- Email: [email protected]
Methane decomposition over high-loaded Ni-Cu-SiO2 catalysts
Fusion Engineering and Design 113 (2016) 279–287
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Methane decomposition over high-loaded Ni-Cu-SiO2 catalysts Jiamao Li, Linjie Zhao, Jianchao He, Liang Dong, Liangping Xiong, Yang Du, Yong Yang, Heyi Wang ∗ , Shuming Peng ∗ Institute of Nuclear Physics and Chemistry, China Academy of Engineering Physics, Mianyang 621900, Sichuan, PR China
h i g h l i g h t s
g r a p h i c a l
• Methane decomposition over Ni-Cu-
Methane decomposition-regeneration with air cycles over 65%Ni-20%Cu-10%SiO2 catalysts.
a b s t r a c t
SiO2 was studied. • The deactivated catalysts were regenerated by air. • Introduction of Cu could enhance the catalytic performance of Ni-SiO2 . • The increase of the Ni-Cu particle influences the performance of the catalysts.
a r t i c l e
i n f o
Article history: Received 14 November 2015 Received in revised form 16 April 2016 Accepted 22 June 2016 Available online 21 July 2016 Keywords: Ni-SiO2 Ni-Cu-SiO2 Methane decomposition Regeneration with air Spherical carbon structure
a b s t r a c t The performance of Ni-SiO2 and Ni-Cu-SiO2 during repeated catalytic decomposition of methane (CDM) reactions and subsequent regeneration of the deactivated catalysts with air has been studied. The catalytic activity of the 75%Ni-25%SiO2 catalyst in the second and third CDM was lower than that during the first, while the lifetime of the catalyst did not change significantly. Both the lifetime and the catalytic activity of 65%Ni-10%Cu-25%SiO2 in the second and third CDM reactions decreased significantly. 55%Ni-20%Cu25%SiO2 showed better performance than the other two catalysts, and its activity and lifetime did not change significantly until the third CDM reaction. The hydrogen yields of 55%Ni-20%Cu-25%SiO2 were 56.8 gH2 /gcat., 42.8 gH2 /gcat., and 2.4 gH2 /gcat. for the first, second, and third CDM reactions, respectively. Spherical carbon structures were observed on the catalysts following all three CDM reactions over 75%Ni25%SiO2 . However, similar carbon structures were only observed following the second and third CDM over 65%Ni-10%Cu-25%SiO2 , and only following the third cycle with 55%Ni-20%Cu-25%SiO2 . The formation of spherical carbon during the repeated CDM reactions strongly influenced the performance of the catalysts. © 2016 Published by Elsevier B.V.
1. Introduction ∗ Corresponding authors. E-mail addresses: [email protected] (H. Wang), [email protected] (S. Peng). http://dx.doi.org/10.1016/j.fusengdes.2016.06.046 0920-3796/© 2016 Published by Elsevier B.V.
In the field of fusion energy, carbon fiber composites consider to be employed as plasma facing components in the design
280
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287
of the International Thermonuclear Experimental Reactor (ITER), and deuterated and tritiated methane may be a considerable constituent of the impurity gas stream [1–6]. The catalytic decomposition of methane (CDM) reaction is a safe and simple method for recovering deuterium and tritium from deuterated and tritiated methane, and an optional catalyst bed for the CDM is currently included in the design of the tokamak exhaust processing (TEP) system in the ITER [7]. The gas exhaust from the D-T fusion reaction in the ITER, which contains Q2 , Q2 O, CQ4 , He, CO, CO2 , Q = H, D, and T, enters the TEP system first. A catalyst bed that contains Ni-based catalysts is included in the 2nd stage of the TEP system, and the function of this catalyst bed is to decompose the deuterated and tritiated methane to recover deuterium and tritium [8]. Much research on the CDM has been reported [9–17]. The catalysts traditionally used in the CDM consist of 3d transition metals (Ni, Fe, Co) supported on different metal oxides (e.g., Al2 O3 or SiO2 ). For CDM reactions, Ni-based catalysts exhibit better performance than Fe-based and Co-based catalysts between 500 ◦ C and 700 ◦ C. Therefore, Ni-based catalysts for the CDM have drawn much attention [18–23]. However, one disadvantage of Ni-based catalysts is that they are deactivated at temperatures above 600 ◦ C [24]. There is a consensus in the literature that catalyst performance in the CDM is highly dependent on the crystallite size of the catalyst particles, where larger particle sizes cause the catalyst to be deactivated more rapidly [25]. Therefore, support materials, such as SiO2 , Al2 O3 , and CeO2 , have been introduced to control particle size and dispersion, which is possible through physical or chemical interactions between the support and the active metallic particle. Takenaka et al. [26] studied the effect of different supports on the performance of Ni catalysts, and they reported that SiO2 , TiO2 , and graphite are effective supports for Ni in the CDM. Although much research on the CDM has been undertaken by many researchers, the regeneration of these catalysts after deactivation has been rarely investigated. In view of this, this study concerns the catalytic performance of high loaded Ni-SiO2 and Ni-Cu-SiO2 catalysts during repeated CDM reactions, and the subsequent regeneration of the deactivated catalysts with air. The carbon structures formed over the catalysts during the repeated CDM reaction is also studied. 2. Experimental 2.1. Catalyst preparation In this work, 75%Ni-25%SiO2 , 65%Ni-10%Cu-25%SiO2 , and 55%Ni-20%Cu-25%SiO2 catalysts (where the percentages in the nomenclature represent mass fraction) were prepared by a heterophase sol-gel technique [27]. Catalysts were prepared by mixing the active precursor (NiO or a mixture of NiO and CuO), in a certain amount of an alcohol solution of tetraethoxisilane (TEOS), which acted as a precursor of the structural promoter. The alcohol solution of TEOS was prepared as described by Wang [21] in detail. The turbid liquids, which contained active precursors and alcosol, were dried in flowing air at room temperature and calcined at 650 ◦ C for 3 h. In addition, the mixtures of NiO and CuO were prepared by a simple and convenient process. First, the mixtures of Ni(NO3 )2 and Cu(NO3 )2 were obtained by evaporation of the mixed solution of Ni(NO3 )2 and Cu(NO3 )2 . Then, the mixture of NiO and CuO was formed by calcining the mixture of Ni(NO3 )2 and Cu(NO3 )2 at 450 ◦ C. 2.2. Catalyst performance tests Tests of the catalytic activity and lifetime of the catalysts during the CDM reaction were carried out in a fixed-bed quartz
Fig. 1. Kinetic curves of methane decomposition over Ni-SiO2 and Ni-Cu-SiO2 catalysts at 650 ◦ C.
