Richard Schrock (2009) - Recent Advances in Olefin Metathesis Catalyzed by Molybdenum and Tungsten Alkylidene Complexes

In the process of preparing M(NR)(CHCMe2R')(OR")2 (R' = Me or Ph) species (or analogs that contain enantiomerically pure biphenolate or binaphtholate ligands) in situ by treating M(NR)(CHCMe2R')(pyrrolide)2 species with alcohols, we have discovered M(NR)(CHCMe2R')(OR)(pyrrolide) (MonoAlkoxidePyrrolide or MAP) complexes (M = Mo or W)'

Thank you very much. Thanks to Countess Bettina for getting me here. I’m sorry I couldn’t be here all week. I was receiving an honorary degree in France. That’s what started the trip so I could not change that, so I apologize but I’m happy to be here now. I guess I’m the only representative for this particular area unlike the previous talks of representing the parties last year, the three talks, beautiful talks. I enjoyed them very much. I want to tell you a little bit of history, not very much but then primarily talk about what has happened in the last two years which has really changed things quite a bit for me and I think, for the next few years I hope, will continue to give rise to new results. Of course this is the reaction, the so-called metathesis dance, the Nobel Prize committee called it. It was actually discovered almost fifty years ago, more than fifty years ago now. Strangely enough at Du Pont where I was from 1972 to ’75 by a man named Eleuterio, Herb Eleuterio in 1956. He first recognized that there was something strange going on and that olefins were chopped up and you could redistribute the ends of them in this not very useful reaction here where two, where unsymmetric and two symmetric olefins are equilibrated. And then you go get both cis or trans olefins. So you get a mixture of six compounds and that’s not very useful but there are many applications that I’ll show you where it is extremely useful. Yves Chauvin as we just mentioned was one of the three recipients and he was the first in 1971 to write down the reaction that turned out to be the correct one. There were others that were written that were not correct as is often the case in science. And that said the simple reaction with a metal carbon double bond, between a metal carbon double bond and a carbon carbon double bond to give a ring. I mean this all seems obvious now but it was not obvious at that time. Of course olefins don’t add to each other to give carbon rings like that so the metal plays a special role here. And then you just break up the ring and you can carry out this scrambling reaction. So between about 1974 when I made a discovery at Du Pont in tantalum chemistry, the first tantalum carbon double bond of the type that might do this. In 1986, well defined catalysts that contained molybdenum and tungsten which was the, the molybdenum was, I think the metal that Eleuterio was using in his experiments, were prepared or designed. Because we really had no idea what was around the metal except for the metal carbon double bond. How to make them, could they be stable and so on. And this is the design that came out of that ten years of work. The imido, so called imido ligand here, of course has a metal nitrogen double bond but that doesn’t take part in any carbon carbon double bond reaction. Only the metal carbon double band takes part and then two alkoxides. All of these ligands have to be large to prevent bio molecular decomposition of this initial alkylidine, this neo pentamidine which is sterically quite crowded and slow to decompose. Decomposition in a bio molecular fashion, especially when this turns into a methylene, so methylene would kind of be the focus of this talk as we go through. Bob Grubbs is also mentioned and is here but is not giving a talk this year as I understand. But he developed ruthenium catalysts which are much more friendly for organic chemists. They’re relatively air moisture and functionality tolerant, unlike my catalysts that tend to react with air and water and active protons. And this is, I guess generation two, called generation two, Grubbs catalyst with an n-heterocyclic carbine which also doesn’t take part in the reaction and a phosphene coordinated to the ruthenium. If you remove that phosphene, this is the active portion which happens to again be a four co-ordinate species and also fourteen electron count as we count electrons in terms of inorganic and organometallic chemistry. The differences between ruthenium and molybdenum or tungsten are now becoming more and more obvious as you’ll see as I go through this talk. So what about the variations which are the interesting things? This is one of many variations. It’s a simple coupling of olefins and it’s good because the equilibrium can be avoided, you remove ethylene and so you can go all the way to the right and end up with an olefin having two R groups on it. Unfortunately, you get mostly the trans olefin and the cis olefin is the one that’s the most valuable. So the cis is Z and this is a problem that is now I think close to being solved as you’ll see by the end of this talk. Z selectivity, you