Reac Kinet Mech Cat DOI 10.1007/s11144-017-1194-y
Modified sepiolite supported nickel and tungsten oxide catalysts for glycerol hydrogenolysis to 1,2-propanediol with high selectivity under mild conditions Wei Long1 • Fang Hao1 • Wei Xiong1 • Pingle Liu1 He’an Luo1
•
Received: 11 February 2017 / Accepted: 8 May 2017 Ó Akade´miai Kiado´, Budapest, Hungary 2017
Abstract Modified sepiolite supported nickel and tungsten oxide catalysts were prepared for glycerol hydrogenolysis to 1,2-propanediol. The effects of nickel dispersion, the amount of WO3 and reaction conditions on catalytic performance were investigated. The Ni-7.5-WO3-1/MSEP catalyst gave the best result of 88.3% glycerol conversion and 96.8% selectivity to 1,2-propanediol under 2.0 MPa and 180 °C. A certain amount of tungsten oxide helps the nickel dispersion and improves the selectivity to 1,2-propanediol. Finally, a possible reaction mechanism was proposed. Keywords 1,2-propanediol Hydrogenolysis Glycerol Tungsten oxide
Introduction As an important chemical intermediate, 1,2-propanediol is widely used in the manufacture of polyester resins, liquid detergents and pharmaceuticals [1]. The current industrial process for the production of 1,2-propanediol is via the route of hydration of prop-2-enal [2]. Glycerol hydrogenolysis is a very important process to produce many products such as 1,2-propanediol and ethylene glycol (EG) [3–5]. Glycerol is an important platform chemical derived from plant sources. The transformation of glycerol into 1,2-propanediol could be carried out through biocatalysis, dehydration and hydrogenolysis processes [6–8]. The hydrogenation of
Electronic supplementary material The online version of this article (doi:10.1007/s11144-017-1194-y) contains supplementary material, which is available to authorized users. & Pingle Liu
[email protected] 1
School of Chemical Engineering, Xiangtan University, Xiangtan 411105, China
123
Reac Kinet Mech Cat
glycerol in the presence of metallic catalysts still rely severe conditions of high temperature and pressure. Hao et al. obtained the results of 90.1% glycerol conversion and 89.7% selectivity to 1,2-propanediol over Cu/Al2O3 with H4SiW12O40 acting as acidic promoter under 240 C and 3.5 MPa [9]. It has been reported that the conversion of glycerol and the selectivity to 1,2-propanediol over Cu/DUSY were affected by the dispersion and the amount of Cu particles [10]. Valencia et al. pointed out that glycerol can be effectively converted to 1,2-propanediol over this catalyst (atomic ratio Cu: Al = 3:1), and the highest glycerol conversion and 1,2propanediol selectivity were 74.26 and 78.84% at 220 °C, respectively [11]. Lee et al. reported the hydrogenolysis of 20 wt% glycerol aqueous solution under 180 °C and 2.5 MPa over 5 wt% Ru supported on hydrotalcite modified with Ca and Zn additives, it gave 85.5% selectivity to 1,2-propanediol at 58.5% conversion of glycerol [12]. Nickel based catalysts are tried in this process [13–15]. The nonnoble metal Ni/NaX catalyst was prepared and used in hydrogenolysis of aqueous glycerol, the conversion of glycerol was 86.6% and the selectivity to 1,2propanediol and ethylene glycol was 94.6% under 6.0 MPa and 200 °C for 10 h [14]. Some researchers reported vapor phase hydrogenolysis of glycerol into 1,2propanediol over a chromium-free Ni-Cu-SiO2 catalyst [16–19]. Sepiolite (Mg8Si12O30(OH)4 (H2O)4.8H2O) can be used as a catalyst support due to its fibrous structure and open porous network [20–22]. Sepiolite supported nickel and potassium bimetallic catalysts presented good catalytic performance in nitrile compound hydrogenation reaction [23]. Tungsten oxide showed great incorporation function in the catalysts through increasing the surface acid amounts [24]. Liu et al. found that the highly dispersed WO3 acted as both chemical and structural promoters by increasing a number of Lewis acid sites [25]. Cecilia et al. pointed out that the catalyst was enhanced when WO3 was supported for glycerol dehydration [26]. WO3 was good promoter and was certified by many researchers [27, 28]. Some researchers studied the reaction mechanism of glycerol hydrogenolysis. Montassier et al. proposed that 2,3-dihydroxypropanal was obtained from dehydrogenation of C–O [29]. Dasari et al. held that 1-hydroxy-2-acetone was the key intermediate over copper-chromite catalyst [30]. Maglinao et al. reported that water promoted glycerol hydrogenolysis [31]. Lan et al. found that the amount of acid sites on the catalyst had influence on the intra-molecular dehydration of _ glycerol [32, 33]. Recently, Zelazny et al. proposed that 2,3-dihydroxypropanal was the important intermediate, then 2-hydroxy-acrolein was formed through dehydration process [34]. Sepiolite supported nickel and tungsten oxide catalysts were prepared and applied in glycerol hydrogenolysis under mild conditions of lower temperature and pressure (Scheme 1). The influences of WO3 promoter on the physicochemical property of the catalyst and catalytic performance were carried out, and the effects of reaction conditions were discussed.
