Quantum geometry and adiabaticity in molecules and in condensed matter
Raffaele Resta · 2025
Read Quantum geometry and adiabaticity in molecules and in condensed matter by Raffaele Resta (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-2503-10672.txt.
Introduction
Raffaele Resta revisits the adiabatic theorem and the hidden geometry inside it, in a 2025 physics paper. The theorem says a system changed slowly enough stays in its ground state. Resta stresses that slow change still adds a geometric term to many measured traits. That term, not the plain ground-state average, is often the part that matters.
He shows this for how materials polarize, how they absorb infrared light, and how molecules twist in magnetic fields, among other traits. Quantum geometry gives short, clear formulas for these effects. The classic approach sums over all excited states, a long route to the same answers. The paper covers both molecules and solids, offering a unified view for chemists and physicists alike.
Catalog record
| CREATOR | Raffaele Resta |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 42 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-10672
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Quantum geometry and adiabaticity in molecules and in condensed matter Raffaele Resta Abstract The adiabatic theorem states that when the time evolution of the Hamiltonian is “infinitely slow”, a system, when started in the ground state, remains in the instantaneous ground state at all times. This, however, does not mean that the adiabatic evolution of a generic observable obtains simply as its expectation value over the instantaneous eigenstate. As a general principle there is an additional adiabatic term, of quantum-geometrical nature, which is the relevant one for several static or adiabatic observables. This is shown explicitly for the cases of polarizability and infrared tensors (in molecules and condensed matter); rotational g g factor and magnetizability (in molecules only). Quantum geometry allows for a transparent derivation and a compact expression for these observables, alternative to the well known sum-over-states Kubo formulas. I Introduction The quantity which today is commonly addressed as a Berry curvature was introduced in the 1980s by Michael Berry [ 1 ] and, independently, by David Thouless and coworkers in a series of papers. The many-body formulation—upon which the present work is rooted—owes to a milestone paper by Niu and Thouless, [ 2 ] hereafter quoted as NT. These papers established the foundations of what nowadays we call “quantum geometry”. The term “adiabatic” intuitively refers to the “infinitely slow” time evolution of a physical system. In thermodynamics the system transforms through a series of continuously evolving equilibrium states; in quantum mechanics through a series of continuously evolving ground states. [ 3 ] I am addressing here the adiabatic response of some observables in condensed matter and in molecules, showing that quantum geometry—and in particular the Berry curvature—allows for a compact and elegant formulation of the corresponding tensors in both cases. All these tensors are a measure of how the many-electron system responds to a slow evolution of the Hamiltonian, and this is precisely what the Berry curvature allows casting in a very compact formalism. In condensed-matter physics the role of quantum geometry could be hardly underestimated. [ 4 , 5 , 6 ] In molecular physics, instead, quantum geometry is endorsed since long time to address conical intersections and related problems, [ 7 ] while its role in the formulation of the adiabatic-response tensors went unnoticed until recently. [ 8 ] The first part of this work is formulated in a generic quantum mechanical setting, i.e. not specifically addressing electrons. In Sec. II I provide some fundamentals of quantum geometry; in Sec. III I give a sharp definition of what “infinitely slow” means, in the sense of Kato’s theorem; [ 3 ] in Sec. IV it is shown that the adiabatic evolution of an observable can be expressed in a very compact form, referred to in the following as the “adiabatic Hellmann-Feynman theorem”, Eq. ( 14 ) below. The two terms therein can be regarded as the nongeometrical and geometrical one, respectively. A generic many-electron system, either condensed-matter or molecular, is addressed since Sec. V onwards. The explicit expressions for the chosen observables are presented in parallel, using—to the extent of possible—a common formulation and common notations for both molecules and condensed matter. Some notable differences, whenever occurring, are emphasized. The response properties addressed in the following are: polarizability and infrared tensors (in molecules and condensed matter); rotational g factor and magnetizability (in molecules only). All these tensor properties have a time-honored expression provided by linear-response theory (i.e. Kubo formulas). It is shown here that they admit an alternative geometrical derivation: they are indeed simple case studies of the adiabatic Hellmann-Feynman theorem at the core of the present work. The expressions in terms of the appropriate Berry curvature are all very compact and transparent. II Quantum geometry Let \hat{H} be a time-independent Hamiltonian and let |\Psi_{n}\rangle be its eigenstates with eigenvalues E_{n} ; when \hat{H} depends on a generic real parameter \lambda , the eigenstates and eigenvalues are parameter-dependent as well. The ground state is assumed to be nondegenerate for all \lambda ; its Berry connection is then defined as {\cal A}_{\lambda}=i\langle\Psi_{0}|\partial_{\lambda}\Psi_{0}\rangle, (1) and measures the infinitesimal phase variation of the eigenstate when \lambda is varied: d\phi=-\mbox{Im ln}\langle\Psi_{0\lambda}|\Psi_{0\lambda+d\lambda}\rangle={\cal A}_{\lambda}d\lambda. (2) Since the eigenstate |\Psi_{0\lambda}\rangle is arbitrary by a \lambda -dependent phase factor, the connection is gauge-dependent and cannot be endowed—as such—with any physical meaning; because of this its role has been overlooked for many years. In a pathbreaking paper [ 1 ] Michael Berry realized that, whenever the Hamiltonian is periodical in \lambda , the loop integral of the connection \gamma=\oint d\phi (3) is a gauge-invariant phase—called nowadays Berry phase—defined modulo 2\pi . The basic tenet of Berry’s paper is that any gauge-invariant quantum-mechanical expression corresponds in principle to an observable. Since then, several observables having the nature of a Berry phase have been identified in electronic structure theory [ 4 , 6 , 5 ] and in many other domains. Next let us suppose that \hat{H} also depends on a second real parameter \kappa ; the ground state is assumed to be nondegenerate for all (\kappa,\lambda) . The Berry curvature is by definition \Omega(\kappa,\lambda)=\partial_{\kappa}{\cal A}_{\lambda}-\partial_{\lambda}{\cal A}_{\kappa}=-2\,\mbox{Im }\langle\partial_{\kappa}\Psi_{0}|\partial_{\lambda}\Psi_{0}\rangle, (4) and is gauge-invariant. Let \Sigma be a domain in the \kappa\lambda plane where \Omega(\kappa,\lambda) is regular, and let \partial\Sigma be its boundary.; then Stokes’ theorem insures that \int_{\Sigma}d\kappa d\lambda\;\Omega(\kappa,\lambda)=\gamma, (5) where \gamma is the Berry phase \gamma=\oint_{\partial\Sigma}({\cal A}_{\lambda}d\lambda+{\cal A}_{\kappa}d\kappa). (6) Therefore the Berry curvature has the meaning of Berry phase per unit area. The parameters may have various physical interpretations and different dimensions; the curvature has the inverse dimensions of the product \kappa\lambda . For a macroscopic homogeneous system \Omega(\kappa,\lambda) is an extensive quantity. When (\kappa,\lambda) becomes time-dependent, the curvature is a quasi-static quantity, in the sense that its definition requires solely the instantaneous ground eigenstate; yet it encodes the lowest-order effect of the excited states on the adiabatic evolution. In the parallel-transport gauge [ 5 ] the \kappa -derivative of the eigenstate can be written as |\partial_{\kappa}\Psi_{0}\rangle=\sum_{n\neq 0}|\Psi_{n}\rangle\frac{\langle\Psi_{n}|\,\partial_{\kappa}\hat{H}\,|\Psi_{0}\rangle}{E_{0}-E_{n}}, (7) and analogously for the \lambda -derivative. Therefore the curvature admits a sum-over-states (i.e. Kubo) formula, first displayed in the original Berry’s paper: \Omega(\kappa,\lambda)=-2\,\mbox{Im}\sum_{n\neq 0}\frac{\langle\Psi_{0}|\partial_{\kappa}\hat{H}|\Psi_{n}\rangle\langle\Psi_{n}|\partial_{\lambda}\hat{H}|\Psi_{0}\rangle}{(E_{0}-E_{n})^{2}}. (8) The formula perspicuously shows how the Berry curvature encodes—to lowest order—the effect of the excited states on the ground state when the quantum system is transported in the parameter space. Eq. ( 8 ) has also the virtue of showing that the curvature becomes ill defined whenever the ground state is degenerate with the first excited state; for instance, at a conical intersection. III What does it mean “adiabatic”? Intuitively, the concept of adiabaticity is related to the time evolution of a system driven out of equilibrium “infinitely slowly”; in quantum mechanics, the Hamiltonian is assumed to have an “infinitely slow” time dependence. [ 3 ] In the current literature the terms “adiabatic” and “nonadiabatic” are sometimes improperly used; it is therefore expedient to start with a sharp definition. Suppose we have a time-independent Hamiltonian which depends on a parameter \lambda . When this parameter acquires a time dependence, \lambda\rightarrow\lambda(t) , the system and its observables evolve in time; saying that such evolution is “infinitely slow” is equivalent to say that \dot{\lambda}(t) is infinitesimal. If \hat{O} is a time-independent operator, the exact time-evolution of