betaFeedback

WorksPhysics and Astronomy

Framework for Polarized Magnetic Neutron Scattering from Nanoparticle Assemblies with Vortex-Type Spin Textures

Abstract

Within the framework of the recently introduced multi-nanoparticle power-series expansion method for the polarized small-angle neutron scattering (SANS) cross section, we present analytical expressions for the polarized SANS observables arising from dilute nanoparticle assemblies with antisymmetric vortex-type spin structures. We establish connections between the magnetic correlation coefficients and the magnetic field-dependent vortex-axes distribution function, which is related to the random orientations of the magnetocrystalline anisotropy axes of the nanoparticles. Our analytical results are validated through a comparative analysis with micromagnetic simulations. This framework contributes to a comprehensive understanding of polarized magnetic neutron scattering from spherical nanoparticle systems exhibiting vortex-type spin structures.

Introduction

Magnetic nanoparticles, which are in the scope of immense interdisciplinary research, offer versatile applications, e.g., in materials science, nanotechnology, and biomedicine [1, 2, 3, 4, 5, 6, 7]. They open up new possibilities in the nanoscopic realm and drive technological advances and breakthrough discoveries. But still, at the current stage of research, it is an immense challenge to characterize their internal spin structure, which is generally to be expected as nonuniform (e.g., [8, 9, 10, 11, 12, 13, 14, 15, 16]).

Magnetic small-angle neutron scattering (SANS) is possibly the only technique to probe the spatial variation of spin structures on a scale of ∼1\sim1−100 nm and in the bulk of the material [17, 18]. Recent advances in the understanding of magnetic SANS from complex nanoparticle systems have been achieved by the marriage of micromagnetic theory and magnetic neutron scattering formalism, through both computer simulations and analytical calculations [19, 20, 21, 22, 23, 24, 25, 26, 27]. Although computer simulations offer considerable potency in predicting neutron scattering observables for intricate nanoparticle assemblies, their drawback lies in their time-intensive nature, vast parameter space, and the inherent challenge of interpreting results. This complexity hinders the derivation of overarching conclusions and poses a substantial obstacle in formulating generalized statements.

To address these challenges, Adams et al. [28] introduced the multi-nanoparticle power-series expansion (MNPSE) method to study the neutron scattering signatures from spherical nanoparticle assemblies featuring diverse types of magnetic surface anisotropy. Here, we use the MNPSE approach to predict the main features of nanoparticle assemblies with inherent vortex-type spin textures as to be seen in the neutron scattering observables. Vortex-type structures are ubiquitous in magnetism research and are encountered in many systems, such as in type-II superconductors [29], GdCoO2_2 micropillars [30], Nd-Fe-B magnets [31], iron oxide nanoparticles [32, 33, 34] and nanoflowers [35], or the very recently discovered topological vortex rings in a chiral magnetic nanocylinder [36]. Our results, which replace the conventional analytical formulation for the superspin model, enable the straightforward prediction of the spin-flip SANS cross section Isf(q)I_{\mathrm{sf}}(q) and the corresponding spin-flip pair-distance distribution function psf(r)p_{\mathrm{sf}}(r) arising from spatially antisymmetric spin structures, such as nanovortices, through easily-applicable analytical expressions.

The latter is organized as follows: We start out by analyzing the main features of the first-order MNPSE method for the spin-flip SANS cross section and the pair-distance distribution function. This approach is valid for an arbitrary linear magnetization distribution. Subsequently, for the particular case of a linear vortex, we derive analytical expressions for the two- and one-dimensional SANS observables. The analytical expressions are compared to the results of micromagnetic computations. We refer to the Supplemental Material [37] for details regarding the analytical derivations and the micromagnetic SANS simulations.

Linear MNPSE method

This approach is based on the following expansion for the magnetization vector field,

M(r′)=[m0xm0ym0z]+[m1xxm1xym1xzm1yxm1yym1yzm1zxm1zym1zz]⋅[x′y′z′],(1)\mathbf{M}(\mathbf{r}') = \begin{bmatrix} m_0^x \\ m_0^y \\ m_0^z \end{bmatrix} + \begin{bmatrix} m_1^{xx} & m_1^{xy} & m_1^{xz} \\ m_1^{yx} & m_1^{yy} & m_1^{yz} \\ m_1^{zx} & m_1^{zy} & m_1^{zz} \end{bmatrix} \cdot \begin{bmatrix} x' \\ y' \\ z' \end{bmatrix}, \tag*{(1)}

where r′=[x′,y′,z′]\mathbf{r}'=[x',y',z'] denotes the position vector in the local particle frame. The model consists of 12 expansion coefficients per particle, i.e., 3 zero-order coefficients m0im_0^i and 9 first-order coefficients m1jkm_1^{jk}. For a dilute assembly of spherical nanoparticles, the MNPSE formalism yields the following expression for the azimuthally-averaged spin-flip SANS cross section [28, 37]:

Isf(q)=Isf0[f(qR)]2+Isf1[f′(qR)]2,(2)I_{\mathrm{sf}}(q)=I_{\mathrm{sf}}^0[f(qR)]^2+I_{\mathrm{sf}}^1[f'(qR)]^2, \tag*{(2)}

where RR is the particle radius, and the field-dependent coefficients Isf0I_{\mathrm{sf}}^0 and Isf1I_{\mathrm{sf}}^1 represent complicated averages of the magnetization coefficients m0im_0^i and m1jkm_1^{jk} over the particle assembly and over the detector plane. The corresponding basis functions are given by (u=qRu = qR):

f(u)=sin⁡u−ucos⁡uu3,f(u) = \frac{\sin u - u \cos u}{u^{3}},
f′(u)=dfdu=(u2−3)sin⁡u+3ucos⁡uu4.f'(u) = \frac{df}{du} = \frac{(u^{2} - 3)\sin u + 3u \cos u}{u^{4}}.

Here, f(u)f(u) is the form-factor function of the unit sphere [18] and f′(u)f'(u) is the related first-order derivative. By inverse Fourier transformation we find from Eq. (2) the related pair-distance distribution function [37]:

psf(r)=Isf0πr26R3[1−3r4R+r316R3]+Isf1πr210R3[1−5r4R+5r316R3−r532R5].(3)p_{\mathrm{sf}}(r) = I_{\mathrm{sf}}^{0}\frac{\pi r^{2}}{6R^{3}}\left[1-\frac{3r}{4R}+\frac{r^{3}}{16R^{3}}\right] + I_{\mathrm{sf}}^{1}\frac{\pi r^{2}}{10R^{3}}\left[1-\frac{5r}{4R}+\frac{5r^{3}}{16R^{3}}-\frac{r^{5}}{32R^{5}}\right]. \tag*{(3)}

While the zero-order contribution (Isf0I_{\mathrm{sf}}^{0}) arises from symmetric (parallel, positive) correlations only, the first-order contribution (Isf1I_{\mathrm{sf}}^{1}) contains antisymmetric (antiparallel, negative) correlations.

By comparison to micromagnetic simulations using Mumax3 [38, 39]—including isotropic exchange, a random cubic anisotropy [40], the Zeeman interaction, and the demagnetizing field—we find that this linear approach [Eqs. (2) and (3)] already captures the main features of vortex-type spin textures seen in the SANS observables.

The results that are embodied by Eqs. (2) and (3) are summarized in Figure 1. Prominent features regarding vortex-type spin structures are the decreased spin-flip scattering intensity Isf(q)I_{\mathrm{sf}}(q) at momentum transfer q=0q = 0 [see Figure 1(a,Figure 1c,Figure 1e,Figure 1g)] and the damped oscillatory behavior of the pair-distance distribution function psf(r)p_{\mathrm{sf}}(r) exhibiting negative (antiparallel) correlations related to a vortex [see Figure 1(b,Figure 1d,Figure 1f,Figure 1h); compare to [23, 31, 41]]. In the limiting case of ι=Isf1/Isf0→∞\iota= I_{\mathrm{sf}}^{1}/I_{\mathrm{sf}}^{0} \to\infty (modeling the remanent state) our linear theory predicts a maximum of Isf(q)I_{\mathrm{sf}}(q) at qmax⁡≅2.50/Rq_{\max} \cong2.50/R [maximum of (f′(u))2(f'(u))^{2}]. This prediction is in excellent agreement with the result from our micromagnetic simulations, where we find $q_{\max}(B_{0} = 0\ \mathrm{T}) \cong 2.50/R2.50/R [see Figure 1(g)]. Furthermore, the relevant zero of psf(r)p_{\mathrm{sf}}(r) is predicted as the result of the following cubic equation that is derived from Eq. (3):

