Catalytic transfer hydrogenation of ethyl levulinate to Γ-valerolactone over a novel porous Zirconium trimetaphosphate

Article abstract of DOI:10.1016/j.mcat.2017.09.011

The synthesis of γ-valerolactone (GVL) from catalytic transfer hydrogenation (CTH) of levulinic acid (LA) and its esters is attracting more and more attention due to its wide application in additive, solvent, and precursor. Herein, a novel Zirconium trimetaphosphate (Zr-TMPA) was successfully synthesized and characterized by Fourier transform infrared, N₂ adsorption-desorption, powder X-ray diffraction, scanning electron microscopy, transmission electron microscopy, and NH₃/CO₂-TPD. The as-synthesized Zr-TMPA was used to catalyze the CTH reaction of EL to GVL using isopropanol as the hydrogen donor and solvent. Experimental results shown that Zr-TMPA could catalyze the CTH reaction and 96.2% GVL yield could be achieved at 160 °C for 8 h. Key to this success was attributed to the acid sites and basic sites of the prepared catalyst (Zr-TMPA). Finally, a possible reaction mechanism was proposed.

Full text of DOI:10.1016/j.mcat.2017.09.011

Molecular Catalysis 442 (2017) 107–114
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Research Paper
Catalytic transfer hydrogenation of ethyl levulinate to ␥-valerolactone
over a novel porous Zirconium trimetaphosphate
Yongdi Xie , Fan Li , Jianjia Wang , Ruiying Wang , Haijun Wanga,∗, Xiang Liua,
a
b
a
a
Yongmei Xiac
a
The Key Laboratory of Food Colloids and Biotechnology, Ministry of Education, School of Chemical and Material Engineering, Jiangnan University, Wuxi
2
14122, China
b
School of Food Science and Technology, Jiangnan University, Wuxi 214122, China
State Key Laboratory of Food Science & Technology, Jiangnan University, Wuxi 214122, China
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 18 July 2017
Received in revised form 8 September 2017
Accepted 11 September 2017
The synthesis of ␥-valerolactone (GVL) from catalytic transfer hydrogenation (CTH) of levulinic acid (LA)
and its esters is attracting more and more attention due to its wide application in additive, solvent,
and precursor. Herein, a novel Zirconium trimetaphosphate (Zr-TMPA) was successfully synthesized
and characterized by Fourier transform infrared, N2 adsorption-desorption, powder X-ray diffraction,
scanning electron microscopy, transmission electron microscopy, and NH3/CO2-TPD. The as-synthesized
Zr-TMPA was used to catalyze the CTH reaction of EL to GVL using isopropanol as the hydrogen donor
and solvent. Experimental results shown that Zr-TMPA could catalyze the CTH reaction and 96.2% GVL
yield could be achieved at 160 C for 8 h. Key to this success was attributed to the acid sites and basic
sites of the prepared catalyst (Zr-TMPA). Finally, a possible reaction mechanism was proposed.
© 2017 Elsevier B.V. All rights reserved.
Keywords:
␥
-Valerolactone
Ethyl levulinate
◦
Catalytic transfer hydrogenation
Zirconium trimetaphosphate
Catalytic mechanism
1
. Introduction
developed a highly active and stable r-Ru-NH -␥-Al O catalyst,
2
2
3
◦
which obtained 99.1% yield of GVL at 25 C for 13 h [19]. Jiang et al.
found that mixed MgO–Al O supported Ni catalysts could achieve
At present, the chemical and energy industries rely heavily on
2
3
◦
the use of fossil resources, which brings a lot of serious problems, so
we need to seek an alternative resource for the production of valu-
able chemicals [1–3]. Biomass, which is a carbon-neutral resource,
has been regarded as the renewable resource for the production of
valuable chemicals [4,5]. Such as the most common levulinic acid
99.7% yield of GVL at 160 C, 1 h, and 3 MPa H2 [20]. Although
these catalysts have achieved good yield, they are still very diffi-
cult to apply in the industry because of the use of precious metals
or molecular hydrogen. On the contrary, transfer hydrogenation,
which using alcohols or formic acid (FA) as hydrogen donors, has
aroused great concern [21]. Ruppert et al. reported the catalytic
hydrogenation of LA with FA as a hydrogen source over Ru/C cata-
lysts [22]. But the corrodibility of FA has limited the application in
the CTH reaction. In recent years, the use of alcohol as a hydrogen
donor has become mainstream in the CTH of LA and its esters to GVL.
