Page 1 of 1

micro extraction

Posted: Fri Dec 09, 2011 5:26 pm
by ym3142
As you know typical GC injects 1 uL. If I extract sample into 1 mL of solvent and make injection out of this 1 mL I lose 1000 times sensitivity.

with the attempts of improve the extaction efficiency I am wondering: 1) what is the minimum volume needed in a GC vial/insert so the GC autosampler can make injections at least once? 50uL, 10 or 5uL?

2) If I have extract solution containing Trichloroanisole and blow it to concentrate into smaller volume, such as from 1mL to 50 uL do I need to worry about loss of the trichloroanisole(the analyte)?

3) another question in 2) is if there will be too much residual analyte sticking on the container wall, instead of dissolved in the solution? I am thinking 1 to 0.1 ug/mL concentration range.

Re: micro extraction

Posted: Fri Dec 09, 2011 6:30 pm
by aldehyde
http://www.jaytee.com/vials_caps/hplc_g ... insert.php

They make vial inserts that let you have a minimal amount of sample in the vial. Looks like 100-300 uL although you may be able to get by with the insert not being completely full. Can't fully answer your second question but the boiling point is 140 C so I would think that you could concentrate the solution without losing analyte.

Re: micro extraction

Posted: Mon Dec 12, 2011 12:23 am
by chromatographer1
It will take GREAT technique at room temperature to perform your concentration of the solution.

You should demonstrate your ability to do this before you spend time testing samples.

I would also recommend an internal std as a check against human error.

This is not a trivial exercise you are attempting! Good luck,

Rod

Re: micro extraction

Posted: Mon Dec 12, 2011 7:03 am
by Peter Apps
Assuming that you can reliably do the evaporation without losing the analyte you will probably find that your extract turns into an oily droplet of other components from the sample (my guess is that you are analysing wine or corks). There have been some good looking trichloroanisole methods by SPME published - try a literature search.

Peter