I have several spectra of different samples which are probably all copper (neutral and/or basic) acetate(s) and which are very similar. I present them overlayed and (some of them) stacked:

Overlayed spectra

Stacked spectra

My question is: can I assume all the samples are qualitatively identical or do subtle/slight intensity differences in the same bands mean a qualitative difference?

These are all spectra made with a pellet of KBr and I intend a qualitative characterisation to compare with literature, not a quantitative one.

  • $\begingroup$ Chemists prefer IR spectra with (remnant) %transmittance along the ordinate axis, and with one (or even two) brakes in the abscissa at about 2200 1/cm (or even a twice at about 1000 1/cm) instead of a straight linear axis; it renders the spectrum more legible. In most cases, the software of the spectrometer offers this and it is only a few clicks away. Is it a transmittance spectrum with a pellet of KBr, or a neat film/ATR? Do you intend a qualitative characterisation to compare with literature, or a quantitative one? Do the traces represent several batches of one (what?) intended product? $\endgroup$
    – Buttonwood
    Apr 21 '17 at 21:06

(Answered after your edit of your question:)

Some variation in the location of the absorption bands (vs. 1/cm) and their intensities (vs. remnant transmittance) is unavoidable due to the interaction of your analyte (here: your copper salt/salts) with the matrix (here: pellet of KBr) compared to recording a neat film of your sample. The spectra depicted by the trace in magenta and blue seem to be a little bit different from the others, perhaps due to some humidity (KCl exposed to air, or a sample not completely dried, or both, ...). On first view of the current representation, the other look very similar, compared with each other.

  • As your data represent infrared spectroscopy, figure out how to change the representation such that your ordinate is in remnant %transmittance instead of absorption. Let the spectrometer's software do this for you. For a visual inspection of the data, a split in the abscissa, too, for example around 2000 1/cm eases the visual inspection a lot -- equally something typically offered by the software of the spectrometer. (On good instruments, both changes do not alter the raw data collected by the spectrometer.) For example:

enter image description here

(source, the picture was edited by addition of horizontal lines)

  • Determine the barycentre of your absorption bands. Likely, as you start with IR spectroscopy, a deviation of about 1/cm is not an issue (ask your instructor). In other words, if the peak pic by the software suggests 1837.54 1/cm, a report like 1838 1/cm likely is good enough (ask your instructor).
  • For each important absorption band, have a look about the shape, note if it is a particular broad absorption band -- in your example, around 3350 1/cm, for example.
  • Beside location, an other important information provided by IR spectroscopy is the intensity of the absorption band. In the representation of %T, imagine three equal parts between the baseline on top, and the most intense absorption band on the bottom. A bit like the three horizontal blue dotted lines added to the picture above. The absorbtion bands in the first strip are then (semiquantitatively) classified as of weak, the ones reaching till the second strip of medium, and the ones reaching down to the third strip of strong intensity. (As an example, the edited spectrum above contains an absorption band of medium intensity at about 3500 1/cm, and two strong ones at about 1700 and 1500 1/cm.)
  • Compare localisation and absorption band intensity of the spectral data recorded by you, with data found in a reference (textbook, journal article, data base). Are all important absorption bands found in your spectrum mentioned in the literature reference? Do you find all important absorption bands described in the literature reference in your spectrum, too? Are there some important bands in addition or missing?
  • Of course you may compare sample #1 with sample #2, with sample #3, etc., too.

Only by growing experience you will learn what important absorption bands in infrared spectroscopy are. Because -- in contrast to other spectroscopic techniques -- it is rather the exception than the general case that you may correlate each absorption band with a structural element in the structure of your compound.

For future reference I suggest to record the IR data either already in the %T mode, or (second best) to convert them into this form. This allows you to check rapidly if even the most intense absorption band does not hit a transmittance of 0% (such an occurence were a rapid indication of a sample with to high concentration of X in your pellet of KBr.)


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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