Copyright 2000, Society of Petroleum Engineers Inc.

This paper was prepared for presentation at the 2000 SPE Permian Basin Oil and Gas
Recovery Conference held in Midland, Texas, 21-23 March 2000.
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300
words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
Abstract
Acidizing is often applied in horizontal wells to improve
well performance. Acidizing in horizontal wells differs
significantly from the acidizing of vertical wells. The length of
acid contact is much longer, the formation properties over the
contact length likely vary in a larger range, and the cost is
generally higher. To enhance the effectiveness of matrix
stimulation, methods to evaluate and optimize acid treatments
similar to those used for vertical wells can be adopted for
horizontal wells.
This paper introduces a simple approach to evaluate
horizontal well response to acid treatments. Skin factor was
calculated explicitly in the method from pressure and flow
rate, which can be directly measured during a treatment. Skin
factor is then used to identify the well response to acid
injection, and therefore to understand and control the
stimulation process. There have been several mathematical
models published to simulate the pressure/flow rate behavior
of horizontal wells. Most of them have complicated solutions
and have to be solved numerically. We selected and modified
analytical solutions for horizontal well performance that can
be implemented for field evaluation of acid treatments.
Field examples presented in the paper show how to apply
the evaluation method for horizontal wells. The method can
assess the effectiveness of acid treatments and can be used to
optimize subsequent treatments.
Introduction
Acidizing in horizontal wells is significantly different
than in vertical well. In a horizontal well, formation contact
area is much longer, and acid distribution inside the wellbore
is a greater issue
1
. The acid injection rate per length of
formation treated is also generally much lower for horizontal
wells than for vertical wells. This reduced injection rate can
reduce the efficiency of the acidizing process, particularly in
carbonate reservoirs, where the creation of wormholes is
desired
2
. Simulating the acidizing process in horizontal wells
involves more uncertainty, which directly affects the
reliability of acidizing treatment design. In addition, acid
treatment costs for horizontal wells are generally higher than
for vertical wells because of higher injection volumes, the
common use of coiled tubing, and more complicated injection
procedures. For these reasons, it is important, but also very
hard, to understand what has happened in horizontal well
acidizing, and to identify the reasons resulting in success or
failure. To enhance the effectiveness of acidizing treatments in
horizontal wells, we introduce a skin analysis method similar
to that used in vertical well acidizing. The method can be
applied to matrix acidizing either during or after the treatment,
to identify the well response to the acidizing treatment
3,4
. The
method is based on transient flow theory for pressure behavior
in horizontal wells. It calculates skin factor from pressure,
flow rate, and other information about the well. In order to use
the method during an acid treatment in real time, the model
has to be simple and analytically solvable. This is a key
criterion when selecting the mathematical models for acidizing
evaluation.
Mathematical Models
Many models for horizontal well flow and pressure
behavior have been developed. Two models will be discussed
in this paper. When used on-site, the time interval to perform
the calculation could be very small if desired. To link the
computation model to an automatic data acquisition system, it
is critical that the model is numerically stable, and the
computing time is short. Therefore, the mathematical model
SPE 59804
Evaluation of Acid Treatments in Horizontal Wells
D. Zhu, SPE, A. D. Hill, SPE, The University of Texas at Austin, and M. D. Looney, SPE, Texaco, Inc.
SPE 59804 Evaluation of Acid Treatments in Horizontal Wells 2
implemented in the evaluation method should be simple, and
an analytical solution is preferred, if available. It is convenient
if the model has a form
b t m
q
p p
i wf
+

sup
(1)
where the left-hand side of Eq. 1 is inverse injectivity, and
t
sup
is a superposition time function. The parameters m and b
in Eq. 1 are the slope and the intercept of the linear
relationship between the inverse injectivity and the time
function. The time function can be different for different
models, as can m and b. The two models we selected for the
evaluation method both have analytical solutions with a linear
relationship between inverse injectivity and a time function.
Early-time linear flow model. One commonly used model
for horizontal well flow is to divide the flow behavior into
four periods: early-time radial flow, early-time linear flow,
pseudo-radial flow and late-time linear flow
5
. The model has
simple analytical solutions for each flow regime, and three of
them can be represented by Eq. 1. The transition time from
one flow regime to another can be calculated by the equations
presented by Joshi
5
. When we tried to apply this model to the
evaluation method, we obtained a discontinuous skin factor
when the flow regime changed from one to another. To
smooth the skin factor, we tried different averaging methods
(linear average, logarithm average, time-weighted average),
but the results could not be practically explained. Since for
most acid treatments the pumping time is on the order of
hours, the early-linear flow equation covers the time period of
most acid treatments. Thus, for many treatments, the entire
time can be approximated by the early-linear flow model. The
constants in Eq. 1 for transient pressure behavior in the early-
linear flow period are
y z w
k k L
B
m
6 . 162
(2)
1
1
]
1