123
Reac Kinet Mech Cat OH HO
OH OH
H2
HO
+
HO
OH
+
Others*
Catalyst 1,2-propanediol
Ethyleneglycol
*Others: methanol, ethanol, 1-propanol, 2-propanol. Scheme 1 The reaction flow chart of glycerol hydrogenolysis reaction
Experimental Materials Sepiolite (silicon dioxide, 42.21%; magnesium oxide, 20.57%; calcium oxide, 18.83%; aluminum oxide, 0.37%; iron oxide, 0.21%; potassium oxide, 0.18%; sodium oxide, 0.08% and manganese dioxide, 0.05%) was purchased from Hunan XiangTan Yuanyuan Sepiolite Limited Company. Analytical grade hydrogen peroxide (30 wt%), Ni(NO3)26H2O, sodium hydroxide, ammonium metatungstate (NH4)10H2(W2O7)6, hydrochloric acid, 1,4-butanediol and glycerol were purchased from Sinopharm Chemical Reagent Corporation Limited. Hydrogen gas (99.99%) was purchased from Zhuzhou Diamond Gas Company. Catalyst preparation Natural sepiolite (SEP) was purified and micronized by wet process. Firstly, the sepiolite was dried for 12 h at 110 °C, and the solids were sieved and placed in 10.0 wt% hydrochloric acid with the weight ratio of 1:20. Then, the mixture was refluxed at 95 °C for 6 h, and the sepiolite was washed by distilled water till the pH of the washed water becomes neutral (measured by acidometer). Then the sepiolite was subsequently calcined in air at 350 °C for 4 h at a heating rate of 2 °C min-1. The modified sepiolite was labled as MSEP. Ni/MSEP catalysts were prepared by wet impregnation method. MSEP was impregnated in nickel nitrate (Ni(NO3)26H2O) aqueous solution for 6 h. The mixture was dried at 110 °C overnight and subsequently calcined in air at 350 °C for 4 h at a heating rate of 2 °C min-1. Then it was reduced under a stream of pure H2 (100 mL/min) at 400 °C for 1 h. The letter x in Ni-x/MSEP represents the mass percent of nickel metal in the catalyst. Ni-WO3/MSEP catalysts were prepared by mixing and co-impregnation method. A certain amount of ammonium metatungstate (NH4)10H2(W2O7)6 was dissolved in 30 mL distilled water at 20 °C, then 3.0 mL hydrogen peroxide was added to promote the dissolution of ammonium metatungstate. At the same time, the predetermined amount of nickel nitrate (Ni(NO3)26H2O) was added. After stirring for 2 h, The MSEP was impregnated in the above mixture for 6 h. Then the mixture was dried at 110 °C overnight and subsequently calcined in air at 350 °C for 4 h at a heating rate of 2 °C min-1. It was also reduced under a stream of pure H2 (100 mL/
123
Reac Kinet Mech Cat
min) at 400 °C for 1 h. The letter z and y in Ni-z-WO3-y/MSEP represents the molar ratio of Ni and WO3 species in the catalyst. The mass ratio of nickel metal is 20.0% in Ni-z-WO3-y/MSEP regardless of the addition of tungsten oxide. Since the amount of tungsten oxide is very small in Ni-z-WO3-y/MSEP, we have chosen the molar ratio of Ni and WO3 to express the content of WO3 in these catalysts. Procedures for catalytic test Catalytic tests were carried out in a 20 mL Teflon-lined stainless steel autoclave with a magnetic stirrer at 400 rpm. Typically, 0.3 g catalyst powder (200–400 lm) was transferred into the autoclave. The reactant solution (20 mL of 20 wt% glycerol aqueous solution) and a certain amount of sodium hydroxide were added. The reactor was sealed and purged with H2 to exclude air for five times, and then it was pressurized to 2.00 MPa with H2 under vigorous stirring after the required temperature reached. After the reaction, the catalysts were separated by filtration from the liquid phase products. The gas phase products were collected in a gas bag. The contents of the reactants and the liquid phase products were determined by gas chromatography (Agilent Technologies, 7890 A) equipped with a DB-Wax capillary column (diameter 0.50 mm, length 30 m) and a flame ionization detector (FID) using 1,4butanediol as the internal standard. The liquid phase products include 1,2propanediol, ethylene glycol, 1-propanol, 2-propanol, ethanol and methanol. The gas phase products were analyzed by gas chromatography, and the major component was methane. The conversion of glycerol and the selectivity to the products was calculated based on the following equations [35–40]. Conversion ð%Þ ¼
Moles of glycerol ðinÞ Moles of glycerol ðoutÞ Moles of glycerol ðinÞ
Selectivityð%Þ ¼ P
C based mol of the product C based mol of all liquid products
Catalyst characterization The specific surface area, pore volume and pore size distribution of the samples were obtained by the nitrogen adsorption–desorption on a Quantachrome NOVA2200e automated gas sorption system. Specific surface areas and pore size distributions were calculated by Brunauer–Emmett–Teller (BET) and Barrett– Joyner–Halenda (BJH) methods. Fourier transform infrared (FT-IR) spectra of the samples were recorded on a Nicolet 380 spectrometer. The spectra of the samples were acquired in the wave number range of 400–4000 cm-1. TG/DTG curves were recorded by a TGA/DCS using air as purge gas (40 mL min-1) over a temperature range of 30–900 °C with a heating rate of 10 °C min-1.
