---
title: Angular Dipole Deformation in Nuclei & Atoms
url: https://www.emergentmind.com/topics/angular-dipole-deformation
type: topic
---

# Angular Dipole Deformation in Nuclei & Atoms

Angular dipole deformation denotes deformation- or orientation-induced anisotropy of dipole response or of dipole-like collective variables. In the nuclear setting treated most explicitly in the axially deformed proton-neutron relativistic quasiparticle random-phase approximation, it is the response of \(L=1\) spin-isospin modes to axial quadrupole deformation: the spherical \(2J+1\) degeneracy is lifted, only the projection \(K\) of angular momentum on the symmetry axis remains a good quantum number, and the \(J=1\) sector splits into \(K=0\) and \(K=\pm1\), with consequent centroid shifts, \(K\)-dependent collectivity, and fragmentation of the strength function [2404.13266]. Closely related usages appear in isovector electric dipole response, magnetic-rotational dipole bands, anisotropic dipole-dipole interactions, beyond-dipole high-harmonic generation, and gauge-invariant \(l=1\) perturbation theory, where reduced symmetry or directional structure resolves a formerly isotropic dipole degree of freedom into distinct angular components [1612.06019][1201.4757][1504.00301][2503.14982][2102.10650].

## 1. Symmetry reduction and the \(K\)-resolved dipole

The central mechanism is symmetry breaking. In a spherical nucleus, dipole excitations are classified by total angular momentum \(J\) and magnetic quantum number \(M\), and the familiar \(2J+1\) degeneracy holds. In an axially deformed nucleus, spherical symmetry is broken and only the projection \(K\) on the symmetry axis remains conserved. For charge-exchange modes this yields the selection rules
\[
K=\Omega_p-\Omega_n,\qquad \pi=\pi_p\times \pi_n,
\]
so the response separates into \(K^\pi\) blocks rather than \(J^\pi\) blocks. For \(J=1\) modes, the spherical triplet becomes \(K=0\) and \(K=\pm1\), with the \(\pm1\) branches degenerate under time reversal. In the deformed Gamow-Teller case,
\[
S(\mathrm{GT}^{\pm},E)=S_{K=0}(E)+2\,S_{K=1}(E),
\]
which makes the \(K\)-decomposition explicit [2404.13266].

The same kinematic idea governs the isovector electric dipole response in an axially symmetric weakly bound nucleus. There the good quantum number is again \(K\), now interpreted as the projection of the dipole multipolarity \(\lambda=1\). The \(K=0\) branch is longitudinal, polarized along the symmetry axis, whereas \(|K|=1\) denotes the two degenerate transverse polarizations. Angular dipole deformation therefore appears as different centroid energies, line shapes, and transition-current topologies for longitudinal and transverse dipole motion [1612.06019].

This suggests a broad but coherent classification. In deformed nuclear spectroscopy, angular dipole deformation usually means \(K\)-splitting and redistribution of strength. In magnetic rotation it refers to the angular orientation of a magnetic dipole vector relative to the total spin. In AMO settings it often denotes anisotropic dipole-dipole coupling or the deflection of an induced dipole away from a principal polarization axis. In gravitational perturbation theory the same language is extended to \(l=1\) axial or mass-dipole sectors, where angular structure encodes spin or center-of-mass dipole charges [1201.4757][1504.00301][2503.14982][2102.10650].

## 2. Axially deformed pnRQRPA and the microscopic treatment of \(L=1\) modes

A fully covariant microscopic realization is provided by the axially deformed relativistic Hartree-Bogoliubov plus proton-neutron relativistic quasiparticle random-phase approximation framework. The ground state is obtained with the axially deformed relativistic Hartree-Bogoliubov model using relativistic density-dependent point-coupling energy density functionals, specifically DD-PC1 and DD-PCX. The excitations are then constructed in an axially deformed pnRQRPA formulated in linear response, so that axial symmetry reduces the problem to separate \(K^\pi\) blocks and enables explicit strength decomposition by \(K\) [2404.13266].

