Sie sind auf Seite 1von 512

Click Here To Continue

Chapter 46
Thermal Recovery
Chieh Chu, Getty Oil Co.*
Thermal recovery generally refers to processes for recov-
ering oil from underground formations by use of heat.
The heat may be supplied externally by injecting a hot
fluid such as steam or hot water into the formations, or
it may be generated internally by combustion. In com-
bustion, the fuel is supplied by the oil in place and the
oxidant is injected into the formations in the form of air
or other oxygen-containing fluids. The most commonly
used thermal recovery processes are steam injection proc-
esses and in-situ combustion.
Two Forms of Steam Injection Processes
In principle, any hot fluid can be injected into the forma-
tions to supply the heat. The fluids used most extensively
are steam or hot water because of the general availability
and abundance of water. Hot water injection has been
found to be less efficient than steam injection and will not
be discussed here. A schematic view of the steam injec-
tion process is shown in Fig. 46.1, together with an ap-
proximate temperature distribution inside the formation.
There are two variations of steam injection processes-
steam stimulation and steam displacement.
Steam Stimulation
This method has been known as the huff n puff method,
since steam is injected intermittently and the reservoir is
allowed to produce after each injection. In this process
the main driving force for oil displacement is provided
by reservoir pressure, gravitational force, rock and fluid
expansion, and, possibly, formation compaction. In the
steam stimulation process only the part of the reservoir
adjacent to the wellbore is affected. After a number of
cycles of injection and production, the near-wellbore
region in reservoirs having little or no dip becomes so
depleted of oil that further injection of steam is futile. In
this case, wells must be drilled at very close spacing to
obtain a high oil recovery.
Now Wlih Texaco inc
Steam Displacement
This process, usually referred to as steamflood or steam-
drive, has a much higher oil recovery than steam stimu-
lation alone. Whereas steam stimulation is a one-well
operation, steamflood requires at least two wells, one
serving as the injector and the other serving as the
producer. The majority of steamflood projects use pat-
tern floods. In many cases, steam stimulation is required
at the producers when the oil is too viscous to flow be-
fore the heat from the injector arrives. Because of the high
oil recovery achievable through steamflooding, many
reservoirs that were produced by steam stimulation previ-
ously now are being steamflooded.
Three Forms of In-Situ Combustion
In-situ combustion usually is referred to as fireflood.
There are three forms of in-situ combustion processes-
dry forward combustion, reverse combustion, and wet
Dry Forward Combustion
In the earlier days, this was the most commonly used form
of the combustion processes. It is dry because no water
is injected along with air. It is forward because combus-
tion starts at the injector and the combustion front moves
in the direction of the air flow.
Fig. 46.2 gives a schematic view of the dry forward
combustion process. The upper part of the figure shows
a typical temperature distribution along a cross section
leading from the injector at the left to the producer at the
right. Two things need to be pointed out. First, the region
near the producer is cold, at the original temperature of
the reservoir. If the unheated oil is highly viscous, it can-
not be pushed forward by the heated oil at its back that
has been made mobile by the high temperature of the com-
bustion zone. This phenomenon is called liquid block-
ing. Second, the temperature of the region in the back
of the combustion zone is high, indicating a great amount
of heat being stored in the region, not used efficiently.
Fig. 46.1-Steam injection processes.
Fig. 46.2-Dry forward combustion
couwslIoN ZONE lnAValNa
Fig. 46.3-Reverse combustion.
The lower part of Fig. 46.2 shows the fluid saturation
distributions inside the formation under the combustion
process. One should note the clean sand in the burned-
out region. Being able to burn the undesirable fraction
of the oil (the heavier portion) is one advantage of the
forward combustion process over the reverse combustion
Reverse Combustion
Strictly speaking, it should be called dry reverse combus-
tion, because normally only air is injected, no water. A
simple example will help to explain how reverse combus-
tion works. In ordinary cigarette smoking, one ignites the
tip of the cigarette and inhales. The burning front will
travel from the tip of the cigarette toward ones mouth,
along with the air. This is forward combustion. The
cigarette also can be burned if one exhales. This way, the
burning front still moves from the tip of the cigarette
toward ones mouth, but the air flow is in the opposite
direction. This is, then, reverse combustion.
Fig. 46.3 shows the various zones inside the formation,
with the cold zone near the injector at the left and the hot
zone near the producer.3 Since the region around the
producer is hot, the problem of liquid blocking mentioned
earlier in connection with the dry forward process has
been eliminated.
In principle, there is no upper limit for oil viscosity for
the application of the reverse combustion process. How-
ever, this process is not as efficient as the dry forward
combustion because a desirable fraction of the oil (the
lighter portion) is burned and an undesirable fraction of
the oil (the heavier portion) remains in the region behind
the combustion front. Besides, spontaneous ignition could
occur at the injector.4 If this happens, the oxygen will
be used up near the injector and will not support com-
bustion near the producer. The process then reverts to for-
ward combustion.
No reverse combustion project has ever reached com-
mercial status. Nevertheless, this process should not be
written off because, in spite of the difficulties facing this
process, it could offer some hope of recovering extremely
viscous oil or tar.
Wet Combustion
The term wet combustion actually refers to wet for-
ward combustion. This process was developed to use the
heat contained behind the combustion zone. In this proc-
ess, water is injected either alternately or simultaneously
with air. Because of its high heat capacity and latent heat
of vaporization, water is capable of moving the heat be-
hind the combustion front forward, and helping to dis-
place the oil in front of the combustion zone.
Fig. 46.4 shows the temperature distributions of the wet
combustion process as the water/air ratio (WAR) in-
creases.5 The curve for WAR=0 refers to dry combus-
tion. With an increase in WAR, the high-temperature zone
behind the combustion zone shortens (WAR=moderate).
With a further increase in WAR, the combustion will be
partially quenched as shown by the curve for
The wet combustion process also is known as the
COFCAW process, which is an acronym for combina-
tion of forward combustion and waterflood. This proc-
ess also can be construed as steamflood with in-situ steam
generation. It should be noted that this method cannot pre-
vent liquid blocking and its application is limited by oil
viscosity, as is the dry forward combustion.
Historical Development
The following lists chronologically some of the major
events that occurred in the development of the thermal
recovery methods.
A steamflood was conducted in Woodson, TX.6
A dry forward combustion proect was started in
Delaware-Childers field, OK.
A dry forward combustion project was conducted
in southern Oklahoma. 8
A reverse combustion project was initiated in Bel-
lamy, M0.9
The steam stimulation process was accidentally
discovered in Mene Grande Tar Sands,
Venezuela. O
Steam stimulation was started in Yorba Linda,
Wet combustion phase of a fireflood project was
started in Schoonebeek, The Netherlands. I2
Current Status
U.S. Oil Production by Enhanced Recovery Methods
The significance of the thermal recovery processes can
be seen from the April 1982 survey of the Oil and Gas
J. I3 As shown in Table 46.1, of the daily U.S. oil pro-
duction with EOR processes, 76.9% comes from steam
injection and 2.7% comes from in-situ combustion, total-
ling 79.6% obtained by thermal recovery processes. The
combustion process, although dwarfed by the steam in-
jection processes, accounts for more than double the pro-
duction of all the chemical floods combined, which
amounts to 1.2 % .
Geographical Distribution of Thermal
Recovery Projects
Potential for Incremental Recovery
Table 46.2, based largely on the 1982 survey, I3 shows
the geographical distribution of the steam injection proj-
According to Johnson et al., l5 vast energy resources ex-
ist in the tar sands in Venezuela and Colombia (1,000 to
ects in the world. Of the daily oil production from steam
injection processes, 71.7% comes from the U.S., 15.4%
1,800 billion bbl), Canada (900 billion bbl) , and the U. S .
(30 billion bbl). These tar sands should be a major target
Total thermal
Other chemicals
Total chemicals
CO, miscible
Other gases
Grand Total
288,396 76.9
10,228 2.7
- ~
298,624 79.6
902 0.2
2,587 0.7
580 0.2
340 0.1
4,409 1.2
21,953 5.9
49,962 13.3
71,915 19.2
374,948 100.0
Fig. 46.4-Wet combustion.
from Indonesia, 7.0% from Venezuela, and 3.0% from
Canada. In the U.S., California accounts for nearly all
the production, with small percentages coming from Loui-
siana, Arkansas, Texas, Oklahoma, and Wyoming.
The daily oil production by in-situ combustion is shown
in Table 46.3. Here, the U.S. accounts for 40.0% of the
total production, followed by Romania (26.0%), Canada
(22.1%), and Venezuela (10.8%). Of the U.S. produc-
tion, nearly one-half comes from California, one-third
from Louisiana, with the rest from Mississippi, Texas,
and Illinois.
Major Thermal Recovery Projects
The major thermal recovery projects, again based largely
on the 1982 survey, t3 are listed in Table 46.4.
Reservoirs Amenable to Thermal Recovery
Table 46.5 shows the ranges of reservoir properties in
which the technical feasibility of steamflood and tireflood
has been proven. I4
Canada (Alberta)
B/D %
288,396 71.7
TABLE 46.3-PRODUCTION BY IN-SITU COMBUSTION (1962) for development of thermal recovery methods, since the
results will be most rewarding if a percentage of these
resources can be tapped economically.
- -
10,228 40.0
5,690 22.1
284 1.1
2,799 10.8
6,699 26.0
- ~
25,760 100.0
Steam stimulation
Field, Location (Operator)
Kern River, CA (Getty)
Duri, Indonesia (Caltex)
Mount Poso. CA (Shell)
San Ardo. CA (Texaco)
Tia Juana Este,
Venezuela (Maraven)
Lagunillas, Venezuela
Duri, Indonesia (Caltex)
Cold Lake, Alberta (Esso)
Suplacu de Barcau,
Battrum No. 1,
Saskatchewan (Mobil)
Bellevue, LA (Getty)
Jobo. Venezuela
Enhanced Oil
Depth, ft
Steamflood Fireflood
160 to 5,000 180 to 11,500
Nei pay, ft 10to1,050 4 to 150
Dip, degrees 0 to 70 0 to 45
Porositv. % 12to39 16to39
Permeability, md 70 to 10,000 40 to 10,000
011 oravitv, OAPI -21044 9.5 to 40
011 iiscosity at initial
temperature, cp
Oil saturation at start, %
OOIP at start, bbl/acre-ft
4 to 106 0.8 to lo6
15to85 30 to 94
370 to 2,230 430 to 2,550
Based on an assumed oil price of $22.OiVbbl, Lewin
and Assocs. Inc. I6 estimated that the ultimate recovery
in the U.S. by thermal recovery methods will amount to
5.6 to 7.9 billion bbl. This includes 4.0 to 6.0 billion bbl
by steamfloods and 1.6 to 1.9 billion bbl by firefloods.
Production Mechanisms
The production mechanisms in steam inection processes
have been identified by Willman et al.
as (1) hot water-
flood, including viscosity reduction and swelling, (2) gas
drive, (3) steam distillation, and (4) solvent extraction ef-
fect. The relative importance of these mechanisms on light
and heavy oil, represented by 37.0 and 12.2 API, respec-
tively, is given in Table 46.6.
In &floods, the above mechanisms are also important.
In addition, the breaking up of heavy oil fractions into
light oil fractions through cracking should have at least
two effects: increase in volume and more drastic reduc-
tion in viscosity. The gas drive effect also should be in-
creased because of the large amount of air injected and
combustion gas produced.
Theoretical Considerations
Surface Line and Wellbore Heat Losses
In current field practice, downbole steam generators are
still in the developmental stage. Surface steam genera-
tors are being used in almost all of the steam injection
projects. Steam from a generator normally is sent to the
injector wellhead through a surface line. Some heat will
be lost to the surrounding atmosphere by convection and
radiation. As steam travels from the wellhead through the
wellbore to the sandface at the pay zone, heat will be lost
to the overburden, mainly by conduction. The method of
calculating surface line and wellbore heat losses is dis-
cussed below.
Surface Line Heat Losses
The steam lines in most of the steam injection projects
are insulated. The heat loss from such a line, Btuihr, is:
Qr,=2ari,U,;(T, -T,,)AL, (1)
= outside radius of the insulation surface, ft,
r, = steam temperature, F,
7,, = atmospheric temperature, F, and
ti = pipe length, ft.
In the above, CT,. is the overall heat transfer coefficient
(based on inside radius of the pipe or tubing), Btu/hr-ft-
F, and can be calculated as follows.
Uti =
+- h+l ) . . . . . . . . . . .
where rro is the outside radius of pipe, ft, and khin is the
thermal conductivity of insulation material, Bm/hr-sq
The convection heat transfer coefficient, h, Btuihr-sq
ft-F, can be calculated thus * :
h=0.75v,o~6/ril10~4, . . . . . , . . . . . . . (3)
where v,+ is the wind velocity, miihr. The radiation heat
transfer coefficient, I, normally can be neglected.
If the pipe is bare, that is, uninsulated, then J-,~ =rin
U,i=h. . . . . . . . . . . . . . . . . . (4)
If the steam is superheated, T, will vary along the line
as heat is being lost to the atmosphere. When the pipe
is long, it needs to be broken up into segments and the
heat loss calculated segment by segment. In each segment,
Ts* =T,$, -Qr,lwsc,s, . . . . . . . . . . (5)
T.vl ,Ts2 =
steam temperatures at the beginning and
the end of the segment, F,
QrI = heat loss along the segment, Btulhr,
w,~ = mass rate of steam, lbm/hr, and
C,, = heat capacity of steam, Btu/lbm-F.
If the steam is saturated, the heat loss will cause reduc-
tion in steam quality.
I,2 =f;, -Qr,lwsLs, . . . . . . . . . . . . .(6)
where f,i and fs2 equal the steam quality at the begin-
ning and the end of the pipe segment, fraction, and L,
is the latent heat of steam, Btu/lbm.
Wellbore Heat Losses
In most of the steam injection projects, saturated steam
at a certain quality is injected into the formation. Here,
we assume a more general case in which the steam first
enters the wellbore as superheated steam, becomes satu-
rated with a gradually diminishing quality, and is further
cooled after its complete condensation into hot water.
Superheated Steam. Assume that when the depth D is
0, the temperature of the steam is T, and varies with
time. Also assume that a linear geothermal gradient ex-
ists so that
Tf=gGD+ T,,,,, . . . . . . (7)
where 7 is the temperature of the formation. Suppose
one starts with the temperature of the steam at a depth
D r , and desires to calculate the temperature at depth Dl
with the length of the depth interval AD= 02 -D 1 . Since
the formation temperature at D is g G D , + T,Y,, Ramey s
equation for the gas case I9 becomes
+[T(D,,t)-gcD, -T,,+gGA+AB]e-hDJA.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A is defined as
A= w~C~[khf+rtiUdit)l
. . . . . . . . . . .
. (9)
2sr,i Urikhf
778c,, . . . . . . . . . . . . . . . . . . . . . . . . . . .
khf =
rtr =
ur, =
thermal conductivity of the formation,
f(r) =
inside radius of the tubing, ft,
overall heat transfer coefficient for the
annular space between inside of the
tubing and outside of the casing based
on rti, Btu/D-ft-F,
transient heat conduction time function for
earth, dimensionless, shown in Fig.
c, = heat capacity of steam, Btu/lbm-F,
gc =
geothermal gradient, F/ft, and
T.m =
surface temperature, OF.
For t>7 days,
f(t)=lnp -0.29, . . . . . . . . . . .
where 1y is the thermal diffusivity, sq ft/D, and rcO is the
outside radius of casing, ft.
Saturated Steam. When the steam is saturated, the well-
bore heat loss will cause changes in the steam quality
whereas the steam temperature, T, , is kept constant. If
Steam injectron pressure, psig
Hot waterflood recovery (Includes viscosity reduction and swelling)
Recovery from gas drive
Extra recovery from steam distillation
Recovery improvements from solventlextractton effects
Total recovery by steam
(% Initial 011 In Place)
Torpedo Sandstone Torpedo Sandstone
Core Core
37OAPI Crude 12 2OAPI Crude
800 (52OOF) 84 (327OF) 800 (52OOF) 84 (327F)
71.0 68.7 68.7 66.0
3.0 3.0 3.0 3.0
18.9 15.6 9.3 4.9
4.7 4.6 3.0 3.7
97.6 91 .9 84.0 77.6
c Vi to
Fig. 46.5-Transient heat conduction in an infinite radial system.
the steam quality at D isf, =f,(D t J), the steam quality
at 02 can be calculated by Satters equation*O:
fsP2J)=fs(D1 A+
AB+aD, +b-T,
a@@ *
+- . . . . . . . . . . . . . . . . . . . . . . . . (12) 2A
In Eq. 12,
. . . . . . . . . . . . . .(13) 2=rli Utrkhf
B=-. . . . . . . . . . . . . . . . . . . . . . . . . . .
Hot Water. For cooling of the hot water, Rameys equa-
tion for the liquid phase I9 applies. To advance from
depth D, to D2,
Wz,t)=g& +T.,, -gcA+[W,,t)-gcD,
+T,y,+g~A]e-aA. . . . . . (15)
Overall Heat Transfer Coeffkient. The temperature dis-
tribution in an annular completion is shown in Fig.
46.6. 2 To evaluate the overall heat transfer coefficient,
U,,, based on the outside tubing surface, the following
procedure developed by Willhite* can be used.
I. Select U,, based on outside tubing surface.
2. Calculate AI), as defined previously.
3. Calculate Tc. at cement/formation interface.
Tfi) +
rlo ulo
. . . . . . . . . . . . . . . . .
rfO (Ito
Fig. 46.6-Temperature distribution in an annular completion.
where Tfi =temperature of fluid, OF.
4. Calculate Tci at casing inside surface.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
r-d = radius to cement/formation interface, ft,
rc( = inside radius of casing, ft,
= thermal conductivity of the cement, Btu/hr-
ft-F, and
= thermal conductivity of the casing material,
5. Estimate I for radiation and h for natural convection.
6. Calculate U,,.
.,-(&+ rr;;ez)-l. . . . . . . . . . .(18)
With commercial insulation of thickness Ar,
rlo In? rlo lncf -
um =
rto +
. . . . . ..1.....................
where h and I are based on insulation outside surface.
Calculations Including Pressure Changes. A more
sophisticated calculation procedure proposed by
Earlougher** includes the effect of pressure changes in-
side the wellbore. The wellbore is divided into a sequence
of depth intervals. The conditions at the bottom of each
interval are calculated, on the basis of the conditions at
the top of that interval. The procedure is as follows.
1. Calculate the pressure at the bottom of the inter-
val, ~2.
p2=pI+1.687x10-2(v,, -v,2);
. . . . . . . . . . ,
; \
= specific volume of the total fluid, cu ft/lbm
(condition 1 is top of interval and 2 is
m = length of depth interval, ft, and 6m.L Ol,rUCT .Io IwccTIo* WCLL -
Ap = frictional pressure drop over interval, psi.
Fig. 46.7-Temperature distribution in Marx-Langenheim model.
The Beggs and Brill correlation23 for two-phase flow can
be used to calculate the Ap in the above equation.
2. Calculate the heat loss over the interval.
-0.5(Tf, +Tp)], . . . . . . . . . . . . . (21)
where U, is the overall heat transfer coefficient based
on outside casing surface, Btulhr-sq ft-F.
3. Calculate the steam quality at the bottom of the in-
terval .
where H,.] and H,,.2 are the enthalpy of liquid water at
top and bottom of the interval, Btu/lbm, and L,,, and L,,*
are the latent heat of vaporization at top and bottom of
the interval, Btu/lbm.
More Recent Developments. A new model has been de-
veloped by Farouq Ali that treats wellbore heat losses
rigorously by using a grid system to represent the sur-
rounding formation. In addition, the pressure calculation
accounts for slip and the prevailing flow regime, based
on well-accepted correlations.
Analytical Models for Steam Injection
For predicting reservoir performance under steam injec-
tion processes, the usual practice is to use three-
dimensional (3D), three-phase numerical simulators.
Where the simulators are unavailable or a quick estimate
of the performance is needed, one can resort to simple
analytical methods. Usually these methods take into ac-
count the thermal aspects of the process only, without
regard to the fluid flow aspects.
Front Displacement Models
Marx-Langenheim Method. 25 Consider that heat is in-
jected into a pay zone bounded by two neighboring for-
mations. The heat-carrying fluid is supposed to advance
with a sharp front perpendicular to the boundaries of the
formation (Fig. 46.7). The heat balance gives: heat in-
jected into the pay zone equals heat loss to the overbur-
den and underlying stratum plus heat contained in the pay
The heated area at any time t can be calculated
A= QriMhao
4k,,, 2AT
(coerfc\l;;; +2Js- 1) , (23)
A = heated area at time t, sq ft,
t = time since injection, hr,
Q,; = heat injection rate, Btu/hr,
M = volumetric heat capacity of the solid
matrix containing oil and water,
Btu/cu ft-F
. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4 = porosity, fraction,
Pr&otPw =
density of rock grain, oil, water,
lbm/cu ft,
C,,C,,C, = heat capacity of rock, oil, water,
S,i,Swi = initial saturation of oil, water, fraction,
h = pay thickness, ft,
= overburden thermal diffusivity, sq
kho = overburden thermal conductivity,
AT = Tid-Tf, F,
Tij = injection temperature, F,
Tr; = initial formation temperature, F,
TV = dimensionless time
= (;;;l,,)
1. .
&y=(~)tB? . . . . . . . . . . . . . . . .
The complementary error function is:
erfcx=l-erfx=l-1S~e-B2d13, . . . . . . . . (27)
where /3 is a dummy variable.
To evaluate eDerfc ,&, one can use the following
approximation. 26
Let y= 1+o.3275911JtD, . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . .
Assume that all the movable oil is displaced in the heated
area. If we assume that all the displaced oil is produced,
we can calcuate the cumulative steam/oil ratio (SOR):
4.275Ahr$(S,, -Sic,) . .
1, =
steam injection rate, B/D, cold water
so, =
initial oil saturation, and
s;, =
irreducible oil saturation.
Differentiation of the expression for A with t gives the
rate of expansion of the heated area. The oil displacement
rate, q(,d, in B/D, is
qod =4.275
Qrr woi -So,)
eroerfcJt,. . . (31)
From this one can calculate the instantaneous SOR
Flo=15. . . . . . . . . . . . . . . . . . . . . . . . (32)
It can easily be shown that
Eh =i eDerfc&+2 . . . . (35)
Rameys Generalization of the Marx-Langenheim
Method. The Marx-Langenheim method can be ex-
tended to the case where a series of constant injection rates
is maintained over various time periods. If the heat in-
jection rate is (Qri)i over the period O<t<r, , and (Q,,),,
over the period t,-1 <t<t,,
i=rl- I
+ C [(Q,i),-(Q,i);+IlF(rD;) 3 . . (36)
i= I
% . . .
and F(~D,)=F(CD~) with i=n. The oil displacement rate
at f; depends on the heat injection rate at that time, in-
dependent of the previous heat injection rates.
Mandl-Voleks Refinement of the Marx-Langenheim
Method.28 Mandl and Volek observed that the heated
area measured in laboratory experiments tends to be lower
than that predicted by the Marx-Langenheim method af-
ter a certain critical time, t,. For t? t,.,
A= QriMhao
eDerfcG+2*- 1 -dE
t D - t cD
Lsfs + 3
ezDerfc tD J-
t D - t cD
3J ?r t D
. . . . . . . . . . . . . . . . . . .
t, is determined by this equation:
ecD erfc& =
. . . . . . . . . ,+ Lag
The relationship between t, and t& is again
. (40)
The thermal (heat) efficiency, Eh, is defined as
Qh- = heat remaining in the heated zone, Btu,
Q;, = total heat injection, Btu, and
Eh = ~ . . . . . . . Q,iT
Myhill and Stegemeier used a slightly different ver-
sion of the Mandl-Volek model and calculated oil/steam
ratio (OSR) for 11 field projects. They found that the ac-
tual OSRs range from 70 to 100% of the calculated ratios.
Steam Chest Models
In contrast to the front displacement models discussed
previously, Neuman30 visualized that steam rises to the
top and grows both horizontally outward and vertically
downward. Doscher and Ghassemi3 took a view even
more drastic than Neumans. They theorized that steam
rises to the top instantly and the only direction of the steam
zone movement is vertically downward. Vogel12 fol-
lowed the same reasoning and developed the following
simple equation for thermal efficiency:
E,, =
. . . . . . . .
Table 46.7 compares the thermal efficiencies calculat-
ed by the Marx-Langenheim method and the Vogel
method. This table shows that the Vogel method predicts
a thermal efficiency that lies between 80 and 100% of that
calculated by the Marx-Langenheim method.
Steam Stimulation
Steam stimulation usually is carried out in a number of
cycles. Each cycle consists of three stages: steam injec-
tion, soaking, and production. The basic concept of this
process follows.
Without stimulation, the oil production rate is
4 oc =
(p, -p ,), . . . . . . . . (42)
4oc =
t-0 =
PC -
P\ , =
cold oil production rate, B/D,
absolute permeability. md,
relative permeability to oil, fraction,
cold oil viscosity, cp,
static formation pressure at external radius
r,, psia, and
bottomhole pressure, psia.
After steam injection, the oil inside the heated region,
r, < r< rh, will have a lower viscosity, p&. The hot oil
production, qoh, is:
qoh =
poh lnrh+poc Ink
(P, -P,), . . (43)
rw r h
Thermal Efficiency
Marx-Langenhelm Vogel Vogel/ML
0.01 0.930 0.900 0.967
0.1 0. 804 0. 737 0.917
1 .o 0. 556 0. 470 0. 845
10.0 0. 274 0.219 0.799
100.0 0.103 0.081 0. 787
where rh equals the radius of the heated region, ft. The
ratio between qoh and qoc is
? oh
. . . . .
lnh lnT
poh r h r r t , I
pm l nr , l nr ,
r w r w
As the reservoir fluids are produced, energy associat-
ed with the fluids are removed from the reservoir. This
causes a reduction in rh and a reduction in temperature,
which increases PO),.
Several methods have been developed for calculating
reservoir performance under steam stimulation. One of
the methods, which has enoyed wide acceptance, is the
Boberg and Lantz method. 3 This method assumes a con-
stant rh, with a changing T inside the heated zone. The
method consists of the following steps.
1. Calculate the size of the heated region using the
Marx-Langenheim method.
2. Calculate the average temperature in this region.
3. Calculate the oil production rate, taking into account
the reduced oil viscosity in this region.
4. Repeat Steps 1 through 3 for succeeding cycles, by
including the residual heat left from preceding cycles.
The average temperature of the heated region is calcu-
lated by
T=TR+(Ts-TR)[V,Vz(l-6)-h], . . . . .(45)
TR =
T, =
v,,v, =
V,,VI =
average temperature of the heated region,
r,<r<rh, at any time t, F,
original reservoir temperature, F,
steam temperature at sandface injection
pressure, F,
average values of V,, V, for O< r< rh and
all hi,*
unit solution for the component conduction
problems in the r and z directions, and
energy removed with the produced fluids,
The quantities s7]. and vZ can be obtained from Fig.
46.8 as functions of dimensionless time, tD. For v,.,
tD =
) . . . . . . . . . . . . . . . . . . . . . . .
r h
These symbols have no physical connolat~on. They are amply mathemailcal symbols
Ij1~11[ ~lllll
S;:t, =
aott-I,)..- %(t-4)
v,:t, =-
o.j#~v;#~ /
0.01 0.1 1.0 10 1ac
Fig. 46.8--Solutions for V, and V,, single sand.
a, =
overburden thermal diffusivity, sq ft/D,
time since start of injection for the current
cycle, D,
ti =
f-h =
For vz,
time of injection for the current cycle, D,
radius of region originally heated, ft.
tD= -
, . . . . . . . . . . . . . . . . . . . . . . . . . .
H, =
m~i t ( . f sLsf Hw3s - Hw~)
Wh 2WVs - TR) N,
) . . . . . . . . . . . . . .
= total mass of steam injected, lbm,
N, = number of sands,
H, vsJ , v~
= enthalpy, Btu/lbm, of water at steam
and reservoir temperatures, F, and
A4 = volumetric heat capacity, Btu/cu ft-F.
The energy removed with produced fluids, 6, can be
calculated thus:
h, = total thickness of all sands, ft,
Q, = heat removal rate at time t, BtulD,
Qr, = qoh(Hog+H,), . _ . . . . . . . . . . .
= (5.6146M, +R,C,)@-TR), . . . . .
H, = 5.6146p,[F,,(hf-H~,R)tRtL,], . . (52)
where hf is the enthalpy of liquid water at T above 32F
(see steam tables), BtuAbm, H, is the enthalpy of oil
and gas based on a STB of oil, BtulSTB oil, and H,. is
the enthalpy of water carried by oil based on a STB of
oil, Btu/STB oil. Also, L, is hfK in the steam tables.
If P Mz >P s and F,, <Fwclt
F,, =0.0001356 .-!A- R,, . . . . . . . .
( >
bbl liquid water at 60F/STB oil.
If F,, (calculated) > F,, ,
F,, =F,,,. . . . . . . . . . . . . . . . . . . . . . . (54)
In the above,
R, = total produced GOR, scf/STB,
= total produced WOR, STBISTB,
F,, = steam/oil ratio, STB/STB,
= producing bottomhole pressure, psia, and
ps = saturated vapor pressure of water at ?,
The rate of hot oil production can be calculated thus:
qoh=FJJcAp, . . . . . . . . . . . . . . . . . . . . . . . . ...(M)
where FJ is the ratio of stimulated to unstimulated
productivity indexes, dimensionless,
FJ= , . . . . . . . . . . . . . . . . . . . . .
EC1 tc2
and J, is the unstimulated (cold) productivity index,
J, =
. . . . . . . . . . . . . . . . . . . . .
pot lnT
If Pe is constant,
c, = rw
- . . . . . . . . . . . . . . . . . . . . . . . . .
c*= rh
-. . . . . . . . . . . . . . . . . . . . . . . . . .
Thus Eq. 55 is identical with Eq. 43 in this case. Ifp,
is declining,
c, =
rw 2re2
@a CP
Fig. 46.9-Correlation of steam stimulation results.
r h
, nr , - I +-
c2 =
rh 2 2re2
. . . . . . . . . . . . . . . . (61)
. ,
This method of calculating oil production rate is proba-
bly the weakest part of the Boberg-Lantz method.
1. It assumes a monotone decline betweenp, andp,.
Actually, because the injected steam is at a high pressure,
there could be a high pressure ps near rh and the pres-
sure declines toward both p ,+ and pe .
2. Only the change in p0 is accounted for in changing
from cold oil productivity to hot oil productivity. Left un-
accounted for is the change in k, , which should change
with changing S,.
Based on the Boberg-Lantz method, a correlation was
developed by Boberg and West34 that allows one to es-
timate incremental OSR with known reservoir properties
(Fig. 46.9).
Numerical Simulation
The analytical models for thermal recovery processes
usually are concerned with the thermal aspects of the proc-
esses only. The fluid flow aspects are neglected. To ac-
count adequately for the fluid flow inside porous media
under a thermal recovery process, numerical simulators
will be needed. In these simulators, the reservoir is divided
into a number of blocks arranged in one, two, or three
dimensions. A detailed study is made of the reservoir by
applying fundamental equations for flow in porous me-
dia to each one of the blocks.
Numerical reservoir simulators are no substitute for
field pilots. They have several advantages, however, over
field pilots. Field conditions are irreversible. It took mil-
lions of years for the field to develop to the present state.
Once disturbed, it cannot revert to the original conditions
and start over again. Furthermore, it takes a long time,
in terms of months or even years, before the pilot results
can be evaluated. The cost for pilots is, of course, enor-
mous. In comparison, a simulated reservoir can be pro-
duced many times, each time starting at the existing state.
This can be done within a short period of time, in terms
of seconds, once the reservoir model is properly set up.
The cost for reservoir simulation is much less than that
of a pilot. However, simulated reservoirs may never dupli-
cate field performance. Modem practice is to use reser-
voir simulation to help design a pilot before launching a
large-scale field development.
Numerical models and physical models are complemen-
tary to each other. As will be detailed later, physical
models can be classified into two types: elemental models
and partially scaled models. In an elemental model, ex-
periments are conducted with actual reservoir rock and
fluids. The results can help explain various fluid flow and
heat transfer mechanisms as well as chemical reaction ki-
netics. In a partially scaled model, reservoir dimensions,
fluid properties, and rock properties are scaled for the
laboratory model so that the ratios of various forces in
the reservoir and the physical model are nearly the same.
One can only build partially scaled models because fully
scaled models are difficult or impossible to construct. One
of the advantages of a numerical model over a physical
model is that there is no scaling problem in numerical
simulation. However, in many cases, a numerical model
needs physical models to validate the formulation or to
provide necessary input data for the simulation.
Steam Injection Model
Numerical simulation models for steam injection proc-
esses have been developed by Coats et a1.35 and
Coats. 36.37 A steam injection model consists of a num-
ber of conservation equations.
1. Mass balance of Hz 0. Both water and steam are in-
2. Mass balances of hydrocarbons. Only one equation
will be necessary for nonvolatile oil. For volatile oil, two
or more pseudocomponents will be needed to describe the
vaporization/condensation phenomenon of the oil and two
or more equations will be needed.
3. Energy balance. The energy balance accounts for
heat conduction, convection, vaporization/condensation
phenomenon, and heat loss from the pay zone to its adja-
cent formations. The need to include an energy balance
in the model sets the thermal recovery processes apart
from isothermal processes for oil recovery.
In addition to the conservation equations, the model
needs to include the following auxiliary equations.
1. If both water and steam coexist, temperature is the
saturated steam temperature for a given pressure. An
equation is needed to describe this relationship between
temperature and pressure.
2. The sum of saturations for the oil, water, and gas
phases equals unity.
3. The mol fractions of hydrocarbon components in the
liquid and gas phases are related through equilibrium
vaporization constants (K-values).
4. The sum of gas-phase mol fractions equals unity.
This includes steam and any volatile components of hydro-
5. The sum of liquid-phase mol fractions for hydrocar-
bons equals unity.
In-Situ Combustion Model
Numerical simulation models have been developed by
Crookston et al., 3g Youngren, 39 Coats,40 and Grabow-
ski et a1.4 The in-situ combustion model is more com-
plicated than the model for steam injection. The
conservation equations are as follows.
1. Mass balance of H20. This equation includes the
water produced from combustion.
2. Mass balances of hydrocarbons. This includes con-
sumption of certain hydrocarbons through cracking and
combustion. It also may include the production of cer-
tain other components through cracking.
3. Mass balance of oxygen. This accounts for the con-
sumption of oxygen by combustion.
4. Mass balance of inert gas. If air is used, the conser-
vation of nitrogen should be accounted for. CO2 pro-
duced from combustion may be included in the equation
for the inert gas or be treated separately.
5. Mass balance of coke. This includes the formation
and burning of coke.
6. Energy balance. This equation now includes the heat
of reaction for the reactions involved in the in-situ com-
bustion process. These reactions may include low-
temperature oxidation of hydrocarbons, high-temperature
oxidation or burning of hydrocarbons, thermal cracking
(which produces coke and other products), and combus-
tion of coke.
This model also needs a number of auxiliary equations,
which include (1) steam/water equilibrium, (2) vapori-
zation equilibrium of hydrocarbons, (3) phase saturation
constraints, (4) mol fraction constraints, and (5) chemi-
cal stoichiometry. An example is:
Oil+a 02 -+b CO2 +c Hz0
This says that one mol of oil reacts with a mols of oxy-
gen to form b mols of CO1 and c mols of HzO.
This model also requires a chemical reaction kinetics
equation. For each reaction involved in the process, an
equation can be written to denote that the reaction rate
varies as a function of temperature and concentrations of
the various reactants. One possible form of the reaction
rate equation is the following Arrhenius equation:
w=k(C,)m(CO,)n exp
This equation says that the reaction rate, w, is propor-
tional to the concentration of oil, C, , raised to the mth
power times the concentration of oxygen, Co, , raised
to the nth power. The temperature dependence ofthe reac-
tion rate is in the given exponential form, where E is the
activation energy, the energy barrier the reactants need
to overcome before being converted to the products, R
is the gas constant, and T is the absolute temperature. The
proportionality constant, k, usually is called the pre-
exponential factor.
The models developed so far are believed to be ade-
quate as far as the formulation of the process mechanisms
is concerned. However, problems abound.
1. Artificial breakdown of the crude oil into two com-
ponents may not be sufficient to describe faithfully the
vaporization/condensation phenomena and the chemical
reactions involved in the combustion process. More com-
ponents mean more equations to be solved and hence
higher computer costs.
2. The grid size problem could be severe. A grid size
large enough for economic computation could greatly dis-
tort the temperature distributions in the simulated reser-
voir. This would lead to erroneous predictions of the
chemical reaction rates and thus of reservoir performance
under combustion.
Laboratory Experimentation
The thermal numerical models have been used widely for
screening thermal prospects, designing field projects, and
formulating production strategies. Still, we cannot com-
pletely dispense with laboratory experiments for several
reasons. First, the numerical models need data that can
be measured only experimentally. These data include rela-
tive permeabilities, chemical kinetics, adsorption of chem-
icals on rocks, etc. Second, the numerical models are valid
only when all the pertinent mechanisms are accounted for.
The currently available models cannot handle adequately
situations such as injection of chemicals along with steam,
swelling of clays, which reduces the permeability, etc.
As previously mentioned, physical models for thermal
recovery processes may be classified into two types,
namely, elemental models and partially scaled models.
The elemental models are used to study the physico-
chemical changes inside a rock-fluid system under cer-
tain sets of operating conditions and are normally zero-
dimensional (OD) or one-dimensional (1D). The partial-
ly scaled models are used to simulate the performance of
a reservoir under thermal recovery operations and are nor-
mally 3D. Although the intent is to scale every physico-
chemical change that takes place in the processes, the
models usually are partially scaled because of the extreme
difficulty in achieving full scaling.
Elemental Models
Elemental models used for steamfloodin can be exem-
plified by those used by Willman er al.
In their clas-
sic work, they used glass bead packs and natural cores
of different lengths to study the recovery of oil under hot
waterflood and steamflood at different temperatures. The
oils used included crudes of different gravities and oil
Fireflood pots and combustion tubes are also elemen-
tal models. In another classic work, Alexander et ~1.~~
used fireflood pots (OD) to study fuel laydown and air re-
quirement, as affected by crude oil characteristics, porous
medium type, oil saturation, air flux, and time-
temperature relationships. The combustion tube (1D) used
by Showalter enabled him to delineate the temperature
profiles at various times, thus giving the combustion front
velocity. More recently, combustion tubes were used to
study the use of water along with airMA and the use of
oxygen-enriched air in combustion. 47
Partially Scaled Models
Partially scaled models have been used to simulate steam-
floods in 5/8 of a five-spot pattern, !4 of a five-spot pat-
tern, etc.48-53 Similar attempts have also been made for
tirefloods. However, it is certainly much more difficult
to include chemical kinetics along with the fluid flow and
heat transfer aspects of the combustion process.
Partially scaled models for steamfloods fall into two
types, namely, high-pressure models and vacuum or low-
pressure models.
High-Pressure Models. All experimental studies on
steamflooding had used high-pressure models until
vacuum models came along and offered an alternative ap-
proach. The scaling laws of Pujol and Boberg normally
were followed in the design. If the dimensions are scaled
down by a factor of F in the model, the steam injection
rate will be scaled down by the same factor and so will
the pressure drop between the injector and the producer.
The permeability will be scaled up by a factor of F, and
the model time will be scaled down by a factor of F*.
Because of the necessity of increasing the permeability
in the model to a great extent, reservoir rock material can-
not be used. Nevertheless, the experiments will be con-
ducted with the actual crude. Also, the steam pressure
and steam quality to be employed in the field will be used
in the model
Vacuum Models. In a small-scale physical model, the
thickness is reduced greatly as compared with that in the
field. To obtain the same gravitational effects as in the
field, the pressure drop from the injector to the producer
also must be reduced greatly. The vaporizationiconden-
sation phenomenon of water and hydrocarbons is governed
by the Clausius-Clapeyron equation, which involves
d In p, or Q/p. Thus, a decrease in the pressure drop (dp)
necessitates a corresponding decrease in the pressure (p)
itself. This is the rationale behind the vacuum-model ap-
proach developed by the Shell group as reported by
Stegemeier et al.
To see the differences between a high-pressure model
and the vacuum model, Table 46.8 has been prepared for
using both models to simulate a hypothetical field element
with a hypothetical oil. The entries for the high-pressure
models were based on the scaling laws of Pujol and
Bobergs and the entries for the vacuum model were
based on the work of Stegemeier et ~1.~~
Length, ft
Steam rate
Pressure 1, psia
Steam quality
Oil viscosity, cp
Temperature, OF
Pressure 2. psia
Steam quality
Oil viscosity, cp
Temperature, OF
High-Pressure Vacuum
Model Model
2 458 1,527
5 yrs 50 min 120 min
300 BID 144.7 cm3/min 263.1 cm3/min
400 400 2.70
0.80 0.80 0.082
3.0 3.0 23.6
445 445 137.5
100 100 1.24
0.80 0.80 0.108
6.3 6.3 38.2
328 328 108.9
The following observations can be made on the high-
pressure and vacuum models.
1. Neither the high-pressure model nor the vacuum
model can accurately simulate the capillary forces and the
relative permeability curves of the actual rock/fluid sys-
tem because, to obtain a very high permeability, actual
rock material is not being used.
2. The high-pressure model does not observe the
Clausius-Clapeyron equation, whereas the vacuum model
follows it to a large extent but not exactly.
3. To use the vacuum model, an oil has to be reconsti-
tuted to obtain the required oil viscosity/temperature rela-
tionship. This is completely different from the actual crude
in many physicochemical aspects, including its vapori-
zation/condensation behavior and chemical kinetics. In
contrast, a high-pressure model normally uses actual
Field Projects
Screening Guides
In dealing with oil prospects, the first step is to find out
whether the field in question can be produced by certain
recovery methods. Screening guides are useful for this
purpose. Screening guides for steamflood and fireflood
processes have been proposed by various authors includ-
ing Farouq Ali, 57 Geffen, 58 Lewin et al., 59 Iyo-
ho, 6oChu, 61-63 and Poettmann. @ These screening guides
are listed in Table 46.9.
A perusal of the various screening guides listed in Ta-
ble 46.9 shows that some of the earlier screening guides
were quite restrictive in selecting oil prospects. Such a
guide tends to minimize the error of the second kind, that
is, the risk of excluding some undesirable prospects. In
so doing, it tends to increase the error of the first kind,
that is, the risk of missing some desirable prospects. Re-
cent changes in the price structure of the crude oil and
improved technology helped to widen the range of applica-
bility for the steamflood and fireflood processes. This is
reflected in the less restrictive screening guides developed
in more recent years. However, in minimizing the error
of the first kind (erroneous rejection), the newer guides
may possibly increase the error of the second kind (er-
roneous acceptance). This should be borne in mind when
applying these guides to oil prospects.
Poettmann @
Lewn and
Year h D 4 h
1973 830 < 3,000 0.30 - 1,000 12 to 15 <I.000 0.15 to 0.22
1973 >20 <4,000 >lO >20 20.10
1976 >20 <5000 > 0.50 >I0 >I00 > 0.065
1979 30 to 400 2.500 to 5.000 >o 30 > 1,000
>0.50 10 to 20 200 to 1,000 >50 > 0.065
1993 >lO >400 >020 > 0.40 ~36 >0.06
1964 >020 >lW
1973 >lO > 500
1976 >lO > 500
1977 2022
1976 5 to 50 200 to 4,500 20.20 >3W
1976 10 to 120 z 0.20
1976 >lO >500 2 0.25
1962 >0.16 >lOO
250 <45 >I00 >005 for COFCAW only
>0.50 10 to 45 >20 >0.05
r0.50 524 < 1,000 >0.13 confidence llmlts approach
> 0 27 regression analysts approach
>050 10 to40 <1,ooo >20 > 0.077 for dry combustion
Reservoir Performance
Performance Indicators Common to Both Steamfloods
and Firefloods. Sweep Efficiency. The area1 and verti-
cal sweep of the steam front or burning front has pro-
nounced influence on the economics of the steamflood or
fireflood projects. Some reported sweep efficiencies of
the steamflood and fireflood projects are given in Table
46.10. 65-81 Whereas the volumetric sweep of steamfloods
varies from 24 to 99 % , that of firefloods appears to be
lower, ranging from 14 to 60%.
Field, Locatlon
Inglewood, CA65
Kern River CA66,67
Kern River, CA68-70
Midway Sunset, CA
El Dorado, KA73
Deerfield, MO 74
South Belridge, CA76
(General Petroleum)
Within Pattern Area
(2.75 acres)
Within Total Burned Area
(7.90 acres)
Sloss, NE-
South Oklahoma
Shannon Pool, WY
(Pan AmericanlCasper)
Areal Vertical Volumetric
60.0 50.0 30.0
- - 80.0
- 100.0 62.8 to 98.8
- 60.0 to 70.0
- <50.0
40.0 34.0
- 24.3 to 41.9
100 59.6 59.6
100 50.4 50.4
50 28 14
85 - 26
43 100 43
(>600 BIAF) (well spacings40 acres)
<lo no upper limit for reverse combustion
<45 <l.wa > 0.064 for wet combustion
<40 >lO >O.iO
Oil Recovery. Table 46.11 lists some of the reported
oil recoveries of steamflood and fireflood projects. 82-21
For the estimation of the oil recovery obtainable in a
steam injection project, the analytical methods discussed
previously can be used. As steam injection continues, the
thermal efficiency will gradually diminish and the instan-
taneous SOR will increase gradually. When this ratio
reaches a certain limit, further injection of steam will be-
come uneconomical and needs to be stopped. The cumula-
tive oil production at that time divided by the original oil
in place (OOIP) will give the oil recovery.
The oil recovery from a fireflood project can be calcu-
lated with the recognition that oil production comes from
both the burned and unburned regions (Nelson and
McNeil 122), Let Evb equal the volumetric sweep effcien-
cy of the burning front and ERu equal the recovery effi-
ciency in the unburned region. The overall oil recovery is:
E,b+(l-Evb)ERu, . . . . . (63)
where C,,, is the fuel content, Ibm/cu ft. In this equation,
the fuel consumed is taken to be a IOAP1 oil with a den-
sity of 62.4 lbm/cu ft.
The equation developed by Satman et al. 23 can be
used to calculate the oil recovery from a dry combustion
Y=47.0 0.427S, -O.O0135h-2.196 -!-
( >
..__,.....,.......... ,.......
rnP +vfb
x100 . . . . . .
[N,,(~S,)l(l -d) . . . . . . . .
Field, Location (Operator)
Smackover, AR (Phillips)e2~83
Kern River, CA (Chevronia4
Kern River, CA (Getty)68
Midway Sunset, CA (CWOD)85
Mount Poso. CA (Shell)*~87
San Ardo, CA (Texaco)
Slocum, TX (She11)89~30
Winkleman Dome, WY (Amoco)g
Tta Juana Estes, Venezuela (Maraven)g3-g5
Brea-Olinda, CA (Union)Q6,97
Midway Sunset, CA (Mobil)gB
Midway Sunset, CA (CWOD)99
South Belridge, CA (General Petroleum)76
South Belridge, CA (Mobil)00
Robinson, IL (Marathon)0~06
Bellevue. LA (Cities) lo7 lo8
Bellevue, LA (Getty)09~*
May Libby, LA (Sun)3
Heidelberg, MS (G$~14~5
Sloss, NE (Amoco)
Glen Hummel, TX \Sun)6,7
Gloriana, TX (Sun) l6
North Tisdale. WY (Continental)g
Suplacu de Barcau, Romania (IFPIICPPG)zo
Miga, Venezuela (Gulf)
Thermal Oil
(Q/o OOIP)
46.6 to 72.6
47.5, 51.2
28.1 *
31 .o
In the above equation,
ANp = cumulative incremental oil production, bbl,
Ve = fuel burned, bbl,
N = OOIP, bbl,
= cumulative air injection, lo3 scf,
= oxygen utilization efficiency, fraction, and
= oil in place at start of project, bbl.
Gates and Ramey 124 developed a correlation between
oil recovery and PV burned at various initial gas satura-
tion, on the basis of field data taken from the South Bel-
ridge tireflood project 76 and laboratory combustion-tube
data. This correlation, shown in Fig. 46.10, should be
useful in predicting current oil recovery as the fireflood
Changes in Oil Property. At the temperatures and pres-
sures prevailing in steamfloods, no changes in the oil prop-
erty are expected to occur because of any chemical
reactions. However, the properties of the recovered oil
could have been changed as a result of steam distillation.
In firefloods, of course, oil properties change considera-
bly because of thermal cracking and combustion, as well
as steam distillation. Changes in oil property in some of
the reported steamfloods and firefloods are shown in Ta-
ble 46.12. 125~30
Performance Indicator Pertaining to Steamfloods
Only. Steam Oil Ratio (SOR). The SOR, F,, , is the most
important factor characterizing the success or failure of
a steamflood project. Its reciprocal, the OSR, F,,v, also
is used commonly. In projects where oil is used as fuel
Fig. 46.10-Estimated oil recovery vs. volume burned.
for steam generation, 1 bbl of oil normally can generate
13 to 14 bbl (cold-water equivalent) of steam. Thus, the
highest SOR that is tolerable without burning more oil
than that produced is 13 to 14. For steamflood operation,
there are other costs than fuel alone. Because of this, steam
injection is normally terminated when the instantaneous
SOR reaches the level of eight or so. Ideally the overall
SOR should be around four. This corresponds to 3 to 4
bbl of oil produced per barrel of oil burned. 13 This ideal
case is, unfortunately, not normally achievable. The SOR
of the majority of the steamflood field projects falls into
the range of 5 to 7.
The following set of regression equations developed by
Chu62 can be used to estimate the SOR with known reser-
voir and crude properties.
1. For F,>5.0 (F,, <0.20),
+0.5 12O$S,). . . . . . . . . . . . . . . . . . (67)
2. For F,, ~5.0 (F,, 20.20),
-0.0005915~, - 14.79S, -O.O002938khl/~~, . (68)
D = depth, ft,
h = reservoir thickness, ft,
8 = dip angle, degrees,
CL0 =
oil viscosity, cp,
k = permeability, md, and
S, = oil saturation at start, fraction.
Another method of estimating F,v, has been given by
Myhill and Stegemeier, 29 based on the Mandl-Volek
46-l 6
Field, Location
Before After
Before After
Brea, CA Z
23.5 25.9
South Belridge, CA76
(General Petroleum)
West Newport, GAlz6
(General Crude)
12.9 14.2 07 2,700 800
120 540 200
160 120 54
15.2 20.0 60 4,585 269
100 777 71
East Venezuela
(Mene Grande)
210 32 IO
9.5 12.2
then 10.5
Kyrock, KY
South Oklahomaso
Asphalt Ridge, UT13
(U.S. DOE)* *
14.2 20.3
Changes I 00 C ~ -C ,* before-21 after-28
Changes in other properf,es
Before After
Pour point. OF 140 25
Residue bolllng above 62 35
1 ,OOOOF, wf%
Performance Indicators Pertaining to Firefloods Only.
Fuel Content. Fuel content (lbm/cu ft of burned volume)
is the amount of coke available for combustion that is
deposited on the rock as a result of distillation and ther-
mal cracking. It is the most important factor influencing
the success of a fireflood project. If the fuel content is
too low, combustion cannot be self-sustained. A high fuel
content, however, means high air requirement and pow-
er cost. Besides, oil production also may suffer.
where C, is the fuel content, lbm/cu ft.
Both laboratory experiments and field projects indicate
that, for a specific reservoir, fuel content decreases as
WAR increases. However, no statistically significant
correlation was found to exist between fuel content and
WAR in the presence of widely varying reservoir prop-
erties . 63
Fuel content can be determined by laboratory tube runs.
Gates and Ramey 24 presented a comparison of the esti-
mated fuel content by use of various methods including
laboratory experiments and field project data from the
South Belridge project. 76 Their comparison shows that
fuel content determined from the tube runs can provide
reasonably good estimation of the fuel content obtaina-
ble in the field.
Air Requirement. As pointed out by Benham and Poett-
mann, 132 air requirement, a, in lo6 scf/acre-ft of burned
volume, can be calculated on the basis of stoichiometric
2Fcc+l +FHC
- ctn
FCC+1 2
x0.04356, . .(70)
In the absence of experimental data, the correlation of
Showalter relating the fuel content to API gravity can
be used. Fig. 46.11 shows a comparison of the Showalter
data and field project data. 63 In addition, the following
regression equation developed by Chu6* based on data
from 17 field projects can be used to calculate the fuel
C, = -0.12+0.00262h+0.0001 14k+2.23S0
+0.000242kh/p, -0.0001890-0.0000652,u0,
. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
where F, is the CO2/CO ratio in produced gas and FHC
is the atomic H/C ratio. In the absence of necessary data
for Eq. 70, the Showalter correlation43 relating air re-
quirement to API gravity can be used. A comparison of
the Showalter data and field project data is given in Fig.
46.12. 63 It can be seen that all the field points fall on the
upper side of the Showalter curve. Air requirement in the
fields can exceed laboratory values because of air chan-
neling and migration. In addition, the following regres-
sion equation developed by Chu63 can be used:
a=4.72+0.03656h+9.996S3,+0.000691k. . (71)
90,000 2,000
120 27
5,000 800
after a month
10 20 30 40
Fig. 46.11-Effect of oil gravity on fuel content.
Air-Oil Ratio (AOR). This important ratio relates air
injection to oil production and usually is expressed in
terms of lo3 scf/bbl. Oil recovery comes from both the
burned and unburned regions. The AOR can be calculat-
ed thus I** :
Fm =
. . . . . . . . . . . . . . . . . . . . . . . .
In the absence of Evb and ER,, , the following regression
equation developed by Chu6* based on 17 field projects
can be used.
F,, =21.45+0.0222h+0.001065k
+0.002645~, -76.76gbS,. . . . . . (73)
Besides, the correlation between oil recovery and PV
burned developed by Gates and Ramey 124 can be used
for estimating the current AOR as the fireflood proceeds.
Both laboratory experiments and field projects indicate
that, for a specific reservoir, AOR decreases as WAR in-
creases. No statistically significant correlation, however,
has been found between AOR and WAR in the presence
of widely varying reservoir properties. 63
Project Design
Design Features Common to Both Steamfloods
and Firefloods
Pattern Selection. For any oil recovery process with fluid
injection, a cardinal rule of pattern selection is that, to
achieve a balance between fluid injection and production,
25 I / I
Fig. 46.12-Effect of oil gravity on air requirement.
the ratio of the number of producers to the number of in-
jectors should be equal to the ratio of well injectivity to
well productivity (Caudle et al. 133). Because of the high
mobility of air or steam compared to that of the oil, the
injectivitylproductivity ratio is high, favoring a high
producer/injector ratio. This rule generally has been fol-
lowed by the various reported steamflood and fireflood
projects. The use of inverted 13-spot, 9-spot, 7-spot, and
6-spot patterns, unconfined five-spot patterns, down-the-
center line of injectors, and single well injection has been
Aside from the injectivity/productivity ratios, other fac-
tors also should enter into consideration in pattern selec-
tion. These factors include: heat loss considerations,
utilization of existing wells, reservoir dip, difficulty in
producing hot wells, etc. Based on these and other con-
siderations, repeated five-spot patterns, updip and crest
injections and line drive also were used. The choice of
pattern or nonpattern floods in the various steamflood and
fireflood projects is shown in Table 46.13. 34-138
Completion Intervals. In most of the steamflood and fire-
flood projects, the producers usually are completed for
the entire sand interval to maximize production. The in-
jectors usually are completed at the lower third or lower
half of the interval, to minimize the override of the steam
or air. In wet combustion projects, it is advisable to com-
plete the lower part for air injection and upper part for
water injection. This is to minimize the underflow of water
as well.
Producer Bottomhole Pressure (BHP). In their study
for a steamflood, Gomaa et al. 139 found that decreasing
the producer BHP lowers the average reservoir pressure,
increases steam volume, and increases predicted oil recov-
ery. It is, therefore, important to keep the producers
pumped off all the time. Without any reason to believe
otherwise, keeping the producers pumped off should
benefit a fireflood as well as a steamflood.
Pattern Types
Inverted 13-spot
Inverted g-spot
Inverted 7-spot
inverted 5-spot
of injectors
Single well injection
Repeated 5-spot
Updip or crest
Downdip injection
Updip and downdip
Line drive
Slocum, TX8g,93
San Ardo, CA8*
Yorba Linda, CA
Kern River, CAM
Slocum, TXm.
Ti!?i!ra, Venezuela35
East Coalinga. CA37
Ke?%er. CAa4
Kern River. CA870
( GeW
Winkleman Dome, WY .
(Pan American)
Brea, CAiz5
Midway Sunset, CA7.72
South Selridge, CAi3
Mount Poso. CA8.87
Design Features Pertaining to Steamfloods Only
SteamInjection Rate. According to Chu and Trim-
ble, I40 the optimal choice of a constant steam injection
rate is relatively independent of sand thickness. As sand
thickness decreases, the total oil content in the reservoir
decreases. This calls for a lower steam rate. At the same
time, a higher steam rate is needed to compensate for the
increased percentage heat loss with a decrease in thick-
ness. These two counteracting factors result in only a small
variation in the optimal steam rate as thickness is changed
from 90 to 30 ft.
The same study with five-spot patterns shows that the
optimal choice of a constant steam rate is proportional
to the pattern size. Furthermore, varying steam rates ap-
pear to be preferable to constant steam rates. An optimal
steam rate schedule calls for a high steam rate in the ini-
tial stage and a decrease in the steam rate with time.
Sellevue, LA07.08
(Cities Service)
Sellevue, LA0g~2
Silverdale, Alta. 34
(General Crude)
West Newport, CA26.27
(General Crude)
Trix-Liz, TX6~36
Gt!sf?~)ummel, TX6,7
Miga, Venezuela*
Sloss, NE77-79
Midway Sunset, CA
Heidelberg, MS4,15
Suplacu de Sarcau,
Romania (IPF/ICPPG)20
Steam Quality. Steam quality refers to the mass fraction
of water existing in vapor form. Gomaa et al. 39 reported
that increasing steam quality increases oil recovery vs.
time but had little effect on recovery vs. Btus injected.
This indicates that heat injection is the important param-
eter in determining steamflood performance.
Just as with steam injection rates, the optimal choice
of steam quality should be studied. High-quality steam
could cause excessive steam override. This may be reme-
died by using lower-quality steam at one stage of a
Design Features Pertaining to Fireflood Only
Dry vs. Wet Combustion. The choice between dry com-
bustion and wet combustion is an important decision to
be made in conducting a field project. Laboratory experi-
ments indicated that the use of water either simultaneously
or alternately with air does reduce the AOR, although the
oil recovery may not be improved significantly. As was
mentioned previously, a correlation between AOR and
WAR, based on data from 21 field projects, was found
to be statistically insignificant in the presence of widely
varying reservoir properties. 63
Cities Service conducted a field comparison test of dry
and wet combustion in the Bellevue field, LA, 14 in
which possible interference by variations in reservoir
properties was essentially circumvented by using two con-
tiguous patterns, one with dry combustion and another
with wet combustion. This test found that, with wet com-
bustion, the volumetric sweep was improved to a great
extent. This indirectly implies an increase in oil recov-
ery. Furthermore, the air requirement for a specific
volume of reservoir was reduced. This reduced the oper-
ating cost and improved the economics. Because of these
encouraging results, the possible advantages of using wet
combustion should be explored.
Air Injection Rate. According to Nelson and McNeil, 22
the air injection rate depends on the desired rate of ad-
vance of the burning front. A satisfactory burning rate
was stated to be 0.125 to 0.5 ft/D. In the design method
proposed by these authors, a maximum air rate is first
determined, based on the minimum burning rate of 0.125
ft/D. They recommended a time schedule such that the
air rate would be increased gradually to the maximum
rate, held at this rate for a definite period, and then re-
duced gradually to zero. The Midway Sunset, CA, proj-
ect of Chanslor-Western99 used a burning rate of 1 in./D
(0.08 ft/D). Gates and Ramey 24 found that the air rate
should provide a minimum rate of burning front advance
of 0.15 ft/D or an air flux of at least 2.15 scfihr-sq ft at
the burning front.
WAR. The reported WAR in various field projects ranged
from 0 (for dry combustion) to 2.8 bbl/103 scf. The
choice of WAR depends on water availability, quality of
the water available, well injectivity, and economic con-
siderations. Combustion tube experiments, properly de-
signed and executed, should be helpful.
Well Completion
Special well completions are needed for injectors and
producers to withstand the high temperatures in steam-
floods, and to withstand the corrosive environment as well
in firefloods.
According to Gates and Holmes, 14* wells used in
steam operations should be completed with due consid-
eration of heat loss with thermal stresses. In deep wells,
tubular goods with high qualities, such as the normalized
and tempered P-105 tubing and P-110 casing, should be
used if the tubing and casing are not free to expand. Ther-
mal stress can be minimized by the proper use of expan-
sion joints. Thermal packers should be used on steam
injection wells and deep wells undergoing cyclic steam-
ing. The cement should include a thermal strength stabiliz-
ing agent, an insulating additive and a bounding additive.
For firefloods, Gates and Holmes42 felt that steel
casing and tubing such as J-55 is suitable for injectors.
These wells can be completed with normal Portland ce-
ment, with high-temperature cement placed opposite and
about 100 ft above the pay zone. The high-temperature
cement recommended for the injectors is calcium
aluminate cement (with or without silica flour), pozzolan
cement, or API Class G cement (with 30% silica flour).
If spontaneous ignition occurs, the use of cemented and
perforated liners is required to prevent well damage re-
sulting from burnback into the borehole. The producers
should be completed to withstand relatively high temper-
atures and severe corrosion and abrasion. These authors
recommended the use of gravel-flow pack, and stainless
steel 316 for both liner and tubing opposite the pay zone.
The well completion methods for injectors and
producers in the various steamflood and fireflood proj-
ects, detailed by Chu6,63 previously, are given in Ta-
ble 46.14.
Field Facilities
Steamflood Facilities
Steam Generation and Injection. Most of the steam in-
jection projects use surface steam generators. The major
difference between oilfield steam generators and indus-
trial multitube boilers is the ability to produce steam from
saline feedwater with minimum treatment. Other features
include unattended operation, portability, weatherproof
construction, and ready accessibility for repairs. The abil-
ity to use a wide variety of fuels including lease crude
is also an important requirement. The capacity of steam
generators used in steamflood projects usually ranges from
12 to 50 X lo6 Btuihr, with 50 x lo6 Btuihr becoming the
industry standard in California.
With surface steam generators, the steam goes from the
generators to the injection wells through surface lines.
Most surface steam lines are insulated with a standard in-
sulation with aluminum housings. The steam is split into
individual injectors through a header system using chokes
to reach critical flow. This procedure requires that the
steam achieve sonic velocity, which, under one field con-
dition, 68 calls for a pressure drop of about 55 % across
the choke. The chokes are sized to each other to give the
desired flow rate into each injector. As long as the pres-
sure drop is greater than 55 X , the flow rate will be in-
dependent of the actual wellhead injection pressure.
A recent development is the use of downhole steam
generators to eliminate wellbore heat losses in deep wells.
There are two basic designs, which differ on the method
of transferring heat from the hot combustion gases to
produce steam. 43 In one design, the combustion gas
mixes directly with feed water and the resulting gas/steam
mixture is injected into the reservoir. Because of this, the
combustion process takes place at the injection pressure.
In another design, there is no direct contact between the
combustion gas and water, just as in the surface genera-
tors. The combustion gas returns to the surface to be
released after giving up much of the heat to generate
steam. A lower pressure than injection pressure can be
used in this case.
Still another new development is cogeneration of steam
and electricity. I44 The effluent gas from a combustor is
used in a gas turbine, which drives an electrical genera-
tor. The exhaust gas from the turbine is then used in steam
generators to produce steam for thermal recovery
Openhole or perforated
Gravel packmg
Openhole or perforated
Gravel packing
Grades: J-55, K-55, and N-80
Sizes: 4%, 5%, 65/8, 7, and 9% in.
Tensile prestressing of casing in deep wells
Both openhole completions with slotted lmers and
solid-string completion with jet perforations
have been reported.
Liner sizes: 4X, 5/2, or 7 in.
Perforations: l/4 or % In., one or two per foot, or
one-half per foot
Some with stamless steel wire-wrapped screens.
Class A, G, and H cement with silica flour (30 to
60% of dry cement).
Use not prevalent
Tubing insulations used in deep wells: asbestos
with calcium silicate, plus alummum radiation
sheld; or jacketed tubing with calcium silicate.
Grade: K-55
Sizes: 4%, 5%, Ss/,, 7, and 85/8 in.
Tensile prestressing of casing in deep wells
Both openhole completion with slotted liners and
solid-string completion with jet perforations
have been reported.
Liner sizes: 4/2, 4X, or 65/a in.
Slot sizes: 40, 60, or 601180 mesh
Perforations: % in., four per foot
Some with stamless steel wire-wrapped screens.
Class G and H cement with silica flour (30 to
60% of dry cement).
Use more prevalent than in injectors
Gravel size: 6/9 mesh flow-packed.
Tubing for rod pump.
Water Treatment. The feedwater treatment for steam
generation consists mainly of softening, usually through
zeolite ion exchange. Some feedwaters may require filtra-
tion and deaeration to remove iron. Still others may need
to use KC1 for control of clay swelling and chlorine to
combat bacteria. Facilities for oil removal also will be
needed if the produced water is to be reused as feedwater
for steam generation.
Fireflood Facilities
Ignition Devices. In many fields, the reservoir tempera-
ture is so high that spontaneous ignition would occur only
a few days after starting air injection. In some projects,
steam, reactive crude, or other fuels will be added to help
Many other fields need artificial ignition devices, which
include electrical heaters, gas burners, and catalytic ig-
nition systems. The various ignition methods, including
equipment and operational data, have been discussed by
Strange. 145
Air Compressors. The air compressors can be gas en-
gine or electrical motor driven. Depending on the total
Grades: J-55 and K-55
Sizes: 41/z, 5%. 7, and 8% in. across the
pay zone
Perforated completion more prevalent than
openhole completion with or without liners.
Liner sizes: 3% or 5% in.
Perforations: l/4 or l / 2 in. (two or four per foot)
Use of high-temperature cement prevalent
Use not prevalent.
Tubing used for air injection or as a thermowell.
In wet combustion, various ways have been used
for injection of air and water.
Grades: H-40, J-55, and K-55
Sizes: 51/z, 7. 85/8, and 95/a in.
Openhole completion with or without slotted liners
and perforated completion are equally
Liner sizes: 4%, 512, or 65/8 in.
Slot sizes: 60-mesh. 0.05, 0.07, or 0.08 in.
Perforations: V2 In. (two or four per foot)
Use of high-temperature cement was reported
Use more prevalent than in Injectors.
Gravel sizes: 20/40 or 619 mesh, flow- or
Tubing for rod pump, to serve as a thermowell. or
for cooling water injection.
injection rate the compressor needs to supply and the out-
put pressure needed, the capacity of the compressors can
range from 1.0 to 20.0~ lo6 scf/D, and the power rat-
ing can range from 300 to 3,500 hp.
Monitoring and Coring Programs
Monitoring Programs
A thermal recovery project could be a complete failure
economically and still be considered a success if it could
provide useful information on the reservoir performance
under steamflood or fireflood. A properly designed
monitoring program carried out during the project and
coring programs during and after the project are impor-
tant in providing the information necessary for evaluat-
ing steamflood or fireflood performance.
The Sample, Control, and Alarm Network (SCAN) au-
tomation system installed by Getty in the Kern River
field 46 illustrates how a large steam injection operation
can be monitored. This system consists of a devoted cen-
tral computer that monitors 96 field sites. At these sites,
the production rates of more than 2,600 producers and
the operating rates of 129 steam generators are gathered.
The SCAN oerforms several functions.
1. It automatically schedules and controls well produc-
tion tests at each site.
2. It monitors results of well production tests, steam
generator operating rates, flow status, and injection sta-
tus of producers, valve positions during well tests, and
various status contact checks.
3. It sounds the alarm upon any malfunctioning at a field
site or a steam generator.
4. It reports necessary operating information routinely
on a daily, weekly, or monthly basis, and other special
reports on demand from the operator.
The Silverdale, Alta., fireflood project of Genera1
Crude 34 also uses an automatic data collection system.
Differential pressure transducers, thermocouple-amplifier
transducers, pressure transducers, and motor load trans-
ducers are used to measure and record data at each well.
These data are transmitted to a central system, which can
be interrogated and can indicate any alarm situation when
pressures, temperatures, or flow rates fall outside certain
specified ranges.
Not all thermal projects call for elaborate automatic
monitoring programs. The following program used in the
Bodcau, LA, fireflood project of Cities Service-DOE I47
typifies one needed for a small-scale pilot.
1. Gas production rates, useful for mass balance cal-
culations, were measured monthly. Monthly analysis of
the produced gas gave data for the calculation of the oxy-
gen utilization efficiency.
2. Oil and water production rates were measured at least
twice each month.
3. Flow line temperatures were measured daily. These
temperatures, in conjunction with the gas production rates,
were useful in determining the amount of quench water
needed at the producers,
4. Downhole temperature profiles were taken monthly
at the observation wells. These profiles helped to deline-
ate the development of the burned volume.
Coring Program
Drilling core holes could be very expensive, depending
on the depths of the pay zones. However, a judiciously
designed and properly executed coring program, either
during a thermal project or afterward, could provide valu-
able information on the project performance. Such a pro-
gram can give the following information: (1) residual oil
saturation (ROS) after steamflood or fireflood, (2) verti-
cal sweep of the injected steam or burned volume, (3) areal
sweep of the steam front or burning front, (4) maximum
temperature distribution, both areally and vertically, and
(5) effective permeability of the rock, and whether any
deposits formed during the process could have reduced
the flow capacity.
A typical coring program, used for postmortem evalu-
ation in the Sloss, NE, fireflood project,79 is summar-
ized next.
Core Analyses. Porosity, permeability, and oil satura-
tions were measured on each foot of the recovered cores.
Oil saturations were determined by the routine Dean-Stark
extraction and weight loss method, and the infrared ab-
sorption method.
Log Analyses. Compensated formation density and dual-
induction laterolog logs were run in the core holes to de-
termine porosity and oil saturation.
Photographs and Visual Examination. Whereas black-
and-white photographs were found to be rather useless,
ultraviolet photographs gave an excellent picture as to
where the oil was removed by the burning process. The
absence of oil also could be seen by visual examination.
In some intervals, the reddish color of the core indicated
that the core had been subjected to a temperature high
enough for iron oxidation.
Mineral Analyses of the Cores. Various minerals, in-
cluding glauconite, illite, chlorite, and kaolinite, under-
went permanent changes with the temperature increase.
The maximum temperature to which the core samples had
been exposed could be determined from the form and
color of these minerals.
Microscopic Studies. The scanning electron microscope
was used to study anhydrite formation and clay altera-
tion in the core samples, which had been subjected to high
The use of tracers helped to monitor fluid movement and
interpret areal coverage in individual steamflood patterns.
According to Wagner, I48 preferred aqueous-phase or
gaseous-phase tracers include radioisotopes, salts with de-
tectable cations and anions, fluorescent dyes, and water-
soluble alcohols. Radioactive tracers include tritium,
tritiated water, and krypton-85. Other tracers include am-
monia, air, sodium nitrite, sodium bromide, and sodium
Operational Problems and Remedies
Operational problems plaguing steamflood and fireflood
projects and their remedies, previously detailed by
Chu . 6 ,63 are summarized next.
Problems Common to Steamfloods and Firefloods
WeII Productivity. Production of the highly viscous crude
may be extremely low before the arrival of the steam front
or burning front. The production rate can be improved
by injecting light oil as a diluent, hot oil treatment, cy-
clic steam injection, or burning at the producers.
When producer temperature exceeds 250F, pump ef-
ficiency decreases to a great extent because of hot pro-
duced fluids flashing to steam or direct breakthrough of
the injected steam or flue gas. The best remedy is to plug
off the hot zone and redirect the steam or flue gas to the
oil section before entering the wellbore.
Sanding. Sanding can be severe even in steamflood proj-
ects. The remedies include the Hyperclean technique,
foamed-in tight-hole slotted liners, a sodium aluminate
sand consolidation technique, and the use of phenolic-resin
gravel packing.
In firefloods, sanding is particularly severe if the sand
is extremely unconsolidated. The erosion can be aggra-
vated further by coke particles and high gas rates. Sand-
blasting could require frequent pulling of wells and
replacement of pumps.
Emulsions. In steamfloods, emulsions sometimes can be
broken easily by chemical treatment. The problem could
Fig. 46.13-Production history of cyclic steam stimulation, TM
sand, Huntington Beach offshore field, CA.
become severe if the emulsion is complicated with the
solids produced and with the continuously changing na-
ture of the produced fluids.
Emulsions found in fireflood projects are formed of
heavy oil, cracked light ends, quench and formation water,
solids, and possibly, corrosion products. They can become
a continual and major problem in some projects, and re-
quire expensive emulsion breakers.
Problems Plaguing Steamfloods Only
Steam Placement. The lack of control of steam place-
ment during steam stimulation is a major problem in
producers with liner completions. The use of solid string
completions will help reduce the problem.
Steam Splitting. The uneven splitting of steam in a two-
phase regime can cause significant differences in steam
quality into different injectors. This can be corrected by
modifying the layout of the steam line branching system.
Depth, fl
Thickness, ft
Porosity, %
Permeability, md
Oil gravity, OAPI
Reservoir temperature, OF
Reservoir pressure at start,
Oil viscosity at 12?F, cp
Oil saturation al start, %
2,000 to 2,300
40 to 58
400 to 800
12 to 15
wig 600 to 800
Problems Plaguing Firefloods Only
Poor Injectivity. Various substances can cause losses in
injectivity for the air injectors. If identifiable, these prob-
lems can be remedied by appropriate means. Injector plug-
ging by iron oxide can be reduced by injecting air into
the casing and bleeding it through the tubing. Asphaltene
buildup can be reduced by squeeze washing with
asphaltene solvent. Emulsion formed in situ can be re-
duced by emulsion breakers. Scale formation caused by
barium and strontium sulfate can be reduced by an or-
ganic phosphate. The injection of NuTriT
(trichloromethylene) and acidizing are useful in improv-
ing the injectivity.
Corrosion. Corrosion can be mild or serious and is caused
by simultaneous injection of air and water, production of
acids, sulfur, oxygen, and CO2 . Corrosion inhibitors are
needed regularly.
Exploration Hazards. To minimize explosion hazards
in the air injection system, an explosion-proof lubricant
should be used. Flushing of the interstage piping with a
nitrox solution is necessary.
Case Histories
Many thermal recovery projects have been reported in
the literature. The following describes a number of select-
ed projects and gives the reasons for their selection.
Steam Stimulation Operations
Huntington Beach, CA (Signal) 149-TypicaI Opera-
tion. The steam stimulation project was conducted in the
TM sand, in the Huntington Beach offshore field, Orange
County, CA, This project typifies the behavior of a heavy-
oil reservoir under cyclic steam stimulation. The reser-
voir properties are given in Table 46.15.
Steam injection was started in nine producers in Sept.
1964, resulting in a large increase in oil production. This
early success prompted the expansion of the project by
drilling wells on 5-acre spacing. The number of wells in-
creased from 9 in 1964 to 35 in 1969. The performance
of the steam stimulation project during the 1964-70 peri-
od is shown in Fig. 46.13. With steam stimulation and
with the almost quadrupling of the number of wells, the
oil rate increased more than lo-fold, from 125 B/D oil
in 1964 to about 1,500 B/D oil in 1970.
The performance of steam stimulation normally deteri-
orates as the number of cycles increases. As shown in
Table 46.16, the OSR changed from the range of 3 to 3.8
bbl/bbl for the first two cycles to the range of 2.4 to 2.5
bbl/bbl for the third and fourth cycles.
Fig. 46.14 shows how oil production in one well
decreases during a cycle and how it varies from one cycle
to another.
Paris Valley, CA (Husky) 50-Co-Iqjection of Gas and
Steam. A wet combustion project was initiated at Paris
Valley, which is located in Monterey County, CA. Be-
fore the arrival of the heat front, the producers were stimu-
lated with steam. A special feature that made this project
interesting was the co-injection of air and steam in three
of the stimulation cycles. The reservoir properties are
given in Table 46.17.
Fig. 46.14-011 production rate, Well J-128, Huntington Beach
offshore field.
