Question

I would like to generate a graph like the link below

http://en.wikipedia.org/wiki/Reaction_coordinate

The graph generated from a calculation of the python library installed.
I would like the line is smooth with type cspline gnuplot

The values E_ads=234.4211 , E_dis=0.730278 and E_reac=-0.8714

Could anyone help me

from ase import *
from ase.calculators.jacapo import *
import Gnuplot as gp
# -- Read in all energies
datadict = {'H2O' :'water.nc',
            'Pt' :'out-Pt.nc',
            'H2OPt' :'H2O.Pt.nc',
            'OHPt' :'OHPt.nc',
            'HPt' :'HPt.nc',
}
E = {}

for label, file in datadict.items():
    print 'Reading energy for %5s from file %20s' % (label, file),
    atoms = Jacapo.read_atoms(file)
    E[label] = atoms.get_potential_energy()
    print '\tE = %14.6f eV'% E[label]
print     

# -- Calculate adsorption and disassociation energies
E_ads = (E['H2OPt'] - 2*E['H2O'] - E['Pt'])/2
print 'H2O adsorption energy on Pt:'
print 'E_ads =', E_ads, 'eV\n'

E_dis = E['HPt'] - E['Pt'] + E['OHPt'] - E['Pt'] - E['H2O']
print 'H2O -> OH + H disassociation energy on Pt:'
print 'E_dis =', E_dis, 'eV\n'

E_reac = E['H2OPt'] - E['HPt'] - E['OHPt'] + E['Pt']
print 'H2O@Pt -> OH@Pt +H@Pt reaction energy on Pt:'
print 'E_reac =', E_reac, 'eV\n'
# -- Collect reaction path
Epath = np.asarray([1.0, E_ads, E_dis, E_reac])
PathLabels= ['']
# -- Plot the reaction path
import pylab as p
import numpy as np
import matplotlib.path as mpath
import matplotlib.patches as mpatches
import matplotlib.pyplot as plt
from scipy.interpolate import spline
import matplotlib.pyplot as plt
from numpy import array, linspace
from scipy.interpolate import spline
fig = p.figure(1)
sp = p.subplot(1,1,1)
p.plot(Epath, color='black', linestyle=':', linewidth=2.0, alpha=0.8)
p.text(0.37, 10.05, 'Free H$_2$O',fontsize=12, color='black',ha='right', va='bottom', alpha=1.0)
p.text(1.1, 238, 'H$_2$O + Pt',fontsize=12, color='black',ha='right', va='bottom', alpha=1.0)
p.title('H$_2$O disassociation')
p.ylabel('Energy [eV]')
p.xlabel('Reaction path')
#p.xlim([-0.5, 2.5])
#p.ylim([-0.5, 1.5])
sp.get_xaxis().set_ticks([]) # Turn off ticks on xaxis
#p.savefig('Teste.png')
p.show()

Answer 1

You can plot a regular cubic spline version of your data by just doing something like this:

plot(np.linspace(0,3),spline([0,1,2,3],Epath,np.linspace(0,3)))

which will yield something like:

But I suspect that isn’t what you want. You might need to resort to something like Monotone splines or shape preserving splines to get the shape which looks the same as those curves shown in your Wikipedia link. I don’t believe either of those interpolation methods are presently implemented in scipy.

If you have a rough idea of the mathematical form of those curves, you could always fit your own approximate function for the continuous section and just clamp the function outside of that range. For example:

plot(np.linspace(0,3),np.maximum(E_react,spline([0,1,2,3],Epath,np.linspace(0,3))))

would yield:

which at least “looks” like the curve you linked to, even if it isn’t the correct fit.