Methane decomposition into COx free hydrogen and multiwalled carbon nanotubes over ceria, zirconia and lanthana supported nickel catalysts prepared via a facile solid state citrate fusion method

Energy Conversion and Management 126 (2016) 302–315

Contents lists available at ScienceDirect

Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman

Methane decomposition into COx free hydrogen and multiwalled carbon nanotubes over ceria, zirconia and lanthana supported nickel catalysts prepared via a facile solid state citrate fusion method Manoj Pudukudy a,⇑, Zahira Yaakob a, Mohd Sobri Takriff b a b

Department of Chemical and Process Engineering, Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia, UKM, Bangi 43600, Selangor, Malaysia Research Center for Sustainable Process Technology (CESPRO), Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia, UKM, Bangi 43600, Selangor, Malaysia

a r t i c l e

i n f o

Article history: Received 24 June 2016 Received in revised form 28 July 2016 Accepted 5 August 2016

Keywords: Methane cracking Solid state citrate fusion Rare earth supports Nickel Hydrogen Multiwalled carbon nanotubes

a b s t r a c t Methane decomposition is the most effective route for the simultaneous production of COx-free hydrogen and nanocarbon. In this work, a set of porous ceria, zirconia and lanthana supported nickel catalysts were successfully synthesized via a facile solid state citrate fusion method and used for the thermocatalytic decomposition of undiluted methane for the first time. The catalysts were completely characterized for their crystalline, structural, textural and reduction properties and correlated to their catalytic performance. The active phase of fresh catalysts was found to be NiO in the CeO2 and ZrO2 supported catalysts whereas the formation of lanthanum nickel oxide solid solution was observed in the Ni/La2O3 catalyst. Various attractive porous morphologies of the fresh catalysts were confirmed by scanning electron microscopic studies. All of the catalysts exhibited high catalytic activity and stability for methane decomposition. The yield of hydrogen and carbon increased significantly with increasing the reaction temperature from 600 °C to 700 °C. A maximum initial hydrogen yield of 62%, 61% and 58% and a final carbon yield of 1360 wt%, 1159 wt% and 1576 wt% was achieved over ceria, zirconia and lanthana supported catalysts respectively, at 700 °C. The surface area of the catalysts could not have any significant effect on the catalytic efficiency and it was fully depended on the metal support interaction. The Ni/La2O3 catalyst showed high catalytic stability than ceria and zirconia supported catalysts due to the enhanced surface dispersion of finely crystallized Ni nanoparticles on the lanthana matrix aroused by the reduction of lanthanum nickel oxide. Moreover, bulk deposition of highly uniform multiwalled carbon nanotubes with high graphitization degree (ID/IG = 0.95) with different diameters depending on the Ni crystalline size was observed over the catalysts. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction In recent years, hydrogen is considered to be an efficient source of clean energy [1]. Currently, the large scale production of hydrogen is executed by the steam reforming or partial oxidation of natural gas [2]. However, the simultaneous production of highly toxic greenhouse gases such as CO and CO2 makes the process less effective, while considering the environmental issues like increased global warming and acid rain [3]. Methane decomposition is an alternative process for the production of pure hydrogen, where the production of greenhouse gases is completely avoided. ⇑ Corresponding author at: Department of Chemical and Process Engineering, Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia, UKM, Bangi 43600, Selangor, Malaysia. E-mail addresses: [email protected], [email protected] (M. Pudukudy). http://dx.doi.org/10.1016/j.enconman.2016.08.006 0196-8904/Ó 2016 Elsevier Ltd. All rights reserved.

In this reaction, methane is directly converted into COx free hydrogen and nanocarbon under anaerobic conditions [4]. For the non-catalytic decomposition reaction, high reaction temperatures of approximately 1200 °C are needed to crack the methane molecules [5]. However, the use of carbonaceous and metallic catalysts was reported to lower the reaction temperatures significantly [6]. Moreover, the nanocarbon deposited after reaction, mainly in the filamentous forms such as fibers or nanotubes, is of great importance in current nanoscience due to their exclusive electrical, chemical and mechanical properties [7]. Graphene is the other form of carbon obtained via this process. A number of Ni, Co and Fe based metallic catalysts were reported for methane decomposition [8]. Among these metals, Ni is identified as the most active metal for methane decomposition due to its exclusive 3d-orbital structures [9]. However, the catalytic performance of bare Ni is related to several parameters

M. Pudukudy et al. / Energy Conversion and Management 126 (2016) 302–315

as reviewed by Li et al. [10]. For an instance, it is reported that the activity of bare nickel catalysts is quite low at lower reaction temperatures and at high temperatures, due to the sintering and encapsulation of nickel by the nanocarbon, a faster deactivation of the catalyst was observed with a low catalytic stability. To improve the efficiency of nickel catalysts at moderate operating temperatures, several support materials were used during the synthesis of catalysts [11]. The use of a support material was already reported to promote the fine surface dispersion of metallic particles on the support and thereby increases the catalytic stability with high methane conversion and carbon yield [12]. Moreover, the textural properties of the support materials were effective to tune the surface chemistry, crystallinity and electronic characters of the metal particles via a migration effect as reported previously [13,14]. The commonly used catalyst supports for nickel in methane decomposition reaction are silica, alumina, titania, zeolites, magnesium aluminates, magnesia and so on [15–17]. Moreover, an appropriate metal-support interaction must be also considered before the selection of the catalyst supports. In addition to support, the catalytic activity of the catalyst mainly depends on the method of its synthesis [18]. A number of works was engrossed on the preparation of supported metal catalysts. Impregnation, deposition-precipitation, co-precipitation, solid state fusion, mechano-chemical activation and sol gel are the mainly used catalyst preparation methods in methane decomposition [19]. Eventhough many of the aforesaid methods are very effective for the development of improved catalytic properties; the solid state fusion is much more attractive since it is the most simple and one pot dry method for the preparation of catalysts with tunable properties such as fine metal dispersion and a proper metal support interaction [20]. Moreover, it devoid many complex and time consuming steps associated with the conventional wet synthesis methods such as washing, filtration, and drying. Thus a combination of the solid state fusion method and use of rarely studied support materials like ceria, zirconia and lanthana for nickel catalysts is highly interesting in the methane decomposition catalysis. So far, there are only limited reports available for methane decomposition using rare earth metal oxides such as ceria and lanthana and transition metal oxide zirconia as the catalyst supports [21]. Ceria as a support material in methane decomposition is highly credited because of its ability to promote the metal dispersion and also the metal support interaction. Li et al. [22] studied methane decomposition over 10%Ni/CeO2 catalysts prepared by different methods. As per their report, at a reaction temperature of 500 °C, the catalyst prepared by the impregnation and deposition-precipitation methods showed high hydrogen yield compared to the catalyst prepared by the co-precipitation method. This is attributed to the strong metal support interaction found in the coprecipitated catalyst, due to the formation of a Ni-O-Ce solid solution. Tang et al. [23] reported a set of Fe loaded CeO2 catalysts for methane decomposition. According to their study, a catalyst with the composition of 60 wt% Fe2O3 and 40 wt% CeO2 showed the highest catalytic activity in terms of hydrogen yield. However, due to the chemical interaction of deposited nanocarbon with the lattice oxygen of ceria, small amounts of COx were detected in the product gas. Odier et al. [24] also reported the same, over Pt/CeO2 catalyst. However, an improved hydrogen formation was observed over the catalyst, due to the spillover effect of noble metals. Zirconia supported catalysts are very rarely reported for methane decomposition. Kogler et al. [25] studied methane decomposition over zirconia, yttria-stabilized zirconia, and yttria catalysts, under dry and steam reforming conditions for the deposition of nanocarbon. They stated that the nanoparticulate zirconia is able to catalyze the growth of single and multiwalled carbon nanotubes via the chemical vapour deposition of methane.

