Al2O3 catalyst

Catalysis Communications 11 (2009) 220–224

Contents lists available at ScienceDirect

Catalysis Communications journal homepage: www.elsevier.com/locate/catcom

Growth of carbon nanofibers synthesized from CO2 hydrogenation on a K/Ni/Al2O3 catalyst Ching-Shiun Chen a,*, Jarrn-Horng Lin b, Jia-Huang Wu a, Cheng-Yu Chiang b a b

Center for General Education, Chang Gung University, 259 Wen-Hwa 1st Road, Kwei-Shan Tao-Yuan 333, Taiwan, ROC Department of Materials Science, National University of Tainan, 33, Sec. 2, Shu-Lin St., Tainan 700, Taiwan, ROC

a r t i c l e

i n f o

Article history: Received 6 June 2009 Received in revised form 2 October 2009 Accepted 7 October 2009 Available online 13 October 2009 Keywords: CO2 hydrogenation Ni/Al2O3 catalyst Carbon nanofibers Potassium Carbon dioxide

a b s t r a c t We used a commercially available Ni/Al2O3 sample containing K additive to enable carbon deposition from CO2 exposure by means of catalytic hydrogenation. Our experimental results suggest that K additives induce the formation of carbon nanofibers (CNFs) or carbon deposition on Ni/Al2O3 during the CO2 hydrogenation reaction. We propose that the rate of carbon deposition depends on the reaction temperature, on H2 and CO2 partial pressures, and on the reactant residence time. Furthermore, we suggest that the creation of a K-relevant active phase may be involved in the carbon deposition process. Our results also indicate that the degree of CNF graphitization can change with reaction time. Ó 2009 Elsevier B.V. All rights reserved.

1. Introduction Carbon dioxide, primarily generated by combustion, is the most important and abundant greenhouse gas and is the main contributor to the greenhouse effect. To address this problem, researchers have proposed the catalytic chemical conversion of atmospheric CO2 into solid carbon [1]. This would be a very powerful CO2 recycling technique with an easily stored end product. The reverse water gas shift (RWGS) reaction is known to provide high-efficiency CO production from CO2 (H2 + CO2 ? CO + H2O) [2–7]. Recently, we have described a novel method for converting CO2 to solid carbon through the hydrogenation of CO2 on K/Ni/Al2O3 catalysts at 500 °C [8]; we discovered that enhancing the Ni catalyst with K can enable the formation of carbon nanofibers (CNFs) from CO2. The rate of formation of solid carbon can be enhanced by increasing the K content of the Ni catalyst. A variety of potential applications for CNTs and CNFs have been reported [9–18], such as catalyst supports, electrode materials, gas storage materials, gas sensors, flat panel displays and adsorbents. Both reactions in CO2 hydrogenation on K/Ni/Al2O3 may occur:

H2 þ CO2 ! CO þ H2 O DH ¼ þ41 kJ=mol 2H2 þ CO2 ! C þ 2H2 O DH ¼ 90:5 kJ=mol

The overall energy of CO2 hydrogenation is 49.5 kJ/mol. Methanol is a leading candidate as hydrogen carrier [19,20], due to its relatively low reforming temperatures, easy availability and yielding large amount of hydrogen. Hydrogen can be extracted from methanol via two process alternatives: partial oxidation and steam reforming reactions.

2CH3 OH þ O2 ! 2CO2 þ 4H2 CH3 OH þ H2 O ! CO2 þ 3H2

DH ¼ 192:2 kJ=mol DH ¼ þ49:5 kJ=mol

One can see that carbon formation in CO2 hydrogenation may be thermodynamically favored in the typical temperature range, when H2 is formed from methanol reactions. In the present paper, we report on our use of a variety of techniques to carry out a detailed kinetic analysis of CNF formation from H2/CO2, and a detailed characterization of CNFs in the growth process using X-ray diffraction spectroscopy (XRD), X-ray photoelectron spectroscopy (XPS) and temperature-programmed desorption (TPD). We also discuss the effects of H2 and CO2 concentrations on carbon deposition and on the growth mechanisms of CNFs in the carbon structure. 2. Experimental 2.1. Catalyst preparation

