Spherical Neutron Polarimetry

• Spherical neutron polarimetry is a technique that measures the complete 3×3 polarization transfer matrix, enabling precise analysis of non-collinear and chiral magnetic structures.
• It utilizes advanced zero-field setups with devices like Cryopad and supermirror optics to control and analyze neutron spin states across all spatial directions.
• The method provides actionable insights into magnetic ordering, magnetoelectric effects, and time-reversal symmetry violations, crucial for identifying multipolar phenomena.

Spherical neutron polarimetry (SNP) is a technique that determines the full polarization transfer properties of neutron scattering by systematically measuring how an incident neutron polarization along any of three orthogonal spatial directions is transformed into an outgoing polarization along any direction after scattering. By enabling access to the $3 \times 3$ polarization transfer matrix, SNP provides unrivaled sensitivity to non-collinear, chiral, and multipolar magnetic structures, as well as to subtle magnetoelectric effects and time-reversal symmetry violation phenomena.

1. Fundamental Principles and Theoretical Framework

SNP extends beyond conventional longitudinal polarization analysis, which measures only spin-flip (SF) and non-spin-flip (NSF) intensities for a single quantization axis, yielding a maximum of four cross-sections and access to only two independent polarization directions. In SNP, the incident neutron polarization is systematically prepared along the $x$, $y$, or $z$ axes (Blume–Maleev frame), and the outgoing polarization is analyzed along each of these axes after interaction with the sample in a zero magnetic field. Measurements for all nine incident–outgoing ($i \rightarrow j$) polarization combinations, and for both spin states, yield up to 36 polarization-resolved intensities and enable full reconstruction of the $3 \times 3$ polarization transfer matrix $P_{ij}$: $$(P_{\rm out})_i = \sum_{j=x,y,z} P_{ij} (P_{\rm in})_j\,.$$ This matrix encodes both the non-spin-flip and spin-flip processes and all off-diagonal (chiral, non-collinear) correlations. The measured polarization transfer, in the absence of external fields at the sample, is uniquely sensitive to the quantum mechanical scattering process and sample correlations (Gorkov et al., 22 Apr 2025, Babkevich et al., 2017, Urru et al., 2022, Qureshi, 2018).

The cross-section for elastic neutron scattering in the Born–Fermi approximation combines nuclear and magnetic contributions: $$\frac{d^2\sigma_{\alpha\to\beta}}{d\Omega\,dE'} = \frac{k_f}{k_i} \left|\langle \beta | N(\mathbf{Q}) + \mathbf{M}_\perp(\mathbf{Q}) \cdot \boldsymbol{\sigma} | \alpha \rangle \right|^2 \delta(E_i - E_f - E)$$ with $N(\mathbf{Q})$ the nuclear structure factor and $\mathbf{M}_\perp(\mathbf{Q})$ the magnetic interaction vector, perpendicular to $\mathbf{Q}$.

In the spherical tensor formalism, the full elastic scattering amplitude is expressed as a sum over irreducible tensor contributions between the neutron spin $\boldsymbol{\sigma}$ and target polarization $\mathbf{I}$, allowing for extraction of time-reversal or parity-violating terms; e.g., the coefficient $D'$ of the T-odd triple correlation $$\boldsymbol{\sigma}\cdot[\mathbf{k}\times\mathbf{I}]$$ is directly accessible by SNP configurations (Gudkov et al., 2019).

2. Experimental Realization: Zero-Field Polarimetry and Instrumentation

Critical to SNP is the maintenance of a genuinely zero external magnetic field at the sample position to exclude spurious precession or depolarization effects. Modern implementations, such as on the KOMPASS spectrometer (Gorkov et al., 22 Apr 2025), use a Cryopad device—a double Meissner-shielded zero-field cavity (superconducting mu-metal or Nb screens) yielding residual fields $\lesssim 10^{-4}$ mT—combined with dual adiabatically rotating guide field coils ("nutators"). The nutators independently control the incoming and outgoing neutron quantization axes, enabling total access to any pair of $(i,j)$ basis directions.