303

Wolfbeisser et al. [26] reported a Ni/ZrO2 catalyst and its Cu promoted one for methane decomposition. As per their report, both of the catalysts showed high activity and stability for methane decomposition. Moreover, an activation process was occurring in methane between 377 °C and 462 °C range, resulting in an irreversible modification of the catalytic activity of the NiCu system towards Ni-like behaviour. The rapid increase catalytic activity was due to the increased concentration of nickel on the surface of the zirconia support. Moreover, the Cu promotion increased the coke resistance capacity of the Ni/ZrO2 catalyst [27]. Kurasawa et al. [28] reported methane decomposition over Ni/ZrO2 catalysts synthesized via a glycothermal method and reported that a highest carbon yield was obtained for the zirconia supported catalyst compared to other support materials. Multiwalled carbon nanotubes with the diameters of 30 nm were formed over the Ni/ZrO2 catalysts via methane decomposition. Lanthana based perovskite catalysts were also reported for methane decomposition. This is based on the fact that the reduction of perovskites produces nanosized metal species, which keeps an appropriate metal support interaction and thereby enhances the efficiency of the catalyst in terms of hydrogen and carbon yield. Liu et al. [29] studied methane decomposition over Fe nanoparticles obtained by the reduction of LaFeO3 perovskites and stated that bulk amounts of single walled carbon nanotubes with a narrow diameter distribution of 0.8–1.8 nm were formed over the catalysts. The growth of SWCTs over the aforesaid catalyst was due to the deposition of highly uniform and closely distributed Fe nanoparticles on the lanthana matrix. Chen et al. [30] reported lanthana supported NiCo bimetallic catalysts for methane decomposition. According to them, a high catalytic activity was obtained for the bimetallic catalysts without any pre-reduction steps. Kuras et al. [31] studied methane decomposition over nickel-lanthana perovskites and reported an improved catalytic activity for the production of multiwalled carbon nanotubes at high temperatures, irrespective of the time for catalyst reduction and temperatures. According to them, the aforementioned parameters do not have any effect on the nickel particle size on the lanthana support. Gallego et al. [32] reported a LaNiO3 perovskite prepared by the self-combustion method for methane decomposition at 600 °C and 700 °C. The stated that the catalysts exhibited high catalytic stability for the production of hydrogen and multiwalled carbon nanotubes, even after a long duration of time on stream of 1320 min at 700 °C. Rivas et al. [33] reported methane decomposition over a nickel catalyst obtained by the activation of LaNiO3 perovskite. The high catalytic stability of the nickel catalyst was due to the surface stabilization of nanosized NiO together with its fine dispersion on the lanthana matrix, which further reduced the sintering of nickel particles. Cunha et al. [34] reported an improved catalytic stability of the Fe/La2O3 catalyst for methane decomposition. They stated that, as a textural promoter, lanthana has the ability to tune the surface area of the catalyst effectively and acts like an electronic promoter. Recently Maneerung et al. [35] reported Ni, Co and Fe based bimetallic lanthana perovskites as catalyst precursors for methane decomposition and reported that the LaNi0.8Fe0.2O3 and LaNi0.8Co0.2O3 perovskites showed high catalytic performance in terms of hydrogen and carbon yield. Both chain and tube -like carbon nanofilaments were deposited over the catalysts after methane decomposition. However, in this manuscript, a set of ceria, zirconia and lanthana supported nickel catalysts were prepared via a facile solid state fusion method using citric acid as the surfactant, and characterized for their crystalline, structural, morphological and textural and redox properties. The catalysts were also successfully used for the thermocatalytic decomposition of undiluted methane at moderate reaction temperatures of 600 °C, 650 °C and 700 °C without any diminishing effect for a period of 360 min of time on the

304

M. Pudukudy et al. / Energy Conversion and Management 126 (2016) 302–315

stream. The properties of the catalysts were tried correlate with their catalytic performance. Moreover, the nanocarbon deposited after methane decomposition reactions were analyzed for their crystalline and structural properties using X-ray diffraction, electron microscopy and Raman analysis. To the best of our knowledge, there is no report on the synthesis, characterization and comparative study of the ceria, zirconia and lanthana supported Ni catalysts prepared via a facile solid state citrate fusion method for methane decomposition. Moreover, a catalyst support dependent characterization of the nanocarbon was studied. 2. Experimental 2.1. Synthesis of ceria, zirconia and lanthana supported nickel catalysts A simple and easily performable citric acid assisted solid state fusion method was employed for the synthesis of supported monometallic nickel catalysts. Cerium nitrate hexahydrate (Ce(NO3)3
6H2O), Zirconyl oxynitrate hydrate (ZrO(NO3)2
xH2O), Lanthanum nitrate hexahydrate (La(NO3)3
6H2O) and Nickel nitrate hexahydrate (Ni(NO3)3
6H2O) were purchased from R&M Chemicals and used as support (CeO2, ZrO2 and La2O3) and metal (Ni) precursors respectively. The amount of Ni in the fresh catalysts was kept 20 wt%. In the typical synthesis, 0.1 mol of support precursor and 0.2 of mol citric acid monohydrate (Fischer Scientific UK limited, UK) were mixed and ground in a mortar using pestle. Next the weighed amount of nickel nitrate hexahydrate was added to the mixture and ground again until it becomes a sticky pasty material. The material was then transferred to a quartz crucible, fused at 100 °C for 1 h in an oven and calcined at 700 °C for 5 h in a muffle furnace to obtain the ceria, zirconia and lanthana supported nickel catalysts. The catalysts were labeled as Ni/CeO2, Ni/ZrO2 and Ni/La2O3. 2.2. Characterization of fresh and spent catalysts To determine the actual amount of nickel in the fresh catalysts, the metal species were extracted using a hot acid solution of HF, HCl and HNO3 in 1:1:3 vol. ratios at 50 °C for 60 min. Afterwards, the solution was diluted to pH 2 using deionized water and was analyzed by an inductively coupling plasma optical emission spectroscopy (ICP-OES). The physicochemical characterization of the fresh catalysts were performed by means of powder X-ray diffraction (XRD), field emission scanning electron microscopy (FESEM), transmission electron microscopy (TEM), nitrogen adsorption– desorption analysis and temperature programmed reduction (TPR) measurements. The XRD patterns of the fresh and spent catalysts were obtained on a D8 Advance Diffractometer (Bruker) using Cu Ka radiation wavelength of k = 1.5406 Å with a step size of 0.025° in the angle range of 5–80°. The average crystallite size was calculated based on the Scherrer’s equation using intense diffraction peaks. The FESEM images of the fresh catalysts and the produced nanocarbon were acquired with a ZEISS MERLIN COMPACT field emission scanning electron microscope, operated at an accelerating voltage of 3 kV. The Energy Dispersive X-ray spectroscopy (EDS) attached to the FESEM microscope was used to verify the catalyst composition and also to determine actual metal content in the catalyst. Mapping analysis was also carried out to observe the metal dispersion of the fresh catalysts. The TEM images of the catalysts and nanocarbon were obtained using a Philips CM-12 TEM instrument controlled at an accelerating voltage of 100 kV. The Nitrogen sorption experiments were carried out in a Micromeritics ASAP 2020 apparatus at 196 °C. The catalyst samples were previously degassed at 300 °C for 6 h before the

analysis. The specific surface area of the catalysts was determined according to the Brunauer-Emmett-Teller (BET) method. The total pore volume and pore diameter distributions were calculated based on the nitrogen desorption isotherms by Barrett-JoynerHalenda (BJH) method. The temperature programmed reduction measurements for the reduction properties of the catalysts were carried out in a Micromeritics Autochem 2920 chemisorption analyzer from room temperature to 800 °C under a flow of 20 ml/min 20% H2/N2 gas mixture and a heating rate of 10 °C per minute. The hydrogen consumption was monitored by a gas chromatograph (GC) equipped with a thermal conductivity detector (TCD). The crystallinity and graphitization degree of the carbon deposited over the catalysts were studied by Raman analyses. The Raman analysis of the deposited nanocarbon was carried out in a WItec Raman spectrometer (Alpha 300R) equipped with an Nd: YAG laser diode and an excitation wavelength of 532 nm from 10 to 4000 cm 1 [36]. 2.3. Methane decomposition experiments The catalytic activity of the synthesized catalysts was evaluated for methane decomposition and the reactions were carried out in an experimental setup as described in our previous works [36–39]. The decomposition experiments were performed in a vertical tubular up flow cracking reactor made of stainless steel 2520 heated by an electric muffle furnace. The tubular reactor (length of 60 cm, outer diameter of 3.1 cm and an inner diameter of 2.5 cm) comprises three compartments and each was coupled with separate thermocouples. The weighed amount of powder catalyst (2 g) was packed in the middle of the reactor using thermal resistant quartz wool, where an internal thermocouple is inserted to monitor the actual reactor temperature. Then the catalyst was reduced in situ with a continuous flow of hydrogen (150 ml/min) for 90 min at 600 °C. After reduction, the reactor was flushed with nitrogen to raise the reactor temperature for an isothermal reaction. The experiments were carried out at 600 °C, 650 °C and 700 °C under atmospheric pressure using undiluted methane (99.95%, NIG Gases, Malaysia) with a flow rate of 150 ml/min for 6 h. During the reaction, outlet gas were collected in gas bags at fixed time intervals and were analyzed by a gas chromatograph (SRI GC 8610C) equipped with a Molecular sieve 5A column connected to a thermal conductivity detector (TCD) with helium as the carrier gas. The hydrogen yield was then calculated using the calibrated data, to express the catalytic activity. No gases other than hydrogen and unreacted methane were detected in the reaction products. After reaction, the reactor was cooled to room temperature with a constant flow of nitrogen (50 ml/min), in order to collect the deposited carbon for further characterization. 3. Results and discussion 3.1. Catalyst characterization ICP-OES analysis was used to study the chemical composition of the freshly prepared catalysts. The actual amount of nickel in the fresh catalysts was estimated using this method and the results are shown in Table. 1. The amount of nickel in the catalysts was

