Proof of the absence of local conserved quantities in the Holstein model

arXiv:2605.19606 · cond-mat.stat-mech, math-ph, math.MP, quant-ph · Submitted 2026-05-19 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: Today's paper: "Proof of the absence of local conserved quantities in the Holstein model".

Mira: The one-dimensional Holstein model possesses no nontrivial local conserved quantities beyond the Hamiltonian and total fermion number operator,

Kai: First, who's behind it and why it matters.

Title and authors: Kai: Now that we know what they’ve proven about the Holstein model, I want to talk about the actual substance of what they did in this paper "Proof of the absence of local conserved quantities in the Holstein model."

Mira: Essentially, they took a k-local quantity and tried to prove it could be conserved by looking at how it commutes with the Hamiltonian <ref:2605.19606#pg2>. They used an operator basis expansion where they look at all possible sequences of input operators to see what constraints that conservation law places on the coefficients <ref:2605.19606#pg2>.

Lev: So, the core of their methodology is setting up a huge system of linear equations based on those constraints, and then showing that these equations only have solutions where all the input coefficients are zero, except for those related to the total particle number <ref:2605.19606#pg2>.

Kai: That’s a very rigorous way to show nonintegrability; it moves beyond just observing dynamics and provides a formal proof that these specific types of local conserved quantities don't exist when t, g, and omega are all nonzero <ref:2605.19606#pg1>.

Mira: They spend a lot of time classifying the input operators into type i, ii, and iii to systematically rule out any possibility of a conserved quantity existing for different support sizes k <ref:2605.19606#pg2>.

Lev: I worry about the practical side here; if this proof is based on an infinite basis expansion, it might be mathematically sound but practically intractable for any simulation you want to run on a real quantum processor <ref:2605.19606#pg2>.

Kai: The paper does acknowledge that they have to handle these constraints step-by-step, moving from examining the constraints on rB k+one in Step one to checking relations like rB kj in Step two and finally eliminating the last candidates in Step three <ref:2605.19606#pg2>.

Mira: Their main result is that for any k-local quantity with support size between three and half the system length, there are no nontrivial local conserved quantities other than those already identified <ref:2605.19606#pg3>.

Lev: That means when you're designing a model for error correction, you can’t just look for a local symmetry of order k to define your code space if k falls in that range <ref:2605.19606#pg3>.

Kai: The paper then shows that for the Holstein–Hubbard model, this absence of conserved quantities also holds, which broadens the scope of this analytical foundation <ref:2605.19606#pg4>.

Mira: So, to sum up, they’ve rigorously shown that in these systems away from specific integrable limits, the local structure is quite simple and lacks those intermediate conservation laws <ref:2605.19606#pg4>.

Lev: It’s a solid piece of work for theoretical physics because it gives us a hard boundary on what symmetries we can expect to find in these correlated systems <ref:2605.19606#pg3>.

The paper's summary: Kai: Moving into the improvements section of this "Proof of the absence of local conserved quantities in the Holstein model," they don't really suggest modifying their mathematical proof structure, but rather they clarify and broaden what their results actually cover.

Mira: They focus on making sure that if you consider trivial conserved quantities with k=one or k=two they are explicitly stated to be just the Hamiltonian and the total fermion number operator <ref:2605.19606#pg3>.

Lev: That clarification is important because it settles the ambiguity around what counts as a "nontrivial" conserved quantity in this context, which is crucial for anyone trying to map this onto physical constraints <ref:2605.19606#pg3>.

Kai: They also show that for k=two every two-local conserved quantity is just a constant multiple of the Hamiltonian, which reinforces that finding <ref:2605.19606#pg3>.

Mira: The real improvement is extending the framework to the Holstein–Hubbard model, demonstrating that this same restriction on local conservation holds even when you introduce an additional term like onsite repulsion <ref:2605.19606#pg4>.

