Adsorption and Activation of O2 on Nitrogen-Doped Carbon

May 10, 2010 - Nanoscale Research Letters 2017 12 (1), .... Carbon 2016 98, 115-125 ..... Nitrogen-doped (6,0) carbon nanotubes: A comparative DFT stu...
0 downloads 0 Views 5MB Size
J. Phys. Chem. C 2010, 114, 9603–9607

9603

Adsorption and Activation of O2 on Nitrogen-Doped Carbon Nanotubes Xingbang Hu,*,† Youting Wu,† Haoran Li,*,‡ and Zhibing Zhang† School of Chemistry and Chemical Engineering, Nanjing UniVersity, Nanjing 210093, and Department of Chemistry, Zhejiang UniVersity, Hangzhou 310027, P. R. China ReceiVed: January 1, 2010; ReVised Manuscript ReceiVed: March 21, 2010

The adsorption and activation of triplet O2 on the surface of nitrogen-doped carbon nanotube (NCNT) with different diameter and length were investigated. It was found that, rather than the unfavorable adsorption on normal carbon nanotube (CNT), the adsorption of O2 on NCNT was obviously exothermic and the electron transition of O2 happened in the adsorption process. The oxygen adsorbed on NCNT showed an interesting electron configuration which was similar to the active oxygen anion. The spin density, charge, and bond length of O2 changed with the size of NCNT. In combination with the recent results reported by Dai et al. (Science, 2009, 323, 760), it is reasonable to believe that these NCNTs should be a potential metal-free catalyst. The results presented here should be useful for designing and developing effective catalyst based on NCNT. 1. Introduction

2. Computational Methods

Reducing or replacing the expensive platinum-based material is one of the most important steps for the large-scale practical application of fuel cells and related catalytic process.1 Some important efforts have been made to develop new electrocatalysts for oxygen reduction reactions, such as Pt-based alloys,2,3 platinum nanoparticles,4 transition metal chalcogenides,5 and carbon nanotube-supported metal particles.6,7 These pioneering works have reduced the amount of platinum used. However, the using of noble metal was still unavoidable.2,6,7 Quite recently, Dai et al. found that the nitrogen-doped carbon nanotube (NCNT) could be used as a metal-free electrode with a much better electrocatalytic activity,1 which may open a way to prepare a new class of efficient metal-free oxygen reduction reaction catalysts. It is crucial to understand the oxygen adsorption and activation processes on the surface of the NCNT by theoretic investigation, because of a widely known fact that there is only weak physisorption between triplet oxygen and the normal carbon nanotube (CNT), and there is little electron transfer from the carbon nanotube to the oxygen by theoretic calculations8-14 and experiment.12 Previously, in order to strengthen the interaction between oxygen and CNT, singlet oxygen must be used in theoretic calculations.14-21 As we know, the ground state of oxygen is triplet (3O2, the value of the left superscript means spin multiplicity). Singlet oxygen (1O2) is the lowest excited state of molecular oxygen. The condition to generate 1O2 is rigorous. Besides this, the lifetime of 1O2 is quite short.22 Hence, developing nanotubes which can adsorb and activate 3O2 is significant. To understand the structure-property relationship of the NCNT and shed some light on why NCNT shows good electrocatalytic activity, we did a theoretical investigation on the adsorption and activation processes of oxygen on a series of NCNTs.

Th density functional theory (DFT) based B3LYP method was used in this investigation. This method has been widely used for the calculation of nanomaterials and shown excellent agreement with the experimental data.23-30 Adopting the 6-31G* basis set is vital to generating satisfying results for the calculation of nanotubes.26-30 Hence, for this work, all the structures were fully optimized at the B3LYP/6-31G* level of theory. Models used in this work are armchair nanotubes terminated with C-H bonds, and they are abbreviated as (N, N)-L. Here, (N, N) means the nanotube type, and L is the length of the nanotube (N ) 3, 4, 5, and 6; L ) 8.0, 10.5, and 12.9). All the calculations were performed with the Gaussian 03 program.31

* To whom correspondence should be addressed. E-mail: [email protected] (X. B. H.); [email protected] (H. R. L.). † Nanjing University. ‡ Zhejiang University.

