Perovskite modified catalysts with improved coke resistance for steam reforming of glycerol to renewable hydrogen fuel

Catalytic steam reforming of renewable feedstock to renewable energy or chemicals always goes with intense coking activities that produce carbonaceous products leading to low performance and eventual catalyst deactivation. Supported metal catalyst such as Ni/Al2O3 is known to catalysed gasification and decomposition of biomass feedstock largely for renewable fuel production with promising results. Catalyst deactivation from high carbon deposition, agglomeration and phase transformations resulting to rapid deactivation are some of the issues identified with the use of the catalyst. In this work, improvement on the coke resistance and catalytic properties of Ni/Al2O3 catalyst is sought via the use of a thermally stable and coke‐resistant perovskite La0.75Sr0.25Cr0.5Mn0.5O3‐δ (LSCM) as catalyst promoter/modifier and involving Zirconia‐doped Ceria (Ce‐Zr) as alternative support in steam reforming of pure and by‐product glycerol. The stabilizing influence of the LSCM on the Ni catalyst has improved stability against agents of deactivation with a consequent significant improvement of catalytic activity of Ni/Al2O3 in H2 production and robust suppression of carbon deposition. Particularly, the synergy between the LSCM promoter and alternative Ce0.75Zr0.25O2 support enhanced the basic and redox properties known for Ce0.75Zr0.25O2 support in contrast to the week acid centres in γ‐Al2O3 support which further improved nickel stability, catalyst–support interaction with a resultant high catalytic activity and robust coke suppression as a result of enhanced oxygen mobility. There is correlation between the product distribution, nature of coke deposited and reforming temperature as well as type of support and structural modification. Hence, integration of a robust perovskite material as a catalyst promoter and choice of support could be tailored in design and development of robust catalyst systems to improve the performance of supported metal catalysts, particularly the suppression of carbon deposition for hydrocarbon and biomass conversion to renewable fuel or chemicals.


| INTRODUCTION
There is much growing interest to secure energy sustainability and clean environment through the emergent technologies in biomass conversion to renewable fuels. Gasification of biomass feedstock at elevated temperatures and then subjecting it to reforming process is capable of producing hydrogen and other related gases such as CO 2 CO, CH 4 and H 2 O reformates that offer many industrial applications in the production of value-added chemicals and energy such as fuel cell technology (Aasberg-Petersen et al., 2011;Moradian et al., 2021;Pandey, 2019). Solid oxide fuel cells (SOFCs) operate at high temperature and are very flexible to choice fuel, hence capable of utilizing these gases as alternative fuel to its conventional fossilbased hydrogen when compared other fuel cell technologies that operate at low temperatures (Hibino et al., 2021;Pongratz et al., 2022). Hence, the readily available and cheap biomass by-product glycerol from biodiesel synthesis is a great energy resource that could provide alternative renewable and sustainable energy. Consequently, reforming of by-product glycerol for the production of hydrogenrich gas or syngas for energy technology and production of renewable chemicals could provide alternative use to its underutilized surplus being wasted from the ever expanding biodiesel production activities (Ismaila et al., 2021). This would make biodiesel production very cheap and enhance its economic viability. More so, it would check all the issues that border on environmental problems of waste by-product glycerol disposal and improve the availability and sustainability of clean energy and safer environment. However, catalytic valorization of biomass feedstock to renewable energy or chemicals always goes with intense coking activities that produce carbonaceous products which require robust catalyst to enhance stability and reusability of catalyst as well as products selectivity and dissemination.
The state-of-the-art catalyst systems mostly used for biomass gasification and reforming processes for the production of hydrogen or syngas are largely nickel-based particularly those supported on Al 2 O 3 . Nickel metal catalysts have great affinity for bond breakage especially C-C, C-H and C-O bonds leading to intense coking activities that could lead to catalyst poisoning and consequent deactivation (Kim et al., 2021;Wu et al., 2013). The catalytic properties of the nickel catalysts and general catalytic behaviour were improved not only through the use of alternative supports but also the use of promoters for structural modifications. Hence, the choice of support and adding the right or suitable amount of a promoter play an important role in shaping and modifying the structural properties and catalytic behaviour. Sufficiently large surface area and thermally stable supports with suitable active sites for instance help a lot in metal catalysts dispersion and distribution. They also offer enhanced metal catalyst-support interaction hence suppress particle size growth and sintering as well as coking activities (Huo et al., 2021). They use rare earth metals and fist raw transition metals such as La, Ce, Zr, Fe, Mn, Co etc. as prompters were demonstrated in different studies to stabilize Ni-based catalyst systems and to modify structural and surface properties and coke suppression with some promising results (Profeti et al., 2009;Therdthianwong et al., 2008). Ceria was also used as promoter in many studies to improve the rate of biomass decomposition and gasification processes which led to suppression of deposition of carbonaceous residues (Gao et al., 2020), oxygen storage-release capacity (Boaro et al., 2019), enhanced Ni-support interaction, dispersion and distribution of Ni particles Toemen et al., 2016) as well as catalyst stability and reusability Toemen et al., 2016;Nie et al., 2017). There are reports on the use of Perovskite materials in reforming of hydrocarbons and are found to have demonstrated very effective performance as results of their good catalytic behaviour, suitable morphology, enhanced oxygen mobility and highly susceptible to structural modification through doping (Mawdsley & Krause, 2008;Kwon et al., 2020;Kousi et al., 2021). Particularly, the chromia-and titaniabased perovskites in combination with some basic metals are found to be robust, stable in fuel streams and catalytically very active with effective resistance to coking activities (Tao & Irvine, 2004;Umar et al., 2021).
