GC-MS Volatile Fatty acids

[ziarehmann]

Hi everybody,

In our lab i am trying to measure the concentration of differerent volatile fatty acids such as acetic, propionic, butyric, valeric, iso valeric and caporic acid using GC-MS. I am using a special capillary that is specified for fatty acids. I have prepared a standard solution using the above mentioned acids. The samples were heated and the gas phase is injected in the GC for analysis (Headspace chromatography). The highest concentration is of acetic acid i.e. 10 g/L and for diluted sample lowest conc. 0.5 g/L. Using my temperature program i am getting all the acids seperated and peaks coming at expected retntion time. I have some basic questions.

1. Split/Splitless: How to decide whether i should use split injection or splitless injection. like in case of acetic acid i have 10-0.5 g/L but in case of Caporic acid i have concentration ranging 0.50-0.025 g/L ?

2. As i said i am getting all the peaks, but i have the peak of Butyric acid very high and sharp where as all other peaks are very small and bit wider although the concentration of Butyric acid is less 1.0-0.025 g/L as compared to acetic acid. (Just for explaning the butyric acid peak is like the first peak in forum logo and rest all the acid peaks like the last two peaks in forum logo.) What i can do to increase the peak size and shape of other acids?
* The equilibration temprature is 85°C
* GC SSL Inlet temp is 200°C
* The GC-MS is from Thermo scientific.

Hope to have some nice comments from the members at this forum.

Regards!
Zia

[Peter Apps]

Details, details

What column are you using, what is the carrier gas and flow rate, and what is the temperature programme ?

How are you doing the headspace injections ? - if an automated headspace sampler which one. What volume are you injecting etc etc etc.

Peter

[ziarehmann]

Hi,

I am using FFAP column from SGE (30 x 0.25 x 0.25)[as prescribed in 2-3 research articles]. Carrier gas is Helium and flow rate is 1.2 ml/min. Initially i used a split ratio of 17. but for diluted samples the results are not good, peaks are not sharp and base line is above 0.
Temperature Program:
1. 80°C-> 1min
2. 10°C/min-> 120°C -> 0 min
3. 15 °C/min-> 180°C -> 1 min

I am doing manual injections no auto sampler. I have started with 5uL but now I am injecting 2uL. I have also option of injection 1uL.

Zia

[MSCHemist]

The split ratio is really a function of if your peak size is sufficient for sensitivity at your lowest level. If not decrease the split or go splitless .

[Peter Apps]

Your first step needs to be to increase your injection volume substantially. The 2 ul you are injecting is appropriate for a liquid injection, not headspace. Try 200 ul.

Also, decrease your initial column temperature to 50 C to sharpen the peaks.

Peter

[ziarehmann]

Hi,

Thanks for your kind reply. The higher i go with volume, the chromatogram gets worse. I have lots of peaks and the peaks are not well seperated. In case of splitless injection i get the later peaks but the initial peaks of acetic acid and propionic acid disappears completely.
Any other ideas.
Zia

[Peter Apps]

Hi Zia

If your samples are equilibriated at 85C, then your syringe has to be at least as hot as the samples. If it is cooler than the sample all that happens is that the water and acids condense inside the syringe. 85C is too hot to handle, and yet you say that you are doing manual injections. How are you doing it ?

How quickly do you inject when you try 200 ul ?

Peter

[ziarehmann]

Hi,

to make things more clear i have uploaded a pic of two chromatograms, one for the highest concentration and the other for lowest concentration. these are the best possible results that i am getting till now. These results are achieved with a column flow of 1.5 ml/min and a splitratio of 7. The temperature program is same as above. And injection volume 2uL.
@Peter: Yes i have thought about that aspect as well. Normally just before the injection i heat up my needle in an oven at 85°C and then take out the samples. Its not the ideal but but at the moment i have no other option.
In case i go for splitless injection i only have the last 4 peaks. The first 3 disappear completely.

Hope to get some new ideas from you ppl. Thanks a lot for the help.

Zia

http://de.tinypic.com/view.php?pic=2hpk ... 1ZSNleDFdV

[Peter Apps]

Hi Zia

After looking at the two chromatograms (which are too small to examine in detail) I am wondering what improvement you expect. You have reasonably shaped peaks that are just about separated for a set of analytes that are difficult to get sharp peaks for, using a sub-optimal method. Actually I am surprised that the peaks are as big as they since you are injecting only about one hundredth of the volume that would be sensible for a headspace sample. What specifically is it that you are not happy with ?

What kind of liner do you have in the inlet, and does it have anyting in it like glass wool ?

