Understanding the CP2K output
General Output
Standard output
The beginning of the standard output contains information about the setup of the run. This will print the important input parameters and details of how CP2K was built and run e.g. the number of processes and threads that were specifed when running. It can be used for future reference.
After this part, the output will then print details of the individual setup for the important sections in the input file. These will appear under clear headings e.g. FIST for the MM section, Quickstep for the DFT section, and QMMM. Input parameters are reported, along with some more information such as for example the number of basis functions.
The report of the calculation then follows in the output file. It will start with the results from the self-consistent Kohn-Sham ground state calculation (SCF optimisation). This will look similar to the snippet below.
SCF WAVEFUNCTION OPTIMIZATION
----------------------------------- OT ---------------------------------------
Minimizer : DIIS : direct inversion
in the iterative subspace
using 7 DIIS vectors
safer DIIS on
Preconditioner : FULL_ALL : diagonalization, state selective
Precond_solver : DEFAULT
stepsize : 0.15000000 energy_gap : 0.08000000
eps_taylor : 0.10000E-15 max_taylor : 4
----------------------------------- OT ---------------------------------------
Step Update method Time Convergence Total energy Change
------------------------------------------------------------------------------
Decoupling Energy: -0.0015860332
Recoupling Energy: 0.0000002197
1 OT DIIS 0.15E+00 1261.9 0.00000101 -412.4425164003 -4.12E+02
Decoupling Energy: -0.0015862541
Recoupling Energy: 0.0000002197
2 OT DIIS 0.15E+00 1.0 0.00000216 -412.4425163985 1.73E-09
Decoupling Energy: -0.0015860586
Recoupling Energy: 0.0000002197
3 OT DIIS 0.15E+00 1.0 0.00000051 -412.4425164055 -6.98E-09
The OT block prints information about the orbital transform solver. This again, reports the options set in the input file.
The SCF steps are then printed as they are calculated, listing the Step number, Method, run Time, Convergence, Total Energy and the Change in energy from the last step. The first step will usually take longer as there are set up overheads.
In an ideal calculation the Total Energy should decrease with each step
as the energy is minimised (this means the Change should be negeative).
The Change in energy should also decrease as it converges to the final value.
The Convergence gives an idea of the accuracy of the SCF energy. When this value
becomes less than the EPS_SCF value specified in the input the SCF calculation will
exit and a message stating that convergence has been acheived is printed.
*** SCF run converged in 3 steps ***
Otherwise, if the SCF does not converge, it continues until the number of steps equals the MAX_SCF in the
input file. The inner SCF loop will finish and then move to the next outer SCF step,
recalculating the preconditioner. Finally, if the number of outer SCF steps equals to
the corresponding MAX_SCF value in the input file, then the SCF calculation has not converged,
a warning will be printed to the standard output and the SCF will be abandoned.
Information on how to handle this is given in the troubleshooting chapter.
Upon successful convergence, electronic densities and energies (and forces if specified) information is printed.
Electronic density on regular grids: -213.9999999972 0.0000000028
Core density on regular grids: 213.9999999999 -0.0000000001
Total charge density on r-space grids: 0.0000000026
Total charge density g-space grids: 0.0000000026
Overlap energy of the core charge distribution: 0.00000923326382
Self energy of the core charge distribution: -943.25295781363013
Core Hamiltonian energy: 284.50150313570356
Hartree energy: 355.46908698187769
Exchange-correlation energy: -85.84000086143999
Hartree-Fock Exchange energy: -18.79624573644062
QM/MM Electrostatic energy: -4.52391134484379
Total energy: -412.44251640550942
outer SCF iter = 1 RMS gradient = 0.51E-06 energy = -412.4425164055
outer SCF loop converged in 1 iterations or 3 steps
ENERGY| Total FORCE_EVAL ( QMMM ) energy (a.u.): -2134.963290942846925
It is a good idea to check that the electronic density corresponds to the number of electrons, and that the charge is as expected.
The Total energy given is the energy from only the QM part i.e from the SCF calculation. A breakdown of its components is printed above it. The total number of outer SCF loops and inner SCF steps that were done is also shown.
The ENERGY (QMMM)` is the QM/MM energy including all its components; the QM energy, MM energy
and QM-MM interation energy. This is the energy you are usually interested in.
Wavefuntions - NAME-RESTART.wfn
Wavefunction files are binary files that contain the wavefunctions obtained from the last SCF step.
They are named with the project_name preceeding '-RESTART.wfn'.
One is written every SCF step, and if a wavefuntion file of the same name
already exists the older version is moved to NAME-RESTART.wfn.bak-1, rather than overwritten.
This is done for up to three files and so you may see the following files, where
the third backup (bak-3) is the oldest.
NAME-RESTART.wfn
NAME-RESTART.wfn.bak-1
NAME-RESTART.wfn.bak-2
NAME-RESTART.wfn.bak-3
Wavefunction restarts are used when restarting a calculation in order to act
as a guide for the first SCF step to speed up the calculation.
In this case the project name should be consistent
and the SCF_GUESS should be set to ‘restart’. Care should be taken that
the wavefunction is a suitable guess for the SCF calculation otherwise it may not
converge.
Output from an MD run
.ener file
The .ener file shows the important information about the system as the simulation progresses. For each step the Temperature, Kinetic energy, Potential energy, CPU time and the Conserved Quantity are printed. The Conserved Quantity can represent different quantities according to the statistical ensemble we are sampling on. For example, if we are running an NVE simulation the Conserved Quantity corresponds to the total energy of the system of interest, while if we run an NVT simulation then the Conserved Quantity is the total energy of a system that includes the system of interest and also the thermostat degrees of freedom.
MD trajectory
This will look similar to the geometry trajectory from a geometry optimation and will appear as a .xyz file as default. The coordinates at each MD step are printed in order in this file. You can control how often this file is updated with new coordinates by using the following settings within the motion section:
&MOTION
&PRINT
&TRAJECTORY
&EACH
MD 10 ! print every 10 steps
&END EACH
&END TRAJECTORY
&END PRINT
&END MOTION
.restart file
The .restart file is written at the end of an MD step and it contains the information to restart the MD simulation from that step. This file is human readable and reports information about the system setup along with the current positions and velocities of the atoms in the system.
To restart an MD simulation the EXT_RESTART section has to be added to the input file to instruct the code
to use the .restart file as the restart point:
&EXT_RESTART
EXTERNAL_FILE_NAME project-1.restart
&END EXT_RESTART
You can also use the .restart file as if the input file for restarting the simulation,
supplying it directly to the CP2K executable in the job submission script:
cp2k.psmp –i project-1.restart