123
Reac Kinet Mech Cat
Hydrogen chemisorption was measured by using Quantachrome ChemBET 3000 instrument. The samples were previously reduced in hydrogen stream and cooled to ambient temperature under the nitrogen stream. The hydrogen chemisorption was performed at 323 K, and the hydrogen pulses (0.02 mL) were injected until the eluted areas of consecutive pulsed became constant. Powder X-ray diffraction (XRD) patterns were determined under a D/max 2500 ˚ ) The tube voltage was TC diffractometer using Cu Ka radiation (k = 1.542 A 40 kv, the current was 30 mA, and the scan range was 5–90° with a scanning rate of 1° min-1. Temperature programmed reduction was carried out in a Quantachrome ChemBET-3000 instrument equipped with a procedure temperature-controlled furnace and a thermal conductivity detector. The sample was first heated at 473 K in argon flow for 2 h. Then, it was heated under 40 mL min-1 and 5 vol.% of H2/Ar flow from 473 to 1173 K at a rate of 5 K min-1. The pyridine adsorbed FT-IR spectra of the samples were recorded on a Nicolet iSTM10 spectrometer. The morphologies of the samples were observed with SEM on a JEOL JSM-6610 LV scanning microscope operating at an accelerating voltage of 5 kV. The microstructure of the samples was observed by transmission electron microscopy (TEM) on a Tecnai G220 ST electron microscope working at less than 200 kV. The instrumental magnification ranged from 2 9 104 to 10 9 106. The samples were deposited on a copper grid and coated with a holey carbon film.
Results and discussion Characterization of catalyst It can be seen from Table 1 that the BET surface area (SBET) increases from 95.6 to 226.7 m2 g-1 after hydrochloric acid treatment at 95 °C, and the pore volume (Vl) also increases. The BET surface area decreases with the increment of the amount of nickel. The surface area and the pore volume of Ni-WO3-MSEP catalysts increase first as the molar ratio of Ni and WO3 decreases to 7.5:1, and then decrease to 80.9 m2 g-1 and 0.193 mL g-1 as the molar ratio of Ni and WO3 decreases to 1:2.5. It can be seen from Fig. S1 that MSEP shows typical type I isotherms with sharp capillary condensation steps at a relative pressure (p/p0) of 0.8–1.0, but the sharp capillary condensation step moves to the relative pressure (p/p0) of 0.5–0.8 as nickel atom was loaded on the support. SEP and MSEP present almost the same pore size distribution, but it changes a little after the nickel is loaded on the support. The sharp capillary condensation steps moves to 0.8–1.0 when tungsten oxide was added, and the average pore size of the samples are concentrated at 2.0–4.0 nm. The FT-IR spectra of the samples are shown in Fig. 1. The peaks of 719 and 1023 cm-1 in all the samples are the bending and stretching vibrations of Si–O, the
123
Reac Kinet Mech Cat Table 1 Textural properties of various catalysts Loading amount of Ni (wt%)
Ni/WO3 molar ratio
SEP
–
–
95.6
0.35
MSEP
–
–
226.7
0.51
Ni-5/MSEP
5
–
210.3
0.43
Ni-10/MSEP
10
–
200.8
0.34
Ni-20/MSEP
20
–
139.6
0.31
Ni-15-WO3-1/MSEP
20
15:1
146.7
0.37
Ni-10-WO3-1/MSEP
20
10:1
165.6
0.48
Ni-7.5-WO3-1/MSEP
20
7.5:1
176.4
0.58
147.8
0.54
94.9
0.35
Ni-5-WO3-1/MSEP
20
5:1
Ni-2.5-WO3-1/MSEP
20
2.5:1
SBET (m2 g-1)
Vl (mL g-1)
Catalyst
Ni-1-WO3-1/MSEP
20
1:1
83.3
0.27
Ni-1-WO3-2.5/MSEP
20
1:2.5
80.9
0.19
(c)
Intensity(a.u.)
