I want to understand why inverting my HPLC C18 column resulted in higher resolution of my peaks.

My eluent is 50/40/10: methanol/water/ 0.5M sodium acetate buffer pH 4.7.

I have been successful in the past in increasing resolution and restoring column performance by removing the top packing and replacing it with packing I collected from old columns.Can I assume from this increase in resolution with inverting column that the top packing is bad?

That's what I would think....

The other possibility is that there is some restriction in the inlet frit of the column which can sometimes change the column performance.

2nd Anonymous: Do you have some data on a restriction interfering with resolution even though flow was unchanged...? I have wondered for a long time whether geometrical, etc., measures (not talking about reducing voids) to "improve" the actual application of sample to the top of the column are meaningful. You may understand that I am somewhat sceptical in view of having applied as much as 8mL sample to analytical columns without change of resolution. It seems one can easily counteract geometric (spatial? or fluidics?) changes.

H. W.
No, I haven't saved any data but have observed several times where peaks have had very poor shapes even to the point of splitting due to a clogged inlet frit. I had heard this was a fairly common phenomenon though.

Regards

Because changing out the top few cm's of packing seemed to improve resolution, I would think that the issue is indeed due to the fouling of the packing near the inlet. Unless, of course, frits were changed when doing this. No mention was made of changing frits in the original post. Also, I think it's the nature of the sample relative to your system rather than the volume of it that determines whether or not you'll foul a column quickly or not. Merely my $0.02...

Chris

I am the originator of this thread,
I have since replaced one half cm of the top packing and improved resolution significantly.

There was no blockaged on the frit. I have not found blockaged to decrease resolution in my application.

A mm of the top packing was yellowish.

I lost resolution on this column (Zorbax ODS) when I used it for gradient starting with 10% Methanol and 90% 0.05M Sodium Acetate Buffer and ramp to 50% Methanol in 10 minutes.The matrix is x-ray processing developer.

It would seem you are getting something on the column from your samples that does not readily come off. What type of rinse procedure are you using when you have finished running samples? You might also try rinsing with a very strong solvent (methylene chloride or hexane but first check the column manufacturer's literature) to see if that can remove the "garbage" from the column, instead of repacking the first few mm. You could also use a guard column so when the performance is unacceptable you simply replace the guard column. Of course, it would be better if you could come up with a way to keep this material from reaching the column. Have you tried any SPE procedures?

If you check many of your other columns, you will find it very common to have different efficiencies when the flow is either forward or reverse. I believe this is due to the way columns are backed under pressure and filled from one end to the other.
Another point would also be that if there is a slight channel you could be shifting the packing bed each time you change the flow direction, the pressure would in fact temporarily eliminate the slight channel thus improving the efficiency and resolution.

There have been many reports on rejuvenating resolution by running columns backward. It is a puzzling phenomenon, the only explanation I could think of is that of the 2nd paragraph of the last anonymous: apparently the column is repacked.
Anybody investigated this?

Can anyone supply me with references on repacking HPLC columns? Or can someone describe a common repacking method? Thanks.

Here's an old one:

J. Chromatogr., v356 (1986) p420