SPE 70043
Saturation Exponent n in Well Log Interpretation: Another Look at the Permissible
Range
Paulus S. Adisoemarta, George A. Anderson, Scott M. Frailey, and George B. Asquith
Center for Applied Petrophysical Studies, Texas Tech University
Copyright 2001, Society of Petroleum Engineers Inc.
This paper was prepared for presentation at the SPE Permian Basin Oil and Gas Recovery
Conference held in Midland, Texas, 15–16 May 2001.
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300
words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
Abstract
Saturation exponent n in Archie’s water saturation equation is
the exponent value in water saturation that establishes the
relationship between water saturation of the rock to the ratio
of fluid filled rock resistivity to the actual rock resistivity.
Most theoretical derivations show that there is no exponent or
that n is equal to one. However, decades of results from core
analyses and openhole log analyses show n greater than one
and most often assumed to be two.
Most of the literature is in agreement that the saturation
exponent is affected by wettability and saturation history
(drainage or imbibition). Furthermore, the saturation exponent
depends on rock type, primarily the manner in which the
conductive water is connected and configured. Several
authors quote values of n ranging from as low as 1 to over 20
for strongly oil wet rocks. Too often an assumed value is
accepted as correct.
This paper reevaluates the relationship between water
saturation and rock resistivities in order to determine the
significance of n and make recommendations on improving an
assumed value for n.
Introduction
The quest to determine and understand n began shortly after
Archie’s earliest work correlating water saturation to
resistivity (or resistivity index). Surprisingly theoretically
based derivations show that n does not exist (or arguably is
one). This paper identifies some of the features that are
influence n and core and logging methods of estimating n.
Measurement Methods
Several methods to determine the value of n are as follows:
core measurement, combination of core measurement and log
data, and log data only.
Core measurement. This method can resolve the value of
n by varying core water saturation and correlate it against the
rock resistivity. The slope of a log-log plot yields n. It is
important to note that for water-wet core samples, the various
saturations must be derived during the drainage process not
the imbibition process. The core starts out 100% saturated
with water and oil is used to displace the water. At various
water saturations resistivity measurements are taken.
Combination of core and log. This method combines
core derives Sw with Rt values from log. One technique in this
method is to log-log crossplot Rt versus Swirr and the slope of
the line is saturation exponent n.
Another technique is to determine Swirr, porosity,
cementation exponent m and tortuosity factor a from a core
sample, and solve for n using Rt values from log. The Swirr is
from capillary pressure data. A variation of this technique is
to use Swirr from relative permeability data [Asquith et al.,
1997].
Log data only. Dielectric logs can independently solve
for Sxo. Using the values of Rxo from resistivity tools, a log-
log plot of Rxo versus Sxo yields a slope of the line equal to the
saturation exponent n.
Field Observed Factors Affecting n
Several factors has been observed in the field that affects the
value of n. These factors are:
• wettability
• texture roughness
• microporosity
Each of these factors is discussed briefly.
Wettability. The preference of grains to contact with a certain
fluid defines wettability. In water-wet rocks, water coats each
grain and provides a continuous path for ion at any saturation.
As the path is relatively unobstructed the resistivity is low
hence n is small. On the other hand, water is trapped in the
middle of oil in the pore space for the case of oil-wet rock. As
water saturation decreases water is separated from neighboring
2 Paulus Adisoemarta, George Anderson, Scott Frailey, and George Asquith
pores and hence breaking the path for ionic conduction,
creating a very high resistivity and so the value of n is high.
Keller [1953] showed the comparison of resistivity-saturation
measurements of water-wet and oil-wet sandstones that proved
the effect of wettability on n. Sweeney and Jennings [1960]
did the same but on carbonates.
Texture Roughness. Diederix [1982] observed the effect of
texture roughness on the value of n. On the North Sea sample
that has clean smooth grains the resistivity-saturation plot
shows a constant value of n. For the case of samples that have
rough grains, the plot shows a knee where the value of n
changes as a function of Sw. A confirmation of this field
observation was made thorough a sandpack modeling where
smooth beads and rough beads was used. The resistivity-
saturation plot of this bead models show the same trend as the
field data, where the smooth bead shows a constant value of n,
and the rough bead has a varying value of n.
Microporosity. Microporosity is defined as pore size of less
than 0.5 micron in diameter [Keith and Pitmann, 1984]. The
effect of microporosity to saturation exponent is a change of
slope on the log-Rt vs. log-Sw [Diederix, 1982]. Diederix
commented that by using a constant value of n across the
whole range of Sw will make the water saturation calculation
too high on the lower end of Sw. Swanson [1985] observed
this phenomenon on both shales, where the micropores occur
between the clay crystals, and clean sands such as chert.
Swanson’s observation showed that a decreasing Sw will
increase the value of n until at a certain Sw where n starts to
decrease.
Theoretical Derivation
The early literature on correlating water saturation to electrical
measurements (Wyllie and Spangler, 1952; Cornell and Katz,
1953; Bassiouni, 1994) provides derivations of the general
form:
R
Sw ∝ o
Rt
Only laboratory and field observations suggest that this Sw
relationship should have an exponent to become an equality.
Archie’s formula for water saturation equation can be used in
any of the general forms below:
R FR w aR
S nw = o = = mw
Rt Rt φ Rt
No matter which parameters are used to find water saturation,
n is required.
Most derivations start with the definition of the resistivity
index, I:
R (Adisoemarta, et al., 2000, Kumar and Frailey, 2001)
I= t
Ro
and the relationship between resistivity and resistance:
rA
R=
L Sw that is too small compared to observations. In other words,
SPE 70043
for a porous sample 100% saturated with water:
r A
Ro = o
L
where ro is
R L
ro = w a
Aa
because only water is considered conductive. The subscript
“a” represents the apparent cross-sectional area and length.
Substituting the last equation into the previous yields:
R ( L / L)
Ro = w a
(A a / A)
By analogy, the same sample saturated with less than 100%
water saturation yields the definition of Rt:
R ( L / L)
R t = w at
(A at / A)
The subscript “at” designates the apparent cross-sectional area
and length in the case of Sw<100%. Substituting the Rt and Ro
equation into the definition of I yields:
(A a / A at )
I=
(L a / L at )
Representing the cross-sectional area, Aat as
Aat = φ SwA
gives the final theoretically derived relation between water
saturation and resistivity as
L R
S w = at o
La R t
(Other derivations show Lat and L instead of La.)
Water saturation-resistivity derivations of this type do not
show Sw with an exponent. Only the statistical approach of
Wyllie and Spangler shows a saturation exponent, but it is
through the derivation of the cementation exponent, m.
An unanswered questions remains. Why do measured Sw-
Rt data require an n-exponent to form an equality?
Basic Problem of Relating Sw to Resistivity
There is a fundamental problem with Sw-R relationships that
basic to the description of porous media. A series of
interconnected pore bodies and pore throats describes many
types of porous media. Water saturation is dominated by the
largest volume of water in porous media, which is the pore
bodies. Resistivity is dominated by the smallest restrictions in
the ion path within the porous media, which is the pore
throats.
The observations that concluded the necessity of n>1 may
be due to relating a pore body dominated property (water
saturation) to a pore throat dominated property (resistivity).
This has been noted with φ-R and k-φ relationships, also.
Effect of n on Sw
Using one of the variations of Archie’s saturation equation
without n (or n=1) as theoretical derivations suggest results in
SPE 70043 SATURATION EXPONENT N IN WELL LOG INTERPREATION: ANOTHER LOOK AT THE PERMISSABLE RANGE
n>1 increases the predicted value of Sw from Archie’s
equation. Figure 1 shows Sw as a function of Ro/Rt for n = 1,
2, and 4. n has the largest correction to Sw at the lower range
of Sw.
As an example, for Ro/Rt = 0.20, theory (n=1) predicts Sw =
20%, while the commonly accepted value of n=2 predicts Sw
= 44.7%. The pore throat dominated resistivity only requires
20% of the pore space to be occupied by water to measure
Ro/Rt = 0.20, while the pore may have a large proportion of
water present (another 24.7% in this example) that does not
influence the resistivity significantly. Again, the difference
relates to the pore body vs. pore throat influenced properties.
Range of n
To quantify and understand the physical meaning of n, a
comparison of the theoretical relation to observation is
necessary. This is difficult for two reasons: 1) n does not
appear in the theoretically derived equation and 2) Lat/La
disappears completely from the equation developed from
observation. In a previous publication that quantified and
explained the physical meaning of m and a (Adisoemarta,
2000), theory could only provide a value of a but not m. In
comparison, theory does not predict the exponent m and m
only exists to correlate theory to observation. n correlates
theory to observation and replaces the term Lat/La. This is
understandable because these terms cannot be measured;
however, a is proportional to La/L which is cannot be
measured but is not dropped from the Fr relationship.
Consequently to identify a range of n by comparing
observation to theory, Lat/La is assumed to be close to one.
The introduction of Sw into the derivation is to approximate
the cross-sectional area to ion movement as φSwA. Because φ
and Sw are dominated by pore bodies, and Ro and Rt are
dominated by the pore throats, n may be correcting the
measured resistivities to the measured Sw due to the difference
in φSw in the pore bodies and pore throats.
From the previous derivation,
S w = A at / A a or A at = S w A a
but observation suggests
A at = S nw A a
Following the example and simple geometric configuration
of porous media that Adisoemarta, et al. (2000) used to
quantify m, similar results are obtained for n. Because the
only factors that influence resistivity are the apparent length
and cross-sectional area, the results for m and n are identical.
(Table 1 is a duplicate of table 2 of Adisoemarta et al (2000)
for the case of n.) This provides a theoretical based
explanation of why historical observations have seen n and m
of similar value (and equal to 2).
Fractured porosity, (Ap=Apt) appears to have the same
affect on n as it does on m. n decreases and approaches one as
Apt approaches Ap. A simple data set shows this, also.
wettability effects may be inferred from this table, also.
The larger values of Ap/Apt represent decreased cross-sectional
area of the water in the pore throats with respect to the area of
3
the water in the pore bodies; this is the case of a oil-wet rock
at low water saturation. The estimated n value for the simple
geometric shaped porous media shows a higher value for all
cases of Lp/Lpt.
Conclusions
Historical observation and the theoretical development of this
paper are in general agreement. Factors that have been
observed to influence n can be simplified and explained in
terms of variations of cross-sectional area of the water.
Many laboratory and field observations observe that
wettability and saturation history (drainage and imbibition)
change n.
Because Lat/La disappears from Archie’s water saturation
equation (when comparing theoretical vs. observation based
derivation), n must compensate for variation in pore throat and
body size and Lat/La. This makes estimating n for simple
geometric difficult without assuming Lat/La of one.
A fundamental inconsistency in Archie’s water saturation
and formation resistivity factor is that water saturation and
porosity are pore body properties, while resistivity
measurements are pore throat properties. Consequently, any
changes or manipulations to Archie’s equation may be
applicable only to a specific formation. In this sense, Archie’s
equations are truly correlations.
Nomenclature
a = tortuosity factor
A = area
Aa = apparent cross-section area, Sw=100%
Aat = apparent cross-section area, Sw<100%
Ap = cross-section area of a pore
Apt = cross-section area of a pore throat
d = grain diameter
F = formation resistivity factor
I = resistivity index
m = cementation factor
n = saturation exponent
r = resistance of the material
R = resistivity
Ro = formation resistivity when 100% saturated with water
of resistivity Rw, Ω.m
Rt = formation resistivity when Sw<100%, Ω.m
Rw = water resistivity, Ω.m
L = length
La = apparent length, Sw=100%
Lat = apparent length, Sw<100%
Lp = pore length
Lpt = pore-throat length
Sw = water saturation, a fraction of the pore volume
α = angle between fluid path and the bulk fluid direction,
degrees
τ = tortuosity
φ = porosity
4 Paulus Adisoemarta, George Anderson, Scott Frailey, and George Asquith SPE 70043

References
Adisoemarta, P.S., Anderson, G.A., Frailey, S.M., and Asquith, G.B.: TABLE 1. VALUES OF n AS A FUNCTION OF Ap / Apt
“Historical Use of m and a in Well Log Interpretation: Is AND Lp / Lpt
Conventional Wisdom Backwards?”, SPE 59699 presented at Values of Values of Lp / Lpt
the 2000 SPE Permian Basin Oil and Gas Recovery Conference,
Midland, March 21-23. Ap / Apt 0.25 0.5 1 2 4
Amyx, J.W., Bass, Jr., D.M., and Whiting, R.L.: Petroleum Reservoir
Engineering, McGraw-Hill, New York (1960). 1 1.00 1.00 1.00 1.00 1.00
“Archie II: Electrical Conduction in Hydrocarbon-Bearing Rock”,
The Technical Review, Vol 36. No. 4; pp 12-21. 2 1.11 1.20 1.33 1.50 1.67
Asquith, G.B., S.P. Dutton, A.G. Cole, Muhammad Razi, and Jose 4 1.18 1.33 1.60 2.00 2.50
Guzman.: “Petrophysics of the Ramsey Sandstone, Ford Gerald
Unit Reeves and Culberson Counties, Texas: in Permian Basin 8 1.21 1.41 1.78 2.40 3.33
Oil and Gas Fields: Turning Ideas into Production”, W.D.
Demis, ed., West Texas Geological Society, Publ. #97-102, 16 1.23 1.45 1.89 2.67 4.00
1997, p. 61-69.
Bassiouni, Z.: Theory, Measurement, and Interpretation of Well 32 1.24 1.48 1.94 2.82 4.44
Logs, Textbook Series, SPE, Richardson, TX (1994) 4.
∞ 1.25 1.50 2.00 3.00 5.00
Cornell, D. and Katz, D.L.: “Flow of Gases Through Consolidated
Porous Media,” Ind. Engr. Chem., (1953) 45.
Diederix , K.M.: “Anomalous Relationships Between Resistivity
Index and Water Saturations in the Roetliegend Sandstone, (The Effect of n on Archie Water Saturation

