- Email: [email protected]
Catalytic activity of several carbons with different structures for methane decomposition and by-produced carbons
Accepted Manuscript Full Length Article Catalytic Activity of Several Carbons with Different Structures for Methane Decomposition and By-Produced Carbons Haruki Nishii, Dai Miyamoto, Yoshito Umeda, Hiroaki Hamaguchi, Masashi Suzuki, Tsuyoshi Tanimoto, Toru Harigai, Hirofumi Takikawa, Yoshiyuki Suda PII: DOI: Reference:
S0169-4332(18)33403-2 https://doi.org/10.1016/j.apsusc.2018.12.073 APSUSC 41170
To appear in:
Applied Surface Science
Received Date: Revised Date: Accepted Date:
9 May 2018 29 August 2018 8 December 2018
Please cite this article as: H. Nishii, D. Miyamoto, Y. Umeda, H. Hamaguchi, M. Suzuki, T. Tanimoto, T. Harigai, H. Takikawa, Y. Suda, Catalytic Activity of Several Carbons with Different Structures for Methane Decomposition and By-Produced Carbons, Applied Surface Science (2018), doi: https://doi.org/10.1016/j.apsusc.2018.12.073
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.
Catalytic Activity of Several Carbons with Different Structures for Methane Decomposition and By-Produced Carbons
Haruki Nishii1, Dai Miyamoto1, Yoshito Umeda2, Hiroaki Hamaguchi3, Masashi Suzuki3, Tsuyoshi Tanimoto1, Toru Harigai1, Hirofumi Takikawa1, Yoshiyuki Suda1*
1
Department of Electrical and Electronic Information Engineering, Toyohashi University of Technology, Toyohashi, Aichi 441-8580, Japan 2
3
Toho Cryogenics Company Limited, Chita, Aichi 456-0004, Japan
Aichi Center for Industry and Science Technology, Kariya, Aichi 470-0536, Japan E-mail: [email protected]
1
Abstract Since methane decomposition has no CO2 emissions, it is attracting attention as a hydrogen production method with a low environmental burden. Understanding the structure of the produced carbon is important for long-term stable production of hydrogen. In this study, methane decomposition was carried out on carbons with several different structures (activated carbon (AC), carbon black (CB), meso-porous carbon (MC), and carbon nanofiber (CNF)). We have found that the carbon produced by methane decomposition decreases activity by covering the catalyst, but itself also acts as a catalyst irrespective of the original carbon catalysts. All of the catalysts continued to maintain a methane conversion ratio of about 17% by catalyzing the produced carbon even after the activity was lowered. By analysis of the catalysts before and after the experiment, it was shown that the produced carbon covered the catalyst surface and resulted in a specific surface area of about 10 m2/g and the intensity ratio of the D band to the G band in the Raman spectra (ID/IG) of around 1.55 irrespective of the original carbons structures. We proposed that Raman spectroscopy is an effective method for evaluating initial catalytic activity for methane decomposition because ID/IG of the catalysts before the experiment have a linear relationship with the methane conversion ratio per unit surface area in the early stage of the reaction.
2
KEYWORDS: methane decomposition, Raman spectroscopy, activation energy, hydrogen production, carbon black, activated carbon.
1. Introduction Hydrogen utilization technology including fuel cells that are characterized by high energy efficiency and no CO2 emission at the utilization stage, can contribute to energy saving and environmental burden reduction. Furthermore, the market size of hydrogen related technology is predicted to be about 1 trillion Japanese Yen by 2030 and about 8 trillion Japanese Yen by 2050 in Japan, making it an important technology. On the other hand, there are many problems in hydrogen utilization: safety, cost, and infrastructure [1,2]. There is a possibility that part of the problem can be solved by on-site hydrogen production using natural gas. By using an existing natural gas supply line, hydrogen transportation is safe and of low cost because it does not require the direct transportation of hydrogen and the laying of new supply lines [3]. The main component of natural gas is methane. Various methods for hydrogen production using methane have been proposed: (a) steam methane reforming, (b) partial oxidation, and (c) autothermal reforming [4,5,6]. However, since all of these hydrogen
3
production methods involve the emission of CO 2, CCS (carbon dioxide capture and storage) is required, which leads to a cost increase. Therefore, methane decomposition without CO and CO2 emission attracts attention as a promising hydrogen production method. Methane is decomposed into carbon and hydrogen according to the following reaction [7]: .
(1)
Various noble metals such as Ni, Cu, and Fe have been used as catalysts for methane decomposition [8,9,10,11,12,13,14]. However, when these noble metals are used as a catalyst, these catalysts are covered by the deposition of carbon produced by the process, so that a regeneration process accompanied by CO2 emission by carbon combustion is required. In recent years, the use of carbons as a catalyst with a lower cost than metal catalysts have attracted attention [15,16,17]. Compared to metal catalysts, carbon catalysts have advantages such as high temperature tolerance, resistance to impurities such as sulfur, and possibility of further cost reduction due to commercial use of high purity carbon [18,19]. Various carbons have been used as catalysts for methane decomposition. Kim, et al.,
4
Kryzynski, et al., and Ashok, et al. reported that AC showed high catalytic activity in the early stage of the reaction, but that the pores on the surface were blocked by produced carbon by methane decomposition and the activity was immediately decreased [20,21,22]. Lazao, et al. and Suelves, et al. reported that CB showed stable catalytic activity for a long time, but the activity decreased when a certain amount of carbon deposits [23,24]. Muradov, et al. evaluated catalytic activity of a wide range of carbon materials for methane decomposition reaction and determine major factors governing their activity. They tested carbons with different structures such as activated carbon (AC), carbon black (CB), carbon nanotubes (CNTs), and glassy carbon and reported that the methane decomposition rate in the early stage of reaction was related to the surface area and that AC and CB were promising catalysts [25]. There is a common point in these reports. As produced carbon by the reaction covered the surface of the catalyst, the activity decreased. On the other hand, the methane conversion ratio continued to show a constant value even after activity was lowered. This fact suggests that the carbon itself produced by the methane decomposition acts as a catalyst. There are few reports on the catalytic activity after the decrease in activity. Especially, the relationship between the structure of the catalyst and the activity of produced carbon by the reaction is not known. If carbon having high catalytic activity can produce, hydrogen can be produced stably
5
for a long period of time. In this study, methane decomposition was carried out on carbons with several different structures including carbon nanofiber (CNF) with high crystallinity, AC, CB, and meso-porous carbon (MC), and the activity of produced carbon by the reaction was investigated. Furthermore, the relationship between the surface structure of carbon and catalytic activity was discussed by analyzing the N2 Brunauer-Emmett-Teller (BET) surface area and Raman spectra.
2. Experimental The carbons tested in this study are listed in Table 1. The BET specific surface areas of the catalysts ranged from 270 to 2208 m2/g. AC and MC are carbons having a pore structure. Most of the pores on the surface of the AC are micropores of 2 nm or less. MC controls surface pores in mesopores of 2 to 50 nm. CB has a structure in which fine particles with a diameter of several nm are aggregated. CNF is a fibrous carbon with a diameter of several nm. SEM micrographs of the carbons before the experiment are presented in Fig. 1. Methane (99.9%) and nitrogen were used without further purification. The methane decomposition was carried out in a horizontal, fixed-bed, 40 mm I.D. quartz-tube flow
6
reactor heated by an electric tube furnace. Quartz wool was used to fix the catalyst. The standard reaction conditions were a methane flow rate in range of 50 to 135 cm3 (under standard temperature and pressure, STP)/min, a catalyst weight of 1.7 g, a space velocity (SV, the amount of gas per hour passing through the reactor divided by the volume of the catalyst in the reactor) of 360 h-1, and the reaction temperature in the range of 1073 to 1173 K. The product gas was analyzed by a hydrogen detector (PD-12, New Cosmos Electric Co., Ltd., Osaka, Japan) and gas chromatography (GC-2014, Shimadzu Corp., Kyoto, Japan), using a SHINCARBON ST (ZT-13, Shinwa Chemical Industries, Ltd., Kyoto, Japan), Ar carrier, and a thermal conductivity detector. The hydrogen detector can continuously measure the hydrogen concentration as compared with gas chromatography, but the initial response is slow (It takes about 20 minutes to accurately measure hydrogen concentration). On the other hand, gas chromatography can measure hydrogen concentration at the initial reaction, but it is measured at intervals of 20 minutes because the retention time of methane is about 15 minutes. Fig. 2 shows the results of the methane decomposition test in which the hydrogen concentration was measured using PD-12 and GC-2014. Except for 30 minutes after the reaction, it was confirmed that the error in hydrogen concentration due to the difference in analyzer used was within 5%.
