State-resolved quantum-information fingerprints of partial dynamical symmetry in the interacting boson model

AI summary

Overview

Research area: Nuclear structure theory (the interacting boson model, IBM) bridged with quantum-information (QI) diagnostics — specifically, whether partial dynamical symmetry (PDS) leaves a state-resolved, measurable trace in eigenstates rather than only in spectra.

Technical level: Advanced. The paper assumes familiarity with algebraic nuclear models (U(6), SU(3), O(6), O(5) chains and their Casimir operators), exact diagonalization in boson Fock space, and standard QI notions (von Neumann entropy, dephasing maps, relative entropy of coherence). The prose is clear and the logic is well signposted, but the formalism is dense.

Scope (one sentence): The paper proposes and tests a single state-by-state diagnostic — the variance of a symmetry Casimir — that separates the symmetry-pure (PDS) eigenstates of $sd$-IBM Hamiltonians from the mixed ones across boson numbers $N=4$ to $16$, at both first- and second-order critical points, under controlled symmetry breaking, and in a proof-of-principle readout on a superconducting processor at $N=3$.

Note on completeness: the supplied text ends mid-sentence in Sec. VI.2. Sections VII–XI (the "purity/coherence" discussion, the contrast with earlier entanglement studies, and the qubit-encoding details) are referenced but their content is not included; only the abstract's claims about them are available.

What This Paper Is About

In the interacting boson model, a partial dynamical symmetry is a halfway house: some eigenstates keep exact quantum numbers and closed-form energies while the rest mix. Until now, PDS has been identified from spectroscopy — fitting level schemes and transition rates. This paper asks a different question: does PDS also leave a fingerprint in the structure of the wavefunctions themselves, one that could be computed or measured state by state?

The answer, for the $sd$-IBM Hamiltonians studied, is yes — but the fingerprint is not the amount of entanglement. It is the vanishing of a symmetry-label fluctuation on a structurally selected subset of states.

Key Contributions

1. A state-resolved classifier for PDS built on a Casimir variance. The central object is $\mathrm{Var}_\psi(\hat{C}_2[G])$ for a chosen subalgebra $G$, shown to share a common zero with a block-coherence entropy $S_G$ and a block impurity $1-\mathcal{P}_G$. All three vanish exactly when a state occupies a single Casimir sector — a single irreducible representation, once conjugate irreps are resolved by the cubic Casimir.

2. Separation of the diagnostic from both entanglement magnitude and closed-form solvability. The paper shows explicitly that static bipartite $s|d$ entanglement does not separate symmetry-pure from mixed eigenstates, and that exact irrep purity is strictly weaker than solvability — the two tiers are distinguished by checking the closed-form energy separately.

3. Robustness and scaling analysis. The separation is tracked from $N=4$ to $N=16$, at Leviatan's first- and second-order critical points, and under a controlled $SU(3)$-breaking perturbation, with the pure-state variance growing as $\epsilon^2$ as perturbation theory predicts.

4. A measurable dynamical form and a hardware demonstration. The diagnostic is recast as a "Casimir Loschmidt echo" that is rigid on symmetry-pure states, and its readout is demonstrated on an IBM superconducting processor at $N=3$, where a single solvable/mixed exemplar reproduces the predicted pattern within finite-shot uncertainties.

Main Findings

• Entanglement magnitude fails as a PDS diagnostic. At a stable $SU(3)$-PDS point ($N=10$), the static $s|d$ entropy distributions of symmetry-pure and mixed eigenstates overlap. Within a fixed angular momentum the pure states are in fact more entangled — described as a Simpson's-paradox reversal.

• Cone rigidity does separate the families. The off-diagonal response $R_D = |Q,D|\psi\rangle|$ measures first-order sensitivity to a deformation along direction $D$. Along the two PDS-preserving directions ($h_0$ and $h_4$), every symmetry-pure state is rigid to $R \lesssim 3\times10^{-10}$ while every mixed state responds with $R \gtrsim 70$ — a separation exceeding ten orders of magnitude with zero misclassifications at $N=8$ and $N=10$. Along the $SU(3)$-breaking direction $\hat{n}_d$ both families respond at $R \sim 1$, showing rigidity is a property of the PDS-preserving operator cone, not of the states in isolation. The solvable subspace overlap under the PDS-preserving flow is $1.000000$ at $N=12$.

