Liner selection help

[cjwitt]

I am running an isothermal method for residual solvents (methanol,acetone, IPA and toluene) on a ZB-624 and need some help selecting the best liner for the job. I am using a 1ul liquid injection with a low split flow on an Agilent 6890. Can anyone help me with this?

[krickos]

Hi

Could you please specify your problem and injection settings?

We routinly do in-process controls with regard to solvents and usually the standard liner works just fine. But then again the split ratio and inlet temp is higher than the default (split ratio 1:5 and 140°C?)ones for residual solvent analysis with headspace injection as described in the pharmacopeias.

What is the sample solvent ? (DMF, DMSO, water other?)

[cjwitt]

My split is 1:10 and inlet temp is 150. When I run the USP Headspace method I use a single gooseneck liner with glass wool and it seems to work just fine. The difference with this particular manufacturers method I am trying to run now is that it is not headspace. I am trying to find out if a 2mm straight liner would be ok for the volume of gas expansion upon injection. My solvent is water. The stock toluene standard is in DMSO and the working is in water.

[krickos]

Hi

Water you say, then you need to be careful. At that temperature 1µl gives close to 1ml of vapour. Which should exceed your liners volume and cause troubles.

If you would inject 0,5µl water on a standard liner (volume about 1ml) then it should work.

As for the headspace injection, you can leave out the glass wool it may work for your application but generally it is only a source for troubles (peak broadning, active sites etc)

[cjwitt]

Kriskos- Thank you for your suggestions.

I am pretty much bound to the 1ul injection volume as I am running a manufacturers method for an API raw material to verify their CofA results. GMP and FDA guidelines (along with my labs general SOP) stipulate that all testing parameters must remain the same as those used by the manufacturer themselves when they tested the product. I will try the single gooseneck liner without the glass wool and see if I like the results better.

Thanks for the insight on the troubles of using the glass wool with residual solvent analysis. That was helpful.

[cjwitt]

Sorry about the typo on my last post on your name Krickos

[Consumer Products Guy]

I am pretty much bound to the 1ul injection volume as I am running a manufacturers method for an API raw material to verify their CofA results. GMP and FDA guidelines (along with my labs general SOP) stipulate that all testing parameters must remain the same as those used by the manufacturer themselves when they tested the product.

"Keep runnin' the ball up the middle, no matter how unsuccessful it is"

Even <USP 621> allows flexibility in the chromatography. Amazing how many will keep doing mediocre chromatography rather than fix it !!!!

[Consumer Products Guy]

I am pretty much bound to the 1ul injection volume as I am running a manufacturers method for an API raw material to verify their CofA results. GMP and FDA guidelines (along with my labs general SOP) stipulate that all testing parameters must remain the same as those used by the manufacturer themselves when they tested the product.

"Keep runnin' the ball up the middle, no matter how unsuccessful it is"

Even <USP 621> allows flexibility in the chromatography. That's why they're called "guidelines". Amazing how many will keep doing mediocre chromatography rather than fix it !!!!

[Peter Apps]

[quote="cjwitt"]
I am pretty much bound to the 1ul injection volume as I am running a manufacturers method for an API raw material to verify their CofA results. GMP and FDA guidelines (along with my labs general SOP) stipulate that all testing parameters must remain the same as those used by the manufacturer themselves when they tested the product. quote]

So, if the manufacturer's results are wrong becuase they used the wrong method parameters, you copy the method and verify the same wrong results

I am SO happy that I don't work in a regulated lab.

Peter

[cjwitt]

If I wanted barbs I would have asked for them. Of course USP allows variation. I am perfectly aware of that myself. Here is an explaination as to why I cannot alter parameters....

1. The testing for residual solvents is not in the USP monograph for this particular API, therefore not required.
2. The testing and a spec for these residual solvents is on the manufacturer's CofA, however, and must be verified and must be done so using their method.
3. Since the validation protocol for their method is unknown, parameters cannot be altered

Know the story and know GMP before dogging on what is being done.

An Yes, sometimes it sucks being in a regulated lab when you know there is a better and more accurate way to do things. Such is life.

[krickos]

Hi

Right so we are talking about a starting material then. This may nevertheless open up for some changes.

First of all find out what is actually filed for the API process. A typical method and specification part of the starting materials are very limited ie critical limits and an extremely breif method section/describtion in best case with wordings like "another suiteble column, different instrument settings giving corresponding selektivity, sensitivity..." (does not mean that methods are not validated, just to keep open for changes)

Secondly, point out that this method may contaminate your instrument gas lines and increase downtime, tends to get management attention

PS
You can edit posts in case you mispell something
DS

[cjwitt]

Thanks Krickos. I was told that the best liner to use for residual solvent analysis was a plain-old single taper liner without the glass wool. I wasn't having the best of luck with the gooseneck liner because it seemed to be inefficient at getting the sample onto my column. I am going to try the liner that was recommended and see what I get.

[Ron]

I would consider a straight liner with glass wool. This will trap nonvolatile material from the API and reduce inlet maintenance. I would not try a 2 mm liner, the volume is too small for a 1 uL injection of water.

[cjwitt]

Hi ron,
Its odd that I contacted 2 very large companies that make consumables and was told to stay away from glass wool for residual solvents. They said it wasn't efficient for trace. They both also recommended 2mm even with water. I agree with you on the gas expansion, but got the best chromatography and response with that liner. Goes to show that its a crap shoot and everyone has something different to offer with gc.

Thanks for the help and input.

[Peter Apps]

Hi cjwitt

I am aiming my barbs at the whole concept of ineffectual methods being repeated in different labs just because that is the way that they have already been done - clearly you are keenly aware of just how pointless such an excercise is. What puzzles me is that on the one hand you are asking what liner you should use, and on the other you tell us that you cannot change anything because you have to use the manufacturer's method.

If you are running split, as you are, you do not need to worry about liner volume - the flow through the split takes care of the expansion volume with a cost in a pulse of flow and pressure that may or may not affect your results. If you can use a megabore column without violating the regulations the higher volume flow rates will help with this even further.

Glass wool has the positive effect of enhancing solvent evaporation (and making it more consistent from run to run) which helps with split injections, but the drawback of adsorptive activity which degrades performance for traces of active components - in your case methanol and IPA. Water can displace active components, so the advantages of wool might outweigh the disadvantages. There are a host of liners with baffles, bulges, sinters and whatnot that are designed to enhance evaporation and gas mixing, I have had good results with one form Varian that has a sintered disc about half way up, I don't know if you can get these for Agilents, or whether you would be allowed to use one if the manufacturer did not.

Peter