Moreover, Ceria-based catalysts have been the subject of many researches because of their suitable structural properties and ability to bind metal catalyst such as nickel particles, enhancing catalyst-support interaction and improving the metal catalyst resistance to sintering and possible deactivation (Aberkane et al., 2019;Bezen et al., 2011;Polychronopoulou et al., 2011). Compared to other support materials like Al 2 O 3 , SiO 2 etc., ceria-zirconia mixed oxides (Ce-Zr) have become more attractive over time as results of their redox potential, improved oxygen storage-release capacity and strong basic-acidic surface and general catalytic behaviour (Kun-udom et al., 2022;Sun et al., 2019). Ceria-zirconia mixed materials owe their great support potentials to strong basic-acidic surface and redox potentials particularly which make them very useful in many reactions and processes such as water-gas shift reaction (WGSR) for hydrogen production (Gorte & Zhao, 2005;Larina et al., 2017). Hence, gradually, ceria-zirconia mixed oxides are replacing pure ceria in many applications (Nunez-Rico et al., 2020;Vinodkumar et al., 2015). Thus, ceria-zirconia mixed oxides were utilized in reforming hydrocarbon and biomass feedstock with promising results. Ceria-doped zirconia was used to suppress the crystal growth and agglomeration of nickel particles due to sintering during reforming process and also to significantly enhance the oxygen mobility. It has also reduced the rate at which fouling of the catalyst surface occurs as a result of coking activities with improved catalytic activity, hydrogen selectivity and reducibility of the catalyst systems (Bampenrat et al., 2010;Łamacz et al., 2011;Zhang et al., 2020). In a new technology, ceriazirconia's mixed oxides sterling properties allow it to be utilized in the three-way catalysts (TWC) production in order to mitigate against emission of harmful gases from car exhaust (Rood et al., 2019;Trovarelli et al., 2001).
Despite the complexity of the composition of bio-oil and its low-energy density, recent reviews show that H 2 , CO, CO 2 and CH 4 could be produced from the reforming of biomass bio-oil using catalyst systems largely Ni, Fe, Co, Zeolite and precious metals usually enhanced or modified with a suitable support such as Al 2 O 3 , SilO 2 , ZrO 2 , TiO 2 to provide good surface area and thermal stability and some promoters such as ZrO 2 , CeO 2 MgO, etc. to enhance catalysts performance and suppression of coking activities Pafili et al., 2021;Santamaria et al., 2021;Zhang, 2022). More interesting results were also obtained from catalytic reforming of bio-oil individual model components such as ethanol, acetic acid and toluene or mixtures of such model components to produce hydrogen and other gaseous fuels with limited carbon deposition and deactivation (Elharati et al., 2022;Santimaria et al., 2020). Catalytic reforming has shown glycerol to be an intense energy resource that could be reformed catalytically to produce hydrogen and other gaseous fuels. Nickel-based supported catalysts largely modified with transition, alkaline or precious metals as modifiers of promoters have shown promising results (Fasolini et al., 2019;Ismaila et al., 2021). Aside the choice catalyst systems, optimization of the pyrolysis and reforming temperature and process technological innovations particularly using inline reforming process has helped to improve catalyst performance, yield and suppression of coke deposition which is identified as the main cause of catalyst deactivation (Arregi et al., 2018;Fernandez et al., 2022;Pafili et al., 2021, Santamaria et al., 2021. Despite all the promising results in the use of these materials particularly ceria-doped zirconia-based catalysts in reforming processes, catalyst deactivation due to coking as results of intermediate and side reactions especially in reforming processes performed at low temperatures (except where they are integrated with expensive precious metals) still remains a major setback (Ocampo et al., 2011;Sundari & Vaidya, 2012;Zhang et al., 2020). More so, it is also clear that the interaction between the Ni metal catalyst and ceriadoped zirconia catalyst though better when compared to using pure CeO 2 is still not robust enough to offer the much desired resistance to coking activities, thermal stability and reducibility (Wang et al., 2011). Interestingly, not much is reported on the use of a perovskite material as promoter to strengthen the metal catalyst-support interaction, improve structural properties, catalyst reducibility and general catalytic behaviour and resistance to coking activities.
