Tingting Yang1, Hu Li1, Jian He1, Yanxiu Liu1, Wenfeng Zhao1, Zhongwei Wang1, Xiaoxiao Ji1, Song Yang1. 1. Center for Research and Development of Fine Chemicals, State-Local Joint Engineering Laboratory for Comprehensive Utilization of Biomass, State Key Laboratory Breeding Base of Green Pesticide and Agricultural Bioengineering (Ministry of Education), Guizhou University (North Campus), Huaxi District, Guiyang, Guizhou 550025, P. R. China.
Abstract
γ-Valerolactone (GVL) is one of the versatile platform molecules and biofuel additives derived from the lignocellulosic biomass. Herein, the efficient synthesis of GVL from biobased ethyl levulinate (EL) using alcohol as both H-donor and solvent without an external hydrogen source has been achieved over porous Ti/Zr microspheres. The catalysts (Ti x Zr y ) with different Ti/Zr molar ratios were synthesized using hexadecylamine (HDA) as a structure-directing agent via a sol-gel process combined with solvothermal treatment and characterized by scanning electron microscopy, transmission electron microscopy, X-ray diffraction, thermal gravimetric analysis, NH3/CO2-TPD, N2 adsorption-desorption, and pyridine-IR. A high GVL yield of 90.1% at 100% EL conversion was obtained at 180 °C for 6 h over Ti2Zr8 in 2-propanol. The microspheric and porous structure, enhanced surface areas, and acid/base contents by the proper introduction of Ti species into Zr oxide were demonstrated to be responsible for the pronounced performance. The microspheric Ti2Zr8 porous catalyst could be reused at least six times with no decrease in catalytic activity.
γ-Valerolactone (GVL) is one of the versatile platform molecules and biofuel additives derived from the lignocellulosic biomass. Herein, the efficient synthesis of GVL from biobased ethyl levulinate (EL) using alcohol as both H-donor and solvent without an external hydrogen source has been achieved over porous Ti/Zr microspheres. The catalysts (Ti x Zr y ) with different Ti/Zr molar ratios were synthesized using hexadecylamine (HDA) as a structure-directing agent via a sol-gel process combined with solvothermal treatment and characterized by scanning electron microscopy, transmission electron microscopy, X-ray diffraction, thermal gravimetric analysis, NH3/CO2-TPD, N2 adsorption-desorption, and pyridine-IR. A high GVL yield of 90.1% at 100% EL conversion was obtained at 180 °C for 6 h over Ti2Zr8 in 2-propanol. The microspheric and porous structure, enhanced surface areas, and acid/base contents by the proper introduction of Ti species into Zr oxide were demonstrated to be responsible for the pronounced performance. The microspheric Ti2Zr8 porous catalyst could be reused at least six times with no decrease in catalytic activity.
With the gradual consumption
of fossil resources and the growing
demand for energy, great efforts have been made to produce biofuels
and chemicals from renewable biomass resources.[1−3] In the past
decades, much attention has been paid to catalytic conversion of biomass
resources into a number of value-added chemicals, such as 5-hydroxymethylfurfural
(HMF),[4] 2,5-dimethylfuran,[5] lactic acid,[6] levulinic acid
(LA) or its esters,[7,8] γ-valerolactone (GVL),[9,10] furfural,[11] and 5-ethoxymethylfurfural
(EMF).[12] Among these platform molecules,
GVL has been considered to be one of the most promising candidates
because of its widespread application.[13] It has been used as a green solvent,[14] fuel additive,[15,16] and food additive.[17] In addition, it can be further hydrogenated
to 2-methyltetrahydrofuran (MTHF) and valeric biofuels.[18−20]Currently, GVL is primarily synthesized from LA or its esters
by
using various hydrogen sources such as molecular hydrogen, formic
acid, and alcohols.[21−24] However, the use of formic acid and hydrogen as H-donors has some
disadvantages, including, for example, costly and unsafe characters,
employment of precious metals (e.g., Pt, Pd, Au, and Ru) and unfriendly
environmental solvents, and involvement of harsh reaction conditions.[25−27] In this respect, increasing attention has been paid to the use of
inexpensive and plentiful alcohols as H-donors, which are free from
