HPLC Column Performance Check.

[iwan.kenten]

Dear colleagues,

I would like to have any comment from forum on handling HPLC's column.
What would be the best practice to decide time to dispose HPLC's
column. I can not find this issue on any compandials I have.

Normally I do performance check to every column we have prior using
according to its manufacturer. Then I record the number of N plates
yield on first usage. After several usages for same material, at
anytime the number of N plates drops to half of its first usage, we
dispose the column.

The problem is sometime it is only 5 to 10 times usage, the number of
N plates for that material already drops to half of its first number,
and it becomes so expensive to purchase another new column.

What would be the better way, continue what we already do, or we have
to decide time interval or usage frequency to repeat column
performance check to get decision to continue using the column or
dispose it.

Thank you in advance,

Iwan Kenten

[tom jupille]

There is no single answer: it depends on what assay you are using the column for. One analysis may require 10,000 plates, while another may work just fine with 500 plates. Ideally, you should have "system suitability" criteria for each analysis. When a column no longer meets system suitability, it should be replaced for use in that assay.

[goxy43]

Hello,
I also do think that you must have a separate column test assay. What we do (proteomic analysis) is injecting a tryptic digest of cytochrom c before every batch. Knowing the peptides and retention times it is easy to check the performance. However, we are running a gradient separation and here plate number does not make much sense, therefore, we check on peak width and retention time. If you run isocratic runs the best way to test the column, from my point of view would be to use two or three substances as standard and run the analysis on them, even daily if necessary.

[mbicking]

First, as others have suggested, you should monitor performance more often. Daily is a good idea. And you should act before your plate number decreases by 50%. John Dolan has written about this in his troubleshooting columns often, and I like his idea of a 10 % change as the indication that a change is necessary.

But you should try to determine why your columns fail after only 10 uses. This fact suggests that there are problems with your method or laboratory procedures that are destroying your column.

[Bryan Evans]

What experimental conditions are you typically using?
Do you see an increase in pressure as well as decrease in N?

You should definetely investigate as to why you are only getting
10 injections (harsh experimental conditions, sample prep, ect.)

[syx]

We decide to dispose a HPLC column when it could not be used in every method which use that column. It is based on the system suitability parameters of the methods. If the column is failed to meet the system suitability test of a method we could not use it in that method. However, there is possibility that the column still could be used for other methods.
If a column becomes worst, one may try to regenerate it and then do the
performance check. The result is compared with initial check when we
received the column.
If your columns have very short lifetime when perform a method maybe there is something wrong with your procedures. If it is occured in every method, perhaps you should evaluate your column maintaining procedures.