A practical and concise homogeneous nickel catalyst for efficient solvent-free synthesis of γ-valerolactone

Article abstract of DOI:10.1039/d0gc00763c

Here, we report an inexpensive nickel-phosphine complex that can catalyse neat levulinic acid (LA) to γ-valerolactone (GVL) without additive at 1 atm H₂, and the catalyst can be reduced to 0.01 molpercent at 30 atm. 5.8 kg LA was successfully converted to GVL in a yield of 94percent with 0.1 molpercent catalyst. DFT calculations andin situAPXPS indicate that L-Ni⁺-H species are the key active intermediates.

Full text of DOI:10.1039/d0gc00763c

View Article Online
View Journal
Green
Chemistry
Cutting-edge research for a greener sustainable future
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: B. Zada, R. Zhu, B.
Wang, J. Liu, J. Deng and Y. Fu, Green Chem., 2020, DOI: 10.1039/D0GC00763C.
Volume 20
Number
May 2018
Pages 1919-2160
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
9
7
Green
Chemistry
Cutting-edge research for a greener sustainable future
rsc.li/greenchem
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 1463-9262
PAPER
Paul T. Anastas et al.
The Green ChemisTREE: 20 years after taking root with the 12
principles
rsc.li/greenchem

Page 1 of 6
Please Gd roe en no tC ha ed mj u iss tt r my argins
View Article Online
DOI: 10.1039/D0GC00763C
Journal Name
COMMUNICATION
A Practical and Concise Homogeneous Nickel Catalyst for Efficient
Solvent-free Synthesis of γ-Valerolactone
Received 00th January 20xx,
Accepted 00th January 20xx
‡
‡
Bakht Zada , Rui Zhu , Bing Wang, Jiao Liu, Jin Deng* and Yao Fu*
DOI: 10.1039/x0xx00000x
www.rsc.org/
Here, we report an inexpensive nickel-phosphine complex that can processes. [ Ruthenium, iridium and other noble metal-based
8]
homogeneous catalysts have been frequently employed for
catalyse neat levulinic acid (LA) toγ-valerolactone (GVL) without
additive at 1 atm H , and the catalyst can be reduced to 0.01mol%
at 30 atm. 5.8 kg LA was successfully converted to GVL in a yield of
4% with 0.1mol% catalyst. DFT calculations and in-situ APXPS
[
9]
the hydrogenation of LA.
3 3
Previously, we used RuCl /PPh
2
based homogenous catalytic system for the hydrogenation of
LA in milder reaction conditions via transfer hydrogenation
9
[9b]
with co-formed FA.
Subsequently, Leitner et al. found that
+
indicate that L-Ni -H species are the key active intermediates.
the use of Ru(acac)
3
/PnOct
3
catalyst can convert LA to GVL in a
. They pointed out
solvent-free condition under 100 atm H
2
Lignocellulosic biomass, as the only renewable organic carbon that the solvent-free reaction conditions greatly facilitated the
resource, is a viable alternative to petroleum derivatives. [
1]
separation of the product and provided the possibility for the
[9c]
recovery of homogeneous catalysts.
Recently, Meer and
Being a significant fraction of biomass, the conversion of
carbohydrates into γ-valerolactone (GVL) is one of the most
pivotal routes of biomass valorization. [ Owing to excellent
Wildeman et al. used Ru-based Shvo catalysts also to achieve
quantitative conversion of the reaction in the solvent-free
2]
[9d]
[9e]
condition.
Similarly, iridium pincer complexes
and half-
[
9f]
physiochemical characteristics, GVL has been extensively used sandwich bipyridyl complexes
have also shown high
[
3]
as a green solvent, polymer monomers
[
4]
[5]
and biofuels.
In this respect, more
O
catalyst / addictives
H2 or HCO2H
O
O
efforts have been devoted to the
preparation of GVL. Generally, GVL is
prepared by the hydrogenation of levulinic
acid (LA), which can be obtained from
carbohydrate biomass via acid hydrolysis.
