Renewable fuels from biomass-derived compounds: Ru-containing hydrotalcites as catalysts for conversion of HMF to 2,5-dimethylfuran

Article abstract of DOI:10.1039/c4cy01376j

Production of transportation fuels from renewable biomass is hugely important considering the current ecological concerns over CO₂ built up in the atmosphere. Ruthenium-containing hydrotalcite (HT) catalysts were applied for the selective hydrogenolysis of biomass-derived 5-hydroxymethylfurfural (HMF) to 2,5-dimethylfuran (DMF). Structural and morphological features of the catalysts were examined using various physico-chemical characterization techniques. The influence of various reaction parameters, such as reaction temperature, solvent, Ru content of the catalyst, etc., was investigated with respect to HMF conversion and DMF yield. The study clearly shows that well-dispersed Ru nanoparticles are highly active and selective in the conversion of HMF to DMF. A catalyst containing only 0.56 wt% Ru converted 100 mol% HMF to yield 58 mol% DMF. This catalyst was found to be recyclable as the activity was retained even after five cycles of reaction. 2-Propanol was found to be a good solvent as it helped to improve DMF yield through transfer hydrogenation. Based on the results of the investigations, a reaction pathway for the conversion of HMF to DMF was proposed for the present Ru-based catalyst system.

Full text of DOI:10.1039/c4cy01376j

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Renewable fuels from biomass-derived
compounds: Ru-containing hydrotalcites
as catalysts for conversion of HMF
to 2,5-dimethylfuran†
Cite this: DOI: 10.1039/c4cy01376j
Atul S. Nagpure, Ashok Kumar Venugopal, Nishita Lucas, Marimuthu Manikandan,
Raja Thirumalaiswamy* and Satyanarayana Chilukuri*
Production of transportation fuels from renewable biomass is hugely important considering the current
2
ecological concerns over CO built up in the atmosphere. Ruthenium-containing hydrotalcite (HT)
catalysts were applied for the selective hydrogenolysis of biomass-derived 5-hydroxymethylfurfural (HMF)
to 2,5-dimethylfuran (DMF). Structural and morphological features of the catalysts were examined using
various physico-chemical characterization techniques. The influence of various reaction parameters, such
as reaction temperature, solvent, Ru content of the catalyst, etc., was investigated with respect to HMF
conversion and DMF yield. The study clearly shows that well-dispersed Ru nanoparticles are highly active
and selective in the conversion of HMF to DMF. A catalyst containing only 0.56 wt% Ru converted
Received 20th October 2014,
Accepted 25th November 2014
1
00 mol% HMF to yield 58 mol% DMF. This catalyst was found to be recyclable as the activity was retained
even after five cycles of reaction. 2-Propanol was found to be a good solvent as it helped to improve DMF
yield through transfer hydrogenation. Based on the results of the investigations, a reaction pathway for the
conversion of HMF to DMF was proposed for the present Ru-based catalyst system.
DOI: 10.1039/c4cy01376j
www.rsc.org/catalysis
to blend with gasoline compared to ethanol. Biomass-derived
DMF has been tested as a biofuel in a single-cylinder gaso-
1
. Introduction
1
3
At present, there is an alarmingly heavy dependence on fossil
fuels, which is not sustainable. Moreover, their indiscriminate
use is leading to ecological problems. Hence, utilization of
biomass to produce renewable fuels and valuable chemical
line direct-injection research engine.
The performance
of DMF was satisfactory against gasoline in terms of igni-
tion, emission and combustion characteristics. These attri-
butes bode very well for the use of DMF as an alternative
fuel for transportation.
1–4
intermediates is gaining increasing attention.
Hydro-
genolysis is an important process in biomass refinement, as
There were several recent reports on the conversion of bio-
mass to DMF. Dumesic and co-workers utilized a two-step
process to convert fructose to DMF. The first step involved
5
,6
biomass-derived materials have high oxygen content.
-Hydroxymethylfurfural (HMF), an important platform chem-
8
5
ical that can be synthesized from hexoses, has been identified
as a key player in the bio-based renaissance. It can be
converted to levulinic acid, ethyl levulinate, γ-valerolactone
and the highly promising transportation fuel additive
the dehydration of fructose to HMF using HCl in a biphasic
solvent system, followed by vapor-phase hydrogenolysis of
HMF to DMF, using a Cu–Ru/C catalyst. Thananatthanachon
and Rauchfuss showed a milder pathway for the production
of DMF using formic acid as a reagent and Pd/C as a
7
–12
2
,5-dimethylfuran (DMF).
DMF is particularly attractive
−
3
14
due to its superior energy density (30 kJ cm ), high research
octane number (RON = 119) and ideal boiling point (92–94 °C).
catalyst. Formic acid functioned as a hydrogen donor dur-
8
ing the second step and also assisted in the deoxygenation of
HMF to DMF. To get a high yield of DMF, formic acid and
H SO have to be used simultaneously. But, both acids are
Further, DMF is immiscible with water and also easier
2
4
not environmentally friendly. Chidambaram and Bell used
Pd/C as a catalyst in ionic liquids to convert 47 mol% HMF,
Catalysis Division, CSIR - National Chemical Laboratory, Dr. Homi Bhabha Road,
Pune-411008, India. E-mail: sv.chilukuri@ncl.res.in, t.raja@ncl.res.in;
Fax: +91 20 25902633; Tel: +91 20 25902019
15
yielding 15 mol% DMF.
A potential drawback of this
in ionic liquids. As a
method is the low solubility of H
2
result, a high pressure of H (62 bar) has to be used, making
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
2
c4cy01376j
the process highly energy intensive.
