Chapter 6

Multiphase Displacement

1
6.0 Historical Perspective Of Water Flooding
6.1 Immiscible Displacement Equations of Motion
6.1.1 Conservation of Mass and Momentum
6.1.2 Fractional Flow
6.2 Buckley-Leverett Theory- One-Dimensional Flow

2
Primary recovery
Secondary recovery

EOR
 Definitions And Classification Of Oil
Recovery Processes
Secondary recovery operations
 Involve the injection of gas, air or water into a reservoir, after
it has reached a state of substantially complete depletion of
its initial content of energy available for oil expulsion or
where the production rates have approached the limits of
profitable operation.
Advantages of Waterflooding
 Availability, low price

 Relatively easy to inject into formation

 Good displacement efficiency

 Definitions And Classification Of Oil
Recovery Processes
• Gas injection is the oldest of the fluid injection processes. The idea of
using a gas for the purpose of maintaining reservoir pressure and
restoring oil well productivity was suggested as early as 1864, just a few
years after the drilling of the Drake well. Since that time, injection of
natural gas or other gases has been used successfully in many oil
reservoirs throughout the world as a secondary recovery method.
• The early gas injection projects were designed to increase the immediate
productivity and should be classified as pressure maintenance projects.
However, recent gas injection applications have been intended to
increase the ultimate recovery and can be considered as enhanced
recovery projects. Gas injection may be either a miscible or an
immiscible displacement process.
 Historical Perspective Of Water
Flooding
 In 1880, Naturally occurring water drive was recognised
as a valid concept and it was suggested that it might be
possible to increase oil production.
 The earliest method of water flooding involved injection of
water into a central well in the reservoir to push oil into the
surrounding wells.
 As these wells watered out, they were converted to injection
wells to gradually increase the water-invaded area.

Peripheral water Outside-edge water Inside-edge water

flooding flooding flooding
1-production well, 2-injection well
 Historical Perspective Of Water
Flooding
 A line drive method in which a series of wells were used for
injection was developed next.
 1922, linedrive pattern was applied
 1924, five-spot pattern was applied
 By 1950, water flooding was being practised in most oil
producing areas.

linedrive pattern five-spot pattern Inverted 5-spot pattern

 Waterflood patterns
 One of the first steps in designing a waterflooding project is flood
pattern selection.
 Waterflood patterns are based on the arrangement of injectors and
producers in a waterflooded reservoir.
 Objective: to select the proper pattern that will provide the
injection fluid with the maximum possible contact with the crude
oil system, thus to improve areal sweep efficiency.

EA is the areal sweep efficiency; Ev is the vertical sweep

efficiency; ED is Displacement efficiency; E is recovery factor
EA is the areal sweep efficiency; Ev is the
vertical sweep efficiency;
 Waterflood patterns
 Flood pattern selection depends on the number and location of existing
wells. In some cases, it can be achieved by
 (1) converting existing production wells into injectors
 (2) drilling infill injection wells
When making the selection, the following factors must be considered:
 Reservoir heterogeneity and directional permeability
Vertical heterogeneity vertical permeability
Areal heterogeneity areal variation in formation properties (h, k, ϕ, Swc);
geometrical factors (faults);boundary conditions (gas cap, aquifer)

 Direction of formation fractures

 Availability of the injection fluid (gas or water)
 Desired and anticipated flood life
 Maximum oil recovery
 Well spacing, productivity, and injectivity
 Common waterflood patterns
 Irregular Injection Patterns
 Willhite (1986) points out that surface or subsurface topology
and/or the use of slant-hole(斜井）drilling techniques may result
in production or injection wells that are not uniformly located. In
these situations, the region affected by the injection well could be
different for every injection well.
 Some small reservoirs are developed for primary production with a
limited number of wells and when the economics are marginal,
perhaps only few production wells are converted into injectors in a
non-uniform pattern. Faulting and localized variations in porosity
or permeability may also lead to irregular patterns.
 Common waterflood patterns

 Common waterflood patterns include:

 Peripheral injection patterns （边缘注水）

 Crestal and basal injection patterns

Many water-injection projects use basal injection patterns

with additional benefits being gained from gravity segregation

 Direct/ staggered(交错）line drive

 Five-spot pattern

 Nine-spot pattern
 Common waterflood patterns
 Peripheral injection patterns
 Injectors are located at the external boundary of the reservoir and
the oil is displaced toward the interior of the reservoir.
 Objective: maintain reservoir pressure, minimize water
production, improve oil recovery.
 Fit for the reservoir with high- or medium permeabilities and
good flow communications between injectors and producers.
 Common waterflood pattern
 Crestal and basal injection patterns
 Crestal injection: the injection is through the wells located at the
top of the structure. Gas injection projects typically use a crestal
injection pattern.
 Basal injection: the fluid is injected at the bottom of the structure.
Many water injection projects use basal injection patterns with
additional benefits being gained from gravity segregation.
Common waterflood patterns
Crestal and basal injection patterns
 Common waterflood patterns
 Direct/ staggered(交错）line drive
 Direct line drive: the lines of injection and
production are directly opposed to each other.
 Staggered line drive: the injectors and
producers laterally displaced by a distance of a/2.
 The pattern is characterized by two parameters:
 a = distance between wells of the same type;
 d = distance between lines of injectors and
producers.
 Injection-Production ratio = 1: 1
 Common waterflood patterns
 Five-spot pattern
 This is a special case of the staggered line drive in which the distance
between all like wells are constant
 Any four injection wells thus form a square with a production well at
the center  ‘element’ of the flood pattern
 Injection-Production ratio = 1: 1
 Applied very broadly, usually results in good areal sweep efficiency.
 Common waterflood pattern
 Nine-spot pattern
 This pattern is similar to that of the five spot but with an extra
injection well drilled at the middle of each side of the square.
Therefore, this pattern contains 8 injectors surrounding 1
producer.
 This pattern is used in low permeability reservoirs where high
water injection volumes are desired.
 Injection-production ratio = 3: 1

