|
Troubleshooting unexpected or unacceptable data is often frustrating because of the variety of possible causes that require assessing or ruling out. The first step in any troubleshooting effort is to identify all possible issues; even the ones that are unlikely as final answers. Once the possibilities are listed, rank them in order from the easiest to reach a decision on, to the most difficult. For instance, GC chromatograms without peaks may be caused by a defective circuit board which reads signals from the detector, or maybe the syringe plunger was bent during the night, preventing sample injection. The considerable effort required to assess a faulty circuit board warrants making sure a bent plunger is not the cause.
This month’s Technical Article describes some helpful methods of quickly assessing certain potential issues. Two specific examples of possible problems are detailed, along with “tricks of the trade” tests that can be applied during troubleshooting experiments. These methods are not manufacturer specific in any case, and the latter may even be applied to several different analysis technologies.
Issue #1: GC/ECD baseline running abnormally high.
Possible cause: A leak due to a loose fitting somewhere in the system.
Troubleshooting method: Turn off the instrument oven and open the door.
A leak caused by a loose fitting increases the GC/ECD baseline because the detector will register response to air entering the system. This situation may cause a baseline that is slightly higher than normal, but in drastic cases the baseline can be several times higher than normal. When the instrument is in operation, the oven fan continually creates a positive pressure which can force air into available leaks. Watch the detector signal output after stopping the oven fan and opening the door. If a leak is the cause of the problem, an immediate, significant drop in signal should be noticed. A variety of leak testing devices are available to further investigate this issue.
Issue #2: Higher than expected, or higher than acceptable, variation in analysis results.
Possible cause: Decreased injection reproducibility.
Troubleshooting method: Weigh the sample.
There are many problems that are possible causes of high variation, making this a sometimes difficult issue to troubleshoot. As well, definitively assessing injection reproducibility can be a daunting task, because the injection action is often difficult to separate from the rest of the analytical system. One simple method is available if an analytical balance is accessible. Weigh a prepared sample, do an injection, and weigh the sample again. The precision of modern analytical balances is sufficient to record weight differences with injections in the microliter range. Collect at least three rounds of these inject and weigh sessions and determine the variation. If the observed variation is significant in comparison to the desired results, consider the injection as a source of the issue. Please note: methods using internal standards are mostly immune to problems caused by injection variation.
These are only two of many troubleshooting techniques applied at APG. APG has a collection of helpful techniques garnered from decades of analyst experience and conversations with other experts in the field of Analytical Science. If your laboratory has troubleshooting issues and would like to see them similarly reviewed in a future Technical Article, please place a call to Technical Support at APG or email support@apgqa.com. |