,
_

s
r c
k
m b
w t
y
87 . 0 23 . 3 log
2

(3)
and

N
j
j N
N
j j
t t
q
q q
t
1
1
1
sup
(4)
where L
w
is the length of horizontal well, k
y
is the
permeability in the horizontal direction perpendicular to the
wellbore, and k
z
is the vertical permeability.
Notice that the superposition time function now is a
square root function of time, rather than a logarithm function,
as is used for the vertical well model. The procedure to use the
evaluation method is to first calculate the slope, m, by Eq. 2,
and at each time increment calculate t
sup
by Eq. 4 and the
inverse injectivity term in Eq. 1. The intercept, b, then can be
calculated from Eq. 1. Finally, skin factor at the current time is
obtained from Eq. 3. The skin evolution shows the well
response to acid injection, which directly reflects the
effectiveness of acidizing.
Even though the early-linear flow equation covers the
time period of most acid treatments, the pressure behavior
may deviate from the equation at very early time since the
radial flow regime may occur.
Semi-Infinite Slab Model. Another model that can be
implemented in the real-time monitoring model is the semi-
infinite slab model
6
. The semi-infinite slab model relaxes
some assumptions made in the four flow-regime model, and it
gives a continuous skin factor, but the analytical solution of
the model is very complicated.
The semi-infinite slab model was developed based on a
flow field being bounded at the top and the bottom, and in the
direction of wellbore length, but not in the third dimension.
Fig. 1 shows a sketch of the geometry for the model. Eq. 1
applies to this model, but the parameters now are defined as
y z x
w
k h h
Br
m
' 4 . 282
(5)
( )

i
j
j D N D
N
j j
t t
q
q q
t
1
1 , ,
1
sup
(6)
and
s
L k k
B
q
q q
v L
hxh
m
q
q q
v
h
m b
w z y
N
j N
j j
z w
z
N
j N
j j
x
x

2 . 141
1
1
1
1
2
2
+

(7)
where the geometry functions and dimensionless groups in the
above equations are defined as
( )

1
]
1

1
2
1 ,
1
n
n j D D x
t t n v erf
n
(8)
( ) ( )

1
]
1

1
1 ,
cos
1
l
e l j D D z
z l t t l v erf
m
(9)
1
1
]
1

,
_

,
_

x
xd
x
xl
w
n
h
L n
h
L n
nL

sin sin
1
(10)
SPE 59804 Evaluation of Acid Treatments in Horizontal Wells 3
( ) ( )
1
1
]
1

,
_

,
_

+ ' 2 sin ' 2 sin

' 4
1
w s
z
w s
z w
l
r h
h
l
r h
h
l
lr

(11)
( ) ' 47 . 1
1
w s
z
e
r h
h
z + (12)
'
000264 . 0
w t
y
D
r c
t k
t

(13)
y
x
x
w
x
k
k
h
r
v
'
(14)
y
z
z
w
z
k
k
h
r
v
'
(15)
( )
xd xl w
L L L (16)
4 / 1
'

