Multiscale Modeling of Phosphorene-Based Sensing Devices for Volatile Organic Compounds

Sensing of volatile organic compounds (VOCs) not only in exhaled air but also in environmental air is essential in rapid diagnostication and in identifying potential hazards related to air quality, respectively. In particular, respiratory diseases like influenza and tuberculosis and other conditions like ketosis and diabetes can be investigated based on the detection of specific biomarkers in the exhaled products. At the same time, airborne pollutant contamination and air toxicity must be strictly and efficiently monitored. A practical and accurate VOC sensing device can be obtained using conductance changes induced in properly functionalized phosphorene active layers, which ensure specific binding sites for the targeted biomolecules. We develop here a multiscale approach that embeds an atomistic description based on density functional theory into a macroscopic transport model. In this way, we account for the electronic structure modifications induced in the active layer by the attached biomarkers with different configurations, which are extracted from ab initio molecular dynamics simulations. For the description of the device operation at the macroscopic level, we introduce a statistical model, which takes into consideration the proportion of the binding sites in the pristine active layer and the fractions of attached molecules. Perturbing factors like carbon dioxide, nitrogen, oxygen, and water molecules are investigated, and biomarker detection limits are evaluated. To improve the specificity of the biosensor, multiple sensing elements are employed, differently customized by transition metal impurities, which enhance the biomarker recognition capability. This allows the precise determination of the proportions of the molecules adhered to the device with the accuracy strictly dependent on the measurement resolution. Our results indicate that acetone and cyclohexanone detection is possible down to a few tens of ppm.