dennisfrancisblewettiii
Hazard to Self
Posts: 61
Registered: 16-12-2005
Location: Madison, Wisconsin
Member Is Offline
|
|
IR spectra: Identifying an unknown?
Hello folks,
In about a couple of days, I will be conducting a lab experiment during which I obtain the IR spectra of an unknown compound. Afterward, I will have
to attempt to identify the compound.
As far as I understand, I would be able to know if I found the right compound if I can match it up with a know IR spectra and they match.
But how do I go about doing that within particular time constraints?
Let me see if this is the right path:
1. Look at function groups within the 4000-3000 wave range.
2. Attempt to figure out which functional groups are present, attempt to determine a basic structure of the molecule...
Yeah, that's great... But what if I get a carbonyl? Or a carboxylic acid?
I mean... there are tons of unknown compounds it could be, especially if I could identify the functional groups attached to the carbonyl carbon.
I would have a lot of molecules to look through and look at.
Ideas to narrow down what possible molecules I might have?
I will have but a week's time to figure it out. I'll only be able to use IR. Personally, I'd like to get it down in about 4 days at most, so I have
two days or so to write up the lab report.
Also, how much does the finger-print region play into all of this? Should I spent a lot of time trying to figure out the finger-print region?
Personally, I believe it's impractical to use only an IR spectrometer to identify a molecule, except to a pure expert. I would think HNMR would
definitely help simplify what the molecule is, but we students won't be using such.
[Edited on 28-8-2011 by Genecks]
|
|
|
DDTea
National Hazard
Posts: 940
Registered: 25-2-2003
Location: Freedomland
Member Is Offline
Mood: Degenerate
|
|
You're on the right track, and yes: mixtures are the bane of structure determination with IR. For structure determination of mixtures, you need to
incorporate some separation scheme first (recrystallization, distillation, solvent extraction, chromatographic separation, etc).
HOWEVER, it's OK to use some chemical intuition here. Because of the nature of undergraduate laboratories, you will not be given something that is a)
highly toxic b) highly reactive c) highly volatile, etc. etc. This will rule out a lot of things. You probably won't be given a mixture, either.
Be sure to consider the physical properties of the material. For example, if you receive a flaky powder, it would be unreasonable to say that it is a
short, straight-chain alkane. If you are doing any wet tests on the material (e.g., pH, Tollen's test, Iodoform test..) be sure to consider the
results of those as well.
The finger-print region (1000-1400 cm^-1 or so) is critical for fine-tuning your structure determination, but turn to it only have you have your
"hunches" down. It is best used to confirm the presence or absence of certain chemical moieties. With practice, you will be able to obtain a lot of
information from this region. The danger is that can be extremely confusing unless you know what you are looking for and it's possible to make
incorrect assignments. Proper assignment of peaks is very context dependent and you must take into account absorptions in other regions of the
mid-IR. Don't be afraid to use it, though: after all, in science, we learn more from our mistakes than our successes.
In my experience with IR spectroscopy, the first thing to do in making a good structure determination is to take pains to acquire the best spectrum
that you can. Do you know what kind of sampling technique you are using? If it's ATR (common in undergrad labs and a nice technique in general), be
sure that the crystal is cleaned and dried after the previous user. If you can, acquire a clean background spectrum--this will mitigate the
contributions from atmospheric CO2 and water vapor (in any case, you should know what these absorptions look like, where they occur, and you should
ignore them). I don't know if you will be doing any sort of post-processing, but a background correction may be useful to you. If you are taking
multiple spectra of different samples, normalizing the data may be a good idea too.
If you want more information/practice using FTIR for structure determination, I strongly recommend Brian C. Smith's book "Infrared Spectral
Interpretation: a Systematic Approach." Brian C. Smith is a good author and he writes for beginners/end-users while still demystifying the "black
box" that FTIR can be.
At a more advanced level (i.e., more mathematically oriented and more steeped in physical chemistry), these books are basically the twin Bibles of
FTIR/vibrational spectroscopy:
Griffiths, Peter R., 1942-
Fourier transform infrared spectrometry / Peter R. Griffiths, James A. de Haseth. - 2nd ed.
ISBN 978-0471-19404-0
Colthup, Norman B.
Introduction to Infrared and Raman spectroscopy / Norman B. Colthup, Lawrence H. Daly, Stephen E. Wiberley. - 3rd ed.
ISBN 0-12-1282554-X
If you need reference spectra, use the NIST chemistry webbook. It's been a valuable tool for me.
[Edited on 8-29-11 by DDTea]
"In the end the proud scientist or philosopher who cannot be bothered to make his thought accessible has no choice but to retire to the heights in
which dwell the Great Misunderstood and the Great Ignored, there to rail in Olympic superiority at the folly of mankind." - Reginald Kapp.
|
|
|
Nicodem
Super Moderator
Posts: 4230
Registered: 28-12-2004
Member Is Offline
Mood: No Mood
|
|
For organic compounds, this instructive text is the best in my opinion:
Interpretation of Infrared Spectra, A Practical Approach
John Coates (In: Encyclopedia of Analytical Chemistry
R.A. Meyers (Ed.) pp. 10815–10837)
|
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