Peter

[ziarehmann]

Hi,

The thing which really disturbs me is the area under the peaks. My thinking is the higher concentration i have the bigger would be the area and height of peak. but in my case its not so. Secondly for a same concentration i have different areas. What could be the reason for that?

From 1 vial how many headspace samples can i take and what would be its effect? I means the difference in peak area and height for first sample and 10th sample from same vial?

Peter: i dont have an issue with the peaks but more with the area. What can i do to improve the reproducability with manual injection.

Hope to get some new ideas. Thanks a lot for the help.

Zia

[CE Instruments]

Manual headspace with a syringe is a PITA, get an autosampler or look for a method to extract. I too am surprised you get much of a chromatogram injecting 1ul Headspace typically uses 100x (or more) your injection volume. What syringe are you using , it is we assume a gas tight syringe ?
Using a syringe based injection you cannot (usually) carry out splitless injection with headspace due to the large injection volume. For reproducabilty I would aim to use a 2.5ml gas tight syringe , inject at least 100ul of gas at a split ratio of at least 10:1 (EPC GCs don't like low split ratios). You will be extremely lucky to get RSDs on replicates any where near as good as 5%.
Sampling multiple times from a single vial will result in a reduction in sample amount each injection. Headspace works by creating an equilibrium and if you extract some sample the equilibrium will shift. This also assumes that the hole you punched in the vial septa seals at the increased pressure of the next incubation. New vial each injection

[Peter Apps]

Hi,

The thing which really disturbs me is the area under the peaks. My thinking is the higher concentration i have the bigger would be the area and height of peak. but in my case its not so. Secondly for a same concentration i have different areas. What could be the reason for that? The size of the peak depends on the concentration of the acid in the headspace, not in the liquid phase. Weaker, longer-chain acids are less hydrophillic, so tend to partition more to the headspace, but are also less voaltile, so tend to partition less. The presence of the relatively strong acetic acid will tend to make the weaker, heavier acids partition more to the headspace. So: do not expect that the peak areas for the different acids will reflect their concentrations in the liquid phase. Unless you lower the pH with a strong acid like hydrochloric the interactions among your analytle acids are very likely to cause any calibrations of peak are vs concentration to be non-linear. If the "different areas" are for the same peak from replicate injections at the same concentration, then this simply reflects the poor repeatability of trying to do manual headspace with a syringe that is colder than the sample

From 1 vial how many headspace samples can i take and what would be its effect? I means the difference in peak area and height for first sample and 10th sample from same vial? If you are only removing 2 ul at a time you could take dozens of samples from one vial (presuming that you have a few mls of headspace).

You will never get repeatable results if the syringe is colder than the sample. You have to find a way of heating the syringe that still allows you to use it, or you have to reduce the temperature of the sample to say 50C, heat the syringe to 55 C which should be possible to handle, and work very quickly before it cools down. With the sample at 50C the peaks will be smaller, so you will have to increase the injection volume. You must understand that manual headspace is only repeatable if the sample and syringe are both at room temperature .

Have you tried reducing the column initial temperature to sharpen the peaks ?

You have not answered my question about the inlet liner

Hope to get some new ideas there is not much that you can do if you are limited to manual headspace - what makes you want to use this technique ?. Thanks a lot for the help.

Zia

[ziarehmann]

Hi,

Now i am really confuse. Yes i am using a Hamilton gastight syringe. About the sample prepration:
I have made a standard solution using the above mentioned acids. Then i put 5 ml of the solution in 10ml vials. Then i put all these vials in a waterbath with a temperature of 85°C (This is the temperature in the headspace of the vial). when the required temperature is achieved i wait for 15 minutes and after that take out the sample using syringe. With no autosampler this is the simplest way in my opinion.
I have normally heared/read a sample volume of 0.5-10 uL is sufficient for GC-MS analysis. I have tried using higher volumes but the chromatogram becomes worse.

Zia

[ziarehmann]

Hi,

I tried to reduce the initial temperature but that also did not result in good peaks. Personally speaking i dont know about the liner, i will look into that and let you know.
I only have the option of manual injection. The autosamplers are very expensive and my prof. has at the moment no budget for that so my only option is manual injection.About the acidification, i have read about it and also the addition of salt but that is in case we have pH higher than 2 but for standard mixture my pH is already 2.

Zia

[MSCHemist]

Headspace is generally sampled in the 100ul-1ml range.

1-2ul of liquid is typically injected because it expands into a few hundred ul of gas in the inlet and the typical liner volume is ~900ul. If you exceed that you make a mess of your inlet.

Headspace is already a gas and the analyte is often much more dilute than in a liquid so a few hundred ul is typically injected. Using a pressure pulse or high split helps empty the liner faster so everything gets to the head of the column at the same time resulting in sharper peaks.