species, the NiMo sulfide prefers to promote the
hydrogenative cleavage of Cβ–OPh bond rather than catalyze the hydrogenation of
Cα=O in the activated βꢀOꢀ4ꢀK.
3.5 The depolymerization of polymeric βꢀOꢀ4 lignin model and birch powders
For the heterogeneous hydrogenation of lignin, the reaction between solid catalyst
and solid substrate remains a challenge.13a In this work, besides the model compounds
of ether monomer, we also checked the application of the selective
oxidationꢀhydrogenation strategy in the conversion of polymeric βꢀOꢀ4 lignin model
(Scheme 4). In the first oxidation of βꢀOꢀ4ꢀAꢀpolymer, no obvious monomer products
were detected in the liquid phase (Figure S14). The solid product was easily filtered
and washed with acetonitrile, and its FTꢀIR spectrum indicated the selective oxidation
of Cα–OH to Cα=O (Figure S15). Further hydrogenation over the NiMo sulfide
catalyst offered a 92% yield of phenolic monomers in 8 h (Scheme 4).
ACS Paragon Plus Environment