Preparation and properties of K₂NiF₄-type perovskite oxides La₂NiO₄ catalysts for steam reforming of ethanol

Zhang Li-feng(张利峰), Wang Yi-ping(王一平), Huang Qun-wu(黄群武)

School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China

Received 10 August 2009; accepted 15 September 2009

Abstract: The performance of La₂NiO₄ perovskite catalysts, prepared using a citric acid complexation method, for the steam reforming of ethanol was studied. The catalysts were characterized by X-ray diffractometry (XRD), specific surface area measurements (BET), thermogravimetric analysis (TGA) and scanning electron microscopy (SEM). The experimental results show that the calcination temperature and the amount of citric acid (CA) have a significant influence on the characteristics of the catalysts and their catalytic activity. Among the catalysts tested, the La₂NiO₄ catalyst calcined at 700 ℃ with n(La)?n(Ni)?n(CA) of 2?1?3 exhibits the best activity and excellent stability as well as very low coke formation.

Key words: hydrogen; ethanol; perovskite-like oxides; La₂NiO₄

1 Introduction

In the wake of the growing demand for carbon- neutral and zero-emission alternative fuels, considerable attention has been paid to transform the biomass-derived compounds into hydrogen-rich gas. Of special interest among such compounds is bioethanol obtained by biomass fermentation, in view of the increasing availability of raw material, ease of handling in the liquid state, high hydrogen content and low toxicity[1-3]. The biomass-derived ethanol, an aqueous solution containing 12%-15% ethanol, requires the distillation for direct use. Hydrogen manufacture by the catalytic steam reforming of ethanol without the distillation can thus be considered to be one of the most effective ways for the utilization of the biomass-derived ethanol. Catalysts using a wide variety of transition metals and supports have previously been reported for the steam reforming of ethanol[4-8]. Among the catalysts reported so far, rhodium, cobalt and nickel are considered to be the most promising metals[9]. Although rhodium seems to be most active, the use of noble metal is considered to be economically unfavorable. On the other hand, more inexpensive nickel catalysts often suffer from catalyst deactivation due to coke formation and sintering. One promising method for the inhibition of carbon deposition over Ni catalysts is using perovskite-type oxides. CHOUDHARY et al[10] reported that complex oxides with a perovskite structure, like LaNiO₃, La_0.8Ca(or Sr)_0.2NiO₃ and LaNi_1-xCo_xO₃ (x=0.2-1.0), were resistant to coking for the partial oxidation of methane to synthesis gas. To inhibit carbon deposition, one must keep the size of the metal clusters smaller than the critical size needed for coke formation. In this work, the Ni catalysts from K₂NiF₄-type perovskite oxides La₂NiO₄ exhibit high activity and a rather high stability. The effects of calcination temperature and citric acid concentration on the structure and the catalytic activity for the steam reforming of ethanol are studied.

2 Experimental

2.1 Catalyst preparation

The La₂NiO₄ catalysts were prepared by a citric acid complexation method. Stoichiometric amounts of Ni(NO₃)₂·6H₂O and La(NO₃)₂·6H₂O were dissolved completely in deionized water, and a stoichiometric amount of citric acid was mixed into the solution. The solution was then heated to 80 ℃ with constant stirring until a gel was formed. The gel was then dried at 110 ℃ for 12 h, and subsequently calcined at different temperatures for 5 h. The calcined samples were pressed, crushed and sieved to obtain a particle size of 0.40-  0.50 mm.

2.2 Catalyst characterization

The powder X-ray diffraction (XRD) experiments were carried out on Philips PANalytical apparatus with Co K_α radiation, at 40 kV and 40 mA. The diffraction angle (2θ) was scanned from 10? to 90?. Specific areas were calculated using the BET method from the nitrogen adsorption isotherms, recorded at the temperature of liquid nitrogen on a NOVA-2000 instrument. The thermogravimetric analysis (TGA) was carried out under an oxidative atmosphere with a Pyris Diamond Analyzer. About 10 mg of sample was heated from room temperature to 850 ℃ at 10 ℃/min. Scanning electron microscopy (SEM) was performed with a PHILIPS XL30 scanning electron microscope operating at 20 kV.

