Progress in Photovoltaics: Research and Applications

Volume 31, Issue 5 pp. 536-545

RESEARCH ARTICLE

Open Access

Temperature and illumination dependence of silicon heterojunction solar cells with a wide range of wafer resistivities

Anh Huy Tuan Le,

Corresponding Author

Anh Huy Tuan Le

[email protected]

orcid.org/0000-0002-9395-0771

School of Photovoltaic and Renewable Energy Engineering, University of New South Wales, Sydney, New South Wales, 2052 Australia

Correspondence

Anh Huy Tuan Le and Ziv Hameiri, School of Photovoltaic and Renewable Energy Engineering, University of New South Wales, Sydney, New South Wales 2052, Australia.

Email: [email protected] and [email protected]

André Augusto, School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, AZ 85287, USA.

Email: [email protected]

Search for more papers by this author

Apoorva Srinivasa,

Apoorva Srinivasa

School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, Arizona, 85287 USA

Search for more papers by this author

Stuart G. Bowden,

Stuart G. Bowden

School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, Arizona, 85287 USA

Search for more papers by this author

Ziv Hameiri,

Corresponding Author

Ziv Hameiri

[email protected]

School of Photovoltaic and Renewable Energy Engineering, University of New South Wales, Sydney, New South Wales, 2052 Australia

Correspondence

Anh Huy Tuan Le and Ziv Hameiri, School of Photovoltaic and Renewable Energy Engineering, University of New South Wales, Sydney, New South Wales 2052, Australia.

Email: [email protected] and [email protected]

André Augusto, School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, AZ 85287, USA.

Email: [email protected]

Search for more papers by this author

André Augusto,

Corresponding Author

André Augusto

[email protected]

School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, Arizona, 85287 USA

Correspondence

Anh Huy Tuan Le and Ziv Hameiri, School of Photovoltaic and Renewable Energy Engineering, University of New South Wales, Sydney, New South Wales 2052, Australia.

Email: [email protected] and [email protected]

André Augusto, School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, AZ 85287, USA.

Email: [email protected]

Search for more papers by this author

Anh Huy Tuan Le,

Corresponding Author

Anh Huy Tuan Le

[email protected]

orcid.org/0000-0002-9395-0771

School of Photovoltaic and Renewable Energy Engineering, University of New South Wales, Sydney, New South Wales, 2052 Australia

Correspondence

Anh Huy Tuan Le and Ziv Hameiri, School of Photovoltaic and Renewable Energy Engineering, University of New South Wales, Sydney, New South Wales 2052, Australia.

Email: [email protected] and [email protected]

André Augusto, School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, AZ 85287, USA.

Email: [email protected]

Search for more papers by this author

Apoorva Srinivasa,

Apoorva Srinivasa

School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, Arizona, 85287 USA

Search for more papers by this author

Stuart G. Bowden,

Stuart G. Bowden

School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, Arizona, 85287 USA

Search for more papers by this author

Ziv Hameiri,

Corresponding Author

Ziv Hameiri

[email protected]

School of Photovoltaic and Renewable Energy Engineering, University of New South Wales, Sydney, New South Wales, 2052 Australia

Correspondence

Anh Huy Tuan Le and Ziv Hameiri, School of Photovoltaic and Renewable Energy Engineering, University of New South Wales, Sydney, New South Wales 2052, Australia.

Email: [email protected] and [email protected]

André Augusto, School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, AZ 85287, USA.

Email: [email protected]

Search for more papers by this author

André Augusto,

Corresponding Author

André Augusto

[email protected]

School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, Arizona, 85287 USA

Correspondence

Anh Huy Tuan Le and Ziv Hameiri, School of Photovoltaic and Renewable Energy Engineering, University of New South Wales, Sydney, New South Wales 2052, Australia.

Email: [email protected] and [email protected]

André Augusto, School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, AZ 85287, USA.

Email: [email protected]

Search for more papers by this author

First published: 22 December 2022

https://doi.org/10.1002/pip.3657

Funding information: Australian Renewable Energy Agency, Grant/Award Number: 2017/RND001; Solar Energy Technologies Office, Grant/Award Number: DE-EE0008549; U.S. Department of Energy, PV Foundry, Grant/Award Number: DE-EE0008975

Share a link

Email
Wechat
Bluesky

Abstract

Recently, the significant improvements in the surface and contact passivation of silicon (Si) solar cells as well as their bulk quality have shifted their operating point to higher injections. Hence, they are less dependent on wafer doping. This shift opens an opportunity of using high-resistivity wafers for practical photovoltaic applications, introducing a promising approach to push the cell efficiency towards the intrinsic limit and to improve the module reliability by increasing the cell breakdown voltage. Therefore, insights into the performance of Si solar cells using high-resistivity wafers at various operating temperatures are of significant interest. In this study, we investigate the temperature- and illumination-dependent performance of Si heterojunction (SHJ) solar cells using a wide range of wafer resistivities (between 3 and 1000 Ω⋅cm). Although a reduction in the passivation quality of the passivating contacts is observed at elevated temperature, the impact on the temperature coefficient of the open-circuit voltage (TC_Voc)—the dominant contributor to the temperature coefficient (TC) of the cell efficiency—is very limited. Their TC_Voc are still dominated by the temperature dependence of the effective intrinsic carrier concentration. Furthermore, we also find that the investigated cells are more sensitive to temperature variation at lower illumination intensities. It is noteworthy that the efficiency of the cells fabricated using high-resistivity wafers is comparable to that of the reference cells at any given temperature, highlighting the potential of using high-resistivity wafers for solar cells.

1 INTRODUCTION

It is well known that the fundamental limit of silicon (Si) solar cells, as an indirect bandgap material, is defined by the Auger recombination.^{1, 2} In principle, this recombination can be reduced by using high-resistivity (low doped) wafers.^{1, 2} However, due to the compromise between the doping density and excess carrier concentration (Δn) [or effective lifetime (τ_eff)] at the maximum power point,³ Si wafers with bulk resistivities below 10 Ω⋅cm are usually used for commercial solar cells.^{4, 5} This compromise can be relaxed if both the bulk quality and the surface and contacts passivation are significantly improved.

