Troubleshooting retention-time problems

We divided this section into two major subsections peak-shape problems and retention-time problems. The last section then covers problems that do not fit into the first two categories, such as pressure problems. This structure should allow you to quickly move to the chapter that covers the problem that you are trying to solve. In the section of peak-shape problems, example chromatograms should also help to point you to the most appropriate section. While this structure is good for rapid troubleshooting, it also makes repetition unavoidable. Therefore, we would like to remind the reader that this part of the book is not designed to be read as a consecutive text. 

Since the response of the detector (and the separation) is a function of a flow rate, it is essential that the standard response curve be determined at the same flow rate as the tablet assay. If retention times differ significantly from the runs of the standards, there is a need to troubleshoot the HPLC to determine where the problem resides. Refer to Chapter 3 for a discussion of retention time precision. 

In the second scenario shown by the data in Table 10.2, poor precision was encountered in both the retention time and peak area of an analyte. Scrutiny of the data showed that both retention time and peak area tended to trend upward. Troubleshooting the autosampler by replacing the sampling syringe yielded no improvements. The poor precision of retention time hinted to possible column problems. The column was therefore replaced and acceptable precision was restored. 

An invaluable tool for troubleshooting and maintenance is the instrument control chart. Anything useful can be control charted, such as detector response factors, deviation from expected retention time, etc., but the analysis results from a check gas is one of the most popular items to chart. The most important information a control chart can show is when the instrument needs maintenance or recalibration. While it may seem like a good idea to recalibrate every day, the act of recalibration has error associated with it, and thus overcalibrating can be as much of a problem as undercalibrating. Statistical tools can... 

---