Team makes powerful water filter with help from light, vibrations

Scientists from the Institute of Nano Science and Technology (INST) in Mohali, IIT-Dharwad, and IIT-Kharagpur have designed a cheap, reusable water filter.

Industrial plants release dyes such as Congo Red and Methylene Blue into rivers and groundwater, from where they can cause stomach, skin, and breathing illnesses. Ozone, Fenton chemistry and other methods work to clean the water, but they burn through chemicals and electricity, expanding cost and the carbon footprint.

The new filter has been designed to sidestep these and other problems. Its development was reported in a paper in the July edition of Nano Energy.

The researchers first 3D printed thin, sponge-like sheets of polylactic acid (PLA), a biodegradable plastic often used in compostable cups. PLA is naturally water-repelling, so the team soaked each sheet in a mild sodium-hydroxide solution to make it water-loving.

Next, they made nanoparticles of bismuth ferrite (BFO) and dipped the prepared PLA sheets into a BFO ink. Treated sheets stayed strong through five reuse cycles, losing only about 3% of their cleaning power.

Under visible light, the BFO acted like a solar-powered catalyst that split water molecules and created highly reactive radicals that shred organic dye molecules. And when shaken by ultrasound, BFO’s piezoelectric nature generated an internal electric field that drove the same radical-making reactions even in the dark. Combining both light and vibration yielded piezo-photocatalysis, a process that worked day or night.

During tests, when light and vibration were used together, the filter removed about 99% of Congo Red and 74% of Methylene Blue in 90 minutes. It also partially cleaned real wastewater collected from a textile plant.

To understand its performance, the authors turned to machine-learning regression models. They fed the computer thousands of experimental data points, including dye concentration, catalyst amount, light intensity, and ultrasound frequency.

Modern algorithms such as random forests, XGBoost, and an artificial neural network learned how these factors interacted. The best models closely matched the experimental results, which they hadn’t seen, well enough to prove artificial intelligence could accurately forecast how fast the dyes vanished in different conditions.

“We are thinking of scaling up production and using the filter near treatment plants, where water bodies are regularly polluted,” Aviru Basu, INST scientist and corresponding author of the paper, said, adding that the team looks forward to its use in Jal Nigam and Namami Gange projects as well.

“Dr. Adreeja Basu, a plant biotechnologist and professor at Chandigarh University, is also helping us a lot in our efforts to make this product more sustainable using plant-derived products,” Dr. Aviru Basu added.