Reactivity of Carboxylic Groups on Armchair and Zigzag Carbon

A theoretical study of gas-phase esterification of monocarboxy fragments of armchair (10,10) and zigzag (16,0) single-walled carbon nanotubes. (SWNTs)...
0 downloads 0 Views 267KB Size
Download PDF
NANO LETTERS

Reactivity of Carboxylic Groups on Armchair and Zigzag Carbon Nanotube Tips: A Theoretical Study of Esterification with Methanol

2002 Vol. 2, No. 8 835-839

Vladimir A. Basiuk* Department of Chemistry and Biotechnology, Faculty of Engineering, Yokohama National UniVersity, Hodogaya-ku, Yokohama 240-8501, Japan Received May 12, 2002; Revised Manuscript Received June 6, 2002

ABSTRACT A theoretical study of gas-phase esterification of monocarboxy fragments of armchair (10,10) and zigzag (16,0) single-walled carbon nanotubes (SWNTs) with methanol was performed by using a two-level ONIOM approach. The results suggest that the reaction is favorable on armchair SWNT tips, contrary to that on the zigzag nanotube tips. Gas-phase esterification might be a route to selective derivatization of different forms of carbon nanotubes.

The finest device applications of carbon nanotubes (CNTs) deal with single nanotubes and depend strongly on their electronic properties, defined first of all by structural features such as nanotube helicity (or chirality). Although it is technically possible to observe and pick up single armchair, zigzag, and intermediate-chirality nanotubes, further development of CNT-based devices demands for more efficient sources. Unfortunately, so far the orientation of carbon hexagons is not an easily controllable parameter of CNT growth. An alternative route to appreciable amounts of different forms of carbon nanotubes might rely upon the development of techniques for their separation and purification from the mixtures produced by conventional methods. One of the most promising approaches to purifying CNTs and expanding their application areas is chemical derivatization.1-20 In particular, reactions of the terminal carboxylic groups (obtained by oxidation of as-prepared capped nanotubes) with amines help to increase the solubility of singlewalled CNTs (SWNTs)1,4,9,14 and to enhance their affinity for metal surfaces.11 Both properties can be useful for the separation of different CNT forms, provided that there is a way to selectively derivatize them. Here experimental data on how the chirality influences reactivity of functional groups on CNT tips would be highly desirable. This, however, requires further sophistication of the methods of handling,

analyzing, and observing individual CNTs. In the meantime, theoretical studies remain the only way to give an insight. Among all the existing methods to study thermodynamics and mechanisms of chemical reactions, not too many of them are suitable for this goal, due to the large size of CNT molecules. Fortunately, a growing need for theoretical approaches capable of treating large (including biochemical) molecular systems gave rise to the development of combined quantum mechanical/molecular mechanical (QM/MM) models, in particular the two- or three-level ONIOM method.21,22 The latter consists of dividing a large molecular system into two or three levels: a relatively small section, essential for a property of interest, is treated at a higher theoretical level (true ab initio or DFT), whereas the remaining one or two levels serve mostly to constrain the general geometry and are described by a computationally less expensive method (molecular mechanics or semiempirical).21,22 Recent examples of using the ONIOM approach to study CNT-based reaction systems are the addition of hydrogen and fluorine atoms to zigzag and armchair SWNT side walls,23-26 and our papers19,20 on the direct amidation of monocarboxy-substituted tips of zigzag (10,0) and armchair (5,5) SWNTs with methylamine, according to the following reaction scheme:

* Permanent address for correspondence: Instituto de Ciencias Nucleares, Universidad Nacional Auto´noma de Me´xico, Circuito Exterior C. U., 04510 Me´xico D. F., Mexico; e-mail: [email protected].

Our results suggested that the direct formation of amides on armchair SWNT tips is much more energetically preferable

10.1021/nl025607n CCC: $22.00 Published on Web 06/26/2002

© 2002 American Chemical Society

SWNT-COOH + H2NCH3 f SWNT-CO-NHCH3 + H2O (1)

Figure 1. Monocarboxy-substituted fragments of armchair (10,10) (a) and zigzag (16,0) (b) SWNTs used for two-level ONIOM calculations of the model gas-phase esterification reaction with methanol. The highlighted (dark) atoms and those belonging to methanol molecule were treated at the B3LYP/6-31G(d) level of theory; the remaining SWNT atoms were treated with UFF molecular mechanics.

