Novel Gas Chromatographic Column Stationary Phase for Carbon Number Grouping and Challenging Industrial Applications

A novel gas chromatographic stationary phase in a wall-coated open tubular column embodiment was employed successfully for carbon number grouping of up to C₁₀. This unique cross-linked and bonded stationary phase, based on alicyclic polysiloxane chemistry, displays a polarity index even lower than the widely used 100% dimethylpolysiloxane phase. Critical pairs of probe compounds employed in a standard capillary column test mixture, such as 2,4-dimethylphenol/undecane and 2,6-dimethylaniline/naphthalene that are co-eluted on a 100% dimethylpolysiloxane phase, are well separated on this unique stationary phase with an R value greater than 1.5.

The stationary phase has a respectable maximum operating temperature of 280 ^oC and is demonstrated to have a high degree of inertness. The gas chromatographic column is suitable for use in determining heat energy value in fossil fuels like natural gas which is highly critical to industry for energy measurement, natural gas trading, and regulatory compliance. The column technology is suitable for other challenging industrial applications involving ultra-volatile and volatile molecules such as sulfur-containing compounds and oxygenated compounds.

Natural gas is classified as "natural" due to its formation from organic matter, including plants and animals, that has been buried beneath the Earth's surface for millions of years. It is one of the most significant energy sources for electricity generation, residential heating, and culinary applications. The demand for natural gas is increasing, attributed to its critical role in mitigating greenhouse gas emissions from other types of higher carbon content fuels; it is recognized as a clean-burning and highly efficient energy source that remains abundant and cost-effective (1-5).

The energy value of natural gas is contingent upon the precise measurement of various hydrocarbons based on the carbon number within the mixture. Composition and caloric value assessments are determined with British Thermal Unit (BTU) analysis using pre-established protocols published by the Canadian Gas Producer Association such as GPA-2145-16, which employs gas chromatography (GC) as an analytical technique. BTU analysis is essential for determining the energy content of fuels, which is critical for energy fair trading and pricing, for regulatory compliance, and for efficiency evaluations in various applications, including power generation and heating.

The primary constituents of natural gas consist of hydrocarbons and branched hydrocarbons. As such, a non-polar stationary phase is optimal for separating these analytes. The interaction between non-polar compounds and a non-polar stationary phase is predominantly dispersive, relying on van der Waals forces. Squalane is the gold standard for the non-polar stationary phase employed in GC; however, it is a non-bonded phase, which limits its maximum operating temperature to less than 150 ^oC, leading to excessive bleeding at elevated temperatures that has the potential of adversely affecting column longevity and analytical system performance reliability (6-12). Furthermore, the restricted availability of raw materials impedes the broader implementation of squalane as a stationary phase. Conversely, polydimethylsiloxane (PDMS) is one of the most widely utilized non-polar cross-linked and bonded stationary phases due to its low bleeding characteristics and inertness (10-19). Despite its non-polar classification, PDMS columns display a higher polarity than squalane, which complicates the grouping of hydrocarbon molecules by carbon number. This limitation may culminate in complicated analyses, the need for sophisticated chromatographic data processing software, and the potential for inaccurate calorific values for the feedstock.

In the present work, we investigate the potential of alicyclic polydimethylsiloxane (ACPDMS) as a stationary phase, aiming to advance the capability to perform distinct separations for carbon number grouping. We evaluate the chromatographic performance of this stationary phase, such as polarity and inertness, compared to the classical PDMS stationary phase. We prove that ACPDMS can successfully render carbon number grouping of at least up to C₁₀ carbon for hydrocarbons. The stationary phase provides high inertness, rendering it suitable for characterizing reactive analytes, such as sulphur-containing and oxygenated compounds.

Experimental

An Agilent 7890A gas chromatograph, equipped with an Agilent 7693 auto-sampler, two split/splitless inlets, and a flame ionization detector (FID) was employed in this study. The inlet temperature was 250 °C, operating in split mode at a split ratio of 3:1. The inlet was equipped with a 4 mm id, packed with quartz wool Ultra-Inert liner. The sample injection size was 1.0 µL for the liquid sample and 1 mL for the gas sample. A 30 m × 0.45 mm-id × 2.55 μm alicyclic polydimethylsiloxane column and a 30 m × 0.45 mm-id × 2.55 μm 100% polydimethylsiloxane column used as a reference column were obtained from Agilent Technologies. Both columns were operating under the same conditions. Unless otherwise stated, the column flow rate was 5.0 mL/min of helium in constant flow mode. The oven temperature was programmed from 40 °C (5 min) to 250 °C at 15 °C /min, and maintained at 250 °C for 10 min. The FID detector was set at 250 °C with an air flow rate of 350 mL/min, a hydrogen flow rate of 30 mL/min, and a nitrogen makeup flow of 25 mL/min. Chromatographic data were obtained with ChemStation software version C.01.07 (Agilent, Waldbronn, Germany).

Chemicals, solvents used for liquid standards, and certified liquid standards were obtained from Sigma-Aldrich (Oakville, Canada). Carrier and fuel gases such as helium, air, hydrogen, nitrogen with a grade greater than 99.999% were acquired from Air Liquide (Edmonton, Canada). Certified gas standards were purchased from BOC (Edmonton, Canada). Samples were collected in new Tedlar bags. Tedlar bags were disposed of after use to prevent potential contamination.

Results and Discussion

Stationary Phase Selectivity and Polarity

Stationary phase selectivity and polarity are paramount in column selection, significantly influencing analyte separation and retention.

From a chemical structure perspective, aliphatic cyclic groups are saturated hydrocarbon rings, comprised of C-C and C-H bonds with low overall polarity due to the small electronegativity differences and the symmetric ring structures that cancel out dipoles. In contrast, with PDMS, siloxy group (Si-O-CH₃) contains Si-O and O-CH₃ bonds. Si-O bond is highly polar due to the large electronegativity difference between Si and O. O-CH₃ bond also contributes to polarity with oxygen pulling electron density. The group has a significant dipole moment and is therefore more polar than hydrocarbon chains or rings, making PDMS more polar when compared to ACPMDS. The polarity of a stationary phase can be accurately assessed by determining the retention indices of selected probe compounds (10-12). We analyzed several probe compounds on the ACPDMS column to evaluate the polarity of this innovative stationary phase, comparing the retention index data with that of a PDMS column, as presented in Table 1.

PDMS has a slightly stronger retention for benzene when compared to ACPDMS due to its interactions with the three delocalized π-bonds in the benzene molecule. This finding illustrates that ACPDMS is indeed slightly less polar than PDMS, validating our initial prediction based on the chemical structure of the stationary phase involved.

Figure 1 showcases chromatograms that illustrate the separation of various probe compounds on both the classical PDMS column and the ACPDMS column. In Figure 1a, the PDMS column reveals that two pairs of compounds: 2,6-dimethylphenol (peak #4) and undecane (peak #5), as well as naphthalene (peak #7) and dodecane (peak #8), are co-eluted. Conversely, the ACPDMS column's higher degree of non-polarity enables the separation of undecane from 2,6-dimethylphenol with an R value exceeding 1.5. Furthermore, naphthalene, a polycyclic aromatic hydrocarbon compound with a delocalized π-bond, are eluted before dodecane, as depicted in Figure 1b. The chromatographic performance further confirms that the ACPDMS stationary phase is notably more non-polar than the classical PDMS operating under the same gas chromatographic conditions.

Also, assessing the inertness of the stationary phase is relevant. A column with a higher degree of inertness not only enables the separation process to transpire, but also effectively minimizes the analytes from being adsorbing or absorbing to the active sites. A column with excessive reactivity, such as having an excess amount of silanol groups in the stationary phase, can lead to peak tailing and/or lower detector response for the targeted compounds. Figure 1b illustrates that several challenging probes, including the basic compound 2,6-dimethylaniline (peak #6), exhibit symmetrical peak shapes, clearly indicating a respectable level of phase inertness.

Column Efficiency and Hydrocarbon Range

We analyzed a mixture of hydrocarbon standards using the ACPDMS column to demonstrate the column's efficiency and hydrocarbon range; Figure 2 highlights the successful separation of alkanes in nitrogen from C₁ to C₆ (Figure 2a) at a concentration of 0.1% (v/v) each in nitrogen and n-alkanes from C₈ to C₂₄ (Figure 2b) with C₈ at 1% (w/w), C₁₀ to C₁₆ each at 0.1% (w/w) and C₁₈ to C₂₄ each at 0.025% (w/w) in hexane using a 30 m × 0.45 mm id × 2.55 µm ACPDMS column, where all hydrocarbon compounds are eluted with symmetrical peaks.

The range of C₁ to C₂₄ covered several classes of hydrocarbon feeds and products, including natural gas, liquefied petroleum gas (LPG), gasoline, aviation gas, jet fuel, and diesel fuel. The column achieved respectable chromatographic efficiency, attaining an average of 1250 theoretical plates per meter (n=3), based on nonane (k=17) at an average linear flow velocity of 36.6 cm/s with helium at 65 °C under isothermal conditions.

Repeatability of retention time for all analytes from C₁ to C₂₄ is quite respectable with an RSD of less than 2% (n=10).With a relatively low phase ratio, this column delivers good efficiency, yielding higher theoretical plates than traditional porous layer open tubular (PLOT) columns and being on par with PDMS columns of similar dimensions (11,12).

Using temperature programming to a final temperature of 250 °C enables the ACPDMS column to elute C₂₄ effectively. The phase has a maximum operating temperature of 280 °C. While the maximum temperature is slightly lower than that of PDMS by approximately 30°C, it is more than sufficient for analyzing analytes with an equivalent boiling point of up to tetracosane (nC₂₄ at 391 °C), all the while maintaining a column bleeding profile of less than 10 pA at 250 °C. The overlaid chromatograms from three replicate injections demonstrate exceptional repeatability in both retention time and peak symmetry for all tested analytes.

Carbon Number Grouping Separation

In various applications, such as feedstock and energy calculation analyses, it is essential to accurately study branched alkanes and unsaturated hydrocarbons and determine their elution order relative to n-alkanes. Here, the PDMS stationary phase struggles as hydrocarbon isomers that differ by one or two carbon numbers can be eluted near one another, which can lead to an inaccurate BTU value for a fuel product characterization (20,21). For instance, a branched-chain compound may be eluted before a linear hydrocarbon with one fewer carbon atom. Figure 3 illustrates the separation of alkanes and alkenes up to C₆ at a concentration of 0.1% (v/v) each in nitrogen, clearly showing that alkenes are eluted earlier than alkanes.

Our studies convincingly demonstrated the ability to separate branched alkanes from straight-chain alkanes on this stationary phase. The overlaid chromatogram in Figure 4 confirms that most selected branched paraffins were successfully separated from n-alkanes, although 2,2-dimethylpropane (peak #6) and butane (peak #5) were not entirely resolved.

Analysis of Hydrocarbon Feedstock

The specified stationary phase is adept at analyzing complex hydrocarbon mixtures. An exemplary application involves cracked gas, which is a product in the manufacturing of ethylene. As depicted in Figure 5, 17 critical components within a synthetic cracked gas mixture sample are categorized into groups according to their carbon number. The repeatability of both retention times and area counts for the analytes involved were excellent with an RSD of less than 1.7% for retention time and 3.4% for area counts (n=10).

Table 2 tabulates the retention times of selected probe compounds, systematically organized by carbon number, thereby confirming a distinctive elution pattern specific to carbon numbers up to C₁₀. It is noteworthy that for carbon numbers exceeding C₁₀, the potential for discrepancies in the elution order might exist. However, within a given carbon number, the sequence of elution from this column adheres to the following hierarchy established: branched paraffins < alkenes < straight-chain alkanes < aromatics.

In numerous industrial applications that employ online analyzers, isothermal GC oven conditions are often preferred, as this approach eliminates cooling periods and enables immediate initiation of subsequent runs, thus maximizing sample throughput to enable data-driven decisions to be made. Also, many online analyzers lack the capability for temperature programming. Even under isothermal conditions, this column facilitates carbon number-specific separation, as evidenced by the overlaid chromatograms of alkanes, aromatics, and a synthetic cracked gas mixture sample shown in Figure 6. Consequently, depending on specific requirements and hardware capabilities, the analysts have the flexibility to opt for either temperature programming or isothermal operation for the application described.

Volatile and Semi-Volatile Sulphur Compounds

The analysis of sulphur compounds is a challenging chromatographic application due to their potential reactivity and adsorption to the active sites in the chromatographic system. As demonstrated in Figure 7, alkyl (C₁ to C₄) mercaptans at a concentration of 0.01% (v/v) in nitrogen exhibited peak symmetry consistent with much less reactive molecules like alkanes.

Additional heavier sulphur compounds, such as alkylated thiophenes and benzothiophenes at a concentration of 0.1% (w/w) in hexane were analyzed. Figure 8 shows the symmetrical peak shapes for all the compounds cited, demonstrating the suitability of using the column for sulphur compound analysis (n=7).

Volatile Oxygenated Compounds

Oxygenated compounds of industrial importance are known to be more active due to the hydrogen bonding reaction that can take place. Analyzing these molecules can be challenging with tailing peaks and loss of analytes at low concentration levels. Therefore, an inert column is critical for trace and accurate quantitative analysis. As shown in Figure 9, respectable separation and peak asymmetry were obtained for some common oxygenated compounds at a concentration of 0.01% (w/w) in water with a 1 mL injection of the headspace of the sample. The new stationary phase showed high inertness even with active molecules such as small chain primary alcohols and aldehydes.

Conclusions

With ACPDMS as a stationary phase, carbon group-specific separation of up to C₁₀ can be achieved. To our best knowledge, the stationary phase is the only cross-linked and bonded phase employed in wall-coated open tubular column technology to attain such a unique advantage. The capability is critical for energy calculations for hydrocarbon feedstock. In terms of polarity, the stationary phase was found to be more non-polar than that of 100% PDMS, a gold standard in high resolution capillary GC. The extra non-polarity can be exploited to resolve critical pairs encountered with 100% PDMS. The inertness of the stationary phase is equivalent to the classical PDMS column, making it suitable for use in other critical applications such as the characterization of reactive sulphur and oxygenated compounds.

Along with other commercially available stationary phases employed in high resolution GC, ACPDMS as a stationary phase has a unique advantage in complementing and contributing to the GC stationary phase portfolio to achieve faster, better, and affordable separation for the applications described herein.

Acknowledgments:

Allen Vickers and Mitch Hastings of Agilent Technologies are acknowledged for fruitful discussions and valuable advice on stationary phase chemistry. Yujuan Hua and Grace Yang are acknowledged for their invaluable help and manuscript preparation. Last, but not least, Narayan Ramesh, Linh Le, Jaime Curtis-Fisk, also of Dow, and Mike Zhang of Agilent Technologies are acknowledged for their support and encouragement.