Image not extracted

Figure 1. “Phase diagram” for the azimuthally-averaged spin flip SANS cross section Isf(q)I_{\mathrm{sf}}(q) [Eq. (2)] and for the spin-flip pair-distance distribution function psf(r)p_{\mathrm{sf}}(r) [Eq. (3)] within the limits of the first-order magnetization model. The left panel shows the analytical results, while the right panel features the corresponding results of the micromagnetic simulations. The ratio ι\iota of the zero-order coefficient Isf0I_{\mathrm{sf}}^{0} and the first-order coefficient Isf1I_{\mathrm{sf}}^{1} determines the appearance of vortex-type spin structures. Field (B0=μ0H0B_{0} = \mu_{0}H_{0}) variations in the simulations correspond to ι\iota variations in the analytical part (zero field: ι→∞\iota\to\infty; saturation: ι→0\iota\to0). (a) Color-coded plot of the normalized Isf(q)I_{\mathrm{sf}}(q) as a function of ι=Isf1/Isf0\iota= I_{\mathrm{sf}}^{1}/I_{\mathrm{sf}}^{0} and qRqR. The black solid line in (a) describes the shift of the maximum in Isf(q)I_{\mathrm{sf}}(q) towards qmax⁡≅2.50/Rq_{\max} \cong2.50/R [white dashed line, compare (e)]. (b) Normalized psf(r)p_{\mathrm{sf}}(r) as a function of ι\iota and r/Rr/R. The black solid line in (b) describes the shift of the zero in psf(r)p_{\mathrm{sf}}(r) towards rz≅1.07Rr_{z} \cong1.07R [white dashed line, compare (f)]. (c) Normalized Isf(qR)I_{\mathrm{sf}}(qR) and (d) normalized psf(r/R)p_{\mathrm{sf}}(r/R) for different ι\iota [see inset in (d)]; the inset in (c) displays Isf(q)/Isfmax⁡I_{\mathrm{sf}}(q)/I_{\mathrm{sf}}^{\max} for 4<qR<104 < qR < 10. The colored horizontal lines in (a) and (b) correspond, respectively, to the curves in (c) and (d).

ν3+4ν2+(2−103ι)ν−(8+403ι)=0,(4)\nu^{3}+4\nu^{2}+\left(2-\frac{10}{3\iota}\right)\nu-\left(8+\frac{40}{3\iota}\right)=0, \tag*{(4)}

where ν=r/R\nu=r/R. For ι→∞\iota\to\infty, Eq. (4) predicts the zero at rz≅1.07Rr_z\cong1.07R, whereas in our micromagnetic simulations we find rz(B0=0 T)≅1.03Rr_z(B_0=0\ \mathrm{T})\cong1.03R [see Figure 1(h)].

Beyond these limits for the momentum transfer qmax⁡q_{\max} and the “zero” correlation length rzr_z we find two specific transition points for ι\iota in the two-dimensional (2D) maps shown in Figure 1(a) and (b). In Figure 1(b) we observe for ι>1\iota>1 the occurrence of negative correlations [negative values of psf(r)p_{\mathrm{sf}}(r)], and in Figure 1(a) we see that the scattering intensity at the origin of reciprocal space, Isf(q=0)/Isfmax⁡I_{\mathrm{sf}}(q=0)/I_{\mathrm{sf}}^{\max}, is constant for ι<5\iota<5 and decreases for ι>5\iota>5. The micromagnetic simulation results reveal an analogous behavior [compare Figure 1(e) and (f)].

MNPSE method: The case of a linear vortex. In the formulation of the linear MNPSE method the parameters Isf0I_{\mathrm{sf}}^{0} and Isf1I_{\mathrm{sf}}^{1} [in Eqs. (2) and (3)] are arbitrary functions of the 12 magnetization expansion coefficients in Eq. (1). In the following, we aim to adapt the linear MNPSE method to include physically motivated parameters (replacing Isf0I_{\mathrm{sf}}^{0} and Isf1I_{\mathrm{sf}}^{1}). This approach allows us to obtain a scattering model that is closer related to the underlying micromagnetic Hamiltonian in the sense that it contains information on the vortex helicity, on the orientation distribution of the vortex axes, and on the transformation behavior of the energies in the Hamiltonian under space inversion.

We consider a dilute assembly of noninteracting spherical nanoparticles that are rigidly embedded in a homogeneous and nonmagnetic matrix. Each particle is assumed to have a random orientation of its (cubic or uniaxial) magnetic anisotropy axis with respect to the externally applied magnetic field H0∥ez\mathbf{H}_0\parallel\mathbf{e}_z, which defines the laboratory frame of reference. In addition to magnetic anisotropy and the Zeeman interaction, we consider an isotropic exchange energy and, most importantly, the magnetodipolar interaction (see the Supplemental Material [37]). When the spin structure of such a spherical nanoparticle is computed starting from saturation, we always find—using the materials parameters of iron—a vortex-type texture at low fields and for particle sizes larger than about 20 nm [23, 41]. It is the dipolar interaction that is responsible for the vortex formation.

