Surface amino-functionalization of Sn-Beta zeolite catalyst for lactic acid production from glucose

Article abstract of DOI:10.1039/c9ra01264h

The catalytic conversion of glucose into lactic acid (LA) provides an alternative approach for LA production to bio-fermentation. Nevertheless, during the process, the dehydration of glucose and fructose into 5-hydroxymethylfurfural catalyzed by Br?nsted acid exists as the main side reaction that leads to the decrease in LA yield. In order to promote the yield and selectivity of LA, a series of acid-base bifunctional Sn-Beta-NH₂ catalysts were prepared by post-grafting aminopropyl groups with the surface silanol groups of Sn-Beta zeolite. The catalysts were systematically characterized by X-ray diffraction, N₂ adsorption-desorption, elemental analysis, X-ray photoelectron spectroscopy, Fourier transform infrared spectra analysis following pyridine adsorption, and CO₂ temperature-programmed desorption. The as-prepared Sn-Beta-NH₂ catalysts exhibiting both Lewis acid and moderate base sites facilitated the conversion of glucose to LA in competition with undesirable side reactions. In addition, effects of reaction parameters including reaction temperature and time, catalyst dosage, and glucose concentration were investigated. A high LA yield up to 56% was achieved under optimized hydrothermal conditions (190 °C, 2 h), along with a complete conversion of glucose and a 5-hydroxymethylfurfural yield of 7%. The result indicated an alternative modification method of Sn-Beta zeolite for a more favorable LA yield.

Full text of DOI:10.1039/c9ra01264h

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Surface amino-functionalization of Sn-Beta zeolite
catalyst for lactic acid production from glucose†
Cite this: RSC Adv., 2019, 9, 18989
ab
a
b
a
b
b
Zheng Shen,
and Yalei Zhang
Ling Kong, Wei Zhang, Minyan Gu, Meng Xia, Xuefei Zhou
*
ab
The catalytic conversion of glucose into lactic acid (LA) provides an alternative approach for LA production
to bio-fermentation. Nevertheless, during the process, the dehydration of glucose and fructose into 5-
hydroxymethylfurfural catalyzed by Brønsted acid exists as the main side reaction that leads to the
decrease in LA yield. In order to promote the yield and selectivity of LA, a series of acid–base
bifunctional Sn-Beta-NH
silanol groups of Sn-Beta zeolite. The catalysts were systematically characterized by X-ray diffraction, N
adsorption–desorption, elemental analysis, X-ray photoelectron spectroscopy, Fourier transform infrared
spectra analysis following pyridine adsorption, and CO temperature-programmed desorption. The as-
prepared Sn-Beta-NH catalysts exhibiting both Lewis acid and moderate base sites facilitated the
2
catalysts were prepared by post-grafting aminopropyl groups with the surface
2
2
2
conversion of glucose to LA in competition with undesirable side reactions. In addition, effects of
reaction parameters including reaction temperature and time, catalyst dosage, and glucose
concentration were investigated. A high LA yield up to 56% was achieved under optimized hydrothermal
Received 19th February 2019
Accepted 5th May 2019
ꢀ
conditions (190 C, 2 h), along with a complete conversion of glucose and a 5-hydroxymethylfurfural
DOI: 10.1039/c9ra01264h
yield of 7%. The result indicated an alternative modification method of Sn-Beta zeolite for a more
rsc.li/rsc-advances
favorable LA yield.
its subsequent separation process are all complex and of low
efficiency. Additionally, the inevitable use of lime as the
Introduction
Biomass, as the most important and perhaps the only sustain- neutralization agent may lead to the production of calcium
able and clean carbon source, is the most abundant and wide- sulfate waste. Compared to bio-fermentation, chemical catalytic
spread renewable resource. It is being intensively studied reactions provide a more acceptable approach for LA produc-
worldwide as raw materials for the production of high value- tion from biomass instead of grain because of its potential for
1
4,15
added chemicals in order to solve the problems of energy large-scale production and higher efficiency.
1–9
crisis and environmental pollution.
Lactic acid (LA) is
Up to date, great efforts have been devoted to synthesizing
a versatile chemical in industry and is regarded as one of the multiple catalysts for the conversion of carbohydrates into LA,
most promising platform molecules because of its wide appli- including homogenous and heterogeneous catalysts. Homoge-
2
+
16
cations in the production of alkyl lactates, propylene glycol, nous catalysts, such as metal salts (Pb ), alkaline-earth metal
10–12
biodegradable plastics, amino acids, etc.
factured either by carbohydrate fermentation or by chemical Er(OTf)
LA can be manu- hydroxides (Ca(OH)
2
and lanthanide catalysts (ErCl
3
and
17–19
)),
always exhibit outstanding catalytic performance
3
synthesis approaches. However, the commercial biotechnolog- in the process of biomass conversion to produce LA. However,
ical routes for LA production are mainly based on the use of the separation and recovery of these catalysts may hinder their
grains including rice, wheat and corn through anaerobic practical application. Another widely studied route is the
10,13
fermentation in the presence of Lactobacillus.