In Table 46.18, Cycles 3 and 5 of Well 20 and Cycle
7 of Well 3 used air-steam injection. For Well 20, oil pro-
duction in Cycle 3 was 4,701 bbl while that in Cycle 2
was 2,449 bbl. Thus, with air-steam injection, oil pro-
duction increased by 92 % . A similar increase was notice-
able for Cycle 5 of Well 20 and Cycle 7 of Well 3 when
compared with their respective preceding cycles, which
used steam only.
Steamflood Projects
Kern River, CA (Getty) 68m70-Largest Steamflood.
The Kern River field is located northeast of Bakersfield,
CA, in the southeastern part of the San Joaquin Valley.
Getty Oil Co.s steam displacement operation in this field
is the largest in the world, based on a 1982 survey. l3 Ac-
cording to this survey, the thermal oil production rate was
83,000 B/D in an area of 5,070 acres.
The Kern River formation consists of a sequence of al-
ternating sand and shale members. The reservoir proper-
ties are given in Table 46.19.
The Kern River field was discovered in the late 1890s.
In the mid-1950s, bottomhole heaters were used to im-
prove the oil productivity. In Aug. 1962, a 2.5-acre nor-
mal five-spot hot waterflood was started. Results showed
that this process was technically feasible but economically
unattractive. In June 1964, the hot waterflood pilot was
converted to a steam displacement test and the number
of injectors was increased from the original 4 wells to 47
wells. Continued expansion through the years has in-
creased the number of injectors to 1,788 wells, with 2,556
producers by 1982. The original Kern project and some
later expansions are shown in Fig. 46.15. The steam dis-
placement operation was in general conducted in 2.5-acre
five-spot patterns.
Getty Oil Co.s steam displacement operation includes
many projects. For illustration purposes, the Kern proj-
ect is presented here with a map showing the well pat-
terns (Fig. 46.16) and a figure showing the injection and
Cycle 1 Cycle 2 Cycle 3 Cycle 4
Number of wells 24 18 11 4
Average cycle length,
months 14 18 15.3 14.5
Average oil recovery
per well, STB 28,900 30,900 24,650 29,225
Average quality of
steam injected, % 71.4 69.3 75.1 78.5
Average volume of
steam injected, bbl 9,590 8,130 10,190 11,760
Ratio of oil recovered
to steam injected,
STBlbbl 3 3.8 2.4 2.5
Depth, R
Net thickness, ft
Dip, degrees
Porosity, %
Permeability, md
Oil gravity, OAPI
Reservoir temperature, OF
Initial pressure, psig
Saturation at start, %
Oil viscosity, cp
Upper Lobe
87F 227,000
lOOoF 94,000
200F 340
Lower Lobe
Well 20 Well 3
Steam volume,
lo3 bbl
Air volume, lo6 scf
Air/steam ratio,
producing days
Oil produced, bbl
Steam/oil ratio,
Oil/steam ratio,
Peak oil production
test, BID
Cycle Cycle Cycle Cycle Cycle Cycle
2 3 4 5 6 7
13.2 16.2 15.7 10.4 8.2 9.2
0 1.5 0 3.7 cl 3.6
0 91 0 355 0 394
161 161 90 90 97 97
2,449 4,701 270 503 2,375 4,203
5.4 3.4 50 21 3.5 2.2
0.19 0.29 0.02 0.05 0.29 0.45
51 81 24 38 60 141
Depth, ft
Thickness, ft
Dip, degrees
Porosity, %
Permeability, md
Oil gravity, OAPI
Reservoir temperature, OF
Reservoir pressure at start, psig
Oil viscosity, cp
Oil saturation at start, %
500 to 1,300
30 to 90
28 to 33
1,000 to 5,000
12.0 to 16.5
35 to 52
production history of the four-pattern pilot (Fig. 46.17).
In this project, the cumulative SOR was 3.8 bbl/bbl and
the production rate reached 100 B/D of oil per pattern.
Core hole data before and after the steamflood showed
an oil recovery of 72 % and also a very high area1 sweep
Brea, CA (Shell) 25-Steam Distillation Drive, Deep
Reservoir, Steeply Dipping. A steam distillation drive
was initiated in 1964 in the Brea field, which is located
about 25 miles east of Los Angeles. This project is in-
teresting because the oil is relatively light with low vis-
cosity, and the reservoir is steeply dipping at a great depth.
The reservoir properties are summarized in Table 46.20.
The dipping reservoir is seen clearly in Fig. 46.18. The
injectors are located updip, as shown in Fig. 46.19. Be-
cause of the depth, insulated tubing was used for the in-
jectors. This figure also shows the area of temperature
response and production response. The injection and pro-
duction rates are given in Table 46.20. As of Dec. 1971,
the steam rate was 1,010 B/D water and the oil rate was
230 B/D, giving an estimated SOR of 4.4 bbl/bbl.
Smackover, AR (Phillips) 82T83-Reservoir With Gas
Cap. The Smackover field is located in Ouachita County,
AR. The steamflood pilot, conducted in the Nacatoch
sand, is worth mentioning because the reservoir has a gas
cap thicker than the oil sand itself. This gas cap can be
seen readily in the log and coregraph of Sidum Well W-35
(Fig. 46.2 1). The reservoir properties are given in Table
Fig. 46.15-Kern River field, CA.
. 312
. l
I .-
Fig. 46.16-Kern steam displacement project, Kern River field.
Depth, ft 4,600 to 5,000
Gross stratigraphic thickness, ft 300 to 800 -
Ratio of net to gross sand, % 63 -
Dip, degrees 66
Porosity, O/O 22 mm
Permeabilitv. md 77 =
Oil gravity, API
Reservoir temperature, OF
Reservoir pressure at start, psi
Oil viscosity at 17YF, cp 6
Saturahon at start, %
Fig. 46.17-Injection and production history, Kern hot water and
steam displacement project (four patterns) Kern
18 River field.
F- 13
F- 12 F- 18 M- 15
Fig. 46.18-Cross section through the lower B sands, Brea
TOP f - 14 SPNO
- - - - - - - - -
- - - _
- - T- - - -
Fig. 46.19-Well locations and area of temperature, tritium, and
production responses, Brea field, CA.
l o 63 64 65 66 67 68 69 70 71
Fig. 46.20-Injection and production history, Brea steam distil-
lation pilot.
I-I 7nnn- -
Fig. 46.21-Log and coregraph, Sidum Well W-35, Smackover
field, AR.
Fig. 46.22 is a map of the lo-acre five-spot pilot, which
was later expanded to a 22-acre nine-spot pattern by add-
ing four more producers. As shown in Fig. 46.23, steam
injection started in Nov. 1964 and stopped in Oct. 1965.
The oil production continued long after steam injection
stopped. As of Aug. 1970, the additional oil produced by
steamflood was 207,000 bbl. With total steam injection
of 860,000 bbl, the cumulative SOR was 4.14 bbl/bbl.
The temperature log in Fig. 46.24 shows that steam
goes to the gas cap. It can be concluded that the increase
in oil production was not caused by frontal displacement.
Rather, the oil zone temperature increased because of con-
duction and convection from the gas cap, thus reducing
the oil viscosity and increasing the oil production.
Depth, ft
Thickness, ft
Dip, degrees
Porosity, %
Permeability, md
Oil gravity, OAPI
Reservoir temperature, -F
Reservoir pressure at start, psia
Oi l viscosity, cp
Saturation at start, %
0 to 5
SCALE I- = 400
Fig. 46.22-Sidum steam injection pilot, Smackover field.
h 300-
/ - /
1964 1965 1966 1967
p7- - -
; loot- )-
1 &VA 4
d 01
d 1966 1969 1970 1971
Fig. 46.23-injection and production history, Sidum steam in-
jection pilot, Smackover field.
125 150 175 200 225 250 275
Fig. 46.24-Temperature log, Sidum Well W-42, Smackover
Slocum, TX (SheU)89~90-Reservoir With Water Sand.
The Slocum field is located in southern Anderson County
in northeast Texas. The steamflood project interests us
since it is conducted in an oil reservoir underlain by a
water sand, as shown in the type log (Fig. 46.25). The
reservoir properties are given in Table 46.22.
The f i el d was discovered in 1955. Only about 1% of
OOIP was produced by primary operation. A small steam-
flood pilot using a N-acre normal five-spot pattern showed
encouraging results. A full-scale seven-pattern project was
initiated in 1966-67, with 5.65acre, 13-s@ patterns (Fig.
46. 26) .
Both injectors and producers were completed a few feet
into the water sand. Steam moves horizontally through
the water layer, rises vertically into the oil layer, and dis-
places oil that had been heated and mobilized. The oil then
falls down and subsequently is swept toward the
producers. The injection and production history is shown
in Fig. 46.27.
Street Ranch, TX (Conoco) I51 -Extremely Viscous
Tar, Fracture-Assisted Steamflood. The Street Ranch
pilot was conducted in the San Miguel-4 tar sand reser-
voir located in Maverick County, TX. This pilot proved
the technical feasibility of the fracture-assisted steamflood
technology (FAST) in recovering extremely viscous heavy
oils and tars. The reservoir properties are given in Table
46.23. The pilot used a 5-acre inverted five-spot pattern.
The four producers were fractured horizontally with cold
water, steam stimulated, perforated, and resteamed. The
injector then was fractured horizontally to establish com-
munication with the producers. The pilot consisted of three
Depth, ft
Thickness, ft
Dip, degrees
Porosity, O/o
Permeability, md
Oil gravity, OAPl
Reservoir temperature, F
Initial reservoir pressure, psig
Oil viscosity, cp
Oil saturation at start, %
0 to 5
> 2,000
18 to 19
1,000 to 3,000
3 to 7
Fig. 46.25-Type log, Slocum field, TX
phases: (1) fracture preheat, (2) matrix steam injection,
and (3) heat scavenging with water injection. The tar pro-
duction and steam/tar ratio during the 3 1 -month history
are shown in Figs. 46.28 and 46.29, respectively. The
average tar production rate was 185 B/D and the cumula-
tive steam/tar ratio was 10.9 bblibbl. Postpilot core holes
showed residual tar saturations as low as 8% and an aver-
age recovery efficiency of 66 % .
Lacq Suphieur, France (Elf Aquitaine) Is2 -Carb-
onate Reservoir. The Lacq Sup&ieur field is on the north
side of the Pyrenees Mts. in southwest France. The steam-
flood pilot is unique because it was conducted in a carbo-
nated, dolomitized, and highly fractured reservoir. The
reservoir properties are given in Table 46.24.
The pilot was located in the central part of the frac-
tured limestone zone, near the top of an anticline. The
pattern area is 35 acres, defined by six old producers, as
shown in Fig. 46.30. The injector was the only one drilled
for the pilot. Steam injection started in Oct. 1977. Oil
production started to increase, only 3 months after steam
injection began. The production history is shown in Fig.
m ElGtlT PATTERN EXPINSlG IIP,PI 46.3 1. By June 1980, incremental oil production amount-
ed to 176,000 bbl with a cumulative steam injection of
926,000 bbl. The cumulative SOR is 5.26 bblibbl. This
Fig. 46.26-Slocum thermal recovery Project.
Fig. 46.27-Injection and production history, Slocum thermal
recovery project.
Depth, ft
Thickness, ft
Dip, degrees
Porosity, %
Permeability, md
Tar gravity, API
Reservoir temperature, OF
Tar viscosity at 95OF. cp (est.)
Tar kinematic viscosity, cSt
Pour pomt. F
Total sulfur, wt%
Initial boiling pomt, OF
Tar saturation at start, %
26.5 and 27.5
250 to 1,000
170 to 180
9.5 to 11.0
54.7 and 38.9
Fig. 46.28-Tar production history, Street Ranch pilot, TX.
Fig. 46.29-Steam/tar ratios, Street Ranch pilot.
\ /I
\ /
0 loo0 ft.
v La4
Fig. 46.30-Pilot area, Lacq Sup&ieur field, France
Depth, ft 1,970 to 2,300
Thickness, ft 400
Oil gravity, OAPI 21.5
Reservoir temperature, OF 140
Reservoir pressure, psi 670
Oil viscosity at 1 40F, cp 17.5
Matrix Blocks Fissure Network
Porosity, % 12 0.5
Permeability, md 1 5,000 to 10,000
Water saturation at start. % 60 100
pilot showed that a strongly fissured reservoir can be ex-
ploited efficiently by the steamflood process, as if it were
a homogeneous reservoir. The dissociation of the car-
bonate rocks by steam apparently produced no unfavora-
ble effects. Rather, the CO2 evolved might have some
positive effect on the process efficiency.
Fireflood Projects
Suplacu de Barcau, Romania (IFP-ICPP) -Largest
Fireflood. The Suplacu de Barcau field lies in north-
western Romania. This is reportedly the largest fireflood
project in the world, producing nearly 6.563 B/D of
15.9API oil. The reservoir properties are given in Ta-
ble 46.25.
The project started with a pilot in 1964 using a 1.24-acre
inverted five-spot pattern that was later expanded into a
4.94-acre inverted nine-spot pattern. This was followed
by a semicommercial operation in the period 1967-71 with
eight 9.88-acre inverted nine-spot patterns. This opera-
tion further expanded into full commercial operation, first
retaining the nine-spot patterns with the same spacing, and
later changing to linedrive operation. The original pilot
and later expansions are shown in Fig. 46.32. Injection
wells numbered 38 in 1979 with 20 using alternate air
and water injection and the balance using straight air. The
production history is given in Fig. 46.33. The WAR was
0.089 to 0.178 bb1/103 scf. As of 1979, the air injection
rate was 63,600~ lo3 scf/D. With an oil rate of 6,563
B/D. the AOR was estimated to be 9.7 X 10 scf/bbl.
West Heidelberg, MS (G~lf)~*~-Deepest Fire-
flood. The West Heidelberg field is located in Jasper
County in eastern Mississippi. With a depth exceeding
2 miles, it is the deepest fireflood project, or the deepest
thermal project, for that matter. The Cotton Valley for-
mation has eight sands. The fireflood was conducted in
Sand No. 5. The reservoir properties are given in Table
As shown in the structure map of Sand No. 5 (Fig.
46.34), only one injector was used, near the top of the
structure, with seven producers located downdip. The in-
jection and production history is given in Fig. 46.35. It
can be estimated from this figure that, during the period
1973-76, the average air injection rate was about
900 x lo3 scf/D whereas the average oil production rate
was about 400 B/D. This gives an AOR of only
2.25 x lo3 scfibbl.
Fig. 46.31-Production history, Lacq Superieur field.
Fig. 46.32-Suplacu de Barcau field, Romania.
Fig. 46.33-Injection and production history, Suplacu de Barcau
Depth, ft
164 to 656
Net thickness, ft
Porosity, %
Permeability, md
Oil gravity, OAPI
Reservoir temperature, OF
Oil viscosity at 64OF, cp
Oil saturation at start, %
Depth, ft
Thickness, ft
Dip, degrees
Porosity, %
Permeability, md
Oil gravity, OAPI
Reservoir temperature, OF
Oil viscosity at 221F, cp
Oil saturation at start. %
20 to 40
5to 15
Gloriana, TX (Sun)6~118-Thinnest Reservoir Pro-
duced by a Fireflood. The Gloriana field is in Wilson
County, TX. The fireflood took place in the Poth A
Sand. It is possibly the thinnest reservoir that has ever
been produced by a fireflood. The reservoir properties
are given in Table 46.27.
The field originally was developed on 40-acre spacing.
A new well, Well 2-8, was ignited in May 1969. Well
2-5, a producer, burned out and was converted to air in-
jection in May 1971. These wells, along with other wells,
Production limit l-1 1,093 Cl
Fig. 46.34--Structure map of Sand No. 5, West Heidelberg field,
are shown in the isopachous map in Fig. 46.36. The in-
jection and production histories are given in Figs. 46.37
and 46.38, respectively. Air injection stopped in Dec.
1974 when the oil production rate declined to the eco-
nomic limit.
Sloss, NE (Amoco) 77-79-Wet Combustion, Tertiary
Recovery. The Sloss field is located in Kimball County,
NE. The pilot used a wet combustion process in a previ-
ously waterflooded reservoir. Here, the pay is thin and
deep, the oil is light, the viscosity is low, and the oil satu-
Depth, ft
Thickness, lt
Dip, degrees
Porosity, O/O
Permeability, md
Oil gravity, OAPI
Reservoir temperature, OF
Oil viscosity, cp
Oil saturation at start, %
0 to 5
70 to 150
250 to 500
ration at the start of the flood was low. The reservoir prop-
erties are given in Table 46.28.
The fireflood started in 1%7 with six 8O-acre five-spots.
Additional wells were included so that it covered 960 acres
in its final stage. The pilot area is shown in Fig. 46.39.
The injection and production data in the 4YGyear period
of its operation are given in Figs. 46.40 and 46.4 1, respec-
tively. Between Feb. 1967 and July 1971, total air injected
was 13,754X lo6 scf and water injected was
10,818 x lo3 bbl, giving a WAR of 0.79 bbl/103 scf. The
total oil production was 646,776 bbl. This gives an AOR
of 21.3~ lo3 scfibbl. The area1 sweep by the greater-
than-350F zone was 50%. Combining with a vertical
sweep of 28%, the volumetric sweep was only 14%.
Asphalt Ridge, UT (DOE) 130-Extremely Viscous
Tar, Combination Reverse/Forward Combustion. The
Northwest Asphalt Ridge deposit is located in northeast
Utah, near the city of Vernal. The fireflood conducted
Fig. 46.35-Injection and production history, West Heide,oerg field,
Fig. 46.36-lsopachous map, net oil, Poth A sand, Gloriana field, TX.
in this deposit is interesting because it attempted to use
a combination of reverse and forward combustion for the
recovery of oil from tar sands. The reservoir properties
are given in Table 46.29.
The U.S. DOE conducted two fireflood tests in the
Asphalt Ridge area. The first, conducted in 1975, dem-
onstrated the feasibility of using reverse combustion to
recover oil in the tar sand. The second tested a combina-
tion of reverse combustion and forward combustion dur-
ing the period from Aug. 1977 to Feb. 1978. The location
of the test sites and well arrangements are shown in Fig.
46.42. In both tests, the line drive was on a small area
of 120~40 ft, covering only 0.11 acres. In the second
test, several echoings of reverse and forward combustion
phases were noticed in the northwest area, as seen from
the temperature variations at observation Well 203 (Fig.
46.43). The reverse combustion phase had an areal sweep
of 95 % and vertical sweep of 9 1% , giving a volumetric
sweep of 86%. The echoing forward combustion phase
had an area1 sweep of 75 % and vertical sweep of 44 % ,
giving a volumetric sweep of only 33%. The produced
oil was of better quality than the original bitumen, with
the pour point reduced from 140 to 25F and the amount
of residue lowered from 62 to 35 wt% .
Forest Hill, TX (Air Products-Greenwich) Is3 -
Oxygen-Enriched Air. The Forest Hill field is located
in Wood County, TX. The significance of the field test
lies in the use of oxygen-enriched air for the fireflood.
The reservoir properties are given in Table 46.30.
The field was on primary production in 1964. Air in-
jection started in 1976. One of the air injectors was
switched to oxygen-enriched air in 1980. The test site is
shown in Fig. 46.44. As seen in Fig. 46.45, during a
2-year period, the oxygen concentration in the injected
gas ranged from 2 1 to 90%. The test showed that essen-
tially pure oxygen can be handled and injected safely in
a typical oilfield environment. Short of any definitive com-
parison, the test only hinted that using enriched air might
produce oil faster than using air only.
Thermal Properties
Only some selected thermal properties of the rock/fluid
systems encountered in the thermal recovery projects will
be presented briefly. A more complete compilation of
tables and figures has been included in Appendix B of
Ref. 154.
Oil Viscosities
The viscosity-temperature relationships for representative
heavy-oil deposits are shown in Fig.46.46. Oil viscosi-
ties should be measured experimentally. In the absence
of experimental data, the viscosities can be estimated by
charts (Fig. 46.47 to 46.49)55-57 and equations. Is8
Beggs and Robinson Is8 suggested the following equa-
tions for estimating viscosities of live oils. Dead-oil vis-
cosity is first calculated:
&,~=lo*-l, . . . . . . . . . . . . .(74)
where pLod equals the viscosity of dead oil (gas-free oil)
at T, cp.
, . . . . . . . . . . . . . . . . . . . . . . . . . .
Y=lO, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..(76)
Z=3.0324-0.02023 yo. . . . . . . . . . . .(77)
where y0 equals the oil gravity, API, and Tis the tem-
perature, F. Live-oil viscosity is calculated next.
p=&,,dB, . . . . . . . . . . . . . . . . . . (78)
Fig. 46.37-GOR and air injection history, Gloriana field.
Fig. 46.38-011 production history, Gloriana field.
Fig. 46.39-Sloss unit, NE.
Depth, ft
Net thickness, ft
Porosity, %
Permeability, md
Oil gravity, OAPI
Reservoir temperature, ' F
Reservoir pressure, psig
Oil viscosity at 200F, cp
Oil saturation al start, O/o
0. 8
2oto 40
rno- nrl "" '
- -

1967 1%8 1969 1970 1971 1972
Fig. 46.40-Injection history, Sloss field COFCAW pilot
I - L -
1967 1968 lW9 1970 1971
Fig. 46.41-Production history, Sloss field COFCAW pilot
Depth, ft
Net thickness, fl
Porosity, 910
Permeability, md
Oil gravity, OAPI
Reservoir temperature, "F
Oil viscosity at 60F, Cp
Pour point, OF
Saturations at start, %
> 1 ,ooo,ooo
Fig. 46.42-LETC field site, Asphalt Ridge deposit, UT.
Depth, ft
Net thickness, ft
Porosity, O/o
Permeability, md
Oil gravity, OAPI
Reservoir temperature, OF
Oil viscosity at 185OF, cp
Saturations at start, %
Fig. 46.43-Maximum temperature vs. time, Welt 203, Asphalt
Ridge deposit.
Fig. 46.46-Viscosity/temperature relationships for representa-
live heavy oil deposits.
A= 10.715@, + 100) -OS~, . . . . . . . . . . . . . . .
Fig. 46.44-Forest Hill field, TX.
B=5.44(R,+150)-0.338, . . . . . . . . (80)
and R, :s the solution gas/oil ratio, scf/STB.
Relative Permeability Curves
Relative permeability data should be determined ex-
perimentally. In the absence of experimental data, the fol-
lowing equations may be used for rough estimation.
According to Brooks and Corey, 159
k, =(S,,,*)5, . . . . . . . . . . . . . . . . . . (81)
k,=(1-S,*)2(1-Sw*2), . . . , . . . . . . .(82)
s *=
SW -Siw
) . . . . . . . . . . . . . . . . . . . . .
1 -ISi,
Fig. 46.45-Injection history, Well 32, Forest Hill field.
where Si, is the irreducible water saturation, percent.
y ,oo
n 400
g 100
; 200
J 100 FORYULA: ABSOLUTE vIscosITY AT l00.F (Cpl* ------27, 29, 420
60 --- EXTQAPOLATED ho-ll)
.01 I I
! 1
60 66
Fig. 46.47-Dead oil viscosity.
Fig. 46.48-Universal temperature/viscosity chart.
Fig. 46.49-Live oil viscosity.
According to Somerton, 160 for unconsolidated sand,
Sj,=0.211+2.0x10-4T+1.1x10-6T2, _._.. (84)
where T is the temperature, F.
The effect of temperature on irreducible water satura-
tion and relative permeability of unconsolidated sands has
been studied by Poston et al. 16 Some of their results are
given in Figs. 46.50 through 46.52. The effect of tem-
perature on relative and absolute permeabilities of con-
solidated sandstones has been studied by Weinbrandt et
al. 62 Some of their results are given in Figs. 46.53
through 46.56.
PV Compressibility
The compressibility of unconsolidated, Arkosic sands was
measured by Sawabini et al. 63 Fig. 46.57 shows that the
effective PV compressibility lies in the range between
10e4 and 10m3 psi-l, about 2 to 3 orders of magnitude
higher than the normally accepted figure of IO -6 psi -I
for consolidated sandstones. In Fig. 46.57, pro is the to-
tal overburden pressure, psi, and pP is the pore pressure,
#) loo Is0 200 300
Fig. 46.50-Effect of temperature on irreducible water satura-
tion, Houston sand, and natural sand.
Thermal Conductivity
Thermal behavior of unconsolidated oil sands was studied
by Somerton et al. K+J Fig. 46.58 shows how thermal
conductivity of Kern River oil sands varies with brine
Vaporization Equilibrium
Vaporization equilibrium of an oil fraction is described
by the equilibrium vaporization constant, K, which is de-
fined as
K=x, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
50 100 IS0 200 26c 300
where y is the mol fraction in vapor phase and x is the
mol fraction in liquid phase.
Fig. 46.51-Effect of temperature on ROS, natural sand.
Poettmann and Mayland I65 in 1949 published a series
of charts on equilibrium constants of various oil fractions
with normal boiling points of 300F, 4OOF, etc., up to
1,OOOF. To illustrate how K values vary with tempera-
ture and pressure, the figure for normal boiling
point=5OOF is shown in Fig. 46.59.
More recently, ef al. 166 presented equilibrium con-
stants of oil fractions with 100F boiling ranges. For ex-
ample, Fraction 1 has the boiling range up to 3OOF,
Fraction 2 boiling between 300 and 400F, and Fraction
6 boiling above 700F. Figs. 46.60 and 46.61 show the
effects of pressure and temperature, respectively. on the
K values for these oil fractions as well as N 2, CH4, and
Chemical Kinetics
Chemical reactions taking place in an in-situ combustion
process are considered to fall into three types: (1) low-
temperature oxidation, (2) fuel deposition or coke forma-
tion, and (3) combustion. The kinetic data of these three
types of reactions reported by various authors have been
summarized in the paper by Fassihi et al. 16 and will not
be reproduced here.
0 20 40 60 80
Fig. 46.52-Water and oil relative permeability at four tempera-
tures, Houston sand, 80-cp 011.
60 loo 140 180
Fig. 46.53-Effect of temperature on irreducible water satura-
tion, sandstone cores.
60 100 140 180
Fig. 46.54-Effect of temperature on ROS, sandstone cores.
zsoo -
2000 k
z 5.
c 29
l 4i
I A1
5 i
l 2
60 100 140 180
Fig. 46.56-Effect of temperature on absolute permeability,
sandstone cores.
% to*
Fig. 46.57-Effective PV compressibility, unconsolidated Arkosic
oil sand.
oo- -
20 40 60 6 100
Fig. 46.55-Water and oil relative permeability at two tempera-
tures, Core 4, Boise sandstone.
hsv .0.?35-r.30~*sw +
a 0.28-0.30 -
0 0.31 -0.33 ----
A 0.34 -0.37 --
Fig. 46.56-Thermal conductivity of Kern River oil sands.
Fig. 46.59-Equilibrium vaporization constant, normal boiling
1000 1500 2000 2500 3000 P(rksld
1 I
5ow loo00 ISCCO zoo00
Fig. 46.60-Effect of pressure on equilibrium K values for Crude
A at 260%.
IO -
500 600 700 T(OF)
I I r
y 0.1 -
0.01 -
0 co2 -
0 cmk n/ 1 , 1 I
x0 250 3al 350 400
Fig. 46.61-Effect of temperature on equilibrium K values of
Crude 8 at 1,514.7 psia.
Specific Volume, cu ftllbm Enthalpy. Btullbm
Absolute Pressure, Temperature Saturated
400. 0
500. 0
700. 0
800. 0
900. 0
544. 58
567. 19
Liquid, Evaporate,
v tg
0.016022 3,302.4
0.016719 26.782
0.017274 8.4967
0.017740 4.4133
0.01809 2.9958
0.01839 2.2689
0.01865 1.82452
0.01889 1.52384
0.01934 1.14162
0.01975 0.90787
0.02013 0.74962
0.02050 0.63505
0.02087 0.54809
0.02123 0.47968
0.02159 0.42436
0.02232 0.34013
0.02307 0.27871
0.02387 0.23159
0.02472 0.19390
0.02565 0.16266
0.02669 0.13603
0.02790 0.11287
0.02938 0.09172
0.05078 0.00000 0.05078
Steam Properties
An abbreviated steam table68 is given in Table 46.31.
a = air requirement, 106, scf/acre-ft
A = heated area at time t, sq ft, or
quantities defined by Eqs. 9 and 79
A = quantity defined by Eq. 13
B = quantities defined by Eqs. 10 and 80
B = quantity defined by Eq. 14
C, = fuel content, lbmicu ft
C,, = heat capacity of oil, Btu/lbm-F, or
concentration of oil, lbm molicu ft
C, = heat capacity of rock, Btuilbm-F
C,, = heat capacity of steam, Btuilbm-F
C,,. = heat capacity of water, Btuilbm-F
Co2 = concentration of oxygen, Ibm molicu ft
C, = quantity defined by Eqs. 58 and 60
C2 = quantity defined by Eqs. 59 and 61
D = depth, ft
E = activation energy, Btuilbm mol
Eh = thermal (heat) efficiency, fraction
Eo2 = oxygen utilization efficiency, fraction
E, = overall oil recovery
= recovery efficiency in the unburned
region, fraction
vh =
volumetric sweep efficiency of the
burning front, fraction
f,, = steam quality at the beginning of the
pipe segment, fraction
0.0003 1,075.5
180.17 970.3
250.2 923.9
298.5 888.6
330.6 863.4
355.5 842.8
376.1 825.0
394. 0 808. 9
424.2 780.4
449.5 755.1
471.7 732.0
491.6 710.2
509.8 689.6
526.7 669.7
542.6 650.4
571.9 613.0
598. 8 576.5
624.2 540.3
648.5 503.8
672.1 466.2
695.5 426.7
719.0 384.8
744.5 337.6
770.7 285.1
801.8 218.4
906.0 0.0
f.Yfs2 = steam quality at the end of the pipe
segment, fraction
f(t) = transient heat conduction time function
for earth, dimensionless
F,., = COZ/CO ratio in produced gas
F HC = atomic H/C ratio
F, = ratio of stimulated to unstimulated
productivity indices, dimensionless
F,?, = steam/oil ratio, STBiSTB
= total produced WOR. STBiSTB
h = pay thickness, ft, or convection heat
transfer coefficient, Btu/hr-sq ft-F
h = convection heat transfer coefficient
based on insulation outside surface,
Btulhr-sq ft-F
hf = cnthalpy of liquid water at T above
32F, Btuilbm
h, = total thickness of all sands, ft
= enthalpy of oil and gas Btuilbm
H,, = enthalpy of water carried by oil based
on a STB of oil, Btu/STB oil
= enthalpy of water at reservoir tempera-
ture, Btu/lbm
H,,, = enthalpy of water at steam tempera-
ture, Btu/lbm
~ cumulative air injection, 10 scf
i, = steam injection rate, B/D
I = radiation heat transfer coefficient,
Btulhr-sq ft-F
I = radiation heat transfer coefficient based
on insulation outside surface. Btuihr-
sq ft-F
J, = unstimulated (cold) productivity Index,
k = absolute permeability, md
k = pre-exponential factor
k,,,., = thermal conductivity of the casing ma-
terial, Btu/hr-sq ft-F
k,IC.P = thermal conductivity of the cement.
k,,f = thermal conductivity of the formation,
k,,;,, = thermal conductivity of insulation ma-
terial, Btuihr-ft-F
ho =
overburden thermal conducitivity,
k,,, = relative permeability to oil, fraction
k,,,. = relative permeability to water, fraction
K = equilibrium vaporization constant
L = pipe length, ft
L,, = latent heat of steam, Btuilbm
= latent heat of vaporization at top of
interval, Btuilbm
L,.? = latent heat of vaporization at bottom of
interval, Btuilbm
= total mass of steam injected, Ibm
M = volumetric heat capacity. Btuicu ft-F
N = OOIP, bbl
N, = number of sands
= oil in place at start of project, bbl
= cumulative incremental oil production,
= static formation pressure at external
radius, psia
= pore pressure, psi
P,~= saturated vapor pressure of water at T,
P ,o
= total overburden pressure, psi
p,,, = bottomhole pressure, psia
p / = pressure at top of interval, psia
p2 = pressure at bottom of interval, psia
Ap = frictional pressure drop over interval,
= oil displacement rate, B/D
qoC. = cold oil production rate, B/D
qnl r = hot oil production rate, B/D
Qi,, = heat remaining in heated zone, Btu
Q;, = total heat injection, Btu
Qri = heat injection rate, Btuihr
Q,., = heat loss along the segment, Btuihr
Q, = heat removal rate at time t. BtuiD
rd = radius to cement/formation interface, ft
r<.; = inside radius of casing. ft
= outside radius of casing. ft
= external radius, ft
= radius of region originally heated. ft
r in
= outside radius of insulation surface, ft
= inside radius of tubing, ft
= outside radius of pipe. tt
r,,. = well radius, ft
R = gas constant
R,, = solution GOR. scf/STB
R, = total produced GOR. scf/STB
S,s = gas saturation, fraction
Sir, = irreducible oil saturation, fraction
S;,,. = irreducible water saturation. percent
S,, = oil saturation at start, fraction
S,; = initial oil saturation, fraction
S,,,; = initial water saturation, fraction
S,,.* = normalized water saturation, fraction
t = time since injection, hours
= critical time, hours
fg = dimensionless time
t; = time of injection for the current cycle,
T = average temperature of the heated
region, rr,. < r< rh , at any time t,
T,, = atmospheric temperature, F
Td = temperature at cement/formation
interface, F
T,.; = temperature at casing inside
surface, F
Tj; = initial formation temperature, F
I$ = temperature of fluid, F
T;,,j = injection temperature, F
TR = original reservoir temperature, F
T, = steam temperature, F
T,s, = formation temperature at ground
surface, F
c 0
= overall heat transfer coefficient based
on outside casing surface, Btuihr-sq
U,i = overall heat transfer coefficient based
on inside radius of pipe or tubing,
Btuihr-sq ft-F
ur, = overall heat transfer coefficient based
on outside tubing surface, Btu/D-sq
specific volume of total fluid, cu
v,,. = wind velocity, mile/hr
FR = fuel burned, bbl
V,., V, = unit solution for component conduction
problems in the r and z directions*
v,, VT = average values of V, and V; for
w = Arrhenius reaction rate
= mass rate of steam, Ibm/hr
x = mole fraction in liquid phase
X = quantity defined by Eq. 66
v = mole fraction in vapor phase
i = quantity defined by Eq. 65
(Y = thermal diffusivity, sq ft/D
= overburnden thermal diffusivity, sq
These symbols have no physical connotation They are simply mathemallcal symbols
0 = dummy variable in Eq. 27
6 = energy removed with the produced
fluids. dimensionless
8 = dip angle, degrees
P 0
= oil viscosity, cp
kLoC = cold oil viscosity, cp
p,,d = viscosity of dead oil (gas-free oil) at T,
poh = hot Oil ViSCOSity, cp
= density of oil, rock grain, and water,
lbmicu ft
4 = porosity, fraction
F,, (in m3im3)=18.744+0.004767D-0.16693h
-0.89814k-0.5915~L,, - 14.79S,
-0.0009767kh. . (68)
C,n (in kg/m 3, = - I .9222 + 0.137695h + 1.85029k
+35.723,, +0.012887kh
/* 0
-0.009930- 1.0444pL.. . . . (69)
Key Equations in SI Units
2Fcc + 1 + FHC
FCC +1
- cm
2 >
. . . . . . .
h=7.165v,.0~6/rin0.4. . . . . . . (3)
p2=pI +7.816x10~r2(v,, -v,,)%
+9.806~lO-~E-Ap. . . . . . . (20)
&I I
F,,,, * =
Ahd(S,; -sju) . . . . (30)
a (in std m3/m3)= 108.356+2.75367/~+229.4773,,
+16.073k. . . . (71)
F,,, =
[(4s,- 1:;oj a
- ELb +@,(I -E,,h)ERu
. . . . . . . . . . . . . . . . . . . . . . . . . . . .
4oc =
( p, - p, , , ) . . . . . . . . . . . . ( 42)
pLoc. lnL
r 111
0.4273, - 0. 004429h
> 1
x, . . . . ( 64)
F,, (in m3/m3)= li -0.011253+0.000091170
F,, (in std m3/m3)=3820.4+12.97h+192.20k
. . . . . . . . . . . . . . . . . . . . . . . . . . . .
a is in std m3/m3,
A is in m2,
C, is in kg/m3,
D is in m,
F,, is in std m3 /m3,
F,, is in m3/m3,
F Jo* is in m3/m3,
h is in kJ/m2 .h*K (Eqs. 2 through 4),
h is in m,
I,~ is in std m3,
k is in pm2,
N, is in m3,
ps are in kPa,
qoc is in m/d,
Q, is in m3/h,
rjn IS m m,
rli IS m m,
t is in h.
+0.00003467~ t-0.5120@,
P 0
v, is in m /kg,
v, is in kg/h,
ws is in kglh,
p. is in Pa.s, and
. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
pot is in Pa.s.
Farouq Ah. SM.: Steam Injection, Sewn&lv uncl Terrrar)
Oil Rrrw~en Processes. Interstate Oil Compact Commission, Ok-
lahoma City (Sept. 1974) Chap. 4.
McNeil. J.S. and Moss. J.T.: Oil Recovery by In-Situ Com-
bustion. Per. Ejtx. (July 19.58) B-29-B-42.
Berry. V.J. Jr. and Parrish. D.R.: A Theoretical Analysis of
Heat Flow in Reverse Combustion, Twns., AIME (1960) 219.
Dietz, D.N. and Weijdema. J.: Reverse Combustion Seldom
Feasible. Producer.s Monrhi~ (May 1968) IO.
Smith. F.W. and Perkins, T.K.: Experimental and Numerical
Simulation Studies of the Wet Combustion Recovery Process.
J . Cdn. Pa. Tuch. (July-Sept. 1973) 44-54.
Stovall. S.L.: Recovery of Oil from Depleted Sands by Means
of Dry Steam, Oil Week/y (Aug. 13, 1934) 17-24.
Grant. B.R. and Szaaz. S.E.: Development of Underground Heat
Wave for Oil Recovery. Tram., AIME (1954) 201. 108-18.
Kuhn. C.S. and Koch, R.L.. In-Situ Combustion-Newest
Method of Increasing Oil Recovery, Oil and Grrs J . (Aug. IO,
1953) 52, 92-96, 113. 114.
Trantham. J.C. and Marx. J.W.: Bellamy Field Tests: Oil from
Tar by Counterflow Underground Burning. J . Per. Tech. (Jan.
1966) 109-15: Trm~.. AIME. 237.
Giusti. L.E.: CSV Makes Steam Soak Work in Venezuela Field,
Oil md Gas 1. (Nov. 4, 1974) 88-93.
Stokes. D.D. and Doscher. T.M.: Shell Makes a Success of
Steam Flood at Yorba Ltnda. Or/ cind Gas J . (Sept. 2. 1974)
Dietz, D.N.: Wet Underground Combustion, State-of-the-Art,
J . Pcjr. Twh. (May 1970) 605-17: Tram., AIME. 249.
Steam Dominates Enhanced Oil Recovery. Oil crud Gw J. (April
5. 1982) 139-59.
Experts Assess Status and Outlook for Thermal, Chemical. and
CO1 Miscible Flooding Processes. J. PPI. Tech. (July 1983)
Johnson. L.A. Jr. rr ul.: An Echoing In-Situ Combustion Oil
Recovery Pmject in the Utah Tar Sand. J . PH. T&I (Feb. 1980)
Enhmcw/ Oil Recoined Porenrid in rhr Uniwd Srcues, Report by
Lewin and Assocs. for Office of Technology Assessment (Jan.
1978) 40-41.
Wdlman. B.T. 61 dl.: Laboratory Studies of Oil Recovery by
Steam Injectton. Truns., AIME (1961) 222. 681-90.
McAdams, W.H.: Hmt Transmission. third edttion. McGraw-Hill
Book Co. Inc., New York City (1954) 261.
Ramey. H.J. Jr.: Wellbore Heat Transmtssion. J. Pet. Tech.
(April 1962) 427-35.
Saiter. A.: Heat Losses During Flow of Steam Down a Well-
bore. J . Pe/. Tech. (July IY65) 845-5 I.
Willhite. G.P.: Overall-All Heat Transfer Coefficients in Steam
and Hot Water Injection Wells. J . Per. Trch. (May 1967) 607-15.
Earlougher. R.C. Jr.: Some Practical Considerations in the De-
stgn of Steam Injection Wells, J . Pet. Tech. (Jan. 1969) 79-86.
Beggs. H.D. and Brill. T.P.: A Study of Two-Phase Flow in
Inclined Pipes. J . Per. Tech. (May 1973) 607-14.
Farouq Ali, S.M.: A Comprehensive Wellbore Steam/Water
Flow Model or Steam Injection and Geothermal Applications.
.Soc~. Pet. En+ J . (Oct. 1981) 527-34.
Marx. J.W. and Langenheim, R N.: Reservoir Heating by Hot
Flutd Injection. Truns.. AIME (1959) 216. 312-15
Evans. J.G.: Heat Loss During the Injection of Steam lnto a
5-Spot. paper presented as term project in PNG 515. Pennsyl-
vania State U.. University Park (June 1960).
Ramey. H.J. Jr.: Discussion of Reservoir Heating of Hot Fluid
Injectton. Trms., AIME (1959) 216. 363-65.
Mandl. G. and Volek. C.W.: Heat and Masg Transport in Steam
Drive Processes. .Soc Per. Ore. J . (March 1969) S9-79.
Myhill, N.A. and Stegemeier, G:i.: Steam Drive Correlattons
and Prediction. J . Per. Ted. (Feb. 1978) 173-82.
Neurnan. C.H.: A Gravity Override Model ofSteamdrive. J .
Per. Twh. (Jan. 1985) 163-69.
Doscher. T.M. and Ghassemi. F.: The Influence ofOil Viscowy
and Thickness on the Steam Drive. J . Pd. Twh. (Feb. 1983)
Vogel, J.V: Stmplrtied Heat Calculations for Steamlloods, J .
Per. Tech. (July 1984) 1127-36.
Boberg. T.C. and Lantz, R.B.: Calculation ot the Production
Rate ofaThermally Stimulated Weli, J . Pet. Tech. (Dec. 1966)
Boberg, T.C. and West, R.C.C.: Correlation of Steam Stimu-
lation Performance. J . Put. Tech. (Nov. 1972) 1367-68.
Coats, K.H. et rrl.: Three-Dimensional Simulation of Steamflood-
ing, Sot. Pet. Enc. J . (Dec. 1974) 573-92.
Coats, K.H.: Simulationof Steamflooding with Diatillatton and
Solution Gas, Sot. Per. Enx. J . (Oct. 1976) 235-47.
Coats. K.H.: A Highly Implicit Steamflood Model. Sot. PC!.
Enn. J . (Oct. 1978) 369-83.
Crookston, R.B., Culham, W.E.. and Chen, W.H.: Numerical
Simulatton Model for Thermal Recovery Processes, Sot,. Per.
Eng. J.(Feb. 1979) 37-58.
Youngren, G.K.: Development and Applications of an In-Situ
Combustion Reservoir Simulator. Sot. Pet. Eng. J . (Feb. 1980)
39-5 I
Coats. K.H.: In-Situ Combustion Model, Sot. Per. En,e. J .
(Dec. 1980) 533-54.
Grabowski, J.W. era/.: A Fully Implicit General Purpose Finite-
Difference Thermal Model for In-Situ Combustion and Steam,
paper SPE 8396 presented at the 1979 SPE Annual Technical Con-
ference and Exhibition, Las Vegas, Sept. 23-26.
Alexander, I.D.. Martin, W.L., and Dew. J.N.: Factors Affect-
ing Fuel Availability and Composition During In-Situ Combua-
tion. J. Per. Tech. (Oct. 1962) 1154-64
Showalter, W.E.: Combustion-Drive Tests. Sot. Pet. Eng J .
(March 1963) 53-58.
Parrish. D.R. and Craig. F.F. Jr.: Laboratory Study of a Com-
bination of Forward Combustion and Waterflooding-The COF-
CAW Process, J. Per. Tech. (June 1969) 753-61.
Burger, J.G. and Sahuquet. B.C.: Laboratory Research on Wet
Combustion. J . Pet. Tech. (Oct. 1973) 1137-46.
Garon, A.M. and Wygal. R.J. Jr.: A Laboratory Investigation
of Firewater Flooding, Sot. Pet. En,q. J . (Dec. 1974) 537-44.
Moss. J.T. and Cady. G.V.: Laboratory Investigation of the Oxy-
gen Combustion Process for Heavy 011 Recovery, paper SPE
10706 presented at the 1982 SPE California Regional Meeting,
San Francisco, March 24-26.
Pursley, S.A.: Experimental Simulation of Thermal Recovery
Processes, Pmc,. , Heavy Oil Sympostum, Maracaibo, Venezuela
Huygen, H.H.A.: Laboratory Steamfoods in Half of a Five-
Spot. paper SPE 6171 presented at the 1976 Annual Technical
Conference and Exhibition. New Orleans, Oct. 3-6.
Ehrlich, R.: Laboratory Investigation of Steam Displacement in
the Wabasca Grand Rapids A Sand, Oil Smds of Canada-
Venezueku 1977, CIM (1978) Special Vol.
Prats. M.: Peace River Steam Drive Scaled Model Experiments,
Oii Sands of Cmada-Ven:uc41 1977. CIM (1978) Special Vol.
Doscher, T.M. and Huane. W.: Steam-Drive Performance
Judged Quickly from Use of Physical Models. Oil and Gas J .
(Oct. 22, 1979) 52-57.
Singhal. A.K.: Physical Model Study of Inverted Seven-Spot
Steamfloods in a Pool Containing Conventional Heavy Oil, J .
C&t. Per. Tech. (July-Sept. 1980) 123-34.
Binder, G.G. era/.: Scaled-Model Tests of In-Situ Combustion
in Massive Unconsolidated Sands, Proc,. , 7th World Pet. Gong..
Mexico City (1967).
Pujol. L. and Boberg. T.C.. Scaling Accuracy of Laboratory
Steamflooding Models, paper SPE 4191 presented at the 1972
SPE California Regional Meeting. Bakersfield. Nov. 8-10.
Stegemeier. G.L., Laumbach. D.D., and Volek, C W.:
Representing Steam Processes with Vacuum Models, Sot. Pa.
Grg, J . (June 1980) 15 l-74.
Farouq Ali. S.M.: Current Status of Steam Injection as a Heavy
Oil Recovery Method. J . G/n. Per. Twh. (Jan.-March 1974)
Geffen. T.M.: Oil Production to Expect from Known Technol-
ogy. Oil md Gas J . (May 7. 1973) 66-76.
Lewin and Assocs. Inc. : Tha Potmricrls und Eummics of En-
howecl Oil Recowr,?, Federal Energy Administration (April 1976)
Report 876122 I, 2-6.
lyoho. A.W.: Selecting Enhanced Oil Recovery Processes.
Wdd Oil (Nov. 1978) 61-64
Chu. C.: State-of-the-Art Review of Steamtlood Field Projects,
paper SPE 11733 presented at the 1983 SPE California Regional
Meeting, Ventura, March 23-25.
62. Chu. C.: A Study of Firetlood F~cld Protect\. J. PHI. &/I.
(Feb. 1977) 171-79.
63. Chu. C. State-ot-the-Art Review of Firetlood Ftcld Prrqccth.
J . Pr,i. Tech. (J an. 1982) 19-X. 91.
Slocutn Field. ltnpm~d Oil RWO I cvy k-w/d Reporr.s. SPE ~
Richerdwn, TX (1976) 2. No I, 11%28. F.&rrlwd Oil Rccvl.-
~!:vFic/l Rqwrrs. SPE, Rtchardson. TX (1979) 5. No. 2. 291-97.
Pollock, C.B. and Buxton, T.S.: Performance of B Forward
Steamdrive Project-Nugget Reservoir, Winkleman Dome Field,
Wyommg, J . Prr. Tech. (Jan. 1969) 35+Kl.
Winkleman Dome Field. frnpwwd Oil Rec~myv Fidd Rcp~m.
SPE. Richardson, TX (1975) 1, No. I, 155-62. E!&~ric(.d Oii
R~c~nw~ Fir/d Rqxvrc, SPE. Richardson. TX (1980) 6, No. I.
Poettmann. F.H. In-Situ Combustion: A Current Appraisal.
W&c/ Oil. Part I (April 1964). 124-28; Pan 2 (May 1963) 95-98.
Blrvina, T.R.. Ascltinc. R.J.. and Kirk, R.S.: Analyv\ ol a Steanl
Drive Protect, Inglcwood Field. California, J . Per. 7ec,i?. (Sept.
1969) Il41-50.
Blevlnb. T.R. and Billingsley. R.H.: The Ten-Pattern Stean-
flood. Kern River Fteld. California. J . Pet. lidi. (Dec. 1975).
I SOS- 14; 7i-<r,z.\. AIME, 259.
Ogleby, K.D. (I trl.: Status of the Ten-Pattern Kern Riwr Field.
California. J. Pet. T&I. (Oct. 1982) 22.51-57.
Buraell. C.G.: Steam Displacement-Kern River Field../. Fe,.
T&r. (Oct. 1970) 1225-31.
Bur\ell. C G. and Pittman, G.M.: Performance of Steam Dis-
placement in the Kern River Field, J . Per Twh. (Aus. 1975)
Greaser, G.R. and Shore. R.A.: Steamflood Performlmcc in the
Kern River Field. paper SPE 8834 prejented al the 19X0
SPE/DOE Symposium on Enhanced Oil Recovery. Tulsa. April
McBean. W.N. Attrc 011 Recovery by Steam Displxcment.
paper SPE 4 170 presented at the 1972 SPE Cahforma Regional
Meetmg. Bakersfteld, Nov 8-10
Rchkopl. B.L.: Me&on Attic Steam Drive. paper SPE S85S
presented at the 1976 SPE California Regional Meetmg. Long
Beach. April X-9.
Hearn. C.L.: The El Dorado Steam Drwe-A Pilot Tertiary
Recovery Test. J . Prr. Tdx (NW 1972) 1377-84
Vnllrroy. V.V. 01 trl.: Deerfield Pilot Test of 011 Rccowry by
Steam Drive. J . Per. Td?. (July 1967) 956-64.
van DtJk. C.: Steam-Drive Projkct in the Schooneheek Field.
The Netherland>, J . PH. Ttrh. (March 1968) 295-302.
Gate\. C F. and Ratney. H.J. Jr.: Fteld Rewlts of South Bel-
ridge Thermal Recovery Experiment. frwz\. , AIME (19.58) 213.
Parrish. D.R. rr [il.: A Tertiary COFCAW Pilot Test m the Sloss
Field, Nchra\ka. J . /<,I. 7d1. (June 1974) 667-75: Trtr,~.,.,
AlME. 257.
Pnrrihh, D.R.. Pollock. C.C.. and Craig. F.F. Jr.. Evaluation
oi COFCAW as a Tertiary Recovery Method. Slow Field. Nchraa-
ka. J . Per. T& (June 1974) 676-86: TUNIS., AIME, 257.
Buxton, T.S. and Polock. C.E.: The Sloss COFCAW PKIJeCt-
Further Evaluation of Performance During dnd After Air In,@-
tion: J . PC,! TwIi. (Dec. 1974) 1439-48: Tru,~r., AIME. 257
Mo$s. J.T.. White. P D.. and McNeil. J.S.: In-Situ Combw
twn Process-Rewlts of a Five-Well Field Experiment in South-
cm Oklahoma. Trw>\. , AIME (1959) 216. 55-64
Parrish. D.R. PI cl/.: Underground Combustion in the Shannon
Pool. Wyoming. J . Per. Tdl. (Feb. 1962) 197~205: Tr<rrl.\.
AIME. 225.
Smuh, R.V. (I cl/.: Recovery of Oil by Swam InJection in the
Smachover Field. Arkansas. J . PH. Tech. (Aug. 1973) X33-89.
Smachover Field, Irri/Jld Oil Rml~c~~ Fir/d Rep?\ ( 1075)
I. No. I. 135-43: E/rktrrtwt/ Oil Rcr~owr? r?r/tl Rqwrr\. SPE.
Richardwn. TX (1982) 8. No I. 685-88.
Kern River Field. Standard Oil Company of Caltfornla. l/w
/whew</ Oii Rrcrwc~r, Ficlrl Rqwwf,\ SPE. Richardson ~ TX ( I Y7S)
1. No I, 83-92: Dtlrtrwrcl U/i R~wwr\. Fidd Rqwm, SPE.
Richardson. TX (1980) 6. No I, 23-24.
Midway-Sunact FLeld. ChanQor-Western Oil and Development
Company . 1~~z~wwl Oil RECOIW~ /+/d Rqxwrv. SPE. Richard-
wn. TX I 1976) 2. No. 3. 445-54; E/~hmcw/ Oil Rec~~wr/~\ Ficltl
R~yom. SPE. Richardson. TX t 1082) 8. No. I. 72%32.
Stohcs. D.D (I (il.: Steam Drive a\ a Supplemental Recovery
Procchh in an Intermediate Viscosity Reservwr. Mount Paso Field.
Richardwn. TX ( 1975) I. No. 7. 277-86: &<r,iw</ Oil Rcui\~-
wv pic,/i/ Rcprw~~. SPE. Richardson. TX ( 1982) 8, No I 70 I -ll3
Travcr\c. E.F., Dethcn. A.D.. and Susteh. A.J.. San Ardrl-A
Cn\s Hl\torq of a Succci\ful Steam~lond. paper SPE I I717
orescnted at the IYX3 SPE Caliliwua Rerrwnul Mectme. Venturs.
il:rrch 73-25.
Hall. A. L. and Bowman. R W.: 0pcr;itwn and Performance o1
3 Slocum Thermdl Recover\) Protect. J . Prr. 7?c,/1. (April 1973)
I I?
I I3
I I4
I I5
Herrerd, A.J.: The M6 Steam Drive ProJect Design and Im-
plementation. J. Cdn. Per. Tee-h. (July-Sept. 1977) 62-83.
Herrera. A.J.: Steam-Drike Recovery Bemg Uxd on Lake
Mardcaibo Coastal Field. Oilcirid GN(. J . (July 17. lY78) 74-80.
Tia Juana Este Field. Marawn, S.A. , Glhowerl Oil Rrw~r~
Fwlcl Report.\, SPE. Richardson. TX (1978) 3, No 4. 751-62:
&hw&/ 011 Rwowrv Field Rqxw/r, SPE. Richardson. TX
(1982) 8. No. I? 75l-S3.
Showalter. W.E. and MacLean. M.A. Fireflood at Brea-Olinda
Field, Orange County. California. paper SPE 4763 presented
at the 1974 SPE Symposium on Improved Oil Recovery, Tulsa.
April 22-24.
Brea Oiinda Field, I*nprowd 011 Rrcrn~~r~ Fdd Rqx~rt.s, SPE,
Richardson. TX (1975) 1, No. I, IS- 18; Orhrrwcd Of/ RPCOW~!
Fiekl R~~pons. SPE, Richardson. TX (1981) 7, No. I, 407-08.
Gates. C.F. and Sklar, I.: Comhubtlon as a Primary Recocery
Process-Midway Sunset Field, J . Per. 7.cvh. (Aug. 1971)
9X1-86: Trur~s.. AIME. 251.
Counihan, T.M.: A Succes\lul In-Situ Combu\tton Pilot In the
Midway Sunset Field, California, paper SPE 6.525 prehented at
the 1977 SPE Califomta Regional Meettng, Baker\fteld, Aprtl
Gates. C.F.. Jung, K.D.. and Surface. R A.: In-Situ Combus-
tion in the Tulare Formation, South Belrtdge Field. Kern County.
CA, .I Per. T&. (May 1978) 798-806: Trcw\. AIME. 265.
Hewitt. C.H. and Morgan. J.T.: The Fry In-Sttu Comhu\tion
Te\t-Reservoir Characteristic\, 1. PH. TK/I. (March lY65)
337-42: Trcrn.c., AIME, 234.
Clark, G.A. e~ol.: The Fry In-Situ Combustion Test-Field Op-
erations. J . Fcr. Teds. (March 196.5) 343-47: Trcrr~s., AIME.
Clark. G.A. ertil.: The Fry In-Situ CombustionTest-Field Op-
erations and Performance, J . Prc Tc,rh. (March 1965) 348-53;
Trcrm., AIME. 234.
Earlougher. R.C. Jr.. Galloway, J.R . and Parsons, R.W.: Per-
formance of the Fry In-Situ Combustmn Project. J . PH. T~~c~/I.
(May 1970) 551-57.
Bleaklcy. W.B.: Fry Unit Fireflood Surwving Economic Pres-
sures.- Oil cm! Gas-J . (May 3. 1971) 92-97:
Howell. J.C. and Peterson. M.E.: The Frv In Situ Comhu\tton
ProJect Performance and Economic Statu;. paper SPE X.181
prebenled at the I979 SPE Annual Technical Confercncc and Ex-
hibttion. Las Vegas. Sept. 23-26.
Little, T.P.: Successti~l Fixflooding of the Bellewe Flcld. &r
Eng. /n/l. (Nov. 1975) 55-56.
Bellewe Field. Cities Service Oil Co.. li~,/~rrwe<l Od Re<o,-
cry FwkI Rcyxux, SPE. Rtchardson. TX (1975) 1, No. 3. 47 I-80:
.&h~~m~~d Oil Rewwr~ Fdd Rcym-r.t , SPE. Richardson. TX
(19X2) 8. No. 1. 713-15.
Cato. R.W. and Fmka. W.A.: Getty 011 Report\ Firetlood Pi-
lot is Successful Protect. O//rrm/ Gtr.c J . [Feb. 12. 19681 93-97.
Getty Expanda Bellevue Fire Flood. Oii cd Gu.\ J . (Jan 13.
lY75) 4.5-49.
Bleakley. W.B.: Getty\ Bellewe Flcld Still Goq and Grow-
ing. Prr. &g. httl. (Nov. 1978) 54-68.
Bellcvue Field, Getty Oil Co.. It,iprrwd Oil Rrcowr~ F-w/d
R~~p~rr.s, SPE, Richardson. TX ( 1976) 2. No. 2. 275-83; Orhrorw~l
Oil Rwcnvry F&d Repma. SPE, Richardwn. TX ( 1982) 8, No.
I. 717-19.
Hardy. W.C. CI tri.. In-Situ Combustion Performance in d Thm
Recervou Contatntnr: Hlrh-Gravitv 011. J . Prv. 7d1. (Feb. 1972)
_ -
IYY-208, Twz\. , AIME. 253.
Mace. C.: Deepest COnlhU~tlOn PrOJcct PrOcWtiin~ %KcehhfUi-
ly. Oil wid Gtrs J . (NW 17. 1975) 74-81.
West Hetdelherg Fxld. //n~wo,~vti U/I Rrcwc~rv Fw/r/ Rqwrrs.
SPE. Richardwn. TX (1975) I. No 2, 35 I -5Y: E!lilrrwr// Oil
Rrtnwr-v Fielrl Rqwrrs. SPE. Richardson. TX (19X0) 7. No. I.
I so.
Gadelle.C.P. (I 01.: .. Heavy Oil Recovery by In-Situ
Comhu&x-Two Field Caaca in Romania. J . PH. Tmh. ( Nov.
1981) 057-66.
Terwilllger. P.L. PI rrl.: Flretlood of the P, 2 Sand Kcscrvoir
in the Miga Field of Eastern Venezuela. J . Pet . Tech. ~. l an. 1975)
N&on. T.W. and McNeil. J.S.: How to Engineer an In-Situ
Comhuwon Project. O! I o, r t / Cm J . (June 5. 1961) 59. No. 3.
S&man. A.. BrIgham. W.E.. and Ramey, H.J. Jr.: In-Situ Com-
hwtion Models ior the Steam Plateau and for Fleldwde 011 Recov-
ery. DOEiETIl2056-1 I. U S. DOE (June 1981).
Gates, C.F. and Ramq, H.J. Jr.: A Method for Engmeerinp
In-Situ Combustion 011 Recoverv Proiects. J . Pc, r Td. (Feb.
1980) 285-94.
V&k. C.W. and Pryor. J.A.: Steam Distillation DTIVL. Brea
Field. Calilhrnia. J. Per. k/ r. (Aug. 1972) X99-906.
Burke. R.E.: Combu&n Proicct 1s Makmp a Profit. Oi/a/zd
Cm J . (Jan. 18. 1965) 44-46.-
Koch. R.L.: Practical Use of Combustion Drive at West New-
port Field. Per. Dz,q. (Jan. 1965) 72-81.
Bowman. C.H: A Two-Spot Comhustmn Recovery Project.
J Per . f dr . (Sept. 1965) 994-98.
Tewilliger. P.L.: Flrctlooding Shallow Tar Sands-A Case His-
tow. J . C&I . Per . Tdz. (Oct.-Dec. 1976) 41-4X.
J&son. L.A. ($1 cri.: An Echomp In-Situ Comhu\tmn Oil-
Recovcrv Proieci in a Utah Tar Sand. J . f cdt . TVI 11 (Feb. 1980)
295sioi . _
Perry, C.W., Hertlherg. R.H.. and Stosur. J.J.: The Status DI
Enhanced Oil Recovery in the United States, Pmt . , Tenth World
Pet. Congres. Bucharest (1980) 3. 257-66.
Benham. A.L. and Poettmann. F.H.. The Thermal Recovery
Process-An Analyhia of Laboratory Combustion Data. Trtr,r.\. ,
AIME (1958) 213, 83-85.
Caudle. B.H.. Hickman. B.M.. and Sllherberg. I.H.: Perform-
ance of the Skewed Four-Sp)t Injection Pattern. J. /cr. Twh.
(Nov. 1968) 1315-19.
Cady. G.V., Hoffman. S J.. and Scarborough, R.M. Silver-
dale Combination Thermal Drive Project, paper SPE 8904
presented at the 1980 SPE California Regional Meetmg. Los An-
geles. April 9m II
de Haan. M.J. and Schenk. L.: Pertormancc Analysis of a MaJor
Steam Drive Project m the Tin Juana Field. Western Venczw
l a. J . f cr . Td7. (Jan. 1969) I I l-19.
Texas Fireflood Looks Like a Winner. Oil crnd Gtry J . (Feb.
17. 1969) 52.
A~~Ju. B.I.: Conversion of Steam Injection to Waterllood. East
Coalinga Field. J . Per . Twh. ( Nov. 1974) 1227-32.
Gate\. C.F and Brewer. S.W.: Steam Iniection into the D and
E Zone, Tulare Formation, South Belridge Field. Kern County,
California. J. Put. TK~. (March 1975) 343-4X.
Gomaa, E.E.. Duerksen. J.H.. and Woo. P.T.: Designing a
Steamtlood Pilot in the Thick Monarch Sand of the Midwy-Sunset
Field, J . PH. Tech. (Dec. 1977) 1559-68.
Chu. C. and Trimble. A.E: Numerical Simulation ofSteam Dis-
placement-Field Performance Applications, J . f e/ . Tkh. (June
1975) 765-74.
Joseph, C. and Pusch. W.H.: A Field Comparison of Wet and
Dry Combustion, J. PC/ . Tech. (Sept. 1980) 1523-28
Gates. C.F. and Holmes, B.C.: Thermal Well Completion\ and
Operation. paper PDI I presented at the Seventh World Pet.
Cmg.. Mexico City (1967).
Fox. R.L.. Donaldwn. A.B., and Mulac, A J. Development
nf Technology for Dwnhole Steam Producti[m. paper SPE 9776
prehcnted at the 1981 SPEiDOE Joint Symposium cm Enhanced
Oil Recovery, Tulsa. April S-8.
I so.
Carrawaq. P.M Kloth. T.L and Bull. A I): Co-Gcneratiirn.
A New Energy Sy\tem to Generate Both Steam and Iilectricitq .
paper SPE 9907. presented ~11 the 1081 C,~lili~rniu Regional Mcel-
1;. Bahcrsticld. March 25-26.
Strange. L.K.: Ignition: Key Phase m Comhu&on Rcwvcry.
f ci . Cr f i . ( Nov. 1964) 105-09, (Dec. 1964) Y7~106.
Shore. R.A.: The Kern River SCAN Automation System Snrw
ple. Control and Alarm Network, paper SPE 3 I73 prcscntcd at
the 1972 SPE California Regional Meeting. Baker~iield. Nov
Bodcau In-Situ Combustion ProJect. iir\t annual report. DOE
Publications SAN-1 189-2. U.S. DOE (Sept. 1977).
Wagner. O.R.: The Use of Tracer5 in Diagnosing lntcrwcll
Reservoir Hetero~eneitics-FIleId Rcwlt\. J . f cv. Tech ( Nov.
1977) 1410-16. -
Yoelin. S.D.: The TM Sand System Stimulation Pqect. J .
Per. Tdz. (Aug 1971) 987-94.
Meldau. R.F., Shipley. R.G.. and Coats. K H : Cyclic
Gas/Steam Stimulation of Heavy-011 Wells. J . Pd. Twk. (Oct.
1981) 1990-98.
B&on, M.W. or (I/.: The Street Ranch Pilot Test ot Frnctuw
Assisted Steamflood Technology. J . PC, / . Tdz. (March 19X3)
Sahuquet. B.C. and Ferrier. J.J.: Steam-Drive Pilot in u Frac-
turcd Carbonate Reservoir: Lacq Superieur Field. J . f vi . T~I .
(April 19X2) X73-80.
Hxjirdos. L.J.. Howard, J.V.. and Roberts. G.W.: Enhanced
Oil Recovery Through Oxygen-Enriched In-Situ Comhu\tion: Test
Results from the Forest Hill Field in Texas. J . PCI . Tdf . (June
lY83) 1061-70
Prata. M,: TIwrw/ R~w\w:~. Monograph Scrlch, SPE. Richard-
son. TX (1983) 201-38.
Beal. C.: The Viscosity of Air. Water. Natural Gas. Crude Oil
and Its Associated Gases at Oil Field Tempxzturcs and Pressures.
Twm. , AIME (1946) 165. 103.
Prfml~ur?! Pmdudon HrznrlhooX. T C. Frick (ed ) Vol. Il.
McGraw-Hill Book Co. Inc.. New York City (1962) 19-39.
Chew, J. and Connally. CA. Jr: A Vi\co\ity Correlation lix
Gas-Saturated Crude Oils. Trcins., AIME (1959) 216, 23-25.
Bcggs. H.D. and Robinson. J.R.: Estimating the Viwosity of
Crude 011 Systems. J . Per . Tech. (Sept. 1975) 1140-41.
Brooks. R.H. and Corey. A.T.: Propertics of Porous Media
Affectinfi Fluid Flow, f r w. , ASCE. lrrlratmn and Drainage\
Div. (19k6) 92. No. 1R2.
Somertan, W.H. and Udell, K.S.: Thermal and High Ternpera-
ture PropertIes of Rock-Fluid Systems. paper prerented at the
1981 OASTRA Workshop on Computer Modeling. Edmonton.
Alta.. Jan. 28-30.
Poston. S.W. CI cl/.: The Effect of Tcmperaturc on lrreduclble
Water Saturation and Relative Permeahdlty of Unconwhdatcd
Sands. .Soc Pet . Eq. J . (June 1970) 171-80.
Weinbrandt, R.M.. Ramey, H.J. Jr., and Cabs& F.J.: The Et-
tcct of Temperature on Relative and Absolute Permeahiluy of Sand-
stones. Sot . f t v. Et ?&J . (Oct. 1975) 376-X4.
Sawahini. C.T., Chilingar. G.V., and Allen. D.R.: Comprca-
slhility of Unconsolidated. Arkosic Oil Sand\. Sot. Pcl. E!I,~.
J . (April 1974) 132-38.
Somerton, W.H., Keese, J.A., and Chu. S.L.: Thermal Behavior
of Unconsolidated Oil Sands. Sot . / <, I . GI , ~. I . (Oct. 1974)
Poettmann. F.H. and Mayland. B.J.: Equilibrium Constants li)r
High Boiling Hydrocarbon Fractions ot Varymg Characteri7dtlon
Factors. Prr. Refiner (July 1949) 101-12.
Lee. S.T. et (il.: Experim>ntal and Theoretical Studies on the
Fluid Properties Required for Simulation of Thermal Procewx
Sot . Per : Or g. J . (bet. 1981) 535-50.
Fassihl. M.R., Brigham. W.E , and Ramcy. H.J. Jr.: The Reac-
tion Kinetics of In-Situ Combustion. Sot,. Per . , ! ?I ~. J . (Aug.
1984) 408- 16.
1967 ASTM Steam Tables. ASME, New York City (1967).
General References
Blblmgraphy of Thermal Methods of Oil Rccoverg, J . C&I . Per .
T&. (April-June 1975) 55-65.
Crawford. P.B.: In-Situ Combustion. Se~omltrr~ trd rertilrr~
Oii R<~ow~~ Pn~w.~.w.~. Interstate Oil Compact Cummlsaion.
Oklahoma City (1974) Chap 5.
Farouq Ah. SM.: Steam tnjecrion. Set~orlllrr~~ lllld Termlr~ 011
Recovery Processes, Interstate Oil Compact Commission, OkId-
homa City (1974) Chap. 6.
Puts. M.: 77wnd Rrw~~r~, Monograph Serw SPE. Richardson.
TX (1982) 7.
7hermul Recowr~, Reprint Ser~cs, SPE, Richardson, TX
(1965. 1985) 7.
7her~w~l Reuwc~~ Techniques, Reprint Series, SPE. Richardson.
TX (1972) 10.
Chapter 47
Chemical Flooding
Larry W. Lake, U. of Texas
Chemical flooding is any isothermal EOR process whose
primary goals are to recover oil by (1) reducing the mo-
bility of the displacing agent (mobility control process),
and/or (2) lowering the oil/water interfacial tension (low-
IFT process).
These classifications do not exclude processes where
both effects are important, such as micellar/polymer (MP)
flooding, nor do they imply that other effects, such as
wettabiiity alteration, extraction, or oil swelling. are not
Mobility control processes inject a low-mobility displac-
ing agent to increase volumetric and displacement sweep
efficiency. The two main techniques are (I) polymer
flooding. whereby a small amount of polymer is added
to thicken brine and (2) foam flooding, through which low
mobilities are attained by injecting a stabilized dispersion
of gas in water. Polymer or mobility buffer drives also
are used to displace micellar and high-pH slugs. Foams
have been used or proposed as driving agents for micel-
lar. solvent, and steam slugs also. For these reasons foam
tlooding could be discussed easily in the chapters on sol-
vent and thermal EOR processes; however. foam stabili-
zation requires surfactants. whose discussion belongs in
this chapter.
Low-IFT methods rely on injecting or forming in-situ
a surface-active agent (surfactant) which lowers oil/water
IFT and, ultimately. residual oil saturation (ROS). Proc-
esscs that inject the surfactant are called MP floods
because of the tendency for surfactants to form micelles
in aqueous solutions and the inevitable need to drive the
micellar solution with polymer. High-pH or alkaline proc-
esses produce a surfactant in situ, since these processes
rely on reactions with acidic components of the crude to
generate the surfactant.
This chapter is divided into sections corresponding to
mobility control and low-IFT processes. Each section be-
gins with a general discussion of how the particular mech-
anism recovers oil, followed by material on the individual
processes, and concludes with typical oil displacement re-
sults. The chapter concludes by giving comparative
screening parameters for each chemical flooding process.
Mobility Control Processes
Effect of Low Mobility on Oil Recovery
Mobility control processes are most applicable to reser-
voirs that have substantial movable oil [oil in place (OIP)
minus ROS] since they displace oil only in excess of ROS.
The mobility ratio concept illustrates how lowering the
mobility improves oil recovery.
The mobility ratio, M, between a displacing agent and
a displaced fluid is
M=hDlh,,, ._____.____................ ._.(l)
where AD is the mobility of displacing agent(s) and A,,
is the mobility of displaced fluid(s). Practical use of Eq.
1 requires specific definitions of the quantities in the
numerator and denominator and of the conditions at which
these quantities are evaluated.
For two-phase waterflooding, common specializations
of Eq. 1 are the endpoint mobility ratio, M. the average
mobility ratio, M, and the shock mobility ratio. M*. Each
definition has been used in particular applications in the
literature: M has been used to correlate areal sweep rela-
tions, M* is the most direct indicator of viscous instability,
and M is the most widely cited value in the literature
for waterfloods. Z For the special case of a piston-like
displacement all three definitions are identical. From the
general definition of Eq. I, lowering the mobility of the
displacing fluid is equivalent to lowering any of the mo-
bility ratios. This recovers oil by increasing both volu-
metric and displacement sweep efficiency.
Volumetric sweep efficiency, EV. is the PV of a reser-
voir contacted by a displacing agent div,ided by the total
PV. Et, is composed of two parts: areal, E/r, and inva-
sion (vertical) sweep efficiency, El.
Fig. 47.1-Schematic of polymer flooding
For more &tailed discuxsions of E,, and El XC Chaps.
30. 43. and 44 and Refs. 1 through 1 I. Both quantities
dcpcnd on throughput and on several fluid. petrophyai-
cal. and geometric factors. usually expressed as dimen-
sionlcsa groups. However. E,.( and El both increase as
Mdecrea~-i.e.. as the mobility of the displacing agent,
A,,. decreases.
When the displacement is piston-like in a homogcnc-
ous. isotropic. horizontal medium and M> I (an UI+I\YIU-
hlr mobility ratio). the displacement front forms small
perturbations. called viscous fingers. which grow dur-
ing propa@on. Such tingcrs also form when M< I. but
they are damped out by the~~~\~cl&le mobility ratio. The
growing viscous fingers can contribute substantially to
poor oil rocovcry because of large-scale bypassing. Mo-
bility control agents prevent viscous fingering cithcr in
il conventional waterflood (polymer flooding) or ah a part
of an otherwise inherently unstable EOR process.
Displacement efficiency. El, (local or microscopic
sweep cfticicncy). is the volume of oil recovered in a dis-
placement divided by the oil volume just before the dis-
placement. the displacement having maximum volumetric
sweep efficiency. The classical solution by Buckley and
Leverett can be used to describe the effect of mobility low-
ering on En. Because of the dependence on several
factors there i+ no unique correspondence between M and
El,: however. lowering M or An results in improved oil
recovery through larger Eo.
Lowering M. then. results in improved oil recovery by
increahins areal. vertical. and displacement sweep effi-
cicncics. Since it is the products of these factors that dc-
tcrminc overall oil recovery. this implies that an incrcasc
in any one may not r-csutt in a larfc overall incrcaso in
oil rccokery. particularly if one of the other efficiencies
ivcre IOU. Even with the combination of the three effi-
clcncIc\. the incremental oil recovered with ;I mobility
control process must he balanced against the additional
eupensc rcquircd to in.jcct the viscous mobility control
Polymer Flooding
Polymers have been used in oil production in three modes.
I. They have been used as near-well treatments to in-
prove the performance of waler injectors or high-watercut
producers by blocking off high-conductivity Loncs.
2. Polymers also are used as agents that may be cross-
linked in situ to plug high-conductivity zones at depth in
the reservoir. These proccsscs require that polymer be
in.jected with an inorganic metal cation. which will cross-
link subsequently injected polymer molccutes with ones
already bound to solid surfaces.
3. The other mode is use as agents to lower M or Xn.
The first mode is not truly a chemical flooding proc-
ess. since the actual oil-displacing agent is not the poly-
mer. The overwhelming majority of polymer EOR
projects have been in the third mode, which is the one
emphasized here.
Fig. 47. I is a schematic of a typical polymer flood in-
jection sequence: a preflush, usually consisting of a low-
salinity brine: an oil bank; the polymer solution itself: a
freshwater buffer to protect the polymer solution from
backside dilution; and, finally. chase or drive water. Many
times the freshwater buffer contains polymer in decreas-
ing amounts (a grading or taper) to lessen the effects ot
unfavorable mobility ratio between the chase water and
the polymer solution. Because ofthe driving nature of the
process. polymer tloods always arc performed through
separate sets of in.jection and production wells.
M is lowered in a polymer tlood by in.jecting water that
contains a high-molecular-weight. water-soluble polymer.
Since the water is usually a dilution of an oillicld brine.
interactions with salinity are important. particularly for
certain classes of polymers. Salinity in this chapter is the
total dissolved solids (TDS) content of the aqueous phase.
Typical values are shown in Fig. 47.2. Virtually all
chemical flooding propertics dcpcnd on the concentrations
ofspccific ions rather than salinity only. It is partcularly
important to monitor the aqueous phahcs total divalcnt
@ Oo
Fig. 47.2-Salinities from representallve oilfield brines
cation content (hardness) separately. since the latter are
usually more crltical to chemical llood properties than the
same TDS concentration. Fig. 47.2 also shows typical
brine hardncsscs.
