Isomerization of glucose into fructose by environmentally friendly Fe/β zeolite catalysts

Article abstract of DOI:10.1016/j.carres.2017.05.006

Herein, the environmentally friendly Fe/β zeolite for glucose isomerization to fructose in aqueous media was reported for the first time. The effects of various reaction conditions including reaction temperature, reaction time, catalyst dosage, etc. on the isomerization reaction over Fe/β zeolite were studied in detail. Under the optimized conditions, yield of fructose higher than 20% were obtained. Moreover, the Fe/β zeolite catalysts were stable and remained constant catalytic activity after five consecutive runs. The possible active Fe species for isomerization of glucose in Fe/β zeolite is also discussed.

Full text of DOI:10.1016/j.carres.2017.05.006

Accepted Manuscript
Isomerization of glucose into fructose by environmentally friendly Fe/β zeolite
catalysts
Siquan Xu, Lei Zhang, Kehao Xiao, Haian Xia
PII:
S0008-6215(17)30123-4
10.1016/j.carres.2017.05.006
CAR 7380
DOI:
Reference:
To appear in: Carbohydrate Research
Received Date: 16 February 2017
Revised Date: 10 April 2017
Accepted Date: 8 May 2017
Please cite this article as: S. Xu, L. Zhang, K. Xiao, H. Xia, Isomerization of glucose into fructose
by environmentally friendly Fe/β zeolite catalysts, Carbohydrate Research (2017), doi: 10.1016/
j.carres.2017.05.006.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
HO
OH HO
O
HO
OH
Glucose
OH
O
Isomerization
Fe
O
Al
O
OH HO
HO
HO
OH
Fructose

ACCEPTED MANUSCRIPT
Isomerization of glucose into fructose by environmentally friendly Fe/β zeolite
catalysts
Siquan Xu, Lei Zhang, Kehao Xiao, Haian Xia*
Jiangsu Key Lab of Biomass-Based Green Fuels and Chemicals,
College of Chemical Engineering, Nanjing Forestry University, Nanjing 210037,
China
*Corresponding Author: Haian Xia
Tel: +86-25-85427635; Fax: +86-25-85428873;
E-mail: haxia@dicp.ac.cn;

ACCEPTED MANUSCRIPT
Abstract
Herein, the environmentally friendly Fe/β zeolite for glucose isomerization to
fructose in aqueous media was reported for the first time. The effects of various
reaction conditions including reaction temperature, reaction time, catalyst dosage, etc.
on the isomerization reaction over Fe/β zeolite were studied in detail. Under the
optimized conditions, yield of fructose higher than 20% were obtained. Moreover, the
Fe/β zeolite catalysts were stable and remained constant catalytic activity after five
consecutive runs. The possible active Fe species for isomerization of glucose in Fe/β
zeolite is also discussed.
Keywords: Fe/β zeolite; glucose; isomerization; fructose; reuse.

ACCEPTED MANUSCRIPT
1. Introduction
The use of alternative energy sources is attracting increasing attention because of
the current high consumption of fossil resources and concomitant serious
1
environmental problems . Biomass is a promising feedstock for the renewable and
sustainable energy source due to its abundance and eco-friendly nature 2, 3. Among the
numerous biomass, fructose can serve as not only a diet sugar for food applications
but also a starting material for the synthesis of value-added chemical compounds,
4
such as 5-hydroxymethylfurfural (HMF) and levulinic acid . However, the natural
abundance of fructose is low. Glucose is the most abundant monosaccharide and the
cheapest hexose, making it a promising candidate as raw material for the production
5
of fructose . Therefore, isomerization of glucose into fructose is regarded as an
important process for the production of fructose in various relevant industries.
This isomerization of glucose to fructose reaction is slightly endothermic and
reversible, which means that the maximum attainable degree of conversion of glucose
to fructose is governed by the thermodynamic equilibrium between both sugars at the
reaction temperature 4. Typical approaches for isomerization glucose include
base-catalyzed and enzymatic techniques 6, 7. However, monosaccharides are unstable
in alkaline media, and a high amount of byproducts are formed due to the side
8
reactions . Enzymatic isomerization can effectively produce fructose, but it is
sensitive to operating conditions, such as pH, temperature, and requires a long
4
treatment time . Recently, Román-Leshkov et al. provided conclusive evidence via
9
NMR that Lewis acid can catalyze sugar isomerization . Moliner et al. reported that