The beam polarization is set by a chain of supermirror (SM) V-cavities or, in some applications, a Heusler (Cu$_2$MnAl) (111) single-crystal analyzer. On KOMPASS, three stacked SM V-cavities (each 1.75 m, Fe/Si $m\sim4.2$) yield $P_{\rm in}>0.99$ for $2 \unicode{x212B}\leq\lambda\leq 6 \unicode{x212B}$ and $\sim25\%$ overall flux transmission. The outgoing polarization is analyzed either by a high-efficiency SM analyzer (15 channels, $>96\%$ for $E_f\leq15$ meV) or by a Heusler analyzer; both can be operated in tandem for precision or dynamic range (Gorkov et al., 22 Apr 2025, Babkevich et al., 2017).

In all cases, control of collimation, beam divergence, and energy transfer settings allows flexible optimization for either flux-limited or resolution-limited experiments. Full-polarized and half-polarized (longitudinal only) modes are accessible, as is the exchange of beamline optics for elastic diffraction or integrated intensity experiments.

3. Mathematical Description: Structure Factors and Polarization Matrices

The analysis of SNP data reduces to accurate computation and refinement of nuclear and magnetic structure factors, as well as the polarization matrix $P_{ij}(\mathbf{Q})$ for each measured Bragg reflection. The nuclear structure factor is

$$N(\mathbf{Q}) = \sum_{j} b_j\,e^{i\mathbf{Q} \cdot \mathbf{r}_j}$$

while the magnetic structure factor is calculated body-by-body using atomic spin and orbital angular momenta, symmetry operators, form factors (including multipole expansions if needed), and finally projected to the direction perpendicular to $\mathbf{Q}$: $$\mathbf{M}_\perp(\mathbf{Q}) = \hat{\mathbf{Q}} \times [\mathbf{M}(\mathbf{Q}) \times \hat{\mathbf{Q}}].$$ For a given incident and detected polarization, the intensity and thus spin-flip and non-spin-flip cross-sections are determined by interference between these terms (Qureshi, 2018, Urru et al., 2022).

The polarization matrix is obtained as: $$P_{ij} = \frac{I_{ij}^{++} - I_{ij}^{+-} - I_{ij}^{-+} + I_{ij}^{--}} {I_{ij}^{++} + I_{ij}^{+-} + I_{ij}^{-+} + I_{ij}^{--}}$$ where $I_{ij}^{\sigma_{\rm in}\,\sigma_{\rm out}}$ are intensities for incident spin $\sigma_{\rm in}$ along $j$ and outgoing spin $\sigma_{\rm out}$ along $i$. The complete $P_{ij}$ encodes all non-spin-flip, spin-flip and chiral correlations required for full structure determination and symmetry analysis (Gorkov et al., 22 Apr 2025, Qureshi, 2018).

4. Data Analysis, Refinement, and Computational Tools

Modern analysis of SNP data employs least-squares refinement frameworks that handle not only the observed SNP matrices but also integrated intensities and flipping ratios, with allowance for structural twins and magnetic domains. Mag2Pol (Qureshi, 2018) is a primary tool for this analysis. It constructs structure factors using explicit crystallographic and magnetic symmetry, supports multipole (beyond dipole) expansions of the magnetization density, and computes both $P_{ij}$ and the flipping ratio $R$.

Refinable parameters include atomic Fourier amplitudes, magnetic basis-vector coefficients, domain populations, extinction corrections, and (for full joint refinements) multipole coefficients $C_l^m$. The minimization target is typically a weighted sum of the squared residuals for integrated intensities and SNP matrix elements,

$$\chi^2_w = w\,\chi^2_{\rm Int} + (1-w)\,\chi^2_{\rm Mat}$$

with proper error estimates from the covariance matrix (Qureshi, 2018). Correction factors for finite beam polarization and analyzer efficiency, as well as statistical and systematic error propagation, are standard in all sophisticated workflows.

