Selective oxidation of furfural in a bi-phasic system with homogeneous acid catalyst

Article abstract of DOI:10.1016/j.cattod.2015.11.036

The selective catalytic oxidation of furfural to 2(5H)-furanone, succinic acid (SA) and maleic acid (MA) was studied. Under optimized conditions, furfural was oxidized to 2(5H)-furanone with a yield of 60–62% in an aqueous/organic bi-phasic system using 1,2-dichloroethane or ethyl acetate as the solvent and formic acid as the catalyst, while the total yield of SA and MA was 15–20%. Compared with other homogeneous and heterogeneous acid catalysts, formic acid gave a much higher selectivity to 2(5H)-furanone because it reacted with hydrogen peroxide to generate performic acid that had a strong oxidizing nature and good solubility in both the aqueous and organic phases. The solvent had a significant influence on the product distribution. A simplified reaction network was established to quantitatively analyze the solvent effect based on the reaction rate constants. In the homogeneous system, the yield of 2(5H)-furanone decreased while the yield of SA increased with an increasing dielectric constant of the solvent. The formic acid/furfural molar ratio, reaction temperature and furfural concentration were optimized for the selective oxidation of furfural to 2(5H)-furanone in the bi-phasic reaction system.

Full text of DOI:10.1016/j.cattod.2015.11.036

G Model
CATTOD-9941; No. of Pages8
ARTICLE IN PRESS
Catalysis Today xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Selective oxidation of furfural in a bi-phasic system with
homogeneous acid catalyst
∗
Xiaodan Li, Xiaocheng Lan, Tiefeng Wang
Beijing Key Laboratory of Green Reaction Engineering and Technology, Department of Chemical Engineering, Tsinghua University, Beijing 100084, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 13 October 2015
Received in revised form
The selective catalytic oxidation of furfural to 2(5H)-furanone, succinic acid (SA) and maleic acid (MA) was
studied. Under optimized conditions, furfural was oxidized to 2(5H)-furanone with a yield of 60–62% in
an aqueous/organic bi-phasic system using 1,2-dichloroethane or ethyl acetate as the solvent and formic
acid as the catalyst, while the total yield of SA and MA was 15–20%. Compared with other homoge-
neous and heterogeneous acid catalysts, formic acid gave a much higher selectivity to 2(5H)-furanone
because it reacted with hydrogen peroxide to generate performic acid that had a strong oxidizing nature
and good solubility in both the aqueous and organic phases. The solvent had a significant influence on
the product distribution. A simplified reaction network was established to quantitatively analyze the
solvent effect based on the reaction rate constants. In the homogeneous system, the yield of 2(5H)-
furanone decreased while the yield of SA increased with an increasing dielectric constant of the solvent.
The formic acid/furfural molar ratio, reaction temperature and furfural concentration were optimized for
the selective oxidation of furfural to 2(5H)-furanone in the bi-phasic reaction system.
2
8 November 2015
Accepted 30 November 2015
Available online xxx
Keywords:
Furfural
Catalytic oxidation
2
(5H)-furanone
Solvent effect
Reaction network
©
2016 Elsevier B.V. All rights reserved.
1
. Introduction
Furanone can be synthesized by deoxygenation of substituted
butanoic acids at high temperature, transformation of hydroxybu-
tyrolactones with acids or amines, hydrolysis of 2-methoxyfuran
and cyclocarbonylation of terminal alkynols in the presence of
palladium [19–22]. Nevertheless, the above methods are deficient
owing to their expensive reagents, complicated processes, extreme
reaction conditions and low yield. Currently, 2(5H)-furanone is
mainly produced by furfural oxidation. Badovskay [23] reported
the oxidation of furfural to 2(5H)-furanone with a yield of 25% by
autocatalysis, and the yield was not significantly improved in the
presence of Mo (VI) or Cr (VI). Cao et al. [13]. found that when the
autocatalytic oxidation of furfural was carried out in a bi-phasic
system using dichloroethane as the solvent, a 37% yield of 2(5H)-
furanone was obtained, however the reaction time was longer than
10 hours. Poskonin [24] reported the synthesis of 2(5H)-furanone
from furfural in water using niobium (V) acetate tetrahydrate as
the catalyst and hydrogen peroxide as the oxidant, but more than
80 hours were needed to obtain a 60% yield. Despite decades of
study, problems with furfural oxidation to 2(5H)-furanone still
exist in the moderate yield (30–50%), long reaction time and the
use of chlorinated solvents. The main utility of 2(5H)-furanone at
present is limited to the intermediate for the synthesis of biologi-
cally active natural products [25].
