Providing a continuum solvent environment for H2O

  • now we are going to put our single water molecule inside a continuum solvent environment. We are going to try two different solvents, water a very polar environment and benzene a very non-polar envrionment
  • copy the water_freq.com to a water_pcm_water.com and water_pcm_benzene.com file, do the same for the corresponding checkpoint files as well.
  • now vi the new *.com files and change the %chk filenames
  • we could just use the structure from the water_freq.chk file, and this is what I would normally do, but I want to show you how to start one job from another if you acciddentally loose the checkpoint file, or it gets corrupted, and you cannot read in geometries/structure
    • vi the water_freq.log file and find the "Input orientation" copy the atomic numbers and xyz coordinates from here and paste into your new *.com files. You will need to remove the center number. 

    
                         Input orientation:
 ---------------------------------------------------------------------
 Center     Atomic      Atomic             Coordinates (Angstroms)
 Number     Number       Type             X           Y           Z
 ---------------------------------------------------------------------
      1          8           0       -0.350229   -0.249260    0.000000
      2          1           0        0.610358   -0.201592    0.000000
      3          1           0       -0.625946    0.672142    0.000000
 ---------------------------------------------------------------------
  • remove the guess=read geom=checkpoint keywords and replace them with geom=cartesian
  • we are going to optimise and then do the frequency analysis in one step this time (this is not normally encouraged, however we have a preoptimised small molecule so we can get away with it). Keep the freq keyword but add opt as well: freq opt.
  • now we need to identify the solvent and method, we are using the default method in gaussian so the new keyword is scrf(solvent=water) or scrf(solvent=benzene) scrf stands for self-consistent reaction field. 
    
%chk=water_pcm_water.com
%mem=3600MB
%nproc=2

# b3lyp/6-311g(d,p) geom=cartesian
  opt freq scrf(solvent=water)

single molecule water frequency analysis

0 1
   8           0       -0.350229   -0.249260    0.000000
   1           0        0.610358   -0.201592    0.000000
   1           0       -0.625946    0.672142    0.000000
  • submit the jobs to the queue!
  • note that if your -N name is too long the submit will fail, the name must be 14 characters or less!
  • check the jobs have terminated normally
  • check the jobs are converged using both gradients (Item) and frequencies (Harmonic)
  • set-up and run a job to compute the properties for these jobs, you will need both the sp keyword and then both pop and scrf keywords and their options.
  • when I did this these are the results I obtained, you should get something very similar!
    system method oxygen hydrogen
    isolated water NBO -0.8767 0.4384
    isolated water ESP -0.72631 0.3632
    water in water NBO -0.9162 0.45809
    water in water ESP -0.8040 0.4020
    water in benzene NBO -0.8946 0.4473
    water in benzene ESP -0.7608 0.3804
  • so the water environment polarizes the individual water molecule, increasing the charge on the oxygena and hydrogen atoms, the benzene environment also affects the single water molecule but not by as much.
  • that water has a greater effect on a polar molecule (the explicit water) is not surprising since Water: ε=78.3553 and Benzene: ε=2.2706
  • Energies are also a prime source of information, the energy you want in this case is identified "SCF Done: E(RB3LYP)"
    isolated water SCF Done: E(RB3LYP) = -76.4474479245
    water in benzene SCF Done: E(RB3LYP) = -76.4507132623
    water in water SCF Done: E(RB3LYP) = -76.4546537827
    so the water molecule is least stable in the gas phase, and most stable in a water environment.
Top