"Discontinuous" (multicolumn) RP-HPLC

[pipettemonkey]

In Chromatographia Vol. 25 189, three RP columns (C1, C4, C18) are connected in a series (in order of increasing retention) and used to separate ribosomal proteins. The authors call this "discontinuous RPC." There is no valve between the columns. Instead, the whole system is run with a single lengthy gradient as in a single column system.

Any thoughts on this? At least in the paper, the 3 column system seemed to increase resolution compared to any single column separation. However, not many people have followed their lead (can't find a paper that cites this work).

[tom jupille]

In a gradient, peaks accelerate as they move down the column, therefore spending a higher percentage of time near the head of the column than near the tail. And big molecules accelerate more precipitously than do small molecules. Without seeing the actual paper, my guess is that coupling columns in this way (less retentive first) tends to even out the time spent in each segment. Interesting and clever approach, but a lot of work for what it does.

[bisnettrj2]

Doesn't Bischoff Chromatography offer a system similar to this, wherein they offer several different phases of columns that couple to each other in order to influence separations?

viewtopic.php?t=5522&highlight=bischoff

viewtopic.php?t=7119&highlight=poplc

Granted, the system Bischoff offers is for different phases, rather than the similar phases offered in the cited paper (C1, C4, C18). I would wonder, then, based on the two threads cited (amongst probably many here and elsewhere) whether band broadening between columns, reduced plate height, and counteractive selectivity might be a problem in a system like this.


Also, from Uwe Neue on another similar thread:

viewtopic.php?t=6859&highlight=poplc

"A bit of caution: The additivity of retention times works well when you combine columns in isocratic separations. It is more complicated in gradients. If the two columns have very similar retention, then there is a reasonable chance that you get new chromatograms with some form of scrambled selectivity. Also, in your studies switch the sequence of the columns. This can give you different retention patterns also. But don't be surprised if 1 + 1 does not equal 2 in gradients with different columns. Good luck!"

Personally, I use a two-column isocratic system in my work that I inherited, where we use a column designed for carbamate pesticides in series with a C-18. Works better than either column alone, except I don't have a 375 mm C-18 or 375 mm carbamate column to test that statement, so...

[Uwe Neue]

I looked at the paper. It is a very complicated technique. From the standpoint of the practitioner, nothing is predictable. You get different results depending on the sequence of the columns. I would call this technique "randomized empiricism", at least in the form presented in these papers.

[Kostas Petritis]

Although I did not read the paper, you would think that if the only thing that changes among those column is increased carbon chain length (i.e. increased hydrophobicity) compounds would just stack at the head of the last column until the appropriate elution strength is applied for them to be eluted, the same as if you would just had only that column (i.e. no resolution gains).

But experimental results are experimental results...

[danko]

I agree with Kostas 100 %. It is just – I would call it - â€

[lmh]

... and of course it increased resolution. How often do you see papers that say "we did something really complicated and the resolution got worse?". It'd be interesting to know if it reliably improves resolution, or if others have tried it and failed? We need that Journal of Negative Results again.

[bisnettrj2]

lmh - would that journal be published weekly or daily? And electronic only, I presume, as I think it would be too cumbersome to ship something that thick that often...

[Bryan Evans]

Or to add:

"The Journal of Proprietary Accomplishments"

What % of scientific achievements in the private industry actually gets published?