**Jani-Petteri Jylhä, Nadir Ali Khan and Ari Jokilaakso \***

Department of Chemical and Metallurgical Engineering, Aalto University, Kemistintie 1, P.O. Box 16100, FI-00076 Aalto, Finland; jani-petteri.jylha@aalto.fi (J.-P.J.); nadir.khan@aalto.fi (N.A.K.)

**\*** Correspondence: ari.jokilaakso@aalto.fi; Tel.: +358-50-3138-885

Received: 12 March 2020; Accepted: 15 April 2020; Published: 22 April 2020

**Abstract:** Computational methods have become reliable tools in many disciplines for research and industrial design. There are, however, an ever-increasing number of details waiting to be included in the models and software, including, e.g., chemical reactions and many physical phenomena, such as particle and droplet behavior and their interactions. The dominant method for copper production, flash smelting, has been extensively investigated, but the settler part of the furnace containing molten high temperature melts termed slag and matte, still lacks a computational modeling tool. In this paper, two commercial modeling software programs have been used for simulating slag–matte interactions in the settler, the target being first to develop a robust computational fluid dynamics (CFD) model and, second, to apply a new approach for molten droplet behavior in a continuum. The latter is based on CFD coupled with the discrete element method (DEM), which was originally developed for modeling solid particle–particle interactions and movement, and is applied here for individual droplets for the first time. The results suggest distinct settling flow phenomena and the significance of droplet coalescence for settling velocity and efficiency. The computing capacity requirement for both approaches is the main limiting factor preventing full-scale geometry modeling with detailed droplet interactions.

**Keywords:** computational fluid dynamics; CFD–DEM; coalescence; settling; funneling flow
