CO_x -free H₂ Production via Catalytic Decomposition of CH₄ over Fe Supported on Tungsten oxide-activated Carbon Catalyst: Effect of Tungsten Loading

2.1 Materials and methods

The following materials were used in the preparation of the newly designed catalysts; Sodium tungstate dehydrate (Na₂WO₄·2H₂O, ≥ 99% Sigma Aldrich), sodium chloride (NaCl; ≥ 99.0%, Sigma Aldrich), hydrated iron nitrate (Fe(NO₃)₃·9H₂O; 99%; Loba Chemie), hydrochloric acid (HCl; 37%, Sigma Aldrich) and waste of date pits (collected from Albaha region, Saudi Arabia).

2.2 Catalyst preparation

2.3 Catalyst characterization

X-ray diffraction (XRD) study of catalyst samples was carried out by Rigaku diffractometer using Cu Kα radiation source operated at 40 kV and 40 mA. 0.01 step size and 5–100 scanning range were set for analysis. Phase analysis was carried out by using X’pert high score plus software and JCPDS database. N₂-physiosorption isotherms study of catalyst sample was carried over Micromeritics Tristar II 3020. Surface area was estimated by Brunauer-Emmet Teller (BET) method whereas pore volume and pore diameter were estimated by Barrett-Joyner-Halenda (BJH) method. The reducibility of the catalyst sample was studied by H₂-temperature-programmed reduction (TPR) over Micromeritics Auto Chem II 2920, USA. 70 mg of the sample was subjected to a heat treatment at 10 °C/min up to 900 °C under 30 ml/min gas flow of 10% H₂/Ar mixture gas. The thermogravimetric analysis (TGA) was carried out over 0.015 g of spent catalyst sample in the temperature range (room temperature to 1000 °C) at heating ramp 20 °C by using Shimadzu TGA-51. The TGA analysis was carried out under oxidizing gas O₂. The weight loss/weight gain of catalyst sample against temperature was monitored continuously. O₂-Temperature programmed oxidation (TPO) was carried out over spent catalyst system in 50–800 °C temperature range by using a 10% O₂/He mixture through by Micromeritics AutoChem II. Before analysis, the spent catalyst was treated under high purity Argon at 150 °C for 30 min and subsequently cooled to room temperature. The morphology of the catalyst sample was investigated by using a field emission scanning electron microscope (FE-SEM, model: JEOL JSM-7100F) and transmission electron microscope (TEM, model: 120 kV JEOL JEM-2100F). Element valance state and binding energy of electron were determined by X-ray photoelectron spectroscopy (XPS) (Themo Fisher Scientific, USA) operated through Al_Kα excitation source and 20 eV pass energy.

