Activity of Cu-based catalysts prepared using adsorption phase reaction technique in first step of methanol synthesis

WANG Zhiyong, DENG Hui, ZHANG Ting, JIANG Xin

?

Activity of Cu-based catalysts prepared using adsorption phase reaction technique in first step of methanol synthesis

WANG Zhiyong1, DENG Hui2, ZHANG Ting1, JIANG Xin1

(College of Chemical and Biological EngineeringZhejiang UniversityHangzhouZhejiangChinaChina Academy of Engineering PhysicsMianyangSichuanChina

The Cu-based catalysts preparedadsorption phase reaction technique (APRT) were characterized using XRD, HRTEM and H2-TPR. The results indicated the copper species of catalysts were well-dispersed on the surface of the supports with the size varied from 5—10 nm. In the catalytic methanol synthesis from syngas in ethanol solvent, the prepared catalysts performed a much higher activity than the commercial catalysts did for the formation of intermediate, ethyl formate. The remarkable difference between APRT catalyst and other Cu-based catalyst (including the commercial catalyst) on two reactions of methanol synthesis manifests that the APRT catalyst has a different structure as well as active sites for the formation and hydrogenolysis of ethyl formate.

microreactor; Cu-based catalyst; methanol synthesis; syngas; catalysis

Introduction

Methanol synthesis has been well studied because of its importance in chemical industries and its potential as an environmentally friendly energy carrier[1]. Commercially, methanol is produced from syngas under high temperature and high pressure, using mainly Cu-based oxide catalysts. However, the efficiency of methanol synthesis is severely limited by thermodynamics because methanol synthesis is an extremely exothermic reaction. Therefore, low-temperature methanol synthesis has attracted great interests because it will greatly reduce the production cost and realize high CO conversion.

Tsubaki[2-3]proposed a new method of low-temperature methanol synthesis from syngas on Cu-based catalyst using alcohol as solvent, by which methanol was produced at 443 K and 3 MPa. For the use of alcohol in system, formate ester, intermediate product, formed and then reduced to methanol by hydrogen. The catalysts in alcohol system could endure the present of CO2and H2O in reaction, comparing with other reaction system[4].

Controversies in the role of the active site of catalyst also existed in study on low-temperature methanol synthesis in alcohol, as traditional methanol synthesis process. Fujimoto[5]pointed out that Cu was active site for the methanol synthesis and the activity of the catalyst was proportional to the surface area of Cu. Tsubaki’s group[6-7]reported that the active sites of catalyst were not only Cu but also special sites such as Cu-ZnOor Cu-Zn site, in which Cu and the special site work cooperatively to catalyze the methanol synthesis reaction. In the homogeneous and heterogeneous synergistically catalytic system, the roles of potassium formate for the esterification as well as Cu/MgO for the hydrogenolysis were approached[8]. However, it is still difficult to distinguish the impact of active site on two reactions in the most of study. Moreover, it is even hard to confirm whether the active site for two steps is identical.

As well known, the methanol synthesis catalysts are structure-sensitive catalysts and the difference in preparation method will lead to the disparities in the catalytic performance and promote the cognition of catalytic process. In this study, a new method, adsorption phase reaction technique (APRT)[9-11], was used to prepare Cu-based catalysts and new catalytic phenomenon in methanol synthesis from syngas was reported.

1 Experimental section

1.1 Materials

Hydrophilic silica (SiO2; Degussa AEROSIL200, average size 12 nm, specific area 200 m2·g-1), ethanol (C2H5OH; AR, Sinopharm Chemical Reagent Co., Ltd.), zinc nitrate (Zn(NO3)2·6H2O; AR, Shanghai Meixing Chemical Reagent Ltd.), cupric acetate (Cu(Ac)2·H2O; AR, Shanghai Zhenxin Chemical Reagent Ltd.), sodium hydroxide (NaOH; AR, Sinopharm Chemical Reagent Co., Ltd.), industrial catalyst NC 308 (Cu/ZnO/Al2O3; Nanjing Chuangming Catalyst Technology Co., Ltd).