4 corner inj wells working for 4 prod wells;

4 side inj wells working for 2 prod wells;
4*(1/4)+4*(1/2)=3
Common waterflood patterns

Four-spot Five-spot
(1:2) (1:1)

Seven-spot Inverted nine-spot Nine-spot

(2:1) (1:3) (3:1)
 The factors affecting the
displacement efficiency
 Characteristics of water flooding in
porous medium
 The pore structure
 The wettability
 The sedimentary rhythm of reservoir
 Characteristics of water flooding in
porous medium
During water drive, the injected water encounters a variety of resisting
forces before it displace the oil out of the pores and channels involved in
the porous medium, and the process also results in changes of the quantities
and distributions of the phases.
For example, the originally continuously distributed oil may be broken
into discontinued oil columns and then gradually into individual oil drops
as the water injection goes on.
In a word, the redistribution of the fluids in a formation results from
the mutual constraints between the oil displacement energies and the
various resisting forces encountered by the flow. At the same time, the
change in fluid distribution conditions also influences in turn the
subsequent process of water displacing oil.
 Non-Piston Like oil Displacement
Essentially speaking, the non-piston-like displacement in formation refers
to the difference in flowing velocity between the differently-sized
conducting channels involved in the porous medium. It can be manifested
in several respects.
(1) The complexity of the pore structure:
(2) The existence of capillary pressure:
(3) Viscosity difference between oil and water: viscous forces are different
(4) The resisting forces caused by the two-phase flowing in capillary tube
水区 两相区 油区
水 油
水+残余油 水+可动油 油

水 油
So
So
Soi
Soi

Sor
L 0
Pistion-Like Oil Displacement Non Pistion-Like Oil Displacement L
 Non-Piston Like oil Displacement

The complexity of the pore structure.

Coordination, pore-throat ratio, wettability

Real rock Simplified ideal model Capillary bundles model for

displacement
 Non-Piston Like oil Displacement
If the density difference between oil and water is very large and the
thickness of reservoir is high, and the velocity is small, then the gravity
segregation is very obvious, oil on the top and water is at the bottom.

oil

water

k rk w
λw = ,λo = k k ,o
µuww uµoo

In most situation, water can flow much easier than oil. There are advantage
water channel, where the following injected water will go through it, causing
serious water fingering.
 Pore structure parameters

(1) Pore-to-Throat Ratio: the ratio of the pore diameter to the throat
diameter.
(2) Coordination Number: the number of the throat channels
interconnected with a pore, normally ranging from 2~15 in sandstones.
(3) Tortuosity τ: The flow of fluid in a porous medium does not go
in a straight line but in a zigzag way forward. Trotuosity describe the
tortuous level (twisted, having many turns) of pores. The tortuosity τ is
defined as the ratio of the real flow length of the fluid particles (l) to
the apparent straight length of the rock (L), and its value usually ranges
within 1.2~2.5 for conventional reservoir rocks. As reported, the
tortuosity of the pore system for some shale rocks can be up to 8.44.
 Pore structure parameters
 The type, size distribution and sorting degree of pores and
throats can be quantitatively described by pore structure
parameters;
 The effective pore throat radius should be the actual pore
throat radius minus the liquid film thickness in the presence
of wetting phase fluid.
 Each throat can be connected to two pores, and each pore is
connected with at least three or more throats, and some are
even connected with six to eight throats, and the coordination
number of the pore throat is a quantitative characterization
of the connectivity of the pore system, which directly affects
the exploitation effect of the oilfield.
 Pore structure parameters

 Pore-throat ratio

Low recovery factor High recovery factor

Pore Pore
Throat
water
oil oil

Large Pore-throat ratio Small

 Clay minerals
 In the process of water flooding development, and the clay minerals and
cements between the rock pores might be partially produced together with
the fluids, so that the rock surface becomes clean and the rock
permeability increases, resulting in improved oil production. But it can also
affect oil production if the fine migration phenomenon is serious.
 For some reservoirs, due to the expansion of clay when exposed to water,
some throat may be plugged in the process of water injection development.

Fractures

Contact point
between
particles Cementing materials

Pores
 Wettability
 Through a large number of analysis of the core of the inspection well,
it is found that the wettability of the oil layer gradually changes with
the increase of the degree of water washing, and generally changes from
lipophilicity to hydrophilicity.
 Factors affecting wettability: Mineral composition, roughness of rock
surface, difference in oil-water properties and fluid saturation, etc. With
the increase of water saturation, the reservoir rock gradually changes
from lipophilic to hydrophilic, development is advantageous.

Water wetting means

the surface tends to be
wetted by water, and
the rock can imbibe
Water Oil water and oil can be
displaced
 Wettability
Wettability affects the distribution of oil & water in pores

Dynamic distribution :
Related with wettability & fluid saturation
Flow regions：oil , displacing front, behind displacing front

Water-wetting pores: water flows along the surface of rock pore.