2.4 Catalyst activity test

The detailed reaction set up for the CH₄ decomposition reaction is shown in Fig. 1. Catalytic decomposition of methane was carried out over 0.15 g catalyst packed in fixed-bed stainless steel tubular micro-reactor (PID Eng & Tech micro activity reference company; L = 30 cm, I.D = 9.1 mm) at atmospheric pressure. The reactor temperature was monitored by an axially positioned K-type stainless steel sheathed thermocouple at the centre of the catalyst bed. Prior to the reaction, the catalyst was activated under 40 ml/min flow of H₂ for 60 min at 600 °C. Futher reactor is purged by N₂ for 15 min to remove the remnant of H₂. Now, the temperature of the reactor was raised to 800 °C under flow of N₂. 15 ml/min CH₄ and 5 ml/min N₂ (total flow rate of feed gas 20 ml/ min) was allowed to pass through the catalyst bed at 800 °C with 8000 ml/hg_cat space velocity of. GC-2014 SHIMADZU (Column: Shin carbon C20380 for gases and Haysepe Q AC0209 column for water analysis; carrier gas: Argon) equipped with conductivity detector was used to analyse the feed and output gas composition. The expression for CH₄ conversion, H₂ yield and Carbon yield (%) are given as $${\mathrm{C}\mathrm{H}}_4\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}=\frac{{\mathrm{C}\mathrm{H}}_{4,\mathrm{i}\mathrm{n}}-{\mathrm{C}\mathrm{H}}_{4,\mathrm{o}\mathrm{u}\mathrm{t}}}{{\mathrm{C}\mathrm{H}}_{4,\mathrm{i}\mathrm{n}}}\times 100\%$$ $$H_2\mathrm{Y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}\%=\frac{\mathrm{M}\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{o}\mathrm{f}{\mathrm{H}}_2\mathrm{i}\mathrm{n}\mathrm{P}\mathrm{r}\mathrm{o}\mathrm{d}\mathrm{u}\mathrm{c}\mathrm{t}}{2\mathrm{x}\mathrm{m}\mathrm{o}\mathrm{l}\mathrm{o}\mathrm{f}{{\mathrm{C}\mathrm{H}}_4}_{\mathrm{i}\mathrm{n}}}\mathrm{x}100$$ $$\mathrm{C}\mathrm{a}\mathrm{r}\mathrm{b}\mathrm{o}\mathrm{n}\mathrm{y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}\left(\%\right)=\frac{\left(Wp-Wcat\right)\times 100}{\mathit{Wcat}}$$

Close

Fig. 1

Reaction set up for CH₄ decomposition reaction.

Export to PPT

(Where W_p is the weight of the product after reaction and W_cat is the weight of the fresh catalyst).

3.1 Characterization results

The X-ray diffraction pattern of 30FexW(100-x) Ac (x = 0–100%) catalysts are shown in Fig. 2 and Fig. 3. 30 wt% Fe supported over activated carbon had only phases related to iron oxide at Bragg angle (2θ) 24.07°, 33.12°, 35.60°, 39.19°, 40.80°, 49.38°, 54.10°, 57.56°, 62.33°, 63.99°, 69.55°, 71.80° (JCPDS reference number 00–024-0072) (Fig. 2 A-C). As activated carbon amount is substituted by tungsten oxide up to 5–10 wt% W, the diffraction peak intensity for iron oxide is suppressed greatly in 30FexW(100-x) Ac (x = 5–10) catalyst. However, upon the 1:3 ratio of W and Ac (25 wt%WO₃-75%Ac), the many peaks related to Fe₂O₃ again appeared with variation in intensity over 30Fe25W75Ac catalyst (Fig. 2 D). The 30Fe25W75Ac catalyst also shows crystalline carbon phases at 2θ values of 27.38°, 36.10°, and 62.7° (JCPDS reference number: 00–018-0311) (Fig. 2 B-C). When “50 wt% WO₃-50 wt% Ac” support is prepared for 30 wt% Fe dispersion, orthorhombic tungsten oxide phases at 22.97°, 24.02°, 33.22°, 35.70°, 49.55°, 54.24°, 62.59° (JCPDS reference number: 00–020-1324), tungsten carbide phase at 35.70°, 64.11° (JCPDS reference number 01–073-0471) and Fe₂(WO₄)₃ phases at 20.33°, 22.45°, 22.97°, 25.45°, 29.98° (JCPDS reference number 00–038-0200) are appeared additionally (Fig. 2 D). Upon 3:1 ratio of W and Ac respectively, the 30Fe75W25Ac catalyst shows the most intense diffraction peak patterns along with additional diffraction peaks for tungsten oxide (monoclinic phase) at 23.61°, 28.70°, 30°, 34.11°, 38.94°, 56.29°, 68.81° (JCPDS reference number 01–072-0677) (Fig. 2 E-F). It may be expected that on further increasing the weight ratio of W and Ac (W/Ac = 75/25, 90/10, 95/5), the peak intensity of tungsten-related phases should be increased but the opposite diffraction results are noticed (Fig. 3 A-C). It indicates either the addition of “5-10 wt% activated carbon in tungsten oxide matrix” or “addition of 5-10 wt% WO₃ in activated carbon” brings a drop of the crystallinity of the catalyst sample. Finally, on complete substitution of activated carbon by tungsten oxide, 30Fe100W catalyst shows iron oxide, tungsten oxide and the intense peak intensity for Fe₂WO₆ mixed oxide (at 20.50°, 27.41°, 31.08°, 33.21°, 35.99°, 39.09°, 49.59°, 54.19°; JCPDS reference number 00–015-0688) (Fig. 3 D-F).

