Dissertations / Theses on the topic 'Thin-film solar cells'

Thin-film silicon photovoltaics are seen as a good possibility for reducing the cost of solar electricity. The focus of this thesis is the ALICIA cell, a thin-film polycrystalline silicon solar cell made on a glass superstrate. The name ALICIA comes from the fabrication steps - ALuminium Induced Crystallisation, Ion Assisted deposition. The concept is to form a high-quality crystalline silicon layer on glass by Aluminium Induced Crystallisation (AIC). This is then the template from which to epitaxially grow the solar cell structure by Ion Assisted Deposition (IAD). IAD allows high-rate silicon epitaxy at low temperatures compatible with glass. In thin-film solar cells, light trapping is critical to increase the absorption of the solar spectrum. ALICIA cells have been fabricated on textured glass sheets, increasing light absorption due to their anti-reflection nature and light trapping properties. A 1.8 μm thick textured ALICIA cell absorbs 55% of the AM1.5G spectrum without a back-surface reflector, or 76% with an optimal reflector. Experimentally, Pigmented Diffuse Reflectors (PDRs) have been shown to be the best reflector. These highly reflective and optically diffuse materials increase the light-trapping potential and hence the short-circuit currents of ALICIA cells. In textured cells, the current increased by almost 30% compared to using a simple aluminium reflector. Current densities up to 13.7 mA/cm2 were achieved by application of a PDR to the best ALICIA cells. The electronic quality of the absorber layer of ALICIA cells is strongly determined by the epitaxy process. Very high-rate epitaxial growth decreases the crystalline quality of the epitaxial layer, but nevertheless increases the short-circuit current density of the solar cells. This indicates that the diffusion length in the absorber layer of the ALICIA cell is primarily limited by contamination, not crystal quality. Further gains in current density can therefore be achieved by increasing the deposition rate of the absorber layer, or by improving the vacuum quality. Large-area ALICIA cells were then fabricated, and series resistance reduced by using an interdigitated metallisation scheme. The best measured efficiency was 2.65%, with considerable efficiency gains still possible from optimisation of the epitaxial growth and metallisation processes.

Kyoto University (京都大学)
0048
新制・課程博士
博士(エネルギー科学)
甲第14742号
エネ博第195号
新制||エネ||44(附属図書館)
UT51-2009-D454
京都大学大学院エネルギー科学研究科エネルギー基礎科学専攻
(主査)教授 八尾 健, 教授 石原 慶一, 教授 辻井 敬亘
学位規則第4条第1項該当

Dielectric thin films have a long history in silicon photovoltaics. Due to the specific physical properties, they can function as passivation layer in solar cells. Also, they can be used as antireflection coating layers on top of the devices. They can improve the back surface reflectance if proper dielectric layers combination is used. What??s more, they can protect areas by masking during chemical etching, diffusion, metallization among the whole fabrication process. Crystalline silicon solar cell can be passivated by two ways: one is to deposit dielectric thin films to saturate the dangling bonds; the other is to introduce surface electrical field and repel back the minority carriers. This thesis explores thermally grown SiO2 and sputtered Si3N4(:H) to passivate n-type and thermal evaporation AlF3 to passivate p-type Float Zone silicon wafers, respectively. Sputtering is a cheap passivation method to replace PECVD in industry usage, but all sputtered samples are more likely to have encountered surface damage from neutral Ar and secondary electrons, both coming from the sputtered target. AlF3/SiO2 multi-layer stack is a negative charge combination; p inversion layer will form on the wafer surface. Light trapping is an important part in solar cell research work. In order to enhance the reflectance and improve the absorption possibility of near infrared photons, especially for high efficiency PERL cell application, the back surface structure is optimized in this work. Results show SiO2/Ag is a very good choice to replace SiO2/Al back reflectors. The maximum back surface reflectance is 97.82%. At the same time, SiO2/Ag has excellent internal angle dependence of reflectance, which is beneficial for surface textured cells. A ZnS/MgF2/SiO2/Al(Ag) superlattice can improve the back reflectance, but it is sensitive to incident angle inside the silicon wafer. If planar wafers are used to investigate all kinds of back reflectors, and an 8 degrees incident angle is fixed for typical spectrometry measurement, the results are easy to predict by Wvase software simulation. If a textured surface is considered, the light path inside the silicon wafer is very complicated and hard to calculate and simulate. The best way to evaluate the result is through experiment.

Thin-film silicon solar cells have the potential to convert sunlight into electricity at high efficiency, low cost and without generating pollutants. However, they need to become more competitive with conventional energy technologies by increasing their efficiency. One of the key efficiency limitations of using thin silicon absorber materials relates to the optical loss of low-energy photons, because the absorption coefficient of silicon decreases strongly for these low-energy photons in the red and near-infrared, such that the absorption length becomes longer than the absorber layer thickness. If, in contrast, the incident light was redirected and trapped into the plane of the silicon slab, a thin-film could absorb as much light as a thick layer. Diffractive textures can not only efficiently scatter the low-energy photons, but are also able to suppress the reflection of the incident sunlight. In order to take advantage of the full benefits that textures can offer, I outline a simple layer transfer technique that allows the structuring of a thin-film independently from both sides, and use absorption measurements to show that structuring on both sides is favourable compared to structuring on one side only. I also introduce a figure-of-merit that can objectively and quantitatively assess the benefit of the structuring itself, which allows me to benchmark state-of-the-art proposals and to deduce some important design rules. Minimising the parasitic losses, for example, is of critical importance, as the desired scattering properties are directly proportional to these losses. To study the impact of parasitics, I quantify the useful absorption enhancement of two different light trapping mechanisms, i.e. diffractive vs plasmonic, based on a fair and simple experimental comparison. The experiment demonstrates that diffractive light-trapping is a better choice for photovoltaic applications, because plasmonic structures accumulate the parasitical losses by multiple interactions with the trapped light. The results of this thesis therefore highlight the importance of diffractive structures as an effective way of trapping more light in a thinner solar cell device, and will help to define guidelines for new designs that may overcome the 30% power conversion efficiency limit.

Cu₂S/CdS solar cells have been studied extensively for the past two decades due to their potentially high efficiencies per unit cost. The operation and characteristics of Cu₂S/CdS solar cells are fairly well understood. However, the properties of the newer Cu₂S/ZnCdS cell type are not well understood. The main goals of this thesis were to compare Cu₂S/CdS and Cu₂S/ZnCdS cells using Cu₂S/CdS cells as a reference, and to understand the operation and properties of Cu₂S/ZnCdS cells in order to improve cell performance. Four different measurements were used in this research to achieve these goals. They were; electrical, spectral, capacitance and deep trap measurements. I-V measurements give important electrical parameters of the cells; cell efficiency, fill factor, short circuit current, open circuit voltage, shunt resistance and series resistance are reported. From a In(I_SC) versus V_OC measurement, the diode factor, A, was found to be about 1 for Cu₂S/CdS, Cu₂S/Zn_0.11Cd_0.89S, and about 1.2 for Cu₂S/Zn_0.25Cd_0.75S cells. The relation between In(J_oo) (current density) and ϕ (potential barrier height) is linear for both types of cells. The slope of this linear relationship increases as the content of Zn increases in Zn_xCd_1-xS. Under air mass 1 (100 mW/cm²) illumination, it was found that V_OC decays and capacitance increases for Cu₂S/ZnCdS cells. This is attributed to electron relaxation from deep traps near the junction. Spectral response with and without bias light were measured for both Cu₂S/CdS and Cu₂S/ZnCdS cells. White and blue bias light enhance the spectral response, while red bias light quenches the response. This is attributed to ionization and filling of deep traps near the junction. Capacitance measurements on both cell types show that 1/C² versus voltage is quite flat, which indicates the existence of an i-layer (insulation layer) in the CdS or ZnCdS near the junction. Three methods–photocapacitance, space-charge-limited current, and thermally stimulated. current techniques–were used for deep trap measurements. Photocapacitance measurements indicate one deep donor energy and two deep acceptor energy levels. These trap energies become larger as the content of Zn in ZnCdS increases. Space-charge-limited current measurements give a trap density of the order of 10¹⁶ cm³ for both cell types. The shallow energy trap is found to be 0.26 eV below the conduction band edge of CdS. The occurrence of a current-saturated region for Cu₂S/ZnCdS is attributed to the filling of the interface traps near the junction. Thermally stimulated current measurements give two energy levels below the conduction band of CdS; 0.05 eV and 0.26 eV. From the above results, several differences between the Cu₂S/CdS and the Cu₂S/ZnCdS cells can be seen. The Cu₂S/ZnCdS cells show stronger red quenching, smaller electron lifetime at the interface near the junction, and deeper traps than the Cu₂S/CdS cells. These differences can account for the decline of I_SC and the V_OC decay. The smaller I_SC for the Cu₂S/ZnCdS cells can also possibly result from smaller electron lifetime at the interface, larger interface recombination velocity, different deep trap levels, and enhanced Zn concentration near the junction. The V_OC decay for the Cu₂S/ZnCdS cells is mostly due to long decay of charge. Longer decay could be attributed to deeper traps.
Ph. D.

The work presented in this thesis is concerned with photovoltaic cells formed by plating CdS single crystals and thin films, and Cd(_y) Zn(1 _ y)S single crystals, with copper sulphide. An electroplating technique has been used to control the phase of copper sulphide by changing the electric field during its formation. Different phases of Cu(_x)S have been identified directly using Reflection High Energy Diffraction (RHEED), and indirectly from spectral response measurements. A dramatic change in the spectral response accompanying the reduction in the covellite response associated with an increase in that from chalcocite following argon heat treatment has been achieved. The change from the djurleite phase to that of chalcocite has also been obtained by using argon heat treatment for 5 minutes at 200 C. This effect was found to be reversible in that layers of chalcocite were converted to djurleite when air was used as the ambient for the heat treatment. C-V measurements have demonstrated that with increasing plating bias the donor concentration decreases at first before it assumes a constant value. This led to the effect of decreasing the junction capacitance as the width of the depletion region changed. The problem of the stability of the CdS-Cu(_2)S photovoltaic devices formed by wet plating" is addressed by studying the combined effects of the substrate onto which the CdS is deposited and the ambient used during annealing. Thin film cells have been prepared on both Ag/Cr and SnO substrates, and the device characteristics for each have been investigated as a function of annealing ambient. The results have shown that devices formed on Ag/Cr substrates were more stable following annealing in air than in argon, while the converse was true for cells fabricated on SnO(_x) substrates. The degradation effects of CdS-Cu(_2) S photovoltaic cells have been investigated. While devices stored in the dark showed little or no degradation, those maintained under illumination exhibited a significant deterioration in all operational parameters over a four week period. As far as the combined effect of temperature and ambient on the stability of cells are concerned, it was found that the ageing of devices in argon at room temperature in the dark was negligible, and moreover the fill factor was observed to improve marginally. When the devices were stored in the same ambient conditions at 50 C, they showed a significant improvement in the fill factor, but simultaneously exhibited a considerable reduction in the short circuit current. This process was reversible, since the sensitivity of degraded devices could be restored by annealing them in a hydrogen/nitrogen mixture. By comparing Electron Spectroscopy for Chemical Analysis (ESCA) studies with solar cell device characteristics, it has been shown that the formation of copper oxide on the Cu(_2)S surface plays a significant role in the degradation of CdS-Cu(_2) S devices. The extent of the cross-over between the dark and light J-V characteristics is a function of the period of etching used prior to junction formation. The variation of current and diode factor has been established as a function of the bias value. The dependence of forward current on the temperature at fixed forward voltage has also been investigated. Finally this work has shown that an increase in V(_oc) can be achieved when Cd(_0◦8)Zn(_0◦2)S is used as a base material for solar cells instead of CdS. Different traps were identified through a photocapacitance investigation. An important trap was found at 0.78eV below the conduction band. It has been demonstrated that the effect of this level was found to be diminished much more slowly when the annealing was carried out in argon rather than in air. This level may play an important role in the Cd(0◦8) Zn(0◦2)S-Cu(_2)S solar cell properties.