(b)
(a)
3375 3580
1610 660 719 1023 1000
1500
2000
2500
3000
3500
wavenumbers(cm-1) Fig. 1 FT-IR spectra of catalysts: SEP (a), MSEP (b), Ni-7.5-WO3-1/MSEP (c)
peak of 719 cm-1 is internal geminal Si–O bond, and the peak of 1023 cm-1 is Si– O–Si bond. The peak of 660 and 1610 cm-1 are attributed to the bending and stretching vibrations of O–H bond. There are two peaks of 3375 and 3580 cm-1 in SEP, which are the bending and stretching vibrations of Mg–OH and Al–OH bonds, respectively. Fig. S2 shows the TG-DTG curves of the samples, the weight loss for SEP below 150 °C is due to the evaporation of the water molecules adsorbed on the surface.
123
Reac Kinet Mech Cat
MSEP is stable and the weight loss percentage is about 6.0% below 100 °C and the TG curve is flat between 130 and 400 °C. It can be seen from Fig. S2c and S2d that the catalysts do not decompose at high temperature. The XRD spectra of the samples in Fig. 2 exhibit characteristic diffraction peaks of SEP at 2h = 8.8°, 26.6°, 50.1° and 68.3 °C [20, 21]. The XRD pattern of Fig. 2a shows the characteristic diffraction peaks of NiO at 2h = 37.2°, 43.0° and 63.1°. The characteristic diffraction peaks at 2h = 44.5°, 51.8°C and 76.4° correspond to (111), (200) and (220) crystalline planes of nickel [14, 15]. In contrast to Fig. 2a, the other catalysts do not have any diffraction peak of NiO. The characteristic diffraction peaks of WO3 at 2h = 22.0° and 36.8° can be seen in Figs. 2b–2f. It is clearly that nickel particles disperse better with the increment of WO3, it means that WO3 is helpful to the dispersion of nickel atom, this is well in accordance with the TEM images in Fig. 5. The hydrogen chemisorption data of the catalysts are summarized in Table 2. It can be seen from Table 2 that Ni-7.5-WO3-1/MSEP demonstrates better dispersion, larger hydrogen uptake quantity and larger metal surface area. Also, the nickel dispersion can be improved effectively with the increment of WO3, but too much WO3 may plug in the channel and reduce the contact area between metal and hydrogen, this is consistent with the characterization results of BET. The TPR profiles of NiO/MSEP and NiO-WO3/MSEP catalysts are given in Fig. 3. The TPR curves show that the initial reduction temperature (TRi) of NiO/ MSEP is 580 °C, which is higher than for other samples. When tungsten oxide was added, the initial reduction temperature (TRi) is about 340 °C, the addition of tungsten oxide is helpful to the nickel oxide reduction.
Fig. 2 XRD patterns of NiO/MSEP (a), Ni-10-WO3-1/MSEP (b), Ni-7.5-WO3-1/MSEP (c), Ni-5-WO31/MSEP (d), Ni-2.5-WO3-1/MSEP (e) and Ni-1-WO3-2.5/MSEP (f)
123
Reac Kinet Mech Cat Table 2 Hydrogen chemisorption data of some different catalysts H2 uptake (lmol g-1)
Catalysts
Metal surface area (m2 g-1)
Dispersion (%)
Ni-20/MSEP
185.7
123.0
10.9
Ni-10-WO3-1/MSEP
204.3
135.3
12.1
Ni-7.5-WO3-1/MSEP
223.9
142.9
13.2
Ni-5-WO3-1/MSEP
210.4
140.3
12.4
Ni-2.5-WO3-1/MSEP
189.7
131.8
11.2
Hyrogen Consumption(a.u.)
e d 343
c
336 b 340 580 345
300
a
400
500
600
700
800
Temperature/°C Fig. 3 TPR profiles of NiO/MSEP (a), NiO-15-WO3-1/MSEP (b),NiO-10-WO3-1/MSEP (c), NiO-7.5WO3-1/MSEP (d) and NiO-5-WO3-1/MSEP (e)
The pyridine adsorbed FT-IR spectra of the samples are shown in Fig. 4. The IR band at 1450 cm-1 is the adsorption of pyridine on Lewis acidic centers, the band at 1490 cm-1 is the interaction of pyridine with Lewis and Brønsted acid sites, and the band at 1540 cm-1 is the adsorption of pyridine on Brønsted acid centers. The results of the pyridine FT-IR spectra indicate that MSEP possesses both Lewis and Brønsted acid sites, and WO3 can increase the Brønsted acid sites, which may have great influence on the catalytic performance. The scanning electron microscope (SEM) images of MSEP with different magnifications are shown in Fig. S3. The typical homogeneous and spine-like structure are observed in Figs. S3a and S3b, and the obvious homogeneous distribution and tree-root fibrous structure can be seen in Figs. S3c and S3d. The TEM images are shown in Fig. 5. It can be seen from Fig. 5 that many uniform flakes form a lot of hollow fabric channels. Both WO3 particles and Ni atoms are dispersed homogeneously, and the WO3 particles are dispersed at the outer layer of fabric surface of the support. Nickel atoms are evenly dispersed at the inner channel of the support.