Recently, significant improvements in the quality of the passivating contacts, used to reduce the recombination at the Si-metal interfaces, have been reported.^6-10 Solar cells that integrate these contacts such as Si heterojunction (SHJ),^{11, 12} interdigitated back contact,^{6, 13} polycrystalline Si on oxide (POLO),⁷ tunnel oxide passivated contact (TOPCon)^{8, 14} and transition-metal-oxide-based cells^{9, 10} exhibit high open-circuit voltages (V_oc) above 720 mV due to high quality passivation of both surfaces and contacts as indicated by extremely low surface saturation current densities (J_0s)¹⁵ below 10 fA/cm². These cell structures also benefit from the high bulk lifetime (τ_bulk) of commercially n-type Czochralski (Cz) substrates. Nowadays, ingot manufacturers often deliver τ_bulk above 3 ms across the entire ingot,¹⁶ with the top parts reaching τ_bulk of 15 ms.¹⁷ High τ_bulk coupled with excellent surface and contact passivation can push those solar cell operating points to the high injection regime at which Δn is usually at the order of 10¹⁵ cm⁻³.^{18, 19} In this case, for high-resistivity wafers, their effective bulk resistivity (the bulk resistivity under the operating conditions) is significantly lower than their dark bulk resistivity as the former is defined by Δn at the high injection regime, which is remarkably higher than the doping density. Therefore, it is expected that the detrimental impacts on the fill factor (FF) under the operating conditions when using high-resistivity wafers for Si solar cells are no longer an issue. Hence, the above solar cell architectures may benefit from using wafers with higher bulk resistivities.^{18, 19}

Studies of solar cells fabricated using high bulk resistivity substrates (>10 Ω⋅cm), beyond the commercial resistivity range, are very limited. Glunz et al.²⁰ investigated the efficiencies of TOPCon cells as a function of bulk resistivity (up to 100 Ω⋅cm) using modelling. The simulations revealed that cell performance is independent of the bulk resistivity in the range of 5–100 Ω⋅cm. Jay et al.¹⁹ reported that the efficiencies of SHJ solar cells with bulk resistivities in the range of 5–10 Ω⋅cm are similar to those of cells fabricated using 367-Ω⋅cm wafers. Simulated efficiencies of other industrially relevant cell structures with bulk resistivities in the range of 10–100 Ω⋅cm are shown to be comparable to those cells that use 1–10 Ω⋅cm wafers when the Shockley–Read–Hall (SRH) lifetimes (τ_SRH) of 1 and 10 ms are assumed as the dominate bulk recombination.²¹ For cells fabricated on wafers with even higher resistivity (>500 Ω⋅cm), it has been shown by simulations that they can outperform cells made on standard wafers (<10 Ω⋅cm) if their τ_SRH is in the millisecond range.²² This offers a great opportunity to significantly reduce the Auger recombination and hence, to push the cell efficiency of current technologies closer to the maximum theoretical efficiency. Furthermore, using high-resistivity wafers can improve the module reliability by increasing the cell breakdown voltage.²³ More importantly, there is a severe lack of experimental data of solar cells with such high bulk resistivities.

The application of high resistivity cells in the field must consider the temperature sensitivity of their electrical parameters. It is well known that the realistic operating temperatures of Si solar cells significantly deviate from the standard testing conditions (STC; at 25°C with an irradiance of 1000 W/m² under the AM1.5G solar spectrum), which are usually used for the characterisation and optimisation of solar cells.²⁴ Therefore, the temperature coefficient (TC) of high resistivity cells is a critical parameter that needs to be studied. More importantly, this parameter enables accurate evaluation of the energy yield of the power systems and hence, plays a crucial role in strategies of optimisation for solar cells and photovoltaic systems at different climatic conditions. Surprisingly, none of the previous studies determine the performance of cells with different bulk resistivities under realistic operating temperatures.

Here, we first investigate the temperature-dependent performance of SHJ solar cells fabricated using a wide range of wafer resistivities at one-sun. We then study the temperature dependence of J_0s of hydrogenated amorphous Si (a-Si:H)-based passivating contacts to gain a deeper understanding regarding its impacts on the TC of the open-circuit voltage (TC_Voc)—the dominant contribution to the TC of the cell efficiency (TC_η)—of the investigated solar cells. Finally, we examine the illumination-dependent TC_Voc of these cells, which allows the determination of their temperature sensitivity under a wide range of illumination intensities.

2 EXPERIMENTAL DETAILS

2.1 Sample preparation

In this study, standard (180 ± 10 μm) and thick (380 ± 10 μm) 4-in. float zone (FZ) n-type Si wafers with crystal orientation of <100> and bulk resistivities of 3 Ω⋅cm (reference), 75 Ω⋅cm (medium), and ≥1000 Ω⋅cm (high) were used to fabricate both SHJ solar cells and test structures. The wafer surfaces were first textured using an alkaline solution, which was followed by a three-step cleaning sequence that includes the Radio Corporation of America (RCA) procedure,²⁵ Piranha etching²⁶ and hydrofluoric acid (HF) etching prior to plasma-enhanced chemical vapour deposition (PECVD).

For the fabrication of the SHJ solar cells, a stack of intrinsic and p-type a-Si:H [a-Si:H(i) and a-Si:H(p), respectively] layers was formed on the front side using a PECVD system. A stack of a-Si:H(i) and n-type a-Si:H [a-Si:H(n)] films was deposited on the rear side using the same system. The thicknesses of the a-Si:H(i) and doped a-Si:H films are between 6–7 and 10–15 nm, respectively. A 75-nm indium tin oxide (ITO) film was then deposited on the front side by a sputtering system through a mask to form active areas of 2 × 2 cm². The front grid metallisation was achieved by screen-printing of low temperature silver (Ag) paste followed by a curing process at 200°C for 30 min in air ambient. The rear contact was formed by a full-area stack of 240-nm sputtered ITO and 240-nm sputtered Ag layers. Further details regarding the PECVD and sputtering deposition conditions can be found in Balaji et al.²⁷

To investigate the temperature-dependent behaviour of the a-Si:H-based passivating contacts, symmetrical lifetime structures for τ_eff measurements and J_0s extraction were prepared. These structures were passivated with either a-Si:H(i/p) or a-Si:H(i/n) stack on both sides.

All the test structures were annealed at 200°C for 30 min in air ambient to mimic the thermal budget of the Ag paste curing process. Sketches of all the studied cells and test structures are shown in Figure 1.