* Corresponding author. Tel.: +886 32118800x5685; fax: +886 32118700. E-mail address: [email protected] (C.-S. Chen). 1566-7367/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.catcom.2009.10.012

The catalyst used in the present study was a commercially available Ni/Al2O3 formulation (12 wt.%) manufactured by

221

C.-S. Chen et al. / Catalysis Communications 11 (2009) 220–224

Süd-Chemie Catalysts, Inc. (catalyst # FCR-4). All K/Ni catalysts were prepared by adding the requisite volume of aqueous KNO3 to the Ni/Al2O3 catalyst without pretreatment; impregnated samples were subsequently air-dried at 80 °C for 10 h. All catalysts were used after calcination in air and reduction in H2 at 500 °C for 5 h. 2.2. Synthesis of carbon nanofibers The RWGS reaction and all carbon syntheses were carried out in a fixed-bed reactor (outer diameter: 0.95 cm) at atmospheric pressure. Fifty milligrams samples of catalyst were used for all reactions, which were carried out by feeding a stream of H2/CO2 over the catalyst at 100 mL/min. The impact of residence time on carbon formation was assessed through experiments in a H2/CO2 (1:1) stream that was controlled by varying the flow rate from 40 to 140 mL/min. Rates of carbon formation were inferred from linear plots of yields (g-carbon/g-cat.) versus reaction time within an initial 5 h period. All gaseous products were analyzed via gas chromatography using a thermal conductivity detector and a 12-ft Porapak-Q column. 2.3. X-ray photoelectron spectroscopy (XPS) XPS data were obtained using a Thermo VG-Scientific Sigma Probe spectrometer at the Precision Instrument Center of the College of Engineering at the National Central University, Taiwan. The spectrometer was equipped with an Al Ka X-ray source (1486.6 eV; 1 eV = 1.602  1019 J) that operated at 108 W and a hemispherical analyzer that operated at a pass energy of 50 eV. The instrument typically operated at an analysis chamber pressure of approximately 1  109 torr. 2.4. Temperature-programmed desorption (TPD) TPD experiments were performed using H2 and He streams at atmospheric pressure in a fixed-bed flow system. The temperature was increased from 25 to 1000 °C at a rate of 10 °C/min over the course of the TPD process. Concentrations of CH4, CO and CO2 were measured with a VG Smart IQ+ 300D mass spectrometer. The desorbed products were admitted into the vacuum chamber through a leak valve using H2 or He as the carrier gas. The chamber operating pressure was approximately 3  107 torr, and the chamber base pressure was approximately 2  109 torr. 3. Results 3.1. Effect of reaction parameters on CNF synthesis Table 1 compares the rates of carbon deposition on 7% K + Ni/ Al2O3 catalyst (K/Ni/Al2O3) obtained with a variety of H2 and CO2 partial pressures. As evident from the data, carbon formation was improved by increasing both H2 and CO2 concentrations. A low ratio of H2/CO2 concentration was insensitive to measurements of carbon deposition in our investigation, thus a partial pressure of CO2 below 380 torr was suggested. Fig. 1A reveals the effect of temperature on the formation rates of carbon, CO and CH4 on the K/Ni/Al2O3 catalyst. The growth of CNFs increased as the temperature was raised from 440 to 500 °C, with the highest rate of formation at 500 °C, consistent with our previous data [8]. A similar temperature dependence was previously reported for CNF formation from methane decomposition [21,22], suggesting that a lower CNF yield from methane at higher temperatures may be due to increased catalyst deactivation. CO formation was apparently enhanced with increasing tem-

Table 1 Effect of H2 and CO2 partial pressures on the formation rate of carbon using a K/Ni/ Al2O3 catalyst. Partial pressure (torr) CO2

