[0:00]For two years, this drug was a miracle. It was introduced in 1996 to treat HIV, and by 1998, 75,000 patients across the country were taking up to 20 of them every day. It's called Ritonavir and it turned a certain death into a manageable condition. This particular pill is on its way to quality control, to a dissolution tester. Here, analysts monitor each batch of capsules, checking that they do dissolve in around 30 minutes, which is quickly enough to be absorbed properly. It's a rigorous precaution for a drug that for two years and 240 consecutive lots has never failed. But now, an analyst sees something unusual, this capsule hasn't dissolved properly. So, they follow protocol and trigger an emergency shutdown. They destroy the entire batch and deep clean the production line to eliminate any possible traces of contamination. But the next day, that quality control, the same thing happens. On the line, the clear capsules are turning white and cloudy. Technicians at the nearby Research and Development Lab study the paste under a microscope and find they're filled with millions of tiny needles. They're crystals, but no one has seen them before. They need a control to compare the needles against, so they make some of their own Retonavir in the lab. But to their horror, it also comes out cloudy. So they try again, but all attempts yield the same result: a white paste every time. The researchers are stumped. They had been making Ritonavir for two years. They knew its exact chemical composition and every part of the process used to make it. So they check all the input ingredients again, all the settings, every temperature setting and procedure, but all of it seems to be done correctly. Yet at the factory, the cloudy capsules are appearing more and more frequently. Within a week, every tablet produced by either the lab or the factory comes out cloudy. Abbott needs to halt all production of Retonavir immediately. But they can't just cut off the supply because people need these tablets. We called on as many resources as we could. We tried everything. We conducted countless experiments, we rebuilt facilities and new lines. We looked at alternative sites to see if we could start clean in a new environment. And they found an alternative site, a factory in Italy. They start Retonavir production there, and to their relief, all the pills pass the dissolution test. This is great news, but it also means that Chicago must have been making a mistake. So a team of scientists flies over to look at what the Italians are doing differently. They check everything: the pressure, temperature, humidity, the exact weight of all the chemicals. But it all matches perfectly with what they're doing in Chicago. None of it makes any sense, but at least Italy can keep making the medicine. But when the Chicago team returns home, they get a call. It's from Italy. Within days after their visit, one of the tablets fails the dissolution test. There was no gradual trend, there was no early warning. In a matter of weeks, maybe five or six weeks, every place the product was became contaminated with the crystals. We did not know how to detect it. We did not know how to test for it. We did not know what caused it. We did not know how to prevent it, we did not know how to get rid of it, and we kept asking the question, why now? They were witnessing a rare disaster. It had happened before and in theory could happen again to just about any drug or chemical compound. It spreads like a disease, but the thing that's getting infected is the medicine. One day you can make it, the next it's gone forever. It is frightening that this could happen to any drug that we take and on which we are dependent. And the scariest part is you can't predict if it will happen, when it will happen, or to which medicine or compound. Overnight, drugs we all rely on might just disappear. So what was happening inside those Retonavir capsules? What were those crystals inside? They appeared to be an entirely new compound, but when they tested them, everything indicated they were Ritonavir. It sounds impossible, but something similar had actually been the center of a heated debate 170 years earlier. In his Paris laboratory, chemist Justus von Liebig was reading a paper. It was about a newly discovered compound and what elements it was made of. This kind of work was at the cutting edge of chemical research, research he knew better than almost anyone, because he had personally pioneered most of it. This had made him highly respected in his field, but he also had a reputation for being difficult to work with. He was arrogant, hot-tempered, and didn't suffer fools. And the more he read this paper, the more incensed he got, because to him, it was clearly written by a fool. Friedrich Wöhler. So, we headed over to the lab at Imperial to recreate what Wöhler claimed to have discovered. So, I've got it here wrapped in foil, because it is a bit photosensitive. It's a bit like, like little rocks in there. Yeah. Beige powder. Okay, made of one silver, one nitrogen, one oxygen, and one carbon. Exactly. You want to light it up? Yeah, sure. Let's do it. You seem quite excited. I am quite excited.
