Etherification and reductive etherification of 5-(hydroxymethyl)furfural: 5-(alkoxymethyl)furfurals and 2,5-bis(alkoxymethyl)furans as potential bio-diesel candidates

Article abstract of DOI:10.1039/c2gc35102a

A low energy intensive process for the production of diesel fuel has been delineated from both 5-(hydroxymethyl)furfural (HMF) and its sugar precursor d-(-)-fructose. Alcoholic solutions of the above produced a mixture of potential bio-diesel candidates namely, 5-(alkoxymethyl)furfural, 5-(alkoxymethyl) furfural dialkylacetal, and alkyl levulinate, in the presence of solid acid catalysts. Sulfonic acid functionalized resins, Amberlyst-15 and Dowex DR2030 showed exceptional reactivity and selectivity for these reactions. Production of another potential diesel candidate 2,5-bis(alkoxymethyl)furan has been optimized through both sequential reduction/etherification and one-pot reductive etherification processes. During the metal catalyzed hydrogenation of HMF, platinum showed an exclusive selectivity for the reduction of the carbonyl functionality of HMF. Both Pt and Pt/Sn supported on Al₂O₃ catalysts have been optimized for the production of 2,5-bis(alkoxymethyl)furan from HMF. The reaction mechanisms of etherification and reductive etherification have been discussed in detail on the basis of intermediates observed during these processes. The Royal Society of Chemistry.

Full text of DOI:10.1039/c2gc35102a

View Online / Journal Homepage / Table of Contents for this issue
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
Volume 14 | Number 6 | June 2012 | Pages 1547–1806
ISSN 1463-9262
COVER ARTICLE
Bell et al.
Etherification and reductive etherification of 5-(hydroxymethyl)furfural:
5-(alkoxymethyl)furfurals and 2,5-bis(alkoxymethyl)furans as potential
bio-diesel candidates

Dynamic Article Links
Green Chemistry
View Online
Cite this: Green Chem., 2012, 14, 1626
www.rsc.org/greenchem
PAPER
Etherification and reductive etherification of 5-(hydroxymethyl)furfural:
5-(alkoxymethyl)furfurals and 2,5-bis(alkoxymethyl)furans as potential
bio-diesel candidates†
Madhesan Balakrishnan, Eric R. Sacia and Alexis T. Bell*
Received 21st January 2012, Accepted 19th March 2012
DOI: 10.1039/c2gc35102a
A low energy intensive process for the production of diesel fuel has been delineated from both
5-(hydroxymethyl)furfural (HMF) and its sugar precursor D-(–)-fructose. Alcoholic solutions of the above
produced a mixture of potential bio-diesel candidates namely, 5-(alkoxymethyl)furfural, 5-(alkoxymethyl)
furfural dialkylacetal, and alkyl levulinate, in the presence of solid acid catalysts. Sulfonic acid
functionalized resins, Amberlyst-15 and Dowex DR2030 showed exceptional reactivity and selectivity for
these reactions. Production of another potential diesel candidate 2,5-bis(alkoxymethyl)furan has been
optimized through both sequential reduction/etherification and one-pot reductive etherification processes.
During the metal catalyzed hydrogenation of HMF, platinum showed an exclusive selectivity for the
reduction of the carbonyl functionality of HMF. Both Pt and Pt/Sn supported on Al2O3 catalysts have
been optimized for the production of 2,5-bis(alkoxymethyl)furan from HMF. The reaction mechanisms of
etherification and reductive etherification have been discussed in detail on the basis of intermediates
observed during these processes.
feedstocks, such as energy grasses that can be produced in high
Introduction
yields per acre with low inputs of fertilizer on land not used for
food crops.6
The conversion of lignocellulosic biomass to diesel fuel is of
considerable interest since worldwide demand for diesel/gasoil
exceeded that for gasoline by nearly 15% in 2009, and the
growth in demand for diesel is expected to outpace that for gaso-
line by over a factor of two through at least 2030.1 With increas-
ing mandates in both the European Union2 and United States3
for the introduction of biomass-derived fuels, innovative sol-
utions are needed to generate transportation fuels on a large
scale. The challenge then is to identify and develop strategies for
producing compounds that could serve as suitable replacements
for typical petroleum-derived diesel, a mixture of C11–C22
alkanes, cycloalkanes, and limited aromatic compounds.
First generation biodiesel has largely been produced through
the transesterification of triglycerides with alcohols to form fatty
acid methyl esters (FAME), which can be used in current
engines with little modification.4 On the other hand, the rela-
tively high cost of the starting material and the low oxidation
stability and poor cold flow properties of these fuels are a
problem, as is the concern of using a food product as a starting
material for a fuel.4,5 For this reason there is considerable incen-
tive to identify routes to diesel starting with lignocellulosic
Several alternatives to the transesterification of triglycerides
with alcohols have been proposed for producing diesel-range
products from cellulosic feedstocks. The first of these is aldol
condensation of 5-(hydroxymethyl)furfural (HMF) or furfural
with acetone.4,7 Both HMF and furfural are readily formed
by the catalyzed dehydration of glucose and xylose obtained by
hydrolysis of the cellulosic components of lignocellulosic
biomass.8 The resulting C9–C15 molecules produced by aldol
condensation must be extensively hydrogenated to form straight
chain alkanes in order to decrease the cloud point and increase
the cetane number of the final products.4 Another method for
converting small molecules into diesel components is the con-
densation of furans. According to this strategy, 2-methylfuran
derived from hemicellulose, is trimerized in the presence of an
acid catalyst to produce bisylvylpentanone.9 Subsequent high-
temperature hydrogenation is used to produce high-cetane
number 6-alkylundecanes. A third approach for making diesel
range compounds starts with the hydrogenation of levulinic acid,
obtained by the decomposition of HMF, to form γ-valerolactone,
which is then decarboxylated to form butene isomers and CO2.10
Butene oligomerization is then carried out to produce liquid
alkenes suitable for jet fuel and diesel.
