Quantification of cocaine in ternary mixtures using partial least squares regression applied to Raman and Fourier transform infrared spectroscopy

2017 ◽  
Vol 48 (12) ◽  
pp. 1732-1743 ◽  
Author(s):  
Ciro Augusto Fernandes de Oliveira Penido ◽  
Marcos Tadeu Tavares Pacheco ◽  
Etelvino Henrique Novotny ◽  
Igor K. Lednev ◽  
Landulfo Silveira
Keyword(s):  
Fourier Transform ◽  
Infrared Spectroscopy ◽  
Least Squares ◽  
Fourier Transform Infrared Spectroscopy ◽  
Partial Least Squares ◽  
Partial Least Squares Regression ◽  
Fourier Transform Infrared ◽  
Ternary Mixtures ◽  
Least Squares Regression

A Novel Calibration Method for the Quantification of Respirable Particles in Mining Scenarios Using Fourier Transform Infrared Spectroscopy

2020 ◽  
pp. 000370282096806
Author(s):  
Robert Stach ◽  
Teresa Barone ◽  
Emanuele Cauda ◽  
Boris Mizaikoff
Keyword(s):  
Particle Size ◽  
Fourier Transform ◽  
Infrared Spectroscopy ◽  
Sample Preparation ◽  
Least Squares ◽  
Partial Least Squares ◽  
Partial Least Squares Regression ◽  
Fourier Transform Infrared ◽  
Respirable Dust ◽  
Least Squares Regression

The exposure of mining workers to crystalline particles, e.g., alpha quartz in respirable dust, is a ubiquitous global problem in occupational safety and health at surface and underground operations. The challenge of rapid in-field monitoring for direct assessment and adoption of intervention has not been solved satisfactorily to date, as conventional analytical methods such as X-ray diffraction and infrared spectroscopy require laboratory environments, complex system handling, tedious sample preparation, and are limited by, e.g., addressable particle size. A novel monitoring approach was developed for potential in-field application enabling the quantification of crystalline particles in the respirable regime based on transmission infrared spectroscopy. This on-site approach analyzes samples of dust in ambient air collected onto PVC filters using respirable dust sampling devices. In the present study, we demonstrate that portable Fourier transform infrared (FT-IR) spectroscopy in combination with multivariate data analysis provides a versatile tool for the identification and quantification of minerals in complex real-world matrices. Without further sample preparation, the loaded filters are immediately analyzed via transmission infrared spectroscopy, and the mineral amount is quantified in real-time using a partial least squares regression algorithm. Due to the inherent molecular selectivity for crystalline as well as organic matrix components, infrared spectroscopy uniquely allows to precisely determine the particle composition even in complex samples such as dust from coal mines or clay-rich environments. For establishing a robust partial least squares regression model, a method was developed for generating calibration samples representative in size and composition for respirable mine dust via aerodynamic size separation. Combined with experimental design strategies, this allows tailoring the calibration set to the demands of air quality management in underground mining scenarios, i.e., the respirable particle size regime and the matrix of the target analyte.

Sign in / Sign up

Export Citation Format

Share Document