- Email: [email protected]
Low temperature dry reforming of methane over Pd-CeO2 nanocatalyst
Catalysis Communications 92 (2017) 19–22
Contents lists available at ScienceDirect
Catalysis Communications journal homepage: www.elsevier.com/locate/catcom
Short communication
Low temperature dry reforming of methane over Pd-CeO2 nanocatalyst Rajib Kumar Singha, Aditya Yadav, Astha Shukla, Manoj Kumar, Rajaram Bal ⁎ Nanocatalysis Area, Refining Technology Division, CSIR-Indian Institute of Petroleum, Dehradun 248005, Uttarakhand, India
a r t i c l e
i n f o
Article history: Received 3 October 2016 Received in revised form 14 December 2016 Accepted 19 December 2016 Available online 21 December 2016 Keywords: Pd-nanoparticles CeO2 Dry reforming Methane
a b s t r a c t Pd-CeO2 nanocrystals were prepared using a simple surfactant induced method, which initiated dry reforming of methane reaction only at 350 °C. Low temperature activity of the prepared catalyst was due to high Pd-dispersion and very good redox property of support Ceria (20–50 nm) nanocrystals (32.8% Ce+3 concentrations). The catalyst was found to be very good coke resisting and inhibited deactivation by coke formation during catalysis. Time on stream (TOS) study was carried out for 12 h and slight decrease in conversion was observed due to sintering, re-oxidation and hydroxylation of the active Pd-nanoparticles. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Dry reforming of methane (DRM) or methane reforming with CO2 is an extensively studied methane reforming reaction to produce much valuable synthesis gas by utilizing two major greenhouse gasses (CH4 and CO2) [1–5]. DRM is a highly endothermic reaction and is one of the better processes to utilize methane and carbon dioxide for synthesis gas production but the reaction requires very high temperature [6–10]. Decades of studies revealed that both noble metals (Rh, Ru, Pt, Pd) and non-noble metals like Co and Ni are highly active for DRM but only Ni is comparable with the noble metals in terms of reactivity [3–7,11,12]. Problem with DRM is intense coking during catalysis due to methane decomposition and Boudouard reaction [1–10]. Non-noble metals are highly coke sensitive but noble metals are reported to be quite good coke resisting elements [2,6,11,12]. So, from the reactivity and stability point of view, noble metal catalysts are better for catalytic DRM reaction. Now, deactivation of DRM catalysts can also occur by re-oxidation or hydroxylation of the active species of the reaction or by sintering of catalyst particles, which is mainly caused by low metal-support interactions between active metal particles and support particles [3–5]. Another problem of DRM is reverse water gas shift (RWGS) reaction, which highly affect H2/CO ratio by consuming the produced hydrogen (CO2 + H2 → CO + H2O) [1–10,13]. For the advantage of utilizing two greenhouse gasses and production of highly valuable syngas, which is the basic raw material for FischerTropsch synthesis for the production of various synthetic fuels [2,3,7], researcher are still involved in making the process much more effective. Researchers are still developing different noble and non-noble metal ⁎ Corresponding author. E-mail address: [email protected] (R. Bal).
http://dx.doi.org/10.1016/j.catcom.2016.12.019 1566-7367/© 2016 Elsevier B.V. All rights reserved.
catalysts for the purpose. It has been found that catalysts in nano-regime show very high activity for different catalytic processes due to its unusual and advantageous behavior [3,4,14–17]. Low metal loading facilitates high chemical potential of the small Pd particles that are bound to Ceria. CeO2 itself is reducible support and provides active surface oxygen during oxidation reduction reaction but addition of small amount of active metal like Pd highly enhances oxygen vacancy [18–20]. Farmer and Campbell reported that smaller active particles supported on partially reduced CeO2 has much higher stability than its larger counterparts [18]. Smaller particles have more interactions with the support as these particles provide much more interfacial surface area than its larger counterparts [3,4,14–23]. Cargnello et al. reported that enhanced metal-support interaction between very small Pd and CeO2 nanoparticles led to exceptionally high methane conversion even below 400 °C [20]. Liu et al. reported that strong metal-support interactions and created oxygen vacancies in the catalyst can enhance the reactant activation during DRM process even at room temperature (at 27 °C) [24]. Here, we are reporting a Pd-CeO2 catalyst, active for DRM even at low temperature as 350 °C, which is theoretically possible [2,25]. 2. Experimental 2.1. Synthesis of Pd-CeO2 nanocrystals In a typical synthesis procedure of Pd-CeO2 catalyst, 1 g cetyltrimethylammonium bromide (CTAB) was dissolved in 100 ml distilled water by vigorous stirring for 30 min. After a homogeneous CTAB solution was formed, 6.5 g of cerium chloride heptahydrate (CeCl3·7H2O) was added to it. The whole mixture was stirred for 2 h to form a homogeneous solution. After that, 0.05 g palladium chloride (PdCl2) dissolved in 20 ml acetonitrile was added drop-wise to the
20
R.K. Singha et al. / Catalysis Communications 92 (2017) 19–22
cerium salt solution. The whole solution was kept on stirring at 50 °C for 30 min. to mix the precursors well (homogeneous solution). 25% NH3 was then used to precipitate the salts maintaining pH at ~ 9. After 1 h of continuous stirring, 10 μl borane triethylammine complex was added to the mixture (Supporting information). The whole mixture was then stirred for 30 min and aged for 4 days (~96 h). After filtering and washing the precipitate with excess water to remove the chloride ions, it was dried at 100 °C for 6 h and then calcined at 600 °C for 6 h in air. The prepared catalyst was denoted as 0.86Pd-CeO2, where 0.86 was the wt% of Pd in the catalyst obtained from ICP-AES analysis.
