assay and impurities of finished product

[moonchips]

greetings.

i am using methanol and 50/50 ACN/H2O as sample diluent.
For a specific RRT 0.8 impurity, 50/50 ACN/H2O gave me no RRt 0.8 impurity.
Methanol gave me 1.5% RRT 0.8 impurity right after i put in the extraction solvent. After shaking one hour, it is growing to 4%. After shaking the extraction vial overnight, it is 6.6%.

My question is: How do I know this is not analytical artifacts generated by using methanol?

[moonchips]

In another word, how do I know if this is due to the stability of the solution or the solubility of the RRT0.8 impurity.

Does filtering work out here?

[juddc]

What's your sample? A pure compound, drug substance, drug product something else?

What is your analyte peak area doing - does it decrease at all in the MeOH solution?

What are you using as a detector? If a PDA, do the spectra of the RRT0 0.8 and your analyte look similar?

We need more information to have even the remotest chance of providing assistance.

[moonchips]

it is drug product with API and 5 other excipients.
the API peak area is not decreasing as time goes by.
It is hard to tell. PDA RRT 0.8 looks not competely alike API, but a bit similar.

The sample prep is as following:
lyophilized sample + acetate buffer, then dilute with 100% methanol.
for API and bulk solution, dilute with 100% methanol directly.

With the same kind of sample preparation (100% methanol):

This RRT 0.8 is not in API (acetate buffer is not involved here)
it is not in bulk active solution (before the lyophilization) (acetate buffer is not involved here)
it is not in bulk placebo solution (before the lyophilization) (acetate buffer is not involved here)
it is not in lyophilized placebo (acetate buffer is involved here).
it is in the lyophilized active (acetate buffer is involved here), and it grows with the extraction time.
It is not stability temperature dependent.

if this impurity is real. it is lyophilization dependent. (can not rule out acetate buffer effect). and it is extraction time dependent.

How do I know it is the solubility but not the solution stability?

Can I filter the reconstituted lyophilized active solution, and then monitor the assay and impurity?