Genericity of universality features in renormalization-group limit cycles: marginal operators in discrete cycles, tunability of irrelevant operators, and multiplicity of geometric spectra

Głazek and Wilson (Phys. Rev. B 69, 094304 (2004)) exactly solved a discrete quantum-mechanical Hamiltonian model whose Wilsonian renormalization group (RG) exhibits a limit cycle, and showed: exactly one marginal operator and an infinite ladder of irrelevant operators with integer critical exponents and Wegner eigenvalues $w_j=r^{-j}$ independent of position on the cycle; a spectrum of $p$ geometric eigenvalue sequences (one negative, $p-1$ positive) with ratio $r=b^p$; and a demonstration that tuning four ultraviolet matrix elements eliminates the two leading irrelevant operators, accelerating convergence of the eigenvalue ratios to $r$ from rate $1/r$ to $1/r^3$ per cycle. The open problem is to determine which of these universality features are generic properties of RG limit cycles in other quantum Hamiltonian systems — most importantly realistic three-body (Efimov) Hamiltonians — and which are artifacts of the model's special solubility. The paper itself left open whether the ultraviolet tuning can be transferred to other limit-cycle systems (the map from Hamiltonian parameters to Wegner-operator coefficients is nonlinear, with no solvability guarantee) and whether a marginal operator must occur when the RG cycle is discrete (a finite set of points) rather than continuous. As of 2025 these four facets — marginal-operator necessity in discrete cycles, tunability of leading irrelevant operators in other systems, the generic multiplicity and sign structure of the geometric sequences, and robustness of integer exponents under deformations of the RG map — remain unresolved: the three-body program of Mohr, Furnstahl, Hammer, Perry and Wilson left the subleading corrections uncomputed, and the most recent three-body limit-cycle work (2025) still defers the inverse-power (irrelevant-operator) corrections to future studies.

Background

The renormalization group (RG) tracks how an effective Hamiltonian of a quantum system changes as a cutoff — the highest energy scale explicitly retained — is lowered by integrating out (or, in Hamiltonian versions, eliminating) degrees of freedom above it. In Wilson's original formulation the RG flow can converge to a fixed point (a Hamiltonian invariant under the transformation), but it can also approach a limit cycle: a closed orbit in the space of Hamiltonians that is traversed periodically, returning to its starting point each time the cutoff changes by a fixed multiplicative factor. Around any fixed point or cycle, infinitesimal deformations of the Hamiltonian are classified by Wegner's linearization: writing the deformation (an 'operator' in this sense) as an eigenvector of the linearized RG transformation, its eigenvalue $w$ labels it relevant ($w>1$: the deformation grows as the RG proceeds), marginal ($w=1$: it persists unchanged), or irrelevant ($w<1$: it decays). Critical exponents are related to these eigenvalues by $w_l=r^{\lambda_l}$, where $r$ is the scale factor of one full RG period. Irrelevant operators govern how fast a finite, cutoff-bounded system approaches the universal behavior — so knowing their spectrum is what lets one accelerate (or merely estimate) convergence to the universal physics.

The physically central example is the Efimov effect: three identical bosons with short-range interactions and a two-body bound state exactly at threshold (infinite two-body scattering length) possess an infinite tower of three-body bound states whose binding energies form a geometric sequence converging on zero (Efimov 1970), a phenomenon called discrete scale invariance. Bedaque, Hammer and van Kolck (1999) showed that renormalizing such a three-body system requires a pointlike three-body interaction whose coupling runs periodically in the logarithm of the cutoff — an RG limit cycle — so the three-body problem is the canonical 'other system' in which limit-cycle universality matters.

Głazek and Wilson (2002; 2004) constructed a solvable discrete Hamiltonian matrix model on a geometric energy lattice $E_n=b^n$ with matrix elements $H_{mn}=(E_mE_n)^{1/2}(\delta_{mn}-g_N-i h_N s_{mn})$ (with $s_{mn}=(m-n)/|m-n|$ for $m\ne n$), whose Wilsonian RG step (Gaussian elimination of the highest-energy basis state) is exactly solvable at criticality, that is, at the coupling values where the cycle occurs. When $h=\tan(\pi/p)$ with integer $p\ge 3$, the real coupling $g$ cycles with period $p$ in the elimination steps, one full cycle corresponding to an energy rescaling by $r=b^p$. The 2004 analysis established, analytically and generically in the model's parameters: exactly one marginal operator ($w_0=1$) and infinitely many irrelevant operators with integer critical exponents and Wegner eigenvalues $w_j=r^{-j}$; no relevant operators; the operators themselves vary with position on the cycle while their eigenvalues do not; a bound-state spectrum forming $p$ distinct geometric sequences with ratio $r$, of which exactly one is negative (bound states) and $p-1$ positive, for all parameter choices including the large-$b$ limit; and an explicit numerical demonstration that altering four ultraviolet matrix elements ($H_{NN}$, $H_{N,N-1}$, $H_{N-1,N}$, $H_{N-1,N-1}$) cancels the coefficients of the two leading irrelevant operators, after which the eigenvalue ratios converge to $r$ at rate $1/r^3$ per cycle instead of $1/r$.

