• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.35 no.2
• /
• pp.134-142
• /
• 2022

As industry and technology go through advancement, it is hard to search new materials which satisfy various standards through conventional trial-and-error based research methods. Crystal Graph Convolutional Neural Network(CGCNN) is a neural network which uses material's features as train data, and predicts the material properties(formation energy, bandgap, etc.) much faster than first-principles calculation. This report introduces how to train the CGCNN model which predicts the formation energy using open database. It is anticipated that with a simple programming skill, readers could construct a model using their data and purpose. Developing machine learning model for materials science is going to help researchers who should explore large chemical and structural space to discover materials efficiently.

• International Journal of Automotive Technology
• /
• v.4 no.2
• /
• pp.77-86
• /
• 2003

How to compare the environmental damage caused by vehicles with different foe]s and drive trains\ulcorner This paper describes a methodology to assess the environmental impact of vehicles, using different approaches, and evaluating their benefits and limitations. Rating systems are analysed as tools to compare the environmental impact of vehicles, allowing decision makers to dedicate their financial and non-financial policies and support measures in function of the ecological damage. The paper is based on the "Clean Vehicles" research project, commissioned by the Brussels Capital Region via the BIM-IBGE (Brussels Institute for the Conservation of the Environment) (Van Mierlo et at., 2001). The VriJe Universiteit Brussel (ETEC) and the universite Libre do Bruxelles (CEESE) have jointly carried out the workprogramme. The most important results of this project are illustrated in this paper. First an overview of environmental, economical and technical characteristics of the different alternative fuels and drive trains is given. Afterward the basic principles to identify the environmental impact of cars are described. An outline of the considered emissions and their environmental impact leads to the definition of the calculation method, named Ecoscore. A rather simple and pragmatic approach would be stating that all alternative fuelled vehicles (LPG, CNG, EV, HEV, etc.) can be considered as ′clean′. Another basic approach is considering as ′clean′ all vehicles satisfying a stringent omission regulation like EURO IV or EEV. Such approaches however don′t tell anything about the real environmental damage of the vehicles. In the paper we describe "how should the environmental impact of vehicles be defined\ulcorner", including parameters affecting the emissions of vehicles and their influence on human beings and on the environment and "how could it be defined \ulcorner", taking into account the availability of accurate and reliable data. We take into account different damages (acid rain, photochemical air pollution, global warming. noise, etc.) and their impacts on several receptors like human beings (e.g., cancer, respiratory diseases, etc), ecosystems, or buildings. The presented methodology is based on a kind of Life Cycle Assessment (LCA) in which the contribution of all emissions to a certain damage are considered (e.g. using Exposure-Response damage function). The emissions will include oil extraction, transportation refinery, electricity production, distribution, (Well-to-Wheel approach), as well as the emission due to the production, use and dismantling of the vehicle (Cradle-to-Grave approach). The different damages will be normalized to be able to make a comparison. Hence a reference value (determined by the reference vehicle chosen) will be defined as a target value (the normalized value will thus measure a kind of Distance to Target). The contribution of the different normalized damages to a single value "Ecoscore" will be based on a panel weighting method. Some examples of the calculation of the Ecoscore for different alternative fuels and drive trains will be calculated as an illustration of the methodology.

• Journal of the Korean Electrochemical Society
• /
• v.16 no.3
• /
• pp.169-176
• /
• 2013

Solid state lithium oxide compounds of layered structure, which has high stability of structure, are mainly used as the cathode materials in lithium-ion batteries (LIBs). Recently, the investigation of Solid Electrolyte Interphase (SEI) between active materials and electrolyte has been focusing to improve the performance of lithium-ion batteries. For the investigation of the SEI, the study of surface properties of cathode materials and anode materials is also required in advance. $LiNiO_2$ and $LiCoO_2$ are very similar layered structure of cathode active materials and representative solid state lithium oxide compounds in LIBs. Various experimental and theoretical studies have been doing for $LiCoO_2$. The theoretical investigation of $LiNiO_2$ is not sufficient, however, even if experimental studies of $LiNiO_2$ are enough. In this study, the surface energies of nine facets of $LiNiO_2$ crystal facets were calculated by Density Functional Theory. In XRD data of $LiNiO_2$, (003), (104), (101), et al. facets are main surfaces in order. However, the results of calculation are different with XRD data. Thus, both (104) and (101) facets, which are energetically stable and measured in XRD, are mainly exposed in the surface of $LiNiO_2$ and it is expected that intercalation and de-intercalation of Li-ion will be affected by them.