Lactobacillus relies on the nutrition including carbohydrates, (Nb
amino acids, nucleotides and minerals, as well as a pH value of (AlW and Er/K10)
The growth of synthesis of heterogeneous catalysts, such as metal oxides
20–22
2
O
5
, CuO and Al–Zr mixed oxides),
solid acid catalysts
23,24
14,25
and zeolite catalysts (Sn-Beta).
5.5–5.8. The fermentation process of lactic acid production and
Among numerous heterogeneous catalysts, Sn-Beta zeolite
has shown benign catalytic performance because of its stability
25–27
a
and various Lewis acid sites generated from Sn.
Therefore,
National Engineering Research Center of Protected Agriculture, Tongji University,
much attention has been paid to the modication of Sn-based
State Key Laboratory of Pollution Control and Resource Reuse, Tongji University, zeolites. In order to accelerate the diffusion of glucose and
Shanghai, 200092, China. E-mail: zhangyalei@tongji.edu.cn
b
Shanghai, 200092, China
fructose molecules, Zhang et al. synthesized a hierarchical Sn-
Beta zeolite catalyst and a 72% alkyl lactate yield was
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/c9ra01264h
1
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 18989–18995 | 18989

View Article Online
RSC Advances
Paper
28
obtained from sucrose. Wan et al. found that Sn-Beta sup- hydroxymethylfurfural (HMF) (97%), levulinic acid (LeA) (98%)
ported gold catalyst could readily convert acetol to LA with were obtained from Alfa Aesar. 3-Aminopropyltrimethoxysilane
29
a yield up to 73% under hydrothermal conditions.
(APTMS) (98%) was obtained from Wako. Formic acid (FA)
During the reaction process of glucose conversion, there (99%) was obtained from J&K China Chemical Co., Ltd. Phos-
exist some side reactions that may lead to the decrease of LA phoric acid (HPLC) was obtained from Sigma-Aldrich. Nitric
yield, including the dehydration of glucose and fructose to 5- acid (GR), ethanol (AR), D-(+)-glucose (AR) were obtained from
hydroxymethylfurfural (HMF) and the rehydration of HMF to Sinopharm Chemical Reagent Co., Ltd.
formic acid and levulinic acid. Moreover, the generated LA may
inevitably act as a Brønsted acid to catalyze these side reactions.
In order to suppress the side reactions, researchers attempted
to add base into the reaction solution, followed by the recovery
Catalysts synthesis
Commercial Beta zeolite was dealuminated by stirring in HNO
solution (13 M) at 80 C for 20 h with a HNO
3
ꢀ
30
3
(mL)/Beta zeolite
of lactic acid with the aid of acid neutralization. But this
approach brings additional cost and may lead to corrosion to
the reactor. To avoid the use of liquid base, Dong et al. rstly
developed a Sn–Zn-Beta catalyst that contained both Lewis acid
(g) ratio of 20. Aer ltrating, washing with a large amount of
deionized water until the supernatant was neutral, drying at
ꢀ
1
50 C overnight, the dealuminated Beta zeolite was obtained
and denoted as deAl-Beta. Solid-state ion exchange (SSIE) was
then performed by grinding 0.2 g of tin(II) acetate with 1.0 g of
dealuminated zeolite for 30 min. Finally, the sample was
14
and base sites and a 48% LA yield was obtained. It was a novel
thought to modify the Sn-Beta zeolite with another metal to
form heterogeneous catalysts with both acidic and basic cata-
lytic sites. Apart from the introduction of metal atoms, the
plenty silanol groups on the surface of Beta zeolite are also
favorable for modication by organic functional groups. Post-
graing method is a common method for organic functionali-
zation. By the reaction of silanol groups and functionalization
reagents to form corresponding covalent bond, the surface
hydrophilicity/hydrophobicity or acidity/basicity of Beta zeolite
could be adjusted. In addition, compared to metal modica-
tion, surface modication with covalently bound organic func-
tional groups could also promote the interaction and collision
opportunity with organic substrates, leading to a further
increase in reaction efficiency.
ꢀ
calcined in an air ow at 550 C for 6 h with a ramp of
ꢀ
ꢁ1
38
5
C min and was denoted as Sn-Beta.
As to the synthesis of surface amino-functionalizated Sn-
Beta zeolite (Sn-Beta-NH2(v), v represents the volume of
APTMS used (mL)), in a typical run, 0.5 g of Sn-Beta zeolite and
a certain volume of APTMS were dissolved into 250 mL of
ethanol in a 500 mL round-bottomed ask equipped with
a reux condenser. Aer stirring at 80 C for 6 h, the white
precipitates were collected by centrifugation and followed by
ꢀ
washing with a large amount of ethanol. Then, the sample was
ꢀ
dried at 80 C overnight to obtain Sn-Beta-NH
2
catalyst.