7
The methane conversion ratio was derived from the following formula using the hydrogen concentration obtained by analysis. (2) In order to clarify the effect of the catalyst, the non-catalytic methane conversion ratio was subtracted from the methane conversion ratio over the catalysts. The non-catalytic methane conversion after reaction for five minutes was 8% at SV = 360 h-1 and T = 1173 K. Carbons before and after the experiment was analyzed by N2 adsorption measurement (BELSORP-max, MicrotracBEL Corp., Osaka, Japan), scanning electron microscopy (SEM, SU8000, Hitachi High-Technologies Corp., Tokyo, Japan), and Raman spectroscopic analysis (NRS-1000, JASCO Corp., Tokyo, Japan).
3. Results The catalytic activity of carbons with different structures at 1173 K were compared and the results are shown in Fig. 3. Temporal change of methane conversion ratio is shown in Fig. 3 (a) at SV = 360 h-1 and methane conversion ratio after five minutes of reaction to the specific surface area is shown in Fig. 3 (b). Methane conversion ratio of
8
all the tested catalysts eventually reached about 17%. In addition, all the catalysts except CNF showed the highest methane conversion ratio at the initial reaction. Particularly in AC, the methane conversion after five minutes of reaction was the highest at 77%, but the decrease in activity was also the fastest among all the carbons. Although MC showed the same tendency as AC, the initial activity was 63%, which was lower than AC and the decrease in activity was also slow. CB showed an initial methane conversion ratio lower than AC, but maintained 32% up to 300 min. After that, the methane conversion ratio decreased, and eventually reached a stable state again at 17% after about 600 min. On the other hand, CNF showed a tendency different from the other carbons. Although the initial methane conversion ratio of CNF was 3%, which is lower than that of other catalysts, the methane conversion ratio gradually increased and reached 13% after 90 min. The difference in the initial activity of each catalyst is related to the specific surface area [25]. The decrease in activity at the initial stage of the reaction and the subsequent stabilization of the activity can be explained by the decrease in the specific surface area due to the deposition of produced carbon by the reaction. Fig. 4 shows the change in specific surface area versus elapsed time. Although AC had the highest surface area, the specific surface area decreased earlier than the other carbons. It is also found that the decrease in specific surface area of CB is slow. These
9
results are similar to the decreasing trend of the methane conversion ratio in Fig. 3 (a). The specific surface area of AC and MC with pore structure sharply decreased at the initial stage of reaction. This result seems to be due to the fact that the produced carbon by the reaction blocks the pores [21]. This also explains that the reduction of the specific surface area of AC having micropores is remarkable compared with MC. It can be seen that the specific surface area of carbon when the methane conversion ratio is sufficiently stabilized around 17% in Fig. 3 (a) shows the same value of around 10 m2/g regardless of the structures of the carbon catalysts. It is considered that the catalyst before the experiment is completely covered with the produced carbon by the reaction and the produced carbon itself continues to show a methane conversion ratio of about 17% as catalyst. On the other hand, there are also phenomena that cannot be explained only by the specific surface area. For example, as shown in Fig. 3 (b), although the specific surface area of AC is about 1.5 times higher than that of CB, the difference in the methane conversion ratio after five minutes of reaction is only about 9.6%. Furthermore, MC has a higher specific surface area but a lower methane conversion ratio than CB. From these facts, it is suggested that the specific surface area is not the only factor determining catalytic activity. The detailed relationship between catalytic activity and the structure
10
of carbon is discussed in Discussion. To investigate CB which showed the longest activity, the reaction order and activation energy of CB before and after the experiment were measured. To determine the reaction order, the activity was measured at 1173 K under different methane partial pressures based on the following formula: (3) where
is the methane decomposition rate,
partial pressure of methane, and
is the rate constant,
is the reaction order. The
is the
was evaluated by
the average partial pressure between the inlet and outlet, because the
at the
reactor inlet varies at the outlet due to the decomposition reaction [26,27]. In order to avoid the influence of produced carbon by the reaction, the α was determined using the value after five minutes of reaction. The result is shown in Fig. 5 (a). The α of CB was determined to be 0.67 before the experiment and 0.63 after the experiment. By using the determined α, the activation energy
was evaluated. The rate constant
was
firstly calculated using the following formula for an integral reactor [28,29]: (4) where
is the feed of methane,
mass of catalysts, and
is the variable of integration, mcat is the initial
is the expansion factor. The temperature dependence of the
11
was calculated by the following equation: (4) where
is the pre-exponential factor, R is the universal gas constant, and
absolute temperature. The result is shown in Fig. 5 (b). The
is the
of CB before and after
the experiment for 700 min in this study obtained from the Arrhenius plot were 143.6 kJ/mol and 167.1 kJ/mol, respectively. These results indicate that the produced carbon by the reaction has lower catalytic activity than the original CB. On the other hand, it has been reported that the
of the carbon nuclei formation during thermal
decomposition of methane is 316.8 kJ/mol and that of the carbon crystallite growth is 227.1 kJ/mol [26,27]. Compared to these
, the
of original CB and CB after the
experiment for 700 min in CH4 decomposition are low. This suggests that although the decreases after the deposition of produced carbon, the produced carbon itself also acts as a catalyst and continues to decompose methane. Table 2 summarizes the carbons reported in the past for reference [20,25,27,29,30,31]. The
of
of the original
CB obtained in this study is close to the result of S. Y. Lee, et al. and E. K. Lee [27,30]. SEM micrographs of the carbons after the experiment are presented in Fig. 6. From the catalysts after the experiment, the structures of the original carbons as seen in Fig. 1 were not observed. As observed for all the catalysts in Fig. 6, the produced carbon by
12
the reaction was spherical in shape with a diameter of about 1 μm and covered the surface of the catalysts while sticking to each other.
4. Discussion Raman spectroscopic analysis was carried out to investigate the structure of produced carbon by the reaction. The samples used for the analysis were all tested at 1173 K under the conditions corresponding to the results in Fig. 3. The Raman spectra of the catalysts before and after the experiment are shown in Fig. 7. In general, the primary Raman spectrum of the carbonaceous material is characterized by the three peaks of the G band at 1580 cm-1, the D band at 1350 cm-1, and the G’ band (2D band) at 2700 cm-1. The G band is attributed to an ideal graphitic lattice vibration mode with E2g symmetry. The D band is attributed to reduction in symmetry due to defects in the graphite lattice. The intensity ratio of the G band and the D band (ID/IG) is used to evaluate the degree of graphitization of carbonaceous materials. Although the G’ band at 2700 cm-1 is an overtone of the D band at 1350 cm-1, it does not depend on the defect density. Since the G’ band is affected by the interaction between the graphene layers, it is a single peak in single-walled CNTs and monolayer graphene, but in the case of
13
graphite and multi-walled CNTs, it is a broad peak due to overlapping plural peaks [32,33]. Curve fitting of the G band and the D band was applied to determine ID/IG. In the case of amorphous carbon such as CB, there are the D2 band around 1188 cm-1, the D3 band around 1500 cm-1, and the D4 band around 1620 cm-1 in addition to the abovementioned three bands [34,35]. However, deconvolution to multiple bands may increase uncertainty. Therefore, in this study, based on the report of Kameya, et al. [29], the spectral data was evaluated using a three-band fitting of G band, D band, and D3 band. The temporal change of the intensity ratio ID/IG of the Raman spectrum is shown in Fig. 8 (a). Lower ID/IG means less graphite lattice defects. Carbon with many lattice defects has many graphite edges compared with fewer lattice defects. It is reported that the active site for methane decomposition is the graphite edge [36,37]. There are BAY, FISSURE, FRONT, COVE, FJORD in the defect site of the graphite lattice [38]. In particular, the cluster of BAY and FISSURE as shown in Fig. 9 are the major structures at the graphite edge. They are also referred to armchair and zigzag faces, respectively. The BAY cluster is more active sites than the FISSURE cluster, because two carbon atoms are needed in order to form a hexagonal ring on the BAY cluster, compared with
14
the necessity of three carbon atoms on the FISSURE cluster [38]. Moreover, it is known that the D band scattering is strongly observed in the BAY cluster, whereas in the FISSURE cluster, the D band scattering is forbidden [39]. Therefore, the ID/IG of each catalyst is closely related to the catalytic activity. From Fig. 8 (a), the ID/IG of original CNF is lower than the other catalysts and there is a possibility of correlation with low methane conversion. Furthermore, the ID/ IG of MC before the experiment was 1.3, lower than those of AC and CB. This result suggests that the defect density on the MC surface is lower than those on AC and CB or that many of the exposed graphite edges on the MC surface are FISSURE cluster. This result seems reasonable that, as shown in Fig. 3 (b), the methane conversion ratio of the MC was lower than that of CB although the MC after the experiment for five minutes had a higher specific surface area than CB. As shown in Fig. 10, a good linear relationship was obtained between the methane conversion ratio per unit surface area of carbons after the experiment for five minutes and ID/IG of the carbon catalysts before the experiment. From these results, we propose that Raman spectroscopy is an effective method for evaluating initial catalytic activity for methane decomposition. From Fig. 8 (a), the ID/IG of catalysts vary as the reaction progresses. When the methane conversion ratio stabilized at around 17%, the ID/IG of each catalyst was in the
15
range of 1.5 to 1.57. In 2009, Alfonso Reina, et al. discovered that IG/IG’ has a better correlation with the number of graphene layers than the spectral shape of the G' band [40]. We evaluated IG/IG’ of the produced carbon as shown in Fig. 8b. A similar trend was seen also in Fig. 8 (b) showing the temporal change of the intensity ratio IG/IG’ of the Raman spectrum. The range of the initial IG/IG’ was from 1.57 to 5.87 became eventually smaller from 2.01 to 3.24. This supports that the produced carbon in this study has eventually the same structure, despite the use of carbons of different structures.
