1.
L Helmich; D C Walter; R Falster; V V Voronkov; J Schmidt
In: Solar Energy Materials and Solar Cells, Bd. 232, S. 111340, 2021, ISSN: 0927-0248.
@article{Helmich2021c,
title = {Impact of hydrogen on the boron-oxygen-related lifetime degradation and regeneration kinetics in crystalline silicon},
author = {L Helmich and D C Walter and R Falster and V V Voronkov and J Schmidt},
doi = {10.1016/j.solmat.2021.111340},
issn = {0927-0248},
year = {2021},
date = {2021-10-01},
journal = {Solar Energy Materials and Solar Cells},
volume = {232},
pages = {111340},
abstract = {We examine the impact of hydrogen on the boron-oxygen-related lifetime degradation and regeneration kinetics in boron-doped p-type Czochralski-grown silicon wafers. We introduce the hydrogen into the silicon bulk by rapid thermal annealing. The hydrogen source are hydrogen-rich silicon nitride (SiNx:H) layers. Aluminum oxide (Al2O3) layers of varying thickness are placed in-between the silicon wafer surfaces and the SiNx:H layers. By varying the Al2O3 thickness, which acts as an effective hydrogen diffusion barrier, the hydrogen bulk content is varied over more than one order of magnitude. The hydrogen content is determined from measured wafer resistivity changes. In order to examine the impact of hydrogen on the degradation kinetics, all samples are illuminated at a light intensity of 0.1 suns near room temperature. We observe no impact of the in-diffused hydrogen content on the degradation rate constant, confirming that hydrogen is not involved in the boron-oxygen degradation mechanism. The regeneration experiments at 160°C and 1 sun, however, show a clear dependence on the hydrogen content with a linear increase of the regeneration rate constant with increasing bulk hydrogen concentration. However, extrapolation of our measurements toward a zero in-diffused hydrogen content shows that the regeneration is still working even without any in-diffused hydrogen. Hence, our measurements demonstrate that there are two distinct regeneration processes taking place. This is in good agreement with a recently proposed defect reaction model and is also in agreement with the finding that the permanent boron-oxygen deactivation also works on non-fired solar cells, though at a lower rate.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
We examine the impact of hydrogen on the boron-oxygen-related lifetime degradation and regeneration kinetics in boron-doped p-type Czochralski-grown silicon wafers. We introduce the hydrogen into the silicon bulk by rapid thermal annealing. The hydrogen source are hydrogen-rich silicon nitride (SiNx:H) layers. Aluminum oxide (Al2O3) layers of varying thickness are placed in-between the silicon wafer surfaces and the SiNx:H layers. By varying the Al2O3 thickness, which acts as an effective hydrogen diffusion barrier, the hydrogen bulk content is varied over more than one order of magnitude. The hydrogen content is determined from measured wafer resistivity changes. In order to examine the impact of hydrogen on the degradation kinetics, all samples are illuminated at a light intensity of 0.1 suns near room temperature. We observe no impact of the in-diffused hydrogen content on the degradation rate constant, confirming that hydrogen is not involved in the boron-oxygen degradation mechanism. The regeneration experiments at 160°C and 1 sun, however, show a clear dependence on the hydrogen content with a linear increase of the regeneration rate constant with increasing bulk hydrogen concentration. However, extrapolation of our measurements toward a zero in-diffused hydrogen content shows that the regeneration is still working even without any in-diffused hydrogen. Hence, our measurements demonstrate that there are two distinct regeneration processes taking place. This is in good agreement with a recently proposed defect reaction model and is also in agreement with the finding that the permanent boron-oxygen deactivation also works on non-fired solar cells, though at a lower rate.
2.
M Winter; D Walter; D Bredemeier; J Schmidt
In: Solar Energy Materials and Solar Cells, Bd. 201, S. 110060, 2019, ISSN: 0927-0248.
