Ni Nanoparticles Inlaid Nickel Phyllosilicate as a Metal-Acid Bifunctional Catalyst for Low-Temperature Hydrogenolysis Reactions

Article abstract of DOI:10.1021/acscatal.5b01080

Hydrogenolysis of carbon-oxygen bonds is a versatile synthetic method, of which hydrogenolysis of bioderived 5-hydroxymethylfurfural (HMF) to furanic fuels is especially attractive. However, low-temperature hydrogenolysis (in particular over non-noble catalysts) is challenging. Herein, nickel nanoparticles (NPs) inlaid nickel phyllosilicate (NiSi-PS) are presented for efficient hydrogenolysis of HMF to yield furanic fuels at 130-150 °C, being much superior with impregnated Ni/SiO₂ catalysts prepared from the same starting materials. NiSi-PS also shows a 2-fold HMF conversion intrinsic rate and 3-fold hydrogenolysis rate compared with the impregnated Ni/SiO₂. The superior performance originated from the synergy of highly dispersed nickel NPs and substantially formed acid sites due to coordinatively unsaturated Ni (II) sites located at the remnant nickel phyllosilicate structure, as revealed by detailed characterizations. The model reactions over the other reference catalysts further highlighted the metal-acid synergy for hydrogenolysis reactions. NiSi-PS can also efficiently catalyze low-temperature hydrogenolysis of bioderived furfural and 5-methylfurfural, demonstrating a great potential for other hydrogenolysis reactions.

Full text of DOI:10.1021/acscatal.5b01080

Research Article
pubs.acs.org/acscatalysis
Ni Nanoparticles Inlaid Nickel Phyllosilicate as a Metal−Acid
Bifunctional Catalyst for Low-Temperature Hydrogenolysis Reactions
Xiao Kong,†,‡,⊥ Yifeng Zhu,†,‡,⊥ Hongyan Zheng,§ Xianqing Li,§ Yulei Zhu,*,†,§ and Yong-Wang Li†,§
†State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, P.R. China
‡University of Chinese Academy of Sciences, Beijing 100049, P.R. China
§Synfuels China Co. Ltd, Beijing, 101407, P.R. China
S
* Supporting Information
ABSTRACT: Hydrogenolysis of carbon−oxygen bonds is a
versatile synthetic method, of which hydrogenolysis of
bioderived 5-hydroxymethylfurfural (HMF) to furanic fuels is
especially attractive. However, low-temperature hydrogenolysis
(in particular over non-noble catalysts) is challenging. Herein,
nickel nanoparticles (NPs) inlaid nickel phyllosilicate (NiSi-
PS) are presented for efficient hydrogenolysis of HMF to yield
furanic fuels at 130−150 °C, being much superior with
impregnated Ni/SiO2 catalysts prepared from the same starting
materials. NiSi-PS also shows a 2-fold HMF conversion
intrinsic rate and 3-fold hydrogenolysis rate compared with
the impregnated Ni/SiO2. The superior performance origi-
nated from the synergy of highly dispersed nickel NPs and substantially formed acid sites due to coordinatively unsaturated Ni
(II) sites located at the remnant nickel phyllosilicate structure, as revealed by detailed characterizations. The model reactions over
the other reference catalysts further highlighted the metal−acid synergy for hydrogenolysis reactions. NiSi-PS can also efficiently
catalyze low-temperature hydrogenolysis of bioderived furfural and 5-methylfurfural, demonstrating a great potential for other
hydrogenolysis reactions.
KEYWORDS: hydrogenolysis, heterogeneous catalysis, nickel, silica, 5-hydroxymethylfurfural
Scheme 1. (a) Reaction Route of HMF Hydrogenolysis to
Furanic Fuels. (b) Ni-Based Catalyst with Protogenetic
Metal−Acid Bifunctionality Derived from Nickel
Phyllosilicate
1. INTRODUCTION
Silica-supported metal NPs are widely employed solid catalysts
for sustainable synthesis of fuels and chemicals. However, SiO2
is always considered as an inert oxide, suffering from the weak
binding and dispersing ability for metal NPs.1 The resulted
SiO2 supported catalysts therefore commonly have low
reactivity. Moreover, it has hardly any surface acidity, making
it a nonpreferred choice for reactions needing acid function-
ality, such as hydrogenolysis of carbon−oxygen bonds.2
Hydrogenolysis of carbon−oxygen bonds (e.g., aldehydes,
ethers, esters, and COx) is an important method for producing
chemicals with lower oxygen contents.2b,3 A typical and
promising case is hydrogenolysis of bioderived HMF to furanic
fuels (e.g., 2,5-dimethylfuran (DMF) and 2,5-dimethyltetrahy-
drofuran (DMTHF) (Scheme 1 and Table S1).3a,4 This
reaction is generally catalyzed by noble metals (e.g., Ru, Pt,
and Pd) at high temperatures (180−220 °C),3a,5 which leaves
ample room for improvements. Low-temperature hydro-
genolysis is charming because it is more energy saving and
produces minimized humins and/or C−C cracking byproducts.
