Studies of the dissociation activities of methane and carbon dioxide over supported nickel catalyst

Accepted Manuscript Title: Studies of the dissociation activities of methane and carbon dioxide over supported nickel catalyst Author: Tsutomu Osawa Ryohei Agata Aki Mouri PII: DOI: Reference:

S1381-1169(15)30009-1 http://dx.doi.org/doi:10.1016/j.molcata.2015.06.034 MOLCAA 9548

To appear in:

Journal of Molecular Catalysis A: Chemical

Received date: Revised date: Accepted date:

2-4-2015 26-6-2015 27-6-2015

Please cite this article as: Tsutomu Osawa, Ryohei Agata, Aki Mouri, Studies of the dissociation activities of methane and carbon dioxide over supported nickel catalyst, Journal of Molecular Catalysis A: Chemical http://dx.doi.org/10.1016/j.molcata.2015.06.034 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Studies of the dissociation activities of methane and carbon dioxide over supported nickel catalyst Tsutomu Osawa*, Ryohei Agata, Aki Mouri Graduate School of Science and Engineering for Research, University of Toyama, Gofuku, Toyama 930-8555, Japan *Corresponding author e-mail: [email protected]

Graphical abstract Highlights Pulse experiment for testing the dissociation activity of CH4 and CO2 on supported Ni Dissociated CO2 easily produced adsorbed carbon on (Ni)/-Al2O3 Ni/CeO2 has a high dissociation activity of CO2 to produce CO Pre-adsorbed CHx species increased the dissociation of CO2 to CO on Ni/-Al2O3  CHx was effectively released as CO from the surface by pre-adsorbed oxygen on

Ni/CeO2 Abstract Pulse experiments for testing the dissociation activity of CH4 and CO2 were carried out on supported Ni catalysts (Ni/Al2O3, Ni/La2O3, and Ni/CeO2) with or without preadsorbed species on the catalysts. The CO2 pulse on Ni/Al2O3 produced a small amount of CO without pre-adsorbed species, while a CO2 pulse produced more CO in the presence of CHx species on Ni/Al2O3. The CHx species on Ni/Al2O3 promote the dissociation of CO2 on the catalyst. The production of CO by the CH4 pulse was low on Ni/Al2O3 regardless of the presence of adsorbed species by the pre-pulse of CO2, while both H2 and CO were effectively produced on Ni/CeO2. Keywords: nickel, methane, hydrogen-deuterium exchange, CO2 dissociation reaction

1

1. Introduction The use of natural gas has been attracting much attention for producing synthesis gas, a mixture of H2and CO, or a process for hydrogen production. Synthesis gas is a major feedstock for important processes, such as the Fisher-Tropsch synthesis, methanol synthesis, and ammonia synthesis [1, 2]. The carbon dioxide reforming of methane is appropriate for these processes, because it produces low H2/CO ratios [3, 4]. Furthermore, the carbon dioxide reforming of methane is attractive for environmental concerns, because it uses two major greenhouse gases. Many types of metals have an activity for the carbon dioxide reforming of methane. Although novel metal catalysts show a low coke deposition activity, application of these catalysts to an industrial process is limited because of their high cost and limited availability. Nickel-based catalysts are preferable for industrial practices from an economical view point and availability, but they have a high activity of coke deposition [5, 6]. Therefore, the development of a nickel catalyst with a high coke resistance is a crucial issue for the industrial use of nickel catalysts for the carbon dioxide reforming of methane. The following elementary reactions of the carbon dioxide reforming of methane has been proposed by Erdöhelyi et al. and Rostrup-Nielsen et al. [7, 8] CH4 + 2*→ CH3* + H* CH3 * + (3-x)*⇄CHx* + (3-x)H* CO2 + H*⇄ CO + HO* 2HO*⇄ H2O + O* + * CHx* + O* + (x-2)*⇄ CO + xH* 2H* ⇄ H2 + 2* For effectively producing CO and H2 from CO2 and CH4, the dissociative adsorption of methane and carbon dioxide on a catalyst surface, and an effective reaction between CHx* species and O* should be needed. When the value of x in CHx* is low, CHx* would form deposited carbon before reacting with O*[9]. Therefore, the catalyst without an activity for producing deposited carbon should have an appropriate degree of dissociative adsorption of CH4 and CO2. In this study, the ability of dissociation of CH4 and CO2 was investigated over -Al2O3, La2O3, CeO2, and Ni on these oxide supports for evaluating the features of these materials. These supports are often used for the 2

