Symmetry Breaking as Predicted by a Phase Space Hamiltonian with a Spin Coriolis Potential

Nadine C. Bradbury, Titouan Duston, Zhen Tao, Jonathan I. Rawlinson, Robert Littlejohn, Joseph Subotnik · 2025

Read Symmetry Breaking as Predicted by a Phase Space Hamiltonian with a Spin Coriolis Potential by Nadine C. Bradbury, Titouan Duston, Zhen Tao, Jonathan I. Rawlinson, Robert Littlejohn, Joseph Subotnik (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-03100.txt.

Introduction

Nadine Bradbury and five co-authors predict symmetry breaking in molecules with degenerate spin states when nuclear motion is treated in phase space. Standard theory separates electronic energy from nuclear kinetic energy. A phase space approach instead lets electronic states and energies depend on both nuclear position and momentum, beyond the Born Oppenheimer picture.

For three small molecules with degenerate spin ground states, the ground energy shows two minima at opposite nonzero momenta rather than one minimum at rest. Barriers between the minima are small in these molecules. The authors hypothesize that barriers become macroscopically large in metallic solids. That would give a new phase space surface for simulating effects like the Einstein-de Haas effect, where magnetization change drives rotation.

Catalog record

CREATORNadine C. Bradbury, Titouan Duston, Zhen Tao, Jonathan I. Rawlinson, Robert Littlejohn, Joseph Subotnik
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE65 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-03100

