Reac Kinet Mech Cat (2013) 109:117–131 DOI 10.1007/s11144-012-0538-x

Effect of zinc incorporation manner on a Cu–ZnO/Al2O3 glycerol hydrogenation catalyst Tingzhen Li • Chuan Fu • Junsheng Qi • Jie Pan Shuhong Chen • Junjie Lin

•

Received: 22 September 2012 / Accepted: 23 December 2012 / Published online: 12 January 2013 Ó Akade´miai Kiado´, Budapest, Hungary 2013

Abstract A systematic study was undertaken to investigate the effects of zinc incorporation manner on the textural properties, bulk and surface phase compositions, reduction behaviors, and surface acidity of a copper based glycerol hydrogenation catalyst. The catalyst samples were characterized by N2 physisorption, X-ray photoelectron spectroscopy (XPS), X-ray diffraction, transmission electron microscopy, air thermal gravimetric analysis (air-TGA), H2 thermal gravimetric analysis (H2-TGA), and NH3 temperature-programmed desorption. The glycerol hydrogenation performance of the catalysts was studied in a fixed-bed reactor. The characterization results indicated that the zinc promoter incorporated by using the co-precipitation method could improve the dispersion of copper oxide, and decrease the particle size of the copper oxide. The zinc incorporated with the impregnation method is enriched on the catalyst surface. The catalyst prepared by adding zinc using the co-precipitation method provides higher glycerol conversion and 1,2-propanediol selectivity, and lower selectivity to acetol. Keywords Glycerol
1,2-Propanediol
Dehydrogenation
Coprecipitation
Impregnation

Introduction In recent years, biodiesel has been considered as a renewable and environmentally friendly energy source [1, 2]. During biodiesel production via the transesterification of triglycerides with methanol, about 1 kg of crude glycerol is formed as a byproduct for every 9 kg of biodiesel [3]. The recent rapid development of biodiesel T. Li
C. Fu
J. Qi (&)
J. Pan
S. Chen
J. Lin College of Chemical and Environmental Engineering, Chongqing Three Gorges University, Wanzhou, Chongqing 404000, People’s Republic of China e-mail: [email protected]

123

118

Reac Kinet Mech Cat (2013) 109:117–131

processes has caused an oversupply of glycerol in the market [4, 5]. Therefore, the utilization of glycerol to produce valuable chemicals becomes crucial to the economic viability of biodiesel. One of the promising routes for glycerol utilization is the conversion of glycerol to 1,2-propanediol by catalytic hydrogenation [6]. 1,2-Propanediol is a commodity chemical used as a raw material in various applications, e.g. unsaturated polyester resins, functional fluids, pharmaceuticals, foods, cosmetics, liquid detergents, tobacco humectants, flavors and fragrances, personal care, paints, and animal feed [6, 7]. Currently, commercial 1,2-propanediol is produced from petroleum-derived propylene via the process involving propylene selective oxidation to propylene oxide and its subsequent hydrolysis [8, 9]. The hydrogenation of glycerol is a two-step process: in the first step, dehydration takes place and is followed by hydrogenation in next step [10]. For the purpose of the catalytic hydrogenation of glycerol to 1,2-propanediol, a lot of efforts have been made. Supported noble metals such as Ru, Rh and Pt have been extensively investigated [9, 11–18], and these catalysts exhibit excellent activity. Unfortunately, these catalysts often promote excessive C–C cleavage, resulting in the formation of degradation products such as ethylene glycol and methane. 1,2-Propanediol formation via glycerol hydrogenation involves the selective cleavage of a C–O bond without breaking C–C bonds. For the purpose, copper based catalysts were explored due to their high efficiency for C–O bond hydrogenation and poor activity for C–C bond cleavage [19], and the copper based catalysts exhibited superior performance in the glycerol hydrogenation reaction [6, 9, 10, 19–29]. Moreover, the Cu based catalysts can be used under mild reaction conditions and do not require a separate solid acid catalyst [20]. Among the copper based catalysts, Cu-Cr and Cu–Zn based catalysts showed exceptionally high catalytic activity in the hydrogenation of glycerol to 1,2-propanediol. For the Cu-Cr based catalysts, which have been extensively investigated by many groups [19, 21–25], the catalysts exhibit excellent catalytic activity in glycerol hydrogenation. However, the increasing concerns over green chemistry make the Cu–Cr catalysts undesirable due to their toxicity associated with chromium, which causes the researchers to focus their work on the Cu–ZnO based catalysts. Chaminand et al. [9] reported that a high selectivity to 1,2-propanediol (90 %) with a glycerol conversion of 20 % was obtained over the Cu–ZnO catalyst in presence of tungstic acid at 180 °C and 8.0 MPa H2 pressure. Wang and Liu [20] prepared a series of Cu–ZnO catalysts using the co-precipitation method with a range of Cu/Zn atomic ratio (0.6–2.0), and examined the catalyst glycerol hydrogenation performance at 180–240 °C under 4.2 MPa H2 pressure. They reported that a high 1,2-propanediol selectivity (84 %) was obtained with 23 % glycerol conversion at 200 °C over a Cu–ZnO catalyst (Cu/Zn = 1.0). Balaraju et al. [10] also synthesized a series of Cu–ZnO catalysts with various Cu to Zn weight ratios using co-precipitation method and reported that Cu–ZnO catalyst with Cu to Zn ratio of 50:50 exhibited high selectivity of 93 % to 1,2-propanediol at ca. 34 % glycerol conversion at H2 pressure of 2.0 MPa and reaction temperature of 200 °C. For the Cu–ZnO based catalysts, support material has a significant effect on the catalyst performance for the glycerol hydrogenation, though the pure support has lower activity for the hydrogenation of glycerol [6, 30].