Most of them are secondary alcohols, for example, Cai et al. reported
(
LA) [6], 5-hydroxymethylfurfural (HMF) [7], 2,5-dimethylfuran [8],
furfural [9], furfuryl alcohol [10], ␥-valerolactone (GVL) [11], and
lactic acid [12], etc. Among them, GVL is one of the most valuable
platform molecules, which can be applied directly as a food and fuel
additive [13,14]. What’s more, it has been used as a precursor for the
production of 1,4-pentanediol, 2-methyltetrahydrofuran, perfume,
and valeric acid [15–17].
that 10Cu-5Ni/Al O3 had the highest activity for the CTH reaction
2
Generally, GVL was mainly prepared from LA and its esters by
direct or transfer hydrogenation. The former was usually conducted
using various metal catalysis (such as Ru, Ni, Au, Ir, Pt, Pd, Rh) in
the presence of external hydrogen [18]. Among them, the Ru- and
Ni-based catalysts showed the highest catalytic activity. Tan et al.
of EL to GVL with 2-butanol as the hydrogen donor [23]. Chia and
Dumesic found that reduction of LA and its esters to GVL could
be achieved using secondary alcohols as the hydrogen donor over
ZrO catalysts [24]. It was not difficult to find that the non-precious
2
metals catalysts have caused widespread concern. Recently, many
®
catalysts, such as RANEY Ni [25], Zr-beta zeolites [26], Zr(OH)4
[
27], Ni-Zr [28], Al-Zr [29], Ni-Fe/AC [30], and SnO /SBA-15 [31],
2
had been successfully applied to the reduction of LA and its esters
to GVL. Especially, song and co-workers continuously reported that
∗ Corresponding author.
E-mail address: wanghj329@outlook.com (H. Wang).
http://dx.doi.org/10.1016/j.mcat.2017.09.011
2
468-8231/© 2017 Elsevier B.V. All rights reserved.

1
08
Y. Xie et al. / Molecular Catalysis 442 (2017) 107–114
Zr-PhyA, Zr-HBA, and Hf-ATMP were used as efficient catalyst for
the CTH of ethyl levulinate (EL) to GVL with isopropanol (IPA) as
the hydrogen source and solvent [32–34]. Xue et al. synthesized a
porous Zr-CA catalyst that has very high activity for the CTH of LA
and its esters to GVL [35]. Although good yield of GVL and conver-
sion of LA and its esters have been achieved, it is still necessary to
overcome the low catalyst stability and harsh reaction conditions.
Therefore, developing more efficient and stability catalysts for the
CTH reaction of LA and its esters to GVL is significant.
Sodium trimetaphosphate (STMP, Schemes 1), which have three
phosphate groups in its structure, is mainly used as a starch mod-
ifier, dispersant, and stabilizer in the food industry. The strong
complexation of phosphoric acid makes it easy to combine with
the four-valent metals [36]. In this wook, we synthesized a novel
Zirconium trimetaphosphate (Zr-TMPA) catalyst by the reaction
of Sodium trimetaphosphate and ZrOCl2 for the CTH reaction of
LA and its esters to GVL in the presence of isopropanol. As far as
we know, Zr-TMPA as a catalyst for this reduction has not been
reported. In addition, the effect of reaction time, temperature, and
catalyst dosage for the CTH reaction of EL to GVL were investigated,
and a possible reaction mechanism was put forward.
2.3. Catalyst characterization
Fourier transform infrared (FT-IR) spectra were recorded on
a Nicolet 360 FT-IR instrument (KBr discs) in the wavenumber
−
1
range of 4000–500 cm . Powder X-ray diffraction (XRD) pat-
terns were obtained on a Bruker D8 Advance diffractometer with
◦
Cu-K␣ radiation with a scanning rate of 4 /min at 40 kV and
20 mA. Scanning electron microscopy (SEM) images were obtained
using a HITACHI S-4800 field-emission scanning electron micro-
scope operated at 15 kV. Transmission electron microscopy (TEM)
measurements were performed on a JEOL JEM-2100 microscope
operated at 120 kV. The N2 adsorption-desorption isotherm using
a Micromeritics ASAP 2020 provided the porosity properties of cat-
alysts. The X-ray photoelectron spectroscopy (XPS) measurements
were implemented on Perkin Elmer PHI 5000 ESCT System. The
contents of P and Zr in Zr-HMPA and Zr-TMPA were determined
by ICP-AES (Optima 8300). Temperature-programmed desorption
of carbon dioxide (CO -TPD) was performed on Micromeritics
2
AutoChem II 2920 Chemisorption analyzer. In the experiment, the
catalyst was charged into the quartz reactor, and the tempera-
◦
ture was increased from room temperature to 300 C at a rate of
◦
3
1
0 C/min under a flow of He (50 cm /min), and then the catalyst
◦
was kept at 300 C for 5 h. After that, the temperature was decreased
to 100 C. CO2 (50 cm /min) was pulsed into the reactor at 100 C
under a flow of He (10 cm /min) until the basic sites were satu-
2
. Experimental section
◦
3
◦
3
2.1. Materials
rated with CO . The adsorbed CO was removed by a flow of He
2
2
3
(
50 cm /min). When the baseline was stable, the temperature was
increased from 60 C to 600 C at a rate of 10 C/min.
