2.5 Complete isolated-water calculation
The inputs on this page are educational templates that have not been executed for this course. Use an authorized installation, verify version-specific options and establish your own convergence.
2.5.1 Define the physical question
Compute PBE electronic energy and forces for fixed neutral closed-shell water in vacuum. This is not liquid water, a molecular crystal, zero-point correction or free energy. The reasonable original geometry is not asserted to be optimized. Save complete water_sp.inp with local verified data files. The 500/60 Ry grid pair is an unvalidated starting choice.
&GLOBAL
PROJECT water_sp
RUN_TYPE ENERGY_FORCE
PRINT_LEVEL MEDIUM
&END GLOBAL
&FORCE_EVAL
METHOD QUICKSTEP
&DFT
BASIS_SET_FILE_NAME BASIS_MOLOPT
POTENTIAL_FILE_NAME GTH_POTENTIALS
CHARGE 0
MULTIPLICITY 1
&QS
METHOD GPW
EPS_DEFAULT 1.0E-12
&END QS
&MGRID
NGRIDS 4
CUTOFF [Ry] 500
REL_CUTOFF [Ry] 60
&END MGRID
&POISSON
PERIODIC NONE
POISSON_SOLVER MT
&END POISSON
&SCF
SCF_GUESS ATOMIC
EPS_SCF 1.0E-7
MAX_SCF 100
&OT
MINIMIZER DIIS
PRECONDITIONER FULL_ALL
&END OT
&END SCF
&XC
&XC_FUNCTIONAL PBE
&END XC_FUNCTIONAL
&END XC
&END DFT
&SUBSYS
&CELL
ABC [angstrom] 18.0 18.0 18.0
PERIODIC NONE
&END CELL
&COORD
UNIT angstrom
O 9.000000 9.000000 9.000000
H 9.758602 9.000000 9.504284
H 8.241398 9.000000 9.504284
&END COORD
&KIND O
BASIS_SET DZVP-MOLOPT-SR-GTH
POTENTIAL GTH-PBE-q6
&END KIND
&KIND H
BASIS_SET DZVP-MOLOPT-SR-GTH
POTENTIAL GTH-PBE-q1
&END KIND
&END SUBSYS
&PRINT
&FORCES ON
&END FORCES
&END PRINT
&END FORCE_EVAL
Verify exact H/O labels in your actual BASIS_MOLOPT and q1/q6 in GTH_POTENTIALS. Standard data provide these GTH-compatible labels, but a file merely containing “DZVP” is not interchangeable. Official first-calculation structural reference has different coordinates/settings.
2.5.2 Geometry and electrostatics have separate periodicity
CELL/PERIODIC NONE governs subsystem geometry; POISSON/PERIODIC NONE specifies electrostatic boundaries. Both are intentional. A large box with periodic electrostatics remains periodic. The Martyna–Tuckerman solver requires a cell at least twice the spatial charge-density extent for its exactness guarantee, including electronic tails rather than only nuclear bounding box. Test finite-box sensitivity. MT requirements.
2.5.3 Execute and inspect
cp2k -i water_sp.inp -o water_sp.out
status=$?
printf 'exit_status=%s\n' "$status"
grep -nE 'SCF run converged|ENERGY\||ATOMIC FORCES|PROGRAM ENDED' water_sp.out
Use the scheduler on clusters. Grep only locates evidence: read context/end and exclude stale logs. Confirm intended version/path, three atoms, H/O kinds, eight valence electrons, singlet, explicit SCF convergence, no later fatal error, total energy and three force vectors with consistent headers. Nonzero forces are expected before optimization. A small force-vector sum is a useful diagnostic, not proof that individual forces vanish.
Absolute energies use method-dependent pseudopotential references. Offsets from all-electron molecular codes do not by themselves diagnose failure. Record output units, model and actual values rather than expected-looking numbers.
2.5.4 Exercise
Create an evidence table for each acceptance check, retaining input, raw output, data hashes and version. Identify how ENERGY_FORCE differs from GEO_OPT. If unexecuted, annotate the complete input and leave energy/force fields blank.
Previous:Environment, GPW and the input tree · Next:SCF, grids and basis: separate tests