Details are in the caption following the image — **FIGURE 1**
Open in figure viewer PowerPoint

Sketches of (A) the cell structure and (B,C) symmetrical test structures for τ_eff measurements

2.2 Characterisation

Current–voltage (I-V) measurements of SHJ solar cells were performed by an I-V tester from 25°C to 70°C, while Suns-V_oc measurements were done using a customised Suns-V_oc system^{28, 29} from 80°C to 30°C. The uncertainty associated with the temperature controller for both measurement tools is estimated at ±0.1°C. After reaching the set temperature, the solar cells were left on the heated stage for 10 min to stabilise before the measurements were done. At the completion of the temperature scan, the measurement was repeated at 25°C to ensure that the properties of the solar cells were not modified by this scan.

TCs are determined from the slopes of the linear fits of the cell parameters as a function of temperature.³⁰ They are normalised to the cell parameters at 25°C and presented as relative TCs in this study.

A spectral response system (QEX7, PV Measurements Inc.) was used to measure the external quantum efficiency (EQE) of the studied solar cells.

The symmetrical lifetime test structures were used for τ_eff measurements in the temperature range from 25°C to 80°C using a WCT-120TS lifetime tester (from Sinton Instruments).³¹ The Kane–Swanson method³² is used to extract J_0s from the τ_eff curves. The model of Richter et al.³³ is used to calculate the intrinsic lifetime, while the bandgap narrowing model of Schenk³⁴ and the model of Klaassen³⁵ are used to determine the effective intrinsic carrier concentration (n_i,eff) and the mobility, respectively.

Based on previous studies,^36-38 light and elevated-temperature induced degradation (LeTID) should be considered when heating and illuminating Si wafers. However, it was reported that in n-type FZ wafers, after 5 s of light soaking at 140°C and light intensity of one-sun, τ_eff reduces by less than 5%.³⁶ In this study, all the measurements were shorter than 5 s and at lower temperatures (25°C–80°C); hence, a significant impact by LeTID is not expected. Furthermore, the measurements at 25°C before and after the temperature scan are identical, confirming that LeTID does not affect the studied cells.

3 RESULTS AND DISCUSSION

3.1 Temperature-dependent performance of SHJ solar cells at one-sun

The cell parameters of the SHJ solar cells with different wafer resistivities fabricated using standard thickness and thick wafers at one-sun as a function of temperature are shown in Figure 2A–H. Overall, as expected, the V_oc, FF, pseudo fill factor (pFF, fill factor without the effects of series resistance), and the efficiency decrease, whereas the short-circuit current density (J_sc) increases at elevated temperatures.³⁰ The reduction of V_oc, FF, and pFF can be explained by the increase of n_i,eff with increasing temperature caused by bandgap narrowing.³⁹ This effect also explains the improvement of J_sc.³⁰ Non-linear behaviour of FF in the temperature range between 25°C and 40°C, regardless of the wafer resistivity and thickness, is observed. This phenomenon commonly occurs in SHJ solar cells and is usually attributed to thermionic barriers at the heterojunctions in these cells.⁴⁰

It is noteworthy that at a given temperature, the V_oc of the 180-μm-thick cells (termed hereafter as ‘standard’ cells, which refer only to their thickness) is higher than that of the thicker cells. The expression for the voltage (V) of n-type solar cells at high injections (Δn ≫ N_D, where N_D is the dopant density) is given by:¹⁷

V_{hi} = \frac{k_{B} T}{q} \ln (\frac{(J_{ph} - J) τ_{eff} Δ n}{{qWn}_{i, eff}^{2}})

(1)

where k_B is the Boltzmann constant, T is the temperature, q is the elementary charge, J_ph is the photogenerated current density, J is the current density, and W is the wafer thickness. Assuming all the other parameters do not change, the reduction of the wafer thickness from 380 to 180 μm contributes to an improvement of ~20 mV in V_oc. However, changes in the numerator on the right side of this expression limit this improvement to the range of 10–12 mV.

Interestingly, for the standard thickness cells, the decrease of V_oc at elevated temperatures is less pronounced with decreasing wafer resistivity while the decrease of V_oc at elevated temperatures is almost identical regardless of the wafer resistivity in the case of the thicker cells. We will discuss further on these observations in the latter part of this section.

At a given temperature, J_sc of the thick cells is higher than that of the standard thickness cells, due to an increase in the optical absorptance at long wavelengths with increasing wafer thickness (see Appendix A, Figure A1).⁴¹ As pointed out by Dupré et al., the temperature sensitivity of Si solar cells' J_sc depends on both the bandgap and the collection efficiency (f_c), as described in the following equation²⁴:

\frac{1}{J_{sc}} \frac{{dJ}_{sc}}{dT} = \frac{1}{J_{sc, ideal}} \frac{{dJ}_{sc, ideal}}{{dE}_{g}} \frac{{dE}_{g}}{dT} + \frac{1}{f_{c}} \frac{{df}_{c}}{dT}

(2)

where J_sc,ideal is the current equivalent to the sum of all photons with energy greater than the bandgap in the incident spectrum and E_g is the bandgap energy. The first term in Equation (2) represents the contribution of the bandgap narrowing to the temperature dependence of J_sc while the second term indicates the contribution of f_c. Although the studied cells use different wafer resistivities, the difference in the reducing rate of the Si bandgap at elevated temperature is negligible (<0.002 meV/°C, see Appendix B, Figure B1). As shown in Figure 2B,F, the increase of J_sc at elevated temperatures is almost identical regardless of the wafer resistivity and thickness. Hence, a similar temperature dependence of f_c for all the studied cells can be assumed.

