AVS 62nd International Symposium & Exhibition
    Selective Deposition as an Enabler of Self-Alignment Focus Topic Thursday Sessions
       Session SD+AS+EM-ThM

Paper SD+AS+EM-ThM1
Fundamental Examinations of Surface Chemistry-Driven Approaches to Selective Area Atomic Layer Deposition

Thursday, October 22, 2015, 8:00 am, Room 210F

Session: Fundamentals of Selective Deposition
Presenter: James Engstrom, Cornell University
Authors: W. Zhang, Cornell University
J.-R. Chen, Cornell University
J.R. Engstrom, Cornell University
Correspondent: Click to Email
Most approaches taken to date concerning selective area growth have involved the use of masking layers consisting of photoresists or self-assembled monolayers (SAMs). While some success has been achieved with this approach there are a number of disadvantages intrinsic to these methods. First, SAMs are very difficult to form defect free, and second, and perhaps more important, these masking layers themselves must be patterned or deposited selectively. A second approach to selective area ALD relies on intrinsic reactivity differences between exposed surfaces, which, unfortunately, may be limited to a few special cases. Here we seek to develop a more general approach to achieving selective area growth. We initially focus on the first half-cycle of ALD, where demonstrating selectivity for this part of the ALD process is a necessary, but not sufficient condition for selective area growth. We are examining two specific approaches to the surface chemistry: (i) the use of adsorption reversal agents; and (ii) the use of molecular blocking agents. Here for a thin film precursor we examine transition metal complexes with the generic structure, M(XRm)n, where M is the transition metal and XRm is the coordinating ligand. In pursuing strategy (i) we examine the introduction of a second species in the first half-cycle that can act as a coordinating ligand, e.g., HXR′m, or HXR′m-1R″, etc. In pursuing strategy (ii) species (Y) are selected that can effectively compete for adsorption sites, dependent on the composition of the substrate. Concerning strategy (i) we are currently investigating reactions between transition metal amido compounds and a series of amines (X = N). We have found a somewhat unexpected result for this chemistry: the partial pressure of H2NR′ produces similar effects on both a metal (Cu) and a dielectric (SiO2) surface, where a low partial pressure attenuates the irreversible adsorption of the thin film precursor, while a high partial pressure results in no adsorption of the precursor. An investigation of the reactions between the thin film precursor and a molecular blocking agent with the structure HSR″, gave much more encouraging results. Here we found complete blockage of chemisorption of the thin film precursor on a Cu surface, whereas on SiO2 the molecular blocking agent had no effect on the adsorption of the metal for the first half-cycle. We will conclude this presentation with a discussion of the mechanisms that are likely operative in both cases, making use of in situ measurements involving X-ray photoelectron spectroscopy, and also initial results concerning the effectiveness of this approach for multiple (5-20) cycles of ALD.