Moisture analysis of ethanol- hydrocarbon blends by Vis-NIR spectroscopy

Significance of the Topic

Fuel contaminants such as free water lead to corrosion, tank damage and engine wear. Modern fuel systems require effectively zero free water. Conventional Karl Fischer titration is time-consuming, uses toxic reagents and demands skilled operators. Vis-NIR spectroscopy offers a rapid, non-destructive alternative for moisture monitoring in ethanol-hydrocarbon blends.

Objectives and Study Overview

The study aimed to evaluate Visible-Near Infrared Spectroscopy (Vis-NIRS) for quantifying moisture content in ethanol-hydrocarbon mixtures. A calibration model was developed on 47 samples and externally validated on 4 independent samples with moisture levels between 0.0026% and 1.2016%, determined by coulometric KF titration.

Methodology and Instrumentation

Samples were measured in transmission mode using a NIRS XDS RapidLiquid Analyzer (400–2500 nm) with 2 mm quartz cuvettes at ambient temperature. Spectral data acquisition and model development were performed with Vision Air 2.0 Complete. A Partial Least Squares (PLS) regression using six factors focused on water absorption bands at 1380–1670 nm and 1875–1990 nm.

Main Results and Discussion

The PLS model achieved a correlation coefficient (R2) of 0.999 with a standard error of calibration (SEC) of 0.0097%, cross-validation error (SECV) of 0.0116% and prediction error (SEP) of 0.0120%. External validation on four samples yielded moisture predictions within ±0.0120% of KF titration. Key water absorption features at 1400 nm and 1900 nm showed linear correlation with moisture levels.

Benefits and Practical Applications

Vis-NIR enables fast, reagent-free moisture analysis, reducing sample preparation and hazards associated with toxic solvents. Its high throughput and ease of use support raw material inspection, process monitoring and final quality control in fuel production and distribution.

Future Trends and Potential Applications

Expected developments include integration of inline Vis-NIR sensors for continuous moisture monitoring, expansion to diverse fuel matrices and blending processes, miniaturized spectrometers for field use, and advanced chemometric algorithms and machine learning to enhance predictive performance.

Conclusion

This application demonstrates that Vis-NIR spectroscopy is a reliable and efficient method for determining moisture in ethanol-hydrocarbon blends, matching the accuracy of Karl Fischer titration while offering significant operational advantages.

References

• MyCleanDiesel. Problem of Water in Diesel.
• MyCleanDiesel. Solutions for Water in Diesel.
• ASTM D6304. Standard Test Method for Determination of Water in Petroleum Products by Coulometric Karl Fischer Titration.