Due to graphene's exceptional properties, graphene and related materials have been studied with great interest in both basic and applied research. In order to study these two-dimensional (2D) materials, Raman spectroscopy has been used most extensively. Despite the various advantages Raman spectroscopy offers in terms of studying and characterizing these materials, Raman spectroscopy could not be a successful technique for industrial control due to the lack of automated procedures for big data analysis.Researchers at the Universidade Federal De Minas Gerais in collaboration with US Army Research Lboratory and Aalto University have recently used parameterized Principal Component Analysis (PCA) to develop a methodology which could be ideal for large scale data treatment.
Graphene Related Materials
Graphene related materials, which include graphene, graphite intercalated compounds (GICs), amorphous carbons, fullerenes, nanotubes and biochar have sp2 hybridized carbon structures. The discovery of graphene also paved the way to explore other 2D materials such as phosphorene, borophene, transition metal dichalcogenides (TMDs), monochalcogenides and other van der waal heterostructures. Due to the special electrical, mechanical, optical and thermal properties these 2D materials offer, these 2D materials have been an area of intense research in the fields of electrical, electronics and material sciences.
.jpg)
Egorov Artem/shutterstock
Structural Characterization Using Raman Spectroscopy
Raman spectroscopy is a spectroscopic technique used in chemistry to study the vibrational, rotational and other low frequency states of molecules in a system. This technique provides a characteristic structural fingerprint for each molecule, therefore serving as a great technique for identifying different molecules. Due to its non-invasive nature and sheer simplicity, Raman spectroscopy has been the leading technique that has been used for studying and characterizing graphene related structures. Even though this spectroscopy technique is sufficient for the laboratory proof-of-concepts, industrial application of Raman spectroscopy has been limited by the lack of automated procedures for big data analysis.
Principal Component Analysis for Large Scale Data Analysis
Establishing automated routines is essential for transforming laboratory proof-of-concepts to techniques for industrial applications that allow the analysis of hundreds of unlabeled spectra. Principal component analysis (PCA) is a one such important technique used to reduce dimensionality and calculate the directions of greatest variance in a data set. By translating and rotating the original coordinate frame, the PCA analysis transforms the coordinates in order to provide relationships between the samples. However, in most cases the PCA analysis only provides nontrivial information and the analysis changes from one dataset to another. To overcome these limitations of PCA and to successfully use PCA for structural analysis of graphene related materials, Ado Jorio’s team introduced a concept of parameterization to develop a methodology for the industrial application of Raman spectroscopy to study sp2 hybridized graphene related materials.
Parameterized PCA Analysis
By constructing an nxm matrix, where n lines represent the sample spectrum and m columns indicating the wave number, the parameterized PCA analysis performs a mathematical transformation to rewrite each Raman spectra into a linear combination of principal components. The Raman spectra from a set of sp2 carbon material samples that has possible variations of a specific well-known properties has been used here to find the orthogonal linear transformation of the data following a PCA algorithm. This PCA algorithm facilitates the classification of an unknown Raman spectrum within the space spanned by the desired properties. Ado Jorio’s team demonstrated two case studies, including 1. characterization of the amorphization structure of sp2 carbons in graphene related material by parameterizing the amount of point-like and line-like defects in the 2D structure and 2. Determined the number of layers in the graphene samples to demonstrate the use of the methodology they developed.
Application of Parameterized Pca Analysis for Large Scale Analysis of Raman Spectra of Graphene Related Materials
The Parameterized PCA analysis methodology developed by Ado Jorio’s team could be used to apply Raman spectroscopy to various industrial procedures, including growth, functionalization, composites synthesis or even large-scale device fabrication graphene related materials. The new methodology could be used in quality control of graphene growth by chemical vapor deposition (CVD) method or by chemical or physical exfoliation methods. The automated model based on parameterized PCA has been used to demonstrate its use in graphene related materials in the present study, this model can be used for large scale analysis of Raman spectra of other relevant materials.
Reference:
- “Applications of Raman spectroscopy in graphene-related materials and the development of parameterized PCA for large-scale data analysis” J. Campos, H. Miranda, et al. Journal of Raman Spectroscopy. DOI: 10.1002/jrs.5225.
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.