• PDS is not generic to a quantum phase transition. On the canonical consistent-$Q$ line $\hat{H}(\xi)=(1-\xi)\hat{n}_d-(\xi/4N)\hat{Q}\cdot\hat{Q}$, the fidelity susceptibility peaks at $\xi_c = 0.500$ and $\langle\hat{n}_d\rangle/N$ jumps from $0$ to $0.66$, yet there are zero nontrivial exact-label states in the interior for every $\xi \in {0.10,\dots,0.90}$ (excluding the trivial $n_d=0$ and $n_d=N$ multiplets).

• First-order critical point: a solvable subset amid strong mixing (type I). At $\hat{H}(\beta_0=\sqrt{2})$ with $N=10$ (a 203-dimensional $M=0$ block), 58 exact-label states coexist with 145 doubly-mixed states. Of the 58, fifty-six are the closed-form $SU(3)$ tower (ground band $(2N,0)$ plus $\gamma^k$ bands $(2N-4k,2k)$ with energy $E=6h_2k(2N-2k+1)+CL(L+1)$), and two are irrep-pure but not closed-form solvable. The latter arise from $K$-band multiplicity (e.g. irrep $(16,2)$ contains $L=16$ in both its $K=2$ and $K=0$ bands; irrep $(12,4)$ contains $L=13$ with multiplicity two). The deformed $L=0$ band head sits at $\langle\hat{n}_d\rangle/N = 0.63$.

• The two-tier structure is structural, not a small-$N$ artifact. Closed-form solvable counts are $22, 37, 56, 79$ for $N=6,8,10,12$; exact-label totals are $24, 39, 58, 81$ over the same $N$. Class II (pure but not solvable) holds steady at two states for every $N \geq 6$.

• Second-order critical point: one shared label (type II). The Hamiltonian of Eq. (6) is an $O(5)$ scalar — its commutator with $\hat{C}2[O(5)]$ has Frobenius norm $2.7\times10^{-12}$ — so every eigenstate keeps exact seniority. Over the whole spectrum at $N=10$, $\max\psi \mathrm{Var},\hat{C}_2[O(5)] = 2.7\times10^{-11}$. The label organizes entanglement into a seniority ladder: mean $s|d$ entropy $\langle S\rangle = 0.71$ at $\tau=0$, $0.16$ at $\tau=6$, and $0.00$ at $\tau=N$.

• The separation does not close with boson number. From $N=4$ to $N=16$ the maximum pure-state residual stays at the diagonalization floor ($\lesssim 3\times10^{-9}$ at $N=16$) while the smallest mixed-state variance stays bounded away from zero ($\gtrsim 22$ at every $N$). The exact-label fraction shrinks from $0.67$ to $0.15$, so the diagnostic selects a progressively smaller subset — it sharpens rather than degrades. The reported separation ratio exceeds $10^9$ throughout, but the authors are careful to describe it as a numerical classification ratio (its denominator is the diagonalization floor), not a physical one. The median mixed variance grows as $N^{2.3}$ by log-log fit, widening toward $N^4$ at the top.

• Threshold robustness. At the stable $SU(3)$ point ($N=10$, 203 states), the largest exact-label variance is $6\times10^{-10}$ and the smallest nonzero one is $24$. The exact-label count is 58 for any threshold from $10^{-8}$ to $1$, dropping to 50 only at $10^{-10}$, the noise floor of the degeneracy refinement.

• Controlled breaking behaves perturbatively. Under $\hat{H}_{\rm PDS}+\epsilon,\hat{n}_d$, the mean label variance rises as $\epsilon^2$; a log-log fit over $\epsilon \leq 0.05$ gives exponent $1.97$. A second, linearly independent breaking direction ($\hat{C}_2[O(5)]$) gives exponent $1.8$ over $\epsilon \leq 0.02$. Both are described as finite-window fits consistent with the perturbative law, not asymptotic values. The pure-state variance stays below the mixed band up to breaking of order ten percent.