In a novel approach, this work attempts to improve on the structural properties and catalytic performance of Ni/ Al 2 O 3 using a perovskite promoter La 0.75 Sr 0.25 Cr 0.5 Mn 0.5 O 3-δ (LSCM) in a new catalyst formulation Ni-LSCM/Al 2 O 3 . It is also to investigate the synergic influence between the LSCM promoter and alternative support Ce 1-x Zr x O 2 (x = 0.25, 0.5 and 0.75) in another catalyst formulation Ni-LSCM/Ce 0.75 Zr 0.25 O 2 on the structural properties and catalytic behaviour. The work focuses largely on catalytic activity and suppression of coking activities in relation to steam reforming of both pure and by-product glycerol. The catalyst could be quite useful for hydrogen-rich gas and syngas biofuels synthesis from biomass and hydrocarbon feedstock for different renewable energy applications and production of other value-added renewable chemicals.

| Synthesis of catalyst systems
All the supported catalyst systems were prepared by method of incipient wet impregnation. Commercial γ-Al 2 O 3 powder was dried at 700°C for 5 h to eliminate possible adsorbed gases and moisture. The colloid mixture was formed by suspending the γ-Al 2 O 3 powder in aqueous solution of Ni(NO 3 ) 2 .6H 2 O at 60°C under stirring for 5 h and baked at 80°C on hot plate and dried further in oven overnight. The sample was air calcined at 500°C for 5 h after it was ground to fine powder. The calcined powder was crushed to fine powder after cooling and kept for further analysis.
The preparation of Ni-LSCM/Al 2 O 3 catalyst followed similar procedure as in above except that as prepared LSCM from combustion synthesis was mixed along with Ni(NO 3 ) 2 .6H 2 O solution before the γ-Al 2 O 3 powder support was added.
Preparation of Ni-LSCM/Ce 0.75 Zr 0.25 O 2 catalyst also followed similar procedure as in above where both Ni and LSCM were deposited on pre-reacted Ce 0.75 Zr 0.25 O 2 . Evaporation-induced self-assembly (EISA) was used for the synthesis of mesoporous Ce 1-x Zr x O 2 (x = 0.25, 0.5 and 0.75) support. Required quantity of Ce(NO 3 ) 3 ·6H 2 O and ZrOCl 2 ·8H 2 O and 0.5 g of P123 pore former was dissolved in 10 mL ethanol and then stirred vigorously at room temperature for 2 h to form a gel. The organic template was removed via slow evaporation of the sol-gel at low temperature of 60°C before calcination at 500°C at the heating rate of 1°C/min and dwelled for 4 h (Yuan et al., 2009).

| Catalyst characterizations
Investigation of physicochemical properties (surface area pore size and volume) was done using Micrometrics TriStar II model via nitrogen adsorption/desorption at 77.35 K. Prior to the analysis, the samples were heated for 3 h at 120°C to degas traces of adsorbed moisture and gases.
Crystallographic data were taken at room temperature using Pan-Analytical Empyrean X-Ray Diffractometer operating on Cu-Kα 1 radiation at λ = 1.5406 Å using reflection mode at 10-90 2θ angle range for 1 h.
High-resolution transmission electron microscopy (HRTEM) was used to investigate the surface morphology and pores of the mesoporous Ce-Zr support.
Used catalyst systems collected after the test were investigated for possible carbon deposition by temperature programmed oxidation (TPO) using NETZSCH TG 209 thermogravimetric analyser integrated with thermostar mass spectrometer. Certain weight of the used catalyst was inserted into the TPO instrument after which argon gas was used to flush and remove residual gases in the system. The surface carbon residues were gasified and oxidized using O 2 to CO 2 at different temperatures up to 900°C depending upon the type. The gases produced were further analysed through the mass spectrometer.