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Symmetry Breaking as Predicted by a Phase Space Hamiltonian with a Spin Coriolis Potential Nadine C. Bradbury, Titouan Duston, Zhen Tao, Jonathan I. Rawlinson, Robert Littlejohn, Joseph Subotnik Abstract We perform electronic structure calculations for a set of molecules with degenerate spin-dependent ground states ( 3 CH 2 , 2 CH ∙ 3 {}_{3}^{\bullet} , 3 O 2 ) going beyond the Born-Oppenheimer approximation and accounting for nuclear motion. According to a phase space (PS) approach that parametrizes electronic states ( | Φ ⟩ \ket{\Phi} ) and electronic energies ( E E ) by nuclear position and momentum (i.e., | Φ ⁡ ( 𝑹 , 𝑷 ) ⟩ \ket{\Phi(\bR,\bP)} and E ⁡ ( 𝑹 , 𝑷 ) E({\bm{R}},{\bm{P}}) ), we find that the presence of degenerate spin degrees of freedom leads to broken symmetry ground states. More precisely, rather than a single degenerate minimum at ( 𝑹 , 𝑷 ) = ( 𝑹 m ​ i ​ n , 0 ) ({\bm{R}},{\bm{P}})=({\bm{R}}_{min},0) , the ground state energy has two minima at ( 𝑹 , 𝑷 ) = ( 𝑹 m ​ i ​ n ′ , ± 𝑷 m ​ i ​ n ) ({\bm{R}},{\bm{P}})=({\bm{R}}_{min}^{\prime},\pm{\bm{P}}_{min}) (where 𝑹 m ​ i ​ n ′ {\bm{R}}_{min}^{\prime} is close to 𝑹 m ​ i ​ n {\bm{R}}_{min} ), dramatically contradicting the notion that the total energy of the system can be written in separable form as E = 𝑷 2 2 ​ M + V e ​ l E=\frac{{\bm{P}}^{2}}{2M}+V_{el} . Although we find that the broken symmetry solutions have small barriers between them for the small molecules, we hypothesize that the barriers should be macroscopically large for metallic solids, thus offering up a new phase-space potential energy surface for simulating the Einstein-de Haas effect. I Introduction Most modern schemes for efficiently propagating mixed quantum-classical nuclear-electronic dynamics do not conserve the total (nuclear + electronic + spin) angular momentum – at least not in a meaningful sense. For instance, within traditional classical Born-Oppenheimer (BO) dynamics, if one studies a system with an even number of electrons and a time-reversal invariant nondegenerate ground state, it is straightforward to show that one conserves the nuclear linear and angular momentum. More precisely, although BO theory formally conserves the total momentum – representing the total momentum by the nuclear momentum in the BO representation [ 1 ] – for a system with an even number of electrons and a time reversal invariant nondegenerate ground state, the electronic momentum is formally set to zero (and so momentum conservation is not really meaningful). In this sense, even if one incorporates some beyond BO physics by including the Berry force (whenever the latter is nonzero) [ 2 , 3 ] and indirectly accounts for how the electronic momentum interacts with the nuclear momentum (leading to a conservation law [ 4 , 5 , 6 ] ), one still cannot use BO theory to accurately capture momentum exchange and conservation in most cases. This state of affairs becomes even worse if we consider angular momentum, where we must account for angular momentum transfer between nuclear, electronic orbital, and electronic spin degrees of freedom. Moreover, in the presence of spin, if one wishes to go beyond single-state dynamics, the most standard nonadiabatic approach, Tully’s original surface hopping algorithm (TSH) [ 7 ] , becomes ill-defined for a doublet in the presence of spin-orbit coupling (SOC); in such a case, not only are the adiabatic states ill-defined because of Kramers’ degeneracy, but one finds complex-valued derivative couplings, and therefore complex nuclear momentum upon momentum rescaling–which requires a new understanding of momentum in surface hopping simulations [ 8 , 9 ] . The difficulties cited above present major problems for ab initio modeling of some of the most interesting and scientifically pressing spin dynamics related phenomena. After all, spin-phonon coupling determines the superconductivity transition temperature in most heavy-fermion materials, [ 10 , 11 ] spin-phonon coupling also defines the spin-lifetime (e.g. usefulness) of spin-qubits based on diamond nitrogen-vacancy centers, [ 12 , 13 ] spin-phonon and electron-phonon coupling in ferromagnetic nickel lead to highly non-thermal distributions and energy transfer during non-equilibrium dynamics, [ 14 , 15 ] and spin-phonon coupling is one putative mechanism behind the origin of the chirality induced spin selectivity (CISS) effect. [ 16 , 17 ] In order to understand the systems above, atomistic modeling is essential. That being said, quite often, the electronic structure of systems at a single geometry is often hard enough (without considering dynamics) [ 18 ] . Nonetheless, for realistic simulations (especially with disorder), there is no alternative to explicitly modeling the coupling of nuclear, electronic and spin degrees of freedom, a dire need that necessitates the development of a computationally feasible theory that goes beyond the BO framework. Let us now directly address the physics that is not captured by BO dynamics. As early as 1929, it was recognized that, in order to recover the correct vibrational spectrum of small molecules, one would need to account for how nuclear motion perturbs the electronic spin. [ 19 ] The resulting spin-rotation coupling [ 20 , 21 , 22 , 23 ] is distinct from spin-orbit coupling, and has been observed spectroscopically and applied to a variety of small molecules, including O 2 [ 24 ] and CH 3 , [ 25 ] – which are treated in this paper. The form of the perturbation is: \displaystyle H_{SR}=-a_{0}\frac{g\alpha^{2}}{2m_{e}}\sum_{A}\sum_{i}\frac{Z_{A}}{|\hat{\bm{r}}_{i}-\bm{R}_{A}|^{3}}((\hat{\bm{r}_{i}}-\bm{R}_{A})\times(\frac{1}{M_{A}}\bm{P}_{A}))\cdot\hat{\bm{s}}_{i} (1) Here, a_{0} is the Bohr radius and \alpha=e/hc\approx 1/137 is the fine structure constant, and g\approx 2.0023 is the Landé g -factor of the electron. This perturbation is almost universally added to the spin-orbit coupling terms \displaystyle H^{(1)}_{\mathrm{SO}} \displaystyle= \displaystyle a_{0}\frac{g\alpha^{2}}{4m_{e}^{2}}\sum_{A}\sum_{i}\frac{Z_{A}}{|\hat{\bm{r}}_{i}-\bm{R}_{A}|^{3}}((\hat{\bm{r}}_{i}-\bm{R}_{A})\times\hat{\bm{p}}_{i})\cdot\hat{\bm{s}}_{i} (2) \displaystyle H^{(2)}_{\mathrm{SO}} \displaystyle= \displaystyle-a_{0}\frac{g\alpha^{2}}{4m_{e}^{2}}\sum_{ij}\frac{1}{|\hat{\bm{r}}_{i}-\hat{\bm{r}}_{j}|^{3}}((\hat{\bm{r}}_{i}-\hat{\bm{r}}_{j})\times\hat{\bm{p}}_{i})\cdot\hat{s_{i}} (3) and the electronic spin-other-orbit coupling term: \displaystyle H^{(2)}_{\mathrm{SOO}} \displaystyle= \displaystyle a_{0}\frac{g\alpha^{2}}{2m_{e}^{2}}\sum_{ij}\frac{1}{|\hat{\bm{r}}_{i}-\hat{\bm{r}}_{j}|^{3}}((\hat{\bm{r}}_{i}-\hat{\bm{r}}_{j})\times\hat{\bm{p}}_{j})\cdot\hat{\bm{s}_{i}} (4) Both Eq. 1 and 4 are “spin-other-orbit coupling” terms that assess how the spin of the target particle precesses in the presence the magnetic field produced by another source particle in the source particles’s rest frame; in Eq. 1 , the “other” particle is a nucleus; in Eq. 4 , the other particle is an electron. Eqs. 2 and 3 arise from investigating the magnetic field that the target spin particle feels in its own rest frame, as caused by the relative motion of a nucleus or electron — noting that the factor of \frac{1}{2} in both of Eq. 2 and 3 come from a Thomas procession correction. Now, it must be noted that, if one seeks to separate the electronic Schrodinger equation from the nuclear Schrodinger equation in the spirit of Born-Oppenheimer theory, there are other problems besides the presence of nuclear momentum in Eq. 1 above. Most importantly, if one seeks the most accurate description of coupled nuclear-electronic dynamics, one must account for the fact that because nuclei are never truly frozen, the electronic Schrodinger equation is being solved in a moving (non-inertial) frame, new forces necessarily arise–in particular a Coriolis term. Now, importantly there are actually two so-called Coriolis forces that arise: (i) an orbital Coriolis force that arises from solving the motion of electrons in a moving frame and (ii) a spin Coriolis force that arise from solving the electronic spin in a moving nuclear

Read the complete text