I prepared an extract by using 100 mg of plant and 2 mL of ethanol, filtered and injected 20 μl into the HPLC system equipped with Waters® XSelect (150 mm × 4.6 mm, 5 μm) column. If I prepare an extract of 50 mg of plant and 1 mL of ethanol (therefore the same plant/solvent ratio) and also inject 20 μl into the HPLC system, can I expect the same peak (or very close areas), right?

Is the most important thing to maintain the solid/solvent ratio and to prevent the process from being “scaled-up”, e.g. 100× to other factors being relevant? Or am I wrong?

  • 2
    $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. @OAS Please append necessary details regarding HPLC method and further findings from the chat discussion to your question (you can edit the question at your own pace). $\endgroup$
    – andselisk
    Apr 20, 2022 at 4:29

2 Answers 2


I would summarize the extensive chatting to this answer:

  • Putting away the constant HPLC part, focusing solely on the sample preparation.
  • Assuming the dry powdered plant leafs samples are properly and representatively prepared.
  • The same extraction procedure, applied on variable mixture volume, can have slightly different extraction efficiency results.
  • E.g. using the same mixing frequency, bigger volume may be in average mixed less vigorously, or may reach different temperature due conversion of mixing energy to heat.
  • Considering all steps contributing to the method repeatability error, the extraction scaling error ( extraction efficiency, evaporation of volatile solvents) may be negligible.
  • Justification may need verification, e.g. by analysis of variance of the least square method for substance content as function of extraction scaling. There are standard formulas to test the statistical significance of the nonzero slope value.
  • If extraction scaling does not have significant impact, than it is trivially obvious the same sample/solvent ratio should lead to the same result. BUT...
  • While it should lead to the same mean result value, the repeatability error may still be affected by the sample size. 50 mg is not much and the sample granularity and particle non-homogenity may manifest more strongly.

If the result is well reproduceable at 100 mg / 2 ml, 50 mg / 1 ml should give very much the same outcome.

Changing the batch size by an order of magnitude could give a significant change. You never know until you tried. ;) Repeatedly!

  • $\begingroup$ So are you saying that I can probably calm down and breathe because the change do not involve an order of magnitude? $\endgroup$
    – OAS
    Apr 20, 2022 at 22:40
  • $\begingroup$ Yea, probably. ;) Some things scale very badly. For example if you compare a process in a half-filled to a fully filled vessel. $\endgroup$
    – Karl
    Apr 21, 2022 at 13:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.