Moisture‐Induced Non‐Equilibrium Phase Segregation in Triple Cation Mixed Halide Perovskite Monitored by In Situ Characterization Techniques and Solid‐State NMR

Abstract

In the last few years, perovskite solar cells (PSCs) became one of the most advanced technology in photovoltaics (PVs), reaching 25.5% certified power conversion efficiency (PCE) in single-junction cells. [1] This high-performance level, exceeding 80% of the thermodynamic bandgap limit of the perovskite, stems from the maximization of the solar spectrum absorption capability and reduction of the open-circuit voltage deficit through well-adjusted interfacial band level alignment and bulk/interface defects passivation to abate nonradiative recombination processes. [2] The reduction of interfacial energy losses due to energy misalignment is facilitated given the large richness of cationic and anionic substitution possibilities starting from the prototypical CH 3 NH 3 PbI 3 (MAPbI 3 ) composition, affording a very precise control of the optoelectronic characteristics. Through a wide exploration of composition, triple cation/double halide formulation Cs 0.05 (MA 0.17-FA 0.83 ) 0.95 Pb(Br 0.17 I 0.83 ) 3 (CsMAFA) rapidly emerged as one of the most efficient compositions for single-junction PSC [3,4] and also as a top cell in a monolithic perovskite/silicon tandem architecture reaching above 29% certified PCE with a slightly bromide richer composition. [5] However, the performance enhancement has progressed more rapidly than improving the stability, inhibiting the technology transfer to larger scale. Many efforts are now turned toward this objective through device encapsulation, [6,7] hydrophobic interfacial layers, [8] nanoscale 2D/3D structuration, [9] and defects passivation. [10][11][12][13] Topdown approaches that give further insights into the degradation pathways of the perovskite absorber and device stacks under operational conditions are highly desirable to propose rational ways to improve the stability at different scales from bulk and surface of the materials, interfaces, and finally on the entire device. Given the complexity in deciphering all possible contributions involved during the degradation, the first step ex situ investigations provide already relevant trends about material weaknesses. For instance, exposure of the conventional MAPbI 3 composition to a humid atmosphere showed rapid decomposition into PbI 2 and gas releases. [14] Depending on the relative humidity (RH)

Solid-State Nuclear Magnetic Resonance of Triple-Cation Mixed-Halide Perovskites

“…These δ­( 133 Cs) values were shifted by +13 ppm with respect to those reported in refs and to be consistent with the IUPAC convention here used for the 133 Cs reference.…”

“…The value of 37 ppm, here measured for TripleMix, allows us to rule out the presence of CsPbX 3 phases and to assess the full incorporation of Cs + in the mixed perovskite phase. It is interesting to observe that δ­( 133 Cs) is also sensitive to the presence and amount of MA, as evinced by comparing the value determined for TripleMix with those, quite different, found for Cs x FA 1– x PbI 3 and that of 44 ppm recently reported for a perovskite with chemical formula Cs 0.05 MA 0.16 FA 0.79 PbI 2.49 Br 0.51 . The line width of the 133 Cs signal in the MAS spectrum of TripleMix (fwhm ≈ 1.5 kHz) is similar to that observed for the other mixed perovskites, − while in the spectra of pure CsPbBr 3 and CsPbI 3 phases the 133 Cs signal is much narrower (fwhm ≈ 200–400 Hz) .…”

“…It is interesting to observe that δ( 133 Cs) is also sensitive to the presence and amount of MA, as evinced by comparing the value determined for TripleMix with those, quite different, found for Cs x FA 1−x PbI 3 and that of 44 ppm recently reported for a p e r o v s k i t e w i t h c h e m i c a l f o r m u l a Cs 0.05 MA 0.16 FA 0.79 PbI 2.49 Br 0.51 . 36 The line width of the 133 Cs signal in the MAS spectrum of TripleMix (fwhm ≈ 1.5 kHz) is similar to that observed for the other mixed perovskites, 34−36 while in the spectra of pure CsPbBr 3 and CsPbI 3 phases the 133 Cs signal is much narrower (fwhm ≈ 200−400 Hz). 40 This finding suggests that, in the mixed samples, a distribution of shifts can arise from slightly different 133 Cs environments.…”

“…While no differences in the dynamic behavior were found in MA x FA 1– x PbI 3 with respect to the parent pure perovskites, in the case of Cs x FA 1– x PbBr 3 the FA dynamics and phase transitions were affected by the presence of Cs + . Increasing the complexity of the systems, the incorporation of Cs + , Rb + , and K + in MA x FA 1– x PbI 3 was investigated by SSNMR in double-, triple-, and quadrupole-cation perovskites, while the only two reports on double/triple-cations and double-halide perovskites by SSNMR concern MA 0.15 FA 0.85 ­PbI 2.55 Br 0.45 and Cs 0.05 MA 0.16 ­FA 0.79 PbI 2.49 Br 0.51 …”

“…While no differences in the dynamic behavior were found in MA x FA 1−x PbI 3 with respect to the parent pure perovskites, in the case of Cs x FA 1−x PbBr 3 the FA dynamics and phase transitions were affected by the presence of Cs + . Increasing the 36 In this article, we applied for the first time multi-nuclear� 133 Cs, 13 C, 1 H, 207 Pb�SSNMR experiments for the characterization of the triple-cation lead mixed-halide perovskite with formula Cs 0.05 MA 0.14 FA 0.81 PbI 2.55 Br 0.45 (in the following indicated as TripleMix), which was found to be in one of the most efficient stoichiometries for single-junction solar cells. 7 In this SSNMR investigation, we exploited several nuclear probes, complementing spectroscopy experiments with the measurement of nuclear relaxation times, with the aim of highlighting structural and dynamic features related to TripleMix specific stoichiometry.…”