Enhancing performance of Ni/La₂O₃ catalyst by Sr-modification for steam reforming of toluene as model compound of biomass tar

Abstract

Steam reforming of biomass tar with toluene as the model compound was studied using Sr-doped Ni/La₂O₃ catalysts prepared using two methods, i.e. co-impregnation of Sr and Ni on La₂O₃ support (Ni–Sr/La₂O₃ catalyst) and sequential impregnation of Sr on Ni/La₂O₃ catalyst (SNL catalyst), which were then calcined at various temperatures (500, 700, and 900 °C). These two types of catalysts are found to possess better catalytic performance than the undoped Ni/La₂O₃ catalyst at the same calcination temperature due to the presence of Sr, which helps in water adsorption at low steam/carbon (S/C) ratio. Moreover, the catalytic performance for catalysts calcined at various temperatures decreases following this trend: 500 °C > 700 °C > 900 °C due to lower BET surface area and lower surface active metal available for reaction. In addition, it is also observed that the Sr/Ni/La₂O₃ catalyst has better performance than the Ni–Sr/La₂O₃ catalyst at the same calcination temperature. Further characterization results suggest that in the Ni–Sr/La₂O₃ catalyst, the Sr is present between Ni and La₂O₃ support. On the other hand, Sr in the Sr/Ni/La₂O₃ catalyst is thought to be located on the surface of Ni/La₂O₃ due to the preparation method. This study shows that more Sr on the catalyst surface has better catalytic activity and stability in steam reforming of toluene.

---

Introduction

The need for a renewable alternative of energy sources is increasingly urgent due to the rapid depletion of fossil fuel reserves and the costly price tag of crude oil. The utilization of biomass has been receiving more attention as an abundant and cheap source of renewable energy. Among them, biomass gasification has garnered considerable attention for hydrogen/synthesis gas production, which can then be used for Fischer Tropsch processes, syntheses of methanol and electricity production in the case of a fuel cell.^1–4

The main challenge in biomass gasification technology is the removal of tar, which is a mixture of various condensable hydrocarbon compounds containing toluene as the major component, followed by naphthalene.³ Furthermore, tar can polymerize to form more complex structures and aerosols. Tar can condense on particulate filters, heat exchangers, and engines, resulting in plugging and attrition. This further causes a decrease in efficiency of the process operations and an increase in maintenance and operating costs. Tar removal methods can be classified as primary or secondary. Primary methods include all measures taken in the gasification reactor to prevent tar from being formed in the gasifier or to convert it. In secondary methods, pretreatment, either chemical or physical, is performed downstream of the gasifier. Both primary and secondary methods require the use of a catalyst.

Various metal catalysts have been investigated, such as Ni^5–12 Co,^13–16 Fe,^{9–12,16–18} and noble metal¹⁹ for the steam reforming of biomass tar using toluene as the model compound. Among those metal catalysts, Ni catalyst has been used extensively not only in biomass tar conversion, but also in other reactions such as reforming of hydrocarbon^20–25 and water gas shift^26,27 reactions because it shows comparable activity and stability with the expensive noble metal catalysts such as Rh, Ir, Pt, and Pd. However, Ni catalyst is susceptible to deactivation mainly through carbon deposition on the surface of the Ni catalyst and sintering of the Ni⁰ metallic species. The agglomeration of the Ni⁰ particles occurs due to the typical reaction temperature (800 °C), which is higher than its Tammann temperature (above which Ni sintering occurs).²⁸ On the other hand, the undesirable coke deposition is due to the high temperature adsorption and dissociation of tars, and unsaturated hydrocarbons on the surface of the catalyst. It has been reported that coke formation can be minimized in various manners using catalyst modification. One of the modifications is to introduce redox^29,30 or oxygen storage properties into the catalyst.^31,32 Some metals such as Cu and Fe are well-known to have reduction–oxidation (redox) properties, which aid in the reduction of carbon deposition. Tomishige et al. reported the use of Fe in Ni–Fe/α-Al₂O₃ catalyst in the steam reforming of biomass tar.²⁹ The addition of Fe to Ni/α-Al₂O₃ enhanced catalytic activity by the formation of Ni–Fe alloy (conversion increase from 64% to 86%) and catalytic stability due to the presence of Fe on the catalyst surface, which increases the oxygen storage/vacancy of the catalyst.^29–32 Kawi et al.⁹ reported that the addition of CeO₂ to the Ni/CaO–Al₂O₃ catalyst resulted in less carbon formation compared to the unpromoted catalysts. A similar conclusion was reported by Atong et al.³³ for the substitution of Ce into LaNiO₃ perovskite-type oxide, Rirksomboon et al.³⁴ for the addition of Mn to Ni/CeO₂–ZrO₂ catalyst, Zhang et al.³⁵ for the addition of CeO₂ to Ni/Al₂O₃ catalyst, and Brown et al.³⁶ for the addition of CeO₂ to Ni/olivine catalyst. Another option is to use basic metal as the promoter for the catalyst.^37–44 Krause et al.⁴³ reported that higher basicity of catalyst is a favorable property for tar decomposition, resulting in the enhancement of catalyst stability. Choong et al.⁴⁴ also reported that the addition of CaO to Ni catalyst promoted water adsorption and provided an abundance of adsorbed OH groups to facilitate the C–C break in Ni catalyst, resulting in higher ethanol conversion in steam reforming of ethanol. The last option to reduce coke deposition is by operating the reaction at high reaction temperatures and excess steam-to-carbon ratio.⁴⁵