Sodium hexametaphosphate (SHMPA, AR), ␥-valerolactone
◦
◦
◦
(
(
98%), Sodium trimetaphosphate (STMPA, 95%), levulinic acid
99%), methyl levulinate (98%), ethyl levulinate (98%), butyl levuli-
Temperature-programmed desorption of ammonia (NH -TPD)
was performed on Micromeritics AutoChem II 2920 Chemisorp-
tion analyzer. The catalyst was charged into the quartz reactor, and
3
nate (98%), ZrOCl ·8H O (AR), and ZrO (AR) were purchased from
2
2
2
Aladdin Industrial Inc. (Shanghai, China). SnCl ·5H O (AR), TiSO
◦
4
2
4
the temperature was increased from room temperature to 300 C
(
AR), pyridine (AR), benzoic acid (AR), ethanol (AR), isopropanol
AR), and naphthalene (AR) were purchased from Sinopharm
◦
3
at a rate of 10 C/min under a flow of He (50 cm /min), and then
(
◦
the catalyst was kept at 300 C for 5 h. After that, the temperature
Chemical Reagent Co. Ltd. (Shanghai, China). Deionized water was
produced with a laboratory water-purification system (RO DI Dig-
ital plus). All reagents were commercially available and without
further purification.
◦
3
was decreased to 100 C. NH /He (10/90, 50 cm /min) was pulsed
into the reactor at 100 C under a flow of He (10 cm /min) until
the acid sites were saturated with NH . The adsorbed NH was
3
◦
3
3
3
3
removed by a flow of He (50 cm /min). When the baseline was sta-
◦
◦
ble, the temperature was increased from 60 C to 700 C at a rate of
◦
1
0 C/min.
2.2. Catalyst preparation
2.2.1. Synthesis of the Zr-TMPA
2.4. Catalytic transfer hydrogenation reaction
In
a typical procedure, 10 mmol STMPA and 30 mmol
ZrOCl ·8H O were dissolved in deionized water (200 mL), respec-
In a typical experiment, EL (1 mmol), isopropanol (5 mL) and
the catalyst (200 mg) were charged into a stainless reactor of 25 mL
equipped with a magnetic stirrer. The reactor was sealed and placed
into a preheated oil-bath at a known temperature for the desired
time. After the reaction, the liquid samples were analyzed quan-
titatively by gas chromatography (GC 9790) using naphthalene as
the internal standard, and identification of the products was done
by GC–MS (ULTRA QP2010). The yield of GVL and the conversion of
EL were calculated using the following equations:
2
2
tively. Then, the solution of STMPA was dropwise added to the
solution of ZrOCl ·8H O in a stirred state. After that, the mixture
2
2
was continuously stirred for 4 h and then aged for 12 h at room
temperature. The white precipitate was separated by centrifuga-
◦
tion, thoroughly washed with water and ethanol, and dried at 80 C
under vacuum for 12 h. For comparison, we synthesized three other
Zr-TMPA with different Lewis acidity and basicity by the reaction of
1
0 mmol STMPA and X mmol ZrOCl ·8H O (X = 10, 20, 40), which
2
2
denoted as Zr-TMPA-1, Zr-TMPA-2 and Zr-TMPA-4, respectively.
Meanwhile, other catalysts with different metal ions were synthe-
sized using a similar route for Zr-TMPA.
Moles of GVL formed
Yield =
× 100%
Moles of EL used
Moles of EL converted
Moles of EL used
Conversion =
× 100%
2
.2.2. Synthesis of the Zr-HMPA
In typical procedure, 10 mmol SHMPA and 60 mmol
a
ZrOCl ·8H O were dissolved in deionized water (200 mL), respec-
2
2
tively. Then, the solution of SHMPA was dropwise added to the
2.5. Reusability of the Zr-TMPA
solution of ZrOCl ·8H O in a stirred state. After that, the mixture
2
2
was firstly stirred for 4 h and then aged for 12 h at room tem-
perature. The white precipitate was separated by centrifugation,
To investigate the reusability of the Zr-TMPA catalyst, the cat-
alyst was recovered by centrifugation and washed with ethanol.
After drying under vacuum at 80 C for 18 h, the catalyst was reused
◦
◦
thoroughly washed with water and ethanol, and dried at 80 C
under vacuum for 12 h.
for the next run.

Y. Xie et al. / Molecular Catalysis 442 (2017) 107–114
109
Fig. 1. Characterization of Zr-TMPA. N2 adsorption-desorption isotherm (a), Powder XRD pattern (b), SEM image (c), and TEM image (d).
Fig. 3. Effect of catalyst amount. Reaction conditions: 1 mmol EL; 5 mL isopropanol;
reaction temperature 160 C; reaction time 8 h.
Fig. 2. FT-IR spectra of Zr-TMPA and STMPA.