want to make cis olefins. You don’t want to make trans olefins and especially you don’t want to make mixtures of the two. The cross coupling is more difficult, you have to selectively cause these, the RR prime olefin to form, not the R prime, R prime, the RR olefin and so on. So this is still a problem that I think we’re close to solving. This reaction hasn’t been around for a long time, especially with fairly strained olefins like norbornadiene, substituted norbornadienes or norborneens, cyclopentanes and so forth and that is to make polymers. Even before we had these very nice catalysts to play with change the functionality and change the role, people were making classic catalysts because these tend to be fairly reactive olefins and so you don’t need much catalyst, you don’t have to know what it is and so you can use a so called classic catalyst not a well-defined species to make polymers. So that’s good. Except that if you really want to make nice polymers you want to have one structure so you want it to be E or Z here between the two pieces or every piece that’s connected and then in some cases there’s tacticity which has to do with which way the rings point, if there are rings that remain and that’s an unsolved problem by and large. Ring closing metathesis was popularized by Bob Grubbs and Greg Fu when he was a post doc in Bob’s group in 1992 using first, one of the molybdenum catalysts I just described. Now I’ll spend the rest of the time on molybdenum and tell you some recent results. The bis alkoxide species although this isn’t really a recent result, part of history I guess. But bis alkoxides are believed to pass through and we have structures of tungsten species that certainly have these structures so that’s soon to be the intermediate as Chauvin proposed, a metallacyclobutane intermediate. And upon reaction of the initial alkylidine with some let’s say mono substituted olefin, in this case, you can make two of these. One with the R group next to the AT Butyl group and one with not and these break up and that gives rise to new alkylidines and then the reaction is off and running. All of these have, most of them, I shouldn’t say all have a trigonal bipyramidal shape, that will turn out to be interesting as we near the end of the talk with one equatorial alkoxide, one axial alkoxide and an axial imido group. Now the main problem as I mentioned is avoiding decomposition. You try to do that as much as possible by creating sterically bulky groups to prevent bio molecular combination especially of methylenes. Of course if you put too much direct bulk around the metal then the metal becomes un-reactive. You just can’t get any olefin in. This is I think where molybdenum and tungsten is really going to shine, are really going to shine in the future because they’re so reactive and so electrophyllic at dragging in the olefin. You can put a lot of steric bulk around them and the olefin wiggles through and you get a beautiful reaction. Generally when you try to substitute chloride, those two chlorides on ruthenium with other groups such as I’m talking about here, then the ruthenium becomes un-reactive for electronic reasons as well as steric reasons. In 1998 about ten years ago I started collaborating with Amir Hoveyda. He’s at Boston college and we’ve had a wonderful collaboration and due to these recent developments I think that will continue for another five or maybe even ten years. On asymmetric metathesis so if you choose now two alkoxides or link them, make them enantiomerically pure chiral bi-phenolates, bi-naphtholates, hydrogenated bi-naphtholates and so on, many, many compounds or ligands of this sort. You can make then enantiomerically pure catalyst, if you make enantiomerically pure catalyst you can generate enantiomerically pure products and of course that’s very nice. You can not only do all these reactions that are orthogonal to organic chemistry, these metathesis reactions, but you can make them enantiomerically pure, you hope. All of these have huge steric and electronic differences and so you can control the reaction by making and choosing the correct, bi-phenolate or bi-naphtholate along with the correct imido. There is a very large variety of imido groups, this is the initial one. There seems to be, in the case of the area of imidos, a need to have something attached, at least to one of the ortho position in order to go through the synthesis or the catalyst which I won’t talk about here. So they all have at least one substituent in the ortho position and then some aliphatic alkiel imido groups such as this anamantle species that turned out to be rather unique, probably for both steric and electronic reasons. Anamantle is actually what we call a small ligand or small, the substitute. The imido group then being a small ligand compared to this one for example. Again huge steric and electronic differences between all of these, in combination then with the bi-phenolate or bi-naphtholate. As an example of an asymmetric reaction I offer this, one reaction out of hundreds that we’ve published by Sarah