Organic solar cells (OSCs) are of interest because the technology offers a significant opportunity to reduce the overall costs of solar energy. OSCs can be very inexpensive to produce given that they rely on non-commodity materials and can use existing manufacturing techniques that are not labour- and capital-intensive. In this research, a financial model, named TEEOS (Technological and Economic Evaluator for Organic Solar), is developed and is used to determine financial indicators, such as simple payback period. These indicators are used to determine technical objectives for the OSCs. Two sample cells are evaluated in Toronto, Canada using historical data. The results show that the cell with a higher efficiency and wider absorptive wavelength range produces a payback period of approximately nine years, while the other cell has a payback period well over 45 years. Stochastic modeling techniques are also used to better replicate electricity price and weather fluctuations.

Increasing global energy demands and diminishing supplies of conventional fuels are forcing the world to focus more on alternative power sources that are both renewable and ecologically benign. Solar energy is clean, regularly available and can be harvested without sacrificing valuable land space. Due to the associated cost of solar cells, however a very small portion of the world’s energy needs are supplied by the sun. Solution-processable organic photovoltaics (OPVs) offer the promise of lower production costs relative to conventional (silicon) solar cell technology. Solution-processing can be performed using reel-to-reel manufacturing, with printing and coating techniques that are significantly cheaper than current processing methods for inorganic semiconductors. Although OPV efficiency values currently remain inferior to those of conventional solar cells, the rate of improvement is much higher in OPVs than in other solar cell technologies. Recently an efficiency exceeding 10% was reported for organic solar cells. An important difference between organic and conventional solar cells is the charge carrier mobility of the semiconductors, which tends to be relatively low in organic semiconductors. Recent advances in molecular design have led to polymer semiconductor materials that possess hole mobility values similar to that of amorphous silicon. The present study investigates potential improvements in OPV devices that can be achieved through the application of high hole mobility polymer semiconductor donors. Two diketopyrrolopyrrole-based polymers, PDQT and PDBFBT, were selected for the role of electron donor in OPV devices due to their high mobilities and their optimum optical and electrical properties. Optimization of the process parameters was performed using PC61BM as the acceptor. A relatively high quantity of PC61BM (3 - 4 × the weight of the donor) is required in the donor-acceptor blends of both polymers in order to balance the high hole mobility. For these donor-acceptor blends, a solvent system consisting of chloroform/ortho-dichlorobenzene (4:1 v/v) is necessary for proper solubility, and an additive, 1,8-diiodooctane, is required to achieve an acceptable morphology. The main benefit expected from the use of high mobility semiconductors is reduced charge recombination. This was studied in relation to the active layer thickness in standard and inverted OPV devices prepared using PC61BM as the acceptor. Normally the thickness of the active layer is required to be low (~100 nm) due to the poor charge transport mobility of the carriers. In this study, rather consistent power conversion efficiencies were achieved throughout a wide range of active layer thicknesses (~100 nm to ~800 nm). A comparison between standard and inverted device configurations demonstrates that the inverted configuration is more suitable for achieving thicker active layers when a high hole mobility donor is used. This is attributed to the longer hole collection path in the inverted structure, which can benefit from using a high hole mobility material. Increasing the absorption spectra of the donor-acceptor blend was studied by substituting PC71BM for PC61BM. The improved absorption leads to greater charge generation. In PDQT devices, the increase in absorption that is contributed by PC71BM appears to be of greatest benefit when active layers are not very thick. Therefore, when thick active layers (>500 nm) are required, the use of PC61BM is sufficient, in conjunction with a high mobility donor. Finally, an increase in a polymer’s crystallinity can often lead to greater mobility. This can be accomplished through various annealing techniques. The improved crystallinity of PDBFBT that occurs as a result of thermal annealing was studied in OPV applications. Although hole mobility of PDBFBT in the lateral direction improves with thermal annealing, mobility in the vertical direction decreases with increasing temperature. This suggests that the crystallinity of PDBFBT is oriented in the lateral direction as opposed to the vertical direction, thereby directing charge flow horizontal to the surface. With thermal annealing, an optimal amount of PC61BM added to PDBFBT can increase the vertical mobility to fairly high values. Nevertheless, the efficiency of standard and inverted OPV devices decreases with increased annealing temperature. This is attributed to agglomeration of PC61BM that occurs from an increase in annealing temperature. The results of this study demonstrate that thermal annealing is not beneficial for PDBFBT:PC61BM films in OPV applications due to the vertical orientation of devices. All of the studies presented in this work involve the use of high hole mobility polymer semiconductors as donor materials for OPV applications. This work will provide a deeper understanding of the properties required for the development of new semiconductor materials in OPV applications. Furthermore, this work will be very useful for the design of device structures for more feasible manufacturing of large area OPV devices via high speed roll-to-roll printing processes.