◦
3
. Results and discussion
particles are shown in Fig. 1c and d. It can be observed that the Zr-
TMPA had no uniform shape and the gaps between the aggregated
particles result to the porosity of the catalyst, which results were
consistent with the XRD results. In addition, the N2 adsorption-
desorption isotherm of the Zr-TMPA belonged to type IV mode with
a hysteresis loop located in the range of 0.4-0.9 relative pressures
(Fig. 1b), indicating the presence of irregular mesoporous struc-
ture in the Zr-TMPA. The porosity of the Zr-TMPA was caused by
the gaps between the aggregated particles. [33] Moreover, the sur-
face area, average pore diameter, and pore volume of Zr-TMPA
3.1. Catalyst characterization
The prepared Zr-TMPA was characterized by N2 adsorption-
desorption isotherm, XRD, SEM, TEM, and FT-IR. The XRD pattern
of the Zr-TMPA was shown in Fig. 1a. It was clear that the Zr-TMPA
had no X-ray diffraction structure. In general, the presence of weak
coordination groups in the catalyst structure resulted in low degree
crystallinity. Therefore, the low degree crystallinity of the Zr-TMPA
was derived from the only weak coordination groups (phosphate
groups) in the structure of the Zr-TMPA. Meanwhile, the XRD pat-
2
−1
3
−1
were 361 m g , 3.7 nm, and 0.28 cm g , respectively. All of
the above parameters proved that Zr-TMPA, which had low crys-
tallinity and porous structure, could serve as a good catalyst. The
FT-IR spectrum of Zr-TMPA and STMPA were shown in Fig. 2. Peaks
◦
tern had one broad diffraction peak at 2␪ = 26 , suggesting that
the Zr-TMPA was amorphous. Simultaneously, SEM and TEM were
conducted to examine the surface morphology and internal com-
position of the prepared catalyst. SEM and TEM images of Zr-TMPA
−
1
−1
were attributed to the stretching
at 3441 cm
and 1627 cm
vibration of hydroxyl groups on the catalyst surface and physically

1
10
Y. Xie et al. / Molecular Catalysis 442 (2017) 107–114
Fig. 4. Effect of reaction time and temperature on the GVL yield (a) and EL conversion (b). Reaction conditions: 0.2 g catalyst; 1 mmol EL; 5 mL isopropanol.
Fig. 5. Leaching experiment of Zr-TMPA for the CTH of EL to GVL (a) and reusability of the Zr-TMPA (b). Reaction conditions: 0.2 g catalyst; 1 mmol EL; 5 mL isopropanol;
◦
reaction temperature 160 C; reaction time 8 h.
adsorbed water. [37] The strong bond at 1062 cm−1 was assigned to
the stretching vibrations of P–O–Zr network, which could be taken
Table 1
a
Performance of different catalysts for the CTH of EL to GVL.
as an evidence of the structural formation of Zr-TMPA. [38] The
Entry
Catalyst
Conversion (%)b
GVL yield (%)b
Selectivity
−1
weak bond at 756 cm should be attributed to the P-O-P bending
1
2
3
4
5
6
7
8
9
None
0
0
96.2
0.9
2.8
0.2
0
96.2
4.7
31.1
25
45.5
95.3
89.3
51.9
95.5
4+
vibration. [39] Thus, it could be deduced that Zr was coordi-
nated to phosphate groups. For comparison, other metal (Ti, Sn)
Zr-TMPA
Ti-TMPA
Sn-TMPA
STMPA
SHMPA
Zr-HMPA
ZrO2
100
19.3
9.0
0.8
1.1
99.1
39.4
88.1
100
−
TMPA catalysts and other ligand catalyst Zr-HMPA were also syn-
thesized with similar textural and structural properties (Figs. S1-S4
and Table S1).
0.5
94.4
35.2
45.7
95.5
ZrOCl2
Zr-TMPA
c
3.2. Conversion of EL to GVL over different catalysts
10
a
Reaction conditions: 0.2 g catalyst; 1 mmol EL; 5 mL isopropanol; reaction tem-
The catalytic research starts from the conversion of EL to GVL
perature 160 ◦C; reaction time 8 h.
b
with isopropanol as the H-donor and solvent. As shown in Table 1,
the CTH reaction did not happen in the absence of catalysts (Table 1,
entry 1). When TMPA-based catalysts were used for this reaction,
it was obvious that Zr-TMPA showed the highest activity with a
Conversion of EL and GVL yield were determined by GC.
c
◦
The reaction temperature was 170 C.
entry 8–9), which revealed Zr-TMPA prepared in this work was an
admirable catalyst for the CTH reaction of EL to GVL. The reason for
the high activity of the catalyst will be discussed below.
◦
full conversion of EL and the GVL yield of 96.2% at 160 C for 8 h
(Table 1, entry 2). In contrary, Ti-TMPA and Sn-TMPA were almost
no activity for this reaction, and the byproduct was mainly ispropyl
levulinate from transesterification of EL with IPA (Table 1, entry
3.3. Effect of the amount of catalyst
3
–4). Afterward, we used STMPA and SHMPA as a catalyst for the
CTH reaction of EL to GVL, only about 1% conversion of EL could be
observed (Table 1, entry 5–6). These results indicated Zr was the
activity center for the CTH reaction of EL to GVL, and the reaction
hardly happens without Zr element. However, Zr-HMPA was also
shown the better activity with 99.1% conversion of EL and 94.4%
yield of GVL at the same condition (Table 1, entry 7). In contrast,
ZrO2 and ZrOCl2 only obtained a moderate yield of GVL (Table 1,
The effect of the Zr-TMPA dosage on the CTH reaction of EL to
GVL was studied at 160 C with a reaction time of 8 h (Fig. 3). The
◦
yield and selectivity of GVL and the conversion of EL increased
rapidly when the amount of catalyst increased from 0.025 g to
0.05 g. After the dosage of catalyst was increased to 0.2 g, the yield of
GVL and the conversion of EL remained almost unchanged, achiev-
ing the maximum yield of GVL (96.2%) and conversion of EL (100%).