Our previous study^11,12 showed that the substitution of a small amount of Sr with La in the La_xSr_1−xNi_0.8Fe_0.2O₃ (LSNFO) catalyst showed good catalytic activity and stability, especially at a low amount of steam (steam/carbon = 1). It is found that Sr in LSNFO catalyst can adsorb and desorb more water at higher temperature than La. However, only La can activate the water. The synergetic effect of Sr and La helps to enhance the catalyst stability. Sekine et al. also reported that the Sr doping in the Ni/La_0.7Sr_0.3AlO_δ catalyst helps to improve the catalytic activity and reduce carbon formation due to lattice oxygen of the catalyst.^46–49 The similar conclusion of the Sr role on the enhancement of catalytic performance was reported in dry CO₂ reforming of methane (DRM) reaction.^20,22,50

Considering the important role of Sr at low steam amount in steam reforming of biomass tar, the catalyst development should be designed by depositing more Sr on the catalyst surface. The higher amount of Sr on the catalyst surface is expected to enhance not only catalyst stability, but also catalyst activity. Hence, in this study, two types of catalysts (Sr/Ni/La₂O₃ and Ni–Sr/La₂O₃ catalysts) were prepared using two conventional impregnation methods, i.e. co-impregnation and sequential impregnation to validate the hypothesis. The catalytic performance of the catalysts in terms of activity and stability is studied for steam reforming of toluene. Various characterizations of fresh, reduced, and spent catalysts are performed to show the crucial location of Sr on the catalyst for obtaining higher catalyst performance.

Experimental

Catalyst synthesis

Sr-doped-Ni/La₂O₃ catalyst was synthesized by either co-impregnation method or sequential impregnation method. For the co-impregnation catalyst, the Ni(NO₃)₂·6H₂O (Merck) and Sr(NO₃)₂ (Sigma Aldrich) were dissolved in deionized water (DI water). The La₂O₃ (Aldrich) was then added into the solution. The mixture was vigorously stirred in a water bath of 80 °C until it dried. Further, drying was performed at 100 °C overnight in an oven before it was subsequently calcined at 500 °C, 700 °C, or 900 °C for 4 hours.

For sequential impregnation catalyst, the Ni/La₂O₃ catalyst was first prepared according to the above procedure but with calcination duration of 2 hours. The calcined Ni/La₂O₃ catalyst was then impregnated with solution containing Sr(NO₃)₂ dissolved in DI water. The dry catalyst was calcined for another 2 hours. The catalysts are designated as NSL 500, NSL 700, and NSL 900 catalysts for co-impregnation Ni–Sr/La₂O₃ catalysts calcined at 500 °C, 700 °C, 900 °C, respectively, and SNL 500, SNL 700, and SNL 900 catalysts for sequential impregnation of Sr/Ni/La₂O₃ catalysts at 500 °C, 700 °C, 900 °C, respectively.

X-ray diffraction (XRD)

A plastic sample holder was filled with the powdered catalyst sample before its surface was pressed flat. Subsequently, the XRD profile was obtained using a Shimadzu XRD-6000 Diffractometer using a Cu target Kα-ray (40 kV and 30 mA) as the X-ray source. A scanning range (2θ) from 10° to 80°, with a scanning speed of 2° min⁻¹ and a step function of 0.02, was used. The receiving slit was set at 0.3 mm while the scattering and divergence slits were set at 1 mm.

Hydrogen temperature-programmed reduction (H₂-TPR)

TPR analysis was performed using a Thermo Scientific TPDRO 1100 system equipped with a Thermal Conductivity Detector (TCD). 50 mg of fresh catalyst was first degassed at 300 °C for 30 minutes with nitrogen gas flushing. This was followed by reduction under 5% H₂/N₂ gas mixture at a flow rate of 30 ml min⁻¹ at which the catalyst samples were heated from room temperature to 1000 °C at a heating rate of 10 °C min⁻¹. The TPR profiles were obtained by monitoring the changes in the signal readings of the TCD against temperature. The TCD signal was initially calibrated by measuring the peak area observed for each injection of known volume of H₂.

In order to calculate the number of active sites and amount of metal dispersion, pulse titration was performed on the Thermo Scientific TPDRO 1100 system according to the method reported in recent literature.⁵¹ The H₂-TPR was initially performed using 50 mg of sample up to 600 °C, followed by cooling down to room temperature. Purified N₂O gas was then introduced to the system by pulse injection until saturation. The H₂-TPR was repeated up to 600 °C. The number of active sites can be calculated from H₂-TPR after N₂O chemisorption, while the dispersion can be calculated by comparing the second TPR to the first TPR results.

X-ray photoelectron spectroscopy (XPS)

Prior to XPS analysis, the sample was reduced at 600 °C for 1 hour, which is the same reduction time as in the reaction study. The XPS was performed on a Kratos AXIS Ultra DLD using a monochromatized Al Kα X-ray source (1486.71 eV photons, 5 mA, 15 kV) at a constant dwell time of 100 ms. The binding energy peaks were fitted using the XPSPEAK software and the calibration of all binding energies was performed by referencing C1s (C–C bond) to 284.5 eV.