carbon dioxide reforming of methane, and especially, the addition of CeO2 as a promoter has been reported to reduce the amount of deposited carbon [10-16]. In order to reveal the features of the effects of the support types on the activity and the amount of deposited carbon, H/D exchange reactions and dissociation reactions of CO2 were carried out. We also carried out pulse experiments of (i) CO2 and (ii) that of CH4, and the two consecutive pulse experiments, i.e., (iii) three CO2 pulses, then CH4 pulse, and (iv) three CH4 pulses, then CO2 pulse. The pulse experiment would be a good technique for analyzing a fresh catalyst surface, because the fresh surface of the catalyst not adsorbed and covered by the reactant can be observed. Furthermore, the pulse experiment would also be appropriate for analyzing the reaction between the substrate and the specified adsorbed species on the surface.

2. Experimental 2.1. Materials CH4 (99.99%), D2 (99.995%), H2 (99.99%), CO2 (99.95%), O2 (99.9%), Ar (99.995%), and He (99.995%) were obtained from the Takachiho Trading Co., Ltd., Japan. -Al2O3 (Showa Light Metals Co., Ltd.) was used after treatment under an Ar atmosphere at 1173 K for 6 h. La2O3 (Wako Pure Chemical Co., Ltd.) and CeO2 (JRCCEO-1, Catalysis Society of Japan) were used as received. 2.2. Preparation of supported nickel catalyst A 5wt% nickel catalyst was prepared by the impregnation method. -Al2O3 (surface area: 69 m2), La2O3 (5 m2), or CeO2 (157 m2)(2.0 g) was immersed in an aqueous solution of Ni(NO3)2 (0.521 g of Ni(NO3)2 •6H2O in 40 cm3 of H2O) and stirred at 353 K for 2 h. The resulting solid was dried at 373 K for 18 h. The dried sample was calcined under O2 (20 cm3 min-1) at 623 K for 1 h and at 1173 K for 24 h to obtain the precursor. The particle size of the precursor was adjusted between 24-35 mesh. 2.3. Hydrogen-deuterium (H/D) exchange of methane A continuous gas-flow fixed bed glass reactor (14 mm id) was used for the catalytic test. The measurement without an exchange reaction was first carried out by a CH4pulse (0.6 cm3) without a catalyst in a He (50 cm3 min-1) stream. The precursor 3

(0.05 g) was placed in the reactor and treated with a D2 stream (mixture of D2 (3 cm3 min-1) and He (50 cm3 min-1)) at atmospheric pressure for 1 h at 1173 K. The H/D exchange reaction was carried out in the same reactor by a CH4 pulse (0.6 cm3) under a continuous gas flow of a mixture of D2 (3 cm3 min-1) and He (50 cm3 min-1) at 1173 K. A small amount of the outlet flow was continuously leaked through a silica capillary tube (0.05 mm id X 2.5 m) into a Q-mass spectrometer (PFEIFFER Vacuum Prisma QMS 200) for isotopic analysis. The data set of the ion current (m/z=12 to 20) was collected every 0.7 seconds. Background corrections of the observed spectra were made (H2O at m/z=18) [17].

