🔢 mathematics

Discontinuity of Continuum Fermion Dynamics

This paper demonstrates that the Heisenberg dynamics of continuum fermions in $\mathbb{R}^d$ interacting via nonconstant pair potentials is not pointwise-norm continuous, thereby proving that neither the CAR algebra nor its gauge-invariant subalgebra remains invariant under the time evolution for almost all times.

Original authors: Oliver Siebert

Published 2026-09-29

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Oliver Siebert

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

In the vast, silent theater of the quantum world, particles do not move like marbles on a table; they exist as a cloud of possibilities, described by mathematical rules that govern how they interact and evolve over time. For decades, physicists have relied on a specific framework to describe systems with many particles, such as the electrons in a metal or the atoms in a gas. This framework treats the collection of particles as a single, unified object that changes smoothly and continuously as time passes. The idea is that if you look at the system at one moment and then a tiny fraction of a second later, the change should be small and predictable, like the gentle turning of a dial. This smoothness is not just a mathematical convenience; it is the foundation that allows scientists to define what it means for a system to be in equilibrium, to have a temperature, or to settle into a stable state. Without this smoothness, the standard tools for understanding how matter behaves in the infinite expanse of space would break down.

A new study challenges the assumption that this smoothness always exists, specifically for systems of fermions, a type of particle that includes electrons. The researchers, working with the mathematics of particles moving through continuous space rather than on a grid, examined what happens when these particles interact through forces that depend on the distance between them. They focused on a scenario where the particles are not artificially restricted or smoothed out, a condition that reflects the raw, unfiltered reality of nature. The team constructed a rigorous mathematical proof showing that for a wide variety of realistic interaction forces, the system does not change smoothly. Instead, the evolution of the system becomes jagged and discontinuous, even when observing the most basic properties of the particles. This means that the standard mathematical description used for these systems fails to capture their true behavior over time, revealing a fundamental flaw in how we model the dynamics of interacting matter in the continuum.

The core of the discovery lies in how the researchers tested the stability of the system. They imagined a scenario where a large number of particles are packed tightly together in a small region, creating a dense cloud. Into this cloud, they introduced a single, additional particle at a specific distance. The researchers then watched how the interaction between this lone particle and the dense cloud evolved over time. In a smooth, well-behaved system, the influence of the cloud on the single particle would change gradually. However, the study proved that when the interaction force varies with distance, the effect on the single particle becomes erratic. As the number of particles in the cloud increases, the difference in how the system evolves at two very close moments in time does not shrink to zero. Instead, it remains significant, creating a sudden jump in the system's state. This jump occurs even though the time interval between the two moments is vanishingly small, effectively breaking the rule of smooth continuity.

This discontinuity has a profound consequence: it means that the mathematical "container" used to hold the description of these particles is not strong enough to survive the passage of time. In the language of the study, the algebra of observables—the set of all measurable properties of the system—is not invariant under the dynamics. In simpler terms, if you start with a valid description of the system and let time pass, the result is no longer a valid description within the same mathematical framework. The system evolves into a state that the original rules cannot describe. This happens for almost every moment in time, not just in rare or special cases. The researchers showed that this failure occurs even for the most basic building blocks of the theory, such as the operators that create or destroy a single particle. The proof holds for a broad class of interaction forces, including the familiar Coulomb force that governs the interaction between charged particles, and applies to both non-relativistic and relativistic descriptions of particle motion.

The study also clarifies why previous attempts to fix this problem by smoothing out the interactions or restricting them to finite volumes were necessary but incomplete. Those earlier methods worked because they removed the very features that cause the discontinuity. By proving that the discontinuity is an inherent feature of the unregularized, infinite system, the researchers demonstrate that the standard approach cannot be extended to the real world without modification. They showed that the only way to restore smoothness and invariance is to either change the interaction to be constant everywhere, which is physically unrealistic, or to expand the mathematical framework to include new types of observables. This finding forces a reevaluation of how we model the time evolution of quantum systems in infinite space, suggesting that the current standard model is fundamentally insufficient for describing the dynamics of interacting fermions without artificial constraints. The work does not suggest that the physics is wrong, but rather that the mathematical language used to describe it needs to be expanded to accommodate the jagged, discontinuous nature of reality at this scale.

=== SUMMARY ===

Technical Summary: Discontinuity of Continuum Fermion Dynamics

Problem Statement
This paper addresses a fundamental issue in the algebraic formulation of quantum statistical mechanics for interacting fermionic systems in infinite volume. In the standard $C^*$ -dynamical system framework, time evolution is required to be a pointwise-norm continuous group of $*$ -automorphisms on the algebra of observables. While free fermionic dynamics satisfy this condition on the Canonical Anticommutation Relations (CAR) algebra, the behavior of interacting continuum fermions remains a subject of rigorous inquiry.

The central problem is whether the Heisenberg dynamics generated by a many-body Hamiltonian with unregularized pair interactions preserves the CAR algebra and remains pointwise-norm continuous. Previous literature has suggested that the "ultraviolet problem"—where interaction energy grows indefinitely with particle density in small regions—might obstruct such continuity. While regularization techniques (e.g., momentum cutoffs or smeared operators) have been shown to yield well-behaved dynamics, the status of the unregularized theory on the standard CAR algebra had not been rigorously settled with a negative result. Specifically, it was unclear if the dynamics fails to be continuous or if the CAR algebra simply ceases to be invariant.

