Application of Adsorption Technology for Gas Separation

Process Examples: Vacuum Pressure Swing Adsorption (VPSA), Vacuum Swing Adsorption
(VSA), Pressure Swing Adsorption (PSA).

Targeted Gases: CO2/N2 separation in the flue gases produced by the combustion of fossil fuels
(coal, natural gas, and oil) for electricity production in power plants, steel industry,
manufacturing of cement, chemical, and petrochemical industries. In addition, CO2/CH4
separation in natural gas processing or biogas. In general, 39% of global CO2 emissions are
generated from the combustion of coal, while 31% of CO2 emissions are produced from oil
combustion, followed by 18% of CO2 emitted from natural gas combustion. Besides, 5-7% of
CO2 emissions are associated with cement kilns and more than 7% from the steel industry.

Targeted Adsorbents (solid materials): Metal-Organic Frameworks (MOFs), Zeolite, or any

other adsorbents.

T - Absorption separation process with an aqueous amine-based solvent is the

most mature carbon-capturing technique, but it suffers from several
drawbacks:
• Equipment corrosion.
• Sensitive to contaminants (SOx, NOx) in the flue gases.
• Solvent loss by aerosols and degradation.
• Energy Intensive (3800-4000 kJ/kgCO2 for regeneration of
(Monoethanolamine-MEA)).
• By-products emissions.
• Solvent make-up.

Ad - Adsorption is a promising alternative technique presenting many advantages:

• Lower cost.
• Lower equipment size.
• Lower environmental impacts.
• Potentially lower energy required.
• Suitable for various operating conditions.
→ Research is required to find the best materials (adsorbents) and the
most suitable process for CO2 capture.
 Methodology:

1 Thermodynamics Study:

Measurements of pure gas adsorption isotherms at different temperatures (Ex., 20, 30,
40°C) to study the effects of temperature on the adsorption using for example the gravimetric
magnetic suspension balance as a measurement apparatus. For instance, measurements of
CO2/N2 adsorption isotherms at (20, 30, 40°C). These isotherms represent the quantity of
CO2 or N2 that could be adsorbed at a given gas pressure for a particular temperature (CO2 or
N2 adsorbed amount in the unit of mmol per gram of adsorbent). The relationship between
the quantity adsorbed of the gas and equilibrium pressure at a constant temperature is known
as an isotherm.

- I can analyze the temperature effects and propose a measurement pressure with
measurement intervals. As well as I can propose the adsorbent regeneration temperature
and the time required for the regeneration before commencing the real measurements.

- I can Simulate, fit, and predict the pure gas adsorption isotherms of different gases using
Matlab software based on the experimental data and some adsorption isotherm models
such as Langmuir, Toth, dual-site Langmuir, or any model. The selection of the
adsorption isotherm model relies on the surface structure of the adsorbent (
heterogeneous, or homogeneous).

* bi Pi
Langmuir model : q i = q s
1+bi Pi

bi =b o e - ΔHi /RT

* bi Pi
Toth model: q i = qs
(1+(b P ) )
i i
t 1/t

* bi Pi dP
Dual-site Langmuir model: q i =q sb + q sd i i
1+bi Pi 1+d i Pi

Where, qi* is the amount of the adsorbate (adsorbed gas) at equilibrium (mg/g or
mmol/g), qs is the saturation capacity or maximum adsorbed amount (mmol/g), bi is the
adsorption equilibrium constant (bar-1), Pi is the equilibrium pressure of component i (bar), ∆Hi
is the heat of adsorption (kJ/mol) for component i, bo is the pre-exponential factor (bar-1), t is
Toth constant or heterogeneity indicator, qsd and di are the saturation capacity (mmol/g) and the
adsorption equilibrium constant for the second site, respectively.

- I can make a Matlab code for adsorption isotherms predictions and fittings. Also, the
code estimates the values of the isotherm parameters (IPs) of any adsorption isotherm
model (Ex., the Toth model has four isotherm parameters: qs, bi, t, ∆H/R).
- In addition, the code estimates R2, normalized root mean square error deviation
(NRMSD), plots the experimented and predicted isotherms at different temperatures, and
presents the values of the required data.
- The example below is CO2 adsorption isotherms on one of the studied MOFs fitted with
the Toth model (R2=99% for all data; NRMSD = 0.0038, 0.0035,0.0046).

Figure 1. CO2 isotherms on MOF (Exp.: experimented; Sim.: simulated in Matlab)

1.1 Estimation of the CO2 Working Capacity

The working capacity is an important parameter that must be evaluated for the adsorption
process performance since a higher working capacity implies that a smaller amount of adsorbent
will be required. It can be defined as the difference between the amount of CO2 adsorbed at high
adsorption pressure and the amount of CO2 adsorbed at the lowest desorption pressure as
illustrated in the expression below:

Working Capacity= ∆q (CO2) = q (CO2)Adsorption – q (CO2)Desorption

- I can make a Matlab code to estimate the CO2 working capacity of any used adsorbent.
Also, I can calculate it using Excel functions.