1.2 Preparation of Cu-based catalysts

Catalysts were preparedon the silica surface by adsorption phase reaction technique. In a typical procedure, silica and copper acetate were added in ethanol-water binary liquid system under a stirred condition. Hydrophilic silica adsorbed water molecule selectively on its surface, as a result, a water- rich adsorbed-layer formed. The cupric acetate distributed itself between the adsorbed-layer and ethanol bulk. Its concentrations in these two regions were tremendously different. After adsorption equilibrium, sodium hydroxide was added into the system. When sodium hydroxide diffused into the adsorbed-layer and reacted with Cu2+, catalyst precursorwas generated.

Preparation of Cu/SiO2catalyst: 1.0 g SiO2, 200 ml ethanol and 2 ml water were added in a triflask under a stirred condition at 303 K. Then 0.30 g Cu(Ac)2·H2O dissolved in 100 ml ethanol was added into system. After 8 h adsorption, 0.24 g NaOH dissolved in 100 ml ethanol was added. After reacting for 8 h, the product was gained by filtering and several washing with ethanol then dried at room temperature followed by different pretreatments.

Preparation of Zn/Cu/SiO2catalyst: The preparation process was similar as the preparation of Cu/SiO2. After 0.24 g NaOH dissolved in 100 ml ethanol was added and reacted for 8 h, 0.45 g Zn(NO3)2·6H2O dissolved in 100 ml ethanol was added and reacted for 8 h. The product was gained by filtering and washing and then dried at room temperature followed by different pretreatments.

Catalysts were pretreated by two methods. Method A: Catalysts were dried at 373 K for 12 h, and reduced by 10% H2flow (H2: 10 ml·min-1, N2: 90 ml·min-1) at 523 K for 10 h, then cooled to room temperature by 10% H2flow. Method B: Catalysts were dried at 373 K for 12 h, and reduced in 10% H2flow (H2: 10 ml·min-1, N2: 90 ml·min-1) at 523 K for 10 h and passivated by 10% O2flow (O2: 4 ml·min-1，N2: 36 ml·min-1) at 523 K for 2 h, then cooled to room temperature by 10% O2flow.

1.3 Characterization

Transmission Electron Microscopy (TEM): One drop of an ultrasonic-mixed, dilute alcohol suspension of the catalyst was placed on a carbon-coated grid. Electron micrographs of the particles retained was taken after evaporation of the solvent. A JEM-2010 (HR) transmission electron microscope was used.

X-ray Diffraction (XRD): The catalysts were analyzed by XRD using a D/max-rA XRD instrument (XD-98, Philips, Eindhoven, The Netherlands) with Cu Kαradiation (0.15406 nm). Patterns were recorded from 10° to 80° (2) at a rate of 4 (°)·min-1at 40 kV and 30 mA. The characteristic peak at 243.3° of Cu(111) were fittedCauchy equation, from which the crystallite size was calculated by Scherrer equation

whereis the Scherrer constant;andare the radiation wavelength and Bragg’s angle, respectively;.

Hydrogen Temperature Programmed Reduction (H2-TPR): The catalysts were analyzed by H2-TPR using a Micromeritics AutoChemⅡ 2920 adsorption instrument. Analysis condition: 30 ml·min-110% H2-Ar mixed gas flow, temperature increasing from 373 to 973 K at a heating rate of 15K·min-1, sample weight: 80.0 mg.

1.4 Catalytic activity test

The catalytic performance of the prepared Cu/SiO2and Zn/Cu/SiO2catalysts for the methanol synthesis using ethanol as solvent was measured. A closed batch reactor with inner volume of 230 ml and a magnetic stir bar was employed. Ethanol solvent and catalyst were added into the reactor at first. Then the reactor was closed and the inside air was purged by the reactant gas of 1 MPa for five times. Reactant gas was introduced and then the reaction took place at the setting temperature. After reaction, the reactor was put into a cooling pool. The gas and liquid products were analyzed qualitatively by GC-MS (HP 6890/5973). All liquid products were analyzed quantitatively by gas chromatograph (Fuli GC-9790, capillary column, ATOV-225).