H2

H2/CO2

Initial rate of carbon formation (g-carbon/g-cat. h) 7% K + Ni/Al2O3

152 152 152 152 380 380 380 304 228 76

608 456 380 304 228 304 380 380 380 380

4.00 3.00 2.50 2.00 0.60 0.80 1.00 1.25 1.67 5.00

0.27 0.23 0.18 0.15 0.05 0.07 0.27 0.19 0.12 0.10

perature, but CH4 exhibited a low formation rate at 440–520 °C. Fig. 1B plots the rate of carbon formation as a function of residence time, suggesting an optimum carbon growth rate at a flow of 80 mL/min (GHSV = 96 L h1 g1). Fig. 1C and D shows relevant kinetic data for carbon deposition on the K/Ni/Al2O3 catalyst. An Arrhenius plot suggests an apparent activation energy for CNF formation at a value of 71.1 kJ/mol between 440 and 500 °C; the reaction rate law for carbon formation 0:55 at 500 °C was r C ¼ kPCO2 P 0:9 H2 , suggesting that the reaction displayed a near-first-order dependence on H2 pressure and a 0.5 order dependence on CO2 pressure. Fig. 1E compares the formation rates of gaseous products in the CO2 hydrogenation reactions for Ni/Al2O3 and K/Ni/Al2O3 catalysts. All experiments were performed using a stream of H2/CO2 (1:1) at a flow rate of 100 mL/min at 500 °C. Ni/Al2O3 with and without K resulted in near-constant CO production, but the rate of CH4 formation was inhibited by the presence of K additives. Fig. 1F depicts the variations of CO and CH4 formation rate on the K/Ni/Al2O3 catalyst with changes in the partial pressures of H2 and CO2 at 500 °C. The red and blue symbols are the results under constant partial pressures of H2 and CO2, respectively. It is clear that CO formation is seemingly enhanced by increasing H2 and CO2 pressures, but CH4 production was virtually insensitive to the reactant concentrations. 3.2. Effect of synthesis time on CNF growth We used XRD spectroscopy to characterize the effect of the total reaction time on CNF formation in the context of K-enhanced Ni/ Al2O3 catalysts. Our results suggest typical XRD patterns for the (0 0 2) graphitic basal plane at approximately 2h = 26.1° across multiple carbon deposition periods, and increased intensity with longer reaction times, as shown in Fig. 2A. All diffraction angles of the (0 0 2) graphite planes were slightly lower than is typically seen for graphite (2h = 26.5°) [23]. Fig. 2B shows the XRD spectra (30–35°) for Ni/Al2O3, K/Ni/Al2O3 and K/Ni/Al2O3 exposed to H2/CO2 at 500 °C for 10 min. Certain weak peaks near 32.1° and 34.1° were identified in the K/Ni/ Al2O3 catalyst, as compared to spectra (a) and (c). We note that the peaks at 32.1° and 34.1° for the K/Ni/Al2O3 catalyst completely disappeared when a trace amount of carbon was formed by H2/CO2 feeding for 10 min, as shown in spectrum (c). We suggest that the new diffraction peaks at 32.1° and 34.1° were the result of new species generated by the addition of K. To date, we have been unable to confidently identify the peaks at 32.1° and 34.1°, whether by examining the existing literature, or by using the International Center for Diffraction Data (ICDD) library. The disappearance of the diffraction peaks at 32.1° and 34.1° suggests that the new K-related species may participate in the carbon formation process. Table 2 lists the interplanar distances d002, the average sizes of coherently scattering domains (CSD) along the c-axis of carbon (Lc),

C.-S. Chen et al. / Catalysis Communications 11 (2009) 220–224

200 CH4 formation X10

0 0.4 carbon formation 0.3 0.2 0.1 0.0 425

450

475

500

0.7

32.1o

30 h 20 h

0.4 0.3 0.2 0.1

Temperature (oC)

60

-1

5h 3h

CO2

22

24

26

28

d002 (nm)

Lc (nm)

Degree of graphitization (g) (%)

Carbon yields (g-carbon/g-cat.)