reactor (10 mm i.d.) under atmospheric pressure. Prior to the reaction, the catalysts were reduced with hydrogen at 650 ◦ C for 1 h. Highly purified methane (99.99%) was passed through the reactor for the CDM reaction to take place. Regeneration of the deactivated 75%Ni-25%SiO2 , 65%Ni-10%Cu-25%SiO2 , and 55%Ni20%Cu-25%SiO2 catalysts was performed through the gasification of the deposited carbon with 20 cm3 /min of air at 600 ◦ C until the formation of CO2 no longer occurred. Then, the catalysts were reduced with hydrogen at 650 ◦ C for 1 h, and the catalytic performance in the CDM by the regenerated catalysts was retested. For each catalyst, the CDM reaction and subsequent regeneration were repeated twice. The gaseous reaction products were monitored by on-line gas chromatography (GC) using a TDX-01 column and a thermal conductivity detector (TCD) for H2 , CH4 , and CO2 analysis. 2.3. Catalyst characterization XRD patterns were recorded using a Bruker D8 Advance diffractometer (Cu Ka radiation at 40 kV and 40 mA). Micrographs of the catalysts and the carbon structures were recorded with an FEI Inspect F scanning electron microscope (SEM). The detailed morphological appearances of the carbon structures were observed using a Tecnai G2 F20 S-Twin transmission electron microscope (TEM) operated at 200 kV. For TEM analysis, samples were prepared by drying a drop of an alcohol suspension of the catalyst particles on a cobalt grid. Surface area (BET) and pore volume (PV) were determined from nitrogen adsorption/desorption isotherms, using a JW-BK200C (JWGB SCI&TECH) gas adsorption device. Before analysis, all samples were out-gassed at 100 ◦ C under vacuum (4 h). 3. Results and discussion 3.1. Effect of Cu on Ni-based catalysts during the CDM reaction Fig. 1 shows the kinetic curves for the CDM reaction over NiSiO2 and Ni-Cu-SiO2 catalysts at 650 ◦ C. Only hydrogen is obtained as a gaseous product over all the catalysts. The lifetime of the Ni-SiO2 catalyst without Cu is very short at 650 ◦ C. After 70 min, the methane conversion of 75%Ni-25%SiO2 decreases to ca. 5% and remains at this value. The introduction of 10% Cu significantly improves the lifetime of the Ni-SiO2 catalyst, and the methane conversion over 65%Ni-10%Cu-25%SiO2 is constant at 40% during the reaction. However, when the Cu content is increased to 20%, the methane conversion decreases to ca. 20%. Therefore, the introduc-
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287
281
Fig. 2. SEM micrographs of carbon deposited over Ni-SiO2 and Ni-Cu-SiO2 catalyst at 650 ◦ C. (a) 75%Ni-25%SiO2, (b) 65%Ni-10%Cu-25%SiO2 , and (c) 55%Ni-20%Cu-25%SiO2 .
Table 1 Cyclic CDM and subsequent regeneration with air over the three catalysts. Cu content
0 10% 20%
m(cat.) mg
10 3 3
The first cycle
The second cycle
The third cycle
Lifetime min
Hydrogen yield gH2 /gcat.
Lifetime min
Hydrogen yield gH2 /gcat.
Lifetime min
Hydrogen yield gH2 /gcat.
20 282 340
0.96 38.6 56.8
20 25 320
0.60 2.66 42.8
20 25 25
0.56 1.83 2.4
tion of Cu into Ni-SiO2 catalysts extends their lifetime; however, when more Cu is introduced into the catalysts, their catalytic activity is reduced. Fig. 2 shows the SEM images of the carbon structures formed over Ni-SiO2 and Ni-Cu-SiO2 catalysts, all of which were prepared at 650 ◦ C. Two kinds of carbon structures are formed over the 75%Ni25%SiO2 catalysts: filamentous carbon and spherical carbon [32]. However, only filamentous carbon is obtained over the Ni-Cu-SiO2 catalysts. Careful examination reveals the textural characteristics and microstructure of the carbon deposited on the Ni-SiO2 catalysts. Some single carbon filaments with smooth and regular shapes form over Ni-SiO2 catalysts, and a pear-shaped Ni metal particle is located at the tip [33] and the particle size is about 50 nm, as shown in Fig. 3(a). However, due to the high activity of Ni metal, pure Ni particles easily sinter into bigger clusters during the reduction process. With an increase in the size of the pure Ni particles, the carbon generation rates exceed that of the rates of carbon diffusion and precipitation, and the enrichment of that carbon leads to the formation of carbon layers on the surfaces of the particles [34]. Consequently, some spherical carbon structures are generated by the Ni-SiO2 catalysts during the CDM reaction, as shown in
Fig. 3(b). Therefore, the bigger Ni-Cu particles form the spherical carbon structures and show poor catalytic performance during the CDM reaction. In the case of the Ni-Cu-SiO2 catalysts, the Ni-Cu particles maintain their size in a suitable range during the reduction process due to the strong influence of Cu on the dispersion of Ni in the catalyst [35]. Therefore, few spherical carbons are observed for these catalysts. Moreover, it found that several filamentous carbon structures deposited on one Ni-Cu particle, and both of their particle sizes are about 100 nm, as shown in Fig. 3(c)–(d). In addition, Cu itself is inactive for methane decomposition and has a negligible effect on the carbon diffusion rate [36]. Therefore, Cu can dilute the planes of Ni, causing segregation of the Ni (111) planes, leading to the formation of several carbon filaments from one particle. This carbon structure is referred to as an octopus-like carbon nano-filament [37,38]. More importantly, the formation of the octopus-like sites are present on the Ni (111), which facilitates carbon transportation. This may be the reason that Ni-Cu-SiO2 catalysts have a better catalytic performance than Ni-SiO2 .
282
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287
Fig. 3. TEM micrographs of carbon deposited over Ni-SiO2 and Ni-Cu-SiO2 at 650 ◦ C. 75%Ni-25%SiO2 , (b) 75%Ni-25%SiO2 , (c) 65%Ni-10%Cu-25%SiO2 , (d) 55%Ni-20%Cu-25%SiO2 .
3.2. Catalytic performance of Ni-SiO2 and Ni-Cu-SiO2 catalysts during repeated CDM Fig. 4 shows the kinetic curves of the CDM reaction over NiSiO2 and Ni-Cu-SiO2 catalysts during three repeated CDM reactions. Comparing to the Ni-SiO2 and Ni-Cu-SiO2 catalysts in the reaction condition as shown in Fig. 1, due to the decrease of the mass of the catalysts and increased the flow rate of methane, the Ni-SiO2 and Ni-Cu-SiO2 catalysts both have shorter catalytic lifetime (from the initial stage to the moment of methane conversion blew 2.5%), especially for the 75%Ni-25%SiO2 catalysts that the methane conversion decreased blew 2.5% after just 20 min. Therefore, the high methane flow rate may promote high decomposition rates (enhance carbon atoms covered the surface of catalysts particles) and cause the activity decreased remarkable. Moreover, the catalytic performance of the Ni-SiO2 and Ni-CuSiO2 catalysts after regeneration in air are quite different from each other, as shown in Fig. 4(a). Compared to the first CDM reaction over 75%Ni-25%SiO2 catalysts, the lifetime of the 75%Ni-25%SiO2 catalyst during the second and third CDM reaction does not change significantly; however, the catalytic activity decreases markedly for the second and third CDM reaction. For the repeated CDM reaction over 65%Ni-10%Cu-25%SiO2 catalysts, the lifetime and the catalytic activity of the catalyst decreases markedly for the second and third CDM, as shown in Fig. 4(b). The 55%Ni-20%Cu-25%SiO2 catalyst shows the longest lifetime during the first CDM, and the lifetime of this catalyst during the second CDM does not decrease significantly, as shown in Fig. 4(c). Although the catalytic activity of the 55%Ni-20%Cu-25%SiO2 catalyst is a little
lower than that for the first CDM. This suggests that the catalytic performance of Ni-Cu-SiO2 during repeated CDM reactions is significantly influenced by the Cu content. However, when the third CDM over 55%Ni-20%Cu-25%SiO2 is performed, both the lifetime and the catalytic activity of the catalyst decrease significantly. The catalytic performances of the prepared catalysts during repeated CDM reactions are summarized in Table 1. 3.3. SEM images of the catalysts and carbon structures after repeated CDM reaction Fig. 5 shows the SEM images of carbon structures generated by 75%Ni-25%SiO2 catalysts during repeated CDM reactions. The images reveal that a large amount of spherical carbon forms over 75%Ni-25%SiO2 catalysts during the three CDM reactions. It may be observed that there are two kinds of carbon structure formed during the three CDM reactions. However, when the regeneration frequency is increased, the size of the carbon spheres increases. Fig. 6 shows SEM images of the carbon structures generated over the Ni-Cu-SiO2 catalyst during the CDM, and the subsequent regeneration with air cycles. Only carbon nano-filaments are generated over 65%Ni-10%Cu-25%SiO2 during the first cycle (Fig. 6(a)). However, when the deactivated 65%Ni-10%Cu-25%SiO2 is regenerated in air, some spherical carbon appears in the second and third cycle, as shown in Fig. 6(b)–(c). The structure of the filamentous carbon generated over 55%Ni-20%Cu-25%SiO2 is a little different from that generated over 65%Ni-10%Cu-25%SiO2 during the three cycles: only carbon filaments are generated during the first and second cycle (Fig. 6(d)–(e)), and some larger spherical carbon is produced
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287
283
Fig. 4. Methane decomposition-regeneration with air cycles over Ni-SiO2 and Ni-Cu-SiO2 catalysts ((a): 75%Ni-25%SiO2 , 10 mg; (b): 65%Ni-10%Cu-25%SiO2 , 3 mg; (c): 55%Ni20%Cu-25%SiO2 , 3 mg).