Bccauxc of the high molecular weight ( I to 3 million
is typical) only a small amount (typically about 500 ppm)
of polJ,mer will bring about a substantial increase in water
viscosity. Furthcrmorc. several types ofpolymcrs lower
mobility by lowcring water relative pcrmcability (pcrme-
ability reduction effect) in addition to incrcaxing the water
viscosity. How polymer lowers mobility. and the inter-
actions with salinity may be qualitatively illustrated with
some di5cusslon of polymer chemistry.
Polymer Types. Several polymers have been considered
for polymer tlooding: xanthan gum, hydrolyzed poly-
acrylamide (HPAM). copolymers of acrylic acid and
acrylamide. copolymers ofacrylamide and 2-acrylamide-2
Imethyl propane sulfonatc (AM/AMPS). hydrouyethyl-
cellulose (HEC). carboxymethylhydroxycthyIccllulose
(CMHEC). polacrylamide (PAM). polyacrylic acid, flu-
can. dextran polycthylencoxide (PEO). and polyvinyl al-
cohol (PA). Only the first three have actually hccn field
tested: however. the variety of entries in this partial list-
ing emphasizes that there arc many potentially suitable
chcmlcals. some of which may prove more cfkctivc than
thoxc currently LI W~. Ncvcrthclcsj. virtually all 01 rhe
commercially attractive polymers fall into two gcncric
classes: polyacrylamidcs and poly\accharide\ (hiopolyme1
or xanthan gum ). The remainder of this discussion deals
with these cxclu\ively. Representative molecular struc-
turn\ are given in Fig. 47.3. .
I 2 I
Fig. 47.3A-Molecular structure for partially hydrolyzed poly-
Fig. 47.3B-Molecular structure for polysaccharlde jbiopolymer).
Polyacrylamides. PAMs are polymers whose mono-
meric unit is the acrylamide molecule. As used in poly-
mer flooding, PAMs have undergone partial hydrolysis.
which causes anionic (negatively charged) carboxyl
groups (-COO-) to be scattered along the chain. The
polymers are called partially hydrolyzed polyacrylamidcs
(HPAMs) for this reason. Typical degrees of hydrolysis
arc 30% or more of the acrylamide monomers: hence.
the HPAM molecule is quite negatively charged. which
accounts for many of its physical properties. The
viscosity-increasing feature of HPAM lies in its large
molecular weight. which is accentuated by the anionic
repulsion between polymer molecules and also between
segments on the same molecule. The repulsion causes the
molecule in solution to elongate or uncoil and snag on
others similarly elongated, an effect that accentuates the
mobility reduction at higher concentrations.
If the brine salinity and/or hardness arc high. however,
this repulsion is decreased greatly through ionic shield-
ing as the freely rotating carbon/carbon bonds allow the
molecule to ball up or coil. (Fig. 47.3 shows the molccu-
lar structure of hydroxyethylcellulose.) This causes a cor-
responding decrease in the effectiveness of the polymer,
since the snagging effect is rcduccd greatly. Virtually all
HPAM properties show a large sensitivity to salinity and
hardness. This is an obstacle to usiiyg,HPAM in many
reservoirs. On the other hand, HPAM 15 inexpensive. rela-
tively resistant to bacterial attack. and exhibits permea-
bility reduction.
Polysaccharides. A second major class of polyniers arc
the polysaccharidcs. which are formed from the polymers-
Fig. 47.4-Polymer solution viscosity vs. shear rate and polymer
zation of saccharide molecules (Fig. 47.3). Polysaccha-
rides or biopolymers are formed from a bacterial
fermentation process. This process leaves substantial
debris in the polymer product that must be removed be-
fore the polymer is injected. I6 The polymer is also sus-
ceptible to bacterial attack after it has been introduced into
the reservoir. These disadvantages are offset by the in-
sensitivity of polysaccharide properties to brine salinity
and hardness. The origin of this insensitivity is seen in
Fig. 47.3, which shows the polysaccharide molecule to
be relatively nonionic and, therefore, free of the ionic
shielding effects of HPAM. Polysaccharides are more
branched than are HPAMs and the oxygen-ringed car-
bon bond does not rotate fully; hence, the molecule in-
creases brine viscosity by snagging and by adding a more
rigid structure to the solution. Polysaccharides do not ex-
hibit permeability reduction, however.
At the present time HPAM is less expensive per unit
amount than polysaccharides; however, when compared
on a unit amount of mobility reduction, particularly at high
salinities, the costs are close enough so that the preferred
polymer for a given application is site-specific. Histori-
cally, HPAMs have been used in about 95% of the report-
ed field polymer floods. I7 Both classes of polymers tend
to degrade chemically at elevated temperatures.
Polymer Properties. Because of the complexity of the
subject, a comprehensive treatment of polymer proper-
ties is not possible. However. qualitative trends, a few
quantitative relations and representative data are presented
on these properties: viscosity relations, non-Newtonian
effects, retention, permeability reduction, and chemical,
biological, or mechanical degradation.
Viscosity Relations and Non-Newtonian Effects. Fig.
47.4 shows polymer solution viscosity. p,,, vs. shear rate
measured in a laboratory viscometer at fixed salinity. I8
At low shear rates, p,, is independent of shear rate.
P(~,,=~,,), and the solution is a Newtonian fluid. At
higher i. p,, decreases, approaching a limiting
(p,, =p, ) value not much greater than the water vis-
cosity, p,,., at some critical high shear rate (not shown
on Fig. 47.4). A fluid whose viscosity decreases with in-
creasing ? is shear thinning. The shear thinning behavior
of the polymer solution is caused by the uncoiling and
Fig. 47.5-Polymer solution viscosity vs. shear rate at various
brine salinities.
unsnagging of the polymer chains when they are placed
in a shear flow. Below the critical shear rate the curve
is reversible.
Fig. 47.5 shows a viscosity/shear-rate plot at fixed poly-
mer concentration with variable NaCl concentration for
an AMPS polymer. I9 Note the profound sensitivity of the
viscosity to salinity: as a rule of thumb the polymer solu-
tion viscosity decreases a factor of 10 for every factor-
of-10 increase in NaCl concentration. The viscosity of
HPAM polymers and HPAM derivatives is even more
sensitive to hardness, but viscosities of polysaccharide so-
lutions are relatively insensitive to both.
The behavior in Figs. 47.4 and 47.5 is favorable be-
cause for the bulk of a reservoirs volume, P is usually
low (about 1 to 5 seconds -I), making it possible to at-
tain a design M with a minimal amount of polymer. Near
the injection wells, however, i can be quite high, which
causes the polymer injectivity to be greater than that ex-
pected on the basis of ~1 The relative magnitude of this
enhanced injectivity ef ect can be estimated once quan- P
titative definitions of shear rate in permeable media and
shear-rate/viscosity relations are given.
The polymer solution viscosity/shear-rate relationship
may be described by a power law model.
pp =K(iy , . . (2)
where K and n are the power-law coefficient and expo-
nent, respectively. Eq. 2 applies only over a limited range
of shear rates: below some low shear rate the polymer
solution viscosity is constant at pp O, and above the crit-
ical shear rate the polymer solution viscosity is also con-
stant p,O. The truncated nature of the power law is
awkward in calculations; hence, another useful relation-
ship is the Meter model. 22
O-P,, m
P p =P p m+ .
If L
c >
p 1%
where 01 is an empirical coefficient and > ,,? is the shear
rate at which CL,, is the average of pLr, and p,, -. As with
all polymer properties, all empirical parameters are func-
tions of salinity, hardness, and temperature.
When applied to permeable media flow thcsc gcncral where
trends and equations continue to apply. cc,, is usually k,, = permeability to polymer solution.
called the r(17/u.~rr viscosity and the effective shear
rate. in,. .
h,, = mobility of water-rich phase. and
is based on capillary tube concepts.
h,, = mobility of polymer solution.
ic =
u ii For constant-flow-rate experiments. FR is the inverse ra-
, . . . . . . . . . . . . . . . . .
tio of pressure drops; for constant-pressure-drop experi-
ments, FR is the ratio of flow rates. F, is an indication
of the total mobility-lowering contribution of a polymer.
To describe the permeability reduction effect alone. a prl--
II,,. = superficial flux of polymer-rich (water)
mwhilit~, rcductiorl ,firctor-. F,.k , is defined as
k,,. = permeability to polymer-rich (water) phase.
S,,. = saturation of polymer-rich (water) phase.
fraction, and
$ = porosity of medium, fraction.
Polymer Retention. All polymers experience retention
on solid surfaces because of adsorption or mechanical
trapping within a permeable medium. Polymer retention
varies with polymer type, molecular weight, rock com-
position, brine salinity and hardness, flow rate, and tem-
perature. Field-measured values of retention range from
20 to 400 Ibm polymer/acre-ft bulk volume with desir-
able retention level being less than about 50 lbm/acre-ft.
Retention causes the loss of polymer from solution, which
can cause the mobility control effect to be destroyed-an
effect that is particularly pronounced at low polymer con-
centrations. Polymer retention also causes a delay in the
rate of polymer propagation.
A final definition is the residual resistctnw jirctor. FRr .
which is the mobility of a brine solution before and after
polymer injection.
Offsetting the delay caused by retention is an accelera-
tion of the polymer solution through the permeable medi-
um. which is caused by inaccessible PV (IPV). The most
common explanation for IPV is that the smaller portions
of the pore space will not allow polymer molecules to enter
because of their size. Thus, a portion of the total pore
space is uninvaded or inaccessible to polymer. and ac-
celerated polymer flow results. Large as they arc, how-
ever, most polymer molecules will fit easily into all but
the smallest pore throats. Hence, a second explanation
of IPV ih based on a wall exclusion effect whereby the
polymer molecules aggregate in the center of a narrow
channel. The polymer pore fluid layer near the wall has
a lower viscosity than the fluid in the center. which causes
an apparent fluid slip.
IPV depends on polymer molecular weight, media per-
meability, porosity, and pore size distribution, becoming
more pronounced as molecular weight increases and the
ratio of permeability to porosity (characteristic pore size)
decreases. IPV can be 30% of the total pore space or
Permeability Reduction. As mentioned previously
HPAM polymers can cause lowered mobility through a
permeability reduction effect. This phenomenon has been
described through three factors. 24 The resistuncc~,fuctor.
FR, is the ratio of the injectivity of a single-phase poly-
mer solution to that of brine flowing under the same con-
A,,. k,,/tL,,,
x,, k,,cl,, .
FR,-=- . _. (7)
where A,,.,, is the mobility of water-rich phase before
polymer injection and A,,,(, is the mobility of water-rich
phase after polymer injection. F,Q is a measure of the
permanence of the permeability reduction effect caused
by the polymer solution. It is the primary measure of the
performance of a channel-blocking application of poly-
mer solutions. For many cases, F,k and FR,- are nearly
equal; however. FR is usually much larger than F,.,: be-
cause the former contains the viscosity-enhancing effect
as well as the permeability-reduction effect.
The most common measure of permeability reduction
is F,.k. Frk is sensitive to polymer type, molecular weight.
degree of hydrolysis, shear rate, and permeable media
pore structure. Polymers that have undergone even a small
amount of mechanical degradation seem to lose most of
their permeability reduction effect. For this reason.
qualitative tests based on screen factor devices are com-
mon to estimate polymer quality. F,.k has been correlat-
ed to polymer adsorption and rheological properties by
Hirasaki and Pope.
Chemical and Biological Degradation. The average
polymer molecular weight can be decreased, to the detri-
ment of the overall process, by chemical. biological, or
mechanical degradation. Chemical degradation can be
minimized by restricting polymer usage to low-
temperature applications, and by adding oxygen
scavengers (e.g., sodium sulfate or sodium sulfite) to the
polymer solution. Biological degradation can be elim-
nated by adding oxygen scavengers and biocides (e.g..
formaldehyde or isopropyl alcohol). In fact, nearly all ap-
plications contain some of these chemicals. usually in very
small quantities (see Ref. 16 or 26).
Mechanical Degradation. Mechanicaldegradation, on
the other hand, is potentially present under all applica-
tions. Mechanical degradation occurs when polymer so-
lutions are exposed to high-velocity flows. These can be
present in surface equipment (valves. orifices, pumps. or
tubing), downhole conditions (perforations or screens).
or in the sandface itself. Perforated completions. partic-
ularly, are a cause for concern. since large quantities of
Fig. 47.6-Tertiary polymer flood response from North Burbank
Unit, Osage County, OK.
polymer solution arc being forced through a number of
small holes. For this reason. most polymer in,jections are
done through openhole or gravel-pack completions. Be-
cause flow velocity falls off quickly with distance from
an injector. little mechanical degradation occurs within
the reservoir itself.
All polymers mechanically degrade under high enough
flow rates. However. HPAMs are most susceptible un-
der normal operating conditions. particularly if the sa-
linity or hardness ofthe brine is large. Evidently, the ionic
coupling of these anionic molecules is relatively fragile
compared with the polysaccharide chains. Also. clonga-
tional stress is ar destructive to polymer solutions as is
shear stress, although the two generally accompany each
other. Maerker and others have correlated permanent vis-
cosity of a polymer solution loss to an elongational stretch
rate times length product. 273 On a viscosity/shear-rate
plot (purely shear flow), mechanical degradation usually
begins at shear rates greater than the minimum-viscosity
critical shear rate.
Field Results. Fig. 47.6 shows a tertiary polymer tlood-
ing field response from the North Burbank unit, Osage
County, OK. The figure shows WOR and oil rate be-
fore and immediately after polymer injection. Polymer
injection began in late 1970, which quickly arrested a
declining oil rate and an increasing WOR. The oil rate
then resumed its prepolymer decline but at a substantial-
ly higher level. The polymer oil or incremental oil recov-
ery (IOR) from a polymer flood is the difference between
the cumulative oil actually produced and that which would
have been produced by a continuing waterflood. Thus.
for a technical analysis of the project it is important to
establish a polymer flood oil rate decline and an accurate
waterflood decline rate. The IOR for North Burbank is
the shaded area in Fig. 47.6.
Table 47. I summarizes other field results on more than
250 polymer tloods on the basis of a comprehensive sur-
vcy by Manning rt rrl. I7 The table emphasizes oil recov-
cry data and screening parameters used for polymer
flooding. Approximately one-third of the statistics arc
from commercial or field-scale floods. The oil recovery
statistics in Table 47. I show average polymer tlood recov-
cries of 3.56% remaining (after watertlood) oil in place.
and 2.69 STB of IOR for each pound of polymer itljcct-
cd. with wide variations in both numbers. The large varia-
bility reflects the emerging nature of polymer tlooding
in the previous decades. Considering the STB IOR per
pound of polymer average and the average costs of crude
and polymer, it appears that polymer flooding could be
a highly attractive EOR process. However. such costs al-
ways should be compared on a discounted basis rctlcct-
ing the time value of money. which will decrease the
apparent attractlvencss of polymer flooding because of the
decreased injectivity of the polymer solutions.
Foam Flooding
Gas/liquid foams offer an alternative to polymers for
providing mobility control in chemical floods, and have
been both proposed and field tested as mobility control
agents in steamfloods.
Foams are dispersions of gas bubbles in liquids. Gas/liq-
uid dispersions are normally unstable and usually will
break in less than I second. If surfactants are added to
the liquid, however, stability is improved greatly so that
some foams can persist indefinitely. To understand foam
Oil recovery, O/o remaining OIP
Oil recovery, STB/lbm
Oil recovery, STBlacre-ft
Permeability variation, fraction
Mobile oil saturation, fraction
Oil viscosity, cp
Resident brine salinity,
Water-to-oil mobility ratio,
Average polymer
concentration, ppm
Temperature, OF
Average permeability, md
Average porosity, fraction
Number of
87 5.86 0.1 51.8 11.05
93 339 51 3,700 343
172 115 46 234 85
187 349 1.5 7,400 720
193 0.20 0.07 0.38 0.20
Mean Minimum Maximum
3.56 0 25.3
2.69 0 36.5
24.0 0 188.7
0.70 0.06 0.96
0.27 0.03 0.51
36 0.072 1,494
40.4 5.0 133.0
propertie rcqulrcc wnc ili5cusvon of surtactunts and
their classifications. Moat of the discussion applies to MP
surfact~nts 3s well.
Surfxtant Chemistq. A typical xurfactant monomet- is
composed of a nonpolar portion (lipophi1c~i.c.. having
strong affinit~~~for oil) and a polar portion (hydrophile-
i.e.. strong attmity for water): the entire monomer is an
amphtphile (has affinity for both oil and vvater) because
ofthis dual nature. Fig. 47.7 shows the molecular struc-
ture of two common surfactanta (top two panels) and il-
lustrates (lowest panel) a shorthand notation for surfactant
monomers: the monomer is represented by a tadpole
sy~dx~l with the nonpolar portion being the tail and the
polar the head. Surfactants are classified into four groups
that depend on their polar portions (see Table 47.1)3.
Anionic.s. As required by electroneutrality. the an
ionic aurfactant molecule is uncharged with an inorganic
metal cation (usually sodium) associated with the
monomer. In an aqueous solution the ~~dx~~le ioniLcs
to free cations and the anionic monomer. Anionic sur-
factants are the most common in EOR because they are
good surfactants. rclativelv resistant to retention. stable.
and relatively inexpensive.
Cationics. In this case the surfactant ~~dxulc con-
tains a cationic hydrophilc and an inorganic anion to bal-
ancc the charpc. Cationic surfactants have little use in
EOR bccausc they are adsorbed highly by the anionic sur-
faces of interstitial clays.
Nonionics. These surfactants do not form ionic
bonds, but when dissolved in aqueous solutions. they cx-
hibit surfactant properties simply by electronegativity con-
trasts betvveen their constituents. Nonionics arc much less
sensittve to high salinities than anionics or cationics.
Anzphoterics. A final class of surfactants are those
that contain aspects of two or more of the previous class-
es. For example. an amphoteric may contain both an an-
ionic group and a nonpolar group. These surfactants have
not been used in EOR.
Within any one class there is a huge variety of possible
surfactants. Fig. 47.7 illustrates differences in Iipophile
molecular weight IC t2 for the sodium dodecyl sulfate
(SDS) vs. C th for Texas No. I]. hydrophilc identity (sul-
fate vs. sulfonate), and tail branching (straight chain for
SDS vs. two tails for Texas No. 1) all within the same
class of anionic surfactants. In addition to these. there are
variations in the position ofthe hydrophile attachment and
the number of hydrophiles (monosulfonates vs. disul-
fonatcs. for example). Even small variations can change
surfactant properties drastically (e.g.. sulfates tend to be
less thermally stable than sulfonatcs). For details on the
c: c
/ \,/ \cA,i\p\cAo _ g _ o- No+
R-S -O-Nat-
(non - polar)
Fig. 47.7-Typical surfactant molecular structures
effect of structure on surfactant propertics see Refs. 31
and 32.
Most commercial surfactants contain distributions of
surfactants and surfactant types that further add to their
complexity. In the following. distinctions between sur-
factant types are ignored by simply treating the surfac-
tant as the tadpole structure of Fig. 47.7.
Foam Stability. The stability of a foam may be under-
stood by viewing the liquid film separating two gas hub-
bles in cross section as in the lower panel of Fig. 47.8. ?
The hydrophiles of the surfactant are oriented into the in-
terior of the film and the lipophiles toward the bulk gas
phase. Suppose that some external force causes the film
to thin as in the lower panel. Since capillary pressure is
inversely proportional to interfacial curvature. the pres-
sure in the thinned portion of the film is lower than in
the ad,iacent flat portion. This causes a pressure differ-
cnce within the film. liquid flow. and healing. The pres-
sure in the gas phase is assumed constant because of its
relatively low density if the foam is static. or low vis-
cosity if in motion.
4-M -eM -e+ d-
Anionics Cationics Amphoterics Nonionics
Sulfonaies Quaternary ammonium Aminocarboxyltc Alkyl-
Sulfates Pyridinum acids Alkyl-aryl
Carboxylates lmidazolrnrum Others Acyl
Phosphates Piperidinium Acylamindo-
Others Sulfononium Acylaminepolyglycol ethers
Compounds Polyol ethers
Others Alkanolamides
ble layer on the interior of the film boundary and the
attractive Van der Waal forces bctwccn the molecules in
the film. If the film becomes substantially smaller than
the equilibrium thickness. the fret energy barrier between
the repulsive and attractive contributions is breached and
the film will collapse.
/ Such thinning can be caused spontaneously by diffu-
/ I
sion of the gas from small to large bubbles and by gravi-
ty drainage. Patton et al. reported on the rate of
3 Hii
spontaneous collapse of a large number of foams as a func-
tion of surfactant type. temperature, and pH. tl The half-
life of the foam heights reported in their static tests ranges
from 1 to about 45 minutes. They report that anionic sur-
factants have greater stability than nonionics. and that the
stability of sulfonate foams is affected greatly by water
hardness. Foams were generally more unstable at high
Thlnner, Lou adrorptlon, Larger aurfece tencllon,
temperatures, and many could be stabilized by adding a
second surfactant.
Fig. 47.8-Upper panel shows surface tension and adsorption
of a surfactant vs. concentration. Lower panel is the
Gibbs-Marangoni effect.
External effects that may cause the foam to collapse are
the presence of a foam breaker (oil or a high clcctrolyte
concentration could do this). local heating. or contact with
a hydrophobic surface.
0 km,
= 4.41 OMCY
* k,,,
* 0.42 DARCY
0 km
0.22 DARCY
!- * A
Fig. 47.9-Effective permeability-viscosity ratio vs foam quality
for consolidated oorous media.
The upper panel in Fig. 47.8 shows that the gas/liquid
surface tension is a decreasing function of surface adsorp-
tion as required by the Gibbs theory. According to this
view the thinned portion of the film will have less specif-
ic adsorption (since the surface area is locally greater) and
greater surface tension. This locally high surface tension
also causes healing.
Clearly. the surface tension at the gas/liquid interfaces
play an important role in film stability. Very low surface
tension would not be favorable; fortunately gas/liquid sur-
face tensions are rarely lower than 20 dynes/cm even with
the best foaming agents. In the absence of external forces.
the film is at an equilibrium thickness caused by a bal-
ance between the repulsion forces of the electrical dou-
Foam Physical Properties. Physically, foams arc char-
acterized by three measures.
Qua&. Foam quality, r, is the percentage of the total
(bulk) foam volume that is gas volume. The quality can
increase with increasing temperature and decreasing pres-
sure both because the gas volume can increase. and also
because gas dissolved in the bulk liquid phase can evolve
from solution. Foam qualities can bc quite high, approach-
ing 97% in many cases. A foam with quality greater than
90% is a dry foam.
Texture. This measure is the average bubble size. The
texture determines how the foam will flow through a
permeable medium. If the average bubble size is larger
than the average pore diameter. the foam flow5 as a
progression of films separating individual gas bubbles.
Given typical foam textures and pore sizes, this condi-
tion is most nearly realized in permeable flow. particu-
larly for high foam qualities.
Bubble Size Range. Foams with a large distribution
range are more likely to be unstable because of the gas
diffusion from large to small bubbles.
Mobility Lowering. Foams can reduce the mobility of
a gas phase drastically. Fig. 47.9 shows the steady-state
mobility of foams of differing quality in Berea cores at
three different permeabilities as a function of quality.
On the extreme right of this figure (r- 100%). the mo-
bility should approach the respective air permeability
divided by the air viscosity: this mobility is two to three
factors of 10 greater than any of the experimental points
on the figure. When r-0 the mobility should approach
the water permeability divided by CL,, Thus. the mobili-
ty of the foam is lower than that of either of its consti-
tuents alone. The mobility of the foam decreases with
increasing quality until the films between the gas bubbles
begin to break and the foam collapses (not shown on Fig.
47.9). Foams are effective in reducing the mobility at all
three permeabilities in Fig. 47.9, but the effect of foam
quality is more pronounced at the highest permeability.
This is a consequence of the contrast between the tcxturc
of the foam and the mean pore size of the mcdium.36
The mobility reduction caused by the foam can be
viewed as an increase in a single-phase viscosity or as
a decrease in the gas-phase permeability. Representative
data of the second type are in Fig. 47.10, which shows
the gas-phase permeability, both with and without foam,
as saturation plotted against the liquid injection
rate. - Note that the foam causes a great decrease in gas
permeability at the same rate and even at the same gas
saturation compared to the nonfoaming displacement. The
analogous analysis performed on the aqueous-phase rela-
tive permeability shows that neither the gas saturation nor
the presence of the foaming agent affects the aqueous-
phase relative permeability. j8
The low foam mobilities in permeable media flow are
postulated to be caused by at least two different mecha-
nisms: (I) the formation of or the increase in a trapped
residual gas phase saturation and (2) a blocking of pore
throats caused by the gas films. From Fig. 47.10 the ef-
fect of a trapped gas saturation, which would lower the
gas mobility through a relative-permeability lowering, is
much smaller than the pore-throat-blocking effect. The
trapped gas-phase saturation effect may become impor-
tant. however, during the later stages of a displacement
where the lower pressures could cause more gas to come
out of solution.
The mobility reduction of foams, viewed as a viscosit
enhancement. has been studied in capillary tubes.
General observations on these data are that foams are
generally shear-thinning fluids whose power law coeffi-
cient increases with the capillary tube radius. Using the-
oretical arguments based on a Newtonian fluid and an
inviscid gas, Hirasaki and Lawson showed that the film
thickness of a single moving bubble increases as the bub-
ble velocity to the two-thirds power. 4o Since shear stress
in a capillary is inversely proportional to film thickness.
the apparent viscosity of a foam in a capillary tube
decreases with increasing velocity. Thus, the shear-
thinning effect observed in capillary tubes is actually a
consequence of the film thickening as velocity increases.
A second implication in the Hirasaki-Lawson theory is
that foam texture occupies as great an importance in de-
termining rheological behavior as does foam quality.
Field Results. Field tests of foam injection alone have
been scarce. Holm reports on the injection of an air/brine
foam into a single well in the Siggins field. Though
there was no measureable oil response, the mobility to
both air and brine were reduced significantly, and the in-
jection profile into the central well became more uniform.
Low-IFT Processes
In addition to stabilizing gas dispersions, surfactants used
in MP flooding or generated in situ can recover oil by
lowering oil/water IFT.
Lowering ROS
The basic tool for illustrating how lowering IFT reduces
ROS is the capillary desaturation curve (CDC) shown in
Fig. 47.1 I. The CDC is a plot of nonwetting- or wetting-
phase residual saturation on the y axis vs. a dimcnsion-
less ratio of viscous to local capillary forces on a logarith-
mic .Y axis. On Fig. 47.1 I, S,,,.,, is the residual oil
(assumed nonwetting), and S,,,. is the irreducible water-
(wetting)-phase saturation. The CDC has been calculat-
l.T Ii iiiiiii iiiiil
0 I 2 3 4 5 6 7 8 9 IO II 12 13 14
Fig. 47.10-Effect of liquid flow rate and gas saturation on gas
permeability with and without foam
Fig. 47.11--Schematic capillary desaturatjon curves
ed theoretically, 4-w but the most common source of this
curve is experimental measurement. . The ratio of vis-
cous to local capillary forces is the capillary number, N,. 1
one form of which is
N,. =N ,, p ,,,/u,,.<, . (8)
For a waterflood, u,,. and p,,, are the superficial flux and
viscosity of the displacing water, and u,!(, is the IFT be-
tween the water and oil phases. For a MP flood a more
general definition is appropriate, but Eq. 8 and the CDC
can illustrate this case also.
The CDC is a nearly horizontal plateau at small N, un-
til a critical value above which both residual phase satu-
rations decrease. At a second critical N,.. the residual
phase saturations are zero and complete recovery of the
Fig. 47.12-Schematic of an MP flooding process.
originally trapped phase occurs. The shape of the CDC
is determined by the pore geometry of the medium and
the wetting behavior of the two phases. The wetting phase
rcquircs a larger N,. for complete recovery. To a lesser
extent. the CDC shape is affected by the mean pore size
and pore size distribution of the medium. and the initial
saturations. Typical N, s for watertlooding are quite
small. which indicates that ROS and S,,, may be assumed
constant for this purpose (XC Fig. 47. I I). Because of the
logarithmic .Y axis. a decrease by several factors of IO in
N, is necessary IO significantly change either residual
phase saturation. Ofthc three quantities in Eq. 8 only the
IFT can be changed this drastically; a typical value for
causing good ROS reduction is in the range of IO
dyne/cm. a value that can be obtained only with a good
Fig. 47.13-Schematic of the CMC
IMP Flooding
MP flooding has appeared in the technical literature un-
der many names: detergent. surfactant, low-tension. solu-
ble oil. microemulsion. and chemical flooding. There are
also several company names that imply a specific sequence
and type of injected fluids as well as the specific nature
of the oil-recovering MP slug itself. Though there are
differences among company processes. the common
aspects are more numerous and important.
An idealized version of a MP flooding sequence is in
Fig. 47.12. The process is applied invariably to tertiary
floods (those producing at high WORs) and is always im-
plemented in the drive mode (not cyclic or huff n puff).
The complete process consists of the following.
Prejlush. A volume of brine whose purpose is to change
(usually lower) the salinity and hardness of the resident
brine so that mixing with the surfactant will not cause the
loss of interfacial activity. Pretlushcs have ranged in size
from 0 to 100% of the reservoir PV. In some processes
a sacrificial agent is added to lessen the subsequent sur-
factant retention and also to precipitate divalent cation.
MP Slug. This volume. ranging from 5 to 40% PV in
field applications. contains the main oil-recovering agent.
the prituut:\~ surfactant, in concentrations ranging from I
to 20 ~0170. Several other chemicals (cosurfactants. al-
cohols. oil. polymer. biocide, and oxygen scavcnger) are
usually necessary for good oil recovery.
LJ40~i~i~ Buffeer. This fluid is a dilute solution of a
water-soluble polymer whose purpose is to drive the MP
\lug and banked-up fluids to the production wells. Much
of polymer flooding technology carries over to dcsign-
ing and implcmcnting the mobility buffer.-18
Freshwater Buffer. This i\ a volume of brine contain-
ing a concentration of polymer grading between that of
the mobility buffer at the front end and zero at the back.
The gradual concentration decrease mitigates the effect
of the unfavorable mobility ratio between the chase water
and mobility buffer.
Chase Water. The purpose of the chase water is sirn-
ply to reduce the expense of continually injecting poly-
mer. If the mobility and freshwater buffers have been
designed properly. the MP slug will be dcplcted before
it is penetrated,by the chase water.
Surfactant Solutions. If an anionic surfactant is dissolved
in brine, the surfactant disassociates into a cation and a
monomer. If the surfactant concentration then is increased,
the lipophilic portions of the surfactant begin to associate
among themselves to form aggregates or micelIes contain-
ing several monomers each. A plot of surfactant monomer
concentration vs. total surfactant concentration (Fig.
47.13) is a curve that begins at the origin, increases with
unit slope, then levels off at the critical micelle concen-
tration (CMC). Above the CMC all further increases in
surfactant concentration cause increases only in the micelle
concentration. CMCs are typically quite small (about
10 - to IO p4 mol/L). At nearly all practical concentra-
tions for MP flooding the surfactant is predominantly in
the micellc form; hence. the name micellar/polymer flood-
ing. The representations of the micelles in Fig. 47.13 and
elsewhere are schematic. The actual micelle structures can
take on various forms, which can fluctuate with time.
When this solution contacts an oleic phase (the term
olcic indicates that this phase can contain more than
oil) the surfactant tends to accumulate at the intervening
interface. The lipophilic tail dissolves in the oleic phase,
and the hydrophilic in the aqueous phase. The accumula-
tion at the intcrfacc causes the IFT between the two phases
to lower. The extent of the IFT lowering is proportional
to the cxccss surface concentration of the surf;lctant-the
difference between the surface and bulk surfactant
concentration-from Gibbs theory. as was the case in Fig.
47.X. The surfactant itself and the attending conditions
should bc adjusted to maximize the excess surface con-
cent]-ation; however, in doing so the solubility of the sur-
tactant in the bulk oleic and aqueous phases also is
affected. Since this solubility also impinges on the mutu-
al volubility of brine and oil, which also affects IFTs,
this discussion leads naturally to the topic of surfactanti
brine/oil phase behavior. Curiously, and this is true of
many micellar properties. the surfactant concentration it-
self plays a rather minor role in what follows, compared
with the temperature. brine salinity, and hardness.
SurfactantlBrineiOil Phase Behavior. Surfactanti
hrinc/oil phase behavior is illustrated conventionally on
a ternary diagram. A ternary diagram is an equilateral
triangle whose apexes represent pure components. bound-
arlcs represent two-component mixtures, and interior rep-
resents three-component mixtures. For complicated
mixtures the three apexes must represent pseudocom-
poncnts whose composition remains constant through-
out the diagram. The pressure and temperature are also
fixed. The diagram can represent both the overall com-
position of a surfactantibrineioil mixture. and the equi-
librium composition of each phase if the mixture forms
more than one phase. Tcrnarica and their accompanying
definitions are discussed in Chap. 23.
MP phase behavior is affected strongly by the salinity
of the brine pseudocomponent. Consider the sequence of
phase diagrams (Figs. 47.14 through 47.16) as the brine
salinity is incrcaxcd. The phase behavior about to bc dc-
hffCROE~LS/ON 011
Fig. 47.14-Schematic of low-salinity surfactantlbrineioil phase
scribed was presented originally by Winsor and adapted
to MP flooding later.50.5
At low brine salinity, a typical MP surfactant will ex-
hibit good aqueous (water-rich) phase solubility and poor
oleic (oil-rich) phase solubility. Thus, an overall compo-
sition near the brine/oil boundary of the ternary will split
into two phases: an e,xcess oil phase that is essentially pure
oil and a (water-external) rnicroemulsio~~phase that con-
tains brine surfactant, and some solubilized oil. The solu-
bilized oil occurs when globules of oil occupy the central
core of the swollen micelles. The tie lines within the two-
phase region have a negative slope. This type of phase
environment is called variously a Winsor Type I system.
a lower-phase microemulsion (because it is more dense
than the excess oil phase), or a Type II system. The lat-
ter terminology is adopted here-11 means that no more
than two phases can (not necessarily will) form and (-)
means that the tie lines have negative slope (Fig. 47.14).
The right plait point in such a system. PK. usually is lo-
cated quite close to the oil apex. Any overall composi-
tion above the binodal curve is single phase.
For high brine salinities (Fig. 47. IS) electrostatic forces
drastically decrease the surfactants solubility in the aque-
ous phase. An overall composition within the two-phase
Iregion now will split into an UUJ,S.\ hr-i/~c phase and an
Fig. 47.15-Schematic of surfactant/brine/oil high-salinity phase
(oil-external) microemulsion phase, which contains most
of the surfactant and some solubilized brine. The brine
is solubilized through the formation of inverted swollen
micelles (Fig. 47.15) with brine globules at their cores.
The phase environment is a Winsor Type II system, an
upper-phase microemulsion, or a Type II( +) system. The
plait point, PI,. is now close to the brine apex.
The two extremes presented thus far are roughly mir-
ror images; note that the microemulsion phase is water-
continuous in the Type II( -) systems and oil-continuous
in Type II(+) systems. The induced solubility of oil in
a brine-rich phase, Type II( -). suggests an extraction
mechanism in oil recovery. Though extraction does play
some role. it is dwarfed by the IFT effect discussed later.
particularly when phase behavior at intermediate salini-
tics is considered.
At salinities between those of Figs. 47. I4 and 47.15.
there is a continuous change between Type II( -) and
II( +) systems and a third surfactant-rich phase is formed.
as shown in Fig. 47.16. An overall composition within
the three-phase region separates into excess oil and brine
phases as in the Type II(-) and II( +) environments. and
into a microemulsion phase whose composition is repre-
sented by an ;U~U~;LUU point. This environment is called
Fig. 47.16-Schematic of surfactant/brine/od phase behavior at
optimal salinity.
a Windsor Type III, a middle-phase microemulsion. or
a Type III system. Above and to the right and left of the
three-phase region are Type II( -) and II( +) lobes wherein
two phases will form as before. Below the three-phase
region there is a third two-phase region (as required by
thermodynamics) whose extent is usually so small that it
is neglected. In the three-phase region there are now
two IFTs between the microemulsion and oil, a,,,,, , and
the microemulsion and water, u,~~,,..
Fig. 47.17, a prism diagram, shows the entire progres-
sion of phase environments from Type II( -) to II( +).
The Type III region forms through the splitting of a criti-
cal tie line that lies close to the brine/oil boundary as the
salinity increases to C,, (low effective salinity limit for
Type III phase environment). 53 A second critical tie line
also splits at C,, (high effective salinity limit for Type
III phase environment) as salinity is decreased from a Type
II( +) environment. Over the Type III salinity range the
invariant point, M, migrates from near the oil apex to near
the brine apex before disappearing at the appropriate crit-
ical tie lines. The migration of the invariant point implies
essentially unlimited solubility of brine and oil in a sin-
gle phase, which has generated an Intense research in-
terest into the nature of the Type III microemulsion.
Fig. 47.17-Prism diagram showing sequence of phase environment transition.
Several variables other than salinity can bring about the
Fig. 47.17 phase environment shifts. In general, chang-
ing any condition that enhances the surfactants oil solu-
bility will cause the shift from Type Il( -) to II( +). Some
of the more important are: (1) decreasing temperature, 5
(2) increasing surfactant molecular weight, (3) decreas-
ing tail branching, (4) decreasing oil specific
gravity55-57 and (5) increasing concentration of high-
molecular-weight alcohols. 58 Decreasing the surfactants
oil solubility will cause the reverse change. Thus, Fig.
47.17 could be redrawn with any of the above variables
(and several others) on the base of the prism with the vari-
able C,, (effective salinity) increasing in the direction of
increased oil solubility.
Nonideal Effects. In much the same manner as the ideal
gas law approximates the behavior of real gases, Fig.
47. I7 is an approximation to actual MP phase behavior.
Some of the more important nonidealities are as follows.
I. At high surfactant concentrations and/or at low
temperaturess.59 or even in the presence of pure surfac-
tants 6. phases other than those on Fig. 47.17 have been
L 3
observed. These phases tend to be high-viscosity liquid
crystals or other condensed phases. The large viscosities
are detrimental to oil recovery since they can cause local
viscous instabilities during a displacement. Frequently,
low- to medium-molecular-weight alcohols (cosolvents)
are added to MP formulations to melt these undesira-
ble viscosities. When the brine contains polymer. a con-
densed phase can be observed at low surfactant
concentration because of exclusion of the polymer from
the microemulsion phase. Cosurfactants can be used to
eliminate this polymerisurfactant incompatibility.
2. When cosurfactants are present it is frequently in-
appropriate to lump all of the chemicals into the surfac-
tant apex of the Fig. 47.17 prism. If the cosurfactants do
not partition with the primary surfactant during a displace-
ment the benefit of adding the chemical is lost; hence.
surfactanticosurfactant separation effects are important.
Efforts to account for the preferential partitioning of the
cosurfactant include a quaternary phase behavior represen-
tation and a pseudophase theory. 62.63
3. The Type III salinity limits (C,, and C,,) are func-
tions of surfactant concentration. This dependency may
be visualized by tilting the vertical triangular planes in
Fig. 47.17 about their bases. This dilution effecth.h5
forms the basis for the salinity requirement diagram de-
sign procedure. UJ The dilution effect is particularly
pronounced when the brine contains significant quanti-
ties of divalent ions.
4. The Fig. 47.17 phase-behavior shifts are specific to
the exact ionic composition of the brine. not simply to
the total salinity. For anionic surfactants. other anions in
solution have little effect on the MP phase behavior: how-
ever. cations readily cause phase-environment changes.
Divalent cations (calcium and magnesium are the most
common) are usually 5 to 20 times as potent as monova-
lent cations (usually sodium). Divalents arc usually pres-
ent in oilfield brines in smaller quantities than monovalents
as shown in Fig. 47.2, but their effect is so pronounced
that it is necessary, as a minimum. to separately account
for salinity and hardness. Nonconstant monovalentidiva-
lent ratios also will cause electrolyte interactions with clay
minerals through cation exchange. The disproportionate
effects of the salinity and hardness are accounted for by
defining a weighted sum of the monovalent and divalcnt
concentrations as an effective salinity. The C,.s in Fig.
47. I7 imply effective salinities.
Phase Behavior and IFT. Early MP flooding literature
contains considerable information about the techniques of
measuring IFTs and what causes them to be low. 6x
IFTs were found to vary with the types and concentra-
tion of surfactant, cosurfactant, electrolyte, oil, and poly-
mer and with temperature. However, in what was surely
one of the most significant advances in all of MP tech-
nology, all IFTs were shown to correlate directly with
the MP phase behavior. The correlation was proposed
originally by Healy and Reed, theoretically substantiat-
ed by Huh, 6y and since verified experimentally by sever-
al others. I. The practical benefit of this correlation is
that relatively difficult measurements of IFTs can be
largely supplanted by simple phase behavior measure-
ments. Indeed, in the recent literature, the behavior of
IFTs has been inferred by a narrower subset of phase-
behavior studies based on the solubilization parameter. hs
Fig. 47.18--Correlation of solubilization ratios with IFT.
1 4 1 1 I 1 1 I I 1 I L I I 1 I
0 0.4 0.6 1.2 I.6 2.0 2.4 2 .a 3.2
Fig. 47.19-IFT and solubilization ratios
To investigate further the relation between IFTs and
phase behavior, let V,,, V,, , and V, be the volume frac-
tions of oil, brine. and surfactant in the microemulsion
phase, respectively. According to Figs. 47. I4 through
47.16, the microemulsion phase is present at all salini-
ties: hence. all three quantities are well-defined and con-
tinuous. Considering the Type II( -) behavior of Fig.
47.14, for example, V,,, V,,., and V,, are the coordinates
of the microemulsion phase composition on the binodal
Solubilization parameters between the microemulsion-
oleic phases, F,,,, , for Type II( -) and III phase behavior.
and between the microemulsion-aqueous phases. F,,,,,.. for
Type II( +) and III are defined as
F,,,,,=V,,iV, . . . . . . . . . . . . . . . . . . . . . . . . . . . ..(9a)
F,,,,,.=l,,./l,. . . . . (9b)
The IFTs between the corresponding phases, u,,,(, and
are functions only of F,,, and F ,,,, ~. Fig. 47.18
shows a typical correlation.
The corresponding behavior of the solubilization pa-
rameters and IFTs are shown in Fig. 47.19 in a differ-
ent manner. Consider a locus at constant oil, brine, and
surfactant overall concentrations in Fig. 47. I7 but with
a variable salinity. If the nonideal effects are unimpor-
tant and the locus is at low surfactant concentration and
intermediate brine/oil ratios, (T,~(, will be defined from
low salinity up to C,,, and uInM. from C,,, to high salini-
ties. Both IFTs are the lowest in the three-phase Type
III region, between C,, and C,,, where both solubiliza-
tion parameters are also large. There is, further, a pre-
cise salinity where both IFTs are equal at values low
enough ( - 10 -3 dyne/cm) for good oil recovery. This
salinity is the optimal saliniv for this particular surfac-
tant/brine/oil combination and the common IFT is the op-
timal IFT. Optimal salinities have been defined on the
basis of equal IFTs, as in Fig. 47.19, equal solubiliza-
tion parameters, maximum oil recovery in corefloods, and
equal contact angles. 50*71~72 All definitions of optima1 sa-
linity give roughly the same value; hence, since optimal
phase behavior salinity is the same as maximum oil recov-
ery salinity, generating an interfacially active MP slug
translates into generating this optimal salinity in situ in
the presence of the surfactant material.
Generating Optimal Conditions. Historically there have
been three techniques for generating optimal conditions
in an MP displacement.
I. The MP systems optimal salinity can be raised to
that of the resident brine salinity in the candidate reser-
voir. This procedure philosophically is the most satisfy-
ing of the three design procedures given here and usually
the most difficult. Though it has been the sub.ject of in-
tensive research, surfactants that have high optimal sa-
linities that are not (at the same time) thermally unstable
at reservoir conditions, excessively retained by the solid
surfaces, or expensive have not yet been discovered. Field
successes with synthetic surfactants have demonstrated the
technical feasibility of this approach. however. 73 A scc-
Fig. 47.20-Schematic of surfactant adsorption on metal oxide surface
ond way to make the optimal salinity of the MP formula-
tion equal to the resident brine salinity is to add
2. Resident salinity of a candidate reservoir can be lo-
wered to match the MP slugs optimal salinity. This is
the main purpose of the pretlush step illustrated in Fig.
37.12. A successful pretlush is appealing because. with
the resident salmtty lowered. the MP slug would displace
oil wherever it goes in the reservoir. Preflushes general-
ly require quite large volumes to lower the resident SB-
linity significantly owing to mixing effects and cation
exchange. 7.75 With some planning, the function of the
pretlush could be accomplished during the watertlood
preceding the MP flood.
3. The salinity gradient design attempts to dynamical-
ly lower the resident salinity to an optimum during the
displacement by sandvviching the MP slug between the
overoptimal rcsidcnt brine and an underoptimal mobility-
buffer salinity. hh.67 The \ucccss of this procedure relies
on it being necessary that only a portion of the MP slug
be in the active region for good oil rccovcry. For salinity
gradient tlonds the salinity of the mobility buffer is the
mot signiticant factor in bringing about good oil rccov
crv x The salinity gradient design has several other ad-
vanrazes in being rcstlient to design and process
unccrtaintics. in providing a favorable cnv~ironmcnt for
the polymer in the mobility buffer. minimizing retention.
and being relatively indifferent to the surfactant dilution
Surfactant Retention. Surfactants are retained through
one of at least four mechanisms.
I, On metal oxide surfaces (Fig. 47.20) the surfactant
monomer will adsorb physically through hydrogen hond-
inp and micelle-like associations with the monomer tails.
and ionically bond vvith cationic surface sites (I). At
higher surtactant concentrations. C,. this association in-
clt~dc?~ tail-to-tail interactions with the solution monomers
with proportionally greater adsorption (II and Ill). At and
above the CMC (IV), the supply of monomers bccomcs
constant as does the retention (c, is the adsorbed surfac-
tant concentration).
2. In hard brines the prevalence of divalent cations
causes the formation of surfactantidivalent complexes,
which have a low solubility in brine. * Precipitation of
this surfactantidivalent complex will lead to retention.
When oil is present this effect is lessened by the surfac-
tams solubility in the oleic phase.
3. At hardness levels somewhat lower than those re-
quired for precipitation, the preferred multivalentisurfac-
tmt complex will be a monovalent cation that can
exchange chemically with cations originally bound to the
reservoir clays just as inorganic cations do.
4. In the presence of oil in a II( +) phase environment
the aurfactant will reside in the oil-external microemul-
sion phase. Because this region is above the optimal sa-
linity, the IFT is relatively large (Fig. 47.18) and this
phase and its dissolved surfactant can be trapped. A
similar phase trapping effect does not occur in the II( - )
environment because the aqueous mobility buffer misci-
bly displaces the trapped aqueous-external microemulsion
phase without permanent retention.
Most studies of surfactant retention have not made the
previously mentioned mechanistic distinctions; thercforc,
which mechanism predominates in a given application is
not obvious. All mechanisms retain more surfactant at
high salinity and hardness, which, in turn. can be attenu-
ated by adding cosurfactants. Precipitation and phase trap-
pin,g,can be eliminated by lowering the mobility buffer
salimty, at which conditions the chemical adsorption
mechanism on the reservoir clays is predominant. There-
fore. there should be some correlation of surfactant rctcn-
tion with reservoir clay content. Fig. 47.2 I is an attempt
to make this correlation by plotting laborator
and field
surfactant retention data against clay fraction. ) The cor-
relation is by no means perfect since it ignores variations
in MP formulation and clay type distribution as well as
salinity effects. HowevJer. the figure does capture a gener-
al trend that is useful for a first-order estimate of reten-
tion in a given reservoir.
Fig. 47.21-Surfactant retention and weight fraction of clays.
Fig. 47.22-MP production response from Well 12-l. Bell Creek
0 0.2-0:4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2 0. 3.2
Recovery Efficiency vs. Mobility Buffer Size
Fig. 47.23--Recovery efficiencies from 21 MP field tests
A useful way to estimate the volume of surfactant re-
quired for an MP slug is through the dimensionless fron-
tal advance lag, FL).
1-o Pr c
4 P,C,
FD = frontal advance lag, dimensionless,
p,. = density of rock, mass/volume rock,
p\- = density of surfactant slug, mass/volume
C, = concentration of surfactant, mass
surfactantimass solution, and
6, = retained surfactant concentration, mass
surfactant/mass rock (includes all forms
of surfactants).
F expresses the volume of surfactant retained at its in-
jected concentration as a fraction of the PV. For best sur-
factant usage, the volume of surfactant injected should
be large enough to contact all of the PV, but small enough
to prevent excessive production of the surfactant. There-
fore, the MP slug size, V,I,v, should be equal to or some-
what larger than FD.
Field Response. Fig. 47.22 shows the produced fluid
analyses of Well 12- 1 in the Bell Creek (Carter and Pow-
der River Counties, MT) MP flood. This tlood used a
high oil content MP slug preceded by a pretlush that con-
tained sodium silicate to lessen surfactant retention and
reduce divalent cation concentration. Well 12-l was a
producer in the center of a unconfined single 40-acre, five-
spot pattern. Further details on the tlood are available in
Refs. 47 and 82.
Before MP slug injection in Feb. 1979, Well 12-l was
experiencing low and declining oil cuts. MP oil response
beginning in late 1980 is superimposed on this decline,
reaching peak cuts of about 13% about 6 months later.
Note that. just as in polymer flooding, the pre-MP decline
must be clearly established for accurate evaluation of the
MP oil recovery. The surfactant is preceding the oil in
Fig. 47.22 because of an excessively large content of
water-soluble, inactive disulfonate in the MP slug. Simul-
taneous oil and surfactant production is a persistent fea-
ture of field MP floods, probably because of
heterogeneities and dispersive mixing. Other significant
features in Fig. 47.22 are the evident presence of the
pretlush preceding the MP slug, inferred from the maxima
in the pH and silicate concentrations. and the very effi-
cient removal of the calcium cations ahead of the sur-
Fig. 47.23 shows oil recovery efficiency, EK (in-
cremental oil recovered/OIP at start of MP process), from
a survey of more than 40 MP field tests correlated as a
function of mobility-buffer slug size. As of the date of
the survey there were no commercial projects reported.
Similar analyses on other process variables showed no
or weak correlation. X3 The strong correlation in Fig.
47.23 indicates the importance of mobility control in MP
design. Note also from this figure that ultimate oil
recovery efficiency averages about 30% in field tests.
Performance Prediction. Generally. three things must
be achieved for efficient oil recovery. X4 ( I ) the MP sur-
factant slug must be propagated in an interfacially active
mode, (2) enough surfactant must be injected so that some
of it is unretained by the permeable media surfaces, and
(3) the MP displacement must be designed so that the ac-
tive surfactant sweeps a large portion of the reservoir
without excessive dissipation (because of dispersion) or
Attaining the first goal is the result of formulation work
based on the phase-behavior concepts discussed previous-
ly. The extent to which the second and third goals are
satisfied depends on prevailing economics, which, in turn,
depends on the oil-recovering ability of the entire proc-
ess. The next few paragraphs describe a simple proce-
dure by which oil recovery and oil rate-time curves may
be estimated for an interfacially active MP process. Since
there are innumerable ways in which interfacial activity
may be lost. the procedure is most accurate for processes
that clearly satisfy the first design goal.
Recovery Efficiency. This procedure has two steps: es-
timating the recovery efficiency of an MP flood and then
proportioning this recovery according to in.jectivity and
fractional flow to give an oil rate-time curve. Space con-
siderations will limit many of the details, which may be
found in Ref. 80.
The recovery efficiency. ER, of an MP flood is the
product of a volumetric sweep efficiency, E. a displace-
ment efficiency. En, and a mobility buffer efficiency.
ER=EDEVEMB, . . . (11)
Each quantity must be calculated independently.
Displacement Efficiency. The displacement efficien-
cy of an MP flood is the ultimate (time-independent)
volume of oil displaced divided by the volume of oil con-
ED=1 -p, . .(l2)
where S,,, and S,,. are the ROS to an MP and water-
flood, respectively. S,,,,. must be known, but S,,, can
be obtained from a large slug (free from the effects of
surfactant retention) laboratory coreflood. Low values of
indicate successful attainment of good interfacial ac-
tivity in the MP slug. If coreflood results arent availa-
ble, S,,
may be estimated from a CDC by using a
field capillary number.
N,=0.565q~L,,y,~l(h~), .._.. ..(13)
N,. = capillary number, dimensionless,
q = injection/production rate,
CL 0
= oil viscosity,
h = net thickness, and
A = pattern area.
Eq. 13 is in consistent units. For screening purposes, as-
sume a controlling IFT of lop3 dyne/cm [ 1 pN/m]; the
CDC curve chosen to estimate S ,,r should be consistent,
Fig. 47.24-Relationship between MP volumetric sweep efficien-
cy, heterogeneity, and slug-size retention ratio.
as much as possible, with conditions of the candidate
Volumetric Sweep Efficiency. Volumetric sweep effi-
ciency, E v, is the volume of oil contacted divided by the
volume of target oil. EV is a function of MP slug size,
V ,,, , retention, FD, and heterogeneity based on the
Dykstra-Parsons coefficient, KDP. Fig. 47.24 shows this
relationship. KDP may be estimated from geologic study,
from matching the previous waterflood, or from core data
(a typical value would be 0.6). The FD is estimated from
Eq. IO on the basis of the retention level, C,, . surfactant
slug concentration, C,, porosity, and rock and fluid den-
sities. C, can come from a laboratory coreflood, Fig.
47.21 (if clay fraction is known), or by using C, =0.4
mglg as a default. L,,s and C,, are from the proposed
MoKlity Buffeer Efficiency. The mobility buffer effi-
ciency, EBB, is a function of EV and KDP.
E,+,Be=0.71-0.6KDP, . . (l4a)
where EMBr is the mobility buffer efficiency extrapolated
to VM~=0 and VM~ is the mobility buffer volume, frac-
tion V,,. Eq. 14a was obtained from a numerical simu-
The recovery efficiency now may be calculated from
Eqs. 1 I through l4a. The reasonableness of the value may
be checked with Fig. 47.23.
Calculation of q,, vs. t Plot. The production function
(oil rate vs. time) is based on ER and the following proce-
dure. The dimensionless production function is assumed
to be triangular with oil production beginning at oil bank
arrival time, increasing linearly to a peak (maximum) oil
cut when the surfactant breaks through, and decreasing
linearly to sweepout time. The triangular shape is imposed
by the reservoir heterogeneity.
The first step is to calculate the dimensionless oil bank
and surfactant breakthrough times for a homogeneous lab-
oratory coreflood.
b I
0 100 roe 300 400 500 WO 700
Fig. 47.25-Comparison between predicted and observed oil
rate-ttme responses for the Sloss MP pilot.
t~,=l+F,-S,,,. , _. __ (14c)
t Lhh
= oil bank arrival time, dimensionless.
s ,,,, = oil bank saturation, fraction,
S,,, = initial oil saturation, fraction,
fr,,, = oil bank oil fractional flow. fraction.
f,,, = initial oil fractional flow, fraction. and
= surfactant arrival time, dimensionless.
S,,h andf,,,, may be estimated from the oil/water relative
permeability curve as described previously in Refs. 80
and 85. or from laboratory experiment.
The second step is to correct these values for the heter-
ogeneity of the candidate reservoir by using an effective
mobility ratio. M,. where
The corrected breakthrough times are now
ttDoh =tDot,lMe . (16a)
tD,=tDs/Mc. (16b)
where tD(,,, is the corrected oil bank arrival time, dimen-
sionless, tlDr is the corrected surfactant arrival time.
dimensionloss. and the peak oil cut. ,f&. i,
M,,-M,, tl)r,h
c >
! ?
.fy,n =
,f;,,, (17)
The final step is to convert the dimensionless production
function to oil rate, yo. vs. time. t. This follows from
q,,=qf;, ---- (l&i)
t=V,,tD/q, . .(18b)
wheref,, is the oil cut. tD is the dimensionless time. and
.f;, and tn are any points on the triangular oil recovery
curve, which begins at (t,,/, ,f;,;). peaks at (tljA .f;,,,k).
and ends at (tn,t., 0). The dimensionless time at com-
plete sweepout, tD,,,,., is selected to make the arca un-
der thef;, -tlD curve equal to ER,
tD,,l.=tnoh +2E,$,,,.,,Jf;,,,,, (19)
A comparison of the results of this procedure with the
Sloss field MP pilot is in Fig. 47.25. Details of this match
and other matches are in the original references.
High-pH Processes
The final chemical flooding EOR process is high-pH
flooding (Fig. 47.26). As in polymer and MP flooding.
there is usually a brine preflush to precondition the reser-
voir, a finite volume of the oil-displacing chemical, a
graded mobility buffer driving agent. and the entire proc-
ess is driven by chase water. Moreover. for both high-
pH and MP flooding the oil-displacing chemical is a sur-
factant; however, for MP flooding the surfactant is in-
jected while in high-pH flooding it is generated in situ.
High-pH Chemistry
High pHs indicate large concentrations of the hydroxide
anions (OH -). The pH of an ideal aqueous solution is
defined as
pH= -log ioc, + , . . (20)
where the concentration of hydrogen ions, CH + , is in
mol/L. As the concentration of OH - is increased, the
concentration of Hi decreases, since the two con-
centrations are related through the dissociation of water,
K,,, = (
, . . . . . . . . . . . .
and the water concentration is nearly constant. These con-
siderations suggest two means for introducing high pHs
into a reservoir: dissociation of a hydroxyl-containing spe-
cies or adding chemicals that preferentially bind hydro-
gen ions.
Many chemicals could be used to generate high pH. but
the most commonly used are sodium hydroxide (caustic,
NaOH). sodium orthosilicate. and sodium carbonate
(Na2C03). NaOH generates OH by dissociation: the
Fig. 47.26-Schematic of high-pH flooding process
latter two through the formation of weakly dissociating
acids (silicic and carbonic acid, respectively) that remove
free H ions from solution. High-pH chemicals gener-
ally have been used in field applications in concentrations
ranging up to 5 wt% (in.jected pHs of 1 I to 13) and with
slug sizes up to 2O%PV. The resulting amounts of chem-
icals are quite comparable with the surfactant usage in
MP flooding: however, high-pH chemicals are substan-
tially less costly. This cost advantage must be discounted
by the historically lower oil recoveries in high-pH
OH by itself is not a surfactant. since the absence of
a lipophilic tail makes it exclusively water-soluble. If the
crude oil contains an acidic hydrocarbon component,
HA,,. some of this, HA,,., can partition to the aqueous
phase where it can react. Xh
HA,,F-?A,;+H+. _. __ .
The exact nature of HA,, is unknown and probably highly
dependent on crude oil type. The deficiency of hydrogen
ions in the aqueous phase will cause the extent of this reac-
tion to be to the right. The anionic species A,; is a sur-
factant that can have many of the properties and enter into
most of the phenomena described above for MP flooding.
If there is no HA,, originally present in the crude. lit-
tle surfactant can be generated. A useful procedure for
characterizing crudes for their attractiveness to high-pH
flooding is through the acid number. The acid number
is the milligrams of potassium hydroxide (KOH) required
to neutralize I gram of crude oil. To make this measure-
ment, the crude is extracted with water until the acldlc
species HA is removed. The aqueous phase containing
HA,, , A,;, and H is then brought to pH=7 by adding
the KOH. For a meaningful value, the crude must be free
of acidic additives (e.g.. scale inhibitor) and acidic gases
(CO? or H?S). A good high-pH flooding crude will have
an acid number of 0.5 mgig or greater. but acid numbers
as low as 0.2 mgig may be candidates, since only a small
amount of surfactant is required to saturate oil/brine in-
terfaces. Fig. 47.27 presents a histogram of acid num-
bers based on the work of Jennings.87.8x
Displacement Mechanisms
Oil recovery mechanisms in high-pH flooding have been
attributed to eight separate phenomena. 89 This chapter
concentrates on only three: IFT lowering. wettability
reversal, and emulsion formation. The last two mecha-
nisms also are present in MP flooding but are dwarfed
by the low-IFT effect. With smaller ultimate oil recover-
ies, the distinction among effects becomes important in
high-pH flooding.
0 I 2 3 4 3
ACID INDEX INTERWL (E&Cl4 05mp. KOH/p RAf f i E)
Fig. 47.27-Histogram of acid numbers
1 M 0.26 wt. 46 NaCI
M 0.50 wt. % NaCI
A 1.00 wt. 16 N&I
z lOOr
lo- -
Fig. 47.28-IFTs for caustic/crude/brine systems
The generated surfactant, A ,, aggregates at oil/water
interfaces, which can lower IFT. 86 In general, such low-
ering is not as pronounced as in MP flooding but, under
certain conditions, can be large enough to produce good
oil recovery. Fig. 47.28 shows IFT measurements of caus-
tic solutions against Long Beach crude oil at various brine
salinities. The IFTs are sensitive to both NaOH concen-
tration and salinity, showing minima in the NaOH con-
centration range of 0.01 to 0.1 wt %. The decrease in IFT
in these experiments is limited by the spontaneous emul-
sification of the oil/water mixture when the IFT reaches
a minimum.
There are many similarities in the low-IFT effects in
MP and high-pH flooding. The data in Fig. 47.28 show
a clear resemblance to the data in the upper plot of Fig.
47.19 except they are plotted vs. NaOH concentration
(presumed proportional to A, concentration) instead of
salinity. This suggests an optimal salinity of about I .O
wt% NaCl for a 0.03 wt% NaOH solution. Indeed, the
work of Jennings et al. 87 has shown that there is an op-
timal NaOH concentration for a given salinity in oil recov-
ery experiments. Moreover, the presence of the
emulsification effect when IFTs are low is exactly what
one would expect from Fig. 47.17 at a surfactant con-
centration above the invariant point surfactant concentra-
tion. This suggests that the data in Fig. 47.28 showing
a Type II( -) phase environment at low NaOH concen-
trations are Type II(+) at high (similar to what would
be expected from the dilution effect in MP flooding). Fur-
ther work is necessary to establish the connection to MP
phase-behavior definitively, since the actual surfactant
concentration A; is likely to be much lower in a high-
pH system. However, Nelson PI a!. )( show that a cosur-
factant can increase the optimal salinity in a high-pH sys-
tem much like in MP systems.
See Chap. 28 for a discussion of wettability and its ef-
fects on petrophysical properties. Owens and Archery
showed that increasing the water wetness increased ulti-
mate oil recovery, where the wettability was reported as
decreasing the water/oil contact angle measured on
polished synthetic surfaces. This has also been shown by
others using high-pH chemicals.9.YX The increased oil
recovery is the result of two mechanisms: a relative per-
meability effect, which causes the mobility ratio of a dis-
placement to decrease, and a shifting of the CDC (see Fig.
47.1 I).
Cooke et al. 94 have reported improved oil recovery
with increased oil wetness. Other data show that oil recov-
ery is a maximum when the wettability of a permeable
medium is neither strongly water- nor oil-wet. )5 Given
the latter information, the important factor may be the
change in the wettability rather than the actual wettabili-
ty of the final state of the medium. In the original wet-
ting state of the medium, the nonwetting phase occupies
large pores and the wetting phase small pores. If the wet-
tability of a medium is reversed, there will be nonwet-
ting fluid in small pores and wetting fluid in large pores.
The resulting fluid redistribution, as the phases attempt
to return to their natural state, would make both phases
vulnerable to recovery through viscous forces.
High-pH chemicals can cause improved oil recovery
through the formation of emulsions. The emulsification
produces additional oil in at least two ways: through a
mobility ratio lowering since many of these emulsions
have a substantially increased viscosity and through
solubilization and entrainment of oil in a flowing aque-
ous stream. The first mechanism improves displacement
and volumetric sweep as do the mobility control agents
discussed previously. Local formation of highly viscous
emulsions should be discouraged, however, as these
would promote viscous fingering from the less viscous
oil-free high-pH solution. The solubilization and entrain-
ment mechanism would be more important when the IFT
between the swollen water phase and the remaining crude
is low. Fig. 47.28 shows that for certain conditions, emul-
sification and low IFTs occur simultaneously. McAuliffe
showed that emulsions injected in a core and those formed
in situ give comparable oil recoveries.96.97
Rock/Fluid Interactions
Interactions of the high-pH chemicals and the permeable
media minerals can cause excessive retardation in the
propagation of these chemicals throughout the permea-
ble medium. This chapter discusses three aspects of rock-
fluid interactions: formation of divalent/hydroxide com-
pounds, cation exchange, and mineral dissolution.
OH - ions themselves are not appreciably bound to the
solid surfaces; however, in the presence of multivalent
cations they can form hydroxyl compounds,
M +-+x(OH -)FiM(OH),,, (23)
which, being relatively insoluble, can precipitate from so-
lution. This reaction, in turn, lowers the pH of the solu-
tion, and also can cause formation damage through pore
blockage and fines migration. The anionic surf&ant spe-
cies A,, can interact with the inorganic cations in solu-
tion just as in MP flooding; however, the interaction with
the divalent cations usually takes precedence, particular-
ly in hard brines (see Fig. 47.3) or where there are sub-
stantial quantities of soluble multivalent minerals. Because
of these interactions and those involving the surfactants
A ,;. high-pH processes are as sensitive to brine salinity
and hardness as are MP processes.
Other high-pH rock/fluid interactions are intimately as-
sociated with the clay minerals. Clays are hydrous alu-
minum/silicate compounds that occupy the smallest (less
than 2 microns) particle size in typical media. Macroscop-
ically. clays occur as segregated streaks of variable degree
of continuity throughout a typical reservoir, or as distrib-
uted clays. which can line pore walls or fill pore throats.
Distributed clays are of most concern here, since these
have quite large surface areas (15 to 40 mig clay), and
therefore can exhibit considerable reactivity. 9X Chemi-
cally, clays can take on a variety of formulas that differ
substantially in their reactivity even though the differences
in their molecular formulas are apparently minor.
The ability of a clay mineral to exchange divalent cat-
ions with an aqueous solution can drastically change the
ionic environment of a solution with which it is in con-
tact. Clays have excess negative charges caused by the
substitution of +2-valence minerals for +3-valence min-
erals within the octahedral or tetrahedral crystal lattice. 99
The cation exchange capacity, Zv, is a measure of this
excess negative charge; typical Zvs are I to 10 meqi 100 g
clay for kaolinite and 100 to 180 meq/lOO g clay for mont-
morillonite. These free anionic sites are covered with cat-
ions from the solution, each of which has a specific degree
of selectivity for the particular clay site. In general, H +
has high clay selectivity, and divalent cations are bound
more strongly than are monovalents. This means that the
anionic sites can be occupied predominantly by H+
and/or divalents even when clays are in contact with rela-
tively soft brines. Any subsequent change in the electro-
lyte environment of the contacting solution can cause the
clays to take or give up these cations with a possible
detrimental effect on high-pH (and MP) flooding.
H cations can exchange on the clay sites with the in-
jected sodium according to
clay-HfNa Sclay-NafH. .(24)
where clay represents a mineral exchange site. I) The
rcverxiblc reaction Eq. 24 will clearly cause the H con-
centration to increase with a resulting pH decline. Fig.
47.29 shows the extent of the OH - retardation caused
by cation exchange in laboratory tloods. Note that many
of the lower pHs may require more than 3 PV of fluid
in-jection to attain the injected pH.
Unlike MP flooding. high-pH chemicals can react
directly with clay minerals and the silica substrate to cause
consumption of OH ~ ions. The reactions with clays are
manifest by the elution of soluble aluminum and silica spe-
cies from core displacements. The resulting soluble
species subsequently can cause precipitates through
hydroxyl reactions as in Eq. 24. lo2 The rate of hydrox-
yl consumption from this slow reaction (cation exchange
is generally fast enough so that local equilibrium applies)
Fig. 47.29-Effluent histories pH from laboratory corefloods. Ex-
perimental curves are solid lines with points, theo-
retical results are dashed lines.
is determined by a dimensionless DamkGhler number,
N,, =+k L/u,,., . . .(25)
where k is the reaction rate constant, time - , and L is
the medium length for a first-order reaction. ND, is the
ratio of the reaction rate to the bulk fluid rate. If all con-
ditions are equivalent between a laboratory experiment
and a prototype field flood, ND, clearly will be much
larger in the field than in the laboratory owing to the much
larger length scale. A larger ND, implies more reaction
relative to the residence time within the system. Thus,
it follows that the penetration distance-the distance
traveled by full-strength OH - ions-will be considera-
bly smaller in the field than in the laboratory. Bunge and
Radke, who illustrate this with several numerical calcu-
lations, caution against extrapolating laboratory-measured
values of OH - consumption to field cases unless the dis-
crepancies in ND, have been taken into account.
Field Results
High-pH field tests of articular interest include a wetta-
bility reversal test,
an emulsion flood,Y6 and a
polymer-driven flood. lo3 Fig. 47.30 shows the produc-
tion data from a high-pH flood conducted in the Whittier
field. 04 The crude oil was 20API with a 40-cp viscosi-
ty, and the 0.2 wt% NaOH chemical was injected as a
0.23.PV slug.
There are many features in these data that are common
to the responses of the other chemical flooding processes
in Figs. 47.6 and 47.22. The oil production rate declines
as the total fluid production increases, indicating a declin-
ing oil cut. The oil rate response to the caustic injection
is again superimposed on the waterflood decline, which
is extrapolated to estimate the IOR. (There are two water-
Hood decline curves in Fig. 47.28. one based on the ac-
tual decline and one based on computer simulation.) The
350,000 to 470,000 STB of oil produced by the caustic
injection was considered a success by the operators.
Table 47.3 shows a summary of data from complctcd
high-pH field floods. Note the wide range in reservoir
and oil characteristics and in oil saturation at the start of
the flood. IOR. expressed as a fraction of PV. ranges
from 0.0006 to 8.0. which translates into recoveries, ex-
pressed as a fraction of the OIP at the project start, that
p. at
Residual Residual
Reservoir Reservoir
Temperature Temperature
Field, Location, Operator
Bradford, PA
iCP, ioF)
(Several tests by several
Southeast Texas (Exxon)
Harrisburg, NE (Amoco)
75 112
1.5 200
Northward-Estes, TX
Singleton, NE (Sinclair)
Whittier, CA (Chevron)
Brea-Oltnda, CA (Umon) 90 135
Orcutt l-111, CA (Union) 17 to 60 140
arc comparable wjith but slightly smaller than those report-
ed from polymer floods (Table 47.1). Of equal impor-
tance is the STB of IOR produced per pound of chemical
in.jccted (0.015 to 0.43 in Table 47.3). This is substan-
tially lower than the polymer flood values; however. the
cost of the high-pH chemical is also substantially lower.
Fif. 47.3 I presents screening guides for the three major
chcmicul flooding processes presented in this chapter
bused on the work of Taber and Martin. I Though there
arc many other possible guidelines. the figure focuses on
three common reservoir parameters: oil viscosity at reser-
voir conditions. permeability. and depth. The guides are
intended as rules of thumb for candidate reservoir selcc-
tion and arc not substltutcs for detallcd reservoir eval-
Chemical flooding application usually is limited to
moderate to low oil viscosities because ofcconomics. As
the oil viscosity increases. more of the rcspcctibr: chem-
cal is rcquircd to attain food mobility control. This causes
a direct penality in chemical cost and an indirect penalty
in increased project life. Similar considerations place a
at Project
Net Water
Porosity Thickness Salinity Depth
Wd ut) mm TW (ft)
16 to 20
33 to 35
64 206 36 850 3.140
40 16
- - -
watered out
51 30 137 fresh 1,500
- -
50 22.5 155 15,000 2,200
lower limit on the reservoir permeability. although in the
case of polymers used in polymer or MP flooding there
is a technical limit imposed by the inability to propagate
large molecules through very small pore spaces. Deep
reservoirs usually imply the ability to apply larger sur-
face pressures with accompanying increase in in,jcction
rates. This beneficial effect is offset by the susceptibility
of polymers to chemical degradation at elevated temper-
atures. High-pH processes arc also more reactive at high
tcmpcrature, which causes excessive consumption.
The information in Fig. 47.3 I suggest the avenues of
future development for chemical tlooding EOR processes:
( I) the development of more cost-effective chemicals such
as surfactants and polymers that are more salinity resis-
tant. have minimal retention. can be manufactured onsite.
or can be recyolcd. (2) development of more efficient de-
sign procedures such as applying MP technology to high-
pH flooding design, (3) removing technical limitations on
the use of water-soluble polymers. particularly as they
relate to temperature sensitivity, and (4) mot-c reliable pre-
diction techniques, particularly as they relate to risk esti-
mates. Each general area will continue to prompt
considerable research so that the economic viability of
chemical flooding will become commonplace.
1363 1964 7363 1366 7367 7363 1969 1370 ,371 1372 1971
Fig. 47.30-ProductIon response from the Whittier field high-pH flood
Type of
(w KOW
2.4 3.2
IOW 2.0
Chemical bbl
Concentration Ibm Recovery, incremental
Injection Slug Size chemical Fraction Oil per Ibm
(WWO) (O/o VP) (bbllbbl VP) PV Chemical
A = pattern area
C,,, = low effective salinity limit
= high effective salinity limit
C, = concentration of surfactant, mass
surfactantimass solution
6, = retained surfactant concentration. mass
surfactantimass rock (includes all of the
previously mentioned forms of
z C
= effective shear rate
e I;5
= shear rate when polymer solution viscosity
is % of high and low values, Eq. 3
= mobility buffer efficiency
= mobility buffer efficiency extrapolated
j;,/, = oil bank oil fractional flow, fraction
f;,, = initial oil fractional flow, fraction
flIL = peak oil cut
FD = frontal advance lag. dimensionless
= solubilization ratios between the
microemulsionioleic phases
= solubilization ratios between the
microemulsioniaqueous phases
F,,: = permeability reduction factor
F, = resistance factor
FRI- = residual resistance factor
k,, = permeability to polymer solution
k,, = permeability to polymer-rich (water) phase
K = power-law coefficient
K IIp = Dykstra-Parsons coefficient
K,, = dissociation coefficient of water
M,, = effective mobility ratio
M* = shock mobility ratio
M" = endpoint mobility ratio
- - -
0.013 0.093 0.003 0.03
15 2.55 8.0 0.03
8 0.55 0.023 0.042
20 0.16 0 05 to 0.07 0.32 to 0 43
- - -
0.017 0.028 0.0006 0.015 to 0.030
Fig. 47.31-Comparative evaluation of reservoir parameters for
chemical flooding.
I, =
N, =
PI. =
P,q =
s ~
illi -
s,,i =
s I(,). =
0,II =
St, =
t ml/l =
t 'DO/~ =
t11\ =
t 'I).! =
I',, =
MB =
v,,, =
z,/ =
hi =
x(J =
A,, =
A,, =
X,,,, =
owl? =
tJ,l =
cc,J -
P,, -
CL\,. =
Pr =
P, =
,,l,i =
(-J,,rn =
u,ic, =
dimensionless time at complete sweepout
superficial flux of polymer-rich (water)
phase. Lit
superficial velocity of water phase
mobility buffer volume. fraction
MP slug size
cation exchange capacity
mobility of displaced fluid(s)
mobility of displacing agent(s)
mobility of polymer solution
mobility of water-rich phase
tnobility of water-rich phase after polymer
mobility of water-rich phase before
polymer injection
apparent viscosity of polymer solution
polymer solution viscosity below some low
shear rate
polymer solution viscosity above the
critical shear rate
viscosity of the displacing water
density of rock. mass/volume rock
density of surfactant slug, mass/volume
IFT between the microemulsion phase and
IFT between the microemulsion phase and
IFT between the water and oil phases
Scvcral useful discussions with R.S. Schechter and G.A.
Pope are gratefully acknowledged.
4. Dyer. A.B Caudlc. B.H.. and Ertchwn. R A : Oil Productnw
r,ltcr Brcnhthrqh ah lnllucnccd hy ,Mobdity Ratio: J. PC/. kc II.
(Apnl 195-l) 27-32: T~,o/i\. AIME. 201
power-law exponent
capillary number. dimensionless
left plait-point designation
right plait-point designation
oil bank saturation, fraction
initial oil saturation. fraction
ROS to an MP flood
ROS to a waterflood
saturation of polymer-rich (water) phase
oil bank arrival time, dimensionless
corrected oil bank arrival time,
surfactant arrival time, dimensionless
corrected surfactant arrival time,
5. ClawJge. E.L. Predictwn (II Recwcry III Unstable Misihlc
Floodmg. Sot. PC/. F,ig. J. (Aprd 1977) 143-SS.
