Techniques of Molecular Modelling: Molecular Mechanics
Case Study 4: Deciding Stable Ring Conformations.
A modelling problem often comprises a complex series of molecules, their
properties and their reactions. The first stage would be to break down the
problem into smaller problems, each of which might require a separate technique
to study them. Take for example the
following reactions. When compound 12 is photolysed in heptane, a solution of "x"
is formed, consisting of an equilibrium mixture of 13 and 14. If the solution is
heated to 98C, "x" then reacts to form a new compound 15 in which the two vinyl
groups are specifically trans. The transformation of 12 to 13, and of 13 to 14
are classical pericyclic reactions where a method appropriate for breaking bonds
and studying transitions states should be used. Molecular mechanics is clearly
NOT appropriate for these steps.

But what about the transformation of 13 to 15? The original authors make
no comment on the origins of the stereoselectivity, but it is clearly tempting to conclude that
this reaction must be a transition state controlled reaction. But it might also depend
on the conformation of the rather unusual ten membered ring, containing two trans
alkene units. Molecular mechanics is in fact an excellent and very rapid probe of
the conformation of such a ring. Its worth finding out how much we can learn about the
system in this manner.
The first step in modelling this reaction would be to define the connectivity
using the molecule editor, and then minimise the structure using a MM force field
suitable for alkenes such as MM2. When 13 is so minimised, one should be aware that in
several conformations might be possible in a 10-membered ring. Two such
conformations are readily located differing in energy by 4.6 kcal/mol. The lowest
has a trans relationship at the bond formed in 15,
stretch = 0.490 angle = 1.353
stretch bend = 0.044 dihedral = 8.366
improp torsion = 0.254 van der Waals = 6.036
electrostatics = 0.133 hydrogen bond = 0.000
The energy of the final structure is 16.6773 kcal/mol.
The higher has a cis
relationship, the difference between the two isomers arising arising purely from the angle term
in the MM2 force field.
stretch = 0.497 angle = 5.453
stretch bend = 0.076 dihedral = 8.415
improp torsion = 0.292 van der Waals = 6.406
electrostatics = 0.146 hydrogen bond = 0.000
The energy of the final structure is 21.2843 kcal/mol.
Thus the
stereospecificity of this reaction is readily explained in terms of angle strain and not
Woodward Hoffmann rules (which by the way would allow both cis and trans products
to form for 15).
Literature Citation. S. W. Staley and T. J. Henry, J. Am. Chem. Soc., 1970, 92, 7613
Case Study 5: Including Metal Ions
Here, one can see that a variety of bond types have to be defined in the mechanics
force field, including values for various Mg...O and Mg...C weak bonds. At best these
are going to be guestimates, but this does allow some exploration of the steric
environment of the reacting site to be made. The original MM2 force field does not
allow such bonds to be calculated, but some more recent modelling programs such as
CAChe do allow such estimates. One interesting conclusion emerging from the calcuations is that
the nature of the 4th coordination site on the Mg (THF is shown
below) does affect the outcome of the calculations, suggesting the reaction may
in fact be solvent sensitive. This in turn might suggest further experiments
to be carried out.
stretch = 14.962 angle = 20.117
stretch bend = 0.023 dihedral = 25.369
improp torsion = 0.307 van der Waals = 16.707
electrostatics = -3.453 hydrogen bond = 0.000
The energy of the final structure is 74.03 kcal/mol.
stretch = 14.734 angle = 19.834
stretch bend = -0.058 dihedral = 27.585
improp torsion = 0.315 van der Waals = 16.633
electrostatics = -3.313 hydrogen bond = -0.529
The energy of the final structure is 75.20 kcal/mol.
Literature Citation. A. G. Shultz, L. Flood and J. P. Springer, J. Org. Chemistry, 1986, 51, 838.
Case Study 6: The Stability of Carbonium Ions
Organic chemists are taught that carbonium ion stability is predominantly due to
electronic factors such as whether they are primary, secondary, tertiary, allylic,
benzylic etc. What is mentioned much less is that their relative stability is also very
sensitive to their geometries, and in particular the angles of the three substituents
at the carbon. So what does Molecular Mechanics make of the following carbonium ions?
They are arranged in the expected order of increasing stability. Notice that
the MM2 method gradually decreases the predicted angle energies. Notice also however
how it predicts that the tertiary carbonium ion is actually LESS stable than the
preceeding primary ion, which is surprising to say the least. Clearly, electronic factors
must be playing an important role as well. Suffice to say that even without ANY
consideration of electronic factors, the MM2 method does not do too badly in its
prediction of relative stabilities. This re-inforces the conclusion that bond angles
in carbonium ions are just as important as substitution!
To make the next jump in realism in our modelling technique, it is clearly time to
move on to the Quantum Mechanics methods, where electronic effects are explictly
calculated. A preview of the results obtained with such a method is shown here for
comparison with the MM results.
stretch = 0.868 angle = 19.173
stretch bend = -0.672 dihedral = 6.497
improp torsion = 0.155 van der Waals = 2.663
electrostatics = -1.341 hydrogen bond = 0.000
The energy of the final structure is 27.342 kcal/mol. (Final QM heat of formation = 249.11 Kcal/mol)
stretch = 0.447 angle = 18.758
stretch bend = -0.078 dihedral = 6.916
improp torsion = 1.124 van der Waals = 1.622
electrostatics = 0.000 hydrogen bond = 0.000
The energy of the final structure is 28.788 kcal/mol. (Final QM heat of formation = 232.62 Kcal/mol)
stretch = 0.571 angle = 5.694
stretch bend = -0.027 dihedral = 7.179
improp torsion = 0.009 van der Waals = 5.678
electrostatics = -0.545 hydrogen bond = 0.000
The energy of the final structure is 18.559 kcal/mol. (Final QM heat of formation = 211.67 Kcal/mol)
stretch = 0.512 angle = 3.439
stretch bend = -0.153 dihedral = 6.662
improp torsion = 0.002 van der Waals = 3.872
electrostatics = 0.683 hydrogen bond = 0.000
The energy of the final structure is 15.017 kcal/mol. (Final QM heat of formation = 195.79 Kcal/mol)
Literature Citation. 2nd Year Organic Problem Sheet, Imperial College, 1991. Set 3.
Return to main course page|Back to Case Studies 1-3| Forward to case Studies 7-9|
(c) H. S. Rzepa 1998. No reproduction rights granted to this material without permission.