Hydrogen Atom Desorption Induced by Electron Bombardment on Si Surface

1. Introduction

Developments of nano-scale quantum device have been expected to open new approaches for information-processing. Among the various quantum devices proposed so far, a solid state qubit is one of the potential candidate devices to bring further progress to the modern computer technology. The nuclear spin qubit proposed by Kane, which is a phosphorus donor atom embedded in a Si lattice, has advantages such as longer relaxation time and higher process compatibility with Si LSI technology [1]. To manipulate a single dopant precisely on a Si substrate, a method employing electron bombardment from an STM tip on hydrogen-terminated Si surface has been proposed [2], [3]. Though successful experimental results including desorption of hydrogen atoms and adsorption of phosphorus atoms with atomic resolution have been reported, the relation between the number of desorbed hydrogen atoms and the number of bombarded electrons, which is one of fundamental issues for fabrication, has not been given in details. In this research, we study hydrogen atom desorption on Si(111) surface quantitatively for aiming at the development of atomic lithography.

2. Method

First, a Si(111) wafer is chemically etched in aqueous NH₄F solution (40%) in order to obtain hydrogen-terminated smooth surface for single atom desorption [4]. To exclude generation of etch pits, N₂ bubbling has been employed prior to the etching to remove dissolved O₂ in the solution.

Next, the wafer is put into a vacuum chamber, and electrons are injected from an STM tip to the surface during STM image scanning. Tunneling electrons with appropriate energy can break Si-H bonds resulting in desorption of hydrogen atoms.

3. Experimental Results

3.1 Flattening of Si Surface

Because the horizontal etching rate of Si(111) is larger than that of vertical one in the aqueous NH₄F solution, flatter surface has been obtained. Our results show that the optimum etching time is about 15 minutes. Moreover, FT-IR result indicates that the surface is terminated by hydrogen atoms successfully.

3.2 Hydrogen Desorption

Fig. 1 shows STM scan images of Si(111) before and after of hydrogen desorption. For this case, hydrogen desorption was done by STM scanning with +3.5 [V] and 4 [nA] sample bias, and all the hydrogen atoms in the scanned area were desorbed. On the other hand, surface images were obtained by STM scanning with -2. 5 [V] and 0.5 [nA] sample bias. Brighter region in the surface image (Fig. 1 (b)) corresponds to the hydrogen desorption area because dangling bonds increase electrical conductivity. Fig. 2 shows the ratio of hydrogen desorption area to the scanning area as functions of the tunneling current and the bias voltage. It can be seen that bias voltage lager than 3.1 [V] is required for desorption and desorption area increases approximately in proportion to the tunneling current. We also carried out electron bombardment without scanning. By applying a voltage pulse to the tip at the center position of an STM image and controlling the applied voltage, tunneling current, and pulse duration time, we can obtained suitable experimental parameters for minimizing the area of hydrogen desorption. The results show that a voltage pulse with 3.6 [V] amplitude and 70 [ms] duration desorbs hydrogens on about 30 sites.

4. Conclusion

We have studied about the desorption of hydrogen atoms on Si(111) surface caused by electron bombardment, and confirmed by utilizing STM scanning that the desorption area is proportional to the tunneling current. Also, we demonstrated that the number of desorption sites can be minimized by applying a voltage pulse and controlling its duration time precisely. By extending the results obtained in this research, it could be possible to develop atomic lithography technique using STM.

References

[1] B. E. Kane, “A silicon-based nuclear spin quantum computer”, Nature, 393(6681), 133-137, 1998.

[2] S. R. Schofield, N. J. Curson, M. Y. Simmons, F. J. Rueß, T. Hallam, L. Oberbeck, and R. G. Clark, “Atomically Precise Placement of Single Dopants in Si”, Physical Review Letters, 91(13), 136104, 2003.

[3] M. Fuechsle, J. A. Miwa, S. Mahapatra, et al. “A single-atom transistor”, Nature Nanotechnology, 7(4), 242-246, 2012.

[4] H. Sakaue, S. Fujiwara, S. Shingubara, T. Takahagi, “Atomic-scale defect control on hydrogen-terminated silicon surface at wafer scale,” Applied Physics Letters, 78(3), 309-311, 2001.