Increasing the share of scrap-based steel production is necessary to achieve CO2 emissions targets. However, the quality of recycled steel is compromised by contaminating elements, of which copper is the most pervasive. Copper from wiring and motors entangles with steel fragments during shredding and is not completely removed by magnetic separation. Beyond hand-picking, no commercial process exists for extraction, but copper in solution with steel segregates during hot rolling, causing surface cracking and defects that are unacceptable for high-quality flat products. This thesis characterizes copper in the global steel system, evaluates the energy requirements of possible extraction processes and presents experimental results to aid in the development of an efficient extraction technique.

Copper contamination is currently managed by globally trading contaminated scrap to tolerant applications and by dilution with primary steel. An evaluation of copper in the global steel system is needed to develop long-term strategies, and this is presented in the first part of this thesis. The copper concentration of flows along the 2008 steel supply chain are estimated from a range of literature sources and compared with the maximum concentration that can be tolerated in all steel products. Quantities of final steel demand and scrap supply by sector from a global stock-saturation model are used to estimate the amount of copper in the future scrap supply, and the total amount tolerable. Assuming current scrap preparation continues, more copper will enter the steel cycle than can be tolerated by demanded products by 2050. This global constraint will set in sooner if primary production is cut to meet climate mitigation targets.

Given the upcoming constraints, improved copper control is necessary. Various techniques for copper separation have been explored in laboratory trials, but as yet no attempt has been made to provide an integrated assessment of these options. The second part of this thesis presents a framework to define the full range of separation routes and evaluate their potential to remove copper, while estimating their energy and material input requirements. The thermodynamic, kinetic and technological constraints of the various techniques are analyzed to show that copper could be removed to below 0.1wt% (enabling the production of high-value flat products) with 5-20% of the melting energy in the electric arc furnace route.

The above analysis reveals a promising and under-explored process route: preferential melting of copper from solid steel scrap, which could be integrated into conventional scrap re-melting with little additional energy. Previous investigations show removal of liquid copper is limited by its adherence to solid scrap. In the third part of this thesis, the individual and combined effects of several parameters (steel carbon content, initial surface oxidation and applied coatings) on the wetting behavior of liquid copper are observed with a heating microscope to understand if a process window to enable separation exists. The most significant factor was carbon content. On medium carbon steel substrates, copper spread rapidly, likely due to reduction of the oxide layer by carbon. Non-wetting copper droplets were observed on low carbon substrates in an inert atmosphere. This indicates a possible process window, but further investigation considering diverse, fragmented end-of-life scrap is needed.

The scrap supply of all metals is expanding. The multi-scale, interdisciplinary method developed in this thesis could be applied to other metal systems to understand the constraints caused by contamination and identify key areas to develop efficient extraction processes, necessary to conserve resources and reduce CO2 emissions.