[5:48]Yeah, not much is happening. Oh, it's melting a little or it's like It is getting a little discolored. Yes, it is.
[5:59]I guess the issue was you said, okay, I found this beige powder, and I know exactly what it's made of. One silver, one carbon, one nitrogen, and one oxygen. He publishes this. The paper reaches Liebig, and he's like, there's no way, because I've just discovered that compound. And when I tried to put a flame to it, it behaves completely differently. And we've got some of that right here, too. Should we try to burn this up? Yeah. I made some fresh this morning. Let's just see how a small amount behaves. So, we're going to go with, I don't know, maybe a few milligrams. I've left it a bit moist. Um, when it's in its moist state, Oh! Oh, my God, I didn't expect that. Man, so it's very sensitive. I'm sorry about that. That is crazy.
[6:47]Yeah, I wasn't ready for that. I wasn't ready for that either. I made it moist, so that it's less likely to self-deflagrate. But clearly, I was wrong. Clearly, Wöhler had made a mistake. These can't possibly be made of exactly the same elements. So, Liebig wrote a paper slamming Wöhler's work, calling him a hopeless analyst and saying he should go back, check his work, and publish again when he's found his mistake. And Wöhler does exactly that. He checks his work, but finds no mistakes, so now he's even more sure that he's correct. So he writes up his results in a second paper. But Liebig wasn't having any of this, and replies with another paper saying he must be wrong. So this public back and forth continues for two years, with each side becoming more and more convinced that the other is out of their mind. Until finally, they agree to meet on neutral ground in Frankfurt to put this whole thing to bed once and for all. They would replicate each other's work and let the results speak for themselves.
[7:51]But when they did, they were stunned. Before we continue, I want to quickly thank SoFi for sponsoring this part of the video. You know, one of the simplest ways to build wealth over time is by earning compound interest on your money. There's this quote from Albert Einstein that says, compound interest is the eighth wonder of the world. He who understands it, earns it. He who doesn't, pays it. And yet, many people don't put their money in a savings account. Which kind of makes sense because most banks give you next to nothing, a fraction of a percent worth of interest. And at that rate, you can't even keep up with inflation. But with today's sponsor, SoFi, you can change that. SoFi is an all-in-one finance app that lets you bank, borrow, and invest. When you open a SoFi checking and savings account and set up an eligible direct deposit, you earn a competitive APY. And you don't have to pay any account fees. That means your money isn't just sitting there, it's working. And every dollar you put in can earn more and more interest. And over time, that really adds up. Einstein called it the eighth wonder for a reason. Of course, the exact amount depends on several factors, like your starting capital, the time period, the interest rate, and any additional deposits. So, check if it's right for you. But right now, SoFi is also offering a special sign up bonus. If you set up a new high yield savings account with an eligible direct deposit of a thousand dollars or more, well then you can get either $50 or $400 cash bonus. So, start putting your money to work by heading over to sofi.com/ve. You can scan this QR code or click the link in the description. I want to thank SoFi for sponsoring this part of the video. And now back to what was happening with Liebig and Wöhler.
[9:44]They were both right. Yeah. Like they both had a compound that was made of exactly one carbon, one nitrogen, one oxygen, and one silver atom. And one could be boring as hell, and the other one can, well, blow up your face. Whoa! Wow! This was surprising because at the time a compound was thought to be just the atoms that made it up, and nothing more. But now this whole conception had to change. Von Liebig and Wöhler had discovered that the way those atoms are arranged also matters. At the time, they had no way to work out that ordering, but today, we can. When you shine light on a molecule, its electric field tugs on the electrons and nuclei. They get pulled back and forth as the field changes direction. This can stretch, squeeze, and bend the bonds in the molecule as the atoms oscillate back and forth. But each bond responds differently to the light, depending on how strong it is and the mass of the atoms it connects. It's like each bond is a boat on the ocean. If the waves are small and rapid, they won't rock it very much. And if the waves are very slow, like the tides coming in and out, that also won't rock the boat very much. The boat just gets lifted up and down. It's only when the waves are just the right size that the boat gets tossed around. Because of this, each bond will react strongest to a specific frequency of light, which we can measure. By hitting the molecule with a range of infrared frequencies, we get a spectrum like this, with peaks that tell us when bonds are reacting. This acts like a fingerprint for the molecule and tells us which bonds are there. Wöhler's compound has a broad peak, which corresponds to bending an N double bonded to C double bonded to O group. Liebig's compound on the other hand, has a spectrum that looks like this, with these two prominent peaks. One at high frequency and one at low frequency. These correspond to stretching a double bond between a carbon and a nitrogen, and a single bond between a nitrogen and an oxygen. We now know that Wöhler's compound was silver cyanate, and it looks like this. The carbon, nitrogen, and oxygen are joined with those two strong double bonds, which is why it's so stable. In contrast, Liebig's compound, silver fulminate, looks like this. The silver is bonded to the carbon instead, so the other elements are arranged this way. The bond between the carbon and nitrogen is a triple bond, but the oxygen and nitrogen are very weakly connected. This single bond is very easy to break, and once it does, the atoms can rearrange into much more stable gases, which is why it's so explosive. They had discovered isomers, that it's not just the atoms in a molecule that dictate how it behaves, but its bonds as well. So naturally, the scientists at Abbott suspected something similar might be happening to Retonavir. They knew that the spectrum of Retonavir should look like this. So they put a sample of white paste into a spectrometer, expecting to see something completely different. But instead, they saw this. The same peaks. The paste had all the same bonds as Retonavir, so it must be Retonavir. But they also noticed it wasn't exactly the same. There were these small deviations between the two. The arrangement of the atoms was the same, but something about the bonds had changed slightly. Well, it turns out there's another way to change the properties. And I can show you how with probably the most delicious demo I'll ever do. Because this, of course, is a piece of chocolate. It's nice, it's shiny, it's durable, and it has that nice snap when you crack it. But you'll notice that if you've ever let your chocolate melt, then it never returns to being quite the same. Suddenly, it melts in your hand when you pick it up, you know, it's dull, it's bendy, and it doesn't quite taste the same. You're not imagining this, there really is a subtle difference. And I can explain what's happening with a little help from my friend Chris over here. Hey, hello. So, Chris runs his own YouTube channel, Chris Young Cooks. And this is way overkill for what we need here probably, because he used to be the head development chef at a three-Michelin star restaurant. We've got some nice shiny chocolate here, but look what happens when we turn up the heat.
[14:17]Ah, yeah. That goes quite quick.