With increasing consumption of energy and diminishing sup-
ply of fossil fuels, substantial interest has been devoted to search
for renewable energy resources in the world. In this respect,
biomass, which is renewable, environmental-friendly, abundant
and widespread, is considered to be an ideal substitute for tradi-
tional fossil fuels [1–4]. In recent years, various catalytic conversion
strategies have been developed for the conversion of biomass to
bio-based platform compounds, which can be further converted to
a wide range of target products [5–9]. Among these renewable plat-
form compounds, furfural, produced by hydrolysis and dehydration
of xylan contained in hemicellulose, is one of the most important
building blocks for bio-refinery [10–12]. Catalytic oxidation has
been used to convert furfural to many chemical intermediates and
end products, including furanone, furoic acid, fumaric acid (FA),
succinic acid (SA) and maleic acid (MA) [6,13–18]. Most of the cur-
rent studies focus on the selective oxidation of furfural to diacids or
acid anhydrides, especially to SA, MA and maleic anhydride. How-
ever, the studies on the oxidation of furfural to 2(5H)-furanone are
very limited.
2
(5H)-Furanone is an ideal material for the production of lac-
∗ Corresponding author.
E-mail address: wangtf@tsinghua.edu.cn (T. Wang).
tones and diols in terms of its chemical structure. In our previous
http://dx.doi.org/10.1016/j.cattod.2015.11.036
0
920-5861/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: X. Li, et al., Selective oxidation of furfural in a bi-phasic system with homogeneous acid catalyst, Catal.
Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.11.036

G Model
CATTOD-9941; No. of Pages8
ARTICLE IN PRESS
X. Li et al. / Catalysis Today xxx (2015) xxx–xxx
2
work, 2(5H)-furanone was hydrogenated to ␥-butyrolactone (GBL)
2.2.2. Oxidation of 2(5H)-furanone
with a high yield using supported metal catalysts [26]. The use of
The oxidation of 2(5H)-furanone was carried out in a sim-
ilar apparatus to that for the oxidation of furfural. For each
experiment, 4.0 g of 2(5H)-furanone was dissolved in 10 mL of 1,2-
dichloroethane, then 1.5 mL of formic acid and 1.5 mL of distilled
water were added to the mixture. When the reaction temperature
reached the set value, 10 mL of hydrogen peroxide was added drop-
wise to the mixture during the first hour while stirring. The reaction
mixture was sampled and analyzed by GC and HPLC.
2
(5H)-furanone as the intermediate for the conversion of furfural
to C4 lactones and diols can improve the atom economy of the con-
version path and reduce the cost caused by harsh conditions, in
comparison with other biomass-derived intermediates like diacids
and acid anhydrides. Thus, the highly selective oxidation of furfural
to 2(5H)-furanone can provide a better approach to produce C4 lac-
tones and diols from furfural, which have been widely used in the
fields of medicine, chemical, plastics, electric and others [27,28].
Hydrogen peroxide is usually used as the oxidant for furfural oxida-
tion to 2(5H)-furanone, because the non-catalytic polymerization
of furfural is severe when furfural interacts with molecular oxy-
gen [29]. The control of selectivity is a great challenge for furfural
oxidation to 2(5H)-furanone, as many side reactions including non-
catalytic polymerization, oxidation to diacids and deep oxidation
may occur.
In this work, different acid catalysts were evaluated for the
oxidation of furfural to 2(5H)-furanone in both the bi-phasic and
homogeneous systems. A simplified reaction network was estab-
lished, and the reaction rate constants and apparent activation
energies were calculated to analyze the solvent effect. The solvent
effect was investigated in homogeneous, bi-phasic and tri-phasic
systems using a variety of solvents with different polarities. The
product distribution was correlated with the solvent property to
guide the selection of solvent for improved yield of 2(5H)-furanone.
Additionally, the effects of the formicacid/furfuralmolar ratio, reac-
tion temperature and furfural concentration were studied in detail.
2.3. Analytical methods
Gas chromatograph (GC 7900 II, Techcomp Instrument
Company) equipped with
a
super-wax capillary column
(30 m × 0.25 mm × 0.5 ␮m) and an FID detector was used to
analyze the contents of furfural and 2(5H)-furanone in the reac-
tion mixture. The diacids including SA and MA were analyzed
with a Shimazu 10 AT–VP HPLC equipped with a C18 column
(150 mm × 4.6 mm) and a UV detector. For HPLC measurement, the
samples were diluted 500 times with distilled water; The mobile
phase was methanol with distilled water (3/97, v/v) containing
KH PO4 (0.02 mol/L), and its pH was adjusted to 2.75 by H PO ;
2
3
4
The flow rate was fixed at 1.0 mL/min and the temperature of the
column was kept at 303 K; The injection volume was 20 ␮L.
The conversion of furfural and the selectivity and yield of prod-
ucts are calculated as follows:
ꢀ
ꢁ
0
n
− n
fur
fur
Furfuralconversion % =
(
)
× 100
(1)
(2)
(3)
0
n
fur
npro
0
fur
Productselectivity (%) = ꢀ
ꢁ × 100
2
. Materials and methods
n
− nfur
2.1. Materials
npro
Productyield (%) =
× 100
0
fur
n
Methanol, tetrahydrofuran, tetrachloromethane, benzene,
where n0 is the initial mole quantity of furfural loaded into the
toluene, cyclohexane, iso-octane, n-heptane, ethyl acetate, sulfuric
acid (98%), phosphoric acid (85%) and sodium sulfate were all
of analytical grade and purchased from Beijing Chemical Works.