,
_

y
z
w w
k
k
r r (17)
In the above equations, r
w
is the equivalent wellbore radius
defined by the last equation in the group. Other geometry
parameters are shown in Fig. 1. h
x
is the length of the modeled
domain, h
z
is the height of the domain, L
xl
is the right
coordinate of the wellbore, and L
xd
is the left coordinate of the
wellbore x. The above equations are complicated, especially
for application in real time. When used in the evaluation
method, the error function is approximated numerically, and
the infinite summation terms are approximated by the first 40
terms, which gives a stable result. To link the model to an
automatic data acquisition system, it is advised that the time
interval should be relatively longer than the one for the other
models (1 minute is recommended).
The early-linear flow model is simpler and more stable
numerically, but the semi-infinite slab model relies on fewer
assumptions when developed, and should be more accurate
than the early-linear model. The calculation results from both
models can be compared in a post-treatment study.
Both models presented here do not consider the effect of
reservoir heterogeneity on skin factor calculated from pressure
and flow rate. In other words, distribution of permeability and
skin along the horizontal lateral of the well is lumped into
three constant values of permeability, k
x
, k
y
and k
z
, and one
total skin factor. As mentioned before, the acid contact length
with the formation in horizontal wells is much longer than in
vertical wells (horizontal section length versus formation
thickness), and in most cases, the permeability and degree of
damage would not be uniform. The evaluation model only
gives a global estimate of the overall response of the well, but
not a zonal description of skin distribution along the wellbore.
It is recommended that other detailed zonal models be used to
check the result from the evaluation model after the treatment
to confirm the interpretation of the treatments effect.
Field Example
To test the horizontal well skin evolution model, we
analyzed a matrix acidizing treatment performed on a
horizontal well in a carbonate formation in the Permian Basin.
Information about the well and the formation are given in
Table 1. With an initial skin factor of near zero, the objective
of this treatment was to create sufficiently long wormholes to
give a post-treatment skin factor of 2 to 3. However, the
injection rate for the treatment was low due to friction
pressure in the coiled tubing used to convey the acid.
The surface injection rate and pressure history for the
treatment are shown in Fig. 2. The skin evolution determined
using the early-linear flow model is presented in Fig. 3. The
analysis shows that the skin factor did not change significantly
throughout the entire treatment. This response was confirmed
by the post-treatment productivity of the well, which was only
3% higher than the pre-treatment productivity. Apparently the
low injection rate used in this treatment resulted in compact
dissolution of the face of the carbonate formation, with no
significant wormhole penetration. This agrees with the
prediction of wormhole models
2
. A subsequent acid fracture
treatment successfully stimulated the well, confirming that
deeper acid penetration was needed.
Conclusions
A simple method of evaluating acid stimulation for
horizontal wells has been developed. The method uses
pressure and flow rate data during an acidizing treatment to
calculate skin factor as a function of injection time. From the
skin evolution, the well response to the treatment can be
diagnosed. It has been concluded that
1. The method is a practical tool to enhance the
effectiveness of acid stimulation in horizontal wells.
2. To apply the method in real time, the model has to be
simple. Two mathematical models that are sufficiently
robust for real-time use were discussed in the paper.
3. The skin analysis model presented yields the global skin
factor. A more detailed model is needed to provide
information about zonal acid coverage.
4. A field example illustrates application of the method.
Comparison of the skin evolution determined with the
wells post-treatment productivity confirms the validity of
the model used. Further tests are needed to confirm its
applicability for real-time monitoring of acidizing of
horizontal wells.
Nomenclature
b = intercept of linear function between flow rate,
pressure and time
SPE 59804 Evaluation of Acid Treatments in Horizontal Wells 4
B = formation volume factor
c
t
= total compressibility of reservoir
h
s
= coordinate of wellbore location in z-direction
h
x
= length of the flow field modeled
h
z
= height of the flow field modeled
k
x
= permeability in x-direction
k
y
= permeability in y-direction
k
z
= permeability in z-direction
L
w
= length of horizontal well
L
xd
= left coordinate of wellbore in x-direction
L
xl
= right coordinate of wellbore in x-direction
m = slope of linear function between flow rate, pressure and
time
p
wf
= bottom flowing pressure
p
i
= reservoir pressure
q = flow rate
r
w
= wellbore radius
r
w
= equivalent wellbore radius
s = skin factor
t
D
= dimensionless time
t
sup
= superposition time function
= porosity
= viscosity of reservoir fluid
References
1. Eckerfield, L. D., Zhu, D., Hill, A. D., Thomas, R. L.,
Robert, J. A., and Bartko, K.: Fluid Placement Model for
Stimulation of Horizontal or Variable Inclination Wells,
paper SPE 49103 presented at the SPE Annual Technical
Conference and Exhibition, New Orleans, LA, 27-30,
September, 1998.
2. Huang, T., Hill, A. D., and Schechter, R. S.: Reaction
Rate and Fluid Loss: The Key to Wormhole Initiation and
Propagation in Carbonate Acidizing, paper SPE 37312
presented at the SPE International Symposium on
Oilfield, Houston, TX, 18-21 February, 1997.
3. Hill, A. D. and Zhu, D., "Real-Time Monitoring of Matrix
Acidizing Including the Effects of Diverting Agents,"
SPEPF (May, 1996) 95-101.
4. da Motta, E. P., dos Santos, J.A.C.M., Zhu, D., and Hill,
A. D., "Field Evaluation and Optimization of Matrix
Acidizing Treatments," paper SPE 37460 presented at the
Society of Petroleum Engineers Production Operations
Symposium, Oklahoma City, Oklahoma, March 9-11,
1997.
5. Joshi, S. D.: Horizontal Well Technology, PennWell
Books, Tulsa, Oklahoma, 1991.
6. Goode, P. A. and Thambynayagam, R. K. M.: Pressure
Drawdown and Buildup Analysis of Horizontal Wells in
Anisotropic Media, SPE Formation Evaluation,
December 1987.
Acknowledgement
The authors thank the sponsors of the Stimulation, Logging,
and Formation Damage Research Program at the University of
Texas at Austin for support of this work. The authors also
thank Texaco Inc. for permission to publish this paper.
Table 1. Well Data for the Example Treatment
Reservoir permeability 0.2 to 2 md
Porosity 10%
Thickness 110 ft
Fluid viscosity 3.3 cp
Reservoir pressure 1800-2000 psi
Formation volume factor 1.2
Total compressibility .0001-.001
Drainage area 20 acres
Lateral length 868 ft
Wellbore radius 4 in
SPE 59804 Evaluation of Acid Treatments in Horizontal Wells 5
L
w
x
y
z
0 L
xd
L
xl
h
x
h
s
h
z
Fig. 1 Schematic of semi-infinite slab model for horizontal well.
0
0.5
1
1.5
2
2.5
3
12:51 13:11 13:31 13:51 14:11 14:31 14:51 15:11 15:31 15:51
time
r
a
t
e
,

b
p
m
0
100
200
300
400
500
600
700
800
900
p
r
e
s
s
u
r
e
,

p
s
i
injection rate
surface pressure
Fig. 2 Injection rate and surface pressure for example treatment.
-1
0
1
12:00 13:12 14:24 15:36 16:48
time
s
k
i
n

f
a
c
t
o
r
Fig. 3 Skin evolution during acidizing for example treatment.