[Next] [Previous] [Up] [Top]

Keywords

SURFACE

This keyword is used to define the amount and composition of each surface in a surface assemblage. The composition of a surface assemblage can be defined in two ways, (1) implicitly, by specifying that the surface assemblage is in equilibrium with a solution of fixed composition or (2) explicitly, by defining the amounts of the surfaces in their neutral form (for example, SurfbOH). A surface assemblage may have multiple surfaces and each surface may have multiple binding sites, which are identified by letters following an underscore.

Example 1

Line 0a:  SURFACE 1 Surface in equilibrium with solution 10
Line 1a:       -equilibrate with solution 10
Line 2a:       Surfa_w   1.0     1000.     0.33
Line 2b:       Surfa_s   0.01
Line 2c:       Surfb     0.5     1000.     0.33
Line 3:        -diffuse_layer    2e-8
Line 0b: SURFACE 2 Ignore electrostatic double layer
Line 1b:       -equilibrate with solution 10
Line 2b:       Surfc     0.5     1000.     0.33
Line 4:        -no_edl 

Explanation 1

Line 0: SURFACE [number] [description]

SURFACE is the keyword for the data block.

number--positive number to designate this surface assemblage and its composition. Default is 1. A range of numbers may also be given in the form m-n, where m and n are positive integers, m is less than n, and the two numbers are separated by a hyphen without intervening spaces.

description--optional character field that describes the surface assemblage.

Line 1: -equilibrate number

-equilibrate--indicates that the surface assemblage is defined to be in equilibrium with a given solution composition. Optionally, equil, equilibrate, or -e[quilibrate].

number--solution number with which the surface assemblage is to be in equilibrium. Any alphabetic characters following the identifier and preceding an integer ("with solution" in line 1a) are ignored.

Line 2: surface name, sites, specific area, mass

surface name--name of a surface binding site (analogous to the name of an element).

sites--total number of sites for this binding site, in moles.

specific area--specific area of surface, in m2/g. Default 600 m2/g.

mass--mass of surface, in g. Default 0 g.

Line 3: -diffuse_layer [thickness]

-diffuse_layer--indicates that the composition of the diffuse layer will be estimated, such that, the net surface charge plus the net charge in the diffuse layer will sum to zero. Optionally, diffuse_layer, -d[iffuse_layer]. See notes following the example. The identifiers -diffuse_layer and -no_edl are mutually exclusive.

thickness--thickness of the diffuse layer in meters. Default is 10-8 m (100 Angstrom).

Line 4: -no_edl

-no_edl--indicates that no electrostatic terms will be used in the calculation. No potential term will be included in the mass-action expressions for the surface species and no charge-balance equations for the surface will be used. The identifiers -diffuse_layer and -no_edl are mutually exclusive.

Notes 1

The default databases contain thermodynamic data for a surface named "Hfo" (Hydrous Ferric Oxides) that are derived from Dzombak and Morel (1990). Two sites are defined in the databases: a strong binding site, Hfo_s, and a weak binding site Hfo_w.

The order of lines 1, 2, 3, and 4 is not important. Lines 1 and, optionally, 3 or 4 should occur only once within the keyword data block. Line 2 may be repeated to define the amounts of all of the binding sites for all of the surfaces. In the example, two surfaces are considered, Surfa and Surfb. Surfa has two binding sites, Surfa_w and Surfa_s; the surface area and mass for Surfa must be defined in the input data for at least one of the two binding sites. Surfb has only one kind of binding site and the area and mass must be defined as part of the input for this binding site.

Lines 1a and 1b require the program to make two calculations to determine the composition of each of the surface assemblages. Before any reaction calculations, two initial surface-composition calculations will be performed to determine the composition of the surface assemblages that would exist in equilibrium with the specified solution (solution 10 for both surface assemblages in this example). The composition of the solution will not change during these calculations. In contrast, during a reaction calculation, when a surface assemblage (defined as in example 1 or example 2 of this section) is placed in contact with a solution with which it is not in equilibrium, both the surface composition and the solution composition will adjust to reach a new equilibrium.