2.4. Quantitative determination of the exchanged species by the correction of the methane isotopologues When the determination of the isotopologues of methane was carried out by quadrupole mass spectroscopy, the methane isotopologues CHxD4-x (x=0-4) have their own characteristic fragmentation patterns, which overlap each other. Therefore, to obtain the pure molecular ion peaks of the methane isotopologues, corrections have to be made for the observed ion peaks. In the present study, to determine each of the isotopologues in the pulse experiment, the pulse areas of the ion current were calculated by our proposed method. The correction was based on the ratios of the ion current of each daughter fragment peak to the parent peak for every data point. The fragment patterns were obtained by CH4 pulse and CD4 pulse without catalyst. The detailed correction method is described elsewhere [18]. The percentage of the exchanged species was expressed by the following equations (Eqs. 1) using the peak area S(CHxD4-x) (x=0-4) of its corrected ion current of the intrinsic parent ion. Exchanged species / % = 100 X {S(CH3D) + S(CH2D2) + S(CHD3) + S(CD4)}/ Ssum (Eqs. 1) Ssum = S(CH4) + S(CH3D) + S(CH2D2) + S(CHD3) + S(CD4) The average value of x in CHxD4-x was obtained using the following equation (Eqs. 2). The x value would represent the degree of the production of deposited carbon from the 4

adsorbed species. The lower the value x in CHx*, the more easily the deposited carbon is produced. Q(CHxD4-x) = S(CHxD4-x) / {S(CH3D) + S(CH2D2) + S(CHD3) + S(CD4)} (Eqs. 2) x(average) = 3 X Q(CH3D) + 2 X Q(CH2D2) + 1 X Q(CHD3) + 0 X Q(CD4)

2.5. Dissociation reaction of CO2 The dissociation reaction of CO2 was carried out using the same reactor for the hydrogen-deuterium exchange reaction. The precursor (0.05 g) was placed in the reactor and treated with a mixture of H2 (3 cm3 min-1) and He (50 cm3 min-1) at atmospheric pressure for 1 h at 1173 K. The dissociation reaction of CO2 was carried out in the same reactor by a CO2 pulse (0.6 cm3) at 1173 K. The ion current (m/z=2, 12, 28, and 44) was collected every 0.7 seconds using a Q-mass spectrometer (PFEIFFER Vacuum Prisma QMS 200).

2.6. Quantitative determination of the dissociated species of CO2 When the outlet gas of the dissociation reaction of CO2 was introduced to the quadrupole mass analyzer, the fragments of CO (m/z=28) and C (m/z=12) as well as CO2 (m/z=44) were observed. In order to obtain the pure molecular ion peaks of CO2 (CO2out) and CO (COout) at the outlet of the reactor, corrections have to be made for the observed ion peaks. The observed ion currents of CO2 (CO2obs, m/z=44), CO (COobs, m/z=28), and C (Cobs, m/z=12) were expressed by the following equations (Eqs. 3), where x and y are the ratio of the dissociation of CO2 to CO and to C, respectively, and z is the ratio of the dissociation of CO to C in the analysis chamber. The x and y were determined by the introduction of CO2 without catalyst, and z was determined by the introduction of CO without catalyst to the analysis chamber. CO2obs = (1-x-y)CO2out COobs = xCO2out + (1-z)COout

(Eqs. 3)

Cobs = yCO2out + zCOout

5

The ratio of the production of CO, the ratio of the adsorbed carbon on the catalyst, the activity of the dissociation of CO2 were express by the following equations (Eqs. 4) using the peak areas of CO2out, COout, Cad, and CO2in (S(CO2out), S(COout), S(Cad), and S(CO2in), respectively). S(CO2in) was determined by the experiment without a catalyst. Ratio of production of CO / % = 100 X S(COout) / S(CO2in) Ratio of adsorbed C / % = 100 X S(Cad) / S(CO2in) Dissociation activity / % = 100 X {S(COout) + S(Cad)} / S(CO2in)

(Eqs. 4)

S(Cad) = S(CO2in) - S(CO2out) - S(COout) S(CO2in) = S(CO2obs) + S(COobs) + S(Cobs)

(measured without catalyst)

2.7. Pulse experiments with or without specific pre-adsorbed species on the catalysts A fixed bed glass reactor (14 mm id) was used for the pulse experiments. For the activation of the catalyst, the precursor (0.05 g) was placed in the reactor and treated with a H2 stream (mixture of H2 (3 cm3 min-1) and He (50 cm3 min-1)) at atmospheric pressure for 1 h at 1173 K. 2.7.1. CO2 pulse CO2 pulse (0.6 cm3) was introduced into the reactor under a continuous gas flow of He (50 cm3 min-1) at 1173 K. The product gas was analyzed by a Q-mass spectrometer (PFEIFFER Vacuum Prisma QMS 200).

