-
Notifications
You must be signed in to change notification settings - Fork 2
/
PySACs.py
180 lines (170 loc) · 9.47 KB
/
PySACs.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
from rdkit.Chem import Draw
from rdkit import Chem
from rdkit.Chem import rdqueries
import numpy as np
import pickle
import matplotlib.pyplot as plt
from mendeleev import element
import os
os.environ["KMP_DUPLICATE_LIB_OK"]="TRUE"
print('************************************************************')
print('Developed by Mohsen Tamtaji ([email protected]) Under supervision of Professor Tom Luo at HKUST')
print('ML algorithm for the prediction of Gibbss free energy for HER, ORR, and CO2RR')
print('SVR model is trained based on 2104 data in the literature for graphene-based single atom catalysts')
print('************************************************************')
Location=r"E:\MLSAC\MlModel" #Define the location of your files (PySAC2.sav and Propertyoriginal.pkl)
Intermediates=[ 'H','N2-side','N2-end','OH','O','OOH','CO','CO2','COOH','CHO']
IntProperties=[[ 1, 2, 2, 1, 1, 1, 1, 1, 1, 1 ],#Numbers of atoms connected to SAC, first ring
[ 1, 14, 7, 8, 8, 8, 7, 7, 7, 7 ],#Atomic_nmber of atoms connected to SAC, first ring
[ 1, 5, 5, 6, 6, 6, 4, 4, 4, 4 ],#Valance_electron of atoms connected to SAC, first ring
[ 0, 0, 1, 1, 0, 1, 1, 2, 2, 2 ],#Numbers of atoms in the second ring
[ 0, 0, 7, 1, 0, 8, 8, 16, 16, 9 ],#Atomic_number of atoms in the second ring
[ 0, 0, 0, 0, 0, 1, 0, 0, 1, 0 ],#Numbers of atoms in the third ring
[ 0, 0, 0, 0, 0, 1, 0, 0, 1, 0 ]]#Atomic_number of atoms in the third ring
MLModel = '\PySACs.sav' #The name of ML model
filename='\Test.mol' # Put the mol file of your SAC structure into your directory
loaded_model = pickle.load(open(Location+MLModel, 'rb'))# load the ML model
mol = Chem.MolFromMolFile(Location+filename)
Draw.MolToFile(mol,Location+'\Test.png') #Save the structure of your SAC
for atom in mol.GetAtoms():
if atom.GetAtomicNum()>20:
elementAN=element(atom.GetAtomicNum())
TMsymbol=elementAN.symbol# the symbol of Transition Metal
NeighborsAN=[x.GetAtomicNum() for x in atom.GetNeighbors()]
NeighborsIndex=[x.GetIdx() for x in atom.GetNeighbors()]
NC,NN,Npr,Npy=0,0,0,0
for x in NeighborsAN:
if x==6: NC+=1 #C
elif x==7: NN+=1 #N
for x in NeighborsIndex:
if str(mol.GetAtomWithIdx(x).GetSymbol())=='N':
if str(mol.GetAtomWithIdx(x).IsInRingSize(5))=='True':
Npr+=1
if str(mol.GetAtomWithIdx(x).IsInRingSize(4))=='True':
Npy += 1
q = rdqueries.AtomNumEqualsQueryAtom(6)#C
CTM=len(mol.GetAtomsMatchingQuery(q))
q = rdqueries.AtomNumEqualsQueryAtom(7)#N
NTM=len(mol.GetAtomsMatchingQuery(q))
Propertyoriginal = pickle.load(open(Location+'\Propertyoriginal.pkl', 'rb'))
Property=np.zeros((len(Intermediates),np.size(Propertyoriginal[0,:])))
i=0
for Species in Intermediates:
elementname = element(TMsymbol)
# print(elementname)
m = str(elementname.ec)
# print(m)
if elementname.atomic_number < 32:
d = '3d'
elif elementname.atomic_number < 51:
d = '4d'
else:
d = '5d'
Property[i][0] = elementname.atomic_number
