The effect of Br- and alkali in enhancing the oxidation of furfural to maleic acid with hydrogen peroxide

Article abstract of DOI:10.1016/j.mcat.2021.111488

This study was focused on investigating a novel catalytic system for the selective conversion of furfural to maleic acid (MA) in an aqueous system with hydrogen peroxide as an oxidant. A series of experiments that study the impacts of catalyst species, furfural concentration, catalyst dosage, reaction temperature, residue time, hydrogen peroxide concentration, excess water content, and solvent types on the oxidation of furfural to MA was carried out. The results showed that the co-existence of Br- and alkali sites might play a vital role in furfural oxidation, which could improve the MA yield remarkably. Under 90 °C for 3 h, 72.4 % MA yield was obtained with KOH and KBr as co-catalyst in an aqueous phase. Moreover, a possible reaction pathway of furfural oxidation was proposed on the basis of our reaction system.

Full text of DOI:10.1016/j.mcat.2021.111488

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Molecular Catalysis 504 (2021) 111488
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.journals.elsevier.com/molecular-catalysis
The effect of Br- and alkali in enhancing the oxidation of furfural to maleic
acid with hydrogen peroxide
,
Tao Yanga, Wenzhi Lia b,*, Ajibola T. Ogunbiyia
a Laboratory of Basic Research in Biomass Conversion and Utilization, Department of Thermal Science and Energy Engineering, University of Science and Technology of
China, Hefei, 230026, PR China
b Institute of Energy, Hefei Comprehensive National Science Center, Hefei, 230031, PR China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Furfural
This study was focused on investigating a novel catalytic system for the selective conversion of furfural to maleic
acid (MA) in an aqueous system with hydrogen peroxide as an oxidant. A series of experiments that study the
impacts of catalyst species, furfural concentration, catalyst dosage, reaction temperature, residue time, hydrogen
peroxide concentration, excess water content, and solvent types on the oxidation of furfural to MA was carried
out. The results showed that the co-existence of Br- and alkali sites might play a vital role in furfural oxidation,
which could improve the MA yield remarkably. Under 90 ◦C for 3 h, 72.4 % MA yield was obtained with KOH
and KBr as co-catalyst in an aqueous phase. Moreover, a possible reaction pathway of furfural oxidation was
proposed on the basis of our reaction system.
Maleic acid
Synergistic effect
Br-
Alkali
1. Introduction
relatively low which generally needed longer reaction time to obtain
high MA yield. The homogeneous catalysts show the properties of high
Maleic acid (MA), as one of the essential C4 intermediates in the
chemical industry, is involved in the synthesis of unsaturated polyester
resins, vinyl copolymers, agricultural chemicals, surface coatings, and
pharmaceuticals [1]. Currently, MA is needed industrially about 1 800
000 tons/year, which is produced mainly via the catalytic oxidation of
petroleum-derived feedstock such as butane, butadiene, and benzene [2,
3]. However, with the depletion of petroleum feedstock and increasing
environmental concerns, some researchers have paid much attention to
developing new and renewable biomass-derived chemicals such as lev-
ulinic acid, 5-hydroxymethylfurfural, and furfural, to prepare MA in the
past few years [4–8]. Among these biomass-derived platform chemicals,
furfural has attracted more and more attention because furfural is a bulk
industrial feedstock produced from lignocellulosic biomass at a rate of
ca. 300 kilotons per year [9–13].
catalytic efficiency and high product selectivity, while there still
remained problems such a recovery and recycling of catalysts. For
´
example, in the field of heterogeneous catalysis, M. Lopez Granados
et al. reported that titanium silicalite (TS-1) could selectively oxidize
◦
furfural to MA with an 80 % yield under 50 C, 28 h [15]. They also
explored TS-1 deactivation, and the reasons were attributed to Ti
leaching and heavy products deposition on the catalyst [16]. Then, by
using γ-valerolactone as a solvent, the deactivation of catalysts can be
suppressed [17]. Furthermore, several catalysts such as heteropoly acid,
sulfonated resin, and carbon-based solid acid catalysts have also been
reported for furfural oxidation [18–21]. In the homogeneous catalytic
field, various novel reaction systems have also been investigated, which
showed higher catalytic activities than heterogeneous catalysts. For
example, Araji et al. used betaine hydrochloride as a catalyst to achieve
a more than 90 % total yield of MA and fumaric acid (FA) [22]. Zhang
et al. found that MA could be efficiently produced from furfural with
formic acid as both a catalyst and solvent [23]. The authors proposed
that carboxylic acids could be oxidized to organic peracids. The latter
may play a vital role in the oxidation of furfural to MA [23,24].