Figure 2C,G shows FF and pFF of the investigated SHJ solar cells as a function of temperature. At STC, the FF of the standard thickness cells slightly decreases with increasing wafer resistivity while the thick cells' FF is almost identical. It can be attributed to the thickness-dependent resistive loss, which has also been reported in Sai et al.⁴² At elevated temperatures, the decrease of FF and pFF is more pronounced with increasing wafer resistivity for the thinner cells. A difference between the FF and pFF trends is observed in the 3- and 1k-Ω⋅cm standard thickness cells. It highlights the contribution of the cells' series resistance (R_s) to the temperature dependence of FF and hence, the TC of the FF (TC_FF). For the thick cells, the decrease of FF and pFF at elevated temperatures is almost identical regardless of the wafer resistivity, indicating that the R_s values of these cells are less sensitive to the temperature compared to the standard thickness cells. As pointed out by Zhao et al.,⁴³ the temperature-dependent behaviour of Si solar cells' FF depends on both V_oc and R_s. To quantify the contributions of these two parameters to the temperature sensitivity of the studied cells' FF, the following equation is used:^{30, 43}

\frac{1}{FF} \frac{dFF}{dT} = (1 - 1.02 {FF}_{0}) (\frac{1}{V_{oc}} \frac{{dV}_{oc}}{dT} - \frac{1}{T}) - \frac{R_{s}}{V_{oc} / I_{sc} - R_{s}} (\frac{1}{R_{s}} \frac{{dR}_{s}}{dT})

(3)

where

{FF}_{0} = \frac{ν_{oc} - \ln (ν_{oc} + 0.72)}{ν_{oc} + 1}

(4)

where FF₀ is the FF in the absence of R_s and shunt resistance (R_sh) effects and ν_oc is the normalised V_oc to the thermal voltage (k_BT/q). The first term in Equation (3) reflects the contribution of V_oc to the temperature dependence of FF while the second term represents the contribution of R_s. From calculations using Equation (3), we find that the contribution of V_oc is dominant and accounts for more than 70% of TC_FF. Hence, the temperature-dependent behaviour of FF reflects the behaviour of V_oc and is dominated by the increase of n_i,eff at elevated temperatures caused by bandgap narrowing.

The efficiency of the cells as a function of temperature is presented in Figure 2D,H. At STC, the efficiency of the standard thickness cells slightly decreases with increasing wafer resistivity, mainly due to FF, whereas the thick cells' efficiency is almost identical. The superior V_oc and FF of the standard thickness cells overcompensate for their lower J_sc, and therefore their efficiency is slightly higher than the thick cells. At elevated temperatures, as expected, the reduction in efficiency is more pronounced with increasing wafer resistivity for the standard thickness cells because of the temperature dependence of both the V_oc and FF. Meanwhile, this reduction is almost identical for the thick cells regardless of the wafer resistivity. Note that the non-linear behaviour of the efficiency in the temperature range between 25°C and 40°C regardless of the wafer resistivity and thickness is due to FF, as depicted in Figure 2C,G.

Figure 3A clearly shows that the V_oc of the standard thickness cells is more sensitive to temperatures with increasing wafer resistivity, while the thick cells' TC_Voc is almost independent of the resistivity. TC_Voc of the latter cells is nearly equal to that of the standard thickness cell with wafer resistivity of 1k Ω⋅cm. The obtained TC_Voc are comparable to those previously reported.⁴² The superior TC_Voc of thinner cells has been also reported in the same reference.

It is well known that the temperature sensitivity of V_oc of a solar cell depends on both the V_oc and the gamma factor (γ) as described in the following equation³:

\frac{{dV}_{oc}}{dT} = - \frac{E_{g 0} / q - V_{oc} + γ k_{b} T / q}{T}

(5)

where E_g0 is the semiconductor bandgap linearly extrapolated to 0 K. γ represents the temperature-dependent diode saturation current density and, hence, contains information about the dominant recombination mechanism in the solar cells.^{3, 24, 44} Since V_oc of the standard thickness cells at 25°C is almost identical regardless of the wafer resistivity (see Figure 2A), we hypothesise that the difference in TC_Voc of these cells might be due to different γ. To calculate γ, E_g0 is required. E_g0 extrapolated from E_g as a function of temperature for the different resistivity wafers are almost identical (not shown here) and are equal to 1.206 eV. Therefore, this value is applied to all the calculations of γ for the different resistivity wafers. The calculated γ of the cells at 25°C based on Equation 5 is summarised in Table 1. We observe that γ of the thick cells is less sensitive to the wafer resistivity, whereas γ of the standard thickness cells shows a larger dependence on the wafer resistivity. The performance of cells made from thinner wafers is more sensitive to the recombination at their surfaces.⁴⁵ Thus, the different γ of the standard thickness cells may indicate a surface-related dominant recombination mechanism that might be impacted by the wafer resistivity. Ponce-Alcántara et al.⁴⁶ reported a wafer resistivity dependence of TC_Voc of Si solar cells, which can be justified by the relative variation of the diode saturation current density with temperature (indicating γ variation). More directly, Berthod et al.⁴⁷ attributed the wafer resistivity dependence of passivated emitter and rear (PERC) cells' TC_Voc to a sensible difference in the value of γ. Note that the solar cells reported in the two studies have the same wafer thickness (180–200 μm) as our standard thickness cells. These results, therefore, support the assumption above.

TABLE 1. γ at STC of the SHJ solar cells with different wafer resistivities using standard thickness and thick wafers

Wafer resistivity (Ω⋅cm)	Gamma factor (γ)
Wafer resistivity (Ω⋅cm)	Standard thickness wafers	Thick wafers
3	1.3	1.8
75	1.8	1.7
1k	2.6	1.7

Abbreviations: SHJ, Si heterojunction; STC, standard testing conditions.

From Figure 3B, it seems that the TC of J_sc (TC_Jsc) is almost identical regardless of the wafer resistivity and thickness, highlighting the interesting point that was discussed above. The obtained TC_Jsc are comparable to that reported in Haschke et al.⁴⁰ and lower than those presented in Sai et al.⁴² This difference might be due to the spectrum-dependent TC_Jsc when using different solar simulators.⁴⁸ The contribution of TC_Jsc to TC_η is the smallest for all the studied cells (see Figure 3E).

In Figure 3C, it can be seen that TC_FF of the 3-Ω⋅cm standard thickness cell is better than its TC of the pFF (TC_pFF), while the opposite trend is observed for the 1k-Ω⋅cm standard thickness cell. Meanwhile, the TC_FF of the standard thickness cell with the wafer resistivity of 75 Ω⋅cm and all the thick cells is almost identical to their TC_pFF. This highlights a different temperature dependence of R_s. Compared to the literature,^{40, 42} the obtained TC_FF of the 3-Ω⋅cm standard thickness cells is slightly better while that of the 3-Ω⋅cm thick cells is slightly worse. Note that Sai et al.⁴² also reported better TC_FF for thinner cells.

