The turning point in aluminum recycling: How innovation is fueling the Green Revolution

Global supply chain disruptions, rising energy costs and increasing environmental pressures are transforming priorities across industries worldwide. This has led to a surging demand for low-carbon ‘green’ aluminum, with recycling becoming the most promising pathway due to aluminum’s minimal energy requirements – just 5% compared to primary production – and its infinite recyclability. Green aluminum is emerging as the cornerstone against market uncertainties – and we believe there is a good reason for that. 

Meanwhile, new waste shipment regulations are changing the landscape for recyclers, fueling the drive for higher-purity feedstock which supports domestic rolling mills and extrusion plants, particularly in Europe. The good news is that the industry has made significant strides in recent years, with new technologies emerging. Ranging from Dynamic LIBS (Laser-Induced Breakdown Spectroscopy) to deep learning, these innovations set the scene for an exciting new era, paving the way for new opportunities. Let’s look into them in more detail. 

Advanced sorting with Dynamic LIBS technology 

A reliable local supply of high-quality recycled aluminum depends on the ability to accurately and consistently separate alloys. Dynamic LIBS technology, featured in TOMRA’s AUTOSORT™ PULSE, marks a significant breakthrough in precision sorting. Launched in 2023, this highly sophisticated system makes the advanced sorting of different aluminum alloys possible, closing another gap towards aluminum circularity. 

What makes Dynamic LIBS technology so groundbreaking? Unlike static LIBS systems, where objects are analyzed as they pass through a fixed laser beam with random targeting, Dynamic LIBS incorporates sophisticated 3D laser measurements and AI algorithms to determine optimal laser targeting positions. 

At the core of this innovation lies its unique single-point focus mode, enabling the laser to repeatedly track and target the same spot on a moving object. Imagine it as if the laser is ‘drilling’ into the material. This approach enhances material penetration, generating considerably more spectral data for analysis which results in the highest possible sorting accuracy. The system excels even when processing items with thick coatings, complex 3D shapes, surface irregularities or longer items, where single-pass systems often struggle to provide reliable readings. 

This robust analysis capability directly translates to improved sorting outcomes: when processing mixed 5xxx and 6xxx stamping scrap, real-world applications of the AUTOSORT™ PULSE have demonstrated both recovery and purity rates exceeding 95% in a single step. Even more impressive results have been achieved with complex streams like Taint Tabor or mixed wrought fractions from Twitch. Here, purity levels can surpass 97% when sorting 6xxx, 5xxx or 3xxx series alloys, or even specific alloys such as 6063. At the same time, maintaining recovery rates of >95% is possible, reaching as high as 97-98% for certain alloy series like 4xxx/cast alloys. 

In terms of throughput capacity, TOMRA’s Dynamic LIBS system can process 4-7 tons per hour for lighter stamping scrap, scaling up to more than 10 tons per hour for typical used aluminum old sheet or mixed wrought fractions. The technology maintains effectiveness even with smaller pieces down to 10 mm, though throughput rates may adjust accordingly for very small items. 

These impressive figures demonstrate the significant advancements achieved by Dynamic LIBS systems, with real-world applications confirming their pioneering potential.  

How new technology can create new markets 

One plant where the groundbreaking impact of our Dynamic LIBS technology can already be seen is Gerhard Lang Recycling GmbH in Germany. As one of the first companies globally to implement this system, they are redefining precision sorting of alloy stamping scrap generated during automotive production. This implementation is part of the KANAL research project, funded by the German Federal Ministry of Economic Affairs and Climate Action, which aims to close the loop on aluminum scrap in automotive production. 

At its Gaggenau facility, Gerhard Lang Recycling has installed the AUTOSORT™ PULSE to establish a scalable solution for sorting stamping scrap, which comprises a mixture of 5xxx (high-magnesium) and 6xxx (low-magnesium) wrought alloys. Previously, the mixed aluminum scrap was sold directly to manufacturers without further separation. Now, with the AUTOSORT™ PULSE and TOMRA's advanced Dynamic LIBS technology, the alloys are sorted into distinct, high-purity 5xxx and 6xxx products.  

The success of Gerhard Lang Recycling underscores the key advantage of Dynamic LIBS technology: it opens up access to previously untapped scrap sources, enabling the recovery of high-quality aluminum alloys and creating new markets. By preventing downcycling and preserving material value, this technology supports full material circularity and a less carbon-intensive aluminum supply chain.  

A revolution around the globe 

This is just the beginning. We can expect to see significantly more of such successful implementations throughout our industry worldwide in 2025. The green aluminum revolution is not limited to Europe but is developing on a global level – something we experience every day when talking to our customers. 

Demand for aluminum sorting technology is expanding beyond established markets like Europe and the UK. We’re witnessing significant demand in markets such as Eastern Europe, the Middle East, the US, and countries like Turkey and Japan. This development underscores the truly global nature of this transformation and its far-reaching impact. 

The next step in aluminum recycling: deep learning 

Another technology that has gained traction in recent years is deep learning. Already integrated into our AUTOSORT™ PULSE, deep learning algorithms enhance the system's ability to identify the shape and position of items on the conveyor belt. For example, when multiple pieces overlap, the AI-based object singulation function determines whether they form a single object or multiple ones in the system’s software. This innovation will help to further improve single object detection beyond traditional processing technology, allowing for higher belt occupancy and faster processing speeds — all while maintaining exceptional sorting purity. 

Deep learning’s transformative potential extends far beyond this application. To learn how, let’s first look at what deep learning actually is. As a subcategory of AI, deep learning imitates the way the human brain processes information. It is a special technique within machine learning that uses artificial neural networks that are trained by huge amounts of data, so they recognize and store certain patterns and later apply them to new data. 

Our AI experts feed thousands of images into the network as training material until it learns to distinguish certain visual characteristics such as shapes, sizes, dimensions or other details. This enables deep learning to solve some of the most complex sorting tasks, which helps us not only to reach even higher sorting granularity but also reduce manual labor in sometimes challenging work environments. 

One prime example is our GAINnext™ technology which is already demonstrating its value in aluminum recovery by efficiently cleaning used beverage can (UBC) fractions or in plastics applications. GAINnext™ is very effective at complementing existing sorting technologies. At TOMRA, we see the vast potential in this solution, and we are actively working on expanding our GAINnext™ ecosystem to address complex sorting needs in the metals sector, which currently rely on manual labor. 

The power of three 

When talking about new technologies, let’s not overlook current solutions. Today, XRT systems such as TOMRA’s X-TRACT™ which sort by atomic density, are widely regarded as the gold standard for separating heavy metals from aluminum. Having proven their value for more than a decade, there is no doubt that XRT solutions will undoubtedly remain central to any aluminum processing plant. The true power, however, if we want green aluminum to become the standard in production processes, lies in combining XRT technology with one or, ideally both the advanced sorting solutions mentioned above.   

By integrating all three technologies, we can take a significant step closer to true material circularity, with sorting playing a critical role in the process.  

Once again, our industry is at a pivotal moment, with technology driving this change. We are convinced that the use of Dynamic LIBS and deep learning will drive the circular economy forward at a time when it is most needed – customer demand for technologically advanced solutions is increasing, companies are aiming to fulfill their net-zero targets and legislation is tightening. Now is the ideal time for new markets to emerge and further boost our industry.  

Here at TOMRA, we are excited to be part of this progress! 

This article was originally published in the March/April 2025 issue of ALUMINUM INTERNATIONAL TODAY