SpatialReactionSolver

This sets up a MOOSE simulation that is spatially and temporally dependent, with no "usual" solution process (with Kernels, etc). This Action adds a GeochemistrySpatialReactor userobject, many AuxVariables corresponding to molality, free-mg, free-cm3, pH, etc using the GeochemistryQuantityAux and a console output object, the GeochemistryConsoleOutput along with its NearestNodeNumberUO UserObject.

The input file must also contain the GeochemicalModelDefinition (as all geochemistry simulations must), along with a definition of the Mesh and Executioner.

An simple example input file is:

# Example demonstrating that controlled-activity can be spatially-dependent
[UserObjects]
  [definition]
    type = GeochemicalModelDefinition
    database_file = "../../../database/moose_geochemdb.json"
    basis_species = "H2O H+ Cl-"
  []
[]

[SpatialReactionSolver]
  model_definition = definition
  charge_balance_species = "Cl-"
  constraint_species = "H2O H+ Cl-"
  constraint_value = "  1                -5            1E-5"
  constraint_meaning = "bulk_composition log10activity bulk_composition"
  constraint_unit = "   kg               dimensionless moles"
  controlled_activity_name = 'H+'
  controlled_activity_value = 'act_H+'
[]

[Mesh]
  type = GeneratedMesh
  dim = 1
  nx = 10
  xmax = 1
[]

[Executioner]
  type = Transient
  num_steps = 1
[]

[AuxVariables]
  [act_H+]
  []
[]

[AuxKernels]
  [act_H+]
    type = FunctionAux
    variable = 'act_H+'
    function = '10^(-5 + x)'
    execute_on = timestep_begin # so the Reactor gets the correct value
  []
[]

[VectorPostprocessors]
  [pH]
    type = LineValueSampler
    start_point = '0 0 0'
    end_point = '1 0 0'
    sort_by = x
    num_points = 11
    variable = pH
  []
[]

[Outputs]
  csv = true
  execute_on = final
[]