Hello Mahsa Lotfi Marchoubeh

Are you using a internal standard? If yes, injecting 1ul won't change the fact that your sample is above the ULOQ. If you're planning to dilute it, you should first make tests to ensure that samples can be diluted without suffering any dilution troubles. I would recommend to dilute a sample with a known concentration and check if you still get the right concentration after calculation.

In any case, you have to make sure that the special treatment you provide to this highly concentrated sample (dilution or 1ul injection) is not changing in any way the recovered concentration.

Mahsa Lotfi Marchoubeh: Your calibration curve would need to be re-done, from scratch, using a new dilution level. *All samples must fall within the expected calibration range of the method without additional modifications or changes. **Changing sample prep for "some" samples means the method is now out-of-compliance. Create a new method that is fit for purpose.

Start by FIRST determining what the best final dilution level and injection volume needs to be for your proposed GC analysis method. Keep the injection volume within the instrument's (A/I) optimal injection volume range (this will be different for different brands/models of injectors, but stay away from small values such as "1 ul"). Once the new injection volume value has been determined, then perform a complete calibration, using the new fixed volume level determined for all standards and save the calibration curve as part of the analysis method. Check the new method against standard mixtures which contain all of the possible interfering compounds to insure it is still selective for the intended analysis and rugged enough to deal with the real world peaks that could interfere with the final results.

If there is an experienced chromatographer at your business, please seek help from them on this topic as learning how to create and standardize calibration tables for samples is one of the most basic and important skills you can learn early on.

Mahsa Lotfi Marchoubeh: This is understood and addressed in my response. As noted, you need to FIRST dilute the sample to a level that can be used for the final GC analysis method. In your case, start with a 10:1 dilution (as you state you need something less than 5x the current value). The final optimized dilution value will be found by experimentation (*YOU will determine and record as part of the sample prep SOP for this method). This will place the data point(s) on scale and within the calibration range. Please speak with an experienced gas chromatographer at your workplace (or hire one) so you can move forward.

Mahsa Lotfi Marchoubeh: Please follow good chromatography practices and consult with an experienced local chromatographer before using the GC-MSD system. For a calibration table to be valid, ALL POINTS (at least five levels for a linear curve, more levels are needed for non-linear) must be measured within the limits of the calibration range. As such, you must construct your calibration table to span above and below the expected sample concentration range. Each calibration level must be repeatable (inject five times). The full calibration range does NOT have to be linear at all. It is best if it is is linear!However, the actual fit must be accurately described mathematically using a proper curve fitting algorithm and your chromatography data software should have many different types of available curve fits to use to find a scientifically valid, "best fit" (e.g. Linear, Polynomial, Exponential, Quadratic etc). (1) Define the calibration range. (2) Run standards at the correct dilution value (all the same volume, conc. should vary). (3) Let the data show you what the actual fit is. Use the correct algorithm to fit the points. Test it with standards to confirm accuracy and repeatability. (4) Make sure you are not overloading the column or detector !!! Doing so will invalidate the data and method used.

Mahsa Lotfi Marchoubeh: Your proposed sample preparation strategy still does not make sense and given the detector choice and method selected (GC-MSD is limited in quat. TC, FID etc are better), you may in fact be trying to use GC-MS to quant a sample range that is not possible. An experienced chromatographer would understand these basics and adapt the techniques used to make the method fit for purpose. Please ask an experienced chromatographer for professional assistance.

For the final analysis method(s) to be valid, you have a few options, but here are some basic steps which will work 100%.

(1) Establish the actual Sample concentration range you wish to analyze. This should be one range and it may cover more than one order-of-magnitude.

(2) You need to create ONE SOP for diluting all "samples", ONE procedure for all of them. The procedure will detail the steps needed to dilute one sample so that it is ALWAYS within the range (established in step one above) of your actual Analysis Method. This is a basic fundamental. This SAME procedure will be used with all unknowns and standards. Again, DO not overload the detector, column or injector. Choose injection volumes and concentrations that do not result in "column effects", resulting in column fouling or poor reproducibility.

(3) Create the analysis method, verify it is selective for the sample(s) under analysis and that all possible sample concentrations within the calibration range can be detected and reported accurately (that cover the full range specified in step two above).

(3a) Use whatever mathematical curve fit is required to obtain an accurate "best fit" for your standards. Note: this may be non-linear fit, including SIGMOIDAL (*I use that one a lot with CAD and ELSD detectors in HPLC), but find one that works well. All unknown samples and of course all standards must fall within, not outside of this range.

(4) IF you find that the "low" and "high" ranges are just too different, then you can create TWO SEPARATE GC analysis methods, each with their own SEPARATE Calibration Table covering a smaller range (SAME SAMPLE PREP SOP is still used. Do not change it), one for the low range and one for the high range. This will mean that you will have two separate calibration tables for each range, each with their own mathematical curve fit for the separate groups of standards. While this approach is not preferred, it does offer you the ability to optimize the TWO SEPARATE GC methods for the two separate ranges (this approach may be useful when samples range across many orders-of-magnitude). Start by using the "High" range GC Analysis Method first with all unknown samples, then if needed, run the low range analysis method next.