Oil-wetting pores： water flows in the middle of pores
 Wetting Reversal

Wetting reversal means the phenomenon of reciprocal transformation

between water wettability and oil wettability on the surface of a solid.

So, surfactant flooding is

developed in the
development of oil reservoir.

1. water; 2. oil; 3. solid.

 imbibition
Wettability
Oil

Water

Empty tube inserts in water

Spontaneous imbibition measure drainage

Oil

Water
Imbibition cell
Tube filled with water inserts in water
 Capillary hysteresis caused by saturation history

Experiments Hollow tube in water Water-filled tube in

water
phenomenon Water level rise Water level drop
Pc is driving force; Pc is resistance force;
advancing angle receding angle decrease, Pc
increase, Pc decrease increase

Conclusions Water drive air Air drive water (non-

(wetting drive non- wetting drive wetting
wetting phase)— phase)—Drainage
Imbibition Process Process

Results Saturation history affects Pc.

Water height in imbibition process < drainage process
That is, Sw (imbibition) <Sw(drainage)
 Capillary pressure
AB: pc↑→So(nonwetting phase) ↑slowly
BC: pc↑little →So↑↑(major section of injecting liquid)
BC the smoother, the longer →the better sorting
BC the lower position →the larger average pore radius r
CD: pc↑sharply→So↑a little (tiny pores)

Typical capillary pressure curve

 Resistances caused by capillary pressure

• Flow resistance in multiphase flow comes from a great

deal of liquid droplets and gas bubbles in pores.
• Resistance effect also means the all kinds of droplet-
bubble effect in multiphase flow.

Jamin effect ：generally refers to various

effects of additional resistances which are
resulted from the deformation of droplet or
bubble while it passes through narrow pores.
 The effect of sedimentary rhythms on RF
• Positive rhythm: lower permeability at the top layer and
higher permeability at the bottom layer.
• At the bottom layer, the water cut can be up to 99% at
the end of water flooding, the oil saturation decreases
the most, and the water flooding efficiency is the
highest.
• The thickness of the flooding is small, and early water
breakthrough.

Water cut for different layers

Water cut
Total
2
1

Water saturation(%)
 The effect of sedimentary rhythm on RF

The compound rhythm with

uniform layer, the thickness
of the flooding is larger,
and the development effect
is better
 Things Reservoir Engineering
Concerned
1. Where is the waterflooding front?

2. What is the water breakthrough time?

3. What is the oil recovery factor?

4. How much water is needed for injection?

 Linear Reservoir Model
One-Dimensional Immiscible, Incompressible Displacement
of Oil by Water.

(a) Plan view (b) Cross section

Linear Reservoir Model
 6.1.1 Conservation of Mass and Momentum

Darcy’s law：

→ kk rw kk rw
water u = −
w ∇ Φ w = − ∇ ( p w + ρ w gh )( 4.1)
(6.1)
uµw uw
w
→ kk ro kk ro
oil u = −
o ∇Φ o = − ∇ ( po + ρ o gh )((6.2)
4.2 )
µuoo
uo

43
6.1.1 Conservation of Mass and Momentum
Considering capillary pressure:

p c = po − p w = pc ( s w ); p w = po − pc ((6.3)
4.3)

→ kk rw kk rw
uw =− ∇Φ w = − ∇ ( p w + ρ w gh )( (6.1)
uww
µ uww
µ
→ kk rw
So: uw =− ∇ ( po − pc + ρ w gh )((6.4)
4.4 )
µuww
44
6.1.1 Conservation of Mass and Momentum

Next, consider conservation of mass

on a control volume:

[Rate of mass change of water in Δx and Δy]

= [Net influx of mass of water]

45
 [Rate of mass change of water in Δx.Δy] = [Net influx of mass of water]
ΔZ is contained in each term, so it is cancelled

ΔZ
ΔZ

ΔZ ΔZ

∂
Figure 6.1[φρw S w ∆x∆y ]= ρ wu w x∆y + ρ wu w y∆x
∂tt (6.5)
∂ ∂
−[ ρ wu w x + ( ρ wu w x ) ∆x ] ∆y −[ ρ wu w y + ( ρ wu w y ) ∆y ] ∆x (
∂x ∂y 46
∂
[φρ w S w ∆x∆y ]= ρ wu wx∆y + ρ wu wy ∆x
∂ tt
∂ ∂
−[ ρ wu wx + ( ρ wu wx ) ∆x ] ∆y −[ ρ wu wy + ( ρ wu wy ) ∆y ] ∆x ( (6.5)
4.5 )
∂x ∂y

∂ ∂ ∂
φ [ ρ w S w ] ∆x∆y = − ( ρ wu wx ) ∆x∆y − ( ρ wu wy ) ∆y∆x ((6.6)
4.6 )
∂t ∂x ∂y

∂ ∂ ∂
φ [ ρ w S w ]= − ( ρ wu wx ) − ( ρ wu wy )((6.7)
4.7 )
∂t ∂x ∂y

∂ ∂ ∂
φ [ ρ w S w ]+ ( ρ wu wx ) + ( ρ wu wy ) = 0 ((6.8)
4.8 )
∂t ∂x ∂y
→ ∇. Nabla
∂
φ [ ρ w S w ]+ ∇ .( ρ w u w ) = 0 ((6.9)
4.9 ) Saturation equation47n
∂t
6.1.1 Conservation of Mass and Momentum
∂ →
φ [ ρ w S w ]+ ∇ .( ρ w u w ) = 0 ( 4.9 )
∂t
∂S w →
For constant water density: φ + ∇ .( u w ) = 0
∂t
Similarly, for constant oil density:
∂S o →
φ + ∇ .( u o ) = 0
∂t
adding the two equations together
Now, since So + Sw = 1
→ →
∂S w ∇ .( u o + u w ) = 0
→
−φ + ∇ .( u o ) = 0
∂t