contamination and are particularly not dependent on a zero valence
metal catalyst.[28]Catalytic transfer
hydrogenation (CTH) of LA or its esters to GVL using alcohols as H-donors[29] typically
proceeded via Meerwein–Ponndorf–Verley (MPV) reduction.[30] This reaction pathway was initiated though the
hydrogenation of LA or its esters, followed by lactonization to produce
GVL.[31] For this transformation route, Zr-containing
oxides [e.g., ZrO2,[32] Zr-HBA,[33] ZrO2/SBA-15,[34] Zr-β,[35] Cu-ZrO2,[36] ZrFeO,[37] Zr-CA,[38] and ZrO(OH)2·xH2O[39]] showed superior activity to other catalysts on account
of their acid/base amphoteric property.[40] In addition, the pore architecture of the catalysts might affect
the flowing rate of reaction systems through the porous network.[41]Physico-chemical natures including good
stability, large surface
area, porous structure, and moderate acid/base density of the catalysts
are most likely to be beneficial for producing GVL from LA or its
esters.[29,35] In this regard, the spheric
catalysts having enhanced surface areas could not only improve flow
properties, but also facilitate the active sites homogeneously distributed
on the catalyst surface,[42,43] which will definitely
promote the MPV reaction. Inspired by the above findings, herein,
a series of porous and microspheric Ti/Zr mixed oxides with different
Ti/Zr molar ratios were synthesized by a sol–gel process combined
with solvothermal treatment using hexadecylamine (HDA) as a structure-directing
agent.[44] After introducing Ti into Zr,
the resultant −Zr–O–Ti–O–Zr–
network not only rendered the mixed oxides adjustable porosity and
a large surface area[45] but also enhanced
acidity and basicity. The direct production of GVL from ethyl levulinate
(EL) in 2-propanol (2-PrOH) without using external hydrogen was further
used to examine the performance of the porous Ti/Zr microspheres.
Results
and Discussions
Catalyst Characterization
Figure shows X-ray diffraction
(XRD) patterns of
sole TiO2 and ZrO2, and Ti/Zr oxides with different
Ti/Zr molar ratios after calcination at 500 °C. It was clearly
observed that anatase crystals were present in the samples of as-prepared
TiO2 and Ti8Zr2, while ZrO2 was composed of both tetragonal and monoclinic crystals.[46] However, when increasing the molar amount of
Zr species, the resultant Ti/Zr samples (i.e., Ti2Zr8 and Ti5Zr5) showed amorphous structure
after calcination at 500 °C. It was proposed that Ti mixed with
Zr species could be well distributed within the microspheres, thus
preventing the formation of Ti/Zr crystals during calcination.[47] However, the probable reason for phase transition
was that the formation of Zr–O–Ti bonds could inhibit
the mobility of a single Ti/Zr atom in the inorganic framework.[48]
Figure 1
XRD patterns of Ti/Zr oxides with different Ti/Zr molar
ratios.
M, T, and A stand for monoclinic zirconia, tetragonal zirconia, and
anatase titania, respectively.
XRD patterns of Ti/Zr oxides with different Ti/Zr molar
ratios.
M, T, and A stand for monoclinic zirconia, tetragonal zirconia, and
anatase titania, respectively.Nitrogen sorption isotherms of Ti/Zr oxides are presented
in Figure a. The isotherms
belonged to type IV isothermal curves,[49] indicating the presence of a porous structure, which was consistent
with the scanning electron microscopic (SEM) image (Figure a,b). Besides, the shift of
the hysteresis loop to lower the relative pressure in Ti/Zr oxides
with the content of zirconia increased, showing the gradual decrease
of pore diameters. The surface areas of Ti/Zr oxides were studied
using the Brunauer–Emmett–Teller (BET) method, as exhibited
in Table . It was
not difficult to find that the surface areas of Ti/Zr mixed oxides
were larger than those of TiO2 and ZrO2 and
much larger than the surface areas of commercial TiO2 and
ZrO2 (Tables and S1). Notably, Ti2Zr8had the highest BET surface area (389 m2/g), which
implied that the combination of Ti with Zr species could adjust the
structural network to generate porous surfaces.[50] Meanwhile, the pore size of Ti2Zr8 was well distributed and centered at around 6.3 nm, while the others
are presented in a relatively scattered form (Figure S1).
