Applied Energy 210 (2018) 317–326

Contents lists available at ScienceDirect

Applied Energy
journal homepage: www.elsevier.com/locate/apenergy

An efficient CO2 adsorptive storage using MOF-5 and MOF-177 MARK

⁎
Naef A.A. Qasem, Rached Ben-Mansour, Mohamed A. Habib
Mechanical Engineering Department and KACST-TIC on CCS, King Fahd University of Petroleum & Minerals, Dhahran 31261, Saudi Arabia

H I G H L I G H T S

• An experimentally validated model was developed using UDF-ANSYS Fluent program.

• Two- and three-dimensional models are presented.
• 2D model is properly adequate to present the adsorptive storage processes.
• MOF-5 is a good choice for CO storage applications at P < 5 bars.
2

• MOF-177 is the perfect storage medium for the same purposes at P ≥ 10 bar.

A R T I C L E I N F O A B S T R A C T

Keywords: Carbon dioxide produced from carbon capture technologies could efficiently be stored into adsorbent materials.
Storage Certain types of metal organic frameworks called MOF-5 and MOF-177 are distinguished adsorbents amongst
Adsorption other porous materials owing to their high CO2 uptake at medium storage pressure (10–50 bars). CO2 adsorptive
Carbon-Dioxide storage and energy-consumption analysis at pressures between 5 and 50 bar were numerically investigated using
MOF-5
MOF-5 and MOF-177. An experimentally validated model was developed using User Defined-Function (written
MOF-177
CFD
in C) linked to ANSYS Fluent program. The CFD two and three dimensional models have been validated against
experimental data for adsorption storage characteristics of hydrogen on activated carbon. The developed model
is feasible to accurately simulate and represent gas adsorptive systems. The results of the multidimensional
modeling showed that the 2D model is properly adequate to present the adsorptive storage processes. The results
obtained from CO2 adsorptive storage using MOF-5 and MOF-177 emphasized that MOF-5 is a good choice for
CO2 storage applications at pressure less than 5 bar, while MOF-177 is a superior storage medium for the same
purposes at higher pressures (≥10 bar). The optimal storage pressure for both adsorbents may be taken as 30 bar
with CO2 uptake and energy consumption about 21.07 mmol g−1 (0.93 kg_CO2 kg_MOF−1) and
231 kJ tonne_CO2−1 for MOF-5, and about 32.5 mmol g−1 (1.43 kg_CO2 kg_MOF−1) and 233 kJ tonne_CO2−1 for
MOF-177, respectively. Importantly, the CO2 storage is the next step after its separation from a flue gas using any
separation method, and the first step prior to underground geological sequestration. The CO2 adsorptive storage
using MOF-177 showed an efficient method to store much more CO2 quantity with lower energy-consumption
than those of traditional storage systems (gas cylinders – without MOFs).

1. Introduction system. In addition, the increase of the atmospheric temperature is

predicted to increase about 6.4% at the end of the current century [1].
In the recent decade, environmental pollution is being identified as In this context, carbon dioxide holds the most significant proportion of
the most significant factor in the context of environmental health and the flue gases being released into the atmosphere from various sources
safety. In this scenario, the major portion of the environmental pollu- [2]. Therefore, in order to minimize and control the amount of CO2
tion arises from fossil fuel burning processes in the form of greenhouse emission in the atmosphere, a considerable amount of effort was made
gases, including carbon dioxide, nitrogen oxide, and methane. These by scientists, institutions, countries and environmental organizations.
gases play a continuous and vital role in facilitating universal en- Carbon dioxide is mainly emitted from fossil fuel combustion processes.
vironmental hazards like global warming that leads to shore floods, This source of energy (fossil fuel) plays a vital role for a comfortable
land droughts, and destruction of cold-marine life and the whole eco lifestyle and will be utilized for the coming few decades. Consequently,

⁎
Corresponding author.
E-mail address: mahabib@kfupm.edu.sa (M.A. Habib).

https://doi.org/10.1016/j.apenergy.2017.11.011
Received 1 July 2017; Received in revised form 8 October 2017; Accepted 2 November 2017
0306-2619/ © 2017 Elsevier Ltd. All rights reserved.
N.A.A. Qasem et al. Applied Energy 210 (2018) 317–326

Nomenclature t time, s
T temperature, K
C concentration, mol m−3 tss stoichiometric time, s
Cp specific heat capacity, J kg−1 K−1 τ stress tensor, N m−2
D, d diameter, size, m →
v velocity vector, m s−1
Ddisp mass dispersion coefficient, m2 s−1
Eg total gas energy, J m−3 Greek letters
Es total adsorbent energy, J m−3
h sensible enthalpy, J kg−1 α enthalpic factor, J mol−1
hext external heat transfer confident, W m−2 K−1 β intropic factors, J mol−1 K−1
hint internal heat transfer coefficient, W m2 K−1 ρ gas density, kg m−3
ΔH heat of adsorption, J mol−1 ε bed porosity
→
g gravity acceleration vector, m s−2 ∊ thermal emissivity
→ dynamic viscosity, Pa s−1
J diffusion flux, kg m−2 s−1 µ
k thermal conductivity, W m−1 K−1 σ Stefan Boltzmann coefficient, W m−2 K−4
kL LDF adsorption time constant, s−1
M molecular weight, kg mol−1 Subscript
n isotherms adsorption constant
q adsorption uptake, mmol g−1 eff effective
q∗ equilibrium adsorption uptake, mmol g−1 p particle
qm adsorption uptake limit, mmol g−1 g gas
QF feed volumetric flow rate at standard conditions, m3 s−1 0 inlet, saturation
P pressure, Pa max maximum
R universal gas constant, J mol−1 K−1 min minimum
r radius, m tot total
Re Reynolds number w wall