Table 1 Actual chemical composition of the fresh catalysts. Catalyst

Nominal (wt%)

Actual (wt%) (ICP-OES)

Actual (wt%) (EDX)

Ni/CeO2 Ni/ZrO2 Ni/La2O3

20 20 20

18.4 22.9 16.4

16.2 25.6 17.6

M. Pudukudy et al. / Energy Conversion and Management 126 (2016) 302–315

found to be quite lower than the nominated values. However, it is only a minimal variation and which is in acceptable range. In addition to ICP analysis, the quantitative and qualitative chemical compositions of the fresh catalysts were studied using EDX analysis and the spectra are shown in Fig. 1. As shown in Fig. 1, the Ni/ CeO2 catalyst contains Ni, Ce and O (Fig. 1(a)), Ni/ZrO2 catalyst contains Ni, Zr and O (Fig. 1(b)) and Ni/La2O3 catalyst contains Ni, La and O (Fig. 1(c)) as the main elements. No other elemental impurities were detected in the EDX spectra, indicating the high purity of the prepared catalysts. The actual surface composition of the metallic nickel species measured by the EDX analysis is also tabulated in Table. 1. Only a minimal variation was observed in the actual and nominated nickel amounts in the catalysts, which is highly consistent with the ICP analysis. The crystal structure and phase purity of the fresh catalysts were studied using XRD and the diffraction patterns are shown

305

in Fig. 2. The sharp diffraction peaks centered at the 2h values of 28.5°, 32.99°, 47.5°, 56.3°, 59.2°, 69.5°, 76.6° and 78.99° in Fig. 2 (a) were directly indexed to the face-centered cubic phase crystalline structure of Ceria (Cerianite, syn) with the lattice parameter of a = 0.5411 nm (JCPDS: 00-043-1002). The other less intense diffraction peaks observed at 37°, 43.2°, 62.8°, 76.7° and 79.1° were ascribed to the presence of NiO in the catalyst. The weak diffraction peaks suggests that the NiO was finely dispersed on the CeO2 support. The absence any other diffraction peaks indicated the phase purity of the catalytic species and confirms the absence of any chemical interaction between CeO2 and NiO species. As shown in Fig. 2(b), the sharp diffraction peaks highlighted at the 2h values of 28.2°, 30.3°, 34.8°, 35.5° 50.4°, 59.5° and 60.3° were directly indexed to the phase pure crystalline structure of tetragonal zirconia with the lattice parameters of a = 0.3598 nm and c = 0.5152 nm (JCPDS: 00-050-1089). The other well resolved intense diffraction

Fig. 1. EDX spectra of the freshly prepared catalysts (a) Ni/CeO2, (b) Ni/ZrO2 and (c) Ni/La2O3.

306

M. Pudukudy et al. / Energy Conversion and Management 126 (2016) 302–315

that the controlled release of huge amounts of gaseous species during the annealing process could be responsible for the porosity in the catalysts [40]. Fig. 3(d–f) represents the FESEM images of the Ni/ZrO2 catalysts. A microflake like morphology was again observed for the catalysts, but it is entirely different from that of Ni/CeO2 catalysts. The thickness of the flakes was found to be quite higher than that of the Ni/CeO2 flakes. Also the size of the flakes was greatly varied from several nanometers to micrometers, both in length and width. The thickness of the microflake shown in Fig. 3(f) was measured to be 600 nm. Furthermore, the flakes were observed to be composed of small pseudo spherical particles with nanosize and which are found to be highly inter-aggregated to provide a porous texture in the cross sectional like image as shown in Fig. 3(f). These small particles were also randomly dispersed on the surface of the flakes as shown in Fig. 3(e). A special porous morphology was observed for the Ni/La2O3 catalyst as shown in Fig. 3(g–i). As like other catalysts, the Ni/La2O3 catalyst also provided a microflake like external morphology with different size and thickness. The thicknesses of the flakes were observed to be ranging from nano to micro scales as shown in Fig. 3(g). A more ordered and homogeneous porous appearance was observed for the catalyst surface. The well-developed ordered pores aroused by the homogeneous inter aggregation of the twinned particles was visible in the images shown in Fig. 3(h, i). These types of porous texture were very rarely reported in literature for the metal loaded catalysts. However, the controlled release of huge amount of gases from the bulk of the fused sample could be responsible for the development of this visible porous morphology. Thus it is clear from this discussion that the surface morphology of the catalyst was fully depended on the nature of support used in the synthesis, irrespective of the role of citric acid. The SEM-mapping analysis was carried out to verify the surface dispersion of the metallic (nickel) species on the support materials. As shown in supplementary data Fig. SD1 to Fig. SD3, the active metal, Ni was found to be finely dispersed on the surface of the catalyst support. The internal structure and porous texture of the fresh catalysts were studied using TEM analysis and images are shown in Fig. 4. All of the fresh catalysts exhibited a sheet like appearance, irrespective of the type of support materials used. The sheet like structures contains pseudo spherical particles and which were found to be highly inter-aggregated to provide a porous texture. The tiny pores were clearly visible in the Ni/ZrO2 catalyst, which are found to be formed from the voids created by the inter-aggregation of nanosized more or less spherical shaped particles as observed in the FESEM images. The particles and pore structure of the Ni/ La2O3 catalyst was quite similar to that observed by the FESEM studies. The twinned particles were appeared to be loosely interaggregated, resulting in the formation of larger pores. The N2 physisorption BET/BJH analysis was used to study the textural properties of the catalysts. Table 2 shows the textural parameters of the catalysts. As shown, the BET surface area of ceria, zirconia and lanthana supported nickel catalysts was found to be 42.1 m2/g, 23.8 m2/g and 10.6 m2/g respectively. Among the samples, the Ni/CeO2 catalyst showed highest specific surface area, with a mean pore size and a pore volume distribution of 13.9 nm

Fig. 2. X-ray diffraction patterns of the fresh catalysts (a) Ni/CeO2, (b) Ni/ZrO2 and (c) Ni/La2O3.

peaks located at the 2h values of 37°, 43.2°, 62.8°, 76.7° and 79.1° were indexed to the rhombohedral crystalline structure of NiO (JCPDS: 01-089-3080). As like Ni/CeO2 catalyst, there is no chemical interaction was observed between NiO and ZrO2 in the catalyst. The diffraction peaks observed at the 2h values of 24°, 28.1°, 31.3°, 32.8°, 42.6°, 43.6°, 47.16°, 53.6°, 55.9°, 57.8°, 65.3°, 68.6°, 75.8° and 78.1° in Fig. 2(c) can be directly indexed to the formation of facecentered orthorhombic crystalline structure of lanthanum nickel oxide (La2NiO4) solid solution with the lattice parameters of a = 0.5459 nm, b = 0.5465 nm and c = 1.2687 nm (JCPDS: 01-0742924). No formation of phase segregated NiO and La2O3 was observed in the Ni/La2O3 catalyst, indicating the strong chemical interaction between the nickel oxide and lanthana support. Furthermore, compared to Ni/CeO2 and Ni/ZrO2 catalysts, the diffraction peak intensities of Ni/La2O3 catalyst were significantly lower and the peaks are broadened. It indicates the reduced crystalline quality of the lanthanum nickel oxide. The mean crystalline size of the various crystalline phases present in the fresh catalysts calculated using the Scherrer’s equation was presented in Table. 2. The surface morphology of the freshly prepared catalysts was studied using FESEM analysis and the images at different magnifications are shown in Fig. 3. Fig. 3(a–c) shows the SEM images of the Ni/CeO2 catalyst, where the microflake- like external morphology of the catalyst was observed. At low magnified images the catalysts were appeared to be bigger clusters and it was realized to be crushed into flakes (Fig. 3(a, b)). The sizes of the flakes were found to be not homogeneous. It is randomly speckled to several micrometers. Furthermore, the surfaces of the microflakes were noted to be very fine and smooth. However, the cross sectional views like images of the flakes shows a special interior porous structure in the catalyst as observed in Fig. 3(c). The pores were clearly visible in the images. The sizes of pores were wide ranging in the nanometer scale. Due to the presence of interior porosity, the flakes provide a crunchy or crispy texture to the catalyst. It is supposed