Figure 2
(a) N2 adsorption–desorption isotherms,
(b) pyridine-IR
spectra, (c) NH3-TPD curves, and (d) CO2-TPD
curves of Ti/Zr oxides with different Ti/Zr molar ratios.
Figure 3
SEM images of Ti2Zr8 microsphere
(a,b), and
TEM images of Ti2Zr8 (c) and Ti8Zr2 (d).
Table 1
Acid/Base
Content, Pore Size, and
BET Surface Area of Ti/Zr Mixed Oxides with Different Ti/Zr Molar
Ratios
catalyst
acidity (mmol/g)a
basicity (mmol/g)a
BET surface area (m2/g)b
pore size (nm)b
ZrO2
0.31
0.21
104
10.8
Ti2Zr8
1.12
0.46
389
6.7
Ti5Zr5
1.34
0.47
302
7.6
Ti8Zr2
0.57
0.34
294
12.4
TiO2
0.79
0.20
115
18.9
Acid and base contents were obtained
from NH3- and CO2-TPD, respectively.
BET surface area and pore size were
measured by N2 adsorption–desorption.
(a) N2 adsorption–desorption isotherms,
(b) pyridine-IR
spectra, (c) NH3-TPD curves, and (d) CO2-TPD
curves of Ti/Zr oxides with different Ti/Zr molar ratios.SEM images of Ti2Zr8 microsphere
(a,b), and
TEM images of Ti2Zr8 (c) and Ti8Zr2 (d).Acid and base contents were obtained
from NH3- and CO2-TPD, respectively.BET surface area and pore size were
measured by N2 adsorption–desorption.As shown in Figure c,d and Table , the
acid and base properties of Ti/Zr oxides were evaluated from NH3- and CO2-TPD, respectively. Both acid and base
contents of Ti/Zr oxides were initially increased and then decreased
with the introduction of Ti species into ZrO2, and the
highest acid and base contents (i.e., 1.34 and 0.47 mmol/g) were observed
for Ti5Zr5. The acid type and strength of Ti/Zroxides were measured by FT-IR spectra of pyridine adsorption (Figures b and S2). The spectra showed three bands at around
1450, 1490, and 1610 cm–1 that were assigned to
Lewis acid, and a small band at 1545 cm–1 was allocated
to Brønsted acid.[51,52] It should also be noted that
Brønsted and Lewis acidity increased with the introduction of
Ti species. Meanwhile, the ZrO2 calcined at 500 °C
was free of Brønsted acid and had the lowest Lewis acidity, while
Ti/Zr oxides remained similar in terms of Brønsted acidity and
TiO2 possessed the highest Lewis acidity.Furthermore,
morphological properties of Ti/Zr catalysts were studied
by SEM and transmission electron microscopy (TEM) (Figures and S2). It was found that Ti/Zr oxides were spherical and commercial TiO2 and ZrO2 were ruleless in shape (Figures a,b, and S2). The visibly porous structure of Ti/Zr oxides was confirmed
by SEM images, which well matched with that of N2 physisorption
isotherms (Figure a). Meanwhile, because of different amounts of Ti species, Ti8Zr2 existed in crystal shape while Ti2Zr8 was amorphous, as illustrated by TEM images (Figure c,d), which were
in agreement with the results of XRD patterns (Figure ).
Conversion of EL to GVL over Different Ti/Zr
Catalysts
As compared with commercial ZrO2 and
TiO2,
microspheric ZrO2 and TiO2 exhibited two times
higher GVL selectivity for the CTH of EL in 2-propanol at 180 °C
for 6 h (Table ),
which could be attributed to the larger surface area, acidity, and
basicity of microspheric Ti/Zr oxides (Tables and S1). In sharp
contrast, Ti/Zr microspheres, by introducing Ti species into ZrO2, exhibited predominant performance in the CTH of EL to GVL
(Table ). Meanwhile,
the synergistic effect of the surface area and Lewis acidity and basicity
have been demonstrated to have a promotion role in CTH reaction.[53] All of the microspheric Ti/Zr mixed oxides possessed
higher GVL yields (>72.9%) than microspheric ZrO2 (71.2%)
and TiO2 (9.4%), which could be ascribed to their improved
surface area and moderate Lewis acidity and basicity (Tables , 2,
and S2). First, Lewis sites can promote
2-propanol deprotonation and hydride transfer through generating surface
2-propoxide, and they play a coordinated role with base sites, in
combination with subsequent lactonization, to boost the production
of GVL from EL. Although ZrO2 has the highest yield per
m2, it has about 1/4 number of acid sites and 1/2 basic
sites of Ti2Zr8; thus, Lewis acid and base sites
play a collaborative role to boost the production of GVL from EL,
so the catalyst Ti2Zr8 gets more GVL yield.
