(105b) Current Modeling and Simulation Challenges in Thin-Film Deposition Processes

The mathematical modeling challenges posed by thin-film deposition processes for microelectronics device manufacturing are well-documented [4]. Simulators created from these models are important tools for understanding the reaction and transport phenomena at work in these processes and for the design, optimization, and control of the deposition reactors to reach desired material composition, uniformity, and electrical property specifications.

New materials, deposition processes, and application areas have rekindled interest in thin-film deposition process modeling. Current development activities in nanoscale photonic structures, flexible electronics, nanolaminates, moisture and barrier films, new biological probes, thin-film solar, and nanostructured electrical storage devices have pushed traditional deposition technologies to their limits and have encouraged the development of new deposition processes.

In this talk, we will focus on the modeling challenges presented by atomic layer deposition (ALD), a deposition method developed in the 1970s, but only recently applied to manufacturing gate dielectrics, diffusion barriers, and seed layers in microelectronics devices. ALD is a thin-film process in which the growth surface is exposed to an alternating sequence of gas-phase chemical precursor species. What distinguishes ALD from CVD and other deposition methods is the self-limiting nature of the precursor chemisorption during each exposure period, giving rise to an operation region where the growth per cycle (GPC) is insensitive to perturbations in precursor partial pressure and exposure time and, for some deposition systems, an "ALD window" where the self-limiting growth also is insensitive to deposition temperature. The net result is highly conformal surface coverage and atomic-level control of film thickness, with equipment-independent GPC for fixed precursor chemistry. It is this set of properties that makes ALD an important tool for fabricating the new thin-film devices described earlier.

The ALD modeling challenges stem from the widely ranging time and length scales of the surface reaction and reactor-phase transport processes, resulting in a true multiscale modeling problem. The nature of the surface reactions make direct experimental measurement of the reaction rates and discrimination of reaction mechanisms difficult, motivating substantial modeling research in quantum chemical computations [2] to unravel the elemental reaction steps. Statistical mechanics methods then can build on the ab initio results to predict the reaction rates based on transition-state chemical kinetics theory [5]. Mathematical descriptions of the amorphous growth surface will be described in this presentation, as well as the prospects for using modeling to predict whether amorphous or crystalline films will be deposited by ALD, an area of considerable current experimental attention.

Moving up the time and length scale range, ballistic transport models for precursor diffusion within nano-structured materials will be discussed [1]. Modeling work in assessing reactor-scale sources of non uniformity in ALD will be described [3]. Continuous ALD reactor operation is characterized by limit-cycle solutions, as seen in Fig. 1. Collocation formulations suitable for these applications will be described, focusing on the challenges posed by the widely varying timescales of the ALD surface reactions.

Fig. 1 Limit-cycle solution of an alumina atomic layer deposition surface reaction system illustrating the rise in -CH₃ surface species during trimethyaluminum exposure and subsequent decrease during the water exposure (top and bottom left), and a representative snapshot of the growth surface state during the processing cycle (bottom right).

References

[1] Adomaitis, R. A., A ballistic transport and surface reaction model for simulating atomic layer deposition processes in high aspect-ratio nanopores. Chem. Vapor Depos. 17 353-365 (2011).


[2] Elliott, S. D., Atomic-scale simulation of ALD chemistry.
Semicond. Sci. Technol. 27 074008 (2012).


[3] Holmqvist, A., T. Torndahl, and S. Stenstrom, A model-based methodology for the analysis and design of atomic layer deposition processes, Part I: Mechanistic modelling of continuous flow reactors.
Chem. Eng. Sci. 81 260-272 (2012).


[4] Middleman, S. and A. K. Hochberg Process Engineering Analysis in Semiconductor Device Fabrication. McGraw-Hill, Inc. (1993).


[5] Travis, C. D. and R. A. Adomaitis, Modeling ALD surface reaction and process dynamics using absolute reaction rate theory.
Chem. Vapor Depos. 19 4-14 (2013).