# Complementary metal-oxide-semiconductor (CMOS) time of evaporation measurement system for binary chemical monitoring

## TL;DR

A new CMOS-based system called ITEMS measures evaporation dynamics to monitor binary chemical mixtures with high accuracy and low cost.

## Contribution

ITEMS introduces a CMOS-based evaporation monitoring system with capacitive sensing and digital interfacing for real-time solvent composition analysis.

## Key findings

- ITEMS successfully monitored ethanol–water, methanol–water, and methanol–ethanol mixtures across various concentrations and temperatures.
- LOESS modeling outperformed linear regression in capturing nonlinear evaporation trends with lower RMSE values.
- The system requires less than 1 μL of sample volume and enables compact, low-cost chemical sensing.

## Abstract

Accurate, real-time analysis of binary liquid mixtures is essential in chemical sensing, especially for miniaturized, low-cost applications. We present a complementary metal-oxide-semiconductor (CMOS)-based platform—ITEMS (Integrated Time-of-Evaporation Measurement System)—designed to monitor binary mixtures via high-resolution capacitive sensing of evaporation dynamics. ITEMS employs an integrated capacitive sensor to detect time-resolved dielectric changes during droplet evaporation under controlled temperatures. By extracting features such as intermediate evaporation time (Δt₂), total evaporation time (ToE), and capacitance variation (ΔCap), ITEMS provides multidimensional insights into solvent composition. We validated the system across ethanol–water, methanol–water, and methanol–ethanol mixtures, with concentrations from 0 to 100% and temperatures between 25 °C and 60 °C. Our analysis reveals that evaporation time and dielectric response exhibit nonlinear dependencies on solvent concentration, particularly at elevated temperatures. Comparative modeling using linear regression and LOESS confirms LOESS’s superiority in capturing these trends, yielding lower Root Mean Square Error (RMSE) values across all datasets. The CMOS integration enables compact packaging, low sample volume requirements (< 1 μL), and direct digital interfacing via a microcontroller and graphical user interface (GUI). These results establish ITEMS as a robust, scalable platform for high-sensitivity, real-time chemical composition monitoring in environmental, biomedical, and industrial applications.

The online version contains supplementary material available at 10.1038/s41598-026-35322-x.