2.3 Catalytic tests

Steam reforming tests were performed in a fixed-bed flow quartz tubular reactor (12 mm of inner diameter) operating at atmospheric pressure. The catalyst (0.10 g) was first subjected to in situ reduction under pure hydrogen flow (3 mL/min) at 500 ℃ for 1 h. Following reduction, hydrogen was replaced by argon (20 mL/h) and temperature was adjusted to reaction temperature. The liquid ethanol-water solution was fed by means of a syringe pum into a heating chamber (150 ℃) and vaporized completely in the stream of argon. The ethanol feed rate was 5.8×10^-5 mol/min in all tests and the ethanol-to-water-to-argon molar ratio of 1?9?14.3 was used. The gas products were analyzed on-line by gas chromatography using both flame ionization and thermal conductivity detector. Results were expressed in terms of molar composition in the gas phase for H₂, CO₂, CO, and CH₄.

3 Results and discussion

3.1 Influence of calcination temperature

The XRD patterns of La₂NiO₄ samples calcined at different temperatures are shown in Fig.1. After the amorphous precursor was calcined at 500 ℃ for 4 h, intense peaks for La₂O₃ were observed. When the temperature was increased to 700 ℃, typical diffraction peaks for the perovskite-like oxide La₂NiO₄, which was identified as single phase with a tetragonal K₂NiF₄ structure, as well as a small amount of La₂O₃ were observed. When the calcination temperature was increased to 850 ℃, the diffraction peaks for the spinel structure significantly intensified and became sharper, which suggests that the crystalline phase of La₂NiO₄ became more perfect. This result indicates that La₂NiO₄ with pure spinel structure can be formed by complexing citric acid with a mixture of La³⁺ and Ni²⁺ and calcining the obtained gel above 700 ℃. The specific surface areas and crystal sizes of the La₂NiO₄, calculated from the half-width of a diffraction peak using Scherrer formula, are listed in Table 1. Upon increasing calcination temperature, the particle sizes increased, and the specific surface area decreased dramatically. The drop in specific surface area may originate from the aggregation of the perovskite phase, which indicates the structure properties are significantly dependent on the calcination temperature.

Fig.1 XRD patterns of La₂NiO₄ calcined at different temperatures

Table 1 Specific surface area and crystal size of La₂NiO₄ calcined at different temperatures

Steam reforming tests were performed at 500 ℃ and 650 ℃ over La₂NiO₄ catalysts calcined at different temperatures. Total ethanol conversion took place all along the 3 h runs at both reaction temperatures. The typical gaseous products H₂, CO₂, CO and CH₄ were obtained in all cases while other common intermediates or by-products like ethylene, diethylether and acetaldehyde were detected only in negligible amounts (＜0.10%). Table 2 shows the typical molar composition of the gaseous mixture obtained with the La₂NiO₄ catalysts calcined at different temperatures. For every catalyst, the H₂ content was found to increase with temperature. At 650 ℃, it was found to be 70.3%- 71.6%, approaching the theoretical reaction stoichiometry (H₂: 75%). Also in agreement with stoichiometry, the H₂-to-CO₂ molar ratio was found to approach 3.0 for the catalysts tested at 650 ℃. In contrast, the CH₄ molar fraction showed a decreasing trend, which is consistent with a higher reaction rate of the reforming reaction(CH₄+2H₂O→CO₂+4H₂), which is favored both thermodynamically and kinetically by the temperature increment. The CO and CO₂ contents were found to have opposite trends with increasing temperature, which is consistent with the temperature dependence of the thermodynamic equilibrium constant of the exothermic water gas shift reaction (WGSR) (CO+H₂O→CO₂+H₂). Ni is known as a not very active catalyst for the WGSR[11]. Thus, under the experimental conditions here the CO-to-CO₂ molar ratios used could be far from the equilibrium values and more dependent on kinetics parameters and textural properties of La₂NiO₄ catalysts. Thus, these La₂NiO₄ catalysts could be explained by differences in the influence of textural properties on the rate of the direct and reverse WGSR.

Table 2 C₂H₅OH reforming using La₂NiO₄ catalysts calcined at different temperatures

By comparing the performance of catalysts at the same reforming temperature, the catalyst La₂NiO₄ (700 ℃) led to a higher H₂ content (71.6%) and a lower CH₄ content (0.10%) in the gaseous mixture than the catalysts La₂NiO₄ (500 ℃) and La₂NiO₄ (85 ℃) at 650 ℃. This is ascribed not only to a higher rate of the CH₄ reforming reaction being linked to a higher surface specific area, but also to that being closely linked to the smaller La₂NiO₄ crystal size in the case of La₂NiO₄ (700 ℃).