Table 2 Crystalline size and textural properties of the fresh catalysts. Catalyst

Ni/CeO2 Ni/ZrO2 Ni/La2O3

XRD crystalline size (nm)

BET/BJH textural properties

NiO

Support

Other phases

Surface area (m2/g)

Mean pore size (nm)

Pore volume (cm3/g)

– 8.2 –

7.6 13.6 –

– – 9.8 (NiLa2O4)

42.1 23.8 10.6

13.9 16.6 23.8

0.1576 0.0932 0.0613

M. Pudukudy et al. / Energy Conversion and Management 126 (2016) 302–315

307

Fig. 3. FESEM images of the fresh catalysts (a–c) Ni/CeO2, (d–f) Ni/ZrO2 and (g–i) Ni/La2O3.

and 0.1573 cm3/g respectively, whereas the lowest surface area was observed for the lanthana supported nickel catalyst with an average pore size and pore volume of 23.85 nm and 0.0613 cm3/g respectively. The Ni/ZrO2 catalyst showed slight lower surface area than the ceria catalyst, with an average pore size of 16.6 nm. The nitrogen adsorption desorption isotherms of the catalysts fully validated this observation and plots are shown in Fig. 5. All of the catalysts were found to exhibit a type IV isotherm with H3 hysteresis loop, indicating the presence of slit-like mesopores in the catalysts, which could be aroused from the voids created by the inter-aggregation of catalyst particles [41]. Moreover it can be seen that the Ni/CeO2 catalyst showed a wide hysteresis loop in the relative pressure range of 0.8–1. The Ni/ZrO2 catalyst also presented a wide hysteresis loop in the P/P0 range of 0.45–1, whereas a narrow hysteresis loop was observed for the Ni/La2O3 catalyst, which is highly consistent with the specific surface area of the catalysts, since the hysteresis loops are generated due to the capillary condensation of nitrogen within the mesopores of the catalysts and hence its surface area was correspondingly affected [42]. The reduction properties of the catalysts were studied using temperature programed reduction analysis and the TPR profiles are plotted in Fig. 6. The support material dependent reduction peaks are observed in all of the catalysts. As shown in Fig. 6, the Ni/CeO2 catalyst showed three reduction peaks in the temperature range of 150 °C to 400 °C. The less intense H2 consumption peaks centered at 162° and 182 °C can be attributed to the partial reduction of surface oxygen of CeO2 and the main peak centered at 306 °C could be ascribed to the reduction of NiO, which had a very weak interaction with the CeO2 support [43–47]. The other broad and weak peak appeared from 550 °C to 800 °C could be assigned to the reduction of lattice oxygen from the bulk of CeO2

as reported by Wang and Lu [48]. Previously Valentini et al. [49] reported that the reduction peaks of pure CeO2 could see at high temperature regions. However, after nickel loading, the reduction temperatures were shifted to low temperature region. This is attributed to the presence of nickel oxide species on the surface of CeO2, which acts like a catalyst for the reduction of ceria and a left shift of the reduction temperature was resulted. It points to the fact that the nickel has a positive impact on the reduction behaviour of CeO2 [49]. The Ni/ZrO2 catalyst also showed one intense reduction peak from 290 °C to 500 °C with a maximum at 371 °C. This intense peak could be ascribed to the reduction of non-interacting NiO species on the surface of ZrO2 support or which is weakly interacted with the support. Moreover, no high temperature reduction peaks were noticed. For Ni/La2O3 catalyst, a wide reduction peak was observed from 220 °C to 620 °C with several well resolved peak maximas at 349 °C, 380 °C, 412 °C and an intense and strong peak was observed at 521 °C. The sharp peaks located in the temperature range of 340–450 °C could be due to the reduction of NiO species on the surface of the catalyst, which is feebly interacted or uninfluenced by the lanthana support. However, the main reduction peak centered at 521 °C could be attributed to the reduction of spinel nickel lanthanum oxide solid solution present in the catalyst [50]. Moreover, it is worth to mention that the reduction of such perovskite species stabilizes the formation of nanosized metal crystallites on the support, under applied reduction conditions and keeps an appropriate metal support interaction and surface chemistry with the support, which may enhance the efficiency of the catalytic reaction [29,31,51]. Fig. 7 shows the XRD patterns of the reduced catalysts. The diffraction peaks in Fig. 7(a), located at the 2h values of 28.5°, 33.1°, 47.5°, 56.4°, 59°, 69.5°, 76.7° and 79.1° were directly indexed

308

M. Pudukudy et al. / Energy Conversion and Management 126 (2016) 302–315

Fig. 4. TEM images of the fresh catalysts (a, b) Ni/CeO2, (c, d) Ni/ZrO2 and (e, f) Ni/La2O3.

Fig. 5. BET - Nitrogen adsorption desorption isotherms of the fresh catalysts.

Fig. 6. TPR profiles of the fresh catalysts (chemisorption analysis).

M. Pudukudy et al. / Energy Conversion and Management 126 (2016) 302–315

309

Fig. 7. XRD patterns of the reduced catalysts (a) Ni/CeO2, (b) Ni/ZrO2 and (c) Ni/ La2O3.

to the face centered cubic phase crystalline structure of CeO2 (JCPDS: 01-075-9470). As shown in Fig. 7(b), the sharp diffraction peaks located at the 2h values of 28.1°, 30.2°, 31.8°, 34.4°, 35.1°, 40.7°, 47.5°, 50.3°, 54.4°, 60.1°, 62.8° and 65.6° were indexed to the crystalline phase of ZrO2. Both monoclinic and tetragonal phases of zirconia were detected (JCPDS: 00-037-1484, 01-0811544). The diffraction peaks centered at the 2h values of 27.6°, 29.2°, 29.9°, 39.4°, 46.1°, 48.5°, 55.5° and 75.3° in Fig. 7(c) were indexed to the formation of hexagonal phase pure crystalline structure of La2O3 (JCPDS: 01-071-5408). Moreover, the diffraction peaks centered at the 2h values of 44.5°, 52.1° and 76.6° were attributed to the presence of metallic Ni, formed by the reduction of NiO in the fresh catalyst. The average crystalline size of the metallic nickel was calculated to be 13 nm, 26 nm and 6 nm for the ceria, zirconia and lanthana supported catalysts respectively. Moreover, it is clear that the nickel lanthanum oxide solid solution was completely reduced into metallic Ni and La2O3, after reduction treatment. The low intensity of the nickel peaks in the reduced catalysts further confirms its fine surface dispersion on the catalyst support. 3.2. Catalytic performance for methane decomposition The kinetic curves of methane decomposition as a function of time on stream at 600 °C, 650 °C and 700 °C were studied and the results are shown in Fig. 8. As shown in Fig. 8, all of the catalysts were found to be highly active and stable for the reaction at 600 °C, 650 °C and 700 °C. No gaseous products other than hydrogen and unreacted methane were detected in the reaction over ceria, zirconia and lanthana supported nickel catalysts. Moreover, it is found that, irrespective of the nature of support material, the hydrogen yield increased significantly with increasing the reaction temperature from 600 °C to 700 °C as shown in Fig. 8. This is attributed to the endothermic nature of the decomposition reaction. As shown in Fig. 8(a), a maximum hydrogen yield of 51% was observed over the Ni/CeO2 catalyst at 600 °C. The hydrogen yield was pointedly increased from 34% to 51% in the initial stage of the reaction, i.e. within the first 60 min of time on stream. After that the yield remained more or less same (50% ± 1) until 150 min of time on stream. A gradual decline in the activity was observed while continuing the reaction further for 360 min of time on stream. At the end of 360 min of reaction, the hydrogen yield was measured to be only 28%. At 650 °C, the hydrogen yield was

Fig. 8. Hydrogen yield as a function of time on stream over (a) Ni/CeO2, (b) Ni/ZrO2 and (c) Ni/La2O3 catalysts at various reaction temperatures of 600 °C, 650 °C and 700 °C.

observed to be increased from 38% to 55%, in the first 60 min of time on stream. After that, the hydrogen yield started to decline and at the end of 360 min, the yield was measured to be 33%. Also it is clear that during the activity abating, the yield was quite similar to the hydrogen yield obtained at 600 °C. However, at 700 °C, a

