Magnetic-biased chiral molecules enabling highly oriented photovoltaic perovskites

ABSTRACT The interaction between sites A, B and X with passivation molecules is restricted when the conventional passivation strategy is applied in perovskite (ABX3) photovoltaics. Fortunately, the revolving A-site presents an opportunity to strengthen this interaction by utilizing an external field. Herein, we propose a novel approach to achieving an ordered magnetic dipole moment, which is regulated by a magnetic field via the coupling effect between the chiral passivation molecule and the A-site (formamidine ion) in perovskites. This strategy can increase the molecular interaction energy by approximately four times and ensure a well-ordered molecular arrangement. The quality of the deposited perovskite film is significantly optimized with inhibited nonradiative recombination. It manages to reduce the open-circuit voltage loss of photovoltaic devices to 360 mV and increase the power conversion efficiency to 25.22%. This finding provides a new insight into the exploration of A-sites in perovskites and offers a novel route to improving the device performance of perovskite photovoltaics.


PSCs
were fabricated using an n-i-p structure: FTO/SnO 2 /FAPbI 3 /BAI/Spiro-OMeTAD/Au.The FTO substrates underwent 15 minutes of cleaning with deionized water, acetone, and ethanol.Then, the substrate was annealed for 30 minutes at 150 °C after being spin-coated with a SnO 2 nanoparticle film (1:4 by water) for 30 seconds at 3,000 rpm. 1 mL of a DMF/DMSO mixed solvent (v/v 4/1) was used to dissolve 1142.39 mg of FAPbI 3 and 36 mg of MACl to create the perovskite solution.The D/L-p-FPhe powders (1 mg/mL) were respectively dissolved in the precursor solutions to prepare two kinds of doping solutions.In the glove box, the FAPbI 3 layer was spin-coated for 30 seconds at 5000 rpm (800 μL of diethyl ether was injected at 15 s).Then, the samples were annealed for 10 minutes at 150 °C.For the target samples, a designed frame with a NdFeB magnet placed on it to give a magnetic field of 50 mT, based on the previous studies focusing on the influence of the magnetic field to the perovskite film [1][2][3].The direction of the magnetic field was confirmed by DFT simulations as shown in Tab.
Finally, the samples were transferred into a thermal evaporation chamber (Suzhou Fangsheng FS380-S12) for the deposition of an Au electrode (100 nm) at a pressure < 2×10 -6 Torr.

Characterizations:
J-V characteristics of the PSCs (0.09 cm 2 ) were taken using a Keithley 2400 source meter under a simulated AM 1.5G spectrum (Enli Technology Co., Ltd).Thin mask was used and performed anti-reflection treatment on devices (blackening).Utilizing an integrated system (Enlitech, Taiwan), EQEs were measured.The EL output characteristics of PSCs were collected using a Photo Research PR655 spectrometer and a Keithley 2400 source meter.
UV-vis absorption spectra were acquired using a Hitachi U-4100 spectrophotometer.
GIWAXS measurements were carried out at the BL14B1 beamline of the China Shanghai Synchrotron Radiation Facility (SSRF).The X-ray used possessed a wavelength of 1.24 Å at a grazing incidence angle of 0.3° and energy of 10 KeV.
Two-dimensional (2D) GIWAXS patterns were collected from a MarCCD 225 detector.UV photoelectron spectroscopy (UPS) was used to measure the work functions of samples.Using PicoHarp 300, which has time-correlated single-photon counting capabilities, time-resolved PL decay profiles were acquired.Steady-state photoluminescence (PL) measurements were carried out using a Horiba Jobin Yvon system.A thermal-electrically cooled CCD (Princeton Instruments, PIX-256E) recorded the microarea photoluminescence (μ-PL) spectra, which were obtained using a handmade optical microscope equipped with a Princeton Instrument grating spectrometer (ARC-SP-2356) for light coupling.An inverted microscope (Olympus, BX43) was used to get the PL microscopy images.
We performed first-principles calculations based on density functional theory (DFT) by using a plane-wave basis set and the projected augmented wave method, as implemented in the VASP program [4,5].The Perdew−Burke−Ernzerhof (PBE) functional with the generalized-gradient approximation (GGA) is used for the exchange-correlation functional in all geometry optimization and self-consistent field calculations [6,7].A 4 × 4 × 1 gamma-centered k-mesh and a plane wave basis with a 400 eV cut-off energy are used for the geometry optimizations.We first relaxed the atomic positions and cell volumes using a conjugate gradient algorithm until all residual forces were smaller than 0.02 eV/Å.Continuation single-point energy calculations on the optimized geometries were performed to create the charge density difference graphs.We insert a vacuum slab with a thickness of 10−15 Å between the periodic slab-molecule structure along the z direction.The effect of the magnetic field is simulated by aligning all spin moments to a specific direction via the constrained spin method [8].Interaction energies between molecules and perovskite slabs were computed using the same level of theory and using the following equation: where , , and are the energies of the molecule/perovskite complex, only molecule, and only perovskite surface, respectively.S5.Binding energy of the chiral molecules with the perovskites in different directions of the magnetic field.A same direction of the magnetic field and the magnetic moment should be confirmed for D-p-FPhe while L-p-FPhe presented similar in two directions.
Free Magnetic Field (eV)

Figure S2 .
Figure S2.Molecular formula and DFT results for the net magnetization of L-p-Fluorophenylalanine molecule.The right diagram corresponds to the DFT results, and the complete chiral molecule lacks any magnetic dipole moments, so there is no mark.

