Organic Chemistry - NMR Problem

[Zipzap]

Hey guys, I'm having trouble determining the structure of a compound based on some NMR that I have been given. Here is what I've been given:

Molecular Formula:

C₈H₆O₄

 

¹H NMR:

1.57 (d, J=8.66 Hz, 1 H) 1.77 (d, J=8.66 Hz, 1 H) 3.51 (d, J=3.00 Hz, 1 H) 3.51 (d, J=3.00 Hz, 1 H) 3.60 (sc,J=7.56, 3.00 Hz, 1 H) 3.60 (sc, J=7.56, 3.00 Hz, 1 H) 6.31 (s, 2 H)

¹³C NMR:

46.10 (s, 1 C) 46.10 (s, 1 C) 47.20 (s, 1 C) 47.20 (s, 1 C) 52.80 (s, 1 C) 135.60 (s, 1 C) 135.60 (s, 1 C) 171.50 (s, 1 C) 171.50 (s, 1 C)

So, here's what I've been able to figure out thus far:

• I decided to first find the degree of unsaturation in this molecule, which comes out as 6.
• From that, my first thought was that I could get 4 degrees from a benzene ring, and 2 degrees from 2 carbonyls (somehow) attached to the ring.
• Based on the highest signal of the carbon NMR, I know that one of the carbons is going to be some sort of R(CO)X derivative, and thus I won't have any hydroxyls within my final structure.
• Given the highest signal on my proton NMR, I figured that there were only going to be 2 symmetrical hydrogen atoms (somewhere) on my ring. Either that, or the remote possibility of putting a CH2 right between two esters.
• My biggest problem comes from the fact that I have two proton signals in the 1.57-1.77 range, and the 3.51-3.60 range. This tells me that I have some carbons NOT involved in the ring, but at the same time I have carbons that are adjacent to either side of the ester groups.

I've tried designing a few structures, but they always keep falling short of either the molecular formula, the integration of each shift, or the group I tried adding doesn't match with the observed deshielding.

What are your thoughts, guys? All feedback is greatly appreciated!

[hypervalent_iodine]

Firstly, I've noticed you wrote in a few of your peaks twice? Was this supposed to mean something? Also curious as to why you've put 1 C next to your carbon peaks, given that you don't integrate carbon NMR.

 

 

I'm not convinced about there being a phenyl ring in your structure. Your carbon spectra doesn't appear to have enough peaks in the aromatic region and your proton NMR doesn't really have any. The closest is one at 6.31 ppm, which is a bit too low, isn't split as you'd expet for a phenyl ring and integrates to one proton. And really, with 8 carbons and 6 hydrogens, this shouldn't be surprising. After you include the ring, you have 2 carbons and maybe 1 hydrogen left to play with, which seems quite the impossible task with your oxygens and another 2 DBE's to incorporate.

[Zipzap]

Thanks for replying, hypervalent_iodine!

 

The peaks are put in there twice because I copied them straight out of the raw data that we were given. If it would help you more, I can post a picture of the actual NMRs we were given.

 

I had the same thought you did about the phenyl ring being unlikely, and your input made me realize that there's most likely not a phenyl ring.

However, I need to somehow create that deshielding at 6.31. My ideas have been to put a CH2 right next to two double bonds, or right next to two esters (I could play around with attaching it to the oxygen or the carbonyl itself in different combinations).

 

I guess what I'm having most trouble with is the individual shifts at the lower ppm's. They all only have one hydrogen, along with having doublets. It doesn't really help. =(

 

Did you have any particular hints that you could lend me so that I can figure it out for myself?

[hypervalent_iodine]

If you can post the spectra, that would be great.

As for the peak at 6.31, you have to consider what kind of bonding structure you might observe given that we aren't looking at an aromatic ring. What kind of bonds will you possibly see with all those DBE's? Where do protons involved in such bonds occur?

[Zipzap]

Here's the images of my spectra: http://imgur.com/m02Qr,yIDM0

 

Well, the only sort of bonds I could see with those DBEs would be a ring structure (not phenyl though), carbonyl, and a double bond. I was considering the possibility of a triple bond, but I don't think that the 13C NMR would agree with me.

 

All I know is that a CH2 HAS to be right beside two deshielding groups, otherwise I wouldn't get the signal at 6.31. Am I on the right track?

 

EDIT: After taking a second look at my proton NMR, what does 'sc' mean (it's assigned to 3.60)?

Edited September 9, 2012 by Zipzap

[hypervalent_iodine]

Right. Your data makes not much sense given your spectra. The peak at 6.51 and 6.30 look to me to part of one and the same doublet, so I am unsure as to why your data lists them separately. Even more confusing are the ranges they appear to have used to calculate coupling constants (the little blue lines above the peaks). I can't really check the calculations, because your spectra only goes to 2 decimal places and I'm not sure what power the machine they used is.

 

The peaks at 1.57 and 1.77 ppm are nice. You'll notice they appear to be pointing to each other. This would typically indicate what's called an AB system. I'm not sure if this is the case for your compound, but it may be something to consider.

 

The other beef I have with your spectra is the integration (red numbers underneath the peaks). They don't match up at all. Do you know why this is?

 

I don't know what 'sc' refers to, as I've never encountered it. If it were cs, you might assume it means complex singlet, but I really don't think that is the case as that particular peak looks more to be a doublet with the peak at 3.51 ppm.

 

Here's the images of my spectra: http://imgur.com/m02Qr,yIDM0

 

Well, the only sort of bonds I could see with those DBEs would be a ring structure (not phenyl though), carbonyl, and a double bond. I was considering the possibility of a triple bond, but I don't think that the 13C NMR would agree with me.

 

Neither your proton nor carbon spectra make sense for an alkyne.

 

All I know is that a CH2 HAS to be right beside two deshielding groups, otherwise I wouldn't get the signal at 6.31. Am I on the right track?

 

EDIT: After taking a second look at my proton NMR, what does 'sc' mean (it's assigned to 3.60)?

 

What makes you think it's a CH₂ group? I would be more inclined to go with 2 x vinylic CH groups. This makes sense with your carbon spectra, as you'll notice that there are (not including double ups) only 5 peaks. You have to have some equivalency somewhere. That being said, I'm still at a loss for where the remaining carbon(s) is. If we assume that you have 1 peak for both of the cabronyl carbons and 1 peak for 2 vinylic CH groups, that only gets you to 7 carbons, not 8. I'll have to sit down and think about it some more.

 

A few other pieces of information:

 

• If we assume that you have 2 x C=CH groups at 6.31 ppm, it is obvious that these two CH groups are not next door to any other protons (but they are next to each other). They have to be near quaternary carbons. This should allow you to work out the structure of a fair chunk of your molecule.
• You have no -CH₃ groups. This is important because it tells you a lot about the ends of the molecule. You either do not have terminating groups (i.e. all the atoms are involved in a ring of some sort) OR your terminating groups have double bonds at the end.
• You have no alkynes, so this means your DBE's are taken up by carbonyls, double bonds and possibly a ring.

[Zipzap]

UPDATE: Turns out that I was given the wrong molecular formula for that spectra, it should be C9H8O3. Would that make things easier? I'm now getting closer to the answer, but I'm still stuck.

[hypervalent_iodine]

Hah. Good, because I was close to having to reassess my career on account of not being able to figure your last one out. I'll have a look at it later today and get back to you.

[Zipzap]

So far, I have experimented with bicyclic compounds as a possibility for my NMR (i.e. two connected rings). I spoke with my professor today, and she said that while I haven't gotten it right yet, I'm definitely on the right track and getting closer to the answer. A helpful hint that I got was "consider the possibility that you could have more than two rings for your compound".

 

I'm currently working on tricyclic compounds and getting much closer to an answer, but I'm still struggling a bit. I'll wait until I see what you've posted, and that'll probably help make a more useful discussion.

 

I also wanted to take the time to say how much I appreciate your help with this, my brain is almost fried from trying to solve it!

[hypervalent_iodine]

Quick question, is it definitely only 3 oxygens? Also, what were the bicyclic structures you proposed?

[Zipzap]

Yeah, it's most definitely only 3 oxygens. I'll post up the bicyclic structures I tried in a few hours, it's midnight where I live. =P

[Zipzap]

So far, me and my fellow classmates have come to this structure:

http://www.sigmaaldrich.com/catalog/product/aldrich/247634?lang=en&region=CA

 

Based on having read its NMR, I can conclude that this is the structure I have. However, it's the explanation of the shifts that's really throwing me off.

[hypervalent_iodine]

Excellent, that was what I came up with as well. The NMR spectra on the sigma site matches it beautifully.

 

As for assigning your peaks.

 

 

¹H NMR

You should be able to fairly easily assign the peak at 6.31 ppm. There aren't many types of protons you would expect to occur in this region and of the ones you have (a methylene CH₂, 2 CH groups near a carbonyl, 2 vinylic CH groups and 2 CH groups at the base of the bridge head) only one of them could possibly occur in such a low field region. If you're having trouble, it might be more obvious if you have a look at a correlation table.

 

 

I mentioned in an earlier post that the peaks at 1.77 and 1.57 ppm were interesting in that they looked to be pointed at each other in a way typical of diastereotopic hydrogens or an AB system. Essentially, these types of systems occur when you have a CH₂ group wherein each hydrogen sits in a different electrostatic environment to the other. This seems counterintuitive because free rotation around neighbouring C-X bonds should mean that each H attached to the carbon atom spends its time in the same average environment as one another. However, in atoms where the C-X bond rotation is inhibited due to nearby steric bulk (phenylsuccinic acid is a good example of this) or fixation within a ring, the hydrogens on the CH₂ group will end up locked in a certain position. This in turn means that each proton is inequivalent in terms of its effective magnetic shielding and will therefore appear on an NMR spectrum as two distinct peaks rather than one. This is why, for instance, axial and equatorial protons on the same carbon atom have totally different chemical shifts. As well as having different shifts, these protons will also show a fairly large through-bond coupling to one another as well as coupling to neighbouring protons. Now, looking at your compound, you should see where this phenomena might apply.

 

 

Discerning between the remaining two peaks is a bit difficult. But, keep in mind (firstly) that they correspond to 2 x CH groups and not CH₂ groups and (secondly) that your bridge head hydrogens should couple to one of those peaks and that peak should (in theory) therefore have a coupling constant that matches with one of the coupling constants of the CH₂ groups.

 

 

Of course, the other way to do this is to do a literature search or use the predictor software on nmrdb.com to help you figure it out.

 

 

Do you also need help with the carbon spectrum, or was it just the proton one giving you trouble?

[Zipzap]

You are amazing, man. The analysis makes total sense!

 

Help with the carbon spectrum would be appreciated, and it seems like it would be a lot easier to come up with given that the peaks are simpler and more obvious.