Proton Radiation Hardness of Perovskite Tandem Photovoltaics

Abstract

Monolithic [Cs_0.05(MA_{0. 17}FA_{0. 83})_0.95]Pb(I_0.83Br_0.17)₃/Cu(In,Ga)Se₂ (perovskite/CIGS) tandem solar cells promise high performance and can be processed on flexible substrates, enabling cost-efficient and ultra-lightweight space photovoltaics with power-to-weight and power-to-cost ratios surpassing those of state-of-the-art III-V semiconductor-based multijunctions. However, to become a viable space technology, the full tandem stack must withstand the harsh radiation environments in space. Here, we design tailored operando and ex situ measurements to show that perovskite/CIGS cells retain over 85% of their initial efficiency even after 68 MeV proton irradiation at a dose of 2 × 10¹² p⁺/cm². We use photoluminescence microscopy to show that the local quasi-Fermi-level splitting of the perovskite top cell is unaffected. We identify that the efficiency losses arise primarily from increased recombination in the CIGS bottom cell and the nickel-oxide-based recombination contact. These results are corroborated by measurements of monolithic perovskite/silicon-heterojunction cells, which severely degrade to 1% of their initial efficiency due to radiation-induced recombination centers in silicon.

Keywords: degradation; multijunction solar cell; perovskite; perovskite/CIGS; perovskite/silicon; perovsktite tandem; radiation hardness; radiation-induced defects; space photovoltaics; tandem solar cell.

Graphical abstract

Figure 1

Probing the Radiation Hardness of Perovskite/SHJ and Perovskite/CIGS Tandem Solar Cells In Operando during Proton Irradiation (A and B) 3D scatter plots of the straggling of 68 MeV protons within the perovskite/CIGS (A) and perovskite/SHJ (B) tandem solar cells. The corresponding energy loss of the incident 68 MeV protons to recoils is plotted as a function of depth based on SRIM simulations with a total of 5 × 10 ⁷ protons. The damage of a real space environment at the orbit of the international space station (ISS) is shown as black line considering polyenergetic and omnidirectional proton irradiation (see Supplemental Information for further details). (C and E) Operando measurements of V_OC, J_SC, FF, and η of the investigated perovskite/CIGS (C) and perovskite/SHJ (E) tandem solar cell as a function of the accumulated proton dose Φ. All values are normalized to their initial value. The proton energy amounted to 68 MeV. (D–G) Normalized short-circuit current of perovskite/CIGS (D) and perovskite/SHJ (F) tandem solar cell under illumination with NIR (λ = 850 nm) and blue LEDs (λ = 450 nm) that were alternatingly set to either 100% or 5/14% ( see Supplemental Information for further details) to mimic current matching under AM0 or forcing one sub-cell into limitation as illustrated in (G).

Figure 2

Proton-Irradiated Perovskite/SHJ and Perovskite/CIGS Tandem Solar Cells (A and C) Current-voltage characteristics of as-prepared (solid lines) and proton irradiated (dashed lines, Φ = 2 × 10 ¹² p⁺/cm², E_p = 68 MeV) perovskite/CIGS (A) and perovskite/SHJ (C) tandem solar cell under AM1.5G and AM0 illumination. The full circles indicate the mean maximum power point (MPP), and the inset depicts the power output at MPP as a function of time. (B and D) External quantum efficiency of the perovskite and the CIGS sub-cell (B) before (solid lines) and after proton irradiation (dashed lines). The EQE was measured using a chopper frequency of 74 Hz and appropriate LEDs to light bias the tandem. In the case of the CIGS bottom cell, EQE measurements were also performed employing higher chopper frequencies and stronger light biasing from a halogen lamp equipped with appropriate filters as indicated. In case of the perovskite/SHJ tandem (D), the irradiated perovskite top cell was also measured employing lower chopper frequencies as indicated. In both cases, the reflection of the tandem solar cells is shown by the blue solid (as-prepared) and dashed (irradiated) lines. The dotted lack line depicts the reflection of the used air-quartz-air encapsulation.

Figure 3

Identification of Radiation-Induced Recombination Pathways in Perovskite/CIGS Tandem Solar Cells after Proton Irradiation (A–C) Photoluminescence spectra (A) and decay (B) of the non-irradiated and irradiated CIGS bottom absorber. As sketched in (C), selective excitation in the CIGS layer was performed through the perovskite top absorber employing either a NIR cw laser at λ = 910 nm (in A) or a pulsed λ = 636 nm laser (in B) at a fluence of 160 mJ/cm² of which 13 mJ/cm² are absorbed within the CIGS in combination with appropriate long-pass filters to detect the emission. (D–F) Photoluminescence spectra (D) and decay (E) of the non-irradiated and irradiated perovskite top absorber. Excitation was performed using cw 405 nm (in D) or pulsed 636 nm (in E) illumination at 380 nJ/cm², as shown in (F).

Figure 4

Photoluminescence Lifetime and Quasi-Fermi-Level-Splitting Mapping of the Perovskite (A–C) (A) Photoluminescence lifetime histogram and (B and C) TRPL lifetime maps of the perovskite top absorber in the as-prepared and proton-irradiated perovskite/CIGS tandem solar cell under excitation with a 636 nm pulsed laser (5 MHz repetition rate, 380 nJ/cm²/pulse fluence). Lifetimes were extracted using single-exponential fitting. (D–F) (D) QFLS histogram and (E and F) QFLS maps of the perovskite top absorber in the as-prepared and proton-irradiated perovskite/CIGS tandem solar cell measured under 405 nm cw laser illumination with an intensity equivalent to 1 sun (see Supplemental Information for details).

Figure 5

Radiation-Induced V_OC Losses in Perovskite/CIGS and Perovskite/SHJ Tandem Solar Cells (A and C) Comparison of perovskite top cell QFLS statistics with the V _OC of perovskite/CIGS (A) and perovskite/SHJ (C) tandem and identically prepared CIGS and SHJ single-junction solar cells before and after irradiation, respectively. (B and D) V_OC as a function of light intensity for as-prepared and proton-irradiated perovskite/CIGS (B) and perovskite/SHJ (D) tandem solar cells, as well as CIGS and SHJ single-junction solar cells, respectively. Open and closed triangles depict the QFLS of the perovskite sub-cell as a function of excitation fluence for the perovskite/CIGS and perovskite/SHJ tandem solar cells. n^∗ denotes the internal ideality factor derived from Suns-QFLS statistics.