5. Applications: Magnetic Structure, Chiral and Multipolar Order

SNP is uniquely suited for magnetic systems with noncollinear, chiral, or multipolar arrangements. In Ba(TiO)Cu$_4$(PO$_4$)$_4$, SNP enabled discrimination between competing $\Gamma_3$ representations, unequivocally identifying a nearly perpendicular "two-in, two-out" ground state stabilized by strong Dzyaloshinskii–Moriya interaction, with chirality and domain populations directly resolved via off-diagonal $P_{ij}$ elements. The resulting magnetic space group permits net quadrupolar but not dipolar order (Babkevich et al., 2017).

In CuO and other antiferromagnetic systems, SNP in conjunction with first-principles calculations allows not only determination of magnetic dipole moments, but direct detection and quantification of magnetoelectric (ME) multipoles (toroidal and quadrupole moments). Ab initio DFT+U computations generate local density matrices, which are decomposed into spherical tensor multipoles. Comparison of measured $P_{ij}(\mathbf{Q})$ matrices against models including or omitting ME contributions yields statistically robust identification of long-range-ordered ME multipoles and their propagation vectors (Urru et al., 2022).

A summary table highlighting example SNP applications:

Material/System	SNP Outcome	Key Feature Resolved
Ba(TiO)Cu$_4$(PO$_4$)$_4$	Noncollinear "two-in, two-out" structure	Chiral, quadrupolar order
CuO	Quantitative fit to DFT-predicted $P_{ij}$	Spontaneous ME multipoles
General magnetics	Full $3 \times 3$ $P$-matrix	Chiral/multipolar discrimination

6. Advanced Methodologies: Tensor Analysis, T-violation, and Multipoles

Advanced applications utilize the irreducible spherical tensor expansion of the neutron–target spin coupling to access higher-rank polarization observables and probe time-reversal or parity violation in neutron scattering. Amplitudes such as $A', B', C', D', H', K'$ encode all possible spin–momentum–target correlations; T-odd observables (e.g., the triple product $$\boldsymbol{\sigma}\cdot[\mathbf{k}\times\mathbf{I}]$$) can be systematically projected by appropriate SNP experimental geometry. This approach enables the search for nonstandard interaction channels and symmetry-breaking effects (Gudkov et al., 2019).

The multipole expansion of the magnetic scattering amplitude explicitly captures ME multipoles via $L=2$ (parity-odd, time-odd) contributions, with SNP providing sensitivity to off-diagonal polarization transfer signals that would be strictly forbidden in dipole-only models (Urru et al., 2022).

7. Experimental Strategies, Resolution Effects, and Optimization

High-precision SNP requires careful management of all experimental parameters. Optimization includes selection of guide and focusing optics (e.g., straight vs. parabolic neutron guides, static vs. exchangeable collimators), control of analyzer–detector distance, and trade-offs between $Q$-resolution and intensity. Resolution functions for polarized triple-axis spectrometers are modeled as multivariate Gaussians in $(Q_x, Q_y, Q_z, E)$, with the resolution matrix accounting for monochromator, collimator, analyzer, and beam-shaping contributions. The net Q-resolution may be modestly broadened for broader divergence in supermirror elements, particularly at higher incident energies ($E_i>12$ meV) due to a drop in polarization efficiency (Gorkov et al., 22 Apr 2025).

Standard workflows for SNP experiments involve: (1) symmetry and representation analysis from powder data, (2) collection of multiple $P_{ij}(\mathbf{Q})$ matrices in multiple sample orientations, (3) explicit calibration of polarization efficiencies, (4) statistical refinement including domain fractions and multipole amplitudes, and (5) simultaneous use of unpolarized and polarized diffraction data for robust parameter estimation and validation (Qureshi, 2018, Babkevich et al., 2017).

---

Spherical neutron polarimetry provides the only direct means to map the full spin-dependent scattering matrix, enabling identification of complex and subtle forms of magnetic order, resolution of chiral and multipolar phenomena, and even the search for time-reversal violating correlations in quantum matter (Gorkov et al., 22 Apr 2025, Urru et al., 2022, Gudkov et al., 2019, Babkevich et al., 2017, Qureshi, 2018).