OH
O
Previous representative work:
H
H
tBu2 P
t
H
Br
Ir P Bu2
P iPr2
N
H
[
6]
This reaction represents a significant
N
N
Fe CO
P
i
H
RuCl3/PPh3
[Ru(acac)3]/PnOct3
3 12
Fe (CO)
HN
synthetic strategy for the efficient use of
renewable resources to produce valuable
Pr2
(Fu & Guo, 2009)
S/C = 1,000
in water
0.1 eq Pyridine
4%, HCOOH (1 eq. )
(Leitner, 2010)
S/C = 1,000
solvent-free
(Zhou, 2012)
S/C = 10,000
in EtOH
(Burtoloso, 2015)
S/C = 25
(Song, 2018)
S/C = 2,000
in MeOH
products.
in water
In recent years, various methods for
producing GVL from LA or its esters using
heterogeneous or homogeneous catalysts
0.01eq NH4PF6
100%, H2 (100atm)
1.2 eq KOH
98%, H2 (50atm)
4.0 eq. imidazole
92%, HCOOH (4 eq. )
1.0 eq. KOH
95%, H2 (50atm)
9
This work:
in the presence of H
2
or formic acid (FA)
without addictive,solvent-free, 98% (S/C = 100), H2 (1 atm)
2
Ph P
[
7]
have been extensively studied.
Ni(OAc)2•4H2O/triphos without addictive,solvent-free, 94% (S/C = 100), HCOOH (2 eq.) triphos: Ph2P
without addictive,solvent-free, 94% (S/C = 10,000), H2 (30 atm)
PPh2
PPh2
Compared with heterogeneous catalysis
systems, homogeneous counterparts not
only have higher catalytic efficiency and
selectivity under mild reaction conditions,
but also generally make it easier to
characterize, understand the catalytic
mechanism, and optimize reaction
Kilogram scale:
Ph2P
Ni(OAc)2•4H2O/DPPP without addictive, solvent-free, 94% (S/C =1,000), H2 (15 atm)
DPPP:
Scheme 1. Homogeneous catalysts for hydrogenation of LA to GVL (S/C = substrate/catalyst molar
ratio).
activities for the conversion of GVL into LA in milder reaction
conditions. Although these noble metal catalysts possess high
catalytic activity, the high costs limit their large-scale
a. B. Zada, R. Zhu, B. Wang, J. Liu, Prof. Dr. J. Deng and Prof. Dr. Y. Fu
Hefei National Laboratory for Physical Sciences at the Microscale, iChEM, CAS Key
laboratory of Urban Pollutant Conversion, Anhui Province Key Laboratory of
Biomass Clean Energy, Department of Chemistry, University of Science and
Technology of China, Hefei, Anhui 230026, China.
Hefei 230026 (PR China)
E-mail for J. Deng: dengjin@ustc.edu.cn
E-mail for Y. Fu: fuyao@ustc.edu.cn
applications.
Alternatively, some non-noble homogeneous metal
catalysts have also been reported for the hydrogenation of LA
into GVL. [ Our previous work showed that iron(II) triflate
10]
[
10a]
with tetraphosphine ligand P(CH
2
CH
2
PPh
)
2 3
and the
Casey’s [
10b]
catalyst could catalyze the conversion of levulinic
[10c]
[10d]
esters (LE) into GVL. Burtoloso et al.
and Song et al.
‡
These authors contributed equally to this work.
reported almost quantitative yields of GVL from LA over
Electronic Supplementary Information (ESI) available: [details of any
supplementary information available should be included here]. See
DOI: 10.1039/x0xx00000x
Fe
3
(CO)12 and pincer iron-complex in the presence of bases
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please Gd roe en no tC ha ed mj u iss tt r my argins
Page 2 of 6
COMMUNICATION
Journal Name
reaction activity of the catalyst is
View Article Online
reduced. The other Ni salts all
Table 1. Screening of ligands for conversion of LA to GVL under solvent-free condition
DOI: 10.1039/D0GC00763C
have good catalytic activity and
good GVL yield. By considering the
reactivity of the catalysts and
optimal reaction conditions, the
O
Ni(OAc)2•4H2O, Ligand
O
OH
+
H2
O
+
2
H O
150ºC, 20h, sovlent-free
O
a
Ni(OAc)
2
was selected as the best
entry
1
2
3
4
5
catalyst in the presence of a ligand
o
Ph
Ph
2
under 1atm H pressure, at 150 C
P
PPh2
2
PPh2
Ph2P
PPh2
Ph P
Ph2P
PPh2 Ph P
2
for 20 h. (See Fig. S2 & S3 for
optimized data of reaction time
Ligand
Yield
Ph
PPh3
DPPM
DPPE
DPPP
DPPB
and
reaction
temperature)
Subsequently, the effect of
b
c
trace
5%
30%
37% (94% , 48% )
8%
phosphine
hydrogenation
ligands
was
on
the
also
a
entry
6
7
8
9
10
investigated under the above
reaction conditions. According to
Table 1, the tridentate phosphine
ligand ‘‘so-called triphos’’ has the
best performance. In contrast to
MeO
Ph
P
Cy
P
Cy
Cy Ph2P
Cy
P
P
P
P
Cy
P
Ph
P
Cy
Cy Ph2P
PPh2
Ligand
Yield
Cy
OMe
DCyPE
DCyPP
triphos
DIPAMP
(R,R)-Me-BPE
3
triphos, mono-dentate PPh ligand
showed very low catalytic activity,
probably due to the poor
coordination ability. While, when
bidentate phosphine ligands were
used, the catalytic activities closely
related to the structure of the
complex (Table 1 entries 2-5). 1,3-
b
c
88%
90%
98% (97% , 94% )
90%
93%
Reaction conditions: a. 1 atm H2, 1 mol% Ni(OAc)2•4H2O, 1 mol% Ligand;
b. 15 atm H2, 0.1 mol% Ni(OAc)2•4H2O, 0.1 mol% Ligand;
c. 30 atm H2, 0.01 mol% Ni(OAc)2•4H2O, 0.01 mol% Ligand.