References

1. Faramawy, S.; Zaki, T.; Sakr, A. A. Natural Gas Origin, Composition, and Processing: A Review. J. Nat. Gas Sci. Eng. 2016, 34, 34–54. https://doi.org/10.1016/j.jngse.2015.06.030
2. Economides, M.; Wood, D. A. The State of Natural Gas. J. Nat. Gas Sci. Eng. 2009, 23 (1), 1–13. https://doi.org/10.1016/j.jngse.2009.03.005
3. Rimos, S.; Hoadley, A. F. A.; Brennan, D. J. Resource Depletion Impact Assessment: Impacts of Natural Gas Scarcity in Australia. Sustain. Prod. Consum. 2015, 3, 45–58. https://doi.org/10.1016/j.spc.2015.08.003
4. Li, L. Development of Natural Gas Industry in China: Review and Prospect. Nat. Gas Ind. B 2022, 9 (2), 187–196. https://doi.org/10.1016/j.ngib.2022.03.001
5. Rivard, C.; Lavoie, D.; Lefebvre, R.; Séjourné, S.; Lamontagne, C.; Duchesne, M. An Overview of Canadian Shale Gas Production and Environmental Concerns. Int. J. Coal Geol. 2014, 126, 64–76. https://doi.org/10.1016/j.coal.2013.12.004
6. Desty, D. H.; Swanton, W. T. Gas–Liquid Chromatography—Some Selective Stationary Phases for Hydrocarbon Separations. J. Phys. Chem. 1961, 65 (5), 766–774. https://doi.org/10.1021/j100823a015
7. Poole, C. F.; Pomaville, R. M.; Dean, T. A. Proposed Substitution of Apolane-87 for Squalane as Nonpolar Reference Phase in Gas Chromatography. Anal. Chim. Acta 1989, 225, 191–203. https://doi.org/10.1016/S0003-2670(00)84607-3
8. Castello, G.; D’Amato, G. Classification of the “Polarity” of Porous Polymer Bead Stationary Phases by Comparison with Squalane and Apolane Standard Liquid Phases. J. Chromatogr. A 1983, 269, 153–160. https://doi.org/10.1016/S0021-9673(01)90798-8
9. Pellizzari, E. D.; Sparacino, C. M. Capillary Gas Chromatography of Linear Alkenes on Squalane. Anal. Chem. 1973, 45, 293–302. https://doi.org/10.1021/ac60324a039
10. Abraham, M. H.; Poole, C. F.; Poole, S. K. Classification of Stationary Phases and Other Materials by Gas Chromatography. J. Chromatogr. A 1999, 842 (1–2), 79–114. https://doi.org/10.1016/S0021-9673(98)00930-3
11. de Zeeuw, J.; Luong, J. Developments in Stationary Phase Technology for Gas Chromatography. TrAC, Trends Anal. Chem. 2002, 21 (9–10), 594–607. https://doi.org/10.1016/S0165-9936(02)00809-9
12. Luong, J.; Gras, R.; Jennings, W. An Advanced Solventless Test for Capillary Columns. J. Sep. Sci. 2007, 30 (15), 2480–2492. https://doi.org/10.1002/jssc.200700131
13. Jennings, W.; Mittlefehldt, E.; Stremple, P. Analytical Gas Chromatography, 2nd ed.; Academic Press: New York, 1997. ISBN 978-0123843579.
14. Jennings, W.; Nikelly, J. Capillary Chromatography—The Applications; Hüthig: Heidelberg, 1991. ISBN 3-7785-2051-2.
15. Bertsch, W.; Jennings, W. G.; Kaiser, R. E. Recent Advances in Capillary Gas Chromatography; Hüthig: Heidelberg, 1981. ISBN 3-7785-0711-7.
16. Bertsch, W.; Jennings, W. G.; Kaiser, R. E. Recent Advances in Capillary Gas Chromatography, Vol. 2; Hüthig: Heidelberg, 1982. ISBN 3-7785-0712-5.
17. Poole, C. F. The Essence of Chromatography; Elsevier Science: Amsterdam, 2003. ISBN 0-444-50198-0.
18. Luong, J.; Gras, R.; Cortes, H.; Shellie, R. A. Multi-Dimensional Gas Chromatography with a Planar Microfluidic Device for the Characterization of Volatile Oxygenated Organic Compounds. J. Chromatogr. A 2012, 1255, 216–220. https://doi.org/10.1016/j.chroma.2012.01.073
19. Luong, J.; Gras, R.; Hawryluk, M.; Shearer, R. A Brief History and Recent Advances in Ozone-Induced Chemiluminescence Detection for the Determination of Sulfur Compounds by Gas Chromatography. Anal. Methods 2016, 8, 7014–7024. https://doi.org/10.1039/C6AY01887D
20. Speight, J. Handbook of natural gas analysis; John Wiley & Son, 2018, ISBN 978119240280.
21. Ulbig, P.; Hoburg, D. Determination of calorific value of natural gas by different methods. Thermochim. Acta, 2002, 382, 27-35. https://doi.org/10.1016/S0040-6031(01)00732-8.