48
6.1.1 Conservation of Mass and Momentum
→ →
∇ .( u o + u w ) = 0 q=u•A

Multiply by area to convert the flux →

velocity to volumetric flow rate q, then: ∇. ( q T ) =0

To obtain an equation in terms

of pressure, substitute Eqs. (6.1)
and (6.2):

→ → kkr w kkr o
uo + u w = − ∇ ( po − pc + ρ w g h) − ∇ ( po + ρ o g h)
µuww µuo o 49
 → → kkr w kkr o
uo + u w = − ∇ ( po − pc + ρ w g h) − ∇ ( po + ρ o g h)
uw uo

λ is Mobility
k krw kk
k kroo
λw = ,λo = ,,λT = λo + λw
uww
µ uµoo

→ → →
q T / A= u o + u w = − λT ∇po + λw∇pc − ( λo ρ o + λw ρ w ) g∇h (

→ →
Substituting to Equation 6.13:
∇ .( u o + u w ) = 0

∇ .[ λT ∇p o − λw∇pc + ( λo ρ o + λw ρ w ) g∇h ]= 0
 Pressure equation

∇ .[ λT ∇p o − λw∇pc + ( λo ρ o + λw ρ w ) g∇h ]= 0

For negligible gravity and capillary forces:

∇ .( λT ∇po ) = 0 ((6.19)
4.19)

• This is a nonuniform Laplace equation for the pressure.

• Laplace equations are elliptic and exhibit smooth “diffusive”

kinds of solution with no internal maxima and minima.

51
6.1.2 Fractional Flow

52
 ∂ →
φ [ ρ w S w ]+ ∇ .( ρ w u w ) = 0 ( 4.9 )
∂t
It is convenient to reformulate the saturation equation in
terms of fractional flow, defined as:
qw qw
fw = = ( (6.20)
q w + qo qT

We can rearrange Eq. (6.16) :

→ → →
q T / A= u o + u w = − λT ∇po + λw∇pc − ( λo ρ o + λw ρ w ) g∇h
∇ .[ λT ∇p o − λw∇pc + ( λo ρ o + λw ρ w ) g∇h ]= 0
as follows:
→
1 qT
− ∇po = [ − λw∇Pc + ( λo ρ o + λw ρ w ) g∇h ]
λT A
53
Substituting into the water flow equation,
→ kk rw kk rw
uw =− ∇Φ w = − ∇ ( p w + ρ w gh )
µuww µ
uw
→
u w = − λw∇ ( po − pc + ρ w g h) = − λw∇po + λw∇Pc − λw ρ w g∇h
→
1 q
−∇po = [ T − λw∇Pc + ( λo ρ o + λw ρ w ) g∇h ]
λT A
→
→ λ q
u w = w [ T − λw∇Pc + ( λo ρ o + λw ρ w ) g∇h ]+ λw∇Pc − λw ρ w g∇h (
λT A

→
→ λ q λw λw λw
uw = w ( T ) −( −1) λw ∇Pc + ( −1) λw ρ w g∇h + λo ρ o g∇h
λT A λT λT λT 54
 λw λw − λw − λo λo
( −1) = =− ((6.25)
4.25)
λT λT λT
→
→ λ q λo λo λw
uw = w ( T )+ λw∇Pc − λw ρ w g∇h + λo ρ o g∇h (
λT A λT λT λT

→
→ λ qT
uw = w [( ) + λo ∇Pc − λo ( ρ w − ρ o ) g∇h ]
λT A
In the absence of Pc and gravity, fw equals water mobility to total mobility.
→
qw uw λw λo A
fw = = = [1+ ( ∇Pc − ∆ρg∇h )]
qT → λT →
( qT / A ) qT 55
 ∂S w →
φ + ∇ .( u w ) = 0 ( (6.29)
4.29)
∂t
Now let’s look again →
∂S w qT
at the equation φ + ∇ .( f w ) = 0 ((6.30)
4.30)
describing ∂t A

conservation of mass → →
∇ .( qT f w ) = qT ∇f w ((6.31)
4.31)
of water, and
substitute the →
∂S w qT
fractional flow: +( ) ∇f w = 0 ( (6.32)
4.32)
∂t φA
→
→ q
Interstitial velocity v = T ( (6.33)
4.33)
φA

qT A ∂S w →
+ v ∇f w = 0 ((6.34)
4.34)
∂t
56
6.1 Immiscible Displacement Equations of Motion
6.2 Buckley-Leverett Theory- One-Dimensional Flow
6.2.1 Equations of Motion
6.2.2 The Buckley-Leverett Solution
6.2.3 Welge method
6.2.4 Oil Recovery Calculations

57
6.2 Buckley-Leverett Theory

– One-Dimensional Flow

6.2.1 Equations of Motion

58
Figure 6.3

59
 Buckley-Leverett Theory
It is necessary to understand what
can be expected to occur in the dip-
normal direction, that is, how the
fluids will be distributed in this
direction.
Assumption:
Homogeneous, water-wet reservoirs
Application of Fractional Theory in Oil Recovery

 Intention: is to describe the physics governing displacement,

as viewed through a linear cross section of the reservoir.

