Font Size: a A A

Numerical Simulation Of Microstructure Evolution Of Directionally Solidified Multicrystalline Silicon Based On LBM-CA Coupling Model

Posted on:2019-05-24Degree:MasterType:Thesis
Country:ChinaCandidate:Y X ZhangFull Text:PDF
GTID:2381330623968710Subject:Engineering
Abstract/Summary:PDF Full Text Request
Multi-crystalline silicon is the mainstream semiconductor material of silicon-based photovoltaic cells.The preparation method mainly uses directional solidification crystal growth technology.Numerical simulation technology can be used to simulate the crystal growth process and explore the macro and microscopic mechanisms of crystal growth.The traditional numerical simulation of crystal growth is limited to the simulation of macroscopic physical fields.Research on the formation and development of microscopic defects relies on new macroscopic and microscopic coupling numerical simulation techniques.Facet growth is a typical behavior during crystal growth.During the growth process,the interface frontier will undergo a transition from facet to equiaxed growth due to the sudden appearance of impurity particles in columnar grains of silicon ingots.The equiaxed crystal grains are small and the grain boundaries are numerous,which has an important influence on the photoelectric conversion efficiency of the final multi-crystalline silicon photovoltaic cell.In this paper,a lattice Boltzmann-cellular automata(LBM-CA)coupled method is used to simulate the growth of facet and facet dendrites of silicon crystals and the transition from a planar faceted front to equiaxed growth(FET)of multicrystal silicon.The LBM was used to calculate the flow field,temperature field and solute field.The CA method is used to calculate the nucleation of silicon melt at the crucible interface and SiC particles,as well as growth and capture.(1)The growth of facet interface.For silicon,the interfacial kinetic coefficient is quite low,which means that its kinetics undercooling will be large.The model takes into account the high anisotropy of surface tension and interface dynamics to simulate the growth of the facet interface.Due to the release of latent heat at high speed,the silicon melt will have a negative temperature gradient at the solid-liquid interface,resulting in a zigzag facet interface.The higher the absolute value of temperature gradient,the faster its growth rate.Due to the difference of undercooling,there will be facet swallowing in front of the solid-liquid interface.(2)The growth of facet dendrite.Different from the non-faceted dendrites,the facet dendrites show stronger anisotropy,and with the increase of undercooling and interfacial energy anisotropy,the loss orientation is more and the angularity is clear.(3)The evolution of equiaxed crystals at the facet interface.A faceted front in conjunction with a sufficiently high carbon content can lead to equiaxed growth by nucleation on SiC.In the polycrystalline silicon directional solidification,when the impurity content is low,a transition from a facet crystal to a thermal equiaxed dendrite is observed;when the impurity content is high,a transition from a facet crystal to a facet equiaxed grains is observed.This model predicts the transition time and location of equiaxed dendrite.More effective control of the final multi-crystalline silicon microstructure can be achieved by changing the nucleation site density and nucleation undercooling.
Keywords/Search Tags:Multi-crystalline silicon, Numerical simulation, Facet, Lattice Boltzmann, Cellular automaton, Temperature field
PDF Full Text Request
Related items