CP2K Open Source Molecular Dynamics

User Tools

Site Tools

Trace:
howto:gfn1xtb

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
howto:gfn1xtb [2022/07/19 12:03] – ahehnhowto:gfn1xtb [2024/01/03 13:22] (current) – oschuett
Line 1: Line 1:
-====== How to run a GFN1-xTB calculation ====== +This page has been moved to: https://manual.cp2k.org/trunk/methods/semiempiricals/xtb.html
- +
-This is a short tutorial on how to run GFN1-xTB computations. The details on the theory and the original implementation by Grimme can be found in [[https://pubs.acs.org/doi/full/10.1021/acs.jctc.7b00118]]. +
-Please cite this paper if you were to use the GFN1-xTB module. +
- +
-===== Brief theory recap ===== +
- +
-The semi-empirical GFN1-xTB energy expression comprises contributions due to electronic (EL), atom-pairwise repulsion (REP), dispersion (DISP), and halogen-bonding (XB) terms, +
- +
-\begin{equation}\label{gfnxtb1_plus_nonbonded} +
-\begin{aligned} +
-E_{\rm{\tiny{GFN1-xTB}}} = E_{\rm{\tiny{EL}}} + E_{\rm{\tiny{REP}}} + E_{\rm{\tiny{DISP}}} + E_{\rm{\tiny{XB}}} \, . +
-\end{aligned} +
-\end{equation} +
- +
-1. The electronic energy contribution, +
- +
-\begin{equation}\label{electronic_energy} +
-\begin{aligned} +
-E_{\rm{\tiny{EL}}} =  \sum_i^{\rm{\tiny{occ}}} n_i \langle \Psi_i | h_0 | \Psi_i \rangle + \frac{1}{2} \sum_{A,B} \sum_{{l}^A}\sum_{{l'}^B} p_l^A p_{{l'}}^B \gamma_{AB,ll'} + \frac{1}{3}\sum_{A} \Gamma_A q_A^3 - T_{\rm{\tiny{el}}} S_{\rm{\tiny{el}}} \, , +
-\end{aligned} +
-\end{equation} +
- +
-contains zeroth-order contributions based on a zeroth-order Hamiltonian $h_0$, the valence molecular orbitals $\Psi_i$, occupation numbers $n_i$ as well as second-order contributions which are optimized self-consistently as well as third-order diagonal contributions. +
-The second order contributions are described using the semi-empirical electron repulsion operator $\gamma_{AB,ll'}$ which depends on the interatomic distance of atoms $A$ and $B$ as well as further empirical parameters that are specific for different angular momenta $l$ and $l'$. The monopole charges of the second-order expression are optimized self-consistently, +
- +
-\begin{equation}\label{scc_charges} +
-\begin{aligned} +
-p_l^A = p_l^{A_0} - \sum_{\nu}^{N_{\rm{\tiny{AO}}}} \sum_{\mu \in A, \mu \in l} S_{\mu \nu } P_{\mu \nu} \, , +
-\end{aligned} +
-\end{equation} +
- +
-referring to the atomic orbital overlap matrix $\mathbf{S}$ and the density matrix $\mathbf{P}$. +
- +
-The remaining diagonal terms represent a cubic charge correction based on the Mulliken charge $q_A$ of atom $A$ and the charge derivative $\Gamma_A$ of the atomic Hubbard parameter $\eta_A$. +
-Furthermore, the electronic temperature times entropy term $T_{\rm{\tiny{el}}}S_{\rm{\tiny{el}}}$ enables fractional orbital occupations. +
-  +
-2. Repulsion is described via an atom-pairwise potential, +
- +
-\begin{equation}\label{repulsion} +
-\begin{aligned} +
-E_{\rm{\tiny{REP}}} = \sum_{AB} \frac{Z_A^{\rm{\tiny{eff}}} Z_B^{\rm{\tiny{eff}}} }{R_{AB}} \exp^{- (\alpha_A \alpha_B)^{1/2} (R_{AB})^{k_f}} \, , +
-\end{aligned} +
-\end{equation} +
-with the effective nuclear charge $\mathbf{Z}^{\rm{\tiny{eff}}}$ as well as the global or element-specific parameters $k_f$ and $\alpha$. +
- +
-3. Dispersion is included by the well-established D3 method in the BJ-damping scheme[[https://aip.scitation.org/doi/10.1063/1.3382344]]. +
- +