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Hansen et al. reported the catalytic transfer hydrogena-
tion (CTH) of HMF over Cu-containing mixed metal oxides
1
6
using supercritical methanol that yielded 48 mol% DMF.
Gallo et al. studied the hydrogenolysis of HMF in the presence
of lactones using a RuSn/C catalyst to obtain DMF yields up
1
7
to 46 mol%. Yang and Sen reported the conversion of
biomass-derived carbohydrates to 2,5-dimethyltetrahydrofuran
(
DMTHF), an alternative fuel that can be obtained by further
hydrogenation of DMF, using homogeneous RhCl and RuCl
The same research group has also reported the
synthesis of 5-methyl furfural (MF) from fructose using hetero-
3
3
1
8,19
catalysts.
2
0
geneous Pd/C as a catalyst. Morikawa et al. studied the CTH
of HMF using cyclohexane over AlCl and Pd/C catalysts with a
3
2
1
DMF yield of 60 mol%. These CTH routes, however, have
several disadvantages including the use of mineral acids as
1
4,22,23
co-catalysts to enhance hydrogenation activity.
Recently, it was suggested that the use of catalysts with a
high amount of precious metal is not conducive to the overall
process economics of the production of liquid transportation
2
4
fuels from biomass-derived compounds. Commercial appli-
cation of the HMF to DMF process will be feasible only by
developing a strategy that minimizes the precious metal con-
tent of the catalyst. The Ru-based catalysts are known for
their effectiveness in the hydrogenolysis of polyols to
Fig. 1 X-ray diffractograms of (a) the as-synthesized precursors and
(b) the calcined–reduced catalysts.
2
5,26
alkanes.
Hence, in the present report we have demon-
strated for the first time, the efficiency of a Ru-doped hydro-
talcite (HT) catalyst with very low (0.56 wt%) Ru content for
the hydrogenolysis of HMF to DMF. This catalyst showed
high catalytic activity and DMF selectivity, in addition to its
excellent reusability for the conversion of HMF to DMF. The
effect of various reaction parameters on the performance of the
catalyst was also investigated by varying the solvent, reaction
Ru content similar to that of RH-1. The reflections belonging to
the periclase phase of Mg(Al)O (JCPDS no. 4-829) were seen pre-
dominantly. In addition to the mixed metal oxide phase,
metallic Ru peaks were also present in the case of the catalyst
with higher Ru content (RH-3). The BET surface areas of
calcined HT (HTcal) and calcined–reduced Ru-containing
temperature, H
2
pressure and the Ru metal content in the
catalysts are given in Table 1. The surface areas of HTcal
,
catalyst. These Ru-based HT catalysts have significant potential
for further development and are expected to pave the way for
realising the goal of renewable liquid fuels from biomass.
RH-1, RH-2, RH-3 and RH-imp were 210, 194, 180, 142 and
2
−1
185 m g , respectively. The surface area decreased with
increasing Ru content of the sample, when compared to
HTcal. The decrease in surface area may either be attributed
to the poor crystallinity, as reflected in lower XRD intensities
2
. Results and discussion
(
Fig. 1), or to the blockage of pores by the segregated Ru
2
.1. Structural and morphological characteristics
27
oxide phases of the sample.
of the catalysts
The morphologies of the HT and Ru-doped HT precursors
were investigated by SEM. Some of their representative
images are shown in Fig. 2. The micrographs demonstrate
The structural, textural and morphological characteristics
of the as-synthesized Ru-doped HT samples (RH-1, RH-2 and
RH-3) along with the corresponding HT precursors
were investigated. Fig. 1a shows the XRD profiles of the
as-synthesized samples, which have characteristic peaks belong-
ing to pure HT IJdIJ003) = 7.65 Å). No extraneous peaks belong-
ing to other phases (JCPDS no. 70-2151) were seen. These
results show that Ru may be present in the brucite-like lattice
of HT, as no peaks corresponding to any of ruthenium or
its oxide phases were seen. The intensities of reflections
pertaining to the layered structure decreased with increasing
Ru content of the sample. The XRD patterns of the calcined–
reduced Ru-doped HT-derived catalysts (RH-1, RH-2 and
RH-3) are given in Fig. 1b, which also includes the spectra of
the Ru-impregnated sample (RH-imp). The RH-imp sample has
2
8
the flower-like morphology of parent HT, which has not
changed even after Ru incorporation. Dispersion of Ru in the
Ru-containing catalysts was investigated by H2 chemisorp-
tion. The Ru metal dispersion, average crystallite size and
metal surface areas are included in Table 1. The metal dis-
persions were: RH-1, 47.6%; RH-2, 33.1%; RH-3, 10.3% and
RH-imp, 32.4%. Good Ru dispersion in the RH-1 catalyst
(47.6%) confirms the homogeneous distribution of Ru on
the metal oxide support with an average Ru crystallite size
of 2.8 nm (Table 1). The Ru particle sizes as well as its distri-
bution were also investigated by TEM. These results are
shown in Fig. 3. The micrograph of RH-1 shows that Ru
nanoparticles are in the 2–6 nm range, dispersed over the
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
Table 1 Chemical composition and structural characteristics of the catalysts
Synthesis
composition
Mg : Al : Ru
Average Ru
crystallite
size (nm)
Average Ru
particle
size (nm)
b
Ru content
(wt%)
BET surface
area (m g
Ru metal d
Ru metal
surface area (m g )
a
c
2
−1
d
e
d
2
−1
Catalyst
)
dispersion (%)
HTcal
RH-1
RH-2
RH-3
RH-imp
3 : 1:0
0
210
194
180
142
185
—
—
—
—
3 : 0.97 : 0.03
3 : 0.94 : 0.06
3 : 0.90 : 0.10
3 : 0.97 : 0.03
0.56
1.0
1.7
0.58
47.6
33.1
10.3
32.4
2.8
4.0
12.9
4.1
3.1
4.3
14.0
—
0.97
1.20
0.56
1.18
f
a
b
c
d
Catalyst precursors prepared by co-precipitation. Estimated by ICP-OES. Calcined–reduced samples. Determined by H
2
chemisorption.
