- Email: [email protected]
The influence of different precipitants on the copper-based catalysts for hydrogenation of ethyl acetate to ethanol
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
The influence of different precipitants on the copper-based catalysts for hydrogenation of ethyl acetate to ethanol Kaili Zhong, Xin Wang* The State Key Laboratory of Chemical Engineering, East China University of Science and Technology, Shanghai 200237, China
article info
abstract
Article history:
Ethanol fuel has become a hot topic in the economic, political, environmental, and sci-
Received 8 March 2014
entific areas. In this work, a new way for the synthesis of ethanol by hydrogenolysis of
Received in revised form
ethyl acetate is introduced and the impact of different precipitants on the ethyl acetate
6 May 2014
hydrogenolysis catalysts is systematically investigated by several considerations, including
Accepted 9 May 2014
dispersion effects, the texture of the catalysts, and the copper phases in the surface layer of
Available online 9 June 2014
the reduced catalysts, etc. These precursors and catalysts are characterized by inductively coupled plasma-atomic emission spectroscopy, N2-adsoption, X-ray diffraction, trans-
Keywords:
mission electronic microscope, H2 temperature-programmed reduction and X-ray photo-
Ethanol fuel
electron spectroscopy. It is confirmed that the choice of precipitant is of great importance.
Precipitant
The samples are classified into two types, depending on the anion of precipitant. Except
Hydrogenolysis
the catalyst prepared by (NH4)2CO3, in which low copper loading is observed, type B cata-
Ethyl acetate
lysts (eCO2 3 ) possess smaller copper particles and larger BET surface than that of type A
Cu/ZnO/SiO2
catalysts (eOH), while the difference of catalysts in the same type is not obvious. Moreover, the coexistence of Cuþ and Cu0 is only detected in reduced type B catalysts. In general, ethyl acetate hydrogenolysis activity varies considerably with the precipitant, in the following order: Na2CO3 NaHCO3 > NaOH KOH > (NH4)2CO3. Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
Introduction As a potential alternative to oil, ethanol fuel has become a hot topic in the economic, political, environmental, and scientific areas [1]. The synthesis of ethanol by hydrogenolysis of ethyl acetate is of great commercial potential, since ethyl acetate can be produced at lower cost and on a large scale.
The catalytic hydrogenesis of esters to their corresponding alcohols [Eq. (1)] has been extensively studied since the first report by Flokers and Adkins [2]. RCOOR0 þ 2H2 /RCH2 OH þ R0 OH
(1)
A mechanism for acetate hydrogenolysis was proposed by Yan et al. [3] and Evans et al. [4] The acetate is adsorbed dissociatively [Eq. (2)]
* Corresponding author. Tel.: þ86 021 64252836. E-mail address: [email protected] (X. Wang). http://dx.doi.org/10.1016/j.ijhydene.2014.05.045 0360-3199/Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
10952
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
RCOOR0 þ 2* /RCO* þ R0 O*
(2)
where * indicates a catalytic site on the surface. Isotopic labeling studies showed that the alkoxy fragment R0 O* reacts quickly to form R0 OH, while the acyl group RCO* is adsorbed longer. Thus the hydrogenation of the acyl fragment is believed to be the rate-determining step during acetate hydrogenolysis. The acyl group can be hydrogenated either to the desired alcohol or to the corresponding aldehyde, which is subsequently hydrogenated [3,5]. An in situ infrared spectroscopy investigation [6] during ethyl acetate hydrogenolysis gained further evidence for the proposed mechanism via dissociatively adsorbed fragments. Furthermore, Kenvin and White [7] could explain their results based on the mechanism proposed by Evans et al. For ester hydrogenolysis, copper-based catalysts including copper chromite, silica-supported copper and Raney copper are well known to exhibit high activities [5,8,9], and the experimental results strongly indicate that copper is the essential catalytic component in the ester hydrogenolysis reaction [8]. Moreover, the addition of metal oxides, such as Cr2O3, ZrO2, Fe2O3 or ZnO, etc. has been extensively investigated to promote the activity of Cu in the hydrogenolysis reaction [10]. When both activity and selectivity were taken into account, Brands [9] and van de Scheur [11] concluded that ZnO was the promoter of choice. It has been reported [10,11] that the addition of ZnO into copper-based catalyst can lead to a higher Cu dispersion (i.e., a smaller Cu particle), and ZnO is believed to act as a structural promoter in enhancing the area of Cu, the active site for the reaction. According to references [5,10,12e14], coprecipitation method is a conventional preparation method for ester hydrogenolysis catalysts, thus, the choice of precipitants is of great importance. Kim et al. [10] and Yuan et al. [12] chose NaHCO3 and Na2CO3 as the precipitant in the references, respectively, while Chen et al. [14] reported that the Cu/SiO2 catalysts precipitated by NaOH showed high activity in hydrogenation of ester. In general, the impact of different precipitants on the performance of ethyl acetate hydrogenolysis catalysts has never systematically investigated. In this work, the silica supported CueZn catalysts are prepared via coprecipitation method for hydrogenation of ethyl acetate, and the main aim of this study is to have a better view of the influence of different precipitants on the catalysts by several considerations, including dispersion effects, metal loading, the texture of the catalysts, and the copper phases in the surface layer of the reduced catalysts, etc. The characterization results are discussed alongside with the catalytic data obtained using a laboratory fixed-bed reactor.
Experimental Catalyst preparation Hydrogenation components CueZn/SiO2 (Cu/Zn/SiO2 molar ratio ¼ 2:1:1) were prepared by coprecipitation method. In a round bottom flask, appropriate amount of silica sol (Sinopharm Chemical Reagent Co., Ltd., China, 30%) was mixed with distilled water under vigorous stirring. Required
amounts of Cu(NO3)2$5H2O and Zn(NO3)2$4H2O (Sinopharm Chemical Reagent Co., Ltd., China, AR) were dissolved in 200 ml distilled water at room temperature, and this solution was referred as A, in which Cu2þ/Zn2þ ¼ 2 and Cu2þ þ Zn2þ ¼ 0.5 mol/L. Solution B (200 ml) was an aqueous of the precipitant [NaOH, KOH, Na2CO3, NaHCO3 or (NH4)2CO3 (Sinopharm Chemical Reagent Co., Ltd., China, AR)]. Then solution A and solution B were simultaneously added dropwise (in around 1 h) to the round bottom flask with vigorously stirring in the ambient temperature, the pH was maintained at 8.0 during dripping process by controlling the flow velocity of solution A and solution B. After completing the process of precipitation, the resulted suspension was stirred for 30 min, filtered and washed five times with distilled water until the pH of filtrate reached 7.0. The precipitate was dried in an oven at 120 C overnight and then calcined in a muffle furnace in air by heating at 500 C for 5 h. Finally, the resulting catalysts were crushed and sieved. The 20e40 mesh particles were used for the activity test. The prepared catalysts were labeled as Cu2Zn1Si1eM, where M stands for the precipitant used in the process of catalysts preparation. Herein, the samples are classified into two types, depending on the anion of precipitant. The Cu2Zn1Si1eNaOH catalyst and Cu2Zn1Si1eKOH catalyst are referred as type A, and the other three are referred as type B.
Characterization of catalyst Inductively coupled plasma-atomic emission spectroscopy Actual metal loading in the calcined catalysts was determined with inductively coupled plasma-atomic emission spectroscopy (ICP-AES) using a multichannel Thermo Jarrel-Ash ICAP 612 E spectrometer.
N2-adsorption Nitrogen isotherms were measured at 196 C with an ASAP 2020 (Micromeritrics). Before experiment, the samples were heated at 120 C and outgassed overnight at this temperature under a vacuum of 105 Torr to a constant pressure. N2 isotherms were obtained in both adsorption and desorption modes. The surface areas of the calcined catalysts were determined by the BET method. The total pore volume (TPV) was calculated from the amount of vapor adsorbed at a relative pressure (p/p0) close to unity (desorption curve), where p and p0 are the measured and equilibrium pressures, respectively. Pore size distribution curves were established from the desorption branches of the isotherms using the BJH model [15].
X-ray diffraction X-ray diffraction (XRD) technique was used to characterize the crystal structure. The experiments were performed using a Siemens D500 diffractometer, using Cu Ka radiation (40 kV, 100 mA) and equipped with a graphite monochromator with a reflected beam. The powder diffraction patterns were recorded in the 2q range from 20 to 80 . An X-ray chamber-reactor was used for the high-temperature registration of the spectra.
Temperature-programmed reduction Temperature programmed reduction (TPR) of H2 was carried out on Auto Chem2910 (Micromeritics) instrument to study
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
the reducibility of the catalysts. In a typical experiment, around 0.1 g of calcined catalyst was exposed to a reducing gas consisting of 5.0 vol% H2 in argon with a temperature ramp from ambient temperature to 600 C at a heating rate of 10 C/ min.
X-ray photoelectron spectroscopy X-ray photoelectron spectroscopy (XPS) and X-ray induced Auger electron spectroscopy (XAES) were obtained using a VG MultiLab 2000 analyzer with a Mg Ka radiation source and the binding energies were referenced to C1s (284.6 eV).