the only feasible solution to continue using fossil fuel with suppressing high porosity associated with high CO2 adsorption at ambient tem-
the CO2 emission is Carbon Capture and Storage (CCS). perature [25,28,29]. Incorporating MOFs with CNT could improve the
Many researchers have been conducted with Carbon Capture and equilibrium CO2 capacity; Cu3(btc)2 adsorbed about 295 mg g−1 at
Storage (CCS). A vast majority of these researchers has studied CO2 298 K and 18 bar, while CNT@Cu3(btc)2 adsorbed about 595 mg g−1 at
separation and storage, using both experimental and simulation the same conditions [30]. Porous carbons with well-developed pore
methods as well as developing novel adsorbents for this purpose [3]. structures showed the CO2 uptake was about 164 mg g−1 of at 1 bar and
The utmost merit of using adsorption for CO2 separation is the ease of about 1045 mg g−1 at 30 bar [31]. A new class of porous materials
regeneration of the adsorbent material by applying heat, decreasing the called covalent-organic frameworks (COFs) showed that COF-102 could
operating pressure, or both heat and decreasing the operative pressure isothermally adsorb 1180 mg g−1 at 35 bar and 298 K [32]. A new two
[4]. The research attempts based on CO2 capture and separation were graphene-based materials was developed [33] for energy-effective CO2
conducted so far in terms of breakthrough, pressure swing adsorption storage: PANI-f-HEG and Fe3O4-HEG. The adsorption uptake was 75
and temperature swing adsorption, which include both experimental and 60 mmol g−1 at 11 bar and 298 K for PANI-f-HEG and Fe3O4-HEG,
methods and numerically validated simulations [5–19]. Other swing respectively. However, these studies conducted with these two ad-
adsorption methods were also reported as: a novel electric swing ad- sorbents did not provide the adsorption heat of both materials; so that,
sorption process to CO2 degassing polyethylenimine-impregnated the thermal and energy performances cannot be predicted. Recently,
carbon spheres [20], a microwave swing regeneration used for CO2 Aminu et al. reviewed a CO2 storage undergrounds; they addressed the
chemical adsorption separation and minimizing consumption energy most challenges that face CO2 sequestration [34].
and amine degradation [21], and an elevated-temperature pressure Carbon dioxide equilibrium adsorption characterization is very
swing adsorption used for integrated gasification combined cycle with important to address the static CO2 adsorption capacities. However, in
carbon dioxide capture unit to improve system performance [22]. real dynamic adsorption applications, the CO2 uptake together with the
Activated carbon and zeolite are the most existent adsorbents that adsorption and thermal diffusions, thermal stability, and energy con-
were conducted in research for CO2 separation and storage. Activated sumption are critical [35]. Unlike CO2 separation processes, the well-
carbon materials have some advantages over zeolites including cost- known published research has not discussed the CO2 adsorptive storage
effectiveness, hydro-stability, lower energy required for regeneration in real applications. The literature has only focused on developing/
due to lower heat of adsorption [23]. A new class of adsorbents ware improving adsorbents in pursuit of adsorbing high quantity of CO2 at
discovered and made of metal-organic materials (called MOFs) [24]. It medium pressures (< 50 bar) which was characterized by adsorption
is recorded for CO2 adsorption since 1998 using MOF-2. The maximum isotherms. The only work concentrated on dynamic CO2 adsorptive
CO2 adsorption capacity was reported as 1470 mg g−1 at 35 bar and storage was 1D lumped simulation using MOF-5 [36]. However, the
room temperature using MOF-177 [25]. Carbon dioxide could be stored CO2/MOF-5 equilibrium adsorption parameters used in that model [36]
to a container contains MOF-177 is equivalent to that stored inside nine did not actually present MOF-5 at all. In addition, there was no thermal
traditional containers (without MOFs) [25]. Recently, many MOFs have and energy analysis research investigated for CO2 adsorptive storage
been synthesized by researchers to reach a proper formulation for se- systems. In simulation side, multi-dimensional CFD modeling for gases
parating and storing efficiently. More specifically, a total of about adsorptive storage was not presented well. Therefore, this work aims at
54,341 MOFs designed for these applications have been recorded in the two- and three-dimensional adsorptive storage modeling. Including two
Cambridge Structure Database before July 2015 [26]. For storage or three dimensions can lead to accurate heat and mass transfer and
purposes, a great potentials for the CO2 capacity at high pressure adsorption predictions. The new developed model will be an excellent
(20–40 bar) was MOF-177 and MOF-5 [27]. Reportedly, MOF-5 showed tool to simulate and represent gas adsorptive storage systems. The

318
N.A.A. Qasem et al. Applied Energy 210 (2018) 317–326

research will highlight the advantage of utilizing MOF-5 and MOF-177 ∂ →

[ερEg + (1−ε ) ρp Es] + ∇ ·[→
v (ρEg + P )] = ∇ ·[k eff ∇T −hJ + τ ·→
v]
for CO2 adsorptive storage and will explore the optimal storage pressure ∂t
for the both adsorbents. Thermal and energy consumption analysis will ∂q
+ (1−ε ) ρp ΔH ,
be assessed as well. The selection of MOF-5 and MOF-177, in this work ∂t (4)
for CO2 storage, is not only because they have distinguished CO2 up-
where Eg (J m−3) is the total gas energy, Es (J m−3) is the total ad-
takes at medium pressures (5–50 bar), but also they are thermally stable
sorbent energy, ΔH (J mol−1) is the heat of adsorption which is dif-
up to 470 °C and 380 °C for MOF-5 and MOF-177, respectively [27].
ferent from material to another and sometimes also varies with varia-
tion of gas adsorbed amounts, T (K) is the equilibrium temperature, h
2. Numerical modeling methods →
(J kg−1) is the sensible enthalpy, and J (kg m−2 s−1) is the diffusion
flux of the gas.
2.1. Model description
The parameter keff is the effective conductivity of the adsorbent bed
and can be expressed as:
The simulation model is based on the momentum, mass and energy
conservation equations. The CFD model was developed using User k eff = εk g + (1−ε ) ks, (5)
Define-Function (UDF, written in C) linked to ANSYS Fluent software
[37] to estimate carbon dioxide adsorption quantities and adjusting the where kg and ks (W m−1 K−1) are the thermal conductivities of the gas
source terms of mass, momentum and energy equations. The model is mixture and adsorbent, respectively.
proposed to be valid for multi-dimensional and various geometries. For the wall, the energy equation through the walls depends upon
For all studied cases, the following assumptions are adopted: the balance of heat stored in the wall and the heat diffusion through it
as follows:

• the porous media is homogenous, ∂

• the physical properties of the adsorbents are constant, and (ρ Cw Tw ) = ∇ (k w ∇Tw ),
∂t w (6)
• the linear driving force (LDF) model is used to account for the mass where ρw (kg m−3) is the wall density, Cw (J kg−1 K−1) is the wall heat
transfer rate during the adsorption process.
capacity, Tw (K) is the local temperature of the wall, and kw
In addition, the isotherms were accurately fitted from the experi- (W m−1 K−1) is the thermal conductivity of the wall material.
mental ones according to Dubnin-Astakhov model.
2.5. Adsorption Isotherm and Kinetics Models
2.2. Mass conservation equation
The equilibrium adsorption isotherms are presented by different
The mass conservation equation might be calculated as the fol- models according to the accuracy of the fitting from the experimental
lowing: isotherms. Dubnin-Astakhov model are exploited for gas equilibrium
adsorption as:
∂ (ρ) ∂q
ε + ∇ ·(ρ→
v ) = −(1−ε ) ρp M , n
∂t ∂t (1) −⎡ RT ln P 0 ⎤
( )
q∗ = qm e ⎣ α + βT P ⎦ , (7)
→
where ρ (kg m−3) is the gas density, v (m s−1) is the velocity vector, ε
−1
is the bed porosity, M (kg mol ) is the molecular weight of the gas, q where qm (mmol g−1) is the limited adsorbed amount of the gas, R
(mmol g−1) is the adsorbent amount of the gas, and t (s) is the time. (J mol−1 K−1) is the universal gas constant, α (J mol−1) and β
(J mol−1 K−1) are enthalpic and intropic factors, P0 (Pa) is the sa-
turation pressure of the pure gas at ambient temperature, and n is a
2.3. Momentum conservation equation
constant.
The adsorption kinetics are evaluated by linear driving force (LDF)
The momentum equation of the gas flow can be expressed as:
model as follows.
∂ (ρ→v)
ε + ∇ ·(ρ→→ → + S,
v v ) = −∇P + ∇ ·τ + ρg ∂qi
∂t (2) = kL (q∗−q)
∂t (8)
→
where P (Pa) is the operating pressure, τ (N m−2) is the stress tensor, g −1
−2 −3 The parameter q (mmol g ) presents the actual dynamic adsorbed
(m s ) is the gravity acceleration vector, and S (N m ) is the mo-
amount while q∗ (mmol g−1) is the equilibrium adsorbed amounts. kL
mentum source tem in the porous media. Its component in i-direction
(s−1) is the adsorption time constant; it is estimated from microporous
could be calculated from Ergun equation including inertia and viscus
resistance of the gas passing through porous materials as:
resistances as:
15Dc
μ 1 kL = ,
Si = −⎛ vi + C2 ρ|→
v |vi⎞, rc2 (9)
⎝κ 2 ⎠ (3)

where µ (Pa s) is the gas dynamic viscosity, 1/κ (m−2) is the porous where rc (m) is the adsorbent crystal radius, Dc (m2 s−1) is the inter-
media viscous resistance, C2 (m−1) is the inertial resistance, |→
v | is the microporous diffusion coefficient. The adsorption isotherm and kinetic
value of the velocity vector, and vi is the velocity component in i-di- properties are usually obtained from an experimental characterization.
rection. The obtained parameters from the experimental fittings are qm, α, β, n
(for calculation equilibrium adsorption uptake, q∗) and D2c (for calcu-
rc
2.4. Energy conservation equation lating LDF constant, kL).

The energy equation for CO2 separation shows the balance between 2.6. Boundary conditions
the energy stored in the adsorbent bed and the change in energy arising
from convective flow, pressure work, thermal diffusion and advection, 2.6.1. External walls
as well as the energy released/consumed as a consequence of adsorp- The heat transfer between the bed wall and the ambient can be
tion/desorption processes: expressed as:

319
N.A.A. Qasem et al. Applied Energy 210 (2018) 317–326

∂T Table 1
k w ⎛ w ⎞ = hext (Tw−Tamb) + ∊ σ (Tw4−Tamb
4
), Polynomial constants for gas components thermal properties.
⎝ ∂n ⎠w (10)

where n (m) is the normal, hext (W m−2 K−1) is the external heat Constants of Cp Constants of k Constants of µ
transfer confident, ∊ is the thermal emissivity, σ (W m−2 K−4) is Stefan equation equation equation

Boltzmann coefficient. A = 491.702 A = −0.00694154 A = −1.10128e−06

The heat transfer from the bed wall to the ambient relies on the heat B = 1.43603 B = 7.53746e−05 B = 5.89863e−08
transfer coefficient, which may be calculated from Nusselt number C = −0.0007742 C = 7.53746e−05 C = −1.74847e−11
using the following correlation [38]:
2
Cp = A + BT + CT 2 (J kg−1 K−1), (21)
⎡ ⎤
h D ⎢ Ra1/6 ⎥
Nu = ext 0 = ⎢0.60 + 0.387 ⎥, k = A + BT + CT 2 (W m−1 K−1), (22)
kair 9/16 8/27
⎢
⎢
⎣
⎡1 +
⎣
( )
0.559
Prair
⎤ ⎥
⎦ ⎥ ⎦ (11) μ = A + BT + CT 2 (Pa s). (23)
−1 −1
where Do (m) is the external wall diameter, kair (W m K ) is the The constants of Eqs. (21)–(23) are shown in Table 1.
thermal conductivity of the air at average temperature (of both the wall For stainless steel material, ρ = 7941 kg m−3,
and the ambient temperatures), Prair is the air Prandtl number, Ra is −1 −1
Cw = 358.98 + 0.487394 T − 2.65708e−07 T J kg K , and k =
2

Rayleigh number at average temperature. 15.14 W m−1 k−1.