310

M. Pudukudy et al. / Energy Conversion and Management 126 (2016) 302–315

maximum hydrogen yield of 62% was obtained within the first 60 min of time on stream. After that, a slight decrease in the hydrogen yield was observed in the whole period of decomposition. Approximately 46% hydrogen yield was measured at the end of 360 min. However, it is worth to mention that the Ni/CeO2 catalyst was not fully deactivated for an undiluted methane feed of 150 ml/min in the whole period of reaction, which indicates the high catalytic stability of the catalyst for an extensive period of reaction. The slight decrease in the hydrogen yield with increasing the time on stream in the reaction can be attributed to the accumulation of the nanocarbon on the catalyst surface, which may prevent the active catalytic sites for the reactant molecules for further reaction. A similar trend was also observed in the case of Ni/ZrO2 catalyst as shown in Fig. 8(b). A maximum hydrogen yield of 48% was obtained within the first 20 min of time on stream at 600 °C, followed by a slight decrease of the yield to 43%. Then the hydrogen yield remained more or less same (43% ± 1) until 150 min of time on stream, followed by a slow decrease in the activity. At the end of 360 min of time on stream, the hydrogen yield was measured to be 30%. At 650 °C, a maximum hydrogen yield of 55% was achieved within the first 30 min of time on stream and after that the yield was pointedly decreased to a final hydrogen yield of 36% at the end of 360 min of time on the stream. Furthermore, at a reaction temperature of 700 °C, 61% hydrogen yield was observed within the first 10 min of time on stream. After a slight decrease of hydrogen to 52%, it remained relatively same until the whole period of reaction and at the end of 360 min of time on stream; the yield was noted to be 49%. As described earlier, no catalyst deactivation was observed in this case also. Furthermore the final hydrogen yield of Ni/ZrO2 catalyst (49%) was higher than that of the Ni/CeO2 catalyst (43%). The activity of Ni/La2O3 catalyst for methane decomposition is shown in Fig. 8(c). Compared to Ni/CeO2 and Ni/ZrO2 catalysts, the Ni/La2O3 catalyst showed a more steady state catalytic activity in the whole period of reaction at all reaction temperatures of 600 °C, 650 °C and 700 °C. The only difference noted was in the yield of hydrogen. At 700 °C, a maximum hydrogen yield of 58% was attained in the first 20 min of reaction. After that, the activity was faintly decreased to 54% and it continued more or less same throughout the entire reaction. At 360 min of time on stream, the yield was measured to be 50%. At 650 °C, an initial hydrogen yield of 47% was increased to 49% in the first 30 min of time on stream and after a faint decrease of hydrogen yield to 45%, it remained almost same throughout the whole period of reaction and a final hydrogen yield of 40% was observed at 360 min of time on stream. A similar trend was also observed at the reaction temperature of 600 °C. A maximum hydrogen yield of 43% was obtained in the first 20 min of reaction and at the end of 360 min of reaction; the yield was measured to be 32%. Table 3 shows the catalytic activity and stability of the catalysts for methane decomposition in terms of highest initial and final hydrogen yield at different reaction temperatures. It was found that the initial hydrogen yield follows the catalyst order of Ni/ CeO2 > Ni/ZrO2 > Ni/La2O3, which is same as in the order of their reduction temperatures. Since the catalytic activity strongly

depends on the number of active Ni sites, a lower reduction temperature often contribute to increased number of reduced Ni species and hence a high catalytic activity. Thus the order of initial hydrogen yield (Ni/CeO2 > Ni/ZrO2 > Ni/La2O3) could be related to the order of their reducibility. However, the difference between the highest initial and final hydrogen yield was observed to be minimum for the Ni/La2O3 catalyst at all reaction temperatures compared to the ceria and zirconia supported catalysts indicating its high catalytic stability compared to other catalysts. The stability of the catalysts follows the order of Ni/La2O3 > Ni/ZrO2 P Ni/CeO2. Besides, the ZrO2 supported catalyst was found to be showing a constant decrease of hydrogen yield irrespective of the reaction temperature. For Ni/CeO2 and Ni/ZrO2 catalyst, quite high catalytic stability was observed at 700 °C. At low reaction temperature of 600 °C, the decline of hydrogen yield was quite high compared to the same at 700 °C. From the discussion, it can be clear that the ceria, zirconia and lanthana supported nickel catalysts were highly active for methane decomposition. However, a support dependent catalytic stability was observed for the catalysts while considering the hydrogen yield. The high activity of the catalysts could be due to their porosity as shown in their surface morphology. It is reported that the nanosized porous materials act as nano-reactors for directing the chemical reactions in limited space, due to their enhanced surface and structural properties with the increased mass transport capacity [52]. Previously it is reported that the ceria based catalysts were expected to produce COx in methane decomposition, due to the presence of lattice oxygens in the ceria crystal structure [22]. However, no formation of COx was observed in the present case after a first few minutes of reaction, which further points to the fact that the complete reduction of the surface and lattice oxygen of ceria was occurred in the applied reduction conditions. It is also possible that the residual lattice oxygen in ceria (if any present after reduction) could be reduced in the first few minutes of decomposition. That’s why no COx was detected from/after the first five minutes of time on stream. Furthermore, an important fact that was noted is the non-correlation of the surface area of the catalysts with catalytic efficiency. It is found that the specific surface area of the catalysts has an insignificant effect on the catalytic stability for methane decomposition over the present catalysts. A high catalytic stability was observed for Ni/La2O3 catalyst having a low surface area of 10.6 m2/g. Thus it can be said that for a reaction like methane decomposition, the surface area is expected to play a significant role only in the initial stage of the reaction, since after an initial high conversion or due to the surface coverage with carbon; the activity could be decreased or the catalyst might be fully deactivated [53,54]. However, in the present study, for over all of the catalysts no such deactivation was noticed for a period of 360 min. However, it should be also pointed that the initial activity of the ceria and zirconia supported catalysts was quite high compared to the lanthana catalyst, which could be due to its high surface area and lower reduction temperatures. However, a quite fast decline in the activity was observed for these catalysts compared to the lanthana supported catalysts. The high catalytic stability of the Ni/La2O3 catalyst can be attributed to its special porous morphology compared to the other catalysts as discussed

Table 3 Tabulated activity and stability of the catalysts in terms of hydrogen yield. Catalysts

H2 yield at 600 °C (%)

H2 yield at 650 °C (%)

H2 yield at 700 °C (%)

Highest initial

Final

Difference

Highest initial

Final

Difference

Highest initial

Final

Difference

Ni/CeO2 Ni/ZrO2 Ni/La2O3

51 48 43

28 30 32

23 18 11

55 55 49

33 36 40

22 19 9

62 61 58

46 49 50

16 19 8

M. Pudukudy et al. / Energy Conversion and Management 126 (2016) 302–315

311

before, which could be enhance the diffusion of methane gas. It is reported that the nanosized metal crystallites generated by the reduction of spinel perovskite materials could enhance the reaction [33,34]. Thus the formation of nickel lanthanum aluminium oxide solid solution and its reduction to Ni and lanthana could be the reason for its high catalytic stability since it is expected to generate nanosized Ni particles and which could be finely dispersed on the surface of the lanthana [51]. 3.3. Characterization of carbon deposited catalysts Carbon yield is defined as the ratio of the total amount of carbon deposited over the catalyst to the amount of metal portion in the catalyst used [55,56]. The yield of nanocarbon deposited over ceria, zirconia and lanthana supported nickel catalysts after 360 min of methane decomposition at 600 °C, 650 °C and 700 °C is plotted in Fig. 9. As shown in Fig. 9, it is clear that the reaction temperature has a significant role in the deposition of nanocarbon. With increasing the reaction temperature, the yield of nanocarbon increased significantly over the catalysts. For all of the catalysts, highest carbon yield was obtained at 700 °C. A maximum carbon yield of 1360 wt% was obtained for the Ni/CeO2 catalyst at 700 °C. At 600 °C and 650 °C, it was found to be only 986 wt% and 1121 wt% respectively. In the case of Ni/ZrO2 catalyst, the carbon yield was increased from 897 wt% to 1159 wt% when the reaction temperature was increased from 600 °C to 700 °C. However, the highest carbon yield was obtained for the lanthana supported catalyst, which may be due to its high catalytic stability. The carbon yield was measured to be 1089 wt%, 1349 wt% and 1576 wt% for the Ni/La2O3 catalyst at 600 °C, 650 °C and 700 °C respectively. The nanocarbon deposited at 700 °C was selected to further characterize them for their crystalline and structural properties. The carbon deposited at 700 °C was further characterized using XRD, to study their crystalline nature and also to evaluate the active phase of the metallic species present in them for further reaction. The XRD patterns of the spent catalysts are shown in Fig. 10. The diffraction peaks centered at the 2h value of 26.6° with the (0 0 2) plane and 42.6° were attributed to the formation of crystalline carbon in all set of catalysts. The other diffraction peaks located at the 2h values of 44.5°, 51.8° and 76.8° were attributed to the presence of face-centered cubic phase crystalline structure of metallic Ni (JCPDS: 01-071-4655) formed after the reduction or

Fig. 9. Carbon yield at various reaction temperatures over Ni/CeO2, Ni/ZrO2 and Ni/La2O3 catalysts.

