add a internal standard

As Prof Zhang suggested, the internal standard is the best way forward. Depending on your gas, you migth have an internal standard already present - such as N2 or CO2 in an air sample or some non-reactive impurity in your process gas.

Prof Zhang, can you please give me a more details.

I do not think GC is suitable for CO2 N2 measurement. The internal standard must be a new compound with a intermediate MW among all targeted compounds

You don't give us much to go on. How are you injecting the sample? What type of column are you using, and what are the analytes?

Respected Wright,

We are manually injecting the sample using micro syringe. We are using Restek made Rt-Alumina bond column and FID detector. Our analysts are Methane, Ethane, Propane and Propolyne gas.

You do better to loop up a manual book and some papers related your work

Forget the internal standard, as I understand you have a very basic equipment and the addition of ISs to gas samples is not a trivial task.

In my long experience, standard glass microsyringes, even the so-called gastight type, are not satisfactorily reproducible, which I think is your main issue. I suggest you to give a try to the extra-cheap plastic insuline type syringes (for injection volumes up to 0.5 mL) or intramuscular 2.5-5 mL type (for higher volumes): I know very well that this is not a mainstream procedure, but I can assure that it saved my day many times, and it WORKS!. I also suggest you to alwais inject in split mode. With a little pratice you should achieve excellent injection reproducibility and trustable calibrations.

Respected Chiavarini,

We have SHIMANZU make GCMS-QP2010 Ultra model equipment. and kindly suggest me if any operating parameter plays a main role in reproducibility?

Manual injection can be difficult to master, especially if you are operating the inlet in split mode. You would greatly improve your chances of success by switching to a gas injection valve.

Michael, it's obvious that a dedicated loop injection valve would be a much better solution, but I tried to give Suresh a quick and easy answer to his problem.

Suresh, I don't know your system, but what I learned on mine (Agilent 7890/5975) was: use a 0.32 ID column at 1.5-2 mL/min flow, split no less than 1:10, very fast inject 0.5 mL max, use a split liner with glass wool packing in the upper part (close to the injection port)-this appears to be crucial to obtain perfect peak shapes and great reproducibility.

Of course your mileage may var, depending upon peculiarities of the instruments at your disposal.

Always keep experimenting....

Thank you for all, for your valuable inputs. I will go with your suggestions.