e
The catalysts were calcined at 450 °C and reduced at 350 °C for metal dispersion studies. Calculated based on the surface area averaged TEM
particle size. Prepared by the dry impregnation method.
f
Fig. 2 SEM images of as-synthesized (a) HT, (b) RH-1, (c) RH-2 and (d) RH-3 catalyst.
HT-derived metal oxide support. The average particle size of
Ru was 3.1 nm, calculated through the surface area averaged
TEM particle size by assuming that the Ru particles are hemi-
spherical in shape, with the flat side on the support. These
values for the three catalysts, RH-1, RH-2 and RH-3, are in
2
close agreement with the values obtained from H chemi-
sorption (Table 1; Table S1, ESI†).
Temperature-programmed reduction (TPR) studies helped
to understand the reducibility of the Ru supported on mixed
oxides. TPR profiles (Fig. 4) showed two H2 consumption
Fig. 3 TEM micrograph and Ru particle size distribution of the RH-1
catalyst.
Fig. 4 TPR profiles of various Ru catalysts.
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
peaks. The first reduction peak is in the range of 120–190 °C,
while the second is in the 200–330 °C range. These peaks
x
must be due to the different interactions of the RuO species
2
9
with the support. The low temperature peak is probably
due to the weak interaction of Ru with the support, while the
high temperature peak may be due to the stronger interac-
tion of Ru with the support. In the case of RH-1, the intensity
of the low temperature peak was weak. But, its intensity
enhanced with increasing Ru content, implying that the con-
centration of loosely bound Ru increased. There is also a
shift of the high temperature peak towards low temperature
at high Ru content, particularly in the case of the impreg-
nated catalyst. In the case of the RH-imp sample, both low
and high temperature peaks have merged.
Fig. 6 Effect of reaction temperature on DMF yield as a function of
reaction time. Reaction conditions: HMF (1 mmol, 126 mg); catalyst
2
(RH-1, 50 mg); H pressure (10 bar); solvent (2-propanol, 25 mL);
2
.2. Catalytic activity in hydrogenolysis of HMF to DMF
.2.1. Effect of reaction temperature. The influence of
stirring speed (500 rpm).
2
reaction temperature on HMF conversion and DMF yield over
the RH-1 catalyst was systematically investigated by varying
the reaction temperature in the 180–230 °C temperature
range. The HMF conversions and DMF yields are shown in
Fig. 5 and 6, respectively. It can be clearly seen (Fig. 5) that
the temperature played an important role with regard to
HMF conversion, as it has increased with increasing reaction
temperature. In addition, HMF conversion also increased
with reaction time. After 4 h of reaction, HMF conversion
increased from 70 to 100 mol%, as the reaction temperature
increased from 180 to 210 °C. On the other hand, complete
conversion of HMF could be seen within 2 h of the reac-
tion, when the reaction was carried out in the 220–230 °C
temperature range. Moreover, the reaction temperature
has a profound influence on DMF yield, as may be seen in
Fig. 6. The DMF yield increased continuously as a function
of reaction time, when the reaction temperature was
raised from 180 to 210 °C, indicating that intermediates
like 2,5-bisIJhydroxymethyl)furan (BHMF), 5-methyl furfuryl
alcohol (MFA) and MF that formed during the course of the
reaction were converted to DMF with increasing reaction tem-
8
,15
perature as well as with reaction time.
On further increas-
ing the reaction temperature to 220 °C, a high DMF yield of
58 mol% could be reached after 4 h of reaction time. But, the
DMF yield has decreased on further increasing the reaction
time (Fig. 6 and entry 3, Table 3). This decrease in the DMF
yield could be attributed to its ring hydrogenation, leading to
the formation of DMTHF. The DMF yield decreased continu-
ously at higher reaction temperature (230 °C), which clearly
implies that the ring hydrogenation is predominant at higher
reaction temperatures (entries 4, 5 and 6, Table 3), leading
1
6
to the formation of DMTHF as a principal product. Thus,
220 °C seems to be the optimum temperature to obtain a
good yield of DMF after 4 h of reaction with this catalyst.
2.2.2. Effect of solvent. The effect of solvent on the cata-
lytic activity in the liquid-phase hydrogenolysis of HMF to
DMF was investigated over the RH-1 catalyst at 220 °C and at
7
2
bar H pressure. Solvents of different chemical nature, such
as protic IJ2-propanol), aprotic polar (tetrahydrofuran (THF)
and 1,2-dimethoxyethane (1,2-DME)) and non-polar (toluene)
solvents, were used to investigate the influence of solvent on
hydrogenolysis activity and DMF selectivity. The results given
in Fig. 7 clearly show that the RH-1 activity is heavily solvent
dependent. It can be seen that the catalytic activity follows
the order: 2-propanol > THF ≈ 1,2-DME > toluene. The
lower HMF hydrogenolysis activity in toluene could be
explained on the basis of the competitive adsorption between
the reactant and the solvent on the active catalytic sites.