Lev: That’s a big deal because it means you don't need to re-run the entire nonintegrability proof every time you tweak the interaction strength U in a simulation; the conclusion is robust across that class of models <ref:2605.19606#pg4>.

Kai: So, the improvement is less about a new technique and more about extending the established conclusion to a wider family of physically relevant models, which makes it much more applicable for condensed matter research <ref:2605.19606#pg4>.

Mira: It solidifies the idea that in these types of systems, you really don't need to search for those intermediate local symmetries; the structure is fundamentally simpler than we might assume <ref:2605.19606#pg4>.

Lev: For hardware development, this means when we design our control pulses, we can focus on driving the known dynamics of the Hamiltonian rather than worrying about some subtle local conservation law that might not even be there <ref:2605.19606#pg3>.

The paper's improvements: Kai: So, wrapping up this discussion on "Proof of the absence of local conserved quantities in the Holstein model," we’ve seen how they rigorously showed that for the one-dimensional Holstein model and its extensions, there are no nontrivial local conserved quantities beyond what is already obvious <ref:2605.19606#pg4>.

Mira: The main implication is that when you look at these systems with electron–phonon coupling, you can stop searching for those intermediate conservation laws to simplify your understanding of their dynamics and thermalization <ref:2605.19606#pg4>.

Lev: From a quantum error-correction viewpoint, this confirms that we can’t rely on finding arbitrary local symmetries to build robust codes in these systems because they simply aren't there for k between three and half the system length <ref:2605.19606#pg3>.

Kai: It really means that when we model charge transport or polaronic systems, we should be prepared for complex relaxation dynamics without expecting those hidden local symmetries to simplify the problem away <ref:2605.19606#pg4>.

Mira: And by extending this result to the Holstein–Hubbard model, they give us confidence that this absence of local conserved quantities is a general feature across a whole class of particle–boson coupled systems <ref:2605.19606#pg4>.

Lev: I just want to say that for real hardware implementation, this theoretical clarity is valuable because it sets clear limits on what kind of auxiliary constraints we might need to consider when designing our simulation protocols <ref:2605.19606#pg3>.

Kai: It’s a solid paper that provides a very firm analytical foundation for why these systems behave the way they do in terms of local structure <ref:2605.19606#pg4>.

Mira: And I think it pushes us to focus our theoretical energy on the actual physics happening within the Hamiltonian rather than chasing phantom symmetries <ref:2605.19606#pg4>.

Conclusion: Kai: So, we've gone through the details of the "Proof of the absence of local conserved quantities in the Holstein model," and essentially, they've proven that away from specific integrable limits, there aren't these intermediate local conservation laws <ref:2605.19606#pg4>.

Mira: Exactly; they rigorously showed that for any k-local quantity with support size between three and half the system length, those nontrivial local conserved quantities just don't exist <ref:2605.19606#pg3>.

Lev: I gotta say, from a hardware standpoint, that means when we're designing our qubit architectures or error correction schemes for these types of coupled systems, we can’t just assume there’s some intermediate symmetry to exploit <ref:2605.19606#pg3>.

Kai: It really highlights the need for us to focus our modeling on the fundamental Hamiltonian structure rather than searching for these kinds of local symmetries <ref:2605.19606#pg4>.

Mira: They also extended this finding to the Holstein–Hubbard model, which is significant because it shows this limitation isn't just a quirk of the simpler one-dimensional case <ref:2605.19606#pg4>.

Lev: That’s solid, because if we move to those more complex models with interactions, we still can't rely on finding these intermediate local quantities to simplify the dynamics <ref:2605.19606#pg4>.

Kai: It means when you're building a simulation or designing a control sequence for a polaronic system, you have to accept that the dynamics are governed by the full, complex Hamiltonian <ref:2605.19606#pg4>.

Mira: That’s the big picture; it forces us to be more careful about what we assume about hidden symmetries when analyzing these strongly correlated materials <ref:2605.19606#pg4>.

