Chemistry @ CSU Stanislaus

The Synthesis and Characterization of Aspirin  Part 3

This requires a Java enabled browser

move me

Purpose

There are three parts to the purpose of this experiment.

 

Procedure

All three of the parts of this experiment will use nuclear magnetic resonance spectroscopy (nmr) as the source of the evidence.  Two nuclei will be used, 13C and 1H.  The actual experiment is simple.  You need to obtain  13C and 1H nmr spectra for your aspirin product and also for your unknown sample.   These spectra will be your data.  They must be included in your notebook.

Sample preparation

Safety:  Whenever a chemist works with a substance for which the hazards are unknown, the substance is used with great caution.   This is especially true when trying to identify an unknown substance because, obviously, you don't know what it is and so you can not determine it's properties and hazards.  In this experiment,  all solution preparation must be carried out in the hood.  Do not even open the solvent or sample container outside of the hood.   Be sure you wear gloves and goggles!  Seal the solvent and sample containers, and be certain they are clean, before removing them from the hood.

The aspirin from your synthesis will be dissolved in a small amount of deuterated methanol.  You can get this from the Stockroom along with an nmr sample tube.  These tubes are fragile and expensive.  You don't want to pay for replacing one, so be careful with it.  Add small amounts of your sample of aspirin to the solvent in a small test tube until a small amount remains undissolved.  This is a saturated solution.  Using a disposable pipet, transfer the solution to an nmr tube.  It is ok to have a very small amount of solid floating in the nmr tube.  There should be about 4 cm of liquid in the tube.

You can also get your unknown from the Stockroom.  It will come with a small amount of solvent (which will depend on the specific substance) and instructions for preparing the solution for analysis. After preparing the solution in a small test tube, use a clean disposable pipet to transfer it to a clean nmr tube for analysis.  Again, you should have about 4 cm of liquid in the tube.

Finally, make an appointment for one of us to run your spectra for you.   If you are reading this now it is a good time to start.  The sooner you have spectra, the more time you will have availabe to interpret the spectra.

Spectral analysis

Analysis of nmr spectra can be done in two ways.  The simplest way is to verify a structure by comparing sample spectra to spectra recorded for authentic samples.  Spectra of samples of aspirin and salicylic acid are found here.

It is not unusual to find it necessary to interpret the spectra and postulate a possible compound which is consistent.   This may be the extent of the analysis.  It is better, though, to then obtain a spectrum for an authentic sample to confirm the assignment of the compound and the specific peaks.   It can also be helpful to compare spectra for unknown compounds with spectra for similar known compounds to get hints for putting together a possible structure.   Identical fragments of molecules will have very similar spectra.  A set of spectra from some very similar compounds are found here.

Spectra of compounds can also be found in the SDBS database.  Here you can obtain molecular structures and  13C and 1H nmr spectra with the peaks attributed to specific atoms.  This database can be searched for compounds by name, molecular formula (even ranges of numbers of C's, H's, O's and N's), molecular weight, and CAS numbers. 

Another helpful source of information about specific compounds is ChemFinder.  This website mines the world wide web for references to compounds.

By combining the information obtained from both the 13C and the 1H nmr spectra from a compound, it is often possible to identify a sample uniquely.  If not, at least it is possible to suggest a reasonable structure based on the spectra.  Negative evidence is often as important as positive evidence.  By knowing what the compound cannot be, it is easier to build a structure that is consistent with the spectra.

The interpretation of spectra is based primarily on several kinds of information.

From 13C spectra we can get the following information:

  1. Each unique environment of a carbon atom results in a peak in the spectrum.  The position of the peak is known as the chemical shift.  If all of the carbons are in unique environments there will be a peak for each carbon.  Carbon atoms with very similar environments will have similar chemical shifts.  The chemicals shifts can be compared with those for authentic samples.  This can be done by comparing the chemical shifts for the peaks in your spectra to a compilation of typical values for each type of carbon.  This compilation is called a correlation diagram.  You can find a simple one of these for carbon nmr here.  You may also find it helpful to compare your spectra to spectra for similar compounds as described above.

  2. If several carbons have very similar chemical shifts (bond types) then the peak heights may be roughly proportional to the number of each type of carbon.   See the carbon spectrum for

From 1H nmr spectra we get the following information:

  1. Unique proton environments also result in different chemical shifts and a correlation diagram is available for these spectra at the same place as the one for carbon.

  2. If a proton in a molecule has a neighboring proton in a different environment (a different chemical shift) the peaks for both protons will be split into two peaks (a doublet) with the same amount of splitting for each set.  This splitting will occur between protons which are within a distance of 3 bonds and generally decrease in magnitude as the number of bonds between the protons increases.  (Oh Oh, I almost forgot to tell you that when we count these bonds, we can ignore double and triple bonds)   From this we can see that a peak for a proton with two different kinds of protons as neighbors will be split into two sets of two peaks. 

    If a proton has two neighboring protons which are in the same environment then the peaks is split into three peaks (a triplet) with the ratio 1:2:1.  A proton with three identical protons as neighbors will be split into four peaks (a quartet) with the ratio 1:3:3:1.   See the spectrum of ethyl benzoate for an example of this.  Note that it is possible to have, for example, more than one kind of neighbor splitting a peak to give a multiplet of multiplets. 

    If there are no neighboring protons to cause splitting then the peak will be a singlet.   Turning this around, a singlet usually means at least one proton on an isolated carbon.

    This splitting results in characteristic patterns of peaks for different substitution on benzene rings.  Compare the proton spectra for compounds with 5 aromatic protons (example) and the three possible examples of 4 aromatic protons (1, 2, 3).  This can be very helpful in deciding the exact structure.

  1. In a proton spectrum the area of the peak set for a particular proton is proportional to the relative number of the protons of that type.   We can determine the area of these peaks by integrating them and so identify the relative number of each type of hydrogen.  This assumes that the spectrum is obtained from a pure substance or that, if there is a mixture, the peaks for the pure compounds can be identified.

Above all, remember that any molecule you propose must have a valid Lewis structure.

 

Conclusion

The conclusions for this experiment are straightforward.  Answer the questions posed in the purpose.  It is very important that you justify your conclusions.  This is done by referring to the spectra and showing how the specific peaks are consistent with those for the compounds in your conclusion.  In other words, explain how all of the peaks in the spectra correspond to those expected for aspirin, salicylic acid or your unknown.

 

Tools

SDBS correlation diagrams chemfinder

Spectra of interest

samples aspirin and salicylic acid

Last modified on 02/03/99