[14:23]Cool. That is surprisingly satisfying. We've obviously melted some of the chocolate, no surprise, but this is what happens, right? You leave it somewhere warm, the chocolate gets above body temperature, it starts to melt, and then as it cools back down, it's going to harden. So, got the chocolate. That looks like heat damage chocolate, right? I know. Like you you've seen this, you've opened a chocolate bar, maybe it was left in your car, sitting in a sunny window. Touch the edge, like you can feel. Feel how that just soft and kind of sticky. Immediately. Compare it to a nicely tempered piece of chocolate, like you can pick that up with your bare hands. It'll eventually melt in your hand, but much more slowly. Now, if this ever accidentally happens to you, we'll show you how to get it back to the nice and shiny form. But what's interesting here is that we didn't change any of the ingredients, and yet the properties changed completely. Chocolate is made of three main ingredients. There are other minor ones as well, but three main ones to focus on. It's got cocoa solids, that's what gives it its color. There's sugar, of course, for sweetness, and then there is cocoa butter, that's what gives it its texture. And this cocoa butter is the culprit. It's a fat made from three long carbon chains bonded together in the middle to make this sort of Y-shape. And that Y-shape can form together to form solids. But there are multiple ways they can stack together. There are many forms the crystal can take, each with different properties, and so we call these polymorphs. Chocolate actually has six polymorphs. The dull chocolate is mostly form IV, and that has a melting point of around 27 degrees Celsius. While the shiny chocolate, which is the one we want, is mostly form V. And that has a higher melting point of around 34 degrees Celsius. So the challenge and the art of chocolate making is managing these polymorphs to get the right form of crystal by managing both temperature and importantly, time. The nice thing about chocolate is you can start over. You just need to heat it back up to 45-50° C to wipe the memory of the wrong crystals. That's hot enough to melt out all of the crystals, but not too hot to start changing the flavor of the chocolate, evaporating a lot of the volatile aromatics. After around 10 minutes at roughly 50 degrees Celsius, all the crystals should have fully melted. So, at this point, we're trying to cool it back down to the temperature where crystals start to form again. Yep. And that's going to start at about 34 degrees Celsius. You'll start getting form five crystals forming at 34. As we cool even lower, we start to get form four and form three, those can all form at these temperatures. And that's okay, we want all of these crystals initially. We want. Oh, really? We we want to have a sort of shotgun of nucleation going on, because we want to make sure we get lots of everything. That's surprising. It does seem surprising. Because we just want form five, right? We do just want form five. The trick is if we just come down to the temperature where form five forms. So if we just went to like 32 degrees, 33 degrees and just waited there, you'd be waiting a very long time and you'd get a very random process of when does that crystal form? And maybe only a few crystals would form and so they would get very large. By bringing the temperature all the way down to 27, we get lots of nucleation really fast. The downside, of course, is we get the crystals we don't want as well. But we get lots of form five and we get lots of small form five. Yeah. So once we have that starting to form, we can select for the ones we want just by raising the temperature back out. And melting the form three and form four, leaving us with only form five, but importantly, lots of form five. Right. After holding the chocolate at around 32 degrees Celsius for five to 10 minutes, we can pour it into the mold. Okay. I think we're going to be okay here, so.
[18:10]Ah, oh, I don't know what I was expecting, but I was not expecting it to go like this. Really comes out as a sort of sheet. One of the things here is I do have some trapped air bubbles. Yep.
[18:27]Ah, yeah. It's like liquefied. Yeah. Yeah.
[18:35]So at this point, seems like we're done, right? Yeah. But actually now we need to lock in that crystal pattern to be created, right? Like as it cools down, there's liquid oil in there, and we're going to drop back down through the temperature where form four and form three can form. So what we need to do is we need to come down through that temperature relatively quickly so that we get mostly form five growing and lock them in by getting rid of most of the liquid oil. So we really need to get this down to about 12 C. So we put it in the fridge and waited for around 20 to 30 minutes. Get the door closed. Great.
[19:09]If we did this correctly, it should be mostly form five, which means all the molecules should have stacked tightly together, resulting in a shiny and snappy bar. Oh, that was satisfying. They're just barely hanging on. Wow. And uh, you can see they're nice. Perfect. Shiny. This is all form five. This is all form five, and we've got a nice shiny surface.
[19:38]Um, got a couple spots where maybe the molds could have been polished a little bit more, but give it a snap, just see how that is. Ooh, that's a very good snap. Yeah, that's a nice. It's very sturdy.