Section VIII of the paper then asked which of these features carry over to general RG limit cycles. It proved two general statements: near-threshold bound states of a limit cycle must form a geometric series (from the scale invariance of the fixed-modulo-cycle Hamiltonian), and critical exponents cannot vary with position on a cycle. But it also made two explicit admissions of ignorance: the tuning procedure cannot be guaranteed to transfer to other limit-cycle systems (such as the three-body Hamiltonians) because the mapping from initial-Hamiltonian parameters to Wegner-operator coefficients is nonlinear, with no assurance that desired output coefficients are reachable; and the marginal operator is guaranteed only when the RG exists in differential form so that the cycle is a continuous curve — 'when the renormalization group is discrete, and the limit cycle is a finite set of points rather than a continuous curve in the space of Hamiltonians, we do not know whether or not a marginal operator has to occur.' Two further facets arise naturally from the model-limited character of the results: the one-negative/$(p-1)$-positive multiplicity of the geometric sequences is proven only within the model, and the persistence of integer critical exponents and exact power-law Wegner eigenvalues under deformations of the RG map (for example when relevant directions are present, as they are in the three-body case, or for continuous rather than discrete transformations) was not studied. The single scientific objective of the present problem is to replace these model-specific facts and admissions with a characterization of the universal operator and spectral structure of RG limit cycles in quantum Hamiltonian systems.

Problem Statement

Determine which of the universality features established for the exactly solvable Głazek-Wilson discrete Hamiltonian model are generic properties of renormalization-group limit cycles in quantum Hamiltonian systems, and establish the conditions under which they hold or fail. Concretely, with a limit cycle understood as a periodic orbit of a spectrum-preserving RG transformation (one full period corresponding to an energy rescaling by a factor $r$), and with Wegner operators understood as eigen-deformations of the linearized RG transformation about the cycle:

(a) Marginal operators in discrete cycles. For a discrete RG transformation whose limit cycle is a finite set of Hamiltonians (a finite periodic orbit) rather than a continuous closed curve, establish whether a marginal operator (Wegner eigenvalue exactly 1 of the linearized cycle-return map) must occur, must not occur, or occurs only under identifiable conditions; either a theorem covering a precisely stated class of spectrum-preserving discrete RG maps, or an explicit counterexample (a genuine discrete RG limit cycle whose linearized cycle-return map has no unit eigenvalue), resolves this facet.

(b) Tunability of leading irrelevant operators in other limit-cycle systems. For at least one other specified limit-cycle system — most importantly a realistic three-body (Efimov) Hamiltonian in a cutoff formulation, or another nontrivial quantum Hamiltonian with a limit cycle — determine whether finitely many local changes to the ultraviolet definition of the Hamiltonian (analogous to the alteration of the four highest-index matrix elements in the model) can set the coefficients of the leading Wegner irrelevant operators to zero or approximately zero, and demonstrate the resulting accelerated convergence of the low-energy spectrum to the universal geometric sequence (in the model, from rate $1/r$ to $1/r^3$ per cycle), or establish that no such finite tuning exists.

(c) Multiplicity and sign structure of the geometric sequences. For classes of limit-cycle Hamiltonians beyond the model, determine the generic number and sign structure of the distinct geometric eigenvalue sequences converging on zero energy (the model has exactly one negative and $p-1$ positive sequences, where $p$ denotes the cycle period in RG steps and $r=b^p$; the Efimov and inverse-square cases have a single tower), in particular whether any general relation exists between the structure of the cycle and the spectral multiplicity, or whether the pattern is model-dependent. This facet is resolved either by proving a general relation between the structure of the cycle and the number and signs of the geometric sequences, or by exhibiting two limit-cycle systems beyond the model with genuinely different multiplicity, thereby establishing model-dependence (neither the model itself nor the multiplicities already stated above supplies such a pair); a multiplicity computed for a single additional system does not by itself resolve it.

(d) Robustness of integer critical exponents. Determine whether integer critical exponents and exact power-law Wegner eigenvalues $w_j=r^{-j}$ persist under deformations of the RG map — for example when relevant directions are present (as in the three-body case, where the two-body interaction strength is relevant), when the RG is continuous rather than a discrete step transformation, or when the regulator is changed — or whether the exponents shift continuously with the deformation. Settling each of these three listed deformation types (determining, for that type, whether the exponents persist as integers and exact powers or shift continuously) resolves this facet; a general statement for arbitrary spectrum-preserving deformations would be stronger but is not required.

These four facets are jointly posed: taken together they constitute the objective of replacing the model-specific evidence of the Głazek-Wilson analysis with a general characterization of the operator and spectral structure of RG limit cycles.