TGA-DTA

The spent catalysts were analysed using a Shimadzu DTG60 TGA-DTA analyser. The spent catalyst was heated from 25 °C to 800 °C at a rate of 10 °C min⁻¹. The weight loss from the spent catalyst was monitored and recorded continuously.

Water temperature-programmed desorption (TPD-H₂O)

To determine the water desorption ability of the catalyst, TPD-H₂O was performed in a fixed-bed reactor equipped with a Carbolite furnace and a Thermostar Mass spectrometer. 50 mg of fresh catalyst was first reduced in 80 ml min⁻¹ of H₂ at 600 °C for 30 minutes before the samples were cooled to 50 °C. Subsequently, deionised water was introduced to the sample through a Helium gas stream of 80 ml min⁻¹ for 30 minutes after which the sample was purged at 100 °C for 20 minutes. The catalyst sample was then heated to 600 °C at a heating rate of 10 °C min⁻¹, in which the effluent gas was continuously monitored using the mass spectrometer. The changes in the H₂O signal (m/z = 18.0) against temperature was recorded by a microcomputer.

Catalytic reaction

Catalytic reactions were carried out in a quartz tube fixed bed micro-catalytic reactor (I.D. 4 mm; length 60 cm), enclosed in a programmable furnace at atmospheric pressure. Approximately 30 mg of catalyst was used for each reaction run, held in a fixed bed using quartz wool. The inlet of the reactor is connected to a pre-heating coil at a temperature of 300 °C, which is used to vaporise the reactants (DI water and toluene). A condenser at a temperature of 5 °C is fitted to the outlet of the reactor to condense any unreacted water vapour and toluene vapour.

The fresh catalysts were first reduced under H₂ gas (30 ml min⁻¹) at 600 °C for 30 minutes. Subsequently, the H₂ was shut off, and the sample was purged by a He gas stream (120 ml min⁻¹) for 10 minutes. Finally, the reactants streams of both steam and toluene were introduced into the micro-catalytic reactor by switching on the pumps of water and toluene. The steam reforming reaction is performed at a temperature of 650 °C at atmospheric pressure. The system was allowed to stabilize for 10 minutes before the start of the analysis of the product gases. The amount and concentration of gaseous products were analysed using a gas chromatograph (HP 6890 series) equipped with a packed column (Carboxen-1000, 60/80mesh sizes) and a TCD detector. The total product volumetric flow rate was measured using a BIOS digital flow meter.

The chromatogram obtained for each reaction run showed peaks for H₂, CO and CO₂ product species. The area of each peak was calculated using the integral function of the offline software coupled to the gas chromatograph. Based on the peak areas, the volume percentage of each product gas was obtained using the calibration curves that were previously determined. The individual volumetric flow rate of the product gases was obtained based on their volume percentage and the total volumetric flow rate of the product gases.

Results and discussion

XRD – catalyst structure analysis

Fig. 1 shows the XRD of fresh Sr-doped Ni/La₂O₃ catalysts prepared using either co-impregnation or sequential impregnation method. Ni/La₂O₃ catalysts calcined at various temperatures were also synthesized and characterized for comparison. It can be seen that Ni/La₂O₃ 500 catalyst shows very broad peaks and significant decrease in crystallinity of La₂O₃ compared to the La₂O₃ support because calcination at 500 °C is unable to completely transform the lanthanum oxide hydroxide to lanthanum oxide. Ekkehard Füglein and Dirk Walter⁵² reported that lanthanum hydroxide can be obtained by the treatment of lanthanum oxide with water at ambient temperature under protective gas. They^52,53 also showed that lanthanum hydroxide will transform to lanthanum oxide in a two step mechanism: 2La(OH)₃ → 2LaOOH + 2H₂O at a temperature of 250–400 °C and 2LaOOH → La₂O₃ + H₂O at a temperature of 400–560 °C. With increasing calcination temperature, the peaks of La₂O₃ in Ni/La₂O₃ 700 catalyst become sharper indicating the increase in crystallinity, but it is still slightly lower than La₂O₃ support. When the Ni/La₂O₃ catalyst was calcined at 900 °C (Ni/La₂O₃ 900), many peaks such as perovskite, spinel, and NiO peaks (43° due to Ni(200)⁵⁴) can be observed. Ni metal has been reported in many literature sources to have interaction with La₂O₃ to form either LaNiO₃ perovskite structure^22,55 or La₂NiO₄ spinel structure depending on the Ni : La ratio and calcination temperatures.^56–58 In addition, the hexagonal phase of La₂O₃ was also observed instead of the cubic La₂O₃ phase, which was used as the catalyst support. It was reported that the cubic La₂O₃ phase can change to hexagonal La₂O₃ when it was calcined at temperatures above 750 °C.⁵⁹

Fig. 1 XRD of fresh catalysts [(●) Perovskite, (♠) Spinel, (♥) hexagonal La₂O₃, and (♦) NiO].

