Simulation of Functional Interfaces

Functional interfaces are of fundamental importance in nano-electronic and photonic devices. Particularly, the interface between oxides and semiconductors is crucial for the overall device performance. Therefore, the modeling of novel semiconductor/oxide interfaces is fundamental to develop future economical, efficient and reliable microelectronic devices. In this thesis, I use molecular dynamics (MD) and density functional theory (DFT) simulations to investigate the oxide/semiconductor interface between different materials. I first investigate the Charge-Optimized Many-Body potential (COMB), an empirical potential with the aim of modeling novel interfaces with realistic features. The COMB potential has been implemented within Physic, a calculator in an object-based Python environment. COMB is found to have substantial shortcomings when describing small systems like silica and hafnia clusters, where the local bonding environment is different compared to the bulk. Other empirical potentials have been used to model silica under high pressure, highlighting the general limitations of this class of potentials. Second, I report on my DFT studies of interfaces between transition metal dichalcogenides (TMDC) monolayers and functionalized graphenes. The excellent physical and semiconducting properties of TMDCs monolayers make them promising materials for many applications, particularly electronic devices. In order to develop logic circuits based on TMDCs, it is necessary to fabricate both n- and p-type field effect transistors (FETs). While monolayer n-FETs have been widely reported, fabrication of p-FETs has been challenging. This is due to the difficulty in designing MoS2/metal contacts with low Schottky barrier heights relative to the valence band. The idea discussed in this thesis consists in inserting a functionalized graphene layer (graphene oxide or graphene fluoride) between the TMDC and the metal contact, in order to modify the work function of the TMDC. I show that in this way it is possible to obtain hole-transport based devices, opening the road for a new, more affordable design of CMOS (complementary metal-oxide-semiconductor) devices. Overall, my results highlight the importance of simulations in helping providing an understanding of experimental results. Limitations should be considered, as in the case of empirical potentials. Through DFT simulations I have conceived a new way to obtain hole-transport in MoS2-based FETs, contributing to the research in facing the continuous miniaturization of electronic devices.