Reported works on low-temperature hydrogenolysis of HMF to
furanic fuels are relatively scarce. A Pd/Zn/C was reported to
catalyze HMF hydrogenolysis with a yield of 85% at 150 °C.6 A
Ru/Co3O4 catalyst could give a yield of 93.4% at 130 °C for
Received: May 22, 2015
Revised: August 2, 2015
© XXXX American Chemical Society
5914
DOI: 10.1021/acscatal.5b01080
ACS Catal. 2015, 5, 5914−5920

ACS Catalysis
Research Article
this reaction.7 Rauchfuss et al. proposed that Pd/C catalyst
combined with formic acid solvent could enable an efficient
synthesis of DMF from fructose at 120−150 °C.8 It seems that
usage of noble metals and/or a specific solvent is a prerequisite
for low-temperature hydrogenolysis. Low-temperature hydro-
genolysis of HMF to furanic fuels using non-noble catalysts has
never been reported.
The formation of nickel phyllosilicate can be proved by FT-
IR (Figure 1), XRD (Figure 2a), and TEM (Figure 2b). The
Among the non-noble catalysts, nickel-based catalysts attract
great interest in the conversion of lignocellulosic biomass,
especially in lignin-derived model compounds.3b,9 Grilc et al.9a,b
reported a series of Ni-based bifunctional catalysts (MIL-
77(Ni), NiMo/Al2O3) for efficient hydrodeoxygenation of
solvolysed lignocellulosic biomass. The group of Lercher
reported a Ni/HZSM-5 catalyst for hydrodeoxygenation of
pyrolysis oil to hydrocarbons under moderate conditions.3b,10
In contrast, HMF hydrogenation over Ni-based catalysts was
scarcely investigated. Fu et al.11 reported that a bifunctional
Ni−W2C/AC catalyst is effective for DMF synthesis at 180 °C.
Recently, we have also shown that nickel-based catalysts (Raney
Ni and Ni−Al2O3 derived from hydrotalcite) are promising
candidates for HMF hydrogenolysis at high temperatures.12
Nevertheless, it is particularly encouraging to convert HMF
under mild conditions over Ni-based catalysts.
Figure 1. FT-IR spectra of different precursors.
Yet, hydrogenation of CC bonds to yield 2,5-dihydrox-
ymethyltetrahydrofuran (DHMTHF) dominates at low tem-
peratures over Ni catalysts (Scheme 1).12a The formed
DHMTHF was rather stable and could not be easily converted,
which are detrimental for production of furanic fuels.11,12b,13
The efficient synthesis of furanic fuels relies on the fast
conversion of C−O to CC bonds under mild conditions, and
the development of bifunctional metal−acid catalyst is a
prerequisite. Herein, we present a nickel NPs inlaid nickel
phyllosilicate (NiSi-PS) catalyst which featured with highly
dispersed nickel NPs and large amounts of proximal acid sites
for HMF hydrogenolysis. The catalyst displays excellent
hydrogenolysis ability at low temperatures (130−150 °C),
which cannot be achieved over supported Ni/SiO2 catalysts.
The metal−acid synergy over NiSi-PS promotes the intrinsic
activity for both HMF conversion and hydrogenolysis reactions
(turnover frequency (TOF), 1404 and 1487 h−1 respectively),
in comparison with Ni/SiO2 catalyst (703 and 479 h−1). NiSi-
PS can also efficiently catalyze hydrogenolysis of bioderived
furfural and 5-methylfurfural at low temperatures.
Figure 2. (a) XRD patterns of precursors and (b) TEM image of NiSi-
PS precursor.
bands at 670, 1024, and 3627 cm−1 are attributed to δOH
vibration, symmetric in-plane Si−O vibration, and νOH
stretching of nickel phyllosilicate (Figure 1).15 No related
bands assigned to phyllosilicate were detected for NiSi-IMP-36
and NiSi-IMP-20 precursors, whereas the bands at 1109 and
1385 cm−1 were observed, which were attributed to amorphous
SiO2 and NO32− respecitively.16 The presence of nickel nitrates
was confirmed by XRD peaks (Figure 2a). Broad XRD peaks at
26.7°, 33.7°, 39.7°, 53.2°, and 60.9° of NiSi-PS precursor can
be attributed to nickel phyllosilicate (PDF#43-0664) (Figure
2a). The characteristic “fibrous” microstructures also confirmed
the presence of phyllosilicate over NiSi-PS precursor (Figure
2b).14a The above results revealed that nickel phyllosilicate was
achieved via ammonia evaporation, which would substantially
change the properties of resulted catalysts.
