Abstract
When temperature programming is performed in supercritical fluid chromatography (SFC) under conditions where the solute‐mobile phase interactions are approximately fixed, that is, with the use of constant mobile phase density, then positive temperature programs can be used. The results resemble those in temperature‐programmed gas chromatography (GC) except that chromatograms are compressed and solutes elute at lower temperatures. For a given temperature‐program rat, the extent of this compression effect is controlled by the (fixed) mobile phase density. When a methylsilicone stationary phase is used along with CO~2~ mobile phase, the compression occurs with little change in selectivity. However, with a biphenyl‐substituted polysiloxane stationary phase, selectivity shifts occur for some solutes (vs. hydrocarbons) as the density is varied. The effect is even greater and affects more solutes when a cyanopropyl‐substituted polysiloxane stationary phase is used.
This selectivity behavior differs from the temperature‐dependent selectivity behavior known in GC. In GC only the forces between solutes and the stationary phase are involved. But in SFC, mobile phase interactions with both the solutes and the stationary phase are also important. The large selectivity shifts we observe in SFC most likely arise because of stationary phase swelling by the mobile phase. This is a function of the temperature and mobile phase density (or pressure) and can be tuned without changing the composition of the mobile phase. It is possible to use either temperature or density to adjust or fine‐tune selectivity between a pair of solutes while using the other parameter to adjust retention of the pair.
These experiments further suggest that GC is not only a special case of SFC (where the solute‐mobile phase interactions are essentially zero), but that SFC possesses an additional and extremely powerful selectivity adjustment mechanism not available in GC.