linear-solvent-strength model
Posted: Mon Aug 13, 2012 4:06 pm
Can linear-solvent-strength model work for all sorts of separation mode? normal phase separation, Hilic separation?
Yes and no.
As the name implies, the model is based on the assumption that in a gradient, the "solvent strength" (log of k') changes linearly with time. This is approximately true in reversed phase, where log(k') is a linear function of %B if you use a linear ramp of %B versus time. It is certainly *not* true in ion exchange, where log(k') is a linear function of the log of the ionic strength, unless you were to use a logarithmic gradient. It is also not true in HILIC / normal phase, where log(k') can be expressed very approximately as a function of the log of the strong solvent concentration.
That said, almost *anything* is linear if you look at a narrow enough range, and LSS predictions for retention from software like DryLab are surprisingly good -- and more than good enough for practical work. In any case, the qualitative insights provided (the way retention and selectivity can vary with gradient steepness) are useful across the board.
If you really want to get into it, the Snyder and Dolan book is what you need. Here's a link to it on Amazon:
http://tinyurl.com/ct68wxz