Navigation Bar
Place mouse on line with 
symbol to expand menu.
|
QUANTUM CHEMISTRY &
CHEMICAL INFORMATICS
Below is a brief introduction of the Quantum Chemistry and Chemical Informatics required to study real-world problems.
I have used Quantum Mechanics, Quantum Chemistry, and Chemical Informatics to aid in the design and analysis of the following systems.
- Calculated UV, IR and NIR Spectra, Reaction Probabilities, Solvating Parameters, and Ground-State Molecular Geometry for Organic and Inorganic Molecules using Ab Initio and Semi-Empirical techniques.
- Catalogued and sorted electronic spectra of dyes.
- Designed Organic Molecules to have certain electronic properties. This involves R&D in NIR absorbers and "Moletronics" molecules.
- Studied Nuclear Structure by developing the Theory and computational Computer Programs to calculate Five-Particle, Spin-Dependent Eigenvalues.
- Created Chemical Informatic Data Base for Disperse Dye Molecules.
Quantum Chemistry is the field of science that studies the theoretical (quantum mechanical) aspects of molecules. Chemical Informatics is the field of endeavor that organizes and sorts the massive amount of chemical data into useful subsets. The data that is experimentally determined and/or calculated, is the source of raw data that is organized by chemical informatics. Both of these efforts rely on the development of modern computer hardware and software.
Fast Personal Computers and new Quantum Chemistry software has allowed complex molecular studies to be performed that could only be dreamed about 30 years ago. One of the most significant improvements is in how quickly the molecular geometries can be set up on these new systems. I performed a five-particle calculation as part of my Ph.D. Dissertation in 1973 and it took approximately 750 hours to just set up the geometry. It took another 100 or so hours to calculate and card-punch my Clebsch-Gordon, 3j, and 6j coefficients to be used in the calculation. I then ran the problem on an IBM 360-65 computer for 12-hours/night, 7 days/week for another 3 months. Further, when I was running, mine was the only job in the computer's CPU.
Now I can set up a complex molecule (100 atoms) in a matter of 15 minutes. I can typically optimize the geometry, and run an electronic spectrum in another 30 minutes using Semi-Empirical techniques with HyperChem.
There are now several good quantum chemistry software packages on the market. They typically can run the following types of problems.
- Ab Initio Quantum Mechanics - These are the fundamental (first principle) calculations of quantum chemistry. They expand molecular orbitals into a linear combination of atomic orbitals (LCAO) and introduce no other approximations.
- Semi-Empirical Quantum Mechanics methods use a rigorous quantum mechanical formulation combined with the use of empirical parameters obtained from comparison with experiment. They use Hartree Fock wave functions in the iterative SCF (Self Consistent Field) process to obtain the correct eigenvalues. The Roothan equations are the basic equations for closed-shell Restricted Hartree Fock (RHF) molecular orbitals and the Pople-Nesbit equations are the basic equations for the open-shell Unrestricted Hartree Fock (UHF) molecular orbitals. The supported semi-empirical methods are briefly summarized below.
- Extended Hückel Theory - It is present in most packages and it is fast. However, it is not recommended because it is not as accurate as other more sophisticated semi-empirical methodologies.
- CNDO/2 - Complete Neglect of Differential Overlap method is the simplest and least accurate of the "post-Hückel" Zero Differential Overlap (ZDO) methods. This methodology should not be used where electron spin is critically important.
- INDO - Intermediate Neglect of Differential Overlap method differs from the CNDO/2 in cases where the electron spin interactions (called electron exchange) are particularly important (NH, or the radical CH3).
- MINDO/3 - Modified Intermediate Neglect of Differential Overlap method is a modification of the INDO method. This is very effective for studies in a wide variety of hydrocarbons , but problems arise in its use on molecules containing heteroatoms (non-hydrocarbons).
- MNDO - Modified Neglect of Diatomic Overlap method. It is widely used to calculate heats of formation (see AM1, below), molecular geometries, dipole moments, ionization energies and electron affinities. It has problems dealing with sterically crowded molecules, four-membered rings, hydrogen bonds and hypervalaent compounds.
- AM1 - This is a modified MNDO method and is generally the most accurate semi-empirical method for most problems. It deals with hydrogen bonds correctly, produces accurate predictions of activation barriers for many reactions and predicts heats of formations better than MNDO.
- PM3 - PM3 is a reparameterization of the AM1 method. PM3 is used primarily for organic molecules, but is also parameterized for main group elements.
- ZINDO/1 - This model is based on the INDO model by Prof. Zerner of the University of Florida, hence the name ZINDO/1. ZINDO/1 is optimized to calculate geometries and energies of molecules with first of second transition row metals. However, it yields less than desirable results for organic molecules.
- ZINDO/S - This model is parameterized to reproduce electronic spectroscopic transitions and should not be used for geometry optimization. It utilizes a configuration interaction methodology to calculate these spectra.
Go to Top of Page
|
