Catalytic Upgrading of 5-Hydroxymethylfurfural to Drop-in Biofuels by Solid Base and Bifunctional Metal-Acid Catalysts

Article abstract of DOI:10.1002/cssc.201501136

Design and synthesis of effective heterogeneous catalysts for the conversion of biomass intermediates into long chain hydrocarbon precursors and their subsequent deoxygenation to hydrocarbons is a viable strategy for upgrading lignocellulose into distillate range drop-in biofuels. Herein, we report a two-step process for upgrading 5-hydroxymethylfurfural (HMF) to C₉ and C₁₁ fuels with high yield and selectivity. The first step involves aldol condensation of HMF and acetone with a water tolerant solid base catalyst, zirconium carbonate (Zr(CO₃)_x), which gave 92 % C₉-aldol product with high selectivity at nearly 100 % HMF conversion. The as-synthesised Zr(CO₃)_x was analysed by several analytical methods for elucidating its structural properties. Recyclability studies of Zr(CO₃)_x revealed a negligible loss of its activity after five consecutive cycles over 120 h of operation. Isolated aldol product from the first step was hydrodeoxygenated with a bifunctional Pd/Zeolite-β catalyst in ethanol, which showed quantitative conversion of the aldol product to n-nonane and 1-ethoxynonane with 40 and 56 % selectivity, respectively. 1-Ethoxynonane, a low oxygenate diesel range fuel, which we report for the first time in this paper, is believed to form through etherification of the hydroxymethyl group of the aldol product with ethanol followed by opening of the furan ring and hydrodeoxygenation of the ether intermediate. Two-stepping to Biofuels! A recyclable and water tolerant heterogeneous base catalyst produced 92 % C₉-aldol product from 5-hydroxymethylfurfural and acetone in water. Subsequent hydrogenation of the isolated aldol product with a metal-acid Pd/zeolite-β catalyst produced gasoline and diesel range n-nonane and 1-ethoxynonane with an overall 96 % yield.

Full text of DOI:10.1002/cssc.201501136

DOI: 10.1002/cssc.201501136
Full Papers
Catalytic Upgrading of 5-Hydroxymethylfurfural to Drop-in
Biofuels by Solid Base and Bifunctional Metal–Acid
Catalysts
[a]
[a, b]
[b, c]
Ashish Bohre, Basudeb Saha,*
and Mahdi M. Abu-Omar*
Design and synthesis of effective heterogeneous catalysts for
the conversion of biomass intermediates into long chain hy-
drocarbon precursors and their subsequent deoxygenation to
hydrocarbons is a viable strategy for upgrading lignocellulose
into distillate range drop-in biofuels. Herein, we report a two-
step process for upgrading 5-hydroxymethylfurfural (HMF) to
C9 and C11 fuels with high yield and selectivity. The first step in-
volves aldol condensation of HMF and acetone with a water
tolerant solid base catalyst, zirconium carbonate (Zr(CO ) ),
tural properties. Recyclability studies of Zr(CO ) revealed a neg-
3
x
ligible loss of its activity after five consecutive cycles over
120 h of operation. Isolated aldol product from the first step
was hydrodeoxygenated with a bifunctional Pd/Zeolite-b cata-
lyst in ethanol, which showed quantitative conversion of the
aldol product to n-nonane and 1-ethoxynonane with 40 and
56% selectivity, respectively. 1-Ethoxynonane, a low oxygenate
diesel range fuel, which we report for the first time in this
paper, is believed to form through etherification of the hydrox-
ymethyl group of the aldol product with ethanol followed by
opening of the furan ring and hydrodeoxygenation of the
ether intermediate.
3
x
which gave 92% C –aldol product with high selectivity at
9
nearly 100% HMF conversion. The as-synthesised Zr(CO ) was
3
x
analysed by several analytical methods for elucidating its struc-
Introduction
The production of renewable fuels from biomass has received
significant attention in the twenty first century because of con-
cerns about environmentally harmful carbon emissions and
rapid depletion of fossil fuels. Biomass, a carbon neutral renew-
able energy source, is widely available for the production of bi-
of furfurals can also produce linear light alkanes of low re-
[
4]
search octane number (RON) values.
The transformation of low carbon containing molecules into
higher carbon chain precursors for jet fuels involves the forma-
tion of new carbon–carbon bonds. The Aldol condensation re-
action allows the formation of longer chain carbon molecules
under mild conditions, but requires a basic or acidic catalyst.
For example, aldol condensation between HMF and acetone
[1]
ofuels and other value-added chemicals. Lignocellulosic bio-
mass can be converted to biofuels and renewable chemicals
through biological, thermal, enzymatic and catalytic process-
[2]
[5]
es. Generally, polysaccharides undergo hydrolysis into their
constituent sugar monomers, which are then dehydrated with
acid catalysts to furanic platform chemicals such as 5-hydroxy-
methyl furfural (HMF) and furfural (Ff). Furfurals are sustainable
renewable feedstock for the manufacture of long chain hydro-
carbon fuels through sequential aldol condensation, hydroge-
forms precursors for C –C hydrocarbons of diesel fractions.
13
15
Most aldol condensation reactions for the production of long
chain hydrocarbon precursors are reported to use homogene-
[
6–9]
ous catalysts, most commonly NaOH.
However, this homo-
geneous base catalyst is corrosive and promotes cross- and
self-condensation reactions. Additionally, neutralisation of the
[3]
[10]
nation and hydrodeoxygenation (HDO) pathways. Direct HDO
generated waste water stream is expensive.
Therefore, it is preferable to use a solid catalyst to avoid ex-
pensive separation steps and minimise the impact on the envi-
ronment. In this context, the design and synthesis of water tol-
erant heterogeneous catalysts for effective aldol condensation
between HMF and acetone or other ketones is desirable.
