Doping beyond the solid solubility limit: from bulk to nanostructured semiconductors


Doping beyond the solid solubility limit: from bulk to nanostructured semiconductors

Berencen, Y.

Hyperdoping has recently emerged as a potential powerful technique to explore new functionalities of semiconductor materials with unique electrical and optical properties [1-3]. Hyperdoping facilitates to introduce dopants into a semiconductor material at concentrations far above those obtained at equilibrium conditions, viz. doping far beyond the solubility limit. Hyperdoped Si with chalcogens or transition metals like Au or Ti has been postulated to be a promising material for many applications, especially for Si-based infrared photodetectors and intermediate band solar cells ([3] and references therein).

Most of the relevant published papers [1-3] have been limited to the study of hyperdoped bulk Si with chalcogens. Particularly, S and Se have primarily been used for this kind of purpose since they can introduce a variety of deep donor states in the upper half of the band gap of Si, which give rise to the formation of an impurity band that allows a strong sub-band gap optical absorption at wavelengths larger than 1 µm [2]. To create such an impurity band in a semiconductor like Si, whose equilibrium solubility limit for the aforementioned dopants is around 1016 cm-3 [1], high dopant concentrations exceeding the so-called Mott limit are required [1]. Therefore, non-equilibrium thermal processing to obtain hyperdoped semiconductor materials is sound.

Chalcogens are commonly introduced by pulsed-laser irradiation of the material under study, which is simultaneously immersed in an atmosphere containing chalcogen atoms [2]. Another singular approach is to use ion implantation followed by either nanosecond (ns)-range pulsed-laser melting (PLM) [1] or millisecond (ms)-range flash lamp annealing (FLA) [3]. Both PLM and FLA are advantageous if compared with conventional annealing techniques. For instance, these techniques offer high crystal quality after recrystallization, high dopant solubility and low heating of the substrate [3]. Moreover, depending on the energy density and the timescale of annealing after implantation (typically, from few ns to dozens of ms), liquid-phase or solid-phase epitaxy can be induced on the implanted material surface, which accounts for the epitaxial recrystallization atop the crystalline substrate [3]. In particular, solid-phase epitaxy via ms-range FLA was reported to be superior to liquid-phase epitaxy through ns-range PLM in terms of less dopant redistribution and high electrical activation of dopants [3].

In the last years, semiconductor nanowires (NWs) have gained increasing importance as building blocks for nanodevices like field-effect transistors, light emitting devices, photovoltaic cells and integrated photodetectors ([4] and references therein). Indeed, their reduced sizes and the physical confinement along two directions allow the tuning of electro-optical properties such as conductivity, optical absorption and photoluminescence, among others. Recently, traditional dopants such as B and P have been introduced into Si NWs by ion implantation followed by conventional annealing [5]. Recrystallization and reactivation of dopants were successfully achieved at low doses. However, both phenomena were not successful at high doses since it resulted in the formation of polycrystals and a low ratio of dopant activation. To date, there are only theoretical examinations of the properties of Si NWs hyperdoped with chalcogens [4]. These results have shown that donor defects give rise to a strong hyperfine interaction in the hyperdoped Si NWs, which can be exploited to develop a Si-based nuclear spin quantum computer [4].

Therefore, the new approach to combine Si NWs and hyperdoping with chalcogens proposed here can provide important contributions to the state-of-the-art and will help to answer important fundamental and practical scientific issues.

Reference
[1] Ertekin E., Winkler M. T., Recht D., Said A. J., Aziz M. J., Buonassisi T., Grossman J. C. (2012): Insulator-to-Metal Transition in Selenium-Hyperdoped Silicon: Observation and Origin. Published in: Physical Review Letters 108, 026401.
[2] Sher M. J., Mazur E. (2014): Intermediate band conduction in femtosecond-laser hyperdoped silicon. Published in: Applied Physics Letters 105, 032103.
[3] Zhou S., Liu F., Prucnal S., Gao K., Khalid M., Baehtz C., Posselt M., Skorupa W., Helm M. (2015): Hyperdoping silicon with selenium: solid vs. liquid phase epitaxy. Published in: Scientific Reports 5, 8329.
[4] Petretto G., Massé A., Fanciulli M., Debernardi A. (2015): Analysis of hyperfine structure in chalcogen-doped silicon and germanium nanowires. Published in: Physical Review B 91, 125430.
[5] Fukata N., Takiguchi R., Ishida S., Yokono S., Hishita S., Murakami K., (2012): Recrystallization and reactivation of dopant atoms in ion-implanted silicon nanowires. Published in: ACS NANO 6, 3278.

Keywords: Chalcogens; silicon; hyperdoping; ion implantation; flash lamp annealing; insulator to metal transition; sub-band gap photoresponse; infrared plasmonics; nanowires

Related publications

  • Lecture (others)
    Leibniz institute seminar, 22.02.2018, Leibniz Institute of Polymer Research Dresden, Germany

Permalink: https://www.hzdr.de/publications/Publ-27187
Publ.-Id: 27187