Theoretical Investigations On Defects And Optoelectronic Properties Of Cu-based Thin Film Solar Cell Materials

Because of the limited resources on earth, especially the non-renewable fossil fuels are running out, solar energy, as a kind of clean and renewable energy, of which the full use has paved the way for solving the energy crisis. Solar cell can convert the solar energy into electricity. Improvement of the conversion efficiency has become a widely investigated target. Cu(In, Ga)Se2 (CIGS) thin film solar cell, as one of the commercialized thin film solar cell, has reached an efficiency of 22.6 %?However,since Cu(In,Ga)Se2 contains rare elements In and Ga, there must exist an obstacle for its further widely application, and there are many people doing reserches on its substitute Cu2ZnSn(S,Se)4 (CZTSe). CZTSe has much richer elements, correspondingly, more defects and complex defect interactions. It will be of great importance to investigate the intrinsic and extrinsic defect properties to further improve the solar cell performance. At the same time, intermediate band solar cell, as one of the third generation high efficiency solar cells, has a theoretical limit conversion efficiency of 47%, higher than that of Schockley-Queisser 31%. CuGaS2, having the same chalcopyrite structure with CIGS,has attracted much attention as a potential intermediate band host material. One way to obtain the intermediate band material is doping. Therefore, one of the theoretical effort is to select appropriate doping elements and study its feasibility. In our paper, with the hybrid density functional, the defect and optoelectronic properties of the mentioned two kind materials are studied against the key problem of efficiency. The main contents have been summarized as follows:1.An investigation of alkaline earth element Na-related defects in solar cell ma-terial CZTSe. We study the formation energy, charge transition levels of Na-related defects,and migration paths of Na in CZTSe as well. We find that NaCu has the lowest formation energy, which means that Na can easily occupy the Cu site. Whereas, NaCu is an equvalent replacement, having no effect on the electrical properties of the material.NaZn has a shallow charge transition level above the valence band maximum (VBM),therefore, may contribute to the hole concentration. NaSn is a deep level defect, which,however, can be inhibited by the Sn-rich growth environment. Besides, the migration properties suggest that it is easy for Na to migrate in the material as intersitial Na and the Cu vacancy mediated mechanism. Our work can supply some explanations for the experimental observations, such as alkaline earth element doping can improve its p-type conductivity and the film quality.2.Investigation of single element doped CuGaS2 for intermediate band material.We investigate the transition metal (Fe, Co and Ni) and IVA element Sn doped CuGaS2,respectively. The results show that Fe and Ni doping can introduce empty band in the bandgap of CuGaS2. The IBs, contributed by their 3d orbitals, are indeed able to enhance the optical absorption. But Co introduces empty states overlaping with valence band. Therefore, Co is not appropriate to achieve IB material. In addition, Sn doping can also introduce intermediate bands, contributed by Sn-s orbital, in the bandgap and enhance the light absorption. Whereas, the chemical potential stability analysis show that Sn doping will reduce the chemical potential stability region of CuGaS2. In order to avoid the formation of SnS and SnS2 secondary phases, it is necessary to control the amount of Sn to be below 50% of Ga.3.Investigation of the same VA group element n-p co-doped CuGaS2 for interme-diate band material. Simultaneously substitution of the Ga and S site with the same element (N, P, As or Sb) will lead to a non-compensated n-p codoping, which can en-sure the appearance of IB, and at the same time, reduce the possible defects brought by doping. The results show that IB can be introduced into the band gap by all the el-ement codopings and enhance the light absorption. But the N doping will lead to the formation of GaN, with no stable chemical potential region. Conversely, P, As and Sb are possible, where P is the best element, since the co-doping of P in the Ga and S sites has the lowest formation energy in the corresponding stable region. This work designs a method of one element n-p codoping to realize IB materials, and investiges the fea-sibility of this method. This method can aslo be used in other functional materials to achieve controlable doping.4.Investigation of defect physics in CuGaS2:SnGa intermediate band material:insights from an optimized hybrid functional. In addition to studying whether doping elements can introduce intermediate bands and enhance light absorption, the interac-tion of dopant with native defects, as well as the effect on material properties is very important as well. Therefore, we use Sn doped CuGaS2 as a case to study the inter-action between SnGa defect and its intrinsic defect with the optimized hybrid density functional HSE(0.26,0.08). The optimized hybrid density functional can reproduce the experimental bandgap and, at the same time, satisfy the generalized Koopmans' theo-rem. It has been used to calculate experimentally observed PL energy and the relative charge transition levels of CuGa and GaCu in CuGaSe2. Our study shows that SnGa is an amphoteric trap, with two largely different energy (+/0) and (0/-) charge transition levels, which can contribute electrons to the CB, and can also receive electrons from the CB, companioned by a radiative recombination. Therefore, the carrier lifetime is limited in such an intermediate band material. In addition, the increase of SnGa shifts the Fermi level up, resulting in the spontaneous formation of CuGa, and the mutual passivation pinned the Fermi level at 1.4 eV above the valence band maximum. At this point, SnGa is positive charged and CuGa is negative (-1, -2) charged. In this case, the electronic transition from the defect state to the CB can be achieved, but not from VB to defect state, therefore only the red light, not the green light, can be utilized. Our work explains,from the perspective of the defect physics, some experimental observations: why do not impurity intermediate band semiconductors work in solar cells as expected, such as why the light absorption in experiment is weak and there is no increase in conversion efficiency.