2.2. Characterization of Resulted Catalysts. The
precursors were calcined at 600 °C. Ni loadings of all catalysts
were in line with the nominal values (Table 1). The N2
adsorption−desorption isotherms of the calcined samples are
shown in Figure S2. All catalysts displayed a type IV isotherms
shape. NiSi-IMP-20 and NiSi-IMP-36 catalysts exhibited a
typical H1 hysteresis behavior, whereas the shape of the
hysteresis loop in NiSi-PS was a superposition of types H1 and
H3. This is because of the formation of fibrous nickel
phyllosilicate over the NiSi-PS catalyst.14a,17 The fibrous
structure also promotes the increase of BET surface area and
pore volume (Table 1). Calcined NiSi-PS exhibited higher
surface area and pore volume than the impregnated catalysts
due to the existence of layered structures.
2. RESULTS AND DISCUSSION
2.1. Formation of Nickle Phyllosilicate. Nickel phyllo-
silicate (tetrahedral/octahedral sheet = 2:1, theoretical Ni
loading is 36 wt %) is a sandwich-like structure with a formula
of Ni3Si4O12H2 (Figure S1). Two sheets of tetrahedral SiO4
sandwich the octahedral Ni2+ species, which are coordinated to
six O atoms or OH groups.14 The material is composed of
fragments of the same layer-like structure, which has large
proportions of edges. An illustration for the dynamic color
changes of suspensions during ammonia evaporation is shown
in Figure S1. With the formation of nickel phyllosilicate, the
dark-blue solution of precursors changed gradually into a light-
green suspension. The impregnated Ni/SiO2 reference with
same Ni loadings of 36 wt % (denoted as NiSi-IMP-36) was
prepared with the same starting materials. The impregnation
method is typically fit for low Ni loading. Therefore, Ni/SiO2
catalyst with Ni loading of 20 wt % (denoted as NiSi-IMP-20)
was also prepared as a reference.
Figure 3a shows the FT-IR spectra of calcined catalysts. The
characteristic bands at 670, 1024, and 3627 cm−1 indicated the
5915
DOI: 10.1021/acscatal.5b01080
ACS Catal. 2015, 5, 5914−5920

ACS Catalysis
Research Article
Table 1. Physicochemical Properties of Catalysts
a
b
b
b
c
d
e
catalyst
SiO2
loading (wt %)
SBET (m2/g)
Vp (cm3/g)
dp (nm)
dNi (nm)
SNi (mmol/g)
Sacid (μmol NH3/g)
-
153.9
352.5
142.3
98.7
0.19
0.76
0.23
0.17
4.7
7.0
5.4
5.4
-
-
-
NiSi-PS
36.1
19.8
34.5
3.3
0.19
0.10
0.02
1.84
0.43
0.31
NiSi-IMP-20
NiSi-IMP-36
15.7
35.1
a
b
Ni loadings were determined by ICP experiments. The BET surface area, pore volume, and pore diameter were determined by N2 physical
c
adsorption. The sizes of Ni were calculated by Ni (200) reflection (2θ = 43.3°) based on the Scherrer equation for impregnated samples and
calculated by HRTEM for NiSi-PS catalyst. SNi was calculated on the basis of H2-TPD results. Sacid was calculated on the basis of NH3-TPD results.
d
e
Figure 4. HRTEM images and particle size distribution histogram of
reduced NiSi-PS.
size distribution histogram showed that NiSi-PS exhibited a
small size of 3.3 nm (Figure 4 and Table 1). Thus, highly
dispersed Ni NPs were formed upon reduction of nickel
phyllosilicate. As evidenced by HRTEM images, the “fibrous”
structure still existed upon reduction. The results indicated the
remnant of nickel phyllosilicate, which was consistent with TPR
Figure 3. (a) FT-IR patterns of calcined catalysts, (b) XRD patterns of
calcined catalysts, (c) TPR results of calcined catalysts, (d) XRD
patterns of reduced catalysts.
results. The unreduced nickel phyllosilicate could act as an
additional host for Ni0 NPs, which would exhibit different
preservation of nickel phyllosilicate after calcination of NiSi-PS
properties with support SiO2 (Figure 5a). Besides, Ni NPs and
precursor, as also evidenced by XRD pattern (Figure 3b). For
NiSi-IMP-36 and NiSi-IMP-20, the disappearance of the band
phyllosilicate were closely contacted, facilitating the binding
at 1385 cm−1 indicated decomposition of nickel nitrates. XRD
results showed the formation of large crystalline NiO particles.
The broad XRD peak around 22° is the feature of amorphous
SiO2. The above results showed that calcined NiSi-PS consisted
of nickel phyllosilicate, whereas impregnated catalysts were
composed of large crystalline NiO and amorphous SiO2.
TPR results can reveal the reduction behavior of Ni2+ species
(Figure 3c). NiSi-IMP-36 and NiSi-IMP-20 catalysts both
exhibited a reduction peak at 350−450 °C and a shoulder
around 500 °C, which were ascribed to the reduction of bulk
NiO and Ni2+ ions interacting with SiO2 respectively. NiSi-PS
showed a broad peak with higher temperatures, which could be
assigned to reduction of Ni2+ located in nickel phyllosilicate.18
The retarded reduction process of Ni species could retain the
phyllosilicate structure partially.
