Clean Hydrogen Production via Methane Cracking Over Ni Supported on Ceria-silica Catalysts

Introduction

The non-catalytic breakdown of methane requires very high temperatures > 1000 °C to occur efficiently to obtain pure hydrogen without CO and CO₂ along with carbon. Such a carbon may be utilized as bulk amorphous activated carbon for various industrial applications ¹. Owing to the limitation of high temperature methane cracking, catalysts have been used during the previous few years to facilitate methane decomposition. Concerning the conversion of methane, generation of hydrogen and carbon, catalytic methane cracking (CMC)can be the choice as it is performed at lower temperatures between 550 to 800 °C ². For this reason, research on carbonaceous catalysts as well as metallic catalysts has been explored ³.Heterogeneous metal-based catalysts significantly facilitated the synthesis of hydrogen as well as the multi or single walled nano-filaments ⁴. These very significant carbon nanostructures are applied in energy, including hydrogen storage and the production of electrode materials for batteries, fuel cells, super capacitors, and other devices⁵. In addition to the reduction in reaction temperature for CMC process, it was found that using a suitable catalyst the ratio of the ordered and/or disordered carbon altered that could enhance the overall yield of the hydrogen⁶. A variety of factors influence the catalytic efficiency of methane decomposition catalysts. For example, metals which are functional, textural supports and or promoters, co-metals, conditions of synthesis and of catalyst preparation techniques would influence the catalyst CMC activity⁷. The most utilized active metals are Ni, Co, and Fe for the CMC process. Other d block elements were used in conjunction with promoters such as noble metals and lanthanum, magnesia, ceria, zirconia oxides etc. as supports for Ni based catalysts ⁸. Furthermore, modification of the support using a suitable metal oxide could affect the catalyst stability and performance^9,10.In the present study ceria modified SiO₂ has been examined as a support for Ni catalysts for the CMC process. Reason for the selection of ceria as a modifier for SiO₂ is that its enhanced textural and redox qualities which could facilitate the dispersion of active Ni metal catalyst. Various investigations in fixed bed reactors have shown that the variables that determine conversion of CH₄ and the carbon formed includes catalyst, the sort of support, the textural properties of the catalyst, and the conditions of operation ^11-13. This study used Ni on ceria and silica to investigate the effects of catalyst textural characteristics and support type in CCM process. Characterization of the catalysts is performed by various techniques such as Temperature programmed reduction by H₂TPR, SA by N₂ physisorption,  spectroscopy study of carbon by Raman, X-ray diffraction study by XRD, carbon, hydrogen, nitrogen, and sulfur content study by CHNS, and chemisorption by hydrogen pulse. The physicochemical characterization data deduced from the above techniques was utilized to correlate with H₂ production rates.

Experimental

Different loadings (2, 4, 6, 8 and 10wt%) of CeO₂is impregnated over fumed silica as a support and calcined in flowing air at 550 °C.  Ni(NO₃)₂.6H₂O is used as a precursor for a Ni loading of 20wt%which was impregnated on the varied loadings of CeO₂-SiO₂ support. At 550 °C, the calcination of the catalysts is performed and reduced at the same temperature in H₂ flow before being examined for the CMC reaction. In a standard procedure a known quantity of nitrate of Ni is dissolved in raw water to which a desired amount of CeO₂-SiO₂ support was added and stirred at 80 °C. Evaporation of water from the samples was at 120 °C in an oven and subsequent calcination at required temperature (550 °C).

Characterization

For conciseness and to prevent self-plagiarism, the experimental details pertaining to the BET surface area, XRD, SEM, TEM, H₂ TPR, H₂-pulse chemisorption, CHNS analysis, and Raman spectroscopy techniques are described in supplementary material.