and FA or H
2
as a hydrogen source, respectively. However, low bis-(diphenylphosphino)ethane
(DPPE)
or
1,3-
catalyst activity, cumbersome catalyst preparation, and having bis(diphenylphosphino)propane (DPPP) ligands capable of forming
to use indirect feedstocks (LE instead of LA) or large amounts stable five- or six-membered ring structures had relatively high
of base in order to maintain catalyst stability are the main catalytic activities. When the benzene ring was replaced with
obstacles to industrial scale GVL production with non-noble cyclohexane ring, the catalytic activities could be significantly
homogeneous metal catalysts.
improved (entry 6 & entry 7). This can be attributed either to
A summary on homogeneous hydrogenation of LA or LE to the electron-donating effect or to the steric hindrance effect.
[
12]
2
GVL is listed in Table S1. Accordingly, there is still lack of an
Firstly, compare to the aryl group possess a C structure,
sp
3
ideal homogeneous catalyst that should possess a non-noble alkyl group with a Csp structure has a stronger electron-
metal catalyst center, inexpensive and easily available ligand, donating ability. Secondly, compare to the planar structure of
high turnover number (TON) value, and high efficiency under aryl group, the steric hindrance of the alkyl group could reduce
low H pressure in solvent-free medium. However, developing the buried volume and increase the cone angle of the ligand,
2
this ideal catalytic system has the following difficulties and which is more conducive to improving the catalytic activity of
the nickel catalyst. [ These explanations can also be verified
13]
challenges:
1) Due to the following chemical equilibrium reactions:
L-Mn+ + H2
by adding an electron donor group on the benzene ring (entry
(
9
) or further increasing the steric hindrance (entry 10) to
L-M(n-1)+-H + H+
improve the catalyst activity. In case of tri-dentate triphos
the potential active intermediate L-M(n-1)+-H should be difficult ligand (entry 8), a high yield of GVL (98%) was obtained. It
to produce in neat levulinic acid. To shift the chemical should be noted that although triphos is a tridentate
equilibrium to the right, H
phosphine ligand, only two P atoms are chelated with the Ni
metal in the complex. [ Therefore, the other P-containing
substituent mainly functioned as an electron-donating group.
In addition, by increasing the hydrogen pressure, the catalytic
reactivity can be improved, and thus the amount of catalyst
2
pressure needs to be increased.
14]
(
2) Relative to the second- and thirdrow transition noble
metal, the firstrow non-noble metal hydrides tend to have
o
relatively lower kinetic and thermodynamic hydricities (∆G H-).
[
11]
can be reduced. For example, by increasing the H
15 atm, only 0.1 mol% of the inexpensive and readily available
Herein, we developed in-situ formed nickel-phosphine ligand DPPP complexed with Ni(OAc) can catalyze the
2
pressure to
L-M(n-1)+-H
L-Mn+ + H- GoH-
2
complexes for hydrogenation of LA into GVL. A robust system hydrogenation of LA to obtain GVL in a yield of 94%. While,
with industrially attractive features such as solvent-free conditions, Ni(OAc)
low H pressure, without any additives, inexpensive and readily
2
/triphos provided an excellent catalytic activity even
2
OH
accessible catalyst with TON value close to 10,000 leading to
complete LA conversion, and through a simple post-treatment
process to purify GVL and recycle catalyst.