Fig. 10. XRD patterns of the spent catalysts (a) Ni/CeO2, (b) Ni/ZrO2 and (c) Ni/La2O3 at 700 °C.

decomposition reaction. These metal species are responsible for the continuation of the decomposition reaction without any deactivation. The other diffraction peaks in Fig. 10(a) were directly indexed to the face centered cubic phase crystalline structure of CeO2. As shown in Fig. 10(b), the sharp diffraction peaks located at the 2h values of 28.1°, 30.2°, 31.8°, 34.4°, 35.1°, 40.7°, 47.5°, 50.3°, 54.4°, 60.1°, 62.8° and 65.6° were indexed to the crystalline phase of ZrO2. Both monoclinic and tetragonal phases of zirconia were detected as in the reduced catalysts (Fig. 7). The other marked diffraction peaks Fig. 10(c) were indexed to the formation of the crystalline La2O3. Among the catalysts, the highest carbon peak intensity was shown by the Ni/ZrO2 catalyst, where the intensity of the peaks of metallic nickel was significantly higher than that of the zirconia peaks and lower than that of carbon peak. This is a clear evidence for the efficiency of the Ni/ZrO2 catalyst for the deposition of nanocarbon with high crystallinity. In Ni/CeO2 catalyst, the support CeO2 showed more intense diffraction peaks than carbon and nickel. The peaks of nickel and carbon were found to be very weak, indicating their fine surface dispersion on the crystalline CeO2 support. However, the carbon peak intensity was significantly higher than that of the peak of nickel in the Ni/CeO2 catalyst. In Ni/ La2O3 spent catalyst, phase segregated Ni and La2O3 was detected instead of lanthanum nickel oxide solid solution. These are supposed to be formed by the reduction of lanthanum nickel oxide after the reduction or methane decomposition (Fig. 7). The diffraction peaks of metallic nickel were found to be very weak and less intense than that of the peaks of carbon and lanthana, indicating its fine dispersion on the lanthana matrix. However, the peak intensities of carbon and lanthana support appeared to be almost same. This is likewise an indication of the surface dispersion of the nickel and also carbon on the lanthana. The average crystalline size of Ni in the spent catalysts was calculated to be 15 nm, 29 nm and 11 nm Ni/CeO2, Ni/ZrO2 and Ni/La2O3 catalysts respectively. The calculated interlayer d-spacing of the nanocarbon deposited over ceria, zirconia and lanthana supported catalysts, at the 2h value of 26.6° was found to be 0.3421 nm, 0.3372 nm and 0.3448 nm respectively. These are very close to the reported value of the ideal distance between graphitic layers (0.3354 nm), further indicating the high crystallinity and graphitization degree of the deposited nanocarbon [57]. The morphological appearance of the nanocarbon deposited over the catalysts was studied using FESEM analysis and the

312

M. Pudukudy et al. / Energy Conversion and Management 126 (2016) 302–315

Fig. 11. FESEM images of the carbon deposited over (a, b) Ni/CeO2, (c, d) Ni/ZrO2 and (e, f) Ni/La2O3.

images are shown in Fig. 11. Irrespective of the catalyst supports used, carbon nanofilaments were produced over all of the nickel based catalysts. The catalyst surfaces were found to be fully covered with bulk amounts of carbon nanofilaments as shown in supplementary data, Fig. SD4. Fig. 11(a, b) shows the images of the nanocarbon deposited over Ni/CeO2 catalyst. The carbon nanotubes over Ni/CeO2 catalysts were found to be thin, uniform and very long. The length of the carbon nanotubes goes to several micrometers. Furthermore, it is found that the carbon nanotubes are highly interwoven in all cases. Therefore, the determination of the actual length of the nanotubes was quite tedious. However, the diameters of the carbon nanotubes were measured to be 20–30 nm. Over Ni/ZrO2 catalyst the same trend was observed, where the surface of the catalyst was completely confined with huge amounts of carbon nanotubes. As shown in Fig. 11(c, d), the diameters of the carbon nanotubes over Ni/ZrO2 catalyst were observed to be quite higher than that of the nanocarbon deposited over Ni/CeO2 catalyst. The diameters were measured to be ranging from 30–40 nm. Open tips of the carbon nanotubes were clearly seen in Fig. 11(d), indicating a base growth mechanism. Moreover, the bulk deposition of carbon nanofilaments with the diameters ranging from 15 to 25 nm was observed over the Ni/La2O3 catalysts (Fig. 11(e, f)). These are also observed to be very thin, homogeneous and long as like the carbon nanotubes over the Ni/CeO2 catalyst. Moreover, the length of the carbon nanotubes over Ni/La2O3 was found to be goes to several micrometers. The diameter distribution histogram of the carbon nanotubes is shown

Fig. 12. The diameter distribution histogram of the multi-walled carbon nanotubes (the diameters were measured using a smartTiff software (Version V02.01) and 50 measurements were considered to plot the diameter distribution histogram).

in Fig. 12. It is clear that the diameters of the carbon nanotubes were varied with respect to the catalyst support and in all cases the diameters were found to be less than 50 nm. The average

M. Pudukudy et al. / Energy Conversion and Management 126 (2016) 302–315

313

Fig. 13. TEM images of the nanocarbon deposited over (a, b) Ni/CeO2, (c, d) Ni/ZrO2 and (e, f) Ni/La2O3.

diameter of the carbon nanotubes follows the catalyst order of Ni/ZrO2 > Ni/CeO2 P Ni/La2O3, which is same as in the order of the average crystalline size of Ni in the reduced/spent catalysts. The internal structure of the carbon nanotubes was studied using TEM analysis and images are shown in Fig. 13. Randomly interwoven carbon nanotubes with different diameters were observed in the images. The dark spots represent the presence of metallic nickel in the carbon nanotubes. The width of carbon nanotubes deposited over Ni/ZrO2 was found to be quite higher than that of the carbon nanotubes deposited over Ni/CeO2 catalysts (Fig. 13(a, c)). Moreover, very thin carbon nanotubes were observed over the Ni/La2O3 catalyst as shown in Fig. 13(e). These results are highly consistent with the FESEM observations. The high magnified images shown in Fig. 13(b, d and f) confirms the formation of multi-walled carbon nanotubes, since the average wall thicknesses were measured to be lies in the range of 7– 9 nm, 12–15 nm and 5–8 nm for the Ni/CeO2, Ni/ZrO2 and Ni/ La2O3 catalysts respectively. A clear internal channel space between the multi-layer walls was observed in the carbon nanotube deposited over Ni/CeO2 catalyst with a channel width of 8.5 nm and a total diameter of 22.8 nm. The average diameter of the carbon nanotube shown over Ni/La2O3 catalyst was measured to be 18 nm with the wall thickness and internal channel space of 6 nm respectively. Over Ni/ZrO2 catalyst, the multiwalled carbon nanotubes with the diameter of 34 nm, wall thickness of 12.5 nm and an internal channel space of 9 nm was observed as shown in Fig. 13(d). Furthermore, a visible cup stack structure was observed

Fig. 14. Raman spectra of the nanocarbon deposited over (a) Ni/CeO2, (b) Ni/ZrO2 and (c) Ni/La2O3.

in the internal channel of the carbon nanotubes produced over Ni/ZrO2 catalyst compared to Ni/CeO2 and Ni/La2O3 catalysts. The graphitization degree and crystalline quality of the multiwalled carbon nanotubes deposited over ceria, zirconia and

314

M. Pudukudy et al. / Energy Conversion and Management 126 (2016) 302–315

lanthana supported catalysts at 700 °C were studied using Raman spectroscopy and the spectra are shown in Fig. 14. For all of the Raman spectra, two major bands were observed, which represents the characteristic D and G bands of the carbon nanomaterials. The d-band was observed at around 1341 cm 1 and it could be assigned to the presence of amorphous carbon or due to the presence of structural imperfections on the graphitic layers [58]. But in this case, the D band is mainly assigned to the structural disorders of the graphitic carbon, since the non-existence of amorphous carbon was confirmed by the XRD analysis. The G band was centered at around 1569 cm 1 and which is attributed to the in plane carbon-carbon stretching vibration of the sp2 atoms in the graphite layers [59]. These two bands clearly indicate the formation of highly graphitized carbon nanotubes. Moreover, the D and G band intensity ratio (ID/IG ratio) was used to estimate the graphitization degree and crystalline quality of the carbon samples. The calculated ID/IG value was found to be 0.95, 0.93 and 0.95 for the ceria, zirconia and lanthana supported nickel catalysts respectively. The values were found to be very small and less than unity, indicating the formation of multiwalled carbon nanotubes with high crystalline quality and better graphitization degree, irrespective of the catalyst supports. However, a much higher crystallinity and graphitization degree was observed for the multiwalled carbon nanotubes deposited over Ni/ZrO2 catalyst since it showed highest carbon peak intensity in the XRD patterns and lowest ID/IG value in the Raman spectra compared to other catalysts.