The catalysts were reduced at 500 °C under H2 atmosphere
for 2 h. XRD results suggested that NiSi-IMP-20 and NiSi-IMP-
36 catalysts were totally reduced (Figure 3d). The Ni particle
sizes of NiSi-IMP-20 and NiSi-IMP-36 were 15.7 and 35.1 nm,
respectively (Table 1). In contrast, no evident XRD peaks
Figure 5. Characterizations of surface acidity. (a) Schematic
representation of reduced impregnated Ni/SiO2 catalysts, NiSi-PS
related to metallic Ni were detected for reduced NiSi-PS,
and origin of surface acidity. (b) NH3-TPD results of reduced
indicating that the metallic Ni particles were too small to be
detected by XRD. We thus performed HRTEM to determine
the distribution of metallic Ni (Figure 4). The corresponding
catalysts, (c) FT-IR and NH3-TPD results of calcined and reduced
NiSi-PS catalysts, and (d) Py-IR spectra of calcined and reduced NiSi-
PS catalysts.
5916
DOI: 10.1021/acscatal.5b01080
ACS Catal. 2015, 5, 5914−5920

ACS Catalysis
Research Article
a
Table 2. Catalytic Performance of Catalysts for HMF Hydrogenolysis to Furanic Fuels
a
b
HMF 1.5 g, 1.5 MPa, 1,4-dioxane 38 mL. Others mainly include furfuryl alcohol and 5-methyltetrahydrofurfuryl alcohol.
and dispersion of Ni NPs. H2-TPD results showed that NiSi-PS
had a surface metallic Ni concentration of 0.19 mmol/g (Table
1), which was greatly higher than impregnated references.
Surface acidity of the reduced catalysts was studied by NH3-
TPD (Figure 5b). The peaks around 200 °C were observed for
all the catalysts, corresponding to the weak acid sites. The
impregnated catalysts displayed small NH3 desorption peak
area due to the slight acidity of SiO2 support. In contrast, NiSi-
PS exhibited a large NH3 desorption peak area, indicating the
significant amounts of acid sites. The amounts of acid sites were
determined through NH3-TPD results and shown in Table 1.
To address the origin of acidity, FT-IR and NH3-TPD of
calcined and reduced NiSi-PS were compared. Upon reduction,
the intensities of peaks at 670 and 1024 cm−1 decreased (Figure
5c). The results indicated that phyllosilicate was partially
destroyed after reduction due to the release of Ni0. The partial
destruction caused a decrease of acid concerntration (inset of
Figure 5c), indicating that surface acidity of NiSi-PS was
associated with phyllosilicate structure. The acid sites could
stem from two distinct locations of phyllosilicate: coordina-
tively unsaturated (cus) Ni(II) sites near the edges/surfaces of
phyllosilicate and surface OH groups (Figure 5a). The former
are Lewis acid centers, whereas the latter may acquire proton
donor (Brønsted acid) and hydrogen bond (H-bond) donor
properties.
The nature of acid sites is determined by the pyridine
adsorbed IR (Py-IR) spectra (Figure 5d). The strong bands at
1449, 1610 and a small band at 1490 cm−1 are consistent with
pyridine adsorbed on Lewis acid sites.19 Our data supported
that the acidity was mainly arised from cus-Ni(II) sites located
in the remnant phyllosilicate structure. The cus-Ni(II) sites are
formed due to the broken bonds near the edges/surfaces of
phyllosilicate.20 The absence of a band at 1540 cm−1 ruled out
Brønsted acidity caused by surface OH groups. The bands at
1577 and 1596 cm−1 were observed, which are assigned to
vibration of pyridine interacting with OH groups via H-
bonds.19 Previous work by Lercher et al.3b on C−O cleavage of
substituted phenols shows that the presence of proximal acid
sites increases the activities of catalyst by a synergistic action.
Similarly, the substantially generated acid sites of NiSi-PS could
cooperate with highly dispersed Ni sites and benefit for HMF
hydrogenolysis.
HMF passing through 2,5-dihydroxymethylfuran (DHMF) and
5-methylfurfuryl alcohol (MFA) intermediates (Scheme 1). To
our knowledge, the low-temperature hydrogenolysis has never
been reported over non-noble catalysts. Table 2 shows
performances of the three catalysts for furanic fuel production.
At 150 °C, NiSi-PS exhibited a conversion of 100% and a high
yield of furanic fuels (DMF and DMTHF, 90.2%) within 3 h,
revealing the high activity toward HMF hydrogenolysis (entry
1). NiSi-IMP-20 showed a HMF conversion of 50.9%, whereas
NiSi-IMP-36 gave a conversion of 41.5% (entry 2, 3). Both
catalysts showed low selectivity toward furanic fuels. The major
byproducts are MFA and DHMF, indicating the insufficient
reactivity of impregnated Ni/SiO2 catalysts. One important
reason could be the low amounts of Ni sites, as revealed by H2-
TPD (Table 1). However, with prolonging reaction time to 20
h (entry 4), the hydrogenolysis products (DMF and DMTHF)
were not promoted selectively over NiSi-IMP-20 catalyst,
despite ofthe complete conversion of HMF. Instead, large
amounts of DHMTHF (50.2%) were obtained, revealing that
total hydrogenation of HMF dominates over the impregnated
catalyst at 150 °C. The hydrogenolysis reaction can be
promoted only at high temperature (180 °C, entry 5), which
was consistent with previous reports.12a The results indicated
that although important, the high dispersion cannot account
alone for the superiority of NiSi-PS compared to impregnated
Ni/SiO2 catalysts.
