79 – Reactor Design & Scale-Up Challenges (S6E4)

Welcome back to the Deep Dive everybody. Today
we're going to be tackling a topic that I think
a lot of chemists probably don't think about
all that much, but it's a pretty big deal. It's
just this idea of taking something that works
really well in like a small flask in the lab
and then trying to scale it up to actually make
a useful amount of material. Yeah, the spail
up problem. It's a huge one and something that
most chemists you know, we'll encounter at some
point. Yeah, absolutely. And it it's something
that I've, you know, I've dabbled in a bit, but
I'm really excited to kind of get your expertise
on this because I think there's a lot more to
it than just, you know, multiplying everything
by 10 or 100. All right. It's not you can't just
take, you know, your lab notebook and just increase
all the numbers and expect it to to work the
same way. Exactly. Exactly. So today we're going
to be digging into the nitty gritty of reactor
design and the challenges of scaling up chemical
reactions. And to kind of give us some real world
examples, we're going to be drawing heavily from
the journal Organ and Metallic Process Research
and Development. That's OPRND for those of you
who like to use the acronym. Which most people
do in the field. It's a great journal, covers
a lot of very practical aspects of process chemistry.
Yeah, it's kind of like the go -to place for
chemists and engineers who are really trying
to take these reactions from the bench top to
you know actual production so We're gonna look
at some of those examples and see how they tackle
these challenges because it's a it's not always
straightforward alright, so Where do we even
begin with this like when you're thinking about
scaling up a reaction? What are some of the first
things that come to mind? Well, you have to,
you know, when you're starting in the lab, you're
probably working with pretty small glassware,
right? So we've seen some examples in the OPRND
literature where they're using like 500 milliliter
round bottom flasks. Yeah, pretty standard. Yeah,
basic glassware. But pretty quickly, as you start
to scale up, you have to think about what kind
of reactor you're going to use. Right, because
the flask isn't going to cut it once you start.
you know, making kilograms of material. Exactly.
Exactly. And the type of reactor you choose is
going to depend a lot on the specific reaction
you're doing. You know, one example we've seen
in OPRND is cycloaddition reactions. Right. And
just for our listeners who might not be, you
know, super familiar with organic chemistry,
cycloaddition reactions are basically where you
have two or more unsaturated molecules, meaning
they have double or triple bonds and they kind
of come together and form a ring. Yeah, they
combine to form a cyclic product. Yeah, right.
And so those reactions, depending on the specifics,
they might require a certain type of reactor
to really work well. Absolutely. And even at
a relatively small scale, like when you're just
going from that small flask to maybe a liter
or two, you have to start thinking carefully
about how you're going to control the reaction.
Yeah, things get a lot more complicated. They
do. Yeah. And one of the things that I always
found kind of interesting, or I guess maybe counterintuitive,
is that heat transfer becomes like a huge deal
as you scale up. Like, why is it that it's so
much harder to control the temperature when you
have a bigger reactor? Well, it really comes
down to the the ratio of surface area to volume.
In a small flask, you have a lot of surface area
relative to the volume of liquid that's in there.
And that means that heat can be transferred in
and out of the reaction mixture pretty easily.
So it's got a lot of contact with the outside
world. Exactly. But as you scale up, the volume
increases much faster than the surface area does.
So it's getting more and more insulated as it
gets bigger. Kind of, yeah. The volume is going
up by a factor of... the cube of the radius,
whereas the surface area is only going up by
the square. So that means that the ability of
the reactor to exchange heat with the surroundings
gets worse and worse as you scale up. So all
that heat from the reaction starts to get trapped
in there. Yeah, and that can lead to some...
some real problems. You can get temperature gradients
within the reactor. Right, so some parts are
hotter than others. Exactly. You might have some
areas where the reaction is running really hot,
maybe even getting out of control. Oh, that's
not good. No, not at all. And then other areas
might be much cooler, so the reaction is going
really slowly. So you end up with an inconsistent
mixture. Right. And that can affect not just
the yield of the reaction, but also the purity
of the product. Right, because if it's too hot,
you might start getting side reactions and all
sorts of junk. Exactly. Yeah. And I can imagine
that, especially with exothermic reactions, that
could be a really big safety concern, right?
