Chromatography Forum

I am comparing the relative amount of FAME's and I have noticed that if I change the injected volume the integrated areas of the peaks don’t change accordingly.
Ex: I injected 1 ul and the 1.5 ul of sample and the integration value in the 1.5 ul injection was lower than when i injected 1 ul...

Is this normal?


I am using a HP -5890
HP-5 column 30 m ; 0.25
Split 1:20
Injector at 220ºC



All help is welcome

Hi Jabber

No, it is not normal.

How many times did you do the "experiment" of comparing 1.5 and 1 ul injections ? Are the results repeatable ? Are the results the same for all the peaks on the chromatogram ?

Is this with an autoinjector or a manual injection ?

What is the solvent ?

What is you lowest and highest FAME molecular weight / chain length / boiling point ?

Do you have a packing in the inlet liner ?

When was the last time you cleaned the inlet ?

Peter

Hi Jabber

No, it is not normal.

How many times did you do the "experiment" of comparing 1.5 and 1 ul injections ? Are the results repeatable ? Are the results the same for all the peaks on the chromatogram ?

Is this with an autoinjector or a manual injection ?

What is the solvent ?

What is you lowest and highest FAME molecular weight / chain length / boiling point ?

Do you have a packing in the inlet liner ?

When was the last time you cleaned the inlet ?

Peter

This effect aparently affects al peaks.
I am doing manual injection with a 10 ul hamilton.
The solvent is hexane.
The lowest compound is a C12 and the highest a C18
No packing in the liner and the liner has been replaced recently.

Ex: Im now doing a 1 ul injection and in the second repetiton the integration values were lower than the first injection. (this started to happen recently, I didnt had this problems before)

Hi Jabber

By far the most likely cause of the problem is manual injection technique. There have been quite a few posts recently about injection technique, check the forum for the various views on which is the "best" way of doing it.

By how much (relative %) do the peak areas differ between duplicates ?

If you do a series of five injections, what is the relative standard deviation of the peak areas ?

Do the relative areas of the peaks stay the same while their areas change ?

Peter

Hi Jabber

By far the most likely cause of the problem is manual injection technique. There have been quite a few posts recently about injection technique, check the forum for the various views on which is the "best" way of doing it.

By how much (relative %) do the peak areas differ between duplicates ?

If you do a series of five injections, what is the relative standard deviation of the peak areas ?

Do the relative areas of the peaks stay the same while their areas change ?

Peter

No the Relative areas change.
EX: 1ºinj
37.7 15.6 9.1 21.9 8.8 6.8
2 injº
31.5 16.1 8.9 20.4 14.2 8.8
the compounds are:
C12:0 C14:0 C16:1 C16:0 C18:1 C18:0

I have previosly said that it afected the all the compouns equally, but a more carefull analisys shows that the behaviour is not same for all compounds. As the relative amounts show the C18 and C18:1 don't seem to change the same way.

As for manual injection I do as follows:

Draw 1ul of air then 1 ul of sample and I pull the sample out of the needle so theres no evaporation in the needle. The sample is quickly injected with air on both sides..

From what I have read this is a good injection technique.

If it didn't happen before, I'd be less inclined to blame manual injection technque, and I'd look closly at the syringe first, espeially for a partially blocked needle ( liquid doesn't squirt straight out ), or a leak back up the plunger or at the needle seal ( if removable needle type ) .

Then I'd do an experiment, I'd inject replicates of 0.2, 0.5. 1.0, 1.5, 2.0 ul. I would be looking for trends. Your data is confusing, there is no trend, which suggests you may have a spurious factor, like leaking septum, or injector flows or temperature not consistent. How does the total area compare with previous times when it was working OK?. If time is important, select C12:0, C18:0 and increase oven temperature for a short 5 - 10 min isothermal analysis. Plot total area against injected amount.

Watch the total area ( excluding solvent ), that tells you how much has been injected, and make sure your instrument has not changed eg the split mode is on. I assume retention times are very constant, if not investigate - leak most likely cause. The fact that there is no trend suggests that a leak ( internal or external - septum, O-ring, bottom seal, syringe, column joint ) may be an issue. Check integration, make sure peak heights and peak areas correlate with earlier data.

I'd also consider using a less volatile solvent for FAMEs, iso-octane or n-Heptane worked OK for me, but if it worked OK in the past with hexane,
I wouldn't change solvent now, except perhaps to flush the syringe with if you still have problems. Make sure you syringe cleaning is OK by injecting solvent and flat-lining.

Please keep having fun,

Bruce Hamilton

Hi Jabber

Having seen the peak areas I agree with Bruce that this particular problem is not due to injection technique.

The only thing that I can think of that would give such eratic changes in different peaks is poor integration - can you post copies of the chromatograms after they have been integrated.

Thanks Peter

Hi Jabber

What is the type of your liner? Maybe your problem is related with the liner. If you are using splitless liner, change your liner with the split one. You can look it to the manual or catalogue.

thanks

Savas
SEM Lab. Co.
Ankara/Turkey