6. Reznik. A.A. 1 Enick. R.M.. and Punvclkcr. S B.: An Analyt-
xal Extemion otthe Dykstra-Parsons Verttcal Stratificatton DIS-
mete Solution to a Contnwous, Real-Tmw Basis. .5x. Per. Gr,q.
J. (Dec. 19X4) 64-56.
7. Dykstra. H. and Parsons. R L.: The Prediction of 011 Recov-
cry by Watertlood: Secondary Recovery of 011 m the United
State\. Bull.. API. Dallas (1950).
8. Johnson, C.E. Jr : Prediction of Oil Rccwcry by Waler Flood-
A Simplilied Graphical Treatment of the Dyk<tra~Parwns
Method. J. PH. i&. (No\ 1956) 55-56: Tranh., AIME. 207.
9. Zapata. V.J.. The Effects of Viscous Cro.raflow on Sharp Front
Displacements in Two-Layered Porous Medta. MS the\lh, U.
ol Texah, Austin (1979).
IO. Hearn. C.L.: Simulation of Stratified Waterfooding by Pwu-
do Relative Permeability Curve\. J. Per. TK/I. (July 197 I )
I I. Dietr, D N : A Tlwrwcri~ol Approach 10 r/w Pr~hlrm o/
O~tn~rr~/~h~~ on Bdc~h~,y Erifir Wort~r. PUK. Ahad. van Weten-
whappen. Amsterdam (1953) 83-91
12. Buckley. S.E. and Lcverctt. M.S.: Mechanlvn rot FluId Di\-
placement m Sands. Tril,?\.. AlME 11942) 146. 107-16.
13. Needham. R.B.. Threlkeld. C.B.. and Gall, J W. Cwmol ot
Water Mobility Using Polymers and Multivalent Cation\. paper
SPE 4747 preented at the 1974 SPE Improwxi Oil Rectwcry Sym-
posium. Tulsa. OK. April 22-24.
14. Gash. B.H.. Griffith. T.D . and Ghan, A.F. Phaw Bchavwr
Etfccts on the Oil Displacement Mechamsms (11 Broad Equiis-
lent Weight Surlactant Systems. paper SPE 98 I2 prcww dl
the 1981 SPE Enhanced 011 Recovery Sympo\~um. Tula. OK.
April S%f.
IS. Wtllhite, G.P. and Domlnquez. J.G.: Mechanisms 01 Polymer
Retention in Porous Mcdiu. hprowd 01 Rr~o~vr) /?,I Surfuc
ran! tofd Polww Noocltriy. D.O. Shah and R.S. Schechter (cd\. ).
Academic Pres\ Inc., New Yorh City (lY77) 51 l-.55.
16. Wcllmgton. S.L.: Bwp,lymer Solutton Vtico\~ty Stabtliratton-
Polymer Degradation and Antioxidant Uw. Six.. PC,/. Etq J.
(Dec. 1983) 901-12.
17. Manmng. R.K., Pope. G.A.. and Lake. L.W.: A Technical Sw
vey of Polymer Flor,dtng Pro~ccta. Contract No.
DOEIBETCIl03?7-IO. U.S. DOE (Sept. lYX3).
18. Tsaur, K.: A Study ofPolymcr/Surfactant Interaction\ liw Micel-
lar/Polymer Floodlnp Application.. MS thwir. U. of Texaj.
Austin (1978).
IO. Martin. F.D.. Donaruma. L G . and Hatch. M J.: Develop-
mcnt ol Improved Mobllitg Control Agent\ ti,r SurtactantiPoly~
mtr Flooding. second annual report. Contract No
DOE/BC/OO047-13, U.S. DOE IOct. I9801.
20. Tinker. G.E Bowman. R.W . and Pope. G A.. Deternnnauon
of In-Smt Mobility and Wellhore Impairment From Polymer In-
jcctivity Data. .I. PH. Td. (May lY76) ~586-96.
21. Sacins. J.G.: Non-Newtonian Flow Through Porous Morct~-
al\. Ind. and Eng. Chcm I 1061)) 61. No, IO. 1X-47.
22. Meter. D.M., and Bard. R.B : Tube FIw (11. Non~Ncwn~ian
Polymer Solutions. Part\ I and 2-Lammar Flw and Rheulugl-
cd Mtxlels. A/C/,E ./. (No\. 1964) X7X-81 and 1143~50.
23. Duda. J.L.. Klaus. E.E.. and Fan. S.K.: lnllucnce oIPolynw
Molecule-W,dI Intcract~on~ on MoblIlt) Cunt~~~l. .5x. P<,r. Tory
J (Ott 1981) 613-Z
2-t. Jcnninp\. R.R., Rogers, J.H . and Wc>r. T.J Factorz Inilw
encing Moblllty Control By Polqmcr S~~lut~on\. J. /cr. TK/I.
(March 1971) 391-401. T/r,w.. AIME. 251.
7.5. Hlrasahi. G.J. and Pope. G.A : Analy\i\ of F&%x\ Intluencq
Mobility and Adwrption in the Flow OF P[,I)mcr Solution Throu:rh
Porow Media. Sor, Per. i?r,v. J. (Aug. 19741 337-46.
76. Shupc. R.D.: Chcmad Stahllity of Poly.srylamidc Polymct-\.
J. PC,/. Tdi. (Aug. I981 ) Ii 13-79.
27. Maerher. J.M.: Mechanlcul Defrnd;mon\ 01 Purtlnll) Hydra-
lyrcd Polyacrylatnidc Solutnw m Unc~n~wl~dated Porou\ ML-
dia. So. Per. k-q,. J. (Aug. lY76) 177-7-1.
2X. Sertght. R.S.: The Eltect\ ot Mechamcal Deqrdatton and Viwo~
&\IIC Behavior on In,jcctivq ~~lP~~I)acrj lam& Solutwn~. .Sfx
PC/. L-q. J. (June 19X3) J75-XS.
20. Clampttt. R.L. and Reid. T.R : An Econwmc Pol~murllwxJ
in the North But-hnnh Unit, 0+x Cwnt>. Ohlahotnn. papcl
SPE 5552 preenccd at the 1075 SPE Annual Technical ConI&-
cncc and Exhthitlon. Dallas. Scpt 7X-Oct. I.
3 I, Gractaa. A.. (I trl.: Criterta for Structurmg Surlactant~ to Max-
imILe Solubilizatton ot Oil and Water: Part I-Commerelal Non-
hllC~. SOCK. Pet. Or,y. 1. (Oct. 1982) 743-49.
32. Bar&at. Y. <f I// : . Crttcrta for Structuring Surtactant\ to Max-
ttntzc Solubilization of 011 and Water. J. Cdlortl Irrterfirw Sri.
(1983) 92, No. 2. X-74.
33. Ovcrbcek. .I. Th. G: Colloidh and Surface Chemistry. A Self-
Study SuhJect Guide. Part 2. Lyophobic Cotloids, 6lt//.. Center
for Advanced Engineermg Study and Dept. of Chemtcal Engi-
neering. Ma\sachusett\ Inst. of Technology. Camhtidge. MA
34. Patton. J.T.. (I ol. Enhanced Oil Recovery by CO z Foam
Flooding. tinal report, Contracr No DOEiMC/O325Y-15. U.S.
DOE (April 1982)
3.5. Khan. S.A.: -The Flow of Foam Through Porous Media. MS
thests. Stanford U.. Stanford. CA (1965)
36. Fried. A.N.: The Foam-Drive Procr\\ tnr Increasing the Recov-
cry d Oil. Report of Invcstigattons 5866. U.S. Dept. of the
lnlerior (IYM))
37 Bernard. G.G. and Helm. L.W.: Effect of Foam on Pcrmea-
hihty of Porous Media to Gas. Snc. Prr. En,is. J. (Sept. 1964)
267-72: Trtrric.. AIME. 231.
38. Bernard. G. G.. H&n. L.W.. and Jacob\. W.L.. Effect of Foam
on Trapped Gas Saturation and on Pcrmcahiltty of Porous Me-
dia to Water. SIX.. Per. &. J. (Dec. 1965) 2Y5-300: Trwts..
AIME. 234.
39. Holbrook. S.T.. Patton, J.T.. and Hhu, W.: Rheology ot
Mohtltty-Control Foam\, Srx. Per Oz,q J. (June 1983) 456-60
40. Hirasaki. G.J. and Lawon. J.B.: Mechanisms of Foam Flow
in Pornw Media: Apparent Viwosity m Smooth Captllartes.
So<.. Per. Gtg, J. (April 1985) 176-90.
41. Helm. L.W.: Foam InJectton Test m the Stggtns Field. Illrnot~.
J Pet. Td?. (Dec. 1970) 1499-1508.
42. Larwn. R.G.. Scrrven. L.E , and Davts. H.T.: Percolation Thc-
ory ot Residual Phases in Porous Media. N~~tr#re (lY77) 268,
43. Stcgemeter. G.L.: Mechanisms ot Entrapment and Mohtltra-
bon of Oil in Porous Media. hnprowtl Oil R~cowry hx Surfiw
1~111, and Polwler F/00&~. D.O. Shah and R.S. Schechter (eds )
Academic Press. New York City (1977) 55-93.
44. Mohanty, K K. and Salter. S.J.: Multiphase Flow in Porous
Media: Part 3-Oil Mobilization. Transverse Dispersion. and Wet-
tuhtlity, paper SPE 12127 presented at the 1983 SPE Annual
Technical Conference and Exhihition. San Francisco. Oct. S-8.
45. Larson. R.G.: From Molecules to Reservoirs. Problems in En-
hanced Oil Recovery. PhD dissertation. U. ofMtnnesota, Min-
neapoli\ (1980).
46. Catnilleri. D.: Micellar/Polymer Flooding Experiments and
Compartson with an Improved I-D Simulator. MS thesis. U.
of Texas, Austin (1983).
47. Helm. L.W.: Design. Performance and Evaluatton of the
Unitlood Micellar-Polymer Process-Bell Creek Field, paper
SPE I I196 prewntcd at the 1982 SPE Annual Techntcal Con-
ference and Exhibition. New Orleans. Sept. 26-29.
48. Goparty. W B Meabon. H.P.. and Milton, H.W. Jr.: Mobil-
ity Control Dc\ign for Miscible-Type Waterflood\ Using Mtcel-
Jar Solutions. J. Per. Tdz. (Feb. 1970) 141-47.
49. Wmsor. P. A, : .!Xwtr Propmrev rfAmphiphi/ic Contppo~rnrl.~. But-
terworth\. London (1954).
50. Healy. R.N.. Reed. R.L., and Stenmark. D.G.: Multiphase
Microemulston Systems. Sot. Per. Gtg. J. (June 1976) 147-60.
51. Nelw, R.C.. and Pope. G.A.: Phase Relationships tn Chem-
cat Floodtng. SM. PH. 01g. J. (Oct. 197X) 325-38.
52. Anderwn. D.F., et a/ : Interfacial Tension and Phase Behavior
tn Surfactant-Brine-Oil Systems. paper SPE SE I I. presented at
the 1976 SPE Symposium on Improved Oil Recovery. Tulsa.
March 22-24.
S3. Bennett, K.E., CI al.: Microemulsion Phase Bchavior-
Observations. Thermodynamic Essentials, Mathematical Sum-
lation. So<.. Per. &,g. J. (Dec. 1981) 747-62.
54. Scrtven. L.E.: Equilibrium Bi-Continuous Structures. M&l-
/i:atlon. So/uhi/&~tion, and Micrmvd~ion.~, K.L. Mittal (ed.).
Plenum Press. New York City (1976).
Puerto. M.C. and Reed. R.L Three~Paratnrter Representation
of Surfactant/Oil/Brme Interaction. So<~ Per. E/II+ J. (Aug.
1983) 669-82.
Caytas. J.L. (I nl.: Modelling Crude Oils for Lo\s Intcrfact,~l
Tension. Sock. Pet. Dl,g. J. (Dec. 1976) 3Sl&S7.
Nelson. R.C.: The Effect of Live Crude on Phase Behavior and
Oil-Recovery Effictcncy of Surfactant Floodmp Systems. Si~c,.
Per. G,,y. .I. (June 1983) 501-10.
Salter. S.J.: The lntlucnce 01 Type and Amount of Alcohol on
Surfactant-Oil-Brine Phase Behavior and Propertie\. paper SPE
6X43 pwented at the 1977 SPE Annual Technical Contcrence
and Exhtbttion. Denver, Oct. 9-12.
Hcaly. R.N. and Reed, R.L.: Physrcochcmical Aspect, 01
Microemulsion Floodmg. Sot Per. Eyq. J (Oct. 1974)
49l-SOI; Twos , AIME, 257.
Salter. S.J.: Opttmizing Surfactant Molccul;tr Weight Dt\tri-
bution: 1. Sulfonatc Phase Behavior and Phy+zll Properties.
paper SPE 12036 presented at the 1983 SPE Annual Technical
Conference and Exhibition. San Franctwo. Ott 5-X.
Trushenshi. S.P.: MicelIar Flooding: Sulfonate-Polymer Intcr-
uctton. . Ir~tpr~wrl Oil Rccow Y /J) SL~I$KI~IIZ~ r~utl Po/wwr F/ad
i/)x. D.O. Shah and R.S. Schcchter (cd\.). Academrc Press. Ncb
York City (1977) 555-75.
Salter. S.J.: Selection ofPheudo~Comp,nents in Surfactant-Oil-
Brine-Alcohol Systems. paper SPE 7056 prcscnted at the 1978
SPE Improved Oil Recovery Symposium. Tulsa. OK. Aprtl
Hirasaki. G.J.: Interpretation ofthe Change in Optimal Salmtty
with Overall Surfactant Concentration. Sot. Pcv. E/IX. J. (Dec.
1982) 971-82.
Clover. C.J. (I cl/.: Surfactant Phase Behavior and Retention
in Porous Media. Sot. Per. Eqq. J. (June 1979) 183-93.
Bourrel. M. et (I/.: Properttes of AmphiphileiOrliWater Sys-
tems at an Optrmum Formulation Ihr Phase Behavtor. paper SPE
7450 presented at the 1978 SPE Annual Tcchntcal Conterence
and Exhibition. Houston, Oct. l-4.
66. Nelson, R.C.: The Saltnity Rcqutrcment Dtagram-A Usetul
Tool in Chemical Flooding Research and Development. So<,.
PP(. 0)~. J. (April 1982) 259-70.
67. Hirasaki. G.J.. van Domwlaar, H.R.. and Nelson. R.C.: Eval-
uatton of the Salinity Gradient Concept in Surfactant Flooding.
Sk. Per. EHR. J. (June 1983) 486-500.
68. Cayias, J.L., Schechter. R.S , and Wade. W.H.: The Meas-
urement of Low lnterfnctal Tension via the Spinning Drop Tech-
nique. Ad,wrprion UI Intrrf~~ws, L.K. Mittal (ed.). Symposium
Series. ACS (1975) No. 8. 1231-39.
69. Huh, C.: Interfacial Tensions and Solubili,@ Ability of a
Microemulsion Phase that Coexists with Oil and Brine. .I Co/-
loid I~~~filcr Sci. (1979) 71. No. 2, 408~26.
70. Glinsmann, G.R.: Surfactant Flooding with Mtcroemulsions
Formed In-situ-Effect of011 Characteristtca. paper SPE 8326
presented at the 1979 SPE Annual Technical Conference and Ex-
hibition. Las Vegas. Sept. 23-26.
71. Reed, R.L. and Healy. R.N.: Some Physico-Chemical Aspects
of Mtcroemulsion Flooding, A Review. Irnprow~l 0ii RKUI~
cry hi Sur$mcrm md Po/wwr Flooding. D.O. Shah and R.S.
Schechter (eds.). Academic Press. New York City (1977)
72. Reed. R.L. and Healy. R.N : Contact Angles for Equtlihrated
Microemulsion Systems. Ser. Pet. Erq J. (June IY84) 342-50.
73. Bragg. J.R. et al.: Loudon Surfactant Flood Pilot Teht. paper
SPE 10862 presented at the 1982 SPE Enhanced Oil Recovery
Symposium, Tulsa, April 4-7.
74. Lake, L.W. and Helfferich. F.: Catton Exchange tn Chemical
Flooding Part 2-The Effect of Dtspersion, Catton Exchange.
and PolymerISurfactant Adsorpbon in Chemical Flood Environ-
ment. Sw. Per. Ot,e. J. (Dec. 1978) 435-44.
75. Pope. G.A.. Lake, L.W., and Helffertch. F G : Catton
Exchange in Chemical Floodmg, Part I-Basic Theory Without
Dtsperston. Sot. Per. f31,q. J. (Dec. 1978) 418-34.
76. Paul. G.W. and Froning. H.R.: Salinity Effects of Mtcellar
Flooding, J. Per. Tech. (Aup. 1973) 957-58.
77. Harwell, J.H.: Surfactant Adsorption and Chromatographic
Movement with Application in Enhanced Oil Recovery. PhD
dtssertation. U. of Texas, Austin (1983).
79 Hill. H J. and L&c. L.W.: Cation Exchanee ,n Chcnucal Flwtl-
lng: Part .i-Experimental. So< PC/. &q. J . (Dec. 1978)
X0. Paul. G.W. (I crl A Slmpllficd Predictive Model lor M~ccl-
IwPolymer Floodmg. paper SPE 1073.1 prewnted at the 1982
SPE Cal~fomia Resmnal Meeting. San Francwx~. March 24-26.
XI L&c. L.W.. Stock. L.G., and Lawson. J.B.: Screening E\ti-
mation of Recovery Efficiency and Chemical Reyulremcnt\ for
Chemical Fiotrding. paper SPE 706Y prtxnted at the 197X SPE
lmprovcd 011 Recover) Symposium. Tuls,~. OK. April l6- IY.
87 Aho. G.E and Bush. J.: Rc\ult\ olthe Bell Creek Unit A
MIcelIar Polymer Pilot. paper SPE I II95 prerentcd at the 1982
SPt Annual Technical Conference and Exhlhitwn. Ncu Orlcdn\.
Sept. 26-20.
Xi Lake. L W. and Pope. G.A. Statu5 of Mwllar-Polymer biekl
Tat\. Prr. G?q /n,/. (Nov. 1979) 51. 3X-60.
X4 ChIliland. H.E. and Conley. F.R.: Surrktanl Waterllwlinp.
paper presenccd xt the 1975 Sympwium on Hydnrarbon Eupk-
r&m. Drillq. and Productwn. Paris. France. Dee IO- I?.
85. Pope. G.A. The Apphcatmn of Fractlonul Flow Thcq to En-
hanccd Oil Rccorery. .%c PC,. Cr,q. ./. (June 1980, 191~205.
X6. R:lm;&ri\hnan. T.S. and Wassan. D.T : A Model for Inta%xd
Activity ol Au&c CNde Od/Caustlc System\ tar Alkdlinc Flood-
mg. S/K Per. hg. J . (Aug. 1983) 602m I?.
87. Jenning\. H.Y. Jr.. Johnxm. C.E. Jr and McAullffc. C.D.: A
Caustic Waterlloodlnp Procw for Heavy Oil\. /. PI,,. T&.
I Dee 1974) 1344-52
90 Nelson.R.C. er rrl.: .Courfact;nt-Enhanced Alhnlinc Flood-
me. paper SPE 12672 prcscntcd at the 1984 SPE Enhnnccd 011
R&very Sympowm. ?ulsa. April I5m I X.
91. Owens. W.W. and Archer. D.L.: The Effect of Rock Wetta-
hility on Oil-Water Relatwe Permeability Relationships, J . Per.
Twh. (July 1971) X73-78: kiwis.. AIME. 251.
floodmg tar Wettahility Alteration-Ev;llu;ltinF a Potential F&l
Application. J. Pcl. 7&. (Dec. 197-I) I33533.
Y4. Cooke. C.E. Jr.. Williams, R.E.. and Kolodz~e. P.A.: Oil
Recovery by Alkalme Waterflooding, J. Per. Twh. (Dec. 1974)
95. Lorenz. P.B.. Donaldson. EC. and Thornag. R.D: U\e ol
Centrifugal Measurcmentr of Wettahility to Predict Oil Rcco%-
cry. Bull. RI 7873. U.S. Dept. of Interior (lY741.
Y6. McAul~ffc, C.D.: Oil-in-Water Emulsions and Their Flow Prop-
erties in Porous Media. J . Pet. Tech. (June 1973) 727-33.
97. McAuliffe. C.D.: Crude-Oil-in-Water Emulww To Improve
Fluid Flaw in an 011 Reaervnir. J . Per. 7i~Ir (June 1973)
98. Grim. R.E.: C/ov Mir~ercrir~~qv, McGraw-Hill Booh Co. Inc.. Nrw
York City (1968)
99. Brownkw. A.H. : Gwr~/~er~~rrrrv. Prentice-Hall. Inc.. Englavood
Cliffs, NJ (1979).
100. Somerton. W.H. and Rndke. C.J.: Role of Clays in the En-
hanced Recovery of Petroleum from Some California Sands.
J. PC/. Ted. (March 1983) 643-54.
101. Bunpe. A.L. and Radkc. C.J : Migratmn of Alkalme Pulvx
in Raervoir Sands. Sot. I<>/. Org. J . (Dee 19x1) 9YX~IOIZ.
102. Sydansk. R.D.: Elevuted-Temperature Castic&mdstonc In-
teraction: Implications for Improving Oil Recovery. Sot,. PO/
E/q. J . (Aug. 1982) 453-62.
103. Sloat. B. and Zlomke. D.: The Isenhnur Unit-A Unique
Polymer-Augment Alkaline Flood, paper SPE I07 Ii) praent-
ed at the 1982 SPE Enhanced Oil Recovery Symposium. Tulsa,
OK. April 4-7.
104. Grwe. D.J. and Johnson. C.E. Jr.: Field Trial of Caustic Flood-
ing Process. .I. Per. TX/I. (Dec. 1974) 1353-.5X.
105. Mayer. E.H.. (1 (I/.: Alkaline InjectIon for Enhanced Oil
Recowx-A Status Report. J . PC/. T~I. (Jun. 1983) 209-21.
106. Taher. J.J. and Martin. F.D.: Tcchmcal Screening Guidcr for
Enhanced Oil Recovery paper SPE I2069 presented at the I983
SPE Annual Technical Conference and Erhihitlon. San Francisco.
Oct. s-x.
Chapter 48
Reservoir Simulation
K.H. Coats, Scientific Software-Intercom
Websters dictionary defines simulate as to assume the
appearance ofwithout the reality. Simulation of petroleum
reservoir performance refers to the construction and
operation of a model whose behavior assumes the ap-
pearance of actual reservoir behavior. The model itself
is either physical (for example, a laboratory sandpack)
or mathematical. A mathematical model is simply a set
of equations that, subject to certain assumptions, describes
the physical processes active in the reservoir. Although
the model itself obviously lacks the reality of the oil or
gas field, the behavior of a valid model simulates (assumes
the appearance of) that of the field.
The purpose of simulation is estimation of field per-
formance (e.g., oil recovery) under one or more produc-
ing schemes. Whereas the field can be produced only
once, at considerable expense, a model can be produced
or run many times at low expense over a short period of
time. Observation of model performance under different
producing conditions aids selection of an optimal set of
producing conditions for the reservoir.
The tools of reservoir simulation range from the in-
tuition and judgment of the engineer to complex
mathematical models requiring use of digital computers.
The question is not whether to simulate but, rather, which
tool or method to use. This chapter attempts to summarize
the evolution and current status of reservoir simulation
practice involving usage of the mathematical, computer-
ized models. The relatively modern nature of this prac-
tice is indicated by the first edition of this handbook (1962)
not including a chapter on reservoir simulation.
The nearly exponential growth in annual rate of
simulation-related publications from the mid-1960s to the
present indicates the industrys widespread acceptance of
mathematical simulation as an engineering tool. This ac-
ceptance has been and remains qualified by questioning
and improvement of accuracy in simulation model results.
Thus a significant portion of the extensive literature deals
with model (1) evaluation or validation through com-
parison of field (laboratory) and model results and (2) im-
provement by use of new techniques related to model
mathematics and representation of reservoir fluid and rock
description parameters.
The volume and increasing complexity of publications
related to the latter item preclude a detailed mathematical
description of current simulation technology in this
chapter. Rather, emphasis is given to a general descrip-
tion of reservoir simulation models, how and why they
are used, choice of different types of models for different
reservoir problems, and reliability of simulation results
in the face of model assumptions and uncertainty in reser-
voir fluid and rock description parameters. The chapter
concludes with an abbreviated description of simulation
model technology consisting of comments on a number
of highly technical publications. Various texts l-4 give
detailed descriptions of simulation technology through me
late 1970s, including finite-difference approximations,
model formulations, iterative solution techniques, and
stability analyses.
A Brief History
In a broad sense, reservoir simulation has been practiced
since the beginning of petroleum engineering in the
1930s. Before 1960, engineering calculations consisted
largely of analytical methods, $ zero-dimensional
material balances, 7,8 and one-dimensional (1 D) Buckley-
Leverett9,10 calculations.
The term simulation became common in the early
1960s, as predictive methods evolved into relatively
sophisticated computer programs. These programs
represented a major advancement because they allowed
solution of large sets of finite-difference equations describ-
ing two- and three-dimensional (2D and 3D), transient,
multiphase flow in heterogeneous porous media. This ad-
vancement was made possible by the rapid evolution of
ii y
Fig. 48.1-l-, 2-, and 3D grids.
large-scale, high-speed digital computer. and development
of numerical mathematical methods for solving large
systems of finite-difference equations.
During the 1960s, reservoir simulation efforts were
devoted largely to two-phase gas/water and three-phase
blackail reservoir problems. Recovery methods simulated
were limited essentially to depletion or pressure
maintenance. It was possible to develop a single simula-
tion model capable of addressing most reservoir problems
encountered. This concept of a single, general model
always has appealed to operating companies because it
significantly reduces the cost of training and usage and,
potentially, the cost of model development and
During tbe 1970s, the picture changed markedly. The
sharp rise in oil prices and governmental trends toward
deregulation and partial funding of field projects led to
a proliferation of enhanced-recovery processes. This led
to simulation of processes that extended beyond conven-
tional depletion and pressure maintenance to miscible
flooding, chemical flooding, CO2 injection, steam or hot-
water stimulation/flooding, and in-situ combustion. A
relatively comfortable understanding of two-component
(gas and oil) hydrocarbon behavior in simple immiscible
flow was replaced by a struggle to unravel and
characterize the physics of oil displacement under the in-
fluence of temperature, chemical agents, and complex
multicomponent phase behavior. In addition to simple
multiphase flow in porous media, simulators had to reflect
chemical absorption and degradation, emulsifying and in-
terfacial tension (IFT) reduction effects, reaction kinetics,
and other thermal effects and complex equilibrium phase
behavior. This proliferation of recovery methods in the
1970s caused a departure from the single-model concept
as individual models were developed to represent each
of these new recovery schemes.
Research during the 1970s resulted in many signifi-
cant advances in simulation model formulations and
numerical solution methods. These advances allowed
simulation of more complex recovery processes and/or
reduced computing costs through increased stability of the
formulations and efficiency of the numerical solution
General Description of Simulation Models
A number of papers-l4 present general, largely non-
mathematical discussions of reservoir simulation. Odeh
gives an excellent description of the conceptual simplici-
ty of a simulation model. He illustrates the subdivision
of a reservoir into a 2- or 3D network of gridblocks and
then shows that the simulation model equations are
basically the familiar volumetric material balance
equation7y8 written for each phase for each gridblock.
The phase flow rates between each gridblock and its two,
four, or six (in lD, 2D, or 3D cases, respectively) adja-
cent blocks are represented by Darcys law modified by
the relative permeability concept. Fig. 48.1 illustrates l-,
2-, and 3D grids representing a portion of a reservoir.
The block and its two or four neighbors are denoted by
B and N in the 1D and 2D grids. One can visualize an
interior block of the 3D grid with its six neighbors, two
on either side of the block in the n, y, and z directions.
The subsea depths to the top surface of each grid in Fig.
48.1 vary with areal position, reflecting reservoir forma-
tion dip. Reservoir properties such as permeability and
porosity, and fluid properties such as pressure,
temperature, and composition, are assumed uniform
throughout a given gridblock. However, reservoir and
fluid properties vary from one block to another; fluid
properties for each gridblock also vary with time during
the simulation period.
A simulation model is a set of partial-difference equa-
tions requiring numerical solution as opposed to a set of
partial differential equations amenable to analytical solu-
tion. Tbe reasons for this are (1) reservoir heterogeneity-
variable permeability and porosity and irregular geometry,
(2) nonlinearity of relative permeability and capillary
pressure vs. saturation relationships, and (3) nonlinearity
of fluid PVT properties as functions of pressure, composi-
tion, and temperature. The models require high-speed
digital computers because of the large amount of
arithmetic associated with the solutions.
The large amount of arithmetic performed by a simula-
tion model stems from the large number of gridblocks
representing the reservoir and from the number and com-
plexity of equations describing the oil-recovery process.
Total arithmetic or computing expense for a given model
run is at least linearly proportional to the total number
of gridblocks, N,N,N,, where N,, NY, and N, are the
numbers of gridblocks specified in the X, y and z direc-
tions , respectively.
The individual gridblocks are customarily identified by
subscripts i, j, k, where blocks are numbered i = 1,2.. .N,
in the x direction, j=1,2...N, in the y direction, and
k= 1,2.. .N, in the z direction. Most simulators use no-
flow or closed boundary conditions at the exterior bound-
aries [x=(O,LX), y=(O,L,) and z=(O,L,)] with provision
for aquifer influx along the areally exterior boundary. I he
nonrectangular, areal (x-y) shapes of most reservoirs are
represented by zero gridblock porosity and permeability
in the appropriate area1 portions of the x-y grid.
Preceding statements described the simulation model
as a set of equations expressing conservation of mass for
each phase for each gridblock. More precisely, the model
equations express conservation of mass of each reservoir
fluid component for each block, The number and identi-
ty of these components depend on the nature of the original
reservoir fluid and the particular oil-recovery process, as
discussed in the following. The total number of mass con-
servation equations is then N,N,N,N, where N is the
number of components necessary to describe the reser-
voir fluids.
Each conservation equation states that the mass rate of
flow into a gridblock minus the mass rate of flow out must
equal the rate of change or accumulation of mass within
the block. These N mass-balance equations (one for each
component) apply to each gridblock. The block is an open
system, in the thermodynamic sense, because of fluid flow
between the block and its six neighbors and fluid injec-
tion or production if a well is perforated in the block.
The center of gridblock (i,j,k) is located at (xi.yj,zk).
This block has six neighboring blocks (i-tl,j,k),(i,j-t 1,k)
and (i,j,kk 1). For brevity and clarity, the interblock flow
rates are written here in terms of only r-direction flow
between blocks (i- 1 ,j,k) and (i,j,k), the indices j and k
are suppressed, and the general symbol C, denotes con-
centration (mass/volume) of component I in the various
phases. The three immiscible phases (water, oil, and gas)
are denoted by subscripts w, o, and g, respectively.
The interblock flow rate of component I, according to
Darcys law modified by relative permeability, is
+%l,(ApO -y,AZ)+&@, -ysAZ) ,
. . . . . . . . . . . . . . . . . . . . . . . . . . . .
41 =
rP =
pp =
IP =
YP =
Ax =
component I interblock flow rate,
absolute permeability,
AyiAZk =cross-sectional area normal to
distance between adjacent block centers,
(AX-1 fAri)/
relative permeability to phase P
viscosity of phase P
concentration of component I in phase P,
pressure of phase P
specific weight of phase P
X,-l -x,,
where x is p or Z, and
subsea depth, measured positively
If subscript J=1,2,3 is used to denote phases w,o,g
respectively, then Eq. 1 simplifies to
The first term in parentheses is the interblock trunsmissi-
b&y, TIJ, for flow of component I in phase J, requir-
ing evaluation here at (i- %,j,k)-i.e., between blocks
i - 1 and i. The M/L portion of T is normally calculated
as the harmonic or series-resistance mean value using
block i- 1 and block i properties. The remaining portion
of T normally is evaluated at the upstream gridblock-
i.e., the block from which the phase is flowing. Thus Eq.
2 becomes simply
representing interblock flow of component I from
gridblock i- 1 to gridblock i.
The right-hand or accumulation terms of the mass
balances are
d c (SJCIJ) , . . . . . . . . . . . . . . , . .
V = grid block volume, hxi AYj AZk
6 = time difference operator, 6X=X,,+, -X,,
n = time level, t,+l =t,+At,
At = timestep,
$I = porosity, fraction, and
SJ = saturation of phase J, fraction of pore
Eqs. 3 and 4 give the final form of the component I
mass-balance equation for gridblock (i,j,k) as
C A[TIJ(APj-r./Az)l-4p~
where q,,I is the mass rate of production of component
I from the block resulting from any well perforated in the
block and the Laplacian term of type A(TAp) is defined as
For the general case where each component is present
(soluble) in all three phases, Eq. 5 is Nequations in 3N+6
unknowns. The unknowns are 3N CIJ values, 3 phase
saturations, and 3 phase pressures. Thus an additional
2N-t 6 equations are required for a determinate or solvable
model having equal numbers of equations and unknowns. a
The N Eqs. 5 are referred to as primary equations while
the additional 2N+6 equations are denoted constraint
equations. The constraint equations are manipulated in the
model programming to eliminate 2N+6 variables
(unknowns) in terms of the remaining N (primary)
unknowns. The result is then the set of N primary Eqs.
5 in N primary unknowns. The constraint equations are
relations between unknowns pertaining only to the par-
ticular gridblock (iJ,k) to which Eqs. 5 apply. The N
primary Eqs. 5, however, involve unknowns (e.g., pi)
at the gridblock (i,j,k) and its six neighboring blocks, ow-
ing to the nature of the interblock flow terms on the left-
hand side.
The 2N+6 constraint equations are illustrated here for
the case of an isothermal, compositional model where the
N components are HZ0 and N- 1 hydrocarbon com-
ponents (e.g., methane, ethane.. .C ,). The first three con-
straints are
S,+S,+S,=l.O, . . . . . . . . . ..(6)
po -pw =Poy,(S,), . . . . . . . . . . . . (7)
pg -p. =P,,(S,), . . . . . . . . . . . . . . (8)
where P,, =water/oil capillary pressure and P,, =gasJ
oil capillary pressure.
These constraints express the requirement that the phase
saturations sum to unity and also eliminate the water and
gas phase pressures in terms of the unknown oil pressure
phase using capillary pressure curves. For this composi-
tional case, concentration C,J =p J x/J where pi is the mo-
lar density of phase J (moMvolume) and xIJ is mol
fraction of component I in phase J. The next three con-
straints require that the mol fractions of all components
sum to unity in each of the three phases,
c x[J=l.o . . . . . . . . . . . . , , . . . . .
where J=w,o,g or 1,2,3. The remaining 2N constraints
express equilibrium of each component among the three
j-r, =fr, . . . . . . . . . . . . .
j - i , =f r , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
( l ob)
where f I J is the fugacity of component I in phase J. These
fugacities can be expressed in terms of mol fractions and
pressure by use of an equation of state (EOS). Altemative-
ly, they can be replaced by equilibrium K-value relation-
ships (e.g., y=Kx) with K-values given as functions of
pressure or of pressure and composition.
The 2N + 6 constraint Eqs .6 through 10 are manipulated
to eliminate one-phase saturation, two-phase pressures and
2N-3 mole fractions (x,J) from the N primary Eqs. 5.
The final result is a model consisting of the N Eq. 5 in
N unknowns consisting of two saturations, one pressure,
and N-3 mole fractions. Each coeffkient or term remain-
ing in the N primary equations is then either one of the
N primary unknowns or a function of one or more of the
primary unknowns.
Types of Models
Different types of simulation models are used to describe
different mechanisms associated with different oil-
recovery processes. The most widely used types are black
oil, compositional, thermal, and chemical flood. The four
basic recovery mechanisms for recovering oil from reser-
voirs are (1) fluid expansion, (2) displacement, (3) gravity
drainage, and (4) capillary imbibition. Simple fluid ex-
pansion with pressure decline results in oil expulsion from
and subsequent flow through the porous matrix. Oil is
displaced by gas and injected or naturally encroaching
water. Gravity drainage, caused by positive (water/oil and
oil/gas) density differences, aids oil recovery by causing
upward drainage of oil from below an advancing bottom-
water drive and downward drainage from above a declin-
ing gas/oil contact. Finally, imbibition, generally normal
to the flow direction, can be an important recovery
mechanism in lateral waterfloods in heterogeneous sands
with large vertical variation of permeability.
Accommodation of compositional and the enhanced-
recovery processes in this discussion requires the addi-
tion of a fifth mechanism, oil mobilization. This loosely
defined term includes widely differing phenomena that
create or mobilize recoverable oil. Some of these
phenomena are not really distinct from the first four.
The black-oil model accounts for the four basic
mechanisms in simulation of oil recovery by natural deple-
tion or pressure maintenance (e.g., waterflooding). This
isothermal model applies to reservoirs containing immisci-
ble water, oil, and gas phases with a simple pressure-
dependent solubility of the gas component in the oil phase.
The two-component representation of the hydrocarbon
content I5 presumes constant (pressure-independent) com-
positions of the oil component and the gas component,
no volatility of the oil component in the gas phase, no
solubility of the oil and gas components in the water phase,
and no volatility of water (H20) in the oil and gas phases.
The oil component is stock-tank oil and its unit of mass
is 1 STB (1 bbl at stock-tank pressure and temperature).
The gas component is surface system gas and its unit of
mass is 1 standard cubic foot (scf). The water component
unit of mass is 1 STB. For water and gas, components
and phases are identical while the oil phase is a mixture
of the oil component and the gas component.
The number of components (N) and therefore the
number of Eqs. 5 per gridblock is three for the black-oil
model. Table 48.1 gives the definitions of component con-
centrations, CIJ, for this model. The water phase, gas
phase, and saturated oil phase, reciprocal formation
volume factors, b, (STB/RB), bR (scf/RB), and b,
(STB/RB), respectively, are given smgle-valued functions
of pressure. For undersaturated oil, b, is dependent on
pressure and solution gas (R, , scf/STB). As discussed
later original work in formulations6m8 led to a number
of papers 9m23 describing black-oil models during the
The remaining model types discussed here account for
some mobilization mechanisms in addition to the four
basic recovery mechanisms. The isothermal compositional
model represents reservoir fluids by N components, in-
cluding water and N- 1 hydrocarbon components.
Generally, but not necessarily, solubilities of water in the
oil and gas phases and of hydrocarbon components in the
water phase are considered negligible. For water, then,
the concentration in Eqs. 5 is as given in Table 48.1. The
hydrocarbon component I concentration CIJ is P J X IJ , as
mentioned earlier, for J=o,g or 2,3. Gas/oil phase
equilibrium and phase densities within each gridblock are
calculated using equilibrium K-values from pressure- and
composition-dependent correlations or, more recently,
from EOSs. 25-28 Unlike the black-oil model, the com-
positional model can represent the mobilization of oil by
outright (single-contact) or dynamic (multicontact)
miscibility, oil swelling and viscosity reduction by solu-
tion of an injected nonequilibrium gas (e.g., CO,), and
stripping or vaporization of an oils lighter ends by in-
jection of a dry gas. With one exception,29 recent
29-33 describing compositional models are based on
equilibrium K-values obtained from EOSs.
A thermal simulation model is a set of N conservation
equations, similar to Eq. 5, which expresses conserva-
tion of mass of H2 0 and N-2 hydrocarbon components
and conservation of energy. With energy designated as
component N, the last (I=N) of Eqs. 5 becomes the
energy balance upon addition of terms representing heat
conduction and overburden heat loss. An additional re-
quirement is the use of pJHJ for cNJ in the well and in-
terblock flow terms and p J UJ for CNJ in the right side
accumulation term. HJ and UJ are enthalpy and inter-
nal energy, respectively, energy/mole. If the in-situ com-
bustion capability is included then the mass conservation
equations include source (sink) terms represented by Ar-
rhenius reaction rate expressions for cracking and oxida-
tion of hydrocarbon components and the energy balance
includes heat of reaction terms. For the same number of
fluid components, a thermal model has one more (energy)
conservation equation than the compositional model and
one additional unknown, temperature T.
For steam-injection processes, thermal model com-
ponents are typically H 2 0, heavy (nonvolatile) and light
(solution gas or distillable) hydrocarbon components and
energy. For in-situ combustion studies, typical com-
ponents are HzO, heavy-oil component, a lighter
(distillable) oil component, solid coke, 02, CO*, N2,
and energy. Frequently CO2 and N2 are lumped as one
component to reduce computing expense. The steam tables
and/or an EOS are used to calculate liquid Hz0 (water
phase) properties and the Hz0 gas/water phase K-value
as functions of pressure and temperature. In most applica-
tions, Hz0 is assumed insoluble in the oil phase. In most
current models, the distribution of other (non-H20) com-
ponents among all phases is represented by user-provided
K-values dependent on only pressure and temperature.
Thermal simulators are applied to steam-injection or in-
situ combustion processes in heavy-oil reservoirs where
oil is mobilized primarily by (1) reduction of oil viscosi-
J =l J =2 J =3
I Component Water Oil Gas
- __ -
1 water bw 0 0
2 oil 0 bo 0
0 b, Rs b,
ty with increased temperature, (2) distillation of in-
termediate hydrocarbon components from the oil phase
to the more mobile gas phase, and (3) cracking of the oil
phase [usually above 500F (26OC)l with subsequent
distillation. Thermal models developed from 1965 to
198234-40 generally exhibit a trend toward inclusion of
more dimensions, more components and dual capability
of steamflood and in-situ combustion.
Chemical flood models include polymer, micellar (sur-
factant), and alkaline (caustic). Polymer waterflooding im-
proves oil recovery by lowering the oil/water mobility
ratio, by reducing the effective permeability to water,
and/or by increasing water viscosity. In micellar flooding,
surfactants greatly reduce oil/water IFT, thereby solubiliz-
ing oil into the micelles and forming an oil bank. 4 The
surfactant slug and mobilized oil normally are propelled
toward the production well by a graded bank of polymer-
thickened water. The mechanisms responsible for im-
proved oil recovery in alkaline flooding are thought to
include low IFT, wettability alteration, and emulsifica-
tion. 42 Chemical flooding processes involve complicated
fluid/fluid and rock/fluid interactions such as adsorption,
ion exchange, viscous shear, and three- (or more) phase
flow. Several recent papers 43-45 describe implementation
of these complex chemical flood mechanisms in numerical
The four types of models described above are defined
or distinguished by the recovery process and the nature
of the original reservoir fluid. Considering the nature of
the reservoir formation leads to a fifth, fractured-matrix
type of simulation model. While in theory any recovery
process can be implemented in a fractured-matrix reser-
voir, most simulation work reported to date is concerned
with black oil fracture&matrix models. Three-dimensional
models are described by Thomas et a1.6 for the three-
phase case and by Gilman and Kazemi4 for two-phase
water-oil flow. Their models consider a discontinuous ar-
ray of matrix blocks in a continuous 3D fracture network.
Flow throughout the reservoir and to the wells occurs in
the fracture system and the matrix blocks are treated as
sink/source terms in that system. Their model equations
include the set of N conservation Eqs. 6 written for each
gridblock in the fracture system. Each gridblock may con-
tain a number of similarly behaving matrix blocks.
However, additional terms are added to Eqs. 6, represent-
ing matrix-fracture flow. Also, for each gridblock addi-
tional equations are required to express mass conservation
of each component in the matrix blocks included in the
gridblock. These additional equations can be eliminated
or combined with the basic N (fracture system) flow
equations4v4 so that the final model includes only N
equations (per block) possessing interblock flow terms.
Blaskovich et a1.48 describe a fractured-matrix model
which allows for reservoir-wide flow through the matrix
as well as the fracture system. This extension leads to a
model including 2N equations (per block) possessing in-
terblock flow terms.
Model Input Data and Calculated Results
A simulation model requires three types of input data.
First, reservoir description data include (1) overall
geometry, (2) grid size specification, (3) permeability,
porosity, and elevation for each gridblock, and (4) relative
permeability and capillary pressure vs. saturation func-
tions or tables. Geological and petrophysical work, which
involves logs and core analyses, is necessary for Items
1 and 3. Laboratory tests on core samples yield estimates
of relative permeability and capillary pressure relation-
ships. Second, fluid PVT properties, such as formation
volume factors, solution gas or component equilibrium
K-values, and viscosities are obtained by laboratory tests.
Finally, well locations, perforated intervals, and produc-
tivity indices (PIs) must be specified. Each well must be
assigned a production (injection) rate schedule and/or a
limiting producing (injecting) pressure for use in
calculating well deliverability (injectivity).
Model output or calculated results include spatial
distributions of fluid pressure, saturations and composi-
tions, and producing GOR and WOR and injection/pro-
duction rate (for wells on injectivitylproductivity) for each
well at the end of each timestep of the computations. In-
ternal manipulation of these results gives average reser-
voir pressure and instantaneous rates and cumulative
injection/production of oil, gas, and water by well and
total field vs. time.
Current models offer various levels of visual output
display features that ease the engineers assimilation and
interpretation of simulator results. Example features are
contour maps of pressure, saturations, compositions and
temperature, concise tabular summaries of individual well
or well-group performance, and field or well timeplots
of quantities such as production rates and WORs and
Purpose of Reservoir Simulation
Reservoir simulation is used to estimate recovery for a
given existing producing scheme (forecasting) to evaluate
the effects on recovery of altered operating conditions,
and to compare economics of different recovery methods.
Black-oil models have been widely applied to forecast oil
recovery and to estimate the effects on oil recovery of
(1) well pattern and spacing, (2) well completion inter-
vals, (3) gas and/or water coning as a function of rate,
(4) producing rate, (5) augmenting a natural water drive
by water injection and desirability of flank or peripheral
as opposed to pattern waterflooding, (6) inlill drilling, and
(7) gas vs. water vs. water-alternating-gas (WAG) in-
A few of many reported studies are briefly mentioned
here. Henderson et al.49 applied a single- (gas) phase
model to optimize the locations and numbers of wells
necessary to meet peak deliverability requirements in a
gas storage field. Mann and Johnson showed good
agreement between model-predicted and actual field per-
formance. Thomas and Driscol15 applied a black-oil
model in estimating locations of bypassed oil for the
pose of designing an infdl drilling plan. Two studies 2,53
report extensive results related to rate-sensitivity of
various Alberta reservoirs subjected to water-oil
displacements. Thakur ef al. 54 applied a black-oil model
to characterize (through history matching) offshore
Nigerian reservoirs and estimate incremental recovery of
waterflooding over natural depletion, with infill drilling
and removal of allowable rates.
Compositional models also are used for most of Pur-
poses 1 through 7 listed previously, but only in cases
where the black-oil assumption of constant composition
oil and gas components is invalid. Example compositional
model applications include (1) depletion of a volatile oil
or gas condensate reservoir where hydrocarbon phase
compositions and properties vary significantly with
pressure below bubble- or dewpoint, (2) injection of non-
equilibrium gas (dry or enriched) into an oil reservoir to
mobilize oil by vaporization into the more mobile gas
phase or by attainment of outright (single-contact) or
dynamic (multicontact) miscibility, and (3) injection of
CO2 into an oil reservoir to mobilize oil by stripping of
light ends, oil viscosity reduction, and oil swelling.
Compositional simulation has been performed to
estimate (1) loss of recovery caused by liquid dropout dur-
ing depletion of retrograde gas condensate reservoirs and
the reduction of this loss by full or partial cycling (rein-
jection of gas from surface facilities) and (2) effects of
pressure level, injected gas composition, and CO2 or N2
injection on oil recovery by vaporization or miscibility.
Graue and Zana55 describe application of a compositional
model in estimating Rangely (CO) field oil recovery by
CO2 injection as a function of injected composition and
pressure level. Results of compositional simulation of a
CO2 project include CO;! breakthrough time and rate and
composition of produced fluids. These are required to
design production facilities and CO2 recycling strate-
gies. 56 Modeling is also useful to optimize pattern size
and CO2 /water-injection rates to overcome the effects of
reservoir heterogeneity. 57
Thermal models are applied in reservoir studies of in-
situ combustion and are used to simulate performance of
cyclic steam simulation and steamflooding. In steam in-
jection, questions addressed by simulation relate to effects
of injected steam quality and injection rate, operating
pressure level, and inclusion of gas with the injected
steam. One question in cyclic stimulation concerns the
optimal time periods per cycle for steam injection, soak,
and production. The flooding case introduces the issues
of well pattern and spacing. A number of steam-injection
field studies using models have been published. Herrera
and HanzlikS8 compare field data and model results for
a cyclic stimulation operation. Williams59 discusses field
performance and model results for stimulation and
flooding, and Meldaum discusses field and model results
related to addition of gas to the inected steam. Gomaa
et al. 6 and Moughamian et al. 61 applied steamflood
simulation in identifying and optimizing operating
parameters in pilot and field drive operations.
Numerical simulation has been used to estimate
chemical flood performance in a reservoir environment
where the processes are very complex and many reser-
voir parameters affect the results. Chemical flood simula-
tion has been used to construct a screening algorithm for
the selection of reservoirs suitable for micellar/polymer
flooding63 and to examine competing EOR strategies-
e.g., CO* vs. surfactant flooding.@ For caustic65 and
polymeP applications, as well as for the micellar proc-
ess, chemical flood modeling is useful to discern control-
ling process mechanisms and to identify laboratory data
required for process description.
In recent years, simulation has been used increasingly
to estimate and compare recoveries from a given reser-
voir under alternative enhanced-recovery processes, such
as CO2 injection, thermal methods (steam injection and
in-situ combustion), and several types of chemical
Considerations in Practical Application of
Simulation Models
This section describes the procedure followed and cer-
tain questions faced by the engineer conducting a reser-
voir simulation study. The engineer must select the
appropriate type of simulation model, select the grid net-
work, and specify rock and fluid description data. Then
the engineer must attempt to reduce or at least estimate
inaccuracies in simulation results which stem from uncer-
tain rock/fluid description data and from spatial trunca-
tion error.
Selection of Model Type
As mentioned earlier, selection of model type may de-
pend on both the nature of the original reservoir fluid and
upon the recovery process(es) to be studied. As a rough
guide, an original reservoir oil solution gas or GOR value,
R,, below 2,000 scf/STB indicates a black oil, whereas
a higher value indicates a volatile oil or retrograde gas
condensate requiring compositional treatment.
For a black-oil reservoir, a black-oil model may be used
to study natural depletion, water injection and/or
equilibrium gas injection operations. However, a com-
positional model is generally necessary to estimate
recovery by injection of dry or enriched gas, solvent, or
CO*. An exception here is the applicability of a modified
black-oil model67 in simulation of CO2 or solvent injec-
tion where outright (single-contact) miscibility occurs.
Compositional simulation is generally employed for
volatile oil reservoirs. However, a less expensive black-
oil model study is adequate for simulating above-
bubblepoint waterflooding performance. Compositional
models have generally been employed to study retrograde
gas condensate reservoirs. The compositional model is
necessary in the case of below-dewpoint cycling.
However, in some cases natural depletion or above-
dewpoint cycling can be simulated at less expense using
a black-oil model modified to account for volatility of the
oil (condensate) component in the gas phase. 6sm70
The alkaline and surfactant-flood processes generally
require use of the complex chemical flood simulation
model. However, some (augmented) black-oil models of-
fer the capability to simulate polymer and alkaline
Selection of Model Grid
Selection of the x-y-z gridblock network involves many
factors, including available budget and the engineers
judgment and experience. For any type of model, the
arithmetic or computing expense per timestep is at least
linearly proportional to the total number of gridblocks
employed. The computing expense of a single model run
is proportional to the product of the number of gridblocks
and the number of timesteps required by the model to
cover the total time period of interest. In many cases, the
timestep size is controlled by the maximum rate of change
(overall gridblocks) in one or more calculated quantities
such as pressure and saturations. This maximum rate of
change generally occurs at or near a well or in the vicini-
ty of a flood front. A doubling of the number of gridblocks
can result approximately in a doubling of this maximum
rate of change since each gridblock is (on the average)
one-half as large. The average timestep size, then, might
decrease by a factor of two if the number of blocks were
doubled. The final result is a computing expense per
model run which can approach a proportionality to the
square of the total number of gridblocks. This indicates
the importance of selecting the smallest number of
gridblocks consistent with reservoir/well description,
recovery process characteristics, and the questions asked
regarding reservoir performance.
The number of gridblocks and resultant study computing
expense are the lowest in cases where the engineer can
justify use of a representative element of the total field
as the basis for the model study. This may be possible
in reservoirs developed with repeated well patterns, for
any recovery process-waterflooding, CO* injection,
steamflooding, etc. In such cases, the representative ele-
ment ideally should be a symmetrical element of the reser-
voir. In strict terms, this requires (1) a repeated, regular
pattern of identically completed and operated wells, (2)
a horizontal, areally homogeneous reservoir formation of
uniform thickness, and (3) areally uniform initial fluid
saturation distributions. If these conditions were met, then
questions regarding total field optimization, forecasting
and comparative evaluation of recovery processes could
be addressed inexpensively by simulation of the single pat-
tern (element).
While actual reservoirs never satisfy these conditions
exactly, representative-element simulation studies are fre-
quently performed for repeated pattern processes. In some
cases, a substantial portion of the reservoir may exhibit
only moderate areal heterogeneity and thickness variation.
Resultant variation in performance from one pattern to
another may be sufficiently small for engineering purposes
to justify scale-up of single-pattern results to total field
Representative-element simulation is often performed
where the study purpose is comparative evaluation of
alternative recovery processes as opposed to forecasting
of total field performance for a specific process and
operating scheme. The justification of single-element
simulation implied in such cases is that the resultant rank-
ing of alternative processes is unaffected by the variations
in pattern (element) properties over the field. This
justification can be and frequently is checked by repeating
the various process simulations for two or more patterns
of different properties representative of different portions
of the reservoir.
Finally, the relatively inexpensive single-element
simulation applies to design or optimization studies of a
specific recovery process operated in a repeated pattern
mode. For a repeated-pattern steamflood, single-pattern
model runs have been performed to optimize pattern
type (e.g., five-, seven- or nine-spot) and size, injected
steam quality and rate, well completions, etc. Occasional
publications describe single-pattern simulation studies us-
ing a one-quarter five-spot or one-quarter nine-spot as the
symmetrical element of the respective pattern. Actually,
a one-eighth five-spot or nine-spot (and xZ seven-spot)
are the smallest symmetrical elements and should be used
to minimize computing expense. 71
Currently, a major portion of the industry-wide effort
and computing expense in simulation studies is associated
with total-field forecasting of black-oil reservoir perfor-
mance under a sequence of recovery processes. Typical-
ly, the engineer must select a 3D grid for a large reservoir
with significant heterogeneity, large areal variation in dip
and thickness, irregular well locations and increasing
numbers of wells with successive development stages. The
engineer may face a several- to many-year period of
historical performance under natural depletion, frequently
with some natural water encroachment. Study objectives
may include history matching, followed by matching and
forecasting for a waterflood period, in turn followed by
forecasting for some tertiary scheme such as CO2 in-
The total number of gridblocks is the product of the
number of areal blocks, N,N,, and the number of grid
layers, N,. Different considerations enter into selection
of these two numbers of spacings. Factors indicating a
need for fine area1 grid spacing are high well density and
sharp or rapid changes (areally) in permeability, porosi-
ty, thickness, and dip. Since these factors frequently vary
over the field, thex- and y-direction grid spacings are often
nonuniform. Grid spacings generally increase toward the
downdip reservoir boundaries and increase greatly with
distance into the aquifer if the latter is present and included
in the grid.
In general, of course, the number of area1 gridblocks
required increases with the size of the reservoir and the
number of wells. However, grid spacings ranging from
very fine to very coarse may be appropriate for different
reservoirs of comparable size. The smallest numbers of
areal blocks (coarsest areal spacings) are associated with
reservoir studies limited to natural depletion and crestal
or flank gas and/or water injection. In such a case, a
coarse grid may result in a number of area1 blocks that
include two or more similar type (e.g., production) wells,
with little loss in engineering significance of the simulator
results. Large numbers of area1 blocks may be required
in cases of pattern waterfloods or enhanced recovery proc-
esses. A rough guide in this case is the need for at least
two, preferably three or more, gridblocks separating each
injection-production well pair. However, recent studies
describe estimation of pseudorelative-permeability curves,
which allow adjacent-block placement of an injectoripro-
ducer well pair. 72,73
The major factors affecting the number of grid layers
(vertical gridblocks) required are the formation stratifica-
tion, vertical communication, and total thickness. Many
reservoirs possess a number of formation layers, which
correlate from well to well over much of or all the field.
Variations of layer thickness, permeability, and porosity
may be significant areally and even greater from one layer
to another. The vertical communication (vertical
permeability) between adjacent layer-pairs may vary from
zero to very high, both areally and from one layer-pair
to another. In general, at least one grid layer should be
used for each correlatable formation layer. However,
common sense and budget constraints argue against detini-
tion of a large number of very thin grid layers. Three-
dimensional reservoir studies typically employ 4 to 12 grid
layers, and one or more of these grid layers may be a
lumped representation of several thin formation layers.
The need for subdivision of one formation layer into
two or more grid layers depends on the layer thickness
and fluid-segregation characteristics of the recovery proc-
ess and operating rates. Most recovery processes result
in moderate to severe gravity segregation of oil and in-
jected fluids; injected water or gas tend to underrun or
override oil, respectively; many steamflood projects ex-
hibit severe override of oil by the steam. A formation layer
that has significant thickness and zero to poor vertical
communication with layers above and below may exhibit
a pronounced phase segregation and require two or more
grid layers. In the idealized example of a fieldwide, pro-
nounced gravity override in a vertically homogeneous
reservoir, a variable grid spacing increasing from top to
bottom might be specified. That is, four layers of
thicknesses 5, 10,20, and 25 ft might give more accurate
results than four layers of equal 15ft thickness.
A customary approach to determining NZ involves use
of the simulation model itself in 2D cross-sectional (X-Z
slice) mode. For the particular recovery process of in-
terest, X-Z model runs are performed by using different
numbers of grid layers. Pseudorelative-permeability
curves reflecting phase segregation are calculated from
model runs performed with fine vertical grid spac-
ing. 74-76 These pseudocurves are then used in equivalent
x-z model runs using fewer grid layers to obtain coarse
(vertical) definition results similar to the fine-spacing
correct results. The fewer grid layers of the coarse
definition are then employed in the 3D reservoir study
grid. This concept of generating pseudocurves for coarse
vertical grids that reproduce vertical fine-grid results (us-
ing rock or laboratory relative permeabilities) has been
extended to the areal spacing problem, 72.75 as mentioned
Obviously, a minimum computing expense follows from
use of a single grid layer representing the entire forma-
tion thickness. This results in a 2D X-Y area1 grid as op-
posed to a 3D grid and occasionally is justified in the two
extremes of a very high vertical permeability and a layered
formation with zero vertical permeability. Pseudorelative
permeability and capillary pressure curves are discussed
for the former case in papers describing the vertical
equilibrium (VE) concept21,77 and for the latter case by
Hearn. 78
Specification of Reservoir Rock and
Fluid Description Data
Geological and petrophysical work based on logs and core
analyses yields maps of structure, net dh, and w1 products
for each of the several reservoir layers. The kh and +h
data often are augmented or modified by results of
drillstem, pressure buildup, and pulse tests. For each
layer, the engineer can overlay his area1 x-y grid spacing
network on these maps and read off the values of subsea
depth, I#& and kh at the center of each gridblock. These
values along with gross thickness of each block are then
transposed to a data file in a format compatible with that
required by the simulation model. Current research ef-
fort is directed toward developing computer programs that
accept digitized core analysis, log and geological data,
the selected grid network, and, through mapping and in-
terpolation techniques, automatically prepare the simula-
tion input data file.
Laboratory core analysis work includes measurement
of relative-permeability, k,., and capillary-pressure, P,,
curves for a number of field cores. Variations in rock
lithology may result in different sets of k, and P, curves
for different layers and/or different areal portions of the
reservoir. Most simulation models allow multiple sets of
such data in tabular form with assignment of each set to
a user-specified layer/portion of the reservoir. If the rock
water/oil (gas/oil) capillary pressure values are small, the
water/oil (gas/oil) transition zone in the reservoir may be
a very small fraction of total formation thickness. In such
cases, pseudocapillary-pressure curve(s) should be
used. l7
For black-oil studies, laboratory tests are performed to
determine gas compressibility factor and saturated oil and
gas viscosities vs. pressure. Differential and/or constant-
composition expansion tests on oil samples yield the
saturated oil pressure-dependent formation volume fac-
tor, B, (RB/STB), and solution gas, R, (scf/STB). The
resulting oil and associated gas properties vs. pressure
are entered in the data file in tabular form compatible with
simulator input requirements. For gas condensate deple-
tion studies, constant-volume and constantcomposition
expansion tests yield the required pressure-dependent liq-
uid content, CL (STB/scf), and condensate density values.
A wide variety of laboratory tests are performed for
compositional model studies that involve injection of a
nonequilibrium fluid (dry or enriched gas, CO?, N2,
etc.). Swelling tests yield relative volumes, saturation
pressures, and equilibrium phase compositions for each
of a sequence of mixtures of, say, 1 mole of original reser-
voir oil and injected fluid. 79 Various single- and
multicontact tests may be augmented by ID corefloods
and/or slim-tube displacements. Orr et al. 8o*81 discuss
a variety of C02-oil laboratory tests. Much of the
laboratory PVT test data must be processed to yield cor-
relations or a calibrated EOSs2-85 for simulator input re-
History Matching
In most simulation studies, reservoirs have some period
of historical performance data that include WOR, GOR,
individual phase rates and cumulatives, and pressure
measurements by well. Ideally, periodic (e.g., monthly),
accurate measurements of all these data would be recorded
and available for all wells. In the typical case, many of
these data are unrecorded or unavailable and some of the
reported values may be of questionable accuracy.
The reservoir description based on log and core analysis
data reflects a very small (volumetric) sampling of the
reservoir. The historical reservoir-performance data
reflect the reservoir description, and its impact on
pressure/fluid movement behavior, on a much larger
scale. The previously mentioned geological and
petrophysical work yields an initial reservoir description.
History matching yields a refinement of that description,
which improves agreement between model results and
observed reservoir behavior. The history-match phase of
the simulation study entails a sequence of model runs in
which input reservoir description parameters are altered
to improve this agreement. This is a trial-and-error pro-
cedure frequently requiring considerable engineering
judgment and experience. The description parameters ob-
tained from the geological/petrophysical work often are
used to establish legitimate ranges of parameter variation
in the history-match model runs.
This history-match phase can consume half or more of
the total simulation study (match plus prediction) com-
puting effort and expense, depending on the length of the
history period, complexity of the reservoir, and amount
of available performance data.
Refs. 86 through 90 describe methods and applications
of inverse simulation or automatic history matching. This
concept requires user-specification of a finite set of reser-
voir description parameters to be determined (e.g., zonal
permeability, porosity values), a finite set of observed
reservoir performance data to be matched, and a regres-
sion procedure coded interactively with the simulator. A
single computer submittal is then performed, which in turn
executes many model history runs. The regression pro-
cedure automatically varies description-parameter values
from run to run to determine that set of description-
parameter values that maximize agreement between model
results and the set of observed data. This concept is
especially appealing to the engineers who have experi-
enced the frequently high frustration levels associated with
trial-and-error matching of complex reservoir behavior.
However, to date the trial-and-error procedure still
predominates with isolated successes reported with
automatic history matching. Two factors complicating the
latter approach are (1) the expense of the required single
computer submittal can be very large, (2) the a priori
choice of description parameters (or zonation) can be dif-
ficult, subjective, and lead to a questionable reservoir
Validity of Simulation Results
Uncertainties or errors in simulation model results may
arise from (1) questionable assumptions or mechanisms
not represented in the differential form of the model, (2)
spatial and time truncation error introduced by replace-
ment of the model differential equations by finite-
difference approximations, and (3) inadequately known
reservoir rock and/or fluid description data. In addition,
the exact solution of the difference equations is not at-
tained because of round-off error introduced by the finite
word length of the computer. Round-off error is general-
ly negligible compared with errors from the other three
sources. With some exceptions, the above sources of er-
ror are listed in order of increasing importance. However,
successful history matching can reverse the importance
of the second and third sources.
Comparisons of model and laboratory experiment
results can indicate model validity in the absence of the
Uncertainty 3 above. Several such comparisons show
good model-experiment agreement for gas/oil sys-
tems,91,92 water/oil coning,93 and fractured-matrix im-
bibition. 94
Model Assumptions
An assumption common to many black-oil models is com-
plete re-solution of free gas in accordance with the
saturated R,(p) curve during repressurization. This may
be a poor assumption in a case where gridblock thickness
is large and gas/oil gravity (vertical) segregation is pro-
nounced. Prior to repressurization in a given block, the
free gas may exist as a high gas saturation in only the up-
per portion of the block. This contradicts its representa-
tion in the model as a lower saturation distributed
throughout the entire block volume. In the segregated
state, the gas will redissolve only in the lower or residual
oil saturation in the upper, gas-occupied portion of the
block volume. However, the model will allow re-solution
in the entire blocks oil volume. Pressure hysteresis in
the R,(p) curve has been used to cope with this problem;
an alternative remedy where the computing budget per-
mits is the use of more grid layers.
An assumption common in early black-oil models was
that the reservoir oil obeyed a single pair of B,(p) and
R,(p) curves. Some black-oil reservoirs exhibit a signifi-
cant variation of oil API gravity and PVT behavior with
depth or with depth and areal location. In some such cases,
this variation can be represented in a black-oil model by
simply allowing initial solution gas R, to vary with depth
in the undersaturated oil column, retaining a single set
of B,(p) and R,(p) curves. In other cases, multiple sets
of these curves and two oil components are necessary and
the single oil-type assumption in a black-oil model can
lead to appreciable error.
Mechanisms or phenomena that are significant in some
reservoirs and may not be represented in the model in-
clude compaction, hysteresis in wetting and nonwetting
relative permeabilities, and interlayer wellbore crossflow.
The latter is a particularly difficult modeling problem and
the subject of continuing research. A production well com-
pleted in a number of layers may exhibit production from
some layers and, simultaneously, injection (backflow or
recirculation) into others. Factors that promote this
possibility are low-pressure drawdown (high PI and/or
low rate) and poor vertical communication between the
reservoir layers in the vicinity of the well. A rigorous
treatment of this problem requires modeling of wellbore
multiphase hydraulics and phase segregation combined
with calculation of correct phase mixtures for the layers
undergoing injection.
Spatial Truncation Error
Spatial and time truncation error theoretically can be
reduced to any desired low level by sufficiently reducing
gridblock dimensions and timestep size. However, the
resultant increased number of blocks and timesteps fre-
quently lead to prohibitive computer expense and memory
storage requirements.
Time truncation error is generally insignificant. In most
applications timestep size is restricted by considerations
other than time truncation error, such as model stability,
frequencies of printout, and frequencies of changes in well
data (rates, completions, new wells, etc.). In any given
case, the level of time truncation error can be estimated
by repeating a run or portion of a run with a smaller (or
larger) timestep. Insensitivity of results to timestep size
indicates low-time truncation error.
Spatial truncation error appears in the forms of
numerical dispersion, grid-orientation effects, and error
in calculated well WOR and GOR values. Spatial trunca-
tion error can be expressed in mathematical terms through
complex manipulation of the model differential equations
and Taylor series expansions. In simpler terms, this er-
ror can be viewed as a consequence of replacing the
physical continuum (reservoir formation) by a 3D network
of mixing cells (gridblocks). This consequence is the con-
tradictory requirements that any variable value (pressure,
saturation, temperature, concentration) simultaneously
represents the value at the grid point (e.g. block center)
and the entire blocks volumetric average value. This re-
quirement is not met (1) during a frontal displacement as
a sharp front enters the gridblock, (2) when gravity forces
result in phase segregation within the blocks thickness,
and/or (3) when area1 cusping or coning causes sharp
localized saturation gradients within the block volume.
Numerical dispersion generally appears as falsely
smeared spatial gradients of water saturation in
waterflooding, temperature in steamflooding, solvent in
miscible flooding, and chemical agent in chemical
flooding. This excessive smearing occurs primarily in the
areal (X or y) directions and, if uncontrolled, results in
too early calculated breakthrough times of water (heat,
solvent, etc.) at production wells. This numerical disper-
sion generally increases with increasing areal gridblock
size (AX and AJJ). Lantz95 quantitatively related the dif-
ference equation truncation error term to an artificial,
second-order diffusion term in the differential equation.
The engineer can anticipate possibly significant
numerical dispersion effects in simulating two types of
miscible displacement. The first type is slug or bank, as
opposed to continuous, injection of solvent or CO?.
Numerical dispersion erodes the calculated solvent con-
centration within the bank. If miscibility requires
maintenance of solvent bank integrity or a certain solvent
peak concentration, then this numerical dispersion can
result in a calculated (false) loss of miscibility. The sec-
ond type is multicontact miscibility For continuous sol-
vent injection in 1D simulations, several studies 3X97
report the need for 100 to 300 gridblocks to reduce the
effect of numerical dispersion on miscible front velocity.
Kyte and Berry 75 describe control of numerical disper-
sion in simulation of waterflooding through large area1
gridblocks. They use pseudorelative-permeability curves
obtained from detailed (fine-grid) cross-sectional simula-
tions. Harpole and Hearn98 used their method in a 3D
black-oil study. To date, steamflood simulation general-
ly has been confined to pattern studies for which a suffi-
cient number of gridblocks between unlike wells is used
to minimize numerical dispersion effects. Killough et
al. 73 describe their reduction of numerical dispersion in
a stratified, heterogeneous, repeated pattern black-oil
reservoir study. They performed fine-grid, 3D single-
pattern simulations and then used regression to determine
pseudorelative permeabilities for a 2 x 2 four-block area1
grid representation of the pattern. Agreement between the
3D fine-grid results and four-block pattern results was
good enough to allow fieldwide simulation by use of the
latter coarse, areal definition. Several recent papers99-01
describe local grid refinement, the method of character-
istics, and other methods to reduce numerical dispersion
Pronounced grid-orientation effects have been noted in
simulation of adverse mobility ratio floods with models
incorporating the commonly used five-point difference
scheme and single-point upstream weighting. The value
/ \
i 2 3
Fig. 48.3-Nine-spot grids
Fig. 48.2-Five-point and nine-point difference schemes
of the coefficient krJCIJIpJ in Eq. 2 obviously affects
the interblock Darcy flow rate from gridblock i-Z to block
i. Intuition might dictate evaluation of this coefficient at
some average of variable values (pressure, saturations,
etc.) in the two blocks. However, considerations of stabili-
ty and numerical dispersion frequently have led to its
evaluation at conditions existing in the block from which
the flow occurs-i.e., the upstream block. This is referred
to as single-point upstream weighting. The five-point dif-
ference scheme is reflected in the form of terms of type
A(7Ap) in Eq. 5 for the case of 2D flow. These terms
represent the interblock Darcy flow rates in the mass
balance equation for each gridblock. The solid arrows of
Fig. 48.2 illustrate these flow rates between the gridblock
and each of its four neighbors.
A strong rid-orientation effect was first reported by
Todd et al. for highly adverse mobility waterfloods
and later observed for pattern steamfloods. lo3 An area1
grid with the usual perpendicular x and y axes may be
placed over a five-spot pattern with the x axis either
parallel to or at a 45 angle to the line connecting the in-
jector to a producer (Fig. 48.3). These parallel and
diagonal grids lo2 can result in markedly different
calculated shapes of the water or steam front and the
breakthrough times. This difference was reduced by the
nine-point finite difference formulation described by
Yanosik and McCracken, to4 illustrated by the four ex-
tra dashed-line diagonal flow terms in Fig. 48.2. The nine-
point scheme has been programmed into many simulators
treating adverse mobility ratio waterfloods, steamfloods,
and CO2 solvent displacements.
As an example of this grid orientation effect, Fig. 48.3
shows a 3-acre nine-spot steamflood pattern with the
diagonal grid and 45-shifted parallel grid. This pattern
has three types of wells-labeled 1 (injector), 2 (near pro-
ducer), and 3 (far producer). Reservoir formation and
fluid properties and well rates used in the simulation model
are reported elsewhere. l4 The calculated results in Table
48.2 show the pronounced effect of grid orientation on
steam breakthrough times calculated by use of the five-
point difference scheme. Obviously, steam should arrive
at the near producer, Well 2, before it reaches the far pro-
ducer, Well 3. The parallel grid with the five-point scheme
actually gives breakthrough at Well 3 at 117 days, before
breakthrough at Well 2 (204 days). Table 48.2 shows that
the nine-point difference scheme virtually eliminates the
effect of grid orientation for this problem. Figure 48.4
uses parallel and diagonal grids to show calculated steam-
front shapes at 80 days for the two different schemes. The
difference between the nine-point fronts for two grids is
small and about equal to the error of manual interpolation.
A two-point, upstream weighting method lo2 was pro-
posed to reduce both numerical dispersion and grid-
orientation effects. Abou-Kassem and Aziz lot discuss this
and other methods 1oe-108 for reducing the orientation ef-
fect. They conclude that the nine-point scheme is the most
effective in reducing steamflood grid-orientation effects.
Two studies105-09 show very significant reduction of
grid-orientation effects in pattern steamflood simulation
results when areally homogeneous, square grids (AX=
Ay=constant) are used with the Yanosik and McCracken
nine-point scheme. However, the effects persist for a non-
square, uniform grid (Ax=2Ay)09 and the latter scheme
yields physically unreasonable results for the cases of
heterogeneity and nonuniform grids where AX (or Ay)
varies with x (y). The latter shortcoming is addressed by
several recent papers110112 that propose new or altered
nine-point schemes. Frauenthal et al. I3 describe a
modified five-point difference scheme and Pruess et
al. It4 present a seven-point, hexagonal gridblock scheme
for reducing grid-orientation effects.
The engineer can anticipate possibly significant grid-
orientation effects in simulating single- or repeated-
pattern, adverse mobility ratio displacements. Preliminary
areal, single-pattern model runs allow estimation of the
level of such effects and the need for use of a nine-point
scheme or other remedy.
The discussion and references cited obviously indicate
the current concern regarding effects of numerical-
dispersion and grid-orientation effects on the validity of
Fig. 48.4-Calculated shape of steamflood front in a nine-spot
simulation results. However, these numerical effects are
not serious in many simulation studies. Numerical-
dispersion effects are generally demonstrated as smear-
ing of theoretically sharp fronts in l- or 2D horizontal
displacements in homogeneous formations. Actual reser-
voir behavior frequently reflects strong gravity effects
such as a gas or solvent override or a water underrun.
These gravity effects combined with reservoir structure
(areal variation in dip angle) can have an influence on fluid
movement patterns, which dominates the numerical ef-
fects just discussed. In addition, reservoir heterogeneity
can play the same relatively dominant role as gravity
forces. In highly stratified or layered reservoirs, the dif-
ferent rates of travel of injected fluid through different
layers can dominate the numerical dispersion effect at the
leading edges of the individual layer displacement fronts.
Finally, a given level of numerical-dispersion or grid-
orientation effect is acceptable if its impact on calculated
reservoir performance is inconsequential in an engineer-
ing sense-i.e., in light of the questions being asked.
The model itself often can be used to estimate the level
of these numerical errors and the degree of their accepta-
bility. Before selecting the full study grid, preliminary
model runs using grids of varying coarseness can be per-
formed for a representative cross-section or 3D portion
Well 2 Well 3
Diagonal Parallel Diagonal Parallel
Five-point 47.0 204 1,400 117
Nine-point 87.7 75.5 900 1,000
of the reservoir. The results can be helpful in selecting
the coarsest grid spacing compatible with acceptably low
numerical dispersion.
In fieldwide simulation, spatial truncation error may af-
fect calculated values of well productivity, wellbore pro-
ducing pressure, and WOR and GOR. Without special
measures, the model calculates the well behavior with only
the gridblocks average values of pressure, saturations,
etc. However, the actual well behavior may reflect near-
well coning, liquid dropout, or gas evolution effects. The
dimensions of this near-well region may be two orders
of magnitude smaller than the areal block dimensions (b,
Ay). Thus the blocks average conditions may provide a
poor basis for calculating well behavior. This problem
can be significant for a well completed throughout for-
mation thickness and even more significant for a partial-
ly penetrating well.