All the catalysts were then reduced and characterized using XRD. The results of XRD characterization of Sr-doped Ni/La₂O₃ catalysts are shown in Fig. 2. The Ni(111) peak at 2θ of 44.6° is observed for all NSL and SNL catalysts with sharper and more prominent peaks for catalysts calcined at higher temperatures, showing that the NSL and SNL catalysts calcined at a higher temperature have bigger crystal sizes than those calcined at a lower temperature.⁶⁰ In addition, it can be seen that reduced Ni/La₂O₃ 500, 700, and 900 catalysts show similar XRD patterns. The hexagonal structure of La₂O₃ is also observed in addition to the cubic La₂O₃ structure, which is the structure of the catalyst support. This result indicates the transformation of cubic La₂O₃ structure to hexagonal La₂O₃ structure through the H₂ reduction at 650 °C. However, different patterns are observed for NSL 500, 700, and 900 catalysts. The reduced NSL 500 and 700 catalysts show a similar pattern but higher intensity compared to the fresh NSL 500 and 700 catalysts, indicating that Sr may strengthen the interaction between Ni and La₂O₃. In contrast, the XRD pattern of reduced SNL 500, 700, and 900 catalysts is very similar to the one of reduced Ni/La₂O₃ 500, 700, and 900 catalysts, suggesting that Sr on these catalysts does not affect the interaction between Ni and La₂O₃. The preparation method in which Sr was intentionally impregnated after Ni also supports this idea.

Fig. 2 XRD of reduced catalysts [(♥) cubic La₂O₃, (♦) hexagonal La₂O_3, and (●) Ni].

TPR – catalyst reducibility analysis

In order to study the reduction properties of the prepared catalyst, the TPR experiment was conducted. The TPR profiles of Ni/La₂O₃ 500, 700, and 900 catalysts are also provided in Fig. 3a for comparison. The TPR results show 3 peaks for all catalysts, which shift to higher temperatures with increasing calcination temperature. The reduction peaks are commonly categorized into three zones: low temperature zone (300–500 °C) assigned to the reduction of NiO with a weak interaction with the catalyst support,²¹ intermediate temperature zone (500 to 600 °C) assigned to the reduction of NiO species with medium strength interaction with the support,⁶¹ and high temperature zone (>600 °C) attributed to the reduction of NiO species with strong interaction with the catalyst support.⁶² In this case, the strong interaction with the catalyst support can be in the form of compound formation such as perovskite or spinel. For Ni/La₂O₃ 500 catalyst, the reduction peaks are observed in the low temperature zone and high temperature zone. With increasing the calcinations temperature to 700 °C, the TPR reduction peaks in Ni/La₂O₃ 700 catalyst are separated clearly into three temperature zones: low, medium and high. Upon further increase in calcination temperature, the reduction peaks in the Ni/La₂O₃ catalyst are observed only in the intermediate and high temperature zones. The catalyst having weakly interacted NiO species generally exist in bigger particle sizes and is possibly easy to migrate and aggregate during reduction reaction.^63,64 In contrast, the catalyst having strong metal–support interaction is usually preferred as it has a small particle size after reduction.⁶⁵ However, it also strongly depends on the reduction temperature in the study. In addition, the H₂ consumption below 600 °C, which is the reduction temperature during the catalytic test, surface Ni amount and dispersion also decrease with increasing calcination temperature, which is well understood in the literature due to a stronger interaction between the metal and the support. This result shows that with increasing calcination temperatures, the metal has a stronger interaction with the support.

Fig. 3 TPR of reduced catalysts. (a) Ni/La₂O₃ catalysts and (b) Sr-doped Ni/La₂O₃ catalysts.

Fig. 3b shows the TPR profile of Sr-doped Ni/La₂O₃ catalyst. It can be seen that all peaks for SNL and NSL catalysts shift to higher temperatures due to a stronger interaction between Ni and the support. The peak shifting trend can be seen evidently, for example, increasing calcination temperature from 500 °C to 700 °C, the low reduction temperature peak zone (300–500 °C) due to weak interaction between NiO and the support in SNL and NSL 500 catalysts shift to a medium reduction temperature peak zone (500–600 °C) in SNL and NSL 700 catalysts. A similar trend is observed for catalysts calcined at 700 °C and 900 °C. Comparing the TPR profile between SNL and NSL catalysts calcined at the same temperature, the H₂ consumption below 600 °C, which is the reduction temperature during the catalytic test, amount of surface Ni and dispersion (Table 1) are about the same, indicating that the amount of Ni metal and its dispersion on both catalysts are the same. However, the peaks of the NSL catalysts for any calcination temperature are located at a higher temperature than those on the SNL catalysts, showing that the Ni has a stronger interaction with La₂O₃ support on the NSL catalyst. For example, the NSL catalyst has a peak at around 250–470 °C with the maximum peak at 420 °C, while the SNL catalyst also shows a peak at around 250–470 °C but with the maximum peak located at 400 °C. This stronger interaction on NSL catalysts compared to SNL catalysts could indicate that Sr helps to strengthen the interaction between Ni and La₂O₃ support on the NSL catalyst, while Sr atoms on the SNL catalyst are thought to be placed on the surface of the Ni/La₂O₃ catalyst due to the preparation method using sequential impregnation.