| Catalyst testing
The catalytic biomass conversion via steam reforming of glycerol was investigated using a quartz tube reactor with the dimension of 10 mm outer diameter, 8 mm inner diameter and 24 cm long at reaction temperatures of 300-600°C and atmospheric pressure. 10% H 2 was used to reduce the nickel particles catalyst at 500°C for 1 h just before the commencement of the test. Harvard Apparatus 22 infusion pump at the flow rate of 0.010-0.019 mL/min was used to supply glycerol/water mixture to a vaporizer/mixer wrapped with heating tape at 250°C. A Helium carrier gas at flow rate of 40 mL/min was used to convey the vapour into the fixed bed reactor. Weight hourly space velocity (WHSV) of 28 h −1 was maintained throughout the test. Steam reforming of pure and by-product glycerol was explored using steamto-carbon ratio (S/C) of 3:1 and 1:1 to screen the catalyst systems for 2 h and durability/stability test for 9 h. Analysis of gaseous products was done on GC equipped with TCD (HP 6890 series) and mass spectrometer residual gas analyser.

| Crystallographic data
The optimized synthesized support materials Ce 1-x Zr x O 2 (where x = 0.25, 0.5 and 0.75) and γ-Al 2 O 3 as well as the catalyst systems in this study-Ni/Al 2 O 3 , Ni-LSCM/Al 2 O 3 and Ni-LSCM/Ce 0.75 Zr 0.25 O 2 were subjected to XRD analysis and the room temperature crystallographic data obtained were used to ascertain their phase purity. The XRD pattern of the supports Ce 1-x Zr x O 2 (where x = 0.25, 0.5 and 0.75) as shown in Figure 1a suggests that they were all successfully synthesized and in pure phase. Ce 0.25 Zr 0.75 O 2 showed extra peak at 43° 2θ angle position probably due to over stretching of the doping limit of ceria.
Indexing of the XRD pattern shown in Figure 1a shows the XRD pattern of the three samples having an intense peaks at 2θ = 28.8°, 33.3°, 47.7° and 56.6° suggests (111), (200), (220) and (311) plane of cubic fluorite structure of ceria (Profeti et al., 2009). In Figure 1b, the diffraction peak due to NiO occurred at 2θ = 37.1, 43.2, 62.7, 75.5 and 79.4°, active γ-Al 2 O 3 at 2θ = 45.7 and 66.9° while peaks due to Ce 1-x Zr x O 2 are as seen in Figure 1a respectively. The LSCM peaks not visible in the pattern because of dispersion or small quantity that was used. Interestingly, to enhance ceria-mediated CO oxidation for an improved catalytic behaviour, low-temperature calcination synthesis of 500°C was adopted in the preparation of the materials (Bunluesin et al., 1998).

| Surface physicochemical properties and microstructure
The Ce 1-x Zr x O 2 (where x = 0.25, 0.5 and 0.75) series support materials were synthesized to optimize surface area to achieve maximum and optimum catalytic activity. The doping ratio of 25%-75% was adopted and sample with 25% Zr, that is, Ce 0.75 Zr 0.25 O 2 gave the highest surface area as seen in Table 1 and was selected for the catalytic studies. The materials have shown suitable pore volume and pore diameter for effective catalysis and dissemination of resultant products. Table 2 Figure 3a and existence of nanofibres of interconnected pore channels of approximately 5 nm in Figure 3b,c as further confirmed by the BET test results of Table 1   results of the synergic influence between the LSCM promoter and the supports on the nickel catalyst and also the fact that this work does not involved the use of expensive precious metal. The synergic influence on the nickel catalyst strengthened the catalyst-support interaction thereby suppressing catalyst deactivation and enhancing glycerol conversion. These have as well reduced the occurrence of side reactions and dehydration that could lead to formation and deposition of carbonaceous residues and consequent deactivation as seen using Ni/Al 2 O 3 catalyst system (Umar & Irvine, 2020;Wu et al., 2013

F I G U R E 4
Influence of promoter and support on product distribution; hydrogen production and glycerol conversion.
less CO in comparison with the LSCM promoted and nonpromoted Al 2 O 3 supported catalyst systems which yielded more CO relatively as evidenced in Figure 4. The observed trend yielding high concentration of H 2 and CO 2 in this work demonstrates LSCM-modified ceria's support properties for water-gas shift reaction (Equation 1; Levalley et al., 2014;Pal et al., 2018) as well as CO oxidation due to improved LSCM-mediated ceria's oxygen storage-release and redox properties that enhanced and facilitated oxygen mobility. The Ni/Al 2 O 3 could also support WGSR because of its Ni content but the formulation in this case favours the reverse water-gas shift reaction which is known to favour more CO yield (Cheng et al., 2011) as shown in Equation 2.