4. Conclusions In summary, ceria, zirconia and lanthana supported nickel catalysts were successfully prepared via a facile solid state citrate fusion method and effectively used for the thermocatalytic decomposition of methane to produce COx free hydrogen and multiwalled carbon nanotubes at 600 °C, 650 °C and 700 °C. Various attractive porous morphologies of the fresh catalysts with respect to the support material, aroused by the release of huge amount of gases from the bulk of the fused sample during the thermal treatment were confirmed by the scanning electron microscopic studies. The phase segregated NiOs with low crystallinity were found to be finely dispersed on the surface of ceria and zirconia supports, whereas the formation of faintly crystallized nickel lanthanum oxide solid solution was observed over the Ni/La2O3 catalyst. The relative low temperature reduction peaks in the TPR profiles of the catalysts corresponded to the reduction of NiO which had a weak interaction with the support and the support depended high temperature reduction peaks were also verified. Moreover the complete reduction of NiLa2O4 into nickel and lanthanum oxide was achieved at/before 620 °C. The nitrogen physisorption analysis confirmed the mesoporous texture of the samples with slit like pores created due to the inter-aggregation of nano-particles. All of the catalysts were found to show high catalytic activity and stability for methane decomposition irrespective of their specific surface area. The yield of hydrogen and nanocarbon increased significantly with increasing the reaction temperature from 600 °C to 700 °C. A maximum hydrogen yield of 62%, 61% and 58% and a carbon yield of 1360 wt%, 1159 wt% and 1576 wt% was achieved over ceria, zirconia and lanthana supported catalysts respectively, at 700 °C. It is found that the Ni/ La2O3 catalyst was more stable than the ceria and zirconia supported catalysts and it could be responsible for the highest carbon yield over Ni/La2O3 catalyst, even it showed low surface area. The high stability of the lanthana supported catalyst could be attributed to the formation of nanosized active Ni and its fine dispersion on lanthana matrix after the reduction of spinel solid solution. Highly uniform multiwalled carbon nanotubes with different

diameters depending on the support were accumulated over the catalysts. Irrespective of the support material, the multiwalled carbon nanotubes showed high graphitization degree with an ID/IG value of  0.95. Acknowledgements This project is financed by Yayasan Sime Darby (YSD) - Universiti Kebangsaan Malaysia and Sime Darby Research, under grants PKT 6/2012 and KK-2014-014. The authors wish to thank the university administration for the financial support and FST and CRIM, UKM for the characterization of the materials. The Institut Teknologi Maju (ITMA) at the Universiti Putra Malaysia is acknowledged for providing the Raman analyses. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.enconman.2016. 08.006. References [1] Pudukudy M, Yaakob Z, Mohammad M, Narayanan B, Sopian K. Renewable hydrogen economy in Asia-opportunities and challenges: an overview. Renew Sustain Energy Rev 2014;30:743–57. [2] Tabrizi FF, Mousavi SAHS, Atashi H. Thermodynamic analysis of steam reforming of methane with statistical approaches. Energy Convers Manage 2015;103:1065–77. [3] Keipi T, Hankalin V, Nummelin J, Raiko R. Techno-economic analysis of four concepts for thermal decomposition of methane: reduction of CO2 emissions in natural gas combustion. Energy Convers Manage 2016;110:1–12. [4] El Naggar AMA, Awadallah AE, Aboul-Enein AA. Novel intensified nanostructured zero-valent nickel alloy based catalyst for hydrogen production via methane catalytic decomposition. Renew Sustain Energy Rev 2016;53:754–65. [5] Gutiérrez F, Méndez F. Entropy generation minimization for the thermal decomposition of methane gas in hydrogen using genetic algorithms. Energy Convers Manage 2012;55:1–13. [6] Amin AM, Croiset E, Epling W. Review of methane catalytic cracking for hydrogen production. Int J Hydrogen Energy 2011;36:2904–35. [7] Uddin MN, Daud WMAW, Abbas HF. Kinetics and deactivation mechanisms of the thermal decomposition of methane in hydrogen and carbon nanofiber Coproduction over Ni-supported Y zeolite-based catalysts. Energy Convers Manage 2014;87:796–809. [8] Ashik UPM, Daud WMAW, Abbas HF. Production of greenhouse gas free hydrogen by thermo catalytic decomposition of methane- a review. Renew Sustain Energy Rev 2015;44:221–56. [9] Abbas HF, Daud WMAW. Hydrogen production by methane decomposition: a review. Int J Hydrogen Energy 2010;35:1160–90. [10] Li Y, Li D, Wang G. Methane decomposition to COx-free hydrogen and nanocarbon material on group 8–10 base metal catalysts: a review. Catal Today 2011;162:1–48. [11] Shen Y, Lua AC. Sol-gel synthesis of titanium oxide supported nickel catalysts for hydrogen and carbon production by methane decomposition. J Power Sources 2015;280:467–75. [12] Ahmed S, Aitani A, Rahman F, Al-Dawood A, Al-Muhaish F. Decomposition of hydrocarbons to hydrogen and carbon. Appl Catal A 2009;359:1–24. [13] Awadallah AE, Aboul-Enein AA, Yonis MM, Aboul-Gheit AK. Effect of structural promoters on the catalytic performance of cobalt based catalysts during natural gas decomposition to hydrogen and carbon nanotubes. Fullerenes, Nanotubes, Carbon Nanostruct 2016;24:181–9. [14] Salmones J, Wang JA, Valenzuela MA, Sánchez E, Garcia A. Pore geometry influence on the deactivation behavior of Ni-based catalysts for simultaneous production of hydrogen and nanocarbon. Catal Today 2009;148:134–9. [15] Uddin MN, Daud WMAW, Abbas HF. Co-production of hydrogen and carbon nanofibers from methane decomposition over zeolite Y supported Ni catalysts. Energy Convers Manage 2015;90:218–29. [16] Nuernberg GDB, Foletto EL, Campos CEM, Fajardo HV, Carreno NLV, Probst LFD. Direct decomposition of methane over Ni catalyst supported in magnesium aluminate. J Power Sources 2012;208:409–14. [17] Zhang J, Jin L, Li Y, Hua H. Ni doped carbons for hydrogen production by catalytic methane decomposition. Int. J. Hydrogen Energy 2013;38:3937–47. [18] Suelves I, Lazaro MJ, Moliner R, Echegoyen Y, Palacios JM. Characterization of NiAl and NiCuAl catalysts prepared by different methods for hydrogen production by thermo catalytic decomposition of methane. Catal Today 2006;116:271–80. [19] Ashok J, Raju G, Reddy PS, Subrahmanyam M, Venugopal A. Catalytic decomposition of CH4 over NiO–Al2O3–SiO2 catalysts: influence of catalyst

M. Pudukudy et al. / Energy Conversion and Management 126 (2016) 302–315

[20] [21]

[22] [23] [24]

[25]

[26]

[27]

[28]

[29]

[30]

[31] [32]

[33]

[34] [35]

[36]

[37]

[38]

