Výzkum potenciálu kvantové simulace nanostruktur pro vývoj senzorů plynů

Abstract

Tato disertační práce představuje numerické atomistické modely struktur, konkrétně komponent pro detekci plynů, které využívají formalismus NEGF, tedy nerovnovážných Greenových funkcí. Výzkum se zaměřuje na různé materiály, jako jsou monokrystalické vrstvy oxidu zinečnatého, grafenové nanovrstvy a nanopásky a vícero molekul soli emeraldinu (polyanilin). Výsledky těchto modelů jsou porovnány s experimentálními údaji o detekci amoniaku, oxidu dusičitého a oxidu uhelnatého. Efektivní odpory (R-R0)/R0, vypočtené z I-V charakteristiky systému s adsorbovanými molekulami plynu a bez nich, pro více koncentrací plynu odpovídaly experimentálním výsledkům. Bylo zjištěno, že nasimulovaná citlivost oxidu zinečnatého na amoniak je asi šestkrát nižší než při experimentu, což naznačuje, že citlivost může mít smíšený původ jen asi z jedné šestiny ovlivněné čistým monokrystalem ZnO.

This thesis presents numerical models of atomistic devices, specifically gas sensing components, which employ the non-equilibrium Green's function formalism. The investigation focuses on various materials such as zinc oxide monocrystalline surfaces, graphene nanosheets, nanoribbon layers, and multiple molecules of emeraldine salt (polyaniline variety). The outcomes of these models are compared with experimental data of ammonia, nitrogen dioxide, and carbon monoxide detection. The effective resistances (R-R0)/R0, calculated from the I-V characteristic of the system with and without adsorbed gas molecules, for multiple gas concentrations corresponded with the experimental results. It was found that the numerical zinc oxide sensibility to ammonia is about sixth times lower than in the experimentation, which suggests that the sensibility may have a mixed background of about one-sixth influenced by pure ZnO material (monocrystalline). The results for graphene nanostructures indicate that the conductance is coupled with distortions, specifically the graphene layer edge. A higher response to disturbance was observed in the armchained geometry than in the zig-zag geometry. The most significant gas response was observed in the narrowest nanoribbon, where the gas molecules settled on the edge. The modeled sensors were most sensitive to NO2 and the least sensitive to CO. Finally, the core of this thesis, numerical polyaniline investigations, fit the measured values well and showed good properties as a gas sensor device for $NH_3$ detection and rather good for NO2 detection. The study on the polyaniline model contaminated with hydrochloric acid and sulfuric acid ions revealed that the addition of dilutants increased the sensor's resistance, causing a slight decrease in sensitivity, which was more pronounced in the case of sulfuric acid. Based on these findings, a recommendation for gas sensor production is provided.