Oh, absolutely. If you have a reaction that's
generating a lot of heat and you can't get rid
of that heat fast enough, things can get very
dangerous very quickly. Right, like a runaway
reaction. Exactly. Yeah. So they talk about that
a lot in OPRND, right? how to manage the heat.
Yeah, they emphasize the importance of careful
temperature control. Right. And they give some
specific examples. For instance, in one paper,
they were making a benzopherin derivative, and
they used reflux conditions. OK, so reflux is
basically where you're boiling the solvent and
then condensing it and letting it drip back into
the reaction mixture, right? Right. And that's
a way to... to maintain a constant temperature.
Right, so it's like a self -regulating system
in a way. Right, yeah, exactly. But even with
reflux, you still have to be very careful, especially
at a large scale. You might need specialized
reactors with jackets that you can pump hot or
cold fluid through. Right, to kind of fine -tune
the temperature. Exactly. OK, that makes sense.
Yeah. So it's not just about throwing the chemicals
together. It's about really managing the energy
flow of the reaction. Very much so. Yeah, that's
really interesting. Now, let's talk about mixing,
because I guess, naively, you might think that
in a bigger reactor, you just need a bigger stir,
right? Right. Just more power, more mixing. You'd
think so, but it's actually a lot more complicated
than that. OK, why that? Well, for one thing,
the... The way that fluids flow in a large reactor
can be very different from how they flow in a
small slask. Right. It's not just like a scaled
up version of the same thing. Oh, no. You can
get all sorts of weird flow patterns. And depending
on the design of the reactor and the type of
impeller you're using, you might have some areas
where the mixing is really good and other areas
where it's practically stagnant. So some parts
are getting, you know, really well mixed, and
other parts are just kind of sitting there. Exactly.
Yeah. That doesn't sound good. No, it's not.
Because if the mixing isn't uniform, you can
get variations in the concentration within the
reactor. Right. So some areas might have a much
higher concentration of reactants. Exactly. And
that could lead to side reactions or other problems.
Yeah, absolutely. Yeah. And I remember, I think
there was an example in OPRD, I think it was
a paper by, get a macedium brach. Yeah, yeah,
I know the one you're talking about. Yeah, where
they had this nitro reduction reaction and they
were having trouble with the catalyst. Right,
the catalyst was getting coated. Yeah, it was
getting coated with something and they couldn't
figure out why. And that was directly related
to mixing and solubility. They found that in
their initial solvent system, the... the starting
material and the product weren't very soluble.
And so as the reaction was proceeding, the product
was starting to precipitate out. Ah, so it was
like coming out a solution. Exactly, and that
was what was coating the catalyst. So they had
to change the solvent system. So they basically
had to find a way to keep everything dissolved.
Yeah. They ended up using a mixture of THF and
water. THF, so that's... tetrahydrofuran. And
that improved the solubility enough that the
catalyst didn't get coated anymore. Wow. So just
that one change made a huge difference. It did.
That's really interesting. So it's like the very
environment of the reaction. when you scale it
up. Yeah, absolutely. Because of the way that
things are flowing and mixing. That's really
fascinating. Okay, so we've talked about heat
transfer, we've talked about mixing. Let's move
on to reaction kinetics, because I guess you
might think that if a reaction takes, let's say
an hour in the lab, it would take roughly an
hour in a larger reactor, maybe a little bit
longer. You'd think so, but it's not always that
simple. Okay, why is that? Well, the... The kinetics
of a reaction, meaning how fast it goes, can
be affected by all sorts of things. Temperature,
concentration, mixing. And all of those things
can be very different in a large reactor compared
to a small flask. Right. We've already talked
about how the temperature and mixing can be less
than ideal in a big reactor. Exactly. So I guess
that means that the reaction could actually slow
down quite a bit. It can, yeah. And sometimes
you have to adjust the reaction time accordingly.