Over the last decades the world has been experimenting an increasing pressure to find solutions to energy crisis issues. As a result, the scientific research has been boosted towards the development of solutions related to alternative energy sources. In this context, several research programs have been supported. Among them, the ENIAC Project Joint Undertaking, Energy for a green society: from sustainable harvesting to smart distribution, equipment, materials, design solutions and their applications, within which the present work has been performed. In particular, this thesis is focused on the optimization of photovoltaic cells, through the use of mathematics tools and optimization techniques based on new theories like the Genetic Algorithm ones. The obtained results are new and they represent a noticeable improvement in the optimization of solar cells design and efficiency.

This thesis discusses the growth of thin-film silicon layers suitable for solar cells using liquid phase epitaxy and the behaviour of oxide LPCVD silicon nitride stacks on silicon in a high temperature ambient.¶ The work on thin film cells is focussed on the characteristics of layers grown using liquid phase epitaxy. The morphology resulting from different seeding patterns, the transfer of dislocations to the epitaxial layer and the lifetime of layers grown using oxide compared with carbonised photoresist barrier layers are discussed. The second half of this work discusses boron doping of epitaxial layers. Simultaneous layer growth and boron doping is demonstrated, and shown to produce a 35um thick layer with a back surface field approximately 3.5um thick.¶ If an oxide/nitride stack is formed in the early stages of cell processing, then characteristics of the nitride may enable increased processing flexibility and hence the realisation of novel cell structures. An oxide/nitride stack on silicon also behaves as a good anti- reflection coating. The effects of a nitride deposited using low pressure chemical vapour deposition on the underlying wafer are discussed. With a thin oxide layer between the silicon and the silicon nitride, deposition is shown not to significantly alter effective life-times.¶ Heating an oxide/nitride stack on silicon is shown to result in a large drop in effective Lifetimes. As long as at least a thin oxide is present, it is shown that a high temperature nitrogen anneal results in a reduction in surface passivation, but does not significantly affect bulk lifetime. The reduction in surface passivation is shown to be due to a loss of hydrogen from the silicon/silicon oxide interface and is characterised by an increase in Joe. Higher temperatures, thinner oxides, thinner nitrides and longer anneal times are all shown to result in high Joe values. A hydrogen loss model is introduced to explain the observations.¶ Various methods of hydrogen re-introduction and hence Joe recovery are then discussed with an emphasis on high temperature forming gas anneals. The time necessary for successful Joe recovery is shown to be primarily dependent on the nitride thickness and on the temperature of the nitrogen anneal. With a high temperature forming gas anneal, Joe recovery after nitrogen anneals at both 900 and 1000oC and with an optimised anti-reflection coating is demonstrated for chemically polished wafers.¶ Finally the effects of oxide/nitride stacks and high temperature anneals in both nitrogen and forming gas are discussed for a variety of wafers. The optimal emitter sheet resistance is shown to be independent of nitrogen anneal temperature. With textured wafers, recovery of Joe values after a high temperature nitrogen anneal is demonstrated for wafers with a thick oxide, but not for wafers with a thin oxide. This is shown to be due to a lack of surface passivation at the silicon/oxide interface.

The main objective of this thesis was to develop high efficiency thin film solar cells based on low-toxic and earth-abundant kesterite Cu2ZnSnSe4 (CZTSe) absorbers through the implementation of innovative doping strategies. Special focus is put on the optimization of reactive thermal processes, followed by the screening of possible doping elements and further analysis of the most promising ones. Additionally, deeper investigations were carried out in order to improve understanding of main loss mechanisms that can degrade device performance and the use of small amounts of Ge to mitigate some of them, including the possible interactions with alkali elements and the effect of post-deposition annealing treatments on devices properties. Most of the results obtained in this thesis have been published as articles in high impact peer-reviewed journals, included in text. In the first part of the thesis, previous study and optimization of the thermal processes were presented, identifying and varying the most critical parameters in a conventional tubular furnace selenization, in order to establish the best performing treatment for our particular precursors and set-up. Following this, after a preliminary screening of possible doping elements in the CZTSe system (including Ag, In, Si, Ge and Pb), In and Ge were both selected as the most promising/interesting ones to further analyze their doping properties. Although In doping did not show any performance improvement, it was demonstrated that CZTSe absorbers can tolerate rather high quantities of this element without significant modifications of their properties, confirming the possibility of using In-containing layers in kesterite CZTSe-based devices. On the other hand, with regard to Ge, a remarkable improvement of solar cells performance (from about 7% efficiency for Ge-free reference samples to more than 10% efficiency for Ge-doped ones) was presented, based on the introduction of nanometric Ge layers into the metallic stack precursors. Several reasons were proposed to explain the great efficiency improvement in spite of the observed Ge loss. In the following optimization, we determined the optimum Ge thickness range, achieving a 10.6% efficiency and open-circuit voltage values around 490 mV for pure selenide CZTSe, leading to voltage deficits around 0.56 V, which are among the best values reported for kesterites. Furthermore, a detailed microstructural analysis of high efficiency Ge-doped CZTSe solar cells was presented, revealing the presence of two distinct types of grain boundaries: one type is meandering in nature and grows largely parallel to the substrate, and denotes the boundary between two CZTSe layers with differing Cu/Zn ratios; and the second type is Cu-enriched and more straight, and predominates in the upper layer. After that, a new approach for obtaining high quality CZTSe layer was presented, by introducing extremely thin Ge nanolayers below and above metallic stack precursors. This strategy led to eliminate the previously characterized meandering horizontal grain boundaries and to obtain huge CZTSe grains. In addition, a deep study of the effect of Ge on the selenization process revealed that Ge strongly affects the in-depth elemental distribution and the phases formation, ultimately, modifying the reaction pathways of CZTSe. Through the optimization of the quantity and location of Ge, a record efficiency of 11.8% was achieved. In the last part of the thesis, the complex Ge-Na interaction was investigated, and a detailed analysis of low temperature post-deposition annealings by multi-wavelength Raman spectroscopy was also presented. On the whole, the work presented in this thesis provides meaningful results and innovative strategies to boost the efficiency of kesterite solar cells, by tackling some limiting factors of this promising material.
El objetivo principal de esta tesis es el desarrollo de células solares de capa fina de alta eficiencia basadas en absorbedores compuestos de elementos de baja toxicidad y abundantes en la corteza terrestre (kesterita, Cu2ZnSnSe4 (CZTSe)), mediante la implementación de estrategias innovadoras de dopaje. En particular, se ha desarrollado un método secuencial basado en el depósito por sputtering de capas metálicas seguido por un proceso térmico reactivo. A través de la optimización de los procesos y la implementación y análisis de diferentes elementos dopantes, se han investigado los mecanismos de pérdida de eficiencia más relevantes en kesteritas, contribuyendo a desarrollar soluciones alternativas. Los resultados obtenidos han sido publicados como artículos en revistas internacionales de alto factor de impacto. En la primera parte de la Tesis, se han estudiado profundamente los procesos térmicos reactivos, variando los parámetros más críticos para adaptarlos a las características particulares de los precursores metálicos. Seguidamente, se ha realizado un análisis de diferentes elementos dopantes (Ag, Si, Ge, Pb e In); In y Ge se han seleccionado como los más prometedores. De estos, Ge ha mostrado excelentes propiedades como dopante, incrementando la eficiencia de los dispositivos desde 7% hasta más de 10%, mediante la introducción de capas nanométricas (10 nm aprox.) en el precursor metálico. A través de una optimización profunda del proceso de dopado con Ge, se ha obtenido una eficiencia de conversión máxima de 11.8% y un déficit de voltaje de alrededor de 0.56 V, que representa uno de los mejores valores reportados para esta tecnología. Estos progresos se han acompañado de una profunda caracterización micro-estructural, que ha facilitado la identificación de importantes características de las kesteritas, como por ejemplo la presencia de dos tipos diferentes de fronteras de grano con distinta composición. Adicionalmente, Ge induce una modificación en los mecanismos de formación de kesteritas, que ha sido clave para mejorar las propiedades de los dispositivos fotovoltaicos basados en estas tecnologías. En resumen, los resultados obtenidos en la presente Tesis han servido para comprender, implementar y demostrar soluciones innovadoras para conseguir avances significativos en el desarrollo de tecnologías fotovoltaicas sostenibles basadas en kesteritas.

Thin-film solar cells based around the absorber material CuIn_1-xGa_xSe₂ (CIGS) are studied with respect to their stability characteristics, and different ways of modelling device operation are investigated. Two ways of modelling spatial inhomogeneities are detailed, one fully numerical and one hybrid model. In the numerical model, thin-film solar cells with randomized parameter variations are simulated showing how the voltage decreases with increasing material inhomogeneities.

With the hybrid model, an analytical model for the p-n junction action is used as a boundary condition to a numerical model of the steady state electrical conduction in the front contact layers. This also allows for input of inhomogeneous material parameters, but on a macroscopic scale. The simpler approach, compared to the numerical model, enables simulations of complete cells. Effects of material inhomogeneities, shunt defects and grid geometry are simulated.

The stability of CIGS solar cells with varying absorber thickness, varying buffer layer material and CIGS from two different deposition systems are subjected to damp heat treatment. During this accelerated ageing test the cells are monitored using characterization methods including J-V, QE, C-V and J(V)_T. The degradation studies show that the typical V_OC decrease experienced by CIGS cells subjected to damp heat is most likely an effect in the bulk of the absorber material.

When cells encapsulated with EVA are subjected to the same damp heat treatment, the effect on the voltage is considerably reduced. In this situation the EVA is saturated with moisture, representing a worst case scenario for a module in operation. Consequently, real-life modules will not suffer extensively from the V_OC degradation effect, common in unprotected CIGS devices.