CH₄ cracking studies

Calcined 20wt%Ni/CeO₂-SiO₂ catalyst (~ 0.1g) is loaded in the centre of a fixed bed quartz reactor. A mixer with 5%H₂ balance Ar was used for the reduction of the catalyst for 3h at 550 °C before the reaction began. Subsequently, the 5%H₂/Ar gas was replaced with 30 mL/min of CH₄ flowing over the catalyst at 550 °C ¹⁴. N₂is employed as the gas carrier and the reactor exit product stream was examined using a Shimadzu gas chromatograph (thermal conductivity detector; column-carbosphere/carboxen). Aautosampler with six ports is used to analyse the methane conversion with a time interval of 20 min. Until the hydrogen peak is negligible and the methane content found constant, the reaction is performed. Carbon oxides (CO and/or CO₂) were not found during the course of reaction emphasizing the methane cracking led to exclusive formation of H₂. Cross-checking was done on the hydrogen yields determined by measuring CH₄ conversion and H₂ production and CHNS analysis for the carbon that accumulated from the recovered catalyst. The mass balance is confirmed by the lack of compounds such as CO and/or CO₂ during the process and these results are in accordance with the recent reports on the CMC reaction¹⁵.

Results and Discussion

N₂ physisorption analysis

N₂ Physisorption results show decrease in the surface area as the percentage loading of CeO₂ increases, possibly because of the silica support pores becoming blocked. The results are tabulated below.

Table 1: Physicochemical characteristics of 20wt%Ni supported on CeO₂-SiO₂

CeO2(wt%) in SiO2	BET-SA (m2/g)a	Crystallite Size (NiO)b	H2 uptake (mmol/g)c	H2 uptake (cm3/g)d	Ni metal surface area (m2/g)e	H2 Yields (molH2/mol Ni)
2 4 6 8 10	142 139 130 128 117	18.4 18.1 18.3 17.9 18.0	3.33 3.41 3.51 3.89 5.23	0.12 0.05 0.09 0.06 0.05	2.75 2.20 2.12 2.01 1.99	2136 1163 1575 767 573

^a calculated from BET surface area; ^b XRD analysis; ^c H₂TPR; ^{d, e} H₂ pulse chemisorption

XRD analysis of the fresh and deactivated catalysts

The 20wt%Ni/CeO₂-SiO₂patterns of XRD that were calcined at 550 °C are shown in Figure 1A. The presence of NiO phase is explained from the reflections appeared at 2θ = 37.28, 43.3 and 62.9°, with1.48, 2.09 and 2.41A°’d’ values[ICDD #01-1239]. The diffraction peak at 14.5°, is corresponding to CeO₂ phase. The deactivated catalysts XRD showed the reflections of graphitic carbon and Ni only (Figure 1B).

Figure 1: XRD patterns of (A) fresh and (B) deactivated samples of 20wt%Ni over (a) 2 (b) 4 (c)6 (d) 8 (e) 10wt%CeO2-SiO2 catalysts Click here to View Figure

H₂-TPR

The metal oxides reduction behaviour is investigated using H₂-TPR (Figure 2) analysis, and Table 1 provides a list of pertinent hydrogen uptakes over 20wt%Ni/CeO₂-SiO₂ catalysts. In every sample, a broad peak was discernible due to the reduction of NiO and certain ratio of CeO₂ species. The bulk feature of the H₂-TPR technique was demonstrated by the increased H₂consumptions with increasing CeO₂ loadings due to the presence of CeO₂ species that were appeared in XRD analysis (Figure 1A). The signals Tmax is somewhat shifted towards a high temperature as CeO₂ loading is increased. This is probably due to an interaction between nickel and ceria particles. XRD analysis specifies that the size of nickel crystal is similar over these catalysts. However, having a constant loading of Ni (20wt%) a raise in H₂ uptakes are found upon increasing the CeO₂ loadings, indicating that ceria species were undergone during the TPR (Table 1). Furthermore, the shoulder peak tail extended towards higher temperatures from CeO₂ loadings 2 to 10wt% is noticed. The interaction of NiO species with the CeO₂could cause this peak shift.

Figure 2: H2 TPR patterns of 20wt%Ni supported on(a)2(b)4 (c) 6(d) 8 and (e)10wt%CeO2-SiO2 catalysts. Click here to View Figure

H₂ chemisorption

Using chemisorption by H₂, the metallic Ni area was determined. It decreased from 2.75 -1.99 m² g^-1, upon increasing the CeO₂ loadings on SiO₂as seen by the H₂ adsorption data (Table 1). The 20wt%Ni/2wt% CeO₂-SiO₂ catalyst showed a high proportion of dispersed nickel species on the surface. Low temperature reduction enhances the Ni dispersion over the  catalyst surface.

CH₄ cracking activity

Figure 3 displays the CCM activity with time operated at 550 °C on the 20wt%Ni supported on various loadings of CeO₂ on SiO₂ catalysts. All the catalysts get deactivated with time. The catalyst with lower CeO2 loading displayed better sustainability compared to other loadings. Table 1 provides the H₂ yields that were obtained during the CMC reaction. High initial conversion, followed by a deactivation within15 h, is evident from the activity data. At a higher loading of CeO2 (10wt%), the catalyst was deactivated quickly within 8 h on stream producing lower hydrogen yields. Compared to the other catalysts, 20Ni/2wt%CeO₂-SiO₂ exhibited higher activity. It has been found that there is a good correlation between the yields of H₂ and SA of nickel over these catalysts (Table 1). All the catalysts recovered after the reaction was analysed by TEM and Raman spectroscopy to understand the nature and type of carbon formed.

Figure 3: Time on stream analysis of CMC activity over 20wt%Ni supported on CeO2-SiO2 Click here to View Figure

Characterization of deactivated catalysts by TEM and Raman spectroscopy

The transmission electron microscopic images of the 20Ni/2wt%CeO₂-SiO₂ catalyst is reported in Figure 4. Figure 4A clearly shows carbon nano fibre with tip growth mechanism. With a diameter of around 40 nm CNFs were seen in TEM images (Figure 4A and 4B). The growth of the CNT is dependent on the particle size of Ni ¹⁶. Catalyst interior, with carbon strand behaviour and diffusion in a ‘V’ shape, was visible in the TEM image (Figure 4C and 4D).The similar size of carbon nanofiber and the Ni particle is observed.

Figure 4: TEM images of the deactivated 20Ni/2wt%-CeO2-SiO2 catalyst recovered after 15 h. Click here to View Figure

Raman spectroscopic results

Figure 5 reports the Raman spectra of the 20Ni/CeO₂-SiO₂ catalysts recovered after the reaction which shows both the “D” and “G” bands of carbons. Two distinct bands were shown, one at approximately 1320 cm^-1which is explained by amorphous and carbon nanoparticles or disordered carbon due to structural defects in graphite. The G- band roughly 1580 cm^-1 is by in-plane stretching of carbon vibrations by ordered structure of C ^17,18. The D band full width at half maximum (FWHM) to the ordered band I_D/I_G is correlated negatively with the graphene ¹⁹. This shows higher ordered nature similar to graphene at lower I_D/I_G values. When FWHM increased, the ratio of disordered to ordered is decreased. For carbon nanofibers, Alvarez et al. discovered a comparable relationship between full width half maxima and the ratio²⁰. The build-up of C versus the width of G-band suggests the formation of ordered carbon over 20Ni/2wt%-CeO₂-SiO₂. The ordered carbon deposition explained due to the high CMC activity of the 20Ni/2wt%-CeO₂-SiO₂ catalyst in comparison to other catalysts.

The carbon form determined with Raman spectroscopic analysis emphasizes reasons why a particular catalyst system show rate of hydrogen production. According to Scott et al., combining silica with ceria for methane dry reforming improves the performance of Ni-based catalysts²¹. Their work elaborates on the features of the catalyst that are determined by the composition and structure of the support. Yoshida and colleagues carried out non-oxidative coupling of methane over supported ceria photocatalysts²². A range of rare earth oxides supported by silica was investigated for photocatalytic methane conversion. Of these, the cerium oxide supported by silica with lower loading demonstrated the highest photo activity to generate hydrogen and ethane²². According to Wang et al., effects of size and sturdy metal support interaction contributes to good performance for methane dry reforming²³. Takriff et al., investigated the reforming of CH₄ using ceria as a support material and Pt as a promoter²⁴. In this study we found an improved rate of hydrogen on a silica that is modified with 2wt%CeO₂ as a support for Ni loading of about 2136 mole of H₂per mole of Ni. Excess addition of CeO₂to SiO₂caused the decrease in surface areas as well as the Ni metal surface areas can be a possible reason for the lower hydrogen rates while maintaining a constant weight percentage of Ni (Table 1). The decrease in surface areas could potentially be due to Ni-CeO₂ interaction. The bulk property of the catalyst is explained by the rise in H₂ uptake values with increase in CeO₂ loading as observed from the H₂-TPR data (Table 1).The interaction between Ni and CeO₂seems to be weak at a loading of 2wt%as the H₂ TPR displayed a broad single peak. However, the peak split is noted from 4-10wt% of CeO₂ loaded SiO₂ supported Ni catalysts (Figure 2). The fine metal dispersion in 2wt%CeO₂–SiO₂can be explained by the high metallic surface area compared to other CeO₂ loadings. The 2wt%CeO₂ produced high H₂ yields of 2136 mol H₂ /mol Ni at 550 °C and atmospheric pressure, wherein excess addition of CeO₂ with 4wt% and higher leads to low H₂ yields of about 573 moles of H₂ per mole of Ni over 20wt%Ni/10wt%CeO₂-SiO₂ catalyst. The 20wt%Ni/10wt%CeO₂-SiO₂ deactivated catalysts displayed more carbon which is graphitic with a lower I_D/I_G ratio.

Figure 5: Raman spectra of the 20wt%Ni supported on (a) 2 (b) 4 (c) 6 (d) 8 and (e)10wt% CeO2-SiO2 catalysts. Click here to View Figure

Conclusions

In summary, the Ni supported on CeO₂ modified SiO₂ catalyst was found to be an active catalyst for CH₄ cracking at moderate temperatures to produce pure hydrogen and carbon nano fibers. Yields of H₂ agree with SA of nickel of the corresponding CeO₂ loaded SiO₂ supported Ni catalysts. The 20Ni/2wt%CeO₂-SiO₂ catalyst exhibited higher methane cracking performance in comparison to other catalysts. The spent catalyst is characterized by XRD and TEM analysis which showed filamentous carbon. The crystalline and amorphous carbon were manifested by Raman spectra. The higher hydrogen yields obtained over 20Ni/2wt%CeO₂-SiO₂was explained due to high nickel SA and formation of well-ordered carbon.

Acknowledgment

The authors acknowledge the Department of Chemistry, Osmania University, Hyderabad and Telangana University, Nizamabad for providing research facilities.

Conflict of Interest

There is no conflict of interest

Funding Sources

The authors acknowledge the financial support from TSCOST, reference Lr. No. 03/TSCOST/DST-PRG/2021-22, Dated: 21-03-2022.

References

1. Soo Kim, I. K.; Eun, S. C.; Sang, W. K.; Seok Lee, C.;  Junyoung, K.; EunByun, K.; Unyong, J.Adv.Mater, 2022, 35, 43.
2. Yousefi, M.; Donne, S.Int. J. Hydrogen Energy, 2022, 47, 699-727.
   CrossRef
3. Torres, D.; Pinilla, J.L.; Lazaro, M. J.; Moliner, R.; Suelves, I.  Int. J. Hydrogen Energy, 2014, 39, 3698–3709.
   CrossRef
4. Suelves, I.;Pinilla, J. L.;Lázaro, M. J.;Moliner,R.; Palacios, J. M. J. Power Sources, 2009, 192, 35–42. 
   CrossRef
5. Abbas,H. F.;.Daud, W. M. A.Int. J. Hydrogen Energy, 2010, 35, 1160–1190.
   CrossRef
6. Kalpana, M.;Manasa, K.;Padmasri, H. A.;Pratyay, B.;Akella, V. S.;Venugopal, A.EP & SE, 2023, 14, 41-49.
7. Li,Y.; Li,D.;Wang,G. Catal. Today, 2011, 162, 1–48. 
   CrossRef
8. Ashik, U. P. M.;Daud,W. M. A. W.;Hayashi, J. Renewable Sustainable Energy Rev.2017, 76, 743–767.
   CrossRef
9. Pinilla, J.L.;Moliner, R.; Suelves, I.;Lazaro, M.J.;Echegoyen,Y.; Palacios, J.M. Int. J. Hydrogen Energy,2007, 32, 4821-4829.
   CrossRef
10. Takenaka, S.;Ogihara, H.; Yamanaka,I.;Otsuka,K.Appl. Catal. A, 2001,217, 101-10.
    CrossRef
11. Borges R.P.; MouraL.G.; Kanitkar S.; Spivey J.J.; Noronha F.B.; Hori C.E.Cat. Today, 2021, 381, 3-12.
    CrossRef
12. Spiess, F. J.;Suib, S. L.;Irie, K.; Hayashi,Y.; Matsumoto, H.Catal. Today, 2004,89, 35-45.
    CrossRef
13. Li, Y.; Zhang, B.; Xie, X.; Liu, J.; Xu,Y.;Shen, W.J. Catal., 2006, 238, 412-24.
    CrossRef
14. Ashok, J.;Naveen Kumar, S.;Venugopal, A.;DurgaKumari,V.; Subrahmanyam, M. J. Power Sources, 2007, 164, 809-14.
    CrossRef
15. Wei,Y.;Kent, H.S. Environ. Prog. & Sustain. 2014, 33, 213-219.
    CrossRef
16. Manasa, K.;Naresh, G.; Kalpana, M.; Sasikumar, B.;Vijay Kumar, V.;Chary, K.V.R.;Michalkiewicz,B.;Venugopal. A.J. Energy Inst.2021, 99, 73-81.
    CrossRef
17. Naresh, G.;Vijay Kumar, V.;Anjaneyulu, C.; Tardio, J.; Bhargava, S. K.;Patel,J.;Venugopal, A. Int. J. Hydrogen Energy, 2016, 41, 19855-62.
    CrossRef
18. Dussault, L.; Dupin, J. C.;Guimon, C.;Monthioux, M.;Latorre, N.;Ubieto, T.; Romeo, E.;Royo,C.; Monzon, A.J. Catal, 2007, 251, 223.
    CrossRef
19. Li, Y.; Zhang, B.C.;Xie, X.W.;Liu, J.L.;Xu,Y.D.;Shen, W.J. J.Catal, 2006, 238, 412.
    CrossRef
20. Darmstadt, H.; Summchen, L.; Ting, J.M.;Roland, U.;Kaliaguine,S.; Roy,C.Carbon, 1997, 35, 1581.
    CrossRef
21. Alvarez, W.E.;Pompeo, F.; Herrera, J. E.;Balzano,L.; Resasco, D.E. Chem Mater, 2002, 14, 1853.
    CrossRef
22. Lovell, E. C.;Horlyck, J.; Scott,J.;Amal, R.Applied Catal. A: General.2017, 547,47-57.
    CrossRef
23. Yuliati, L.;Hamajima,T.;Hattori, T.;  Yoshida, H.J. Phys. Chem. C , 2008, 112, 18, 7223–7232.
    CrossRef
24. Han, K.; Xu, S.;Wang, Y.;Wang, S.;Zhao, L.J. Power Sources, 2021, 506, 230-232.
    CrossRef
25. Pudukudy, M.; Yaakob, Z.; Jia, Q.; Takriff, M. S. New J. Chem.2018, 42, 14843-14856.
    CrossRef

---

This work is licensed under a Creative Commons Attribution 4.0 International License.