The complexity of crude oil
Crude oil contains thousands of organic compounds ranging from light hydrocarbons to complex biomolecules. Speciation of hydrocarbons is of great interest, as certain isomers may have an adverse effect on engine performance if present in high quantities in the final fuel. Furthermore, chemical fingerprinting of crude oil is extremely important for assessing the level of oil maturation and biodegradation, with the analysis of biomarkers, in particular, being used for this purpose. These are breakdown products of the biomolecules in the original oil producing organisms, and are known as ‘chemical fossils’ because of their resistance to degradation. The presence of these compounds (in certain ratios) can allow the analyst to gauge the thermal maturity and extent of biodegradation of the oil, and can even reveal information on the source rock.
However, there are many challenges involved in the analysis of crude oils, including sample complexity, diversity and trace concentrations. In the past, such complex samples were subjected to chemical fractionation followed by one dimensional gas chromatography with mass spectrometry (GC–MS). However, fractionation is a time consuming, labour intensive process, which uses large volumes of solvents and can be prone to error.
Two dimensional gas chromatography
Two dimensional gas chromatography coupled with time of flight mass spectrometry (GC×GC–TOF MS) is now used routinely in the oil industry to combat this problem. The coupling of two columns of different selectivity provides GC×GC with the capacity to resolve an order of magnitude more compounds than traditional gas chromatography 1, making it ideal for the analysis of complex mixtures. The sample is separated in two dimensions, across a retention plane instead of along a retention line (as in conventional gas chromatography), allowing all chemical classes to be monitored in a single analytical run.
The output of GC×GC combines a series of fast second dimension separations stacked side by side to produce a two dimensional retention time plane. A colour gradient is then added to indicate signal intensity, allowing the retention plane to be viewed as a 3D surface plot. An alternative representation is the colour contour plot, which provides a birds eye view looking down on the peaks (Figure 1). An additional benefit of GC×GC–TOF MS is the structured ordering of chromatograms. A phenomenon known as ‘roof tiling’ causes compounds in the same/similar chemical classes to elute in bands, allowing ‘at a glance’ overview of oil composition.
Figure 1: Comparison of GC×GC–TOF MS contour plots of a crude oil analysed at 70 eV (top) and 14 eV (bottom), with expansions highlighting the improved chromatographic resolution for hydrocarbons at low ionisation energies.
Despite the superior separation afforded by GC×GC, the identification of individual compounds in complex samples remains challenging when multiple compounds in a chemical class have similar spectra at conventional (70 eV) ionisation energies. Branched alkanes are a prime example, with weak molecular ions that further complicate the process. As a result, group type analysis is often performed, whereby individual compounds are not identified, but simply reported as being part of the same chemical class, e.g. ‘C30 iso-alkanes’. Spectral similarity can be addressed by the use of soft ionisation to reduce the degree of ion fragmentation, but this approach has been cumbersome to implement until now.
In this article, we show how the problem of compound identification in petrochemical samples can be resolved without recourse to the time consuming analysis of multiple standards. We achieve this by the application of a MS ion source technology that enhances compound identification by enabling efficient electron ionisation at much lower (softer) energies, without compromising sensitivity. This low energy electron ionisation reduces analyte fragmentation, which in turn enhances selectivity, sensitivity, and aids structural elucidation. In contrast to other soft ionisation techniques for GC–MS, low energy ionisation is achieved by changing a parameter in the analytical method, so avoiding the need for reagent gases, ion source pressurisation, or changes in hardware setup.
A 100 mg/mL dilution of crude oil in dichloromethane was injected into a 7890A GC (Agilent Technologies, USA) with a DB-5 (28 m × 0.25 mm × 0.25 µm) 1st dimension column and a SGE BPX50 (3.3 m × 0.1 mm × 0.1 µm) second dimension column, with a total run time of 79.75 min. The detector was a BenchTOF-Select time of flight mass spectrometer (Markes International, UK) acquiring in the mass range m/z 40–600 at a rate of 50 Hz (with 200 spectral accumulations per data point). Image processing employed GC Image (GC Image, LLC).
The GC×GC–TOF MS contour plots shown in Figure 1 provide a comparison of the analysis of crude oil at 70 eV and 14 eV. The results show an improvement in selectivity at low energy ionisation, particularly within the aliphatic region, due to dramatically reduced fragmentation of these compounds within the ion source.
The enhanced selectivity provided by low energy ionisation allowed the automated detection of over 1000 additional ‘blobs’ in the 14 eV GC×GC contour plot, compared to the same sample run at 70 eV. This extra information provides more robust statistical comparisons for chemical fingerprinting of whole oils, an advantage that is also important in environmental forensic investigations, where spilled oils need to be matched to potential sources.
Enhancement of structurally significant fragment ions
Figure 2 demonstrates how the reduced fragmentation resulting from soft electron ionisation can improve the differentiation of isomeric species, by increasing confidence in compound identification. In effect, the spectrum at 14 eV provides both additional information and independent corroboration of compound identification using the reference quality 70 eV spectrum. The enhancement of molecular ion and structurally-significant fragments at low energies can also aid structural elucidation of compounds that are not present in commercial libraries.
It is worthwhile noting that the low energy spectra generated are repeatable, making it possible to generate searchable libraries of mass spectra at a given ionisation energy.
Figure 2: Top: GC×GC–TOF MS contour plot of a crude oil, showing overlaid EICs (m/z 268 + 282 + 296 + 310 + 324 + 338 + 352), analysed using Select-eV with an ionisation energy of 14 eV. The panels for each of the four labelled peaks A–D show the significant differences in fragmentation between 70 eV and 14 eV, and that it is much easier to discriminate between these compounds using the low-energy spectra.
Identification of biomarkers
Variable energy ion source technology can also aid the identification of crude oil biomarkers. Figure 3 compares the 70 eV and 14 eV mass spectra for two sterane type biomarkers in the crude oil. In both cases, the mass spectra at low energies exhibit dramatically reduced fragmentation, as well as enhanced molecular ion signals, resulting in improved signal to noise values and lowered limits of detection. However, unlike other soft ionisation techniques that just produce the molecular ion, the technology described here results in significant fragments that can aid structural elucidation.
Figure 3: Mass spectral comparisons for two sterane-type biomarkers at 70 eV and 14 eV, showing improved signal-to-noise ratios for the molecular ion at the lower energy.
In this article, we have shown how advances in analytical technology allow us to extend our knowledge of oil composition further than ever before. We have demonstrated how a GC×GC–TOF MS approach provides both the sensitivity and chromatographic resolving power necessary to deliver highly structured, data rich chromatograms, alongside mass spectra that provide a high degree of confidence in analyte identification.
Furthermore, soft electron ionisation has been shown to provide improved sample characterisation for these complex crude oils, by reducing fragmentation and increasing molecular ion response, with no inherent loss in sensitivity. A key benefit of this, the ability to differentiate between compounds with very similar spectra, may help to uncover additional information on oil composition and geochemistry that has previously gone unnoticed. The enhanced sensitivity and selectivity stemming from the dramatic reduction in fragmentation at low energies also greatly increases the number of compounds identified, permitting robust statistical comparisons essential for successful chemical fingerprinting.
1. J.B. Phillips and J. Beens, 'Comprehensive two-dimensional gas chromatography: A hyphenated method with strong coupling between the two dimensions', Journal of Chromatography A, 1999, 856: 331–347.
Written by Laura McGregor and David Barden, Markes International, UK.
Edited by Claira Lloyd
Read the article online at: https://www.hydrocarbonengineering.com/gas-processing/31072014/crude-oil-markes-analysis/