1–15 nm of Pd-species deposited on the surface of CeO2. Pd-dispersion analysis showed much lower size of Pd-nanoparticles. The high value of Pd-size measured from TEM analysis was due to the fact that the technique was not capable of detecting ultrasmall particles, which also contributed during H2-chemisorption analysis. Fig. 2 (inset) shows the fringes of PdO nanocrystal. The d-spacing of 1.66 Å and 2.63 Å was attributed to (112) and (101) plane of PdO. We also carried out TEM analysis for the spent catalyst (Fig. S4), which showed sintered catalyst particles. TGA/DTG analysis of uncalcined and spent catalyst was performed to check thermal stability and carbon deposition on the catalyst surface (Figs. S5, S6).
2.2. Characterization and activity measurements 3.2. Catalytic performance Provided in Supporting information of the manuscript (Figure and Table numbers are designated as S1, S2 etc.). 3. Results and discussions 3.1. Characterizations Surface area analysis of prepared 0.86Pd-CeO2 catalyst was carried out, which showed 63 m2/g surface area of it, whereas surface area of the spent catalyst was found to be 38 m2/g. The observed result could be due to sintering of catalyst particles [4,15,16]. H2-chemisorption analysis was carried out for determination of Pd dispersion in the catalyst. The analysis showed 33.9% Pd dispersion and about 2.9 nm average Pd-particles size. XRD analysis of the fresh catalyst showed only characteristics peaks for cubic CeO2 but no peaks for any Pd-species was observed. The observation indicated about good dispersion of Pd-species but the absence of any Pd-species peaks can also be due to very low loading (0.86 wt%). CeO2 particles size of fresh catalyst calculated from XRD using Scherrer equation was found to be 37.1 nm. XRD pattern of the spent catalyst showed much more intense CeO2 peaks but still there was no peak for Pd-species observed. Increase in XRD peak intensity indicated increase in crystallinity of CeO2 particles [4] but absence of any Pd-species peak indicated that small amount of Pdspecies was not detectable as XRD is a bulk sensitive technique [4,15]. Absence of peak for graphitic carbon also revealed an important feature of the catalyst i.e. coke resistibility. Surface analysis to determine the Pd species present in the catalyst was carried out by XPS technique. From the binding energy (B.E.) values, the analysis revealed that fresh PdCeO2 catalyst contain both metallic Pd (B.E. 334.8 eV) and Pd-oxide (PdO) (B.E. 336.7 eV) species. Presence of metallic Pd could be due to presence of CTAB during calcination (Supporting Information) [15]. XPS analysis of the spent catalyst showed metallic Pd (B.E. 334.9 eV), PdO (B.E. 336.6) and PdO2 and/or Pd(OH)4 (B.E. 337.5 eV) species. Presence of Pd(OH)4 species was due to hydroxylation of the active Pd particles by produced water during catalysis [4]. To be sure about the hydroxide species in the spent catalyst we also analyzed the O 1S XPS pattern. The peak at B.E. value 531.4 eV confirmed the presence of hydroxide species in the spent catalyst [4]. XPS survey and Ce 3d core level spectra is also provided in Supporting information (Fig. S1). We have calculated Ce+3 concentration form XPS analysis [26,27], which was found to be 32.8% (Fig. S2). H2-TPR profile of 0.86Pd-CeO2 catalyst showed 4 peaks in the temperature range 50–800 °C. First peak at 110 °C was due to reduction of easily reducible very small Pd-oxide nanoparticles, second peak at 140–280 °C was due to little bit larger Pd-oxide particles, third peak (300–500 °C) was due to reduction of surface oxygen of support CeO2 [15]. Normally the peak shows at about ~500 °C but the presence of Pd-species on the CeO2 surface makes the oxygen at the metal-support interface much easily reducible [16]. The last peak at above 600 °C was due to bulk reduction of CeO2 [15] (see Fig. 1). To analyze the morphology of the prepared catalyst we have carried out both SEM (Fig. S3) and TEM analysis (Fig. 2) of the fresh catalyst catalyst, which revealed that CeO2 particles were about 20–50 nm size and
Catalytic activity of the prepared catalyst was tested in the temperature range 300–800 °C at optimized weight hourly space velocity (WHSV) value 70,000 ml·g−1·h−1. Observed results for heat transfer and mass transfer (HT-MT) limitations experiments showed negligible HT-MT limitations (Figs. S7, S8). Carbon and material balance for the experiments was in between 98 and 102% (or ± 2%). Thermodynamic analysis says, DRM is highly endothermic reaction and requires very high temperature (800–1000 °C) to obtain high conversions as both the molecules are highly stable at normal condition [2,25,28]. CH4 þ CO2 →2CO þ 2H2 ΔH298K ¼ 247 kJ=mol With increasing pressure, reactant conversions and H2/CO yields drops drastically. So, according to thermodynamics, it is essential to operate DRM at lower pressure [2,25,28]. At these conditions carbon deposition is inevitable [2,25,28]. These problems can be overcome by operating the reaction at lower temperatures and with the catalysts which are resistant towards coking during DRM [2,25,28]. Rate of reaction over 0.86Pd-CeO2 catalyst was also calculated and reported (Table S1). Fig. 3a shows reactant conversions and H2/CO ratios at different temperature. It was found that experimental conversions values are lower than equilibrium conversion values (Fig. S9). Fig. 3b shows TOS stability of the catalyst at 800 °C. It can be seen from Fig. 3a that the reaction was initiated at 350 °C with methane and CO2 conversions 3.56% and 1.96% respectively and very slow increase in conversions were observed up to 500 °C. In our recent articles, we have showed that use of CeO2 as support or promoter and strong metal-support interactions activates methane at lower temperatures [3,4,14–17]. Initiation of DRM reaction at such a low temperature was the main advantage of the catalyst (Table S2) and it could be due to high activity of highly dispersed Pd-nanoparticles and high metal (Pd)-support (CeO2) interactions [3,4,14–16]. High concentration (32.8%) of Ce+3 in 0.86Pd-CeO2 was highly influential for activating methane at low temperature [2,5,8,12,15]. Deposition of Pd particles on the surface of CeO2 created oxygen vacancies and these oxygen vacancies highly improved the anchoring of Pd-nanoparticles and increased Pd-dispersion [28]. Methane dissociates over Pdnanoparticles and produced carbon and hydrogen. The produced carbon is then reduced the oxygen species of CeO2 adjacent to the Pd-nanoparticles producing CO [29]. Produced carbon can also take part in CO formation by dissociative adsorption of CO2, thereby removing the deposited carbon and initiating the DRM reaction. Very slow increase in reactant conversions up to 500 °C could be due to the inherent endothermic nature of DRM reaction. It could also be a reason that methane can be dissociated over Pd-oxide nanoparticles in an oxidizing environment created by CeO2 by its oxygen storage property but CO2 requires metallic Pd particles to dissociate and in the oxidizing environment its conversion was low at lower temperatures [3,4]. With increasing temperature and production of hydrogen by methane dissociation led to a reducing environment, which facilitated CO2 dissociation and increase in CO2 conversions was observed. After 550 °C, sharp increase in reactant conversions were observed may be due to favorable
R.K. Singha et al. / Catalysis Communications 92 (2017) 19–22
21
Fig. 1. (A) XRD pattern of (a) fresh and (b) spent catalysts, (B) XPS pattern of (a) fresh and (b) spent catalysts, (C) O 1S XPS of spent catalyst and (D) H2-TPR pattern of fresh 0.86PD-CeO2 catalyst. Spent catalyst condition: Temperature (800 °C), pressure (1 atm), WHSV (70,000 ml·g−1·h−1), reaction time (12 h).
thermodynamic conditions [2,3]. H2/CO ratio over the prepared 0.86PdCeO2 catalyst was found to be decreasing with temperature as with increase in temperature CO2 conversions were found to be increasing may be due to increasing dominance of RWGS reaction [2–4]. At the temperature range between 350 and 550 °C, higher H2/CO ratio was due to high methane conversion, low CO2 conversion and the absence of RWGS reaction. The results also indicates that there were less CO production by CO2 decomposition (CO2 + C → 2CO) and/or removal of deposited carbon (C + O → CO) by surface oxygen of CeO2. This observation led to the
conclusions that after 600 °C RWGS reactions was dominating and controlled the H2/CO ratio by consuming produced hydrogen and producing CO and H2O [13]. Now, production of H2O can lead to hydroxylation of active Pd-particles as it was also evidenced from the XPS analysis of spent 0.86Pd-CeO2 catalyst. Time on stream (TOS) stability of the catalyst was tested at 800 °C (Fig. 3b). It was found that initially methane and CO2 conversions were above 90% but after 9 h, slight decrease in reactant conversion was observed. Catalyst stability during DRM gets affected mainly by coking but also depends on sintering of catalyst particles, re-oxidation and hydroxylation of active species [3–5]. Sintering of the catalysts particles were observed from XRD and TEM analysis. In this case, catalyst deactivation was found not to be due to coking but due to sintering of catalyst particles, re-oxidation and hydroxylation of active Pd-species. Coking mainly occurs due to methane decomposition and Boudouard reaction [2–5], whereas re-oxidation of the active species occurs by oxidants like CO2, or surface oxygen of the catalyst [4,5]. Re-oxidation of active particles is quite possible in case of catalysts with high OSC [4, 5]. Hydroxylation of the active Pd-species could have occurred by H2O molecules, produced during DRM by RWGS reaction [4]. XPS analysis of the spent 0.86Pd-CeO2 catalyst confirmed the presence of PdO2 and Pd(OH)4 in the catalyst. As Oyama et al. reported that RWGS reaction is the main problem of DRM for synthesis gas production [13]; we have also experienced the same in this case. A comparative analysis was also carried out for catalytic activity and TOS stability of impregnated catalyst (1Pd-CeOImp 2 ) and 0.86Pd-CeO2 catalyst (Table S3, Figs. S10, S11). 4. Conclusions
Fig. 2. TEM images of fresh 0.86Pd-CeO2 catalysts.
Low temperature active Pd-CeO2 catalyst for dry reforming of methane was prepared by single step surfactant induced method. The main feature and advantage of the prepared catalyst was its low temperature activity for dry reforming of methane. The catalyst showed syngas
22
R.K. Singha et al. / Catalysis Communications 92 (2017) 19–22
Fig. 3. (a) Methane, CO2 conversions and H2/CO ratios, (b) time on stream (TOS) stability at 800 °C over 0.86Pd-CeO2 catalyst. Reaction condition: Pressure (1 atm), WHSV (70,000 ml·g−1·h−1), feed ratio (CH4:CO2:He = 1:1:12).
formation at about 350 °C due to presence of highly dispersed and very small Pd-nanoparticles (~2.9 nm). High metal-support interactions, and redox property of support CeO2 was the main driving force for the low temperature activity of the catalyst. About 32.8% Ce+3 concentrations (high reducibility) in 0.86Pd-CeO2 catalyst influenced methane activation at as low as 350 °C. Deactivation of the catalyst was found to be due to re-oxidation and hydroxylation of the active Pd-species and sintering of catalyst particles but not due to coking. RWGS reaction was the main problem with this catalyst, which highly affected H2/CO ratio by consuming produced hydrogen. Acknowledgements R.S. thanks University Grants Commission (UGC) , New Delhi, India for their fellowships. R.B. thanks CSIR for funding in the form of Network Project (CSC-0125, CSC-0117), DST Government of India for funding in the form of AISRF Grand Challenge Project (DST/INT/Aus/GCP-4/ 13(C)), Nanomission Scheme (SR/NM/NS-1105/2015). The Director, CSIR-IIP, is acknowledged for his help and encouragement. The Bilateral Collaboration Program supported by Italian CNR and Indian CSIR is kindly acknowledged. The authors thank the Analytical Science Division, Indian Institute of Petroleum for analytical services. Appendix A. Supplementary data Details of catalyst characterizations and experimental data. Supplementary data associated with this article can be found in the online version, at http://dx.doi.org/10.1016/j.catcom.2016.12.019. References [1] R. Bouarab, O. Cherifi, A. Auroux, Green Chem. 5 (2003) 209–212. [2] D. Pakhare, J. Spivey, Chem. Soc. Rev. 43 (2014) 7813–7837. [3] R.K. Singha, A. Yadav, A. Agrawal, A. Shukla, S. Adak, T. Sasaki, R. Bal, Appl. Catal. B Environ. 191 (2016) 165–178. [4] R.K. Singha, A. Yadav, A. Shukla, Z. Iqbal, C. Pendem, K. Sivakumar, R. Bal, ChemistrySelect 1 (2016) 3075–3085.
[5] S. Das, S. Thakur, A. Bag, M.S. Gupta, P. Mondal, A. Bordoloi, J. Catal. 330 (2015) 46–60. [6] D. Pakhare, V. Schwartz, V. Abdelsayed, D. Haynes, D. Shekhawat, J. Poston, J. Spivey, J. Catal. 316 (2014) 78–92. [7] A.I. Paksoy, B.S. Caglayan, A.E. Aksoylu, Appl. Catal. B Environ. 168 (2015) 164–174. [8] M. Yu, Y.A. Zhu, Y. Lu, G. Tong, K. Zhu, X. Zhou, Appl. Catal. B Environ. 165 (2015) 43–56. [9] A. Albarazi, M.E. Gálvez, P. Da Costa, Catal. Commun. 59 (2015) 108–112. [10] L. Xu, H. Song, L. Chou, ACS Catal. 2 (2012) 1331–1342. [11] L.C.P.F. Júnior, S. de Miguel, J.L.G. Fierro, M. do Carmo Rangel, Natural Gas Conversion VIII, 2007 499–504. [12] S.M. Stagg-Williams, F.B. Noronha, G. Fendley, D.E. Resasco, J. Catal. 194 (2000) 240–249. [13] S.T. Oyama, P. Hacarlioglu, Y. Gub, D. Lee, Int. J. Hydrog. Energy 37 (2012) 10444–10450. [14] R.K. Singha, A. Shukla, A. Yadav, L.N.S. Konathala, R. Bal, Appl. Catal. B Environ. 202 (2017) 473–488. [15] R.K. Singha, S. Ghosh, S.S. Acharyya, A. Yadav, A. Shukla, T. Sasaki, A.M. Venezia, C. Pendem, R. Bal, Catal, Sci. Technol. 6 (2016) 4601–4615. [16] R.K. Singha, A. Shukla, A. Yadav, S. Adak, Z. Iqbal, N. Siddiqui, R. Bal, Appl. Energy 178 (2016) 110–125. [17] R.K. Singha, S. Das, M. Pandey, S. Kumar, R. Bal, A. Bordoloi, Catal, Sci. Technol. 6 (2016) 7122–7136. [18] J.A. Farmer, C.T. Campbell, Science 329 (2010) 933–936. [19] S.D. Senanayake, J. Zhou, A.P. Baddorf, D.R. Mullins, Surf. Sci. 601 (2007) 3215–3223. [20] M. Cargnello, J.J. Delgado Jaén, J.C. Hernández Garrido, K. Bakhmutsky, T. Montini, J.J. Calvino Gámez, R.J. Gorte, P. Fornasiero, Science 337 (2012) 713–717. [21] P.G. Lustemberg, P.J. Ramírez, Z. Liu, R.A. Gutiérrez, D.G. Grinter, J. Carrasco, S.D. Senanayake, J.A. Rodriguez, M.V. Ganduglia-Pirovano, ACS Catal. 6 (2016) 8184–8191. [22] Z. Zuo, P.J. Ramírez, S.D. Senanayake, P. Liu, J.A. Rodriguez, J. Am. Chem. Soc. 138 (2016) 13810–13813. [23] G. Pantaleo, V. La Parola, F. Deganello, R.K. Singha, R. Bal, A.M. Venezia, Appl. Catal. B 189 (2016) 233–241. [24] Z. Liu, D.C. Grinter, P.G. Lustemberg, T.D. Nguyen-Phan, Y. Zhou, S. Luo, I. Waluyo, E.J. Crumlin, D.J. Stacchiola, J. Zhou, J. Carrasco, H.F. Busnengo, M.V. Ganduglia-Pirovano, S.D. Senanayake, J.A. Rodriguez, Angew. Chem. 128 (2016) 7581–7585. [25] M.K. Nikoo, N.A.S. Amin, Fuel Process. Technol. 92 (2011) 678–691. [26] A. Gupta, S. Das, C.J. Neal, S. Seal, J. Mater. Chem. B 4 (2016) 3195–3202. [27] P. Pal, R.K. Singha, A. Saha, R. Bal, A.B. Panda, J. Phys. Chem. C 119 (2015) 13610–13618. [28] X. Du, D. Zhang, L. Shi, R. Gao, J. Zhang, J. Phys. Chem. C 116 (2012) 10009–10016. [29] A. Luengnaruemitchai, S. Osuwan, E. Gulari, Catal. Commun. 4 (2003) 215–222.
Contents lists available at ScienceDirect
Catalysis Communications journal homepage: www.elsevier.com/locate/catcom
Short communication
Low temperature dry reforming of methane over Pd-CeO2 nanocatalyst Rajib Kumar Singha, Aditya Yadav, Astha Shukla, Manoj Kumar, Rajaram Bal ⁎ Nanocatalysis Area, Refining Technology Division, CSIR-Indian Institute of Petroleum, Dehradun 248005, Uttarakhand, India
a r t i c l e
i n f o
Article history: Received 3 October 2016 Received in revised form 14 December 2016 Accepted 19 December 2016 Available online 21 December 2016 Keywords: Pd-nanoparticles CeO2 Dry reforming Methane
a b s t r a c t Pd-CeO2 nanocrystals were prepared using a simple surfactant induced method, which initiated dry reforming of methane reaction only at 350 °C. Low temperature activity of the prepared catalyst was due to high Pd-dispersion and very good redox property of support Ceria (20–50 nm) nanocrystals (32.8% Ce+3 concentrations). The catalyst was found to be very good coke resisting and inhibited deactivation by coke formation during catalysis. Time on stream (TOS) study was carried out for 12 h and slight decrease in conversion was observed due to sintering, re-oxidation and hydroxylation of the active Pd-nanoparticles. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Dry reforming of methane (DRM) or methane reforming with CO2 is an extensively studied methane reforming reaction to produce much valuable synthesis gas by utilizing two major greenhouse gasses (CH4 and CO2) [1–5]. DRM is a highly endothermic reaction and is one of the better processes to utilize methane and carbon dioxide for synthesis gas production but the reaction requires very high temperature [6–10]. Decades of studies revealed that both noble metals (Rh, Ru, Pt, Pd) and non-noble metals like Co and Ni are highly active for DRM but only Ni is comparable with the noble metals in terms of reactivity [3–7,11,12]. Problem with DRM is intense coking during catalysis due to methane decomposition and Boudouard reaction [1–10]. Non-noble metals are highly coke sensitive but noble metals are reported to be quite good coke resisting elements [2,6,11,12]. So, from the reactivity and stability point of view, noble metal catalysts are better for catalytic DRM reaction. Now, deactivation of DRM catalysts can also occur by re-oxidation or hydroxylation of the active species of the reaction or by sintering of catalyst particles, which is mainly caused by low metal-support interactions between active metal particles and support particles [3–5]. Another problem of DRM is reverse water gas shift (RWGS) reaction, which highly affect H2/CO ratio by consuming the produced hydrogen (CO2 + H2 → CO + H2O) [1–10,13]. For the advantage of utilizing two greenhouse gasses and production of highly valuable syngas, which is the basic raw material for FischerTropsch synthesis for the production of various synthetic fuels [2,3,7], researcher are still involved in making the process much more effective. Researchers are still developing different noble and non-noble metal ⁎ Corresponding author. E-mail address: [email protected] (R. Bal).
http://dx.doi.org/10.1016/j.catcom.2016.12.019 1566-7367/© 2016 Elsevier B.V. All rights reserved.
catalysts for the purpose. It has been found that catalysts in nano-regime show very high activity for different catalytic processes due to its unusual and advantageous behavior [3,4,14–17]. Low metal loading facilitates high chemical potential of the small Pd particles that are bound to Ceria. CeO2 itself is reducible support and provides active surface oxygen during oxidation reduction reaction but addition of small amount of active metal like Pd highly enhances oxygen vacancy [18–20]. Farmer and Campbell reported that smaller active particles supported on partially reduced CeO2 has much higher stability than its larger counterparts [18]. Smaller particles have more interactions with the support as these particles provide much more interfacial surface area than its larger counterparts [3,4,14–23]. Cargnello et al. reported that enhanced metal-support interaction between very small Pd and CeO2 nanoparticles led to exceptionally high methane conversion even below 400 °C [20]. Liu et al. reported that strong metal-support interactions and created oxygen vacancies in the catalyst can enhance the reactant activation during DRM process even at room temperature (at 27 °C) [24]. Here, we are reporting a Pd-CeO2 catalyst, active for DRM even at low temperature as 350 °C, which is theoretically possible [2,25]. 2. Experimental 2.1. Synthesis of Pd-CeO2 nanocrystals In a typical synthesis procedure of Pd-CeO2 catalyst, 1 g cetyltrimethylammonium bromide (CTAB) was dissolved in 100 ml distilled water by vigorous stirring for 30 min. After a homogeneous CTAB solution was formed, 6.5 g of cerium chloride heptahydrate (CeCl3·7H2O) was added to it. The whole mixture was stirred for 2 h to form a homogeneous solution. After that, 0.05 g palladium chloride (PdCl2) dissolved in 20 ml acetonitrile was added drop-wise to the
20
R.K. Singha et al. / Catalysis Communications 92 (2017) 19–22
cerium salt solution. The whole solution was kept on stirring at 50 °C for 30 min. to mix the precursors well (homogeneous solution). 25% NH3 was then used to precipitate the salts maintaining pH at ~ 9. After 1 h of continuous stirring, 10 μl borane triethylammine complex was added to the mixture (Supporting information). The whole mixture was then stirred for 30 min and aged for 4 days (~96 h). After filtering and washing the precipitate with excess water to remove the chloride ions, it was dried at 100 °C for 6 h and then calcined at 600 °C for 6 h in air. The prepared catalyst was denoted as 0.86Pd-CeO2, where 0.86 was the wt% of Pd in the catalyst obtained from ICP-AES analysis.
