Defect engineering in kesterite solar cells using cation substitution

• Main Author: Hadke Shreyash Sudhakar
• Other Authors: Lydia Helena Wong
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Materials
• Online Access: https://hdl.handle.net/10356/146400
• Institution: Nanyang Technological University

Cu2ZnSnS4 (CZTS) is a promising photovoltaic material due to its easy processability, earth-abundant constituent elements, high absorption coefficient, and a suitable bandgap. Further development of CZTS-based solar cells is currently limited by a low open-circuit voltage (VOC). Although there is a lack of consensus regarding the loss mechanisms contributing to this low VOC, it is generally agreed that point defects contribute to most of the VOC loss through mechanisms such as interface recombination, bulk recombination, bandgap fluctuations, etc. Hence, suppressing and passivating point defects is crucial for further development of CZTS solar cells. One of the proposed reasons for the presence of point defects in CZTS is the similarity in the ionic radii of Cu+1, Zn+2, and Sn+4 cations. Hence, the substitution of one of these cations with another isovalent cation having dissimilar ionic radius provides a way to incorporate structurally and chemically varied cations in CZTS. This concept is known as cation substitution and has been used to improve the performance of CZTS solar cells. However, the experimental evidence for the relationship between cation substitution – for example, the substitution of Zn with Cd, or that of Cu with Ag, etc. – and defect characteristics is scarce. In this thesis, the effect of cation substitution is studied using experimental and theoretical methods, with an emphasis on correlating the theoretically calculated defect characteristics with experimentally measured optoelectronic properties. The results obtained allow for the engineering of defect characteristics in CZTS using cation substitution. Specifically, it is shown how (i) the substitution of Zn with Cd thermodynamically suppresses 2CuZn+SnZn defects, (ii) the substitution of Zn with Cd kinetically reduces the CuZn+ZnCu disorder, and (iii) the double substitution of Cd and Ag reduces the depth of acceptor defects and increases the photoluminescence yield. These three results are discussed below. 2CuZn+SnZn is a neutral compensated defect cluster in CZTS that leads to a severe conduction band downshift. Hence, a large density of these defects can lead to electron trapping states and bandgap fluctuations. In Chapter 4, it is shown using theoretical calculations how the substitution of Cd can increase the formation energy of these defects. Further, it is also shown that the suppression of these defects is more prominent in the presence of VCu. Using optoelectronic characterization, it is also shown that the suppression of 2CuZn+SnZn correlates with longer carrier lifetime and reduced bandgap fluctuations. Finally, a Cu2CdSnS4 (CCTS) solar cell with 8% efficiency is fabricated, which is a record for novel quaternary materials based on Cu2ZnSnS4. CuZn+ZnCu cluster is among the most studied defects in CZTS. This defect cluster has low formation energy, leading to a large Cu-Zn disorder in the CZTS structure. It has also been reported in the literature that the formation energy of CuCd+CdCu defects is similar to that for CuZn+ZnCu defects. In Chapter 5, it is shown that Cd can reduce the Cu-Zn disorder even though the thermodynamic defect formation energy of CuZn/Cd+Zn/CdCu remains similar. This is because of the structural differences between CZTS and CCTS. CZTS adopts a kesterite structure at room temperature and undergoes an order-disorder phase transformation at ~520 K, while CCTS adopts a stannite structure at room temperature and does not undergo an order-disorder phase transformation at a specific temperature. Rather, the order in CCTS increases gradually with decreasing temperature. The nucleation and growth process required for the ordering of CZTS at a relatively low temperature of ~520 K is circumvented for CCTS, possibly leading to a reduced disorder. Finally, it is shown in Chapter 6 that multiple cation substitution can be used to address multiple performance limiting factors. Using partial substitution of Zn and Cu with Cd and Ag, the depth of the shallow acceptor defects is reduced, and the photoluminescence yield is improved. The improved defect characteristics lead to an efficiency of 10.1% (total area) for AgxCu2-xZn1-xCdxSnS4 device. The methods used in this thesis show that cation substitution can be used not only to engineer the defect characteristics but also to understand them.