Kirkpatrick Kyle M, Zhou Benjamin H, Bunting Philip C, Rinehart Jeffrey D
Department of Chemistry and Biochemistry, University of California - San Diego La Jolla California 92093 USA
Materials Science and Engineering Program, University of California - San Diego La Jolla California 92093 USA.
Chem Sci. 2023 Jun 15;14(27):7589-7594. doi: 10.1039/d3sc02113k. eCollection 2023 Jul 12.
Magnetization is a common measurable for characterizing bulk, nanoscale, and molecular materials, which can be quantified to high precision as a function of an applied external field. These data provide detailed information about a material's electronic structure, phase purity, and impurities, though interpreting this data can be challenging due to many contributing factors. In sub-single-domain particles of a magnetic material, an inherently time-dependent rotation of the entire particle spin becomes possible. This phenomenon, known as superparamagnetism (SPM), simultaneously represents a very early size-dependent property to be considered, while being one of the least explored in the current quantum materials era. This discrepancy is, at least in part, due to the need for models with less built-in complexity that can facilitate the generation of comparative data. In this work, we map an extensive dataset of variable-size SPM FeO (magnetite) to an intrinsic statistical model for their field-dependence. By constraining the SPM behavior to a probabilistic model, the data are apportioned to several decorrelated sources. From this, there is strong evidence that standard measures such as saturation magnetization, , are poor comparative parameters, being dependent on experimental knowledge and measurement of the magnetic mass. In contrast, parameters of the intrinsic probability distribution, such as the maximum susceptibility, , are far better suited to describe the SPM behavior itself and do not propagate unknown magnetic mass error. By confining the data fitting to intrinsic variables of the model distribution, scaling parameters, and linear contributions, we find greater value in magnetic data, ultimately aiding potential synthesis diagnostics and prediction of new properties and functionality.
磁化是表征块状、纳米级和分子材料的一种常见可测量量，它可以作为外加磁场的函数被高精度地量化。这些数据提供了有关材料电子结构、相纯度和杂质的详细信息，不过由于许多因素的影响，解释这些数据可能具有挑战性。在磁性材料的亚单畴颗粒中，整个颗粒自旋的固有时间依赖性旋转成为可能。这种现象，称为超顺磁性（SPM），同时代表了一个非常早期的尺寸依赖性特性，需要加以考虑，而在当前的量子材料时代，它是研究最少的特性之一。这种差异至少部分是由于需要具有较低内在复杂性的模型，以便于生成比较数据。在这项工作中，我们将可变尺寸的SPM FeO（磁铁矿）的大量数据集映射到其场依赖性的固有统计模型。通过将SPM行为约束到一个概率模型中，数据被分配到几个不相关的源。由此有强有力的证据表明，诸如饱和磁化强度等标准测量值是较差的比较参数，它们依赖于实验知识和磁质量的测量。相比之下，固有概率分布的参数，如最大磁化率，更适合描述SPM行为本身，并且不会传播未知的磁质量误差。通过将数据拟合限制在模型分布的固有变量、缩放参数和线性贡献上，我们在磁性数据中发现了更大的价值，最终有助于潜在的合成诊断以及新特性和功能的预测。