Moreover, with toluene as a solvent, hydrogenated com-
pounds of toluene were observed under the reaction condi-
tions studied. Competitive adsorption between toluene and
HMF on the active catalytic sites results in reduced availabil-
ity of active sites for HMF. As a result, the hydrogenolysis
activity of HMF drops down. The strength of toluene adsorp-
tion is controlled by the degree of overlap of the carbon π
Fig.
function of reaction time. Reaction conditions: HMF (1 mmol, 126 mg);
catalyst (RH-1, 50 mg); H pressure (10 bar); solvent (2-propanol,
5 mL); stirring speed (500 rpm).
5 Effect of reaction temperature on HMF conversion as a
2
3
0
2
molecular orbitals with the d bands of the Ru metal.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
that of CTH (difference in DMF yield in Fig. 7 and 8). To
understand this aspect further, the effect of temperature, the
effect of Ru content and the effect of H pressure were inves-
2
tigated with THF as a solvent, which is not expected to be
reactive in the given conditions. These results are given in
the ESI (Fig. S3, S4 and S5†). Unlike with 2-propanol, varia-
tion of any of the reaction parameters did not lead to a
change in DMF yield, clearly demonstrating the advantage
of 2-propanol as a solvent. In view of the higher apparent
catalytic activity and superior DMF yield, 2-propanol was
chosen as the solvent for further investigations. It is well
known that there are differences between heterogeneous
catalytic hydrogenation using hydrogen donor molecules as
the source of hydrogen and hydrogenation using molecular
Fig. 7 Effect of solvent on DMF yield as a function of reaction time.
Reaction conditions: HMF (1 mmol, 126 mg); catalyst (RH-1, 50 mg);
2
temperature (220 °C); solvent (25 mL); H pressure (7 bar); stirring
speed (500 rpm).
3
1
H
2
.
The CTH reaction could occur through direct hydride
transfer from 2-propanol to HMF. Further investigations are
required to delineate the mechanism for hydrogenolysis
using 2-propanol as the hydrogen source.
To understand the reason behind the superior activity in
the presence of 2-propanol, experimental runs were carried
out in the absence of H at 220 °C under 5 bar N pressure
2 2
over RH-1, RH-2, RH-3 and RH-imp catalysts (Fig. S1, ESI†).
Catalyst RH-1 exhibited superior activity in terms of CTH,
compared to other catalysts, probably due to the smaller Ru
crystallite size. The results of the CTH experiments, using
Compounds: 5-hydroxymethylfurfural (HMF); 2,5-
bisIJhydroxymethyl)furan (BHMF); 5-methyl furfural (MF);
5-methyl furfuryl alcohol (MFA); furfuryl alcohol (FA); 2,5-
bisIJhydroxymethyl)tetrahydrofuran (BHMTHF); 5-methyl-
tetrahydrofurfuryl alcohol (MTHFA); 5,5′-IJoxybisIJmethylene))-
bisIJ2-methylfuran) (OMBM); 2,5-dimethylfuran (DMF); cis-2,5-
dimethyltetrahydrofuran (cis-DMTHF); 2-methylfuran (MFU);
trans-2,5-dimethyltetrahydrofuran (trans-DMTHF); 2-methyl-
tetrahydrofuran (MTHF).
RH-1 in the absence of H , with different solvents are given
2
in Fig. 8. About 14 mol% DMF yield was seen after 5 h of
reaction with 2-propanol as a solvent. Acetone was detected
as one of the products in this experiment, implying hydrogen
transfer from 2-propanol. However, the 2-propanol consumed
for the purpose of CTH was low (<1 mol%), as it was used as
a solvent at a high solvent-to-substrate molar ratio (>300).
No CTH was seen when THF or 1,2-DME was used as a sol-
vent under similar conditions. These results strongly suggest
that in 2-propanol, HMF is hydrogenated over the RH-1 cata-
lyst via an additional reduction mechanism involving hydro-
gen transfer from 2-propanol to HMF in the presence of the
2.2.3. Effect of Ru metal content of catalysts. To optimise
the Ru content of the catalyst, a series of catalysts with differ-
ent wt% of Ru were prepared and evaluated for HMF hydro-
genolysis. The HMF conversions and product yields are sum-
marized in Table 2 (Fig. S2, ESI†). Catalyst RH-1 showed the
highest activity in terms of HMF hydrogenolysis, yielding
58 mol% DMF at 100 mol% HMF conversion (entry 1,
Table 2). However, under similar reaction conditions, DMF
yields were only 48, 35 and 45 mol% over RH-2, RH-3 and
RH-imp catalysts, respectively. When Ru content was low
(RH-1 with 0.56 wt% Ru), smaller yields of the hydrogenated
products, BHMF and DMF, were observed, leading to high
yield of DMF. On the other hand, over catalysts with high
Ru content (RH-2 and RH-3), a significant increase in
the yields of ring hydrogenated products such as 2,5-
bisIJhydroxymethyl)tetrahydrofuran (BHMTHF) and DMTHF
was observed. This could be a result of the larger Ru crystal-
lite size in these catalysts (Table 1). However, the metal
surface areas of RH-2 and RH-imp were higher than that of
RH-1 (Table 1). Even then, RH-1 was found to be superior,
as the smaller crystallite size (2.8 nm) suppresses ring hydro-
genation, thus leading to a better DMF yield over RH-1. This
is a welcome result, as the hydrotalcite catalyst with low Ru
content (RH-1) can achieve a high DMF yield (58 mol%),
which may be helpful for the economic production of trans-
portation fuels from biomass.