At 130 °C, a furanic fuel yield of 68.0% was received over
NiSi-PS within 3 h (entry 6). The yield can be improved to
78.9% when the reaction time was prolonged to 6 h (entry 7).
High yield of furanic fuels (83.4%) can also be realized after
increasing the catalyst dosage from 0.08 to 0.15 g (entry 8).
Combined with the above characterizations, the results revealed
that the synergy effects of Ni and acid sites over NiSi-PS are
responsible for the superior reactivity. Besides, DHMTHF
selectivity over NiSi-PS increased slightly at 130 °C compared
with that at 150 °C, due to the favorable hydrogenation of
HMF to DHMTHF at lower temperature.
The NiSi-PS catalyst was also compared with impregnated
Ni/Al2O3 (20 wt %), Ni/ZrO2 (20 wt %) and commercial
Raney Ni catalysts (Table S2). Among the reference catalysts,
Ni/Al2O3 showed the highest selectivity of furanic fuels
(38.0%), which was still lower than that for NiSi-PS. Raney
Ni exhibited high selectivity for DHMTHF, demonstrating the
high reactivity toward CC hydrogenation and low reactivity
2.3. Catalytic Behavior. Furanic fuels (DMF and
DMTHF) can be obtained via tandem hydrogenolysis of
5917
DOI: 10.1021/acscatal.5b01080
ACS Catal. 2015, 5, 5914−5920

ACS Catalysis
Research Article
for C−O hydrogenolysis. High selectivity of furanic fuels can be
achieved over Ni/Al2O3 upon prolonging the reaction time.
The characterization results revealed that Ni/Al2O3 has similar
acid concentration with NiSi-PS (Figure S3) and lower
dispersion of Ni species than that of NiSi-PS (Figure S4).
The comparisons hence further highlighted the synergy
between metal and acid sites for C−O hydrogenolysis. Further
optimizations of metal−acid sites over NiSi-PS catalyst are
needed.
NiSi-IMP-20 was chosen as reference to further elucidate the
effect of acid sites over NiSi-PS. The catalytic behavior of NiSi-
PS and NiSi-IMP-20 under low conversions was compared
(Figure 6). To obtain the intrinsic activity of catalysts, the yield
about 3 times that of NiSi-IMP-20 catalyst. Finally, TOFC−O/
TOFHMF was larger over NiSi-PS as compared with that of
NiSi-IMP-20, demonstrating that the hydrogenolysis step could
be promoted more significantly by the synergy of highly
dispersed Ni sites and acid sites over NiSi-PS.
NiSi-PS catalyst was also compared to previously reported
catalysts especially that operated under mild conditions,
including Pd/Zn/C,6 Ru/Co3O4,7 Pd/C with formic acid8
(Table S2). The NiSi-PS, with the moderate HMF/active metal
ratio (52.0), can give similar fuel yields, which was comparable
to the reported noble metal catalysts. In all, we report an
efficient catalytic system for low-temperature hydrogenolysis of
HMF to furanic fuels without usage of noble metals and specific
solvents. Design of an efficient Ni catalyst with lower loadings is
also an important direction.
In terms of the different properties of DMF and DMTHF as
fuels (Table S1), the time course for tunable production of
DMF and DMTHF was tested (Figure 7). DMF yield increased
Figure 6. HMF hydrogenation on NiSi-PS (a, c) and NiSi-IMP-20 (b,
d) catalyst as a function of reaction time. Reaction conditions: 1.5 g of
HMF, 38 mL of 1,4-dioxane, 150 °C, 1.5 MPa H2. NiSi-PS 0.02 g,
NiSi-IMP-20 0.08 g.
Figure 7. Conversion and yield of DMF/DMTHF over NiSi-PS versus
reaction time. Reaction conditions: 1.5 g of HMF, 38 mL of 1,4-
dioxane, 0.08 g of NiSi-PS, 150 °C, 1.5 MPa H2.
first and reached the maximum at 2 h (∼79.3%). Further
increase of reaction time resulted in the conversion of DMF to
DMTHF. Finally, a high yield of DMTHF (89.7%) was
achieved over NiSi-PS catalyst. These results demonstrated the
synthesis of furanic fuels with tunable properties.
