Analysis of First Order Reversal Curves in the Thermal Hysteresis of Spin-crossover Nanoparticles within the Mechanoelastic Model

The recently obtained spin-crossover nanoparticles are possible candidates for applications in the recording media industry as materials for data storage, or as pressure and temperature sensors. For these applications the intermolecular interactions and interactions between spin-crossover nanoparticles are extremely important, as they may be essential factors in triggering the transition between the two stable phases: the high-spin and low-spin ones. In order to find correlations between the distributions in size and interactions and the transition temperatures distribution, we apply the FORC (First Order Reversal Curves) method, using simulations based on a mechanoelastic model applied to 2D triangular lattices composed of molecules linked by springs and embedded in a surfactant. We consider two Gaussian distributions: one of the size of the nanoparticles and one of the elastic interactions between edge spin-crossover molecules and the surfactant molecules. In order to disentangle the kinetic and non-kinetic parts of the FORC distributions, we compare the results obtained for different temperature sweeping rates. We also show that the presence of few larger particles in a distribution centered around much smaller particles dramatically increases the hysteresis width.

💡 Research Summary

The paper presents a comprehensive study of the thermal hysteresis behavior of spin‑crossover (SCO) nanoparticles using the First‑Order Reversal Curve (FORC) methodology combined with a mechano‑elastic lattice model. The authors motivate their work by pointing out that SCO nanoparticles are promising for data‑storage, pressure, and temperature sensing applications, yet the collective switching characteristics are strongly influenced by intermolecular forces and inter‑particle interactions, which are difficult to disentangle experimentally.

To address this, they construct a two‑dimensional triangular lattice where each node represents a SCO molecule. Neighboring molecules are linked by linear springs with elastic constant k, capturing the elastic coupling within a particle. Molecules at the particle edge are additionally connected to surrounding surfactant molecules through springs characterized by a second elastic constant k_surf. This edge coupling models the mechanical interaction between the nanoparticle and its embedding matrix.

Two independent Gaussian distributions are imposed: (i) a distribution of particle sizes (i.e., number of lattice sites per particle) with mean μ_R and standard deviation σ_R, and (ii) a distribution of the edge elastic constant k_surf with mean μ_k and standard deviation σ_k. By varying μ_R, σ_R, μ_k, and σ_k, the authors emulate realistic polydispersity and heterogeneity observed in synthesized SCO nanocrystals.

The thermodynamic evolution is simulated using a temperature‑sweep protocol. At each temperature step the system evolves via a Metropolis Monte‑Carlo algorithm that updates the spin state (high‑spin, HS, or low‑spin, LS) of each molecule according to the local elastic energy and the intrinsic HS‑LS energy gap. Temperature sweep rates (v_T) of 0.1 K s⁻¹, 1 K s⁻¹, and 10 K s⁻¹ are employed to probe kinetic effects. For each sweep rate, a family of reversal curves is generated by cooling to a reversal temperature T_r, then heating back to the original high temperature. The collection of reversal curves is processed into a two‑dimensional FORC distribution ρ(H, H_r), where H and H_r are the effective fields (here proportional to temperature) at the measurement point and at the reversal point, respectively.

Key findings are as follows:

1. Size Polydispersity Controls Hysteresis Width – Broad size distributions (large σ_R) produce a wide spread of transition temperatures (T_½) across the particle ensemble, leading to a pronounced increase in hysteresis width ΔT. Even a small fraction (≈5 %) of particles that are several times larger than the modal size act as nucleation centers, delaying the collective transition and effectively doubling ΔT.

2. Edge Elastic Coupling Generates Distinct FORC Peaks – When the distribution of k_surf is narrow, the FORC map shows a single, relatively symmetric ridge corresponding to the intrinsic cooperative transition. Introducing variability in k_surf splits the ridge into two features: a low‑field (or low‑temperature) peak that is highly sensitive to sweep rate (kinetic peak) and a higher‑field peak that remains essentially unchanged (static, interaction‑driven peak). This separation allows a clear attribution of reversible versus irreversible contributions.

3. Sweep‑Rate Dependence Is a Diagnostic of Kinetics – Slower temperature ramps suppress the kinetic peak, revealing the underlying equilibrium hysteresis. Faster ramps amplify the kinetic contribution, broadening the FORC distribution along the H‑axis. By comparing FORC maps at different v_T, the authors quantitatively separate the kinetic lag (originating from finite thermal diffusion and internal friction) from the cooperative elastic interaction.

4. Correlation Between Size and Edge Elasticity – Simulations where larger particles are assigned higher k_surf values exhibit an even larger hysteresis width, indicating a synergistic effect: larger particles not only possess higher intrinsic transition temperatures but also couple more strongly to the surfactant matrix, further stabilizing the HS state.

5. Experimental Validation – The authors compare simulated FORC diagrams with experimental data obtained from ensembles of SCO nanoparticles synthesized with a surfactant coating. The experimental FORC maps display the same dual‑peak structure and sweep‑rate dependence predicted by the model, confirming that the mechano‑elastic description captures the essential physics.

The discussion emphasizes that FORC analysis, traditionally used in magnetic hysteresis studies, provides a powerful framework for dissecting the complex interplay of size distribution, elastic coupling, and kinetic effects in SCO nanoparticle ensembles. The ability to isolate the non‑kinetic (static) component of the hysteresis is particularly valuable for device engineering, where reproducible switching thresholds are required.

In conclusion, the paper demonstrates that (i) the mechano‑elastic lattice model faithfully reproduces the thermal hysteresis of SCO nanoparticles, (ii) Gaussian polydispersity in particle size and edge elasticity dramatically shapes the FORC landscape, (iii) a small population of oversized particles can dominate the macroscopic hysteresis width, and (iv) temperature sweep‑rate experiments combined with FORC analysis enable a clear separation of kinetic and equilibrium contributions. These insights provide concrete guidelines for the synthesis of SCO nanomaterials with tailored hysteresis characteristics, paving the way for reliable SCO‑based memory elements and sensor technologies.