Real-time inline predictions of jet fuel properties by NIRS

Significance of the Topic

Jet fuel performance and safety depend on multiple physical and chemical properties. Conventional ASTM/ISO laboratory methods are accurate but time-intensive and resource-heavy. Inline near-infrared spectroscopy (NIRS) offers rapid, non-destructive analysis enabling real-time quality control.

Objectives and Study Overview

This Application Note demonstrates development of quantitative models for key jet fuel parameters such as API gravity, density, aromatic content, cetane index, boiling profiles, flash point, freezing point, hydrogen content and viscosity using a NIRS XDS Process Analyzer. The goal is to replace slow wet-chemical tests with fast inline predictions.

Methodology and Used Instrumentation

A set of 105 diverse jet fuel samples from various suppliers and grades was analyzed. Spectra were acquired in the 800–2200 nm range with 32 scans per measurement using a NIRS XDS Process Analyzer equipped with a stainless steel immersion probe (6 mm gap, 12 mm path length). Each sample was measured in triplicate with stirring. Spectral data were pretreated using second-derivative transformation to correct baseline and multiplicative effects. Calibration models were developed in Vision chemometric software using leave-one-out cross-validation and independently validated.
Used Instrumentation

• NIRS XDS Process Analyzer – Article No. 29280120
• NIRS Immersion Probe SS – Article No. 67440010

Main Results and Discussion

Calibration performance matched or exceeded ASTM precision for all parameters:

• API gravity: R2 = 0.986, SEC = 0.26, SECV = 0.27, SEP = 0.29
• Density at 15 °C: R2 = 0.984, SEC = 0.0013, SECV = 0.0013, SEP = 0.0016
• Aromatic content: R2 = 0.962, SEC = 0.5, SECV = 0.51, SEP = 0.72
• Cetane index: R2 = 0.934, SEC = 0.85, SECV = 0.89, SEP = 0.84
• Boiling profiles at 10%, 20%, 50%, 90% recovery: R2 range 0.839–0.952, SEC 1.9–3.2 °C
• Flash point: R2 = 0.925, SEC = 1.9 °C
• Freezing point: SEC = 1.9 °C (precision 0.8 °C)
• Hydrogen content: R2 = 0.939, SEC = 0.05 wt%
• Viscosity at –20 °C: R2 = 0.905, SEC = 0.2139 cSt

The spectral variations corresponded to overtone and combination bands of chemical bonds, enabling accurate physico-chemical predictions.

Benefits and Practical Applications

NIRS delivers results within 30 seconds from a single measurement without specialized operators. Inline implementation of NIRS accelerates process monitoring and supports consistent product quality in refinery and distribution settings.

Future Trends and Possibilities

• Extending model libraries to cover broader fuel types and additives.
• Integrating NIRS with process control systems and digital twins for automated feedback.
• Advancing chemometric algorithms to improve robustness under variable process conditions.
• Coupling NIRS data with machine learning for predictive maintenance and real-time optimization.

Conclusion

The study confirms that NIRS XDS Process Analyzer provides fast, accurate inline predictions of critical jet fuel parameters, matching standard lab method precision. This enables real-time quality assurance and more efficient fuel management.