of furanic fuels was lower than 20%. Thus, during the
investigated reaction time, MFA and DHMF increased
monotonously. Most importantly, the catalysts exhibited
different product distributions at similar conversions (Figure
6c,d). NiSi-PS exhibited higher DMF yield, whereas NiSi-IMP-
20 showed a higher DHMF yield. For example, at a HMF
conversion of ∼35%, the yield of DMF to DHMF over NiSi-PS
was 1.33; however, the ratio was only 0.48 for NiSi-IMP-20,
indicating the stronger hydrogenolysis ability of NiSi-PS
catalyst. TOFs based on numbers of Ni sites were calculated
for HMF conversion (TOFHMF) and hydrogenolysis of C−O
bonds (TOFC−O), respectively (see Supporting Information for
details). About 2-fold TOFHMF was obtained over NiSi-PS, in
compared with NiSi-IMP-20 (Table 3). Moreover, the attained
TOFC−O value of NiSi-PS catalyst was 1487 h−1, which was
On the basis of the catalytic results and the reported works,21
we proposed the possible mechanism for HMF hydrogenolysis
over NiSi-PS catalyst. The initial hydrogenation of HMF to
DHMF occurred over metallic Ni sites which have strong C
O hydrogenation ability. Lewis acid, presenting in the preserved
nickel phyllosilicate can facilitate polarization and activation of
C−O bonds of DHMF and MFA. Thus, DMF/DMTHF can be
produced through the synergy of Ni and proximal acid sites.12b
DMF/DMTHF can also be generated by direct clevage of C
O bonds. The CO bonds of HMF might be adsrobed onto
the surface via an η2(C,O) configuration. The C atom is
bonded to a metallic Ni site, and the O atom is bonded to acid
sites. The formed CH2OH-(C4H3O)−CH intermediate was
then hydrogenated into furanic fuels.22 In all, compared with
other references, NiSi-PS catalyst provides the required
proximity of metallic and acid sites and catalyzes the
hydrogenolysis of HMF effectively.
Table 3. Parameters in the Hydrogenolysis of HMF under
Low Conversions
DMF/
TOFHMF
TOFC−O
TOFC−O
a
b
c
catalyst
NiSi-PS
DHMF
(h−1
)
(h−1
)
/TOFHMF
1.33
0.48
1404
703
1487
479
1.1
0.7
NiSi-IMP-
20
NiSi-PS was extended to hydrogenolysis of 5-methylfurfural
and furfural (Table 4) which were generally performed at high
temperatures (e.g., 180 °C).5c,23 The catalyst gave a conversion
of 100% and high selectivities toward hydrogenolysis products
at 150 °C. For example, 2-methylfuran (2-MF) and 2-
a
b
The ratio of yield of DMF to DHMF. Calculated on the basis of
c
converted HMF. Calculated on the basis of DMF, DMTHF, and
MFA products.
5918
DOI: 10.1021/acscatal.5b01080
ACS Catal. 2015, 5, 5914−5920

ACS Catalysis
Research Article
NiSi-IMP-20. The reference Ni/Al2O3 and Ni/ZrO2 catalysts
were also prepared by impregnated method with Ni loadings of
20 wt %.
Table 4. Hydrogenolysis of Bioderived Chemicals over the
NiSi-PS Catalyst
a
feedstock
conv. (%)
products (sel. (%))
4.3. Evaluation. The tests were performed in a 100 mL
tank reactor. Prior to the test, the catalysts were reduced in a
quartz tube under H2 flow at 500 °C for 2 h. The reduced
catalysts were then protected by 1,4-dioxane and transferred to
the reactor. For a typical procedure based on our previous
works,12 the reactor was fed with HMF (1.5 g), 1,4-dioxane (38
mL), and a certain amount of prereduced catalyst and then
sealed and purged by H2 (5 times). After that, the reactor was
filled with 1.5 MPa H2 and heated to the desired temperature.
The timing was then started at the moment the temperature
reached the desired value. After the test, the reactor was
quenched in ice−water, and then the liquid and gas products
were analyzed by a GC instrument with a FID detector and a
capillary column (J&W DB-WAX). The conversion and
selectivity were determined by calibrated area normalization.
4.4. Catalyst Characterization. The catalysts were
characterized by inductively coupled plasma-mass spectrometry
(ICP), N2 physical adsorption, X-ray diffraction (XRD), high-
resolution transmission electron microscope (HRTEM), infra-
red spectra (IR), pyridine adsorbed IR (Py-IR), temperature-
programmed reduction (TPR), H2 temperature-programmed
desorption (H2-TPD), and NH3 temperature-programmed
desorption (NH3-TPD). More details could be found in
Supporting Information.
5-methylfurfural
furfural
100
100
DMF (15.7), DMTHF (76.9)
2-MF (0), 2-MTHF (78.3), THFA (17.4)
a
1.5 MPa, 150 °C, 3 h, feedstock 1.5 g, catalyst 0.08 g 1,4-dioxane 38
mL.
methyltetrahydrofuran (2-MTHF) were obtained with a total
yield of 78.3% from furfural. The major byproduct was
tetrahydrofurfuryl alcohol (THFA). The results showed the
great potential of NiSi-PS for low-temperature hydrogenolysis
reactions.