[
a] Dr. A. Bohre, Prof. Dr. B. Saha
Laboratory of Catalysis
University of Delhi
[
11]
North Campus, Delhi 110007 (India)
A few examples of such catalysts are zeolites, NH -ZSM-5(23),
4
[
b] Prof. Dr. B. Saha, Prof. Dr. M. M. Abu-Omar
NH -ZSM-5(50), NH -BEA(25), NH -BEA(38), NH -MOR(20),
4
4
4
4
Department of Chemistry and the Center for Direct Catalytic Conversion of
[12]
HSDUSY(80) and HSUSY(5). Natural dolomitic rocks are also
reported as a potential base catalyst for aldol condensation of
Ff with acetone in the temperature range of 33–1508C with
Biomass to Biofuels (C
Purdue University
3
Bio)
Indiana 47907 (USA)
E-mail: sahab@purdue.edu
mabuomar@purdue.edu
[
13]
a 73% yield of the condensation product.
Among other
basic catalysts, Mg–Al mixed oxides, hydrotalcites (HT) and
MgAl O are reported to show good activity in aldol condensa-
[
c] Prof. Dr. M. M. Abu-Omar
School of Chemical Engineering
Purdue University
2
4
[
14–17]
tion reactions.
However, these materials, particularly HT,
Indiana 47907 (USA)
have high sensitivity towards ambient CO and hence deacti-
2
ChemSusChem 2015, 8, 4022 – 4029
4022
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[
18]
vate quickly. Therefore, there is a need to develop effective
solid base catalysts for aldol condensation that can be reused
for many cycles.
Zirconium carbonate (Zr(CO ) ), a basic salt of zirconia, is
3
x
[
19,20]
used in the textile, cosmetics, paper and paint industries.
Son and co-workers have reported the use of Zr(CO3)x as
[
21]
a basic catalyst for isomerisation of glucose to fructose.
A Zr(CO ) supported metal catalyst has also been used for the
3
x
production of value-added chemicals. Recently Son et al. inves-
tigated the catalytic activities of Ru/C, Ru/SBA, Au/Zr(CO ) and
3
x
Au/ZrO materials for liquid phase hydrogenation of biomass
2
derived levulinic acid into g-valerolactone (GVL) using formic
[22]
acid as a hydrogen source.
Among these catalysts, Au/
Zr(CO3)x and Au/ZrO2 demonstrated comparable activities,
giving 97% GVL.
In this paper, we report the synthesis of Zr(CO ) and its
3
x
structural properties and catalytic effectiveness for aldol con-
densation of HMF with acetone. The results show that the cat-
Figure 1. Electron Microscopic study of Zr(CO
image (c) TEM image with SAED pattern (d) EDS spectrum.
3 x
) (a) TEM image (b) STEM
alyst is effective for high yield and selectivity of the C aldol
9
product, 4-[5-hydroxymethyl)furan-2-yl]but-3-en-2-one (HAc),
under mild conditions in water. Recyclability studies show that
the catalyst can be reused without significant loss in activity.
Reactions are also conducted to optimise the yield of HAc and
to understand the reaction insights to make Zr(CO ) a promis-
[
26]
cles are amorphous (Figure 1c).
Elemental mapping per-
formed using energy dispersive X-ray spectroscopy (EDS) in
SEM suggests the presence of C, O and Zr in the Zr(CO3)x
sample (Figure 1d). The other peaks in the EDS image are from
the copper grid used for the analysis, which is not a compo-
nent of the Zr(CO ) nanostructure. Elemental analysis further
3
x
ing replacement for NaOH for long chain hydrocarbon synthe-
sis. We extended the scope of the present study by upgrading
the isolated HAc product to the corresponding long chain
fuels with a metal–acid bifunctional catalyst, which produced
gasoline and diesel range fuels, n-nonane and 1-ethoxynonane,
with a total yield of 96%.
3
x
confirmed that the sample contained 3.47 wt% carbon and
1.46 wt% hydrogen, against theoretical values of carbon and
hydrogen of 3.9 and 0.7 wt%, respectively. The difference in
the actual and theoretical carbon and hydrogen content could
be because of the presence of a small amount of moisture in
the sample.
Results and Discussion
The presence of carbonate anions in the as-synthesised
Zr(CO ) was confirmed by FTIR spectroscopy (Figure 2). The
Characterisation of the catalyst
3
x
À1
The crystalline phase of the catalyst, determined by X-ray dif-
fraction, is shown in Figure S1 in the Supporting Information.
The XRD pattern shows two broad peaks in the 2q range of
transmission bands at 1552 and 1349 cm represent the asym-
2
À
metric stretching vibrations of CO3 , while two very weak
À1
bands at 1037 and 1094 cm represent the symmetric stretch-
2
À [27]
À1
1
5–408 and 40–708 indicating that the sample is amorphous.
ing modes of CO3
.
Two other peaks at 2334 and 2378 cm
[
28]
The observed pattern agrees with characteristic reflections of
reported zirconium carbonate materials, although the Joint
Committee on Powder Diffraction Standard (JCPDS) file of pure
may be from the adsorbed CO in the sample. An intense
2
À1
broad band in the region of 3540–3200 cm corresponds to
[21,23]
zirconium carbonate has not been generated yet.
The
broad peaks also represent the nanophase nature of the
[
24]
Zr(CO ) . The higher order reflections were not observed in
3
x
the XRD pattern, which could be either because of the amor-
phous nature of the pore walls or a lack of long range ordering
[
25]
in the Zr(CO ) .