Isopropanol (SCR Beijing, 99.5%), 1,2-dichloroethane (J&K Chem-
ical, 99%), ␥-butyrolactone (Xiya Reagent, 99%), formic acid (SCR
Beijing, 88%) and hydrogen peroxide (Beijing Modern Oriental
Fine Chemistry Co., Ltd., 30%) were purchased. The above reagents
were used without further purification. Furfural (SCR Beijing, 99%)
was purified by vacuum distillation before use. The strong acid
ion exchange resin Amberlyst-15 was purchased from Aladdin
Industrial Corporation, and ZrTiO4 was prepared according to the
literature [30]. The corresponding precursors were co-precipitated
to synthesize Ti(OH) –Zr(OH) , and this hydroxide was calcined at
fur
reactor, and nfur and npro are the moles of furfural and products,
respectively. The mole quantity of each component n in the homo-
i
geneous system and bi-phasic system was calculated based on Eq.
(4) and Eq. (5), respectively:
n = C × V
(4)
i
i
where Ci is the concentration of the component measured by GC or
HPLC, and V is the volume of the reaction mixture at a given time.
ni = Ci,aq × Vaq + Ci,or × Vor
(5)
where Ci,aq and Ci,or refer to the concentration of the component in
the aqueous and organic phases, respectively, and Vaq and Vor refer
to the volume of the aqueous and organic phases at a given time,
respectively.
4
4
9
23 K to obtain ZrTiO4.
2.2. Oxidation reaction
3. Results and discussion
2
.2.1. Oxidation of furfural
All the furfural oxidation reactions were carried out in a three-
3.1. Homogeneous and heterogeneous acid catalysts
necked flask of 50 mL, equipped with a condenser to reduce
volatilization loss of the solvent. An oil bath was used to control
the reaction temperature. In a typical experiment, 10 mL of sol-
vent, 1.5 mL of distilled water, 4.0 g of furfural, 4.0 g of sodium
sulfate and an appropriate amount of acid catalyst were added to
the flask and heated to the reaction temperature. When the reaction
mixture reached the set temperature, 10 mL of hydrogen peroxide
The furfural oxidation reaction was first studied in a bi-phasic
system with 1,2-dichloroethane as the solvent and in a homoge-
neous system with methanol as the solvent, focusing on the effect
of acid catalysts. The results are shown in Fig. 1. Among the selected
homogeneous catalysts, formic acid had the best performance for
furfural oxidation to 2(5H)-furanone in both the bi-phasic and
homogeneous systems. The yield of 2(5H)-furanone reached 60.3%
in the bi-phasic system, while it decreased to 13.1% in the homo-
(
30%) was added to the reactor dropwise during the first hour while
stirring. The reaction mixture was sampled and analyzed by a gas
chromatograph (GC) and a high performance liquid chromatograph
geneous system. When inorganic acid, H PO4 or H SO , was used
3
2
4
as the catalyst, the yield of 2(5H)-furanone was lower than 10% in
both the bi-phasic and homogeneous systems. For heterogeneous
(
HPLC).
Please cite this article in press as: X. Li, et al., Selective oxidation of furfural in a bi-phasic system with homogeneous acid catalyst, Catal.
Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.11.036

G Model
CATTOD-9941; No. of Pages8
ARTICLE IN PRESS
X. Li et al. / Catalysis Today xxx (2015) xxx–xxx
3
8
6
4
2
0
0
0
0
0
the homogeneous system, the produced 2(5H)-furanone, SA and
MA were easier to be further oxidized to small molecular acids as
secondary byproducts.
Homogeneous system
Bi-phasic system
3.2. Effect of formic acid/furfural molar ratio
The effect of formic acid/furfural molar ratio was studied in
the bi-phasic system with 1,2-dichloroethane as the solvent at
33 K. The furfural conversion, selectivity to 2(5H)-furanone and
3
selectivity to SA and MA are shown in Fig. 3. As a function of the
reaction time. The reaction rate of furfural significantly increased
with an increasing amount of formic acid. For example, the reac-
tion time required for complete conversion of furfural was 180 min
at a formic acid/furfural molar ratio of 0.8, and was only 60 min
at a molar ratio of 2.0. With an increasing amount of formic acid,
the concentration of performic acid increased and the mass trans-
fer rate of this active oxidant from the aqueous phase to the
organic phase also increased. As a result, the furfural oxidation was
enhanced in the organic phase. During the reaction, the selectivity
to 2(5H)-furanone first increased and then decreased. The high-
est selectivity to 2(5H)-furanone (62.3%) was obtained at a formic
acid/furfural molar ratio of 0.8 and furfural conversion of 96.9%.
Note that the selectivity to 2(5H)-furanone slightly decreased in
the later stage of the reaction due to deep oxidation. The selectivity
to diacids (SA and MA) decreased gradually during the reaction, as
small molecular byproducts were formed.
