Difference between revisions of "Comp Chem 02"
Line 79: | Line 79: | ||
:The Hartree-Fock (HF) theory makes 3 approximations: | :The Hartree-Fock (HF) theory makes 3 approximations: | ||
− | ::# | + | ::# Nuclei are stationary relative to the motion of the electrons; this is the ''Born-Oppenheimer Approximation.'' |
− | ::# | + | ::# Electrons act independently of each other; this is the ''Hartree-Fock Approximation.'' <-- this is totally not correct, but makes the calculations very doable. |
− | ::# | + | ::# The solution (ie. wavefunctions or orbitals) to the multi-electron schrodinger equation should be similar to the atomic solutions (wavefunctions or orbitals). So atomic orbitals (AO) form the basis for the multi-electron/molecular orbitals (MO); The MOs are a linear combination of atomic orbitals (LCAO); this approximation has no name. |
===Basis Set=== | ===Basis Set=== | ||
===Charge/Multiplicity=== | ===Charge/Multiplicity=== |
Revision as of 12:35, 16 April 2020
(4/10/20, bes)
Previous pages:
In the previous discussion you were taught how to submit a WebMO/Gaussian job and analyze results related to measuring of bond length and angles, viewing molecular orbitals, and determining the rotation barrier energy about a carbon-carbon bond. We will continue to use WebMO/Gaussian to develop chemistry insights, but first we must discuss further the nature of the calculations that were done in the previous exercises.
Review
WebMo is an excellent tool (graphical interface) to build an input file for the Gaussian engine. WebMO assisted in the construction of the input file; you are familiar with the figure on the left where you select your job/calculation parameters. The middle figure is generated if you choose to "preview" the job parameters, and the figure to the right shows how Gaussian interpreted the WebMO input file. The figure on the right is an excerpt from the "Raw output" "Actions" item accessed through the "View Job" window.
Job Options
As shown above, the user must specify certain job options. Each of these options will be discussed briefly.
Calculation
Molecular Energy
- This option calculates the energy, or "single-point (SP) energy" of the structure "as drawn" in the Build Molecule window. After building a molecule, structures are "cleaned up" and default bond lengths and bond angles are used. This calculation can be used to see how the energy changes as bond lengths/angles are changes. Overall, this is not a very useful type of calculation.
Geometry Optimization
- This option requests that a geometry optimization (OPT) be performed. The geometry (all bond length and bond angles) will be adjusted until the lowest energy configuration is found. There are multiple ways to adjust the geometry and there are multiple ways to evaluate when the lowest energy has been found, more on this later. This is probably the most used calculation option.
Vibrational Frequencies
- This option computes force constants and the resulting vibrational frequencies (FREQ).
Optimize + Vib Freq
- Because it is often necessary to optimize geometry (OPT) prior to doing a vibrational frequency (FREQ) analysis, an option exists to do these operations in series.
Excited States and UV-Vis
- For now, the name says a lot...calculates UV-Vis spectrum.
NMR
- Calculates the expected NMR spectrum. This does not work real well unless you use a very high level of theory. Please note that there are many NMR "simulation" packages...these use empirical (experimental) data to make an educated guess at the NMR spectrum. The math used to simulate is different than the math (QM) used in computational approaches.
Coordinate Scan
- A coordinate scan carries out a set of calculations on a molecule that differ in a single "coordinate," like bond angle. In the previous Exercise 3 - Study of Rotation Barrier Energy about Carbon-Carbon Bonds we performed a coordinate scan to show how the energy is dependent on the "torsion angle" of the central C-C bond.
Bond Orders
- skip for now
Molecular Orbitals
- Everybody knows about atomic orbitals, 1s, 2s, 2p, 3s, 3p, 3d, 4s, 4p, 4d, 4f, --> 7f, and we have even plotted the hydrogen atom atomic orbitals using Mathematica (or at least shown in the text). Again, i hate to be the bearer of bad news, but when a second electron is added to the picture, ie. helium, these well loved atomic orbitals (solutions to the Schrodinger Equation) are no longer valid. When two elements come together to form a bond, the resulting molecule orbitals (solutions to the Schrodinger Equation) look nothing like the atomic orbitals. We did do an introductory molecular orbital calculation in Lab 1, titled Exercise 2 - Study of the Molecular Orbitals in Ethene, CH2CH2.
Natural Bond Orders
- skip for now
Transition State Optimization
- skip for now
Saddle Calculation
- skip for now
IRC Calculation
- skip for now
CBS-QB3 High Accuracy Energy
- skip for now
Other
- WebMO has the above "common" calculation types hard-wired into the drop down menus. Gaussian can do many other, less common calculations and if you wish to do these you need to find the "keyword" to enter here.
Theory
The term "theory" make reference to how the Schrodinger equation is written/interpreted. As we will find out in Chapter 10: Multi-electron systems, we can easily write out the full Schrodinger equation (there are many terms and it might take a few minutes to write out), but as noted, this cannot be solved. So this is where we make approximations. Each "level of theory" has a different set of assumptions. The more approximations made (ie, the lower the level of theory) the less time it takes to complete the calculation. When less approximations are made (ie. higher level of theory) the longer the calulations take to complete.
Below are the most common levels of theory
Hartree-Fock
- The details of this level of theory is described in Sec 15.3.
- The Hartree-Fock (HF) theory makes 3 approximations:
- Nuclei are stationary relative to the motion of the electrons; this is the Born-Oppenheimer Approximation.
- Electrons act independently of each other; this is the Hartree-Fock Approximation. <-- this is totally not correct, but makes the calculations very doable.
- The solution (ie. wavefunctions or orbitals) to the multi-electron schrodinger equation should be similar to the atomic solutions (wavefunctions or orbitals). So atomic orbitals (AO) form the basis for the multi-electron/molecular orbitals (MO); The MOs are a linear combination of atomic orbitals (LCAO); this approximation has no name.