The GVL yields obtained with Ti2Zr8 (90.1%)
and Ti5Zr5 (83.3%) were >10% higher than
that
with Ti8Zr2 (72.9%), implying that the higher
acid/base density and enhanced surface area of the catalysts (i.e.,
Ti5Zr5 and Ti2Zr8) were
beneficial for GVL formation (Table , Figures and 3c,d). The EL conversion (100%)
and average rate (37.6 μmol g–1 min–1) with Ti2Zr8 were higher than that with Ti5Zr5 (94% conversion and 31.1 μmol g–1 min–1 average rate), which was probably attributed
to the relatively improved surface area of Ti2Zr8 than that of Ti5Zr5 (Table ). In addition, the relatively higher Lewis
acidity in Ti2Zr8 seemed to be more beneficial
for producing GVL, as compared with Ti5Zr5 (Figure b, Table S2). As a result, Ti2Zr8 (with
appropriate acidity/basicity and enhanced surface area) possessed
the highest activity (100% EL conversion, 90.1% GVL yield) among the
different mole Ti/Zr catalysts. Meanwhile, Ti2Zr8 (90.1% GVL yield, 180 °C) showed better performance than Zr/B
(88.5% GVL yield, 200 °C), Al–Zr (83.2% GVL yield, 220
°C), and ZrO2 (81.5% GVL yield, 250 °C) under
the same reaction system.[31,32,54] Therefore, the Ti2Zr8 catalyst was chosen
for subsequent investigations.
Table 2
Results of EL-to-GVL
Conversion with
Different Ti/Zr Catalystsa
catalyst
EL conv (%)
GVL
yield (%)
GVL selec (%)
average rate (μmol g–1 min–1)b
commercial ZrO2
7.3
2.9
39.7
1.1
microspheric ZrO2
86.0
71.2
82.8
27.9
Ti2Zr8
100
90.1
90.1
37.6
Ti5Zr5
94.0
83.3
88.6
31.1
Ti8Zr2
87.4
72.9
83.4
28.3
microspheric
TiO2
16.2
9.4
58.0
3.5
commercial TiO2
4.5
1.1
24.4
0.4
Reaction conditions: 1 mmol EL,
0.072 g catalyst, 5 mL 2-PrOH, 180 °C, and 6 h.
Average rate is defined as GVL mol/(catalyst
weight × time).
Reaction conditions: 1 mmol EL,
0.072 g catalyst, 5 mL 2-PrOH, 180 °C, and 6 h.Average rate is defined as GVL mol/(catalyst
weight × time).
Effect of Reaction
Time and Temperature
The CTH reactions
of EL to GVL were further carried out over Ti2Zr8 at different reaction temperatures in the range of 160–200
°C with the reaction time of 1–10 h (Figure ). Gas chromatography (GC)–mass
spectrometry (MS) showed that isopropyl levulinate (IPL) was the primary
byproduct (20% yield) after reacting at 160 °C for 4 h (Figure S3), which could be promoted by acid and
base sites of Ti2Zr8 through transesterification.
A moderate GVL yield of 84% was obtained for a long duration (10 h)
at 160 °C (Figure a), and IPL (∼14% yield) without other byproducts was obtained.
It seemed that the generation of IPL (9%) could be suppressed at a
relatively higher temperature of 180 °C, giving higher yields
of GVL (∼90% after 6 h; Figure b). However, the IPL yield slightly decreased with
no visible increase in EL conversion and GVL yield after further increasing
the temperature to 200 °C for 6 h (Figures S4 and 4c). Hence, 180 °C and 6
h were chosen for the following measurements.
Figure 4
Catalytic production
of GVL from EL by varying reaction time at
temperatures of 160 (a), 180 (b), and 200 °C (c). Reaction conditions:
1 mmol EL, 0.072 g Ti2Zr8, and 5 mL 2-PrOH.