H PO
H SO
4
HCOOH
Acid catalysts
ZrTiO4 Amberlyst-15
3
4
2
Fig. 1. Yield of 2(5H)-furanone using homogeneous and heterogeneous acids.
catalysts, Amberlyst-15 with strong acidity was tested considering
its proven capability of converting furfural to C4 diacids [18], and
ZrTiO with moderate acid sites was chosen because of its catalytic
4
activities for aldehyde oxidation [30]. The yield of 2(5H)-furanone
over ZrTiO (7.6% in the bi-phasic system and 10.2% in the homoge-
4
neous system) was similar to that over Amberlyst-15 (5.1% in the
bi-phasic system and 9.0% in the homogeneous system).
Fig. 4 shows the influence of the formic acid/furfural molar
ratio on the final yields of 2(5H)-furanone and diacids. With an
increasing amount of formic acid, the yield of diacids monotonically
increased, while the yield of 2(5H)-furanone first increased and
then decreased. The highest yield of 2(5H)-furanone was obtained
at a formic acid/furfural molar ratio of 0.8. At a low amount of formic
acid, the catalytic oxidation of furfural was very slow and the non-
catalytic polymerization decreased the yield of 2(5H)-furanone.
However, a high amount of formic acid may accelerate side reac-
tions such as deep oxidation of the produced organics.
Fig. 2 compares the catalytic mechanism of formic acid with that
of the other selected catalysts in the bi-phasic system and homo-
geneous system. When formic acid was used as the catalyst in the
bi-phasic system, it reacted with hydrogen peroxide to generate
performic acid, which was the real active oxidant and soluble in the
organic phase. Then performic acid transferred from the aqueous
phase to the organic phase and reacted with furfural to form 2(5H)-
furanone following the proposed reaction pathway A (Scheme 1)
[
31]. Furfural and 2(5H)-furanone mainly existed in the organic
phase, thus their direct contact with hydrogen peroxide and deep
oxidation were effectively reduced. The water soluble byproducts,
such as succinic acid (SA, 12.0%) and maleic acid (MA, 6.3%), were in
the aqueous phase. The strong oxidizing nature and good solubility
in both aqueous and organic phases of performic acid contributed to
the better performance of formic acid [32]. When the reaction was
carried out in the homogeneous system with methanol as the sol-
vent, the yield of 2(5H)-furanone significantly decreased and more
SA was formed. The reason was that the effective concentration
of perfomic acid decreased due to the esterification of formic acid
with methanol occurring under the reaction conditions. In addi-
tion, furfural also reacted with hydrogen peroxide directly in the
presence of H+ following the reaction pathway B, contributing to
an increased yield of SA. When other catalysts were used in the
bi-phasic system, the oxidation reaction occurred in the aqueous
phase and the organic phase served as a reservoir to gradually
release the reactant furfural via phase equilibrium. Meanwhile, the
non-catalytic polymerization of furfural occurred in the organic
phase, and the color turned brown gradually. Consequently, the
direct oxidation with hydrogen peroxide in the aqueous phase and
the severe polymerization of furfural in the organic phase resulted
in a much lower yield of 2(5H)-furanone. When inorganic acids or
solid acids were used in the homogeneous system, the polymeriza-
tion of furfural was slowed down. The concentration of furfural in
the homogeneous system was lower than that in the organic phase
of the bi-phasic system. With the assumption that furfural oxida-
tion reaction is first-order and furfural polymerization reaction is
second-order, the decrease of the concentration of furfural can pro-
mote the ratio of oxidation reaction to polymerization reaction. In
3
.3. Reaction network
To further analyze the oxidation reaction of furfural, a simplified
reaction network is proposed, as shown in Scheme 2. This reaction
network consists of the following four Reactions: (1) oxidation of
furfural to 2(5H)-furanone, (2) further oxidation of 2(5H)-furanone
to diacids including SA and MA, (3) direct conversion of furfural to
diacids, and (4) deep oxidation of the produced diacids to some
small molecular acids like malic acid, lactic acid and propionic
acid. In the bi-phasic system, the concentration of performic acid
in the organic phase is assumed to be constant, while the concen-
tration of hydrogen peroxide in the aqueous phase is considerably
excessive with respect to the dissolved organic compound. In the
homogeneous system, the concentration of hydrogen peroxide is
hypothesized to be unchanged considering that hydrogen perox-
ide is continuously added during the reaction. For simplicity, all
the reactions in Scheme 2 are considered to be irreversible first-
order with respect to the organic reactant. The mass balance of
each component leads to Eqs. (6)–(9).
dCF1
=
=
−k C − k C
(6)
(7)
(8)
(9)
1
F1
3 F1
dt
dCF2
dt
k C − k C
1
F1
2 F2
dCF3
dt
= k C + k C − k C
2
F2
3
F1
4 F3
dCF4
dt
=
k C
4
F3
Please cite this article in press as: X. Li, et al., Selective oxidation of furfural in a bi-phasic system with homogeneous acid catalyst, Catal.
Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.11.036

G Model
CATTOD-9941; No. of Pages8
ARTICLE IN PRESS
4
X. Li et al. / Catalysis Today xxx (2015) xxx–xxx
Fig. 2. Catalytic mechanism of acid catalysts in bi-phasic system and homogeneous system.