Furthermore, inspired by a previous report, Yan et al. developed the
Various alternative and renewable furfural oxidation routes have
been reported for MA production, particularly those that employed
hydrogen peroxide (H2O2) as an oxidant. During these oxidation pro-
cesses, heterogeneous and homogeneous catalytic systems have been
developed for MA production [14]. The heterogeneous catalysts can be
recovered easily from reaction system, however, the reaction efficacy is
* Corresponding author at: Laboratory of Basic Research in Biomass Conversion and Utilization, Department of Thermal Science and Energy Engineering, Uni-
versity of Science and Technology of China, Hefei, 230026, PR China.
E-mail address: liwenzhi@ustc.edu.cn (W. Li).
https://doi.org/10.1016/j.mcat.2021.111488
Received 23 December 2020; Received in revised form 4 February 2021; Accepted 14 February 2021
Available online 3 March 2021
2468-8231/© 2021 Elsevier B.V. All rights reserved.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
T. Yang et al.
Molecular Catalysis 504 (2021) 111488
deep eutectic solvent (DES) as a solvent and catalyst, which consisted of
choline chloride and oxalic acid, to obtain a 95.7 % total yield of MA and
FA [25]. Different from the above reports, Zhu et al. used an alkali
catalyst to investigate the effect of Mg(OH)2 on furfural oxidation, which
showed that the selectivity of 2(5 H)-furanone (FRO) and succinic acid
(SA) could be enhanced by the presence of an alkali [26]. The author
proposed that the OHꢀ can facilitate the attack of the carbon in ꢀ HC = O
by HOOꢀ , which is beneficial for the formation of FRO and SA. As
demonstrated, it has shown that the currently reported catalysts for MA
production have not reflected the obvious similar properties because
various types of catalysts (e.g., TS-1, organic acid, Mg(OH)2) have been
used during furfural oxidation. Furthermore, some studies have reported
that an alkali presence could remarkably improve formic acid yield
during the oxidation of glucose. For example, Jin et al. used KOH as a
catalyst and H2O2 as an oxidant to convert glucose to formic acid, and a
75 % yield was obtained under 250 ◦C [27]. Wang et al. found that LiOH
showed excellent catalytic activity for glucose conversion at room
temperature with a 91.3 % formic acid yield [28]. According to the
above reports, the presence of a base may be beneficial for accelerating
the generation of HOOꢀ , hydroxyl, or superoxide ion species from H2O2,
which could play a vital role in promoting the oxidation of furfural’s
carbonyl group. Thus, various types of alkali were used to investigate its
possible effect in the furfural oxidation process. Furthermore, consid-
ering the diverse products in a single Mg(OH)2 system and the positive
effect of bromide for carboxy groups formation in previous reports, the
bromide salts would also be added into the reaction system [29,30].
In this regard, the study was focused on a novel catalytic system for
MA production from furfural oxidation with H2O2 as an oxidant, cata-
lyzed by alkali and a Br-containing salt. A series of alkali species and
various bromide-containing compounds were used as catalysts to
explore the possible activity sites during the oxidation of furfural to MA.
Furthermore, the effects of reaction time, temperature, catalyst con-
tents, excess of water contents, H2O2/furfural concentration, and sol-
vent types on the yield of MA were investigated systematically.
Moreover, a possible reaction pathway was proposed based on these
results. In conclusion, this study provides a novel and highly selective
route to transform furfural to MA (Scheme 1).
2.2. Experimental procedure
All catalytic oxidation reactions were conducted in a 15 ml thick-
walled glass tube. Typically, 1 mmol furfural, 0.5 mmol KBr, 0.5
mmol mg KOH, 2 ml distilled water (DIW), and 1 ml H2O2 were added
into the glass tube. Then the tube was heated in a preheated oil bath with
magnetic stirring. After the specified reaction time elapsed, the tube was
put in flowing water to end the reaction. Finally, the resultant products
were stored in a 4 ◦C refrigerator for further analysis.
2.3. Product analysis
The quantitative analysis of MA, SA, FRO, and FA were determined
by high-performance liquid chromatography (HPLC, Waters 515 pump),
which has a HPX-87H column (Bio-Rad, USA) and a differential
refractive index detector (Waters 2414). The column and detector
temperature was set at 63 and 50 ◦C, respectively. Furthermore, a 5 mM
H2SO4 solution was chosen as the mobile phase, and the flow rate was
controlled at 0.6 mL/min.
The substrates, such as furfural and 5-HMF, were performed by HPLC
equipped with a Symmetry-C18 column (30 ◦C) and an Ultraviolet De-
tector (Waters 2489) (detector wavelength 280 nm). Methanol and
water (2/3, v/v) were used as mobile phase with a 0.4 mL/min flow rate.
The product yields were calculated by the external standard curves
and the detailed information was added in the support information.
moles of product produced
moles of starting substrate
Yield of product (m ol%) =
× 100%
Product = MA, SA or FRO
Substrate = furfural, 5-HMF or furoic acid
3. Results and discussion
3.1. Influence of catalyst species on furfural oxidation experimental
In previous reports, Zhu et al. used Mg(OH)2 as a catalyst, and they
found that there was a synergy between Mg(OH)2 and H2O2 in
enhancing the yield of FRO and SA [26]. Furthermore, inspired by the
high yield of formic acid from glucose catalyzed by KOH and LiOH [27,
28]. Therefore, KOH and NaOH were preliminarily used as catalysts in
this study to investigate the effect of an alkali on furfural oxidation.
When KOH and NaOH were added into the reaction system, three main
products (MA, SA, FRO) were detected, which indicated that furfural
oxidation with H2O2 as an oxidant and a base as a catalyst showed a lack
of selectivity. Furthermore, Yan et al. have reported a new chemical
route to produce terephthalic acid from corn-stover-derived lignin oil
with Co-Mn-Br system and we wonder if the presence of KBr would be
beneficial for the formation of carboxy groups on the furfural oxidation.
Interestingly, when KBr was added into the reaction system, only MA
was detected without the formation of SA and FRO, implying that the
presence of KBr could inhibit the formation of SA and PRO. Considering
the alkali role in promoting H2O2 activation, we then added KBr and
KOH simultaneously into the reaction system. It was surprising that the
MA yield could be improved to 64.3 % without SA and FRO. Thus, we
guessed that halogen might play a vital role in enhancing MA selectivity.
However, when KCl replaced KBr, 41.2 % MA, 24.2 % SA, and 10.4 %
FRO yield were obtained, suggesting that the Clꢀ could not keep MA
selectivity. Thus, it was proposed that the inhibition of SA and FRO
formation may be ascribed to the effect of Brꢀ . Therefore, NaBr, NH4Br,
LiBr, and CaBr2 were used as catalysts to verify our hypothesis, which
gave 29.9 %, 27.7 %, 28.2 %, and 27.7 % MA yield, respectively.
Similarly, SA and FRO could not be produced with these Br-containing
catalysts. Furthermore, the MA yield also maintained similar results of
about 30 %.
2. Experimental
2.1. Materials
Furfural (99 %), 5-hydroxymethylfurfural (5-HMF, 98 %), 2(5 H)-
furanone (FRO, 98 %), furoic acid (98 %), maleic acid (99 %), suc-
cinic acid (99 %), fumaric acid (98 %), lithium bromide (LiBr), and
butyrrolactone (GBL) were purchased from Aladdin. (Shanghai, China).
Potassium bromide (KBr), ammonium bromide (NH4Br), sodium bro-
mide (NaBr), potassium hydroxide (KOH), sodium hydroxide (NaOH),
ammonia (NH3⋅H2O), potassium chloride (KCl), potassium nitrate
(KNO3), formic acid (FA), 1,4-dioxane, γ-valerolactone (GVL), hydrogen
peroxide (H2O2, 30 %), sulphuric acid (98 %, H2SO4), and dimethyl
sulfoxide (DMSO) were supplied by Sinopharm Chemical Co., Ltd.
(Shanghai, China).
Scheme 1. Conversion of furfural to MA with Brꢀ /OHꢀ in the presence
of H2O2.
Furthermore, to investigate the role of alkali, KOH was added into
2

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
T. Yang et al.
Molecular Catalysis 504 (2021) 111488
the above four types of Br-containing catalysts, and the MA yield
improved to 64.6 %, 61.6 %, 63.1 %, and 54.3 % (Table 1, entries
11–14), respectively. Therefore, the high MA yield may be attributed to
the role of Brꢀ and OHꢀ . Moreover, to exclude the effect of K+ on the
catalytic process, NH4Br-NH4OH (Table S1, entry 1) and LiBr-LiOH
(Table S1, entry 2) were also used as catalysts, and 61.8 % and 59.8 %
MA yields, respectively, were achieved in these conditions, which
proved that the K+ is not the crucial activity species for enhancing MA
selectivity. To determine if there was any synergistic effect between
other types of alkali and KBr, we also tested some common bases such as
NaHCO3, K2CO3, Na2CO3 (Table S1, entries 4–6). There were also no SA
and FRO with the presence of these catalysts, but the yields of MA
changed slightly with varying types of base precursors. Obviously, the
yield of MA improved up to 63.9 % for NaHCO3, 62.6 % for K2CO3, and
67.8 % for Na2CO3 added, while only 27.5 % of MA yield was obtained
with only KBr as a catalyst, suggesting that the introduction of a strong
or weak alkali may have a synergistic effect with Brꢀ . In conclusion, the
results demonstrated that the presence of Brꢀ and alkali sites played a
vital role in enhancing MA yield from furfural.
Table 2
Furfural oxidation using different furfural concentrations.
Entry
Furfural (mmol)
H2O2 (mL)
MA yield (%)
1
2
3
4
5
6
7
8
0.5
1
1
1
1
1
2
3
2
3
66.3
67.4
51.5
37.9
61.4
63.0
58.8
64.8
2
3
2
2
3
3
Reaction conditions: KBr (0.5 mmol), KOH (0.5 mmol), 2 ml DIW, 100 ◦C, 3 h.
content further increased to 0.75 and 1.0 mmol (with a constant KOH
contents of 0.75 mmol), the MA yield slightly decreased to 67.8 % and
70.0 %, respectively. Thus, excessive amounts of catalyst would result in
by-reactions such as condensation, producing by-products [17,31]. So,
an optimum catalyst ratio of 0.5/0.75 (mmol/mmol) was used in our
subsequent studies. Furthermore, interestingly, when the KBr/KOH was
0.05 and 0.1 mmol (Table 3, entry 1–2), 14.7 % and 7.1 % FRO yields
were obtained besides the presence of MA, respectively. However, when
FRO was used as a substrate, there was no MA formation, as shown in
Table 4 (Entries 1, 2), implying that the formation of FRO and MA are
parallel reactions in our reaction system. Under the above conditions,
which suggested that either there were no sufficient activity sites for MA
formation or there was a lack of enough Br- for the inhibition of FRO
formation. Thus, FRO can be completely inhibited only when the
amounts of KBr and KOH reach a certain level.
