The aim of this study is to develop the first system based on chemical MOX sensors for monitoring ethanol on the skin after consumption. Thus, non-invasive measurements to monitor alcohol concentration and their correlation with blood and breathe Alcohol become possible. Transdermal alcohol emissions by perspiration have been investigated during clinical trials to demonstrate the relevance of this method. First, three commercial MOX sensors were used after calibration in respect of the thermodynamic conditions of the skin surface. Then, six volunteers have been selected for clinical trials with two cohorts of BAC target: 0.5 g/l and 0.8 g/l, and the three different sensors have been integrated in a wristband. We observed that the skin emits ethanol concentrations and we found a consistent correspondence between the kinetics of blood, breath and perspiration. MOX sensors can reliably estimate whether a drinker consumed a low or significant amount of alcohol. After validating this method, we developed our sensors based on tin dioxide thin sensitive layers with thickness of 50nm.Sensitive properties of the films to ethanol were studied in the conditions of the skin surface.

Gas concentration measurements by means of metal oxide microsensors represent a promising issue due to several advantages (size, low cost, power consumption, reliability…).However, improvements are required to increase performances of complete experimental systems including microsensor and testing chamber at least. This paper deals with the study of different size and shape configurations of gas testing chamber, by coupling 3D unsteady modelling and experiments in the case of a SnO 2 sensor with ethanol gas flow. The influence of the testing-chamber design on the gas flow hydrodynamics and on the system response is shown. A new 3D-printed prototype chamber (boat-shape design), as compared to the commonly used testing chamber (cross-shape design), leads to an increase of the dynamics, an enhancement of the gas concentration homogeneity and a significant reduction of flow recirculation and dead volumes. In this work we have shown that the optimization of the test chamber (volume and shape) makes it possible to get as close as possible to the real electrical characteristics of the sensor. Consequently thanks to these new achieved characteristics, the performances of the whole system are improved.