The Leverett J-Function

Part 11, Chapter 11: Capillary Pressure and Saturation-Height

Learning objectives

  • State the dimensionless Leverett J-function and its terms
  • Explain how dividing out the square root of k over phi cancels pore size
  • Collapse many rock types' capillary curves onto one J-Sw curve
  • Apply one cored well's capillary data across the whole field

One Curve for Many Rocks

Every rock type draws its own capillary-pressure curve, and they are not gentle variations on one shape. A tight rock needs many times the pressure of a clean sand to reach the same water saturation, so a field of mixed rock is a tangle of curves with no single one to carry into a saturation-height model. The question Leverett answered in 1941 is how to fold all of them into one.

The Leverett J-function0204060801000.00.51.01.52.02.5water saturation Sw (%)Leverett J (dimensionless)good k=1000 phi=28%medium k=100 phi=22%poor k=10 phi=15%one master curveDivide Pc by sqrt(k/phi): three very different rocks collapse onto one J-Sw curve.

Cancelling the Pore Size

The trick is that two things in the rock both track the pore-throat radius. Capillary pressure runs as Pc1/rP_c \sim 1/r by Young-Laplace, while the quantity k/ϕ\sqrt{k/\phi}, the reservoir quality index from the permeability chapter, runs as rr. Multiply one by the other and the radius cancels, leaving a dimensionless number that depends only on the shape of the pore system, not its size:

J(Sw)=0.21645Pcσcosθkϕ.J(S_w) = 0.21645\,\frac{P_c}{\sigma \cos\theta}\sqrt{\frac{k}{\phi}}.

The constant 0.21645 makes JJ dimensionless with PcP_c in psi, the interfacial tension σ\sigma in dyne/cm, and kk in millidarcies. Rocks that share a pore-system shape, however far apart their raw PcP_c curves sit, collapse onto one master JJ-SwS_w curve.

The Payoff and the Limit

This is what makes capillary pressure usable at field scale. Measure PcP_c on a handful of cores, fit one JJ curve, and apply it to every porosity-permeability pair the logs deliver, turning a few laboratory samples into a saturation-height model for the whole reservoir. The limit is honest: real rocks differ in pore geometry, not only in k/ϕk/\phi, so the collapse is never quite perfect. The cure is to group the rock into a small number of types and fit one JJ curve per type, which the next sections turn into the saturation-height function itself.

References

  • Leverett, M. C. (1941). Capillary behavior in porous solids. Transactions of the AIME, 142, 152-169.
  • Tiab, D. and Donaldson, E. C. (2015). Petrophysics, 4th ed. Gulf Professional Publishing.

This page is prerendered for SEO and accessibility. The interactive widgets above hydrate on JavaScript load.