75% of clothing is discarded within one year of production.
The growing need for chemical recycling techniques
TI DLP NIRscan Nano can be classified as Sagitto

The burn test is a classic method for determining what a fabric is made of. It involves burning a sample of fabric to see if it shrinks, melts, or burns, and smelling the resulting odor.

This is a cumbersome process. However, using the TI DLP NIRscan Nano evaluation module and the Sagitto system, textile composition analysis can be easily performed. Sagitto combines an ultra-small near-infrared sensor with a machine learning model, enabling unprecedented analysis. Each fabric has a unique near-infrared fingerprint based on its fiber composition. Most clothing is composed of various fibers, and accurately understanding the fiber composition is crucial throughout the garment's lifespan.
Many countries require clear labeling of the fiber composition of textiles. However, in some cases, these labels can be misleading. For example, in the picture below, the dishcloth set is labeled as 100% cotton, but when tested with Sagitto, it was found to be a blend of 67% cotton and 33% polyester.
Why is fiber composition important?

Each year, 80 billion pieces of clothing are produced worldwide, 75% of which end up in landfills or incineration. Many are pressuring major clothing companies to find alternatives to reduce resource waste caused by this high waste rate. Governments around the world are promoting the "circular economy" and introducing measures to recycle clothing that would otherwise end up in landfills.

Clothing made of acrylic and polyester has a significant environmental impact, releasing hundreds of thousands of microfibers into sewage treatment plants with each wash. Of this, 40% ultimately flows into rivers, lakes and oceans.
This is why pressure is growing to develop new chemical recycling techniques for textiles. These techniques can return polyester and cotton garments to their original chemical components: cellulose fibers and polyester monomers and polymers. To do so, recyclers seeking chemical recycling must first accurately sort garments by fiber composition.

Previously, workers sorted waste textiles by sight and feel. They picked up each garment, inspected it, and assessed it. However, it's impossible for the human eye to determine the fiber composition with the level of accuracy required for chemical recycling techniques.

Conversely, by equipping a robotic arm with a TI DLP NIRscan Nano and combining it with sophisticated machine learning, we can develop an accurate robotic sorting system for chemical recycling facilities.

Sagitto combines cloud-based artificial intelligence with the DLP NIRscan Nano. With Sagitto, you don't need your own data scientists. There's no need to collect data separately to train machine learning models. Therefore, Sagitto removes the barriers associated with equipment cost, expertise, and data, enabling a wider range of manufacturers and producers to optimize their processes using the DLP NIRscan Nano EVM. You can experiment with demo models of fiber composition using Sagitto artificial intelligence software and the DLP NIRscan Nano EVM.

This article is based on a contribution from Texas Instruments.