1.2 Heat of the adsorption

It is an essential factor that presents the amount of energy released during the adsorption
process. Moreover, it is possible to determine the regeneration energy and the interaction
between the adsorbent and gaseous molecules by estimating the heat of adsorption. It can be
represented by the isosteric heat of adsorption at a given adsorption amount using the Clausius-
Clapeyron equation

 ln p 
ΔH isos = -R * 
 1/T  qi

Where, ∆Hisos is the isosteric heat of adsorption (kJ/mol), p is the gaseous component
pressure (bar), T is the general temperature (K), R is the universal gas constant (8.314 J/mol.K),
and qi is the gaseous adsorbed amount (mol/kg). The heat of adsorption can be estimated from
the slope of the linear plot of ln P versus 1/T at a given adsorbed amount.

- I can make a Matlab code to estimate the heat of adsorption of the gaseous component
adsorbed on any used adsorbent. Also, I can calculate it using Excel functions.
2 Kinetic Study

It can be done by breakthrough curves of CO2 and N2 measurements on the adsorbents

using a single bed (column) adsorption unit. The breakthrough curves show the gas mole
fractions against the contact time at a certain temperature. The selected kinetics model
represented by the linear driving force overall mass transfer coefficient (KLDF) can be estimated
by the minimization of the differences between the experimented and simulated breakthrough
curves as depicted in figure 2.

qi
= k LDF (q*i − qi )
t
Where, KLDF is the overall mass transfer coefficient (s-1), q *i is the adsorbed phase

amount at equilibrium with the amount in the gas phase (mmol/g) and qi is the amount of solid
loading (mmol/g).

Figure2. Experimented (Exp.) and simulated (Sim.) CO2 and N2 breakthrough curves on MOF

- I can simulate the breakthrough curves and minimize the squared errors between the
experimented and predicted breakthrough curves to estimate the mass transfer
coefficients of CO2 and N2 (KLDF) using Aspen Adsorption software.
3 CO2 Selectivity

The evaluation of an adsorbent for CO2 capture must consider its selectivity for CO2
relative to other components in the gas mixture. It is a valuable initial tool to identify the
adsorbent that is suitable for a given process and it is related to the difference in the affinity of
the adsorbent to CO2 over other components in the gas mixture. It can be estimated from the co-
adsorption measurements or breakthrough curves, or using the Ideal Adsorbed Solution Theory
(IAST) by predicting the co-adsorption of gaseous components based on pure gas adsorption
isotherms.

x1 / x 2
S1/2 =
y1 / y 2

Where, S is the selectivity factor, y and x are the molar fractions of gaseous components
in the gas phase and adsorbed phase, respectively. IAST is a thermodynamic approach, basically
an adsorption analog to Raoult’s law for vapor-liquid equilibrium.

- I can write a robust Matlab code using the Ideal Adsorbed Solution Theory (IAST) for:
➢ Predicting the co-adsorption of gaseous components.
➢ Calculating the selectivity using a wide range of pressures.
➢ Estimating the equilibrium gas phase pressure (bar) of the pure component i at the
temperature and spreading pressure of the mixture in the standard state.
➢ Estimating xi the mole fractions of component i in the adsorbed phase since yi
and the total pressure (P) are known.
➢ Estimating the total adsorbed amount qT.
➢ Estimating the standard state loading q i (Pio ) , and the adsorbed amount for each

component in a mixture.
 Cyclic Adsorption Process Simulation and Design

The carbon dioxide capture unit must have a performance of 90% minimum CO2
recovery and a CO2 purity level of 95% to align with the requirements of the efficient capture
process based on the requirements of the US Department of Energy. The energy consumption is
expected to be less than 2600 kJ/kgCO2.

- The evaluation of the adsorptive cyclic process will be based on four key indicators :
➢ CO2 purity in the product stream.
➢ CO2 recovery.
➢ Process productivity in kgCO2/kgads.h (defined as the amount of CO2 produced
per mass of adsorbent and per unit time of cycle).
➢ The total energy consumption in kJ/kgCO2.
- I can design and generate different flowsheets and process configurations using Aspen
Adsorption software.
- Simulating different process configurations ( ex., Two-bed, three-bed, two-unit) to obtain
high process performance.
- Manipulating the operation conditions to improve the process performance.
- Improving the basic Skarstrom cycle ( pressurization, adsorption, blowdown, purge) by
adding some steps such as pressure equalization, rinse, and light blowdown depending on
the process configuration adapted.
- Estimation of some adsorption parameters that are necessary for process simulation.
- Selection of the proper adsorption isotherm model for the simulation.
- Selection of an adequate equation of state for gas properties estimation such as molecular
weight, viscosity...etc.
- Writing a code in Aspen Adsorption software to make the adsorption isotherm model as
temperature dependence in the case of the model in the software showed temperature
independence.
- Choosing suitable energy, heat, mass balance assumptions, flows, and other inputs such
as proper equation for pressure drop in the bed.
- Analyze and elaborate the effects of adsorption parameters such as adsorption time, purge
time, purge flowrate, feed flowrate, feed pressure, and desorption pressures…etc.
- I have designed a Two-stage (unit) VPSA process with a two-bed in each stage
employing a five-step Skarstrom cycle (pressurization, adsorption, pressure equalization,
blowdown, purge). I used zeolite 13X and MOF (I cannot mention its name) as studied
adsorbents to separate CO2 from flue gases of power plants and energy-intensive
industries with a feed containing (15%CO2-85%N2). Regardless of the adsorbent used, I
have successfully obtained > 95% CO2 purity, > 90% CO2 recovery, and total energy
consumption of less than 1000 kJ/kgCO2. Hence, the objectives of an efficient capture
unit have been met.