Systematic Search for Isomerization Pathways of Hexasilabenzene for

Mar 19, 2009 - Systematic Search for Isomerization Pathways of Hexasilabenzene for Finding Its Kinetic Stability. Masahiro Moteki, Satoshi Maeda ... T...
0 downloads 4 Views 1MB Size
Download PDF
2218

Organometallics 2009, 28, 2218–2224

Systematic Search for Isomerization Pathways of Hexasilabenzene for Finding Its Kinetic Stability Masahiro Moteki, Satoshi Maeda, and Koichi Ohno* Department of Chemistry, Graduate School of Science, Tohoku UniVersity, Aramaki, Aoba-ku, Sendai 980-8578, Japan ReceiVed September 10, 2008

Isomerization pathways of hexasilabenzene (Si6H6) were theoretically searched by using the anharmonicdownward-distortion-following (ADD-following) method, which is a recently established systematic and automatic transition-state (TS) finding method. The search revealed the lowest barrier from the sixmembered-ring structure in hexasilabenzene, and the barrier height was computed to be only 74 kJ mol-1. It follows that the six-membered ring in hexasilabenzene can easily be deformed via the TS structure. The present study identified three lower-lying TS structures around hexasilabenzene by the systematic search. These TS structures will be useful for designing a suitable substituent in a hexasilabenzene derivative for its isolation in the future, since such a substituent should be designed so that all TSs deforming the six-membered-ring backbone in a hexasilabenzene derivative are destabilized. Introduction Silicon-hydrides (SinHm) show different characteristics from hydrocarbons: weaker π-π interaction,1-3 considerably bent Si-Si multiple bonds,1-12 formation of electron-poor bonds,6-8,12-21 and so on. Moreover, the hydrocarbon analogues satisfying the octet rule are not the most stable isomers on the potential energy surface (PES) of SinHm.21-29 Hence, synthesis and isolation of such structures have been done under stereoprotected conditions.30-32 Although five (CH)6 isomers, i.e., benzene, benzvalene, Dewar-benzene, prismane, and bicyclopropenyl, have been * Corresponding author. Tel: +81-22-795-6576. Fax: +81-22-795-6580. E-mail: [email protected]. (1) Power, P. P. Chem. ReV. 1999, 99, 3463. (2) Kira, M. Pure Appl. Chem. 2000, 72, 2333. (3) Frenking, G.; Krapp, A.; Nagase, S.; Takagi, N.; Sekiguchi, A. ChempPhysChem 2006, 7, 799. (4) Poirier, R. A.; Goddard, J. D. Chem. Phys. Lett. 1981, 80, 37. (5) Luke, B. T.; Pople, J. A.; Kroghjespersen, M. B.; Apeloig, Y.; Karni, M.; Chandrasekhar, J.; Schleyer, P. v. R. J. Am. Chem. Soc. 1986, 108, 270. (6) Malrieu, J.-P.; Trinquier, G. J. Am. Chem. Soc. 1989, 111, 5916. (7) Jacobsen, H.; Ziegler, T. J. Am. Chem. Soc. 1994, 116, 3667. (8) Nagase, S.; Kobayashi, K.; Takagi, N. J. Organomet. Chem. 2000, 611, 264. (9) Malcolm, N. O. J.; Gillespie, R. J.; Popelier, P. L. A. J. Chem. Soc., Dalton Trans. 2002, 3333. (10) Kosa, M.; Karni, M.; Apeloig, Y. J. Am. Chem. Soc. 2004, 126, 10544. (11) Veszpre´mi, T.; Petrov, K.; Nguyen, C. T. Organometallics 2006, 25, 1480. (12) Power, P. P. Organometallics 2007, 26, 4362. (13) Ko¨hler, H.-J.; Lischka, H. Chem. Phys. Lett. 1984, 112, 33. (14) Trinquier, G. J. Am. Chem. Soc. 1990, 112, 2130. (15) Bogey, M.; Bolvin, H.; Demuynck, C.; Destombes, J. L. Phys. ReV. Lett. 1991, 66, 413. (16) Cordonnier, M.; Bogey, M.; Demuynck, C.; Destombes, J. L. J. Chem. Phys. 1992, 97, 7984. (17) Grev, R. S.; Schaefer, H. F., III. J. Chem. Phys. 1992, 97, 7990. (18) Zyubin, A. S.; Dembovsky, S. A. Solid State Commun. 1993, 87, 175. (19) Srinivas, G. N.; Jemmis, G. D. J. Am. Chem. Soc. 1997, 119, 12968. (20) Srinivas, G. N.; Yu, L. W.; Schwartz, M. J. Chem. Soc., Dalton Trans. 2002, 1857.