than that on the zigzag nanotube tips, which might open, in principle, a new route to selective derivatization of different forms of CNTs. However, before making any generalizations, it is necessary to investigate more systematically this type of reaction by varying nanotube size and, what is especially important, derivatizing reagents. A good candidate for further studies is esterification with aliphatic alcohols, a close analogue of the amidation reaction (eq 1). Despite the synthetic purposes both of them usually require a chemical activation of the carboxylic groups, at elevated temperatures one-step direct condensation is possible where water is released. In the present paper we report on a comparative QM/MM study of the gas-phase esterification with methanol on monocarboxy-substituted armchair and zigzag SWNT tips, according to the following reaction scheme: SWNT-COOH + HOCH3 f SWNT-CO-OCH3 + H2O (2) The SWNT models employed in this work were in many regards similar to those used in our previous papers.19,20 In particular, the high (quantum mechanics) level included the atoms belonging to methanol molecule and the terminal carboxylic group, together with their adjacent C and H-atoms, as shown in Figure 1. In this way, an important parameter 836

such as the number and nature of high level-treated atoms was kept the same in the armchair and zigzag models. On the other hand, we rejected the nanotube diameter used previously, i.e., (10,0) for zigzag and (5,5) for armchair SWNT models, since a detailed inspection27 revealed that within such narrow cavities, the opposite wall attracts methylamine molecule, making it equilibrate close to the nanotube axis. In turn, this can dramatically influence its position with respect to the carboxylic group, resulting in unrealistically long N-H‚‚‚O separations, which are supposed to be hydrogen bonds, and apparently producing significant errors in energy estimates for the reaction complexes. To avoid this sort of undesirable effect, in this work we used armchair (10,10) and zigzag (16,0) nanotube models, with calculated diameters of ca. 1.3-1.4 nm approaching commonly observed SWNT diameters. The fragment length was also increased, to three complete rows of carbon hexagons, as shown in Figure 1. The reactions were studied using a two-level ONIOM approach21,22 implemented in the Gaussian 98W suite of programs.22,28 The universal force field29 (UFF) molecular mechanics was used for the low-level treatment, and the Becke’s three-parameter hybrid method30 with the exchange functional of Lee, Yang, and Parr31 (B3LYP) was used for the high-level description, in conjunction with the 6-31G(d) basis set by Pople et al.32-34 The stationary point geometries were fully optimized and characterized as minima (0 imaginary frequencies) or first-order saddle points (1 imaginary frequency) by vibration frequency calculations.35 All the optimizations met the default convergence criteria set in Gaussian 98W. We present not only the calculated B3LYP energies (as it was in the previous works) but also extrapolated ONIOM energies as well as the corresponding zeropoint energy-corrected (ZPE) values for comparison. As it will be seen from the further discussion, despite that they differ (sometimes significantly) in most cases, main conclusions on the reaction thermodynamics remain the same, regardless of either B3LYP, ONIOM, or ZPE-ONIOM energies are considered. Total optimized geometries and fragments treated at the B3LYP/6-31G(d) theoretical level corresponding to stationary points for the gas-phase reaction of monocarboxysubstituted (10,10) armchair and (16,0) zigzag SWNT with methanol are shown in Figures 2 and 3, respectively. The reaction complexes (A-RC and Z-RC, respectively) form in both cases due to hydrogen bonding between the hydroxy group of methanol with the carbonyl group of COOH, with OH‚‚‚OdC separation of 2.090 and 2.004 Å for the armchair and zigzag SWNT, respectively. The carboxylic group is more twisted out of the aromatic ring plane in the armchair case, as can be seen from O-C-C-C dihedral angles of >40°. In terms of the B3LYP energies, Z-RC is noticeably more stable than A-RC, by ca. 3 kcal mol-1 (Table 1); however, for the corresponding ONIOM and ZPE-ONIOM values this difference is insignificant. The esterification reaction in both cases has to pass through a transition state having a high positive energy. The B3LYP, ONIOM, and ZPE-ONIOM values obtained show a reNano Lett., Vol. 2, No. 8, 2002