5. Conclusions We conducted a methane decomposition experiment using four kinds of carbon materials with different structures as a catalyst. We found that all of these catalysts continued to maintain a methane conversion of about 17% for longer than 600 min by catalyzing the produced carbon. By analysis of the carbon catalysts before and after the experiment, we found that the produced carbon by the reaction covered the catalyst surface and resulted in a specific surface area of about 10 m2/g and ID/IG of about 1.5 to 1.57 irrespective of the original carbon structures. We proposed that Raman
16
spectroscopy is an effective method for evaluating initial catalytic activity of carbon catalysts for methane decomposition.
Acknowledgements A part of this work was supported by “Development project of near-term hydrogen economy forming technology” for priority of research project of Knowledge Hub Aichi. The authors would like to thank K. Asato for emeritus prof. of Gifu Univ. and H. Yamashita for emeritus prof. of Nagoya Univ. with discussion about correlation between produced carbon by the reaction and methane conversion.
References [1] Ministry of Economy, Trade and Industry, Revised Version of the Strategic Roadmap for Hydrogen and Fuel Cells [http://www.meti.go.jp/english/press/2016/0322_05.html]. [2] M. Nishikawa, Research issues for development of hydrogen energy system, Journal of Plasma and Fusion Research. 80 (2004) 5-11 [in Japanese]. [3] International Energy Agency, Hydrogen Production and Storage [https://www.iea.org/publications/freepublications/publication/hydrogen-production-and -storage.html]
17
[4] F. Besenbacher, I. Chorkendorff, B. S. Clausen, B. Hammer, A. M. Molenbroek, J. K. Norskov, I. Stensgaard, Design of a surface alloy catalyst for steam reforming, Science 279 (1998) 1913-1915. [5] B. C. Enger, R. Lodeng, A. Holmen, A review of catalytic partial oxidation of methane to synthesis gas with emphasis on reaction mechanisms over transition metal catalysts, Applied Catalysis A: General 346 (2008) 1-27. [6] S. Ayabe, H. Omoto, T. Utaka, R. Kikuchi, K. Sasaki, Y. Teraoka, K. Eguchi, Catalytic autothermal reforming of methane and propane over supported metal catalysts, Applied Catalysis A: General 241 (2003) 261-269. [7] H. F. Abbas, W. Daud, Hydrogen production by methane decomposition : A review, International Journal of Hydrogen Energy 35 (2010) 1160-1190. [8] Yongdan Li, Douxing Li, Gaowei Wang, Methane decomposition to CO x-free hydrogen and nano-carbon material on group 8-10 base metal catalysts: A review, Catalysis Today 162 (2011) 1-48. [9] H. Y. Wang, A. C. Lua, Deactivation and kinetic studies of unsupported Ni and Ni-Co-Cu alloy catalysts used for hydrogen production by methane decomposition, Chemical Engineering Journal 243 (2014) 79-91. [10] Dan Ping, Chaoxian Wang, Xinfa Dong, Yingchao Dong, Co-production of hydrogen and carbon nanotubes on nickel foam via methane catalytic decomposition, Applied Surface Science 369 (2016) 299–307. [11] A. A. Ibrahim, A. H. Fakeeha, A. S. Al-Fatesh, A. E. Abasaeed, W. U. Khan, Methane decomposition over iron catalyst for hydrogen production, International Journal of Hydrogen Energy 40 (2015) 7593-7600. [12] Manoj Pudukudy, Zahira Yaakob, Zubair Shamsul Akmal, Direct decomposition of methane over SBA-15 supported Ni, Co andFe based bimetallic catalysts, Applied Surface Science 330 (2015) 418–430. [13] Lu Zhou, Linga Reddy Enakonda, Moussab Harb, Youssef Saih, Antonio 18
Aguilar-Tapia, Samy Ould-Chikh, Jean-louis Hazemann, Jun Li, Nini Wei, Daniel Gary, Pascal Del-Gallo, Jean-Marie Basset, Fe catalysts for methane decomposition to produce hydrogen and carbon nano materials, Applied Catalysis B: Environmental 208 (2017) 44-59. [14] D. Torres, J.L. Pinilla, I. Suelves, Screening of Ni-Cu bimetallic catalysts for hydrogen and carbon nanofilaments production via catalytic decomposition of methane, Applied Catalysis A: General 559 (2018) 10-19. [15] Yi Shen, Aik Chong Lua, A trimodal porous carbon as an effective catalyst for hydrogen production by methane decomposition, Journal of Colloid and Interface Science 462 (2016) 48-55. [16] Marwa Mahmoudi, Joseph Dentzer, Roger Gadiou, Abdelmottaleb Ouederni, Evaluation of activated carbons based on olive stones as catalysts during hydrogen production by thermocatalytic decomposition of methane, International Journal of Hydrogen Energy 42 (2017) 8712-8720. [17] Jianbo Zhang, Xing Li, Huiyong Chen, Meng Qi, Guorong Zhang, Haoquan Hu, Xiaoxun Ma, Hydrogen production by catalytic methane decomposition: Carbon materials as catalysts or catalyst supports, International Journal of Hydrogen Energy 42 (2017) 19755-19775. [18] B. Gaudernack, S. Lynum, Hydrogen from natural gas without release of CO 2 to the atmosphere, International Journal of Hydrogen Energy 23 (1998) 1087-1093. [19] R. Guil-Lopez, J. A. Botas, J. L. G. Fierro, D. P. Serrano, Comparison of metal and carbon catalysts for hydrogen production by methane decomposition, Applied Catalysis A: General 396 (2011) 40-51. [20] M. H. Kim, E. K. Lee, J. H. Jun, S. J. Kong, G. Y. Han, B. K. Lee, T. J. Lee, K. J. Yoon, Hydrogen production by catalytic decomposition of methane over activated carbons: kinetic study, International Journal of Hydrogen Energy 29 (2004) 187-193. [21] S. Kryzyzynski, M. Kozlowski, Activated carbons as catalysts for hydrogen production via methane decomposition, International Journal of Hydrogen Energy 33 19
(2008) 6172-6177. [22] J. Ashok, S. N. Kumar, A. Venugopal, F. D. Kumari, S. Tripathi, M. Subrahmanyam, COx free hydrogen by methane decomposition over activated carbons, Catalysis Communications 9 (2008) 164-169. [23] M. J. Lazaro, J. L. Pinilla, I. Suelves, R. Moliner, Study of deactivation mechanism of carbon blacks used in methane decomposition, International Journal of Hydrogen Energy 33 (2008) 4104-4111. [24] I. Suelves, J. L. Pinilla, M. J. Lazaro, R. Moliner, Carbonaceous materials as catalysts for decomposition of methane, Chemical Engineering Journal 140 (2008) 432-438. [25] N. Muradov, F. Smith, A. and T-Raissi, Catalytic activity of carbons for methane decomposition reaction, Catalysis Today 102-103 (2012) 225-233. [26] N. Muradov, Catalysis of methane decomposition over elemental carbon, Catalysis Communications 2 (2001) 89-94. [27] E. K. Lee, S. Y. Lee, G. Y. Han, B. K. Lee, T. J. Lee, J. H. Jun, K. H. Yoon, Catalytic decomposition of methane over carbon blacks for CO 2-free hydrogen production, Carbon 42 (2004) 2641-2648. [28] O. Levenspiel, Chemical Reaction Engineering, 3rd ed. p. 111 (1972). [29] Y. Kameya, K. Hanamura, Kinetic and raman spectroscopic study on catalytic characteristics of carbon blacks in methane decomposition, Chemical Engineering Journal 173 (2011) 627-635. [30] S. Y. Lee, B. H. Ryu, G. Y. Han, T. J. Lee, K. J. Yoon, Catalytic characteristics of specialty carbon blacks in decomposition of methane for hydrogen production, Carbon 46 (2008) 1978-1986. [31] J. L. Pinilla, I. Suelves, M. J. Lazaro, R. Moliner, Kinetic study of the thermal decomposition of methane using carbonaceous catalysts, Chemical Engineering Journal 20
138 (2008) 301-306. [32] L. M. Malard, M. A. Pimenta, G. Dresselhaus, M. S. Dresselhaus, Raman spectroscopy in graphene, Physics Reports 473 (2009) 51-87. [33] S. Ijima, M. Endo, Carbon nanotube graphen handbook, p. 174 (2011) [in Japanese]. [34] L. Bokobza, J. L. Bruneel, M. Couzi, Raman spectroscopic investigation of carbon-based materials and their composites. Comparison between carbon nanotubes and carbon black, Chemical Physics Letters 590 (2013) 153-159. [35] M. Pawlyta, J. N. Rouzaud, S. Duber, Raman microspectroscopy characterization of carbon blacks: Spectral analysis and structural information, Carbon 84 (2015) 479-490. [36] S. Y. Lee, J. H. Kwak, G. Y. Han, T. J. Lee, K. J. Yoon, Characterization of active sites for methane decomposition on carbon black through acetylene chemisorption, Carbon 46 (2008) 342-348. [37] T. Tokunaga, N. Kishi, K. Yamakawa, K. Sasaki, T. Yamamoto, Methane decomposition for hydrogen production by catalytic activity of carbon black under low flow rate condirions, Journal of the Ceramic Society of Japan 125 (2017) 185-189. [38] S. Wauters, G. B. Marin, Computer generation of a network of elementary steps for coke formation during the thermal cracking of hydrocarbons, Chemical Engineering Journal 82 (2001) 267-279. [39] A. Jorio, M. Dresselhaus, R. Saito, G. F. Dresselhaus, Raman Spectroscopy in Graphene Related Systems, p. 332-333 (2012). [40] A. Reina, X. Jia, J. Ho, D. Nezich, H. Son, V. Bulovic, M. S. Dresselhaus, J. Kong, Large Area, Few-Layer Graphene Films on Arbitrary Substrates by Chemical Vapor Deposition, Nano Letters 9 (2009) 30-35.