More so, the LSCM-promoted systems have yielded more hydrogen with both Ce 0.75 Zr 0.25 O 2 and Al 2 O 3 supported catalysts showing better performance notably in hydrogen production and glycerol conversion when compared to non-promoted Ni/Al 2 O 3 . In comparison, this shows that the synergic influence between the LSCM and the supports on the nickel catalyst has significantly enhanced catalytic behaviour of Ni/Al 2 O 3 catalyst systems in relation to hydrogen production, glycerol conversion and suppression of side reactions that yield carbon deposition in this approach relatively (Wolfbeisser et al., 2016). Consequently, the result further reveals an improved general catalytic behaviour for Ce 0.75 Zr 0.25 O 2 support in oxidation of hydrocarbon and CO and support for watergas shift reaction to generate hydrogen when combined with LSCM hence corroborating the versatilities of ceriabased catalysts (Vita, 2020). The resultant performance of Ni-LSCM/Al 2 O 3 in H 2 and CO production is a result of the catalyst support for steam reforming of methane and perhaps dry reforming of methane (Adhikari et al., 2007;Umar & Irvine, 2020) as well as seen in Equations 3 and 4, respectively, due to the integration of LSCM promoter in the system hence could be useful for syngas production.
Moreover, the promoting behaviour of LSCM promoter in this work could be compared to a reported work involving Mn-doped Ce-Zr where the Mn not only significantly improved the performance of the catalyst Ce-Zr system but suppressed the carbon deposition significantly. The behaviour was attributed to Mn tendency to enhance the oxygen storage capacity as well as oxygen mobility on the surface of Ce-Zr for effective oxidation of carbonaceous residues (Bampenrat et al., 2010). The LSCM promoter in this study showed similar effect and despite the fact that this test was done at 500°C while 700°C was used in the other case, the results in this work compare favourably. The robust catalytic behaviour of the catalysts in this work is also traced to its La and Sr ions from the LSCM that further improved surface basicity as well as the contribution of Cr which facilitated the suppression of coking activities as seen in a related work involving chromia and basic metals containing materials (Tao & John, 2004;Umar et al., 2021). Suffices to observe that the glycerol conversion was not complete due to factors which includes the amount of catalyst used that resulted into thin catalyst bed hence short contact time and much throughput and glycerol passing unconverted. Nonetheless, the conversion obtained is reasonable and corroborates the robust property of the new catalyst systems.

| Stability of the catalyst systems to prolong usage in gaseous streams
Hydrogen production monitored over time for a longer period in a stability test at 500°C has further revealed difference in durability and stability among the catalyst systems and the robust catalytic influence of the LSCM promoter. Figure 5 shows how the hydrogen generated in mole per mole of glycerol was sustained among the different catalysts over a period of 9 h. The non-promoted Ni/ Al 2 O 3 shows some evidence of fast deactivation over time (1 during the prolonged 9-h test compared to the more stable LSCM-promoted catalysts relatively. The deactivation of Ni/Al 2 O 3 was attributed to rapid coking activities as a result of dehydration leading to the formation of ethene and other carbon residues that could poison active site and causing pore blockage, reduced surface area, agglomeration of crystal particles, restricted diffusion of reactants and products and overall poor catalytic activity (Kim et al., 2021;Wu et al., 2013). The enhanced properties of the supports due to the influence of LSCM promoter have slowed the deactivation process and significantly sustained the hydrogen yield with the Ni-LSCM/Al 2 O 3 and Ni-LSCM/Ce 0.75 Zr 0.25 O 2 catalysts for the 9-h prolonged test. Consequently, it is important to observe that the influence of LSCM as a promoter has stabilized hydrogen production due to the stability of the nickel nanoparticles catalyst in Ni-LSCM/Al 2 O 3 and Ni-LSCM/Ce 0.75 Zr 0.25 O 2 catalyst systems, respectively, against re-oxidation, sintering and coking activities that might lead to catalyst deactivation (Boaro et al., 2019;Silva et al., 2015). The synergy between the LSCM promoter and Ce-Zr support in Ni-LSCM/Ce 0.75 Zr 0.25 O 2 catalyst has yielded improved metal catalyst-support interaction which strengthened stability and the oxygen storage-release capacity of the catalyst system. These facilitated the oxidation of carbonaceous residues as a result of increased oxygen mobility on the catalysts surface hence suppressed catalyst deactivation and improved stability compared to other work reported using alternative materials as supports and promoters (Boaro et al., 2019;Montini et al., 2016). The good surface area and small particles of the Ce 0.75 Zr 0.25 O 2 as seen in Table 1 has helped to improve the observed catalytic behaviour. Therefore, perovskite promoter and choice of suitable support could be tailored to improve stability and reusability of nickel-based catalysts for reforming of hydrocarbons and biomass feedstock for many applications.