The simplest remedy to this problem is applicable in
some cases where good vertical communication results
in a high degree of vertical phase segregation. In this case,
well pseudorelative-permeability curves have been
used. 15,16 These curves reflect the location of the com-
pletion interval and relate well behavior to average block
conditions. A more complicated approach requires use of
multivariable correlations relating well WOR and GOR
to average block conditions. g118 These correlations are
developed from a number of single-well r-z (radial-depth)
model runs with fine grid spacing near the wellbore. The
most rigorous treatment of this problem incorporates in-
dividual 1D radial or 2D ~-2 simulations for each well
simultaneously within the fieldwide 3D simulation. 19,120
Again, in any given case, the model itself can be used
to estimate the severity of this problem through com-
parison of single-well, r-z and representative 3D (portion
of reservoir) model results.
Uncertain Reservoir Description Data
Errors in reservoir description data clearly contribute to
errors in simulation model results. Since the description
data are never exactly known, one might infer that model
results are necessarily erroneous and unreliable. A number
of considerations contribute, in contradiction of this in-
ference, to model results being widely used to select and
to design oil-recovery processes and to forecast oil
Accurate determination of all reservoir description data
is not necessary for reliability of model results. The re-
quired accuracy of any description parameter is propor-
tional to its influence on computed results (reservoir
performance). The simulation model should be used to
perform preliminary sensitivity runs to determine which
description data are important. Expense and effort should
then be concentrated on obtaining or refining only those
sensitive description data. The particular parameters
found to be important will vary from study to study,
depending on the nature of the reservoir, the recovery
process(es) of interest, and study objectives or questions.
For example, if computed oil recovery is insensitive to
wide variations in the gas relative-permeability curve, then
the accuracy of this curve might deserve little attention.
In a case where the gravity drainage mechanism is domi-
nant, the oil relative-permeability curve at low and
midrange oil saturations has a large effect on oil recovery
and deserves effort of definition. Gas viscosity, relative
permeability, and capillary pressure may play virtually
no role and their accuracies are irrelevant. All phase
relative permeabilities may be unimportant in a natural
or flank waterflood of a relatively clean, thick high-relief
sand where gravity forces are dominant with pronounced
phase segregation. Only relative-permeability curve end-
points may be important in such cases. However, thinner
sand or stratification, lower permeability and/or higher
rates can increase the importance of water and oil relative-
permeability curve shapes. While capillary pressure is
unimportant in many reservoir studies, it can provide the
dominant, cross-imbibition, mechanism in waterflooding
thin, heterogeneous water-wet sands.
In some studies, the engineer is less concerned with the
absolute accuracies of both model results and description
data than with the sensitivity of calculated results to varia-
tions in those data. An example is a study performed to
compare oil recoveries under alternative recovery proc-
esses. Model runs performed for each process, with reser-
voir description data varied over estimated ranges of
uncertainty, may yield substantially invariant process
rankings and incremental oil recovery differences. If so,
any significant history-matching effort may be un-
necessary and the only concern regarding accuracy of
description data should be the estimated ranges of uncer-
tainty. Another example is a design study of a given
recovery process performed to optimize pattern type and
size, well completions, and rates. Model runs, as just
described, may show that minimal history-match and/or
laboratory efforts for reservoir description are necessary
to meet the study objective.
As previously mentioned, reservoir description data are
altered through history matching to improve agreement
between model results and reservoir performance data.
Frequently, the study objectives involve estimation of
reservoir performance under displacement conditions not
present or recovery processes not active during the his-
tory period. In such cases, some description parameters
that significantly influence future performance may not
be reflected in the historical performance. An example
is a heavy-oil reservoir that was produced for nearly 40
years under natural depletion with no water drive. Solu-
tion gas was very low and interstitial water saturation was
immobile. A 50% water cut developed in time as a large
pressure decline caused water mobility through water ex-
pansion and porosity reduction. Performance data includ-
ed WOR, GOR, and pressure data for a number of wells.
The only description parameters influencing these data
were formation permeability, compressibility, critical gas
saturation, water relative permeability at saturations slight-
ly above SwC, and gas relative permeability at saturations
slightly above S,,. The history-match effort gave a good
match of performance with a unique set of these parameter
values. However, it provided no information regarding
the full-range relative-permeability curves necessary to
estimate oil recovery under waterflood and steam stimu-
lation or flooding. Laboratory relative-permeability meas-
urements and waterflood and thermal pilots were
conducted in this case.
Laboratory work and well pressure testing can be per-
formed to estimate values of some reservoir description
parameters that are not reflected in performance data.
These parameters, together with others determined by
history matching, can be used in model runs to estimate
oil recovery under the various alternative recovery
schemes within the study scope. Field pilot tests then may
be planned for one or more of the recovery processes,
subject to the model results and engineering judgment.
Argument has persisted for years regarding uniqueness
of the reservoir description obtained by history matching.
A thorough treatment of this question requires length and
mathematical complexity beyond the scope of this chapter.
Any such treatment requires careful definition of terms.
For example, define a reservoir description as a bounded
set of m numbers {Xi} representing selected zonal
permeabilities and porosities and parameters characteriz-
ing relative permeability curves. Let the sets of N numbers
{dj*}, {dj} represent observed and model calculated per-
formance data where dj=dj(x, ,x2.,.x,). If N>m, each
xi affects one or more dj, and the d. are independent .
functions of {Xi} (in a mathematics i sense undefined
here), then with rare exceptions a unique set of parameter
values {Xi} will minimize the difference between the
observed and calculated data. An altered zonation gives
a physically different parameter set {ii}. Again, a unique
set of values of {ai} generally will minimize the dif-
ference between observed and calculated data. However,
for this two-parameter set experiment, comparable
matches of observed data would allow a claim of non-
As a practical matter, study budget and time constraints
prevent exhaustive trials of different parameter sets and
even limit the number of model runs with different com-
binations of parameter values within a given set. General-
ly, difficulty encountered in a history-match effort is that
of finding any reasonable description that gives good
agreement with history. The effort rarely ends with dif-
ficulty in selecting among significantly different reservoir
descriptions that give comparably good matches. In any
event, the pertinent question regarding reservoir-
description data is not related to correctness or uniqueness
in an absolute sense. The pertinent question concerns the
engineering significance of variations in parameter values
within ranges of uncertainty. As discussed previously, the
model itself is useful in estimating this significance.
Simulation Technology
Simulation technology can be divided roughly into the
categories of model definition, model formulation, solu-
tion techniques, and special techniques related to
numerical dispersion control, viscous fingering, and grid-
orientation effects. Model definition includes specifica-
tion of the problem (process) addressed, component iden-
tities, mass transport laws or expressions, fluid PVT and
rock property relationships and, finally, the set of finite-
difference equations expressing conservation of mass for
each gridblock. These equations are generally nonlinear.
Before they can be solved for pressures, saturations, etc.,
they must be linearized and manipulated into a set of
simultaneous linear algebraic equations. The term for-
mulation refers to these manipulations and the final form
of this set of linearized equations. In a general sense, this
set of equations can be expressed in the matrix form Ap = b
where A is a very sparse, banded Nb xN~ matrix and the
known t, and unknown e are column vectors of dimen-
sion Nb . A rapidly expanding portion of the simulation
literature describes increasingly efficient, iterative solu-
tion techniques for this problem.
Model Formulations
In 1959, Douglas et al. I6 proposed leap-ffog and
simultaneous formulations for incompressible 2D two-
phase flow. During 1960-69 a number of authors21-24
described two- and three-phase, 2D and 3D black-oil
models based on this simultaneous formulation. In 1960,
Stone and Garder I8 and Sheldon et al. I7 introduced the
concept of eliminating saturation derivatives among the
black-oil model equations to obtain a sin le difference
equation in pressure. Fagin and Stewart 1$ in 1966 and
Breitenbach et al. *O in 1968 described three-phase black-
oil models based on this implicit-pressureiexplicit-
saturation (IMPES) formulation. The IMPES formulation
is explicit in saturation and composition in that relative
permeabilities and concentrations are expressed explicit-
ly in the interblock flow terms. Solution of the pressure
equations over the grid is followed by an explicit updating
of phase saturations and compositions in each gridblock.
than does IMPES. For each gridblock, the adaptive im-
plicit method internally senses (without user intervention)
which dependent variables (e.g., saturations, pressure,
mole fractions) require implicit dating for stability. For
most practical reservoir problems this results in one equa-
tion per gridblock for a major fraction of the grid and an
overall average number of equations per block con-
siderably less than N. Thus the method can attain the
stability of the implicit formulation with considerably less
computing expense. Also, computer storage requirements
are reduced significantly. Future implementations of this
formulation may contribute to increased model reliabili-
ty (stability) and efficiency in simulations of all types of
recovery processes.
In 1969, Blair and Weinaug12 published a fully im-
plicit formulation which expresses all terms in the inter-
block flow and well production expressions implicitly.
This requires simultaneous solution of all N model equa-
tions. A number of later papers describe implementation
of the implicit formulation in black-oil, * composi-
tional 3 and thermal 39 models.
In 1970, MacDonald23 improved the stability of the
IMPES method for the two-phase water/oil case by
following the pressure equation solution with solution of
a water-saturation equation over the grid using implicit
(new-time-level or end-of-timestep) values of relative
permeabilities in the interblock flow terms. Spillette et
al. 124 extended this concept to the three-phase case and
called the formulation sequential.
Single-well coning studies generally involve radial grid
spacings, resulting in very small gridblocks near the well
and large throughput ratios. For these studies, the IMPES
formulation is unsuitable, and the implicit formulation is
generally the most efficient. ** For field-scale, 3D black-
oil studies, the overall computing time is frequently less
with the sequential than with the IMPES or implicit for-
mulation. The typical black-oil simulator applied today
in 1 ,OOO- or more gridblock, field-scale studies is an IM-
PES model with a user-specified option of sequential solu-
tion. Smaller black-oil studies and preliminary cross-
sectional, coning, and sensitivity studies associated with
the large problems frequently employ the implicit formula-
tion. Recent thermal models involve implicit formulations.
With one exception 3 recent compositional models29-33
are based on the IMPES formulation.
The IMPES formulation can become unstable if the
volumetric flow through a gridblock in a timestep exceeds
a small fraction of the block PV. The more stable sequen-
tial formulation remains stable to much larger ratios of
gridblock volumetric throughput/PV. The tolerable
throughput ratio for the implicit formulation is significant-
ly larger than that of the sequential method. Arithmetic
(or computing cost) per timestep and timestep size both
increase from IMPES to sequential to implicit formula-
tions. Since the total cost of simulating a given time period
is proportional to the product of arithmetic per timestep
and timestep size, all three formulations are used widely
The popular IMPES and more recent implicit formula-
tions are illustrated here for the case of 3D two-phase flow
of water and undersaturated oil. This illustration is in the
form of the Newton-Raphson procedure, which Blair and
Weinaug l2 used in describing their implicit formulation.
For clarity, rock compressibility, gravity, and capillary
pressure are neglected, and phase (component) produc-
tion rates are fixed, independent of pressure and satura-
tions. The terms explicit and implicit refer to the time level
of evaluation for variables or terms in the left side, in-
terblock flow terms of Eqs. 5. Explicit dating denotes
evaluation at the beginning of the timestep, t, (level n),
while implicit dating denotes evaluation at the end of the
timestep, t,+i (level n+ 1). The implicit formulation of
Eq. 5, then, appears for each gridblock as
The sequential formulation can fail to preserve material
balances in some problems where adjacent
ridblock com-
positions differ greatly. 25 Meijerink * described a
stabilized IMPES formulation, which improves the stabili-
ty of IMPES to a lesser extent than the sequential method
but reduces material balance error in regions of steep com-
position gradients.
=f(S,,p)=O . . . . ., . . . . . . . . . . . . (lla)
Thomas and Thurnau 127 describe an adaptive implicit
formulation, which allows different levels of implicitness
in different gridblocks. These various levels may change
with timestep number and with iteration number within
a given timestep. As previously mentioned, the implicit
formulation 12 requires simultaneous solution of N equa-
tions for each block over the entire grid. The correspond-
ing arithmetic effort of solution is proportional to N3.
Since N= 1 for IMPES, the implicit formulation obviously
requires considerably more computing time per timestep
-+,S, -@Jo),1
=g(S,,p)=O, . . . . . . . . . . . . . . . . . . . (llb)
T = interblock transmissibility,
p = pressure,
qpp = production rate of phase P, oil or water
VP = PV of gridblock,
48-I 5
Atp = timestep
bp = reciprocal formation volume factor of
phase P, ST vol/res vol,
Sp = saturation of phase P, and
f, g = function of.
Gridblock indices ij, and k on all terms are suppressed
for clarity. All terms at time level n are known from the
previous timesteps calculations. The absence of time level
subscript denotes the implicit level, n + 1. Thus all terms
T,,T,,b,,b,,S,,S,,p in Eq. 11 represent unknown
values at time level n + 1, For all Nb gridblocks, Eqs. 11
are ~Nz. eouations in the 2Nh unknowns
11 a;k two e&rations
}. For a particular gridblock, Eqs.
in the 14 unknowns consisting of
(S,,p) pairs in the block and its six neighbors. Thesesix
neighbor pairs are introduced by the Laplacian interblock
flow terms. Oil saturations are not additional unknowns
since S, =I-S,; the transmissibilities T and reciprocal
formation volume factors (b) are functions of S ,,, and p.
Application of the well-known Newton-Raphson itera-
tive procedure to Eq. 11 gives
f6p=o . . . . . . . . . . . . . . . . . . . . . . . .
b? p
g(s,,p)=g(sppf)+ - W,
( 1
6S, w
where P is iteration number, superscript P denotes evalua-
tion at (S&), hp=p+ -pp, and S,f,,pp approach the
desired S, ,p values as 4 increases. The terms containing
derivatives are actually sums of seven terms because of
the previously mentioned functional dependence on
neighboring block unknowns. Substituting from Eqs. 11
into Eqs. 12, performing the differentiation and rearrang-
ing the result gives
- pt(S,bc,)Y6P+f(S~,PY)=O .
A(TpZ, A6S,)+A(Tp*2A6p)+
At (S,b,c,)pGp+g(S~,pe)=O, .
or in condensed matrix form,
A(Tc)-Cc+B=O, . . . . . . .
. (14)
where T and C are 2 X 2 matrices and P and R are 2 X 1 -
column vectors:
. . . . . . . . . . . . . . . . . . . . . . . . . . . .
The coefficients Tll , TzI arise from saturation deriva-
tives of relative permeabilities in the transmissibilities.
The C matrix elements are obvious upon inspection of
Eqs. 13 (e.g., c,* = VplA,(S,b,c,). The compressibil-
ities (c, ,c,) appear through the definition c = (1 ib)dbidp.
Eq. 14 written for all gridblocks is a set of Nb linear
parabolic difference equations in the Nt, unknowns
{Pi,.,k}. Following solution by a direct or iterative tech-
nique, the new iterates are calculated as S,$+ =S$+
6Sw ,P
p+ =pp+Gp. Coefficients are recalculated at
(SK, .p
I ) for all blocks and Eq. 14 is solved again.
These outer or Newton iterations are continued until the
maximum values over the grid of { I AS,,, I , I Sp I } are less
than some prescribed tolerances. The term inner itera-
tions refers to iterations performed by an iterative solu-
tion technique in solving Eq. 14 for a given Newton
The IMPES formulation treats transmissibilities ex-
plicitly SO that the first terms in Eqs. 11 are A(T, Ap)
and A(T,,Ap), where T,, and T,, are calculated from
known saturations at time level n. This results, upon ap-
plication of Eq. 12, in zero values for transmissibility
derivatives with respect to neighboring block saturations
and Eq. 16 replaces Eq. 13:
-z(S b ,+ ,c,)6p+f(S$,,pp)=0.. .(16a)
A(T,,Asp) + %b$S,
-~(S,b,c,)6p+g(S&pp)=0. . . (16b)
The single saturation unknown, 6S,, can be eliminated
by multiplying Eqs. 16a and 16b by B,fz and BJ, respec-
tively, and adding to obtain
B,$ A(T,.,A&p) + B,PA(T,,,A&p)
-J$S ,L,cM.+SSUcO)E~p+B~~fY+BljigY=O,. (17)
where B is the formation volume factor, l/b.
This is a set of Nb single or scalar, linear parabolic dif-
ference equations in the Nb pressure unknowns {Gp+ }
As described before, a number of outer or Newton itera-
tions are performed with pressure updated and coefficients
recalculated after each iteration. After convergence,
saturation S, is explicitly calculated, block by block,
from Eq. 1 la (with T,, in the first term). Eq. 17 can be
written more simply for the section to follow as
A(TAGp)-c++r=O, . . . . . . (18)
where T, 6p,c, and r are scalars.
Solution Techniques
Eq. 18 written for all Nb gridblocks can be expressed in
matrix form as
AP=b, . . _ . . . . . . . . . . . . . . . . . . . . (19)
where A is a nonsymmetric, sparse Nb xN~ matrix and
& is the Nb x 1 column vector (Sp Uk ) . Direct solution
(Gaussian elimination) or iterative methods can be used
to solve Eq. 19. The arithmetic effort required in direct
solution strongly depends on the pattern of nonzero
elements in the A matrix. This pattern in turn depends
upon the particular linear ordering or numbering of the
Nb gridblocks. An ordering is simply a one-to-one cor-
respondence between a linear index m = 1,2 . Nh and the
gridblock indices {ij,k}, i= 1,2. . .N,, j=1,2,. . .N,,
k = 1,2. . NZ . Here the term natural ordering denotes
numbering the blocks consecutively first in the shortest
direction, then in the next shortest direction, and finally
in the longest direction. For example, if N, >N, > N,,
m=k+(j-l)N,+(i-l)NYN,. . . . . . . (20)
Breitenbach et al., 29 Peaceman, and others illustrate
the diagonal-band form of the A matrix and minimum
direct solution effort which result from this natural
The half bandwidth of the A matrix is N,N, and the
arithmetic effort (number of multiplications) of direct solu-
tion is roughly proportional to Nb(NyN,)*. Iterative
methods require an arithmetic effort roughly proportional
to Nb. Thus increasing problem size, Nh, renders
iterative solution increasingly preferable to direct solu-
tion. For large problems, computer storage requirement
is also significantly less for iterative than direct solution.
Price and Coats I30 described reduced bandwidth direct
solution methods based on diagonal (D2) and alternate-
diagonal (D4) gridblock orderings. For certain test prob-
lems and iterative methods, they showed a D4
direct/iterative work ratio less than unity for half band-
widths up to about 30. Compared with natural ordering,
D4 ordering can reduce direct solution computational ef-
fort by factors up to four and six for the 2D and 3D cases,
respectively. Woo et al. 13 described other techniques
that take advantage of matrix sparsity to reduce direct
solution effort. In spite of these advancements in direct
solution, iterative methods remain preferable for large
reservoir studies.
solution must be performed for tridiagonal matrices in
LSOR and pentadiagonal matrices in planar SOR. As the
block size is increased the arithmetic work per iteration
increases because of the increased arithmetic associated
with this direct solution within each block. However, con-
vergence rate generally increases and the total number
of iterations correspondingly decreases with increasing
block size. To some extent, optimal block size can be
determined by mathematical analysis 33,134 of this
tradeoff between work per iteration and number of itera-
tions. The SOR methods remain popular because of ease
of coding, low computer storage requirement and
automatic determination of the optimum value of the single
iteration parameter. Varga ~4 describes the power method
for this parameter determination and Breitenbach et al. 129
illustrate its application.
In 1971, Watts135 presented an additive correction
method which improves LSOR convergence rate in highly
anisotropic problems. A highly anisotropic problem is one
where, throughout the grid, transmissibibties in one direc-
tion are much
reater than those in other direction(s). Set-
tari and Aziz 6 extended Watts method to other iterative
solution techniques.
Alternating-direction iterative methods (ADI) were
developed for 2D by Peaceman and Rachford 137 in 1955
and for 3D by Douglas and Rachford 38 in 1956. These
methods were widely used in simulation throughout the
1960s and into the 1970s. The AD1 methods require a
sequence or set of iteration parameters. While mathe-
matical analysis yields an optimal parameter set for cer-
tain cases, 1,137 actual reservoir cases frequently require
some trial-and-error effort.
In 1968, Stone 139 described the strongly-implicit pro-
cedure (SIP); Weinstein et al. t4 described SIP in 3D.
Again, a set of iteration parameters is required. Parameter
estimation methods associated with AD1 have proved
useful for SIP * but, again, some trial-and-error effort is
required or beneficial in man reservoir studies.
A number of studiest29~136~15~141 compare direct solu-
tion, LSOR, ADI, and SIP methods for a variety of test
and reservoir problems. There is no simple answer to
which method is best. The ranking of the methods is
problem-dependent in that it depends on the range of varia-
tion in coefficient (transmissibility) values in the A matrix
and the pattern (e.g., highly anisotropic), if any, of their
variation. In general, the more difficult reservoir problems
have a very large range from the smaller to larger
transmissibilities and this large ratio is not uniformly
associated with a particular direction throughout the grid.
SIP became widely used throughout the 1970s and re-
mains in use today because it frequently outperforms the
other methods in these difficult cases.
A new class or type of iterative methods is the subject
of a number of papers4-50 published since the mid
1970s. Basically, the methods involve approximate fac-
torization of the A matrix into an LU product, followed
by an iterative sequence
Successive-overreluxation (SOR) iterative methods
described bv Young 132,33 have been used in simulators
wpk+ -pk)=rk, , . . . . . . . . .
from the early 1960s. Block SOR (BSOR) methods, in-
cluding line (LSOR), two-line, and planar SOR, have
proved especially popular. The BSOR methods require
L = a lower triangular matrix (all entries Pii =0
direct solution within each block, which means that direct
for j>i),
V = an upper triangular matrix (all entries uii
= 0 forj<i),
k = iterate number, and
rk = b-Apk.
Convergence is accelerated by the conjugate gradi-
14 orthomin, 143 or other techniques. Because of the
sparse, banded nature of the lower and upper triangular
matrices L and V, the arithmetic work per iteration of
solving Eq. 21 is a very small fraction of that required
in direct solution of the original problem, Eq. 19.
These new methods seem very attractive in that they
require no iteration parameter and are more robust than
previous methods. That is, they generally exhibit fast or
reasonable convergence rates, even for difficult reservoir
problems where older methods fail or converge slowly.
Two studies 148,50 showed convergence for the difficult
thermal (steamflood) problem where negative transmis-
sibilities can occur. 15 The interested reader should con-
sult the reference sections of Refs. 141 through 150 for
a number of equally good papers dealing with these new
iterative methods.
Code Vectorization
The computational speed, storage, and vectorization
capabilities of computer hardware have increased sharp-
ly in the past few years. As an example, the Cray-I S com-
puter provides up to 4,000,OOO decimal-words of storage,
compared with a typically available 100,000 words on
most machines used until 1975. Recently introduced com-
puters offer sharp increases in computational speeds and
speed/cost ratios. In addition, vector processing
capabilities of Control Data Corp. and Cray supercom-
puters significantly increase the efficiency of simulators
coded to use this vectorization.
This increased machine size, speed, and vector process-
ing contribute to the feasibility of larger reservoir studies.
Until the middle 1970s, most black-oil studies involved
grids of 3,000 or fewer blocks. In 1980, Mrosovsky et
al. 15* described a Prudhoe Bay field study with more
than 16,000 active gridblocks. Studies on the supercom-
puters with grids of more than 30,000 blocks have been
performed recently.
The vector-processing capability has had a strong im-
pact on simulation technology. The resultant increase in
feasible study size has spurred the development of new,
faster iterative solution techniques. Several papers 52-56
describe the contribution of code vectorization to reduced
computing expense and the need to develop or redesign
model code to take advantage of the vectorization.
A = a nonsymmetric, sparse Nb x Nh matrix
A = Ay,AZk =cross-sectional area normal to
bp = reciprocal formation volume factor of
phase P, ST vol/res vol
C = 2x2 matrix
C,J = concentration of component I in phase J
C,p = concentration of component I in phase P
CNJ = concentration of component N in phase .I
f,J = fugacity of component I in phase J
HJ = enthalpy, energy/mole
rP =
relative permeability to phase P (P=w,o,g)
L= a lower triangular matrix (all entries e,=O
forj>i), see Eq. 21
distance between adjacent block centers,
Nb =
N, =
N, =
N, =
ego =
two =
9I =
(AT~-~ +Ax;)/2, see Eq. 1
1, 2. _ .Nb, linear index
total number of gridblocks, N,N,N,
number of gridblocks in x direction
number of gridblocks in y direction
number of gridblocks in z direction
time level, t n+, =t, +A\t
Nbxl column vector {6pijk}, see Eq. 19
gas/oil capillary pressure
water/oil capillary pressure
component I interblock flow rate,
qpi =
mass rate of production of component I
qpp =
production rate of phase P (water or oil)
R= 2 X 1 column vector
s,, = critical gas saturation
SJ = saturation of phase J
WC =
critical water saturation
T= interblock transmissibility and T=2 x 2
TN =
interblock transmissibility for flow of
component I in phase .I
T,, = oil transmissibility at time level 12
= water transmissibility at time level n
V = an upper triangular matrix (all entries
uij=O forj<i)
UJ = internal energy, energy/mole
I/ = grid block volume, hi A!J kzk
Ax = xi-1 -Xi, where x is p or Z
x/J = mol fraction of component I in phase J
yp = specific weight of phase P
6 = time difference operator, 6X=X,+, -X,
pp = viscosity of phase P
i, j, k = gridblock indices
J = w,o,g or 1,2,3,
n = step number
k = iteration number
P = iteration number
1. Peaceman, D.W. : Fundamenta1.s of Numerical Reservoir Simula-
tion, Elsevier Scientific Pub. Co., New York City (1977).
2. Criehlow, H.G.: Modem Reservoir Engineering-A Simulation Ap-
proach, Prentice-Hall, Inc., Englewood Cliffs, NJ (1977).
3. Aziz, K. and Set&i, A.: Petroleum Reservorr Simulation. Applied
Science Publishers, London (1979).
4. Thomas, G.W.: Principles of Hydrocarbon Reservoir Simulation,
second edition, Intl. Human Resources Dev. Corp.. Boston (1982).
5. Muskat, M.: The Flow of Homogeneous Fluids Through Porous
Media, J.W. Edwards, Inc., Ann Arbor, MI (1946).
6. Muskat, M,: Physical Properties ofOil Production, McGraw-Hill
Book Co. Inc., New York City (1949)
7. Muskat, M.: The Production Histories of Producing Gas-Drive
Reservoirs, J. Applied fhys. (1945) 16, 147.
8. Katz, D.L.: Methods of Estimating Oil and Gas Reserves,
Trans., AIME (1936) 118, 18-32.
9. Buckley, S.E. and Leverett, M.C.: Mechanism of Fluid Displace-
ment in Sands, Trans., AIME (1942) 146, 107-17.
10. Welge, H.J.: A Simplified Method for Computing Oil Recoveries
by Gas or Water Drive, Trams, AIME (1952) 195, 91-98.
11. Odeh, A.S.: Reservoir Simulation-What Is It? J. Pet. Tech.
(Nov. 1969) 1383-88.
12. Staggs, H.M. and Herbeck, E.F.: Reservoir Simulation
Models-An Engineering Overview, J. Pet. Tech. (Dec. 1971)
13. ODell, P.M.: Numerical Reservoir Simulation: Review and State
of the Art, paper presented at the 1974 National AIChE Meeting,
Tulsa, OK, March 11-13.
14. Coats, K.H.: Reservoir Simulation: State of the Art, J. Per.
Tech. (Aug. 1982) 1633-42.
15. Craft, B.C. and Hawkins, M.F. Jr.: Applied Petroleum Reservoir
Engineering, Prentice-Hall, Inc., Englewood Cliffs, NJ (1959).
16. Douglas, I. Jr., Peaceman, D.W., and Rachford, H-H., Jr.: A
Method for Calculating Multi-Dimensional Immiscible Displace-
ment, Trans., AIME (1959) 216, 297-308.
17. Sheldon. J.W.. Harris. C.D.. and Bavlv, D.: A Method for
GeneralReservoir Behavior Simulation dn Digital Computers,
paper SPE 1521-G presented at the 1960 SPE Annual Meeting,
benver. Oct. 2-5. -
18. Stone, H.L. and Garder, A.O. Jr.: Analysis of Gas-Cap or
Dissolved-Gas Drive Reservoirs, Sac. Pet. Eng. J. (June 1961)
92-104; Trans., AIME, 222.
19. Fagin, R.G. and Stewart, C.H. Jr.: A New Approach to the Two-
Dimensional Multiphase Reservoir Simulator, Sot. Per. Eng. J.
(June 1966) 175-82; Trans., AIME, 237.
20. Breitenbach, E.A., Thumau, D.H., and Van Poolen, H. K.: The
Fluid Flow Simulation Equations, paper SPE 2020 presented at
the 1968 SPE Symposium on Numerical Simulation of Reservoir
Performance, Dallas. April 22-23.
21. Coats, K.H. etal.: Simulation ofThree-Dimensional. Two-Phase
Flow in Oil and Gas Reservoirs, Sot. Pet. Eng. J. (Dec. 1967)
377-88; Trans., AIME, 240.
22. Peery, J.H. and Herron, E.H. Jr.: Three-Phase Reservoir Simula-
tion, J. Pet. Tech. (Feb. 1969) 21 l-20; Trans., AIME, 246.
23. Snvder. L.J.: Two-Phase Reservoir Flow Calculations. Sot.
Pe;. Eng. J. (June 1969) 170-82.
24. Sheffield, M.: Three-Phase Fluid Flow Including Gravitational,
Viscous and Capillary Forces, Sot. Pet. Eng. 2. (June 1969)
255-69, Trans., AIME, 246.
25. Zudkevitch, D. and Joffe, J.: Correlation and Prediction of Vapor-
Liquid Equilibria with the Redlich-Kwong Equation of State,
AlChE J. (Jan. 1970) 16, 112-19.
26. Soave, G.: Chem. Eng. Sci. (1972) 27, 1197.
27. Peng, D.Y. and Robinson, D.B.: A New Two-Constant Equa-
tion of State, Ind. Eng. Chem. Fund. (1976) 15, 59.
28. Martin, J.J.: Cubic Equations of State-Which? Ind. Eng. Chem.
Fund. (May 1979) 18, 81.
29. Kazemi, H., Vestal, C.R., and Shank, G.D.: An Efficient
Multicomponent Numerical Simulator, Sot. Per. Eng. J. (Oct.
1978) 355-68.
30. Fussell, L.T. and Fussell, D.D.: An Iterative Technique for Com-
positional Reservoir Models, Sot. Pet. Eng. J. (Aug. 1979)
31. Coats, K.H.: A Equation of State Compositional Model, Sot
Per. Eng. J. (Oct. 1980) 363-76.
32. Nghiem, L.X., Fong, D.K., and Aziz, K.: Compositional Model-
ing With an Equation of State, Sot. Pet. Eng. J. (Dec. 1981)
33. Young, L.C. and Stephenson, R.E.: A Generalized Composi-
tional Approach for Reservoir Simulation, Sot. Pet. Eng. J. (Oct.
1983) 727-42.
34. Gottfried, B.S.: A Mathematical Model ofThermal Oil Recovery
m Linear Systems. Sot. Pet. Eng. J. (Sept. 1965) 196-210;
Trans., AIME, 234.
35. Shutler, N.D.: Numerical Three-Phase Model of the Two-
Dimensional Steamflood Process, Sot. Pet. Eng. J. (Dec. 1970)
405-17; Trans., AIME, 249.
36. Weinstein, H.G., Wheeler, J.A., and Woods. E.G. : Numerical
Model for Thermal process, Sot. Pet. Eng. J. (Feb. 1977) 65-78;
Trans., AIME, 263.
37. Crcokston. H.B., Culham, W.E., and Chen, W.H.: A Numerical
Simulation Model for Thermal Recovery Processes, Sac. Pet.
Eng. J. (Feb. 1979) 37-58; Trans., AIME, 267.
38. Youngren, G.K.: Development and Application of an In-Situ
Combustion Reservoir Simulator, Sot. Pet. Eng. J. (Feb. 1980)
39-5 1.
39. Coats, K.H.: In-Situ Combustion Model, Sot. Pet. Eng. .I. (Dec.
1980) 533-53.
40. Hwang, M.K., Jines, W.R., and Odeh, A..%: An In-Situ Com-
bustion Process Simulator With a Moving-Front Representation,
Sot. Pet. Eng. J. (April 1982) 271-79.
41. Gogarty, W.B.: Status of Surfactant or Micellar Methods, J.
Pet. Tech. (Jan. 1976) 93-102.
42. Johnson, C.E. Jr.: Status of Caustic and Emulsion Methods,
J. Pet. Tech. (Jan. 1976) 85-92.
43. Pope, G.A. and Nelson, R.C.: A Chemical Flooding Composi-
tional Simulator, Sot. Pet. Eng. J. (Oct. 1978) 339-54.
44. Todd, M.R. and Chase, CA.: A Numerical Simulator for Predict-
ing Chemical Flood Performance, Proc., Fifth SPE Symposium
on Reservoir Simulation (1979) 161-74.
45. Fleming, P.D. III, Thomas, C.P., and Winter, W .K.: Formula-
tion of a General Multiphase, Multicomponent Chemical Flood
Model, Sot. Per. Eng. J. (Feb. 1981) 63-76.
46. Thomas, L.K., Dixon, T.N., and Pierson, R.G.: Fractured Reser-
voir Simulation, Sot. Pet. Eng. J. (Feb. 1983) 42-54.
47. Gilman, J.R. and Kazemi, H.: Improvements in Simulation of
Naturally Fractured Reservoirs, Sot. Pet. Eng. J. (Aug. 1983)
48. Blaskovich, F.T. er al.: A Multicomponent Isothermal System
for Efficient Reservoir Simulation, paper SPE 11480 presented
at the 1983 SPE Middle East Oil Show, Manama, March 14-17.
49. Henderson, J.H.. Dempsey, J.R. and Tyler, J.C.: Use of
Numerical Models to Develop and Operate Gas Storage Reser-
voirs, J. Pet. Tech. (Nov. 1969) 1239-46.
50. Mann, L.D. and Johnson, G.A.: Predicted Results of Numeric
Grid Models Compared With Actual Field Performance. J. Pet.
Tech. (Nov. 1970) 1390-98.
51. Thomas, J.E. and Driscoll, V.J.: A Modeling Approach for Op-
timizing Waterflood Performance, Slaughter Field Chickenwire
Pattern, J. Pet. Tech. (July 1973) 757-63.
52. A Study of the Sensitivity of Oil Recovery to Production Rate.
Proc., Alberta Energy Resources Conservation Board, No. 75 11
(Feb. 1974) Schedule 1, Shell Canada Ltd.
53. Stright, D.H. Jr.. Bennion, D. W., and Aziz, K.: Influence of
Production Rate on the Recovery of Oil From Horizontal
Waterfloods, J. Pet. Tech. (May 1975) 555-63.
54. Thakur, G.C., et al. : G-2 and G-3 Reservoux, Delta South Field,
Nigeria: Part 2-Simulation of Water Injection, J. Pet. Tech.
(Jan. 1982) 148-58.
55. Graue, D.J. and Zana, E.T.: Studv of a Possible CO? Flood
in Rangely Field, Colorado, J. Pet.Tech. (July 1981) 1~12-18.
56. Bloomquist, C.W., Fuller, K.L., and Moranville, M.B.: Miscible
Gas Enhanced Oil Recovery Economics and the Effects of the
Windfall Profit Tax, paper SPE 10274 presented at the 1981 SPE
Annual Technical Coifeience and Exhibition, San Antonio, Oct.
57. Todd, M.R., Cobb, W.M., and McCarter, E.D. : CO2 Flood
Performance Evaluation for the Cornell Unit, Wasson San Andres
Field, J. Pet. Tech. (Oct. 1982) 1583-90.
58. Herrera, J.Q. and Hanzlik, E.J.: Steam Stimulation History Match
ofMultiwel1 Pattern in the Sl-B Zone, Cat Canyon Field, paper
SPE 7969 presented at the 1979 California Regional Meeting, Ven-
tura, April 18-20.
59. Williams, R.L.: Steamflood Pilot Design for a Massive, Steep-
ly Dipping Reservoir, paper SPE 10321 presented at the 1981
SPE Annual Technical Conference and Exhibition, San Antonio,
Oct. 5-7.
60. Meldau. R. F., Shipley, R.G., and Coats, K.H .: Cyclic Gas/Steam
Stimulation of Heavy-Oil Wells, J. Pet. Tech. (Oct. 1981)
61. Gomaa, E.E., Duerksen, J.H., and Woo, P.T.: Designing a
Steamflood Pilot in the Thick Monarch Sand of the Midway-Sunset
Field, J. Per. Tech. (Dec. 1977) 1559-68.
62. Moughamian, J.M., et al. : Simulation and Design of Steam Drive
in a Vertical Reservoir, J. Per. Tech. (July 1982) 1546-54.
63. Selection of Reservoirs Amenable to Micellar Flooding, First
Annual Report, Dept. of Energy (Dec. 1980) BC/OOO48-20.
64 Fayers, F.J., Hawes, R.I., and Mathews, J.D.: Some Aspects
of the Potential Application of Surfactants or CO2 as EOR Proc-
essesinNorth Sea Reservoirs, J. Per. Tech. (Sept. 1981) 1617-27.
65. dezabala, E.F., er al.: A Chemical Theory for Linear Alkaline
Flooding, Sot. Pet. Eng. J. (April 1982) 245-58.
66. Patton, J.T., Coats, K.H., and Colegrove, G.T.: Prediction of
Polymer Flood Performance, Sot. Per. Eng. J. (March 1971)
72-84; Trans., AIME, 251.
67. Todd, M.R. and Longstaff, W.J.: The Development. Testing,
and Application of a Numerical Simulator for Predicting Misci-
ble Flood Performance, J. Per. Tech. (July 1972) 874-82; Trans.,
AIME, 253.
68. Cook, R.E., Jacoby, R.H., and Ramesh, A.B.: A Beta-Type
Reservoir Simulator for Approximating Compositional Effects Dur-
ing Gas Injection, Sot. Per . Eng. J. (Oct. 1974) 471-81.
69. Patton, J.T., Coats, K.H., and Spence, K.: Carbon Dioxide Well
Simulation: Part I-A Parametric Study, J. Per. Tech. (Aug.
1982) 1798-1804.
70. Coats, K.H.: Simulation of Gas Condensate Reservoir Perfor-
mance, paper SPE 10512 presented at the 1982 SPE Reservoir
Simulation Symposium, New Orleans, Feb. 1-3.
71. Coats, K.H.: Simulation of l/8 Five-/Nine-Spot Patterns, Sot.
Pet. Eng. J. (Dec. 1982) 902.
72. Killough, I.E., et al.: The Kuparek River Field: A Regression
Approach to Pseudorelative Permeabilities, paper SPE 1053 1,
presented at the 1982 SPE Symposium on Reservoir Simulation,
New Orleans, Feb. 1-3.
73. Killough, I.E., er al.: The Prudhoe Bay Field: Simulation of a
COIIIpleX ReSeNOir, Proc., Intl. Petroleum Exhibition and
Technical Symposium, Beijing (1982) 777-94.
74. Jacks, H.H., Smith, O.E., and Mattax, C.C. : The Modeling of
a Three-Dimensional Reservoir With a Two-Dimensional Reser-
voir Simulator-The Use of Dynamic Pseudo Functions, Sot.
Pef. Eng. J. (June 1973) 175-85.
75. Kyte, J.R. and Berry, D.W.: New Pseudo Functions to Control
Numerical Dispersion, Sot. Pet. Eng. J. (Aug. 1975) 269-76.
76. Killough, J.E. and Foster. H.P. Jr.: Reservoir Simulation of the
Empire Abe Field: The Use of Pseudos in a Multilayered System.
Sot. Per. Ert~. J. (Oct. 1979) 279-88.
77. Coats, K.H., Dempsey. J.R.. and Henderson, J.H.: The Use of
Vertical Equihbrrum in Two-Dimensional Simulation of Three-
Dimensional Reservoir Performance, Sot. Per. Enx. J. (March
1971) 63-71; Trans.. AIME, 251.
78. Hear, C.L.: Simulation of Stratified Waterflooding by Pseudo
Relative Permeability Curves, J. Per. Trch. (July 1971) 80-13.
79. Simon. R., Rosman. A.. and Zana. ET.: Phase-Behavior Proper-
ties of CO*-Reservoir Oil System, Sot. Pet. Eng. J. (Feb. 1978)
80. Orr, F.M. and Silva, M.K.: Equilibrium Phase Compositions
of CO 2 /Hydrocarbon Mixtures-Part 1: Measurement by a Con-
tinuous Multiple-Contact Experiment, Sot. Per. Eng. J. (April
1983) 272-80.
81. Orr, F.M., Silva, M.K., and Lien, C.: Equilibrium Phase Com-
positions of CO2iCrude Oil Mixtures-Part 2: Comparison of Con-
tinuous Multiple-Contact and Slim-Tube Displacement Tests, Sot.
Pet. Eng. J. (April 1983) 281-91.
82. Katz, D.L. and Firoozabadi, A.. Predicting Phase Behavior of
Condensate/Crude-Oil Systems Using Methane Interaction Coef-
ficients, J. Pet. Tech. (Nov. 1978) 1649-55; Trans., AIME, 265.
83. Yarborough, L.: Application of a Generalized Equation of State
to Petroleum Reservoir Fluids, Equations of St&e in Engineer-
ing, Advances in Chemisrgv Series, K.C. Chao and R.L. Robn-
son (eds.), American Chemical Society, Washington, D.C. (1979).
182, 385-435.
84. Whitson, C.H. and Tarp, S.B.:
Evaluating Constant-Volume
Depletion Data, J. Per. Tech. (March 1983) 610-20.
85. Coats, K.H. and Smart. G.T.: Application of a Regression-Based
EOS PVT Program to Laboratory Data, paper SPE 11197,
presented at the 1982 SPE Annual Technical Conference and EX-
hibition, New Orleans, Sept. 26-29.
86. Coats, K.H., Dempsey, J.R., and Henderson, J.H.: A New
Technique for Determining Reservoir Description from Field Per-
formance Data, Sot. Pet. Eng. J. (March 1970) 66-74; Trans.,
AIME, 249.
87. Thomas, L.K. and Hellurns, L.J.: A Nonlinear Automatic History
Matching Technique for Reservoir Simulation Models, Sot. Pet.
Eng. J. (Dec. 1972) 508-14; Trans.. AIME, 253.
88. Wasserman, M.L., Emanuel, A.S., and Seinfeld, J.H.: Prac-
tical Application of Optimal-Control Theory to History-Matching
Multiphase Simulator Models, Sot. Pet. Eng. J. (Aug. 1975)
347-55; Trans., AIME, 259.
89. Bobcrg, T.C., et al.: Application of Inverse Simulation to a Com-
plex Multireservoir System, J. Pet. Tech. (July 1974) 801-08;
Trans., AIME, 257.
90. Watson, A.T., et al. : History Matching Two-Phase Petroleum
Reservoirs, Sot. Per. Eng. J. (Dec. 1980) 521-32.
91. Blair, P.M. and Peaceman, D.W.: An Experimental Verifica-
tion of a Two-Dimensional Technique for Computing Performance
of Gas-Drive Reservoirs, Sot. Pet. Eng. J. (March 1963) 19-27;
Trans., AIME, 228.
92. Ridings, R.L., et al. : Experimental and Calculated Behavior of
Dissolved-Gas-Drive Systems, Sot. Pet. Eng. J. (March 1963)
41-48; Trans., AIME, 228.
93. Mungan, N.: A Theoretical and Experimental Coning Study,
Sot. Pet. Eng. J. (June 1975) 247-54: Trans., AIME, 259.
94. Kazemi, H. and Merrill, L.S.: Numerical Simulation of Water
Imbibition in Fractured Cores, Sot. Per. Eng. J. (June 1979)
95. Lantz, R.B.: Quantitative Evaluation of Numerical Diffusion
(Truncation Error), Sm. Pet. Eng. J. (Sept. 1971) 315-20:
Trans., AJME. 251.
96. Van-Quy, N., Simandoux, P., and Corteville. I.: A Numerical
Study of Diphasic Multicomponent Flow, Sot. Pet. Eflg. J. (April
1972) 171-84; Trans., AIME, 253.
97. Fussell, D.D., Shelton, J.L., and Griffith, J.D. : Effect of Rich
Gas Composition on Multiple-Contact Miscible Displacement-
A Cell-to-Cell Flash Model Study, Sot. Per. Eng. J. (Dec. 1976)
310-16; Trans., AIME, 261.
98. Harpole, K.J. and Heam, C.L.: The Role of Numerical Simula-
tion in Reservoir Management of a West Texas Carbonate Reser-
voir, froc., Intl. Exhibition and Technical Symposium. Beijing
(1982) 759-76.
99. Heinemann, Z.E., et al.: Using Local Grid Refinement in a
Multiple-Application Reservoir Simulator, Proc., SPE Sym-
posium on Reservoir Simulation, San Francisco (1983) 205-18.
100. Ewing, R.E., Russell, T.F., and Wheeler, M.F.: Simulation of
Miscible Displacement Using Mixed Methods and a Modified
Method of Characteristics, Proc., SPE Symposium on Reser-
voir Simulation, San Francisco (1983) 71-82.
101. Carr, A.H. and Christie, M.A.: Controlling Numerical Diffu-
sion in Reservoir Simulation Using Flux-Corrected Transport.
Proc., SPE Symposium on Reservoir Simulation, San Francisco
(1983) 25-32.
102. Todd, M.R., ODell, P.M., and Hirasaki, G.J.: Methods for In-
creased Accuracy in Numerical Reservoir Simulators, Sot. Pet.
Eng. J. (Dec. 1972) 515-30; Trans.. AIME, 253.
103. Coats, K.H., et al.: Three-Dimensional Simulation of
Steamflooding, Sot. Pet. Eng. J. (Dec. 1974) 573-92; Trans.,
AIME, 257.
104 Yanosik, J.L. and McCracken, T.A.: A Nine-Point, Finite-
Difference Reservoir Simulator for Realistic Prediction of Adverse
Mobility Ratio Displacements, Sot. Pet. Eng. J. (Aug. 1979)
253-62; Trans., AIME, 267.
105. Abou-Kassem, J.H. and Aziz. K.: Grid Orientation During Steam
Displacement, paper SPE 10497 presented at the 1982 SPE Sym-
posium on Reservoir Simulation, New Orleans, Feb. 1-3.
106 Holloway, C.C., Thomas, L.K., and Pierson, R.G.: Reduction
of Grid Orientation Effects in Reservoir Simulation, paper SPE
5522 presented at the 1975 SPE Annual Technical Conference and
Exhibition, Dallas, Sept. 28-Oct. 1.
107. Robertson, G.E. and Woo, P.T.: Grid-Orientation Effects and
the Use of Orthogonal Curvilinear Coordinates tn Reservon Simula-
tion, Sot. Pet. Eng. J. (Feb. 1978) 13-19.
108. Vinsome, P.K.W. and Au, A.D.K.: One Approach to the Grid
Orientation Problem in Reservoir Simulation, paper SPE 8247
presented at the 1979 SPE Annual Technical Conference and Ex-
hibition, Las Vegas, Sept. 23-26.
109. Coats, K.H. and Ramesh, A.B.: Effects of Grid Type and Dif-
ference Scheme on Pattern Steamflood Simulation Results, paper
SPE 11079 presented at the 1982 SPE Annual Technical Conference
and Exhibition, New Orleans, Sept. 26-29.
110. Bet-tiger, W.I. and Padmanabhan. L.: Finite-Difference Soh-
tions to Grid Orientation Problems Using IMPES, paper SPE
12250 presented at the 1983 SPE Symposium on Reservoir Simula-
tion, San Francisco, Nov. 16-18.
111. Shah, P.C.: A Nine-Point Finite Difference Gperator for Reduc-
tion of the Grid Orientation Effect, paper SPE 12251 presented
at the 1983 SPE Symposium on Reservoir Simulation, San Fran-
cisco, Nov. 16-18.
112. Coats, K.H. and Modine, A.D.: A Consistent Method for
Calculating Transmissibilities in Nine-Point Difference Equations,
paper SPE 12248 presented at the 1983 SPE Symposium on Reser-
voir Simulation, San Francisco, Nov. 16-18.
113. Frauenthal, J.C., Towler. B.F., and diFranco, R.: Reduction
of Grid-Orientation Effects in Reservoir Simulation By Generalized
Upstream Weighting, paper SPE 11593 presented at the 1983
SPE Symposium on Reservoir Simulation, San Francisco, Nov.
114. Pruess, K. and Bodvarsson, G.S.: A Seven-Point Finite-
Difference Method for Improved Grid Orientation Performance
in Pattern Steamfloods, paper SPE 12252 presented at the 1983
SPE Symposium on Reservoir Simulation, San Francisco, Nov.
115. Emmanuel, A.S. and Cook, G.W.: Pseudo-Relative Permeability
for Well Modeling, Sm. Pet. Eng. J. (Feb. 1974) 7-9.
116. Chappelear, J.E. and Hirasaki, G.J.: A Model of Oil-Water Con-
ing for Two-Dimensional, Area1 Reservoir Simulation, Sot. Per.
Eng. J. (April 1976) 65-72; Trans., AIME, 261.
117. Woods, E.G. and Khurana, A.K.: Pseudofunctions for Water
Coning in a Three-Dimensional Reservoir Simulator, Sot. Pet.
Enn. J. (Aun. 1977) 251-62.
118. Ad&gton, D.V.: An Approach to Gas-Coning Correlations for
a Large Grid Cell Reservoir Simulator, J. Pet. Tech. (Nov. 1981)
119. Akbar, A.M., Arnold, M.D., and Harvey, A.H.: Numerical
Simulation of Individual Wells in a Field Simulation Model, Sot.
Pet. Eng. J. (Aug. 1974) 315-20.
120. Mrosovsky. I. and Ridings, R.L.: Two-Dimensional Radial Treat-
ment of Wells Within a Three-Dimensional Reservoir Model,
Sot. Pet. Eng J. (April 1974) 127-31.
121. Blair, P.M. and Weinaug, C.F.: Solution of Two-Phase Flow
Problems Using lmphcit Difference Equations, Sot. Per. Eng.
J. (Dec. 1969) 417-24; Trans., AIME, 246.
122. Bansal, P.P. et a[.: A Strongly Coupled, Fully Implicit. Three-
Dimensional, Three-Phase Reservoir Simulator, paper SPE 8329
presented at the 1979 SPE Annual Technical Conference and Ex-
hibition, Las Vegas, Sept. 23-26.
123. MacDonald, R.C. and Coats, K.H.: Methods for Numerical
Simulation of Water and Gas Coning, Ser. Pet. Eng. J. (Dec.
1970) 425-36; Trans., AIME, 249.
124. Spillette. A.G.. Hill&ad. J.G., and Stone, H.L.: A High-Stabilitv
Sequential-Solution Approach to Reservoir Simulation,-paper SPE
4542 presented at the SPE 1973 Annual Meeting, Las Vegas, Sept.
30-Oct. 3.
125. Coats, K.H.: A Highly Implicit Steamflood Model, Sot. Pet.
Eng. J. (Oct. 1978) 369-83.
126. Meijerink, J.A.: A New Stabilized Method for Use in IMPES-
Type Numerical Reservoir Simulators, paper SPE 5247 presented
at the 1974 SPE Annual Meeting, Houston, Oct. 6-9.
127. Thomas, G.W. and Thumau, D.H.: Reservoir Simulation Us-
ing an Adaptive implicit Method, Sot. Pet. Eng. J. (Oct. 1983)
128. Trimble, R.H. and McDonald, A.E.: A Strongly Coupled, Ful-
ly implicit, Three-Dimensional, Three-Phase Well Coning Model,
Sot. Pet. Eng. J. (Aug. 1981) 454-58.
129. Breitenbach, E.A., Thurnau, D.H., and Van Poollen, H.K.: Solu-
tion of the Immiscible Fluid Flow Simulation Equations, Sot.
Per. Eng. J. (June 1969) 155-69.
130. Price H.S. and Coats, K.H.: Direct Methods in Reservoir Simula-
tion, Sot. Pet. Eng. J. (June 1974) 295-308; Trans., AIME, 257.
13 I, Woo, P.T., Roberts, S.J., and Gustavson, F.G.: Apphcation of
Sparse Matrix Techniques in Reservoir Simulation, Sparse Matrix
Computations, J.R Bunch and D.E. Rose (eds.), Academic Press
Inc.; Washington, D.C. (1976) 427-38.
132. Young, D.M.: The Numerical Solution of Elliptic and Parabolic
Partial Differential Equations, Survey ofNumerical Analysis, J.
Todd (ed.), McGraw-Hill Book Co. Inc., New York City (1963)
135. Watts, J.W : An Iterative Matrix Solution Method Suitable for
Anisotropic Problems, Sot. Per. Eng. J. (March 1971) 47-5 I;
Trans., AIME, 251.
136. Settari, A. and Aziz, K.: A Generalization of the Additive Cor-
rection Methods for the lterative Solution of Matrix Equations,
Sm. Ind. Appl. Math. J. Number Analysis (1973) 10, 506-21.
137. Peaceman, D.W. and Rachford, H.H.: The Numerical Solution
of Parabolic and Elliptic Differential Equations, Sot. Ind. Appi.
Math. J. (1955) 3, 28-41.
138. Douglas, J. and Rachford, H.H.: On the Numerical Solution of
Heat Conduction Problems in Two and Three Space Variables,
Trans., American Math. Sot. (1956) 82, 421-39.
139. Stone, H.L.: Iterative Solution of lmpliclt Approximation of
Multidimensional Partial Differential Equations, Sot. Ind. Appl.
Math. J. Number Analysis (1968) 5, 530-58.
140. Weinstein, H.G.. Stone, H.L., and Kwan. T.V.: Iterative Pro-
cedure for Solution of Systems of Parabolic and Elliptic Equations
in Three Dimensions, IEC Fundamentals (1969) 8, 281-87.
141. Watts, J.W. III: A Conjugate Gradient-Truncated Direct Method
for the Iterative Solution of the Reservoir Simulation Pressure Equa-
tion, Sot. Per. Eng. J. (June 1981) 345-53.
142. Hestenes, M.R. and Stiefel, E.: Methods of Conjugate Gradients
for Solving Linear Systems, J. of Research (1952) 49, 509-36.
143. Concus, P. and Golub, G.H.: A Generalized Conjugate Gradient
Method for Nonsymmetric Systems of Linear Equations, Report
STAN-CS-76-535, Stanford U.. Stanford, CA (Jan. 1976).
144. Vinsome, P.K.W.: Orthomin, an Iterative Method for Solving
Sparse Banded Sets of Simultaneous Linear Equations. paper SPE
5729 presented at the 1976 SPE Symposium on Numerical Simula-
tion of Reservoir Performance, Los Angeles, Feb. 19-20.
145. Meijerink, J.A. and Van Der Worst, H.A.: An Iterative Solu-
tion Method for Linear Systems of Which the Coefficient Matrix
is a Symmetric M-Matrix, Math. of Camp. (Jan. 1977) 148-62.
146. Kershaw, D.S.: The Incomplete Cholesky-Conjugate Gradient
Method for the Iterative Solution of Systems of Linear Equations,
J. Compt. Physics (1978) 26, 43-65.
147. Young, D.M. and Jea. N.C.: Generalized Conjugate Gradient
Acceleration of Nonsymmetric Iterative Methods, Linear
Algebraic Applicarions (1980) 34, 159-94.
148. Tan, T.B.S. and Letkeman, J.P.: Application of D4 Ordering
and Minimization in an Effective Partial Matrix Inverse Iterative
Method, paper SPE 10493 presented at the 1982 SPE Symposium
on Reservoir Simulation, New Orleans, Feb. 1-3.
149. Behie, A. and Forsyth, P.A.: Practical Considerations for In-
complete Factorization Methods in Reservoir Simulation. paper
SPE 12263 presented at the 1983 SPE Symposium on Reservoir
Simulation, San Francisco, Nov. 16-18.
150. Wallis, J.R.: Incomplete Gaussian Elimination as a Precondi-
tioning for Generalized Conjugate Gradient Acceleration, paper
SPE 12265 presented at the 1983 SPE Symposium on Reservoir
Simulation, San Francisco, Nov. 16-18.
151. Coats, K.H.: Reservoir Simulation: A General Model Formula-
tion and Associated Physical/Numerical Sources of Instability,
Bounaivy and Interior Layers-Computational and Asymptotic
Methods, J.J. Miller (ed.), Boole Press, Dublin (1980) 62-76.
152. Mrosovsky, I., Wong, J.Y., and Lampe. H.W.: Construction
of a Large Field Simulator on a Vector Computer, J. Pet. Tech.
(Dec. 1980) 2253-64.
153. Woo, P.T.: Application of Array Processor to Sparse Elimina-
tion, Proc., paper SPE 7674 presented at the 1979 SPE Sym-
posium on Reservoir Simulation, Denver, Jan. 31-Feb. 2.
154. Nolen, J.S., Kuba, D.W., and Kasic, M.J. Jr.: Application of
Vector Processors to Solve Finite Difference Equations, Sot. Pet.
Eng. J. (Aug. 1981) 447-53.
155. Calahan, D.A.: Performance of Linear Algebra Codes on the
CRAY-I, Sot. Pet. Eng. J. (Oct. 1981) 558-64.
156. Killough, J.E. and Levesque, J.M.: Reservoir Simulation and
the In-House Vector Processor: Experience for the First Year,
paper SPE 10521 presented at the 1982 SPE Symposium on Reser-
voir Simulation, New Orleans, Feb. 1-3.
133. Young, D.M : Iterative Solution of Large Linear Systems.
Academic Press Inc., Washington, D.C. (1971).
General Reference
134. Varga, R.S.: Matrix I&rat& Analysis, Prentice-Hall, Inc., Mattox. C.C. and Dalton, R.L.: Rescwoir Simtdariort. Monograph Ser-
Englewood Cliffs, N.J. (1962) 322. ies. SPE, Dallas: to be published in 1986.
Chapter 49
Electrical Logging
M.P. Tixier, Consulting Engineer *
Well logging is an operation involving a continuous
recording of depth vs. some characteristic datum of the
formations penetrated by a borehole. The record is called
a log. In addition, a magnetic tape is usually made.
Many types of well logs are recorded by appropriate
downhole instruments called sondes, lowered into the
wellbore on the end of a cable. The winch of the logging
cable is generally brought to the well on a special log-
ging truck (Fig. 49. l), which also carries the recorders,
power sources, and auxiliary equipment. The parameters
being logged are measured in situ as the sonde is moved
along the borehole. The resulting signals from the sonde
are transmitted through electrical conductors in the cable
to the surface, where the continuous recording, or log, is
Electrical logging is an important branch of well log-
ging. Essentially, it is the recording, in uncased sections
ofa borehole,of the resistivities (or their reciprocals, the
conductivities) of the subsurface formations, generally
along with the sponfaneous potentials (SP) generated in
the borehole.
Electrical logging has been accepted as one of the most
efficient tools in oil and gas exploration and production.
When a hole has been drilled, or at intervals during the
drilling, an electrical survey is run to obtain quickly and
economically a complete record of the formations
penetrated. This recording is of immediate value for
geological correlation of the strata and detection and
evaluation of possibly productive horizons. The informa-
tion derived from the electrical logs may at the same time
be supplemented by sidewall samples of the formations
taken from the wall of the hole or by still other types of
borehole investigations that can be performed by using
additional wireline equipment available for use with the
logging truck [deviation surveys, caliper (hole-diameter)
surveys, dipmeter surveys, temperature surveys,
radioactivity (gamma ray, density, neutron, and nuclear
Authors of rhe orlglnal chapler on this top!c in the 1962 edltion included fhts author.
Ii G Doll, M. MarIm. and F Segesman.
spectrometry) surveys, acoustic surveys, wireline forma-
tion tester, etc.].
As explained later, several types of resistivity-
measuring systems are used that have been designed to
obtain the greatest possible information under diverse
conditions-e.g., conventional devices (normals and
laterals), induction log (IL), Laterolog (LL),
microresistivity devices, and electromagnetic propaga-
tion logs. Table 49.1 gives the service company
nomenclature for various logging tools.
The typical appearance of a standard electrical log is
illustrated in Fig. 49.2. The left track of the log contains
the SP curve. The middle track contains a l&in. short
normal (shallow-investigation resistivity curve), record-
ed on both regular and amplified sensitivity scales as
solid curves, and a 64-in. normal (medium-investigation
resistivity curve, dashed curve). The right track contains
an 1%ft 8-in. lateral (deep-investigation curve).
Logs recorded with other combinations of resistivity-
measuring devices have a similar general appearance,
although the corresponding devices differ in principle
and performance. Microresistivity logs generally include
a microcaliper curve (hole-diameter recording), which is
useful in the location of permeable zones. Of late, four-
logarithmic tracks are often replacing the two-arithmetic
track mentioned previously.
The curves are recorded on the most appropriate of
several available sensitivity scales. The usual depths of
scales are 2 in. = 100 ft (regular) and 5 in.=100 ft
(detail). Less frequently a scale of 1 in. = 100 ft is used.
For cases where great detail is involved, as in microlog-
ging and dipmeter logging, special expanded scales are
available. In many parts of the world, metric depth
scales are used instead of English scales.
Earth Resistivities
Formation resistivities are important clues to probable
lithology and fluid content. With a few exceptions that
are rare in oilfield practice, such as metallic sulfides and
graphite, dry rocks are very good insulators but, when
their pores are impregnated with water, they conduct
electric current. Subsurface formations in general have
finite measurable resistivities because of the water con-
tained in their pores or adsorbed on their interstitial clay.
Formation resistivity also depends on the shape and the
interconnection of the pore spaces occupied by the
water. These depend on the formation lithology and, in
the case of reservoir rocks, on the presence of noncon-
ductive oil or gas.
Units of Resistivity and Conductivity. In electrical log-
ging, the resistivity is usually measured. An exception is
induction logging, in which the conductivity is recorded
along with its reciprocal, the resistivity. Measurements
made with electromagnetic propagation are discussed
Fig. 49-l-Setup for wireline logging operations in wells
The resistivity (specific resistance) of a substance to
the flow of electrical current, at any given temperature,
is the resistance measured between opposite faces of a
unit cube of that substance. In electrical-logging work,
Schlumberger Gearhart
Electrical Log Electrical Lag
Induction Electric Log (IEL) Induction Electric Log
Induction Spherically Focused Log (ISF)
Dual Induction Spherically Focused Log Dual Induction-Laterolog
Laterolog.3 (LL3)
Dual Laterolog
Microlog (ML)
Microlaterolog (MLL)
Proximity Log (PL)
Microspherically Focused Log (MSFL)
Borehole Compensated Sonic Log
Long Spaced Sonic Log
Cement Bond/Variable Density Log
Gamma Ray Neutron
Sidewall Neutron Porosity Log
Compensated Neutron Log (CNL)
Thermal Neutron Decay Time Log
Dual Spacing TDT
Compensated Formation Density Log
Litho-Density Log
High Resolution Dipmeter
Formation Interval Tester
Repeat Formation Tester
Sidewall Sampler
Electromagnetic Propagation Log
Bore Hole Geometry Tool
Ultra Long Spacing Electric Log
Natural Gamma Ray Spectrometry
General Spectroscopy Tool
Well Seismic Tool
Fracture Identification Log
Dual Laterolog
Micro Electrrc Log
Sorehole Compensated Sonic Log
Sonic Cement Bond System
Gamma Ray Neutron
Sidewall Neutron Porosity Log
Compensated Neutron Log
Compensated Density Log
Four Electrode Dipmeter
Selective Formation Tester
Sidewall Core Gun
X-Y Caliber
Fracture Detection Log
Dresser Atlas
Induction Electrolog
Dual Induction Focused Log
Focused Log
Dual Laterolog
Proximity Log
Sorehole Compensated Sonic Log
Long Spacing BHC Acoustilog
Acoustic Cement Bond Log
Gamma Ray Neutron
Sidewall Epithermal Neutron Log
Compensated Neutron Log
Neutron Lrfetime Log
Dual Detector Neutron
Compensated Densilog
Formation Tester
Formation Multi Tester
Caliper Log
Carbon/Oxygen Log
Electric Log
Induction Electric Log
Dual Induction Log
Guard Log
Dual Guardlog
Contact Log
F,R,, Log
Acoustic Velocrty Log
Gamma Ray Neutron
Srdewall Neutron Log
Dual Spaced Neutron Log
Density Log
Formation Tester
Sidewall Coring
Caliper Log
the meter was chosen as the unit of length; so the unit of
resistivity is taken as the (Q.m)/m, or more simply, the
ohm-meter, 52.m.
Since conductivity is the reciprocal of resistivity
(C=lIR), a possible unit of conductivity would be
l/(Q.m), or G/m. However, since this unit would
necessitate extensive use of decimal fractions, a unit
one-thousandth as large, the millimho/meter (mu/m), is
employed. Thus, formations having resistivities of 10,
100, or 1,000 Q*rn have conductivities of 100, 10, or 1
mu/m, respectively.
Dependence of Water Resistivity on Salinity and
Temperature. The resistivity of an electrolytic solution
decreases as the amount of chemicals therein increases.
At any given temperature the electrical conductivity of a
formation water or a drilling mud will depend on the
concentration and nature of the dissolved chemicals.
In most cases the predominant solute is sodium
chloride (NaCl); therefore, the NaCl conversion chart
(Fig. 49.3) may generally be used to obtain resistivity
from concentration. If other chemicals are present in
relatively large amounts, it is possible to convert the con-
centrations of such chemicals into equivalent concentra-
tions of NaCl to find the resistivity. To make the conver-
sion, apply the appropriate multipliers given in Table
49.2 for the concentration of each separate ion [in parts
per million (ppm) or (m 3 /m 3 ) by weight, or in grains per
gallon (gr/gal) or (kg/m)], and add the products. Note
that concentrations expressed in milligrams per liter
(mg/L) and in ppm may be appreciably different at high
concentrations. Below about 50,000 ppm, however,
measurements at room temperature in the two units may
be used interchangeably without serious error.
tt f
Fig. 49.2-Typical electrical log.
Fig. 49.3~Resistivity vs. concentration for NaCl solutions at various temperatures.
The resistivity of an electrolytic solution decreases as
its temperature increases. This is of great importance,
since temperature in the earth increases with depth.
Before the resistivity of the drilling mud (measured at
surface temperature) can be compared with that of a for-
mation (measured at a much higher temperature in a deep
well) the resistivities must be converted to values that
would have been observed at a common temperature.
The temperature conversion is accomplished by means
of Fig. 49.3, which shows for NaCl solutions the effects
of both salinity and temperature on resistivity . Downhole
temperatures may be estimated from a so-called bot-
tomhole temperature (BHT) obtained by means of a
maximum-reading thermometer inserted in the body of
the sonde.
Resistivities of Formation Waters. Formation waters
can vary remarkably with geographic location, depth,
and geological age. Shallow groundwaters are usually
fresh (not saline), with resistivities sometimes exceeding
20 to 50 !l. m at room temperature. They also may con-
tain appreciable amounts of calcium and magnesium
salts, which make them hard. At great depths, forma-
tion waters generally tend to be more saline. In deep
wells, formation-water resistivities sometimes may cor-
respond to complete saturation (0.014 O.rn at 200F).
A knowledge of R,. , the formation-water resistivity. is
important in electrical-log interpretation. R,, may be ob-
tained from the readings of the SP curve (Eq. 9) or from
resistivity measurements on samples of formation water
recovered from production or in drillstem tests. It also
may be estimated from measurements of the resistivity of
the permeable formations of interest when they are 100%
water-saturated, Ro, if the porosity or formation factor is
known (Eqs. 1 and 2). R, may be computed, as has been
explained, from analyses of formation waters. Resistivi-
ty of formation waters is discussed further in Chap. 24.
Mud, Mudcake, and Mud-Filtrate Resistivities.
Resistivities of the mud, R,, the mudcake, R,, , and the
mud filtrate, R,,,f, are all important in log interpretation.
R, is obtained by direct measurement on a mud sample.
R,,,f and R, are obtained by direct measurements on
filtrate and mudcakes pressed from a sample of the mud,
or they can be estimated from average statistical data on
the basis of mud resistivity. 2-4 Correction for the varia-
tion of these resistivities with temperature is made by use
of Fig. 49.3.
Formation Resistivity Factor. If R. is the resistivity of
a clean (nonshaly) formation completely saturated with
water of resistivity R,, the ratio Ro/R, will be a con-
stant that depends on the lithologic structure of the for-
Cations Anions
Na 1.0 Cl 1.0
Ca 0.95 so4 0.5
WI 2.0 co3 1.26
HCO, 0.27
mation and not on the resistivity, R,, of the saturating
water. This constant is the formation resistivity factor,
FR, commonly called formation factor.
FR=- R, . . . . . . . . . . . . . . .
Dependence of Formation Factor on Porosity and
Lithology. The formation factor, F, , of a clean forma-
tion can be related to its porosity, 6. by an empirical for-
mula of the form F~=alc$'?', where a and m are con-
stants. The exponent m, sometimes called the cementa-
tion exponent or factor, varies with the lithology.
In the construction of many graphs for log interpreta-
tion, 2 the Humble formula proposed by Winsauer et
a1.5 has been generally adopted:
FR= ~2.,5 . . . . . . . . . . . . . . . . . I . . .
An early formula proposed by Archie, which fits par-
ticularly well for consolidated formations such as hard
sandstones and limestones, is
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...(3)
Limestones often contain vugs, interconnected with
fissures, which add their porosity to that of the matrix.
When the vugs and fissures are spaced closely, com-
pared with the spacings of the resistivity-measuring
devices, Eq. 3 often can be used as in the case of sand-
stones or limestones with only granular porosity. Never-
theless, it is sometimes advisable to use values of m
greater than two as required to fit local observations.
Shaly (Dirty) Formations. Shales and clays are
themselves porous and are generally impregnated with
mineralized water. Therefore, they have appreciable
conductivity, which is enhanced by ion-exchange con-
duction through the shale matrix. (This shale conduction
is sometimes, though not quite properly, referred to as
resulting from conductive solids. ) On the other hand,
the size of the shale pores is so small that practically no
movement of fluid is possible. Accordingly, shale,
whether deposited in thin laminations or dispersed in the
interstices of the sand, contributes to the conductivity of
the formation without contributing to its effective
The relation between formation resistivity and porosity
becomes more complex for shaly formations than for
clean formations. Because of the additional shale con-
ductance, the ratio of formation resistivity to water
resistivity (i.e., the formation factor) is not constant
when the resistivity of the impregnating water changes. 6
Nevertheless, if the shale content is not too great, ex-
perimental observations show that for low enough values
of water resistivity this ratio is almost constant, as
though the conductance of the shale were then negligible
in comparison with that of the water; and a limiting for-
mation factor is found, which is related approximately to
the effective porosity in the same way as the formation
factor of a clean sand.
Relation Between Formation Resistivity and Satura-
tion. When a part of the pore space is occupied by an in-
sulating material such as oil or gas, the resistivity of the
rock, R,, is greater than the resistivity that it has when
100% water-bearing, R,. The resistivity of such rock is
a function of the fraction of the PV occupied by water.
For substantially clean formations, the water satura-
tion, S,, is related to R, (resistivity of formation con-
taining hydrocarbons and formation water, with a water
saturation S,) and R,J (resistivity of same formation
when 100% saturated with the same water) b an em-
pirical relation known as the Archie equation.
. . . . . . . . . . . . . . . . . . . .
Empirically determined values of n range between 1.7
and 2.2, depending on the type of formation. Experience
shows that n =2 should give a sufficiently good approx-
imation. Then, combining Eqs. 4 and 1 gives
SW=(+) 1/i =(F) I/?. . . . . . . . . . . .
The ratio RJR0 is sometimes designated as the resistivi-
ty index, 1~; accordingly, S, =(ZR) -I.
The relation between formation resistivity and water
saturation is more complex when the formations contain
some shale or clay because of the additional conductance
resulting from the interstitial shale. sv9
Ranges of Resistivity-Formation Classifications.
Clays and shales are porous, practically impervious for-
mations and are often very uniform throughout their
mass. Their resistivity is comparatively low and prac-
tically constant over wide intervals. Compact and imper-
vious rocks, such as gypsum, anhydrite, dense
calcareous formations, or certain kinds of coal, are
highly resistive because of their very small interstitial
water content.
Resistivities of porous and permeable formations, such
as sands, vary widely, depending on their lithology and
fluid content. In electrical logging it is convenient to
classify reservoir rocks as follows.
Soft Forma&ions. These formations are chiefly poorly
consolidated sand/shale series. The porosity of the sands
is intergranular and exceeds 20%. Resistivities range
from 0.3 !2* m for saltwater-bearing sands to several fl. m
for oil-saturated sands.
Intermediate Formations. These are chiefly
moderately consolidated sandstones but frequently
limestones and/or dolomites. Reservoir porosity is
generally intergranular, ranging from about 15 to 20 % .
The reservoir formations are interbedded with shales and
very often with tight rocks. Resistivities range from 1 to
about 100 Q-m.
Hard Formations. These are chiefly limestones andior
dolomites, and also consolidated sandstones. They con-
sist mostly of tight rocks containing porous and
permeable zones, and shale streaks. The porosity of
reservoirs is less than 15 % . Most often, the porous and
permeable zones contain fissures and vugs. Resistivity
range is from 2 to 3 Q. m to several hundred. For the
completely tight formations, such as salt and anhydrite,
the resistivity may be practically infinite.
Anisotropy. In many sedimentary strata, the mineral
grains have a flat or plate-like shape with an orientation
parallel to the sedimentation. Current travels with great
facility along the water-filled interstices, which are
mostly parallel to the stratification. These strata,
therefore, do not possess the same resistivity in all direc-
tions. Such microscopic anisotropy is observed mostly in
Moreover, in electrical logging, the distance between
electrodes or coils on the measuring devices is great
enough that the volume of formation involved in the
measurements very often includes sequences of interbed-
ded resistive and conductive streaks. Since current flows
more easily along the beds than perpendicular to them,
the formation has macroscopic anisotropy.
Both kinds of anisotropy may add their respective ef-
fects to influence the apparent resistivity. The
longitudinal, or horizontal, resistivity , RH, measured
along the bedding planes is always less than the transver-
sal, or vertical (perpendicular) resistivity, Rv.
Resistivity-measuring devices whose readings are not
appreciably affected by the borehole [the deep induction
log (IM), and under certain conditions, the laterolog
(LL), and the long lateral when the ratio RHIR, is low
or moderate] will read RH. Because of the borehole ef-
fect, the short-spacing-electrode devices usually read
values greater than RH. lo
Distribution of Fluids and Resistivities in Permeable
Formations Invaded by Mud Filtrate. Inasmuch as the
hydrostatic pressure of the mud is usually maintained
greater than the natural pressure of the formations, mud
filtrate (forced into the permeable beds) displaces the
original formation fluids in the region close to the
borehole. Solid materials from the mud deposited on the
wall of the hole form a mudcake, which tends to impede
and reduce further infiltration.
The thickness and the nature of the mudcake depend
on the kind of mud and on the drilling conditions rather
than on the formations. The thickness, h,,,,, is usually
between /s and 1 in. For water-based muds the mudcake
resistivity, R,, , is about equal to one or two times the
mud resistivity, R,. In some oil-emulsion muds, R,,
may be somewhat greater.
Fig. 49.4a represents a schematic cross section of an
oil-bearing permeable bed penetrated by a borehole. Fig.
49.4b and 49.4~ show the corresponding radial distribu-
tion of fluids in formation and resistivities.
As indicated in Fig. 49.4a, the zones of different
resistivity may be divided into the drilling mud within
the borehole (of resistivity R,); the mudcake R,,, the
flushed zone R,,; a transition zone; in some cases an
annulus. R, (present only in certain oil- or gas-
bearing formations); and the uncontaminated zone (of
resistivity R,). The invaded zone (of average
resistivity, Ri) includes the flushed zone and the transi-
tion zone.
I nvaded Zone. This zone is behind and close to the
wall of the hole; it is believed that most of the original
interstitial fluids have been flushed out by the mud
rANNULU5 (Ran)
Fig. 49.4-a. Horizontal section through a
permeable oil-bearing bed
(S, < 60%); b. radial distribu-
tion of fluids in formation
(qualitative); c. radial distribu-
tion of resistivities.
filtrate. This flushed zone, of resistivity R,,, is con-
sidered to extend, under usual conditions of invasion, at
least 3 in. from the wall. Exceptions to this rule can
If the bed is water bearing, the pores in the flushed
zone are completely filled with the mud filtrate, and for
clean formations R,, is nearly equal to F,R,f; FR being
the formation factor and R mf the mud-filtrate resistivity
If the bed is oil bearing, the flushed zone contains
some residual oil saturation, S,, . From Eq. 5, S,, , the
water saturation in the flushed zone is
R,,=T, . . . . . . . . . . . . . . . . . . . . . . . . . . . .
where S,=l-S,,.