Cadmium Telluride (CdTe) is a leading thin film photovoltaic (PV) material due to its near ideal bandgap of 1.45 eV and its high optical absorption coefficient. The typical CdTe thin film solar cell is of the superstrate configuration where a window layer (CdS), the absorber (CdTe) and a back contact are deposited onto glass coated with a transparent electrode. Substrate CdTe solar cells where the above listed films are deposited in reverse are not common. In this study substrate CdTe solar cells are fabricated on flexible foils. The properties of the Molybdenum back contact, Zinc Telluride (ZnTe) interlayer and CdTe absorber on the flexible foils were studied and characterized using X-Ray Diffraction (XRD), and Scanning Electron Microscopy (SEM). Substrate curvature and film flaking was observed during the fabrication as a result of differences in thermal expansion coefficients between the substrate and the deposited films, and also due to impurity diffusion from the foil into the film stack. In order to overcome this problem diffusion barriers where used to eliminate contamination. Silicon dioxide (SiO2), silicon nitride (Si3N4) and molybdenum nitride (MoxNy) were used as such barriers. Electrical characterization of completed devices was carried out by Current-Voltage (J-V), Capacitance-Voltage (C-V) and Spectral Response (SR) measurements. Roll-over was observed in the first quadrant of J-V curves indicating the existence of a back barrier due to a Schottky back contact. The formation of non-rectifying contact to p-CdTe thin-film is one of the major and critical challenges associated with the fabrication of efficient and stable solar cells. Several materials (ZnTe, Cu, Cu2Te, and Te) were studied as potential candidates for the formation of an effective back contact.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Materials Science and Engineering, 2012.
This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.
Cataloged from student-submitted PDF version of thesis.
Includes bibliographical references (pages 153-159).
The photovoltaic technology has been attracting widespread attention because of its effective energy harvest by directly converting solar energy into electricity. Thin-film silicon solar cells are believed to be a promising candidate for further scaled-up production and cost reduction while maintaining the advantages of bulk silicon. The efficiency of thin-film Si solar cells critically depends on optical absorption in the silicon layer since silicon has low absorption coefficient in the red and near-infrared (IR) wavelength ranges due to its indirect bandgap nature. This thesis aims at understanding, designing, and fabricating novel photonic structures for efficiency enhancement in thin-film Si solar cells. We have explored a previously reported a photonic crystal (PC) based structure to improve light absorption in thin-film Si solar cells. The PC structure combines a dielectric grating layer and a distributed Bragg reflector (DBR) for effcient light scattering and reflection, increasing light path length in the thin-film cell. We have understood the operation principles for this design by using photonic band theories and electromagnetic wave simulations. we discover that this DBR with gratings exhibit unusual light trapping in a way different from metal reflectors and photonic crystals. The light trapping effects for the DBR with and without reflector are numerically investigated. The self-assembled anodic aluminum oxide (AAO) technique is introduced to non- lithographically fabricate the grating structure. We adjust the AAO structural parameters by using different anodization voltages, times and electrolytes. Two-step anodization is employed to obtain nearly hexagonal AAO pattern. The interpore periods of the fabricated AAO are calculated by fast Fourier transform (FFT) analysis. We have also demonstrated the fabrication of ordered patterns made of other materials like amorphous Si (a-Si) and silver by using the AAO membrane as a deposition mask. Numerical simulations predict that the fabricated AAO pattern exhibits light trapping performance comparable to the perfectly periodic grating layer. We have implemented the light trapping concepts combining the self-assembled AAO layer and the DBR in the backside of crystalline Si wafers. Photoconductivity measurements suggest that the light absorption is improved in the near-IR spectral range near the band edge of Si. Furthermore, different types of thin-film Si solar cells, including a-Si, mi- crocrystalline Si ([mu]-Si) and micromorph Si solar cells, are investigated. For demonstration, the designed structure is integrated into a 1:5 [mu]m thick [mu]c-Si solar cell. We use numerical simulations to obtain the optimal structure parameters for the grating and the DBR, and then we fabricate the optimized structures using the AAO membrane as a template. The prototype devices integrating our proposed backside structure yield a 21% improvement in efficiency. This is further verified by quantum efficiency measurements, which clearly indicate stronger light absorption in the red and near-IR spectral ranges. Lastly, we have explored the fundamental light trapping limits for thin-film Si solar cells in the wave optics regime. We develop a deterministic method to optimize periodic textures for light trapping. Deep and high-index-contrast textures exhibit strong anisotropic scattering that is outside the regime of validity of the Lambertian models commonly used to describe texture-induced absorption enhancement for normal incidence. In the weak ab- sorption regime, our optimized surface texture in two dimensions (2D) enhances absorption by a factor of 2.7[pi]n, considerably larger than the classical [pi]n Lambertian result and exceeding by almost 50% a recent generalization of Lambertian model for periodic structures in finite spectral range. Since the [pi]n Lambertian limit still applies for isotropic incident light, our optimization methodology can be thought of optimizing the angle/enhancement tradeoff for periodic textures. Based on a modified Shockley-Queisser theory, we conclude that it is possible to achieve more than 20% efficiency in a 1:5 [mu]m thick crystalline Si cell if advanced light trapping schemes can be realized.
by Xing Sheng.
Ph. D.

Thesis advisor: Michael J. Naughton
Geometrically generalized analytical expressions for device transport are derived from first principles for a photovoltaic junction. Subsequently, conventional planar and unconventional coaxial and hemispherical photovoltaic architectures are applied to detail the device physics of the junction based on their respective geometry. For the conventional planar cell, the one-dimensional transport equations governing carrier dynamics are recovered. For the unconventional coaxial and hemispherical junction designs, new multi-dimensional transport equations are revealed. Physical effects such as carrier generation and recombination are compared for each cell architecture, providing insight as to how non-planar junctions may potentially enable greater energy conversion efficiencies. Numerical simulations are performed for arrays of vertically aligned, nanostructured coaxial and hemispherical amorphous silicon solar cells and results are compared to those from simulations performed for the standard planar junction. Results indicate that fundamental physical changes in the spatial dependence of the energy band profile across the intrinsic region of an amorphous silicon p-i-n junction manifest as an increase in recombination current for non-planar photovoltaic architectures. Despite an increase in recombination current, however, the coaxial architecture still appears to be able to surpass the efficiency predicted for the planar geometry, due to the geometry of the junction leading to a decoupling of optics and electronics
Thesis (PhD) — Boston College, 2012
Submitted to: Boston College. Graduate School of Arts and Sciences
Discipline: Physics

The response of thin-film photovoltaic devices to changes in the environment is not well understood. There are a large number of conflicting reports, reflecting largely the superimposed nature of the environmental effects. A separation of the effects is not often attempted mainly because of the lack of appropriate spectral data. An experimental system has been designed and operated to facilitate the separation of the environmental effects, including spectral effects. This involves measurements in a controlled laboratory environment as well as outdoor monitoring. Furthermore, a number of analysis tools have been developed and tested for their suitability. In order to develop a system model, the applicability of parametric models for thin-film devices is probed. The thermal variation of the underlying physical parameters is investigated and problems of describing thin-film devices with parametric models are discussed.

Microcrystalline silicon solar cells require an enhanced absorption of photons in the near-bandgap region between 700-1150nm. Conventional textured mirrors scatter light and increase the path length of photons in the absorber by total internal reflection. However, these mirrors exhibit a high surface roughness which degrades the performance of the microcrystalline silicon device. An alternative solution is to use metal nanoparticles with low surface roughness to scatter light. An illuminated metal nanoparticle exhibits a resonant or plasmonic excitation which can be tuned to enable a strong scattering of light. This work aims to develop an efficient near-infrared light-scattering system using randomly arranged metal nanoparticles near a mirror. Situating the nanoparticles at the rear of the solar cell helps to target weakly absorbed photons and eliminate out-coupling losses by the inclusion of a rear mirror. Simulation results show that the electric field driving the plasmonic resonance can be tuned with particle-mirror separation distance. The plasmonic scattering is maximised when the peak of the driving field intensity coincides with the intrinsic resonance of the nanoparticle. An e-beam lithography process was developed to fabricate a pseudo-random array of Ag nanodiscs near a Ag mirror. The optimized plasmonic mirror, with 6% coverage of 200nm Ag discs, shows higher diffusive reflectivity than a conventional textured mirror in the near-infrared region, over a broad angular range. Unlike a mirror with self-organised Ag islands, the mirror with Ag nanodiscs exhibits a low surface roughness of 13.5nm and low broadband absorption losses of around 10%. An 8.20% efficient thin n-i-p μc-Si:H solar cell, with the plasmonic mirror integrated at the rear, has been successfully fabricated. The optimised plasmonic solar cell showed an increase of 2.3mA in the short-circuit current density (Jsc), 6mV in the open-circuit voltage (Voc) and 0.97% in the efficiency (η), when compared to the planar cell counterpart with no nanodiscs. The low surface roughness of the plasmonic mirror ensures no degradation in the electrical quality of the μc-Si:H layer – this is also confirmed by the constant value of the fill factor (FF). The increase in Jsc is demonstrated to be mainly due to optical absorption enhancement in the near-infrared region as a result of plasmonic scattering, by detailed calculation of the exact photogenerated current in the plasmonic and planar devices, for the 700-1150nm wavelength range.

This thesis presents two models of a dye-sensitized solar cell (DSC): diffusion model and electrical model. The main purpose is to investigate interfacial charge transfer and charge transport within the semiconductor/electrolyte layer under illuminated conditions. These two interrelated models confirm that diffusion is the major driving force for electron and ion transport, while the drift of electrons is negligible. The diffusion model was utilized to simulate the temperature influence on the overall efficiency of DSC with a consideration of the voltage loss at titanium dioxide (TiO2)/ transparent conductive oxide (TCO) interface. It reveals that low temperature conditions have serious detrimental effects on the DSCs' performance. Further the electrical model was used to analyze the effect of diffusion/drift, dye loading, and electrode thickness on DSC performance. The predicted optimal electrode thickness ranges between 10-15 μm which is consistent with the thickness (10 μm) used in experimental studies published in the literature.

