< Back to previous page

Project

Impact of external factors on band alignment of TMDs with SiO2: WS2 versus WSe2 and MoS2

Over the past 50 years Si based integrated circuits have been downscaling tremendously to create faster logic chips and high-volume memory circuits. Such scaling required significant technological advances and innovative designs. Unfortunately, as transistor channel lengths have been approaching 10 nm, it became clear that such aggressive scaling would not be sustainable for Si based transistors. In looking for solutions one addresses more exotic materials namely two dimensional (2D) crystals, such as graphene and transition metal dichalcogenides (TMDs). These TMD structures are particularly auspicious as all atomic bonds are saturated within a monolayer which consists of a transition metal atoms sandwiched in between two chalcogen atom layers. As a result, in an ideal TMD crystal no dangling bonds are expected to be present at its surfaces or interface with other solids. Therefore, single or few-monolayer TMD films have been suggested as potential semiconductor channel materials enabling further transistor downscaling.

A fundamental factor of such 2D based devices that can’t be overlooked are the semiconductor/insulator band offsets, as they control the leakage currents and built-in potentials. The most straightforward way to determine these interface band offsets is by internal photoemission (IPE) of electrons and will be employed here to investigate the band offsets for synthetic MX2 TMD layers with the industry accepted insulator (SiO2) where M=Mo, W and X=S, Se. We shall focus on the WS2/SiO2 interface as it is projected as the most promising candidate for field-effect transistor (FET) applications. The main goal of this dissertation is to trace parameters possibly influencing reproducibility, stability, and variability of the MX2/SiO2 interface band alignment.

In this work, a novel technique is introduced to characterize excitonic features of the 2D films using transient photoconductivity observations. An expansion of the technique is developed, that allows for the observation of excitonic features through a semi-transparent top contact layer, allowing for 2D film monitoring upon processing. A demonstration of the film quality assessment after such processing steps or extended storage is given. Furthermore, the transient photoconductivity method can be used to evaluate the built-in potentials at the interfaces of 2D semiconductors. As interface band alignment schemes are procured in this dissertation, an accurate determination of the WS2 band gap is achieved through transient photoconductivity spectroscopy using electric field-dependent measurements.

IPE measurements are used to determine the barrier height between the WS2 valence band (VB) top edge position and the SiO2 conduction band (CB) bottom for several WS2 synthesis techniques, this in an attempt to trace parameters causing interface variability. Analysis of WS2/SiO2 interface barriers indicates that the band alignment is not only sensitive to the number of monolayers in the semiconductor film but, also, to the WS2 synthesis technique. In particular, for a certain novel low temperature deposition method it is found over various film depositions that despite similar intrinsic properties of WS2 films are achieved, the band alignment may change up to ≈ 0.5 eV, pointing to a crucial impact of the SiO2 surface or the interlayers present. This kind of "electrostatic" instabilities emerges as the major extrinsic factor affecting TMD/SiO2 band alignment. At the same time, for monolayer WS2 films the band offsets with SiO2 are found to be less affected by processing steps compared to the earlier results on MoS2. The resilience against oxidation is found to be highly dependent on the film deposition conditions.

Although, WS2 is considered as the most promising candidate to replace Si in FETs WSe2 exhibits a higher electron mobility, yet its band offsets with SiO2 remain unexplored. Compared to WS2, IPE analysis of the WSe2/SiO2 interface reveals a ≈ 0.4 eV upwards shift in energy of the VB with respect to the SiO2 CB.

Seed promoters have been found to enhance the MoS2 film quality significantly, however, the effect of introducing additional atomic species into the interface with the underlying oxide are often ignored and its effect on the VB position remains unknown. Therefore, IPE is used to study the effect of seed promoters on the VB position of MoS2 films. These reveal that films grown without seed promoter suffer from lateral non-uniformities in the band offsets with SiO2 and electrostatic potential, related to the film discontinuities. However, despite the fact that closed MoS2 films can be grown with a seed promoter, the latter is not only found to change the band offsets by up to ≈ 0.4 eV but, also, introduce lateral band alignment non-uniformities.

Date:20 Aug 2018 →  10 Jan 2023
Keywords:2D semiconductors
Disciplines:Condensed matter physics and nanophysics
Project type:PhD project