andrewtarzia
commented
3 years ago Hey @Jaloca2 - Welcome and thank you!
This is an interesting structure problem! I have previously written code that performs a similar construction, where a MOC is reacted at functional groups on its ligand backbone. I modified that code to do the reaction at the axial position (clearly, I have used a weird set of building blocks, but you can figure out how to replace them with what you want; let me know how it goes for you!).
Note that the way I have done this was to actually modify the Paddlewheel topology to have an axial position. i hope the code makes sense! I will suggest that the selection of COF topology may be a problem here, so let us know what you are trying to build and we can try to help!
@lukasturcani - what do you think of this solution??
import numpy as np
import stk
import stko
class AxialPaddlewheel(stk.metal_complex.MetalComplex):
"""
Represents a metal complex topology graph.
Two distinct metal building blocks, one with five functional
groups, one with four.
Binding ligand building blocks with two functional groups are
required for this topology graph and axial ligand building
blocks with one functional group are required.
When using a :class:`dict` for initialization, a
:class:`.BuildingBlock` needs to be assigned to each of the
following numbers:
| axial_metal: (0, )
| other_metal: (1, )
| ligands: (0, 1, 2, 3)
| axial_ligand: (4, )
See :class:`.MetalComplex` for more details and examples.
"""
_metal_vertex_prototypes = (
stk.metal_complex.vertices.MetalVertex(0, [0, 1, 0]),
stk.metal_complex.vertices.MetalVertex(1, [0, -1, 0]),
)
_ligand_vertex_prototypes = (
stk.metal_complex.vertices.BiDentateLigandVertex(2, [2, 0, 0]),
stk.metal_complex.vertices.BiDentateLigandVertex(3, [0, 0, 2]),
stk.metal_complex.vertices.BiDentateLigandVertex(4, [-2, 0, 0]),
stk.metal_complex.vertices.BiDentateLigandVertex(5, [0, 0, -2]),
stk.metal_complex.vertices.MonoDentateLigandVertex(6, [0, 2, 0]),
)
_edge_prototypes = (
stk.Edge(
id=0,
vertex1=_metal_vertex_prototypes[0],
vertex2=_ligand_vertex_prototypes[0],
position=[0.1, 0.5, 0],
),
stk.Edge(
id=1,
vertex1=_metal_vertex_prototypes[1],
vertex2=_ligand_vertex_prototypes[0],
position=[0.1, -0.5, 0],
),
stk.Edge(
id=2,
vertex1=_metal_vertex_prototypes[0],
vertex2=_ligand_vertex_prototypes[1],
position=[0, 0.5, 0.1],
),
stk.Edge(
id=3,
vertex1=_metal_vertex_prototypes[1],
vertex2=_ligand_vertex_prototypes[1],
position=[0, -0.5, 0.1],
),
stk.Edge(
id=4,
vertex1=_metal_vertex_prototypes[0],
vertex2=_ligand_vertex_prototypes[2],
position=[-0.1, 0.5, 0],
),
stk.Edge(
id=5,
vertex1=_metal_vertex_prototypes[1],
vertex2=_ligand_vertex_prototypes[2],
position=[-0.1, -0.5, 0],
),
stk.Edge(
id=6,
vertex1=_metal_vertex_prototypes[0],
vertex2=_ligand_vertex_prototypes[3],
position=[0, 0.5, -0.1],
),
stk.Edge(
id=7,
vertex1=_metal_vertex_prototypes[1],
vertex2=_ligand_vertex_prototypes[3],
position=[0, -0.5, -0.1],
),
stk.Edge(
id=8,
vertex1=_metal_vertex_prototypes[0],
vertex2=_ligand_vertex_prototypes[4],
position=[0, 1.5, 0],
),
)
def main():
rdkit_opt = stko.MetalOptimizer(
metal_binder_distance=1.9,
metal_binder_forceconstant=1.0e2,
max_iterations=200,
)
fgs = [
stk.SmartsFunctionalGroupFactory(
smarts='[#6]~[#8]~[#1]',
bonders=(1, ),
deleters=(2, ),
),
stk.SmartsFunctionalGroupFactory(
smarts='[#6]~[#8X1]',
bonders=(1, ),
deleters=(),
),
]
ligand_bb = stk.BuildingBlock.init_from_file('org_carb.mol', fgs)
axial_bb = stk.BuildingBlock(
smiles='C(N)O',
functional_groups=(stk.AlcoholFactory(), ),
)
axial_bb.write('axial_bb.mol')
print(ligand_bb, axial_bb)
copper_atom = stk.BuildingBlock(
smiles='[Cu+2]',
functional_groups=(
stk.SingleAtom(stk.Cu(0, charge=2))
for i in range(5)
),
position_matrix=np.array([[0, 0, 0]]),
)
# Use a custom Paddlewheel topology that has axial positions.
copper_pw = stk.ConstructedMolecule(
AxialPaddlewheel(
metals={copper_atom: (0, 1)},
ligands={
ligand_bb: (0, 1, 2, 3),
axial_bb: (4, ),
},
optimizer=stk.MCHammer(),
)
)
copper_pw.write('copper_pw.mol')
copper_pw = rdkit_opt.optimize(copper_pw)
copper_pw.write('copper_pw_opt.mol')
copper_pw = stk.BuildingBlock.init_from_molecule(
copper_pw, (stk.BromoFactory(), )
)
ligand_bb = stk.BuildingBlock.init_from_file(
'org.mol', (stk.BromoFactory(), )
)
print(copper_pw, ligand_bb)
cage = stk.ConstructedMolecule(
topology_graph=stk.cage.M2L4Lantern(
building_blocks=(
copper_pw,
ligand_bb,
),
optimizer=stk.MCHammer(num_steps=2000),
)
)
cage.write('cage.mol')
cage = rdkit_opt.optimize(cage)
cage.write('cage_opt.mol')
cage_bb = stk.BuildingBlock.init_from_molecule(
molecule=cage,
functional_groups=(stk.PrimaryAminoFactory(), ),
)
linker_bb = stk.BuildingBlock(
smiles='C1(C(C(C1N)N)N)N',
functional_groups=(stk.PrimaryAminoFactory(), ),
)
print(cage_bb, linker_bb)
cof = stk.ConstructedMolecule(
topology_graph=stk.cof.Square(
building_blocks=(cage_bb, linker_bb),
lattice_size=(2, 2, 1),
# Setting scale_steps to False tends to lead to a
# better structure.
optimizer=stk.Collapser(scale_steps=False),
),
)
cof.write('cof.mol')
if __name__ == '__main__':
main()
Jaloca2
First of all, thank you for the excellent tutorials and stk code. I'm a beginner at that, interested in material science mainly in extended frameworks. I can form metal-organic cages based on paddle-wheel units (M2L4, M3L6). However, I cannot extend the cage in a coordination network using the axial position (my purpose is to use the cage as bb). I tried to put a Br\NH2 functional group with materials studio, and after calling the .mol file and trying a reaction, I obtain an error (I think related to the metal valence). I tried also to construct a paddle-wheel unit using the range(6) such as octahedral metal complexes, but I obtain the same error at the same point. Do you know how I can form an extended network using metal-organic cages as bb through the axial positions? A 2D extended framework for example (using cof code)
Thank you in advance.