March 28, 2024, 04:57:34 PM
Forum Rules: Read This Before Posting


Topic: GC-MS Manual Injection: Internal standard a good idea?  (Read 2012 times)

0 Members and 1 Guest are viewing this topic.

Offline Omega Glory

  • Regular Member
  • ***
  • Posts: 21
  • Mole Snacks: +0/-0
GC-MS Manual Injection: Internal standard a good idea?
« on: October 01, 2016, 05:11:42 AM »
A man came to me the other day with a problem. He needs to run a number of samples by GC-MS. The samples are highly complex, but due to certain features of his experimental design, he can only run them by manual injection. No autosampler.

I suggested he include an internal standard in his samples as a way to ensure reliable and reproducible results, as he's interested in quantitation. In rebuttal, he asked me if such was really necessary when he could just as easily attempt to normalize his data in post by dividing by the sum of all peaks or by the square root of the sum of squares.

What would be the pitfalls in doing it his way? In skirting an internal standard? Curious to know what the consensus is on this, when an internal standard is recommended and when it's permissible to do a quick and dirty normalization after the fact.

Offline Arkcon

  • Retired Staff
  • Sr. Member
  • *
  • Posts: 7367
  • Mole Snacks: +533/-147
Re: GC-MS Manual Injection: Internal standard a good idea?
« Reply #1 on: October 01, 2016, 07:21:25 AM »
Here's a good resource for internal standards:

http://www.chromatographyonline.com/when-should-internal-standard-be-used-0?id=&pageID=1&sk=&date=

We've also had an incredible diversity of information on our own forum:  just search for internal standard.

There is much to talk about, but lets try to keep the discussion simple for starters.  I use internal standard when ...
1). I'm required to because the method was written and validated that way.  This adds tedium, makes more opportunities for mistakes and, as the reference above states may not even help.
2). When I was new to science, and my pipetting was in doubt, we used an internal standard to normalize my standards results.  Again, I was new, and it was necessary, but it shouldn't have been, or eventually didn't need to be, or ... etc.
3). You suspect instrument variability.  Which should never be the case, says the instrument manufacturer.  In your case, the manual injection is a combination of this sort of concern and the previous concern.
4).  There's serious variability in the separation method with regard to samples and their matrix.  And in an ideal world, that's always fixable.

Its hard therefore to recommend conclusively that an internal standard is needed.  And yet it often is.  You'll have to apply my caveats above, and others you can find here and elsewhere, point by point, the this situation as specifically as possible.  And then you can argue from strength.
Hey, I'm not judging.  I just like to shoot straight.  I'm a man of science.

Sponsored Links