The growing volume of end-of-life (EoL) photovoltaic (PV) panels presents a significant challenge and an opportunity for sustainable resource recovery. This thesis investigates the application of Magnetic Density Separation (MDS) to recover and separate valuable non-ferrous metals, such as silicon, copper, zinc, tin, and silver, from crystalline silicon PV waste. The objective of the study is to evaluate the separation efficiency and recovery rates of MDS while addressing economic feasibility and process optimization challenges.
The experimental work involved the use of MDS to fractionate PV waste into light, medium, and heavy-density ranges based on particle size distributions: 200–500 µm, >500 µm–3 mm, and >3 mm-6 mm. X-ray fluorescence (XRF) was employed to analyze the elemental composition of the separated fractions, with leaching processes using hydrochloric acid (HCl) and citric acid applied to assess the purity of magnetic samples. Calibration and instrumental errors, as well as the effects of alloying and particle motion, were accounted for in the analysis. A financial assessment of the MDS process was conducted, examining revenues generated from metal recovery and associated costs, including the economic implications of contamination and refining. The results revealed that MDS achieved moderate separation efficiencies, with high recovery rates of silica (SiO2) in the form of glass and sand in the light fractions and silver (Ag) and copper (Cu) in the heavy fractions. However, significant contamination was observed due to overlapping densities of alloyed metals, such as Tin-silver-copper (Sn-Ag-Cu) and Zinc-copper (Zn-Cu), and particle agglomeration during feeding. A financial snapshot demonstrated that medium- and large-sized fractions contributed the most to revenue, driven by silver recovery, with a break-even point requiring the processing of at least 114 tons of PV waste. In conclusion, MDS shows significant potential as a low-energy, cost-effective solution for recycling EoL PV panels. However, achieving industrial viability requires addressing contamination issues and optimizing the separation process. Future work should focus on integrating complementary refining techniques and scaling up operations to meet purity standards and economic demands. The study also recommends the need for improved ferrofluid magnetization and alloy-specific calibration to enhance detection limits.

This study investigates the economic performance of waste electrical and electronic equipment (WEEE) collection schemes in Europe, focusing on two case studies: Cyclad, France, and Helsinki, Finland. Guided by the European Union’s WEEE Directive, the research employs cost-benefit analysis (CBA), sensitivity analysis, and stakeholder interviews to evaluate collection systems under varying socio-economic and geographic conditions. Key metrics such as collection rates, investment costs, operational expenses, and producer responsibility organization (PRO) fees were analyzed across standard, best-case, and worst-case scenarios.
The results highlight the influence of regional variables, including population density and geographic area, on the financial and operational performance of WEEE collection systems. Cyclad, operating in a rural setting with a dispersed population, faced higher logistical costs and required additional investments in theft prevention and public awareness. In contrast, Helsinki benefited from its urban density and centralized infrastructure, which facilitated cost reductions and increased collection efficiency. While Cyclad’s financial performance showed greater sensitivity to fluctuations in PRO fees, Helsinki demonstrated resilience across scenarios due to its mature system.
This study emphasizes the importance of tailored financing models, public engagement strategies, and technological innovations to optimize WEEE collection systems. These findings offer valuable insights for policymakers and stakeholders aiming to enhance the sustainability and cost-effectiveness of WEEE collection practices across diverse European regions.