123

Reac Kinet Mech Cat (2013) 109:117–131

119

Sato et al. [30] indicated that the addition of a basic support material can decrease the acetol selectivity of the Cu–ZnO based catalyst, while the presence of acidic supports such as Al2O3, ZrO2 and SiO2 can improve the acetol selectivity. Va´zquez et al. [26] indicated that the addition of Al2O3 can improve the Cu–ZnO catalyst stability. Thus, in order to improve the catalyst performance for glycerol hydrogenation, the support material has been incorporated into the Cu–ZnO based catalysts. Feng et al. [27] studied the hydrogenation of glycerol over Cu–ZnO catalysts supported with different supports (Al2O3, TiO2 and ZrO2) at 240–300 °C under 0.1 MPa of H2. They indicated that the presence of the support material supplied the acid sites for the dehydration of glycerol, and the Cu–ZnO catalyst supported with Al2O3 exhibited the highest selectivity to acetol (81–40 %) at reaction temperatures of 240–300 °C. Huang et al. [28] investigated the glycerol hydrogenation over the Cu–ZnO/Al2O3 catalyst (Cu/Zn/Al = 54.5:27.3:18.2) prepared with the continuous precipitation method, and a high 1,2-propanediol selectivity of 92 % was achieved at 190 °C, 0.64 MPa and 0.08 h-1. At the same time, the hydrogenation of glycerol to 1,2-propanediol was also investigated by Suchart et al. [6] over Cu–ZnO/Al2O3 catalysts prepared by three different methods (incipient wetness impregnation, co-precipitation and sol–gel) at 250 °C and hydrogen pressure of 3.2 MPa in a continuous flow fixed-bed reactor. They reported that the Cu–ZnO/Al2O3 catalyst prepared by the incipient wetness impregnation method exhibited the highest catalytic performance. In addition, a series of Cu–ZnO/Al2O3 catalysts with different Cu/Zn/Al metal compositions were prepared by Zhou et al. [29] using the co-precipitation method, and the performance of the catalysts for the glycerol hydrogenation was tested at a LHSV (liquid hourly space velocity) of 4.6 h-1, temperature of 220 °C, H2 pressure of 4.0 MPa and H2/glycerol molar ratio of 5:1. They reported that the catalyst with a Cu/Zn/Al molar ratio of 1:1:0.5 exhibited the best performance, glycerol conversion being 81.5 % and selectivity of propylene glycol 93.4 %. At the same time, the glycerol hydrogenation reaction kinetics was also studied over the best performance catalyst in an isothermal fixed-bed reactor at a hydrogen pressure range of 3.0–5.0 MPa and a temperature range of 220–240 °C, and indicated that the activation energies for dehydration and hydrogenation reactions were 86.56 and 57.80 kJ/mol, respectively. Although the Cu–ZnO/Al2O3 catalyst for the glycerol hydrogenation has been intensively investigated, the manner of zinc incorporation is rarely reported. Moreover, different zinc incorporation manners can lead to different physicochemical properties of the catalysts. Therefore, it is significant to study the effect of the zinc incorporation manner on the performance of the Cu–ZnO/Al2O3 glycerol hydrogenation catalyst. The present study is undertaken to investigate the effect of the zinc incorporation manner on the performances of a Cu–ZnO/Al2O3 glycerol hydrogenation catalyst. Particular attention is focused on the effect of the zinc incorporation manner on the textural properties, reduction, bulk phase and surface compositions of the catalyst as prepared and after reduction. The glycerol hydrogenation activity and product selectivity of the catalyst are well correlated with the characterization results. In addition, the calcination process of the catalyst is also investigated by TG technique in detail.

