Low Bleed Columns

As the column temperature is increased, there is an increase in the steady-state baseline signal. The chemical bonds of the stationary phase are under increased thermal stress, degradation fragments are produced at an increased but constant rate, and the baseline signal rises and remains steady as long as that temperature is maintained. In polysiloxane stationary phases, the degradation fragments consist largely of cyclic siloxanes, dominated by trimers and tetramers of (-Si-O-) to which the methyl (or other) substitu-ents occupying the remaining two bonds of the tetra-valent silicon atom may remain attached, or may cleave. True column bleed does not generate peaks or humps; a signal that rises and then falls must have a 'point source'. Bleed signal is annoying with any bleed-sensitive detector, and can be the limiting factor in GC-MS, especially with the newer ion trap mass spectrometers (ITD). For a given quality of stationary phase under a given set of conditions (temperature, carrier gas flow), bleed is always a function of the mass of stationary phase in the flow path. Hence, shorter, smaller diameter, thinner film columns will exhibit lower bleed levels than longer, larger diameter, thicker film columns.

At the present time, column bleed is usually reported in terms of pA of FID (flame ionization detector) signal at a given temperature, but this is an imprecise specification. It would be better to report bleed as 'pg carbon emitted per unit time' as measured on a calibrated detector. Only where the same detector is used under the same conditions for both determinations can different columns be truly compared in terms of 'pA of FID signal'. Bleed signal from polysiloxane columns is generally attributed to cyclic siloxanes that usually arise from thermal and/or oxi-dative degradation of the phase, but contaminants in the detector or in gas lines supplying that unit, materials outgassing from septa and ferrules, and contaminating oils from column installation also contribute to what is perceived as column bleed. The latter sources are usually (but not always) dwarfed by the former, but their significance increases in the case of low bleed columns. It should also be noted that at elevated temperatures, even a pristine FID without any column generates 1-2 pA signal.

Some years prior to the invention of gas chromatography, Sveda, a Du Pont chemist working on bulk polymers, filed two patents on silarylene-siloxane polymers. Shown in Figure 2 are generic structures of (A) the conventional 95% dimethyl-5% diphenylpolysiloxane (in which the phenyl groups are pendant to the siloxane chain), and (B) Sveda's poly(tetramethyl-1,4-silphenylene siloxane).

Not surprisingly, the two forms display somewhat different selectivities toward solutes. More than 40 years after Sveda's work, several column manufacturers began offering proprietary 'low-bleed' columns that would appear to be based on the silarylene siloxanes. The lengthy delay may have been partly due to the fact that, until quite recently, bleed rates from high quality 'conventional' columns were not considered excessive in most applications. A more probable cause is that silarylene polymers are not yet commercially available, so that those wishing to utilize them must resort to a sequential series of in-house syntheses that are both materials and labour intensive, each step of which is usually characterized by low yields.

Figure 2 Generic structures of (A) the conventional 95% dimethyl - 5% diphenylsiloxane and (B) Sveda's poly (tetramethyl-1,4-silphenylene siloxane).

Several pathways have been postulated for the thermal degradation of siloxanes, some of which require a proximity of two normally separated groups that would require folding of the siloxane chain. It has been suggested that such reactions might be blocked by the insertion of groups that would make the chain more rigid and restrict its flexibility. Such efforts have led to the introduction, by more than one manufacturer, of stationary phases that are characterized by the generation of lower bleed signals, even at elevated temperatures, e.g. 360°C.

Figure 3 contrasts the bleed profiles of examples of 'first generation' (a silphenylene siloxane) and 'second generation' (a silphenylene siloxane containing 'chain-stiffening' groups) low-bleed columns. Note that the latter not only exhibits a bleed level about half that of the former, but the bleed pattern is simpler, making the phase especially valuable for those utilizing bleed-sensitive detectors such as the ion trap mass spectrometer. It also possesses a unique selectivity that has excited great interest among those interested in a variety of problematic separations, including the polychlorinated biphenyl congeners.

Column performance is also influenced by the deactivation, or surface preparation treatment. The observation that thin-film columns often exhibit adsorption toward active solutes, and that thicker films of stationary phase result in more inert columns,

Figure 3 Chromatograms of a column test mixture on 30 mx 0.25 mm, d 0.25 |im columns. Test mixtures were run at 130°C isothermal, and columns ramped to higher temperatures to determine bleed profiles. Note that the 'delta bleed' is 2.32 pA at 320°C for the 'second generation' column, versus 14.42 pA for the 'conventional' column. This is essentially the 'maximum temperature' of the latter column, which has reached a point where bleed increases exponentially with temperature. The latter column is capable of still higher temperatures, however, and exhibits a bleed signal of 5.47 pA at 340°C. Key: —, first generation 35% diphenylpolysiloxane column;-, second generation 35% phenyl low-bleed column.