Table 1 Physicochemical properties of the prepared catalysts

Catalysts	Content (mmol g^−1)		H2 consumption^a (μmol g^−1)	BET surface area (m^2 g^−1)	Amount of surface Ni^b (μmol g^−1)	Dispersion^c (%)
	Ni	Sr
a H2 consumption below 600 °C in TPR profiles shown in Fig. 3.b Amount of surface Ni was calculated by H2-TPR after N2O chemisorption.c Calculated by comparing TPR-H2 before and after N2O pulse titration up to 600 °C.
Ni/La2O3 500	0.85	—	324.1	15.19	33.9	10.5
Ni/La2O3 700	0.85	—	215.0	10.43	19.3	8.9
Ni/La2O3 900	0.85	—	189.1	5.07	12.1	6.4
NSL 500	0.85	0.57	512.2	16.90	76.3	14.9
NSL 700	0.85	0.57	260.6	12.08	33.7	12.9
NSL 900	0.85	0.57	130.6	4.65	12.3	9.4
SNL 500	0.85	0.57	515.3	17.81	76.8	14.8
SNL 700	0.85	0.57	269.5	12.33	34.1	12.7
SNL 900	0.85	0.57	196.2	2.76	16.1	8.2

---

XPS of reduced catalysts

Fig. 4a shows the XPS Ni 3p profile of reduced undoped and Sr-doped Ni/La₂O₃ catalysts. The XPS Ni 2p profile is not shown here because the Ni 2p_3/2 peak overlaps with the La 3p_3/2 peak at around 853 eV. It can be seen that the main deconvoluted peak at around 65.8 eV in Ni/La₂O₃ 500 catalyst shifts to a lower binding energy for the Ni/La₂O₃ 700 catalyst (65.3 eV) and considerably lower binding energy for the Ni/La₂O₃ 900 catalyst (65.1 eV) due to stronger interaction between Ni and the La₂O₃ support. The similar peak location is observed for SNL catalysts. The binding energy for the SNL 500 catalyst is located at 65.8 eV, and it shifts to a lower binding energy (65.3 eV) for the SNL 700 catalyst due to stronger interaction between Ni and the La₂O₃ support. The peak shifting is more intense for the SNL 900 catalyst, which has binding energy around 65.1 eV due to a much stronger metal–support interaction. The same peak position between Ni/La₂O₃ catalysts and SNL catalysts shows that Sr does not significantly affect the electronic state of Ni on La₂O₃ support. Similarly, the peak location in the NSL catalysts decreases in the following order: NSL 500 catalyst (66.3 eV) > NSL 700 catalyst (65.4 eV) > NSL 900 catalyst (65.3 eV), but the peak position is located at a considerably higher binding energy, showing that the Ni in the NSL catalysts has transferred more electrons than the Ni in the SNL catalyst as an indication of stronger interaction of Ni with either Sr or La in NSL catalysts. This result also shows that with increasing calcination temperature, the interaction between Ni and the support becomes stronger, which is a similar observation with XRD and TPR results.

Fig. 4 XPS of reduced Sr-doped Ni/La₂O₃ catalysts for (a) Ni 3p and (b) Sr 3d.

The XPS Sr 3d profiles of reduced Sr-doped Ni/La₂O₃ catalysts are shown in Fig. 4b. The binding energy of the NSL catalysts shifts to a higher binding energy from NSL 500 to NSL 700 to NSL 900 catalysts, suggesting significant transfer of electrons for Sr in the catalyst calcined at a higher temperature. Because Sr has higher electronegativity than La and Ni, the electron transfer should be from Sr to La and/or Ni. However, the binding energy of the SNL catalysts is observed to be constant at 133 eV, which is due to SrO binding energy,⁶⁶ indicating that there is no electron transfer from Sr. This also indicates that the Sr is only located on the catalyst surface without any strong interaction with either Ni and La₂O₃ support. This observation is in-line with observations in the TPR and XRD experiments.

Catalytic activity and stability

Fig. 5 shows the catalytic activity of Sr-doped Ni/La₂O₃ catalysts prepared by either co-impregnation (NSL catalysts) or sequential impregnation (SNL catalysts) and calcined at various temperatures compared to the undoped Ni/La₂O₃ catalysts calcined at various temperatures. It can be seen that the catalytic activity of NSL and SNL catalysts is consistently higher than that of undoped Ni/La₂O₃ catalysts calcined at similar temperatures because of the higher amount of surface Ni in the Sr-doped Ni/La₂O₃ catalysts (Table 1). Comparing the effect of calcinations temperature on catalytic performance, one can clearly see that the catalytic performance of undoped Ni/La₂O₃ catalysts decreases in the order of 500 °C > 700 °C > 900 °C, due to a lower amount of surface Ni metal on the catalyst for the higher calcination temperature (Table 1) because higher calcination temperature can lead to stronger interaction between metals and the support, resulting in higher difficulty for reduction. This result is also in-line with H₂ consumption of the catalysts from the TPR result.

Fig. 5 Catalytic activity of Sr-doped Ni/La₂O₃ catalysts at 650 °C and S/C = 1. Reaction condition: toluene 188 μmol min⁻¹; steam 1316 μmol min⁻¹; He 5357 μmol min⁻¹; W = 30 mg.

Fig. 5 also shows that the SNL catalysts prepared using the sequential impregnation method always exhibit higher activity than NSL catalysts prepared using the co-impregnation method for the same calcination temperature. Because the amount of metal surface area is similar for NSL and SNL catalysts calcined at the same temperature, the sole difference lies in the preparation method, resulting in the disparity in the location of Sr. The above characterization results and the preparation method (sequential impregnation) suggest that all Sr is on the catalyst surface without strong interaction with Ni for the SNL catalysts because Sr was impregnated at the last step using the sequential impregnation method. However, for NSL catalysts, some of the Sr is located between the Ni and the La₂O₃ support, resulting in stronger interaction between the Ni and La₂O₃ support. Our previous study^11,12 reported that Sr can adsorb and desorb more water at higher temperature than La, but only La can activate the water. Therefore, this study shows that the crucial role of Sr to strongly adsorb and desorb water can be enhanced by arranging Sr on the catalyst surface.

