GC goes 2-D
Gas chromatography for analyzing petrochemical samples

Two-dimensional gas chromatography is a powerful analytical technique that separates compounds according to their volatility and polarity. It is based on the use of two columns with different retention mechanisms, so that different physical and physico-chemical properties such as boiling point and polarity are treated independently during the separation process.

Two-dimensional (2-D) gas chromatography (GCxGC) is particularly useful for complex hydrocarbon mixtures containing a number of homologous compounds [1]. The ability to view the results in a two-dimensional plane instead of along a one-dimensional line produces clearly-recognizable patterns, which in turn make the identification of similar compounds much easier and more reliable. This article describes the characterization of petrochemical samples using the HyperChrom Data System from Thermo Electron.

The technique has several advantages for the oil industry, notably its ability to characterize complex mixtures in terms of groups rather than individual compounds. The total number of separate compounds in an oil fraction is typically enormous, but the number of homologous classes is much smaller. Group type characterization is therefore a good way to correlate the composition of a product mixture with its physical and chemical properties [2].
Compounds that are part of homologous families usually form bands according to boiling point along the first retention axis and to polarity along the second retention axis. In particular, within the
same number of carbon atoms, branched homologues show a typical “roof tile” pattern [1] which becomes more evident for classes featuring higher retention in the second dimension. This clustering shown by homologous compounds is a great help in the qualitative interpretation of chromatograms, since the 2-D plot gives an immediate impression of the sample composition. Similarly, the quantitative evaluation of the relative amounts of different groups becomes simpler when the chemical classes are well separated in the 2-D retention plane.
In addition, quantitative results obtained with a flame ionization detector (FID) have already proven to be as good as those from a conventional GC [3]. The HyperChrom Data System has been used for this purpose, and the results for different petrochemical fractions are illustrated here.
Experimental setup
The Thermo Electron Trace 2DGC system with a CO2 Dual-Jet modulator [4] was used for the analyses. The modulator is entirely controlled by the GC, allowing the modulation time to be stored as part of the GC protocol, and the embedded electronics allow the modulation to be synchronized with the acquisition frequency.
The Trace 2DGC was configured with an SSL injector and a FID with an acquisition frequency of up to 300 Hz. The AI3000 autosampler was used to automate sample injection. A fast flame photometric detector (FFPD) with a sampling rate of 200 Hz was also used for selective sulfur measurements.
The work here was done using two sets of columns:
A) Column 1: RTX-5 30m x 0.32mm i.d., 0.25mm df
Column 2: BPX-50 0.8m x 0.1mm i.d., 0.1mm df
B) Column 1: RTX-1 30m x 0.32mm i.d., 0.25mm df
Column 2: DB-17 1m x 0.1mm i.d.,
0.2mm df
In each case the second column was chosen to provide suitable separation between homologous groups [5].
The HyperChrom Data System was used for data acquisition and processing for the group type analysis. The calculation of the relative amounts of different groups is
done by drawing a box around each group of homologous compounds. The system then calculates the total peak area within the box and expresses this as a percentage of the total peak area for the complete sample.
Experimental results
Figure 3 shows the results for a light cycle oil, with some of the compound groups shown. The use of a selective detector such as the FFPD allows us to highlight and confirm the presence of sulfur compounds such as thiophenes and benzothiophenes.
Figure 4 is the chromatogram of a kerosene showing the main groups. It is possible to combine this family grouping with more detailed analysis by carbon number within groups, as shown for paraffins and naphtenes. Figure 2 shows the relative amount of each group present, as calculated from the peak areas by the system.
In conclusion, comprehensive two-dimensional GC provides in a single run much more information than conventional GC for the characterization of complex hydrocarbon mixtures. The ability of this technique to generate an orthogonal plot based on separations in two different columns allows homologous compounds to be grouped into bands on the retention plane, according to their volatility along the first axis and polarity along the second. This makes group identification much easier, giving immediate qualitative information on the sample composition. The HyperChrom data system adds an accurate and easy-to-use tool for quantitative calculation, allowing the comprehensive characterization of petrochemical fractions.