3
x
The morphology and elemental composition analyses of the
Zr(CO ) were performed by TEM and STEM techniques, and
3
x
the characteristic images are shown in Figure 1. The high reso-
lution TEM image reveals that the material is composed of uni-
form 3–4 nm spherical particles (Figure 1a). The STEM image
shows an ordered arrangement of Zr(CO ) nanoparticles and
3
x
that the particles are mostly amorphous (Figure 1b). The short-
range order of broad rings in the selected area electron diffrac-
tion (SAED) pattern also suggests that the Zr(CO ) nanoparti-
3 x
Figure 2. FTIR Spectrum of as-synthesised Zr(CO ) catalyst.
3
x
ChemSusChem 2015, 8, 4022 – 4029
www.chemsuschem.org
4023
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
the asymmetric stretching and bending modes of the OH
acetone at 408C yielded 63% HAc at 80% HMF conversion
after 24 h. Upon further increase of the temperature to 548C,
a significant improvement in the HAc yield (92%) was noted
with quantitative conversion of HMF.
À1
groups. A weak band at 841 cm represents the ZrÀO stretch-
[
28]
ing mode.
Thermogravimetric analysis (TGA) of the synthesised and
used catalysts were performed in air for measuring the weight
loss as a function of temperature (Figure S2). The results show
two mass loss steps at 228 and 5728C with a corresponding
mass loss of 8.85 and 24.37%, respectively. The mass loss in
the first step is attributed to the
Usually, the aldol condensation between HMF and ketones
progresses via the formation of a C alcohol as an intermedi-
9
ate, followed by dehydration of the intermediate alcohol to an
[
18]
a, b-unsaturated ketone (Scheme 1).
The present reaction
loss of adsorbed moisture and
the structured water, while the
second mass loss can be ex-
plained by the decomposition of
the interlayer carbonate. The
TGA data suggest that the cata-
lyst is stable over a broad tem- Scheme 1. Reaction Scheme for aldol condensation between HMF and acetone over Zr(CO
perature range, although its cat-
3 x
) catalyst.
alytic activity was investigated
up to 548C. Similar mass loss was also noted for the recovered
with the Zr(CO ) catalyst is also believed to proceed via a simi-
3 x
catalyst after three consecutive cycles.
lar pathways. A mixture of monomer (HAc) and dimer 1,5-bis-
2-furanyl-5-hydroxymethyl)-1,4-pentadien-3-one (H Ac) are re-
(
2
ported to form as the major products from aldol condensation
Catalysis
[29]
of HMF with acetone over MgO–ZrO catalyst. Similarly, Ff
2
condensation with acetone over 20% MgO/NaY catalyst is also
reported to form a mixture of monomer 4-(2-furanyl)-3-butene-
We evaluated the catalytic activity of as-synthesised Zr(CO3)x
for aldol condensation between HMF and acetone in water in
varying reaction conditions, such as variation of the HMF to
acetone molar ratio, catalyst loading, reaction time and tem-
perature. The purpose for designing these experiments was to
optimise the reaction conditions and maximise the yield of
HAc.
2
-one (FAc) and dimer 1,5-bis-(2-furanyl)-1,4-pentadien-3-one
[
30]
(
F2Ac).
When comparing these literature observations with
our results, the Zr(CO ) catalysed reaction produced only one
3 x
product (HAc) with high selectivity, which is important for up-
grading the aldol precursor to the corresponding hydrocarbon
fuels selectively through catalytic HDO.
Effect of reaction temperature
Effect of catalyst loadings
The reaction temperatures for condensation of HMF
(
0.5 mmol) with acetone (10 mmol) using 100 mg catalyst in
mL water was varied from 20 to 548C at a fixed reaction
The yield of HAc from aldol condensation between HMF and
acetone was found to increase by increasing the amount of
Zr(CO ) catalyst from 60 mg to 100 mg (Figure 4). Using 60 mg
1
time (24 h) using an oil bath for heating. The temperature was
found to have a significant effect on both HMF conversion and
HAc yields. A preliminary reaction between HMF and acetone
formed no product at 208C. However, higher HMF conversions
and HAc yields were obtained at higher temperatures
3
x
Zr(CO3)x catalyst, a reaction between 0.5 mmol HMF and
1
7
0 mmol acetone in water (1 mL) produced 56% HAc with
2% conversion of HMF at 548C for 24 h. When the catalyst
(
ꢀ408C). As shown in Figure 3, a reaction between HMF and
Figure 4. Effect of catalyst loading on HAc yields and HMF conversions. Error
Figure 3. Effect of reaction temperature on HAc yield and HMF conversion.
bars represent experimental uncertainty.
ChemSusChem 2015, 8, 4022 – 4029
www.chemsuschem.org
4024
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
loading was increased from 60 mg to 80 mg, a higher conver-
sion of HMF (90%) with higher HAc yield (72%) was recorded.
Upon further increasing the catalyst loading to 100 mg, the re-
action achieved quantitative HMF conversion with 92% HAc
yield. Under comparable conditions, a reaction at lower cata-
lyst loading (20 mg) gave lower HAc yield (30%) and HMF con-
version (51%). Mass transport of the heterogeneous catalyst
was not a limiting factor in this reaction as we found an in-
crease in HAc yields with increased catalyst loadings.
Effect of reaction time
The effect of reaction time on the aldol condensation reaction
of HMF (0.5 mmol) with acetone (10 mmol) over Zr(CO3)x
(
100 mg) at 548C was investigated by varying the reaction
time from 2 to 24 h. As shown in the reaction profile
Figure 5), both HMF conversions and HAc yields were found
Figure 6. Effect of substrate variation on HAc yield and HMF conversion.
