Sciencemadness Discussion Board

Desymmetrization through chiral catalysts.

zeppelin69 - 2-11-2010 at 17:52

This summer I was fortunate enough to land an undergrad research position in an organic lab at the university I attend. My job up until now has been building up starting materials for a grad student who has been struggling with his project for a while, but recently my professor decided that my time would be better spent working on my own project. I gave him a few of my own ideas for what I could do, and he liked one of them so much that he told me to get started right away. I was pretty excited at first, but I am realizing now that I'm in over my head.

The basic idea is to make the Diels-Alder product of cyclopentadiene with maleic anhydride, then use a chiral catalyst to split open the maleic anhydride ring asymmetrically. The broken ends of the ring should have selectivity for which groups they get (one being an ester and the other remaining as the acid), so this way I would have control over the stereochemistry in the proceeding steps. Then I can modify each of these chains separately, eventually leading to intra molecular metathesis reactions and so on.

The problem is that I rushed into this project without doing enough research into these chiral catalyst. Does anyone on this board have any experiences with them? More specifically, I am looking at thiourea based catalyst like these.
http://en.wikipedia.org/wiki/Thiourea_organocatalysis
Would it be feasible to attempt to make my own catalyst? Any advice or insight would be greatly appreciated.


desymmerization.gif - 5kB

DDTea - 2-11-2010 at 21:04

You may be in over your head. I don't know about you, but I do my most transcendental and amazing work when I'm on the verge of a nervous breakdown. Don't ever be afraid to push yourself...hard.

Anyhow, I have no experience with these particular catalysts. What I would do is first of all, a lot of brainstorming. Go back to the fundamentals: how do catalysts work? By a) raising the free energy of the reactants or b) lowering the energy of the relevant transition state, leading to favorable yields of one product over another (via Curtin-Hammett principle) or increased reaction rate. In any case, I would spend a lot of time thinking about how your transition state will look (the one that connects your diels-alder product to your cleaved ring) and how you can stabilize that transition state with your thiourea catalysts.

I hope that helps a little, or at least gave you a warm-fuzzy feeling. Good luck!

Nicodem - 3-11-2010 at 11:05

The desymmetrization of symmetric succinic anhydrides should already be described in the literature, so the first thing you should do, is to do a thorough literature search and review.

I doubt chiral catalysts that interact with the substrate via H-bonds would do much in this case, but you can always try (but don't use solvents that compete as H-bond acceptors or donors!). What you need are catalysts that interact with the carbonyls and modify their electrophilicity. Chiral Lewis acids should do this best. There are many that are commercially available and others that can be easily made (like BINOL + AlCl3 or Ti(OiPr)4, etc.). Consider also that the solvent should not be competing with the catalyst (thus destabilizing the transition state) or else you will lose on the enantiomeric excess (if there is any at all), so you should not just use methanol as the solvent for the asymmetric methanolysis. Using an inert solvent like CH2Cl2 and only a slight excess of MeOH would probably give better results. In short, for a good e.e. you will need good reaction conditions optimizations, so you should not discard catalysts that initially give poor results without evaluating what causes good or poor results. Like for everything, you need to actually understand what is going on.

Methanolysis of anhydrides can be catalysed by tertiary amines and DMAP-type of bases as well, so you might also consider testing chiral bases like the ones of the cinchonidine-type. In any case, only those catalysts which form a diastereoisomeric transition state can induce an asymmetry in the reaction outcome, so you can predict which chiral reagents can not act as desymmetrization catalyst by considering the structure of the transition state for the catalysed methanolysis of your substrate, for which you need to know the exact mechanism of catalysis (which you can mostly find in the literature for the related achiral catalysts - luckily there are only about three mechanisms).

zeppelin69 - 6-11-2010 at 16:36

Thanks for the insight and advice. I was able to find a publication that was extremely relevant, and it's been helpful, but I am still lost on how I should go about obtaining the catalyst. It seems too me that they aren't being sold anywhere, but that's just as well because I doubt my research group could afford it and I like to make as many of my own reagents as I can anyway.

Just looking at catalyst 1 in the pdf, I wonder if I were to attempt to synthesize this myself, could I substitute the trifluoromethyl groups for something like nitro? I assume that the CF3's are being used for their electron withdrawing properties, and considering my lab's financial situation, it would be great if I could use a cheaper EWG.


Attachment: CC1.pdf (780kB)
This file has been downloaded 578 times


DDTea - 6-11-2010 at 17:47

Quote: Originally posted by zeppelin69  

Just looking at catalyst 1 in the pdf, I wonder if I were to attempt to synthesize this myself, could I substitute the trifluoromethyl groups for something like nitro? I assume that the CF3's are being used for their electron withdrawing properties, and considering my lab's financial situation, it would be great if I could use a cheaper EWG.


It's true that both nitro and trifluoromethyl are electron-withdrawing groups, but the mechanism by which they do that is very different. -NO<sub>2</sub> withdraws electron density primarily through resonance while -CF<sub>3</sub> withdraws electron density primarily through inductive effects.

zeppelin69 - 6-11-2010 at 21:48

I understand how they are different and why, but do the differences between the two matter for this purpose? The point of using the trifluoromethyl groups here is to pull electrons to one side of the molecule and create some polarity, is it not? If that is the case, then I shouldn't I be able to use any strong EWG? I could imagine there being a problem if nitro groups were significantly more bulky than CF3, creating some hinderence and not allowing the catalyst to "fit up" to the anhydride properly, but I imagine that they are relatively similar in size. Is there something else I am not seeing that wouldn't allow this to work? I should probably mention that I have not yet completed my second Ochem lecture course, so I still have a great deal of basics left to learn...

EWGs aside, for synthesizing the rest of the catalyst L-valine would probably be the most economical starting point, but I am at a loss for where to go from there. At some point, I would have to form a dimethylamine group in place of the carboxylate on the amino acid. This would be simple if I were using an acyl chloride. I would just form the amide and reduce to the amine, but as far as I know, there aren't any protecting groups that would allow me to halogenate an amino acid to some sort of amino acyl chloride for this to work. What I find the most baffiling at this point though, is how in the world do I attach the amino acid (or its derivitive) to an aromatic ring with thiourea inbetween :o. I can see how adding the thiourea group to my amino acid alone wouldn't be too difficult, but if I were to do that how can I attach the NH2 to the aromatic ring afterwards?

Thanks again for the input so far DDTea and Nicodem .



[Edited on 7-11-2010 by zeppelin69]

zeppelin69 - 11-11-2010 at 15:51

I could really use a little help with the synthesis of this catalyst. I'm not looking to be spoonfed, I'm just not familiar with the reactions used to make something like this. I'd really appreciate it if someone could point me in the right direction. Thanks again.

DDTea - 12-11-2010 at 01:25

Truthfully, zeppelin, I'm not familiar with this either! As my adviser, a theoretical chemist, once said: "If two things look different, they probably are." In regard to -CF<sub>3</sub> vs. -NO<sub>2</sub>, I genuinely do not know; I would have to thoroughly peruse the paper and do the same research you would. However, the best pointer that I can give you: don't take anything for granted. If something looks funny, investigate it. Don't make assumptions and if you do, at least know what assumptions you are making (after all, assumptions define the limits of every model).

As far as synthesizing the catalyst goes: could you use some derivative of thiocarbamic acid, with L-valine attached to the nitrogen, in your synthesis? Again, I'm taking a shot in the dark.