Figure 3 Chromatograms of a column test mixture on 30 mx 0.25 mm, d 0.25 |im columns. Test mixtures were run at 130°C isothermal, and columns ramped to higher temperatures to determine bleed profiles. Note that the 'delta bleed' is 2.32 pA at 320°C for the 'second generation' column, versus 14.42 pA for the 'conventional' column. This is essentially the 'maximum temperature' of the latter column, which has reached a point where bleed increases exponentially with temperature. The latter column is capable of still higher temperatures, however, and exhibits a bleed signal of 5.47 pA at 340°C. Key: —, first generation 35% diphenylpolysiloxane column;-, second generation 35% phenyl low-bleed column.

suggests that activity sometimes depends on whether solutes 'see' the surface. As solute molecules migrate through the stationary phase toward the siliceous surface, the carrier flow sweeps the solute molecules in the gas phase downstream. To re-establish the distribution constant in that portion of the column, solute molecules in the stationary phase reverse direction and migrate toward the mobile phase. Whether this impetus to change the direction of migration occurs before or after the solute has reached the surface would, under a given set of conditions, depend on the thickness of the stationary phase film.

The affinity of the surface for polar stationary phases is sometimes estimated by measuring the surface energy of the prepared surface. While both the nature and strength of surface-to-polymer interactions are important to column performance, they are not necessarily predictable. A given 'high energy' surface is not always suitable for the deposition of a 'high energy' polymer. Architectural 'tailoring' of the surface can be as difficult as tailoring stationary-phase selectivity. The concept of 'coating efficiency', used in Figure 4, (CE = 100 xH^m^/ [H^ved]) is often used to measure the compatibility of a surface for a given stationary phase.

Figure 4 shows four dimensionally identical columns under the same operational conditions and coated with the same experimental high phenyl stationary phase; each column received a different deactivation treatment. Note the third example (C),

Figure 4 Chromatograms of a test mixture on four dimensionally identical columns, all coated with the same experimental high phenyl stationary phase, but subjected to different deactivation pretreatments. Note the disappearance of 2-ethylhexanoic acid in (D) and the intercolumn variations in retention indices (/) for methylnaphthalene and undecanol. See text for discussion of the effects of the surface energies (y) on coating efficiencies (CE). Solutes in order of elution: 1, 2-ethylhexanoicacid (totally adsorbed in chromatogram D); 2, 1,6-hexanediol; 3, 4-chlorophenol; 4, tridecane; 5, 2-methylnaphthalene; 6, 1-undecanol; 7, tetradecane; and 8, dicyclo-hexylamine.

Figure 4 Chromatograms of a test mixture on four dimensionally identical columns, all coated with the same experimental high phenyl stationary phase, but subjected to different deactivation pretreatments. Note the disappearance of 2-ethylhexanoic acid in (D) and the intercolumn variations in retention indices (/) for methylnaphthalene and undecanol. See text for discussion of the effects of the surface energies (y) on coating efficiencies (CE). Solutes in order of elution: 1, 2-ethylhexanoicacid (totally adsorbed in chromatogram D); 2, 1,6-hexanediol; 3, 4-chlorophenol; 4, tridecane; 5, 2-methylnaphthalene; 6, 1-undecanol; 7, tetradecane; and 8, dicyclo-hexylamine.

which has high surface energy but very poor coating efficiency. (A), (B) and (D) all are nonbeaded surfaces, but the coating efficiencies of (A) and (B) are significantly better than that of (D). The coating efficiencies of these three columns vary in the order 2 > 1 > 4, while the surface energies vary in the order 1 > 4 > 2. The column with the lowest surface energy of the coatable surfaces (column B), yields the highest coating efficiency. In column (D), the acid peak disappeared. In every case, there is almost surely at least a slight effect on selectivity. One of our better ways for estimating this quality is the duplicability of retention indices for polar and apolar compounds, and column-to-column variations in these comparisons imply that closely eluting solutes of different functionalities may exhibit a given elution order on the one column, and a different elution order on the other column. These data indicate that both the stationary phase (which may be proprietary) and column pre-treatments (which are almost always proprietary and vary from manufacturer to manufacturer) affect retention factors, separation factors, and even the elu-tion order. Surface pretreatments, including but not limited to deactivation, can, and often do, exert profound effects on overall column performance.

Solar Panel Basics

Solar Panel Basics

Global warming is a huge problem which will significantly affect every country in the world. Many people all over the world are trying to do whatever they can to help combat the effects of global warming. One of the ways that people can fight global warming is to reduce their dependence on non-renewable energy sources like oil and petroleum based products.

Get My Free Ebook


Post a comment