[ES] Desde hace una década se esta investigando intensamente la forma de mejorar la eficiencia de conversión de energía (PCE) de las células solares de silicio (Si) y reducir sus precios. Sin embargo, a pesar de las mejoras obtenidas, la fabricación de células solares de Si sigue siendo costosa y puede rebajarse usando materiales en forma de capa fina. Por ello la búsqueda de materiales absorbentes alternativos, no tóxicos, abundantes en la naturaleza y con buenos rendimientos de conversión se ha intensificado en los últimos años. Entre los diferentes materiales absorbentes el sulfuro de estaño (SnS), con una banda prohibida de 1.3 eV cercana a la óptima, es un candidato adecuado para la conversión fotovoltaica. Pero para células experimentales de SnS el rendimiento alcanzado hasta ahora es de 4.6%, que es mucho menos que el PCE para dispositivos de silicio, mientras que entre otras células híbridas (orgánicas-no orgánicas) como la perovskita de metilamonio de plomo y yodo (MAPbI3) se demuestra que es un candidato adecuado con PCE que alcanza un valor del 23%. Aparte de la estabilidad, uno de los problemas para la comercialización de células de MAPbI3 es la naturaleza tóxica del plomo (Pb). Por este motivo, se ha utilizado el análisis numérico para revisar los parámetros de diseño de las células solares de perovskita híbrida sustituyendo el absorbente MAPbI3 por MASnI3 y estudiar el efecto del resto de parámetros de diseño en el rendimiento de estas células solares. Hay varios softwares de simulación disponibles que se utilizan para el análisis numérico de células solares. En este trabajo hemos usamos un software llamado "A Solar Cell Capacitance Simulator" (SCAPS), está disponible de forma gratuita y es muy popular entre la comunidad científica y tecnológica. Para lograr un diseño efectivo para una célula solar eficiente, se propuso una aproximación numérica basada en la mejora de la PCE de una célula solar experimental. Esto se hizo reproduciendo los resultados para la célula solar diseñada experimentalmente en un entorno SCAPS con estructura p-SnS / n-CdS con una eficiencia de conversión del 1,5%. Después de la reproducción de los resultados experimentales, el rendimiento del dispositivo se optimizó ajustando el grosor de la capa absorbente y la capa tampón, la el tiempo de vida de los portadores minoritarios, la concentración del dopado en las capas absorbente, tampón y en la capa de la ventana. Mediante la optimización gradual de los parámetros del dispositivo, se alcanzó un valor de 14.01% en PCE de células solares diseñadas con SCAPS con arquitectura p-SnS / n-CdS / n-ZnO. A partir del análisis, se encontró que la PCE de una célula solar depende en gran medida de la concentración de dopaje de la capa absorbente, el espesor de la capa absorbente y los defectos de la interfaz. Sobre la base de los resultados obtenidos, se realizó un análisis para determinar el efecto de la recombinación de la interfaz en el rendimiento de las células solares y cómo se puede controlar. Para realizar esta tarea, se realizó un análisis para la selección de la capa tampón adecuada para la célula solar de perovskita metilamonio de estaño y yodo (MASnI3) y se encontró que el PCE de la célula solar también depende de la alineación de la banda entre el absorbedor y la capa de tampón. Por otra parte, se ha propuesto una nueva estructura para la célula solar de perovskita libre de Pb (contacto posterior / MASnBr3 / MASnI3 /CdZnS / FTO) con un PCE de 18.71% para un espesor del absorbedor de 500 nm y una concentración de dopado en el aceptor de 1x1016 cm-3. Los resultados obtenidos en esta tesis proporcionarán una guía para que los investigadores experimentales puedan construir células solares más eficientes.
[CAT] Des de fa una dècada s'està investigant intensament la forma de millorar l'eficiència de conversió d'energia (PCE) de les cèl·lules solars de silici (Si) i reduir els seus preus. No obstant això, tot i les millores obtingudes, la fabricació de cèl·lules solars de Si segueix sent costosa i pot rebaixar-se usant materials en forma de capa fina. Per això la recerca de materials absorbents alternatius, no tòxics, abundants en la naturalesa i amb bons rendiments de conversió s'ha intensificat en els últims anys. Entre els diferents materials absorbents, el sulfur d'estany (SnS), amb una banda prohibida de 1.3 eV propera a l'òptima, és un candidat adequat per a la conversió fotovoltaica. Però per a cèl·lules experimentals de SnS el rendiment assolit fins ara és de 4.6%, que és molt menor que el PCE per a dispositius de silici, mentre que entre altres cèl·lules híbrides (orgàniques-no orgàniques) com la perovskita de metilamonio de plom i iode ( MAPbI3) es demostra que és un candidat adequat amb PCE que arriba a un valor del 23%. A part de l'estabilitat, un dels problemes per a la comercialització de cèl·lules de MAPbI3 és la naturalesa tòxica del plom (Pb). Per aquest motiu, s'ha utilitzat l'anàlisi numèrica per revisar els paràmetres de disseny de les cèl·lules solars de perovskita híbrida substituint l'absorbent MAPbI3 per MASnI3 i estudiar l'efecte de la resta de paràmetres de disseny en el rendiment d'estes cèl·lules solars. Hi ha diversos programaris de simulació disponibles que s'utilitzen per a l'anàlisi numèric de cèl·lules solars. En aquest treball hem fem servir un programari anomenat "A Solar Cell Capacitance Simulator" (SCAPS), està disponible de forma gratuïta i és molt popular entre la comunitat científica i tecnològica. Per aconseguir un disseny efectiu per a una cèl·lula solar eficient, es va proposar una aproximació numèrica basada en la millora de la PCE d'una cèl·lula solar experimental. Això es va fer reproduint els resultats per a la cèl·lula solar dissenyada experimentalment en un entorn SCAPS amb estructura p-SnS / n-CdS amb una eficiència de conversió de l'1,5%. Després de reproduir els resultats experimentals, el rendiment del dispositiu es va optimitzar ajustant el gruix de la capa absorbent y de la capa tampó, el temps de vida dels portadors minoritaris, la concentració del dopatge en les capes absorbent, tampó i en la capa finestra. Mitjançant l'optimització gradual dels paràmetres del dispositiu, es va assolir un valor de 14.01% en PCE de cèl·lules solars dissenyades experimentalment en SCAPS amb arquitectura p-SnS / n-CdS / n-ZnO. A partir de l'anàlisi, es va trobar que la PCE d'una cèl·lula solar depèn en gran mesura de la concentració de dopatge de la capa absorbent, el gruix de la capa absorbent i els defectes de la interfície. D'altra banda, es va realitzar una anàlisi per determinar l'efecte de la recombinació de la interfície en el rendiment de les cèl·lules solars i com es pot controlar. Per realitzar aquesta tasca, es va realitzar una anàlisi per a la selecció de la capa tampó adequada per a la cèl·lula solar de perovskita de metilamoni d'estany i iode (MASnI3) i es va trobar que el PCE de la cèl·lula solar també depèn de l'alineació de la banda entre l'absorbidor i la capa de tampó.
[EN] A decade of extensive research has been conducted to enhance the power conversion efficiency (PCE) of silicon (Si) solar cells and to cut their prices short. But still, the fabrication of Si solar cells are costly. So, to reduce the fabrication cost of the solar cell search for alternate earth abundant and non-toxic absorber materials is thriving. Among different absorber materials tin sulfide (SnS) is found to be a suitable candidate for the non-organic solar cell with a band gap of 1.3 eV. But the PCE achieved for SnS is 4.6% that is far less from the PCE of (Si), whereas among other organic non-organic solar cells like methylammonium lead halide perovskite ({\rm MAPbI}_3) is proven to be a suitable candidate with PCE reaching to a value of 23%. The problem with the commercialization of {\rm MAPbI}_3 is due to the toxic nature of lead (Pb). So, in dealing with these issues of solar cell numerical analysis can play a key role as numerical analysis allows flexibility in the design of realistic problem and experimentation with different hypotheses can easily be performed. Complete set of device characteristic can often be easily generated by consuming less amount of time and effort. Because of this reason numerical analysis was used to revisit solar cells design parameters and the effect of solar cell physical parameters on solar cell performance. There are various simulation software's available that are used for solar cell numerical analysis. Here in this work, we used Solar cell capacitance simulator (SCAPS) software, it is freely available and is most popular among the research community. To achieve effective design for efficient solar cell a numerical guide was proposed based on which PCE of an experimental designed solar cell can be enhanced. This was done by reproducing results for the experimentally designed solar cell in SCAPS environment with structure p-SnS/n-CdS having a conversion efficiency of 1.5%. After reproduction of experimental results device performance was optimized by varying thickness of (absorber layer, buffer layer), minority carrier lifetime, doping concentration (absorber, buffer), and adding window layer. By stepwise optimization of device parameters, PCE of an experimental designed solar cell in SCAPS with architecture p-SnS/n-CdS/n-ZnO was reached to a value of 14.01%. From the analysis, it was found that PCE of a solar cell is highly depended upon doping concentration of the absorber layer, the thickness of the absorber layer and interface defects. Based on the results evaluated an analysis was performed for tin based organic non-organic methylammonium tin halide perovskite solar cell ({\rm MASnI}_3) to find the effect of interface recombination on solar cell performance and how it can be governed. The reason for this transition from SnS to {\rm MASnI}_3 was because {\rm MASnI}_3 can be fabricated simply by spin-coating methylammonium iodide (MAI) over SnS layer. To perform this task analysis was performed for the selection of suitable buffer layer for Pb free methylammonium tin halide perovskite solar cell ({\rm MASnI}_3) and it was found that PCE of the solar cell is also depended upon band alignment between absorber and buffer layer. Based on the results a new structure was proposed for Pb free perovskite solar cell (Back\ contact/{\rm MASnBr}_3/{\rm MASnI}_3/CdZnS/FTO) with PCE of 18.71% for absorber thickness of 500 nm and acceptor doping concentration of 1x10^{16}\ {\rm cm}^3. The results achieved in this thesis will provide an imperative guideline for researchers to design efficient solar cells.
Baig, F. (2019). Numerical analysis for efficiency enhancement of thin film solar cells [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/118801
TESIS