during the third cycle (Fig. 6(f)). Comparing the rates of methane conversion described in Fig. 6 and Table 1, it can be concluded that when the Ni-Cu particle sinter into a big cluster that it will generate the spherical carbon and show poor catalytic performance during the CDM reaction. Introduction of Cu not only promotes the catalytic performance of Ni-based catalysts, but also helps preserve the activity of Ni-based catalysts during the regeneration in air. However, with an increase in the number of regenerations, the NiCu-SiO2 catalysts lose their ability to crack methane, even when the Cu content is increased to 20%. Fig. 7 shows the XRD patterns of the fresh Ni-Cu-SiO2 catalysts and the samples after regeneration in air at 600 ◦ C, and the crystal sizes of the three samples are calculated by Debye-Scherrer formula from the most intense peak (Ca. 43.25◦ ) of NiO. As shown in Fig. 7(a)–(b), the Ni and Cu species in the regenerated Ni-Cu-SiO2 catalysts mainly exist in the form of NiO and CuO. Moreover, due to carbon phase is not noted to exist, so it can prove that carbon atoms have been removed from the surface of the catalysts. The XRD characterization results also show the change of the NiO particle size, as shown in Fig. 7(a)–(b). For the 65%Ni-10%Cu-
25%SiO2 catalysts, the particle size is changed from 24.7 to 27.5 nm after the first regeneration, however, it barely increases after the second regeneration. Different from the 65%Ni-10%Cu-25%SiO2 catalysts, the particles size is not changed after the first regeneration, but it remarkable increased after the second regeneration (from 22.3 to 24.7 nm). Therefore, it can be concluded that the aggregation process of NiO particles can be reduced in some extent through increasing of Cu content after the first regeneration. However, with an increase in the number of regenerations, the average of the NiO particles increases over Ni-Cu-SiO2 catalysts even the Cu content increases to 20%. Fig. 8 shows the XRD spectra of the fresh Ni-Cu-SiO2 catalysts and the regenerated catalysts after hydrogen reduced, and the crystal sizes of the three samples are calculated by DebyeScherrer formula from the most intense peak (Ca. 44.5◦ ) of Ni-Cu alloy. The diffraction peaks due to Cu is not detected in NiCu-SiO2 catalysts even if the Cu content reaches 20% or the catalysts regenerated twice. Moreover, the Ni-Cu particle size over 55%Ni-20%Cu-25%SiO2 catalysts is much smaller the 65%Ni-10%Cu25%SiO2 catalyst in each the same cycle. Similar with Fig. 7, with
284
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287
Fig. 5. SEM images of carbon structures formed over 75%Ni-25%SiO2 catalysts during repeated methane decomposition ((a): the first reaction; (b): the second reaction; and (c): the third reaction).
an increase in the number of regenerations, the average of the Ni-Cu particles increased over 65%Ni-10%Cu-25%SiO2 and 55%Ni20%Cu-25%SiO2 catalysts. However, the average particle size of 55%Ni-20%Cu-25%SiO2 catalysts does not obviously increased after the first regeneration in air at 600 ◦ C, which might prove that the generated catalysts can maintain good catalytic performance during the CDM reaction. To detailed investigated the relationship between the Ni-CuSiO2 catalysts and catalytic performance, the nitrogen adsorption/desorption isotherms is used to evaluate their surface structures. Although the reaction and regeneration temperature might be a little higher for Sol-gel derived materials, the BET surface area, pore volume and micropore volume of Ni-Cu-SiO2 catalysts increased with the number of regenerations Table 2. For the NiCu-SiO2 catalyst, the BET surface area of 65%Ni-10%Cu-25%SiO2 catalyst increases from 37.701 to 49.332 m2 /g after two regeneration processes, and the BET surface area of 55%Ni-20%Cu-25%SiO2 catalysts increased from 27.834 to 45.023 m2 /g. Moreover, no matter how many times the catalyst is generated, it has the same average pore diameter. Therefore, the results can states that the pore structures of the Ni-Cu-SiO2 catalyst do not break during the reaction and regeneration process. Therefore, the increase of the Ni-Cu particle size is the key factor to cause the catalysts exhibited poor catalytic performance after the regeneration in air. Several studies [25] have reported that the carbon filaments formed over Ni-Cu-SiO2 catalysts obeyed a “tip-growth” model, and that the Ni-Cu particle was located at the tip of the filaments; therefore, the Ni-Cu clusters did not interact with the SiO2 support during the carbon filament growth process. In this work, when the deac-
tivated Ni-Cu-SiO2 catalysts are regenerated in air at 600 ◦ C, the carbon is removed from the Ni-Cu species by combustion. In this situation, Ni-Cu species easily sinter into larger clusters because of the weak interaction between the particle and SiO2 support. As described in Section 3.2, due to the Ni-Cu clusters sintering into larger particles, the carbon diffusion and deposition rate is lower than the methane decomposition rate over these catalysts during the CDM process. Consequently, the enrichment of the carbon encapsulates the active sites of the metal by the formation of graphite layers or destruction of the active metal [12]. Therefore, spherical carbon is generated during this process. However, though introduction of Cu enhances the dispersion of Ni particles, the NiCu particles without support still exhibit an increase in particle size during the regeneration process to some extent, even when the Cu content is 20%. 4. Conclusions The catalytic performance of 75%Ni-25%SiO2 , 65%Ni-10%Cu25%SiO2 , and 55%Ni-20%Cu-25%SiO2 during repeated methane decomposition decreased with the number of repetitions for all three catalysts. 55%Ni-20%Cu-25%SiO2 exhibited better catalytic performance than the other catalysts, affording hydrogen yields for the repeated methane decomposition of 56.8, 42.8, and 2.4 gH2 /gcat. The temperature programmed reduction of the Ni-CuSiO2 catalysts indicated that Ni-Cu species were present in the fresh Ni-Cu-SiO2 catalyst, although this phenomenon was not observed in the XRD spectra of the fresh Ni-Cu-SiO2 catalyst. The XRD studies of Ni-Cu-SiO2 catalysts reduced by H2 indicated that a Ni-Cu
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287
285
Fig. 6. SEM images of the Ni-Cu-SiO2 catalyst after methane decomposition-regeneration with air (65%Ni-10%Cu-25%SiO2 : a: the first cycle; b: the second cycle; c: the third cycle; 55%Ni-20%Cu-25%SiO2 : d: the first cycle; e: the second cycle; f: the third cycle).