21
Methane decomposition using carbon catalyst for CO2–free hydrogen production The carbon produced by methane decomposition also acts as a catalyst Raman spectroscopy is an effective method for evaluating initial catalytic activity
22
Figure 1. SEM micrographs of carbons before the experiment. (a) AC, (b) MC, (c) CB, and (d) CNF.
23
Hydrogen concentration (%)
100
PD-12 GC-2014
80 60 40 20 0 0
20
40
60
80
100
Elapsed time (min) Figure 2. Temporal changes of methane conversion ratio over CB measured by PD-12 and GC-2014.
24
Methane conversion ratio (%)
80
(a)
AC MC CB CNF
70 60 50 40 30 20 10 0 0
100
200
300
400
500
600
Elapsed time (min)
Methane conversion ratio (%)
100
(b) CB 2 -1 (1520 m g )
80
AC 2 -1 (2208 m g )
60
MC 2 -1 (1700 m g )
40 CNF 2 -1 (270 m g )
20 0 0
500
1000
1500
2
2000 -1
2500
Specific surface area (m g ) Figure 3. (a) Temporal change of methane conversion ratio at T = 1173 K and SV = 360 h-1 and (b) methane conversion ratio at five minutes of reaction vs. specific surface area at T = 1173 K and SV = 360 h-1.
25
2
Specific surface area (m /g)
10
4
10
3
10
2
10
1
10
0
AC MC CB CNF
0
100
200
300
400
500
600
700
800
Elapsed time (min) Figure 4. Temporal change of specific surface area by methane decomposition at T = 1173 K and SV = 360 h-1.
26
(a)
1
CB
0.1
4
rCH (mmol/min gcat)
10
Produced carbon (over CB)
0.01 0.1
1
PCH
4
10
kp (mmol/min gcat atm)
(b)
CB
1
Produced carbon (over CB) 0.1
0.01 0.84
0.86
0.88
0.90
0.92
0.94
-1
1/T * 1000 (K ) Figure 5. (a) Methane decomposition rate over CB as a function of methane partial pressure at T = 1173 K and (b) Arrhenius plot for methane decomposition over CB at SV = 360 h-1.
27
Figure 6. SEM micrographs and their magnified images of carbons after the experiment at T = 1173 K and SV = 360 h-1. (a) AC, (b) MC, (c) CB, and (d) CNF.
28
(a)
Intensity (a.u.)
CNF
CB MC
AC
0
1000
2000
3000
4000
-1
Raman shift (cm )
(b)
Intensity (a.u.)
CNF at 640 min CB at 700 min
MC at 640 min
AC at 640 min
0
1000
2000
3000
4000
-1
Raman shift (cm ) Figure 7. Raman spectra of (a) original carbons and (b) carbons after the experiment at T = 1173 K and SV = 360 h-1.
29
2.0
(a)
ID/IG (a.u.)
1.5
1.0 AC MC CB CNF
0.5
0.0 0
100
200
300 400 500 600 Elapsed time (min)
700
800
8
(b)
AC MC CB CNF
IG/IG' (a.u.)
6
4
2
0 0
100
200
300 400 500 600 Elapsed time (min)
700
800
Figure 8. Time variation of spectral parameters of Raman spectra at T = 1173 K and SV = 360 h-1 of (a) peak intensity ratio of the D band to the G band (ID/IG), (b) peak intensity ratio of the G band to the G’ band (IG/IG’).
30
SURE
BAY
FISSURE
Figure 9. BAY and FISSURE clusters at the coke surface [38].
31
BAY
Methane conversion ratio 3 2 per unit surface area *10 (g/m )
0.3
CB AC 0.2
MC
0.1
CNF 0.0 0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
ID/IG (a.u.) Figure 10. Methane conversion ratio per unit surface area after the experiment for 5 min at T = 1173 K and SV = 360 h-1 vs. ID/IG of original catalyst.
32
Table
Table 1. Tested carbon catalysts. Use
BET surface area (m2/g)
Bulk density (g/cm3)
Pore diameter (nm)
UESb TOYO TANSO
2208 1700
0.21 0.07
1.8 5.6
6.5 5
Cabot Showa Denko
1520 270
0.17 0.09
― ―
2.2×10-3 ―
Name
Manufacturer
Activated carbon (AC) Mesoporous carbon (MC)
UCG-CPT P (3)010
Carbon black (CB) Carbon nanofiber (CNF)
BP2000a VGCF-H
a BP = black pearls. b UES = Ueda environmental solution Co., Ltd.
Particle size (μm)
Table 2. Activation energy of carbons.
Author
Catalysts
Name
Measurement time (min)
Specific surface area (m2/g)
Temperature range (K)
This study
CB
BP-2000
After 5 min
1520
1073–1173
0.67
144
E. K. Lee E. K. Lee S. Y. Lee S. Y. Lee S. Y. Lee S. Y. Lee S. Y. Lee
CB CB CB CB CB CB CB
DCC-N330 BP-2000 HI-170 HI-20L HI-900L EC-600JD BP-2000
20–30 20–30 20–60 20–60 20–60 20–60 20–60
79 1475 23 86 300 84 240
1123–1273 1023–1273 1123–1223 1123–1223 1123–1223 1123–1223 1023–1223
0.919 0.984 0.72 0.86 1.04 0.71 0.84
183 143 185 181 178 135 143
S. Y. Lee S. Y. Lee
CB CB
DCC-N774 DCC-N330
20–60 20–60
1270 1500
1123–1223 1123–1223
0.69 0.71
233 213
J. L. Pinilla Y. Kameya Y. Kameya Y. Kameya
CB CB CB CB
BP-2000 SB285 SB905 SB285
Initial reaction After 60 min After 60 min After 60 min
1337 81 212 81
973–1223 1013–1113 1013–1113 1133–1233
0.6 0.59 0.62 0.59
238 221 179 192
Y. Kameya N. Muradov J. L. Pinilla M. H. kim M. H. kim N. Muradov
CB CB AC AC AC AC
SB905 BP-2000 CG-Norit CCN-DR CL-SCR KBB
After 60 min Initial reaction Initial reaction Initial reaction Initial reaction Initial reaction
212 1650 1300 725 912 1500
1133–1233 973–1173 973–1223 923–1173 923–1173 973–1173
0.62 0.5±0.1 0.48 0.51 0.49 0.6±0.1
157 205–236 141 194 186 160–201
Reaction order
Activation energy (kJ/mol)
S0169-4332(18)33403-2 https://doi.org/10.1016/j.apsusc.2018.12.073 APSUSC 41170
To appear in:
Applied Surface Science
Received Date: Revised Date: Accepted Date:
9 May 2018 29 August 2018 8 December 2018
Please cite this article as: H. Nishii, D. Miyamoto, Y. Umeda, H. Hamaguchi, M. Suzuki, T. Tanimoto, T. Harigai, H. Takikawa, Y. Suda, Catalytic Activity of Several Carbons with Different Structures for Methane Decomposition and By-Produced Carbons, Applied Surface Science (2018), doi: https://doi.org/10.1016/j.apsusc.2018.12.073
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.
