Dottorato in TECNOLOGIE DELL'INFORMAZIONE E DELLE COMUNICAZIONI (ICT)

Titolo della tesi: Transparent selective contacts for silicon-based heterojunction solar cells

Renewable energies are nowadays recognised to be the protagonists of a necessary sustainable transition in the energy production process. Among the existing technologies, photovoltaic (PV) stands out for the infinite availability of the source, the direct efficient energy transformation and for low global energy costs. These reasons drive the global research toward materials and methods to obtain higher and higher efficiencies. Beyond the well-assessed silicon-based homojunction technology, that covers the 95% of the total PV production, the amorphous/crystalline-silicon heterojunction concept is gaining increasing market share. Heterojunctions (HJs) are able to overcome some shortcomings of homojunctions, such as the high thermal budget connected to the diffusion process to achieve the doping and to the recombination losses that limit the VOC. However HJs still require complex architectures and the use of toxic and dangerous gases, which demand costly measures for a safe industrial manufacturing. A new approach to circumvent some of the above-mentioned issues is currently explored with increasing interest, and involves the use of non-doped, non-silicon films to replace the amorphous-silicon ones. A great variety of materials has already been proposed and tested for the purpose, but some challenges still have to be completely solved, first of all the resistance to thermal annealing at 200-250°C, that is a necessary step in module assembling. Moreover, in view of an industrial application, the manufacture of these novel materials needs to satisfy several constrains and then guarantee an easy and high-throughput process. In this thesis the attention is focused on some of these novel materials for both emitter and base contacts of a heterostructure solar cell based on n-type doped crystalline silicon. In particular nickel oxide and titanium oxide or zinc sulphide are exploited as selective emitter and base contact respectively. These materials have to be more transparent than the doped amorphous layer counterpart to enhance the quantum yield of the solar cell in the blue region of the solar spectrum where the amorphous films introduce parasitic absorptions. Moreover the substitute layers must be deposited on the well-assessed thin buffer layer of amorphous silicon suboxide used in our laboratory to passivate the surfaces of the silicon wafer, therefore the most relevant constrain to fulfil is to avoid any contamination and any detrimental thermal stress of the surface passivation. On the basis of these requirements, particular care will be given to the choice of the deposition procedures as well as to the choice of the doping elements to be introduced into the selective contacts. The measurement of silicon surface passivation will be adopted as the most relevant method to evaluate the suitability of materials and manufacturing methods. Numerical simulations will also be used to evaluate the ideal band diagram of the proposed heterostructures and to verify how differences in the experimental properties of the materials with respect to ideality could affect the heterostructure device performances. In analogy with the doped amorphous layers used in conventional HJs, also the proposed novel materials for selective contacts need to be coupled to a transparent conductive oxide to make up for the poor lateral conductivity that still remains a basic issue of any selective layer proposed in the literature up to now. In this thesis another promising material is tested as an alternative to indium tin oxide, namely tungsten-doped indium oxide. It is found that non-reactive radio-frequency sputtering is suited to grow transparent films with very high mobility and low resistivity due a low carrier concentration that reduces parasitic absorptions. Most importantly, the sputtering damage to surface passivation of the crystalline silicon substrates can be completely recovered after a suitable thermal annealing. All the discussed results show that ideal materials or totally qualified deposition methods for obtaining alternative selective contacts have yet to be found, mostly from the point of view of the resistance to thermal treatments. This thesis adds some tiles to the road to high-efficiency, low-thermal budget, low-toxicity, high-throughput silicon-based heterojunctions.