Alternating and Modified Alternating Least Squares Applied to Raman Spectra of Finished Gasolines

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Chemical Engineering

Background:

• Characterizing finished gasoline composition is crucial for quality control and process optimization.
• Traditional methods for component analysis can be complex and time-consuming.
• Raman spectroscopy offers a non-destructive approach to chemical analysis.

Purpose of the Study:

• To extract and characterize individual components from finished gasoline using Raman spectroscopy.
• To compare the effectiveness of Modified Alternating Least Squares (MALS) against Alternating Least Squares (ALS) for multicomponent resolution.
• To demonstrate the advantages of MALS in handling complex spectroscopic data.

Main Methods:

• Raman spectroscopy was employed to collect spectral data from finished gasoline.
• Modified Alternating Least Squares (MALS) was utilized for separating spectral data into pure component spectra.
• Alternating Least Squares (ALS) was used as a benchmark for comparison.

Main Results:

• MALS demonstrated superior accuracy and resolution capabilities compared to ALS.
• MALS proved more robust in resolving components from collinear data.
• MALS successfully extracted components present at low concentrations (near noise level), which ALS could not resolve.

Conclusions:

• MALS is a more effective and robust method than ALS for multicomponent resolution of Raman spectroscopic data from gasoline.
• The stabilization techniques in MALS overcome limitations of ALS, particularly with noisy or collinear data.
• This study validates MALS as a powerful tool for detailed chemical analysis of complex mixtures like gasoline.