(
over 24 h. HMF conversion was poor (20%) at a 1:1 molar ratio
of HMF to acetone. A change in the molar ratio of HMF to ace-
tone from 1:1 to 1:5 resulted in a significant improvement in
HMF conversion (from 20 to 47%) and HAc yield (from 5 to
2
4%). When the ratio of the reactants was 1:10, the conversion
of HMF reached 70% with 54% HAc yield. A further decrease
in the HMF to acetone molar ratio to 1:20 gave an optimal
HAc yield and HMF conversion. While the results show that
a high amount of acetone is necessary to achieve a high yield
of aldol product, low boiling point acetone can be easily sepa-
rated from the process and recycled. A previous report also
demonstrated a similar reaction trend showing the yield of
aldol product improved at higher molar ratios of the reac-
[
31]
tants. A control experiment in neat acetone produced no
aldol product, which suggests that water plays an important
role in this reaction. It is proposed that water increased
the formation of the desired products via an enolate inter-
mediate and suppressed the formation of by-products
Figure 5. Effect of reaction time on HAc yield and HMF conversion.
[
13,32–33]
to increase with increasing reaction time. After 2 h the reaction
produced only 30% HAc at 51% HMF conversion. Upon in-
creasing the reaction time from 2 to 12 h, the HAc yield and
HMF conversion increased from 30 to 67% and 51 to 85%, re-
spectively. The results show that the reaction achieved a maxi-
mum HAc yield (92%) and nearly 100% HMF conversion after
(Scheme 2).
Catalyst recyclability
To test the catalyst stability, a recyclability study was per-
formed for a reaction between 0.5 mmol HMF and 10 mmol
2
4 h.
The effect of HMF to acetone
ratio
The molar ratio of HMF to ace-
tone was found to influence the
overall yield of HAc (Figure 6). To
study this effect, the starting
acetone concentrations were
varied from 0.5 to 10 mmol at
a fixed HMF concentration of
0
.5 mmol. Under a 1:1 molar
ratio of HMF to acetone, a reac-
tion with 100 mg Zr(CO ) cata-
3
x
3 x
lyst produced 5% HAc at 548C Scheme 2. Proposed pathways for Zr(CO ) catalysed aldol condensation.
ChemSusChem 2015, 8, 4022 – 4029
www.chemsuschem.org
4025
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
acetone using 100 mg catalyst at 548C for 24 h. After each
cycle, the catalyst was recovered, washed with distilled water,
dried and reused for the next cycle by adding fresh HMF, ace-
tone and water. Fresh catalyst was not added to replenish any
loss of the catalyst mass during recovery. As shown in Figure 7,
the yield of HAc decreased from 91% (1st cycle) to 84% (5th
cycle). While the XRD pattern and TGA data of the reused cata-
lyst were comparable to that of the as-synthesised catalyst
sponding long chain alkanes with moderate yields and selec-
[
38]
tivity.
To show the potential integration of our aldol condensation
step with a hydrogenation step for upgrading HAc product to
fuels, we studied HDO of isolated HAc with a bifunctional
metal–acid catalyst, Pd/Zeolite b (5 wt% Pd). The catalytic ef-
fectiveness of Pd/Zeolite-b for HDO of Me -furoin in ethanol
2
has been reported in an earlier publication which demonstrat-
(
Figure S1 and S2), a slight loss in activity of the re-used cata-
ed 70% n-dodecane yield with 96% selectivity at 2308C,
[
38b]
lyst in the 5th cycle could be because of a slight loss in basicity
60 bar H and 8 h reaction time.
Under a comparable reac-
2
À1
of the recovered Zr(CO ) catalyst from 0.19 mmolg in as-syn-
tion temperature (2308C) and at a lower H pressure (50 bar),
3
x
2
À1
thesised catalyst to 0.15 mmolg after the 5th cycle (Table S1).
Moreover, loss of catalyst mass during recovery could also ac-
count for such variation in catalytic activity.
we found complete conversion of HAc with Pd/Zeolite-b in
ethanol with the formation of two main products, as shown in
the GC chromatogram (Figure 8). Qualitative and quantitative
Figure 8. GC chromatogram of the product solution obtained from HDO of
the isolated aldol product with Pd/Zeolite-b.
Figure 7. Recyclability study of the Zr(CO
between acetone and HMF.
3
)
x
catalyst for aldol condensation
analysis of the product solution by GC/MS and GC methods re-
vealed the two products are n-nonane and 1-ethoxynonane
with a selectivity of 40 and 56%, respectively. While n-nonane
is a gasoline range fuel, the low oxygenated ethoxy derivate
can be used as a diesel fuel, like 5-ethoxymethylfurfural with
Catalytic HDO of HAc to fuels
À1 [39]
an energy density of 30 MJL . Thus, the overall yield of gas-
Deoxygenation of aldol products to fuel range hydrocarbons is
a viable strategy for the production of drop-in biofuels from
lignocellulose. This process has received significant attention in
recent years because of drawbacks of other processes. For ex-
ample, upgrading biomass to hydrocarbons through gasifica-
tion, followed by Fischer–Tropsch synthesis is energy and capi-
oline and diesel range fuels from our reaction is 96%, which is
higher than those reported from HDO of HAc using Pd/Al O
2
3,
[
40]
Pd/C-La(OTf) and Pt/Al O catalysts. A small amount of 1-
2
3
nonanol was also detected in the GC/MS spectrum, which
could account for the balance 4% of the carbon mass.
The formation of 1-ethoxynonane in our reaction suggests
that HDO of HAc with the Pd/Zeolite-b catalyst progressed in
two possible pathways (Scheme 3); etherification of the À
[
34]
tal intensive. Another approach is hydrogenation of furfurals
to low oxygenated 2-methylfuran (2-MF) and 2,5-dimethylfuran
[35]
(
DMF) for alternate fuels or additives, but these furanic mole-
CH OH group of HAc with ethanol prior to opening of the
2
cules have a low carbon density. Catalytic upgrading of aldol
products to long chain alkanes has been studied by using sev-
eral heterogeneous catalysts such as bifunctional Pt/Al O -
furan ring and subsequent HDO (pathway A), or involvement
of direct transformation of HAc through ring opening followed
by HDO (pathway B). Another pathway involving HDO of HAc
to 1-nonanol followed by etherification of 1-nonanol to 1-
ethoxynonane cannot be excluded. The detection of a small
amount of 1-nonanol in the GC/MS spectrum supports possi-
bility of the latter pathway.
2
3
[7]
SiO2, Pd/Al-MCM-41, Pd/H-ZSM-5, Pd/H-b, Pd/Al O Pd/SiO2-
2
3,
[36]
Al O , Pd/H-ZSM-5, Pt/C, Pt/Al O and Pt/ZrP. However, the
2
3
2
3
yields of the desired hydrocarbons in these catalytic HDO pro-
cesses were either poor or the reactions required harsh reac-
tion conditions (>3008C, >50 bar H ) to achieve reasonable
Interestingly, we found no alkane product when Pd/Zeolite-
b catalysed HDO of HAc was carried out in THF under the con-
2
[
36c,37]
hydrocarbon yields.
Recently, metal–acid tandem catalysis
[
35b]
using bifunctional catalysts have been reported to be effective
for the conversion of oxygenated intermediates to the corre-
ditions reported earlier
and the GC chromatogram showed
multiple peaks. This observation agreed with reported results
ChemSusChem 2015, 8, 4022 – 4029
www.chemsuschem.org
4026
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
an important role in facilitating
condensation. Recyclability stud-
ies indicate a slight deactivation
of the catalyst during the fifth
cycle; in terms of yield of the
aldol product the loss was 7%,
which could be because of
a slight loss of basic density or
catalyst mass during recovery.
XRD patterns and TGA data of
the re-used catalyst after the
third cycle were comparable to
those of the as-synthesised ma-
Scheme 3. Possible pathways for HAc HDO with Pd/Zeolite-b to n-nonane and 1-ethoxynonane.
demonstrating that the formation of alkanes is more selective
in ethanol and that a high temperature and pressure are nec-
essary for opening the furan ring of the oxygenated intermedi-
terial. Further investigation on catalytic HDO of the aldol prod-
uct with a Pd/Zeolite-b catalyst revealed that isolated aldol
product can be selectively upgraded to gasoline and diesel
range fuels, n-nonane and 1-ethoxynonane, with an overall
yield of 96%.
[38b]
ate and its subsequent HDO.
of the Pd catalyst for HDO of HAc was performed in ethanol at
308C and 50 bar H for 8 h. In this experiment, the catalyst
The recyclability experiment
2
2
was filtered from the product solution by filtration, washed
and dried before reuse for the next cycle. The GC chromato-
gram of the product solution obtained from the second cycle
showed multiple peaks. Although the peak at 10.5 min for 1-
ethoxynonane was not observed, the n-nonane peak at
Experimental Section
Materials and catalyst preparation
ZrOCl ·8H O and Na CO were purchased from Thomas Baker,
2
2
2
3
India. Dried acetone was purchased from Merck, India. HMF of
99% purity was purchased from Sigma–Aldrich. All the reagents
6
.1 min was broad, which could be because of the overlap of
~
the peaks of the oxygenated intermediates that are formed
after opening of the furan ring of HAc with the n-nonane
peak. A couple of other peaks in the lower molecular weight
region were also observed, suggesting possible fragmentation
of the oxygenated intermediate, owing to the partial deactiva-
tion of the reused catalyst, which was not observed with the
fresh catalyst. The observed deactivation of the reused catalyst
could be because of coke formation on the catalyst surface,
were used without further purification. Unless otherwise men-
tioned, deionised water was used for all reactions. Zirconium car-
[
21]
bonate catalyst was prepared by following a reported method.
According to this method, a solution of 0.5 m ZrOCl ·8H O (40 mL)
was added slowly into a solution of 0.25 m Na CO (100 mL) with
2
2
2
3
constant starring over a period of 1 h. The resultant white gel was
aged at 708C for 12 h. The aged solid was then recovered by filtra-
tion, washed with distilled water until the pH of the filtrate
reached 7 and dried in a muffle furnace at 1508C for 3 h.
[38b]
which blocks facile access to the acidic sites.
Further re-
Pd/Zeolite-b catalyst was prepared by incipient wetness impregna-
tion of Pd (5 wt%) on a zeolite-b (H-form, 360:1 SiO /Al O , ob-
search is in progress to understand the deactivation of the Pd/
Zeolite-b catalyst and to develop more stable catalysts by
using different acidic supports.
2
2
3
tained from Alfa Aesar) by following our previously reported meth-
[38b]
od.
In this method, an aqueous HCl solution of palladium di-
chloride was slowly added to the zeolite support with continuous
stirring over 12 h. The mixture was then placed in an oven and al-
lowed to dry at 508C overnight. The dried material was then trans-
ferred to a Schwartz-type drying tube and calcined under air for
Conclusions
We have synthesised a water tolerant heterogeneous basic cat-
alyst, zirconium carbonate (Zr(CO ) ) and studied its structural
2
h at 4508C. The gas feed was then switched to H and the cata-
2
3
x
lyst was reduced for 2 h at the same temperature.
properties and catalytic effectiveness for aldol condensation
between HMF and acetone in water. Detailed characterisation
data suggests that the as-synthesised nanoparticulate material
Physico-chemical characterisation
À1
is amorphous and has a basic density of 0.19 mmolg . Under
Powder X-ray diffraction (PXRD) patterns of as-synthesised Zr(CO3)x
mild reaction conditions, the catalyst showed excellent activity
enabling quantitative HMF conversion and 92% yield of C9
were recorded on a Bruker d-8 Advance diffractometer using Cu
radiation (l=0.15406 nm) at a scanning rate of 1.28min and 2q
Ka
À1
aldol product, a precursor for upgrading to C hydrocarbon
range of 5–808. Transmission electron microscope (TEM) analysis of
9
fuel. Besides the high yield (92%), the Zr(CO ) catalysed reac-
the sample was performed on a JEOL TEM/STEM-2010F (accelera-
3
x
tion voltage: 200 kV) instrument by dispersing Zr(CO ) powder on
tion also improved selectivity of the aldol product by produc-
ing only one product with high selectivity. Several reactions
were performed under variable reaction conditions such as,
catalyst loading, reaction time, temperature and molar ratio of
HMF to acetone, to optimise the yield of the aldol product. In-
terestingly, a reaction in neat acetone without water did not
show any condensation reaction, suggesting that water plays
3 x
a Cu grid pre-coated with a thin carbon film. To confirm the func-
tional compositions of the Zr(CO ) , an FTIR spectrum was recorded
3
x
on a KBr pellet by using a PerkinElmer 1752X FTIR spectrophotom-
1
eter. H NMR spectral analysis was recorded on a JEOL JNM ECX-
4
00 P 400 MHz instrument and the collected data were processed
by using JEOL DELTA program version 4.3.6. A Vario EL III element
analyser was used for elemental analysis of the catalyst. The ther-
ChemSusChem 2015, 8, 4022 – 4029
www.chemsuschem.org
4027
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
mogravimetric analysis of the catalyst was performed by using a Di-
amond TG-DTA (PerkinElmer, USA) instrument from room tempera-
Hydrogenation of aldol product (HAc)
À1
Catalytic HDO of HAc was carried out in a stainless steel Parr reac-
tor equipped with a stirring impeller, gas line and programmable
control device for setting reaction time and temperature. In this
experiment, the reactor was loaded with 35 mg isolated HAc,
ture to 7008C at a heating rate of 108Cmin . The basicity of the
[41]
catalyst was determined by the benzoic acid titration method. In
this method, 0.1 g of catalyst was suspended in 2 mL of water/eth-
anol (4:1 v/v) solvent with constant stirring for 30 min. Using phe-
nolphthalein as an indicator, this solution was then titrated against
1
00 mg Pd/Zeolite-b (5 wt% Pd) and 20 mL ethanol and then
properly sealed with the reactor head. After purging the mixture
0
.01m benzoic acid (in toluene) until the pink color of the basic so-
with UHP grade H for a couple of minutes, the reactor was pres-
2
lution disappeared. The concentration of the total basic density
was calculated from the volume of benzoic acid solution. The titra-
tion was repeated 3 times to obtain an average basic density of
the catalyst.
surised with 50 bar H and heated with stirring at 2308C for 8 h.
2
The reactor was then cooled to room temperature and the pres-
sure was released. The solution was filtered and the collected fil-
trate was analysed by GC and GC–MS.
The yield of the hydrogenation products of HAc were measured by
analysing the product solutions on a GC instrument (Agilent
6
890N) equipped with an FID detector and DB-5 capillary column
Aldol condensation reaction
of dimension 0.25 mm ID ” 0.25 mm”30 m. The essential parame-
ters of the GC analysis were as follows: injection volume 1.0 mL,
inlet temperature 2508C, detector temperature 2508C and a split
ratio 1:5. The initial column temperature was 508C (2 min) with
Zr(CO ) -catalysed aldol condensation between HMF and acetone
3
x
was carried out in the temperature range of 20–548C. All reactions
were performed in a 5 mL round-bottom flask. In a typical experi-
ment, HMF (63 mg), acetone (580 mg) and catalyst (100 mg) were
mixed with 1 mL distilled water. The reaction mixture was then
heated with constant stirring in a temperature-controlled oil bath
at the desired temperature. Upon completion of reaction for the
desired time of 2–24 h, the reaction mixture was cooled to room
temperature and the catalyst was separated by filtration. The or-
ganic product was extracted by liquid–liquid separation using
ethyl acetate as an organic phase. Upon removal of ethyl acetate
by rotary evaporator under vacuum, the crude product was collect-
À1
a temperature rise of 108Cmin and the final temperature was
3008C. n-Nonane was identified by its retention time in compari-
son with an authentic sample and by GC–MS analysis. 1-Ethoxyno-
nane was characterised by GC–MS. Both peaks in the GC chroma-
togram were properly integrated and the actual concentration of
n-nonane was obtained from a pre-calibrated plot of peak area
against concentrations. The yield of 1-ethoxynonane was deter-
mined from its peak area and the response factor, calculated by
considering n-nonane as a standard. GC–MS spectrometry analyses
were carried out on an Agilent 5975C (Agilent Labs, Santa Clara,
CA) mass spectrometer system. The typical electron energy was
70 eV with the ion source temperature maintained at 2508C. The
individual components were separated using a 30 meter DB-5 ca-
pillary column (250 mm ID ” 0.25 mm film thickness). The initial
column temperature was set at 358C (for 3 min) and programmed
1
13
ed and analysed by NMR. Both H and C NMR spectra of the
brown oily liquid revealed the formation of 4-(5-hydroxymethyl)fur-
an-2-yl) but-3-en-2-one (HAc) as a sole product, and its observed
1
characteristic signals in the NMR spectra are as follows; H NMR
(
400 MHz, CDCl ): d=2.29 (s, CH ), 4.62 ppm (s, CH OH), 6.37 (d,
3 3 2
13
ÀCH), 6.58 (d, ÀCH), 6.60 (d, ÀCH), 7.22 (d, ÀCH); C NMR
À1
(
400 MHz, CDCl ): d=27.51, 110.14, 116.92, 123.59, 129.44, 150.24,
to reach 2808C at 10.08Cmin . The flow rate is typically set at
1 mLmin . The injector temperature was set at 2508C.
3
À1
1
57.38, 198.42 ppm (Figure S3 and S4). A few reactions were also
replicated three times to determine the experimental uncertainty
in the HMF conversions and HMF yields, which were found to be
within Æ4%.
Acknowledgements
B.S. acknowledges the Council of Scientific and Industrial Re-
search (CSIR), New Delhi, Government of India, for funding this
research (Project Number No. 01 (2686)/12/EMR-II). M.M.A.-O. ac-
knowledges the Center for Direct Catalytic Conversion of Biomass
to Biofuels (C3Bio), an Energy Frontier Research Center funded by
the U.S. Department of Energy, Office of Science and Office of
Basic Energy Sciences under Award Number DE-SC0000997 for
supporting this research. A.B. acknowledges the University Grant
Commission, India for a DS Kothari Postdoctoral Research Fellow-
ship.
Determination of HAc yield
1
[42]
The HAc yield was measured using H NMR spectroscopy. In this
method, a known concentration of mesitylene (internal standard)
was added into the crude HAc product in CDCl . The yield of HAc
3
was calculated from the integrated values of the ÀCH OH proton
2
(d=4.62 ppm) of HAc and the three aromatic ring protons of mesi-
tylene (d=6.79 ppm).
Recyclability study of Zr(CO3)x
Keywords: aldol condensation · 5-hydroxymethylfurfural ·
hydrodeoxygenation · zeolites · zirconium carbonate
The recycling efficiency of the Zr(CO ) catalyst was determined for
3
x
the reaction between HMF and acetone under the following condi-
[
1] M. Stçcker, Angew. Chem. Int. Ed. 2008, 47, 9200–9211; Angew. Chem.
tions: HMF=63 mg, acetone=580 mg, Zr(CO ) =100 mg, reaction
3
x
2008, 120, 9340–9351.
time=24 h and T=548C using 1 mL distilled water as a solvent.
After the reaction, the reaction mixture was cooled down to room
temperature and the catalyst was separated by filtration. The re-
[
[
2] J. R. Regalbuto, Science 2009, 325, 822–824.
3] G. Li, N. Li, J. Yang, L. Li, A. Wang, X. Wang, Y. Cong, T. Zhang, Green
Chem. 2014, 16, 594–599.
covered Zr(CO ) catalyst was washed with distilled water, dried
3
x
[4] T. Buntara, S. Noel, P. H. Phua, I. Melian-Cabrera, J. G. de Vries, H. J.
Heeres, Angew. Chem. Int. Ed. 2011, 50, 7083–7087; Angew. Chem. 2011,
123, 7221–7225.
and reused for three consecutive cycles by adding fresh HMF, ace-
tone and water. The HAc yield was determined after each run.
ChemSusChem 2015, 8, 4022 – 4029
www.chemsuschem.org
4028
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[
[
[
[
[
5] G. W. Huber, J. N. Chheda, C. J. Barrett, J. A. Dumesic, Science 2005, 308,
446–1450.
6] H. Liu, W. Xu, X. Liu, Y. Guo, Y. Guo, G. Lu, Y. Wang, Kinet. Catal. 2010,
1, 75–80.
[29] W. Shena, G. A. Tompsett, K. D. Hammond, R. Xing, F. Dogan, C. P. Grey,
W. C. Conner, S. M. Auerbach, G. W. Huber, Appl. Catal. A 2011, 392, 57–
68.
[30] X.-m. Huang, Q. Zhang, T.-j. Wang, Q.-y. Liu, L.-l. Ma, Q. Zhang, J. Fuel
Chem. Technol. 2012, 40, 973–978.
1
5
7] R. Xing, A. V. Subrahmanyam, H. Olcay, W. Qi, G. P. van Walsum, H.
Pendse, G. W. Huber, Green Chem. 2010, 12, 1933–1946.
8] P. A. Zapata, J. Faria, M. P. Ruiz, D. E. Resasco, Top.Catal. 2012, 55, 38–
[31] R. M. West, Z. Y. Liu, M. Peter, C. A. Gartner, J. A. Dumesic, J. Mol. Catal.
A 2008, 296, 18–27.
[
[
32] C. Wang, J. Liu, W. Leng, Y. Gao, Int. J. Mol. Sci. 2014, 15, 1284–1299.
33] E. L. Coitino, J. Tomasi, O. N. Ventura, J. Chem. Soc. Faraday Trans. 1994,
5
2.
9] N. Fakhfakh, P. Cognet, M. Cabassud, Y. Lucchese, M. D. de Los Rios,
9
0, 1745–1755.
34] a) J. N. Chheda, G. W. Huber, J. A. Dumesic, Angew. Chem. Int. Ed. 2007,
6, 7164–7183; Angew. Chem. 2007, 119, 7298–7318; b) D. M. Alonso,
Chem. Eng. Process. 2008, 47, 349–362.
[
[
[
[
[
[
10] C. J. Barrett, J. N. Chheda, G. W. Huber, J. A. Dumesic, Appl. Catal. B
006, 66, 111 –118.
11] E. Dumitriu, V. Hulea, I. Fechete, A. Auroux, J. F. Lacaze, C. Guimon, Mi-
croporous Mesoporous Mater. 2001, 43, 341–359.
12] O. Kikhtyanin, V. Kelbichova, D. Vitvarova, M. Kubu, D. Kubicka, Catal.
Today 2014, 227, 154–162.
