No Arabic abstract
Electron transfer (ET) in biological molecules such as peptides and proteins consists of electrons moving between well defined localized states (donors to acceptors) through a tunneling process. Here we present an analytical model for ET by tunneling in DNA, in the presence of Spin-Orbit (SO) interaction, to produce a strong spin asymmetry with the intrinsic atomic SO strength in meV range. We obtain a Hamiltonian consistent with charge transport through $pi$ orbitals on the DNA bases and derive the behavior of ET as a function of the injection state momentum, the spin-orbit coupling and barrier length and strength. A highly consistent scenario arises where two concomitant mechanisms for spin selection arises; spin interference and differential spin amplitude decay. High spin filtering can take place at the cost of reduced amplitude transmission assuming realistic values for the spin-orbit coupling. The spin filtering scenario is completed by addressing the spin dependent torque under the barrier, with a consistent conserved definition for the spin current.
The strength of the spin-orbit interaction relevant to transport in a low dimensional structure depends critically on the relative geometrical arrangement of current carrying orbitals. Recent tight-binding orbital models for spin transport in DNA-like molecules, have surmised that the band spin-orbit coupling arises from the particular angular relations between orbitals of neighboring bases on the helical chain. Such arrangement could be probed by inducing deformations in the molecule in a conductive probe AFM type setup, as it was recently reported by Kiran, Cohen and Naamancite{Kiran}. Here we report deformation dependent spin selectivity when a double strand DNA model is compressed or stretched. We find that the equilibrium geometry is not optimal with respect to the SO coupling strength and thus spin selectivity can be tuned by deformations. The latter can be increased by stretching the helical structure taking into account its elastic properties through the Poisson ratio. The spin filtering gap is also found to be tunable with uniaxial deformations.
Chirality induced spin selectivity, discovered about two decades ago in helical molecules, is a non-equilibrium effect that emerges from the interplay between geometrical helicity and spin-orbit interactions. Several model Hamiltonians building on this interplay have been proposed and while these can yield spin-polarized transport properties that agrees with experimental observations, they simultaneously depend on unrealistic values of the spin-orbit interaction parameters. It is likely, however, that a common deficit originates from the fact that all these models are uncorrelated, or, single-electron theories. Therefore, chirality induced spin selectivity is, here, addressed using a many-body approach, which allows for non-equilibrium conditions and a systematic treatment of the correlated state. The intrinsic molecular spin-polarization increases by two orders of magnitudes, or more, compared to the corresponding result in the uncorrelated model. In addition, the electronic structure responds to varying external magnetic conditions which, therefore, enables comparisons of the currents provided for different spin-polarizations in one of the (or both) leads between which the molecule is mounted. Using experimentally feasible parameters and room temperature, the obtained normalized difference between such currents may be as large as 5 - 10 % for short molecular chains, clearly suggesting the vital importance of including electron correlations when searching for explanations of the phenomenon.
The comment by O. Entin-Wohlman, A. Aharony, and Y. Utsumi, on our paper S. Varela, I. Zambrano, B. Berche, V. Mujica, and E. Medina, Phys. Rev. B 101, 241410(R) (2020) makes a few points related to the validity of our model, especially in the light of the interpretation of Bardarsons theorem: in the presence of time reversal symmetry and for half-integral spin the transmission eigenvalues of the two terminal scattering matrix come in (Kramers) degenerate pairs. The authors of the comment first propose an ansatz for the wave function in the spin active region and go on to show that the resulting transmission does not show spin dependence, reasoning that spin dependence would violate Bardarsons assertion. Here we clearly show that the ansatz presented assumes spin-momentum independence from the outset and thus just addresses the spinless particle problem. We then find the appropriate eigenfunction contemplating spin-momentum coupling and show that the resulting spectrum obeys Bardarsons theorem. Finally we show that the allowed wavevectors are the ones assumed in the original paper and thus the original conclusions follow. We recognize that the Hamiltonian in our paper written in local coordinates on a helix was deceptively simple and offer the expressions of how it should be written to more overtly convey the physics involved. The relation between spin polarization and torque becomes clear, as described in our paper. This response is a very important clarification in relation to the implications of Bardarsons theorem concerning the possibility of spin polarization in one dimensional systems in the linear regime.
We report a new type of spin-orbit coupling (SOC) called geometric SOC. Starting from the relativistic theory in curved space, we derive an effective nonrelativistic Hamiltonian in a generic curve embedded into flat three dimensions. The geometric SOC is $O(m^{-1})$, in which $m$ is the electron mass, and hence much larger than the conventional SOC of $O(m^{-2})$. The energy scale is estimated to be a hundred meV for a nanoscale helix. We calculate the current-induced spin polarization in a coupled-helix model as a representative of the chirality-induced spin selectivity. We find that it depends on the chirality of the helix and is of the order of $0.01 hbar$ per ${rm nm}$ when a charge current of $1~{rm mu A}$ is applied.
The chirality-induced spin selectivity (CISS), demonstrated in diverse chiral molecules by numerous experimental and theoretical groups, has been attracting extensive and ongoing interest in recent years. As the secondary structure of DNA, the charge transfer along DNA hairpins has been widely studied for more than two decades, finding that DNA hairpins exhibit spin-related effects as reported in recent experiments. Here, we propose a setup to demonstrate directly the CISS effect in DNA hairpins contacted by two nonmagnetic leads at both ends of the stem. Our results indicate that DNA hairpins present pronounced CISS effect and the spin polarization could be enhanced by using conducting molecules as the loop. In particular, DNA hairpins show several intriguing features, which are different from other chiral molecules. First, the local spin currents can flow circularly and assemble into a number of vortex clusters when the electron energy locates in the left/right electronic band of the stem. The chirality of vortex clusters in each band is the same and will be reversed by switching the electron energy from the left band to the right one, inducing the sign reversal of the spin polarization. Interestingly, the local spin currents can be greater than the corresponding spin component of the source-drain current. Second, both the conductance and the spin polarization can increase with molecular length as well as dephasing strength, contrary to the physical intuition that the transmission ability of molecular wires should be poorer when suffering from stronger scattering. Third, we unveil the optimal contact configuration of efficient electron transport and that of the CISS effect, which are distinct from each other and can be controlled by dephasing strength. The underlying physical mechanism is illustrated.