Figure S3 .
Figure S3.DFT results of the net magnetization when the p-Fluorophenylalanines are introduced to the FAPbI 3 cage, and the net magnetization mainly originated by N ion of the alanine.The result revealed the presence of magnetic dipole moments between chiral molecules and FA ions, with values of 0.156 μ B and -0.112 μ B (μ B representing the Bohr magneton) along the z-axis for D/L-p-fluorophenylalanine (D/L-p-FPhe) molecules.

Figure S4 .
Figure S4.The charge density profiles for the D/L-p-FPhe molecules on the perovskite surface with or without magnetic field based on CCD calculations.Specifically, all spin magnetic moments were restricted in the molecule/perovskite system to align along the z-axis direction.

Figure S5 .
Figure S5.DFT results of the angles when the p-Fluorophenylalanines are introduced to the FAPbI 3 cage.

Figure S6 .
Figure S6.XPS spectra of perovskite film tread with L-p-FPhe under magnetic field and control FAPbI 3 perovskite film.

Figure S7 .
Figure S7.2D plots of the azimuthally integrated scattering intensity varies with annealing timealong the ring at q = 10 nm −1 for the L-p-FPhe doping samples with/without magnetic field.To compare the outcomes of these two effects, L-p-FPhe based samples were fabricated, which were demonstrated to solely possess passivation effects without exhibiting magnetic dipole moments.

Figure S8 .
Figure S8.Corresponding variations in the (001) peak intensity with time during FAPbI 3 film growth under annealing conditions.

Figure S9 .
Figure S9.In situ GIWAXS monitoring the crystal growth process of D/L-p-FPhe and control perovskite films covering q of 0-24 nm -1 .Top view image of FAPbI 3 (001) peaks in GIWAXS images.Expanding the analysis beyond the core (001) crystal face, the GIWAXS results encompassing multiple crystal faces within a broader range consistently yielded similar outcomes.

Figure S10 .
Figure S10.2D-GIWAXS patterns of the D-p-FPhe doped perovskite films with/without magnetic field during spin-coating time, before and after anti-solvent dropping, annealing for 30, 60 and 120 s.

Figure S11 .
Figure S11.2D-GIWAXS patterns of the perovskite films with/without magnetic field during the spin-coating time, before and after anti-solvent dropping, annealing for 30, 60 and 120 s.

Figure S12 .
Figure S12.2D-GIWAXS patterns of the L-p-FPhe doped perovskite films with/without magnetic field during spin-coating time, before and after anti-solvent dropping, annealing for 30, 60 and

Figure S14 .
Figure S14.Polar images of the control samples with annealing time of 0, 60, 180, 300, 600 s under the polarized incident light.

Figure S15 .
Figure S15.Normalized emission intensities as a function of detection polarization angle (ϕ) for the final target sample and control sample.Solid curves are fits to cos2 ϕ.

Figure S16 .
Figure S16.Steady state PL spectra of target and control films.

Figure S17 .
Figure S17.Absorption spectra of target and control films.

Figure S19 .
Figure S19.UPS spectra of target and control films and the schematic illustration of the corresponding energy band alignment of perovskite solar cells.

Figure S20 .
Figure S20.Schematic illustration of the corresponding energy band alignment of perovskite solar cells.

Figure S21 .
Figure S21.EL curves of the D-p-FPhe (with MF) and control devices.

Figure S22 .
Figure S22.EQE diagram and corresponding integral current of the control device.

Figure S24 .
Figure S24.DLCP curves of the D-p-FPhe (with MF) and control devices.

Figure S26 .
Figure S26.Evolution of PCE relative to the initial PCE for the D-p-FPhe-doped devices with magnetic field application, L-p-FPhe-doped devices without magnetic field application and FAPbI 3 devices over 4200 hours of storage in dry air.Each average (symbol) and standard deviation (error bar) was calculated from six solar cells.

Table S1 .
Carrier lifetime of the D-p-FPhe (with MF) and control films extracted from TRPL

Table S2 .
Fitting parameters of the formula for photocurrent measurement of the D-p-FPhe (with MF) and control devices.

Table S3 .
Performance factors of the D-p-FPhe (with MF) and control devices.

Table S4 .
V oc loss analysis of D-p-FPhe (with MF) and control devices.