I am using a HP 7694 Headspace Sampler for residual solvents on a capillary column. The procedure I am using was created on a similar system but had a seperate HSS carrier flow. A pulsed splitless injection was used. The system I am using has the column carrier gas flow inline with the HSS. When I try to use a splitless injection the peaks are split and ugly. But when I use a split injection, even just 0.1/1 split ratio, I get great peak shape. The peaks are not that big, to over load the column. Why does the splitless injection cause my peaks to split?
By travis luncan on Sunday, September 9, 2001 - 07:44 pm:
Part of the problem could be with the dead space in the sample loop on the head space. This volume can be anywhere from .5ml to 5ml. Low GC flow and no split means the carrier gas pressure is going to drop off when it is in line with the the HSS.
By Beppe on Monday, September 10, 2001 - 08:27 am:
I have no experience of HS, but remember splitless injection must follow specific rules I tried to sum up in a previous discussion :
Splitless mode is very powerfull, but is a little bit tricky to use; do not forget there are two possible modes : "cold trapping" or "solvent effect".
For cold trapping, the analytes ebulition point should be at least 150°C greater than the solvent one. The injector temp should be higher than the highest analyte ebulition point and the oven temp during injection and purge activation time should be somewhere between solvent and analytes EP, so that analytes are trapped in the column; elution will really start with oven gradient.
If you don't have these 150°C between solvent and analytes, use sovent effect : you have to set your oven temp ~10°C below your solvent EP so that part of the solvent condense in the column, making a liquid trap for the analytes; here again oven gradient will start the elution (it may be useful to have a short steep ramp to volatilize solvent before the "true" gradient ramp).
I do not know if it is possible to combine these rules with the HS mode requirements.
By Anonymous on Monday, September 10, 2001 - 09:22 am:
You need a certain velocity to sweep the loop and get the analytes to the column in a tight band. It is a little strange that a very low split ratio works much better than splitless. Splitless usually doesn't work well for headspace because the valve switching and pressure changes associated with putting the sample loop in-line usually cause drastic oscillations in pressure and flow, leading to the poor chromatography you describe. Usually headspace is performed at a split ratio of 10:1 or above to get rapid loop sweeping and good peak shape.
Try different split ratios to find the optimum conditions for analyte transer to the column.
Posting is currently disabled in this topic. Contact your discussion moderator for more information.