Liquid-Metal Droplet Coalescence in Metallurgical Slags: A Critical Review of Interfacial Phenomena, Slag Properties, Hydrodynamics, Kinetics, and Metal Recovery
Year: 2026
Type: article
Abstract: Metal losses to metallurgical slags arise from fundamentally different mechanisms: chemical dissolution in the oxide or salt phase and physical entrainment of discrete liquid-metal or matte droplets. This distinction is decisive because only the physically entrained fraction can be directly targeted by collision, coalescence, settling, or forced-separation strategies. This structured critical narrative review integrates a retained contemporary corpus of 108 publications from 2020 to August 2026, supplemented by four foundational pre-2020 mechanistic sources used only to anchor classical droplet-transport, breakup, coalescence, and population-balance concepts and excluded from the corpus count. The review examines how droplet generation, interfacial chemistry, slag rheology, hydrodynamics, and residence time jointly determine whether dispersed metal is recovered or discarded with slag. The review develops a Droplet–Interface–Slag–Hydrodynamics–Recovery (D–I–S–H–R) framework and emphasizes five coupled controls: droplet-size distribution, dynamic interfacial tension and wetting, slag viscosity and solid fraction, collision-versus-breakup hydrodynamics, and the time/force available for phase separation. Evidence from copper-slag cleaning, aluminum salt-flux processing, iron-droplet assembly, steel–slag emulsification, and supergravity separation shows that lower viscosity and larger droplets generally favor separation, but the response is non-monotonic because additives and mixing can simultaneously alter liquidus temperature, interfacial reactions, droplet breakup, oxidation, refractory interaction, and metal quality. A central conclusion is that “more mixing” is not equivalent to “more recovery”: useful operation requires a finite hydrodynamic window in which collision frequency is increased without creating a finer steady-state dispersion, followed by sufficient low-turbulence or forced-separation time. The literature remains limited by scarce in-situ droplet-size measurements, dynamic high-temperature interfacial data, closed metal balances, industrial validation of multiphase models, and standardized reporting of uncertainty. The proposed framework converts these gaps into a process decision sequence for diagnosing metal loss, conditioning slag, managing interfaces, tuning mixing, and providing adequate separation before tapping.
Author: Antonio Clareti Pereira
Institution: Universidade de São Paulo
Language: English
Subfield: Metals and Alloys
Field: Materials Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W7203766821
