Summaries of research and development activities by using supercomputer system of JAEA in FY2018 (April 1, 2018 - March 31, 2019)

HPC Technology Promotion Office*

JAEA-Review 2019-017, 182 Pages, 2020/01

JAEA-Review-2019-017.pdf:11.11MB

Japan Atomic Energy Agency (JAEA) conducts research and development (R&D) in various fields related to nuclear power as a comprehensive institution of nuclear energy R&Ds, and utilizes computational science and technology in many activities. As shown in the fact that about 20 percent of papers published by JAEA are concerned with R&D using computational science, the supercomputer system of JAEA has become an important infrastructure to support computational science and technology. In FY2018, the system was used for R&D aiming to restore Fukushima (environmental recovery and nuclear installation decommissioning) as a priority issue, as well as for JAEA's major projects such as research and development of fast reactor cycle technology, research for safety improvement in the field of nuclear energy, and basic nuclear science and engineering research. This report presents a great number of R&D results accomplished by using the system in FY2018, as well as user support, operational records and overviews of the system, and so on.

Stable structure of hydrogen atoms trapped in tungsten divacancy

Osawa, Kazuhito*; Toyama, Takeshi*; Hatano, Yuji*; Yamaguchi, Masatake; Watanabe, Hideo*

Journal of Nuclear Materials, 527, p.151825_1 - 151825_7, 2019/12

https://doi.org/10.1016/j.jnucmat.2019.151825

no abstracts in English

Seven-year temporal variation of cesium-137 discharge inventory from the port of Fukushima Dai-ichi Nuclear Power Plant; Continuous monthly estimation of cesium-137 discharge in the period from April 2011 to June 2018

Machida, Masahiko; Yamada, Susumu; Iwata, Ayako; Otosaka, Shigeyoshi; Kobayashi, Takuya; Watanabe, Masahisa; Funasaka, Hideyuki; Morita, Takami*

Nippon Genshiryoku Gakkai Wabun Rombunshi, 18(4), p.226 - 236, 2019/12

https://doi.org/10.3327/taesj.J18.030

After direct discharges of highly-contaminated water from Unit 2 and 3 in Fukushima Daiichi Nuclear Power Plant (1F) in April to May 2011, Kanda suggested that relatively small run-off of radionuclides from 1F port into Fukushima coastal region has subsequently continued by using his estimation scheme. However, the estimation period was limited until September 2012, and there has been no report on the issue since the work. Therefore, this paper focuses on discharge inventory from 1F port until June 2018. In the missing period, the central government and Tokyo Electric Power Company Holdings have done continuous efforts to stop the discharge, and consequently sea water concentration inside 1F port has diminished gradually. We show monthly discharge inventory of Cs-137 until June 2018 by two schemes, i.e., Kanda's scheme partially improved by authors and more sophisticated one using Voronoi tessellation reflecting the increment of the number of monitoring points inside 1F port. The results show that the former always presents overestimated results compared to the latter but the ratio of former to latter is less than one order. Based on these results, we evaluate impact of discharge inventory from 1F port into the coastal area and radiation does via fish digestion.

Isotope and plasma size scaling in ion temperature gradient driven turbulence

Idomura, Yasuhiro

Physics of Plasmas, 26(12), p.120703_1 - 120703_5, 2019/12

https://doi.org/10.1063/1.5133084

This Letter presents the impacts of the hydrogen isotope mass and the normalized gyroradius on L-mode like hydrogen (H) and deuterium (D) plasmas dominated by ion temperature gradient driven (ITG) turbulence using global full-f gyrokinetic simulations. In ion heated numerical experiments with adiabatic electrons, the energy confinement time shows almost no isotope mass dependency, and is determined by Bohm like scaling. Electron heated numerical experiments with kinetic electrons show clear isotope mass dependency caused by the isotope effect on the collisional energy transfer from electrons to ions, and the H and D plasmas show similar ion and electron temperature profiles at an H to D heating power ratio of . The normalized collisionless ion gyrokinetic equations for H and D plasmas become identical at the same , and collisions weakly affect ITG turbulence. Therefore, the isotope mass dependency is mainly contributed by the scaling and the heating sources.

GPU acceleration of communication avoiding Chebyshev basis conjugate gradient solver for multiphase CFD simulations

Ali, Y.*; Onodera, Naoyuki; Idomura, Yasuhiro; Ina, Takuya*; Imamura, Toshiyuki*

Proceedings of 10th Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems (ScalA 2019), p.1 - 8, 2019/11

https://doi.org/10.1109/ScalA49573.2019.00006

Iterative methods for solving large linear systems are common parts of computational fluid dynamics (CFD) codes. The Preconditioned Conjugate Gradient (P-CG) method is one of the most widely used iterative methods. However, in the P-CG method, global collective communication is a crucial bottleneck especially on accelerated computing platforms. To resolve this issue, communication avoiding (CA) variants of the P-CG method are becoming increasingly important. In this paper, the P-CG and Preconditioned Chebyshev Basis CA CG (P-CBCG) solvers in the multiphase CFD code JUPITER are ported to the latest V100 GPUs. All GPU kernels are highly optimized to achieve about 90% of the roofline performance, the block Jacobi preconditioner is re-designed to extract high computing power of GPUs, and the remaining bottleneck of halo data communication is avoided by overlapping communication and computation. The overall performance of the P-CG and P-CBCG solvers is determined by the competition between the CA properties of the global collective communication and the halo data communication, indicating an importance of the inter-node interconnect bandwidth per GPU. The developed GPU solvers are accelerated up to 2x compared with the former CPU solvers on KNLs, and excellent strong scaling is achieved up to 7,680 GPUs on the Summit.

Research and development behind a computation system for 3D distributions of air dose rates in the environment; Estimating environmental radiation doses using PHITS together with remote sensing data

Kim, M.; Malins, A.; Sakuma, Kazuyuki; Kitamura, Akihiro; Machida, Masahiko; Hasegawa, Yukihiro*; Yanagi, Hideaki*

Isotope News, (765), p.30 - 33, 2019/10

https://www.jrias.or.jp/books/cat3/2019/765.html

Here we outline a system for generating three dimensional models of urban and rural areas in Fukushima Prefecture. The Cs and Cs radioactivity distribution can be set flexibly across the different components of the model. The models incorporate realistic representations of local buildings, individual conifer and broadleaf trees, and the topography of the land surface. The system is demonstrated by modelling a suburban area 4 km from the Fukushima Daiichi Nuclear Power Plant that has yet to be decontaminated. Air dose rates calculated in PHITS were correlated with measurements taken across the site in a car-borne survey.

Quantum chemical calculations for the norbadione A complexes with Cs, K, and Na in gas and aqueous phases

Suno, Hiroya; Machida, Masahiko

Chemical Physics Letters, 730, p.26 - 31, 2019/09

https://doi.org/10.1016/j.cplett.2019.05.012

We perform quantum chemical calculations for the Cs, K, and Na complexes of norbadione A (NBA), a pigment molecule in mushrooms known to accumulate Cs. A numerical two-step approach, by Ota , is employed to examine its alkali-metal-cation complexation selectivity in aqueous solutions. Applying it to the neutral, di- and tetra-deprotonated NBAs, we confirm that the complexation selectivity on Cs emerges only in high pHs, in which the di-protonated NBA dominates, in agreement with experimental results. This is the first demonstration of the approach for a biological molecule whose selectivity is known to be anomalous.

First-principles calculation of mechanical properties of simulated debris ZrUO

Itakura, Mitsuhiro; Nakamura, Hiroki; Kitagaki, Toru; Hoshino, Takanori; Machida, Masahiko

Journal of Nuclear Science and Technology, 56(9-10), p.915 - 921, 2019/09

https://doi.org/10.1080/00223131.2019.1604271

To elucidate the mechanical properties of fuel debris inside the Fukushima Daiichi Nuclear Power Plant, we use first-principles calculations to evaluate mechanical properties of cubic ZrUO, which is a main component of the fuel debris. We focus on the dependence of mechanical properties on the fraction x of zirconium, compare our results with recent experiment of simulated debris, in which dependences of elastic moduli and fracture toughness on the ZrO content showed deviation from a simple linear relation. We show that elastic moduli drop at around x=0.25 and increase again for larger values of x, as has been observed in experiments. The reason of the drop is a softening owing to disordered atomistic structures induced by the solute zirconium atoms. We also find that stress-strain curves for the x=0.125 case show marked hysteresis owing to the existence of many meta-stable states. We show that this hysteresis leads to slightly increased fracture toughness, but it is not enough to account for the significant increase of fracture toughness observed in experiments.

Analyzing the cross slip motion of screw dislocations at finite temperatures in body-centered-cubic metals; Molecular statics and dynamics studies

Suzudo, Tomoaki; Onitsuka, Takashi*; Fukumoto, Kenichi*

Modelling and Simulation in Materials Science and Engineering, 27(6), p.064001_1 - 064001_15, 2019/08

https://doi.org/10.1088/1361-651X/ab235e

Plasticity of body-centered-cubic (BCC) metals at low temperatures is determined by screw dislocation kinetics. Because the core of screw dislocation in these metals has non-planar structure, its motion is complex and unpredictable. For example, although density functional theory (DFT) predicts slip on a { 110 } plane, the actual slip plane at elevated temperatures departs from the prediction, its mechanism having been a mystery for decades. Here we conduct a series of molecular dynamics simulations to track the screw dislocation motion and successfully reproduced the transition of the slip plane. We then devised an algorithm to scrutinize the activation of dislocation jump over the Peierls barrier and discovered the possible origin of this unexpected phenomenon, i.e., a large fluctuation leads to the kink-pair nucleation for the cross-slip jump without transition of dislocation core structure.

Characterisation of structural similarities of precipitates in Mg-Zn and Al-Zn-Mg alloys systems

Bendo, A.*; Maeda, Tomoyoshi*; Matsuda, Kenji*; Lervik, A.*; Holmestad, R.*; Marioara, C. D.*; Nishimura, Katsuhiko*; Nunomura, Norio*; Toda, Hiroyuki*; Yamaguchi, Masatake; et al.

Philosophical Magazine, 99(21), p.2619 - 2635, 2019/07

https://doi.org/10.1080/14786435.2019.1637032