Chromatography Forum

We have a product that sometimes oxidises during analysis. I have traced the oxidation to occur in the very short time interval when the product is opened (ampoule) and transferred to an LC-vial.

There are long treads about my problem here on Chromforum, but we never really found out why this happened. For a while, the ambient ozone in the air was the main suspect since we could see that it occurred more often during summer. But we have seen this also during winter days when the ozone level was close to zero (it is very dark up here in the winter time).

The other day I noticed that my newly washed lab coat had a faint smell of bleach. Could this be responsible for the oxidation? So I opened two ampoules, put one in my lab coat locker and one of the lab bench. The next morning I analysed the samples for oxidation, and the locker sample was heavily oxidised. The lab bench sample was not.

Have you ever heard anything like this before?

Try to handle the product under inert gas atmosphere. Ambient oxygen already may act as an oxidizing agent.

humidity could also be a factor accelerating the oxidation. A locker with a recently-worn coat will be more humid than an open, air-conditioned lab

On the other hand, if you could *smell* the bleach in the open, that suggests that the concentration in the closed locker could get fairly high. I know everyone is trying to be thorough, but Occam's razor suggests that there is no need to invoke humidity or atmospheric oxygen in this case.

I have done many investigations of this phenomenon. It is not a big deal if the oxidation happens in my lab (development), but a disaster when it happens at QC (we have rejected many batches over the years).

Some basic facts: The molecule is a small protein with a sulfurbond that is very easy to oxidise. On the other hand, it is not possible to oxidise it with oxygen only (I have bubbled 100% oxygen through the solution and nothing happens).

In the past, I have more or less analysed a whole batch of rejected product. The oxidation pattern was very different depending on what day I analysed it (and all ampoules were randomly picked for a big plastic bag). Some days there were 0% oxidised samples, the next day it could be 20% that were oxidised. This proved that the product was not oxidised during production, it was destroyed in the lab.

I have no idea if the "lab coat finding" is relevant at all for this problem, and I am not sure how to investigate it. I cannot ask the entire lab to take off the lab coats...

Maybe you need to forgoe the "how" and purge the vial with nitrogen before you seals it (or after, even) to prevent oxidation.

I forgot to mention that QC is doing the transfer from ampoule to vial under nitrogen, since about six months.

It appears as the OOS rate has increased after this was introduced (hard to tell). At least there is no positive effect.

The drug is very potent = low dose + high molecular mass = not much oxidant needed for OOS. I have calculated the amount needed for OOS to be less than 1 ng per ampoule (calculated as H2O2).

It feels as I am chasing ghosts...

Update:

I have placed open ampoules in plastic bags containing different lab coats. I found that the smell of chlorine was very different on different lab coats. Possibly they only use sodium hypochlorite on the most stained ones? Some coats did not smell at all.

Anyhow: The ampoules in the bags that contained a single-use coat or a lab coat without any chlorine smell did not oxidise. The samples in the bag with a chlorine stinking coat were heavily oxidised.

So I guess that I have proven now that the lab coat can indeed induce oxidation. Then it is a question if this is a problem in real life. It is not realistic conditions to put the samples in a small volume of air together with a lab coat.

We will now state in the method that single-use lab coats must be used. I guess it must be the first method in history that has this recommendation

Hi Mattias

That's a pretty impressive bit of troubleshooting !

If laminar flow cabinets are available they might also solve the problem by blowing the bleach away from the sample instead of sucking it towards the sample as an ordinary fume cupboard does.

Peter

Amazing and congratulations, follwed your earlier threads on the issue.

Thanks! I have put so much work in this over the years, testing more and more crazy things.

I am still not certain that this is the actual cause of our problems, but we have to have a "better safe than sorry" approach. I guess that I will know in a month or two if the OOS rate has gone down (or hopefully disappeared).

I did one last thing today, and that was to let the solution touch the lab coat (by pouring solution and collecting was was coming through the fabric). The solution was analysed immediately and showed a lot of oxidation (more than was achieved overnight when the samples were standing next by).

The part of the coat that got wet smelled a lot more of chlorine, compared to the dry parts. Could maybe be a clue?

hehe, maybe so.

Seroiusly, regardless of this particular intriging issue, I would call the washing department or firm and firmly suggest increased rinsing of laundry. Can not be healthy in long run.