-2.2

1 3 5 7 16 20 30 50 70

0.337 0.338 0.339 0.341 0.341 0.341 0.341 0.342 0.342

11.6 8.9 9 8.3 7.5 6.7 7.4 6.1 6.8

81.4 69.8 58.1 34.9 34.9 34.9 34.9 23.3 23.3

0.23 0.81 1.02 2.12 2.50 2.80 3.05 3.54 4.11

4

6

7

(F) 600 CO formation CH4 formation

80

60

40

20

0

Ni/Al2O3 Ni-K/Al2O3

CO formation at PH2=380 torr

Formation Rate (mol/s.g-cat.)

Formation Rate (mol/g-cat. s)

5

Ln(P/Torr)

-1

1/T (K )

100

34

Time (h)

-2.4

120

33

-2.0

0.0014

(E)

32

2θ

-1.6

0

31

Table 2 Effect of time on the characterization of CNF on 7% K + Ni/Al2O3 via analyses of XRD spectra and carbon yield.

-1.4

-2.0

30

Fig. 2. (A) Time-dependent XRD spectra of carbon deposited on 7% K + Ni/Al2O3 catalyst. (B) XRD spectra of Ni/Al2O3 and K + Ni/Al2O3 catalysts after calcination in air and reduction in H2 at 500 °C for 5 h: (a) Ni/Al2O3, (b) 7% K + Ni/Al2O3 and (c) 7% K + Ni/Al2O3 exposed to H2/CO2 at 500 °C for 10 min.

-1.8

-1.8

30

2θ

-1.6

-1.2

(a)

0 min

H2

-1.4 Ea=71.1 kJ/mol

(b)

10 min

20

-1.2

-1

7h

GHSV (L h-1 g-cat.-1)

-0.6

Ln(rate/g-carbon g-cat. h )

16 h

90 120 150 180

(D)

-1.0

(g)

0.0 30

-0.8

o

50 h 0.5

525

(C)

34.05

(c)

70 h

0.6

Intensity (Arb. Unit)

CO formation

Intensity (Arb. Unit)

400

(B)

(A)

(B)

(A) Formation Rate of Carbon (g carbon/g-cat. h)

Formation Rate (g /h. g-cat. ) Formation Rate (mol/s.g-cat)

222

500

CH4 formation at PH2=380 torr CO formation at PCO2=380 torr CH4 formation at PCO2=380 torr

400

300

200

100

0 100

200

300

400

Pressure (torr)

Fig. 1. (A) Comparison of the catalytic activity for CNF, CO and CH4 formation from the hydrogenation of CO2 on 7% K + Ni/Al2O3. (B) Effect of GHSV on formation rates of carbon. (C) Arrhenius plots for the rate of carbon formation on 7% K + Ni/Al2O3 at 440–500 °C. (D) Dependence of the formation rate of carbon on the partial pressures of H2 and CO2. (E) Comparing the formation rates of CO and CH4 on Ni/Al2O3 and 7% K + Ni/Al2O3. (F) Dependence of the formation rates of CO and CH4 on the partial pressures of H2 and CO2.

graphitization percentages (g) and yields of solid carbon with various reaction times. The carbon yield at 500 °C increased with time over the course of the CO2 hydrogenation reaction within the first 70 h. The interplanar distance d002 was determined from the position of reflection at the (0 0 2) plane. The value of Lc was calculated from the full width at the half-maximum of the diffraction peak of k , where k the (0 0 2) face (b) using the Sherrer equation, Lc ¼ b cos h and h are the wavelengths of the X-rays and the diffraction angle, respectively [24]. The d002 of carbon deposition in our study was greater than that of typical graphite with a characteristic distance of 0.335 nm. The carbon graphitization degree (g) was calculated according to the formula d002(Å) = 3.354(Å) + 0.086(1  g) [21], where g is the graphitization percentage in the carbon samples. We determined that solid carbon with high graphitization may initially form over shorter synthesis periods. However, the degree of graphitization was observed to decrease with increasing reaction times. 3.3. Characterization of CNF The thermal stability of the functional groups for the CNF sample synthesized from H2/CO2 at 500 °C for 10 min was analyzed using TPD in H2 and He streams, as shown in Fig. 3A and B. In

223

C.-S. Chen et al. / Catalysis Communications 11 (2009) 220–224

(B)

o

Mass Intensity (Arb. Unit)

