Commercially available mild steel (CMS) electrodes and electrodes recycled from waste tomato paste cans (recycled mild steel, RMS) were compared in an electrocoagulation reactor treating amoxicillin in aqueous solution. Both materials were characterised by scanning electron microscopy, and the operating conditions were optimised one factor at a time: initial pH (5, 7, 8), initial concentration (10, 25, 50, 100 mg/L), inter-electrode distance (1.0, 1.5, 2.0 cm) and applied voltage (5, 10, 15, 20 V). The optimum was pH 7, 1 cm spacing and 15 V, under which removal exceeded 99% for both materials: 99.4% for the recycled electrode and 99.3% for the commercial one. Over the full 90-minute treatment the corresponding specific energy consumption was 32.06 and 32.63 kWh/m³ respectively. Removal followed pseudo-first-order kinetics on both materials, the apparent rate constant falling from 0.038 to 0.019 min⁻¹ as the initial concentration rose from 10 to 100 mg/L. The concentration dependence at 90 minutes was described best by the Langmuir isotherm, with an apparent monolayer capacity of 76.5 mg/g of dissolved iron for the recycled electrode and 72.3 mg/g for the commercial one; equilibrium was not reached within this time, so these capacities are lower bounds. Electrodes recovered from discarded food cans therefore remove amoxicillin from the aqueous phase as effectively as commercial mild steel while diverting a waste stream from disposal, a practical contribution to the circular economy. Removal denotes transfer of the antibiotic into the iron hydroxide floc rather than degradation, which was not measured.