3
0
Ru metal catalyst. Thus, HMF is hydrogenated by molecular
as well as by hydrogen transfer from 2-propanol over the
H
2
catalyst (Fig. S2, ESI†). However, the rate of HMF hydrogena-
tion using molecular H is comparatively much higher than
2
Fig. 8 CTH of HMF as a function of reaction time over the RH-1 cata-
lyst. Reaction conditions: HMF (1 mmol, 126 mg); catalyst (50 mg);
To ascertain the desirability of Ru in the catalyst, HMF
hydrogenolysis was carried out in its absence, using only
HTcal, which gave <1 mol% DMF yield (entry 5, Table 2).
2
temperature (220 °C); solvent (25 mL); N pressure (5 bar); stirring
speed (500 rpm).
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 2 Product distributions during HMF hydrogenolysis over different catalystsa
Yield (mol%)
HMF conv.
b
c
−1
Entry
Catalyst
(mol%)
DMF
DMTHF
BHMF
MFA
MF
BHMTHF
MFU
Others
TOF (h
)
1
2
3
4
5
6
RH-1
RH-2
RH-3
RH-imp
HTcal
none
100
100
92
100
32
58
48
35
45
<1
0
6
10
9
8
0
5
8
8
7
1
0
8
11
8
12
1
4
5
4
4
0
0
2
6
15
7
0
0
3
1
2
2
0
0
14 (3)
11 (4)
11 (7)
15 (4)
29 (29)
8 (8)
52.3
24.4
10.4
39.0
—
8
0
0
—
a
2
Reaction conditions: HMF (1 mmol, 126 mg); catalyst (50 mg); temperature (220 °C); H pressure (10 bar); solvent (2-propanol, 25 mL); reaction
b
time (4 h); stirring speed (500 rpm). It includes 5-methyltetrahydrofurfuryl alcohol (MTHFA), furfuryl alcohol (FA), 5,5′-IJoxybisIJmethylene))bisIJ2-
methylfuran) (OMBM), 2,5-bisIJhydroxymethyl)tetrahydrofuran (BHMTHF) and other unidentified products (condensation products, values in
brackets correspond to the unidentified products). TOF = turnover frequency (moles of DMF produced per mole of Ru).
c
Blank experiments with HMF in 2-propanol, in the absence
of a catalyst, showed no activity for DMF formation (entry 6,
Table 2). But, minor amounts of by-products, probably as a
result of HMF polymerization and condensation reactions,
were observed in the absence of the catalyst.
(MTHFA) increased at higher pressure (13 bar). These results
clearly show that at higher H pressures, ring hydrogenation
2
is the predominant reaction, leading to the lower yield of
DMF (Table 3). Hence, 10 bar of H pressure was found to be
2
the optimum, which was used for further investigations.
2.2.5. Effect of catalyst content. The amount of catalyst
used in the reaction is an important parameter that needs to
be optimized to get high DMF yield. Experiments were
conducted by varying the amount of the RH-1 catalyst from
15 to 70 mg at 220 °C, while maintaining the same substrate
content (Fig. 10). These experiments showed that the DMF
yield increased initially with catalyst content. When the
catalyst was 15 mg, the DMF yield reached a maximum of
40 mol% after 12 h of reaction, which declined on further
increasing the reaction time. When 30 mg of catalyst was
used, the DMF yield reached 46 mol% in a slightly shorter
duration (10 h) of reaction time. With 50 mg of catalyst, a
maximum DMF yield of 58 mol% was accomplished only in
4 h of reaction time, but it decreased when the reaction was
continued further, mostly due to ring hydrogenation of DMF.
However, when the catalyst content was further increased to
70 mg, a much lower DMF yield of 44 mol% was achieved,
which decreased with further increasing time on stream.
These results show that with increased duration of reaction
even at low catalyst content, DMF undergoes consecutive
hydrogenation, leading to DMTHF formation. At higher cata-
lyst content, more active sites are available, which drive the
formation of side products such as MFU and MTHF in addi-
tion to hydrogenated products MTHFA and DMTHF, leading
to a lower DMF yield. Hence, optimization of catalytic sites
and reaction time is essential to obtain the maximum DMF
yield. Based on these studies, 50 mg of catalyst was found to
be the optimum under the given reaction conditions.
2
2 2
.2.4. Effect of H pressure. The influence of H pressure
on the HMF hydrogenolysis was studied by varying the pres-
sure in the range of 1–13 bar. Results of these experiments
are given in Fig. 9. When the reaction was carried out in the
low pressure range (1 to 7 bar), the intermediate products
such as BHMF, MFA and MF formed in significant quantities,
showing that the hydrogenolysis was incomplete. As a result,
only 48 mol% DMF yield was obtained even after 8 h of reac-
2 2
tion under 7 bar H pressure. On increasing the H pressure
to 10 bar, the DMF yield reached the maximum (58 mol%)
over the RH-1 catalyst within 4 h of reaction. But, the DMF
yield decreased on continuation of the reaction for further
duration (Fig. 9 and entry 3, Table 3). Similarly, increasing
the H pressure further to 13 bar has an adverse affect on the
2
DMF yield, mostly due to the increased rate of consecutive
ring hydrogenation of DMF, leading to the formation of
DMTHF in significant quantities (entries 8, 9 and 10, Table 3).
Moreover, the concentration of other unwanted by-products
such BHMTHF and 5-methyltetrahydro-furfuryl alcohol
2.2.6. Proposed reaction pathway. To understand the reac-
tion network of hydrogenolysis of HMF to DMF, studies were
carried out at lower temperature (180 °C) while monitoring
the products as a function of time using the RH-1 catalyst.