Catalytic production
of GVL from EL by varying reaction time at
temperatures of 160 (a), 180 (b), and 200 °C (c). Reaction conditions:
1 mmol EL, 0.072 g Ti2Zr8, and 5 mL 2-PrOH.
Effect of Different H-Donors
The influence of various
alcohols including methanol (MeOH), ethanol (EtOH), n-propanol (n-PrOH), 2-PrOH, n-butanol
(n-BuOH), and 2-butanol (2-BuOH) on the CTH of EL
to GVL was investigated. Compared with primary alcohols, the secondary
alcohols showed an outstanding performance for the CTH reaction because
of their lowest reduction potential (Figure ).[55] It was not
difficult to find that 2-BuOH (86.1%) and 2-PrOH (90.1%) afforded
high selectivity toward GVL at near-complete EL conversions. Although
a high EL conversion could be obtained by using methanol as the H-donor,
extremely poor GVL selectivity with the majority of methyl levulinate
formed via transesterification was observed. Owing to stronger steric
effects of n-propyl levulinate (PL) and n-butyl levulinate (BL),[39] the CTH of PL
in n-PrOH and BL in n-BuOH to GVL
was more difficult than EL-to-GVL conversion in EtOH, thus resulting
in inferior selectivity toward GVL in n-PrOH and n-BuOH. Meanwhile, it was clear that some other side reactions
took place in EtOH besides transesterification and MPV reaction.[39] In addition, 2-PrOH possessed the lowest reduction
potential, resulting in the best reduction capacity that ensures the
highest GVL yields. Therefore, 2-PrOH was the best H-donor in this
reaction system.
Figure 5
The effect of different alcohols as H-donors on the production
of GVL from EL. Reaction conditions: 1 mmol EL, 0.072 g Ti2Zr8, 5 mL alcohol, 180 °C, and 6 h.
The effect of different alcohols as H-donors on the production
of GVL from EL. Reaction conditions: 1 mmol EL, 0.072 g Ti2Zr8, 5 mL alcohol, 180 °C, and 6 h.
Effect of Catalyst Dosage and Substrate
Catalyst dosage
is one of the important parameters to measure the activity of the
reaction system among various factors. As shown in Figure , the effect of the mass ratio
of EL to the catalyst in the range of 1:1 to 8:1 on GVL production
was examined. When the EL/catalyst mass ratio increased from 1:1 to
2:1, GVL selectivity and EL conversion remained almost constant. However,
both EL conversion and GVL yield showed a remarkable reduction as
the mass ratio of EL/catalyst reached 4:1, and IPL with a yield of
about 10% was also formed. When the mass ratio of EL/catalyst further
increased to 6:1 and 8:1, both EL conversion and GVL yield revealed
continuous decrease, while the IPL yield remain unchanged (∼10%).
Hence, the subsequent experiments were set the mass ratio of EL and
catalyst at 2:1. Notably, LA was also used as a substrate under the above-mentioned
identical reaction conditions (Figure S3), and complete LA conversion with comparable yields of GVL (81.2%)
and IPL (11.6%) were achieved.
Figure 6
The effect of different catalyst dosage
on the production of GVL
from EL. Reaction conditions: 1 mmol EL, catal: Ti2Zr8, 5 mL 2-PrOH, 180 °C, 6 h.
The effect of different catalyst dosage
on the production of GVL
from EL. Reaction conditions: 1 mmol EL, catal: Ti2Zr8, 5 mL 2-PrOH, 180 °C, 6 h.
Recyclability of the Catalyst
The multiple reusability
and long-term stability of the catalyst are extraordinarily significant
considerations for the production of GVL from EL in order to save
the economic costs. In this recycling test, the reused catalyst was
separated from the reaction mixture by centrifugation, washed by methanol
and N,N-dimethylformamide (DMF),
dried for 2 h at 80 °C, and subsequently calcined for 2 h at
500 °C. The catalytic performance of regenerate catalyst for
the next cycle reaction was conducted under the same conditions. The
catalyst recycling study under a relatively low EL conversion was
also conducted with the decrease of the catalyst dosage to 0.018 g;
the results are presented in Figure . EL conversion and GVL yield declined by only ∼6%.