Close

Fig. 2

X-ray diffraction of (A-C) 30FexW(100-x) Ac; (x = 0, 5, 10, 25) catalyst (D-) 30FexW(100-x) Ac; (x = 25, 50) catalyst (E-F) 30FexW(100-x) Ac (x = 50, 75) catalyst; C = Carbon, W = WO₃, @ = Fe₂O₃, FW = Fe₂(WO₄)_3.

Export to PPT

Close

Fig. 3

X-ray diffraction of (A-C) 30FexW(100-x) Ac (x = 75, 90, 95) catalyst (D-F) 30FexW(100-x) Ac (x = 95, 100); C = Carbon, W = WO₃, fw = Fe₂WO₆, @ = Fe₂O_3.

Export to PPT

The support “activated carbon” has 0.8348 m²/g surface area, 0.002550 cm³/g pore volume, and 393 Å pore diameter. For 30FexW(100-x) Ac (x = 0–100) catalyst system, adsorption isotherm, pore size distribution, surface area, pore volume, and pore diameter are shown in Fig. 4 and Fig. S2. The catalyst system belongs to the type IV isotherm having an H₃ hysteresis loop. It indicates the presence of non-rigid aggregate-like mesopores. Upon incorporation of 5-10 wt% tungsten oxide, the surface area of 30FexW(100-x)Ac (x = 5, 10) catalyst is 2.5 times than 30Fe100Ac indicating expansion of framework (Kumar et al., 2016). Upon 25 wt% tungsten oxide incorporation, the surface area is noticed to decrease to 30% but pore volume is increased by 46%. However, upon further loading up to 50 wt% W; the surface area and pore volume of 30Fe50W50Ac are increased to 3 times and 1.8 times (with respect to 30Fe100Ac) respectively. The pore size distribution plot (dV/dlogW vs W) indicates that up to 50 % incorporation of tungsten oxide, pore size distributions remain bimodal. In the 30Fe50W50Ac catalyst, the intensity of the low pore-width range is more pronounced than the higher pore-width range. In 30FexW(100-x)Ac (x = 0–100) catalyst systems, when support is made up by major WO₃ than Ac (upon > 50 wt% W incorporation), the pore size distribution becomes multimodal. In 30FexW(100-x)Ac (x = 75–100),the surface area decreases suddenly to 11–22 m²/g (against 59.15 m²/g in 30Fe50W50Ac) due to deposition of various crystallite inside the pore (Rahman et al., 2015).

Close

Fig. 4

N₂-adsorption isotherm and porosity distribution profile of (A) 30FexW(100-x)Ac(x = 0–25) catalyst (B) 30FexW(100-x)Ac(x = 50) catalyst (C) 30FexW(100-x)Ac(x = 75–100) catalyst (D) Surface parameters of 30FexW(100-x)Ac(x = 0–100) catalyst.

Export to PPT

The H₂-Temperatured programmed reduction profile of 30FexW(100-x) Ac (x = 0–100) catalyst systems are shown in Fig. 5 A and Fig. S3. The total H₂-consumption during the H₂-TPR experiment is shown in Table S4. H₂-TPR of Fe₂O₃ is constituted by two sharp peaks at 355 °C and 577 °C and a broad peak between 624 °C and 934 °C (Fig. S3). The three peaks are correlated with the sequential reductions Fe₂O₃ → Fe₃O₄ → FeO → Fe respectively (Ibrahim et al., 2015; Jozwiak et al., 2007). The activated carbon-supported Fe (30Fe100Ac) catalyst shows shifting of lower temperature reduction peak to relatively higher temperature (at 385 °C) indicating interaction of Fe₂O₃-species with support. The peak at 385 °C belongs to reduction of interacted-Fe₂O₃-species into Fe₃O₄. Upon incorporation of just 5 wt% WO₃, a merged peak maximum at 420 °C for reduction of Fe₂O₃ → Fe₃O₄ → FeO and a broad peak at a higher temperature for the reduction of FeO → Fe are observed. Further incorporation of 10 wt% WO₃, the reduction peak of 30Fe10W90Ac catalyst is shifted towards more higher temperature (444 °C). It is noticeable that the amount of reducible iron species had decreased upon providing support as well as increasing the proportion of tungsten oxide (up to 10 wt%) in the support. It indicates that the total reducible quantity has decreased due to the interaction of Fe-species with the new support composed of xW(100-x) Ac (x = 0, 5, 10) catalyst. As well as WO₃ incorporation is increased to 25 wt%, the reduction peak maxima of 30Fe25W75Ac catalyst are shifted to a higher temperature and the amount of reducible iron-species is increased to ∼ 34% with respect to 30Fe10W90Ac catalyst (Table S4). Shifting of reduction peak to a higher temperature also indicates increased metal support interaction upon tungsten oxide loading. 30Fe50W50Ac catalyst has the lower temperature reduction peak (for reduction of Fe₂O₃ → Fe₃O₄ → FeO at 470 °C and broad higher temperature reduction peak with comparable amount of reduceable species than 30Fe25W75Ac catalyst. 30FexW(100-x) Ac (x = 90–100 wt%) showed the reduction peak about 530–560 °C. It again shows a general trend of increasing metal-support interaction upon tungsten oxide loading. The peak pattern of 30FexW(100-x)Ac (x = 90–100 wt%) has also an additional peak in the temperature region of 593 to 720 °C for reduction of WO₃ crystallite or “WO₃ interacted species”(Ramanathan et al., 2013). It is noticeable that the total concentration of reducible species at the catalyst surface is decreased sharply above 50 wt% tungsten oxide incorporation. 30Fe50W50Ac, 30Fe90W10Ac, 30Fe95W5Ac, and 30Fe100W catalysts had 174.2 cm³/g, 72.28 cm³/g, 68.24 cm³/h, and 46.32 cm³/g consumption of hydrogen. The H₂-TPR pattern of the 30Fe25W75Ac catalyst is needed to address separately. It has the highest amount of reducible species over the surface (183 cm³/g H₂ consumption in H₂-TPR result) among other tungsten oxide incorporated catalysts. The H₂-TPR peak pattern is constituted by five peaks enveloping each other at 433 °C, 506 °C, 664 °C, 816 °C and 929 °C. It indicates the 30Fe25W75Ac catalyst had the highest concentration of “Fe-related” reducible species which interacted with the support to different extents.

Close

Fig. 5

(A) H₂-TPR profile of 30FexW(100-x)Ac(x = 5–100) catalyst (B) TGA profile of spent 30FexW(100-x)Ac(x = 5–95) catalyst (C) O₂-TPO of 30FexW(100-x)Ac(x = 5, 50, 90) (D) O₂-TPO of 30FexW(100-x)Ac(x = 10, 25, 95, 100).