The results are shown in Fig. 11. When the reaction time was
increased from 1 to 10 h, the conversion of HMF increased
from 40 to 97 mol% and the concentration of the products
changed in a complex manner. The yield of DMF increased
Fig. 9 Effect of H
time. Reaction conditions: HMF (1 mmol, 126 mg); catalyst (RH-1,
0 mg); temperature (220 °C); solvent (2-propanol, 25 mL); stirring
2
pressure on DMF yield as a function of reaction
5
speed (500 rpm).
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
Table 3 Effect of reaction temperature and hydrogen pressure on DMTHF yielda
Entry
Time (h)
Temp (°C)
DMF yield (mol%)
DMTHF yield (mol%)
Cis : trans ratio of DMTHF
1
2
3
4
5
6
4
4
6
4
6
8
4
2
4
6
4
210
220
220
230
230
230
220
220
220
220
220
36
58
49
38
29
20
38
47
41
35
30
3
6
6.3
6.2
6.7
6.6
6.4
6.5
6.1
6.8
6.6
6.6
6.7
15
17
26
32
2
b
7
c
8
7
c
9
12
21
19
c
1
1
0
1
d
a
Reaction conditions: HMF (1 mmol, 126 mg); catalyst (RH-1, 50 mg); H
2
pressure (10 bar); solvent (2-propanol, 25 mL); stirring speed
b
c
d
(500 rpm).
H
2
pressure (7 bar).
H
2
pressure (13 bar).
H
2
pressure (16 bar).
entire duration of the reaction. The low yield of MF may be
attributed to the fact that it does not form fast enough. Based
on these results, a reaction network of the hydrogenolysis of
HMF to DMF is proposed (Scheme 1).
To validate the proposed reaction sequence in Scheme 1,
hydrogenolysis experiments were conducted with MF as the
starting substrate. The results (Fig. 12) demonstrate that MF
was quite reactive. Within 3 h of reaction time, this interme-
diate was converted principally to MFA and DMF. With
increasing reaction time, the yield of MFA decreased but the
DMF yield increased. The results of this experiment show
that the rate of MF hydrogenation to MFA is fast, with the
subsequent hydrogenolysis of MFA to DMF being very rapid.
Low yields of DMTHF and MFU were also observed, whose
concentration increased with time followed by the formation
of DMF.
Fig. 10 Effect of catalyst content on DMF yield as
of reaction time over the RH-1 catalyst. Reaction conditions: HMF
a function
(
1 mmol, 126 mg); temperature (220 °C); H
2
pressure (10 bar); solvent
(
2-propanol, 25 mL); stirring speed (500 rpm).
The cis- and trans-DMTHF isomers have different physical
properties, e.g. boiling points, which can be distinguished
and quantified by GC-FID. Our experiments show that the
formation of cis-DMTHF is favoured over trans-DMTHF, with
a cis : trans molar ratio of ≈6.5 : 1 under the given experimen-
tal conditions (Table 3). This may be explained in terms of
steric crowding. During the course of the reduction reaction,
addition of the second hydrogen molecule takes place on the
same face as the first would result in less steric hindrance,
whereas the hydrogen molecule may be sterically crowded by
the methyl group, restricting the formation of trans-DMTHF
(
Scheme 2).
.2.7. Recyclability of the catalyst. Catalyst recyclability is
Fig. 11 Conversion of HMF and product yields as a function of time at
80 °C. Left: conversion of HMF. Right: yields of various products.
Reaction conditions: HMF (1 mmol, 126 mg); catalyst (RH-1, 50 mg);
pressure (10 bar); solvent (2-propanol, 25 mL); stirring speed
500 rpm).
2
1
of great importance in order to apply the best found catalytic
system and convert it into an industrial process. The recycla-
bility of the RH-1 catalyst was evaluated by repeating the
reaction with the same catalyst at least five times without any
regeneration/activation (Fig. 13 and S7, ESI†). The results
show that the catalyst remains active even after five cycles,
though a minor drop in DMF yield was observed probably
due to blockage of some catalytic sites. However, on regenera-
tion by calcination–reduction steps, 100% activity was
restored. These results indicate good stability of the catalyst.
The product mixture at the end of each recycle was analyzed
H
(
2
continuously with reaction time, while the yield of BHMF
reached a plateau at 38 mol% after 5 h and then decreased
with increasing reaction time. The MFA yield also showed a
similar trend to that of BHMF, passing through the maxi-
mum at 4 h. Though MF, DMTHF and MFU were also
observed in the product, their yields were low during the
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Scheme 1 Reaction network of the hydrogenolysis of HMF to DMF over the RH-1 catalyst.
Fig. 12 Conversion of MF and product yields as a function of time at
80 °C. Left: conversion of MF. Right: yields of various products.
Reaction conditions: MF (1 mmol, 110 mg); catalyst (RH-1, 50 mg);
pressure (10 bar); solvent (2-propanol, 25 mL); stirring speed
500 rpm).
1
H
(
2
Fig. 13 The recyclability experiments of the RH-1 catalyst in HMF
hydrogenolysis. Reaction conditions: solvent (2-propanol, 25 mL);
molar ratio of HMF to Ru (360); temperature (220 °C); H
2
pressure
(10 bar); reaction time (4 h); stirring speed (500 rpm).
concentration of Ru in the catalyst was similar to that of the
starting catalyst even after five cycles.
3
. Conclusions
The present investigations demonstrate that highly dispersed
Ru-containing mixed metal oxide catalysts can be obtained
by the calcination of hydrotalcite-like precursors, which
were obtained through co-precipitation. As a result, even cata-
lysts with low Ru metal (0.56 wt%) content were highly active
in the conversion of HMF to DMF. The possible reaction
pathway for the conversion of HMF to DMF was explored by
using the intermediate compound MF as a starting material.