3. Results and Discussion The adsorbing potential energy curves of CNT/NCNT · · · O2 are presented in Figure 1. The potential energy curve has been widely used to study the adsorbing process of oxygen8-21,32 and it can give us a better understanding of this process. The potential energy scan was performed at the B3LYP/6-31G* level of theory. The geometric structure of the nanotube was fixed and the distance between oxygen and nanotube (R) was set as a variable in the calculation. The initial distance investigated was 5 Å and step size was 0.1 Å (when the distance between O2 and CNT/NCNT is larger than 5 Å, there is almost no interaction). As we can see, the adsorption of O2 on the surface of NCNT is quite different from that on CNT. The adsorption of 3O2 on CNT is energy-unfavorable, whereas the adsorption of 1O2 on CNT is a little exothermic. These results agree with the previous research on the interaction between O2 and CNT.8-21 However, as shown in Figure 1b, the energy of the 1O2 system at each point of the potential energy curve is much higher than that of the initial state of the 3O2 system (R ) 5.0 Å), which makes the adsorption of 1O2 quite difficult under normal conditions. A spin transition of O2 happens on the surface of NCNT, which greatly favors the adsorption process of 3O2. Though the

10.1021/jp1000013  2010 American Chemical Society Published on Web 05/10/2010

9604

J. Phys. Chem. C, Vol. 114, No. 21, 2010

Hu et al.

Figure 1. The adsorbing potential energy curves of (5,5)-12.9 NCNT/CNT · · · O2. Red: oxygen. Blue: nitrogen. Gray: carbon. White: hydrogen.

Figure 2. Optimized geometric configuration of (5,5)-12.9 NCNT/CNT · · · O2 systems. Red: oxygen. Blue: nitrogen. Gray: carbon. White: hydrogen. Bond lengths are in Å. Italic values in parentheses mean binding energy in kJ/mol. E(Binding energy) ) E(NCNT/CNT · · · O2 - E(NCNT/CNT) E(triplet O2).

energy of the 3O2 system increases as the distance R is shortened, the triplet curve crosses the singlet one by overcoming a small barrier (about 61.2 kJ/mol) when the distance is 2.1 Å. On the basis of the two spin state reactivity theory,33 it is reasonable to predict that a spin transition of O2 happens and the 3O2 will be adsorbed along the curve of the singlet system when R < 2.1 Å. As a result, the adsorption of 3O2 is exothermic. A similar spin transition of O2 on the surface of normal CNT is impossible, and the 3O2 system retains its spin state when it interacts with the nanotube.11,17 A series of structures of both NCNT · · · O2 and CNT · · · O2 systems were fully optimized (Figure 2). The optimized results also suggested that the absorption of O2 on the surface of CNT was energy-unfavorable. In the most stable CNT · · · O2 system (CNT1), the distance between O2 and the CNT is about 3.4 Å, which agrees with the previous results.10

Previous research has concluded that the interaction between 3 O2 and CNT is a weak physisorption and there is no charge transfer between the O2 and nanotubes.8-14 On the contrary, the absorption of 3O2 on the surface of NCNT is energyfavorable. The C-O distances range from 1.5 to 3.5 Å in these exothermic interactions. The O-O bond is distinctly lengthened when the C-O distance is smaller than 1.6 Å. This indicates a complicated electron rearrangement happening in the absorption process and this absorption should belong to chemisorption. It is interesting to investigate the electric structure of the O2 in the most stable structures (NCNT1 and CNT1 in Figure 2). The natural electron configuration of isolated O2 in different states and O2 in CNT1/NCNT1 were presented in Table 1. The most interesting result is that the electron configuration of O2 in NCNT system is quite different from those of both 3O2 and

Adsorption/Activation of O2 on NCNTs

J. Phys. Chem. C, Vol. 114, No. 21, 2010 9605

TABLE 1: Natural Electron Configuration of Isolated O2 and O2 on the Surface of NCNT1 and CNT1a R spin 1

O2 O2 1 O222 O2ONCNT b,c ONCNT b,d OCNT b,c OCNT b,d 3

0.90

2.08

0.01

2S 2P 3S 2S0.912P2.583S0.013P0.01 2S0.962P2.53 2S0.932P2.56 2S0.862P2.37 2S0.932P2.50 2S0.912p2.583S0.013p0.01 2S0.912p2.583S0.013p0.01