As expected, the TC_η of all the investigated cells is dominated by TC_Voc, with a contribution above 60%, regardless of the wafer resistivity and thickness (see Figure 3E). It is therefore not surprising that the TC_η of the standard thickness cells becomes worse with increasing wafer resistivity, while it is almost identical for the thick cells. The obtained TC_η are slightly lower compared to the values reported in Sai et al.⁴² (3-Ω⋅cm cells), mainly due to the higher TC_Jsc of the cells in this reference.

From the above discussion, we find that the temperature dependence of V_oc dominates that of the cell performance regardless of the wafer resistivity and thickness. Therefore, it is important to gain a deeper understanding regarding the impacts of passivating contacts on the cells' TC_Voc as well as knowledge regarding their performance at different illumination intensities (as usually occur in the field). In the next sections, we will focus on these two aspects.

3.2 Temperature dependence of surface passivation

To assess the temperature dependence of the surface passivation, J_0s values were extracted from τ_eff measurements of symmetrical test structures and are presented in Figure 4A–D. The τ_eff at Δn = 10¹⁵ cm⁻³ and 25°C are summarised in Table C1 (Appendix C). At 25°C, the extracted J_0s values are comparable to those reported in Le et al.⁴⁹ At elevated temperatures, as expected, J_0s significantly increases regardless of the wafer thickness or the configuration of the test structures. It is due to an increase of n_i,eff by several orders of magnitude.³² The J_0s/n_i,eff² ratio as a function of temperature is also presented in the figures. This ratio increases with increasing temperature indicating a slight reduction of the passivation quality at elevated temperatures. Bernardini et al.⁵⁰ also reported an increase of the surface recombination velocity of FZ n-type Si wafers passivated by 50-nm a-Si:H(i) layers with increasing temperature, which is consistent with our findings. ${T C}_{J_{0 s} / n_{i, eff} 2}$ values extracted from the linear fits of J_0s/n_i,eff² ratio as a function of temperature are provided in Table 2.

TABLE 2. Extracted TCs of the J_0s/n_i,eff² ratio and their statistical errors determined from the standard deviation of the linear regression for the symmetrical test structures

Wafer resistivity (Ω⋅cm)	${T C}_{J_{0 s} / n_{i, eff} 2}$ (%/°C)
Wafer resistivity (Ω⋅cm)	Standard thickness wafers i/p	Standard thickness wafers i/n	Thick wafers i/p	Thick wafers i/n
3	0.43 ± 0.03	0.48 ± 0.05	0.35 ± 0.05	0.43 ± 0.08
75	0.21 ± 0.04	0.41 ± 0.07	0.21 ± 0.01	0.46 ± 0.06
1k	0.24 ± 0.03	0.53 ± 0.04	0.18 ± 0.03	0.51 ± 0.08

Abbreviation: TC, temperature coefficients.

For standard thickness wafers, the reduction of the passivation quality of the a-Si:H(i/p)-based test structures is less pronounced for wafers with high resistivity (75 Ω⋅cm and above). Meanwhile, the reduction of the passivation quality of the a-Si:H(i/n)-based test structures is almost identical regardless of the wafer resistivity. Similar trends are observed for the test structures using thick wafers. Interestingly, ${T C}_{J_{0 s} / n_{i, eff} 2}$ values of the a-Si:H(i/p)-based and a-Si:H(i/n)-based test structures are almost the same for the standard thickness and thick wafers. Nevertheless, it seems the observed reduction in the passivation quality in all the cases has only an insignificant impact on TC_Voc. The temperature-dependent behaviour of V_oc and hence TC_Voc, is still dominated by that of n_i,eff.

3.3 Injection and temperature dependence of SHJ solar cells

In the previous sections, we discussed the temperature-dependent performance of the investigated cells and test structures at one-sun. However, in the field, solar cells operate not only at different temperatures, but also under a wide range of illumination intensities. Therefore, insights into their performance at combinations of various temperatures and intensities are of significant interest. In this section, we use temperature-dependent Suns-V_oc measurements in between 0.001 and >100 suns to extract TC_V_oc as a function of illumination intensity. This information can be used to predict the temperature sensitivity of the cells at different illumination intensities as TC_V_oc dominates TC_η for all the investigated cells.

Figure 5A–F presents the Suns-V_oc measurements in the temperature range from 30°C to 80°C. A significant reduction of V_oc with increasing temperature at low illumination intensities can be observed regardless of the wafer resistivity and thickness. This reduction becomes less pronounced at higher illumination intensities (above one-sun). The extracted relative TC_V_oc values from the Suns-V_oc measurements as a function of illumination intensity are presented in Figure 5G,H as open symbols. For comparison, TC_V_oc values obtained from I-V measurements (solid symbols) at one-sun are also shown. As can be seen, TC_V_oc values at one-sun extracted from both methods match very well (in the range of 2% and 0.2% for the standard and thick cells, respectively), supporting the observed trend of the temperature-dependent V_oc as discussed in Section 3.1. All the investigated cells are more sensitive to temperature variation at lower illumination intensities as indicated by the more negative TC_V_oc. Interestingly, below one-sun, the illumination intensity dependence of TC_V_oc is more pronounced for the standard thickness cells using higher resistivity wafers. Meanwhile, the TC_V_oc values of the thick cells behave the same across the entire measured intensity range regardless of the wafer resistivities. To clarify this observation, the extracted V_oc values at 25°C are presented in Figure 5I,J. In most of the illumination range, the trend of TC_V_oc can be explained by that of the initial V_oc. However, for the standard thickness cells, it seems that γ has a significant impact on their TC_V_oc in the illumination range of 0.3–2 suns in which their TC_V_oc become more negative with increasing wafer resistivity despite their similarly initial V_oc.

4 CONCLUSION

The temperature- and illumination-dependent performance of SHJ solar cells using high wafer resistivities (up to 1000 Ω⋅cm) was investigated. The TCs of the main electrical cell parameters of these cells are almost independent of the wafer resistivity. Furthermore, we also found that the investigated cells are more sensitive to temperature variation at lower illumination intensities. The findings of this study demonstrate that there is no limitation in using very high-resistivity wafers for SHJ solar cells under field operating conditions. The study, therefore, set a premise for using such wafers for reducing the Auger recombination and to push SHJ solar cell efficiency towards their theoretical efficiency limit.

