HPLC Anomally???

[Jetgel]

An hplc method was ran on a waters 717 plus system. Mobile phase: Methanol + H2O. Three samples with the same list number but different lot number were tested - samples were injected in duplicate and bracketed by standards.

Results:
1st sample - 0%
2nd Sample - 100%
3rd sample - 200%

How can this anomaly be explained if the sample prep/mobile phase was correctly prepared? What if any problems can occur with the instrument that would produce such results?

Any theories would be greatly appreciated.

[tom jupille]

How much did the duplicates vary?

Assuming relatively little difference between duplicates, the most likely explanations would seem to be:
- the lots are in fact different
- the sample preps were not, in fact, identical
- peaks were misidentified

[Gerhard Kratz]

Or the column was not equilibrated

[carls]

Did the first sample (0% result) have a peak at the correct retention time? If so, how does the peak area compare to that of the standard? How do the peak areas of the different samples compare to each other? How do the peak areas of all standards compare to one another, i.e. %RSD of peak areas of all standards? You'll also need to look carefully at the integration of all samples and standards. The baseline should be drawn similarly for samples and standards. Of course the retention time of the analyte peak should vary by less than 2-3% across the analysis

[Jetgel]

I know this sounds crazy but the sample preps were indeed correct, column was equilibrated, %RSD less than 1 %, and retention time of the sample and standard were almost identical.

We were thinking that it could be a problem with the syringe or rotary valves?

[Peter Apps]

I am intrigued by the complete absence of units in your result - even if % was an acceptable unit (which it isn't) it has to be a percentage of something.

And how do you know that the sample prep was correct ?, obviously something is wrong, why not the sample prep ? At the moment (and in the complete absence of anything about what you are analysing and how) it looks like the most plausible explanation.

Peter

[Jetgel]

A remeasurement was done on the same exact vials and of the contents in the flasks of the original prep and all results yielded at or around 100%.