60
 Buckley-Leverett Theory
 A better way is to determine the derivative mathematically
by recognizing the concept of relative permeability ratio
(kro /krw)
 The relative permeability ratio expresses the ability of a
reservoir to permit flow of one fluid as related to its ability
to permit flow of another fluid under the same
circumstances.
 The permeability ratio is usually plotted on the log scale of
semi-log paper as a function of the saturation. Like many
relative permeability ratio curves, the central or the main
portion of the curve is quite linear.
Buckley-Leverett Theory
Express the central straight-line
portion ofthe relationship in the
following analytical form:
 Buckley-Leverett Theory
 The constants a and b may be determined by selecting
the coordinate of two different points on the straight-line
portion of the curve.
 The resulting two equations can be solved
simultaneously for the constants a and b. Notice that the
slope b in the above expression has a negative value.
 Buckley-Leverett Theory
This equation

can be substituted into fw Equation to give.

 Buckley-Leverett Theory

 The derivative of may be obtained

mathematically by differentiating
with respect to Sw to give:
 Buckley-Leverett Theory
Water+Oil

Supply boundaries

Drainage Channel
water Oil
The distribution curve
of water saturation

Swc- Connate water saturation Non-piston like displacement

Swf - Water saturation at front

Sor – Residual oil saturation

Swmax =1-Sor, – The maximum water

saturation in two-phase zone
The distribution curve of Sw
Drainage
Channel
Supply
boundaries
Buckley-Leverett Theory
Drainage
Channel
Supply
boundaries
Buckley-Leverett Theory
Drainage
Channel
Supply
boundaries
Buckley-Leverett Theory
Drainage
Channel
Supply
boundaries
Buckley-Leverett Theory
 Buckley-Leverett Theory
 There is a sudden change at
the front of water displacing
oil . Swf is constant, it does not
change with time.
 As water continues to be
injected, two phase zone
increases, and the Sw at any
cross-section is increasing. Sw
µr3 > µr2 > µr1
1
µr1
 Think: What is the Sw profile µr2
µr3
change at different oil-
water viscosity ratio? Swc
0 xo xe x
 Buckley-Leverett Theory
→
qw uw λw λo A
fw = = = [1+ ( ∇Pc − ∆ρg∇h )]
qT → λT →
( qT / A ) qT

since dh/dx = sin θ.

where the fractional flow, fw, is now given by:

λw λo A ∂Pc
f w = [1 + ( − ∆ρg sin θ )]
λT qT ∂x
72
 oil

water

Clearly the fractional flow, fw, is a function of saturation

and dip angle:

f w = f w ( S w ,θ ) (6.40)
• For flow updip (θ>0) gravity retards the water flow.
• For flow downdip (θ<0) gravity increases the water flow.
• If qT/A is large, then the flow is viscous-dominated,
capillary pressure and gravity have less influence.
73
The effect of gravity and Pc
The total effect of gravity and Pc makes the Sw
front change more smoothly

Residual oil

gravity

capillary
Displacement front
6.2.2 The Buckley-Leverett Solution

75
 Buckley-Leverett
 One of the simplest and most widely used methods of
estimating the advance of a fluid displacement front in an
immiscible displacement process.
 The B-L theory (1942) estimates the rate at which an
injected water bank moves through a porous medium.
 The approach uses fractional flow theory and is based on
the following assumptions:
 Flow is linear and horizontal

 Water is injected into an oil reservoir

 Oil and water are both incompressible

 Oil and water are immiscible

 Gravity and capillary pressure effects are negligible

 Fractional flow equation

Without considering Pc and gravity, then:

Permeability ratio

Semilog Plot of Relative Permeability

Ratio versus Saturation
 Fractional flow curve

Swi 1-Sor
The rule of fw varying with Sw:
Slow changes at two ends and fast
change in middle section.
 Fractional flow curve

Fractional flow curve for horizontal flow

Favorable mobility ratio is when M<1
 Saturation equation

Saturation equation

for this one-dimensional problem the equation becomes:

∂S w ∂f w
+v = 0(
∂t ∂x

It is not possible to solve for the general distribution of water

saturation Sw( x,t ) in most realistic cases because of the
nonlinearity of the problem.

80
 To simplify the problem:

∂S w ∂f w
+v = 0
∂t ∂x

Where the interstitial velocity v=qT/ϕA (6.55)

we can write this as: ƒw = F( Sw ) and Sw = G( t,x ).

∂S w df w ∂S w
+v = 0 ((6.56)
4.56)
∂t dSw ∂x

81
 B-L frontal advance equation
∂S w d fw ∂S w
+v =0
∂t d Sw ∂x

The full differential of Sw is:

Since it is the intention to study the

rate of advance of a constant
saturation plane, that is: dSw / dt = 0

 Buckley-Leverett frontal
advance equation:
-

VSw = Is the velocity of the moving plane with water saturation Sw

 B-L solution

The distance a plane of constant saturation

If qt is constant, then we have:

Xsw is proportional to time and to the

value of the derivative
Water Fractional Flow and its Derivative

xSw is the distance traveled by a particular Sw contour

Qinj (=qt×t) is the cumulative water injection at reservoir conditions.
 B-L solution
 Example: The following figure indicates that the water
saturation after 240 days at 330 ft is 20, 39, and 69%. But Sw
can be only one value at any place and time.