The objective of this thesis is to understand and improve optical and electrical confinement to achieve cost-effective high-efficiency thin p-type Si solar cells. Optical confinement is achieved by front surface texturing in conjunction with an internal reflective layer on the back surface. Electrical confinement is obtained through the use of a high-lifetime material coupled with high-quality passivation on both surfaces. This research is divided into five tasks. In the first task, Ga-doped Cz Si was investigated to achieve a high and stable lifetime. It was found that for 1 ohm-cm nominal-resistivity screen-printed Al-back surface filed (BSF) cells, the Ga-doped ingot gave ~1.5% higher absolute efficiency after light-soaking relative to the B-doped counterpart. The benefit of using Ga is therefore quite explicit. In the second task, the screen-printed Al-BSF was investigated to explore its potential and limitations for achieving high-efficiency cells. It was found that there exists a critical alloying temperature for a given Al-thickness, above which the Al-BSF becomes non-uniform and cell performance starts to degrade. This puts a limit on the quality of the Al-BSF that can be achieved. An alternative way of back passivation involving dielectric/metal layers was therefore explored. In Task three, two key requirements for achieving high-efficiency dielectric back-passivated cells were established through device modeling. These are (1) a formation of a high-quality BSF underneath the local back contact through vias in the dielectric and (2) a high-quality dielectric passivation with either a moderate positive charge density or a high negative charge density. Task four involved the development of a metallization technique through vias in the dielectric to achieve a high-quality contact and an efficient internal reflector in conjunction with a high-quality local BSF. Further, a novel dielectric system composed of a spin-on SiO<SUB>2</SUB> layer capped with SiN<SUB>x</SUB> was developed that exhibited excellent passivation and a moderate positive charge density. The final task involved fabrication and analysis of dielectric back-passivated cells. The new dielectric and process sequence developed in this thesis resulted in screen-printed solar cells with efficiency as high as 19% with the potential for 20% efficient cells on 100-µm thick Si substrates.

京都大学 / 新制・課程博士 / 博士(工学) / 甲第23901号 / 工博第4988号 / 新制||工||1779(附属図書館) / 京都大学大学院工学研究科材料工学専攻 / (主査)教授 田中 功, 教授 杉村 博之, 准教授 野瀬 嘉太郎 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

Photovoltaics (PV) is fast emerging as an attractive renewable energy technology due to concerns of global warming, pollution and scarcity of fossil fuel supplies. However to compete in the global energy market, solar cells need to be cheaper and more energy efficient. Silicon is the favorite semiconductor used in solar photovoltaic cells because of its ubiquity and established technology, but due to its indirect bandgap silicon is a poor absorber and light emitter. Thin film cells play an important role in low cost photovoltaics, but at the cost of reduced efficiencies when compared to wafer based cells. There remains much untapped potential in thin-film solar cells which this work has attempted to exploit through exploring novel approaches of enhancing the efficiency of thin film cells using the optical properties of sub-wavelength metal nanoparticles. Metals are considered as strong absorbers of light because of their large free-electron density. How can metals improve light trapping in solar cells? This question has raised several eyebrows and this thesis is an attempt to show that metal nanoparticles can be useful in producing efficient solar cells. Subwavelength metal particles support surface modes called surface plasmons when light is incident on them, which cause the particles to strongly scatter light into the underlying waveguide or substrate, enhancing absorption. The process of coupling thin film silicon waveguide modes to plasmonic metals using unpolarised light at normal incidence is applied to silicon-based solar cells and light emitting diodes, and enhanced photocurrent and electroluminescence is realized with potential for further optimisation and improvement. The results from this study correspond to a current increase of up to 19% from planar wafer based cells and up to 33% increase from 1.25 micron thin-film silicon-on-insulator structures for the AM1.5 global spectrum. We also report for the first time an up to twelve fold increase in electroluminescence signal from 95nm thick light-emitting diodes. From the results we conclude that this method which involves simple techniques of nanoparticle deposition and characterization could hold important implications in the improvement of thin-film silicon cell absorption / emission efficiencies where conventional methods of light trapping are not feasible, resulting in promising near-term applications of surface plasmons in photovoltaics and optoelectronics.

In polycrystalline silicon (pc-Si) thin-film solar cells, defect passivation is critical to device performance. Isoelectronic or covalently bonded impurities, hydrogenic, extended defects and defects with localized levels in the bandgap (deep level defects) are typically introduced during the fabrication of, and/or are inherent to, pc-Si thin-film solar cells. These defects dramatically affect minority carrier lifetimes. Removing and/or passivating these defects is required to maximize minority carrier lifetimes and is typically done through thermal annealing and passivation techniques. For pc-Si thin-film solar cells on low temperature glass superstrates, rapid thermal annealing (RTA) and hydrogen plasma passivation (hydrogenation) are powerful techniques to achieve effective removal and passivation of these defects. In this thesis, three silicon thin-film solar cells structures on low-temperature glass are subjected to variations in RTA high-temperature plateaus, RTA plateau times, and hydrogen plasma passivation parameters. These solar cells are referred to as ALICIA, EVA and PLASMA. By varying the RTA plateau temperature and time at plateau, the trade-off between extensive dopant diffusion and maximum defect removal is optimized. To reduce the density of point defects and to electrically activate the majority of dopants, an RTA process is shown to be essential. For all three of the thin-film solar cell structures investigated in this thesis, a shorter, higher-temperature RTA process provides the best open-circuit voltage (Voc). Extensive RTA plateau times cause excessive dopant smearing, increasing n = 2 recombination and shunt resistance losses. Hydrogenation is shown to be an essential step to achieve maximum device performance by `healing' the defects inherent to pc-Si thin-film solar cells. If the hydrogen concentration is about 1-2 times the density of oxygen in the cells as measured by secondary ion mass spectroscopy (SIMS), the cells seem to respond best to hydrogenation, with good resultant Voc and short-circuit for all cells investigated in this thesis. The effect of hydrogen passivation on the Voc is spectacular, typically increasing it by a factor of 2 to 3.5. Hydrogen de-bonding from repeated thermal treatments at increasing temperature provides a deeper understanding of what defects exist and the nature of the defects that limit the cell voltage. The variation in RTA and hydrogenation process parameters produces significant empirical insight into the effectiveness of RTA processes for point defect removal, dopant activation, point defect and grain boundary passivation, and impurity passivation. SIMS measurements are used to determine the impurities present in the cells' bulk and the amount of hydrogen available to passivate defects. From the results presented it appears that pc-Si thin-film solar cells on low-temperature glass are a promising, and potentially lower-cost, alternative to Si wafer based cells.

The cost of crystalline silicon solar cells must be reduced in order for photovoltaics to be widely accepted as an economically viable means of electricity generation and be used on a larger scale across the world. There are several ways to achieve cost reduction, such as using thinner silicon substrates, lowering the thermal budget of the processes, and improving the efficiency of solar cells. This thesis examines the use of plasma enhanced chemical vapour deposited silicon nitride to address the criteria of cost reduction for n-type crystalline silicon solar cells. It focuses on the surface passivation quality of silicon nitride on n-type silicon, and injection-level dependent lifetime data is used extensively in this thesis to evaluate the surface passivation quality of the silicon nitride films. The thesis covers several aspects, spanning from characterisation and modelling, to process development, to device integration. The thesis begins with a review on the advantages of using n-type silicon for solar cells applications, with some recent efficiency results on n-type silicon solar cells and a review on various interdigitated backside contact structures, and key results of surface passivation for n-type silicon solar cells. It then presents an analysis of the influence of various parasitic effects on lifetime data, highlighting how these parasitic effects could affect the results of experiments that use lifetime data extensively. A plasma enhanced chemical vapour deposition process for depositing silicon nitride films is developed to passivate both diffused and non-diffused surfaces for n-type silicon solar cells application. Photoluminescence imaging, lifetime measurements, and optical microscopy are used to assess the quality of the silicon nitride films. An open circuit voltage of 719 mV is measured on an n-type, 1 Ω.cm, FZ, voltage test structure that has direct passivation by silicon nitride. Dark saturation current densities of 5 to 15 fA/cm2 are achieved on SiN-passivated boron emitters that have sheet resistances ranging from 60 to 240 Ω/□ after thermal annealing. Using the process developed, a more profound study on surface passivation by silicon nitride is conducted, where the relationship between the surface passivation quality and the film composition is investigated. It is demonstrated that the silicon-nitrogen bond density is an important parameter to achieve good surface pas-sivation and thermal stability. With the developed process and deeper understanding on the surface passivation of silicon nitride, attempts of integrating the process into the fab-rication of all-SiN passivated n-type IBC solar cells and laser doped n-type IBC solar cells are presented. Some of the limitations, inter-relationships, requirements, and challenges of novel integration of SiN into these solar cell devices are identified. Finally, a novel metallisation scheme that takes advantages of the different etching and electroless plating properties of different PECVD SiN films is described, and a preliminary evalua-tion is presented. This metallisation scheme increases the metal finger width without increasing the metal contact area with the underlying silicon, and also enables optimal distance between point contacts for point contact solar cells. It is concluded in this thesis that plasma enhanced chemical vapour deposited silicon nitride is well-suited for n-type silicon solar cells.

Thin film silicon on low temperature glass substrates is currently seen as the best path toreduce the $/W cost of photovoltaic (PV) modules. However, producing thin film polysilicon, on glass, is an ongoing research challenge. Laser crystallisation of a-Si is one of the possible methods. Typically excimer (XMR) lasers are used for laser crystallisation. This thesis introduces the copper vapour laser (CVL) as a viable alternative for thin film photovoltaic applications. The CVL, like the XMR, is a high powered, pulsed laser. However, the CVL has higher pulse rates (4-20kHz), better beam quality and a visible wavelength output (578 & 511nm). Preliminary experiments, using 600K-heated silicon-on-quartz samples, confirmed that CVL crystallisation can produce area weighted average grain size of 0.1-0.15??m, which is comparable to results reported for XMR??? s. Importantly, the CVL results used thicker films (1??m), which is more applicable to thin photovoltaic devices that need 1-10??m of silicon to be viable. The CVL??? s longer wavelength and therefore longer penetration depth (1/alpha) are proffered as the main reason for this result. Extensive laser-thermal modelling highlighted further opportunities specific to CVL crystallisation. Through-the-glass doublesided irradiation was shown in simulations to reduce thermal gradients, which would enhance crystal growth. The simulations also produced deeper melts at lower surface temperatures, reducing the thermal stress on the sample. Subsequent experiments, using silicon-on-glass, confirmed the benefit of through-the-glass doublesided irradiation by maintaining grain sizes without the usual need for substrate heating. Furthermore, Raman analysis showed that doublesided crystallisation achieved full depth crystallisation, unlike single side irradiation which produced partial crystallisation. A new mode of crystallisation, stepwise crystallisation, was also postulated whereby a series of CVL pulses could be used to incrementally increase the crystallisation depth into the silicon. Simulations confirmed the theoretical basis of the concept, with HeNe Raman spectroscopy and analysis of surface grain sizes providing indirect experimental support. The CVL??? s ability to crystallise thicker films more directly applicable to photovoltaic devices secures its viability as an alternative laser for photovoltaic applications. The through-the-glass doublesided irradiation and the stepwise crystallisation provide additional potential for increased process flexibility over XMR???s.

One of the poly-Si thin-film cells developed at the University of New South Wales (UNSW) is the EVA cell. In this work, SiN films for EVA cells as an antireflection/barrier coating were investigated. In addition, the effect of hydrogenation pre-treatment of solid phase crystallisation (SPC) on grain size and open-circuit voltage (Voc) was investigated. The SiN films deposited by PECVD were examined for uniformity of the thickness and the refractive index of the films across the position of the samples in the PECVD deposition system. A spectrophotometric analysis was used to determine these film properties. It was found that these properties were very uniform over the deposition area. Good repeatability of the depositions was also observed. A series of SiN film depositions by reactive sputtering were also performed to optimize the deposition process. Parameters adjusted during the deposition were nitrogen flow rate, substrate bias, and substrate temperature. By investigating the deposition rate, refractive index, and surface roughness of the films, the three deposition parameters were optimised. The effects of post SiN deposition treatments (a-Si deposition, SPC, RTA, and hydrogenation) on thickness and refractive index of both SiN films deposited by PECVD and reactive sputtering were investigated by using samples which have the same structure as the EVA cells. The thickness of the PECVD SiN films decreased about 6 % after all the treatments. On the other hand, the thickness reductions of the reactively sputtered SiN films were very small. The refractive index of the PECVD SiN films increased about 0.6 % after the treatments, whereas that of the reactively sputtered SiN films decreased 1.3 % after the treatments. As a possible method to improve the performance of EVA cells, hydrogenation of a-Si was investigated as a pre-treatment of SPC process. There were no obvious differences in the grainsize and the Voc of the EVA cells with and without the hydrogenation. Therefore it is likely that the hydrogenation pre-treatment of SPC does not have a beneficial effect on the performance of EVA cells.

The cost of photovoltaic electricity needs to be significantly reduced in order to achieve a high electricity market penetration. Thin-film solar cells have good potential to achieve such cost savings though (i) large-area deposition onto low-cost foreign substrates, (ii) more streamlined processing, (iii) monolithic cell interconnection, and very efficient use of the expensive semiconductor material. Polycrystalline silicon (poly-Si) on glass is a promising technology for the cost-effective large volume production of PV modules since it (i) makes use of an abundant raw material, (ii) is non-toxic, (iii) does not suffer from light-induced degradation, and (iv) does not rely on TCO layers. Usually plasma enhanced chemical vapour deposition (PECVD) is used for the layer formation. This thesis explores the use of e-beam evaporation as deposition method since it is potentially much faster and cheaper than PECVD. The resulting solar cells are referred to as EVA (from EVAporation). Two inherent shunting mechanisms in EVA cells are demonstrated to be shunting through sub-micron sized pinholes when the back electrode is deposited and shunting between the emitter and the absorber layer at the glass-side electrode. Through the improved understanding of these shunting mechanisms it was possible to develop a suitable metallisation scheme for EVA cells using an aligned deposition of emitter and back surface field line contacts and a specially developed shunt mitigation etching technique. For the first time appreciable efficiencies of up to 5.2% were demonstrated on this material. It was also shown that only very lightly doped absorber layers can lead to the required high short-circuit currents in EVA cells. The resulting cells are currently completely limited by space charge region recombination occurring with comparatively low ideality factors of only ~ 1.4 This thesis also demonstrates the usefulness of Jsc-Suns measurements and investigates optical loss mechanisms in the current devices. Advanced modelling of distributed series resistance effects, influencing Suns-Voc, m-Voc and Jsc-Suns curves, is employed. PC1D modelling is used to extract relevant device parameters. In this work it was found that the diffusion length in the best EVA cells is longer than the absorber layer and that insufficient light trapping is currently the major hurdle to higher cell efficiencies. From the obtained results it can be concluded that EVA solar cells are promising candidates for the low-cost and high-volume production of solar modules.