Researchers at Nagoya University, in collaboration with the National Institute for Materials Sciences in Japan, have developed a groundbreaking machine learning technique that can accurately detect microplastics. This new method addresses the challenges associated with traditional detection methods that are time-consuming and costly.
Microplastics are difficult to distinguish from natural organic compounds found in water samples, making their detection a complex process. The new machine learning technique utilizes a porous metal foam to capture and detect microplastics optically using surface-enhanced Raman spectroscopy (SERS).
The data obtained from SERS is complex, but researchers have developed a neural network computer algorithm called SpecATNet to interpret the optical measurements. This algorithm can quickly and accurately detect six key types of microplastics: polystyrene, polyethene, polymethylmethacrylate, polytetrafluoroethylene, nylon, and polyethene terephthalate.
The researchers hope that this innovative method will aid in monitoring microplastics in water samples directly from the environment without the need for pretreatment. The ability to detect microplastics rapidly and accurately will play a crucial role in assessing the impact of microplastic pollution on public health and marine life.
In addition to creating inexpensive microplastic sensors, the researchers aim to expand the capabilities of the SpecATNet neural network to detect a broader range of microplastics. By providing open-source algorithms and enabling the detection of microplastics even in resource-limited labs, this new technique has the potential to revolutionize environmental monitoring efforts.