123

120

Reac Kinet Mech Cat (2013) 109:117–131

Experimental Catalyst preparation The catalysts used in the present study were prepared using the combination of precipitation and impregnation. During the catalyst preparation, two methods (co-precipitation and impregnation) were used in the zinc incorporation. For the zinc co-precipitated catalyst (COP), a solution containing Cu(NO3)2
3H2O, Zn(NO3)2
6H2O and Al(NO3)3
9H2O with a desired molar ratio (5:1:1) was introduced into a precipitation vessel containing deionized water at 80 ± 1 °C. A sodium carbonate solution (1 M) was added simultaneously into the vessel to maintain the pH at a constant value of 8.0 ± 0.1. After the precipitation, the precipitate was washed thoroughly with deionized water (80 °C), and subsequently filtered. After the filtration, the catalyst sample was dried at 120 °C overnight and calcined at 400 °C for 4 h. For the zinc impregnated catalyst (IMP), a solution containing Cu(NO3)2
3H2O and Al(NO3)3
9H2O with a molar ratio of 5:1 was precipitated by a sodium carbonate solution (1 M) at 80 ± 1 °C and pH value of 8.0 ± 0.1. After the precipitation, the precipitate was washed thoroughly with deionized water (80 °C), filtered and subsequently dried overnight at 120 °C. After drying, the catalyst sample was impregnated by Zn(NO3)2
6H2O solution with a desired Cu:Zn:Al molar ratio of 5:1:1. After the impregnation, the catalyst sample was dried at 120 °C overnight, and subsequently calcined at 400 °C for 4 h. In addition, in order to study the effect of the zinc incorporation manner on the catalyst performance conveniently, a benchmark Cu/Al2O3 catalyst (BEN) with a desired Cu:Al molar ratio of 5:1 was also prepared with the coprecipitation method under same preparation conditions.

Catalyst characterization The textural properties (BET surface area, average pore size and pore volume) of the catalyst samples were measured by N2 physisorption at its normal boiling point (-196 °C) using a Micromeritics ASAP 2500 instrument. Prior to the measurement, the catalyst samples were degassed at 120 °C for 6 h. The surface atomic concentrations of the fresh catalysts were determined by XPS using a PHI Quantera SXM spectrometer with Al Ka radiation and the spectrometer resolution of energy was 0.5 eV. The peak positions were corrected for sample charging by setting the C 1s binding energy at 284.8 eV. The relative surface concentrations of the elements were determined using the whole peak area of Cu 2p, Zn 2p and Al 2p regions and the corresponding sensitivity factors, respectively. The crystalline structure of the catalyst samples were measured by powder X-ray diffraction (XRD) on a D/max-RA X-ray diffractometer (Rigaku, Japan), equipped ˚ ) at 40 kV and 150 mA. The patterns were with Cu Ka radiation (k = 1.5406 A scanned at a rate of 2°/min from 2h = 10° to 80°. The phases were identified by comparing diffraction patterns with those on the standard powder XRD cards compiled by the Joint Committee on Powder Diffraction Standards (JCPDS).

123

Reac Kinet Mech Cat (2013) 109:117–131

121

Transmission electron microscopy (TEM) analysis was performed on a Hitachi H-600 electron microscope operated at an accelerating voltage of 100 kV. A small amount of specimen was prepared by ultrasonically suspending the powder sample in ethanol, and drops of the suspension were deposited on a carbon-coated copper grid dried at room temperature before analysis. The thermogravimetric analysis (air-TGA, H2-TGA) experiments were performed in Mettler Toledo TGA/DSC1 thermogravimetry. For the air-TGA, the sample holder was loaded with 10 mg of catalyst precursor, and the decomposition of the sample was monitored under a flow of air (50 mL/min) and heated up to 800 °C with a heating rate of 10 °C min–1. For the H2-TGA, 20–30 mg sample was treated in 5%H2/Ar (on mole basis) at 100 °C for 20 min and then temperature was increased to 700 °C with a heating rate of 10 °C min–1. The acidities of the reduced catalysts were determined by temperatureprogrammed desorption of ammonia (NH3-TPD) in a fixed-bed continuous flow micro-reactor at atmospheric pressure. Prior to the test, the catalyst samples (100 mg) were reduced at 350 °C for 4 h, and then were ammonia-saturated in a NH3 stream at 50 °C for 0.5 h. After purging with helium (30 mL/min) at 50 °C for 0.5 h to remove the physically adsorbed NH3, the samples were heated up to 750 °C with a heating rate of 10 °C/min. Catalytic activity measurements The catalytic activity was examined in a stainless steel tubular fixed-bed reactor (length of 700 mm and i.d. of 13 mm) with an ice trap, packed with 5.0 g of catalyst (20–40 mesh), operating at 250 °C under 0.1 MPa of H2 pressure. Prior to the reactions, the catalysts were in situ reduced at 350 °C for 4 h in a flowing of H2/N2 (1:9 mol/mol). After that, the reactor was cooled to the reaction temperature of 250 °C, and an aqueous solution of glycerol (80 wt%) was continuously introduced to the reactor using a syringe pump with a weight hourly space velocity (WHSV) of 0.05 h-1. At the same time, pure hydrogen was introduced to the reactor with a molar ratio of H2/glycerol = 150:1. The tail gas products were analyzed by GC (6890N and 4890D; Agilent). The liquid products were identified by a gas chromatograph (6890N, Agilent, USA) with a coupled mass spectrometer (5973, Agilent). The conversion was calculated based on the amount of glycerol actually reacted per fed glycerol, while the selectivity based on the amount of products observed to be formed per the amount of glycerol actually reacted.