Based on these simulation results, and with the aim to obtain an approximate expression for the spin-flip SANS cross section of an ensemble of vortex-carrying randomly-oriented nanoparticles, we introduce a magnetization model with a uniform (constant) part of magnitude m0m_0 and a linear vortex term of magnitude m1m_1. More specifically, the basic magnetization vector field is written as:

M′(r′)=m0ez′+m1v(r′),(5)\mathbf{M}'(\mathbf{r}')=m_0\mathbf{e}'_z+m_1\mathbf{v}(\mathbf{r}'), \tag*{(5)}

where ez′=[0,0,1]\mathbf{e}'_z=[0,0,1] is the unit vector in z′z' direction, v(r′)=[−y′,x′,0]\mathbf{v}(\mathbf{r}')=[-y',x',0] is the linear vortex field, and r′=[x′,y′,z′]\mathbf{r}'=[x',y',z'] is the position vector with reference to the local vortex frame. Compared to Eq. (1) the number of expansion coefficients in Eq. (5) has been reduced to two. A positive m1m_1 indicates a counterclockwise (CCW) or right-handed sense of rotation, while a negative m1m_1 corresponds to a clockwise (CW) or left-handed sense of rotation. We note that for a micromagnetic Hamiltonian that contains the isotropic exchange interaction, magnetic anisotropy, the Zeeman, and magnetodipolar interaction, there exists no preference for CCW or CW vortex rotation senses in the particles. CCW and CW vortices appear with equal probability so that the chiral function averages to zero (see below). However, by including the Dzyaloshinskii-Moriya interaction (DMI), which breaks space-inversion symmetry, chirality selection takes place and leads to a nonzero chiral function [27].

Equation (5) models a linear vortex in the local vortex frame. We introduce a zyzy rotation matrix R(α,β)\mathbf{R}(\alpha,\beta) that transforms the local magnetization M′\mathbf{M}' into the laboratory frame of reference, where α\alpha and β\beta denote the (global) polar and azimuthal angles, respectively. The resulting global magnetization vector field is then obtained as:

M(r;α,β)=R(α,β)⋅M′(RT(α,β)⋅r).(6)\mathbf{M}(\mathbf{r};\alpha,\beta)=\mathbf{R}(\alpha,\beta)\cdot\mathbf{M}'\left(\mathbf{R}^{\mathrm{T}}(\alpha,\beta)\cdot\mathbf{r}\right). \tag*{(6)}

Using Eq. (6) in the MNPSE method [37], we define the ensemble-averaged (dilute) SANS cross sections as:

⟨dΣsf,χdΩ⟩=12∫04π[dΣsf,χCCWdΩ+dΣsf,χCWdΩ]ψ(α,β) dΓ(7)\left\langle\frac{\mathrm{d}\Sigma_{\mathrm{sf},\chi}}{\mathrm{d}\Omega}\right\rangle=\frac{1}{2}\int_{0}^{4\pi}\left[\frac{\mathrm{d}\Sigma_{\mathrm{sf},\chi}^{\mathrm{CCW}}}{\mathrm{d}\Omega}+\frac{\mathrm{d}\Sigma_{\mathrm{sf},\chi}^{\mathrm{CW}}}{\mathrm{d}\Omega}\right]\psi(\alpha,\beta)\,\mathrm{d}\Gamma \tag*{(7)}

where dΓ=sin⁡α dαdβ\mathrm{d}\Gamma=\sin\alpha\,\mathrm{d}\alpha\mathrm{d}\beta is the solid-angle differential, and dΣsf,χCCWdΩ(q;α,β)\frac{\mathrm{d}\Sigma_{\mathrm{sf},\chi}^{\mathrm{CCW}}}{\mathrm{d}\Omega}(\mathbf{q};\alpha,\beta) and dΣsf,χCWdΩ(q;α,β)\frac{\mathrm{d}\Sigma_{\mathrm{sf},\chi}^{\mathrm{CW}}}{\mathrm{d}\Omega}(\mathbf{q};\alpha,\beta) are the SANS cross sections referring to two nanoparticles with the same orientation (α,β)(\alpha,\beta), but opposite senses of vortex rotation (m1CCW=−m1CWm_1^{\mathrm{CCW}}=-m_1^{\mathrm{CW}}). The function ψ(α,β)\psi(\alpha,\beta) is a field-dependent probability distribution that models the orientation of both the CCW and CW vortex rotation axes (no distinction between the different polarities); its origin is related to the distribution of the net magnetization vectors of the nanoparticles. For simplicity, we assume a uniform distribution ψu\psi_u on the spherical surface, which is limited by a field-dependent conical opening angle 0∘≤αc≤90∘0^\circ\leq\alpha_c\leq90^\circ. The azimuthally-symmetric distribution is then given by (see [37] for details):

