Philosophical Magazine Letters, 65(6), 291-298 (1992)
In tables 3 and 4 we compare the calculated modes with experiment and other ab initio density-functional calculations (Adams et al. 1991, Feuston, Androni, Feuston and Clementi 1991). The greated error in the Raman modes is nearly 90 cm-1 and it occurs for the lowest Ag mode. In the calculation of Feuston et al. (1991) there was a problem in determining the higher Hg modes. It is clear that our errors are somewhat less than these.
The infra-red modes are given in table 4. Again our error is less than about 90 cm-1 compared with 80 cm-1 for Feuston et al. and 200 cm-1 for Adams et al. (1991). It is interesting that the Musgrave-Pople potential is able to account for almost all the modes with a similar error.
We then projected the forces deduced from the addition of three electrons to the ball onto each normal mode coordinate calculated using the Musgrave-Pople potential above. The most significant overlap occurs with the lowest Ag mode at 580 cm-1 and to a lesser extent the upper Ag mode. The normal coordinates of these modes are shown in Figs. 2a and 2b. Clearly the lower frequency one involves a breathing mode of the molecule and the higher a tangential displacement field. Thus the electron-phonon coupling will be strongest for these modes. The high values of the superconducting transition temperature may then be related to these high frequencies.
In conclusion, we have shown that the addition of three electrons to the molecule results in the shortening of the pentagonal bonds. We have evaluated, for the first time, all the dynamical modes by an ab initio method and shown that this distortion is strongly coupled with the lowest vibratory Ag mode.
[Abstract] [Introduction] [Method] [Vibrational modes] [Conclusions] [Acknowledgements] [References] [Table 1] [Table 2] [Table 3] [Table 4] [Figure 1] [Figure 2]
| Christopher D. Latham | HTML 3.2: [W3C][WDG] |