Initially, we examined the effects of various low-cost Ni salts
on the hydrogenation of LA into GVL. The results of Ni salts
screening are listed in Table S2. We find that when the anions
of Ni salts are halogen atoms, the yield of GVL is generally low.
We speculate that this is mainly due to halogen atoms
participating in competitive coordination with Ni atom, leading
to the ligand exchange difficult to occur. As a result, the
Path I
OH
-
H
H2
2 O
O
H +
O
4-Hydroxypentanoic acid
[
Ni/P]-cat.
O
O
OH
-
H
O
2O
H2
Levulinic acid
O
-Valerolactone
O
[Ni/P]-cat.
Path II
Angelica lactone
Scheme 2. Possible reductive transformation pathways from LA to GVL.
2
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 6
Please Gd roe en no tC ha ed mj u iss tt r my argins
Journal Name
COMMUNICATION
These indicated that the protons of
View Article Online
O
levulinic acidDO
c
I
ou
: 1
0
ld
.
103
pr
9/D
omote e
076th3C
0G
C
0


G(L) = -1.2 kcal/mol
G(L') = -0.9 kcal/mol reaction.
OH
LNi(OAc)2
+
2
LNi(OOCCH CH COCH ) + 2 HOAc
2
2
3 2
O
LNi(A)2
To further understand the
reaction mechanism, density
functional theory (DFT) calculations
were performed (Scheme 3). First,
2
Ni(OAc) is formed a paramagnetic
complex with a phosphine ligand,
which can be confirmed by EPR (Fig.
S5), indicates that the complex is an
outer-orbital coordination, that is, a
high-spin with unpaired electrons
compound. As the anion-exchange
between levulinate and acetate is
quite facile, the nickel(II) levulinate

G (kcal/mol)
L=triphos H: proton
H: hydride
H: neutral hydrogen

H
O

L'=DPPP

O
L
Ni
O
TS2'
7.4
O
H
O
O
O
O
O
H
O

3
O
H
H H
Ni
A
O
O
O
H
OH
H
O
L
HA
I3'
26.9
O
O
O
TS2
34.8
O
H
O
TS1'
4.1
HA
TS4
24.6
2
TS3
2.3
I2
1.4
2
2
I3
25.7
HA
TS1
2.5
2
I2'
H
L
Ni
A
OH
O
HA
20.8
L Ni
O
HA
H2O
H2
A
H
H
I4
L
Ni
A
A
I7
4
.9
I1' (triplet)
.0
HA
10.3
O
I6
-0.4
0
LNi(A)2HA
I4'
.7
OH
OH
3
I1(triplet)
.0
LNi(A)2
O
0
O
I5
2
with the phosphine ligand (LNi(A) )
-2.4
OH
HA
=
OH
O
H
was considered to be the active
catalytic species. It was found that
the triplet form is more stable than
the singlet form, thus was taken as
the energy reference. As clearly
illustrated in Scheme 3, hydrogen
O
I8
-9.9
O
OH
O
OH
O
H
O
O
O
Scheme 3. Energies profile for proposed mechanism. TS: transition state, HA: levulinic acid, L: ligand.
at 0.01 mol% under 30 atm H
,400.
To investigate the reaction pathways, validation
2
pressure with TON value of
was first coordinated to metal Ni and the heterolytic H-H bond
dissociation occurred with the assistance of a neighboring
carboxyl group (via TS1/TS1’), leading to the formation of the
Ni-H compound. It was worth noting that the activation of
9
experiments were performed and two possible reaction
pathways were proposed (Scheme 2): Path I: The ketone
carbonyl group was hydrogenated to a hydroxyl group,
followed by intramolecular esterification into GVL; Path II: LA
underwent intramolecular dehydration to form α-angelica
lactone, and then α-angelica lactone was hydrogenated to
form GVL. However, when α-angelica lactone was used as a
substrate under the specific reaction conditions, only 3% yield
of GVL was obtained. This figures that the hydrogenation of LA
under nickel/phosphine catalysis went through pathway 1. In
addition, when methyl levulinate was used as a substrate, 35%
yield of GVL was produced (Scheme S1). When we added 4
mol% of n-valeric acid to this reaction system to provide
additional protons, the GVL yield increased from 35% to 54%.
