GCEP Progress Report

“Collaborative Research on Carbon Sequestration

in Saline Aquifers in China”

Investigators
Dongxiao Zhang, Professor, Hydrogeology and Petroleum Engineering; Kristian Jessen,
Assistant Professor, Peteoleum Engineering; Cheng Chen, Post-doctoral Researcher;
Dalad Nattwongasem, Graduate Researcher, University of Southern California;

Bin Gong, Research Scientist, Energy & Resources Engineering; Qingdong Cai,
Associate Professor, Mechanics; Yi Zheng, Research Scientist, Energy & Resources
Engineering; Da Huo, Xuan Liu, Ming Lei, Zhijie Wei, Xiang Li, Lingyu Zhang, Yongbin
Zhang, Graduate Researchers, Peking University;

Yilian Li, Professor, Geochemistry; Yanxin Wang, Hydrogeology and Geochemistry;

Jianmei Cheng, Hydrogeology; Wei Zhang, Ling Jiang, Gengbiao Qiu, Qi Fang, Yibing
Ke, Peng Cheng, Sanxi Peng, Ronghua Wu, Jianxiong Dong, Anne Nyatichi Omambia,
and Kabera Telesphore, Graduate Researchers, China University of Geosciences.

Abstract
Injection of CO2 into saline formations represents a sequestration option of large
potential capacity in China. As a leading research, this work represents a unique
international collaborative effort by professors and researchers from Peking University,
China University of Geosciences at Wuhan, and University of Southern California to
address fundamental issues associated with large-scale sequestration of CO2 in saline
formations with emphasis on developing the potential for CO2 sequestration projects in
China.

The collaborative research aims to address 1) a comprehensive review of

sedimentary basins in China in terms of basin characteristics and its proximity to CO2
point sources; 2) experimental study to investigate the dynamic behavior of CO2
migration in the context of storage in saline aquifers; 3) modeling and simulation for
microscopic interactions and macroscopic long-term fate of injected CO2 versus the host
environment.

For sedimentary basin study, a regional study on potential CO2 sequestration in the
Jianghan Basin was performed. The two formations, Qianjiang and Xingouzui
Formation, were firstly identified and investigated with respect to the lateral extend and
quality of reservoirs, the geological and hydrogeological properties of the formation
including the depth, thickness, porosity, permeability and hydrochemical conditions.
Preliminary analysis indicated that the two formations could be suitable candidate sites
for CO2 injection.

For dynamic experimental study, we are currently investigating the dynamics of

countercurrent flows in well-defined porous materials (glass bead packs) using an analog
isooctane/brine fluid system. The experimental work utilizes dynamic resistivity
measurements to monitor the evolution of an isooctane plume initially located at the
bottom of the column. Time estimates for residual entrapment via scaling analysis for
counter-current flow and design of segregation experiments based on an analog fluid
system in glass-bead pack were studied.

For microscopic modeling and simulation, a coupled multiphase Lattice Boltzmann

(LB) model was developed to simulate the dissolution of immiscible liquid droplets in
another liquid during the rising process resulting from buoyancy. It was found that there
exists a terminal rise velocity for each droplet, and there was a power law relationship
between the Eötvös (Eo) number and the terminal Reynolds (Re) number. The
simulation results were in agreement with the empirical correlation derived for predicting
bubble rise. The Damkohler (Da) and Peclet (Pe) numbers were varied to investigate
the coupling between droplet size, flow field, dissolution at the interface, and solute
transport.

For macroscopic modeling and simulation study, a CO2 sequestration simulation

framework that can account for natural and drilling/injection induced fractures and
chemical reactions was developed. This framework is capable of modeling dynamic
flow in micro-scale and reservoir-scale simultaneously. Simulations have been done
using Discrete Fracture Modeling (DFM). Systems with fractures in CO2 injection
formation versus in cap rocks are compared. Results have shown that the existence of
mudstone layers could prevent injected CO2 from leaking outside the reservoir when no
fractures are present. While vertical fractures intersecting with mudstone layers will
cause significant leakage increase as fractures form extremely preferential pathways for
CO2 transport.

Introduction
Injection of CO2 into saline formations represents a sequestration option of large
potential capacity in China. However, no systematic investigation of this potential
including specific site selection and characterization has yet been performed. In these
formations, the main CO2 sequestration mechanisms are structural trapping (due to
low-permeability traps), capillary trapping due to interfacial phenomena, solubility
trapping (CO2 dissolution into brine), and mineral trapping (reactions with minerals to
form new, permanent mineral products). Gravity segregation, viscous fingering,
solubility, reaction kinetics, and possible leakage through natural/artificial pathways are
some of the factors that may significantly affect the sequestration capacity as well as the
long-term fate and redistribution of the injected CO2. These factors must all be included
in a scientifically sound assessment of large-scale storage potentials and long term fate of
any injected CO2.

This collaborative research integrates geological modeling, reservoir simulation and

laboratory experiments to identify the appropriate scientific approach for quantifying the
storage capacity in these highly complex geological settings.

In Geology and hydrogeology, we aim to investigate the geology, stratigraphy and

hydrogeology of the Jianghan Basin in detail to identify combinations of strata that may
provide suitable reservoir and seal pairs with respect to the CO2 geosequestration
potential and further quantify the storage capacity by integrating geological modeling,
reservoir simulation and laboratory experiments. The Jianghan Basin is a representative
salt-lake rift basin with the Cretaceous and Quaternary sedimentary deposits up to 10km
in depth. CO2 emissions in the region directly overlying the Jianghan-Nanyang Basins
contributed approximately 116 Mt CO2 per year to China’s estimated total stationary
source CO2 emissions of 2970 Mt/yr. In a 300 km radius from the two basins there are
sources totaling approximately 604 Mt/yr and there are four power plants and a steel mill
releasing about 25 Mt of CO2 per year in the vicinity of Wuhan city. The gypsum-salt
rock and mudstone are well developed in the basin, and hydrocarbon resources are in
great abundance as the well-known Jianghan Oilfield is located in the central of the basin.

Along our research in dynamic experiments, we focus on the dynamics of CO2 plume
migration after the injection phase has ended. It is well established (e.g. Zhou et al.
2002) that the relative permeability behavior observed in co-current immiscible
displacement experiments (e.g. water displacing oil) differs significantly from the
relative permeability behavior observed in counter-current flow experiments such as
counter-current imbibition processes of relevance to oil production from fractured
reservoirs. To date, however, all published efforts towards understanding the time
scales associated with immobilization of CO2 by capillary entrapment in brine formations
(e.g. Kumar et al. 2005; Ide et al. 2007) have utilized existing simulation technology
that does not distinguish between co-current and counter-current in the representation of
relative permeability.
 To improve the fundamental understanding of the time scales associated with residual
entrapment of CO2 in aquifers and to strengthen the state-of-the-art simulation
technology by honoring different flow regimes, we combine experimental investigations
of gravity segregation processes with analytical and numerical modeling efforts. Time
estimates for residual entrapment via scaling analysis for counter-current flow and design
of segregation experiments based on an analog fluid system in a well-define porous
medium (glass-bead pack) were studied.

In modeling and simulation, of the collaborative research project, during the first year
we investigated the motion and dissolution of rising CO2 droplet/bubble during
geological carbon sequestration in saline aquifers. During carbon sequestration, the
upward motion of supercritical carbon dioxide (scCO2) due to buoyancy leads to brine
flowing around it, which accelerates the dissolution of scCO2. In addition, the dissolved
scCO2 slightly increases the density of brine, which might give rise to downward
density-driven flow at the large temporal and spatial scales. This process also favors the
dissolution of scCO2. The dissolution of scCO2 in water is an important sequestration
mechanism and is usually referred to as solubility trapping.

For microscopic simulation, we proposed a coupled LB model that takes into account
two-phase flow, dissolution at the interface, and solute transport during the rising process
of immiscible liquid droplets. The two-phase LB model accounting for droplet motion
is based on the code of Kang et al (2002a). The dissolution simulator accounts for the
dissolution taking place at the liquid-liquid interface, based on the method of Kang et al.
(2002b, 2003). The solute transport simulator describes the transport of the dissolved
mass (Dawson et al., 1993) and is based on the single-phase LB model of Chen et al.
(2008). The rise and dissolution of the same amount of dispersed phase were simulated,
when it was in a single, two, and three droplets. By means of simultaneous numerical
simulations of the multiple processes, we are able to better understand the complicated
interplay and coupling between droplet number, size, rise velocity, dissolution, and solute
transport. Specifically, this study aims to describe the rise of one or more scCO2
droplets in brine, with mass transfer across the interface due to dissolution.
Understanding of this coupled process is important for geological carbon sequestration.

