Laser Induced Breakdown Spectroscopy to Classify Recycling


A group of scientists in China are using LIBS technology to help accurately classify recyclable waste.

The classification of recyclable waste is a challenge for environmental protection. Properly classifying refuse means that it is less likely to be contaminated by other waste during the recycling process and can effectively be reused.

A group of researchers at Hefei University of Technology in China, are using laser-induced breakdown spectroscopy (LIBS) technology for the classification and identification of recyclable waste (1). The scientists outlined their findings in the journal AIP Advances earlier this year.

LIBS is an atomic emission spectroscopy technology that detects the elemental composition of the sample based on emission spectra, the researchers wrote in the study. Unlike many spectroscopic techniques, LIBS technology is not impacted by the environment or lighting, or other factors such as the sample’s shape or color. LIBS is already used to classify iron ore, coal, plastic, ceramics, and more, the authors wrote, which makes it ideal for classifying waste. More traditional methods of waste classification include gravity separation, magnetic separation, and electrostatic separation. These methods are often only able to detect a few categories of waste.

“We have used LIBS technology for the first time to identify and classify recyclable waste,” said Lei Yang, a study author and professor at Hefei University, in a press release. “This method has accurate, reliable, fast detection results, and can achieve automatic detection.”

The researchers analyzed the spectra of 80 recyclable waste samples and using LIBS classified each sample into paper, plastic, metal, glass, textile, and wood. From there, the team further classified each sample into a more specific sub-category. 

The team also used machine learning (ML) including linear discriminant analysis (LDA) and random forest (RF) to further classify the waste. ML involves the use of algorithms to learn from data to classify samples without being explicitly programmed to do so. LDA is a supervised classification method that aims to find the linear combination of features that best separates different classes by maximizing the ratio of between-class variation to within-class variation. RF is an ensemble learning method that constructs multiple decision trees during training and combines their outputs to improve classification accuracy, while combining the individual predictions from multiple classifier models to arrive at a final classification decision.

While LIBS proved to be a promising technology for categorizing waste, there’s still more work to be done. In the future, they plan on increasing the number of waste samples and incorporating new types of waste such as kitchen waste, they wrote in a press release.

“What surprised us the most was that by using LIBS technology for classification and recognition without any preprocessing of the waste object, the results are satisfactory,” Yang said.


1. Yang, L.; Xiang, Y.; Li, Y.; Bao, W.; Ji, F.; Dong, J.; Chen, J.; Xu, M.; Lu, R. Identification and Classification of Recyclable Waste Using Laser-Induced Breakdown Spectroscopy Technology. AIP Advances 2023, 13 (7). DOI:10.1063/5.0149329.

Related Content