Dolman who is actually an organic chemist in my group at MIT who collaborated then heavily with the Hoveyda group, especially Elizabeth Sattely in doing ring closing metatheses or desymmetrisations as we call them, where this carbon becomes connected to that one or that one so you get, you generate a chiral carbon here and in particular because nitrogen containing compounds are more interesting alkaloids and so on, I’ll mention a couple as I go along. And so that’s why we chose that project. Eight membered rings are very hard to make in general. This one is made in very high yield and it’s ninety seven percent EE. So these are very, relatively easy ways to make some pretty interesting and enantiomerically, almost pure compounds. So there are many variations and that’s the good news. The bad news is that there are many variations and it’s going to get worse as you’ll see, or better. I think better. And if you want to make all these catalysts, you have to make them and store them and use them one by one and that’s painful even for an organometallic chemist. So we wanted to make catalysts in situ some way. Find a way to take a generic pre cursor and add alkoxides, bi-phenolates, bi-naphtholates whatever, and make catalysts in situ and we settled on bispyrrolides which was not an obvious choice. Now a pyrrolide is of course the anine of pyrrole and pyrrole is an aromatic like benzene. And it’s a six bi-electron system and the pyrrolide is an analogous to C5H5 minus. So it’s C4 NH4 minus. So nitrogen analogue of cyclopentadieum. So they’re easy to make, that’s one reason why we chose them. There were no pyrrolides before we chose that pyrrolide. This is work by Adam Hock originally who’s in this fairly recent paper where Smaranda Marinescu, a Romanian graduate student is the first author. You can make a wide variety of pyrrolides here. There are the two five di-substituted pyrrolides. One is bound like a, A to 5 cyclopentadienyl through all five atoms here as shown and one is A to 1 bound. This is an eighteen electron species, should not be very reactive but it’s fluctional, that is the A to 5 goes to the A to 1 so you get an A to 1, A to 1 intermediate, that’s a fourteen electron species. That’s highly reactive fortunately. This is a crystal structure showing that A to 5, A to 1 arrangement. Only SYN, I didn’t talk about isomers in this particular game, but the SYN isomer is the isomer where this group, whatever it is on the alkylidine points towards the imido group, the one that points away is called the anti isomer and those two have very different re-activities and very different rates of inner converting with each other. So it’s an additional enormous complication and benefit, I think, in many ways once you learn how to control it in this reaction that I don’t have time to touch on. Well these bispyrrolides do react with a lot of phenols, diphenols and so forth to make the catalyst that we had made before. And the product in this reaction, is just a prothenatian reaction. The pyrrolide doesn’t protenate at nitrogen because that pyelectron pair is bound up in the six pyelectron system. It protenates actually at carbon. Either that carbon or that carbon to give what’s called a pyrrolinine, that pyrrolinine comes off and that then pyrrolinine then rearranges to pyrrole. Pyrroles then are very poor ligands. They are very poor. There is no way to bind the pyrrole to the metal, not through the nitrogen lone pair because that’s part of the six bi system. They actually bind pyrroles bind through carbon-carbon bonds to very electron rich metals. Not the kind of metals I’m talking about here. So these pyrroles are formed and they turn out to be not a problem. So we can prepare catalysts that can’t even be prepared the normal way. These reactions are fast and not always but often quantitative, pyrroles are very poor ligands as I just said. And we’ve shown that you can make catalysts that we have made and used isolated in situ and you get the same results as you do with isolated catalysts in all the reactions that we’ve tried. We didn’t try all the possible reactions but I think we tried enough to sort of convince ourselves that was true. So that was, that was the good news. We could make catalysts that we wanted to make. What we didn’t realize is that that wasn’t the best part of this story and this is now the new chemistry that I want to talk about. As you might expect if you use a mono alcohol, there’s no reason to do this unless the mono alcohol might be chiral as far as asymmetric synthesis is concerned and I’ll mention that. You’re going to get an intermediate probably. If you add one equivalent of alcohol, you’re going to protenate off one of these and then you’re going to be left with a monopyrrolide, monoalkoxide and then the next one will be protenated if you want to make the bisalkoxide, the catalyst is now the traditional catalyst. Well this was done by Rojendra Singh in my group, a Nepalese student. Probably the only Nepalese student that I will ever have. A very good student, excellent student, he realized, this