Energy Biosciences Institute, Department of Chemical and
Biomolecular Engineering, University of California, Berkeley, CA
94720, USA. E-mail: bell@cchem.berkeley.edu
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c2gc35102a
An alternative method requiring less hydrogen to produce
usable diesel-range products is the etherification of the alcoholic
functionality of HMF. Previous work in this area has operated by
first replacing the OH group of HMF by a Cl atom, thereby
1626 | Green Chem., 2012, 14, 1626–1634
This journal is © The Royal Society of Chemistry 2012

View Online
forming 5-(chloromethyl)furfural (CMF).11 The nucleophilic
substitution of CMF with ethanol forms 5-(ethoxymethyl)-
furfural (EMF) and HCl. While this process gives high yields of
EMF, there are concerns about HCl recyclability and the intro-
duction of unreacted halides into automobile fuel systems, which
can cause premature wear. As the combustion of 17 wt% blends
of furanyl ethers in diesel have already been shown to perform
well in diesel engines with a 16% decrease in soot, there is con-
siderable interest in the direct formation of furanyl ethers for
fuels without formation of CMF.12 The high cetane number of
ethers13 and high energy density of EMF have also been cited as
a reason for implementation of these ethers, as EMF has an
energy density only 3% lower than typical gasoline and is 29%
greater than that of ethanol.11c,14 Furthermore, as EMF is already
widely established as a flavor and aroma component in wines
and beers, concerns about its toxicity are smaller compared to
other potential diesel additives.15 For this reason, ethers of HMF
and HMF derivatives were investigated to determine reaction
pathways and methods to maximize yields for these potential
alternative diesel molecules.
4 mL screw thread, phenolic cap, glass vials with a magnetic
spin bar. Reactions at elevated temperatures and/or under hydro-
gen atmosphere were carried out in a HEL Chem-Scan high
pressure parallel synthesizer equipped with 8 × 16 mL capacity
independently controlled stainless steel reactors. Reaction mix-
tures were analyzed using a Varian CP-3800 gas chromatograph
(GC) equipped with a flame ionization detector (FID) coupled to
a Varian 320-MS mass spectrometer (MS). Products were separ-
ated using a FactorFour capillary column (VF-5 ms, 30 m
length, 0.25 mm diameter) coated with a 0.25 mm thick station-
ary phase (5% phenyl and 95% dimethylpolysiloxane). Products
identified by GC/MS were confirmed by analysis of pure com-
ponents, which were also used to develop calibration curves
along with an internal standard (anthracene/dodecane) for quan-
titative analysis by GC/FID. Components with hydroxyl
functionalities were silylated in situ with N,O-bis(trimethylsilyl)-
trifluoroacetamide at 50 °C for 15 min before injection. The
silylation procedure was mandatory especially in case of ethoxy
analogues, to differentiate EMF (2a), HMF (1) and BHMF (7),
which eluted at the same retention time in the gas chromatogram
without silylation.
Experimental
Etherification of 5-(hydroxymethyl)furfural (1) in alcohols
Materials
A solution of 1 (0.126 g, 1 mmol) in alcohol (2 g) was added
with either soluble or solid acid catalyst (5 mol%) to a 4 mL
screw-cap vial equipped with a magnetic spin bar. The reaction
mixture was sealed, stirred (250 rpm) and heated at 75 °C on the
deck of the Symyx Core Module for 24 h before cooling to
ambient temperature. To neutralize the acid, solid anhydrous
Na2CO3 (50 mg) was added and stirred for an additional 15 min.
The products were then passed through a small pad of silica gel
and washed with ethyl acetate (3 × 10 mL) to remove solid par-
ticles, after which the solvents were rotary evaporated at reduced
pressure (150 mbar) to produce a crude reaction mixture which
was quantified after in situ silylation using anthracene as an
internal standard.
Chemicals were used as received without further purification.
