MRSF TDDFT MECI Optimization - Open-Quantum-Platform/openqp GitHub Wiki
An Examples of a MRSF-TDDFT Minimum Energy Conical Intersection(MECI) Optimization calculation
[input]
system=
C -1.6699351346837055 0.1537249235528157 -1.5459803491111643
C -1.8079415266835852 -0.0386075716896284 -0.1602069788110266
H -2.6609567768367581 0.2572290722092156 -2.0290359598415040
H -1.2898503996116444 -0.7568524635289917 -2.0470428696820342
H -1.3096398768036397 0.6557118321425524 0.5396052278505126
H -2.3820842951209360 -0.7983813277099963 0.4308517619153288
runtype=meci
functional=bhhlyp
charge=0
method=tdhf
basis=6-31g*
[scf]
type=rohf
maxit=50
multiplicity=3
[tdhf]
type=mrsf
nstate=5
[optimize]
optimizer=bfgs
maxit=50
istate=1
jstate=2
energy_shift=1e-5
energy_gap=1e-4
rmsd_grad=5e-4
max_grad=1e-3
[input] Section
- system: This is the coordination of your system, your system's coordination can be achieved through two primary methods for OpenQP. The first method employs the Standard Cartesian format, as illustrated in the provided example. The second method utilizes the .XYZ format. To leverage the .XYZ format, save your coordination details in this format within your input folder, for instance, as H2O.xyz. Subsequently, it can be easily applied by specifying system=H2O.xyz in your input file. Here is an example of .XYZ format:
3
symmetry c1
O 0.000000000 0.000000000 -0.041061554
H -0.533194329 0.533194329 -0.614469223
H 0.533194329 -0.533194329 -0.614469223
-
charge: The total charge of the system. A value of 0 indicates that the molecule is neutral, with no net charge.
-
basis: The basis set used for the calculation, 6-31gs in this case. The 6-31gs is a split-valence basis set with polarization functions on heavy atoms, designed to provide a good balance between accuracy and computational cost. You can find all of the basis-sets supported by OpenQP within the basis_set folder. Note that the basis-set file names are like 6-31g* -> 6-31g*, 6-31g** -> 6-31g(d,p) so there are no * at all.
-
functional: Specifies the DFT functional to be used. bhhlyp is a hybrid functional that combines Hartree-Fock exchange with B88 exchange and LYP correlation. It's known for its good performance in a variety of systems.
-
runtype: Set to meci, indicating that the calculation aims to find a Minimum Energy Conical Intersection, a key feature in the study of excited-state dynamics and photochemistry.
-
method: Indicates the computational method, tdhf for time-dependent Hartree-Fock.
[scf] Section
-
type: The SCF calculation type, rohf for Restricted Open-Shell Hartree-Fock. ROHF methods are used for molecules with open-shell electronic configurations, providing a way to handle both closed-shell (paired electrons) and open-shell (unpaired electrons) components of the wavefunction.
-
maxit: Sets the maximum number of iterations for the SCF procedure to 50, ensuring the electronic structure converges to a stable solution.
[tdhf] Section
-
type: The type of time-dependent calculation, mrsf for mixed-reference spin-flip. This advanced approach is designed to accurately describe excited states, particularly those involving changes in electron spin states. It allows for the treatment of systems where single-reference methods might fail, especially for complex excited-state phenomena.
-
nstate: Considers five states in the calculation, likely including the ground state and four excited states, to comprehensively explore the system's electronic landscape.
[optimize] Section
-
optimizer: Chooses bfgs(Broyden–Fletcher–Goldfarb–Shanno algorithm), a robust optimization algorithm well-suited for finding local minima in complex potential energy surfaces.
-
maxit: Specifies the maximum number of optimization iterations as 100, allowing for extensive exploration of the energy landscape.
-
mep_maxit: Limits the MEP calculation iterations to 2, focusing the search on connecting the specified states efficiently.
-
istate: Targets the second excited state (istate=2) for the MEP calculation, aiming to explore the transition pathway from this specific state.
-
energy_shift, step_size, step_tol, rmsd_grad, max_grad: These parameters fine-tune the optimization process, adjusting the energy criteria, step sizes, and convergence thresholds to ensure accurate and efficient mapping of the Minimum Energy Path.