2 Results and discussion

2.1 Catalysts characterization

The surface morphology of the fresh and used catalysts was showed in Fig.1. In the both pictures, the copper species (dark particles) were well-dispersed on the surface of SiO2and the size of particles varied from 5 to 10 nm. The XRD showed the difference: no characteristic diffractive peaks of copper species were observed in fresh catalysts, whereas the characteristic diffractive peaks of metallic Cu appeared in used catalysts (not given).

The H2-TPR profiles of catalysts were shown in Fig.2. The reduction temperature of the industrial catalyst (NC 308) was the lowest. The addition of Zn decreased the reduction temperature, in agreement with the promoting effect played by zinc species on the reducibility of copper[8,12]. The reduction temperature of Cu/SiO2and Zn/Cu/SiO2is higher than industrial catalyst, suggesting the interaction between copper species and support was stronger in the APRT catalysts. The areas of reduction peak of the APRT catalysts were much lower than that of the NC 308, indicating the copper content of the APRT catalysts was lower.

Fig.1 TEM pictures of non-pretreated catalysts

Fig.2 H2-TPR profiles for industrial catalyst (NC 308), Zn/Cu/SiO2 and Cu/SiO2

2.2 Non-pretreated APRT Cu-based catalyst for methanol synthesis

The activity of the non-pretreated APRT Cu/SiO2catalyst for the methanol synthesis from syngas using ethanol as solvent was investigated. Methanol was synthesizedthe formation and hydrogenolysis of ethyl formate, which was the key intermediate and can be further hydrogenated to produce methanol and ethanol. Table 1 showed the batch reaction results of the non-pretreated APRT Cu/SiO2with different feed gas compositions. The catalyst performed high activity for the formation of ethyl formate. It seemed that H2had no influence on the production of ethyl formate, because ethyl formate can be produced without H2in feed gas. As the initial pressure of CO increased from 1 to 2 and 3 MPa, the productivity of ethyl formate increased 45.9% and 98.7%, respectively. The average size of metallic Cu in used catalysts also increased as the increase of CO initial pressure.

Zeng[3]reported that no ethyl formate was produced when CO was used alone, which was different with our results. Based on the reaction mechanism proposed by Yang[13-14]ethyl formate could be producedthe reaction between ethanol and adsorbed formate species, which can be produced by inserting of CO into adsorbed hydroxide. We inferred that adsorbed hydroxide might largely exist on the surface of the APRT catalyst, sequentially promoted the formation of ethyl formate.

Table 1 Batch reaction results for methanol synthesis with non-pretreated APRT catalyst

①No peaks of metallic Cu were observed.

Note: Reaction conditions: catalyst weight: 0.2 g, ethanol solvent: 100 ml, 443 K, 4 h.

It was interesting that no methanol was detected for all five experiments, which was quite different with literature. In the similar study, the first step, the formation of ethyl formate, was considered as the rate-determining reaction, while the second step, the hydrogenolysis of ethyl formate, was easily occurred over Cu-based catalysts[6-7,15]. Therefore, a series of experiments were designed to test and verify the reliability of this new phenomenon.

2.3 Influence of pretreatment on catalysts

The APRT catalysts were pretreated with different methods mentioned in the experimental section of catalysts preparation. As shown in Fig. 3, no characteristic diffractive peaks of copper species were observed for the Cu/SiO2catalyst reduced at 523 K. It might be the size of grains was too small to be detected. The peaks of CuO were detected for the Cu/SiO2and Zn/Cu/SiO2catalysts reduced then passivated at 523 K, and the average size was 5.9 nm and 6.5 nm, respectively. For the Zn/Cu/SiO2catalyst reduced at 523 K, the copper species was Cu2O, with the average size of 2.9 nm.