Scheme 1. Illustration of the proposed reaction pathways for furfural oxidation. Pathway A: furfural is transformed into 2-formyloxyfuran via Baeyer–Villiger rearrangement
at first. Then 2-formyloxyfuran is hydrolyzed to 2-hydroxyfuran and its tautomers, 2(5H)-furanone and 2(3H)-furanone, which can be further oxidized to maleic acid and
succinic acid, respectively. Pathway B: the furan ring of furfural is opened to form a dienol which is ketonised into diketo aldehyde. The diketo aldehyde is then oxidized to
succinic acid by hydrogen peroxide.
Scheme 2. The reaction network for furfural oxidation.
Please cite this article in press as: X. Li, et al., Selective oxidation of furfural in a bi-phasic system with homogeneous acid catalyst, Catal.
Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.11.036

G Model
CATTOD-9941; No. of Pages8
ARTICLE IN PRESS
X. Li et al. / Catalysis Today xxx (2015) xxx–xxx
5
1
00
2
(5H)-Furanone
Diacids (MA+SA)
80
60
40
20
0
0.0
0.5
1.0
1.5
2.0
Formic acid/ furfural molar ratio
Fig. 4. Effect of formic acid/furfural molar ratio on the final yields of 2(5H)-furanone
and diacids.
Table 1
Reaction rate constants and apparent activation energies based on the proposed
reaction network in the bi-phasic system.
Entry
k (10-2 min-1
)
T (K)
23
Ea (kJ/mol)
3
333
343
1
2
3
k1
k2
k2’
k3
1.36
0.18
0.25
1.23
1.86
0.40
0.48
1.66
3.63
0.61
0.69
2.15
45.1
55.8
46.9
25.7
b
4
aReaction conditions: 1,2-dichloroethane (10 mL), furfural (4.0 g), sodium sulfate
4.0 g), H2O2 (30%, 10 mL), distilled water (1.5 mL), formic acid (88%, 1.5 mL), 3 h.
(
b
ꢀ
ꢀ
The reaction rate constant k2 and apparent activation energy Ea2 were calculated
based on the experimental data of the oxidation of 2(5H)-furanone.
k1
6
5
4
3
k2
k3
k'2
k
0
.0029
0.0030
0.0031
-
1
1
/T (K )
Fig. 5. Arrhenius plots of -ln(k) versus 1/T.
Fig. 3. Furfural oxidation with different formic acid/furfural molar ratios, (a) con-
version of furfural,(b) selectivity to 2(5H)-furanone, and (c) selectivity to diacids.
for the bi-phasic system. In this case, the temperature range of
23–343 K was used for collecting the kinetics data, because the
3
where k , k , k and k4 are the reaction rate constants and CF1
,
loss of solvent could not be neglected at a higher temperature while
the non-catalytic polymerization was severe at a lower tempera-
ture. The results are listed in Table 1. For the same temperature,
the reaction rate constants followed the trend of k > k > k , and
1
2
3
CF2, CF3 and CF4 are the concentrations (mol/mL) of furfural, 2(5H)-
furanone, diacids and secondary byproducts, respectively. The
reaction rate of each component was obtained from derivation of
its concentration-time curve. Then the reaction rate constants were
calculated by solving Eqs. (6)–(9).
1
3
2
k3 was three to five times larger than k , indicating that diacids
2
were mainly formed via the direct conversion of furfural rather than
the oxidation of 2(5H)-furanone. Fig. 5 demonstrates the Arrhe-
nius plots of -ln(k) versus 1/T, showing a good linear relationship.
Using 1,2-dichloroethane as the solvent and formic acid as the
catalyst, the reaction rate constants k , k and k3 were estimated
1
2
Please cite this article in press as: X. Li, et al., Selective oxidation of furfural in a bi-phasic system with homogeneous acid catalyst, Catal.
Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.11.036

G Model
CATTOD-9941; No. of Pages8
ARTICLE IN PRESS
X. Li et al. / Catalysis Today xxx (2015) xxx–xxx
6
The apparent activation energies Ea1, Ea2 and Ea3 were calculated
from the slope of the Arrhenius plots. The activation energy for
furfural oxidation to 2(5H)-furanone (Ea1, 45.1 kJ/mol) was higher
than that for furfural oxidation to diacids (Ea3, 25.7 kJ/mol), there-
fore increasing temperature was beneficial to improve the yield
of 2(5H)-furanone. However, the yield of 2(5H)-furanone slightly
decreased with a further increase in the reaction temperature,
because Ea2 (55.8 kJ/mol) was even higher than Ea1. To verify the
reliability of the proposed reaction network, the oxidation of 2(5H)-
furanone was also carried out under the same conditions. The
100
(a)
2(5H)-Furanone
MA
SA
80
Diacids (MA+SA)
60
4
2
0
0
0
ꢀ
results showed that the reaction rate constant k2 and apparent
ꢀ
activation energy Ea2 directly determined from 2(5H)-furanone
oxidation reaction were very similar to those calculated based on
the proposed reaction network.
