高等專題討論二期中報告

# 高等專題討論二期中報告 ###### tags: `學習` `完成` `高等專題討論二` **Speaker**: `Prof. Szu-Yen Huang 黃斯衍` **Author**: `D06222003` `Kuan-Chia Chiu 邱冠嘉` ## Title ### ==Thermally induced magnonic spin current== ## The content and My Understanding ### Introduction and Motivation From the first industrial revolution, the evolution of human life had been changed a lot by ourselves. Now, we are on the wave of the "Third" industrial revolution, which includes IoT (Internet of Things), AI, big data and digital currency. These are not the end of this wave, it just began. ^[[數字革命](https://zh.wikipedia.org/wiki/%E6%95%B8%E5%AD%97%E5%8C%96%E9%9D%A9%E5%91%BD)] Particularly, for the development of IoT, AI, big data needs tons of large storage units to keep the useful information. Nevertheless, the storage device must response rapidly, having enough endurance of heat and read/write times and non-volatile (not easy to lose the information inside). To confront these difficulties, the scientists are finding some physics states which can preserve information for a relative long time, such as "the magnetization". Different from stored by a voltage, magnetization does not need an extra electronic energy to keep the state. It only needs energy when we read or write the state. Now, they exist in all your memory devices, like MRAMs (magnetoresistive random access memory). However, the techonology of writing and reading state in MRAM has not refined yet. It can be categorized different writing methods: "current-driven", "spin-transfer torque (STT)" and "field-assist spin-transfer torque". The biggest issue is the heat produced by the current. STT effect may be the candidate to reduce the maximum current density of state-writing process, because it use the "spin-polarized" current to change magnetization rather than a magnetic field generated by current. ### Experimental Results Based on the introduction, Prof. Huang is interested in how heat can generated a "pure spin currents" called "spin Seebeck effect (SSE))". A heat gradient can produce spin unbalance in a ferromagnetic material, and be probed by inverse spin Hall effect in a heavy metal (Pt, Pd). Later, I would focus on Prof. Huang's experimental results the past few years. #### Noncollinear magnetization between surface and bulk Y~3~Fe~5~O~12~ Prof. Huang's group studied the magnetic properties of Y~3~Fe~5~O~12~ (a ferromagnetic insulator). This material is charming because it is both ferromagnetic and insulating. And, they found a unusual behavior of magnetoresistance in different geometry and deposition of YIG slabs. This effect is also found in Kerr signal. After they tried different experiments and measurements. They concluded it is because the thickness of YIG changing the magnetization status of both surface and bulk YIG. ^[[Noncollinear magnetization between surface and bulk Y~3~Fe~5~O~12~](https://journals.aps.org/prb/abstract/10.1103/PhysRevB.94.024405)] #### Absence of the Thermal Hall Effect in Anomalous Nernst and Spin Seebeck Effects The anomalous Nernst effect (ANE) is important when it is performed SSE. This effect is similar with anomalous Hall effect that the magnetization will produce a transverse voltage from the spin injection. The difference is ANE with a heat gradient. In this work, they found the thermocouple effect is inevitable in the ANE and SSE configuration. But, how much it affects the SSE signal still needs to test in this project. Using different geometries and different Seebeck coefficients between the electrodes and wires to check. And, they found the thermocouple effect can be neglect in this kind of measurements. ^[[Absence of the Thermal Hall Effect in Anomalous Nernst and Spin Seebeck Effects](https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.117.247201)] #### Enhancement of the anomalous Nernst effect in ferromagnetic thin films Later, they have measured the thickness dependence of the ANE in several ferromagnetic metal, such as Fe, Co, Ni and Permalloy (Py). The sign and magnitude of the ANE show different behavior against the thickness of different ferromagnetic materials. With Studying the thickness of the longitudinal Seebeck coefficient (S~xx~), the found that the anomalous Nernst coefficient $\theta_{ANE}$'s are significantly enhanced with decreasing Their results shew that the spin-polarized current excited by the ANE, which is strongly correlated with the band structure and the Berry curvature around the Fermi level, could be changed dramatically in an ultrathin film. ^[[Enhancement of the anomalous Nernst effect in ferromagnetic thin films](https://journals.aps.org/prb/abstract/10.1103/PhysRevB.96.174406)] ## My experience after this talk Actually, Prof. Huang's study is related to my topic a little bit. My previous topic is studying the behavior of spin current tunneling through a metal oxide barrier. After this talk, I found the logic of Prof. Huang's research is pretty clear. I may review the study of YIG geometry part, which would be useful to me. And, I also learned some attitudes to do researches: Never to ignore something strange in the experimental results, even it affects almost nothing much. ## Introduction Articles - [Emergent phenomena induced by spin–orbit coupling at surfaces and interfaces](https://www.nature.com/articles/nature19820) - [Self-consistent determination of spin Hall angles in selected 5d metals by thermal spin injection](https://journals.aps.org/prb/abstract/10.1103/PhysRevB.89.140407)