Вы находитесь на странице: 1из 7

J. Anal. Appl. Pyrolysis 73 (2005) 335341 www.elsevier.

com/locate/jaap

Catalytic decomposition of methane over activated carbon


Zongqing Bai a,b, Haokan Chen a,*, Baoqing Li a, Wen Li a
a

State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, P.O. Box 165,Taiyuan, Shanxi 030001, PR China b Graduate School of Chinese Academy of Sciences, Beijing 100039, China Received 24 September 2004; accepted 18 March 2005 Available online 18 April 2005

Abstract Methane decomposition over activated carbons was carried out in a xed bed, quartz-tube ow reactor. The kinetics of methane decomposition and surface properties changes before and after reaction was investigated. As a non-carbon based material, active alumina was also used to compare and understand the catalytic decomposition mechanism of methane over different materials. A reaction order of 0.5 is obtained for methane decomposition over ACs and activation energies in the range from 117 to 185 kJ mol1. The pore size change in the course of methane decomposition over activated carbons indicates that the catalytic reaction occurs mainly in the micropores. Activated alumina has different pore properties and carbon deposition in mesopores of the alumina is responsible for the catalytic activity. # 2005 Elsevier B.V. All rights reserved.
Keywords: Catalytic decomposition; Methane; Activated carbons

1. Introduction Recently, people show great concern about the climate changes resulting from the emission of greenhouse gases in particular CO2 and at the same time environmental friendly energy source is strongly required. Hydrogen is the best choice because its combustion product is H2O. However, conventional hydrogen production methods, such as methane steam reforming and methane partial oxidation, have disadvantages of hydrogen purication because of byproducts of CO, CO2 and the consequence of CO2 storage. While the direct thermal decomposition of methane has advantages over other methods because of the simple product of C and H2 based on which the carbon black production has been industrialized. As it is known, methane is the most stable hydrocarbon having a high decomposition temperature without catalyst. So the activation of methane i.e., hydrogen production by catalytic decomposition of methane has been widely researched [16]. The catalysts used in these researches are mostly based on transition metal such as Ni, Fe, Co etc., which show remarkable activity. But
* Corresponding author. Tel.: +86 351 404 8967; fax: +86 351 404 8967. E-mail address: chhk@sxicc.ac.cn (H. Chen). 0165-2370/$ see front matter # 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.jaap.2005.03.004

the regeneration of the deactivated catalyst is always by burning off or gasifying the deposited carbon on the catalyst surface leading to CO2 production again. A new process of hydrogen production by methane decomposition over carbon-based catalyst was proposed by Muradov recently [79]. The carbon catalyst shows several advantages over metal catalyst: (i) low price; (ii) needless regeneration of the catalyst; and (iii) valuable carbon byproduct; and (iv) hydrogen production without COx emission during the process. Muradov [9] investigated several kinds of carbon including activated carbon (AC), carbon black (CB), graphite, diamond, carbon bers and the carbon nanotubes in a xed bed reactor among which the ones having disordered forms of carbon, such as AC, CB, are more catalytically active than the ordered ones (e.g. graphite and diamond). Lee et al. [10] used a uidized bed reactor to produce CO2 free hydrogen by decomposition of methane with activated carbon catalyst. The authors concluded that the reaction rate and deactivation of carbon catalyst were similar to those of xed bed reactor system and the deactivation of catalyst was resulted from the carbon deposition. However, there are few references in literature on the catalytic decomposition of methane over activated carbon