The external charge-exchange operators are the Fermi operator,
\[
\hat{O}_F=\sum_i \tau_\pm(i),
\]
the Gamow-Teller operator,
\[
\hat{O}_{GT}=\sum_i \boldsymbol{\sigma}(i)\tau_\pm(i),
\]
and the spin-dipole family,
\[
\hat{O}_{SD}^{J}=\sum_i r_i\,[Y_1(\hat r_i)\otimes \boldsymbol{\sigma}(i)]_J\,\tau_\pm(i),\qquad J=0,1,2.
\]
The associated strength function is
\[
S(E)=\sum_\nu \big|\langle \nu|\hat O|0\rangle\big|^2\delta(E-E_\nu),
\]
and in axial symmetry the operator and response are resolved into \(K\)-components evaluated with axially deformed Dirac spinors [2404.13266].

Self-consistency is enforced at the level of both particle-hole and particle-particle channels. The particle-hole interaction contains the isovector-vector term inherited from the ground-state functional and an isovector-pseudovector pion-like term added only in the residual interaction; its Landau-Migdal parameter is adjusted to reproduce Gamow-Teller centroids, with \(g_0=0.734\) for DD-PC1 and \(0.621\) for DD-PCX. The pairing sector uses a separable finite-range interaction in both isovector \(T=1,S=0\) and isoscalar \(T=0,S=1\) channels, and the application to unnatural-parity transitions sets the isoscalar pairing scale to \(V_0^{pp}=1.0\) [2404.13266].

Computationally, the formal pnRQRPA amplitudes satisfy the standard QRPA matrix equation in each \(K^\pi\) block, but the practical solver avoids explicit diagonalization of the full two-quasiparticle space. Instead it solves the Bethe-Salpeter equation for the response matrix in a reduced space of separable interaction channels. This substantially lowers the cost, and singular value decomposition further reduces the interaction-channel dimension without loss of accuracy. The axial harmonic-oscillator basis converges around \(N_{\mathrm{osc}}\approx 12\) for \(A\approx 70\) nuclei, while production calculations use \(N_{\mathrm{osc}}=16\); strength distributions are smeared with \(\eta=1\) MeV, and anti-particle contributions in the charge-exchange channel are neglected as negligible [2404.13266].

Benchmarking in the spherical limit is an essential part of the construction. For doubly magic \(^{28}\)O, the deformed solver constrained to spherical shape reproduces spherical pnRQRPA results for the isobaric analog state and GT\(^-\) response, including the expected \(K=0\) and \(K=\pm1\) degeneracy and the factor-of-three relation between total and \(K\)-resolved Gamow-Teller strength [2404.13266].

## 3. Spin-isospin angular dipole deformation in deformed nuclei

The resulting phenomenology is sharply mode dependent. Fermi transitions are almost shape independent because only \(K=0\) contributes and the operator carries isospin but no spin or coordinate dependence. In \(^{56\text{--}62}\)Fe the isobaric analog state energy differs by at most \(0.05\) MeV between oblate, spherical, and prolate shapes. For \(^{56}\)Fe, the experimental IAS centroid is around \(8.9\) MeV, about \(0.5\) MeV above calculated values, and DD-PCX shifts the theory upward by about \(0.13\) MeV [2404.13266].

Gamow-Teller transitions are markedly more sensitive. In \(^{60}\)Ni GT\(^+\), the spherical strength is a single peak at \(\omega\approx 2.3\) MeV. Constraining the nucleus to the oblate minimum at \(\beta_2\approx-0.19\) produces broad fragmentation, whereas the prolate shape at \(\beta_2\approx+0.13\) generates two peaks near \(\omega\approx 1.7\) MeV and \(3\) MeV; with \(\eta=1\) MeV these merge into a broadened resonance, and with \(\eta=0.5\) MeV they separate clearly. Across the deformed ground states of \(^{60,62,64}\)Ni and \(^{64}\)Zn, which are oblate, and \(^{56}\)Fe and \(^{46}\)Ti, which are prolate, the deformed pnRQRPA strength functions display fragmented main peaks of reduced height and agree better with measured GT\(^+\) distributions than spherical calculations. DD-PCX generally shifts centroids a few hundred keV higher than DD-PC1 [2404.13266].

The \(K\)-resolved GT\(^-\) pattern in the Fe chain shows a robust correlation with the sign of \(\beta_2\). For prolate shapes, low-lying GT\(^-\) strength is dominated by \(K=0\), whereas the giant GT resonance region is dominated by \(K=1\). For oblate shapes the trend is reversed: \(K=0\) is pushed to lower energies and \(K=1\) to higher energies, so the high-energy giant-resonance region carries more \(K=0\) content. The magnitude of the \(K\)-splitting grows with \(|\beta_2|\). The mechanism is explicitly traced to Nilsson-level splitting and mixing, the increased two-quasiparticle level density in a deformed mean field, and pairing-induced changes in the coherence of low-lying configurations; the TV and TPV residual interactions set collectivity and centroid energies, while isoscalar pairing modifies low-lying GT strength in the unnatural-parity channel [2404.13266].

