Capillary Pressure in Petroleum Engineering: Concepts and Applications

Capillary Pressure in Petroleum Engineering - From Pore-Scale Physics to Reservoir Engineering Applications

Capillary pressure is one of the most misunderstood parameters in reservoir engineering - not because the concept is difficult, but because engineers routinely use capillary pressure data without understanding what it actually measures at the pore scale and what assumptions are embedded in the laboratory methods. A capillary pressure curve from a Mercury Injection test tells you something fundamentally different from a centrifuge test, and using the wrong dataset in a reservoir simulator can cause errors in initial fluid distribution that propagate through every production forecast you ever make. This guide gives you the complete physical framework, the measurement methods with their limitations, and the practical applications that matter most.

1. The Physics of Capillary Pressure - What It Actually Measures

1.1 Definition and Young-Laplace Equation

Capillary pressure (Pc) is the pressure difference across a curved interface between two immiscible fluids in a porous medium:

Pc = Pnw - Pw

Where:
Pnw = pressure in the non-wetting phase (oil or gas)
Pw = pressure in the wetting phase (water)

From the Young-Laplace equation for a cylindrical pore:
Pc = 2 x sigma x cos(theta) / r

Where:
sigma = interfacial tension between the two fluids (mN/m)
theta = contact angle (wettability indicator, degrees)
r = pore throat radius (microns)

Physical interpretation: Capillary pressure is controlled by three factors - how strongly the fluids repel each other (IFT), how strongly the rock prefers one fluid over the other (wettability/contact angle), and how small the pore throats are (r). A water-wet rock (theta near 0°, cos(theta) near 1.0) has high capillary pressure that strongly retains water in small pores. An oil-wet rock (theta near 180°, cos(theta) near -1.0) has negative capillary pressure - water is actually expelled from small pores.

1.2 Wettability - The Most Important Variable

Wettability State	Contact Angle	Pc Behavior	Common Reservoir Type
Strongly water-wet	0 - 30°	High positive Pc, water in small pores	Clean sandstones, carbonates
Weakly water-wet	30 - 75°	Moderate Pc, mixed distribution	Most reservoir rocks
Neutral wettability	75 - 105°	Near-zero Pc	Some aged crude oil systems
Weakly oil-wet	105 - 150°	Negative Pc drainage	Carbonates with asphaltenes
Strongly oil-wet	150 - 180°	Highly negative Pc	Some carbonate reservoirs

Why wettability matters for production: In a water-wet reservoir, injected water preferentially enters the small pores first, displacing oil from large pores - this is efficient displacement. In an oil-wet reservoir, injected water bypasses small pores entirely and channels through large pores - poor sweep efficiency. Identical porosity and permeability but opposite wettability can give recovery factors differing by 20-30 percentage points.

2. The Capillary Pressure Curve - Reading and Interpreting It

2.1 The Drainage and Imbibition Cycle

Every capillary pressure measurement involves either drainage (non-wetting phase displacing wetting phase - analogous to oil migrating into a water-saturated reservoir) or imbibition (wetting phase displacing non-wetting phase - analogous to waterflooding).

Process	Reservoir Analogy	Pc Sign	Engineering Use
Primary drainage	Oil migration into water-saturated rock	Positive	Initial fluid distribution, OWC
Spontaneous imbibition	Water invasion at zero capillary pressure	Positive to zero	Wettability index (Amott)
Forced imbibition	Waterflooding above capillary entry pressure	Negative	Waterflood performance

2.2 Key Points on the Drainage Pc Curve

Entry pressure (Pd): The minimum Pc required for the non-wetting phase to enter the largest pore throats. In a tight gas reservoir, this can be 50-500 psia. In a high-permeability sandstone, it may be less than 1 psia. Entry pressure is the most important single value on the curve for seal evaluation and trap integrity assessment.

Plateau region: The flat section of the curve where large increments of non-wetting phase enter at nearly constant pressure. Wide, flat plateaus indicate well-sorted, uniform pore systems (good reservoir quality). Steep, irregular plateaus indicate poorly sorted rock with complex pore geometry.

Irreducible water saturation (Swirr): The minimum water saturation - the point where the curve becomes nearly vertical. Water is trapped in the smallest pores and cannot be displaced no matter how high the capillary pressure. Swirr ranges from 5% in clean carbonates to 40%+ in shaly sands.

