座長：久保田 樹、小林 優斗
|氏名： 上野 祥広
指導教員名： 鹿野田 一司 教授
発表題目（英語）： NMR study of doped spin liquid material κ-(ET)4Hg2.89Br8
要旨（英語）： κ-(ET)4Hg2.89Br8 is a layered material, in which conducting ET layers and insulating Hg2.89Br8 layers are alternately stacked. Due to incommensurability between the ET and Hg sublattices, the ET layer with a half-filled band is 11% hole-doped and is metallic. In the conduction layer, the ET molecules form dimers, which constitute a triangular lattice. Since the spin susceptibility of κ-(ET)4Hg2.89Br8 is similar to that of the spin-liquid Mott insulator, κ-(ET)2Cu2(CN)3, with a triangular lattice, κ-(ET)4Hg2.89Br8 is considered to be a "doped spin liquid" in which the spin degree of freedom is in a spin-liquid state even though it is a metal. In the present study, I have performed 13C NMR measurements to investigate the magnetism of this material in detail and obtained the following results.
First, I confirmed that the NMR spectral shift (Knight shift), which measures the magnitude of the local magnetic field created by the electron spins, is scaled to the spin susceptibility and found that the spectrum is widened with decreasing temperature, suggesting an increasing inhomogeneity of the local spin susceptibility upon cooling.
Second, the nuclear spin-lattice relaxation rate 1/T1, which measures the intensity of electron spin fluctuations, showed a temperature dependence similar to in the spin-liquid compound κ-(ET)2Cu2(CN)3. A previous study found both materials to behave similar in spin susceptibility; thus, κ-(ET)4Hg2.89Br8 is regarded as a doped spin liquid hosting both mobile carriers and a quantum spin liquid.
Furthermore, a decreases in the Knight shift and a cubic temperature dependence of 1/T1 below 10 K suggest that superconductivity emerging in this material has spin-singlet d-wave symmetry with highly enhanced fluctuations above Tc.
|氏名： 大西 由吾
指導教員名： 永長 直人 教授
発表題目（英語）： Feedback-type thermoelectric effect in correlated solids
要旨（英語）： The second law of thermodynamics is seemingly violated when feedback is applied at a microscopic level. This paradox is now resolved by considering the entropy when the information is erased at the memory. Several systems including this Maxwell’s demon have been proposed theoretically to improve performances of heat engines, mainly for microscopic or mesoscopic systems.
Meanwhile, in the field of materials science, materials that exhibit good thermoelectric performances have been explored for a long time and many strategies or mechanisms to realize efficient thermoelectric materials are proposed theoretically. However, the role of information or feedback in the thermoelectric effect in solid is rarely discussed.
In this presentation, I will talk about a new mechanism of thermoelectric effect inspired by Maxwell's demon proposed in our work . We formulate a specific model for the thermoelectric effect in solid in the framework of stochastic thermodynamics and calculate its response. The result shows that the figure of merit ZT, which represents the efficiency of the thermoelectric material, can be relatively high, i.e., of the order of 1, in this model within a range of realistic parameters. This result suggests that the role of feedback is important to realize high-ZT materials.
|氏名： 海老原 豪
指導教員名： 吉岡 孝高 准教授
発表題目（英語）： Introduction to dual-comb spectroscopy and its application to observing molecules
要旨（英語）： In this presentation, I will introduce dual-comb spectroscopy comparing with a conventional spectroscopic method and give two examples.
The optical frequency comb was initially developed as a ruler for optical frequencies whose spectrum consists of evenly spaced narrow lines. Recently it has begun to be applied as a new tool of spectroscopy. DCS(dual-comb spectroscopy) allows us to directly observe optical frequencies as beat signals in the RF frequency domain when two optical frequency combs with slightly different repetition rates interfere with each other. Because it eliminates moving parts, DCS provides a much higher spectral and temporal resolution than conventional methods.
I will present two examples of applications: (i) Time-resolved DCS of molecules to measure the relaxation time of the transition on the order of 10 ns, and (ii) parallel detection of 22 trace molecular species in a gas mixture with part-per-billion sensitivity and sub-Doppler resolution.
|氏名： 荻原 琢磨
指導教員名： 山地 洋平 特任准教授
発表題目（英語）： Efficient real-frequency solver for dynamical mean-field theory
要旨（英語）： Dynamical mean field theory(DMFT) is a powerful method for strongly correlated system, which maps a correlated lattice model onto a quantum impurity model. Therefore, we can solve strongly correlated systems with DMFT and how to how to solve an Anderson impurity model. Since the introduction of DMFT, there has been an enormous development on impurity solver. Several method for impurity solver are available. Numerical renormalization group(NRG) theory and density renormalization group theory are solutions for the one-site Hubbard model. However, These methods are hard to apply to situation with multiple interacting orbitals or sites. Then, Hirsch Fye QMC and continuous time (CT) QMC method can be applied to some extensions including several correlated fermions, but have problems with low symmetry systems. Another draw back of QMC is the use of imaginary frequencies, which leads to an ill-conditioned inversion problem. Exact digonalization (ED) methods are implemented using real frequencies. This problem with ED method is that bath has to be represented by small number of discrete state in order to keep the exponentially growing many-body Hilbert space.
In this presentation, I will intoroduce how to implement the DMFT loop on real-frequency repsesentations of the Green's function and how to solve the Anderson impurity model using ED including several hundred bath sites.
 A. Georges, G. Kotliar, W. Krauth, and M. J. Rozenberg, Rev. Mod. Phys. 68, 13 (1996).
 Y. Lu, M. Höppner, O. Gunnarsson, and M. W. Haverkort, Phys. Rev. B 90, 085102 (2014).
|氏名： 加藤 啓輔
指導教員名： 中村 泰信 教授
発表題目（英語）： Nuclear magnetic resonance (NMR) has prospered as one of the probes in material science for decades. NMR signals are usually weak, hence this method has been applied only to bulk or liquid form materials consisting of a sufficient number of nuclei.
Recently, thin-film materials such as graphene and van der Waals materials have attracted research interests due to their novel physical properties. Because of their dimensionality, however, it has been unable to apply NMR spectroscopy to these novel thin materials, so the measurement method for them is basically limited to optical spectroscopy like reflectance or photoluminescence spectroscopy.
There are two approaches for detecting weak NMR signals. One is improving the signal-to-noise ratio and recently a new optical method was proposed. The other is increasing signal through aligning nuclear spin on the use of spin transfers between electron and nuclei. This technique is called dynamical nuclear polarization.
In this presentation, I briefly explain the dynamical nuclear polarization of some kind of thin-film material and experiments under preparation.
|氏名： 越智 友崇
指導教員名： 長田 俊人 准教授
発表題目（英語）： Low temperature thermoelectric measurement of 2D materials
要旨（英語）： The thermoelectric effect as typified by the Seebeck effect is attracting attention in modern society from the point of view of promoting the utilization of waste heat. Now, this interest is shifting from 3D materials to 2D materials because of the miniaturization of modules. Therefore, it is very important to understand the fundamental properties of the thermoelectric effect in 2D materials.
The twisted bilayer graphene constitutes a flat band of the so-called pudding mold at a particular magic angle. The steep rise of these bands is expected to be accompanied by a strong thermoelectricity. In fact, it is reported that the thermoelectricity is enhanced in the bilayer graphene whose gap is opened by the application of top gate and back gate.
However, low-temperature thermoelectric measurements of atomic layer thin films have several difficulties. There are two major difficulties. The first one is the complexity of the design and fabrication process. This measurement needs a heater, two thermometers, four terminals for measurement, top gate and back gate. This makes it extremely difficult in the design, and the number of making processes will increase. Secondly, the diffusion and conduction of heat outside the sample occur. Ideally, the thermoelectricity measurement should be performed in a vacuum with the sample suspended in midair, but in reality, the sample contacts with the substrate. Therefore, heat diffusion and conduction to the substrate are inevitable. To reduce this effect, it is effective to use a material with low thermal conductivity as a substrate. However, it is more important to select a material which is convenient for the sample preparation than the thermal ideal substrate in the experiment.
The thermoelectricity measurement of the atomic layer is accompanied by such difficulties as described above. In order to tackle this problem, as a first step, I tried low temperature thermoelectric measurements of milli-scale bulk graphite to reproduce previous study. In this presentation, the result of the experiment and future prospects are discussed.
 K. Kuroki and R. Arita, J. Phys. Soc. Jpn. *76*, 083707 (2007)
 C. R. Wang *et al.,* Phys. Rev. Lett. *107*, 186602 (2011)
 Z. Zhu *et al., *Nat. Phys. *6*, 26–29 (2010)