| Trend in catalytic activity in relation to temperature changes
Changes in products distribution particularly the production of hydrogen with the reforming temperature was investigated at low and moderate temperatures of 300-600°C on the more active Ni-LSCM/Ce 0.75 Zr 0.25 O 2 catalyst system. Steam to carbon ratio (S/C) ratio of 3:1 using pure glycerol at WHSV of 28 h −1 and glycerol solution molar flow rate of 2.60 × 10 −4 mol/min was maintained throughout the test. The results obtained at the steady time of the test as in Figure 6 shows the pattern and trend of catalytic activity and products distribution and glycerol conversion as the reforming temperature changes. There was a steady hydrogen and carbon dioxide production as well as glycerol conversion with increase in reforming temperature to a maximum level of production at 500°C. Interestingly, while Ni/Al 2 O 3 catalyst system is not very active at low reforming temperature due to dehydration and formation of ethylene and propylene hence gets deactivated rapidly from coking activities (Umar & Irvine, 2020;Wu et al., 2013), the precious metal-free Ni-LSCM/Ce 0.75 Zr 0.25 O 2 in this test yielded a steady glycerol conversion from temperature as low as 300°C to moderate temperature of 500°C. The synergic influence between the LSCM promoter and Ce-Zr support on the nickel catalyst boosted the catalyst-support interaction which suppressed other reactions that would have deactivated the catalyst. The enhanced Ni-support interaction changed the reaction kinetics by increasing gasification rate and decomposition of glycerol and coke resistance though oxidation of carbonaceous products resulting to the steady glycerol conversion and hydrogen production even at low temperature compared to Ni/Al 2 O 3 (Paweewan et al., 1999). The hydrogen yield and glycerol conversion reflect what thermodynamic predicted at that temperature due to the influence of the catalyst (Adhikari et al., 2007).
While H 2 and CO 2 profile in Figure 6 and corroborated by the H 2 /CO 2 ratio of Table 3 show a rise to maximum at 500°C before falling at 600°C, the CO profile showed near steady condition or very little rise to maximum at 600°C. The CH 4 profile got to maximum at 500°C after steady rise with temperature. The steady rise in H 2 and CO 2 profile was attributed to the endothermic property of steam reforming reaction and glycerol pyrolysis which both increased with increase in temperature under the influence of the catalyst generating lots of H 2 and CO 2 as demonstrated in Equations 5 and 6, respectively, and also evidenced by the H 2 /CO 2 mole ratio in Table 3. The low or slight rise in CH 4 production could be attributed to little methanation reaction due to the exothermic property of the reaction (Equation 7). F I G U R E 6 Influence of temperature on hydrogen production and glycerol conversion.
The exothermic nature of water-gas shift reaction kept H 2 and CO 2 production steady under the influence of the catalyst at temperatures up to 500°C and perhaps methane reforming or dehydrogenation as shown by thermodynamic evaluation (Adhikari et al., 2007). The trend of catalytic activities shown by Ni-LSCM/Ce 0.75 Zr 0.25 O 2 catalyst in this study compares more favourable with a reported work for H 2 production via glycerol steam reforming over Ni/Al 2 O 3 by Umar and Irvine (2020) and Wang et al. (2022) looking at the reaction temperatures. The lower value of the evaluated CO/CO 2 ratio at 500°C in Table 3 is a clear evidence of intense WGS reaction yielding lots of H 2 and CO 2 at that temperature while the relatively higher values of CO/CO 2 ratio at other temperatures suggest occurrence of some possible unwanted reactions yielding a corresponding lower performance (Kamonsuangkasem et al., 2011;Kauppi et al., 2010). Consequently, the optimum temperature for high hydrogen production and glycerol conversion as seen in this work is 500°C. Deactivation due to reversible self-poisoning that occurred at 600-650°C during reforming processes is responsible for the drop in activity observed at 600°C in this work (Breen et al., 2002). Therefore, reforming of biomass or hydrocarbon feedstock at low temperatures could be achieved and with a good result on nickel-containing catalysts using perovskite promoter developed from cheap metals without the use of expensive precious metals.

| Influence of steam-to-carbon (S/C) ratio on the catalytic activity in relation to pure and by-product glycerol
Influence of S/C ratio and glycerol type on hydrogen production via steam reforming of glycerol on Ni-LSCM/Ce 0.75 Zr 0.25 O 2 catalyst was investigated to further strengthen the robust properties of the catalysts system and the results obtained are shown in Figure 7. S/C ratio of 3:1 and 1:1 at the required WHSV of and glycerol/water mixture molar flow rate using both pure and by-product glycerol was used for the test. Hydrogen generated in mol per mol of glycerol and glycerol conversion was used as index for catalytic activity. The results obtained as shown in Figure 7 have demonstrated the influence of S/C ratio on the catalytic activity of the catalyst in the steam reforming of glycerol and suppression of coking activities. Hydrogen production as well as the glycerol conversion has reduced when S/C ratio lower than stoichiometric amount was used in the reforming test suggesting that forward reaction of WGSR was favoured by water which led to more hydrogen production and glycerol conversion (Pal et al., 2018;Vita, 2020). Hence, use of optimum dilution ratio enhances catalyst performance and product yield though the analysis above shows that the water dilution has negative influence on the production rate of CH 4 .
The glycerol conversion and hydrogen production using glycerol by-product despite its impurities such as biodiesel, methanol, unconverted vegetable oil, intermediate products (mono-, di-and triglycerides) residues that could reduce activity, the performance compares very well with that of pure glycerol. Going by the reaction conditions and catalyst amount used in this test, results obtained using by-product glycerol show good correlation with what Kamonsuangkasem et al. (2011) reported using different grades of glycerol and as others reported in related work (Adeniyi & Ighalo, 2019;Schwengber et al., 2016). The result of this test suggests that the waste by-product glycerol glut from the fast-growing biodiesel industry is a good energy resource that could be harnessed for different energy application to provide the much desired renewable energy and value-added chemicals. This would also reduce the T A B L E 3 Molar ratio of CO to CO 2 produced as a function of temperature on Ni-LSCM/Ce-Zr catalyst system with pure glycerol and S/C 3:1showing level of occurrence of unwanted reactions as temperature changes. cost of biodiesel production and boost full commercialization of biodiesel.

| Analysis of carbon deposition on the catalyst surface
Coking or carbon deposition leading to catalyst deactivation has always been a major problem in reforming of hydrocarbons and biomass feedstock such as glycerol (Adeniyi & Ighalo, 2019;Schwengber et al., 2016). Coking has negative influence on catalyst activity largely through poisoning of active sites and pore blockage by carbonaceous residues from biomass decomposition resulting into reduced surface area and restricted movement of reactants and products with consequent low performance (Cheng et al., 2011;Umar & Irvine, 2020). TPO profile gives CO 2 peaks due to oxidation of carbon at different temperatures from the catalyst surface depending upon the location, type or nature (amorphous, aromatic, polymeric or graphitic) of the carbon on the catalyst surface hence represents an index for the characterization of carbon deposited during steam reforming. Amorphous coke deposited on the active phase or metal centres of the catalyst are easy to remove hence yield a low-temperature CO 2 peaks compared to more severe polymeric or graphitic coke deposited at metal-support interface and bulk coke deposited on the support that are more difficult to remove and therefore yield medium-and high-temperature CO 2 peaks respectively (Mawdsley & Krause, 2008). More so, the TGA profile on the TPO profile could also be used as index for quantifying the amount of carbon deposited which reflects the extent of the damage as a result of coking activities.
The CO 2 peaks and TGA profile of Figure 8a were obtained from the TPO test carried out on all the used or post-test catalyst systems tested at 500°C reaction temperatures. Generally, the CO 2 peaks occur at approximately 550°C with all the samples suggesting that most of the coke is polymeric or graphitic deposited at the interface between the active phase and support. Polymeric and graphitic carbon are difficult to remove as such oxidizes at slightly high temperatures. Interestingly, unlike the Ni/Al 2 O 3 catalyst system having only one (difficult polymeric or graphitic) type of carbon, the Ni-LSCM/ Ce 0.75 Zr 0.25 O 2 catalyst system showed two distinct CO 2 peaks depicting two types of carbon species on the catalyst surface.
The low-temperature CO 2 peak shows oxidation of amorphous carbon usually deposited on the metal surface and the other CO 2 peak at intermediate temperature indicates graphitic carbon. These suggest a mixture of polymeric or graphitic and amorphous carbons. Unlike polymeric or graphitic carbon, amorphous carbon is easy to remove and therefore oxidizes at low temperature hence improving reusability of the catalyst. This confirms that the properties and type coke deposited is a subset of the type of substrate used as support, its interaction with a promoter and extent of its involvement in the reaction cycles. Therefore, while Ni/Al 2 O 3 catalyst system is associated largely with severe polymeric or graphitic, conversely, the novel Ni-LSCM/Ce 0.75 Zr 0.25 O 2 is associated largely with simple amorphous carbon which allows the catalyst to be regenerated and reused. Hence, suffices to say the synergy between the LSCM promoter and the Ce 0.75 Zr 0.25 O 2 support enhances reusability of the catalyst making it more economically viable compared to Ni/Al 2 O 3 or even Ni/Ce-Zr catalyst F I G U R E 8 Change in weight of the catalyst samples used for the test in oxidizing atmosphere monitored in parallel with CO 2 produced as a function of temperature (a) with pure glycerol and 1:3 S/C ratio for all the catalyst systems, (b) with pure and by-product glycerol and S/C ratio of 3:1 and 1:1 for the Ni-LSCM/Ce-Zr catalyst. that is not promoted. Therefore, the behaviour of this new catalyst system Ni-LSCM/Ce 0.75 Zr 0.25 O 2 shows an improvement over Ni/Al 2 O 3 and compares with a better performance in terms of coke resistance and stability with what is reported using other supports and promoters (Adeniyi & Ighalo, 2019;Boaro et al., 2019;Gao et al., 2020;Schwengber et al., 2016).