Netherlands)”, presented at the 23rd SPWLA Annual 1

Symposium, June 6-9, 1982.

0.9
Keith, B.D., and E.D. Pittman,1984.: “Bimodal Porosity in Oolitic
Reservoirs-effect on Productivity and Log Response, Rodessa 0.8

Limestone (lower Cretaceous), East Texas Basin”, American 0.7

Association of Petroleum Geologists Bull. v. 67, no. 9, p. 1391-

Archie Water Saturation

0.6
1399.
Keller, G.V.: “Effect of Wettability on the Electrical Resistivity of 0.5

Sand”, Oil & Gas Journal 51 (January 5, 1953), pp 62-65. 0.4

Kumar, N. and Frailey, S.M.: “Using Well Logs to Infer
0.3
Permeability: Will There Ever Be a Permeability Log?”,
n=1
Southwestern Petroleum Short Course, Lubbock, TX (April 24- 0.2
n=2

5, 2001). 0.1
n=4

Mungan, N, and Moore, E.J.: “Certain Wettability Effects on

Electrical Resistivity in Porous Media”, presented at the 18th 0
0.000 0.100 0.200 0.300 0.400 0.500 0.600 0.700 0.800 0.900 1.000
Petroleum Society of CIM, Banff, Canada (May 1967), pp 20- Ro/Rt

25.
Swanson, B.F.: “Microporosity in Reservoir Rocks – Its Figure 1: Impact of n on calculated water saturation using Archie’s
Measurement and Influence of Electrical Resistivity”, equation
Transactions of the SPWLA 26th Annual Logging Symposium,
Dallas (June 17-20, 1985)
Tiab, D. and Donaldson, E.C.: Petrophysics: Theory, and Practice
of Measuring Reservoir Rock and Fluid Transport Properties,
Gulf Publishing, Houston (1996).
Sweeney, S.A., and Jennings, H.Y. Jr.: “The Electrical Resistivity of
Preferentially Water-Wet and Preferentially Oil-Wet Carbonate
Rock”, Producers Monthly no.7 (May 1960), pp 29-32
Wyllie, M.R.J., and Gardner, G.H.F.: “The Generalized Kozey-
Carman Equation,” World Oil (March and April 1958).
Wyllie, M.R.J., and Spangler, M.B.: “Application of Electrical
Resistivity Measurements to Problem of Fluid Flow in Porous
Media,” Bull. Am. Assoc. Petrol. Geologists (February 1952) p.
359.