Spin-dipole transitions exhibit a related but more structured behavior because they combine \(\Delta L=1\), \(\Delta S=1\), and explicit radial dependence. In \(^{58}\)Fe, the centroid energies \(E_{\mathrm{cent}}=m_1/m_0\) satisfy the ordering
\[
E_{\mathrm{cent}}(2^-)<E_{\mathrm{cent}}(1^-)<E_{\mathrm{cent}}(0^-),
\]
with the oblate values \(22.04\), \(27.07\), and \(28.33\) MeV, respectively. This ordering persists across shapes and isotopes. With increasing neutron number from \(^{58}\)Fe to \(^{60}\)Fe to \(^{62}\)Fe, all three spin-dipole centroids move to lower energy and the total strengths increase. The \(K\)-decomposition is again shape dependent: \(0^-\) carries only \(K=0\); \(1^-\) contains \(K=0\) and \(K=1\), with the relative ordering of their centroids reversing between prolate and oblate \(^{58}\)Fe; \(2^-\) contains \(K=0,1,2\), with \(K=2\) dominating near \(\omega\approx 30\) MeV in prolate \(^{58}\)Fe but controlling low-lying strength near \(\omega\approx 10\) MeV in the oblate shape. Within the deformed pnRQRPA, the non-energy-weighted SD sum rule \(m_0\) is preserved, consistent with the self-consistent residual interaction [2404.13266].

These redistributions have direct astrophysical implications. Deformation-induced fragmentation modifies the low-energy GT\(^-\) and SD\(^-\) strength available within the \(Q_\beta\) window and therefore can change beta-decay half-lives. The improved match to GT\(^+\) strength in deformed \(pf\)-shell nuclei indicates that deformation is needed for reliable electron-capture rates. The deformation dependence of SD modes also affects first-forbidden contributions relevant to neutrino-nucleus reactions and hot astrophysical environments [2404.13266].

## 4. Electric dipole anisotropy, giant resonances, and shape diagnostics

The same angular logic governs the isovector electric dipole response, but here weak binding and surface diffuseness can substantially alter the deformation pattern. In \(^{40}\)Mg, a fully self-consistent continuum finite-amplitude QRPA on a large deformed coordinate-space mesh was used to study both prolate and oblate shapes. The prolate ground state has \(\beta_2=0.39\), while an oblate local minimum at \(\beta_2=-0.32\) lies about \(1.9\) MeV higher. The \(K\)-resolved strengths show a conventional giant dipole splitting of about \(5\) MeV for both shapes, consistent with the near-linear hydrodynamic expectation. The pygmy dipole resonance behaves differently: in the prolate case the lowest \(K=0\) branch appears at \(E\approx 2.5\) MeV and the \(|K|=1\) branch is centered around \(4.0\text{--}4.5\) MeV, giving \(\Delta E_{\mathrm{PDR}}\approx 1.4\) MeV, larger than the simple hydrodynamic estimate of about \(0.95\) MeV; in the oblate case the \(|K|=1\) component dominates the low-energy strength and the splitting is compressed to about \(0.45\) MeV, much smaller than the estimate of about \(1.05\) MeV. Transition-current analysis identifies the lowest prolate \(K=0\) pygmy mode as a compressional core-halo oscillation with a single nodal boundary around \(12\) fm, whereas higher-energy PDR and GDR branches display progressively more complicated nodal and circulation-like patterns. Accurate angular decomposition requires a very large box, with \(27.6\) fm found necessary; smaller boxes generate false peaks and distort the surface flow [1612.06019].

