发表机构
WPI Advanced Institute for Materials Research (AIMR), Tohoku University; Graduate School of Engineering Science, The University of Osaka; Nano-materials Micro-devices Research Center, Osaka Institute of Technology; Spintronics Research Network Division, Institute for Open and Transdisciplinary Research Initiatives, The University of Osaka; Center for Spintronics Research Network (CSRN), Graduate School of Engineering Science, The University of Osaka; National Institute of Advanced Industrial Science and Technology (AIST), Research Institute for Hybrid Functional Integration; Center for the Promotion of Advanced Interdisciplinary Research, The University of Osaka; Center for Science and Innovation in Spintronics, Core Research Cluster, Tohoku University; Graduate School of Engineering, The University of Osaka; ULVAC– The University of Osaka Joint Research Laboratory for Future Technology, The University of Osaka; Institute for Chemical Research, Kyoto University; Center for Spintronics Research Network (CSRN), Institute for Chemical Research, Kyoto University; Faculty of Advanced Engineering, Tokyo University of Science; International Center for Synchrotron Radiation Innovation Smart, Tohoku University; Department of Electronic Science and Engineering, Kyoto University（东北大学物质材料研究机构; 大阪工业大学研究生院; 大阪工业大学纳米材料微器件研究中心; 大阪工业大学开放与跨学科研究倡议研究所自旋电子学研究网络部门; 大阪工业大学研究生院自旋电子学研究网络中心; 产业技术综合研究所混合功能集成研究所; 大阪工业大学高级跨学科研究促进中心; 东北大学自旋电子学科学与创新中心核心研究集群; 大阪工业大学工程学院; 爱发科-大阪工业大学未来技术联合实验室; 京都大学化学研究所; 京都大学化学研究所自旋电子学研究网络中心; 东京理科大学先进工学部; 东北大学同步辐射创新智能国际中心; 京都大学电子科学与工程系）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究通过室温下两个磁性斯格明子的布朗运动实现信息引擎，实验验证信息流与信息亲和力成正比，确立信息欧姆定律，并修正了真实电路中的基尔霍夫电压定律。
AI 中文摘要
降低计算能耗——在人工智能时代一个紧迫的问题——需要一个基于信息热力学的电路设计框架。通过将非平衡统计力学扩展到信息热力学，已证明携带信息的概率流（以下称为“信息流”）由其共轭热力学力，即“信息亲和力”驱动。由于信息亲和力在电路中的作用类似于电压，因此提出信息传输也遵循类似于欧姆定律的线性响应关系。然而，尽管有这一理论预测，实验上建立信息传输的欧姆类本构关系仍是一个重要的未解问题。在此，我们实现了一个由两个在室温下进行布朗运动的磁性斯格明子组成的信息引擎。我们证明了所产生的信息流在相对广泛的范围内与麦克斯韦妖产生的信息亲和力成正比，从而确立了信息的欧姆定律。此外，我们阐明，在真实信息电路中，信息欧姆定律和基尔霍夫电压定律都必须稍作修改，以考虑信息载体之间的相互作用以及信息流的散射。
英文摘要
Reducing the energy cost of computation-a pressing issue in the age of AI-requires a framework for circuit design based on information thermodynamics. By extending non-equilibrium statistical mechanics to information thermodynamics, it has been shown that a probability flow carrying information-hereinafter referred to as an 'information current'-is driven by its conjugate thermodynamic force, namely 'information affinity'. Since information affinity plays a role analogous to that of voltage in an electrical circuit, it is suggested that information transport also follows a linear response relationship analogous to Ohm's law. However, despite this theoretical prediction, experimentally establishing an Ohm-like constitutive relation for information transport remains an important unsolved problem. Here, we realise an information engine consisting of two magnetic skyrmions undergoing Brownian motion at room temperature. We demonstrate that the resulting information current is proportional to the information affinity generated by Maxwell's demon over a relatively wide range, establishing Ohm's law for information. Furthermore, we clarify that, in real information circuits, both Ohm's law for information and Kirchhoff's voltage law must be slightly modified to account for interactions between information carriers and the scattering of the information current.
Comments18 pages, 4 figures