hydrogen in this step was reversible. Therefore, increasing H
2
pressure was conducive to the activation of catalyst. After
that, the concerted hydride transfer/protonation of the
coordinating levulinate anion was taken place. Herein, the
protonated carbonyl group was attacked by hydride and
converted to a hydroxyl group via TS2/TS2’, which was the
rate-determining step of the reaction. The subsequent anion-
exchange between intermediate I4/I4’ and levulinic acid
2
regenerated the LNi(A) catalyst and released the alcohol
compound. And finally, GVL was produced through the ring-
forming (TS3) and dehydration (TS4) processes with the aid of
protonic acid LA. This was also supported experimentally,
when a methyl levulinate was used 36% yield of GVL was
obtained in same reaction conditions. The whole reaction
process is exergonic by 9.9 kcal/mol and the reaction energy
barrier was 34.8 kcal/mol (for triphos) or 37.4 kcal/mol (for
2
DPPP). Additionally, the energy barrier of Ni(OAc) /DPPP is
higher than that of Triphos/Ni, either through the hydrogen
activation step (via TS1/TS1’) or the hydride transfer step (via
TS2/TS2’), which also proves that the catalytic activity of
Ni(OAc)
hydride intermediate I2' is relatively stabler than I2, which
indicates increasing H pressure is more favourable to improve
catalytic efficiency of Ni(OAc) /DPPP. Of course, the value
2 2
/DPPP is lower than Ni(OAc) /triphos. While, the
2
2
calculated by DFT is just for reference. We have also used
Eyring formula (detail process could be seen in Fig. S4) to
measure the reaction activation energy at 33.4 kcal/mol
(
triphos) and 35.1 kcal/mol (DPPP) respectively, which also
proves that the catalytic activity of Ni(OAc) /triphos is higher
than Ni(OAc) /DPPP.
2
2
In addition, in order to confirm the existence of the active
intermediate L-Ni -H, we employed atmospheric pressure X-
2
Figure 1. XPS spectra Ni2P of [Ni(OAc) ] /DPPP catalyst (a) in the absence
+
of hydrogen (b) under hydrogen atmosphere at 150ºC.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please Gd roe en no tC ha ed mj u iss tt r my argins
Page 4 of 6
COMMUNICATION
Journal Name
Next, in order toViewstAurtdicyle Ontlhinee
(
a)
Ni(OAc) /triphos
2
(b)
Ni(OAc)
2
/DPPP
DOI: 10.1039/D0GC00763C
applicability of the nickel-phosphine
Post-treatment in air
Post-treatment in N₂
Post-treatment in air
Post-treatment in N₂
1
00%
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
complex catalyst to the conversion of
LA directly derived from dehydration
of carbohydrates, 9 g (50 mmol) of
glucose was dissolved in 50 mL of 0.8
M HCl and heated at 200 °C for 1 h.
The reaction mixture was filtered and
distilled under reduced pressure, ca. 2
ml (2.38 g) of crude LA was obtained
9
8
7
6
5
4
3
2
1
0%
0%
0%
0%
0%
0%
0%
0%
0%
0
%
fresh
1st
2nd
3rd
4th
fresh
1st
2nd
3rd
4th
Run
Run
as
a
yellow liquid.
•4H (51.2
1
mol%
and
Figure 2. Reuse of catalyst for the hydrogenation of LA in solution-free after a vacuum distillation to recycle
the catalyst (a) Ni(OAc) /triphos (b) Ni(OAc) /DPPP. Reaction conditions: LA (10 mmol), Ni(OAc) •4H O (0.01
mmol), triphos (0.01mmol) or DPPP (0.01mmol), H (15 atm), 150°C, 20h.
Ni(OAc)
2
2
O
mg)
2
2
2
2
triphos(128 mg) were added to the
distilled product, and the mixture was
2
ray photoelectron spectroscopy (AP-XPS) to test the charge
density in Ni valence before and after the hydrogenation. [
2
stirred at 150 °C for 20 h under 1atm H . Under these conditions,
15]
GVL was obtained in a yield of 93%. The overall GVL yield from
glucose was about 38% (Table 2, entry 1). For the generation of
GVL from other carbohydrate biomass resources, the catalytic
system can also be successfully applied as shown in Table 2. We
also tried to use the acid hydrolysis solution of carbohydrates as
raw materials directly to perform the catalytic reduction under high
pressure hydrogen, but GVL can be obtained in a yield of less
than 10%. Water as a solvent and formic acid as a by-product of
hydrolysis have no obviously negative effect on the reaction, but
hydrochloric acid has a strong inhibitory effect (Scheme S1). This
The XPS signal peaks of Ni(OAc)
cannot be resolved well. Whereas, Ni(OAc)
demonstrated clear high-resolution XPS spectra of Ni2p before
2
/triphos are so weak that
2
/DPPP
2
+
2
and after H exposure as shown in Figure 1. The L-Ni species
was confirmed from the bonding energies observed at 855.97
and 873.66 eV respectively. While, additional peaks at low
bonding energies (i.e., 852.98 and 870.06 eV) were appeared
when the XPS spectrum was recorded in the presence of H
2
atmosphere (Figure 1b). The new signal peaks illustrate the
+
emergence of L-Ni -H species.
