🔢 mathematics

Eigenvalue asymptotics for the one-particle density matrix and one-particle kinetic energy density operator

This paper establishes explicit asymptotic formulas for the eigenvalues of the one-particle density matrix and kinetic energy density operators associated with $N$ -particle atomic eigenfunctions, demonstrating that their decay rates are determined by particle coalescence singularities and are significantly faster when the eigenfunction is totally antisymmetric.

Original authors: Søren Fournais, Alexander V. Sobolev

Published 2026-08-11

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Søren Fournais, Alexander V. Sobolev

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a universe where everything is made of tiny, jittery dancers called electrons. These dancers don't just move randomly; they follow strict, invisible choreography dictated by the laws of quantum mechanics. When they gather around an atomic nucleus, they form a complex, multi-dimensional dance floor described by a mathematical object called a "wavefunction." This wavefunction tells us the probability of finding a dancer at any specific spot. But here's the catch: calculating the behavior of even a few of these dancers is so incredibly hard that computers can't solve the exact equations for real atoms. Instead, scientists use shortcuts, approximating the infinite possibilities with a finite number of "steps" or "modes," kind of like trying to describe a swirling ocean using only a few flat tiles.

To know how good these shortcuts are, scientists need to understand the "eigenvalues" of the system. Think of eigenvalues as the volume knobs on a stereo system for the atom. Each knob controls a specific frequency or pattern of the electron dance. The most important knobs are loud (large eigenvalues), and the less important ones are quiet (small eigenvalues). If the quiet knobs fade away very quickly, you can turn them off without losing much sound, meaning your shortcut is excellent. If they fade slowly, you need thousands of knobs to get the sound right, and your shortcut might be terrible. The big question in this field is: exactly how fast do these volume knobs turn down as you go to the quieter, higher-numbered ones?

This paper, written by Søren Fournais and Alexander V. Sobolev, dives deep into the math of these volume knobs for atoms. They focus on two specific "dials" that control how we see the atom: the one-particle density matrix (which tells us where the electrons are likely to be found) and the one-particle kinetic energy density (which tells us how much energy they have). The authors prove that the speed at which these dials fade away depends entirely on a very specific, messy part of the electron dance: the moment two electrons crash into each other.

The paper's main finding is a precise formula for how fast these eigenvalues drop. For a general atom, the authors prove that the $k$ -th volume knob (where $k$ is a huge number) gets smaller at a rate proportional to $k^{-8/3}$ . In plain English, if you double the number of the knob, the volume doesn't just drop a little; it drops by a factor related to that specific power, which is a very fast decline. However, the paper also discovers a special case: if the electrons are "totally antisymmetric" (a fancy way of saying they are fermions, like real electrons, and follow the rule that no two can be in the exact same state in a specific way), the dance becomes smoother at the crash points. In this special case, the volume knobs fade away even faster, at a rate of $k^{-10/3}$ .

The authors didn't just guess this; they proved it rigorously. They showed that the "roughness" of the electron wavefunction exactly where two particles collide is the only thing that matters for this calculation. Other messy spots, like where an electron hits the nucleus or where three electrons crash at once, don't actually change the speed of the fade-out. By isolating these specific collision points and using advanced mathematical tools called "Birman-Solomyak spectral asymptotics," they derived exact constants ( $A$ and $B$ ) that tell us exactly how loud or quiet these knobs will be based on the specific shape of the electron wavefunction.

In the world of quantum chemistry, this is a huge deal. It means that when scientists build computer models to simulate atoms and molecules, they now have a mathematical guarantee on how many "steps" they need to take to get an accurate result. If they know the electrons are behaving in that special antisymmetric way, they can be even more confident that their approximations will converge quickly, saving massive amounts of computing power. The paper essentially hands chemists a ruler to measure the error in their simulations, ensuring that the digital models of the atomic dance are as close to the real thing as possible.

Technical Summary: Eigenvalue Asymptotics for the One-Particle Density Matrix and One-Particle Kinetic Energy Density Operator

Problem Statement
This paper investigates the spectral asymptotics of the eigenvalues $\lambda_k(\Gamma)$ and $\lambda_k(K)$ associated with the one-particle density matrix operator $\Gamma$ and the one-particle kinetic energy density operator $K$ . These operators are derived from an eigenfunction $\psi$ of the $N$ -particle atomic Schrödinger operator $H = -\Delta + V$ , describing $N$ electrons interacting with a nucleus of charge $Z$ and with each other.

The eigenvalues of $\Gamma$ (often called occupation numbers) and $K$ are central to quantum chemistry computations, particularly for estimating errors arising from finite-dimensional approximations in bound state calculations. While previous works [42, 43] established the decay rates $\lambda_k(\Gamma) \sim k^{-8/3}$ and $\lambda_k(K) \sim k^{-2}$ for general eigenfunctions, the coefficients in these asymptotic formulas were not explicitly expressed in terms of the eigenfunction $\psi$ . Furthermore, the behavior of these eigenvalues in the case where $\psi$ is totally antisymmetric (representing spinless fermions) remained an open question, as the standard leading-order terms vanish in this scenario.