ψu(α,β)=Θ(1−α/αc)2π(1−cos⁡αc),(8)\psi_u(\alpha,\beta)=\frac{\Theta(1-\alpha/\alpha_c)}{2\pi(1-\cos\alpha_c)}, \tag*{(8)}

where Θ(ξ)\Theta(\xi) is the Heaviside function. In the fully saturated case (B0→∞B_0\to\infty) it follows that αc→0\alpha_c\to0, and $\alpha_c increases with decreasing applied magnetic field. By inserting Eqs. (5)–(8) into the formalism of the MNPSE method, we obtain the following final expressions for the randomly-averaged 2D spin-flip SANS cross section and chiral function [37]:

Image not extracted

Figure 2. Illustration of the 2D (normalized) spin-flip SANS cross section and chiral function computed from Eqs. (9) and (10) reflecting the saturation and remanence cases (particle size: D=2R=40D = 2R = 40 nm) (linear color scale). (a,b) and (g,h) show snapshots of the underlying real-space spin structures. The left panel shows the analytical results, while the right panel features the results of the micromagnetic simulations. The incoming neutron beam (∥ex\parallel e_x) is perpendicular to the applied magnetic field H0∥ez\mathbf{H}_0 \parallel e_z (B0=μ0H0B_0 = \mu_0 H_0). The maximum of the spots in (d) and (j) are found at qy,max⁡≅2.50/Rq_{y,\max} \cong2.50/R. (e,f) and (k,l) display the respective chiral functions in the remanent state for counterclockwise (CCW) and clockwise (CW) vortex rotations. Note that the specific values for αc=27∘\alpha_c = 27^\circ and for the ratio m1/m0=0.7m_1/m_0 = 0.7 in (d) are based on a fit of the analytical function [Eq. (9)] to the 2D simulation data shown in (j).

⟨dΣsfdΩ⟩(q,θ)=W8[m0f(qR)]2×[12−(cos⁡2αc+cos⁡αc)(3cos⁡2(2θ)+2cos⁡(2θ)+3)+4cos⁡(2θ)]+W2[Rm1f′(qR)]2×[3−(2cos⁡2αc+2cos⁡αc−1)cos⁡(2θ)],(9)\left\langle\frac{\mathrm{d}\Sigma_{\mathrm{sf}}}{\mathrm{d}\Omega}\right\rangle(q,\theta) = \frac{W}{8}\left[m_0 f(qR)\right]^2 \times\left[12 - \left(\cos^2\alpha_c + \cos\alpha_c\right)\left(3\cos^2(2\theta) + 2\cos(2\theta) + 3\right) + 4\cos(2\theta)\right] + \frac{W}{2}\left[Rm_1 f'(qR)\right]^2 \times\left[3 - \left(2\cos^2\alpha_c + 2\cos\alpha_c - 1\right)\cos(2\theta)\right], \tag*{(9)}
⟨dΣχCCW,CWdΩ⟩(q,θ)=±W[Rm0∣m1∣f(qR)f′(qR)cos⁡θ]×[4+cos⁡2αc+cos⁡αc−3(cos⁡2αc+cos⁡αc)cos⁡2θ],(10)\left\langle\frac{\mathrm{d}\Sigma_{\chi}^{\mathrm{CCW,CW}}}{\mathrm{d}\Omega}\right\rangle(q,\theta) = \pm W\left[Rm_0\lvert m_1\rvert f(qR)f'(qR)\cos\theta\right] \times\left[4 + \cos^2\alpha_c + \cos\alpha_c - 3\left(\cos^2\alpha_c + \cos\alpha_c\right)\cos^2\theta\right], \tag*{(10)}

where WW is a scaling constant. In Eq. (10) we have separated the chiral function into CCW (“+” sign) and CW (“−” sign) contributions. For the here-considered micromagnetic energy contributions [37] (with no chirality selection taking place), it then follows that

⟨dΣχdΩ⟩=12[⟨dΣχCCWdΩ⟩+⟨dΣχCWdΩ⟩]=0.(11)\left\langle\frac{d\Sigma_{\chi}}{d\Omega} \right\rangle= \frac{1}{2}\left[\left\langle\frac{d\Sigma_{\chi}^{\mathrm{CCW}}}{d\Omega} \right\rangle+ \left\langle\frac{d\Sigma_{\chi}^{\mathrm{CW}}}{d\Omega} \right\rangle\right] = 0. \tag*{(11)}

