Quantum confinement in semiconductor random alloys: a case study on Si/SiGe/Si

Daniel Dick, Florian Fuchs, Jörg Schuster, Sibylle Gemming · 2026

Read Quantum confinement in semiconductor random alloys: a case study on Si/SiGe/Si by Daniel Dick, Florian Fuchs, Jörg Schuster, Sibylle Gemming (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-18747.txt.

Introduction

Dick, Fuchs, Gemming, and Schuster study quantum confinement in silicon-germanium random alloys. Local swings in germanium content matter once layers shrink to nanoscale thickness. Extended Hückel theory models a SiGe layer sandwiched between silicon layers. The team varies composition, layer thickness, and local germanium fluctuation. Outputs include band alignment and band gap shifts. A finite quantum well model is checked against the atomistic results. That simple model captures the core physics and runs much faster, making it a practical stand-in for device work.

Catalog record

CREATORDaniel Dick, Florian Fuchs, Jörg Schuster, Sibylle Gemming
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE33885
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.18747

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Quantum confinement in semiconductor random alloys: a case study on Si/SiGe/Si Daniel Dick, Florian Fuchs, Jörg Schuster, Sibylle Gemming Abstract Local composition fluctuations in random alloys become crucial when one or more dimensions are reduced to the nanoscale. Using extended Hückel theory, we study the semiconductor random alloy SiGe sandwiched between Si due to its relevance for transistor devices. We evaluate the effects of the alloy composition, layer thickness, and local fluctuations of the Ge concentration on the band alignment and the band gap. The results are compared with the finite quantum well model. That model captures the essential physics and can act as a computationally faster alternative. I Introduction Semiconductor alloys allow tuning electronic properties, such as the band gap, by changing the alloy composition. This adjustability makes them attractive materials for a wide range of applications. Thin layers of SiGe, for example, are of interest in semiconductor devices such as heterojunction bipolar transistors 25 ; 24 ; 28 , field-effect transistors 33 ; 4 ; 10 ; 18 and quantum cascade lasers 39 . With the ongoing downscaling of such devices, quantum confinement increases the band gap. Additionally, local fluctuations in alloy composition have an increasing impact as the layer thickness is reduced. Studying that effect therefore requires a method by which fluctuations on the atomic scale can be considered. Quantum confinement in Si-based nanowires and nanoslabs has been studied previously 8 ; 15 ; 34 ; 17 , but such nanostructures are often simulated in vacuum, without the focus on layer stacks, and the effect of fluctuations in the atomic structure has not been intensively investigated yet 14 ; 23 . In the present study, we address the topics of quantum confinement in ultra-thin SiGe layers in-between Si and the role of local stoichiometry fluctuations by simulating different atomic structures and layer thickness values using extended Hückel theory (EHT). The paper is structured as follows: Following this overview, Section II introduces the methodology of this study and the simulated atomic structure models. Afterwards, in Section III , we showcase the calculated band gaps of nm-thin SiGe layers, sandwiched between Si, in dependence on the layer thickness and the alloy composition. We consider the topic of band alignment and give a parameterized description of the conduction band and valence band edges with regard to the Ge concentration. Next, we study the effect of quantum confinement and compare EHT results with the models of the finite and the infinite potential well. Finally, we quantify the effect of local fluctuations of the Ge content. II Methodology In view of the large number of atoms to be considered in few-nm-thick layers, the computationally efficient extended Hückel theory (EHT) 12 ; 3 ; 32 was chosen for the present study. We employ a parameterization of Si and Ge that was shown to be suited for describing the properties of SiGe alloys by comparison with self-consistent density functional theory 5 . The vacuum level of Ge was shifted by -0.55 eV compared to the cited parameterization to more accurately describe the band offset (see section III.1 for further discussion). The calculations are done using Synopsys QuantumATK X-2025.06 30 ; 32 . Figure 1: The supercell of a 3 nm SiGe layer used for the structural relaxation, with randomly distributed Si and Ge atoms and an average Ge concentration of 30% (top). It is divided into 36 sub-cells (bottom) of appropriate size for EHT calculations. The simulation cell is assembled in the following way: First, a single layer is defined by a 12\times 12\times 1 grid of conventional unit cells (roughly 6.5 nm \times 6.5 nm \times 0.5 nm). Each atom is randomly selected as either Si or Ge, depending on the chosen alloy composition. Multiples of such layers are then stacked on top of each other in r_{z} -direction until the chosen layer thickness t is reached. Then, this cell is sandwiched between two Si layers of the same cross-sectional area and a thickness of 8 conventional unit cells (roughly 6.5 nm \times 6.5 nm \times 4.3 nm) each, acting as a separator between periodic images of the SiGe layers in the r_{z} -direction (see Figure 1 ). For a comparison to bulk alloys, additional simulations without the Si layers are performed as well. The supercell with more than 20,000 atoms is relaxed using a parameterization of the Stillinger-Weber potential 31 by Laradji et al. 21 . The lattice constants in the in-plane directions, r_{x} and r_{y} , are fixed to the lattice constant of pristine Si, resembling a SiGe layer inside a pure Si matrix. Only the lattice constant in the confined out-of-plane direction changes during the relaxation step. The lattice constant, which is not further discussed in this study, closely follows Vegard’s law, in agreement with recent DFT simulations 9 ; 36 . The relaxed supercell is then divided into 36 cells of \sim 500 - 1000 atoms (depending on layer width, see Figure 1 ). Because of the random distribution of Si and Ge atoms in the supercell, each of those sub-cells has a slightly different alloy composition. This allows to quantify the effect of local stoichiometry fluctuations. There is a small difference between the mean Ge concentration of all cells \langle x\rangle and the chosen target Ge content x , most notably for the thinnest layers studied in this work. This leads to an uncertainty of \sim 2 meV of the mean band gap, which is within the accuracy of our method. Appendix E describes this in more detail. Figure 2: The band alignment of the structure depicted on the bottom of Figure 1 and its atom-projected density of states (dos). The band gap of the layer, E_{\text{gap}}(x,t) , is larger than the corresponding band gap of the bulk material, E_{\text{gap}}(x,\text{bulk}) , due to quantum confinement. As indicated in Figure 2 , the band gap of each structure is calculated as the smallest difference between valence band and conduction band edge. As shown by the local density of states, the band alignment between Si and SiGe leads to a quantum well for both electrons and holes. III Results and discussion Figure 3 a) shows how the band gap of SiGe alloys changes with the Ge content and the layer thickness. We focus on the technologically relevant alloy compositions of up to 30% Ge and average the band gap values of 36 atomic structures each. Two trends can be observed. First, the band gap decreases with higher Ge content. This finding agrees well with prior research on bulk SiGe alloys on Si substrate 9 ; 5 ; 19 ; 20 ; 2 ; 38 . Secondly, the band gap increases with decreasing layer thickness due to quantum confinement. This counteracts to the effect of Ge content and needs to be considered in smaller devices. For example, as indicated by the contour lines in the figure, a 3 nm thin layer with 30% Ge has approximately the same band gap as a bulk alloy with a Ge content of 20%. Figures 3 b) and 3 c) showcase the position of the conduction band edge ( E_{\text{c}} ) and valence band edge ( E_{\text{v}} ). The conduction band offset is significantly smaller,

Read the complete text