Project Details
Projekt
Exploration of the compositional phase space of metallurgical slag models for a rational design of processes of refractory metal recovery through smelting and recrystallization
Subject Area
Mechanical Process Engineering
Term
since 2021
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 470366582
The diversity of multicomponent oxides in slag systems offers significant possibilities in terms of a wide range of crystalline phases that can be selectively recovered from the slag matrix. Metallurgical tin and copper slags represent waste streams with great recycling potential for refractory metals (Co, Ta, Nb), which can be recovered from the slag matrix through Engineered Artificial Mineral (EnAM) phases. In the first phase, a rational methodology was implemented to identify new EnAM phases (e.g., CoFe2O4) with a supervised exploration of the phase space of a slag. However, the nucleation and growth mechanisms of these crystals within the slag matrix are largely unknown. This poses significant challenges for process design that ensures (i) sufficient loading of the crystals with valuable elements worth being recovered and (ii) cost-effectiveness in terms of technically relevant time scales (process speed) and conversion (process efficiency). This project focuses on experimental and simulation investigations of the crystallization mechanisms of identified EnAM phases rich in refractory metals under non-isothermal processing, with the aim of developing continuous cooling transformation (CCT) diagrams for multi-crystalline oxide systems. The work combines simulations using molecular dynamics methods with slag synthesis and melting experiments, as well as experimental material characterization, to gain a comprehensive understanding of the crystallization behavior of a complex slag system. Our main hypothesis is that the formation of a potential EnAM phase at the molecular level allows us to evaluate the thermodynamics and kinetics of phase formation and derive process guidelines for controlled melting and cooling that lead to EnAM formation with high recovery rates. The investigations towards this hypothesis are structured in four steps: (1) implementation of an efficient approach using machine learning (ML) and molecular dynamics (MD) for simulating EnAM nucleation; (2) slag synthesis using double flame spray pyrolysis (DFSP) and in-situ mixing with crystal seeds for EnAM formation as model material; (3) ML-MD simulations of EnAM interfacial growth in the slag matrix; and (4) controlled crystallization of model slags under defined cooling conditions within a laser-induced smelting process and heat treatments. We expect that our study will provide in-depth knowledge of the thermodynamics and kinetics of formation of refractory metal-enriched crystal phases from multicomponent systems, which can be transferred to real slags and processes on an industrial scale.
DFG Programme
Priority Programmes