3.2. Furfural oxidation under varying substrate concentrations and
catalyst contents
As we all know, high furfural concentrations are vital to enhance the
industrial economic benefits of MA production. Thus, the effect of the
initial furfural concentration (0.5ꢀ 3 mmol) on MA production in re-
actions catalyzed by KBr-KOH was investigated, and the results are listed
in Table 2. The MA yields are 66.3 %, 67.4 %, 51.5 %, and 37.9 % with
0.5, 1, 2, and 3 mmol furfural, respectively. According to the above
results, increasing furfural contents from 0.5 to 3 mmol harmed the MA
yield. We guessed that the low MA yield might be caused by the lack of
H2O2 oxidant. Thus, an additional H2O2 was added to the reaction sys-
tem. Under the conditions of 2 mmol furfural substrate, the MA yield
increased to 61.4 % (Table 2, entry 5) and 63.0 % (Table 2, entry 6) with
the increase of H2O2 to 2 ml and 3 mL, respectively. Similarly, when the
3 mmol furfural was used as a substrate, the MA yield increased to 58.8
% (Table 2, entry 7) and 64.8 % (Table 2, entry 8) with the increase of
H2O2 to 2 ml and 3 mL. Therefore, with the further addition of furfural,
the MA yield mainly depended on the amount of excess H2O2 added.
Besides furfural concentration, the catalyst dosage also has economic
importance in the industrial production of MA, therefore, the effect of
catalyst contents on catalytic properties was studied in detail. As shown
in Fig. S1, the yield of MA increased from 53.0% to 72.2% as the KOH
contents increased from 0.25 to 1.0 mmol with a constant KBr contents
of 0.5 mmol, and the same phenomenon could be also observed with the
constant KBr of 0.75 or 1.0 mmol. The best MA yield can be achieved
with 0.5 mmol KBr and 0.75 mmol KOH. Furthermore, as the KBr
3.3. The effects of reaction temperature and retention time on furfural
oxidation to maleic acid
The effects of reaction temperature and residue time on the con-
version of furfural to MA in a water-H2O2 system were investigated by
conducting the experiments under 70–110 ◦C within a time range of 2–5
h. As shown in Fig. 1, both reaction temperature and time affect the MA
yield. The catalytic reaction at 90, 100 and 110 ◦C gave high MA yields
within only 2 h, quicker than that of 70 and 80 ◦C. The possible reason is
that a higher reaction temperature could increase the transformation
rate of H2O2 (evident by the bubbles formed at the end of the reaction),
thereby promoting the oxidation of furfural. Under low reaction tem-
perature (70, 80 ◦C), a longer reaction time of up to 5 h would favor MA
formation. The optimal reaction temperature is 90 ◦C, and the highest
MA yield reached was 72.4 % at 3 h, indicating that the KBr-KOH was
efficient for such an oxidation reaction. Although higher reaction tem-
peratures can reduce the reaction time, yield loss reactions might occur
since the MA yield decreased when the temperature increased to 100
and 110 ◦C.
Table 1
3.4. The effect of H2O2/furfural ratios and water contents on furfural
oxidation
Furfural oxidation using various catalysts.
Entry
catalyst
MA yield (%)
SA yield (%)
FRO yield (%)
As described earlier, H2O2 plays a vital role in the conversion of
furfural to MA. Thus, the effect of H2O2 concentration on MA formation
was investigated by varying the H2O2/furfural mole ratio from 1.7 to
1
blank
24.5
27.6
29.8
27.5
64.3
41.2
29.9
27.7
28.2
27.7
64.6
61.6
63.1
54.3
24.0
20.3
2
KOH
22.9
14.6
3
NaOH
21.3
15.9
4
KBr
–
–
–
–
5
KOH/KBr
KOH/KCl
NaBr
Table 3
6
24.2
10.4
7
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Effect of catalyst concentrations on the furfural oxidation.
8
NH4Br
Entry
KBr/KOH (mmol-mmol)
MA yield (%)
FRO yield (%)
9
LiBr
10
11
12
13
14
CaBr2
1
2
3
4
5
6
0.05ꢀ 0.05
0.1ꢀ 0.1
38.1
51.9
58.7
67.4
67.8
67.1
14.7
KOH/NaBr
KOH/NH4Br
KOH/LiBr
KOH/CaBr2
7.1
–
0.25ꢀ 0.25
0.5ꢀ 0.5
–
0.75ꢀ 0.75
–
1.0ꢀ 1.0
–
Reaction conditions: catalyst 0.5 mmol (2–4, 7–10) or 0.5/0.5, mmol/mmol (5,
6, 11–14), furfural (1 mmol), DIW (2 mL), H2O2 (1 mL), 100 ◦C, 3 h.