Results and discussion Textural properties The textual properties (BET surface area, pore volume and average pore diameter) and elemental compositions in the bulk and the surface of the fresh catalysts are shown in Table 1. It can be seen that the COP catalyst has the largest BET surface area (75 m2/g) and pore volume (0.34 cm3/g), which are much higher than those of

123

122

Reac Kinet Mech Cat (2013) 109:117–131

the benchmark Cu/Al2O3 (BEN) catalyst. This result may be due to the fact that the zinc added with the co-precipitation method has a strong interaction with copper, which improves the dispersion of copper, suppresses the agglomeration of copper oxides during calcination, and results in a high BET surface area and large pore volume. However, the BET surface area and pore volume of the IMP catalyst are slightly lower than those of the BEN catalyst. Compared with the benchmark Cu/Al2O3 (BEN) catalyst, the slight decrease of IMP catalyst in the BET surface area and pore volume may be caused by the zinc loading [6]. The results of TEM in the present study also show that the COP catalyst has the smallest CuO particle size. As shown in the results of XPS characterization, the Zn/Cu and Al/Cu atomic ratios on the catalyst surface are much higher than those in the bulk, which indicates that the added zinc and aluminum are enriched on the catalyst surface during the catalyst preparation. At the same time, the surface Zn/Cu atomic ratio of the IMP catalyst is much higher than that of COP catalyst, which indicates that the zinc added with the impregnation method is mainly concentrated on the catalyst surface. Structural properties The XRD patterns of the catalysts with different treatment procedures (after drying, after calcination and after reduction) are presented in Fig. 1. As shown in Fig. 1, after drying, the benchmark Cu/Al2O3 (BEN) catalyst is mainly composed of Cu2(OH)3NO3 and Cu2(OH)2CO3. For the COP catalyst, besides Cu2(OH)3NO3 and Cu2(OH)2CO3, a hydrotalcite-like phase, Cu2Zn4Al2(OH)16CO3
4H2O, is also present. However, after drying, the IMP catalyst is mainly composed of Zn3(OH)4(NO3)2, Cu2(OH)3NO3 and Cu2(OH)2CO3. The results suggest that the zinc incorporated with co-precipitation method can form hydrotalcite-like phase with copper and aluminum. Similar results were also reported by Li and Inui [31] in the study of the Cu–ZnO/Al2O3 methanol synthesis catalyst. After the calcination, the BEN catalyst is mainly composed of CuO, while the main phases of the COP catalyst are transformed into CuO and Cu2Zn4Al2(OH)16CO3
4H2O, and no ZnO phase is detected. The results indicate that the phases of Cu2(OH)3NO3 and Cu2(OH)2CO3 are transformed into CuO during the calcination, while the phase of Cu2Zn4Al2(OH)16CO3
4H2O cannot be decomposed completely in the process of Table 1 Textural properties and elemental composition in the bulk and the surface of the fresh catalysts

Catalysts

COP

IMP

BEN

BET surface area (m2/g)

75

25

28

Pore volume (cm3/g) Average pore size (nm)

0.34 17.7

0.07 10.5

0.08 10.0

Zn/Cu atomic ratio Bulka

0.20

0.20

–

Surface

0.28

0.87

–

Al/Cu atomic ratio Bulka

0.19

0.19

0.19

Surface

1.03

0.78

0.92

a

Bulk Zn/Cu and Al/Cu atomic ratios measured by AAS

123

Reac Kinet Mech Cat (2013) 109:117–131

123

I

Cu2Zn4Al2(OH)16CO3.4H2O Zn3(OH)4(NO3)2

Intensity (a.u.)