Methodology
The authors employ a multi-step analytical approach combining regularity theory for elliptic equations, factorization techniques, and spectral asymptotics for pseudodifferential operators.

Regularity and Singularity Analysis:
The core of the analysis relies on the specific singular behavior of the eigenfunction $\psi$ at "pair coalescence points" where two particles occupy the same position ( $x_j = x_k$ ). The authors utilize the representation $\psi = e^F \phi$ , where $e^F$ is a Jastrow factor capturing the main singularities (specifically the Coulomb cusp), and $\phi$ is a more regular function.
- General Case: Using recent regularity results [21, 28], the authors establish bounds for derivatives of $\phi$ near coalescence points. They show that the leading singularity of $\psi$ behaves like $|x_j - x_k|$ .
- Antisymmetric Case: For totally antisymmetric $\psi$ , the authors demonstrate enhanced regularity. By exploiting the antisymmetry condition $\psi(\dots, x_j, x_k, \dots) = -\psi(\dots, x_k, x_j, \dots)$ , they show that the leading singularity is of higher order (specifically, the function vanishes at the coalescence point, and the leading term behaves like $|x_j - x_k|^2$ or involves gradients).
Operator Factorization:
The integral operators $\Gamma$ and $K$ are factorized into products of the form $\text{Int}(\Psi)^* \text{Int}(\Psi)$ and $\text{Int}(V)^* \text{Int}(V)$ , where $\Psi$ and $V$ are vector-valued integral kernels derived from $\psi$ and its gradient. This reduces the problem of finding eigenvalue asymptotics to finding singular value asymptotics for the operators $\text{Int}(\Psi)$ and $\text{Int}(V)$ .
Spectral Asymptotics via Birman-Solomyak Theory:
The authors isolate the leading singularities of the kernels near the coalescence points. They decompose the kernels into a leading singular part (homogeneous in the relative coordinates) and a smoother remainder.
- The singular parts are treated as integral operators with homogeneous kernels (or pseudodifferential operators with asymptotically homogeneous symbols).
- The authors apply the spectral asymptotic results of Birman and Solomyak [3, 4, 5, 6] for such operators. This theory links the decay rate of singular values to the homogeneity order of the kernel symbol.
- Nikol'skii-Besov space estimates are used to bound the contribution of the smoother remainder terms, ensuring they do not affect the leading asymptotic coefficient.

Key Contributions and Results

Explicit Asymptotic Coefficients (General Case):
The paper provides explicit formulas for the constants $A$ and $B$ in the asymptotic relations:
$\lim_{k \to \infty} k^{8/3} \lambda_k(\Gamma) = A^{8/3}, \quad \lim_{k \to \infty} k^2 \lambda_k(K) = B^2$
Unlike previous results, $A$ and $B$ are given explicitly in terms of integrals involving the eigenfunction $\psi$ evaluated at coalescence points (specifically $|\psi(\dots, x, x)|^2$ ). The authors prove these formulas hold under natural polynomial decay assumptions on the one-particle density, removing the need for the stricter exponential decay assumptions used in earlier works.
New Results for Totally Antisymmetric Functions:
The paper establishes entirely new asymptotic formulas for the case where $\psi$ is totally antisymmetric. In this case, the standard coefficients $A$ and $B$ vanish, leading to a faster decay of eigenvalues. The authors derive:
$\lim_{k \to \infty} k^{10/3} \lambda_k(\Gamma) = (A_{\text{asym}})^{10/3}, \quad \lim_{k \to \infty} k^{8/3} \lambda_k(K) = (B_{\text{asym}})^{8/3}$
The new coefficients $A_{\text{asym}}$ and $B_{\text{asym}}$ are explicitly expressed in terms of the gradient of the eigenfunction, $\nabla \psi$ , at the coalescence points. This result relies on the enhanced smoothness of antisymmetric wavefunctions, which shifts the leading singularity order.
Refined Regularity Estimates:
The paper contributes detailed regularity estimates for the eigenfunction $\psi$ and its derivatives near coalescence points, particularly for the antisymmetric case. These estimates are derived using scale analysis and Jastrow factorization, providing the necessary bounds to apply the Birman-Solomyak theory.

Significance
The authors state that the primary significance of this work lies in the explicit identification of the asymptotic coefficients. By expressing $A$ , $B$ , $A_{\text{asym}}$ , and $B_{\text{asym}}$ directly in terms of the eigenfunction and its derivatives, the results provide a more transparent connection between the microscopic structure of the wavefunction (specifically its behavior at particle coalescence) and the macroscopic spectral properties of the density operators.

The paper notes that these asymptotics are determined solely by the singularities at pair coalescence points ( $x_j = x_k$ ), while higher-order coalescences or nucleus-electron coalescences do not contribute to the leading order. This clarifies the specific physical mechanisms driving the spectral decay.

Furthermore, the extension to the totally antisymmetric case is significant for quantum chemistry, as it addresses the specific behavior of fermionic systems where the standard leading-order terms vanish, revealing a different scaling law governed by the gradient of the wavefunction. The results are intended to serve as a theoretical basis for error estimation in finite-dimensional approximations used in atomic and molecular bound state computations.

More like this