ACKNOWLEDGEMENTS

This work was funded by the Australian Government through Australian Renewable Energy Agency [ARENA; project 2017/RND001]. The views expressed herein are not necessarily the views of the Australian Government, and the Australian Government does not accept responsibility for any information or advice contained herein. This material is also based upon work supported by the U.S. Department of Energy's Office of Energy Efficiency and Renewable Energy (EERE) under the Solar Energy Technology Office (SETO), award number DE-EE0008549 and by the U.S. Department of Energy, PV Foundry, grant DE-EE0008975. Open access publishing facilitated by University of New South Wales, as part of the Wiley - University of New South Wales agreement via the Council of Australian University Librarians.

APPENDIX A: EQE SPECTRA OF THE INVESTIGATED CELLS

APPENDIX B: BANDGAP NARROWING OF SI AT ELEVATED TEMPERATURES

APPENDIX C: τ_eff OF THE SYMMETRICAL LIFETIME TEST STRUCTURES AT Δn = 10¹⁵ cm⁻³ AND 25°C

TABLE C1. τ_eff of the symmetrical lifetime test structures at Δn = 10¹⁵ cm⁻³ and 25°C

Wafer resistivity (Ω⋅cm)	τ_eff at Δn = 10¹⁵ cm⁻³ and 25°C (ms)
Wafer resistivity (Ω⋅cm)	Standard thickness wafers i/p	Standard thickness wafers i/n	Thick wafers i/p	Thick wafers i/n
3	1.41	2.72	2.21	4.72
75	2.40	2.75	2.14	8.64
1k	1.48	1.28	2.45	5.52