Actually, the intermediate values of Sw have the maximum velocity, will initially
tend to overtake the lower saturations resulting in the formation of a saturation
discontinuity or shock front.
 B-L solution
 A discontinuity in Sw at a flood front location Xƒ is needed to
make the water saturation distribution single valued and to
provide a material balance for displacing fluid.

(a) Saturation derivative of a typical fractional flow curve

(b) Resulting water saturation distribution in the displacement path.
 6.2.3 Welge method

A more elegant method of achieving the same result was

presented by Welge in 1952. The Welge technique is a
simplified form of the B-L or frontal advance method.

This consists of integrating the saturation distribution

over the distance from the injection point to the front,
thus obtaining the average water saturation behind the
front Swf.

86
 Water Breakthrough
After breakthrough
 No sharp saturation front present in the system and the water
saturation is higher than Swf everywhere.
 Oil displacement continues at high fractional flow of water.
 The saturation at the producing end is related to the volume of water
injected.

Water Breakthrough is when the water front arrives at the production well
 Average water saturation
Water Saturation Distribution Before breakthrough
0
There is a discontinuity in
the water saturation which
increases abruptly from
Swc to Swf, the flood
cumulative w
saturation. (Shock Front).

• At this time the maximum water saturation, Sw = 1 – Sor, has moved to X1,
• The flood front saturation Swƒ is located at position X2.

Where, Qinj is cumulative water injection.

Qinj/X2Aϕ is pore volume (PV) injected
 Average water saturation
At breakthrough time:
Before breakthrough:
PVI: Pore Volume of Injection
 dx  qt  d f w 
  =  
 dt S w Aφ  d S w 
 Average water saturation
The average Sw in the
reservoir at the time of
breakthrough is given by material
balance as:

Tangent to the Fractional Flow Curve from Sw = Swi

 Example
• A well will produce water-free oil until the flood front approaches
the well. When a producing formation consists of two or more
rather definite strata of different permeabilities, the overall effect
will be a combination of several separate displacements.

Saturation Distribution After 240 Days

 6.2.4 Oil Recovery Calculations

This simple graphical technique of Welge has much

wider application in the field of oil recovery calculations.

92
 Oil Recovery Calculations

Figure 6.20: Saturation determined at outlet, xD = 1.

93
 Oil Recovery Calculations
Since all the oil produced has been replaced by water, the average
saturation Swave can be related to the cumulative oil production, Np:

Np Np
Sw a v =
e S w i+ Sw = S w i + N p D , w h er eNp D = (
Vp a ve V p

N p D= S w a v e− S w i

Prior to breakthrough the average saturation is given by:

S w i (φA L) + qT t S w a v =e S w i+ t D
t q
T 1 t
S w a v e= ( tD = ∫ ( ) d =t ∫ qT d =
φA L Vp 0
φA L 0

tD is pore volume injected

 Oil Recovery Calculations
In particular, Swave behind
the shock front is given by:

1− f w f
= f w' f
Sw a v −
e Sw f

this slope is also equal to

1
= f w' f
Sw a v −
e Sw i
Average saturation determined from
intercept with fw = 1
95
 Oil Recovery Calculations
After breakthrough: 1 − f w1
= f w' 1
Swave is given by the intersection of tangent to ∧
S w (t D ) − S w1
the fractional flow curve to the axis fw=1

Average saturation determined from intercept with fw = 1. 96

SUMMARY and EXERCISES
 Fractional flow equation

 Frontal advance equation

 dx  qt  d f w 
  =  
 dt S w Aφ  d S w 
 A Typical Fractional Flow Curve

 Below Swi, fw is zero since the Sw = 1-Sor

relative permeability to water is 100 100
zero and only oil can flow. 90 ko
k ro = 90
k

Relative Permeability (%)

Above Sw = 1-Sor , fw=1 since kro 80 80

(Expressed as % of Kabs.)

is zero and only water can flow. 70 Oil
70
Two-Phase Flow
60 Region 60
 Between Swi and 1-Sor, fw 50
Water
50
fw
increases continuously. 40 40
30 30
 This type of fractional flow curve 20 20
can be calculated directly from
10
the relative permeability data. 10
0 0
0 Swc 20 40 60 80 100
Water Saturation (%PV)
Connate Water Saturation
(Single-Phase flow region)
Residual Oil Saturation
(Single-Phase flow region)
Frontal Advance Equation For 1D
Displacement

Buckley-Leverett model:
 Provides a graphical methodology for predicting
the volume of oil displaced as a function of the
volume of water injected.
 Not an exact solution because it neglects the
effect of capillary pressure gradient.
  Problem solved by Buckley & Leverett along with the saturation
distribution within the linear reservoir at later stages of water flood

 Fig. a. Initially, the whole reservoir is at Swi.

 Fig. b. Injection of water qt , A sharp
saturation gradient develops.
 Fig. c. Ahead of the sharp front, the water
saturation remains at Swi, and only oil is
flowing. Behind the saturation front oil and
water flow simultaneously. Eventually the
front arrives at the end of the reservoir.
This is called "water breakthrough".
 Fig. d. After breakthrough, the fraction of
oil in the effluent decreases as the
remaining oil is displaced by flowing water.
Finally, the reservoir is at 1-Sor
 dx  qt  d f w 
  =  
 dt S w Aφ  d S w 