ACCEPTED MANUSCRIPT
the Lewis acid Sn-β zeolite and Ti-β zeolite show a high catalytic activity in the
isomerization reaction of glucose into fructose, achieving fructose yields of 32% and
14%, respectively 10,11. The Sn-β zeolite is particularly effective for catalyzing the
isomerization of a series of pentose and hexose sugars with activities comparable to
biological processes 10. The open Lewis acidic Sn4+ sites has been proposed as the
11
active site for the isomerization of glucose in Sn-β zeolite . However, tin is a toxic
heavy metal, in violation of the standards of modern green chemistry.
In our previous work, we found that Lewis acidic Fe sites derived from the
hydrolysis of FePO4 exhibited highly catalytic performance in the conversion of
carbohydrates into HMF 12. Moreover, the production of HMF and levulinic acid
catalyzed by Fe-containing zeolite has also been reported recently, with a large
number of Lewis acid sites being created after Fe was incorporated 13, 14. Therefore, in
this work, we use Fe/β zeolite with Lewis acid site to catalyze the glucose
isomerization to fructose in aqueous media. To the best of my knowledge, the
isomerization of glucose into fructose catalyzed by Fe/β zeolite is here for the first
time studied. Compared with known Sn-β zeolite catalyst, Fe/β zeolite catalyst is
more “green”, and the synthesis method is more facile. Therefore, a more than 20% of
the fructose yield can be considered considerable.
2. Results and dicussion
XRD and UV-Vis DRS were applied to characterize the structure and iron
species of the Fe/β zeolite samples (Figures S1 and S2). The results of XRD and

ACCEPTED MANUSCRIPT
UV-Vis DRS indicated that the crystalline nature of zeolites was maintained without
any structural change after incorporation of Fe, and that these samples could contain
some oligomers FeOx species besides isolated Fe species as the major Fe species.
Table 1 Acidity of H-β and Fe/β zeolite catalysts
Brønsted acid sites
(mmol/g) a
0.225
Lewis acid sites
(mmol/g) a
0.178
Total
acidity(mmol/g)
0.403
Catalysts
H-β
Fe/β-0.06
Fe/β-0.10
Fe/β-0.14
Fe/β-0.18
0.168
0.200
0.368
0.161
0.208
0.368
0.151
0.196
0.347
0.191
0.166
0.357
a The amount of acid sites from IR-Py at 150 ℃
The FTIR spectroscopy of pyridine adsorption was carried out to evaluate acid
sites of the catalysts, as shown in Figure 1. The Lewis acid sites can be observed at
band 1450 cm-1 while the band at 1545 cm-1 referred to the Brønsted acid sites. In
addition, a band at 1490 cm-1 was correlated to both Brønsted and Lewis acid sites 15,
16. The amount of acid sites of the catalysts is illustrated in Table 1, from which we
obtained the sequence of the Lewis acid sites ranked as Fe/β-0.10 > Fe/β-0.06 >
Fe/β-0.14 > H-β > Fe/β-0.18. As shown in the Figure 2, the amount of Lewis acid
sites showed a significant correlation with fructose selectivity. Insufficient Lewis acid
sites are detrimental to the catalyzing the isomerization of glucose to fructose. The

