Page 171 - Multidimensional Chromatography

Unified Chromatography 163
content is more difficult. In theory, GC performed with a highly deactivated non-
polar column, would provide a high degree of orthogonalilty to normal-phase LC.
Retention on a non-polar GC column would correlate strongly with the molecular
weight (and thus the carbon number) of the fatty acids.
Now, let us expand this thinking into other realms of unified chromatography.
Suppose we have a problem similar to the fatty acids, but with solutes which for
some reason are not amenable to GC. What other techniques provide a high degree
of orthogonality to normal-phase LC but are more widely applicable than GC? We
would need a separation dominated by dispersion interactions. Thus, a highly deacti-
vated, non-polar stationary phase is still appropriate. The mobile phase should also
be either non-polar or low in polarity so that solutes are distributed on the basis of
dispersion interactions. If we needed to separate quickly in the second dimension, as
the case would be in a comprehensive, two-dimensional separation, we would need
to select a non-polar mobile phase with the highest possible diffusion rate. If GC
were not applicable, our needs would lead us toward a low-polarity mobile phase
such as CO 2 or perhaps a small alkane used at the highest temperature and lowest
pressure that would achieve our needs. Of course, there may be an overload problem
in a case like this where we have purposely mismatched very polar solutes such as
fatty acids with non-polar stationary and mobile phases in the second dimension.
Some trade-off may be necessary between orthogonality and practicality, here per-
haps by selecting a somewhat more polar stationary phase (for example, with a little
phenyl substitution) to reduce overloading. Small concentrations of a polar modifier
or of an acid or base may also be necessary in the mobile phase to produce symmet-
ric peaks, depending on the characteristics of the stationary phase and the solute
retention mechanism.
There is much that needs to be explored here. In general, it seems that to make the
most of two-dimensional unified chromatography we will have to choose the two
dimensions to have the biggest differences in selectivity that we can manage while
trying to maximize the analysis speed.
Rather than looking just for new opportunities for orthogonality, it may also be
fruitful to look for ways to make existing approaches work faster. For example, if
reversed-phase LC is already used with some degree of success in one dimension of
a multidimensional separation, it can often be made faster by increasing the tempera-
ture and lowering the viscosity of the mobile phase, such as with the addition of CO 2
or fluoroform (29). Normal-phase LC can benefit from the addition of CO 2 to the
mobile phase, up to total replacement, as long as solute retention does not go too
high. Additional selectivity may be possible by switching to a mobile phase that will
not freeze or become so viscous at low temperatures that pumping becomes impossi-
ble, then lowering the column temperature significantly. There are over 40 current
references regarding temperature effects on chiral separations using unified chro-
matography techniques, with several examples at temperatures as low as 50 °C
(30, 31).
Therefore, with unified chromatography we are not really reinventing separations
so much as taking advantage of the widening possibilities of conditions. It is