β spin 0.90

2.08

TABLE 2: Spin Density and Charge of O2 on the Surface of NCNTa

spin density 0.01

2S 2P 3S 2S0.902P1.593S0.01 2S0.962P2.53 2S0.922P2.07 2S0.862P2.06 2S0.902P1.81 2S0.902p1.593S0.01 2S0.902p1.593S0.01

0.00 1.00 0.00 0.50 0.31 0.69 1.00 1.00

a

Calculation with 6-31G* basis set; because of the symmetry, only one oxygen of 1O2, 3O2, 1O22-, and 2O2- was presented. b Using (5,5)-12.9 NCNT/CNT. c The oxygen close to the nanotube. d The oxygen away from the nanotube.

O2 but similar to that of 2O2-. As we know, 2O2- is an oxygen molecule with one more electron and is quite reactive.22 A compound which can transfer 3O2 to 2O2- should be a potentially powerful catalyst.32,34-36 NCNT has shown wonderful electrocatalytic activity.1 On the basis of previous electrochemistry experiments and theoretic results obtained here, it is reasonable to believe that NCNT could also be used as an efficient metalfree catalyst. As a contrastive example, the electric structures of O2 on the surface of CNT are almost the same as that of the normal 3O2. More NCNTs with different diameters and lengths were investigated to confirm the above results (Figure 3). For all of the models used here, no matter what diameter or length of the NCNT is, there is a mass of electron transferring from the NCNT to the O2 (Table 2). It can make the electron configuration of O2 change toward more a reactive one. It has been known that the finite-length nanotubes show an oscillating and downward HOMO-LUMO gap when the length of the nanotube increases, and the one with the smallest HOMO-LUMO gap may best represent the metallic character of the carbon nanotube.37,38 For the NCNT, it can be found that the HOMO-LUMO gap decreases with increase of the diameter. Furthermore, the HOMO-LUMO gap also shows an oscillating character as the length of the nanotube increases. NCNT (6, 6)-10.5 possesses the smallest HOMO-LUMO gap in the current study. The structure

spin density size of NCNT gap (3, 3)-8.0 (3, 3)-10.5 (4, 4)-8.0 (4, 4)-10.5 (5, 5)-8.0 (5, 5)-10.5 (5, 5)-12.9 (6, 6)-8.0 (6, 6)-10.5

2.02 1.97 1.70 1.32 1.54 1.09 1.21 1.39 0.92

b

c

charge b

OC

OA

OC

OAc

BEd (kJ/mol)

0.292 0.295 0.298 0.304 0.341 0.313 0.310 0.377 0.324

0.698 0.696 0.692 0.691 0.666 0.686 0.690 0.649 0.681

-0.207 -0.193 -0.200 -0.187 -0.196 -0.185 -0.179 -0.199 -0.184

-0.163 -0.161 -0.168 -0.165 -0.189 -0.168 -0.165 -0.200 -0.171

-71.4 -36.0 -8.1 -18.8 70.0 -6.8 -3.2 68.4 13.5

a Calculation with 6-31G* basis set. b The oxygen close to the NCNT. c The oxygen away from the NCNT. d Binding energy.

1

parameters obviously change with the diameters of NCNT. With NCNT (N, N)-8.0 (N ) 3, 4, 5, 6) as example, the following values decrease as the diameter increases: bond length of the adsorbed oxygen, spin density, and charge carried by OA. At the same time, the following values increase as the diameter increases: the distance between the oxygen and NCNT, spin density, and charge carried by OC. Similarly, the same trend can be found for NCNT (N, N)-10.5 (N ) 3, 4, 5, 6). On the basis of the analysis of NCNT (N, N)-8.0 and NCNT (N, N)-10.5 (N ) 3, 4, 5, 6), it seems that the binding energy decreases as the NCNT diameter increases, except that NCNT (5, 5)-8.0 has almost the same binding energy value as NCNT (6, 6)-8.0. At the same time, the interaction may be influenced by the periodic character of finite nanotubes with different lengths. The NCNT (5, 5)-L (L ) 8.0, 10.5, 12.9) shows an oscillation O2 binding energy as the length increases. The origin of the difference between CNT and NCNT is the nitrogen doping (Figure 4). First, the nitrogen doping lifts the energy level of the highest-occupied molecular orbitals (HOMO) by 0.75 eV and reduces the gap between the HOMO and the lowest-unoccupied molecular orbitals (LUMO) by 0.49 eV, which will make it easier for the electron to transfer from the nanotube to the oxygen adsorbed.

