Paper SS-TuP6
Hydrogen Termination of Si(110) Surface in UHV Conditions

Tuesday, October 19, 2010, 6:00 pm, Room Southwest Exhibit Hall

Hydrogen termination on silicon surfaces is very important in science and technology. Hydrogen (H) is used for surface passivation, and may play role of surfactant for epitaxial growth and also present in many technological processes including chemical vapor deposition. The behavior of H on Si(100) and Si(111) surfaces is extensively studied. However, the information on interaction of H with Si(110) is very limited. The (1x1) bulk-terminated structure was reported for chemically hydrogenated Si(110) in solution [1,2]. The surface structure and processes occur on Si(110) hydrogenated in ultra-high vacuum (UHV) may be very different from chemically treated samples. There are very few reports on in situ hydrogenation of Si(110) surface. Here, we report scanning tunneling microscopy (STM) study of atomic H adsorption on Si(110) in UHV.

The high efficiency hot capillary source was used to produce atomic H flux. The large concentration of highly reactive H atoms in the flux allows us to reach surface saturation for hydrogenation at ~2 L exposure. Chemically etched tungsten tips have been used for STM (Omiron VT-system).

The initial adsorption process of hydrogen reveals itself as suppression of individual protrusions of pentagon pairs on STM image. The number of these “missing” protrusions increases with H coverage, although the overall “16x2” structure of clean Si(110) preserves. More drastic changes were observed on unreconstructed areas exposed to small amount of H at elevated temperatures. Si pentagon pairs, usually randomly distributed on these areas, were arranged in patches with local “5x4” periodicity. Further exposure at room temperature (RT) led to loosing of atomic resolution on STM images. At higher temperatures (400~500ºC) the surface have reconstructed to the zigzag row structures. Depending on relative shift of neighboring zigzag rows the local periodicity was either (2x3) or (2,1)x(0,3). The up-and-down terrace structure of “16x2” reconstruction could still be observed, probably, due to insufficient temperature to initiate massive Si transport to smooth out the surface. At temperatures more than ~550ºC the H atoms start to desorb from Si(110). These kinds of structure have not been observed before for H/Si(110) surface. No (1x1) reconstruction reported for chemically hydrogenated Si(110) was confirmed. Only small patches of (1x1) were observed in very narrow temperature window at ~350ºC.

References

[1] K. Arima, J. Ktoh, and K. Endo, Appl. Phys. Lett. 85 (2004) 6254.

[2] S. Horie, K. Arima, K. Hirose, J. Katoh, T. Ono, and K. Endo, Phis. Rev. B 72 (2005) 113306.