• Journal of the Korean Magnetics Society
• /
• v.26 no.6
• /
• pp.179-184
• /
• 2016

For an industrial spin-transfer torque (STT) MRAM, low switching current and high thermal stability are required, simultaneously. For this point of view, it is essential to find magnetic materials which satisfy high spin polarization and strong perpendicular magnetocrystalline anisotropy (MCA). In this paper, we investigate electronic structures and MCA energies of perovskite $CoFeX_3$ (X = O, F, S, Cl). For X = F and Cl, spin polarization at the Fermi level are 97 % and 96 %, respectively, which are close to a half metal. Furthermore, Co-terminated 5-monolayer (ML) $CoFeX_3$ (X = O, F, S, Cl) films show perpendicular MCA. In particular, the MCA energy of the Co-terminated $CoFeCl_3$ is about 1.0 meV/cell which is three times larger than that of a 5-ML CoFe film. Therefore, we expect to realize a magnetic material with high spin polarization and strong perpendicular MCA energy by utilizing group 6 and 7 elements in the periodic table, and to contribute to commercializing of the STT-MRAM.

• Journal of the Korean Magnetics Society
• /
• v.19 no.5
• /
• pp.165-170
• /
• 2009

It is well-known that a meta-stable fcc bulk Fe has an antiferromagnetic (AFM) ground state and could be synthesized by growing Fe on a proper fcc metal substrate. In this study magnetism of Fe monolayers on nonmagnetic fcc transition metal (Cu, Rh, Pd, and Ag) (001) surfaces has been investigated using the all-electron full-potential linearized augmented plane wave method. The Fe monolayers on Rh(001) and Pd(001) surfaces were calculated to be stabilized in an AFM state, whereas the Fe monlayers on Cu(001) and Ag(001) surfaces are stabilized in a ferromagnetic (FM) state. Noting that Cu and Ag have the smallest and largest lattice constants and the fcc bulk Fe with a larger lattice constant is getting stabilized in a ferromagnetic state, it is unexpectable and interesting. The calculated magnetic moments of the Fe atoms on Cu, Rh, Pd, and Ag(001) surfaces are 2.811, 2.945, 2.987, and 2.990 $_{{\mu}B}$ in FM states and 2.624, 2.879, 2.922, and 3.001 $_{{\mu}B}$ in AFM states.

• Journal of the Korean Magnetics Society
• /
• v.18 no.5
• /
• pp.163-167
• /
• 2008

The electronic structure and magnetism of superlattice systems consisted of Heusler compound $Co_2MnSi$ (CMS) and zinc-blende MnAs (MA) are investigated by means of the all-electron full potential linearized augmented plane wave method within the generalized gradient approximation. Four superlattice systems are considered, that is CMS(m)/MA(n), where m and n, being either 2 or 4, denote the number of alternatingly arrayed layers of the compounds in a superlattice along [001] direction. From the calculated total magnetic moments as well as the total density of states, it is found that neither of the four systems is half-metallic. It is also found that the Mn atoms are antiferromagnetically coupled in the systems of CMS2/MA2 and CMS2/MA4. The total and atom-resolved density of states of the four superlattices are compared with those of the bulk $Co_2MnSi$ and MnAs, and the influences of the change in the systems symmetry on the magnetism and half-metallicity are discussed.

• Journal of the Korean Magnetics Society
• /
• v.25 no.6
• /
• pp.175-179
• /
• 2015

The half-metallicity and magnetism of the (001) surfaces of $(AlP)_1/(CrP)_1$ superlattice were investigated by means of FLAPW (Full-potential Liniarized Augmented Plane Wave) method. We considered four types of (001) surface termination, i.e., Al(S)-, Cr(S)-, P(S)Al(S-1)- and P(S)Cr(S-1)-term systems. We found that only Cr(S)-term system maintains the half-metallicity at the surface as only this system has the calculated magnetic moment of integer number of bohr magnetons. The magnetic moment of Cr(S) atom in the system was $3.02{\mu}_B$ which was increased from the bulk value by the effects of band narrowing and increased spin-splitting at the surface. The electronic density of states of the P(S) atom in the P(S)Al(S-1)-term showed very sharp surface states due to the broken $p_z$ bonds at the surface. We found there is still a strong p-d hybridization between the P(S) and Cr(S-1) layers in the P(S)Cr(S-1)-term which causes a considerable increase of magnetic moment of P(S) atom.