Catalytic Activity of Several Carbons with Different Structures for Methane Decomposition and By-Produced Carbons
Haruki Nishii1, Dai Miyamoto1, Yoshito Umeda2, Hiroaki Hamaguchi3, Masashi Suzuki3, Tsuyoshi Tanimoto1, Toru Harigai1, Hirofumi Takikawa1, Yoshiyuki Suda1*
1
Department of Electrical and Electronic Information Engineering, Toyohashi University of Technology, Toyohashi, Aichi 441-8580, Japan 2
3
Toho Cryogenics Company Limited, Chita, Aichi 456-0004, Japan
Aichi Center for Industry and Science Technology, Kariya, Aichi 470-0536, Japan E-mail: [email protected]
1
Abstract Since methane decomposition has no CO2 emissions, it is attracting attention as a hydrogen production method with a low environmental burden. Understanding the structure of the produced carbon is important for long-term stable production of hydrogen. In this study, methane decomposition was carried out on carbons with several different structures (activated carbon (AC), carbon black (CB), meso-porous carbon (MC), and carbon nanofiber (CNF)). We have found that the carbon produced by methane decomposition decreases activity by covering the catalyst, but itself also acts as a catalyst irrespective of the original carbon catalysts. All of the catalysts continued to maintain a methane conversion ratio of about 17% by catalyzing the produced carbon even after the activity was lowered. By analysis of the catalysts before and after the experiment, it was shown that the produced carbon covered the catalyst surface and resulted in a specific surface area of about 10 m2/g and the intensity ratio of the D band to the G band in the Raman spectra (ID/IG) of around 1.55 irrespective of the original carbons structures. We proposed that Raman spectroscopy is an effective method for evaluating initial catalytic activity for methane decomposition because ID/IG of the catalysts before the experiment have a linear relationship with the methane conversion ratio per unit surface area in the early stage of the reaction.
2
KEYWORDS: methane decomposition, Raman spectroscopy, activation energy, hydrogen production, carbon black, activated carbon.
1. Introduction Hydrogen utilization technology including fuel cells that are characterized by high energy efficiency and no CO2 emission at the utilization stage, can contribute to energy saving and environmental burden reduction. Furthermore, the market size of hydrogen related technology is predicted to be about 1 trillion Japanese Yen by 2030 and about 8 trillion Japanese Yen by 2050 in Japan, making it an important technology. On the other hand, there are many problems in hydrogen utilization: safety, cost, and infrastructure [1,2]. There is a possibility that part of the problem can be solved by on-site hydrogen production using natural gas. By using an existing natural gas supply line, hydrogen transportation is safe and of low cost because it does not require the direct transportation of hydrogen and the laying of new supply lines [3]. The main component of natural gas is methane. Various methods for hydrogen production using methane have been proposed: (a) steam methane reforming, (b) partial oxidation, and (c) autothermal reforming [4,5,6]. However, since all of these hydrogen
3
production methods involve the emission of CO 2, CCS (carbon dioxide capture and storage) is required, which leads to a cost increase. Therefore, methane decomposition without CO and CO2 emission attracts attention as a promising hydrogen production method. Methane is decomposed into carbon and hydrogen according to the following reaction [7]: .
(1)
Various noble metals such as Ni, Cu, and Fe have been used as catalysts for methane decomposition [8,9,10,11,12,13,14]. However, when these noble metals are used as a catalyst, these catalysts are covered by the deposition of carbon produced by the process, so that a regeneration process accompanied by CO2 emission by carbon combustion is required. In recent years, the use of carbons as a catalyst with a lower cost than metal catalysts have attracted attention [15,16,17]. Compared to metal catalysts, carbon catalysts have advantages such as high temperature tolerance, resistance to impurities such as sulfur, and possibility of further cost reduction due to commercial use of high purity carbon [18,19]. Various carbons have been used as catalysts for methane decomposition. Kim, et al.,
4
Kryzynski, et al., and Ashok, et al. reported that AC showed high catalytic activity in the early stage of the reaction, but that the pores on the surface were blocked by produced carbon by methane decomposition and the activity was immediately decreased [20,21,22]. Lazao, et al. and Suelves, et al. reported that CB showed stable catalytic activity for a long time, but the activity decreased when a certain amount of carbon deposits [23,24]. Muradov, et al. evaluated catalytic activity of a wide range of carbon materials for methane decomposition reaction and determine major factors governing their activity. They tested carbons with different structures such as activated carbon (AC), carbon black (CB), carbon nanotubes (CNTs), and glassy carbon and reported that the methane decomposition rate in the early stage of reaction was related to the surface area and that AC and CB were promising catalysts [25]. There is a common point in these reports. As produced carbon by the reaction covered the surface of the catalyst, the activity decreased. On the other hand, the methane conversion ratio continued to show a constant value even after activity was lowered. This fact suggests that the carbon itself produced by the methane decomposition acts as a catalyst. There are few reports on the catalytic activity after the decrease in activity. Especially, the relationship between the structure of the catalyst and the activity of produced carbon by the reaction is not known. If carbon having high catalytic activity can produce, hydrogen can be produced stably
5
for a long period of time. In this study, methane decomposition was carried out on carbons with several different structures including carbon nanofiber (CNF) with high crystallinity, AC, CB, and meso-porous carbon (MC), and the activity of produced carbon by the reaction was investigated. Furthermore, the relationship between the surface structure of carbon and catalytic activity was discussed by analyzing the N2 Brunauer-Emmett-Teller (BET) surface area and Raman spectra.
2. Experimental The carbons tested in this study are listed in Table 1. The BET specific surface areas of the catalysts ranged from 270 to 2208 m2/g. AC and MC are carbons having a pore structure. Most of the pores on the surface of the AC are micropores of 2 nm or less. MC controls surface pores in mesopores of 2 to 50 nm. CB has a structure in which fine particles with a diameter of several nm are aggregated. CNF is a fibrous carbon with a diameter of several nm. SEM micrographs of the carbons before the experiment are presented in Fig. 1. Methane (99.9%) and nitrogen were used without further purification. The methane decomposition was carried out in a horizontal, fixed-bed, 40 mm I.D. quartz-tube flow
6
reactor heated by an electric tube furnace. Quartz wool was used to fix the catalyst. The standard reaction conditions were a methane flow rate in range of 50 to 135 cm3 (under standard temperature and pressure, STP)/min, a catalyst weight of 1.7 g, a space velocity (SV, the amount of gas per hour passing through the reactor divided by the volume of the catalyst in the reactor) of 360 h-1, and the reaction temperature in the range of 1073 to 1173 K. The product gas was analyzed by a hydrogen detector (PD-12, New Cosmos Electric Co., Ltd., Osaka, Japan) and gas chromatography (GC-2014, Shimadzu Corp., Kyoto, Japan), using a SHINCARBON ST (ZT-13, Shinwa Chemical Industries, Ltd., Kyoto, Japan), Ar carrier, and a thermal conductivity detector. The hydrogen detector can continuously measure the hydrogen concentration as compared with gas chromatography, but the initial response is slow (It takes about 20 minutes to accurately measure hydrogen concentration). On the other hand, gas chromatography can measure hydrogen concentration at the initial reaction, but it is measured at intervals of 20 minutes because the retention time of methane is about 15 minutes. Fig. 2 shows the results of the methane decomposition test in which the hydrogen concentration was measured using PD-12 and GC-2014. Except for 30 minutes after the reaction, it was confirmed that the error in hydrogen concentration due to the difference in analyzer used was within 5%.