336

Z. Bai et al. / J. Anal. Appl. Pyrolysis 73 (2005) 335341

and many are not well understood about the process of catalytic decomposition of methane, such as the characteristic of the active site of carbon, catalytic performance differences between the carbon based catalyst and the non-carbon based catalyst, surface properties changes of activated carbon before and after the reaction and so on. In this work, methane decomposition in a xed bed reactor was studied to understand the differences of catalytic performance between carbon-based catalyst and the noncarbon based catalystactive alumina. At the same time the kinetics of methane decomposition over activated carbon and the surface properties changes of carbon and the active alumina were also investigated. 2. Experimental Several different commercial activated carbons (DX40, CB10, MZ10, ZL30) and an active alumina were employed as catalysts in methane decomposition. AC MZ10 is derived from hardwood and other activated carbons (ACs) are from coal with steam activation for all carbons. The surface area for MZ10 is 783 m2/g, for CB10 is 734 m2/g, for ZL30 is 738 m2/g and for active alumina is 152 m2/g. All catalysts were grinded and sieved to 246833 mm and dried at 383 K for 24 h before the experiment. Methane (99.99%) was used directly without further purication. The decomposition reaction was carried out in a quartztube xed bed reactor heated by an electric furnace. The experiments were conducted at temperature ranges from 1023 to 1173 K and the volumetric hourly space velocity (VHSV) was from 3.0 to 18.0 l g1 h1. Before introducing the reactant gas, the system was heated up to the set temperature under Ar. In the kinetic study, the reaction conditions for all catalysts are the same as follows: catalyst charge of 1g, methane ow rate of 100 ml/min, i.e., VHSV of 6.0 l g1 h1, which has been proved eliminating the inuence of diffusion. The activated carbon and the active alumina were weighed before and after each experiment. All experiments were conducted at atmospheric pressure. The outow gas was analyzed by gas chromatography equipped with a 5A molecular sieve column and TCD detector using Ar as a carrier. Another gas chromatography with a n-octadecane column and FID detector was used to analyze C1C4 hydrocarbon. The rst analysis was done 5 minutes after methane was owed to ush out the argon gas. Very small amount of hydrocarbon other than methane (<0.1 vol.%) is detected in the efuent gas during the experiment and there are no liquid products. So carbon and hydrogen can be regarded as the only products of methane decomposition, that is, only the following reaction occurs during the process: CH4 ! C 2H2 ; DH 0 75:6 kJ=mol (1)

The conversion of methane can be calculated using the concentration of hydrogen in the efuent gas as the following formula: H2 % 100% (2) XCH4 % 200 H2 % where XCH4 % is methane conversion at given time and H2% the hydrogen concentration in the efuent gas (% by volume) determined by gas chromatography. The surface properties of the fresh and the used samples were measured by N2 adsorption at 77 K in a Tristar 3000 physical adsorption apparatus. The specic surface area and pore volume were calculated applying the BET method and BarrettJoynerHalenda (BJH) numerical integration method, respectively. BJH desorption model was used to calculate the pore size distribution of the samples. The SEM images of the samples were analyzed by H600 electron microscope.

3. Results and discussion 3.1. Methane decomposition over activated carbons Fig. 1 shows the general performance of ve activated carbons on methane decomposition at 1123 K. For all catalysts methane conversion shows the highest value of 25 35% at the beginning of the reaction and then it gradually decreases. The highest activity of ACs in the initial stage follows a rapid drop and then a smooth decrease until to a steady state which is similar as another work [9]. AC CB10 shows the highest initial activity on methane decomposition while AC ZL30 shows the lowest. Methane decomposition over ZL30 shows lower conversion than other four ACs during the whole process, which is not proportional to its surface area of 738 m2/g. One reason is that only a part of the surface of AC ZL30 is active for methane decomposition [11] even though its whole surface area is only a little smaller than others The deactivation behavior of catalyst is

Fig. 1. Methane conversion as a function of time over ve ACs at 1123 K (VHSV = 3.0 l g1 h1).

Z. Bai et al. / J. Anal. Appl. Pyrolysis 73 (2005) 335341

337

Fig. 2. Methane conversion as a function of time over active alumina at 1123 K (VHSV = 3.0 l g1 h1).

Fig. 5. Arrhenius plot of ln k as a function of 1/T for four ACs.

Fig. 3. Initial rate of methane decomposition over ACs at different methane partial pressure.

apparently due to the carbon deposition in pores of ACs produced by methane decomposition, which is proved by the increasing weight of catalysts after reaction. 3.2. Methane decomposition over active alumina To well understand the mechanism of methane catalytic decomposition, methane decomposition over non-carbon

based material, active alumina with a surface area of 152 m2/ g was conducted in the xed bed reactor under the same conditions as over ACs. Fig. 2 shows the result of methane conversion over active alumina. It shows a completely different trend and much lower catalytic activity than those of activated carbons under the same experimental conditions. In the initial 20 min methane conversion quickly increases to about 11% and maintains for about 10 min then gradually decreases to about 6% at the end of the reaction. Methane decomposition over alumina can be simply divided into two steps: carbon nuclei formation and carbon crystallites growth, and it has been known that the activation energy of carbon nuclei formation is 317 kJ mol1 and that of carbon crystallites growth is 227 kJ mol1 [6,7,9]. Hence, the apparent decomposition rate is the sum of two steps. So the special trend of methane decomposition over active alumina can be possibly explained as follows: the rate of rst step of methane decomposition on the surface of alumina is slow because of higher activation energy, but once the carbon nuclei have been formed, the carbon crystallites growth with lower activation energy becomes much faster resulting in the increase of methane decomposition rate and the increasing of methane conversion. This is followed by more and more carbon deposition, which has lower activity

Fig. 4. Effect of temperature on methane decomposition over ACs (a): AC MZ10, (b): AC CB10.

338

Z. Bai et al. / J. Anal. Appl. Pyrolysis 73 (2005) 335341

Table 1 Activation Energies of methane decomposition over ACs AC AC AC AC AC DX40 MZ10 CB10 ZL30 Ea (kJ mol1) 116.9 133.6 140.0 184.9

area, which is important for the methane adsorption capacity of catalysts. 3.3. Catalytic decomposition kinetics 3.3.1. Conrm of reaction order The rate of methane decomposition over ACs can be represented as follows because the catalyst deactivated rapidly [12]:   t Rt R0 exp (3) t0 where R0 and Rt are the initial rate and the rate at a given time in the unit of mmol/min gcat, t is the time on-stream, and t0 the time constant for deactivation. R0 can be

than the fresh carbon on the active carbon sites in the pores of alumina and gradually decreases methane decomposition rate [9]. This is an autocatalytic reaction of carbon deposition. The whole catalytic activity of alumina is lower than that of AC because of the lack of the active sites at the beginning of the reaction, and the considerably lower population of active carbon sites in the course of the reaction. Another important factor is the much lower surface

Table 2 Surface properties changes of three ACs and alumina before and after methane decomposition for 2 h Sample Al2O3 Before reaction BET surface area (m2/g) Average pore diameter (nm) Pore volume (cm3/g) Micropore volume (103 cm3/g) 152.2 19.2 0.73 7.10 After reaction 148.7 13.7 0.51 2.38 AC MZ10 Before reaction 783.0 2.2 0.43 260 After reaction 11.9 6.6 0.02 0.112 AC CB10 Before reaction 734.7 2.0 0.36 284 After reaction 3.6 14.6 0.01 0.258 AC ZL30 Before reaction 738.0 2.1 0.40 85 After reaction 2.3 14.3 0.01 0.100

Fig. 6. Pore size distribution of three ACs and alumina before and after methane decomposition for 2 h.

Z. Bai et al. / J. Anal. Appl. Pyrolysis 73 (2005) 335341

339

estimated by plotting ln Rt versus t through extrapolating to zero time. To determine the reaction order, methane decomposition over ACs was performed at different methane partial pressure using N2 as the diluent. The initial rate was obtained according to the method mentioned above. Plotting ln R0 against methane partial pressure as a ln form derives a line and the slope is just the reaction order. The results are shown

in Fig. 3 in which three slopes are all about 0.5. So the reaction order of methane decomposition over ACs is 0.5. The initial rate equation can be written as follow: R0 k p0:5 (4)

where k is the reaction rate constant and p the partial pressure of methane. The reaction order of 0.5 of methane

Fig. 7. SEM images of the catalyst surface before and after the reaction (a): MZ10, (b): ZL30, (c): Al2O3.

340

Z. Bai et al. / J. Anal. Appl. Pyrolysis 73 (2005) 335341

decomposition over ACs is consistent with the result in the previous study [11]. 3.3.2. Effect of temperature The inuences of temperature on methane conversion over AC MZ10 and CB10 are presented in Fig. 4. As the temperature increases, the initial rates of methane decomposition over both two ACs increase because the methane decomposition is an endothermic reaction. While the deactivation of ACs become more rapidly at high temperature because of rapid carbon deposition in the active sites [11], which is especially apparent at 1173 K. The initial rates of methane decomposition at different temperature are estimated by the extrapolation used in Section 3.3.1 and then the reaction rate constants are calculated by using Eq. (4). The activation energies over ACs are determined by using Arrhenius plot of ln k against 1/T in which the slope of the linear plot is Ea/R. The results are shown in Fig. 5 and summarizing of the results is shown in Table 1. For the four ACs, there is a wide range of activation energies from 117 to 185 kJ mol1 possibly caused by difference in the structure of active sites. The activation energy obtained by other work was reported to be about 190 kJ mol1 [11] which is similar to that of AC ZL30 in this study. Also, AC ZL30 with relative low methane decomposition activity has the highest activation energy than other three. But this is much lower than the methane CH bond dissociation energy 440 kJ mol1 and ACs obviously have catalytic effect on methane decomposition. 3.4. Surface properties changes To well understand the process of methane decomposition over ACs, the surface properties changes including pore properties and surface images are measured for the fresh and used ACs and alumina. Some of their surface properties are summarized in Table 2 and the pore size distribution changes are shown in Fig. 6. For three ACs, the BET surface areas are all greatly decreased after reaction in the methane stream at 1123 K for 2 h. For example, the BET surface area of ZL30 decreases the most from 738 to 2.3 m2/g. The pore volume and micropore volume also decrease greatly for three ACs and the decrease of micropore volume is especially great. These are all resulting from the carbon deposition especially in the micropores of ACs, which contributes greatly to the surface area and pore volume of ACs. From the pore size distribution changes of ACs in Fig. 6 the decrease of micropore volume and the increase of average pore diameter can be easily seen. So the adsorption and decomposition of methane occur mainly in the micropores of ACs because of all above changes. Interestingly, the surface properties changes of alumina are quite different from ACs. After reaction for 2 h, the surface area, pore volume and the micropore volume of alumina all only decrease a little. A large decrease is found for the average pore diameter of alumina, while it is a large increase for that of ACs. Due to

the lower activity for methane decomposition, the smaller amount of carbon deposition found in alumina leads to the less decrease of surface area and pore volume. At the same time, the carbon deposition in the mesopores of alumina leads to the partial blockage of the mesopores and the consequent decrease of the average pore diameter. The SEM image changes of the surfaces of two ACs and alumina are shown in Fig. 7. Carbon deposition is clearly seen on the surfaces of MZ10 and ZL30 while it is nearly invisible on alumina surface because of low activity for methane decomposition. MZ10 and ZL30 show different images because they are derived from two different materials of hardwood and coal. At the same time, carbon produced at two ACs also shows different types distinguished by naked eyes. A part of the carbon deposition on ZL30 is something in the form of carbon lament, which has been observed on metal catalyst [13]. While the carbon produced on MZ10 is irregular agglomerate form and no carbon lament is observed. The formation of carbon lament is possibly due to the ash in the AC derived from coal such as Fe, Na and K [10], which needs further investigation in detail.

4. Conclusion Activated carbons show higher activity on methane decomposition than active alumina and the catalytic behavior is different too. The difference is from the two steps of methane decomposition over catalyst: carbon nuclei formation and carbon crystallites growth with activation energies of 317 and 227 kJ mol1, respectively. Methane decomposition over ACs shows a reaction order of 0.5 and the activation energies of about 117 to 185 kJ mol1 which is much lower than the methane CH bond dissociation energy of 440 kJ mol1. The decreases of surface area, pore volume and micropore volume while the increase of average pore diameter show that methane decomposition over ACs occurs mainly in micropores. The quite different surface properties changes of active alumina show that the reaction mainly occurs in mesopores, which are the majority of alumina. The carbon deposition over ACs shows either carbon lament or carbon agglomerate because of the difference resource of activated carbon.

Acknowledgement This work was nancial supported by Shanxi Natural Science Fund (NO 20041033).

References
[1] M.G. Poirier, C. Sapundzhiev, Int. J. Hydrogen Energy 22 (1997) 429. [2] A. Steinfeld, V. Kirillov, G. Kuvshinov, Y. Mogilnykh, A. Reller, Chem. Eng. Sci. 52 (1997) 3599.

Z. Bai et al. / J. Anal. Appl. Pyrolysis 73 (2005) 335341 [3] R. Aello, J.E. Fiscus, H.C. Loye, M.D. Amiridis, Appl. Catal. A192 (2000) 227. [4] T.V. Choudhary, C. Sivadinarayana, C.C. Chusuei, A. Klinghoffer, D.W. Goodman, J. Catal. 199 (2001) 9. [5] M. Serban, M.A. Lewis, C.L. Marshall, R.D. Doctor, Energy Fuels 17 (2003) 705. [6] N. Muradov, in: Proceedings of the 13th World Hydrogen Energy Conference, vol. 1, Beijing, China, 2000, p. 428. [7] N. Muradov, Energy Fuels 12 (1998) 41.

341

[8] N. Muradov, Int. J. Hydrogen Energy 26 (2001) 1165. [9] N. Muradov, Catal. Commun. 2 (2001) 89. [10] K.K. Lee, G.Y. Han, K.J. Yoon, B.K. Lee, Catal. Today 9395 (2004) 81. [11] M.H. Kim, E.K. Lee, J.H. Jun, G.Y. Han, B.K. Lee, T.J. Lee, K.J. Yoon, Int. J. Hydrogen Energy 29 (2004) 187. [12] C. Naccache, in: J. Oudar, H. Wise (Eds.), Deactivation and Poisoning of Catalysts, Marcel Dekker, New York, 1985, p. 185. [13] T. Zhang, M.D. Amirids, Appl. Catal. A: Gen. 167 (1998) 161.

Вам также может понравиться