Y. Xie et al. / Molecular Catalysis 442 (2017) 107–114
111
Table 2
CTH of LA and its esters into GVL over Zr-TMPA.a
Entry
Substrate
Time (h)
Conversion (%)b
GVL yield (%)b
Selectivity (%)b
1
2
3
4
Levulinic acid
4
8
8
8
100
100
99.9
90.2
97.7
96.5
96.2
83.3
97.7
96.5
96.3
92.4
Methyl levulinate
Ethyl levulinate
Butyl levulinate
a
◦
Reaction conditions: 0.2 g catalyst; 1 mmol EL; 5 mL isopropanol; reaction temperature 160 C.
b
Conversion of EL and GVL yield were determined by GC. The other products were mainly isopropyl levulinate obtained by esterification or transesterification.
3.6. The leaching and reusability of the catalyst
The stability and reusability of the catalyst were an important
criterion for measuring the quality of a catalyst, hence a leaching
experiment was carried out. For compared with the reaction that
◦
conducted at 160 C for 9 h in the presence of Zr-TMPA, another
reaction was continued after removing the catalyst at 3 h. As shown
in Fig. 5a, the yield of GVL was no further increase after removing Zr-
TMPA, suggesting that the active site of Zr-TMPA was not lost and
Zr-TMPA was heterogeneous in the reaction. In addition, ICP-AES
analysis also shown that Zr leaching was negligible in the solution
after the reaction.
To investigate further the reusability of the Zr-TMPA catalyst,
recycle tests for the CTH reaction of EL to produce GVL were per-
formed under the optimized reaction condition. EL conversion and
GVL yield only slightly decreased after five repeated runs (Fig. 5b).
Meanwhile, the reused Zr-TMPA after five cycles was character-
ized by XRD, FT-IR, SEM, TEM (Fig. S5), N2 adsorption-desorption,
and XPS (Table S2), these results suggested that the structure and
properties of the Zr-TMPA were almost unchanged after five cycles.
Thence, Zr-TMPA was stable in the reaction system.
Fig. 6. Effect of catalysts with different Lewis acidity and basicity. Reaction con-
ditions: 0.2 g catalyst; 1 mmol EL; 5 mL isopropanol; reaction temperature 160 C;
reaction time 8 h.
◦
Another 3.8% EL was shown to be converted to isopropyl levulinate
that was the leading by-product produced by the transesterifica-
tion of EL with isopropanol. Hence, 0.2 g of the Zr-TMPA would be
the optimum dosage in the subsequent experiments.
3.7. Effect of catalysts with different lewis acidity and basicity
3.4. Effect of reaction temperature and time
In order to explain the importance of Lewis acidity and basic-
We studied the effect of reaction time and temperature (Fig. 4).
ity to the CTH reaction of EL to GVL, the other three Zr-TMPA
with different Lewis acidity and basicity were synthesized and
characterized by N2 adsorption-desorption isotherm, CO2-TPD and
NH3-TPD (Fig. S6 and Table S3). It shows clearly that there are a
big difference between the structural properties and acid/base sites
density of the different samples. Although the strength and sun of
acid sites decrease with increasing Zr, the base sites density of Zr-
TMPA higher than others catalysts, indicating that the content of
acid and base sites could be adjusted by varying the molar ratio
of SMTPA and zirconium. When the four catalysts were used for
the CTH reaction of EL to GVL, it was obvious that the catalytic
performance of catalysts followed the order: Zr-TMPA > Zr-TMPA-
2 > Zr-TMPA-1 > Zr-TMPA-4 (Fig. 6). These results were consistent
with the results of characterization, manifesting that the activity
of the catalyst is closely related to the amount of acid and base
sites. But the excellent catalytic activity can only be achieved at the
optimum acid and base sites ratio.
As shown in Fig. 4, it was clear that the reaction time and temper-
ature had a significant effect on the CTH reaction. The yield of GVL
and the conversion of EL gradually increased when the temperature
◦
◦
increased from 140 C to 160 C, which indicated that high temper-
ature was favorable for the CTH reaction of EL to GVL. However, the
yield and selectivity of GVL decreased slightly when the reaction
◦
proceeded at 170 C (Table 1, entry 10). This was possibly attributed
to by-product generated easily at the higher reaction temperature.
