Abstract: In this article, a thorough 3D numerical simulation using COMSOL Multiphysics of a TGS1820 gas sensor for acetone detection in human breath is presented. The sensor works on the basis that acetone adsorption causes a change in the conductivity of tin dioxide (SnO₂). The model simulates gas diffusion and electrical reaction by combining the mechanics of electric currents with transport of diluted species. A three-dimensional architecture was created that included an air domain, platinum electrodes, and a SnO₂ sensing layer. A breath pulse (one to three seconds) was used to apply acetone concentrations ranging from 0.5 to 3.0 ppm. According to the results, the concentration of acetone on the SnO₂ surface peaks at 0.022 mol/m³, which results in a voltage drop at the output electrode from 2.3 V to 1.8 V, With a sensitivity of 500 mV/ppm and R2 = 1.00, the sensor displays a linear response. Acetone diffusion from the inlet to the sensor surface is visible in the 3D visualization, with streamlines displaying flow patterns. The TGS1820 sensor design for non-invasive diabetes monitoring applications is validated by this COMSOL-based simulation.

Keywords: COMSOL Multiphysics, TGS1820, Acetone Detection, Gas Sensor, Breath Analysis, Diabetes Monitoring, Numerical Simulation, SnO₂ Sensor, Electric Currents, Transport Physics.

Downloads: | DOI: 10.17148/IJIREEICE.2026.14359