Parabens - stability studies

[Aleksandra]

Hi everyone,

during stability studies of parabens in gel samples i got unexpected results. I observed peaks of hydroxybenzoic acid, methyl paraben and ethyl paraben. Methyl paraben was the only one used in formulation, hydroxybenzoic acis was a product of hydrolysis of methyl paraben and ethyl paraben was a product of transestrification of methyl paraben (ethanol comes from formulation) but after calculations i didn't get "sufficient recovery" of them.
I know that at time 0 there was 0,1% of methyl paraben (method recovery around 99%). After over a year there is only 70% of this initial value of methyl paraben, 19% of hydroxybenzoic acid and around 0,8% of ethyl paraben.There is 9% somewhere missing. I consider aromatic ring as stable and do not think that I miss this 9% becouse of degradation of parabens to forms which would be not detected on UV detector.
On the chromathograms there is no other peaks observed, and as I used DAD detector I've checked full UVspectra looking for missing 9%.
Column was C18 (lenght 12,5 and 25), mobile phase 1%CH3COOH:ACN 83:17, temperature of column was ambient, samples were prepared in methanol.
I've also tried Phenyl-column and CN-column, 0,1%TFA instead od acetic acid and results were always the same. Have introduced sample heating to the method and it didn't help either.
Can anyone please suggest me some solutions of this problem or give me some advice what could be the cause of this lost in mass recovery?
Thank you in advance!!!!!

[tom jupille]

First of all, you need to clarify your percentages. I'm not sure where you got the 9% missing.

Second, are you referring to weight percent or mole percent? If you are referring to weight percent, are you correcting for molecular weight differences when you compute your recovery (the molecular weight of p-hydroxybenzoic acid is 9.2% lower than the molecular weight of methyl paraben).

Third, hopefully you kept time 0 reference samples in the deep freeze. What do you see when you re-analyze using one of these?

[BB!]

Hi everyone,
In my opinion the problem looks like that:

At the beginnig (time 0) you have 100% of methyl paraben in your gel. Then after some time the methyl paraben is decomposed into hydroxybenzoic acid (lower molecular mass) and ethyl paraben (higher molecular mass). The mass of these three substances should give you almost 100% of mass of methyl paraben, but you obtained only 91% of methyl paraben mass at the time 0. Is that true?

Of course considering loss caused by differences in molecular mass of methyl paraben and hydroxybenzoic acid you may suppouse to lose around 2% of overal mass. Why?
Since if your methyl paraben decomposed into hydroxybenzoic acid in 100% you would obtain 9.1% (difference in mass) mass loss (CH3OH cannot be seen using DAD i supose). Unfortunatelly you obtained, let me say "only" 19 % of hydroxybenzoic acid what means the mass difference should reach around 1.8%, but some of CH3OH was used to produce ethyl paraben so the mass loss caused by differences in molar mass diminish to around 1.5 %. Where is yours 7.6 %? I do not know.

Maybe package (tubes?) is a problem?

What is the point of checking samples from deep freeze? Do you think that something wrong happens with the system? I am asking since I usually analyse freshly made samples to check if standards are all right and then analyse samples from stability studies, but it might not be the proper way.

//Sorry for my english but it is not my mother tongue

[Aleksandra]

I apologizes for writing back so late,

First of all, BB!'s way of thinking about the nature of the problem is correct. I referred to weight percentages but after recalculation to molar percentages my result improve only by 1 % giving 92 % of mass recovery (should be >98 %).

Second, I do not have samples stored in deep freeze.Thus to check system and standards I just use freshly made samples. My time 0 values are these obtained at the beginning of stability studies.

I tried to make it a bit more clear, hope it helps.

[BB!]

I have been investigating your problem but it is not easy to say what is going on with your samples. Maybe try to tell us more about composition of the sample.

Package, any other substances presence, conditions and time of samples storage etc.

[tom jupille]

The reason I was hoping that you had preserved frozen T=0 samples is to allow a re-check in case there was an error in the original analysis.

[Aleksandra]

First:the method i used was previously validated. second:I obtained good results for all test until 18th month of stability studies. Unfortunately don't store time 0 sample frozen.

[DSP007]

First:the method i used was previously validated. second:I obtained good results for all test until 18th month of stability studies. Unfortunately don't store time 0 sample frozen.

The water wears away the stone.
Hydrolisis and reeterification wich etanole. End oxidation oxybezoic acid wich air oxigen/ Experience to get dirty. But you can calculate the reaction rate constants for unknown conditions
PS
Judging by the name - Slavjan lady ( russian or ucraine,or belarus or bulgar).
Dobro pojalovat na anchem.ru

[Aleksandra]

Sample was kept in tubes not permeable for air oxygen, so there should be no oxygene inluence.

Sorry don't understand russian:(

[DSP007]

Aluminium tube or polymer tube? Or in sealed (in argone) glass ampules ?
Oxygen diffuses through the polymer is very good.



Also unclear recipe mixes recipe- only ethanol, water, and parabens, or something else?

[Aleksandra]

It was polymer with EVOH membrane. It was gel sample, consist of Carbomer, NaOH solution, Menthol, Aseptin M, EtOH and Aluminium Acetotartrate as drug substance

[DSP007]

May be many variants of problems!
1) Unsoluble aluminium salt
2) Sorbtion on plastics
3) Oxidaton
4)Other
Aseptin is metilparabene TM &?

[DSP007]

Oxigen diffusion throght polimers
http://www.sciencedirect.com/science?_o ... archtype=a

Oxbenzoic acid oxidaton throht quinone acid.
Should be unidentifiable peaks in early chromatogram.

Sorbtion on tube - Rinse with water pipe, cut to pieces and extracted in a Soxhlet

[Peter Apps]

Hi Aleksandra

How did you calibrate the detector response to analyte quantity for the hydroxybenzoic acid and the ethyl paraben ?

Peter

[Aleksandra]

Calibration curves for each: hydroxybenzoic acid, methyl paraben, ethyl paraben standards were prepered and were analyzed in the same sequence as samples.

Will try to extract parabens from the tubes tomorrow and will let you know how it goes. Thank you for ideas btw by Aseptin I meant methyl paraben.