Figure 2 displays Eqs. (9) and (10). At saturation [Fig. 2(a,c) and (g,i)], with αc=0∘\alpha_c = 0^\circ and m1/m0=0m_1/m_0 = 0, the spin-flip SANS cross section exhibits the well-known sin⁡2θcos⁡2θ\sin^2 \theta\cos^2 \theta angular anisotropy. At remanence [Fig. 2(b,d) and (h,j)], with αc=27∘\alpha_c = 27^\circ and m1/m0=0.7m_1/m_0 = 0.7, we observe for the spin-flip signal an anisotropy that strongly differs from the saturated case, with maxima for θ=90∘\theta= 90^\circ. This observation strongly suggests that the magnetization Fourier components are anisotropic, i.e., M⃗x,y,z=M⃗x,y,z(q,θ)\vec{M}_{x,y,z} = \vec{M}_{x,y,z}(q,\theta) (compare to Eq. (4) in [37]). Consequently, to not lose this information, the experimental data analysis should be concentrated on 2D spin-flip data rather than on the azimuthally-averaged 1D data. In micromagnetic simulations of spherical nanoparticles, very similar scattering patterns were observed [27, 42].

Averaging Eq. (9) over the angle θ\theta, i.e., (2π)−1∫02π(...)dθ(2\pi)^{-1}\int_{0}^{2\pi} (...)d\theta, yields the 1D quantity [37]

⟨Isf⟩(q)=3W16[m0f(qR)]2(8−3cos⁡2αc−3cos⁡αc)+83W16[m1Rf′(qR)]2.(12)\langle I_{\mathrm{sf}}\rangle(q) = \frac{3W}{16}[m_0 f(qR)]^2(8 - 3\cos^2\alpha_c - 3\cos\alpha_c) + 8\frac{3W}{16}[m_1 R f'(qR)]^2. \tag*{(12)}

By comparison to Eq. (2) we note that the new parameters m0m_0, m1m_1, and αc\alpha_c in Eqs. (9) and (10) are related to the coefficient ratio Isf1/Isf0I_{\mathrm{sf}}^1/I_{\mathrm{sf}}^0 as follows:

ι=Isf1Isf0=8m12R2m02(8−3cos⁡2αc−3cos⁡αc).(13)\iota= \frac{I_{\mathrm{sf}}^1}{I_{\mathrm{sf}}^0} = \frac{8m_1^2R^2}{m_0^2(8 - 3\cos^2\alpha_c - 3\cos\alpha_c)}. \tag*{(13)}

which emphasizes the importance of the vortex-axes distribution function. The angle αc\alpha_c may be obtained from the analysis of (preferentially 2D) experimental spin-flip SANS data (compare to Fig. 2(d) and (j) and the video clip in [37]).

Conclusion. In this Letter we have demonstrated that the linear MNPSE approach captures the main effects in the spin-flip SANS cross section and pair-distance distribution function stemming from dilute assemblies of spherical nanoparticles exhibiting vortex-type spin textures. A crucial insight is that the linear functionality represents the most important contribution to the magnetic neutron scattering cross section. Based on the specific case of a linear vortex model, we have derived analytical expressions for the 2D and 1D spin-flip and chiral cross sections of an ensemble of randomly oriented vortex-carrying nanoparticles. The maximum of the spin-flip scattering intensity and the zero of the pair-distance distribution function appear, respectively, at momentum transfer qmax⁡≃2.50/Rq_{\max} \simeq2.50/R and position rz≃1.07Rr_z \simeq1.07R, where RR denotes the radius of the spherical nanoparticles. The analytical predictions, which enable e.g. the determination of the field-dependent conical opening angle αc\alpha_c of the vortex-axes distribution from experimental data, are in very good agreement with the results of micromagnetic simulations. The chiral SANS cross section is sensitive to the vortex rotation sense, but in a many-particle system with no chirality selection, it averages to zero (as expected). A candidate for chirality selection is the DMI interaction that breaks space-inversion symmetry.

We acknowledge financial support from the National Research Fund of Luxembourg (AFR Grant No. 15639149, CORE Grant DeQuSky, and PRIDE MASSENA Grant).

