However, in case of not getting appropriate external calibration curve at the analysis stage, we do need to add internal standard to delete possible systematic analytical instrument system errors.
Generally, it is not require to add internal standards at extraction stage, because if you add internal sanded at the extraction stage, there is a risk to loss Internal stranded with the extraction and cleanup process. so that It should be add to the sample at the analysis stage of sample (that mean after extraction and cleanup process).
It is good practice to add a known amount of internal standard before the sample preparation stage (before extraction, etc.). In this way, losses of analytes at the sample preparation stage are compensated as they are assumed to be proportional to the losses of the internal standard. In addition, another internal standard can be added to an already prepared sample - just before instrumental analysis. This allows to take into account losses of analytes during sample injection (e.g. analytes losses in the GC injector).
It is truth for the recovery studies and quality controlling of the analytes considerd. But It is not required for internal stand. Because, It is used for the calibratinan and to undestsnd the chnges of detecter responce and retention time etc.
Addition of internal standard is very important in residue analysis, and the stage at which you add the internal standard to the matrix depends on what you want to achieve. When added before sample preparation stage, then in addition you are checking the effectiveness of the sample preparation and processing technique. When added to the final extract, then probably your interest is checking the instrumentation efficiency, it could also aid in situations where matrix interference is an issue and thus the ratio of internal standard to the chemical of interest is used.
So all am saying is that, the point of addition of internal standard is dependent on what you want to check in the whole residue determination process
If you need to study 'the effectiveness of the sample preparation and processing technique" mentioned by Samuel Kofi Frimpong, you can do it using the recovery studies (considered analyte) by spike sample.
Further, If you need to study about the recoveries of internal stranded you can do so, otherwise it is not important to add internal standard to the sample prior to extraction and cleanup steps.
In addition, for my satisfaction and for my knowledge, I would like to ask acceptable reference for the Samuel Kofi Frimpong's explanation while thanking him, if he do so.
Could you please notice to the Piyal concern and let us know whether there is/are any reference(s) available, or more scientific reasons behind your explanation.
Maybe I need to explain more, I'm talking practical here and not theory. Just ask yourself, why do we use IS in chromatography? Mostly IS aids in cases where there is the possibility of volumetric recovery losses. In situations with multi-steps sample preparation and processing, for instance, if you are doing liquid-liquid extraction, followed by evaporation and then re-dissolution of dry extract for injection. For such, if you add the known amount of IS before extraction, you don't bother too much of say taking 99% or 70% of the top layer (organic) phase, because the ratio of the IS to the analyte of interest (peak area or height) at any point in terms of volume recovered is constant. It also happens, should there be inaccurate aspiration for injection, that too, the ratio of internal standard to analyte of interest in whatever amount injected is constant.
So I still insist, when to add IS in residue determination is solely dependent on what you want to check or achieve. In any case, you could read more or check from any good separation science publications.
I think it is important to note where the reference is taken from, and not to take our contribution out of context. You may realized that, your reference talks more of gas chromatography and its interference that can be eliminated by the use of IS but never talks of volume recovery error during sample preparation and processing that can also be eliminated by the use of IS. I am not saying IS cannot be added to final extract before injection just as the reference said, but I'm saying it can also be added before extraction to aid elimination of errors for sure. And we should know the difference between IS and spiked/recovery/fortification, and what each can help achieve.
You need to know what IS is for the method. In the past, when a chemist used manual injection on the GC analysis, they added IS at the end to compensate for the error on injection volume (it is hard to read or inject 1 uL every time). IS will correct for this error. Later on, the autosampler helped to fix this issue. IS was also used to correct matrix-enhancement in hot GC. When you inject std in solvent in hot GC, the response will be less than the same standard in the sample matrix. The hot GC injector will cause compound degradation and sample matrix protects the analyte. In LC-MS, you have the opposite issue where you have matrix suppression (std in solvent has a higher response than the standard in the matrix) and you use IS to fix this issue. However, many methods added IS during sample extraction to fix the losses during sample extraction/cleanup. I think this is not an appropriate approach. If you pick the wrong solvent and the recovery is 50% or has poor cleanup step and lost more analyte at 25%, your total losses will be 75%. However, since IS and native standard will lost at the same rate (75%), the "apparent recovery" will be good because the ratio of native std/IS will be the same e.g. 25/25 or 100/100. It will guarantee that you will get perfect % recovery every time, no matter how lousy the method is. I usually try to fix the extraction/cleanup procedure to get the maximum recovery and add IS at the end to fix the matrix effect. Again, this is just what I will do.
Thank you very much for your comment, However, my question was
whether it is needed to add internal standard at extraction procedure stage, NOT in the GC ingestion process.
As you know, sample preparation is a per-treatment stage used before instrumental analysis in order to make the sample as purified/concentrated as possible.
Dear Narong Chamkasem, I can understand your indirect idea. Thank you for comprehensive explanation.
Dear Seyed Jamaleddin Shahtaheri,
I would like to give further explanation for this targeting your question.
Narong Chamkasem has given clear idea that what is the importance of adding IS. Yes it is correct.
Since, IS is considered naturally not available, What is the importance of doing the recovery studies for IS?
What is the important of compering recovery of IS and the analytes considered ?
If anybody sagest this, It is important to give clear explanation for doing that and its importance .
I would like to say that the Chemical structure of IS is not similar to the compound (that you are going to analyze) hundred percent. Because of this different nature, Interaction of
IS-Sample matrix may be not similar to the interaction of Targeted compound-Sample matrix. Therefore, at such condition, no meaning adding IS at per-treatment stage.
Yes, if you add IS at the extraction step, it will help to "correct" for the losses during extraction (one shake of liquid-liquid extraction will not give you 100% recovery) and during sample cleanup. Even if you drop half of the sample on the floor or use the wrong solvent, it will fix that. Assume that you use QuEChERS extraction with methylene chloride (instead of acetonitrile) and it will extract polar analyte only 10%, it will extract 10% of added IS as well and you will get 100% recovery. To make a rugged method, I think you have to use the proper extraction procedure and proper sample cleanup (if needed) to get maximum recovery (> 80%) and use IS at the end to fix for issue that you cannot avoid such as matrix enhancement in GC or matrix suppression in LC/MS. Again, it is your choice.
One followup question: Once a method is developed and optimized to get 80-120% recovery without the aid of an IS, wouldn't it still be advisable to add the IS during extraction to account for any other errors made during prep?
Ex: Sample is extracted correctly but a pipetting error is made during dilution (say 15% too much) into ALS vial. If the IS is already in the extraction solution, this error is accounted for. If the IS is added with or in the diluent, the error could go unnoticed. In this case, would it not be better to have the IS in the extraction solvent already? Adding the IS directly into the bulk extraction solvent has been a technique I've used with good success to date, but I've always wondered if there was a compelling reason not to do so.
Again, this is my opinion. The IS has the benefit, and it will fit for purpose. IS should be used to cover the error that you cannot control such as manual pipette to GC injector port (1 uL is challenging to do it precisely), change in volume during the extraction due to the moisture in the sample (we don't know from one sample to another), pipette error that you cannot control (not your sloppiness), matrix suppression for LC/MS. As long as you know the possible source of error and try to minimize it by using IS, then it is ok. It makes your method a bit user-friendly. The issue of adding IS at the beginning is the cost because you have to add a lot and it can be expensive. It can disguise the poor extraction method which will not work with incurred residue that we know the concentration before such as proficiency samples. My point is, you have to use it properly. Sometimes, we add different IS just to do a specific job and called it a different name, such as surrogate standard. You can add before extraction to check for recovery of the set such as d10-chlorpyrifos to verify that the set is under control for pesticide screening (you would not see this compound in nature). You may use deuterated-naphthalene just before injection to fix the matrix enhancement in the GC injector port.
I agree with you that you have to prove that your method is good (recovery > 80%) before using IS to fix an uncontrollable error, not to use to get good "apparent recovery) of the poor extraction method. The freshly spike is not an ideal recovery check because the analytes don't have time to stick to the sample like "incurred" residue. The analytes are at the surface in a short time and extract with solvent. It just proves that no surface absorption and you go a good mass transfer from one phase to another. The ideal way to check how good your extraction is to applied radio-labeled compound (C-13) to the matrix (spray to plant and let it go inside the cell). First, you count the radioactivity of the sample and compare with the activity of the final sample extract. It is not easy to do and expensive. I did it a lot when I worked for a pesticide manufacturer company because they have money and expertise.
Thanks for your fast reply! I've always operated under the assumption that the extraction efficacy is shown during initial method development, but never had the resources for the radioactivity test. Certainly a neat way to go about it! In the high-volume production labs I've worked in, mass transfer was always a threat to our accuracy, hence the strategy of adding the IS to the extraction solvent.
Thanks again for your notes on the topic, I appreciate the help.
Actually, there are two parts a) method development and b) routine analysis
a) a chemist must ensure that the method is rugged, accurate, and precise over time, particularly for the incurred residue. Ideally, use the sample with incurred residue with the known amount.
b) during the production, the chemist must have a QC/QA protocol to prove that the analytical set using that method is under control and usable. This is by doing the fresh spike to prove that you get the recovery accurately (get the amount you expected). The chemist needs to show that during that set, the instrument is working properly e.g. get acceptable sensitivity, instrument gives a linear response, the standard solution in the set was prepared accurate (compare ICV and CCV) and there is no error in dilution or sample transfer.