[discodermolide]

Under or over charging can also be a safety hazard, never mind quality.
I don't think the FDA would approve of just dumping in "half a bag".
The operator is not jumping through hoops when he is doing his job properly. If the plant procedure calls for 525 lbs of material to be added then it should be weighed. Otherwise it is just guesswork and when something goes wrong you will never find out what happened.

[curiouscat]

Under or over charging can also be a safety hazard, never mind quality.

Where it is a hazard one absolutely shouldn't do it.

I don't think the FDA would approve of just dumping in "half a bag".

Right. No doubt. OTOH there's a big chunk of the industry that's not pharma.

[discodermolide]

Sure there is a big industry that is not pharma. However the compounds delivered for public use still need to be approved. S there is really no use for sloppiness. As a customer I would expect a product that I ordered to conform.
It can be likened to the garage replacing the oil in your car with some other oil that was "not quite correct", what would you say when the engine crapped up? Or something in the petrol that should not be there just because the operator dumped in half a bag instead of doing it properly.

[curiouscat]

If you're an engineer, what would you most want a new process chemist to understand, or what misconceptions does a new process chemist need to get cleared up?
 

I cringe at giving advice not being wise and experienced like @disco and others.

But for what it's worth I'd add a small point: Treat everything mentally as a cost-benefit analysis: Rather than ask "Can I replace the 5% accurate flowmeter with a 1% accurate flowmeter?" I'd start thinking as "If I could replace the 5% meter with a 1% meter what is my process improvement and how much would I be willing to spend on it?"

OR
"Maybe three extra washes will give me 99% purity material but if the 95% grade sells for almost the same price were my washes worth it?"
OR
"Does this process spec. really need to be 'Maintain at 55°C for 6 hours' or can I write it as 'Maintain between 50-60°C for 6 hours'"

Apologies if I was stating the obvious or the indefensible. 

[curiouscat]

S there is really no use for sloppiness. As a customer I would expect a product that I ordered to conform.

That's not what I meant at all. Apologies if it came out that way. It wasn't an excuse for sloppiness.

How do I put it better? Maybe, "Do everything you need to make sure your product confirms but do not overdo things that are not necessary to the product confirming."

Or "Acknowledge that the sensitivity of a process to different inputs is not the same and a good operating procedure needs to be designed to reflect this."

[discodermolide]

We used a quality risk analysis to define the bandwidth of process parameters. Late in the development the product specs were set then we started to fiddle about with things like time, temperature, over/under dosing, solvent and so on.
This way we could define that parameter which was critical for quality. This was kept up even in chemical production.
I have several examples on this computer but they are in german. I'll send you one if you want.

[stewie griffin]

Excellent advice from both disco and curiouscat. Thank you much!

[eazye1334]

Just to expound a little on curiouscat's previous advice: if you ever have the opportunity, physically get in there and help them with the job if you can.

For example, I'm trying to introduce a chemical change to improve safety here at our electroforming shop. I've been met with some resistance from the guys who have been here a long time and it doesn't help that I'm a good 20 years younger than several of them. What has helped immensely is that I've gone in and done their everyday work using my new chemical process on my own. This shows that I'm willing to do the work, I'm willing to use the chemical myself, and I'm willing to get dirty myself to get the work done. On certain days when we get really backed up, I'll put my work on hold and become essentially another worker so we can get back on track. Naturally I understand this isn't possible everywhere, but I would really suggest trying to find some way to help out a few times. It really gets the guys on your side and helps immensely in the long run.

Also, be willing to answer why. Trying to convince a technician that something needs to be changed or that process instructions need to be amended can be quite a chore. If you explain why and what benefit comes from the changes, many are willing to give it a shot. They may not understand entirely, but they at least  appreciate you spending the time with to them to try and help their understanding. I've found that those workers are the ones who come to you to explain problems or difficulties that you may not know about otherwise; it really opens up communication lines.

[DrCMS]

I've been reading this whole thread and it strikes me as to how different the outlook is between industrial chemists working in pharma to GMP and those of us who do not.

However the compounds delivered for public use still need to be approved. S there is really no use for sloppiness. As a customer I would expect a product that I ordered to conform.
It can be likened to the garage replacing the oil in your car with some other oil that was "not quite correct", what would you say when the engine crapped up? Or something in the petrol that should not be there just because the operator dumped in half a bag instead of doing it properly.