Export to PPT

The thermogravimetric analysis (TGA) of spent catalysts is shown in Fig. 5B. The weight gain over the catalyst system in TGA analysis may be due to oxidation of “reduced metal species” (like iron or tungsten-related metal oxide) over the catalyst surface (Ibrahim et al., 2015). In the case of spent 30Fe25W75Ac and spent 30Fe50W50Ac, weight loss due to oxidation of deposit carbon is optimum and so weight gain due to oxidation of “reduced metal species” over these catalyst systems is not evident on TGA analysis. The carbon yield % over the different catalysts is found in the following order; 30Fe25W75Ac (140%) > 30Fe50W50Ac (107%) > 30Fe10W90Ac (1 2 0) 30Fe95W5Ac (93.3%) > 30Fe90W10Ac (66.6%) > 30Fe5W95Ac (13.3 %) > 30Fe75W25Ac (6.7%) (Table S5). Clearly, carbon yield % over 30Fe25W75Ac and 30Fe50W50Ac are higher than other catalysts. Previously, tungsten species were claimed to generate additional CH₄ decomposition sites (Patel et al., 2021). It seems that the presence of 25-50 wt% of WO₃ in the catalyst system cultivates the optimum amount of catalytic active sites which leads potential dissociation of CH₄ into carbon and H₂. It resulted in an excellent carbon yield and severe weight loss. These findings also give the sign of higher activity toward CH₄ decomposition reaction over 30Fe25W75Ac and 30Fe50W50Ac catalysts. Severe weight loss over spent 30Fe25W75Ac and spent 30Fe50W50Ac catalysts also indicates that the carbon deposits over these catalysts are not inert, it is oxidizable under O₂ stream. Inert carbon deposit may shade the catalytic active site permanently and causes fast deactivation. The non-inert carbon deposit over 30Fe25W75Ac and spent 30Fe50W50Ac catalysts may cause slower deactivation than other catalysts.

O₂-TPO of spent 30FexW(100-x) Ac (x = 0–100) catalyst system is shown in Fig. 5C-5D. In the literature, the O₂-TPO peak profile is differentiated into three regions 300–500 °C for easily oxidizable α-carbon (amorphous carbon) species (Al-Fatesh et al., 2021), 500–600 °C for moderately oxidizable β-carbon species (Patel et al., 2021) and > 600 °C for higher crystallization degree of carbon species (Zhang et al., 2015). In our catalyst system, TPO peak maxima is found about ∼ 600 °C in spent-30Fe5W95Ac, spent-30Fe50W50Ac, and spent-30Fe90W10Ac catalysts whereas, for spent-30Fe10W90Ac, spent-30Fe25W75Ac, spent-30Fe95W5Ac and spent-30Fe100W catalysts, TPO peaks are at about ∼ 650 °C. This observation indicates that a particular amount of tungsten oxide (spent-30Fe5W95Ac, spent-30Fe50W50Ac, and spent-30Fe90W10Ac catalyst) in the support induces less crystallization degree of carbon (than30Fe10W90Ac, 30Fe25W75Ac, and 30Fe100W catalyst).

The morphology of catalyst samples is shown by SEM images in Fig. S6. The catalyst morphology of fresh low tungsten-containing samples (30Fe10W90Ac) or high tungsten-containing samples (30Fe90W10Ac) is not differentiable. In the case of spent 30Fe10W90Ac catalyst, carbon tubes are easily observed than in 30Fe90W10Ac catalyst. The morphology of the catalyst and carbon tube is evident in TEM images under Fig. 6. TEM image indicates particle size over the 30Fe25W75Ac has grown from 5.57 nm to 5.8 nm after the reaction (Fig. 6 A-D). Fig. 6 E shows the presence of carbon nanotubes of varying diameters. A typical multiwalled carbon tube having wall width of 3.81–4.74 nm and total tube width of 13.13 nm is evident in Fig. 6F.

Close

Fig. 6

(A) TEM image of fresh 30Fe25W75Ac at 100 nm scale (B) Particle size distribution curve of fresh 30Fe25W75Ac (C) TEM image of spent 30Fe25W75Ac at 100 nm scale (D) Particle size distribution curve of spent 30Fe25W75Ac (E) TEM image of carbon nanotubes in spent 30Fe25W75Ac catalyst at 20 nm scale (D) Single multiwalled carbon nanotube in spent 30Fe25W75Ac catalyst at 5 nm scale.