Water desorption (TPD-H₂O)

In order to understand the characteristics of NSL and SNL catalysts, which have higher catalytic performance than Ni/La₂O₃ catalysts, TPD-H₂O was performed on reduced catalysts, and the result is shown in Fig. 6. It can be seen that one peak at around 300–400 °C due to desorption of water from La₂O₃ is observed for the Ni/La₂O₃ catalysts. This peak shifts to a higher temperature for catalysts calcined at lower temperature as the La is not strongly interacted with the Ni, and hence it has stronger desorption capacity. For NSL and SNL catalysts, two peaks are observed for all catalysts. The peak at lower temperature can be attributed to the water desorption from La₂O₃, while the one at higher temperature is due to water desorption from SrO.¹¹ In addition, both peaks shift to higher temperature for catalysts calcined at lower temperature as the La and Sr are not strongly interacted with the Ni. The water desorption from La₂O₃ and SrO in La_0.8Sr_0.2Ni_0.8Fe_0.2O₃ perovskite oxide catalyst was reported at around 270–350 °C and 400–500 °C, respectively.¹² Compared to this literature, the peaks in SNL catalysts appear at a higher temperature, showing the stronger adsorption of water on La and Sr in SNL catalysts. The difference can be explained by the absence of the Sr-incorporated perovskite structure in the SNL catalyst. The shifting of peaks to higher temperature in the SNL catalysts is even more intense as the Sr in the SNL catalysts was intentionally impregnated after Ni such that more Sr is present on the surface to adsorb and desorb water to enhance the catalytic activity and stability of the SNL catalysts. This result shows that a higher amount of strongly adsorbed water due to the presence of a greater amount of Sr on the surface results in enhancement of catalytic performance in the SNL catalyst.

Fig. 6 TPD-H₂O profiles of reduced Sr-doped Ni/La₂O₃ catalysts.

Carbon formation on spent catalysts

The main problem of catalyst deactivation in steam reforming of the toluene reaction is high carbon deposition on the catalyst surface from toluene decomposition. It is, therefore, important to characterize the spent catalyst using various methods to observe the type and amount of carbon deposition. Fig. 7 shows the FESEM images of spent NSL 500 and SNL 500 catalysts after reaction for 8 h. It can be seen that the spent NSL 500 catalyst clearly produces filamentous carbon, which is graphitic carbon that is difficult to oxidize at low temperature. Although this type of carbon did not result in deactivation of the nickel surface, it blocked the reactor and caused catalyst destruction.⁶⁷ However, it is hard to find any filamentous carbon on spent SNL 500 catalyst, indicating that the SNL 500 catalyst has higher stability than the NSL catalyst.

Fig. 7 FESEM images of spent SNL 500 and NSL 500 catalysts.

The quantitative analysis of carbon deposition on spent Ni/La₂O₃, NSL and SNL catalysts was performed using TGA analysis, and the results are shown in Fig. 8a. It can be seen that the amount of carbon deposition on spent Ni/La₂O₃ catalysts is considerably lower than either NSL or SNL catalysts due to its lower catalytic performance. However, the carbon formation on SNL catalysts is considerably lower than the one on spent NSL catalysts, even though the catalytic activity of the SNL catalysts is higher than that of the NSL catalysts. This result indicates that by allocating more Sr on the catalyst surface (SNL catalyst), it can not only enhance the activity, but also significantly reduce the carbon formation. In addition, the order of carbon formation rate in those catalysts calcined at various temperatures (NSL/SNL 500 ≈ NSL/SNL 700 > NSL/SNL 900) also follows the order of catalytic performance temperatures (NSL/SNL 500 > NSL/SNL 700 > NSL/SNL 900), showing that the carbon deposition rate is highly dependent on the catalytic activity.

Fig. 8 (a) Carbon formation rate on spent Sr-doped Ni/La₂O₃ catalysts. (b) Carbon-based yield of reaction products from Sr-doped Ni/La₂O₃ catalysts.

The carbon-conversion yield of deposited carbon compared to the CO and CO₂ produced from reaction and unreacted toluene is shown in Fig. 8b. The result clearly shows that the low amount of deposited carbon on Ni/La₂O₃ catalysts is due to low catalytic performance of the catalysts, which can be seen from the high amount of unconverted toluene. However, the low amount of deposited carbon on the SNL catalysts and low amount of unconverted toluene show that SNL catalysts can enhance the catalytic activity and reduce the carbon formation at the same time. On the other hand, the high amount of deposited carbon on NSL catalysts and relatively low amount of unconverted toluene show that NSL catalysts only enhance the catalytic activity without suppressing the deposited carbon. As a result, the deposited carbon on NSL catalysts increases.

