Dear all!
i am newbie to QM so my question might be naive.
-I'm optimizing a complex of 3-Au ring at the center surrounded by 3 citrate molecules.
my question is
1. if I manually place the 3 citric acid groups around the Au ring, is the final geometry independent of the initial positioning of the citric acid molecules. (I mean, does gaussian check all combinations of bond formations by rotating a molecule to report the final results or the calculations are guided by the initial placement of a molecule)
2. In gaussian optimization, I am using B3LYP gen pseudo=read > B3LYP 6-31G** for C H and O atoms, and core potential LANL2DZ for Au atoms. Are these setting optimal for running the complex optimization?
3. while using the gaussian optimization, is it possible to get new bonds formed (covalent)?
Many thanks!
Hi,
It is better to think of a potential energy surface including many minima (potential wells) to figure out how geometry optimization process works. Let's say A, B, C, etc for different potential wells and each of these corresponds to a different minimum energy structure, but only one of them is the lowest energy one! Let's say this "structure B".
if you start with a A-like structure the geometry optimization will end into structure A. If you start with C-like structure the optimization let it converge into the potential well of structure C, etc. So, to find the lowest energy structure of your molecule (structure B) it is necessary to start with the true initial structure. That requires a bit chemical intuition and of course some experience.
Citric acid is a small molecule but has many torsional angles. So you should first perform a "conformational analysis" of citric acid molecule. Geometry optimization does not make this! I suggest spartan program for this analysis. After you find the lowest energy conformer of your ligand, you should perform a geometry optimization for it. Now it is time to prepare the initial structure of your Au complex for geometry optimization by using the optimized citric acid structure.
I did not test LANL2DZ basis set for Au atom. It may be better to check different basis sets. But i suggest you adding the diffuse functions to your 6-31G basis at least for carbon and oxygen atoms.
In the geometry optimization process many bonds are formed and broken until it reaches to the final structure. You can check them by reading intermediate geometries from the output file. But of course you should only be interested in the final structure.
hope this helps..
Hi,
It is better to think of a potential energy surface including many minima (potential wells) to figure out how geometry optimization process works. Let's say A, B, C, etc for different potential wells and each of these corresponds to a different minimum energy structure, but only one of them is the lowest energy one! Let's say this "structure B".
if you start with a A-like structure the geometry optimization will end into structure A. If you start with C-like structure the optimization let it converge into the potential well of structure C, etc. So, to find the lowest energy structure of your molecule (structure B) it is necessary to start with the true initial structure. That requires a bit chemical intuition and of course some experience.
Citric acid is a small molecule but has many torsional angles. So you should first perform a "conformational analysis" of citric acid molecule. Geometry optimization does not make this! I suggest spartan program for this analysis. After you find the lowest energy conformer of your ligand, you should perform a geometry optimization for it. Now it is time to prepare the initial structure of your Au complex for geometry optimization by using the optimized citric acid structure.
I did not test LANL2DZ basis set for Au atom. It may be better to check different basis sets. But i suggest you adding the diffuse functions to your 6-31G basis at least for carbon and oxygen atoms.
In the geometry optimization process many bonds are formed and broken until it reaches to the final structure. You can check them by reading intermediate geometries from the output file. But of course you should only be interested in the final structure.
hope this helps..
As Berkay Sütay mentioned, you'll get the optimized structure which is closest to your initial guess structure (The geometry which is provided in the input file). So, you need to optimize all the possible conformers by yourself (and compare there energies) since Gaussian will not do that for you. Now you can do it manually, e.g by placing citric acids (CA) at all possible orientations surrounding the gold ring and generate a set of Gaussian input file and separately optimize all of them. However, this is inefficient and if there are many conformers you'll definitely miss some of them.
A better choice is to use automated algorithm. Check out the 'Lego' module of ABCluster (http://www.zhjun-sci.com/software-abcluster-download.php). You can combine ABCluster with Gaussian. It will automatically generate different guess structures of CA surrounding the central molecule and perform the optimization using Gaussian and finally report all the energies in increasing order.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Atomic, Molecular and Optical Physics
- Atoms