In addition, the reusability of the unregenerate Ti2Zr8 is shown in Figure S5b, wherein
EL conversion decreased only by 10% and GVL selectivity remained balanced
in six consecutive cycles, showing a little loss of the active sites
in the Ti2Zr8 catalyst. Figure S5a shows that there is no obvious difference of EL
conversion (∼100%) and GVL yield (decreasing only by 5%) after
six consecutive cycles, showing no loss of the active sites in the
Ti2Zr8 catalyst. XRD patterns of the fresh and
used Ti2Zr8 indicated no obvious structural
change (Figure S6), and the morphology
and porous structure of used Ti2Zr8 were not
alerted at all, as confirmed by SEM (Figure S7). No
obvious change in the acid (0.98 mmol/g) and base (0.45 mmol/g) density
of the used Ti2Zr8 was observed, compared with
those (1.08 and 0.46 mmol/g) of fresh Ti2Zr8 (Figure S8). The thermal stability of
the fresh and used Ti2Zr8 was verified by TG
analysis, which showed only 2.3% weight loss difference between them
(Figure S9). All of the above observations
demonstrate that the Ti2Zr8 catalyst is very
stable during the reaction and can be used for multiple recycles.
Figure 7
Recyclability
of regenerate Ti2Zr8 in the
production of GVL from EL. Reaction conditions: 1 mmol EL, 0.018 g
Ti2Zr8, 5 mL 2-PrOH, 180 °C, and 6 h.
Recyclability
of regenerate Ti2Zr8 in the
production of GVL from EL. Reaction conditions: 1 mmol EL, 0.018 g
Ti2Zr8, 5 mL 2-PrOH, 180 °C, and 6 h.
Reaction Mechanism
On the basis of above discussions
and previous reports,[28,54] two possible reaction pathways
(i.e., hydrogenation–cyclization and cyclization–reduction)
can be involved for producing GVL from EL.[56] In the present catalytic systems, no angelica lactone was detected
by GC–MS, and no GVL was obtained when angelica lactone was
directly used as a substrate. Therefore, it can be speculated that
the CTH of EL to GVL over Ti2Zr8 microspheres
proceeds through sequential transfer hydrogenation and intramolecular
transesterification (i.e., cyclization), as shown in Scheme . The formation of 2-propoxide
and the activation of carbonyl group of EL are promoted by both acid
and base sites (Zr4+–O2–) of the
Ti/Zr catalyst, resulting in the production of 4-hydroxypentanoate
(4-HPE) through a six-membered transition state, which can be further
transformed into GVL via a intramolecular transesterification process.
Scheme 1
Possible Reaction Mechanism for Producing GVL from EL over Ti/Zr
Microsphere
Conclusion
In
summary, the efficient synthesis of GVL from EL via MPV reduction
catalyzed by microspheric Ti/Zrporous oxides has been developed by
using 2-PrOH simultaneously as H-donor and solvent. Among various
Ti/Zr microspheres with different Ti/Zr molar ratios, Ti2Zr8 possessing the largest surface area (385 m2/g), and appropriate acidity (1.12 mmol/g) and basicity (0.46 mmol/g),
was more beneficial for the MPV reaction. A GVL yield of 90.1% at
an EL conversion of 100% was achieved at 180 °C for 6 h over
Ti2Zr8, which can be ascribed to moderate physico-chemical
properties. Moreover, the Ti2Zr8 catalyst showed
excellent stability and reusability in six consecutive cycles, with
a GVL yield of no less than 84.4% at complete EL conversion.
Experimental
Section
Materials
EL (99%), LA (99%), GVL (98%), titanium(IV)isopropoxide
(TiPP, 97%), zirconium(IV)propoxide (ZrPP, 70 wt % in 1-propanol),
hexadecylamine (HDA, 90%), sec-butyl alcohol (98%),
zirconium oxide (≥99.9%), and titanium oxide (≥99%)
were purchased from Aladdin Reagent Co. Ltd. (Shanghai, China). Absolute
ethanol (≥99.5%), 2-PrOH (≥99.5%), and potassium chloride
(KCl, ≥99.5%) were obtained from Chongqing Chuandong Chemical
Reagent Company. Other chemicals were of analytical grade and used
without further purification.
