Help on: When Should I Throw my HPLC Column Away?

MTS HELPDESK

Do you have any problems relating to analytical chemistry for pharmaceuticals or training? Send your questions to the MTS helpdesk using our contact form.

Question:
"If the value for the theoretical plates for an HPLC column decreases below 2000, should the column be discarded? And which parameters should be considered when deciding when to discard a column?"

Answer:
"Unfortunately the answer to your question is - it depends. Deciding when to get rid of an HPLC column is difficult and may depend on what the column is being used for and whether it is being used for a single type of method or lots of different methods and samples. The theoretical plate count (N) is a good indication of how well a column is working but it is relative, this means that you need to monitor the value over the course of the column lifetime and correlate the value with the decrease in separating power.

If the chromatography for a method looks good and the peaks are still separated adequately for quantitative analysis then I would not discard a column because the value of N went below 2000. However, from experience for a particular column you may know that the method does not work as well once N is below a certain value and thus this could be your indicator that the column should be discarded. It is definitely easier to use this approach if the column has only been used for one type of analysis. Once a column is used for lots of different samples then it is very difficult to predict when it will fail. 

There are a few things to be aware of if you are using plate number. The value is specific to a particular analyte and thus it will be different for each peak in your chromatogram, this can make it tricky to say a column has a particular number. Also, it is only useful if you are using isocratic elution methods, it is not meaningful for gradient methods.

Typical indicators that a column has reached the end of its life are:

Resolution – if the peaks that you are interested in are not separated adequately then the column may need to be discarded, you would expect this to build up over time. A sudden loss of resolution may have a different source. A typical value of resolution when the separation is satisfactory is Rs > 2 although a value of around 1.5 is usually considered baseline separated. Therefore a value < 1.5 indicates that the separation is not adequate.

Peak shape – peak tailing (especially on large peaks) is usually associated with the age of the column and at some stage will be too great for reasonable quantitative analysis. A typical value of an acceptable tailing factor is T < 2. Therefore when the tailing factor is above this value the column may be no longer performing at an adequate level. Peak shoulders may also appear (on all peaks) which indicate that there is a build-up on the column inlet that means the column is no longer fit for use.

Theoretical plate count, N – in the region of about 2000 is usually accepted as a limit but be careful not to throw away a column that it still working.

Pressure – The back pressure of the column will usually build up over time and may determine the lifetime of the column."

  Tweet