Open Research

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

REFERENCES

1Richter A, Hermle M, Glunz SW. Reassessment of the limiting efficiency for crystalline silicon solar cells. IEEE J Photovolt. 2013; 3(4): 1184-1191. doi:10.1109/JPHOTOV.2013.2270351
10.1109/JPHOTOV.2013.2270351
Web of Science® Google Scholar
2Veith-Wolf BA, Schäfer S, Brendel R, Schmidt J. Reassessment of intrinsic lifetime limit in n-type crystalline silicon and implication on maximum solar cell efficiency. Sol Energy Mater sol Cells. 2018; 186: 194-199. doi:10.1016/j.solmat.2018.06.029
10.1016/j.solmat.2018.06.029
CAS Web of Science® Google Scholar
3Green MA. Solar cells: Operating principles, technology, and system applications; 1982.
Google Scholar
4Zhao L, Li H, Zhou C, Diao H, Wang W. Optimized resistivity of p-type Si substrate for HIT solar cell with Al back surface field by computer simulation. Solar Energy. 2009; 83(6): 812-816. doi:10.1016/j.solener.2008.11.007
10.1016/j.solener.2008.11.007
CAS Web of Science® Google Scholar
5Srinivasa A, Herasimenka S, Augusto A, Bowden S. Effect of ingot variability on performance of silicon heterojunction solar cells. 47^th IEEE Photovoltaic Specialists Conference; 2020: 2238-2241.
Google Scholar
6Yoshikawa K, Kawasaki H, Yoshida W, et al. Silicon heterojunction solar cell with interdigitated back contacts for a photoconversion efficiency over 26%. Nat Energy. 2017; 2(5): 17032. doi:10.1038/nenergy.2017.32
10.1038/nenergy.2017.32
CAS Web of Science® Google Scholar
7Haase F, Hollemann C, Schafer S, et al. Laser contact openings for local poly-Si-metal contacts enabling 26.1%-efficient POLO-IBC solar cells. Sol Energ Mat sol C. 2018; 186: 184-193. doi:10.1016/j.solmat.2018.06.020
10.1016/j.solmat.2018.06.020
CAS Web of Science® Google Scholar
8Richter A, Benick J, Feldmann F, Fell A, Hermle M, Glunz SW. n-type Si solar cells with passivating electron contact: Identifying sources for efficiency limitations by wafer thickness and resistivity variation. Sol Energ Mat sol C. 2017; 173: 96-105. doi:10.1016/j.solmat.2017.05.042
10.1016/j.solmat.2017.05.042
CAS Web of Science® Google Scholar
9Bullock J, Wan YM, Hettick M, et al. Dopant-free partial rear contacts enabling 23% silicon solar cells. Adv Energy Mater. 2019; 9(9):1803367. doi:10.1002/aenm.201803367
10.1002/aenm.201803367
Web of Science® Google Scholar
10Dréon J, Jeangros Q, Cattin J, et al. 23.5%-efficient silicon heterojunction silicon solar cell using molybdenum oxide as hole-selective contact. Nano Energy. 2020; 70:104495. doi:10.1016/j.nanoen.2020.104495
10.1016/j.nanoen.2020.104495
CAS Web of Science® Google Scholar
11Adachi D, Hernandez JL, Yamamoto K. Impact of carrier recombination on fill factor for large area heterojunction crystalline silicon solar cell with 25.1% efficiency. Appl Phys Lett. 2015; 107(23):233506. doi:10.1063/1.4937224
10.1063/1.4937224
Web of Science® Google Scholar
12Tanaka M, Taguchi M, Matsuyama T, et al. Development of new a-Si/c-Si heterojunction solar cells: ACJ-HIT (artificially constructed junction-heterojunction with intrinsic thin-layer). Jpn J Appl Phys. 1992; 31: 3518-3522. doi:10.1143/JJAP.31.3518
10.1143/JJAP.31.3518
CAS Web of Science® Google Scholar
13Masuko K, Shigematsu M, Hashiguchi T, et al. Achievement of more than 25% conversion efficiency with crystalline silicon heterojunction solar cell. IEEE J Photovolt. 2014; 4(6): 1433-1435. doi:10.1109/JPHOTOV.2014.2352151
10.1109/JPHOTOV.2014.2352151
Web of Science® Google Scholar
14Feldmann F, Simon M, Bivour M, Reichel C, Hermle M, Glunz SW. Carrier-selective contacts for Si solar cells. Appl Phys Lett. 2014; 104(18):181105. doi:10.1063/1.4875904
10.1063/1.4875904
Web of Science® Google Scholar
15McIntosh KR, Black LE. On effective surface recombination parameters. J Appl Phys. 2014; 116(1):014503. doi:10.1063/1.4886595
10.1063/1.4886595
CAS Web of Science® Google Scholar
16Zhang Y, Wang L, Chen D, Kim M, Hallam B. Pathway towards 24% efficiency for fully screen-printed passivated emitter and rear contact solar cells. J Phys D Appl Phys. 2021; 54(21):214003. doi:10.1088/1361-6463/abe900
10.1088/1361?6463/abe900
CAS Web of Science® Google Scholar
17Augusto A, Karas J, Balaji P, Bowden S, King R. Exploring the practical efficiency limit of silicon solar cells using thin solar-grade substrates. J Mater Chem a. 2020; 8(32): 16599-16608. doi:10.1039/D0TA04575F
10.1039/D0TA04575F
CAS Web of Science® Google Scholar
18Veirman J, Varache R, Albaric M, et al. Silicon wafers for industrial n-type SHJ solar cells: Bulk quality requirements, large-scale availability and guidelines for future developments. Sol Energy Mater sol Cells. 2021; 228:111128. doi:10.1016/j.solmat.2021.111128
10.1016/j.solmat.2021.111128
CAS Web of Science® Google Scholar
19Jay F, Veirman J, Najid N, Muñoz D, Dubois S, Jouini A. Exclusively thermal donor-doped Cz wafers for silicon heterojunction solar cell technology. Energy Procedia. 2014; 55: 533-538. doi:10.1016/j.egypro.2014.08.020
10.1016/j.egypro.2014.08.020
CAS Google Scholar
20Glunz S, Feldmann F, Richter A, et al. The irresistible charm of a simple current flow pattern−25% with a solar cell featuring a full-area back contact. 31^st European photovoltaic solar energy conference and exhibition; 2015: 259-263.
Google Scholar
21Steinkemper H, Hermle M, Glunz SW. Comprehensive simulation study of industrially relevant silicon solar cell architectures for an optimal material parameter choice. Prog Photovolt. 2016; 24(10): 1319-1331. doi:10.1002/pip.2790
10.1002/pip.2790
CAS Web of Science® Google Scholar
22Augusto A, Srinivasa A, King RR, Bowden SG. Performance of silicon heterojunction solar cells using high resistivity substrates. 46^th IEEE Photovoltaic Specialists Conference; 2019:0300-0303.
Google Scholar
23Srinivasa A, Bowden S, Augusto A. Silicon heterojunction solar cells with 1k Ω.cm bulk resistivity wafers. 48^th IEEE Photovoltaic Specialists Conference; 2021: 1916-1918.
Google Scholar
24Dupré O, Vaillon R, Green MA. Thermal Behavior of Photovoltaic Devices. Springer; 2017.
10.1007/978-3-319-49457-9
Google Scholar
25Kern W. The evolution of silicon wafer cleaning technology. J Electrochem Soc. 1990; 137(6): 1887-1892. doi:10.1149/1.2086825
10.1149/1.2086825
CAS Web of Science® Google Scholar
26Schneider TW, Buttry DA. Electrochemical quartz crystal microbalance studies of adsorption and desorption of self-assembled monolayers of alkyl thiols on gold. J am Chem Soc. 1993; 115(26): 12391-12397. doi:10.1021/ja00079a021
10.1021/ja00079a021
CAS Web of Science® Google Scholar
27Balaji P, Dauksher WJ, Bowden SG, Augusto A. Improving surface passivation on very thin substrates for high efficiency silicon heterojunction solar cells. Sol Energy Mater sol Cells. 2020; 216:110715. doi:10.1016/j.solmat.2020.110715
10.1016/j.solmat.2020.110715
CAS Web of Science® Google Scholar
28Le AHT, Basnet R, Yan D, Chen W, Seif JP, Hameiri Z. Temperature dependence of polysilicon pasivating contact and its device performance. 47^th Photovoltaic Specialists Conference; 2020: 1020-1023.
Google Scholar
29Le AHT, Basnet R, Yan D, et al. Temperature-dependent performance of silicon solar cells with polysilicon passivating contacts. Sol Energ Mat sol C. 2021; 225:111020. doi:10.1016/j.solmat.2021.111020
10.1016/j.solmat.2021.111020
Web of Science® Google Scholar
30Green MA. General temperature dependence of solar cell performance and implications for device modelling. Prog Photovolt. 2003; 11(5): 333-340. doi:10.1002/pip.496
10.1002/pip.496
CAS Web of Science® Google Scholar
31Zhu Y, Hameiri Z. Review of injection dependent charge carrier lifetime spectroscopy. Prog Energy. 2021; 3(1):012001. doi:10.1088/2516-1083/abd488
10.1088/2516?1083/abd488
Google Scholar
32Kane D, Swanson R. Measurement of the emitter saturation current by a contactless photoconductivity decay method. 18^th IEEE Photovoltaic Specialists Conference; 1985: 578-583.
Google Scholar
33Richter A, Glunz SW, Werner F, Schmidt J, Cuevas A. Improved quantitative description of auger recombination in crystalline silicon. Phys Rev B. 2012; 86(16):165202. doi:10.1103/PhysRevB.86.165202
10.1103/PhysRevB.86.165202
CAS Web of Science® Google Scholar
34Schenk A. Finite-temperature full random-phase approximation model of band gap narrowing for silicon device simulation. J Appl Phys. 1998; 84(7): 3684-3695. doi:10.1063/1.368545
10.1063/1.368545
CAS Web of Science® Google Scholar
35Klaassen DBM. A unified mobility model for device simulation—I. model equations and concentration dependence. Solid State Electron. 1992; 35(7): 953-959. doi:10.1016/0038-1101(92)90325-7
10.1016/0038-1101(92)90325-7
CAS Web of Science® Google Scholar
36Kang D, Sio H, Xinyu Z, Wang Q, Jin H, Macdonald D. 36^th European Photovoltaic Solar Energy Conference and Exhibition; 2019: 318-321.
Google Scholar
37Schmidt J, Bredemeier D, Walter DC. On the defect physics behind light and elevated temperature-induced degradation (LeTID) of multicrystalline silicon solar cells. IEEE J Photovolt. 2019; 9(6): 1497-1503. doi:10.1109/JPHOTOV.2019.2937223
10.1109/JPHOTOV.2019.2937223
Web of Science® Google Scholar
38Niewelt T, Selinger M, Grant N, Kwapil W, Murphy J, Schubert M. Light-induced activation and deactivation of bulk defects in boron-doped float-zone silicon. J Appl Phys. 2017; 121(18):185702. doi:10.1063/1.4983024
10.1063/1.4983024
Web of Science® Google Scholar
39Pässler R. Semi-empirical descriptions of temperature dependences of band gaps in semiconductors. Physica Status Solidi. 2003; 236(3): 710-728. doi:10.1002/pssb.200301752
10.1002/pssb.200301752
CAS Web of Science® Google Scholar
40Haschke J, Seif JP, Riesen Y, et al. The impact of silicon solar cell architecture and cell interconnection on energy yield in hot & sunny climates. Energ Environ Sci. 2017; 10(5): 1196-1206. doi:10.1039/C7EE00286F
10.1039/C7EE00286F
CAS Web of Science® Google Scholar
41Sai H, Umishio H, Matsui T, et al. Impact of silicon wafer thickness on photovoltaic performance of crystalline silicon heterojunction solar cells. Jpn J Appl Phys. 2018; 57(8S3):08RB10. doi:10.7567/JJAP.57.08RB10
10.7567/JJAP.57.08RB10
Web of Science® Google Scholar
42Sai H, Sato Y, Oku T, Matsui T. Very thin crystalline silicon cells: A way to improve the photovoltaic performance at elevated temperatures. Prog Photovolt. 2021; 29(10): 1093-1104. doi:10.1002/pip.3443
10.1002/pip.3443
CAS Web of Science® Google Scholar
43Zhao J, Wang A, Robinson S, Green M. Reduced temperature coefficients for recent high-performance silicon solar cells. Prog Photovolt. 1994; 2(3): 221-225. doi:10.1002/pip.4670020305
10.1002/pip.4670020305
CAS Google Scholar
44Dupré O, Vaillon R, Green MA. Experimental assessment of temperature coefficient theories for silicon solar cells. IEEE J Photovolt. 2016; 6(1): 56-60. doi:10.1109/JPHOTOV.2015.2489864
10.1109/JPHOTOV.2015.2489864
Web of Science® Google Scholar
45Battaglia C, Cuevas A, De Wolf S. High-efficiency crystalline silicon solar cells: Status and perspectives. Energ Environ Sci. 2016; 9(5): 1552-1576. doi:10.1039/C5EE03380B
10.1039/C5EE03380B
CAS Web of Science® Google Scholar
46Ponce-Alcántara S, Connolly JP, Sánchez G, Miguez JM, Hoffmann V, Ordás R. A statistical analysis of the temperature coefficients of industrial silicon solar cells. Energy Procedia. 2014; 55: 578-588. doi:10.1016/j.egypro.2014.08.029
10.1016/j.egypro.2014.08.029
CAS Web of Science® Google Scholar
47Berthod C, Søndergaard ST, Odden JO. Experimental investigation of the optimal ingot resistivity for both the cell performance and the temperature coeffcients for different cell architectures. 7^th World Conference on Photovoltaic Energy Conversion; 2018: 0293-2097.
Google Scholar
48Zhang SM, Seif JP, Allen TG, et al. Temperature- and illumination-dependent characterization of solar cells using Suns-V_oc(T) and I-V(T). 48^th IEEE Photovoltaic Specialists Conference; 2021: 0737-0740.
Google Scholar
49Le AHT, Dréon J, Michel JI, et al. Temperature-dependent performance of silicon heterojunction solar cells with transition-metal-oxide-based selective contacts. Prog Photovolt. 2022; 30(8): 981-993.
10.1002/pip.3509
CAS Web of Science® Google Scholar
50Bernardini S, Bertoni MI. Insights into the degradation of amorphous silicon passivation layer for heterojunction solar cells. Physica Status Solidi. 2019; 216:1800705. doi:10.1002/pssa.201800705
10.1002/pssa.201800705
Web of Science® Google Scholar

