著者

藤 正督 ラザヴィ ホソロシャヒ ハディ 高井 千加 佐藤 知広 尾畑 成造 立石 賢司

出版者

一般社団法人 粉体粉末冶金協会

雑誌

粉体および粉末冶金 (ISSN:05328799)

巻号頁・発行日

vol.65, no.10, pp.609-615, 2018-10-15 (Released:2018-11-08)

参考文献数

12

被引用文献数

2 2

---

Sintering of most ceramic and composite materials are only possible at temperatures over 1000°C due to their high melting point. Sintering is acknowledged as an expensive process, which takes several hours to several days. In addition, high temperature sintering affects the final product by causing undesired grain coarsening or changing the initial chemical stoichiometry. Our research group has proposed a “non-firing sintering”, where no firing process is required for achieving high densities. The underlying idea of this method involves the chemical activation of powder surface via ball milling, where the surface of particles is rubbed against balls. In this review, we will introduce the mechanism of the method as well as some process know-how, with some examples of preparing solidified bodies of silicon carbide, composite of carbon nanotube (CNT) and silica, and organic/inorganic composite materials.

著者

佐藤 知広 久保田 敦斗 齋藤 賢一 藤 正督 高井 千加 瀬名 ハディ 宅間 正則 高橋 可昌

出版者

公益社団法人 日本材料学会

雑誌

材料 (ISSN:05145163)

巻号頁・発行日

vol.71, no.2, pp.167-174, 2022-02-15 (Released:2022-02-20)

参考文献数

14

被引用文献数

1

---

The industrial manufacturing methods for ceramics are powder mixing, molding, and firing. Ceramics are fired at a higher temperature than metal sintering. For this reason, in the ceramics manufacturing process, a large amount of energy is consumed, and a large amount of carbon dioxide is also emitted, especially in the firing process. Therefore, attention is focused on the non-firing solidification process of ceramics. In this method, after the molding process, there is a solidification process using a solvent instead of firing. In order to realize this solidification process, a grinding process is required to increase the activation energy of the surface of the raw ceramics particle. Therefore, in this study, we set up a molecular dynamics model that simulated grinding and calculated the activation of the silica surface. The grinding of the material surface was modified by the cylindrical indenter of LAMMPS, the material surface was constantly activated by passing multiple indenters continuously instead of a single indenter. As a result, a clear increase in energy was observed. It was suggested that continuous energy input is more effective than local energy input to the surface when reproducing surface activity. Furthermore, activation of the internal structure was observed as in the experiment. Adding water molecules in the relaxation calculation on the activated surface, binding through and without water molecules was observed. It was clarified that there are hydrogen bonds and siloxane bonds in this bond.

著者

栗林 大樹 佐藤 知広 齋藤 賢一 宅間 正則 高橋 可昌

出版者

一般社団法人 粉体粉末冶金協会

雑誌

粉体および粉末冶金 (ISSN:05328799)

巻号頁・発行日

vol.68, no.8, pp.317-323, 2021-08-15 (Released:2021-08-15)

参考文献数

16

---

In recent years, materials infomatics (MI), a technology that combines materials engineering and machine learning, has become popular and is used for discovering new materials. In this research, we aimed to verify whether MI can be applied to the problem of “development and maintenance of technology,” which is becoming more difficult due to the decrease in the number of engineers caused by the declining birthrate and aging population in Japan. We selected “discrimination of optical electron microscope images” as the verification target, and used Convolutional Neural Networks (CNNs) as the machine learning technology to discriminate between seven types of sintered metal objects under different sintering conditions, hoping for general applicability to the discrimination problem, and confirmed a discrimination accuracy of 98.5%. In addition, we verified the effectiveness of using pseudo-samples for the discrimination problem using Generative Adversarial Networks (GANs) in the hope of improving accuracy by increasing the number of samples, and confirmed the improvement of accuracy by adding pseudo-samples to the training data.