[Faris]

Hi Everyone,

I've been wondering - when scaling up a reaction to industrial scale (say, going from mg to kg quantities) does everything usually scale linearly ?  Now I know that this will always be determined in an empirical way ultimately but I'm just curious.

Maybe another way to phrase the question: is there any reason why the ratio of catalyst to reactant or ratio of reactants to solvent would change during scale up ?

I was reading about some reactions that involve mg quantities of reactant in say 10ml of solvent and it struck me, if someone wanted to scale that up to  kgs it would mean thousands of litres of solvent.  I'd love to hear any experiences.

Thanks,

Faris

[discodermolide]

Usually the research reactions are run very dilute as you noticed. So on scale-up you would try to make things much more concentrated.
The ratio of starting materials, usually remains the same, or gets better. Lots of lab experiments are run to optimise a reaction, and there are lots of parameters to look at. It really depends on the reaction. Work-up becomes crucial and things should crystallise, no chromatography but even that can be done , at least in the Pharma business where costs of product are not limited.
But I have gone from mg to kg and it's good fun.

[curiouscat]

Its a very good question. Like @disco said lots of factors are involved. Some reactions can be highly exo / endo thermic. In which case the heat transfer area of large batches is smaller relatively so that can get limiting.

Often you might need a solvent because you are refluxing it to take heat away. Stirring on scale is also never as good as in a lab so that can affect yields.

About the large solvent excesses in lab scale I think that's a flaw in the system. People often publish patents and papers with concentrations so low that the scheme is entirely unworkable. Perhaps if you are making a $2000/kg compound you could afford it but for most chemicals those solvent excesses are just totally impractical commercially.

[TwistedConf]

I've been wondering - when scaling up a reaction to industrial scale (say, going from mg to kg quantities) does everything usually scale linearly ?

Sometimes it requires completely changing the synthesis.

Process chemists do some great things...  more than just running a bunch of already worked-out reactions on larger scale.

[zsinger]

See: Bosch-Haber Process.  GREAT example of a industrial scale up of an idea which was Fritz Haber's.
          -Zack

[Arkcon]

If you want to learn more, you can browse our chemical engineering sub-forum, and see the sorts of topics that come up.  I know I don't actually understand most things, but its fun to see what I don't know, and tease out a glimmer of understanding.

http://www.chemicalforums.com/index.php?board=33.0