PCSI2024 Session PCSI-MoM1: Semiconductor Heterostructures (Growth, Nanostructures & Interfaces) I

Remote epitaxy on monolayer graphene is promising for synthesis of highly lattice mismatched materials, exfoliation of free-standing membranes, and re-use of expensive substrates. However, due to contaminants at the transferred graphene/substrate interface, other mechanisms such as pinhole-seeded lateral epitaxy often dominate rather than the intrinsic growth via remote interactions [1]. I will describe our understanding of the synthesis science of remote epitaxy, focusing on III-V semiconductors and Heusler compounds [1,2]. I will also show how exfoliated free-standing membranes of rare earth Heusler compounds can be used to tune (flexo)magnetism and novel superconductivity [3,4]

[1] S. Manzo, et. al.,Nature Commun., 13, 4014 (2022). https://doi.org/10.1038/s41467-022-31610-y

[2] D. Du et. al., Nano Lett. 22, 21, 8647 (2022). https://doi.org/10.1021/acs.nanolett.2c03187

[3] D. Du, et. al., Nature Commun., 12, 2494 (2021). https://doi.org/10.1038/s41467-021-22784-y

[4] D. Du, e. al. APL, 122, 170501 (2023). https://doi.org/10.1063/5.0146553

Transistor applications of semiconducting oxides require, both high room-temperature electron mobilities (µ_RT) and high charge carrier densities (CCDs), ideally realized with a two-dimensional electron gas (2DEG). So far, prototype oxide 2DEG systems have either high µ_RT but limited CCD such as modulation-doped (Al_xGa_1-x)₂O₃/Ga₂O₃, or a high CCD but low µ_RT such as the polar-discontinuity doped LaAlO₃/SrTiO₃ interface. Interfacing the more suitable, wide-bandgap, nonpolar semiconductor BaSnO₃ (BSO), having high bulk µ_RT (up to 320 cm²/Vs), with polar LaInO₃ (LIO) is predicted to create and confine a 2DEG with CCD up to 2x10¹⁴ cm^-2 for the SnO₂/LaO interface termination.

We demonstrate the adsorption-controlled growth of the LIO[1] on BSO heterostructure by molecular beam epitaxy using a shutter sequence to control the SnO₂/LaO interface termination. The films were analyzed by reflection high-energy electron diffraction (RHEED) [Figs. 1(a) and 1(b)], x-ray diffraction, atomic force microscopy (AFM) [Fig. 1(c)]. The interface structure is investigated by cross-sectional transmission-electron microscopy. The formation of the quantized 2DEG at their interface is confirmed by capacitance-voltage (CV) [Fig. 1(d)] and angular-resolved photo-electron spectroscopy (ARPES) [Fig. 1(e)]. Van der Pauw-Hall measurements confirm CCD>10¹³cm^-2 and µ_RT>100cm²/Vs.

[1] G. Hoffmann et al., Phys. Rev. Mater. 7, 084606 (2023).

Low-dimensional nanomaterials, such as one-dimensional (1D) or two-dimensional (2D) systems, assembled in vertical or lateral arrangements, often lead to enhanced properties and new functionalities. Nanotubes, nanowires (NWs), and 2D layered structures (graphene and graphene like materials) are emerging as key building blocks for the next generation devices and emerging technologies. Practical implementation of such nanomaterials necessitates their successful incorporation with well-established processes for fabricating electronic and/or mechanical devices. While preparing layered architectures usually involves multi-step fabrication processes but relies on mask-assisted fabrication techniques. Here, we present a methodology for the controlled and selective preparation of nanostructures such as 1D NWs on 2D materials-substrates in various controlled geometric assemblies by employing direct write patterning (DWP) of custom ink precursors on supported or suspended architectures for subsequent chemical vapor deposition (CVD) synthesis.Our two-step fabrication approach enables simple and flexible routes to produce various architectures in a precisely controlled fashion. Location-specific materials synthesis provides access to as-grown interfaces and rapid testing of materials’ quality, crystallinity, chemical composition, etc.

