Liquid Injection using Agilent 7683B ALS and 6890N GC

[Jana P]

Hi

I am currently running a Liquid injection method used to separate methanol, toluene and DMSO in an API. The diluent used is Dichloromethane and heptane is used as an internal standard. Samples are prepared by dissolving 1 g of API into 10 mL of dichloromethane.

The problem we are experiencing is that the system is shutting down randomly when sample is injected onto the column and we are also experiencing occasional poor reproducibility (precision). The needle tends to be quite stiff and is normally the cause of the instrument failure.

We are not overly familiar with the instrument and is there anything that can be done to minimise the needle becoming "sticky" or Jamming? Any ideas on how we can improve the precision of the method?
The method when it works is quite good and so we would prefer not to have to change it in a large way.

If you can help it would be appreciated.

Thanks

[GaryR]

Hi Jana

Do you have any post injection solvent washes in your injection program?
Does your shutdown problem require changing the syringe to restart the run? (If you remove the syringe and rinse manually with acetone or DCM does it free up the needle?)
You say that the syringe/needle tends to be quite stiff, and I would suggest that this is caused by the high sample concentration and insufficient rinsing of the syringe post-injection.
Is it only for this analysis that you are having problems with the liquid sampler, or is this a common problem - maybe it requires a service call?

Hope some of this is useful to you

GaryR

[Jana P]

Hi Gary

We are runnning with pre and post injection solvent washes.

Once the system has shut it self down the syringe does need to be replaced and we have used both new and syringes that we had washed and then used again. rinsing/washing the syringes does seem to free the needle up and make them useable once more.

Agilent have not been to helpful in regards to the problem, not for lack of trying, mainly they just don't seem to be able to have any ideas that we haven't tried for improving the method/system etc.

thanks for your help, i will try increasing the post and pre injection rinses.

[Consumer Products Guy]

Jana (had a huge crush on the first Jana I knew, way back in high school) - there is nothing inherently "bad" about your Agilent system. I agree with GaryR that dissolving 1 g of API into 10 mL of dichloromethane is extemely concentrated, and the injector feels too much physical syringe resistance and is shutting down automatically so it doesn't bend your syringe plunger. Since you appear to be assaying for the residual solvents here and not the API, what about dispersing the sample with a solvent which will dissolve those solvents but not the API itself, then filtering? Or, if you absolutely cannot change anything in the method try rinse solvents other than dichloromethane I'm assuming you currently use dichloromethane (DCM) as the rinse solvent, and that the DCM is not evaporating out of your rinse vials (do you use the tiny funnels in them to reduce evaporation?). And, if your sensitivity is adequate, couldn't you just use 25 ml or more DCM and just make everything correspondingly more dilute?

[chromatographer1]

You might try putting another solvent like acetonitrile both as a wash and as a co-dissolution solvent (20%-60%). Then perhaps the needle will be better lubricated and flushed clean. IPA is another candidate. Additional washes post injection are called for here as a minimum to assuage your problem. Be aware of tiny peaks in many lots of acetonitrile which may elute near methanol.

best wishes,

Rod

[Terry]

My 7683B injector got a similar problem 3 days ago. The solvent I used to dissolve the sample, Dimethylsulfoxide, partially freezed, and the injector was jammed. After I turned on the air-conditoner, the DMSO melted down, then the injector got to work smoothly again.