Scheme
2 Mechanism for the preferential formation of the
cis-DMTHF isomer over the trans-DMTHF isomer from DMF over the
RH-1 catalyst.
by ICP-OES, for the presence of any Ru due to leaching out of
the catalyst. No such leaching was observed. Moreover, the
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Paper
Analysis of intermediate products at different stages of reac-
tion showed that DMF is formed via BHMF followed by MFA.
Under optimized reaction conditions, a maximum DMF yield
X-ray diffraction patterns were obtained using a PANalytical
X'Pro dual goniometer equipped with an X'celator solid state
detector. Nickel filtered Cu Kα (λ = 1.5406 Å, 40 kV, 30 mA)
radiation was used and the data collection was carried out
using a flat holder in Bragg–Brentano geometry. The data
were recorded in the 2θ range of 5–90° with 0.02° step size.
The surface areas of all the samples were investigated by N2
sorption at −196 °C (Quanta chrome Autosorb IQ). Prior to
sorption, the samples were evacuated at 200 °C for 3 h to a
of 58 mol% was achieved at 220 °C, at 10 bar H pressure
2
with 2-propanol as the solvent. The catalyst can be recycled
without any significant loss in activity. Higher DMF yields
were seen with 2-propanol as the solvent, as a result of hydro-
gen transfer from 2-propanol to HMF over the Ru metal. This
study clearly shows that Ru-containing mixed metal oxide-
derived catalysts have excellent potential for the conversion
of biomass oxygenates to biofuels.
−
3
residual pressure of 2 × 10 torr. The isotherms were ana-
lyzed in the relative pressure IJp/p ) range of 0.05 to 0.30.
o
Hydrogen chemisorption was also conducted using
a
Quantachrome autosorb iQ instrument. Before chemisorp-
4
. Experimental section
tion at 40 °C, the catalyst was reduced in situ in H flow at
2
4
.1. Chemicals
350 °C. The amount of Ru present in the samples was esti-
All the chemicals were reagent grade and used without fur-
ther purification. HMF (99%), DMF (99%), DMTHF (99%),
MFU (99%), MTHF (99%) and THF (98%) were procured from
mated by ICP-OES (Spectro Arcos, FHS-12). The SEM images
of the samples were recorded using a JEOL-JSM-5200 to study
the morphology. The samples were prepared by dispersing
them ultrasonically in isopropyl alcohol and transferring a
portion of it onto a silicon wafer, and then subsequently
dried and gold coated before study. TEM images were
collected using a FEI Technai TF-30 instrument operating at
Sigma-Aldrich. MF (99%) and RuIJNO)IJNO
form Alfa Aesar, while MgIJNO Ĵ6H O, AlIJNO
-propanol and 1,2-dimethoxyethane were purchased from
Loba Chemicals, India.
3
)
3
were purchased
3 2
)
2
3
)
3
Ĵ9H O, toluene,
2
2
300 kV. The samples for TEM measurement were prepared by
4
.2. Preparation of catalysts
placing a droplet of the highly diluted suspension of the sam-
ple in isopropyl alcohol on a carbon-coated copper grid and
allowing to dry at room temperature. TPR studies of the
catalysts were carried out using a Micromeritics Autochem-
Ruthenium-doped HT catalyst precursors were prepared
by the co-precipitation method at a constant pH of 9.5–10.
In a typical synthesis, an aqueous solution containing
2
920 instrument in the temperature range 50–600 °C at a
3 2 2 3 3 2
MgIJNO ) Ĵ6H O (0.25 mol), AlIJNO ) Ĵ9H O (0.25 mol) and
−
1
−
1
heating rate of 5 °C min using 5% H
2
in He as the probe
RuIJNO)IJNO3)3 (1.5 wt%, Ru content 15 mg mL ) was pre-
pared in 50 mL of double distilled deionized water. This solu-
tion was added dropwise to a second solution containing
Na CO (0.125 M) in 50 mL of double distilled deionized
water under vigorous stirring at 30 °C. The pH of the mixture
was maintained constant while adding aqueous 0.25 M
NaOH. The precipitate formed was filtered, washed thor-
oughly and dried at 100 °C for 10 h. Subsequently, it was cal-
gas. The H consumption in the TPR study was estimated
2
quantitatively by a thermal conductivity detector that was cal-
ibrated before the TPR study. Prior to the TPR, the catalyst
was pre-treated at 300 °C for 1 h using 5% oxygen in a
helium gas mixture.
2
3
4.4. Evaluation of catalysts
cined in air at 450 °C for 4 h followed by reduction in H
2
Evaluation of catalysts was carried out using a 100 mL capac-
ity Parr autoclave (SS316). In a typical experiment, 1 mmol
−
1
stream (30 mL min ) at 350 °C for 3 h. These samples with
different Ru contents were designated as RH-1, RH-2 and RH-3.
A similar procedure was adopted for the preparation of the
Mg–Al HT sample with a Mg/Al mole ratio of 3, without
(
126 mg) of HMF, 25 mL of solvent and the required amount
of the freshly reduced catalyst were introduced into the reac-
tor vessel. After closing the reactor, it was purged two to three
times with hydrogen and filled with the same gas to the
required hydrogen pressure. Subsequently, the reaction vessel
was heated under stirring to the required temperature. Dur-
ing the reaction, the liquid samples were withdrawn periodi-
cally and analyzed by a GC (Agilent 7890A) equipped with a
flame ionization detector and a CP Sil 8 CB capillary column
using RuIJNO)IJNO
RH-imp) was prepared by the pore filling dry impregnation
method. For this, an aqueous solution of RuIJNO)IJNO ) was
3 3
) solution. The Ru-impregnated catalyst
(
3
3
added dropwise to the freshly dried HT sample. The resultant
wet solid was initially dried at ambient temperature for 12 h
and subsequently dried at 100 °C for 10 h. Finally, the cata-
lyst was calcined in air at 450 °C for 4 h followed by reduc-
(
30 m length, 0.25 mm diameter). Product identification was
−
1
tion in H (30 mL min ) at 350 °C for 3 h.
