Molecule Synthesis

[ekay]

I'm looking to synthesize the product here using the starting material (glucose?) shown below. I'm able to use any reagent but I just can't figure out the steps to do so. Should I open the ring first? Substitute the substituents or . . I'm really stumped here. Any help is appreciated!

 

[ekay]

Anyone?

[mississippichem]

I'm looking to synthesize the product here using the starting material (glucose?) shown below. I'm able to use any reagent but I just can't figure out the steps to do so. Should I open the ring first? Substitute the substituents or . . I'm really stumped here. Any help is appreciated!

 

 

Well since you don't have any stereochemistry shown , your substrate is a generic hexopyranose. I would protect the hydroxyl groups first. If you protect the hydroxyl group after opening the ring, you'll end up protecting the hydroxyl that you need to reform the cyclic hemiacetal.

[ekay]

That makes sense. I'm going to use MOM to protect the hydroxyl groups first then try to open the ring. Am I now able to protonate the oxygen in the ring and remove it as water?

[Mr Skeptic]

Looks like you have at least three different things to do, and the order to do them would be important. Methylating the C1 hydroxyl is used to prevent rings from changing the stereochemistry (alpha <--> beta). But what about removing that one specific hydroxyl on C6? Can you make it different than the others, say by doing something to C5?

[Horza2002]

Tbh, making it from glucose would be pretty difficult and also probably very ineffcient. If I where you, I'd try to come up with another way to make it.

[hypervalent_iodine]

I agree with Horza, to an extent. Having worked with sugars extensively in the past, I can promise you that the reaction you're doing is best not done with glucose. Asides from the fact that sugars are a pain to work with - you can't heat reactions above 35 degrees celsius, protection and deprotection steps are a pain (low yields, shifting protecting groups etc) and commercially, not viable (if you have to do a lot of them) - your end product is a 5 membered sugar. Why are you starting with glucose, when arabinose and ribose are readily available?

 

Regardless, your ring opening reaction (using glucose) will be difficult and you cannot simply 'protect the C6 hydroxyl once it's open", as glucose will spontaneously ring close when in solution. Your product has one less carbon in the main ring (I'm not including the ether linked methyl groups in that count), so instead of protecting it, you need to change it completely and shorten the chain. You can do this by treatment with NH2OH to generate an oxime, then generate the cyanohydrin with acetic anhydride and sodium acetate. From there it's simply a case of treating it with sodium methoxide in methanol to expel the cyanide and generate a mixture of ribose and arabinose. This will then cyclise to give the corresponding furanose. You can work from there pretty easily.

 

Also, protonating the oxygen to remove as water isn't the most efficient way to go. You're best of activating it with Tosyl or something like that.

 

As I say, long winded and painful. As well as that, reactions with cyanide leaving groups in them are not ideal due to toxicity issues. You'd be much better off actually starting with arabinose or ribose than you are glucose.

[mississippichem]

 

Also, protonating the oxygen to remove as water isn't the most efficient way to go. You're best of activating it with Tosyl or something like that.

 

Wow, do you work in organic synthesis? I do coordination chem, so I'm easily tripped up by these carbohydrate reactions . I was wondering if there is a way to selectively activate the 6-hydroxyl to a tosyl? Will tosyl chloride sterically prefer that position?

[hypervalent_iodine]

I certainly do

 

I haven't looked much in to adding a Tosyl group to C6, but I am almost sure that if you did it very slowly and very carefully that you could selectively add to the C6 without disrupting any of the other exposed hydroxyls. And yes, you would start with a Tosyl chloride. The other problem you have with this is that you cannot simply 'eliminate' the hydroxy this way. Doing so would require temperatures in excess of 200 degrees C and you would end up with an alkene, which then needs further reduction. I would try use something like the Barton-McCombie deoxygenation. Traditionally it doesn't use a Tosyl group on the OH, but instead converts it to a thiocarbonyl. I think using tosyl should be fine though. The next step uses AIBN and Bu3SnH, which I would be concerned about using with around all the other OH groups exposed, so I would go about protecting them after you've added the tosyl. Probably as benzyl ethers, since you'll have to remove them before you shorten the chain. You can use Bn-Cl for the protecting group reactions, but be careful to check with NMR etc that you have fully protected every hydroxyl group. A lot of the time when you're simultaneously protecting a bunch of different hydroxyls, they tend not to all go on and you have to repeat the reaction.

 

Again, I would really not use glucose for this. Asides from all the horribly long and painful synthetic steps, you will end up with two different products. Are you able to use ribose or arabinose at all?

[Horza2002]

Sterically, the 6 position would be favoured for the tosylation because its a primary alcohol where as all the others are secondary. However in practise, you'd probably end up with a mixture of tosylated products. The problem here is that primary and secondary hydroxyl groups reactivity is still very similar so getting seletvity would be a problem.

 

Seeing as no stereochemistry is shown, you could oxidise all of the hydroyl group to the ketones (for the secondary) and carboxylic acid (for the primary). Then convert the acid to a methyl ester and then reduce all the ketones back to hydroxyl groups with sodium borohydride. This is not exactly a great way to do it as you would lose all stereocontrol of the reaction not to mention the potential for transesterification reactions leading to polymers.

 

You might be able to selectively protect the secondary alcohol here because there are essentially two sets of 1,2-diols. With this, you could potentially make two acetonides with them. This would then allow you to use a different protecting group for the primary alcohol (say benzyl). You could then remove the acetonide and you'd have selectively protected the primary. However, again this route is not exactly efficient at making sure you doubley protect the two 1,2-diols.

[hypervalent_iodine]

That's not really true, Horza. I have often protected sugars in the C6 position with all the other groups unexposed. It works fine if you do it properly - or it did for what I was working with. Tosyl may be a problem, but it then again, it may not.

 

And you don't really seem to have taken into account that there is one less carbon in the product? Oxidising all those hydoxys to ketones would be slow and painful and probably not work. Also, what do you intend to oxidise with - remembering that the aldehyde functionality can also oxidise quite easily to the acid. Whether or not you can make two acetonides depends on stereochemistry as well. You'd be better off going with Bn or Bz groups. It does depend on how you remove the OH in the end product due to different sensitivities of the various PG's. Oh, and though it is slow, you may also end up reducing the methyl esters, especially since you would have to use so much to reduce all those ketones you somehow plan on making.

[Horza2002]

I've just had an idea how to make it from Maleic anyhdride (easily and cheaply avaliable).

 

The first step involves the di-hydroxylation of the alkene with osmium tetroxide (giving a racemic mixture)

Melthylation of these two hydroxyl groups should be easy enough by deprotonating (NaH) them and then treating them with methyl iodide.

Conversion of one of the carbonyls to an alkene could then be done using one equivalent of Tebbe olefination (uses the metal carbene generated from titanocene dichloride and trimethyl aluminium).

Reduction of this alkene with hydrogen gas on palladium should be easy enough.

Reduction of the carbonyl (I can't remember the conditions that I've used to do this before) and finally methylating again like for the diol.

 

Again though, as I don't know what stereochemistry you want, this route might not work.

idea.doc

[hypervalent_iodine]

Or you could just start with ribose/arabinose, etc, eliminate the primary hydroxyl, then methylate.

 

EDIT: probably better methylate and then eliminate actually, which would mean an extra protection/deprotection step for the primary OH.

Edited January 12, 2011 by hypervalent_iodine

[Horza2002]

I agree, methylate firt would be so you don't have to worry about accidentally eliminating any of the other hydroyl groups.

[hypervalent_iodine]

that won't be too hard. Protecting groups coupled with the added reactivity of the primary spot will make it a fairly simple process.

 

You know, the funny thing is that ekay is probably not even going to read this. I have enjoyed the exercise though.

[Horza2002]

You said you had some experince with this sort of chemistry? Is there a major difference betweent he reactivity of the primary compared to the secondary alcohol. I always thought that they would be too similar to get them to react differently; I tend to avoid them in my reaction routes. Is it a case of carefully adding 1 equivalent and temperature control to selectively get the primary portection.

 

Yer, I doubt ekay will read this, but its good to have a synthetic discussion...there tend not to be too many around here!

[hypervalent_iodine]

One of the reactions I did was with a glucosamine hydrochloric acid salt, which was to be the second monomer unit of a tri-beta-1,6-N-acetyl-D-glucosamine. I was in the process of synthesising a potential vaccine for a particular biofilm forming bacteria and so I needed the synthesis to be really short for it to work commercially. The people I spoke to about trying to cut down my protecting group steps said that I could protect my C-6 with a TBDPS group, in the presence of the amine and 3 hydroxyls, without any trouble. I did pretty much what you said - I added it (under nitrogen) drop wise at room temp over about 10 or so minutes for 4 mL of TBDPS-Cl and it worked beautifully. I didn't get to analyse the yields until 2 products later because I carried my products straight through, but after 3 steps I had almost 70% yield and the TBDPS didn't appear to have gone anywhere else I didn't want it to. The main difference in reactivity I think is the steric affect. You could also argue that, depending on reaction conditions, the stability of their respective deprotonated forms may also have an affect on reactivity - i.e. a secondary or tertiary centre would get a heavier pull of electrons from all the extra alkyl groups than would a primary hydroxy, thus making the primary anion more stable. In my experience, I have found that you can often use the difference in reactivity to your advantage, if you do it carefully enough.

 

And yes, there are so very few synthetic chemistry posts on here, which is unfortunate. In any case, it's always nice to talk to an organic chemist who understands the jargon!

[mississippichem]

That's not really true, Horza. I have often protected sugars in the C6 position with all the other groups unexposed. It works fine if you do it properly - or it did for what I was working with. Tosyl may be a problem, but it then again, it may not.

 

Have you used any other reagents to selectively protect a primary hydroxyl ( selectively against a secondary)? benzyl, dimethyl silyl, O-t-butyl? I was wondering because this is quite interesting, taking advantage of a small steric and electronic difference to get good selectivity.

 

I'm not an organic chemist but I end up making organo-ligands all the time for my metal complexes. I'm really interested if this selective protection can be done with organo-silyl protecting groups. I like those groups because they are easily hydrolyzed off which is crucial when I have a metal center with a pH sensitive redox equilibrium, usually vanadium (III), ruthenium (III, IV), or osmium (II, III).

 

And yes, there are so very few synthetic chemistry posts on here, which is unfortunate. In any case, it's always nice to talk to an organic chemist who understands the jargon!

 

Yes, good chemistry talk is hard to come by here. The best I get is usually with Horza here and confused freshmen biology majors who need help with general chem. I jump at the chance to talk synthesis or p-chem (which is my true love) with pros. This is the part of the thread where the chemists come out.

Edited January 12, 2011 by mississippichem

[Horza2002]

I would have guessed that the sterics were the major issue controlling where the protection occured. Espcially using TBDPS groups which are sterically large as well! 70% over three steps is pretty good as well! I've also found that to given my limited experience (only done 4 months of my PhD so far) that slight difference in reactiity can be useful for selectivity..although so far I don't get great selectivity in some of my reactions.

 

Mississippichem, when you remove the silyl ether protecting groups, do you use acidic conditions to do it? If you have some pH sensitive metals, then you can also use flouride to remove the silyl group. Adding NaF is often enough to remove them, but there are some better flouride releasing groups as well (e.g. TBAF). The flouirde adds to the silicon (forming an extremely strong Si-F bond) followed by elimination of the alcohol. That might be useful for you if using some pH sensitive metals...not sure how adding a fluoride donor would affect metals though...

[mississippichem]

I would have guessed that the sterics were the major issue controlling where the protection occured. Espcially using TBDPS groups which are sterically large as well! 70% over three steps is pretty good as well! I've also found that to given my limited experience (only done 4 months of my PhD so far) that slight difference in reactiity can be useful for selectivity..although so far I don't get great selectivity in some of my reactions.

 

Mississippichem, when you remove the silyl ether protecting groups, do you use acidic conditions to do it? If you have some pH sensitive metals, then you can also use flouride to remove the silyl group. Adding NaF is often enough to remove them, but there are some better flouride releasing groups as well (e.g. TBAF). The flouirde adds to the silicon (forming an extremely strong Si-F bond) followed by elimination of the alcohol. That might be useful for you if using some pH sensitive metals...not sure how adding a fluoride donor would affect metals though...

 

I usually use a neutral fluoride donor, like NaF. The fluoride anion is good because it is largely non-coordinating to d-centers (bad orbital overlap with d-hybrids, not willing to back donate an electron pair). There are some steps though where I need to leave one deprotonated hydroxide (secondary) and protect a primary hydroxide so it can last through a substitution I'm doing on the other end of the (very large) mixed metal macrostructure.

 

And yes, I would guess that steric factors do influence that protection's selectivity more so than anion stability. Especially for bulky attackers like TBDPS.

 

*My ligands are fairly labile in polar solvents, so I'm worried that a pre-protecting group that could attack my coordinated oxygen could separate my ligand from the center, and cause a drastic geometry shift! Coordination chemistry with mixed metal complexes is very unforgiving. The metal centers can redox each other, and some geometry shifts caused by equilibrium shifts of labile ligands end up being practically irreversible (though not in theory irreversible, you're going to have to start over).[Mississippichem pulls his hair and sighs]

Edited January 12, 2011 by mississippichem

[hypervalent_iodine]

Have you used any other reagents to selectively protect a primary hydroxyl ( selectively against a secondary)? benzyl, dimethyl silyl, O-t-butyl? I was wondering because this is quite interesting, taking advantage of a small steric and electronic difference to get good selectivity.

 

I'm not an organic chemist but I end up making organo-ligands all the time for my metal complexes. I'm really interested if this selective protection can be done with organo-silyl protecting groups. I like those groups because they are easily hydrolyzed off which is crucial when I have a metal center with a pH sensitive redox equilibrium, usually vanadium (III), ruthenium (III, IV), or osmium (II, III).

 

 

 

Yes, good chemistry talk is hard to come by here. The best I get is usually with Horza here and confused freshmen biology majors who need help with general chem. I jump at the chance to talk synthesis or p-chem (which is my true love) with pros. This is the part of the thread where the chemists come out.

 

I keep forgetting to reply to this post.

 

Yes, you can do it with TBDMS as well as TBDPS. I have not tried reacting benzyl or t-butyl, but from memory it can be done (I'd have to check my PG book). I tend to like silyl protecting groups on C6 anyway - they stay put and do what they're told more often than not, which is great. There is nothing worse than a migrating protecting group that thinks it can go where ever it damn well pleases after I've gone to all the effort of attaching it - not normally too much of a problem with C6, but it does happen. Also, selective deprotection is nice and easy and doesn't require anything harsh, which would be great for what you're talking about. I haven't had to work with transition metal chemistry since undergrad. I have to say, I'm happy for it. It was never a strong suit for me.