The Lipkin method emulates exact variation after projection (VAP) in nuclear mean-field theory by minimizing a corrected energy of the symmetry-broken state, with corrections fixed by transition energy and overlap kernels at small symmetry shifts; for translational symmetry the correction involves the Peierls-Yoccoz mass, and Dobaczewski (2009) showed within Skyrme energy-density-functional (EDF) calculations that the Gaussian overlap approximation works excellently and that the Peierls-Yoccoz mass differs from the true mass by only a few percent, while explicitly leaving open its extension to rotation, isospin, higher orders, and its general reliability. Later work partially closed that list: particle-number restoration via gauge-angle kernels was implemented and benchmarked to sixth order (Wang et al., 2014), and a quadratic Lipkin operator for rotation was benchmarked against exact angular-momentum projection in well-deformed erbium isotopes only (Gao, Dobaczewski, Toivanen, 2015, unpublished). The problem posed here is the surviving open core: quantitatively implement and assess Lipkin-method restoration of rotational and isospin symmetry in realistic EDFs — including shell dependence, closed-shell and weakly-deformed nuclei, and the impact on mass systematics — substantiate or refute the claim that small-shift kernel evaluations avoid the singularities of density-dependent EDF kernels, and establish criteria for when Lipkin-VAP-optimized states plus a single projection substitute for exact VAP.

Background

Low-energy nuclear structure is dominated by self-consistent mean-field methods built on energy density functionals (EDFs), such as the Skyrme and Gogny functionals: the total energy is minimized over product-type Hartree-Fock or Hartree-Fock-Bogolyubov (HFB) quasiparticle vacua rather than over expectation values of a genuine Hamiltonian. Such product states generically break the symmetries of the exact ground state: translational invariance (the center of mass is localized), rotational invariance (deformed intrinsic shapes), the gauge symmetry associated with conserved particle number (superfluid pairing), and, near the N=Z line, isospin symmetry (the classification of nuclear states by total isospin T, broken mainly by the Coulomb interaction between protons).

Exact symmetry restoration is achieved by projection: with a projection operator, the component of the broken-symmetry state carrying the desired quantum number (linear momentum, angular momentum J, particle number N, or total isospin T) is extracted. Minimizing the projected energy over the intrinsic state — variation after projection (VAP) — gains correlation energy of order MeV per restored symmetry and yields the optimal intrinsic state for each quantum number; the cheaper projection after variation (PAV) recovers only part of it. VAP requires integrating transition kernels (overlap and energy matrix elements between the state and its symmetry-transformed images) over the whole symmetry group, which for several symmetries simultaneously is at, or beyond, the limit of current computational capability.

Already in 1960, Lipkin observed that exact VAP can be emulated by minimizing, over the symmetry-broken state alone, a 'corrected' energy in which a Lipkin operator K^\hat{K} — a low-order polynomial in the symmetry generators — is subtracted so as to flatten the spectrum of symmetry-projected energies. For a quadratic Lipkin operator the correction involves the Peierls-Yoccoz (PY) mass or moment of inertia: the inertial parameter obtained from the norm of the symmetry-projected state, which differs from the true inertial mass of the nucleus. The coefficients of K^\hat{K} can be determined from transition energy and overlap kernels between the state and small symmetry shifts of it (a spatial translation, a gauge-angle rotation, an Euler-angle rotation), which is why the method is called kernel-based; a related scheme is due to Kamlah, and the widely used Lipkin-Nogami (LN) method — the standard approximate particle-number VAP in large EDF codes — replaces the kernels by expectation values.

Dobaczewski (2009) systematically developed the kernel-based Lipkin method for translational-symmetry restoration within Skyrme-EDF calculations: the Gaussian overlap approximation (GOA — the assumption that the norm kernel is Gaussian in the collective coordinate) works excellently, the PY mass differs from the true mass by only a few percent, and replacing the exact mass by the PY mass in the Lipkin correction induces an almost constant energy shift of the order of 0.5 MeV with very small shell effects in lead isotopes. Because density-dependent EDFs are not generated from a Hamiltonian, their energy kernels develop poles (singularities) under the generalized Wick theorem — the exchange terms fail to cancel the zeros of the overlap kernel — which plague exact projection; whether these singularities contaminate the small-shift kernels used by the Lipkin method is a nontrivial question that the paper explicitly left open.

