How crystallization additives govern halide perovskite grain growth

Abstract

The preparation of perovskite solar cells from the liquid phase is a cornerstone of their immense potential. However, a clear relationship between the precursor ink and the formation of the resulting perovskite is missing. Established theories, such as heterogeneous nucleation and lead complex colloid formation, often prove unreliable, which has led to an overreliance on heuristics. Most high-performing perovskites use additives to control crystallization. Their role during crystallization is, however, elusive. Here, we provide evidence that typical crystallization additives do not predominantly impact the nucleation phase but rather facilitate coarsening grain growth by increasing ion mobility across grain boundaries. Drawing from the insights of our broad, interdisciplinary study that combines ex and in situ characterization methods, devices, simulations, and density function theory calculation, we propose a concept that proves valid for various additives and perovskite formulations. Moreover, we establish a direct link between additive engineering and perovskite post-processing, offering a unified framework for advancing material design and process engineering.

Fig. 1. Solvent chemistry—solvent type and concentration.

a Schematic illustration of the perovskite deposition process accompanied by the measurement approach. b UV-Vis absorbance measurements on MAPbI₃ precursor inks with different concentrations in DMF. Note that reflection was neglected and absorbance was calculated as 1 − T, where T is transmittance. UV-Vis for dimethylsulfoxid (DMSO) and n-methyl-2-pyrrilidone (NMP) are shown in Fig. S4. c Illustration of the lead complex evolution in dependency of concentration, as also proposed by ref. . d ²⁰⁷Pb NMR studies of MAPbI₃ precursor inks based on dimethyl formamide (DMF), DMSO, and NMP in dependency of precursor concentration. The dotted line represents the solid-state MAPbI₃ signal derived from ref. . Chemical shifts are reported on a ppm scale relative to the standard Pb(CH₃)₄. e Conductance study comparing stoichiometric MAPbI₃ precursor to MAI dissolved in DMF, NMP, or DMSO at different concentration. The bottom panel shows the relative difference of conductance (G) between the dissociated MAI and the colloidal MAPbI₃ ink, referenced by MAPbI₃, calculated as ${\mathrm{\Delta }}_{\mathrm{EC}}=\frac{G_{\mathrm{MAI}}-G_{{\mathrm{MAPbI}}_3}}{G_{{\mathrm{MAPbI}}_3}}$, following the apparent evolution in charge-to-volume ratio in the colloidal ink with increasing concentration. For details about the conductance measurement setup see the Methods section as well as Fig. S5.

Fig. 2. Impact of thiourea additive.

a Grain size distribution and scanning electron microscopy (SEM) images of MAPbI₃ layers desposited from an DMF:NMP precursor ink by gas-quenching with a concentration of 1.0 M with and without (w/o) 0.1 M of thiourea additive. Topography recorded by atomic force microscopy and perovskite layers deposited from other solvent systems can be found in Fig. S8. b XRD patterns of respective layers with and without 0.1 M thiourea. c Solar cell characteristics employing an ITO/PTAA/MAPbI₃/PCBM/AZO/Ag device stack using MAPbI₃ with and w/o thiourea additive as active layers. Here, measurements in the stabilized state (after light soaking) are shown. Statistics, other solar cell characteristics and solar cells based on a FA_0.94Cs_0.06PbI₃ active system with and w/o 0.1 M thiourea are shown in Fig. S9, Fig. S10, Fig. S11, and Fig. S12. d ²⁰⁷Pb NMR spectra showing the chemical shift of a 1.0 M MAPbI₃ ink with and w/o addition of 0.1 M thiourea in common solvents DMF, DMSO, and NMP, as well as our reference solvent system DMF:NMP. e Electrical conductance (EC) measurements of concentration series of the same perovskite and solvent systems as probed by ²⁰⁷Pb NMR.

Fig. 3. In situ investigations of the deposition process.

a Illustration of the measurement setup for in situ grazing incidence wide angle X-ray scattering (GIWAXS) characterization of the perovskite formation process. Exemplary plots for b radial and c azimuthal profiles that are plotted versus time in (e, f), and (h, i), respectively. Reciprocal space maps of MAPbI₃ perovskite thin films deposited from precursors without (d) and with 0.1 M (g) of thiourea additive. In d, also the plot direction for the time-dependent plots is highlighted. Radial Intensities (azimuthally integrated) of reciprocal space maps obtained by in situ GIWAXS as a function of time (e, without thiourea, h, with thiourea). The decline of the intermediate (cross) and formation of the perovskite phase (circles) in both cases is visible. The orange asterisk marks PbI₂. f, i Azimuthal angular intensity of the perovskite  <110> signal (around Q_r = 1 Å⁻¹) as a function of time, representing orientational order (f, without thiourea, i, with thiourea). A contrast is visible only after the annealing step has started. Vertical lines in all panels indicate the start of spinning (dotted), gas-quenching (GQ, dot-dashed) and annealing (heat, dashed). For all samples, a solvent mixture of 7:3 DMF:NMP was used.