synthesized,33-36 only the hexasilaprismane derivative with six sp3 Si atoms and a prism-like backbone was synthesized37 for all Si-substituted analogues. Concerning silabenzene derivatives with a cyclic SinC6-n backbone, the compounds with n ) 138 and 239 have been synthesized to date.40 The (SiH)6 species have been investigated by theoretical calculations.41-46 Among (21) Sari, L.; McCarthy, M. C.; Schaefer, H. F., III.; Thaddeus, P. J. Am. Chem. Soc. 2003, 125, 11409. (22) Binkley, J. S. J. Am. Chem. Soc. 1984, 106, 603. (23) Sax, A. F. J. Comput. Chem. 1985, 6, 469. (24) Nagase, S.; Nakano, M. Angew. Chem., Int. Ed. Engl. 1988, 27, 1081. (25) Koseki, S.; Gordon, M. S. J. Phys. Chem. 1989, 93, 118. (26) Yates, B. F.; Schaefer, H. F., III. Chem. Phys. Lett. 1989, 155, 563. (27) Miyazaki, T.; Uda, T.; Sˇtich, I.; Terakura, K. Chem. Phys. Lett. 1996, 261, 346. (28) Ge, Y.; Head, J. D. Chem. Phys. Lett. 2004, 398, 107. (29) Kosa, M.; Karni, M.; Apeloig, Y. J. Chem. Theory Comput. 2006, 2, 956. (30) West, R.; Fink, M. J.; Michl, J. Science 1981, 214, 1343. (31) Ishida, S.; Iwamoto, T.; Kabuto, C.; Kira, M. Nature 2003, 421, 725. (32) Sekiguchi, A.; Kinjo, R.; Ichinohe, M. Science 2004, 305, 1755. (33) First synthesis: Foote, J. K.; Mallon, M. H.; Pitts, J. N. J. Am. Chem. Soc. 1966, 88, 3698. (34) First synthesis: van Tamelen, E. E.; Pappas, S. P. J. Am. Chem. Soc. 1963, 85, 3297. (35) First synthesis: Katz, T. J.; Acton, N. J. Am. Chem. Soc. 1973, 95, 2738. (36) First synthesis of (CH)6: Billups, W. E.; Haley, M. M. Angew. Chem., Int. Ed. Engl. 1989, 28, 1711. (37) Sekiguchi, A.; Yatabe, T.; Kabuto, C.; Sakurai, H. J. Am. Chem. Soc. 1993, 115, 5853. (38) (a) Wakita, K.; Tokitoh, R.; Okazaki, R.; Nagase, S. Angew. Chem., Int. Ed. Engl. 2000, 39, 634. (b) Wakita, K.; Tokitoh, R.; Okazaki, R.; Takagi, N.; Nagase, S. J. Am. Chem. Soc. 2000, 122, 5648. (39) Kinjo, R.; Ichinohe, M.; Sekiguchi, A.; Takagi, N.; Sumimoto, M.; Nagase, S. J. Am. Chem. Soc. 2007, 129, 7766. (40) Lee, V. Y.; Sekiguchi, A. Chem. Soc. ReV. 2008, 37, 1652. (41) Nagase, S.; Kudo, T.; Aoki, M. J. Chem. Soc., Chem. Commun. 1985, 1121. (42) Clabo, D. A. Jr.; Schaefer, H. F., III. J. Chem. Phys. 1986, 84, 1664. (43) Sax, A. F.; Janoschek, R. Angew. Chem., Int. Ed. Engl. 1986, 25, 651. (44) Nagase, S.; Teramae, H.; Kudo, T. J. Chem. Phys. 1987, 86, 4513. (45) Slanina, Z. Chem. Phys. Lett. 1989, 161, 175.