Starting materials 5-(hydroxymethyl)furfural (5-HMF) and
D-(−)-fructose were purchased from Sigma-Aldrich, USA and
Acros, USA respectively. Acid catalysts such as sulfuric acid,
p-toluenesulfonic acid monohydrate (p-TSA·H2O), Amberlite
IR120 (4.4 meq g−1) and Dowex 50WX8 (4.4 meq g−1) were
from obtained from Acros, USA. Amberlyst-15 (4.7 meq g−1
)
and Dowex DR2030 (4.7 meq g−1) were obtained from Sigma-
Aldrich, USA. Silica sulfuric acid (SSA, 2.63 meq g−1) was
prepared using a literature procedure (see ESI†). Hydrogenation
catalysts including Pd/carbon (10 wt%), Pt/carbon (5 wt%),
Pt/alumina (5 wt%), Ru/carbon (5 wt%) and Rh/carbon (5 wt%)
were received from Sigma-Aldrich, USA. Pt/Sn alloys were pre-
pared with various proportions (1 wt% Pt in all cases) supported
on alumina (ESI†). Ethanol and n-butanol were obtained from
Acros, USA and HPLC grade ethyl acetate and hexanes were
obtained from Fisher, USA. Triethyl orthoformate and N,O-bis-
(trimethylsilyl)trifluoroacetamide with 1% TMSCl, and anhy-
drous solids (Na2CO3 and Na2SO4) were purchased from Acros,
USA. Ethyl levulinate (EL, 3a) was obtained from Sigma-
Aldrich. 5-(ethoxymethyl)furfural (EMF, 2a), 5-(ethoxymethyl)-
furfural diethylacetal (EMFDEA, 4a), 5-(hydroxymethyl)furfural
diethylacetal (HMFDEA, 6a), 2,5-bis(hydroxymethyl)furan
(BHMF, 7), 2,5-bis(ethoxymethyl)furan (BEMF, 8a), 5-(ethoxy-
methyl)furfuryl alcohol (EMFA, 9a) as well as corresponding n-
butyl analogues 2b, 3b, 4b, 6b, 8b and 9b were synthesized and
characterized by NMR and mass spectroscopic techniques (see
ESI†) and were used for quantification studies.
One-pot dehydration/etherification of D-(−)-fructose (5)
A suspension of 5 (0.180 g, 1 mmol) in alcohol (2 g) and the
desired acid catalyst (10 mol%) were introduced into a HEL
Chem-Scan batch reactor and flushed (3×) with nitrogen. The
sealed reactor was heated to specified temperature (100–120 °C)
with stirring (500 rpm) for 24–30 h. After cooling, the depressur-
ized reaction mixture was worked up in a manner similar to that
discussed above.
Hydrogenation of 5-(hydroxymethyl)furfural (1) over platinum
catalysts
A solution of 1 (0.126 g, 1 mmol) in ethanol (2 g) and the
platinum catalyst (0.25 mol%) were introduced into an HEL
batch reactor, sealed, flushed with 200 psi nitrogen (1×) as
well as hydrogen (3×) and pressurized with hydrogen (200 psi).
Reaction conditions were maintained as listed in Table 2. After
reaction, the mixture was cooled, depressurized, and filtered
through a small pad of silica gel by washing with ethyl acetate
Analysis of reaction products
Reactions at temperatures below the boiling point of the solvent
were carried out on the deck of a Symyx Core Module Robot
(CMR) equipped with heating and stirring capabilities using
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 1626–1634 | 1627

View Online
Scheme 1 Etherification of 5-(hydroxymethyl)furfural in alcohols. Reaction conditions: 5-HMF (1 mmol), H2SO4 (5 mol%), ROH (2 g), 75 °C,
24 h.
Table 1 Etherification of 5-HMF and one-pot dehydration/etherification of D-(−)-fructose in alcohol solvents
2 (%)
3 (%)
4 (%)
No
Starting materiala
Reaction conditions
R = Et
n-Bu
R = Et
n-Bu
R = Et
n-Bu
1.
2.
3.
4.
5.
6.
1
1
1
1
5
5
H2SO4 (5 mol%), 75 °C, 24 h
p-T SA (5 mol%), 75 °C, 24 h
H2SO4 (5 mol%), 120 °C, 30 h
p-TSA (5 mol%), 120 °C, 30 h
H2SO4 (10 mol%), 100 °C, 24 h
H2SO4 (10 mol%), 120 °C, 30 h
81
75
—
—
70
—
76
78
—
—
63
—
16
15
62
58
18
56
15
13
73
69
15
64
Trace
Trace
—
—
Trace
—
Trace
Trace
—
—
Trace
—
a Starting material (1 mmol) was dissolved or suspended (in case of 5) in 2 g of ROH.
(3 × 10 mL) in order to remove the solid catalyst particles. The
solvents were rotary evaporated and the products were quantified
after silylation.
reaction was continued for an additional 18 h while maintaining
these conditions. At the end of reaction, the products were
worked up in the same manner as described above.
Etherification of 2,5-bis(hydroxymethyl)furan (7)
Results and discussions
An alcoholic solution of 7 (0.128 g, 1 mmol dissolved in 2 g of
solvent) and Amberlyst-15 were placed in a 4 mL glass vial,
capped, then heated under magnetic stirring using the Symyx
Core Module. The reaction conditions used for these exper-
iments are given in Table 3. The cooled reaction mixture was
worked up as detailed above to give a crude product mixture
which was then silylated and quantified.