Export to PPT

The X-ray photo-electron spectra of 30Fe25W74Ac catalyst is shown in Fig. 7. Fe (2p_3/2) peak at 711 eV and Fe (2p_1/2) peak at 725 eV and O (1 s) peak at 530.1 confirms the presence of Fe⁺³ oxidation state (Allen et al., 1974; Konno and Nagayama, 1980) (Fig. 7A- B). The presence of W(4f_7/2) peak at 35.4 eV and W(4f_5/2) peak at 37.6 eV confirm the presence of WO₃ or W⁺⁶ oxidation state (Fig. 7C) (Barreca et al., 2001). The C(1 s) XPS spectra is observed at 284.7 (Barreca et al., 2001; Grünert et al., 1987) (Fig. 7). The 30Fe25W75Ac catalyst has both carbon and WO₃ but absence of carbidic carbon peak at 282.7 eV indicates that WC like species are not formed over the catalyst surface (Katrib et al., 1994). Overall, from the XPS spectra presence of Fe (III) (as Fe₂O₃), W (IV) (as WO₃) species are confirmed. Fe₂O₃ and WO₃ phases are already confirmed during the XRD analysis of sample.

Close

Fig. 7

X-ray photo-electron spectra of 30Fe25W74Ac catalyst (A) Fe(2p) spectra (B) O (1 s) spectra (C) W (4f) spectra (D) C (1 s) spectra.

Export to PPT

3.2 Results and discussion

The CH₄ conversion, H₂-yield and ratio of H₂-yield/CH₄ conversion ( $Y_{H_2}$ / $C_{C{H}_4}$ ) are shown in Fig. 8A-8C. The initial conversion of CH₄ and H₂-yield at 30wt.%Fe supported over activated carbon (30Fe100Ac) is just 4.43% and 1.96% respectively whereas 30wt.%Fe supported over tungsten oxide (30Fe100W) shows 20.5% initial CH₄ conversion and 19.43% initial H₂-yield. Markedly tungsten-oxide supported iron has a higher catalytic activity as well as ratio of $Y_{H_2}$ / $C_{C{H}_4}$ is > 0.94 whereas activated-carbon-supported iron has low catalytic activity and a ratio of $Y_{H_2}$ / $C_{C{H}_4}$ is just half (Fig. 8C). During the entire time on stream (420 min), $Y_{H_2}$ / $C_{C{H}_4}$ ratio remains close to 0.9 over 30Fe100W whereas $Y_{H_2}$ / $C_{C{H}_4}$ ratio drops to ∼ 0.2 over 30Fe100Ac at the end of 420-minutes time on stream. A high $Y_{H_2}$ / $C_{C{H}_4}$ ratio indicates a higher accumulation of CH₄ or CH_x over the catalyst surface followed by higher H₂ release to the gas phase during the reaction (Łamacz and Łabojko, 2019). The higher activity of 30Fe100W than 30Fe100Ac catalyst toward methane decomposition reaction can be explained by X-ray diffraction and H₂-TPR results. 30Fe100Ac catalyst had only reducible iron oxide as surface active species whereas 30Fe100W has iron oxide, tungsten oxide, and Fe₂WO₆ mixed oxide phases. Patel et. al found “additional CH₄ decomposition sites” over tungsten oxide-zirconia in CH₄-temperature programmed surface reaction experiment (Patel et al., 2021). It was reported that Fe₂WO₆ may also be reduced to respective Fe and W under the hydrogen stream at reaction temperature (Pak et al., 2009). In H₂-TPR results, we found the reduction peak for WO₃ at 593 °C to 720 °C over 30Fe100W catalyst. That means 30Fe100W has a variety of reducible surface-active species that markedly influence the CH₄-decomposition reaction. Overall, it can be said that tungsten oxide is promising support for Fe-based catalyst for CH₄ decomposition reaction.

Close

Fig. 8

Catalytic activity results of 30FexW(100-x) Ac (x = 0–100) catalyst (A) CH₄ conversion (B) H₂-yield (C) $Y_{H_2}$ / $C_{C{H}_4}$ ratio.