For adiabatic walls: Generally, the thermal inertia of the bed is:

⎛ ∂Tw ⎞ = 0. (ρCp)bed = (ρCp)gas + ρs (Cp,s + qMCp,g ). (24)

⎝ ∂n ⎠w (12)
The last term in Eq. (24) contains the heat capacity of adsorbed
amounts. The molecular weight, M, should be in kg mol−1.
2.6.2. Interfaces between walls and adsorbent
The convection heat transfer through the interface between the bed 2.8. Model validation and adsorptive storage cases
and wall is estimated by the local heat conduction Eq. (13).
∂T ∂T 2.8.1. Validation
k eff ⎛ ⎞ = hint (T−Tw ) = klocal ⎛ ⎞ The adoptive storage model can be used to simulate mass, heat and
⎝ ∂n ⎠w ⎝ ∂n ⎠local (13)
adsorption behaviors of any gas/adsorbent pair (such as H2/AC, CO2/
Here, hint (W m2 K−1) is the internal convection heat transfer coef- AC, CO2/MOFs, CH4/MOFs). The difference from gas/adsorbent storage
ficient calculated automatically by Fluent from the last term in Eq. (13). system to another is only the adsorption and thermal properties.
Also In this text, we have validated the present two- and three-dimen-
∇ρ = 0. (14) sional models against the experimental data of H2 adsorptive storage
using activated carbon, Ye et al. [39]. We have selected H2 adsorptive
storage system due to unavailability of experimental data for CO2 ad-
2.6.3. Inlet boundary conditions during gas charging sorptive storage tank in the published literature. The storage tank is
The mass and heat transfer at the bed inlet could be expressed as the shown in Fig. 1 with small inlet tube for feeding the gas to the reservoir.
diffusion quantities in balance with those corresponding to advection. The tank, activated carbon, and H2 properties are described in Table 2.
The inlet mass fluxes during tank charging, dormancy, discharging and
∇ ·(ρ→
v)=0 (15)
dormancy are listed in Table 3.
The isotherm model used here is Dubnin-Astackov (Eq. (7)). The
−εk eff ∇T = →
v ρ (hi|− −hi|+ ) (16)
main parameters of this model for AC are listed in Table 4. LDF coef-
ficient was taken as 0.15 s−1 [39].
2.6.4. Outlet boundary conditions during gas discharging
The mass and heat fluxes at the outlet can be written as: 2.8.2. Carbon dioxide storage in MOF-5
∇ρ = 0, The same tank used in validation case was used for MOF-5 to si-
(17)
mulate the effect of pressure on the CO2 adsorptive storage. Thermal
∇T = 0. (18) properties of MOF-5, isotherm parameters, and LDF model are de-
scribed in Table 5. The heat transfer from the wall of the cylinder to the
ambient was evaluated by free convection and radiation mechanisms,
2.6.5. Centerlines (in 2D) and symmetry planes (in 3D)
while the small inlet tube wall was adiabatic.
The mass and heat fluxes at the axisymmetric axis and symmetry
Six cases have been investigated by changing the inlet pressure as 5,
planes are:
10, 20, 30, 40, and 50 bars as shown in Fig. 2. The first 500 s were
∇ρ = 0, (19) dedicated for CO2 charging process, while the remains were left for
dormancy.
∇T = 0. (20)

2.8.3. Carbon dioxide storage in MOF-177

2.7. Materials properties MOF-177 has been used for CO2 adsorptive storage; it has been si-
mulated using the same cases and bed configuration that used for MOF-
The used materials in the models were adsorbents (AC, MOF-5 and 5. The important thermal properties of MOF-177, CO2 isotherms para-
MOF-177), H2 and CO2 gases, and stainless steel wall. The thermal meters, and LDF models are tabulated in Table 6.
properties for CO2 and wall material have been taken as a polynomial Like MOF-5, six cases have been investigated by changing the inlet
equation as a function of temperature. The gas density has been cal- pressure as 5, 10, 20, 30, 40, and 50 bars as shown in Fig. 2.
culated using ideal gas law. Fig. 3 shows a general scheme of the model details that were used
Thermal properties for CO2 are: for both H2 (for validation using AC) and CO2 (for study cases using

320
N.A.A. Qasem et al. Applied Energy 210 (2018) 317–326

Fig. 1. Scheme of adsorptive storage tank.

(a) 2D adsorptive tank.

(b) 3D adsorptive tank.

Table 2 Table 5
Tank, AC, and H2 properties used in the numerical modeling at the same conditions of the Thermal and adsorptive properties of MOF-5/CO2.
experimental work [39].
Properties Value
Tank diameter (m) 0.0928 Pipe diameter (m) 0.008
Tank wall thickness (m) 0.0039 Pipe wall thickness (m) 0.001 Particle density, ρ 621 [40]
Tank length (m) 0.4 Pipe length (m) 0.05375 Bed void fraction, ε 0.31
Wall density (kg m−3) 7830 Wall specific heat 468 Solid Specific heat, Cs (J kg−1 K−1) 750 [41]
(J kg−1 K−1) LDF coefficient 0.025 [27]
Wall thermal conductivity 13 AC density (kg m−3) 517.6 CO2 adsorption heat ΔH (J mol−1) −15675.971−9.2906q−19.001q2 [27]
(W m−1 K−1) qm (mmol g−1) 23 [27]
AC specific heat capacity 825 AC thermal conductivity 0.646 P0 (MPa) 72.14 [27]
(J kg−1 K−1) (W m−1 K−1) α (J mol−1) 20 [27]
Bed porosity 0.49 H2 specific heat capacity 14,700 β (J mol−1 K−1) 37.8 [27]
(J kg−1 K−1) n 7 [27]
H2 thermal conductivity 0.206 H2 viscosity (Pa s) 8.411e−6 Bed density (kg m−3) 428.5
(W m−1 K−1) MOF-5 thermal conductivity (W m−1 K−1) 0.32 [42]

Table 3 60
Inlet mass flux of PSA validation [39].
50 50 bar
Time (s) Mass flux (kg s−1 m−2)
Storage pressure (bar)

0–442 0.866399 40 40
442–3046 0
3046–3907 −0.43489069
30 30
3907–6000 0

20 20
MOF-5 and MOF-177) adsorptive storage tanks.
10 10
2.9. Adsorptive storage performance 5
0
For more useful performance analysis, the estimation of CO2 storage 0 500 1000 1500 2000 2500 3000 3500
energy consumption has been considered as written in Eq. (25) [46]. Time (s)
γ−1 Fig. 2. Inlet pressure for CO2 adsorption storage cases at 300 K.
γ QF εtot CR T ⎡ ⎛ Pmax ⎞ γ ⎤
Energy Consumption = ∫ γ −1 η
⎢ ⎜ ⎟ −1⎥ dt
⎢ ⎝ Pmin ⎠ ⎥
⎣ ⎦ (25) from the inlet velocity and the cross section area of the inlet pipe; the
terms “QF C” represents the molar flow rate of the gas. εtot is the total
Here, η is the compressor efficiency which assumed to be 0.72, γ is
bed porosity (εtot = ε + (1−ε ) εp ); it is added here to only consider the
the specific heat capacity of gas, C (mol m−3) is the gas concentration,
tank places where the compressed CO2 occupied. Pmin is the CO2
QF (m3 s−1) is the volumetric flow rate inters the tank – it is calculated

Table 4
Dubnin-Astackov parameters for H2 isotherms of AC [39].