The paper closed by formulating the open issues that define this problem: rotational restoration, where the Lipkin operator contains J^2\hat{J}^2 and, being included before variation, may induce deformation even in spherical (magic) nuclei, with strongly shell-dependent corrections that could affect the agreement of calculated masses with data; isospin restoration, where the Lipkin operator is T^2\hat{T}^2 and the PY isospin mass must be built from nuclear (Coulomb-excluded) energy kernels so that the Coulomb energy induces the correct isospin mixing; higher-order Lipkin operators (polynomials of order M determined from M shifts, with criteria for shift selection and convergence); particle-number restoration via gauge-angle shifts as an alternative to LN; and the general question of when Lipkin-optimized states followed by a single projection can substitute for exact VAP. The particle-number items were subsequently resolved in the cited Wang et al. (2014) work; the rotational, isospin, kernel-singularity, and applicability-limit items were not, and they jointly define the single objective of this problem: establishing the quantitative range of applicability of the Lipkin VAP approximation in EDF theory beyond translational symmetry.

Problem Statement

For the Lipkin kernel-based approximate variation-after-projection method in nuclear energy-density-functional theory, beyond the translational symmetry for which it was established, determine quantitatively its applicability and reliability, specifically:

(1) Rotational symmetry: implement the rotational Lipkin method self-consistently in realistic EDF calculations (Lipkin operator built on J^2\hat{J}^2, with Peierls-Yoccoz moments of inertia determined from transition kernels at small Euler-angle shifts) and determine the magnitude and shell dependence of the rotational Lipkin corrections, their behavior in spherical and closed-shell nuclei — where inclusion of J^2\hat{J}^2 before variation may induce deformation — and in weakly-deformed (transitional) nuclei where the projected spectrum is not parabolic in JJ, and their net impact on the agreement of calculated masses with experimental data.

(2) Isospin symmetry: construct and implement the Lipkin operator built on T^2\hat{T}^2 with the Peierls-Yoccoz isospin mass computed from nuclear (Coulomb-excluded) energy kernels, so that the Coulomb energy induces the correct isospin mixing, and quantify the resulting isospin-restoration corrections and their dependence on shell structure and on the neutron-to-proton ratio, including near the N=Z line.

(3) Kernel singularities: substantiate or refute the claim that Lipkin evaluations at small collective-coordinate shifts avoid the singularities (poles) that plague exact projection with density-dependent EDFs, by analyzing the singularity-free or regularized kernels proposed in the literature and showing whether the required corrections are appropriately small for the shifts used in Lipkin procedures.

(4) Range of applicability: on the basis of (1)-(3), and of the already-available particle-number results, establish criteria — choice of the Lipkin-operator order MM, selection of the shifts {Rn}\{R_n\} avoiding ill-conditioning, convergence diagnostics such as shift-independence of the parameters — that certify when a Lipkin-VAP-optimized state followed by a single projection quantitatively reproduces the exact variation-after-projection result, and for which symmetries, nuclei, and observables the method fails.

These four items are jointly required, and together serve one overarching objective: to determine the range of applicability and the quantitative reliability of the Lipkin kernel-based VAP approximation in EDF theory beyond translational symmetry. No restriction to particular nuclei, EDF parametrizations, or operator orders is imposed; those choices belong to the answer, provided the assessment covers the regimes (well-deformed, transitional/weakly-deformed, and closed-shell nuclei; N=Z and neutron-rich nuclei for isospin) in which the method is claimed to be tested.

References

  1. H. J. Lipkin, 'Collective motion in many-particle systems', Annals of Physics 9, 272-291 (1960). DOI: 10.1016/0003-4916(60)90032-4, https://doi.org/10.1016/0003-4916(60)90032-4
  2. R. E. Peierls and J. Yoccoz, 'The Collective Model of Nuclear Motion', Proceedings of the Physical Society. Section A 70, 381-387 (1957). DOI: 10.1088/0370-1298/70/5/309, https://doi.org/10.1088/0370-1298/70/5/309
  3. J. Dobaczewski, M. V. Stoitsov, W. Nazarewicz, and P.-G. Reinhard, 'Particle-number projection and the density functional theory', Physical Review C 76, 054315 (2007). DOI: 10.1103/PhysRevC.76.054315, https://doi.org/10.1103/PhysRevC.76.054315
  4. J. Dobaczewski, 'Lipkin translational-symmetry restoration in the mean-field and energy-density-functional methods', Journal of Physics G: Nuclear and Particle Physics 36, 105105 (2009). DOI: 10.1088/0954-3899/36/10/105105, arXiv:0906.4763, https://arxiv.org/abs/0906.4763
  5. D. Lacroix, T. Duguet, and M. Bender, 'Configuration mixing within the energy density functional formalism: Removing spurious contributions from nondiagonal energy kernels', Physical Review C 79, 044318 (2009). DOI: 10.1103/PhysRevC.79.044318, https://doi.org/10.1103/PhysRevC.79.044318
  6. M. Bender, T. Duguet, and D. Lacroix, 'Particle-number restoration within the energy density functional formalism', Physical Review C 79, 044319 (2009). DOI: 10.1103/PhysRevC.79.044319, https://doi.org/10.1103/PhysRevC.79.044319
  7. T. Duguet, M. Bender, K. Bennaceur, D. Lacroix, and T. Lesinski, 'Particle-number restoration within the energy density functional formalism: Nonviability of terms depending on noninteger powers of the density matrices', Physical Review C 79, 044320 (2009). DOI: 10.1103/PhysRevC.79.044320, https://doi.org/10.1103/PhysRevC.79.044320
  8. W. Satuła, J. Dobaczewski, W. Nazarewicz, and M. Rafalski, 'Isospin Mixing in Nuclei within the Nuclear Density Functional Theory', Physical Review Letters 103, 012502 (2009). DOI: 10.1103/PhysRevLett.103.012502, https://doi.org/10.1103/PhysRevLett.103.012502
  9. W. Satuła, J. Dobaczewski, W. Nazarewicz, and M. Rafalski, 'Isospin-symmetry restoration within the nuclear density functional theory: Formalism and applications', Physical Review C 81, 054310 (2010). DOI: 10.1103/PhysRevC.81.054310, https://doi.org/10.1103/PhysRevC.81.054310
  10. T. Duguet and J. Sadoudi, 'Breaking and restoring symmetries within the nuclear energy density functional method', Journal of Physics G: Nuclear and Particle Physics 37, 064009 (2010). DOI: 10.1088/0954-3899/37/6/064009, arXiv:1001.0673, https://arxiv.org/abs/1001.0673
  11. X. B. Wang, J. Dobaczewski, M. Kortelainen, L. F. Yu, and M. V. Stoitsov, 'Lipkin method of particle-number restoration to higher orders', Physical Review C 90, 014312 (2014). DOI: 10.1103/PhysRevC.90.014312, arXiv:1403.6364, https://arxiv.org/abs/1403.6364
  12. W. Satuła and J. Dobaczewski, 'Simple regularization scheme for multireference density functional theories', Physical Review C 90, 054303 (2014). DOI: 10.1103/PhysRevC.90.054303, arXiv:1407.0857, https://arxiv.org/abs/1407.0857
  13. Y. Gao, J. Dobaczewski, and P. Toivanen, 'Approximate restoration of translational and rotational symmetries within the Lipkin method', arXiv preprint arXiv:1511.02814 (2015), https://arxiv.org/abs/1511.02814 (unpublished preprint)
  14. J. A. Sheikh, J. Dobaczewski, P. Ring, L. M. Robledo, and C. Yannouleas, 'Symmetry restoration in mean-field approaches', Journal of Physics G: Nuclear and Particle Physics 48, 123001 (2021). DOI: 10.1088/1361-6471/ac288a, arXiv:1901.06992, https://arxiv.org/abs/1901.06992
  15. M. Kafker and A. Bulgac, 'Impact of the Center of Mass Fluctuations on the Ground State Properties of Nuclei', arXiv preprint arXiv:2503.09470 (2025), https://arxiv.org/abs/2503.09470