Donar DE-XRT Pro AI-Driven Sorting System

Aluminum recycling presents a sorting challenge where alloy composition and impurity content directly determine commercial value. Zorba, the mixed non-ferrous shredder product recovered from end-of-life sources, not only contains heavy metal impurities such as copper, brass, zinc, and stainless steel, but is also mixed with wrought aluminum alloys and cast aluminum alloys of varying values. In particular, aluminum and zinc both present highly similar silvery-gray appearances under visible light, and wrought aluminum alloys and cast aluminum alloys are difficult to distinguish by surface color alone. Relying solely on surface features makes it impossible to effectively identify and remove these 'visually identical' impurities, let alone achieve refined classification of aluminum fractions. 

The Donar DE-XRT Pro series is an intelligent dual-sensor sorting system tailored for this exact scenario. It adopts deep fusion of dual-energy X-ray transmission (DE-XRT) technology and high-resolution RGB (HR-RGB) VIS imaging, powered by proprietary AI deep learning algorithms, to transform Zorba into high-purity aluminum fractions. By analyzing differential X-ray attenuation across two energy levels, the system identifies and ejects heavy metal impurities, including copper, brass, zinc, and stainless steel, based on their distinct atomic density signatures. Furthermore, the AI-driven visual recognition module analyzes surface characteristics such as texture, shape, and color to efficiently separate wrought aluminum alloys from cast aluminum alloys.

Empowered by AI-driven surface feature recognition, the system continuously learns from real production data, steadily refining its sorting accuracy. This means the equipment's performance is not static after deployment; instead, it continuously adapts and improves as production data accumulates, delivering stable and reliable sorting performance. This enables aluminum scrap recyclers and secondary smelters to produce high-purity aluminum fractions that meet stringent downstream remelting specifications, unlocking premium market pricing and maximizing recovery value.

Sorting Principle of Donar DE-XRT Pro AI-Driven Sorting System

Purifying Aluminum from Zorba (Heavy Metal Impurity Removal)

The separation of aluminum fractions from heavy metal impurities in Zorba is fundamentally based on the significant differences in effective atomic number and material density between these materials.

When dual-energy X-rays penetrate the material stream, the system measures the differential attenuation between high-energy and low-energy X-rays. Aluminum, as a light metal with low atomic number, exhibits relatively weak X-ray attenuation. In contrast, heavy metal impurities such as copper, brass, zinc, and stainless steel possess significantly higher effective atomic numbers and densities, resulting in strong X-ray absorption. By calculating the Dual-Energy Ratio (DER), the system accurately determines the intrinsic material signature of each individual particle, effectively eliminating interference from variations in material thickness or surface conditions such as paint, oil, or oxidation. Based on these density and atomic number differences, the system efficiently identifies heavy metal impurities and triggers pneumatic ejection to remove them, leaving behind purified aluminum fractions.

Separating Wrought Aluminum from Cast Aluminum

The separation of wrought aluminum alloys from cast aluminum alloys is primarily based on subtle differences in both material density and surface characteristics.

Although both wrought and cast aluminum are aluminum-based materials, cast aluminum (commonly used in components such as engine blocks) contains a higher proportion of alloying elements, particularly silicon (Si), along with copper and zinc in many cases. This results in a distinct material density and X-ray absorption signature compared to wrought aluminum (which typically has lower alloy content). Leveraging high-precision X-ray transmission technology, the system can identify and eject high-alloy cast aluminum particles, significantly reducing their content in the material stream. However, some low-alloy cast aluminum grades have densities very close to those of wrought aluminum, making them difficult to distinguish via X-ray transmission alone. Cast aluminum surfaces, however, typically exhibit distinctive visual features imperceptible to the human eye—such as mold casting traces (parting lines, ejector pin marks), characteristic textures, and specific surface sheens. By utilizing AI deep learning to detect these minute visual and morphological features, the system can efficiently eject low-alloy cast aluminum that shares similar density but differs in material composition and surface morphology. This serves as a powerful complement to X-ray transmission sorting, significantly enhancing the value of the recycled aluminum stream. (At this stage, the material stream consists almost exclusively of wrought aluminum, though it may still contain a mixture of different wrought alloy grades. Further refined separation of these alloy grades is achieved through more advanced LIBS technology to unlock even higher value.)

Technical Features

● Dual-Sensor Fusion Technology

Deep integration of dual-energy X-ray transmission (DE-XRT) and high-resolution RGB (HR-RGB) VIS imaging enables efficient identification and removal of heavy metal impurities, significantly improving aluminum fraction purity. By combining effective atomic number and material density analysis with surface characteristic recognition, the system achieves intelligent, high-efficiency separation of wrought aluminum alloys from cast aluminum alloys.

● Proprietary AI Deep Learning Algorithms

Trained against comprehensive identification and sorting standards, the system uses deep learning to replicate human visual judgment for material classification. The AI model can be continuously retrained and optimized with new production data, driving ongoing improvements in sorting accuracy.

● High-Performance Air Ejector

Equipped with in-house, high-power, high-speed air ejectors, the system delivers high-precision material rejection and ensures long-term stable operation even under high-throughput sorting conditions.

● Multi-Specification Model Portfolio

Available in two models with belt widths of 1300 mm and 2000 mm, meeting diverse throughput requirements. The system is suitable for auto dismantlers, aerospace aluminum recyclers, packaging material recyclers, and incineration bottom ash (IBA) recyclers of all scales.

✦ Recovery of Aluminum Scrap

✩ Building Window/Door Recycling

✩ Beverage Can Recycling

✩ E-Waste Aluminum Recycling

✦ Separation of Wrought Aluminum and Cast Aluminum

✩ Auto Shredder Residue Sorting

✩ Industrial Machinery Scrap

✩ Aerospace Aluminum recycling

✩ Consumer Electronics Housing

✦ Recovery of Stainless Steel

✩ IBA (Incinerator Bottom Ash) Recycling

✩ ASR (Automotive Shredder Residue) Recycling

Specifications

Specifications are approximate and subject to change without notice. Lauffer Vision reserves all rights for modifications.