Beyond the region of maximum flushing, R,, , there is
a more or less extended transition region, the nature of
which depends on the characteristics of the formation,
the speed of invasion, and the hydrocarbon content. The
invaded zone includes the flushed zone and the part of
the transition zone invaded by filtrate. In the case of
water-bearing sands and oil-bearing sands of high water
saturation, the invaded zone extends up to the uncon-
taminated zone, R,.
There can be no exact definition of the depth of the in-
vaded zone, but it is convenient to introduce a factor di,
called the electrically equivalent diameter of inva-
sion, corresponding to an average invaded zone of
resistivity Ri , which has the same effect as the actual in-
vaded zone on measurements made in the borehole. The
depth of invasion is variable. It depends on the plastering
properties of the mud, pressure differences between the
mud column and the formation, time elapsed since the
formation was drilled, porosity of the formation, propor-
tion and nature of the fluids (water, oil, gas) present in
the pores, reaction of any interstitial clays with the mud
filtrate, etc.
All other conditions being the same, the greater the
porosity, the smaller the depth of invasion. With usual
muds, di seldom exceeds 2dh (dh =hole diameter) in
high-porosity sands, but it may exceed 5dh and even
1Odh in low-porosity formations such as consolidated
sandstones or limestones. In some cases, invasion can be
extremely shallow in very permeable formations and in
gas-bearing formations.
In very permeable beds, when there is an appreciable
difference between the specific gravities of the mud
filtrate and the salt-laden interstitial water, gravity-
segregation effects may occur, with the fresher filtrate
tending to accumulate at the top boundary of the bed,
resulting in a decrease in the depth of invasion in the
lower part of the bed. I
In fissured formations, the permeability is quite often
enormous because of the fissures-much greater than the
permeability of the matrix material surrounding them.
Suppose that a formation is composed of a porous but
relatively impermeable material, broken by networks of
roughly parallel fissures. Mud filtrate penetrates the
fissures easily and deeply, driving out much of the
original fluids (oil and formation water). On the other
hand, the matrix itself may be penetrated hardly at all by
the filtrate. Since the l%.sures constitute a small part of
the total PV, only a vety small portion of the total
original fluids is displaced. As a result, R, is little dif-
ferent from R,, and the ratio R,IR,,f is no longer
representative of the formation factor.
Annulus. When the formation contains hydrocarbons,
the process of invasion is complex. The distribution of
fluids is then affected by the two-phase permeabilities,
relative densities (gravities) and viscosities of the fluids,
capillary forces, etc.
When the initial water saturation is low (less than
about 50%), one important feature is the existence of an
annular region just inside the uncontaminated zone, con-
taining mainly formation water and some residual oil.
This annulus is explained as follows. The mud filtrate
penetrates the formation radially, sweeping the
removable oil and formation water ahead of it. For large
oil saturation, the relative permeability to oil is ap-
preciably greater than that to water. Therefore, the oil
moves faster, leaving a zone (the annulus) enriched in
formation water behind it.
It seems likely that, because of the effects of diffusion,
capillary pressure, gravity, etc., the existence of a well-
defined annulus is a transitory phenomenon. Field log
experience nevertheless seems to show that the annulus
does very often exist at the time the logs are run. Com-
putations have shown that the presence of the annulus
has a practically negligible effect on the response of the
devices with electrodes (normals, laterals, and
laterolog) It may have an effect on the induction log, but
this can be taken care of for practical purposes by means
of appropriate interpretation charts. 2
Uncontaminated Zone. For clean formations, from
Eq. 5,
s,2. . . . . . . . . . . . . I . .
In the usual case, R,,,f is 10 to 25 times as large as R w.
Thus, comparing Eqs. 6 and 7 with usual values of S,
and S,, ,
R,, even in oil-bearing formations, is often less
than R,, as represented in Fig. 49.4~.
Apparent Resistivity. Since any resistivity measure-
ment is affected in some degree by the resistivities of all
the media in the immediate vicinity of the sonde (i.e.,
mud, different parts of the formation that vary in
resistivity, adjacent formations if the bed measured is
thin), any given device records an apparent resistivity.
Each resistivity device is calibrated so that when the
sonde is in a homogeneous medium (or in some other
condition appropriate to practice, specified for the par-
ticular device) the apparent resistivity reading is equal to
the actual resistivity.
Requirements for and Types of Resistivity Devices.
Inspection of basic relations in Eqs. 1, 2, 5, and 6 shows
that a determination of S, and 4 requires a knowledge of
R, and R,, (or R i, in certain cases where R,, is not easi-
ly determined). Thus, for the reservoir-evaluation prob-
lem, it is necessary to have resistivity-measuring devices
with different depths of investigation to obtain values in-
dicative of the resistivities of the invaded zone and the
uncontaminated zone. The readings of the deep- and
shallow-investigation curves may often be used to cor-
rect each other, through correction charts or departure
curves, to obtain better values of R, and Ri
Another function of resistivity recording is to provide
an accurate definition of bed boundaries, particularly of
permeable beds. Finally, it is desirable that the readings
not be influenced by the effect of the mud column or, in
case of thin beds, by the adjacent formations.
These requirements are only partly satisfied with the
conventional resistivity devices. The introduction of
microdevices and focused devices has brought about an
appreciable improvement.
Currently used resistivity devices may be classified in
two categories.
1. Macrodevices, which derive their reading from
about 10 to 100 cu ft of material around the sonde (useful
for R, and Ri evaluation), and include unfocused-
electrode devices, focused-electrode devices, and induc-
tion logging devices.
2. Microdevices (also called wall-resistivity devices),
which derive their readings from a few cubic inches of
material behind or close to the wall of the hole. Since the
electrodes are mounted on an insulating rubber pad
pressed against the wall of the hole, measurements are
affected only marginally by the mud column.
Microdevices arc of unfocused and focused types.
Resistivity devices that have electrodes may be used in
holes filled with water or water-based drilling mud,
which provides the electrical contact necessary between
electrodes and formation. The induction log can also be
used in empty holes or in holes filled with nonconductive
oil-based mud. The various resistivity devices are
described later.
Spontaneous Potential (SP) Log
The SP log is a record of the naturally occurring poten-
tials in the mud at different depths in a borehole. The
measurement is made in uncased holes containing water-
based or oil-emulsion muds between an exploratory elec-
trode on the sonde in the borehole and a stationary
reference electrode at the surface.
Usually the SP curve (Fig. 49.2) consists of a more or
less straight baseline (corresponding to the shales) hav-
ing excursions or peaks to the left (opposite the
permeable strata). The shapes and the amplitudes of the
excursions may be different, according to the forma-
tions, but there is no definite correspondence between
the magnitudes of the excursions and the values of
permeability or porosity of the formation.
The principal uses of the SP curve are to (1) detect the
permeable beds, (2) locate their boundaries (except when
the formations are too resistive), (3) correlate such beds,
and (4) obtain good values for R,, the formation-water
Origin of the SP. The character of the potentials
measured in the mud results from ohmic drops produced
by the flow of SP currents through the mud resistance. If
the mud is extremely conductive, these ohmic drops may
be insignificant, and the variations in the SP curve may
be too small to be useful.*
The SP currents flow as a result of electromotive
forces (EMFs) existing within the formations or at the
boundaries between formations and mud. One
phenomenon that could cause an EMF to appear across
the mudcake opposite a permeable bed is electrofiltra-
hon. The mud filtrate, in being forced through the mud-
cake, would tend to produce an EMF, positive in the
direction of flow. According to experiments, I2 the EMF
across the mudcake may be quite sizable, but there is
also an electrotiltration EMF generated across the adja-
cent shales. Thus, the net effect of electrotiltration in
causing variations of SP is small and in most cases
negligible for all practical purposes-a conclusion
verified by field experience.**
Most important ate the EMFs of electrochemical
origin, which occur at the contacts between the drilling
mud (or its filtrate) and the formation water, in the pores
of the permeable beds, and across the adjacent shales. I6
In a clean sand lying between shale beds, all penetrated
by a borehole containing conductive (water-based) mud,
the total electrochemical EMF, E,, is produced in the
chain (Fig. 49.5): Mud/mud filtrate/formation water/
shale/mud. The EMF of the junction, mud/mud filtrate,
is taken to be practically nil because, although the
resistivities of the mud and its filtrate may differ, their
electrochemical activities should be the same.
The part of the chain consisting of formation
water/shale/mud gives rise to the shale-membrane
EMF, Em. The part mud ftltrateiformation water
gives rise to the liquid-junction EMF, EJ . For NaCl
(monovalent-ion) solutions, at 75F,
*In such acase the gamma ray log. which distinguishesshales from nonshale beds,
IS sometimes recorded as a subslltute for the SP
Further information on the electrof~ltralion EMF, or streaming potential, may be
found in Refs. 13 through 15.
Fig. 49.5--Schematic representation of
electrochemical chain and
SP current path at boundary
between permeable bed and
adjacent shale.
where a, and a,f are the chemical activities of the for-
mation water and mud filtrate, respectively (at 75 F),
and EM and EJ are in millivolts. The total E, is the sum
of E,+, and E,:
E,=K, log,,% . . . . . . . . , . . . . . . . . . . . (8)
where K, is the electrochemical coefficient and is equal
to 71 at 75F.
Eq. 8 is general, provided that both formation water
and mud filtrate are essentially NaCl solutions of any
concentration. The values of K, are directly proportional
to the absolute temperature. Thus, at 150F the coeffi-
cient K, in Eq. 8 becomes 81 instead of 71, and at 300F
it becomes 101 (see Fig. 49.6).
From Eq. 8, in the usual case of a, greater than a,,f,
E, is positive. However, if a,f is greater than a,, cor-
responding to mud mom saline than formation water,
then E, is negative and the SP deflections corresponding
to permeable beds are then reversed on the log.
Effect of Invasion on Generation of the EMF. In the
explanation of the electrochemical potential, it has been
assumed that no shale-type potential is created by the
mudcake. In the normal case, mud filtrate bathes both
sides of the mudcake and no shale-type potential can
arise. In some formations, there is only a little filtrate
behind the mudcake. Such small amount of filtrate will
be contaminated easily by the formation water. In this
case, one face of the mudcake is wetted by the filtrate in
the hole, the other face by contaminated filtrate of dif-
ferent activity. This will give rise to a shale-type poten-
tial of the same polarity as the main shale potential, and
the SP curve will be decreased. This explains the
decreasing of the SP curve with time in very highly
permeable beds. I7 The filtrate is evacuated by gravity
Fig. 49.6--R, determination from the SP. The inset chart of true
R, vs. R, applies to formation waters of average
segregating forces and the formation fluids tend to come
back toward the hole with time.
Conversely, an increase in SP with time is observed
often in low-permeability water-bearing formations.
Very little filtrate invades the formation in a freshly
drilled hole and the filtrate is contaminated by the forma-.
tion water. As the invasion proceeds, more and more
filtrate goes into the formation and the mudcake is wet-
ted on both faces by the mud filtrate. When the mudcake
does not contribute any shale-type potential, the SP
curve, recorded on the front of a thick permeable sand, is
said to be fully developed.
Effect of Interstitial Shales on the SP. Increasing
amounts of shale or clay in a permeable bed effectively
result in a reduction of the SP curve. At the limit, for
100% shaliness, E, becomes zero; that is, the sand is
then all shale and indistinguishable from the surrounding
The presence of oil in a shaly sand tends to enhance
the effect of the shale. All other conditions being the
same, the total E, of a shaly sand will be smaller if oil
bearing than if water bearing.
The effect of interstitial shale is also greater in low-
porosity formations. In these cases, only a small amount
of shale reduces the SP deflection appreciably. Con-
versely, the E, of shaly water-bearing sands of high
porosity remains practically equal to the E, of a clean
sand, as long as the shale content is reasonably
low-i.e., does not exceed a few percent.
Geometric Effect Influencing the SP Curve
Circulation of the SP Current, The various EMFs add
their effects to generate the SP currents, which follow
the paths represented schematically in Fig. 49.7 (right)
by solid lines. Each current line encircles the junction of
mud, invaded zone, and uncontaminated zone. In the
usual case where the formation waters are saltier than the
mud, E, is positive and the current circulates in the
direction of the arrows. The potential of a point in the
mud column opposite the sand is negative with respect to
one opposite the shale.
Along its path, the SP current forces its way through a
series of resistances, both in the ground and in the mud.
Along a closed line of current flow, the total of the
ohmic-potential drops is necessarily equal to the
algebraic sum of the EMFs encountered. Moreover, the
total potential drop is divided between the different for-
mations and the mud in proportion to the resistance of
the path through each respective medium.
Static SP (Clean Formations) and Pseudostatic SP
(Shaly Formations). It is convenient to use an idealized
representation in which the SP current is prevented from
flowing by means of insulating plugs placed across hole
and invaded zones, as shown in Fig. 49.7a (right).
Under these conditions, a plot of the potential in the mud
column would appear as the dashed cross-hatched curve
on the left of Fig. 49.7a, with a maximum negative
deflection opposite the permeable bed equal to the
algebraic sum of all the EMFs of various origins. This is
the maximum SP that could be measured. It is therefore
convenient to use this theoretical value as a reference. In
the case of a clean sand, it is called the static SP, ESp. If
the sand is shaly, it is called the pseudostatic SP. Epsp.
For given values of the activities of mud and formation
water, the pseudostatic SP of a shaly sand is smaller than
the static SP of a clean sand. The ratio E,rplEsp is
called the reduction factor or ratio and is designated by
the symbol (Y sp.
The SP log records only that portion of the potential
drop occurring in the mud. When the bed is sufficiently
thick the amplitude of the SP deflection approaches the
static SP (or EpsP in case of shaly formations), hccause
then the resistance offered to the current by the bed itself
is negligible compared with the resistance of the path
through the mud in the borehole.
Factors Influencing the Shape and Amplitude of SP
Deflections. As seen in Fig. 49.7b, the current circulates
in the hole not only opposite the permeable formation but
also a short distance beyond its boundaries. As a result,
although on the static SP diagram the boundaries of a
permeable bed are indicated by sharp breaks, those on
the actual SP curve show a more gradual change in
potential. An analysis of the circulation of the current
shows that, for uniform resistivity in the formations, the
bed boundaries are located at the inflection points on the
SP log. This fact provides a means of determining the
thickness of a bed from the SP log.
Both the shape of the SP deflection and its relative
amplitude (in fractional parts of the Essp or EpsP) are in-
fluenced by four factors, which determine the conditions
for the circulation of SP currents: (1) bed thickness, (2)
resistivities of the bed, the adjacent formations, and the
Fig. 49.7-a. Static SP diagram (left) that would be
observed in hole when current IS prevented
from flowing by means of insulating plugs
(right); b. actual SP diagram (solid curve,
left) and schematic representation of SP
current distribution in and around perme-
able bed (right).
mud, (3) borehole diameter, and (4) depth of invasion.
All other factors remaining the same, a change of the
total EMFs affects the amplitude but does not modify
the general shape of the SP log.
Influence of Mud Resistivity and Hole Diameter. The
mud resistivity has a predominant influence on the SP
curve. If the mud is of about the same degree of salinity
as the formation water, electrochemical EMFs are
small. If the mud is more saline than the formation
water, the SP may be reversed (sand deflections toward
the positive side of the log). Moreover, the lower the
mud resistivity (compared with the formation resistivity)
the broader the deflection above and below the
permeable bed and, because the ohmic drops in the mud
are decreased, the smaller the amplitude of the
An increase in hole diameter acts approximately like
an increase in the ratio of formation resistivity to mud
resistivity. It tends to round off the deflections on the SP
log and reduce the amplitude of the deflections opposite
thin beds. A decrease in hole diameter has the same ef-
fect as a decrease in the ratio of formation resistivity to
mud resistivity.
The SP log would also be influenced by a lack of
homogeneity of the mud-a change in salinity of the mud
the permeable sections. In between, they flow through
the mud rather than through resistive sections close to the
borehole, because of the large resistances the latter paths
offer. However, within the formation at a distance from
the borehole, where the paths through the resistive beds
have larger cross sections and hence lower resistances,
the SP currents can complete their circuits from
permeable beds to shale. They cannot return to the mud
through adjacent permeable beds because there they en-
counter EMFs opposing them.
Opposite a given resistive bed, the SP current in the
mud column remains essentially constant along the
borehole. This means that the potential drop per unit
length of hole is also constant,-thus giving g constant
slope on the SP log as shown by the straight-line portions
of the SP in Fig. 49.8. At the level of each conductive
bed, some SP current will enter or leave the mud col-
Fig. 49.8-SP phenomena in highly resis-
tive formations (schematic).
at a certain level would result in an SP baseline shift at
that level. However, it has been found in practice that
such changes in salinity are rare.
Effect of Invasion. Permeable beds in general arc invad-
ed by mud filtrate. Because the boundary between mud
filtrate and interstitial water is somewhere inside the for-
mation, a fraction of the SP current flows directly from
the shale into the invaded zone, without penetrating the
mud column. As a result, the presence of the invaded
zone has an effect on the SP log similar to that of an in-
crease in hole diameter.
SP in Soft Formations. Theory and field experience
have shown that the amplitude of the SP deflection is
practically equal to the static SP (of a clean sand) or to
the pseudostatic SP (of a shaly sand) when the permeable
beds are thick and the resistivities of the formations are
not too great compared with that of the mud. Moreover,
the SP curves define the boundaries of the bed with great
accuracy. The amplitude of the deflection is less than the
static SP or pseudostatic SP for thin beds, and the thinner
the bed, the smaller the deflection.
On the other hand, when the resistivity of the forma-
tion, R,, is considerably greater than that of the mud,
R,, the SP curves are rounded off, the boundaries are
marked less accurately, and all other conditions being
the same, the amplitude of the peak is less than when the
ratio R,I R, is close to unity.
For the case of shaly sands, the SP curve may also be
affected by the presence of oil. A change in the
magnitude of the SP deflection occurs very often when
passing an oil/water contact in a shaly sand. This change
is not a positive criterion for the detection of oil because
the same effect would be obtained if the salinity of the
interstitial water were reduced or if the percentage of
shale were increased.
umn, thus modifying the slope of the SP log. For in-
stance, the slope of the SP log changes at the level of the
permeable bed, P2, because part of the current leaves the
hole and flows into the bed. *
As a general rule, in hard formations the permeable
beds are characterized on the SP log by slope changes or
curvatures that are convex toward the negative side of
the log. Shales are characterized by curvatures that are
convex toward the positive side of the log. Highly
resistive beds correspond to essentially straight parts of
the SP log.
Determination of Static SP (SSP). The SP deflection is
measured with respect to the shale baseline, a reference
line which can generally be traced along the extreme
positive edges of the SP curve. Usually the shale line is
straight and vertical. *
In any given well, since the mud salinity is constant
and the interstitial waters may tend to be constant, there
is often a definite tendency for the maximum SP deflec-
tions to be the same for the same types of permeable for-
mations at comparable depths. Thus, it is usually possi-
ble to draw, parallel to the shale line, a sand line on the
log along the maximum negative deflections of the clean
sands of sufficient thickness.
It is very likely that, for all the beds where the SP
peaks reach the sand line, (1) the formation-water
resistivity is practically the same, (2) the beds are vir-
tually free from shaly material, and (3) the amplitude of
the deflection is equal to the SSP. For thin beds in cases
where the SSP cannot be determined as above (or for a
thin shaly sand), the SP reading from the log must be
corrected by means of appropriate charts in order to ob-
tain the Essp or Epsp. 2
Determination of R, from SSP
Since the variations of electrotiltration potential from
sand to shale can generally be neglected, the SSP is taken
in practice as equal to the corresponding value of -EC as
long as the SP is fully developed.
It is convenient to replace Eq. 8 by
SP in Hard Formations. Hard formations are highly
resistive except for permeable beds, whether oil- or
water-bearing, and shales, which are impervious. The
Essp =-Kc log+ . . . . . . . . . . .
. (9)
SP currents generated by the different EMFs flow into
field experience has shown that in certa#n regions there may be shifts of the shale
the hole out of the shale sections and out of the hole into
line. Sometimes rhese shifts are found systemattcatty at the ?.ame places in the
geologlcal column and can be used as markers.
where R, is an equivalent formation-water resistivity .
The computation of R, is given in the chart of Fig.
49.6, and R, is derived from R, by means of the aux-
iliary chart at the lower right of Fig. 49.6. The solid
curves on this auxiliary chart correspond to highly saline
formation waters, where the presence of salts different
from NaCl is negligible in practice. They are derived
from the known activity/resistivity relationships for pure
NaCl solutions. The dashed curves correspond to forma-
tion waters of low salinity, where the presence of other
salts (calcium and magnesium chlorides, sulfates, and
bicarbonates) have an important bearing on the activity
values. These curves are derived from empirical obser-
vations and cover formation waters of average composi-
tion. I9 Note that, for intermediate salinities (0.08<
R, ~0.3 at 75F), the value of R,, is practically equal
to R,.
The mud filtrate is taken here as an NaCl solution, and
this is generally done in practice, except for muds con-
taining gypsum, CaC12, or NaOH. In such cases, the
determination of R, from the SP curve requires the
measurement of the activity of the mud. A field instru-
ment is provided for this purpose.
Resistivity Logging Devices*
A general classification of the types of rcsistivity logging
devices was given previously.
Electrical Survey (ES). During the first 25 years of log-
ging practice, the standard ES (Fig. 49.2) usually includ-
ed, in addition to the SP, three conventional (unfocused)
resistivity curves; namely, a short normal curve
(distance between electrodes A and M is 16 in.), a long
normal (AM=64 in.) or a short lateral (distance be-
tween electrodes and A and 0 is 6 to 9 ft), and a stan-
dard lateral (AO= 18 ft, 8 in. in general), all recorded
simultaneously. In some regions, such as the Permian
basin (west Texas and New Mexico), the short-normal
spacing was reduced to 10 in., and the limestone sonde
was recorded instead of the long normal. The ES log is
rarely run today, but it was the standard log for many
Induction-Electrical Surveys (IES) (Figs. 49.9 and
49.10). The simultaneous recording of induction (con-
ductivity and resistivity) curves, 16-in. normal, and SP
curve, is a good combination for the logging of fonna-
tions of low to moderate resistivities in fresh muds. Of
late, the 16-in. normal has been replaced by a focused
electrode system, and two induction logs of different in-
vestigations may also replace the single induction.
Focused Electrode Devices. In wells drilled with very
saline mud, or in high resistivity formations, a laterolog
or dual laterolog is used with a gamma ray tool. Fullest
benefit of these combinations usually is derived if a
microresistivity survey is also run. Microresistivity
surveys generally include a microcaliper curve (hole-
diameter recording) (see Figs. 49.9 and 49.10). To avoid
multiple runs, many of the above devices are combined
with porosity logs-acoustic, density, and neutron logs.
These porosity logs are discussed in other chapters.
See Table 46.6 for the names of the various service companies 1ools.
49-l 1
Bed Thxkncrr, Feel
1 1 I 1
I 2 3456 6 IO 20 30 4050
------+ 1
Bed Thdner!, Feet 1
2 3 4 56 6 IO 20 M 4050
Fig. 49.9-Shoulder-bed corrections, LLS (top) and LLD
Flg. 49.10-Principle of Delaware effect.
T-- - A
Fig. 49.11 -Normal device
Fig. 49.12-Lateral device (schematic).
0 2 4 6 8 IO
Fig. 49.13-Laboratory curves for normal
sonde of spacing AM =2d
through uninvaded beds
more resistive than adjacent
Conventional Resistivity Devices
During the first quarter century of well logging, the only
electrical surveys available were the conventional
resistivity logs plus the SP. Thousands of them were run
each year in holes drilled all over the world. Since then,
new logging methods have been developed to measure
values much closer to R, and R,, which are the values
sought. Nevertheless, the conventional ES (consisting of
SP, 16-in. normal, 64-m. normal, and 18-ft 8-in. lateral)
still is being run in some parts of the world. For this
reason, and also because new information can often be
obtained by reinterpreting old ES logs, this chapter in-
cludes discussion on the principles and responses of the
ES measurements.
Principles: Normal and Lateral Devices. In conven-
tional resistivity logging a current of known intensity is
sent between two electrodes, A and B (A on the sonde, B
on the sonde or at the surface), and the resulting potential
difference is measured between other electrodes M and
N. The apparent resistivity is proportional to the
measured potential difference.
For normal devices, the distance AM is small (1 to 6
ft) compared with MN, MB, and BN. In practice N or B
may be placed in the hole at a large distance above A and
M (Fig. 49.11). The voltage measured is practically the
potential of M (because of current from A), referred to
an infinitely distant Point. The distance AM of a normal
device is its spacing. The point of measurement is mid-
way between A and M.
For lateral devices, measuring electrodes M and N are
close to each other and located several feet below current
electrode A. Current-return electrode B is at a great
distance above A or at the surface. The voltage measured
is approximately equal to the potential gradient at the
point of measurement 0, midway between M and N.
The distance A0 is the spacing of the lateral device. The
two arrangements shown in Fig. 49.12 (in which current
and measuring electrodes are interchanged) are
equivalent as regards measured potentials (and
Curve Shapes-Laboratory Results. Fig. 49.13 shows
laboratory curves from a normal device for
homogeneous resistive layers between adjacent beds of
low resistivity. The curves are symmetrical with respect
to the center planes of the layers. The same curves are
recorded if M is above A instead of, as in the figure, A
above M.
The upper part of Fig. 49.13 shows a resistive bed
thicker than the spacing (bed thickness, h, is 6dh ; spac-
ing AM is 2dh ; where dh is the hole diameter). At the
boundaries of the bed the curve tends to be rounded off
owing to the influence of the borehole. Moreover, the in-
dicated bed thickness (distance between the inflection
points P and P) is less than the actual thickness. Normal
curves tend to show resistive beds thinner than they ac-
tually are (and conductive beds thicker than they actually
are) by an amount equal to the spacing AM. The error in
picking the boundaries of thick resistive beds is small for
short-spacing normals, which is one reason for the re-
cording of a short normal.
As shown in the lower part of Fig. 49.13, for a
resistive layer thinner than the spacing, the curve shows
01 5 io 15 20
Fig. 49.14-Laboratory curves for lateral
sonde of spacing A0 = 1 Id
through uninvaded beds
more resistive than adjacent
a depression opposite the layer with two symmetrical
small peaks, c and d, on either side. The main disadvan-
tage of the normal device is that beds thinner than the
spacing, no matter how resistive they may be, appear on
the log as being conductive.
Fig. 49.14 shows similar curves for a lateral. The
lateral curves are markedly dissymmetrical, and their
features are more complex. Again the transitions in the
curves at the boundaries have been rounded off by the ef-
fect of the borehole. When the bed is thicker than the
spacing, the upper boundary of the bed is not well de-
fined on the lateral curve, and, as a whole, the bed ap-
pears as being displaced downward by a distance equal
to the spacing AO.
In the lower part of Fig. 49.14 the lateral indicates a
resistive layer thinner than the spacing by a sharp peak of
relatively low apparent resistivity. Below the layer is a
low-resistivity blind zone, followed by a reflection
peak at a distance A0 below the bottom boundary of
the layer. The blind zone is recorded when the resistive
streak is located between the current electrode and the
measuring electrodes.
The lateral is useful for the location of thin, highly
resistive streaks, although interpretation may be difficult
if several resistive streaks are close together. A lower
streak located in the blind zone of an upper resistive
streak may be missed, and the reflection peaks may be
mistaken for actual resistive streaks in the formation. For
a resistive layer of thickness approximately equal to the
spacing (cnticul rhickness), the lateral is almost com-
pletely flattened.
Similar generalizations are possible for lateral curves
recorded for beds more conductive than the surrounding
formations. Whether the layer is thick or thin, the shape
of the curve is dissymmetrical and the anomalies are
spread downward, outside the bottom boundaries. The
apparent increase of bed thickness is roughly equal to
Normals and Laterals in Hard Formations. Fig.
Fig. 49.15--Responses of normals and
laterals in hard formations
49.15 shows schematically the behavior of the normals
and laterals in thick, highly resistive formations contain-
ing porous or shaly (that is, more conductive) zones. In a
highly resistive formation most of the current from elec-
trode A flows up or down the borehole, dividing in in-
verse proportion to the resistances of the two paths,
which are determined mostly by the resistance of the
mud column in the hole between the current electrode
and the nearest conductive beds. At the conductive beds,
depending on their thickness and conductivity, the cur-
rent has low-resistance paths from the hole. The lopsided
appearance of the normal and lateral curves is ex-
plainable in terms of the unequal division of current
flowing up and down the hole.
The normal, for example, has M and N above the cur-
rent electrode. The voltage measured is the ohmic poten-
tial drop in the hole resulting fmm current flowing in the
mud between M and N. When the device is near the bot-
tom of a resistive bed, most of the current flows down to
the conductive bed just below, and there is little potential
drop between M and N because the current up is small.
When the device has moved farther up in the bed, the
current down decreases because the resistance of that
path has increased. Also, since the resistance of the up-
ward path has decreased, the current up increases.
Therefore, the potential drop between M and N increases
as the device moves upward until electrode N reaches the
next conductive bed, where the upward current is
diverted from the hole. Above that level the normal
reading decreases.
The explanation of the shape of the lateral curve is
similar. The direction of the lopsidedness for either
device depends on whether the measuring electrodes are
above or below the current electrode. The depressions
read on the curves opposite the conductive beds are
smooth and, in the case of the lateral, much broadened
and displaced downward. Accurate determinations of
bed boundaries from the curves are practically
Fig. 49.16-Limestone sonde
Limestone Sonde. Four current electrodes (A, A, B,
and B), connected as shown in Fig. 49.16 by insulated
wires of negligible resistance, are symmetrically ar-
ranged so that AB=AB. A measuring electrode, M, is
placed in the middle of the device. Depths are measured
from electrode M. In practice AM = AM =30 or 3.5 in.,
and AB =AB =4 or 5 in. The device is therefore a sym-
metrical double lateral.
Opposite a thick, highly resistive layer (upper part of
Fig. 49.17) practically all the flow of current is confined
to the spaces between A and B and between A and B.
No current flows from B or B, up or down the hole away
from the device. Hence, from Ohms law, B and B are
at zero potential. Similarly M is at the same potential as
A and A.
The potential of M is, therefore, equal to the potential
drop in the mud, because of the flow of current, between
A and B (or A or B). As long as all the electrodes of the
devices are opposite the resistive formation this potential
difference is dependent only on hole size and mud
resistivity; if these are constant, a constant apparent
resistivity is recorded.
If the device is located just above a conductive streak
(as in the lower part of Fig. 49.17), the streak is effec-
tively a low resistance connecting adjacent portions of
the device to points at zem potential. Part of the current
now flows in the paths indicated by the arrows, and the
potential of electrode M is correspondingly decreased.
The conductive streak is indicated on the log by a
relatively sharp, symmetrical depression.
The limestone sonde gives clearer and simpler logs in
hard formations, but measurements with the limestone
sonde arc strongly affected by the mud column. When
the formations are much more resistive than the mud, the
readings are appreciably lower than the formation
Application of Conventional Resistivity Logs. The
three devices that are generally recorded in the conven-
tional electrical log (16-in. normal, 64in. normal, and
18-ft, 8-in. lateral) were designed to provide the most
complete information with a system of nonfocused
Fig. 49.17-Principle of limestone sonde
The short normal is well adapted for bed definition,
boundary determination, and correlation of formations of
low or moderate resistivities (sand-shale series). The
lateral generally shows sharp peaks at the level of thin
resistive beds, but the definition of these beds is often
obscured by blind zones and spurious peaks.
The precision of the normal and lateral curves for bed
definition is limited in hard formations and is quite poor
when salty muds are used. Somewhat better resolution is
obtained with the limestone sonde. In all cases, forma-
tion delineation is more detailed and accurate with the IL
and focused devices (LL) and with the microdevices. In
hard formations the 16-in. normal and the limestone
sonde can provide an approximation to the value of Ri
and hence an approach for formation factor evaluation.
The capabilities of the conventional tools for the deter-
mination of R, are discussed later in this chapter.
Induction Logging
The IL was first developed to measure formation
rcsistivity in boreholes containing oil-based muds. O
Electrode devices do not work in these nonconductive
muds, and attempts to use wall-scratcher electrodes
proved unsatisfactory. Experience soon demonstrated
that the induction tools had many advantages over the
conventional ES for logging wells drilled with water-
based muds _ 2
Induction logging devices are focused to minimize the
influence of the borehole and of the surrounding forma-
tions. They are designed for deep investigation and
reduction of the influence of the invaded zone.
Practical induction sondes include a system of several
transmitter and receiver coils. However, the principle
can be understood by considering a sonde with only one
transmitter coil and one receiver coil (Fig. 49.18).
High-frequency AC of constant intensity is sent
through the transmitter coil. The alternating magnetic
field thus created induces secondary currents in the for-
mations. These currents flow in circular ground-loop
paths coaxial with the transmitter coil. These ground-
loop currents, in turn, create magnetic fields that induce
signals in the receiver coil. The induced receiver signals
are essentially proportional to the conductivity of the for-
mations. Any signal produced by direct coupling of
transmitter and receiver coils is balanced out by the
measuring circuits.
The IL operates to advantage when the borehole fluid
is an insulator-even air or gas. But when properly
designed the tool also will work very well when the
borehole contains conductive mud, provided that the
mud is not too salty, the formations are not too resistive,
and the borehole diameter is not too large.
Four types of induction equipment are now in use.
1. The 6FF40 IES tool includes a six-coil induction
device of 40-in. normal spacing, a 16-in. normal, and an
SP electrode. The induction array provides the greatest
lateral depth of investigation presently available with in-
duction tools.
2. The 6FF28 IES is a small-diameter (2% in.) tool
for use in slim holes. It is a scaled-down version of the
6FF40, having a 28-in. coil spacing, and incorporates a
standard 16-in. normal and an SP.
3. The Dual Induction-Laterolog 8 (DILTM) or
Spherically Focused Log (SFL) system uses a deep-
reading induction device (ID, similar to the 6FF40), a
medium induction device (IM), an LL8 (or an SFL), and
an SP electrode. The IM device has vertical resolution
similar to that of the 6FF40 tool but about half the depth
of investigation. It is much more affected by large hole
diameters and/or salty muds. The DIL log, with its three
focused resistivity readings of different depths of in-
vestigation, is superior to the IES log for determination
of R, and R, in extreme ranges of invasion depths and
in cases of annulus.
4. The ISF/Sonic combination incorporates an ID
measurement similar to that from the 6FF40 tool, the
new ISF log, an SP curve that may be electronically cor-
rected for noise, a borehole compensated (BHC) sonic
log, and an optional gamma ray curve. Of late, the BHC
sonic tool can be replaced in this tool string by a com-
bination neutron/density device.
Log Presentation and Scales
The SP and/or gamma my curve are recorded in Track 1
for all tools; they can be recorded simultaneously with
ISFisonic equipment. A gamma ray curve may also be
run with 6FF40 or DIL equipment.
Fig. 49.19 illustrates the standard IES presentation.
The induction conductivity curve is sometimes recorded
over both Tracks 2 and 3. The linear scale is in
millimhos per meter (mu/m), increasing to the left. In
Track 2 both the 16-in. normal and the reciprocated in-
duction curves are recorded on the conventional linear
rcsistivity scale.
The DIL introduced the logarithmic grid for resistivity
presentations. The current form is the log-linear grid
shown in Fig. 49.20. In this, the resistivity curves on the
detail log (5 in. I100 ft) have a split 4-decade logarithmic
scale. On the correlation log (1 or 2 in./100 fi), the scale
is linear. This presentation offers several advantages
over the other alternatives. The detail log has good
readability in low resistivities, a wide range without
backup traces, and the ease of reading resistivity ratios
Fig. 49.18-Induction-logging equipment
ml l l l vol t r
ohm, m' m m, l l , mhos m . $&
I .
Fig. 49.19-Induction-electrical log presentation.
mi l l i vol t s
: ::::,:
i lid :
.-.- _ _-*.
L!-8 _: * :.;..:
v; 1,
)>+.h i
3-N -!
: :
,:. :
:: :
j y;*
: ::
: 7:
I.. ::
,:: :
:. :
Fig. 49.21~ISFkonic presentation.
Fig. 49.20-Dual Induction Laterolog 8 presentation.
if it were from a homogeneous medium. A secondary
skin-effect correction may be required when the media
surrounding the sonde are not of uniform conductivity.
Such corrections are incorporated in various interpreta-
tion charts.
directly from the logarithmic scale. The linear scale is
more easily correlated with earlier logs. This format has
gained acceptance as the standard for resistivity logs.
The ISF log in combination with the sonic log requires
a modification of this grid usage because Track 3 is
needed for the sonic At curve. The grid selected is shown
in Fig. 49.21.
Skin Effect
In very conductive formations the induced secondary
currents are large, and their magnetic fields are impor-
tant. The magnetic fields of their ground loops induce
additional EMFs in other ground loops. This interaction
between the loops causes a reduction of the conductivity
signal recorded by the induction log. This signal reduc-
tion is known as skin effect.
Induction logs usually are automatically corrected for
skin effect during recording. The correction is based on
the magnitude of the uncorrected tool response treated as
Geometrical Factor
When conductivities are not high, skin effect may be
neglected, and the response of induction logs can be
described in terms of conductivities and geometrical
factors of the volumes surrounding the tool. The
geometrical factor, G, of a volume having a specific
geometrical orientation with the sonde is simply the frac-
tion of the total signal that would originate with that
volume in an infinite homogeneous medium. For com-
putation of geometrical factor to be practical, it is
necessary to assume that the volumes conform to sym-
metry of revolution about the sonde.
The magnitude of the signal in conductivity units is the
product of the geometrical factor and the conductivity of
the material, and the total signal sensed by the tool is the
-No Skin Effect
----Skin Effect Included: Case of
-0 40 00 120 160 200 240 280 320 360
49.22-Geometrical factor. Dashed curve includes skin effect under conditions shown, for the
ILd device.
sum of these products for all volumes within range
(which extends to infinity, but can be circumscribed to
practical limits). Since the Gs add up to unity by defini-
tion, this can be stated:
CtL=CtG, +C2G2+C3G3...+CNGN, . . . (10)
where C and G refer to the zones of differing conductivi-
ty and N is the total number of such zones.
The chief significance of this concept is the fact that a
volume of space defined only by its geometry relative to
the sonde has a fixed and computable geometrical factor.
This permits the construction of mathematically sound
correction charts to account for the effects of borehole
mud, the invaded zone, and adjacent beds on the R,
measurement, providing symmetry of revolution exists. 2
These charts incorporate the secondary skin effect cor-
rection mentioned above.
Invasion Effects
Fig. 49.4 illustrates an invaded formation. It includes
volumes having several conductivities, C,, C,, Ci,
and C, (corresponding to R,, R,,, Ri, and R,). The
total conductivity signal, CT, received from this zone by
the induction tool is
C,=C,G,+C,G,+CiGi+C,G,. . . . . . .(ll)
If the zone were infinitely thick, this would be the only
signal received, and CT = C,L. If the tool is a 6FF40, the
hole size moderate, and the mud relatively fresh, the
borehole signal is negligible, and the C,, and Ci zones
can be merged into one for this example.
If a moderate diameter of invasion, say 65 in., is
assumed, Fig. 49.22 reveals that the geometrical factor
of all material within the 65in. diameter is 0.2. If R,r, is
taken equal to 4 R,, then C,, =C,/4, and the induction
tool response is
=(C,/4) (0.2)+C,(O.S),
=0.85 C,.
In the same conditions, but using salty mud so that
R,, =R,i4, the response is
GIL =4 C,(O.2) + C, (0.8),
=1.6 C,,
which illustrates the conductivity-seeking character-
istic of the induction devices, and shows why they must
be used with discretion in salt-mud environments. As a
rule of thumb, R, should be less than about 2.5 R,, and
di (diameter of invaded zone) no greater than 100 in. for
satisfactory R, determination from 6FF40-type induction
In oil-bearing formations of low S, and high permeabili-
ty, an annulus of low resistivity , R, , may exist between
the flushed zone, R,,, and the virgin formation. When
R, is greater than R,, R,, is less than R,, and the ef-
fects of the two on the induction log tend to cancel.
However, the high conductivity of the annulus has more
effect in medium invasion ranges (2dh < di < 4 or 5dh),
and it may cause a single ID to read resistivities lower
than either R,, or R, . The DIL 8 tool is often capable of
detecting the presence of annuli, since in these cir-
cumstances the IM measurement reads lower than either
the LL8 or the ID values,
Thin Bed Corrections
The skin-effect correction accomplished automatically in
the induction tools assumes infinitely thick beds. Skin
effect in thin beds may require additional corrections,
and these are provided in Ref. 2.
Fig. 49.23-Schematic of focusing-electrode devices.
Borehole Corrections
Conductivity signals from the mud can be evaluated us-
ing geometrical factors. Chart Rcor-4 (Ref. 2) gives cor-
rections for the various induction tools and standoffs. On
the basis of bit size the nominal borehole signal is
sometimes removed from the recorded log. When the
hole signal is significant, log headings should always be
consulted to ascertain whether this was done. This
precaution applies most frequently to the IM measure-
ment of the DIL tool.
For hole diameters in the range of 7 to 13 in. there is
an uncertainty of about f0.0003 on the geometrical fac-
tor of the borehole for the 6FF40 sonde. This results
from several factors, including diameter and shape of the
borehole, mudcake thickness, standoff, and sonde tilt.
To preclude the possibility of cumulative errors ex-
ceeding 20% of the 6FF40 reading, the tool should not
be used where the resistivity to be measured is greater
than about 500 R,.
Very Resistive Formations
There is an uncertainty of about 52 mO/m on the zero of
the present induction sondes (6FF40, ID, and IM), and
consequently the resistivity error may be great as con-
ductivity approaches zero. To preclude an error of more
than 20%, the formation conductivity should be greater
than 10 mu/m (i.e., the resistivity less than 100 Q.m).
This error can sometimes be practically eliminated by
downhole calibration techniques if suitable formations
are present.
Primary calibration is performed by placing a test loop
around the sonde. The conductive loop has a resistance,
which has been adjusted to produce a certain conductivi-
ty signal in the sonde. An additional calibration pro-
cedure has a signal produced internally in the sonde to
adjust the control-panel sensitivity for proper galvanom-
eter deflections. The zero errors of the electronics in
the equipment arc also checked and balanced out.
Calibrate tails, usually attached to the log, serve as a
record of the calibration tests made before and after the
logging run.
In some regions it is possible to check the calibration
of the IL by observing that the conductivity reading op-
posite an impervious formation of exceedir,gly high
resistivity (such as anhydrite) represents the sum of all
spurious signals. If the hole diameter is known, it is then
possible to correct the IL reading so that the range of
uncertainty is reduced and greater accuracy is obtained in
formations of practical interest.
1. The IL can be used most effectively in holes filled
with moderately conductive drilling muds, nonconduc-
tive muds, and in empty holes.
2. Vertical focusing is good, making possible reliable
evaluation of beds down to about 5 ft thick with 6FF40,
ID, and IM devices, and down to about 3% ft thick with
6FF28 tools.
3. The deep induction logs (ILd) are only moderately
affected by invasion in relatively fresh muds, and good
R, determinations are possible where R, is less than
about 2.5 R, and di is less than 100 in.
4. The three curves of the DIL give more precise
knowledge of invasion profiles and hence better R,
values in the cases of deep invasion or annulus.
5. The log-linear presentation of the DIL and other
ILs meets most log requirements better than alternative
Focused-Electrode Logs
The responses of conventional ES can be greatly affected
by the borehole and adjacent formations. These in-
fluences are minimized by a family of resistivity tools
that use focusing currents to control the path taken by the
measure current. These currents are supplied from
special electrodes on the sondes.
The focused-electrode tools include the Latero!ogs (LL)
and SFLs. These tools are much superior to the ES
devices for large R&R, values (salt muds and/or highly
resistive formations) and for large resistivity contrasts
with adjacent beds (RJR, or R,IR,). They are much bet-
ter for resolution of thin to moderately thick beds.
Focusing-electrode systems are available wit! deep,
medium, and shallow depths of investigation.
Devices using this principle have as quantitative ap-
plications the determination of R, and of R,. The R,
tools are Laterolog 7 (LL7), Laterolog 3 (LL3), and
LATERLOG 0,0,=32=4d,A,A,=80=10d,l
AC,= 18% = 28d
Fig. 49.24--Response of Laterolog 7 and ES opposite a thin, resistive, noninvaded bed, with very-Sal-
ty mud (laboratory determination).
LLD of the deep dual laterolog. The medium-to-
shallow-reading devices, all integral with combination
tools, are Laterolog 8 (LL8) of the DIL, LLS of the
shallow dual laterolog, and the SFL of the ISFisonic
Laterolog 7. This deviceI comprises a center electrode
Ao, and three pairs of electrodes: M 1 and Mz; M 1 and
Mz; and Al and A2 (Fig. 49.23). The electrodes of
each pair are symmetrically located with respect to A0
and are connected to each other by a short-circuiting
A constant current 1, is sent through electrode Ao.
Through bucking electrodes A, and AZ, an adjustable
current is developed; the bucking current intensity is ad-
justed automatically so that the two pairs of monitoring
electrodes, M , and Mz and M , and M 2, are brought
to the same potential. The potential drop is measured
between one of the monitoring electrodes and an elec-
trode at the surface (i.e., at infinity). With a constant IO
current, this potential varies directly with formation
Since the potential difference between the M 1 -Mz
pair and the M I -M 2 pair is maintained at zero, no cur-
rent from A0 is flowing in the hole between M I and M 1
or between M2 and M2. Therefore, the current from
A0 must penetrate horizontally into the formations.
Fig. 49.23 shows the distribution of current lines when
the sonde is in a homogeneous medium; the sheet of
10 current, indicated by the hatched area, retains a fairly
constant thickness up to a distance from the borehole
somewhat greater than the total length A, A2 of the
sonde. Experiments have shown that the sheet of IO cur-
rent retains substantially the same shape opposite thin
resistive beds. The thickness of the IO current sheet is 32
in. (distance 0 I 02 on Fig. 49.23), and the length A I A2
of the sonde is 80 in.
Fig. 49.24 compares the curves obtained experimen-
tally opposite a thin resistive bed using the conventional
devices (16 and 64-in. normals and 18-ft, g-in. lateral)
with the corresponding LL7 recording. The conventional
devices give poor results; the LL7, in spite of difficult
conditions (RJ R,,, is 5,000), shows the bed very clearly
and reads close to R,. An SP curve may be recorded on
depth simultaneously with the LL7.
Laterolog 3
Like LL7, LL3 also uses currents from bucking elec-
trodes to focus the measuring current into a horizontal
sheet penetrating into the formation. However, as seen in
Fig. 49.23, large electrodes are used. Symmetrically
placed, on either side of the central A0 electrode, are
two very long (about 5-ft) electrodes, Al and AZ, which
are shorted to each other. A current, 10, flows from the
A0 electrode whose potential is fixed. From Al and A2
flows a bucking current, which is automatically adjusted
to maintain A, and A2 at the potential of A0 All elec-
trodes of the sonde are thus held at the same constant
potential. The magnitude of the IO current is then pro-
portional to formation conductivity.
The IO current sheet is constrained to the shaded, ap-
proximately disk-shaped area in Fig. 49.23. The
thickness, 0, 02, of the IO current sheet is usually about
12 in., much thinner than for LL7. As a result, LL3 has a
better vertical resolution and shows more detail than
LL7. Furthermore, the influences of the borehole and of
the invaded zone are slightly less.
The simultaneous recording of an SP curve is possible,
but the SP has to be displaced in depth, usually by about
25 ft, because of the large mass of metal in the sonde.
However, a gamma ray curve is normally run
simultaneously with the LL3 for lithology definition,
since the SP has very little character in the salt muds
where the LL is used. There is also available a
simultaneous LL3-neutron/gamma ray combination tool.
a- -
Fig. 49.25-Schemalic of the dual laterolog.
Guard-Electrode Device. In the guard-electrode
system, the surveying current flows into the adjacent for-
mations from a measuring electrode disposed between
relatively long upper and lower guard electrodes from
which current also flows. The guard electrodes tend to
confine the current from the measuring electrode to a
generally horizontal path. The measuring and guard elec-
trodes are connected through a very low impedance, as
necessary to measure the surveying current supplied to
the measuring electrode.
A resistivity value is obtained by recording the ratio of
the voltage of an electrode in the assembly (referred to a
distant point) to the current emitted from the measuring
electrode. The guard-electrode device is used mostly in
hard-rock territories for detailed bed definition, correla-
tion, and as a help in reservoir evaluation. For the deter-
mination of R, it is preferable that R,fIR, be small (less
than 4), as in the case of salty muds.
Laterolog 8. The shallow-investigation LL8 device is
recorded with small electrodes on the DIL sonde. It is
similar in principle to LL7 device except for its shorter
spacings. The thickness of the 10 current sheet is 14 in.,
and the distance between the two bucking electrodes is
somewhat less than 40 in. The current-return electrode is
located a relatively short distance from Ao. With this
configuration, the LL8 tool gives sharp vertical detail,
and the readings are more influenced by the borehole and
the invaded zone* than arc those for LL7 and LL3. The
LL8 data are recorded with the DIL on a split 4-decade
logarithmic scale.
Dual Laterolog. Since the measure current of an LL has
to traverse mud and invaded zone to reach the undis-
turbed formation, the measurement is necessarily a com-
bination of effects. With only one resistivity measure-
ment, the invasion profile and R,, had to be known or
Thick Beds
8 Ho
--- RX,= 0.1 Rt
- Rx0 Rt
8 20 40 60 80
Diameter d i 1 inches)
Fig. 49.26-Radial pseudogeometrical factors, fresh muds
(solid) and salty muds (dashed).
estimated in order to calculate R, . The need for a second
measurement at a different depth of investigation
resulted in the dual laterolog/gamma ray tools (Fig.
One version of the tool records the two laterologs se-
quentially; another does it simultaneously and has added
a shallow MICROSFL (MSFL) for R,, information. Both
can record a gamma ray curve on depth, simultaneously
with the resistivity curves. An SP can also be run.
By use of effectively longer bucking electrodes and a
longer spacing, the LLD (deep latemlog) has been given
a deeper investigation than either LL7 or LL3 devices.
The LLS (shallow laterolog) uses the same electrodes in
a different manner (Fig. 49.25 right) to achieve a current
beam equal in thickness to that of LLD, 24 in., but hav-
ing a much shallower penetration. The LLS depth of in-
vestigation lies between those of the LL7 and LL8
devices (Fig. 49.26).
Spherically Focused Log. The SFL log is part of the
ISF/sonic combination, and it was developed as an im-
provement over both the 16-in. normal and the LL8 as a
short-spacing companion to the deep induction log.
Normal resistivity devices rely on the concept of equal
intensity of current radiation in all directions, as would
happen in a homogeneous isotropic medium. When the
current distribution is distorted from the spherical model,
as by the presence of a borehole, the readings must be
corrected by departure curves. The SFL device uses
focusing currents to enforce an approximately spherical
Fig. 49.27-Laterolog recorded on hybrid scale.
shape on the equipotential surfaces over a wide range of
wellbore variables. The borehole effect is virtually
eliminated where dh 5 10 in., yet most of the response is
from the invaded zone in all but extreme conditions.
Scales. The linear resistivity scales originally used for
LL data were poorly adapted to record the wide range of
measurements characteristic of these tools. Although
linear scales are still used occasionally, compressed
scales of either the hybrid or logarithmic type have sup-
planted the linear for quantitative work.
The hybrid scale, first used on the LL3 log, presented
linear resistivity on the first half of the grid track and
linear conductivity on the last half. Thus, one
galvanometer could record all resistivities from zero to
infinity. (See Fig. 49.27.)
The logarithmic scale was used first with the dual in-
duction tool, and it has also been adapted for the LL and
the SFL devices (Fig. 49.28). It combines readability
and detail in low msistivities with a wide range of
values, and it also offers the advantage of graphic
(quick-look) interpretations.
Influence of Wellbore Variables
These logging devices can be significantly affected by
the borehole mud, the invaded zone, and adjacent beds.
Charts Rcor-1 and Rcor-2 of Ref. 2 provide needed cor-
rections. Where only one measurement is available,
some knowledge or assumption of depth of invasion
must be used for deriving R,. Readings must be cor-
rected for borehole effect before the shoulder-bed charts
are entered. (See Figs. 49.29 and 49.9.) These figures
J --I
: :
. I
: r---
I<- -L-
: :
-1. ._:

: :
: i I
.: ::I
. ._.~ --.:.
.: ::;f
: :
z; ;r;i
fT ::Y
. .
7 ::::
;- .ii
i j
: :
: :
::.:I :
:: ::: :
,:..: :
T I :-j ;
1- i
T..... -:
I . :
T r..... f
: : :::: :
L.... .
! I..!
Fig. 49.29-Laterolog over same interval as Fig. 49.22, record-
ed on logarithmic scale.
Fig. 49.29--Shoulder-bed corrections, LL3 (top) and LL7
give R, /R, as a function of bed thickness, R, being the
formation waters-and (2) to provide correlation and R,,
corrected resistivity and R, the apparent resistivity cor-
determinations in conjunction with deeper-reading R,
rected for borehole effect.
Pseudogeometrical Factors
Microresistivity Devices
Geometrical factor may be defined as that fraction of the
total signal that would originate from a volume having a
specific geometrical orientation with the sonde in an in-
finite homogeneous medium.
The only well-logging devices for which this concept
is sound are the induction tools, because only with these
is the measuring geometry independent of variations in
RJ R,. However, it is useful to construct charts based
on pseudogeometrical factors for other resistivity
devices, for purposes of comparative evaluation. Such a
chart is shown in Fig. 49.26 in which the integrated
pseudogeometrical factors of progressively larger
cylinders are plotted vs. the diameters of the cylinders.
The apparent resistivity, R,, measured in a thick bed is
given approximately by
Microresistivity devices are used to measure R,,
(resistivity of the flushed zone) and to delineate
permeable beds by detecting the presence of mudcake. In
the discussion that follows, the micmlog (ML) will be
treated in considerable detail, not because of its relative
importance-the micmlaterolog (MLL), the proximity
log (PL), and the MSFL are superior tools for obtaining
R,,-but because its principle is fundamental, and it is
still the best of the three microresistivity devices for
delineating permeable-bed boundaries, hence for making
sand counts.
Measurements of R,, are important for several
reasons. When invasion is moderate to deep, knowledge
of the R, value makes possible more accurate deter-
minations of true resistivity and hence of saturation.
Also, some methods for computing saturation are
entered with the ratio RJ R,. Also, in clean formations,
a value of F may be computed from R,, and R,,f, if S,r,
is known or estimated. From F, a value for porosity may
be found.
R,=R,,G,i+R,(l-G,;), . . . . (12)
where G,; is the pseudogeometrical factor. It must be
emphasized that a pseudogeometrical factor relating to
an electrode-type resistivity device is applicable in only
one set of conditions, and therefore charts of this type are
not valid as general-purpose invaded-zone correction
charts. The most useful feature of the Fig. 49.26 chart is
its graphic comparison of the relative contribution of in-
vaded zones to the responses of the various tools.
The Delaware Effect
If both B and N electrodes are placed downhole as in
Fig. 49.10, LL data may exhibit Delaware effect* (or
gradient) in sections located just below thick noncon-
ductive beds such as anhydrite. It appears as abnormally
high resistivity for 80 ft or so below the resistive bed.
The LL3 is the only field tool now using this
Fig. 49.10 illustrates the effect and its causes. As B
enters the thick anhydrite, the current flow is confined to
the borehole, and if the bed is thick enough (several hun-
dred feet) practically all the current will flow in that part
of the hole below B. Then when N enters the bed, it can
no longer remain near zem potential as intended. It is ex-
posed to an increasing negative potential as it rises far-
ther from the bed boundary. This potential appears at the
surface as an increase in the resistivity measurement.
LL7 and LLD devices normally use surface electrodes
for current return. so they are not subject to Delaware ef-
feet. However, a small anti-Delaware effect has been
observed, which produces resistivities that are too low
just below the resistive beds.
To measure R, a shallow-investigation tool is
desirable, since the R,, zone may sometimes extend on-
ly a few inches beyond the hole wall. Also, the reading
should preferably not be affected by the borehole.
A sidewall-pad tool is indicated. The pad, carrying
short-spacing electrode devices, is pressed against the
mudcake, thus reducing the short-circuiting effect of the
mud. Currents from the electrodes on the pad must pass
through the mudcake to reach the R,, zone.
Microresistivity readings are more or less affected by
mudcake, depending on its resistivity, R,,, and
thickness, h,, Moreover, mudcakes can be anisotmpic,
with mudcake resistivity parallel to the borehole wall
less than that across the mudcake. Mudcake anisotropy
increases the mudcake effect on micmresistivity
readings, so that the effective, or electrical, mudcake
thickness is greater than that indicated by a caliper. The
micmresistivity tools incorporate two-arm calipers,
which show the size and condition of the borehole.
Present equipment includes a combination tool with two
pads mounted on opposite sides. One is the ML pad, and
the other may be for either the MLL or the PL, as re-
quired by mudcake conditions. The measurements are
recorded simultaneously. The MSFL is a combination
tool, which can be run with either formation density or
dual latemlog equipment.
Conclusions Conclusions
Log Presentation Log Presentation
Resistivity devices with the focusing electrode principle Resistivity devices with the focusing electrode principle
The microresistivity logs are scaled, of course, in
The mirrnr&rtivitv lnuc nn- a-sled, of course, in
meet certain logging requirements better than other types meet certain logging requirements bette- +hon n+hpr +l~noc ~
resistivity units. When recorded by itself, the ML data itself, the ML data
now available. These requirements now available. These requirements are (1) to take
are usually recorded over Tracks 2 and 3 on a linear
measurements leading to determination measurements leading to determination of R, in condi-
resistivity scale. The microcaliper data are in Track 1.
tions for which the induction tool tions for which the induction tools are not well
The MML or PL is recorded on a four-decade
suited-i.e., R, values in excess of 100 suited-i.e., R, values in excess of 100 Q-m and/or mud
logarithmic scale on the right of the depth track (Fig.
resistivities of the same order or lower resistivities of the same order or lower than those of the
49.30). The caliper is recorded in Track 1. When the ML
So-called because 1 was ilrst observed ,n Ihe Delaware saw v, ,,,r Y So-called because 1 was ilrst observed ,n Ihe Delaware sand of the Delaware basin
data are also recorded, they are in Track 1 on a linear
(west Texas) This sand underltes a very thick anhydrite bed (west Texas) This sand underltes a very thick anhydrite bed scale. The MSFL data are also recorded on the
logarithmic grid. When run with the dual laterolog, it is
presented on the same film as the LL curves. With the
Compensated Formation Density (FD?) log, it must be
presented on a separate film, since the FDC log uses a
linear grid; the two logs arc normally recorded
With the microlog too1,23 two short-spacing devices
with different depths of investigation provide resistivity
measurements of a very small volume of mudcake and
formation immediately adjoining the borehole. They
readily detect the presence of any mudcake, indicating
invaded (hence permeable) formations.
Principle. Arms and springs press a rubber pad against
the hole wall. In the face of the pad are inserted three
small electrodes in line, spaced 1 in. apart. With these
electrodes a 1-X l-in. micminverse, R, Xl, and a 2-in.
micronormal, R2, are recorded simultaneously.
As drilling mud filters into the permeable formations,
mud solids accumulate on the hole wall, forming a mud-
cake. The resistivity of the mudcake is about equal to, or
slightly greater than, the resistivity of the mud. Mudcake
resistivity is usually considerably smaller than the
resistivity of the invaded zone near the borehole.
The 2-in. micronormal has a greater depth of in-
vestigation than the 1-x l-in. micminverse. It is
therefore less influenced by the mudcake and reads a
higher resistivity, producing positive curve separa-
tion. In the presence of the low-resistivity mudcake, both
devices will measure moderate resistivities, usually
ranging from about 2 to 10 times R,
I nterpretation. Positive separation in a permeable
zone is illustrated in Fig. 49.30, at Level A. The caliper
shows evidence of mudcake. However, quantitative in-
ferences of permeability are not possible from the ML
data. When no mudcake exists, for whatever reason, the
ML data may yield useful information as to borehole
condition or lithology , but the log is not quantitatively
Under favorable circumstances, R, values can be
derived from the ML by use of Fig. 49.31. R,, values
for this purpose can be measured directly or estimated
fmm Ref. 2, and h,, is obtained from the caliper.
Limitations of the method are (1) the ratio R,,lR,, must
be lower than about 15 (porosity more than 15%), (2)
h ,,,= must be no greater than /2 in., and (3) depth of inva-
sion has to be over about 4 in., otherwise the ML
readings are affected by R,. Eqs. 6 or 14 permit the
porosity derivation from the ML measurements. For
this, the value of S,, must be reasonably well known.
Mud Log
The ML sonde is lowered into the hole with arms closed.
Except in holes smaller than 8 in., the measuring pad
will randomly face away from the wall part of the time,
and its reading then will be determined mostly by the
mud. A recording of these readings, conveniently made
going down, serves as a mud log, on which the
lowest resistivities correspond to the upper limit of the
in-situ value of R,. This log has several potential ap-
plications, including crosschecking the surface R,
measurement, detecting mud-system changes, and iden-
tifying downhole water flows.
yll, II -
hrn, - r
,I 3 70 803 moo ?
Fig. 49.30-Presentation of proximity log-microlog
Ftg. 49.31-Interpretation chart for 8-in.-hole series
C microlog for which adjacent elec-
trodes are 1 in. apart. R, in designates
reading of 2-in. micronormal (AM, =2
in.) and RIx, ,, is I- by I-in. microin-
verse (AM, = 1 in., M, M, = 1 in.). Type
I hydraulic pad, noninsulated sonde.
Fig. 49.32-Microlaterolog pad showing electrodes (left) and
schematic current lines (right).
Microlaterolog (MLL)
Fig. 49.33-Comparative distribution of current lines of
Microlaterolog and Microlog.
On ML (Fig. 49.31) for values of RI O/R,, greater than
about 15, the curves for constant values of R,,lR,, arc
crowded; as a result, the accuracy of the determination
of R, from the ML is poor in this region. With the MLL
method, it is possible to determine R,, accurately for
higher values of R,,lR,,, provided, however, that the
mudcake thickness does not exceed X in.
Proximity Log (PL)
Principle. The MLL pad is shown in Fig. 49.32.24 A
small electrode, Ao, and three concentric circular elec-
trodes are embedded in a rubber pad applied against the
hole wall. A constant current, I,, is emitted through
electrode AO. Through the outer electrode, A I , is sent a
current automatically adjusted so that the potential dif-
ference between the two monitoring electrodes is main-
tained essentially equal to zero. The 10 current flowing
past the M 1 electrode cannot reach M2 and is forced to
flow in a beam into the formations. The current lines are
shown on the figure. The IO current near the pad forms a
narrow beam, which opens up rapidly at a few inches
from the face of the pad. The MLL reading is influenced
mostly by the formation within this narrow beam.
Principle. The proximity tool is similar in principle to
the MLL tool. The electrodes are mounted on a
somewhat wider pad, which is applied to the wall of the
borehole; the system is automatically focused by
monitoring electrodes.
Response. Pad and electrode design arc such that
isotropic mudcakes up to % in. have very little effect on
the measurements (see Chart Rxo-2 in Ref. 2). If the in-
vasion is shallow, the reading of the proximity device is
influenced by R,. The resistivity measured can be ex-
pressed as
RfL=GpiRxo+(l-Gpi)R,, . . .
Fig. 49.33 compares qualitatively the current-line
distributions of the MLL and the ML when the cor-
responding pad is applied against a permeable formation.
The greater the value of R,,lR,, the greater the tenden-
cy for the microlog 10 current to escape through the mud-
cake to reach the mud in the borehole. Consequently, for
high R,I R,, values, the readings of the ML respond
very little to variations of R,. On the contrary, all the
MLL I0 current flows into the formation and the MLL
reading will depend mostly on the value of R,,.
where G,i is the pseudogeometrical factor of the invaded
zone. The value of G,i as a function of invasion
diameter di is given in Fig. 49.34; this chart gives only
an approximate value of G,; , which in fact also depends
to some extent on the diameter of the borehole and on the
ratio R,, lR, .
If di is greater than 40 in., GpI is very close to 1 and,
accordingly, RPL will differ little from R,, . If di is less
than 40 in., R, is somewhere between R,, and R,,
usually much closer to R,, than to R, . RPL can be fairly
close to R, only if the invasion is very shallow (of course
when R,, and R, are nearly equal, the value of RPL will
depend very little on d,).
Response. Laboratory tests have shown that the virgin This is a pad-mounted SFL device. It embodies two
formation has practically no influence on the MLL distinct advantages over other microresistivity devices:
readings if the invasion depth is more than 3 or 4 in. The (1) it is compatible with other logging tools, specifically
influence of mudcake is negligible up to h mr = % in. but
the FDC and the simultaneous dual laterology (SDL),
increases rapidly with greater thicknesses. Chart Rxo-2 which eliminates the need for a separate logging run to
in Ref. 2 gives appropriate corrections; however, if mud- obtain R, information, and (2) it responds to shallow
cakes thicker than % in. arc anticipated, the PL is prc- R, zones in the presence of mudcake. The MSFL gives
ferred for R,, determination.
good R.w
resolution in thick-mudcake conditions, but
Mud Coke
does not require as great an invasion depth as does the
PL. This characteristic makes it useful in a wider range
of conditions than either the PL or the MLL. The effect
of mudcake on the MSFL data is shown in Ref. 2.
Principle. Spherical focusing is the shaping of the
equipotential surfaces produced by the resistivity device
to approximately spherical form. The focusing is ac-
complished by auxiliary electrodes, as in the MLL and
PL; but, instead of being forced into a narrow beam, the
measure current is merely prevented from following the
borehole mud or mudcake paths. A careful selection of
electrode spacings achieves an optimal compromise be-
tween too much and too little depth of investigation.
The ML permits a very accurate delineation of
permeable beds in all types of formations. It can also
provide satisfactory R,, and porosity determinations
under favorable conditions, which are (1) R,,I R,, <15,
(2) h,, c % in., and (3) depth of invasion greater than
about 4 in.
The focused microresistivity tools can provide good
R,, values under a much wider range of conditions. The
MLL is limited chiefly by mudcake thickness, but is well
adapted to salt-base muds. When h,, exceeds % in., the
PL or the MSFL log is preferable.
Uses and Interpretation of Well Logs
Bed Detection and Definition
Formations encountered in uncased boreholes may be
detected and their boundaries defined by a number of dif-
ferent logs. The SP and short-investigation curves are
most commonly used. For great detail, the microdevices
are superior-ML in fresh muds and MLL in very salty
muds. A substitute for the SP in salt muds is the gamma
ray log, which distinguishes shales from nonshale beds.
Also, the sonic and density logs could be used in all
types of formations, with any type of mud.
In holes filled with nonconductive muds, or in empty
holes, the induction, radiation, temperature, sonic (not
applicable in empty holes), and perhaps section-gauge
logs can provide useful information. Sometimes conven-
tional devices with wall scratchers, for contact with the
formation, may be run.
Porous and Permeable Beds-Sand Count. Porous
and permeable beds are of primary interest since they are
the potential oil and gas reservoirs. A sand-count deter-
mination of the total effective thickness of a permeable
section, excluding shale streaks and other impermeable
zones, can be derived from electrical logs.
Fresh Muds. The SP and microresistivity devices are
the principle curves for locating and defining permeable
beds. The SP has a good resolving power in formations
of low and moderate resistivity. In very resistive forma-
tions it can still detect shales and permeable beds, but it
cannot define their boundaries accurately. In most types
of formations the ML is best for establishing the detailed
location of the boundaries of permeable beds.
The microcaliper helps in the location of permeable
beds because it can usually detect the mudcakes. par-
ticularly if the adjacent formations are approximately at
bit size. Shales, on the other hand, generally tend to cave
-- R X,,= 0 I I,
Depth of Invasion, in
Fig. 49.34-Pseudogeometrical factors, MLL and PL.
or erode more than the permeable or hard impervious
beds. Very often, and mostly in sand-shale series, the
difference between the readings of a short-investigation
and a long-investigation device is a clear indication of
the presence of a permeable bed.
Salt Muds. The combination of very salty mud with
moderate-to-high formation resistivity adversely affects
the conventional (nonfocused) devices and the SP curve.
Also, mudcakes are thin; as a consequence, the
microcaliper (or also the section gauge) is not of great
help for the detection of permeable sections but detects
clearly the caved shales. In conjunction with the LL and
MLL the gamma ray log distinguishes between shales
and nonshales, and the neutron, sonic log, and density
log provide indications of porosity.
The process of correlating two (or more) logs in different
wells depends essentially on the similarity in shapes or
outlines of the curves (Fig. 49.35). In some regions such
correlations may be easily made for wells miles apart; in
other cases (where serious faulting, lenticular deposits,
or unconformities are present) it may be difficult to cor-
relate logs from wells only a few hundred feet apart. The
knowledge of formation dip from dipmeter determina-
tions assists in correlation.
Correlations are facilitated when the curves show
characteristic markers such as a well-known chalk or
shale section. Thin resistive beds such as lignites or
evaporites frequently furnish valuable correlation points.
For long-range correlation, the devices that investigate
large volumes of formation are best. However, when the
Fig. 49.35-Example of long-distance correlation.
pay zones are known in a field, they can be correlated in
great detail from well to well by comparing the ML or
MLL; thus, the geometry of the reservoir can be
Investigation of Porosity
Evaluation of formation porosity from electrical logs is
based on determination of the formation factor, FR,
which is related to porosity, 4, by Eq. 2. FR can be
found from the value of R,, as determined from the ML
or MLL with Eq. 6 rewritten as
FR=- Rx* (S,)2, . . . . . . (14)
where S,, = 1 -S,, .
To determine R, from microresistivity measurc-
merits, it is necessary that invasion be deep enough that
the measurements are not affected by the formation
resistivities beyond the flushed zone. An exceedingly
great mudcake thickness may also limit the accuracy of
the R, determination. For reliable porosity interpreta-
tion from ML data the porosity should be greater than
about 12 to 15%.
In an oil-bearing sand, the residual oil saturation
(ROS) must be surmised for use in Eq. 14. Taking a
value of 20% for the ROS would not usually entail too
great an error in most cases, at least in formations with
granular porosity and containing light oils. Greater
ROSs are frequent, however, chiefly in the cases of
highly viscous oil. Gas-bearing formations also seem to
display high residual gas saturations in the flushed zone
if the permeability is high. This is because of segregation
effects resulting from a combination of gravity and
capillary forces. Because of these residual saturations,
the ML may show water contacts.
To check the results of the microresistivity devices, or
to replace them in case they are not available, FR may be
determined as equal to RoI R,, R. being the true
resistivity of the formation at a level where it is 100%
water bearing. Extrapolation of a value of FR determined
at a water-bearing level to the oil-bearing section within
the same formation implies that the lithological
characteristics are approximately constant, which is not
always the case, however.
Still another approach often used in hard formations is
based on the value of Ri, the average resistivity of the in-
vaded zone estimated from the readings of the limestone
sonde or the short normal.25,26 This procedure has
proved valuable in many cases.
Other Tools for Obtaining Porosity. The sonic log, the
neutron log, and the density log are increasing in impor-
tance as auxiliaries to the electrical logs for improved
porosity determination.* In addition, with the present
sample taker, reliable determination of formation factor
and porosity are often possible from sidewall cores.
Investigation of Fluid Content
Quite often good qualitative judgments about a forma-
tion as a potential producing zone can be had by direct
visual inspection of the log. Basically such judgments
usually depend on (1) identification of a permeable for-
mation by means of the SP, mudcake indications on
microcaliper, positive separation on ML, indication of
invasion by separation between shallow- and deep-
investigation macmresistivity curves, etc., and (2) in-
dication by the deep-investigation resistivity devices that
R, in the permeable formation is appreciably larger than
the resistivity, Ro, that the formation would be expected
to have if water bearing.
When it can be assumed that there is no abrupt change
in the salinity of the formation waters, the radius of inva-
sion, etc., qualitative evaluations of saturation may be
made by comparing a porosity log with a resistivity log
over a large enough section of formations. Such a
method, in which the neutron log was the source of the
porosity data, was proposed several years ago.27 Also,
the possibility of accurate measurements of porosity with
the sonic and density logs has
mmpted similar
qualitative interpretation procedures.
When invasion is not too deep, so that the deep-
investigation resistivity devices read fairly close to R,,
the following procedures are applicable.
1. Sonic** transit time may be plotted vs. a deep IL
resistivity (in case of fresh muds) or vs. the LLD
resistivity (in case of salty muds). 29 By a proper choice
of scales, lines of constant resistivity index, I,, become
straight lines on the chart, as in Fig. 49.36. The line of
100% water saturation (ZR = 1) is constructed as the line
bounding the leftmost plotted points and passing through
the point corresponding to infinite resistivity and zero
sonic log porosity. Lines for other values of iR may then
be constructed. It is possible to distinguish between oil-
or gas-bearing and water-bearing zones by the relative
positions of the plotted points. For a zone to be oil- or
gas-productive the plotted point should fall appreciably
to the right of and below the line of 100% saturation.
This method is useful even if the formation-water
resistivity is not initially well known.
2. By use of porosity derived from the sonic or neutron
log and the resistivity from the deep-investigation log,
apparent values of the formation-water resistivity
The shon-interval sonic velocity log (Sony log) IS discussed I Chap. 50. The
neutron lag IS dlscusssd ln Chap. 51.
With some precautons. den%ty or newon data may be employed I place 01 90~
data in a similar crossplot technique
R, =R,I FR may be computed (using Eqs. 2 or 3 to ob-
tain FR). When tabulated in terms of depth, those ap-
parent values that stand definitely above the average
trend indicate the presence of hydrocarbon saturation.
This procedure is rapid and applicable in shaly sands. It
is also applicable when formation-water salinity varies
appreciably with depth, provided that the variation is
gradual and continuous.
3. The readings of a short-investigation resistivity log
should give indications of the variations in the formation
factor (and porosity). Accordingly, hydrocarbon satum-
tion can also be qualitatively determined by examining
the ratios of the readings of a short-investigation device
(such as a short normal, LLS, or PL) to those of the deep
induction log (IM) or LLD. The two sets of readings
may be used to locate points on a graphic plot, as
described previously, or to construct a continuous curve
of the ratio vs. depth. In either case the comparison is
facilitated if the logs have been recorded on a
logarithmic sensitivity scale.
The following procedure is helpful when, because of
deep invasion, the value of the true resistivity is in
doubt. The porosity from the sonic log is compared with
the porosity computed from a shallow-investigation
resistivity log (such as short normal, limestone sonde,
LLS or PL). The latter porosity is affected by any
residual oil saturation, whereas the sonic log porosity is
not. Zones where the two porosities differ should ac-
cordingly indicate potentially productive formations.
Also, in hard formations (where permeability is
generally low) a resistivity gradient (corresponding to a
saturation gradient) is observed on the logs between the
zone of lowest water saturation and the level of 100%
water saturation. The existence of this gradient is the
basis of a method for the delineation of intervals
saturated with hydrocarbons. 3o
Quantitative Interpretation*
Quantitative determinations of the hydrocarbgn satura-
tion from electrical logs are essentially determinations of
the water saturation, S,, in the uncontaminated zone.
Recalling Eq. 5
The evaluation of R, is a major step in the determina-
tion. Evaluation of FR is equivalent to an evaluation of
porosity, as already discussed. R. may be determined in
a 100% water-saturated section of the same sand or in a
lithologically similar sand.
Some interpretation procedures do not require explicit
evaluation of R, and FR . Along with a knowledge of the
SSP, or equivalently the ratio R&R,, they are based on
the ratios of readings taken with deep- and shallow-
investigation devices [e.g., R,,I R, method, induction-
electrical log (IEL) interpretation, Rocky Mountain
method]. A special approach is applied in case of shaly
Only ~omp~tatmns that can be made by hand wll be studled here Latest ~nterpreta-
ban techniques ate made by computers and are explalned under The Digital Age.
(TAKING YM*,R,X 21.000 FT/SEC)
F,(FOR rn: 2 1
Fig. 49.36--Qualitative method for dif-
ferentiating water-bearing
and oil- or gas-bearing
zones by plotting deep-
investigation resistivity
values against sonic-log
transit times. The ordinate
scale is proportional to
Determination of Rt. Conventional resistivity logs can
provide for the value of R, only for thick beds, since the
readings in general cannot be reasonably corrected for
the effect of adjacent formations. The long lateral,
because of its large radius of investigation, is practically
unaffected by the mud column and the invaded zone and
gives a good approximation of R, when the beds are
reasonably uniform over a thickness of at least 30 ft.