3. CONCLUSION
We have reported that nickel nanoparticles (NPs) inlaid nickel
phyllosilicate (NiSi-PS) catalyst could effectively catalyze
hydrogenolysis of HMF to yield furanic fuels at low
temperatures (130−150 °C), whereas supported Ni/SiO2
catalyst only shows weak hydrogenolysis ability. At 150 °C,
the overall yield of furanic fuels can be up to 90.2% within 3 h,
whereas the yield could also be around 83.4% at 130 °C. In
particular, NiSi-PS shows a higher HMF conversion intrinsic
rate (1404 h−1) and hydrogenolysis rate (1487 h−1) in
comparison with supported Ni/SiO2 (703 h−1 and 479 h−1,
respectively). We have clarified that synergy of highly dispersed
nickel NPs and substantially formed acid sites due to
coordinatively unsaturated Ni (II) sites located at remnant
nickel phyllosilicate structure mainly contributes to the
realization of low-temperature hydrogenolysis. In addition,
the NiSi-PS demonstrated a great potential for low-temperature
hydrogenolysis of furfural and 5-methylfurfural. The attractive
features of NiSi-PS might be also promising for hydrogenolysis
of lignin-derived model compounds, providing guiding
principles for future catalyst design for biomass conversion.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.5b01080.
Experimental details, Table S1 to S3, Figure S1 to S4
(PDF)
AUTHOR INFORMATION
■
4. EXPERIMENTAL SECTION
Corresponding Author
*(Y. Zhu) E-mail: zhuyulei@sxicc.ac.cn. Tel.: +86 351
7117097. Fax: +86 351 7560668.
4.1. Chemicals. The following materials have been used for
the preparation of catalysts: nickel(II) nitrate hexahydrate
(Sinopharm. Co. Ltd.), silica sol (Qingdao Ocean Chemical
Co., Ltd.), and ammonia aqueous solution (Sinopharm. Co.
Ltd.). 5-Hydroxymethylfurfural was purchased from Shanghai
De-Mo Pharmaceutical Science and Technology Co. Ltd. 1,4-
Dioxane was purchased from Sinopharm. Co. Ltd. 5-
Methylfurfural was purchased from T.C.I. These chemicals
were used without further purification. Furfural was used after
distillation.
4.2. Catalyst Preparation. NiSi-PS catalyst was prepared
by a modified ammonia evaporation procedure.14a A certain
amount of Ni(NO3)2·6H2O and a 25 wt % ammonia aqueous
solution dissolved in water were mixed and stirred for 15 min.
Silica sol (40 wt %) was then added to the solution and stirred
for 12 h at room temperature. Then the suspension was heated
in a water bath preheated to 80 °C to evaporate ammonia until
the pH value of the suspension decreased to 6−7. Then the
precipitate was filtered, washed with distilled water, and then
dried at 80 °C for 12 h. A green material was received and
calcined at 600 °C in air. For detailed comparison, two
supported Ni/SiO2 catalysts were prepared by impregnation
method with a Ni loading of 36 and 20 wt % respectively. The
support SiO2 for impregnation was obtained by drying silica sol
gel at 80 °C. The catalysts were denoted as NiSi-IMP-36 and
Author Contributions
⊥These authors contributed equally (X.K. and Y.Z.).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. G. Ding for helpful suggestions. This project was
supported by the Major State Basic Research Development
Program of China (No. 2012CB215305).
REFERENCES
■
(1) Kennedy, G.; Baker, L. R.; Somorjai, G. A. Angew. Chem. 2014,
126, 3473−3476.
(2) (a) Vasiliadou, E. S.; Heracleous, E.; Vasalos, I. A.; Lemonidou, A.
A. Appl. Catal., B 2009, 92, 90−99. (b) Zhu, Y.; Kong, X.; Cao, D.-B.;
Cui, J.; Zhu, Y.; Li, Y.-W. ACS Catal. 2014, 4, 3675−3681.
(3) (a) Roman
Nature 2007, 447, 982−985. (b) Song, W.; Liu, Y.; Barat
́
-Leshkov, Y.; Barrett, C. J.; Liu, Z. Y.; Dumesic, J. A.
h, E.; Zhao,
́
C.; Lercher, J. A. Green Chem. 2015, 17, 1204−1218. (c) Serrano-Ruiz,
J. C.; Dumesic, J. A. Energy Environ. Sci. 2011, 4, 83−99. (d) Chheda, J.
N.; Huber, G. W.; Dumesic, J. A. Angew. Chem., Int. Ed. 2007, 46,
7164−7183.
5919
DOI: 10.1021/acscatal.5b01080
ACS Catal. 2015, 5, 5914−5920

ACS Catalysis
Research Article
(4) (a) Grochowski, M. R.; Yang, W.; Sen, A. Chem. - Eur. J. 2012, 18,
12363−12371. (b) Chidambaram, M.; Bell, A. T. Green Chem. 2010,
12, 1253−1262.
(5) (a) Wang, G.-H.; Hilgert, J.; Richter, F. H.; Wang, F.; Bongard,
H.-J.; Spliethoff, B.; Weidenthaler, C.; Schuth, F. Nat. Mater. 2014, 13,
̈
293−300. (b) Jae, J.; Zheng, W. Q.; Lobo, R. F.; Vlachos, D. G.