[ES] Las energía solar fotovoltaica ha emergido como una fuente de energía nueva y sostenible, que es ecológica y rentable si la producción es a gran escala. En el escenario actual, los dispositivos fotovoltaicos económicos y de alta eficiencia de conversión sin que se degraden sus componentes están bien posicionados para la generación de electricidad. Las células solares basadas en silicio dominan este mercado desde hace muchos años. Para la fabricación y producción de células solares basadas en silicio, se requieren sofisticadas técnicas de fabricación que hacen que el panel solar sea costoso. Por otra parte estan las células solares de película delgada, las cuales, debido a la intensificación de las capacidades de fabricación están ganando importancia. La tecnología de película delgada es una de las tecnologías más rentables y eficientes para la fabricación de células solares, y es un tema de intensa investigación en la industria fotovoltaica. La tecnología de película delgada es más económica que otras tecnologías porque los dispositivos utilizan menos material y están basados en varios tipos de materiales semiconductores que absorben la luz. Entre estos materiales, las células solares de kesterita que utilizan CZTS, CZTSe y sus aleaciones CZTSSe pueden convertirse en el reemplazo óptimo a los absorbentes de calcopirita. Estos materiales presentan unas características ópticas y eléctricas sobresalientes y tienen un gap óptico directo con una banda prohibida que oscila entre 1,4\ eV\ y 1,5\ eV y un coeficiente de absorción, \alpha>{10}^4{cm}^{-1}. Estas características han propiciado que las kesteritas esten siendo muy investigadas por la comunidad fotovoltaica de películas delgadas. De acuerdo con el límite de Shockley-Queisser, la eficiencia de conversión para una célula solar basada en CZTS\ es alrededor del 28%. Esta eficiencia es teóricamente posible mediante el ajuste de la banda prohibida, pero aún así, todavia no se ha podido alcanzar experimentalmente, probablemente debido a la falta de comprensión de las características de los dispositivos.Para una mejor comprensión de las características de los dispositivos, la modelación numérica puede jugar un papel importante al perimitir estudiar diferentes estructuras de dispositivos que pueden ahorrar tiempo y costos a la comunidad científico-técnica. En este trabajo, se ha llevado a cabo una modelazación numérica para estimar y analizar el efecto de parámetros físicos como el espesor y la concentración de dopado de la capa absorbente, la capa tampón y las capas ventana, además de estudiar el efecto de la temperatura y el efecto de la potencia de iluminación del sol en el rendimiento del dispositivo. El análisis numérico de los dispositivos se realizó con el software de simulación denominado "Solar Cell Capacitance Simulator" (SCAPS-1D). Para ello se analizó una estructura simple p-n-n^+ usando molibdeno como contacto posterior y FTO como ventana óptica y contacto frontal y siguiendo la secuencia de materiales Mo/CZTS/CdS/ZnO/FTO. A través del análisis, se estudió el rendimiento de las células solares con la variación en el espesor del absorbente para encontrar el espesor óptimo de la capa absorbente. También se estudió el efecto de la concentración del dopado y de la función de trabajo del metal. Después de la visualización de una estructura de dispositivo básica en SCAPS-1D, se modelo una célula solar experimental basada en CZTS. Los resultados de las células solares CZTS diseñados experimentalmente se simularon por primera vez en el entorno SCAPS-1D. Los resultados simulados de SCAPS-1D se compararon con los resultados experimentales. Después de la optimización de los parámetros de la celda, se incrementó la eficiencia de conversión de un dispositivo optimizado y, a partir del modelado, se descubrió que el rendimiento del dispositivo mejora al aumentar el tiempo de vida de los porta
[CAT] L'energia solar fotovoltaica ha emergit com una font d'energia nova i sostenible, que és ecològica i rendible si la producció és a gran escala. En l'escenari actual, els dispositius fotovoltaics econòmics i de gran eficiència de conversió estan ben posicionats per a la generació d'electricitat neta i sostenible. Les cèl·lules solars basades en silici dominen aquest mercat des de fa molts anys. Per a la fabricació i producció de cèl·lules solars basades en silici, es requereixen tècniques de fabricació sofisticades que fan que el panell solar sigui costós. Per altra banda estan les cel·les solars de capa fina, que estan guanyant importància a causa de l'intensificació de les capacitats de fabricació. La tecnologia de capa fina és una de les tecnologies més rentables i eficients per a la fabricació de cel solars, i és un tema d'intensa investigació en la fotovoltaica industrial. La tecnologia de capa fina és més econòmica que altres tecnologies perquè els dispositius utilitzen menys material i estan basats en diversos tipus de materials semiconductors que absorbeixen la llum. Entre aquests materials, les cèl·lules solars de kesterita que utilitzen CZTS, CZTSe i les seves aleacions CZTSSe poden convertir-se en el reemplaçament òptim als absorbents de calcopirita. Aquests materials presenten unes característiques òptiques i elèctriques sobresalientes i tenen un gap òptic directe amb una banda prohibida que oscil·la entre 1,4eV i 1,5eV i un coeficient d'absorció, \alpha>{10}^4{cm}^{-1}. Aquestes característiques han propiciat que les Les kesteritas estan sent molt investigades per la comunitat fotovoltaica de capes primes. D'acord amb el límit de Shockley-Queisser, l'eficiència de conversió per a una cel·la solar basada en CZTS és d'aproximadament 28%. Aquesta eficiència és teòricament possible a través de l'ajust de la banda prohibida, però tot i així, encara no s'ha pogut assolir experimentalment, probablement a causa de la incomprensió del funcionament dels dispositius. Per a una millor comprensió de les característiques i funcionament dels dispositius, la modelització numèrica pot jugar un paper important al permetre estudiar diferents estructures de sistemes que poden estalviar temps i costos a la comunitat científica-tècnica. En aquest treball, s'ha dut a terme una modelització numèrica per estimar i analitzar l'efecte de paràmetres físics com l'espessor i la concentració de dopatge de la capa absorbent, la capa tampó i la capa finestra, a més d'estudiar l'efecte de la temperatura i l'efecte de la potència d'il·luminació del sol en el rendiment del dispositiu. L'anàlisi numèrica dels dispositius es va realitzar amb el programari de simulació denominat "Solar Cell Capacitance Simulator" (SCAPS-1D). Per això es va analitzar una estructura senzilla p-n-n^+ utilitzant molibdé com contacte posterior i FTO com a finestra òptica i contacte frontal i seguint la seqüència de materials Mo/CZTS/CdS/ZnO/FTO. A través de l'anàlisi, es va estudiar el rendiment de les cel·les solars amb la variació en l'espessor de l'absorbent per trobar l'espessor òptim de la capa absorbent. També es va estudiar l'efecte de la concentració del dopatge i de la funció de treball del metall. Després de la visualització d'una estructura de dispositiu bàsic en SCAPS-1D, es model una cel·la solar experimental basada en CZTS. Els resultats de les cel·les solars CZTS dissenyats experimentalment es simularen per primera vegada en l'entorn SCAPS-1D. Els resultats simulats de SCAPS-1D es van comparar amb els resultats experimentals. Després de l'optimització dels paràmetres de la celda, es va incrementar l'eficiència de conversió d'un dispositiu optimitzat i, a partir del modelatge, es va descobrir que el rendiment del dispositiu es millora a l'augmentar la vida útil dels minoritaris, cosa que es aconsegueix amb la incorporació d'un camp elèctric a la superfície del con
[EN] The solar cell has emerged as a newer and a relatively sustainable energy source, that is eco-friendly and cost-effective if the production is on a larger scale. In the current scenario, the economic and high-power conversion efficiency photovoltaic devices without degradation of materials are designed for the generation of electricity. The silicon-based solar cells dominated the market for many years. For the manufacturing and production of silicon-based solar cells, sophisticated fabrication techniques are required that make the solar panel costly. Due to intensification in manufacturing capabilities, thin film solar cells are gaining significance. Thin film technology is one of the most cost-effective and efficient technologies for the manufacturing of solar cells, and it is an excellent subject of intense research in the photovoltaic industry. Thin film technology is economical than other technologies because devices have relatively less material and are based on various types of light absorbing semiconductor materials. Among these materials, kesterite solar cells utilizing CZTS, CZTSe and their alloys CZTSSe are emerging as the most auspicious replacement for the chalcopyrite absorbers. The outstanding electrical and optical features having direct optical band gap ranges among 1.4eV to 1.5eV and large absorption coefficient \alpha\ >{10}^4{cm}^{-1} of CZTS have made it very interesting in the thin film community. According to the Shockley-Queisser limit, the optimum conversion efficiency of around 28\ % is theoretically possible from a CZTS based solar cell by tuning the band gap, but still, it is not experimentally possible to achieve 28% conversion efficiency from a solar cell due to lack of understanding of device characteristics. For a better understanding of device characteristics, numerical modeling can play a significant role by modeling different device structures that can save time and cost of the research community. In this work, numerical modeling was carried out for estimating and analyzing the effect of physical parameters such as thickness and doping concentration of absorber, buffer and window layers, temperature effect and effect of illumination power of the sun on device performance. Device modeling had performed on the dedicated simulation software "Solar Cell Capacitance Simulator" (SCAPS-1D). To achieve this task first, a simple {p-n-n}^+ structure for Mo/CZTS/CdS/ZnO/FTO had been analyzed with molybdenum as back contact and FTO as a front contact. Through analysis, it had been found that solar cell performance was affected by variation in absorber thickness, doping concentration, and metal work function. After visualization of a basic device structure in SCAPS-1D, CZTS based experimental solar cell had been modeled. Experimentally designed CZTS solar cell results were first simulated in SCAPS-1D environment. The SCAPS-1D simulated results were then compared with experimental results. After optimization of cell parameters, the conversion efficiency of an optimized device was increased and from modeling, it had been found that device performance was improved by improving minority carrier lifetime and integration of back surface field at the back contact. Based on the results presented, it was found that recombination in a solar cell can greatly affect the performance of a solar cell. Therefore, a new structure (Back\ contact/CFTS/ZnS/Zn(O,S)/FTO) was modeled and analyzed in which interface recombination is reduced by optimizing the band gap of Zn(O,S) layer. Based on different device structure modeling, it was found that solar cell with structure CFTS/ZnS/Zn(O,S)/FTO can exhibit an efficiency of 26.11% with optimized physical parameters like absorber thickness layer of 4\mu m and acceptor concentration density of 2\times{10}^{18}\ {cm}^{-3}. The proposed results will give a valuable guideline for the feasible fabrication and designing of high-power conversion efficiency solar cells.
Khattak, YH. (2019). Modeling of High Power Conversion Efficiency Thin Film Solar Cells [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/118802
TESIS

This thesis describes the development of a fast rate method for the deposition of high quality CdS and CdTe thin films. The technique uses Pulsed DC Magnetron Sputtering (PDCMS). Surprisingly, the technique produces highly stable process conditions. CREST is the first laboratory worldwide to show that pulsed DC power may be used to deposit CdS and CdTe thin films. This is a very promising process technology with potential for eventual industrial deployment. The major advantage is that the process produces high deposition rates suitable for use in solar module manufacturing. These rates are over an order of magnitude faster than those obtained by RF sputtering. In common with other applications it has also been found that the energetics of the pulsed DC process produce excellent thin film properties and the power supply configuration avoids the need for complex matching circuits. Conventional deposition methodologies for CdS, Chemical Bath Deposition (CBD) and CdTe thin films, Electrodeposition (ED), have been chosen as baselines to compare film properties with Pulsed DC Magnetron Sputtering (PDCMS). One of the issues encountered with the deposition of CdS thin films (window layers) was the presence of pinholes. A Plasma cleaning process of FTO-coated glass prior to the deposition of the CdS/CdTe solar cell has been developed. It strongly modifies and activates the TCO surface, and improves the density and compactness of the deposited CdS thin film. This, in turn, improves the optical and morphological properties of the deposited CdS thin films, resulting in a higher refractive index. The pinhole removal and the increased density allows the use of a much thinner CdS layer, and this reduces absorption of blue spectrum photons and thereby increases the photocurrent and the efficiency of the thin film CdTe cell. Replacing the conventional magnetic stirrer with an ultrasonic probe in the chemical bath (sonoCBD) was found to result in CdS films with higher optical density, higher refractive index, pinhole and void-free, more compact and uniform along the surface and through the thickness of the deposited material. PDCMS at 150 kHz, 500 W, 2.5 μs, 2 s, results in a highly stable process with no plasma arcing. It allows close control of film thickness using time only. The CdS films exhibited a high level of texture in the <001> direction. The grain size was typically ~50 nm. Pinholes and voids could be avoided by reducing the working gas pressure using gas flows ii below 20 sccm. The deposition rate was measured to be 1.33 nm/s on a rotating substrate holder. The equivalent deposition rate for a static substrate is 8.66 nm/s, which is high and much faster than can be achieved using a chemical bath deposition or RF magnetron sputtering. The transmission of CdS can be improved by engineering the band gap of the CdS layer. It has been shown that by adding oxygen to the working gas pressure in an RF sputtering deposition process it is possible to deposit an oxygenated CdS (CdS:O) layer with an improved band gap. In this thesis, oxygenated CdS films for CdTe TF-PV applications have been successfully deposited by using pulsed DC magnetron sputtering. The process is highly stable using a pulse frequency of 150 kHz and a 2.5 μs pulse reverse time. No plasma arcing was detected. A range of CdS:O films were deposited by using O2 flows from 1 sccm to 10 sccm during the deposition process. The deposition rates achieved using pulsed DC magnetron sputtering with only 500 W of power to the magnetron target were in the range ~1.49 nm/s ~2.44 nm/s, depending on the oxygen flow rate used. The properties of CdS thin films deposited by pulsed DC magnetron sputtering and chemical bath deposition have been studied and compared. The pulsed DC magnetron sputtering process produced CdS thin films with the preferred hexagonal <001> oriented crystalline structure with a columnar grain growth, while sonoCBD deposited films were polycrystalline with a cubic structure and small grainy crystallites throughout the thickness of the films. Examination of the PDCMS deposited CdS films confirmed the increased grain size, increased density, and higher crystallinity compared to the sonoCBD CdS films. The deposition rate for CdS obtained using pulsed DC magnetron sputtering was 2.86 nm/s using only 500 W power on a six inch circular target compared to the much slower (0.027 nm/s) for the sonoChemical bath deposited layers. CdTe thin films were grown on CdS films prepared by sonoCBD and Pulsed DC magnetron sputtering. The results showed that the deposition technique used for the CdS layer affected the growth and properties of the CdTe film and also determined the deposition rate of CdTe, being 3 times faster on the sputtered CdS. PDCMS CdTe layers were deposited at ambient temperature, 500 W, 2.9 μs, 10 s, 150 kHz, with a thickness of approximately 2 μm on CdS/TEC10 coated glass. The layers appear iii uniform and smooth with a grain size less than 100 nm, highly compact with the morphology dominated by columnar grain growth. Stress analysis was performed on the CdTe layers deposited at room temperature using different gas flows. Magnetron sputtered thin films deposited under low gas pressure are often subject to compressive stress due to the high mobility of the atoms during the deposition process. A possible way to reduce the stress in the film is the post-deposition annealing treatment. As the lattice parameter increased; the stress in the film is relieved. Also, a changing the deposition substrate temperature had an effect on the microstructure of CdTe thin films. Increasing the deposition temperature increased the grain size, up to ~600 nm. CdTe thin films with low stress have been deposited on CdS/TEC10 coated glass by setting the deposition substrate temperature at ~200°C and using high argon flows ~ 70 sccm Ar. Finally, broadband multilayer ARCs using alternate high and low refractive index dielectric thin films have been developed to improve the light transmission into solar cell devices by reducing the reflection of the glass in the extended wavelength range utilised by thin-film CdTe devices. A four-layer multilayer stack has been designed and tested, which operates across the wavelength range used by thin-film CdTe PV devices (400 850 nm). Optical modelling predicts that the MAR coating reduces the WAR (400-850 nm) from the glass surface from 4.22% down to 1.22%. The application of the MAR coating on a thin-film CdTe solar cell increased the efficiency from 10.55% to 10.93% or by 0.38% in absolute terms. This is a useful 3.6% relative increase in efficiency. The increased light transmission leads to improvement of the short-circuit current density produced by the cell by 0.65 mA/cm2. The MAR sputtering process developed in this work is capable of scaling to an industrial level.

The global need for a clean, sustainable and affordable source of energy has triggered extensive research especially in renewable energy sources. In this sector, photovoltaic has been identified as a cheapest, clean and reliable source of energy. It would be of interest to obtain photovoltaic material in thin film form by using simple and inexpensive semiconductor growth technique such as electroplating. Using this growth technique, four semiconductor materials were electroplated on glass/fluorine-doped tin oxide (FTO) substrate from aqueous electrolytes. These semiconductors are indium selenide (InxSey), zinc sulphide (ZnS), cadmium sulphide (CdS) and cadmium telluride (CdTe). lnxSey and ZnS were incorporated as buffer layers while CdS and CdTe layers were utilised as window and absorber layers respectively. All materials were grown using two-electrode (2E) system except for CdTe which was grown using 3E and 2E systems for comparison. To fully optimise the growth conditions, the as-deposited and annealed layers from all the materials were characterised for their structural, morphological, optical, electrical and defects structures using X-ray diffraction (XRD), Raman spectroscopy, scanning electron microscopy (SEM), atomic force microscopy (AFM), optical absorption (UV- Vis spectroscopy), photoelectrochemical (PEC) cell measurements, current-voltage (I-V), capacitance-voltage (C-V), DC electrical measurements, ultraviolet photoelectron spectroscopy (UPS) and photoluminescence (PL) techniques. Results show that InxSey and ZnS layers were amorphous in nature and exhibit both n-type and p-type in electrical conduction. CdS layers are n-type in electrical conduction and show hexagonal and cubic phases in both the as-deposited and after annealing process. CdTe layers show cubic phase structure with both n-type and p- type in electrical conduction. CdTe-based solar cell structures with a n-n heterojunction plus large Schottky barrier, as well as multi-layer graded bandgap solar cells were fabricated. This means that the solar cells investigated in this thesis were not the conventional p-n junction type solar cells. The conventional cadmium chloride (CdCl2 or CC) treatment was applied to the structures to produce high performance devices; however, by modifying the treatment to include cadmium chloride and cadmium fluoride (CdCl2+CdF2 or CF) device performance could be improved further. The fabricated devices were characterised using I-V and C-V measurement techniques. The highest cell efficiency achieved in this research was ~10%, with an open circuit voltage of 640 mV, short-circuit current density of 38.1 mAcm-2, fill factor of 0.41 and doping concentration of 2.07xl016 cm'3. These parameters were obtained for the glass/FTO/n-InxSey/n- CdS/n-CdTe/Au solar cell structure.

Electrodeposition was used to deposit thin film semiconductor materials for use in solar cell devices. Copper indium diselenide (CuInSe₂) was deposited from ethylene glycol at 150°C with the aim of improved material properties due to the elevated temperature. The broad nature of the X-ray diffraction (XRD) peaks before and after annealing indicated the layers were comprised of multiple phases identified as CuInSe₂ and Cu-Se binaries. Insufficient indium inclusion for device quality materials was incorporated into the layers over the explored growth range of -0.800 to -1.000 V vs Se reference electrode. The layers deposited at more positive deposition voltages were metallic and contained mainly Cu-Se binary phases. At more negative deposition voltages the formation of CuInSe₂ was confirmed although above -1.000 V vs Se the layers were often powdery and disintegrated on removal from the electrolyte. There were no noticeable improvements in the CuInSe₂ layers deposited from ethylene glycol compared to reports from aqueous media, which is less toxic and lower cost, therefore electrodeposition from aqueous solution is preferable. Undoped zinc oxide (ZnO) and aluminium doped ZnO (ZnO: Al) were deposited from zinc nitrate solutions with the aim of using electrodeposition for the ZnO bilayer in CuInSe₂ devices to unify the production process. ZnO was deposited at a range of deposition voltages from -0.900 to -1.050 V vs silver/silver chloride reference electrode as identified using XRD. Various morphologies were observed using scanning electron microscopy (SEM) and the electrical resistivity was determined at 6.9x 10⁶ Ω cm and decreased to 3.4x10⁵ Ω cm after Al doping. To make this method suitable for commercialisation more work would need to be carried out to address consistency issues mainly regarding the electrolyte conditions, including pH and oxygen concentration as a function of growth time. A comparison was made between electrodeposited and sputtered ZnO and ZnO: AI. Some differences in the material properties were found; all layers were identified as hexagonal wurtzite ZnO. A considerable change in morphology was observed by SEM between the electrodeposited and sputtered materials. Little change in the electrical resistivity was observed between electrodeposited and sputtered undoped ZnO, having 6.9x 10⁶ and 6.2x 10⁷ Ω cm. The electrical resistivity of ZnO: AI was 3.4 x l 0⁵ and 2.3 x 10⁵ Ω cm for electrodeposited and sputtered materials respectively. Further work would need to be carried out to quantify the concentration of Al dopant in the electrodeposition solution as a function of growth time if this method were to be used for commercialisation. Cadmium telluride (CdTe) was electrodeposited from aqueous solution onto glass/fluorine doped tin oxide/cadmium sulphide substrates. Little improvement in XRD spectra was observed for annealed layers compared to the as-deposited material and the CdTe was identified exhibited cubic phase having (111) preferential orientation. Working solar cell devices were fabricated over a range of growth voltages with superior performance being observed for materials deposited between -0.620 to -0.650 V vs saturated calomel electrode (SCE). Furthermore high uniformity over a2 cm² area completed with an array of 2 mm diameter contacts was observed for devices deposited in this growth voltage range. All devices fabricated using CdTe grown at -0.610 to -0.690 V vs SCE indicated photovoltaic activity although layers deposited between -0.630 and -0.650 V vs SCE indicated the highest performance, with device parameters of open circuit voltage = 420-540 mV, short circuit current density = 3.2-19.1 mA cm^-2 and fill factor = 0.48.

The power conversion efficiency of organic thin film solar cells must be improved before they can become commercially competitive alternatives to silicon-based photovoltaics. Exciton diffusion and charge carrier migration in organic films are strongly influenced by film morphology, which can be controlled by the substrate temperature during film growth. Zinc-phthalocyaninelbuckminsterfullerene bilayer film devices are fabricated with substrate temperatures between 25°C and 224°C and their solar cell performance is investigated here. The device open-circuit voltage, efficiency, and fill factor all exhibit peaks when films are grown at temperatures between 160°C and 180°C, which is likely a result of both the increase in shunt resistance and reduction in undesirable back diode effects which occur between l00°C and 180°C. The device performance can also be attributed to changes in the film crystallite size, roughness, and abundance of pinholes, as well as the occurrence of crystalline phase transitions which occur in both zinc-phthalocyanine and buckminsterfullerene between 150°C and 200°C. The unusually high open-circuit voltage (1.2 V), low short-circuit current density (0.03 mA/cm²), and low device efficiency (0.04%) reported here are reminiscent of single layer phthalocyanine-based Schottky solar cells, which suggests that pinholes in bilayer film devices can effectively lead to the formation of Schottky diodes.

To cope with energy requirements the utilization of renewable energies, particularly the Sun supplies the biggest and abundant energy source in Earth. Photo-voltaic and solar cell are the well advance and burning technology and a field of hot research. Majority of research centers and universities are working in this field. 1G, 2G, 3G and next generation of photo-voltaic cells have been developed and still to improve its efficiency and to decrease it 0.2 $/W cost. Our work mainly based on the theoretical and physical analysis of thin-film Photovoltaic devices. We will explore different software used for the analysis of PV cells, and will analyse different simulation related to solar cells like open circuit voltage VOC, Short circuit current JSC, Fill Factor FF (%) and external Quantum efficiency (%) for thin film solar cell including CIGS, CIS, CGS, CdTe, SnS/CdS/ZnO etc. To have different analysis for different combination and different replacement for materials used in the solar cell fabrication. To cope with the PV cost and environmental hazards we have to find alternate solutions.
Ullah, H. (2015). Simulation studies of photovoltaic thin film devices [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/48800
TESIS

A study has been made of a variety of factors influencing the efficiency and operational stability of front-wall CdS-Cu2S solar cells. In the course of this work -1 cm2 cells were fabricated with conversion efficiency of up to 8% without attempting to reduce reflection losses.The CdS films were produced by vacuum evaporation and the electrical and structural characteristics of these films were studied as a function of the rate and temperature of the deposition. Previously there had been some controversy concerning the nature of the CdS source material required for fabricating high performance CdS-based solar cells, but this work has shown that a variety of CdS sources can be employed successfully provided that the film deposition parameters are suitably chosen.A conventional chemical exchange technique was employed to convert the CdS film surface to Cu2SI with the thickness and stoichiometry of the resultant Cu2S layer being examined by means of electrochemical analysis.Changes in the electrical properties of the CdS-Cu2S cells due to post- fabrication anealing under a variety of different conditions were studied and correlated with structural changes monitored by means of Auger electron spectroscopy with the aid of argon ion etching. Depth profiles of the constituent element concentrations indicate that, for samples annealed in air, a deep penetration of copper into the CdS layer occurs together with a significant out-diffusion of cadmium from the CdS after only a few minutes at 1000C. In contrast, the copper penetration which results from vacuum or hydrogen annealing treatment is substantially less and no significant out-diffusion of cadmium is observed for annealing temperatures up to 4000C. Two different diffusion processes, one in the grain boundaries and one in the mid-grain regions, have been identified and their relative importance has been studied for annealing cycles performed under the same three different ambient atmospheres (air, vacuum or hydrogen). The normally rapid and undesirable grain boundary diffusion of copper was found to be significantly inhibited by the use of flowing hydrogen during annealing. A further technologically important observation concerns the effect of the deposition of a film of copper over the copper sulphide layer of a cell and subsequent annealing of it in air. The improved electrical stability which this treatment yields has been shown to be directly associated with reduced interdiffusion at the CdS-Cu2S interface. This interfacial diffusion has also been shown to be influenced by the CdS stoichiometry in the vicinity of the junction.Finally, a brief investigation was made into the use of the ion implantation technique as a means of doping the upper layer of the OdS film with copper without annealing the completed cell. The results have demonstrated the feasibility of this technique, with the best results being obtained using a copper ion fluence of 5.1014 ions cm-2 at 50 keV ion energy.

