Chemical Analysis. Francis Rouessac

for Carbowax 20M) can be used for deposition, impregnation or as grafted phases, for temperatures not exceeding 260°C, depending on the type (Figure 2.8).

      Column catalogues present optimized phases for special applications: separation of sulfur products, chlorinated pesticides, permanent gases, triglycerides, PAHs, petroleum products or fatty acid esters. Some columns accept high temperatures up to 450°C (e.g. DEXSIL 400 or PETROCOL). The original applications include simulated distillation in the oil industry, replacing conventional distillation, which can take up to 100 hours per analysis.

      

      2.6.3 Ionic Liquids (IL)

These stationary phases, developed more recently than the previous ones, are made of di‐ or tri‐cationic salts (ammonium or phosphonium) associated with sulfonate‐ or imide‐type organic anions (Figure 2.8). For GC, we use viscous ionic liquids with a melting point lower than ambient temperature, due to the fact that these are stable from a heating standpoint and not very sensitive to oxygen. These polar phases, deposited and not grafted on the column surface, are suitable for the analysis of neutral, basic or diverse compounds (fatty acid esters). Elution diagrams are different from those obtained with Carbowax, which makes them useful in two‐dimensional GC. Their near‐zero vapour pressure makes them useful for GC‐Mass Spectrometry.

Figure 2.8 Polarity scale of stationary phases in GC and summary of the three types of stationary phase. The polarity scale goes from squalane (polarity 0 by definition) to polarity 100 for TCEP (tricyanoethoxypropane). (a) Polysiloxane structure (silicones), (b) polyethylene glycols. Many phase compositions of this type, used in impregnation or grafting. (c) e.g. a dicationic tetraalkylphosphonium type ionic liquid associated with an imide (here, IL60); max. temperature 300°C.

      

      2.6.4 Chiral Stationary Phases

      An organic molecular compound, whose formula demonstrates a centre of asymmetry, leads at our scale, i.e. macroscopic, to a mixture of the two possible enantiomers R and S, in equal quantities or not. Their physical properties are identical, but their behaviour regarding polarized light is what distinguishes them, hence the name of optical activity. In space, these two forms cannot be superimposed. One is the mirror image of the other.

      The separation of optical isomers is an important application of chromatography. GC optimizes separations more quickly than other chromatographic techniques (see Chapters 4, 5, and 6). Therefore, it holds a privileged spot in this field, even though the selectivity factor for enantiomer pairs remains close to 1.

      Two separations methods exist: direct process and indirect process.

      Direct process

The analyte, supposedly composed of two enantiomers (R and S), is chromatographed on a chiral stationary phase composed of a single enantiomer (e.g. R‐type). Between the analyte enantiomers and the stationary phase, reversible interactions of the R (analyte)/R (stationary phase) and S (analyte)/R (stationary phase) are created, with slightly different stabilities. This leads to distinct migration speeds for the two enantiomers. Finally, this leads to two peaks on the chromatogram for this single analyte (Figure 2.9), whose areas reflect the abundances of the R and S forms, from which the analyte’s optical purity is calculated.

      The optical purity of the analyte refers to its enantiomeric excess (e.e.), calculated from the following relationship, where S_R and S_S refer to the areas of the two enantiomer peaks:

      (2.2)

      According to this formula, a pure chemical compound present as a racemic mixture will yield two peaks equal in size, each corresponding to an enantiomer. Its optical purity will thus be equal to 0.

Figure 2.9 Example of a separation with a chiral phase which contains grafted cyclodextrins. The use of a chiral column to separate a racemic mixture of compounds (alcohols 2 and 4). This chromatogram in isothermal mode enables the calculation of retention indexes for the separated compounds.

      Indirect process

      Unlike the direct process, this process consists in creating a chemical reaction between the analyte and a chiral reagent, prior to the injection, to form diastereomers. As these have different physical properties, they should be separable with conventional phases. This longer process is not used as often as the direct one. It may cause a partial racemization of the analyte.

Chiral stationary phases for GC are mainly obtained from cyclodextrins (Figure 2.10), ring‐shaped macromolecules composed of six, seven or eight units of D(+)‐glucose. They look like cones, with a hydrophobic central cavity, which enables the selective and reversible inclusion of a large variety of compounds, such as diastereomeric complexes.

      After possible chemical transformation, cyclodextrins are either deposited on the internal surface of the capillary column or incorporated into a poly(dimethylsiloxane)‐type polymer or grafted onto the silanol functions of silica via a short carbonated chain. These can be used up to 200°C. Above this temperature, analytes can racemize.

      As a reminder, the other chiral vectors for GC, much less in use, are crown ethers and some asymmetrical diamides, which are progressively being abandoned (Figure 2.10).

      2.6.5 Solid Stationary Phases

These phases are made from a variety of adsorbent materials: silica or alumina deactivated by mineral salts, molecular sieves 5 Å (0.5 nm), porous glass or polymers, or graphite (e.g. Chromosorb^® 100, Porapak^®). Capillary columns made by deposition of these materials in the form of a fine porous uniform layer are called PLOT. They are employed to separate gaseous or highly volatile samples. Columns containing graphitized carbon black have been developed for the separation of N₂, CO, CO₂, and very light hydrocarbons. The efficiency of these columns is very high (Figure 2.11).

Figure 2.10 Chiral stationary phase in GC. Among the three chiral vectors encountered in GC (1 and 2, β‐cyclodextrins,