◦
Hence, 160 C was chosen as the best reaction temperature. Simi-
larly, it was found that the yield of GVL and the conversion of EL
were relatively low before 5 h due to the formation of isopropyl
levulinate. The yield of GVL and the conversion of EL gradually
increased by prolonging the reaction time. Finally, 96.2% yield of
GVL and 100% conversion of EL could be obtained within 8 h at
◦
1
60 C.
3.5. Catalytic effect of Zr-TMPA on different substrates
3.8. Analyze the high activity of the catalyst
The CTH reaction of LA and its esters to GVL were further investi-
gated over Zr-TMPA under optimized reaction conditions. As listed
in Table 2, LA was easily converted to GVL with 97.7% GVL yield
and full LA conversion for only 4 h (Table 2, entry 1). This might be
due to the acidity of LA itself that could accelerate the lactonization
of 4-hydroxypentanoate. [34] In addition, methyl levulinate could
also achieve good results (Table 2, entry 2). However, compared
with methyl levulinate and ethyl levulinate, butyl levulinate shown
a low activity because of the steric hindrance of butyl (Table 2,
entry 4). The above description fully illustrates that Zr-TMPA was
an excellent catalyst for the CTH of LA and its esters to GVL.
As can be seen from the previous analysis that Zr-TMPA showed
a high reactivity for the CTH reaction of LA and its esters to produce
GVL. Compared with ZrO , this high activity may be attributed to
2
their improved surface area and high Lewis acidity and basicity.
That being the case, the acidity of ZrO and Zr-TMPA were first ana-
2
lyzed by the XPS and NH -TPD (Fig. 7a and b). In general, metal-ions
3
in metal-ligand coordination polymers exhibit Lewis acidity. [40]
Hence, the local environment of Zr species in ZrO2 and Zr-TMPA
were inspected by XPS in order to detect the Lewis acidity caused
by Zr element. As displayed in Fig. 7a, the binding energies of Zr

1
12
Y. Xie et al. / Molecular Catalysis 442 (2017) 107–114
Fig. 7. XPS spectra of Zr 3d (a), NH3-TPD (b), and CO2-TPD (c) of Zr-TMPA, Zr-HMPA and ZrO2.
3
d5/2 and Zr 3d3/2 in Zr-TMPA at 183.1 and 185.5 eV were higher
this structure with phosphate groups could significantly increase
than in ZrO , manifesting higher positive charge on Zr atoms in the
the basicity of the catalyst.
2
Zr-TMPA, which resulted in stronger Lewis acidity. [41] This was
mainly due to the decrease of electron density of Zr caused by the
introduction of phosphorus. [32] In addition, the strength and con-
tents of the acid sites of Zr-TMPA and ZrO2 were determined by
The acidity and basicity of catalysts we discussed above were
very important for the CTH reaction and the structure of the cata-
lysts was also equally important. As shown in Table S1, the surface
area, average pore diameter, pore volume of Zr-TMPA were greater
NH -TPD (Fig. 7b). It can be found that Zr-TMPA have two desorp-
than ZrO . The more the three parameters was more beneficial to
3
2
◦
◦
tion peaks located at 210 C and 500 C, corresponding to the weak
and moderate acid strength, and the amounts of acids were 17.6
increase the diffusion of reactants to the activity center (Zr) in Zr-
TMPA. In summary, the high activity of Zr-TMPA is attributed to the
large number of acid-base sites and excellent pore structure.
3
and 11.6 cm /g, respectively. However, ZrO2 had no appreciable
desorption peaks. So it could be inferred that Zr-TMPA had higher
acidity than ZrO , which was consistent with the results of XPS. The
3.9. Poisoning experiments and reaction mechanism
2
higher Lewis acidity of Zr-TMPA could not only activate carbonyl
groups of LA and its esters and also increase the total reaction rate.
According to the above discussion and previous studies, [32–35]
the acid-base site in the catalyst are critical for the CTH reaction of
LA and its esters. From the results of NH -TPD and CO -TPD, we
[
32]
It has been reported that the basicity of catalyst was crucial for
3
2
CTH reaction due to it could increase the dissociation of hydroxyl
groups in isopropanol. [33] Therefore, the basicity of Zr-TMPA and
ZrO were examined by CO -TPD (Fig. 7c) and the amounts of basics
can clearly see that there are plenty of acid and basic sites on the
2−
Zr-TMPA. The basic sites are derived from O in the phosphate
groups and acid sites originate from the Zr4+. In order to gain more
insight toward the role of active sites in Zr-TMPA, Lewis acid and
base sites were separately poisoned with pyridine and benzoic acid,
respectively. As shown in Table S4, a significant reduce in EL con-
version and GVL selectivity was observed when benzoic acid was
added (Table S4, entry 2). It is probable that the strong interaction
between benzoic acid and base sites. However, the catalytic activ-
ity of Zr-TMPA was slightly decreased in the presence of pyridine
(Table S4, entry 3). This can be explained by the poor adsorption
performance of pyridine on the surface of the catalyst led to the
weak poisoning effect. [42] It is concluded that the activity of cata-
lyst is more closely related to the base sites than the acid sites. But
the outstanding catalytic effect for the CTH reaction of EL to GVL
2
2
3
3
were 28.6 cm /g and 0.5 cm /g, respectively. Song and co-workers
had reported that the phosphate groups could increase the basicity
of the catalysts. [32,34] Therefore, the high basicity of Zr-TMPA was
due to the presence of phosphate groups in its structure. Addition-
ally, Zr-HMPA have similar structures with Zr-TMPA (Scheme S1),
which was also used to catalyze LA and its esters to GVL and found
that 99.1% conversion of EL and 94.4% yield of GVL were obtained
under the optimized reaction conditions. Zr-HMPA was character-
ized by XRD (Fig. S1), SEM, TEM (Fig. S3), N adsorption-desorption
2
(
Fig. S4), XPS (Fig. 7a), and NH /CO -TPD (Fig. 7b and c). From these
3 2
characterization results, it was not difficult to find that the structure
and property of Zr-HMPA were similar to Zr-TMPA, suggesting that

Y. Xie et al. / Molecular Catalysis 442 (2017) 107–114
113
Scheme 1. Possible mechanism for CTH of LA and its esters to produce GVL over Zr-TMPA.
