Q. Cao et al. / Applied Catalysis A: General 403 (2011) 98–103
99
were shown to efficiently catalyse the fructose dehydration solely
or by using NaHSO ·H O as a catalyst. The partitioning of HMF from
4
2
◦
the highly concentrated TEAC/fructose/NaHSO ·H O melt at 120 C
4
2
and the recycling of TEAC were also demonstrated by continuously
adding THF as an extracting phase.
2
. Experimental
2.1. Materials and experimental procedures
Tetramethyl ammonium chloride (TMAC) (analytical grade),
tetraethyl ammonium chloride (TEAC) (98%), benzyltrimethy-
lammonium chloride (BeTMAC) (98%), benzyltriethylammonium
chloride (BeTEAC) (98%), trimethylphenylammonium chloride
(
PhTMAC) (98%), trimethylamine hydrochloride (TMHC) (chemi-
cally pure), and choline chloride (ChoCl) (analytical grade) were
purchased from Aladdin Reagent (China) Co., Ltd. Tetrabutyl
ammonium chloride (TBAC) (analytical grade), dimethylamine
hydrochloride (DMHC) (99.0%) were purchased from Tianjin
Institute of Fine Chemicals Retrocession. Fructose (≥99%) was pur-
chased from Solarbio Science & Technology Co., Ltd. and used as
a standard. CrCl ·6H O (≥99%), anhydrous ferric chloride (≥97%)
Fig. 1. HMF yield in the fructose/TEAC system as a function of the fructose to TEAC
weight ratio. Conditions: TEAC 1 g, 120 C, 70 min reaction time, fructose conversion
◦
100%.
ous solution with a flow rate of 1.0 mL/min. The amount of fructose
was determined using the external standard.
3
2
and CuCl ·2H O (analytical grade) were purchased from Sinopharm
1
2
2
H NMR spectra were recorded by using a Bruker AVANCE III
Chemical Reagent Co., Ltd. Tetrahydrofuran (THF) (99%) was pur-
NMR spectrometer (600 M Hz) and CDCl3 as the solvent.
chased from Tianjin Fuyu Fine Chemicals Co., Ltd. NaHSO ·H O
4
2
(
≥98.5%) was purchased from Guangdong Guanghua chemical
3. Results and discussion
reagent Co., Ltd. Aluminum chloride anhydrous was purchased
from Tianjin Bodi Chemical Co., Ltd. MoCl3 (99.5%) was purchased
from Alfa Aesar. 5-Hydroxymethylfurfural (≥99%) was obtained
from Wujiang Yingchuang Chemical Reagent Co., Ltd. and used as a
standard. Anhydrous glucose (analytical grade) was obtained from
Guangdong Guanghua Chemical Reagent Co., Ltd.
3.1. The reactivity of fructose in the TEAC/fructose systems
Low melting systems have already been obtained by mixing
ChoCl with fructose [15] or other polar organic components like
carboxylic acids [32,33]. The formation of the TEAC/fructose low
melting mixtures is not unexpected because of the similar struc-
tures between ChoCl and TEAC. However, the conversion of fructose
In a typical experiment, a 25 mL reaction tube was charged with
1
.0 g TEAC, 0.5 g fructose and 0.074 g CrCl ·6H O (10 mol% cor-
3
2
responding to 0.5 g fructose) in a glovebox and then sealed. The
reaction mixture was heated in an oil bath preset at 100 C for
◦
to HMF in the TEAC/fructose mixtures without a catalyst at 120 C
◦
as shown in Fig. 1 is new. While some solvent-catalysts in fructose
dehydration do exist, such as DMSO [22,34], DMA [20], 1-ethyl-3-
methylimidazolium chloride (EmimCl) [24] and BmimCl [28], the
chemical structure of TEAC is very different from those of these
solvents. A novel type of solvent-catalyst has therefore been intro-
duced in this work.
7
0 min and then cooled to room temperature.
The biphasic reaction was carried out in an autoclavevessel lined
with Teflon. 1.0 g TEAC, 1.0 g fructose, 0.038 g NaHSO ·H O (5 mol%
4
2
corresponding to 1.0 g fructose) and 10 mL THF were added into the
◦
vessel. The vessel was sealed and heated at 120 C with magnetic
stirring in an oil bath for 30 min and then cooled to room tempera-
ture to refresh the THF for the next batch. The THF layer recovered
from the reactor vessel after each batch was flashed in vacuum to
remove the volatile components for reuse. The residue was then
analysed using HPLC to determine the content of HMF. For batch
run 1, the residue containing HMF in the reactor vessel was subse-
The reactivity of fructose in TEAC was first investigated by
◦
changing the weight ratio of fructose to TEAC at 120 C. Fig. 1 shows
that over 80% HMF yield could be achieved when the fructose to
TEAC weight ratio was less than 0.52. As the fructose concentra-
tion was further increased, a gradual decrease in the HMF yield
occurred. This observation may be attributed to the fact that higher
fructose concentration would lead to a higher local HMF concentra-
tion, causing undesired side reactions [15]. The side reactions, such
as the formation of insoluble humins or soluble polymers [28], lead
to products in addition to HMF, thus causing a decrease in the HMF
yield. Speculation as to why TEAC, a neutral compound, can catalyse
1
quently analysed using H NMR. Nitrogen and sulphur contained in
the residue was further analysed using an elemental analyser (Vario
EL cube, ELEMENTAL) and an inductively coupled plasma optical
emission spectroscopy (ICP-OES, Vista Axial CCD Simultaneous),
respectively.
−
the fructose dehydration, it has been known that Cl does not play
2.2. Analytical methods
a significant role in fructose dehydration as observed by Zhao et al.
[
24]. It is therefore hypothesised that the catalytic effect of TEAC is
+
HMF was quantified using a Waters HPLC with an UV detector at
due to the N (CH CH ) cation which interacts with the fructose
2
3 4
TM
◦
2
84 nm and a SunFire C18 column (4.8 mm × 150 mm) at 35 C. A
molecule leading to the dehydration of fructose. This mechanism
warrants further studies in the future.
mixture of methanol and water at a volumetric ratio of 1:4 was used
as the carrying solvent. The flow rate was 0.6 mL/min. The amount
of HMF was calculated using the external standard as described in
Section 2 above.
The fructose concentration was analysed using a high perfor-
mance ion-exchange chromatograph with a pulsed amperometric
Temperature is a significant parameter influencing the rates of
reactions [29]. In the conversion of fructose to HMF, increasing
the reaction temperature not only speeds up the reaction of HMF
formation, but also the competing side reactions [24,35,36]. Thus,
an optimal temperature should be determined for a fixed reaction
time. Visual inspection of the reaction systems confirmed that in
TM
detector (Dionex, ICS-3000) and a CarboPac
PA10 column
◦
◦
(
4 mm × 250 mm) at 30 C. The eluent used was 7.5 mM NaOH aque-
the temperature range from 100 to 140 C, homogeneous phase