S Bordihn; B Min; R Peibst; R Brendel
Modelling of Passivation and Resistance of n-Type poly-Si Layers by Trained Artificial Neural Networks Proceedings Article
In: WIP, (Hrsg.): Proceedings of the 36th European Photovoltaic Solar Energy Conference and Exhibition, S. 176-179, Marseille, France, 2019, ISBN: 3-936338-60-4.
@inproceedings{Bordihn2019b,
title = {Modelling of Passivation and Resistance of n-Type poly-Si Layers by Trained Artificial Neural Networks},
author = {S Bordihn and B Min and R Peibst and R Brendel},
editor = {WIP},
doi = {10.4229/EUPVSEC20192019-2BO.3.1},
isbn = {3-936338-60-4},
year = {2019},
date = {2019-10-23},
booktitle = {Proceedings of the 36th European Photovoltaic Solar Energy Conference and Exhibition},
pages = {176-179},
address = {Marseille, France},
abstract = {This paper studies the passivation quality and sheet resistance of n-type poly-Si on oxide (POLO) layers that we prepare at various post-deposition annealing temperatures. The n-type poly-Si layers are 50 nm-thin and grown on textured and planar Cz Si substrates. The samples are annealed from 880 °C to 1000 °C to transform the amorphous Si to poly-crystalline Si. The surface passivation quality is evaluated after the aforementioned anneal step, after an additional hydrogenation step (induced by Al2O3 capping layers and subsequent low temperature anneal), and after firing at 840 °C peak temperature. The optimal surface passivation quality is found for annealing at 960 °C and hydrogenation, resulting in iVoc-values of 710 mV. The hydrogenation step improves the iVoc by ~20 mV depending on the post-deposition annealing temperature. We find that surface passivation and sheet resistance of the poly-Si layers increased with increasing anneal temperature up to 960 °C and starts to decline above 980 °C. This trend is found to correlate with the amount of in-diffused dopants that depends also on the annealing temperature. We show that artificial neural network based model can serve as a fast tool for predicting layer properties that depend on multiple process parameters. The quality of the modelling is the same as that using the Design of Experiment method.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
S Bordihn; B Min; R Peibst; R Brendel
Modelling of Passivation and Resistance of n-Type poly-Si Layers by Trained Artificial Neural Networks Vortrag
Marseille, France, 10.09.2019, (36th European Photovoltaic Solar Energy Conference and Exhibition).
@misc{Bordihn2019,
title = {Modelling of Passivation and Resistance of n-Type poly-Si Layers by Trained Artificial Neural Networks},
author = {S Bordihn and B Min and R Peibst and R Brendel},
year = {2019},
date = {2019-09-10},
address = {Marseille, France},
abstract = {This paper studies the passivation quality and sheet resistance of n-type poly-Si on oxide (POLO) layers that we prepare at various post-deposition annealing temperatures. The n-type poly-Si layers are 50 nm-thin and grown on textured and planar Cz Si substrates. The samples are annealed from 880 °C to 1000 °C to transform the amorphous Si to poly-crystalline Si. The surface passivation quality is evaluated after the aforementioned anneal step, after an additional hydrogenation step (induced by Al2O3 capping layers and subsequent low temperature anneal), and after firing at 840 °C peak temperature. The optimal surface passivation quality is found for annealing at 960 °C and hydrogenation, resulting in iVoc-values of 710 mV. The hydrogenation step improves the iVoc by ~20 mV depending on the post-deposition annealing temperature. We find that surface passivation and sheet resistance of the poly-Si layers increased with increasing anneal temperature up to 960 °C and starts to decline above 980 °C. This trend is found to correlate with the amount of in-diffused dopants that depends also on the annealing temperature. We show that artificial neural network based model can serve as a fast tool for predicting layer properties that depend on multiple process parameters. The quality of the modelling is the same as that using the Design of Experiment method.},
note = {36th European Photovoltaic Solar Energy Conference and Exhibition},
keywords = {},
pubstate = {published},
tppubtype = {presentation}
}
B Min; H Wagner; M Müller; D -H Neuhaus; P P Altermatt
Incremental efficiency improvements of mass-produced PERC cells up to 24%, predicted solely with continuous development of existing technologies and wafer materials Proceedings Article
In: WIP, (Hrsg.): Proceedings of the 31st European Photovoltaic Solar Energy Conference, S. 473-476, Hamburg, Germany, 2015, ISBN: 3-936338-39-6.
@inproceedings{Min2015,
title = {Incremental efficiency improvements of mass-produced PERC cells up to 24%, predicted solely with continuous development of existing technologies and wafer materials},
author = {B Min and H Wagner and M Müller and D -H Neuhaus and P P Altermatt},
editor = {WIP},
doi = {10.4229/EUPVSEC20152015-2DO.3.3},
isbn = {3-936338-39-6},
year = {2015},
date = {2015-09-14},
booktitle = {Proceedings of the 31st European Photovoltaic Solar Energy Conference},
journal = {Proceedings of the 31st European Photovoltaic Solar Energy Conference},
pages = {473-476},
address = {Hamburg, Germany},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}