LITMIR.BIZ

Figure 2.7 Capillary columns. Representation of a 50 m long commercial capillary column coiled around its metal support (Document from the Alltech Associates Inc.) and close‐up of a column. At this scale, the thickness of the stationary phase will scarcely be visible.

While less efficient than capillary columns, owing to preferential paths (Eddy factor of the Van Deemter curve), they are still used for some standardized routine analyses. Nevertheless, these are not adapted to trace analyses, and the progressive generalization of GC‐MS coupling has greatly reduced their use.

      Porous supports are made of spherical particles of around 0.2 mm in diameter. These are obtained from fossilized silicates (kieselguhr, tripoli) whose backbone is chemically comparable to that of amorphous silica. One of the most common goes by the name of Chromosorb^®. Their specific surface area is variable (between 2 and 8 m² /g).

      2.5.2 Capillary Columns (Open Tubular)

They are usually made of the highest purity fused silica obtained by the combustion of tetrachlorosilane (SiCl₄) in an oxygen‐rich atmosphere. The internal diameter of the tube used for these columns varies from 100 to 530 μm. The technology is particularly delicate in order to obtain perfectly cylindrical columns with a thickness of 50 μm and lengths going up to 100 m (Figure 2.7). These columns have a polyimide outer coating, which is a thermally stable polymer (T_max = 370°C), to make them less fragile, both chemically and mechanically. They can be wound into coils around a lightweight metal circular support. Some manufacturers offer columns made from a metal capillary (aluminium, nickel or steel) which tolerates high operating temperatures on the order of 450°C, providing that the stationary phase is stable enough. The internal surface of the column is usually treated to promote good bonding for the stationary phase. This could be a chemical treatment or the deposit of a thin layer of alumina or silica gel.

      The regular thickness of the stationary phase can vary between 0.05 and 5 μm. It is either simply deposited or better yet grafted with covalent bonds, possibly followed by a polymerization with cross‐linking on the wall. This deposit is obtained by evaporating a solution or by polymerization in situ in contact with the wall. These are WCOT (wall‐coated open tubular) or PLOT (porous layer open tubular) columns, depending upon the nature of the stationary phase employed. Columns are particularly stable and can be rinsed periodically with solvents, which enable them to recover their initial performance levels.

      To compare or anticipate the behaviour of capillary columns, it is useful to know the phase ratio β = V_M / V_S. By designating ID as the internal diameter of the column and d_f as the thickness of the deposited film, an approximate calculation leads to:

      (2.1)

      The column’s capacity is related to the phase ratio but also, for each solute, to its retention time, since k is inversely proportional to it. The phase ratio β, accessible from the physical characteristics of the column, and k (retention factor) from the chromatogram, help us calculate the partition coefficient K of a solute, whose value is generally quite high (1,000 for example) owing to the nature of the mobile phase (gas).

      WCOT columns are well suited to mass spectrometry detection and are thus used often. To deposit a film of known thickness, a method consists in filling the column with a solution of stationary phase of known concentration (e.g. 0.2% in ether), so that the desired thickness is obtained after solvent evaporation. This layer can then be cross‐linked by a peroxide or by γ irradiation. The process is similar to the application of paint on a surface that has been pretreated to obtain good adherence.

2.6 STATIONARY PHASES

      For packed columns, impregnation or deposit techniques can lead to many different stationary phases from a large selection of low‐volatility organic compounds. On the other hand, for capillary columns, manufacturing constraints require a much more limited selection of compounds. The current phases correspond in principle to three families: polysiloxanes, polyethylene glycols, and ionic liquids. Each category can have many structural variants. For the study of optically active compounds, specific phases are used.

Each of these phases can be used between a minimum temperature, below which concentration equilibria occur too slowly, and a maximum temperature, above which degradation of the polymer occurs. The high limit depends on the film thickness and the nature of the polymer.

      Squalane is used as a reference phase, since it is the only one that is perfectly defined. On the McReynolds scale, squalane has a polarity of zero (Section 2.10.3). This saturated hydrocarbon (C₃₀H₆₂) is derived from squalene, a natural terpene extracted from shark’s liver. On this stationary phase, which can be used between 20 and 120°C (using either deposition or impregnation), the compounds are eluted in increasing order of their boiling points (retention time being inversely proportional to vapour pressure). Various grafted phases based upon polyalkylsiloxanes are also almost apolar.

      2.6.1 Polysiloxanes

For capillary columns, polysiloxanes (also known as silicone oils and gums) are the mostly commonly used stationary phases. They are based upon a repetitive backbone that consists of two hydrocarbon chains per silicon atom (Figure 2.8). Dozens of different compositions of alkyl or aryl chains (methyl or phenyl), onto which can be incorporated further functional groups (e.g. cyanopropyl, trifluropropyl), are available on the market. Monomers combined in various proportions also convey changes in the properties of stationary phases (polarity, stability range). As such, dimethylpolysiloxanes have a very large temperature range, from −50 to 300/325°C.

      2.6.2 Polyethylene Glycols (PEG)

      The best known representative of this family is Carbowax^®. These polar polymers (M_r

Скачать книгу