3.2 Influence of amount of citric acid

Citric acid (CA) is used as a chelating reagent in the synthesis of the La₂NiO₄ powders. The amount of citric acid plays an important role in the morphology and activity of the catalysts. If the amount of citric acid is too low, some of the ions may not chelate with the citric acid and resulting gel may not be homogeneous. On the other hand, too much citric acid causes waste and precipitation from unchelated citrate. The SEM micrographs of the catalysts with different molar ratios of La to Ni to CA are shown in Fig.2. An homogeneous phase is visible when molar ratios are 2?1?1 and 2?1?3. Thus, we propose that this technique is a good and convenient method to disperse the active metal. While the catalyst powders with n(La)?n(Ni)?n(CA) of 2?1?4 have the relatively large average granularity due to significant agglomeration of fine powders.

Fig.2 SEM images of La₂NiO₄ catalysts prepared with different n(La)?n(Ni)?n(CA): (a) 2?1?1; (b) 2?1?3; (c) 2?1?4

Fig.3 shows the XRD patterns of the La₂NiO₄ catalysts calcined at 700 ℃ with different molar ratios of La to Ni to CA. All these catalysts exhibited the characteristic diffraction peaks for the La₂NiO₄ spinel structure. Weak diffraction lines of NiO were also observed in the catalysts with n(La)?n(Ni)?n(CA)=2?1?1 and n(La)?n(Ni)?n(CA)=2?1?4. This phenomenon indicates that the optimal amount of citric acid needed to obtain a pure perovskite structure of La₂NiO₄ is n(La)?n(Ni)?n(CA)=2?1?3.

Fig.3 XRD patterns of La₂NiO₄ prepared with different n(La)?n(Ni)?n(CA): (a) 2?1?3; (b) 2?1?1; (c) 2?1?4

The surface areas of fresh and used La₂NiO₄ catalysts prepared with different amounts of citric acid are listed in Table 3. As the amount of citric acid increased from n(La)?n(Ni)?n(CA)=2?1?1 to 2?1?4, the specific surface areas of the fresh catalysts decreased from 7.78 to 6.48 m²/g. After the ethanol steam reforming reaction, the specific surface area for n(La)?n(Ni)?n(CA)=2?1?1 decreased dramatically, and that for n(La)?n(Ni)?n(CA)=2?1?4 decreased slightly, but the specific surface area for n(La)?n(Ni)?n(CA)=2?1?3 was almost the same. This suggests that the La₂NiO₄ catalyst with n(La)?n(Ni)?n(CA)=2?1?3 has the most stable structure.

Table 3 Specific surface area of La₂NiO₄ prepared with different n(La)?n(Ni)?n(CA)

The conversion of C₂H₅OH and the content of products at 500 ℃ and 650 ℃ over La₂NiO₄ with different molar radios of La to Ni to CA were studied and the results are shown in Table 4. Total ethanol conversion took place all along the 3 h runs at both reaction temperatures. The typical gaseous products H₂, CO₂, CO and CH₄ were obtained in all cases while other common intermediates or by-products like ethylene, diethylether and acetaldehyde were detected only in negligible amounts (＜0.10%). For every catalyst, the H₂ content was found to increase with temperature. By comparing the performance of catalysts at the same reforming temperature, the catalyst La₂NiO₄ with n(La)?n(Ni)?n(CA)=2?1?3 led to a higher H₂ content (71.6%) and a lower CH₄ content (0.10%) in the gaseous mixture than the catalysts La₂NiO₄ with n(La)?n(Ni)?n(CA)=2?1?1 and 2?1?4 at 650℃. This is ascribed to a higher rate of the CH₄ reforming reaction linked to a higher active metal dispersion degree seen from Fig.2.

