Page 1 of 1

RSD for GC

Posted: Mon May 30, 2011 3:57 pm
by vishnu
can anyone please help me?
I am getting RSD for my samples (mostly methanol or xylene) below 6.
i have to get below 2.
i am using manual injector - Ezchrome for agilent.
is there any tricks in injection or at anything to get the RSD below 2?

Thanks

Re: RSD for GC

Posted: Mon May 30, 2011 5:02 pm
by DSP007
Its determination of residual solvents ( WHO Guidelines , ICH,USP, EP/BP) ?
No panic
1) How many its solvent - if you determine 1 ppm of methanol - you RSD isn't principial 0.94-1.06 << 3000 ppm sign in guidelines ,product is OK
2) You method is headspace wich split ingection? Its tipical RSD. Tou inject probe "handmake wich surgine" if surgine is 10 mkl and probe 1 mkl- its good RSD , "a sure eye and steady hand" as Robin Good, Wilhelm Tiles and Slava Zaicev (http://ru.wikipedia.org/wiki/%D0%97%D0% ... 0%B8%D1%87) "
3)To reduce the RSD is good to use the method of internal standard. But I understand you are not Chief TCI, and yet simple statement?Of course no one method will not change.
4) You inject "check test mix for equipment qualification", as I recomendate in parallel You post? What is result?
5)It could be a hardware problem. Where is the leak. Can be broken gasket or bad tighten the nut.
Here you must decide for yourself - how critical the situation is and whether you are ready to carry out maintenance work on the device.
All the bumps (the problems) will be in your if you will failure without bureocratic note. Everybody loves a winner (с). Just do not judge, but not the fact that the awards

However, your tendency is very commendable, as the Chinese say - " And the way of a thousand li begins with the first step".

Re: RSD for GC

Posted: Tue May 31, 2011 2:04 am
by pincer
DSP007, are you a Chinese?

Re: RSD for GC

Posted: Tue May 31, 2011 2:04 am
by vishnu
Thanks for the response,
No sir, i got the new instrument so there may not be any maintenance problem.
My problem i think is the injection method or Integration method.

Is there any auto integration by software in Ezchrome?
Is there any trick in injection methods to minimize RSD?

Re: RSD for GC

Posted: Tue May 31, 2011 7:03 am
by DSP007
Thanks for the response,
No sir, i got the new instrument so there may not be any maintenance problem.
My problem i think is the injection method or Integration method.

Is there any auto integration by software in Ezchrome?
Is there any trick in injection methods to minimize RSD?
1) The fact that the new tool - does not mean there are no problems.
In addition, persistence of some supplies is negligible. Septa maintains no more than 1000 injections, and recommend it to change even at one-hundredth, not waiting for when it gets dirty or her needle pierce through-hole
2) What is injection method ? Liquid handmake (RSD about 10% to 1 mkl), Liquid autosampler (RSD about 1,5% to 1 mkl) , Automatically headspace (RSD about 2% from unpolar low boiling liquid in water , to 10% from vary polar high boiling liquide in water) , handmake headspace (RSD more 10%).
If you use Automatical static Headspace as 1888
a) One flack - only one injection.
b) Liqud volume in headspace flack should be the same
c) Heat time need more 15 min.
d) T in transfer line and loop should be > boil temperature of probe analising components
Toluene boil in 110C - transfer line need 120 C
I think that all this is done, and here the problem can not be.

3) What problem in Integration ?I don,t job by Ezchrome , but think that all this is done, and here the problem can not be also. Peak of methanol and toluene full separate , and in good chromatography metod - vary simmetry and have small tailing factor (0.9-1.5).
Problem may be only in vary small concentration , around limit of sencivity - the noise peak (jump) summed whith peak . But in subtancies limit of sencivity isnt critical - its 0,1-1 ppm on methanol and about 0.5 ppm on toluene (FID , SP 1:10). This is less than the requirements of WHO is a thousand times, so even a 10% deviation in this case is irrelevant.

(!) May be problem in calculate if you do not take into account the true mass of the samples.
In this case, the error in the hitch is summed with the other errors.
For electrified substances attempt to take a sample of the eye usually leads to very large variant of mass.


The same problem may be, if you do an analysis of packaged products, without weighing it.

A well-tuned machine by pouring into bottles has precision of about 5%, well-tuned machine for pressing pellets/ tablets - about 3%. On the production line (in farmaceutical factory) there may be several, and each mistakes.In addition, the production always acts the human factor - the working-apparatchik may be inexperienced (not to sleep, go to the binge) and the proper setup will not have to say.

Re: RSD for GC

Posted: Tue May 31, 2011 7:08 am
by DSP007
DSP007, are you a Chinese?
Not I'm russian. But is good proverb: if you do not make the first step, you never you reach the end of the road. But taking the first step, you're sure you reach the end.

Vishnu make first step...

Re: RSD for GC

Posted: Tue May 31, 2011 7:46 am
by Peter Apps
can anyone please help me?
I am getting RSD for my samples (mostly methanol or xylene) below 6.
i have to get below 2.
i am using manual injector - Ezchrome for agilent.
is there any tricks in injection or at anything to get the RSD below 2?

Thanks
Hi Vishnu

Please post full details of what you are doing - what sample, what concentration, injection volume, inlet temperature, split or splitless, split ratio, etc etc etc. Then we can help.

Peter