First-principles Study On Defect Properties Of Low-symmetry Semiconductors

Point defects are critical factors in controlling the fundamental optical and electronic properties of semiconductors,so the optimization of device performance will largely depend on defect control.Currently,there are diverse experimental measurements that can characterize the defects in semiconductors,which can achieve some physical properties,e.g.,defect concentration and defect level.However,there usually exists only one defect among ten thousand atoms in a lattice,so an accurate prediction of the microscopic configuration of the point defect can be very challenging.In recent years,as the great improvement of the defect calculation with the aid of first-principles calculation,predicting the defect properties by theoretical calculation becomes a useful tool to study the defects in semiconductors.In the past,researchers usually focus on the materials with high-symmetry,e.g.,zinc-blende and wurtzite structures,and yet the related researches have almost reached saturation.Recently,some researchers began to study the physics in low-symmetry semiconductors,because there might be more novel properties as the structure and element become richer.Meanwhile,the number of the defect types will also increase in low-symmetry semiconductors.Consequently,it is necessary to study the defect properties when studying new low-symmetry semiconductors.Based on the background stated above,I made the related researches.1.We develop a code named SDASC(Semiconductor Defect Ab initio Simulation Code),which can calculate the defect properties of a semiconductor.This code has 5 modules,including the calculation of 1)phase stability and chemical potential;2)defect formation energy and defect level;3)self-consistent Fermi level calculation;4)simulation of photoluminescence(PL)lineshape;5)non-radiative transition rate.SDASC can automatically generate very rich defect configurations,and can call VASP code to perform atomic relaxations and electronic self-consistent calculations.Afterwards,SDASC can output the defect formation energies and defect levels,and it can also proceed with the module 3),4)and 5)based on the defect results.We have successfully applied for the computer software copyright of SDASC.2.We study the intrinsic defects in low-dimensional photovoltaic semiconductor Sb₂Se₃,finding the abnormal defect behavior,revealing the influence of the defects on photovoltaic performance,and proposing new schemes to improve the photovoltaic performance.Sb₂Se₃ is a binary semiconductor with a quasi-one-dimensional(Q1D)structure,which has attacted much attention in recent years.Although a certified efficiency of 9.2%was achieved,it is still much lower than that of mature photovoltaic materials,and the point defects are believed to be the critical factor that affects its photovoltaic performance.Hence,we study the defects in Sb₂Se₃ and find the defects are quite unexpected compared to those in conventional semiconductors.Through self-consistently solving the charge-neutrality equation based on the defect formation energies,we find as Se changes from poor to rich conditions,the concentration of V_Se firstly undergoes a decrease,then unexpectedly increase,which is apparently contrary to the simple chemical intuition.In order to understand the defect physics,we propose a new physical concept“Defect-Correlation”to explain the abnormal concentration increase.This concept can also be generalized for other charge-compensated semiconductors,e.g.,its sulfide counterpart Sb₂S₃.After identifying the deep-level defects,to quantitatively evaluate the influence of the non-radiative recombination on the efficiency of the Sb₂Se₃ solar cell,we calculate the electron and hole capture rate,and then propose a new scheme to assess the defect-recombination limited solar cell efficiency.As the concentration of V_Se exceeds 10¹⁶cm^-3,the efficiency of p-type Sb₂Se₃ solar cell will be limited to 17%;meanwhile we also find the detrimental effect caused by the recombination will be greatly weakened if a weakly p-type Sb₂Se₃ is used,but it also suffers from low open circuit voltage.Hence,it is important to find the scheme to overcome the bottleneck.We firstly test the doping effect of Pb.Unfortunately,Pb is not a good choice to enhance the photovoltaic performance.Although it can truly increase the doping concentration and then improve the p-type conductivity,it also leads to a higher concentration of V_Se,aggravating the recombination process.Since Sb₂S₃ is also a promising photovoltaic material,we also try to understand the fundamental electronic and defect properties of Sb₂(S,Se)₃ alloy,and find a high miscibility and linear band gap change.Sb-poor condition should be used to decrease the concentration of deep-level defects.3.We study the defects in chalcopyrite semiconductor ZnGeP₂,and find the origin of the infrared absorption that affects its opto-electronic properties is actually the anion-cation antisites.Zn Ge P₂ is a mature nonlinear optical material,but it is usually subject to the infrared absorptions in high-power devices.Up to now,the origins of the absorptions are still not very clear.Many researchers attributed it to the existence of cation disorders(Ge _Zn and Zn _Ge antisite)and vacancies.Howver,after exploring all possible defect configurations by SDASC,we show the anion-cation antisites Ge _P and P_Ge can be the dominant defects in Zn Ge P₂,where the concentrations can reach 10¹⁸ cm^-3.The importance of anion-cation antisites were neglected not only in the experimental studies in the past decades,but also in other II-IV-V₂semiconductors.Furthermore,by simulating the PL lineshape of the defect,we find Ge _Pand P_Ge are actually the defect origins of the experimentally-observed 1.4 e V and 1.6e V PL peaks,which challenges the experimental opinion that V_P should be responsible for the PL peaks.Although the defect concentration is high,the carrier concentration is quite low,implying that there is serious donor-acceptor compensation.Due to the strong correlation between different defects,it is quite challenging to suppress the defects by changing the growth conditions.Consequently,doping under non-equilibrium condition should be adopted to suppress the defects,so that the Zn Ge P₂-based optical device performance can be improved.4.We proposed two schemes to design defect-tolerant semiconductors.Defects can exist in semiconductors,so it is important to suppress the negative influence of the detrimental defects.Designing semiconductors with high defect-tolerance can satisfy this requirement.The first scheme is choosing the material with lone-pair element,to produce s-p hybridization in the valence band,similar to the role of p-d hybridization in Cu-based chalcopyrite semiconductors.Na Sb S₂ satisfies this requirement and its elements are earth-abundant and environmentally-friendly.We find it is a quasi-direct semiconductor with high absorption coefficient.S-rich condition can effectively suppress the deep-level defects meanwhile achieve a benign p-type conductivity.Consequently,we suggest it can be a promising p-type solar cell absorber.The second scheme is substituting the light element for heavy element in topological materials,which opens a“natural”band gap different from the inverted band gap.Taking the newly-discovered topological semi-metal Na₃Bi as a benchmark,we show Na₃P,Na₃As,and Na₃Sb have superior opto-electronic properties and high defect tolerance,i.e.,as long as Na-rich condition is used during growth,the concentration of deep-level defects can be negligible.Meanwhile,a large number of shallow donor defects can act as the effective conduction band edge,producing sufficient n-type carriers over 10¹⁹cm^-3.As a result,we propose they can be good thermoelectric materials.