Effects of a traveling magnetic field on vertical gradient freeze growth of cadmium zinc telluride Andrew Yeckel a and Jeffrey J. Derby a a Department of Chemical Engineering and Materials Science, University of Minnesota, 421 Washington Avenue SE, Minneapolis, MN 55455-0132, USA ABSTRACT The effects of a traveling magnetic field (TMF) on vertical gradient freeze (VGF) growth of cadmium zinc telluride (CZT) are studied using a coupled model of magnetic induction, fluid dynamics, and heat transfer. Simulations are performed to determine the influences of current and frequency on melt flow and growth interface shape. A downward traveling electromagnetic wave drives flow downward at the wall, which tends to flatten the interface, whereas an upward traveling wave has the opposite effect. TMF makes a significant impact on interface shape in the absence of thermal buoyancy, but is ineffectual under realistic conditions in a 4 inch diameter ampoule, for which buoyancy dominates Lorentz force throughout the melt. Keywords: Crystal growth, traveling magnetic field, vertical gradient freeze, cadmium zinc telluride 1. INTRODUCTION Cadmium zinc telluride (CZT) is one of a few semiconductor crystals from which portable, room-temperature gamma radiation detectors have been successfully built. 1 Affordable material of sufficient crystallographic perfec- tion remains unavailable, however, particularly for large field-of-view imaging and high-resolution spectroscopic analysis. 2 The difficulties in growth of single crystal CZT have been well documented, 3–5 not all of which have been overcome yet. In this work we focus on the feasibility of manipulating the shape of the crystal-melt interface during growth by applying a traveling magnetic field (TMF) 6–9 to the vertical gradient freeze (VGF) process widely used in growing CZT. A key issue facing production of large-area CZT is defect formation caused by thermal stresses and wall interactions. Thermal stresses are smallest when the interface is flat, but a slightly convex interface is preferred under the notion that defects generated from deleterious wall interactions grow outward to terminate at the wall. Our previous simulations of VGF 10–12 and the closely related vertical Bridgman (VB) system 13–17 show that the growth interface maintains an undesirable concave shape due to three factors: the jump in thermal conductivity between crystal and melt, the release of latent heat, and the heat transfer characteristics typical of crystal growth furnaces. Here we consider whether TMF can be used to flatten or invert the interface by modifying convection in the melt, when a typical VGF temperature profile is used. Electromagnetic stirring in melt crystal growth has drawn interest since the 1950’s, but the first attempts at theoretical modeling of TMF appear to date from the late 1990’s, 18 starting with Ono and Trapaga, 19 who modeled time-dependent magnetic induction in Czochralski growth. Since then several theoretical studies have been published on TMF in VGF or VB systems. 6–9 Recent modeling work related to the KRISTMAG project at IKZ 20, 21 is the most detailed and integrated to date, though a decoupled approach has been used to solve the induction problem separately from the transport model. Here we develop a fully coupled model that incorporates rigorous and accurate treatments of both induction and transport physics. Further author information: (Send correspondence to A.Y.) A.Y.: E-mail: yecke003@umn.edu J.J.D.: E-mail: derby@umn.edu Submitted to the proceedings of Hard X-Ray, Gamma-Ray, and Neutron Detector Physics XIII, SPIE Optical Engineering + Applications, 21-25 August 2011,San Diego, CA.