[39]

preparation conditions on the production of H2. Int J Hydrogen Energy 2008;33:4809–18. Zhou L, Enakonda LR, Saih Y, Loptain S, Gary D, Gallo PD, et al. Catalytic methane decomposition over Fe-Al2O3. Chem Sus Chem 2016;9:1243–8. Ying Y, Meisheng C, Minglai L, Na Z, Zhiqi L, Yongxi S. Rare earth modified Ni-Si catalysts for hydrogen production from methane decomposition. J Rare Earths 2014;32:709–14. Li Y, Zhang BC, Tang XL, Xu YD, Shen WJ. Hydrogen production from methane decomposition over Ni/CeO2 catalysts. Catal Commun 2006;7:380–6. Tang LG, Yamaguchi D, Burke N, Trimm D, Chiang K. Methane decomposition over ceria modified iron catalysts. Catal Commun 2010;11:1215–9. Odier E, Schuurman Y, Mirodatos C. Non-stationary catalytic cracking of methane over ceria-based catalysts: mechanistic approach and catalyst optimization. Catal Today 2007;127:230–7. Kogler M, Köck EM, Perfler L, Bielz T, Pollach MS, Hetaba W, et al. Methane decomposition and carbon growth on Y2O3, yttria- stabilized zirconia, and ZrO2. Chem Mater 2014;26:1690–701. Wolfbeisser A, Klötzer B, Mayr L, Rameshan R, Zemlyanov D, Bernardi J, et al. Surface modification processes during methane decomposition on Cupromoted Ni–ZrO2 catalysts. Catal Sci Technol 2015;5:967–78. Wolfbeisser A, Kovács G, Kozlov SM, Föttinger K, Bernardi J, Klötzer B et al. Surface composition changes of CuNi-ZrO2 during methane decomposition: an operando NAP-XPS and density functional study. Catal Today, http://dx.doi. org/10.1016/j.cattod.2016.04.022. Kurasawa S, Iwamoto S, Inoue M. Synthesis of carbon nanotubes by the catalytic decomposition of methane on Ni based catalysts. Mol Cryst Liq Cryst 2002:381 [347]/123-[352]128. Liu BC, Tang SH, Yu ZL, Zhang BL, Chen T, Zhang SY. Catalytic growth of singlewalled carbon nanotubes with a narrow distribution of diameters over Fe nanoparticles prepared in situ by the reduction of LaFeO3. Chem Phys Lett 2002;357:297–300. Chen L, Liu HT, Yang K, Wang JK, Wang XL. Catalytic synthesis of carbon nanotubes from the decomposition of methane over a Ni–Co/La2O3 catalyst. Can J Chem 2009;87:47–53. Kuras M, Zimmermann Y, Petit C. Reactivity of perovskite-type precursor in MWCNTs synthesis. Catal Today 2008;138:55–61. Gallego GS, Barrault J, Dupeyrat CB, Mondragon F. Production of hydrogen and MWCNTs by methane decomposition over catalysts originated from LaNiO3 perovskite. Catal Today 2010;149:365–71. Rivas ME, Fierro JLG, Guil-Lopez R, Pena MA, La Parola V, Goldwasser MR. Preparation and characterization of nickel-based mixed-oxides and their performance for catalytic methane decomposition. Catal Today 2008;133– 135:367–73. Cunha AF, Mahata N, Orfao JJM, Figueiredo JL. Methane decomposition on La2O3-promoted Raney-type Fe catalysts. Energy Fuels 2009;23:4047–50. Maneerung T, Hidajat K, Kawi S. Co-production of hydrogen and carbon nanofibers from catalytic decomposition of methane over LaNi(1 x)MxO3 a perovskite (where M = Co, Fe and X = 0, 0.2, 0.5, 0.8, 1). Int J Hydrogen Energy 2015;40:13399–411. Pudukudy M, Yaakob Z, Akmal ZS. Direct decomposition of methane over SBA15 supported Ni, Co and Fe based bimetallic catalysts. Appl Surf Sci 2015;330:418–30. Pudukudy M, Yaakob Z. Methane decomposition over Ni, Co and Fe based monometallic catalysts supported on sol gel derived SiO2 microflakes. Chem Eng J 2015;262:1009–21. Pudukudy M, Yaakob Z, Takriff MS. Methane decomposition over Pd promoted Ni/MgAl2O4 catalysts for the production of COx free hydrogen and multiwalled carbon nanotubes. Appl Surf Sci 2015;356:1320–6. Pudukudy M, Yaakob Z, Akmal ZS. Direct decomposition of methane over Pd promoted Ni/SBA- 15 catalysts. Appl Surf Sci 2015;353:127–36.

315

[40] Pudukudy M, Yaakob Z, Narayanan B. Selective vapour phase oxidation of benzyl alcohol to benzaldehyde over mesoporous ceria-zirconia solid solution synthesized via a facile citrate route. J Clust Sci 2014;25:1599–614. [41] Bayat N, Rezaei M, Mesh F. Methane decomposition over Ni-Fe/Al2O3 catalysts for production of COx-free hydrogen and carbon nanofiber. Int J Hydrogen Energy 2016;41:1574–84. [42] Mantri K, Selvakannan PR, Tardio J, Bhargava SK. Synthesis of very high surface area Au-SBA-15 materials by confinement of gold nanoparticles formation within silica pore walls. Colloids Surf A 2013;429:149–58. [43] Zhang C, Li S, Li M, Wang S, Ma X, Gong J. Enhanced oxygen mobility and reactivity for ethanol steam reforming. AIChE 2012;58:516–25. [44] Zou W, Ge C, Lu M, Wu S, Wang Y, Sun J, et al. Engineering the NiO/CeO2 interface to enhance the catalytic performance for CO oxidation. RSC Adv 2015;5:98335–43. [45] Trovarelli A. Catalytic properties of ceria and CeO2-containing materials. Catal Rev 1996;38:439–520. [46] Tada S, Shimizu T, Kameyama H, Haneda T, Kikuchi R. Ni/CeO2 catalysts with high CO2 methanation activity and high CH4 selectivity at low temperatures. Int J Hydrogen Energy 2012;37:5527–31. [47] Fajardie F, Tempere O, Manoli JM, Mariadassou GD, Blanchard G. Ceria lattice oxygen ion substitution by Cl during the reduction of Rh(Cl)/CeO2 catalysts. Formation and stability of CeOCl. J Chem Soc Faraday Trans 1998;94:3727–35. [48] Wang SB, Lu GQM. Role of CeO2 in Ni/CeO2–Al2O3 catalysts for carbon dioxide reforming of methane. Appl Catal B 1998;19:267–77. [49] Valentini A, Carreño NLV, Probst LFD, Barison A, Ferreira AG, Leite ER, et al. Ni: CeO2 nanocomposite catalysts prepared by polymeric precursor method. Appl Catal A 2006;310:174–82. [50] Nabgan W, Abdullah TAT, Mat R, Nabgan B, Jalil AA, Firmansyah L et al. Production of hydrogen via steam reforming of acetic acid over Ni and Co supported on La2O3 catalyst. Int J Hydrogen Energy, http://dx.doi.org/10.1016/ j.ijhydene.2016.04.176. [51] Benito M, Garcıa S, Aparicio PF, Serrano LG, Daza L. Development of biogas reforming Ni-La-Al catalysts for fuel cells. J Power Sources 2007;169:177–83. [52] Beretta M, Sozzani P. Nanostructured mesoporous materials obtained by template synthesis and controlled shape replica, PhD Thesis; 2009. [53] Awadallah AE, Enein AAA, Aboul-Gheit AK. Effect of progressive Co loading on commercial Co–Mo/Al2O3 catalyst for natural gas decomposition to COx-free hydrogen production and carbon nanotubes. Energy Convers Manage 2014;77:143–51. [54] Awadallah AE, Ahmed W, Noor El-Din MR, Enein AAA. Novel aluminosilicate hollow sphere as a catalyst support for methane decomposition to COx-free hydrogen production. Appl Surf Sci 2013;287:415–22. [55] Chai SP, Seah C, Mohamed AR. A parametric study of methane decomposition into carbon nanotubes over Co-Mo/Al2O3 catalyst. J Nat Gas Chem 2011;20:84–9. [56] Fakeeha AH, Khan WU, Al-Fatesh AS, Abasaeed AE, Naeem MA. Production of hydrogen and carbon nanofibers from methane over Ni-Co-Al catalysts. Int J Hydrogen Energy 2015;40:1774–81. [57] Awadallah AE, Mostafa MS, Enein AAA, Hanafi SA. Hydrogen production via methane decomposition over Al2O3-TiO2 binary oxides supported Ni catalysts: effect of Ti content on the catalytic efficiency. Fuel 2014;129:68–77. [58] Pudukudy M, Yaakob Z, Takriff MS. Methane decomposition over unsupported mesoporous nickel ferrites: effect of reaction temperature on the catalytic activity and properties of the produced nanocarbon. RSC Adv 2016;6:68081–91. [59] Nuernberg GB, Foletto EL, Probst LFD, Carreno NLV, Moreira MA. MgAl2O4 spinel particles prepared by metal-chitosan complexation route and used as catalyst support for direct decomposition of methane. J Mol Catal A: Chem 2013;370:22–7.