GC compounds - poor peak shapes and missing peaks

One of the most common questions I am asked by customers is “Why is my peak shape so poor?” Another is “I do not see a peak for my compound, what should I do?”

Sometimes they are developing a new method and have no experience with the compound(s) of interest; other times things worked well in the past and now they don’t. Depending upon the exact circumstances my questions/suggestions may vary somewhat, but the first thing I usually do is to ask a few questions (Q) to help me narrow down what may be the issue.

Q1.  Has this analysis worked well on this specific instrument in the past?  If so, when was the last time?  What has changed from when it worked well until now?  Has the instrument recently been serviced or had extremely “dirty” samples analyzed on it?

Q2. Has this analysis worked well on any instrument in the lab in the past?  If so, when and on which instrument make/model?  Is the current GC similarly equipped to the previous model?

Q3. Is this a new analysis for this lab?  If so, what method is being followed?  Is it a common, well-publicized method like those from EPA, ASTM, USP, NIOSH, etc., or is it an obscure method that was found searching the web?

Q4. How does one know for certain that this particular analysis is going to work well on a specific instrument?  In other words, have you found any data/chromatogram/method showing that this analysis should work well (or is even possible) on your instrument?

Guide to GC Column Selection and Optimizing Separations

Some of the questions listed above may seem somewhat repetitive, but they really are not. Depending upon the answers to the questions above, possible suggestions (S) are listed below.

S1. If a specific instrument has provided good peak shape(s) in the past, we would suspect that it should be able to do so again.  I suggest changing the column, cleaning the injection port, replacing the injection port liner, cleaning and/or replacing the detector and associated consumables, verifying all gas flows, etc.

S2. If this analysis has not been performed successfully on this particular instrument, but it has on other similarly-equipped instruments, I suggest trying to perform this analysis on the instrument where it last worked.  Assuming this instrument is working well, if the same issues occur, there likely is something else going wrong that is not related to a single instrument.  Verify your chemical reference standard is OK.  Verify the instrument gases are clean.  Look at whatever these two instruments have in common and investigate further.

S3. If this analysis has never been performed in your lab but you are following a method, I suggest contacting the author(s) or agency that developed the method.  It’s possible they may have heard of similar issues and be able to quickly provide you a solution.

S4. If you are not certain that your instrument is capable of performing a particular analysis, I suggest searching for chromatograms/methods showing the analysis using a similarly-equipped instrument. You need to make sure your peak of interest isn’t eluting under the solvent peak (if a solvent is involved), co-eluting with other peaks, and that you have allowed enough time for your compound to elute from the column.

In summary, by reviewing the answers to these questions, it usually becomes easier to determine a logical approach to finding a solution. For example, if you know the analysis has worked in the past, you know it should be able to work in the future, and a methodical step-by-step troubleshooting approach will usually provide the solution.  But if you are uncertain if a particular analysis will work using your column/instrument/detector, your time may be better spent researching to make certain you have chosen the proper column & instrument & detector.

If you have any questions, feel free to send me an email.

You may find the related articles helpful.

Thank you for your time.