123
Reac Kinet Mech Cat
1450
1490
1540
a
b c
1400
1500
1600
1700
Wavenumber(cm-1) Fig. 4 Pyridine adsorbed FT-IR spectra of samples: MSEP (a), Ni-7.5-WO3-1/MSEP (b) and Ni-1-WO31/MSEP (c)
a 200 nm
b 100 nm
c 100 nm
d 100 nm
e 50 nm
f 50 nm
Fig. 5 TEM images of MSEP (a, b) and Ni-7.5-WO3-1/MSEP (c, d, e, f)
Catalytic performance Effects of the loading amount of Ni and WO3 Effects of various catalysts on the conversion of glycerol and the selectivity to the products were examined, and the results are summarized in Table 3. The primary
123
Reac Kinet Mech Cat Table 3 Effects of loading amount of Ni and WO3 Catalyst
The molar ratio of Ni and WO3
Conversion (%)
Selectivity (%) 1,2-propanediol
Ethylene glycol
Others
Ni-5%/MSEP
–
52.2
78.1
15.6
6.3
Ni-10%/MSEP
–
74.4
76.4
16.3
7.3
Ni-20%/MSEP
–
82.1
72.5
18.5
9.0
Ni-15-WO3-1/MSEPa
15:1
78.2
82.3
9.5
8.2
Ni-10-WO3-1/MSEPa
10:1
79.1
90.3
5.3
4.4
Ni-7.5-WO3-1/MSEPa
7.5:1
87.0
96.7
3.2
0.1
Ni-5-WO3-1/MSEPa
5.0:1
59.0
93.2
4.7
2.1
Ni-2.5-WO3-1/MSEPa
2.5:1
58.3
90.7
6.6
2.7
Ni-1-WO3-1/MSEPa
1:1
53.7
88.4
8.3
3.3
Ni-1-WO3-2.5/MSEPa
1:2.5
53.3
87.2
8.3
4.5
Reaction conditions 0.3 g catalyst, 20 mL glycerol aqueous solution (20 wt%), 0.64 g NaOH and 2.0 MPa H2 at 180 °C for 6.0 h a
The mass amount of nickel metal in various catalysts is 20.0%
product is 1,2-propanediol over all the different catalysts. As to Ni/MSEP, the conversion of glycerol increases significantly with the increment of nickel loading amount, but the selectivity to 1,2-propanediol decreases a little. Clearly, the active sites on Ni/MSEP increase as the rise of nickel loading amount. Ni-WO3/MSEP catalysts show better catalytic performance. It can be seen from Table 3 that the addition of WO3 is in favor of the production of 1,2-propanediol. The addition of a certain amount of WO3 is helpful to the nickel dispersion, which is beneficial to the hydrogenation process. The interaction between nickel and WO3 may prevent the cracking of C–C bond, thus the selectivity to the by-product of ethylene glycol decreases obviously, and the selectivity to 1,2-propanediol increases a lot. The conversion of glycerol increases to 87.0% and the selectivity to 1,2-propanediol is up to 96.7% as the molar ratio of nickel to WO3 reaches to 7.5:1. It can be seen from the characterization results that Ni-7.5-WO3-1/MSEP presents better nickel dispersion, larger hydrogen uptake quantity and larger metal surface area. However, both the conversion and the selectivity decrease as the added amount of WO3 continues to increase. It can be seen from Table 1 that the BET surface area decreases when too much WO3 is added, too much tungsten oxide may block the pores of the support. It can be seen from Fig. 3 that the reduction temperature rises when too much tungsten oxide is added. Hence, the addition of excessive amount of WO3 has negative effects on the active sites of nickel, and the conversion of glycerol has a precipitous decline. Table 4 shows the data from the literature of the catalytic performance over different metal catalysts. We can find that Cu/TiO2 and Cu/Al2O3 give better results of 40–52% glycerol conversion and 85–92% selectivity to 1,2-propanediol. Obviously, either from the results of the literature or the results of our study, the addition of tungsten oxide in Ni/MSEP is better than other transition metals. It has been confirmed that the addition of tungsten oxide promoter in the catalyst may help to the dispersion of active metal and increase the acid sites
123
Reac Kinet Mech Cat
