Conical Intersections and Electronic Momentum As Viewed From Phase Space Electronic Structure Theory

Titouan Duston, Nadine Bradbury, Zhen Tao, Joseph E. Subotnik · 2025

Read Conical Intersections and Electronic Momentum As Viewed From Phase Space Electronic Structure Theory by Titouan Duston, Nadine Bradbury, Zhen Tao, Joseph E. 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-2506-11963.txt.

Introduction

Titouan Duston, Nadine Bradbury, Zhen Tao, and Joseph Subotnik rethink how electrons behave where two energy states meet, in a 2025 chemistry paper. Such meeting points, called conical intersections, steer many light-driven reactions. The standard theory treats the nuclei as fixed points and tracks only their positions.

The authors use a phase space theory that tracks nuclear momentum as well as position. For a small test molecule, the meeting region gains an extra dimension under this view. At a fixed shape, scanning the momentum reveals two low-energy wells away from zero. That means the steady electron states themselves carry motion, which the standard theory cannot show. The findings could sharpen models of fast chemical change.

Catalog record

CREATORTitouan Duston, Nadine Bradbury, Zhen Tao, Joseph E. Subotnik
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE44 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2506-11963

Browse the full library · About the Tower of Babel

Full text

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

Conical Intersections and Electronic Momentum As Viewed From Phase Space Electronic Structure Theory Titouan Duston, Nadine Bradbury, Zhen Tao, Joseph E. Subotnik Abstract We investigate the structure of a prototypical two-state conical intersection (BeH 2 ) using a phase space electronic Hamiltonian H ^ P ​ S ​ ( 𝑹 , 𝑷 ) \hat{H}_{PS}({\bm{R}},{\bm{P}}) that goes beyond the Born-Oppenheimer framework. By parameterizing the electronic Schrödinger equation by both nuclear position ( 𝑹 {\bm{R}} ) and momentum ( 𝑷 {\bm{P}} ), we solve for quantum electronic states in a moving frame that can break time reversal symmetry and, as a result, the branching plane of the conical intersection within a phase space framework now has dimension three (rather than dimension two as found within the standard Born-Oppenheimer framework). Moreover, we note that, if one fixes a geometry in real space that lies in the conical intersection seam and scans over the corresponding momentum space, one finds a double well (with minima at ± 𝑷 m ​ i ​ n ≠ 0 \pm{\bm{P}}_{min}\neq 0 ), indicating that the stationary electronic states of the phase space electronic Hamiltonian carry electronic momentum – a feature that cannot be captured by a Born-Oppenheimer electronic state. Interestingly, for B ​ e ​ H 2 BeH_{2} , this electronic momenta (as calculated with full configuration interaction) agrees with what is predicted by approximate complex restricted Hartree-Fock calculations, indicating a physical interpretation of complex instabilities in modern electronic structure calculations. Altogether, this study suggests that we have still have a lot to learn about conical intersections when it comes to electronic momentum, and highlights the urgent need for more experiments to probe what photochemical observables can and/or cannot be captured by standard electronic structure that isolates conical intersections within the Born-Oppenheimer framework. I Introduction: Electronic Momentum in Nonadiabatic Processes There are many fundamental processes in chemistry that require advanced nuclear-electronic dynamics methods that allow for motion on multiple electronic surfaces. In particular, tractable and accurate nonadiabatic techniques for dynamics on multiple surfaces are required to study photochemistry Joalland et al. 2014 ; Domcke and Yarkony 2012 , electron transfer Duncan, Craig, and Prezhdo 2007 , and inter-system crossings. Mai, Marquetand, and González 2015 To that end, great strides have been taken over the last few decades to develop efficient nonadiabatic algorithms and nowadays it is possible to navigate somewhat successfully on multiple surfaces. That being said, it must be noted that computational chemistry today still remains tied to the Born-Oppenheimer (BO) framework insofar as, with only a few exceptions Jasper and Truhlar 2011 , standard nonadiabatic approaches (e.g., fewest switches surface hopping Tully 1990 ; Sholl and Tully 1998 and multiple spawning Ben-Nun and Martınez 1998 ; Ben-Nun, Quenneville, and Martínez 2000 ) run along BO surfaces. Linear response Ehrenfest dynamics Li et al. 2005 is also tied to the BO representation. In short, while theoretical chemists have long sought to go beyond the BO approximation and have achieved many successes in that quest over the past several decades, the fact remains that the vast majority of theoretical chemistry remains driven by the BO framework . As far as understanding the strengths and limitations of the above algorithms to solve nonadiabatic problems, most analysis heretofore has focused on understanding the energy flow between nuclei and electrons. For instance, the surface hopping and multiple spawning algorithms were designed to capture how electronic transitions absorb or release energy to and from the nuclear degrees of freedom. The redistribution of energy is achieved within Ehrenfest dynamics by walking along average potential energy surfaces. In order to quantify the flow of energy between nuclei and electrons, chemists nowadays routinely calculate branching ratios Tully 1990 ; Zheng, Papajak, and Truhlar 2009 ; Cotton and Miller 2016 from scattering scenarios and kinetics (e.g. Marcus theory rates Marcus and Sutin 1985 ; Landry and Subotnik 2011 ; Chen and Reichman 2016 ; Jain and Subotnik 2015 ). There has also been a focus on the final equilibrium state Parandekar and Tully 2005 ; Schmidt, Parandekar, and Tully 2008 ; Tully 2012 as predicted by different nonadiabatic schemes. For example, it is well known that Ehrenfest does not obey detailed balance, whereas surface hopping does (but only approximately Barbatti 2011 ). Now, while the focus above is on energy conversion, one missing ingredient in the discussion above is momentum: the redistribution of momentum between nuclei and electrons is rarely discussed in the literature (except within the community of exact-factorization theorists Abedi, Maitra, and Gross 2010 ; Li, Requist, and Gross 2022 ). Ultimately, the source of the problem is clear: by making a separation between nuclei and electrons, and calculating electronic orbits in the frame of frozen nuclei, BO theory does not treat electronic and nuclear momentum equivalently. More precisely, while BO theory conserves momentum with the proper interpretation Littlejohn, Rawlinson, and Subotnik 2023 , in practice, momentum conservation is often achieved by setting the electronic momentum to zero. Therefore, one cannot reliably calculate the flow of momentum between nuclei and electrons (though the total should obviously be conserved); for meaningful momentum-conserving dynamics within a BO framework, one must include non-BO terms. Bian et al. 2023 ; Tao et al. 2024a At this point, in order to understand the implications of this error in detail, let us analyze two separate cases in BO theory: the cases of dynamic vs. static coupling. I.1 The Limit of Dynamic Correlation Consider first the case where a molecule evolves on one BO surface that is well separated from the other surfaces. For motion on a single surface (here labeled I ), it is obvious that there can be no electronic momentum within BO theory. After all, \left<\Phi_{I}\middle|\hat{{\bm{p}}}_{e}\middle|\Phi_{I}\right>=0 for any real electronic wavefunction \ket{\Phi_I} . Of course, the result above is incomplete. A more accurate result was found by Nafie, who used perturbation theory. Nafie 1983 The argument is as follows: If we regard the term -i\hbar\frac{{\bm{P}}\cdot\hat{\bm{d}}}{{\bm{M}}} as the perturbation to the standard H_{el} within the BO framework, then we can construct the perturbed electronic wavefunction \ket{\Psi_I} as \displaystyle\ket{\Psi_I}=\ket{\Phi_I}-i\hbar\sum_{J\neq I}\frac{\langle\Phi_{J}|\sum_{A}\frac{{\bm{P}}^{A}\cdot\hat{{\bm{d}}}^{A}}{M_{A}}|\Phi_{I}\rangle}{E_{I}-E_{J}}\ket{\Phi_J} (1) where \left<\Phi_{I}\middle|\hat{\bm{d}}^{A}\middle|\Phi_{J}\right>\equiv\bm{d}^{A}_{IJ} is the derivative coupling vector. Evaluating the expectation value of \langle\Psi_{I}|\hat{{\bm{p}}}_{e}|\Psi_{I}\rangle , we find: \displaystyle\langle\Psi_{I}|\hat{{\bm{p}}}_{e}|\Psi_{I}\rangle= (2) \displaystyle 2\hbar\text{Im}\sum_{J\neq I}\langle\Phi_{I}|\hat{{\bm{p}}}_{e}|\Phi_{J}\rangle\frac{\langle\Phi_{J}|\sum_{A}\frac{{\bm{P}}^{A}\cdot\hat{{\bm{d}}}^{A}}{M_{A}}|\Phi_{I}\rangle}{E_{I}-E_{J}} (3) If we assume a complete basis, we may use the commutator: \displaystyle\left[{\hat{H}}_{e},\hat{{\bm{r}}}_{e}\right]=-i\hbar\frac{\hat{{\bm{p}}}_{e}}{m_{e}}, (4) From this exact relationship, it follows that \displaystyle(E_{I}-E_{J})\langle\Phi_{I}|\hat{\bm{r}}_{e}|\Phi_{J}\rangle=-i\hbar\frac{1}{m_{e}}\langle\Phi_{I}|\hat{{\bm{p}}}_{e}|\Phi_{J}\rangle. (5) Then, if we plug Eq. 5 into Eq. 3 and use the fact that \bra{\Phi_{I}}\hat{{\bm{d}}}^{A}\ket{\Phi_{I}}=0 Lipkowitz 2007 (so that we can pull out a resolution of the identity), we find: \displaystyle\langle\Psi_{I}|\hat{{\bm{p}}}_{e}|\Psi_{I}\rangle (6) \displaystyle=2m_{e}\text{Re}\sum_{J}\langle\Phi_{I}|\hat{{\bm{r}}}_{e}|\Phi_{J}\rangle\langle\Phi_{J}|\sum_{A}\frac{{\bm{P}}^{A}\cdot\hat{{\bm{d}}}^{A}}{M_{A}}|\Phi_{I}\rangle (7) \displaystyle=2m_{e}\text{Re}\langle\Phi_{I}|\hat{{\bm{r}}}_{e}\sum_{A}\frac{{\bm{P}}^{A}\cdot\hat{{\bm{d}}}^{A}}{M_{A}}|\Phi_{I}\rangle (8) \displaystyle=2m_{e}\text{Re}\langle\Phi_{I}|\hat{{\bm{r}}}_{e}|\frac{d}{dt}\Phi_{I}\rangle (9) \displaystyle=m_{e}\frac{d}{dt}\langle\Phi_{I}|\hat{{\bm{r}}}_{e}|\Phi_{I}\rangle (10) If the math above is tedious, the reader might gain intuition from considering the very simplest case, a hydrogen atom, with Hamiltonian: \displaystyle\hat{H}=\frac{1}{2M_{H}}\hat{{\bm{P}}}^{2}_{H}+\frac{1}{2m_{e}}\hat{{\bm{p}}}^{2}_{e}-\frac{e^{2}}{|{\bm{R}}_{H}-{\bm{r}}_{E}|} (11) A naive view of BO theory would freeze the nucleus and solve for the electron in that frozen frame. Of course, such a perspective misses the fact that the motion of the nuclei likewise imparts momentum on the electron. One can understand these dynamical correlation effects as capturing the “electronic inertia”–which are present even for the hydrogen atom. Kutzelnigg 2007 In such a case, the electron inertial effects can be separated simply by going to the center-of-mass frame. \displaystyle\hat{H}_{com}=\frac{1}{2M_{H}+m_{e}}\hat{{\bm{P}}}^{2}_{com}+\frac{1}{2\mu}\hat{{\bm{p}}}^{2}_{\mu}-\frac{e^{2}}{|\hat{{\bm{r}}}|} (12) The eigenvalues

Read the complete text