As shown in Table 2, the activity of the Cu/SiO2or Zn/Cu/SiO2for the formation of ethyl formate did not change remarkably, although the valence states of copper were different (including Cu+and Cu2+) by different pretreatments. No methanol was detected for the all of experiments, and ethyl formate was produced without H2in feed gas, similar with results in Table 1. For the catalysts pretreated with method A, the size of metallic Cu in used catalysts was far smaller than that pretreated with method B, suggesting the different pretreatments had influence on the interaction between copper species and support. Traditionally, addition of zinc can improve the copper dispersion of Cu-based catalysts and promote the catalytic activity. However, in our experiments, the addition of zinc in the APRT catalyst seriously decreased the productivity of ethyl formate, which seems mean that the interaction between Cu and ZnO does not bring a positive effect on first step reaction although it could be in favor of the second step reaction.

Fig.3 XRD patterns for catalysts pretreated with different methods

●?SiO2; ◇?CuO; ◆?Cu2O

a—Cu/SiO2, dried at room temperature; b—Cu/SiO2, reduced at 523 K;c—Cu/SiO2, reduced then passivated at 523 K; d—Zn/Cu/SiO2,reduced at 523 K; e—Zn/Cu/SiO2, reduced then passivated at 523 K

Table 2 Reaction results for methanol synthesis with Cu/SiO2 and Zn/Cu/SiO2 catalysts pretreated with different methods

①Preteated with method A: reduced at 573 K for 10 h.② Preteated with method B: reduced for 10 h then passivated for 2 h at 573 K.③ No peaks of metallic Cu were observed.

Note: Reaction conditions: catalyst weight: 0.2 g, ethanol solvent: 100 ml, 443 K, 4 h.

The results showed that different pretreatments of catalysts had no obvious influence on the activity for the formation of ethyl formate but influence the average size of Cu remarkably. And no methanol detected in reaction suggested that the APRT Cu-based catalyst only has activity for the formation of ethyl formate while no activity for the hydrogenolysis of ethyl formate.

2.4 Comparison experimental

The comparison experiments between the APRT catalyst and industrial catalyst (NC 308) were carried out to probe other features of the APRT catalyst.

Firstly, the activity of the second reaction was directly test by using pure ethyl formate as raw materials and solvent under atmosphere of H2. As shown in Table 3, the NC 308 performed a high activity for the hydrogenolysis of ethyl formate. Ethanol and methanol were produced through the hydrogenolysis of ethyl formate, and methyl formate formed through the ester-exchange reaction between methanol and ethyl formate. But the APRT catalyst showed no activity for hydrogenolysis.

Table 3 Results of hydrogenolysis of ethyl formate with NC 308 and Cu/SiO2

① NC 308 (reduced then passivated at 523 K ): 2.0 g.② Cu/SiO2(reduced then passivated at 523 K): 0.5g.

Note: Reaction conditions: HCOOC2H5: 40 ml;(H2)3 MPa (initial pressure), 443 K, 4 h.

Secondly, the activity of the NC 308 for the methanol synthesis from syngas using ethanol as solvent was tested. As shown in the first row of Table 4, both ethyl formate and methanol formed, and the selectivity of methanol was higher than that of ethyl formate, which was similar with the reported literature[16-18]. It indicated that industrial catalyst NC 308 had active sites for the both two steps. Comparing with Table 1 and Table 2, the APRT catalysts, especially the Cu/SiO2, manifest much higher activity for the first reaction than industrial catalyst NC 308.

Table 4 Reaction results for methanol synthesis with NC 308 and mixed catalyst

①Catalyst weight: 1.0 g (reduced then passivated at 523 K).② NC 308: 1.0 g, Cu/SiO2: 0.2 g (reduced then passivated at 523 K).③ Calculated based on the weight of NC 308 only.④Calculated based on the weight of APRT Cu/SiO2only.

Note: Reaction temperature: C2H5OH: 100 ml, 443 K, 4 h.

Thirdly, to avoid the interference of second reaction, the activity of the NC 308 for the formation of ethyl formate from CO without H2also was investigated as showed in the second row of Table 1. By contrast, the productivity of ethyl formate for the APRT Cu-based catalyst in Table 1 was 29 times more than that of NC 308 at the same reaction condition ((CO)3 MPa). Considering the much lower copper content in APRT catalyst, the difference of the activity based on the weight of Cu is huge. The higher activity of APRT catalyst for formation of ethyl formate, on one hand, might be attributed to the better dispersion of copper species on support. On the other hand, the richer adsorbed hydroxide on the surface of APRT catalysts could promote the formation of ethyl formate.

Fourthly, the activity of the mixed catalyst of the APRT Cu-based catalyst and NC 308 for methanol synthesis from syngas was investigated. The productivities of ethyl and methanol for the NC 308 increased as the APRT catalyst mixed. It might be the increase of concentration of key intermidiate ethyl formate that resulted in the increase of productivity of methanol. The productivity of ethyl formate for the Cu/SiO2could be recalculated as 900 mmol·kg-1·h-1for methanol synthesisthe hydrogenation of ethyl formate, closed to 1063 mmol·kg-1·h-1for the non-pretreated Cu/SiO2catalyst listed in Table 1. It suggested the Cu/SiO2provided most active site for the formation of ethyl formate in mixed catalyst.

All experiments showed the APRT catalysts had higher activity for the formation of ethyl formate, which was considered as the rate-determining step for other Cu-based catalysts. The unexpected low activity for the hydrogenolysis of APRT catalysts manifest that, at least for the APRT catalysts, the active sites for formation and hydrogenolysis of ethyl formate are different, which suggests we could explore the active sites of two reactions separately.

3 Conclusions

The Cu-based catalysts were preparedadsorption phase reaction technique. The copper species in as-prepared catalysts were well-dispersed on the surface of support with size varied from 5—10 nm. The catalysts performed high productivity of key intermediate ethyl formate, and the productivity was proportional to the initial pressure of CO. It was found that ethyl formate could form without H2in feed gas, while no methanol was detected in product. The results of a series of experiments indicated that the APRT catalysts had the active site for formation of ethyl formate but no active site for hydrogenolysis, which implied the active sites for the two steps were different.

[1] Liu X M, Lu G Q, Yan Z F, Beltramini J. Recent advances in catalysts for methanol synthesishydrogenation of CO and CO2[J]., 2003, 42 (25): 6518-6530.

[2] Tsubaki N, Ito M, Fujimoto K. A new method of low-temperature methanol synthesis [J]., 2001, 197 (1): 224-227.

[3] Zeng J, Fujimoto K, Tsubaki N. A new low-temperature synthesis route of methanol: catalytic effect of the alcoholic solvent [J]., 2002, 16 (1): 83-86.

[4] Palekar V M, Tierney J W, Wender I. Alkali compounds and copper chromite as low-temperature slurry phase methanol catalysts [J].:, 1993, 103 (1): 105-122.

[5] Hu B, Fujimoto K. Low temperature methanol synthesis in slurry phase with a hybrid copper-formate system [J]., 2009, 129 (3/4): 416-421.

[6] Bao J, Liu Z, Zhang Y, Tsubaki N. Preparation of mesoporous Cu/ZnO catalyst and its application in low-temperature methanol synthesis [J]., 2008, 9 (5): 913-918.

[7] Yang R Q, Yu X C, Zhang Y, Li W Z, Tsubaki N. A new method of low-temperature methanol synthesis on Cu/ZnO/Al2O3catalysts from CO/CO2/H2[J]., 2008, 87 (4/5): 443-450.