Catalysts characterizations
Here, we developed a new method to modify Sn-Beta zeolite
with amino groups, in order to strengthen its basicity while not The X-ray diffraction (XRD) patterns of the catalysts were
decrease its acidity drastically. By post-graing organoamines recorded on a Bruker D8 Advance X-ray powder diffractometer
˚
with the surface silanol groups of the zeolite materials, highly with Cu Ka radiation of l ¼ 1.54 A (40 kV and 40 mA) from 2q 5¼
ꢀ
ꢀ
ꢀ
ꢁ1
stable catalysts containing solid acid and base catalytic sites 10 to 90 with a scan speed of 2 min . N adsorption–
2
could be synthesized. The surface amino-functionalization of desorption isotherms measurements were performed using
ꢀ
supporters has already been conducted to the surface modi- a Micromeritics ASAP 2460 analyzer at ꢁ196 C. Prior to the
3
1
32,33
34,35
36
ꢀ
cation of silica, MCM-48,
SBA-15,
MCM-41 and ion analysis, samples were degassed at 150 C for 6 h under
3
7
exchange resins. The structural and physicochemical proper- vacuum. The total surface area was calculated by employing the
ties of the catalysts were systematically characterized upon multipoint BET method and the microporous pore volume was
functionalization and the functionalized catalysts exhibited determined using the t-plot method. The C and N contents of
improved activity for the self-condensation of cyclohexanone, the Sn-Beta-NH catalysts were determined on Elementar Vario
2
32–37
ꢀ
Knoevenagel reaction, and Henry reaction.
Considering that EL III elemental analyzer at 1000 C. The contents of Si and Al of
the Sn-Beta zeolite contains surface silanol groups, it could also the catalysts were measured by inductively coupled plasma
be modied by surface amino-functionalization. Herein, we optical emission spectroscopy (ICP-OES, PerkinElmer Optima
report that glucose can be efficiently converted to LA under 2100DV). Prior to the measurements, the samples were digested
ꢀ
hydrothermal conditions over amino groups functionalized Sn- in an acidic mixture (HCl–HNO –HF) at 150 C for 12 h. X-ray
3
Beta catalyst. A systematical optimization of reaction parame- photoelectron spectroscopy (XPS) studies were performed on
ters is conducted and the possible reaction pathway is also a RBD upgraded PHI-5000C ESCA system (PerkinElmer) with Al
proposed based on structural and acid–base characterization.
Ka radiation (hn ¼ 1486.6 eV). Charge referencing was done
against adventitious carbon (C1s at 284.6 eV). The acidic prop-
erties of the catalysts were tested by adsorption, followed by
temperature programmed desorption (TPD) of pyridine using
Fourier transform infrared spectroscopy (Py-IR). The Py-IR
Experimental
Materials
All reagents were used without further purication. Beta zeolite spectra were recorded on a PerkinElmer Frontier FT-IR in the
ꢁ
1
ꢁ1
was obtained from the Catalyst Plant of Nankai University. range of 1400–1700 cm with a resolution of 2 cm . A 10 mg of
Tin(II) acetate (95%), lactic acid (LA) (1.0 N), 5- sample was pressed into a self-supported wafer with a diameter
18990 | RSC Adv., 2019, 9, 18989–18995
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
of 13 mm. Then, the wafer was set in a quartz IR cell which was
sealed with CaF windows and connected to a vacuum system.
2
ꢀ
The samples were evacuated at 450 C for 2 h under vacuum.
Aer cooling, pyridine vapor was employed into the cell and the
adsorption was performed for 0.5 h. Subsequently, desorption
ꢀ
ꢀ
steps were conducted at different temperatures (25 C, 150 C
ꢀ
and 250 C) for 1 h. The alkaline properties of the samples were
performed on Micromeritics AutochemII 2920 chemisorption Fig. 1 Proposed procedures for preparation of the Sn-Beta-NH
catalyst.
2
2
analyzer using the CO -TPD method. The sample was rst
ꢀ
ꢀ
ꢁ1
purged with Ar and then heated to 550 C at a rate of 10 C min
ꢀ
for 1 h and nally cooled to 45 C. The surface of the catalyst was graed on the Sn-Beta zeolites surface in formation of Si–O–Si
saturated with CO for 30 min at this temperature, and then, the bonds.
2
excess CO
2
was purged with Ar blowing. The TPD curve of the
The XRD patterns in Fig. 2A conrmed that the topology of
sample was obtained at the temperature range from 45 to 800 C Beta zeolite was well preserved for the prepared deAl-Beta in
with a heating rate of 10 C min under an Ar ow and HNO solution. Then the structure of deAl-Beta changed during
3
ꢀ
ꢀ
ꢁ1
3
ꢁ1
a owing rate of 30 cm min
.