-4. Corrections for element-specific interactions are possible using either a halogen-bonding correction term (XB) or a generic nonbonding potential correction (NONBOND). Note that the generic nonbonding potential correction is CP2K specific and thus the so-obtained energy differs from the original GFN1-xTB method, +
- +
-\begin{equation}\label{gfnxtb1_energy_expression} +
-\begin{aligned} +
-E_{\rm{\tiny{GFN1-xTB+NONBOND}}} = E_{\rm{\tiny{GFN1-xTB}}} + E_{\rm{\tiny{NONBOND}}} \, . +
-\end{aligned} +
-\end{equation} +
- +
- +
-===== The GFN1-xTB input section ===== +
- +
-The most important keywords and subsections of section ''xTB'' are: +
-  * ''DO_EWALD'': keyword to activate Ewald summation for periodic boundary conditions (PBC); has to be switched to true in case of PBC +
-  * ''USE_HALOGEN_INTERACTION'': keyword to switch off contribution $E_{\rm{\tiny{XB}}}$ to correct halogen interactions, default is to include this correction +
-  * ''CHECK_ATOMIC_CHARGES'': the cubic charge diagonal contribution is checked to be numerically stable by switching the keyword to true.  +
-  * ''DO_NONBONDED'': add a generic correction potential to correct bond- or atomic-specific interactions +
-  * ''PARAMETER'': it is possible to add this section with corresponding keywords to modify xTB parameters +
- +
-The additional keywords ''COULOMB_INTERACTION'', ''COULOMB_LR'' and ''TB3_INTERACTION'' are for debugging purposes only and it is recommended to use the default options here. +
- +
- +
-===== Simple examples ===== +
- +
-==== GFN1-xTB ground-state energy for  ==== +
- +
-The following input is an examplary standard input for calculating GFN1-xTB ground-state energies.  +
- +
-<code - periodic.inp> +
- +
-&GLOBAL +
-  RUN_TYPE  ENERGY +
-  PROJECT_NAME xtb +
-  PRINT_LEVEL  MEDIUM +
-  PREFERRED_DIAG_LIBRARY SL +
-&END GLOBAL +
-&FORCE_EVAL +
- METHOD QS +
-&DFT +
-  &QS +
-   METHOD XTB +
-   &XTB +
-    CHECK_ATOMIC_CHARGES F               ! Keyword to check if Mulliken charges are physically reasonable +
-    DO_EWALD  T                          ! Ewald summation is required for periodic structures +
-    USE_HALOGEN_CORRECTION T             ! Element-specific correction for halogen interactions (Cl, Br) with (O, N)      +
-   &END XTB +
-  &END QS +
-  &SCF +
-   SCF_GUESS RESTART +
-   MAX_SCF 50 +
-   EPS_SCF 1.E-6 +
-   &OT ON +
-     PRECONDITIONER FULL_SINGLE_INVERSE +
-     MINIMIZER DIIS +
-   &END +
-   &OUTER_SCF +
-     MAX_SCF 200 +
-     EPS_SCF 1.E-6 +
-   &END OUTER_SCF +
-  &END SCF +
- &END DFT +
-  &SUBSYS +
-    &TOPOLOGY +
-      COORD_FILE_FORMAT  xyz +
-      COORD_FILE_NAME  input.xyz +
-      CONNECTIVITY OFF +
-      &CENTER_COORDINATES +
-      &END CENTER_COORDINATES +
-     &END TOPOLOGY +
-    &CELL +
-      ABC  21.64 21.64 21.64 +
-      ALPHA_BETA_GAMMA 90.0 90.0 90.0 +
-      PERIODIC XYZ +
-    &END CELL +
-  &END SUBSYS +
-&END FORCE_EVAL  +
- +
-</code> +
- +
-The so-obtained output is listing information on the chosen system-specific parameters. Note that parameters can be changed manually by adding a ''PARAMETER'' section to the ''xTB'' section and specifying corresponding keywords for the specific parameters with the adjusted values or by giving the path to the modified parameter file, adding the keywords ''PARAM_FILE_PATH pathname'' and ''PARAM_FILE_NAME filename''.  +
- +
-<code - cp2k> +
- +
-                  #####   #####        #          ####### ######    +
-                 #     # #     #      #              #    #     #   +
-                 #     # #           #    ##   ##    #    #     #   +
-                 #     #  #####     #      ## ##     #    ######    +
-                 #   # #       #   #        ###      #    #     #   +
-                 #    #  #     #  #        ## ##     #    #     #   +
-                  #### #  #####  #        ##   ##    #    ######    +
-                                                                    +
- +
- xTB| Parameter file                                              xTB_parameters +
- xTB| Basis expansion STO-NG                                                   6 +
- xTB| Basis expansion STO-NG for Hydrogen                                      4 +
- xTB| Halogen interaction potential                                            F +
- xTB| Halogen interaction potential cutoff radius                         20.000 +
- xTB| Nonbonded interactions                                                   F +
- xTB| D3 Dispersion: Parameter                                         dftd3.dat +
- xTB| Huckel constants ks kp kd                        1.850     2.250     2.000 +
- xTB| Huckel constants ksp k2sh                                  2.080     2.850 +
- xTB| Mataga-Nishimoto exponent                                            2.000 +
- xTB| Repulsion potential exponent                                         1.500 +
- xTB| Coordination number scaling kcn(s) kcn(p) kc     0.006    -0.003    -0.005 +
- xTB| Electronegativity scaling                                           -0.007 +
- xTB| Halogen potential scaling kxr kx2                          1.300     0.440 +
- +
-</code> +
- +
-Analogously to any other self-consistent field optimization (SCF) method, the output also includes the energy and convergence during the SCF steps with the finally converged GFN1-xTB energy. +
- +
-<code - cp2k> +
- +
- SCF WAVEFUNCTION OPTIMIZATION +
- +
-  ----------------------------------- OT --------------------------------------- +
-  Minimizer      : DIIS                : direct inversion +
-                                         in the iterative subspace +
-                                         using   7 DIIS vectors +
-                                         safer DIIS on +
-  Preconditioner : FULL_SINGLE_INVERSE : inversion of  +
-                                         H + eS - 2*(Sc)(c^T*H*c+const)(Sc)^T +
-  Precond_solver : DEFAULT +
-  stepsize       :    0.08000000                  energy_gap     :    0.08000000 +
-  eps_taylor     :   0.10000E-15                  max_taylor     :             4 +
-  ----------------------------------- OT --------------------------------------- +
- +
-  Step     Update method      Time    Convergence         Total energy    Change +
-  ------------------------------------------------------------------------------ +
-     1 OT DIIS     0.80E-01    0.5     0.01213502      -947.7483409153 -9.48E+02 +
-     2 OT DIIS     0.80E-01    0.3     0.00675007      -951.5762826800 -3.83E+00 +
-     3 OT DIIS     0.80E-01    0.3     0.00092877      -953.2164544959 -1.64E+00 +
-     4 OT DIIS     0.80E-01    0.3     0.00034159      -953.2591478247 -4.27E-02 +
-     5 OT DIIS     0.80E-01    0.3     0.00018348      -953.2687102329 -9.56E-03 +
-     6 OT DIIS     0.80E-01    0.3     0.00009265      -953.2707750500 -2.06E-03 +
-     7 OT DIIS     0.80E-01    0.3     0.00005495      -953.2714236504 -6.49E-04 +
-     8 OT DIIS     0.80E-01    0.3     0.00002612      -953.2716704946 -2.47E-04 +
-     9 OT DIIS     0.80E-01    0.3     0.00001585      -953.2717390500 -6.86E-05 +
-    10 OT DIIS     0.80E-01    0.3     0.00001020      -953.2717664315 -2.74E-05 +
-    11 OT DIIS     0.80E-01    0.3     0.00000564      -953.2717774258 -1.10E-05 +
-    12 OT DIIS     0.80E-01    0.3     0.00000354      -953.2717818198 -4.39E-06 +
-    13 OT DIIS     0.80E-01    0.3     0.00000206      -953.2717839406 -2.12E-06 +
-    14 OT DIIS     0.80E-01    0.3     0.00000127      -953.2717844831 -5.42E-07 +
-    15 OT DIIS     0.80E-01    0.3     0.00000077      -953.2717846336 -1.51E-07 +
- +
-  *** SCF run converged in    15 steps *** +
- +
- +
-  Core Hamiltonian energy:                                   -962.45147378153547 +
-  Repulsive potential energy:                                   8.84897617161771 +
-  Electronic energy:                                            0.76461561909348 +
-  DFTB3 3rd order energy:                                       0.33228335538302 +
-  Dispersion energy:                                           -0.76618599817727 +
- +
-  Total energy:                                              -953.27178463361872 +
- +
-  outer SCF iter =    1 RMS gradient =   0.77E-06 energy =       -953.2717846336 +
-  outer SCF loop converged in   1 iterations or   15 steps +
- +
-</code>  +
- +
-==== Adding a generic correction potential ==== +
- +
-It is possible to add a generic non bonded correction potential. The potential form can be chosen freely and needs to be specified by adding the keyword ''FUNCTION''. Included parameters and variables have to be specified using the keywords ''VARIABLES'' and ''PARAMETERS''. The section can be repeated as often as required and enables to include pairwise, element-specific correction potentials. The implementation also features analytic gradients for this option.    +
- +
-<code - periodic_with_generic_correction_potential.inp> +
- +
-&GLOBAL +
-  RUN_TYPE  ENERGY +
-  PROJECT_NAME xtb +
-  PRINT_LEVEL  MEDIUM +
-  PREFERRED_DIAG_LIBRARY SL +
-&END GLOBAL +
-&FORCE_EVAL +
- METHOD QS +
-&DFT +
-  &QS +
-   METHOD XTB +
-   &XTB +
-    CHECK_ATOMIC_CHARGES F               +
-    DO_EWALD  T                          +
-    USE_HALOGEN_CORRECTION T                   +
-    DO_NONBONDED T                      ! Possible option to include a generic non-bonded potential +
-     &NONBONDED                         ! Specification of the potential, keyword can be repeated  +
-      &GENPOT +
-       ATOMS Kr Br +
-       FUNCTION Aparam*exp(-Bparam*r)-Cparam/r**8  ! Potential formula has to be specified +
-       PARAMETERS Aparam Bparam Cparam             ! Parameters included in the formula above +
-       VALUES 70.0 1.0 0.0                         ! Explicit values for the parameters  +
-       VARIABLES r +
-       RCUT 40.5 +
-      &END GENPOT +
-     &END NONBONDED +
-   &END XTB +
-  &END QS +
-  &SCF +
-   SCF_GUESS RESTART +
-   MAX_SCF 50 +
-   EPS_SCF 1.E-6 +
-   &OT ON +
-     PRECONDITIONER FULL_SINGLE_INVERSE +
-     MINIMIZER DIIS +
-   &END +
-   &OUTER_SCF +
-     MAX_SCF 200 +
-     EPS_SCF 1.E-6 +
-   &END OUTER_SCF +
-  &END SCF +
- &END DFT +
-  &SUBSYS +
-    &TOPOLOGY +
-      COORD_FILE_FORMAT  xyz +
-      COORD_FILE_NAME  input.xyz +
-      CONNECTIVITY OFF +
-      &CENTER_COORDINATES +
-      &END CENTER_COORDINATES +
-     &END TOPOLOGY +
-    &CELL +
-      ABC  21.64 21.64 21.64 +
-      ALPHA_BETA_GAMMA 90.0 90.0 90.0 +
-      PERIODIC XYZ +
-    &END CELL +
-  &END SUBSYS +
-&END FORCE_EVAL  +
- +
-</code> +
- +
- +
howto/gfn1xtb.1658232223.txt.gz · Last modified: 2022/07/19 12:03 by ahehn