API Degredation Product in HS-GC Residual Solvent Analysis

[Terry]

Hello everyone,

I am developing HS-GC test method for residual solvents in an API. There is a huge extraneous peak in the chromatogram obtained with the test solution.

With the help of HS-GC-MS, the peak is identified. I have to investigate its origin.

The production people tell me it is neither a process byproduct nor an impurity from other solvents used. The only reasonable explanation seems like: while being heated in the HS oven, API degrades to give rise to this substance.

The problem is that I can not come up with a scientifically sound degradation pathway. I have consulted some senior chemists, they have no idea either.

Here is my question: could I just perform some experiment to substantiate my wild guess? and how?

Many thanks in advance.

Terry

[Peter Apps]

The first thing is to verify that the peak is coming from the API by running blanks and standards with no API in them.

Peter

[Terry]

I have run air blank, blank solution and standard solution, there was no corresponding peak.

Terry

[Peter Apps]

Hi Terry

API from different batches, would be expected to have slightly different impurities, but at a given temperature for a given time in the headspacer the breakdown should be consistent. If the large peak varies form batch to batch it is probably an impurity, if not it is decomposition.

Can you get hold of pure (as opposed to production) API ? If the pure stuff gives the peak it is breakdown, if not it is production impurity.

What are you headspace conditions ?

Peter

[Terry]

Can you get hold of pure (as opposed to production) API ? If the pure stuff gives the peak it is breakdown, if not it is production impurity.

LC impurity test of the API shows a nice result, all test items are within the specification. I was told that it was pure.

I have tested as many as ten batches, the peak area is quite consistent. I have also tested the API obtained from another supplier, it gives the peak too.

What are you headspace conditions?

Oven: 80 ℃, Loop: 90 ℃, Transfer line: 105 ℃, equlibration time: 30 min

I have used water and DMSO as diluent, respctively, the peak is always there.

[Peter Apps]

Hi Terry

From this evidence it certainly looks as if you have a breakdown product. Try lowering the equilibrium temperature, and shortening the equilibrium time.

Peter

[krickos]

Hi Terry

Yeay would agree with Peter, I would set a temp gradient for the HS oven ie starting low at 35/40°C and step up 5-10°C for each injection to see when the problem starts. and as Peter said kept equilibtrum time short (max 15min if posssible).

Do you mind sharing the APIs name if you really want to dig into the pathway? seems like a generic drug as you tested another suppliers.

Seen similar problems at elevated temparatures for some "omeprazole" type APIs and other more "complicated/strained" structures.
As a general rule, stay away from elevated temperatures if sensitivity is not crucial,=> less problems


By the way if reading right you have done HS-GC-MS and LC/MS of the API, but did you try injecting the HS sample solution after heating in HS on the LC/MS? There might be a non volatile "intermediate" break down product.

[Terry]

Peter and Krickos,

Excese me for absence on this board. I have been enjoying a long-awaited new year's trip.

I would try setting lower HS oven temp and shorter equilibration time in the following days, hopefully it will make a difference.

Do you mind sharing the APIs name if you really want to dig into the pathway? seems like a generic drug as you tested another suppliers.

I am not sure whether it is against the forum rule to disclose the name of the API and its degradation product. I may send it to your mailbox.

but did you try injecting the HS sample solution after heating in HS on the LC/MS? There might be a non volatile "intermediate" break down product.

I do not have a LC/MS

[Terry]

My boss have just given me a degradation pathway, so no experiment is needed any more.

A happy ending