@article{Winter2019c,
title = {Light-induced lifetime degradation effects at elevated temperature in Czochralski-grown silicon beyond boron-oxygen-related degradation},
author = {M Winter and D Walter and D Bredemeier and J Schmidt},
doi = {10.1016/j.solmat.2019.110060},
issn = {0927-0248},
year = {2019},
date = {2019-10-01},
journal = {Solar Energy Materials and Solar Cells},
volume = {201},
pages = {110060},
abstract = {The effect of ‘Light and elevated Temperature Induced Degradation’ (LeTID) of the carrier lifetime is well known from multicrystalline silicon (mc-Si) wafers and solar cells. In this contribution, we perform a series of carrier lifetime measurements to examine, whether the same effect may also be observable in boron-doped Czochralski-grown silicon (Cz-Si). The Cz-Si samples of our study are illuminated (i) at room temperature, (ii) under standard regeneration conditions eliminating the boron-oxygen (BO) related defect (i.e. at 185 °C) and (iii) at a temperature of 80 °C, typical for the examination of the LeTID effect in mc-Si. We observe the typical decay of the carrier lifetime due to the activation of the BO-related defect. Beyond the BO degradation, applying standard solar cell processes, there is no indication for the activation of a second defect. On samples, whose surfaces are passivated by fired hydrogen-rich silicon nitride layers, an additional bulk lifetime degradation effect on a long timescale is observed in the Cz-Si material. However, defect generation rate and injection dependence of the lifetime suggest another defect type than the mc-Si-specific LeTID defect. We conclude that by applying processing steps that trigger LeTID in mc-Si, the same defect does not occur in the Cz-Si samples examined in this study. On a long timescale, however, a hitherto unknown type of defect is activated, which is different from the mc-Si-specific LeTID defect. A careful differentiation between the various kinds of recombination centres which may form during illumination at elevated temperatures is hence of utmost importance.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The effect of ‘Light and elevated Temperature Induced Degradation’ (LeTID) of the carrier lifetime is well known from multicrystalline silicon (mc-Si) wafers and solar cells. In this contribution, we perform a series of carrier lifetime measurements to examine, whether the same effect may also be observable in boron-doped Czochralski-grown silicon (Cz-Si). The Cz-Si samples of our study are illuminated (i) at room temperature, (ii) under standard regeneration conditions eliminating the boron-oxygen (BO) related defect (i.e. at 185 °C) and (iii) at a temperature of 80 °C, typical for the examination of the LeTID effect in mc-Si. We observe the typical decay of the carrier lifetime due to the activation of the BO-related defect. Beyond the BO degradation, applying standard solar cell processes, there is no indication for the activation of a second defect. On samples, whose surfaces are passivated by fired hydrogen-rich silicon nitride layers, an additional bulk lifetime degradation effect on a long timescale is observed in the Cz-Si material. However, defect generation rate and injection dependence of the lifetime suggest another defect type than the mc-Si-specific LeTID defect. We conclude that by applying processing steps that trigger LeTID in mc-Si, the same defect does not occur in the Cz-Si samples examined in this study. On a long timescale, however, a hitherto unknown type of defect is activated, which is different from the mc-Si-specific LeTID defect. A careful differentiation between the various kinds of recombination centres which may form during illumination at elevated temperatures is hence of utmost importance.
3.
L Helmich; D C Walter; D Bredemeier; R Falster; V V Voronkov; J Schmidt
In-situ characterization of electron-assisted regeneration of Cz-Si solar cells Artikel
In: Solar Energy Materials and Solar Cells, Bd. 185, S. 283-286, 2018, ISSN: 0927-0248.
@article{Helmich2018b,
title = {In-situ characterization of electron-assisted regeneration of Cz-Si solar cells},
author = {L Helmich and D C Walter and D Bredemeier and R Falster and V V Voronkov and J Schmidt},
doi = {10.1016/j.solmat.2018.05.023},
issn = {0927-0248},
year = {2018},
date = {2018-10-01},
journal = {Solar Energy Materials and Solar Cells},
volume = {185},
pages = {283-286},
abstract = {Abstract We examine the regeneration kinetics of passivated emitter and rear solar cells (PERCs) fabricated on boron-doped p-type Czochralski-grown silicon wafers in darkness by electron injection via application of a forward bias voltage at elevated temperature (140 °C) in order to discriminate between electronic and photonic effects. Based on these dark regeneration experiments, we address the existing inconsistency regarding the measured linear dependence of the regeneration rate constant on the excess carrier density. Using the method of dark regeneration by current injection into the solar cell, we are able to measure the total recombination current of the solar cell at the actual regeneration temperature under applied voltage, i.e., at the physically relevant regeneration conditions. The direct comparison of the regeneration rate constant as a function of electronically injected carrier concentration in the dark and the regeneration rate constant during illumination clearly shows that the regeneration is a purely electronically stimulated effect and that photons are not directly involved.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Abstract We examine the regeneration kinetics of passivated emitter and rear solar cells (PERCs) fabricated on boron-doped p-type Czochralski-grown silicon wafers in darkness by electron injection via application of a forward bias voltage at elevated temperature (140 °C) in order to discriminate between electronic and photonic effects. Based on these dark regeneration experiments, we address the existing inconsistency regarding the measured linear dependence of the regeneration rate constant on the excess carrier density. Using the method of dark regeneration by current injection into the solar cell, we are able to measure the total recombination current of the solar cell at the actual regeneration temperature under applied voltage, i.e., at the physically relevant regeneration conditions. The direct comparison of the regeneration rate constant as a function of electronically injected carrier concentration in the dark and the regeneration rate constant during illumination clearly shows that the regeneration is a purely electronically stimulated effect and that photons are not directly involved.
