Glass or quartz liners in GC

wool is often used to increase the injectors heat transfer capacity. The evaporation of higher boiling compounds also requires more energy and in those cases quartz wool is often used. It also collects non-volatile heavy molecular weight residues which might harm the column. Besides these benefits glass or quartz wool also has its disadvantages. The wool can become adsorptive especially if some fibers are broken or when it has become dirty. It should be exchanged at a regular basis to prevent chromatographic problems. Avoid the use of glass wool when it is not advantageous.

The use of quartz wool in a liner has been a subject of much debate and there are many advantages to using wool. The quartz wool acts as a crude filter for the analytical column and minimizes the chances of any particulate or

non-volatile material from reaching the column. If the quartz wool is positioned so that the needle tip is in the centre of the wool mass, the wools large surface area helps in the efficient vaporization of the sample. The wool also promotes mixing in the liner. The needle tip is wiped on the wool as the needle is withdrawn, enabling the entire sample to be vaporized. To reduce sample degradation the wool in the liner should be deactivated. The quartz wool used in SGE liners is fully deactivated after it has been placed in the liner. This eliminates active surfaces being exposed during the process of inserting the wool into the liner, which inevitably involves breaking the strands of the wool. The use of wool is not recommended when analyzing low level pesticides such as DDT or Endrin.

One option is to deactivate the liner after some glass wool has been added by a silanization procedure. We have used this approach for the determination of volatile mercaptans that are very critical compounds as considers adsorption on active sites. The liner (together with the glass wool) is acivated by soaking it in 6N HCl followed by a washing procedure to emove excessive acid. After that, it is soaked in a solution of a silylating agent. Using this approach, any active sites of the glass wool created by rupture of fibers are deactivated.

What do You exactly mean with bad reproducibility ?

One compound or all compounds? Area or peak shape?

The remarks of Michael and Robin are quiet complete but bad area reproducibility can have different causes. Without any idea of your experiment the first thought is : if all the compounds suffer try an internal standard .

we are already using an internal standard and doing a manual injection. However the reproducibility in the area of consecutive injection is poor. The deviation is high as 10%. An autosampler will improve the precision. Nevertheless with manual injection, the deviation should not be so high. The analyte is ethanol and internal std. is butanol.

Some more questions:

-Does the IS normalization decrease or increase the RSD

- What is your solvent, how much do you inject , at witch temperature and what is the liners internal volume

-Witch column and initial temperature ?

The RSD will be as high as 30% if we do not use IS normalization. Solvent is water. Injection vol 1 microlit. Initial oven tempt 50C, for 2 min. 20C /min to 200C. Injector 300C. Column BP 624 (0.53 mm id) Nitrogen carrier, flow 3 ml/min. I hope I have answered all the qquestions....

-If Your column is 30m then the pressure will be about 15kPa. under these conditions the volume of the evaporated water sample is~ 2.3 ml. Liner and split have to be chosen to deal with this expansion. A lower temperature will reduce the volume.

-The ethanol will be vaporized just before the water explosion happens,I fear that the amount to the split is suffering from this.

-At 50°C oven temperature a water film will remain in your column, changing temporarily the stationary phase, the effects may stay unnoticed on an FID but are clearly visible with MS. Starting closer to 100°C solves this.

-For this kind of job Static Headspace is (in my opinion) the best sampling technique . With a vial temperature of 80°C the amount of water on your system is a lot lower. We use the 'pressure balance' approach on a wax column and for ethanol in water the RSD is < 2% without internal standard

Thanks Kris. Shall try the oven tempt close to 100C. Shall also try doing the headspace injection.

Kris, i did try using a static headspace with the vial temperature of 85 degree Celcius. However the peak areas of ethanol and butanol are still varying although the ratio is almost constant. But I am still not satisfied with the method.......

Priti, If your ratio's are constant you can always use the internal standard approach.

There are several way's of implementing HS injection : sample loop, syringe, pressure balance. Each of them has different figures regarding maintenance, repeatability and carry over. Specially the syringe approach can be very troublesome, As a consequence the use of Internal standard is common practice.

You can always provide me with a full description (a chromatogram, RSD, equipment, system, pressures, temperatures, split, transferline if used,...) then I can see if there is room for improvement. Feel free to use email (see profile)if you do not want to disclose your work at this moment .

Dear Kris, thanks a lot for your comments. We have always been using internal std.for analysis of ethanol ie for liquid injection as well. At that time also the ratio was constant but i don't understand how can the peak area vary so much for a liquid injection....

I fear that you where overloading your injection liner (see previous reply), water is a tricky solvent to inject. There is a handy free software tool from Agilent to calculate how much you can inject. It allows you also to calculate flows and pressure.

http://www.chem.agilent.com/Library/usercontributedsoftware/Public/Flow_Calculator_00010924.exe