Printing High-Efficiency Perovskite Solar Cells in High-Humidity Ambient Environment-An In Situ Guided Investigation

Abstract

Extensive studies are conducted on perovskite solar cells (PSCs) with significant performance advances (mainly spin coating techniques), which have encouraged recent efforts on scalable coating techniques for the manufacture of PSCs. However, devices fabricated by blade coating techniques are inferior to state-of-the-art spin-coated devices because the power conversion efficiency (PCE) is highly dependent on the morphology and crystallization kinetics in the controlled environment and the delicate solvent system engineering. In this study, based on the widely studied perovskite solution system dimethylformamide-dimethyl sulfoxide, air-knife-assisted ambient fabrication of PSCs at a high relative humidity of 55 ± 5% is reported. In-depth time-resolved UV-vis spectrometry is carried out to investigate the impact of solvent removal and crystallization rate, which are critical factors influencing the crystallization kinetics and morphology because of adventitious moisture. UV-vis spectrometry enables accurate determination of the thickness of the wet precursor film. Anti-solvent-free, high-humidity ambient coatings of hysteresis-free PSCs with PCEs of 21.1% and 18.0% are demonstrated for 0.06 and 1 cm² devices, respectively. These PSCs exhibit comparable stability to those fabricated in a glovebox, thus demonstrating their high potential.

Keywords: air‐knife assisted drying; blade coating; crystallization; nucleation; perovskite solar cells; scalable ambient fabrication.

Figure 1

a) Schematic illustration of room‐temperature air‐knife‐assisted blading of a perovskite light‐harvesting layer in air with relative humidity of 55% (this study). b) Representative PCE of PSCs fabricated by a spin coating method (open symbol), and blade coating method (solid symbol) under different relative humidities.

Figure 2

a) Cross‐sectional SEM image of ambient‐air‐blade‐coated PSC with air‐knife blowing at 40 m s⁻¹, and b) PCE distribution as a function of air‐knife blowing velocity of the film prepared in ambient air and the control prepared in glovebox. c) PCE distribution histogram of 59 ambient‐air‐processed devices. d) J–V characteristics, e) EQE spectrum and the corresponding integrated J _SC, and f) stability test at the maximum power point of the champion device with an area of 0.06 cm².

Figure 3

a) In situ UV–vis transmission spectra of the as‐bladed precursor wet film in ambient air from two different samples. b) Plot of 1/λ versus maxima order extracted from the transmission spectrum exhibiting interference oscillations. c–f) Top‐view SEM images of perovskite thin‐films fabricated upon different blowing rates (0, 14, 28, and 40 m s⁻¹) and the associated images of higher magnification in the area marked by the red rectangle. g–j) Illustration of different nucleation and perovskite crystal growth behavior under varied blowing rates based on the classical LaMer model.

Figure 4

Time‐resolved absorption spectra measured by in situ UV–vis spectrometry for the blade‐coated wet perovskite film sample a) dried naturally and b) thermally dried at 105 °C without gas blowing in ambient conditions. Photographs of the samples are shown in Figure S2a,e, Supporting Information.

Figure 5

Time‐resolved absorbance spectra measured by in situ UV–vis spectrometry of air‐knife‐assisted drying at a) different gas‐blowing velocities, b) different extracted wavelengths, and c) the corresponding rate of change of absorbance.

Figure 6

a) Steady‐state and b) TRPL spectra for the samples prepared by air‐knife‐assisted drying at different gas‐blowing velocities. c) EL spectra of the device prepared by air‐knife‐assisted drying at 40 m s⁻¹ operating as a LED under different forward voltage bias. d) EQE _EL of devices prepared in ambient air by different gas blowing velocities. e) XRD patterns of perovskite thin‐films fabricated in ambient conditions by air‐knife‐assisted blading with different blowing rates, as labeled. Diffraction peaks associated with FTO are marked by an asterisk.

Figure 7

Performance of device with an area of 1 cm²: a) J–V characteristics, b) EQE, and c) stability test at maximum power point of d) the champion 1 cm² device. The integrated current density derived from the EQE spectrum is 21.65 mA cm⁻².