As mentioned above, the inherit limitations for downstream is either due to excessive acidity leading to decomposition of the
separation of individual products in homogenous catalysis can be catalyst or due to chloride ions participating in competitive
easily overcome by using solvent free conditions. [9c] Therefore, coordination resulting in reduced catalyst activity. It is worth to
we conducted catalysts cycling to test the reusability and the mention that although FA can also be used as a hydrogen source
stability of catalysts. After the reaction, the product GVL was to reduce LA. Due to the saturated vapor pressure curves of FA
separated from the catalyst by distillation under reduced pressure, and water almost overlap, most FA and water are removed by
and the distillation residue was used as catalyst in the next round. distillation during the concentration of the hydrolysate. Therefore,
The results showed that after reused twice, the catalytic efficiency the actual subsequent reduction must use external FA instead of
2 2
decreased significantly. By adding Ni(OAc) •4H
O and phosphine concomitant hydrolyzed by-product FA. Secondly, according to
ligand separately, it was found that only the addition of phosphine the analysis of the life cycle assessment (LCA) of the reaction
ligand could restore the catalytic activity, which proved that the pointed by Meer and Wildeman et al., [9d] the use of excess FA
change of the ligand affected the activity of the catalyst. could lead to its global warming potential (GWP) to be higher than
Speculating the phosphine ligand in the complex may be oxidized
by air, [16] a post-treatment of separating and recycling the catalyst
H
2
, which means the H
2
is the "greenest" choice.
Because of the absence of mass transfer and diffusion
by vacuum distillation in N
2
atmosphere was implemented, which resistances, facilitate efficient transfer of heat and the inherent
could greatly alleviate the catalyst deactivation (Fig. 2).
similarity between lab and industrial conditions, homogeneous
catalytic hydrogenation systems are usually easier to scale up
than heterogeneous counterpart. [8] Considering the cost of the
catalysts, we selected the commercial obtainable inexpensive
ligand DPPP (100 $/kg) instead of triphos (50 $/g), and carried out
the catalytic hydrogenation of 5.8 kg LA (50mol) at the dosage of
Table 2. Dehydration of carbohydrates [a] and the subsequent reduction of
carbohydrate-derived LA in the presence of Ni(OAc)
2
/triphos [b]
entry
Carbohydrate
LA yield [mol%] [c]
GVL yield [mol%] [d]
1
2
3
4
5
glucose
fructose
sucrose
starch
41
45
43
36
27
38 (93)
43 (95)
41 (95)
34 (94)
26 (96)
0.1mol% of DPPP (20 g) and Ni(OAc)
2
•4H
2
O (12.5 g). After
pressure, about
stirring at 150 °C for 18 hours under 13-16 atm H
2
95% of LA was successfully converted to GVL. GVL with a purity
of more than 98% can be obtained by simple vacuum distillation.
The cost of the catalyst was less than 0.52 $/kg GVL per batch
cellulose
[
a] Reaction conditions: carbohydrate (9.0 g), HCl aqueous (0.8 N, 50 mL),
N
2
, 200°C, 1h. [b] Reaction conditions: vacuum distillate from hydrolysis,
•4H O (1.0 mol%), triphos (1.0 mol%), 1atm of H , 150°C, 20h. [c]
(See SI for details), and an efficient recycling of catalyst could
Ni(OAc)
2
2
2
further reduce overall cost of the process (based on five cycles,
the catalyst cost will be less than 0.1 $/kg GVL).
Isolated yields of vacuum distillation based on carbohydrate. [d] Overall yield
for the two-stage conversion of carbohydrate. The yield based on the
conversion of LA is given in parenthesis.
4
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 6
Please Gd roe en no tC ha ed mj u iss tt r my argins
Journal Name
COMMUNICATION
Marrodan, P. Barbaro, ACS Catal. 2015, 5, 1882Vi–ew1A8r9tic4le;Odn)liZne.
In summary, the results presented here describe a reusable
catalytic system based on an inexpensive and abundant firstrow
transition metal Ni, which is capable of very efficiently
Zhang, ChemSusChem 2016, 9, 156-171.