• Journal of Elementary Mathematics Education in Korea
• /
• v.13 no.2
• /
• pp.285-304
• /
• 2009

The operations of fractions are the main contents of number and operations in the elementary mathematics curriculum. They are also difficult for students to understand conceptually. Nevertheless, there has been little study on the addition and subtraction of fractions. Given this, this paper explored the connection between the national mathematics curriculum and its concomitant textbooks, the adequacy of when to teach, and the method of constructing each unit to teach addition and subtraction of fractions. This paper then analyzed elementary mathematics textbooks and workbooks by three parts aligned with the general instructional flow: 'introduction', 'activity', and, 'exercise'. First, it was analyzed with regard to the introduction part whether the word problems of textbooks might reflect on students' daily lives as intended, how different meanings of operations would be expected to be taught, and how the subsequent activities were connected with the original word problems. Second, the main analysis of activity part of the textbooks dealt with how to use concrete or iconic models to promote students' conceptual understanding of operations and how to formalize the calculation methods and principles with regard to addition and subtraction of fractions. Third, the analysis of the part of exercise in the textbooks and workbooks was conducted with regard to problem types and meanings of operations. It is expected that the issues and suggestions stemming from this analysis of current textbooks and workbooks are informative in developing new instructional materials aligned to the recently revised mathematics curriculum.

• Journal of Science Education
• /
• v.44 no.3
• /
• pp.331-344
• /
• 2020

This study aims to find meaningful implications for the development of Korean elementary school math education courses and textbooks by comparing and analyzing the number and arithmetic areas of Korean and Canadian math textbooks in fifth and sixth grades. To this end, the textbook composition system of Korean and Canadian elementary schools was compared and analyzed, and the number and timing of introduction of math textbooks and math textbooks by grade, and the number in fifth and sixth grade and the learning contents of math textbooks were compared and analyzed. The following conclusions were obtained from this study: First, it is necessary to organize a textbook that can solve the problem in an integrated way by introducing the learned mathematical concepts and computations naturally in the context of problems closely related to real life, regardless of the type of private calculation or mathematics area. Second, it is necessary to organize questions using materials such as real photography and mathematics, science, technology, engineering, art, etc. and to organize textbooks that make people feel the necessity and usefulness of mathematics. Third, sufficient learning of the principles of mathematics through the use of various actual teaching aids and mathematical models, and the construction of textbooks focusing on problem-solving strategies using engineering tools are needed. Fourth, in-depth discussions are needed on the timing of learning guidance for fractions and minority learning or how to organize and develop learning content.

• Korean Journal of Materials Research
• /
• v.31 no.12
• /
• pp.682-689
• /
• 2021

NbC, HfC, TaC, and their solid solution ceramics have been identified as the best materials for ultrahigh-temperature ceramics. However, their structural stability and elastic properties are mostly unclear. Thus, we investigated structure and elastic properties of (Nb_1-xTa_x)C and (Nb_1-xHf_x)C solid solutions via ab initio calculations. Our calculated results show that the stability of (Nb_1-xTa_x)C and (Nb_1-xHf_x)C increases with the increase of Hf and Ta content, and (Nb_1-xHf_x)C is more stable than (Nb_1-xTa_x)C at the same content of Hf and Ta. The lattice constants decrease with increasing of Hf and Ta content. (Nb_1-xTa_x)C and (Nb_1-xHf_x)C carbides are mechanically stable and brittle. Bulk modulus of (Nb_1-xTa_x)C increases with increasing Ta content. In contrast, bulk modulus of (Nb_1-xHf_x)C decreases with increasing Hf content. Hardness of solid solutions shows the highest values at the (Nb_0.25Ta_0.75)C and (Nb_0.75Hf_0.25)C. In particular, (Nb_0.75Hf_0.25)C shows the highest hardness for the current system. The results indicate that the overall mechanical properties of (Nb_1-xHf_x)C solid solutions are superior to those of (Nb_1-xTa_x)C solid solutions. Therefore, controlling the Hf and Ta element and content of the (Nb_1-xTa_x)C and (Nb_1-xHf_x)C Solid solution is crucial for optimizing the material properties.