10.1021/om800881y CCC: $40.75  2009 American Chemical Society Publication on Web 03/19/2009

Isomerization Pathways of Hexasilabenzene

Organometallics, Vol. 28, No. 7, 2009 2219

Figure 1. Equilibrium structures and relative energies (kJ mol-1) of Si6H6 at the B3LYP/6-311G** level, where (A) shows the 15 most stable isomers and (B) the four (SiH)6 isomers.

the (SiH)6 isomers, hexasilaprismane was computed to be the most stable,41,43,46 although prismane is much less stable than benzene in the (CH)6 case.44,47 On the other hand, hexasilabenzene has been studied extensively41,42,44 as an analogue of benzene, and its aromaticity has also been discussed.48-53 It was predicted to have a nonplanar and chairlike structure because of weaker π-electron donation to bond than that of benzene,44 while it has been shown that the planar structure can be stabilized at the dianion state.54 Small SinHm have been studied as a microscopic model of semiconductor devices.27,28 Although bare Sin clusters have been extensively investigated as the microscopic model,55-67 surfaces (46) Zhao, M.; Gimarc, B. M. Inorg. Chem. 1996, 35, 5378. (47) Dinadayalane, T. C.; Priyakumar, U. D.; Sastry, G. N. J. Phys. Chem. A 2004, 108, 11433. (48) Ohanessian, G.; Hiberty, P. C.; Lefour, J.-M.; Flament, J.-P.; Shaik, S. S. Inorg. Chem. 1988, 27, 2219. (49) Shaik, S. S.; Hiberty, P. C.; Ohanessian, G.; Lefour, J.-M. J. Phys. Chem. 1988, 92, 5086. (50) Schleyer, P. v. R.; Jiao, H.; Hommes, N. J. R. v. E.; Malkin, V. G.; Malkina, O. N. J. Am. Chem. Soc. 1997, 119, 12669. (51) Baldridge, K. K.; Uzan, O.; Martin, J. M. L. Organometallics 2000, 19, 1477. (52) Sakai, S. J. Phys. Chem. A 2002, 106, 10370. (53) Engelberts, J. J.; Havenith, R. W. A.; van Lenthe, J. H.; Jenneskens, L. W.; Fowler, P. W. Inorg. Chem. 2005, 44, 5266. (54) Takahashi, M.; Kawazoe, Y. Comput. Mater. Sci. 2006, 36, 30. (55) Toma´nek, D.; Schlu¨ter, M. A. Phys. ReV. Lett. 1986, 56, 1055. (56) Raghavachari, K. J. Chem. Phys. 1986, 84, 5672. (57) Ballone, P.; Andreoni, W.; Car, R.; Parrinello, M. Phys. ReV. Lett. 1988, 60, 271. (58) Cullis, A.; Canham, L. Nature 1991, 353, 335. (59) Grossman, J. C.; Mita´sˇ, L. Phys. ReV. Lett. 1995, 74, 1323. (60) Scha¨fer, R.; Schlecht, S.; Woenckhaus, J.; Becker, J. A. Phys. ReV. Lett. 1996, 76, 471.

