I am trying to develop a headspace residual solvent method and been having great difficulty. The solvents are IPA, ACN, and DME. When injecting the linearity for the validation the results come out satisfactory. The sample is injected and the known amount of the organics are determined. The sample is then spiked with the standard linearity solution. Instead of having an increase in organic they all actually decrease. The linerity data of the spiked samples were fairly linear. Once I get half way through the spiked samples with the linearity the area counts were aabout the same as the neat.
It appears the addition of more organic causes the matrix to retain the additional amount plus what it already has. How can this be? Any help would be appreciated.
By Anonymous on Tuesday, July 16, 2002 - 11:32 am:
You found the answer you seek:
"It appears the addition of more organic causes the matrix to retain the additional amount plus what it already has."
You must keep the amount of liquid in the vial constant and the ratio of solvents used constant to achieve linearity of results using headspace.
You must also test each vial under the same conditions as far as temperature and pressure and time.
You might wish to review the technique and its theoretical aspects again before you design your experiment. Let me explain a little.
If, for example, your sample is dissolved into water and you add your 'std addition' portions dissolved in neat DMSO or some other non-aqueous organic solvent, you will find the partition of the analytes will NOT be linear from differing solutions of different composition. Your drug's presence in solution will also affect the recovery of the solvents you seek when compared to the solution without the drug.
Every and each vial when tested must contain the same amount of organic-water at the same ratio of composition.
On the bright side you have ascertained the problem by yourself, and you have confirmed that chemistry in your part of the universe works the same as everywhere else.
( please smile at my poor joke )
Just redesign your testing and std solutions to meet this requirement and you will get the results you desire. Email me with any other specific questions you may have.
By chris on Tuesday, July 16, 2002 - 11:57 am:
I have a follow-up question. I have not run head space in many years, but I seem to remember that adding sodium sulfate helped to overcome this kind of problem (the USP called for addition of this salr into it's method IV OVI methods). Am I wrong on this one?
By Anonymous on Tuesday, July 16, 2002 - 12:22 pm:
I appreciate your timly response. I wasn't sure what your email address was so I will state additional information here.
The samples were desolved in DMSO.
Only 1 mL of standard was placed in vials and only 1 ml of the deluted sample was place in vials.
Since the standard and neat samples were both deluted with DMSO the sample was spiked with 1 mL of the standard.
I assume that this covers your argument or did I still miss something.
By Anonymous on Wednesday, July 17, 2002 - 05:13 am:
Thanks for the additional but critical information.
This changes the situation drastically. It helps to hear the whole story.
But the steps you perform to make the sample, sample spiked and the standards are still unclear to me, but I believe I can help you through this.
RGEORGE@SIAL.COM IS MY EMAIL ADDRESS.
Please email me so I can respond privately.
By dan on Thursday, July 18, 2002 - 06:28 pm:
This is a good one. If I read your post properly the volume in the vial for a spiked sample is 2ml. This would affect signal by decreasing the headspace volume and increaseing the liquid volume. This technique is governed by Henry's law. The sodium sulfate is added in OVI methods to "salt out" the organics and increase the amount in the headspace. Although this can help it does not account for why there would be less in the spiked sample compared to the sample. Also, how low are the concentrations in the sample? Increased organic solvents would make the solvents more soluble in the DMSO relative to the headspace. Also, what is DME? If my comments are not sufficient to point you in the right direction please type out the current method details. Drug concentration equipment type temperatures and any of the parameters that influence the results. Your problem is complicated and something as simple as incubation temperature and time could be the source of this problem. Good Luck and please post your solution.
By Anonymous on Friday, July 19, 2002 - 07:37 am:
In such cases you can use same amount of some internal standard in standard and sample and take ration of your peak of interest to internal standard peak,so that you can confirm that there are no handling or process.
Many times in GC analysis internal standard can help.
Posting is currently disabled in this topic. Contact your discussion moderator for more information.