Such favorable conditions are seldom met in practice.
The long-normal reading usually requires a correction
for the borehole effect. This correction is sufficiently ac-
curate in fresh mud. The reading is close to R, when the
bed thickness is at least 10 to 15 ft and invasion is
shallow. The effect of deep invasion can be accounted
for, to some extent, by means of departure curves. 31,32
Correction for invasion requires the help of other devices
with shorter radii of investigation (ML and/or short nor-
which provide values of the invaded-zone
The advantage of the ILD and the LLD is that under
usual conditions of application their readings are practi-
cally unaffected by the mud column or by the adjacent
formations for bed thicknesses greater than about 5 to 6 ft.
The ILd, with the present technique, is appropriate for
logging in fresh mud. Its accuracy is excellent for forma-
tions reading up to 50 0-m and reasonable up to 200
n-m. Above that figure the accuracy is not so good. The
ILd gives R, when the invasion diameter does not exceed
25 to 40 in. and can be corrected for deep invasion. The
auxiliary readings are the ML and/or the short normal, or
better, the LL8 or SFL, particularly in case of con-
solidated formations.
The LL is the essential tool for the logging of hard for-
mations in salty mud. It gives R, for shallow invasion
(d; less than 15 to 25 in.). Correction for invasion is
made with the help of the MLL or MSFL. The most effi-
cient way to correct for invasion and thus obtain R, is the
use of a three-log combination: (1) DIL or dual induc-
tion/SFL in fresh muds and in soft and intermediate for-
mations and (2) dual laterolog/R, in hard formations or
salty muds. The so-called butterfly or tornado charts
in Ref. 2 provide a simple procedure to ob-
tain R,.
Maximum Producible Oil Index
If the porosity, 4, of a formation is uniform and in-
tergranular, the quantities S,4 and S,$ represent the
amounts of water per unit of bulk volume present in the
pores of the flushed and uncontaminated zones, respec-
tively. The difference Y=(S,, -S,)+ is the amount of
oil per bulk volume displaced by the mud filtrate. Y is
called maximum producible oil index because it ap-
proximates the maximum amount of oil per bulk volume
producible with water drive. Y is given approximately by
the equation 29
Y=@ lh -(-+) %. . . . . . . . .
Evaluation of Y by this equation does not require a direct
knowledge of porosity, 4, formation saturation, S,, or
flushed-zone saturation, S,, (or ROS).
Ratio Methods
It is impossible, in the space available, to present all log-
interpretation techniques or to present all charts and
tabulations made up thereof. For the practicing log inter-
preter most of the charts needed are included in Ref. 2.
The R,I R, Method. For clean formations, saturations
can be computed from the empirical relationship found
by combinations Eqs. 5 and 6,
sw=sxo(~.~) A . . . . . . . . . (16)
Any method which gives correct values for R,, , R,,
R mfi and R, may be used to determine S,, provided S,,
is known or can be reasonably surmised.
The chart of Fig. 49.37 provides a convenient way for
solving Eq. 16. The ratio R&R,* is entered in abscissas
(upper scale) and the ratio RIO/R, in ordinates. From the
point so plotted, an oblique line is extended up to its in-
tersection with the edge of the chart; from this intersec-
tion, a horizontal line is drawn which gives the values of
S, for different values of S,, .
The value of S, found directly by interpolation be-
tween the two nearby oblique lines is based on the
assumption that S,, , on the average, is related to 5,
through the empirical equation S,, = a,. Within the
limits where R ,+ = R,, (i.e., R, between 0.08 and 0.3
O-m at 75F), the Essp can be entered, instead of
R&R,, using the lower grid.
Microlog Shaly-Sand Method. The amplitudes of both
the SP deflection and the resistivities are reduced by the
presence of interstitial shale in a formation. In some ex-
Fig. 49.37-Interpretation chart for RJR, and shaly-sand
methods. (Working lines illustrate example of
Fig. 49.38).
treme cases shaly oil sands are hardly differentiated from
the adjacent shales on the logs. Furthermore, the ML
does not show much separation between the curves. In-
terpretations in shaly sands, therefore, are generally
made under comparatively unfavorable conditions. The
practical method of interpretation described hereafter is
quite approximate and is applicable only when the
amount of shaliness is not too great.*
The practical method of interpretation in shaly forma-
tions33 is based on the observations made on field logs
that in 100% water-saturated shaly sands the PSP
(pseudostatic SP) is given by
Epsp=-Kc log?, _.
For an interlaminated sand-shale formation containing
oil and/or gas, the PSP can be expressed by the equation
E PSP=-Kr. logf-201srK,. lo+, .
, 1,
where asp is the SP reduction factor and is defined as
Epp/E~p. For clean sands, asp = 1 and Eq. 18 reduces
to Eq. 17 as a special case.
According to Eq. 18 the determination of S,,, in a shaly
sand requires the knowledge of R,, , S,, , R,, E,, and
ESSP. The Essp can be determined from clean sands
reasonably close to the shaly formations considered. The
Epsp is given by the deflection of the SP curve at the
level of the bed, after correction for bed thickness, if
necessary. The value of S,,, =l -S,,, again must be
The solution of the equation for S,,. is given by Fig.
49.37, which applies to both clean and shaly formations.
For a shaly sand, the PSP and R,,/R, are first entered in
abscissa and ordinate, respectively. This gives us the ap-
Sane other shaly-sand references are Refs. 8 and 9.
__- .___ L __._.._
Fig. 49.38-Shaly-sand example: induction-electrical survey and microlog-microcaliper
survey in well on Louisiana Gulf Coast. I?, = 1.45, R,, = 1.2, and R,, = 1.3 at
BHT (14OOF). d= 83/i in.
parent saturation. To find the true saturation, a line is
drawn through the origin (circled), and the point
representing the apparent saturation; and this line is ex-
tended until it intercepts the Essp or the corresponding
value of R,fI R,. The saturation at this point is treated in
the same manner as in the R,,I R, method above.
If the Essp is not known and R,%, is unobtainable, it is
not possible to determine the water saturation, S,,..
Nevertheless, plotting the ratio R,,I R, vs. PSP on Fig.
49.37 will show whether or not the sand falls on the
100% water-saturation line. If it falls some distance
below this line, there is a chance for production.
Note that in sand-shale laminations the value of S,,, is
the water saturation of the sand itself. In sands contain-
ing disseminated shale, it is the water bound by the
quartz grains and does not include the water held by the
Example Problem 1. Fig. 49.38 shows the IEL and
the ML over a portion of a well in Jefferson Davis
Parish, LA. The shaly sand from 8,046 to 8,054 ft will
be interpreted. The mud resistivity at formation
temperature of 140F is 1.45 Q-m, R,, = 1.3 Q-m, and
R,,,,= 1.2 9-m at 140F. Bit size is 8% in.
From the IEL, the Epsp is -55 mV. The Essp (sand
at 7,830 to 7,850 ft) is - 130 mV. Thus cusp is 0.42. The
short-normal resistivity R 16 is 2.2 n-m, and the
induction-log resistivity R,L is 1.9 B-m. Invasion is
known to be quite shallow, as further substantiated by
the readings of the microlog. Therefore R,L may be
taken as equal to R,.
As shown by the positive separation on the ML, three
distinct porous intervals, A, B, and C, appear between
8,046 and 8,054 ft. The values of R,, and h,, for each
interval are found by use of Fig. 49.31. Thus, for Inter-
val A R,, is 4.0, for Interval B it is 5.6, and for Interval
C it is 6.8. The average R,, is 5.5, and for all three inter-
vals the mudcake thickness is % in. Note that the
microcaliper indicates a hole diameter of 7 3/4 in., exactly
1 in. smaller than bit size, corresponding to a %-in.
mudcake, thus verifying the h,, values from Fig. 49.3 1.
I I , I
IO 2 3 4 5 6jBYlO
7 _
6 BO NOT USE WHEN -- . ~~~
R,,/R,<~ FOR d, = 3d,
\ -
0 7 OHM
Fig. 49.39-40-in. induction log, l&in. normal interpretation chart. (Working lines il-
lustrate example of Fig. 49.40.)
All necessary data are now at hand:
= 5.5 D-m
R,(=R,L) = 1.9 n-m
-55 mV
- 130 mV
asp = 0.42
normal reading must not be perturbed by the presence of
resistive streaks.
K, = (for the SP) at 140F is close to 79
R,r,I R, = 2.9, and
R.,,,lR,,,f = 4.6
The chart incorporates approximate compensation for
the presence of an annulus, assuming equal viscosities
for the oil and water and an S,, in the flushed zone of 15
to 30%. To use the chart the value of the ratio R~~IRIL
(resistivity fmm 16-in. normal divided by the IL
resistivity), corrected, if necessary, for effect of
borehole and adjacent formations, is plotted either vs.
the SSP (lower grid) entered opposite the appropriate for-
mation temperature or vs. R,fIR, (upper scale).
Fig. 49.37 is entered with R,,I R, =2.9 and PSP= -55 Points falling within the shaded area correspond to
at K, =79, locating a point on the chart. Then a line is water-bearing sands. Points falling below the shaded
drawn through this point from the origin and extended area correspond to oil saturation. Approximate saturation
until it intersects the vertical line erected at Essp =- 130 scales are provided for dj values of 3dh and 5dh. The
mV for K, =79. This intersection gives a value of S, dashed lines represent lines of equal saturation on the
equal to 43 %, assuming that S,, is 15%, which is chart. The critical saturation corresponds to S,U =60% in
reasonable. The sand was perforated from 8,046 to soft formations and 50% in consolidated sandstones.
8,050 ft and made 75 BOPD and 280,000 cu ft/D gas These may not be the correct critical saturations for
through a 7/64-in. choke. many limestones.
IEL Interpretation Method. A chart for practical inter-
pretation of fluid saturation from the readings of the IEL
combination is shown in Fig. 49.39, Ref. 2. The condi-
tions limiting valid use of the chart are: (1) the invasion
diameter di must be between 2dh and 1Odh (dh is the
hole diameter), (2) R,, must be greater than R,, which is
generally fulfilled if R,,fI R, is greater than three to live,
as for fresh muds and saline formation waters, and (3)
the beds must be fairly homogeneous-i.e., the short-
For shaly formations a construction like that used in
the shaly-sand method can be used in Fig. 49.39.
Although not rigorously correct, this procedure should
give acceptable results if the shaliness is not too great.
Since it is not always known beforehand whether the
interpretation chart (Fig. 49.39) is within its limits of ap-
plicability, it is useful to employ the value of porosity or
formation factor, when known from independent
sources, to check the results by means of Eq. 5. Details
of this porosity balance check are given in Ref. 34.
Fig. 49.40-Induction-electrical log and microlog-microcaliper log on a well in Young Coun-
ty, TX. I?,,, = 0.95 and R ,,,, = 0.61 at BHT (117OF). d= 7% in.
Example Problem 2. Fig. 49.40 shows part of the
induction-electrical log and ML on a well in Young
County, TX. The sand at the bottom of the figure is divid-
ed into two parts, A and B, by a thin, hard streak, as
shown on the ML. A and B will be interpreted separate-
ly. Necessary well data are R, =0.95 and R,,lf=0.61, at
BHT of 117F; hole diameter, d,, , is 7% in.
Interval A. From the induction-electrical log,
RIL=17.5 R-m, R16=32.5 R-m, SP=-95 mV, and
BHT is 117F. Applying Fig. 49.39, the ratio
R,6I R1~=32.5/17.5=1.85. The SP is -95 mV, so
the point is fixed on the chart, giving an average S, of
about 30 % .
Interval B. Here the RI L=26.0 R-m and R16=37.0
n-m. R161RI L=37.0/26.0=1.42. The SP is -90 mV at
117F. This fixes a point on Fig. 49.39, which gives a
value of S, of about 25 % . Drillstem tests on the two in-
tervals showed that the sand contained gas and distillate,
with a higher flowing pressure over Interval B.
Rocky Mountain Method. The Rocky Mountain
method* was developed for the interpretation of con-
ventional electrical logs to yield values of S, and 4 in
invaded, clean, hard formations when they are suffi-
ciently thick and homogeneous. It is necessary that the
invasion be such that the average resistivities recorded
by the short normal and the lateral (when corrected for
borehole effect) approximate the average values of Rj
ami R,,
respectively. For this to be true, the short-
normal readings should be at least 10 times the mud
resistivity. Only average values, over thick intervals, can
be obtained with this method.
F, . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. R,i is the resistivity of the water found in the invad-
ed zone. It is usually made up of filtrate and interstitial
(connate) water mixed in such proportion that
1 fw
(1 -fw)
-, . . . . . . . . . . . . . . . . . .
Rwi Rw Rnzf
wheref, is the fraction of interstitial (connate) water in
the total mixture (usually 5 to 10%).
2. Si is the water saturation in the invaded zone. It has
been found by experience that Si * =S,; so
Ri=FR,ISw. Inasmuch as R,=FR~I S,*, we obtain
Since R,I Rwi =f,+(l -f,)R,/R,,,f, it follows that
R,iI R, is a function of the SP value. The upper part of
Fig. 49.41 gives a graphical solution of the equation for
S,. The SP is entered in ordinate and the ratio RiIR, on
oblique lines. The intersection gives the abscissa S,.
The lower part of the chart is used to obtain the porosi-
ty by using the water saturation, S,, just found, with the
Fig. 49.41-Interpretation chart for
Rocky Mountain method.
value of R,I R,. The intersection falls on or between
oblique lines that are graduated into porosity values, ac-
cording to the Humble formula, Eq. 2. This method
should not be used in salt muds.
Electromagnetic Propagation Tool
Principle. The EPT (electromagnetic propagation
too13) measures the travel time and attenuation rate of
an electromagnetic wave through the formation near the
borehole. In addition, a caliper and a ML can also be
recorded. The tool can be combined with the gamma ray,
neutron, or density instruments.
The propagation time of water differs sharply from
those of gas, oil, or matrix rock and is, moreover, little
affected by the salinity of the water. This tool permits the
evaluation of water saturation that is relatively indepen-
dent of water resistivity (salinity) and, in fact, is most ac-
curate in the fresher waters.
Dielectric permittivity is one of the main factors deter-
mining electromagnetic propagation in a material.
Dielectric permittivity of any medium is proportional to
the electric dipole moment per unit volume. The electric
Relative Dielectric
Gas or air 1 .o
Oil 2.2
Water 56-60
Quartz 4.7
Limestone 7.5
Dolomite 6.9
Anhydrite 6.5
Propagation Time,
dipole moment is made up of one or more effects: elec-
tronic, ionic, interfacial, and dipolar. Since each of these
dominates over a certain range of the electromagnetic
spectrum, they can be separated experimentally. *
The electronic contribution results from the displace-
ment of electron clouds, and is the only one that operates
at optical frequencies. The ionic and interfacial contribu-
tions come from displacement and movement of ions,
hence arc confined to low frequencies. The dipolar con-
tribution is from permanent electric dipoles, which orient
themselves in the direction of an applied electric field.
With the exception of water, there are very few materials
abundantly found in nature that have permanent electric
dipoles. A borehole dielectric measurement in the
109-Hz frequency region, where the dipole polarization
of water dominates, should lead to a measurement of
water content that is independent of salinity.
Table 49.3 gives laboratory-measured values of propa-
gation time and dielectric permittivity (relative to air) of
typical reservoir materials.
Tool. The tools now in the field carry two transmitters
and two receivers on a wall-contact pad, configured as
shown in Figs. 49.42 and 49.43. These transmitters and
receivers must be antennas to operate as they do in the
microwave frequency range. The tool uses a differential
measurement based on the signals detected by near and
far receivers, similar in principle to the widely used
method of measuring At with a two-receiver sonic tool.
In a similar manner, the two receivers produce cancella-
tion of any effects caused by mudcake or variations in
signal coupling (so long as both receivers are affected
To reduce any error caused by sonde tilt, the EPT uses
an antenna configuration similar to the transducer array
used in a borehole-compensated sonic tool. Transmitting
antennas are placed above and below the receiver pair
and are pulsed alternately. Simple geometric considera-
tions show that if these two transmission modes arc
averaged, the first-order effects of pad tilt will be
The basic principle of the tool involves a surface or
lateral electromagnetic wave launched along the surface
of a conducting pad. In the absence of mudcake, the
electromagnetic wave would move along the pad face
past two receiving antennas, but in the normal borehole
case with mudcake present, propagation takes place on
the surface between mudcake and formation. The phase
shift and attenuation per unit distance along the surface
of the pad are proportional to E and C (as shown in
theory) for a plane wave.
It has been demonstrated both theoretically and ex-
perimentally that for mud cakes up to % in. the travel
time measured by the EPT is essentially the same as the
travel time in the invaded zone without any mud cake.
Above such thickness the measurement deteriorates
rapidly until the tool responds only to the mud. Limited
experience with air- and oil-based mud-filled tools in-
dicates that even very thin layers of these fluids between
the pad and the formation cause the tool to respond only
to the fluid and not the formation. This is because of the
short travel time of these fluids. The tool contains a
1 .l-GHz microwave transceiver. The transmitter is
capable of generating more than 2 W of output power
while the receiver can process a 0.3 pico watt (pW)
signal. This allows accurate measurements in formations
when R, approaches 0.3 a-m.
Theory. Assuming a plane wave varying sinusoidally in
time, the electric field, E, at the second receiver is given
E=E,,e - YL+bJt
. . . . . . . . . . . . . . . . . . . . . . . . . .
where E, is the electric field at the first receiver; L is the
distance between the two receivers; j is the vectorial
operator fi; o is the angular frequency; t is the time
of travel of the waves over a distance L in the formation;
and y is the complex propagation factor given by
y=cY+jp, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..(22)
where CY is the attenuation factor (coefficient) in
neper/m, and /3 is the phase factor in rad/m.
For a lossless formation,* or=O. From Eq. 21, the
phase velocity vPO is given by
L Cd 1
-=- . . . . . . . . . . . . . . . . . . . . .
where the subscript o indicates loss-free conditions and
tpo is the loss-free propagation time of a given medium
in ns/m.
From Maxwell s equations, it can be shown that
y. =jw& =jwtpo, . . . . . . . . . . . . . . . . . (24)
where is the magnetic permeability (H/m). Since most
formations of interest are nonmagnetic, of the forma-
tion is the same as that of free space (
=47rX 10
H/m), and E is the dielectric permittivity (F/m). When
the formation is lossy, y and E are complex.
Fig. 49.42Schematic of EPT antenna
pad, showing principle of
2-receiver measurement of
*Formation with no electromagnetic energy losses
transit time.
Fig. 49.43The EPT antenna pad.
Squaring Eq. 22 and Eq. 24 and equating the real and
imaginary terms, we have
w2/.Loc=p2 -a?, . . . . . . . . . . . . (25)
wp,C=2aUp, . . . . . . . (26)
where C is the equivalent conductivity (a/m) of the
losses in the formation. Dividing Eq. 25 by w2,
p2 a2
w2 ,2 . . . . . . . I.. . . . . . . .
From Eq. 24, 40~=tpo2. Since filw=t,l is the travel
time in the lossy medium,
2- I?--*
- tp1
w2 . . . . . . . . . . . . . . . (28)
Remembering that 01 is the attenuation factor, Eq. 28 im-
plies that the actual propagation time in a conductive for-
mation is longer than that of a corresponding loss-free
formation. If the propagation wave is not a plane wave,
suitable spreading-loss corrections to the measured at-
tenuation (A,,,) are made before Eq. 28 is applied. Thus
the corrected attenuation A,. =Aiog -G,Y (dB/m) where
G,, is the geometrical spreading loss and AI,,~the re-
corded attenuation in dB/m. G,? is about 50 dB/m in air
or G,f =45.0+ 1.3r,/ +O. 18r,/. Here tp/ is the recorded
travel time in ns/m.
Interpretation. The range of travel time encountered in
the borehole in common reservoir rock varies from 6.3
ns/m for a 40-pu sandstone filled with hydrocarbons to
17.2 ns/m for a 40-pu water-filled limestone. In terms of
phase shift, this corresponds to angles between 100 and
270 when computed over a 4-cm receiver spacing.
The EPT log responds primarily to the bulk volume of
water in the formation. Since the tool has a relatively
shallow depth of investigation-about 1 to 6 in. depend-
ing on the conductivity-it normally responds to the
flushed zone of an invaded section.
Eq. 28 can be transformed into
tpo= (tJ-&) lh . . . . . . . . . . . . .
The apparent water-filled porosity (dEPT) can be derived
in a way similar to the derivation of the porosity from
sonic At. Thus
tpo - tpm
. . . . . . . . . . . . . . . . . . . . . . .
tpwo - tpm
tpwo, the loss-free propagation time of water, varies with
the temperature and slightly with pressure and can be ob-
tained as
where T is the temperature. F. Knowledge of water
salinity is not required to obtain r,,,,.,, .
*a I neperlm and A ,~~ IS dWm Smce 3 neper = 6.666 dB, A ,og =6.66&t
The loss-free propagation time of matrix, tpm, is in-
dicated on Table 49.2. The nature of the matrix can be
determined by the knowledge of the apparent matrix den-
sity and interpolating between lithology density values.
The water saturation S,, is given by
tpo = Sxo4tpwo
+(l -Sxo)4tph f(1 -+p,, . . (31)
s.w =
tpo -tpm +d(tpm -tph)
, . . . . . . .
4(tp,, _ tph)
where tph is the propagation time for hydrocarbon and 4
is the porosity of the formation. Since tpm and tph are
fairly close, we can estimate S,, roughly as
S,= !-. . . . . . . . . . . . . . . . (33)
Example Problem 3. Fig. 49.44 is an example of the
log presentation currently in use. Track 1 contains a con-
ventional caliper curve, taken from the motion of the
backup pad, and the attenuation curve, scaled in dB/m.
Tracks 2 and 3 are given to the principal measurement,
travel time (tp/), in ns/m. Track 3 also presents the signal
levels from the two receivers. The chief use of these
curves is to monitor the primary signal detection at the
receivers, which provides an indication of the relative
reliability of the log parameters at any level.
A self-evident and very real advantage may be inferred
from the 4-cm spacing between receivers; the tool has
excellent vertical resolution. The log of Fig. 49.44 ac-
tually looks overactive in spots, but its repeatability
testifies that the recording is valid. In fact, the data
recorded by the tool are too detailed for direct merging
with other logs by means of computer. Averaging
(smoothing) subroutines are thus required preliminaries
for programs using EPT data.
Example Problem 4. Fig. 49.45 shows an ISF log and
the porosity computed from density. neutron, and EPT
logs. Zone A is obviously gas bearing, as evidenced by
the neutron porosity reading much less than the density.
The EPT porosity is a little higher than the neutron
porosity and much less than the density porosity, confir-
ming the presence of hydrocarbons.
Zone B exhibits a different porosity profile. Once
again the neutron porosity is less than the density porosi-
ty, indicating the presence of some light hydrocarbons,
but now the EPT porosity is less than both neutron and
density pomsities. Since the total porosity from neutron
and density logs is roughly
and the hydrocarbon volume is
the volume of hydrocarbons affecting the three porosity
tools is about the same for Zones A and B as determined
from the EPT data. However, there is a much stronger
light hydrocarbon effect on the neutron and density logs
in Zone A. Thus, it would be expected that Zone B con-
tains more condensate or oil than Zone A.
3 5oc
5n Caliper l5l
Travel Time,Jt,l 1, nsec/m
Fig. 49.44-An unaveraged EPT log shows fine detail. Repeat sections (faint curves) of the attenuation
and t,, curves show excellent repeatability.
60 +40
________ !F:
I - - -
~1~0 6--w-----------
Fig. 49.45-ISF and EPTICNLIFDC logs.
Zone C appears to contain some residual hydrocar-
bons, since the EPT porosity is often slightly less than
the neutron and density porosities. There is a little more
hydrocarbon in the shalier bottom part of Zone C than in
the cleaner top portion.
The very top of Zone D contains oil since the EPT
measurement is much lower than the neutron and density
porosities, which read about the same. Water-bearing
zones are identified when the EPT porosity is about the
same or higher than the neutron and density porosity.
Thus, Zone E is clearly water bearing, as is the bottom
of Zones B and D.
Machine programs are available to give complete
quantitative interpretation of all these logs. This is
especially important in the study of tar sands or crude
oils where the hydrocarbon is not flushed by the mud
filtrate and where S,, is very near S, . These studies arc
also of great interest to provide values of residual oil
The Digital Age
Before 1960, all logs were recorded in analog form on
film or paper. Magnetic tape recording was introduced in
1960 to record dipmeter. Shortly thereafter, various
other logs were also recorded on tape, thus permitting
the use of computers for various purposes. Before long,
computers were made an intrinsic part of the recording
systems on the logging trucks. This has revolutionized
the capacity for data acquisition at the wellsite. At the
same time, many computer-processed products have
become available in real time or only a short time after
logging is completed at the wellsite.
Generic Name Derivation Log Input Required Presentation Schlumberger Gearhart Dresser Welex
R... Assumes all formations contain 100% water. Simultaneous Sinale curve in R,,, R
Computes apparent R w, R wa = R ,/F
Uses deep resistivity and shallow focused
resistivity to estimate RJR, ratio.
.._ resistivity and Track 1 on
porosity. Usually logarithmic
sonic and scale.
Simultaneous deep Single curve in R,IRt
resistivity and Track 1
shallow focused compatibly
F overlay or Derives F from a porosity curve that IS played Deep resistivity
onto logarithmic resistivity as R,. and porostty
Any combination of porosity logs with same Simultaneous
lithology assumptions to compute porosity. porosity logs
Hole volume Uses caliper logs to compute hole volume for Caliper curve-
cement calculation. preferably 2
curves at 90
such as 4.arm
Fracture Uses differences in adjacent pad readings &arm dipmeter
locating log from 4-arm dipmeter to infer fractures.
scaled as a
pseudo SP
Single dashed R,
curve in
Tracks 2 and
3 on
Coded curves in Compatible
Tracks 2 and porosity
3 with gamma scales
ray and
caliper in
Track 1.
Pips or tic Borehole
marks in volume
depth column
at every 10 cu
ftand 1OOcu
Adjacent pad Fracture
F curve F Curve
Compatible Compatible Compatible
porosity porosity porosity
scales scales scales
Borehole Borehole Borehole
volume volume volume
Fracture Fracture Dipmeter
readings are
and any
separation is
Generic Name Derivation
Merged and Replays all logs, shifts depths and makes
depth shifted sample calculattans such as R,,, R xo,
data R,/Rr, compatible porosity scales or
cross plot porosities.
Wellsite Uses all logs to provide a first order
evaluation log computer analysis.
Formation dip Computes formatron dtp from 4-arm
computations dipmeter
True vertical Computes lVD of any point from dipmeter
depth log orientation data
Log Input Requrred Presentation Schlumberger
Any logs run on Usually 3 to 5 Cyberlook
tape. tracks. Varies Pass 1
by service
diplays only
log data.
Resistivity and Usually 3 or 4 Cyberlook
porosity tracks. Has
reservoir data
derived from
log data.
4-arm dipmeter Formation dip, Cyberdip
hole deviation,
Contmuous Replay of any TVD
dipmeter plus log on TVD
any log to be scale
Gearhart Dresser Welex
cross-plot Prolog Computer
(x-plot) Van
Well Prolog CAL
converted lo
All logs available in real time are also avarIable in replay time If recorded on magnetic tape
Without any doubt, the digital age is responsible for
the creation of new equipment deemed impossible
before. Many interpretation techniques and studies today
could not be made without the use of computers. Finally,
electronic transmission of log data is a present reality,
facilitating exchange between wells and offices, towns
and continents. An overview of this vast field is
Magnetic Tapes
API-recommended standard format permits logging ser-
vice company tapes to be read by most computers. More
exhaustive treatment of the subject is available directly
from the service companies. Quality control of the
magnetic tape is ensured in real time in integrated log-
ging systems having on-board computers.
Computed Log Products
Log analysis performed by a computer is available to the
log user at three different levels.
1. Real-time quick-look products, summarized in
Table 49.4, run at the same time the log is being run.
Many of these curves, such as R, and F curves, are
recorded on the standard logs and are often placed in the
SP or resistivity tracks.
2. Wellsite log analysis products, summarized in
Table 49.5, are generally available at the wellsite in
replay time. Wellsite analysis is made after the logging is
completed. The process involves playing back taped logs
and using an appropriate log analysis program, such as a
shaly-sand analysis or a dipmeter computation.
3. Computing center products am provided well after
the logging is finished (days or weeks later) and are
generally more comprehensive than either of the wellsite
products. In general, these computations fall into three
categories: shaly-sand analysis, complex lithology
study, and dipmeter processing. The most-used products
are summarized in Table 49.6. Other, less frequently
used products such as tar sand analysis or mechanical
properties are not included; details on these may be ob-
tained directly from the service company.
Generic Name Derivation Log Input Required Presentation Schlumberger Gearhart
Advanced Uses most sophisticated analytical and Rssistivhy, density, Usually 4 tracks SARABAND Comsand
sandstone statistical methods to correct and compute neutron, gamma presentation
analysis logs in sandstones and shaly sandstones. ray with sonic of lithology, YOUN F Pairs
desirable saturation,
porosity and
bulk volume.
Advanced Uses most sophisticated analytical and Resistivity, density, Usually 4 tracks COAIBAND Comlith
carbonate statistical methods to correct and compute neut r on, gamma presentation
analysis logs in carbonate and lithologically complex ray with sonic of lithology, GLOBAl . Frax
reservoirs and saturation,
microresistivity porosity, and
desirable bulk volume.
Advance Uses most advanced correlation logic to *-arm dipmeter Arrow plot with CLUSTER NEXUS
dipmeter compute dips followed by a statistical caliper, (structural) and
computa- sorting to retain the most reliable data. correlation GEODl P
tions curve and (stratigraphic)
hole deviation.
Also available:
Schmidt plots,
and listings.
Dresser Welex
Dresser Diplog
computed analysis
DENTaT mrom
NElJmoN roIosm
Fig. 49.46-CYBERLOOK Pass One log.
An example of wellsite log analysis is the
CvBERLOOK"pmgram, 36 which requires as a minimum
suite of logs a deep investigation resistivity,
CNL-FDC (compensated neutron/density logs), a
gamma ray, or SP curve. The CYBERLOOK computation
is based on the dual-water model and is normally made
in two passes.
On the Pass One log (Fig. 49.46), the SP curve is on
Track 1 with the gamma ray. On Track 2, in four cycles,
are found the R, curve, the R, (computed fmm R,),
and the porosity given by the CNL. Track 3 shows the
porosity given by the density, 40, the porosity given by
the neutron, fpN, and a cmssplot porosity computed from
#D ami #N.
On Pass Two log (Fig. 49.47), the Track 1 gives the
shale index, which is the minimum shale index of several
shale indicators obtained from the SP curve, the gamma
ray, and the maximum and minimum neutron readings.
Track 2 shows R, as a dashed curve and Ro as a solid
curve. The left half of Track 3 has the water saturation
and the right half has the porosity and bulk volume free
water. A differential caliper is presented as a dotted
curve with bit size in the middle of Track 3. A gas flag
appears in the depth track when a large hydrocarbon cor-
rection was necessary to obtain the porosity from
neutron/density logs.
VOLAP is an example of a complex analysis program
(Fig. 49.48). It is based on the dual-water model, as is
the CYBERLOOK program mentioned previously, but the
computations are far more refined and the results more
accurate. For a detailed study of the dual-water model,
see Refs. 37 and 38. The dual water model simply says
that in a shaly sand, its equivalent formation water con-
ductivity is dependent on the relative amount of
bound water and free water.
The conductivity of the bound water is found by the
use of the nearby shale resistivity and the total porosity
given by the average of CNL. In a like manner, the free
v (*
-------- ----
owIuY* Y
Fig. 4%47-CYBERLOOK Pass Two log.
water conductivity is found by use of the resistivity of
the clean water sand and its total porosity. In a shaly
water sand, the equivalent water conductivity is found in
the same way by using the resistivity of the shaly water
sand and its total porosity. Knowing the bound and free
water conductivities, it is easy to compute their fractions
of the total porosity that are necessary to obtain the same
equivalent water conductivity of the shaly water sand.
The fractions of bound and free water can be related to
the relative deflections of the gamma ray or SP curve,
etc., thus permitting the use of such calibrations when
analyzing hydrocarbon saturated zones. The analysis is
done by using a dispersed-clay-type equation.
a,,f = chemical activity of mud filtrate
a,. = chemical activity of formation water
A,. = corrected attenuation of a formation
A log
= recorded attenuation of a formation
C = equivalent conductivity of losses in the
C,t = conductivity as given by induction log
CT = total conductivity signal
Ci = conductivity of invaded zone
C, = conductivity of mud
C, = true conductivity of formation
= conductivity of flushed zone
E = electric field
E,. = total electrochemical EMF
EJ = liquid-junction EMF
EM = shale-membrane EMF
Epsp = pseudostatic SP
Esp = static SP
i l
Fig. 49.48-High-porosity presentation.
En = electric field at the first receiver
FR = formation resistivity factor
= resistivity factor of formation water
= resistivity factor of water in invaded zone
f,,. = fraction of interstitial (connate) water in the
total mixture
Gi = geometrical factor of invaded zone
G,, = geometrical factor of mud
G,j = pseudogeometrical factor of the invaded
G,! = geometrical spreading loss
G, = geometrical factor, true formation
G,,, = geometrical factor of flushed zone
iR = resistivity index
j = vectorial operator J-l
K,. = electrochemical coefficient
m= cementation exponent or factor
n = saturation exponent
R, = apparent resistivity
R = annulus resistivity
i: = corrected resistivity
RIL = induction-log resistivity
RpL = proximity-log resistivity
R, = resistivity of invaded zone
R,,. = resistivity of mudcake
= resistivity of mud filtrate
R, = true formation resistivity
R, = formation-water resistivity
R = equivalent formation-water resistivity
RI = resistivity of the water found in the
invaded zone
R = resistivity of flushed zone
i: = resistivity of a clean (nonshaly) formation
saturated with 100% water
= resistivity of 1- X 1 -in. microinverse
R2 = resistivity of 2-in. micronormal
16 =
short-normal resistivity
Si = water saturation in the invaded zone
S,, = residual oil saturation
S, = formation water saturation
= water saturation in the flushed zone
= propagation time for hydrocarbon
= travel time in the lossy medium
= loss-free propagation time of matrix
= loss-free propagation time
= loss-free propagation time of water
Y = maximum producible oil index
= SP reduction factor
y = complex propagation factor
t = relative dielectric permittivity
p = magnetic permeability
4~ = density porosity
WEPT = electromagnetic propagation tool porosity
$HC = hydrocarbon porosity
4~ = neutron porosity
w = angular frequency
CNLTM = compensated neutron log
DIL = dual induction-laterolog 8
EPT = electromagnetic propagation tool
ES = electrical survey
FDC rM = compensated density log
ID = deep-reading induction device
IEL = induction-electrical log
IES = induction-electrical survey
IL = induction log
ILd = deep induction log
IM = medium-reading induction device
ISF = induction spherically-focused log
LL = laterolog
LLD = deep laterolog
LLS = shallow laterolog
ML = microlog
MLL = microlaterolog
PL = proximity log
SDL = simultaneous dual laterolog
SFL = spherically focused log
SSP = static SP
Dunlap, H.F. and Hawthorne. H.R.: Calculation of Water
Resistivities from Chemical Analysis, J. Per. Tech. (July 1957)
202-17; Trans., AIME, 192.
a. Log Interpretation Charts, Schlumberger Well Services
b. Log Interpretation Charts, Dresser-Atlas (1981).
c. Charts for the Interpretation of Well Logs, Welex (1979)
d. Chart Book. Gearhart (1982).
Lamont. N.: Relationships Between the Mud Resistivity, Mud
Filtrate Resistivity, and the Mud Cake Resistivity of Oil Emulsion
Mud Systems, J. Pet. Tech. (Aug. 1957) 51-52; Trans., AIME.
Mounce, W.D. and Rust, W.M. Jr.: Natural Potentials in Well
Logging, Per. Tech. (Sept. 1943); Trans., AIME. 6.
Winsauer, W.O., er al.: Resistivity of Brine-satured Sands in
Relation to Pore Geometry, Bull., AAPG (Feb. 1952) 253-77.
patnode, H.W. and Wyllie, M.R.J.: Presence of Conductive
Solids in Reservoir Rocks as a Factor in Electric Log Interpreta-
tion, J. Pet. Tech. (Feb. 1950) 47-52; Trans., AIME, 189.
Archie. G.E.: Classification of Carbonate Resetvotr Rocks and
Petrophysical Considerations, Bull., AAPG (Feb. 1952) 36,
Waxman, M.H. and Thomas, E.C.: Electrical Conductivittes m
Shaly Sands-I. The Relation Between Hydrocarbon Saturation
and Resistivity Index; II. The Temperature Coefficient of Elec-
trical Conductivity, J. Pet. Tech. (Feb. 1974) 213-23; Trans.,
AIME, 257.
Waxman, M.H. and Smits, L.J.M.: Electrical Conductivities in
Otl-Bearing Shaly Sands, Sue. Pet. Eng. J. (June 1968) 107-22;
Trans., AIME, 243,
Kunz, K. and Moran, J.: Some Effects of Anisotropy on
Resistivity Measurements in Boreholes, Geophpics (Oct. 1958)
23, 770-94.
Doll, H.G.: Filtrate Invasion in Highly Permeable Sands, Pet.
Engr. (Jan. 1955) 27, BJ3-66.
Gondouin, M. and Scala, C.: Streaming Potential and the SP
Log. J. Pet. Tech. (Aug. 1958) 170-79; Trans., AIME, 213.
Hill, H.J. and Anderson, A.E.: Streaming Potential Phenomena
in SP Log Interpretation, J. Pet. Tech. (Aug. 1959) 203-08;
Truns., AIME, 216.
Wyllie, M.R.J.: Investigatron of Electrokmetic Component of
the Self-Potential Curve,J. Pet. Tech. (Jan. 1951) l-18; Truns.,
AIME, 192.
Wyllie, M.R.J., de Witte, A.J.. and Warren. J.E.: On the
Streaming Potential Problem in Well Logging, Trans., AIME
(1958) 213, 409-17.
Wyllie, M.R.J.: Quantitative Analysis of the Electrochemical
Component of the SP Curve. J. Per. Tech. (Jan. 1949) 17-26:
Trans., AIME, 186.
Segesman, F. and Tixier, M.P.: Some Effects of Invasion on the
SP Curve, /. Pet. Tech. (June 1959) 138-46; Trans., AIME.
Doll, H.G.: SP Log: Theoretical Analysis and Principles of ln-
tetpretation, J. Pet. Tech. (Sept. 1948) 146-85; Truns., AIME,
Goudouin, M., Tixier. M.P., and Simard. G.L.: Experimental
Study on Influence of Chemical Composition of Electrolytes on
SP Curve, J. Pet. Tech. (Feb. 1957) 58-72: Trans., AIME. 210.
Doll, H.G.: Introduction to Induction Logging and Application
to Logging of Wells Drilled with Oil-base Mud. J. Pet. Te&.
(June 1949) 148-62; Truns.. AIME, 186.
Dumanoir, J.L., Tixier. M.P.. and Martin, M.: Interpretation of
the Inductton-Electrical Log in Fresh Mud, J. Pet. Tech. (July
1957) 202-17: Trans., AIME. 210.
Doll, H.G.: Laterolog-A New Resistivity Loggmg Method
with Electrodes Using an Automatic Focusing System, J. Pet.
Tech. (Nov. 1951) 305-16; Trans., AIME, 192.
Doll, H.G.: Micro Log-A New Electrical Logging Method for
Detailed Determinations of Permeable Beds, J. Pet. Tech. (June
1950) 155-64; Trans., AIME, 189.
Doll, H.G.: The MicroLaterolog, J. Per. Tech. (Jan. 1953)
17-32; Trans., AIME, 198.
Tixier, M.P.: Electrical Log Analysis in the Rocky Mountains.
Oil and Gas J. (June 1949) 48, 143-48.
Tixier, M.P.: Porosity Index in Ltmestone from Electrical
Logs,, Oil and Gas J. (Nov. 1951) 140-42, 169-73.
Wyllie, M.R.J.: Procedures for the Direct Employment of
Neutron Log Data in Electnc Log Interpretation, Geophysics
(Oct. 1952) 17, 790-805.
Tixier, M.P., Alger, R.P., and Tanguy, D.R.: New Develop-
ment in Induction and Sonic Logging, J. Per. Tech. (May 1960)
79; Trans., AIME, 219.
Doll, H.G. and Martin, M.: How to Use Electric Log Data to
Determine Maximum Producible 011 Index in a Formatton. Oil
ad Gas J. (July 1954) 53, 120-26.
Tixier, M.P.: Evaluation of Permeability from Electric Log
Resistivity Gradient, Oil and Gas J. (June 1949) 48, 11323.
a. Resistivity Depanure Curves, Bull., Schlumberger Well
Surveying Corp. (1949).
b. Interpretation Charts for Electric Logs and Contact Logs,
Bull, Welex Inc., A-101.
a. Resistivity Depanure Curves (Beds of infinite Thickness).
Bull., Schlumberger Well Surveying Corp. (1955).
b. Fundamentals of Quantitative Analysis of Electric Logs.
Ed., Welex Inc., A-132.
Poupon, A., Loy, M.E., and Tixier, M.P.: A Contribution to
Electrical Log Interpretation in Shaly Sands, J. Per. Tech. (June
1954) 138-45; Trans., AIME, 201.
Tixier, M.P.: Porosity Balance Verifies Water Saturation Deter-
mined From Logs, .I. Pet. Tech. (July 1958) 161-69; Truns.,
AIME, 213.
Wharton, R.P., er al.: Electromagnetic Propagation Logging:
Advances in Technique and Interpretation, paper SPE 9267
presented at the 1980 SPE Annual Technical Conference and Ex-
hibition, Dallas, Sept. 21-24.
Best, D.L., Gardner, J.S., and Dumanoir, J.L.: A Computer-
Processed Wellsite Log Computation, paper presented at the
1978 SPWLA Annual Logging Symposium, June 13-16.
Coates, G.R., Schulze, R.P., and Throop. W.H.: Volan*-An
Advanced Computational Log Analysis. paper presented at the
1982 SPWLA Annual Logging Symposium, July 669.
Clavier. C.. Coates. G.R., dnd Dumannir. J Thcorcticai and
Expertmental Bases tar the Dual-W&r Model for lntcrpretation 01
Shaly Sands, Sw. Prr. E,tg. J. (April 1984) I S3-6X.
General References
Alger, R.P.: Interpretation of Electrical Logs in Fresh Water Wells in
Unconsolidated Formations, paper presented at the 1966 SPWLA
Annual Logging Symposium, Tulsa, OK, May 8-l I.
Departure Curves for Laterolog, Bull., Schlumberger Well Suney-
mg Corp. (Aug. 1952).
DeWttte. L.: A Study of Electric Log Interpretation Methods in Sha-
ly Formations, J. Per. Tech. (July 1955) 103-10; Trans., AIME,
Doll. H.G.: SP Log in Shaly Sands. J. Per. Tech. (July 1950)
205514; Trun.~. I AIME, 189.
Guyed, H.: Electric Analog of Resistivity Logging, Grophwics
(1955) 615-29.
Guyed. H.: Electric Log Interpretation, Oil Week/~ (Dec. 1955).
Guyeds Electrical Well Logging, Bull., Wellex Inc.. A-132.
Interpretation Handbook for Resistivity Logs, Bull., Schlumberger
Well Surveying Corp. (1949).
Johnson, H.M.: A History of Well Logging, Geophysics (1962)
Jorden, J.R. and Campbell, F.L.: Well Logging I-Rock Properties,
Borehole Environmenr, Mud and Temperature Logging, Monograph
Series, SPE, Dallas (1984).
Keller, G.V.: Modified Mono-Electrodes for Improved Resistivity
Logging, Prod. Monthly (July 1950) 14, 13-16.
Kewer, J.K. and Pmkop, C.L.: Effect of the Presence of Hydmcar-
bons on Well Logging Potential, Oil and Gas J. (Dec. 1955)
Lipson, L.B. and Overton, H.L.: The Effect of Treating Agents on
the Electrochemical Activities of Drilling Mud Filtrates, paper
SPE 867G presented at the 1957 SPE Annual Meeting, Dallas, Oct.
Mayer, C. and Sibbit, A.: Global, A New Approach to Computer
Processed Log Interpretation, paper SPE 9341 presented at the
1980 SPE Annual Technical Conference and Exhibition, Dallas,
Sept. 21-24.
Millican, M.L., Raymer, L.L., and Alger, R.P.: Wellsite Record-
ings of the Movable Oil Plot, paper presented at the 1964 SPWLA
Annual Logging Symposium, Midland, TX, May 13-15.
Morris, R.L. and Biggs, W.P.: Using Log-Derived Values of Water
Saturation and Permeability, paper presented at the 1967 SPWLA
Annual Logging Symposium.
Pitson, S.J.: Formation Evaluation by Log Interpretation, World
Oil (May 1957) 170-83.
Tixier, M.P., Morris, R.L., and Connell, J.G.: Log Evaluation of
Low Resistivity Pay Sands in the Gulf Coast, Log Analyst
(Nov./Dee. 1968).
Log Interpretation, Vol. I-Principles, Vol. II-Applications, Wyllie, M.R.J.: The Fundamentals of Electric Log Interpretation. sec-
Bull., Schlumberger Well Services (1974). ond edition, Academic Press Inc., New York City (1957).
Chapter 50
Nuclear Logging Techniques
Darwin V. Ellis, Schlumberger-Doll Research *
In this chapter, the use of nuclear radiation in wireline
logging will be presented. To avoid repetition, the reader
is referred to Chap. 49 for a basic introduction to the
principles of wireline logging in terms of the operation
and genera1 types of devices used in the area of electrical
logging; Chap. 51 discusses the third major area-
acoustic well logging.
To introduce the general subject of nuclear logging, it
is appropriate to provide a motivation for the use of
nuclear measurement techniques in well logging. This
can be done best by constructing a list of petrophysical
parameters of interest in the evaluation of hydrocarbon-
bearing formations. In the most straightforward applica-
tion, the purpose of well logging is to provide
measurements that can be related to the volume fraction
and type of hydrocarbon present in porous formations. In
the case of openhole logging (as distinguished from
measurements made in a production well with steel cas-
ing), there are four principal parameters of interest: (I)
presence of hydrocarbons, (2) porosity, 4, (3) water
saturation, S,, and (4) permeability, k. To this list, ad-
ditional parameters or descriptors can be added: (5)
lithology, (6) clay identification, and (7) pore fluid iden-
tification. For cased-hole logging the same list of
petrophysical parameters of interest may hold, but with
perhaps more emphasis on fluid identification.
Relationship of Petrophysical Parameters and
Physical Parameters
These petrophysical parameters of interest are derived
normally from a number of measurements provided by
logging services. For the moment, we will concentrate
on some bulk physical parameters associated with them
that may be amenable to measurement through the use of
nuclear techniques.
Aulhcf of the original chapter on this toptc in the 1962 edition was John L.P.
Presence of Hydrocarbons. An obvious method for the
detection of hydrocarbons is based on their chemical
compositions. Since there is no oxygen in most
hydrocarbons, the ratio of the atomic concentration of
carbon to oxygen for a hydrocarbon is significantly dif-
ferent from the ratio for most sedimentary rocks and for-
mation fluids. Thus, a measurement of the ratio of the
number of carbon atoms to oxygen atoms (C/O) con-
tained in a formation would indicate directly the
presence of hydrocarbons when no carbon is present in
the matrix. This is to be contrasted with the method of
electrical measurements, where detection of hydrocar-
bons is based on the contrast of conductivities between
saline water and hydrocarbon in a porous medium.
Porosity. The porosity or nonmatrix portion of a rock
sample can be determined from a measurement of its
bulk density. The fundamental equation that relates the
bulk density, Pb, to the solid matrix, which has a density
pmrr, and the porosity or volume fraction, $, which con-
tains a fluid of density pf, is
Pb=&f+(l-~)Pma. ___. . ...(l)
From this relationship, the porosity, 4, can be deter-
mined from a measurement of bulk density, assuming
that the matrix density and fluid density are known.
These will be known with any precision only if the fluid
type and properties and the lithology are known. In prac-
tical terms, the density range of fluids is between 0.8 and
1.2 g/cm3 (although calcium chloride solutions may
reach 1.4 g/cm3), and most matrix densities are between
2.60 and 2.96 g/cm3.
Another means of detection and quantification of
porosity is based on the fact that the formation porosity is
filled with liquid or gas, all of which contain dispropor-
tionate amounts of hydrogen. This hydrogen may be
associated with the brackish formation water or with the
hydrocarbons. Thus, detection of hydrogen is a means of
inferring porosity in an otherwise solid rock matrix.
Hydrocarbon Saturation. The determination of
hydrocarbon saturation can proceed once the porosity of
a formation has been determined. It can be done (1) by
the direct measurement of C/O and comparison to the
value expected for fully oil- and water-saturated cases,
or (2) by a more indirect measurement of the effective
salinity of the formation in question.
Permeability. There is no clear-cut physical parameter
amenable to nuclear measurements that will predict for-
mation permeability accurately. However, there is one
measurement technique-nuclear magnetic resonance-
that can be related to permeability; it is discussed in
Chap. 53.
Lithology, Clay Types, and Fluid Identification.
These parameters have been grouped together because of
a common approach for their determination, which is
basically some aspect of their chemical composition.
There are two principal interests in identifying the
lithology. One is for a reasonable matrix density to be
assigned to a formation so that porosity can be extracted
from the density measurement. The other is to provide
identification of formations for use in well-to-well cor-
relation. Since the neutron properties of rocks are
somewhat dependent on the type of lithology, it is possi-
ble to determine the three principal lithological matrices
by comparison of the gamma ray attenuation and
neutron-slowing-down properties of the medium. Well-
to-well correlation often is done most simply by com-
parison of the natural radioactivity of the formations.
However, a more direct approach for the identification
of the lithology (i.e., sandstone, limestone, or dolomite)
is not based on the density but rather on the unique
chemical composition of each of these matrices. One
method of identifying the lithology would be to make a
chemical identification of the various elements
associated with the matrix. Another slightly more refined
approach to the determination of the lithology depends
on another bulk property of the material: its average
atomic number. The average atomic number of the for-
mation, which reflects to some extent the lithological
composition, can be obtained by measuring the low-
energy gamma ray absorption properties of the material.
Identification and quantification of clays are much
more difficult, since the chemical compositions of clays
are so varied. Hence, the chemical composition of clays
is a key to their detection. A measurement of the
presence of elements such as Al, Si, Fe, and K must be
counted as a means to their identification. An earlier
technique, which measured the total natural gamma ray
activity of earth formations, was based on the fact that
the naturally occurring radioactive daughter products
(subsequent products of radioactive decay of an element)
of uranium, thorium, and potassium were associated
with clay minerals. Sometimes, however, one or more of
these elements (U, Th, K) is present in a formation con-
taining no clay. Examples of this include the case of
potassium feldspar in the rock matrix or uranium dis-
solved in the formation water.
A third property of clays is the great abundance of
hydrogen associated with the clay mineral structure.
Thus, detection of the presence of hydrogen is another
means of clay identification.
Pore fluid identification is based on indirect
measurements and inferences. The presence of gas in the
formation pores will have a significant impact on the
bulk density for reasonable porosities as well as on the
neutron-slowing-down properties. The distinction be-
tween oil and water again is measured most directly by
the atomic C/O density ratio or based on the thermal
neutron absorption properties of the water phase, which
generally contains chlorides
Fig. 50.1 summarizes the relationships between
petrophysical descriptors and physical parameters,
which can be determined quantitatively through the use
of nuclear radiation and measurement techniques. A
third column has been added to indicate the additional in-
formation necessary to interpret the suggested bulk prop-
erty measured to obtain the desired petrophysical
Physical Parameters and Nuclear Radiation
Before presenting the basic nuclear phenomena
necessary to describe the operation of most of the com-
mon nuclear logging devices, we need to link, in general
terms, the physical parameters discussed previously to
the types of general nuclear techniques that will be
described later in mote detail. To be specific, it should
be pointed out that the types of nuclear radiation used in
well logging are gamma radiation and neutrons. These
two types of penetrating radiation are the only ones that
are able to traverse the pressure housings of the logging
tools and the formation of interest and still return a
measurable signal. It is for this reason that (Y and fi radia-
tion are of no particular interest for exploring formation
characteristics; their penetration ranges are much too
small to be of any practical use.
In the preceding section, it is clear that many of the
proposed parameters to be measured are. in fact, no
mom than the chemical composition of the earth forma-
tion. Instead of the obvious but time-consuming and ex-
pensive chemical analysis of formation samples, a
technique of gamma ray spectroscopy can be used. This
is based on the fact that the nucleus of any atom, after
having been put into an excited state by a previous
nuclear reaction, can emit gamma rays of characteristic
energies, which uniquely identify the atom in question.
Gamma ray spectroscopy refers to the detection and
identification of these characteristic gamma rays.
Another use of gamma rays is in conjunction with the
measurement of bulk density. The bulk density of a
material has a significant influence on the scattering and
transmission of gamma rays through it. At very low
energies, the transmission of gamma rays is influenced
additionally by the chemical composition. This addi-
tional absorption is related to the atomic number, Z, of
the absorber.
The interest in using neutrons in well logging tech-
niques comes from several properties of their interaction
with matter. First, the transmission and moderation of
neutrons are influenced by the bulk properties of the
medium and, in particular, by the amount of hydrogen
present. The scattering of neutrons by hydrogen is very
Petroph ysical Physical
Descriptors Parameter
Clay Identification
Cl0 Ratio
w Lithology
Bulk Densitv
Hydrogen Content
Apparent Salinity
Bulk Density
Hydrogen Content
Average Atomic
t Porosity
Th, U, K
Ca, Si, S, Fe
Al, .
Fig. 5&l--Relationship between petrophysical descriptors and measurable physical parameters
efficient in reducing the neutron energy. Second, in-
teraction of high-energy neutrons with certain nuclei can
excite characteristic gamma rays for subsequent elemen-
tal identification by gamma ray spectroscopy. At very
low energies, neutrons can be absorbed, thereby reduc-
ing the flux, and as a byproduct, another set of
characteristic gamma rays may be emitted. Some of
these capture gamma rays are emitted after some delay
and are referred to as activation gamma rays. So there
are two types of measurements that can be based on the
use of neutrons: the scattering or slowing-down proper-
ties of formations and neutron production of gamma rays
(either by absorption or inelastic high-energy reactions
with elements) of characteristic energies for use in spec-
troscopic identification. Fig. 50.2 illustrates the types of
nuclear measurement techniques that can be used to
measure physical parameters related to the relevant
petrophysical descriptors sought.
Nuclear Physics for Logging Applications
Nuclear Radiation
Nuclear radiation refers to the transport of energy by a
nuclear particle. In the earliest investigation of radioac-
tive materials, three types of radiation were identified
and named, quite unimaginatively, CY, /3, and y radiation.
It subsequently was discovered that (Y radiation consisted
of fast-moving He particles stripped of their electrons
and that /3 radiation consisted of energetic electrons. The
gamma rays were found to be packets of electromagnetic
radiation, also referred to as photons.
The discovery of this radiation then provoked its quan-
tification, namely the measurement of the amount of
energy transported. The unit chosen is known as the
electron-volt (eV), which is equal to the kinetic energy
acquired by an electron accelerated through an electric
potential of 1 V. For the types of radiation discussed in
the following sections, the range of energies is between
fractions of an eV and millions of electrons volts (MeV).
Another convenient multiple for discussing gamma ray
energies is the kiloelectron volt (keV).
Since a! and /3 radiation consist of energetic charged
particles, their interaction with matter is primarily
ionization. That is, they interact with the electrons of
material by losing energy rapidly during their passage
and transferring it to electrons. In most materials their
range is rather limited and is a function of the material
properties (2, the atomic charge or number of electrons
per atom, and its density) and the energy of the particle.
They consequently have not been of any practical impor-
Physical Parameter
Nuclear Measurement
Technique or Parameter
: ,C/O Rati?:-
l Induced Inelastic Gamma Ray Spect.
@Gamma Ray Attenuation
Neutron Slowing-Down Length
Hydrogen Capture Gamma Ray Spect.
*Induced Inelastic Gamma Ray Spect.
Apparent Salinity*
b Thermal Neutron Absorption X-Section
Bulk Density ;~;i. .. 6
Average Atomic Number b
Th. U, KF .. -.
Gamma Ray Attenuation
Gamma Ray Photoelectric Absorption
Natural Gamma Ray Spectroscopy
)lnduced Capture Gamma Ray Spect.
Activiation & High Resolution Spect.
w Gamma Ray Attenuation
b Neutron Diffusion Length
*Thermal Neutron Absorption X-Section
---+ Chlorine Capture Gamma Ray Spect.
Fig. 50.2-Nuclear measurement techniques linked to measurable physical parameters of petro-
physical interest.
tance for well logging applications. Gamma rays, on the
other hand, are extremely penetrating radiation, which
makes them of great importance for well logging
Radioactive decay of certain naturally occurring
substances such as radium was responsible for the
developments mentioned and needs further discussion.
Radioactive decay is a time-varying pmperty of nuclei
whereby a transition from one nuclear energy state to
another lower one is made spontaneously. The result is
that the excess energy is shed by the nucleus by one or
more of the types of radiation previously mentioned. The
basic experimental fact of radioactivity is that the proba-
bility of any one nucleus decaying, within an interval of
time dt, is proportional to dt-i.e., it is independent of
external influences, including the decay of another
nucleus. This probability is proportional only to the time
interval of observation. So for a single radioactive atom
the probability P(dt) of decaying in the interval of time dr
is expressed as
P(dr)=Xdr, . . . . . . .(2)
where X is the decay constant. For a collection of Np
particles, the number decaying, dN, is just
dN= - AdrN,, , . . . . . . . . . . . . . . . . . (3)
resulting in the expression for radioactive decay, namely
where N,, now is the number of particles remaining at
time r, of the initial number of particles Ni . The constant
of proportionality, X, is related to the better-known
parameter, the half-life, t ,,z ,
tlh =-. . . . . . . . . . . . . . . . . . . . . . . . . . .
No physical quantity can ever be measured exactly,
but in the case of nuclear processes where the number of
events observed is small, randomness is important. The
practical complication of this statistical process of
nuclear decay is that only the bulk or average properties
can be predicted with any certainty. We can talk only
about the measurement of a group of particles together
and the distribution of the measured value about some
TO understand one important property of nuclear radia-
tion, it is necessary to digress a moment for a quick
review of the binomial distribution,
which was
discovered in the 18th century by Bernoulli. It describes
the probability, P, , that an event that has a probability P
of occurring will occur x times when the observation is
repeated z times. The probability thus specified was
identified with the binomial expansion of (P+q)i, where
q=l-P, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...(6)
so that the general term of the expansion is
p,= z!
-P( 1 -,)ZPX,
. . , . . . . . . . . . (7)
which gives the probability of x occurrences in z trials.
This expression can be applied to radioactive decay, in
which P, represents the probability of having x nuclei
decay in time dt when there are z atoms present. For this
case, generally the probability P is very small but the
number of particles observed (z) is very large so that Eq.
7 simplifies to
P,=/L~--, . . . . . . .(8)
which is known as the Poisson distribution. It gives the
probability of observing x decays in a given time where
an average of ji decays is to be expected. Fig. 50.3
shows the general form of the Poisson distribution with
the maximum probability at the mean value, which was
taken as 100 for this example. The curve resembles the
usual bell-shaped distribution curve with a width
specified by a parameter u, the standard deviation.
The practical importance of this discussion is that the
appropriate u for the Poisson distribution that
characterizes the statistics of counting random nuclear
events is not an independent parameter (as is the case for
most measurements) but is related to the mean value ,%
0=x$ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...(9)
This means that if N,. counts from a radiation detector
are expected per time interval, then, in repeated observa-
tions, about 32% of the measurements will deviate
beyond the value of N, + 6. This is a quantitative
description of the statistics associated with all nuclear
logging techniques.
Number of Occurrences, x
Fig. 50.3-The Poisson distribution for the case of mean value
(CL) of 100. The probability, P,, is shown for values
of x around the expected mean value.
There have been a number of clever techniques for
dealing with the statistical fluctuations inherent in
nuclear counting rates. Probably best known is the R-C
circuit, which has a time constant associated with it and
permits the recording of a continuous moving average.
There are now more modem digital signal processing
techniques (such as Kalman filtering) to provide more
refined filtered counting rates or outputs derived from
statistically varying counting rates. However, the only
sure approach to reduce the fluctuations is to increase the
average number measured, either by using higher-output
sources, more efficient counters, or longer counting
times per sample.
Particle Reactions. There are certain nuclear particle in-
teractions of interest to well logging. To discuss these as
necessary in the following sections, a few mathematical
definitions are presented here to help the mechanisms of
the reactions.
As in radioactive decay, the process of nuclear reac-
tions is also statistical in nature. The question of interest
is how readily these reactions will take place. Fig. 50.4
shows the idealization of the nuclear reaction process. A
beam of radiation (it may be gamma rays or neutrons) of
an intensity pi is seen to enter the slab of material. The
intensity of the radiation, 9 i , is called the flux and has
units of numbers of particles per unit surface area per
unit time.
The slab of material is characterized by N,, the
number of particles per unit volume with which the flux
of radiation may interact. The experimental fact ob-
served is that after passing through a slab of material of
thickness 6/z, a certain fraction of the incident particles
have undergone interactions, and that number is propor-
tional to the thickness and the number of target nuclei
and the incident flux. This is expressed mathematically
6\k=a~Np6h, . . . . . . . (10)
N, nuclei cm3
This forms the basis for the cheapest, easiest, and most
reliable method for neutron production. The physical ex-
planations of this reaction are beyond the scope of in-
terest of the present work and may be found in Refs. 1
and 2. The practical construction of a neutron source
consists of finding a naturally occurring cr emitter and
mixing it with an appropriate light element having a
large (cy,n)* cross section. Some possibilities for (Y emit-
ters are Pu, Ra, Am, and PO. Three target elements are
Be, B, and Li. The actual spectrum (energy distribution)
of emitted neutrons is quite complicated and depends
somewhat on the geometric details of the cy emitter and
target; but generally speaking, the peak of the neutron
distribution is around 4 MeV.
Another method of exploiting particle-induced reac-
tions is by use of charged particle accelerators. In one
realization currently used in well logging, deuterium and
tritium ions are accelerated toward a target impregnated
Fig. 50.4-Idealized view of nuclear radlation interactlng with
with the hydrogen isotopes deuterium (D) and tritium
matter showing the reduction in flux traversing a thick-
(T). The reaction is written as
ness of material characterized by the number of in-
teracting particles per cubic centimeter.
D+T-+4He+n+ 17.6MeV. . . .(l3)
where the constant of proportionality, (T, is called the
cross section of the interaction. The units of this
microscopic cross section u are area/interacting target
nucleus. Cross section is used because in a classical
sense it is the apparent area each target nucleus presents
to the incoming beam. In effect, it collects all the nuclear
interaction details into one useful number. The practical
unit is called the barn and is equal to lO-24 cm. The
macroscopic cross section, C, is the product of (T and N,
and has the dimension of area/particle times par-
ticles/volume or inverse length. In practical terms, C can
be calculated easily because Ni, is related to Avogadros
number, NA , and the material density, ph. by
. . ..___...____.__. (11)
where M is the molecular weight of the target for a single
particle per molecule.
In general, the cross sections for most reactions must
be determined experimentally and are often available in
graphical or tabular form. The quantity &N,, in Eq. IO
has dimensions of (cm . set) - and has the interpreta-
tion of the reaction rate per unit volume resulting from
the incident flux.
Nuclear Reactions. To discuss the second type of radia-
tion of great importance to well logging applica-
tions-neutrons. which are not the result of naturally oc-
curring radioactive decay schemes-a brief discussion of
artificial or induced radioactivity is given here.
The classic reaction that inspired the discovery of the
neutron was the bombardment of beryllium by alpha par-
ticles and can be written as
The cross section for this reaction has a maximum at
about 100 keV of D projectile energy, which dictates the
accelerating voltages in such a device.
Despite the engineering difficulties of constructing
such a device, the advantages for logging are many. One
is the relative high energy of the produced neutrons.
They are emitted at 14.1 MeV (nor 17.6 MeV, because
some of the energy of this reaction is given up to the
alpha particle). These high-energy neutrons are useful
for producing other interesting nuclear reactions in the
formation, as discussed later. The other advantage is that
a source of this type can be controlled-i.e., switched off
and on at will. This provides a degree of safety un-
paralleled for radioactive sources as well as opening the
door to measurements involving timing as a means of
determining some interesting nuclear properties of the
Now that we have covered the two types of nuclear
radiations currently used in logging devices, let us ex-
amine how gamma rays and neutrons interact with matter
and define some macroscopic properties of matter that
can be used to characterize this behavior.
Fundamentals of Gamma Ray Interactions
For the purposes of our discussion there are three types
of gamma ray interactions that are of interest: the
photoelectric effect, Compton scattering, and pair pro-
duction. The type of interaction a gamma ray will
undergo depends on the properties of the material and the
energy of the gamma ray. The ordering of these three in-
teractions reflects the transition of the dominant process
as the gamma ray energy increases.
The photoelectric effect concerns the interaction of a
gamma ray with an atomic electron in the material. In
this process the incident gamma ray disappears and
transfers its energy to the bound electron. Depending on
This shorthand (a, n) tndlcates a reamon of an /a parWe wth an unspec,f,ed
9Be+4He+ C+n+5.76 MeV. .(12)
nucleus, n. resulttng I the produclm of a neulron and anolher uns,,ec,f,ed
the energy of the incident gamma ray, generally the elec-
tron is freed from its nucleus and begins collisions with
the adjacent material. Normally the ejected electron is
replaced by another electron with the accompanying
emission of a characteristic fluorescence X-ray with an
energy dependent on the atomic number of the material
and generally below 100 keV.
The cross section for the photoelectric effect, P,,~~,
varies strongly with energy. falling off as nearly the cube
of the gamma ray energy (EGO). It is also highly depen-
dent on the atomic number (Z) of the absorbing medium.
In the energy range of 40 to 80 keV, the cross section per
atom of atomic number Z is given by
For most earth formations the photoelectric effect
becomes the dominant process for gamma ray energies
below about 100 keV.
The photoelectric effect is an important process in
understanding one of the conventional gamma ray detec-
tion devices and a well logging tool that is sensitive to
the lithology of the scattering formation. The tool in
question makes a measurement of the photoelectric ab-
sorption factor, F,, , which is proportional to the
photoelectric cross section per electron.
Since this quantity is very sensitive to the average
atomic number of the medium, Z, it can be used to obtain
a direct measurement of the lithology of the scattering
medium. This is because the principal rock matrices
(sandstone, limestone, and dolomite) have considerably
different photoelectric absorption characteristics and the
pore fluids play only a minor role because of their
low Zs,
The Compton scattering process involves the interac-
tion of a gamma ray and an electron. It is a process in
which only part of the gamma ray energy is imparted to
the electron and the energy of the gamma ray, conse-
quently, is reduced. Unlike the photoelectric effect, the
probability for Compton scattering changes relatively
slowly with energy.
To see the bulk effect of Compton scattering in a
material consisting of nuclei of atomic mass A and
atomic number Z, one can use the linear absorption coef-
ficient, which is just the Compton cross section, gc(,,
multiplied by the number of electrons per cubic
cc0 =u& Ap~Z, . . (15)
The final Z factor in this equation takes into account that
there are Z electrons per atom. Consequently, the at-
tenuation of gamma rays resulting from Compton scat-
tering will be some function of the bulk density (ph) and
the ratio of Z/A. The fact that Z/A = % for most elements
of interest is the basis for the determination of bulk den-
sity from gamma ray scattering devices.
. . . . .
where h, in this case, is very nearly the source-to-
detector spacing.
It is convenient to define the electron density index as
The third and final gamma ray interaction is that of
pair production. It, like the photoelectric process, is one
of absorotion rather than scattering. In this case the eam-
. . . . (18)
1 c n
.Ol 0.1 1.0 10
Gamma Ray Energy (MeV)
Fig. 50.5~Regions of dominance of the three major gamma ray
interactions in terms of the gamma ray energy, E,,.
and the atomic number, 2, of the target material. The
two lines separating the three regions indicate where
the two adjacent interactlons occur with equal prob-
ma ray interacts with the electric field of the nucleus. and
if the gamma ray energy is above the threshold energy of
1.022 MeV, it disappears and an electron/positron pair is
formed. The subsequent annihilation of the positron
(positively charged electron) results in the emission of
two gamma rays of 5 1 I keV each. The cross section of
this process is somewhat energy-dependent and is zero
below the required threshold energy of 1.022 MeV. In
addition, it also depends on the charge of the nucleus.
To establish the regions of dominance of the three
types of interactions, refer to Fig. 50.5. It shows, as a
function of gamma ray energy and atomic number of the
absorber, the regions in which the probabilities of the
various processes are equal. The regions of dominance
are quite clear.
From the earlier definition of cross section, the fun-
damental law of gamma ray attenuation can be stated as
l=P,e-h, _. _. _. . _. .(l6)
where *; is the flux incident on a scatterer of thickness
h, n is the number of scatterers per unit volume, and u is
the cross section for scattering per scatterer.
In the case of gamma-gamma density devices, the
source of gamma rays is chosen to have an energy for
which the primary interaction is Compton scattering. In
this case the scatterers are electrons and u refers to the
Compton cross section per electron. This results in the
following expression for the attenuation of the source
energy gamma rays:
0.1 1 10
Gamma Ray Energy, MeV
Fig. 50.6-The mass absorption coefficient for gamma ray
Fig. 60.7-The relationship between neutron energy and speed
interactions in aluminum. for the three broad classifications of neutron energies.
so that the attenuation of the gamma rays is seen to be
proportional to the spacing, h, between source and detec-
tor and the electron density index, which in turn can be
related to the bulk density if the properties (specifically
Z/A) of the scattering material are known. For most
sedimentary rocks the ratio of Z/A is nearly r/z so that p r
is very nearly equal to ph.
Another unit for measuring the gamma ray attenuation
properties of a material is the mass absorption coeffi-
cient, K, ,* which regroups the constants in Eq.
K,=ZNAo, . . . . . . . . . . . . . . . ..I.. . . (19)
so that the gamma ray attenuation equation can be writ-
ten as
*=9$-K-. . . . . . . . . . . . .(20)
The convenience of the mass absorption coefficient for
Compton scattering is that it is remarkably similar for all
materials since Z/A = % and the density dependence has
been eliminated. Fig. 50.6 shows the mass attenuation
coefficients (in cm*/& for aluminum. This element,
with a density of 2.7 g/cm3 and atomic number of 13, is
quite typical of earth formations. The average atomic
number ranges from 11 to 16 between quartz and
limestone, while the grain densities are between 2.65
and 2.71 g/cm3.
Fundamentals of Neutron Interactions
As in the case of gamma rays, the interaction of neutrons
with materials can be categorized by the types of interac-
tions with the appropriate cross sections that describe
' The symbol used i n physi cs tormass absorpti on cOef f l Ci enl I s P
Neutron Speed
Source :. ::
2. 2 -
0. 22 -
O. i l 0. 1 t 10 102 103 104 105 106 10' Energy (eV)
1 keV 1MeV
each one. The reactions of neutrons with matter are
much more varied and complex than those of the gamma
rays. For simplification we will confine ourselves to four
principal types of interactions of neutrons with matter.
Fig. 50.7 defines in broad terms the energy range of
interest for neutrons. For logging applications it can be
seen that the energy range of interest is over about 9
decades: from source neutrons of 5 to 15 MeV in the
broad fast neutron range above 10 eV, to epithermal
neutrons in the range of 0.2 to 10 eV and thermal
neutrons, which are distributed around 0.025 eV at room
To have an idea of the time scale for later discussions
of the themralization process, it is useful to note the rela-
tionship between neutron energy and its associated
velocity. To evaluate the velocity of a neutron, we can
use (at low energies) the classical relationship between
kinetic energy, Ek, velocity, v, and mass, m,
Ek=%mv2, . . . . . . . . . . . . . . . . . . . . . . . . . . . ..(21)
so that the velocity, v, is given by
v= -. . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..(22)
If this expression for velocity is evaluated for thermal
energies (0.025 eV), the result is 2200 m/s or 0.22
cm/ps. Thus, the velocity at any energy E (in eV) is
given by
v=o.22 ~ o,025 . . . . . . . . . . . . . . . . . .
where v is the velocity, cm/pLs. Therefore, the speed of
an epithermal neutron of 2.5 eV is 2.2 cmips, and for a
near-source energy neutron of 2.5 MeV the velocity is
2200 cmips. These velocities are also noted on Fig.
Of the four principal types of interactions, the first two
generally are referred to as moderating interactions, or
interactions in which the energy (or speed) of the neutron
is reduced. One of these is known as elastic scattering
and the other, inelastic scattering. Classical mechanics
(elastic billiard ball analysis) can describe the
moderating power of the struck nucleus. The energy of
the neutron is reduced more efficiently as the mass of the
struck nucleus approaches the mass of the neutron. Thus,
hydrogen and other low-atomic-mass elements are quite
efficient in reducing fast-neutron energy. Fig. 50.8 il-
lustrates, for elastic neutron scattering with several
elements, the range of reduction in neutron energy
available for a single collision. It is seen that for the most
common earth formation elements the maximum energy
reduction per collision for the heavy elements is about 10
to 25 %, However, for the case of hydrogen it is seen that
the entire neutronenergy can be lost in a single collision.
In the case of inelastic scattering, a portion of the
energy of the incident neutron goes into exciting the
target nucleus. This reduces the energy of the incident
neutron and, in addition, the target nucleus usually will
produce a characteristic gamma ray upon de-excitation.
This type of reaction always has a threshold energy
(below which it will not happen) associated with it and is
exploited in the measurement of the C/O ratio in earth
The second general category of neutron interactions is
known as absorptive interactions. The two general types
are radiative capture and reactions in general. In
radiative capture, unlike the moderating interactions
considered above, the neutron (usually near thermal
energies) is absorbed by the target nucleus and then
disappears, and subsequent characteristic gamma rays
are produced. .The general category of neutron reactions
is quite broad; it will be sufficient to say that the interac-
tion of neutrons with other nuclei can provoke the emis-
sion of other particles such as alphas, protons, /3s, or
even several subsequent neutrons. All these reactions,
although common, have a very small probability for hap-
pening relative to the other interactions of interest to us
and usually occur over a restricted and high-energy
To show the complexity of the cross sections for
neutron interactions see Fig. 50.9, which schematically
indicates the variations with energy. The top figure
refers to the total cross section as a function of neutron
energy, EN, and the four following figures indicate how
this can be decomposed. The first line (n,n) refers to
elastic scattering, which is shown to be rather constant
with energy except for some resonances at low energies.
The next line shows inelastic interactions (n,n) showing
some characteristic threshold below which this reaction
is not possible; the fourth line is one of the many particle
reactions possible (n,(y); and the final line (although
there could be others) is the radiative capture (n,?),
which is seen to increase in probability at low energies.
Despite these complexities, there are some gross prop-
erties that can be assigned to materials on the basis of
I 1 I I
0 H 10 20 30 40
A, Atomic Mass
Fig. 50.8-The range of energy reduction possible for neutron
elastic scattering with several important elements for
formation evaluation. E, is the energy before scat-
tering, and E is the energy after scattering. Hydro-
gen is seen to provide the greatest possible energy
reduction for a single collision.
Total 11
tn. n)
(n, n)
,::, ( 1
Neutron Energy, EN
Fig. 50.9-A schematic illustration of the components involved
for the total neutron cross sectlon as a function of
energy. The characteristics of four specific cross sec-
tions are shown.
/-,40 p*u.
,:::i:::-20 p,u,
0 p.u.
0.1 ~
0.1 1.0
Neutron Energy, MeV
Fig. 50.10-The mean free path of fast neulrons in water and water-filled limestone at several porosities as a function of energy.
their neutron cross sections. The first is the macroscopic
cross section, which is defined as the product of the cross
section in question times the number of atoms per cubic
centimeter, N,-i.e.,
Ei =NpUi =- CTi . . . . . . . .
The dimensions of the macroscopic cross section C; are
inverse centimeters and the interpretation is that its
reciprocal is the mean-free-path length in centimeters be-
tween interactions of Type i. Frequently in logging, use
is made of the macroscopic absorption cross section at
thermal energies. The units of this (so-called capture units,
c.u.) are just 1,000 times the Cj defined previously,
where Ui refers to the thermal absorption cross section,
which is dominant at thermal energies for most elements.