2.7.2. CO2 pulse after three successive CH4 pulses CH4 pulse (0.6 cm3) was introduced into the reactor under a continuous gas flow of He (50 cm3 min-1) at 1173 K. The CH4 pulses were carried out three times with 5 min intervals. After the CH4 pulses, the CO2 pulse (0.6 cm3) experiment was carried out under a continuous gas flow of He (50 cm3 min-1) at 1173 K. The product gas was analyzed by a Q-mass spectrometer.

2.7.3. CH4 pulse 6

A CH4 pulse (0.6 cm3) was introduced into the reactor under a continuous gas flow of He (50 cm3 min-1) at 1173 K. The product gas was analyzed by a Q-mass spectrometer.

2.7.4. CH4 pulse after successive three CO2 pulses A CO2 pulse (0.6 cm3) was introduced into the reactor under a continuous gas flow of He (50 cm3 min-1) at 1173 K. CO2 pulses were carried out three times with 5 min intervals. After the CO2 pulses, a CH4 pulse (0.6 cm3) experiment was carried out under a continuous gas flow of He (50 cm3 min-1) at 1173 K. The product gas was analyzed by a Q-mass spectrometer. 3. Results and discussion 3.1. Activity of the CO2 reforming of methane For the preliminary experiment, the CO2 reforming reaction of methane was carried out over various types of catalysts. Table 1 shows the results of the activity and the amount of the deposited carbon on the catalysts.

The conversions of CH4 and that of CO2 were very low when the support oxides (Al2O3, La2O3, and CeO2) were used as the catalysts. A large quantity of deposited carbon was observed on -Al2O3 compared to La2O3 and CeO2. On the other hand, the activities of the supported nickel catalysts using the same supports as examined above were much higher than those without nickel on the support surface. The activity of Ni/-Al2O3 was higher, while the amount of the deposited carbon was more than those of Ni/La2O3 and Ni/CeO2. Although the activity of Ni/La2O3 and Ni/CeO2 were almost the same, the amount of carbon on Ni/CeO2 was less than that on Ni/La2O3. The smaller the particle size of Ni, the higher the activity of reforming reaction of methane would be. On the contrary, the bigger the particle size of Ni, larger amount of carbon is produced [19]. From the results of Table 1, the activity was Ni/Al2O3 >Ni/La2O3≈Ni/CeO2, and the deposited carbon was produced on Ni/Al2O3 >Ni/La2O3>Ni/CeO2. As the surface area of the supports used (before calcination) were CeO2 (157 m2)>Al2O3 (69 m2)>La2O3 (5 m2), the effects of the types of supports rather 7

than the particle size of Ni would be significant for the catalysts used in this study. This would be partly due to the sintering of the supports during the calcination at 1173 K for 24 h. 3.2. H/D exchange of methane and the dissociation of CO2 When CH4 was pulsed onto the catalyst, the dissociation of CH4 occurred to produce adsorbed species (CH3*,CH2*, CH*, C*) on the surface. These adsorbed species were reacted with D2 gas and the exchanged product CHxD4-x (CH3D, CH2D2, CHD3, CD4) was desorbed into the gas phase. Table 2 shows the percentage of the exchanged species and the average value of x in CHxD4-x over the -Al2O3, La2O3, CeO2 and Ni supported catalysts on these supports.