[36, 37, 39, 41]. In this study, the catalyst characterization results show that the addition of tungsten oxide increases the Brønsted acid sites and improve the nickel dispersion. Compared with Ni/MSEP, Ni-7.5-WO3-1/MSEP gives much higher conversion of glycerol and selectivity to 1,2-propanediol. The reason may be better nickel dispersion and the increment of the Brønsted acid sites. Effects of reaction temperature Fig. 6 shows the effects of reaction temperature on glycerol hydrogenolysis over Ni7.5-WO3-1/MSEP. It is clearly found that reaction temperature has great effect on catalytic performance. The conversion of glycerol increases rapidly from 53.0 to 87.0% when the reaction temperature rises from 120 to 180 °C, and the selectivity to 1,2-propanediol increases from 89.5 to 96.7%. As the reaction temperature continues to rise, the conversion of glycerol will increase, but the selectivity to 1,2propanediol begins to decrease when the reaction temperature is higher than 180 °C. At the same time, higher reaction temperature may promote the cracking of C–C bond, which cause the increment of ethylene glycol. Effects of H2 pressure The effects of H2 pressure on glycerol hydrogenolysis are shown in Fig. 7. The conversion of glycerol increases sharply when H2 pressure rises from 1 to 2 MPa and it increases slowly when H2 pressure is greater than 2 MPa. The selectivity to 1,2-propanediol also presents the highest value at the H2 pressure of 2 MPa, excessively high H2 pressure is in favor of the production of ethylene glycol and other by-products. Table 4 Conversion of glycerol and selectivity to 1,2-propanediol over various metal catalysts Catalyst
Conversion (%)
Selectivity 1,2-propanediol (%)
References [30]
5% Ru/Al2O3
23.1
59.7
Ru/Al2O3?HZSM5(25)
6.4
36.0
[33]
5% Pd/C
5.0
72.0
[30] [30]
5% Pt/C
34.6
82.7
4.5% Ru-4.5% Re/SiO2
51.7
25.3
[32]
Raney Co
48.9
69.1
[30]
2Cu/TiOa2
52.0
92.0
[34]
4Cu/Al2Oa3
40.0
85.0
[34]
Raney Ni
49.5
52.7
[30]
Ni/SiO2- Al2O3
45.1
29.1
[30]
a
2Cu/TiO2 and 4Cu/Al2O3 is function as 2 and 4 mmol metal/1 g of support
123
Reac Kinet Mech Cat
Conersion or selectivity(%)
100
80
60 Conversion of glycerol Selectivity of 1,2-propanediol Selectivity of ethylene glycol Selectivity of others
40
20
0 120
140
160
180
200
220
Reaction temperature(°C) Fig. 6 Effect of reaction temperature. Reaction conditions 0.3 g Ni-7.5-WO3-1/MSEP, 20 mL of glycerol, aqueous solution (20 wt%), 0.64 g NaOH and 2.0 MPa H2 for 6.0 h
Effects of the amount of sodium hydroxide The effects of the amount of sodium hydroxide are shown in Table 5. It has been found that the addition of sodium hydroxide has great influence on glycerol hydrogenolysis. Glycerol almost does not transform without the addition of sodium hydroxide. As it can be seen from the subsequent possible reaction path, sodium hydroxide plays an important role in the formation of the intermediates. The conversion of glycerol increases sharply with the increment of sodium hydroxide. It can be seen from Table 5 that the conversion of glycerol and the selectivity to 1,2-
Coversion or Selectivtiy(%)
100
80 60 Conversion of glycerol Selectivity of 1,2-propanediol Selectivity of ethylene glycol Selectivity of others
40
20 0 1
2
3
4
Pressure(MPa) Fig. 7 Effects of H2 pressure. Reaction conditions 0.3 g Ni-7.5-WO3-1/MSEP, 20 mL of glycerol, aqueous solution (20 wt%), 0.64 g NaOH and temperature (180 °C) for 6.0 h
123
Reac Kinet Mech Cat
propanediol reaches the highest value when the amount of sodium hydroxide rises to 0.64 g. Effect of the amount of catalyst Table 6 shows the effects of the amount of catalyst. It can be seen that the conversion of glycerol increase rapidly from 35.5 to 88.3% as the amount of catalyst rises from 0.1 to 0.4 g. The conversion of glycerol and the selectivity to 1,2propanediol do not change much when the amount of the catalyst continues to increase. Hence, the suitable amount of the catalyst is about 0.3–0.4 g for 20 mL 20 wt% glycerol aqueous solution. Effect of reaction time Table 7 summarizes the effects of reaction time on glycerol hydrogenolysis. It can be seen that the conversion of glycerol gradually increases with the prolonged reaction time, while the selectivity to 1,2-propanediol does not change much. Moreover, the conversion of glycerol changes a little when the reaction time is greater than 6 h. Hence, the suitable reaction time for glycerol hydrogenolysis over Ni-7.5-WO3-1/MSEP under 180 °C and 2 MPa is 6 h. The possible reaction path On the basis of the results in this work, the possible reaction path for glycerol hydrogenolysis over Ni-WO3/MSEP catalyst was proposed in Scheme 2. The process of glycerol hydrogenolysis is complex, 1,2-propanediol and ethylene glycol exist in this reaction [5, 42]. Acrolein and acetol may be the important intermediates due to the acid sites on the catalyst [43]. In this work, we have found that dehydrogenation, dehydration, decarbonylation and hydrogenolysis processes exist Table 5 Effects of the amount of sodium hydroxide Amount of NaOH (g)
Conversion (%)
Selectivity (%) 1,2-propanediol
Ethylene glycol
Others
0.08
32.1
91.0
7.5
1.5
0.22
58.2
92.1
6.4
1.5
0.36
68.0
93.5
4.9
1.6
0.50
72.4
94.5
4.2
1.3
0.64
87.0
96.7
3.2
0.1
0.78
87.3
96.7
3.2
0.1
Reaction conditions 0.3 g Ni-7.5-WO3-1/MSEP, 20 mL of glycerol aqueous solution (20 wt%), 2.0 MPa H2 and temperature(180 °C) for 6.0 h
123
Reac Kinet Mech Cat Table 6 Effects of the amount of catalyst Amount of catalyst (g)
Conversion (%)
Selectivity (%) 1,2-propanediol
Ethylene glycol
Others
0.10
35.5
94.3
5.3
0.4
0.20
52.2
95.5
4.3
0.2
0.30
87.0
96.7
3.2
0.1
0.40
88.3
96.8
3.0
0.2
Reaction conditions Ni-7.5-WO3-1/MSEP, 20 mL of glycerol aqueous solution (20 wt%), 0.64 g NaOH, 2.0 MPa H2 and temperature(180 °C) for 6.0 h
Table 7 Effects of the reaction time Reaction time (h)
Conversion (%)
Selectivity (%) 1,2-propanediol
Ethylene glycol
Others
2.0
31.0
95.4
3.0
1.6
4.0
73.2
96.1
3.1
0.8
6.0
87.0
96.7
3.2
0.1
8.0
87.3
96.7
3.1
0.2
12.0
88.0
96.3
3.5
0.2
Reaction conditions 0.3 g Ni-7.5-WO3-1/MSEP, 20 mL of glycerol aqueous solution (20 wt%), 0.64 g NaOH, 2.0 MPa H2 and temperature (180 °C)
Scheme 2 Possible reaction path of glycerol hydrogenolysis
in this reaction, which is in accordance with the literature [5, 43]. Glycerol can be transformed to the intermediates such as glyceraldehyde and dihydroxyacetone firstly, and the intermediates hydrogenate to 1,2-propanediol or ethylene glycol under the reaction conditions. Glycerol can also lose a water molecule and then hydrogenate to form other by-products such as 1-propanel and 2-propanel which has been testified by GC–MS.
123
Reac Kinet Mech Cat
Conclusions In conclusion, MSEP supported nickel and tungsten oxide catalyst Ni-WO3/MSEP was prepared and characterized. The catalyst presents good catalytic performance in glycerol hydrogenolysis to 1,2-propanediol under mild conditions. The addition of a certain amount of WO3 is in favor of the nickel dispersion and may prevent the cracking of C–C bond and thus improve the selectivity to 1,2-propanediol. Ni-7.5WO3-1/MSEP gives the best catalytic performance of 88.3% conversion of glycerol and the selectivity to 1,2-propanediol is up to 96.8% under 180 °C and 2 MPa. The possible reaction path is proposed based on GC–MS, FT-IR and work of other researchers. Acknowledgements This work was supported by NSFC(U1662127), Project of Hunan province Science and Technology Department (2015GK1060) and Project of Xiangtan University(2015SEP04, 2015SEP05), and Collaborative Innovation Center of New Chemical Technologies for Environmental Benignity and Efficient Resource Utilization.
References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.
