If you search the web for “how to determine what a fabric is made of”, you’ll likely find web content for “burn tests.” In a burn test, a small sample of fabric is taken, placed over an open flame, and watched to see if it shrinks, melts, or burns, and pays attention to the odor that is produced.
Now, with the TI DLP® NIRscan® Nano Evaluation Module (EVM) and the Sagitto system, it is easier and more accurate to determine fabric and textile composition. The Sagitto system combines tiny near-infrared sensors and machine learning models to help businesses simplify the measurement process. Each type of fabric has a unique near-infrared fingerprint due to different components. Garments often contain different types of fibers, and the precise composition of the composition is important throughout the life of the garment.
Figure 1: Near-infrared absorption spectra of textiles with different fiber content
Many countries require clear identification of the fiber content of textiles. Sometimes these labels are misleading. For example, in the image below we see a set of dish towels labeled 100% cotton, but tested by Sagitto and found to be a mix of 67% cotton and 33% polyester.
Figure 2: The dish towel is shown as 67% cotton and 33% polyester, not 100% cotton on the label
But why does fiber composition matter? It is estimated that 80 billion pieces of clothing are produced each year, of which 75% end up in landfill or incineration. More and more consumers are asking large clothing retailers to find alternative ways to deal with the high volume of waste generated in the high-turnover fashion retail industry. The government has also begun to develop regulations to encourage a “circular economy” and divert clothing from waste.
Acrylic and polyester clothing can have serious environmental impacts, for example, releasing hundreds of thousands of microfibers to local wastewater treatment plants with each wash cycle. As much as 40 percent of these microfibers may end up in rivers, lakes and oceans.
Figure 3: Textile waste is becoming a global problem
Therefore, the market urgently needs to develop new textile chemical recycling technology. For example, these recycling technologies require the breakdown of polyester and cotton garments into their chemical constituents – cellulose fibers, polyester monomers and oligomers. But first, recyclers using chemical recycling need to precisely sort raw materials by fiber composition.
In traditional operations, waste textiles are usually sorted by feel and vision, i.e. the composition of the textile is determined when each garment is picked up. Unfortunately for humans, it is simply impossible to precisely determine the composition of textiles containing fiber mixtures and meet the requirements of modern chemical recycling technologies.
By integrating the TI DLP NIRscan Nano into a robotic arm, coupled with sophisticated machine learning capabilities, it is possible to develop a precise robotic sorting system for chemical recycling plants.
Sagitto combines DLP NIRscan Nano with cloud-based artificial intelligence. With Sagitto, you don’t need to hire your own data scientists or even collect your own data to train machine learning models. Sagitto removes barriers such as equipment cost, skills and data, enabling a new class of manufacturers and producers to optimize production processes using the DLP NIRscan Nano EVM.
Using the Sagitto artificial intelligence software and the DLP NIRscan Nano evaluation module, you can demonstrate models with unique fabric compositions. Register on the Sagitto website and request access to a Sagitto demo account to make 50 free predictions using the DLP NIRscan Nano evaluation module.