Please, can somebody help me with this problem.
During validation of a reverse phase isocratic HPLC method for quantifying impurities in a tablet formulation a set of response factors for the various known impurities was determined.
Later when the method was transferred the response factors could not be reproduced even though a identical make of column and HPLC equipment was used. A third attempt to determine the response factors gave yet a different result. Is this a common problem and if so, is there a correct way to deal with the problem?
Thanks in advance.

Dear anonymus;

Your problem is a common one, and it basically depends on how different are the response factors from system to system. If you used the same standard compounds, then most likely you are seeing a difference in responses due to wavelength changes, in other words, not all detectors have the same vawelength accuracy.

Changes in response due to differences in wavelength can be very drastic, in some cases they could change by an order of magnitude. Obviously, this problem is very prominent if you are using a wavelength set on the tail of the absorption bands of your compounds.

There are a couple of ways to solve this problem, you can first test the wavlength accuracy of the detectors employed and compensate for the differences found. Another way, would be to introduce a suitability test solution containing your compounds, and running it at different wavelength settings until you obtain the desired responses.

Better yet is just to specify in your method that response factors should be determined for each system and employ in the calculations.

I hope this information helps you.

Good Luck;

Benjamin

Anon,

It is possible if you have the same make of equipment. You will have to control all of the detector parameters and make sure all of the data collection is done using the same slit width, band width, response time, collection rate and reference wavelength. Also, procedures would need to be inplace to recalculate the response factors in the event of a wavelength recalibration.

Basically using response factors is tough because the spec. for wavelength accuracy is usually someting like +/- 2nm and you need accuracy more like 0.2 nm to have stable response factors. When I write transfer protocols for methods using response factors I always leave open the escape route of recalculating the response factors in the lab we are transfering to. good luck.

How different are your responses? Do they have any significant impact on your final results e.g. individual related substances? Is the difference has any significant impact on the method variability (precision)?

I agree with Benjamin's comments. However, checking response for each system is a bit too extreme but may be the only choice.