Herbert Brown (1998) - A General Asymmetric Synthesis Based on Chiral Organoboranes

I should clarify two things in this introduction. My parents came from the town Zhitomir in the Ukraine. It was not they who changed the name. But in those days when you came across immigration, when the immigration officer had difficulty spelling a name, he simplified it and so that’s how we got the name Brown. The first 4 letters are the same as Brovarnik. Anyhow, another statement I want to make, in 1936 I got a bachelor’s degree from the University of Chicago. It was at that time that my girlfriend, not yet my wife, gave me a graduation present, a book by Alfred Stock on the Hydrides of Boron and Silicon. At that time diborane was a chemical rarity available only in 2 laboratories in the entire world. And why did she pick this particular book to give to me? We were very poor in those days; it was the time of the great depression. And she got the cheapest chemistry book in the bookstore. (laughter). This is one way to find a rich new research field that will lead you to Nobel. As it happened I took my PhD in 1938, that’s 60 years ago. And I deliberately decided to speak on this topic because I feel it is inspirational to students. When I received my PhD degree, I thought that organic chemistry was a mature science and there was relatively little left to be discovered. And in all probability I would be spending my time working out reaction mechanisms, working to improve yields. But I was completely wrong and in the 60 years since then I’ve seen discovery after discovery come along. I have no evidence at all that things are slowing down. So I want to leave you with a message. I want to show you some of the things that we have done. Leave you with a message that there are still many new continents out there in science awaiting discovery by young enthusiastic explorers. Now as I told you diborane was a chemical rarity available only in 2 laboratories in the entire world. World War II research led to the practical synthetic methods for diborane and to the discovery of sodium borohydride. These turned out to be excellent reducing agents in organic chemistry. That’s the beginning of the use of hydrogen compounds, hydrides of boron for organic reductions. Exploration of their reducing action led us to discover the hydroboration reaction, the addition of HB bonds, diborane and similar compounds to carbon-carbon double bonds to give us organoboranes. Now, when I substituted... When I sent in a communication to the journal reporting we had made this discovery, the referee recommended that the journal not published this discovery. They said organoboranes were known for almost 100 years. And nobody had found anything useful that they would do that would be useful for organic chemistry. Since the main value of this reaction is to produce organoboranes, why publish it? I persuaded the editor that it’s true that no one had published anything useful to be done with organoboranes. But I pointed out that it’s not that anybody had tried and failed, there was no evidence that anyone ever tried. So I persuaded the editor, he published it and you’ll see what happened from then on. So this is a general asymmetric synthesis based on chiral organoboranes. We discovered hydroborations that made organoboranes readily available. These are some of the characteristics. We found in general in ether solvents diborane and similar BH compounds add with great ease to carbon-carbon double bonds to give us the organoboranes. If we took 2 butane and treated with diborane, 3 mol of olefin reactant and we end up with trisegment butyl boron. We oxidise it with hydrogen peroxide, we got secondary butyl alcohol, 3 mol. The boron tends to add to the terminal atom of a terminal olefin. So for the first time we could make primary alcohols readily from such olefins. We found that the addition to methyl cyclopentene and similar cyclic compounds goes with a cis addition of the H and B atoms. And then when we oxidised and went with retention so made pure trans-alcohol. One time it used to be several weeks work to make pure trans-alcohols of this kind. Now we could make them easily. Norborene with hydroboration gave us 99.6% exo. This is one of the things that got me started and my questioning of the non-classical structure because in the case of solvolysis of noroborene with tosylate it was the fact that you get exo compound, both from the endo and endo derivatives. That people propose as something new and different. But here is a molecule where there is no carbonium ion and where you're getting entirely exo. Then we took alpha-pinene, a very cheap olefin and very easily rearranged. And we questioned whether we wanted to try and see whether hydroboration would rearrange this compound. It didn’t rearrange it, it went simply and smoothly and we were able to get the corresponding alcohol. But one thing we found that was unexpected. Only 2 mol of alpha-pinene reacted making the diisopinocampheylborane. That gave us a new hydroborane agent, optically active. And that’s where my story of today, really starts from there. So we began to study the chemistry of organoboranes. And investigation revealed that organoboranes possess an exceptional personal chemistry for organic synthesis. Remember what the referee had said. Referees aren’t always right. You can see here we have 24 major reactions, new