Preparation of Microspheric
Ti/Zr Porous Oxides
The
TiZr microspheres
[with Ti/Zr molar ratios of 2:8 (Ti2Zr8), 5:5
(Ti5Zr5), and 8:2 (Ti8Zr2)] were prepared through the sol–gel process combined with
solvothermal treatment using HAD as a structure-directing agent. In
a typical procedure for the synthesis of Ti2Zr8, 1.21 g of HDA was dissolved in 100 mL ethanol, followed by the
addition of 0.43 mL KCl aqueous solution (0.1 M) and 0.54 mL H2O. To the resultant solution, a mixture containing 10 mL of
ethanol, 0.57 g TiPP (2 mmol), and 2.62 g ZrPP (8 mmol) was added
under vigorous stirring at room temperature. The milky white suspension
was acquired and kept static for 18 h, followed by centrifugation
and washing by ethanol and dried in air at room temperature. The resultant
white solids (1.2 g) were further dispersed into a mixed solution
of 7.5 mL H2O and 15 mL ethanol and then subjected to a
solvothermal process in Teflon-lined autoclave for 16 h at 160 °C.
After suction filtration, the solvothermally treated
solids were immersed in ethanol, followed by desiccation at room temperature.
Finally, the obtained white powder was calcined at 500 °C at
a heating ramp of 2 °C/min for 2 h in air, affording the porous
Ti/Zr oxide microspheres by elimination of organic components. The
sole TiO2 and ZrO2 microspheres were synthesized
with 10 mmol TiPP and 10 mmol ZrPP, respectively, and the synthetic
procedures were similar to those of TiZr microspheres.
Catalysts Characterization
XRD patterns of the catalysts
were recorded on a Rigaku diffractometer with Cu Kα radiation
source (λ = 1.5406 Å), with 2θ from 5 to 80°.
Thermal gravimetric analysis was performed on a 1000 thermal analyzer
under a dynamic N2 atmosphere. SEM was performed on EHT
with accelerating voltage of 15 kV. TEM images were obtained from
FEI Tecnai G2 F20. BET surface areas of the catalysts were determined
by the nitrogen adsorption–desorption method using a Micromertics
ASAP2020 instrument. NH3 and CO2 temperature-programmed
desorption (TPD) was used to measure the acidic and basic properties
of the catalysts, respectively, and the profiles were obtained by
an AutoChem 2920 apparatus equipped with a thermal conductivity detector.
In a general procedure, the sample was preheated at 300 °C for
1 h in helium flow, and then cooled to 50 °C under helium atmosphere,
followed by exposure to NH3 or CO2 gases at
50 °C for 1 h with a flow rate of 30 mL/min and subsequent desorption
by heating from 50 to 500 °C. FT-IR spectra of pyridine adsorption
were measured by a Bruker Tensor 27 FT-IR spectrometer. Before pyridine
adsorption, the sample was heated to 250 °C and degassed until
the residual pressure was <0.05 Pa. Then pyridine was introduced
for complete adsorption, followed by degassing and heating at 110
°C to remove physisorbed pyridine, which was finally subjected
to desorption at 250 °C (heating rate: 5 °C/min) for 45
min and in situ recorded with FT-IR.
General Procedures for
Catalytic Conversion of EL to GVL
The CTH of EL to GVL was
conducted in a 25 mL stainless steel autoclave
containing a Teflon-lined reactor, equipped with a magnetic stirrer.
In a typical run, 1 mmol of EL, 0.02 g of naphthalene as an internal
standard, 5 mL alcohol as a solvent and hydrogendonor, and 0.0721
g catalyst were put into the vessel. The reaction was controlled in
a desired temperature and time. After the reaction, the reactor was
cooled by flowing water, and the liquid samples were filtered out
and analyzed by GC.
Sample Analysis
The concentrations
of EL, GVL, and
byproducts were analyzed by GC (Agilent 7890) equipped with FID detector
and HP-5 capillary column (30 m × 0.32 mm × 0.25 μm).
The carrier gas was nitrogen with a flow rate of 1.0 mL/min, the injection
volume of 1 μL, injector temperature of 250 °C, detector
temperature of 270 °C, and oven temperature programming from
60 °C at a rate of 10 °C/min to 230 °C and maintained
for 5 min. The identification of the liquid products was conducted
by GC (Agilent 6890)–MS (5973).