Here is the $\pu{2000-4000 cm^{-1}}$ spectral region (known for showing hydrogen bonding interactions, $\ce{-H\bond{<-}X-}$) for three different compounds, each showing a different functional group:

enter image description here


  • A is a carboxylic acid,
  • B is an amine, and
  • C is an alcohol.

Is it possible to uniquely identify these three compounds just from peak positions, shapes, and intensities within this window?

| improve this question | | | | |
  • $\begingroup$ Not quite sure anyone is really going to understand what your exact problem is? Do you have any specific reason to doubt your choices? If so, what was your line of thought in solving these? Give us specifics - we are not here to do your homework. $\endgroup$ – Bob Mar 12 '17 at 22:36
  • $\begingroup$ @Bob I just looked at pictures of graphs of similar groups and assigned them based on shape. We weren't told if we should do anything more analytic than that. $\endgroup$ – redbull123 Mar 12 '17 at 23:53
  • $\begingroup$ I would suggest reading your textbook to figure out why specific patterns appear in a spectrum. $\endgroup$ – Bob Mar 13 '17 at 0:08
  • $\begingroup$ @Bob I remember Pretsch et al., Structure Determination of Organic Compounds (link.springer.com/book/10.1007%2F978-3-540-93810-1) includes even "hypothetical IR spectra". To clarify this: how the spectrum should look like if only this functional group (amine, aldehyde, ester, ...) were present; presenting the whole range from about 4000 to down 600 (or so) 1/cm; followed by examples. Still today I find this presentation more instructive and helpful than "the simplier tables" alone. $\endgroup$ – Buttonwood Mar 13 '17 at 0:28
  • $\begingroup$ @Buttonwood Then they're based off the max's. $\endgroup$ – redbull123 Mar 13 '17 at 2:34

Perhaps you do not know that the title of your question already contains an important key to interpret IR-spectra: not only to inspect rather small, isolated regions of the spectrum, but to see the spectrum as a whole. By doing so, you have to watch out for additional absorption bands that were equally providing evidence for one, or the other structure; by scrutinizing both the position ($\tilde\nu$), shape, and relative intensity of the absorption bands in question.

As an example: is there evidence for a carbonyl group, indicative for the acid?

Why? Because -- depending on the mode of data collection (e.g., transmission in a matrix of a KBr pellet, or neat film by ATR) the shape of the spectra recorded may differ. In contrast to 1H NMR, IR offers you a view on structural elements .and. the whole molecule. Then, indeed, IR spectroscopy may offer much insight.

P.S.: From the baseline of the spectra shown it is likely the data were recorded on a ATR-FT-IR spectrometer. If you look into the data recorded, you will find each point is typically about 0.43 wavenumbers a part from the next one. Hence it may be more sound to perform the "peak pic" / stating the barycentre of the absorption bands only by whole wavenumbers, without two digits behind the decimal point (significant figures).

| improve this answer | | | | |

Not the answer you're looking for? Browse other questions tagged or ask your own question.