the process of incorporation of Sn species. The intensity of the
ꢀ
peak located at 2q ¼ 22.5–22.6 decreased aer Sn incorpora-
tion, indicating a lower crystallinity. The Sn-Beta-NH possessed
2
Catalytic reactions and products analysis
a similar crystallinity to Sn-Beta, showing that the further post-
graing of organoamines process did not destroy the Sn-Beta
zeolite structure. Meanwhile, a slight amount of tin oxide was
The catalytic conversion of glucose to LA was carried out in
a Teon-lined stainless steel autoclave reactor with an inner
volume of 50 mL. In a typical procedure, 225 mg of glucose
ꢀ
observed, as evidenced by the occurrence of the peaks at 26 ,
(7.5 mmol of carbon) and 160 mg of catalyst were mixed in
ꢀ ꢀ
3
4 and 51 , possibly due to the high dosage of tin(II) acetate
10 mL of H O in the autoclave, followed by heating to the
39
2
used during the SSIE procedure. Fig. 2B presents the N2
adsorption–desorption isotherms of all catalysts. Typical type-I
and type-IV isotherms with two steep steps at the relative
pressure of P/P0 < 0.01 and 0.6 < P/P0 < 0.9, corresponding
respectively to micropore lling and capillary condensation in
the mesopores, were observed. It can also be conrmed that the
most probable pore sizes of all catalysts were about 3.9 nm
ꢀ
desired reaction temperature (190 C) and keeping for the
appropriate time (2 h). Aer reaction, the autoclave was cooled,
and the liquid mixture was separated by centrifugation and

ltration. The concentrations of organic acids, HMF and sugars
were analyzed on an Agilent 1200 series HPLC. For organic acids
and HMF, a Gemini-NX 5u C18 column was used with a UV
detector (210 nm) using a 0.1% aqueous phosphoric acid as the
mobile phase with a ow rate of 1 mL min . The column
temperature was 30 C. The glucose concentration was analyzed
using a RI detector (Bio-Rad HPX-87H column) and the column
and RI detector temperature were 55 C and 45 C, respectively.
The mobile phase was a 5 mM aqueous sulphuric acid solution
with a ow rate of 0.5 mL min . All the concentrations were
determined based on the external standard. The conversion of
sugars was dened as the moles of sugars reacted divided by
initial sugars moles used in the system and the product yields
were determined based on carbon. All experiments were per-
formed for more than three times and the quantitative data
were the average values of the analytical results. The relative
errors were less than 10% for all experiments.
(
Fig. S1†), indicating that the post-graing of organic functional
ꢁ1
groups did not inuence the mesopore size of Sn-Beta zeolite.
Table 1 summarizes the physicochemical properties of the
Beta-type zeolites. The N contents of Sn-Beta-NH2(30) and Sn-
Beta-NH2(200) catalysts measured by CHN elemental analysis
were approximately 0.2 wt% and 0.8 wt%, respectively. The
former value was similar to the theoretical value on the grounds
of the dosage of APTMS, while the latter one was lower than the
theoretical value. The results of ICP-OES show that the
elemental ratio of Si/Al increased from 25 for commercial Beta
ꢀ
ꢀ
ꢀ
ꢁ
1
zeolite to over 1700 for Sn-Beta and Sn-Beta-NH , indicating that
2
Al was almost completely removed during the nitric treatment
process. Fig. 3A gives the XPS spectra of Sn-Beta-NH2(30) catalyst
with ve main elements including Si, C, N, Sn and O. The peaks
Results and discussion
Catalysts preparation and characterizations
The Sn-Beta-NH
2
catalyst was synthesized from a combined
strategy of both Sn loading and organoamines functionaliza-
3
2,33,36,37
tion.
As proposed in Fig. 1, commercial Beta zeolite was
to remove most of Al species in the
pretreated using HNO
3
framework. Then, the Sn sites were incorporated into the
dealuminated zeolite by SSIE. The synthesis of the bifunctional
catalyst, or the post-graing of organoamine groups, was
carried out by stirring a certain amount of APTMS with Sn-Beta
Fig. 2 (A) Powder XRD patterns of the samples. (B) N
isotherms of the samples. The samples are: (a) Beta, (b) deAl-Beta, (c)
2
sorption
zeolite in a polar solvent ethanol. Thus, organoamines were Sn-Beta, (d) Sn-Beta-NH2(30) and (e) Sn-Beta-NH2(200)
.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 18989–18995 | 18991

View Article Online
RSC Advances
Paper
Table 1 Physicochemical properties of the Beta-type zeolites
a
b
c
d
e
e
f
S
BET
V
micro
V
total
D
V
(d)
B
L
N
2
ꢁ1
3
ꢁ1
ꢁ1
Samples (m g ) (cm g
)
(nm)
(mmol g
)
(wt%)
Beta
deAl-
Beta
509
576
0.17
0.21
0.39
0.50
3.89
3.89
0.26
0
0.15
0
0
0
Sn-Beta 451
Sn-Beta- 348
NH2(30)
Sn-Beta- 320
NH2(200)
0.15
0.11
0.35
0.29
3.91
3.90
0
0
0.47
0.40
0
0.2
ꢀ
Fig. 4 (A) Py-IR spectra of the samples at 150 C. (B) CO
2
-TPD profiles
0.11
0.26
3.89
0
0.20
0.8
of the samples. The samples are (a) Beta, (b) deAl-Beta, (c) Sn-Beta, (d)
Sn-Beta-NH2(30) and (e) Sn-Beta-NH2(200)
.
a
b
BET surface area. Micropore volume, calculated by the t-plot
c
d
method. Single point total pore volume at P/P
0
¼ 0.95. The most
ꢁ1
e
at 1448 and 1610 cm ) disappeared aer dealumination by
3
. The subsequent incorporation of Sn species enhanced
the Lewis acid sites, conrmed by the strong absorption band at
probable aperture. The number of Brønsted acid and Lewis acid
sites based on pyridine IR adsorption at 423 K. Determined by CHN HNO
elemental analysis.
f
ꢁ1
1
448 cm . Aer post-graing organoamine groups, the Lewis
acidity was weakened due to the alkalinity of the amine groups
that were observed in the Fig. 3B at 487.1 and 495.5 eV could be
assigned to Sn 3d3/2 and Sn 3d5/2, respectively. These two peaks
were obviously different from the peaks of extra-framework
octahedral Sn species, as oen observed at 494.5 and
and the screening of Sn by the amine groups. When the dosage
ꢁ1
ꢁ1
of APTMS was 30 mL, the Lewis acid sites (0.40 mmol g
)
decreased slightly compared to Sn-Beta zeolite (0.47 mmol g ).