References

1. Stanisław D. Głazek and Kenneth G. Wilson, 'Universality, marginal operators, and limit cycles', Phys. Rev. B 69, 094304 (2004). DOI: 10.1103/PhysRevB.69.094304; arXiv:cond-mat/0303297. https://arxiv.org/abs/cond-mat/0303297
2. Stanisław D. Głazek and Kenneth G. Wilson, 'Limit Cycles in Quantum Theories', Phys. Rev. Lett. 89, 230401 (2002). DOI: 10.1103/PhysRevLett.89.230401. https://doi.org/10.1103/PhysRevLett.89.230401
3. Franz J. Wegner, 'Corrections to Scaling Laws', Phys. Rev. B 5, 4529 (1972). DOI: 10.1103/PhysRevB.5.4529. https://doi.org/10.1103/PhysRevB.5.4529
4. V. Efimov, 'Energy levels arising from resonant two-body forces in a three-body system', Phys. Lett. B 33, 563 (1970). DOI: 10.1016/0370-2693(70)90349-7. https://doi.org/10.1016/0370-2693(70)90349-7
5. P. F. Bedaque, H.-W. Hammer, and U. van Kolck, 'Renormalization of the Three-Body System with Short-Range Interactions', Phys. Rev. Lett. 82, 463 (1999). DOI: 10.1103/PhysRevLett.82.463. https://doi.org/10.1103/PhysRevLett.82.463
6. Thomas Barford and Michael C. Birse, 'Effective theories of scattering with an attractive inverse-square potential and the three-body problem', J. Phys. A: Math. Gen. 38, 697 (2005). DOI: 10.1088/0305-4470/38/3/009; arXiv:nucl-th/0406008. https://arxiv.org/abs/nucl-th/0406008
7. R. F. Mohr, R. J. Furnstahl, H.-W. Hammer, R. J. Perry, and K. G. Wilson, 'Precise numerical results for limit cycles in the quantum three-body problem', Ann. Phys. 321, 225 (2006). DOI: 10.1016/j.aop.2005.10.002; arXiv:nucl-th/0509076. https://arxiv.org/abs/nucl-th/0509076
8. Eric Braaten and H.-W. Hammer, 'Universality in few-body systems with large scattering length', Phys. Rep. 428, 259 (2006). DOI: 10.1016/j.physrep.2006.03.001; arXiv:cond-mat/0410417. https://arxiv.org/abs/cond-mat/0410417
9. H.-W. Hammer and Brian G. Swingle, 'On the limit cycle for the 1/r^2 potential in momentum space', Ann. Phys. 321, 306 (2006). DOI: 10.1016/j.aop.2005.04.017; arXiv:quant-ph/0503074. https://arxiv.org/abs/quant-ph/0503074
10. Alberto Anfossi, André LeClair, and Germán Sierra, 'The elementary excitations of the exactly solvable Russian doll BCS model of superconductivity', J. Stat. Mech. P05011 (2005). DOI: 10.1088/1742-5468/2005/05/P05011; arXiv:cond-mat/0503014. https://arxiv.org/abs/cond-mat/0503014
11. Stanisław D. Głazek, 'Limit cycles of effective theories', Phys. Rev. D 75, 025005 (2007). DOI: 10.1103/PhysRevD.75.025005; arXiv:hep-th/0611015. https://arxiv.org/abs/hep-th/0611015
12. Stanisław D. Głazek, 'Limit Cycles in Quantum Mechanics', in Mathematical Physics of Quantum Mechanics, Lecture Notes in Physics 690, pp. 65-78 (Springer, 2006). DOI: 10.1007/3-540-34273-7_8. https://link.springer.com/chapter/10.1007/3-540-34273-7_8
13. S. M. Dawid, R. Gonsior, J. Kwapisz, K. Serafin, M. Tobolski, and S. D. Głazek, 'Renormalization group procedure for potential -g/r^2', Phys. Lett. B 777, 260 (2018). DOI: 10.1016/j.physletb.2017.12.028; arXiv:1704.08206. https://arxiv.org/abs/1704.08206
14. Michael C. Birse, 'The renormalization group and nuclear forces', Phil. Trans. R. Soc. A 369, 2662 (2011). DOI: 10.1098/rsta.2010.0381; arXiv:1012.4914. https://arxiv.org/abs/1012.4914
15. Lauro Tomio, Tobias Frederico, Varese S. Timóteo, and Marcelo T. Yamashita, 'Fixed-Point Few-Body Hamiltonians', Few-Body Syst. 63, 12 (2021). DOI: 10.1007/s00601-021-01714-0; arXiv:2111.05954. https://arxiv.org/abs/2111.05954
16. Langxuan Chen, Feng Wu, Xincheng Lin, Sebastian König, Ubirajara van Kolck, and Pengfei Zhang, 'The Three-Body Limit Cycle: Universal Form for General Regulators' (2025). arXiv:2509.04746. https://arxiv.org/abs/2509.04746
17. M. Girguś and S. D. Głazek, 'Spiral cutoff-flow of quantum quartic oscillator', Nucl. Phys. B 1010, 116776 (2025). DOI: 10.1016/j.nuclphysb.2024.116776; arXiv:2404.17446. https://arxiv.org/abs/2404.17446
18. André LeClair, 'A rich structure of renormalization group flows for Higgs-like models in 4 dimensions' (2024). arXiv:2411.07476. https://arxiv.org/abs/2411.07476