Gaussian: Make Energy Convergence in Optimization Easy !
Make Energy Convergence in Optimization Easy !
In this post I would like to present a technique that can make your geometry optimization job to converge easily. Route portion in Gaussian input controls the job calculations, level of theory, etc. The followings are some important keywords you should know and we will play around with them.
Printing Output Level
#P will provide somewhat more detailed output (recommended)
# will provide normal output
#T will provide somewhat less output
Geometry Optimization
opt=Z-Matrix request Gaussian to optimize the geometry in internal coordinates (as provided in the input file)
opt or opt=Redundant request Gaussian to optimize the geometry in redundant internal coordinates (chosen automatically)
opt=Cartesian request Gaussian to optimize the geometry in Cartesian coordinates
Work around when Energy does not converge.
Item Value Threshold Converged? Maximum Force 0.021672 0.000450 NO RMS Force 0.018596 0.000300 NO Maximum Displacement 0.038954 0.001800 NO RMS Displacement 0.033876 0.001200 NO Predicted change in Energy=-1.250480D-03The energy will converge when the values of quantities in first column is less than or equal to the threshold (3rd column)
To gain the accuracy of geometry optimization, one can reduce the threshold, but the cost of computational will also be increased.
1. Use OPT optional keyword.
Default threshold of Maximum Force and RMS are set to 4.5 x 10-4 and 3 x 10-4ถ้าเป็น, adding opt=tight will request Gaussian to make
the energy convergenceopt more tight. Since tight is added, the threshold of RMS becomes 1 x 10-5.
Moreover, you can use reduce more by using verythight, which will produce more accuracy, threshold of RMS is set to 1 x 10-6),
but the calculation would be running very slow. ได้ถึง 1*10-6
2. Use IOP keyword
IOP is internal option that users can define or set values of parameter used in calculation themselves.
For increasing the accuracy of convergence, threshold will be reduced, you can use iop(1/7=x), whre x is the number of negative power of 10.
For example, iop(1/7=5) is the same meaning of opt=tight, which the threshold is set to 1 x 10-5.
Rangsiman Ketkaew