Mass Intensity (Arb. Unit)

(C)

o

500 C

630 C

sp3 C-C

CO

sp2 C=C

Intensity (Arb. Unit)

(A)

432oC

104oC

C-O

5h

CO2

10min

CH4

200 400 600 800 1000 o

Temperature ( C)

288

150 300 450 600 750 o

285

282

Binding Energy (eV)

Temperature ( C)

Fig. 3. (A) TPD spectrum of CH4 desorbed from surface carbon in a H2 stream. (B) TPD spectra of CO2 and CO desorbed from surface carbon in a He stream. (C) XPS spectra of carbon deposited at 500 °C for 10 min and 5 h.

the H2 stream, carbon was hydrogenated with the result that CH4 desorbed from the surface carbon to a maximum extent at 630 °C (Fig. 3A). CO2 and CO were also extracted from the CNF samples (Fig. 3B) by means of the TPD process with He as a carrier gas. Desorptions of both CO2 and CO are related to the destruction of oxygen-containing functional groups on the CNF surface; as such, the desorption maxima of CO2 (at 104 and 432 °C) and CO (at 500 °C) can be ascribed to carboxylic acid functional groups [25]. However, no detectable XRD peak for the trace amounts of carbon deposition could be identified with a reaction time of 10 min, as shown in Fig. 2. Thus, XPS C1s spectroscopy was used to evaluate the time-dependent carbon structures, as shown in Fig. 3C. The XPS spectrum in Fig. 3C, which was obtained using a longer CNF synthesis time of 5 h, can be fitted into three overlapping peaks. The peaks at 284.6, 285.3 and 286.7 eV can be subsequently assigned to sp2 C@C, sp3 CAC and CAO, respectively [26,27]. We note that the XPS C1s peak for trace amounts of carbon deposition after 10 min synthesis time appeared primarily at 286.7 eV and was linked with the CAO functional group. Related sp2 C@C and sp3 CAC diagnostic peaks were not evident over the course of the initial carbon formation process. Initial carbon deposition might preferentially create amorphous carbon structures with an oxygenbonded configuration. 4. Discussion Based on the kinetic results, an Eley–Rideal model is used to interpret the mechanism for the overall reaction: 2H2 + CO2 ? C + 2H2O. The proposed sequence of elementary steps shows as followed: (1) CO2 þ 2S CO2  2S (2) H2 þ CO2  2S H2 O þ CO  S þ S (3) H2 + CO  S ? H2O + C  S S refers to a surface site. Based on the mechanisms above, step (3) is assumed to be the rate-determining step. The equation, r ¼ k3 P H2 hCO , can be attributed to the rate of carbon formation. Use a steady-state approximation to obtain hCO :

dhCO ¼ k2 PH2 hCO2  k2 PH2 O hCO hs  k3 PH2 hCO ¼ 0 dt k2 PH2 hCO2 hCO ¼ k2 PH2 O hs þ k3 PH2

Step (1) is assumed to be in quasi-equilibrium. It can obtain the hs  h 12 K 1 P CO CO 2 and hCO2 as followed: hs ¼ K 1 PCO2 and hCO2 ¼ 1þ2K 1 PCO , thus



