TDDFT-ris 2.0 更新，200原子体系无压力, 支持Multiwfn - 计算化学公社

git clone git@github.com:John-zzh/pyscf_TDDFT_ris.git

复制代码

export PYTHONPATH=absolue_path_to_ris_repo:$PYTHONPATH

复制代码

conda create -n ris-mokit-pyscf-py39 python=3.9
conda activate ris-mokit-pyscf-py39
conda install mokit pyscf -c mokit/label/cf -c conda-forge

复制代码

%chk=gau.chk
%nprocs=8
%mem=16GB
#p PBE1PBE/def2SVP
retinal
0 1
xyz略...
g16 gau.gjf

复制代码

conda activate ris-mokit-pyscf-py39
export PYTHONPATH=absolue_path_to_ris_repo:$PYTHONPATH
export MKL_NUM_THREADS=4
export OMP_NUM_THREADS=4
python absolute_path_to_ris_repo/main.py -f gau.fch -func pbe0 -n 5 -M 4000 -pt 0.0001

复制代码

orca_2mkl molecule -mkl
mkl2fch molecule.mkl molecule.fch

复制代码

args.nroots 5
Woring directory: /home/jojo/ris/gaussian_multiwfn/TDDFT
nbf = 434
nif = 434
mo_coeff.shape (434, 434)
mo_energy.shape (434,)
Restricted Kohn-Sham
functional: pbe0
get_mf_from_fch_time: 0.73 seconds
This job can use: 24CPUs
use new code
use different J and K fitting basis: J with sp and K with s
self.nroots 5
use single precision? True
loading default XC functional paramters from parameter.py
use hybrid XC functional
self.a_x 0.25
cartesian or spherical electron integral = _sph
n_occ = 78
n_vir = 356
MO truncation in K with threshold 40 eV above HOMO and below LUMO
rest_occ = 57
rest_vir = 169
==================== RIJ ====================
Asigning minimal auxiliary basis set: sp
The exponent alpha: 0.2/R^2
auxmol.cart = False
Full eri3c in shape (np.int64(434), np.int64(434), np.int64(112)), will take 161 MB
max_mem_for_one_batch 2000.0
small 3c2e, just incore
pre_T_pq time: 0.1 seconds
pre_T_pq_to_Tpq one batch time 0.0 seconds
T_ia_J time 0.5 seconds
==================== RIK ====================
Asigning minimal auxiliary basis set: s
The exponent alpha: 0.2/R^2
auxmol.cart = False
Full eri3c in shape (np.int64(434), np.int64(434), np.int64(49)), will take 70 MB
max_mem_for_one_batch 2000.0
small 3c2e, just incore
pre_T_pq time: 0.1 seconds
pre_T_pq_to_Tpq one batch time 0.0 seconds
pre_T_pq time: 0.0 seconds
pre_T_pq_to_Tpq one batch time 0.0 seconds
pre_T_pq time: 0.0 seconds
pre_T_pq_to_Tpq one batch time 0.0 seconds
T_ia_K T_ij_K T_ab_K time 0.3 seconds
======= TDDFT Eigen Solver Statrs =======
size of A matrix = 27768
Using non-orthogonalized Krylov subspace (nKs) method.
Furche, Filipp, Brandon T. Krull, Brian D. Nguyen, and Jake Kwon.
Accelerating molecular property calculations with nonorthonormal Krylov space methods.
The Journal of Chemical Physics 144, no. 17 (2016).
iter: 1 max|R|: 1.27e-01 new_vectors: 10 MVP: 0.1 seconds
iter: 2 max|R|: 3.37e-02 new_vectors: 5 MVP: 0.1 seconds
iter: 3 max|R|: 1.28e-02 new_vectors: 5 MVP: 0.1 seconds
iter: 4 max|R|: 5.19e-03 new_vectors: 5 MVP: 0.1 seconds
iter: 5 max|R|: 1.70e-03 new_vectors: 4 MVP: 0.1 seconds
iter: 6 max|R|: 9.69e-04 new_vectors: 1 MVP: 0.0 seconds
Finished in 6 steps, 0.72 seconds
final subspace = 30
max_norm = 9.69e-04
MVcost 0.4901s 67.86%
fill_holder_cost 0.0892s 12.35%
subgencost 0.0266s 3.69%
subcost 0.0267s 3.69%
full_cost 0.0362s 5.01%
======= TDDFT Eigen Solver Done =======
check norm of X^TX - Y^YY - I = 4.38e-08
print RKS transition coefficients larger than 1.00e-01
index of HOMO: 78
index of LUMO: 79
Excited State 1: Singlet-A 2.9053 eV 427.04 nm f=0.0001 <S**2>=0.000
76 -> 79 0.65193
76 -> 80 -0.23655
Excited State 2: Singlet-A 2.9834 eV 415.87 nm f=1.3705 <S**2>=0.000
77 -> 79 -0.14597
78 -> 79 0.69388
Excited State 3: Singlet-A 3.8900 eV 318.95 nm f=0.4812 <S**2>=0.000
77 -> 79 -0.64082
78 -> 79 -0.15515
78 -> 80 -0.23941
Excited State 4: Singlet-A 4.3099 eV 287.87 nm f=0.0396 <S**2>=0.000
75 -> 79 0.35318
77 -> 79 0.19885
78 -> 80 -0.57193
Excited State 5: Singlet-A 4.9689 eV 249.69 nm f=0.0002 <S**2>=0.000
76 -> 79 0.27008
76 -> 80 0.59980
76 -> 81 -0.19345
76 -> 83 -0.11782
transition coefficient data also written to gau_TDDFT_ris_coeff_Multiwfn.txt
================================================
eV nm cm^-1 oscillator strength
2.905 427 23433 0.00012678
2.983 416 24063 1.37051890
3.890 319 31375 0.48124768
4.310 288 34762 0.03955060
4.969 250 40077 0.00018934
spectra data also written to gau_TDDFT_ris_eV_os_Multiwfn.txt
total ris time: 2.48 seconds
Please cite the TDDFT-ris method:
1. Zhou, Zehao, Fabio Della Sala, and Shane M. Parker.
Minimal auxiliary basis set approach for the electronic excitation spectra
of organic molecules. The Journal of Physical Chemistry Letters
14, no. 7 (2023): 1968-1976.
(must cite)
2. Zhou, Zehao, and Shane M. Parker.
Converging Time-Dependent Density Functional Theory Calculations in Five Iterations
with Minimal Auxiliary Preconditioning. Journal of Chemical Theory and Computation
20, no. 15 (2024): 6738-6746.
(for efficient orbital truncation technique)
2. Giannone, Giulia, and Fabio Della Sala.
Minimal auxiliary basis set for time-dependent density functional theory and
comparison with tight-binding approximations: Application to silver nanoparticles.
The Journal of Chemical Physics 153, no. 8 (2020).
(The idea of TDDFT-ris originates from TDDFT-as)
And cite the pyscf-TDDFT-ris package:
1. Zehao Zhou, pyscf-TDDFT-ris, (https://github.com/John-zzh/pyscf_TDDFT_ris)
2. Jingxiang Zou, Molecular Orbital Kit (MOKIT), (https://gitlab.com/jxzou/mokit)
3. PySCF: the Python-based simulations of chemistry framework,
Q. Sun, et. al., and G. K.-L. Chan, WIREs Comput. Mol. Sci. 8, e1340 (2018)
(https://pyscf.org/about.html)

复制代码