Impact of Magneto‑Dipole Interactions on the Specific Absorption Rate of Iron Oxide Nanoparticle Assemblies

Background

Magnetic hyperthermia is a leading strategy for cancer treatment, relying on heat generated by magnetic nanoparticles (MNPs) in an alternating magnetic field. The heating efficiency, expressed as SAR, depends on particle size, material, field amplitude, frequency, and the surrounding medium’s properties. Iron‑oxide nanoparticles are particularly attractive due to their biocompatibility, biodegradability, and visibility in MRI.

In vivo, MNPs often become tightly bound to tissues, limiting their physical rotation. Consequently, Brownian relaxation is negligible, and only Néel relaxation—i.e., reorientation of the magnetic moment within the particle—contributes to heating. Moreover, magnetic interactions between closely spaced particles can significantly alter the SAR, especially in dense aggregates that form fractal structures inside tumors.

Previous work has highlighted the role of thermal fluctuations in SAR for dilute, non‑interacting assemblies and identified an optimal particle diameter (~20 nm) for maximal heating at typical hyperthermia frequencies (200–500 kHz). However, experimental SAR values frequently fall below theoretical predictions, likely due to strong magneto‑dipole coupling in dense assemblies. This study aims to quantify that effect using a rigorous stochastic Landau–Lifshitz framework.

Numerical Simulation

Non‑Interacting Nanoparticles

For a dilute, randomly oriented superparamagnetic assembly, the Fokker–Planck equation yields an approximate kinetic description (Eq. 1) for the populations n₁ and n₂ of the two energy wells. The reduced magnetization follows Eq. (2). These equations, validated against direct stochastic Landau–Lifshitz simulations, enable rapid calculation of hysteresis loops for various particle diameters, frequencies, and field amplitudes.

Nanoparticle Clusters

We model two cluster morphologies: (i) quasi‑spherical 3‑D random clusters and (ii) fractal clusters typical of intracellular aggregates. In the 3‑D case, particles of diameter D are placed at random positions inside a sphere of radius R_cl while ensuring no direct contact. The packing density η controls the mean inter‑particle distance D_av via D_av = (6V_cl/(πN_p))^1/3 and η = (D/D_av)³. Fractal clusters are generated using the Filippov algorithm with descriptors D_f and k_f, yielding a particle count N_p = k_f(2R_g/D)^D_f.

The stochastic Landau–Lifshitz equation (Eq. 3) governs the dynamics of each particle’s unit magnetization vector α_i, with effective fields derived from anisotropy (Eq. 5), Zeeman coupling (Eq. 6), and dipole–dipole interactions (Eq. 7). Thermal fields obey the fluctuation–dissipation relation (Eq. 9). The numerical scheme follows Refs. [13, 40, 41].

Results and Discussion

Non‑Interacting Iron Oxide Nanoparticles

Using typical iron‑oxide parameters—saturation magnetization M_s = 70 Am²/kg, anisotropy constant K = 10⁴ J/m³, temperature 300 K, and diameters 10–30 nm—we computed SAR for a fixed field amplitude H₀ = 8 kA/m at frequencies 200–500 kHz. The SAR peaks for diameters 20–21 nm, reaching 350–450 kW/kg, in agreement with experimental reports for optimally sized particles. However, many in‑vivo measurements report lower SAR (~20–50 kW/kg), suggesting additional interaction effects.

Assembly of 3‑D Clusters

Figure 3 illustrates how increasing the cluster packing density η diminishes the hysteresis loop area and SAR. For N_p = 40, η values of 0.005, 0.04, and 0.32 correspond to inter‑particle ratios D_av/D of 1.46, 2.92, and 5.84, respectively. At η ≥ 0.04, magneto‑dipole coupling significantly reduces SAR, with a six‑fold drop when η rises from 0.005 to 0.32. The optimal diameter remains near 20 nm, though the peak becomes less pronounced.

Assembly of Fractal Clusters

Fractal aggregates show an even stronger SAR suppression. The absorption peak shifts toward smaller diameters for all but the highest fractal dimension (D_f = 2.7). Non‑magnetic shell thicknesses t_Sh of 1 nm mitigate nearest‑neighbour dipole coupling, partially restoring SAR values and narrowing the diameter dependence, as depicted in Figure 5.

Conclusions

Our simulations demonstrate that magneto‑dipole interactions within nanoparticle clusters—quantified by the packing density η—dramatically reduce the SAR of iron‑oxide assemblies. For conventional 3‑D clusters, increasing η from 0.005 to 0.32 lowers SAR by roughly a factor of six, aligning theoretical predictions with experimentally observed values (~50–100 kW/kg). Fractal clusters exhibit similar SAR suppression and a systematic shift of the optimal diameter to smaller sizes. Introducing non‑magnetic surface shells can counteract dipole coupling and recover higher SARs, which is promising for biomedical applications.

These findings reinforce the importance of controlling particle size, aggregation state, and surface chemistry when designing MNPs for hyperthermia. Future work should incorporate realistic size distributions and potential exchange interactions for a more comprehensive model.