Integrating with respect to time,

starting from the beginning of
water injection :
t
1  df w 
xS w =   ∫ qt dt
Aφ  dS w at S w 0
The location of different
saturation planes, after
injecting a given volume of water,
Wi, can be determined from the
slope of the fractional flow curve
W i  df w 
xS w =   evaluated at the saturation value
Aφ  dS w at S w of interest
 The tangent to the fractional flow
curve drawn from the point (Swc, 0) 1.00

must intercept the line fw = 1 at the 0.90 Sw

average water saturation, Sw. 0.80
 Provides a very convenient method fwf
0.70
for calculating Sw behind the flood

fw (fraction)
front and hence for the location of 0.60
Swf
the front for a given Wi before 0.50
breakthrough. 0.40
 Allows one to calculate the volume of
0.30
water that can be injected (and hence
the volume of oil that can be 0.20
Swc
produced) before the water breaks 0.10
through at the producing end . Wi 0.00
can be calculated from the 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80
following equation, when x=L.
Water Saturation (fraction)

i  df w 
Fractional flow curve with tangent
W
xS w =   drawn from point (Swi.0) to determine
Aφ  dS w at S w Swf, fwf and (Sw)avg.
If the rate of water injection, qi, is kept constant, the time of
water breakthrough can be calculated from:
ALφ ( S w - S wc )
tb =
qi
L - length of the one dimensional reservoir (the distance between the
injection and production ends in the linear system).

After breakthrough:
 No sharp saturation front present in the system and the
water saturation is higher than Swf everywhere.
 Oil displacement continues at high fractional flow of water.
 The saturation at the producing end is related to the volume
of water injected.
Note: Value of xsw is now equal to the length of the linear system, L.

L Aφ  df w 
=  (Eqn. 20)
Wi  dS w  S w

The material balance equation for calculating the average water

saturation after water breakthrough:
W 1− fw2
Sw = Sw2 +
i
(1 - f w2) = Sw 2 + (Eqn. 21)
AφL f w′ 2
The total volume of oil displaced at reservoir condition since the
start of water injection:
N p = A φ L( S w - S wc ) (Eqn. 22)

Volume at surface conditions:

A φ L( S w - S wc )
( N p )surface = (Eqn. 23)
Bo
Exercise 1
Calculate the Sw distribution in two-phase zone
An active oil reservoir by waterflooding, three production wells are located
and the production rate for each well is q=31.8 m3/d.

Width b=402m
Thickness h=6.1m
Φ=0.25
Bo=1.5
µo/ µw=2
Kro,Krw~Sw is known.

What is the Sw profile after 60days, 120days and 240days?

Exercise 1
Calculate the Sw distribution in two-phase zone
The distance of a plane of constant Sw

Solution:

Step 1

Reservoir condition
Exercise 1
Calculate the Sw distribution in two-phase zone
The distance of a plane of constant Sw
Solution:

Step 1

Step 2

Step 3 The Sw profile after 60days, 120days and 240days

Exercise 1
Calculate the Sw distribution in two-phase zone
Step 4
 Exercise 1
Calculate the Sw distribution in two-phase zone

Step 5

It means at the same location, there are two

Sw, which is impossible.
 Exercise 2

Procedure For Predicting Linear Waterflood Performance

 Objective: Estimate the cumulative oil produced, oil production
rate and water/oil ratio as a function of time from the known
reservoir characteristics and the selected water injection rate.

 Assumptions:
 Flow is one dimensional.
 Connate water saturation is uniform and immobile.
 Both oil and water are incompressible.
 Capillary pressure effects can be neglected.
 The selected water injection rate is compatible with the total
injectivity of the available injection wells.
Graphical Procedure:

Requires a simple calculator and a graph paper.

1. Divide the Sw range between Swc and (1-

Sor) into several nearly equal segments (typically
20). Prepare a table listing the values of Sw, kro and
krw for each value of Sw.

2. Calculate the fw corresponding to each value of Sw.

The value of flood velocity, ut, is equal to qi/A.

3. Plot the fw ~Sw on a square grid graph paper.

Draw a smooth curve through the points. Mark Swc
on the x axis (line fw = 0). Draw a line parallel to x
axis at fw = 1.
 Graphical Procedure:

4. Draw a tangent to the fractional flow curve from Swc and

extend the tangent line to intersect the line fw = 1. The
point where the tangent meets the fractional flow curve
gives the fractional flow and the water saturation at the
front. The point where the tangent intersects the line fw =
1 gives the average water saturation behind the front.

5. Calculate the volume of water injected (which is equal to

the volume of oil produced) up to the time of
breakthrough from the average water saturation
calculated in previous step. Calculate the time for water
breakthrough by dividing total volume injected with flow
rate, qi.
 N p = A φ L( S w - S wc )
Graphical Procedure:
6. Re-plot the upper part of the
fractional flow curve
(containing values higher than
the frontal fractional flow) on
an expanded scale.

7. Select a Sw>Swf (say higher

than 1%). Draw a tangent to
the fw curve at Sw. For
example: the point gives Sw2,
and fw2, at the producing wells.
The point where this tangent
intersects the fw = 1 line,
gives .
Fractional flow curve drawn on an
expanded scale for computation of
performance after breakthrough.
 Graphical Procedure:
8. The total volume of water injected
to reach this condition, Wi.

W i = L A φ ( S w - S wc )
Calculate the oil production rate by
multiplying the water injection rate
with (1-fw2).

