[OrganicH2O]

I am questioning the supposedly given correct answer for an H-NMR question. I am also wondering if this question is reasonable to solve in some way without a chemical shift additivity calculation, or perhaps even better, 2-D NMR. Supposedly, the correct answer can be reached without knowledge of either of these things (only 1-D NMR was provided, and the student has not been taught about chemical shift additivity).

I attached a picture of two different constitutional isomers that are consistent with all of the splitting and integration, and most of the chemical shift data. The question lies with the connectivity of the propyl group and methyl group. The propyl group could be bonded to the aromatic ring and the methyl could be bonded to the ketone carbonyl, or vice versa.

I did a quick additivity calculation and it appears the aryl substituent will have a higher PPM. This is consistent with the data, but I am wondering how we are supposed to know this without precise knowledge of chemical shift values, given that the two PPM's are so close.

[OrganicH2O]

Integration, splitting, and the number of unique hydrogens. I used the NMR data to construct those two possibilities. I think both possibilities are consistent with integration/splitting/number of signals. But the only way I can see to differentiate them is based on the (very slightly) different chemical shift values for the indicated protons in the picture that I made.

In the picture I made, the two carbons are highlighted in red in both possible structures.

[Babcock_Hall]

Upon further consideration, this problem is more difficult than I thought it was.

[Babcock_Hall]

I sometimes use model compounds when I assign the shifts in an unknown.  The actual CH₃ is at 2.4 ppm, and the actual CH₂ is at 2.6.  In the case of ethyl benzene, the -CH₂-Ph group is 2.62 ppm.  The methyl group of methyl benzyl ketone is 2.02 ppm, and the methyl group of methyl ethyl ketone is 2.13 ppm.  I would say that the agreement between the models and the correct structure (2) regarding the CH₂ (methylene) group is very good, but the agreement regarding the methyl group is no better than fair.  If we now look at the incorrect structure (1), we can use toluene (2.32) or ortho-xylene (2.22) as models for methyl groups on an aromatic ring and dipropyl ketone (2.38 ppm for the CH₂C(O) group) as the other model, we obtain adequate agreement between the data and the proposed structure.  We could also use ortho-ethyltoluene as a model compound for both methyl and methylene groups at the benzylic position.

Based on the above exercise, I would be hesitant to say that one structure clearly outperformed the other one in terms of agreement with the data.  Perhaps the models I chose do not fully account for the effect of branching near the ketone, although the relevant methyl group in 3-methyl-2-butanone is only about 0.1 ppm more downfield than other methyl ketones I have seen.  Is there any chance that someone has mistranscribed the actual NMR shifts?

[OrganicH2O]

Thank you for the analysis. We were not even given the ppm numbers, or even a very precise scale, so I am only estimating the ppm values. Given that, it's probably not worth it to try to justify anything too precisely. Your logic makes sense.

The best way to solve this problem was with faulty backwards "mind reading" logic where we are required to synthesize a different molecule using this molecule as a starting material. The correct structure led to a very easy synthesis, and the wrong structure led to a lengthy synthesis using reactions that had not been taught in class.