10953
Table 1 e Metal loading determined using ICP-AES in the calcined catalysts. Type
Catalysts
Measured copper loading (wt%)
Measured zinc loading (wt%)
Type Type Type Type Type e
Cu2Zn1Si1eNaOH Cu2Zn1Si1eKOH Cu2Zn1Si1e(NH4)2CO3 Cu2Zn1Si1eNaHCO3 Cu2Zn1Si1eNa2CO3 CTV
41.9 41.3 23.7 42.4 41.9 42.5
19.4 18.6 25.7 19.7 18.2 21.6
A A B B B
Transmission electron microscopy Transmission electron microscopy (TEM) images of reduced catalysts were recorded using an accelerating voltage of 200 kV (JEOC-2010F) equipped with energy-dispersive X-ray (EDX) facility. Before TEM analysis, catalysts were reduced in a stream of H2 at 300 C for 2 h, the reduced powders were poured into alcohol under the protection of H2 and suspended with an ultrasonic dispersion for 15 min, and then the resulted solution was dropped on carbon film of nickel grid.
Catalytic test The catalytic activity tests were carried out in a laboratory fixed-bed reactor operated in the down-flow mode. The temperature of the reactor was controlled using a PID controller. The temperature of the catalyst bed was measured using a NieCreNi thermocouple. Isothermal operating conditions were maintained by a cooling coil inside the reactor and an electrical heating jacket covering the outer side of the reactor wall. The pipings of the system were heated with three independent heating sets. The reactant was ethyl acetate (100%), which was transported into the reactor with a micro-syringe pump. For each experiment, 0.9 g of unreduced catalyst (20e40 mesh) was loaded in the middle of the reactor, with the spare spaces were filled with quartz sand (40e60 mesh). The catalyst was reduced at 300 C and a pressure of 1 MPa with H2 (flow rate 30 cm3/min) for 2 h. The reactions were carried out at a reaction temperature of 523 K, a reaction pressure of 2 Mpa, and a WHSV of 2 h1 with g (molar ratio of hydrogen to ester) ¼ 15. The products were analyzed using an Agilent 6890 N gas chromatograph (flame ionization detector, HP-5 column, 30 m 5 mm 0.25 mm) and confirmed by gas GCeMS.
It is clear from Table 1 that the metal loading found by ICPAES is close to the theoretically expected values for all catalysts, except for Cu2Zn1Si1e(NH4)2CO3 catalyst, that show 25.7 wt% zinc loading and 23.7 wt% copper loading, which is considerably lower than the theoretically expected 42.5 wt%. The low copper loading measured for Cu2Zn1Si1e(NH4)2CO3 catalyst is due to the formation of copper complex (Cu(NH3)2þ 4 ) [16] in the process of coprecipitation, which is soluble in water. In view of the low copper weight content in the catalyst, it could be expected that the activity of Cu2Zn1Si1e(NH4)2CO3 catalyst would be low. Indeed, the experimental data (see Table 5) proves it. Hence, it reveals that (NH4)2CO3 is not a suitable precipitant for copper ions and Cu2Zn1Si1e(NH4)2CO3 catalyst is not worth studying.
XRD characterization Fig. 1 compares the XRD patterns of Cu2Zn1Si1eNaOH, Cu2Zn1Si1eKOH, Cu2Zn1Si1eNaHCO3 and Cu2Zn1Si1eNa2CO3 that have the same Cu/Zn/Si molar ratio (Cu/Zn/Si ¼ 2/1/1). Note that no SiO2 is detected in all samples, which may due to its amorphous phase or high dispersion [12]. Interestingly, the XRD patterns of reduced Cu2Zn1Si1eNaOH catalyst and Cu2Zn1Si1eKOH catalyst both display distinct reflection peaks which could be attributed to ZnO, whereas for the reduced
Results and discussion The metal loading in the calcined catalysts The metal loading in the calcined catalysts, as determined using ICP-AES, is shown in Table 1. The theoretically expected value of metal loading (CTV) was calculated according to Eq. (3). CTV ¼
2*MCu or 1*MZn 2*MCuO þ 1*MZnO þ 1*MSiO2
(3)
With: MCu or MZn ¼ relative atomic mass ¼ 64 (g/mol) or 65 (g/mol), MCuO ¼ 80 (g/mol), MZnO ¼ 81 (g/mol), MSiO2 ¼ 60 (g/ mol).
Fig. 1 e X-ray diffraction (XRD) pattern of the samples reduced. (a) Cu2Zn1Si1eNaOH, (b) Cu2Zn1Si1eKOH, (c) Cu2Zn1Si1eNaHCO3 and (d) Cu2Zn1Si1eNa2CO3.
10954
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
Cu2Zn1Si1eNaHCO3 catalyst and Cu2Zn1Si1eNa2CO3 catalyst, the XRD ZnO peaks disappear partially. This result means that ZnO is better dispersed in latter two catalysts than the first two. The characteristic peaks of Cu0 at 43.3 , 50.4 and 74.1 are detected in all samples, which are ascribed to diffraction of (1 1 1), (2 0 0) and (2 2 0) planes. The intensity of characteristic reflection peaks of Cu0 for the samples is quite different. The peak intensity may reflect the crystal size and induce of active phase dispersion. It has been reported by van der Grift et al. [17,18]. Thus, the average crystallite size of copper particles is calculated according to the (1 1 1) diffraction peak at 2q ¼ 43.3 , using the Scherrer equation: D ¼ 0.89l/(bcos q), where l is the x-ray wavelength (0.154 nm) and b is the full width at halfmaximum (FWHM) of the (1 1 1) line. Table 2 displays data about the average crystallite size of copper in the reduced samples. It can be noted that the size of the copper crystallite size is apparently dependent on the type of samples, as the mean particle sizes of type A catalysts are around 25 nm, which is much larger than 19 nm for the type B catalysts. In other words, the type B catalysts achieve better dispersion of active species than that of type A catalysts. However, the difference of average copper metal size between the catalysts in the same type is not obvious. In view of this, the Cu2Zn1Si1eNaOH catalyst and Cu2Zn1Si1eNa2CO3 catalyst in the text that follows are chose as the representative for type A catalysts and type B catalysts, respectively, so as to make a better investigation on the effect of precipitants.
TEM analysis The TEM photographs of reduced Cu2Zn1Si1eNaOH and Cu2Zn1Si1eNa2CO3 catalysts are shown in Fig. 2. In the TEM images, dark particles are dispersed on light-gray particles. The former are deduced to be copper particles and the later as oxide of zinc or silica with the aid of EDX analysis, which is in agreement with a previous study [19]. Moreover, EDX analysis of the samples (several spots were analyzed) confirms that the compositions of the catalysts are similar as their bulk content mentioned above. Comparison of the results obtained for the two catalysts suggests more densely packed copper particles in Cu2Zn1Si1eNaOH than in Cu2Zn1Si1eNa2CO3, which confirms that copper particles are better dispersed in Cu2Zn1Si1eNa2CO3 catalyst. A lot of separated copper particles sized in 20e30 and 10e20 nm are detected in the TEM images of reduced Cu2Zn1Si1eNaOH and Cu2Zn1Si1eNa2CO3. Furthermore, the average particle size of Cu detected in high resolution TEM images is calculated as: P ni d3i dTEM ¼ P ni d2i
(4)
Table 2 e Cu particle size calculated by XRD in the reduced catalysts. Type Type Type Type Type
A A B B
Sample Cu2Zn1Si1eNaOH Cu2Zn1Si1eKOH Cu2Zn1Si1eNaHCO3 Cu2Zn1Si1eNa2CO3
FWHM (rad) 3
6.10*10 5.76*103 7.74*103 7.78*103
Average Cu particle size (nm) 24.1 25.6 19.1 18.9
Fig. 2 e TEM images of the reduced catalysts. (a) Cu2Zn1Si1eNaOH, (b) Cu2Zn1Si1eNa2CO3.
where ni is the number of particles having a characteristic diameter di. The calculated copper particle sizes in reduced Cu2Zn1Si1eNaOH catalyst and Cu2Zn1Si1eNa2CO3 catalyst in the HRTEM images are 23 nm and 17 nm, respectively.
Catalyst texture The information about the texture of the calcined catalysts is shown in Table 3. The N2 adsorptionedesorption isotherms of Cu2Zn1Si1eNaOH catalyst and Cu2Zn1Si1eNa2CO3 catalyst are shown in Fig. 3. As evidenced by the shape of the isotherms and by data reported in Table 3, the precipitant used in the process of catalyst preparation has a drastic influence on textural properties of different samples. According to the IUPAC classification, Cu2Zn1Si1eNa2CO3 catalyst exhibit type Ⅳ isotherms with a regular H1-type hysteresis loop at relative pressures (p/p0) of 0.72e1.0 due to the N2 capillary condensation in cylindrical channels [20,21]. Cu2Zn1Si1eNaOH catalyst show the same type of N2-sorption isotherms with a higher H1-type hysteresis loop at relative pressures (p/p0) of 0.6e1.0 but a wider interval between two closed points, indicating that this sample has more regular cylindrical channels and larger pore size [22].
Table 3 e Texture of the calcined samples. Samples Cu2Zn1Si1eNaOH Cu2Zn1Si1eNa2CO3 a b c
SBET (m2/g)a
TPVb (cm3/g)c
Mean pore size (nm)c
63.6 80.4
0.38 0.29
21.9 17.0
BET surface area. Total pore volume. Calculated based on desorption branch of isotherm.
10955
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
dispersed copper oxide species on outside surface of catalyst [24] and the second peak (at around 210 C) can be assigned as reduction of Cu2þ ions reduced with more difficulty [23,25]. Furthermore, the metal reduction degrees calculated from the hydrogen consumption peak area in the temperature region from 50 to 300 C by H2-TPR for Cu2Zn1Si1eNaOH and Cu2Zn1Si1eNa2CO3 are 98.2% and 96.9%, respectively. This result implies either the presence of some Cuþ in the samples treated at 300 C or a not complete reduction of Cu2þ. Therefore, more information about metal species in the surface layer of the reduced samples is obtained from XPS experiments in the next section.
3 Adsorbed Volume/cm /g STP
Full symbols: adsorption Empty symbols: desorption
Cu2Zn1Si1- NaOH
Cu2Zn1Si1- Na2CO3
0.0
0.2
0.4
0.6
0.8
1.0
0
Relative Pressure/(P/P )
Fig. 3 e Nitrogen adsorptionedesorption isotherms. The BET surface and pore volume of Cu2Zn1Si1eNaOH catalyst determined by the BarretteJoynereHalenda (BJH) method are 63.6 m2/g and 0.38 cm3/g, respectively (Table 3) and the average pore diameter of the catalyst is 21.9 nm. On the other hand, Cu2Zn1Si1eNa2CO3 catalyst possesses surface area of 80.4 m2/g (Table 3), smaller pore volume (0.29 cm3/g) and mean pore size (17.0 nm).
H2-TPR characterization Fig. 4 illustrates H2-TPR profiles of calcined catalysts prepared by two types of precipitants. An asymmetric peak at 210 C with a distinct shoulder peak in the lower temperature region is detected for Cu2Zn1Si1eNa2CO3 catalyst, indicating that different kinds of Cu species formed in the catalysts [23]. On the other hand, reduction of Cu2Zn1Si1eNaOH catalyst also proceeds in two steps with a sharp maximum at 210 C and a not obvious shoulder at a slightly lower temperature. The first hydrogen consumption peak observed in two samples (at around 170 C) can be ascribed to the reduction of highly
Surface analysis XPS has provided additional information about copper phases in the reduced catalysts. XPS and X-ray induced Auger spectra (XAES) of the reduced samples are shown in Figs. 5 and 6, respectively. The whole Cu2p region comprises two components, 2p3/2 and 2p1/2, due to the spin-orbit splitting for orbital having l > 1. These two components are well separated (around 20 eV) so only the Cu2p3/2 peaks can be considered for the assignment of copper chemical states [26]. In Fig. 5, the catalysts display the peaks at ca. 934.8 ± 0.1 eV, which are assigned to Cu2þ. Moreover, the presence of the characteristic shakeup 2p / 3d satellite peaks between 940 and 945 eV also suggests the existence of Cu2þ species in the two catalysts [27]. Since the binding energy values for Cuþ and Cu0 are almost identical, the peaks at 932.5 eV are assigned to 2p3/2 of copper in the form of Cu0 and/or Cuþ [28]. Thus, the distinction between these two species on the catalyst surface needs a modified Auger parameter, a0 , defined as Cu2p3/2 (BE) þ Cu LMM (KE) [29]. The deconvolution of the original Cu LMM peaks is thus carried out and the peak positions as well as their contributions are extracted from the deconvolution and are listed in Table 4. The a0 value at ca. 1851.0 eV is due to Cu0 and that at ca. 1847.0 eV to Cuþ according to reference [30]. The results (Table 4) indicate that both Cu0þ and Cuþ coexist on the surface of the reduced Cu2Zn1Si1eNa2CO3 catalyst, while no Cuþ species is observed in the Cu2Zn1Si1eNaOH
Cu2p region
Cu2Zn1Si1- NaOH Cu2Zn1Si1-Na2CO3
Cu2p3/2
Cu2p1/2
932.5
934.8 2+
(b)
Intensity (a.u.)
Intensity (u.a.)
Cu shake up
(a) 0
100
200
300
400
500
Temperature (ºC)
Fig. 4 e H2-TPR profiles of the calcined samples. (a) Cu2Zn1Si1eNaOH and (b) Cu2Zn1Si1eNa2CO3.
600
965
960
955
950
945
940
935
930
Binding Energy (eV)
Fig. 5 e Cu2p XPS spectra of reduced samples. (a) Cu2Zn1Si1eNaOH, (b) Cu2Zn1Si1eNa2CO3.
925
10956
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
CuLMM
Table 5 e Catalytic performances of catalysts prepared by different precipitants. Type
Catalysta
Type Type Type Type Type
Cu2Zn1Si1eNaOH Cu2Zn1Si1eKOH Cu2Zn1Si1e(NH4)2CO3 Cu2Zn1Si1eNaHCO3 Cu2Zn1Si1eNa2CO3
Intensity (a.u.)
(b) 914.4
918.7
915.8
(a) 912
914
916
918
920
922
924
926
A A B B B
Ethyl acetate Ethanol Ethanol conversion selectivity Yieldb (%) (%) (%) 83.20 78.96 69.41 88.15 90.48
98.63 99.24 99.03 98.94 98.95
82.06 78.36 68.74 87.22 89.53
Catalyst 0.9 g, in situ reduction in H2 at 300 C for 2 h before reaction. b Reaction conditions: Temp. ¼ 250 C, pressure ¼ 2 Mp, g (molar ratio of hydrogen to ester) ¼ 15 and 2 h1 WHSV of ethyl acetate. a
Kinetic Energy (eV)
Fig. 6 e XAES spectra of different catalysts after reduction. (a) Cu2Zn1Si1eNaOH, (b) Cu2Zn1Si1eNa2CO3.
catalyst. Compared with bulk value, the smaller a0 values calculated for Cuþ (bulk value is 1848.8 [31]) and Cu2þ (bulk value is 1852.0 [31]) are attributed to the well dispersed copper species and a strong interaction between species and the support. It has been reported that when copper (in þ2, þ1, and 0 valence) is in the highly dispersed state and in intimate contact with the support, a0 can be 2e3 eV lower than the bulk value [32,33].
Catalytic performance The catalytic data of the catalysts prepared by different precipitants are shown in Table 5. As expected, all the catalysts show an excellent selectivity towards ethanol over the whole conversion range in the hydrogenolysis of ethyl acetate, which has been described in the report before [5]. Nevertheless, the experimental data of acetate conversion reveal that the precipitants lead to a drastic influence on the catalytic activity of the catalysts for ethyl acetate hydrogenolysis reaction. Indeed, the Cu2Zn1Si1e(NH4)2CO3 catalyst exhibits low activity for the reason of the low copper loading measured by ICP-AES (Table 1), which is due to the formation of copper complex (Cu(NH3)2þ 4 ). Additionally, the activity of type B
Table 4 e XPS parameters of the reduced samples. Samples Cu2Zn1Si1eNaOH Cu2Zn1Si1eNa2CO3
KEb (eV)
(eV)
932.5 934.9 932.5
918.9 915.7 918.5 914.4 916.0.
1851.4 1850.6 1851.0 1846.9 1850.7
934.7 a
a0 c
BEa (eV)
Assignment Xd (%) Cu0 Cu2þ Cu0 Cuþ Cu2þ
77.4 22.6 71.3 6.7 22.0
Binding Energy. b Kinetic Energy. c a0 ¼ BE þ KE. d Intensity ratio of Cu0 (or Cu2þ, Cuþ)/(Cu0 þ Cuþ þ Cu2þ) by deconvolution of CuLMM XAES spectra.
catalysts, except Cu2Zn1Si1e(NH4)2CO3, is much higher than that of type A catalysts, and the comparison between the catalysts in the same type (except Cu2Zn1Si1e(NH4)2CO3) reveals that the precipitants also shows some influence on the performance of the catalysts, while the influence is not remarkable. As mentioned above, it is observed that the activity of type B catalysts, except Cu2Zn1Si1e(NH4)2CO3, is much higher than that of type A catalysts. Hence, more attention is focused on the impact of the different types of precipitants on the copperbased catalysts. As the representative for type B catalysts, Cu2Zn1Si1eNa2CO3 catalyst possesses smaller copper particles than that of Cu2Zn1Si1eNaOH, which is confirmed by XRD (Fig. 1) and TEM (Fig. 2) results, indicating that the catalysts with better dispersed active species (i.e., a smaller Cu particle) show higher acetate conversion levels. Namely, the conversion of ethyl acetate depends strongly on the particle size of copper, smaller sized copper is extremely active [34]. Moreover, with respect to the texture of the samples (Table 3), larger BET surface of Cu2Zn1Si1eNa2CO3 is observed. In many reports [35,36], large BET surface is of great importance for the design of efficient catalysts, which is an important reason for the high activity of Cu2Zn1Si1eNa2CO3 catalyst. From the XPS and X-ray induced Auger spectra of reduced Cu2Zn1Si1eNaOH catalyst and Cu2Zn1Si1eNa2CO3 catalyst, the conclusion obtained is that well dispersed Cu2þ species (around 22%) and Cu0 species (>71%) both exist on the surface of the two catalysts, while only the Cu2Zn1Si1eNa2CO3 catalyst shows the presence of highly dispersed Cuþ species (6.7%). Despite the high metal reduction degrees (by H2-TPR) for the two samples reduced at 300 C for 2 h, some unreduced Cu2þ exist in the surface layer, which may have negative influence on the catalyst activity. With respect to Cu0 species and Cuþ species, it has been suggested that only Cu0 acts as the active site in ester hydrogenation [37], and this is also evidenced by the experimental date, from which it is observed that Cu2Zn1Si1eNaOH catalyst, though no Cuþ exists on the surface, shows the conversion of ethyl acetate 83.2% and high selectivity to ethanol. The existence of Cuþ for the reduced catalysts signifies the stronger interaction between copper and silica [32]. Moreover, Cuþ may function as electrophilic or Lewis acid sites to polarize the C]O bond via the electron lone pair on oxygen [30], thus improving the reactivity of the ester group in ethyl acetate. The cooperative effect between Cu0 and
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
Cuþ is responsible for the higher hydrogenation activity of the Cu2Zn1Si1eNa2CO3 catalyst.
Conclusion The present work demonstrated that the precipitants used in the coprecipitation method have profound effects on the texture, composition, and structure of the calcined Cu/ZnO/ SiO2 and reduced Cu/ZnO/SiO2 catalysts. (NH4)2CO3 can lead to the formation of copper complex (Cu(NH3)2þ 4 ), indicating that it is not a suitable precipitant for Cu ions. The different types of precipitants (except (NH4)2CO3), show a drastic influence on the performance of copper-based catalysts, whereas the difference of catalysts in the same type is not obvious. The type B catalysts (except Cu2Zn1Si1e(NH4)2CO3) possess smaller copper particles and larger BET surface than that of type A catalysts, which is the main reason for the higher activity of type B catalysts. Furthermore, the cooperative effect between Cu0 and Cuþ can lead to higher hydrogenation activity. On the Cu/ ZnO/SiO2 catalyst prepared by NaCO3, an ethanol yield of 89.53% was obtained under the hydrogenation conditions.
references
[1] Luchansky MS, Monks J. Supply and demand elasticities in the U.S. ethanol fuel market. Energy Econ 2009;31:403e10. [2] Adkins H, Folkers K. The catalytic hydrogenation of esters to alcohols. J Am Chem Soc 1931;53:1095e7. [3] Yan TY, Albright LF, Case LC. Hydrogenolysis of esters, particularly perfluorinated esters. Ind Eng Chem Prod Res Dev 1965;4:101e7. [4] Evans JW, Wainwright MS, Cant NW, Trimm DL. Structural and reactivity effects in the copper-catalyzed hydrogenolysis of aliphatic esters. J Catal 1984;88:203e13. [5] Claus P, Lucas M, Lucke B. Selective hydrogenolysis of methyl and ethyl acetate in the gas phase on copper and supported group Ⅷ metal catalysts. Appl Catal A 1991;79:1e18. [6] Kohler MA, Cant NW, Wainwright MS, Trimm DL. The mechanism of the catalytic chemistry of ester hydrogenolyisis on copper surfaces. Proc Int Congr Catal 1988;9:1043e50. [7] Kenvin JC, White MG. Supported catalysts from mononuclear copper complexes: catalytic properties. J Catal 1992;135:81e91. [8] Evans JW, Cant NW, Trimm DL, Wainwright MS. Hydrogenolysis of ethyl formate over copper-based catalysts. Appl Catal 1983;6:355e62. [9] Brands DS, Poels EK, Bliek A. Ester hydrogenolysis over promoted Cu/SiO2 catalysts. Appl Catal A 1999;184:279e89. [10] Kim SM, Lee ME, Choi JW, Suh DJ, Suh YW. The role of ZnO in Cu/ZnO/Al2O3 catalyst for hydrogenolysis of butyl butyrate. Catal Commun 2011;12:1328e32. [11] van de Scheur FT, Staal LH. Effects of zinc addition to silica supported copper catalysts for the hydrogenolysis of esters. Appl Catal A 1994;108:63e83. [12] Yuan P, Liu ZY, Zhang WQ, Sun HJ, Liu SC. Cu-Zn/Al2O3 catalyst for the hydrogenation of esters to alcohols. Chin J Catal 2010;31:769e75. [13] Huang H, Cao GP, Fan CL, Wang SH, Wang SJ. Effect of water on Cu/Zn catalyst for hydrogenation of fatty methyl ester to fatty alcohol. Korean J Chem Eng 2009;26:1574e9.
10957
[14] Chen HM, Zhu YL, Ding GQ, Zheng HY, Li YW. Study on hydrogenation of dimethyl oxalate to ethylene glycol. J Fuel Chem Tech 2011;39:519e26. [15] Barrett EP, Joyner LG, Halenda PP. The determination of pore volume and area distributions in porous substances. J Am Chem Soc 1951;73:373e80. [16] Holland EB, Gilligan GM. Solubility of copper sulphate, malachite, and burgundy precipitate in ammonium hydroxide. J Phys Chem 1927;31:728e41. [17] van der Grift CJG, Wielers AFH, Mulder A, Geus JW. The reduction behaviour of silica-supported copper catalysts prepared by deposition-precipitation. Thermochim Acta 1990;171:95e113. [18] van der Grift CJG, Wielers AFH, Mulder A, Geus JW. Characterization of silica-supported copper catalysts by means of temperature-programmed reduction. Appl Catal 1990;60:181e92. [19] Chinchen GC, Waugh KC, Whan DA. The activity and state of the state of the copper surface in methanol synthesis catalysts. Appl Catal 1986;25:101e7. [20] Groen JC, Peffer LAA, Ramirez JP. Pore size determination in modified micro- and mesoporous materials. Pitfalls and limitations in gas adsorption data analysis. Microporous Mesoporous Mater 2003;60:1e17. [21] Li CL, Zhang QH, Wang Y, Wan HL. Preparation, characterization and catalytic activity of palladium nanoparticles encapsulated in SBA-15. Catal Lett 2008;120:126e36. [22] Wang JJ, Chernavskii PA, Wang Y, Khodakov AY. Influence of the support and promotion on the structure and catalytic performance of copper-cobalt catalysts for carbon monoxide hydrogenation. Fuel 2013;103:1111e22. [23] Yuan ZL, Wang LN, Wang JH, Xia SX, Chen P, Hou ZY, et al. Hydrogenolysis of glycerol over homogenously dispersed copper on solid based catalysts. Appl Catal B 2011;101:431e40. [24] Lin L, Pan PB, Zhou ZF, Li ZJ, Yang JX, Sun ML, et al. Cu/SiO2 catalysts prepared by the sol-gel method for hydrogenation of dimethyl oxalate to ethylene glycol. Chin J Catal 2011;32:967e9. [25] Esposito S, Turco M, Bagnasco G, Cammarano C, Pernice P. New insight into the preparation of copper/zirconia catalysts by sol-gel method. Appl Catal A 2011;403:128e35. [26] Salvi AM, Langerame F, Macchia A, Sammartino MP, Tabasso ML. XPS characterization of (copper-based) coloured stains formed on limestone surfaces of outdoor Roman monuments. Chem Cent J 2012;6:1e13. [27] Sun KP, Lu WW, Qiu FY, Liu SW, Xu XL. Direct synthesis of DME over bifunctional catalyst: surface properties and catalytic performance. Appl Catal A 2003;252:243e9. [28] Kaddouri A, Dunpont N, Gelin P, Delichere P. Methane combustion over copper chromites catalysts prepared by the sol-gel process. Catal Lett 2011;141:1581e9. [29] Moretti G. Auger parameter and Wagner plot in the characterization of chemical states: initial and final state effects. J Electron Spectrosc 1995;76:365e70. [30] Chen LF, Guo PJ, Qiao MH, Yan SR, Li HX, Shen W, et al. Cu/ SiO2 catalysts prepared by the ammonia-evaporation method: texture, structure, and catalytic performance in hydrogenation of dimethyl oxalate to ethylene glycol. J Catal 2008;257:172e80. [31] Fantauzzi M, Atzei D, Elsener B, Lattanzi P, Rossi A. XPS and XAES analysis of copper, arsenic and sulfur chemical state in energies. Surf Interface Anal 2006;38:922e30. [32] Yin AY, Guo XY, Dai WL, Fan KL. Effect of initial precipitation temperature on the structural evolution and catalytic behavior of Cu/SiO2 catalyst in the hydrogenation of dimethyloxalate. Catal Commun 2011;12:412e6.
10958
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
[33] Kohler MA, Curry-Hyde HE, Hughes AE, Sexton BA, Cant NW. The structure of Cu/SiO2 catalysts prepared by the ionexchange technique. J Catal 1987;108:323e33. [34] Zhu YM, Shi L. Zn promoted Cu-Al catalyst for hydrogenation of ethyl acetate to alcohol. J Ind Eng Chem; 2013. http:// dx.doi.org/10.1016/j.jiec.2013.10.010. [35] Tang XL, Meng X, Shi L. Desulfurization of model gasoline on modified bentonite. Ind Eng Chem Res 2011;50:7527e33.
[36] Huang H, Yi DZ, Lu YN, Wu XL, Bai YP, Meng X, et al. Study on the adsorption behavior and mechanism of dimethyl sulfide on silver modified bentonite by in FTIR and temperatureprogrammed desorption. Chem Eng J 2013;225:447e55. [37] Mokhtar M, Ohlinger C, Schlander JH, Turek T. Hydrogenolysis of dimethyl maleate on Cu/ZnO/Al2O3 catalysts. Chem Eng Technol 2001;24:423e6.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
The influence of different precipitants on the copper-based catalysts for hydrogenation of ethyl acetate to ethanol Kaili Zhong, Xin Wang* The State Key Laboratory of Chemical Engineering, East China University of Science and Technology, Shanghai 200237, China
article info
abstract
Article history:
Ethanol fuel has become a hot topic in the economic, political, environmental, and sci-
Received 8 March 2014
entific areas. In this work, a new way for the synthesis of ethanol by hydrogenolysis of
Received in revised form
ethyl acetate is introduced and the impact of different precipitants on the ethyl acetate
6 May 2014
hydrogenolysis catalysts is systematically investigated by several considerations, including
Accepted 9 May 2014
dispersion effects, the texture of the catalysts, and the copper phases in the surface layer of
Available online 9 June 2014
the reduced catalysts, etc. These precursors and catalysts are characterized by inductively coupled plasma-atomic emission spectroscopy, N2-adsoption, X-ray diffraction, trans-
Keywords:
mission electronic microscope, H2 temperature-programmed reduction and X-ray photo-
Ethanol fuel
electron spectroscopy. It is confirmed that the choice of precipitant is of great importance.
Precipitant
The samples are classified into two types, depending on the anion of precipitant. Except
Hydrogenolysis
the catalyst prepared by (NH4)2CO3, in which low copper loading is observed, type B cata-
Ethyl acetate
lysts (eCO2 3 ) possess smaller copper particles and larger BET surface than that of type A
Cu/ZnO/SiO2
catalysts (eOH), while the difference of catalysts in the same type is not obvious. Moreover, the coexistence of Cuþ and Cu0 is only detected in reduced type B catalysts. In general, ethyl acetate hydrogenolysis activity varies considerably with the precipitant, in the following order: Na2CO3 NaHCO3 > NaOH KOH > (NH4)2CO3. Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
Introduction As a potential alternative to oil, ethanol fuel has become a hot topic in the economic, political, environmental, and scientific areas [1]. The synthesis of ethanol by hydrogenolysis of ethyl acetate is of great commercial potential, since ethyl acetate can be produced at lower cost and on a large scale.
The catalytic hydrogenesis of esters to their corresponding alcohols [Eq. (1)] has been extensively studied since the first report by Flokers and Adkins [2]. RCOOR0 þ 2H2 /RCH2 OH þ R0 OH
(1)
A mechanism for acetate hydrogenolysis was proposed by Yan et al. [3] and Evans et al. [4] The acetate is adsorbed dissociatively [Eq. (2)]
* Corresponding author. Tel.: þ86 021 64252836. E-mail address: [email protected] (X. Wang). http://dx.doi.org/10.1016/j.ijhydene.2014.05.045 0360-3199/Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
10952
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
RCOOR0 þ 2* /RCO* þ R0 O*
(2)
where * indicates a catalytic site on the surface. Isotopic labeling studies showed that the alkoxy fragment R0 O* reacts quickly to form R0 OH, while the acyl group RCO* is adsorbed longer. Thus the hydrogenation of the acyl fragment is believed to be the rate-determining step during acetate hydrogenolysis. The acyl group can be hydrogenated either to the desired alcohol or to the corresponding aldehyde, which is subsequently hydrogenated [3,5]. An in situ infrared spectroscopy investigation [6] during ethyl acetate hydrogenolysis gained further evidence for the proposed mechanism via dissociatively adsorbed fragments. Furthermore, Kenvin and White [7] could explain their results based on the mechanism proposed by Evans et al. For ester hydrogenolysis, copper-based catalysts including copper chromite, silica-supported copper and Raney copper are well known to exhibit high activities [5,8,9], and the experimental results strongly indicate that copper is the essential catalytic component in the ester hydrogenolysis reaction [8]. Moreover, the addition of metal oxides, such as Cr2O3, ZrO2, Fe2O3 or ZnO, etc. has been extensively investigated to promote the activity of Cu in the hydrogenolysis reaction [10]. When both activity and selectivity were taken into account, Brands [9] and van de Scheur [11] concluded that ZnO was the promoter of choice. It has been reported [10,11] that the addition of ZnO into copper-based catalyst can lead to a higher Cu dispersion (i.e., a smaller Cu particle), and ZnO is believed to act as a structural promoter in enhancing the area of Cu, the active site for the reaction. According to references [5,10,12e14], coprecipitation method is a conventional preparation method for ester hydrogenolysis catalysts, thus, the choice of precipitants is of great importance. Kim et al. [10] and Yuan et al. [12] chose NaHCO3 and Na2CO3 as the precipitant in the references, respectively, while Chen et al. [14] reported that the Cu/SiO2 catalysts precipitated by NaOH showed high activity in hydrogenation of ester. In general, the impact of different precipitants on the performance of ethyl acetate hydrogenolysis catalysts has never systematically investigated. In this work, the silica supported CueZn catalysts are prepared via coprecipitation method for hydrogenation of ethyl acetate, and the main aim of this study is to have a better view of the influence of different precipitants on the catalysts by several considerations, including dispersion effects, metal loading, the texture of the catalysts, and the copper phases in the surface layer of the reduced catalysts, etc. The characterization results are discussed alongside with the catalytic data obtained using a laboratory fixed-bed reactor.
Experimental Catalyst preparation Hydrogenation components CueZn/SiO2 (Cu/Zn/SiO2 molar ratio ¼ 2:1:1) were prepared by coprecipitation method. In a round bottom flask, appropriate amount of silica sol (Sinopharm Chemical Reagent Co., Ltd., China, 30%) was mixed with distilled water under vigorous stirring. Required
amounts of Cu(NO3)2$5H2O and Zn(NO3)2$4H2O (Sinopharm Chemical Reagent Co., Ltd., China, AR) were dissolved in 200 ml distilled water at room temperature, and this solution was referred as A, in which Cu2þ/Zn2þ ¼ 2 and Cu2þ þ Zn2þ ¼ 0.5 mol/L. Solution B (200 ml) was an aqueous of the precipitant [NaOH, KOH, Na2CO3, NaHCO3 or (NH4)2CO3 (Sinopharm Chemical Reagent Co., Ltd., China, AR)]. Then solution A and solution B were simultaneously added dropwise (in around 1 h) to the round bottom flask with vigorously stirring in the ambient temperature, the pH was maintained at 8.0 during dripping process by controlling the flow velocity of solution A and solution B. After completing the process of precipitation, the resulted suspension was stirred for 30 min, filtered and washed five times with distilled water until the pH of filtrate reached 7.0. The precipitate was dried in an oven at 120 C overnight and then calcined in a muffle furnace in air by heating at 500 C for 5 h. Finally, the resulting catalysts were crushed and sieved. The 20e40 mesh particles were used for the activity test. The prepared catalysts were labeled as Cu2Zn1Si1eM, where M stands for the precipitant used in the process of catalysts preparation. Herein, the samples are classified into two types, depending on the anion of precipitant. The Cu2Zn1Si1eNaOH catalyst and Cu2Zn1Si1eKOH catalyst are referred as type A, and the other three are referred as type B.
Characterization of catalyst Inductively coupled plasma-atomic emission spectroscopy Actual metal loading in the calcined catalysts was determined with inductively coupled plasma-atomic emission spectroscopy (ICP-AES) using a multichannel Thermo Jarrel-Ash ICAP 612 E spectrometer.
N2-adsorption Nitrogen isotherms were measured at 196 C with an ASAP 2020 (Micromeritrics). Before experiment, the samples were heated at 120 C and outgassed overnight at this temperature under a vacuum of 105 Torr to a constant pressure. N2 isotherms were obtained in both adsorption and desorption modes. The surface areas of the calcined catalysts were determined by the BET method. The total pore volume (TPV) was calculated from the amount of vapor adsorbed at a relative pressure (p/p0) close to unity (desorption curve), where p and p0 are the measured and equilibrium pressures, respectively. Pore size distribution curves were established from the desorption branches of the isotherms using the BJH model [15].
X-ray diffraction X-ray diffraction (XRD) technique was used to characterize the crystal structure. The experiments were performed using a Siemens D500 diffractometer, using Cu Ka radiation (40 kV, 100 mA) and equipped with a graphite monochromator with a reflected beam. The powder diffraction patterns were recorded in the 2q range from 20 to 80 . An X-ray chamber-reactor was used for the high-temperature registration of the spectra.
Temperature-programmed reduction Temperature programmed reduction (TPR) of H2 was carried out on Auto Chem2910 (Micromeritics) instrument to study
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
the reducibility of the catalysts. In a typical experiment, around 0.1 g of calcined catalyst was exposed to a reducing gas consisting of 5.0 vol% H2 in argon with a temperature ramp from ambient temperature to 600 C at a heating rate of 10 C/ min.
X-ray photoelectron spectroscopy X-ray photoelectron spectroscopy (XPS) and X-ray induced Auger electron spectroscopy (XAES) were obtained using a VG MultiLab 2000 analyzer with a Mg Ka radiation source and the binding energies were referenced to C1s (284.6 eV).
10953
Table 1 e Metal loading determined using ICP-AES in the calcined catalysts. Type
Catalysts
Measured copper loading (wt%)
Measured zinc loading (wt%)
Type Type Type Type Type e
Cu2Zn1Si1eNaOH Cu2Zn1Si1eKOH Cu2Zn1Si1e(NH4)2CO3 Cu2Zn1Si1eNaHCO3 Cu2Zn1Si1eNa2CO3 CTV
41.9 41.3 23.7 42.4 41.9 42.5
19.4 18.6 25.7 19.7 18.2 21.6
A A B B B
Transmission electron microscopy Transmission electron microscopy (TEM) images of reduced catalysts were recorded using an accelerating voltage of 200 kV (JEOC-2010F) equipped with energy-dispersive X-ray (EDX) facility. Before TEM analysis, catalysts were reduced in a stream of H2 at 300 C for 2 h, the reduced powders were poured into alcohol under the protection of H2 and suspended with an ultrasonic dispersion for 15 min, and then the resulted solution was dropped on carbon film of nickel grid.
Catalytic test The catalytic activity tests were carried out in a laboratory fixed-bed reactor operated in the down-flow mode. The temperature of the reactor was controlled using a PID controller. The temperature of the catalyst bed was measured using a NieCreNi thermocouple. Isothermal operating conditions were maintained by a cooling coil inside the reactor and an electrical heating jacket covering the outer side of the reactor wall. The pipings of the system were heated with three independent heating sets. The reactant was ethyl acetate (100%), which was transported into the reactor with a micro-syringe pump. For each experiment, 0.9 g of unreduced catalyst (20e40 mesh) was loaded in the middle of the reactor, with the spare spaces were filled with quartz sand (40e60 mesh). The catalyst was reduced at 300 C and a pressure of 1 MPa with H2 (flow rate 30 cm3/min) for 2 h. The reactions were carried out at a reaction temperature of 523 K, a reaction pressure of 2 Mpa, and a WHSV of 2 h1 with g (molar ratio of hydrogen to ester) ¼ 15. The products were analyzed using an Agilent 6890 N gas chromatograph (flame ionization detector, HP-5 column, 30 m 5 mm 0.25 mm) and confirmed by gas GCeMS.
It is clear from Table 1 that the metal loading found by ICPAES is close to the theoretically expected values for all catalysts, except for Cu2Zn1Si1e(NH4)2CO3 catalyst, that show 25.7 wt% zinc loading and 23.7 wt% copper loading, which is considerably lower than the theoretically expected 42.5 wt%. The low copper loading measured for Cu2Zn1Si1e(NH4)2CO3 catalyst is due to the formation of copper complex (Cu(NH3)2þ 4 ) [16] in the process of coprecipitation, which is soluble in water. In view of the low copper weight content in the catalyst, it could be expected that the activity of Cu2Zn1Si1e(NH4)2CO3 catalyst would be low. Indeed, the experimental data (see Table 5) proves it. Hence, it reveals that (NH4)2CO3 is not a suitable precipitant for copper ions and Cu2Zn1Si1e(NH4)2CO3 catalyst is not worth studying.
XRD characterization Fig. 1 compares the XRD patterns of Cu2Zn1Si1eNaOH, Cu2Zn1Si1eKOH, Cu2Zn1Si1eNaHCO3 and Cu2Zn1Si1eNa2CO3 that have the same Cu/Zn/Si molar ratio (Cu/Zn/Si ¼ 2/1/1). Note that no SiO2 is detected in all samples, which may due to its amorphous phase or high dispersion [12]. Interestingly, the XRD patterns of reduced Cu2Zn1Si1eNaOH catalyst and Cu2Zn1Si1eKOH catalyst both display distinct reflection peaks which could be attributed to ZnO, whereas for the reduced
Results and discussion The metal loading in the calcined catalysts The metal loading in the calcined catalysts, as determined using ICP-AES, is shown in Table 1. The theoretically expected value of metal loading (CTV) was calculated according to Eq. (3). CTV ¼
2*MCu or 1*MZn 2*MCuO þ 1*MZnO þ 1*MSiO2
(3)
With: MCu or MZn ¼ relative atomic mass ¼ 64 (g/mol) or 65 (g/mol), MCuO ¼ 80 (g/mol), MZnO ¼ 81 (g/mol), MSiO2 ¼ 60 (g/ mol).
Fig. 1 e X-ray diffraction (XRD) pattern of the samples reduced. (a) Cu2Zn1Si1eNaOH, (b) Cu2Zn1Si1eKOH, (c) Cu2Zn1Si1eNaHCO3 and (d) Cu2Zn1Si1eNa2CO3.
10954
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
Cu2Zn1Si1eNaHCO3 catalyst and Cu2Zn1Si1eNa2CO3 catalyst, the XRD ZnO peaks disappear partially. This result means that ZnO is better dispersed in latter two catalysts than the first two. The characteristic peaks of Cu0 at 43.3 , 50.4 and 74.1 are detected in all samples, which are ascribed to diffraction of (1 1 1), (2 0 0) and (2 2 0) planes. The intensity of characteristic reflection peaks of Cu0 for the samples is quite different. The peak intensity may reflect the crystal size and induce of active phase dispersion. It has been reported by van der Grift et al. [17,18]. Thus, the average crystallite size of copper particles is calculated according to the (1 1 1) diffraction peak at 2q ¼ 43.3 , using the Scherrer equation: D ¼ 0.89l/(bcos q), where l is the x-ray wavelength (0.154 nm) and b is the full width at halfmaximum (FWHM) of the (1 1 1) line. Table 2 displays data about the average crystallite size of copper in the reduced samples. It can be noted that the size of the copper crystallite size is apparently dependent on the type of samples, as the mean particle sizes of type A catalysts are around 25 nm, which is much larger than 19 nm for the type B catalysts. In other words, the type B catalysts achieve better dispersion of active species than that of type A catalysts. However, the difference of average copper metal size between the catalysts in the same type is not obvious. In view of this, the Cu2Zn1Si1eNaOH catalyst and Cu2Zn1Si1eNa2CO3 catalyst in the text that follows are chose as the representative for type A catalysts and type B catalysts, respectively, so as to make a better investigation on the effect of precipitants.
TEM analysis The TEM photographs of reduced Cu2Zn1Si1eNaOH and Cu2Zn1Si1eNa2CO3 catalysts are shown in Fig. 2. In the TEM images, dark particles are dispersed on light-gray particles. The former are deduced to be copper particles and the later as oxide of zinc or silica with the aid of EDX analysis, which is in agreement with a previous study [19]. Moreover, EDX analysis of the samples (several spots were analyzed) confirms that the compositions of the catalysts are similar as their bulk content mentioned above. Comparison of the results obtained for the two catalysts suggests more densely packed copper particles in Cu2Zn1Si1eNaOH than in Cu2Zn1Si1eNa2CO3, which confirms that copper particles are better dispersed in Cu2Zn1Si1eNa2CO3 catalyst. A lot of separated copper particles sized in 20e30 and 10e20 nm are detected in the TEM images of reduced Cu2Zn1Si1eNaOH and Cu2Zn1Si1eNa2CO3. Furthermore, the average particle size of Cu detected in high resolution TEM images is calculated as: P ni d3i dTEM ¼ P ni d2i
(4)
Table 2 e Cu particle size calculated by XRD in the reduced catalysts. Type Type Type Type Type
A A B B
Sample Cu2Zn1Si1eNaOH Cu2Zn1Si1eKOH Cu2Zn1Si1eNaHCO3 Cu2Zn1Si1eNa2CO3
FWHM (rad) 3
6.10*10 5.76*103 7.74*103 7.78*103
Average Cu particle size (nm) 24.1 25.6 19.1 18.9
Fig. 2 e TEM images of the reduced catalysts. (a) Cu2Zn1Si1eNaOH, (b) Cu2Zn1Si1eNa2CO3.
where ni is the number of particles having a characteristic diameter di. The calculated copper particle sizes in reduced Cu2Zn1Si1eNaOH catalyst and Cu2Zn1Si1eNa2CO3 catalyst in the HRTEM images are 23 nm and 17 nm, respectively.
Catalyst texture The information about the texture of the calcined catalysts is shown in Table 3. The N2 adsorptionedesorption isotherms of Cu2Zn1Si1eNaOH catalyst and Cu2Zn1Si1eNa2CO3 catalyst are shown in Fig. 3. As evidenced by the shape of the isotherms and by data reported in Table 3, the precipitant used in the process of catalyst preparation has a drastic influence on textural properties of different samples. According to the IUPAC classification, Cu2Zn1Si1eNa2CO3 catalyst exhibit type Ⅳ isotherms with a regular H1-type hysteresis loop at relative pressures (p/p0) of 0.72e1.0 due to the N2 capillary condensation in cylindrical channels [20,21]. Cu2Zn1Si1eNaOH catalyst show the same type of N2-sorption isotherms with a higher H1-type hysteresis loop at relative pressures (p/p0) of 0.6e1.0 but a wider interval between two closed points, indicating that this sample has more regular cylindrical channels and larger pore size [22].
Table 3 e Texture of the calcined samples. Samples Cu2Zn1Si1eNaOH Cu2Zn1Si1eNa2CO3 a b c
SBET (m2/g)a
TPVb (cm3/g)c
Mean pore size (nm)c
63.6 80.4
0.38 0.29
21.9 17.0
BET surface area. Total pore volume. Calculated based on desorption branch of isotherm.