was achieved without adding pyridine and benzoic acid (Table S4,
entry 1), which results from the synergistic effect of the Lewis acid
and base sites.
was put forward. Hence, this catalyst has great potential for the
catalytic conversion of biomass.
Based on this knowledge, a possible reaction mechanism for the
CTH of LA and its esters to produce GVL over Zr-TMPA was put
forward. As shown in Scheme 1, isopropanol first interacts with
Acknowledgment
This work was financially supported by the MOE & SAFEA for the
4
+
2−
the acid-base site (Zr -O ) on Zr-TMPA leading to the separa-
tion of isopropanol into the corresponding alkoxide. At the same
time, the carbonyl group in LA and its esters can be activated by
1
11 Project (B13025).
Appendix A. Supplementary data
4+
2−
on Zr-TMPA. Then, hydrogen transfer occurs between
Zr -O
the activated carbonyl group and the separated alcohol to form
-hydroxypentanoate or its esters through a six-member interme-
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.mcat.2017.09.
4
diate. Finally, 4-hydroxypentanoate or its esters could be converted
to the desired product GVL via intramolecular esterification or
transesterification.
0
11.
References
[
[
1] M.Y. He, Y.H. Sun, B.X. Han, Angew. Chem Int. Ed. 52 (2013) 9620–9633.
2] M.J. Climent, A. Corma, S. Iborra, Green Chem. 16 (2014) 516–547.
4
. Conclusion
[3] H. Li, Z. Fang, J. Luo, S. Yang, Appl. Catal. B: Environ. 200 (2017) 182–191.
[
4] Q. Xu, X.L. Li, T. Pan, C.G. Yu, J. Deng, Q.X. Guo, Y. Fu, Green Chem. 18 (2015)
287–1294.
1
In summary, we prepared an efficient and robust Zr-TMPA cat-
[
5] L. Wang, H. Wang, F.J. Liu, A.M. Zheng, J. Zhang, Q. Sun, James P. Lewis, L.F.
Zhu, X.J. Meng, F.-S. Xiao, ChemSusChem 7 (2014) 402–406.
6] F.D. Pileidis, M.-M. Titirici, ChemSusChem 9 (2016) 562–582.
7] R.-J. van Putten, J.C. van der Waal, E. de Jong, C.B. Rasrendra, H.J. Heeres, J.G.
de Vries, Chem. Rev. 113 (2013) 1499–1597.
alyst for the catalytic transfer hydrogenation reaction of LA and its
esters to produce GVL. EL conversion of 100% and GVL yield of 96.2%
can be obtained under optimized reaction conditions. The Lewis-
acidity of Zr and the Lewis-basicity of phosphate groups were
responsible for the high reactivity of catalyst. Moreover, repeated
tests shown that the Zr-TMPA catalyst could be reused at least five
times and its activity is hardly reduced. Finally, a probable reaction
mechanism for LA and its esters catalytic transfer hydrogenation
[
[
[8] G.-H. Wang, J. Hilgert, F.H. Richter, F. Wang, H.-J. Bongard, B. Spliethoff, C.
Weidenthaler, F. Schüth, Nat. Mater. 13 (2014) 293–300.
[
9] R. Mariscal, P. Maireles-Torres, M. Ojeda, I. Sadaba, M.L. Granados, Energy
Environ. Sci. 9 (2016) 1144–1189.
[10] E. Appiani, R. Ossola, D.E. Latch, P.R. Erickson, K. McNeill, Environ. Sci.:
Process. Impacts 19 (2017) 507–516.

1
14
Y. Xie et al. / Molecular Catalysis 442 (2017) 107–114
[
11] X. Tang, X. Zeng, Z. Li, L. Hu, Y. Sun, S. Liu, L. Lin, Renew. Sustain. Energy Rev.
[28] H. Li, Z. Fang, S. Yang, ChemPlusChem 81 (2016) 135–142.
[29] J. He, H. Li, Y.M. Lu, Y.X. Liu, Z.B. Wu, D.Y. Hu, S. Yang, Appl. Catal. A Gen. 510
(2016) 11–19.
