regarding to geometry optimization

[arpit]

hello sir
my question is about optimization..i am try to optimize a geometry and the out primitive vectors are always same as my input lattice parameters....second time i chance the lattice parameters for the same geometry and again the resultant parameters are same as my input parameters....so my question is that how we can optimize our geometry??

[Anders Blom]

If you set up the calculation via the Scripter, the default is to constrain the lattice vectors. You should remove the corresponding cell constraints, and then try again.

[arpit]

sir to remove the cell constraints....what i have to do....plz explain it???

[Anders Blom]

In the Scripter, under Geometry Optimization, remove the tick from the checkbox "Constrain cell".

[arpit]

sir the output is same.....i m not getting the proper result......now what else i can do ...plz suggest me......

[Anders Blom]

Impossible to say without seeing your script

[arpit]

sir actually i m optimize the geometry of graphene nano-ribbon having chiral vector n=3, m=0 and repetition=2...............
my input lattice parameters are a=10 , b= 18.0595, c=8.52516 all in Ang..and then i move to scripter  where in new calculator i select ATK-DFT and then select the option optimize geometry..in which i remove the tick mark as you suggest me....for more detail i m attaching my output script...plz help me i m realy in a fix...............

[Anders Blom]

Although you didn't post your script, but the log file, it can be seen from it that the stress on the initial cell is smaller than the stress tolerance. Thus there is no reason for the program to change the cell.

[arpit]

sir here is my script...i have an other question if i reduce the unit cell size then is their any change on the stability... ?


# -------------------------------------------------------------
# Bulk configuration
# -------------------------------------------------------------

# Set up lattice
vector_a = [10.0, 0.0, 0.0]*Angstrom
vector_b = [0.0, 18.0594922152, 0.0]*Angstrom
vector_c = [0.0, 0.0, 8.52516]*Angstrom
lattice = UnitCell(vector_a, vector_b, vector_c)

# Define elements
elements = [Carbon, Carbon, Carbon, Carbon, Carbon, Carbon, Carbon, Carbon,
            Carbon, Carbon, Carbon, Carbon, Hydrogen, Hydrogen, Hydrogen,
            Hydrogen, Carbon, Carbon, Carbon, Carbon, Carbon, Carbon, Carbon,
            Carbon, Carbon, Carbon, Carbon, Carbon, Hydrogen, Hydrogen,
            Hydrogen, Hydrogen]

# Define coordinates
cartesian_coordinates = [[  5.        ,   5.95349397,   2.84172   ],
                         [  5.        ,   7.18399482,   3.55215   ],
                         [  5.        ,   5.95349397,   1.42086   ],
                         [  5.        ,   8.41449568,   2.84172   ],
                         [  5.        ,   9.64499654,   3.55215   ],
                         [  5.        ,   7.18399482,   0.71043   ],
                         [  5.        ,   8.41449568,   1.42086   ],
                         [  5.        ,  10.87549739,   2.84172   ],
                         [  5.        ,  12.10599825,   3.55215   ],
                         [  5.        ,   9.64499654,   0.71043   ],
                         [  5.        ,  10.87549739,   1.42086   ],
                         [  5.        ,  12.10599825,   0.71043   ],
                         [  5.        ,   5.        ,   3.39222   ],
                         [  5.        ,   5.        ,   0.87036   ],
                         [  5.        ,  13.05949222,   3.00165   ],
                         [  5.        ,  13.05949222,   1.26093   ],
                         [  5.        ,   5.95349397,   7.1043    ],
                         [  5.        ,   7.18399482,   7.81473   ],
                         [  5.        ,   5.95349397,   5.68344   ],
                         [  5.        ,   8.41449568,   7.1043    ],
                         [  5.        ,   9.64499654,   7.81473   ],
                         [  5.        ,   7.18399482,   4.97301   ],
                         [  5.        ,   8.41449568,   5.68344   ],
                         [  5.        ,  10.87549739,   7.1043    ],
                         [  5.        ,  12.10599825,   7.81473   ],
                         [  5.        ,   9.64499654,   4.97301   ],
                         [  5.        ,  10.87549739,   5.68344   ],
                         [  5.        ,  12.10599825,   4.97301   ],
                         [  5.        ,   5.        ,   7.6548    ],
                         [  5.        ,   5.        ,   5.13294   ],
                         [  5.        ,  13.05949222,   7.26423   ],
                         [  5.        ,  13.05949222,   5.52351   ]]*Angstrom

# Set up configuration
bulk_configuration = BulkConfiguration(
    bravais_lattice=lattice,
    elements=elements,
    cartesian_coordinates=cartesian_coordinates
    )

# -------------------------------------------------------------
# Calculator
# -------------------------------------------------------------
calculator = LCAOCalculator()

bulk_configuration.setCalculator(calculator)
nlprint(bulk_configuration)
bulk_configuration.update()
nlsave('analysis.nc', bulk_configuration)

bulk_configuration = OptimizeGeometry(
        bulk_configuration,
        maximum_forces=0.05*eV/Ang,
        maximum_stress=0.05*eV/Ang**3,
        trajectory_filename=None,
        )
nlsave('analysis.nc', bulk_configuration)
nlprint(bulk_configuration)

[Anders Blom]

Not sure I understand. Stability of calculations, or of structure? And reduce in which direction? And why?

[arpit]

sir actually i m changing "a" and by changing "a" i m calculating the energy of the structure and ploting energy vs volume curve......according to my knowledge the minima of that curve will tell us the most stable size of unit cell........but this process takes a lot of time...so i tried optimization...so that i will directly get the optimized and most stable structure of that cell.sir i want to know that whether the GNR  is stable on default parameters can i quote these parameters in my paper......(stability of structure)

[Anders Blom]

As mentioned already, the stress on the cell is already rather small. If you want to relax it to a smaller value, then lower the tolerance and run like you did.

[kstokbro]

your geometry is periodic in the z direction. if you want to optimize by hand  you must calculate the energy as function of the length of the z vector.

[arpit]

sir thank you but i am a new user can you plz tell how i calculate the energy as function of the length of the z vector...

[Anders Blom]

sir actually i m changing "a" and by changing "a" i m calculating the energy of the structure and ploting energy vs volume curve

Change vector_c instead. But you should use fractional coordinates then, otherwise the atoms may end up outside the cell. It's best to use the stress optimization, just lower the tolerance.