PARALLEL-PLATE CAPACITIVE SENSOR FOR SCREENING METHANOL ADULTERATION IN CACHAÇA

Abstract

This paper reports the design, construction, and characterization of a low-cost parallel-plate capacitive sensor intended for screening gross methanol adulteration in cachaça. The transducer, built from copper-clad boards held by a 3D-printed PLA frame, operates without direct fluid–electrode contact: samples are sealed in low-density polyethylene pouches and inserted between the plates. Coupled to an NE555 timer in a stable configuration, the sensor converts changes in the effective permittivity of the medium into oscillation frequency, which is read by a microcontroller through pulse counting. Twenty-eight samples covering 0–30% (v/v) methanol in four commercial cachaças were measured between 15 °C and 26 °C, yielding 2,014 readings. Methanol addition consistently increased the oscillation frequency, consistent with the partial replacement of the high-permittivity aqueous fraction of the beverage by methanol. Univariate linear calibration, however, proved insufficient (best R² = 0.49): the signal is confounded by a thermal drift of up to −273 counts/°C and by baseline shifts of about 22% caused by dissolved sugars. The results support the hardware as a viable, low-cost data-acquisition platform for adulteration screening and characterize the interferences that any quantification model must compensate, indicating multivariate approaches as the natural next step. The system's selectivity limits and the gap between the tested concentrations and regulatory limits are explicitly discussed.