本文へジャンプ

Advanced Material Design based on
Synergetic Exploration, Learning, and Prediction

TANIIKE Lab.
Professor: TANIIKE Toshiaki

E-mail:
[Research areas]High-throughput experimentation, materials informatics, computational chemistry
[Keywords]Catalysis, polymer, membrane, nanocomposite, graphene

Skills and background we are looking for in prospective students

Students who have skill or background in either physical chemistry, quantum chemistry, polymer science, mechanical engineering or so on. Motivation to challenge a new field is most desired.

What you can expect to learn in this laboratory

Through research activities including the attendance of experimental report meeting, colloquium and conferences, we wish a student to acquire
1. Abilities to find out a bottleneck problem in materials science and to design research strategy and methodology to the problem.
2. Abilities related to scientific presentation, writing, and discussion.

【Job category of graduates】
Companies for chemistry and chemical engineering

Research outline

Facing rapid acceleration in science and technology, TANIIKE laboratory aims to develop and implement innovation-oriented materials science based on the combination of three concepts: Exploration for high-throughput experimentation, Learning with data mining, and Prediction based on high-precision molecular modeling.

1. Exploration

Combination of different elements and substances can bear an astronomical number of materials. One of the purposes in materials science is to explore good combinations and a novel way (process) to do that. In order to find out serendipity among numerous material candidates, we conduct high-throughput experiments using automated and/or parallelized instruments and tools. Including the developments of new high-throughput modules and efficient protocols for screening, researchers maximize the throughput of the experiments and change the research style from routine works to examination and consideration.

2. Learning

High-throughput experiments generate not only a huge number of materials, but also a huge quantity of data which list synthetic conditions, structural characteristics and performances of the materials. In order to implement efficient exploration of materials, it is not enough only to pick up good ones: It is necessary to clarify a structure-property relationship, i.e. to find which factors are relevant to a performance. Since the performance is rarely dominated by a single factor, multivariate analysis or even more sophisticated machine learning techniques are required. We learn from all of the data to significantly improve the quality of the next high-throughput experiments and to discover a novel principle connecting the structure and performance of materials.

3. Prediction

Developments of computers and computational chemistry have enabled realistic simulation of complicated materials. Nonetheless, computer-aided materials design (i.e. in-silico design) is still far to be realized. The most difficult and important task is how to construct a molecular model which represents a material. We implement experiments and construct a high-precision molecular model which can explain these experimental results toward in-silico materials design. Practical computational chemistry is intended with deep understanding of materials science and experiments. With the above three concepts, we are pursuing a systematic methodology to efficiently discover breakthrough, a rational solution from a wide range of materials (rather than concentrating on a specific material), and cultivation of materials scientists who can design strategy and methodology of research and development.

Key publications

  1. A. Thakur, R. Baba, P. Chammingkwan, M. Terano, T. Taniike, Synthesis of aryloxide-containing half-titanocene catalysts grafted to soluble polynorbornene chains and their application in ethylene polymerization: integration of multiple active centres in a random coil, Journal of Catalysis, 2018, 357, 69-79.
  2. A.T.N. Dao, K. Nakayama, J. Shimokata, T. Taniike, Multilateral characterization of recombinant spider silk in thermal degradation, Polymer Chemistry, 2017, 8, 1049-1060.
  3. L.H. Le, D.X. Trinh, N.B. Trung, T.P.N. Tran, T. Taniike, Fabrication of assembled membrane from malonate-functionalized graphene and evaluation of its permeation performance, Carbon, 2017, 114, 519-525.

Equipment

Various high-throughput modules such as automated balancer, liquid handler, different types of parallelized/automated reactors, micro-extruder, automated sampling and QMS/GC analysis, microplate reader, parallelized high-pressure filtration, high-throughput/operando chemiluminescence imaging instruments, and so on. These instruments were and are being developed in our laboratory. Software for data analysis and computational chemistry.

Teaching policy

There is no core time in our laboratory: The throughput of experiments and the research itself is maximized in order to assure the work-life balance. On the other hand, extensive discussion is regularly made based on well-written reports and papers in experimental group meeting (once a few weeks) and colloquium (once a month). The attendance of domestic and international conferences is fully supported.

[Website] URL:http://www.jaist.ac.jp/ms/labs/taniike/en/

PageTop