Chromatography Forum

Hello there, I am new in this forum. So I am learning to use it.

Well I am developing GC analysis for ethanol in a distillery. We are using an Agilent 6850 with FID detector and manual injection. We use an HP-Innowax column 530µm internal diameter.

We want to go as low as possibly detecting the contaminants in ethanol. We are working with: Acetaldehyde, Acetone, Ethyl acetate, methanol, Secbutyl alcohol, propyl alcohol, isobutyl alcohol, 3-pentanol, n-butyl alcohol, n-amyl acetate, 2,4-dimethyl 3 pentanol, isoamyl alcohol, n-amyl alcohol, furfurol, acetic acid, isopropyl alcohol.

We are working with concentrations below 2 ppm. We are using teramyl alcohol as our internal standard (ISTD).

We prepared 4 levels of calibration standards all levels has the same concentration of ISTD. However the chromatograms of the four levels show the ISTD as a growing peak in each level. I mean level 4 is the biggest and level 1 the shorter ISTD. I understand that the ISTD sould show a similar size peak in all levels, but this is not the case.

I have changed liner and septa, have cleaned the FID parts and change de JET, but still thè problem is that the ISTD behaves like stayed above.

I am thinking that the problem is the manifold (the EPC controller) and have requested a change in everything, the EPC module, liner, septa, gold seal and washer, and the collector and jet in the FID.

The same method has been installed in a 6890 in a laboratory next to this one. The only difference is that this is a 6850, and that the 6890 has automatic injection.

Have anyone some idea about that.

Thank you.

How repeatable is this effect - how many series of standards have you made up, and how many times have you injected them ? How repeatable are replicate injections from a single vial ? Is the increase in internal standard peak size (area or height ?) linear vs the concentration of the standard ?

Peter

Hi adalme,
Sorry didn't read question - I guess you are selling ethanol off as by-product.
Some of those contaminants are flavour components and might be of use to someone.
WK

What would appear to be the most obvious answer is that as you increase your sample concentration there is a peak present which coelutes with your internal standard.

Now is it something else, or is it tert-amyl alcohol that is in the ethanol?

You might wish to try a SPB-20 column. See application note 164 from Supelco.

I believe ispropanol elutes just before the acetone peak although it is not documented.

best wishes,

Rod

adalme,
A 624 type column might work well - have a look at some manufacturers application sections
WK

Hi,
Thank you for your answers. The ethanol is a byproduct at a sugar cane facility. The main product is sugar and ethanol is a byproduct.

I don't think that there is a other component, eluting there with my internal standard but I would check for it. The bad thing is that I don't go to the lab everyday and it is out of the city two hours away so is difficult to check it.

I cannot use another column because we have no money for it. And my boss wants to use this column because there is in use in other chromatograph and functions very well.

I have made a series of five standards for each level, but the funny thing is that the three lower levels are linear but the fourth level has a response bellow response of the third level (the levels goes 1 to 4 with 1 less concentration and 4 the higher concentration).

I think that the equipment has a problem.

Regards,



AAM

Hi,
Thank you for your answers. The ethanol is a byproduct at a sugar cane facility. The main product is sugar and ethanol is a byproduct.

I don't think that there is a other component, eluting there with my internal standard but I would check for it. The bad thing is that I don't go to the lab everyday and it is out of the city two hours away so is difficult to check it.

I cannot use another column because we have no money for it. And my boss wants to use this column because there is in use in other chromatograph and functions very well.

I have made a series of five standards for each level, but the funny thing is that the three lower levels are linear but the fourth level has a response bellow response of the third level (the levels goes 1 to 4 with 1 less concentration and 4 the higher concentration).

I think that the equipment has a problem.

Regards,



AAM

We need more information, but here are some suggestions. I would initially focus only on the actual injector setup, syringe and technique, and quality of your reagents.

One obvious issue is carryover, which could be sourced from several places (syringe, injector, etc. ). The other is that one of your standard components has an impurity that elutes with the IS, or perhaps even is the IS.

Did you run the standards with blanks between them, or reverse the order 4 - 1 and 1 - 4?. If the problem still persists ( larger with higher concentration std ), then you need to invesigate your impurity stds. Ensure you inject plenty of blanks to eliminate any carryover issues.

Are the same solutions used on the other GC?. If so, check all the trace peaks and confirm their elution is the same, and that the other GC doesn't have different retention or improved resolution from a different column. Can you run the solutions on the other GC if they are not routinely used there - just to get some some different data on the peaks and carryover.

Check your syringe and injection technique, ensure that you are thoroughly flushing the syringe with good quality ethanol after each injection. Ethanol is fairly viscous and does require a reasonable amount of energy to vaporise, so ensure that you are not overloading the injector, and the injector temperature settings/flows are OK.

The low response for the highest standard could be caused by injection discrimination, so ensure the split/splitless or on-column system is correctly set up.

Please keep having fun,

Bruce Hamilton

Please can you tell us your procedure for making up the standards and for adding the internal standard - with exactly how much of each compound is added to how much ethanol, and how you do your dilutions, what you use to measure and dispense liquids etc.

When you say that you made up five series of standards, did you start from the pure compounds each time, or did you start from stock solutions ? Do you have stock solutions of the internal standard ?.

How do the sizes of the target analyte peaks change with concentration ? Does their size increase linearly with concentration ? How repeatable (in terms of relative standard deviation) are peak areas from both the internal standard and the analytes ?

Peter