was curious as to what the reactivity might be of this monoalkoxide, monopyrrolide and whether that pyrrolide was A to 1 as I’ve drawn there or A to 5 and so on. Well he showed that you could, by adding various alcohols here, tibutenal and so forth, exofloral, isopropanal, exofluortibutenol, you can make and isolate crystallize and characterize these monoalkoxides. Now initially I didn’t think this was going to be very interesting because we know that bispyrrolides basically don’t react with olefins. They’ve too much electron donation from the nitrogen. Of course this was an eighteen electron species so it has to form the A to 1 A to 1 intermediate, but basically these don’t react with olefins. These are beautiful compounds, tyrol or not and we thought these would be somewhere in between, in between zero and high reactivity. It turns out that they are probably on the average a hundred times more reactive than these are. So that was a big surprise, a huge surprise. It didn’t make much sense. There are some electronic reasons for that. There’s some theoretical calculations recently by Odile Eisenstein that rationalize why a particular molecule with four different groups and in particular what she called an acceptor and what we might think is a donor, along with the imido and alkylidine, are especially good for the metathesis reaction. So we call these MAP Monoalkoxide Pyrrolide catalysts. So this was reported first back in this paper in 2007. So that was the beginning of MAP catalyst chemistry. Now what is obvious if you look at this crystal structure. Two things, one is the pyrrolide is A to 1, not A to 5. So this is a fourteen electron species, should be reactive, no problem there. And of course there are four different groups bound to the metal. Okay that’s not very unusual and in organometallic chemistry, you often see four different groups bound to the metal. But this is a catalyst and that’s what makes it different from a lot of the other, which we all, other compounds where there are four different groups and you can therefore in this crystal structure, of course this is not a chiral space group. It has a mirror, a crystal graphic mirror in it and so it’s a mixture of enantiomers, that’s not new. But it is something to note. A third thing I should mention is that all of these are SYN isomers that we’ve seen. Bispyrrolides, monopyrrolides, everything that we’ve, crystallographically characterized, don’t have any anti isomers in them so it’s a simplification in a way. So there was a man named Henry Brunner who proposed that of all the asymmetric reactions in organometallic chemistry and catalyses which by and large now all use chiral ligands often with C2 symmetry, bidente, binap, very famous things, nuryore ligands and so on. They all basically use a chirality in a ligand. None of them with a few exceptions use any chirality at the metal and you thought chirality at the metal was the most powerful chirality to induce in some asymmetric reaction, chirality into the product. Because you have four different ways to approach this metal and so you approach trans to let’s say an alkylidine theory, trans to pyrrolide, trans to an imido group or trans to alkoxide and make some intermediate. We know that what we want to make is a metallacycle butane. And each of those approaches is dramatically different. So if you can make a pure enantiomer which is what Henry Brunner did and use that enantiomer in a catalytic reaction, it would be the most powerful way to make an asymmetric product. Well it turns out that nature as always has other things in mind and usually what happens is whenever you make a pure enantiomer and you try to do anything, you get the other enantiomer disasteromers. So you’re right back to the point where you have a racemic mixture and of course you can’t do anything in terms of asymmetric reactions with a racemic mixture. So you attach a chiral auxiliary. So this is not a pure demonstration of the power of chirality at the metal because you get disasteromers. So this is a fixed chirality. This actually is a half protected by tetraline so it’s a monophenol with a big group there. It sits there. And it has the R configuration. So each of these has the R configuration, then the metal has the S configuration here or the R configuration there. So you’re working with different compounds, this one and this one, and they’re not new images, they are disasteromers. You can make them from a bispyrrolide and isolate them or not, as you’ll see, you don’t have to. And you get a seven to one mixture in this particular case. The thermo dynamic mixture is two. The reason why it’s seven to one and you can actually isolate and crystalographically characterize each of these is that they’re configurationally stable. And that’s not a surprise. The ligands are covalently bound. No ligand can come off. There is no way except to squash the compounds and make square plainer species to inter convert the configuration of the metal. Carbons don’t do that and certainly the metal is not going to do that in this particular case. These metals are not prone to make square plain or compounds in their oxidation state. So they are configurationally stable in the absence of an olefin and I stress that point. Now there was a beautiful project in Amir Hoveyda’s lab and I should stress that the heavy lifting in organic chemistry is in Hoveyda’s lab, that simple stuff, we do in our lab. So things like this, this is Hoveyda chemistry. This is a synthesis of a natural product called quebrachamine, it’s a tetracyclic alkaloid. It’s reasonably complex. Maybe not by today’s standards but I thought it was a pretty, a nice goal. And so did Amir. And Elizabeth Sattely worked out a way to use metathesis to shorten the synthesis, the known synthesis of that molecule and to do it asymmetrically. And the key compound is this one here. So this is another sort of de-symmetrisation reaction where the metal attacks, you form the first metathesis product here at the more exposed allyl functionality and then closes on one of these phenol groups, this one or this one, a two step reaction is typical de-symmetrisation ring closing metathesis reaction. So that should if you do it asymmetrically then, in the right way you, you had won an enantimor, double bonds can be hydrogenated and that would give the product. Unfortunately this is a very difficult reaction. This is a preternary centre, it’s very crowded here. You can add the metal but then it doesn’t close and in fact even, the only reaction that she got to work from here to here wasn’t, an asymmetric reaction it was a symmetric reaction. With bis hexafluoro-t-butoxide. What people usually call the Schrock catalyst but that’s only one of thousands. And it’s a terrible, terrible reaction. Nobody would use it fifteen mol percent, twelve hours, seventy one percent yield okay but it’s very expensive in catalysts. Fifteen mol percent, nobody can do that. Or would do that. So if you just make the mono hexafluoro-t-butoxide, pyrrolide as I mentioned, you can make that, you find out that that species is, this is where the factor of a hundred comes from, is approximately a hundred times better. One mol percent instead of fifteen and one hour instead of twelve hours. You get seventy nine percent yield, greater than ninety eight percent conversion. But it’s not asymmetric so that’s not going to help you. Every asymmetric catalyst we tried gave zero product. Except these MAP catalysts. So we were extraordinarily lucky. There is something very interesting about these molecules and we don’t know all the secrets yet but we’re learning. So in this case the dibromo ligand that I mentioned, bromide in three and three prime positions, two mol percent, one hour, ninety eight percent conversion, ninety five percent EE and the dichloride version about the same, ninety six percent EE. So we reported this first in ‘Nature’ and then earlier this year in Jacks in another format, the complete paper. I don’t have a story about science like the earlier speakers do. But I do have a short story about ‘Nature’. It was a pain to get this published. It was because again referees, you know, you got to have referees and one referee said you’re not doing anything differently here. I mean you’ve got a chiral ligand on the metal, I mean so what, you’ve got a chiral centre at the metal, so what? You’re using disasteromers so what? Ninety six percent EE. That’s pretty good for a mixture of disasteromers, you expect one of them to give one result and the other one to get maybe the negative of that and so you wouldn’t get a very good result. So how does this work, how the hell can you have a mixture of disasteromers and get a ninety six percent EE? Well we wanted to know what the intermediates were in these reactions so we made some metallacyclobutanes. They had to be of tungsten, won’t go into the details but basically they’re trigonal bipyramids, here’s this big chiral ligand, yes at the bottom, the imido group at the top as I draw it. Here is the un-substituted metallacyclobutane and it’s fairly small by comparison to five dimetal pyrrolide in an equatorial position. Note that a loss of ethylene from one side or the other side would give the two configurations at the metal. And this is then our sort of working proposal as to how this system works. You can invert the configuration at the metal because the olefin attacks to give this sort of pseudo symmetric metallacycle from one side and then you generate ethylene in this case leaving from the other side, and that inverts the configuration at the metal from R to S. Well that’s not really the secret, because that’s going the wrong way. I mean that’s how can you use that in any way if you racemise at the metal. The configuration therefore inverts with each forward metathesis step. Only SYN alkylidines are involved as far as we can tell. So how does it work? Well one disasteromer simply doesn’t react for various reasons. We think we know, we’ve done some crystal structures and we think it’s at this one, the one that has the, the enantiomerically pure R ligand bound then to a metal that has the R configuration. That