Machine Learning for Shape Memory Graphene Nanoribbons and Applications in Biomedical Engineering

• DOI: 10.3390/bioengineering9030090
• Publication Date: 2022-02-23T14:34:38Z
• Authors: Carlos León; Roderick Melnik
• Keywords: Metastability

Shape memory materials have been playing an important role in a wide range of bioengineering applications. At the same time, recent developments graphene-based nanostructures, such as nanoribbons, demonstrated that, due to unique properties graphene, they can manifest superior electronic, thermal, mechanical, and optical characteristics ideally suited for their potential usage next generation diagnostic devices, drug delivery systems, other biomedical One most intriguing parts these new lies fact that certain types graphene nanoribbons exhibit shape effects. In this paper, we apply machine learning tools build interatomic from DFT calculations highly ordered oxide material had effects with recovery strain up 14.5% 2D layers. The layer shrink metastable phase lower constant lattice through application electric field, returns initial external mechanical force. deformation leads electronic rearrangement induces magnetization around oxygen atoms. show no sufficiently narrow while model predict suppression narrower nanoribbons. We improve prediction accuracy by analyzing only evolution phase, where is found according calculations. developed here allows also us study phases wider would be computationally inaccessible pure approach. Moreover, extend our analysis realistic systems include vacancies boron or nitrogen impurities at atomic positions. Finally, provide brief overview current applications exhibiting fields, focusing on data-driven approaches potentials.