交换多重态框架中的交变磁性及广义张量性质
Altermagnetism and generalised tensorial properties in the exchange multiplet framework
查看机构详情
- Clarendon Laboratory, Department of Physics, University of Oxford（牛津大学物理系克拉伦登实验室）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
本文提出基于交换多重态形式论的表示论框架，将磁序参量视为连续变量，用于计算磁性材料的时间反演奇张量性质，并适用于交变磁体与传统反铁磁体。
中文摘要 AI 辅助
本文发展了一个基于Bertaut–Izyumov交换多重态形式论的表示论框架，用于计算磁性材料的时间反演奇张量性质。与针对磁序参量不同取向采用不同磁点群或自旋群的方法不同，交换多重态框架允许将序参量视为连续变量。研究表明，对于恒定磁矩的共线结构，协变性对Landau–Lifshitz展开中出现的张量交织子施加了一组简单的对称性约束。这导致了规范交织子形式，将张量构造的磁性和晶体学部分分解开来。张量场可以构建为在物理变量中保持笛卡尔表示，同时通过对称适配的多项式基获得对序参量方向的系统性依赖。对于任意序参量方向，磁点群选择规则可自动恢复，同时该框架还提供了共旋转（无自旋轨道耦合）与非共旋转贡献的自然分离。该方法同样适用于交变磁体和传统反铁磁体，特别适用于磁序参量可连续变化的实验和第一性原理计算。
英文摘要
A representation-theoretic framework for the calculation of time-reversal-odd tensorial properties of magnetic materials based on the Bertaut--Izyumov exchange-multiplet formalism is developed. In contrast to approaches based on distinct magnetic point groups or spin groups for different orientations of the magnetic order parameter, the exchange-multiplet framework permits the order parameter to be treated as a continuous variable. It is shown that, for constant-moment collinear structures, covariance imposes a simple set of symmetry constraints on the tensor intertwiners appearing in a Landau--Lifshitz expansion. This leads to canonical intertwiner forms that factorise the magnetic and crystallographic parts of the tensor construction. Tensor fields can be built to retain a Cartesian representation in the physical variables while acquiring a systematic dependence on the order-parameter direction through symmetry-adapted polynomial bases. Magnetic-point-group selection rules are recovered automatically for arbitrary order-parameter directions, while the framework also provides a natural separation of co-rotating (SOC-free) and non-co-rotating contributions. The approach is applicable to altermagnets and conventional antiferromagnets alike and is particularly suited to experiments and first-principles calculations in which the magnetic order parameter can be continuously varied.