[19:48]Delicious. That's a good chocolate bar. This is how you can temper a chocolate bar. Amazing. But the stacking of the molecules in the crystals also changes something else. Since each molecule is surrounded by other molecules, it changes how the bonds inside can move. This is what the scientists at Abbott had seen in the spectrum. The needles they had seen under the microscope were a new polymorph of Retonavir, and a more stable one at that. Form one crystals look like this instead. Now at first, this might seem like good news. It was still Retonavir, even if it looked a bit different. It's just like how dull chocolate, even if it's not quite as nice, is still chocolate. But the problem was this new polymorph was far too stable. Retonavir form two is substantially more stable than form one. And the way we know it's more stable is because it's less soluble. But if that crystal structure happens to be much more stable, then it won't dissolve properly, and then it's a bit like you haven't taken the drug at all. But with chocolate, we can change which polymorph we have. We just have to heat it up to switch it from shiny to dull, and then by cooling it down again in a specific way, we could get back to shiny. So, you might expect that Abbott could just do something similar with Ritonavir, and they tried. But the problem was that no amount of heating or cooling could turn form two back into form one. They were stuck. We can see what's happening by taking a look at this here. See, each polymorph has different energy levels, and in the case of chocolate, that looks something like this, where form four has a higher energy level and form five a lower one. And they're separated with this sort of hill in between. Now, after heating up the chocolate bar, we were mostly left with form four. So, let's drop this little ball in there. Uh, and then you'll see it will slowly settle down into that valley. But not to the more stable form five, and that's because there's this little hill in between. But now imagine adding some heat to this. It's like giving the ball a little bit of a kick, and you can see that the ball will suddenly start to move around. And if I give it enough of a kick, it will roll down into form five. And now it is stuck there. Now, you could keep adding more heat, and you could get it back over the hill back to form four. But then you would just end up with a mass of both forms. Because whenever you start cooling it down again, you know, the ball could just randomly settle in one of the two valleys. So that's what happened when we melted the chocolate uncontrol a bit. You just get a mixture of these two forms. But with Ritonavir, the situation is a little different. The hill between the two forms is now much taller, but the form two valley is also much deeper. So, once the ball does get down there, it's basically impossible to get it back out of there. Which is why, no matter what the scientists at Abbott tried, they couldn't get back to form one. But this still doesn't explain why form two was suddenly everywhere. Nothing had changed in their procedures, the barrier between the two forms should still be there, so it shouldn't have been possible to make this much form two at all. And yet, 300 years earlier, legends of such a transformation spread across Northern Europe. It was a bitter winter morning, and it had been like this for months. The organist was on his way to a cathedral. The cold had been messing with the organ pipes, it's gone out of tune again. But that wasn't what the congregation thought. There were stories of other organs getting sick with warts or leprosy eating away at the pipes. Some thought it was the devil attacking the organ to punish an unfaithful flock. It was even said that when it was very quiet, you could hear these organs screaming and groaning in pain from the lesions. Nonsense, of course. It was just the metal contracting and expanding. Except these pipes weren't just contracting. They were cracked, and others are indeed covered in what looks like these lesions. Black growths all over the organ.
[24:17]Now, originally when this happened, people thought that this was the work of Satan. Of course, that's not what was going on, and we can explain what is actually happening. So, we've got some normal tin right here, which is what those organ pipes were made out of. It kind of looks silver, it feels pretty strong, and it's sort of the form we're used to. Exactly. But here we have a slightly different form of tin. You can look at it. It's a bit more gray, it's a bit more crumbly. And at room temperature, normally, the silver tin is sort of more stable. But if you cool tin down to something like below 13 degrees Celsius and ideally way colder, then it can transform into this new kind of gray tin. We're going to see what happens when we put it on top of the silver tin. We want to try and get it to around minus 30 degrees Celsius. Yep. And what better to get us to those temperatures is dry ice. So, dry ice is frozen carbon dioxide, and that is around minus 78 degrees Celsius. Now, we've taken a thermo flask and filled it up with dry ice, and then put a platform on top on which we'll put our tin. This should cool it down to around minus 30 degrees Celsius. Now, we left this here for around 14 hours, and what you'll see is that initially, there's a very tiny speck of tin that suddenly transformed into gray tin, and then it spreads from there. Almost like an infectious disease, which is why this is also known as tin pest. And because gray tin is less dense, the tin expands. And so, if you look closely, you can see it's started to tear apart the metal. Now, normally it takes a lot of energy to transform some silver tin into gray tin. But once