2.3 Converting Laboratory Pc to Reservoir Pc

Laboratory measurements use different fluid pairs (mercury-air, brine-air) at different conditions than the reservoir. Before using lab data in a reservoir simulator, you must convert:

Pc(reservoir) = Pc(lab) x (sigma x cos(theta))_reservoir / (sigma x cos(theta))_lab

Common conversion factors:
Mercury-air to oil-water: multiply by 0.085 - 0.10
Mercury-air to gas-water: multiply by 0.11 - 0.15
Brine-air to oil-water: multiply by 0.60 - 0.80

This conversion is critical. Mercury injection Pc of 1,000 psia translates to only 85-100 psia for oil-water in the reservoir. Using raw mercury injection values in a reservoir model without conversion will dramatically overestimate capillary trapping and produce incorrect initial water saturations.

3. Measuring Capillary Pressure - Methods and Their Limitations

3.1 Mercury Injection Capillary Pressure (MICP)

Mercury is injected into a dried core plug at incrementally increasing pressures. At each pressure step, the volume of mercury entering the sample is recorded. MICP accesses pore throats from 0.003 to 400 microns, making it uniquely suited for tight reservoirs and seal characterization.

Advantages: Covers the widest pressure range (1-60,000 psia), rapid measurement (hours vs days), provides full pore throat size distribution, low cost.

Critical limitations:

• Mercury is non-reactive with rock - it does not replicate the wettability of the actual reservoir fluid system
• The core must be dried and cleaned - wettability alteration from cleaning can be irreversible
• Mercury injection is irreversible - you cannot measure imbibition on the same sample
• At very high pressures (>20,000 psia), mercury can compress into intergranular cement pores that are not part of the connected pore network

3.2 Centrifuge Method

Core plugs saturated with the wetting phase are spun in a centrifuge containing the non-wetting phase at the outside. As centrifuge speed increases, the centrifugal force overcomes capillary pressure at progressively smaller pore throats, forcing wetting phase out. Fluid production is measured optically at each speed step.

Advantages: Uses actual reservoir fluids, measures both drainage and imbibition, provides wettability information, relatively fast (days).

Limitations:

• Pc is calculated from centrifuge speed and fluid densities - not directly measured
• The Pc value represents an average over the sample length - end effects can cause significant errors in short plugs
• Limited to low-to-moderate Pc values (typically <300 psia) - cannot characterize tight rocks

3.3 Porous Plate (Restored State) Method

A core plug is placed on a semi-permeable porous plate saturated with the wetting phase. Gas or non-wetting liquid pressure is applied above the plate, forcing wetting phase through the plate while retaining non-wetting phase in the plug. Each pressure step is maintained until equilibrium is reached, which can take days to weeks.

Advantages: Most accurate method for water-wet systems, uses actual reservoir fluids and conditions, measures true equilibrium saturations.

Limitations:

• Extremely slow - a single drainage curve can take 3-6 months
• Limited to Pc < 100 psia due to plate breakthrough pressure
• Not suitable for tight or heterogeneous rocks

3.4 Method Selection Guide

Reservoir Type	Recommended Method	Reason
High-permeability sandstone (>100 md)	Centrifuge + porous plate	Low Pc range, need actual fluid wettability
Moderate-permeability sandstone (1-100 md)	Centrifuge primary, MICP secondary	Balance between accuracy and range
Tight sandstone / carbonate (<1 md)	MICP primary	High Pc needed to access tight pore throats
Shale / unconventional	MICP (high pressure, >30,000 psia)	Only method that accesses nano-darcy pore throats
Seal/cap rock evaluation	MICP + mercury seal capacity	Entry pressure determines column height supported

4. Applications in Reservoir Engineering

4.1 Initial Fluid Distribution - Height Above Free Water Level

The most direct application of Pc data is calculating water saturation as a function of height above the Free Water Level (FWL). At the FWL, Pc = 0 and Sw = 100%. Above the FWL, water saturation decreases as capillary pressure increases with height:

Pc = (rho_w - rho_o) x g x h / 144

Where:
rho_w = water density (lbs/ft3)
rho_o = oil density (lbs/ft3)
g = gravitational acceleration (32.2 ft/s2)
h = height above FWL (ft)
144 = conversion to psi

Simplified: Pc (psi) = 0.433 x (SG_water - SG_oil) x h(ft)

Worked example: Water density 1.05 g/cc, oil density 0.82 g/cc, height above FWL = 120 ft:

Pc = 0.433 x (1.05 - 0.82) x 120 = 0.433 x 0.23 x 120 = 11.95 psia

From the Pc curve, Sw at Pc = 11.95 psia = 35% - this is the initial water saturation at 120 ft above FWL.