Gas species qm (mmol g−1) P0 (MPa) α (J mol−1) β (J mol−1 K−1) n ΔH (J mol−1)

H2 71.6 1408 3080 18.9 2 3185

321
N.A.A. Qasem et al. Applied Energy 210 (2018) 317–326

Table 6 10
Thermal and adsorptive properties of MOF-177/CO2. 9 Exp.
2D Sim.
8
Properties Value 3D Sim.
7

Pressure (MPa)
Particle density, ρ 477 [40] 6
Bed void fraction, ε 0.44
5
Solid Specific heat, Cs (J kg−1 K−1) 490 [43]
LDF coefficient 0.1597 [44] 4
CO2 adsorption heat ΔH (J mol−1) −14,000 [45] 3
qm (mmol g−1) 37.4 [27]
2
P0 (MPa) 72.14 [27]
α (J mol−1) 100 [27] 1
β (J mol−1 K−1) 34.8 [27] 0
n 7 [27]
0 1000 2000 3000 4000 5000 6000 7000
Bed density (kg m−3) 239.1 Time (s)
MOF-177 thermal conductivity (W m−1 K−1) 0.3
Fig. 4. A comparison of storage pressure between experimental work [39] and present 2D
and 3D simulations for H2 adsorptive storage.

3.1. Model validation and 2D and 3D comparison

A numerical modeling has mainly been developed for CO2 ad-

sorptive storage processes. Firstly, the 2D and 3D CFD models were
compared to the experimental data [39] that carried out for storing
hydrogen into activated carbon at almost 100 bar. The adsorbed bed
parameters and adsorbent/adsorbate characterization are described in
Tables 2–4. The measured parameters, as the average operating pres-
sure (Fig. 4) and the temperature at the bed center point
(z = 0.27875 m, x = 0 m, y = 0 m) (Fig. 5), have been compared well
with those of the present 2D and 3D simulations for the all stages:
charging, dormancy, discharging, and dormancy. This confirms that the
UDF-Fluent multi-dimensional modeling can efficiently predict the
adsorption storage behaviors.
Fig. 6 displays the temperature contours of the 2D and 3D simula-
tion during charging and cooling (dormancy) stages at some selected
times (110, 290, 500, 1010, 2000, and 2990 s). The temperature values
have increased due to increasing H2 uptake values, which were arisen
by increasing the pressure values. The maximum temperature (about
340 K) have been addressed immediately after the end of the charging
time (about 500 s); and then, the storage tank was cooled down to
about 300 K by the surrounding ambient – during 2500 s. The tem-
perature difference between the 2D and 3D modeling was not sig-
nificant. That small difference occurs due to respecting the thermal and
mass diffusion in all the three dimensions (for 3D case). Hydrogen
uptake contours are presented in Fig. 7. It is obvious that the adsorbed
H2 amounts have gradually grown from start-to-end of the CO2 char-
ging process. Then, during the cooling process, the H2 amounts were
almost constant (about 8 mmol g−1) due to no H2 entered to or left from
the tank. A little increase in H2 uptake values, occurred during the

355

345 Exp.
Fig. 3. Schematic details of the modeling system.
2D Sim.
335
Temperature (K)

3D Sim.
pressure coming from separation processes; it is assumed to be as the 325
ambient pressure (101.325 kPa). Pmax is the compressed CO2 pressure 315
that inters the storage tank. The term “C R T” can also be presented by
305
the bed average pressure (P = CRT) which is very close to Pmax.
295

285
3. Results and discussion
275
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500
Firstly, the code results have been compared to those of experi- Time (s)
mental data for showing the reliability of UDF-Fluent model, and then
Fig. 5. A comparison of temperature history of the bed (at x = 0 m, y = 0 m,
the CO2 adsorptive storage has been investigated. The fine mesh of
z = 0.27875 m) between experimental data [39] and present 2D and 3D simulations for
4847 cells was selected for 2D geometry after ensuring that no effect on H2 adsorptive storage.
the results when adding more cells [47]. For 3D bed, the cell size have
been taken similar to that used for 2D.

322
N.A.A. Qasem et al. Applied Energy 210 (2018) 317–326

Fig. 6. 2D and 3D temperature contours of H2

adsorptive storage tank for AC at some selected
times, during charging and dormancy processes.

dormancy process, was because of that some amounts of H2 in the tank 460
20, 30, 40, 50 bar
void were being adsorbed again. 440
The produced CO2 from separation processes should be firstly stored
to be utilized later for different applications (such as enhancing oil 420
production by CO2 sequestration process). The best way to store more
Temperature (K)

400
quantities of CO2 in a confined container is by adsorbing it into a high
CO2-capacity adsorbent such as MOF-5 and MOF-177. 380
As shown in Figs. 4–7, the two-dimensional adsorption storage 360 10 bar
model (in cylindrical tank) is sufficient to simulate such systems, hence
we have only carried out a 2D simulations for the CO2 adsorptive sto- 340
rage using the same geometry of the adsorptive tank (Fig. 1(a)). 320 5 bar

300
3.2. CO2 adsorptive storage using MOF-5 0 10 20 30 40 50 60
Time (min)
Adsorbent MOF-5 exhibits a substantial CO2 adsorption capacity at Fig. 8. Temperature profiles of CO2 adsorptive tank at the tank center for different sto-
high storage pressures; so that, CO2 adsorptive storage using MOF-5, at rage pressures for MOF-5.
different pressures, have been concentrated in this research. The

Fig. 7. 2D and 3D H2 uptake into AC at some

selected times, during charging and dormancy
processes.

323
N.A.A. Qasem et al. Applied Energy 210 (2018) 317–326

charging process has taken place at the first 500 s, and has been fol- best CO2 adsorption capacity and energy performances.
lowed by cooling process for about 2500 s for all the investigated ad- The temperature profiles at the middle of the tank point for all the
sorptive pressures (5, 10, 20, 30, 40, and 50 bar), as shown in Fig. 2. investigated pressure values are plotted in Fig. 12. The temperature
Fig. 8 shows the temperature profiles at the middle of the tank for values have increased rapidly from the exothermic reaction (adsorption
all the studied pressure values. The maximum temperature values have process) as the CO2 was adsorbed more into MOF-177: when increasing
been associated with the highest storage pressure due to the increase of the charging pressure. In contrast, the temperature values has explicitly
CO2 adsorbed amounts into MOF-5: which in turn leads to release more dropped down more for the higher pressure cases (such as the cases of
heat. The adsorption process is exothermic; therefore, the more CO2 50, 40, and 30 bar) than those of the lower pressure cases (such as the
adsorbed, the higher heat released. Oppositely, during the dormancy case of 20 bar) due to reaching adsorption limits (saturations) earlier,
process, the middle of the bed was rapidly cooled down for higher and thereby starting cooling formerly (for high pressure cases). How-
pressure cases (such as the case of 50 bar) in comparison to those of ever, for the low pressure cases, a few amounts of CO2 were adsorbing
lower pressure cases (such as the case of 20 bar). This effect is due to from the tank void into MOF-177 leading to slow down the cooling
that the central region of the tank reached adsorption saturation earlier, process.
and hence stopping the adsorption (for the high pressure cases); while a Fig. 13 illustrates the CO2 uptake for all studied storage pressures.
little amount of CO2 was still being adsorbed to the adsorbent from the During charging period (< 8.3 min), the CO2 uptake has inherently
tank void – regarding to the low pressure cases – leading to slow down grown up (for high-pressure cases). Generally, The CO2 uptake values
the bed cooling. The dramatic increase in the temperature values for increase dramatically when increasing the pressure values from 5 to
cases 20–50 bar – during charging process – is mainly because of that 30 bar, while those have steadily grown when the pressure values
MOF-5 adsorbs more CO2 at higher pressures, as shown in Fig. 9. climbed from 30 to 50 bar. This is due to the relating trends of CO2
Fig. 9 explains the CO2 uptake for all aimed storage pressures. The equilibrium adsorption (q∗, obtained from the isotherms [27]) with
higher is storage pressure (up to 50 bar), the more is CO2 uptake. The actual dynamic adsorption (q) against the storage pressure for MOF-
CO2 adsorbed amounts have increased sharply when the pressure in- 177. Fig. 14 shows the CO2 uptake as mass ratio (CO2 adsorbed amount
creased from 5 to 20 bar. After that, the CO2 uptake values rose slowly per kilogram of the adsorbent (MOF-177)). It is shown that the case of
for the pressure values allocated form 20 to 50 bar due to the asymp- 5 bar has only enabled the adsorbent to adsorb about 0.07 kg kg−1
totic behavior of equilibrium CO2 uptake (q∗) against pressure for MOF- while that has been harshly increased to about 1.45 kg kg−1 for 30 bar.
5, which was taken from the adsorption isotherms [27]. Indeed, the CO2 The CO2 uptake had a slight increase after 30 bar reaching 1.58 kg kg−1
isotherm of MOF-5 shows that the equilibrium CO2 uptake (q∗) in- at 50 bar. In addition, the storage power consumption has grown from
creases sharply between 5 and 20 bar and slightly after 20 bar [27]. 121 kJ tonne_CO2−1 for 5 bar to about 233 kJ tonne_CO2−1 for 30 bar,
Then, the actual dynamic CO2 uptake (q) values vary as close trend as Fig. 15. This has been followed by a little increase in the energy con-
those of equilibrium adsorption values (q∗). sumption to about 255 kJ tonne_CO2−1 for 50 bar. Like MOF-5, the
Fig. 10 exhibits the percentage of the adsorbed CO2 amounts per adsorptive pressure 30 bar is fairly satisfactory for CO2 adsorptive sto-
kilogram of the adsorbent (MOF-5). It is obvious that MOF-5 can only rage with an excellent energy and storage performances. It is clear, also,
adsorb about 0.13 kg kg−1 (at 5 bar), while that increased sharply to that MOF-177 is superior for CO2 storage purposes more than those of
about 0.93 kg kg−1 (at 30 bar). Then, the CO2 uptake has slowly MOF-5, especially at high pressure values. Nevertheless, the storage
reached 0.99 kg kg−1 at 50 bar. Moreover, the power consumption due performance at low pressure conditions (5 bar) is higher for MOF-5.
to adsorptive storage has dramatically increased from 119 kJ ton-
ne_CO2−1 (at 5 bar) to about 231 kJ tonne_CO2−1 (at 30 bar) and; fol-
lowed by a little increase to about 255 kJ tonne_CO2−1 (at 50 bar), 4. Conclusions
Fig. 11. All these figures recommend that the storage pressure of 30 bar
is fairly sufficient for CO2 adsorptive storage using MOF-5 – with a good Carbon dioxide adsorptive storage into MOF-5 and MOF-177 has
energy and storage performance. been numerically investigated in this work. A CFD model was devel-
oped using User-Defined-Function (written by C) linked to ANSYS
3.3. CO2 adsorptive storage using MOF-177 Fluent program for implementing the adsorption isotherm and kinetic
models and; adjusting the source terms of mass, momentum and energy
Metal organic framework, MOF-177, has an excellent CO2 adsorp- equations. The developed model has been used to carry out two- and
tion capacity at high pressures [27]; so that, in this numerical modeling, three-dimensional simulations. The results of the developed UDF-Fluent
it is exploited to investigate the optimal storage pressure to obtain the have been validated against experimental data conducted for hydrogen
adsorptive storage using activated carbon. The multidimensional
30 modeling has shown that the 2D model is accurately sufficient to pre-
20, 30, 40, 50 bar sent the CO2 adsorptive storage processes for the geometry chosen.
25 MOF-5 has shown a good performance of CO2 uptake and energy con-
sumption at high pressures: for the optimal pressure (about 30 bar), the
20 CO2 uptake and energy consumption were about
q (mmol/g)

0.93 kg_CO2 kg_MOF−1 and about 231 kJ tonne_CO2−1, respectively.

15 MOF-177 has revealed a high CO2 adsorption capacity of about
10 bar
1.43 kg_CO2 kg_MOF−1 with reasonable energy consumption, about
10 233 kJ tonne_CO2−1 at 30 bar. MOF-5 is a good choice for CO2 storage
at low pressures which is equivalent to 5 bar, while MOF-177 is the
5 5 bar perfect adsorbent to store CO2 at high pressures (≥10 bar). The sepa-
rated CO2 from the flue gas can be efficiently stored into MOF-177
0 storage tanks instead of the traditional tanks (cylinders).
0 10 20 30 40 50 60
Time (min)
Fig. 9. CO2 uptake at different storage pressures for MOF-5.

324
N.A.A. Qasem et al. Applied Energy 210 (2018) 317–326

120% 180%
153% 158%
97% 99% 160%
% Adsorbed amounts

100% 93% 143%

(kg_CO2 kg_MOF-1)

82% 140%

% Adsorbed amounts
(kg_CO2 kg_MOF-1)
80% 119%
120%
60% 49% 100%
80%
40%
60% 54%
20% 13%
40%
0% 20% 7%
5 10 20 30 40 50
Storage Pressure (bar) 0%
5 10 20 30 40 50
Fig. 10. Adsorbed CO2 amounts per MOF-5 amount ratio for different storage pressures.
Storage Pressure (bar)
Fig. 14. CO2 adsorptive amounts per MOF-177 amounts for different storage pressures.
300
255
244 300
250 231
Energy consumption

211 251

Energy consumption
243
(kJ tonne_CO2-1)

250 233

(kJ tonne_CO2-1)
200 169 215
200 174
150 119
150 121
100
100
50
50
0
5 10 20 30 40 50 0
Storage Pressure (bar) 5 10 20 30 40 50
Storage Pressure (bar)
Fig. 11. CO2 storage energy consumption at different storage pressures for MOF-5.
Fig. 15. CO2 storage energy consumption at different storage pressures for MOF-177.

440
20, 30, 40, 50 bar Acknowledgements
420
We thank and acknowledge support from King Abdulaziz City for
400
Science and Technology (KACST), the Carbon Capture and
Temperature (K)

380 Sequestration Technology Innovation Center (CCS-TIC #32-753) under

Project CCS10, and the Deanship of Research at KFUPM.
360
References
340 10 bar

320 [1] Lee S-Y, Park S-J. A review on solid adsorbents for carbon dioxide capture. J Ind Eng
5 bar
Chem 2015;23:1–11.
[2] D'Alessandro DM, McDonald T. Toward carbon dioxide capture using nanoporous
300
0 10 20 30 40 50 60 materials. Pure Appl Chem 2010;83(1).
[3] Ben-Mansour R, Habib MA, Bamidele OE, Basha M, Qasem NAA, Peedikakkal A,
Time (min) Laoui T, Ali M. Carbon capture by physical adsorption: Materials, experimental
Fig. 12. Temperature profiles of CO2 adsorptive tank at the tank center for different investigations and numerical modeling and simulations – a review. Appl Energy
2016;161:225–55.
storage pressures for MOF-177.
[4] Songolzadeh M, Ravanchi MT, Soleimani M. Carbon dioxide capture and storage: a
general review on adsorbents. World Acad Sci, Eng Technol 2012;70:225–32.
[5] Biswas P, Agrawal S, Sinha S. Modeling and simulation for pressure swing ad-
sorption system for hydrogen purification. Chem Biochem Eng Quart
2010;24(4):409–14.
[6] Casas N, Schell J, Pini R, Mazzotti M. Fixed bed adsorption of CO2/H2 mixtures on
activated carbon: experiments and modeling. Adsorption 2012;18(2):143–61.
40 [7] Cavenati S, Grande CA, Rodrigues AE. Separation of mixtures by layered pressure
30, 40, 50 bar
swing adsorption for upgrade of natural gas. Chem Eng Sci 2006;61(12):3893–906.
35 [8] Chaffee AL, Knowles GP, Liang Z, Zhang J, Xiao P, Webley PA. CO2 capture by
adsorption: Materials and process development. Int J Greenhouse Gas Control
30 2007;1(1):11–8.
[9] Cho S-H, Park J-H, Beum H-T, Han S-S, Kim J-N. A 2-stage PSA process for the
25
recovery of CO2 from flue gas and its power consumption*. In: Carbon Dioxide
q (mmol g-1)

20 bar Utilization for Global Sustainability, Proceedings of 7ththe International Conference

20
on Carbon Dioxide Utilization. Elsevier BV; 2004. p. 405–10.
[10] Choi W-K, Kwon T-I, Yeo Y-K, Lee H, Song HK, Na B-K. Optimal operation of the
15
10 bar pressure swing adsorption (PSA) process for CO2 recovery. Korean J Chem Eng
2003;20(4):617–23.
10
[11] Dantas TL, Amorim SM, Luna FMT, Silva Jr IJ, de Azevedo DC, Rodrigues AE,
Moreira RF. Adsorption of carbon dioxide onto activated carbon and nitrogen-en-
5
5 bar riched activated carbon: surface changes, equilibrium, and modeling of fixed-bed
0 adsorption. Sep Sci Technol 2009;45(1):73–84.
0 10 20 30 40 50 60 [12] Dantas TLP, Luna FMT, Silva IJ, de Azevedo DCS, Grande CA, Rodrigues AE,
Time (min) Moreira RFPM. Carbon dioxide–nitrogen separation through adsorption on acti-
vated carbon in a fixed bed. Chem Eng J 2011;169(1–3):11–9.
Fig. 13. CO2 uptake at different storage pressures for MOF-177. [13] Dantas TLP, Luna FMT, Silva IJ, Torres AEB, de Azevedo DCS, Rodrigues AE,
Moreira RFPM. Carbon dioxide–nitrogen separation through pressure swing