Property[i][1] = float(m[m.index(d) + 2:m.index(d) + 4])
Property[i][2] = elementname.atomic_radius
Property[i][3] = elementname.en_pauling
Property[i][4] = elementname.evaporation_heat
if m== 'Ru': Property[0][4] = 595
Property[i][5] = elementname.ionenergies[1]
Property[i][6] = elementname.electron_affinity
if m== 'Mn' or m== 'Zn':
Property[i][6] = 0
elif m== 'Cd':
Property[i][6] = -0.7
elif m== 'Hg':
Property[i][6] = -0.5
Property[i][7] = NTM
Property[i][8] = CTM
Property[i][9] = CTM/(CTM+NTM)*100
Property[i][10] = NN #
Property[i][11] = NC #
Property[i][12] = Npy #
Property[i][13] = Npr #
Property[i][14] = (elementname.nvalence()%2)/2*2+1 # Spin multiplicity of SACs
Property[i][15] = IntProperties[0][Intermediates.index(Species)]
Property[i][16] = IntProperties[1][Intermediates.index(Species)]
Property[i][17] = IntProperties[2][Intermediates.index(Species)]
Property[i][18] = IntProperties[3][Intermediates.index(Species)]
Property[i][19] = IntProperties[4][Intermediates.index(Species)]
Property[i][20] = IntProperties[5][Intermediates.index(Species)]
Property[i][21] = IntProperties[6][Intermediates.index(Species)]
i+=1
#Normalizing the input data between 0 and 1
for i in range(np.size(Propertyoriginal[0,:])):
Property[:,i]=(Property[:,i]-np.min(Propertyoriginal[:,i]))/(np.max(Propertyoriginal[:,i])-np.min(Propertyoriginal[:,i]))
Response=loaded_model.predict(Property)
with open(Location + "\Results.txt", 'w+') as f:
i=0
for Species in Intermediates:
f.write(Species+','+str(Response[i]))
f.write("\n")
i+=1
#ORR limiting potential
ULP=np.min([Response[3]-0,Response[4]-Response[3],Response[5]-Response[4],4.92-Response[5]])
UOP=4.92/4
print('Limiting potential for ORR is = ',np.round(ULP,2),' V')
print('Overpotential for ORR is = ',UOP,' V')
print('Find the free enegies in your directory')
print('Enjoy ...')
#Plot figure for HER
plt.figure()
plt.rcParams["font.family"] = "Arial"
plt.rcParams['axes.linewidth'] = 2
plt.rc('legend', fontsize=16)
plt.plot([0,1],[0,0],'b',[2,3],[Response[0],Response[0]],'b',[4,5],[0,0],'b',linewidth=2)
plt.plot([1,2],[0,Response[0]],'b--',[3,4],[Response[0],0],'b--',linewidth=2)
plt.xticks(fontsize=16,fontweight='bold',color='black')
plt.xticks([])
plt.yticks(fontsize=16,fontweight='bold',color='black')
plt.ylabel('Free energy (eV)', fontsize=20,fontweight='bold',color='black')
plt.xlabel('Reaction pathway', fontsize=20,fontweight='bold',color='black')
plt.tick_params(axis="x", direction="in", length=6, width=2, color="black")
plt.tick_params(axis="y", direction="in", length=6, width=2, color="black")
plt.text(2.3,Response[0],'H$_{ads}$*',fontsize=14,fontweight='bold',color='black')
plt.text(4.1,0,'1/2H2',fontsize=14,fontweight='bold',color='black')
plt.legend(['U=0 V'],loc='upper left',frameon=True)
plt.savefig(Location+"\HER.png",dpi=500)
plt.show
#[str(TMsymbol)+'N'+str(NN)+'C'+str(NC)]
#Plot figure for ORR
plt.figure()
plt.plot([9,8],[0,0],'k',[7,6],[Response[3],Response[3]],'k',[5,4],[Response[4],Response[4]],'k',[3,2],[Response[5],Response[5]],'k',[1,0],[4.92,4.92],'k',linewidth=2)
plt.plot([8,7],[0,Response[3]],'k--',[6,5],[Response[3],Response[4]],'k--',[4,3],[Response[4],Response[5]],'k--',[2,1],[Response[5],4.92],'k--',linewidth=1.5)
plt.plot([9,8],[0,0],'r',[7,6],[Response[3]-ULP*1,Response[3]-ULP*1],'r',[5,4],[Response[4]-ULP*2,Response[4]-ULP*2],'r',[3,2],[Response[5]-ULP*3,Response[5]-ULP*3],'r',[1,0],[4.92-ULP*4,4.92-ULP*4],'r',linewidth=2)
plt.plot([8,7],[0,Response[3]-ULP*1],'r--',[6,5],[Response[3]-ULP*1,Response[4]-ULP*2],'r--',[4,3],[Response[4]-ULP*2,Response[5]-ULP*3],'r--',[2,1],[Response[5]-ULP*3,4.92-ULP*4],'r--',linewidth=1.5)
plt.plot([9,8],[0,0],'r',[7,6],[Response[3]-UOP*1,Response[3]-UOP*1],'b',[5,4],[Response[4]-UOP*2,Response[4]-UOP*2],'b',[3,2],[Response[5]-UOP*3,Response[5]-UOP*3],'b',[1,0],[4.92-UOP*4,4.92-UOP*4],'b',linewidth=2)
plt.plot([8,7],[0,Response[3]-UOP*1],'b--',[6,5],[Response[3]-UOP*1,Response[4]-UOP*2],'b--',[4,3],[Response[4]-UOP*2,Response[5]-UOP*3],'b--',[2,1],[Response[5]-UOP*3,4.92-UOP*4],'b--',linewidth=1.5)
plt.xticks(fontsize=16,fontweight='bold',color='black')
plt.xticks([])
plt.yticks(fontsize=16,fontweight='bold',color='black')
plt.ylabel('Free energy (eV)', fontsize=20,fontweight='bold',color='black')
plt.xlabel('Reaction pathway', fontsize=20,fontweight='bold',color='black')
plt.tick_params(axis="x", direction="in", length=6, width=2, color="black")
plt.tick_params(axis="y", direction="in", length=6, width=2, color="black")
plt.text(8,0.1,'H2O(l)',fontsize=14,fontweight='bold',color='black')
plt.text(6,Response[3]+0.1,'OH*',fontsize=14,fontweight='bold',color='black')
plt.text(4,Response[4]+0.1,'O*',fontsize=14,fontweight='bold',color='black')
plt.text(2,Response[5]+0.1,'OOH*',fontsize=14,fontweight='bold',color='black')
plt.text(0,4.95,'O2',fontsize=14,fontweight='bold',color='black')
plt.legend(['U=0 V','U='+str(np.round(ULP,2))+' V','U='+str(np.round(UOP,2))+' V'],loc='upper right',frameon=True)
plt.savefig(Location+"\ORR.png",dpi=500)
plt.show
#Plot figure for CO2RR
plt.figure()
plt.plot([0,1],[0,0],'b',[2,3],[Response[8],Response[8]],'b',[4,5],[Response[6],Response[6]],'b',[6,7],[Response[9],Response[9]],'b',linewidth=2)
plt.plot([1,2],[0,Response[8]],'b--',[3,4],[Response[8],Response[6]],'b--',[5,6],[Response[6],Response[9]],'b--',linewidth=1.5)
plt.xticks(fontsize=16,fontweight='bold',color='black')
plt.xticks([])
plt.yticks(fontsize=16,fontweight='bold',color='black')
plt.ylabel('Free energy (eV)', fontsize=20,fontweight='bold',color='black')
plt.xlabel('Reaction pathway', fontsize=20,fontweight='bold',color='black')
plt.tick_params(axis="x", direction="in", length=6, width=2, color="black")
plt.tick_params(axis="y", direction="in", length=6, width=2, color="black")
plt.text(0,0,'*+CO2',fontsize=14,fontweight='bold',color='black')
plt.text(2,Response[8],'*COOH',fontsize=14,fontweight='bold',color='black')
plt.text(4,Response[6],'*CO',fontsize=14,fontweight='bold',color='black')
plt.text(6,Response[9],'*CHO',fontsize=14,fontweight='bold',color='black')
plt.legend(['U=0 V'],loc='upper right',frameon=True)
plt.savefig(Location+"\CO2RR.png",dpi=500)
plt.show()