[30] L. Cai, G.Y. Xu, Y.X. Zhai, X.H. Liu, Y.F. Ma, Y. Zhang, Fuel 203 (2017) 23–31.
[31] S.D. Xu, D.Q. Yu, T. Ye, P.P. Tian, RSC Adv. 7 (2017) 1026–1031.
[32] J.L. Song, B.W. Zhou, H.C. Zhou, L.Q. Wu, Q.L. Meng, Z.M. Liu, B.X. Han, Angew.
Chem. Int. Ed. 54 (2015) 9399–9403.
[33] J.L. Song, L.Q. Wu, B.W. Zhou, H.C. Zhou, H.L. Fan, Y.Y. Yang, Q.L. Meng, B.X.
Han, Green Chem. 17 (2015) 1626–1632.
[34] C. Xie, J.L. Song, B.W. Zhou, J.Y. Hu, Z.R. Zhang, P. Zhang, Z.W. Jiang, B.X. Han,
ACS Sustain. Chem. Eng. 4 (2016) 6231–6236.
4
0 (2014) 608–620.
[
[
12] D. Esposito, M. Antonietti, ChemSusChem 6 (2013) 989–992.
13] I.T. Horváth, H. Mehdi, V. Fábos, L. Boda, L.T. Mika, Green Chem. 10 (2008)
2
38–242.
[
[
[
[
14] Z.H. Zhang, ChemSusChem 9 (2016) 156–171.
15] D.M. Alonso, S.G. Wettstein, J.A. Dumesic, Green Chem. 15 (2013) 584–595.
16] M.G. Al-Shaal, A. Dzierbinski, R. Palkovits, Green Chem. 16 (2014) 1358–1364.
17] L. Xin, Z.Y. Zhang, J. Qi, D.J. Chadderdon, Y. Qiu, K.M. Warsko, W.Z. Li,
ChemSusChem 6 (2013) 674–686.
[
[
18] W.R.H. Wright, P. Ralkovits, ChemSusChem 5 (2012) 1657–1667.
19] J.J. Tan, J.L. Cui, T.S. Deng, X.J. Cui, G.Q. Ding, Y.L. Zhu, Y.W. Li, Catal. Sci.
Technol. 6 (2016) 1469–1475.
[35] Z.M. Xue, J.Y. Jiang, G.F. Li, W.C. Zhao, J.F. Wang, T.C. Mu, Catal. Sci. Technol. 6
(2016) 5374–5379.
[
20] K. Jiang, D. Sheng, Z.H. Zhang, J. Fu, Z.Y. Hou, X.Y. Lu, Catal. Today 274 (2016)
[36] M.Y. Gu, D.H. Yu, H.M. Zhang, P. Sun, H. Huang, Catal. Lett. 133 (2009)
214–220.
5
5–59.
[
[
21] D. Wang, D. Astruc, Chem. Rev. 115 (2015) 6621–6686.
22] M.R. Agnieszka, J. Marcin, S.-P. Olga, K. Nicolas, S.D. Alexandre, M. Carine, S.
Philippe, G. Jacek, Green Chem. 18 (2016) 2014–2028.
[37] T.Z. Ren, Z.Y. Yuan, B.L. Su, Chem. Phys. Lett. 374 (2003) 170–175.
[38] Y.J. Jia, Y.J. Zhang, R.W. Wang, J.J. Yi, X. Feng, Q.H. Xu, Ind. Eng. Chem. Res. 51
(2012) 12266–12273.
[
23] B. Cai, X.C. Zhou, Y.C. Miao, J.Y. Luo, H. Pan, Y.B. Huang, ACS Sustain. Chem.
[39] J. Jiménez-Jiménez, P. Maireles-Torres, P. Olivera-Pastor, E.
Rodríguez-Castellón, A. Jiménez-López, D.J. Jones, J. Rozière, Adv. Mater. 10
(1998) 812–815.
[40] H.L. Liu, Y.W. Li, H.F. Jiang, C. Vargas, R. Luque, Chem. Commun. 48 (2012)
8431–8433.
Eng. 5 (2017) 1322–1331.
[
[
[
24] M. Chia, J.A. Dumesic, Chem. Commun. 47 (2011) 12233–12235.
25] Z. Yang, Y.B. Huang, Q.X. Guo, Y. Fu, Chem. Commun. 49 (2013) 5328–5330.
26] L. Bui, H. Luo, W.R. Gunther, Y. Román-Leshkov, Angew. Chem Int. Ed. 52
(
2013) 8022–8025.
[41] B. Tang, W.L. Dai, X.M. Sun, G.J. Wu, N.J. Guan, M. Hunger, L.D. Li, Green Chem.
17 (2015) 1744–1755.
[
27] X. Tang, H. Chen, L. Hu, W. Hao, Y. Sun, X. Zeng, L. Lin, S. Liu, Appl. Catal. B:
Environ. 147 (2014) 827–834.
[42] V.A. Ivanov, J. Bachelier, F. Audry, J.C. Lavalley, J. Mol. Catal. 91 (1994) 45–59.