To further extend the applicability of first-principles electronic structure calculations based on density functional theory (DFT) to large-scale systems containing more than ten thousands of atoms, here we present development of low-order scaling DFT methods: one is a numerically exact one, the other is approximate O(N) methods. Though the conventional DFT calculations based on semi-local functionals scale as the third power of number of atoms, it will be shown that the computational complexity of DFT calculations can be reduced to a low-order scaling in a numerically exact sense [1,2]. We further discuss an efficient O(N) divide-conquer (DC) method based on localized natural orbitals (LNOs) for large-scale DFT calculations of gapped and metallic systems [3], where the LNOs are noniteratively calculated by a low-rank approximation via a local eigendecomposition of a projection operator for the occupied space. In addition to the low-order scaling methods, efficient parallelization methods for massively parallel computers will be presented for atom decompositions [4] and fast Fourier transforms [5,6].
References:Quasi-2D atomically thin materials display a number of unique properties not found in their bulk counterparts, such as large self-energy and excitonic effects due to the quantum confinement and reduced screening with layer number close to the 2D limit. These atomically thin layer structures demonstrate rich physics and pave the way for emerging fields, such as excitonics and valleytronics, with great potential for applications in next-generation devices. To probe the dimensionality effects, we use ab initio GW+BSE methods based on many-body perturbation theory (MBPT) to explain and predict the quasiparticle and optical properties of potential quasi-2D semiconductors: monolayer group VI monochalcogenides (Ge,Sn/S,Se) and transition metal dichalcogenides (MoSe2 and Janus MoSSe). Significant exciton binding energy, layer-controlled bandgap, anisotropic optical response [1], and possible valley polarization [2] serve as a convenient and efficient method for engineering the excited-state properties of quasi-2D systems.
References:The studies of dispersion-less bands revealed in electronic and photonic systems have caught great attention recently. Many exotic quantum phenomena, for example, the high-transition-temperature superconductivity associated with the infinitely large density of states of the flat bands, are proposed. In this talk, I will begin with an introduction to the flat bands using Wannier functions. Then I will introduce three tight-binding models, namely the Lieb, kagome, and checkerboard lattices, by considering only the nearest-neighbor hopping parameters and demonstrate that the recognized flat bands associated with the three lattices can be ideally embedded into new structures, respectively [1]. Finally, I will provide several examples for the appearance of nearly flat bands realized in two-dimensional materials with long-range hopping beyond the simplified tight-binding models based on our first-principles calculations for the systems composed of Ge atoms.
Our study clearly demonstrates that the flat bands given by the well-known lattices, namely the Lieb, kagome, and checkerboard lattices, can be ideally embedded into the new structures that cannot be recognized as the original ones. Therefore, the amount of materials that can give interesting flat-band physics could be much larger.
この十数年で2次元物質はその多様な物性と幅広い応用可能性から多くの注目を集めている。特に近年の剥離・合成技術の発展により、1T(trigonal monolayer) や 1H (hexagonal monolayer)といった特徴的な構造を持つAB2型の遷移金属ダイカルコゲナイド(e.g. MoS2) やMXenes (遷移金属カーバイド/ナイトライド) などに関する研究が多数報告されている。多くのAB2型単層物質が既に報告されているが[1]、AB2型の化合物だけであっても元素の組み合わせは膨大であるため、まだまだ計算物質科学の立場から未知の2次元物質を探索する余地が十分にあると考えられる。そこで本研究ではOpenMX[2]を用いた網羅的DFT計算によりAB2型2次元物質の構造マップを作成した[3,4]。具体的には、希ガスやランタノイドやアクチノイドを除いた62元素の組み合わせからなる3844化合物(=62元素×62元素)に対して初期構造として1T・1H・planarの3種類を用意し、2×2のスーパーセルに対して構造最適化を行い、得られた構造を空間群で分類することにより構造マップを作成した。また、この構造マップを用いて周期表の族の組み合わせごとに1T/1H構造を持つ化合物の数をまとめることにより、1T/1H構造になりやすい族の組み合わせを明らかにした。さらに、原子サイズのバイナリ記憶素子として機能すると考えられる「メモリ構造」や、その他特徴的な未知のAB2型2次元物質が構造最適化により得られた。
このような構造マップは、未知の2次元物質の構造探索に新しい視点を与え、さらなる実験的な構造探索を促進するとともに新規物性の探索にも非常に有用であると考えられる。
Metallic nanoparticles are widely used for technological applications in catalysis, data storage and solar energy. However, the performance of nanoparticles is usually determined by the shape of nanoparticles. Therefore, understanding the morphology and composition of the metallic nanoparticles changed by environment is important. In this presentation, we will discuss the external factors, such as adsorbed molecules and substrate material, on nanoparticle using density functional theory (DFT) calculations. First, we present the morphology changing of L10 ordered FePt epitaxial growth on Mg(1-x)TixO substrates [1]. Second, we demonstrate the investigation on Ti nanoparticles oxidation, strain and oxygen penetration [2]. Next, we investigate the atomic arrangement of TiPt nanoparticles under different oxygen adsorption [3]. Finally, we study the strong metal-support interaction (SMSI) between Au nanoparticles and ZnO substrates and partly explain the enhanced catalytic reaction (CO oxidation) by the ZnO encapsulation. The investigations show computational calculations can be used to model modification of nanoparticles by adsorbed molecules or supports, and study the properties changing, such as morphology, energy barrier, atomic arrangement and catalytic performance. In summary, the study demonstrates the functional characteristics of nanoparticles highly depend on their nanostructures.
References:日程: | (July 2nd-6th) The Summer School on DFT: Theories and Practical Aspects |
(July 9th and 10th) The 3rd OpenMX developer's meeting (July 11th and 12th) Advanced Lecture Series |
|
場所: | 物性研究所本館6階 |
![]() |