Etherification of 5-HMF (1) to 5-(alkoxymethyl)furfural (2) and
alkyl levulinates (3) with homogeneous acid catalysts
Initial experiments on the etherification of HMF (1) by ethanol
and butanol were carried out in the presence of sulfuric acid
(5 mol%) under the conditions shown in Scheme 1. The reaction
products observed were 5-(alkoxymethyl)furfural (2), alkyl levu-
linate (3) and trace amounts of dialkylacetal (4) (Scheme 1). The
distribution of products varied depending upon reaction con-
ditions (see Table 1). Alkyl ethers were the major products when
reactions were carried out at lower temperatures (Table 1, entries
1 and 2), whereas higher reaction temperatures (Table 1, entries
3 and 4) favored the formation of 3 exclusively. Alkyl levulinates
(3) result from the alcoholysis of 5-HMF (1) and should
be accompanied by an equivalent amount of ethyl formate.
While ethyl levulinate was readily observed in the ethanolic
reaction mixture, ethyl formate elutes with the solvent peak in
the gas chromatogram and was, therefore, not directly observed.
However, the presence of n-butylformate was readily observed
when n-butanol was used as the solvent for etherification.
Reductive etherification of 5-(hydroxymethyl)furfural (1)
A solution of 1 (0.126 g, 1 mmol) in alcohol (2 g), with the
platinum catalyst (0.25 mol%), and Amberlyst-15 (11 mg, 5 mol
%) were placed in an HEL Chem-Scan reactor. The reaction
mixture was sealed, flushed with 200 psi nitrogen (1×) as well as
hydrogen (3×), and pressurized with hydrogen (200 psi). The
reaction conditions used for these experiments are given in
Table 4. The work up of the reaction mixture and the quantitative
analysis of the products were identical to those described above.
Recognizing that 5-HMF is the dehydration product of
hexoses like glucose and fructose and that the dehydration of
these sugars is promoted by acid catalysts similar to those used
to promote the etherification of HMF, next we examined the pos-
sibilities of carrying out a one-pot dehydration/etherification
starting from D-(−)-fructose (5). Under conditions similar to
those used for etherification, 5 did not react. Exploration of
various reaction conditions revealed that the dehydration of fruc-
tose demands a higher acid loading (10 mol%) and temperature
One-pot dehydration/reductive etherification of
D-(−)-fructose (5)
A suspension of 5 (0.180 g, 1 mmol) in alcohol (2 g), Amber-
lyst-15 (10 mol%), and Pt1Sn1/Al2O3 (0.5 mol%) were placed in
an HEL Chem-Scan reactor and flushed (3×) with nitrogen. The
sealed reactor was heated to 110 °C with stirring (500 rpm) for
30 h. The operating temperature was then reduced to 60 °C,
flushed (3×), and pressurized with hydrogen (200 psi). The
1628 | Green Chem., 2012, 14, 1626–1634
This journal is © The Royal Society of Chemistry 2012

View Online
Table 2 Platinum catalyzed hydrogenation of 5-HMF (1) in ethanol
Yields (%)
Entry
Catalyst
Temp. (°C)
Time (h)
7
6
1.
2.
3.
4.
5.
6.
7.
Pt/C
Pt/Al2O3
Pt/C
Pt/Al2O3
Pt1Sn0.25/Al2O3
Pt1Sn0.5/Al2O3
Pt1Sn1/Al2O3
23
23
60
60
60
60
60
18
18
5
5
5
5
5
82
85
17
45
55
80
82
4
3
3
5
16
5
8
Table 3 Optimization for Amberlyst-15 catalyzed etherification of BHMF (7) in ethanol
Conditions
Yields (%)
Entry
Cat. loading
Temp. (°C)
Time (h)
8a
9a
1.
2.
3.
4.
5.
6.
7.
5 mol%
10 mol%
20 mol%
5 mol%
5 mol%
5 mol%
5 mol%
40
40
40
40
60
60
75
5
5
5
16
5
16
16
30
60
71
80
74
66
39
43
16
3
6
2
1
0
presence of solid catalysts (5 mol% acid loading) as illustrated in
Scheme 3.
Table 4 Optimization of reaction conditions for one-pot reactions from
1 with Pt/alumina and Amberlyst-15 in ethanol
Resins with sulfonic acid functionalities showed promising
Conditions
Yields (%)
activities for converting
1 to its etherified derivatives
(Scheme 3). Significant amounts (up to 33%) of the diethylacetal
product, 4a, were found in the product, demonstrating the reac-
tivity of the aldehyde functionality of 1 in the presence of such
catalysts. The most active resin catalyst was Dowex DR2030, fol-
lowed by Amberlyst-15, Dowex 50WX8, and Amberlite IR120.
Silica sulfuric acid exhibited activity slightly higher than Amber-
lite IR120 (Fig. 1). Interestingly, the distribution of products pro-
duced was nearly independent of the type of catalyst use. An
attempt to use H-FAU for the etherification of 1 with ethanol
was made but not pursued further, since this catalyst deactivated
rapidly due to catalyst pore plugging by bulky reaction inter-
mediates 4a, 6a and other oligomers.
Entry Pt/Al2O3
A-15
Temp. (°C) Time (h) 8a 9a 4a 6a
1.
2.
3.
4.
5.
0.25 mol% 5 mol% 60
0.5 mol% 5 mol% 60
1.0 mol% 5 mol% 60
0.5 mol% 5 mol% 60
0.5 mol% 5 mol% 40
18
18
18
5
50
55
59
31 15
33 26
4
7
7
2
2
1
6
8
6
—
—
—
—
18
than those used for etherification in order to form 2, 3 and 4
(Table 1, entry 5 and Scheme 2). At a still higher temperature
and longer reaction time (Table 1, entry 6) only alkyl levulinates,
3, were observed.
Encouraged by these results, a number of solid acid catalysts
were screened, including zeolites, resins and, silica based acid
catalysts. Ethanol solutions of 1 were heated to 75 °C in the
A plausible mechanism for the etherification of 1 is shown in
Scheme 4. The diethylacetalization of 5-HMF (1) to 6a is almost
instantaneous in the presence of an acid catalyst, as determined
by gas chromatographic analysis of the products. Product 6a is
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 1626–1634 | 1629

View Online
Scheme 2 One-pot dehydration/etherification of D-(−)-fructose in alcohols. Reaction conditions: D-(−)-fructose (5, 1 mmol), H2SO4 (10 mol%),
ROH (2 g), 100 °C, 24 h.
Scheme 3 Etherification of 5-HMF in ethanol using heterogeneous catalysts. Reaction conditions: 5-HMF (1 mmol), solid acid cat. (5 mol%), EtOH
(2 g), 75 °C, 24 h.
reaction conditions were optimized for complete conversion of 5
in the case of Amberlyst-15 (10 mol%, 110 °C, 30 h) and then
used for other catalysts. As seen in Fig. 2, all of the solid cata-
lysts tested gave high yields of products, principally 2a, with the
exception of Amberlite IR120, which also gave a modest yield
of 1, formed by dehydration of 5. It should also be noted that
because the reaction was carried out at 110 °C only a small yield
of diethylacetal 4a was observed (5–10%), consistent with what
is projected by Scheme 4. A comparison of the yield of 3a
reported in Fig. 1 and 2 suggests that formation of ethyl levuli-
nate (3a) is favored at elevated temperatures. Attempts to
carry out the corresponding one-pot, dehydrative etherification
of glucose in alcohol was not successful and produced alkyl-D-
Fig. 1 Product distributions during etherification of 1 in ethanol with
various solid acid catalysts. Reaction conditions: 5-HMF (1 mmol),
solid acid cat. (5 mol%), EtOH (2 g), 75 °C, 24 h.
glucopyranoside as the only major product. While the ketohex-
ose (furanose) readily undergoes dehydration (as is the case
for fructose), aldohexose (pyranose) must first isomerize to keto-
hexose and then undergo dehydration to 1. Most likely, due to
the ease of formation of stable alkylglucopyranoside, the isomer-
ization and successive dehydration pathway is least preferred in
alcoholic solutions of glucose in the presence of an acid catalyst.
Application of the optimized etherification reaction conditions
for 1 and 5 in n-butanol produced very similar results.
formed by protonation of 1, followed by nucleophilic attack of
ethanol to produce the hemiacetal 1a. The formation of 4a from
6a is envisioned to occur via similar sequence of steps.
The distribution of final products depends on the reaction
temperature and the presence of water. In experiments carried
out with Amberlyst-15 as the catalyst, EMFDEA (4a) was found
to equilibrate with 2a, the extent depending on the temperature
and the concentration of water content in the reaction medium.16
Increasing reaction temperature from 75 °C to 110 °C increased
the yield of 2a from 55% (compare Fig. 1 and 2) to 71% and the
yield of 3a from 8% to 16%, whereas the yield of 4a decreased
from 31% to 10%. Addition of a water scavenger, such as
triethylorthoformate (2.5 equivalents), to the reaction mixture
under optimized conditions (5 mol% solid acid cat., 75 °C) gave
exclusively 4a in 71% yield after 1.5 h of reaction. Ethoxy-
methylfurfural (2a) underwent further ethanolysis to produce
equimolar amounts of ethyl levulinate (3a) and ethyl formate
(3a′) at elevated temperature (120 °C).
Chemo-selective carbonyl reduction of 5-HMF under metal
catalyzed hydrogenation
Once an optimized set of reaction conditions was found for
converting either 5-HMF (1) or D-(−)-fructose (5) to the alkoxy-
methylfurfurals (2a,b), dialkylacetals (4a,b) and alkyl levulinates
(3a,b) using heterogeneous catalysts, attention was focused on
producing compounds containing only ether linkages. The
motivation for this choice is the ethers should be acceptable can-
didates for biodiesel, since such compounds are relatively inert,
have low freezing points, and high cetane numbers.13 With this
aim in mind, we sought catalysts for the selective hydrogenation
of the aldehyde functionality of 1 in order to convert it into
dihydroxyl derivative 7 (2,5-bis(hydroxymethyl)furan, BHMF).
One-pot dehydration/etherification of fructose (5) suspended
in ethanol was probed using solid acid catalysts as well. The
1630 | Green Chem., 2012, 14, 1626–1634
This journal is © The Royal Society of Chemistry 2012

View Online
Scheme 4 Plausible mechanistic pathways for the observed products during etherification of 1.
likely reason why the results reported in ref. 17 could not be
reproduced. Constrained by this stringent requirement, we
systematically screened a number of catalysts including carbon-
supported ruthenium, rhodium, palladium, and platinum. Pd/C
and Rh/C exhibited relatively little functional group selectivity,
and produced numerous products under the reaction conditions
used (0.5–1.0 mol% cat., 15–200 psi of H2 pressure, and
ambient temperature). Under similar reaction conditions, Ru/C
produced only 6a along with starting material 1.
Reactions carried out using Pt/C or Pt/Al2O3 proved to be
more promising, providing 7 in high yields (see entries 1 and 2,
Table 2) even for low catalyst loadings (0.25 mol%). Attempts to
carry out hydrogenations at elevated temperature (60 °C) in
order to minimize the duration of the reaction (5 h), reduced
the yield of 7 significantly. High yields of 7 could, however, be
obtained with PtSn bimetallic catalysts, which have been
reported to show superior activity compared to platinum catalysts
for the chemo-selective reduction of furfural to the correspond-
ing alcohol.18 Table 2 (entries 5, 6, and 7) show that the addition
of Sn increases its activity and selectivity for the formation of 7.
In fact yields of up to 82% of 7 could be obtained with a Pt1Sn1/
Al2O3 operated at 60 °C for 5 h.
Fig. 2 Product distributions during one-pot dehydrative etherification
of 5 in ethanol with various solid acid catalysts. Reaction conditions: D-
(−)-fructose (1 mmol), solid acid cat. (10 mol%), EtOH (2 g), 110 °C,
30 h.
Although palladium catalyzed reduction of 5-HMF in 1-propanol
containing soluble acid catalyst is reported to yield 19% of 5-(n-
propoxymethyl)furfuryl alcohol,17 these results could not be
reproduced in ethanol. Indeed, palladium catalyzed reduction of
1 in ethanol produced compounds with either partial or complete
reduction of the furan ring. Once the aromaticity of the furan
ring is destroyed, ether formation no longer readily occurs due to
the lack of formation of the oxocarbocation intermediate in the
furan ring. Over-hydrogenation of the ring is, therefore, the
Acid catalyzed etherification of 2,5-BHMF in alcoholic solvents
Optimization of the reaction conditions for etherification of
BHMF (7) was then undertaken using Amberlyst-15 due to its
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 1626–1634 | 1631

View Online
Scheme 5 One-pot reductive etherification of 5-HMF (1). Reaction conditions: 5-HMF (1 mmol), Amberlyst-15 (5 mol%), Pt/alumina (0.25 mol%),
H2 (200 Psi), EtOH (2 g), 75 °C, 24 h.
outstanding performance for the etherification of 5-HMF. Unlike
1, we anticipated an enhanced reactivity for 7 towards acid cata-
lyzed transformations facilitated by ease of formation of oxocar-
bocation intermediate as a result of electron richness of the furan
ring. Our initial experiments on etherification of 7 at lower temp-
eratures (40 °C) for short durations (5 h) in ethanol revealed the
presence of 5-(ethoxymethyl)furfuryl alcohol (EMFA, 9a) and
2,5-bis(ethoxymethyl)furan (BEMF, 8a) in 43, 30% yields
respectively (entry 1, Table 3) in the reaction mixture. To maxi-
mize the desired product yield (diether 8a), 20 mol% loading of
Amberlyst-15 was required under these conditions (entry 3).
Nevertheless, it was possible to achieve these yields with lower
acid loadings (5 mol%) by either prolonged stirring at 40 °C
Fig. 3 Performance of various platinum catalysts (0.25 mol%) during
one-pot reductive etherification of 1 (60 °C, 18 h) in ethanol.
(entry 4) or slightly elevating the temperature for a short duration
(60 °C, 5 h, entry 5). It is appropriate to mention that selectivity
to 8a diminished significantly upon prolonged heating at higher
temperatures as shown in entries 6 and 7. Under the reaction
(Scheme 6). The mono-ether 9a could also undergo further
conditions shown in entry 5, Dowex DR2030 also produced 8a
etherification to give 8a under same reaction conditions.
and 9a in yields of 66% and 3%, respectively. The correspond-
Since alumina-supported PtSn alloys had been proven to be
very active for the reduction of the carbonyl group of 1 even
ing yields of n-butyl ethers 8b (56%), 9b (10%) could be
obtained using similar conditions (5 mol% Dowex DR2030,
under low loadings, a systematic screening was undertaken to
60 °C, 5 h), and n-butanol as the solvent.
establish their efficiency for one-pot reductive etherification of
5-HMF (1). Based on product distributions observed in the
experiments carried out with different PtSn catalysts together
with Amberlyst-15 under identical reaction conditions (Fig. 3), it
was established that the best activity occurred for Pt/Sn = 1.
Using the optimized conditions for reductive etherification
(0.25 mol% Pt1Sn1/Al2O3, 5 mol% Amberlyst-15, 60 °C, 18 h)
of 1 in n-butanol a 47% yield of 2,5-bis(n-butoxymethyl)furan
(BBMF, 8b) and a 12% yield of 5-(n-butoxymethyl)furfuryl
alcohol (BMFA, 9b) were obtained.