ACCEPTED MANUSCRIPT
highest selectivity of 52% was obtained when Fe/β-0.06 zeolite with the 0.200
mmol/g Lewis acid sites was used. In addition, only 25% of the fructose selectivity
was obtained under the same reaction conditions (130 ꢀ, 90 min) in the blank
control without any addition of the catalyst. This result further demonstrates that the
Fe/β-zeolites can effectively promote the formation of fructose from glucose.
a
c
b
1545 1490 1450
1545 1490 1450
350
1545 1490 1450
350
250
ꢀ
ꢀ
350
ꢀ
250
150
ꢀ
ꢀ
250
150
ꢀ
ꢀ
ꢀ
150
ꢀ
25
ꢀ
25
ꢀ
25ꢀ
1550 1500 1450 1400 1550 1500 1450 1400 1550 1500 1450 1400
-1
-1
-1
Wavenumber(cm )
Wavenumber(cm )
Wavenumber(cm )
Figure 1. FTIR spectra of pyridine adsorption on (a) H-β zeolite, (b) Fe/β-0.06 zeolite
and (c) Fe/β-0.18 zeolite at different temperature.
60
55
50
45
40
35
30
25
0.250
0.225
0.200
0.175
0.150
0.125
Amount of Lewis acid sites
Fructose selectivity
H-β
Fe/β−0.06 Fe/β−0.10 Fe/β−0.14 Fe/β−0.18
Catalyst

ACCEPTED MANUSCRIPT
Figure 2. Fructose selectivity distribution with Lewis acid sites for H-β zeolite and
Fe/β zeolites
Reaction condition: Glucose (1.0 g), catalyst (0.10 g), H2O (50mL), 130ꢀ , 90 min
We first investigated the effects of reaction temperature and reaction time on the
isomerization of glucose to fructose, and the results are shown in Table 2. When the
glucose isomerization was carried out at 90 ꢀfor 90 min, the reaction was very slow
and the yield of fructose was only 4% (Table 2, entry 1). When the reaction was
further carried out at 110 ꢀ for 90 min, the fructose yield increased to 7% (Table 2,
entry 2). When it was 130 ꢀ, the 17% fructose yield was obtained (Table 2, entry 3).
4
The isomerization of glucose to fructose is slightly endothermic , so the yield of
fructose can be improved by increasing the reaction temperature. The fructose yield
increased to 22% but the selectivity decreased when the temperature was further
increased to 150 ꢀ (Table 2, entry 4). Unknown products such as dehydration
by-products and humins from fructose produced at higher temperature may account
for the decreased selectivity of fructose 17. Therefore, 130 ꢀwas determined to be the
optimal reaction temperature. In addition, the progress of glucose isomerization to
fructose is generally accompanied by mannose production since the reversibility of
the reaction 9, 10. Here, a mannose yield of 7% was also obtained when the
isomerization of glucose occurred at 150 ꢀ for 90 min , while almost no mannose
was generated below 150 ꢀ (Table S1). This fact seems to indicate that the high
temperature favors the formation of mannose in the isomerization of glucose with
Fe/β zeolite catalysts.

ACCEPTED MANUSCRIPT
Table 2 The effects of reaction conditions on the isomerization of glucose to fructosea
Entery
Reaction
Reaction Conversion/%
time/min
Fructose
temperature/ꢀ
yield/mol%
1
2
3
4
5
6
7
90
90
90
13
18
33
55
29
30
31
4
110
130
150
130
130
130
7
90
17
22
11
12
10
90
60
120
150
a Reaction condition: Glucose (1.0 g), Fe/β-0.06 zeolite (0.10 g), H2O (50 mL)
The impact of the reaction time on the isomerzation of glucose to fructose was
also evaluated at 130 ꢀ, and the results were shown in Figure 3. The yield of fructose
improved from 11% to 17% upon the increase of the reaction time from 60 min to 90
min. However, when the reaction time extended to 120 and 150 min, fructose yield
was decreased to 12 and 10%, respectively. The gradual decrease in fructose yield is
likely due to the further catalytic degradation or dehydration of fructose by Brønsted