Fig. 4. Perovskite formation process.

a Schematic illustration of the deposition process during the in situ experiment. b (top) Integrated GIWAXS signal intensities of perovskite and PbI₂ ⋅ NMP intermediate as a function of time without and with thiourea. The intensities that correspond to the respective phases are indicative of the relative material portions. (bottom) Degree of orientational order, represented by the ratio of the integrated GIWAXS signal intensity at 90° against the intensity at 45° azimuthal angle. Orientational order is used as an indicator for the formation of large grains with preferentially oriented crystallites as ultimately observed in layers employing the thiourea additive. For all samples, a solvent mixture of 7:3 DMF:NMP was used. Artefacts at around 200 s are related to the stop of the spin-coating process. c Sequence of processes as derived from precursor analysis (UV-Vis absorption, NMR & electrical conductance) and layer formation studies (GIWAXS & SEM), both in situ and ex situ. In the perovskite precursor, polynuclear colloidal structures form and increase in size during the spin-coating and drying process. Rapid solvent evaporation during gas-quenching leads to the formation of the PbI₂ ⋅ NMP intermediate, which transitions into the perovskite phase with continued nitrogen flow. During the annealing step, the last residues of PbI₂ ⋅ NMP intermediate are eliminated and the impact of the additive unfolds: In the absence of the additive, the randomly oriented perovskite phase with small grains is retained, while in the presence of the thiourea additive, highly oriented and large crystallites on the micrometer scale emerge. Coordination of the thiourea additive to the lead site likely occurs shortly before or during the annealing step.

Fig. 5. Universal coarsening mediated growth of perovskite grains.

a Evaluated GIWAXS measurements plotting the root-mean-square (rms) of azimuthal intensity fluctuations of characteristic diffraction signals as a function of time, obtained from time-resolved GIWAXS measurements, providing an indirect measure of the grain size. The plots show MAPbI₃ with and without 0.1 M thiourea and 0.1 M urea additive, as well as FA_0.94Cs_0.06PbI₃ with and without 0.1 M thiourea additive. The insets show SEM images of the respective perovskite layers. b Illustration and results of the post-processing approaches, where the crystallization agent was either spin-coated from isopropanol (IPA) solution on top of the pristine layer with subsequent heating (see also Fig. S23) or planar hot-pressing was applied at different temperatures. c Phase-field simulations illustrating a coarsening grain growth with different grain boundary mobilities as a function of time. For details, see Supplementary Note 3. d In situ ion mixing experiment observed by solid-state NMR to track the ion transport kinetics with and without the thiourea additive. (top) Mixing of MAPbI₃ and MAPbBr₃ without additive before and after 210 min annealing at 80 °C. (bottom) Normalized NMR signal intensities of the [PbI_6−xBr_x] environment ("mixed phase'') during the mixed-halide perovskite phase formation process from physical mixtures of MAPbI₃ and MAPbBr₃ containing 0 mol-%, 5 mol-%, and 10 mol-% thiourea fitted by Johnson-Mehl-Avrami-Kolmogorov model; the parameter k is reported for quantification. Fitting parameters and further details are given in Table S2 and Supplementary Note 4. The red arrows illustrate the NMR signals that are tracked for the bottom panel. A similar study for MACl additive is shown in Fig. S33. e (Top) Ultraviolet photoelectron spectroscopy (UPS) measurements of MAPbI₃ with and w/o thiourea, showing a shift of the Pb 5d_5/2 semi-core level to lower binging energy in the presence of thiourea, indicating electron transfer from the additive to the lead atoms at the sample surface. The full valence band scans and data for MACl treatment are included in Fig. S34. (Bottom) Temperature-dependent solid-state NMR measurements focusing on the ¹H signal of NH₂ groups (~7 ppm). The grey arrow indicates the position and development of the thiourea signal.

Fig. 6. Atomistic mechanism of how additives can increase ion mobility and grain coarsening.

Schematic illustration of the coarsening effect a without and b with the presence of additives at the grain boundary that enhance ion mobility. c–e Mechanisms of enhanced ion mobility in the presence of additives and elevated temperatures, as derived from FTIR, UPS, solid-state NMR measurements, and DFT calculations. Both processes, defect opening and ion shuttling, appear to increase ion mobility and further grain coarsening, which is why urea and thiourea, which can act as ion shuttles, generate higher ion mobilities than MACl, which is only expected to open defects. Furthermore, the lower adsorption energy of thiourea in comparison to urea makes it more likely to detach and gain mobility, as also indicated in Fig. 5e at elevated temperatures.