Chromatography Forum

Hi

Well although most of generally used reverse phase column are C18 or C4, other possible combinations such as C1, C2, C12 etc.. Can be made/used…I would like to know what types of chain length matters??? How much variation can be expected from use of lets say C8, C12 C18 and C20..??(where should we draw the line ???)


Thanks a lot

Chain length matters, but much less than other variables such as pore size, surface area, surface coverage, use of monofunctional vs. di- or trifunctional bonding reagents, etc. For example, there is no clear-cut distinction between C18 and C8 columns as a group; there is a tremendous amount of overlap in any of their characteristics.

hi tom jupille,

well i would like to know why we stick on to C8, C18 (in few cases C4
) columns??? why are there no columns such as C10, C14, C20 etc...
do such bond phase help in anyway??

Let me ask the other way round: what is the point of a C16 phase, if there is little difference from an equally prepared C18?

You can get much larger differences between different C18s: Resolve C18, microBondapak C18, Spherisorb ODS1, Spherisorb ODS2, Spherisorb ODSB, Nova-Pak C18, Symmetry C18, XTerra MS C18, XBridge C18, Atlantis dC18, Atlantis T3, Sunfire C18, Delta-Pak C18 - oh, I forgot the different pores sizes, and the versions with an embedded polar group...

do such bond phase help in anyway??

The short answer is "no".

Tom, are you saying it's not the size of the chain but what you do with it that matters?

Tom, are you saying it's not the size of the chain but what you do with it that matters?

Ouch! That one belongs down in the "Around the Water Cooler" section with all the other chromatography jokes!

Actually, though, the longer answer is that even C18 and C8 columns are more alike than they are different. A couple of years ago I had a chance to look at an early version of the "PQRI" column selectivity database that's now on the USP web site (http://www.usp.org/USPNF/columnsDB.html). There was a tremendous overlap between C8 and C18 columns; the scatter within each category is much larger than the average difference between the categories. Including C16, C14, ... would merely be adding data points in an already crowded region.

I found out first hand the wide variations you can get from C18 columns some time ago. We had a method that used an older type C18 column and were having problems getting column-to-column reproduciblity. One column would be fine, the next one borderline when new. I kept trying different C18's with newer silica, better bonding, etc. Unfortunately, none of them gave the separation of the active peaks from the matrix and degradation peaks we needed. Perhaps we needed the reactive silanols present on the older type column.