Quantification of Diesel Additives in Co-Processed Fuels via NIR Spectroscopy and Chemometric Modeling

The growing global population and increasing energy demands are placing immense pressure on the sustainability of fossil fuels due to resource depletion and environmental concerns. Adopting a circular economy approach, which repurposes waste materials into alternative fuels, presents a promising solution for reducing both fossil fuel dependency and waste generation. Diesel additives play a critical role in enhancing fuel quality and performance; however, accurately monitoring their concentrations remains a challenge, often leading to overuse and elevated costs. This study utilizes near-infrared (NIR) spectroscopy coupled with partial least-squares (PLS) modeling to enable real-time monitoring of additive concentrations, optimizing their application, and reducing costs. We tested waste-derived oils─such as tire oil, plastic oil, and cooking oil─alongside conventional diesel, incorporating three key additives: cold-flow improver (CFI), conductivity-lubricating improver (CLI), and cetane number improver (CNI). The developed predictive models demonstrated calibration coefficients exceeding 0.98 and RPD values of 4.5 for CFI, 2.7 for CLI, and 4.4 for CNI. Repeatability tests produced RSD values of 3.9% for CFI, 6.9% for CLI, and 7.2% for CNI. Comprehensive validation, including external testing and statistical evaluations (RMSEC, RMSEP, and RMSECV), confirmed the models’ strong accuracy and reliability. These findings highlight the significant potential of PLS models for the precise determination of diesel additive concentrations across a range of fuel matrices, making them highly applicable for both industrial use and further research.