• Anchoring in a real nucleus. The analysis is anchored in $^{168}$Er, where block probabilities match Leviatan's fitted $SU(3)$ admixtures and reproduce his Hamiltonian-parameter-independent $\gamma \to g$ $B(E2)$ ratios.

• Conjugate-irrep caveat handled explicitly. The $SU(3)$ quadratic Casimir $f_2(\lambda,\mu)=\lambda^2+\mu^2+\lambda\mu+3(\lambda+\mu)$ is symmetric under $(\lambda,\mu)\leftrightarrow(\mu,\lambda)$, so conjugate irreps are degenerate whenever $N$ is a multiple of three — occurring at $N=6,9,12,15$ within the studied range (e.g. $(6,0)$ and $(0,6)$ at $N=6$; four pairs at $N=12$). A Hermitian cubic Casimir was constructed and used to recheck every classification; every state with vanishing $SU(3)$ quadratic-Casimir variance also has vanishing cubic-Casimir variance, so no conjugate-irrep mixture masquerades as symmetry-pure.

• Convention checks. The operator identity of Eq. (4) is reproduced to $\leq 2\times10^{-13}$ for $N=3,4,5,8,10$; the first-order critical Hamiltonian equals half the $h_2$ term of the stable Hamiltonian to $3\times10^{-14}$; Eq. (10) is verified numerically to $3\times10^{-10}$.

Methodology in Plain English

The author works with a well-known algebraic model of nuclear collective motion in which "$s$" bosons (spin 0) and "$d$" bosons (spin 2) represent correlated nucleon pairs. Fixing the total boson number $N$ and restricting to the $M=0$ block (which contains exactly one member of every angular-momentum multiplet) keeps the matrices manageable, and everything is computed by exact diagonalization — no approximation.

The key move is to pick a symmetry subgroup, compute its quadratic Casimir operator, and ask a simple question of each eigenstate: does this state live entirely inside one symmetry sector, or is it spread across several? If it lives in one sector, the Casimir has a single definite value and its variance is zero. The author proves this is equivalent to two other statements — a block-coherence entropy being zero and a block purity equalling one — so one physical condition can be read three ways.

To check the answer is meaningful and not a tautology, the author then does several independent tests. First, a "decision gate" showing entanglement magnitude does not work, which rules out later separations being trivially explained. Second, a rigidity test: nudge each state along directions that are supposed to preserve the symmetry and see whether it moves. Pure states do not move; mixed states move a lot. Crucially, the same pure states do move when nudged along a symmetry-breaking direction, which shows the rigidity is real and not an artifact of being an eigenstate. Third, the separation is tracked as $N$ grows, as the symmetry is deliberately broken by small amounts, and at two different critical points constructed by Leviatan.

For the quantum-information side, the model is encoded on a qubit register and the fingerprint prepared and read out variationally — with the explicit caveat that time evolution is used only as a tool and no quantum advantage is claimed.

Why This Matters

Impact on research. PDS has been a spectroscopic concept — something inferred from level patterns and transition-rate ratios measured in experiments. This work turns it into a wavefunction property that can be computed directly and, in principle, measured. That opens a two-way street: theory can predict which eigenstates of a candidate PDS nucleus are symmetry-pure, and experiment (or quantum simulation) can in principle test the prediction without a full spectroscopic fit. The paper also supplies a clean conceptual separation that the field needed: irrep purity, closed-form solvability, and entanglement magnitude are three different things, and conflating them leads to wrong inferences. The negative results (entanglement magnitude does not separate; PDS is not generic to a QPT) are as useful as the positive ones.

Potential applications (not claimed as demonstrated by the paper):

• Quantum-hardware benchmarking. A symmetry label with an exactly known answer is a natural calibration target for near-term processors: a device that correctly reproduces the solvable/mixed pattern at small $N$ is doing something structurally right, not just reproducing a smooth observable.
• Guiding spectroscopic analysis of deformed nuclei. The $^{168}$Er anchoring suggests the diagnostic could be used to flag which states in a measured band are likely symmetry-pure before committing to a full PDS fit.
• Quantum simulation of algebraic models. The Casimir Loschmidt echo is a form that can be implemented as a circuit primitive, suggesting this class of algebraic nuclear models as a testbed for symmetry-resolved quantum algorithms.
• Transferable symmetry diagnostics. The variance/coherence/purity triple is not specific to the IBM; it applies to any model with a nested subalgebra structure, including molecular and condensed-matter models with similar algebraic underpinnings.