ACCEPTED MANUSCRIPT
acid sites derived from zeolite over time at high temperatures, as demonstrated by the
HMF detected in the liquid product (Figure S3 and S4). The results show that both the
reaction temperature and time had significant influences on the isomerization of
glucose to fructose. Therefore, the optimum reaction conditions for glucose
conversion and fructose selectivity were at 130 ꢀfor 90 min.
45
Glucose conversion(%)
40
35
30
25
20
15
10
5
Fructose yield(%)
60
90
120
150
Reaction time(min)
Figure 3. The effects of reaction time on the isomerization of glucose to fructose
Reaction condition: Glucose (1.0 g), 0.06-Fe/β zeolite (0.10 g), H2O (50 mL), 130 ꢀ.
To improve fructose yield, we explored the effects of the Fe/Al mole ratios of the
Fe/β zeolites on the isomerization of glucose to fructose. The results are listed in
Table 3. The 17% fructose yield were obtained when using Fe/β-0.06 zeolite as the
catalyst (Table 3, entry 2). In contrast, when H-β zeolite as the catalyst was added to
the glucose solution, the fructose yield is 5%, because extraframework Al species is

ACCEPTED MANUSCRIPT
able to catalyze the isomerization of glucose (Table 3, entry 1). The incorporation of
Fe element resulted in the formation of more active Lewis acid sites, which is
consistent with the FTIR spectra of pyridine adsorption. However, the yield of
fructose decreased with increasing of Fe/Al mole ratio. The possible explanation is
that oligomer Fe species could be unfavorable for the isomerization of glucose 18. This
means that mononuclear Fe species is most likely the active site for the isomerization
of glucose into fructose.
Table 3 The effects of catalyst dosage and Fe/Al mole ratio on the isomerization of
glucose to fructose a
Entry
Catalyst
Amount of
catalyst/g
Conversion/%
Fructose
yield/mol%
1
2
3
4
5
6
H-β
0.10
0.10
0.10
0.10
0.10
0.05
16
33
25
27
24
26
5
17
9
Fe/β-0.06
Fe/β-0.10
Fe/β-0.14
Fe/β-0.18
Fe/β-0.06
9
7
11

ACCEPTED MANUSCRIPT
7
8
Fe/β-0.06
Fe/β-0.06
/
0.20
0.30
0.00
34
36
4
20
17
1
9 b
a Reaction condition: Glucose (1.0 g), H2O (50 mL), 130 ꢀ, 90 min
b Reaction condition: Glucose (1.0 g), no catalyst, H2O (50 mL), 130 ꢀ, 90 min
We also investigated the effects of catalyst dosage on the glucose conversion and
fructose selectivity and the results are listed in Table 3. A blank test showed that the
isomerization of glucose to fructose could almost not proceed without a catalyst in an
aqueous medium (Table 3, entry 9). The fructose yield initially increased with
increasing catalyst dosage (Table 3, entries 2, 6 and 7). When the catalyst dosage was
increased to 0.20 g, the 20% fructose yield were obtained (Table 3, entry 7). Further
increases in the catalyst dosage had a negligible effect on the conversion of glucose,
but the yield of fructose was obviously decreased, possibly because the amount of
Fe/β zeolite catalyst (0.20 g) was already sufficient for glucose isomerization under
the specified condition, and the excessive Lewis acids caused more side reactions.

ACCEPTED MANUSCRIPT
50
40
30
20
10
0
Conversion
Fructose yield
1
2
3
4
5
Reuse sequence
Figure 4. Fe/β zeolite reusability for isomerization of glucose to fructose.
Reaction condition: Glucose (1.0 g), 0.06-Fe/β zeolite (0.10 g), H2O (50 mL), 130 ꢀ,
90 min
In order to investigate the sustained catalytic activity of the Fe/β catalyst for
glucose isomerization to fructose after several cycles, the Fe/β-0.06 zeolite was
selected as the catalyst for the isomerization of glucose to fructose at 130 ꢀ for 90
min, using 0.1 g catalyst (for the first cycle), and 1 g glucose. The catalyst after each
run was recovered by filtration, and then was washed with water and dried in an oven
at 100 ꢀ. Subsequently, the recovered catalyst was calcined at 400 ꢀ for 4 h to
eliminate the effects of carbonaceous deposits and used in the next cycle. The
reusability of the Fe/β-0.06 zeolite was examined for five cycles. From the results
shown in the Figure 4, the fructose yield remained constant at about 10%-20% in all
five consecutive runs. This clearly demonstrates that the Fe/β zeolite was relative
stable and could be reused at least five runs without significant loss in activity.