Figure 3. Optimized geometric configuration of different NCNT/CNT · · · O2 systems. Red: oxygen. Blue: nitrogen. Gray: carbon. White: hydrogen.

9606

J. Phys. Chem. C, Vol. 114, No. 21, 2010

Hu et al.

Figure 4. (a) LUMO/HOMO and (b) charge distribution of (5,5)-12.9 CNT and NCNT.

On the other hand, the carbon on the normal CNT only carries neglectable positive charge. However, when there is a nitrogen atom doped on the nanotube, it induces a notable electron transfer. The nitrogen atom on the NCNT shows remarkable negative charge and the carbon atoms around the nitrogen show obvious positive charge. That is, the nitrogen doping introduces electron-deficient carbon and facilitates the binding between NCNT and O2. Besides, it has been known that nitrogen doping in silicon or carbon nanotubes can generate delocalized electrons and make the material metallic.38,39 These may be the cause of oxygen activation. 4. Conclusions In summary, the adsorption and activation of O2 on the surface of NCNT were investigated with the comparison of the same processes on CNT. The obtained results indicated that nitrogendoping facilitated the adsorption of O2 on carbon nanotubes. Electronic transition of O2 accompanies with the adsorption process on NCNT. The electron configuration of O2 on NCNT is similar to that of active 2O2-. Changing the diameter and length of NCNT can influence the bond length of O2 and the binding energy between O2 and NCNT, but the O2 on different NCNTs remains reactive. NCNT has shown good electrocatalytic activity.1 With the characteristics of NCNT obtained here and previously, it is reasonable to predict that NCNT should be an efficient metalfree catalyst for oxidation reaction. Designing and developing a catalyst based on NCNT is a challenge and requires a great effort from both theory and experiment. Though there have been many effective methods for growing different nitrogendoped carbon nanotubes, the theoretical research is deficient.1,40,41 We are expecting that the results of our work presented here

will provide a clue for the future development of a metalfree catalyst based on NCNT. Acknowledgment. This work was supported by the National Natural Science Foundation of China (no. 20773109 and no. 20803062). References and Notes (1) Gong, K. P.; Du, F.; Xia, Z. H.; Durstock, M.; Dai, L. M. Science 2009, 323, 760. (2) Zhang, J.; Saski, K.; Sutter, E.; Adzic, R. R. Science 2007, 315, 220. (3) Shao, M. H.; Sasaki, K.; Adzic, R. R. J. Am. Chem. Soc. 2006, 128, 3526. (4) Wang, C.; Daimon, H.; Lee, Y.; Kim, J.; Sun, S. J. Am. Chem. Soc. 2007, 129, 6974. (5) Gong, K. P.; Yu, P.; Su, L.; Xiong, S.; Mao, L. J. Phys. Chem. C 2007, 111, 1882. (6) Che, G.; Lakshmi, B. B.; Fisher, E. R.; Martin, C. R. Nature 1998, 393, 346. (7) Kongkanand, A.; Kuwabata, S.; Girishkumar, G.; Kamat, P. Langmuir 2006, 22, 2392. (8) Mowbray, D. J.; Morgan, C.; Thygesen, K. S. Phys. ReV. B 2009, 79, 195431. (9) Ulbricht, H.; Moos, G.; Hertel, T. Phys. ReV. B 2002, 66, 075404. (10) Sorescu, D. C.; Jordan, K. D.; Avouris, P. J. Phys. Chem. B 2001, 105, 11227. (11) Giannozzi, P.; Car, R.; Scoles, G. J. Chem. Phys. 2003, 118, 1003. (12) Tchernatinsky, A.; Desai, S.; Sumanasekera, G. U.; Jayanthi, C. S.; Wu, S. Y.; Nagabhirava, B.; Alphenaar, B. J. Appl. Phys. 2006, 99, 034306. (13) Lee, K.; Sinnott, S. B. Nano Lett. 2005, 5, 793. (14) Ricca, A., Jr.; Bauschlicher, C. W. Phys. ReV. B 2003, 68, 035433. (15) Akdim, B.; Duan, X.; Pachter, R. Nano Lett. 2003, 9, 1209. (16) Chan, S. P.; Chen, G.; Gong, X. G.; Liu, Z. F. Phys. ReV. Lett. 2003, 90, 086403. (17) Jhi, S. H.; Louie, S. G.; Cohen, M. L. Phys. ReV. Lett. 2000, 85, 1710. (18) Zhang, Y. F.; Liu, Z. F. Carbon 2006, 44, 928. (19) Zhang, Y. F.; Liu, Z. F. J. Phys. Chem. B 2004, 108, 11435.

Adsorption/Activation of O2 on NCNTs (20) Liu, H. J.; Chan, C. T.; Liu, Z. Y.; Shi, J. Phys. ReV. B 2005, 72, 075437. (21) Liu, H. J.; Chan, C. T. Phys. ReV. B 2006, 73, 113405. (22) Schweitzer, C.; Schmidt, R. Chem. ReV. 2003, 103, 1685. (23) Kutana, A.; Giapis, K. P. J. Phys. Chem. C 2009, 113, 14721. (24) Wang, L. Y.; Roudgar, A.; Eikerling, M. J. Phys. Chem. C 2009, 113, 17989. (25) Zhou, C. G.; Li, J. Y.; Chen, S.; Wu, J. P.; Heier, K. R.; Cheng, H. S. J. Phys. Chem. C 2008, 112, 14015. (26) Gao, X. F.; Wang, L.; Ohtsuka, Y.; Jiang, D.; Zhao, Y.; Nagase, S.; Chen, Z. F. J. Am. Chem. Soc. 2009, 131, 9663. (27) Lu, X.; Chen, Z. F.; Schleyer, P. V. R. J. Am. Chem. Soc. 2005, 127, 20. (28) Nikawa, H.; Yamada, T.; Cao, B.; Mizorogi, N.; Slanina, Z.; Tsuchiya, T.; Akasaka, T.; Yoza, K.; Nagase, S. J. Am. Chem. Soc. 2009, 131, 10950. (29) Chen, Z. F.; Nagase, S.; Hirsch, A.; Haddon, R. C.; Thiel, W.; Schleyer, P. V. R. Angew. Chem., Int. Ed. 2004, 43, 1552. (30) Barone, V.; Peralta, J. E.; Wert, M.; Heyd, J.; Scuseria, G. E. Nano Lett. 2005, 5, 1621. (31) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;

J. Phys. Chem. C, Vol. 114, No. 21, 2010 9607 Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, reVision B.01; Gaussian, Inc.: Pittsburgh, PA, 2003. (32) Hu, X. B.; Li, H. R.; Wang, C.; Han, S. J. Chem. Phys. Lett. 2006, 426, 39. (33) Schro¨der, D.; Shaik, S.; Schwarz, H. Acc. Chem. Res. 2000, 33, 139. (34) Meunier, B.; de Visser, S. P.; Shaik, S. Chem. ReV. 2004, 104, 3947. (35) Hu, X. B.; Li, H. R.; Wang, C. M. J. Phys. Chem. B 2006, 110, 14046. (36) Hu, X. B.; Li, H. R. J. Phys. Chem. A 2007, 111, 8352. (37) Rochefort, A.; Salahub, D. R.; Avouris, P. J. Phys. Chem. B 1999, 103, 641. (38) Teo, B. K.; Huang, S. P.; Zhang, R. Q.; Li, W. K. Coord. Chem. ReV. 2009, 253, 2935. (39) Li, Y. F.; Zhou, Z.; Wang, L. B. J. Chem. Phys. 2008, 129, 104703. (40) Lee, D. H.; Lee, W. J.; Kim, S. O. Nano Lett. 2009, 9, 1427. (41) Koos, A. A.; Dowling, M.; Jurkschat, K.; Crossley, A.; Grobert, N. Carbon 2009, 47, 30.

JP1000013