The lower the value x in CHx*, the more easily the deposited carbon is produced from the adsorbed species. The -Al2O3 support was less active for the H/D exchange of methane than La2O3 and CeO2. In the case of the supported Ni catalysts, Ni/La2O3 was more active than Ni/-Al2O3 and Ni/ CeO2. Concerning the average value x, x for Al2O3was rather higher than or equal to the other supports, irrespective of the absence or the presence of Ni on the supports. Based on the results listed in Table 1, -Al2O3 and Ni/-Al2O3 produced the most deposited carbon compared to the other catalysts. As the results of the H/D exchange reaction of methane did not show low x values for both -Al2O3 and Ni/-Al2O3, a large amount of deposited carbon for -Al2O3 and Ni/Al2O3 during CO2 reforming of methane could not be explained by the dissociation activity of methane. Table 3 shows the results of the dissociation reaction of CO2 on -Al2O3, La2O3, CeO2 and Ni supported catalysts on these supports. Among the three supports, -Al2O3 was the most active for the dissociation of CO2. However, most of the dissociated CO2 (12%) was transformed into the adsorbed carbon (11%), which was a significantly high ratio compared to the other supports. Concerning the supported Ni catalysts, Ni/CeO2 showed 94% of the dissociation activity and the very high ratio of production of CO (76%). These results would indicate that the adsorption/dissociation of CO2 and the desorption in the form of CO easily occurred on Ni/CeO2, and that at the same time, an oxygen species (O*) was produced on the 8

surface. Meanwhile, Ni/La2O3 had a low dissociation activity. This could be due to the strong interaction of CO2 to Ni/La2O3[20, 21]. It has been reported that the deposition of carbon on the catalyst surface during the CO2 reforming of methane is due to the two reactions, i.e., the decomposition of methane (Eq. 5) and CO disproportionation (Eq. 6) [3]. CH4→C* + 2H2 (Eq. 5) 2CO→C* + CO2 (Eq. 6) For the CO2 reforming of methane, -Al2O3 produced more deposited carbon on the surface than La2O3 and CeO2 (Table 1). Although the value x in CHx* of -Al2O3 during the H/D exchange reaction was almost the same as the other supports (Table 2), the ratio of the adsorbed carbon to the dissociated CO2 on -Al2O3 and Ni/-Al2O3 during the dissociation reaction of CO2 was much more than those on the other supports (Table 3). One of the reasons for producing a large amount of deposited carbon on Al2O3 and Ni/-Al2O3 would be that the dissociation of CO2 on (Ni)/-Al2O3 was ready to produce carbon and adsorbed oxygen (O*), but not to produce carbon monoxide and O*. Most of the deposited carbon on (Ni)/-Al2O3 could be due to the dissociation of CO2 rather than the decomposition of CH4. On the other hand, Ni/CeO2 produced the least deposited carbon during the CO2 reforming of methane. This would be attributed to the features of the high dissociation activity of CO2 and the high production ratio of CO on Ni/CeO2. The oxygen species (O*) that originated from the dissociation of CO2 or lattice mobile oxygen species (O*lattice) that supplied by the dissociation of CO2 could play an important role in diminishing the deposited carbon (Eqs. 7 and 8). C* + O* (or O*lattice) →CO *

*

CHx + O (or

O*lattice)

(Eq. 7)

→CO + xH*

(Eq. 8)

3.3. Effects of the adsorbed species on the change in the ion current of the product gas 3.3.1. CO2 pulse experiment with or without adsorbed species produced by successive CH4 pulses Fig. 1 shows the ion current obtained by a CO2 pulse over three supported Ni 9

catalysts. Fig. 2 shows the ion current obtained by a CO2 pulse after three successive CH4 pulses.

Over the Ni/-Al2O3, a small amount of CO was produced by the CO2 pulse without pre-pulses (Fig. 1 (a)), while a comparable amount of CO was produced by the CO2 pulse after the CH4 pulses (Fig. 2 (a)). These results could be explained by the promotion of the CO2 dissociation by CHx* species adsorbed on Ni/-Al2O3. The increase in the production of CO on Ni/-Al2O3 with adsorbed CHx*could be due to the reaction of CO2 with adsorbed H* (Eq. 9) [3] and/or CHx (Eq. 10), and/or the Eq. 11 reaction could more easily occur than the simple dissociation of CO2. CO2 +H* →CO + OH* CO2 +CHx* → 2CO + xH* CO2+

