You have been redirected to your local version of the requested page

Fuels can incorporate traces of water during the production process, in transport, and while in storage. Excessive water in fuels poses several problems. For example, elevated water content in diesel fuel promotes biological growth in storage tanks, which could lead to metal corrosion and formation of sludge and biofilms. This in turn can cause blockage of fuel filters and therefore damage vehicle fuel injection systems. 

The standard specification for diesel fuel quality includes multiple parameters, but water contamination is the biggest risk factor. According to the European Committee for Standardization, the maximal acceptable amount of water in diesel for commercialization is 200 mg/L (ppm) (EN 590). Usually, this is determined by Karl Fischer (KF) titration, yet this method requires chemicals and takes about five minutes to perform. This Application Note describes how near-infrared spectroscopy (NIRS) is a faster and more cost-efficient alternative to KF titration for the prediction of water content in diesel fuel.

DS2500 Liquid Analyzer and a sample filled in a disposable vial.
Figure 1. DS2500 Liquid Analyzer and a sample filled in a disposable vial.

Samples of diesel with varying water contents (from 103 to 379 mg/L) were measured with a DS2500 Liquid Analyzer in transmission mode (400–2500 nm). Reproducible spectrum acquisition was achieved using the built-in temperature control at 40 °C. For convenience, disposable vials with a pathlength of 8 mm were used, which made cleaning of the sample vessels unnecessary. The Metrohm software package Vision Air Complete was used for all data acquisition and prediction model development. 

Table 1. Hardware and software equipment overview
Equipment Metrohm number
DS2500 Liquid Analyzer 2.929.0010
DS2500 Holder 8 mm vials 6.7492.020
Disposable vials, 8 mm 6.7402.000
Vision Air 2.0 Complete 6.6072.208

The obtained Vis-NIR spectra (Figure 2) were used to create a prediction model for quantification of the moisture content in diesel samples. The quality of the prediction model was evaluated using the correlation diagram, which displays a very high correlation between the Vis-NIR prediction and the reference values. The respective figures of merit (FOM) display the expected precision of a prediction during routine analysis.

Figure 2. Vis-NIR spectra of diesel samples analyzed on a DS2500 Liquid Analyzer.
Figure 3. Correlation diagram for the prediction of water content in diesel using a DS2500 Liquid Analyzer. The lab value was evaluated using KF titration.
Table 2. Figures of merit for the prediction of water content in diesel using a DS2500 Liquid Analyzer.
Figures of merit Value
R2 0.9776
Standard error of calibration 16 ppm
Standard error of cross-validation 21 ppm

This application note demonstrates the feasibility to determine a key parameter of the quality control of diesel fuel (water content) with NIR spectroscopy. The main advantages of Vis-NIR spectroscopy over wet chemical methods are that running costs are significantly lower and time-to-result is significantly reduced. Additionally, no chemicals are required and the technique is non-destructive to samples.

Table 2. Time to result overview for KF titration
Parameter Method Time to result
Water Karl Fischer titration ∼ 5 minutes

Metrohm AG

9100 Herisau


Internal reference: AW NIR CH-0064-112021