k2 PH2 0 K

hCO ¼

k2 PH2 O @

) hCO ¼

2



2

¼

A þ k3 PH 2 2

2O

2O

1

1

1

1

1

k2 P H

2O

2



K 1 P CO

2

1þ2K 1 PCO

2

1 þ k3 P H2

ð1þ2K 1 PCO2 Þ2

2 1

ð1þ2K 1 PCO2 Þ2

1þ2K 1 P CO Þ2 2ð 2

k2 PH2





K 1 P CO

1þ2K 1 PCO

PH

hCO

112

k2 P H 2

PH

1þ2K 1 P CO

K 1 P CO

k3 P H

when

2

1 PCO2 1þ2K 1 PCO 2

k2 k3

2



K 1 P CO

þ1

 1 and K 1 PCO2  1

K 21 K 2 P H2 P2CO2 ¼ pffiffiffi 2P H2 O

k3 K 21 K 2 P2H2 P2CO2 pffiffiffi r¼ 2P H 2 O The reaction order of H2 may be adjusted with partial pressure of CO2. This article given the reaction order of H2 with 0.9 is obtained under the constant partial pressure of the CO2 at 152 torr. As the constant pressure of CO2 increasing to 380 torr, the order of H2 can increase to be ca. 3. Our experimental results demonstrate that pure Ni and K surfaces do not induce carbon deposition over the course of the CO2 hydrogenation reaction [4,8]. However, the CNF yield increases with increasing K content in the case of Ni/Al2O3 exposed to a feed of H2/CO2 [8]. This suggests that carbon formation should correlate with the formation of new active phases as K is added to the system and as it participates in the initial carbon deposition process (Fig. 2B). During the reaction on the Ni/Al2O3 and K/Ni/Al2O3 catalyst, CO and CH4 were formed as major gas products, while no detectable carbon was formed in the CH4 stream using Ni/Al2O3 and K/Ni/Al2O3 catalysts at 500 °C. It is suggested that differently dual sites on K/Ni/Al2O3 catalyst may form to participate in the CO and CH4 formation and carbon deposition, respectively. It could reasonably interpret why no observable carbon deposition was identified on the Ni/Al2O3 catalyst during CO2 hydrogenation. The carbon formation on the special active phase could be proposed to depend on the reaction pathway: CO2 ? CO ? CNF.

224

C.-S. Chen et al. / Catalysis Communications 11 (2009) 220–224

In our experience, pure CO2 failed to deposit carbon on the K/Ni/ Al2O3 catalyst. Therefore, the observed positive effect of H2 concentration on the carbon deposition rate might involve the formation of a reaction intermediate that leads to accelerated CNF generation. Based on previous literature reports, a key intermediate, namely the formate species resulting from the association of H atoms and CO2, may be created on Ni catalysts [28]. This conclusion is similar to our early study with respect to intermediate of RWGS reaction on Cu surface. So far, we can conclude that the carbon source is deduced to be obtained from the CO2 reactant and new active phases generated from K adding from present data, but no further evidence can confirm an intermediate depending on carbon formation, even if a formate species has been found on Ni surface. Increasing the partial pressures of H2 and CO2 can lead to enhanced CNF growth on the K/Ni/Al2O3 catalyst, especially at high H2 pressures. In general, high H2 concentrations inhibit the growth of surface carbon for other carbon sources [22,29], and increasing H2 leads to elevated gasification rates of surface carbon that subsequently hinder the catalytic action of the metal surface and the chemisorbed species. In the present study, we failed to observe H2 gasification of CNF synthesized from H2/CO2. Fig. 3A reveals that the TPD of CNF in H2 leads to maximum CH4 desorption at 630 °C, while surface carbon was difficult to convert to CH4 using H2 at 500 °C. Our XPS and XRD spectroscopy data offer insight into CNF growth over time. First, an initial growth stage takes place during which surface carbon forms amorphous carbon containing CAO groups that lack sp2 C@C and sp3 CAC structures. Second, a high degree of CNF graphitization may be generated at the start of graphene layer formation (carbon deposition for 1 h). Finally, the a more disordered sp3 carbon structure becomes apparent prior to the formation of a more ordered sp2 structure over a longer period of time. 5. Conclusions

ture growth over time. We suggest that carbon formation may depend on the reaction pathway: CO2 ? CO ? CNF. Acknowledgment Financial support from the National Science Council of the Republic of China (NSC 96-2113-M-182-002-MY2) is gratefully acknowledged. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24]

We have demonstrated that CNF formation depends strongly on temperature, H2 and CO2 partial pressures and on reactant residence time during the CO2 hydrogenation reaction. Certain K-related active phases obtained by calcination and reduction pretreatments may participate in the carbon deposition process. Furthermore, we obtained kinetic data for CNF synthesis in the context of H2/CO2, suggesting an apparent activation energy of 0:55 71.1 kJ/mol and a reaction rate law of rC ¼ kPCO2 P 0:9 H2 . XRD and XPS spectroscopic analyses indicated a change in CNF carbon struc-

[25] [26] [27] [28]

[29]

X. Xiaoding, J.A. Moulijn, Energy Fuel 10 (1996) 305. C.S. Chen, J.H. Lin, J.H. You, C.R. Chen, J. Am. Chem. Soc. 128 (2006) 15950. C.S. Chen, W.H. Cheng, S.S. Lin, Appl. Catal. A 257 (2004) 97. C.S. Chen, W.H. Cheng, S.S. Lin, Appl. Catal. A 238 (2003) 55. C.S. Chen, W.H. Cheng, Catal. Lett. 83 (2002) 126. C.S. Chen, W.H. Cheng, S.S. Lin, Chem. Commun. (2001) 1770. C.S. Chen, W.H. Cheng, S.S. Lin, Catal. Lett. 68 (2000) 45. C.S. Chen, J.H. Lin, A.S. Lin, H.T. Huang, H.L. Ma, Carbon 46 (2008) 369. H. Tang, J.H. Chen, Z.P. Huang, D.Z. Wang, Z.F. Ren, L.H. Nie, Carbon 42 (2004) 191. A. Chambers, C. Park, R.T.K. Baker, N.M. Rodriguez, J. Phys. Chem. B 102 (1998) 4253. G.G. Tibbetts, G.P. Meisner, C.H. Olk, Carbon 39 (2001) 2291. C.W. Huang, H.C. Wu, Y.Y. Li, Sep. Purif. Technol. 58 (2007) 219. J. Kong, N.R. Franklin, C.W. Zhou, M.G. Chapline, S. Peng, K.J. Cho, Science 287 (2000) 622. E. Frackowiak, F. Beguin, Carbon 39 (2001) 937. W.A. Deheer, A. Chatelain, D. Ugarte, Science 270 (1995) 1179. S.S. Fan, M.G. Chapline, N.R. Franklin, T.W. Tombler, A.M. Cassell, H.J. Dai, Science 283 (1999) 512. Y.Y. Li, K. Mochidzuki, A. Sakoda, M. Suzuki, Carbon 39 (2001) 2143. R. Ryoo, S.H. Joo, M. Kruk, M. Jaroniec, Adv. Mater. 13 (2001) 677. B. Frank, F.C. Jentoft, H. Soerijanto, J. Kröhnert, R. Schlögl, R. Schomäcker, J. Catal. 246 (2007) 177. L.C. Wang, Y.M. Liu, M. Chen, Y. Cao, H.Y. He, G.S. Wu, W.L. Dai, K.N. Fan, J. Catal. 246 (2007) 193. M.A. Ermakova, D.Y. Ermakov, A.L. Chuvilin, G.G. Kuvshinov, J. Catal. 201 (2001) 183. I. Kvande, D. Chen, Z. Yu, M. Rønning, A. Holmen, J. Catal. 256 (2008) 204. T.I.T. Okpalugo, P. Papakonstantinou, H. Murphy, J. Mclaughlin, N.M.D. Brown, Carbon 43 (2005) 2951. T.V. Reshetenko, L.B. Avdeeva, Z.R. Ismagilov, V.V. Pushkarev, S.V. Cherepanova, A.L. Chuvilin, V.A. Likholobov, Carbon 41 (2003) 1605. J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Órfão, Carbon 37 (1999) 1379. J.H. Zhou, Z.J. Sui, J. Zhu, P. Li, D. Chen, Y.C. Dai, W.K. Yuan, Carbon 45 (2007) 785. S.Y. Brichka, G.P. Prikhod’ko, Y.I. Sementsov, A.V. Brichka, G.I. Dovbeshko, O.P. Paschuk, Carbon 42 (2004) 2581. E. Vesselli, L. De Rogatis, X. Ding, A. Baraldi, L. Savio, L. Vattuone, M. Rocca, P. Fornasiero, M. Peressi, A. Baldereschi, R. Rosei, G. Comelli, J. Am. Chem. Soc. 130 (2008) 11417. N. Jeong, J. Lee, J. Catal. 260 (2008) 217.