null-7
null-7
null-7
Molecular Modeling
1. Using Vibrational Frequencies to Verify an Equilibrium also presumably to bond lengthening. This in turn suggests
Structure. that Hartree-Fock bond distances are too short.
Each of the frequencies that make up the vibrational
(infrared/Raman) spectrum of a molecule is proportional to Compare the experimental bond length for hydrogen fluoride
the square root of the ratio of the curvature of the potential (0,917Ǻ) with that obtained from the Hartree-Fock model
surface in the vicinity of the structure (the second derivative of using the cc-pVTZ basis set. This is a sufficiently large basis set
the energy) and a mass. As mass is positive, this means that the to approximate the limit of the Hartree-Fock model for
resulting frequency will be a real number only if the surface equilibrium geometries. Is the calculated distance shorter than
curves upward (the second derivative is positive). Therefore, the experimental value?
the presence of one or more imaginary frequencies in a Repeat your calculations for lithium hydride (the experimental
calculated spectrum indicates that the structure is not an bond length is 1.596Ǻ). Is the Hartree-Fock distance shorter
energy minimum. than the experimental length? If it is not, provide an
explanation why not. Hint: examine both highest-occupied and
Use the HF/6-311G** model to obtain equilibrium geometries lowest-unoccupied molecular orbitals.
and vibrational spectra for the chair and boat forms of
cyclohexane. Are all frequencies real for both molecules? 4. Bond Lengths in Cyclopropane and Cyclobutane.
Which if either molecule is not a minima on the energy Use the Hartree-Fock model with the 6-31G* basis set to
surface? Elaborate. obtain equilibrium geometries for propane and cyclobutane.
Are the carbon-carbon bond lengths in the cycloalkane
2. Equilibrium Geometry of Disilylene (H2Si=SiH2). shorter, longer or about the same as those in propane (the
As discussed in the previous problem, a structure that gives “standard)? Is what you find consistent with the fact that CH
rise to a vibrational spectrum with an imaginary frequency is bonds on adjacent CH2 groups nearly eclipse each other
not an energy minimum. However, a minimum energy leading to an increase in non-bonded repulsion? Elaborate.
structure can be located from such a structure by “walking Given your result for cyclobutane, what you expect the carbon-
along” the geometrical coordinate associated with the carbon bond lengths in cyclopropane to be? Is your
imaginary frequency. expectation supported by the calculated equilibrium
structrure? If it is not, provide a plausible explanation as to why
Use the B3LYP/6-311G** model to obtain the equilibrium not.
geometry of planar (like ethylene) disilylene, the simplest
molecule with a silicon-silicon double bond, and calculate the 5. Water Dimer.
vibrational spectrum. Are all frequencies real numbers? If they The water dimer exhibits a structure with a single hydrogen
are not, perform the following operations: First, animate any bond. While the H --- O distance is not known experimentally,
imaginary frequencies to see how disilylene wants to distort to the distance between the two oxygen atoms has been
move it toward an energy minimum. Next, distort your measured as 2.98Ǻ.
structure accordingly. Finally, reoptimize the geometry (of the
distorted molecule) and again calculate vibrational
frequencies. Is it now an energy minimum? If not, repeat the
process until you get a minimum energy structure. Describe
the equilibrium geometry of disilylene.
The Hartree-Fock model with the very large cc-pVQZ basis set
3. Limiting Hartree-Fock Bond Lengths. shows a similar overall geometry but with an OO distance that
The Hartree-Fock model almost always yields bond lengths is 0.05Ǻ longer than the experimental value. This suggests that
that are shorter than experimental distances. This may be electron correlation shortens the hydrogen bond, the opposite
explained by noting that improvements to the Hartree-Fock of what is normally observed for covalent bonds. To model the
model involve promotion of electrons from molecular orbitals effect of correlation on the geometry of water dimer, consider
that are occupied to orbitals that are unoccupied. Occupied the consequences of promoting electrons from filled to empty
orbitals are typically either bonding or non-bonding whereas molecular orbitals, most simply from the highest-occupied to
unoccupied orbitals are typically antibonding. Therefore, the lowest-unoccupied molecular orbital.
electron promotion not only leads to a better description but
Obtain the geometry of water dimer using the Hartree-Fock/6-
31G* model, and display the highest-occupied and lowest-
1
unoccupied molecular orbitals. Is the HOMO bonding, larger dimerization energy also that with exhibits the larger
antibonding or non-bonding with respect to the H --- O changes in structure?
hydrogen bond? Would loss of an electron lead to shortening
or lengthening of the hydrogen bond or would it be expected 8. Infrared Spectrum of Acetic Acid Dimer.
to have little effect? Is the LUMO bonding, antibonding or non- Use the B3LYP/6-31G* model to calculate infrared spectra for
bonding with respect to the H --- O hydrogen bond? Would acetic acid and its dimer. (You already have equilibrium
gain of an electron lead to shortening or lengthening of the structures if you completed the previous problem.)
hydrogen bond or would it be expected to have little effect?