10955
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
dispersed copper oxide species on outside surface of catalyst [24] and the second peak (at around 210 C) can be assigned as reduction of Cu2þ ions reduced with more difficulty [23,25]. Furthermore, the metal reduction degrees calculated from the hydrogen consumption peak area in the temperature region from 50 to 300 C by H2-TPR for Cu2Zn1Si1eNaOH and Cu2Zn1Si1eNa2CO3 are 98.2% and 96.9%, respectively. This result implies either the presence of some Cuþ in the samples treated at 300 C or a not complete reduction of Cu2þ. Therefore, more information about metal species in the surface layer of the reduced samples is obtained from XPS experiments in the next section.
3 Adsorbed Volume/cm /g STP
Full symbols: adsorption Empty symbols: desorption
Cu2Zn1Si1- NaOH
Cu2Zn1Si1- Na2CO3
0.0
0.2
0.4
0.6
0.8
1.0
0
Relative Pressure/(P/P )
Fig. 3 e Nitrogen adsorptionedesorption isotherms. The BET surface and pore volume of Cu2Zn1Si1eNaOH catalyst determined by the BarretteJoynereHalenda (BJH) method are 63.6 m2/g and 0.38 cm3/g, respectively (Table 3) and the average pore diameter of the catalyst is 21.9 nm. On the other hand, Cu2Zn1Si1eNa2CO3 catalyst possesses surface area of 80.4 m2/g (Table 3), smaller pore volume (0.29 cm3/g) and mean pore size (17.0 nm).
H2-TPR characterization Fig. 4 illustrates H2-TPR profiles of calcined catalysts prepared by two types of precipitants. An asymmetric peak at 210 C with a distinct shoulder peak in the lower temperature region is detected for Cu2Zn1Si1eNa2CO3 catalyst, indicating that different kinds of Cu species formed in the catalysts [23]. On the other hand, reduction of Cu2Zn1Si1eNaOH catalyst also proceeds in two steps with a sharp maximum at 210 C and a not obvious shoulder at a slightly lower temperature. The first hydrogen consumption peak observed in two samples (at around 170 C) can be ascribed to the reduction of highly
Surface analysis XPS has provided additional information about copper phases in the reduced catalysts. XPS and X-ray induced Auger spectra (XAES) of the reduced samples are shown in Figs. 5 and 6, respectively. The whole Cu2p region comprises two components, 2p3/2 and 2p1/2, due to the spin-orbit splitting for orbital having l > 1. These two components are well separated (around 20 eV) so only the Cu2p3/2 peaks can be considered for the assignment of copper chemical states [26]. In Fig. 5, the catalysts display the peaks at ca. 934.8 ± 0.1 eV, which are assigned to Cu2þ. Moreover, the presence of the characteristic shakeup 2p / 3d satellite peaks between 940 and 945 eV also suggests the existence of Cu2þ species in the two catalysts [27]. Since the binding energy values for Cuþ and Cu0 are almost identical, the peaks at 932.5 eV are assigned to 2p3/2 of copper in the form of Cu0 and/or Cuþ [28]. Thus, the distinction between these two species on the catalyst surface needs a modified Auger parameter, a0 , defined as Cu2p3/2 (BE) þ Cu LMM (KE) [29]. The deconvolution of the original Cu LMM peaks is thus carried out and the peak positions as well as their contributions are extracted from the deconvolution and are listed in Table 4. The a0 value at ca. 1851.0 eV is due to Cu0 and that at ca. 1847.0 eV to Cuþ according to reference [30]. The results (Table 4) indicate that both Cu0þ and Cuþ coexist on the surface of the reduced Cu2Zn1Si1eNa2CO3 catalyst, while no Cuþ species is observed in the Cu2Zn1Si1eNaOH
Cu2p region
Cu2Zn1Si1- NaOH Cu2Zn1Si1-Na2CO3
Cu2p3/2
Cu2p1/2
932.5
934.8 2+
(b)
Intensity (a.u.)
Intensity (u.a.)
Cu shake up
(a) 0
100
200
300
400
500
Temperature (ºC)
Fig. 4 e H2-TPR profiles of the calcined samples. (a) Cu2Zn1Si1eNaOH and (b) Cu2Zn1Si1eNa2CO3.
600
965
960
955
950
945
940
935
930
Binding Energy (eV)
Fig. 5 e Cu2p XPS spectra of reduced samples. (a) Cu2Zn1Si1eNaOH, (b) Cu2Zn1Si1eNa2CO3.
925
10956
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
CuLMM
Table 5 e Catalytic performances of catalysts prepared by different precipitants. Type
Catalysta
Type Type Type Type Type
Cu2Zn1Si1eNaOH Cu2Zn1Si1eKOH Cu2Zn1Si1e(NH4)2CO3 Cu2Zn1Si1eNaHCO3 Cu2Zn1Si1eNa2CO3
Intensity (a.u.)
(b) 914.4
918.7
915.8
(a) 912
914
916
918
920
922
924
926
A A B B B
Ethyl acetate Ethanol Ethanol conversion selectivity Yieldb (%) (%) (%) 83.20 78.96 69.41 88.15 90.48
98.63 99.24 99.03 98.94 98.95
82.06 78.36 68.74 87.22 89.53
Catalyst 0.9 g, in situ reduction in H2 at 300 C for 2 h before reaction. b Reaction conditions: Temp. ¼ 250 C, pressure ¼ 2 Mp, g (molar ratio of hydrogen to ester) ¼ 15 and 2 h1 WHSV of ethyl acetate. a
Kinetic Energy (eV)
Fig. 6 e XAES spectra of different catalysts after reduction. (a) Cu2Zn1Si1eNaOH, (b) Cu2Zn1Si1eNa2CO3.
catalyst. Compared with bulk value, the smaller a0 values calculated for Cuþ (bulk value is 1848.8 [31]) and Cu2þ (bulk value is 1852.0 [31]) are attributed to the well dispersed copper species and a strong interaction between species and the support. It has been reported that when copper (in þ2, þ1, and 0 valence) is in the highly dispersed state and in intimate contact with the support, a0 can be 2e3 eV lower than the bulk value [32,33].
Catalytic performance The catalytic data of the catalysts prepared by different precipitants are shown in Table 5. As expected, all the catalysts show an excellent selectivity towards ethanol over the whole conversion range in the hydrogenolysis of ethyl acetate, which has been described in the report before [5]. Nevertheless, the experimental data of acetate conversion reveal that the precipitants lead to a drastic influence on the catalytic activity of the catalysts for ethyl acetate hydrogenolysis reaction. Indeed, the Cu2Zn1Si1e(NH4)2CO3 catalyst exhibits low activity for the reason of the low copper loading measured by ICP-AES (Table 1), which is due to the formation of copper complex (Cu(NH3)2þ 4 ). Additionally, the activity of type B
Table 4 e XPS parameters of the reduced samples. Samples Cu2Zn1Si1eNaOH Cu2Zn1Si1eNa2CO3
KEb (eV)
(eV)
932.5 934.9 932.5
918.9 915.7 918.5 914.4 916.0.
1851.4 1850.6 1851.0 1846.9 1850.7
934.7 a
a0 c
BEa (eV)
Assignment Xd (%) Cu0 Cu2þ Cu0 Cuþ Cu2þ
77.4 22.6 71.3 6.7 22.0
Binding Energy. b Kinetic Energy. c a0 ¼ BE þ KE. d Intensity ratio of Cu0 (or Cu2þ, Cuþ)/(Cu0 þ Cuþ þ Cu2þ) by deconvolution of CuLMM XAES spectra.
catalysts, except Cu2Zn1Si1e(NH4)2CO3, is much higher than that of type A catalysts, and the comparison between the catalysts in the same type (except Cu2Zn1Si1e(NH4)2CO3) reveals that the precipitants also shows some influence on the performance of the catalysts, while the influence is not remarkable. As mentioned above, it is observed that the activity of type B catalysts, except Cu2Zn1Si1e(NH4)2CO3, is much higher than that of type A catalysts. Hence, more attention is focused on the impact of the different types of precipitants on the copperbased catalysts. As the representative for type B catalysts, Cu2Zn1Si1eNa2CO3 catalyst possesses smaller copper particles than that of Cu2Zn1Si1eNaOH, which is confirmed by XRD (Fig. 1) and TEM (Fig. 2) results, indicating that the catalysts with better dispersed active species (i.e., a smaller Cu particle) show higher acetate conversion levels. Namely, the conversion of ethyl acetate depends strongly on the particle size of copper, smaller sized copper is extremely active [34]. Moreover, with respect to the texture of the samples (Table 3), larger BET surface of Cu2Zn1Si1eNa2CO3 is observed. In many reports [35,36], large BET surface is of great importance for the design of efficient catalysts, which is an important reason for the high activity of Cu2Zn1Si1eNa2CO3 catalyst. From the XPS and X-ray induced Auger spectra of reduced Cu2Zn1Si1eNaOH catalyst and Cu2Zn1Si1eNa2CO3 catalyst, the conclusion obtained is that well dispersed Cu2þ species (around 22%) and Cu0 species (>71%) both exist on the surface of the two catalysts, while only the Cu2Zn1Si1eNa2CO3 catalyst shows the presence of highly dispersed Cuþ species (6.7%). Despite the high metal reduction degrees (by H2-TPR) for the two samples reduced at 300 C for 2 h, some unreduced Cu2þ exist in the surface layer, which may have negative influence on the catalyst activity. With respect to Cu0 species and Cuþ species, it has been suggested that only Cu0 acts as the active site in ester hydrogenation [37], and this is also evidenced by the experimental date, from which it is observed that Cu2Zn1Si1eNaOH catalyst, though no Cuþ exists on the surface, shows the conversion of ethyl acetate 83.2% and high selectivity to ethanol. The existence of Cuþ for the reduced catalysts signifies the stronger interaction between copper and silica [32]. Moreover, Cuþ may function as electrophilic or Lewis acid sites to polarize the C]O bond via the electron lone pair on oxygen [30], thus improving the reactivity of the ester group in ethyl acetate. The cooperative effect between Cu0 and
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
Cuþ is responsible for the higher hydrogenation activity of the Cu2Zn1Si1eNa2CO3 catalyst.
Conclusion The present work demonstrated that the precipitants used in the coprecipitation method have profound effects on the texture, composition, and structure of the calcined Cu/ZnO/ SiO2 and reduced Cu/ZnO/SiO2 catalysts. (NH4)2CO3 can lead to the formation of copper complex (Cu(NH3)2þ 4 ), indicating that it is not a suitable precipitant for Cu ions. The different types of precipitants (except (NH4)2CO3), show a drastic influence on the performance of copper-based catalysts, whereas the difference of catalysts in the same type is not obvious. The type B catalysts (except Cu2Zn1Si1e(NH4)2CO3) possess smaller copper particles and larger BET surface than that of type A catalysts, which is the main reason for the higher activity of type B catalysts. Furthermore, the cooperative effect between Cu0 and Cuþ can lead to higher hydrogenation activity. On the Cu/ ZnO/SiO2 catalyst prepared by NaCO3, an ethanol yield of 89.53% was obtained under the hydrogenation conditions.
references
[1] Luchansky MS, Monks J. Supply and demand elasticities in the U.S. ethanol fuel market. Energy Econ 2009;31:403e10. [2] Adkins H, Folkers K. The catalytic hydrogenation of esters to alcohols. J Am Chem Soc 1931;53:1095e7. [3] Yan TY, Albright LF, Case LC. Hydrogenolysis of esters, particularly perfluorinated esters. Ind Eng Chem Prod Res Dev 1965;4:101e7. [4] Evans JW, Wainwright MS, Cant NW, Trimm DL. Structural and reactivity effects in the copper-catalyzed hydrogenolysis of aliphatic esters. J Catal 1984;88:203e13. [5] Claus P, Lucas M, Lucke B. Selective hydrogenolysis of methyl and ethyl acetate in the gas phase on copper and supported group Ⅷ metal catalysts. Appl Catal A 1991;79:1e18. [6] Kohler MA, Cant NW, Wainwright MS, Trimm DL. The mechanism of the catalytic chemistry of ester hydrogenolyisis on copper surfaces. Proc Int Congr Catal 1988;9:1043e50. [7] Kenvin JC, White MG. Supported catalysts from mononuclear copper complexes: catalytic properties. J Catal 1992;135:81e91. [8] Evans JW, Cant NW, Trimm DL, Wainwright MS. Hydrogenolysis of ethyl formate over copper-based catalysts. Appl Catal 1983;6:355e62. [9] Brands DS, Poels EK, Bliek A. Ester hydrogenolysis over promoted Cu/SiO2 catalysts. Appl Catal A 1999;184:279e89. [10] Kim SM, Lee ME, Choi JW, Suh DJ, Suh YW. The role of ZnO in Cu/ZnO/Al2O3 catalyst for hydrogenolysis of butyl butyrate. Catal Commun 2011;12:1328e32. [11] van de Scheur FT, Staal LH. Effects of zinc addition to silica supported copper catalysts for the hydrogenolysis of esters. Appl Catal A 1994;108:63e83. [12] Yuan P, Liu ZY, Zhang WQ, Sun HJ, Liu SC. Cu-Zn/Al2O3 catalyst for the hydrogenation of esters to alcohols. Chin J Catal 2010;31:769e75. [13] Huang H, Cao GP, Fan CL, Wang SH, Wang SJ. Effect of water on Cu/Zn catalyst for hydrogenation of fatty methyl ester to fatty alcohol. Korean J Chem Eng 2009;26:1574e9.
10957
[14] Chen HM, Zhu YL, Ding GQ, Zheng HY, Li YW. Study on hydrogenation of dimethyl oxalate to ethylene glycol. J Fuel Chem Tech 2011;39:519e26. [15] Barrett EP, Joyner LG, Halenda PP. The determination of pore volume and area distributions in porous substances. J Am Chem Soc 1951;73:373e80. [16] Holland EB, Gilligan GM. Solubility of copper sulphate, malachite, and burgundy precipitate in ammonium hydroxide. J Phys Chem 1927;31:728e41. [17] van der Grift CJG, Wielers AFH, Mulder A, Geus JW. The reduction behaviour of silica-supported copper catalysts prepared by deposition-precipitation. Thermochim Acta 1990;171:95e113. [18] van der Grift CJG, Wielers AFH, Mulder A, Geus JW. Characterization of silica-supported copper catalysts by means of temperature-programmed reduction. Appl Catal 1990;60:181e92. [19] Chinchen GC, Waugh KC, Whan DA. The activity and state of the state of the copper surface in methanol synthesis catalysts. Appl Catal 1986;25:101e7. [20] Groen JC, Peffer LAA, Ramirez JP. Pore size determination in modified micro- and mesoporous materials. Pitfalls and limitations in gas adsorption data analysis. Microporous Mesoporous Mater 2003;60:1e17. [21] Li CL, Zhang QH, Wang Y, Wan HL. Preparation, characterization and catalytic activity of palladium nanoparticles encapsulated in SBA-15. Catal Lett 2008;120:126e36. [22] Wang JJ, Chernavskii PA, Wang Y, Khodakov AY. Influence of the support and promotion on the structure and catalytic performance of copper-cobalt catalysts for carbon monoxide hydrogenation. Fuel 2013;103:1111e22. [23] Yuan ZL, Wang LN, Wang JH, Xia SX, Chen P, Hou ZY, et al. Hydrogenolysis of glycerol over homogenously dispersed copper on solid based catalysts. Appl Catal B 2011;101:431e40. [24] Lin L, Pan PB, Zhou ZF, Li ZJ, Yang JX, Sun ML, et al. Cu/SiO2 catalysts prepared by the sol-gel method for hydrogenation of dimethyl oxalate to ethylene glycol. Chin J Catal 2011;32:967e9. [25] Esposito S, Turco M, Bagnasco G, Cammarano C, Pernice P. New insight into the preparation of copper/zirconia catalysts by sol-gel method. Appl Catal A 2011;403:128e35. [26] Salvi AM, Langerame F, Macchia A, Sammartino MP, Tabasso ML. XPS characterization of (copper-based) coloured stains formed on limestone surfaces of outdoor Roman monuments. Chem Cent J 2012;6:1e13. [27] Sun KP, Lu WW, Qiu FY, Liu SW, Xu XL. Direct synthesis of DME over bifunctional catalyst: surface properties and catalytic performance. Appl Catal A 2003;252:243e9. [28] Kaddouri A, Dunpont N, Gelin P, Delichere P. Methane combustion over copper chromites catalysts prepared by the sol-gel process. Catal Lett 2011;141:1581e9. [29] Moretti G. Auger parameter and Wagner plot in the characterization of chemical states: initial and final state effects. J Electron Spectrosc 1995;76:365e70. [30] Chen LF, Guo PJ, Qiao MH, Yan SR, Li HX, Shen W, et al. Cu/ SiO2 catalysts prepared by the ammonia-evaporation method: texture, structure, and catalytic performance in hydrogenation of dimethyl oxalate to ethylene glycol. J Catal 2008;257:172e80. [31] Fantauzzi M, Atzei D, Elsener B, Lattanzi P, Rossi A. XPS and XAES analysis of copper, arsenic and sulfur chemical state in energies. Surf Interface Anal 2006;38:922e30. [32] Yin AY, Guo XY, Dai WL, Fan KL. Effect of initial precipitation temperature on the structural evolution and catalytic behavior of Cu/SiO2 catalyst in the hydrogenation of dimethyloxalate. Catal Commun 2011;12:412e6.
10958
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 9 ( 2 0 1 4 ) 1 0 9 5 1 e1 0 9 5 8
[33] Kohler MA, Curry-Hyde HE, Hughes AE, Sexton BA, Cant NW. The structure of Cu/SiO2 catalysts prepared by the ionexchange technique. J Catal 1987;108:323e33. [34] Zhu YM, Shi L. Zn promoted Cu-Al catalyst for hydrogenation of ethyl acetate to alcohol. J Ind Eng Chem; 2013. http:// dx.doi.org/10.1016/j.jiec.2013.10.010. [35] Tang XL, Meng X, Shi L. Desulfurization of model gasoline on modified bentonite. Ind Eng Chem Res 2011;50:7527e33.
[36] Huang H, Yi DZ, Lu YN, Wu XL, Bai YP, Meng X, et al. Study on the adsorption behavior and mechanism of dimethyl sulfide on silver modified bentonite by in FTIR and temperatureprogrammed desorption. Chem Eng J 2013;225:447e55. [37] Mokhtar M, Ohlinger C, Schlander JH, Turek T. Hydrogenolysis of dimethyl maleate on Cu/ZnO/Al2O3 catalysts. Chem Eng Technol 2001;24:423e6.