Industry relevance. Direct industrial relevance is limited and the paper does not claim any quantum advantage. The most plausible engagement points are quantum-computing vendors interested in symmetry-based verification and benchmarking of processors, and research groups in nuclear data and nuclear energy modeling who rely on collective-model descriptions of heavy nuclei.

Future Directions

1. Extend beyond the $sd$ model. The paper works entirely in the $sd$-IBM; the natural test is whether the same label-variance classifier fires in the proton–neutron $s$-$d$ IBM-2, in the $spdf$ extension, or in fermionic shell-model settings where PDS structures have also been proposed.

2. Scale the hardware readout. The processor demonstration is at $N=3$ with a single solvable/mixed exemplar. Whether the pattern survives at larger qubit counts and under realistic noise — and whether the Casimir Loschmidt echo can be implemented without full state tomography — is unresolved in the supplied text.

3. Make the diagnostic experimentally actionable for real nuclei. The $^{168}$Er anchoring is a consistency check against existing fits. Turning this into a predictive tool requires connecting the block probabilities to measurable electromagnetic transition strengths beyond the $\gamma \to g$ $B(E2)$ ratios already reproduced.

4. Characterize the mixed-state sector structure. The paper reports the mixed states' variance distribution widening toward $N^4$ but explicitly declines to claim a specific power law. What the mixed sector's internal structure looks like — and whether it carries its own organizing principle — is left open.

Target Audience

Most beneficial: researchers in nuclear structure theory working on algebraic models and partial/quasi dynamical symmetries, who will find the state-resolved classifier directly usable and the purity-versus-solvability distinction clarifying. Quantum-information researchers interested in symmetry as a resource or as a benchmark will find the variance/coherence/purity unification and the hardware demonstration relevant, though they should be prepared for the nuclear-physics formalism. Graduate students in either area looking for a well-defined bridge problem between algebraic many-body models and QI diagnostics would get a great deal from it, provided they have the algebraic background. Experimental nuclear spectroscopists working on deformed rare-earth nuclei will find the $^{168}$Er section the most immediately accessible entry point.

Less suited to: readers seeking near-term quantum-advantage claims (none are made), or those without comfort in Casimir operators and irreducible-representation labels.

Authors’ abstract

Partial dynamical symmetry (PDS) is an algebraic structure in which a symmetry is neither exact nor fully broken: a subset of eigenstates keeps good quantum numbers and stays solvable while the rest mix. It is currently identified from spectroscopy. We ask whether it also leaves a purely structural mark in the eigenstates, and find, for the $sd$-IBM Hamiltonians studied here, that it does, though not in the magnitude of entanglement. The diagnostic is the variance of a symmetry Casimir, the label variance $\Var\,\C[G]$, which shares its zero with a block-coherence entropy and a block impurity: all three vanish when a state occupies a single Casimir sector, a single irreducible representation once conjugate representations are resolved by the cubic Casimir. Resolved state by state, the variance is numerically zero on the symmetry-selected subset and large on the mixed states, isolating the PDS states from the wavefunctions alone; the separation does not close as the boson number grows and persists at Leviatan's first- and second-order critical points, whereas the magnitude of bipartite $s|d$ entanglement does not separate the two families. We anchor the analysis in $^{168}$Er, reproducing Leviatan's fitted $\su3$ admixtures and his Hamiltonian-parameter-independent $γ\!\to\! g$ $B(E2)$ ratios. The diagnostic also has an exact dynamical form, a Casimir Loschmidt echo rigid on the symmetry-pure states, and we demonstrate its readout on an IBM superconducting processor at $N=3$, where a single solvable/mixed exemplar reproduces the predicted pattern within finite-shot uncertainties.