Thin film photovoltaics encompass a group of technologies able to harvest light within a few microns thickness. The reduced thickness allows a low cost of manufacture while making the films flexible and adaptable to different surfaces. This, combined with their low weight, positioned thin film solar cells as ideal candidates for building integrated photovoltaics. For the latter, organic solar cells (OSC) can provide a high quality semi-transparency that closely mimics the aesthetics of standard windows. Indeed, some unique features of organic solar cells make them the optimal solution for applications where standard Si technology cannot be used. However, for large-scale electricity production where efficiency is, perhaps, the most determining factor, newer thin film technologies like perovskites solar cells may be a more adequate option. At the moment of writing this thesis, state of the art efficiencies of single junction perovskites nearly double that of the best single junction organic solar cell. A limitation found in both technologies, especially in organics and to a lesser degree in perovskites, is the low mobility of the carriers. This, together with other processing shortcomings in the organic absorbers and perovskites limit their thickness to 100-130 nm, and 500-600 nm, respectively. In summary, light management must be an essential ingredient when designing device architectures to achieve the optimal performance in the specific application being considered. In this thesis, in order to achieve an optimal light harvesting and therefore increase the performance of thin film solar cells, we take two approaches. On one hand, we increase the total thickness of the absorber material used in the device without increasing the thickness of the single active material layer and, on the other hand, we combine complementary absorbers to cover a wider portion of the solar spectra. These approaches pose the double challenge of finding the optimal electromagnetic field distribution within a complicated multilayer structure containing two or more active layers, while at the same time implementing an effective charge collection or recombination in the intermediate layers connecting two adjacent sub-cells. In the case of OSC, we consider multi-junction cells where the same active material is used in all the junctions. This can be implemented by fabricating structures where the active layer thickness in each sub-cell does not exceed the 100 nm. For other types of thin film solar cells, we consider configurations using complementary absorbers. In both cases, but particularly in the former one, a systematic approach to optimize light absorption is needed. In order to obtain such optimal configurations, we implement an inverse integration approach combined with a transfer matrix calculation of the electric field. Furthermore, we develop several new approaches to optimize charge collection in the sub-cell interconnection layers which we apply to tandem, triple, 4-terminal and series-parallel configurations. The thesis has been organized into five chapters. Chapter 1 introduces concepts required for the development of the thesis work including the optical model. Chapter 2 describes the optical optimization and experimental implementation of current-matched multi-junction devices using PTB7:PC71BM, including applications. In order to profit from the advantage of electrically separated devices, Chapter 3 evaluates different types of 4-terminal architectures using PTB7:PC71BM and PTB7-Th:PC71BM. In one of the architectures we establish a serial-connection between sub-cells while in other we leave the sub-cells completely independent. Chapter 4 theoretically proposes a novel monolithic architecture combining perovskites and CIGS which does not require current-matching. Finally, in Chapter 5, an in-depth study of the semi-transparent inner electrodes is given that include vacuum-based and solution-processed layers.
La fotovoltaica de capa delgada engloba un grupo de tecnologías capaces de capturar la luz en tan sólo unos pocos nanómetros de espesor. Su bajo costo de manufactura, flexibilidad y bajo peso, hace a las capas delgadas candidatas ideales para la integración en edificios. En particular, las celdas orgánicas pueden proveer una transparencia de alta calidad similar a las ventanas convencionales irrealizable con tecnologías basadas en Silicio. Sin embargo, para la producción de electricidad a gran escala en donde la eficiencia es, tal vez, el factor determinante, existen nuevas tecnologías como las celdas solares de perovskita que pueden resultar más adecuadas. Al momento de escribir esta tesis, las eficiencias de celdas de perovskita de simple unión casi duplican la de las mejores celdas orgánicas de simple unión. Una limitante de ambas tecnologías, en especial de las celdas orgánicas y en menor medida de las perovskitas, es la baja movilidad de las cargas. Esta, junto a otras desventajas de los absorbentes orgánicos y perovskitas limita su espesor al rango de los 100 a los 130 nm, y entre los 500 a 600 nm, respectivamente. En resumen, el manejo de la luz debe constituir un ingrediente esencial para el diseño de los dispositivos, tal que se consiga un desempeño óptimo en la aplicación para la cual sean considerados. En esta tesis, con el fin de alcanzar un aprovechamiento óptimo de la luz y por ende aumentar el desempeño de las celdas solares de capa delgada, utilizamos dos enfoques. Por un lado, aumentamos el espesor total de material absorbente dentro del dispositivo sin incrementar el espesor de las capas actives individuales y por otro lado, combinamos absorbentes complementarios para cubrir una porción más amplia del espectro solar. Estos enfoques conllevan al doble reto de encontrar la distribución de campo electromagnético óptima dentro de una estructura compleja de multicapas con dos o más capas activas, junto a la implementación de una recolección o recombinación de cargas efectiva por parte de las capas intermedias encargadas de conectar dos subceldas adyacentes. En el caso de las celdas orgánicas, consideramos celdas de multiunión usando el mismo material activo para todas las subceldas. Para implementarlas, se realizan estructuras cuyas capas activas no excedan los 100 nm. También estudiamos configuraciones donde los materiales tienen absorciones complementarias usando perovskitas. En ambos casos, sobretodo en el primero, se requiere un método sistemático para optimizar el aprovechamiento de la luz. Para obtener las configuraciones óptimas empleamos una estrategia de integración inversa junto con un cálculo del campo eléctrico basado en el modelo de matriz de transferencia. Además, desarrollamos nuevas estrategias para optimizar la colección de cargas en las capas de interconexión de las subceldas aplicables a dispositivos tipo tandem, triple, 4-terminales y serie-paralelo.

碩士
國立臺灣海洋大學
光電科學研究所
98
Organic photovoltaic devices have gained a broad interest due to their potential for large-area low-cost solar cells. In this thesis, we used a series of novel p-type small organic molecules as the electron donor and fullerene derivatives (C60, PCBM) as the electron acceptor to form simple bulk heterojunction solar cells. Among them, LCC1 shows high hole mobility (h ca.10-4~ 10-5 cm2/Vs), good absorption coefficients, small molecular weight and favourable solution proceessability, which can be applied in thermal evaporation and solution process. Here, we successfully deposited photoactive layer by cosublimation of LCC1:C60 and by spin-coating mixtures of LCC1:PCBM. In thermal evaporation, LCC1:C60 in ratio of 3:1 has best performance. Short circuit current density is 2.74 mA/cm2, while the maximum efficiency could be 0.76 %. In solution process, LCC1:PCBM in ratio of 1:1 that has best performance. Short circuit current density is 3.21 mA/cm2, while the maximum efficiency could be 1.06 %.

碩士
國立中興大學
電機工程學系所
101
In this thesis, hydrogenated amorphous silicon carbide (a-SiC:H) thin-film solar cells were prepared by using 13.56 MHz pulse-modulation plasma-enhanced chemical vapor deposition (Pulse-PECVD). (1) Modulation of plasma turn-on time (ton) to fabricate single i-layer, (2) inserting buffer layers at the p/i and i/n interfaces, and (3) step change ton to prepare graded bandgap (Eg) i-layer, using these experimental design, the influence of different bandgap single i-layer, buffer layers, and graded bandgap i-layer on the performance of a-SiC: H solar cells were investigated. For pulse modulation of ton to produce the single i-layer a-SiC: H with various Eg solar cells, as ton changes from 5 ms to 40 ms, the deposition rate is increased from 0.046 nm/s to 0.199 nm/s, and the Eg is increased from 1.70 eV to 1.765 eV. The absorption coefficient and the density of the film were decreased. The open-circuit voltage (Voc) of the solar cells is increased from 0.822 V to 0.872 V, but short-circuit current density (Jsc) is decreased from 11.91 mA/cm2 to 10.24 mA/cm2, and fill factor (FF) is also decreased from 72.8 % to 66.5 %. Due to the reduction of Jsc and FF, the energy transfer efficiency is decreased from 7.12% to 5.94%. For insertion of a thin (15 nm) a-SiC:H buffer layer at p/i interface, as ton increases from 10 ms, to 20 ms and to 40 ms, the Voc and FF are increased.The performance of a-SiC: H solar cell can be improved with the Voc, Jsc, FF, and energy transfer efficiency f 0.837 V, 11.87 mA/cm2,73.7%, and 7.32 %.For insertion of a thin (15nm) a-SiC: H buffer layer at i/n interface, as ton increases from 5 ms,to 10 ms, and to 20 ms, the Voc,Jsc, and FF are decreased. The performance of a-SiC: H solar cell can be improved with the Voc, Jsc, FF, and energy transfer efficiency of 0.86 V, 10.41 mA/cm2,70.1 %, and 6.72 %. For step change ton to prepare graded bandgap a-SiC:H i-layer solar cells, the graded bandgap solar cell has the effective Eg of 1.732 eV, that the Jsc and FF relative to those of the i-layer with the Eg of 1.765 eV solar cell are increased from 10.41 mA/cm2 to 11.19 mA/cm2,and 70% to 71.7%, but are not higher than the i-layer with the Eg of 1.70 eV solar cell, which the Jsc and FFare 11.87 mA/cm2 and 73.7%. The Voc of graded bandgap solar cell relative to that of the i-layer with the Eg of 1.70 eV solar cellis increased from 0.837 V to 0.860 V, which is very close tothe value (0.862V) of i-layer with the Egof 1.765 eV solar cell. The energy transfer efficiency of graded bandgap solar cell with respect to the i-layer with the Eg of 1.765 eV solar cell is increased from 6.27% to 6.90%, but is not higher than the value (7.32%) of the i-layer with the Eg of 1.70 eV solar cell.

博士
國立東華大學
電機工程學系
106
The two-step process including the deposition of the metal precursors followed by heating the metal precursors in a vacuum environment of Se overpressure was employed for the preparation of Cu(In,Ga)Se2 (CIGS) films. The correlations among the two-step process parameters, film properties, and cell performance were studied. The results demonstrated that the CIGS films selenized at the relatively high Se flow rate of 25 Å/s exhibited the improved surface morphologies. With the given selenization conditions, the efficiency of 12.5% for the fabricated CIGS solar cells was achieved. The features of co-evaporation processes including the single-stage, bi-layer, and three-stage process were discussed. The characteristics of the co-evaporated CIGS solar cells were presented. Not only the surface morphologies but also the grading bandgap structures were crucial to the improvement of the open-circuit voltage of the CIGS solar cells. Efficiencies of over 17% for the co-evaporated CIGS solar cells have been achieved. Furthermore, the critical factors and the mechanisms governing the performance of the CIGS solar cells were addressed. In addition, a baseline model and an advanced model of CdTe solar cells with the selected semiconductor properties fitting to the performance parameters of the champion CdTe solar cells were developed. The responsible factors for the efficiency improvement of the high-performance CdTe solar cell were analyzed. The thin CdS films, and the low defect densities and high carrier mobilities of the CdS films were the crucial factors for the enhancement of the short-circuit current density. With the suppression of carrier recombination, the open-circuit voltage and the fill factor of the CdTe solar cells with the low defect densities in either CdTe films or interdiffusion layer were enhanced. Furthermore, the carrier collection was impeded for the interdiffusion layer with a high defect density, leading to a decrease in the short-circuit current density. Moreover, the simulation results revealed that the efficiency of 20-21% was achieved for the CdTe solar cells with the low defect densities and the high carrier concentrations of the CdTe films.

Perovskite is a class of materials named after their characteristic crystal structure, presenting excellent optoelectronic properties. These properties (particularly, ambipolar charge transport) allow these materials to integrate solar cells with and without the addition of selective layers, like Electron-Transport-Layer and Hole-Transport-Layer, possessing also the ability to be fabricated using simple, non-expensive, solution processing techniques, like Spin-Coating. Coupled with its facile production, the steep rise in SC efficiency over the last years, makes these materials a strong candidate to replace Silicon in the photovoltaic market, both in solid-state and flexible devices. Despite its many advantages, Perovskite SC still face considerable costs regarding processing conditions and HTL materials. By fabricating these devices under ambient air conditions and using Copper(I) Thiocyanate as the material for the HTL, the fabrication costs are significantly reduced. To further lower fabrication cost, Methylammonium Chloride is studied as a replacement for Methylammonium Iodide in Perovskite precursor solution fabrication. Using this solution, crystalline films were obtained, studying several deposition parameters, and the best reported ones, should provide a starting point for further optimization under similar fabrication conditions. The main goal for this work was the optimization of SC, using Spin-Coating technique, keeping the devices as low-cost as possible. Improving Perovskite film quality is detrimental to enhance SC performance, which is why efforts were made to minimize film degradation during Perovskite film fabrication and HTL deposition steps. Due to the hygroscopic nature of the organic component in Perovskite films, the influence of humidity levels was tested, and methods to reduce thin film degradation via humidity exposure were also evaluated. Overall, device optimization was successful, with Perovskite films reaching >95% bulk density and the champion device presenting PCE of 2.65%, with Jsc of 15.11 mA/cm2 and Voc of 0.701V.

The focus of this thesis is on the optimization and fabrication of p-i-n amorphous silicon (a-Si:H) solar cells both on glass and transparent plastic substrates. These solar cells are specifically fabricated on transparent substrates to facilitate the integration of thin film batteries with these solar cells. To comply with plastic substrates, different silicon layers are optimized at the low processing temperature of 135 C. In the first part of the optimization process, the structural, electronic, and optical properties of boron- and phosphorous-doped, hydrogenated nanocrystalline silicon (nc-Si:H) thin films deposited by plasma-enhanced chemical vapor deposition (PECVD) at the substrate temperature of 135 C are elaborated. Additionally, in this part, the deposition of protocrystalline silicon (pc-Si) films on glass substrates are investigated. In the device integration and fabrication part of this thesis, the optimization process is continued by fabricating single junction devices with different hydrogen dilution ratios for the cell absorber layer. The optimum device performance is achieved with an absorber layer right at the transition from amorphous to microcrystalline silicon. To further improve the performance of the fabricated solar cells, amorphous silicon carbide buffer layers are introduced between the nc-Si p-layer and the undoped pc-Si absorber layer. Single junction p-p'-i-n solar cells are fabricated and characterized both on glass and plastic substrates. Our measurements show conversion efficiencies of 7.0% and 6.07% for the cells fabricated on glass and plastic substrates, respectively. In the last part of this research, the light trapping enhancement in amorphous silicon solar cells using Distributed Bragg Reflectors (DBRs) are experimentally demonstrated. Reflectance characteristics of DBR test structures, consisting of amorphous silicon (a-Si) / amorphous silicon nitride (SiN) film stacks are analysed and compared with those of conventional ZnO/Al back reflectors. DBR optical measurements show that the average total reflectance over the wavelength region of 600-800 nm is improved by 28% for DBR back structures. Accordingly, single junction amorphous silicon solar cells with DBR and Al back reflectors are fabricated both on glass and plastic substrates. Our results show that the short-circuit current density and consequently the conversion efficiency is enhanced by 10% for the cells fabricated on textured transparent conductive oxide substrates. In addition, these DBR back structures are designed and employed to improve the efficiency of semi-transparent solar cells. In this application, the optimized DBR structures are designed to be optically transparent for the part of the visible range and highly reflective for the red and infra-red part of the spectrum. Using these DBR structures, the efficiency of the optimum semi-transparent solar cell is enhanced by 5%.

博士
國立中興大學
電機工程學系所
99
Due to the material cost dropping and the efficiency improving, the manufacturing cost can be reduced under US$ 0.7/W. Thin film solar cells become the most potential photovoltaic cells. In this dissertation, the basic theorem and the electrical character of photovoltaic are first introduced; and, the advantage/disadvantage and the limitation of traditional thin film solar cells are discussed. For the solutions, novel thin film solar cells are then provided. The main idea of novel solar cells is based on the dividing methods of photon absorption and carrier transport to improve the photo voltaic efficiency. TCAD, a popular tool for device simulations, is used to verify the idea. Before the simulation, the material character and the model parameters of the silicon-based solar cells, such as density of state for the a-Si(a-SiC) band structure, generation and recombination model and so on, are first studied to calibrate and simulate the device electrical behavior. The AM 1.5 solar radiation and transfer matrix (TMM) are further used to simulate the optical field in the novel cells. Finally, these models are used to simulate, to compare the traditional cells with the novel cells, and to find the efficiency improvement of the novel cells. The different cells dimensions are also used to see the change of the efficiency. The results can prove the thinking for this study and provide the consultation for the future work.

碩士
國立臺北科技大學
材料科學與工程研究所
99
The Cu2O thin films were prepared on Corning Eagle 2000 glass substrate by reactive direct current magnetron sputtering. The influences of oxygen flow rate, argon flow rate and annealing temperature on the structures and properties of deposited films were investigated. Mixtures of Cu, Cu2O, Cu4O3 and CuO with different optical and electrical properties were found to form under different process conditions. Cu2O thin films that are suitable for absorption layer in solar cell can be obtained by reactive magnetron sputtering at 200 °C under 1.65 × 10-2 Torr oxygen pressure and annealed by rapid thermal annealing (RTA) system at different temperatures. Various degree of reduction is also observed by varying annealing temperatures. These films are characterized using UV-VIS photometer, four-point probe system, Hall measurement system, and solar cell efficiency measurement system. It is found that 600 °C annealing is optimum to obtain the required Cu2O thin film structures which also contains trace amount of Cu phase. Such formed p-Cu2O was combined with n-AZO film to form hetrojunction. It demonstrated an open circuit voltage (Voc), short circuit current (Isc) and fill factor (FF) of 0.263 V, 5.72 mA and 0.323, respectively. The value of conversion efficiency is smaller than expected which could possibly be improved by reducing reverse leakage current.

碩士
國立雲林科技大學
電子工程系
102
The indium and gallium of CIGS are rare materials in the earth, and the hydrogen selenide (H2Se) is a toxic gas in selenization process. In recent years, some alternative absorbent materials have been developed for the thin film solar cells, such as Cu2ZnSnS4 (CZTS), Cu2ZnSnSe4 (CZTSe), and other similar quaternary compounds of chalcopyrite structures. In this work, copper zinc tin sulfide (Cu2ZnSnS4) films were investigated as the absorbent layer for the thin film solar cells. The CZTS film is of p-type direct bandgap semiconductor and is suitable as the absorbent layer for the thin-film solar cells due to its low cost, non-toxicity, abundant material and direct bandgap. In this work, the structure of the thin film solar cell with CZTS film is Al/ITO/ZnO/ZnS/CZTS/Mo/SLG. The physical properties of the deposited CZTS films were characterized with Field-Emission Scanning Electron Microscope (FESEM), XRD, UV-VIS analysis. The photovoltaic conversion efficiencies of the CZTS solar cells were measured with solar simulator. The experimental results showed that the CZTS film has good surface morphology with sputtering power of 80W. The surface morphology of the CZTS film is relatively flat which is consistent with the XRD analysis, and after annealing 500℃,there is no material of secondary phase formed. The EDS results show that with increased annealing temperatures, the Cu/Zn+Sn ratio is increased that the ratios of Cu/Zn+Sn, Zn/Sn, S/metal are closer to the ideal values. The optimized annealing temperature is 500℃ that the atomic ratios of Cu/Zn+Sn, Zn/Sn, and S/metal are 0.725, 0.894, and 0.951, respectively. The absorption coefficients of the prepared CZTS films are about 104 cm-1. The energy band gaps of the prepared CZTS films are about 1.5 eV. However, the photovoltaic conversion efficiency of the CZTS thin film solar cell still needs to be improved.

碩士
國立臺灣科技大學
光電工程研究所
102
In this work, GaN surface roughening techniques were developed and used in solar cells to increase the photocurrent density. Typical commercial GaN LED wafers that we used to fabricate GaN solar cells have very thin active layers resulting in quite low solar light absorption. Therefore we added a surface scattering layer by surface roughening to increase the optical length of the incident light in the active layer in order to increase light absorption and photocurrent density. In this work, SiO2 sub-micron spheres of approximately 300 nm in diameter were used as a mask and spin-coated on GaN wafers of more than 70 % in area coverage percentage. Followed by ICP-RIE (inductively-coupled plasma reactive ion etching) dry etching, scatters of various sizes and depths on GaN surface were produced under different setting of processing variables. Then we continuously processed the textured wafers into solar cells and compared their photocurrents with the ones without surface textures. In results, because the roughening of GaN surface also increased Schottky-contact areas and surface defects/surface recombination centers, the open circuit voltage (Voc) was slightly decreased and the series resistance was increased; consequently, the fill factor and the power conversion efficiency were also affected. When the wafer surface was not overly etched, and the width and depth of the scatters were about 174 nm and 64 nm, respectively, the short circuit current density and the external quantum efficiency were improved by ~18% compared to the ones without the surface textures

碩士
國立中興大學
光電工程研究所
98
In this thesis,at first we utilize the method of forming neat PS nanoball lithograth to apply on a-Si glass substrate,and then fabricate a-Si nanorod by using ICP-RIE.Second,we apply hydrogen surface plasma on a-Si nanorod.And last,we utilize PECVD system deposit N-type a-Si thin film, to fabricate nanostructure silicon thin film solar cells. My experiment result show that: in proper the hydrogen plasma power,processing time,pressure.We can get the optimal processing condition of hydrogen surface plasma.Using the optimal hydrogen surface plasma to deal with the a-Si nanorod which had discussed in proper the ICP power,RF power,etching pressure,etching gas flow,etching time.The nanostructure device can be improved by hydrogen surface plasma. Then,we utilize the UV-VIS,and IPCE to disscuss the property of absorptive absorpticn of i nanorod or not,at the same time,we analyze the property of electric to the different type device which are PI/N , P/I nanorod/N , P/I nanorod/H2 plasma/N.