Conclusions

Two types of catalysts were prepared using co-impregnation of Sr and Ni on La₂O₃ support (Ni–Sr/La₂O₃ catalyst/NSL) and sequential impregnation of Sr on Ni/La₂O₃ catalyst (Sr/Ni/La₂O₃ catalyst/SNL). The SNL catalysts exhibit higher performance than the NSL catalysts calcined at the same temperature due to the presence of more Sr on the catalyst surface for SNL catalysts because Sr was impregnated after Ni during the sequential impregnation method. The characterization using TPR, XRD, XPS, and TPD-H₂O suggests that some of the Sr in the NSL catalysts is located between Ni and La₂O₃ support and affects the electronic state of Ni. On the other hand, the preparation method and characterization data suggest that Sr in the SNL catalysts is located on the catalyst surface without strong interaction with Ni. This study shows that more Sr on the catalyst surface induces better catalytic activity and stability in steam reforming of toluene.

Acknowledgements

The authors acknowledge financial support from National University of Singapore and NEA-ETRP (Project no. 1002 114, Research Grant no. 279-000-333-490). Usman Oemar sincerely thanks Dr Yasotha Kathiraser and Dr Jangam Ashok for technical support and discussions.

Notes and references

1. G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106, 4044.
2. J. Han and H. Kim, Renewable Sustainable Energy Rev., 2008, 12, 397.
3. H. de Lasa, E. Salaices, J. Mazumder and R. Lucky, Chem. Rev., 2011, 111, 5404.
4. P. R. Buchireddy, R. M. Bricka, J. Rodriguez and W. Holmes, Energy Fuels, 2010, 24, 2707.
5. T. Kimura, T. Miyazawa, J. Nishikawa, S. Kado, K. Okumura, T. Miyao, S. Naito, K. Kunimori and K. Tomishige, Appl. Catal., B, 2006, 68, 160–170.
6. Z. A. El-Rub, E. A. Bramer and G. Brem, Ind. Eng. Chem. Res., 2004, 43, 6911.
7. D. Li, M. Koike, J. Chen, Y. Nakagawa and K. Tomishige, Int. J. Hydrogen Energy, 2014, 39, 10959.
8. C. Phuhiran, T. Takarada and S. Chaiklangmuang, Int. J. Hydrogen Energy, 2014, 39, 3649.
9. J. Ashok and S. Kawi, Int. J. Hydrogen Energy, 2013, 38, 13938.
10. U. Oemar, P. S. Ang, K. Hidajat and S. Kawi, Int. J. Hydrogen Energy, 2013, 38, 5525.
11. U. Oemar, M. L. Ang, W. F. Hee, K. Hidajat and S. Kawi, Appl. Catal., B, 2014, 148–149, 231.
12. U. Oemar, M. L. Ang, K. Hidajat and S. Kawi, AIChE J., 2014, 60, 4190.
13. C. Soongprasit, D. D. Aht-Ong and D. Atong, Adv. Mater. Res., 2013, 747, 690.
14. M. Koike, Y. Hisada, L. Wang, D. Li, H. Watanabe, Y. Nakagawa and K. Tomishige, Appl. Catal., B, 2013, 140–141, 652.
15. K. Soongprasit, D. Aht-Ong, V. Sricharoenchaikul and D. Atong, Top. Catal., 2013, 56, 339.
16. L. Wang, Y. Hisada, M. Koike, D. Li, H. Watanabe, Y. Nakagawa and K. Tomishige, Appl. Catal., B, 2012, 121–122, 95.
17. A. Laobuthee, C. Veranitisagul, W. Wattanathana, N. Koonsaeng and N. Laosiripojana, Renewable Energy, 2015, 74, 133.
18. N. Laosiripojana, W. Sutthisripok, S. Charojrochkul and S. Assabumrungrat, Fuel Process. Technol., 2014, 127, 26.
19. K. Tomishige, T. Miyazawa, M. Asadullah, S.-I. Ito and K. Kunimori, Green Chem., 2003, 5, 399.
20. K. Sutthiumporn, T. Maneerung, Y. Kathiraser and S. Kawi, Int. J. Hydrogen Energy, 2012, 37, 11195.
21. U. Oemar, K. Hidajat and S. Kawi, Appl. Catal., A, 2011, 402, 176.
22. K. Sutthiumporn and S. Kawi, Int. J. Hydrogen Energy, 2011, 36, 14435.
23. K. Yoshida, N. Begum, S. Ito and K. Tomishige, Appl. Catal., A, 2009, 358, 186.
24. L. Mo, K. K. M. Leong and S. Kawi, Catal. Sci. Technol., 2014, 4, 2107.
25. X. Wu and S. Kawi, Energy Environ. Sci., 2010, 3, 334.
26. E. T. Saw, U. Oemar, X. R. Tan, Y. Du, A. Borgna, K. Hidajat and S. Kawi, J. Catal., 2014, 314, 32.
27. M. L. Ang, U. Oemar, E. T. Saw, L. Mo, B. H. Chia and S. Kawi, ACS Catal., 2014, 4, 3237.
28. D. L. Trimm, Catal. Today, 1999, 49, 3.