Meanwhile, when the dosage was further increased to 200 mL,
40
4
86.1 eV. Combined with the XRD results, it can be conrmed
that Sn-Beta zeolite synthesized by SSIE method contained
some tetrahedrally coordinated Sn species different from SnO
ꢁ1
the quantity of Lewis acid sites decreased to 0.20 mmol g
(Table 1). The decrease in active sites was ascribed to the bulky
2
.
aminopropyl groups, which were spatially spaced on the surface
The C1s spectra of Sn-Beta-NH2(30) catalyst was shown in Fig. 3C.
The most intensive peak observed at 284.6 eV was attributed to
34
of catalysts. Actually, as the distances between the post-graed
36
aminopropyl groups were small on average, it may lead to
C–H in APTMS. The presence of amine groups (–NH ) was
2
a negative effect on the functionality of Sn species both in the
characterization and reaction, especially under the APTMS
dosage of 200 mL. But the Lewis acid sites quantity of Sn-Beta-
NH2(30) catalyst was only a little less than the one of Sn-Beta
catalyst, which would not inuence the process to yield LA
evident from the N1s spectra shown in Fig. 3D. The standard
binding energy for the free amine group fell in the region of
3
–
99–401 eV. Thus, the peak at 399.9 eV was assigned to the
NH , while the peak at 401.9 eV may be attributed to proton-
ated amine group.
2
41
obviously. The CO -TPD proles of the Beta zeolites and APTMS-
2
The pyridine-probed FTIR analysis can distinguish the
Brønsted and Lewis acid sites readily. Shown in Fig. 4A are the
functionalized catalysts were presented in Fig. 4B. It can be seen
that the APTMS-functionalized catalysts exhibited the new
ꢀ
IR spectra of pyridine adsorbed on various catalysts at 150 C.
ꢀ
strong desorption peak located at about 400 C associated with
For deAl-Beta zeolite, all bands corresponding to the Brønsted
ꢀ
22
moderate base sites at 300–600 C, while the zeolites
unmodied with amino groups possessed no basic sites obvi-
ously. These phenomena conrmed the incorporation of basic
ꢁ1
acid (bands at 1544 and 1633 cm ) and Lewis acid sites (bands
2
groups (–NH ).
Catalytic conversion of glucose to lactic acid
According to above discussion, we discovered that Sn-Beta
zeolite, equipped with Lewis acid sites, could be an acid–base
bifunctional catalyst by simply modifying with APTMS. Fig. 5
showed the glucose conversion and product yields using
different Beta zeolite catalysts. When unmodied commercial
Beta zeolite was used as the catalyst, the yields of main products
consisting of LA, FA, LeA and HMF were all negligible while the
glucose conversion was about 80%. The poor catalytic activity of
the commercial Beta zeolite should be attributed to its plentiful
Brønsted acid sites and triing Lewis acid sites internally. While
in the presence of deAl-Beta zeolite, the yields of the main
products were still low (LA yield 4.78%) and the conversion of
glucose decreased to 40% simultaneously. By modifying the
deAl-Beta zeolite with APTMS to obtain deAl-Beta-NH2(30)
3
Fig. 3 XPS spectra of (A) Sn-Beta-NH2(30), (B) Sn d, (C) C1s and (D) N1s.
18992 | RSC Adv., 2019, 9, 18989–18995
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
Fig. 6 Proposed reaction pathways in the conversion of glucose to
lactic acid.
Fig. 5 Conversion of glucose and yield of LA, FA, LeA and HMF over
Beta, deAl-Beta, Sn-Beta and APTMS-functionalized catalysts. Reac-
organoamines indeed inhibit the formation of side products.
With a further increase in the APTMS dosage, the glucose
conversion remained 100% but LA yield slightly decreased.
When the volume of APTMS used was 200 mL, the yield of LA
decreased to 30%, and the HMF yield was still in decline, down
to 4% approximately. As the APTMS dosage increased, the
amount of Lewis acid sites decreased. The present Lewis acid
sites could still lead to a glucose conversion of 100%. Mean-
while, the production of LA is closely related to the amount of
Lewis acid sites. The sharp decrease of LA yield could be
attributed to the excessive alkalinity and the screening of Lewis
acid sites caused by amino group on the surface, which also led
to the decline in HMF yield. Based on these results, the Sn-Beta-
NH2(30) catalyst was selected for subsequent studies since this
system resulted in the highest LA yield from glucose.
tion conditions: 225 mg of glucose, 160 mg of catalyst, 10 mL of H
2
O,
ꢀ
1
90 C, 2 h.
zeolite, the LA yield slightly changed to 6.54%. The extremely
low LA yields obtained by deAl-Beta-NH2(30) indicated that the
amino group exhibited no catalytic efficiency towards the main
reaction to produce LA. Using Sn-Beta zeolite, however, the
conversion of glucose reached up to 90% and the yield of LA
increased to 23%, along with larger amounts of HMF (16%),
14,25,29
which is similar to the results reported in literatures.
