The study aims to enhance the durability of thin-film flexible solar cells by developing a barrier layer that will act as a protective layer against chemical etching during the manufacturing process of the solar cells. This thesis is in collaboration with a Dutch flexible solar company, HyET Solar, which uses a temporary substrate to stack all the layers of the solar cell. The main problem arises when the temporary substrate is required to be removed and replaced by transparent encapsulation. The first layer after the temporary substrate is the TCO that shows defects such as pinholes, which might allow the chemical to penetrate the device and degrade it. The proposed solution is the fabrication of a barrier layer that is deposited before TCO to protect the cell stack from the etching process. Barrier layers were deposited using a scalable, industry-compatible technique, and their performance was assessed through a systematic evaluation of etch resistance and optical transparency. This was followed by optimisation of layer parameters. The findings demonstrate that properly engineered barrier layers can effectively resist chemical etching while maintaining high transparency, thereby enabling their seamless integration into flexible solar modules.

The first amorphous silicon solar cell using a p-i-n structure was developed in 1976 by Carlson and Wronski. In the following decades techniques such as hydrogenation and light trapping texturing were used to raise the efficiency of a-Si thin film solar cell from 2.4% to 9.3%. The research focus has since then shifted from single junction to multi junction solar cells. In a single junction solar cell much of the solar spectrum is underutilised. As photon energy that exceeds the bandgap is wasted as heat, and photons with energy below the bandgap can not be absorbed. In a multi junction structure this problem can be solved by tailoring each layer of the solar cell to a specific range of the solar spectrum, thereby increasing the efficiency.

The focus of this project is to study the potential of germanium tin alloy as a low bandgap material for the bottom layer of a multi junction solar cell. This is done by depositing the a-GeSn:H layers through plasma enhanced chemical vapour deposition(PECVD) with germane and tetramethyltin(TMT) as precursor gas, and examining the optical and electrical properties of the samples made under various deposition conditions. This project will also study the effect of carbon inclusion in germanium, due to it being a side effect of using TMT as a precursor gas.

Photovoltaic solar energy is one of the most powerful renewable sources, playing a key role in transitioning the global energy sector from fossil fuels to zero-carbon emissions. The second generation of photovoltaic technology, such as thin-film silicon-based devices, offers potential for significant performance improvements. Among the layers that can be optimized, the front contact, composed of transparent conductive oxide (TCO) materials, is crucial. As the first layer to encounter incident light, TCO has to meet the high transparency requirement. Additionally, it has to offer high conductivity to transport carriers from the absorber layers to the metal contact. However, there is a trade-off between transparency and conductivity—improving one often compromises the other. Instead of seeking a single layer that balances these properties, an alternative approach involves the use of bilayers. Bilayers consist of two TCOs: one optimized for transparency and the other for conductivity. This thesis study focus on the design of bilayer configurations using two distinct materials, one optimized for transparency and the other for conductivity. Those designs aim to overcome the limitations of single-layer TCOs and achieve superior performance. Indium Cerium Oxide (ICO) and Hydrogenated Indium Oxide (IOH) were selected as conductive materials due to their excellent electrical properties, while intrinsic Zinc Oxide (i-ZnO), known for its high transparency, was used as the transparent layer. Tin Oxide (SnOx), a promising alternative, was also explored and integrated into bilayer configurations. All depositions were performed using Radio Frequency magnetron sputtering. The design process began by optimizing deposition conditions for each material. For ICO, power and process pressure were the examined variables, while for IOH, power and partial water pressure were studied. For SnOx, power and gas composition during deposition were evaluated. After determining the optimal deposition parameters based on opto-electrical properties examination, the following bilayer combinations were produced: IOH/i-ZnO, ICO/i-ZnO, and IOH/SnOx. In these bilayer depositions, the conductive material (IOH or ICO) was deposited first, followed by the transparent layer (i-ZnO or SnOx). The bilayer configurations exhibited superior opto-electrical properties compared to single layers. Specifically, the bilayers maintained high transparency with carrier densities of 1-2×1019 cm−3, minimizing parasitic absorption in the near-infrared region of the solar spectrum. Mobility values of 50-60 cm2/Vs ensured excellent lateral conductivity. These results demonstrated that bilayers preserve the best attribute of each composed material and the combination is a superior option for silicon-based solar cells. The initial depositions were conducted on flat glass substrates. Since silicon-based solar cells use textured substrates, the same bilayer configurations were also deposited on textured surfaces. The results showed that mobility values remained consistent at 50-60 cm2/Vs, though free carrier density was increased to 7-9×1019 cm−3 . However, those bilayers still outperformed single layers, and with further optimization, the issue of reduced free carrier density can be addressed.

This thesis aimed to process and characterize hydrogenated germanium (Ge:H) films to use them as a low-bandgap intrinsic absorber layer. The bandgap of Germanium can go down to 0.67 eV. A p-i-n solar cell consisting of intrinsic germanium film produced using Plasma Enhanced Chemical Vapour Deposition (PECVD) can be inexpensively paired with crystalline silicon and other semiconductor materials to produce a multi-junction solar cell with high efficiencies. Such a cell would utilize the lower part of the solar spectrum currently untapped in commercially available mass-produced solar modules. Previous research had deposited Ge:H films using a counterflow PECVD reactor, as compared to a showerhead configuration in the reactor used in this research.

The study consisted of experiments that explored the properties of the germanium film deposited. First, the deposition parameters of PECVD were varied to study its effect on important film properties. Since the germanium layer will be used in a solar cell as the absorber layer, the band-gap and activation energy were the most critical parameters to monitor film intrinsicity. Experiments were performed by depositing germanium layers on glass substrates and silicon wafers to understand the properties of the films when deposited at different temperatures and also when the germane flow was altered. The effect of the power and pressure on the properties of the germanium samples was studied too. In all these experiments, the deposition time and electrode gap are kept constant. The processing window of 3-4.5 mbar and 12-20 W where intrinsic amorphous germanium (a-Ge:H) films can be produced was identified.

Single junction solar cells in p-i-n and n-i-p configuration were then incorporated with the intrinsic amorphous germanium. The p-i-n configuration displayed solar cell characteristics, whereas the n-i-p configuration cells did not, despite the changes in doping of the p-layer. Further experiments were carried out to look into the stability of the material by repeating the J-V curve measurement of the first cell that showed solar cell characteristics, but similar behaviour was not observed. Additionally, Raman spectroscopy was carried out, which revealed that the material is amorphous.

In summary, the results look promising in proving these films as a suitable material for an intrinsic absorber layer. It also strongly suggests that the showerhead PECVD configuration can lead to better control of reaction mechanics, giving films with desired material properties. The layer can be used as an intrinsic layer in a p-i-n structure device, but further investigation is required to look into the interaction of the layers of the solar cells.

Thin film solar cells are devices that make use of thin layers to generate electricity from solar energy. Among various layers present in a thin film solar cell architecture, transparent conductive oxide (TCO) is crucial. TCOs are a set of materials with a unique set of properties. As a front contact, TCOs are required to offer high transparency to allow photons to reach the absorber layer to generate charge carriers. At the same time, TCOs need to be conductive as well to transport charge carriers from absorber to the metal electrode. However, there exists a trade-off where improvement to one property often comprises the other.
The concept of a bi-layer TCO configuration is explored in this project which makes use of two single layer TCOs: one with superior electrical properties and the other with superior optical properties. Hydrogenated indium oxide (IOH) is investigated as the conductive layer while the intrinsic zinc oxide (i-ZnO) as the transparent layer. Various other conductive layers are also tried in combination with i-ZnO, those being - Cerium doped indium oxide (ICO) and Cerium and hydrogen co-doped indium oxide (ICOH).
The first step to produce highly conductive bi-layer films is optimise post-deposition annealing parameters. Emphasis is placed on optimising annealing time, which tracing out the behavior of change in opto-electrical properties of the bi-layer after every 10 min increment in annealing time. 140 minutes is found as the optimum annealing time for IOH/i-ZnO bi-layers on both flat glass and textured glass substrates.
Novel highly conductive IOH/i-ZnO bi-layer were deposited demonstrating mobility of 143 𝑐𝑚^2/𝑉 𝑠 and carrier concentration of 1.05𝑥10^19 𝑐𝑚^−3 on flat glass. The bi-layer has the ability to not only retain individual constituent TCO properties and also to surpass the limitations of their individual TCOs. This attributes to the fact that the free charge carriers transport takes place a t the interface between TCO layer in the bi-layer stack. The ICO/i-ZnO and ICOH/i-ZnO structures were deposited, both bi-layers demonstrated mobility below 100 𝑐𝑚^2/𝑉 𝑠, however they outperformed their layers both electrically and optically.
IOH/i-ZnO bi-layer construction was also deposited on textured glass substrates. Best sample demonstrates mobility as high as 110 𝑐𝑚^2/𝑉 𝑠 with a carrier concentration of 3.63𝑥10^19 𝑐𝑚^−3. The reason for slightly lower mobilities is attributed to the inconsistency in shape and size of the textures present on the substrate, which leads to non-uniform deposition of IOH across the entire surface. Further investigation will lead to understanding hydrogen’s role in the IOH/i-ZnO bi-layer and the extent to which bi-layer performance on textured substrates can be improved.

Thin film silicon solar cells are a type of photovoltaic technology with the advantage of having thin layers of silicon to generate electricity. Unlike traditional crystalline silicon solar cells, which use thick silicon wafers, thin-film silicon solar cells use amorphous silicon that can be made with much thinner layers of silicon. This allows them to be more lightweight, flexible, and potentially less expensive to manufacture. However, their efficiency has traditionally been lower than crystalline silicon solar cells, which are more
commonly used in large-scale installations. This is primarily due to the material property of amorphous silicon, which is laden with defects and voids. Nonetheless, ongoing research and development aim to improve the efficiency and commercial viability of thin-film silicon solar cells.

HyET Solar B.V. is a company based in the Netherlands which employs a Roll to Roll (R2R) technology to produce such flexible solar cells. A temporary aluminum foil is used as the substrate on which the solar cell stack is deposited. The temporary foil is etched away, and the layers are encapsulated in low-cost polymer foils. This thesis is part of the ongoing FlamingoPV (Flexible Lightweight Advanced Materials In Next Generation of PV) project in collaboration between HyET Solar and TU Delft, to develop single, tandem, and triple junction cells with 12, 13, and 14% efficiencies, and a lifetime longer than 35 years. Part of the ongoing research to improve the performance of thin-film silicon solar cells is to understand the difference between the thin films deposited on a rigid glass substrate and flexible aluminum substrate and to investigate why the performance is lower on the aluminum substrate. In
particular, special emphasis is given to the origin of shunts in thin-film silicon solar cells, and conducting a top-down root cause analysis to investigate the origin of these shunts. Mitigation strategies are suggested to improve the solar cells fabricated on foil. Apart from this, the solar cell layers are optimized for improved electrical and optical performance on the glass substrate, to ultimately implement it on the aluminum substrate.

The key takeaways from this research are that aluminum foil was found to be a major culprit for the origin of the shunts. The foil consists of alloying elements of iron and copper, which get exposed to the surface when the foil is cleaned. When the subsequent layers are grown on the foil, the alloying elements were observed to short-circuit the device, thus causing leakage currents. Another major culprit was the formation and accumulation of silicon dust on the samples during the PECVD deposition, which was more prevalent on devices fabricated on aluminum foil. The key takeaways from the optimization experiments are that by the band-gap profiling of i-layer at a lower thickness (230nm) than the standard (300nm), we could maintain the initial electrical and optical properties of the devices. This gives us room to reduce material usage and costs at the same output performance. A permanent degradation was observed in the metal contacts of these devices, apart from the temporary light-induced degradation commonly seen in a-Si:H based solar cells. The tests conducted to improve the electrical performance did give the desired results, with an increase in the efficiency of the devices. The tests conducted to improve the optical performance did not give the desired results, with a decrease in the electrical performance and no significant increase in the optical response of the devices.

This report emphasizes the importance of implementing quality control methods in the production machines at HyET Solar for assessing the quality of deposited materials. The focus is on two key steps: the deposition of the front transparent conductive oxide (TCO) layer and the deposition of silicon layers. Models were developed to characterize the quality of the TCO layer, including layer thickness, carrier concentration, and mobility of the electrons. A Drude model-based approach was chosen for in-line implementation. The sheet resistance of the TCO layer was measured using an improved tool, by enhancing its robustness and accuracy. To assess the quality of the deposited silicon layers, a non-contacting capacitive device was developed and calibrated (to measure the spatial gap between the photovoltaic layer and the device). Results showed that the Drude model fitting provided sensible results for carrier concentration and mobility with the former having a better correlation than the latter, with the standard Hall effect measurements. The results from the optical thickness model showed overestimation of values as compared to the scanning electron microscopy (SEM) results. The electrical sheet resistance values correlated well with Hall effect measurements than the optically obtained sheet resistance values. The calibrated non-contacting capacitive device demonstrated its ability to measure opto-electrical properties of the silicon layers, accurately. This research contributes to enhancing the efficiency and reliability of roll-to-roll production process of thin-film solar modules.