Table 2 the surface structures of the fresh Ni-Cu-SiO2 catalysts and the samples after regeneration by air. Samples Regeneration times
BET surface areas (m2 /g) Average pore diameter (nm) Pore volume (cm3 /g) Micropore volume (10−3 cm3 /g)
65%Ni-10%Cu-25%SiO2 catalysts
55%Ni-20%Cu-25%SiO2 catalysts
0
1
2
0
1
2
37.701 18.673 0.225 0.0153
40.684 20.848 0.239 0.0109
49.332 20.917 0.587 0.0153
27.834 16.925 0.131 0.0108
40.926 19.979 0.227 0.0125
45.023 18.389 0.168 0.0157
286
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287
(a)
(a)
(b)
(b)
Fig. 8. XRD patterns of the fresh Ni-Cu-SiO2 catalysts and the regenerated catalysts after hydrogen reduced ((a): 65%Ni-10%Cu-25%SiO2 catalysts; (b): 65%Ni-10%Cu25%SiO2 catalysts). Fig. 7. XRD patterns of the fresh Ni-Cu-SiO2 catalysts and the samples after regeneration by air ((a): 65%Ni-10%Cu-25%SiO2 catalysts; (b): 65%Ni-10%Cu-25%SiO2 catalysts).
alloy was formed in the Ni-Cu-SiO2 catalyst reduced by H2 . Spherical carbon structures were observed following all three repeated CDM reactions for 75%Ni-25%SiO2 . However, similar carbon structures were only observed following the second and third CDM for 65%Ni-10%Cu-25%SiO2 , and only following the third one for 55%Ni20%Cu-25%SiO2 . Moreover, the BET surface area, pore volume and micropore volume of Ni-Cu-SiO2 catalysts increased with the number of regenerations. The increase of the Ni-Cu particle during the regeneration process strongly influenced the performance of the catalysts, and Ni-Cu particle which suffered an aggregation process (sinter into big cluster) exhibited poor performance and formed the spherical carbon structures during the CDM reaction. Acknowledgements The authors are grateful for the financial support by the Science and Technology Development Foundation (No.2013B0301036) and
the Programs of Laboratory of Advanced Nuclear Energy (LANE), managed by the Institute of Nuclear and Chemistry(No. R20140303-03). References [1] M. Glugla, L. Dörra, R. Lässera, D. Murdoch, H. Yoshidac, Fusion Eng. Des. 61 (2002) 569–574. [2] C. Rizzelloa, F.B.T. Pinnab, S. Tosti, Fusion Eng. Des. 85 (1) (2010) 58–63. [3] M. Glugla, D.K. Murdoch, H. Geißler, P. Herrmann, Kraemer, Fusion Eng. Des. 39–40 (1) (1998) 893–899. [4] R. Lasser, M. Glugla, K. Gunther, T.L. Le, D. Niyongabo, R.D. Penzhorn, K.H. Simon, Fusion Eng. Des. 58–59 (2001) 371–375. [5] S. Tosti, L. Bettinali, F. Borgognoni, D.K. Murdoch, Fusion Eng. Des. 82 (2) (2007) 153–161. [6] M. Glugla, A. Perevezentsev, D. Niyongabo, R.D. Penzhorn, A. Bell, P. Herrmann, Fusion Eng. Des. 49 (-50) (2000) 817–823. [7] M. Glugla, D.K. Murdoch, Fusion Eng. Des. 39-40 (1998) (893–869). [8] M. Glugla, A. Antipenkov, S. Beloglazov, Fusion Eng. Des. 82 (5–14) (2007) 472–487. [9] A.L. Dicks, J. Power Sources 61 (1996) 113–124. [10] S.S. Penner, Energy 31 (2006) 33–43. [11] R.A. Lemons, J. Power Sources 29 (1990) 251–264. [12] B.C.H. Steele, A. Heinzel, Nature 414 (2001) 345–352.
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287 [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25]
J.H. Wee, K.Y. Lee, J. Power Sources 157 (2006) 128–135. J.N. Coleman, U. Khan, W.J. Blau, Y.K. Gun’ko, Carbon 44 (2006) 1624–1652. W. Zhang, Q.J. Ge, H.Y. Xu, J. Energy Chem. 20 (2011), 339–334. A. Chatla, V.K. Velisoju, K.B. Suresh, V. Akula, J. Energy Chem. 22 (2013) 853–860. S. Tajammul Hussain, S. Gul, M. Mazhar, D.H. Anjum, F. Larachi, J. Nat. Gas Chem. 17 (4) (2016) 374–382. Y. Wang, J. Peng, C. Zhou, Z.Y. Lim, Int. J. Hydrogen Energy 39 (2014) 778–787. S.K. Saraswat, K.K. Pant, Int. J. Hydrogen Energy 13 (2013) 52–59. J.L. Pinilla, R. Utrilla, M.J. Lázaro, Fuel Process. Technol. 2111 (92) (2016) 1480–1488. W.H. Wang, H.Y. Wang, Y. Yang, Int. J. Hydrogen Energy 37 (2012) 9058–9066. N.Z. Muradov, T.N. Veziroglu, Int. J. Hydrogen Energy 30 (2005) 225–237. Y. Zhang, K.J. Smith, Catal. Lett. 95 (2004) 7–12. H. Ogihara, S. Takenaka, I. Yamanaka, E. Tanabe, A. Genseki, K. Otsuka, J. Catal. 238 (2006) 353–360. A.C. Lua, H.Y. Wang, Appl. Catal. B Environ. 132–133 (2013) 469–478.
[26] [27] [32] [33] [34] [35] [36] [37] [38]
287
Y. Li, J. Chen, L. Chang, Y. Qin, J. Catal. 178 (1998) 76–83. M.A. Ermakova, D.Y. Ermakov, Catal. Today 77 (3) (2002) 225–235. N.Z. Muradova, T.N. Veziroglu, Int. J. Hydrogen Energy 30 (2005) 225–237. J.C. Guevara, J.A. Wang, L.F. Chen, M.A. Valenzuela, P. Salas, Int. J. Hydrogen Energy 35 (8) (2010) 3509–3521. T.V. Reshetenko, L.B. Avdeeva, Z.R. Ismagilov, A.L. Chuvilin, V.A. Ushakov, Appl. Catal. A Gen. 247 (1) (2003) 51–63. Y. Echegoyen, I. Suelves, R. Lázaro Moliner, J.M. Palacios, J. Power Sources 169 (2007) 150–157. H.Y. Wang, A.C. Lua, Chem. Eng. J. 169 (2007) 150–157. L.B. Avdeeva, O.V. Goncharova, D.I. Kochubey, V.I. Zaikovskii, L.M. Plyasova, B.N. Novgorodov, S.K. Shaikhutdinov, Appl. Catal. A Gen. 141 (1996) 117–129. T.V. Aapg BulletinReshetenko, L.B. Avdeeva, Z.R. Ismagilov, A.L. Chuvilin, V.A. Ushakov, Appl. Catal. A Gen. 247 (2003) 51–63.