Reaction conditions: catalyst (KBr-KOH), 2 mL DIW, H2O2 (1 mL), 100 ◦C, 3 h.
3

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
T. Yang et al.
Molecular Catalysis 504 (2021) 111488
mole ratio of H2O2 to furfural was 6.8. Therefore, this optimum ratio was
used for further investigation.
Table 4
Oxidation of various substituted furan compounds.
The amount of excess water is also a vital parameter influencing the
reaction [25], and the results of this investigation are shown in Fig. 2B.
The MA yield increased with the increase of excess water up to 2 mL.
However, as the amount of excess water increased further to 3 mL, the
yield of MA has a slight decrease, suggesting that proper water content is
beneficial for MA production, as excess water may affect H2O2 and the
catalyst properties. The similar results have been explored in previous
reports, for example, the MA yield decreased remarkably with additional
water adding into formic acid or deep eutectic solvent catalytic system
[23,25]. Moreover, when no excess water was added to the reaction
system, the MA yield only reached to 62.8 %. In this regard, water is
essential, and suitable water content would also play a crucial role in the
whole process of furfural oxidation.
Entry
Substrate
Catalyst
MA yield (%)
SA yield (%)
FRO yield
(%)
1
2
FRO
FRO
KBr
–
–
–
–
73.3a
KBr-
KOH
KBr
–
3
4
Furoic acid
Furoic acid
28.9
87.1
–
–
–
–
KBr-
KOH
Reaction conditions: substrate (1 mmol), DIW (2 mL), H2O2 (1 mL), 100 ◦C, 3 h.
a
the remained FRO.
3.5. Effect of solvent types on the production of maleic acid
Solvent types may also play an important role in furfural oxidation to
MA because it can influence the dispersion of furfural and MA’s distri-
bution and isolation. The previous report has demonstrated that solvents
such as GVL can improve the di-acids yield with TS-1 as a catalyst. Thus,
to explore the effect of solvent on furfural oxidation reaction, various
common solvents such as water, GVL, GBL, 1, 4-dioxane, and DMSO
Fig. 1. Effect of the reaction temperature and time on the oxidation of furfural
in a KBr-KOH system. Reaction conditions: Furfural (1 mmol), DIW (2 mL),
KBr/KOH (0.5/0.75 mmol/mol), H2O2 (1 mL).
10.2, and the results are shown in Fig. 2A. The increase in the mole ratio
of H2O2 to furfural resulted in the variation of the maleic acid yield.
When the mole ratio increased from 1.7 to 5.1, the MA yield increased
from 31.0% to 71.6%, and a slight increase to 72.4 % with the mole ratio
up to 6.8. While the MA yield are 71.7 % and 72.6 % respectively with
the mole ratio reaching to 8.5 and 10.1. According to previous reports
[15], the stoichiometric amount of H2O2 needed for transforming
furfural to MA is 3; however, a higher H2O2/furfural mole ratio was
required to supplement the underside consumption reactions of H2O2. It
should be mentioned here that the use of ratios higher than 6.8 did not
give obvious further increase in MA yield, suggesting that the optimum
Fig. 3. Effect of solvent types on furfural oxidation.
Reaction conditions: furfural (1 mmol), solvent (2 mL), KBr/KOH (0.5 /0.75
mmol), H2O2 (1 mL), 100 ◦C, 3 h.
Fig. 2. Effect of H2O2/furfural ratio (A) and excess water on furfural oxidation (B).
Reaction conditions: (A) 1 mmol furfural, 0.5 mmol KBr, 0.75 mmol KOH, 2 mL DIW, 0.25–1.5 mL H2O2, 100 ◦C, 3 h; (B) Furfural (1 mmol), excess water (0–3.5 mL),
KBr/KOH (0.5 mmol/0.75 mmol), H2O2 (1 mL), 100 ◦C, 3 h.
4