Lev: So, to wrap up on this paper, the main implication is that we need a more direct approach to understanding the long-time behavior of these models without relying on those specific intermediate conservation laws <ref:2605.19606#pg3>.

Kai: Absolutely; it sets a new boundary for what symmetries we can expect to find in these fermion–boson coupled systems <ref:2605.19606#pg4>.

Mira: And that's why the work on "Proof of the absence of local conserved quantities in the Holstein model" is so important for shaping how we approach theoretical characterization in condensed matter <ref:2605.19606#pg4>.

Lev: It’s a very clear statement about what we can and cannot expect from these models when looking for local symmetries to simplify things <ref:2605.19606#pg3>.

Department of Integrated Sciences, College of Arts and Sciences, The University of Tokyo · Department of Basic Science, Department of Multidisciplinary Sciences, Graduate School of Arts and Sciences, The University of Tokyo

cond-mat.stat-mech, math-ph, math.MP, quant-ph

Submitted: 2026-05-19

Updated: 2026-10-07

Comments: 35 pages, 2 figures, 1 table

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 83/100

The gist: The one-dimensional Holstein model possesses no nontrivial local conserved quantities beyond the Hamiltonian and total fermion number operator, establishing a rigorous analytical foundation for its

Key concepts

k-local conserved quantity: A physical quantity in the system that remains unchanged over time, defined by an operator involving at most k sites. The paper investigates these quantities to determine if any exist for a range of support sizes.
Nonintegrability: The mathematical property of a physical system where it lacks enough conserved quantities to be solved exactly using standard methods. The absence of nontrivial local conserved quantities proves the Holstein model is nonintegrable in certain parameter regimes.
Operator Basis Expansion: A technique used in the proof where a potential conserved quantity is written as a sum of terms involving various local operators. The conservation law requires that when this expression commutes with the Hamiltonian, specific coefficients in this expansion must be zero.
Input Operators (Ak_i): These are the specific types of local operators used to construct the potential conserved quantity Q_k. The proof systematically analyzes three categories of these input operators—Type i, Type ii, and Type iii—to show that they all lead to contradictions unless the conserved quantity is trivial.

Terminology

Hubbard Model — 24× in this paper · explained in 3 episodes
Conservation Laws — 9× in this paper · explained in 3 episodes
Core — 4× in this paper · explained in 2 episodes
Ab Initio — 1× in this paper · explained in 1 episode
Gibbs State — 1× in this paper · explained in 2 episodes
Phase Transitions — 1× in this paper · explained in 2 episodes
Physical Constraints — 1× in this paper · explained in 7 episodes

Summary

The one-dimensional Holstein model possesses no nontrivial local conserved quantities beyond the Hamiltonian and total fermion number operator, establishing a rigorous analytical foundation for its nonintegrability in regimes where all parameters are nonzero.

Model Definition and Nonintegrability Criterion

The study focuses on the one-dimensional (spinless) Holstein model, defined by the Hamiltonian:

Hˆ = Hˆ hop + Hˆ int + Hˆ pho, where Hˆ hop describes fermion hopping, Hˆ int represents electron–phonon coupling, and Hˆ pho describes phonon dynamics. The core of the result is that away from known integrable parameter regimes (where at least one of t, g, or ω is zero), the model exhibits nonintegrability in the sense that it lacks nontrivial local conserved quantities with support size 3 to L/2. This finding provides a rigorous analytical foundation for the nonintegrable behavior previously inferred only from dynamical and transport properties.

Classification of Local Conserved Quantities

The paper rigorously classifies all k-local conserved quantities for the Holstein model when t, g, and ω are nonzero:

Theorem 1 states that there is no k-local conserved quantity with 3 ≤ k ≤ L/2.
Theorem 2 provides a complete classification for k ≤ L/2, restricting nontrivial local conserved quantities to linear combinations of the Hamiltonian itself (Hˆ) and the total fermion number operator (Nˆ = X̂Li n̂i).