When the -diffuse_layer identifier is used, the composition of the diffuse layer is calculated. The moles of each aqueous species in the diffuse layer are calculated according to the method of Borkovec and Westall (1983) and the assumption that the diffuse layer is a constant thickness (optional input with -diffuse_layer, default is 10-8 m). The net charge in the diffuse layer exactly balances the net surface charge. Conceptually, the results of using this alternative approach are correct. Charge imbalances on the surface are balanced in the diffuse layer and the solution remains charge balanced. There still exist great uncertainties in the true composition of the diffuse layer and the thickness of the diffuse layer. The ion complexation in the bulk solution is assumed to apply in the diffuse layer, which is unlikely because of changes in the dielectric constant of water. The thickness of the diffuse layer is purely an assumption that allows the volume of water in the diffuse layer to remain small relative to the solution volume. It is possible, especially for solutions of low ionic strength, for the calculated concentration of an element to be negative in the diffuse layer. In these cases, the assumed thickness of the diffuse layer is too small or the entire diffuse-layer approach is inappropriate. The calculation of the diffuse-layer composition involves a computer intensive integration and an additional set of iterations. The -diffuse_layer identifier causes calculations to be 5 to 10 times slower than calculations with the default approach.

The -diffuse_layer identifier is a switch that activates a different model to account for the accumulation of surface charge. An additional printout of the elemental composition of the diffuse layer is produced. When -diffuse_layer is not used (default), to account for the charge that develops on the surface, an equal, but opposite, amount of charge imbalance is attributed to the solution. Thus, charge imbalances accumulate in the solution and on the surface when surfaces and solutions are separated. This handling of charge imbalances for surfaces is physically incorrect. Consider the following, where a charge-balanced surface is brought together with a charge-balanced solution. Assume a positive charge develops at the surface. Now remove the surface from the solution. With the present formulation, a positive charge imbalance is associated with the surface, Zs, and a negative charge imbalance, Zsoln, is associated with the solution. In reality, the charged surface plus the diffuse layer surrounding it would be electrically neutral and both should be removed when the surface is removed from solution. This would leave an electrically neutral solution. The default formulation is workable; its main defect is that the counter-ions that should be in the diffuse layer are retained in the solution. The model results are adequate, provided solutions and surfaces are not separated or the exact concentrations aqueous counter-ions are not critical to the investigation.

A third alternative for modeling surface-complexation reactions, in addition to the default and -diffuse_layer, is to ignore the surface potential entirely. The -no_edl identifier eliminates the potential term from mass-action expressions for surface species, eliminates any charge-balance equations for surfaces, and eliminates any charge-potential relationships. The charge on the surface is calculated and saved with the surface composition and an equal and opposite charge is stored with the aqueous phase. All of the cautions about separation of charge, mentioned in the previous two paragraphs, apply to the calculation using -no_edl.

For transport calculations, it is much faster in terms of cpu time to use either the default (no explicit diffuse layer calculation or -no_edl). However, -diffuse_layer can be used to test the sensitivity of the results to diffuse-layer effects. All solutions should be charge balanced for transport calculations.

Example 2

Line 0:  SURFACE 1 Measured surface composition
Line 1a:      Surf_wOH     0.3     660.     0.25
Line 1b:      Surf_sOH     0.003

Explanation 2

Line 0: SURFACE [number] [description]

Same as example 1.

Line 1: formula, sites, specific area, mass

formula--formula of the surface binding site in its OH form, Surf_sOH and Surf_wOH in this example. It is important to include the OH in the formula or hydrogen and oxygen will be extracted from the solution during the reaction step, which will cause unexpected redox or pH reactions.

sites--total number of sites for this binding site, in moles.

specific area--specific area of surface, in m2/g.

mass--mass of surface, in g.

Notes 2

Although this example only defines one surface with two binding sites, Surf_s and Surf_w, other surfaces with one or more binding sites could be defined by repeating line 1. The -diffuse_layer or -no_edl identifier can also be included in this example.

After a reaction has been simulated, it is possible to save the resulting surface composition with the SAVE keyword. If the new composition is not saved, the surface composition will remain the same as it was before the reaction. After it has been defined or saved, the surface composition may be used in subsequent simulations through the USE keyword.

Example problems

The keyword SURFACE is used in example problems 8 and 10.

Related keywords

SURFACE_MASTER_SPECIES, SURFACE_SPECIES, SAVE surface, and USE surface.

Example 1
Explanation 1
Notes 1
Example 2
Explanation 2
Notes 2
Example problems
Related keywords

User's Guide to PHREEQC - 07 MAY 96
[Next] [Previous] [Up] [Top]

Generated with CERN WebMaker