[8] Zhao T S, Zhang K, Chen X R, Ma Q X, Tsubaki N. A novel low-temperature methanol synthesis method from CO/H2/CO2based on the synergistic effect between solid catalyst and homogeneous catalyst [J]., 2010, 149 (1/2): 98-104.

[9] Jiang X, Wang T. Influence of preparation method on morphology and photocatalysis activity of nanostructured TiO2[J]., 2007, 41 (12): 4441-4446.

[10] Wang T, Jiang X, Mao C. Influence of an adsorption layer and its evolvement on the formation of Ag [J]., 2008, 24 (24): 14042-14047.

[11] Jiang X, Deng H. Synthesis of Au-CeO2/SiO2catalystadsorbed- layer reactor technique combined with alcohol-thermal treatment [J]., 2011, 257 (24): 10883-10887.

[12] Fierro G, Jacono M L, Inversi M, Tsubaki N. Study of the reducibility of copper in CuO-ZnO catalysts by temperature-programmed reduction [J].:, 1996, 137 (2): 327-348.

[13] Yang R Q, Fu Y L, Zhang Y, Tsubaki N.DRIFT study of low-temperature methanol synthesis mechanism on Cu/ZnO catalysts from CO2-containing syngas using ethanol promoter [J]., 2004, 228 (1): 23-35.

[14] Yang R Q, Zhang Y, Iwama Y, Tsubaki N. Mechanistic study of a new low-temperature methanol synthesis on Cu/MgO catalysts [J].:, 2005, 288 (1): 126-133.

[15] Reubroycharoen P, Vitidsant T, Yoneyama Y, Tsubaki N. Development of a new low-temperature methanol synthesis process [J]., 2004, 89 (4): 447-454.

[16] Hu B, Yamaguchi Y, Fujimoto K. Low temperature methanol synthesis in alcohol solvent over copper-based catalyst [J]., 2009, 10 (12): 1620-1624.

[17] Xu B L, Yang R, Meng F Z, Reubroycharoen P, Vitidsant T, Zhang Y, Yoneyama Y, Tsubaki N. A new method of low temperature methanol synthesis [J]., 2009, 13 (3): 147-163.

[18] Hu B, Fujimoto K. High-performance Cu/MgO-Na catalyst for methanol synthesisethyl formate [J].:, 2008, 346 (1): 174-178.

吸附相反應技術制備Cu基催化劑在甲醇合成第一步反應中的活性

汪志勇1，鄧輝2，張挺1，蔣新1

（1浙江大學化學工程與生物工程學院，浙江杭州 310027；2中國工程物理研究院，四川綿陽 621900）

采用吸附相反應技術（APRT）制備了Cu基催化劑，并用XRD、HRTEM、H2-TPR等表征手段進行了分析。結果表明催化劑中的Cu良好分散于載體表面，粒徑在5～10 nm。在液相乙醇體系合成氣制甲醇的反應中，該Cu基催化劑對第一步形成中間產物甲酸乙酯的催化活性遠高于工業催化劑。APRT制備的催化劑與其他催化劑（包括工業催化劑）在液相合成氣制甲醇的兩步反應中表現出的顯著差異，不僅說明APRT催化劑具有不同的結構特點，也表明甲酸乙酯的形成和進一步的加氫的活性位是不同的。

微反應器；Cu基催化劑；甲醇合成；合成氣；催化

10.11949/j.issn.0438-1157.20150805

TQ 426.8

A

0438—1157（2015）08—3050—07

蔣新。

汪志勇（1988—），男，碩士研究生。

國家自然科學基金項目(21276223)；國家高技術研究發展計劃項目(2010AA064905)；浙江省重點科技創新團隊計劃項目(2009R50020)。

2015-06-02.

JIANG Xin, jiangx@zju.edu.cn

supported by the National Natural Science Foundationof China (21276223), the National High Technology Research and Development Program of China (2010AA064905) and the Program for Zhejiang Leading Team of Science & Technology Innovation (2009R50020).

2015-06-02收到初稿，2015-06-09收到修改稿。