2
done using authentic standards and by using GC-MS (Varian,
Saturn 2200).
4
.3. Characterization techniques
The physico-chemical characterization of the catalyst and
support was carried out by X-ray diffraction (XRD), scanning
electron microscopy (SEM), transmission electron microscopy
Acknowledgements
Atul S. Nagpure and Ashok Kumar acknowledge the Council
of Scientific and Industrial Research, New Delhi, for providing
(TEM), temperature-programmed reduction (TPR), etc. The
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
a senior research fellowship. The authors also acknowledge
financial support from CSIR Network project CSC-0122.
15 M. Chidambaram and A. T. Bell, Green Chem., 2010,
12, 1253.
1
6 T. S. Hansen, K. Barta, P. T. Anastas, P. C. Ford and
A. Riisager, Green Chem., 2012, 14, 2457.
References
1
7 J. M. R. Gallo, D. M. Alonso, M. A. Mellmer and
J. A. Dumesic, Green Chem., 2013, 15, 85.
1
G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006,
06, 4044.
1
18 W. Yang and A. Sen, ChemSusChem, 2010, 3, 597.
19 A. Sen and W. Yang, US0307050A1, 2010.
20 W. Yang and A. Sen, ChemSusChem, 2011, 4, 349.
21 S. Morikawa, Noguchi Kenkyusho Jiho, 1980, 23, 39.
22 T. Thananatthanachon and T. B. Rauchfuss, Angew. Chem.,
2010, 122, 6766.
23 S. De, S. Dutta and B. Saha, ChemSusChem, 2012, 5, 1826.
24 D. J. Braden, C. A. Henao, J. Heltzel, C. C. Maravelias and
J. A. Dumesic, Green Chem., 2011, 13, 1755.
25 L. Chen, Y. Zhu, H. Zheng, C. Zhang, B. Zhang and Y. Li,
J. Mol. Catal. A: Chem., 2011, 351, 217.
2
3
A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107, 2411.
J. N. Chheda, G. W. Huber and J. A. Dumesic, Angew. Chem.,
Int. Ed., 2007, 46, 7164.
4
5
K. Shimizu and A. Satsuma, Energy Environ. Sci., 2011, 4, 3140.
E. L. Kunkes, D. A. Simonetti, R. M. West, J. C. Serrano-Ruiz,
C. A. Gärtner and J. A. Dumesic, Science, 2008, 322, 417.
A. J. Ragauskas, C. K. Williams, B. H. Davision, G. Britovsek,
J. Cairney, C. A. Eckert, W. J. Frederick Jr., J. P. Hallett,
D. J. Leak, C. L. Liotta, J. R. Mielenz, R. Murphy, R. Templer
and T. Tschaplinski, Science, 2006, 311, 484.
6
7
8
B. F. M. Kuster, Starch, 1990, 42, 314.
Y. Román-Leshkov, C. J. Barrett, Z. Y. Liu and J. A. Dumesic,
Nature, 2007, 447, 982.
L. Deng, J. Li, D. M. Lai, Y. Fu and Q. X. Guo, Angew. Chem.,
Int. Ed., 2009, 48, 6529.
26 T. Miyazawa, S. Koso, K. Kunimori and K. Tomishige,
Appl. Catal., A, 2007, 318, 244.
27 F. Basile, L. Basini, G. Fornasari, M. Gazzano, F. Trifiro and
A. Vaccari, Chem. Commun., 1996, 2435.
28 Q. Wang, H. H. Tay, Z. Guo, L. Chen, Y. Liu, J. Chang,
Z. Zhong, J. Luo and A. Borgna, Appl. Clay Sci., 2012,
55, 18.
9
1
1
1
1
1
0 J. Q. Bond, D. M. Alonso, D. Wang, R. M. West and
J. A. Dumesic, Science, 2010, 327, 1110.
1 I. T. Horváth, J. Mehdi, V. Fábos, L. Boda and L. T. Mika,
Green Chem., 2008, 10, 238.
2 J. P. Lange, W. D. van de Graaf and R. J. Haan,
ChemSusChem, 2009, 2, 437.
3 S. Song, R. Daniel, H. Xu, J. Zhang, D. Turner, M. L. Wyszynski
and P. Richards, Energy Fuels, 2010, 24, 2891.
4 T. Thananatthanachon and T. B. Rauchfuss, Angew. Chem.,
Int. Ed., 2010, 49, 6616.
29 (a) R. Lanza, S. G. Jaras and P. Canu, Appl. Catal., A, 2007,
325, 57; (b) M. G. Cattania, F. Parmigiani and V. Ragaini,
Surf. Sci., 1989, 211, 1097; (c) S.-H. Lee and D. J. Moon,
Catal. Today, 2011, 174, 10.
30 N. M. Bertero, A. F. Trasarti, C. R. Apesteguía and
A. J. Marchi, Appl. Catal., A, 2011, 394, 228.
31 R. A. W. Johnstone, A. H. Wilby and I. D. Entwistle, Chem.
Rev., 1985, 85, 129.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2014