Numerical simulation is the key approach to study the long-term fate of injected CO2
and potential impact of escape of CO2 due to the failure of natural barriers or migration
along natural/anthropogenic escape paths such as faults or wells. As we view the static
modeling of saline aquifer would be following the well established industry standard
using commercial tools such as GOCAD and Petrel, the focus of this research is put on
designing the simulation framework, code development and its implementation to
China’s candidate formations.
Background
1. Geological Indicators for Storage Site Suitability
The regional potential of CO2 geological storage can be evaluated by five key factors
such as storage capacity, injectivity potential, site details, containment and existing
natural resources. Key geological indicators for storage site suitability are shown in
Table I (EU Geocapacity, 2009).

Table I: Key geological indicators for storage site suitability.

Influential Criteria to investigate in the screening process
Basic,
geological and
geological
physical Positive indicators Cautionary indicators
related criteria
parameters
Pressure and Depth of crest of Depth of crest of
Sufficient depth Temperature reservoir >1000m reservoir <800m
of reservoir Porosity and Depth of base of Depth of base of
Permeability reservoir <2500m reservoir >2500m
Petrophysic Porosity >20% <10%
reservoir
Permeability >300mD <200mD
properties
Lithology Low permeable
(Porosity, lithologies, such as
Permeability) clay
Integrity of seal Thickness >100m <20m
Faults Unfaulted Faulted
Heterogeneity Homegenous Heterogeneous
Tectonic activity No tectonic activity Tectonic activity
Total capacity of
Total capacity of
reservoir estimated to
reservoir estimated to
be similar to less than
Reservoir: be much larger than the
the total amount
total amount produced
produced from CO2
from CO2 source
source
Storage Thickness >50m <20m
capacity
Areal extent Well defined Not well defined
Heterogeneity
Faults Unfaulted Faulted
Trap type Well defined structure Not well defined
Petrophysic
Values given above Value given above
prop
2. Numerical Modeling for Carbon Dioxide Storage
For the study of long-term behavior on carbon dioxide sequestration, which occurs in
a long geohistory time, the numerical method has certain advantages than the
experimental method, but because the sequestration of carbon dioxide subsurface is a
complex physics-chemical problem which includes:
i. The co-existence of supercritical carbon dioxide and subsurface flow (water, oil or
else). The transport of the carbon dioxide and the interaction between the two
fluids is in the scope of multiphase flow.
ii. During the transportation, chemical reaction happened between carbon dioxide
and subsurface strata.
iii. Ions, the production of the chemical reaction, diffuse and advect in the multiphase
flow.
iv. The special structures, for example the fractures or karsts, makes it easy for
transportation of carbon dioxide. Meanwhile, the chemical reaction enlarge the
fractures (sometimes plug). The leakage of carbon dioxide began while the
fracture development to certain extent.
v. Multi-scale exists in the system. The exquisite description of the problem
expands from porous scale to field scale.
vi. The system is thermal.

Therefore, the corresponding numerical simulators should be able to address these

issues. Recent numerical modeling for CO2 storage applications can be split into three
categories: (i) hydrodynamic modeling simulating structural, residual gas and dissolution
trapping processes, (ii) geochemical modeling simulating batch geochemical reactivity
(closed system without any fluid flow) and (iii) reactive transport modeling combining
the two previous types of simulations and therefore providing a complete calculation over
time of the amount of CO2 trapped through structural, dissolution or mineral trapping.
Reactive transport modeling is the most realistic modeling technique to quantify the
long-term fate of CO2 as well as other aspects during geological storage, but also the
most challenging to perform (Gaus et al, 2008).

Until now, several simulators for geologic CO2 sequestration are being developed by
some research groups. In this part, we generally summarize three major simulators,
ECLIPSE CO2STORE module, TOUGHREACT ECO2N module, and GPRS CO2
modeling part. For comparison of the three simulators, see Table II.

Table II: Comparison of CO2STORE, ECO2N and GPRS

CO2STORE ECO2N
GPRS
(ECLIPSE) (TOUGHREACT)

Compositional model 3 3 3
 Fully implicit solver 3 3 3

Oil phase inclusion ¯ ¯ 3

CO2 dissolve in water phase 3 3 3

H2O dissolve in gas phase 3 3 3

Molecular diffusion 3 3 3

physical dispersion ¯ ¯ 3

Hysteresis effect 3 3 3

Non-isothermal ¯ 3 ¯

Parallelize 3 3 3

Chemical reaction Bad Good Normal

Computational speed Fast Slow Fastest

Expansibility Bad Normal Good

Redevelop No Yes Yes

3. Discrete Fracture Modeling (DFM)

Reservoirs with fractures may cause leakage problems which will be detrimental to
the environment and safety to human race. It is found in Algerian In Salah CCS project
that CO2 prefers to go through fractures than matrix. Research on CO2 flow in fracture
reservoirs is urgent as well as significant. In the work of Nelson et al. (2005), natural
and artificially-induced hydraulic fractures have been identified as the most important
factors in risk assessment of CO2 storage, and regarded as potential escape routes for CO2,
which could damage the prospective storage ability of a specific storage site. Natural
fractures could potentially allow CO2 to quickly migrate through the cap rock to the
surface, while local pressure increase caused by CO2 injection can also lead to hydro
fracturing in the vicinity of wells.

Discrete Fracture Modeling (DFM) begins to be used more and more in reservoir
simulation because of isolated and disconnected fractures. In DFM, the fracture plays
as a standalone grid block but a secondary factor in the matrix-fracture system, which
allows us to separately deal with matrix and fracture properties and flows occurring in
them. DFM can also be used in combination with dual porosity model. DFM models
have proven to be useful for measuring leakage risks associated with hydraulic fracturing
and coal bed methane production (Pashin, Jin, and Payton, 2004). It is for this reason
that DFM is expected to become an effective modeling approach for measuring the risks
associated with carbon sequestration.

Results
1. Jianghan Basin Evaluation
The Jianghan Basin lies in the south-central part of the Hubei Province, south China,
the Jianghan Plain between Yangtzi River and Han River, west to Yichang city, east to
Yingcheng city, south to Huhong city and north to the north of Jingzhou city covering an
area of 36350km2. CO2 emissions in the region directly overlying the
Jianghan-Nanyang Basins contributed approximately 116 Mt CO2 per year to China’s
estimated total stationary source CO2 emissions of 2970 Mt/yr.

Figure 1: Location map of the Jianghan Basin (Li et al., 2002)

Major point sources in the vicinity of Wuhan city amount to approximately 25 Mt of

generation (including four power plants and a steel mill) corresponding to approximately
25 Mt of CO2 per year.
 The Jianghan Basin is a representative salt lake rift basin developed on the
Zhongyangzi Paraplatform during the Cretaceous and Tertiary Period covering an area of
36350 km2. The tectonic structure location is in the central part of the Zhongyangzi
Depression in Yangzi Paraplatform, west to Western Hubei fold belt, east to
Wuhan-Tongshan fold belt, north to Bahong fold belt and south to Jianghan fold belt (Pan
and Zhu, 1988). The Early Cretaceous Basement of the basin is a double-deck structure
consisting of Proterozoic metamorphic rocks in the lower part and sedmentary rocks
deposited from Sinian to Mesozoic Jurassic in the in the upper part acting as the caprock
of the Zhongyangzi Depression of Yangzi Paraplatform. The development process of
the Jianghan Basin contains seven different phases controlled by tectonic movements
which can be grouped into two tectonic cycles characterized by tension
crack-rift-depression. In addition, together with the controls of the northeast and
northwest basement faults, the basin forms eight- depression and four-uplift (or low uplift)
tectonic pattern as shown in Table III (Li et al., 2002).

Table III: The secondary tectonic units of the Jianghan basin

Sedimentary
Category Name Area(km2) Characteristics
thickness (m)
Jiangling depression 7400 6500 Depression sag
Chentuokou Single fault type
1400 4500
depression depression
Qiangjiang Double fault type
2400 8500
depression depression
Mianyang
3400 4000 Depression sag
depression
Single fault type
Xiaoban depression 500 6000
depression
Depression Zhijiang depression 3300 2000 Depression sag
Single fault type
Yunying depression 4500 --
depression
Herong depression 1600 -- Depression sag
Single fault type
Jinmeng depression 3200 --
depression
Single fault type
Hanshui depression 3600 3000
depression
Double fault type
Yuanan depression 600 --
depression
Tonghaikou uplift 1000 -- Uplift
Longsaihu lower
2300 -- Lower uplift
uplift
Uplift
Yuekou lower uplift 300 -- Lower uplift
Yajiao-Xingou
500 -- Lower uplift
lower uplift
Figure 2: Tectonic map of the Jianghan Basin (Pan and Zhu, 1988
Figure 3: Interpretation of the profile lines in Northeast and Northwest in the Jianghan
Basin (Yang et al.2009)
 The Jianghan Basin contains Late Cretaceous to Quaternary sedimentary deposits up
to 10km thick, which are grouped into two predominant sedimentary cycles. The first
sedimentary cycle contains the sediments from the Upper Cretaceous to Xingouzui
formation of Lower Eocene in Paleogene period. Upper Cretaceous is composed of red
terrestrial clastic rocks formation with the thickness ranging from 500 to 2000m. Shashi
Formation of Paleocene in Paleogene period consists of gypsum-salt and red mudstone
formation with the thickness varying from 200 to 900 meters. Xingouzui Formation of
low Eocene is composed of sandstone and mudstone with interbedded gypsum salt,
which is a petroliferous formation with the thickness ranging from 600 to 1900 meters.
The second sedimentary cycle consists of the sedimentary deposits from Jingsha
Formation of middle Eocene in Paleogene period to the Jinghezhen Formation of
Oligocene. Jingsha Formation is composed of red terrestrial clastic rocks formation
with the thickness ranging from 600 to 1870 meters. Qianjiang Formation is a red
terrestrial clastic rocks formation with the thickness varying from 750 to 5000 meters.
Jinghezhen Formation deposits the grey-green and green-grey siltstone. The gypsum
salt is widely distributed extending from the Jiangling Depression to Mianyang
Depression with the area up to 9000 km2 in the first sedimentary cycle. However, the
salt rock deposits very well in the second sedimentary cycle, with the accumulative
thickness up to 1800 meters, and is widely distributed in the Qianjiang, Xiaoban and
Yunying Depression. Figure 4 and 5 give a generalised stratigraphy and distribution of
Jianghan Basin.

Figure 4: Stratigraphy from Triassic to Paleocene of the Jianghan Basin (Pan and Zhu,
1988)
 Figure 5: Generalised stratigraphy of the Jianghan Basin (Li et al., 2002)

The sandstones of Haiyang formation of Cretaceous; the sandstones and siltstones of

Jinghenzhen Formation, Qianjiang Formation, Jingsha Formation, Xingouzui Formation
and Shashi Formation of Paleogene; the sandy conglomerate of Guanghuasi Formation of
Neocene; and the surfical sediments of Pingyuan Formation of Quaternary form the main
aquifers in the Jianghan Basin. To be pointed out, the shallow reservoir bed is
composed of the sandstones and gravel bed of the Quaternary Pingyuan Formation and
the middle and upper part of the Neocene Guanghuasi Formation, which usually has a
close connection with the surface water since the runoff condition of the surface is quite
poor due to the low ground and accumulated rivers and lakes. However, in the north
area of the Qianjiang Depression, the lower and middle Guanghuashi Formation can act
as the regional confining bed since it contains varicolored clay layer and minor
calcareous mudstone with the thickness of 100m. And the logging information shows
that there is a great distinction between of the freshwater and saline water between the
upper and lower of the Guanghuasi Formation. The groundwater in most of the region
of Guanghuasi Formation is in free water exchange zone. Below the plane of
unconformity between Neocene and Paleogene, however, the groundwater is usually
brackish water and saline water. The salinity both rises in vertical from above to below
and in horizon from the edge to center of the depressions. Whatever in vertical or in
horizon, the hydrogeology of the reservoirs both present closed to semi-closed states.

The Cretaceous aquifer and the Paleogene aquifer can be divided into three formation
water belt in depth, namely alternate belt, alternate and stagnant belt, and alternate and
stasis belt. In the alternate belt, as the burial depth is less than 1100m, the groundwater
is influenced by the surface water and mainly is sulfate-sodium type. The salinity is less
than 200000mg/L. In the alternate and stagnant belt, the groundwater is mainly
sulfate-sodium type as the burial depth between 1100 to 2000m where is the mixed zone
of surface and sedimentary formation water. The salinity varies from 2000000 mg/L to
340000 mg/L. In the alternate and stasis belt, the burial depth is deeper than 2000m,
and the water type is mainly magnesium chloride, and then sodium sulfate. The
distribution of salinity of reservoir water is concentrated, varying from 100000 mg/L to
340000 mg/L. Water chemical symbol shows that the belt is nearly not influenced by
the infiltration of the surface water. Table IV provides the data on the salinity of the
information water in Paleogene of the Jianghan Basin.

Table IV: Salinity of the formation water in Paleogene of the Jianghan Basin
Paleosalinity %
Formation Number of samples
Lowest Highest Range Average
Jinghenzhen Fm 5 2.00 3.69 2.4~3.5 3.37
Qian 1 28 2.40 5.21 2.6~4.5 3.84
Qianjiang Qian 2 24 3.03 9.10 3.3~5.5 4.42
Fm Qian 3 58 0.86 8.01 3.0~4.7 3.85
Qian 4 32 0.96 7.01 1.8~3.6 3.04
Jingsha Fm 5 1.40 3.13 1.4~3.1 2.53
Xingouzui Fm 9 1.73 3.30 2.0~3.2 2.70

The reservoir rocks in the Jianghan Basin include sandstone, globulitic marl,
fractured mudstone and basalt, sandstone reservoir as the major. Xingouzui Formation
and Qianjiang Formation is the predominant petroliferous and saliniferous sandstone
reservoir as shown in Figure 5. The sediment supply is unidirectional and adequate,
mainly from the Jingmen, Hanshui and Dangyang palaeodrainage pattern in the north of
the basin during the Qianjiang and Xingouzui Formation deposits.

The main cap rocks in the basin involve mudstone and gypsum-slat bed. The lower
Xingouzui Formation contains two sets of the regional seals, one referring to mudstone
confining bed with 20~40 meters thick lying between Sandstone Ⅱ and Sandstone Ⅲ,
another referring to gypsum on the top of the low Xingouzui Formaiton with 5~20 meters
thick. What’s more, the upper Xingouzui Formation depositing red mudstone with
minor sandstone actually can act as the regional seal since its sediment thickness
surpasses 200 meters, and even up to 600 meters in some area. Table V provides a
simple generalized model of the reservoir and seal pairs of Xingouzui Formation in the
Jianghan Basin. Qianjiang Formation, however, contains four sets of the regional seals,
namely 1~ 6 rhythms in the Upper Qian Four, 4~ 8 rhythms in the Lower Qian Three, 11~
15 rhythms in the Qian Two and argillaceous gypsum bed on the top of Qian One. The
rhythm gaprock is composed of the grey mudstone, salt rock and argillaceous gypsum.
Table V gives a simple generalized model of the reservoir and seal pairs of Qianjiang
Formation in Jiangling Depression of the Jianghan Basin.

Table V: Reservoir and seal pairs of Xingouzui Formation in the Jianghan Basin

The sandstone in Qianjiang Formation is not well deposited because of the

complicated epalaeogeographic setting and palaeoclimate, poor hydrodynamic force and
great density of aqueous medium in saline lake, but concentrated distributes in Qianjiang
Depression, Jiangling Depression and Xiaoban Depression covering the area of the 6078
km2, accounting for 16.9 percent of the basin. Moreover, the thickness of the sandstone
is comparatively large, ranging from 30m to 300m, with the maximum thickness up to
649 m. The favorable facies belt lies in the north of the Qianjiang Depression with the
overlay area of 1630 km2. The petrophysical property of the sandstone is good. The
average porosity is 19.2% and the permeability is 325.8×10-3µm2. Figure 6 gives the
sandstone isopachs of Paleogene Qianjiang Formaion of in the Jianghan Basin.

The formation water in Qianjiang formation mainly is sulfate-sodium type. As the

depth of burial lower than 1250m, water salinity increases with depth; as the burial depth
deeper than 1250m, the water salinity vary from 250000 mg/L to 340000 mg/L, almost in
the saturation state; when the burial depth deeper than 1650m, the chemistry of formation
water appears the same features of sedimentary storage water. Table VI shows the
major species of formation groundwater in Qianjiang Formation from different regions.

Table VI: Reservoir and seal pairs of Qianjiang Formation in Jiangling Depression,
Jianghan Basin
 Figure 6: Sandstone isopachs of Qianjiang Formaion of in Jianghan Basin (Yang et
al.2009)

Xingouzui Formation distributes three delta sand bodies in the Mashan, Hougang and
Hanchuan region from west to east in the north of the basin and they connect together
into a reservoir development belt in horizon and get thin to the south. The sand body is
widely distributed with the area of 11000 km2, and the thickness of the sandstone is
generally 20 to 140 meters in range, with the maximum thickness up to 237 meters.
Siltstone is the main rock type, with the average porosity of 17% and the average
permeability of 98×10-3µm2. Figure 7 shows the sandstone isopachs of Lower Xingouzui
Formation in the Jianghan Basin. In Jiangling Depression, the major component of the
sandstone is feldspar with the quartz content of 50.7% to 66.1%. Porous cement acts as
the main cementation type containing low impurity less than 5%. The formation water
in Xingouzui Formation is mainly calcium chloride and sodium sulfate type. As the
burial depth deeper than 1000m, the water salinity appears little differences with the
increasing depth, mainly between 100000 mg/L to 260000 mg/L. Table VII shows the
major species of formation groundwater in Xingouzui Formation from different regions.
TableVII: Major Species of formation groundwater in Qianjiang Formation from different region
Na+ Ca2+ Mg2+ Cl- SO42- HCO3- I- Br- B- Li+ K+ Salinity Others
Region Relative density
(mg/l) mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l (mg/l) (mg/l)
H2S
Xijiakou 1.0986 56932 1566 491 88957 3594 659 152199
228
Yajiao 1.1101 66563 1608 444 102970 4640 722 11.6 392 53 56 481 177946
Haokou 1.1776 107385 661 104 162220 6093 656 10.4 461 161 57 2126 279934
Zhongshi 1.1774 111101 861 129 166853 8093 622 12.1 288 173 63 1647 289842
Gaochang 1.2054 123835 685 150 185836 8903 654 14.0 558 172 51 1883 322741
Guanghua 1.2078 126220 490 170 189423 8432 645 9.7 226 349 69 3898 329932
Wangchang 1.1812 111975 416 138 161482 17423 1048 6.9 272 208 57 1195 294221
Huangchang 1.2023 1.7507 403 23 171555 13565 830 10.3 268 149 53 919 305282
Zhanggang 1.1684 91592 277 53 125419 21045 1628 3.8 82 240100
Guangmingtai 1.1908 117772 1231 227 184800 3402 689 12.8 745 125 54 2425 311483
H2S
Tankou 1.1482 90825 515 163 136297 6666 418 8.2 196 136 61 673 235958
1098
Xiongkou 1.2167 130570 156 34 187400 19040 987 11.5 611 102 60 1090 340062

TableVIII: Major Species of formation water in Xingouzui Formation

Region Density Salinity Na+ Mg2+ Ca2+ Cl- SO42- HCO3-
3
g/cm mg/L
Xingou Oilfield 1.1134 166417 62675 401 991 91679 10500 168
Xingou Oilfield 1.1221 229616 86569 554 2426 137368 2563 133
Laoxin Region 1.0834 125721 46618 136 2304 75483 1020 157
Jiangling Region 1.1326 203932 71012 1240 6569 123552 1470 87
Shashi Region 1.1295 192563 69197 793 4759 116271 1399 143
 Figure 7: Sandstone Isopachs of Lower Xingouzui Formation in Jianghan Basin

The Jianghan Basin is a representative petroliferous basin bearing salt lake sedimentary
deposit with the maximum thickness up to 10 km deposited from Cretaceous to Paleogene,
characterized by the development of gypsum, mudstone and salt rock. The caprock includes
gypsum rock, salt rock, and mudstone, mudstone acting as the prominent caprock distributed in
each formation in vertical, deposited in the river alluvial plain, river delta and lacustrine
environment which is highly favourable for forming good seals. The parameters of mudstone
as caprock in different depositional environments are shown in Table IX.

Table IX: Parameters of mudstone as caprock in different depositional environments (Yin and Li,
2005)
Porosity Breakthrough Median Radius
Sedimentary facies
(%) pressure(MPa) (nm)
Interval value 1.31~5.41 0.131~10.2 3.98~25
Fluvial facies
Medium value 3.10 5.02 13.54
Interval value 0.97~13.95 4.18~13.29 2.42~10.64
Delta facies
Medium value 9.39 8.10 5.46
Lacustrine Interval value 3.97~14.35 2.94~14.46 1.97~10.16
facies Medium value 8.35 9.40 5.76

2. Experimental and Modeling Investigation of Post-Injection Plume Migration

 Scaling analysis for crossflow mechanisms has been widely used to characterize the
contribution of relevant physics including viscous, gravity and capillary forces. In this study,
we have focused on the use of the Bond number, , (e.g. Zhou et al. 1994) to characterize the
magnitude of gravity and capillary forces in counter-current vertical flows. The scaling
term is given by the ratio of gravity to capillary forces and has a general form

∆
(1)
where is the characteristic capillary pressure and can be expressed as

1 (2)

Previous studies (e.g. Zhou et al. (1994), Cinar et al. (2006)) have investigated the
transition from capillary to gravity dominated flows: If 5, gravity dominates the flow
whereas for 0.2, capillary forces dominate the flow. Accordingly, when 0.2 0.5,
a displacement is in the transition between capillary and gravity dominated flow regimes.

To correctly capture the interplay between capillary and gravity forces, a new dimensionless
time has been derived for counter-current vertical flows. The scaling is based on the
assumption that the displacement of CO2 plume in brine aquifers is completely immiscible with
no dispersion and diffusion. The proposed dimensionless timescale is given by

λ λ 1 exp 1 exp
λ σ θ (3)
Δρ

This dimensionless time includes the impact of effective permeability, porosity, phase
mobilities, characteristic length, interfacial tension, fluids densities and the ratio of gravity to
capillary forces (NB). A full derivation of our proposed dimensionless time scale is presented in
Nattwongasem and Jessen (2009).

To test the proposed dimensionless time we start by considering the idealized model system
shown in Fig 1 and use the fluid properties listed in Table X. The CO2 plume initially occupies
the lower 20% of the pore volume.
 Sw,i = 1.0 40 m.

SCO2,i = 1- Swc 8 m.

40 m.
Figure 8: Model dimensions and initial fluid distribution

Table X: Fluid properties based on the aquifer setting of Ide et al. (2007),

w 1054.8 kg/m3
CO2 728.8 kg/m3
w 0.4921 mPa.s
CO2 0.0612 mPa.s
0.025 N/m

Simulations of the segregation process were performed for a) homogeneous and isotropic
setting, b) homogeneous and anisotropic settings and c) heterogeneous, anisotropic settings with
three different correlation lengths. In this example, the Bond number was calculated to 1.34 and
hence the flow is in the capillary-gravity transition zone, shifted towards the capillary dominated
regime. Figure 9 compares the simulation results in terms of the fraction of gas that is trapped as
a function of the proposed dimensionless time on time scales proposed by Ma et al. (1997), Xie
and Morrow (2000) and Zhou et al. (2002).
1.0
Fraction of Gas Trapped

0.8

0.6
Homo
0.4 K1
K2
0.2 K3
Homo_Isotro
0.0
0 5 10 15 20
Dimensionless time by Ma et al. (1997)
 1.0

Fraction of Gas Trapped

0.8

0.6 Homo
K1
0.4
K2
0.2 K3
Homo_Isotro
0.0
0 50 100 150 200
Dimensionless Time by Xie & Morrow (2000)
1.0
Fraction of Gas Trapped

0.8

0.6 Homo
K1
0.4 K2
K3
0.2
Homo_Isotro
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Dimensionless Time by Zhou et al. (2002)

1.0
Fraction of Gas Trapped

0.8

0.6
Homo
0.4 K1
K2
0.2 K3
Homo_Isotro
0.0
0 1 2 3 4 5 6 7 8
Proposed Dimensionless Time

Figure 9: Comparison of the fraction of trapped gas vs. Dimensionless time scales for
homogeneous and heterogeneous anisotropic models: K1-3 represent short, medium and long
correlation lengths respectively.

It is clear from Figure 2 that the proposed dimensionless time more accurately captures the
dynamics of the plume migration for this set of model parameters. This observation
demonstrates the importance of including the magnitude of both gravity and capillary forces and
that appropriate scaling by use of the Bond Number, NB, allow us to improve the accuracy of an
estimate of trapped gas as well as the associated immobilization time.

Next, we consider a range or relevant aquifer settings reported in the work of Kopp et al.
(2009a, 2009b). Table reports the fluid properties and NB for different relevant aquifers based
on H = 100m, porosity of 0.3, and kr and Pc functions that are similar to the base case model
discussed previously. The initial plume is assumed to occupy the 20% of the available pore
volume and to be located at the bottom of the model at t = 0. From the scaling groups of Zhou
et al. (1994), we estimate the range of NB to span the capillary/gravity transition zone and the
gravity dominated flow regime (see Table XI).

Table XI: Fluid properties and Bond number (NB) at different aquifer conditions reported by
Kopp et al. (2009a, 2009b).

Depth P T w CO2 w CO2 σ NB M*

3 3
Aquifer ID. (m) (MPa) (⁰C) (kg/m ) (kg/m ) (Pa.s) (Pa.s) (dyne/cm)
Median M 1524 15.47 55.13 1025.5 622.8 0.5429 0.0530 27.28 8.645 0.150
Warm W 1524 14.95 104.49 994.5 313.1 0.29775 0.0300 30.84 12.939 0.155
Cold C 1524 15.51 37.43 1031.7 779.5 0.72827 0.0670 27.01 5.468 0.141
Shallow S 386 4.06 21.58 1032.9 100.2 1.02001 0.0170 40.32 13.546 0.026
Deep D 3495 35.31 115 995.2 666.1 0.27259 0.0551 27.21 7.083 0.310

Numerical calculations based on the above properties were performed and Figure 3 compares,
the fraction of gas trapped as a function of the proposed dimensionless time for the five aquifer
settings.

Figure 10: Comparison of the fraction of gas trapped for various Bond Numbers for a range of
relevant aquifer settings reported by Kopp (2009a, 2009b).

From Figure 3, we observe that the propose time scale substantially collapses the relevant
range of aquifer settings onto a single curve with moderate deviation of the shallow aquifer
setting. Accordingly, the proposed scaling can be used to estimate the immobilization time
based on NB and the characteristic mobility ratio from a given aquifer setting.
3. Lattice Boltzmann simulation of the rise and dissolution of two-dimensional immiscible
droplets
The methodology and main results are summarized in the publication of Chen and Zhang
(2009). We developed a LB model that is able to handle an arbitrary number of fluid
components and different density ratios. By simulating the rise of droplets with different sizes,
we found a power law relationship between the Eo number and terminal Re number. Our
simulation results can be described by Rodrigue’s empirical correlation, which was originally
derived for predicting gas bubble motion. Also, by simulating the simultaneous rise of two and
three identical droplets in a finite domain, we found that the average rise velocity was lower than
that of a single droplet with the same size, due to the mutual resistant interactions. In addition,
the trajectories of droplet except for the central one were not straightly upward, but moved away
from each other gradually, resulting from the non-zero net horizontal forces acting on the
droplets.

The two-phase LB model was combined with a solute transport model, as well as a boundary
condition handling dissolution at the liquid-liquid interface, in order to investigate the
complicated coupling between droplet size, flow field, solute transport, and dissolution at the
interface. For a given Pe, the case with a higher Da always had a faster dissolution rate
compared to the case with a lower Da. This implies that the physicochemical property of the
liquid-liquid interface is the fundamental factor affecting the whole dissolution process. For a
given Da, the case with a higher Pe always had a faster dissolution rate compared to the case
with a lower Pe. This was because with the higher Pe the advective transport relative to the
droplet was able to move the dissolved mass into the bulk solution quickly, which favored the
subsequent dissolution.

For a large Da and a small Pe, the process near the interface was diffusion-limited, and there
was a thick layer of dissolved mass covering around the droplet. The advective flow at the top
side of the droplet resulting from droplet rise was unable to clean up the solute near the interface
quickly. In order to accelerate the dissolution process, it was favorable to split one single
droplet into two or more small droplets, by increasing the interface area per unit mass of droplet.
In contrast, for a small Da and a large Pe, the process near the interface was dissolution-limited.
The dissolved mass accumulated near the interface was little, which was sensitive to the
advective flow at the top side of the droplet. The advective flow relative to a big droplet was
able to quickly clean the accumulated solute away from the interface, which enhanced the whole
dissolution process. Thus, in this case it was favorable to keep the droplet as a single one to
accelerate the dissolution process.

Based on the investigations of both Pe and Da numbers, we constructed a Da-Pe phase plane,
and proposed there exists an interface that divides the plane into region 1 and 2. In region 1 it
is favorable to split the single droplet into as many small ones as possible in order to accelerate
dissolution, while in region 2 it is favorable to keep the droplet as a single one for the same
purpose. This interface is expected to be an increasing function of the Pe number, which
implies that a small Pe number and large Da number lead to a region 1 case, while a large Pe
number and a small Da number lead to a region 2 case. This research has important
applications in many engineered and medical processes where fast dissolution of immiscible
liquid droplets is desired, such as geological carbon sequestration and drug delivery in blood
vessels. Different combinations of Pe and Da numbers can be investigated by numerical
simulation presented in this study, and a Da-Pe phase plane as well as the interface can be
obtained. After the real Pe and Da numbers of the system are determined, we will have a clear
idea whether it is favorable to break down the droplet or not in order to accelerate dissolution, by
judging whether the point (Pe, Da) falls in region 1 or 2.

4. General Framework of Simulator for Carbon Storage

A modular simulator, which is available under various circumstances, consists of three
pretreatment modules and eleven main program modules, the structural drawing as shown in
Figure 11.

Figure 11: Framework of CCS simulator

The three pretreatment modules are particularly used for preparing EOS parameters,
fluid-solid coupling parameters and special structure change.
 EOS parameter preparation module is built based on experiment data. It is used to generate
EOS parameters concerning carbon dioxide and brine, and the relationship between physical
property of formation and pressure and temperature. Any differences between the measured
and calculated data are minimized using regression facility which adjusts various Equation of
State parameters. This ‘tuned’ model is then exported in a form suitable for the simulator.

Fluid-solid coupling module is used to prepare parameters for relative permeability and
capillary curves. This module is prepared for the purpose of figuring out the relationship
between relative permeability, capillary pressure and rock porosity, pore-throat ratio,
throat-to-pore coordination number through applying the corresponding theory and formula,
together with the figure of core experiments in the lab. While observing the long-term state of
carbon dioxide, we have to put into consideration the dissolution and deposition of the mineral
compositions brought about by the existence of carbon dioxide. The dissolution and deposition
of mineral compositions cause the change of formation porosity, which is obvious. Besides,
dissolution of mineral compositions (mainly carbonatite) brings about secondary pores (also
dissolved pores). Mineral deposition leads to the jamming of pore throat. In general, the
geologic chemical reaction concerning carbon dioxide will change the natures of rock porosity,
pore-throat ratio, throat-to-pore coordination number of formation. Yet the changes of the
nature will accordingly cause deviations of residual water saturation, residual gas saturation,
threshold pressure and trends of relative permeability and capillary pressure curves. Therefore,
this module ought to be prepared for the purpose of generating a set of formulas revealing the
relationship between relative permeability and capillary pressure curves and rock porosity,
pore-throat ratio, throat-to-pore coordination number, so as to update relative permeability and
capillary pressure curves when necessary.

In the main program, we need a module that can be used for analyzing gridding flexibly
according to specific geologic construction (such as fracture and fault). It is a plain fact that
geologic chemical reaction will probably cause new fracture in the reservoir. Or it will enlarge
the original fracture. But it is indicated in the core experiment that carbon dioxide will erode
the core of the rocks and thereby, produce worm hole. So we had better build up our gridding
model by using the method of discrete fracture. This module will be applied to other modules
during the calculation process, so as to update the gridding model and make it better serve the
purpose of portraying the local geologic construction. What is worth being pointed out is that
we don’t need to update the whole reservoir, but only the concerned area and the gridding
suggesting the changes of geological construction. That is due to the large quantity of the
updating work, including updating cell list and connection list, working out the compositional
distribution and physical property parameters of the new gridding according to the compositional
distribution and physical property parameters based on the original gridding.
 As in the regular compositional model, we need modules for flash calculation, Jacobi matrix
construction, linear solver and fluid property update. We also need an explicit geological
chemical reaction module whose calculation is processed by applying the composition
distribution worked out by and mineral ion concentration with the method of chemical
thermodynamic. It will help to figure out the direction and degree of the chemical reaction
happening in each grid block. This module surely can be added to Newton iteration in an
implicit way, which can make the result of chemical reaction accurate, yet increase calculation
work to a great extent.

Then we need another module updating the porosity of the grid block based on the
calculation result of the above module. This module should also be available in adding or
deleting the special geological structure (such as caverns and fracture) in the grid blocks
according to the changes of porosity. Scale changes of the special geologic structure must also
be acceptable in this module. These operations ought to be based on the statistic regulation of a
series of micro-block simulation.

Likewise, we can build up an independent module to update the curve of absolute

permeability curves. This is not only reasonable but also advantageous for openness and
maintenance of the program. Yet since the relationship between absolute permeability and
porosity can be expressed with a simple log relationship, this very module can be included into
another module which is applied for the purpose of updating the porosity of reservoir according
to the result of the chemical reaction.

We also need a module to update relative permeability and capillary pressure curve base on
chemical reaction progress. The module is base on the relationship between fluid-solid
coupling parameters and porosity.

Finally, we need a module to update the special structure in certain block base on chemical
reaction calculation.

5. CO2 Sequestration Simulation Using Discrete Fracture Modeling (DFM)

CO2 sequestration without chemical reactions is considered. The chemical reaction module
is still in development.

In this part, DFM is used to characterize the reservoir geometry. We firstly discretize the
reservoir into triangles and calculate the transmissibility between each grid block. The
triangulation is prolonged into a one-dimension volume reservoir and simulated. In this
simulation, we use GPRS as the reservoir simulator.
5.1 Multi fracture system
The system has a sandstone reservoir with many fractures. Sandstone porosity is 0.25.
Fracture porosity is 1. Sandstone permeability is 10mD. Fracture permeability is 1000000mD.
Fracture aperture is 0.1mm. Medium is firstly filled with water. CO2 is injected into the left
bottom corner Injection water at bottom left corner. CO2 injection rate fixed at 0.01PVI/d.

The system has 83 fractures. The fractures are distributed randomly in the reservoir. 83
fractures are discretized into 814 fractures, as shown in Figure 13.

Figure 12: Fracture System

100

90

80

70

60

50

40

30

20

10

0
0 10 20 30 40 50 60 70 80 90 100
Figure 13: triangulation of fractured reservoir
 Figure 14 represents the CO2 saturation profiles after several time steps of water injection.
After CO2 has arrived at the fractures, it will firstly flow through fractures rather than sandstone.
Finally we could find CO2 is distributed all around the fractures.

Saturation distribution Saturation distribution

100 1 100 1

90 0.9 90 0.9

80 0.8 80 0.8

70 0.7 70 0.7

60 0.6 60 0.6

50 0.5 50 0.5

40 0.4 40 0.4

30 0.3 30 0.3

20 0.2 20 0.2

10 0.1 10 0.1

0 0 0 0
0 20 40 60 80 100 0 20 40 60 80 100

0.005PVI 0.08PVI

Saturation distribution
100 1 Saturation distribution
100 1
90 0.9
90 0.9
80 0.8
80 0.8
70 0.7
70 0.7
60 0.6
60 0.6
50 0.5
50 0.5
40 0.4
40 0.4
30 0.3
30 0.3
20 0.2
20 0.2
10 0.1
10 0.1
0 0
0 20 40 60 80 100
0 0
0 20 40 60 80 100

0.14PVI 0.22PVI
Figure 14: CO2 saturation distribution

5.2 Multi fracture system with mud layers

The system has 10 layers, including reservoir : 0-10m, 20-30m, 40-50m, 60-70m, 80-90m
and mud: 10-20m, 30-40m, 50-60m, 70-80m, 90-100m. Sandstone porosity is 0.25. Mud
porosity is 0.01. Fracture porosity is 1. Sandstone permeability is 0.1mD. Sandstone
permeability is 0.001mD. Fracture permeability is 1000000mD. Fracture aperture is 0.1mm.
Medium is firstly filled with water. CO2 is injected into the left bottom corner Injection water
at bottom left corner. CO2 injection rate fixed at 0.01PVI16/d.

100

90

80

70

60

50

40

30

20

10

0
0 10 20 30 40 50 60 70 80 90 100

Figure 15: triangulation of fractured reservoir

Figure 16 represents the CO2 saturation profiles after several time steps of water injection.
CO2 will preferentially pass through the fractures, then sandstone, and then mud.

Saturation distribution Saturation distribution

100 1 100 1

90 0.9 90 0.9

80 0.8 80 0.8

70 0.7 70 0.7

60 0.6 60 0.6

50 0.5 50 0.5

40 0.4 40 0.4

30 0.3 30 0.3

20 0.2 20 0.2

10 0.1 10 0.1

0 0 0 0
0 20 40 60 80 100 0 20 40 60 80 100

0.008PVI 0.04PVI
 Saturation distribution Saturation distribution
100 1 100 1

90 0.9 90 0.9

80 0.8 80 0.8

70 0.7 70 0.7

60 0.6 60 0.6

50 0.5 50 0.5

40 0.4 40 0.4

30 0.3 30 0.3

20 0.2 20 0.2

10 0.1 10 0.1

0 0 0 0
0 20 40 60 80 100 0 20 40 60 80 100

0.08PVI 0.22PVI

Figure 16: CO2 saturation distribution

5.3 Leakage along wellbore

Preexisting wells and well bores may play major pathways through the crust as their high
permeability. Current well closure and abandonment technology seems sufficient to confine
CO2, however, individual wells may suffer from a variety of factors that limit their integrity,
including improper cementation, improper plugging, overpressure, corrosion, and other failure
conditions.

In this case, we describe a wellbore leakage system. The blue line in the middle of the
reservoir is a well, which has main fractures along the wellbore.

The system has 10 layers, including reservoir : 0-10m, 20-30m, 40-50m, 60-70m, 80-90m
and mud: 10-20m, 30-40m, 50-60m, 70-80m, 90-100m. Sandstone porosity is 0.25. Mud
porosity is 0.01. Fracture porosity is 1. Sandstone permeability is 0.1mD. Sandstone
permeability is 0.001mD. Fracture permeability is 1000000mD. Wellbore fracture aperture is
0.1mm. Medium is firstly filled with water. CO2 is injected into the left bottom corner
Injection water at bottom left corner. CO2 injection rate fixed at 0.01PVI/d.
 100

90

80

70

60

50

40

30

20

10

0
0 10 20 30 40 50 60 70 80 90 100

Figure 17: triangulation of reservoir with wellbore leakage

We could clearly find that CO2 will move up along the wellbore and disperse into the
sandstone. While in the mud, CO2 does not move that clearly. We may conclude that
wellbore leakage induces main movement of CO2 in the sandstone. Figure 18 represents the
CO2 saturation profiles after several time steps of water injection.

Saturation distribution
100 1 Saturation distribution
100 1

90 0.9
90 0.9

80 0.8 80 0.8

70 0.7 70 0.7

60 0.6 60 0.6

50 0.5 50 0.5

40 0.4 40 0.4

30 0.3 30 0.3

20 0.2 20 0.2

10 0.1
10 0.1
0 0
0 0 0 20 40 60 80 100
0 20 40 60 80 100

0.002PVI 0.02PVI
 Saturation distribution Saturation distribution
100 1 100 1

90 0.9 90 0.9

80 0.8 80 0.8

70 0.7 70 0.7

60 0.6 60 0.6

50 0.5 50 0.5

40 0.4 40 0.4

30 0.3 30 0.3

20 0.2 20 0.2

10 0.1 10 0.1

0 0 0 0
0 20 40 60 80 100 0 20 40 60 80 100

0.05PVI 0.11PVI

Figure 18: CO2 saturation distribution

Progress
For basin evaluation, a regional study on potential CO2 sequestration in the Jianghan Basin
was performed. The two formations, Qianjiang and Xingouzui Formation, were firstly
identified and investigated with respect to various factors and conditions. Preliminary analysis
indicated that the two formations could be suitable candidate sites for CO2 injection.

For dynamic experiments, the experimental work was carried out utilizing dynamic
resistivity measurements to monitor the evolution of an isooctane plume initially located at the
bottom of the column. Time estimates for residual entrapment via scaling analysis for
counter-current flow and design of segregation experiments based on an analog fluid system in
glass-bead pack were studied.

For microscopic modeling and simulation, a coupled multiphase Lattice Boltzmann (LB)
model was developed to simulate the dissolution of immiscible liquid droplets in another liquid
during the rising process resulting from buoyancy. It was found that there exists a terminal rise
velocity for each droplet, and there was a power law relationship between the Eötvös (Eo)
number and the terminal Reynolds (Re) number. The simulation results were in agreement with
the empirical correlation derived for predicting bubble rise. The Damkohler (Da) and Peclet
(Pe) numbers were varied to investigate the coupling between droplet size, flow field, dissolution
at the interface, and solute transport.

For macroscopic modeling and simulation study, a CO2 sequestration simulation framework
that can account for natural and drilling/injection induced fractures and chemical reactions was
developed. This framework is capable of modeling dynamic flow in micro-scale and
reservoir-scale simultaneously. Simulations have been done using Discrete Fracture Modeling
(DFM). Systems with fractures in CO2 injection formation versus in cap rocks are compared.
Results have shown that the existence of mudstone layers could prevent injected CO2 from
leaking outside the reservoir when no fractures are present. While vertical fractures intersecting
with mudstone layers will cause significant leakage increase as fractures form extremely
preferential pathways for CO2 transport.

This work is designed to accelerate translation of data into scientific understanding,

simulation of site performance, and establishment of best standards. For example, this work is
highly recognized by Shenhua Group. For the first CO2 sequestration in a saline aquifer project
in China, we are working with Shenhua on building the geological model, performing long-term
simulations and history match with monitored data. The ultimate goal of this research is to
develop realistic simulation tools, provide specific guidance and establish standards for the
design and evaluation of CO2 sequestration in saline aquifers in China.

Future Plans
Along dynamic experiments, investigations on the dynamic behavior of CO2 migration in
saline aquifers and the long-term experiments on CO2-brine-rock interactions will be performed
to examine the geochemical reactions in selected saline aquifers. The numerical simulation will
be performed to study the long-term fate of injected CO2 and potential impact of escape of CO2
due to the failure of natural barriers or migration along natural/anthropogenic escape paths such
as faults or wells.

To improve the current state-of-the-art simulation technology, we are currently studying the
dynamics of countercurrent flows in well-defined porous materials (glass bead packs) using an
analog isooctane/brine fluid system. The experimental work utilizes dynamic resistivity
measurements to monitor the evolution of an isooctane plume initially located at the bottom of
the column.
 Figure 19: Counter-current flow experiment in glass bead pack.

The experimental observations from the segregation experiments will be supplemented with
steady state relative permeability measurements (co- and counter-current) to provide input to our
planned modeling efforts. The combined effort will allow us to test the accuracy of using
co-current relative permeability functions to estimate the displacement behavior of gravity driven
counter-current displacement experiments. It is expected that the traditional modeling approach
will fail to represent the dynamic behavior of the gravity driven displacement with any
reasonable accuracy. The experimental observations will form the basis for development of
new models that account for a dynamic transition from viscous dominated flow (injection period)
and gravity dominated flow (post injection period). A detailed investigation and analysis of this
transition will improve our predictive capabilities and strengthen our confidence in time-scale
estimates for CO2 sequestration projects. Interpretation of the experimental observations and
the development of new models will by aided by pore- and core-scale numerical calculations.

We will continue to study the effects of pore-scale structures and processes on macroscopic
coefficients (e.g., permeability, dispersivity, mechanical, and reaction coefficients) and identify
key microscopic parameters (e.g., pore-size distribution, diffusivity, and surface reaction rates)
and predominant processes that control the macroscopic quantities. In the third year, we will
develop upscaling strategies for deriving macroscopic kinetic and thermodynamic models for the
interactions and for driving flow parameters such as relative permeability and capillary pressure
functions. We will develop upscaling strategies for deriving macroscopic kinetic and
thermodynamic parameters as well as flow parameters for the complex system. The upscaling
techniques for non-reactive flows in porous media will be evaluated and modified to account for
the interplay of convection, diffusion, reaction, and pore-geometry evolution for reactive flows.
The upscaled constitutive relations will be tested on the experimental observation from and
included in the analysis and model development for counter-current relative permeability.

We will study more from micro-scale simulation to reservoir-scale simulation. In the

micro-scale simulation, Lattice Boltzmann Method (LBM) will be utilized to simulation
long-term interactions between rock and CO2. In a micro-scale system, the grid can be fine
enough to capture the dissolution and dissipation of crystal particles, and then predict the
pore-throat variation and the appearance, enlargement of micro fractures. Patterns between the
general condition change of the reservoir property and the general extent of the chemical
reactions will be drawn and then applied to reservoir-scale simulation. Re-arrangement of the
grid-block, in hope of revealing the temporary status of grid in an accurate way, will be
performed for better description.

For DFM modeling and simulation on CO2 sequestration, we plan to study more cases,
including induced fractures by hydraulic fracturing, shale confinement, and a localized fractured
region with strong capillary pressure effects. More complex 3D fracture network system will
also be considered in the future. Realistic modeling and simulation based on regional study
results of Jianghan Basin will also be conducted.

Publications
1. [Link], [Link], [Link], [Link]. Numerical Simulation of Potential Effects of CO2 Leakage on Shallow
Potable Aquifers. Accepted by IAHS Red Book publication entitled “Managing Groundwater and the
Environment”.
2. Wei Zhang, Yilian Li, Tianfu Xu, et al. Numerical simulation on effects of convective mixing on long term CO2
geological storage in deep saline formations. Environmental Earth Science (the revised paper has been
submitted)
3. Anne Nyatichi Omambia, Yilian Li, Wei Zhang. Numerical simulation of carbon dioxide injection in
Wangchang Oilfield - Jianghan Basin, China. Accepted by Journal of American Science.
4. Ling Jiang. Research on the Environmental Effects on the CO2 geological storage, Jianghan Case[D], China
University of Geosciences, 2010,6
5. Gengbiao Qiu. Design of the Injection Technique of CO2 in Xingou Olifield of the Jianghan Basin [D], China
University of Geosciences, 2010,6
6. Anne Nyatichi Omambia. Design of the carbon dioxide injection in Wangchang Oilfield-Jianghan Basin, China
[D], China University of Geosciences, 2010,6
7. Nattwongasem, D. and Jessen, K: "Residual Trapping of CO2 in Aquifers during the Counter-Current Flow",
Paper SPE 125029, SPE Annual Technical Conference and Exhibition, New Orleans, Louisiana, USA, 4–7
October 2009.
8. Gong, B., Huo, D., Discrete Modeling and Simulation on Potential Leakage Through Fractures and Wells in
CO2 Sequestration, SPE135507, presented at the SPE Annual Technical Conference and Exhibition held in 19-
 22 September 2010 in Florence, Italy.
9. Gong, B., Huo, D., Numerical Simulation on CO2 Leakage through Fractures along Wells using Discrete
Fracture Modeling, SPE133986, presented at SPE Asia Pacific Oil & Gas Conference and Exhibition held in
18-20 October 2010 in Brisbane, Australia.
10. Chen, C., and D. Zhang, Lattice Boltzmann simulation of the rise and dissolution of two-dimensional
immiscible droplets, Physics of Fluids, 21, 103301, DOI: 10.1063/1.3253385, 2009.
11. Chen, C., and D. Zhang, Pore-scale simulation of density-driven convection in fractured porous media during
geological CO2 sequestration, Water Resources Research, under review, 2010.

References
1. Altunin, V.V. Thermophysical Properties of Carbon Dioxide, Publishing House of Standards, 551 pp., Moscow,
1975 (in Russian).
2. Andersen, G., A. Probst, L. Murray and S. Butler. An Accurate PVT Model for Geothermal Fluids as
Represented by H2O-NaCl-CO2 Mixtures, Proceedings 17th Workshop on Geothermal Reservoir Engineering,
pp. 239 - 248, Stanford, CA, 1992.
3. Battistelli, A., C. Calore and K. Pruess. The Simulator TOUGH2/EWASG for Modeling Geothermal
Reservoirs with Brines and Non-Condensible Gas, Geothermics, Vol. 26, No. 4, pp. 437 - 464, 1997.
4. Cao, H.:” Development of Techniques For General Purpose Simulators,” PhD thesis, Stanford University,
2002.
5. Fan, Y.:”Development of CO2 Sequestration Modeling Capabilities In Stanford General Purpose Research
Simulator,” PhD thesis, Stanford University, 2006.
6. Fenghour, A., Wakeham, W.A., Vesovic, V. The Viscosity of Carbon Dioxide, J. Phys. Chem. Ref. Data, Vol.
27, No.1, 1999
7. Kell, G.S and Whalley, E. Reanalysis of the density of liquid water in the range 0-150 C and 0-1 kbar , J. Chem.
Phys., Vol 62, No. 9, May 1975.
8. Nelson, C.R., Evans, J.M., Sorensen, J.A., Steadman, E.N., Harju, J.A., Factors affecting the potential for CO2
leakage from geological sinks, PCOR partnership, 2005.
9. Pashin, J. C., Jin. G, and Payton, J. W., 2004, Three-dimensional computer models of natural and induced
fractures in coalbed methane reservoirs of the Black Warrior basin: Alabama Geological Survey Bulletin 174,
62 p.
10. Potter, Babcock and Brown. A new method for determining the solubility of salt in aqueous solutions at
elevated temperatures, Research U.S. Geol. Surv. 5, no. 3, Page 389-395, 1977.
11. Spycher, N. and Pruess, K. CO2-H2O mixtures in the geological sequestration of [Link]. Partitioning in
chloride brines at 12-100 C and up to 600 bar, Geochimica et Cosmochimica Acta, Vol. 69, No. 13, P
3309-3320, 2005.
12. Vesovic, V. Wakeham, W. A., Olchowy, G.A., Sengers, J.V., Watson, J.T.R., Millat, J. The Transport
Properties of Carbon Dioxide, J. Phys. Chem. Ref. Data, Vol. 19, No.3, 1990.
13. Zaytsev, I.D., and Aseyev, G.G. Properties of Aqueous Solutions of Electrolytes, CRC Press, 1993.
14. EU Geocapacity, 2009. Assessing European Capacity for Geological Storage of Carbon Dioxide.
15. Li Guoyu, et al., 2002. Atlas of China’s petroliferous basins [D]. Petroleum Industry Press.
16. Pan Guosi, Zhu Zhendong, 1988. The Symposium on structural features of oil gas bearing areas in China. The
Petroleum Industry Press.
17. Yang Panxin, et al.2009, Structure Model and Evolution of the Jianghan Basin and Relation with Moderate to
Strong Earthquakes [J], Earthquake, Vol.29, No.4, P123-131.
18. Jianghan Oilfield Company of China Petroeum. Petroleum Geology of Jianghan Oilfield [D]. Petroleum
Industry Press.
19. Yin Wenjie, Li Hui, 2005. Research on some favourable phase belts about the reservoir caprocks of Cretaceous
– Paleocene in Jianghan Basin [J]. Journal of Jianghan Petroleum University of Staff and Workers, 18(4): 12-14
20. Zhou, D., Jia, L., Kamath, J. and Kovscek, A. R.: “Scaling of counter-current imbibition processes in
low-permeability porous media”, Journal of Petroleum Science and Engineering, Volume 33, Issues 1-3, p.
61-74, 2002.
21. Kumar, A., Noh, M., Pope, G.A., Sepehmoori, K., Bryant, S. and Lake, L.W.: “Reservoir Simulation of CO2
Storage in Deep Saline Aquifers”, Society of Petroleum Engineers (SPE) Journal, Vol. 10(3), 2005.
22. Ide, S. T., Jessen, K. and Orr, F.M: “Storage of CO2 in saline aquifers: Effects of gravity, viscous, and capillary
forces on amount and timing of trapping” International Journal of Greenhouse Gas Control, Vol. 1(4), p.
481-491, 2007.
23. Nattwongasem, D. and Jessen, K: "Residual Trapping of CO2 in Aquifers during the Counter-Current Flow",
Paper SPE 125029, SPE Annual Technical Conference and Exhibition, New Orleans, Louisiana, USA, 4–7
October 2009.
24. Ma, S., X. Zhang, and N.R. Morrow. "Influence of Fluid Viscosity on Mass Transfer between Fractures and
Matrix." The 1995 Petroleum Society of CIM Annual Technical Meeting. Banff, Alberta: 14-17 May, 1997.
25. Xie, X., and N.R. Morrow. "Oil Recovery by Spontaneous Imbibition from Weakly Water-Wet Rocks." the
International Symposium of the Society of Core Analysis Meeting. Abu Dhabi, UAE, 2000.
26. Kopp, A., H. Class, and R. Helmig. "Investigation on CO2 storage capacity in saline aquifers Part1.
Dimensional analysis of flow processes and reservoir characteristics." International Journal of Greenhouse Gas
Control 3 (2009a): 263-276.
27. Kopp, A., H. Class, and R. Helming. "Investigations on CO2 storage capacity in saline aquifers-Part 2:
Estimation of storage capacity coefficients." Internation Journal of Greenhouse Gas Control 3 (2009b):
277-287.
28. C. Chen, B. L. T. Lau, J. Gaillard, and A. I. Packman, “Temporal evolution of pore geometry, fluid flow, and
solute transport resulting from colloid deposition,” Water Resour. Res., 45, W06416,
doi:10.1029/2008WR007252 (2009).
29. S. P. Dawson, S. Chen, and G. D. Doolen, “Lattice Boltzmann computations for reaction-diffusion equations,” J.
Chem. Phys., 98, 1514, (1993).
30. Q. Kang, D. Zhang, and S. Chen, “Displacement of a two-dimensional immiscible droplet in a channel”, Phys.
Fluids 14(9), 3203 (2002a).
31. Q. Kang, D. Zhang, S. Chen, and X. He, “Lattice Boltzmann simulations of chemical dissolution in porous
media,” Phys. Rev. E 65(3), 036318 (2002b).
32. Q. Kang, D. Zhang, and S. Chen, “Simulation of dissolution and precipitation in porous media”, J. Geophys.
Res. 108(B10), 2505, doi:10.1029/2003JB002504 (2003).

Contacts:
Dongxiao Zhang donzhang@[Link]
Kristian Jessen jessen@[Link]
Cheng Chen chen54@[Link]
Dalad Nattwongasem nattwong@[Link]
Bin Gong gongbin@[Link]
Qingdong Cai caiqd@[Link]
Yi Zheng yizheng@[Link]
Da Huo danielhuo@[Link]
Xuan Liu occultliu@[Link]
Ming Lei leiming0461@[Link]
Zhijie Wei wjzweizhijie@[Link]
Xiang Li yesyou@[Link]
Lingyu Zhang lbzhang1984@[Link]
Yongbin Zhang zybpkucoe@[Link]
Yilian Li [Link]@[Link]
Yanxin Wang [Link]@[Link]
Jianmei Cheng jmcheng@[Link]
Wei Zhang zhangwei_cug@[Link]
Ling Jiang jiangling84yy@[Link]
Gengbiao Qiu 276450108@[Link]
Qi Fang frances2009@[Link]
Yibing Ke [Link]@[Link]
Peng Cheng 042041cheng@[Link]
Sanxi Peng heiyingpsxq@[Link]
Ronghua Wu wuronghua050@[Link]
Kabera Telesphore kaberacris@[Link]
Anne Nyatichi Omambia tichiomambia@[Link]
Jianxiong Dong dong12205206@[Link]