Thin single crystal films of Al can be epitaxially grown on Si(111) [1]. When taken out of ultra-high vacuum, Al oxidizes and forms a quantum well structure Si-Al-Al₂O₃. Theoretical calculations can shed light on finding the band alignment across the hetero-structure and its electronic and optical properties. However, one needs to build a structural model. The structure of Al on Si(111) has been experimentally determined via the electron microscopy by McSkimming et al. [1]. The Al layer is found to be (111) oriented, with a linear ratio of 3 Si to 4 Al atoms along the interface. Using this information, we construct a periodic model of Si-Al interface and perform density functional theory (DFT) calculations to find the relaxed structure, layer-projected density of states, and planar averaged potential across the interface. From there, we find the band alignment using the valence band offset from the bulk averaged potential. When aluminum metal is exposed to air, a several nanometers thick layer of alumina rapidly forms [2]. This oxide layer is amorphous. We simulate the structure of amorphous alumina using ab-initio molecular dynamics with the melt-and-quench technique [3]. This amorphous alumina can then be directly combined with the previous system to form an asymmetrical quantum well, where the well states and band alignment can be directly obtained from the DFT calculations.

[1] McSkimming, B. M., Alexander, A., Samuels, M. H., Arey, B., Arslan, I., & Richardson, C. J. (2016). Metamorphic growth of relaxed single crystalline aluminum on silicon (111). J. Vac. Sci. & Technol. A 35, 021401 (2017).

[2] Jeurgens, L. P., Sloof, W. G., Tichelaar, F. D., & Mittemeijer, E. J. Growth Kinetics and mechanisms of aluminum-oxide films formed by thermal oxidation of aluminum. Journal of Applied Physics, 92, 1649 (2002).

[3] Gutiérrez, G., asnd Johansson, B. Molecular dynamics study of structural properties of amorphous Al₂O₃. Physical Review B 65, 104202(2002).

Silicene is a 2D material which poses significant challenges in synthesis and device integration but offers unique electronic properties which are not yet fully understood. Challenges remain in silicene synthesis on non-metallic substrates and the control of buckling. The degree of in-plane buckling is tied to the emergence of a Dirac point and intriguing quantum phases have been predicted.[1-4]In our previous work we discovered a new pathway to silicene synthesis using h-MoSi₂ (0001) surfaces as templates,[5]and we present here the formation of silicene nanoribbons on the same surface, and measure their geometric and electronic structure with STM and STS at 77 K. Our work promises a new pathway to create silicene layers and nanoribbons with molecular beam epitaxy directly on silicon wafers.

The h-MoSi₂ (0001) crystallites are synthesized on Si(100) by deposition of a thin Mo film with electron beam evaporation, and subsequent annealing to form well-defined crystallites with 15 - 120 nm in diameter. Nanoribbons cover the entire surface for all “flat top” silicide crystallites with (0001) surfaces. Several surface reconstructions present intermediate superstructures, and are precursors for the nanoribbons. The STM images in Fig. 2 illustrate the geometric structure of the nanoribbons which are 1.8 nm in width and separated by a distinct groove of roughly a single atomic row in width. Models of the nanoribbon’s geometric structure will be discussed. The atoms located at the ribbon edge appear brighter which is likely due to density of states (DOS) modulation by edge states. Characteristic defects are seen in the ribbons and express identical STM. The average DOS (dI/dV) measured with STS is indicative of a Dirac type electronic signature with a V-type dip around E_F, and the position dependent DOS across the ribbons allows to identify the Dirac type regions. We will discuss all aspects of electronic and geometric structure of the nanoribbons including electronic confinement, and localized Dirac type signatures. In addition, point defects, and long range deformations induced by strain are addressed and inform our structural models.

[1] C. Grazianetti et al.. ACS Nano 11, 3376-3382 (2017). [2] C.-C. Liu et al.. Physical Review Letters 107, 076802 (2011). [3] A. Molle et al. Chemical Society Reviews 47, 6370-6387 (2018). [4] H. J. W. Zandvliet. Nano Today 9, 691-694 (2014). [5] C. Volders et al. Nano Letters 17, 299-307 (2017).