Density functional theory and beyond—opportunities for quantum methods in materials modeling semiconductor technology

Abstract

In the semiconductor industry, the use of new materials has been increasing with the advent of nanotechnology. As critical dimensions decrease, and the number of materials increases, the interactions between heterogeneous materials themselves and processing increase in complexity. Traditionally, applications of ab initio techniques are confined to electronic structure and band gap calculations of bulk materials, which are then used in coarse-grained models such as mesoscopic and continuum models. Density functional theory is the most widely used ab initio technique that was successfully extended to several applications. This paper illustrates applications of density functional theory to semiconductor processes and proposes further opportunities for use of such techniques in process development.

First-principles simulations of copper diffusion in tantalum and tantalum nitride

“…However, these materials may also be used as diffusion barriers under special conditions. 21 For each of them, Cu/barrier material interfaces are constructed by depositing Cu thin layers on top of the barrier substrates. The detailed interfacial structures will be discussed in Sec.…”

“…It is clear however from both experimental and theoretical results that Cu forms stable pseudomorphic films on top of Ta, i.e., films which have the same structure as the Ta substrate. 21,23 Since 3D simulations of Cu islands or droplets on Ta based on first-principles are not feasible computationally, we focus on the pseudomorphic thin-film mode in this study. This is a reasonable approximation because the interfacial diffusions are highly localized, taking place near the interface.…”

Modeling Mechanism and Growth Reactions for New Nanofabrication Processes by Atomic Layer Deposition

“…A similar model involving two ligand elimination reactions for each half reaction is presented by Shankar et al to predict the reaction kinetics and the kinetics are fed into a continuum-based ALD reactor simulation that also includes fl uid fl ow, mass and heat transport, so as to predict growth rate as a function of reactor conditions. [ 16 ] Heyman and Musgrave also explore the AlCl 3 +H 2 O half reactions for alumina ALD with DFT, comparing the computed reaction mechanisms to those obtained for the TMA+water process. [ 17 ] Both half reactions are predicted to have similar reaction kinetics at T = 0 K involving only a single transition state for each discrete reaction step.…”

Dynamic Modeling for the Design and Cyclic Operation of an Atomic Layer Deposition (ALD) Reactor

“…Additionally, see [33] for a similar approach and a more general discussion of ALD reactor modeling. Furthermore, [34] describe a dynamic reactor model for alumina ALD processing that combines precursor flow modeling through the reactor and a ballistic flux model for precursor transport and reaction in substrate trenches.…”