1–15 nm of Pd-species deposited on the surface of CeO2. Pd-dispersion analysis showed much lower size of Pd-nanoparticles. The high value of Pd-size measured from TEM analysis was due to the fact that the technique was not capable of detecting ultrasmall particles, which also contributed during H2-chemisorption analysis. Fig. 2 (inset) shows the fringes of PdO nanocrystal. The d-spacing of 1.66 Å and 2.63 Å was attributed to (112) and (101) plane of PdO. We also carried out TEM analysis for the spent catalyst (Fig. S4), which showed sintered catalyst particles. TGA/DTG analysis of uncalcined and spent catalyst was performed to check thermal stability and carbon deposition on the catalyst surface (Figs. S5, S6).
2.2. Characterization and activity measurements 3.2. Catalytic performance Provided in Supporting information of the manuscript (Figure and Table numbers are designated as S1, S2 etc.). 3. Results and discussions 3.1. Characterizations Surface area analysis of prepared 0.86Pd-CeO2 catalyst was carried out, which showed 63 m2/g surface area of it, whereas surface area of the spent catalyst was found to be 38 m2/g. The observed result could be due to sintering of catalyst particles [4,15,16]. H2-chemisorption analysis was carried out for determination of Pd dispersion in the catalyst. The analysis showed 33.9% Pd dispersion and about 2.9 nm average Pd-particles size. XRD analysis of the fresh catalyst showed only characteristics peaks for cubic CeO2 but no peaks for any Pd-species was observed. The observation indicated about good dispersion of Pd-species but the absence of any Pd-species peaks can also be due to very low loading (0.86 wt%). CeO2 particles size of fresh catalyst calculated from XRD using Scherrer equation was found to be 37.1 nm. XRD pattern of the spent catalyst showed much more intense CeO2 peaks but still there was no peak for Pd-species observed. Increase in XRD peak intensity indicated increase in crystallinity of CeO2 particles [4] but absence of any Pd-species peak indicated that small amount of Pdspecies was not detectable as XRD is a bulk sensitive technique [4,15]. Absence of peak for graphitic carbon also revealed an important feature of the catalyst i.e. coke resistibility. Surface analysis to determine the Pd species present in the catalyst was carried out by XPS technique. From the binding energy (B.E.) values, the analysis revealed that fresh PdCeO2 catalyst contain both metallic Pd (B.E. 334.8 eV) and Pd-oxide (PdO) (B.E. 336.7 eV) species. Presence of metallic Pd could be due to presence of CTAB during calcination (Supporting Information) [15]. XPS analysis of the spent catalyst showed metallic Pd (B.E. 334.9 eV), PdO (B.E. 336.6) and PdO2 and/or Pd(OH)4 (B.E. 337.5 eV) species. Presence of Pd(OH)4 species was due to hydroxylation of the active Pd particles by produced water during catalysis [4]. To be sure about the hydroxide species in the spent catalyst we also analyzed the O 1S XPS pattern. The peak at B.E. value 531.4 eV confirmed the presence of hydroxide species in the spent catalyst [4]. XPS survey and Ce 3d core level spectra is also provided in Supporting information (Fig. S1). We have calculated Ce+3 concentration form XPS analysis [26,27], which was found to be 32.8% (Fig. S2). H2-TPR profile of 0.86Pd-CeO2 catalyst showed 4 peaks in the temperature range 50–800 °C. First peak at 110 °C was due to reduction of easily reducible very small Pd-oxide nanoparticles, second peak at 140–280 °C was due to little bit larger Pd-oxide particles, third peak (300–500 °C) was due to reduction of surface oxygen of support CeO2 [15]. Normally the peak shows at about ~500 °C but the presence of Pd-species on the CeO2 surface makes the oxygen at the metal-support interface much easily reducible [16]. The last peak at above 600 °C was due to bulk reduction of CeO2 [15] (see Fig. 1). To analyze the morphology of the prepared catalyst we have carried out both SEM (Fig. S3) and TEM analysis (Fig. 2) of the fresh catalyst catalyst, which revealed that CeO2 particles were about 20–50 nm size and
Catalytic activity of the prepared catalyst was tested in the temperature range 300–800 °C at optimized weight hourly space velocity (WHSV) value 70,000 ml·g−1·h−1. Observed results for heat transfer and mass transfer (HT-MT) limitations experiments showed negligible HT-MT limitations (Figs. S7, S8). Carbon and material balance for the experiments was in between 98 and 102% (or ± 2%). Thermodynamic analysis says, DRM is highly endothermic reaction and requires very high temperature (800–1000 °C) to obtain high conversions as both the molecules are highly stable at normal condition [2,25,28]. CH4 þ CO2 →2CO þ 2H2 ΔH298K ¼ 247 kJ=mol With increasing pressure, reactant conversions and H2/CO yields drops drastically. So, according to thermodynamics, it is essential to operate DRM at lower pressure [2,25,28]. At these conditions carbon deposition is inevitable [2,25,28]. These problems can be overcome by operating the reaction at lower temperatures and with the catalysts which are resistant towards coking during DRM [2,25,28]. Rate of reaction over 0.86Pd-CeO2 catalyst was also calculated and reported (Table S1). Fig. 3a shows reactant conversions and H2/CO ratios at different temperature. It was found that experimental conversions values are lower than equilibrium conversion values (Fig. S9). Fig. 3b shows TOS stability of the catalyst at 800 °C. It can be seen from Fig. 3a that the reaction was initiated at 350 °C with methane and CO2 conversions 3.56% and 1.96% respectively and very slow increase in conversions were observed up to 500 °C. In our recent articles, we have showed that use of CeO2 as support or promoter and strong metal-support interactions activates methane at lower temperatures [3,4,14–17]. Initiation of DRM reaction at such a low temperature was the main advantage of the catalyst (Table S2) and it could be due to high activity of highly dispersed Pd-nanoparticles and high metal (Pd)-support (CeO2) interactions [3,4,14–16]. High concentration (32.8%) of Ce+3 in 0.86Pd-CeO2 was highly influential for activating methane at low temperature [2,5,8,12,15]. Deposition of Pd particles on the surface of CeO2 created oxygen vacancies and these oxygen vacancies highly improved the anchoring of Pd-nanoparticles and increased Pd-dispersion [28]. Methane dissociates over Pdnanoparticles and produced carbon and hydrogen. The produced carbon is then reduced the oxygen species of CeO2 adjacent to the Pd-nanoparticles producing CO [29]. Produced carbon can also take part in CO formation by dissociative adsorption of CO2, thereby removing the deposited carbon and initiating the DRM reaction. Very slow increase in reactant conversions up to 500 °C could be due to the inherent endothermic nature of DRM reaction. It could also be a reason that methane can be dissociated over Pd-oxide nanoparticles in an oxidizing environment created by CeO2 by its oxygen storage property but CO2 requires metallic Pd particles to dissociate and in the oxidizing environment its conversion was low at lower temperatures [3,4]. With increasing temperature and production of hydrogen by methane dissociation led to a reducing environment, which facilitated CO2 dissociation and increase in CO2 conversions was observed. After 550 °C, sharp increase in reactant conversions were observed may be due to favorable
R.K. Singha et al. / Catalysis Communications 92 (2017) 19–22
21
Fig. 1. (A) XRD pattern of (a) fresh and (b) spent catalysts, (B) XPS pattern of (a) fresh and (b) spent catalysts, (C) O 1S XPS of spent catalyst and (D) H2-TPR pattern of fresh 0.86PD-CeO2 catalyst. Spent catalyst condition: Temperature (800 °C), pressure (1 atm), WHSV (70,000 ml·g−1·h−1), reaction time (12 h).
thermodynamic conditions [2,3]. H2/CO ratio over the prepared 0.86PdCeO2 catalyst was found to be decreasing with temperature as with increase in temperature CO2 conversions were found to be increasing may be due to increasing dominance of RWGS reaction [2–4]. At the temperature range between 350 and 550 °C, higher H2/CO ratio was due to high methane conversion, low CO2 conversion and the absence of RWGS reaction. The results also indicates that there were less CO production by CO2 decomposition (CO2 + C → 2CO) and/or removal of deposited carbon (C + O → CO) by surface oxygen of CeO2. This observation led to the
conclusions that after 600 °C RWGS reactions was dominating and controlled the H2/CO ratio by consuming produced hydrogen and producing CO and H2O [13]. Now, production of H2O can lead to hydroxylation of active Pd-particles as it was also evidenced from the XPS analysis of spent 0.86Pd-CeO2 catalyst. Time on stream (TOS) stability of the catalyst was tested at 800 °C (Fig. 3b). It was found that initially methane and CO2 conversions were above 90% but after 9 h, slight decrease in reactant conversion was observed. Catalyst stability during DRM gets affected mainly by coking but also depends on sintering of catalyst particles, re-oxidation and hydroxylation of active species [3–5]. Sintering of the catalysts particles were observed from XRD and TEM analysis. In this case, catalyst deactivation was found not to be due to coking but due to sintering of catalyst particles, re-oxidation and hydroxylation of active Pd-species. Coking mainly occurs due to methane decomposition and Boudouard reaction [2–5], whereas re-oxidation of the active species occurs by oxidants like CO2, or surface oxygen of the catalyst [4,5]. Re-oxidation of active particles is quite possible in case of catalysts with high OSC [4, 5]. Hydroxylation of the active Pd-species could have occurred by H2O molecules, produced during DRM by RWGS reaction [4]. XPS analysis of the spent 0.86Pd-CeO2 catalyst confirmed the presence of PdO2 and Pd(OH)4 in the catalyst. As Oyama et al. reported that RWGS reaction is the main problem of DRM for synthesis gas production [13]; we have also experienced the same in this case. A comparative analysis was also carried out for catalytic activity and TOS stability of impregnated catalyst (1Pd-CeOImp 2 ) and 0.86Pd-CeO2 catalyst (Table S3, Figs. S10, S11). 4. Conclusions
Fig. 2. TEM images of fresh 0.86Pd-CeO2 catalysts.
Low temperature active Pd-CeO2 catalyst for dry reforming of methane was prepared by single step surfactant induced method. The main feature and advantage of the prepared catalyst was its low temperature activity for dry reforming of methane. The catalyst showed syngas
22
R.K. Singha et al. / Catalysis Communications 92 (2017) 19–22
Fig. 3. (a) Methane, CO2 conversions and H2/CO ratios, (b) time on stream (TOS) stability at 800 °C over 0.86Pd-CeO2 catalyst. Reaction condition: Pressure (1 atm), WHSV (70,000 ml·g−1·h−1), feed ratio (CH4:CO2:He = 1:1:12).
formation at about 350 °C due to presence of highly dispersed and very small Pd-nanoparticles (~2.9 nm). High metal-support interactions, and redox property of support CeO2 was the main driving force for the low temperature activity of the catalyst. About 32.8% Ce+3 concentrations (high reducibility) in 0.86Pd-CeO2 catalyst influenced methane activation at as low as 350 °C. Deactivation of the catalyst was found to be due to re-oxidation and hydroxylation of the active Pd-species and sintering of catalyst particles but not due to coking. RWGS reaction was the main problem with this catalyst, which highly affected H2/CO ratio by consuming produced hydrogen. Acknowledgements R.S. thanks University Grants Commission (UGC) , New Delhi, India for their fellowships. R.B. thanks CSIR for funding in the form of Network Project (CSC-0125, CSC-0117), DST Government of India for funding in the form of AISRF Grand Challenge Project (DST/INT/Aus/GCP-4/ 13(C)), Nanomission Scheme (SR/NM/NS-1105/2015). The Director, CSIR-IIP, is acknowledged for his help and encouragement. The Bilateral Collaboration Program supported by Italian CNR and Indian CSIR is kindly acknowledged. The authors thank the Analytical Science Division, Indian Institute of Petroleum for analytical services. Appendix A. Supplementary data Details of catalyst characterizations and experimental data. Supplementary data associated with this article can be found in the online version, at http://dx.doi.org/10.1016/j.catcom.2016.12.019. References [1] R. Bouarab, O. Cherifi, A. Auroux, Green Chem. 5 (2003) 209–212. [2] D. Pakhare, J. Spivey, Chem. Soc. Rev. 43 (2014) 7813–7837. [3] R.K. Singha, A. Yadav, A. Agrawal, A. Shukla, S. Adak, T. Sasaki, R. Bal, Appl. Catal. B Environ. 191 (2016) 165–178. [4] R.K. Singha, A. Yadav, A. Shukla, Z. Iqbal, C. Pendem, K. Sivakumar, R. Bal, ChemistrySelect 1 (2016) 3075–3085.
[5] S. Das, S. Thakur, A. Bag, M.S. Gupta, P. Mondal, A. Bordoloi, J. Catal. 330 (2015) 46–60. [6] D. Pakhare, V. Schwartz, V. Abdelsayed, D. Haynes, D. Shekhawat, J. Poston, J. Spivey, J. Catal. 316 (2014) 78–92. [7] A.I. Paksoy, B.S. Caglayan, A.E. Aksoylu, Appl. Catal. B Environ. 168 (2015) 164–174. [8] M. Yu, Y.A. Zhu, Y. Lu, G. Tong, K. Zhu, X. Zhou, Appl. Catal. B Environ. 165 (2015) 43–56. [9] A. Albarazi, M.E. Gálvez, P. Da Costa, Catal. Commun. 59 (2015) 108–112. [10] L. Xu, H. Song, L. Chou, ACS Catal. 2 (2012) 1331–1342. [11] L.C.P.F. Júnior, S. de Miguel, J.L.G. Fierro, M. do Carmo Rangel, Natural Gas Conversion VIII, 2007 499–504. [12] S.M. Stagg-Williams, F.B. Noronha, G. Fendley, D.E. Resasco, J. Catal. 194 (2000) 240–249. [13] S.T. Oyama, P. Hacarlioglu, Y. Gub, D. Lee, Int. J. Hydrog. Energy 37 (2012) 10444–10450. [14] R.K. Singha, A. Shukla, A. Yadav, L.N.S. Konathala, R. Bal, Appl. Catal. B Environ. 202 (2017) 473–488. [15] R.K. Singha, S. Ghosh, S.S. Acharyya, A. Yadav, A. Shukla, T. Sasaki, A.M. Venezia, C. Pendem, R. Bal, Catal, Sci. Technol. 6 (2016) 4601–4615. [16] R.K. Singha, A. Shukla, A. Yadav, S. Adak, Z. Iqbal, N. Siddiqui, R. Bal, Appl. Energy 178 (2016) 110–125. [17] R.K. Singha, S. Das, M. Pandey, S. Kumar, R. Bal, A. Bordoloi, Catal, Sci. Technol. 6 (2016) 7122–7136. [18] J.A. Farmer, C.T. Campbell, Science 329 (2010) 933–936. [19] S.D. Senanayake, J. Zhou, A.P. Baddorf, D.R. Mullins, Surf. Sci. 601 (2007) 3215–3223. [20] M. Cargnello, J.J. Delgado Jaén, J.C. Hernández Garrido, K. Bakhmutsky, T. Montini, J.J. Calvino Gámez, R.J. Gorte, P. Fornasiero, Science 337 (2012) 713–717. [21] P.G. Lustemberg, P.J. Ramírez, Z. Liu, R.A. Gutiérrez, D.G. Grinter, J. Carrasco, S.D. Senanayake, J.A. Rodriguez, M.V. Ganduglia-Pirovano, ACS Catal. 6 (2016) 8184–8191. [22] Z. Zuo, P.J. Ramírez, S.D. Senanayake, P. Liu, J.A. Rodriguez, J. Am. Chem. Soc. 138 (2016) 13810–13813. [23] G. Pantaleo, V. La Parola, F. Deganello, R.K. Singha, R. Bal, A.M. Venezia, Appl. Catal. B 189 (2016) 233–241. [24] Z. Liu, D.C. Grinter, P.G. Lustemberg, T.D. Nguyen-Phan, Y. Zhou, S. Luo, I. Waluyo, E.J. Crumlin, D.J. Stacchiola, J. Zhou, J. Carrasco, H.F. Busnengo, M.V. Ganduglia-Pirovano, S.D. Senanayake, J.A. Rodriguez, Angew. Chem. 128 (2016) 7581–7585. [25] M.K. Nikoo, N.A.S. Amin, Fuel Process. Technol. 92 (2011) 678–691. [26] A. Gupta, S. Das, C.J. Neal, S. Seal, J. Mater. Chem. B 4 (2016) 3195–3202. [27] P. Pal, R.K. Singha, A. Saha, R. Bal, A.B. Panda, J. Phys. Chem. C 119 (2015) 13610–13618. [28] X. Du, D. Zhang, L. Shi, R. Gao, J. Zhang, J. Phys. Chem. C 116 (2012) 10009–10016. [29] A. Luengnaruemitchai, S. Osuwan, E. Gulari, Catal. Commun. 4 (2003) 215–222.