3.4. Solvent effect
EAC
DCE
CCl4
Benzene
Toluene
Solvent
The solvent had a strong effect on the furfural oxidation reac-
tion [26], therefore a variety of solvents were investigated using
formic acid as the catalyst at 333 K. When nonpolar solvents such
as cyclohexane, iso-octane and n-heptane were used, a tri-phasic
reaction system consisting of an aqueous phase with little furfural,
an organic phase with a low concentration of furfural and another
organic phase rich in furfural was formed. Formic acid and per-
formic acid existed in all the three phases. In such a tri-phasic
system, the yield of 2(5H)-furanone was lower than 10%, and the
organic phase rich in furfural became brown gradually. The addition
of a nonpolar solvent like cyclohexane, iso-octane and n-heptane
cannot dilute furfural, so the polymerization of furfural phase was
still severe. In addition, the formation of the second organic phase
decreased the concentration of performic acid in the furfural phase,
thus reduced the reaction efficiency of furfural oxidation.
When other solvents such as ethyl acetate, 1,2-dichloroethane,
tetrachloromethane, benzene and toluene were introduced, a bi-
phasic system was formed. In this system, the performic acid was
formed in the aqueous phase and transferred to the organic phase,
where furfural was oxidized to 2(5H)-furanone by performic acid.
The product distributions with different solvents are shown in
Fig. 6(a). When using 1,2-dichloroethane as the solvent, the yield of
1
00
2(5H)-Furanone
(
b)
MA
SA
8
6
4
2
0
0
0
0
0
Diacids (MA+SA)
GBL
THF
Isopropanol
Methanol
Solvent
Fig. 6. Product yields with different solvents, (a) in the bi-phasic system and (b) in
the homogeneous system.
2
(5H)-furanone was 60.3%, and the yields of MA and SA were 12.0%
and 6.3%, respectively. Ethyl acetate with lower toxicity was found
to be an ideal substitute for the toxic solvent 1,2-dichloroethane,
giving 61.5%, 6.7% and 8.5% yields of 2(5H)-furanone, MA and SA,
respectively. With the utilization of another chlorinated solvent,
tetrachloromethane, a 40.2% yield of 2(5H)-furanone was obtained.
In contrast, when benzene and toluene were used as the solvent,
the yield of 2(5H)-furanone was only 8.9% and 12.2%, respectively.
A possible reason for these low yields was that the interaction of
furan ring with benzene ring promoted the polymerization of fur-
fural. With different weak polar solvents, the yields of MA and SA
were 5–15%, which was less dependent on the solvent, because
the direct oxidation of furfural to diacids mainly occurred in the
aqueous phase.
Table 2
Reaction rate constants based on the proposed reaction network in the homoge-
neous system.
k (10-2min-1
)
Entry
Solvent
ε
k1
k2
k3
k4
1
2
3
4
Methanol
Isopropanol
THF
33
19
8
1.26
1.37
1.60
1.00
0.35
0.30
0.39
0.38
2.58
2.37
2.12
3.73
0.89
1.59
1.10
0.92
GBL
39
aReaction conditions: solvent (10 mL), furfural (4.0 g), sodium sulfate (4.0 g), H2O2
30%, 10 mL), distilled water (1.5 mL), formic acid (88%, 1.5 mL), 333 K, 3 h.
(
The solvent effect on furfural oxidation was also studied in the
homogeneous system using methanol, isopropanol, tetrahydrofu-
ran (THF) and GBL as solvents. As shown in Fig. 6(b), SA was the
major product and the product distribution strongly depended on
the solvent. The highest yield of SA was 38.3% with GBL as the sol-
vent, and the yield of SA decreased with the decreasing dielectric
constant of the solvent. The highest yield of MA (17.3%) was also
obtained with GBL, and the yield of MA was 10.0–12.0% with the
other three solvents. When THF was used as the solvent, an 18.2%
yield of 2(5H)-furanone was obtained, while methanol, isopropanol
and GBL gave similar yields of 2(5H)-furanone (13.1%, 11.3% and
the bi-phasic system with 1,2-dichloroethane or ethyl acetate as
the solvent. In the homogeneous system, the products were easier
to be excessively oxidized to small molecular acids. The incom-
plete carbon balance was caused by the formation of some GC and
HPLC-silent products via deep oxidation or polymerization.
The reaction rate constants were calculated for the homo-
geneous system to quantify the solvent effect. The results are
summarized in Table 2. It can be seen that the values of k2 were
similar with different solvents, but k and k strongly depended on
the solvent. The polarity of the solvent, which can be described by
the dielectric constant, is an important factor for the influence of
solvent. As shown in Fig. 7, when the dielectric constant increased
from 8 of THF to 39 of GBL, the reaction rate constant k3 increased
1
3
1
2.8%, respectively). The total yield of 2(5H)-furanone and diacids
(
68.4% in GBL) in the homogeneous system was lower than that in
Please cite this article in press as: X. Li, et al., Selective oxidation of furfural in a bi-phasic system with homogeneous acid catalyst, Catal.
Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.11.036

G Model
CATTOD-9941; No. of Pages8
ARTICLE IN PRESS
X. Li et al. / Catalysis Today xxx (2015) xxx–xxx
7
100
(
a)
8
6
4
2
0
0
0
0
0
3
13
323
333
343
Reaction Temperature (K)
Fig. 7. Correlation of reaction rate constants with dielectric constant of the solvents.
1
00
80
60
−
1
−1
−
from 0.0212 min to 0.0373 min , while the reaction rate con-
(b)
1
−1
stant k1 decreased from 0.0160 min to 0.0100 min . It indicates
that increasing polarity of the solvent promotes the direct conver-
sion of furfural to diacids and decreases the oxidation of furfural
to 2(5H)-furanone. This is consistent with the experimental results
that a higher yield of SA was obtained using a solvent with a larger
dielectric constant.
4
2
0
0
0
3
.5. Effect of reaction temperature
The influence of the reaction temperature on furfural oxidation
was investigated in the bi-phasic system using 1,2-dichloroethane
as the solvent with a formic acid/furfural molar ratio of 0.8.
As shown in Fig. 8(a), the conversion of furfural increased with
increasing reaction temperature. When the reaction temperature
increased from 313 K to 333 K, the yield of 2(5H)-furanone sig-
nificantly increased from 9.0% to 60.3%, whereas a slight decrease
was observed when the reaction temperature further increased to
3
13
323
333
339
Reaction Temperature (K)
Fig. 8. Effect of reaction temperature on furfural oxidation in the bi-phasic system
a) with 1,2-dichloroethane as the solvent, and (b) with ethyl acetate as the solvent.
3
3
43 K. In particular, the yield of 2(5H)-furanone increased from
4.9% at 323 K to 60.3% at 333 K. The non-catalytic polymeriza-
(
tion of furfural was the main reaction under a lower temperature,
until the reaction temperature reached a certain value to pro-
mote the catalytic oxidation of furfural. The slight decrease of the
2
(5H)-furanone yield at 343 K was probably caused by the fast
1
00
decomposition of hydrogen peroxide and reduced stability of per-
formic acid. The yield of diacids only slightly changed with the
reaction temperature. The total yield of 2(5H)-furanone, SA and MA
increased from 29.5% at 313 K to 86.4% at 333 K. For the bi-phasic
system using ethyl acetate as the solvent, the reaction tempera-
ture varied in the range of 313–339 K, considering that the boiling
point of ethyl acetate is lower than that of 1,2-dichloroethane. As
shown in Fig. 8(b), the conversion of furfural and the yield of 2(5H)-
furanone increased with increasing reaction temperature, while the
yield of diacids was nearly unchanged. At the same reaction tem-
perature, the yield of 2(5H)-furanone in the bi-phasic system using
ethyl acetate as the solvent was higher while the yield of diacids
was lower than that using 1,2-dichloroethane as the solvent.
1,2-Dichloroethane
Ethyl acetate
Toluene
8
6
4
2
0
0
0
0
0
3.6. Effect of furfural concentration
0
.20
0.25
0.30
0.35
0.40
To investigate the influence of furfural concentration, the oxida-
Furfural concentration (g (furfural)/ mL (solvent))
tion reaction was carried out with different initial concentrations of
furfural at 333 K in the bi-phasic systems using 1,2-dichloroethane,
ethyl acetate and toluene as solvents. The formic acid/furfural
molar ratio was maintained at 0.8. It can be seen from Fig. 9
that the yield of 2(5H)-furanone was almost unchanged when
Fig. 9. Yield of 2(5H)-furanone with different initial furfural concentrations.
Please cite this article in press as: X. Li, et al., Selective oxidation of furfural in a bi-phasic system with homogeneous acid catalyst, Catal.
Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.11.036

G Model
CATTOD-9941; No. of Pages8
ARTICLE IN PRESS
X. Li et al. / Catalysis Today xxx (2015) xxx–xxx
8
1
,2-dichloroethane or ethyl acetate was used as the solvent. When
References
toluene was used as the solvent, however, the yield of 2(5H)-
furanone decreased gradually with increasing concentration of
furfural, which aggravated the second-order polymerization reac-
tion in the presence of benzene ring.
[
[
1] D.M. Alonso, J.Q. Bond, J.A. Dumesic, Green Chem. 12 (2010) 1493–1513.
2] D. Tilman, R. Socolow, J.A. Foley, J. Hill, E. Larson, L. Lynd, S. Pacala, J. Reilly, T.
Searchinger, C. Somerville, R. Willams, Science 325 (2009) 270–271.
[3] G.W. Huber, S. Iborra, A. Corma, Chem. Rev. 106 (2006) 4044–4098.
[
[
4] A.V. Bridgwater, Biomass Bioenergy 38 (2012) 68–94.
5] L. Hu, G. Zhao, W. Hao, X. Tang, Y. Sun, L. Lin, S. Liu, RSC Adv. 2 (2012)
4
. Conclusions
11184–11206.
[
[
[
[
6] P. Gallezot, Chem. Soc. Rev. 41 (2012) 1538–1558.
7] S. Dutta, S. De, B. Saha, M.I. Alam, Catal. Sci. Technol. 2 (2012) 2025–2036.
8] P. Gallezot, Catal. Today 121 (2007) 76–91.
The selective oxidation of furfural to 2(5H)-furanone was stud-
ied using hydrogen peroxide as the oxidant and acid as the catalyst.
