Related Experiment Video
Updated: Jan 6, 2026
08:59
Dry Oxidation and Vacuum Annealing Treatments for Tuning the Wetting Properties of Carbon Nanotube Arrays
Published on: April 15, 2013
15.4K
Multivariate Calibration for Carbon Nanotubes in the Environment Using the Microwave Induced Heating Method
Yang He1, Souhail R Al-Abed2, Dionysios D Dionysiou1
1Environmental Engineering and Science program, Department of Chemical and Environmental Engineering, University of Cincinnati, 2600 Clifton Ave., Cincinnati, Ohio 45221, United States.
Environmental Nanotechnology, Monitoring & Management
|October 5, 2019
Summary
Chemometrics models like LS-SVM and ANN can quantify carbon nanotubes (CNTs) in environmental samples using microwave heating. LS-SVM demonstrated superior accuracy for simultaneous determination of multiple CNT types.
Area of Science:
- Environmental Science
- Analytical Chemistry
- Materials Science
Background:
- Carbon nanotubes (CNTs) are increasingly used, necessitating accurate environmental monitoring.
- Quantifying individual CNT types in complex matrices poses analytical challenges.
Purpose of the Study:
- To develop and compare chemometrics-based multivariate calibration models for simultaneous CNT determination.
- To utilize microwave-induced temperature rise data for quantitative analysis of CNTs.
Main Methods:
- A multifactor, multilevel experimental design was employed to create calibration datasets.
- Partial Least Square Regression (PLS), Least Square-Support Vector Machine (LS-SVM), and Artificial Neural Networks (ANN) were applied.
- Microwave-induced temperature rise (ΔT) spectral data were collected for samples containing single-walled CNTs (SWCNTs), multi-walled CNTs (MWCNTs), and carboxylated MWCNTs (MWCNT-COOH).
Main Results:
- The LS-SVM model exhibited the highest predictive accuracy (R² = 0.64-0.95) with the lowest prediction error (RMSEP = 0.0243-0.0410 mg).
- The ANN model showed good accuracy for two-component mixtures (R² = 0.77-0.89).
- The PLS model was least effective, with low R² values (0.20-0.87) and high RMSEP (0.0209-0.1021 mg).
Conclusions:
- Chemometrics, particularly LS-SVM, offers a viable approach for simultaneously quantifying multiple CNT types in environmental samples.
- Microwave-induced temperature rise data, when analyzed with appropriate models, can be effectively used for CNT mass determination.
- LS-SVM and ANN show promise for environmental monitoring of CNTs, outperforming PLS in this application.
