авг 6, 2019

Identify minerals in real time? It is possible, with a laser developed in Portugal

Researchers from INESC TEC have submitted a European patent application for laser technology based on Artificial Intelligence, which allows minerals to be distinguished in real time.


The system developed by the researchers of the Institute of Systems and Computer Engineering, Technology and Science (INESC TEC) has the capacity of self-learning and, through new data, it learns what new behavior to acquire.

Laser technology uses Artificial Intelligence to identify and quantify minerals in real time. The concept that supports the innovation is that of LIBS, laser-induced breakdown spectroscopy or laser-induced plasma spectroscopy in Portuguese. The solution aims to solve a real problem in the mining industry and the prototype was created in the context of an abandoned mining project and may have various applications in various sectors of activity, such as environment, agriculture, health, cultural heritage, among others.

The system pulverizes the mineral sample placed in the laser, generating a plasma that, when it cools down, emits the energy stripes specific to each element and it is from these data that it perceives which element to identify and how much, explains the press release. "A single element can have hundreds of energy stripes and a raw material can have thousands. To understand the energy stripes of each element is, in essence, to have access to a kind of fingerprint of the element", explains Pedro Jorge, researcher at the Centre for Applied Photonics (CAP) of INESC TEC and one of the inventors of this technology.

This solution allows to obtain results in real time, being faster than the alternatives that require samples to be sent to the laboratory and offers even less prone to errors than portable devices.

Researchers explain that this technology allows better use of materials, saves resources and reduces environmental impact. The team received the green light and funding to study the technology in contexts such as lithium mining, soil contamination and precision farming.

The goal now is to transfer this technology to industry by integrating this transparent artificial intelligence software into smaller, portable devices.