Figure 2. General views along SWNT axis (upper structures) of the reaction complex (A-RC), transition state (A-TS), and products (A-P) for the model gas-phase esterification reaction of monocarboxy-substituted armchair (10,10) SWNT with methanol, and fragments treated at the B3LYP/6-31G(d) theoretical level (lower structures) with some details of their geometry (interatomic distances in Å, bond angles and dihedrals in degrees). Table 1: Calculated B3LYP, Extrapolated ONIOM Energies, and the Corresponding Zero-Point Energy-Corrected Values (ZPE-ONIOM) for Stationary Points in the Model Gas-Phase Esterification Reaction of Terminal Carboxylic Groups of Armchair (10,10) and Zigzag (16,0) SWNTs with Methanola

A-RC A-TS ∆EA-TS A-P A-Psep Z-RC Z-TS1 ∆EZ-TS1 Z-P Z-Psep Z-TS2 ∆EZ-TS2 Z-SP Z-SPsep

B3LYP

ONIOM

ZPE-ONIOM

-3.5 38.7 42.2 -3.3 4.7 -6.8 39.3 46.1 7.4 16.0 17.3 24.1 -21.1 -14.2

-12.7 29.9 42.6 -15.5 -3.4 -13.4 29.2 42.6 2.7 11.3 14.2 27.6 -30.5 -22.7

-12.4 27.3 39.7 -16.9 -5.7 -12.7 26.0 38.7 2.0 8.8 13.0 25.7 -30.0 -24.4

a All energies are relative to the reactant level (except for barrier heights ∆E), in kcal mol-1. Abbreviations: A, armchair; Z, zigzag; RC, reaction complex; TS, transition state; P, product; SP, side product; Psep, separated products.

markable similarity between the two nanotube forms: 38.7, 29.9, and 27.3 kcal mol-1 for the armchair (A-TS; Table 1) and 39.3, 29.2, and 26.0 kcal mol-1 for the zigzag (Z-TS1), respectively. For A-TS, the corresponding B3LYP, ONIOM and ZPE-ONIOM values for barrier heights (∆ETS) are also very close to each other, of 42.2, 42.6, and 39.7 kcal mol-1; whereas it is not the case for its zigzag counterpart (46.1, 42.6, and 38.7 kcal mol-1, respectively). The differences in transition state geometry worth mentioning are longer distances from the carboxylic C atom to the oxygen atoms of OH and OCH3 in Z-TS1 (1.985 and 2.108 Å, respectively; Nano Lett., Vol. 2, No. 8, 2002

Figure 3) as compared to A-TS (1.840 Å both; Figure 2). Also, in the latter case, the migrating H atom remains at an approximately equal distance from the OH and OCH3 groups (1.180 and 1.220 Å, respectively), whereas in Z-TS1 it is noticeably closer to the hydroxy moiety (at 1.105 Å) than to the methoxy group (at 1.338 Å). The above transition states were initially found with smaller basis sets for the B3LYP level, to save computation cost. For A-TS, this procedure was going smoothly, by gradually applying STO-3G, 3-21G, 4-31G, and finally 6-31G(d). In the structures preoptimized with the smallest two basis sets, the methyl group of OCH3 moiety is directed toward the nanotube cavity; with 4-31G, the geometry changed to A-TS shown in Figure 2, where the methyl group is directed outward the SWNT cavity. Similar lower-theory (STO-3G and 3-21G) “toward-cavity” structures were found for the zigzag transition state as well. However, after applying the 4-31G basis set we obtained a quite unexpected result. Instead of producing a refined (but essentially the same) structure, the methoxy moiety inevitably moved to the neighboring terminal C atom (CH group). The resulting firstorder saddle point (Z-TS2; Figure 3) is no more related to the ester formation but leads to a ketene-ether side product Z-SP. Interestingly, all Z-TS2 relative energies (B3LYP, ONIOM, and ZPE-ONIOM of 17.3, 14.2, and 13.0 kcal mol-1, respectively; Table 1) are roughly half as large as those for Z-TS1; the barrier height values drop in a similar way, by 1.5-2 times, as compared to ∆EZ-TS1. Nevertheless, although this side reaction pathway is not senseless from common chemical considerations, it can hardly have practical implications for the derivatization purposes, since the presence of terminal CH groups seems unlikely after the strongacid treatment employed for the carboxylation of nanotube tips. (The “correct” transition state structure Z-TS1 was 837

Figure 3. General views along SWNT axis of the reaction complex (Z-RC), transition states (Z-TS1 and Z-TS2), products (Z-P) and side-products (Z-SP) for the model gas-phase esterification reaction of monocarboxy-substituted zigzag (16,0) SWNT with methanol, and fragments treated at the B3LYP/6-31G(d) theoretical level with some details of their geometry (interatomic distances in Å, bond angles and dihedrals in degrees).

finally obtained manually, by turning the methyl group outward the nanotube cavity.) The Z-SP formation is highly exothermic: the B3LYP, ONIOM, and ZPE-ONIOM relative energies are -21.1, -30.5, and -30.0 kcal mol-1 for the H-bonded products (ZSP; Table 1), and -14.2, -22.7, and -24.4 kcal mol-1 for the separated products (Z-SPsep). In the case of armchair SWNT we also obtained mostly negative values, in particular B3LYP, ONIOM, and ZPE-ONIOM of -3.3, -15.5, and -16.9 kcal mol-1 for the H-bonded products (A-P), respectively. It is the energies of ester products where the most striking difference in the armchair and zigzag nanotube behavior manifests: none of the B3LYP, ONIOM, and ZPEONIOM relative energies for Z-P (7.4, 2.7, and 2.0 kcal mol-1) and Z-Psep (16.0, 11.3, and 8.8 kcal mol-1) are negative, contrary to the values discussed above. Thus, the esterification reaction on the zigzag SWNT tips turns out to be thermodynamically unfaVorable. At the same time, A-P and Z-P do not exhibit any substantial geometric differences. In particular, dihedral angles reflecting orientation of the ester fragment with respect to the nanotube wall differ by a few 838

degrees only, and variations of the H-bond lengths between the water molecule and the ester fragment do not exceed 0.16 Å. The thermodynamic results obtained here for esterification reaction 2 are more explicit than our previous calculations for the direct amidation of carboxylated SWNT tips (reaction 1).19,20 From the practical point of view, they might imply that treating the oxidized CNTs with aliphatic alcohols, under the gas-phase high-temperature conditions similar to the ones described previously by us,19 would derivatize selectively COOH groups at the armchair nanotube tips. If a long chain alcohol is used for the treatment, the resulting esterderivatized armchair CNTs would acquire much better solubility in organic solvents than the underivatized zigzag nanotubes have, that in turn opens a facile route to their separation. For a final conclusion, our theoretical results need to be verified experimentally. Acknowledgment. The author thanks the Japan Society for the Promotion of Science (JSPS) for the JSPS Invitation Fellowship (grant No. L01536), the National Autonomous Nano Lett., Vol. 2, No. 8, 2002

University of Mexico for financial support through DGAPAIN102900 grant, and the Faculty of Engineering of Yokohama National University for hospitality. References (1) Chen, J.; Hamon, M. A.; Hu, H.; Chen, Y.; Rao, A. M.; Eklund, P. C.; Haddon, R. C. Science 1998, 282, 95. (2) Wong, S. S.; Joselevich, E.; Woolley, A. T.; Cheung, C. L.; Lieber, C. M. Nature 1998, 394, 52. (3) Wong, S. S.; Woolley, A. T.; Joselevich, E.; Cheung, C. L.; Lieber, C. M. J. Am. Chem. Soc. 1998, 120, 8557. (4) Hamon, M. A.; Chen, J.; Hu, H.; Chen, Y.; Itkis, M. E.; Rao, A. M.; Eklund, P. C.; Haddon, R. C. AdV. Mater. 1999, 11, 834. (5) Chen, Y.; Chen. J.; Hu, H.; Hamon, M. A.; Itkis, M. E.; Haddon, R. C. Chem. Phys. Lett. 1999, 299, 532. (6) Mickelson, E. T.; Chiang, I. W.; Zimmerman, J. L.; Boul, P. J.; Lozano, J.; Liu, J.; Smalley, R. E.; Hauge, R. H.; Margrave, J. L. J. Phys. Chem. B 1999, 103, 4318. (7) Boul, P. J.; Liu, J.; Mickelson, E. T.; Huffman, C. B.; Ericson, L. M.; Chiang, I. W.; Smith, K. A.; Colbert, D. T.; Hauge, R. H.; Margrave, J. L.; Smalley, R. E. Chem. Phys. Lett. 1999, 310, 367. (8) Tang, B. Z.; Xu, H. Macromolecules 1999, 32, 2569. (9) Ausman, K. D.; Piner, R.; Lourie, O.; Ruoff, R. S.; Korobov, M. J. Phys. Chem. B 2000, 104, 8911. (10) Riggs, J. E.; Guo, Z.; Carroll, D. L.; Sun, Y.-P. J. Am. Chem. Soc. 2000, 122, 5879. (11) Liu, Z.; Shen, Z.; Zhu, T.; Hou, S.; Ying, L.; Shi, Z.; Gu, Z. Langmuir 2000, 16, 3569. (12) Riggs, J. E.; Walker, D. B.; Carroll, D. L.; Sun, Y.-P. J. Phys. Chem. B 2000, 104, 7071. (13) Jin, Z.; Sun, X.; Xu, G.; Goh, S. H.; Ji, W. Chem. Phys. Lett. 2000, 318, 505. (14) Chen, J.; Rao, A. M.; Lyuksyutov, S.; Itkis, M. E.; Hamon, M. A.; Hu, H.; Cohn, R. W.; Eklund, P. C.; Colbert, D. T.; Smalley, R. E.; Haddon, R. C. J. Phys. Chem. B 2001, 105, 2525. (15) Chen, Q.; Dai, L.; Gao, M.; Huang, S.; Mau, A. J. Phys. Chem. B 2001, 105, 618. (16) Kong, J.; Dai, H. J. Phys. Chem. B 2001, 105, 2890. (17) Niyogi, H.; Hu, H.; Hamon, M. A.; Bhowmik, P.; Zhao, B.; Rozenzhak, S. M.; Chen, J.; Itkis, M. E.; Meier, M. S.; Haddon, R. C. J. Am. Chem. Soc. 2001, 123, 733. (18) Fu, K.; Huang, W.; Lin, Y.; Riddle, L. A.; Carroll, D. L.; Sun, Y.-P. Nano Lett. 2001, 1, 439. (19) Basiuk, E. V.; Basiuk, V. A.; Ban˜uelos, J.-G.; Saniger-Blesa, J.-M.; Pokrovskiy, V. A.; Gromovoy, T. Yu.; Mischanchuk, A. V.; Mischanchuk, B. G. J. Phys. Chem. B 2002, 106, 1588.

Nano Lett., Vol. 2, No. 8, 2002

(20) Basiuk, V. A.; Basiuk, E. V.; Saniger-Blesa, J.-M. Nano Lett. 2001, 1, 657. (21) Svensson, M.; Humbel, S.; Froese, R. D. J.; Matsubara, T.; Sieber, S.; Morokuma, K. J. Phys. Chem. 1996, 100, 19357. (22) Dapprich, S.; Koma´romi, I.; Byun, K. S.; Morokuma, K.; Frisch, M. J. J. Mol. Struct.: THEOCHEM 1999, 461-462, 1. (23) Bauschlicher, C. W. Chem. Phys. Lett. 2000, 322, 237. (24) Bauschlicher, C. W. Nano Lett. 2001, 1, 223. (25) (a) Froudakis, G. E. Nano Lett. 2001, 1, 179. (b) Froudakis, G. E. J. Phys.: Condens. Matter 2002, 14, 453. (26) Bauschlicher, C. W.; So, C. R. Nano Lett. 2002, 2, 337. (27) Basiuk, V. A., in preparation. (28) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, Jr., J. A.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; 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. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98W, Revision A.9, Gaussian, Inc.: Pittsburgh, PA, 1998. (29) Rappe, A. K.; Casewit, C. J.; Colwell, K. S.; Goddard, W. A.; Skiff, W. M. J. Am. Chem. Soc. 1992, 114, 10024. (30) Becke, A. D. J. Phys. Chem. 1993, 98, 5648. (31) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785. (32) Hariharan, P. C.; Pople, J. A. Chem. Phys. Lett. 1972, 66, 217. (33) Binkley, J. S.; Pople, J. A. Int. J. Quantum Chem. 1975, 9, 229. (34) Krishnan, R.; Frisch, M. J.; Pople, J. A. J. Chem. Phys. 1980, 72, 4244. (35) In some cases of hydrogen-bonded reaction complexes and products, using the default criteria does not allow to get rid of one small (ca. 20 cm-1) imaginary frequency in the calculated IR spectrum. It always corresponds to rotation of the complex around the CringC(dO)O bond. Setting tight convergence criteria can eliminate the imaginary frequency. However, this considerably increases computation time, and on the other hand produces a negligible effect on the energy values.

NL025607N

839