A common misconception is that deformation should always appear as a clean double-hump IVGDR in well-deformed nuclei. Proton inelastic scattering at \(200\) MeV and \(0^\circ\) on \(^{144,146,148,150}\)Nd and \(^{152}\)Sm shows a more nuanced behavior. Along the chain the adopted axial quadrupole deformation increases from \(0.13\) in \(^{144}\)Nd to \(0.31\) in \(^{152}\)Sm. In the well-deformed nuclei \(^{150}\)Nd and \(^{152}\)Sm, the extracted photoabsorption cross sections show a pronounced asymmetry rather than the distinct double-hump structure expected from simple \(K\)-splitting. The interpretation advanced in the self-consistent SLy6 Skyrme separable RPA analysis is that these nuclei lie near the critical point of the phase transition from vibrators to rotors and possess a soft quadrupole deformation potential; shape fluctuations smear the \(K=0\) and \(K=1\) branches and reduce the distinctness of the low-energy component [1612.06819].

Polarized \(\gamma\)-decay provides an even more differential probe of angular dipole deformation. In \(^{154}\)Sm, linearly polarized quasimonochromatic photon beams were used to disentangle elastic scattering and \(2_1^+\)-Smekal-Raman scattering from the isovector giant dipole resonance through their distinct angular distributions,
\[
W_{0^+\to 1^-\to 0^+}^{(\mathrm{ES})}(\theta,\phi)
= \frac{3}{4}\left(1+\cos^2\theta-\sin^2\theta\cos 2\phi\right),
\]
\[
W_{0^+\to 1^-\to 2^+}^{(2_1^+\text{-SRS})}(\theta,\phi)
= \frac{3}{40}\left(13+\cos^2\theta-\sin^2\theta\cos 2\phi\right).
\]
A simultaneous fit to \(\gamma\)-decay branching ratios and photoabsorption data yielded three IVGDR components and extracted the ground-state deformation parameters
\[
\beta = 0.2925 \pm 0.0025,\qquad \gamma = 5.0^\circ \pm 1.5^\circ.
\]
Photoabsorption alone gave a flat \(\gamma\) posterior with a \(99.7\%\) credible upper limit \(\gamma\le 28.6^\circ\), whereas inclusion of the \(\gamma\)-decay observables reduced that upper limit to \(8.3^\circ\). Here angular dipole deformation is not only the splitting of dipole eigenmodes along principal axes, but also the anisotropic emission pattern of their decay amplitudes [2406.19695].

At finite temperature and angular momentum, deformation can also be inferred indirectly from the giant dipole width. Pandit and collaborators established a non-linear, mass-dependent relation for \(A\approx 59\text{--}208\),
\[
\beta_{\mathrm{exp}}
=0.18+\beta_{\mathrm{GDR}}
+0.7\left(\frac{\Gamma(J,T,A)-\Gamma_0}{E_0}\right)^{D(A)},
\]
with
\[
\beta_{\mathrm{GDR}}=0.04+\frac{4.13}{A},\qquad
D(A)=2-0.0036A,
\]
\[
E_0=31.2\,A^{-1/3}+20.6\,A^{-1/6},\qquad
\Gamma_0=0.05\,E_0^{1.6}.
\]
The subtraction of \(\beta_{\mathrm{GDR}}\) encodes the GDR-induced quadrupole moment and explains why the width-deformation relation is not simply linear. The correlation holds for \(T\ge T_c\simeq 0.7+37.5/A\) MeV and below the approximate Jacobi-transition scale \(J_c\sim 1.2A^{5/6}\) [1305.3360].

## 5. Intrinsic dipoles, rotation, and parity mixing

In magnetic-rotational spectroscopy, angular dipole deformation refers less to collective \(K\)-splitting than to the directional geometry of a magnetic dipole built from high-\(j\) quasiparticles. In \(^{112}\)In, three dipole bands were observed up to \(E_x\approx 5.64\) MeV and \(I\approx 20\hbar\). The positive-parity band A shows a monotonic decrease of \(B(M1)\) with spin, from \(3.28(14)\,\mu_N^2\) at \(15^+\) to \(1.89^{+0.11}_{-0.10}\,\mu_N^2\) at \(16^+\), \(1.52^{+0.20}_{-0.14}\,\mu_N^2\) at \(17^+\), and \(>0.63\,\mu_N^2\) at \(18^+\). Tilted-axis cranking assigns this band the configuration \(\pi g_{9/2}^{-1}\otimes \nu(d_{5/2}/g_{7/2})(h_{11/2})^2\), with a favored static solution near \(\epsilon_2=0.08\), \(\gamma=5^\circ\), corresponding to \(\beta_2\approx 0.076\), and an almost constant tilt angle of the rotational axis near \(55^\circ\). Band C is assigned \(\pi g_{9/2}^{-1}\otimes \nu(h_{11/2})^3\), with \(\epsilon_2=0.13\) and \(\gamma=0^\circ\). The shears interpretation uses
\[
\cos\theta=\frac{J^2-j_\pi^2-j_\nu^2}{2j_\pi j_\nu},
\]
so the decrease of \(B(M1)\) tracks the closing of the proton and neutron blades and the concomitant reduction of the transverse magnetic moment. The inferred intrinsic quadrupole moments remain small, \(Q_0\approx 1.0\) eb for band A and \(Q_0\approx 1.6\) eb for band C, which is why the angular, rather than strongly collective quadrupole, character dominates the dipole bands [1201.4757].

A different but related body-fixed dipole appears in octupole-deformed nuclei. For reflection-asymmetric, axially symmetric shapes, the surface is expanded as
\[
R(\theta)=c_V(\beta)R_0\left(1+\sum_{l\ge 1}\beta_lY_{l0}(\theta)\right),
\]
and a nonzero octupole parameter \(\beta_3\) induces a collective intrinsic electric dipole moment aligned with the symmetry axis. To leading order in \(\beta_2\beta_3\), the intrinsic electric dipole and Schiff moments are
\[
d_{\mathrm{int}}
= AZ\frac{e^3}{C}\,\frac{3}{40\pi}
\sum_{l\ge 2}
\frac{(l^2-1)(8l+9)}{\big[(2l+1)(2l+3)\big]^{3/2}}\,
\beta_l\beta_{l+1},
\]
\[
S_{\mathrm{int}}
\approx eZR_0^3\,\frac{3}{20\pi}
\sum_{l\ge 2}
\frac{(l+1)\beta_l\beta_{l+1}}{\sqrt{(2l+1)(2l+3)}}.
\]
In a laboratory eigenstate of good parity and angular momentum, rotational averaging makes these expectation values vanish. They reappear only when \(T,P\)-violating interactions mix opposite-parity rotational doublets, with the laboratory moments
\[
d_I\simeq 2\alpha\,\frac{I}{I+1}\,d_{\mathrm{int}},\qquad
S_I\simeq 2\alpha\,\frac{I}{I+1}\,S_{\mathrm{int}}.
\]
For \(^{225}\)Ra the tabulated values are \(d_{\mathrm{int}}=0.14\,e\,\mathrm{fm}\), \(S_{\mathrm{int}}=11\,e\,\mathrm{fm}^3\), \(\beta_3=0.088\), \(d_{\mathrm{lab}}=0.35\times 10^{-2}\,\bar\theta\,e\,\mathrm{fm}\), and \(S_{\mathrm{lab}}=0.28\,\bar\theta\,e\,\mathrm{fm}^3\). The quoted uncertainty of the estimates may exceed a factor of \(10\), but the formal point is clear: angular dipole deformation here is the alignment problem of a collective intrinsic dipole with respect to the spin axis under parity mixing [2411.18943].

## 6. Atomic, optical, and relativistic extensions

In Rydberg physics, angular dipole deformation becomes the anisotropic reshaping of the interaction graph by the angle between the internuclear axis and the quantization axis. At a Stark-tuned Förster resonance between two individual \(^{87}\)Rb atoms, the isolated exchange channel obeys
\[
J(\theta)=\frac{C_3}{R^3}\,\frac{1-3\cos^2\theta}{2},
\]
with a magic angle
\[
\theta_m=\arccos\!\left(\frac{1}{\sqrt 3}\right)\approx 54.735^\circ
\]
where the resonant coupling vanishes. For \(R=9.1\,\mu\mathrm m\), the measured oscillation frequency at \(\theta=0^\circ\) is \(\nu\approx 8.6\pm 0.4\) MHz, minimal oscillations are seen near \(57^\circ\), and the \(\theta=90^\circ\) frequency is reduced consistently with the \((1-3\cos^2\theta)\) form. In many-atom simulations of resonant \(s\leftrightarrow p\) transport, the same anisotropy enters through
\[
f_0(\theta)=\frac{1-3\cos^2\theta}{2},\qquad
f_2(\theta)=\frac{3\sin^2\theta}{2},
\]
leading to faster propagation for transverse fields than for axial fields and to a calculated localization length of \(6.3\,\mu\mathrm m\) in one representative geometry [1504.00301][1510.07914].

In strong-field and structured-light physics, the phrase denotes a dynamical rotation of the driven electronic dipole away from the principal polarization axis. Beyond the electric-dipole approximation, a linearly polarized OAM Bessel beam is treated with the full minimal-coupling Hamiltonian,
\[
H=\frac{(\mathbf p+e\mathbf A(\mathbf r,t))^2}{2m_e}+e\phi(\mathbf r,t)+V_{\mathrm{nuc}}(\mathbf r),
\]
so that spatial gradients and magnetic fields induce nonzero transverse dipole components. A practical deformation angle is defined by
\[
\theta_d(t)=\tan^{-1}\!\left(\frac{|a_\perp(t)|}{|a_\parallel(t)|}\right),
\qquad
a_\perp=\sqrt{a_y^2+a_z^2}.
\]
Using the reported off-axis dipole amplitudes at the hot spot, one infers \(\theta_d\sim 10^{-3}\) rad \(\approx 0.057^\circ\). Despite this small time-domain deflection, the spectral consequences are large: the second-harmonic emission along \(y\) is approximately doubled relative to a matched plane wave, and the beyond-dipole correction reaches about \(160\%\) in amplitude, while the third-harmonic correction remains about \(0.07\%\). Angular dipole deformation is therefore observed through symmetry breaking, transverse emission, and the appearance of even harmonics [2503.14982].

In fast charged-particle scattering from atoms, the angular distribution of secondary electrons relative to the momentum-transfer direction \(\hat{\mathbf q}\) departs from the familiar photoionization dipole pattern because the transition operator contains a full multipole expansion,
\[
e^{i\mathbf q\cdot \mathbf r}
=
\sum_{L=0}^\infty
(4\pi i^L)\,j_L(qr)
\sum_{M=-L}^L
Y_{LM}^*(\hat{\mathbf q})Y_{LM}(\hat{\mathbf r}).
\]
The differential generalized oscillator strength is expanded in Legendre polynomials,
\[
\frac{dF_{nl}(q,\omega)}{d\Omega_e}
=
\frac{F_{nl}(q,\omega)}{4\pi}
\Big[
1+\beta_{nl}^{(\mathrm{in})}P_2
+q\,\gamma_{nl}^{(\mathrm{in})}P_1
+q\,\eta_{nl}^{(\mathrm{in})}P_3
+s_{nl}^{(\mathrm{in})}P_4+\cdots
\Big].
\]
For \(s\)-subshells in the optical limit, the scattering pattern is \(1+2P_2\), not the photoionization form \(1-P_2\), and the non-dipole terms scale with \(qR\) rather than \(\omega/c\). Quadrupole contributions are therefore much less suppressed than in photoionization. Calculations for He, Ar, and Xe, including the collectivized Xe \(4d^{10}\) subshell, show prominent deviations even at very small \(q\) [1012.5465].

Relativistic perturbation theory extends the terminology further. In Nakamura’s gauge-invariant any-order treatment of perturbations on Schwarzschild spacetime, angular dipole deformation is the \(l=1\) odd-parity sector. In vacuum and the stationary limit it yields
\[
g_{t\phi}=-2J\,r^{-1}\sin^2\theta,
\]
equivalently \(g_{t\phi}=-r^2\sin^2\theta\,\omega(r)\) with \(\omega(r)=2J/r^3\), reproducing the slow-rotation limit of Kerr. In the eikonal framework for classical gravitational scattering, the same language is applied to changes in the system’s spatial angular momentum and center-of-mass dipole charges. The ambiguity-free center-of-mass dipole is
\[
M_i \equiv J_{i0}-\int dt\, t\,\frac{dP_i}{dt},
\]
which removes recoil drift and Coulombic ambiguities from the boost component \(J_{i0}\). Here angular dipole deformation no longer refers to a material dipole oscillation, but to the \(l=1\) charge content of the gravitational field and its radiative loss [2102.10650][2406.03937].

Across these domains, the unifying structure is the same. A system that is isotropic or effectively dipolar in a high-symmetry limit acquires direction-dependent branches, couplings, or observables once symmetry is reduced by deformation, orientation, gradients, or rotation. In nuclear structure this appears most canonically as \(K\)-splitting and strength fragmentation of \(L=1\) modes; in neighboring fields it persists as anisotropic polarization, angularly selective coupling, or \(l=1\) charge transport.

Source: https://www.emergentmind.com/topics/angular-dipole-deformation