13] R. E. O’neill, L. Vanoye, C. D. Bellefon, F. Aiouache, Appl. Catal. B 2014,
4
2
J. Q. Bond, J. A. Dumesic, Green Chem. 2010, 12, 1493–1513.
35] a) B. Saha, M. M. Abu-Omar, ChemSusChem 2015, 8, 1133–1142; b) B.
Saha, C. M. Bohn, M. M. Abu-Omar, ChemSusChem 2014, 7, 3095–3101;
c) J. Jae, W. Zheng, R. F. Lobo, D. G. Vlachos, ChemSusChem 2013, 6,
1158–1162; d) W. Lee, Z. Wang, W. Zheng, D. G. Vlachos, A. Bhan, Catal.
Sci. Technol. 2014, 4, 2340–2352; e) P. Panagiotopoulou, D. G. Vlachos,
Appl. Catal. A 2014, 480, 17–24; f) S. De, S. Dutta, B. Saha, Chem-
SusChem 2012, 5, 1826–1833.
1
44, 46–56.
[
[
14] F. Cavani, F. Triffiro, A. Vaccari, Catal. Today 1991, 11, 173–301.
15] K. K. Rao, M. Gravelle, J. Valente, F. Figureueras, J. Catal. 1998, 173, 115–
[
[
36] a) M. Chatterjee, K. Matsushima, Y. Ikushima, M. Sato, T. Yokoyama, H.
Kawanami, T. Suzuki, Green Chem. 2010, 12, 779–782; b) J. Yang, N. Li,
S. Li, W. Wang, L. Li, A. Wang, X. Wang, Y. Cong, T. Zhang, Green Chem.
1
21.
[
[
16] K. Pupovac, R. Palkovits, ChemSusChem 2013, 6, 2103–2110.
17] L. Hora, V. Kelbichova, O. Kikhtyanin, O. Bortnovskiy, D. Kubicka, Catal.
Today 2014, 223, 138–147.
18] C. Xu, Y. Gao, X. Liu, R. Xin, Z. Wang, RSC Adv. 2013, 3, 793–801.
19] J. Fenner, (Goldschmidt AG), US 4283377, 1980.
20] V. N. Pokhodenko, N. G. Kisel, K. A. Solomko, L. I. Kisina, V. N. Zinkov,
N. A. Demidova, Refract. Ind. Ceram. 1980, 21, 266–268.
21] P. A. Son, S. Nishimura, K. Ebitani, React. Kinet. Mech. Catal. 2014, 111,
2
014, 16, 4879–4884; c) A. Corma, O. De La Torre, M. Renz, N. Villandier,
Angew. Chem. Int. Ed. 2011, 50, 2375–2378; Angew. Chem. 2011, 123,
423–2426; d) G. Li, N. Li, Z. Wang, C. Li, A. Wang, X. Wang, Y. Cong, T.
[
[
[
2
Zhang, ChemSusChem 2012, 5, 1958–1966.
37] a) C. Wen, E. Barrow, J. Hattrick-Simpers, J. Lauterbach, Phys. Chem.
Chem. Phys. 2014, 16, 3047–3054; b) G. Li, N. Li, J. Yang, A. Wang, Biore-
sour. Technol. 2013, 134, 66–72.
[
1
83–197.
[38] a) D. Liu, E. Y.-X. Chen, ChemSusChem 2013, 6, 2236–2239; b) B. L. We-
genhart, L. Yang, S. C. Kwan, R. Harris, H. I. Kenttmaa, M. M. Abu-Omar,
ChemSusChem 2014, 7, 2742–2747; c) A. D. Sutton, F. D. Waldie, R. Wu,
M. Schlaf, L. A. Petesilks, J. C. Gordon, Nat. Chem. 2013, 5, 428–432.
[39] G. J. M. Gruter, F. Dautzenberg, (Avantium) EP 1834950 A1, 2007.
[40] a) W. Dedsuksophon, V. Champreda, N. Laosiripojana, Eng. J. 2010, 14,
1–10; b) D. Liu, E. Y.-X. Chen, ACS Catal. 2014, 4, 1302–1310; c) I. Faba,
E. Diaz, S. Ordonez, Appl. Catal. B 2014, 160, 436–444.
[
[
22] P. A. Son, S. Nishimura, K. Ebitani, RSC Adv. 2014, 4, 10525–10530.
23] A. Sinhamahapatra, N. Sutradhar, B. Roy, A. Tarafdar, H. C. Bajaj, A. B.
Panda, Appl. Catal. A 2010, 385, 22–30.
24] G. Balakrishnan, P. Kuppusami, D. Sastikumar, J. Song, Nanoscale Res.
Lett. 2013, 8, 82–87.
[
[
[
25] M. J. Hudson, J. A. Knowles, J. Mater. Chem. 1996, 6, 89–95.
26] S. K. Das, M. K. Bhunia, A. K. Sinha, A. Bhaumik, J. Phys. Chem. C 2009,
1
13, 8918–8923.
27] a) Q. Tao, B. J. Reddy, H. He, L. Frost, P. Yuan, J. Zhua, Mater. Chem. Phys.
008, 112, 869–875; b) T. Gougousi, D. Niu, R. W. Ashcraft, G. N. Parsons,
[41] R. Rahul, J. K. Satyarthi, D. Srinivas, Indian J. Chem. Sect. A 2011, 50,
1017–1025.
[42] S. De, S. Dutta, B. Saha, Green Chem. 2011, 13, 2859–2868.
[
[
2
Appl. Phys. Lett. 2003, 83, 3543–3545.
28] V. Bolis, G. Magnacca, G. Cerrato, C. Morterra, Thermochim. Acta 2001,
Received: August 22, 2015
3
79, 147–161.
Published online on November 9, 2015
ChemSusChem 2015, 8, 4022 – 4029
www.chemsuschem.org
4029
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim