First-principles Studies On Photovoltaic Semiconductor Materials And Cu-based Memory Materials

This thesis is mainly constituent of two parts of work:the first part is about photovoltaic semiconductor materials and we will introduce the research and analysis routines for the fundamental properties of semiconductors, as well as the chemical trends of semiconductor properties as the constituent elements varies. Then we will turn to the importance of semiconductor alloys in tuning semiconductor properties and introduce two common ways of forming alloys, i.e., one based on isovalent element substitution and the other based on nonisovalent element mutation. By specific systems, we will see the key role that alloys play in improving the properties of photovoltaic semiconductors, searching and de-signing new and novel photovoltaic materials. After studying properties of these perfect semiconductor systems, we will focus on defect systems and introduce the research routines of defect properties of a system and the external effects on defect properties. The second part is work on Cu-based memory materials and will introduce two materials that might have potential applications for memories, that is, Cu2OSeO3, a multiferroic material, and Cu2O, a binary transition oxides.the specific contents of each chapters are as follows:In the first chapter, we will give a simple introduction of the two research hot spots in the current material fields:photovoltaic materials for solar cells and new generation materials for memory. After showing the importance of these materials in solving our social problems, we will introduce their working principles and already known fundamental properties, as well as the current status of their research.The second chapter will give a brief introduction of the basic theories of first-principles calculations, including the ideas and approximations of transforming a many-body problem to a single-electron problem, and the density functional theories, etc.The first part of our work involves chapter3,4.5, and6. In chapter3, we have studied properties of three Ⅱ-Ⅵ tellurides, i.e., MgTe, ZnTe, and CdTe, including their structure properties, electronic structure properties, and band offsets. Then the properties of their random alloys are investigated, based on the special quasi-random structure theory. The bowing parameters and structural stabilities of random alloys are given and (Mg, Cd)Te random alloys are found to be good for improving or expanding the applications of CdTe-based photovoltaic materials. In chapter4, we have continued to propose another way of forming semiconductor alloys by element mutations based on the ideas of band structure engineering and apply this to Si. A novel alloy between Si and AlP, SisAlP, is studied, including its structural stabilities, electronic structures and optical properties. This new alloy is found to have better optical absorption properties than Si, which is explained by both its electronic properties and the band align-ments of Si3AIP and Si. In chapter5, we have systematically investigated the defect properties of MgTe, including the defect formation energies and transition energy levels of intrinsic defects, impurities, and defeet complexes. The defect compensation effects in this system are also considered carefully and the most, suitable acceptors and donors are suggested. In chapter6, we have studied the relationships between the transition energy levels of VCu and CuZn and lattice constants in Cu2ZnSnS4and Cu2ZnSnSe4. The Vcu acceptor level is found to increase as the lattice is expanded while the CuZn level is found to decrease. We have explained these trends by considering pressure effect and p-d coupling effect.The second part of our work involves chapter7and8. In chapter7, we have studied the multiferroic properties of Cu2OSeO3. By starting from the spin Hamiltonian. we have designed and developed methods, named four-state energy mapping methods, to derive all the spin interaction parameters from first-principles calculations and these parameters for this system are given. Strong DM interaction is found to exist in Gu2OSeO3. leading to the helical ground spin state instead of ferrimagnetic state. Besides, the helical state is found degenerate, that’s, independent of the directions of the spin propagation vector. So under certain external conditions, i.e., magnetic field or temperature, different helical states tend to coexist or couple with each other, leading to the observation of skrymions in this system. Learning from the idea of cluster expansions, we have developed and improved the theory of spin-induced ferroelectricity. According to our polarization model, the spin-induced polarizations come from the sum of single-spin contributions and spin-pair contributions, omitting the higher orders. The polarization coefficient matrix is derived from our four-state energy map-ping methods. Based on the symmetry analysis, the polarization in this system is found to be mainly from single-spin contributions. Our polarization model, di-rect DFT calculation results, and the experimental results agree with each other well and prove the generals and Tightness of our model. In chapter8. we have studied the switching mechanism of Cu2O-based resistive random access mem-ory. Different from other binary oxides, the most popular defect in CU2O is Vcu By studying the migration barriers of Vcu m Cu2O and at the interface between Cu2O and Cu, we have proposed our switching mechanism based on the filament formation and broken by VCu.which is in good agreement with the experiments.