Is it accurate and acceptable to calculate sample concentration using only two standards in LC-MS?
Yes and no! Typically, during any HPLC run a reference standard is used to calculate the concentration of the analyte. However, during method validation you have already demonstrated that you can accurately and acceptably predict the analyte concentration (through best fit analysis, eg linear curve or quadratic).
Jumana Ghannam: Accurate calibration table creation of HPLC or LC-MS data requires that basic fundamentals are followed. Here are just a few of the basics.
- For a statistically significant result (for linear plots. Non-linear curves may require many more points), use at least five different calibration levels of standards (levels = concentration amounts), per order of magnitude, that cover the expected range of concentrations.
- Make sure the range extends 10 to 20% above and below the expected range measured.
- Replicates should be used to demonstrate reproducibility and standard deviation.
I would have to agree with both answers above. It is ultimately a question of accuracy and the experimental question at hand. If you want some degree of confidence in how close to the truth you are, it would be advisable to establish the linearity (non-linearity) of your analyte of interest using a minimum 5 pt curve bracketing the expected range of your target. The more linear the behavior, the better your results will be with a limited calibration. Once you know this, if you want to move forward with a two point curve, at least you have a sense of the magnitude of error and if that is acceptable with respect to your question. If you need less accuracy in your results to answer the question, you can be less conservative in your approach. But remember the opposite is also true.
- Badges
- Science topic
- Similar topics
- Mathematics