So what might take an hour in the lab could take
several hours or even longer. Oh, yeah, absolutely.
Yeah, there was one example in OPR &D where they
were making a cyclic ester. OK. And the reaction
time was something like 20 hours. 20 hours. That's
a long time. It is, yeah. But that's what they
needed to get to get a good yield and purity.
So it's not just about making more of it. It's
about making sure that it's still the right stuff.
Exactly. Yeah, that's really important. All right,
so we've kind of laid the groundwork here. We've
talked about some of the general challenges of
scale up. Let's get into some of those specific
troubleshooting examples that we were talking
about earlier, those real world stories from
OPRND. OK, sounds good. Yeah, so let's start
with that nitro reduction that we were talking
about before. Right, so that was the work by
Garmesetti and Brayish. Yeah, and they were having
that problem with the catalyst getting coated.
Right. And as we discussed, the root cause of
that problem was the poor solubility of the starting
material and the product in the original solvent
system. Right, it was like coming out a solution
and gumming up the catalyst. Exactly. Yeah. So
how did they actually fix that? Well, their solution
was pretty clever. They decided to add some water
to the solvent mixture. They were originally
using just tetrahydrofuran, or THF, and they
found that by adding, I think it was about 10
% water, they were able to... significantly improved
the solubility of both the starting material
and the product. So it was like just enough water
to keep everything dissolved. Right, and that
prevented the catalyst from getting coated. Wow,
so just that simple change made all the difference.
It did, and it also allowed them to lower the
reaction temperature. Oh, really? Yeah. They
were originally running the reaction at 70 degrees
Celsius, and they were able to lower it to 50
degrees Celsius. That's a pretty big drop. It
is, yeah. And that actually helped to further
reduce the formation of impurities. So it was
like a win -win. Yeah, that's a really good example
of how sometimes the solution can be kind of
unexpected. Yeah. It's not always obvious what's
going to work. Absolutely. Yeah, that's really
cool. OK, what's another example of a real -world
problem that they tackled in OPR &A? Let's see.
Oh, there was a good one involving the quenching
of a reaction. OK, quenching. So that's when
you're basically stopping the reaction usually
by adding something to it. So in this case, they
were making a pretty complex molecule. chloropropoxybenzoyl
benzo -furanol isoindole -dione. Wow, that's
a mouthful. It is, yeah. Okay, so they were making
this complicated molecule and they had to quench
the reaction. Right. And what was the problem
there? Well, the problem was that when they scaled
up the reaction, the quenching process generated
a lot of heat. Ah, so it was like an exothermic
quenching. Exactly. And if they weren't careful,
the temperature could get too high, and they
could start to get decomposition of the product.
Oh, so it was like the reaction was still going,
even though they were trying to stop it? Kind
of, yeah. Okay, so how did they deal with that?
Well, they had to be very careful about how they
added the quenching region. They had to add it
slowly, and they had to keep the temperature
below 15 degrees Celsius. Wow, so they had to
really keep it cold? Yeah, it was tricky. Yeah,
and I imagine that's not easy to do when you're
dealing with large volumes. Oh no. It's not.
Yeah, that's really impressive that they were
able to figure that out. Yeah, they had to do
a lot of optimization to get the conditions just
right. Right, because you don't want to add it
too quickly and have it overheat, but you also
don't want to add it too slowly and have the
reaction keep going. Exactly. Yeah, so it's a
delicate balance. Okay, so that's another really
good example, and I think it kind of highlights
the fact that it's not just the reaction itself
that you have to worry about. Right. It's all
the steps around it. Yeah, the workup is just
as important as the reaction itself. Yeah, absolutely.
Now, for our last example, I wanted to touch
on something a little bit different. Okay. Which
is this idea of continuous flow reactors. Ah,
yes. And I know we've talked about this before
on the show. I think it was back in season six
when we did that. a deep dive on drug manufacturing.