Cu2(OH)2CO3 Cu2(OH)3NO3 BEN IMP COP

10

20

30

40

II

50

2θ (°)

60

70

80

Cu2Zn4Al2(OH)16CO3.4H2O

Intensity (a.u.)

CuO

ZnO

BEN

IMP COP

10

20

30

40

50

2θ (°)

60

70

80

ZnO

III

Intensity (a.u.)

Cu

BEN

IMP

COP

10

20

30

40

50

2θ (°)

60

70

80

Fig. 1 XRD patterns of the catalysts at different stages: I (after dryness), II (after calcination), and III (after reduction)

123

124 Fig. 2 TEM images of the fresh catalysts

123

Reac Kinet Mech Cat (2013) 109:117–131

Reac Kinet Mech Cat (2013) 109:117–131

125

calcination at 400 °C. Similar phenomena were also observed by Cabrera et al. [32] for the thermal decomposition of Cu2Zn4Al2(OH)16CO3
4H2O. However, after the calcination, the IMP catalyst is mainly composed of CuO and ZnO, which indicates that the phase of Zn3(OH)4(NO3)2 is transformed into ZnO during the calcination. In addition, the average CuO particle sizes for the COP, IMP and BEN catalysts are 15, 70 and 50 nm, estimated by TEM (Fig. 2). The smaller CuO particle size for the COP catalyst can be explained by the synergy between CuO and ZnO. It is reasonable to suppose that the interdispersion of CuO and ZnO in COP catalyst was more intimate than that in IMP catalyst because the former resulted mainly from an isomorphous substance (Cu2Zn4Al2(OH)16CO3
4H2O), in which copper and zinc were in a greater interaction state. Upon calcination, ZnO and CuO had to segregate because of their different crystal structures and limited solubilities, forming a very fine interdispersion of CuO. After reduction, the BEN catalyst is mainly composed of Cu, and at the same time, the main phases of the IMP catalyst are converted to Cu and ZnO, indicating that CuO is reduced to Cu during the catalyst reduction. For the COP catalyst, after reduction, the diffraction peaks of Cu2Zn4Al2(OH)16CO3
4H2O disappear, and only metallic copper can be detected, which suggests that the phase of Cu2Zn4Al2(OH)16CO3
4H2O is gradually decomposed during the catalyst reduction. Similar results were also reported in the Cu–ZnO/Al2O3 methanol synthesis catalyst preparation [32]. In addition, there are no diffraction peaks of Al2O3 detected by XRD analysis, revealing that Al2O3 is in amorphous phase in the catalyst. Air-TGA analysis The results of thermogravimetric analysis (TGA) for the catalyst precursors are shown in Fig. 3. For the BEN catalyst, there is only a pronounced weight loss in the range of 150–350 °C, corresponding to a clearly visible weight loss peak at around 238 °C in the DTG curve. Based on the results of XRD, this weight loss can be assigned to the decompositions of Cu2(OH)3NO3 and Cu2(OH)2CO3 structures [33, 34]. At the same time, the TG is flat after 700 °C suggesting the formation of stable CuO. The TG/DTG curves of IMP catalyst closely match with those of BEN catalyst except that the weight loss temperature range (200–300 °C) becomes much narrower. The weight loss of IMP catalyst can be ascribed to the decompositions of Cu2(OH)3NO3, Cu2(OH)2CO3 and Zn3(OH)4(NO3)2 structures. Compared with the BEN catalyst, the decrease of weight loss temperature range for the IMP catalyst may be explained as that in the process of zinc impregnation, the addition of zinc nitrate solution improves the transformation of Cu2(OH)2CO3 formed during the precipitation to Cu2(OH)3NO3 [31], due to the acidity of zinc nitrate solution. At the same time, Cu2(OH)2CO3 has a much higher thermal stability than Cu2(OH)3NO3 [31]. Thus, the precursor of the IMP catalyst can be decomposed in a narrower temperature range. However, for the COP catalyst, the precursor undergoes four weight loss processes upon heating in a flow of air. The first weight loss (below 100 °C) can be ascribed to the mass loss of physically adsorbed water. The second weight loss process in the 100 to 150 °C temperature range is associated with the removal of water of crystallization present in the phase of Cu2Zn4Al2(OH)16CO3
4H2O. The third weight loss located in the temperature range of

123

126

Reac Kinet Mech Cat (2013) 109:117–131

COP 100

Weight loss (%)

DTG 90

80

TG 70

60 100

200

300

400

500

600

700

800

o

Temperature ( C) 110

IMP DTG

Weight loss (%)

100

90

80

70

TG

60 100

200

300

400

500

600

700

800

o

Temperature ( C) 110

BEN DTG

Weight loss (%)

100

90

80

TG

70

60 100

200

300

400

500 o

Temperature ( C) Fig. 3 TG/DTG curves of the catalyst precursors

123

600

700

800

Reac Kinet Mech Cat (2013) 109:117–131

127

COP

dW/dT

IMP

BEN

100

200

300

400

500

600

700

o

Temperature ( C) Fig. 4 H2-DTG profiles of the catalysts

150 to 500 °C can be attributed to the decompositions of Cu2Zn4Al2(OH)16CO3
4H2O and Cu2(OH)2CO3 [32, 33, 35]. The final weight loss in the 500 to 800 °C temperature range can be assigned to the decomposition of CO23 groups derived from the decomposition of the phase of Cu2Zn4Al2(OH)16CO3
4H2O [32].

Reduction properties The H2-DTG profiles of the catalysts after calcination are illustrated in Fig. 4. It can be seen that the reduction peaks of the catalysts are mainly located in the range of 200–350 °C. Based on the results of XRD, the reduction peak in all the samples can be assigned to the reduction of copper oxide. For the BEN and IMP catalysts, only a single reduction peak is observed at 298 and 317 °C, respectively. The temperature of the reduction peak of the IMP catalyst is higher than that of the BEN catalyst, which suggests that the copper oxide of the IMP catalyst is more difficult to reduce in H2. As illustrated by the XPS measurement (mentioned above), zinc oxide is mainly concentrated on the catalyst surface when zinc is added using the impregnation method. At the same time, it is known that zinc oxide cannot be reduced by H2. Thus, the enrichment of zinc oxide on the catalyst surface suppresses the reduction of copper oxide in H2 for the IMP catalyst. For the COP catalyst, two reduction peaks are observed in the range of 150–300 °C. The low temperature shoulder peak can be assigned to the reduction of CuO with smaller particle size and/or the reduction of CuO to Cu?, and the high temperature peak can be assigned to the reduction of CuO with larger particle size and/or the reduction of Cu? to metallic copper [10, 36, 37]. Compared with the BEN catalyst, the reduction peaks of the COP catalyst shift to the lower temperature range. This result may be explained as that the zinc incorporated with the co-precipitation method improves

123

128

Reac Kinet Mech Cat (2013) 109:117–131

619 518 COP

Intensity

136

465

562

125 IMP 612

111

BEN

100

200

300

400

500

600

700

o

Temperature ( C) Fig. 5 NH3-TPD profiles of the catalysts

the dispersion of CuO, reduces the CuO particle sizes, and results in a relatively lower CuO reduction temperature [27]. Surface acidity The acidity of the reduced catalysts was estimated by NH3-TPD, and the strength of the acid sites was determined by the temperature at which the adsorbed NH3 desorbs. According to the NH3 desorption temperature in NH3-TPD profile, the acid sites can be classified into three types, weak (below 250 °C), middle (256–367 °C), and strong (higher 427 °C) [38, 39]. The NH3-TPD profiles of the reduced catalysts are shown in Fig. 5. As shown in Fig. 5, the NH3 desorption peaks of all the catalysts are located in the ranges of 111–136 °C and 465–619 °C, meaning that all the reduced catalysts possess both weak and strong acid sites. Similar results were also reported by Feng et al. [27] in the Cu–ZnO/Al2O3 catalyst investigation. It also can be seen that the NH3-TPD profile of Cu/Al2O3 (BEN) catalyst shows two NH3 desorption peaks at 111 and 612 °C. However, the NH3-TPD profiles of Cu–ZnO/ Al2O3 (COP and IMP) catalysts show three NH3 desorption peaks at 136, 518, 619 °C and 125, 465, 562 °C. This result suggests that the NH3 desorption peaks located at 111, 125, 136, 562, 612 and 619 °C of the reduced catalysts can be attributed to desorption of NH3 adsorbed by the Al2O3 support, and the desorption peaks located at 465 and 518 °C can be attributed to the desorption of NH3 adsorbed by ZnO, which are consistent with the results of earlier studies [39, 40]. The area of the NH3 desorption peaks (136 and 619 °C) for the COP catalyst are larger than that of the peaks (125 and 562 °C) for the IMP catalyst. As indicated by the XPS analysis, the Al2O3 surface content of the COP catalyst is higher than that of the IMP catalyst, which results in a larger amount of NH3 adsorption for the COP catalyst. At the same time, the area of the NH3 desorption peak (465 °C) for the IMP

123

Reac Kinet Mech Cat (2013) 109:117–131

129

100

Glycerol conversion (%)

90 80 70 60 50 40 30

COP IMP BEN

20 10 0 0

20

40

60

80

100

Time on stream (h) Fig. 6 Glycerol hydrogenation conversion and stability of the catalysts. Reaction conditions: 250° C, 0.1 MPa, WHSV = 0.05 h-1, H2/glycerol = 150:1 (mol/mol) Table 2 Glycerol hydrogenation conversions and selectivities Catalysts

COP

IMP

BEN

Time on stream (h)

48.5

96.5

46.5

97.0

49.5

97.5

Glycerol conversion (%)

85.5

85.3

63.7

58.0

48.2

44.5

1,2-propanediol

83.7

82.5

68.4

63.7

75.8

74.6

Acetol

11.4

12.1

26.8

31.2

15.5

16.3

Othersa

4.9

5.4

4.8

5.1

8.7

9.1

Selectivity (%)

-1

Reaction conditions: 250 °C, 0.1 MPa, WHSV = 0.05 h , H2/glycerol = 150:1 (mol/mol) a

Methanol, ethanol and unknown products

catalyst is larger than that of the peak (518 °C) for the COP catalyst, which can be ascribed to the fact that the surface ZnO content of the IMP catalyst is higher than that of the COP catalyst. In addition, the total acidity of the reduced COP catalyst is higher than that of the reduced IMP and BEN catalyst, which may be ascribed to the fact that the acid strength of Al2O3 is stronger than that of ZnO [40], and higher surface area can provide more acid sites. Glycerol hydrogenation performance of the catalysts The glycerol hydrogenation performances (activity, stability and selectivity) are shown in Fig. 6 and Table 2. It can be seen that the COP catalyst exhibits the highest glycerol conversion while the BEN catalyst shows the lowest activity, meaning that the addition of zinc improves the catalyst activity. The glycerol conversion of the COP catalyst is almost constant, while the activities of the IMP and BEN catalysts decrease gradually with increasing time on stream. It is known

123

130

Reac Kinet Mech Cat (2013) 109:117–131

that the hydrogenation of glycerol is a two-step process involving a first step of dehydration to acetol and subsequent hydrogenation to 1,2-propanediol over the acid catalyst [10, 41, 42]. For the copper based catalysts, the dehydration of glycerol proceeds on the acid sites of the metal oxides, and the acetol hydrogenation occurs on the copper metal site [20, 27, 28]. As shown in the surface acidity estimation, the COP catalyst has the highest surface acidity, and the presence of a large quantity of acid sites facilitates the dehydration of glycerol. At the same time, for the COP catalyst, it has been illustrated that in the N2 physisorption and TEM analyses, the incorporation of zinc with the coprecipitation method improves the dispersion of copper and reduces the copper particle size, which provides more active copper metal sites for the acetol hydrogenation [20]. Thus, the COP catalyst has the highest glycerol conversion and 1,2-propanediol selectivity, and the lowest acetol selectivity among the catalysts. In addition, the strong interaction between copper and zinc incorporated with the coprecipitation method in the COP catalyst suppresses the sintering of Cu and ZnO during the reaction, which leads to a stable conversion in the glycerol hydrogenation reaction [6]. Compared with the BEN catalyst, the IMP catalyst has a higher glycerol conversion and acetol selectivity, and a lower 1,2-propanediol selectivity. This result may be due to the fact that the zinc added with the impregnation method for the IMP catalyst is mainly concentrated on the catalyst surface, which supplies more acid sites for the glycerol dehydration to acetol [20, 28], while the surface copper content decrease caused by the surface enrichment of zinc suppresses the hydrogenation of acetol to 1,2-propanediol [20].

Conclusions The effects of zinc incorporation manner on the Cu–ZnO/Al2O3 glycerol hydrogenation catalysts have been systemically studied. The zinc incorporated with the co-precipitation method can improve the dispersion and reduction of copper oxide, and the COP has smaller CuO particle size, larger BET surface area and higher surface acidity. The zinc incorporated with the impregnation method is mainly concentrated on the catalyst surface. The COP catalyst provides higher glycerol conversion and 1,2-propanediol selectivity, and lower selectivity to acetol. Acknowledgments We thank the financial support from the Education Department Science Foundation of Chongqing (KJ08112) and Talent Introduction Foundation of Chongqing Three Gorges University (2007-SXXYRC-010).

References 1. 2. 3. 4. 5. 6. 7.

Chheda JN, Huber GW, Dumesic JA (2007) Angew Chem Int Ed 46:7164–7183 McKendry P (2002) Bioresour Technol 83:37–46 Karinen RS, Krause AOI (2006) Appl Catal A 306:128–133 Zhou CH, Beltramini JN, Fan YX, Lu GQ (2008) Chem Soc Rev 37:527–549 Behr A, Eilting J, Irawadi K, Leschinski J, Lindner F (2008) Green Chem 10:13–30 Suchart P, Siriporn J, Thammanoon S, Thirasak R, Somchai O (2011) Catal Today 174:59–64 Kenar JA (2007) Lipid Technol 19:249–253

123

Reac Kinet Mech Cat (2013) 109:117–131 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42.

131

Cortright RD, Sanchez-Castillo M, Dumesic JA (2002) Appl Catal B 39:353–359 Chaminand J, Djakovitch L, Gallezot P, Marion P, Pinel C, Rosier C (2004) Green Chem 6:359–361 Balaraju M, Rekha V, Sai Prasad PS, Prasad RBN, Lingaiah N (2008) Catal Lett 126:119–124 Miyazawa T, Kusunoki Y, Kunimori K, Tomishige K (2006) J Catal 240:213–221 Maris EP, Davis RJ (2007) J Catal 249:328–337 Lahr DG, Shanks BH (2005) J Catal 232:386–394 Maris EP, Ketchie WC, Murayama M, Davis RJ (2007) J Catal 251:281–294 Furikado I, Miyazawa T, Koso S, Shimao A, Kunimori K, Tomishige K (2007) Green Chem 9:582–588 Kurosaka T, Maruyama H, Naribayashi I, Sasaki Y (2008) Catal Commun 9:1360–1363 Balaraju M, Rekha V, Sai Prasad PS, Prabhavathi Devi BLA, Prasad RBN, Lingaiah N (2009) Appl Catal A 354:82–87 Feng J, Fu HY, Wang JB, Li RX, Chen H, Li XJ (2008) Catal Commun 9:1458–1464 Dasari MA, Kiatsimkul PP, Sutterlin WR, Suppes GJ (2005) Appl Catal A 281:225–231 Wang S, Liu H (2007) Catal Lett 117:62–67 Liang C, Ma Z, Ding L, Qiu J (2009) Catal Lett 130:169–176 Nam DK, Seogil O, Ji BJ, Kwang SJ, Jongheop Y (2010) Top Catal 53:517–522 Nam DK, Seogil O, Ji BJ, Kwang SJ, Jongheop Y (2010) Korean J Chem Eng 27:431–434 Zhenle Y, Junhua W, Lina W, Weihui X, Ping C, Zhaoyin H, Xiaoming Z (2010) Bioresour Technol 101:7088–7092 Mane RB, Ghalwadkar AA, Hengne AM, Suryawanshi YR, Rode CV (2011) Catal Today 164:447–450 Va´zquez MS, Rojas AR, Collins-Martinez V, Ortiz AL (2005) Catal Today 107–108:831–837 Feng Y, Yin H, Wang A, Shen L, Yu L, Jiang T (2011) Chem Eng J 168:403–412 Huang L, Zhu YL, Zheng HY, Li YW, Zeng ZY (2008) J Chem Technol Biotechnol 83:1670–1675 Zhou Z, Li X, Zeng T, Hong W, Cheng Z, Yuan W (2010) Chin J Chem Eng 18:383–390 Sato S, Akiyama M, Takahashi R, Hara T, Inui K, Yokota M (2008) Appl Catal A 347:186–191 Li JL, Inui T (1996) Appl Catal A 137:105–117 Cabrera IM, Granados ML, Fierro JLG (2002) Phys Chem Chem Phys 4:3122–3127 Zhu CL, Chen CN, Hao LY, Hu Y, Chen ZY (2004) Solid State Commun 130:681–686 Audebrand N, Auffre´dic JP, Loue¨r D (1998) Chem Mater 10:2450–2461 Voyer N, Soisnard A, Palmer SJ, Martens WN, Frost RL (2009) J Therm Anal Calorim 96:481–485 Guerreiro ED, Gorriz OF, Rivarola JB, Arrfia LA (1997) Appl Catal A 165:259–271 Oguchi H, Nishiguchi T, Matsumoto T, Kanai H, Utani K, Matsumura Y, Imamura S (2005) Appl Catal A 281:69–73 Barrio VL, Arias PL, Cambra JF, Gu¨emez MB, Pawelec B, Fierro JLG (2003) Appl Catal A 242:17–30 Lu H, Yin H, Liu Y, Jiang T, Yu L (2008) Catal Commun 10:313–316 Sharath RK, Boris GS, Guggilla VS, Komandur VRC, Abraham C (2006) J Catal 242:319–331 Mane RB, Hengne AM, Ghalwadkar AA, Vijayanand S, Mohite PH, Potdar HS, Rode CV (2010) Catal Lett 135:141–147 Yuan Z, Wu P, Gao J, Lu X, Hou Z, Zheng X (2009) Catal Lett 130:261–265

123