One-pot reductive etherification protocol for the production of
2,5-bis(alkoxymethyl)furan
We also investigated the possibility of producing 8a directly
from 1. For this purpose, the hydrogenation and etherification
steps were carried out together in the presence of Amberlyst-15
(5 mol%) and Pt/Al2O3 (0.25 mol%) in the presence of hydrogen
(200 psi) for 18 h (Scheme 5). A 50% yield of the bis-ether 8a
could be obtained along with small yields of intermediates 4a,
6a, and 9a (see Table 4, entry 1). Increasing the platinum
catalyst loading (up to 1 mol%, entry 1–3) slightly improved
the yield of 8a, whereas decreasing either the temperature or the
duration of reaction led to incomplete reactions by producing a
mixture of mono and bis ethers (Table 4, entries 4 and 5).
As discussed earlier, the formation of diethylacetals 6a and
4a, presumably through the intermediate 1a, occurs readily in
ethanolic solutions of 1 under acidic conditions. The presence of
acetals 6a and 4a in the products obtained using Pt/C and
Pt/Al2O3 (see Fig. 3) suggests that the reaction follows a reductive
etherification pathway rather than sequential carbonyl reduction
to BHMF (7) followed by etherification during one-pot reac-
tions.19 According to reductive etherification protocol, protona-
tion of 1 by the acid catalyst in ethanol leads to intermediates 1a,
6a and 4a which then undergo hydrogenolysis over the platinum
catalyst to produce the corresponding ethers 9a and 8a
One-pot sequential dehydration/reductive etherification of
D-(−)-fructose (5) to 2,5-bis(alkoxymethyl)furan
A further objective was to determine whether furfuryl ethers 8
and 9 could be obtained directly from fructose (5). Towards
this end, an ethanolic suspension of 5, Amberlyst-15 (5 mol%),
and Pt1Sn1/alumina (0.25 mol%) was heated to 110 °C under a
hydrogen atmosphere (200 psi). A number of products was
observed, but few of the desired products, suggesting that the
operating temperature was too high for the reductive etherifica-
tion step of the sequence (entry 7, Table 3). Nevertheless, a one-
pot conversion of 5 to desired products could be realized by car-
rying out each of the reactions in sequence, as depicted in
Scheme 7. While both catalysts were present during the reaction,
the lack of H2 led to dehydration/etherification of 5 to produce 2,
3 and 4 at 110 °C. In the next step of the sequence, the operating
1632 | Green Chem., 2012, 14, 1626–1634
This journal is © The Royal Society of Chemistry 2012

View Online
Scheme 6 Plausible reaction pathways for one-pot reductive etherification of 5-HMF (1).
Scheme 7 One-pot sequential dehydration/reductive etherification of D-(−)-fructose (5) to furfuryl ethers in alcohols. Reagents and conditions: D-
(−)-fructose (1 mmol), Amberlyst-15 (5 mol%), Pt1Sn1/alumina (0.25 mol%), ROH (2 g), 110 °C, 30 h then H2 (200 psi), 60 °C, 18 h.
temperature was reduced to 60 °C and 200 psi of H2 was pro-
vided. Both ethanol and n-butanol produced the corresponding
alkoxymethylfurans 8a,b and 9a,b in reasonable yields follow-
ing this procedure. During the optimization of these reactions we
found an increase in the PtSn catalyst loading (0.5 mol%) helped
facilitate higher yields (Table 4). It should also be noted that the
yield of alkyl levulinate 3 formed during the first step was unaf-
fected by the reduction conditions imposed for the second step.
levulinates 3 were identified during etherification of 1, as well as
one-pot dehydrative etherification of D-(−)-fructose (5) in the
presence of an acid catalyst. The distribution of products could
be tuned by changing the conditions. For example, lower temp-
eratures favored alkoxymethylfurfural 2 and dialkylacetal 4,
whereas elevated temperatures produced levulinates 3 exclu-
sively. The dehydrative etherification of 5 occurred at a slightly
higher temperature than the etherification of 1, indicating that
multiple dehydrations are necessary to transform fructose into its
furanic counterpart. Of the various heterogeneous acid catalysts
screened, Amberlyst-15 was found to be highly active for these
reactions. Alumina-supported Pt and PtSn alloys were found to
be very active and selective for the reduction of the carbonyl
functionality of 1. Interestingly, one-pot reductive etherification
and one-pot sequential dehydration/reductive etherification pro-
tocols produced 2,5-bis(alkoxymethyl)furan 8 directly from 1
and 5, respectively, in reasonable yields. The molecular structure
Conclusions
Acid catalyzed etherification and reductive etherification of 5-
(hydroxymethyl)furfural (1) in alcoholic solvents provide routes
to 5-(alkoxymethyl)furfurals (2a,b) and 2,5-bis(alkoxymethyl)-
furans (8a,b), respectively, products that are potential bio-diesel
fuels. Alkoxymethyl derivatives 2, dialkylacetals 4, and alkyl
This journal is © The Royal Society of Chemistry 2012
Green Chem., 2012, 14, 1626–1634 | 1633

View Online
8 (a) S. Dee and A. T. Bell, Green Chem., 2011, 13, 1467–1475;
(b) S. J. Dee and A. T. Bell, ChemSusChem, 2011, 4, 1166–1173;
(c) M. Chidambaram and A. T. Bell, Green Chem., 2010, 12, 1253–1262.
