When we first started in flow chemistry about 10 years ago, a lot of people thought this is not going to go anywhere, because they thought, well, you're going to get a lot of clogging, lots of problems. You have to build these instruments from scratch.

And now if you look at it, you can buy commercial instruments to perform reactions that are too dangerous otherwise, whereas Caleb is difficult. Or they have very, very precious intermediates. So I think what you're going to show people today is that this now also works in photochemistry. So good luck with that.

See you. Welcome to our lab here in Berlin. I'm going to tell you today about our recent publication, which concerns continuous flow photochemistry. And we've used a continuous flow photochemical reactor to prepare a number of 3H-azepanones, which have

some pharmaceutical relevance. And we've done that using this system here, which consists of a pumping module, which is commercially available, and this reactor, which we build ourselves. Now, this consists of FEP tubing, which is transparent to UV light,

wrapped around a medium pressure mercury lamp. And this is cooled by this cryostat here. And we'll tell you more about this later. But what I should really start with is to tell you why it's worth running a reaction or a photochemical reaction in continuous flow compared to running it in an immersion well in batch.

So I can do that by showing you some diagrams over here. So in a traditional batch reactor setup, we have an immersion well like the one I've shown here. And this consists of a lamp, which is immersed inside the reaction mixture.

Now, we know from the Beer-Lambert law that the measured intensity of light decreases exponentially as we move away from the light source. And that means in a batch reactor case, all the reactivity is happening very close to the lamp. Now, we can overcome that by working in the space very close to the lamp. And indeed, that would require a very small reactor.

And we wouldn't be able to get very much done in that. So we can overcome this by having either a falling film of our reaction mixture being past the light source, or we can approximate that with a channel like the one I showed you over there, where we have tubing wrapped around the light source. There's a second benefit here, and that's

that we're continuously removing the products of the reactions that are formed. And this can help to prevent them undergoing secondary photochemical reactions. And that's something we'll talk to you about today. The paper we're presenting today is an example of this, where we've been able to minimize the formation of a byproduct in a photochemical reaction by running it

in a system of this type. So I'll pass you over now to my colleague Farhan, who will tell you in-depth about the photochemistry we've been performing. Now, I'd like to move to the board to give you more details and better understanding of the reaction that we're performing in the paper. As my colleague Alex mentioned, we're

performing the photolysis of aryl azides and the subsequent formation, a rearrangement into 3-H AZP nodes. And I'll draw the structure of the starting material and the product on the board and explain to you the importance of this reaction.

Upon photolysis with UV light, aromatic azides decompose into singlet nitrenes with concomitant release of nitrogen gas. Singlet nitrenes in return undergo a rearrangement and during expansion to give the dihydroazepine rings, which probably arise from the ring expansion

of these intermediary 2-H azurines. These dihydroazepines could be trapped by different nucleophiles, which could be either water, alcohol, or an amine, which in this case, if it was a protein, it's very important reaction that is used in protein photo and radio labeling.

And the reason why we were interested in the 3-H AZP nodes is that there has been a class of drugs that has been used as inhibitors for Alzheimer's disease. And now I would like to give the camera to my friend Francois Levesque, who is going to introduce the experimental details of this paper. Now please follow me. I will show you how we actually did the experiment.

So as you do when you bake a cake, you first need to pre-eat the oven. We need to pre-eat the lamp. So first of all, we need to start the cooling system. So we just started the cooling system, which is connected to the cooling jacket right here.

Just remember, the lamp is inside the cooling jacket, which is surrounded by a Pyrex filter, which cuts out the light below 300 nanometer. Another important part of the cooling system are those two fans right here. And to start them, you just need to plug them.

So now we need to close the box before turning on the lamp. So this is for safety issue because the lamp is really powerful. And then we will switch on the lamp by turning it on the power supply, just like that. So now the lamp is on. And we need now to prime the system.

The lamp will take about 30 minutes to warm up. So meanwhile, we will prime the system. So the way to do that is we start the pumps. Then we prime with this valve. We open it. Then you pull a few milliliters of solvent just to make sure that all the bubbles are removed

from the pump and from the line. Then you can close it. In that case, there's two lines. So we need to switch to the second line. And so we use this. Then we need to prime the second pump. So we open the valve.

Again, we need to pull some solvent. And finally, we close the valve. So now the system is primed and ready to go. OK, so now we will need to change the condition of the pumping from the one that

were used for the priming to the one that will be used for the reaction. So to do so, we use this computer. So we change the water from 2 mil per minute to 0.2 and the THF to 0.267, which will give us a residence time of 30 minutes.

So now we need to change the line from the THF to the region. So we just pull it out and put it inside like that.

Now we will need to wait a little bit to see that the system is stabilizing. Now we will be ready to pump the region into the flow reactor. To do so, we will need to change the pumping

from this line to this line. To do so, we will activate this valve, which is controlled by this bottom. So pushing here will start the reaction. We're using two solutions, and we need to mix them at one point. So we are using a T-mixer to do so. As you can see, there's two lines, one for the water,

one for the THF. Both of them are combined and are pushed into the reactor from to this side.

Well, we can see the reaction is working very nicely as nitrogen is developing. And this is just one reaction where flow chemistry really shows a lot of promise. Whereas dangerous reactions, reaction would have to be scaled up. But also, reactions that are dangerous to use

can be perfectly used at large scale in these flow reactors. I don't think it's saying too much, but there will be a lot of developments happening very soon, and you'll hear about that from us.