Export to PPT

Up to incorporation of 5–10 wt% WO₃ in 30FexW(100-x) Ac (x = 5–10) catalyst, X-ray diffraction peaks are suppressed and surface area is increased up to 2.5 times (with respect to 30Fe100Ac catalyst). It indicates the suppression of crystallinity upon expansion of the surface (Khalid et al., 2013). H₂-TPR result of 30Fe10W95Ac showed increased metal-support interaction over than 30Fe5W95Ac catalyst. Increased metal-support interaction over an expanded surface is the ideal condition for exposing more active sites for CH₄ decomposition. Overall, it can be said that 30Fe5W95Ac catalyst has an expanded surface (than 30Fe100Ac) and 30Fe10W90Ac catalyst has an expanded surface (than 30Fe100Ac) as well as increased interaction of reducible Fe-species (Fe₂O₃, Fe₃O₄, FeO) with support (than 30Fe5W95Ac). 30Fe5W95Ac catalyst has 29.5% initial CH₄ conversion (against 4.43% in 30Fe100Ac) and 22.5% initial H₂-yield (against 1.96% in 30Fe100Ac). 30Fe10W90Ac catalyst shows 50.2% initial CH₄ conversion and 48.39% initial H₂-yield. The $Y_{H_2}$ / $C_{C{H}_4}$ ratio over 30Fe10W90Ac also remains ∼ 0.9 up to 250 min whereas, over 30Fe5W95Ac, $Y_{H_2}$ / $C_{C{H}_4}$ ratio falls to 0.5 within 70 min time on stream. Here, the role of tungsten in CH₄ decomposition, higher surface area, and more “surface interacted reducible Fe-species” are evident over 30Fe10W90Ac catalyst which achieves higher CH₄ conversion and higher $Y_{H_2}$ / $C_{C{H}_4}$ ratio.

Up to 25 wt% W incorporation; prominent iron oxide phase is evident over 30Fe25W75Ac catalyst. The presence of Fe₂O₃ and WO₃ phases are also confirmed by Fe(2p), W(4f) and O (1 s) XPS spectra. The decrease in surface area (up to 30%) of 30Fe25W75Ac (with respect to 3010W90Ac) is compensated by an increase in average pore diameter (up to 46%). 30Fe50W50Ac has again expanded surface area (three times than 30Fe100Ac) and other tungsten-related phases like tungsten oxide, tungsten carbide, and Fe₂(WO₄)₃ over the surface. 30Fe25W75Ac catalyst has the highest amount of reducible species (183 cm³/g H₂ consmption in H₂-TPR result) over the surface among the rest tungsten oxide incorporated catalysts. These reducible iron species have interacted with the support along a wide range of temperatures as per the extent of interaction with support. 30Fe50W50Ac catalyst has also good number of reducible species (174 cm³/g H₂ consumption in H₂-TPR result) after 30Fe25W75Ac catalyst. Previously, tungsten species were claimed to generate additional CH₄ decomposition sites (Patel et al., 2021). 30Fe25W75Ac and 30Fe50W50Ac catalysts show severe weight loss and higher carbon yield. The catalyst activity of 30Fe50W50Ac and 30Fe25W75Ac towards the CH₄ decomposition reaction are also close to each other initially. It indicates that the presence of 25-50 wt% of WO₃ in the catalyst induces an optimum amount of catalytic active sites leading to potential CH₄ dissociation, excellent carbon yield, severe weight loss and optimum H₂-yield over 30Fe25W75Ac and 30Fe50W50Ac catalysts. The initial CH₄ conversion of 30Fe25W75Ac and 30Fe50W50Ac catalysts are 66.04% and 64.82 % respectively. Again, the initial H₂-yield over 30Fe25W75Ac and 30Fe50W50Ac catalyst is found 63.12% and 59.81 % respectively.

