[flkpm]

I'm trying to protect an OH group with a silyl ether, protect the amine with BOC, deprotect the OH group, oxidize, and then deprotect the amine, however, I can't make it past BOC protection. The protection of the OH is quantitative and the RF moves completely from 0.2 to 0.5.  But when I perform BOC protection all I get is the same 0.5 no matter what I do. I have tried THF, DCM, DMF, and methanol and it doesn't matter. I have tried the catalysts of DMAP, TEA, and Amberlyst and it doesn't matter. I have tried RT up to heating to the boiling point of BOC and it doesn't matter (I realize boiling BOC is lethal). I have run it from 2 hours to 4 days and it doesn't matter. And I have tried different TLC solvent systems, spotting strengths, and even the "Z" test and it doesn't matter. No matter what I try I cannot get the starting RF to change. Now, when I make some of these attempts (particularly the lengthy procedures with heat) I do get a very very very faint 0.7 after letting the stain develop for a while but it is so faint and takes so long to develop I would say that there is almost nothing there even though this would confirm my assumption that the OH-protected/N-protected product would be higher on the plate than just the OH-protected product. But these BOC protection procedures are very straight forward and simple so I don't know why it would not take completely unless there is a chance that the RF would be EXACTLY the same under several different solvent systems. I think that this is far fetched but I just wanted to check with you guys if this is any way possible.

So could a N-BOC/TBDMS-OR molecule have the exact same RF value as just the TBDMS-OR molecule?

[Dan]

It's unlikely but not impossible, especially if the amine is an aniline (can you show the structure?). NMR will give you an answer in 5 minutes.

As a side note, you should be able to Boc the N in the presence of a free OH (i.e. TBS protection may not be necessary).

What is "the "Z" test"?

[flkpm]

It's unlikely but not impossible, especially if the amine is an aniline (can you show the structure?). NMR will give you an answer in 5 minutes.

Unfortunately I don't have access to an NMR.... I don't feel comfortable showing the structure.

As a side note, you should be able to Boc the N in the presence of a free OH (i.e. TBS protection may not be necessary).

I tried that several ways as well but I would end up with FOUR spots. And since the strongest spot was always what I started with and the weakest spot being what I suspected the N-protected compound was I figured that it wasn't even worth doing a column chromatography on. I assume the four spots were; the original compound (which I know for sure), OH protected with BOC (which lined up with the OH-TBDMS product), N-BOC product, and then N-BOC/OH-BOC product. Whatever all the spots were it was always a mess so I just gave up with it.

What is "the "Z" test"?

When you have two compounds that run very similar RF values and you can't tell if there is a difference then on a TLC you can run what I have heard referred to as the "Z test". That's when you draw a line across the plate and alternate spotting of the two different samples across the line. So sample A and sample B would be spotted "A B A B A B A B". Then you run the TLC and if there is any difference you should see a slight zig-zag or Z occur. You are probably familiar with it, I have just heard it called the Z test so that's what I called it. I did it after I did a column chromatography and thought I might had separated two different fractions.


So I guess if it shouldn't run at the same RF then the only thing I would be looking at would be really bad sterical hindrance??? This would seem odd though because I have found the very exact compound in literature going through these procedures no problem except that in literature they were using the dihydro form. But if it is the case then I suppose my only option would be maybe try to find a different protecting group in different conditions?

[Dan]

So sample A and sample B would be spotted "A B A B A B A B". Then you run the TLC and if there is any difference you should see a slight zig-zag or Z occur. You are probably familiar with it, I have just heard it called the Z test so that's what I called it.

Never heard of that one, thanks for the explanation.

Another thing to try is developing the plate with a vanillin stain, because it gives a range of different colours. I have had reactions with very close Rf values before, but the starting material and product stained different colours.

Can you get mass spec?
Without NMR, are you 100% sure the starting material (including the material before TBS protection) is what you think it is?

What TLC solvent combinations have you tried? I ask because sometimes people tell me they've tried "all the solvents" when what they mean is EtOAc/pet ether in 10 different ratios. I'd recommend trying these in various combinations:

Strong: Et₂O, EtOAc, acetone, (1:4 MeOH/EtOAc), EtOH, MeOH
Weak: pet ether, (1:1 DCM/pet ether), DCM, toluene, (1:1 toluene:pet ether)

So I guess if it shouldn't run at the same RF then the only thing I would be looking at would be really bad sterical hindrance??? This would seem odd though because I have found the very exact compound in literature going through these procedures no problem except that in literature they were using the dihydro form. But if it is the case then I suppose my only option would be maybe try to find a different protecting group in different conditions?

Hard to say without seeing a structure (I understand you can't show it). You could try a smaller protecting group, like TES, but it will limit how hard you can push the Boc protection conditions.

For some more information, you could try reacting the amine with an unhindered chloroformate, e.g. EtOCOCl, just to see if you get a reaction and Rf change. Run it in THF with triethylamine - triethylammonium chloride is insoluble in THF and the precipitate will give you visual confirmation of a reaction. The product will probably not be of use synthetically, but if it works you know you have an alternative option and can choose a carbamate protecting group that can be introduced with a chloroformate and is compatible with your synthesis.

[flkpm]

Thanks for your response.

There is actually a guy on Ebay now selling NMR's and MS's for $50 each so I suppose I could do that.

When I say I tried all solvents I mean it. I originally started with 9:1 chloroform:methanol with a dash of ethanol with anhydrous ammonia in it, but then I switch to various different combos of chloroform, benzene, ethyl acetate, toluene, DCM, ethanol, methanol, hexane, acetone, with the ammonia buffer TEA. I'd get the fraction to go from 0.1 to 0.9 and everywhere in between but never see any separation.

After I performed on a single batch 2 column chromatographies (first one was to get rid of the salts and some junk that was running all down the column), and then I performed several different washes and extractions with different solvents, and THEN after I let it sit in DMF/hexane for 4 days I was able to get some clear crystallization develop around what otherwise would be brown goo. I did a TLC on the crystals and for the first time ever not only did I get that 0.7 fraction stain nicely but it actually stained quicker and better than the normal 0.5 fraction (which was still present in the crystals). So I am going to go ahead and conclude that the N-BOC/TBDMS-OR is the 0.7 fraction and is in fact a different RF than just the TBDMS-OR.

This leaves me with the problem of unbelievably poor yield on a BOC protection procedure that should be simple and quantitative. I mean the crystals I got probably represent 500mg out of 15g. Not exactly the %99 yield I'd expect. Which leaves me wondering wtf. Steric hindrance I suppose would be the main suspect were it not for the fact that the N atom on the amine is sitting out there in the open all alone. It's not like it's hidden in between 3 different functional groups or anything. Considering that I have tried every different solvent that has solubility, and that I have done all procedures under excess time and heat, I suppose this means I have no choice left but to maybe try a different protecting group on that N even though BOC is suppose to be the easiest one (although adding hydroxylamine is suppose to make BOC take to amines 2.5 times faster so maybe I will try that).

[kriggy]

Why do you need to protect the amine for alcohol oxidation?