Energy Conversion and Management 118 (2016) 204222

Contents lists available at ScienceDirect

Energy Conversion and Management

journal homepage: www.elsevier.com/locate/enconman

Review

Property impacts on Carbon Capture and Storage (CCS) processes:

A review
Yuting Tan a,,1, Worrada Nookuea b,,1, Hailong Li b, Eva Thorin b, Jinyue Yan a,b,
a
School of Chemical Science and Engineering, Royal Institute of Technology, SE 100 44 Stockholm, Sweden
b
School of Business, Society and Engineering, Mlardalen University, SE 721 23 Vsters, Sweden

a r t i c l e i n f o a b s t r a c t

Article history: The knowledge of thermodynamic and transport properties of CO2-mixtures is important for designing
Received 21 January 2016 and operating different processes in carbon capture and storage systems. A literature survey was con-
Received in revised form 23 March 2016 ducted to review the impact of uncertainty in thermos-physical properties on the design and operation
Accepted 25 March 2016
of components and processes involved in CO2 capture, conditioning, transport and storage. According to
Available online 5 April 2016
the existing studies on property impacts, liquid phase viscosity and diffusivity as well as gas phase dif-
fusivity significantly impact the process simulation and absorber design for chemical absorption.
Keywords:
Moreover, the phase equilibrium is important for regenerating energy estimation. For CO2 compression
Property impact
CO2 mixtures
and pumping processes, thermos-physical properties have more obvious impacts on pumps than on com-
Impurities pressors. Heat capacity, density, enthalpy and entropy are the most important properties in the pumping
Thermodynamic and transport properties process, whereas the compression process is more sensitive to heat capacity and compressibility. In the
Carbon Capture and Storage (CCS) condensation and liquefaction process, the impacts of density, enthalpy and entropy are low on heat
exchangers. For the transport process, existing studies mainly focused on property impacts on the perfor-
mance of pipeline steady flow processes. Among the properties, density and heat capacity are most
important. In the storage process, density and viscosity have received the most attention in property
impact studies and were regarded as the most important properties in terms of storage capacity and
enhanced oil recovery rate. However, for physical absorption, physical adsorption and membrane sepa-
ration, there has been a knowledge gap about the property impact. In addition, due to the lack of exper-
imental data and process complexity, little information is available about the influence of liquid phase
properties on the design of the absorber and desorber for chemical absorption process. In the CO2 condi-
tioning process, knowledge of the impacts of properties beyond density and enthalpy is insufficient. In
the transport process, greater attention should focus on property impacts on transient transport pro-
cesses and ship transport systems. In the storage process, additional research is required on the disper-
sion process in enhanced oil recovery and the dissolution process in ocean and saline aquifer storage.
2016 Elsevier Ltd. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
2. CO2 capture by chemical absorption. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
2.1. Process description and related properties of chemical absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
2.1.1. Process description of chemical absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
2.1.2. Key components and related properties of chemical absorption. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
2.2. Existing studies of property impacts on chemical absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
2.2.1. Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
2.2.2. Heat capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
2.2.3. Viscosity and diffusivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209

Corresponding authors at: School of Chemical Science and Engineering, Royal Institute of Technology, SE 100 44 Stockholm, Sweden (Y. Tan, J. Yan); School of Business,
Society and Engineering, Mlardalen University, SE 721 23 Vsters, Sweden (W. Nookuea, J. Yan).
E-mail addresses: tany@kth.se (Y. Tan), worrada.nookuea@mdh.se (W. Nookuea), jinyue@kth.se (J. Yan).
1
These two authors contributed equally to this work.

http://dx.doi.org/10.1016/j.enconman.2016.03.079
0196-8904/ 2016 Elsevier Ltd. All rights reserved.
 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222 205

Nomenclature

Symbol P pressure
Qsen sensible heat for solvent heating PC-SAFT Perturbed-Chain Statistical Associating Fluid Theory
Qlatent latent heat for water evaporation PR PengRobinson
Qrxn heat of reaction for the desorption of CO2 PRMHV2 PengRobinson with modified HuronVidal mixing
 specific heat ratio rules
q density PROPATH PROgram PAckage on the THermo-physical proper-
ties of fluid
Abbreviation PRSV PengRobinsonStryjekVera
3P1T three-parameter cubic equation of state with one PRWS PengRobinson with WongSandler mixing rules
parameter temperature-dependent PSA Pressure Swing Adsorption
BWRS BenedictWebbRubinStarling PSRK Predictive SoaveRedlichKwong
CCS Carbon Capture and Storage PT PatelTeja
e-NRTL Electrolyte Nonrandom Two-liquid REFPROP REFerence fluid PROperties
EOR enhanced oil recovery RK RedlichKwong
EOS equation of state RK-ASPEN RedlichKwong-ASPEN
GDPC generalized pressure drop correlation RKS RedlichKwongSoave
GHG greenhouse gas RKSWS RedlichKwongSoave with WongSandler mixing
IFT interfacial tension rules
ISRK Improved SoaveRedlichKwong SRK SoaveRedlichKwong
KSMHV2 KwongSoave with modified HuronVidal mixing SR-POLAR SchwartzentruberRenon
rules SW SpanWagner
LK LeeKesler T temperature
LKP LeeKesler Plcker TSA temperature swing adsorption
MEA monoethanolamine UNIQUAC Universal Quasi Chemical
MMP Minimum Miscible Pressure VSA vacuum swing adsorption
MPR Modified PengRobinson
MSRK Modified SoaveRedlichKwong

2.2.4. Phase equilibrium. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209

3. CO2 capture by physical absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
3.1. Process description and related properties of physical absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
3.1.1. Process description of physical absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
3.1.2. Key components and related properties of physical absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
3.2. Existing studies of property impacts on physical absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
4. CO2 capture by physical adsorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
4.1. Process description and related properties of physical adsorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
4.1.1. Process description of physical adsorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
4.1.2. Key components and related properties of physical adsorption and existing studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
5. CO2 capture by membrane separation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
5.1. Process description and related properties of membrane separation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
5.1.1. Process description of membrane separation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
5.1.2. Key components and related properties of membrane separation and existing studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
6. CO2 conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
6.1. Process description and related properties of CO2 conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
6.1.1. Process description of CO2 conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
6.1.2. Key component and related properties of CO2 conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
6.2. Existing studies of property impacts on CO2 conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
6.2.1. Property impacts on compression and pumping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
6.2.2. Property impacts on condensation and liquefaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
7. CO2 transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
7.1. Process description and related properties of CO2 transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
7.1.1. Process description of CO2 transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
7.1.2. Key component and related properties of CO2 transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
7.2. Existing study of the property impacts on CO2 transport. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
7.2.1. Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
7.2.2. Viscosity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
7.2.3. Phase characteristic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
7.2.4. Heat capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
7.2.5. Thermal conductivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
8. CO2 storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
8.1. Process description and related properties of CO2 storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
8.1.1. Process description of CO2 storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
8.1.2. Key components and related properties of CO2 storage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
8.2. Existing study of the property impacts on CO2 storage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
8.2.1. Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
206 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222

8.2.2. Viscosity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218

9. Discussion and conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
10. Future studies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220

1. Introduction turn will alter the properties. The deviation of properties may sig-
nificantly impact the design and operation of and the investment
As a greenhouse gas, carbon dioxide (CO2) makes a major con- in the system. Taking CO2 storage as an example, a deviation of
tribution to the climate change. Approximately 78% of the total
5% in density of CO2 mixtures will reduce the storage capacity
greenhouse gas (GHG) emissions increase during the period by approximately 20% and increase storage cost by approximately
20002010 derived from the CO2 emissions from fossil fuel com- 15%. For a CO2 storage capacity of 3.5 million tons per year, the cost
bustion and industrial processes [1]. It is important to develop difference caused by the calculation uncertainty will be 0.5 million
technologies to reduce CO2 emissions and mitigate the latters EUR [6].
impact on climate change. According to the IEA Technology Road- In our previous work [7,8], we collected and evaluated the avail-
map 2013, a portfolio of technological options include end-use fuel able experimental data and existing property models for both ther-
and electricity efficiency, renewables, Carbon Capture and Storage modynamic and transport properties for CO2 mixtures and
(CCS), end-use fuel switching, nuclear power, power generation identified knowledge gaps concerning the experimental data and
efficiency and fuel switching, in which CCS accounts for 14% of property model development. Although all the properties can
the total GHG mitigation potential to achieve the 2 degree target impact the process design and performance, different properties
by 2050 [2]. will impact the process to different degrees. To address knowledge
CCS technologies have attracted much attention because they gaps in property impacts, it is important to prioritize research on
are technically feasible methods to significantly reduce CO2 emis- the most important properties. In addition, understanding the
sions from sources such as power generation, natural-gas process- impact of properties can provide a guideline of the accuracy when
ing and industrial processes in a brief period. The CCS chain
comprises four steps: capture, conditioning, transport and storage.
Capture technologies can be divided into three categories:
post-combustion capture, pre-combustion capture and oxy-fuel
combustion. Compared with the other two technologies, post- Chemical Absorption (Section 1)
Physical absorption (Section 2)
combustion capture is easier to implement as a retrofit option
Physical adsorption (Section 3)
for existing plants because the radical changes in the plants are Capture
Membrane separation (Section 4)
not required. A common separation approach for post-
combustion capture is chemical absorption, which can feasibly
capture CO2 at a low concentration. The pre-combustion capture Compression and pumping (Section 4)
technology can be more efficient due to the high concentration Condensation and liquefaction (Section 5)
Conditioning
of CO2 in the syngas produced from gasification. However, the cap-
ital cost of the gasification process is often larger compared with a
traditional pulverized coal combustion power plant. Feasible sepa-
ration methods for pre-combustion capture are physical sorption Pipeline transport (Section 6)
and membrane separation. Oxy-fuel combustion, because the com- Ship transport (Section 6)
Transport
bustion occurs in a nitrogen-less environment, results in flue gases
mainly containing CO2 and steam (H2O). CO2 of high purity can be
easily separated by simple water condensation and downstream
Geological storage (Section 7)
purification. Before CO2 is transported, it typically passes through Ocean storage (Section 7)
a conditioning process to meet transportation requirements. CO2 Storage
conditioning involves processes such as compression/pumping,
condensation and liquefaction [3]. CO2 can be transported to the
storage sites either by pipeline or in vessels. CO2 can either be
stored geologically by injecting CO2 into geologic formation deep Fig. 1. CCS processes in this study.
underground or be sequestrated in the ocean. Alternatively, it
can be stored in an underground reservoir, thus in ocean storage,
trapped in minerals. CO2 transport and storage combines technical
costs and climate impact assessments, and the simulation consid-
ers a wide range of variables and parameters such as flow rates,
ship size, transport distances, and storage capacity [4].
A better understanding is required of the design and operation
of different CCS processes in terms of thermo-physical properties
of the CO2 mixtures, namely, thermodynamic and transport prop-
erties. These properties are prerequisites for the numerical simula-
tions of all CCS processes. For example, in development, the design
and simulation of amine-based chemical absorption columns
require accurate information on viscosities, thermal conductivities
and diffusion coefficients [5]. However, the CO2 captured from an
energy-conversion process always contains impurities, which in Fig. 2. Chemical absorption process flow sheet (adapted from [18]).
 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222
developing property models. However, no systematic study on
property impacts on CCS has been conducted.
The purpose of this paper is to investigate the property impacts
on the design and operation of different CCS processes by review-
ing existing work. Based on the results, the knowledge gaps will be
identified and the priority of properties in model development will
be analyzed. This work will establish the state of the art of property
impact studies and provide suggestions for future research on the
properties of CO2 mixtures. The CCS processes focused on in this
paper are chemical/physical absorption, physical adsorption, mem-
brane separation, CO2 conditioning, transport and storage pro-
cesses, as listed in Fig. 1.
2. CO2 capture by chemical absorption
2.1. Process description and related properties of chemical absorption
2.1.1. Process description of chemical absorption
Chemical absorption using amine solvents is widely used and
commercially mature for separating CO2. It can be applied to gas
with a low CO2 concentration due to its high reaction rate with
CO2 [912]. The chemical absorption process is illustrated in
Fig. 2. An absorber, a heat exchanger, a desorber and a reboiler
are the key pieces of equipment. The flue gas is passed into the
absorber at the bottom and comes into contact with the solution
such as aqueous monoethanolamine (MEA) injected from the top.
The separation of CO2 from the flue gas occurs as mass transfer is
enhanced by chemical reaction. The CO2 rich solvent leaves the
absorber at the bottom and enters through the top of the desorber
where it flows downward and comes into contact with the strip-
ping steam generated in the reboiler. Most of the CO2 is desorbed
inside the desorber and partly in the re-boiler. The desorbed CO2
leaves the top of the desorber at a high temperature and most of
water vapor is condensed in the condenser. The regenerated lean
solution is then returned to the absorber through the rich/lean heat
exchanger and cooling unit to lower its temperature. Solvent and
water make-up are passed to the lean solution to keep the circulat-
ing solvent flow rate constant and to maintain its amine concentra-
tion [1317].
2.1.2. Key components and related properties of chemical absorption
To achieve a high recovery rate, large volumes of the absorber
and desorber are typically required. Consequently, the regenera-
tion process consumes a significant amount of thermal energy.
Therefore, the sizing of the absorber/desorber tower and the oper-
ating condition and energy requirement of the reboiler are impor-
tant issues for process application [19]. Designing the columns,
such as determining their height and diameter and their packing
type and flooding velocities, requires knowing the properties of flu-
ids. According to the review on modeling CO2 capture by chemical
absorption [20], thermo-physical properties are crucial to achieve
accurate simulations of pressure drop, CO2 recovery rate and purity
[21].
2.1.2.1. Columns diameter. CO2 absorption and desorption are gen-
erally operated in a counter-current flow. The diameters of the
absorber and desorber columns can be estimated using a similar
methodology. The column diameter is determined by the proper-
ties, flow rate and packing specification. The flooding condition
can be determined by the maximum gas velocity because the
down-flowing liquid resistance and the column pressure drop
increase with increasing gas velocity. The diameter design of the
packing column is typically based on 7080% gas velocity or flux
[21,22]. The flooding velocity can be retrieved from the gas flux
207
at flooding when the mass flow rate of gas phase is known. From
the well-known generalized pressure drop correlation (GDPC) for
packed columns by [23], the gas flux at flooding is calculated from
the packing factor, the flooding capacity parameter, the density
from both phases, and the liquid phase viscosity. Moreover, the
flow parameter is a function of mass flow rate and the density of
the fluids. The same properties are used to modify the GPDC corre-
lation proposed by Strigle for a more accurate design of the new
generation of packing [24].
The pressure drop of the two-phase flow in a packed column
can be obtained from GDPC graphs when the flow parameter and
the capacity parameter are calculated. The densities of both phases
are required to calculate the flow parameter. For capacity parame-
ter calculation, both phase density and liquid phase viscosity are
required. Numerical method are also used to calculate the pressure
drop, such as the methods proposed by, for example [25,26].
According to [25], the pressure drop inside the desorber varies
with the geometric surface and the void volume of the column
packing, the gas load factor, the wall factor and the resistance coef-
ficient. Because the free-cross section for the gas flow is reduced by
the column holdup, and the surface structure changes as a result of
coating with the liquid film, the column holdup must be consid-
ered in performing the pressure drop calculation. To obtain the
pressure drop by this method, the density and viscosity of both
phases are required. To use the method purposed by [26], the liq-
uid phase density and viscosity are required to calculate the liquid
loading factor, while the gas phase density is required to calculate
the gas loading factor. The gas loading factor is used to calculate
the pressure drop of the dry packing bed. The pressure drop
of the wet packing can be estimated subsequently from the
pressure drop of the dry packing bed together with the liquid load-
ing factor [21].
2.1.2.2. Packing height. To design the packing height of the absorber
and desorber columns, electrolyte thermodynamics, chemical reac-
tions, and heat and mass transfer across the gasliquid interface
and in the bulk gas and liquid phases must be considered. Mathe-
matical models to design the packing height have been developed
in several studies. In general, to calculate the CO2 absorbed by
amine solution, the gas/liquid-phase mass transfer coefficients,
the effective interfacial area, the enhancement factor, and the
phase equilibrium of the system are required [21,27,28]. Onda
et al. [29,30] developed the first and still widely used correlations
for mass transfer and effective interfacial area based on wetted-
wall columns for random packing. The gas phase mass transfer
coefficient is directly proportional to the diffusivity and density,
but inversely proportional to the viscosity. For liquid phase, the
effective interfacial area is considered for completing the mass
transfer estimation. This parameter is obtained from a detailed
investigation of the effect of hydrodynamics and liquid physical
properties, namely density, viscosity and surface tension. There-
fore, the liquid phase mass transfer coefficient is directly propor-
tional to the diffusivity, density and surface tension, but
inversely to the viscosity. Several correlations proposed by other
studies [31,32] for random packing show similar relationships
between mass transfer correlations and the fluid properties. For
structured packing, in the correlations proposed by [33,34], the
gas phase mass transfer coefficients vary with both gas and liquid
phase properties. They are directly proportional to the diffusivity
and densities of the gas phase, but inversely proportional to the
liquid phase density and the viscosity of both phases. Moreover,
the liquid phase coefficient is directly proportional to the diffusiv-
ity, but inversely proportional to the viscosity and the density.
2.1.2.3. Simulation and operation. Models to simulate the columns
are developed based on mass and energy balances. To calculate
208 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222

Table 1
Existing studies of property impacts on chemical absorption process.

Property Phase Impact Study Mixture Working Condition Ref.

Density Liquid Temperature profile and CO2 Comparison of correlations by [42,43] CO2/MEA/H2O 313350 K [44]
recovery rate
Temperature profile, rich Comparison of correlations by [42,43] 316350 K [45]
loading, and CO2 recovery rate
Gas Required packing height Variation from
20% to +20% of the base CO2/H2O/N2/O2 313351 K, 0.100.11 MPa [46]
case values
Heat capacity Liquid Temperature profile, and CO2 Variation of the calculated value from CO2/MEA/H2O 313350 K [44]
recovery rate temperature-dependent correlation
Temperature profile Variation of the calculated value at
15% 288357 K [47]
and
20%
Temperature profile, rich Comparison of correlations by [42] with 316350 K [45]
loading, and CO2 recovery rate constant values applied by [44]
Viscosity Liquid Temperature profile, rich Comparison of correlations by CO2/MEA/H2O 316350 K [45]
loading, and CO2 recovery rate [44,48,43,49]
Gas Required packing height Variation from
20% to +20% of the base CO2/H2O/N2/O2 313351 K, 0.100.11 MPa [46]
case values
Diffusivity Liquid Rich loading and CO2 recovery Comparison of calculated values from CO2/MEA/H2O 316350 K [45]
rate different viscosity correlations using the
Modified Einstein correlation
Gas Required packing height Variation from
20% to +20% of the base CO2/H2O/N2/O2 313351 K, 0.100.11 MPa [46]
case values
Phase equilibrium Liquid Heat of absorption Comparison of the e-NRTL and CO2/MEA/H2O 313393 K [50]
DeshmukhMather models

the mass balance, the mass transfer of gas absorption is affected by
the chemical reaction. The Hatta number compares the maximum
chemical conversion in the mass transfer film to the maximum dif-
fusion flux through the liquid film. This number is directly propor-
tional to the diffusion coefficient, but inversely proportional to the
liquid phase mass transfer coefficient of CO2 in aqueous amine
solution. It is further used to estimate the enhancement factor nec-
essary for estimating the volumetric mass transfer in the liquid
phase [35]. The phase equilibrium of CO2 and aqueous solvent sys-
tem are also essential to calculate the mass transfer rate along the
column height. The Henry constant method and more rigorous
methods such as the Electrolyte Nonrandom Two-liquid (e-NRTL)
activity coefficient model and Universal Quasi Chemical
(UNIQUAC) activity coefficient model have been used to retrieve
the phase equilibrium of the system [28,36,37]. To calculate the
energy balance, the temperature and pressure profiles along the
column must be accurately simulated. The heat transfer coefficient,
heat capacity and effective interfacial area are required. The heat
transfer coefficient can be obtained from dimensionless analogies.
In [21], the ChiltonColburn dimensionless analogy is used, which
requires the parameters of diffusivity, heat capacity, thermal con-
ductivity, density and mass transfer coefficient of the gas phase
to perform the calculation.
To estimate the energy demand of the reboiler, the phase equi-
librium of the CO2amineH2O system, heat capacity and thermal
conductivity are important. The total energy demand for regenera-
tion consists of the following three thermal energy parameters: the
sensible heat for solvent heating (Qsen), the latent heat for water
evaporation (Qlatent) and the heat of reaction for the desorption of
CO2 (Qrxn). The heat capacity and thermal conductivity are required
in the enthalpy balance for estimating the required Qsen. The phase
equilibrium is required to determine the liquid phase driving force
and solution speciation [36,38,39]. Therefore, the phase equilib-
rium is required to estimate the Qlatent, because the amount of
water content at the top of the desorber can be estimated using
the relationship between the partial pressure of CO2 and water
using the top column pressure. It is also important for estimating
the Qrxn with the GibbsHelmholtz equation [40,41].
2.2. Existing studies of property impacts on chemical absorption
Based on the discussion in Section 2.1, the properties required
for the design, simulation and operation of the chemical absorption
process are summarized in Table 1, together with the existing
studies on the property impacts on chemical absorption.
2.2.1. Density
Kvamsdal and Rochelle [44] studied the impact of liquid phase
density on the liquid temperature profile and CO2 recovery rate by
comparing the results of the temperature bulge of the liquid phase
in the packed column absorber calculated from a rate-based
absorption model using different liquid phase density correlations
proposed by [42,43]. An 11% difference in density was observed to
affect the magnitude and location of the temperature bulge. How-
ever, the difference in CO2 recovery rate between the two cases
was found to be less than 1%, and no difference was detected in
the liquid temperature at the bottom of the column.
The impacts of density on a rate-based CO2 absorber model
were further studied by [45]. The effects of density on the simu-
lated column performance expressed by the rich stream tempera-
ture, the rich stream loading and the CO2 recovery rate were
investigated. Using different liquid phase density correlations pro-
posed by [42] based on an unloaded solution and a correlation by
[43] produced significant deviations in the liquid temperature pro-
files, particularly in the area near the temperature bulge. However,
the deviation in density impacts the rich CO2 loading and CO2
recovery rate slightly differently. The deviation on the rich stream
temperature, the rich loading and the recovery rate between the
two cases were 0.7%, 0.2% and 0.7%, respectively.
For the gas phase, the influence of the gas phase properties on
the design of a random packing height for chemical absorption
by MEA was studied by [46]. The work calculated the required
packing height of the absorber based on a combination of the reac-
tive absorption process model and the integral rate based model
and then conducted a sensitivity analysis of the gas phase
properties by varying those property values from +20% to
20%.
It was found that a 5% overestimation of density results in a 2%
 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222
Fig. 3. Process flow sheet of physical absorption with two-stage flash desorption
(adapted from [55]).
underestimation of packing height. However, from our ongoing
study, the reported impact of gas phase density in [46] is found
to be overestimated.
2.2.2. Heat capacity
Kvamsdal and Rochelle [44] investigated the impact of the liq-
uid phase heat capacity by comparing three constant values based
on a temperature-dependent correlation. The differences in the
value of the heat capacity significantly impact the temperature
profile of the liquid phase along the column height, including to
the location and magnitude of the bulge. Moreover, a reduction
of 40% and 77% in heat capacity reduced the CO2 recovery rate
by 1.7% and 4.9%, respectively.
Faramarzi et al. [47] studied the impacts of liquid phase heat
capacity by varying the predicted values at
15% and
20%. They
concluded that the variation of liquid phase heat capacity signifi-
cantly impacts the temperature profile, not only its magnitude,
but also its location. The reduction in heat capacity clearly
increases the magnitude of the temperature bulge at the rich end
of the column.
Kvamsdal and Hillestad [45] studied the impact of liquid phase
heat capacity on a rate-based CO2 absorber model. The calculated
results as a function of temperature using the correlation by [42]
were compared with the constant value applied by [44], and the
former were found to yield approximately 16% higher heat capac-
ity values. As with the density, the effect of different heat capacity
correlations on the temperature profile was significant, particu-
larly in the area near the temperature bulge. However, the impacts
on the other parameters were insignificant. The 16% difference in
heat capacity produced deviations in the rich stream temperature,
the rich stream loading, and the CO2 recovery rate that were 3.0%,
0.5%, and 1.3%, respectively.
2.2.3. Viscosity and diffusivity
Kvamsdal and Hillestad [45] studied the impact of viscosity and
diffusivity on the rich stream loading, the temperature and the CO2
recovery rate. Different values from four liquid phase viscosity cor-
relations proposed by [44,48,43,49] were used. The calculated vis-
cosities were further used to estimate the diffusivity values with a
modified Einstein correlation. For viscosity, the deviations in the
simulated absorption process using correlations of [48,43,49] com-
pared with the correlations by [44] for the same simulation were
approximately 1% for the rich stream temperature and the rich
209
stream loading and 23% for the CO2 recovery rate. No quantitative
results for the diffusivity impact were reported. However, the
researchers concluded that the simulation results seem to be
slightly more sensitive to variations in the liquid phase viscosity
and diffusivity than to variations in the other properties.
Nookuea et al. [46] studied the influence of gas phase viscosity
and diffusivity on absorber design. The viscosity was found to have
an opposite impact to that the impacts of density and diffusivity.
However, the impact magnitude was in the same range as that of
density. The diffusivity was found to exert the most significant
effect on the packing height. A 5% overestimation of viscosity
was found to spur a 2% overestimation of the packing height,
whereas a 5% overestimation of diffusivity results in a 3% underes-
timation of the packing height. However, from our ongoing study,
the reported impacts of gas phase viscosity and diffusivity in [46]
are found to be overestimated.
2.2.4. Phase equilibrium
In the case of the thermodynamic properties, Kim et al. [50]
compared the heat of absorption from the different thermody-
namic methods used to retrieve the phase equilibrium data neces-
sary for calculating the activity coefficients of the species. The
values from the e-NRTL model and the DeshmukhMather models
at 313, 353, and 393 K were compared. The enthalpy of the reac-
tion is important for evaluating the heat of absorption, which heat
is directly related to the steam requirements for the solvent regen-
eration process. It was found that both models yield acceptable
prediction of the heat of absorption compared with experimental
data. The e-NRTL model generates a better fit to the experimental
heats of absorption than does the DeshmukhMather model, par-
ticularly at high values of CO2 loading. However, the quantitative
result of the property impacts on the heat of absorption was not
reported.
3. CO2 capture by physical absorption
3.1. Process description and related properties of physical absorption
3.1.1. Process description of physical absorption
Physical absorption is also a leading commercial option for CO2
capture technology, particularly in natural gas purification and
pre-combustion capture. Unlike chemical absorption, the absorp-
tion mechanism of physical absorption occurs in one step, which
dissolves CO2 into the liquid solvent [51]. The main advantage of
physical absorption is that the absorption is not limited by the
fixed stoichiometry of the reaction as in the case of chemical
absorption. The processes are proven to be efficient and economi-
cal when operated at higher pressures and lower temperatures
[52,53]. Compared with chemical absorption, less energy is
required for physical absorption, and it is also more profitable to
treat the feed gas with a high CO2 concentration [51]. Physical
absorption mainly consists of two steps: CO2 absorption and sol-
vent regeneration by desorption and dehydration [54]. Due to the
high vapor pressure of the solvent, the absorption process must
be operated at a low temperature and a high pressure to avoid sig-
nificant solvent loss at a high temperature while maintaining the
absorption capacity. The high pressure syngas is first cooled to
the design absorption temperature before being absorbed to the
solvent that has a high solubility for CO2 in the absorber column.
The reaction of CO2 absorption in the physical solvent is an
exothermic reaction. Therefore, the rising of solvent temperature
inside the column must be controlled. The CO2 loaded solvent
can be regenerated with three different methods: flash at reduced
210 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222
Table 2
Existing studies of property impacts on physical absorption process.
Property Phase Impact
Liquid phase density, Liquid/gas Top column
heat capacity, temperature, pressure,
enthalpy, and phase stream flow rate and
equilibrium composition
Stage temperature and
stream composition
Fig. 4. Flow diagram of a conventional PSA CO2 capture process (adapted from [63]).
pressure, as shown in Fig. 3; stripping; or distillation [53,54]. The
process used depends on the required purity of the treated gas.
3.1.2. Key components and related properties of physical absorption
The key components of physical absorption are the flue gas
absorber, regeneration units, and the refrigeration system. The pro-
cess power and energy demand are located primarily in the refrig-
eration system, the feed gas compressor and the solvent pump.
3.1.2.1. Column design. The column design for physical absorption
is similar to that used in chemical absorption, but less complicated.
For example, the flooding velocity and the pressure drop are
required to design the packed column diameter, and mass transfer
coefficient of both the gas and liquid phases are required to design
the packing height. To determine the liquid phase mass transfer
coefficient in packed columns, the values of the diffusivity, density,
viscosity and surface tension are necessary. In addition, the phase
equilibrium of the system is important because the saturated
absorption amount of the solute in the solvent depends on the
vaporliquid equilibrium of the system.
3.1.2.2. Flash regeneration. The CO2 loaded solvent can be regener-
ated with three different methods: flash at reduced pressure, strip-
ping, and distillation [53,54]. The process used depends on the
required purity. Because the design of the desorber column is sim-
ilar to the design of the absorber column, identical properties are
required. For the flash regeneration, the rich solvent is flashed
through several stages to reduce the solubility at a lower partial
pressure. Therefore, the phase equilibrium of the mixture is
important.
3.1.2.3. Operation. The operation of the physical absorption process
occurs at a very low temperature, particularly in the Rectisol pro-
cess [52,56]. The main energy demands of this process are made
by the refrigeration system, including the feed gas and the solvent
transportation. The required heat transfer area and the cooling load
Model Mixture Working condition Ref.
PR, PSRK, RK-ASPEN, CO2/CH3OH/H2/N2/H2S/CH4/CO/COS 223254 K [53]
and PC-SAFT and 3.34.4 MPa
Modified PC-SAFT EOS
PSRK, RK-ASPEN, CO2/CH3OH/H2/N2/H2S/H2O/Ar/CH4/CO/COS 245 K [60]
RKSWS, SR-POLAR, 3.63 MPa
PRWS, KSMHV2 and
PRMHV2
are determined by the heat transfer coefficient, which is related to
the thermal conductivity, the heat capacity, the density, and the
viscosity of the fluids in the heat exchanger [57,58]. For shell and
tube heat exchangers, the tube-side heat transfer coefficient is a
function of the Reynolds number, the Prandtl number, and the tube
diameter. The Reynolds number and the Prandtl number can be
unpacked into the parameters of density, viscosity, thermal con-
ductivity, heat capacity, and superficial velocity of the stream [59].
3.2. Existing studies of property impacts on physical absorption
Table 2 lists the existing studies of property impacts on physical
absorption processes. Sun and Smith [53] assessed the impacts of
thermodynamic properties, which were liquid phase density, heat
capacity, enthalpy, and phase equilibrium, on the operation of the
Rectisol process. However, the assessment did not consider the
impact of each property separately, and the deviation in property
values from each model was not mentioned. The impact results
of implementing different property models on the top column tem-
perature, the pressure, the mole flow and the mole fraction were
reported. Among the evaluated models, the Perturbed-Chain Sta-
tistical Associating Fluid Theory equation of states (PC-SAFT EOS)
yielded simulation results with the smallest differences compared
with the design data, except the CO2 concentration of the output
purified syngas (approximately 15% error) and the mole flow of
the rich solvent (approximately 30% error). However, a revised
PC-SAFT presented in the same study provides better prediction
results, yielding, for example, only a 0.07% error in the CO2 concen-
tration of the output purified syngas and a 4.9% error in the mole
flow of the rich solvent.
Liu et al. [60] stated that the results from their preliminary
study show that most of the possible thermodynamic models in
commercial process simulators, such as ProII with Provision and
ASPEN, do not generate workable results for the Rectisol absorp-
tion process applied to raw gases compared with on-site data. It
is because the system is operated under extreme condition, there-
 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222
Fig. 5. Two-stage membrane system for CO2 capture from flue gas (adapted from [94]).
fore the systematic thermodynamic behavior is far from normal.
However, the uncertainty of the models in the commercial simula-
tors were not reported. The work evaluated the impacts of thermo-
dynamic properties on the stage temperature and stream
composition for the CO2 absorption units using different property
models to conduct the process simulation. However, as in [53],
the quantitative impact of each property was not presented.
4. CO2 capture by physical adsorption
4.1. Process description and related properties of physical adsorption
4.1.1. Process description of physical adsorption
Physical adsorption is another technology for facilitating CO2
capture from flue gas using a solid adsorbent. Pressure/vacuum
swing adsorption (PSA/VSA) and temperature swing adsorption
(TSA) are the two main feasible adsorption technologies for
large-scale CO2 separation and purification [61,62]. In a typical
adsorption process, as shown in Fig. 4, the flue gas is compressed
and sent to the adsorption column. The CO2 in the gas stream is
captured by the adsorbent, and the impurities are expelled from
the top of the adsorption tower. After the adsorption process, the
CO2 is separated, and the adsorbent is regenerated in the desorp-
tion tower. The regenerated adsorbent is reused, and the CO2 is
recovered and undergoes further processing [63,64]. The typical
adsorption system consists of the following main components:
the compressor, the cooler, the adsorption column, and the desorp-
tion column. The compressor is an important component because
it demands the most energy in the adsorption process and will
be elaborated in the paper when discussing CO2 conditioning.
The adsorption column is also a significant component of the pro-
cess, and its design and operation greatly impacts the performance
of the adsorption rate and the purity of the CO2 product [6567].
4.1.2. Key components and related properties of physical adsorption
and existing studies
In most adsorption processes, the adsorption of the fluid on the
adsorbent occurs in a fixed bed. An accurate mathematical model
for predicting the performance of the adsorptive capture of CO2
is necessary for process design and optimization [68,69]. A
comprehensive model consisting of coupled partial differential
equations of mass, energy and the momentum balance with
equilibrium equations is necessary [61]. Different properties are
required. Specifically, in the mass balance of the adsorption pro-
cess, the gassolid adsorption equilibrium is the most important
211
parameter; it controls the dynamic behavior and the adsorption
capacity of a packed column. This parameter also determines the
mass transfer coefficient. According to the existing models [70
72], the mass transfer coefficient is a function of the Sherwood
number, which is further related to the Reynolds number and the
Schmidt number. Therefore, the properties of CO2 mixtures, such
as their density, viscosity, and diffusivity, are required for deter-
mining the mass balance of the adsorption column. In determining
the energy balance in the gas phase, it should be noted that the
energy flow consists of a heat transfer between CO2 mixtures and
the surface of the solid adsorbent as well as the energy transferred
to the column wall. The energy balance can be expressed by an
energy conservation equation and is related to the heat transfer
coefficient, the bulk density of the CO2 mixtures, and the heat dis-
persion. To calculate the heat transfer coefficient in the adsorption
column, the properties of CO2 mixtures such as their density, vis-
cosity, thermal conductivity, and heat capacity are necessary
[7375]. In determining momentum balance, the profile of pres-
sure and velocity can be obtained from the momentum conserva-
tion equations [7678] and the Ergun equation [79]. The pressure
drop of the CO2 mixtures due to viscous energy losses can also
be estimated. The pressure drop could reduce the working capacity
and is also required for performing the compression work. The
pressure drop of the structures in the column can be expressed
by the viscosity, density, velocity and properties of adsorbents
[80,81].
To summarize, to accurately design the adsorption column, the
aforementioned properties of CO2 mixtures, including the density,
viscosity, heat capacity, phase behavior, thermal conductivity, and
diffusivity, are required. However, to the best of the authors
knowledge, there is no research on the property impacts of the
CO2 mixtures on CO2 adsorption column design and operation.
By comparison, the properties of the adsorbent as well as improved
adsorption column models have received greater attention. There-
fore it is essential to study the property impacts on CO2 adsorption
design and operation.
5. CO2 capture by membrane separation
5.1. Process description and related properties of membrane
separation
5.1.1. Process description of membrane separation
The polymeric and organic membranes are two types of mem-
brane used in post- and pre-combustion capture. The polymeric
212 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222
Fig. 6. Flow diagram of the CO2 conditioning process for pipeline transport (adapted from [97]).
membranes are predominantly used due to their higher selectivity
compared with organic membranes, which selectivity gives the
permeate gas a higher purity [8284]. However, it is noticeable
that the use of inorganic membranes such as zeolite and ceramic
membranes have become more interesting because they have
higher gas permeability and greater chemical, physical, and
mechanical stability [85]. In using a polymeric membrane, CO2 is
separated from the flue gas by a solutiondiffusion mechanism.
The mechanism starts with the sorption of CO2 in the membrane
structure, followed by diffusion through the membrane, and ends
with desorption on the permeate side of the membrane [82,86].
More complicated mechanisms are involved in separation using
inorganic membranes. The gas separation by inorganic membranes
is governed by molecular sieving, surface diffusion, Knudsen diffu-
sion, and capillary condensation mechanisms [87,88].
As shown in Fig. 5, the separation process generally consists of
compressors, membrane modules, vacuum pumps, heat exchang-
ers and expanders [8993]. In post-combustion capture, a low con-
centration of CO2 flue gas enters the separation process at ambient
pressure. Compressors and vacuum pumps are required to main-
tain the separation driving force, which is a partial pressure differ-
ence between the feed and the permeate side. The compression
increases the gas temperature significantly. Therefore, before
entering the membrane modules, the gas must be cooled to the
optimal operating temperature, between 40 C and 70 C for poly-
meric membranes, though cooling might be unnecessary for inor-
ganic membranes [84]. The expander is used to recover some work.
5.1.2. Key components and related properties of membrane separation
and existing studies
Because the membranes used for CO2 removal operate on the
principle of solutiondiffusion through a non-porous polymeric
membrane, the separation is based on how well different com-
pounds dissolve into the membrane and then diffuse through the
membrane. The selectivity and permeability of membranes are
important properties, which indicate the purity and degree of sep-
aration of the CO2 [90,95]. The solubility of gases in membranes
and the diffusion coefficient of the separated species through the
membranes are the key properties of this separation process, both
of which properties vary with the gas and polymer type [96].
The main capital cost of membrane separation process depends
considerably on the required membrane surface area. The poly-
meric membrane permeability depends on working conditions
and the mixed gas composition. Zhanat and Sharipzhan [95]
derived a method for calculating the membrane surface area
required for mixture separation. The gas flux through the mem-
brane is a key parameter used in the calculation and is directly pro-
portional to the diffusion coefficient of CO2 through the polymer.
Similar to the physical adsorption, to the best knowledge of the
authors, no studies have been conducted on the property impacts
of the CO2 mixtures on the membrane CO2 separation processes.
6. CO2 conditioning
6.1. Process description and related properties of CO2 conditioning
6.1.1. Process description of CO2 conditioning
CO2 conditioning is the process during which the temperature
and pressure of CO2 streams are adjusted for the next operational
step. The design of a conditioning system must carefully consider
the composition of CO2 mixtures and process constraints, including
the working condition and impurity limits. In general, as shown in
Fig. 6, which represents the flow sheet of the CO2 conditioning pro-
cess for pipeline transport, the conditioning system consists of the
following processes: the compression and cooling process, water
removal by separator drums and adsorption, the removal of
unwanted components by chemical or physical treatment, CO2
condensation by external cooling system or by expansion via
JouleThomson valves, the removal of volatile gases by flash tanks
or distillation columns, and the pumping process to achieve trans-
port pressure [97]. Among the components in the CO2 conditioning
processes, compressors and pumps, as well as heat exchangers for
CO2 condensation and liquefaction, are the most important compo-
nents because they require the greatest amount of energy and
investment [3,98102]. Therefore, increased attention should be
given to the compression/pumping process and the CO2 condensa-
tion process. In CO2 conditioning, the property impacts will be sep-
arated and illustrated by two parts: the compression/pumping
process and the condensation/liquefaction process.
6.1.2. Key component and related properties of CO2 conditioning
6.1.2.1. Compressor and pump. Compressors and pumps are used to
elevate the pressure of gaseous CO2 and liquid CO2, respectively.
There are many types of industrial compressors, including axial,
positive displacement, and centrifugal compressors [103]. It is pos-
sible to use either a centrifugal or an axial compressor for the first
stage. For the later stage, centrifugal compressors appear to be the
most appropriate choice [104]. For pumps, most current pumps are
not used for pumping CO2 or operating at cryogenic conditions, but
 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222 213

Table 3
Existing studies of property impacts on compression and pumping processes.

Property Impact Model Mixture Working condition Ref.

Heat capacity Compression work/discharging PR, PT, RK, SRK, 3P1T CO2/Ar/O2/N2/SO2 219423 K, 011 MPa [107]
temperature
Pump efficiency REFPROP/PROPATH 298311 K, 6.759.25 MPa [111]
Pressure ratio/choking Pure Ar/CO2/Air 0.82 MPa [106]
mass/compressor efficiency
Speed of sound Energy requirement and cost CO2/O2/N2/Ar/H2/CO/ 273323 K, 15 MPa [112]
H2S/CH4
Density Compression work/discharging PR, PT, RK, SRK, 3P1T, MPR, MSRK, CO2/Ar/O2/N2/SO2 219423 K, 011 MPa [107]
temperature ISRK
Compressor outlet temperature & IDG, LKP, LK, PR, SRK CO2/H2O/O2 3001800 K, 0.110 MPa [113]
compression work
Pump efficiency REFPROP/PROPATH 298311 K, 6.759.25 MPa [111]
Phase Compression work/discharging PR, PT, RK, SRK, 3P1T CO2/Ar/O2/N2/SO2 219423 K, 011 MPa [107]
characteristic temperature
Enthalpy & Compression work/discharging PR, PT, RK, SRK, 3P1T CO2/Ar/O2/N2/SO2 219423 K, 011 MPa [107]
entropy temperature
Compressor outlet temperature & IDG, SRK CO2/H2O/O2 3001800 K, 0.110 MPa [113]
compression work
Pump efficiency REFPROP/PROPATH 298311 K, 6.759.25 MPa [111]
Viscosity Energy requirement and cost CO2/O2/N2/Ar/H2/CO/ 273323 K, 15 MPa [112]
H2S/CH4
Thermal Energy requirement and cost CO2/O2/N2/Ar/H2/CO/ 273323 K, 15 MPa [112]
conductivity H2S/CH4
Compressibility Mass flow per unit area LK, SW Pure CO2 304433 K, 7.420.3 MPa [114]
Energy requirement and CO2/O2/N2/Ar/H2/CO/ 273323 K, 15 MPa [112]
compressor cost H2S/CH4
Polytropic head & efficiency, gas RK, BWRS, LKP, PR 0.783.10 reduced T, 0.0550.5 [115]
horsepower, inlet volumetric flow reduced P

might be adaptable [104]. The CO2 compression and pumping pro-
cess requires significant amounts of energy to achieve a final pres-
sure for transport or re-injection. Therefore, it is crucial to
accurately estimate the performance of the compression and
pumping process. The key properties in CO2 compression and
pumping processes are density, speed of sound, specific heat, phase
characteristic, heat capacity, enthalpy, entropy, phase equilibrium,
viscosity, thermal conductivity, and compressibility [105108]. For
example, the influence on the phase equilibrium of mixtures
resulting from the variation of impurity content has been identified
[107]; the operating conditions of the CO2 compression must be
changed accordingly. The density and speed of sound can impact
the selection of impellers (flow coefficient and peripheral Mach
number) [108]. The heat capacity, enthalpy, and entropy are
important parameters because they affect the heat transfer and
energy demand of the CO2 compression processes [107]. In addi-
tion, the specific heat ratio is regarded as one of the four key
non-dimensional parameters that affect the performance of a tur-
bomachine [106]. The transport properties of CO2 mixtures, such
as viscosity and thermal conductivity, are also important for eval-
uating the compression processes [107].
6.1.2.2. Heat exchanger. The heat exchanger is the key component
in CO2 condensation and liquefaction processes. In the heat
exchanger, the CO2 streams can be cooled by external or internal
cooling sources. Specifically, refrigeration systems, for example,
with ammonia as the working fluid, can be used as external cooling
sources [3]. In internal cooling, heat transfer is conducted between
CO2 streams, and JouleThomson valves are used for CO2 mixture
expansion to cool the mixture to a low temperature [99]. In CO2
condensation and liquefaction processes, parameters such as the
liquefaction rate, the pressure drop, and the heat transfer coeffi-
cient are key in designing heat exchangers. Different parameters
require different properties. Specifically, the viscosity and density
are relevant to the pressure drop, which in turn affects the satura-
tion temperature [109]. In addition, the liquefaction rate is influ-
enced by the temperature gradients between the vapor and
cooling media as well as by the thermal conductivity and heat
capacity [109]. The heat transfer coefficient is significant for deter-
mining the heat transfer area and the cost of heat exchanger. The
heat transfer coefficient is the function of the Reynolds number
and the Prandtl number, and can be calculated based on knowl-
edge of the density, viscosity, thermal conductivity and heat capac-
ity [110].
6.2. Existing studies of property impacts on CO2 conditioning
6.2.1. Property impacts on compression and pumping
Table 3 lists the existing studies on property impacts on the
compression and pumping process.
6.2.1.1. Heat capacity. Li [107] evaluated the impurity effect on the
heat capacity and further impacts on the compression process, the
discharging temperature and the work required for compressing
different CO2 streams isothermally at different outlet pressures
and different CO2 compositions. The compression work and dis-
charging temperature were found to decrease with the increment
of heat capacity. The two property databases REFPROP (reference
fluid properties) and PROPATH (program package on the
thermos-physical properties of fluid) were compared concerning
the determination of heat capacity by [111], and the effect of prop-
erty deviation on pumping efficiency was discussed. For the heat
capacity model, up to a 2% deviation was found between the data-
bases near the saturation line, and an increase in deviation was
observed near the critical point, thus resulting in a maximum
12% deviation in pumping efficiency for the 1.5 kg/s mass flow rate
case. Roberts and Sjolander examined the effect of the specific heat
214 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222
Table 4
Existing studies of property impact on condensation and liquefaction processes.
Property Impact
Heat capacity Refrigeration energy consumption
Density Exhaust gas enthalpy drop of economizer and condenser
Enthalpy & entropy Exhaust gas enthalpy drop of economizer and condenser
ratio (c) on the aerodynamic performance of compressors experi-
mentally and numerically [106]. The performance of a centrifugal
compressor stage was measured and simulated for air (c = 1.4),
CO2 (c = 1.29) and Ar (c = 1.67). The results showed that in the
middle of the mass flow range (mass flow parameter from 0.04
to 0.12), a higher c leads to a higher pressure ratio. Performance
differences are predominantly due the differences in c, and the
maximum deviation in pumping efficiency could reach 5%.
6.2.1.2. Speed of sound. The properties of CO2 mixtures, speed of
sound included, were calculated in different impurity scenarios
to evaluate the impurity impacts and property impacts on com-
pression process by [112]. It was found that the CO2 composition
significantly affects the speed of sound, with a maximum deviation
of 19% at 15 MPa, but has a relatively minor effect on CO2 compres-
sion power, with a maximum deviation less than 7%. Nonetheless,
given that all properties change during the simulation, the impact
of a single property on the compression process remains unclear.
6.2.1.3. Density. The density impact on discharging temperature
and compression work was studied by [107]. It was found that
the discharging temperature and compression work decrease
when density increases. The findings implied that a deviation of
515% in calculating the density could result in an error in com-
pression work of 3.86.5%. Ulfsnes et al. [113] evaluated the impact
of the density on the compression process and compared the found
impact with collected experimental data. The compressor outlet
temperature and compression work were calculated to analyze
property impacts on the compression process. The results showed
that the maximum deviation between calculations with different
models was approximately 2% for both the compressor outlet tem-
perature and compression work. This deviation is small and not
vital for achieving calculation accuracy in compression processes.
The pumping efficiency was also evaluated by [111]. It was found
that up to a 2% of deviation exists in different models and that the
maximum deviation of pump efficiency calculated with different
property databases can reach 12% in the 1.5 kg/s mass flow rate
case [111].
6.2.1.4. Phase characteristic. The impacts of phase equilibrium and
phase diagram on performance of compressor were studied by
[107]. It was found that with increases in the boiling pressure
and condensing pressure of CO2 mixtures, the compression work
and the discharging temperature also increase.
6.2.1.5. Enthalpy and entropy. The enthalpy of CO2 mixtures was
evaluated using different equations of state (EOS) by [107,113]
and by using different property databases by [111]. The property
impacts on the compression process concerning discharging tem-
perature and required compression energy [107,113], as well as
pumping efficiency [111], were evaluated. The maximum deviation
of enthalpy and entropy using different models was found to be 2%,
which causes a 12% difference in pumping efficiency [111]. How-
ever, for the compression process, the maximum deviation
between calculations with ideal gas law and SRK (SoaveRdlich
Kwong) is only approximately 2% for both the compressor outlet
temperature and the compression work [113].
Model Mixture Working condition Ref.
PR, PT, RK, SRK, 3P1T CO2/Ar/O2/N2/SO2 219.15423.15 K, 011 MPa [107]
IDG, LKP, LK, PR, SRK CO2/H2O/O2 3001800 K, 0.110 MPa [113]
IDG, SRK CO2/H2O/O2 3001800 K, 0.110 MPa [113]
6.2.1.6. Viscosity. The synergic effects of properties, including vis-
cosity, were conducted by [112] on compression power. The devi-
ation of compression work was found to be less than 7%, with
variations of different properties caused by expected amounts of
different impurities in CO2 mixtures. Among the variations, a max-
imum variation of viscosity of 25% at 15 MPa was mentioned.
6.2.1.7. Thermal conductivity. The impacts of thermal conductivity
and other properties were evaluated by [112]. The properties
change alongside variations in the impurities of CO2 mixtures,
and the deviation of compression work ranges from 2.6% to 6.8%
with different impurities. The maximum deviation of thermal con-
ductivity is 15% at 15 MPa. Because the property impacts were
evaluated together with other properties, the single property
impact of thermal conductivity is unclear.
6.2.1.8. Compressibility. Baltadjiev [114] studied the impact of com-
pressibility on the isentropic relations, the definition of the speed
of sound, compressible flow equations, and the corrected mass
flow per unit area. Wetenhall et al. [112] also studied the com-
pressibility of different CO2 mixtures and the impact on compres-
sion performance and on other properties. Sandberg [115]
calculated and compared compressibility factors from four differ-
ent equations of state. The performance of the compressor was also
evaluated and compared using the values of the polytropic effi-
ciency and gas horsepower. It was found that compressibility is
modeled accurately, with an average discrepancy of approximately
1%, and that the isentropic exponent varies by less than 3%. There-
fore, real gas effects are expected to exert no significant impact on
the compressor performance and stage matching [114]. Wetenhall
et al. [112] also found that although the deviation of compressibil-
ity between different CO2 mixtures ranges from 8% to 30% at
15 MPa, the maximum deviation of compression work is less than
7%. However, it should be studied whether the deviation is caused
by a variation in compressibility or by a synergetic effect of other
properties. The deviation of the compressibility compared with
empirical values varied from
8% to 14%, in turn causing the devi-
ation of the polytropic head to vary from
4% to 12%, that of the
polytropic efficiency from
2.5% to 4%, the deviation of gas horse-
power from
1% to 5%, and the deviation of inlet volumetric flow
from
1% to 7% [104].
6.2.2. Property impacts on condensation and liquefaction
Table 4 lists the existing studies of property impacts on conden-
sation and liquefaction.
6.2.2.1. Heat capacity. Li [107] evaluated the impact of heat capac-
ity on refrigeration in a CO2 purification system and found that the
refrigeration duty decreases with decreases in the heat capacity of
CO2 mixtures with non-condensable gases, whereas the refrigera-
tion duty increases with increases in the heat capacity of CO2 mix-
tures with impurities such as SO2.
6.2.2.2. Density. Ulfsnes et al. evaluated the density models with
collected experimental data and calculated the compressor outlet
temperature and compression work to analyze the property
impacts on compression processes [113]. Relative errors in density
 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222
Table 5
Existing studies of the property impacts on transport process.
Property Impact
Density Transportation cost
Pressure drop/pipeline transport
Pressure loss & temperature drop
Viscosity Pressure loss & temperature drop
Phase characteristic Pressure drop
Heat capacity Temperature drop
Thermal conductivity Temperature drop
calculations with different EOS were found to be within 1%. The
errors cause between a 0.2% and 2.3% deviation in exhaust gas
enthalpy drop in the economizer and condenser.
6.2.2.3. Enthalpy and entropy. The impacts of enthalpy and entropy
on heat exchangers were studied by [113] in a O2/CO2 gas turbine
cycle with CO2 capture. The values of enthalpy and entropy were
calculated with a SRK model and were compared with experimen-
tal data. It was found that the maximum error is less than 0.8% and
that the maximum deviations in enthalpy drops over the econo-
mizer and condenser are 0.9% and 2.3%, respectively.
7. CO2 transport
7.1. Process description and related properties of CO2 transport
7.1.1. Process description of CO2 transport
CO2 is transported from capture plants to storage sites by
pipelines, ships or tanker trucks, with the selection of transport
method based mainly on the distance [116]. Pipelines today oper-
ate as a mature market technology and are the most common
method for transporting CO2. A CO2 stream is typically pumped
to a pressure greater than 8 MPa to increase the density of CO2
and to avoid two-phase flow regimes, thereby making it easier
and less costly to transport. Transport by ship could be cheaper
than pipeline transport under some circumstances, particularly
for longer distances, as ships can also be more flexible than pipeli-
nes [117,118]. The equipment used in ship transport includes an
intermediate storage tank, loading/unloading facilities, a transport
ship and CO2 carrier tanks. The carrier tanks can be generally clas-
sified into three types, pressurized, fully refrigerated and semi-
refrigerated, depending on the pressure and temperature of the
transported CO2. Optimal working conditions correspond to either
1.52.0 MPa and approximately 243 K [119,120] or 0.60.7 MPa
and approximately 223 K [4,121,122]. Both for pipeline transport
and ship transport, different properties are involved in designing
and operating the system.
7.1.2. Key component and related properties of CO2 transport
7.1.2.1. Pipeline transport. The pipeline transport capacity is one of
the first design criteria required for CO2 transport cost estimation.
Pipeline capacity is a function of both pipeline diameter and oper-
ating pressure, and pipelines must be appropriately sized based on
corresponding CO2 emission sources. The diameter calculation
method specified by the MIT Carbon Capture and Sequestration
Program shows that the pipeline diameter is a function of maxi-
mum allowable pressure drop, the CO2 mass flow rate, the CO2
density, and the Fanning friction factor. Therefore the key proper-
ties for determining the pipeline diameter are density and viscosity
[123].
215
Model Mixture Working condition Ref.
Corrected SRK CO2/O2/Ar/N2 273423 K, <125 MPa [124]
SRK, PR, PRSV, BWRS CO2/H2O/N2/CH4/Ar/SO2 273333 K, <20 MPa [137]
REFPROP <298 K, 815.2 MPa [138]
REFPROP <298 K, 815.2 MPa [138]
PR, RKS, BWRS, LK, GERG CO2/N2/Ar/O2/CH4 150350 K, <10 MPa [139]
REFPROP <298 K, 815.2 MPa [138]
REFPROP <298 K, 815.2 MPa [138]
For steady-state flow processes, the pressure and temperature
profile are important for transporting CO2 in its dense phases.
Therefore the most important parameters are pressure loss and
temperature drop, which are related to momentum and heat trans-
fer processes. The key properties are the phase characteristic, den-
sity, viscosity, thermal conductivity and heat capacity. The phase
characteristic is one of the important properties because CCS
streams are designed to be transported in their liquid or dense
phases; therefore, knowledge of the phase enveloping the CO2 mix-
ture is essential [124]. Thermal conductivity and heat capacity are
physical properties that govern heat diffusion in the steady state
and are important for temperature drop estimation [125]. In addi-
tion to heat transfer effects in the pipe flow, the main properties in
pressure calculations are the fluid viscosity and the fluid density
[126].
For the transient transport processes, in addition to the proper-
ties mentioned in steady flow process, other properties including
the speed of sound, diffusivity, and the JouleThomson inversion
curve are significant. The speed of sound is used for the prediction
of wave propagation in a pipeline, leakage tests, monitoring com-
positional changes, as well as 4D seismic studies [127130]. Diffu-
sivity is a physical property governing transient heat diffusion
[125]. The JouleThomson inversion curve is important for hazard
assessment studies of pipeline depressurization because it indi-
cates whether the outflow stream will follow a cooling or a heating
path upon pressure drops [131].
7.1.2.2. Ship transport. For ship transport, designing the cargo tank
or the CO2 carrier requires the values of the boiling temperature,
the density of the intended cargoes, and the internal pressure.
The design temperature is decisive for the choice of tank material
and the cargo density for the scantlings of the tanks and its support
[132,133]. To reduce the heat transfer from the surrounding envi-
ronment to the tank trough conduction, convection and radiation
and to keep the state stable inside the tank, the cryogenic insula-
tion of the cargo tank must be known. Selecting appropriate ther-
mal insulation material and determining insulation thickness
require knowing the thermal conductivity of fluids [134,135]. In
addition, the internal pressure of the tank is designed based on
the vapor pressure and liquid pressure [135], and the tank thick-
ness is affected by the density because greater densities require
thicker shells [136].
7.2. Existing study of the property impacts on CO2 transport
Table 5 lists the existing studies of the property impacts on the
transport process.
7.2.1. Density
Chapoy et al. [124] studied the effect of multi-component impu-
rities on the density and the impact on the transportation cost. The
results showed that the density of the mixtures can be as much as
216 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222

35% lower than that of pure CO2 at the same temperature and pres- Table 6
sure, which can have a significant impact on transportation costs. Existing studies of the property impacts on storage process.

Zhao and Li [137] investigated the impacts of nonpolar and polar Property Impact Model Mixture Working Ref.
impurities on the mass density at critical and supercritical temper- condition
atures and the property impact on pipeline transportation. It was Density Storage PR CO2/O2/Ar/N2/ 320350 K, [146]
found that higher transport temperatures and pressures will min- capacity H2 525 MPa
imize the chance that the density changes abruptly for Geological PR with Oxy-fuel 303393 K, [160]
storage exp. data mixtures 1050 MPa
supercritical-phase CO2 pipeline transportation. Tan et al. [138] Storage CO2/H2S/non- [161]
investigated the impacts of density on pressure loss and tempera- capacity condensable
ture drop in CO2 pipeline transport. A sensitivity analysis was con- Oil recovery [147]
ducted by varying the value of density and the maximum factor
deviations of the existing models of density were reviewed as well. Viscosity Geological PR with CO2/O2/Ar/N2/ 303393 K, [160]
This paper used the REFPROP [140] model to calculate the density storage exp. data H2/SO2 1050 MPa
Injectability Oxy-fuel 261367 K, [126]
and varied the value from
20% to 20% to perform the sensitivity
and cost mixtures <21 MPa
analysis. It was shown that overestimating the density increases Oil recovery [147]
the pressure loss and decreases the temperature drop, whereas factor
underestimating the density decreases the pressure loss and
increases the temperature drop. The maximum modeling devia-
tions for density were found to be 18%, and, correspondingly, a for the transport of CO2 mixtures by varying the thermal conduc-
maximum deviation of approximately 16% and 20% was found tivity from
20% to 20%. It was shown that the impacts of the ther-
for the pressure loss and temperature drop, respectively. From mal conductivity on the temperature drop are small and can be
the perspective of property modeling, this behavior shows that neglected.
developing more accurate density models should be prioritized.

7.2.2. Viscosity
Tan et al. [138] investigated the impacts of viscosity on pressure
loss and temperature drop in CO2 pipeline transport. A sensitivity
analysis was conducted by varying the value of viscosity from

20% to 20%. The results showed that overestimating the viscosity
increases the pressure loss, whereas underestimating viscosity
decreases it. In addition, pressure loss and temperature drop are
shown to not be sensitive to viscosity, and an overestimation of
the viscosity by 20% leads to an increment in pressure loss of
approximately 3% and a negligible variation in the temperature
drop.
7.2.3. Phase characteristic
Mazzoccoli et al. [139] compared the PqT experimental data
of CO2 binary mixtures with different EOS. Based on that compar-
ison, pipelines were simulated to illustrate the impacts of the pre-
dicted properties on the pressure drop. The pressure drops for
transportation in the supercritical region, in the dense region and
as a subcooled liquid were compared by using different EOS as Red
lichKwongSoave (RKS) and GERG. In the supercritical region, the
difference in pressure drop calculated by RKS and GERG is from
10.5% to 43.8%, in the dense region, the difference is from 11.8%
to 22.9% and in subcooled liquid, the difference is from 1.7% to
3.3%.
7.2.4. Heat capacity
To evaluate the impact of heat capacity on the transport pro-
cess, the temperature drop was calculated by varying the value
of heat capacity, as reported in the work of [138]. The accuracy
of the existing models of heat capacity was reviewed to study
the real deviation of the temperature drop. The value was varied
from
20% to 20% for the sensitivity analysis. The over-
estimation of the heat capacity leads to a lower temperature drop,
whereas the underestimation of the heat capacity results in a
higher temperature drop. For predicting the heat capacity, the
maximum modeling deviation is 5%; correspondingly, the
maximum deviation leads to the variation of the temperature drop
of 3%.
7.2.5. Thermal conductivity
Tan et al. [138] conducted a sensitivity analysis on the influence
of the thermal conductivity on the temperature drop in a pipeline
8. CO2 storage
8.1. Process description and related properties of CO2 storage
8.1.1. Process description of CO2 storage
CO2 storage can be classified as geological storage or ocean stor-
age. Geological storage of CO2 is accomplished by injecting it CO2 in
dense form into a rock formation below the earths surface [117].
The geological storage technologies can be sorted into different
categories, such as onshore/offshore storage. They can also be
sorted based on types of reservoirs. A number of reservoir types
are available worldwide, including (depleted) oil and gas fields, sal-
ine aquifers, unmined coal seams and cavities, such as coal mines
[141]. Specifically, CO2 can be injected into depleted oil/gas reser-
voirs to increase their pressure and provide the driving force for
enhanced oil recovery (EOR), whereas the injected CO2 remains
stored there permanently. CO2 also can be injected into deep coal
beds to recover methane that is trapped in the porous structure
of coal seams. In addition, injecting CO2 into deep saline aquifers
is one of three main options for implementing the geological stor-
age of CO2 to decrease anthropogenic greenhouse gas emissions
into the atmosphere [142]. Dispersion models are normally used
to study the plume dispersion due to a particular atmospheric con-
dition and for assessing the effect of plume dispersion on the envi-
ronment. In ocean storage, the captured CO2 is deliberately
injected into the ocean at great depth, where most of it will remain
isolated from the atmosphere for centuries [117]. Additionally, the
potential leakage from storage sites needs to be avoided and the
negative impacts such as ocean acidification require careful mea-
surements [143].
8.1.2. Key components and related properties of CO2 storage
8.1.2.1. Geological storage. The properties of the storage reservoir
and injectants are both significant for the CO2 geological storage
process. The former one includes the characteristics of the reser-
voir, including pressure and temperature profile, porosity, and per-
meability of the porous medium [144,145]. For the CO2 mixtures,
the injectants in the storage process have different properties that
are necessary for different technologies. For the geological storage
process, the efficiency of CO2 storage is commonly essential and is
 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222
Table 7
Status of the existing studies of the property impacts.
Property Process
Chemical Physical
absorption absorption
Thermodynamic property
p p
Density
Phase characteristic
p p
Heat capacity
Enthalpy and entropy
Transport property
p p
Viscosity
Speed of sound  
p
Diffusivity
Thermal conductivity
Surface tension  
Joule-Thomson inversion  
curve
Compressibility  
defined as the ratio of the volume accessible or occupied by CO2 in
a given pore volume of a porous medium. The volume of the CO2
mixture depends on its density, which is a function of the reservoir
pressure and temperature. In real cases, non-condensable impuri-
ties would increase the pressure for injection and reduce the
capacity of the storage sites by decreasing the density of the CO2
stream [146]. Therefore, density is an important property for effec-
tive storage [126]. Specifically, in the EOR process, the relevant
properties can be divided into thermodynamic properties and
transport properties. The involved properties are Minimum Misci-
ble Pressure (MMP), solubility, phase behavior, interfacial tension,
density, formation volume factor and swelling factor. The related
transport properties are viscosity or mobility ratio, thermal con-
ductivity, and diffusive coefficient or dissolution rate [147150].
During the injection process in a deep saline aquifer, the flow of
CO2 is driven by hydrodynamic and buoyant forces and is opposed
by viscous and capillary forces. Buoyant forces result from the den-
sity difference between CO2 and the aquifer water, which is a func-
tion of the relative permeability and the dynamic viscosity [151].
Therefore, the density and viscosity are necessary parameters in
the simulation of the process. Compared to buoyancy and viscous
forces, capillary forces have a smaller effect on the evolution of a
CO2 plume, hence on storage efficiency [152154]. Among the
properties, small variations of the interfacial tension (IFT) have
an impact on capillary pressure and permeability, and then the
flow process through capillary of porous medium is also affected,
especially under the high pressure [155]. Having injected CO2 into
a saline aquifer, the CO2 starts dissolving into the brine and forms a
solution with the dissolved CO2 that has a higher density than that
of pure water. Thus, a density-driven fluid flow occurs that will
enhance the dissolution rate of CO2, reduce time for solutal trap-
ping and increase storage security. Therefore, density is important
for the mechanism called density-driven natural convection [156].
In addition, diffusion is another process occurring after injection
that will take a long time for CO2 to be completely dissolved into
the formation brine [157]. In the convection and diffusion process,
density, viscosity, and diffusion coefficient are the important prop-
erties [158].
8.1.2.2. Ocean storage. For ocean storage, there are different meth-
ods of CO2 injection into the seawater: the first option is to inject
the carbon dioxide directly into seawater, where it will be dis-
solved. In this method, hydrate formation can occur at lower tem-
peratures (510 C) and higher pressures (approximately 0.4 MPa).
The dependence of liquid droplets and their rinsing or sinking rate
on the release depth is very important. The free liquid droplets will
217
Physical Membrane Compression and Condensation Transport Storage
adsorption separation pumping
p p p
p p
p
p p

p p
p
   
 
p p
   
   
p
   
rise and evaporate into gas bubbles as the pressure drops below
the saturation pressure, and the density of the hydrate mass
depends on the CO2 mixtures density; therefore, the phase behav-
ior and density are important in retaining the hydrate [159]. The
dissolution rate is also a parameter studied as well as droplet raft-
ing and formation of a gas phase as a droplet crosses the liquefac-
tion pressure depth [117]. The second option provides the
possibility of storage of CO2 as a liquid lake in the ocean floor. This
lake can be formed by either releasing the liquid directly into a
depression or by releasing negative buoyancy droplets or hydrated
particles close to the bottom before dissolution is completed [159].
In this case, the liquefaction pressure at a given temperature is sig-
nificant which determines the buoyancy relative to seawater [147].
8.2. Existing study of the property impacts on CO2 storage
Table 6 illustrates the existing studies of the property impacts
on the storage processes, and the previous studies mainly concen-
trated on the density and the viscosity.
8.2.1. Density
For the impact of density on the storage processes, Wang et al.
[146] studied the density impact on the storage capacity. The
results showed that the normalized capacity of the injected CO2
mixtures with 50% deviation in the density values can cause a
40% loss in the normalized storage capacity. Numerical experi-
ments aiming at the impacts of CO2 stream impurities on the
thermo-physical properties and geologic carbon storage were con-
ducted by [160]. The density of CO2 mixtures were compared with
that of pure CO2 as a function of depth. In addition, the static stor-
age capacity variation caused by changing the density and the vis-
cosity was studied. Similar results were found that the loss of
capacity can be over 50% at shallow depths, with 60% deviation
in the density of the CO2 mixtures; however, the difference of stor-
age capacity quickly decreases with depth. Jacquemet et al. [161]
analyzed the qualitative effects of the change of density on the
storage capacity. The methods of determination of the physical
properties of the gas mixtures and the implications of their
changes on the storage capacity were presented. It was found that
the presence of the associated components in the CO2 is likely to
lower its density and, as a consequence, lowers the storage capac-
ity available for CO2 as well as the migration of CO2 within the
reservoir. The decreased density may reduce the overall storage
capacity in the range of the yearly injection volumes and lead to
the maximum storage capacity being reached faster than it would
have been by injection of pure CO2. Nasir and Amiruddin [147]
218 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222
used an extended black oil simulator to investigate the impacts of
density and viscosity on the oil recovery factor in miscible CO2
injection. A sensitivity analysis was conducted by varying the
properties from
30% to 30%. The gas phase density affects the
gas phase viscosity, which changes the mobility ratio. It is observed
that as the density increases, the oil recovery increases. The frac-
tional change of the density is found to be 9%, and the uncertainty
in gas phase density is found to be 3%; thus, it is anticipated that
the oil recovery will vary within 27%.
8.2.2. Viscosity
The impacts of viscosity on the geologic storage capacity were
conducted by [160]. It is found that a decrease in viscosity of the
CO2 mixtures of 50% results in an approximately 50% loss in the
storage capacity in the shallow depth. Sass et al. [126] discussed
the viscosity effects of flue gas properties on storage capacity. It
is found that the kinematic viscosity (the ratio of viscosity and den-
sity) of the dry flue gas mixture can range between 0.7 and 1.4
times that of pure CO2, which can affect the injectability of CO2 into
the geologic formation. The added costs with the deviations are
likely to be a fraction (1025%) of the capital and operation costs
associated with an injection system. The sensitivity study of the
property impacts on the oil recovery factor was conducted by
[147]. The results showed that the oil recovery increases with
increasing gas phase viscosity. Changing the gas phase viscosity
by 1% affects the recovery by approximately 6%. As the uncertainty
of viscosity is in the range of 1230%, it is expected the oil recovery
prediction changes by more than 60%.
9. Discussion and conclusion
The involved CCS processes and properties (including thermo-
dynamic properties and transport properties) are listed in Table 7.
The relationship between the processes and the properties, as well
as the status of the property impacts in existing studies are also
p
identified in the Table. Symbol indicates that the property has
been studied and quantitative results are available. Symbol
indicates that no qualitative results. Symbol shows that there
is no existing research on the impact of the property. Symbol 
denotes that the property is not related to the process.
For chemical absorption, the impacts of the properties on the
process parameters were studied quantitatively. In the case of
amine absorption in a packed column, deviation of the density
and the heat capacity of the liquid phase were found to have nota-
ble impacts on the location and magnitude of the temperature
bulge in the liquid phase temperature profile inside the absorber.
However, the deviation of the density and the heat capacity causes
less significant impacts on the rich stream loading and CO2 recov-
ery rate compared with the impacts from viscosity and diffusivity.
Apart from the liquid phase, the impacts of the gas phase proper-
ties, including density, viscosity and diffusivity, were also evalu-
ated. The deviation of gas phase diffusivity was found to have a
more significant impact on the packing height design than the den-
sity and viscosity. Moreover, a study was found regarding the
impacts of the phase equilibrium used to retrieve the activity coef-
ficients of the species for the calculation of the heat of absorption.
Phase equilibrium deviation results in different values of the heat
of absorption of CO2 in the amine solution and will directly affect
the determination of the energy requirement for the solvent regen-
eration. Knowledge gaps on the quantitative studies of the prop-
erty impacts on chemical absorption processes are still required
for both transport and thermodynamic properties. There are sev-
eral parts of the design and operation that have not been studied
concerning the property impacts, such as column diameter and
pressure drop, especially in the desorption part, which is more
complicated than the absorption. Moreover, studies on the impacts
of several properties, such as surface tension and thermal conduc-
tivity, were not found.
Few existing studies on the property impacts for the physical
absorption process were found. Only the impacts of liquid phase
density, heat capacity, enthalpy and phase equilibrium on the Rec-
tisol process were studied. The evaluated process parameters are
the temperature, pressure, mole flow and mole fraction of each
component of the streams leaving the absorbers. The studies
focused mainly on the accuracy of different property models, and
the quantitative results of the property impacts are not available.
Moreover, the impacts of the transport properties and the impacts
on the other parts of the process, such as the refrigeration system
and flash regeneration, were not found. Therefore, in this capture
process, quantitative evaluations of the property impact on the
process parameters are required for both thermodynamic and
transport properties.
For the CO2 adsorption process, there are no studies on the
property impacts on the process. However, the properties of CO2
mixtures, including density, viscosity, heat capacity, phase behav-
ior, thermal conductivity, and diffusivity, are necessary for calcu-
lating the momentum, mass and energy balance for evaluation of
the performance of adsorption columns.
The evaluation of the performance of membrane separation is
dependent on the knowledge of several properties. However, as
listed in Table 7, the study of the property impact on the process
parameters, such as the required area of membrane and the com-
pression duty, is still lacking.
For CO2 compression and pumping processes, the CO2 pumping
process is more sensitive to the uncertainty of the properties.
Among the properties, heat capacity, density, enthalpy and entropy
are the most important ones, and it has been shown that 2% devi-
ation in these properties can result in up to 12% difference in the
calculation of the pumping efficiency. Compared with the pumping
process, the compression performance is less sensitive to the prop-
erties and impurities. The compression performance is most sensi-
tive to heat capacity and compressibility. The impacts of thermal
conductivity, speed of sound, viscosity and density still must be
evaluated. However, in some cases, the impacts of different prop-
erties were evaluated in one simulation, which gives the synergetic
effect of property groups, for example, the study of [112] on ther-
mal conductivity, speed of sound, viscosity and density. Therefore,
further research should be performed to investigate the impacts of
single property to rank the properties by importance.
For the property impacts on condensation and liquefaction, the
impact of density, enthalpy and entropy were studied quantita-
tively. The deviations of density, enthalpy and entropy using differ-
ent models are low and give small deviations in the determination
of the exhaust gas enthalpy drop in heat exchangers. However, for
other related properties, such as phase characteristic, heat capac-
ity, viscosity and thermal conductivity, the impacts must be fur-
ther investigated.
For the transport process, existing studies mainly focused on
property impacts on the performance of pipeline steady flow pro-
cesses. The studied parameters include pressure loss, temperature
drop and transport safety. Among the properties, density and heat
capacity are the most important in the pipeline steady flow pro-
cess. For other properties, including viscosity and thermal conduc-
tivity, the property impacts are much smaller. In addition, the
phase characteristics were studied mainly from the aspect of
avoiding two-phase flow and securing transport safety; these
phase characteristics are also very important in pipeline transport.
However, the research involving the property impacts on transient
pipeline transport processes is insufficient, and the properties,
such as speed of sound, diffusivity and JouleThomson inversion
curves, must be evaluated in the future. In addition, there are no
 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222 219

Table 8
Ranking of the properties according to their impacts on the processes.

Process Phase Mixture Working High impact Low impact

condition
Chemical Gas CO2/H2O/N2/O2 313351 K, Diffusivity, phase equilibrium Density and viscosity
absorption 0.10
0.11 MPa
Liquid CO2/H2O/MEA 288357 K Viscosity, diffusivity, phase Density, heat capacity,
equilibrium thermal conductivity
Physical absorption Gas/liquid CO2/CH3OH/H2/N2/H2S/CH4/CO/COS, CO2/ 223254 K Thermodynamic properties
CH3OH/H2/N2/H2S/H2O/Ar/CH4/CO/COS 3.3
4.4 MPa
Physical adsorption Gas CO2/H2O/CO/SO2/O2/NO2/NO/N2/CH4 273673 K,
0.015 MPa
Membrane Gas CO2/CH4/N2/H2/CO/O2 293343 K,
separation 0.1
3.5 MPa
Compression and Gas/liquid CO2/O2/N2/Ar/H2/CO/H2S/CH4/H2O/SO2 219423 K, Heat capacity, density,
pumping 011 MPa enthalpy, compressibility
Condensation and Gas/liquid CO2/H2O/O2/N2/Ar/SO2 219423 K, Density, enthalpy and
liquefaction 011 MPa entropy
Transport Liquid/supercritical CO2/H2O/N2/CH4/Ar/SO2 223350 K, Density, heat capacity Viscosity, thermal
0.120 MPa conductivity
Storage Liquid/supercritical CO2/O2/Ar/N2/H2/SO2/H2S 261393 K, Density, viscosity
1050 MPa

studies conducted on the property impacts on ship transport,
which is also a knowledge gap.
In the storage process, the density and viscosity have received
most of the attention in the property impact studies, and they were
regarded as most important properties concerning the storage
capacity and enhanced oil recovery rate. In the storage capacity
calculation, the existence of 15% impurities can change the density
and viscosity values up to 50%, which leads to 60% difference in the
normalized storage capacity. In addition, the density and viscosity
have larger impacts on the EOR process, and 1% change in the den-
sity and viscosity can lead to 9% and 6% variation in oil recovery
factor, respectively. However, the impacts of the properties of
phase characteristic, heat capacity, enthalpy and entropy, diffusiv-
ity and thermal conductivity require further studies in the next
step in the dispersion process in the EOR process and in the disso-
lution process in the ocean or the saline aquifer.
10. Future studies
The properties are grouped in Table 8 according to their impacts
on each CCS process in the existing studies. The properties are clas-
sified as high impact and low impact. For the high impact proper-
ties, the uncertainty of properties will result in great deviations in
the system design and operation; therefore, the property model
with high accuracy should be selected or developed.
For chemical absorption, the viscosity and diffusivity of the liq-
uid phase have more impacts on the absorption simulation than
the other properties. Moreover, compared with the density and vis-
cosity of the gas phase, the diffusivity shows a more significant
impact on the column design. In addition, phase equilibrium is
found to have an important influence on the regeneration energy
calculation. Therefore, from the perspective of property modeling,
developing more accurate diffusivity, viscosity and thermody-
namic models should be prioritized. In addition, more studies
focusing on the property impacts on the final design values, such
as the column diameter or the packing height, are still required.
Apart from the process design, the quantitative impacts of the ther-
modynamic properties in the evaluation of the regeneration energy
requirement should be studied.
In physical absorption, there is no available study about the
property impact on the process simulation, and only some studies
about different thermodynamic property models were found.
Therefore, the future study of physical absorption should cover
the quantitative evaluation on the impact of each property on
the process to highlight the potential property model
development.
For CO2 adsorption and membrane separation processes, the
impacts of the CO2 mixture properties are not clear due to the lack
of studies. For these two processes, it is important to quantitatively
evaluate the property impacts to highlight the impact level of each
property.
For CO2 compression and pumping processes, heat capacity,
density, and enthalpy have high impacts on pumping performance,
and heat capacity and compressibility are the most important in
the compression process. Therefore, the model accuracy of the
above properties must be improved. In addition, the impacts of
other properties are unknown and must be studied in the future.
In the condensation and liquefaction process, density, enthalpy
and entropy have low impacts on the heat exchanger design and
operation. However, other properties need to be evaluated to iden-
tify the key properties in the future.
In the transport process, the properties of density and heat
capacity are found to have strong impacts on pipeline transporta-
tion, whereas the effects of viscosity and thermal conductivity
can be ignored. More research should be focused on vessel trans-
portation regarding the property impacts.
In the storage process density and viscosity are the most impor-
tant properties, based on the existing studies; therefore, these two
properties should be prioritized in property modeling to reduce the
uncertainty. In addition, more studies should be conducted on
other impacts of the storage process.
Different methodological approaches can be applied to the
future studies. For the CCS processes in which the property impacts
have not been identified, such as physical adsorption and mem-
brane separation processes, the approach of sensitivity study can
be employed in process simulation and design in order to identify
the key properties. For the processes in which the important prop-
erties have been identified, for example density and viscosity in
220 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222

dense phase in storage process, modeling of properties need to be [24] Strigle RF. Packed tower design and applications: random and structured
packings. 2nd ed. Gulf Pub. Co.; 1994.
conducted. Specifically, the experimental data needs to be carefully
[25] Billet R, Schultes M. Prediction of mass transfer columns with dumped and
checked and collected to evaluate the accuracy of property models. arranged packings: updated summary of the calculation method of billet and
To improve the property models, the existing models which have Schultes. Chem Eng Res Des 1999;77:498504.
the potentials to give accurate results can be modified, otherwise [26] Robbins LA. Improve pressure-drop prediction with a new correlation. Chem
Eng Prog 1991;87:8790.
the new models should be developed if the existing models with [27] Zhang Y, Chen H, Chen C-C, Plaza JM, Dugas R, Rochelle GT. Rate-based
modifications cannot satisfy the requirements. process modeling study of CO2 capture with aqueous monoethanolamine
solution. Ind Eng Chem Res 2009;48:923346.
[28] Khan FM, Krishnamoorthi V, Mahmud T. Modelling reactive absorption of CO2
Acknowledgement in packed columns for post-combustion carbon capture applications. Chem
Eng Res Des 2011;89:16008.
[29] Onda K, Sada E, Takeuchi H. Gas absorption with chemical reaction in packed
The financial support by Swedish Energy Agency and Swedish columns. J Chem Eng Jpn 1968;1:62.
Research Council is appreciated. Yuting Tan also appreciates schol- [30] Onda K, Takeuchi H, Okumoto Y. Mass transfer coefficients between gas and
liquid phases in packed columns. J Chem Eng Jpn 1968;1.
arship from China Scholarship Council (CSC). Worrada Nookuea
[31] Billet R, Schultes M. Predicting mass transfer in packed columns. Chem Eng
would like to appreciate the foundation of tandlkare Gustav Dahls Technol 1993;16:19.
memory. [32] Billet R, Schultes M. Advantage in correlating packing column performance.
Inst Chem Eng Symp Ser 1992;128.
[33] Bravo JL, Rocha JA, Fair JR. Mass transfer in gauze packings. Hydrocarbon
References Process 1985;64:91.
[34] Rocha JA, Bravo JL, Fair JR. Distillation columns containing structured
packings: a comprehensive model for their performance. 2. Mass-transfer
[1] Edenhofer O, Pichs-Madruga R, Sokona Y, Minx JC, Farahani E, Kadner S, et al.
model. Ind Eng Chem Res 1996;35:16607.
IPCC: climate change 2014: mitigation of climate change. Cambridge (United
[35] Roizard C, Wild G. Mass transfer with chemical reaction: the slow reaction
Kingdom): Cambridge University Press; 2014.
regime revisited. Chem Eng Sci 2002;57:347984.
[2] International Energy Agency. Technology roadmap carbon capture and
[36] Austgen DM, Rochelle GT, Peng X, Chen CC. Model of vapor-liquid equilibria
storage 2013 Edition; 2013.
for aqueous acid gas-alkanolamine systems using the electrolyte-NRTL
[3] Fu C, Gundersen T. Techno-economic analysis of CO2 conditioning processes
equation. Ind Eng Chem Res 1989;28:106073.
in a coal based oxy-combustion power plant. Int J Greenhouse Gas Control
[37] Haghtalab A, Dehghani Tafti M. Electrolyte UNIQUAC
NRF model to study
2012;9:41927.
the solubility of acid gases in alkanolamines. Ind Eng Chem Res
[4] Roussanaly S, Jakobsen JP, Hognes EH, Brunsvold AL. Benchmarking of CO2
2007;46:605360.
transport technologies: Part IOnshore pipeline and shipping between two
[38] Freguia S, Rochelle GT. Modeling of CO2 capture by aqueous
onshore areas. Int J Greenhouse Gas Control 2013;19:58494.
monoethanolamine. AIChE J 2003;49:167686.
[5] Hoff KA, Juliussen O, Falk-Pedersen O, Svendsen HF. Modeling and
[39] Notz R, Tnnies I, Mangalapally HP, Hoch S, Hasse H. A short-cut method for
experimental study of carbon dioxide absorption in aqueous alkanolamine
assessing absorbents for post-combustion carbon dioxide capture. Int J
solutions using a membrane contactor. Ind Eng Chem Res 2004;43:490821.
Greenhouse Gas Control 2011;5:41321.
[6] Yan J, Anheden M, Bernstone C, Liljemark S, Pettersson H, Li H, et al. Impacts
[40] Oexmann J, Kather A. Minimising the regeneration heat duty of post-
of non-condensable components on CO2 compression/purification, pipeline
combustion CO2 capture by wet chemical absorption: the misguided focus
transport and geological storage. In: Proceedings of the 1st oxyfuel
on low heat of absorption solvents. Int J Greenhouse Gas Control 2010;
combustion conference, Cottbus, Germany; 2009.
4:3643.
[7] Li H, Jakobsen JP, Wilhelmsen , Yan J. PVTxy properties of CO2 mixtures
[41] Kim H, Hwang SJ, Lee KS. Novel shortcut estimation method for regeneration
relevant for CO2 capture, transport and storage: review of available
energy of amine solvents in an absorption-based carbon capture process.
experimental data and theoretical models. Appl Energy 2011;88:356779.
Environ Sci Technol 2015;49:147885.
[8] Li H, Wilhelmsen , Lv Y, Wang W, Yan J. Viscosities, thermal conductivities
[42] Cheng S, Meisen A, Chackma A. Predict amine solution properties accurately.
and diffusion coefficients of CO2 mixtures: review of experimental data and
Hydrocarbon Process 1996;75:81.
theoretical models. Int J Greenhouse Gas Control 2011;5:111939.
[43] Weiland RH, Dingman JC, Cronin DB, Browning GJ. Density and viscosity of
[9] Olajire AA. CO2 capture and separation technologies for end-of-pipe
some partially carbonated aqueous alkanolamine solutions and their blends. J
applications a review. Energy 2010;35:261028.
Chem Eng Data 1998;43:37882.
[10] Nuchitprasittichai A, Cremaschi S. Sensitivity of amine-based CO2 capture
[44] Kvamsdal HM, Rochelle GT. Effects of the temperature bulge in CO2
cost: the influences of CO2 concentration in flue gas and utility cost
absorption from flue gas by aqueous monoethanolamine. Ind Eng Chem Res
fluctuations. Int J Greenhouse Gas Control 2013;13:3443.
2008;47:86775.
[11] Warudkar SS, Cox KR, Wong MS, Hirasaki GJ. Influence of stripper operating
[45] Kvamsdal HM, Hillestad M. Selection of model parameter correlations in a
parameters on the performance of amine absorption systems for post-
rate-based CO2 absorber model aimed for process simulation. Int J
combustion carbon capture: Part I. High pressure strippers. Int J Greenhouse
Greenhouse Gas Control 2012;11:1120.
Gas Control 2013;16:34250.
[46] Nookuea W, Tan Y, Li H, Thorin E, Yan J. Sensitivity study of thermo-physical
[12] Wang M, Lawal A, Stephenson P, Sidders J, Ramshaw C. Post-combustion CO2
properties of gas phase on absorber design for CO2 capture using
capture with chemical absorption: a state-of-the-art review. Chem Eng Res
monoethanolamine. Energy Proc 2015;75:230510.
Des 2011;89:160924.
[47] Faramarzi L, Kontogeorgis GM, Michelsen ML, Thomsen K, Stenby EH.
[13] Supap T, Idem R, Tontiwachwuthikul P. Mechanism of formation of heat
Absorber model for CO2 capture by monoethanolamine. Ind Eng Chem Res
stable salts (HSSs) and their roles in further degradation of
2010;49:37519.
monoethanolamine during CO2 capture from flue gas streams. Energy Proc
[48] Glasscock DA. Modeling and experimental study of carbon dioxide absorption
2011;4:5918.
into aqueous alkanolamines. Texas (United States): University of Texas at
[14] Astarita G. Mass transfer with chemical reaction. Amsterdam, London, New
Austin Press; 1990.
York: Elsevier Publishing Company; 1967.
[49] Hartono A, Mba EO, Svendsen HF. Physical properties of partially CO2 loaded
[15] Ahn H, Luberti M, Liu Z, Brandani S. Process simulation of aqueous MEA plants
aqueous monoethanolamine (MEA). J Chem Eng Data 2014;59:180816.
for post-combustion capture from coal-fired power plants. Energy Proc
[50] Kim I, Hessen ET, Haug-Warberg T, Svendsen HF. Enthalpies of absorption of
2013;37:152331.
CO2 in aqueous alkanolamine solutions from e-NRTL model. Energy Proc
[16] Oyenekan BA, Rochelle GT. Energy performance of stripper configurations for
2009;1:82935.
CO2 capture by aqueous amines. Ind Eng Chem Res 2005;45:245764.
[51] Kidnay J, Parrish WR, McCartney DG. Fundamentals of natural gas processing.
[17] Yeh JT, Pennline HW, Resnik KP. Study of CO2 absorption and desorption in a
2nd ed. Boca Raton (United States): CRC Press; 2011.
packed column. Energy Fuels 2001;15:2748.
[52] Kohl AL, Nielsen RB. Physical solvents for acid gas removal. In: Kohl AL,
[18] Nagy T, Mizsey P. Model verification and analysis of the CO2-MEA absorber
Nielsen RB, editors. Gas Purification. Houston: Gulf Professional Publishing;
desorber system. Int J Greenhouse Gas Control 2015;39:23644.
1997. p. 1187237 [chapter 14].
[19] Tontiwachwuthikul P, Meisen A, Lim CJ. Novel pilot plant technique for sizing
[53] Sun L, Smith R. Rectisol wash process simulation and analysis. J Clean Prod
gas absorbers with chemical reactions. Can J Chem Eng 1989;67:6027.
2013;39:3218.
[20] Koronaki IP, Prentza L, Papaefthimiou V. Modeling of CO2 capture via
[54] Scholes CA, Anderson CJ, Stevens GW, Kentish SE. Membrane gas separation
chemical absorption processes
an extensive literature review. Renew
physical solvent absorption combined plant simulations for pre-combustion
Sustain Energy Rev 2015;50:54766.
capture. Energy Proc 2013;37:103949.
[21] Khalilpour R, Abbas A. Optimal synthesis and design of solvent-based PCC
[55] Park SH, Lee SJ, Lee JW, Chun SN, Lee JB. The quantitative evaluation of two-
process using a rate-based model. Sep Purif Technol 2014;132:14967.
stage pre-combustion CO2 capture processes using the physical solvents with
[22] Cussler EL. Diffusion: mass transfer in fluid systems. Cambridge (United
various design parameters. Energy 2015;81:4755.
Kingdom): Cambridge University Press; 1997.
[56] Stehlk P. Contribution to advances in waste-to-energy technologies. J Clean
[23] Eckert JS. Design techniques for sizing packed towers. Chem Eng Prog
Prod 2009;17:91931.
1961;57:548.
 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222
[57] Yang J, Oh S-R, Liu W. Optimization of shell-and-tube heat exchangers using a
general design approach motivated by constructal theory. Int J Heat Mass
Transfer 2014;77:114454.
[58] Chiu L-F, Li M-H. Heat capacity of alkanolamine aqueous solutions. J Chem
Eng Data 1999;44:1396401.
[59] Mukherjee R. Effectively design shell-and-tube heat exchangers. Chem Eng
Prog 1998;94.
[60] Liu L, Cai F, Sun J, Han W. Simulation and optimization of Rectisol using
revised thermodynamic methods. In: A.I.o.C. Engineers, editor. 2012 spring
meeting & 8th global congress on process safety; 2012.
[61] Shafeeyan MS, Wan Daud WMA, Shamiri A. A review of mathematical
modeling of fixed-bed columns for carbon dioxide adsorption. Chem Eng Res
Des 2014;92:96188.
[62] Ducroux R, Jean-Baptiste P. Technologies, methods and modeling for CO2
capture. In: Greenhouse gas control technologies, vol. 2; 2005. p. 18359.
[63] Song C, Kansha Y, Ishizuka M, Fu Q, Tsutsumi A. Conceptual design of a novel
pressure swing CO2 adsorption process based on self-heat recuperation
technology. Chem Eng Process 2015;94:208.
[64] Gomes VG, Hassan MM. Coalseam methane recovery by vacuum swing
adsorption. Sep Purif Technol 2001;24:18996.
[65] Kacem M, Pellerano M, Delebarre A. Pressure swing adsorption for CO2/N2
and CO2/CH4 separation: comparison between activated carbons and zeolites
performances. Fuel Process Technol 2015;138:27183.
[66] Chue KT, Kim JN, Yoo YJ, Cho SH, Yang RT. Comparison of activated carbon
and zeolite 13X for CO2 recovery from flue gas by pressure swing adsorption.
Ind Eng Chem Res 1995;34:5918.
[67] Cavenati S, Grande CA, Rodrigues AE. Removal of carbon dioxide from natural
gas by vacuum pressure swing adsorption. Energy Fuels 2006;20:
264859.
[68] Afzal S, Rahimi A, Ehsani MR, Tavakoli H. Modeling hydrogen fluoride
adsorption by sodium fluoride. J Ind Eng Chem 2010;16:97885.
[69] Leinekugel-le-Cocq D, Tayakout-Fayolle M, Le Gorrec Y, Jallut C. A double
linear driving force approximation for non-isothermal mass transfer
modeling through bi-disperse adsorbents. Chem Eng Sci 2007;62:404053.
[70] Jin X, Malek A, Farooq S. Production of argon from an oxygen
argon mixture
by pressure swing adsorption. Ind Eng Chem Res 2006;45:577587.
[71] Wakao N, Funazkri T. Effect of fluid dispersion coefficients on particle-to-fluid
mass transfer coefficients in packed beds: correlation of sherwood numbers.
Chem Eng Sci 1978;33:137584.
[72] Green DW, Perry RH. Perrys chemical engineers handbook. 8th ed. New York
(United States): McGraw-Hill; 1934.
[73] Grande CA, Rodrigues AE. Adsorption kinetics of propane and propylene in
zeolite 4A. Chem Eng Res Des 2004;82:160412.
[74] Wakao N, Kaguei S, Funazkri T. Effect of fluid dispersion coefficients on
particle-to-fluid heat transfer coefficients in packed beds: correlation of
Nusselt numbers. Chem Eng Sci 1979;34:32536.
[75] Rezaei F, Grahn M. Thermal management of structured adsorbents in CO2
capture processes. Ind Eng Chem Res 2012;51:402534.
[76] Alpay E, Kenney CN, Scott DM. Simulation of rapid pressure swing adsorption
and reaction processes. Chem Eng Sci 1993;48:317386.
[77] Lu ZP, Loureiro JM, Rodrigues AE, LeVan MD. Pressurization and blowdown of
adsorption bedsII. Effect of the momentum and equilibrium relations on
isothermal operation. Chem Eng Sci 1993;48:1699707.
[78] Yang J, Park M-W, Chang J-W, Ko S-M, Lee C-H. Effects of pressure drop in a
PSA process. Korean J Chem Eng 1998;15:2116.
[79] Macdonald IF, El-Sayed MS, Mow K, Dullien FAL. Flow through porous media-
the Ergun equation revisited. Ind Eng Chem Fund 1979;18:199208.
[80] Rezaei F, Webley P. Optimum structured adsorbents for gas separation
processes. Chem Eng Sci 2009;64:518291.
[81] Richardson JT, Peng Y, Remue D. Properties of ceramic foam catalyst supports:
pressure drop. Appl Catal A 2000;204:1932.
[82] Vaughn JT, Koros WJ. Analysis of feed stream acid gas concentration effects on
the transport properties and separation performance of polymeric
membranes for natural gas sweetening: a comparison between a glassy and
rubbery polymer. J Membrane Sci 2014;465:10716.
[83] Koros WJ, Fleming GK. Membrane-based gas separation. J Membrane Sci
1993;83:180.
[84] Favre E. Membrane processes and postcombustion carbon dioxide capture:
challenges and prospects. Chem Eng J 2011;171:78293.
[85] Brunetti A, Scura F, Barbieri G, Drioli E. Membrane technologies for CO2
separation. J Membrane Sci 2010;359:11525.
[86] Lee H-J, Suda H, Haraya K. Gas permeation properties in a composite
mesoporous alumina ceramic membrane. Korean J Chem Eng 2005;22:7218.
[87] Wang K, Suda H, Haraya K. The characterization of CO2 permeation in a CMSM
derived from polyimide. Sep Purif Technol 2003;31:619.
[88] Uchytil P, Petrickovic R, Thomas S, Seidel-Morgenstern A. Influence of
capillary condensation effects on mass transport through porous
membranes. Sep Purif Technol 2003;33:27381.
[89] Bounaceur R, Lape N, Roizard D, Vallieres C, Favre E. Membrane processes for
post-combustion carbon dioxide capture: a parametric study. Energy
2006;31:255670.
[90] Zhao L, Riensche E, Menzer R, Blum L, Stolten D. A parametric study of CO2/N2
gas separation membrane processes for post-combustion capture. J
Membrane Sci 2008;325:28494.
[91] Hasan MMF, Baliban RC, Elia JA, Floudas CA. Modeling, simulation, and
optimization of postcombustion CO2 capture for variable feed concentration
221
and flow rate. 1. Chemical absorption and membrane processes. Ind Eng
Chem Res 2012;51:1564264.
[92] Merkel TC, Lin H, Wei X, Baker R. Power plant post-combustion carbon
dioxide capture: an opportunity for membranes. J Membrane Sci
2010;359:12639.
[93] Zhao L, Menzer R, Riensche E, Blum L, Stolten D. Concepts and investment
cost analyses of multi-stage membrane systems used in post-combustion
processes. Energy Proc 2009;1:26978.
[94] He X, Hgg M-B. Energy efficient process for CO2 capture from flue gas with
novel fixed-site-carrier membranes. Energy Proc 2014;63:17485.
[95] Zhanat U, Sharipzhan E. Mechanism of gases transfer through polymer
membranes and membrane module calculation model to separate gases. Proc
Technol 2012;1:35661.
[96] McCabe WL, Smith JC, Harriott P. Unit operation of chemical engineering. 7th
ed. McGraw-Hill; 2005.
[97] Aspelund A, Jordal K. Gas conditioningthe interface between CO2 capture
and transport. Int J Greenhouse Gas Control 2007;1:34354.
[98] Xu G, Li L, Yang Y, Tian L, Liu T, Zhang K. A novel CO2 cryogenic liquefaction
and separation system. Energy 2012;42:5229.
[99] Posch S, Haider M. Optimization of CO2 compression and purification units
(CO2CPU) for CCS power plants. Fuel 2012;101:25463.
[100] Besong MT, Maroto-Valer MM, Finn AJ. Study of design parameters affecting
the performance of CO2 purification units in oxy-fuel combustion. Int J
Greenhouse Gas Control 2013;12:4419.
[101] Li H, Hu Y, Ditaranto M, Willson D, Yan J. Optimization of cryogenic CO2
purification for oxy-coal combustion. Energy Proc 2013;37:13417.
[102] Pipitone G, Bolland O. Power generation with CO2 capture: technology for
CO2 purification. Int J Greenhouse Gas Control 2009;3:52834.
[103] Ludington AR. Tools for supercritical carbon dioxide cycle analysis and the
cycles applicability to sodium fast reactors. Massachusetts (United
States): Massachusetts Institute of Technology Press; 2009.
[104] Lindroth M, Munktell M. Preliminary scoping and thermodynamic modelling
of CO2 compressors for carbon capture and storage. Lund (Sweden): Lund
University Press; 2009.
[105] Goos E, Riedel U, Zhao L, Blum L. Phase diagrams of CO2 and CO2N2 gas
mixtures and their application in compression processes. Energy Proc
2011;4:377885.
[106] Roberts SK, Sjolander SA. Effect of the specific heat ratio on the aerodynamic
performance of turbomachinery. J Eng Gas Turbines Power
2002;127:77380.
[107] Li H. Thermodynamic properties of CO2 mixtures and their applications in
advanced power cycles with CO2 capture processes. Stockholm
(Sweden): Royal Institute of Technology Press; 2008.
[108] Weatherwax M, Patel V, Musardo A, Giovani G, Plella M, Cipriani S. CO2
compression at worlds largest carbon dioxide injection project. In: Forty-first
turbomachinery symposium Texas, United States; 2012.
[109] Kster T, Eggers R. Liquefaction of oxyfuel flue gases experimental results
and modeling of heat transfer coefficients for pure CO2. Energy Proc
2014;51:10918.
[110] Prasad BSV, Gurukul SMKA. Differential methods for the performance
prediction of multistream plate-fin heat exchangers. J Heat Transfer
1992;114:419.
[111] Lee J, Lee JI, Ahn Y, Kim SG. Effect of uncertainties in CO2 property database
on the S-CO2 compressor performance. In: Transaction of the Korean Nuclear
Society Spring Meeting, Gwangju, Korea; 2013.
[112] Wetenhall B, Aghajani H, Chalmers H, Benson SD, Ferrari MC, Li J, et al. Impact
of CO2 impurity on CO2 compression, liquefaction and transportation. Energy
Proc 2014;63:276478.
[113] Ulfsnes RE, Karlsen G, Jordal K, Bolland O. Investigation of physical properties
for CO2/H2O mixtures for use in semi-closed O2/CO2 gas turbine cycle with
CO2-capture. Copenhagen (Denmark): ECOS; 2003.
[114] Baltadjiev ND. An investigation of real gas effects in supercritical CO2
compressors. Massachusetts (United States): Massachusetts Institute of
Technology Press; 2012.
[115] Sandberg MR. Equation of state influence on compressor performance
determination. In: Proceedings of the thirty-fourth turbomachinery
symposium, college station. Texas (United States): Texas A&M University
Press; 2005.
[116] Luo X, Wang M, Oko E, Okezue C. Simulation-based techno-economic
evaluation for optimal design of CO2 transport pipeline network. Appl
Energy 2014;132:61020.
[117] Metz B, Davidson O, de Coninck H, Loos M, Meyer L. IPCC: special report on
carbon dioxide capture and storage. Cambridge (United
Kingdom): Cambridge University Press; 2005.
[118] IEAGHG R&D Programme. Ship transport of CO2. Report number PH4/30;
2004.
[119] Seo Y, You H, Lee S, Huh C, Chang D. Evaluation of CO2 liquefaction processes
for ship-based carbon capture and storage (CCS) in terms of life cycle cost
(LCC) considering availability. Int J Greenhouse Gas Control 2015;35:112.
[120] Decarre S, Berthiaud J, Butin N, Guillaume-Combecave J-L. CO2 maritime
transportation. Int J Greenhouse Gas Control 2010;4:85764.
[121] Aspelund A, Mlnvik MJ, De Koeijer G. Ship transport of CO2: technical
solutions and analysis of costs, energy utilization, exergy efficiency and CO2
emissions. Chem Eng Res Des 2006;84:84755.
[122] Lee U, Lim Y, Lee S, Jung J, Han C. CO2 storage terminal for ship transportation.
Ind Eng Chem Res 2012;51:38997.
222 Y. Tan et al. / Energy Conversion and Management 118 (2016) 204222
[123] Carbon Management GIS. CO2 pipeline transport cost estimation. Carbon
capture and sequestration technologies program. Massachusetts (United
states): Massachusetts Institute of Technology Press; 2009.
[124] Chapoy A, Nazeri M, Kapateh M, Burgass R, Coquelet C, Tohidi B. Effect of
impurities on thermophysical properties and phase behaviour of a CO2-rich
system in CCS. Int J Greenhouse Gas Control 2013;19:92100.
[125] Clauser C. Thermal storage and transport properties of rocks, II: Thermal
conductivity and diffusivity. In: Gupta H, editor. Encyclopedia of solid earth
geophysics. Netherlands: Springer; 2014. p. 143148.
[126] Sass BM, Farzan H, Prabhakar R, Gerst J, Sminchak J, Bhargava M, et al.
Considerations for treating impurities in oxy-combustion flue gas prior to
sequestration. Energy Proc 2009;1:53542.
[127] Liu M, Zang S, Zhou D. Fast leak detection and location of gas pipelines based
on an adaptive particle filter. Int J Appl Math Comput Sci 2005;15:54150.
[128] Lund H, Fltten T, Tollak Munkejord S. Depressurization of carbon dioxide in
pipelines models and methods. Energy Proc 2011;4:298491.
[129] Stoianov I, Nachman L, Madden S, Tokmouline T, Csail M. PIPENET: a wireless
sensor network for pipeline monitoring. In: Sixth international symposium
on information processing in sensor networks; 2007. p. 26473.
[130] Zhang J. Designing a cost-effective and reliable pipeline leak-detection
system. Pipes Pipelines Int 1997;42:206.
[131] Vrabec J, Kumar A, Hasse H. Joule-Thomson inversion curves of mixtures by
molecular simulation in comparison to advanced equations of state: natural
gas as an example. Fluid Phase Equilib 2007;258:3440.
[132] Abdumula FM. Optimization of paraffin removal from the Algyo
Szazhalombatta transporting crude oil pipeline. Miskolc
(Hungary): University of Miskolc Press; 2005.
[133] Lvstad M, Veritas DN. Design & operation of LPG ships: best practices for
maintenance of LPG ships hull integrity management. London, United
Kingdom; 2008.
[134] Li R, Jin W, Zhang J, Jin J, Xu E. Thermal study on LNG cargo tank. Int J Autom
Control Eng 2015;4:58.
[135] Germanischer Lloyd SE. Rules for classification and construction ship
technology. Hamburg (Germany): Germanischer Lloyd SE; 2008.
[136] Senjanovic I, Slapnicar V, Mravak Z, Rudan S, Ljustina AM. Structure design of
cargo tanks in liquefied gas carriers. In: International design conference
design 2006, Croatia; 2006.
[137] Zhao Q, Li Y-X. The influence of impurities on the transportation safety of an
anthropogenic CO2 pipeline. Process Saf Environ Prot 2014;92:8092.
[138] Tan Y, Nookuea W, Li H, Thorin E, Zhao L, Yan J. Property impacts on
performance of CO2 pipeline transport. Energy Proc 2015;75:22617.
[139] Mazzoccoli M, Guidob GD, Bosioa B, Aratoa E, Pellegrini Laura A. CO2-mixture
properties for pipeline transportation in the CCS process. Chem Eng Trans
2013;32:18616.
[140] Lemmon EW, Huber ML, McLinden MO. NIST standard reference database 23,
NIST reference fluid thermodynamic and transport properties-REFPROP,
version 9.1. Standard reference data program. Gaithersburg (MD): National
Institute of Standards and Technology; 2013.
[141] Welkenhuysen K, Ramrez A, Swennen R, Piessens K. Strategy for ranking
potential CO2 storage reservoirs: a case study for Belgium. Int J Greenhouse
Gas Control 2013;17:43149.
[142] Michael K, Golab A, Shulakova V, Ennis-King J, Allinson G, Sharma S, et al.
Geological storage of CO2 in saline aquifersa review of the experience from
existing storage operations. Int J Greenhouse Gas Control 2010;4:65967.
[143] Queirs AM, Taylor P, Cowles A, Reynolds A, Widdicombe S, Stahl H. Optical
assessment of impact and recovery of sedimentary pH profiles in ocean
acidification and carbon capture and storage research. Int J Greenhouse Gas
Control 2015;38:11020.
[144] White CM, Strazisar BR, Granite EJ, Hoffman JS, Pennline HW. Separation and
capture of CO2 from large stationary sources and sequestration in geological
formationscoalbeds and deep saline aquifers. J Air Waste Manage Assoc
2003;53:645715.
[145] Wei N, Li X, Wang Y, Dahowski RT, Davidson CL, Bromhal GS. A preliminary
sub-basin scale evaluation framework of site suitability for onshore aquifer-
based CO2 storage in China. Int J Greenhouse Gas Control 2013;12:23146.
[146] Wang J, Ryan D, Anthony EJ, Wildgust N, Aiken T. Effects of impurities on CO2
transport, injection and storage. Energy Proc 2011;4:30718.
[147] Nasir FM, Amiruddin NA. Miscible CO2 injection: sensitivity to fluid
properties. Society of Petroleum Engineers; 2008.
[148] Wilkinson JR, Leahy-Dios A, Teletzke GF, Dickson JL. Use of CO2 containing
impurities for miscible enhanced oil recovery. Society of Petroleum
Engineers; 2010.
[149] Al-Siyabi I. Effect of impurities on CO2 stream properties. Edinburgh
(Scotland, United Kingdom): Heriot-Watt University Press; 2013.
[150] Zhang PY, Huang S, Sayegh S, Zhou XL. Effect of CO2 impurities on gas-
injection EOR processes. In: SPE/DOE symposium on improved oil
recovery. Oklahoma (United States): Society of Petroleum Engineers; 2004.
[151] Bachu S. Review of CO2 storage efficiency in deep saline aquifers. Int J
Greenhouse Gas Control 2015;40:188202.
[152] Taku Ide S, Jessen K, Orr Jr FM. Storage of CO2 in saline aquifers: effects of
gravity, viscous, and capillary forces on amount and timing of trapping. Int J
Greenhouse Gas Control 2007;1:48191.
[153] Kopp A, Class H, Helmig R. Investigations on CO2 storage capacity in saline
aquifers: Part 1. Dimensional analysis of flow processes and reservoir
characteristics. Int J Greenhouse Gas Control 2009;3:26376.
[154] Kopp A, Class H, Helmig R. Investigations on CO2 storage capacity in saline
aquifersPart 2: Estimation of storage capacity coefficients. Int J Greenhouse
Gas Control 2009;3:27787.
[155] Xing W, Song Y, Zhang Y, Nishio M, Zhan Y, Jian W, et al. Research progress of
the interfacial tension in supercritical CO2-water/oil system. Energy Proc
2013;37:692835.
[156] Ranganathan P, Farajzadeh R, Bruining H, Zitha PJ. Numerical simulation of
natural convection in heterogeneous porous media for CO2 geological
storage. Transp Porous Med 2012;95:2554.
[157] Schmid M, Busbridge M, W est A. Double-diffusive convection in Lake Kivu.
Limnol Oceanogr 2010;55:22538.
[158] Green CP, Ennis-King J. Steady dissolution rate due to convective mixing in
anisotropic porous media. Adv Water Resour 2014;73:6573.
[159] Rackley SA. Carbon capture and storage. 1st ed. Oxford (United
Kingdom): Butterworth-Heinemann; 2009.
[160] Nicot J-P, Solano S, Lu J, Mickler P, Romanak K, Yang C, et al. Potential
subsurface impacts of CO2 stream impurities on geologic carbon storage.
Energy Proc 2013;37:45529.
[161] Jacquemet N, Le Gallo Y, Estublier A, Lachet V, von Dalwigk I, Yan J, et al. CO2
streams containing associated componentsa review of the thermodynamic
and geochemical properties and assessment of some reactive transport codes.
Energy Proc 2009;1:373946.