ChemSusChem 2013, 6, 1158−1162. (c) Scholz, D.; Aellig, C.;
Hermans, I. ChemSusChem 2014, 7, 268−275.
(6) Saha, B.; Bohn, C. M.; Abu-Omar, M. M. ChemSusChem 2014, 7,
3095−3101.
(7) Zu, Y.; Yang, P.; Wang, J.; Liu, X.; Ren, J.; Lu, G.; Wang, Y. Appl.
Catal., B 2014, 146, 244−248.
(8) Thananatthanachon, T.; Rauchfuss, T. B. Angew. Chem. 2010,
122, 6766−6768.
(9) (a) Celic, T. B.; Grilc, M.; Likozar, B.; Tusar, N. N.
ChemSusChem 2015, 8, 1703−1710. (b) Grilc, M.; Likozar, B.;
Levec, J. Appl. Catal., B 2014, 150−151, 275−287. (c) Sergeev, A. G.;
Hartwig, J. F. Science 2011, 332, 439−443. (d) Sergeev, A. G.; Webb, J.
D.; Hartwig, J. F. J. Am. Chem. Soc. 2012, 134, 20226−20229.
(e) Grilc, M.; Likozar, B.; Levec, J. Biomass Bioenergy 2014, 63, 300−
312.
(10) Zhao, C.; Lercher, J. A. Angew. Chem. 2012, 124, 6037−6042.
(11) Huang, Y. B.; Chen, M. Y.; Yan, L.; Guo, Q. X.; Fu, Y.
ChemSusChem 2014, 7, 1068−1072.
(12) (a) Kong, X.; Zhu, Y.; Zheng, H.; Dong, F.; Zhu, Y.; Li, Y.-W.
RSC Adv. 2014, 4, 60467−60472. (b) Kong, X.; Zheng, R.; Zhu, Y.;
Ding, G.; Zhu, Y.; Li, Y.-W. Green Chem. 2015, 17, 2504−2514.
(13) Chen, J.; Liu, R.; Guo, Y.; Chen, L.; Gao, H. ACS Catal. 2015, 5,
722−733.
(14) (a) Zhang, C.; Yue, H.; Huang, Z.; Li, S.; Wu, G.; Ma, X.; Gong,
J. ACS Sustainable Chem. Eng. 2012, 1, 161−173. (b) Burattin, P.; Che,
M.; Louis, C. J. Phys. Chem. B 1998, 102, 2722−2732.
(15) Kermarec, M.; Carriat, J. Y.; Burattin, P.; Che, M.; Decarreau, A.
J. Phys. Chem. 1994, 98, 12008−12017.
(16) Cavani, F.; Trifiro,
301.
̀
F.; Vaccari, A. Catal. Today 1991, 11, 173−
(17) (a) Chen, L.; Guo, P.; Qiao, M.; Yan, S.; Li, H.; Shen, W.; Xu,
H.; Fan, K. J. Catal. 2008, 257, 172−180. (b) Yue, H.; Zhao, Y.; Zhao,
S.; Wang, B.; Ma, X.; Gong, J. Nat. Commun. 2013, 4, 2339−2346.
(18) Zhang, C.; Zhu, W.; Li, S.; Wu, G.; Ma, X.; Wang, X.; Gong, J.
Chem. Commun. 2013, 49, 9383−9385.
(19) Zaki, M. I.; Hasan, M. A.; Al-Sagheer, F. A.; Pasupulety, L.
Colloids Surf., A 2001, 190, 261−274.
(20) (a) Peri, J. Discuss. Faraday Soc. 1966, 41, 121−134. (b) Giese,
R.; Costanzo, P.; Van Oss, C. Phys. Chem. Miner. 1991, 17, 611−616.
(c) Shirai, M.; Aoki, K.; Torii, K.; Arai, M. Appl. Catal., A 1999, 187,
141−146. (d) Arai, M.; Kanno, M.; Nishiyama, Y.; Torii, K.; Shirai, M.
J. Catal. 1999, 182, 507−510.
(21) (a) Panagiotopoulou, P.; Martin, N.; Vlachos, D. G.
ChemSusChem 2015, 8, 2046−2054. (b) Gilkey, M. J.;
Panagiotopoulou, P.; Mironenko, A. V.; Jenness, G. R.; Vlachos, D.
G.; Xu, B. ACS Catal. 2015, 5, 3988−3994.
(22) Shi, D.; Vohs, J. M. ACS Catal. 2015, 5, 2177−2183.
(23) (a) Stevens, J. G.; Bourne, R. A.; Twigg, M. V.; Poliakoff, M.
Angew. Chem., Int. Ed. 2010, 49, 8856−8859. (b) Sitthisa, S.; An, W.;
Resasco, D. E. J. Catal. 2011, 284, 90−101.
5920
DOI: 10.1021/acscatal.5b01080
ACS Catal. 2015, 5, 5914−5920