Contents lists available at ScienceDirect
Fusion Engineering and Design journal homepage: www.elsevier.com/locate/fusengdes
Methane decomposition over high-loaded Ni-Cu-SiO2 catalysts Jiamao Li, Linjie Zhao, Jianchao He, Liang Dong, Liangping Xiong, Yang Du, Yong Yang, Heyi Wang ∗ , Shuming Peng ∗ Institute of Nuclear Physics and Chemistry, China Academy of Engineering Physics, Mianyang 621900, Sichuan, PR China
h i g h l i g h t s
g r a p h i c a l
• Methane decomposition over Ni-Cu-
Methane decomposition-regeneration with air cycles over 65%Ni-20%Cu-10%SiO2 catalysts.
a b s t r a c t
SiO2 was studied. • The deactivated catalysts were regenerated by air. • Introduction of Cu could enhance the catalytic performance of Ni-SiO2 . • The increase of the Ni-Cu particle influences the performance of the catalysts.
a r t i c l e
i n f o
Article history: Received 14 November 2015 Received in revised form 16 April 2016 Accepted 22 June 2016 Available online 21 July 2016 Keywords: Ni-SiO2 Ni-Cu-SiO2 Methane decomposition Regeneration with air Spherical carbon structure
a b s t r a c t The performance of Ni-SiO2 and Ni-Cu-SiO2 during repeated catalytic decomposition of methane (CDM) reactions and subsequent regeneration of the deactivated catalysts with air has been studied. The catalytic activity of the 75%Ni-25%SiO2 catalyst in the second and third CDM was lower than that during the first, while the lifetime of the catalyst did not change significantly. Both the lifetime and the catalytic activity of 65%Ni-10%Cu-25%SiO2 in the second and third CDM reactions decreased significantly. 55%Ni-20%Cu25%SiO2 showed better performance than the other two catalysts, and its activity and lifetime did not change significantly until the third CDM reaction. The hydrogen yields of 55%Ni-20%Cu-25%SiO2 were 56.8 gH2 /gcat., 42.8 gH2 /gcat., and 2.4 gH2 /gcat. for the first, second, and third CDM reactions, respectively. Spherical carbon structures were observed on the catalysts following all three CDM reactions over 75%Ni25%SiO2 . However, similar carbon structures were only observed following the second and third CDM over 65%Ni-10%Cu-25%SiO2 , and only following the third cycle with 55%Ni-20%Cu-25%SiO2 . The formation of spherical carbon during the repeated CDM reactions strongly influenced the performance of the catalysts. © 2016 Published by Elsevier B.V.
1. Introduction ∗ Corresponding authors. E-mail addresses: [email protected] (H. Wang), [email protected] (S. Peng). http://dx.doi.org/10.1016/j.fusengdes.2016.06.046 0920-3796/© 2016 Published by Elsevier B.V.
In the field of fusion energy, carbon fiber composites consider to be employed as plasma facing components in the design
280
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287
of the International Thermonuclear Experimental Reactor (ITER), and deuterated and tritiated methane may be a considerable constituent of the impurity gas stream [1–6]. The catalytic decomposition of methane (CDM) reaction is a safe and simple method for recovering deuterium and tritium from deuterated and tritiated methane, and an optional catalyst bed for the CDM is currently included in the design of the tokamak exhaust processing (TEP) system in the ITER [7]. The gas exhaust from the D-T fusion reaction in the ITER, which contains Q2 , Q2 O, CQ4 , He, CO, CO2 , Q = H, D, and T, enters the TEP system first. A catalyst bed that contains Ni-based catalysts is included in the 2nd stage of the TEP system, and the function of this catalyst bed is to decompose the deuterated and tritiated methane to recover deuterium and tritium [8]. Much research on the CDM has been reported [9–17]. The catalysts traditionally used in the CDM consist of 3d transition metals (Ni, Fe, Co) supported on different metal oxides (e.g., Al2 O3 or SiO2 ). For CDM reactions, Ni-based catalysts exhibit better performance than Fe-based and Co-based catalysts between 500 ◦ C and 700 ◦ C. Therefore, Ni-based catalysts for the CDM have drawn much attention [18–23]. However, one disadvantage of Ni-based catalysts is that they are deactivated at temperatures above 600 ◦ C [24]. There is a consensus in the literature that catalyst performance in the CDM is highly dependent on the crystallite size of the catalyst particles, where larger particle sizes cause the catalyst to be deactivated more rapidly [25]. Therefore, support materials, such as SiO2 , Al2 O3 , and CeO2 , have been introduced to control particle size and dispersion, which is possible through physical or chemical interactions between the support and the active metallic particle. Takenaka et al. [26] studied the effect of different supports on the performance of Ni catalysts, and they reported that SiO2 , TiO2 , and graphite are effective supports for Ni in the CDM. Although much research on the CDM has been undertaken by many researchers, the regeneration of these catalysts after deactivation has been rarely investigated. In view of this, this study concerns the catalytic performance of high loaded Ni-SiO2 and Ni-Cu-SiO2 catalysts during repeated CDM reactions, and the subsequent regeneration of the deactivated catalysts with air. The carbon structures formed over the catalysts during the repeated CDM reaction is also studied. 2. Experimental 2.1. Catalyst preparation In this work, 75%Ni-25%SiO2 , 65%Ni-10%Cu-25%SiO2 , and 55%Ni-20%Cu-25%SiO2 catalysts (where the percentages in the nomenclature represent mass fraction) were prepared by a heterophase sol-gel technique [27]. Catalysts were prepared by mixing the active precursor (NiO or a mixture of NiO and CuO), in a certain amount of an alcohol solution of tetraethoxisilane (TEOS), which acted as a precursor of the structural promoter. The alcohol solution of TEOS was prepared as described by Wang [21] in detail. The turbid liquids, which contained active precursors and alcosol, were dried in flowing air at room temperature and calcined at 650 ◦ C for 3 h. In addition, the mixtures of NiO and CuO were prepared by a simple and convenient process. First, the mixtures of Ni(NO3 )2 and Cu(NO3 )2 were obtained by evaporation of the mixed solution of Ni(NO3 )2 and Cu(NO3 )2 . Then, the mixture of NiO and CuO was formed by calcining the mixture of Ni(NO3 )2 and Cu(NO3 )2 at 450 ◦ C. 2.2. Catalyst performance tests Tests of the catalytic activity and lifetime of the catalysts during the CDM reaction were carried out in a fixed-bed quartz
Fig. 1. Kinetic curves of methane decomposition over Ni-SiO2 and Ni-Cu-SiO2 catalysts at 650 ◦ C.
reactor (10 mm i.d.) under atmospheric pressure. Prior to the reaction, the catalysts were reduced with hydrogen at 650 ◦ C for 1 h. Highly purified methane (99.99%) was passed through the reactor for the CDM reaction to take place. Regeneration of the deactivated 75%Ni-25%SiO2 , 65%Ni-10%Cu-25%SiO2 , and 55%Ni20%Cu-25%SiO2 catalysts was performed through the gasification of the deposited carbon with 20 cm3 /min of air at 600 ◦ C until the formation of CO2 no longer occurred. Then, the catalysts were reduced with hydrogen at 650 ◦ C for 1 h, and the catalytic performance in the CDM by the regenerated catalysts was retested. For each catalyst, the CDM reaction and subsequent regeneration were repeated twice. The gaseous reaction products were monitored by on-line gas chromatography (GC) using a TDX-01 column and a thermal conductivity detector (TCD) for H2 , CH4 , and CO2 analysis. 2.3. Catalyst characterization XRD patterns were recorded using a Bruker D8 Advance diffractometer (Cu Ka radiation at 40 kV and 40 mA). Micrographs of the catalysts and the carbon structures were recorded with an FEI Inspect F scanning electron microscope (SEM). The detailed morphological appearances of the carbon structures were observed using a Tecnai G2 F20 S-Twin transmission electron microscope (TEM) operated at 200 kV. For TEM analysis, samples were prepared by drying a drop of an alcohol suspension of the catalyst particles on a cobalt grid. Surface area (BET) and pore volume (PV) were determined from nitrogen adsorption/desorption isotherms, using a JW-BK200C (JWGB SCI&TECH) gas adsorption device. Before analysis, all samples were out-gassed at 100 ◦ C under vacuum (4 h). 3. Results and discussion 3.1. Effect of Cu on Ni-based catalysts during the CDM reaction Fig. 1 shows the kinetic curves for the CDM reaction over NiSiO2 and Ni-Cu-SiO2 catalysts at 650 ◦ C. Only hydrogen is obtained as a gaseous product over all the catalysts. The lifetime of the Ni-SiO2 catalyst without Cu is very short at 650 ◦ C. After 70 min, the methane conversion of 75%Ni-25%SiO2 decreases to ca. 5% and remains at this value. The introduction of 10% Cu significantly improves the lifetime of the Ni-SiO2 catalyst, and the methane conversion over 65%Ni-10%Cu-25%SiO2 is constant at 40% during the reaction. However, when the Cu content is increased to 20%, the methane conversion decreases to ca. 20%. Therefore, the introduc-
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287
281
Fig. 2. SEM micrographs of carbon deposited over Ni-SiO2 and Ni-Cu-SiO2 catalyst at 650 ◦ C. (a) 75%Ni-25%SiO2, (b) 65%Ni-10%Cu-25%SiO2 , and (c) 55%Ni-20%Cu-25%SiO2 .
Table 1 Cyclic CDM and subsequent regeneration with air over the three catalysts. Cu content
0 10% 20%
m(cat.) mg
10 3 3
The first cycle
The second cycle
The third cycle
Lifetime min
Hydrogen yield gH2 /gcat.
Lifetime min
Hydrogen yield gH2 /gcat.
Lifetime min
Hydrogen yield gH2 /gcat.
20 282 340
0.96 38.6 56.8
20 25 320
0.60 2.66 42.8
20 25 25
0.56 1.83 2.4
tion of Cu into Ni-SiO2 catalysts extends their lifetime; however, when more Cu is introduced into the catalysts, their catalytic activity is reduced. Fig. 2 shows the SEM images of the carbon structures formed over Ni-SiO2 and Ni-Cu-SiO2 catalysts, all of which were prepared at 650 ◦ C. Two kinds of carbon structures are formed over the 75%Ni25%SiO2 catalysts: filamentous carbon and spherical carbon [32]. However, only filamentous carbon is obtained over the Ni-Cu-SiO2 catalysts. Careful examination reveals the textural characteristics and microstructure of the carbon deposited on the Ni-SiO2 catalysts. Some single carbon filaments with smooth and regular shapes form over Ni-SiO2 catalysts, and a pear-shaped Ni metal particle is located at the tip [33] and the particle size is about 50 nm, as shown in Fig. 3(a). However, due to the high activity of Ni metal, pure Ni particles easily sinter into bigger clusters during the reduction process. With an increase in the size of the pure Ni particles, the carbon generation rates exceed that of the rates of carbon diffusion and precipitation, and the enrichment of that carbon leads to the formation of carbon layers on the surfaces of the particles [34]. Consequently, some spherical carbon structures are generated by the Ni-SiO2 catalysts during the CDM reaction, as shown in
Fig. 3(b). Therefore, the bigger Ni-Cu particles form the spherical carbon structures and show poor catalytic performance during the CDM reaction. In the case of the Ni-Cu-SiO2 catalysts, the Ni-Cu particles maintain their size in a suitable range during the reduction process due to the strong influence of Cu on the dispersion of Ni in the catalyst [35]. Therefore, few spherical carbons are observed for these catalysts. Moreover, it found that several filamentous carbon structures deposited on one Ni-Cu particle, and both of their particle sizes are about 100 nm, as shown in Fig. 3(c)–(d). In addition, Cu itself is inactive for methane decomposition and has a negligible effect on the carbon diffusion rate [36]. Therefore, Cu can dilute the planes of Ni, causing segregation of the Ni (111) planes, leading to the formation of several carbon filaments from one particle. This carbon structure is referred to as an octopus-like carbon nano-filament [37,38]. More importantly, the formation of the octopus-like sites are present on the Ni (111), which facilitates carbon transportation. This may be the reason that Ni-Cu-SiO2 catalysts have a better catalytic performance than Ni-SiO2 .
282
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287
Fig. 3. TEM micrographs of carbon deposited over Ni-SiO2 and Ni-Cu-SiO2 at 650 ◦ C. 75%Ni-25%SiO2 , (b) 75%Ni-25%SiO2 , (c) 65%Ni-10%Cu-25%SiO2 , (d) 55%Ni-20%Cu-25%SiO2 .
3.2. Catalytic performance of Ni-SiO2 and Ni-Cu-SiO2 catalysts during repeated CDM Fig. 4 shows the kinetic curves of the CDM reaction over NiSiO2 and Ni-Cu-SiO2 catalysts during three repeated CDM reactions. Comparing to the Ni-SiO2 and Ni-Cu-SiO2 catalysts in the reaction condition as shown in Fig. 1, due to the decrease of the mass of the catalysts and increased the flow rate of methane, the Ni-SiO2 and Ni-Cu-SiO2 catalysts both have shorter catalytic lifetime (from the initial stage to the moment of methane conversion blew 2.5%), especially for the 75%Ni-25%SiO2 catalysts that the methane conversion decreased blew 2.5% after just 20 min. Therefore, the high methane flow rate may promote high decomposition rates (enhance carbon atoms covered the surface of catalysts particles) and cause the activity decreased remarkable. Moreover, the catalytic performance of the Ni-SiO2 and Ni-CuSiO2 catalysts after regeneration in air are quite different from each other, as shown in Fig. 4(a). Compared to the first CDM reaction over 75%Ni-25%SiO2 catalysts, the lifetime of the 75%Ni-25%SiO2 catalyst during the second and third CDM reaction does not change significantly; however, the catalytic activity decreases markedly for the second and third CDM reaction. For the repeated CDM reaction over 65%Ni-10%Cu-25%SiO2 catalysts, the lifetime and the catalytic activity of the catalyst decreases markedly for the second and third CDM, as shown in Fig. 4(b). The 55%Ni-20%Cu-25%SiO2 catalyst shows the longest lifetime during the first CDM, and the lifetime of this catalyst during the second CDM does not decrease significantly, as shown in Fig. 4(c). Although the catalytic activity of the 55%Ni-20%Cu-25%SiO2 catalyst is a little
lower than that for the first CDM. This suggests that the catalytic performance of Ni-Cu-SiO2 during repeated CDM reactions is significantly influenced by the Cu content. However, when the third CDM over 55%Ni-20%Cu-25%SiO2 is performed, both the lifetime and the catalytic activity of the catalyst decrease significantly. The catalytic performances of the prepared catalysts during repeated CDM reactions are summarized in Table 1. 3.3. SEM images of the catalysts and carbon structures after repeated CDM reaction Fig. 5 shows the SEM images of carbon structures generated by 75%Ni-25%SiO2 catalysts during repeated CDM reactions. The images reveal that a large amount of spherical carbon forms over 75%Ni-25%SiO2 catalysts during the three CDM reactions. It may be observed that there are two kinds of carbon structure formed during the three CDM reactions. However, when the regeneration frequency is increased, the size of the carbon spheres increases. Fig. 6 shows SEM images of the carbon structures generated over the Ni-Cu-SiO2 catalyst during the CDM, and the subsequent regeneration with air cycles. Only carbon nano-filaments are generated over 65%Ni-10%Cu-25%SiO2 during the first cycle (Fig. 6(a)). However, when the deactivated 65%Ni-10%Cu-25%SiO2 is regenerated in air, some spherical carbon appears in the second and third cycle, as shown in Fig. 6(b)–(c). The structure of the filamentous carbon generated over 55%Ni-20%Cu-25%SiO2 is a little different from that generated over 65%Ni-10%Cu-25%SiO2 during the three cycles: only carbon filaments are generated during the first and second cycle (Fig. 6(d)–(e)), and some larger spherical carbon is produced
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287
283
Fig. 4. Methane decomposition-regeneration with air cycles over Ni-SiO2 and Ni-Cu-SiO2 catalysts ((a): 75%Ni-25%SiO2 , 10 mg; (b): 65%Ni-10%Cu-25%SiO2 , 3 mg; (c): 55%Ni20%Cu-25%SiO2 , 3 mg).
during the third cycle (Fig. 6(f)). Comparing the rates of methane conversion described in Fig. 6 and Table 1, it can be concluded that when the Ni-Cu particle sinter into a big cluster that it will generate the spherical carbon and show poor catalytic performance during the CDM reaction. Introduction of Cu not only promotes the catalytic performance of Ni-based catalysts, but also helps preserve the activity of Ni-based catalysts during the regeneration in air. However, with an increase in the number of regenerations, the NiCu-SiO2 catalysts lose their ability to crack methane, even when the Cu content is increased to 20%. Fig. 7 shows the XRD patterns of the fresh Ni-Cu-SiO2 catalysts and the samples after regeneration in air at 600 ◦ C, and the crystal sizes of the three samples are calculated by Debye-Scherrer formula from the most intense peak (Ca. 43.25◦ ) of NiO. As shown in Fig. 7(a)–(b), the Ni and Cu species in the regenerated Ni-Cu-SiO2 catalysts mainly exist in the form of NiO and CuO. Moreover, due to carbon phase is not noted to exist, so it can prove that carbon atoms have been removed from the surface of the catalysts. The XRD characterization results also show the change of the NiO particle size, as shown in Fig. 7(a)–(b). For the 65%Ni-10%Cu-
25%SiO2 catalysts, the particle size is changed from 24.7 to 27.5 nm after the first regeneration, however, it barely increases after the second regeneration. Different from the 65%Ni-10%Cu-25%SiO2 catalysts, the particles size is not changed after the first regeneration, but it remarkable increased after the second regeneration (from 22.3 to 24.7 nm). Therefore, it can be concluded that the aggregation process of NiO particles can be reduced in some extent through increasing of Cu content after the first regeneration. However, with an increase in the number of regenerations, the average of the NiO particles increases over Ni-Cu-SiO2 catalysts even the Cu content increases to 20%. Fig. 8 shows the XRD spectra of the fresh Ni-Cu-SiO2 catalysts and the regenerated catalysts after hydrogen reduced, and the crystal sizes of the three samples are calculated by DebyeScherrer formula from the most intense peak (Ca. 44.5◦ ) of Ni-Cu alloy. The diffraction peaks due to Cu is not detected in NiCu-SiO2 catalysts even if the Cu content reaches 20% or the catalysts regenerated twice. Moreover, the Ni-Cu particle size over 55%Ni-20%Cu-25%SiO2 catalysts is much smaller the 65%Ni-10%Cu25%SiO2 catalyst in each the same cycle. Similar with Fig. 7, with
284
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287
Fig. 5. SEM images of carbon structures formed over 75%Ni-25%SiO2 catalysts during repeated methane decomposition ((a): the first reaction; (b): the second reaction; and (c): the third reaction).
an increase in the number of regenerations, the average of the Ni-Cu particles increased over 65%Ni-10%Cu-25%SiO2 and 55%Ni20%Cu-25%SiO2 catalysts. However, the average particle size of 55%Ni-20%Cu-25%SiO2 catalysts does not obviously increased after the first regeneration in air at 600 ◦ C, which might prove that the generated catalysts can maintain good catalytic performance during the CDM reaction. To detailed investigated the relationship between the Ni-CuSiO2 catalysts and catalytic performance, the nitrogen adsorption/desorption isotherms is used to evaluate their surface structures. Although the reaction and regeneration temperature might be a little higher for Sol-gel derived materials, the BET surface area, pore volume and micropore volume of Ni-Cu-SiO2 catalysts increased with the number of regenerations Table 2. For the NiCu-SiO2 catalyst, the BET surface area of 65%Ni-10%Cu-25%SiO2 catalyst increases from 37.701 to 49.332 m2 /g after two regeneration processes, and the BET surface area of 55%Ni-20%Cu-25%SiO2 catalysts increased from 27.834 to 45.023 m2 /g. Moreover, no matter how many times the catalyst is generated, it has the same average pore diameter. Therefore, the results can states that the pore structures of the Ni-Cu-SiO2 catalyst do not break during the reaction and regeneration process. Therefore, the increase of the Ni-Cu particle size is the key factor to cause the catalysts exhibited poor catalytic performance after the regeneration in air. Several studies [25] have reported that the carbon filaments formed over Ni-Cu-SiO2 catalysts obeyed a “tip-growth” model, and that the Ni-Cu particle was located at the tip of the filaments; therefore, the Ni-Cu clusters did not interact with the SiO2 support during the carbon filament growth process. In this work, when the deac-
tivated Ni-Cu-SiO2 catalysts are regenerated in air at 600 ◦ C, the carbon is removed from the Ni-Cu species by combustion. In this situation, Ni-Cu species easily sinter into larger clusters because of the weak interaction between the particle and SiO2 support. As described in Section 3.2, due to the Ni-Cu clusters sintering into larger particles, the carbon diffusion and deposition rate is lower than the methane decomposition rate over these catalysts during the CDM process. Consequently, the enrichment of the carbon encapsulates the active sites of the metal by the formation of graphite layers or destruction of the active metal [12]. Therefore, spherical carbon is generated during this process. However, though introduction of Cu enhances the dispersion of Ni particles, the NiCu particles without support still exhibit an increase in particle size during the regeneration process to some extent, even when the Cu content is 20%. 4. Conclusions The catalytic performance of 75%Ni-25%SiO2 , 65%Ni-10%Cu25%SiO2 , and 55%Ni-20%Cu-25%SiO2 during repeated methane decomposition decreased with the number of repetitions for all three catalysts. 55%Ni-20%Cu-25%SiO2 exhibited better catalytic performance than the other catalysts, affording hydrogen yields for the repeated methane decomposition of 56.8, 42.8, and 2.4 gH2 /gcat. The temperature programmed reduction of the Ni-CuSiO2 catalysts indicated that Ni-Cu species were present in the fresh Ni-Cu-SiO2 catalyst, although this phenomenon was not observed in the XRD spectra of the fresh Ni-Cu-SiO2 catalyst. The XRD studies of Ni-Cu-SiO2 catalysts reduced by H2 indicated that a Ni-Cu
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287
285
Fig. 6. SEM images of the Ni-Cu-SiO2 catalyst after methane decomposition-regeneration with air (65%Ni-10%Cu-25%SiO2 : a: the first cycle; b: the second cycle; c: the third cycle; 55%Ni-20%Cu-25%SiO2 : d: the first cycle; e: the second cycle; f: the third cycle).