The weight loss on the TGA profile also confirmed carbon deposition with all the catalyst systems. The Ni/Al 2 O 3 catalyst system has the highest carbon deposition of 25% which is attributed to strong adsorption of glycerol on the alumina support week acid site resulting to formation of polymeric carbon species due to dehydration. The synergy between the LSCM and supports (Al 2 O 3 and Ce 0.75 Zr 0.25 O 2 ) suppressed the coking activities on both Ni-LSCM/Al 2 O 3 as well as Ni-LSCM/Ce 0.75 Zr 0.25 O 2 catalyst systems due to the influence of chromia and manganesia (Tao & John, 2004;Umar et al., 2021). The improved oxygen storagerelease of Ce 0.75 Zr 0.25 O 2 which enhanced oxygen mobility that facilitated oxidation of carbonaceous residues gave an added advantage to it which explains its superior performance (Adeniyi & Ighalo, 2019;Schwengber et al., 2016). There is also an improved acidic-basic surface property of Ni-LSCM/Ce 0.75 Zr 0.25 O 2 due to La and Sr ions in the LSCM promoter which helped in the suppression of carbon deposition through enhanced oxidation of carbon residue despite the rigorous catalytic activity (Schwengber et al., 2016;Vinodkumar et al., 2015). The extent of carbon deposition reduces from 25% in Ni/Al 2 O 3 to 15% and 2% in Ni-LSCM/Al 2 O 3 and Ni-LSCM/Ce 0.75 Zr 0.25 O 2 respectively. Thus, suppression of carbon deposition by the catalysts followed the order Ni/Al 2 O 3 > Ni-LSCM/ Al 2 O 3 > Ni-LSCM/Ce 0.75 Zr 0.25 O 2 .
The influence of S/C ratio and grade of glycerol on the carbon deposition was also invested on the used catalyst and the result obtained is shown in Figure 8b. The CO 2 peak on the TPO profile revealed a medium-temperature carbon dioxide peak irrespective of the S/C ratio used or glycerol type. Use of stoichiometric S/C ratio of 3:1 gave another CO 2 peak at low temperature using either pure or by-product glycerol suggesting formation of amorphous carbon apart from the polymeric or graphitic carbon associated with S/C ratio of 1:1. Therefore, more severe carbon of polymeric or graphitic type is generated when steamto-carbon ratio below the stoichiometric 3:1 is used due to probable formation of intermediates and by-products which signifies the influence of water in suppressing carbon deposition. This observation was corroborated by the TGA profile of Figure 8b which shows significant increase in carbon deposition when S/C ratio less than the stoichiometric value was used. Therefore, use of a perovskite promoter under appropriate S/C ratio and choice of support system could be tailored to design and develop catalyst system with improved catalytic behaviour and robust suppression of coking activities.

PERSPECTIVES
This work has demonstrated that both pure and cheap, readily by-product glycerol could be converted catalytically to hydrogen-rich gas and syngas for different energy applications such as fuel cell technology as renewable fuel to the fossil-based conventional hydrogen and possibly for the production of other value-added chemicals. The work has revealed that the synergy between the promoting activity of the LSCM and that of the Al 2 O 3 support in Ni/LSCM-Al 2 O 3 has significantly improved the catalytic behaviour of Ni/Al 2 O 3 by hindering crystal growth/agglomeration and re-oxidation of nickel catalyst as well as enhancing catalyst-support interaction and more importantly suppression of carbon deposition activities. The synergic influence was further improved using alternative support Ce 0.75 Zr 0.25 O 2 in a novel formulation Ni-LSCM/Ce 0.75 Zr 0.25 O 2 . Hence, integration of a robust perovskite material and use of alternative support such as Ce 0.75 Zr 0.25 O 2 could be tailored in design and development of robust catalyst systems to improve performance of metal-supported catalysts. Furthermore, both the promoter and the support were formulated using low cost transition and rare earth metals as against the use of expensive precious metals. In future work, therefore, other perovskite materials based on transition metals such as titanium, magnesium, iron etc. on the B-site and rare earth metal in combination with basic metals such as groups 1 and 2 metals could be explored as promoters in the design and development of catalyst systems for reforming of biomass feedstock. They could also be explored as alternative supports in a related work.