Besson M, Gallezot P, Pinel C (2014) Chem Rev 114:1827–1870 Huber GW, Jporra S, Corma A (2006) Chem Rev 106:4044–4098 Mehdipour P, Omrani A, Rostami H, Rostami AA (2016) J Mol Liq 219:165–172 Carric¸o CS, Cruz FT, Santos MB, Oliveira DS, Pastore HO, Andrade HMC, Mascarenhas AJS (2016) J Catal 334:34–41 Wang SH, Yin K, Zhang YC, Liu HC (2013) ACS Catal 3:2112–2121 Hong E, Kim D, Kim J, Yoon S, Rhie S, Ha S, Ryu Y (2015) Biomass Bioenergy 77:177–186 Lauriol-Garbey P, Loridant S, Bellie`re-Baca V, Rey P, Millet JMM (2011) Catal Commun 16:170–174 Hirasawa S, Watanabe H, Kizuka T, Nakagawa Y, Tomishige K (2013) J Catal 300:205–216 Hao SL, Peng WC, Zhao N, Xiao FK, Wei W, Sun YH (2010) J Chem Technol Biot 85:1499–1503 Niu L, Wei RP, Jiang F, Zhou MH, Liu CY, Xiao GM (2014) React Kinet Mech Cat 113:543–556 Valencia R, Tirado JA, Sotelo R, Trejo F, Lartundo L (2015) React Kinet Mech Cat 116:205–222 Lee SH, Moon DJ (2011) Catal Today 174:10–16 Yu WQ, Xu J, Ma H, Chen C, Zhao J, Miao H, Song Q (2010) Catal Commun 11:493–497 Zhao J, Yu WQ, Che C, Miao H, Ma H, Xu J (2010) Catal Lett 134:184–189 Perosa A, Tundo P (2005) Ind Eng Chem Res 44:8535–8537 Nie Y, Shang S, Xu X, Hua W, Yue Y, Gao Z (2012) Appl Catal A 433:69–74 Hernandez-Pichardo ML, de Montoya la Fuente JA, Del Angel P, Vargas A, Navarrete J, Hernandez I, Lartundo L, Gonza´lez-Brambila M (2012) Appl Catal A Gen 431:69–78 Taylor MN, Zhou W, Garcia T, Solsona B, Carley AF, Kiely CJ, Taylor SH (2012) J Catal 285:103–114 Lee M, Hwang YK, Chang JS, Chae HJ, Hwang DW (2016) Catal Commun 84:5–10 Alkan M, Tekin G, Namli H (2005) Microp Mesopor Mat 84:75–83 ´ vila P, Banu´s ED, Miro´ EE, Milt VE (2015) Appl Catal A 498:41–53 Tuler FE, Portela R, A Yang SB, Ding CC, Cheng WC, Jin ZX, Sun YB (2015) J Mol Liq 204:170–175 Lv Y, Hao F, Liu PL, Xiong SF, Luo HA (2017) J Mol Catal A: Chem 426:15–23 Zhang YH, Zhao XC, Wang Y, Zhou LK, Zhang JY, Wang J, Wang AQ, Zhang T (2013) J Mater Chem A 1:3724–3732 Liu ZM, Su H, Chen B, Li JH, Woo SI (2016) Chem Eng J 299:255–262 Cecilia JA, Garcia-Sancho C, Me´rida-Robles JM, Gonza´lez JS, Moreno-Tost R, Mairles-Torres P (2016) Appl Catal A Gen 516:30–40 Dalil M, Carnevali D, Dubois JL, Patience GS (2015) Chem Eng J 270:557–563
123
Reac Kinet Mech Cat 28. Liebig C, Paul S, Katryniok B, Guillon C, Couturier JL, Dubois JL, Dumeignil F, Hoelderich WF (2014) Appl Catal B Environ 148–149: 604–605 29. Montassier C, Giraud D, Barbier J (1988) Heter Catal Fine Chem 165–168 30. Dasari MA, Kiatsimkul PP, Sutterlin WR, Suppes GJ (2005) Appl Catal A 281:225–231 31. Maglinao RL, He BB (2011) Ind Eng Chem Res 50:6028–6033 32. Lan MA, Li YM, He DH (2011) Chin J Catal 32:872–876 33. Li YM, Ma L, Liu HM, He DH (2014) Appl Catal A 469:45–51 _ 34. Zelazny A, Samson K, Grabowski R, S´liwa M, Ruggiero-Mikolajczyk M, Kornas A (2017) Reac Kinet Mech Cat 116:1–15 35. Liebig C, Paul S, Katryniok B, Guillon C, Couturier JL, Dubois JL, Dumeignil F, Hoelderich WF (2013) Appl Catal B Environ 132:170–182 36. Garcı´a-Ferna´ndez S, Gandarias I, Requies J, Gu¨emez MB, Bennici S, Auroux A, Arias PL (2015) J Catal 323:65–75 37. Gong LF, Lu Y, Ding YJ, Lin RH, Li JW, Dong WD, Wang T, Chen WM (2010) Appl Catal A Gen 390:119–126 38. Dam JT, Djanashvili K, Kapteijn F, Hanefeld U (2013) Chem Cat Chem 5:497–505 39. Zhu SH, Gao XQ, Zhu YL, Li YW (2015) J Mol Catal A: Chem 398:391–398 40. Seretis A, Tsiakaras P (2016) Fuel Process Technol 142:135–146 41. Liu LJ, Zhang YH, Wang AQ, Zhang T (2012) Chin J Catal 33:1257–1261 42. Wang XC, Meng LQ, Wu F, Jiang YJ, Wang L, Mu XD (2012) Green Chem 14:758–765 43. Zhu SH, Qiu YN, Zhu YL, Hao ShL, Zheng HY, Li YW (2013) Catal Today 212:120–126
123