DOI: 10.1039/D0GC00763C
[
[
8] C. H. Bartholomew, R. J. Farrauto, Fundamentals of industrial
catalytic processes. John Wiley & Sons, 2011.
9] a) U. Omoruyi, S. Page, J. Hallett, P. W. Miller,
ChemSusChem 2016, 9, 2037 – 2047. b) L. Deng, J. Li, D. Lai,
Y. Fu, Q. Guo, Angew. Chem. Int. Ed. 2009, 48, 6529 – 6532;
c) F. M. A. Geilen, B. Engendahl, A. Harwardt, W. Marquardt,
J. Klankermayer, W. Leitner, Angew. Chem. Int. Ed. 2010, 49,
hydrogenating LA to GVL without solvent and additive, using H
2
as the reductant. These advantages give the Ni/phosphine, the
robust catalytic system, a strong industrial potential to apply in the
+
key step of biomass conversion. In addition, L-Ni -H species, the
5510 – 5514; d) C. A. M. R. van Slagmaat, M. A. F. Delgove, J.
key active intermediate indicated by DFT calculations and in-situ
APXPS, can have the catalytic activity equivalent to that of noble
metal catalyst by means of regulating steric and electronic effects
of the substituents on the ligand.
Stouten, L. Morick, Y. van der Meer, K. V. Bernaerts, S. M. A.
De Wildeman, Green Chem. 2020, 22, 2443 – 2458; e)W. Li, J.
Xie, H. Lin, Q. Zhou, Green Chem. 2012, 14, 2388 – 2390; f) J.
Deng, Y. Wang, T. Pan, Q. Xu, Q. Guo, Y. Fu,
ChemSusChem 2013, 6, 1163 – 1167.
[
10]a) M. Fu, R. Shang, Z. Huang, Y. Fu, Synlett 2014, 25, 2748 –
2752; b) N. Dai, R. Shang, M. Fu, Y. Fu, Chin. J. Chem. 2015,
33, 405 – 408; c) G. Metzker, A. C. B. Burtoloso, Chem.
Commun. 2015, 51, 14199 – 14202; d) Y. Yi, H. Liu, L. Xiao, B.
Wang, G. Song, ChemSusChem 2018, 11, 1474 – 1478; e) C.
A. M. R. van Slagmaat, S. M. A. D. Wildeman, Eur. J. Inorg.
Chem. 2018, 694 – 702; f) Z. Liu, Z. Yang, P. Wang, X. Yu, Y.
Wu, H. Wang, Z. Liu, ACS Sustainable Chem. Eng. 2019, 7,
ACKNOWLEDGMENT
This work was supported by the National Key R&D Program of
China (2018YFB1501604), the National Natural Science
Foundation of China (21875239) and the Fundamental Research
Funds for the Central Universities (WK3530000004). The authors
thank the Hefei Leaf Biotech Co., Ltd and Anhui Kemi Machinery
Technology Co., Ltd for free samples and equipment that
benefited our ability to conduct this study.
1
8236 – 18241.
11]a) S. Chakraborty, H. Guan, Dalton Trans. 2010, 39, 7427 –
436; b) K. M. Waldie, A. L. Ostericher, M. H. Reineke, A. F.
[
[
7
Conflicts of interest
Sasayama, C. P. Kubiak, ACS Catal. 2018, 8, 1313 – 1324.
12]X. Qi, Y. Fu, L. Liu, Q. Guo, Organometallics 2007, 26,
There are no conflicts to declare.
4197−4203.
[
[
13] K. Wu, A. G. Doyle, Nat. Chem. 2017, 9, 779 – 784.
14]
a) M. Kandiah, G. S. McGrady, A. Decken, P. Sirsch, Inorg.
Chem. 2005, 44, 8650 – 8652; b) J. Mautz, K. Heinze, H.
Wadepohl, G. Huttner, Eur. J. Inorg. Chem. 2008, 1413 –
1422.
Notes and references
[
1] a) A. Farrán, C. Cai, M. Sandoval, Y. Xu, J. Liu, M.J.Hernáiz,
R. J. Linhardt, Chem. Rev. 2015, 115, 6811 – 6853; b) L. T.
Mika, E. Cséfalvay, Á. Németh, Chem. Rev. 2018, 118, 505 – [15]Because
13; c) W. Leitner, J. Klankermayer, S. Pischinger, H. Pitsch,
K. Kohse-Höinghaus, Angew. Chem. Int. Ed. 2017, 56, 5412 –
452.
