Synthesizability indexing of machine learning algorithm suggested novel compounds

• Main Author: Koh, Joel Bo Jun
• Other Authors: Kedar Hippalgaonkar
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Materials
• Online Access: https://hdl.handle.net/10356/166928
• Institution: Nanyang Technological University

The creation of new materials is a critical aspect of scientific research and development, as it plays a significant role in improving the quality of life for people all over the world. From advanced medical technologies to the development of sustainable energy sources, new materials have a vast range of applications that have the potential to revolutionize multiple industries. New materials are essential for technological advancements, as they are often designed to be stronger, more durable, and more versatile than existing materials. For example, carbon fiber composites have replaced metal alloys in the construction of high-performance sports cars and aerospace components, resulting in lighter and more fuel-efficient vehicles [1]. Similarly, the development of new materials such as perovskite solar cells has led to significant improvements in solar panel efficiency, making renewable energy more cost-effective and accessible. Furthermore, the creation of new materials can also address specific challenges facing society, such as climate change and disease prevention. For example, researchers are currently developing nanomaterials that can selectively remove carbon dioxide from the atmosphere, potentially mitigating the effects of global warming [2]. In the field of medicine, researchers are exploring the use of biomaterials to develop advanced drug delivery systems that can more effectively target specific diseases and reduce the risk of side effects [3]. While significant progress has been made in the development of new materials, there is a need for continuous innovation to address emerging challenges and improve existing technologies. As technologies evolve and become more complex, the materials used to create them must also become more advanced to meet the demands of these new applications. For example, the emergence of electric vehicles has created a need for new materials with high energy storage capacities to develop better and more efficient batteries (Zhang et al., 2020). 3 Furthermore, materials science is a rapidly evolving field, with new discoveries and advancements being made all the time. Continuously creating new materials allows researchers to explore new possibilities and overcome limitations that may have previously hindered progress. Additionally, advancements in materials science can often have significant economic benefits, as new materials can open up new markets and create new job opportunities [4]. The purpose of this report is to document the research, development, and evaluation of a machine learning-based synthesizability indexing system for 3-element compounds. The project aims to explore the correlation between the physical and chemical properties of the constituent elements in 3-element compounds and their ability to be synthesized. The report will discuss the use of machine learning techniques and formulas to create a model that accurately predicts the synthesizability of 3-element compounds based on the properties of the constituent elements