Well Fluids¶

Overview¶

Every well in BORES needs to know the properties of the fluids it handles. Production wells define ProducedFluid objects for each phase (oil, water, gas) that may flow into the wellbore. Injection wells define a single InjectedFluid object describing the fluid being pumped into the reservoir. These fluid objects provide the specific gravity and molecular weight needed to compute PVT properties (formation volume factors, viscosities, densities) at wellbore conditions.

The distinction between produced and injected fluids matters because of how PVT properties are used. For production wells, the simulator looks up oil, water, and gas properties from the reservoir model's PVT grids. For injection wells, the fluid properties may differ significantly from the in-situ reservoir fluids, particularly for gas injection (where the injected gas may be CO2 or nitrogen with very different properties from the reservoir's solution gas).

ProducedFluid¶

A ProducedFluid describes a fluid phase that flows from the reservoir into a production well. It is a simple data class with four required fields:

import bores

oil = bores.ProducedFluid(
    name="Oil",
    phase=bores.FluidPhase.OIL,
    specific_gravity=0.85,    # Oil gravity relative to water (~35 API)
    molecular_weight=200.0,   # Average molecular weight (g/mol)
)

water = bores.ProducedFluid(
    name="Formation Water",
    phase=bores.FluidPhase.WATER,
    specific_gravity=1.02,    # Slightly saline
    molecular_weight=18.015,  # H2O
)

gas = bores.ProducedFluid(
    name="Associated Gas",
    phase=bores.FluidPhase.GAS,
    specific_gravity=0.65,    # Light hydrocarbon gas
    molecular_weight=16.04,   # Close to methane
)

For most simulations, the produced fluid properties should match the reservoir fluid properties you used when building the reservoir model. The specific gravity of oil should match the oil_specific_gravity_grid values, and the gas specific gravity should match the gas_specific_gravity parameter.

A production well typically includes all three phases, even if you expect only one phase to be produced initially. During a waterflood, for example, water will eventually break through and the well needs water fluid properties to compute the water production rate correctly.

InjectedFluid¶

An InjectedFluid describes the fluid being pumped into an injection well. It extends the base WellFluid class with additional parameters for miscible flooding, salinity, and property overrides.

Water Injection¶

For water injection, the InjectedFluid is straightforward:

import bores

water_fluid = bores.InjectedFluid(
    name="Injection Water",
    phase=bores.FluidPhase.WATER,
    specific_gravity=1.0,     # Fresh water
    molecular_weight=18.015,
)

If you are injecting saline water (brine), specify the salinity:

brine_fluid = bores.InjectedFluid(
    name="Seawater",
    phase=bores.FluidPhase.WATER,
    specific_gravity=1.03,
    molecular_weight=18.015,
    salinity=35000.0,  # ppm NaCl
)

The salinity affects water density and viscosity calculations. Seawater typically has about 35,000 ppm total dissolved solids, while formation brines can range from 10,000 to over 200,000 ppm.

Hydrocarbon Gas Injection¶

For injecting hydrocarbon gas (methane, natural gas, or enriched gas), use standard gas properties:

import bores

natural_gas = bores.InjectedFluid(
    name="Natural Gas",
    phase=bores.FluidPhase.GAS,
    specific_gravity=0.70,
    molecular_weight=20.0,
)

The gas properties (compressibility factor, density, viscosity) are computed from correlations using the specific gravity and molecular weight. These correlations work well for hydrocarbon gases because they were developed from large datasets of natural gas measurements.

CO2 Injection¶

CO2 injection requires special attention because CO2 is a non-ideal gas whose properties deviate significantly from the standard gas correlations used for hydrocarbon gases. At typical reservoir conditions (above 1100 psi and 90 degrees F), CO2 exists as a supercritical fluid with a density 5 to 10 times higher than what the gas correlations predict, and a viscosity 3 to 5 times higher.

BORES provides density and viscosity override parameters on InjectedFluid to handle this:

import bores

co2_fluid = bores.InjectedFluid(
    name="CO2",
    phase=bores.FluidPhase.GAS,
    specific_gravity=1.52,      # CO2 gravity relative to air
    molecular_weight=44.01,     # CO2 molecular weight
    density=35.0,               # lbm/ft³ - from EOS or lab data
    viscosity=0.05,             # cP - from EOS or lab data
)

When density or viscosity is set, BORES uses those values directly instead of computing them from correlations. This is critical for CO2 because the standard gas correlations (Lee-Gonzalez-Eakin for viscosity, Hall-Yarborough for Z-factor) were developed for hydrocarbon gases and produce grossly inaccurate results for CO2 at reservoir conditions.

The density and viscosity values should come from equation-of-state calculations (using a tool like CoolProp, NIST REFPROP, or commercial PVT software) or from laboratory measurements at your specific reservoir temperature and pressure. As a rough guide for CO2 at typical reservoir conditions:

Miscible Gas Injection¶

For miscible flooding (CO2 or enriched gas that mixes with oil above the minimum miscibility pressure), set is_miscible=True and provide the MMP and Todd-Longstaff mixing parameter:

import bores

co2_miscible = bores.InjectedFluid(
    name="CO2",
    phase=bores.FluidPhase.GAS,
    specific_gravity=1.52,
    molecular_weight=44.01,
    density=35.0,
    viscosity=0.05,
    is_miscible=True,
    minimum_miscibility_pressure=1200.0,  # psi
    todd_longstaff_omega=0.67,            # Mixing parameter (0-1)
    miscibility_transition_width=500.0,   # Pressure range for smooth transition
)

The minimum_miscibility_pressure (MMP) is the pressure above which the injected gas develops first-contact or multi-contact miscibility with the reservoir oil. Below the MMP, the gas displaces oil immiscibly with high residual oil saturation. Above the MMP, the displacement approaches piston-like efficiency with near-zero residual oil.

The todd_longstaff_omega parameter controls the degree of mixing between solvent and oil at the sub-grid scale. A value of 1.0 means complete mixing (the solvent and oil are fully miscible within each grid cell), while 0.0 means no mixing (the fluids remain segregated). The default value of 0.67 is the most commonly used in industry. See Miscible Flooding for detailed physics.

The miscibility_transition_width controls the pressure range over which miscibility transitions from immiscible to fully miscible. BORES uses a smooth hyperbolic tangent transition centered at the MMP, with the width controlling how abrupt the transition is. A width of 0 gives a sharp step change; values of 300-500 psi give a gradual transition that is more numerically stable.

Nitrogen Injection¶

Nitrogen is sometimes used for pressure maintenance or immiscible gas injection. Like CO2, its properties differ from hydrocarbon gas, but the deviation is less severe:

import bores

n2_fluid = bores.InjectedFluid(
    name="Nitrogen",
    phase=bores.FluidPhase.GAS,
    specific_gravity=0.967,     # N2 gravity relative to air
    molecular_weight=28.013,    # N2 molecular weight
)

For nitrogen, the standard gas correlations are reasonably accurate because nitrogen is closer to ideal gas behavior than CO2. However, for high-pressure applications (above 5000 psi), you may want to provide density and viscosity overrides from EOS calculations for improved accuracy.

Common Fluid Property Values¶