18
Y. Shen et al. / Catalysis Communications 50 (2014) 17–20
2. Experiment
2.1. Materials and instruments
The chemicals used in this study were analytic grade and without
further purification. InCl3·4H2O was supplied by Tianjin Jinke Fine
Chemical Research Institute (Tianjin, China). Glucose, fructose, tetrahy-
drofuran (THF) and NaCl were purchased from Beijing Chemicals Co.
Ltd. (Beijing, China). The standard 5-HMF (99%) was purchased from
Acros Organics. LA (98%) was obtained from Alfa Aesar. Nanopure
water was used for all reactions. The dehydration experiments of mono-
saccharides were performed in a pressure-tight stainless autoclave
(50 mL) with a magnetic stirrer and a temperature controller ( 1 °C).
2.2. General procedure for the dehydration of monosaccharides
In a typical reaction, fructose (1.0 g, 5.6 mmol) and InCl3 (30 mg,
0.14 mmol) in 20 mL water were charged into the reactor (50 mL).
The reactor was sealed and stirred at 180 °C. After the certain reaction
time, the reactor was removed from the heat and quenched in an ice
cool water bath. Then, the liquid samples were collected, filtered, and
diluted to 20 mL for analysis. The solid products were collected and
dried in a vacuum oven overnight to constant weight for analysis.
Fig. 1. Fructose dehydration with reaction time catalyzed by InCl3. Conditions: fructose
(1.0 g), InCl3 (0.03 g, 2.5 mol%) in water (20 mL) at 180 °C (the reaction time began to
be calculated when the system reaches the reaction temperature).
Product yields ðmol%Þ :
Y ¼ ðMoles ð5‐HMF or levulinic acidÞ=Moles ðstarting amount of fructoseÞÞ
ꢀ100%
2.3. Recycling experiments
3. Results and discussion
THF was used as an extractant with 2:1 ratio of organic/aqueous
when biphasic systems were studied. The fructose dehydration in the
biphasic system with THF and 25 wt.% of NaCl was similar to that in
the single-phase reaction. The aqueous phase was recycled and the cat-
alyst InCl3 was reused.
During the production of 5-HMF as secondary product from glucose
[24], the decomposition process of glucose is: glucose ⇌ fructose →
HMF → LA + FA. Normally, fructose will be converted to downstream
chemicals more completely and quickly than glucose. Comparatively,
the dehydration of glucose and fructose was separately performed in
InCl3–H2O system under the same conditions (Table 1, entries 1–4). It
can be observed that the conversion of fructose were better than that
of glucose. Nearly 76% of the fructose conversion was achieved at
180 °C in 10 min. At the same time, the yield of 5-HMF obtained from
fructose went over those originated from glucose, which could be attrib-
uted to the stable ring structure of glucose. This stability led to a lower
degree of the enolization that determined 5-HMF formation based on
the reaction mechanism for aldose (glucose) to ketose (fructose).
Subsequently, a systematic study on the dehydration of fructose was
conducted and the results are shown in Fig. 1. It illustrated that the yield
of 5-HMF ascended to nearly 80% along with the reaction time up to
15 min, then started to descend. This indicated that the produced
5-HMF was subsequently converted into LA and FA by rehydration as
shown in Fig. 2, which was confirmed by the gradually increased LA
and FA in Fig. 1. These results were close to the fructose conversions
in ionic liquid as reported by Zhao et al. [25]. Moreover, fructose not
only dehydrated to 5-HMF in a parallel path but also isomerized to glu-
cose. This was also supported by the data in Table 1. Clearly, nearly
24.4 mg of starting fructose and 20.3 mg of glucose isomerized from
fructose were left after 5 min (Table 1, entry 3), and just 32.0 mg of
transformed glucose was detected and fructose was totally gone after
10 min (entry 4). Above all, the dehydration of fructose to produce
2.4. Analytical methods
The quantitative analysis of products was performed on high-
performance liquid chromatography (HPLC; Agilent 1200, USA) with a
Bio-Rad Aminex HPX-87H analytical column and a refractive index detec-
tor. The eluent was 5 mM sulfuric acid solution with a volumetric flow
rate of 0.5 mL min−1. In addition, the amount of products was also ana-
lyzed by HPLC fitted with a Lichrospher C18 column and an ultraviolet
detector at 284 nm. The mobile phase was methanol/water = 35:65
(v/v) at a flow rate of 0.8 mL min−1. The elemental composition of the
solid products was determined using a Vario EL Elemental Analyzer
(Elementar, Germany).
2.5. Calculation of conversion and yield
The conversion of fructose and the yields of products were evaluated
on a carbon basis respectively as shown below:
Fructose conversion ðmol%Þ :
C ¼ ð1−Moles ðfructoseÞ=Moles ðstarting amount of fructoseÞÞ ꢀ 100%
Table 1
Dehydration of monosaccharides to HMF and LAa.
Entry
Substrate
Tb (min)
Glu (mg)
Fru (mg)
Conv. (%)
Yield of HMF (%)
Yield of LA (%)
1
2
3
4
Glucose
Glucose
Fructose
Fructose
5
10
5
181.2
86.1
20.3
32.0
45.4
60.4
24.4
0
81.9
91.4
97.6
100
52.0
59.8
72.0
76.0
21.5
27.0
16.0
17.0
10
a
Reaction conditions: 1.0 g substrate and 0.03 g InCl3 (3 wt.%) in 20 mL of water at 180 °C.
The reaction time began to be calculated when the system reaches the reaction temperature.
b