Formic acid showed the best catalytic performance for furfural oxi-
dation to 2(5H)-furanone when the reaction was carried out in
a bi-phasic system using 1,2-dichloroethane or ethyl acetate as
the solvent, giving a 60–62% yield of 2(5H)-furanone. The catalytic
mechanism of formic acid in comparison with other acid catalysts in
both homogeneous and bi-phasic systems was explained in detail.
The good solubility in the organic phase and the strong oxidiz-
ing nature of performic acid, which was the real active oxidant
generated from formic acid and hydrogen peroxide, contributed
to the better performance of formic acid. The formation of a bi-
phasic system using 1,2-dichloroethane or ethyl acetate as the
solvent was beneficial for the selective oxidation of furfural to
9] I. Agirrezabal-Telleria, I. Gandarias, P.L. Arias, Catal. Today 234 (2014) 42–58.
[10] J.P. Lange, E. van der Heide, J. van Buijtenen, R. Price, ChemSusChem 5 (2012)
50–166.
11] A.S. Mamman, J.M. Lee, Y.C. Kim, I.T. Hwang, N.J. Park, Y.K. Hwang, J.S. Chang,
J.S. Hwang, Biofuels Bioprod. Bioref. 2 (2008) 438–454.
[12] K. Yan, G. Wu, T. Lafleur, C. Jarvis, Renew. Sustain. Energy Rev. 38 (2014)
663–676.
13] R. Cao, C. Liu, L. Liu, Org. Prep. Proced. Int. 28 (1996) 215–216.
14] M. Krystof, M. Perez-Sanchez, P. Dominguez de Maria, ChemSusChem 6
2013) 826–830.
[15] J. Lan, Z. Chen, J. Lin, G. Yin, Green Chem. 16 (2014) 4351–4358.
16] N. Alonso-Fagúndez, I. Agirrezabal-Telleria, P.L. Arias, J.L.G. Fierro, R. Mariscal,
M.L. Granados, RSC Adv. 4 (2014) 54960–54972.
17] N. Alonso-Fagundez, M.L. Granados, R. Mariscal, M. Ojeda, ChemSusChem 5
(2012) 1984–1990.
1
[
[
[
(
[
[
[
[
18] H. Choudhary, S. Nishimura, K. Ebitani, Appl. Catal. A: Gen. 458 (2013) 55–62.
19] J.W.E. Glattfeld, G. Leavell, G.E. Spieth, D. Hutton, J. Am. Chem. Soc. 53 (1931)
2
(5H)-furanone, while the oxidation to diacids and polymeriza-
3164–3171.
tion of furfural were the dominant reactions in the homogeneous
system and tri-phasic system, respectively. To further discuss the
solvent effect, a simplified reaction network was established. In the
homogeneous system, the yield of SA increased gradually, while the
yield of 2(5H)-furanone first increased and then decreased with
an increasing dielectric constant of the solvent. Additionally, the
reaction conditions were studied in the bi-phasic system, show-
ing that the optimum reaction temperature was 333 K and formic
acid/furfural molar ratio was 0.8.
[20] C.C. Price, J.M. Judge, Org. Synth. Coll. 5 (1973) 22.
[
[
[
21] J.E. Garst, G.L. Schmir, J. Org. Chem. 39 (1974) 2920–2923.
22] W.Y. Yu, H. Alper, J. Org. Chem. 62 (1997) 5684–5687.
23] E.P. Grunskaya, L.A. Badovskaya, V.V. Poskonin, Y.F. Yakuba, Chem. Heterocycl.
Compd. 34 (1998) 775–780.
[
[
[
24] V.V. Poskonin, Chem. Heterocycl. Compd. 45 (2009) 1177–1183.
25] A. Hashem, E. Kleinpeter, Adv. Heterocycl. Chem. 81 (2001) 107–165.
26] X. Li, X. Lan, T. Wang, Green Chem. (2015), http://dx.doi.org/10.1039/
C5GC02411K.
[27] H. Gräfje, W. Körnig, H.M. Weitz, W. Reiß, G. Steffan, H. Diehl, H. Bosche, K.
Schneider, H. Kieczka, Ullmann’s Encycl. Ind. Chem. (2000).
[
28] T. Haas, B. Jaeger, R. Weber, S.F. Mitchell, C.F. King, Appl. Catal. A: Gen. 280
2005) 83–88.
(
Acknowledgement
[29] H.D. Mansilla, J. Baeza, S. Urzúa, M. Maturana, J. Villasefior, N. Nelson Durfin,
Bioresour. Technol. 66 (1998) 189–193.
[
[
30] B.M. Reddy, A. Khan, Catal. Rev. 47 (2007) 257–296.
31] H. Choudhary, S. Nishimura, K. Ebitani, Chem. Lett. 41 (2012) 409–411.
This work was supported by Program for New Century Excellent
Talents in University of China (NCET-12-0297).
[32] C.H.W. Hirs, Methods Enzymol. 11 (1967) 197–199.
Please cite this article in press as: X. Li, et al., Selective oxidation of furfural in a bi-phasic system with homogeneous acid catalyst, Catal.
Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.11.036