one towards substrate is un-reactive. Substrate reacts twice, that is, once you put the metal on and then once to make the ring. The other one surprisingly does and even more surprisingly these two rapidly inter convert in the presence of small olefins, especially ethylene which is a product in this ring closing reaction. So these two are in rapid equilibrium. One doesn’t react, this one does and each step that gives rise to this product is a two step reaction so you go from S to the R configuration and then you close the ring and you give the S configuration back. So retention overall for that particular reaction, so this is rather rare and in catalytic chemistry, I call it stereogenic metal control. The metal is not a pure stereogenic metal control because you do have a chiral auxiliary stuck to the metal but it’s relatively rare in transition metal catalysis, perhaps two or three examples in fact. So that, that’s what’s new, these reactions take advantage of asymmetry at the metal which you know, Henry Brunner was right I think. The four different ligands aren’t connected and they’re covalently bound. So there is no disassociation. There is no linking of ligands to keep them on the metal, you don’t have to worry about that. You want to have all geometries accessible in these five co-ordinate and metallacycles for olefins to leave and come back so that’s all pretty new. The most successful catalysts I think are relatively successful because methylenes are stable. I don’t have time to talk about it but it seems like for, for whatever reason, whether it’s because the metal is chiral or just for steric reasons, we’ve got all these bulky ligands on the metal, the methylenes are quite stable. You can heat some of them up to ninety degrees and observe no decomposition. So we don’t know why that’s true, but that’s obviously key. Methylene death is what is limited metathesis pretty much in this area for decades. Now asymmetric catalysts could be useful in symmetric reactions just because of their properties. So the, and of course that half protected bi-tetralin is actually accessible. You can buy the, the basic ligand and in two or three steps you can make that ligand so it’s not a hard thing to do. But what this did is it started giving us some ideas and that is how to solve this remaining problem. There may be a couple of other remaining problems but this is a huge problem in metathesis. You want to make it Z selective. So E or trans olefins aren’t nearly as useful in organic chemistry as Z, but the thermo dynamically most stable one is trans usually, about eighty percent in most cases and this is pretty much an equilibrium reaction so you end up making the trans olefin. You don’t want the trans olefin in for example this cross coupling reaction which is more difficult because you have to prevent formation of each, a homo couple product. So if you want to make this product, let’s say you have the red alkylidine here and you add the blue olefin you want to make this intermediate with these R groups pointing up and next to one another so you can give the Z alkine, the methylene which then reacts with the red olefin and goes back and makes this and generates ethylene. How do you do that? Well if this is the intermediate with this imido group pointed up and this great big ligand pointed down, perhaps you should just make a big ligand like that. We already have one so let’s go ahead and use it. Let’s make sure that it’s going to work though by making the imido group small. Now small as I said is adamantyl imido in our book. So adamantyl imido is the small group pointing up, the big phenoxide is this chiral one as it so happens. So no computers, no nothing, just sort of back of the envelope type calculations as to what might result in the formation of Z alkine exclusively and that is an intermediate in which the, our groups are next to one another and can only point up. So large small and it turns out that that works beautifully. We published this fact in, earlier this year. And the key is of course this cis or Z olefin formed in this ring opening cross metathesis. So not only is it Z selective, ninety eight percent Z, it’s ninety five percent EE which is not new to us because we made many compounds which have high EE and this ring opening cross reaction. And highly efficient, one percent of this catalyst, one hour, boom. And you can make this catalyst in situ. Why? Because the bispyrrolide which is the source of it and bisalkoxide which is very, a few little form because this phenol is so big, don’t react, compared to this very reactive species, for reasons we don’t completely understand. So this is the one that does the job even though it formed only sixty five percent yield as well so you can generate your chiral catalyst in situ and perform these very nice reactions enantioselective and Z selective. Now you don’t need an enantiomerically pure phenoxide as I mentioned, so we thought well let’s look in the literature and see what other big phenoxides we can find or big phenols and here’s one made by Phil