References

References

  1. [1]M. De, P. S. Ghosh, and V. M. Rotello, Adv. Mater. 20, 4225 (2008).
  2. [2]Y. Diebold and M. Calonge, Prog. Ret. Eye Res. 29, 596 (2010).
  3. [3]S. C. Baetke, T. Lammers, and F. Kiessling, Br. J. Radiol. 88, 20150207 (2015).
  4. [4]W. J. Stark, P. R. Stoessel, W. Wohlleben, and A. Hafner, Chem. Soc. Rev. 44, 5793 (2015).
  5. [5]X. Han, K. Xu, O. Taratula, and F. Khashayar, Nanoscale 11, 799 (2019).
  6. [6]A. Lak, S. Disch, and P. Bender, Adv. Sci. 8, 2002682 (2021).
  7. [7]X. Batlle, C. Moya, M. Escoda-Torroella, Ò. Iglesias, A. Fraile Rodríguez, and A. Labarta, J. Magn. Magn. Mater. 543, 168594 (2022).
  8. [8]J. A. De Toro, M. Vasilakaki, S. S. Lee, M. S. Andersson, P. S. Normile, N. Yaacoub, P. Murray, E. H. Sánchez, P. Muñiz, D. Peddis, R. Mathieu, K. Liu, J. Geshev, K. N. Trohidou, and J. Nogués, Chem. Mater. 29, 8258 (2017).
  9. [9]A. Lappas, G. Antonaropoulos, K. Britanakis, M. Vasilakaki, K. N. Trohidou, V. Iannotti, G. Ausanio, A. Kostopoulou, M. Abeykoon, I. K. Robinson, and E. S. Bozin, Phys. Rev. X 9, 041044 (2019).
  10. [10]D. Zákutná, D. Nižňanský, L. C. Barnsley, E. Babcock, Z. Salhi, A. Feoktystov, D. Honecker, and S. Disch, Phys. Rev. X 10, 031019 (2020).
  11. [11]J. T. Batley, M. Nguyen, I. Kamboj, C. Korostynski, E. S. Aydil, and C. Leighton, Chem. Mater. 32, 6494 (2020).
  12. [12]C. Kons, M.-H. Phan, H. Srikanth, D. A. Arena, Z. Nemati, J. A. Borchers, and K. L. Krycka, Phys. Rev. Mater. 4, 034408 (2020).
  13. [13]T. Köhler, A. Feoktystov, O. Petracic, E. Kentzinger, T. Bhatnagar-Schöffmann, M. Feygenson, N. Nandakumaran, J. Landers, H. Wende, A. Cervellino, U. Rücker, A. Kovács, R. E. Dunin-Borkowski, and T. Brückel, Nanoscale 13, 6965 (2021).
  14. [14]Ö. Iglesias and H. Kakchachi, Single Nanomagnet Behaviour: Surface and Finite-Size Effects, in New Trends in Nanoparticle Magnetism, edited by D. Peddis, S. Laureti, and D. Fiorani (Springer International Publishing, Cham, 2021) pp. 3–38.DOI
  15. [15]E. M. Jefremovas, M. de la Fuente Rodríguez, D. Alba Venero, C. Echevarría-Bonet, P. Bender, B. Fåk, J. A. Blanco, and L. Fernández Barquín, Commun. Mater. 4, 56 (2023).
  16. [16]M. Gerina, M. Sanna Angotzi, V. Mameli, V. Gajdošová, D. N. Rainer, M. Dopita, N.-J. Steinke, D. Aurélio J. Vejpravová, and D. Zákuťná, Nanoscale Adv. 5, 4653 (2023).
  17. [17]S. Mühlbauer, D. Honecker, E. A. Périgo, F. Bergner, S. Disch, A. Heinemann, S. Erokhin, D. Berkov, C. Leighton, M. R. Eskildsen, and A. Michels, Rev. Mod. Phys. 91, 015004 (2019).
  18. [18]A. Michels, Magnetic Small-Angle Neutron Scattering: A Probe for Mesoscale Magnetism Analysis (Oxford University Press, Oxford, 2021).
  19. [19]K. L. Metlov and A. Michels, Sci. Rep. 6, 25055 (2016).
  20. [20]A. Michels, D. Mettus, D. Honecker, and K. L. Metlov, Phys. Rev. B 94, 054424 (2016).
  21. [21]I. Mirebeau, N. Martin, M. Deutsch, L. J. Bannenberg, C. Pappas, G. Chaboussant, R. Cubitt, C. Decorse, and A. O. Leonov, Phys. Rev. B 98, 014420 (2018).
  22. [22]A. A. Mistonov, I. S. Dubitskiy, I. S. Shishkin, N. A. Grigoryeva, A. Heinemann, N. A. Sapole­tova, G. A. Valkovskiy, and S. V. Grigoriev, J. Magn. Magn. Mater. 477, 99 (2019).
