Kinetics of Scrap Melting in Liquid Steel: Multipiece Scrap Melting JIANGHUA LI and NIKOLAS PROVATAS This article extends the study of single-steel bar melting discussed in a previous article [1] to the investigation of two-bar and multibar melting kinetics. Experiments involving multiple bars reveal that the interbar spacing and the initial solid and liquid steel temperatures influence the final melting time by altering the degree of ‘‘steel iceberg’’ formation. Simulations of scrap melting using a recently developed phase-field model of steel scrap melting [1] are shown to follow the trends of the two-bar melting experiments. The phase-field methodology is also extended to examine melting of randomly distributed scrap in the liquid steel bath, a poorly understood situation that is difficult to access experimentally. Two types of simulations were performed. The first type assumed a constant heat-transfer coefficient and liquid steel temperature, corre- sponding to the limiting case of melting with perfect stirring in the liquid steel bath. Results for this case reveal that the final melting time was controlled by the largest of a group of isolated steel icebergs, which formed in regions of low scrap porosity. The second type examined the case of melting dominated by heat conduction, using an effective thermal conductivity to model low- level natural convection in the liquid steel. In this case, phase-field simulations show that, under certain conditions, melting could be well approximated by a simple one-dimensional (1-D) analytical melting model with effective parameters related to the scrap distribution, scrap pre- heating, and liquid bath temperatures. DOI: 10.1007/s11663-007-9102-x Ó The Minerals, Metals & Materials Society and ASM International 2008 I. INTRODUCTION IN recent years, industrial melting of steel scrap in a liquid steel bath in an electric arc furnace (EAF) has become increasingly important as large flat baths become common. A ‘‘hot heel’’ operation utilizes the molten steel left in the bottom of the furnace from the prior heat to assist in the melting of fresh scrap entering the EAF. Some new processes such as CONSTEEL and ECOARC employ a permanent liquid bath to melt steel scrap. [2] Scrap melting in a liquid steel bath is a process involving multiple scrap pieces. The interaction between scrap pieces plays an important role in establishing the final scrap melting time. In particular, the formation of solidified shells around original scrap pieces, and agglom- eration of these shells (termed as ‘‘steel icebergs’’), may dominate the melting process. The present work was undertaken with two aims. The first was to conduct a comprehensive experimental study on multipiece steel scrap melting kinetics. Following the single-bar melting experiments introduced in a previous article, [1] more complex melting experiments involving two and more bars were explored to investigate how parameters such as spacing between bars and initial steel sample temperature affect steel iceberg formation and the melting time. The details of this experimental investigation are presented in Section II. The second aim of this work was to extend a phenomenological phase-field model developed in a previous article, [1] which was shown to successfully capture the kinetics of single-bar melting, to the case of two-bar and randomly distributed steel scrap melting. After comparing model simulations against the two-bar melting experiments, two types of simulations involving multipiece randomly distributed scrap were performed. The first of these corresponded to the limiting case of scrap melting in a perfectly stirred bath, while the second corresponded to the case of conduction or low- level, natural-convection-dominated heat transfer in the liquid steel bath. The following specific issues associated with scrap melting practice were addressed: (1) the effect of scrap porosity on melting, (2) the effect of preheating scrap on melting, (3) the effect of liquid steel bath temperature on melting, (4) the effect of stirring on melting, and (5) the effects of scrap size on melting. In addition, for the case of conduction-dominated melting, a simple mean-field analytical model was developed and found to predict the kinetics of the melting front through the steel bath quite well. Details of the two-bar melting simulations are presented in Section III. Phase-field simulations of randomly distrib- uted scrap melting and the analytical one-dimensional (1-D) melting model are presented in Section IV. JIANGHUA LI, formerly a Graduate Student with Materials Science and Engineering, McMaster University, is presently a Post- doctoral Fellow with Department of Materials Science and Engineering, University of Toronto, Toronto, ON, Canada M5S 3E4. Contact e-mail: jianghua.li@utoronto.ca NIKOLAS PROVATAS, Associate Profes- sor, is with Material Science and Engineering, McMaster University, Hamilton, ON, Canada L8S 4L7. Manuscript submitted May 8, 2007. Article published online March 20, 2008. 268—VOLUME 39B, APRIL 2008 METALLURGICAL AND MATERIALS TRANSACTIONS B