Simultaneous quantification of uranium(<scp>vi</scp>), samarium, nitric acid, and temperature with combined ensemble learning, laser fluorescence, and Raman scattering for real-time monitoring

Sadergaski, Luke R.; Andrews, Hunter

doi:10.1039/d2an00998f

Cited by 14 publications

(33 citation statements)

References 43 publications

(131 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The 6 H 9/2 band changed the most in shape and relative intensity to the other peaks in the spectrum. Band intensities increased with a more coordinating environment as LiCl concentration increased, which suggests that the transitions have an electric dipole (ED) character …”

Section: Resultsmentioning

confidence: 99%

“…9−12 Even though f-element absorption and emission spectra are described as "line-like," the fine structure of these spectra has proven to be useful for determining the symmetry and coordination environment of lanthanide species. 15,17 The emission spectra of Eu(III) and Sm(III) in differing concentrations of LiCl (3−12 M) exhibit numerous lines between 550 and 725 nm (Figure 1). The emission spectra are likely affected by long-range interaction between the cations and solvent molecules and outer-sphere complexation with chloride anions.…”

Section: Resultsmentioning

confidence: 99%

“…This procedure was based on previous studies with LV selection and allowed for automation when building and evaluating PLSR models. 17,27 Model prediction performance was compared using Tukey− Kramer significance tests. 34,35 Full details are provided elsewhere.…”

Section: Partial Least Squaresmentioning

confidence: 99%

“…Although the time dimension provides a unique fingerprint for various species, the analysis time is often ∼10–15 min, which is too slow for processes requiring real-time feedback . Laser-induced fluorescence (LIF) spectra can be measured directly, using a charge-coupled device, to quantify complex systems with overlapping bands, matrix effects, chemical interaction(s), and baseline offsets using multivariate chemometrics …”

Section: Introductionmentioning

confidence: 99%

“…16 Laser-induced fluorescence (LIF) spectra can be measured directly, using a charge-coupled device, to quantify complex systems with overlapping bands, matrix effects, chemical interaction(s), and baseline offsets using multivariate chemometrics. 17 One of the most robust supervised regression techniques, partial least squares regression (PLSR), 18 covariance between two matrices corresponding to the spectra (X) and concentrations (Y) using combinations of latent variables (LVs). PLSR models require a representative training set that can be efficiently selected using design of experiments.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Pursuit of the Ultimate Regression Model for Samarium(III), Europium(III), and LiCl Using Laser-Induced Fluorescence, Design of Experiments, and a Genetic Algorithm for Feature Selection

2023

Self Cite

View full text Add to dashboard Cite

Laser-induced fluorescence spectroscopy, Raman scattering, and partial least squares regression models were optimized for the quantification of samarium (0−150 μg mL −1 ), europium (0−75 μg mL −1 ), and lithium chloride (0.1−12 M) with a transformational preprocessing strategy. Selecting combinations of preprocessing methods to optimize the prediction performance of regression models is frequently a major bottleneck for chemometric analysis. Here, we propose an optimization tool using an innovative combination of optimal experimental designs for selecting preprocessing transformation and a genetic algorithm (GA) for feature selection. A D-optimal design containing 26 samples (i.e., combinations of preprocessing strategies) and a userdefined design (576 samples) did not statistically lower the root mean square error of the prediction (RMSEP). The greatest improvement in prediction performance was achieved when a GA was used for feature selection. This feature selection greatly lowered RMSEP statistics by an average of 53%, resulting in the top models with percent RMSEP values of 0.91, 3.5, and 2.1% for Sm(III), Eu(III), and LiCl, respectively. These results indicate that preprocessing corrections (e.g., scatter, scaling, noise, and baseline) alone cannot realize the optimal regression model; feature selection is a more crucial aspect to consider. This unique approach provides a powerful tool for approaching the true optimum prediction performance and can be applied to numerous fields of spectroscopy and chemometrics to rapidly construct models.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%