you get a tiny bit of gray tin, something strange happens, because now it acts as this nucleation site that other tin can attach to, and it effectively brings that hill way down. It lowers the activation energy. And so now it becomes very easy to switch from silver tin to gray tin, and so it starts to spread, it starts to take over. And the same thing was happening to those organ pipes. Once you got a lesion on one of those pipes, well, then it would grow and spread everywhere. Little flakes would come off the pipes and seed all the others, and it would spread. And that's also exactly what happened with Retonavir. Once a tiny bit of form two appeared, it acted as a nucleation site, lowering that massive activation energy and causing all the form one to crystallize into form two. Tiny seed crystals then broke off, could become airborne and spread, attaching themselves to people's clothes and making it into other parts of the production line, effectively seeding them. So that when new Retonavir was synthesized, it contained these seed crystals and turned the entire capsules into form two. And because everyone likely had these seed crystals on their clothes, when the Chicago team flew over to Italy, they seeded that factory too. And in this way, soon not a single place was able to manufacture form one. Retonavir is arguably the most dramatic case of what we now call a disappearing polymorph. The YouTube channel Reactions made a great video about this that involves lots of physical demos, so I highly recommend you check it out. When this happened to us, we conducted an extremely thorough investigation to see if there was something that we did that could have caused this. While we have speculated on the cause of this chemical transformation, we don't have conclusive proof what happened. It might be that a mistake on the production line caused some chemicals to dry out. This might have created a new crystal, similar in shape to form two Retonavir, which acted like a seed. Or it might have just been a bad luck that a sea crystal formed on its own purely by chance. Even if you have a sea crystal if there's some particles or some scratches in the recipient, where actually crystals can start to nucleate, that can induce then different crystal structures. So it happens that in some pharmaceutical companies where they produce the same polymorph for years and years, that suddenly there is, I would say a hair or some other particle that kind of gets into the process and will change the entire crystallization of of the compound and that is then very difficult to control. And once a more stable form has appeared, it can spread and quickly seed the entire planet. It might be that you will never ever get the initial polymorph again. After five months of research, Abbott's researchers held a press conference to share their findings. Good afternoon. My colleagues and I are here today to explain what has happened, why it has happened, how we have responded to the problem, and what we are doing to correct the problem. Sometime during this summer, the semisolid formulation of Ritonavir began to change into a crystal form, a transformation that we believed was a scientific and chemical impossibility.
[29:39]You are a large, multinational company. Your scientists are obviously smart. How could this happen? A company's size and the collective IQs of their scientists have no relationship to this problem. This phenomenon is, I believe, unpredictable. We are, in some sense, the victim of bad luck. There are many mysteries of nature that we have not solved. Hurricanes, for example, continue to occur and often cause massive devastation. There is nothing that we can do today to prevent a hurricane from striking any community or polymorphysm from striking any drug. Science cannot provide a solution to all of our problems.
[32:43]The liquid formulation was not ideal. It had worse side effects and not all patients could tolerate it well, but it worked. It is frightening that this could happen to any drug that we take and on which we are dependent, even though it is not that common. This time it happened to Abbott and to the tens of thousands of people taking the semisolid capsule. Thankfully, we had the liquid formulation as a safety net. Next time it may happen to another drug that may not have a safety net. Now, here's a good question. Is everything polymorphic? So, nobody had discovered a polymorph of aspirin. Right, so it had been around as one of the earliest drugs. It had been crystallized in industry for, well, 130, 140 years. And so, can you say because nobody had discovered a polymorph of aspirin, therefore no. Right? The only problem is, uh, I discovered form two of aspirin. By accident, right? It turns out, over half of all compounds are known to be polymorphic, and there could be more. The number of polymorphs is proportional to the amount of time and money you spend researching that compound. In fact, nowadays, we know there are not two forms of Retonavir, but at least five. So are new cases of disappearing polymorphs something to worry about? It's quite, quite rare. We certainly know a lot more than we did when Retonavir occurred. But I wouldn't be surprised to see it happen again. If there's a 1% chance the world's going to end, you're going to do something about it. Right? If there's a 1% chance a plane is going to crash, you're not going to fly. Right? Uh, so, so yeah, so we we're at that situation where it might only be in that order of 1%, but if it happens, it's going to cost you a hell of a lot more than a few weeks of research on polymorphs. Ritonavir was, uh, one of the red flags that that caused a lot of regulatory activity and a lot of scientific activity around polymorphs.