Proof Strategy: Operator Basis Expansion

The proof proceeds by expanding a k-local quantity Qˆk in an operator basis as Qˆk = X̂k l=1 X̂Li n Ali, where Ali are l-support input operators. The conservation law [Qˆk, Hˆ] = 0 implies that the output operators B l must have zero coefficients (rB l = 0). This leads to a large set of linear constraints on the coefficients qA in the input expansion. If these constraints admit no solution other than qAk = 0 for all k-support inputs Ak, then Qˆk cannot be a k-local conserved quantity.

Analysis of Input Types

The proof systematically examines three types of k-support inputs Ak i:

Type i inputs are severely restricted by the condition rB k+1 = 0, leading to Lemma 1, which shows that coefficients vanish unless both endpoints are ĉ†(0, 0) and ĉ(0, 0).
Type ii inputs are shown to have zero coefficients in Step 1 analysis (Lemma 2), owing to inversion symmetry.
Type iii inputs are shown to have zero coefficients by analyzing commutators with the chemical potential term and using constraints from Step 1 analysis (Lemma 5).

Elimination via Iterative Constraints

The proof proceeds through three steps:

(Step 1)

Focuses on rB k+1 = 0, narrowing down candidates for Ak i. Type i inputs are restricted to those where middle terms e2,..., ek−1 are identity I(0, 0).

(Step 2)

Focuses on rBk j = 0. For type i inputs, this yields relations such as q̂ĉ†i cˆi+k-1 = t g q̂ĉ†i+1 cˆ(0,1)i+k-1 (Eq. 38), and constraints on (k−1)-support inputs Ak−1 j.

(Step 3)

Focuses on rBk−1 = 0, using the constraints from Step 2 to eliminate the last remaining type i candidates. For k=3, this shows qAk=3 i = 0 (Lemma 6). For k ≥ 4, similar arguments are used to show that all remaining type i inputs have zero coefficients (Lemma 7).

Trivial Conserved Quantities

For trivial conserved quantities with k = 1 and k = 2, Theorem 3 establishes that the only independent local conserved quantities are the Hamiltonian Hˆ and the total fermion number Nˆ. This confirms Theorem 2, showing that up to the freedom of adding 1-local conserved quantities, every 2-local conserved quantity is a constant multiple of the Hamiltonian.

Holstein–Hubbard Model Extension

The result extends to the Holstein–Hubbard model (Hˆ = Hˆ'hop + Hˆ'int + Hˆpho + Hˆe−e) via Theorem 4, which states that k-local conserved quantities with k ≤ L/2 are restricted to linear combinations of Hˆ and N̂(σ,σ′), regardless of the presence of the Coulomb interaction term U. This demonstrates that nontrivial local conserved quantities are absent in representative fermion–boson coupled systems.

Improvements for AI systems

As a fastidious and diligent researcher, I have analyzed this seminal work on nonintegrability in fermion-boson coupled systems. The core finding is that local conserved quantities are absent in the 1D Holstein model (and Holstein-Hubbard model) except for the Hamiltonian and total particle number operator, provided the parameters are away from specific integrable limits.

Here are the specific improvements for AI systems derived from this scientific understanding:

The improved AI systems can perform tasks requiring rigorous analysis of complex, non-integrable many-body dynamics where standard thermalization assumptions might fail due to hidden conservation laws. Specifically:

Perform high-fidelity simulations and modeling of condensed matter systems (e.g., quantum magnets, superconducting materials, charge transport in polaronic systems) where the underlying physics is expected to be nonintegrable but exhibits complex relaxation dynamics.
Develop more accurate theoretical frameworks for characterizing long-time behavior in these systems that explicitly account for the absence (or presence) of local conserved quantities, rather than relying solely on generic thermalization expectations.

The specific improvements are as follows:

The improved AI system can specifically: