Numerical Front Contact Grid Optimization for Photovoltaics
Nikander, Veikka (2023)
2023
Teknis-luonnontieteellinen DI-ohjelma - Master's Programme in Science and Engineering
Tekniikan ja luonnontieteiden tiedekunta - Faculty of Engineering and Natural Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2023-05-19
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202304274800
https://urn.fi/URN:NBN:fi:tuni-202304274800
Tiivistelmä
Photovoltaic converters are a key part of the power by light systems, which transfer energy via light rather than electrical conduction or induction. They excel in applications requiring galvanic isolation, electromagnetic interference-free operation, and ultra-lightweight transmission lines. Power density is a key performance indicator for any photovoltaic chip, but this is even more true for the photovoltaic converter. Two main loss mechanisms start to play a limiting role as higher power densities are used: heating-related power loss and resistive losses. A common way to tackle the resistive loss of photovoltaic devices is to introduce front contact metallization on top of the chip to aid in the charge carrier transport at the lateral direction of the cell.
Numerical front contact grid optimization for two single junction photovoltaic converters is done with a novel Hybrid quasi-3D simulation model. The model is constructed by a collection of established optical and semiconductor simulation models giving an efficient yet reliable and accurate description of the photovoltaic device. Double busbar finger grid design’s finger spacing is optimized for a similar but inverted p-type on n-type (p–n) and n-type on p-type (n–p) structures. Optimization is done for four relevant illumination profiles and three operation temperatures, to yield a comprehensive picture of the behavior between the two structures and different grid spacings. Analysis of the photovoltaic cell performance parameters V_oc, I_sc, and FF is done the gain a deeper knowledge of the cell behavior.
Results predict that the n–p structure performs considerably better at high illumination power compared to its p–n counterpart. The n–p structure yielded twice as large spacings and 8.5-9.6% higher output powers. Tolerance to spacing values was also considerably higher for n–p structures than p–n. Both illumination non-uniformity and temperature were found to decrease the optimal spacing values, with illumination having a stronger negative effect. Non-uniform illumination and temperature also had a negative effect on the spacing tolerance. Analysis of FF values revealed that with p–n structure, the device’s low sheet resistance of the p-GaAs emitter starts to limit the cell performance as wider spacings are used. Also, results predicted V_oc values to decrease with increased spacing with non-uniform illumination. This effect was explained by a thought experiment of a voltage source matrix with non-uniform voltages trying to equalize their potential difference.
The Hybrid quasi-3D model proved to be a highly useful tool for simulating photovoltaic devices. The efficient 3D model has high value when the chip designer wants to move beyond the 1D layer level design to chip and module level design. The ability to consider the layer structure, illumination profile, and metal grid designs in the same model, gives the designer access to results needed to create world-class photovoltaics and to strive in the highly competitive industry of photovoltaics.
Numerical front contact grid optimization for two single junction photovoltaic converters is done with a novel Hybrid quasi-3D simulation model. The model is constructed by a collection of established optical and semiconductor simulation models giving an efficient yet reliable and accurate description of the photovoltaic device. Double busbar finger grid design’s finger spacing is optimized for a similar but inverted p-type on n-type (p–n) and n-type on p-type (n–p) structures. Optimization is done for four relevant illumination profiles and three operation temperatures, to yield a comprehensive picture of the behavior between the two structures and different grid spacings. Analysis of the photovoltaic cell performance parameters V_oc, I_sc, and FF is done the gain a deeper knowledge of the cell behavior.
Results predict that the n–p structure performs considerably better at high illumination power compared to its p–n counterpart. The n–p structure yielded twice as large spacings and 8.5-9.6% higher output powers. Tolerance to spacing values was also considerably higher for n–p structures than p–n. Both illumination non-uniformity and temperature were found to decrease the optimal spacing values, with illumination having a stronger negative effect. Non-uniform illumination and temperature also had a negative effect on the spacing tolerance. Analysis of FF values revealed that with p–n structure, the device’s low sheet resistance of the p-GaAs emitter starts to limit the cell performance as wider spacings are used. Also, results predicted V_oc values to decrease with increased spacing with non-uniform illumination. This effect was explained by a thought experiment of a voltage source matrix with non-uniform voltages trying to equalize their potential difference.
The Hybrid quasi-3D model proved to be a highly useful tool for simulating photovoltaic devices. The efficient 3D model has high value when the chip designer wants to move beyond the 1D layer level design to chip and module level design. The ability to consider the layer structure, illumination profile, and metal grid designs in the same model, gives the designer access to results needed to create world-class photovoltaics and to strive in the highly competitive industry of photovoltaics.