Table 4 C₂H₅OH reforming using La₂NiO₄ catalysts prepared with different n(La)?n(Ni)?n(CA)

3.3 Stability of La₂NiO₄ catalyst

Fig.4 shows the conversion of CH₄, and the content of H₂, CO, CH₄ and CO₂ as a function of time on stream at 500 ℃ with the ethanol feed rate of 5.8×10^-5 mol/min and the ethanol-to-water-to-argon molar ratio of 1?9?14.3 over a La₂NiO₄ catalyst calcined at 700 ℃ with n(La)?n(Ni)?n(CA)=2?1?3. It is clear that during the first 55 h reaction ethanol was completely converted. When the operation time was longer than 56 h, the ethanol conversion started to decrease slightly and arrived about 97.5% at 80 h, and the selectivities of H₂, CO, CH₄ and CO₂ were stable at around 67.5%, 1.44%, 2.73% and 28.3%. Meanwhile, other common intermediates or by-products like ethylene, diethylether and acetaldehyde were detected only in negligible amounts (＜0.10%) in the whole 80 h, which suggests that the La₂NiO₄ catalysts are quite stable and catalytically highly active. In agreement with these results, Ni-La catalysts also showed constant activity under the same reforming conditions[12].

Fig.4 Stability of La₂NiO₄ catalyst for ethanol steam reforming reaction at 500 ℃

The X-ray diffraction patterns of the fresh, reduced and used La₂NiO₄ catalysts are shown in Fig.5. For the fresh La₂NiO₄ catalyst, typical diffraction peaks for the La₂NiO₄ spinel structure were observed. After the catalyst was reduced by hydrogen at 500 ℃ for 1 h, most of the diffraction peaks attributed to La₂NiO₄ disappeared, whereas those belonging to La₂O₃ appeared. And nickel existed chiefly as Ni⁰, and the average crystal size of Ni⁰ (21 nm) for La₂NiO₄ (after reduction), obtained from the XRD line broadening, was found to be much smaller than that (125 nm) for Ni-La₂O₃[13]. This suggests that the La₂NiO₄ has been reduced and decomposed by hydrogen to form nanoscale Ni particles, which are segregated by La₂O₃. LIU and AU[14] suggested that La₂O₃ can prevent transition metals from agglomeration and promote the dispersion of nanoscale Ni⁰ particles, resulting in an enhancement of catalytic activity and stability. After reaction, the catalysts showed similar patterns with the results of ZHANG and VERYKIOS[15]. The La₂O₃ phase that existed in the reduced catalysts disappeared, and the La₂O₂CO₃ phase was formed due to the adsorption of CO₂ on La₂O₃ (CO₂+La₂O₃→La₂O₂CO₃). The La₂O₂CO₃ phase that existed in the used La₂NiO₄ catalyst was mainly hexagonal. Ni⁰ particles were not observed, thus, Ni⁰ particles might be present in an amorphous form or they were highly dispersed. ZHANG and VERYKIOS[15] have noted that the carbon species formed on the Ni sites were easily removed by the oxygen species originating from La₂O₂CO₃ (La₂O₂CO₃+C*→La₂O₃+2CO+*), thus producing an active and stable catalyst due to the existence of synergetic sites which consist of Ni and La elements. As indicated in this experiment, La₂O₂CO₃ may play a crucial role in the ethanol steam reforming reaction using La containing catalyst prepared from perovskite precursors.

Fig.5 X-ray diffraction patterns of La₂NiO₄: (a) After calcination at 850 ℃; (b) After 1 h reduction at 500 ℃; (c) After reaction

The TG/DTG curves for the La₂NiO₄ catalyst after ethanol steam reforming reaction at 500 ℃ for 80 h are shown in Fig.6. The DTG curve of the used catalyst distinctly indicated a mass loss due to the removal of carbon. Three DTG features were observed at 363, 513 and 700 ℃. This indicated that at least three kinds of carbon depositions were formed on the La₂NiO₄. One of the carbon depositions was likely attributed to La₂O₂CO₃,

Fig.6 TG/DTG profiles of La₂NiO₄ catalysts after ethanol steam reforming reaction at 500 ℃ for 80 h

which had been verified by XRD. La₂O₂CO₃, formed by the interaction of La₂O₃ with CO₂, may decompose into CO and produce oxygen species, which react with the surface carbon species on the Ni sites, thus giving active and stable catalytic performance for the ethanol steam reforming reaction, indicating La₂NiO₄ catalyst has higher résistance to coke formation, as well as higher catalytic stability as shown in Fig.4 and Fig.6.

4 Conclusions

1) Better catalytic performances can be achieved using the perovskite La₂NiO₄ as catalyst precursor. Among the catalysts tested, the catalyst La₂NiO₄ prepared with n(La)?n(Ni)?n(CA)=2?1?3 calcined at   700 ℃ exhibits the best activity with excellent stability.

2) The XRD results confirm that La₂NiO₄ exhibits a typical spinel structure.