of practical devices are passivated by inert species such as H atoms. Hence, SinHm is an alternative limit of the microscopic model.68-70 In connection with the interest, experimental syntheses and observations of SinHm+ clusters were performed,71-74 and the global minima on the PESs of SinHm have been explored by the plane-wave pseudopotential calculation27 and the semiempirical AM1 theory.28 Stable SinHm clusters found in these ¨ gu¨t, S.; Chelikowsky, J. R. Phys. ReV. Lett. 1997, (61) Vasiliev, I.; O 78, 4805. (62) Ho, K. M.; Shvartsburg, A. A.; Pan, B.; Lu, Z. Y.; Wang, C. Z.; Wacker, J. G.; Fye, J. L.; Jarrold, M. F. Nature 1998, 392, 582. (63) (a) Bazterra, V. E.; Caputo, M. C.; Ferraro, M. B.; Fuentealba, P. J. Chem. Phys. 2002, 117, 11158. (b) Bazterra, V. E.; On˜a, O.; Caputo, M. C.; Ferraro, M. B.; Fuentealba, P.; Facelli, J. C. Phys. ReV. A 2004, 69, 053202. (c) On˜a, O.; Bazterra, V. E.; Caputo, M. C.; Facelli, J. C.; Fuentealba, P.; Ferraro, M. B. Phys. ReV. A 2006, 73, 053203. (64) Sun, Q.; Wang, Q.; Jena, P.; Waterman, S.; Kawazoe, Y. Phys. ReV. A 2003, 67, 063201. (65) Maroulis, G.; Be´gue´, D.; Pouchan, C. J. Chem. Phys. 2003, 119, 794. (66) (a) Yoo, S.; Zeng, X. C.; Zhu, X. L.; Bai, J. J. Am. Chem. Soc. 2003, 125, 13318. (b) Yoo, S.; Zeng, X. C. J. Chem. Phys. 2005, 123, 164303. (c) Yoo, S.; Shao, N.; Koehler, C.; Fraunhaum, T.; Zeng, Y. C. J. Chem. Phys. 2006, 124, 164311. (67) (a) Nigam, S.; Majumder, C.; Kulshreshtha, S. K. J. Chem. Phys. 2004, 121, 7756. (b) Nigam, S.; Majumder, C.; Kulshreshtha, S. K. J. Chem. Phys. 2006, 125, 074303. (68) Canham, L. T. Appl. Phys. Lett. 1990, 57, 1046. (69) (a) Hirao, M.; Uda, T. Surf. Sci. 1994, 306, 87. (b) Hirao, M.; Uda, T. Int. J. Quantum Chem. 1994, 52, 1113. (70) Onida, G.; Andreoni, W. Chem. Phys. Lett. 1995, 243, 183. (71) Haller, I. Appl. Phys. Lett. 1980, 37, 282. (72) Murakami, M.; Kanayama, T. Appl. Phys. Lett. 1995, 67, 2341. (73) Watanabe, M. O.; Murakami, H.; Miyazaki, T.; Kanayama, T. Appl. Phys. Lett. 1997, 71, 1207. (74) Rechtsteiner, G. A.; Hampe, O.; Jarrold, M. F. J. Phys. Chem. B 2001, 105, 4188.

2220 Organometallics, Vol. 28, No. 7, 2009

Moteki et al.

Figure 2. Energy profile for conversion pathways among hexasilabenzene 2-EQ17, hexasilaprismane 2-EQ12, and the global minimum of Si6H6 2-EQ1 at the B3LYP/6-311G** level.

studies are not similar to any stable CnHm isomers satisfying the octet rule, and it was suggested that they have chemical properties resembling the Sin clusters.27 For the above two purposes, two different energy ranges on the PES of Si6H6, i.e., the high-energy region for the (SiH)6 isomers and the low-energy area including the cluster-like global minimum, have been investigated separately. Although hexasilabenzene is a local minimum on the PES of Si6H6,42,44,45 it may be just a metastable structure that immediately isomerizes to the more stable cluster-like structures. However, it might be isolated at least in the gas phase if it is sufficiently stable against the lowest transition state (TS) for its isomerization. Although a thorough search for isomerization pathways of hexasilabenzene is necessary to elucidate its kinetic stability from theoretical calculations, it has not yet been performed hitherto because of the difficulty in finding many TS structures in a systematic way. Since the topography on PESs of SinHm is much more complicated than those of CnHm because of the existence of the cluster-like structures, electron-poor bonds, bend multiple bonds, etc., a search using initial guesses based on intuition or experience is likely to overlook important TSs in comparison with the CnHm case. Moreover, a high-quality quantummechanical calculation should be employed in such TS searches to describe subtle interactions related to the characteristics of SinHm that cannot easily be modeled empirically. However, there has been no practical systematic TS-finding method that can be applied to PESs based on such expensive calculations. Recently, we discovered that the anharmonic downward distortion (ADD) can be a signpost of chemical reactions leading to TSs from an arbitrary equilibrium (EQ) structure.75 Base on this principle, we successfully developed the ADD-following method, which enables one to systematically search for TSs around an EQ starting from the EQ,75,76 and it has been employed in automated global reaction route mapping (GRRM) (75) Maeda, Maeda, (76)

(a) Ohno, K.; Maeda, S. Chem. Phys. Lett. 2004, 384, 277. (b) S.; Ohno, K. J. Phys. Chem. A 2005, 109, 5742. (c) Ohno, K.; S. J. Phys. Chem. A 2006, 110, 8933. Maeda, S.; Ohno, K. J. Phys. Chem. A 2007, 111, 4527.

on the PES of many species.75,77-80 In this study, we applied the ADD-following method to hexasilabenzene and hexasilaprismane and systematically located their isomerization pathways as well as interconversion pathways among hexasilabenzene, hexasilaprismane, and the cluster-like global minimum. On the basis of the information, we discuss the kinetic stability of hexasilabenzene for its isolation.

Methodology The ADD-following method was proposed to globally search for EQs and TSs in an automatic way. Although the PES around an EQ point can be approximated by the harmonic function, the real PES should always distort downward from the harmonic surface in the direction of other EQs or dissociation channels (DCs) due to stabilization effects related to other potential minima or flat regions leading to a DC. Hence, the ADD can be a signpost of chemical reactions, and many reaction channels can be discovered by following the ADDs.75 When all ADDs are followed by the full-ADD-following method, almost all of the reaction channels including very high-energy processes can be searched around an EQ.75 By applying the fullADD-following to all EQs obtained during the process, a global reaction route map for a given chemical formula can be obtained automatically.75,77-80 On the other hand, the large-ADD-following method was proposed for quick exploration of low-barrier pathways only,76 and it was applied to hydrogen-bond clusters76,81,82 and organometallic catalytic systems.83-85 In this study, we employed both the full-ADD-following and the large-ADD-following methods as described below. (77) Yang, X.; Maeda, S.; Ohno, K. J. Phys. Chem. A 2005, 109, 7319. (78) Yang, X.; Maeda, S.; Ohno, K. Chem. Phys. Lett. 2006, 418, 208. (79) Yang, X.; Maeda, S.; Ohno, K. J. Phys. Chem. A 2007, 111, 5099. (80) Watanabe, Y.; Maeda, S.; Ohno, K. Chem. Phys. Lett. 2007, 447, 21. (81) (a) Luo, Y.; Maeda, S.; Ohno, K. J. Phys. Chem. A 2007, 111, 10732. (b) Luo, Y.; Maeda, S.; Ohno, K. J. Comput. Chem., in press. (82) Maeda, S.; Ohno, K. J. Phys. Chem. A 2008, 112, 2962. (83) Maeda, S.; Ohno, K. J. Phys. Chem. A 2007, 111, 13168. (84) Luo, Y.; Maeda, S.; Ohno, K. J. Phys. Chem. A 2008, 112, 5720. (85) Maeda, S.; Ohno, K. J. Am. Chem. Soc. 2008, 130, 17228.

Isomerization Pathways of Hexasilabenzene

Organometallics, Vol. 28, No. 7, 2009 2221

Figure 3. Transition-state structures for the reaction pathways in Figure 2. The ADD-following method is available in the GRRM program75,76 developed by two of the present authors. We employed the GRRM program in this study throughout, where energy values, gradient vectors, and Hessian matrices were obtained from outputs of the GAUSSIAN03 program.86

(86) 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.; Milliam, 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.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich; 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 C.02; Gaussian, Inc.: Wallingford, CT, 2004.

Results and Discussions At first, we made the large-ADD-following search starting from 12 randomly generated structures at the HF/6-31G(d) level to list the lower-energy cluster-like species. Then, all of the obtained EQs were reoptimized at the B3LYP/6-311G** level, and 125 EQs were located at the B3LYP level. The identical global minimum has been reported by two groups using different computation methods,27,28 and the present search also located the same cage-like structure as the global minimum on the B3LYP surface. Figure 1A shows the 15 most stable EQs obtained in this study. All of these stable species do not satisfy the octet rule. Although benzene is more stable than fulvene of the second most stable isomer of C6H6 by about 130 kJ mol-1,47 the energy gap between 1-EQ1 and 1-EQ2 is only 59.1 kJ mol-1. Figure 1B lists the (SiH)6 isomers at the B3LYP level, where hexasilabicyclopropenyl of a (CH)6 analogue was not optimized at the B3LYP level. As can be seen in the figure, many clusterlike species have much lower energy than these (SiH)6 isomers. In these figures, thick lines represent Si-Si bonds shorter than 2.5 Å and Si-H bonds shorter than 1.7 Å, and thin lines

2222 Organometallics, Vol. 28, No. 7, 2009

Moteki et al.

Figure 4. One-step isomerization pathways and relative energies (kJ mol-1) around (A) 4A-EQ0 and (B) 4B-EQ0 at the G3//B3LYP/6311G** level, where B3LYP/6-311G** values are shown in parentheses.

represent Si-Si bonds shorter than 3.0 Å (longer than 2.5 Å) and Si-H bonds shorter than 2.2 Å (longer than 1.7 Å). We then performed the large-ADD-following search starting from hexasilaprismane and hexasilabenzene to find the conversion pathways leading to the global minimum at the B3LYP/ 6-31G* level. All EQs and TSs obtained by the search were refined at the B3LYP/6-311G** level. Figure 2 represents energy profiles along the intrinsic reaction coordinate87 (IRC) pathways obtained by the search. TS structures for the pathways are also shown in Figure 3, in which broken lines represent generating or dissociating chemical bonds along the IRCs (87) Fukui, K. Acc. Chem. Res. 1981, 14, 363.

starting from the TSs. These IRC routes include hexasilabenzvalene (2-EQ14) and hexasila-Dewar-benzene (2-EQ21) as intermediates, and hence these can also be pathways from 2-EQ14 and 2-EQ21. The pathways from hexasilaprismane (2EQ12) and hexasilabenzene (2-EQ17) intersect at hexasilabenzvalene (2-EQ14) and then undergo the same reaction routes. Two pathways were found between 2-EQ6 and 2-EQ1, and the longer one has the lower maximum energy point along the routes between 2-EQ6 and 2-EQ1. The highest barrier along the pathways is 2-TS6 in Figure 3 (next to 2-EQ3), which is 128.6 kJ mol-1 higher in energy than 2-EQ17. It should be noted that the neighboring isomers along the pathways can easily convert to each other with barriers of ca. 60 kJ mol-1 on average, while

Isomerization Pathways of Hexasilabenzene

Organometallics, Vol. 28, No. 7, 2009 2223

Table 1. Computational Level Dependence of Energetics for Structures in Figure 4

1-EQ1 4A-EQ0 4A-EQ1 4A-EQ2 4A-EQ3 4A-EQ4 4A-EQ5 4A-EQ6 4A-TS1 4A-TS2 4A-TS3 4A-TS4 4A-TS5 4A-TS6 4B-EQ0 4B-EQ1 4B-EQ2 4B-EQ3 4B-EQ4 4B-TS1 4B-TS2 4B-TS3 4B-TS4

B3LYP/ 6-311G**

MP2/ 6-311+G**

G3//B3LYP/ 6-311G**

G3//MP2/ 6-311+G**

CCSD(T)//MP2/ 6-311+G**

ZPE (B3LYP)

ZPE (MP2)

0.0 131.9 141.4 144.5 153.3 164.7 198.7 226.5 207.5 227.8 215.2 177.9 224.8 277.7 150.9 212.7 225.3 232.8 392.6 263.6 237.7 250.8 413.9

0.0 62.3 148.2 133.8 164.4 181.5 220.0 253.6 209.9 228.3 209.2 186.8 228.9 291.6 185.5 242.7 270.5 255.2

0.0 149.0 152.7 139.4 163.5 174.6 227.9 245.0 211.4 220.8 227.3 175.5 245.2 280.3 215.0 233.9 284.6 257.5 490.1 304.9 298.2 289.1 501.6

0.0 149.3 152.5 138.9 162.4 172.3 225.8 243.7 213.1 220.3 213.3 173.4 231.0 280.5 216.6 233.0 286.5 256.6

0.0 155.3 154.1 150.9 168.8 180.2 219.8 244.0 220.2 241.9 215.5 195.1 232.9 297.5 178.0 225.3 256.9 250.4

153.4 143.3 146.4 147.8 144.0 147.5 142.4 145.4 145.3 142.9 143.2 144.7 141.8 141.2 145.9 144.0 145.0 141.3

306.8 300.0 289.6

285.9 273.6 285.4

148.4 140.6 140.9 142.4 139.5 143.8 137.7 139.4 139.0 138.1 136.7 139.6 136.5 136.7 140.6 138.8 138.8 136.5 130.9 136.5 134.9 135.7 129.7

306.4 291.1 292.8

barriers for interconversion among the C6H6 isomers are mostly larger than 300 kJ mol-1.88 It follows that Si6H6 is much more flexible than C6H6. Finally, we applied both the full-ADD-following and the large-ADD-following method at the B3LYP/6-31G* and B3LYP/6-31G** levels, respectively, to hexasilaprismane and hexasilabenzene to systematically find the isomerization pathways starting from them. Geometries for species concerned with the pathways were reoptimized at the B3LYP/ 6-311G** level, and then single-point energies for the B3LYP/6-311G** structures were refined at the G389 level (G3//B3LYP/6-311G**). Figure 4A,B presents all isomerization pathways around hexasilaprismane (4A-EQ0) and hexasilabenzene (4B-EQ0), respectively, obtained by the present search. The lowest barriers from the structures of 4A-EQ0 and 4B-EQ0 are only 26.5 and 74.2 kJ mol-1, respectively. Although most of the products of these reactions are higher in energy than 4A-EQ0 and 4B-EQ0, they can further isomerize into the cluster-like structures without significant barriers, as seen in Figure 2. Activation energies to overcome the barriers in Figure 4 are much lower than usual C-H, C-C, CdC, and CtC bond dissociation energies of 300-800 kJ mol-1, while they are closer to typical hydrogen-bond dissociation energies of 10-100 kJ mol-1. It follows that hexasilaprismane and hexasilabenzene do not have large kinetic stability, unlike (CH)6, while benzene is lower in energy than the neighboring lowest TS by 373.7 kJ mol-1.88 It is surprising that the lowest barrier for 4A-EQ0 is much lower than that of 4B-EQ0 in spite of the synthesis of a hexasilaprismane derivative.37 This may be because the substitution group introduced in the derivative destabilized the TSs in Figure 4A. Although hexasilabenzene derivatives have not yet been synthesized, they can be isolated when substitution groups are designed so that all TSs in Figure 4B are destabilized to avoid unwanted rearrangements in the six-membered-ring backbone. To support the above discussions, we performed geometry optimizations at the MP2 level and energy refinement calculations at the G3 and CCSD(T) levels for structures shown in Figure 4. Table 1 compares energies of the structures relative to the global minimum (1-EQ1 in Figure 1) at different computational levels. Two unimportant

142.5 140.9 141.4

structures, i.e., 4B-TS4 and 4B-EQ4, at the MP2 level are missing in the table since optimizations starting from the B3LYP structures did not give converged MP2 structures. Although MP2 significantly underestimates the relative energy of hexasilaprismane, single-point G3 energies at geometries of the two levels show very similar values. This implies that these two geometries are almost the same. Actually, these geometries are very similar: Si-Si bond lengths in triangles are 2.371 and 2.359 Å for the B3LYP and MP2 geometries, respectively, distances between the top and bottom triangles are 2.377 and 2.361 Å, respectively, and Si-H bond distances are 1.492 and 1.482 Å, respectively. Since higher-level single-point energies at the G3 and CCSD(T) levels are close to the B3LYP one, we estimate that the big difference is due to overestimation of electron correlation energy by the low-order perturbation theory. When energies of B3LYP are compared with other ab initio energies, one can find that B3LYP underestimates energies in a high-energy part of the PES including hexasilabenzene by about 20-50 kJ mol-1. However, the unsigned average and maximum deviations between G3//B3LYP and G3//MP2 are only 2.5 and 14.2 kJ mol-1, respectively. It follows that structures at the B3LYP level are very similar to those of the MP2 level for all structures in Figure 4. When energies at the G3 and CCSD(T) levels are compared, CCSD(T) energies for structures around hexasilabenzene are lower than those of G3 by about 30 kJ mol-1. The minimum activation energy for deformation of hexasilabenzene is estimated to be 74 and 95 kJ mol-1 by G3 and CCSD(T), respectively. Concerning the activation energy for deformation of hexasilaprismane, the minimum one is calculated to be 27 and 40 kJ mol-1 by G3 and CCSD(T), respectively. It follows that the energetics of our final results still have an ambiguity of about 20-30 kJ mol-1, although it does not change the conclusions of this paper at least from a qualitative point of view.

Conclusions Isomerization reaction pathways of hexasilabenzene were explored by using the ADD-following method,75,76 which can systematically and automatically find isomerization channels

2224 Organometallics, Vol. 28, No. 7, 2009

starting from an arbitrary equilibrium structure. Hexasilabenzene was found to isomerize to the more stable clusterlike structures with a barrier of ca. 100 kJ mol-1. The lowest barrier retaining the Si six-membered-ring backbone was computed to be only 74 kJ mol-1 at the G3//B3LYP/6311G** level. It follows that introduction of a suitable substituent that destabilizes all isomerization TSs discovered in this study is necessary for practical isolation of the Si six-membered ring. Theoretical trials for finding such a substituent are under way based on the present TSs and an approach combining the ONIOM method90 and the ADDfollowing method.83,85

(88) Kislov, V. V.; Nguyen, T. L.; Mebel, A. M.; Lin, S. H.; Smith, S. C. J. Chem. Phys. 2004, 120, 7008.

Moteki et al.

Acknowledgment. S.M. is supported by a Research Fellowship of the Japan Society for the Promotion of Science for Young Scientists. Supporting Information Available: A list of energies and xyz coordinates for all structures at the B3LYP/6-311G** level discussed in this paper. This material is available free of charge via the Internet at http://pubs.acs.org. OM800881Y (89) Curtiss, L. A.; Raghavachari, K.; Redfern, P. C.; Rassolov, V.; Pople, J. A. J. Chem. Phys. 1998, 109, 7764. (90) (a) Svensson, M.; Humbel, S.; Froese, R. D. J.; Matsubara, T.; Sieber, S.; Morokuma, K. J. Phys. Chem. 1996, 100, 19357. (b) Dapprich, S.; Koma´romi, I.; Byun, K. S.; Morokuma, K.; Frisch, M. J. J. Mol. Struct. (THEOCHEM) 1999, 461-462, 1. (c) Vreven, T.; Morokuma, K. J. Comput. ¨ .; Schlegel, Chem. 2000, 21, 1419. (d) Vreven, T.; Morokuma, K.; Farkas, O H. B.; Frisch, M. J. J. Comput. Chem. 2003, 24, 760. (e) Morokuma, K.; Wang, Q.; Vreven, T. J. Chem. Theory Comput. 2006, 2, 1317.