Theses Doctoral

Pulsed-Laser-Induced Melting and Solidification of Thin Metallic Films

Choi, Min Hwan

This thesis focused on investigating excimer-laser induced melting and solidification of thin metallic films on SiO2. Two distinct topics were pursued: (1) microstructural manipulation and optimization of Cu films via SLS of as-deposited Cu films on SiO2, and (2) investigation of oriented heterogeneous nucleation via complete melting and subsequent nucleation-initiated solidification of Ni films on SiO2. The work on SLS of Cu films is motivated in large part by the need to improve the properties of Cu films which, among other applications, constitute an essential element in the continued evolution of microelectronic products. The experiments we have conducted show clearly that the film can be, without much difficulty, melted and solidified using pulsed-laser irradiation.

Based on the findings from a series of systematic single-shot experiments, we show that SLS can be properly implemented to obtain large-grained Cu films with controlled microstructures and restricted textures. The lateral growth distance was found to increase as a function of increasing incident energy density. This observation is consistent with the findings that were made previously using other materials, and basically indicates that lateral solidification continues until the interface is halted by the interfaces growing from nucleated solids, which are triggered within the liquid matrix ahead of the growing interface. Close examination of the laterally grown grains, which quickly develop 100 rolling direction crystallographic orientation texture due to occlusion of differently oriented grains, reveal, furthermore, that low-angle grain boundaries as well as twins can be generated during the growth.

These intra-grain defects are found to appear in a systematic manner (as they are located at specific regions and observed under specific incident energy densities). Significantly longer lateral growth distances observed in Cu films (compared to that of Si films) was attributed to the fact that substantially higher growth rates are expected with simple metallic films at a given interfacial undercooling. The implementation of SLS using Cu films was accomplished quite effectively, as can be expected from the above lateral-growth-related results involving single-shot experiments. We were able to achieve a variety of large-grained, grain-boundary location and grain-orientation controlled Cu films via various SLS techniques.

When performed optimally in accordance with the findings made in chapter 2, the resulting microstructure exhibits large grains, which are primarily devoid of intra-grain defects. For example, 2-shot SLS processed Cu films led to strong 100 rolling direction orientation, while avoiding the formation of low-angle grain boundaries and twin-boundaries. The highlight of SLS on Cu films correspond to a version of SLS (referred to as "2-Shot plus 2-Shot" SLS) in which the second 2-shot SLS is performed in the direction perpendicular to the first one. Through this approach, we were able to achieve grain-boundary-location controlled microstructure with a strong 100 orientation texture in all three dimensions (forming, effectively, an ultra-large quasi-single crystal material). Nucleation of solids in laser-quenched Ni films was investigated using EBSD analysis.

The surface orientation analysis of nucleated grains obtained within the complete melting regime reveal a clear sign of texture. From these and additional findings from previous work involving Al films, we were able to conclude that systematic heterogeneous nucleation has taken place, and, furthermore, that oriented nucleation of the solids must have taken place. Although always suspected to be the case, it is typically extremely challenging to prove with certainty, in conventional nucleation experiments, that the mechanism of nucleation corresponds to that of a heterogeneous one. Furthermore, although it has been suspected theoretically for over 50 years, experimental results that clearly show that oriented nucleation actually transpires have not been obtained until our work involving Al films; the present findings involving Ni films further strengthen this conclusion as the Ni system removes some of the experimental uncertainties that are associated with Al films, and, furthermore, suggests that the process of oriented nucleation is a general and rather pervasive phenomenon.

Additionally, it was observed that the selected orientation changed as a function of incident energy density; in the low energy density regime (above the completed melting threshold) {110}-surface texture was observed, while {111}-surface texture became more dominent at higher densities. Motivated by our experimental work involving Al and Ni that clearly indicates that oriented heterogeneous nucleation is a major path through which heterogeneous nucleation of solids occurs, we have also carried out a 2-dimensional Winterbottom-type thermodynamic analysis that can be used to obtain a better understanding of the phenomenon. In contrast to the previous work on the subject, we consider in our modelling the anisotropic nature of both the solid-liquid and solid-substrate interfacial energy; we advocate that this is the only physically consistent combination. The results show that oriented nucleation can be systematically accounted for as stemming from the expected anisotropic nature of the involved interfacial energies. Furthermore, the analysis also suggests possible reasons for observing a transition in surface texture from one orientation to another.

Files

  • thumnail for Choi_columbia_0054D_11049.pdf Choi_columbia_0054D_11049.pdf application/pdf 35.8 MB Download File

More About This Work

Academic Units
Applied Physics and Applied Mathematics
Thesis Advisors
Im, James Sungbin
Degree
Ph.D., Columbia University
Published Here
October 17, 2012