29. L. Wang, D. Li, M. Koike, S. Koso, Y. Nakagawa, Y. Xu and K. Tomishige, Appl. Catal., A, 2011, 392, 248.
30. M. Koike, C. Ishikawa, D. Li, L. Wang, Y. Nakagawa and K. Tomishige, Fuel, 2013, 103, 122.
31. C. Li, D. Hirabayashi and K. Suzuki, Appl. Catal., B, 2009, 88, 351.
32. J. Nishikawa, K. Nakamura, M. Asadullah, T. Miyazawa, K. Kunimori and K. Tomishige, Catal. Today, 2008, 131, 146.
33. K. Soongprasit, D. Aht-Ong, V. Sricharoenchaikul and D. Atong, Curr. Appl. Phys., 2012, 12, S80.
34. A. Bampenrat, V. Meeyoo, B. Kitiyanan, P. Rangsunvigit and T. Rirksomboon, Appl. Catal., A, 2010, 373, 154.
35. Y. Lu, S. Li, L. Guo and X. Zhang, Int. J. Hydrogen Energy, 2010, 35, 7161.
36. R. Zhang, Y. Wang and R. C. Brown, Energy Convers. Manage., 2007, 48, 68.
37. D. Li, L. Wang, M. Koike, Y. Nakagawa and K. Tomishige, Appl. Catal., B, 2011, 102, 528.
38. K. Nakamura, T. Miyazawa, T. Sakurai, T. Miyao, S. Naito, N. Begum, K. Kunimori and K. Tomishige, Appl. Catal., B, 2009, 86, 36.
39. D. Świerczyński, S. Libs, C. Courson and A. Kiennemann, Appl. Catal., B, 2007, 74, 211.
40. J. S. Lisboa, D. C. R. M. Santos, F. B. Passos and F. B. Noronha, Catal. Today, 2005, 101, 15.
41. F. Bimbela, D. Chen, J. Ruiz, L. Garcia and J. Arauzo, Appl. Catal., B, 2012, 119–120, 1.
42. A. Carrero, J. A. Calles and A. J. Vizcaino, Chem. Eng. J., 2010, 163, 395.
43. T. Viinikainen, H. Rönkkönen, H. Bradshaw, H. Stephenson, S. Airaksinen, M. Reinikainen, P. Simell and O. Krause, Appl. Catal., A, 2009, 362, 169.
44. C. K. S. Choong, L. Huang, Z. Zhong, J. Lin, L. Hong and L. Chen, Appl. Catal., A, 2011, 407, 155.
45. D. Świerczyński, C. Courson and A. Kiennemann, Chem. Eng. Process., 2008, 47, 508.
46. Y. Sekine, D. Mukai, Y. Murai, S. Tochiya, Y. Izutsu, K. Sekiguchi, N. Hosomura, H. Arai, E. Kikuchi and Y. Sugiura, Appl. Catal., A, 2013, 451, 160.
47. D. Mukai, S. Tochiya, Y. Murai, M. Imori, T. Hashimoto, Y. Sugiura and Y. Sekine, Appl. Catal., A, 2013, 453, 60.
48. D. Mukai, S. Tochiya, Y. Murai, M. Imori, Y. Sugiura and Y. Sekine, Appl. Catal., A, 2013, 464–465, 78.
49. Y. Sugiura, D. Mukai, Y. Murai, S. Tochiya and Y. Sekine, Int. J. Hydrogen Energy, 2013, 38, 7822.
50. G. Valderrama, A. Kiennemann and M. R. Goldwasser, J. Power Sources, 2010, 195, 1765.
51. S. Tada, M. Yokoyama, R. Kikuchi, T. Haneda and H. Kameyama, J. Phys. Chem. C, 2013, 117, 14652.
52. E. Füglein and D. Walter, Z. Anorg. Allg. Chem., 2006, 632, 2154.
53. A. Neumann and D. Walter, Thermochim. Acta, 2006, 445, 200.
54. M. A. Gondal, T. A. Saleh and Q. A. Drmosh, Appl. Surf. Sci., 2012, 258, 6982.
55. C. Batiot-Dupeyrat, G. Valderrama, A. Meneses, F. Martinez, J. Barrault and J. M. Tatibouët, Appl. Catal., A, 2003, 248, 143.
56. Y. Kathiraser, W. Thitsartarn, K. Sutthiumporn and S. Kawi, J. Phys. Chem. C, 2013, 117, 8120.
57. B. S. Liu and C. T. Au, Appl. Catal., A, 2003, 244, 181.
58. W. D. Zhang, B. S. Liu, C. Zhu and Y. L. Tian, Appl. Catal., A, 2005, 292, 138.
59. N. Imanaka, T. Masui and Y. Kato, J. Solid State Chem., 2005, 178, 395.
60. W. Lin, H. Cheng, L. He, Y. Yu and F. Zhao, J. Catal., 2013, 303, 110.
61. A. Jones and B. D. Mc Nicol, Temperature-programmed Reduction for Solid Materials Characterization, Marcel Dekker, New York, 1986.
62. G. B. Sun, K. Hidajat, X. S. Wu and S. Kawi, Appl. Catal., B, 2008, 81, 303.
63. J. Gao, Z. Hou, J. Guo, Y. Zhu and X. Zheng, Catal. Today, 2008, 131, 278.
64. Z. Xu, Y. M. Li, J. Y. Zhang, L. Chang, R. Q. Zhou and Z. T. Duan, Appl. Catal., A, 2001, 210, 45.
65. Y. H. Hu and E. Ruckenstein, Catal. Lett., 1997, 43, 71.
66. Handbook of X-ray Photoelectron Spectroscopy, ed. J. F. Moulder, J. Chastain and R. C. King, Perkin-Elmer, 1992.
67. A. P. E. York, T. Xiao, M. L. H. Green and J. B. Claridge, Catal. Rev., 2007, 49, 511.

---