Bridging or in-Plane Oxygen Vacancies for a High-Performance Chemoresistive Gas Sensors? a First-Principles Study

Abstract

Introduction:The use of Density Functional Theory (DFT) simulations to predict the chemical-physical properties of nanostructured compounds and heterogeneous interactions (solid-gas) has become almost essential today, both to predict the properties of technologically advanced nanomaterials and to overcome the limits of experimental characterizations.Specifically, the use of DFT calculations can likewise be used in chemoresistive gas sensor applications, to investigate the physical-chemical properties of nanostructured semiconductors and their possible catalytic activity. Among the various gas sensing materials used, the Metal Oxide Semiconductors (MOXS) are the most investigated since the excellent sensitivity, good stability and low cost of production [1,2]. Nevertheless, besides all of these advantages they present also some drawbacks such as signal drift over the time, lack of selectivity and high operating temperature.Investigators have tried to address these problems in several manners, including the synthesis of solid solutions composed of different metal oxides, doping these materials with deferent kinds of elements and also with controlled concentrations of oxygen vacancies. The latter is a field still little explored, but preliminary experimental and theoretical works highlighted interesting results [3,4].When the sensor is exposed to a reducing (oxidizing) gas, the gas molecules interact with the chemisorbed oxygen species from the surrounding atmosphere and then decrease (increase) the resistance of a n-type based gas sensor, and vice versa for a p-type semiconductor. This change in resistance is due to the exchange of electrons between the surface and the conduction band. For a stoichiometric semiconductor, the electrons involved in this reaction that are in the conduction band are coming from the valence band after heating the material, in this case they need a high energy to blow up from the valence band to the conduction band. But when one introduces some defects, new energy level will be created and then they will be a new electrons source, thus the energy that the sensor needs to be thermoactivated decrease; that is energy that electrons need to jump to the conduction band. Scope and Methods: The purpose of this work is to investigate the impact of oxygen vacancies inside MOXS on their physical-chemical properties in general and on the electrons exchange in particular. Indeed, the creation of oxygen vacancies influences the physical-chemical properties of metal oxides. i.e. The bridging oxygen vacancies lead to the formation of energy level in the band gap region (Fig.1), these energy levels will behave as electrons donors that decrease the energy (ε1) needed for the electrons to be in the conduction band. While creation of oxygen vacancies from in-plane positions drives to the formation of impurity levels located slightly below the conduction band minimum (CBM), in this case the energy (ε2) that they need to jump to the conduction band is lower than ε1. Therefore, we can say that in-plane oxygen vacancies can overcome the disadvantage of high working temperatures for MOXS; as much as the electrons resource is close to the conduction band, as much as the ionization of these electron levels will be easy and then enhance the reactions (adsorption process) on the surface of the active element at low temperature.Tin dioxide is a typical n-type semiconducting material with a wide band gap of 3.6 eV. It has attracted the attention of many researchers because of its broad spectrum of physicochemical properties. It has also been used in several fields such as optoelectronic devices, electrocatalysis, ceramics and gas sensors. It is the semiconductor most studied as active element for the production of chemoresistive gas sensors. Then, it represents the best candidate for the proposed innovative work.In this work, DFT was used to compute the physical-chemical properties of the considered material. The Local Density Approximation (LDA) and the Generalized Gradient Approximation (GGA), which improve the LDA results taking into account the electron density gradient due to the non-homogeneity of the electron density, are the principal basis of the exchange-correlation functional within the framework of DFT. In this work, the GGA parameterized PBE (Perdew-Burke-Ernzerhof) has been used for structural properties prediction. This approximation has demonstrated, beside the LDA, their ability to predict the structural properties of materials. Nevertheless, their applications underestimate the band gap energy of materials compared to experimental results. To enhance the results concerning the band gap the Tran-Blaha’s modified Becke-Johnson exchange potential model (TB-mBJ) has been applied, this approximation has proved its ability to predict the band gap of materials.A series of first-principles studies has been carried out using the Full Potential Linearized Augmented Plane Wave (FPLAPW) method within the framework of the DFT, as implemented in the Wien2k code. This method is the most effective method to solve the Khon Sham equation.