But I thought it would be worth bringing up again
here because it kind of offers a different way
of thinking about scale up. Yeah, it's a very
different approach. So can you just remind us
what continuous flow is all about? Sure. So in
a traditional batch reactor, you put all your
reactants in, you let the reaction run, and then
you empty the reactor and start over. That's
like one big pot. Exactly. But in a continuous
flow reactor, you're constantly flowing the reactants
through the reactor. It's like a pipeline almost.
Yeah, kind of. And the reaction is happening
as the reactants are flowing through. Ah, so
it's not like a static thing. It's a dynamic
process. Exactly. And how does that help with
scale up? Well, for one thing, it allows you
to... to control the reaction conditions much
more precisely, because you're not dealing with
this huge volume of material all at once. Right,
it's kind of like you're breaking it down into
smaller chunks. Yeah, exactly. And that can help
to improve the heat transfer and the mixing.
Ah, so you're kind of getting around those problems
that we were talking about before. Right. That's
really interesting. And then to scale up, you
basically just make the flow rate bigger, right?
Well, you can do that, or you can. you can run
multiple reactors in parallel. Ah, so you're
kind of like numbering up instead of scaling
up. Okay, that's really clever. So continuous
flow is kind of like a whole different way of
approaching this problem. Yeah, it's a really
exciting technology. Yeah, and it's becoming
more and more common in the pharmaceutical industry,
right? It is, yeah. And I think it has the potential
to... To really revolutionize how we make chemicals.
Yeah, that's really cool Alright, so we've covered
a lot of ground today. Let's uh, let's take a
moment to just kind of summarize the key takeaways
here Okay, so we've seen that scaling up a chemical
reaction is a lot more complicated than just
you know multiplying everything by a factor of
ten or a hundred right there are all these These
new challenges you have to deal with. Yeah, absolutely.
And one of the biggest ones is heat transfer,
right? As a reactor gets bigger, it gets harder
and harder to get the heat in or out. Yeah, and
that can lead to all sorts of problems, like
temperature gradients and runaway reactions.
Exactly. And then there's mixing, which is also
really important. Yeah, you have to make sure
that the reactants are well mixed so that the
reaction can proceed uniformly. Great. Otherwise,
you might end up with some parts of the reactor
where the reaction is going really fast and other
parts where it's barely going at all. Exactly.
And then we talked about how the reaction kinetics
can change when you scale up. Right. The reaction
might take longer to go to completion. Yeah,
because of all those heat and mass transfer limitations.
Exactly. And then, of course, we looked at some
some real world examples from OPRD, where they
had to troubleshoot all sorts of problems like
that catalyst coating issue and that exothermic
quenching. Yeah, and those examples really highlight
the importance of careful planning and experimentation.
Right. You can't just assume that everything's
going to work the same way at a larger scale.
Absolutely not. Yeah, that's a really, really
good point. So I guess the overall message here
is that scale up is a It's a complex process.
It is. But it's also a really important one.
Because if we want to make these important chemicals,
like pharmaceuticals and other materials, we
have to be able to scale up the reactions. Absolutely.
So it's definitely worth the effort. Yeah, for
sure. All right. So I think we've covered a lot
of good stuff today. But I want to leave our
listeners with one final thought. OK. And that
is this. We've talked a lot about the technical
challenges of scale up. Right. But there are
also a lot of other factors that come into play
when you're trying to take a reaction from the
lab to commercial production. Like safety regulations,
environmental concerns, cost considerations,
all sorts of things. So I guess the question
for our listeners is what are some of those other
factors that you think are important to consider
when you're scaling up a chemical reaction? It's
a it's a really good question. Yeah, it's it's
not just about the chemistry. It's about the
whole The whole picture right the bigger picture.
It's a it's a very multifaceted problem. Absolutely
All right. Well, that's all the time we have
for today. Thanks so much for joining us on the
deep dive Thanks for having me and we'll see
you next time