Spin polarization in chiral molecules, extensively explored under the framework of chirality-induced spin selectivity (CISS), has attracted significant attention in recent research [1]. A notably perplexing issue is the observed substantial variance in CISS-induced magnetoresistance (MR) ratios between spin-polarized conductive atomic force microscopy measurements and multilayer device assessments, despite using identical material systems. This discrepancy might be linked to pinholes within multilayer film devices, though the specifics remain unresolved [2]. Our objective is to delve into CISS-related phenomena absent of electric current injection into chiral molecules. By omitting the bias electric current, we sidestep the complications associated with pinholes in chiral molecule films. Our approach involves a bilayer system composed of two-dimensional (2D) chiral hybrid lead-iodide perovskite [2] and NiFe, where spin pumping facilitates the pure spin current injection into chiral molecules [3].

2D chiral hybrid lead-iodide perovskites, ($R$)-(+)-α-methylbenzylammonium lead iodide and ($S$)-(-)-α-methylbenzylammonium lead iodide (($R''-''$MBA)₂PbI₄ and ($S''-''$MBA)₂PbI₄) [2], were synthesized and employed in this work. The spin pumping experiments involved multilayer films, comprising a 5 nm thick Ni₈₁Fe₁₉ layer deposited on the approximately 60 nm thick 2D chiral perovskite films, as depicted in Fig. 1(a). Time-resolved magneto-optical Kerr effect (TRMOKE) measurements were conducted employing a femtosecond pulse laser at a wavelength of 515 nm. Figure 1(b) shows the magnetization curves for NiFe films deposited on 2D chiral perovskites, ($R''-''$MBA)₂PbI₄ and ($S''-''$MBA)₂PbI₄, as measured by the magneto-optical Kerr effect with perpendicular magnetic field (${\theta }_{\mathrm{B}}=0°$). In Fig. 1(c), the TRMOKE measurements results for the ($R''-''$MBA)₂PbI₄/NiFe are presented. These measurements, conducted under a constant magnetic field of ±2 T angled 25° from the film normal, reveal a pronounced initial drop in the Kerr rotation angle at zero delay time, followed by distinct oscillations as the Kerr rotation magnitude gradually recovers. Figure 1(d) shows similar TRMOKE measurement outcomes for the ($S''-''$MBA)₂PbI₄/NiFe, almost mirroring the results observed for ($R''-''$MBA)₂PbI₄/NiFe.

The impact of magnetic field direction on the resonant frequency and effective damping constants is shown in Fig. 2. As can be seen for ($R''-''$MBA)₂PbI₄/NiFe, the damping constant at +2 T is slightly ($\sim$0.001) higher than at −2 T. Conversely, for ($S''-''$MBA)₂PbI₄/NiFe, the damping constant at +2 T is marginally ($\sim$0.001) lower than at −2 T. Although the variations in the damping constant fall within the error range, the experiment demonstrates that the additional damping of NiFe, induced by the chirality of perovskite layer, is small. Considering the intrinsic damping constant of NiFe (5 nm) on a Si/SiO₂ substrate, as determined from separate measurements (not shown), to be 0.005±0.002, the increase in intrinsic damping of NiFe due to spin pumping is estimated at $\sim$0.04. Consequently, the damping difference of $\sim$0.001, when considered in the context of spin injection from NiFe to ($R''-''$MBA)₂PbI₄, is likely to represent at most a few percent of the difference of the spin injection. Note that this observation is not consistent with the spin filtering effect deduced from the substantial CISS-induced MR effect [2]. The minimal difference in damping constants, indicating a minor CISS, implies that the differences in CISS observed between the spin-polarized conductive atomic force microscopy and multilayer device methods may stem from fundamentally distinct physical mechanisms. The origin of the substantial CISS-induced MR using the spin-polarized conductive atomic force microscopy is unlikely to be due to spin filtering effect.

References

• [1] R. Naaman et al., Nat. Rev. Chem. 3, 250-260 (2019).
• [2] H. Lu et al., Sci. Adv. 5, eaay0571 (2019).
• [3] T. Hatajiri et al., Phys. Rev. B 110, 054435 (2024).

Authors

• T. Hatajiri, S. Sakamoto, H. Kosaki, Z. Tian, M. Tanaka, T. Ideue, K. Inoue, D. Miyajima^a,b, and S. Miwa
• ^aRIKEN
• ^bThe Chinese University of Hong Kong