To find a really valuable information on the topic, or at leat a good reference which could list the modifiers, is not too easy...
There is *a lot* of textbooks on organic modifiers just being %B. But I couldn't find a textbook on "how to prepare a TERNARY mobile phase with, e.g.:
H20/MeOH(ACN, THF)/...? (see topic).
It is only stated, for instance, that THF allows for creating "interesting" compositions of solvent by addition of some strong organic additive - but WHAT IS IT????????????

Thanks in advance,
ANY info (incl. exemp. applications) would be appreciated,

There is a fairly thorough discussion on the whole issue of solvent selectivity, ternary/quaternary blends and so on in Practical HPLC Method Development by Snyder. GLajch, and Kirkland. Check the link to http://www.lcresources.com/books.htm for more information (truth in advertising: I work for LC Resources).

There is a lot of mumbo-jumbo about ternary mobile phases, but the short answer to your question is:

The three water-miscible organic solvents that are the most different from one another are methanol (MeOH), acetonitrile (ACN), and tetrahydrofuran (THF). Since no one *wants* to use a ternary (two organics plus water) mobile phase if a binary phase will work (simpler is better!), the systematic strategy for developing these phases is:

1. Start with one of the binary mixtures (e.g, ACN/H2O).
2. Adjust the composition (%B = % organic) to give reasonable retention of your sample.
3. See if you can get adequate selectivity (spacing of the peaks) to meet your resolution needs by varying either %B or temperature. If you succeed, there is little point in going further (we're talking about "adequate" here, not necessarily "optimum"!).
4. If you don't succeed start over again with one of the other two solvents (e.g., MeOH/H2O). Repeat steps 2 and 3.
5. If you don't succeed this time, try again with the third solvent (presumably, THF/H2O) and repeat steps 3 and 4.
6. If you don't succeed this time, look at the best results obtained with each of the three binaries and check to see if the selectivity was significantly different (i.e., if different pairs of peaks were inadequately resolved). If the answer is "no", then there is little point in trying a ternary blend (if ACN/H2 doesn't resolve your problem peaks and THF/water doesn't resolve your problem peaks, then is it unlikely that ACN/THF/water will resolve the problem peaks!). If, on the other hand, selectivity changed significantly from one binary mobile phase to another, then it is likely that an "intermediate" ternary blend will improve things, so you try blending the best binary phases.

This is obviously an oversimplification, but I hope you get the idea. We have developed software (DryLab) that makes this approach a lot less time intensive. You can read a more detailed description of the approach at http://www.lcresources.com/dlmd.htm

Hope this helps!

-- Tom Jupille / LC Resources Inc.

This is a bit apocryphal but there was an article
that suggested getting the best sepn with ACN/H2O
and then try adding THF -no more than to 10%- to
optain the desired separation.

Ive tried it and it did wok in some cases.

At the beginning of the nineties we elaborated a terner gradient system especially for the separation of complex carotenoid mixtures:
A: water/methanol
B: methanol
C: methanol/acetone
I admit this is an unusual system but it works!
The system is described in:
Chromatographia 30(1990)318
J. Agric. Food Chemistry 39(1991)1907 and 40(1992)2072
In our more recent papers we only refer to the above publications. If you wish I send you a complete list of publications, where this terner gradient was used.
Best regards!