3) During the ethanol steam reforming reaction, the active Ni⁰ particles do not aggregate and sinter, and the catalyst shows good stability as well as higher resistance to coke formation within an on-stream time of 80 h.

References

[1] FRENI S, CAVALLARO S, MONDELLO N, SPADARO L, FRUSTERI F. Steam reforming of ethanol on Ni/MgO catalysts: H₂ production for MCFC [J]. Journal of Power Sources, 2002, 108(1/2): 53-57.

[2] BREEN J P, BURCH R, COLEMAN H M. Metal-catalysed steam reforming of ethanol in the production of hydrogen for fuel cell applications [J]. Applied Catalysis B: Environmental, 2002, 39(1): 65-74.

[3] VIZCA?NO A J, CARRERO A, CALLES J A. Hydrogen production by ethanol steam reforming over Cu-Ni supported catalysts [J]. International Journal of Hydrogen Energy, 2007, 32(10/11): 1450-1461.

[4] LIGURAS D K, KONDARIDES D I, VERYKIOS X E. Production of hydrogen for fuel cells by steam reforming of ethanol over supported noble metal catalysts [J]. Applied Catalysis B: Environmental, 2003, 43(4): 345-354.

[5] KUGAI J, SUBRAMANI V, SONG C, ENGELHARD M H, CHIN Y H. Effects of nanocrystalline CeO₂ supports on the properties and performance of Ni-Rh bimetallic catalyst for oxidative steam reforming of ethanol [J]. Journal of Catalysis, 2006, 238(2): 430-440.

[6] BATISTA M S, SANTOS R K S, ASSAF E M, ASSAF J M, TICIANELLI E A. Characterization of the activity and stability of supported cobalt catalysts for the steam reforming of ethanol [J]. Journal of Power Sources, 2003, 124(1): 99-103.

[7] FATSIKOSTAS A N, VERYKIOS X E. Reaction network of steam reforming of ethanol over Ni-based catalysts [J]. Journal of Catalysis, 2004, 225(2): 439-452.

[8] SHENG P Y, YEE A, BOWMAKER G A, IDRIS H. H₂ production from ethanol over Rh-Pt/CeO₂ catalysts: The role of Rh for the efficient dissociation of the carbon-carbon bond [J]. Journal of Catalysis, 2002, 208(2): 393-403.

[9] HARYANTO A, FERNANDO S, MURALI N, ADHIKARI S. Current status of hydrogen production techniques by steam reforming of ethanol: A review [J]. Energy Fuels, 2005, 19(5): 2098-2106.

[10] CHOUDHARY V R, UPHADE B S, BELHEKAR A A. Oxidative conversion of methane to syngas over LaNiO₃ perovskite with or without simultaneous steam and CO₂ reforming reactions: Influence of partial substitution of La and Ni [J]. Journal of Catalysis, 1996, 163(2): 312-318.

[11] AUPRETRE F, DESCORME C, DUPREZ D, CASANAVE D, UZIO D. Ethanol steam reforming over Mg_xNi_1-xAl₂O₃ spinel oxide- supported Rh catalysts [J]. Journal of Catalysis, 2005, 233(2): 464-477.

[12] SUN Jie, QIU Xin-ping, WU Feng, ZHU Wen-tao. H₂ from steam reforming of ethanol at low temperature over Ni/Y₂O₃, Ni/La₂O₃ and Ni/Al₂O₃ catalysts for fuel-cell application [J]. International Journal of Hydrogen Energy, 2005, 30(4): 437-445.

[13] CHOUDHARY V R, RANE V H, RAJPUT A M. Selective oxidation of methane to CO and H₂ over unreduced NiO-rare earth oxide catalysts [J]. Catalysis Letter, 1993, 22(4): 289-297.

[14] LIU B S, AU C T. Sol-gel-generated La₂NiO₄ for CH₄/CO₂ reforming [J]. Catalysis Letters, 2003, 85(3/4): 165-170.

[15] ZHANG Z, VERYKIOS X E. Carbon dioxide reforming of methane to synthesis gas over Ni/La₂O₃ catalysts [J]. Applied Catalysis A: General, 1996, 138(1): 109-133.

Corresponding author: Huang Qun-wu; Tel: +86-22-27404771; E-mail: huangqw@tju.edu.cn

DOI: 10.1016/S1003-6326(09)60048-0

(Edited by YANG Bing)