  23. [23]L. G. Vivas, R. Yanes, D. Berkov, S. Erokhin, M. Bersweiler, D. Honecker, P. Bender, and A. Michels, Phys. Rev. Lett. 125, 117201 (2020).
  24. [24]V. D. Zaporozhets, Y. Oba, A. Michels, and K. L. Metlov, J. Appl. Cryst. 55, 592 (2022).
  25. [25]C. Kons, K. L. Kryvcka, J. R. Robles, N. Ntallis, M. Pereiro, M.-H. Phan, H. Srikanth, J. A. Borchers, and D. A. Arena, ACS Appl. Nano Mater. 6, 10986 (2023).
  26. [26]V. Ukleev, F. Ajejas, A. Devishvili, A. Vorobiev, N.-J. Steinke, R. Cubitt, C. Luo, R.-M. Abrudan, F. Radu, V. Cros, N. Reyren, and J. S. White, STAM Methods 25, 2315015 (2024).
  27. [27]E. P. Sinaga, M. P. Adams, E. H. Hasdeo, and A. Michels, (2024), arXiv:2402.00558 [cond-mat.mes-hall].arxiv.org/abs/2402.00558
  28. [28]M. P. Adams, E. P. Sinaga, H. Kachkachi, and A. Michels, Phys. Rev. B 109, 024429 (2024).
  29. [29]M. R. Eskildsen, E. M. Forgan, and H. Kawano-Furukawa, Rep. Prog. Phys. 74, 124504 (2011).
  30. [30]C. Donnelly, K. L. Metlov, V. Scagnoli, M. Guizar-Sicairos, M. Holler, N. S. Bingham, J. Raabe, L. J. Heyderman, N. R. Cooper, and S. Gliga, Nat. Phys. 17, 316 (2021).
  31. [31]M. Bersweiler, Y. Oba, E. P. Sinaga, I. Peral, I. Titov, M. P. Adams, V. Rai, K. L. Metlov, and A. Michels, Phys. Rev. B 108, 094434 (2023).
  32. [32]A. S. Samardak, A. V. Davydenko, A. V. Ognev, Y. S. Jeon, Y. S. Choi, and Y. K. Kim, Jap. J. Appl. Phys. 55, 100303 (2016).
  33. [33]N. A. Usov, M. S. Nesmeyanov, and V. P. Tarasov, Sci. Rep. 8, 1224 (2018).
  34. [34]G. Niraula, D. Toneto, G. F. Goya, G. Zoppellaro, J. A. H. Coaquira, D. Muraca, J. C. Denardin, T. P. Almeida, M. Knobel, A. I. Avesh, and S. K. Sharma, Nanoscale Adv. 5, 5015 (2023).
  35. [35]C. Moya, M. Escoda-Torroella, J. Rodríguez-Álvarez, A. I. Figueroa, Í. García, I. B. Ferrer-Vidal, A. Gallo-Cordova, M. Puerto Morales, L. Albella, A. Fraile Rodríguez, A. Labarta, and X. Batlle, Nanoscale 16, 1942 (2024).
  36. [36]Y. Liu and N. Nagaosa, Phys. Rev. Lett. 132, 126701 (2024).
  37. [37]See the Supplemental Material [URL] for further details on the numerical micromagnetic calculations and on the MNPSE method used for the analytical calculation of the spin-flip SANS cross section and chiral function of vortex-carrying nanoparticles. We also provide a video clip that features the field evolution of the spin structure, the 2D and 1D spin-flip SANS cross section, as well as the pair-distance distribution function along the hysteresis loop.
  38. [38]A. Vansteenkiste, J. Leliaert, M. Dvornik, M. Helsen, F. Garcia-Sanchez, and B. Van Waeyenberge, AIP Advances 4, 101733 (2014).
  39. [39]J. Leliaert, M. Dvornik, J. Mulkers, J. De Clercq, M. V. Milošević, and B. Van Waeyenberge, J. Phys. D: Appl. Phys. 51, 123002 (2018).
  40. [40]In the numerical micromagnetic simulations we use the materials parameters for (cubic) iron, and we also consider the case of a random uniaxial anisotropy (see [37]).
  41. [41]E. P. Sinaga, M. P. Adams, M. Bersweiler, L. G. Vivas, E. H. Hasdeo, J. Leliaert, P. Bender, D. Honecker, and A. Michels, Phys. Rev. B 107, 014416 (2023).
  42. [42]M. P. Adams, E. P. Sinaga, and A. Michels, IUCrJ 10, 261 (2023).

Paper details

vlnc:W_bpcux8aj
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents