Quadrupole–Hexadecapole Configuration Mixing

• Quadrupole–hexadecapole mixing is the nonseparable coupling between rank-2 and rank-4 multipole modes, crucial for understanding nuclear deformations and collective excitations.
• Methodologies such as two-dimensional GCM, interacting-boson models, and surface–volume analyses reveal tilted collective valleys and explicit off-diagonal coupling terms.
• This mixing significantly alters correlation energies, excitation spectra, and transport phenomena in nuclei, heavy-ion collisions, and antiferromagnetic systems through symmetry-driven interactions.

Quadrupole–hexadecapole configuration mixing denotes the nonseparable coupling between rank-2 and rank-4 multipole degrees of freedom, most commonly the axial mass moments $Q_{20}$ and $Q_{40}$ or the corresponding deformation parameters $\beta_2$ and $\beta_4$. In nuclear structure it appears as a tilted collective valley in the $(\beta_2,\beta_4)$ plane, as off-diagonal couplings in a two-dimensional Generator Coordinate Method (GCM), or as admixture between $d$- and $g$-boson configurations in $sdg$ interacting-boson models. In deformed-density parameterizations it also appears as a mismatch between volume multipoles and Woods–Saxon surface parameters, so that a surface hexadecapole term contributes to the volume quadrupole moment. In correlated electron systems, an analogous phenomenon occurs when magnetic quadrupole and hexadecapole order parameters belong to the same irreducible representation and therefore admix once symmetry is lowered. Across these settings, the central theme is that quadrupole and hexadecapole modes are often interwoven rather than independent, with consequences for correlation energies, excitation spectra, transition strengths, rotational response, flow observables, and magnetoelectric transport (Kumar et al., 2023, Ryssens et al., 2023, Watanabe et al., 2017, Rodriguez-Guzman et al., 7 Feb 2025).

1. Operator content and symmetry conditions

The standard intrinsic mass multipole operators are

$$\hat Q_{20}=\sum_{i=1}^A r_i^2\,Y_{20}(\hat r_i),\qquad \hat Q_{40}=\sum_{i=1}^A r_i^4\,Y_{40}(\hat r_i),$$

with associated deformation coordinates

$$\beta_\ell=\frac{\sqrt{4\pi(2\ell+1)}}{3\,R_0^\ell\,A}\, \langle\varphi|\hat Q_{\ell 0}|\varphi\rangle,\qquad R_0=1.2\,A^{1/3}\,\mathrm{fm},\qquad \ell=2,4.$$

Equivalent definitions are used in Gogny-HFB, GCM, and boson-mapping studies, sometimes with $Q_{40}$0 and $Q_{40}$1 in Cartesian form (Kumar et al., 2023, Rodriguez-Guzman et al., 7 Aug 2025).

A recurrent microscopic signature of mixing is that the minimum-energy path in $Q_{40}$2 space is tilted rather than aligned with either axis. In $Q_{40}$3Sm, the bottom of the prolate valley follows roughly $Q_{40}$4, so quadrupole and hexadecapole deformations vary in lockstep; in actinides, the principal valley direction in $Q_{40}$5U is reported as $Q_{40}$6 with $Q_{40}$7 and $Q_{40}$8 (Kumar et al., 2023, Rodriguez-Guzman et al., 7 Feb 2025). Such relations are the geometric expression of a nonzero mixed curvature in the collective potential.

In magnetic systems the same logic is formulated group-theoretically. For Ba$Q_{40}$9K$\beta_2$0Mn$\beta_2$1As$\beta_2$2, the crystal space group is $\beta_2$3 with point group $\beta_2$4, while G-type antiferromagnetism breaks $\beta_2$5 and global inversion $\beta_2$6 but preserves $\beta_2$7, reducing the point symmetry to $\beta_2$8. Under $\beta_2$9, only the $\beta_4$0 irrep reduces to the totally symmetric $\beta_4$1 of $\beta_4$2, so the antiferromagnetic order parameter must transform as $\beta_4$3. The lowest-rank magnetic multipoles in this irrep are the magnetic quadrupole $\beta_4$4 and the magnetic hexadecapole $\beta_4$5, which therefore can admix on symmetry grounds (Watanabe et al., 2017).

2. Collective and microscopic formulations of the mixing problem

The most explicit formulation in nuclear structure is the two-dimensional GCM built from Hartree–Fock–Bogoliubov intrinsic vacua constrained simultaneously in $\beta_4$6 and $\beta_4$7: $\beta_4$8 with weights determined by the Hill–Wheeler–Griffin equation

$\beta_4$9

The normalized amplitudes $(\beta_2,\beta_4)$0 then provide a probability distribution in the $(\beta_2,\beta_4)$1 plane (Kumar et al., 2023). Closely related Gogny-EDF studies of rare-earth and actinide nuclei use the same structure without full angular-momentum projection, again interpreting $(\beta_2,\beta_4)$2 as the collective probability density (Rodriguez-Guzman et al., 7 Aug 2025, Rodriguez-Guzman et al., 7 Feb 2025).

Near a minimum, the microscopic origin of mixing is often summarized by a quadratic expansion of the collective potential,

$(\beta_2,\beta_4)$3

where $(\beta_2,\beta_4)$4. The cross-term $(\beta_2,\beta_4)$5 is the static quadrupole–hexadecapole coupling. In the adiabatic description of actinides there is also an off-diagonal inertia $(\beta_2,\beta_4)$6, so coupling enters both the potential and kinetic sectors of the collective Hamiltonian (Rodriguez-Guzman et al., 7 Feb 2025).

Not every study constructs an explicit off-diagonal collective kernel. In the macroscopic–microscopic Woods–Saxon plus HFBC treatment of the $(\beta_2,\beta_4)$7 region, the coupling is introduced directly through the simultaneous $(\beta_2,\beta_4)$8 and $(\beta_2,\beta_4)$9 dependence of the deformed surface in the mean-field Hamiltonian. The work explicitly states that it does not build a GCM-type mixing Hamiltonian in collective coordinates, but instead explores the coupling through total-energy-surface scans and the induced single-particle admixtures at a fixed equilibrium shape (Li et al., 15 Jan 2026).

In odd-parity magnetic metals, a Landau description is natural. Writing $d$0 and $d$1, the free energy is

$d$2

In the insulating limit, $d$3 condenses at $d$4 and the order is pure hexadecapolar; in the metallic state, LS coupling generates $d$5, so a nonzero $d$6 generically induces $d$7 (Watanabe et al., 2017). A plausible implication is that “configuration mixing” across subfields refers less to a unique formalism than to a common structural feature: symmetry-allowed off-diagonal coupling between rank-2 and rank-4 sectors.

3. Beyond-mean-field nuclear structure: correlation energies, wave functions, and shape coexistence

The axial HFB+GCM study of even-even Sm and Gd isotopes finds strong coupling between the quadrupole and hexadecapole degrees of freedom. For nuclei around $d$8, the static HFB surface shows a well-developed prolate valley with positive $d$9, a much higher-lying oblate pocket with $g$0, and near-spherical minima at the semi-magic nucleus $g$1Gd. In $g$2Sm, the ground-state collective wave function is a tilted two-dimensional Gaussian elongated along the valley direction, the first excited $g$3 is a one-phonon excitation along that same mixed $g$4–$g$5 direction, and the second $g$6 is predominantly one-phonon along the perpendicular direction (Kumar et al., 2023).

The same study reports that the gain from the $g$7–$g$8 space is $g$9 MeV in Sm/Gd, about twice the $sdg$0-only result, implying a non-negligible hexadecapole contribution of about $sdg$1 MeV. The first and second $sdg$2 excitations appear at $sdg$3 MeV in most cases, while in isotopes such as $sdg$4Gd and lighter Sm/Gd near $sdg$5 the oblate pocket comes down, so the first excited $sdg$6 often localizes in the oblate minimum at $sdg$7 MeV with $sdg$8 and $sdg$9 (Kumar et al., 2023). This distinguishes genuine mixed-mode vibrational states from low-lying shape-coexisting configurations.

For Yb, Hf, W, and Os isotopes in the interval $$\hat Q_{20}=\sum_{i=1}^A r_i^2\,Y_{20}(\hat r_i),\qquad \hat Q_{40}=\sum_{i=1}^A r_i^4\,Y_{40}(\hat r_i),$$0, a 2D-GCM analysis finds that ground and excited states of the lighter isotopes are associated with diamond-like shapes, while a region of square-like shapes occurs below the neutron shell closure $$\hat Q_{20}=\sum_{i=1}^A r_i^2\,Y_{20}(\hat r_i),\qquad \hat Q_{40}=\sum_{i=1}^A r_i^4\,Y_{40}(\hat r_i),$$1. The quadrupole and hexadecapole degrees of freedom are reported to be interwoven up to about $$\hat Q_{20}=\sum_{i=1}^A r_i^2\,Y_{20}(\hat r_i),\qquad \hat Q_{40}=\sum_{i=1}^A r_i^4\,Y_{40}(\hat r_i),$$2–$$\hat Q_{20}=\sum_{i=1}^A r_i^2\,Y_{20}(\hat r_i),\qquad \hat Q_{40}=\sum_{i=1}^A r_i^4\,Y_{40}(\hat r_i),$$3, after which the collective wave functions align predominantly along the $$\hat Q_{20}=\sum_{i=1}^A r_i^2\,Y_{20}(\hat r_i),\qquad \hat Q_{40}=\sum_{i=1}^A r_i^4\,Y_{40}(\hat r_i),$$4 axis and the motion becomes effectively quadrupole-dominated. The total correlation energy $$\hat Q_{20}=\sum_{i=1}^A r_i^2\,Y_{20}(\hat r_i),\qquad \hat Q_{40}=\sum_{i=1}^A r_i^4\,Y_{40}(\hat r_i),$$5 ranges from about $$\hat Q_{20}=\sum_{i=1}^A r_i^2\,Y_{20}(\hat r_i),\qquad \hat Q_{40}=\sum_{i=1}^A r_i^4\,Y_{40}(\hat r_i),$$6 MeV up to about $$\hat Q_{20}=\sum_{i=1}^A r_i^2\,Y_{20}(\hat r_i),\qquad \hat Q_{40}=\sum_{i=1}^A r_i^4\,Y_{40}(\hat r_i),$$7 MeV, with an additional hexadecapole gain $$\hat Q_{20}=\sum_{i=1}^A r_i^2\,Y_{20}(\hat r_i),\qquad \hat Q_{40}=\sum_{i=1}^A r_i^4\,Y_{40}(\hat r_i),$$8 MeV (Rodriguez-Guzman et al., 7 Aug 2025).

Actinide calculations based on Gogny HFB and GCM reach a parallel conclusion. For selected Ra, Th, U, and Pu isotopes, static hexadecapole deformations are sizable near $$\hat Q_{20}=\sum_{i=1}^A r_i^2\,Y_{20}(\hat r_i),\qquad \hat Q_{40}=\sum_{i=1}^A r_i^4\,Y_{40}(\hat r_i),$$9U, and a region with small negative hexadecapole deformation just below $$\beta_\ell=\frac{\sqrt{4\pi(2\ell+1)}}{3\,R_0^\ell\,A}\, \langle\varphi|\hat Q_{\ell 0}|\varphi\rangle,\qquad R_0=1.2\,A^{1/3}\,\mathrm{fm},\qquad \ell=2,4.$$0 remains stable once zero-point quadrupole–hexadecapole fluctuations are included. The quoted correlation energies are $$\beta_\ell=\frac{\sqrt{4\pi(2\ell+1)}}{3\,R_0^\ell\,A}\, \langle\varphi|\hat Q_{\ell 0}|\varphi\rangle,\qquad R_0=1.2\,A^{1/3}\,\mathrm{fm},\qquad \ell=2,4.$$1 MeV and $$\beta_\ell=\frac{\sqrt{4\pi(2\ell+1)}}{3\,R_0^\ell\,A}\, \langle\varphi|\hat Q_{\ell 0}|\varphi\rangle,\qquad R_0=1.2\,A^{1/3}\,\mathrm{fm},\qquad \ell=2,4.$$2 MeV, so the extra gain from including $$\beta_\ell=\frac{\sqrt{4\pi(2\ell+1)}}{3\,R_0^\ell\,A}\, \langle\varphi|\hat Q_{\ell 0}|\varphi\rangle,\qquad R_0=1.2\,A^{1/3}\,\mathrm{fm},\qquad \ell=2,4.$$3 is $$\beta_\ell=\frac{\sqrt{4\pi(2\ell+1)}}{3\,R_0^\ell\,A}\, \langle\varphi|\hat Q_{\ell 0}|\varphi\rangle,\qquad R_0=1.2\,A^{1/3}\,\mathrm{fm},\qquad \ell=2,4.$$4 MeV. For $$\beta_\ell=\frac{\sqrt{4\pi(2\ell+1)}}{3\,R_0^\ell\,A}\, \langle\varphi|\hat Q_{\ell 0}|\varphi\rangle,\qquad R_0=1.2\,A^{1/3}\,\mathrm{fm},\qquad \ell=2,4.$$5U specifically, the reported values are $$\beta_\ell=\frac{\sqrt{4\pi(2\ell+1)}}{3\,R_0^\ell\,A}\, \langle\varphi|\hat Q_{\ell 0}|\varphi\rangle,\qquad R_0=1.2\,A^{1/3}\,\mathrm{fm},\qquad \ell=2,4.$$6 MeV, $$\beta_\ell=\frac{\sqrt{4\pi(2\ell+1)}}{3\,R_0^\ell\,A}\, \langle\varphi|\hat Q_{\ell 0}|\varphi\rangle,\qquad R_0=1.2\,A^{1/3}\,\mathrm{fm},\qquad \ell=2,4.$$7 MeV, and therefore $$\beta_\ell=\frac{\sqrt{4\pi(2\ell+1)}}{3\,R_0^\ell\,A}\, \langle\varphi|\hat Q_{\ell 0}|\varphi\rangle,\qquad R_0=1.2\,A^{1/3}\,\mathrm{fm},\qquad \ell=2,4.$$8 MeV (Rodriguez-Guzman et al., 7 Feb 2025). The data are consistent with the repeated observation that hexadecapole correlations contribute an amount of binding comparable to the quadrupole correlation energy itself.

A more mean-field-oriented treatment of $$\beta_\ell=\frac{\sqrt{4\pi(2\ell+1)}}{3\,R_0^\ell\,A}\, \langle\varphi|\hat Q_{\ell 0}|\varphi\rangle,\qquad R_0=1.2\,A^{1/3}\,\mathrm{fm},\qquad \ell=2,4.$$9Yb, $Q_{40}$00Hf, and neighboring nuclei also emphasizes the structural role of $Q_{40}$01, even though it does not implement explicit GCM mixing. In $Q_{40}$02Hf the global minimum is found at $Q_{40}$03, whereas setting $Q_{40}$04 shifts the quadrupole minimum and raises the total energy by $Q_{40}$05–$Q_{40}$06 MeV. At finite $Q_{40}$07 a pronounced shell gap at $Q_{40}$08, $Q_{40}$09 opens, and the proton $Q_{40}$10 orbital with $Q_{40}$11 shows increased $Q_{40}$12 admixture, with a $Q_{40}$13 component of about $Q_{40}$14 instead of about $Q_{40}$15 at $Q_{40}$16 (Li et al., 15 Jan 2026). This suggests that even without collective superposition, quadrupole–hexadecapole coupling can be read directly from deformation-induced orbital mixing.

4. Boson realizations and spectroscopic fingerprints

In interacting-boson descriptions, quadrupole–hexadecapole mixing is represented by the coexistence of $Q_{40}$17 ($Q_{40}$18), $Q_{40}$19 ($Q_{40}$20), and $Q_{40}$21 ($Q_{40}$22) bosons. A widely used Hamiltonian is

$Q_{40}$23

with

$Q_{40}$24

and a quadrupole operator containing the mixed $Q_{40}$25 terms (Lotina et al., 2024, Lotina et al., 2024). These are the explicit off-diagonal pieces that connect configurations differing by one $Q_{40}$26 and one $Q_{40}$27 boson and thereby mix quadrupole and hexadecapole structures.

In neutron-rich Sm and Gd isotopes, Gogny-D1S mapped $Q_{40}$28-IBM calculations show that the largest differences between $Q_{40}$29-IBM and $Q_{40}$30-IBM occur in transitional nuclei around $Q_{40}$31. In Sm with $Q_{40}$32, ground-state-band states carry $Q_{40}$33, while for $Q_{40}$34 they have $Q_{40}$35. The excited $Q_{40}$36 and $Q_{40}$37 states have $Q_{40}$38–$Q_{40}$39 $Q_{40}$40-boson fractions for $Q_{40}$41, but only $Q_{40}$42–$Q_{40}$43 for $Q_{40}$44. The mapped $Q_{40}$45-IBM improves the description of high-spin yrast states in lighter Sm and Gd and enhances non-yrast $Q_{40}$46 strengths, whereas for $Q_{40}$47 the main residual sensitivity is in the $Q_{40}$48 monopole systematics (Lotina et al., 2024).

A related mapped $Q_{40}$49-IBM study of Nd, Sm, Gd, Dy, and Er near $Q_{40}$50 concludes that the inclusion of the $Q_{40}$51 boson is necessary to improve the $Q_{40}$52 yrast energies in nuclei with $Q_{40}$53 and $Q_{40}$54, near the neutron shell closure. In the well-deformed $Q_{40}$55 and $Q_{40}$56 systems, the model increases the quadrupole transition strengths between yrast states, in better agreement with experiment, while the monopole strengths do not differ significantly from those of the simpler $Q_{40}$57 model (Lotina et al., 2024).

For axially deformed Gd isotopes, the explicit inclusion of the hexadecapole degree of freedom is reported not to affect most low-spin and low-lying states qualitatively, but to improve the description of high-spin states of ground-state bands in nearly spherical vibrational nuclei and to generate $Q_{40}$58 bands with strong $Q_{40}$59 transitions in strongly deformed nuclei. Illustrative values include downward shifts of roughly $Q_{40}$60 keV, $Q_{40}$61 keV, and $Q_{40}$62 keV for the $Q_{40}$63, $Q_{40}$64, and $Q_{40}$65 levels in $Q_{40}$66Gd, and for $Q_{40}$67Gd a $Q_{40}$68 bandhead at about $Q_{40}$69 MeV, close to the experimental $Q_{40}$70 MeV, together with a predicted $Q_{40}$71 W.u. as opposed to $Q_{40}$72 W.u. in the $Q_{40}$73-IBM (Lotina et al., 2023).

Mixed-symmetry hexadecapole states provide a particularly transparent example. In $Q_{40}$74Ru, the $Q_{40}$75-IBM-2 yields for the $Q_{40}$76 state an approximate composition of $Q_{40}$77 $Q_{40}$78 bosons, $Q_{40}$79 $Q_{40}$80 bosons, and $Q_{40}$81 $Q_{40}$82 bosons. The observed strong $Q_{40}$83 transition $Q_{40}$84 is analyzed as $Q_{40}$85 arising from the $Q_{40}$86-boson term and $Q_{40}$87 from the $Q_{40}$88-boson term, while the moderate branch $Q_{40}$89 W.u. is tied to the $Q_{40}$90-boson admixture. The model also predicts $Q_{40}$91 W.u. (Hennig et al., 2015). In this setting, configuration mixing is directly observable in the decomposition of the wave function and in the relative strength of $Q_{40}$92, $Q_{40}$93, and $Q_{40}$94 channels.

5. Surface–volume mixing and the heavy-ion-collision realization

A distinct but conceptually related form of quadrupole–hexadecapole mixing arises when one compares volume multipole moments with surface deformation parameters in Woods–Saxon densities. The volume deformations

$Q_{40}$95

characterize the density throughout the nucleus, whereas the Woods–Saxon surface parameters $Q_{40}$96 enter the angle-dependent radius

$Q_{40}$97

These two notions are not identical for a nucleus that carries both quadrupole and hexadecapole deformation (Ryssens et al., 2023).

For an axially symmetric nucleus with only $Q_{40}$98 and $Q_{40}$99 nonzero, expanding the quadrupole volume moment in the sharp-surface limit gives

$\beta_2$00

The last term is the explicit quadrupole–hexadecapole mixing term: a surface hexadecapole distortion contributes to the volume quadrupole moment (Ryssens et al., 2023). One common misconception is therefore that low-energy tabulated $\beta_2$01 values can be inserted directly as Woods–Saxon surface parameters in high-energy collision simulations.

The case of $\beta_2$02U is the canonical demonstration. Using 21 Skyrme parameter sets in 3D Skyrme-HFB, the quoted ensemble averages are $\beta_2$03 and $\beta_2$04 for the volume multipoles, with corresponding best-fit Woods–Saxon parameters $\beta_2$05 and $\beta_2$06 (Ryssens et al., 2023). In other words, reproducing a true volume $\beta_2$07 requires a smaller surface quadrupole once the hexadecapole deformation is included.

This correction alters the initial conditions used in hydrodynamic simulations of U+U collisions. The older practice of taking $\beta_2$08 and neglecting $\beta_2$09 led to an overestimate of elliptic flow. Using the ratio

$\beta_2$10

the reported estimate is $\beta_2$11 for the old choice, compared with the STAR measurement $\beta_2$12. Replacing it by $\beta_2$13 and $\beta_2$14 yields $\beta_2$15, and full IP-Glasma + MUSIC + UrQMD calculations recover the observed reduction of $\beta_2$16 in central U+U collisions (Ryssens et al., 2023). The same work emphasizes that $\beta_2$17 has virtually no direct effect on $\beta_2$18 fluctuations in central events; its importance is indirect, through the redefinition of the effective surface quadrupole.

6. Magnetic quadrupole–hexadecapole mixing in odd-parity antiferromagnets

In BaMn$\beta_2$19As$\beta_2$20 and Ba$\beta_2$21K$\beta_2$22Mn$\beta_2$23As$\beta_2$24, the relevant order is not a mass deformation but an odd-parity magnetic multipole. The magnetic quadrupole and hexadecapole operators in the $\beta_2$25 representation are given in cubic-harmonic form as

$\beta_2$26

and

$\beta_2$27

The ground state of semiconducting BaMn$\beta_2$28As$\beta_2$29 is identified as a magnetic hexadecapole ordered state, and microscopic analysis shows ferroic ordering of a leading magnetic hexadecapole moment together with an admixed magnetic quadrupole moment (Watanabe et al., 2017).

In the metallic regime the admixture becomes explicit. LS coupling induces a nonzero $\beta_2$30 while the leading order remains $\beta_2$31, and the effective single-band Hamiltonian is built on two Mn sublattices with odd-parity antisymmetric spin–orbit coupling and opposite antiferromagnetic molecular fields on the two sublattices (Watanabe et al., 2017). In Landau language, once the hexadecapole condenses, the linear coupling term $\beta_2$32 generically induces the quadrupole component as well.

The mixed order produces characteristic electromagnetic and transport responses. By symmetry, the uniform magnetoelectric tensor has only $\beta_2$33 in the hexadecapole state; in the insulating case the reported magnitude is $\beta_2$34, enhanced in the metal. The antiferromagnetic Edelstein response is expressed as

$\beta_2$35

with an intraband contribution proportional to the relaxation time $\beta_2$36. A further consequence is current-induced nematic order: tetrahedral Fermi-surface warping proportional to $\beta_2$37 implies that a current along the $\beta_2$38 axis induces $\beta_2$39-plane nematicity, a $\beta_2$40 coupling of $\beta_2$41 symmetry, described as magnetopiezoelectricity (Watanabe et al., 2017).

The proposed experimental signatures are correspondingly diverse: neutron scattering should detect the $\beta_2$42 spin-density anisotropy associated with the primary hexadecapole form factor, with the small quadrupole admixture shifting intensity between $\beta_2$43 reflections; an out-of-plane current should drive an in-plane shear strain observable by X-ray or ultrasonic probes; $\beta_2$44 may be accessed by polarized optics or dielectric tuning under magnetic field; and a basal-plane current should generate a staggered sublattice magnetization that may be detectable by XMCD or second-harmonic magneto-optics (Watanabe et al., 2017). In this condensed-matter realization, quadrupole–hexadecapole mixing is not a secondary analogy to nuclear collectivity but a symmetry-equivalent instance of the same general principle: rank-2 and rank-4 multipoles of the same symmetry class do not remain independent once the microscopic Hamiltonian permits their coupling.