Floating Silicon Method single crystal ribbon – observations and proposed limit cycle theory

Abstract

In the Floating Silicon Method (FSM), a single-crystal Si ribbon is grown while floating on the surface of a Si melt. In this paper, we describe the phenomenology of FSM and the observation of approximately regularly spaced "facet lines" on the ribbon surface whose orientation aligns with (111) crystal planes.Sb demarcation experiments sectioned through the thickness of the ribbon reveal that the solid/melt interface consists of dual (111) planes and that the leading edge facet growth is saccadic in nature, rather than steady-state.To explain this behavior, we propose a heuristic solidification limit cycle theory, using a continuum level of description with anisotropic kinetics as developed by others, and generalizing the interface kinetics to include a roughening transition as well as a re-faceting mechanism that involves curvature and the Gibbs-Thomson effect.

Experimental and numerical investigation of the horizontal ribbon growth process

“…Thus, the model predicts a growth and spread picture of the solidification; 2D Nucleation occurs at the triple junction creating steps that then propagate down the facet, slowing at the elbow such that the misalignment angle increases to the point that roughened growth takes over and the interface continuously bends towards horizontal. Our more recent experimental data [25] indicate that this is an oversimplified view of the process especially very near the triple junction (point A), however as we show below, this kinetic model does well-predict the response of the system to changes in pull-speed.…”

“…These observations led us to question the wedge model of growth and postulate that a faceted growth process was more likely. Further evidence confirming this fact is given in a companion paper that reports more of the experimentally observed details [25]. Previous authors have discussed faceting effects for silicon grown on free-surfaces [12,26], but they did not relate this to the limits of pull speed for a horizontal ribbon growth process.…”

NeoGrowth silicon: A new high purity, low‐oxygen crystal growth technique for photovoltaic substrates

“…In the last 10 years, other new liquid crystallization options have arisen, undaunted by the market dominance of Czochralski (Cz) and multicrystalline (mc) ingot growth. Cast monocrystalline, Direct Wafer, low‐angle sheet silicon, and mushroom‐shaped crucible growth have each demonstrated promise in quality and/or silicon utilization, although none has broken through to large‐scale production. Meanwhile, the 2 big contenders, Czochralski pulling and multicrystalline casting, have seen tremendous amounts of continuous improvement and have grown to an enormous scale without a clear winner emerging .…”

“…However, even if this can be demonstrated, the growth of a new technology to gigawatt scale and beyond also demands a low capital cost and a backer with the wherewithal to take the risk on the first large factory operation. Promising results are consistently presented by technologies looking to cut out significant and expensive parts of the process while maintaining high quality, such as the epitaxial methods and low‐angle sheet silicon . These would be disruptive entrants into the industry, and each have some remaining challenges.…”

Modeling and Analysis of Novel Horizontal Ribbon Growth of Silicon Crystal

“…In general, growth techniques enable realization of continuous growth via continuous extraction of silicon ribbon. To date, a variety of different growth techniques have been fabricated as based on this concept [1][2][3][4], e.g., the Horizontal supported web (HSW) [5], Ribbon growth on substrate (RGS) [6], Horizontal ribbon growth (HRG) [7,8], and Floating silicon method (FSM) [9,10]; all of these methods have provided excellent ideas and references for the method we have developed and described in this paper.…”