Methodology
The author employs a constructive approach to demonstrate the failure of continuity and invariance. The methodology relies on the following key components:

Construction of High-Density Clusters: The proof utilizes a packing argument to construct a sequence of $n$ -particle Slater determinants ( $\Psi_n$ ) confined to a bounded region. These states are designed to have kinetic and interaction energies scaling at most as $O(n^2)$ , ensuring they remain within the domain of the Hamiltonian's quadratic form.
Leakage and Stability Estimates: Using techniques related to Lieb-Robinson bounds and propagation estimates (specifically the ASTLO technique), the author establishes that these particle clusters remain localized for short time scales ( $t \sim n^{-1}$ ). A "leakage estimate" proves that the probability of particles escaping the cluster region is negligible ( $O(n^2 t^2)$ ) on this time scale.
Relative Phase Argument: The core mechanism involves placing two additional test particles (via creation operators $a^*(f_1)$ and $a^*(f_2)$ ) at distinct positions $z_1$ and $z_2$ relative to the cluster. If the pair potential $W$ is non-constant, $W(z_1) \neq W(z_2)$ . The interaction between the test particles and the dense cluster generates distinct dynamical phases, $e^{-2inW(z_1)t}$ and $e^{-2inW(z_2)t}$ .
Antisymmetrization Control: A critical technical step involves a localized antisymmetrization estimate (Lemma 3.3). This avoids the standard norm bound of $\sqrt{n}$ associated with adding a particle to an $n$ -particle state, showing instead that the norm growth is controlled by the local particle number in the disjoint region, allowing the relative phase to be detected in a fixed observable.
Automatic Continuity Argument: To bridge the gap between norm discontinuity and non-invariance, the paper employs a classical automatic-continuity argument (Lemma 3.5). This lemma states that if a strongly continuous unitary group acts on a separable Hilbert space, and the orbit of an element in a norm-closed, separable subspace is measurable, then norm continuity on a set of positive measure implies continuity everywhere.

Key Results
The paper establishes the following main theorems under general assumptions on the dispersion relation $\omega$ , the pair potential $W$ , and the external potential $U$ :

Theorem 2.4 (Discontinuity): For a large class of non-constant pair potentials (including singular potentials like Coulomb interactions in $d \ge 3$ ), the Heisenberg dynamics $\tau_t$ fails to be pointwise-norm continuous on the CAR algebra. Specifically, there exist normalized orthogonal functions $f_1, f_2$ such that for a specific sequence of times $t_n = \theta/n$ , the norm difference $\|\tau_{t_n}(A) - A\|$ does not vanish as $n \to \infty$ for the observable $A = a^*(f_1)a(f_2)$ . This discontinuity occurs even within the gauge-invariant subalgebra.
Corollary 2.5 (Non-invariance): As a consequence of the discontinuity and the automatic-continuity argument, neither the full CAR algebra nor its gauge-invariant subalgebra is invariant under the dynamics for almost all times $t \in \mathbb{R}$ . That is, $\tau_t(A) \notin \mathcal{A}_{CAR}$ for almost every $t$ .
Proposition 2.6 (Constant Interactions): In the special case of a constant interaction potential $W \equiv c \neq 0$ , the full CAR algebra is not invariant (discontinuity occurs for creation/annihilation operators), but the gauge-invariant CAR algebra is invariant, and the dynamics restricted to this subalgebra coincides with the free Bogoliubov dynamics, which is continuous.
Remark 2.7 (Dichotomy): The paper summarizes a dichotomy: for continuous potentials away from the origin, the gauge-invariant algebra is invariant if and only if the potential is constant; the full algebra is invariant if and only if the potential is zero.

Significance and Claims
The paper claims to provide the missing rigorous negative result regarding the behavior of unregularized continuum fermion dynamics on the standard CAR algebra. Its significance lies in:

Rigorous Justification for Extensions: The results justify the necessity of extending the observable algebra (as proposed in prior work like [Sie25]) to define a well-behaved $C^*$ -dynamical system for interacting continuum fermions. The standard CAR algebra is shown to be insufficient for unregularized interactions.
Limitations of Lieb-Robinson Bounds: The findings indicate that the standard application of Lieb-Robinson bounds to construct pointwise-norm continuous infinite-volume dynamics on the CAR algebra (as done in [GNRS20; HLS25]) cannot be extended to unregularized interactions. The obstruction is intrinsic to the ultraviolet behavior of the unregularized theory.
Clarification of the Ultraviolet Problem: The work confirms the heuristic expectation that the interaction strength seen by a particle in a high-density cloud grows indefinitely, leading to a breakdown of the standard algebraic continuity properties.

The author explicitly states that the paper does not propose new regularization schemes or experimental applications but rather serves as a "negative result" clarifying the mathematical limitations of the current framework for unregularized systems. The results are derived for a broad class of Hamiltonians, including non-relativistic and relativistic dispersion relations and singular potentials like the Coulomb interaction.

Technical Summary: Discontinuity of Continuum Fermion Dynamics

More like this