ACCEPTED MANUSCRIPT
In this study, Fe/β catalysts have been shown to be effective catalysts for the
isomerization of glucose to fructose. It is well establishing that the iron species mainly
include mono- and bi-nuclear iron species, and few oligomer FeOx species for most
19
Fe-containing zeolite catalysts . UV-Vis DRS result of the Fe/β catalysts revealed
that the main iron species are highly isolated extra-framework Fe species due to low
Fe exchanged content. Moreover, the result of FTIR of pyridine adsorption indicated
that incorporation of Fe species produced new Lewis acid sites in Fe/β zeolites. The
reaction results showed oligomer FeOx species is unfavorable for the isomerization of
glucose into fructose. It has also been reported that, the partially hydrolyzed
framework Sn site (i.e. a (SiO)3Sn(OH) center) which has been proposed as the Lewis
acid site in Sn-beta catalyzed glucose to fructose isomerization reactions 20. Therefore,
we propose that the active sites for the isomerization of glucose to fructose are
partially hydrolyzed extra-framework isolated Fe species on the Fe/β catalysts. It
should be noted that the structure of the active Fe sites for the isomerization of
glucose to fructose is different from that of the open Sn sites in Sn-Beta zeolite due to
the fact that the Fe/β zeolites contains framework Al sites, whereas Sn-Beta zeolite is
a pure silica materials. Thus, we further speculate that the isolated Fe species is bound
to framework or extra-framework Al sites, which is significantly different from the
structure of the open Sn sites in Sn-beta zeolite. The further systematic in-depth
research is being carried out.
3. Conclusion

ACCEPTED MANUSCRIPT
In conclusion, we have developed a Fe/β zeolite which are able to effectively
isomerize glucose to fructose in aqueous media. A fructose yield of over 20% was
achieved from glucose under the optimized reaction conditions. Beside known Sn and
Ti-atoms, we show that the Fe element is an active Lewis acid sited capable of
catalyzing the isomerization of glucose into fructose. Partially hydrolyzed
extra-framework isolated iron species could be active site for the isomerization of
glucose to fructose. The deeper understanding of the nature of active Fe sites and the
mechanism for the isomerization glucose to fructose are under investigation.
4. Experimental
4.1 Catalyst preparation
Fe/β zeolites was prepared by liquid ion-exchange method from H-β zeolite
(Si/Al=33; The Catalyst Company of Nankai University) with FeSO4. Solution
containing a known amount of FeSO4 to H-β zeolite was stirred at room temperature
for 24 h. Then, the mixture was dried overnight 120 . The Fe/β zeolite was formed
after calcination at 500 for 2 h. Samples with different Fe/Al mole ratios (0.06, 0.10,
0.14, 0.18) are denoted as Fe/β-0.06 zeolite, Fe/β-0.10 zeolite, Fe/β-0.14 zeolite and
Fe/β-0.18 zeolite.
4.2 Catalytic runs
The conversion of glucose to fructose was conducted a 100 mL high-pressure
autoclave at the desired temperature with stirring speed of 500 rpm. The reactor was
pressurized at 5 bar of N2 and heated to the desired reaction temperature. Once the

ACCEPTED MANUSCRIPT
reaction temperature was reached, the monitoring of the reaction started. After the
reaction was completed, the mixture was cooled to room temperature. All samples
were filtered before further analysis using HPLC.
4.3 Product analysis
The liquid products were analyzed by high performance liquid chromatography
(HPLC, Agilent 1200) using a column (Agilent Hi-Plex H) with a Refractive Index
detector. The column temperature is 60 ꢀ, and pure water as mobile phase. The
conversion of glucose was determined as follows: conversion (%) = (moles of glucose
reacting) / (moles of glucose staring) × 100. The yields of fructose and mannose was
calculated as: yield (%) = (moles of yield in the products) / (moles of feedstocks put
into the reactor) × 100.
Acknowledgements
This work was financially supported by the National Natural Science Foundation
of China (grant no. 31200445), and the Natural Science Foundation of Jiangsu
province (grant no. BK2012416). This work was also financially supported by the
open foundation of Jiangsu Key Laboratory of Biomass Energy and Materials
(JSBEM201503) and the Priority Academic Program Development of Jiangsu Higher
Education Institutions (PAPD).
References

ACCEPTED MANUSCRIPT
Nigam, PS.; Singh. Prog, A., Energy Combust. 2011, 37, 52-68.
Bridgwater, AV., Biomass Bioenergy. 2012, 38, 68-94.
Hall, DO.; Scrase, JI., Biomass Bioenergy. 1998, 15, 357-367.
Moliner, M.; Roman-Leshkov, Y.; Davis, ME., Proc. Natl. Acad. Sci. U.S. 2010,
107, 6164-6168.
1
2
3
4
5
Nikolla, E.; Román-Leshkov, Y.; Moliner, M.; Davis, ME., ACS Catal. 2011, 4,
408-410.
6
7
8
9
Tewari, YB,. Appl. Biochem. Biotech. 1990, 23, 187-203.
Lima, S.; Dias, AS.; Lin, Z.; Brandão, P., Appl. Catal., A. 2008, 339, 21-27.
Yang, BY.; Montgomery, R., Carbohydr. Res. 1996, 1, 27-45.
Román-Leshkov, Y.; Moliner, M,; Labinger, JA., Davis, ME., Angew. Chem. Int.
Ed. 2010, 49, 8954-8957.
10 Holm, MS.; Pagán-Torres, YJ.; Saravanamurugan, S., Green Chem. 2012, 14,
702-706.
11 Harris, JW.; Cordon, MJ.; Di Iorio, JR., J. Catal. 2016, 335, 141-154.
12 Yang, L.; Yan, XP.; Xu, SQ., RSC Adv. 2015, 5, 19900-19906.
13 Moreno-Recio, M.; Santamaría-González, J.; Maireles-Torres, P., Chem.Eng.
2016, 303, 22-30.
14 Ramli, NAS.; Amin, NAS., Appl. Catal., B., 2015, 163, 487-498.
15
Li, W.; Zheng, J.; Luo, Y. Applied Surface Science., 2016, 382, 302-308.
16 Triwahyono, S.; Yamada, Y.; Hattori, H., Appl. Catal., A.. 2003, 242, 101-109.

ACCEPTED MANUSCRIPT
17 Qi, XH.; Watanabe, M.; Aida, TM.; Smith, RL., Catal. Commun. 2009, 10,
1771-1775.
18 Weingarten, R.; Tompsett, GA.; Conner Jr, WC.; Huber, GW., J. Catal. 2011,
279, 174-182.
19 Berlier, G.; Pourny, M.; Bordiga, S.; Spoto, G.; Zecchina, A.; Lamberti, C., J.
Catal. 2005, 229, 45-54.
20 Bermejo-Deval, R.; Assary, RS.; Nikolla, E., Proc. Natl. Acad. Sci. USA. 2012,
25, 9727-9732.

ACCEPTED MANUSCRIPT
Highlights
ꢀ Under the optimized conditions, yield of fructose higher than 20% were obtained.
ꢀ Fe/β zeolite catalysts maintained a constant catalytic activity after five
consecutive cycles.
ꢀ Isolated Fe species can be acted as active sites for glucose isomerization into
fructose.