Fig. 50.10 shows the total mean free path in limestone
of 0, 20, 40, and 100 porosity units (PU) as a function
of energy for fast neutrons. At the energy of chemical
source emission (2 to 4 MeV), it is seen that there is very
little porosity dependence. It is only as the neutrons are
slowed down that the mean free path becomes strongly
As mentioned earlier, in the case of elastic scattering.
low-mass nuclei are more efficient in reducing the ener-
gy of the ccattered neutron. As can be inferred from Fig.
50.8, the result of a collision can be considered. on aver-
age, as a percentage decrease of the neutron energy. This
usually is expressed as the average logarithmic energy
decrement, [ :
=I@;)--in(E)= -ln(EIE;). . . (25)
It can be shown from classic mechanics that the average
log energy decrement is simply related to the atomic mass,
A, of the struck nucleus by
A+2/3 . . . . . . . . . . . . . . . . . . . . . . . . . . . .
for large values of atomic mass A. The average log ener-
gy decrement allows an estimation of the average num-
ber of collisions, n, to reduce the neutron from an initial
energy Ei to some lower energy E from the following
reasoning. If the sequence E 1, E2 . . E, represents the
average energy after each collision, then we can write
ln(%) =ln($$. .F) . . .
=ln($) =n In($) _.,
=nf. . . . . . . . . . . . . . . . . . . . . . . . . . ..(29)
20-PU limestone
0-PU limestone
0. 12
0. 05
0. 92
0. 514
14. 5
15. 8
29. 7
Average number of col l wons f rom4. 2 MeV to 1 eV
Thus, the average number of collisions is given by
. . . . . . . . . . . . . . . . . . . . . . . . . .
The constant .$ can be computed for a mixture of elements
by weighting the value of each 2: i for element i with the
appropriate total scattering cross section u;. Table 50.1
shows some typical values for the average logarithmic
energy decrement and the number of collisions necessary
to reduce source energy neutrons (4.2 MeV) to 1 eV.
There are two more parameters that help to character-
ize neutron interactions with bulk material. One parame-
ter is known as the slowing-down length, L,Y, and the
other as the diffusion length, Ld. L,s can be described as
roughly proportional to the average distance a neutron (in
an infinite homogeneous medium) travels from its emis-
sion at high energy until it arrives at the lower edge of
the epithermal energy region. This distance can be
calculated4 with a detailed knowledge of the cross sec-
tions of the constituent elements. Fig. 50.11 shows the
variation of L, as a function of water-filled porosity for
limestone, sandstone, and dolomite.
Ld can be thought of as the distance a thermal energy
neutron travels between the point at which it became ther-
mal until its final capture. This distance is given by
Ld=v@E), . . . . . . . . . . . . . . . . . . . . . . . . . . ..(31)
. Sandstone
30 40 50 60 I D 80 90
Water- Fi l l ed Porosi ty. p. u.
Fig. 50.11-The calculated slowing-down length, L,. as a func-
tion of water-filled porosity for three rock matrices:
sandstone, limestone and dolomite.
where D is the thermal diffusion coefficient and c is the
macroscopic thermal absorption cross section of the ma-
terial. The diffusion coefficient, D, also can be calculated
from the knowledge of the cross sections of the material
and is shown in Fig. 50.12 as a function of porosity for
the three principle matrices.
Since thermal neutrons will be affected strongly by the
presence of thermal absorbers, it is interesting to look at
an abbreviated list of elements that frequently are found
in formations that have large macroscopic thermal absorp-
tion cross sections, This is found in Table 50.2, where
the units are capture cross section (c.u.) per gram of ma-
terial. Of particular interest is chlorine, the implication
being that salt water will have some measurable effect on
the thermal neutron population as well as iron and bo-
ron, which frequently are associated with clays.
Fig. 50.12-The calculated thermal diffusion coefficient, D, as
a function of water-filled porosity for three rock ma-
trices: sandstone, limestone, and dolomite.
Gamma Ray
lnsulat& End Plate
Fig. 50.13-Components of a gas-discharge radiation detector.
Multiplier Tube
Fig. 50.14-Schematic of the steps involved in gamma ray de-
tection by the production of a measurable electri-
cal signal in a photomultiplier coupled to an Nal
Another auxiliary parameter, the migration length
(L,), has been defined as
L; =L;+L~. . . . . . . . (32)
It can be viewed as a distance that represents the combi-
nation of the path traveled during the slowing-down phase
(L,) and the distance traveled in the thermal phase be-
fore being captured (Ld). The use of this parameter pro-
vides a convenient way of predicting the response of a
thermal neutron porosity device, which is discussed in
more detail in a later section.
Nuclear Radiation Detectors
Gamma Ray Detectors. The devices for the detection
of gamma rays involve the exploitation of one or more
of the three processes of gamma ray interactions with mat-
ter described earlier. Three general types of gamma ray
detectors in current use will be described next. The first
variety, the ionized-gas counter, is a direct descendant
of the earliest efforts in nuclear radiation detection. The
second and most common present-day gamma ray detec-
tor used in well logging applications is the scintillation
detector. The third type of device, the solid-state detec-
tor, is just beginning to be used in logging applications.
The common form of the ionized-gas or gas-discharge
counter consists of a metal cylinder with an axial wire
passing through it (Fig. 50.13) and insulated from it. The
cylinder is filled with a gas that is normally nonconduc-
tive, and some moderate (several hundreds of volts) elec-
trical potential is maintained between the central wire and
the cylinder. The detection process is initiated by the for-
mation of some ionized gas molecules. These freed elec-
trons are accelerated by the radial electric field and in
successive collisions produce additional free electrons,
which finally results in a measurable charge collection on
the central wire.
For gamma rays to be detected with such a device, the
gas somehow must be ionized initially. Since the gas den-
sity is moderate, even at rather high pressure available
in commercial tubes, and the atomic number of useful
gases is relatively low, there is little possibility of the gam-
ma rays interacting directly with the gas. The main de-
tection mechanism is photoelectric absorption or recoil
electron ejection from Compton scattering in the metal
shield. For the gamma rays absorbed near the inner radius
of the cylinder, there is some probability of the ejected
electron escaping into the gas and providing the initial
ionization. This also is illustrated in Fig. 50.13.
It should be evident from the foregoing discussion that
the detection efficiency of such detectors is not high. It
can be improved somewhat by the incorporation of con-
ductive high-atomic-number gamma absorbers, such as
silver, as an inner lining of the cylinder. Although they
can be operated in a proportional mode, the energy reso-
lution of these detectors is not of great practical use. The
most positive aspects of gas-discharge counters are their
simplicity, ruggedness, and reliability for functioning in
the hostile environment of well logging. Because of their
poor efficiency and inapplicability to spectroscopic gam-
ma ray detection, they are being replaced rapidly by a
newer generation of scintillation detectors.
A more common type of gamma ray detector uses a
scintillation crystal. Once again, the active detector ele-
ment is sensitive to ionizing radiation, such as energetic
electrons. When these particles travel within the crystal
lattice, they impart their energy to a cascade of secon-
dary electrons, which finally are trapped by impurity
atoms. As the electrons are trapped, visible or near-visible
light is emitted. The light flashes are then detected by a
photomultiplier tube optically coupled to the crystal and
transformed into an electrical pulse. This is indicated sche-
matically in Fig. 50.14. The output pulse height can be
related to the total energy deposited in the crystal by the
initial high-energy electron. The great advantage of such
a detection scheme is the possibility of performing gam-
ma ray spectroscopy--that is, to detect the actual energy
of the incident gamma ray, which, in some cases, will
identify uniquely the source of the emitted gamma ray,
as in the case of induced gamma ray logging.
However, a scintillation detector is a detector of gam-
ma rays only to the extent that an electron is produced
in the crystal through one or more of the three basic gam-
Boron 42 300
Chlorine 564
Hydrogen 198
Manganese 146
Iron 27.5
ma ray interaction mechanisms: photoelectric absorption,
Compton scattering, and pair production. Thus, the gam-
ma ray detection efficiency of a scintillator will depend
on its size, density, and average atomic number (for photo-
electric absorption). A scintillator in common use is a
crystal of sodium iodide doped with a thallium impurity,
NaI (Tl), which has good gamma ray absorption proper-
ties and a fairly rapid scintillation decay time (- 0.23
psec) to allow for high-counting-rate spectroscopy.
The use of such a device for gamma ray spectroscopy
implies that the output light pulse is proportional to the
incident gamma ray energy; however, this is possible only
for the case of total absorption of the gamma ray. Some
of the difficulties that can complicate the detected spec-
trum are shown in Fig. 50.15, for the case of a tool de-
signed to look for the unique gamma rays emitted by
inelastic neutron reactions with carbon and oxygen. The
figure illustrates what might happen to an inelastic car-
bon gamma ray that is produced at the site marked (IS)
with an initial energy of 4.44 MeV. It first makes a Comp-
ton scattering in the borehole fluid (CS) and loses 90 keV
of energy before traversing the tool housing and entering
the NaI detector with an energy of 4.05 MeV. At the point
marked (PP) it suffers a pair-production interaction,
producing one electron and one positron with energies of
2.00 and 1.03 MeV, respectively, the missing 1.02 MeV
having gone into the creation of the electron/positron pair.
Both particles impart their energy to the scintillation proc-
ess indicated by the dashed lines. When the positron has
given up all its kinetic energy, it annihilates with an elec-
tron to produce two gamma rays, each of 0.51 MeV ener-
gy. One of the gamma rays undergoes Compton scattering
at (CS), and the reduced-energy gamma ray (0.41 MeV)
is finally absorbed photoelectrically within the crystal at
point (Ph.A). The other 0.51-MeV gamma ray is shown
escaping the crystal, to the right, and being absorbed in
the tool housing without contributing to the total energy
transferred to the crystal. The energy recorded by the
crystal for the event depicted is 3.54 MeV (4.05
MeV - 1.02 MeV pair-production $0.511 MeV annihi-
lation) instead of the 4.44 MeV that we would like to be
Optical Coupling
Na(TI) Crystal
Reflective Coating
Fig. 50.15~Illustration of the possible sources of gamma ray
energy degradation in an Nal detector system.
Thus, %. t e degradation of the structure of the incident
gamma ray spectrum is seen to be inherent in the physics
of the many processes involved in the detection. Only if
the gamma ray is absorbed totally by the detector is the
light output of the scintillator proportional to the incident
gamma ray energy. This would be the case for the photo-
electric absorption, for example. Fig. 50.16 shows the
energy deposited in this case as the single line to the right
marked E,.
Full Energy
Light Flashes
Produced by
Compton Recoil
Electrons /
If only a Compton interaction occurs, then a fraction
of the energy will be registered. The possible range of
energy deposition in this case follows the distribution
shown in Fig. 50.16 from zero to the Compton edge,
which corresponds to the maximum energy being trans-
ferred from the gamma ray to the electron. Additionally,
if the gamma ray is of sufficiently high energy there may
be a pair-production reaction, and if one or more of the
5 1 1-keV photons escapes the detector without interaction,
the so-called first and second escape peaks will be pro-
duced in the detected spectrum. Fig. 50.17 indicates the
additional distortion introduced by this process.
Energy Transferred Compton Ey
to Crystal Edge
In addition to the distortions in the measured spectrum
produced by the possible interactions within the detector,
Fig. 50.16~Idealized response from a scintillation detector to
mono-energetic gamma rays of energy E, showing
the photo-peak and the Compton tail.
Energy Transferred to Crystal
Fig. X1.17-Idealized spectrum distortion in a scintillator caused
by pair-production. The highest energy peak corre-
sponds to photoelectric absorption or the full ener-
gy of the incident monoenergetic gamma ray, and
the two lower energy escape peaks correspond to
escape from the crystal of one or two of the annihi-
lation gamma rays of 511 keV.
a dominant perturbation to the measurement is the detec-
tor resolution. This refers to the broadening of the line
spectra, as can be observed clearly in Fig. 50.17. The
width of the observed gamma ray lines is, in the case of
an NaI detector, primarily a function of the gamma ray
energy, the size of the crystal, and the optical coupling
between the crystal and photomultiplier, as well as the
characteristics of the photomultiplier. One of the major
drawbacks of the scintillation detectors is their poor ener-
gy resolution. The reason is that detection in this type of
device requires a number of inefficient steps, the result
being that the energy required to produce one informa-
tion carrier (a photo-electron in the photo-multiplier) is
about 1,000 eV. Thus, the number of carriers for a typi-
cal radiation detection is rather small; the statistical fluc-
tuations on such a small number place an inherent
limitation on the energy resolution.
The use of semiconductor materials as radiation detec-
tors can produce many more information carriers per de-
tected event and, thus, can achieve a very-high-energy
resolution. In a solid-state device such as the germanium
detector, the semiconductor properties are used to trans-
fer the charged-particle energy into a usable electrical
pulse in a much more direct manner. When a gamma ray
interacts with the detector, charged particles are produced.
These, in turn, transfer energy to electrons bound (by only
0.7 eV for Ge) in the crystal lattice, enabling many of
them to become free. Each free electron leaves a positive
hole in the electron structure of the crystal. Under a strong
electrical field applied to the detector crystal, the free elec-
trons and holes migrate quickly to the electrodes and cre-
ate an electrical impulse.
The excellent resolution arises because the band gap
is so small. About 3.5 x lo5 electrons are freed by the
detection of a l-MeV gamma ray to contribute to the re-
sulting pulse with no intervening inefficient steps. The
result is sharp energy resolution. Another result is that
the detector must be operated at extremely low tempera-
tures. This is because at room temperatures the electrons
have sufficient energy to cross the 0.7.eV band gap and
camouflage those freed by gamma ray interactions.
Although the gamma ray spectra obtained with Gc detec-
tors are superb, their overall counting rates are less than
those obtained by NaI detectors. Application of solid-state
detectors is limited to devices concerned with precise spec-
troscopic elemental definition or in-situ chemical analysis.
Neutron Detectors. Neutrons are detected through
nuclear reactions in which energetic charged particles are
produced. Thus, most neutron detectors consist of a tar-
get material for this conversion coupled with a conven-
tional detector, such as a proportional counter or
scintillator, to achieve the measurement. Since the cross
section for neutron interactions in most materials is a
strong function of neutron energy, different techniques
have been developed for different energy regions. For well
logging applications, at present, it is the detection of ther-
mal and epithermal neutrons that is of interest. The de
tection schemes considered in this section are appropriate
for these low-energy neutrons.
The determination of useful nuclear reactions for neu-
tron detectors involves satisfying several criteria: the cross
section for reaction must be very large, the target nuclide
should be of high isotopic abundance. and the energy
liberated in the reaction following the neutron capture
should be high for subsequent ease of detection by con-
ventional means. Three target nuclei have been found
generally to satisfy these conditions: B, Li, and He.
In the case of the first two targets, the (n,ol) reaction is
used, and for 3He it is the (n,p) reaction.
The boron reaction is exploited widely in the form of
BF3 in a proportional tube. In this case the boron trifluo-
ride serves as the target and as the proportional tube gas.
For this application the gas is enriched in B, to attain
a high detection efficiency. Another approach is to use
boron as the inner coating of a proportional counter, which
may use some other proportional gas more suitable than
BFJ for applications involving fast timing, for example.
Since a suitable lithium compound gas does not exist,
the lithium reaction is not exploited in a proportional coun-
ter. However, lithium scintillators, similar to those of so-
dium iodide for gamma ray detection, are available.
Because of the large amount of energy released by the
(n,cr) reaction, neutrons are registered at an energy of
about 4. I MeV, which provides a means of discriminat-
ing against the gamma rays, which also will be detected
readily by the LiI crystal.
The most common neutron detector in well logging,
however, is based on the He (n,p) reaction. In this case
He is used as the target and proportional gas in a coun-
ter. It is preferred to BF3 because it has a higher cross
section than the boron reaction and the gas pressure can
be made much higher without degradation of its propor-
tional operation. The overall simplicity of a proportional
tube is preferred to the additional complications associated
with a scintillator.
For the three reactions discussed, the cross sections vary
inversely with the square root of the neutron energy 50
that the detection efficiency for neutrons will vary in the
same manner. The detectors using these reactions, then,
are basically thermal neutron detectors. For some logging
applications, it is desirable to measure the cpithcrmal neu-
tron flux while being insensitive to thermal neutrons. This
can be achieved by making a minor modification to any
of the three types of detectors previously mentioned. It
consists of using an exterior thermal-neutron-absorbing
material with a large cross section, such as cadmium. to
shield the detector. Thermal neutrons will be absorbed
in the shield, but the reaction particles. whose range is
small (on the order of tenths of millimeters). will not reach
the counter. The higher-energy neutrons that manage to
penetrate the shield will be detected by the thermal neu-
tron detector with somewhat reduced efficiency.
Nuclear Radiation Logging Devices
The logging devices discussed in the following section fall
under two general categories: those that measure natural
radiation fields and those that produce radiation fields and
measure some aspect of their interaction with the forma-
tion. The first group contains tools that measure the natur-
al gamma ray activity of earth formations resulting from
the spontaneous decay of radioactive materials. The sec-
ond category can be broken down into the type of radia-
tion used-gamma rays or neutrons. The latter may be
subdivided further into the use of chemical or steady-state
neutron sources or pulsed particle accelerator-based
sources described earlier.
Rather than trace the historical development, which has
been well documented by Segesman, only the most re-
cent logging devices will be discussed. Both neutron
porosity and gamma-gamma density devices have under-
gone substantial evolution since their respective introduc-
tions as commercial services. The earliest devices
invariably used a single detector. As the use of these types
of measurements grew, more emphasis was put on im-
proving the quantitative nature of the measurements and
a better appreciation of environmental effects was gained.
This led to the development of borehole-compensated
devices generally using a second detector at a lesser spat-
ing from the source that, because of its larger sensitivity
to environmental effects, provides a correction to be ap-
plied to the principal detector.
Gamma Ray Devices
There are two series of naturally occurring radioactive
isotopes that occur in significant quantities in sedimen-
tary rocks: the uranium and thorium series. The only other
significant naturally occurring radioisotope is that of
potassium (40K). Clay minerals that are formed during
the decomposition of igneous rocks in general have a very
high cation exchange capacity. Because of this property
they are able to retain trace amounts of radioactive min-
erals that originally may have been components of the
feldspars and micas that go into the production of clay
minerals. This process generally results in a higher con-
centration of radioactive elements in shales than in sand-
stones or carbonate rocks not produced by weathering.
However, some radioactivity can be associated with car-
bonate rock and sandstones because of transport of radi-
oactive minerals in solution in the formation waters.
The principal use of the gamma ray log is to distinguish
between the shales and the nonshales. Historically, the
first gamma ray devices measured only the total gamma
0 0.5 1.5 2 2.5 3
Gamma Ray Energy, MeV
Fig. 50.18-Theoretical gamma ray emission spectra from the
three naturally occurring radioactive products.
ray flux emanating from the formation. However, it is
now known that different types of shale have different total
gamma ray activity because of the Th, U, and K concen-
trations. Fig. 50.18 shows the various gamma ray line
emissions associated with each. This indicates that by de-
termining the intensity of the particular gamma ray ener-
gies it is possible to identify the quantity of each
radioactive emitter in the formation. With the develop-
ment of improved spectroscopic-quality gamma ray de-
tectors, it became natural to refine the gamma ray
measurement into a measurement of the actual concen-
trations of the three components.
The measurement element for recent gamma ray or
spectral gamma ray logging devices is the NaI detector.
The gamma ray devices measure the total number of gam-
ma rays above some practical lower limit (on the order
of 100 keV). This total counting rate will be (I) a func-
tion of the distribution and quantity of radioactive mate-
rial in the formation and (2) influenced by the size and
efficiency of the detector used. For this reason some
calibration standards have been established by the API,
and all total-intensity gamma ray logs are recorded in API
The definition of the API unit of radioactivity comes
from the artificially radioactive formation constructed at
the U. of Houston facility. A formation containing ap-
proximately 4% K, 24 ppm Th, and 12 ppm U was con-
structed and defined to be 200 API units. The details of
this calibration facility can be found in Ref. 6.
Spectral gamma ray devices basically use the same type
of detection system as the total gamma ray devices, but
instead of one broad energy region for detection. the gam-
ma rays are analyzed into several different energy bins.
n emitters/cm3 This simply says that the total counting rate is propor-
tional to nlpb, which can be expressed as the weight per-
cent of the material that is radioactive. Consequently, the
utility of expressing the radioactive contents as weight
fractions is seen.
One of the fundamental difficulties in the interpretation
of the gamma ray device measurements is inherent in its
very concept. There are nonradioactive clays and there
are hot dolomites. The use of spectral gamma ray
devices can often point out an anomaly such as a hot
dolomite or other formation with some unusual excess of
U, or in other cases K or Th.
To Emitters in Shell
Both types of devices suffer to some small degree from
the borehole environment. Because of mud in the bore-
hole and varying hole diameters, the gamma rays emit-
ted from the formation must pass through different
amounts of gamma ray absorbers to reach the detector.
Additional complications can arise because of mud addi-
tives such as barite or KCl. In the first case, the barium
Emitters in Shell
content of the mud becomes a very efficient absorber of
low-energy gamma rays emanating from the formation.
In the second case, the borehole fluid is also a source of
Fig. 50.19-Geometry for the gamma ray flux al a point from uni-
radioactive potassium, which is contained in the KC1 ad-
formly distributed sources in an infinite medium. ditive. Ref. 7 discusses a method for correcting for these
This allows for the determination (after comparing to nor-
malized standard formations where the concentrations of
K, U, and Th are known) of the concentrations of these
elements present in the measured formation. These log
output quantities usually are expressed as a percent by
weight of the total material.
It is of interest to note that the gamma ray intensity from
a uniformly distributed source (whose concentration is
maintained at a constant value when expressed in weight
percent of the medium in which it is embedded) is in-
dependent of the formation density, even though the at-
tenuation is a direct function of the formation density. This
can be seen from the following argument.
Consider an infinite homogeneous medium containing
n gamma ray emitters per cubic centimeter, each with a
source strength of emission of one gamma ray per sec-
ond. To calculate the total gamma ray flux that would be
seen by a detector at a given point in this medium, refer
to Fig. 50.19. The contribution to the total counting rate
from a spherical shell of thickness dr at a distance r from
the detector would be the number of emitters contained
in this shell multiplied by the attenuation over the path
length r to the detector,
. . . . . . . . . . . . . . . .
Gamma-Gamma Density Devices
As noted in an earlier section, the transmission of gam-
ma rays through matter can be related to the electron den-
sity if the predominant interaction is Compton scattering.
Thus, a gamma ray transmission-type measurement
through a formation can be used to determine its density
and with some information on the material composition
(lithology and pore fluids) the porosity can be determined.
The gamma ray source usually used in density devices
is 37Cs, which emits gamma rays at 662 keV, well be-
low the limit for pair production. This isotope has a half
life of about 30 years, which provides a usable, stable
intensity during a reasonable period. Some devices use
6oCo, which emits two gamma rays at 1332 and 1173
The earliest devices consisted of the gamma ray source
and a single detector, which initially was called a Geiger-
Miiller tube. However, to compensate for the frequent oc-
currence of intervening mudcake, modern devices incor-
porate two detectors (generally both NaI) in a housing that
shields them from direct radiation from the source and
is forced up against the formation with a hydraulically
operated arm. This arm provides a force of application
as well as a measurement of the diameter (along one axis)
of the borehole.
The measurement principle derives from the fact that
the counting rate of a detector will vary exponentially with
the density of the formation. Consequently, the forma-
and the total counting rate is just the integral
tion density can be determined simply from an observed
counting rate. However, in the case of intervening mud-
cake of unknown density and thickness, there will be a
-eeKaPbrdr, . . . . . . . . . (34)
disturbance of the counting rate. Fig. 50.20 shows the
usual logging situation. To correct for this intervening
mudcake, the apparent density of the long- and short-
spacing devices can be derived. Laboratory measurements
then are used to define the correction, Ap, that must be
*=n- KoPb. . . . . . . . . . . . . . . . . . . . . . . . . . . . (35) applied to the apparent density from the long-spacing de-
tector to read the value of the formation density behind it.
In at least one device the shape of the gamma ray spec-
trum is measured and correlated with the photoelectric
absorption parameters of the formation, which, in turn,
can be linked with the lithology of the formation. The
photoelectric factor, Fpe, is proportional to the photo-
electric cross section per electron. Fig. 50.21 shows the
utility of such a measured parameter for distinguishing
between the three principal matrices.
Since the Fpe of mixtures does not combine volumetri-
cally, a new parameter, U, which has the property of com-
bining linearly, has been developed for interpretation
purposes. The definition of U is the product of Fpp and
electron density,
Long Spacing
U=Fpepe, . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..(36)
and corresponds to the macroscopic photoelectric cross
section. This follows from the definition of Fpe, which
is the photoelectric cross section per electron, and pp,
which is proportional to the number of electrons per cu-
bic centimeter. Thus, the value of U for any mixture can
be computed by making a simple volumetric addition of
the Uassociated with each of the pure components of the
mixture. Table 50.3 lists some useful lithology parame-
ters for a number of commonly encountered minerals. Fig.
50.22 shows the use of U and density in the determina-
tion of lithology once the effect of porosity has been elim-
Short Spacing
Examination of Table 50.3 shows the enormous sensi-
tivity of the parameter U or Fpe to elements with a large
atomic number. In particular, note the values of Fpe for
the several iron compounds and for barium. In the case
of iron, this sensitivity can be exploited to make a deter-
mination of shale content of the formation if there is iron
associated with the clay mineral. This is discussed in the
section on interpretation. However, the sensitivity to barite
makes the Fpe measurement difficult in heavily weight-
ed barite muds. If there is a substantial thickness of barite
mudcake between the tool skid and the formation or if
there is invasion of BaS04 particles into the formation,
the resultant photoelectric absorption can seriously dis-
turb the measurement.
Fig. 50.20--Schematic of a compensated density device in a
borehole with mudcake.
Neutron Porosity Devices
Historically, the neutron device was the first nuclear
device to be used to obtain an estimate of formation
porosity. The principle of operation is based on the fact
that hydrogen, with its relatively large scattering cross
section and small mass, is very efficient in the slowing
of fast neutrons. Thus, a measurement of the flux of
epithermal neutrons resulting from the interaction of high-
energy source neutrons with a formation will be related
to its hydrogen content. If the hydrogen (in the form of
water or hydrocarbons) is contained within the pore space,
then the measurement will yield porosity. The simplest
version of the device consists of a source of fast neutrons
such as Pu-Be or Am-Be with average source energies
of several MeV and a detector of much lower-energy neu-
trons at some distance from the source. Two general
categories will be considered on the basis of the types of
neutrons detected-epithermal or thermal.
Fig. 50.21-Values of the photoelectric factor, Fpe, [or the three
principal matrices showing the relative Insensitivity
to porosity.
To be a little more quantitative about the response of
neutron porosity devices, we can use the results of two-
group diffusion theory, * which show that the flux of
50-l a
Salt water
CaSO s
CaCO 3
CaCO, .MgCO,
FeS 2
FeS 2
Clean sandstone
Dirty sandstone
Average shale
1 0.00025
6 0.15898
8 0.44784
11 1.4093
12 1.9277
13 2.5715
14 3.3579
16 5.4304
17 6.7549
19 10.081
20 12.126
22 17.089
26 31.181
38 122.24
40 147.03
56 493.72
2.700 2.602
2.070 2.066
136.146 5.055 2.960 2.957 14.95
233.366 266.8 4.500 4.011 1070.0
100.09 5.084 2.710 2.708 13.77
277.88 4.089 1.61 1.645 6.73
183.696 55.13 3.960 3.708 204.0
101.90 1.552 3.970 3.894 6.04
I 84.42 3.142 2.870 2.864 9.00
172.18 3.420 2.320 2.372 8.11
58.45 4.65 2.165 2.074 8.65
159.70 21.48 5.210 4.987 107.0
151.75 16.63 4.70 4.46 74.2
84.33 0.829 3.037 3.025 2.51
231.55 22.08 5.180 4.922 109.0
i 19.98 16.97 4.870 4.708 79.9
I 19.98 16.97 5.000 4.834 82.0
60.09 1.806 2.654 2.650 4.79
79.90 10.08 4.260 4.052 40.8
74.557 8.510 1.984 1.916 16.3
183.31 69.10 4.560 4.279 296.0
18.016 0 358 1.000 1.110 0.40
0.807 1.086 I.185 0.96
0.119 0.850' 0.948 0.11
0.125 0.850" 0.970 0.12
2.308 2.330
2.394 2.414
2.650' 2.645'
1.700' 1.749
1.400 1.468
5 6 7
I I i
8 9 10
Fig. 50.22-Matrix identification chart from U and density once the effect of porosity has been eliminated
epithermal neutrons, in an infinite medium containing a
point source of fast neutrons, falls off exponentially with
the distance from the source, L, with a characteristic
length, L,s, which is determined by the constituents of the
where D is the epithermal diffusion coefficient, which is
related to the transport mean free path of neutrons. At
a fixed spacing, counting rates should vary nearly ex-
ponentially with the slowing-down length of the forma-
tion. An indication of this type of behavior can be seen
in Fig. 50.23, which shows, on the left, the counting rate
~ ,=L eeLILs
of one of the early epithermal neutron devices as a func-
cp, D , . .
L tion of porosity, and on the right as a function of slowing-
down length. The matrix effect is much reduced in the
Sl owmg- Down Length. Ls. cm
Fig. 50.23-Counting rates for a long-spacing epiihermal detector in test formations of varrous porositres.
On the left, the data are plotted as a function of the formation porosity with th:ee data trends
resulting that car respond to the three matrices of the test formattons. On the right, the same
data have been replotted as a function of the corresponding slowing-down length
for each of
the test formattons.
50- 20
' 6
. E
a 4
k 2
6 '
. o
z 0
LT 0 10 20 30 40 50 60 70 90 100
Porosity, %
Fig. 50.24-The ratio of near to far detector counting rates as
a function of porosity for a thermal neutron
ty device.
- A-
90 l oo
Porosity, p.u,
Fig. 50.25-The calculated migration length, L, , as a function
of porosity for the three principal matrices: sand-
stone, limestone, and dolomite.
second presentation. It also shows how the counting rate
in any other material, once its slowing-down length has
been calculated, then can be estimated or, conversely, how
the slowing-down length of the formation can be deter-
mined from a measurement of the epithermal flux. It was
seen in Fig. 50.11 that the slowing-down length is strongly
dependent on the amount of hydrogen present in the mix-
ture for which it is calculated.
Fig. 50.11 also shows the slowing-down length as a
function of porosity for three common matrix materials:
limestone, dolomite, and sandstone. From this presenta-
tion, it can be seen that if the matrix is known, the appro-
priate true porosity can be determined. As an operational
expedient it has been convenient to convert the epither-
mal counting rate into porosity directly, assuming a lime-
stone matrix with a slight correction to be made for the
other two matrices. The separation of the three curves
in the previous figure suggests how such a correction is
One of the first really quantitative devices of this type
used a single epithermal detector in a skid applied mechan-
ically against the borehole wall. This sidewall epithermal
neutron device had the advantage of minimizing borehole
effects, although it is sensitive to the actual size of the
borehole and can be disturbed by the presence of mud-
cake between the pad surface and the borehole wall.
A more recent development is the dual-detector com-
pensated neutron device. This type of device uses a pair
of thermal detectors for increased counting rate to improve
the statistical uncertainty of the derived porosity values
at high porosity. The second detector, the nearer to the
source, is used to provide compensation for borehole ef-
fects. Although thermal neutron detection is used, it can
be shown8 that if the source-to-detector spacings are ap-
propriately chosen, the ratio of the two counting rates
should vary exponentially as the inverse of the slowing-
down length just as in the case of the single epithermal
detector. In practice, however, it is found that some ad-
ditional corrections must be made to the measurements,
which can deviate from expected values if the thermal cap-
ture properties of the borehole and formation are signifi-
cantly different. These generally are provided by the
service companies in the forms of charts or nomographs.
More recently they have been provided as a part of com-
puterized interpretation.
The migration length, discussed in an earlier section,
provides a convenient way to characterize the response
of the thermal neutron device. Fig. 50.24, taken from Ed-
mundson, shows the ratio of the near to the far count-
ing rate of such a device for three types of lithologies as
a function of porosity. If the porosity values on the points
of this plot are converted through the use of Fig. 50.25,
which shows the migration length, L, , as a function of
porosity, then the counting rates for the three lithologies
lie on a single ljne, as seen in Fig. 50.26. This demon-
strates that the response characteristics of the neutron
porosity tools are given by some function of the slowing-
down length and diffusion length rather than porosity.
Although the API committee6 that set up the gamma
ray calibration standards also took some steps to stan-
dardize neutron log responses, their recommendations for
API units have not been implemented. The conventional
approach to neutron log output is to calibrate the tool in
limestone primary formations and to report all readings
in apparent limestone porosity. Conversion charts are then
necessary to correct the apparent limestone porosity for
the matrix in which the measurement actually was made.
(As noted previously, some consideration should be given
to using the slowing-down length and diffusion length as
the units for reporting the log measurements.) These meas-
urements would be converted to porosity by use of charts
similar to Figs. 50.11 and 50.25, lying entirely in the
realm of interpretation.
q Sandslone (Quartz)
0 Limestone
4 Dolomite
@ Water (100%)
One of the biggest limitations of the thermal porosity
device is the disturbance on the measurement that can be
caused by shale, either from its iron or potassium con-
tent or associated trace elements with high thermal cap-
ture cross section. However, even without the additional
disturbance of thermal absorbers, clays and shales pre-
sent a problem for all neutron porosity interpretation be-
cause of the hydroxyls associated with the clay mineral
structure. Fig. 50.27 illustrates this point by showing the
variation of slowing-down length of a sand/illite and
sand/kaolinite mixture as a function of porosity. In both
cases the sand and shale volumes are in equal proportions.
It is clear that if the presence of clay in addition to the
sand is not taken into account, large errors in porosity
can result. Also note that the apparent porosity of kaolinite
is much larger than that of illite. The examples shown
in the figures indicate that a 20-PU sand/illite mixture will
appear to be about 2.5 PU, whereas the 20-PU
sandjkaolinite mixture will have an apparent porosity of
about 36 PU. This will be seen in a later section to be
caused by the differing hydroxyl content of these two clay
l -
01 I
7. 5 10 15 23 25 30 35
Migration Length, Lm, cm
Fig. 50.26-The ratio data of Fig. 50.24 plotted as a function of
the migration length, L, , corresponding to the
matrix and porosity of the test formation meas-
32. 1
27. 5
It is also of note that the most recent neutron porosity
device consists of a pair of thermal and a pair of epither-
mal detectors. This enables measuring an apparent porosi-
ty unaffected by thermal absorbers and simultaneously
obtaining a measurement of the macroscopic thermal ab-
sorption coefficient, C, which additionally describes the
Pulsed Neutron Logging Devices
Pulsed neutron logging devices respond to the macroscop-
ic thermal absorption capture cross section. The macro-
scopic thermal absorption cross section depends on the
chemical constituents of the matrix and pore fluids. Chlo-
rine, which is nearly always a constituent of formation
waters, has a large absorption cross section. Thus, a meas-
urement of the absorption cross section can provide the
means of identifying salt water and measuring formation
fluid saturation.
12. 5
10. 0
7. 5
5. 0
0 5 10 15 20 25 30 35 40
Porosity, p.u.
To determine the macroscopic thermal cross section,
the actual phenomenon being measured is the lifetime of
thermal neutrons in an absorptive medium. In a manner
analogous to radioactive decay, we can predict the time-
dependent behavior of thermal neutrons. The reaction rate
for thermal neutron absorption is given by the product
of the macroscopic cross section C and the velocity of the
neutron, v. So for a system of NN neutrons the rate of
thermal absorption is given by
Fig. 50.27-The calculated slowing-down length as a function
of porosity for sand and sand/clay mixtures. For the
two lower lines the matrix is composed of equal mix-
tures, by volume, of sand and kaolinite or illite.
dNN= -Cvdt, . . . . . . . . . . . . . . . . . . . (38)
which when integrated yields
NN=NiC, . . . . . . . . . . . . , . . . . . . . . . (39)
t n
Fast Neutron
Burst from
Pulsed Source
Build-up & Decay
of Thermal Neutrons
0 100 200 300
Time, ks
Fig. 50.28--Schematic timing diagram of a pulsed neutron cap-
ture/gamma ray device.
which relates the number present at time t to an initial
number Ni at time zero. The exponential decay constant
is seen to depend inversely on the desired quantity C.
The practical realization of such a pulsed neutron log-
ging device depends on a pulsed source of high-energy
neutrons. Such a device was discussed previously. The
basic mode of operation consists of pulsing the source of
14-MeV neutrons for a brief period. This forms a cloud
of high-energy neutrons in the borehole and formation,
which then becomes thermalized through multiple colli-
sions. This process is illustrated in Fig. 50.28. Only at
thermal energies does the absorption become important
and the neutrons begin to disappear in accordance with
Eq. 39. As each neutron is captured, whether it be by
hydrogen or chlorine, gamma rays are emitted, and the
decay of the gamma ray counting rate is the actual meas-
urement that reflects the decay of the neutron population.
As seen from Eq. 39, the decay constant for a particu-
lar formation is given by 1lv.L The value of the capture
cross section C is listed in Table 50.4 for a number of
pertinent cases. Included in the table is the decay time
associated with the particular matrix, which was computed
from the relationship
?d=- c , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
where K is 4550 psec, because v for thermal neutrons is
0.22 cmlpsec and Cabs, the thermal absorption cross sec-
tion, is in capture units.
- -
Quartz 4.26 1066
Dolomite 4.7 966
Lime 7.07 643
20-PU lime 10.06 452
Waler 22 206
Salt water (26% NaCI) 125 36
Since the derivation of the decay-time measurement is
based on the simple model of a cloud of thermal neutrons
being present and then decaying, it is of some interest to
see just how long it is after the burst of 14-MeV neutrons
that they become thermalized. To estimate this time we
need only to refer to the section on neutron physics, where
the average number of collisions for thermalization and
the mean free path were discussed. The simplest estimate
of the time required is to suppose that between each col-
lision the average distance traveled is the mean free path
(l/C,). The time between one collision and the next, At,
is then given approximately by
c, v, . . . . . . . . . . . . ..I............ ..(41)
where C, is the total cross section and v is given in terms
of the energy E by Eq. 23. The 1 Iv factor can be replaced
by using Eq. 23 in conjunction with the expression (Eq.
30) for the average number of collisions, a, yielding
-oQ~'*, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..(42)
where E is the average logarithmic energy decrement de-
fined in Eq. 25.
The value of an average mean free path for formations
of interest can be estimated from Fig. 50.10. From the
above information, the total time, t, (psec), from emis-
sion to thermal energies is given by
Evaluation of this expression for 20%-porosity limestone
gives an estimate of about 2.8 psec and in water it is only
0.5 psec, both of which are much smaller than the decay
times shown in Table 50.4.
Numerous measurement schemes are used for control-
ling the period during which the 14-MeV neutrons are
produced and the period during which the gamma rays
are measured. Some devices use dual-detector systems in
an attempt to correct for the small disturbance that can
be introduced by the borehole size and salinity as well
as to provide some measure of the porosity.
Inelastic and Capture Gamma Ray Spectrometry
The primary motivation for the development of induced
gamma ray spectroscopy devices was the possibility of
performing in-situ chemical analysis of the formation con-
stituents. The tantalizing possibility of directly measuring
the ratio of the number of carbon atoms to oxygen atoms
and thus providing the first direct downhole measurement
of the presence of hydrocarbons spurred the development
of a number of technologically sophisticated devices. This
is to be contrasted with a traditional approach that depends
on the analysis of a core or side-wall sample.
Tools of this type are based on a type of chemical anal-
ysis that can be performed through the use of neutrons
and gamma ray spectroscopy. Neutrons are used to ex-
cite the nuclei, which then emit gamma rays of precise
0 2 4 6 8 10 12 14
Neutron Energy, E,, MeV
Fig. 50.29-Cross sections for the production of inelastic gam-
ma rays by carbon and oxygen as a function of the
incident neutron energy.
Ge Detector
Gamma Ray Energy. MeV
Fig. 50.30-Comparison spectra from a high-resoluhon solid-
state detector (Ge) compared to the gamma ray
spectrum detected with a more conventional Nal de-
energies uniquely identifying the isotope in question.
There are two neutron reactions that can produce such
gamma ray emissions: inelastic scattering, which can
occur with very-high-energy neutrons, and capture reac-
tions with thermalized neutrons as exploited in pulsed neu-
tron logging.
Few elements for well-logging applications have large
inelastic cross sections, but fortunately carbon and oxy-
gen do. Fig. 50.29 shows the cross sections for the pro-
duction of gamma rays from inelastic scattering from
carbon and oxygen. These inelastic induced gamma rays
are observed not only by spectroscopic gamma ray de-
tection but also in conjunction with timing. To avoid con-
fusion with gamma rays produced from thermal capture,
the inelastic gamma rays are detected during the burst of
14-MeV neutrons. At some later time, gamma rays aris-
ing from thermal absorption are detected, providing sen-
sitivity to a large number of elements such as H, Fe, Cl,
Si, Ca, S, etc. At least two different neutron pulsing and
gamma ray detection sequences are currently in use. One
method lo uses a fixed time of about 50 to 100 psec be-
tween neutron bursts, and another method, which col-
lects information on the capture gamma rays, uses a
variable neutron-pulse interval that is controlled by the
characteristic decay time of the thermalized neutrons.
Tool design differences can optimize the detection of
the inelastic or capture gamma rays. Some designs incor-
porate the measurement of both through appropriate tim-
ing cycles. In addition to the measurement of the gamma
ray yields of the various elements, the macroscopic cross
section C can be determined from an analysis of the de-
cay of the total gamma ray signal, which is also meas-
ured. The only practical limitation on the number of
elements is determined by counting statistics and the in-
herent detector resolution.
An experimental tool that uses a high-resolution Ge de-
tector has measured more than two dozen different ele-
ments in borehole logging. This type of detector must be
operated at very low temperatures (- I96C), which in-
troduces a number of technological problems for bore-
hole measurements. However. the advantage that it brings
is in the much improved resolution, which increases the
number of distinct gamma rays that can be distinguished
in the spectrum. Fig. 50.30 shows the dramatic improve-
ment compared to a conventional NaI detector for the ex-
amination of a natural uranium sample. Two elements that
are readily detectable with such a device are aluminum,
which is very useful in the classification and quantifica-
tion of in-situ clays, and vanadium, which can be
correlated I2 with the API gravity of the associated oil.
Interpretation of Nuclear Logs
The following section discusses how the previously men-
tioned nuclear logging tools are used in interpretation.
However, it should be stated at the outset that this is not
intended to be a self-contained log interpretation course.
A number of references- of such works should be
consulted for further information.
The approach taken here is more or less a stand-alone
interpretation of each tool. An examination of all the com-
bination measurements and the interpretation techniques
used with each tool is beyond the scope of this chapter.
Nonetheless, several of the more standard tool combina-
tion interpretation approaches are discussed. Fig. 50.31
50- 24
Interpretation Steps Nuclear Measurements Used
: I Cor;el~iions of : jj . I : ~:.: i : : : : jGamma Ray j j I
. ,.
; : : :
/ Clean Zones :: . 1 j
,. : .:
: i Natural Th, U.K. Spectrosc&
. .
. G&n&Density : :
Neutron Pordsity : : :
Lithology &
-Photoelectric F&&r ip,betisity/Neutron
Clay Typing
Induced Gamma Spectroscopy:
High ResolutionGamma Ray Spectroscopy
Pulsed Neutron
C/O~(Gamma Spectroscopy)
Fig. 50.31-four interpretation tasks and the nuclear measurements associated with them.
indicates four of the steps in the interpretation process
and the nuclear measurements used to obtain the desired
results. The indicated list of measurement devices or tech-
niques is in the order of the discussion that follows.
Interpretation of Gamma Ray Measurements
The gamma ray log traditionally has been used for cor-
relating zones from well to well, crude lithology identifi-
cation, and rough volume of shale estimation. With the
current state of knowledge of clay composition and other
more refined lithology determinations, it is clear that the
surest use of the gamma ray is, indeed, for correlation.
For estimating the volume of shale, the approach is to
scan the log for minimum and maximum gamma ray read-
ings, Ymin and ymax.
The minimum reading then is
assumed to be the clean point and the maximum reading
is taken as the shale point. Then the gamma ray reading
in API units at any other point in the well, ylog , is scaled
sh a
Ylog -Ymin
. . . . _. . . . . . . . . . . . . . . . . . .
Ymax -Ymin
This ratio usually is referred to as the gamma ray index
and can be scaled into percent shaliness according to
charts I3 depending o n rock type. This method is some-
times appropriate, if in fact the maximum gamma ray
reading corresponds to the same type of shale as the values
that are being interpreted.
Numerous examples show the deficiencies of this
method, and for this reason the spectral gamma ray tool
was developed. Tools of this type measure, in fact, the
relative concentration of three radioactive components of
the total gamma ray signal. It is interesting to point out
the relationship between the concentration of the three
radioactive components and the total gamma ray signal
in API units. It is given by
YAPt=AxTh+BxU+CxK. . . (45)
When thorium and uranium are measured in ppm and
potassium in weight percent, it is found that ratios of the
coefficients (A:B:C) are 1:2:4; i.e., 1 wt % K contributes
four times more to YAPI than 1 ppm of thorium. It is ob-
vious from this that in a shaly sand, if a mineral rich in
potassium (such as mica) is present, the total gamma ray
signal will increase and give a false indication of pcrcen-
tage shale when, in fact, this additional radioactivity is
caused by the mica.
There are two solid reasons for using a spectral gam-
ma ray measurement over the standard gamma ray, which
is really reliable only for correlation. The first is for the
detection of radioactive anomalies, such as referred to
previously, and the second is to make some estimation
of the clay types by classifying them in terms of the rela-
tive contributions of the three radioactive components. For
this second point the reader is referred to publications on
spectral gamma ray interpretation. I*-*
Fig. 50.32 shows a log example from the North Sea
in a micaceous sand. At 10,612 to 10,620 ft, a shale is
indicated that has a total gamma ray signal of about 90
API units. With just the total gamma ray as an indicator
it appears that the zone 10,568 to 10,522 ft contains about
half the amount of shale estimated for the lower zone.
However, the decomposition of the gamma ray signal
shows quite clearly that the amounts of U, Th, and K in
these two zones are quite different. In fact, the upper zone
is a mixture of sand and mica, whereas the lower zone
is, indeed, shale.
In the next example, Fig. 50.33, the gamma ray alone
would indicate that below the lower boundary of the shale
bed at 12,836 ft, there is a relatively clean sand. It can
be seen, however, from the K trace that the high level
of potassium in the shale zone persists several feet below
12,836. This excess potassium was found to be caused
by feldspar, which has considerable impact on the grain
density to be used in the interpretation of density logs.
The third example, Fig. 50.34, shows how a uranium-
rich formation would be misinterpreted (in simple gam-
ma ray interpretation) as being shale. The sudden increase
Gamma Ray
Depths Thorium Uranium Potassium
50 4J 0
0 API 200 0 ppm200 wmloo % 5 50 %o 0
Fig. 50.32-A spectral gamma ray log from the North Sea in a micaceous sand.
Gamma Ray Depths Thorium Uranium Potassium
Porosity and
Fluids Analysis
bv Volume
Fig. 50.33-A spectral gamma ray log from Nigeria where a continuing trace of K is attributed to the presence of feldspar.
in uranium content alone signals that this is not a simple
shale of the variety in the adjacent levels. Core analysis
showed this zone, however, to be rich in organic materi-
al and U is often trapped in organic complexes.
Porosity Determination
Gamma-Gamma Density Devices. The basic output of
the gamma-gamma density device, bulk density, is con-
ceptually the simplest measured parameter to interpret in
terms of porosity. The basic relation (Eq. 1) is
Pb =d&f+(l -dhma,
which volumetrically links the density of the pore fluid,
pf, and the rock matrix density, pmo, to the bulk densi-
ty, Pb.
However, there are a few difficulties to overcome to
interpret the output of such a density device, especially
the matter of electron density index, oe. It was shown
in an earlier discussion that the gamma-gamma density
device is measuring the electron density index. Table 50.3
shows a comparison between bulk density and electron
density index. It is in close correspondence for nearly all
the compounds listed except for water and hydrocarbons.
This is because the average value of Z/A is about % for
all elements except hydrogen. It is seen that because of
Tppm Uwm
0 GRAPl 150 0 20, 0 10,o 5 0 100 %
B i i i i i i / i 4
Fig. 50.34-A spectral gamma ray log showing a zone of anomalous U concentration
this discrepancy for hydrogen, the bulk density and elec-
tron density index of water differ by about 11%
To compensate for this fact, a simple transform of the
electron density index is made so that in water-filled lime-
stone the transformed or log density, p log. agrees with
the bulk density. Fig. 50.35 shows this simple transform
where the bulk density for 0-PU limestone and water are
plotted against the electron density values. The equation
of the straight line connecting these two points,
pb=1.07~p,-o.188, . . . . . . . . . (46)
corresponds to the published transform I4 used by the log-
ging companies. It is worthwhile to point out that this
transformed density, p l0g, will also agree to within 0.004
g/cm3 for the other principal matrix materials listed in
the table.
Returning to the interpretation problem, the solution of
Eq. 1 for porosity yields
Pm -Pb
p*-Pf .
. . . . . . . . . . . . . . . . . . . . .
so the problem rests on knowing the values to insert for
fluid and matrix density. Before examining the means for
determining these values, it is perhaps of some interest
to know to what precision these two parameters must be
It is interesting to look at the uncertainty that can be
tolerated on the value used for the matrix density, pm0 .
From the previous equation we can write
Pb -Pmo
dC$= -~
Pf -Pm2
1 .o 1.5 2.0 2.5 3
Electron Density, per g/cc
Fig. 50.35-The transform between the measured electron den-
sity index, pe, and bulk density, pb.
If this is evaluated for the case of a sand of about 30%
porosity we can use the values pb =2.16, pf= 1 .OO, and
pmn =2.65 and obtain
d$=0.43dp,. . . . . . . . . . . . . (49)
Thus, for the uncertainty in 4 to be less than 0.02, the
uncertainty in p ma must be less than 0.05 g/cm3. A more
detailed analysis of the uncertainty in grain density that
can be tolerated can be found in Ref. 2 1.
For values of fluid density it is necessary to know the
type of fluid in the pores. The fluid density for hydrocar-
bon may range from 0.2 to 0.8. Salt-saturated water
(NaCI) may be as high as 1.2 g/cm 3, and with the pres-
ence of CaCl2, values even as great as 1.4 g/cm 3 may
occur. However, the uncertainty that can be tolerated in
pf is much greater than that for p ma. An error analysis
of Eq. 41 for the values chosen above shows that
&#J =O. 188pf, which allows about double the margin of
The value for matrix density in simple cases can be
taken from Table 50.3, which shows a rather narrow var-
iation between 2.65 g/cm3 for quartz and 2.96 g/cm3 for
anhydrite. Grain densities for shales are an entirely
separate matter and will not be covered here. The obvi-
ous problem left is assigning a matrix density, which can
only be done with a knowledge of the lithology. In es-
sence, all density interpretation aimed at porosity deter-
mination revolves on this.
Before going on to the determination of lithology and
thus an estimate of grain density, quality control of the
density measurement should be discussed.
As mentioned earlier, the modern gamma-gamma den-
sity devices are compensated measuring devices. They use
two detectors at different spacings from the gamma ray
source to compensate for the possible intervening pres-
ence of mudcake or drilling fluid. Normally. in addition
to the density curve the log will also show a trace of the
compensation, generally referred to as the Ap curve. This
curve represents the correction made to the apparent den-
sity seen by the long-spacing detector (P,,~) based on the
discrepancy between the long- and short-spacing meas-
The counting rate from either detector can be convert-
ed to an ap arent density after a series of laboratory
* If there is not any intervening material be-
tween the tool surface and the formation being measured,
then the two values will be equal. As the mudcake thick-
ness increases, the two density values will diverge for
some reasonable value of thickness (generally less than
1 in.). The quantity Ap is determined by experimentation,
as a function of the density differences, to be the amount
to be added to the apparent long-spacing density, pls, to
match the bulk density of the formation-i.e.,
p,,=pls +Ap. . . . , . . . . . . . . (50)
Although it commonly is thought that Ap is a measure
of the mudcake thickness, h,,, it is, in fact, proportion-
al to the product of mudcake thickness and the density
contrast between the mudcake, pmc, and the formation
Apcxh,,(pb -pmc). . . . . . . . . (51)
Beyond some thickness (- 1 in.), the compensation
scheme will break down and the ph value will be in
doubt. However, this point cannot be identified simply
by use of a cutoff value of Ap. A very small gap of water
(p,, = 1) in front of a low-porosity formation would yield
a large value of Ap and yet be perfectly compensated for,
whereas a 1-in.-thick mudcake of medium density in front
of a high-porosity zone may yield a small Ap with some
residual error in the compensation. Nonetheless, it is cer-
tain that the Ap curve will be used as quality control on
the bulk density with some fixed cutoff value despite this
Neutron Porosity Devices. Modem neutron porosity
devices are of two types, depending on the energy range
of detection: thermal or epithermal. By convention. the
log output usually is scaled in equivalent limestone porosi-
ty units. Under appropriate circumstances, this would cor-
respond to the true porosity of a clean, water-filled
limestone formation. In the following discussion it is
assumed that the log reading already has been corrected
for the various environmental effects. In addition to the
environmental corrections that must be made before the
porosity values are interpreted, there are three effects that
must be considered in more detail.
Matrix Efsect. As in density logging, it is necessary
to know the rock matrix to make any practical use of the
app:lent limestone porosity value measured. Fig. 50.36
shows, for a thermal and an epithermal device, the matrix
correction necessary to transform the measured units into
the appropriate porosity units. It should be noted that these
- - CNL.
Neutron Porosi ty I ndex I Li mestone~. p. u
- - r&l CNLl cor Neutron Porosi ty I ndex ( Li mestonel . p. u
Fig. 50.36-Matrix correction chart for two specific neutron
porosity tools.
two charts are for two very specific tools (CNL* and
SNP*). When such tools are involved, the appropriate
chart for the corresponding tool should be used, since
some of the so-called matrix effect is tool-design depend-
ent. However, a large part of the matrix effect can be un-
derstood in terms of the two basic parameters used to
describe the bulk parameters of the formation (i.e., the
slowing-down length and the migration length).
First consider the case of epithermal detection. To dem-
onstrate the construction of the matrix-effect correction
curves, there are four steps to consider.
1. The first step is the link between the measurement
(for this case we take the simple case of using the ratio
of near- to far-detector counting rate) and porosity in the
primary laboratory calibration standards. This is a
freshwater-filled limestone with g-in. borehole. Fig. 50.37
shows the behavior of the ratio, F=NN,, lNNe as a func-
tion of limestone porosity, +rS. From this plot a fit can
be established to ascribe a functiqnal relationship between
the measured parameter F and the limestone porosity @jS,
F=J@,,). . . . . . . . . . . . . . . . . . . . . . . (52)
2. In the second step, the relationship between the meas-
ured parameter F and the slowing-down length, L, , must
be established for measurement in all three of the prin-
cipal matrices. That this can be done easily is shown in
Fig. 50.38, where measurements in quartz, dolomite, and
limestone are shown for a range of porosities. From this
plot a new fit can be found,
F=f(L,), . . . . . . . . . . . . . . . . . . . (53)
Mark of Schl umberger
0 10 20 30 40
Porosity, p.u.
Fig. 50.37-Experimental values for the ratio of epithermal
counting rates as a function of porosity in limestone
calibration formations.
that allows the prediction of the measured ratio from the
known slowing-down length of the formation.
3. The next step is to establish the connection between
the slowing-down length, L, , and porosity for limestone
as shown in Fig. SO.39 and seen earlier in Fig. 50.11.
This curve now represents the limestone transform of
Eq. 52 and the porosity axis represents true porosity.
4. The slowing-down lengths of sandstone and dolo-
mite now are calculated as a function of porosity and are
shown also in Fig. 50.39. They fall on either side of the
limestone response because of their different chemical
compositions, which influence their slowing-down
Fig. 50.38-Data of Fig. 50.37 plotted as a function of the equiva-
lent slowing-down length, L, , of the corresponding
25. 0
The apparent limestone porosity for either formation
can be found by selecting a porosity, lo-PU sandstone
for example, and finding the corresponding slowing-down
length (approximately 15.5 cm). The apparent limestone
porosity then is obtained by finding the porosity associated
with the limestone formation of the same 15, value. In this
case it is 8.5 PU. The same case for dolomite indicates
14 PU instead of 10 PU.
A similar procedure can be used for a thermal neutron
porosity device. As shown earlier, it has been found use-
ful to cast the results in terms of the migration length
L m. 9 Because the L, contains some information con-
cerning the macroscopic thermal absorption cross section,
the results are qualitatively similar to the preceding but
differ slightly in magnitude. This can be seen by perform-
ing the preceding exercise on the plots of L, vs. porosity
of Fig. 50.25.
0 5 10 15 20 25 30 35 40
Fig. 50.40 shows the neutron/density crossplot for a
dual-detector thermal device. The matrix effect can be ob-
served by comparing the equiporosity points on the in-
dividual lithology lines with the apparent limestone
porosity scale on the abscissa. These discrepancies are
the same as indicated in the portion of Fig. 50.36 indi-
cated as thermal. Ref. 23 discusses the effect of ab-
sorbers on the thermal tool response.
Gas Effect. Neutron porosity devices have been
calibrated for liquid-filled porosity. However, the replace-
ment of the liquid in the pores by gas will have a con-
siderable impact on the slowing-down length of the
formation and thus on the apparent porosity. In general
terms, partial replacement of the water component of the
mixture by a much lighter gas will increase the neutron
slowing-down length and thus the apparent porosity will
decrease. The actual apparent porosity decrease will be
a function primarily of the true porosity, the water satu-
ration, the gas density, and, to some extent, the litholo-
gy. In a situation such as this, replacement of fluid in the
pores by a less-dense gas will decrease the bulk density
of the formation. These two effects have been exploited
in well logging by making the density and neutron porosity
measurements in a single measurement pass. On the log
presentation these two effects cause the density and neu-
tron traces to separate, which can be recognized easily
as being caused by the presence of gas, if the invasion
is less than 6 in.
To quantify the traditional neutron-density pas separa-
tion indication and to illustrate the possibility ofcstimat-
ing gas saturation from an epithermal neutron
mcasuremcnt and a density mcasurcmcnt. conxidcr Figs.
SO.41 and 50.42. In both these figures the slowing-down
length of a sand formation has been computed between
Lcro and 40 PU at 2-PU increments. These values of
slowing-down length have been plotted as a function ol
the corresponding bulk density of the formation for the
five gas saturations indicated on the figures. In the caxc
of Fig. 50.41. the gas density, P,~% has been taken to bc
O.OOI g/cm. and in Fig. S0.47. the gas dcnslty has been
taken to be about 0.25 g/cm3. which covers the cntirc
range possible under normal rcscrvoir conditions. In cithcr
cam it is clear that. for the presence 01 gas at a fixed
poroGty. the slowing-down length is lqcr than th;lt ah-
sociated with the water-tilled porosity. The mterpretation
of this larger value of L, ih an apparent decrease in
porosity. In the caxc of the total ,:a\ saturation curve of
Porosity, p.u
Fig. 50.39-Estimation of the epithermal matrix effect from
slowing-down length, L,. as a function of porosity
for sandstone and dolomite
Equtvalent Limestone Porosity, %
Fig. 50.40-A neutron/density cross plot for a thermal neutron
porosity device. The lithology of a formatlon can be
identified by plotting points representlng o, (In
limestone porosity units) and pb. Point A may rep-
resent a 22.p.u. limestone or. less likely, a mixture
of sandstone and dolomite.
Fig. 50.41-A plot of slowing-down length, L,, vs. bulk densi-
ty, pb, for sandstone formations with varying gas
saturation. Invasion of borehole fluid is assumed to
be negligible and the gas density is taken to be 0.001
Fig. 50.41, L, values greater than - 28 cm would cor-
respond to zero or negative apparent porosities on the log
On the plots of L, vs. ph. a pair of points (L,, , p,,) will
yield the saturation and porosity corresponding to the con-
ditions of gas density specified. For both figures the as-
sumption for the calculation is that the matrix is sandstone
with no shale. The gas density difference for the two plots
spans the range of expected gas densities. The saturation
referred to on the figures is with reference to a gas/water
mixture. For the lowest curve on both figures, the porosity
is entirely filled with water.
Since the slowing-down length normally is not presented
on a neutron porosity log, an example taken from the log
of Fig. 50.43 can serve as a guide. From the gas zone
clearly indicated on this figure we can use the following
values for the illustration: @N, neutron porosity (sand)=
27 PU and $0, density porosity (sand)=35 PU.
The lowermost curves on Figs. 50.41 and SO.42 serve
to determine the correspondence between +N and the
slowing-down length. L, It is indicated by the horizon-
tal line to be about I I cm for this example. The density
porosity of 35 PU can be seen to correspond to a density
of about 2.07 g/cm. as indicated by the vertical lint.
The intersection of these two points at the coordinate
(2.07. I I), shown in the figure, indicates about 25% gas
saturation in the case of a gas density of 0.25 g/cm (Fig.
50.42), and a saturation of about 12.5% if the gas dcnsi-
ty ix taken to be nearly zero. The porosity of the forma-
tion can be found by following the slope of the
equiporosity lines from the example coordinate to the low-
er liquid-saturated line and is seen to indicate a value of
about 33 PU for both casts.
With more information concerning the gas properties
of a reservoir, this interpretation can be used instead to
yield an indication of the invasion of the drilling fluid into
the zone of investigation of the neutron and density
Shale Effect. Generally. the presence of shale tends to
increase the neutron porosity values. The reason for this
lies in two effects: additional hydrogen resulting from the
Fig. 50.42-Effect of gas saturation in sandstone for a gas den-
sity of 0.25 g/cm 3.
Fig. 50.43-Log example of the neutron/density combination ex-
hibiting the characteristic crossover behavior In a
gas zone.
Fig. 50.44-Comparison of (1/L,)3 for sandstone, kaolinite, and
illite as a function of porosity. The apparent porosi-
ty of pure illite is seen to be about 53 p.u. and for
illite, 12 p.u.
Flg. 50.45-Variation of (l/L,)3 vs. porosity for sand and
sandlkaolinite mixtures. A line at 30 p.u. indicates
the effect of changing the matrix from sand to
kaolinite, and another indicates the change in
(l/L,)3 expected when the 30 p.u. is filled progres-
sively with kaolinite.
I 1
0 LM+6w
q so+sw
l LM+Fw+B
I 6 I
0 10 20 30
Ax, cu
hydroxyls in the clay minerals, and in the case of ther-
mal porosity devices, the possibility of additional ther-
mal neutron absorbers such as boron associated with the
clay minerals.
Initially, consider the case of thermal absorbers and con-
sider just two types of clay minerals, kaolinite and illite.
For these two clays (and many others) we refer to the data
compiled by Edmundson et al. 9 The chemical formula
for kaolinite is A14Si40 tc(OH)s and the formula for il-
lite is K2Si6A1sFe302e(OH)4.
The important point to note about these formulations
is the differing amount of (OH). A rough calculation in-
dicates that kaolinite has about one-third the hydrogen den-
sity of water and that illite, with a lesser amount of
hydroxyls, has about one-tenth the hydrogen density of
water. In the case of either of these pure clay miner-
als, a neutron porosity device will detect some large value
of porosity, since the slowing-down length will be influ-
enced greatly by the presence of the (OH) components.
That the slowing-down length is considerably affected
can be seen in Fig. 50.44, which shows the value of L,
of sand, illite, and kaolinite from 0 to 100 PU. The or-
dinate in this figure is (1/L,)3, which has been chosen
because it tends to vary linearly with the combination of
two materials of differing values of L c. Values for the
two pure clay minerals are indicated at 0 PU. From the
figure, the apparent porosity would be about 53 PU for
pure kaolinite and about 12 PU for illite. This is consis-
tent with the trend expected from the hydrogen density.
For the slowing-down length, the choice of the replace-
ment of three Fe atoms for three Al atoms is immaterial,
but it will have an impact on the thermal neutron absorp-
tion because of capture by the iron.
By using a plot similar to Fig. 50.44, the expected
response of an epithermal neutron porosity device to a
mixture of kaolinite and sand can be predicted. Fig. 50.45
shows an example of this. In the case of a true 30 PU
(that is, 30% of the volume is water-filled), the mixed-
matrix line indicates the change of L, with a change in
the sand/kaolinite matrix. Following the indicated line,
a matrix of 70% sand and 30% kaolinite would have an
apparent porosity of about 41% compared to the actual
value of 30 % .
On the other hand, we can take the case of 30 PU, or
more precisely the case of 70 ~01% being sand, and ask
what happens to the value of L, as the porosity is filled
with kaolinite. This process also is indicated by the
porosity-infilling line in Fig. 50.45. It is clear that the
endpoint of this line must lie on the line of the 30/70
kaolinite mixture. As indicated in this case, the minimum
apparent porosity as deduced from the slowing-down
length will be about 12 PU where, in fact, there will be
no porosity available for fluids.
The impact on thermal neutron porosity devices is
somewhat more difficult to predict and will depend in de-
tail on the tool design and consequent response. A simi-
lar type of graphical construction using the migration
length instead of the slowing-down length can be used to
see that for the normal neutron absorbers, such as iron,
in the clay minerals there is not a large effect. However,
if boron is present in any substantial quantity, this will
not be the case. To deal with this problem Arnold et al. 24
have determined experimentally a correction for the ap-
parent porosity of a dual-detector thermal neutron porosity
Fig. 50.46-Correction for a thermal neutron porosity tool as a
function of the formation, C.4
device as a function of the macroscopic absorption cross
section KL of the formation. It is shown in Fig. 50.46.
Despite the foregoing discussion. it must be recognized
that neutron porosity devices do. in fact, respond to
porosity. among other things. The dynamic range of the
measurement at low porosity is excellent because of the
sensitivity of the slowing-down length in this range. The
problems of interpretation caused by the influence of shale
are tractable when the neutron measurements arc com-
bined with other tools.
Lithology Determination
Neutron/Density Combination. One of the traditional
methods of lithology interpretation is the neutron/densi-
ty combination. The useful property of combining these
two measurements can be seen in Fig. 50.40: the three
principal lithologies form three different response lines.
The grain density. P,,~~,. increases in an almost linear
fashion from sand to dolomite. so that it is tempting to
draw lines of equal grain density for intermediate points
that do not lie on any of the three lines. This type of
presentation is shown in Fig. 50.47. With this approach,
the particular lithology mixture is not of great importance
but a fair estimation of the appropriate grain density is
obtained despite the fact that p,,],, is not a characterizing
parameter for neutron logs.
The combination of the neutron and density measure-
ments in this fashion can solve only simple binary miner-
al combinations. To interpret the results unambiguously,
the two minerals must be known. For example, one can
imagine a result. marked as Point A on Fig. 50.40, aris-
ing from a formation consisting of a sand and dolomite
mixture. In this case, with no additional knowledge. the
interpretation of the indicated point, intermediate between
dolomite and sandstone. would be limestone.
Photoelectric Factor. Discrepancies such as these can be
resolved by the use of additional information. In particu-
lar. one such nuclear measurement is the photoelectric
factor. F,,, It is convenient for purposes of interpreta-
tion to use the quantity U (Eq. 36). It has the property
of combining volumetrically for the case of several sub-
stances being present in the scattering region whose in-
dividual absorption characteristics can be computed. For
a two-component system (fluid and matrix) of porosity
C#J this can be written as
u=ufc#l+u,,,,,(1 -f$). (54)
F,,,. for this mixture then can be determined from Eq. 36,
P~=(Pc)l.~+(Pr),,,n(l -4). .(55)
The results of this type of computation are shown in the
crossplot of Fig. 50.48, which shows how F,,,. varies as
a function of density for the three principal matrices: lime-
stone. dolomite, and sandstone. The interesting thing
about this presentation is the order in which the three lines
fall; dolomite now is bracketed by limestone and sand-
stone. It is obvious that in the hypothetical case of the
sandy dolomite. given previously. the ambiguity about the
presence of limestone would be removed when inspect-
ing the corresponding value of F,,,, In this case it would
clearly indicate a sand/dolomite mixture.
Neutron Porosi ty I ndex. 4, . , , . P. U.
Fig. 50.47-Extracting apparent grain density, prna, from the
neutron/density cross plot.
One elementary use of the Fpr curve in conjunction
with density is to make a simple two-mineral model. This
is no more than formalization of the relationships seen
in the preceding figure. This is done by solving the fol-
lowing set of equations for U and p, and using the fact
that the volumes sum to unity:
u,,,,=iJ,v, +U2V2+iJf$,
plog=p,V, +prV2+pf.$. (56)
I=V, + v2 ++.
Bul k Omsi ty, pb, g/ cc
Fig. 50.46-A cross plot of the photoelectric factor vs. density.
Lines for the three principa! matrices are indicated.
The points represent the sampled log data.
Iron Content
D-l% by Weight) -10% 1 F
Fig. 50.49-A log example showing the conversion of the excess
F,, in a shaly sand to the weight fraction of iron.
In Track 1, the estimate from Fpe (Eq. 61) is com-
pared to values obtained by core analysis.
The indicated parameters for the two suspected minerals
can be inserted in this equation to solve for the appropri-
ate porosity and lithology mix.
However, there is no reason to limit this type ofanaly-
sis to two minerals. It can be expanded to many more with
the addition of supplementary measurements such as Th,
U. K. Al. Fe, and Si. The natural combination with the
gamma ray spectral information has been shown useful
for complex mineralogy in which two different three-
mineral combinations can be treated. 2s
Another interesting use of F,, can be to get an esti-
mation of shale in sand. The F,,(> values measured in
shales are related primarily to their iron content. For an
example of this application, refer to Fig. 50.48. The data
points shown on the crossplot are from some sand and
shale sequences. The clean-sand points are evident by their
placement with respect to the indicated sand line. The
shale points are those scattered to higher values of F,, .
In this case the lithology measurement is responding to
the photoelectric absorption resulting from the iron as-
sociated with the shale. In this particular case the shale
is known to be a mixture of kaolinite and illite and the
iron is associated with the illite.
To demonstrate the quantitative nature of the F,,,, mcas-
urement, consider the following argument for the deter-
mination of the weight percent of a trace photoelectric
absorber in an otherwise two-component system. If F,,,,.,
represents the photoelectric factor of the trace absorber
and the associated volume fraction is V, (which is con-
sidered to be much smaller than unity), the following ap-
proximate equation applies:
(1~~,~+CI,,,(,(l~~)+II,V,, . . . . . . . . (57)
where V., < < 1 and
U,=p,,F,,<~,. _. _. .(58)
Note that the third term in this equation represents the
product of the trace element absorption factor and its as-
sociated mass contained in I cm3:
U,V., =F,,<.\.p.,V\. _. _. _. .(59)
Thus. the weight per cubic centimeter of the absorber can
be determined from the measured parameter (i.e.. from
the product of the measured p and F,,,,) Cl from the fol-
p, v\ =
The fraction by weight of the solid fraction. f\, , is then
given by
An interesting application of this idea is shown for the
log data of Fig. 50.48. The measured F,,,, curve is shown
in Track 2 of Fig. 50.49 along with the curve of the an-
ticipated F,, based on pure sandstone. The shading rep-
resents the excess photoelectric factor, which can be
converted to percent weight of iron by use of the above
argument. The resulting curve of the iron concentration
is found in Track 1 of Fig. 50.49, along with a series of
discrete measurements of the iron concentration made on
the core samples. The good argument demonstrates the
validity of the supposition that, in this case, the F,,(, meas-
urement is responding primarily to the iron concentration
in the shale and the volume of shale present. A similar
procedure could be followed for any other suspected trace
element of sufficient concentration and photoelectric ab-
sorption factor.
Induced-Gamma-Ray Spectroscopy. The induced gam-
ma spectroscopy tools are very important for lithology
determination in complex situations. In the case of induced
capture gamma rays a large number of elements can be
identified: hydrogen, silicon. calcium. iron, sulfur, and
chlorine. Lithology identification can be made by com-
paring the yields of particular elements. For example, an-
hydrite is identified easily by the strong gamma ray yield
from sulfur and calcium associated with this mineral.
AVOGADROS NUMBER (6.023 x 10z3)
Density Oxygen Carbon
(g/cm 7
(atoms/cm3) (atoms/cm 3,
2.71 0.081 0.027
2.87 0.094 0.031
2.65 0.088
2 96 0.087
1.0 0.056
Limestone can be distinguished from sandstone by com-
paring the silicon and calcium yields. Numerous
references. show examples of these types of proce-
dures as well as interpretation schemes proposed by some
service companies.
High-resolution spectroscopic devices will provide
enormous advances in mineralogical identification. Such
devices use solid-state gamma ray detectors instead of the
conventional Nat detectors, which suffer from poor reso-
lution. One of the most promising applications of such
devices is the possibility of making aluminum activation
measurements. Because aluminum is a common consti-
tuent of clay minerals. it is well adapted to quantitative
clay content determination. As an example of the type of
information that is obtainable from a high-resolution gam-
ma ray spectroscopy device. refer to Fig. 50.50. which
shows the gamma ray spectrum obtained after irradiation
of the formation with a zsCf source. Along with the
gamma rays from the naturally occurring thorium, ura-
nium. and potassium, the artificially produced radiation
from manganese, sodium, aluminum, and vanadium is
The gamma ray intensities from the aluminum and
vanadium isotopes can be used to produce a log calibrat-
ed in weight percent of the two elements. In the case of
the aluminum, the measurement is a continuous one made
at about 400 ftihr and is shown in Fig. 50.5 I along with
core measurements from the same well. To detect the
vanadium, since it is present in rather small amounts, sta-
tionary readings were taken. These values are indicated
along with error bars representing the uncertainty of the
measurement caused by counting statistics. These two
traces illustrate the type of detailed evaluations that can
be made by logging devices, which previously could be
made only with extensive core analysis.
Saturation Determination
Inelastic Gamma Ray Spectroscopy. Two nuclear tools
are well adapted for the determination of saturation. The
first is based on the determination of the ratio of carbon
to oxygen atoms in the formation. The use of such a ratio
is shown clearly in Table 50.5. where it is seen that thcrc
is a significant difference between the atomic C/O ratios
of oil, water, and the common matrix minerals.
Fig. 50.52 shows how the atomic C/O ratio changes
as a function of porosity and water saturation for the three
principal matrices. From this figure it is clear that for
clean formations of a given lithology the interpretation
is relatively straightforward. An inherent difficulty in the
measurement is immediately obvious. At low porosities
the dynamic range of the C/O ratio as a function of water
Fig. 50.50-Gamma ray spectrum obtained with a high resolu-
tion Ge gamma ray spectrometer and Cf neutron
Vanadium Aluminium
200 0
Fig. 50.51 -Aluminum and vanadium logs of an oil-bearing shaly
sand derived from a high-resolution gamma ray
spectrometer. The solid bars represent the results
of core analysis. The dashed bars indicate the log
value of vanadium and its statistical uncertainty.
0 0.5
g 0.4
/* ,./-
*-I .
I 1 I 1
10 20 30 40
Porosity, 4, %
Fig. 50.52-Variation of C/O as a function of porosity and water
saturation shrinks to zero. Examples of interpretation of
this ratio in more complex situations can be found in the
references. 28Zu
Pulsed Neutron Logging. The second type of nuclear
device used for the determination of water saturation. par-
ticularly in cased wells, is the pulsed neutron device. This
log responds primarily to the time taken for the thermal
neutrons to be absorbed by the formation. The most com-
mon important thermal neutron absorber is chlorine.
which is present in most formation waters. Hence, the
response of the pulsed neutron tool resembles the usual
opcnhole resistivity measurements.
The advantage. however, of the pulsed neutron tech-
nique is the ability to log in cased hole. It can distinguish
between oil and salt water contained in the pores. If the
porosity is known, gas/oil interfaces can be distinguished.
Under ideal conditions of salinity. porosity, and litholo-
gy. the water saturation, S,,.. can be computed.
In the simplest case of a single mineral, the measured
value of C 1,j(, can be thought of as consisting of two com-
ponents. on; from the matrix and the other from the for-
mation fluid:
&,jg =(I -r$)z:,,,,, +4C,. (62)
Porosity, +
Fig. 50.53--Solution of S, with Z.
In terms of wate