Ni-CHx* →CO+NiO-CHx*

(Eq. 9) (Eq. 10) (Eq. 11)

The pulse rise times of CO and CO2 were the same, while the CO peak reached the top of the peak earlier than CO2. This showed that the ratio of CO to CO2 decreased during the passage of CO2 through the catalyst bed. Based on the results of Table 3, when Al2O3 or Ni/-Al2O3 was used as a catalyst, the ratio of the adsorbed carbon to the dissociated CO2 was high. It was revealed that the adsorbed carbon was reduced and the production of CO increased when the CHx* and/or H* species were on the catalyst surface. The enhanced production of CO in the presence of CHx would be due to a reduced yield of carbon-forming reaction, so that more CO is desorbed. In the case of La2O3 used as a support (Fig. 1 (b) and Fig. 2 (b)), the same tendency was observed when using Al2O3, but the production of CO was less than the case of Al2O3. The reason that the production of CO was less on La2O3 would partly due to the small surface area of La2O3 used in this study. On the contrary, when CeO2 was used as a support (Fig. 1 (c) and Fig. 2 (c)), the main product was CO irrespective of the presence of the adsorbed species. This would be explained by the significant ability of adding or removing oxygen species produced by the dissociation of CO2 into or out of the CeO2 support [22]. CeO2 possesses large 10

oxygen storage capacity, that promotes the formation of oxygen vacancies and increases the mobility of oxygen [23-25]. As lewis bases, such as CeO2 have a strong affinity for the chemisorption of CO2, dissociation of CO2 leaves oxygen on the catalyst surface and the adsorbed mobile oxygen would react with CHx species [26].

3.3.2. CH4 pulse experiment with or without adsorbed species produced by successive CO2 pulses Fig. 3 shows the ion current obtained by the CH4 pulse over three supported Ni catalysts. Fig. 4 shows the ion current obtained by the CH4 pulse after three successive CO2 pulses.

Over Ni/Al2O3 without the adsorbed species (Fig. 3(a)), H2 and a lower amount of CO were observed. CH4 would be dissociated into H2 and carbon species on Ni/Al2O3 (Eq. 12), and the carbon species would be reacted with the lattice oxygen of Al2O3 to produce CO (Eq. 13). Meanwhile, when the dissociated species of CO2 were adsorbed on Ni/Al2O3 (Fig. 4(a)), the production of H2 and CO was significantly increased. This could be due to the adsorbed oxygen produced by the dissociation of CO2. The adsorbed carbon species would be reacted with the adsorbed oxygen (Eq. 14) in addition to the lattice oxygen or lattice mobile oxygen species supplied by the dissociation of CO2, and hence the production of CO and H2 increased. The dissociation of CH4 could be increased by the removal of CHx* on the surface by the adsorbed oxygen. CH4→CHx* + (4-x)/2H2

(Eq. 12)

CHx* + O(lattice oxygen) →CO+ x/2H2 *

(Eq. 13)

*

CHx + O (adsorbed oxygen) →CO+ x/2H2 (Eq. 14) When Ni/La2O3 was used, the amount of produced H2 and CO did not change irrespective of the CO2 pulses before the CH4 pulse. These results indicate that the CO2 pulses over Ni/La2O3 produced a low amount of adsorbed species from CO2. This 11

agreed with the results of Fig. 1(b). The production of H2 and CO over Ni/CeO2 without the CO2 pulses before the CH4 pulse was greater than that over Ni/Al2O3 and Ni/La2O3 (Fig. 3(c)). CeO2 has a high oxygen storage capacity and the ability of reversibly adsorbing and releasing oxygen [22]. As the lattice oxygen of CeO2 was easily mobile, the reactions of Eq. 13 and hence Eq. 12 would be enhanced. Under the conditions of the presence of the adsorbed species produced by the CO2 pulses before the CH4 pulse, the production of H2 and CO significantly increased (Fig. 4(c)). Compared to the amount of produced H2 and CO over Ni/Al2O3, the amount of H2 was almost the same, while the amount of CO produced over Ni/CeO2 was much more than that over Ni/Al2O3. These results indicate that Ni/Al2O3 has the ability of dissociating CH4 to produce H2(Eq. 12) comparable to Ni/CeO2, however, the ability of producing CO by Eqs. 13 and 14 is less than that of Ni/CeO2. The large amount of deposited carbon on Ni/Al2O3 for the CO2 reforming of methane would be due to these features of Ni/Al2O3. The CH4 pulse after three CO2 pulses on Ni/CeO2 resulted in the production of CO2. This would be due to the excess oxidation reaction of the CHx* species by the adsorbed oxygen on CeO2. 4. Conclusion The reactions involve in the carbon dioxide reforming of methane over oxides (Al2O3, La2O3, and CeO2) and Ni on these supports revealed that -Al2O3 and Ni/Al2O3 produced a large amount of deposited carbon on the surface. In order to reveal the features of the carbon deposition on these catalysts, pulse experiments for testing the dissociation activity of CH4 and CO2 were carried out.For (Ni)/-Al2O3, the highest value of x of the adsorbed CHx* species was observed during the dissociation experiments of the CH4 pulse, while the ratio of deposited carbon to dissociated CO2 was the highest during the dissociation experiments of the CO2 pulse.One of the reasons for producing a large amount of deposited carbon on (Ni)/-Al2O3 during the CO2 reforming of methane would be that the dissociation of CO2 on (Ni)/-Al2O3 was ready to produce deposited carbon, but not to produce carbon monoxide. The existence of pre-adsorbed CHx* species on (Ni)/-Al2O3 inhibited the dissociation of CO2 into deposited carbon, and hence promoted the production of CO. On the other hand, H2 and a lower amount of CO were produced by the CH4 pulse irrespective of the pre-adsorbed species by the CO2 pulses on Ni/Al2O3, meanwhile both H2 and CO were effectively 12

produced by the CH4 pulse on Ni/CeO2. This would be due to the active adsorbed oxygen species and mobile lattice oxygen of Ni/CeO2.

References

13

[1] J.R. Rostrup-Nielsen, J. Sehested, J.K. Norskov, Adv. Catal., 47 (2002) 65-139. [2] J.R.H. Ross, A.N.J. van Keulen, M.E.S. Hegarty, K. Seshan, Catal. Today, 30 (1996) 193-199. [3] M.C.J. Bradford, M.A. Vannice, E. Ruckenstein, Catal. Rev. - Sci. Eng., 41 (1999) 1-42. [4] Y.H. Hu, E. Ruckenstein, Adv. Catal., 48 (2004) 297-345. [5] K. Tomishige, O. Yamazaki, Y. Chen, K. Yokoyama, X. Li, K. Fujimoto, Catal. Today, 45 (1998) 35-39. [6] J.M. Wei, B.Q. Xu, J.L. Li, Z.X. Cheng, Q.M. Zhu, Appl. Catal., A: Gen., 196 (2000) L167-L172. [7] A. Erdohelyi, J. Cserenyi, F. Solymosi, J. Catal., 141 (1993) 287-299. [8] J.R. Rostrup-Nielsen, J.H. Bak Hansen, J. Catal., 144 (1993) 38-49. [9] H. Matsumoto, Shokubai, 18 (1976) 71-77. [10] J. Chen, R. Wang, J. Zhang, F. He, S. Han, J. Mol. Catal. A: Chem., 235 (2005) 302-310. [11] H. Li, H. Xu, J. Wang, Journal of Natural Gas Chemistry, 20 (2011) 1-8. [12] K.Y. Koo, H.-S. Roh, U.H. Jung, W.L. Yoon, Catal. Today, 185 (2012) 126-130. [13] W. Chen, G. Zhao, Q. Xue, L. Chen, Y. Lu, Appl. Catal., B: Env., 136-137 (2013) 260-268. [14] E.C. Faria, R.C.R. Neto, R.C. Colman, F.B. Noronha, Catal. Today, 228 (2014) 138-144. [15] T. Osawa, Y. Nakai, A. Mouri, I.Y.S. Lee, Appl. Catal., A: Gen., 474 (2014) 100106. [16] C.G. Rotaru, G. Postole, M. Florea, F. Matei-Rutkovska, V.I. Pârvulescu, P. Gelin, Appl. Catal. A: Gen., 494 (2015) 29-40. [17] T. Osawa, Y. Hamano, S. Saga, O. Takayasu, J. Mol. Catal. A: Chem., 298 (2009) 111-114. [18] T. Osawa, K. Sugimoto, M. Miyamoto, O. Takayasu, React. Kinet. Catal. Lett., 90 (2007) 93-100. [19] M.C.J. Bradford, M.A. Vannice, Appl. Catal. A: Gen., 142 (1996) 73-96. [20] Y. Takita, F. Hori, N. Yamazoe, T. Seiyama, Bull. Chem. Soc. Jpn., 60 (1987) 2757-2762. [21] V.R. Choudhary, V.H. Rane, J. Catal., 130 (1991) 411-422. 14

