Intermediate Band Solar Cells: Simulations and Quantum Dot Studies

Abstract

Silicon solar cell technology is now a mature technology and competitive with other energy technologies in many energy markets worldwide. However, to have major contribution to the electricity generation higher energy conversion efficiencies are highly desirable. The intermediate band solar cell (IBSC) is one of the solar cell concepts that have been proposed to increase the efficiency significantly. IBSCs are based on materials where an additional, intermediate energy band (IB) is present inside the semiconductor. IBSCs were proposed as late as in 1997, and different research groups have attempted to realize them using various materials. So far, the device performance has been affected by losses due to defects in the materials, and in addition there is a risk that the IB itself will lead to increases losses due to so-called Shockley- Read-Hall (SRH) recombination. The objective of this thesis is two-fold: firstly to carry out a theoretical study of the impact of SRH recombination on the IBSC performance, and secondly to investigate experimentally the material system that has been proven to result in the desired improved performance, but only at low temperature and a concentrated light, namely InAs/GaAs quantum dots. It is now well known that this material is not suitable for use in IBSCs under normal operating conditions. However, since the next version of QDbased IBSC most likely will be a modification of InAs/GaAS QDs, we need to study this material system to establish a starting point for the materials development. In the first part of the work, the IBSC performance has been calculated using detailed balance analysis, and adding the effect of SRH recombination. The defect levels causing the SRH recombination have been placed energetically in the middle of the two subbandgaps of the IBSC, as well as at the same energy as the IB. The theoretical work is an extension of the work of a former PhD student in the group, Rune Strandberg, who developed a detailed balance model for (defect free) IBSCs, where so-called photofilling effects were included. With photo-filling, the IB is assumed to be empty in thermal equilibrium, and then the IB will be filled as light is absorbed and carriers are excited to the IB from the valence band. In many cases the performance of IBSCs is calculated assuming the IB is half-filled (by pre-filling), but that might be far from the filling in realistic materials. In the current thesis the performance of IBSCs with SRH recombination is therefore calculated for both pre-filled and photo-filled IBSCs. The main finding is that SRH recombination can be detrimental to the IBSC performance, similar to what has been observed experimentally by many groups, but not in all cases. Defects above the IB might aid the filling of a photo-filled IBSC, and might lead to “over-filling” of the IB if it is pre-filled. In many cases the SRH recombination has only a minor effect, especially for concentrated light. Also the case where states in the IB itself lead to SRH recombination does not have a dramatic effect on the performance, as long as the radiative recombination dominates. In the experimental, second part of the work, InAs/GaAs QDs have been grown using molecular beam epitaxy (MBE) and studied using atomic force microscopy (AFM), scanning electron microscopy (SEM) and photoluminescence spectroscopy (PL). QD samples with a wide range of growth parameters have been made and studied, with the aim to make ultra small, high density quantum dots. Small dots are needed to have only one confined state for electrons in the QDs, thus limiting the thermal escape out of the QDs. If thermal escape is possible, the quasi-Fermi level of the IB, made up of the confined electron states, will be coupled to the quasi-Fermi level of the conduction band (CB). With coupled quasi-Fermi levels, the cell would behave as a single junction, low bandgap cell, and not an IBSC. High density is needed to maximize the sub-bandgap absorption. With too low sub-bandgap absorption, there will be a too small increase in the photo-generated current, and the IBSC will have poor performance. The experimental part of the thesis shows, as is known from literature, that it is possible to grow small QDs of high density, but that the QD properties are very sensitive to all growth parameters. It also shows that improvement in one QD parameter often is accompanied with a decrease in another parameter, making the growth optimization very challenging. An important point that seems to be overlooked by many groups working with QD based IBSCs, is the high Auger recombination rates that can occur in QD materials, especially when illuminated with high intensity light. Nevertheless, the experimental work in this thesis has formed a basis for further development of QD materials for IBSCs based on the InAs/GaAs system.