4.2 Oil-Water Contact and Transition Zone

The Oil-Water Contact (OWC) is commonly defined as the depth where Sw = 1.0 - but in a reservoir with significant capillary pressure, there is no sharp OWC. Instead there is a transition zone where Sw gradually decreases from 100% at the FWL to Swirr at some height above the FWL.

Transition zone thickness calculation:

h_transition (ft) = Pc_irreducible / (0.433 x (SG_water - SG_oil))

Example: Pc at Swirr = 35 psia, density difference = 0.23 g/cc:

h_transition = 35 / (0.433 x 0.23) = 35 / 0.0996 = 351 ft

A 351 ft transition zone means that a well perforated 200 ft above the FWL would still produce significant water. This is why logging-based Sw measurements must be calibrated to Pc data - resistivity logs detect Sw but not whether that water is mobile or capillary-bound.

4.3 Seal Capacity and Column Height

The entry pressure of the cap rock determines the maximum hydrocarbon column the trap can hold:

Max column height (ft) = Pc_entry(reservoir) / (0.433 x (SG_water - SG_HC))

Field significance: A shale seal with mercury injection entry pressure of 5,000 psia, converted to gas-water = 5,000 x 0.13 = 650 psia reservoir Pc. For a gas reservoir with SG_gas = 0.25 and SG_water = 1.05:

Max column = 650 / (0.433 x (1.05 - 0.25)) = 650 / 0.346 = 1,878 ft of gas column supported

4.4 Reservoir Simulation - J-Function Normalization

Different core plugs from the same formation have different Pc curves due to variation in permeability and porosity. The Leverett J-function normalizes these curves to a single dimensionless function for use in reservoir simulators:

J(Sw) = Pc x sqrt(k/phi) / (sigma x cos(theta))

Where:
k = permeability (md)
phi = porosity (fraction)
sigma = IFT (mN/m)
theta = contact angle (degrees)

When J-function curves from multiple plugs collapse onto a single curve, the formation is petrophysically homogeneous and the J-function can be used to populate Sw throughout the reservoir model. When they scatter, it indicates distinct rock types (flow units) that require separate Pc functions.

5. Common Errors in Capillary Pressure Applications

Error	Consequence	Prevention
Using MICP data without fluid conversion	10x overestimate of reservoir Pc, wrong Sw-depth profile	Always apply IFT and contact angle conversion
Ignoring wettability alteration during core cleaning	Artificially water-wet Pc curve - underestimates transition zone	Use preserved cores, measure Amott wettability index
Single rock type Pc curve for heterogeneous reservoir	Wrong initial Sw - OOIP error of 15-30%	Define rock types, use separate Pc per flow unit
Confusing FWL with OWC in simulation	Misplaced contacts, wrong OOIP calculation	FWL is where Pc = 0, OWC is where Sw = 1.0 - they differ by transition zone thickness

Conclusion

Capillary pressure is not just a laboratory measurement - it is the physical mechanism that controls where hydrocarbons are stored in a reservoir, how efficiently they can be produced, and whether a trap can hold a commercial column. Every initial water saturation you assign in a reservoir model, every transition zone thickness you estimate, every seal integrity evaluation you perform passes through capillary pressure data.

The engineers who apply Pc correctly understand two things: the measurement method determines what the data represents and what conversions are required, and wettability is not a constant - it varies with depth, lithology, and fluid contact history in ways that a single average Pc curve cannot capture. Integrate your Pc data with resistivity logs, NMR, and fluid composition data before trusting any Sw-depth profile in your model.

Want to discuss capillary pressure measurement selection for a specific reservoir type, or access our J-function normalization spreadsheet? Join our Telegram group for reservoir engineering discussions, or visit our YouTube channel for step-by-step tutorials on Pc data interpretation and reservoir simulation applications.