I'm sorry disco but this is utter tripe. 
The vast majority of chemicals produced by the chemical industry are used by other industries not by the public.  They are used to do a job and batch to batch variability is built into the product specification.  If the customers process/product can not work with material from either ends of the spec they need to work on their process/product.
Chemicals are tested for their physical and chemical properties but they do not have to be approved by anybody and minor batch to batch variability is not part of that test regime.
We are not talking about sloppiness we are talking about efficient manufacture and not weighing to the nth degree just because a bench chemist did so when they first made the product 10 years ago.
As a customer you have the right to a product that meets the specification and that's it.

With the products I make the process can be changed every batch if I feel like it as long as we meet the final product spec agreed with the customer (which I wrote to match with what I judged we can produce consistently).  I'm looking to make each process as cheap as possible.  That might be reducing raw material usage or batch times.  Using a small excess of a cheap reagent may be cheaper than the extra time taken to weigh it out accurately.

Pharmaceuticals are different because they interact with a very complicated chemical factory in the human body.  As we do not understand all those interactions we seek to standardise the drug manufacture from batch to batch hence GMP.  Once the drug has been approved the process is pretty much fixed.

I have deliberately set up processes to work to full bags/drums/IBC's of reagent to remove possible charging errors and to save time.  If you understand what is going on in your process that is a perfectly acceptable step to take for some reagents.   You might do it for operator ease of charging or because the reagent has a short shelf life etc etc. 

It's a lot easier to throw in 10 full 25kg bags of sodium bicarbonate to make the pH alkaline in every batch than it is to add enough to give a pH of 8-9 for each indiviual one.  If having collected the data you find the charges have all been between 230 and 240kg that 10-20kg of bicarb in excess costs peanuts but the time and hassle saved is worth much much more.

Do not assume that everything needs to be precise, in reality most things do not.  With drugs we're not as sure which things really are important so GMP makes them all important but for the rest of the chemical industry that is overkill that would cost most things out of existance.

[curiouscat]

@DrCMS

Ha! Now this sounds like someone from exactly the sort of plants I've worked in! As I was reading your post I was going "Yes!" on every line.

Thanks! I couldn't have put it down that well myself. I think it is indeed a wide gulf between Pharma and the rest.

disco is probably exactly right too but in his domain. The procedural and cultural schism between the two sectors is just huge; so large that SOP's from one might seem totally alien to the other.

[DrCMS]

@DrCMS

Ha! Now this sounds like someone from exactly the sort of plants I've worked in! As I was reading your post I was going "Yes!" on every line.

Thanks! I couldn't have put it down that well myself. I think it is indeed a wide gulf between Pharma and the rest.

Thanks for the support.

disco is probably exactly right too but in his domain. The procedural and cultural schism between the two sectors is just huge; so large that SOP's from one might seem totally alien to the other.

I'm sure disco is completely and utterly right about pharma and GMP but there is just such a huge difference in approach between the two sides,  It is difficult to understand how one side can possible be right if you're from the other camp.

We do some work for a major international pharma company that does not need to be to GMP and would not be viable for them if it required GMP but it's still hard work dealing with them because all their internal systems are set up to expect GMP and that is the only the way they think.  We need to remind them every now and again that we're not working to GMP and they specifically came to us to get the right product at the right price without GMP.

[curiouscat]

Even in the Pharma industry, it is only the last API stage manufacturer that is tightly controlled, right?

Anyone upstream in the chain operates like a regular chemical firm?  Or am I mistaken?

[discodermolide]

Sure the steps from the API to the drug substance are tightly controlled.
The preceding part of the synthesis depends upon the company, I would guess?
I the company I worked for the attitude was;"If you are manufacturing your intermediates in a GMP Plant then, the synthesis of these intermediates will be carried out with all the normal GMP requirements."
But then our company had a bug up it's corporate ass concerning cGMP.

[curiouscat]

Sure the steps from the API to the drug substance are tightly controlled.
The preceding part of the synthesis depends upon the company, I would guess?

Right. OTOH these days a lot of the intermediates get contract-outsourced or purchased. So that's another can of worms.

I always had the impression that GMP is in theory great and well-intentioned. In practice though, often it skews heavily towards documentation and the underlying purpose is often diluted.

What was your experience?

[discodermolide]

"I always had the impression that GMP is in theory great and well-intentioned. In practice though, often it skews heavily towards documentation and the underlying purpose is often diluted."

What was your experience?
[/quote]




Exactly right. Remember those papers I sent you? Well we made 60g (actually 120g, but only 60g was used). I guesstimated that we produced around 7kg paper/gram drug substance.
The FDA is not interested in the chemistry, unless something is obviously wrong, it cares more about your SOP's, and were they followed to the letter. It cares more about the analytical data of your compounds in the GMP steps and the cleaning records of the plant.
Chemistry, what's that? It's a means to an end.