The need to recover and recycle material towards building a circular economy is increasingly a global imperative. Non-ferrous metals in particular are highly recyclable and can be extracted using processes such as eddy current separation. However, their further separation into recyclable groups based on metal or alloy continues to pose a challenge. Recently, we proposed a new technique to discriminate between non-ferrous metals: Magnetic induction spectroscopy (MIS) measures how a metal fragment scatters an excitation magnetic field over different frequencies. MIS is related to conductivity, which can be used to classify the fragment according to this property.In this paper, we demonstrate for the first time the use of MIS with machine learning to classify non-ferrous scrap metals drawn from commercial waste streams. Two approaches are explored: (1) MIS over a bandwidth from 3 kHz to 90 kHz, and (2) the combination of MIS with physical colour of the metal samples. We show that MIS alone can obtain purity and recovery rates >80% for most metal groups and waste streams, rising to >93% for stainless steel. The exception was the Zorba waste stream where the mix of aluminium alloys within the sample set led to poor conductivity contrasts. The introduction of colour substantially improved results in this case, increasing purity and recovery rates by 20-35 percentage points. Of the machine learning models tested, we found that random forest, extra trees and support vector machine algorithms consistently achieved the highest performance.

Magnetic induction is widely used to detect and classify metal objects over a range of applications; this paper considers the potential of this technique to inspect for the presence and characteristics of batteries within waste streams. As the number of batteries used across the world increases, an efficient method is needed to ensure batteries can be classified to allow for more efficient recycling. The detection of batteries would also reduce the risk of fire and pollution by identifying the battery before they are crushed or shredded.In this study, a magnetic induction sensor measured the batteries and scrap metal between 781 Hz and 95282 Hz to allow a significant frequency range to be observed. The real component (in-phase) of a battery's electromagnetic response is different from scrap metal; this could allow for an algorithm to be trained to detect batteries within metal waste or when inside non-metallic objects. The response observed shows that batteries could be grouped into size, which is useful if no line of sight is available, which a traditional camera system requires. Once grouped into size the batteries could be further separated according to their internal contents when the real component is used; this would reduce the risk of cross-contamination when they are recycled. The real component response of lithium and NiMH batteries is different when compared to the other batteries; this could allow them to be detected and removed from a waste stream, which is important as lithium batteries can set on fire.