7
The methane conversion ratio was derived from the following formula using the hydrogen concentration obtained by analysis. (2) In order to clarify the effect of the catalyst, the non-catalytic methane conversion ratio was subtracted from the methane conversion ratio over the catalysts. The non-catalytic methane conversion after reaction for five minutes was 8% at SV = 360 h-1 and T = 1173 K. Carbons before and after the experiment was analyzed by N2 adsorption measurement (BELSORP-max, MicrotracBEL Corp., Osaka, Japan), scanning electron microscopy (SEM, SU8000, Hitachi High-Technologies Corp., Tokyo, Japan), and Raman spectroscopic analysis (NRS-1000, JASCO Corp., Tokyo, Japan).
3. Results The catalytic activity of carbons with different structures at 1173 K were compared and the results are shown in Fig. 3. Temporal change of methane conversion ratio is shown in Fig. 3 (a) at SV = 360 h-1 and methane conversion ratio after five minutes of reaction to the specific surface area is shown in Fig. 3 (b). Methane conversion ratio of
8
all the tested catalysts eventually reached about 17%. In addition, all the catalysts except CNF showed the highest methane conversion ratio at the initial reaction. Particularly in AC, the methane conversion after five minutes of reaction was the highest at 77%, but the decrease in activity was also the fastest among all the carbons. Although MC showed the same tendency as AC, the initial activity was 63%, which was lower than AC and the decrease in activity was also slow. CB showed an initial methane conversion ratio lower than AC, but maintained 32% up to 300 min. After that, the methane conversion ratio decreased, and eventually reached a stable state again at 17% after about 600 min. On the other hand, CNF showed a tendency different from the other carbons. Although the initial methane conversion ratio of CNF was 3%, which is lower than that of other catalysts, the methane conversion ratio gradually increased and reached 13% after 90 min. The difference in the initial activity of each catalyst is related to the specific surface area [25]. The decrease in activity at the initial stage of the reaction and the subsequent stabilization of the activity can be explained by the decrease in the specific surface area due to the deposition of produced carbon by the reaction. Fig. 4 shows the change in specific surface area versus elapsed time. Although AC had the highest surface area, the specific surface area decreased earlier than the other carbons. It is also found that the decrease in specific surface area of CB is slow. These
9
results are similar to the decreasing trend of the methane conversion ratio in Fig. 3 (a). The specific surface area of AC and MC with pore structure sharply decreased at the initial stage of reaction. This result seems to be due to the fact that the produced carbon by the reaction blocks the pores [21]. This also explains that the reduction of the specific surface area of AC having micropores is remarkable compared with MC. It can be seen that the specific surface area of carbon when the methane conversion ratio is sufficiently stabilized around 17% in Fig. 3 (a) shows the same value of around 10 m2/g regardless of the structures of the carbon catalysts. It is considered that the catalyst before the experiment is completely covered with the produced carbon by the reaction and the produced carbon itself continues to show a methane conversion ratio of about 17% as catalyst. On the other hand, there are also phenomena that cannot be explained only by the specific surface area. For example, as shown in Fig. 3 (b), although the specific surface area of AC is about 1.5 times higher than that of CB, the difference in the methane conversion ratio after five minutes of reaction is only about 9.6%. Furthermore, MC has a higher specific surface area but a lower methane conversion ratio than CB. From these facts, it is suggested that the specific surface area is not the only factor determining catalytic activity. The detailed relationship between catalytic activity and the structure
10
of carbon is discussed in Discussion. To investigate CB which showed the longest activity, the reaction order and activation energy of CB before and after the experiment were measured. To determine the reaction order, the activity was measured at 1173 K under different methane partial pressures based on the following formula: (3) where
is the methane decomposition rate,
partial pressure of methane, and
is the rate constant,
is the reaction order. The
is the
was evaluated by
the average partial pressure between the inlet and outlet, because the
at the
reactor inlet varies at the outlet due to the decomposition reaction [26,27]. In order to avoid the influence of produced carbon by the reaction, the α was determined using the value after five minutes of reaction. The result is shown in Fig. 5 (a). The α of CB was determined to be 0.67 before the experiment and 0.63 after the experiment. By using the determined α, the activation energy
was evaluated. The rate constant
was
firstly calculated using the following formula for an integral reactor [28,29]: (4) where
is the feed of methane,
mass of catalysts, and
is the variable of integration, mcat is the initial
is the expansion factor. The temperature dependence of the
11
was calculated by the following equation: (4) where
is the pre-exponential factor, R is the universal gas constant, and
absolute temperature. The result is shown in Fig. 5 (b). The
is the
of CB before and after
the experiment for 700 min in this study obtained from the Arrhenius plot were 143.6 kJ/mol and 167.1 kJ/mol, respectively. These results indicate that the produced carbon by the reaction has lower catalytic activity than the original CB. On the other hand, it has been reported that the
of the carbon nuclei formation during thermal
decomposition of methane is 316.8 kJ/mol and that of the carbon crystallite growth is 227.1 kJ/mol [26,27]. Compared to these
, the
of original CB and CB after the
experiment for 700 min in CH4 decomposition are low. This suggests that although the decreases after the deposition of produced carbon, the produced carbon itself also acts as a catalyst and continues to decompose methane. Table 2 summarizes the carbons reported in the past for reference [20,25,27,29,30,31]. The
of
of the original
CB obtained in this study is close to the result of S. Y. Lee, et al. and E. K. Lee [27,30]. SEM micrographs of the carbons after the experiment are presented in Fig. 6. From the catalysts after the experiment, the structures of the original carbons as seen in Fig. 1 were not observed. As observed for all the catalysts in Fig. 6, the produced carbon by
12
the reaction was spherical in shape with a diameter of about 1 μm and covered the surface of the catalysts while sticking to each other.
4. Discussion Raman spectroscopic analysis was carried out to investigate the structure of produced carbon by the reaction. The samples used for the analysis were all tested at 1173 K under the conditions corresponding to the results in Fig. 3. The Raman spectra of the catalysts before and after the experiment are shown in Fig. 7. In general, the primary Raman spectrum of the carbonaceous material is characterized by the three peaks of the G band at 1580 cm-1, the D band at 1350 cm-1, and the G’ band (2D band) at 2700 cm-1. The G band is attributed to an ideal graphitic lattice vibration mode with E2g symmetry. The D band is attributed to reduction in symmetry due to defects in the graphite lattice. The intensity ratio of the G band and the D band (ID/IG) is used to evaluate the degree of graphitization of carbonaceous materials. Although the G’ band at 2700 cm-1 is an overtone of the D band at 1350 cm-1, it does not depend on the defect density. Since the G’ band is affected by the interaction between the graphene layers, it is a single peak in single-walled CNTs and monolayer graphene, but in the case of
13
graphite and multi-walled CNTs, it is a broad peak due to overlapping plural peaks [32,33]. Curve fitting of the G band and the D band was applied to determine ID/IG. In the case of amorphous carbon such as CB, there are the D2 band around 1188 cm-1, the D3 band around 1500 cm-1, and the D4 band around 1620 cm-1 in addition to the abovementioned three bands [34,35]. However, deconvolution to multiple bands may increase uncertainty. Therefore, in this study, based on the report of Kameya, et al. [29], the spectral data was evaluated using a three-band fitting of G band, D band, and D3 band. The temporal change of the intensity ratio ID/IG of the Raman spectrum is shown in Fig. 8 (a). Lower ID/IG means less graphite lattice defects. Carbon with many lattice defects has many graphite edges compared with fewer lattice defects. It is reported that the active site for methane decomposition is the graphite edge [36,37]. There are BAY, FISSURE, FRONT, COVE, FJORD in the defect site of the graphite lattice [38]. In particular, the cluster of BAY and FISSURE as shown in Fig. 9 are the major structures at the graphite edge. They are also referred to armchair and zigzag faces, respectively. The BAY cluster is more active sites than the FISSURE cluster, because two carbon atoms are needed in order to form a hexagonal ring on the BAY cluster, compared with
14
the necessity of three carbon atoms on the FISSURE cluster [38]. Moreover, it is known that the D band scattering is strongly observed in the BAY cluster, whereas in the FISSURE cluster, the D band scattering is forbidden [39]. Therefore, the ID/IG of each catalyst is closely related to the catalytic activity. From Fig. 8 (a), the ID/IG of original CNF is lower than the other catalysts and there is a possibility of correlation with low methane conversion. Furthermore, the ID/ IG of MC before the experiment was 1.3, lower than those of AC and CB. This result suggests that the defect density on the MC surface is lower than those on AC and CB or that many of the exposed graphite edges on the MC surface are FISSURE cluster. This result seems reasonable that, as shown in Fig. 3 (b), the methane conversion ratio of the MC was lower than that of CB although the MC after the experiment for five minutes had a higher specific surface area than CB. As shown in Fig. 10, a good linear relationship was obtained between the methane conversion ratio per unit surface area of carbons after the experiment for five minutes and ID/IG of the carbon catalysts before the experiment. From these results, we propose that Raman spectroscopy is an effective method for evaluating initial catalytic activity for methane decomposition. From Fig. 8 (a), the ID/IG of catalysts vary as the reaction progresses. When the methane conversion ratio stabilized at around 17%, the ID/IG of each catalyst was in the
15
range of 1.5 to 1.57. In 2009, Alfonso Reina, et al. discovered that IG/IG’ has a better correlation with the number of graphene layers than the spectral shape of the G' band [40]. We evaluated IG/IG’ of the produced carbon as shown in Fig. 8b. A similar trend was seen also in Fig. 8 (b) showing the temporal change of the intensity ratio IG/IG’ of the Raman spectrum. The range of the initial IG/IG’ was from 1.57 to 5.87 became eventually smaller from 2.01 to 3.24. This supports that the produced carbon in this study has eventually the same structure, despite the use of carbons of different structures.
5. Conclusions We conducted a methane decomposition experiment using four kinds of carbon materials with different structures as a catalyst. We found that all of these catalysts continued to maintain a methane conversion of about 17% for longer than 600 min by catalyzing the produced carbon. By analysis of the carbon catalysts before and after the experiment, we found that the produced carbon by the reaction covered the catalyst surface and resulted in a specific surface area of about 10 m2/g and ID/IG of about 1.5 to 1.57 irrespective of the original carbon structures. We proposed that Raman
16
spectroscopy is an effective method for evaluating initial catalytic activity of carbon catalysts for methane decomposition.
Acknowledgements A part of this work was supported by “Development project of near-term hydrogen economy forming technology” for priority of research project of Knowledge Hub Aichi. The authors would like to thank K. Asato for emeritus prof. of Gifu Univ. and H. Yamashita for emeritus prof. of Nagoya Univ. with discussion about correlation between produced carbon by the reaction and methane conversion.
References [1] Ministry of Economy, Trade and Industry, Revised Version of the Strategic Roadmap for Hydrogen and Fuel Cells [http://www.meti.go.jp/english/press/2016/0322_05.html]. [2] M. Nishikawa, Research issues for development of hydrogen energy system, Journal of Plasma and Fusion Research. 80 (2004) 5-11 [in Japanese]. [3] International Energy Agency, Hydrogen Production and Storage [https://www.iea.org/publications/freepublications/publication/hydrogen-production-and -storage.html]
17
[4] F. Besenbacher, I. Chorkendorff, B. S. Clausen, B. Hammer, A. M. Molenbroek, J. K. Norskov, I. Stensgaard, Design of a surface alloy catalyst for steam reforming, Science 279 (1998) 1913-1915. [5] B. C. Enger, R. Lodeng, A. Holmen, A review of catalytic partial oxidation of methane to synthesis gas with emphasis on reaction mechanisms over transition metal catalysts, Applied Catalysis A: General 346 (2008) 1-27. [6] S. Ayabe, H. Omoto, T. Utaka, R. Kikuchi, K. Sasaki, Y. Teraoka, K. Eguchi, Catalytic autothermal reforming of methane and propane over supported metal catalysts, Applied Catalysis A: General 241 (2003) 261-269. [7] H. F. Abbas, W. Daud, Hydrogen production by methane decomposition : A review, International Journal of Hydrogen Energy 35 (2010) 1160-1190. [8] Yongdan Li, Douxing Li, Gaowei Wang, Methane decomposition to CO x-free hydrogen and nano-carbon material on group 8-10 base metal catalysts: A review, Catalysis Today 162 (2011) 1-48. [9] H. Y. Wang, A. C. Lua, Deactivation and kinetic studies of unsupported Ni and Ni-Co-Cu alloy catalysts used for hydrogen production by methane decomposition, Chemical Engineering Journal 243 (2014) 79-91. [10] Dan Ping, Chaoxian Wang, Xinfa Dong, Yingchao Dong, Co-production of hydrogen and carbon nanotubes on nickel foam via methane catalytic decomposition, Applied Surface Science 369 (2016) 299–307. [11] A. A. Ibrahim, A. H. Fakeeha, A. S. Al-Fatesh, A. E. Abasaeed, W. U. Khan, Methane decomposition over iron catalyst for hydrogen production, International Journal of Hydrogen Energy 40 (2015) 7593-7600. [12] Manoj Pudukudy, Zahira Yaakob, Zubair Shamsul Akmal, Direct decomposition of methane over SBA-15 supported Ni, Co andFe based bimetallic catalysts, Applied Surface Science 330 (2015) 418–430. [13] Lu Zhou, Linga Reddy Enakonda, Moussab Harb, Youssef Saih, Antonio 18
Aguilar-Tapia, Samy Ould-Chikh, Jean-louis Hazemann, Jun Li, Nini Wei, Daniel Gary, Pascal Del-Gallo, Jean-Marie Basset, Fe catalysts for methane decomposition to produce hydrogen and carbon nano materials, Applied Catalysis B: Environmental 208 (2017) 44-59. [14] D. Torres, J.L. Pinilla, I. Suelves, Screening of Ni-Cu bimetallic catalysts for hydrogen and carbon nanofilaments production via catalytic decomposition of methane, Applied Catalysis A: General 559 (2018) 10-19. [15] Yi Shen, Aik Chong Lua, A trimodal porous carbon as an effective catalyst for hydrogen production by methane decomposition, Journal of Colloid and Interface Science 462 (2016) 48-55. [16] Marwa Mahmoudi, Joseph Dentzer, Roger Gadiou, Abdelmottaleb Ouederni, Evaluation of activated carbons based on olive stones as catalysts during hydrogen production by thermocatalytic decomposition of methane, International Journal of Hydrogen Energy 42 (2017) 8712-8720. [17] Jianbo Zhang, Xing Li, Huiyong Chen, Meng Qi, Guorong Zhang, Haoquan Hu, Xiaoxun Ma, Hydrogen production by catalytic methane decomposition: Carbon materials as catalysts or catalyst supports, International Journal of Hydrogen Energy 42 (2017) 19755-19775. [18] B. Gaudernack, S. Lynum, Hydrogen from natural gas without release of CO 2 to the atmosphere, International Journal of Hydrogen Energy 23 (1998) 1087-1093. [19] R. Guil-Lopez, J. A. Botas, J. L. G. Fierro, D. P. Serrano, Comparison of metal and carbon catalysts for hydrogen production by methane decomposition, Applied Catalysis A: General 396 (2011) 40-51. [20] M. H. Kim, E. K. Lee, J. H. Jun, S. J. Kong, G. Y. Han, B. K. Lee, T. J. Lee, K. J. Yoon, Hydrogen production by catalytic decomposition of methane over activated carbons: kinetic study, International Journal of Hydrogen Energy 29 (2004) 187-193. [21] S. Kryzyzynski, M. Kozlowski, Activated carbons as catalysts for hydrogen production via methane decomposition, International Journal of Hydrogen Energy 33 19
(2008) 6172-6177. [22] J. Ashok, S. N. Kumar, A. Venugopal, F. D. Kumari, S. Tripathi, M. Subrahmanyam, COx free hydrogen by methane decomposition over activated carbons, Catalysis Communications 9 (2008) 164-169. [23] M. J. Lazaro, J. L. Pinilla, I. Suelves, R. Moliner, Study of deactivation mechanism of carbon blacks used in methane decomposition, International Journal of Hydrogen Energy 33 (2008) 4104-4111. [24] I. Suelves, J. L. Pinilla, M. J. Lazaro, R. Moliner, Carbonaceous materials as catalysts for decomposition of methane, Chemical Engineering Journal 140 (2008) 432-438. [25] N. Muradov, F. Smith, A. and T-Raissi, Catalytic activity of carbons for methane decomposition reaction, Catalysis Today 102-103 (2012) 225-233. [26] N. Muradov, Catalysis of methane decomposition over elemental carbon, Catalysis Communications 2 (2001) 89-94. [27] E. K. Lee, S. Y. Lee, G. Y. Han, B. K. Lee, T. J. Lee, J. H. Jun, K. H. Yoon, Catalytic decomposition of methane over carbon blacks for CO 2-free hydrogen production, Carbon 42 (2004) 2641-2648. [28] O. Levenspiel, Chemical Reaction Engineering, 3rd ed. p. 111 (1972). [29] Y. Kameya, K. Hanamura, Kinetic and raman spectroscopic study on catalytic characteristics of carbon blacks in methane decomposition, Chemical Engineering Journal 173 (2011) 627-635. [30] S. Y. Lee, B. H. Ryu, G. Y. Han, T. J. Lee, K. J. Yoon, Catalytic characteristics of specialty carbon blacks in decomposition of methane for hydrogen production, Carbon 46 (2008) 1978-1986. [31] J. L. Pinilla, I. Suelves, M. J. Lazaro, R. Moliner, Kinetic study of the thermal decomposition of methane using carbonaceous catalysts, Chemical Engineering Journal 20
138 (2008) 301-306. [32] L. M. Malard, M. A. Pimenta, G. Dresselhaus, M. S. Dresselhaus, Raman spectroscopy in graphene, Physics Reports 473 (2009) 51-87. [33] S. Ijima, M. Endo, Carbon nanotube graphen handbook, p. 174 (2011) [in Japanese]. [34] L. Bokobza, J. L. Bruneel, M. Couzi, Raman spectroscopic investigation of carbon-based materials and their composites. Comparison between carbon nanotubes and carbon black, Chemical Physics Letters 590 (2013) 153-159. [35] M. Pawlyta, J. N. Rouzaud, S. Duber, Raman microspectroscopy characterization of carbon blacks: Spectral analysis and structural information, Carbon 84 (2015) 479-490. [36] S. Y. Lee, J. H. Kwak, G. Y. Han, T. J. Lee, K. J. Yoon, Characterization of active sites for methane decomposition on carbon black through acetylene chemisorption, Carbon 46 (2008) 342-348. [37] T. Tokunaga, N. Kishi, K. Yamakawa, K. Sasaki, T. Yamamoto, Methane decomposition for hydrogen production by catalytic activity of carbon black under low flow rate condirions, Journal of the Ceramic Society of Japan 125 (2017) 185-189. [38] S. Wauters, G. B. Marin, Computer generation of a network of elementary steps for coke formation during the thermal cracking of hydrocarbons, Chemical Engineering Journal 82 (2001) 267-279. [39] A. Jorio, M. Dresselhaus, R. Saito, G. F. Dresselhaus, Raman Spectroscopy in Graphene Related Systems, p. 332-333 (2012). [40] A. Reina, X. Jia, J. Ho, D. Nezich, H. Son, V. Bulovic, M. S. Dresselhaus, J. Kong, Large Area, Few-Layer Graphene Films on Arbitrary Substrates by Chemical Vapor Deposition, Nano Letters 9 (2009) 30-35.
21
Methane decomposition using carbon catalyst for CO2–free hydrogen production The carbon produced by methane decomposition also acts as a catalyst Raman spectroscopy is an effective method for evaluating initial catalytic activity
22
Figure 1. SEM micrographs of carbons before the experiment. (a) AC, (b) MC, (c) CB, and (d) CNF.
23
Hydrogen concentration (%)
100
PD-12 GC-2014
80 60 40 20 0 0
20
40
60
80
100
Elapsed time (min) Figure 2. Temporal changes of methane conversion ratio over CB measured by PD-12 and GC-2014.
24
Methane conversion ratio (%)
80
(a)
AC MC CB CNF
70 60 50 40 30 20 10 0 0
100
200
300
400
500
600
Elapsed time (min)
Methane conversion ratio (%)
100
(b) CB 2 -1 (1520 m g )
80
AC 2 -1 (2208 m g )
60
MC 2 -1 (1700 m g )
40 CNF 2 -1 (270 m g )
20 0 0
500
1000
1500
2
2000 -1
2500
Specific surface area (m g ) Figure 3. (a) Temporal change of methane conversion ratio at T = 1173 K and SV = 360 h-1 and (b) methane conversion ratio at five minutes of reaction vs. specific surface area at T = 1173 K and SV = 360 h-1.
25
2
Specific surface area (m /g)
10
4
10
3
10
2
10
1
10
0
AC MC CB CNF
0
100
200
300
400
500
600
700
800
Elapsed time (min) Figure 4. Temporal change of specific surface area by methane decomposition at T = 1173 K and SV = 360 h-1.
26
(a)
1
CB
0.1
4
rCH (mmol/min gcat)
10
Produced carbon (over CB)
0.01 0.1
1
PCH
4
10
kp (mmol/min gcat atm)
(b)
CB
1
Produced carbon (over CB) 0.1
0.01 0.84
0.86
0.88
0.90
0.92
0.94
-1
1/T * 1000 (K ) Figure 5. (a) Methane decomposition rate over CB as a function of methane partial pressure at T = 1173 K and (b) Arrhenius plot for methane decomposition over CB at SV = 360 h-1.
27
Figure 6. SEM micrographs and their magnified images of carbons after the experiment at T = 1173 K and SV = 360 h-1. (a) AC, (b) MC, (c) CB, and (d) CNF.
28
(a)
Intensity (a.u.)
CNF
CB MC
AC
0
1000
2000
3000
4000
-1
Raman shift (cm )
(b)
Intensity (a.u.)
CNF at 640 min CB at 700 min
MC at 640 min
AC at 640 min
0
1000
2000
3000
4000
-1
Raman shift (cm ) Figure 7. Raman spectra of (a) original carbons and (b) carbons after the experiment at T = 1173 K and SV = 360 h-1.
29
2.0
(a)
ID/IG (a.u.)
1.5
1.0 AC MC CB CNF
0.5
0.0 0
100
200
300 400 500 600 Elapsed time (min)
700
800
8
(b)
AC MC CB CNF
IG/IG' (a.u.)
6
4
2
0 0
100
200
300 400 500 600 Elapsed time (min)
700
800
Figure 8. Time variation of spectral parameters of Raman spectra at T = 1173 K and SV = 360 h-1 of (a) peak intensity ratio of the D band to the G band (ID/IG), (b) peak intensity ratio of the G band to the G’ band (IG/IG’).
30
SURE
BAY
FISSURE
Figure 9. BAY and FISSURE clusters at the coke surface [38].
31
BAY
Methane conversion ratio 3 2 per unit surface area *10 (g/m )
0.3
CB AC 0.2
MC
0.1
CNF 0.0 0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
ID/IG (a.u.) Figure 10. Methane conversion ratio per unit surface area after the experiment for 5 min at T = 1173 K and SV = 360 h-1 vs. ID/IG of original catalyst.
32
Table
Table 1. Tested carbon catalysts. Use
BET surface area (m2/g)
Bulk density (g/cm3)
Pore diameter (nm)
UESb TOYO TANSO
2208 1700
0.21 0.07
1.8 5.6
6.5 5
Cabot Showa Denko
1520 270
0.17 0.09
― ―
2.2×10-3 ―
Name
Manufacturer
Activated carbon (AC) Mesoporous carbon (MC)
UCG-CPT P (3)010
Carbon black (CB) Carbon nanofiber (CNF)
BP2000a VGCF-H
a BP = black pearls. b UES = Ueda environmental solution Co., Ltd.
Particle size (μm)
Table 2. Activation energy of carbons.
Author
Catalysts
Name
Measurement time (min)
Specific surface area (m2/g)
Temperature range (K)
This study
CB
BP-2000
After 5 min
1520
1073–1173
0.67
144
E. K. Lee E. K. Lee S. Y. Lee S. Y. Lee S. Y. Lee S. Y. Lee S. Y. Lee
CB CB CB CB CB CB CB
DCC-N330 BP-2000 HI-170 HI-20L HI-900L EC-600JD BP-2000
20–30 20–30 20–60 20–60 20–60 20–60 20–60
79 1475 23 86 300 84 240
1123–1273 1023–1273 1123–1223 1123–1223 1123–1223 1123–1223 1023–1223
0.919 0.984 0.72 0.86 1.04 0.71 0.84
183 143 185 181 178 135 143
S. Y. Lee S. Y. Lee
CB CB
DCC-N774 DCC-N330
20–60 20–60
1270 1500
1123–1223 1123–1223
0.69 0.71
233 213
J. L. Pinilla Y. Kameya Y. Kameya Y. Kameya
CB CB CB CB
BP-2000 SB285 SB905 SB285
Initial reaction After 60 min After 60 min After 60 min
1337 81 212 81
973–1223 1013–1113 1013–1113 1133–1233
0.6 0.59 0.62 0.59
238 221 179 192
Y. Kameya N. Muradov J. L. Pinilla M. H. kim M. H. kim N. Muradov
CB CB AC AC AC AC
SB905 BP-2000 CG-Norit CCN-DR CL-SCR KBB
After 60 min Initial reaction Initial reaction Initial reaction Initial reaction Initial reaction
212 1650 1300 725 912 1500
1133–1233 973–1173 973–1223 923–1173 923–1173 973–1173
0.62 0.5±0.1 0.48 0.51 0.49 0.6±0.1
157 205–236 141 194 186 160–201
Reaction order
Activation energy (kJ/mol)