Ni(OAc) /triphos
and
Ni(OAc) /DPPP
are
2
2
6
paramagnetic, they are not suitable to test with in-situ nuclear
magnetic resonance (NMR) under hydrogen atmosphere to
directly prove the existence of hydride.
5
[
2] a) I. T. Horváth, H. Mehdi, V. Fábos, L. Boda, L. T. Mika, [16]S. A. Buckler, J. Am. Chem. Soc. 1962, 84, 3093 – 3097.
Green Chem. 2008, 10, 238 – 242; b) D. M. Alonso, S. G.
Wettstein, J. A. Dumesic, Green Chem. 2013, 15, 584 – 595;
c) L. Bui, H. Luo, W. R. Gunther, Y. Román-Leshkov, Angew.
Chem. Int. Ed. 2013, 52, 8022 – 8025; d) L. Yan, Q. Yao, Y.
Fu, Green Chem. 2017, 19, 5527 – 5547; e) Z. Yu, X. Lu, C.
Liu, Y. Han, N. Ji, Renew. Sustain. Energy Rev. 2019;112:140
–
157.
[
3] a) J. S. Luterbacher, J. M. Rand, D. M. Alonso, J. Han, J. T.
Youngquist, C. T. Maravelias, B. F. Pfleger, J. A. Dumesic,
Science 2014, 343, 277 – 280; b) E. I. Gürbüz, J. M. R. Gallo,
D. M. Alonso, S. G.Wettstein, W. Y. Lim, J. A. Dumesic,
Angew. Chem. Int. Ed. 2013, 52, 1270 – 1274; c) G.
Strappaveccia, L. Luciani, E. Bartollini, A. Marrocchi, F. Pizzo,
L. Vaccaro, Green Chem. 2015, 17, 1071 – 1076.
[
[
4] a) J.-P. Lange, J. Z. Vestering, R. J. Haan, Chem. Commun.
2007, 3488 – 3490; b) A. Marckwordt, F. E. Ouahabi, H.
Amani, S. Tin, N. V. Kalevaru, P. C. J. Kamer, S. Wohlrab, J.
G. de Vries, Angew. Chem. Int. Ed. 2019, 58, 3486 – 3490.
5] a) J.-P. Lange, R. Price, P. M. Ayoub, J. Louis, L. Petrus, L.
Clarke, H. Gosselink, Angew. Chem. Int. Ed. 2010, 49, 4479 –
4483; b) J. Q. Bond, D. M. Alonso, D. Wang, R. M. West, J. A.
Dumesic, Science 2010, 327, 1110 – 1114; c) R. Palkovits,
Angew. Chem. Int. Ed. 2010, 49, 4336 – 4338.
[
[
6] a) X. Du, L. He, S. Zhao, Y. Liu, Y. Cao, H. He, K. Fan, Angew.
Chem. Int. Ed. 2011, 50, 7815 – 7819; b) D. Lai, L. Deng, Q.
Guo, Y. Fu, Energy Environ. Sci. 2011, 4, 3552 – 3557; c) W.
Luo, M. Sankar, A. M. Beale, Q. He, C. J. Kiely, P. C.A.
Bruijnincx, B. M. Weckhuysen, Nat. Commun. 2015, 6: 6540.
7] a) W. R. H. Wright, R. Palkovits, ChemSusChem 2012, 5,
1657 – 1667; b)K. Yan, Y. Yang, J. Chai, Y. Lu, Appl. Catal. B:
Environ. 2015, 179, 292 – 304; c) F. Liguori, C. Moreno-
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please Gd roe en no tC ha ed mj u iss tt r my argins
Page 6 of 6
View Article Online
DOI: 10.1039/D0GC00763C
Journal Name
COMMUNICATION
Just a few nickels
Ni2+
O
Phosphine ligand
O
OH
+
H2
O
+
H2O
solvent-free
without additive
O
9
9
8% (S/C = 100), H2 (1 atm)
4% (S/C = 10,000), H2 (30 atm)
Ni(OAc) /triphos
2
Kilogram scale Ni(OAc) /DPPP
94% (S/C = 1,000), H2 (15 atm)
2
Catalyst cost: <0.1$/kgGVL
(S/C = substrate/catalyst molar ratio)
An inexpensive and readily available nickel-phosphine catalyst is used to efficiently catalyze the reduction of
levulinic acid to produce γ-valerolactone and the catalyst cost only needs a few nickel coins (< 0.1 $/kg GVL).
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 6
Please do not adjust margins