4.
D C Walter; L Helmich; D Bredemeier; J Schmidt; R Falster; V V Voronkov
Lifetime Evolution during Regeneration in Boron-Doped Czochralski-Silicon Proceedings Article
In: WIP, (Hrsg.): Proceedings of the 35th European Photovoltaic Solar Energy Conference and Exhibition, S. 522-526, Brussels, Belgium, 2018.
@inproceedings{Walter2018,
title = {Lifetime Evolution during Regeneration in Boron-Doped Czochralski-Silicon},
author = {D C Walter and L Helmich and D Bredemeier and J Schmidt and R Falster and V V Voronkov},
editor = {WIP},
doi = {10.4229/35thEUPVSEC20182018-2AV.1.29},
year = {2018},
date = {2018-09-24},
booktitle = {Proceedings of the 35th European Photovoltaic Solar Energy Conference and Exhibition},
pages = {522-526},
address = {Brussels, Belgium},
abstract = {We measure the evolution of the carrier lifetime in boron-doped Czochralski-grown silicon wafers for the first time in-situ during permanent deactivation of the boron-oxygen defect under the applied conditions, i.e. illumination with a halogen lamp at elevated temperatures. Applying illumination intensities ≥1 sun on standard 1.5 Ωcm p-type Cz-Si, the lifetime (measured at the regeneration temperature) changes only negligibly during this regeneration process. As expected, in this case, the time dependence of the defect concentration (measured at room temperature) follows a single-exponential decay function during regeneration. For light intensities << 1 sun on the same material, the lifetime shows a significant change during the regeneration conditions and the evolution of the defect concentration does no longer follow a single-exponential decay curve. In addition, on 0.5 Ωcm p-type Cz-Si we observe a non-exponential decay of the defect concentration for a regeneration treatment performed at 1 sun illumination intensity and a single-exponential decay for a higher illumination intensity of 2.9 suns. These observations are well compatible with a deactivation rate increasing proportionally with the excess carrier concentration.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
We measure the evolution of the carrier lifetime in boron-doped Czochralski-grown silicon wafers for the first time in-situ during permanent deactivation of the boron-oxygen defect under the applied conditions, i.e. illumination with a halogen lamp at elevated temperatures. Applying illumination intensities ≥1 sun on standard 1.5 Ωcm p-type Cz-Si, the lifetime (measured at the regeneration temperature) changes only negligibly during this regeneration process. As expected, in this case, the time dependence of the defect concentration (measured at room temperature) follows a single-exponential decay function during regeneration. For light intensities << 1 sun on the same material, the lifetime shows a significant change during the regeneration conditions and the evolution of the defect concentration does no longer follow a single-exponential decay curve. In addition, on 0.5 Ωcm p-type Cz-Si we observe a non-exponential decay of the defect concentration for a regeneration treatment performed at 1 sun illumination intensity and a single-exponential decay for a higher illumination intensity of 2.9 suns. These observations are well compatible with a deactivation rate increasing proportionally with the excess carrier concentration.
5.
D C Walter; V Steckenreiter; L Helmich; T Pernau; J Schmidt
Production-Compatible Regeneration of Boron-Doped Czochralski-Silicon in a Combined Fast-Firing and Regeneration Belt-Line Furnace Proceedings Article
In: WIP, (Hrsg.): Proceedings of the 33rd European Photovoltaic Solar Energy Conference and Exhibition, S. 377-381, Amsterdam, The Netherlands, 2017, ISBN: 3-936338-47-7.
@inproceedings{Walter2017,
title = {Production-Compatible Regeneration of Boron-Doped Czochralski-Silicon in a Combined Fast-Firing and Regeneration Belt-Line Furnace},
author = {D C Walter and V Steckenreiter and L Helmich and T Pernau and J Schmidt},
editor = {WIP},
doi = {10.4229/EUPVSEC20172017-2CO.9.4},
isbn = {3-936338-47-7},
year = {2017},
date = {2017-09-27},
booktitle = {Proceedings of the 33rd European Photovoltaic Solar Energy Conference and Exhibition},
pages = {377-381},
address = {Amsterdam, The Netherlands},
abstract = {Boron-doped Czochralski-grown silicon (Cz-Si), as used in the industrial production of solar cells today, is suffering from a light-induced degradation (LID) of the carrier lifetime during illumination. It has been known since a decade that under lab conditions, this degradation can be permanently cured by illumination at increased temperature, which has been known as ‘regeneration’ treatment. In this contribution, we examine the permanent regeneration of the carrier lifetime in standard boron-doped Cz-Si using an industrial combined fast-firing and regeneration furnace - the c.FIRE REG of centrotherm photovoltaics. Our study reveals that for an optimized firing and regeneration process, very high implied open-circuit voltages Voc.impl exceeding 740 mV are achieved on standard 1-2 cm p-type Cz-Si wafers. These very high measured Voc.impl values clearly indicate the suitability of the process for highly efficient PERC cell production. Additionally, as the c.FIRE REG furnace is realized in a belt-line configuration, which exposes each wafer for less than 70 s to the firing and regeneration conditions, our experiments clearly demonstrate the excellent suitability of the examined tool for the implementation into industrial solar cell production lines. We also demonstrate the successful application to industrial PERC solar cells produced by two different manufacturers.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Boron-doped Czochralski-grown silicon (Cz-Si), as used in the industrial production of solar cells today, is suffering from a light-induced degradation (LID) of the carrier lifetime during illumination. It has been known since a decade that under lab conditions, this degradation can be permanently cured by illumination at increased temperature, which has been known as ‘regeneration’ treatment. In this contribution, we examine the permanent regeneration of the carrier lifetime in standard boron-doped Cz-Si using an industrial combined fast-firing and regeneration furnace - the c.FIRE REG of centrotherm photovoltaics. Our study reveals that for an optimized firing and regeneration process, very high implied open-circuit voltages Voc.impl exceeding 740 mV are achieved on standard 1-2 cm p-type Cz-Si wafers. These very high measured Voc.impl values clearly indicate the suitability of the process for highly efficient PERC cell production. Additionally, as the c.FIRE REG furnace is realized in a belt-line configuration, which exposes each wafer for less than 70 s to the firing and regeneration conditions, our experiments clearly demonstrate the excellent suitability of the examined tool for the implementation into industrial solar cell production lines. We also demonstrate the successful application to industrial PERC solar cells produced by two different manufacturers.
6.
V Steckenreiter; D C Walter; J Schmidt
In: Energy Procedia, Bd. 124, Nr. Supplement C, S. 799-805, 2017, ISSN: 1876-6102, (7th International Conference on Silicon Photovoltaics, SiliconPV 2017, 3-5 April 2017, Freiburg, Germany).
@article{Steckenreiter2017c,
title = {Two-stage permanent deactivation of the boron-oxygen-related recombination center in crystalline silicon},
author = {V Steckenreiter and D C Walter and J Schmidt},
doi = {10.1016/j.egypro.2017.09.350},
issn = {1876-6102},
year = {2017},
date = {2017-09-21},
journal = {Energy Procedia},
volume = {124},
number = {Supplement C},
pages = {799-805},
abstract = {We analyze the lifetime evolution during permanent deactivation of the boron-oxygen-related defect center (BO defect) in boron-doped, oxygen-rich Czochralski-grown silicon (Cz-Si). In particular, we examine the impact of the samples’ states prior to the permanent deactivation process. Samples that were initially fully degraded show a two-stage deactivation process consisting of a fast and a slow deactivation component, which can be fitted by two exponential functions with their respective rate constants. For both components, we find a pronounced increase of the rate constants with illumination intensity. In addition, we observe that the rate constant describing the slow deactivation component of samples deactivated after complete degradation is identical to the rate constant determined on samples, which were deactivated immediately after annealing in darkness. In the latter case, a purely mono-exponential deactivation behavior was observed. Our study clearly demonstrates that the asymptotic deactivation behavior does not depend on the initial state of the lifetime sample. We prove that the same is valid for initially degraded and dark-annealed PERC solar cells. Hence, it is not necessary to first degrade the sample to realize a fast BO deactivation.},
note = {7th International Conference on Silicon Photovoltaics, SiliconPV 2017, 3-5 April 2017, Freiburg, Germany},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
We analyze the lifetime evolution during permanent deactivation of the boron-oxygen-related defect center (BO defect) in boron-doped, oxygen-rich Czochralski-grown silicon (Cz-Si). In particular, we examine the impact of the samples’ states prior to the permanent deactivation process. Samples that were initially fully degraded show a two-stage deactivation process consisting of a fast and a slow deactivation component, which can be fitted by two exponential functions with their respective rate constants. For both components, we find a pronounced increase of the rate constants with illumination intensity. In addition, we observe that the rate constant describing the slow deactivation component of samples deactivated after complete degradation is identical to the rate constant determined on samples, which were deactivated immediately after annealing in darkness. In the latter case, a purely mono-exponential deactivation behavior was observed. Our study clearly demonstrates that the asymptotic deactivation behavior does not depend on the initial state of the lifetime sample. We prove that the same is valid for initially degraded and dark-annealed PERC solar cells. Hence, it is not necessary to first degrade the sample to realize a fast BO deactivation.
7.
D Walter; B Lim; J Schmidt
In: Progress in Photovoltaics: Research and Applications, Bd. 24, Nr. 7, S. 920-928, 2016.
@article{Walter2016b,
title = {Realistic efficiency potential of next-generation industrial Czochralski-grown silicon solar cells after deactivation of the boron-oxygen-related defect center},
author = {D Walter and B Lim and J Schmidt},
doi = {10.1002/pip.2731},
year = {2016},
date = {2016-07-01},
journal = {Progress in Photovoltaics: Research and Applications},
volume = {24},
number = {7},
pages = {920-928},
abstract = {We measure carrier lifetimes of different Czochralski‐grown silicon (Cz‐Si) materials of various boron and oxygen concentrations and determine the maximum achievable lifetime after an optimized thermal treatment. We obtain very high and stable bulk lifetimes of several milliseconds, virtually eliminating the boron–oxygen (BO) defect complex, which previously limited the carrier lifetime in boron‐doped Cz‐Si materials after prolonged illumination. Based on these experimental results, we introduce a new parameterization of the bulk lifetime of B‐doped Cz‐Si after permanent deactivation of the BO center. Notably, we measure lifetimes up to 4 ms on 2‐Ωcm Cz‐Si wafers at an injection level of 1/10 of the doping concentration. Importantly, these high lifetime values can be reached within 10 and 20 s of BO deactivation treatment. A detailed analysis of the injection‐dependent lifetimes reveals that the lifetimes after permanent deactivation of the BO center can be well described by a single‐level recombination center characterized by an electron‐to‐hole capture cross‐section ratio of 12 and located in the middle of the silicon band gap. We implement the novel parameterization into a two‐dimensional device simulation of a passivated emitter and rear solar cell using technologically realistic cell parameters. The simulation reveals that based on current state‐of‐the‐art solar cell production technology, efficiencies reaching 22.1% are realistically achievable in the near future after complete deactivation of the BO center.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
We measure carrier lifetimes of different Czochralski‐grown silicon (Cz‐Si) materials of various boron and oxygen concentrations and determine the maximum achievable lifetime after an optimized thermal treatment. We obtain very high and stable bulk lifetimes of several milliseconds, virtually eliminating the boron–oxygen (BO) defect complex, which previously limited the carrier lifetime in boron‐doped Cz‐Si materials after prolonged illumination. Based on these experimental results, we introduce a new parameterization of the bulk lifetime of B‐doped Cz‐Si after permanent deactivation of the BO center. Notably, we measure lifetimes up to 4 ms on 2‐Ωcm Cz‐Si wafers at an injection level of 1/10 of the doping concentration. Importantly, these high lifetime values can be reached within 10 and 20 s of BO deactivation treatment. A detailed analysis of the injection‐dependent lifetimes reveals that the lifetimes after permanent deactivation of the BO center can be well described by a single‐level recombination center characterized by an electron‐to‐hole capture cross‐section ratio of 12 and located in the middle of the silicon band gap. We implement the novel parameterization into a two‐dimensional device simulation of a passivated emitter and rear solar cell using technologically realistic cell parameters. The simulation reveals that based on current state‐of‐the‐art solar cell production technology, efficiencies reaching 22.1% are realistically achievable in the near future after complete deactivation of the BO center.
8.
D C Walter; B Lim; K Bothe; R Falster; V V Voronkov; J Schmidt
Lifetimes exceeding 1 ms in 1-Ohm cm boron-doped Cz-silicon Artikel
In: Solar Energy Materials and Solar Cells, Bd. 131, S. 51-57, 2014, (SiliconPV 2014).
@article{Walter2014c,
title = {Lifetimes exceeding 1 ms in 1-Ohm cm boron-doped Cz-silicon},
author = {D C Walter and B Lim and K Bothe and R Falster and V V Voronkov and J Schmidt},
doi = {10.1016/j.solmat.2014.06.011},
year = {2014},
date = {2014-12-01},
journal = {Solar Energy Materials and Solar Cells},
volume = {131},
pages = {51-57},
note = {SiliconPV 2014},
keywords = {},
pubstate = {published},
tppubtype = {article}
}