Researchers at Sungkyunkwan University, a South Korean university, have built an electrochemical sensor that detects vitamin C at very low concentrations, and they used AI rather than repeated lab experiments to work out how to make it

2

1

. The study, led by Professor Jinsung Park of the Department of Biomechatronic Engineering, was published online November 9, 2026 in Chemical Engineering Journal

2

.

Why the usual way is slow

The sensor reached a calculated detection limit of 778 picomolar, a measure of concentration so low it corresponds to roughly a few hundred molecules of vitamin C in a liter of fluid

2

1

. The usual route is slow: conventional vitamin C analysis often needs sophisticated equipment, and building a sensitive electrochemical sensor normally means repeated trial-and-error experiments

2

1

.

The core of the sensor is a molecularly imprinted polymer, a material studded with recognition sites shaped to match the target molecule, which is what makes it pick vitamin C out of a sample

2

1

. Fabricating it well depends on several settings: the concentrations of the materials, the number of electrochemical polymerization cycles, the solution's pH and one more parameter the outlets do not name. Five possible values for each gives 625 possible combinations. The team trained its model on just 25 initial conditions, so the AI found good settings without the lab work that testing nearly all of them would have taken

2

1

.

Source: Phys.org

Source: Phys.org

What the numbers rest on

The team says the most significant part is that AI can spot promising fabrication conditions without researchers testing every combination in the lab

2

1

. The sensor also worked outside the lab. It performed promisingly in human serum and artificial urine, and it tracked urinary vitamin C changes after intake, correlating strongly with results from a conventional lab assay

2

1

. Both sets of results come from the researchers themselves. The detection limit is a calculated value, and the team calls the biological-sample findings promising; no independent validation has appeared

2

1

.

The work was funded by Korea's science ministry and the National Research Foundation of Korea

2

. The researchers hope to extend the approach beyond vitamin C, to sensors for disease markers and harmful pollutants

2

1

. "Our next goal is to expand the role of AI beyond optimizing fabrication conditions, enabling it to recommend the most suitable sensing materials for different target molecules," Daeil Park, the doctoral student who led the experimental work, said

2

.

© 2026 TheOutpost.AI All rights reserved