Affiliates 2006 Hill

Download as pdf or txt
Download as pdf or txt
You are on page 1of 39

Acid Transport and Reaction in

Heterogeneous Media
Dr. A. D. Hill
Robert L. Whiting Chair in
Petroleum Engineering
Texas A&M University

CSM Meeting Oct. 17, 2006

Acid Stimulation
Matrix acidizing (below frac pressure)
Inject relatively small volumes of acid
Objective is to overcome damage effects
Acid only affects the formation for a few
inches to a few feet

Acid Fracturing
Create a hydraulically induced fracture,
just like proppant fracturing
Conductivity is created by acid etching the
faces of the fracture in a non-uniform way

CSM Meeting Oct. 17, 2006

Acid Stimulation
Matrix acidizing (below frac pressure)
Inject relatively small volumes of acid
Objective is to overcome damage effects
Acid only affects the formation for a few
inches to a few feet

Acid Fracturing
Create a hydraulically induced fracture,
just like proppant fracturing
Conductivity is created by acid etching the
faces of the fracture in a non-uniform way

CSM Meeting Oct. 17, 2006

Acid Stimulation Models


Matrix acidizing Do models
consider small scale
heterogeneity?
Only at large scales

Acid Fracturing - Do models


consider heterogeneity?
No

Is it important?
Yes at all scales
CSM Meeting Oct. 17, 2006

Modeling Acid Transport and Resulting Rock


Dissolution in Heterogeneous Media

Matrix stimulation of
sandstones
Creation of conductivity in
acid fractures

CSM Meeting Oct. 17, 2006

Matrix Acidizing of
Sandstone
Complex chemical interactions between HF/HCl and
many minerals
Acid typically penetrates only a few inches into the
formation
Propagation of acid through the damaged zone is
critical to success
Models generally treat the formation as 2
homogeneous regions in series damaged zone and
undamaged region
Small scale heterogeneity can change acid
penetration by an order of magnitude

CSM Meeting Oct. 17, 2006

Matrix Acidizing of Sandstone


Geometry of the standard acidizing model

CSM Meeting Oct. 17, 2006

Property Initialization

Sandstone core with small-scale heterogeneity


Completely Random Porosity
Sample
(Mean porosity: 0.18, SD : 0.03)

CSM Meeting Oct. 17, 2006

Property Initialization

Sandstone core with small-scale heterogeneity


Correlated Porosity Distribution
( (inches): 20 in x, 0.05 in y, 1 in z)

CSM Meeting Oct. 17, 2006

Basic Case

HF Front Propagation
(Homogeneous porosity and mineral)

A: 5 PV

PV

C: 25 PV

CSM Meeting Oct. 17, 2006

B: 15 PV

D: 35 PV

Basic Case

Distribution During the


Acid Flooding
(Homogeneous porosity and mineral)

A: 5 PV

PV

C: 25 PV

CSM Meeting Oct. 17, 2006

B: 15 PV

D: 35 PV

Heterogeneous Porosity

Sandstone core with small-scale heterogeneity

Distribution ( >= 0.02)


Random Initial Porosity

A: 5 PV

C: 25 PV

CSM Meeting Oct. 17, 2006

B: 15 PV

D: 35 PV

Heterogeneous Porosity

Sandstone
core
small-scale heterogeneity
Comparison
ofwith
Breakthrough

Time (PVbt) in
Cores with Different
Heterogeneities
35

break through pore volume

33
31
29
27
25
23
21
19
17
15
0

0.01

0.02

0.03

0.04

0.05

standard deviation of porosity distribution

CSM Meeting Oct. 17, 2006

0.06

0.07

0.08

Correlated Porosity

Sandstone core with small-scale heterogeneity

Distribution ( >= 0.02) Correlated


distribution

A: 5 PV

B: 15 PV

C: 25 PV

D: 35 PV

CSM Meeting Oct. 17, 2006

Correlated Porosity

Sandstone core with small-scale heterogeneity

Breakthrough Time in Cores with Different


Correlation Lengths
4

3.5

PVbt

2.5

1.5
1

10
correlation scale parameter in axial direction (inch)

CSM Meeting Oct. 17, 2006

100

Matrix Acidizing of
Sandstone
Small-scale heterogeneities allow acid to
penetrate deeper into the formation than
predicted by homogeneous models
Sandstones with correlated permeability in
the main flow direction (laminated systems)
show the greatest effect acid penetrates
ten times as fast
Heterogeneity results in better stimulation
results than predicted by homogeneous
models

CSM Meeting Oct. 17, 2006

Creation of Conductivity in
Acid Fracturing
After hydraulic pressure is released during a
fracture treatment, the fracture will close
under the force of the confining stresses
Acid creates conductivity through
differential etching the acid must remove
rock in an uneven manner to create lasting
conductivity
Paradox Acid creates conductivity by
heterogeneously removing rock; acid frac
models simulate a homogeneous process

CSM Meeting Oct. 17, 2006

Creation of Conductivity in Acid Fracturing


Current acid fracture modeling approach

2D fracture geometry
Homogeneous formation no
variations in rock properties,
stress state, leakoff conditions
Local conductivity based on total
amount of rock dissolved and
empirical correlation (NierodeKruk correlation)
No effect of rock dissolution on
fracture mechanics
CSM Meeting Oct. 17, 2006

Scaling Up Local
Conductivity - Background
Laboratory conductivity
measurements are for samples a
few inches in size
Fracture simulators have grid
blocks 5 30 feet in size
Channels that create conductivity
are intermediate in size, are
completely missed if we directly
apply lab-scale models
CSM Meeting Oct. 17, 2006

Intermediate scale
corelation used in
frac model

Intermediate
Scale Model:
Simulation
studies to
develop new
correlation

Acid fracture
model

Lab-scale
Conductivity
correlations

Experimental
study

Conductivity Model Development


CSM Meeting Oct. 17, 2006

Creation of Conductivity in Acid Fracturing


Intermediate Scale Model - Approach

Simulate acid transport and


dissolution in a fracture domain
the size of a frac model grid
block
From extensive simulation studies,
derive a correlation of fracture
conductivity appropriate for use
in a frac model

CSM Meeting Oct. 17, 2006

Intermediate Scale Model

The task:
3D Navier-Stokes (NS)
Acid transport
Acid/rock reaction->acid etching profile

Complications
Fracture domain is irregular
Domain is growing with time

Method:
Front fixing method (body-fitted
coordinate transformation)
CSM Meeting Oct. 17, 2006

Problem Description

z
y

x
y

CSM Meeting Oct. 17, 2006

Mathematical Model
Assumptions
Incompressible
Consecutive steady state (NS calculation)
Newtonian
Gravity is neglected

CSM Meeting Oct. 17, 2006

Mathematical Model
NS

v
u = 0
v
v
2v
(u )u = p + [ u ]

Acid transport
c
c
+ u.c =
De
t
y y
Fracture width
y0
c

vl c Deff

=
t
y y =0
(1 )

CSM Meeting Oct. 17, 2006

y1

c
vl c Deff

=
t
(1 )
y y = w

Coordinate Transformation
Boundaries and of the mesh points x y z moving ->
fixed points at each time step

y
CSM Meeting Oct. 17, 2006

Coordinate Transformation
Boundaries and of the mesh points x y z moving ->
fixed points at each time step

CSM Meeting Oct. 17, 2006

Front Fixing Method


Front fixing method
- The irregular moving boundary-> a regular fixed
boundary (body fitted coordinate transformation)
- The transformed PDE is complex (i.e contains
mixed derivatives, boundary condition)

CSM Meeting Oct. 17, 2006

Coordinate Transformation
Irregular moving boundary (physical) -> a
regular fixed boundary (calculation)
x
L
y y0
=
y1 y 0

z
=
H
=t

0 1
(b( x, z , t ) = y1 y 0 )
0 1

y0

y1
CSM Meeting Oct. 17, 2006

0 1

Slot Flow with Uniform Leakoff


Terrill, 1964

Y (inch)

-0.1

-0.1

10

15

20

X (feet)

CSM Meeting Oct. 17, 2006

25

30

35

Velocity Profile of Ux
Ux at x=10m
0.015
Analytical

Ux (m/s)

0.012

Numerical

0.009
0.006
0.003
0
-0.1

-0.05

0
Y (inch)

CSM Meeting Oct. 17, 2006

0.05

0.1

Velocity Profile of Uy
0.00002

Uy (m/s)

0.00001

0
-0.1

-0.05

0.05

-0.00001
Analytical
Numerical
-0.00002
Y (inch)

CSM Meeting Oct. 17, 2006

0.1

Velocity Field

CSM Meeting Oct. 17, 2006

Velocity Contour

CSM Meeting Oct. 17, 2006

Initializing a Fracture
1. Variables
Permeability and width

2. Characteristics
Not completely random, spatially correlated

3. Model
Semi-variogram model
FFT simulator (Jennings 2003): correlation scale parameters

CSM Meeting Oct. 17, 2006

Permeability Distribution
Z

30

k (md)
140
130
120
110
100
90
80
70
60
50
40
30
20

20

Z (feet)
10

0
10
-0.2

-0.1

CSM Meeting Oct. 17, 2006

20
0.1

Y (in
ch)

0.2

30

X (fe

et)

What if you are


injecting acid into
this?

CSM Meeting Oct. 17, 2006

What if you are


injecting acid into this?

CSM Meeting Oct. 17, 2006

The Effect Of Heterogeneity on


Acid Stimulation
Acknowledgements:
Contributors to this work include
Dr. Ding Zhu
Dr. Chun-lo Li
Dr. Cheng-Li Dong
Tao Xi
Maysam Pournik
Jianye Mou
Maria Melendez
and many other former students

CSM Meeting Oct. 17, 2006

You might also like