The
catalytic activity of the Sn-Beta zeolite should be ascribed to the
incorporation of Sn species into the deAl-Beta zeolite, and the
resulted generation of various Lewis acid sites. For comparison,
when the Sn-Beta zeolite was modied with 30 mL of APTMS to
form Sn-Beta-NH2(30), the reaction proceeded with a surprising
LA yield (56%) and a decreasing HMF (7%) yield. The LA yield
using Sn-Beta-NH2(30) was also slightly higher than that using
Fig. 8 and S2† revealed that the reaction time and reaction
temperature had a great effect on both the conversion of glucose
and LA yield. At a relatively low reaction temperature (below 190
ꢀ
C), the yield of LA rst increased and then became almost
14
the bimetallic Zn–Sn-Beta catalyst in our previous work (48%).
constant with the increase of reaction time. When the reaction
The HMF yield decreased obviously compared to which was
obtained by Sn-Beta zeolite. These results suggested that the LA
yield was improved signicantly upon small addition of APTMS,
accompanied by a complete conversion of glucose. Further-
more, the introduction of organoaminos drastically restricted
the formation of the main side product HMF.
ꢀ
temperature was 150 C, the LA yield increased from 10 to 32%,
in parallel with the increase in the conversion of glucose (91–
98%) as the reaction time prolonged (Fig. S2†). The poor yields
should be attributed to the unfavorable temperature condition.
It is generally known that the glucose-to-LA proceeds
through multiple steps of isomerization of glucose to fructose
and then conversion of fructose to LA, together with side reac-
tion of the formation of HMF and other subsequent side
14,28
products.
The main reaction to produce LA, is signicantly
enhanced by the generation of Lewis acid sites. In the mean-
time, the side reaction to form HMF, could be restrained by the
basic –NH group (Fig. 6).
2
Reaction parameter studies
The performances of Sn-Beta-NH
2
catalysts with various
dosages of APTMS were studied, as shown in Fig. 7. Without
APTMS, only 23% yield of LA was obtained and the yield of HMF
was about 16%. The presence of APTMS with a volume of 30 mL
remarkably promoted the LA yield to 56% with glucose
conversion up to 100%. Meanwhile, the HMF yield decreased to
Fig. 7 Conversion of glucose and yield of LA, FA, LeA and HMF over
various NH
2
-supporting catalysts. Reaction conditions: 225 mg of
ꢀ
7%
and this result indicated that the post-graed glucose, 160 mg of catalyst, 10 mL of H
2
O, 190 C, 2 h.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 18989–18995 | 18993

View Article Online
RSC Advances
Paper
LA yield was 29%. In this case, the amount of catalytic sites was
relatively insufficient, so it might took a very long period of
reaction time to reach a high LA yield. Obviously, compared to
other catalyst dosages, a reaction time of 2 h was not enough for
the catalyst dosage of 40 mg. In the reaction process, the glucose
was nearly completely converted, and some intermediates were
formed, mainly including pyruvaldehyde and dihydroxyacetone
14
according to our previous work. Further increase in catalyst
amount (>160 mg), an inconspicuous change in LA yield was
observed. The maximum LA yield of 56% was obtained in the
presence of 160 mg of catalyst. The promoted LA yield was
attributed to the increase in the accessibility of active sites.
Overall, the conversion of glucose increased from 98 to 100%,
the yield of LA increased from 29 to 56% and the change in
yields of other products was insignicant with increasing
Fig. 8 Effect of reaction temperature and reaction time on yield of LA catalyst dosage. Additionally, as part of catalyst efficiency, the
over Sn-Beta-NH2(30) catalyst. Reaction conditions: 225 mg of
glucose, 160 mg of catalyst, 10 mL of H O.
2
effect of initial glucose concentration on LA yield was also
investigated. As the glucose dosage increased from 45 to
75 mg, glucose conversion decreased slightly from 100 to 96%,
and LA yield decreased from 67 to 23% in the presence of
60 mg of Sn-Beta-NH2(30) catalyst (Fig. S3†). The possible
6
ꢀ
When the reaction temperature increased to 190 C, a remark-
able LA yield of 56% was achieved for a 2 h reaction. As the
reaction proceeded, no signicant differences in the yield of LA
were observed for a longer reaction time. However, a sharp
decrease in LA yield was observed when the reaction tempera-
1
reason accounting for the decrease in LA yield was the reduced
availability of active sites to convert glucose to LA. For practical
application, apart from LA yield, the products distribution, the
reaction parameters optimization and the cost of catalyst recy-
cling are also of vital importance, thus more systematical
investigation is still needed.
ꢀ
ture was over 190 C, which might be caused by the inactivation
ꢀ
of the supporting amino groups. It was suggested that 190 C
and 2 h were sufficient for the catalytic reaction, and LA was
stable in the system. Moreover, it could be seen that an increase
in the reaction temperature not only increased glucose
conversion and maximum LA yield, but also shortened the
reaction time needed to reach the maximum yield of LA.
Fig. 9 showed the changes in the glucose conversion and LA
yield with catalyst dosage over the Sn-Beta-NH2(30) system. The
conversion of glucose increased from 98 to 100% when the
dosage of catalyst increased from 40 to 160 mg for a 2 h reac-
tion, indicating that the inuence of increasing catalyst amount
to glucose conversion was negligible. Nevertheless, a sharp
increase in LA yield (29–56%) was observed. When the catalyst
dosage was 40 mg, the conversion of glucose was 98%, while the
Following the study of the effect of reaction parameters, the
regeneration and reuse of Sn-Beta-NH2(30) catalyst were also
examined (Fig. S4†). Aer each run, the catalysts were separated
by centrifugation and regenerated by washing with ethanol
several times. Although the recycled catalysts exhibited similar
crystalline structure according to their XRD patterns (Fig. S5†),
the yield of LA decreased notably to 22% aer recycling for 3
runs. In order to gain further insight into the reason for the
decreasing LA yield, Sn content in the reaction solution aer
reaction and centrifugation. The result of ICP test showed that
ꢁ1
the Sn content was about 2 mg L , indicating that the leaking
percent of Sn was below 0.3%, so the decrease of LA yield was
not caused by Sn leaking. The decrease might be attributed to
the screening of active sites and the prevention of glucose
diffusion caused by the coke and humins attached on the
surface of catalysts and the leaching of amine groups, which
was in accordance with the C and N contents shown in Table
S1.† Notably, the surface coke and humins of the catalysts
cannot be efficiently removed even by the hydrothermal method
as mentioned in the catalytic reaction above. Moreover, aer the
ꢀ
catalyst was calcined at the range of 300–550 C for 3–6 h, the
yield of LA was still 20–25%. Therefore, more systematic studies
on this issue are required.
Conclusions
In conclusion, we synthesized a series of Sn-Beta-NH catalysts
2
by means of SSIE and post-graing organoamines. The catalysts
contained both Lewis acid sites and moderate base sites
according to the Py-IR and CO -TPD characterization. The acid–
2
Fig. 9 Effect of catalyst dosage on yield of LA over Sn-Beta-NH2(30)
catalyst. Reaction conditions: 225 mg of glucose, 10 mL of H
2
O,
ꢀ
1
90 C, 2 h.
18994 | RSC Adv., 2019, 9, 18989–18995
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
RSC Advances
base bifunctional catalyst could efficiently convert glucose to LA 16 T. L. Wang, W. P. Deng, B. J. Wang, Q. H. Zhang, X. Y. Wan,
and the side reaction was well suppressed. Using Sn-Beta-
Z. C. Tang, Y. Wang, C. Zhu, Z. X. Cao, G. C. Wang and
NH2(30) catalyst, 56% LA yield was obtained under optimized
hydrothermal conditions (190 C, 2 h) along with only 7% HMF 17 X. Lei, F. F. Wang, C. L. Liu, R. Z. Yang and W. S. Dong, Appl.
H. L. Wan, Nat. Commun., 2013, 4, 2141.
ꢀ
yield. It was evidenced that by the introduction of base sites into
the Lewis acid catalyst, such a high yield of LA could be ach- 18 S. Lux and M. Siebenhofer, Catal. Sci. Technol., 2013, 3,
ieved. We hope that this work would be helpful for synthesizing 1380–1385.
highly efficient heterogeneous catalysts for biomass conversion. 19 F. F. Wang, C. L. Liu and W. S. Dong, Green Chem., 2013, 15,
091–2095.
Catal., A, 2014, 482, 78–83.
2
2
0 D. Cao, W. J. Cai, W. T. Tao, S. Y. Zhang, D. Z. Wang and
D. Z. Huang, Catal. Lett., 2017, 147, 926–933.
Conflicts of interest
The authors declare no competing interests.
21 Y. Q. Wang, F. M. Jin, M. W. Sasaki, F. W. Wang, Z. Z. Jing
and M. Goto, AIChE J., 2013, 59, 2096–2104.
2
2
2
2
2
2 W. Zeng, D. G. Cheng, F. Q. Chen and X. L. Zhan, Catal. Lett.,
Acknowledgements
2009, 133, 221–226.
This work is supported by the National Natural Science Foun-
dation of China (21676205), the National Science Fund for
Distinguished Young Scholars (51625804), the National Key
R&D Program of China (2018YFD1100503) and the Funda-
mental Research Funds for the Central Universities
3 F. Chambon, F. Rataboul, C. Pinel, A. Cabiac, E. Guillon and
N. Essayem, Appl. Catal., B, 2011, 105, 171–181.
4 F. F. Wang, J. Liu, H. Li, C. L. Liu, R. Z. Yang and W. S. Dong,
Green Chem., 2015, 17, 2455–2463.
5 M. S. Holm, S. Saravanamurugan and E. Taarning, Science,
(22120190012, 22120190054).
2010, 328, 602–605.
6 J. Dijkmans, D. Gabri ¨e ls, M. Dusselier, F. Clippel,
P. Vanelderen, K. Houthoofd, A. Maliet, Y. Pontikes and
B. F. Sels, Green Chem., 2013, 15, 2777–2785.
Notes and references
1
2
3
A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107, 2411–
502.
M. Dusselier, P. Van Wouwe, A. Dewaele, E. Makshina and
B. F. Sels, Energy Environ. Sci., 2013, 6, 1415–1442.
27 M. Moliner, Y. Roman-Leshkov and M. E. Davis, Proc. Natl.
Acad. Sci. U.S.A., 2010, 107, 6164–6168.
2
2
8 J. Zhang, L. Wang, G. X. Wang, F. Chen, J. Zhu, C. T. Wang,
C. Q. Bian, S. X. Pan and F. S. Xiao, ACS Sustainable Chem.
Eng., 2017, 5, 3123–3131.
G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106,
4044–4098.
2
3
3
3
9 Y. Wan, M. Q. Zhuang, S. P. Chen, W. D. Hu, J. Sun, J. D. Lin,
S. L. Wan and Y. Wang, ACS Catal., 2017, 7, 6038–6047.
0 X. C. Wang, Y. L. Song, C. P. Huang, F. B. Liang and
B. H. Chen, Green Chem., 2014, 16, 4234–4240.
4
5
A. J. Straathof, Chem. Rev., 2014, 114, 1871–1908.
S. Dutta, J. Kim, Y. Ide, J. H. Kim, M. S. A. Hossain, Y. Bando,
Y. Yamauchi and K. C. W. Wu, Mater. Horiz., 2017, 4, 522–
5
45.
S. Dutta, A. Bhaumik and K. C. W. Wu, Energy Environ. Sci.,
014, 7, 3574–3592.
J. Kuo, N. Suzuki, Y. Yamauchi and K. C. W. Wu, RSC Adv.,
013, 3, 2028–2034.
Y. Y. Li and K. C. W. Wu, Phys. Chem. Chem. Phys., 2012, 14,
3914–13917.
1 L. D. White and C. P. Tripp, J. Colloid Interface Sci., 2000, 232,
6
7
8
9
400–407.
2
2 J. Yang, B. Tang, W. L. Dai, L. D. Li, G. J. Wu and N. J. Guan,
Sci. Sin.: Chim., 2015, 45, 396–404.
2
3
3
3 S. G. Wang, Catal. Commun., 2003, 4, 469–470.
4 J. C. Hicks, R. Dabestani, A. C. Buchanan and C. W. Jones,
Chem. Mater., 2006, 18, 5022–5032.
1
W. H. Hsu, Y. Y. Lee, W. H. Peng and K. C. W. Wu, Catal.
Today, 2011, 174, 65–69.
3
3
3
3
3
5 M. W. McKittrick and C. W. Jones, J. Am. Chem. Soc., 2004,
1
26, 3052–3053.
6 K. K. Sharma and T. Asefa, Angew. Chem., Int. Ed., 2007, 119,
937–2940.
7 B. Xu, Y. Q. Liu, J. G. Wang, X. Q. Ren, W. M. Ma and
W. L. Wu, J. Chem. Eng. Chin. Univ., 2015, 29, 335–340.
8 C. Hammond, S. Conrad and I. Hermans, Angew. Chem., Int.
Ed., 2012, 51, 11736–11739.
9 P. Wolf, M. Valla, A. J. Rossini, A. Comas-Vives, F. Nunez-
Zarur, B. Malaman, A. Lesage, L. Emsley, C. Coperet and
I. Hermans, Angew. Chem., Int. Ed., 2014, 53, 10179–10183.
0 H. Y. Luo, L. Bui, W. R. Gunther, E. Min and Y. Rom ´a n-
Leshkov, ACS Catal., 2012, 2, 2695–2699.
1
1
1
0 R. Datta and M. Henry, J. Chem. Technol. Biotechnol., 2006,
1, 1119–1129.
1 P. Maki-Arvela, I. L. Simakova, T. Salmi and D. Y. Murzin,
Chem. Rev., 2014, 114, 1909–1971.
2 W. P. Deng, Y. Z. Wang, S. Zhang, K. M. Gupta, M. J. H u¨ lsey,
H. Asakura, L. M. Liu, Y. Han, E. M. Karp, G. T. Beckham,
P. J. Dyson, J. W. Jiang, T. Tanakac, Y. Wang and N. Yan,
Proc. Natl. Acad. Sci. U. S. A., 2018, 115, 5093–5098.
3 X. Y. Lv, B. Yu, X. W. Tian, Y. Chen, Z. J. Wang, Y. P. Zhuang
and Y. H. Wang, J. Taiwan Inst. Chem. Eng., 2016, 61, 124–
8
2
1
4
4
131.
14 W. J. Dong, Z. Shen, B. Y. Peng, M. Y. Gu, X. F. Zhou, B. Xiang
and Y. L. Zhang, Sci. Rep., 2016, 6, 26713.
1 M. Chong and X. S. Zhao, J. Phys. Chem. B, 2003, 103, 12650–
12657.
15 W. P. Deng, Y. Wang and N. Yan, Curr. Opin. Green Sustain.
Chem., 2016, 2, 54–58.
This journal is © The Royal Society of Chemistry 2019
RSC Adv., 2019, 9, 18989–18995 | 18995