GC Instrument Qualification

[sakthidasan]

Hi,

Could you please let me know the tests to be performed with its sample and pass/fail criteria for GC instrument performance qualification?

GC instrument Configuration/ Tests required

Inlet : Split and Splitless, PPIP
Detector: FID, TCD and ECD
Injector : ALS and Headspace

If you have any document capturing the qualification details, it would be fine.

Thanks and regards
Sakthi Dasan.S

[sam.pedraglio]

Usually, for each kind of instrument, the best practice is ask a draft of the qualification protocol to the manufacturer. If you have any particular requirments in your work, you could start a discussion with the manufacturer based on his protocol and try to integrate your requirments wit it.

[sakthidasan]

Thanks Sam for your reply.

Could you please let me know the standards that can be used for the qualification of GC as per the tests that needs to be performed?. This will really help me a lot.

Thanks once again

[skunked_once]

Sakthidasan,

You have posted a simple question to which there is not a simple answer.

As Sam has replied, each instrument manufacturer has a set of parameters that are used to verify the performance of their instrument. These parameters depend on the type of inlet and detectors so there is more than one set of parameters. You will need to check your instrument manuals for details on instrument setup or contact the manufacturer of the GC. For example, the performance test for an HP 5890 with split injection and FID uses a 10M x 0.53mm methyl silicone column and a test mixture containing 0.1% 1-octanol, nonanol, n-undecane, naphthalene, 2-decanone, n-dadecane, n-tridecane, 1-pentadecane in tetradecane. Different test mixes are used for different detectors.

If you are trying to qualify a GC for an assay in a regulated test environment, there should be specific tests to document the validity of your instrument. There is no way for us to answer that question because of the many different requirement for each agency.

You will have to obtain details from the instrument manufacturer or from the government agency you will be working with. If you reply with more specific requirements someone may be able to help. Good luck!

[sakthidasan]

Thanks Skunked_Once.

Standards details provided by you may help me better.

To put my question straight, What are the tests that will be done by an Agilent service engineer when he qualifies the instrument.

Hope fully, he may perform for 6890 and 7890 [Please correct me if i'm wrong]


Manual OQ for,

1. Flow Accuracy
2. Temperature Stability
3. Inlet pressure accuracy
4. Inlet pressure decay
5. Oven temperature accuracy

Automated QO for,

1. Signal to noise
2. Noise and Drift
3. Injector Linearity [Only for ALS, Will it be applicable for HS also?]
4. Detector Linearity [Hope this is applicable only for FID and not for TCD and ECD]
5. Carry Over [ Head space and ALS]

Suppose if i have to follow the above tests [if i'm correct] in Agilent 6890 and in 7890, then what are the standards used by an Agilent engineer to Qualify the results [for Automated OQ].

Do they follow standard reports to show the test are pass/fail. If yes, Could you please post some sample reports on that.

Thanks
Sakthi Dasan.S

[Don_Hilton]

Check the Agilent web site. It seems to me that I found a copy of Agilent's OQ procedures two or three years ago. It inlcuded forms, specifications, part numbers for standards and equipment, and any other details you would need.

[sakthidasan]

Thanks Don,

I got a OQ procedure document from Agilent website. But it doesn't have any information about the standards that can be used for each test.

Do you have any idea what samples shall Agilent engineer use to qualify the instrument?

[Don_Hilton]

It was about three years ago. I remember ordering standards that matched the descritions from the Agilent catalog - and then I dropped the particualar project. So, my reccomendation is that you look at the Agilent web site and see if there are matching standards. Which is as far as I could take it, myself, today.

[sakthidasan]

Thanks Don.

Would you like to suggest any standard samples for the tests mentioned?

Also, I couldn't able to make any view from their website what the exact samples Agilent engineer would prefer when they qualify the instruments?

[skunked_once]

You will probably have to contact Agilent technical service directly for the details of the tests you wish to do. Naturally they will want to send out one of their service engineers a major cost but perhaps you will be able to get them to give you the details you require. Good luck

[stepluke]

Hello sakthidasan

You might consider using a kit that allows you to perform the qualification yourself.

http://www.v-kit.com/products/v-kit/ini ... bundle.htm

Cheers

Steve Luke
Technical Manager
v-kit

[aldehyde]

You will probably have to contact Agilent technical service directly for the details of the tests you wish to do. Naturally they will want to send out one of their service engineers a major cost but perhaps you will be able to get them to give you the details you require. Good luck

Well yeah, of course they'll send some one out and it costs money. Its a service provided by Agilent and the software, methods etc are proprietary . You can always do the qualification yourself and deal with the QA/regulatory implications. Keep in mind that there are other benefits that come with getting an official Agilent OQ, for example: the way we run the protocol creates an audit trail for all data, methods, and sequences. All original data/methods/sequences are also stored with the results for future review.

DISCLAIMER: I am an Agilent service provider and the majority of my work is running qualifications, everything I say here is subject to guesstimation and is not The Law. Contact Agilent, get the ACE EQP for your system to get some official information. If you need to make changes to the limits or configuration that can all be discussed with the engineer before you approve the protocol. If I was going to do your system this is what the setup would be:


Channel 1: ALS -- PP -- TCD
Channel 2: HS --- SS -- FID
Channel 3: ALS -- SS -- ECD (depending on how you use the ECD, if you use it with the headspace I'd test it with the headspace attached) *see note at bottom

Tests to be run on the headspace include heated zone accuracy for the oven and sample loop, solenoid valve function test to check for leaks with the vent and pressurization valves. A series of injections would be made to confirm reproducibility of injection area counts and retention times. Another injection is made to check for carryover/contamination.

For the GC:

Inlets are leak tested and the pressure accuracy is also tested. Oven temperature accuracy is checked at a low and high set point, and stability is checked by comparing drift at an isothermal setting. Detector flow accuracy would also be checked for the FID and TCD. I haven't run an OQ on an instrument with an ECD in a little while but I don't think it has a flow test.

Injection precision is checked for the ALS by making another series of inections, signal to noise would be tested for FID/TCD using both the ALS and headspace and then the ECD would also be tested (probably with the ALS but this is up to the customer). A noise and drift test would be performed for all detectors as well.

* Note that a separate charge is included for testing the 3rd detector, linearity is also an additional test which requires an extra charge.

Let me know if you have more questions.