9 A. Corma, O. de la Torre, M. Renz and N. Villandier, Angew. Chem., Int.
Ed., 2011, 50, 2375–2378.
of 8 features only ether linkages. Such compounds are poten-
tially good candidates for diesel fuel because they are stable,
have low freezing points, and high cetane numbers. Both the
etherification and reductive etherification of 1 in ethanol and
butanol is believed to be initiated by the dialkylacetal formation
from the aldehyde functionality of 6. A detailed study of the
mechanism and kinetics of acid catalyzed etherifications of bio-
mass derived platform molecules has been initiated and will be
communicated soon.
10 J. Q. Bond, D. M. Alonso, D. Wang, R. M. West and J. A. Dumesic,
Science, 2010, 327, 1110–1114.
11 (a) M. Mascal and E. B. Nikitin, Angew. Chem., Int. Ed., 2008, 47,
7924–7926; (b) M. Mascal and E. B. Nikitin, ChemSusChem, 2009, 2,
859–861; (c) M. Mascal and E. B. Nikitin, ChemSusChem, 2009, 2, 423–
426; (d) M. Mascal and E. B. Nikitin, Green Chem., 2010, 12, 370–373.
12 (a) P. Imhof, A. S. Dias, E. de Jong and G.-J. Gruter, presented in part at
the NAM Abstract, 2009; (b) O. Casanova, S. Iborra and A. Corma,
J. Catal., 2010, 275, 236–242; (c) P. Lanzafame, D. M. Temi,
S. Perathoner, G. Centi, A. Macario, A. Aloise and G. Giordano, Catal.
Today, 2011, 175, 435–441; (d) G. J. M. Gruter and F. Dautzenberg, Eur.
Pat. Appl, 2007, EP1834950; (e) G. J. M. Gruter and L. E. Manzer, PCT
Int. Appl, 2009, WO2009030510; (f) G. J. M. Gruter and F. Dautzenberg,
US Pat. Appl, 2011, US20110082304.
Acknowledgements
This work was supported by BP through the Energy Biosciences
Institute.
13 M. J. Murphy, J. D. Taylor and R. L. McCormick, Compendium of Exper-
imental Cetane Number Data, National Renewable Energy Laboratory,
Golden, CO, 2004.
14 E.-J. Ras, S. Maisuls, P. Haesakkers, G.-J. Gruter and G. Rothenberg,
Adv. Synth. Catal., 2009, 351, 3175–3185.
15 (a) B. Vanderhaegen, H. Neven, S. Coghe, K. J. Verstrepen, H. Verachtert
and G. Derdelinckx, J. Agric. Food Chem., 2003, 51, 6782–6790;
(b) J. S. Câmara, M. A. Alves and J. C. Marques, Anal. Chim. Acta,
2006, 563, 188–197; (c) H. Oliveira e Silva, P. Guedes de Pinho,
B. P. Machado, T. Hogg, J. C. Marques, J. S. Câmara, F. Albuquerque
and A. C. Silva Ferreira, J. Agric. Food Chem., 2008, 56, 11989–11996.
16 M. R. Altıokka and H. L. Hoşgün, Ind. Eng. Chem. Res., 2007, 46,
1058–1062.
17 G. C. A. Luijkx, N. P. M. Huck, F. van Rantwijk, L. Maat and H. van
Bekkum, Heterocycles, 2009, 77, 1037–1044.
18 V. Vetere, A. B. Merlo, J. F. Ruggera and M. L. Casella, J. Braz. Chem.
Soc., 2010, 21, 914–920.
19 (a) S. H. Lee, Y. J. Park and C. M. Yoon, Tetrahedron Lett., 1999, 40,
6049–6050; (b) L. J. GooBen and C. Linder, Synlett, 2006, 3489–3491;
(c) T. T. Pham, S. P. Crossley, T. Sooknoi, L. L. Lobban, D. E. Resasco
and R. G. Mallinson, Appl. Catal., A, 2010, 379, 135–140.
References
1 N. A. Al-Soof, F. Siala, B. Aklil, M. Taeb, M. Khesali, M. Mazraati,
B. Lubiantara, T. Najah, A. Alawami, C. Clemenz, N. Guerer,
G. Brennand, J. Ban, J. Spitzy, D. Linton, J. Griffin, M. Tallett, P. Steiner
and U. Szalkowska, World oil outlook 2010, 2010.
2 Renewable Energy Road Map, Renewable energies in the 21st century:
building a more sustainable future, Communication from the Commission
to the Council and the European Parliament, Brussels, 2007.
3 J. R. Regalbuto, Science, 2009, 325, 822–824.
4 D. M. Alonso, J. Q. Bond and J. A. Dumesic, Green Chem., 2010, 12,
1493–1513.
5 G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106, 4044–
4098.
6 C. Somerville, H. Youngs, C. Taylor, S. C. Davis and S. P. Long, Science,
2010, 329, 790–792.
7 (a) G. W. Huber, J. N. Chheda, C. J. Barrett and J. A. Dumesic, Science,
2005, 308, 1446–1450; (b) C. J. Barrett, J. N. Chheda, G. W. Huber and
J. A. Dumesic, Appl. Catal., B, 2006, 66, 111–118.
1634 | Green Chem., 2012, 14, 1626–1634
This journal is © The Royal Society of Chemistry 2012