Hello,
I have run into a technical problem during some recovery experiments involving an analyte behavior I haven't seen before, and I thought maybe someone might be able to shed some light on the problem, or perhaps make a suggestion on how to solve it.

The analyte is a barbituate and exists as the free acid form or the salt form.

The Problem: As measured against a standard preparation (5mg of standard in 100mL of Diluent), a drug substance assay (5mg of sample in 100mL of Diluent) gives a result of ~95%, while a series of dilutions to the same level (0.05mg/mL)of a formulation sample (100mg/mL to 250mg/mL) made from the very same batch of drug substance tested in the drug substance assay gives a result of ~105%. These results are repeatable using HPLC or UV, at pH's ranging from 1 to 12.

Related Details: The standard is available in the free acid form. The drug substance samples are in the salt form. A correction factor is used to calculate the equivalent concentration of the standard as the salt form.

Gravimetric Assay of the drug substance is 99%.

The detection wavelength is changed depending on whether the diluent pH is adjusted to low or high pH. The results are obtained independent of diluent pH. Low pH diluent must have some organic content (acetonitrile) to solvate the free acid. High pH measurements were performed on the UV only.

Pretreatment of solubulized standard or drug substance in acid or base (0.2N) does not affect the results.

Thanks for taking the time to look at this & have a nice day.

Hard to see a single cause. Assay of substance is low while assay of sample made from substance is high.

How are you doing the gravimetric assay?
Have you done Karl Fischer on standard and substance?
Is this a solid dosage?
How are you performing the dilutions to get a sample to 0.05 mg/ml?
What is the percentage of drug substance in the formulation?
What happens if you just take 5mg of your formulation and dilute to 100mL and correct for percentage in formulation?

Right now the only possibilities I can see involve a combination of problems. Like water or a correction factor error causing the substance to come back low. Then perhaps the initial dilution of the formulation sample occured in a small volume and/or had a large amount of insoluble excipients that caused a volumetric error causing samples to come back high. Good luck.

Hi Tom.
Gravimetric Assay is USP. Add 1mL Conc. HCl to .5g of the sample dissolved in 15mL of Water. Extract with 4 X 25mL of Chloroform. Evaporate, Dry @ 105C for 1/2 Hr and weigh. Three preps gave 98.96% with 0.33% RSD. (Pretty Good, 'eh?)

Standard is dried 4Hrs @ 105C and used as is to assume 100%. The Drug Substance gives 1.6% Loss On Drying (1 prep), and 1.5% Water by KF (3 prep's). (The COA says 0.3%.)

Drug substance assay prep and standard prep for HPLC are both 5mg QS to 100mL with diluent. The formulation is 100mg/mL made by 1000mg of the same drug substance QS to 10 mL in water. The assay prep for the formulation is 1mL of the formulation QS to 100mL, then 1mL of that to 20mL with diluent.

There are no excipients in the formulation.

I appreciate your time. Thanks.
jcb

Hi John,

I think you got bit by a change in the partial molar volume of your drug substance as you went from a dilute solution (assay prep) to a 2000 fold more concentrated solution (formulation prep). Prep your standard like the formulation sample and this error should go away.

Still not quite sure why substance is coming back low. I guess the water could account for some of it. What is the CofA purity of the substance? What do the chromatograms look like?

Hello Tom,
I've never had this issue with the hundred or so compounds I have worked with in the past.
COA Purity is 98.7% based on the above USP Method.
Chromatograms show two related manufacturing impurities for which I have standards. The total %Area of the two peaks is less than 0.1%. The analyte peak has a tailing factor of ~1.0.
Still no solution.
If I assay the free acid from the USP assay using HPLC I get~100%, but the Drug Substance gives only ~95%. Even trying to make the acid under harsh conditions in situ (concentrated acid pre-treatment) doesn't give the expected result.
How can a Partial Molar Volume change affect the results when the API is formulated using a known weight in a volumetric flask followed by dilutions using a volumetric pipette into other volumetric flasks to get the final assay concentration?
jcb

Years ago we had a problem related to solubility of a drug substance. It had a larger particle distribution than the RSM and therefore needed more time to be dissolved. After increasing sonication time, everything was fine. Otherwise we would get a assay of 95% or so.

John,

Sorry, I think you are right. Dim recollection of a problem we had with some small polymers a long time ago. Our initial preps were gravimetric (w/w) followed by volumetric dilutions. I guess in your case any change in volume of the substance should wash.

If there is a cheap supplier of the free acid form I would still prep it like a formulation sample, maybe something else is going on. Maybe someone else will post with a real idea :-) Good luck.

Just as a final note.
Storage of the aqueous recovery solutions (100mg/mL) at room temperature for 1 day or at 5C for ~7 days gives 100% of the theoretical recovery.
Making the recovery solutions with diluent instead of water gives immediate 100% recovery.
Pretty strange behavior for a solution.

Looks like a solubility problem as mentioned above.