reactions. Each of these was a discovery in our laboratory and each of these was essentially published in a separate paper. Now we got a number of other reactions but we don’t consider them the major reactions. So this was something which the referee had said there was no future in organoborane chemistry. There was unexpected development. When we studied the substitution reactions of boron compounds, we find that the substitution groups usually precede with complete retention configuration. That’s different than substitution carbon compounds which usually go with inversion in configuration or racemisation. Now as an example, if we take and make the dimethyl derivative BH and add it to this, we get this. Treat it o-hydroxylamine sulphonic acid, we get the corresponding. The group goes with a pair of electrons from boron to nitrogen. And therefore you end up with the amine, pure transamine. And in general the reactions occur by primary coordination of the reagents with boron and then a rearrangement. So this is why we account for the retention and configuration. Now, if we could do asymmetric hydroboration we would have a general asymmetric synthesis. I had a student from the ETH who had come just at the time we discovered hydroboration, George Zweifel. He had worked there at sugar chemistry and had spent 3 years as a postdoctorate in England and came there. And I persuaded him to give up sugar chemistry and look at this new field of boron. And he was a godsend. Almost everything he tried worked like a charm. So this shows you that if we take... if we can make an optically active group attached to boron, we go through all these reactions and make optically active groups out of them. This is what George Zweifel did and I said to him: “Look we have an optically active hydroborane agent. Let’s take... We can’t hydroborate a third mole of alpha-pinene, it’s too hindered. Let’s take a less hindered olefin such as cis-2-butene, hydroborate it and see whether we can find any activity.” I was looking for the usual 10 to 20%, which people were getting in those days. And he ran the experiment, he came running into my office and says the compound shows an optically activity of 87% EE. Since the alpha pinene we started with was only 92% EE, we had achieved an almost 100% asymmetric synthesis. Now we improved the method and you see here in a number of cases we get close to... I used to say 100% but someone will always give me an argument. So there’s maybe a 100th of 1% there, we don’t see it in GC and so on. So it’s equal or greater than 99% EE. Now we couldn’t do it through trans or tri-substitute olefins. So we needed a compound that would have less ... (inaudible 10.50) than the di-compound. But when you try to hydroborate alpha-pinene, you go right past the mono to the di. So we had to go back, remove one group. If we put in a base like tetramethylethylene diamine, we remove the alpha-pinene, we get the monoisopinocampheylborane addition compound. This is crystalline, it septates right out. If we take this and treat it with BF3, the BF3 compound of this base also is inside the bond, crystallises out and you’re left with this reagent in ether solution. So we applied it to these compounds and you see we got 53, 62, 66, 72. But we found an interesting thing. When we carried out the reaction... Remember even if you get, let’s say, 60% EE, that means 80% is one isomer and 20% is the undesired isomer. If we allowed it to crystallise out, 1 isomer ... (inaudible 11.52) at 100% purity. And the other... The same solution didn’t do us any harm so that we had an easy way then to bring these up to 100%. So therefore, if we take this and hydroborate it to this and add to this, we have these 2 groups attached we don’t want. We found an easy way to remove them. If we treat them with acid aldehyde, they come off and give us alpha-pinene back again, which can be recycled, so that we now got boronic ester, optically active. We had to study the reactions of these. Previously we had been using R3B to get these 24 major reactions. We now had to learn how to do it with boronic esters. But we solved those problems and we could then make a series of boronic esters. And we could use them then to make all optically active compounds. Now, Don Matheson at Washington State University has come up with another approach to make optically active boron compounds, asymmetrical moligation. He took alpha-pinene and made a diol out of it, reacted that with boronic acid, made this derivative. He found if he added lithium CHCL2 at -100 degrees, he got the addition compound attached to boron, added a little zinc to help one chlorine ionise off and therefore he came to this compound, optically active. If you treat this with lithium alkyl or Grignard, you replace the chlorine by SN2 displacement reaction. And you end up with an optically active derivative, the kind we can’t get by hydroboration. So we have another approach to make these compounds. And here is an example where we have taken a cyclohexyl or a benzyl or the ci-tert-butyl derivatives we made the aldehyde, we made the carboxylic acid, we made the ketone and the amines and so on. So now we can have these compounds made by the Don Matheson procedure and we can apply up to 24 different reactions. Actually we’ve only applied about 7 or 8 because we got tired. Everything was working so there was no desire there. But let’s see what the scope is now. Examination of scope with boron based approach to the synthesis of pure enantiomer reveals that the number of pure enantiomers which can be readily synthesised by this approach is over 100,000. So we have taken and made in our own laboratory 34 different R star Bs. If we use minus alpha-pinene instead of plus alpha-pinene, you get a total of 64 starting materials. Now, we can also take and make the corresponding things through the homologation and this will... Let’s assume that you get the same number, for simplicity, you get the same number by the Matheson procedure. So you double the number 136. But then you can put on 1 methyl group or 2 methyl groups or 3 methyl groups and you get different compounds. Each of these is optically active and has a different structure. Or you can do it in one step by using Al-chloride and going in one step to the three. So now we have to add these to the list. If we add these to the list and now we got to multiply them by the 24 reactions. So we’ll have first 34, minus alpha-pinene, doubles that to 68, homologation gives you an equal number. Then adding 1 carbon atom, 2 carbon atoms, 3 to any one of these gives it new structures. And then, so you have a total of 544 at 24 major reactions, that’s 13,056. But each of these can make many compounds. For example, if you make a carboxylic acid you can make many esters. If you make the corresponding mean, you can make primary, secondary, tertiary means putting methyl, ethyl, isopropyl, tert-butyl, cyclohexyl with an optically active group. So let’s take roughly 10 per reaction, we will then get 130,000 optically pure compounds. Remember this is a rather new thing because all this optically active work was done actually since the Nobel. So that this represents a development after I retired in 1978 and was awarded the Nobel Prize the following year in 1979. Now, another approach is to do asymmetric reduction. Asymmetric reduction provides still with other boron based synthesis of pure enantiomers. One of my students, Mark Midland, took alpha-pinene; he added it to 9BBN and got this compound known as alpine borane. He found that if he reduces deuteral aldehydes he got 100% EE there. Or if he applied it to acetylene ketones, again he got close to 100% EE. But these were very fast reactions. This was much slower, acetophenone, and he got only 10% EE. The trouble is that if you got a cyclic mechanism you retain the optical activity and you get close to 100% EE. But if you have a slow reaction, another pathway takes place. There’s partial association of this into 9BBN of this. This reacts rapidly to give inactive products. And then you don’t get any. Fortunately we found a way around it. If we took alpha-pinene and made a diisopinocampheylborane and added ACL, we got diisopinocampheylborane, now sold commercially as DIP-chloride. And this then would give us... We took this and reacted with acetophenone. This is rather fast. And we got this intermediate. Alpha-pinene came off. And if we treated it with diaformin we could precipitate the IPCB boronic acid as the ester. And you got this and this was 98% EE. So we’re now getting up very high in the EE by reduction. So we applied a large number of compounds. You see this is the acetophenone, the ethyl derivative, the monopropyl, the isopropyl, all that reacted comparably. If you put in a tert-butyl group it goes the opposite way. Instead of giving you the S-isomer, you get the R-isomer. That’s because apparently phenol is larger than isopropyl but tert-butyl is larger than phenol so that changes the direction of the reduction curves. Now, many people have been working on producing reagents but usually they apply it to one type of ketone where it works well, ignore all the others. So you never really know how good the reagent is for all ketones. We suggested taking 10 different ketones and applying each new reagent to it so we can compare results. So here is with isopinocampheyl, methyl isopropyl ketone, no good. Here 98% good, here is good, good, good here, not particularly good here and so on. So we go through that way. Here I have listed the various reagents, alpine borane, Mark Midland’s reagent, our reagent, ... (inaudible 20.29) reagent, ... (inaudible), glucaride and by now NLH (inaudible) which Naiari had introduced. And you see that I’m putting a double plus with those which are best. Now, the way we analysed it was that when we go through the transition state, the ketone comes in and coordinates with the boron and you have 2 groups here, a large group and a small group. And that is affected by the methyl group and the alpha-pinene. And therefore usually a smaller group will be here and a larger group will be out here away and it will give you this. But if you try to make the other isomer, then the large group was being pressed against the methyl group and that’s undesirable. So we said: Why don’t we make that methyl group a larger group, make it into an ethyl? It’s very easy to metalate alpha-pinenes. You go to the ethyl derivative. And look what happens. When we apply this reagent, now go out up from 32% previously to 95%, equal or greater than 99, equal or greater than 99, equal or greater than 99 etc. So we now have the best one, 1, 2, 3, 4, 5, 6, 7 does with this epine borane chloride. Then we want to do an asymmetric allyl total boration. And this