Design and Optimisation of Nanomaterials with Artificial Intelligence

Kalay M.

4.Uluslararası Mühendislik ve Doğa Bilimleri Çalışmaları Kongresi, Ankara, Turkey, 24 - 25 May 2024, pp.586-599

  • Publication Type: Conference Paper / Full Text
  • City: Ankara
  • Country: Turkey
  • Page Numbers: pp.586-599
  • Kayseri University Affiliated: Yes


This study focuses on the design and optimisation of nanomaterials using artificial intelligence and deep learning techniques as well as traditional materials science methods. Nanomaterials are materials produced by nanotechnology and generally have dimensions in the nanometre range. With the advancement of nanotechnology, nanoscale products are spreading from electronics to healthcare products and pharmaceuticals, initiating a radical change in the world order. The properties of these materials are determined by many factors, ranging from their chemical composition to their crystal structure and surface morphology. These properties are critical in determining the performance of materials. Traditionally, the design and synthesis of nanomaterials has been carried out through experimental studies. However, this process is quite time-consuming and costly. With the introduction of artificial intelligence and deep learning techniques, this process can be made more efficient. Artificial intelligence and deep learning algorithms can identify complex relationships and recognise patterns based on large amounts of data. This enables them to analyse and optimise the relationships between nanomaterial properties and synthesis conditions. This area of research has great potential in materials science and can play an important role in the discovery and design of next-generation materials. It can also contribute to the dissemination of nanomaterials to a wider range of applications, enabling the emergence of innovative solutions in many fields, from electronics to biomedicine. To this end, the results of previous studies have been evaluated, analysing the advantages of these techniques and the shortcomings of existing methods.