Yours Chromatographically, Kaushik Zala

A reader recently asked a question about the use of the peak-purity function of his diode-array UV detector (DAD). The question related to
whether or not he could detect the presence of enantiomers, stereoisomers, diastereoisomers, or epimers with the peak-purity function.


The peak-purity function of DADs is quite simple in concept. UV spectra are
taken at various points across a chromatographic peak and compared. If the
spectra are sufficiently alike, the peak is considered pure; if the differences in
spectra are large enough, the peak is not pure. With some peak-purity outputs,
all you get is an indication of the peak purity, sometimes called a match angle
or purity angle. With sophisticated peak-deconvolution software, the peaks
can be measured separately, in much the same manner as an LC-MS detector
can distinguish between two co-eluting peaks of different molecular weights.
Peak-purity measurements sound good in principle, but most workers are
disappointed in the actual results. The reason for this is that several challenges
exist in most chromatograms. First, the peak shape of most chromatographic
peaks are not an ideal Gaussian distribution, but tend to tail, so peak distortion
does not necessarily mean that more than one compound is present. More
important, though, is the nature of co-eluting peaks. If two peaks co-elute or
have very similar retention times in a single sample, there is a good chance
that they are chemically related. Compounds that are chemically related
usually have similar UV spectra. Further complicating spectral similarity is
the fact that many of the compounds that we’re interested in analyzing have
very “uninteresting” UV spectra. That is, they don’t have much spectral detail,
with strong absorbance <210 nm, but little to distinguish them at higher
wavelengths, so even under ideal conditions, spectral differences may be
minimal. And one more complicating factor is concentration differences. It
may be possible to distinguish between two fairly similar compounds if their
concentrations are large enough, but if one compound has a concentration
<1% of the other, it will be much more difficult to distinguish between two
overlapping peaks with peak-purity algorithms.


The bottom line is that the various isomers mentioned in the reader’s
question are likely to have very similar, if not identical, UV spectra. If this is the
case, even under ideal circumstances (large concentrations, similar peak sizes),
it will not be possible to differentiate between isomers using a peak-purity function. Even LC-MS is likely to fail to distinguish the peaks in many such
cases.
This is a place where chromatography wins! Use a chiral column to
help separate chiral compounds. Normal-phase chromatography (NPC)
usually will be more successful at separating isomers than reversed-phase
chromatography (RPC). This is because RPC tends to separate on sample
polarity, which will differ very little between isomers. In contrast, NPC is based
on adsorptive interactions between the solute and the column surface. These
interactions tend to differ based on the shape of the molecule, so one isomer
tends to interact more strongly with the column than another, allowing
separation to take place.


YOURS CHROMATOGRAPHICALLY

KAUSHIK ZALA