Potential Oxide-Based Intermediate Band Materials: Fabrication and Photoemission Studies
MetadataShow full item record
- Institutt for fysikk 
Intermediate band solar cells (IBSCs) is a so-called third generation solar cell concept, and are based on a new type of semiconductor materials which have an intermediate band (IB) within the bandgap. This IB works as a stepping stone, allowing absorption of photons with energy less than the bandgap, making use of a broader range of sunlight compared to solar cells without this IB. IBSCs have a rather simple design compared to other third generation concepts like tandem cells and a theoretical efficiency 50% higher than conventional single junction solar cells. However, since it is a relatively new concept, there are still several challenges to overcome, such as to identify and fabricate sustainable intermediate band materials with the desired properties. Many of the explored intermediate band materials are based on toxic and/or non-abundant elements. In contrast, this thesis work focuses mainly on studying sustainable, meaning non-toxic and abundant, IB materials. The selected materials studied in this thesis work are transition metal oxides, such as molybdenum, titanium and cuprous oxides, MoO3, TiO2 and Cu2O, respectively. Photoemission spectroscopy (PES) is a powerful technique used to determine the surface composition and to understand the electronic structure of materials. Combined with Inverse photoemission spectroscopy (IPES), it can give insight not only on the occupied and unoccupied states in the valence band (VB) and conduction band (CB), respectively, but also determine the bandgap of the material in study. These techniques were used in this thesis work because they detect states in the band gap and, ultimately, can contribute towards the realisation of an intermediate band for IB materials. Two main strategies were attempted to realise an IB in the host oxide materials: incorporation of defects (vacancies and/or interstitials) during the materials fabrication and incorporation of dopants by ion implantation into undoped oxide samples. The defects makes the samples deviate from the oxide’s stoichiometric composition, and the non/sub-stoichiometric MoO3 and TiO2 samples were fabricated using pulsed laser deposition. States in the bandgap were detected by PES, for both molybdenum and titanium oxides. An accelerated ageing study of MoO3 was carried out, to test the material stability and lifetime upon X-ray/UV light exposure using synchrotron light, and heat. MoO3 turns into metallic MoO2 after approximately 100 days of sunlight if in a satellite outside atmosphere and, similarly, for a temperature range comparable to the conditions in which photovoltaics may be required to operate. In both cases, creation of oxygen vacancies is responsible and puts into question the use of unprotected molybdenum oxide for solar cell in space applications. The second work on MoO3 had emphasis on sub-stoichiometric samples, in which PES probed extra oxidation states besides lattice for the lowest oxygen flow growth, and also higher intensity for states in the bandgap. With sub-stoichiometry confirmed by PES, ellipsometry showed a large contribution of free carriers, with substantial amount of MoO2. These properties are also important to understand efficiency limit for a future IBSC device. For TiO2 samples, the identification of the crystalline phases was achieved through analysis of the measured PES VB shape: anatase, rutile or a mixed phase, using a new methodology only recently reported. Moreover, this work also uses binding energy differences between core levels and VB, to infer the dominating crystal phase present in each sample. The main findings include the possibility to control the crystal phase and the non-stoichiometry by a careful tuning of the TiO2 sample fabrication, and that PES techniques can be used to probe these material properties, and similar for the other materials. Moreover, PES was combined with optical techniques, to confirm the phase identification and to relate non-stoichiometry with sub-bandgap absorption. The third oxide-based material studied, cuprous oxide, Cu2O, was made by thermal oxidation of copper foil. Shallow and deep doping of Cu2O was achieved by low and higher energy ion implantation, respectively. For the former, an in-situ p-type doping method by N ion implantation was demonstrated to be robust by PES, meaning it withstands a wide range of temperatures, air and radiation. For the latter, Cu2O was single and codoped with nitrogen and potassium ions. PES and IPES were combined to fully probe states in the VB, CB and bandgap. It includes also Density Functional Theory calculations, to help better understand the changes dopants bring to the electronic structure of Cu2O. Findings include partial replacement of oxygen by nitrogen for N-doped Cu2O, detection of higher potassium concentration than expected at the surface for K-doped Cu2O, and presence of oxygen vacancies for all doped samples but not for undoped Cu2O, specially for K-doped and N,K-codoped. K-doping gives the most metallic behaviour, with a very narrow bandgap, and this can be related to oxygen vacancies which may act as electron donors. To conclude, this thesis work used two different methods in an attempt to make oxide-based IB materials. A set of candidate materials were fabricated and characterised using multiple techniques, with emphasis on PES-based techniques, and the results form a basis for a better understanding on how oxide-based IB materials with the desired properties can be fabricated.