Volume31, Issue5

May 2023

Pages 536-545

Temperature and illumination dependence of silicon heterojunction solar cells with a wide range of wafer resistivities

Abstract

1 INTRODUCTION

2 EXPERIMENTAL DETAILS

2.1 Sample preparation

2.2 Characterisation

3 RESULTS AND DISCUSSION

3.1 Temperature-dependent performance of SHJ solar cells at one-sun

3.2 Temperature dependence of surface passivation

3.3 Injection and temperature dependence of SHJ solar cells

4 CONCLUSION

ACKNOWLEDGEMENTS

APPENDIX A: EQE SPECTRA OF THE INVESTIGATED CELLS

APPENDIX B: BANDGAP NARROWING OF SI AT ELEVATED TEMPERATURES

APPENDIX C: τ_eff OF THE SYMMETRICAL LIFETIME TEST STRUCTURES AT Δn = 10¹⁵ cm⁻³ AND 25°C

Open Research

DATA AVAILABILITY STATEMENT

REFERENCES

Figures

References

Information

About Wiley Online Library

Help & Support

Opportunities

Connect with Wiley

Temperature and illumination dependence of silicon heterojunction solar cells with a wide range of wafer resistivities

Abstract

1 INTRODUCTION

2 EXPERIMENTAL DETAILS

2.1 Sample preparation

2.2 Characterisation

3 RESULTS AND DISCUSSION

3.1 Temperature-dependent performance of SHJ solar cells at one-sun

3.2 Temperature dependence of surface passivation

3.3 Injection and temperature dependence of SHJ solar cells

4 CONCLUSION

ACKNOWLEDGEMENTS

APPENDIX A: EQE SPECTRA OF THE INVESTIGATED CELLS

APPENDIX B: BANDGAP NARROWING OF SI AT ELEVATED TEMPERATURES

APPENDIX C: τeff OF THE SYMMETRICAL LIFETIME TEST STRUCTURES AT Δn = 1015 cm−3 AND 25°C

Open Research

DATA AVAILABILITY STATEMENT

REFERENCES

Figures

References

Related

Information

APPENDIX C: τ_eff OF THE SYMMETRICAL LIFETIME TEST STRUCTURES AT Δn = 10¹⁵ cm⁻³ AND 25°C