Table 2 the surface structures of the fresh Ni-Cu-SiO2 catalysts and the samples after regeneration by air. Samples Regeneration times
BET surface areas (m2 /g) Average pore diameter (nm) Pore volume (cm3 /g) Micropore volume (10−3 cm3 /g)
65%Ni-10%Cu-25%SiO2 catalysts
55%Ni-20%Cu-25%SiO2 catalysts
0
1
2
0
1
2
37.701 18.673 0.225 0.0153
40.684 20.848 0.239 0.0109
49.332 20.917 0.587 0.0153
27.834 16.925 0.131 0.0108
40.926 19.979 0.227 0.0125
45.023 18.389 0.168 0.0157
286
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287
(a)
(a)
(b)
(b)
Fig. 8. XRD patterns of the fresh Ni-Cu-SiO2 catalysts and the regenerated catalysts after hydrogen reduced ((a): 65%Ni-10%Cu-25%SiO2 catalysts; (b): 65%Ni-10%Cu25%SiO2 catalysts). Fig. 7. XRD patterns of the fresh Ni-Cu-SiO2 catalysts and the samples after regeneration by air ((a): 65%Ni-10%Cu-25%SiO2 catalysts; (b): 65%Ni-10%Cu-25%SiO2 catalysts).
alloy was formed in the Ni-Cu-SiO2 catalyst reduced by H2 . Spherical carbon structures were observed following all three repeated CDM reactions for 75%Ni-25%SiO2 . However, similar carbon structures were only observed following the second and third CDM for 65%Ni-10%Cu-25%SiO2 , and only following the third one for 55%Ni20%Cu-25%SiO2 . Moreover, the BET surface area, pore volume and micropore volume of Ni-Cu-SiO2 catalysts increased with the number of regenerations. The increase of the Ni-Cu particle during the regeneration process strongly influenced the performance of the catalysts, and Ni-Cu particle which suffered an aggregation process (sinter into big cluster) exhibited poor performance and formed the spherical carbon structures during the CDM reaction. Acknowledgements The authors are grateful for the financial support by the Science and Technology Development Foundation (No.2013B0301036) and
the Programs of Laboratory of Advanced Nuclear Energy (LANE), managed by the Institute of Nuclear and Chemistry(No. R20140303-03). References [1] M. Glugla, L. Dörra, R. Lässera, D. Murdoch, H. Yoshidac, Fusion Eng. Des. 61 (2002) 569–574. [2] C. Rizzelloa, F.B.T. Pinnab, S. Tosti, Fusion Eng. Des. 85 (1) (2010) 58–63. [3] M. Glugla, D.K. Murdoch, H. Geißler, P. Herrmann, Kraemer, Fusion Eng. Des. 39–40 (1) (1998) 893–899. [4] R. Lasser, M. Glugla, K. Gunther, T.L. Le, D. Niyongabo, R.D. Penzhorn, K.H. Simon, Fusion Eng. Des. 58–59 (2001) 371–375. [5] S. Tosti, L. Bettinali, F. Borgognoni, D.K. Murdoch, Fusion Eng. Des. 82 (2) (2007) 153–161. [6] M. Glugla, A. Perevezentsev, D. Niyongabo, R.D. Penzhorn, A. Bell, P. Herrmann, Fusion Eng. Des. 49 (-50) (2000) 817–823. [7] M. Glugla, D.K. Murdoch, Fusion Eng. Des. 39-40 (1998) (893–869). [8] M. Glugla, A. Antipenkov, S. Beloglazov, Fusion Eng. Des. 82 (5–14) (2007) 472–487. [9] A.L. Dicks, J. Power Sources 61 (1996) 113–124. [10] S.S. Penner, Energy 31 (2006) 33–43. [11] R.A. Lemons, J. Power Sources 29 (1990) 251–264. [12] B.C.H. Steele, A. Heinzel, Nature 414 (2001) 345–352.
J. Li et al. / Fusion Engineering and Design 113 (2016) 279–287 [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25]
J.H. Wee, K.Y. Lee, J. Power Sources 157 (2006) 128–135. J.N. Coleman, U. Khan, W.J. Blau, Y.K. Gun’ko, Carbon 44 (2006) 1624–1652. W. Zhang, Q.J. Ge, H.Y. Xu, J. Energy Chem. 20 (2011), 339–334. A. Chatla, V.K. Velisoju, K.B. Suresh, V. Akula, J. Energy Chem. 22 (2013) 853–860. S. Tajammul Hussain, S. Gul, M. Mazhar, D.H. Anjum, F. Larachi, J. Nat. Gas Chem. 17 (4) (2016) 374–382. Y. Wang, J. Peng, C. Zhou, Z.Y. Lim, Int. J. Hydrogen Energy 39 (2014) 778–787. S.K. Saraswat, K.K. Pant, Int. J. Hydrogen Energy 13 (2013) 52–59. J.L. Pinilla, R. Utrilla, M.J. Lázaro, Fuel Process. Technol. 2111 (92) (2016) 1480–1488. W.H. Wang, H.Y. Wang, Y. Yang, Int. J. Hydrogen Energy 37 (2012) 9058–9066. N.Z. Muradov, T.N. Veziroglu, Int. J. Hydrogen Energy 30 (2005) 225–237. Y. Zhang, K.J. Smith, Catal. Lett. 95 (2004) 7–12. H. Ogihara, S. Takenaka, I. Yamanaka, E. Tanabe, A. Genseki, K. Otsuka, J. Catal. 238 (2006) 353–360. A.C. Lua, H.Y. Wang, Appl. Catal. B Environ. 132–133 (2013) 469–478.
[26] [27] [32] [33] [34] [35] [36] [37] [38]
287
Y. Li, J. Chen, L. Chang, Y. Qin, J. Catal. 178 (1998) 76–83. M.A. Ermakova, D.Y. Ermakov, Catal. Today 77 (3) (2002) 225–235. N.Z. Muradova, T.N. Veziroglu, Int. J. Hydrogen Energy 30 (2005) 225–237. J.C. Guevara, J.A. Wang, L.F. Chen, M.A. Valenzuela, P. Salas, Int. J. Hydrogen Energy 35 (8) (2010) 3509–3521. T.V. Reshetenko, L.B. Avdeeva, Z.R. Ismagilov, A.L. Chuvilin, V.A. Ushakov, Appl. Catal. A Gen. 247 (1) (2003) 51–63. Y. Echegoyen, I. Suelves, R. Lázaro Moliner, J.M. Palacios, J. Power Sources 169 (2007) 150–157. H.Y. Wang, A.C. Lua, Chem. Eng. J. 169 (2007) 150–157. L.B. Avdeeva, O.V. Goncharova, D.I. Kochubey, V.I. Zaikovskii, L.M. Plyasova, B.N. Novgorodov, S.K. Shaikhutdinov, Appl. Catal. A Gen. 141 (1996) 117–129. T.V. Aapg BulletinReshetenko, L.B. Avdeeva, Z.R. Ismagilov, A.L. Chuvilin, V.A. Ushakov, Appl. Catal. A Gen. 247 (2003) 51–63.