[22] H.C. Yao, Y.F. Yu Yao, J. Catal., 86 (1984) 254-265. [23] X. Cai, X. Dong, W. Lin, Journal of Natural Gas Chemistry, 17 (2008) 98-102. [24] Ş. Özkara-Aydınoğlu, A.E. Aksoylu, Catal. Commun., 11 (2010) 1165-1170. [25] F. Pompeo, D. Gazzoli, N.N. Nichio, Int. J. Hydrogen Energy, 34 (2009) 22602268. [26] J. Zhu, X. Peng, L. Yao, X. Deng, H. Dong, D. Tong, C. Hu, Int. J. Hydrogen Energy, 38 (2013) 117-126.

15

Figure 1. Ion current obtained by CO2 pulse a)

Ni/-Al2O3, b) Ni/La2O3, c) Ni/CeO2

16

Figure 2. Ion current obtained by CO2 pulse after three CH4 pulses a) Ni/-Al2O3, b) Ni/La2O3, c) Ni/CeO2

17

Figure 3. Ion current obtained by CH4 a) Ni/-Al2O3, b) Ni/La2O3, c) Ni/CeO2

18

Figure 4. Ion current obtained by CH4 pulse after three CO2 pulses a) Ni/d-Al2O3, b) Ni/La2O3, c) Ni/CeO2

19

Table 1 Activity of CO2 reforming of methane and the amount of deposited carbon on the catalyst a) Catalyst

Conversion of Conversion of Amount of deposited CH4 / % b) CO2 / % b) carbon / a.u. c) 0.6 0.3 214 -Al2O3 La2O3 2.2 4.8 1.8 CeO2 3.4 7.2 0.8 68 77 100 5wt% Ni/-Al2O3 5wt% Ni/La2O3 33 50 30 5wt% Ni/CeO2 31 47 14 3 -1 a) Reaction conditions; catalyst (0.2 g), CH4 (300 cm min ) + CO2 (300 cm3 min-1), reaction temperature (1173 K) b) Average conversion for 6 h c) Relative amount of deposited carbon after a 6-h reaction based on the carbon on 5wt% Ni/-Al2O3=100

Table 2 Percentage of the exchanged species and the average value of x in CHxD4-x Exchanged species Average value of x in Catalyst /% CHxD4-x 43 2.62 -Al2O3 La2O3 54 2.51 CeO2 57 2.50 46 2.60 5wt% Ni/-Al2O3 5wt% Ni/La2O3 64 2.41 5wt% Ni/CeO2 57 2.53

Table 3 Dissociation reaction of CO2 on the catalyst Ratio of Ratio of production of adsorbed Catalyst CO / % C/% -Al2O3 La2O3 CeO2

1 3 4

11 1 5

20

Dissociation activity / %

12 4 9

Adsorbed Carbon / Dissociated CO2 0.92 0.25 0.56

5wt% Ni/-Al2O3 5wt% Ni/La2O3 5wt% Ni/CeO2

7 3 76

15 1 18

21

22 4 94

0.68 0.25 0.19