Backflash and split/splitless injections and FID air grade

[Roxanne42]

Hi there,

A couple of quick questions...

(1) So if we're injecting samples in splitless mode - let's say 1 uL of a sample in DCM with an inlet T of 250, head pressure of 10 psi in an inlet liner of 493 uL effective volume - this should give a 399 uL vapour cloud which should be fine. Now if one wanted to inject 2 uL instead, the vapour cloud would be 797 uL, which would probably cause backflash.

But let's say you are injecting in split mode, with for example a 50:1 split ratio - does the ratio apply directly now so that a 2 uL injection would be fine and reproducible? I'm a little confused on how to know whether backflash should be an issue with split injections and if so, at which point (injection volume, conditions and split ratio) does it become an issue...?

(2) Technically, all gases we use on our GC-FID are UHP grade. But the last few cylinders of air that were used were actual industrial grade. The thing is we haven't noticed any major problems with using this lower (and much cheaper!) grade... Is it possible that it only raises the baseline? By how much? Anyone have any experience with this?

Thanks,

Roxanne.

[chromatographer1]

The issue of vaporization of DCM and other solvents is dependent upon the rate of vaporization as well as the injector volume. Using a 'burst' mode of injection can cause a very rapid increase in volume, especially if the split is low and the injector temperature is high. Increasing the split and/or decreasing the temperature will decrease the likelihood of backflashing.

On the other hand if a very slow injection technique is used then the vaporization of the solvent may be carried out in a more timely manner and even if the temperature of the injector held low, done effectively without backflashing.

Not all autosamplers can be adjusted to parameters which allow this flexibility.

Air for a FID that is lower grade can give you a higher signal and more noise than when using high purity air. One has to determine the requirements of their instrument and their analyses before deciding upon which grade suits best. Don't spend money needlessly, but don't cheat on consumables if it means that your work is of less or no value.

best wishes,

Rod

[Peter Apps]

Hi Roxanne

The vapour volumes thrown out by various GC softwares etc are, like everything else to do with flash vaporizing inlets (!) a dramatic simplification of extremely complex and unpredictable circumstances. Use them as a guide and not as gospel. My thoughts are that many of the problems laid at the door of vapour backflash are due to the inability of EPCs to control flows and pressures when faced with the pressure pulses that injecting volatile solvents into hot inlets produces. It might just be co-incidence that backflash and vapour volume calculations started to worry people after GCs were routinely equipped with electronic rather than mechanical gas controls, but then again .......

On the issue of air quality - the problem with industrial grade is that its quality will be inconsistent, you could run quite happily for months and then you will get a cylinder with something horrible in it that will invalidate a stack of work and crud up the supply lines. In my student days (long past) I worked in a lab where we used industrial hydrogen as carrier gas !!. Not knowing any better, all was fine until a cylinder that was contaminated with ammonia (but still within industrial spec) wiped out a whole lab full of columns (fortunately home made glass capillaries) in one evening.

Peter

[chromatographer1]

I agree with you Peter.

Today trying to describe the injection performance with an EPC injector is "a dramatic simplification of extremely complex and unpredictable circumstances".

My description is based on using the older technology (not necessarily inferior technology) of maintaining a constant pressure on the inlet.

The difference in using low temperature injection with a sample value and a heated injection with a syringe is very obvious over a period of months and becomes a critical factor in maintenance of the hardware.

Applying this experience to inlets that use syringe sample introduction has allowed me to see vast improvements in analytical results when different parameters are used to introduce liquid samples into heated inlets.

This is very pronounced when injecting volumes of liquids that normally would flood the internal volume of port.

As a side issue I have even seen differences in results when the needle tip type is changed.

As Peter has said so well, it is a complicated matter.

best wishes,

Rod