Published November 8, 2022 | Version v1
Dataset Open

Novel techniques for characterising graphene nanoplatelets using Raman spectroscopy and machine learning

  • 1. Henry Royce Institute, National Graphene Institute, and Department of Materials, The University of Manchester, Oxford Road, M13 9PL, UK.
  • 2. Graphene Engineering Innovation Centre, Masdar Building, The University of Manchester, Sackville Street, M1 3BB UK.
  • 3. Centre for Imaging Sciences, Division of Informatics, Imaging and Data Sciences, School of Health Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, Oxford Road, M13 9PL, UK.

* Contact person

Description

A significant challenge for graphene nanoplatelet (GNP) suppliers is the meaningful characterisation of platelet morphology in an industrial environment. This challenge is further exacerbated to platelet surface chemistry when scalable functionalisation processes such as plasma treatment are used to modify the GNPs to improve the filler-matrix interphase in nanocomposites. The costly and complex suite of analytical equipment necessary for a complete material description makes quality control and process optimisation difficult. Raman spectroscopy is a facile and accessible characterisation technique with recent advancements unlocking fast mapping for rapid data collection. In this work we detail several big-data methodologies that extract full value out of Raman spectra so that GNP morphology and surface chemistry can be characterised. A unsupervised peak fitting and processing algorithm was used to extract crystallinity data rapidly and accurately and correlate it with laser-diffraction derived lateral size values for a commercial set of GNPs. Classical machine learning was used to find the most informative Raman features for classifying plasma functionalised GNPs. The initial material properties were found to affect which peak features are most useful for classification. In low-defect density and low specific surface area GNPs, the D peak FWHM is found the most useful, whereas the 2D/G ratio is most useful in the opposite case. Finally, a convolutional neural network was trained to discern between different GNP grades with 86% accuracy. This work demonstrates how computer vision could be deployed for rapid and accurate quality control on the factory floor.

Files

File preview

files_description.md

All files

Files (719.5 MiB)

Name Size
md5:e37eb2703944d33eb91b07473a16ccee
302 Bytes Preview Download
md5:225c0b068f1893c25b06607092a45849
26.5 MiB Preview Download
md5:3de7a6d96a7803b035ec1ed17356cf72
354.4 MiB Preview Download
md5:bdc4ea183465f7815493c0c607e1cdc6
82.0 MiB Preview Download
md5:48674f0db35f432200081cbfd97f94f1
256.6 MiB Preview Download

References

Journal reference (Journal Article)
Orts Mercadillo V. et al., 2D Materials, 10 (2023), doi: 10.1088/2053-1583/acc080

Journal reference (Journal Article)
Orts Mercadillo V. et al., 2D Materials, 10 (2023)