Clean IPC Column

[Sneezy]

Running IPC HPLC analysis using octanesulfonic acid sodium salt about 15mM(buffer) 90:10 (buffer:MeOH) as MP A, MP B = 100% MeOH
Linear Gradient like:
Time MPA MPB
0 100 0
4 100 0
35 95 5
55 75 25
70 40 60
71 100
80 100

Flow Rate 2.0 mL/min
Lambda: 210nm
Column: Phenomenex Luna C18(2) 3um, 4.6 x 75mm
Column Pressure: 2300 to 4300 psi
Column Temp: 40 min
Inj Vol: 10 uL

Getting about 50 injections per column. Thinking the IP reagent is not being properly removed. Cleaning involves cleaning through 20:80 MeOH:Milli-Q, to 80:20 MeOH:Milli-Q, 60:40 MeOH:Milli-Q @about 0.5 ml/min for 180 min.
IPC analysis must separate 9 compounds where R=NLT 1.0.
New out of box column failed. Less than 50 injections.
Fresh 2nd (New Bed) new column failed. LT 50 inj.

Method Validated on one column. Suggestions?

[Gerhard Kratz]

Using IP reagent means a modification of the stationary phase. Mostly irreversible.
No chance to clean the column material. Sorry.
The bonding of the IP reagent is not very reproducible. A column you used with IP reagent should not be used for other method development.

[tom jupille]

Gerhard's point is a good one, but if it's a validated method, you have no choice.

A bit more information would help:
1. Is the loss in resolution due to changes in retention time or broadening of the peaks?
2. Is the change gradual over 50 injections or is it abrupt?
3. If it's gradual, is the deterioration more closely related to number of injections or to time in the mobile phase (e.g., if you did 25 injections, then ran 25 "dummy" injections, would the column be as dead as if you had run 50 real samples?

[Gerhard Kratz]

Once I had a similar problem when I did a validation with a column which was already used with IP reagent. I got an excellent baseline separation with sharp peaks, it looked wonderful. Suddenly I saw peak splitting and peak shoulders. I had to use a brand new column and now no baseline separation and broad peaks. I had to start new with the method development.
I decided to use 3 columns, if possible with 3 different lots of the packing material, to do validation. Lab manager was not happy but later he agreed that it was the right decision.
50 injections on one column, and you cannot change the method, than you are a good column customer.........
Take it easy, if the final product is generating good margin for your company than it is ok, but I'm not your lab manager.
Do you use a guard column? Maybe a so called saturator column will increase the number of injections.
Take a plain silica guard column and install it between pump and injection system. Since it is not part of the separation you can do that without doing a revalidation. Good luck.

[danko]

How does a saturation column fit into the picture here?
Do we fear silica dissolution, or C18 ligands maybe?
I don’t think so.

Anyway the solution of the problem here as I see it, is not cleaning/removal of the IP. When using IP on a column dedicated to the method, you do not need to remove the IP and introduce it again next tame you run the method. It only prolongs the start-up phase/step.
Experiencing short column life is not always related to stationary phase degradation. Often it is seen when the sample (or a part of it) is not completely eluted and that is a completely different problem, necessitating completely different actions.
If we get some more information there could be another solution than trying to remove IP residues from the column.

Best Regrds

[Jeffetw]

Are you able to try other solvents (acetonitrile, isopropanol) for cleaning and then re-equilibrate the column in the methanol/buffer solution?

[HPLCaddict]

I'm completely with Danko here. Getting rid of the IP reagent is not the solution. You WANT to have the IP reagent on the column, it's a crucial part of the method.
Are you maybe tricked by the "old" model of ion-pairing chromatography, i.e. the reagent and the analyte form an ion-pair, which travels down the column? This is only part of reality. It's a bit more complex, actually a considerable amount of the IP reagent is adsorbed to the column, forming a dynamic ion-exchange stationary phase. So the IP reagent HAS to stick to the column for your method to work.
You have to investigate, if it's part of your samples which are fouling the column, or if it's a component of the mobile phase (could be, of course, some impurity in the IP reagent). Tom pointed out, what to do. Once you know if it's your samples or the mobile phase, more systematic troubleshooting will be possible. THEN it's the time to either adress your sample preparation or the mobile phase components.
What's the pH of your mobile phase, by the way?

[Gerhard Kratz]

http://www.nestgrp.com/pdf/Vapp/Vcolcare/Vcolcare.shtml Some helpful hints to column care.