is another way. Organoboranes do not react like Grignard reagents, Mikhailov in Russian had made allylboron and found they added easily to aldehydes. Then Hoffmann at Marburg took and made this optically active derivative and put allyl in front of it and he got there reasonable results but not in a range that we today would like to have, from 36 to 86% EE. Well, it occurred to us that we ought to try our reagent. It’s good for hydroboration. Let’s put on an allyl group on it and try it and in our first effort we got 93% EE here. And we tried it and it seemed to be generally good. You see we could take metalate isobutylene, carry it out and we get this in 90% EE. Or metalate, the methyl allyl ether and we get this compound, go through it and you get the same compound you see with 2 asymmetric centres and again it’s 98% EE. Or we could take this allene and hydroborate it, we get the allyl derivative, carry it out. Now this is a natural product and we can in a one pot reaction go to this alpha product in 96% EE. And we can take the corresponding diene, cyclohexadiene, hydroborate it and if we keep the temperature below -25 there is no isomerisation of this. And add the aldehyde to it and we go directly to this compound, an excellent 97% EE. Now, this has been a favourite reaction of literature and I believe there are well over several hundred of the applications of this allyl total boration. For example here to make a total we take the 2 butane, we can metalate it. Again Schlosser had done this but he said that if you make the trans it isomerises as cis. But if we keep the temperature below -45 it metalates and does not rearrange, so that we can make the corresponding BIPC2, both the cis and the trans. And you see if we now use plus alpha-pinene or D-alpha-pinene or L-alpha pinene and the cis and trans we’ve got 4 reagents. Treat them with acid aldehyde and we get each of the 4 reagents, pure. Now we explored some possibilities for commercial applications. I mention this one, Prozac here and Ipsenol and Ipsdienol. So in this case, you see, we carried out, we could make this. I mention this because this compound was proposed as an antidepressant. And unfortunately it could not be... they couldn’t resolve it. Today we can make it 100% EE. Now Prozac is a very important reagent for anti-depressant brought out by Eli Lilly and they tried to resolve their compound, couldn’t do it. After 18 crystallisations they got it up to 80% EE and they gave up. Here we take this very simple process, take this, treat it with our reducing agent and we get this... (inaudible 26.03) reaction we don’t get an aversion, we get this. And then treat it with methylamine and we get this. And we get 100% EE. This is one that the Bristol-Myers Squibb organisation is making, an antipsychotic reagent. And again we can do it in a much simpler way. Finally I’d like to mention this one here. I think at one time in the Black Forrest in Germany they were having an attack by this insect which was eating up all the trees and they were looking for a way of making these pheromones which can control the insect. But it’s a very difficult compound to make. But we found an easy way. If we take a methylate isoprene put the potassium salt, put it on the methyl group and on to IPC2B there. And if you added this aldehyde, isobutyl aldehyde you get Ipsenol. And this aldehyde you get, now once you're up to 96% a simple crystallisation usually brings you up to 100%. Now many chiral auxiliaries have been examined and explored. This was Hoffmann’s reagents. and then Rausch(inaudible) proposed the titrate or this compound. And Corey came up with this. But alpha-pinene is a very cheap starting material and there’s practically no synthesis of reagents. And it provides a basis for optimism anticipating development, practical, economical, asymmetric synthesis. So here I’ve shown using the alpha-pinene, the things we can do with it. It can be for hydroboration, asymmetric hydroboration, asymmetric reduction, asymmetric propargylation, allylation, corellation. We can open epoxides and so on. When we take this, we’re getting a much greater improvement. This is only a start. For example, if we take the methyl-S-propyl ketone and use DIP-chloride we get only 32% EE. Use DIP-bromide 95%. If we take methyl-cyclohexyl ketone we get only 27% EE. Whereas with the... (inaudible) you get 97% EE. If we open up the epoxide with the IPC we get here 84% EE. Apply it to the cyclopentyl ether, only 48%. But with chloride gives us 99% for both. And finally this is the things we would like to do but I’m 86 now. I don’t think... There’s lots of things for the rest of you to do. So here shows what you ought to explore, what will do and there. Finally, I thought I would show you something that happened very recently. You all heard about the Nobel medal. Now the ACS has decided to have an AC Brown medal. So this is the AC Brown medal. And now the Nobel medal is 23 carat gold. The only thing I’ve ever seen that has 23 carat gold. This is only 14 but it’s still a nice medal to have. So with this then I will close my talk and I hope I haven’t run over time. But I wanted to give you a bird’s eye view of this thing so that young people you will be encouraged that there’s still a whole continent, many continents of knowledge out there waiting to be discovered by enthusiastic explorers. Thank you. Applause.

Herbert Brown (1998)

A General Asymmetric Synthesis Based on Chiral Organoboranes

Herbert Brown (1998)

A General Asymmetric Synthesis Based on Chiral Organoboranes

Comment

This is the last talk of H. C. Brown in Lindau, for which a recording is currently available. Brown lectured at Lindau Meetings during twenty years and while his topic, organoboron chemistry, remained a constant during all this time, its dimension expanded significantly. So did the technical possibilities: while the pictures and schemes supporting Browns first Lindau lecture were analogously projected by an aide, the present lecture relies on a computer based slideshow. Unfortunately, Brown’s slides are not available anymore and so, Brown’s 1980 and 1998 lectures have to serve as a case example of the detrimental effect of information technology on the structure of language. While the 1980 lecture is easy to follow, even without seeing the actual projections, the 1998 lecture is almost impossible to grasp in its entire detail.In any case, unless educated in organic chemistry, you will probably ask yourself what that mysterious “ee” is, that Brown repeatedly mentions. One thing that seems to be certain is that the higher the ee the better... and indeed: the ee refers to a special kind of selectivity of an organic reaction, the enantiomeric excess. A reaction with a high ee selectively yields one of two possible enantiomers and may hence be used for so-called asymmetric syntheses. A nice way to understand what enantiomers are is to think of a pair of gloves: they are made from the same materials, they weigh the same and they feel the same but yet they are different, as one fits the right hand and one the left. In other words: they are mirror images. Still, the differences are crucial: fighting cold hands with left hand gloves only would not solve the problem, no matter how many of them are available.What appears to be an idle thought in the field of gloves is of prime importance in pharmaceutical research. Much like gloves, certain drug molecules can only fulfil their purpose if their chirality “fits”, i.e. if they occur as the correct enantiomer. In other instances, they may do a lot of damage. Contergan, a drug given to pregnant women as treatment of morning sickness around 1960 is now well-known for causing severe birth defects. The active substance, thalidomide, occurs as two enantiomers, only one of which can sustain contergan’s disruptive effect on child development.Due to the significant influence of chirality on the efficiency of many bioactive substances, pharmaceutical companies nowadays are required to strictly control impurities by undesired enantiomers. This is why reactions with high ee’s are so desirable. And Brown and his team have developed quite a few of them, as he points out in his talk. An efficient synthetic route to the well-known antidepressant Prozac is only one of many significant results mentioned.The reactions discussed thereby all have one thing in common: the chemical element boron. Brown, who had worked with boron all his scientific life, systematically built up its application in synthetic organic chemistry - from a landmark synthesis of one of the simplest boron compounds, diborane (B2H6), published in 1944, to the asymmetric syntheses of complex pharmaceuticals discussed in this talk.Although it might appear that the end of the flagpole has been reached in this particular area, Brown liked to repeatedly point out in his talks that even supposedly well-researched fields offer a lot of room for surprises. In 1959, when he tried to publish the hydroboration reaction, which can be considered the basis for his share of the 1979 Nobel Prize in Chemistry, reviewers were not in favour, stating that boron compounds had been around for a hundred years already and that no significant effect on organic chemistry could be expected from them. Some 40 years later, Brown’s lecture is a late triumph over this scepticism.Another 12 years later, a former PostDoc in Brown’s lab, Akira Suzuki, should share the 2010 Nobel Prize in Chemistry for his work on organoboron-based, palladium-catalysed cross-couplings. A further success of boron, which Brown was not able to witness anymore: in 2004, at the age of 92, he passed away after a heart attack.David Siegel

Cite


Specify width: px

Share

COPYRIGHT

Cite


Specify width: px

Share

COPYRIGHT


Related Content