Partial least squares methods in open-source software for identification of adulterated creatine using infrared...

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Chemometrics

Background:

• Creatine is a widely used dietary supplement for physical activity.
• Ensuring the purity of creatine supplements is crucial for consumer safety and efficacy.
• Adulteration of supplements with cheaper substances like corn starch is a concern.

Purpose of the Study:

• To develop and validate a novel analytical methodology for detecting creatine adulteration.
• To differentiate pure creatine from creatine adulterated with corn starch (CS).

Main Methods:

• Association of Attenuated Total Reflection-Fourier Transform Infrared (ATR-FTIR) spectroscopy with multivariate analyses.
• Development of Partial Least Squares Regression (PLS) and Partial Least Squares Discriminant Analysis (PLS-DA) models.
• Utilized GNU Octave for model development and analysis.

Main Results:

• Key spectral features, including asymmetric stretching bands of the carboxyl group in creatine and glycosidic bonds in CS, were identified for discrimination.
• The PLS model demonstrated a strong experimental fit with a Root Mean Square Error of Prediction (RMSEp) of 9.15%.
• The PLS-DA model achieved 97% accuracy in distinguishing pure from adulterated creatine samples, with 95% accuracy, 90% sensitivity, and 100% specificity in blind testing of commercial samples.

Conclusions:

• The combined ATR-FTIR and multivariate analysis approach is effective for screening adulterated creatine supplements.
• The developed PLS-DA model provides a reliable tool for quality control and identification of creatine adulteration in the food supplement market.