In situ dynamic observations of perovskite crystallisation and microstructure evolution intermediated from [PbI6]4− cage nanoparticles

Hybrid lead halide perovskites have emerged as high-performance photovoltaic materials with their extraordinary optoelectronic properties. In particular, the remarkable device efficiency is strongly influenced by the perovskite crystallinity and the film morphology. Here, we investigate the perovskites crystallisation kinetics and growth mechanism in real time from liquid precursor continually to the final uniform film. We utilize some advanced in situ characterisation techniques including synchrotron-based grazing incident X-ray diffraction to observe crystal structure and chemical transition of perovskites. The nano-assemble model from perovskite intermediated [PbI6]4− cage nanoparticles to bulk polycrystals is proposed to understand perovskites formation at a molecular- or nano-level. A crystallisation-depletion mechanism is developed to elucidate the periodic crystallisation and the kinetically trapped morphology at a mesoscopic level. Based on these in situ dynamics studies, the whole process of the perovskites formation and transformation from the molecular to the microstructure over relevant temperature and time scales is successfully demonstrated.

devices, separately. The Guassian distribution fitting curves were also provided.

Supplementary Note 1 | Perovskites crystallization and structure transformation.
In Supplementary Fig. 6, Gaussian-function fitting was applied to determine the peak position, peak area and the full width at half maximum (FWHM) of the special peaks through Igor software. The crystal size was obtained by Debye-Scherer equation, D = κλ/βcos(θ), where D is the mean size of the ordered crystalline domains (which could be equal to or smaller than the grain size), κ is a dimensionless shape factor, λ is X-ray wavelength, β is the line broadening  (Fig. 3 in the maintext) in slot die printing from precursor solution to colloidal intermediated composition, to the dried polycrystalline perovskite film and to degrade to the lead iodide are applicable to the simple drop-casting or common spin-coating procedure. The fundamental chemistry will not change with different process. From the previous XRD results reported by Dr. Zhang Wei and his colleagues 1 , we can conclude that the wet film after spin coating has similar diffraction peak positions with the slot-die-printing film at room temperature, indicating the formation of intermediated MAI· PbI2· DMF composition, which is the similar process of stage 1 in our study. Supplementary Fig. 4 and 8 show that the final dried films of spin coating and slot die printing annealed at 80 °C have almost the same XRD and FTIR spectra, proving the same final product, namely, the same stage 2. When the perovskite film are annealing at high temperature and prolonged time, the CH3NH3PbI3 could decompose to PbI2, which is the stage 3. In summary, the formation mechanism and fundamental chemistry is universal to different procedure.
As Supplementary Fig. 15 shows, the morphology could be various with different procedures. The drop casting leads to large domain size of tens micrometer, which is similar to slot die printing films at low temperature (Fig. 4 in the maintext), while the domain size is much larger than the grain size of hundreds nanometer on the spin-coating film. This major differences in morphology are resulted from the different drying kinetics. A slow solution drying in drop casting leads to prolonged crystal growth and thus form large spherulite crystals. Increasing drying kinetics using slot die processing under elevated temperatures could lead to a balanced crystallization between material diffusion and solvent evaporation, then obtaining a rhythmic crystallization process. In spin coating, the dramatically faster solvent removal could kinetically trap the crystallization in an earlier stage and thus reduces crystal sizes.

Supplementary Note 3 | Perovskite solar cells based on slot die printing.
The optoelectronic properties of the perovskite films strongly depend on the crystallinity and morphology. We fabricated the planar heterojunction perovskite solar cells (PSCs) based on this research to investigate the film quality and characterize the device performance. PSCs were fabricated using a device structure of ITO/PEDOT:PSS/CH3NH3PbI3/PCBM/BCP/Ag 2 ( Supplementary Fig. 16). The PEODT:PSS and PC61BM were selected as the hole transport and electron transport layers, respectively. The CH3NH3PbI3 perovskite photoactive layer was processed by HAQ with a ~350 nm thickness. The inset of Supplementary Fig. 16a shows that large grains, hundreds of nanometers in size were formed in lateral direction, which are larger than those observed in conventional spin-coating fabrication ( Supplementary Fig. 17). The grain size in the vertical direction to the perovskite film is comparable to film thickness, thus efficient carrier transport can be achieved. Supplementary Fig. 16b Supplementary Fig. 16c. The incident photon to current conversion efficiency shown in Supplementary Fig. 16d was examined and the integrated current density was 20.6 mA cm -2 , which was in good agreement with the short-circuit current density. Large-area devices with active area 1.00 cm 2 were also fabricated basing on the slot die printing and HAQ method.
A champion PCE of 11.6% under standard illumination with Jsc, Voc, and FF reaching values of 21.4 mA cm -2 , 0.92 V and 0.59 were achieved (Supplementary Fig. 19). In addition, we also fabricated the device based on the slot die printing without HAQ method, however, the device 24 performances were very poor. That is because of the non-continuous and rugged film morphology. As Fig. 4 in the maintext and Supplementary Fig. 10 show, the height variation of these periodic structure at the printing perovskite film surface, i.e. the amplitude, could be as large as hundreds of nanometers to micrometer. On one hand, there must be uniform PC61BM layer as thick as micrometer to cover the whole perovskite film, on the other hand, it is not easy to achieve high device performance based on such rough perovskite active layer even though the perovskite layer could be fully covered by PC61BM.
In fact, from slot die printing to HAQ experiment, we are running experiment under the same controlling parameter and the same drying kinetics. We see by detailed morphology characterization and in situ crystal growth study that increase the drying speed of perovskite solution could suppress the rhythmic crystallization and form smooth films, which yield ideal morphology to fabricate solar cells. In morphology investigation and in-situ study, we have clearly elaborated their differences. We conclude that at ~80 o C printing temperature, film got dried within seconds, leaving us no much room to observe the morphology evolution details. HAQ method makes film dried even quicker, and we cannot conduct the in-situ experiment.
Thus, we only provide static HAQ thin film morphology results. These results imply that the perovskite films via HAQ method in this research have comparable optoelectronic properties as the films obtained using traditional spin-coating fabrication. Furthermore, the uniform perovskite films could also have wider applications in other large-area optoelectronic devices based on slot die printing process and HAQ method.