325
N.A.A. Qasem et al. Applied Energy 210 (2018) 317–326

adsorption. Chem Eng J 2011;172(2–3):698–704. capacities by lithium doping. Angewandte Chemie International Edition 2011;
[14] Gomes VG, Yee KWK. Pressure swing adsorption for carbon dioxide sequestration 50(2): p. 491–494.
from exhaust gases. Sep Purif Technol 2002;28(2):161–71. [31] Meng L-Y, Park S-J. Influence of MgO template on carbon dioxide adsorption of
[15] Krishnamurthy S, Rao VR, Guntuka S, Sharratt P, Haghpanah R, Rajendran A, cation exchange resin-based nanoporous carbon. J Colloid Interface Sci
Amanullah M, Karimi IA, Farooq S. CO 2 capture from dry flue gas by vacuum swing 2012;366(1):125–9.
adsorption: A pilot plant study. AIChE J 2014;60(5):1830–42. [32] Furukawa H, Yaghi OM. Storage of hydrogen, methane, and carbon dioxide in
[16] Lee C-H, Yang J, Ahn H. Effects of carbon-to-zeolite ratio on layered bed H2 PSA for highly porous covalent organic frameworks for clean energy applications. J Am
coke oven gas. AIChE J 1999;45(3):535–45. Chem Soc 2009;131(25):8875–83.
[17] Park J-H, Kim J-N, Cho S-H. Performance analysis of four-bed H2 PSA process using [33] Szczesniak B, Choma J, Jaroniec M. Gas adsorption properties of graphene-based
layered beds. AIChE J 2000;46(4):790–802. materials. Adv Coll Interface Sci 2017;243(Supplement C):46–59.
[18] Wang L, Liu Z, Li P, Yu J, Rodrigues AE. Experimental and modeling investigation [34] Aminu MD, Nabavi SA, Rochelle CA, Manovic V. A review of developments in
on post-combustion carbon dioxide capture using zeolite 13X-APG by hybrid VTSA carbon dioxide storage. Appl Energy 2017.
process. Chem Eng J 2012;197:151–61. [35] Belmabkhout Y, Guillerm V, Eddaoudi M. Low concentration CO2 capture using
[19] Wang L, Yang Y, Shen W, Kong X, Li P, Yu J, Rodrigues AE. Experimental evaluation physical adsorbents: Are metal–organic frameworks becoming the new benchmark
of adsorption technology for CO2 capture from flue gas in an existing coal-fired materials? Chem Eng J 2016;296:386–97.
power plant. Chem Eng Sci 2013;101:615–9. [36] Ben-Mansour R, Olufemi Eyitope B, Antar MA. Simulation of adsorptive storage of
[20] Wang M, Yao L, Wang J, Zhang Z, Qiao W, Long D, Ling L. Adsorption and re- CO2 in fixed bed of MOF-5. J Energy Res Technol 2015;138(1). 012001-012001-12.
generation study of polyethylenimine-impregnated millimeter-sized mesoporous [37] Ansys-Fluent. http://www.ansys.com; 2016.
carbon spheres for post-combustion CO2 capture. Appl Energy 2016;168:282–90. [38] Incropera; Frank PD, David P. Fundamentals of Heat and Mass Transfer. J. Wiley;
[21] McGurk SJ, Martín CF, Brandani S, Sweatman MB, Fan X. Microwave swing re- 2007.
generation of aqueous monoethanolamine for post-combustion CO2 capture. Appl [39] Ye F, Xiao J, Hu B, Benard P, Chahine R. Implementation for Model of Adsoptive
Energy 2017;192:126–33. Hydrogen Storage Using UDF in Fluent. Phys Procedia 2012;24:793–800.
[22] Zhu X, Shi Y, Cai N. Integrated gasification combined cycle with carbon dioxide [40] Mason JA, Veenstra M, Long JR. Evaluating metal-organic frameworks for natural
capture by elevated temperature pressure swing adsorption. Appl Energy gas storage. Chem Sci 2014;5(1):32–51.
2016;176:196–208. [41] Kloutse FA, Zacharia R, Cossement D, Chahine R. Specific heat capacities of MOF-5,
[23] Plaza MG, González AS, Pevida C, Pis JJ, Rubiera F. Valorisation of spent coffee Cu-BTC, Fe-BTC, MOF-177 and MIL-53 (Al) over wide temperature ranges:
grounds as CO2 adsorbents for postcombustion capture applications. Appl Energy Measurements and application of empirical group contribution method. Micropor
2012;99:272–9. Mesopor Mater 2015;217:1–5.
[24] Li J-R, Sculley J, Zhou H-C. Metal-organic frameworks for separations. Chem Rev [42] Liu D, Purewal JJ, Yang J, Sudik A, Maurer S, Mueller U, Ni J, Siegel DJ. MOF-5
2012;112(2):869–932. composites exhibiting improved thermal conductivity. Int J Hydrogen Energy
[25] Millward AR, Yaghi OM. Metal−organic frameworks with exceptionally high ca- 2012;37(7):6109–17.
pacity for storage of carbon dioxide at room temperature. J Am Chem Soc [43] Mu B, Walton KS. Thermal analysis and heat capacity study of metal-organic fra-
2005;127(51). 17998-17999. meworks. J Phys Chem C 2011;115(46):22748–54.
[26] CCDC. Support solution. www.ccdc.cam.ac.uk; 2014. [44] Saha D, Bao Z, Jia F, Deng S. Adsorption of CO2, CH4, N2O, and N2 on MOF-5,
[27] Millward AR, Yaghi OM. Metal-organic frameworks with exceptionally high capa- MOF-177, and Zeolite 5A. Environ Sci Technol 2010;44(5):1820–6.
city for storage of carbon dioxide at room temperature. J Am Chem Soc [45] Mason JA, Sumida K, Herm ZR, Krishna R, Long JR. Evaluating metal–organic
2005;127(51). 17998 9. frameworks for post-combustion carbon dioxide capture via temperature swing
[28] Britt D, Tranchemontagne D, Yaghi OM. Metal-organic frameworks with high ca- adsorption. Energy Environ Sci 2011;4(8):3030.
pacity and selectivity for harmful gases. Proc Natl Acad Sci 2008;105(33):11623–7. [46] Susarla N, Haghpanah R, Karimi IA, Farooq S, Rajendran A, Tan LSC, Lim JST.
[29] Tranchemontagne DJ, Hunt JR, Yaghi OM. Room temperature synthesis of metal- Energy and cost estimates for capturing CO2 from a dry flue gas using pressure/
organic frameworks: MOF-5, MOF-74, MOF-177, MOF-199, and IRMOF-0. vacuum swing adsorption. Chem Eng Res Des 2015;102:354–67.
Tetrahedron 2008;64(36):8553–7. [47] Xiao J, Peng R, Cossement D, Bénard P, Chahine R. CFD model for charge and
[30] Xiang ZHZ, Cao D, Yang W, Lu J, Han B, Wang W. Metal–organic frameworks with discharge cycle of adsorptive hydrogen storage on activated carbon. Int J Hydrogen
incorporated carbon nanotubes: improving carbon dioxide and methane storage Energy 2013;38(3):1450–9.

326