Power. It’s a two six di-substituted phenol and in the two and six positions there is a two four six tri isopropanol. So this is perpendicular to that ring, we have another one over here and so you have a big bowl. And if you imagine that in an axio position of a metallacyclobutane intermediate, you can see that there’s, I don’t have time to show structures. We have done some structures. There is no way for anything to point down towards the phenoxide. So you can make the species, you can do some varied, now these are obviously not heavy lifting reactions. We do these in our laboratory. You can do a very simple reaction between a three hexene and cis four octene. And generate this equilibrium mixture which is not a useful reaction because it’s kind of a mess except that it’s all Z. So you go from cis to cis and back and forth. So you get a one to two to one mixture of these three products, not six products. It takes a while and that’s because if you want to, you know making trans olefins is fast, relatively fast. So if you want to make cis olefins with the two substituents next to one another, that’s going to be a slow reaction. So that’s why it takes eight hours at room temperature. But it’s totally cis selective. There is some isomerisation in three days so perhaps we have to beef this up a little bit but we’re still working on that. Now what, how do you really take advantage of chirality at the metal without an auxiliary alkoxide or phenoxide. And that is to do ROMP reactions and so this is the typical substituted norbonadyine and this double bond is open. Now remember the mechanism is the olefin adds trans to the pyrrolide. So this comes in on this side, the other constituents point up. The ring opens, you get this ring pointing out. But you’ve changed the configuration so the next one comes in from this side and that gives the ring pointed in the other direction so you get all cis and this is a syndiotactic arrangement and unknown structure for polymer chemistry. Now in polymer chemistry you want to make one structure. So we were very excited to find that this prediction actually turned out to be true. You can make this brand new structure using these new catalysts and there is no chiral group here except the metal itself, as a mixture of enantiomers. That each step then generates this structure, involves this syndiotactic, this was reported a couple of issues ago. The last thing I want to mention is other directions, one of them being ethenolysis so running coupling reactions in the reverse way. You can’t do that very well with ruthenium because ruthenium is not stable to ethylene. These molecules are very stable at ethylene and very reactive. The metallacycles will break up very readily and what you would like to do is, you would like to take a natural product like methaloliate and add ethylene to it and chop it up to give one decenylene and methionine-decanoate, you can put these in polyethylene and make fancy polyethylenes. You can make pheromones with this piece. And so on, detergents maybe with this piece. We can do that now with our molybdenum catalysts, five thousand turnovers at room temperature, just by pressurizing up metaloliate with ethylene, in atmospheres, because it’s not very soluble in metaloliate. Ninety nine percent selectivity, that mean you only get this product and that product. Ninety five percent yield so that’s a beautiful beginning for green chemistry you might say in terms of using natural products as sources of useful chemicals. So catalyst engineering is really the answer. As you can see, but you have thousands of variations possible and sometimes you make a very small change, it makes a very big difference so we have a lot to do here to understand all about this chemistry that we have to and of course in order to obtain the desired result. So that’s where we’re headed, more Z selective catalysts. This is a key issue here, people do not like to use molybdenum because it’s air sensitive for one. It would be nice to support all of these catalysts that we’ve developed in homogenous reactions so that the function is retained. That is important. You can’t mess around with the ligand too much. You want to maintain the function that you’ve designed so beautifully and then of course in this challenge for convinced users that air sensitive catalysts are not bad but good because of what they do. If you’re going to make a billion dollar drug it doesn’t matter if it’s an air sensitive catalyst, I mean that’s, that’s humorous. It doesn’t matter. So we hope that we’ll be able to really do some unique things here next couple of years. So everybody has support that does work, we don’t do the work as was pointed out. So I’ve been supported by the NSF and the NIH, I want to acknowledge certainly Amir Hoveyda and his group at Boston College. It’s been a great collaboration in the last ten years and I’m sure it’s going to continue. Here’s some former group members that I mentioned and some that are presently in my group working on this new area. So thank you very much for the opportunity to be here.

Richard Schrock (2009)

Recent Advances in Olefin Metathesis Catalyzed by Molybdenum and Tungsten Alkylidene Complexes

Richard Schrock (2009)

Recent Advances in Olefin Metathesis Catalyzed by Molybdenum and Tungsten Alkylidene Complexes

Abstract

In the process of preparing M(NR)(CHCMe2R')(OR")2 (R' = Me or Ph) species (or analogs that contain enantiomerically pure biphenolate or binaphtholate ligands) in situ by treating M(NR)(CHCMe2R')(pyrrolide)2 species with alcohols, we have discovered M(NR)(CHCMe2R')(OR)(pyrrolide) (MonoAlkoxidePyrrolide or MAP) complexes (M = Mo or W)'. MAP complexes are chiral at the metal by virtue of having four different covalently bound ligands. When the alcohol is enantiomerically pure then M(NR)(CHCMe2R')(OR*)(pyrrolide) diastereomers are formed. MAP species are remarkably active olefin metathesis catalysts (as much as 100x more active than the best bisalkoxide or diolate catalysts for a given transformation) and highly enantioselective. Catalyst loadings are typically <1%. A dramatic example of the efficacy of these new catalysts is Mo(NAr)(CHCMe2Ph)(Me2Pyr)(1) (where Ar = 2,6-diisopropylphenyl, Me2Pyr = 2,5-dimethylpyrrolide, and 1 is derived from (R)-1H where R = Si(t-Bu)Me2), which serves as an initiator for an asymmetric ring-closing of an intermediate in the enantioselective synthesis of the Aspidosperma alkaloid, quebrachamine, a reaction that yielded no product when asymmetric Mo(NR)(CHR')(diolate) catalysts were employed². Mo(NAr)(CHCMe2Ph)(Me2Pyr)(1) can be prepared in situ and is effective (95% ee) at relatively low loadings as a 7:1 mixture of diastereomers. New findings include (inter alia) the use of MAP complexes for Z- and enantioselective ring-opening/cross metathesis reactions,³ for the ROMP synthesis of polymers with new structures, and for Z-selective metatheses of internal olefins.

___________

'Schrock, R. R. Chem. Rev. 2009, 109, online 3/13.
²Malcolmson, S. J.; Meek, S. J.; Sattely, E. S.; Schrock, R. R.; Hoveyda, A. H. Nature 2008, 456, 933-937.
³Ibrahem, I; Yu, M.; Schrock, R. R.; Hoveyda, A. H. J. Am. Chem. Soc. 2009, 131, 3844-3845.

Cite


Specify width: px

Share

COPYRIGHT

Cite


Specify width: px

Share

COPYRIGHT


Related Content