Fe supported over “xW(100-x)Ac (x = 10–50)” are able to show > 50% CH₄ conversion, ≥50% H₂-yield and ∼ 0.9 $Y_{H_2}$ / $C_{C{H}_4}$ ratio initially. Severe weight loss in TGA profile is obtained over 30Fe25W75Ac and spent 30Fe50W50Ac catalysts. It indicates that carbon deposits over these catalysts are oxidizable/not inert/active. The non-inert carbon deposit deactivates 30Fe25W75Ac and spent 30Fe50W50Ac catalyst slowly than the rest catalysts. After 160 min, CH₄ conversion and H₂ yield of the 30Fe25W75Ac catalyst drop to 37.61% (against 66% initial CH₄ conversion) and 35.2% (against 63.12% initial H₂ yield) respectively. 30Fe10W90Ac catalyst is found the second best as the CH₄ conversion and H₂-yield don’t fall below 25% after 160 min time on stream. The $Y_{H_2}$ / $C_{C{H}_4}$ ratio of both 30Fe25W75Ac and 30Fe10W90Ac catalyst is also ≥ 0.9 up to 240 min. At the end of 160 min 30Fe50W50Ac catalysts showed ∼ 19% CH₄ conversion, ∼11% H₂ yield and 0.56 $Y_{H_2}$ / $C_{C{H}_4}$ ratio. Overall, at the end of 420 min time on stream, 30Fe25W75Ac is found best. It has ∼ 14% CH₄ conversion, ∼6% H₂-yield and > 0.4 $Y_{H_2}$ / $C_{C{H}_4}$ ratio at 420 min time on stream.

TGA results indicate severe coke decomposition over 30FexW(100-x) Ac (x = 25, 50) catalyst. O₂-TPO result indicates that coke over the 30Fe25W75Ac catalyst has a higher crystallization degree than the 30Fe50W50Ac catalyst. The carbon yield calculation of the spent 30FexW(100-x) Ac (x = 5, 10, 25, 50, 90, 95) catalyst system is shown in Table S5. Here also, the carbon yield over the spent 30Fe25W75Ac catalyst is greater than the 30Fe50W50Ac catalyst. Interestingly, the catalytic activity of the 30Fe25W75Ac catalyst is less affected by severe coke deposition but the activity of 30Fe50W50Ac drops suddenly on increasing time on stream. Initially, the activity of both catalysts is close to each other but after the end of 200 min, 30Fe50W50Ac has only ∼ 16% CH₄ conversion (against 30% in 30Fe25W75Ac), 9% H₂-yield (against 29% in 30Fe25W75Ac) and 0.56 $Y_{H_2}$ / $C_{C{H}_4}$ ratio (against 0.97 in 30Fe25W75Ac). It indicates that coke decomposition affects the performance of 30Fe50W50Ac to a great extent but not the performance of 30Fe25W75Ac catalyst. It seems that over 30Fe25W75Ac, rate of CH₄ decomposition (carbon formation) is well matched with the rate of diffusion of carbon species from metal-gas interface (where decomposition of CH₄ took place) to the metal-nanofiber interface (where carbon precipitates to form carbon nanofibers). Over highly crystalline 30Fe50W50Ac catalyst (than 30Fe25W50Ac), the rate of carbon formation may not properly match the rate of carbon diffusion. So, carbon species isn’t able to be transferred away in time and would cover the catalyst’s active sites leading to catalyst deactivation (Chen et al., 1997).

Tungsten oxide incorporation of>50 wt% in 30FexW(100-x)Ac (x = 75, 90, 95) causes a fast drop of surface area due to the deposition of various crystallite inside the pore (Rahman et al., 2015) constituted by major-tungsten oxide and minor-activated carbon. These catalysts systems have also low density of reducible reducible-species over the catalyst surface. Low surface area catalyst and few catalytic active sites on the surface conveys less initial CH₄ conversion. 30Fe75W25Ac has the highest crystallinity among rest catalyst systems. 30FexW(100-x) Ac (x = 75–95) catalysts has low initial CH₄ conversion (14–36%) and initial H₂-yield (13–29%).