Overall, would you expect the limiting Hartree-Fock hydrogen
bond distance in water dimer to be shorter or longer than or
unchanged from the experimental value?
6. Alternative Structure of the Water Dimer. Point out and rationalize any significant differences between
As detailed in the previous problem, water forms a dimer with the frequencies and/or intensities associated with the (two)
a single hydrogen bond. However, because water incorporates OH stretching motions in the dimer and the OH stretching
two electron pairs that may act as hydrogen bond acceptors frequency in acetic acid.
and two OH bonds that may act as hydrogen-bond donors, it
should be possible to construct an alternative dimer with two 9. Radical Cation of Diborane.
hydrogen bonds. Does such a structure actually exist? One of the valence molecular orbitals of diborane closely
resembles the π orbital in ethylene.
2
ethylene. Do both structures appear to be minima on the C2H5+
potential surface or does one of the structures “collapse” to
the other? Elaborate. If there is only one energy minimum, is it
open or bridged?
4
21. Comparison of Measured and Calculated Frequencies. 23. Lithium Aluminum Hydride.
Frequencies routinely obtained from quantum chemical Lithium aluminum hydride (LiAlH4) is among the most
calculations assume that the potential in the vicinity of the important reducing agents, that is, source of “hydride”. Is it
minimum is a quadratic function. This means that they are best described as an ion pair between lithium cation and
generally larger than measured frequencies. (The correct aluminum hydride anion (AlH4— ), a weak complex between
potential for say a bond stretching motion goes to zero lithium hydride (LiH) and aluminum hydride (AlH 3), or are one
whereas a quadratic potential goes to ∞. It is possible to or more hydrogen atoms “shared” by the two metals? To
correct calculated frequencies for cubic and higher-order decide, obtain the equilibrium geometry using the B3LYP/6-
terms, but this is generally prohibitive in terms of computer 31G* model. Are the calculated atomic charges consistent with
time. It is also possible to correct measured frequencies for your geometry? Elaborate. Is there more than one stable
anharmonic behavior, leading so-called “harmonic structure for lithium aluminum hydride? Start from one or
frequencies”. These exist only for diatomic and very small more structures that are different from what you have found
polyatomic molecules, and include: LiF, 914; CO, 2170; N 2, and see if you get a different result.
2360 and F2, 923 (all in cm-1).
24. Sodium Cyclopentadienide:
Use the HF/6-311+G** model to calculate equilibrium Is sodium cyclopentadienide accurately represnted in terms of
geometries and vibrational frequencies for LiF (914), CO a structure in which sodium cation associates with
(2170), N2 (2360) and F2 (923). (Harmonic frequencies in cm-1 cyclopentadienyl anion?
are given in parentheses.) What is the average percentage
error between calculated and measured (harmonic)
frequencies? Are calculated frequencies uniformly too small or
too large?
Repeat your calculations using the B3LYP/6-311+G** model. Obtain the equilibrium geometry of sodium cyclopentadienide
What is the average percentage error between calculated and using the B3LYP/6-31G* model. Is the charge on sodium close
measured (harmonic) frequencies? Is this smaller than the to unity? Calculate the equilibrium geometry of
average error for Hartree-Fock calculations? Are calculated cyclopentadienyl anion and compare it with the
frequencies uniformly too small or too large? cyclopentadienyl fragment in sodium cyclopentadide. Is what
you find indicative of a weak complex? Elaborate. Calculate the
22. Methyl Lithium: energy of dissociation to cyclopentadienyl anion and sodium
The structure of methyl lithium appears to be “normal” insofar cation (you need to obtain the energy for sodium cation). Is the
as it incorporates a carbon with a roughly tetrahedral binding energy weak (< 100 kJ/mol), comparable to that of a
geometry. This could either mean that the C-Li bond is normal covalent bond (~400 kJ/mol) or somewhere in
covalent, or that lithium cation is loosely associated with between? Is your result consistent with the calculated charges
methyl anion (which also incorporates a tetrahedral carbon). and geometries?
Which of these descriptions is more consistent with atomic
charges obtained from the HF/6-31G* model? Elaborate. 25. Proton NMR Spectrum of Cyclohexane.
At very low temperatures, the proton NMR spectrum of
Experimentally, methyl lithium exists most simply as a cyclohexane shows two equal intensity lines at 1,12 and 1,60
tetramer with the four lithium atoms and four methyl groups ppm. Use the HF/6-31G* model to obtain the equilibrium
at the corners of a cube. geometry and the proton chemical shifts. Which line
corresponds to the equatorial protons and which corresponds
to the axial protons?