6.2 NVT and NpT answer different ensemble questions

These are unexecuted teaching inputs and starting models. Original diagrams are schematics, not calculated results. Validate version-specific syntax, licensed or authorized data, numerical convergence and the scientific model before using this workflow.

6.2.1 Model, units and provenance

Keep basis/potential files and executable versions traceable. Grid controls use Ry; common energy/force outputs use hartree and hartree/bohr. Read the unit in each output heading.

Shared inputs, conventions and evidence

Original schematic: NVT and NpT answer different ensemble questions. No numerical results are claimed.

6.2.2 Unexecuted inputs and explicit deltas

Use the accompanying instructions to identify the parent calculation and placement of every delta; a snippet is not automatically a standalone input. Preserve all blank-line and file-provenance requirements.

6.2.2.1 Input block 1

&MD
  ENSEMBLE NVT
  STEPS 2000
  TIMESTEP [fs] 0.5
  TEMPERATURE [K] 300
  &THERMOSTAT
    TYPE CSVR
    REGION GLOBAL
    &CSVR
      TIMECON [fs] 100
    &END CSVR
  &END THERMOSTAT
&END MD

6.2.2.2 Input block 2

&BAROSTAT
  PRESSURE [bar] 1.0
  TIMECON [fs] 1000
&END BAROSTAT

6.2.3 Worked investigation

Intuition. A thermostat targets a temperature distribution, not a perfectly flat temperature line. A barostat lets the volume respond to pressure and therefore depends on reliable stress. NVT at a chosen density and NpT at a chosen pressure are different physical constraints. Use case 21's stable force/timestep settings, periodic bulk water, and stress convergence before attempting NpT.

Original NVT replacement under MOTION/MD. Use this in place of the NVE MD section, preserving trajectory output. The coupling time is a starting value, not a universal optimal thermostat strength.

See input block 1 above.

NpT delta. After validating stress and an appropriate initial density, change ENSEMBLE to NPT_I for an isotropic bulk-cell experiment, add STRESS_TENSOR ANALYTICAL under FORCE_EVAL, and add BAROSTAT under MD:

See input block 2 above.

Confirm the build/method supports the requested ensemble. A fully flexible NPT_F cell is a different experiment and can be inappropriate for a small liquid cell or a surface. Save cell vectors at the same cadence as coordinates using MOTION/PRINT/CELL with EACH/MD 1.

Workflow. 1. Establish static force and pressure sanity before dynamics. 2. Run NVT equilibration and monitor temperature, potential energy and structural statistics. 3. Compare a second thermostat coupling time to assess sensitivity of equilibrium averages; stronger coupling may distort dynamics even if averages are acceptable. 4. For NpT, monitor volume, density and pressure, allowing for large instantaneous fluctuations in a small cell. 5. Define an equilibration interval from stationary behavior and independent starts. 6. Collect production with an ensemble and coupling appropriate to the desired observable; diffusion and vibrational correlation functions need additional dynamical sensitivity tests.

Interpretation and checks. The conserved extended-system quantity differs from raw K+U in thermostatted/barostatted dynamics. Pressure fluctuates strongly and its instantaneous value need not equal the target. A correct mean temperature does not validate density, structure or mixing. Do not use a barostat to correct a wrong electron count or unresolved Pulay/grid stress. Changing from NVT to NpT is a planned ensemble transition, not a seamless continuation of the same statistical distribution.

Exercise and diagram. Compare block-averaged temperature, volume and a structural observable across two coupling choices. Explain why pressure needs longer averaging and why a vacuum-containing slab should not use an isotropic bulk barostat.