PETROLEUM ENGI NEERI NG

Numerical and Laboratory Study of Gas Flow

through Unconventional Reservoir Rocks
RPSEA Piceance Basin Tight Gas Research Review
Xiaolong Yin, Assistant Professor
Petroleum Engineering, Colorado School of Mines
April 21 2011
PETROLEUM ENGI NEERI NG
Presentation Outline
Challenges and our research activities
Our RPSEA project and collaborators
Preliminary results
Goals and objectives
RPSEA Piceance Basin Research Review
April 21, 2011, Denver, Colorado
PETROLEUM ENGI NEERI NG
Background
In very tight reservoir rocks, both measurement techniques
and our understanding of the physics are being challenged
Our groups research activities focus on pore-scale physics
and flowusing direct simulation and experiments
Specifically, we investigate what makes unconventional rocks
unconventional
Surface interaction
Non-continuum slippage
Heterogeneity in pore structure and rock-fluid interactions
RPSEA Piceance Basin Research Review
April 21, 2011, Denver, Colorado
PETROLEUM ENGI NEERI NG
Our RPSEA Project
RPSEA 09122-29 02/2011 02/2014
Use nanofluidic chips and single-molecule detection techniques to
visualize fluid flow in nano-sized pores
Combine core flooding test and SEM imaging to correlate fluid flow in
tight rocks to pore structures
Develop pore-scale numerical models to provide information that
cannot be easily obtained from experiments, such as three-
dimensional motion of fluids
Our Team
Missouri University of Science & Technology B. Bai (core flooding),
Y. Ma (single-molecule detection)
Colorado School of Mines X. Yin (pore-scale models), K. Neeves
(nanofluidic chips)
RPSEA Piceance Basin Research Review
April 21, 2011, Denver, Colorado
PETROLEUM ENGI NEERI NG
Traditional laboratory studies
Pulse-decay permeability measurement
Mercury porosimetry
Linear core flooding
Ultra-centrifuge
PVT (CSM)
These equipments allow us to study
Porosity and permeability
Storage capacity and transport
Multiphase flows and formation damage
A CMS-300 Pulse-Decay Permeameter
RPSEA Piceance Basin Research Review
April 21, 2011, Denver, Colorado
PETROLEUM ENGI NEERI NG
Non-traditional laboratory studies
What makes flow in unconventional reservoir rocks
unconventional?
Surface interactions
Non-continuum slippage
Heterogeneity in surface
properties
Fractures and cracks
In-situ stress

Some of these effects
can be studied using
micro (right) and nano-
scale (below) porous
media analogs
constructed on silicon /
polymer chips
Photo courtesy
of Keith Neeves
Chemical
Engineering
Colorado School
of Mines
This list is probably far
from complete.
RPSEA Piceance Basin Research Review
April 21, 2011, Denver, Colorado
PETROLEUM ENGI NEERI NG
Preliminary results from micro-chip experiments
Data and photo provided by
Keith Neeves and Melissa Wu ,
Chemical Engineering, Colorado
School of Mines
RPSEA Piceance Basin Research Review
April 21, 2011, Denver, Colorado
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
1 2 3 4 5
P
e
r
m
e
a
b
i
l
i
t
y

(
D
a
r
c
y
)
Experiment LB Simulation
Permeability comparison
Air-water two phase
flow test: Left: The
geometry is initially
saturated with
water
Below: Air is
injected in the
lower left corner
forcing water out of
the channels
PETROLEUM ENGI NEERI NG
Simulation of micro- and nano-scale flows
Numerical tools have been and are being developed to study fluid flow
with non-continuum effects in nano-sized pores
Representative porous media geometry models
Experimentally
measured and k
and pore structure
Direct numerical simulation models:
Lattice-Boltzmann (for Navier-Stokes)
DSMC (for non-continuum flows being
developed)
Pore structure
Relative permeability
Effect of stress
Adsorption / Desorption
RPSEA Piceance Basin Research Review
April 21, 2011, Denver, Colorado
PETROLEUM ENGI NEERI NG
Preliminary results from pore-scale modeling
Porosity-permeability relation is the key to rock typing and
understanding geomechanical effects on fluid flow
0.0000
0.0005
0.0010
0.0015
0.0020
0.0025
0.0030
0.00 0.10 0.20 0.30 0.40
ks
2
Porosity
Homogeneous Geometry
Heterogeneous Geometry
Ideal, channelized
geometry porosity-
permeability shows a
strong correlation
More realistic,
heterogeneous geometry
porosity-permeability
shows large scattering
RPSEA Piceance Basin Research Review
April 21, 2011, Denver, Colorado
PETROLEUM ENGI NEERI NG
Preliminary results from pore-scale modeling
A universal porosity-permeability correlation can be
developed by recognizing that the large pores do not
contribute to permeability
0.0000
0.0005
0.0010
0.0015
0.0020
0.0025
0.0030
0.00 0.10 0.20 0.30 0.40
ks
2
Conductive Porosity
Homogeneous Geometry
Heterogeneous Geometry
These data are from our scoping
studies using 2D geometries
3D simulations and experiments are
underway
Such a correlation can be used to
determine the geometry of pores
from bulk measurement without
resorting to image analysis
RPSEA Piceance Basin Research Review
April 21, 2011, Denver, Colorado
PETROLEUM ENGI NEERI NG
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35
0
0.005
0.01
0.015
0.02
0.025
Porosity
k
a
v 2
Parallelized, 3D pore-scale simulator
3D Representative
Geometries
3D Porosity-Permeability Data
3D Simulator Parallel Speedup
10
0
10
1
10
2
10
3
10
-3
10
-2
10
-1
10
0
10
1
Number of Cores
S
i
m
u
l
a
t
i
o
n

t
i
m
e
-
1

(
s
-
1
)
Slope = 1
RPSEA Piceance Basin Research Review
April 21, 2011, Denver, Colorado
PETROLEUM ENGI NEERI NG
Goals and objectives
Simulations with
controlled
geometries and
assumed physics
Micro- and nano-
fluidic experiments
with controlled
geometries and real
physics
Core-scale
experiments with
real geometries and
real physics
Increasing level of
reality
The combined approach will
significantly improve our
understanding of fluid flow on
nano-scale and unconventional
reservoir dynamics
RPSEA Piceance Basin Research Review
April 21, 2011, Denver, Colorado
PETROLEUM ENGI NEERI NG
Acknowledgements
Collaborators
Baojun Bai, Yinfa Ma (MUST)
Keith Neeves (CSM, ChemE)
Qinjun Kang (Los Alamos National Lab)
Students
Feng Xiao (PE)
Lei Wang (PE)
Melissa Wu (ChemE)
Funding
RPSEA
PETROLEUM ENGI NEERI NG
Simulated vs. measured permeability
Simulation of flow through digital cores from CT-scan
CT and experimental data from Imperial College
Numerical simulations are done in CSM
A1 BSS C1 F42A S1 S2
Porosity 42.9 19.6 23.3 33 14.1 24.6
Resolution 3.85 5.345 2.85 9.996 8.683 4.956
K (exp) 7,220 1,286 1,102 59,000 1,678 3,898
K (sim) 8,675 1,507 1,192 59,331 2,006 4,076
% error 20.2% 17.2% 8.1% 0.6% 19.6% 4.6%