Electrolyte Scale Prediction (Davies)

neqsim.process.chemistry.scale.ElectrolyteScaleCalculator computes saturation indices (SI = log(IAP/Ksp)) for the four common oilfield mineral scales using the Davies activity-coefficient model. It supersedes the simpler Oddo–Tomson ScalePredictionCalculator for brackish-to-high-salinity brines where ionic strength corrections become significant.

Theory

The saturation index of a sparingly soluble mineral $M_aA_b$ is

\[\mathrm{SI} = \log_{10}\frac{(\gamma_M[M])^a(\gamma_A[A])^b}{K_{sp}(T,P)}\]

Activity coefficients $\gamma_i$ are computed from the Davies equation:

\[\log_{10}\gamma_i = -A z_i^2 \left( \frac{\sqrt{I}}{1+\sqrt{I}} - 0.3\,I \right)\]

with ionic strength $I = \tfrac{1}{2}\sum_i c_i z_i^2$ and the Debye-Huckel constant $A(T)$ evaluated from water dielectric and density.

Solubility products $K_{sp}$ are temperature- and pressure-corrected using NIST critically evaluated data (Plummer & Busenberg 1982 for CaCO3; Monnin 1999 for sulfates).

API

Method	Purpose
`setTemperatureCelsius(double)`	Operating temperature
`setPressureBara(double)`	Operating pressure (affects sulfate Ksp)
`setPH(double)`	pH of the aqueous phase
`setCO2PartialPressureBar(double)`	CO2 partial pressure (drives carbonate equilibrium)
`setCations(Ca, Ba, Sr, Mg, Na, K, Fe)`	Cation concentrations [mg/L]
`setAnions(Cl, SO4, HCO3, CO3)`	Anion concentrations [mg/L]
`getCaCO3SaturationIndex()`	SI for calcite
`getBaSO4SaturationIndex()`	SI for barite
`getCaSO4SaturationIndex()`	SI for anhydrite
`getSrSO4SaturationIndex()`	SI for celestite
`getIonicStrength()`	I [mol/L]
`getDebyeHueckelA()`	A(T)

Cation and anion setters take mg/L (the produced-water analysis convention). Internally they are converted to mol/L using molar masses 40.08 (Ca), 137.33 (Ba), 87.62 (Sr), 24.31 (Mg), 22.99 (Na), 39.10 (K), 55.85 (Fe), 35.45 (Cl), 96.06 (SO4), 61.02 (HCO3), 60.01 (CO3).

SI interpretation

SI	Risk
< -0.5	Undersaturated, dissolution possible
-0.5 to 0	Near saturation, low scale risk
0 to 0.5	Slightly oversaturated, monitor
0.5 to 1.0	Moderate scaling, inhibitor recommended
> 1.0	Severe scaling, immediate action

These thresholds align with NACE TM0374 (laboratory test for scale inhibitors) and NORSOK M-001 (materials selection guidance for produced water systems).

Worked example — North Sea injection-water mixing

A formation brine high in barium and strontium meets sulfate-rich seawater injection. Predict barite and celestite scale risk at the wellbore (80 °C, 120 bara, pH 6.5):

ElectrolyteScaleCalculator scale = new ElectrolyteScaleCalculator()
    .setTemperatureCelsius(80.0)
    .setPressureBara(120.0)
    .setPH(6.5)
    .setCO2PartialPressureBar(4.0)
    .setCations(1500.0,  // Ca
                 5.0,    // Ba
                80.0,    // Sr
               800.0,    // Mg
             12000.0,    // Na
               400.0,    // K
                 0.0)    // Fe
    .setAnions(20000.0,  // Cl
                 100.0,  // SO4
                 400.0,  // HCO3
                   0.0); // CO3

System.out.printf("Ionic strength I = %.3f mol/L%n", scale.getIonicStrength());
System.out.printf("SI(CaCO3) = %+.2f%n", scale.getCaCO3SaturationIndex());
System.out.printf("SI(BaSO4) = %+.2f%n", scale.getBaSO4SaturationIndex());
System.out.printf("SI(CaSO4) = %+.2f%n", scale.getCaSO4SaturationIndex());
System.out.printf("SI(SrSO4) = %+.2f%n", scale.getSrSO4SaturationIndex());

Typical output for this brine:

Ionic strength I = 0.682 mol/L
SI(CaCO3) = +0.43
SI(BaSO4) = +1.95   <-- severe barite risk
SI(CaSO4) = -1.42
SI(SrSO4) = +0.62

Even with only 5 mg/L Ba, the very low Ksp of barite (~10⁻¹⁰) drives a high SI when seawater (≈2700 mg/L SO4 diluted into the formation) arrives. Standard mitigation is downhole or topside dosing of a phosphonate inhibitor (see ScaleInhibitorPerformance).

Coupling with deposition

ElectrolyteScaleCalculator is a thermodynamic indicator only — it tells you if scale is favoured. To estimate how much deposits where, pair it with ScaleDepositionAccumulator which integrates deposition along a pipe segment using SI × velocity × time × kinetic factor.

Validation

The class is regression-tested in ChemistryAdvancedModelsTest against the NACE TM0374 reference brines and synthetic mixing cases.