Fixed and deforming grid solutions of solidification in an undercooled melt: A benchmark problem

M. Liang, V. R. Voller

Research output: Contribution to journalArticlepeer-review

2 Scopus citations


A problem involving the solidification of a solid seed in an undercooled melt is posed. A sharp interface between solid and liquid is assumed, and the equilibrium solidification temperature at this interface is controlled by both curvature and kinetic undercooling. Numerical solutions based on a fixed grid and on a deforming grid are developed. In the limits of a vanishing surface energy and large molecular mobility, these solutions are verified by comparing with known analytical solutions. In general applications, with curvature and kinetic undercooling present, predictions from the fixed and deforming grid solutions are in close agreement. The results obtained highlight the role of the surface curvature and kinetic undercooling in controlling solidification speed and acceleration. In addition, both the fixed and deforming solution methods provide verification tools for more general 2-D and 3-D crystal growth simulations.

Original languageEnglish (US)
Pages (from-to)1-17
Number of pages17
JournalNumerical Heat Transfer, Part B: Fundamentals
Issue number1
StatePublished - Jul 2011

Bibliographical note

Funding Information:
Received 23 August 2010; accepted 25 March 2011. Although this work is focused on the thermal modeling of crystal growth, a significant motivation was to explore how undercooling thermal models could be adapted to model the formation of channels in sedimentary deltas, ongoing work supported by the STC program of the National Science Foundation via the National Center for Earth-Surface Dynamics under agreement EAR-0120914. Address correspondence to V. R. Voller, Department of Civil Engineering, Saint Anthony Falls Laboratory, University of Minnesota, Minneapolis, MN 55455-0116, USA. E-mail:


Dive into the research topics of 'Fixed and deforming grid solutions of solidification in an undercooled melt: A benchmark problem'. Together they form a unique fingerprint.

Cite this