Recycling Automation: Spectroscopy & Multispectral Sorting of Plastics & Metals
According to a 2025 Nature paper, the global recycling rate remains stagnant at 9%. With contamination from food traces, label remnants, and other impurities being one of the main reasons for this shortfall, the accuracy of recycling automation becomes a key lever for improving recovery. In response, hyperspectral imaging (HSI) is actively raising the bar for what automated sorting can achieve – and that increased precision relies on high-performance optical components to reliably detect and sort materials .
Identifying Materials on the Line
Recent peer-reviewed UCL research (2025) identifying contaminated plastic packaging with up to 97% accuracy demonstrates why HSI has established itself as one of the leading optical sorting techniques in automated recycling.
It works in the visible (VIS), near-infrared (NIR) and short-wave infrared (SWIR) bands for plastic sorting – the wavelength ranges where these materials reveal their chemistry. Every polymer has a distinct spectral fingerprint from its chemical composition – specifically, carbon-hydrogen (C-H) and oxygen-hydrogen (O-H) absorption in the NIR and SWIR regions. For example, polyethylene terephthalate (PET), high-density polyethylene (HDPE), polyvinyl chloride (PVC), polypropylene (PP), polystyrene (PS) and acrylonitrile butadiene styrene (ABS) each have their own profiles. These spectral signatures are captured by multispectral and hyperspectral cameras pixel-by-pixel across the conveyor, with the data feeding the sorting system’s decision in real time. Bandpass filters sit at the heart of these systems, isolating target NIR and SWIR wavebands that carry every polymer’s signature and blocking unwanted external light, ensuring readings are clean.
Metal sorting, meanwhile, generally follows a slightly different process; while laser-induced breakdown spectroscopy (LIBS) and X-ray fluorescence (XRF) handle alloy-level identification, HSI and multispectral imaging (MSI) distinguish metals from non-metals at earlier stages.
Optical Challenges on the Sorting Line
Modern-day waste-sorting facilities operate at high throughput. Because conveyors can reach speeds of 3 m/s, optical components require high transmission to ensure sufficient light reaches the sensor during short integration times. What’s more, mixed and overlapping materials combined with bright industrial lighting and shiny surfaces cause stray reflections that produce flare – often suppressed through anti-reflective (AR) coatings.
These processes also take place in harsh commercial environments, meaning optics can’t be compromised by dust, shock, vibration, and other contaminants that degrade sorting accuracy. To help protect internal components, durable optical windows, such as sapphire windows, are typically integrated to withstand abrasion and impact, helping to keep the optical path clear.
How Engineered Optics Drive Recovery Rates
Optical components directly influence recovery. HSI and MSI systems rely on them to capture accurate spectral data, and poor-quality optics lead to misclassification of materials with the potential to contaminate entire bales. Well-specified components provide cleaner separation, higher recyclate purity, and a stronger recycling yield, turning sorted material into commercially viable output.
Enabling the Circular Economy
Looking ahead, machine vision – built on HSI and MSI, and the optics behind them – is a critical step in lifting recycling rates beyond the current 9% baseline. Specifying the right optical components, such as precision filters and impact-resistant protective windows, is what will drive the industry’s next gains.
To learn more about our optical components for hyperspectral imaging and multispectral imaging, contact a member of our team today.