Regarding Linearity Range

[AdrianF]

HWM - If you can find the perfect IS that is good but if you are having to deal with a multitude of NCEs it can be extremely time consuming to find a suitable compound.

As I said in my previous post this is not relevant as the matrix will have none of the NCE present.

In my experience internal standards add unecessary complications - results are better without them. The only time I have used them wilingly is with GC where the volume of injection was uncertain.

[aniket]

Hi All,


I am getting rally confused by the talk of Adrian and HW as well as Mark, regarding the use of an IS in Discovey stage!!! I use these IS as my extraction as well as sample prepration steps are not optimized so i feel the use of an IS in such situations is justified. Also i ahve a bank of compounds and their retention behaviour so i can easily select an IS, many times i have seen many of calibration lines would not meet the acceptance criteria if i exclude the IS from calculation, i have observed great deviations for the calibration without an IS, so my approach is "It is better to be safe than sorry!" Can i ask you all once again what is the best strategy for Discovey Lab?? I am sorry if i am repeating my question?
But this is really important and a great learning experience for me so please will all please discuss this issue in detail.

Thanking You,

Aniket

Keep Smiling!!

[HW Mueller]

AdrianF, the initial post by aniket is apparently talking about bilogical matrices.

aniket, The sentence of Mark which I quotet above explains it all.
Maybe a hypothetical example is in order: Lets say you calibrated your system with a standard solution, including an internal standard. Now you use this calibration to evaluate your unknowns to which you added an IS. I would call this the external standard (ES) method. It turns out that the IS, which you calculated via the ES method (just like the unknowns), varies all over. Something is wrong. You can now either use the IS to calc the unknowns (you better make sure that the IS variation parallels that of the unknowns) or you can improve your whole method. If the IS "behaves" there is a chance that your unknowns also "behave". If you add IS to all your samples you have a check on all your samples, even if you didn´t use the IS to calc the unknown concentrations.
If you want to be sure of your results you have to determine the unknowns with at least two widely differing methods.
If this is still confusing, just revert to the mentioned sentence by Mark.

[AdrianF]

I have reread Mark's contribution - it says it all - only use IS if you have to.

Aniket it would help the discussion if you outlined your general approach to these analytical problems. If you could list the steps you take we would be able to give you clearer advice.

[Mark Tracy]

Calibration by standard additions will let you compensate for sample recovery problems, but it is tedious, time-consuming, and the cost of reduced systematic error is increased random error. Neither is it a substitute for adequate chromatographic resolution. It is useful for situations where the matrix is unique (forensic toxicology) or you simply can't find a good IS, and you know you need to correct for recovery.

On the subject of IS, don't think of it as a security blanket. Think of it as a tool. The benefits of IS are not automatic, and they come with certain costs and prerequisites. Evaluate the needs of your work, and go where the data takes you.

[aniket]

I am confused by all the Differnt spectrum of views, i am quiet a novice at this, so will please tell explian to me simple terms!!

I have always been tutored the importnace of an IS in an extracticion procedure by my seniors, so it has got stuck in my head, so it seems pretty difficult for me a digest doing a analysis without using an IS.

I many times use a fast common gredient for many compounds, so i can select for different compounds, is this strategy right??

I just wanted to know from experts working in Discovery labs what is their strategy for screeing large numbe of compounds? do you go for cassete dosing and using longer gredients? or you just shift to new tech such as UPLC? what is the best approach??

Compared to all this new technology it feels like i am living some very different old world!!!

Ok Thanks all for replying.

Keep Injecting!!

Aniket

[Mark Tracy]

If you wish to become a senior chemist yourself, you must think your way through these issues. There is no simple advice. The best you can hope for is clear advice; I apologize if my advice has not been clear. Since you are asking these kind of questions, it is clear that your employers are paying you to know, understand and think; do not disappoint them. We on the forum can help you with knowledge and with explanations, but the thinking has to be yours.

As an exercise, sit down and look at the calculations for your method. Ask yourself what assumptions have been made in those equations. Check to see if any assumptions might be violated. Ask yourself where every number comes from. Think about what might influence the reliability of that number. Check to see if the influences have been controlled in your experimental method.

Don't give up!