9. Repeat step 7 at progressively

higher values of water
saturations, say in increments
of 1%. Continue the process
until the fractional flow of
water exceeds the specified Fractional flow curve drawn on an expanded
criterion for abandoning the scale for computation of performance after
water flood. breakthrough.
Example Problem 6-1.
A line-drive water flood is being considered for a reservoir having
the following characteristics.
Distance between 200m
injection and •Oil water relative permeability curves
production wells as shown below
Average reservoir 75m
width
Average formation 20m Estimate the following for this
height proposed waterflood project.
Porosity 20% (a)Volume of oil (at surface conditions)
Formation dip angle 0 degree recoverable before water breakthrough.
Average permeability 50 md (b)Time of water breakthrough.
Proposed water 150 m3/d (c)Total volume of oil (at surface)
injection rate recoverable assuming abandonment at
Swc 10% water cut of 97%.
Oil viscosity at 1.0 mPa.s
reservoir P, T (d)Rate of oil and water production
Water viscosity at 0.5 mPa.s (at surface conditions) as a function of
reservoir P, T
Bo at current 1.2
time.
conditions (e)Water/oil ratio (at surface
Bw 1.0 conditions) as a function of time.
 Solution
1. Calculate the Water Relative Permeability Fractional
Saturation Flow of
fractional flow Sw Oil, kro Water, krw Water
curve using: (fraction) (fraction) (Fraction) fw

0.10 1.000 0.000 0.0000

1
f w= 0.30 0.373 0.070 0.2729
 ko   µw 
1+     0.40 0.210 0.169 0.6168

 k w   µo  0.45 0.148 0.226 0.7533

0.50 0.100 0.300 0.8571

qw / Bw
watercut = 0.55 0.061 0.376 0.9250
qw / Bw + qo / Bo
0.60 0.033 0.476 0.9665

f w / Bw 0.65 0.012 0.600 0.9901

watercut =
f w / Bw + f o / Bo 0.70 0.000 0.740 1.0000
 Solution
2. Plot the fractional flow
curve and draw a tangent
to determine the water
saturation at the front &
the average water
saturation at the time of
breakthrough
3. Calculate the volume of water injected and oil
produced up to breakthrough and the time of water
breakthrough using:
W i = L A φ ( S w - S wc ) =27780 m3

ALφ ( S w - S wc )
tb = =185.2 days
qi

4. Calculate the post breakthrough performance:

1
Pore Volumes Injected = W i =
L A φ  df w 
 
 dS w S w2

(1 - f w2 )
S w = S w2 +
f ′w2
A φ L( S w - S wc )
( N p )surface =
Bo
f /B w
WOR = w2
（1 - f w2 ）/ Bo
 (1 - f w2 )
S w = S w2 +
f ′w2
Wi __
Sw2
Fw2 Cumulative Sw (Np)surface
Exit-end dfw/dSw
Exit-end Water cut Water Average m3
Water Slope of
Flowing at Injected Water Cumulative WOR
Saturation Fraction at
Fraction of surface (PV) Saturatio Oil
(Fraction Flow Curve
Water Wi=1/(dfw/ n Produced
PV)
dSw) (PV)
0.469 0.798 0.826 2.16 0.463 0.563 23150 4.74

0.495 0.848 0.870 1.75 0.472 0.582 24100 6.696

0.520 0.888 0.905 1.41 0.711 0.600 25000 9.516

0.546 0.920 0.932 1.13 0.887 0.617 25850 13.8

0.572 0.946 0.955 0.851 1.176 0.636 26800 21.024

0.597 0.965 0.971 0.649 1.540 0.652 27600 33.084

0.622 0.980 0.983 0.477 2.100 0.666 28300 58.8

0.649 0.990 0.992 0.317 3.157 0.681 29050 118.8

0.674 0.996 0.997 0.195 5.13 0.694 29700 298.8

0.700 1.000 1.000 0.102 9.80 0.700 30000

A φ L( S w - S wc ) WOR =
f w2 /B w
( N p )surface = （1 - f ）/ Bo
Bo w2
 Exercise 3
 The following data are available for a linear reservoir
system:

 Calculate and plot the water saturation profile after 60, 120,
and 240 days.
Exercise 3
 Step 1. Plot the relative permeability ratio kro /krw vs. water
saturation on a semi-log paper and determine a and b.

a=537.59
b=-11.51

Therefore
Exercise 3
 Step 2. Assume several values of water saturation and
caIculate the fraction flow curve at its derivatives by applying
Exercise 3

 Step 3. Plot fw and vs. Sw on a

Cartesian scale as shown in the figure. Draw a
straight line from Swc and tangent to the fw
curve.

 Determine the coordinates of point of tangency

and the slope of the tangent to give:
Exercise 3

 This means that the

leading edge of the
waterfront (stabilized
zone) has a constant
saturation of 0.596
and water cut of 78%
Exercise 3

 Step 4. When constructing the water saturation

profile, it should be noted that no water
saturations with a value less than Swf i.e., 59.6%,
exits behind the leading edge of the water bank.

 Assume water saturation values in the range of

Sw to (1 –Sor) , i.e., 59.6 to 75% and calculate
the water saturation profile as a function of time
Exercise 3
Exercise 3

Step 5. Plot the water saturation profile as a function of

distance and time
Exercise 3

 To determine the time to breakthrough, tBT,

simply set Xswf equal to the distance between the
injector and producer L and solve for the time:
Exercise 3

 Note that the pore volume(PV) is given by

 solving for the time to breakthrough tBT gives:

Exercise 3

 Step 6. Calculate the reservoir pore volume:

Step 7. Calculate the time to breakthrough from:

Exercise 3

 Step 8.Determine cumulative water injected at

breakthrough:
Exercise 3

 Step 9. Calculate total pore volumes of water

injected at breakthrough: