Powder Technology 373 (2020) 446–458

Contents lists available at ScienceDirect

Powder Technology

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

Fixed bed adsorption as affected by thermodynamics and kinetics:

Yellow tuff for CO2 capture
Paola Ammendola a, Federica Raganati a,⁎, Riccardo Chirone a, Francesco Miccio b
a
Istituto di Ricerche sulla Combustione (IRC) - CNR, Piazzale V. Tecchio 80, 80125 Naples, Italy
b
Istituto di Scienza e Tecnologia dei Materiali Ceramici (ISTEC) - CNR, via Granarolo, 64, 48018 Faenza, (RA), Italy

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

Article history: In this work, a yellow tuff has been proposed for the first time as sorbent for CO2 capture. Dynamic breakthrough
Received 31 January 2019 experiments have been performed in a fixed bed reactor at different temperatures (25–150 °C) and CO2 partial
Received in revised form 9 December 2019 pressures (0.01–0.20 atm), focusing on the process thermodynamics and kinetics. The thermodynamic and ki-
Accepted 28 June 2020
netic studies highlighted that the CO2 adsorption on the tuff in the low pressure region typical of combustion
Available online 04 July 2020
flue gases can be properly described by Freundlich's isotherm model and by the pseudo-first order kinetic
Keywords:
model, respectively, thus indicating a physical, multilayer and heterogeneous surface binding mechanism.
Fixed bed adsorption Based on the results obtained, it has been demonstrated that, in the framework of a real application for post-
Thermodynamics/kinetics combustion CO2 capture, the natural tuff, i.e. a low-cost natural sorbent, could be more conveniently employed
Tuff in vacuum swing or mixed mode in order to minimize the energy penalty of the process.
Zeolites © 2020 Elsevier B.V. All rights reserved.
Langmuir/Freundlich isotherms
Pseudo-first/pseudo-second order kinetic
models

1. Introduction reactor is taken off-line and the feed is switched to the second bed. Si-
multaneously, the first bed is regenerated by either increasing the tem-
Among all the CO2 separation strategies, i.e. post-combustion cap- perature (temperature swing adsorption, TSA) or decreasing the
ture, pre-combustion capture and oxy-fuel combustion [1], developed pressure (pressure swing adsorption, PSA) [90]. However, for the suc-
in the framework of Carbon Capture and Storage (CCS) [2], post- cess of this approach the sorbent should combine low cost with versatil-
combustion capture, which consists in the selective removal of CO2 ity and good performances at low CO2 pressure (up to 0.2 atm in typical
from the gaseous effluent, represents a near-term solution for stationary post-combustion conditions [3]) in terms of a high CO2 adsorption ca-
fossil fuel-fired plants, since it does not need any radical change to pacity, fast adsorption/desorption kinetics, high CO2 selectivity, mild re-
existing facilities and processes [3]. Even though being the most mature generation conditions, high stability to cyclic adsorption–desorption
thechnology for CO2 separation [4], amine-based or ammonia-based ab- operations, tolerance to the presence of moisture and other impurities
sorption methods suffer from severe drawbacks when applied to post- in the feed and adequate mechanical strength [5]. In this context, sor-
combustion applications: high energy requirement for sorbent regener- bents are classified as either physical or chemical sorbents [6]. In partic-
ation, corrosion issues, thermal/chemical degradation of the amines due ular, physisorption arises from relatively weak interactions such as van
to the presence of oxygen and amine losses due to evaporation, causing der Waals force, while chemisorption involves stronger chemical inter-
environmental impact [4]. In this framework, adsorption using solid actions (chemical bonding) with consequent transfer of electrons be-
sorbents has attracted great interest since it is characterized by low re- tween the adsorbent and adsorbate [6]. Common materials acting as
generation energy consumption, high selectivity, easiness in handling, physisorbent are porous carbonaceous materials [5,7–12], zeolites
no liquid waste streams and applicability over a relatively wide range [13,14], alumina [15], silica gels [16] and geopolymers [17,18]. On the
of operating temperatures [5]. Adsorption is typically carried out in contrary, chemical sorbents, such as amine-functionalized sorbents,
two fixed bed reactors. The feed stream containing adsorbate is passed have basic sites (carbonates and amino groups) capable of strongly in-
through the first bed and, then, when breakthrough is about to occur, teract with the acidic CO2 molecules [19,20]. Other sorbents, such as
i.e. once the percentage of the adsorbate exiting the column rises metal organic frameworks (MOFs), are capable of providing both phys-
above a certain threshold value (i.e. at the breakthrough value), the ical and chemical interactions with CO2 [21,22]. In this framework,
physical adsorption of CO2 is an attractive possibility because, the pro-
⁎ Corresponding author. cess is totally reversible, i.e. the sorbent can be easily and effectively re-
E-mail address: federica.raganati@irc.cnr.it (F. Raganati). generated with relatively small energy requirements due to the lower

https://doi.org/10.1016/j.powtec.2020.06.075
0032-5910/© 2020 Elsevier B.V. All rights reserved.
 P. Ammendola et al. / Powder Technology 373 (2020) 446–458 447

adsorption enthalpy in comparison to chemical sorbents [3]. Synthetic Table 1

zeolites are highly ordered microporous crystalline materials, Physical properties of the natural tuff.

synthetized with a very precise and suitable molecular cell, pore size Particle size, μm 400–600
and nature of extra-framework cations, thus specifically affecting their Particle porosity, − 0.339a
adsorption performance [23]. However, their high synthesis cost and Absolute density, g mL−1 2.241a
the environmental footprint represent a negative contribute to their Tap density, g mL−1 0.810
sustainability [17,24]. A possible alternative is provided by natural oc- Surface area, m3 g−1 141b
Specific heat, J g−1 K−1 1.3
curring zeolites, as they are available in large quantities in different re-
Minimum fluidization velocity, m s−1 0.168
gions of the world in the natural composition of different types of
a
rocks [25,26]. Their abundance and low or null price, somehow, offset MIP.
b
ref. [23].
some unavoidable drawbacks such as variable composition that may
change even in a single deposit, low purity and likely poorer separation
performance compared to the more-performant synthetic zeolites [25].
In this framework, the volcanic tuffs are the most valuable deposits of 2.2. Experimental apparatus and procedure for CO2 adsorption tests
natural zeolites [27] and they are characterized by the presence of dif-
ferent minerals, such as: clinoptilite, mordenite, chabazite, quartz, The CO2 dynamic adsorption breakthrough tests were carried out at
cristobalite, feldspar, etc. [28]. Natural zeolites were promisingly tested atmospheric pressure and at different temperatures (25, 40, 70, 100 and
for biogas purification [29,30]. Nonetheless, the literature about the use 150 °C) in a laboratory-scale fixed bed reactor (ID = 10 mm, column
of natural tuff as CO2 adsorbent at low pressure is still lacking. length = 600 mm) reported in Fig. 1.
After the selection of the adsorbent material, a combined study of A type K thermocouple, with a diameter of 1 mm, connected to a PID
the equilibrium and kinetics is necessary for the design of the specific controller is used to monitor the temperature and keep it at the desired
adsorption system (such as pressure swing and temperature swing). In- value, by means of an electric heating jacket (Tyco Thermal Controls
deed, the accurate mathematical representation of both the adsorption GmbH).
equilibrium and kinetics is crucial for the identification of the adsorp- Separate high purity N2 and CO2 cylinders (99.995% vol.) are used to
tion performances, by providing useful insight into the adsorbate- prepare the gas feed, using two mass flow controllers (Brooks 8550S) to
adsorbent interactions [31,32]. In particular, a good adsorbent should set and control the inlet flowrates. The analysis system consists of a con-
provide the right compromise between high equilibrium adsorption ca- tinuous gas analyzer equipped with an infrared detector (ABB AO2020,
pacity and fast kinetics [31,32] that are requisites for process URAS 14), monitoring the outlet CO2 concentration during the transient
implementation. adsorption process.
This work is aimed at investigating the low-pressure CO2 adsorption In a typical adsorption test, the sorbent (5 g, corresponding to about
on a natural zeolite tuff, with particular attention to its mechanism, in 5 cm of bed height) is subjected to a drying/cleaning step by flowing N2
terms of thermodynamics and kinetics. Dynamic breakthrough tests (15 L h−1) for 60 min at 150 °C and atmospheric pressure. This step is
have been carried out in a lab-scale fixed bed reactor at different adsorp- necessary since zeolites are very sensitive to the presence of water,
tion temperatures (25–150 °C) and CO2 partial pressures (1–20%vol.). which strongly inhibits their CO2 adsorption performances [25]. After-
Langmuir and Freundlich isotherm models have been used to fit the wards, during a pre-conditioning step, the temperature is set to the de-
equilibrium experimental equilibrium data, thus providing insights sired adsorption temperature (25, 40, 70, 100 and 150 °C) and N2 is
into the feasibility and strength of adsorbate–adsorbent interaction. continuously fluxed through the bed (15 L h−1). Clearly, the actual du-
Pseudo-first and pseudo-second order kinetic models have been used ration of this pre-conditioning step depends on the adsorption temper-
to fit the experimental CO2 uptake data, thus obtaining the rate of CO2 ature, i.e. it lasts until the sorbent reaches the set adsorption
adsorption for possible equipment design. The unpublished results of temperature starting from the drying temperature (150 °C). Then,
CO2 adsorption by means of natural tuff are reported and discussed in
the article, along with theoretical speculations for practical use in
post-combustion CO2 removal.

2. Experimental

2.1. Materials

A natural yellow tuff from Campania (I) region has been used as ad-
sorbent material. The sample has been crushed and sieved to obtain a
400–600 μm fraction. The main physical properties of the yellow tuff
are reported in Table 1. Prior to the CO2 adsorption study, the tuff phys-
ical and structural properties have been characterized. The internal po-
rosity and the absolute density of the material have been determined by
employing mercury intrusion porosimetry (MIP, ThermoFinnigan 240).
XRD analysis has provided the phase composition of the tuff by using a
Bruker D8 Advance powder diffractometer with Cu Kα radiation. In par-
ticular, the sample has been analyzed after drying at 100 °C and heating
up to 250 °C in order to check possible changes of the phases in the ma-
terial due to the thermal treatment. The results are discussed in section
5.1. The specific heat, determined with calorimetric procedure at room
Fig. 1. Experimental apparatus: 1) N2 cylinder; 2) CO2 cylinder; 3) N2 mass flow controller;
temperature, is in the typical range of lapideous materials. Also the min- 4) CO2 mas flow controller; 5) multichannel control instrument; 6) 6 mm ID fixed bed
imum fluidization velocity of the tuff has been theoretically evaluated reactor; 7) thermocouple; 8) temperature controller; 9) heating jacket; 10) CO2
[33,87,88,91]. analyzer; 11) stack.
448 P. Ammendola et al. / Powder Technology 373 (2020) 446–458

once the desired adsorption temperature is reached, the adsorption is The Langmuir model [47] is the simplest theoretical model to de-
started, in which 15 L h−1 of the CO2/N2 gas mixture (1, 3, 5, 10, 15 scribe monolayer adsorption onto homogeneous surfaces (i.e. the ad-
and 20% vol. of CO2) is fed through column. By continuously monitoring sorption sites are equivalent from an energetic point of view) and it is
the CO2 concentration in the outlet stream until saturation of the bed based on the following mathematical expression [47]:
(i.e. the outlet CO2 concentration approached the inlet value) the break-
through curves are obtained. K L P CO2
qe ¼ qL ð2Þ
Adsorption tests have been also carried out to measure the selectiv- 1 þ K L P CO2
ity of the tuff for CO2 over CH4. The CH4 adsorption isotherm has been
where qe (mmol g−1) is the amount of CO2 adsorbed per mass of sor-
obtained at ambient temperature by using the same operating condi-
bent at equilibrium, qL (mmol g−1) is the maximum monolayer adsorp-
tions as those of the CO2 adsorption tests (CH4/N2 mixture inlet flow
tion capacity of the adsorbent, PCO2 (atm) is the equilibrium pressure of
rate of 15 L h−1; 1, 3, 5, 10, 15 and 20% vol. of CH4); the outlet CH4 con-
the gas adsorbed, whilst KL (atm−1) is the Langmuir adsorption con-
centration has been monitored using the same analysis system as that
stant or affinity constant (larger values of KL indicate a stronger adsor-
used for the CO2 adsorption tests. Then, the CO2/CH4 selectivity factor
bate/adsorbent affinity with a wider surface coverage). Another
has been estimated based on the measurement of pure fluids equilib-
important parameter is the dimensionless separation factor or equilib-
rium capacities. In particular, it is the ratio between the equilibrium
rium parameter, RL, defined as:
molar uptakes of CO2 and CH4 at a given pressure taken from the corre-
sponding single component isotherms [34]. 1
For each CO2 adsorption test the breakthrough curve has been ob- RL ¼ ð3Þ
1 þ K L P CO2
tained, i.e. a plot reporting C/C0 versus time, being C and C0 the volumet-
ric CO2 concentration in the outlet and inlet stream, respectively. By The value of RL indicates the shape of the isotherm (RL > 1 unfavor-
subsequent elaboration of the data the most important adsorption pa- able; RL = 1 linear; RL = 0 irreversible; 0 < RL < 1 favorable) [47].
rameters have been calculated [5,35,36]: Adsorption of molecules on surfaces having constant energy of inter-
i) the amount of CO2 adsorbed per unit mass of adsorbent, qe, ob- action is very rare in practice, i.e. most solids are very heterogeneous in
tained from the integration of the breakthrough curves in accor- contrast to the hypothesis made by Langmuir. Therefore, the likely het-
dance to a mass balance equation applied to the adsorption erogeneity of the sorbent surface is taken into account by assuming that
column the interaction energy follows some kind of mathematical distribution.
One of the most used isotherm models to take into account this (also
Z ts  in the case of CO2 adsorption) [37–45] is the Freundlich model [48], pro-
1 
qe ¼ F CO2,in −F CO2,out dt ð1Þ viding the first empirical equation to describe non-ideal multilayer ad-
m 0
sorption (i.e. several layers of adsorbate can be attached on the
adsorbent) onto heterogeneous surfaces [46,49]. The main assumption
where m is the mass of sorbent in the bed, FCO2,in and FCO2,out are to the
is that the adsorption energy exponentially decreases as the number
molar flowrate of CO2 at the inlet and outlet of the bed, respectively, ts is
of available adsorption sites decreases (i.e. with increasing surface cov-
to the time needed to saturate the bed of sorbent;
erage) [48], in contrast to the Langmuir model. Its mathematical expres-
ii) the breakthrough time, tb, i.e. the time in which the CO2 concen-
sion is [48]:.
tration at the outlet of the column is the 5% of the inlet concentra-
tion; it is the time at which the reactor is taken off-line for qe ¼ K F P CO2 =n
1
ð4Þ
regeneration, i.e. the higher tb is the higher the effective capture
capacity is; where KF is the Freundlich isotherm constant (mmol g−1 atm-1/n) and n
iii) the time parameter Δτ = t70 − t10 (with t10 and t70 being the is the heterogeneity factor (Freundlich coefficient). The ratio 1/n is the
time for which CO2 reach 70% and 10% of the inlet concentration Freundlich intensity parameter and it is a measure of the adsorbate/ad-
at the adsorption column outlet) that is tightly linked to the slope sorbent binding energy and surface heterogeneity [48]. Moreover, the
of the linear part of the sigmoid (the smaller Δτ is, the steeper magnitude of 1/n gives insight on whether the process is favorable or
the breakthrough curve and consequently the faster the adsorp- not, the adsorption being favorable at 1/n < 1 [48].
tion will be); The quality of the isotherm fit by Lamgmuir and Frenundlich models
iv) the fraction of bed used at breakpoint ψ, i.e. the percentage of CO2 to experimental data has been assessed by evaluating the coefficient of
adsorbed until tb with respect to the total amount adsorbed upon correlation, R2, changing between 0 and 1, and the HYBRID error func-
saturation. tion, proposed for improving the fit of the sum of the squares of errors
(SSE) method at low partial pressure values [39], where the lower the
3. Adsorption thermodynamics HYBRID value the better the fitting quality is. In particular, it is defined
as:
3.1. Adsorption isotherms 2 2 3
6 qmodi −q exp i
i¼n
100 7
The adsorption equilibrium is typically described by adsorption iso- HYBRIDð%Þ ¼ ∑4 5 ð5Þ
n−p i¼1 q exp i
therms, i.e. graphical representation of the relationship between the
equilibrium CO2 uptake (i.e. the amount of CO2 adsorbed at equilibrium
by a unit weight of the sorbent) and the CO2 partial pressure at equilib- where qexp and qmod are the experimental and modeled adsorption up-
rium and at fixed temperature. The equilibrium CO2 uptake for a certain takes, respectively, n is the number of experimental points and p is the
temperature and CO2 partial pressure is computed by integrating the number of parameters of the equation.
breakthrough curves [31]. In particular, among all the available iso-
therms models, Langmuir and Freundlich models have been employed 3.2. Evaluation of thermodynamic properties
to fit the experimental adsorption isotherms of the tuff, according
with most of the works available on CO2 adsorption on physical sor- The most important thermodynamic parameters, i.e. the standard
bents (such as activated carbons and zeolites) [37–45]. A non-linear re- Gibbs free energy change (ΔG0, kJ mol−1), the enthalpy change (ΔH0,
gression analysis has been used because of the issues associated with kJ mol−1), and the entropy change (ΔS0, kJ mol−1) have been evaluated
the linearization of non-linear isotherm equations [46]. combining the thermodynamic laws with the experimental data
 P. Ammendola et al. / Powder Technology 373 (2020) 446–458 449

obtained from the Langmuir isotherms and by applying the van't Hoff   
equation [26,50]. The analysis of these parameters can give information qt ¼ qe 1− exp −kf t ð7Þ
on the nature of the adsorbent-adsorbate interactions [50]. The adsorp-
tion process is spontaneous if the total free energy of the system de- This model can properly describe a reversible type of adsorbate/ad-
creases (ΔG0 < 0). The process is either exothermic or endothermic if sorbent interactions, as the case of CO2 adsorption on physical adsor-
ΔH0 < 0 and > 0, respectively [50]. Moreover, for physical adsorption bents like activated carbons and zeolites [58].
|ΔH0| < 20 kJ mol−1, whereas for chemisorption |ΔH0| > 40 kJ mol−1 The pseudo-second order kinetic model was firstly proposed by
[50]. The organization of the adsorbate at the solid/gaseous interface Blanchard et al. [59] and it assumes that the adsorption rate is propor-
during the adsorption process becomes less random if ΔS0 < 0 or tional to the square of the number of available adsorption sites on the
more random if ΔS0 > 0. adsorbent surface. Its mathematical expression in the differential form
Another key parameter is the isosteric heat of adsorption (Qst) [51], is [59]:
defined as the heat of adsorption at constant amount of adsorbed adsor-
dqt
bate. It can provide an estimation of the strength of molecular-scale in- ¼ ks ðqe −qt Þ2 ð8Þ
dt
teractions between the adsorbate molecules and the adsorbent surface,
Qst < 80 kJ mol−1 applying for physisorption and Qst = 90–100 kJ mol−1 where ks (g mmol−1 min−1) is the second order rate constant. Integrat-
for strong chemisorption [46,52,53]. Moreover, information regarding ing and applying boundary conditions t = 0, qt = 0, and t = ∞, qt = qe,
its magnitude and dependence on the surface coverage can be used as Eq. (8) becomes:
criteria for determining the energetic heterogeneity of a solid surface
[46,52,53]. Indeed, Qst is independent of the surface coverage when no qe 2 ks t
qt ¼ ð9Þ
interaction occurs between adsorbed molecules, and the surface is ener- 1 þ qe ks t
getically homogeneous [46,52,53]. On the contrary, a variation of Qst
with the surface loading indicates the existence of different levels of sur- This model assumes that, besides weak physical interactions, ad-
face energy and heterogeneity of the adsorbent surface [46,52,53]. Qst at sorption is also associated to chemical adsorbate/adsorbent interac-
a given qe can be calculated, after integrating the Clausius–Clapeyron tions, thus being generally most suited to describe chemisorption [59].
equation, from the slopes of the plot of lnPCO2 against 1/T at a fixed spec- The fitting quality of the kinetic models to the experimental data has
ified adsorbed amount of CO2 (i.e. slopes of adsorption isosteres) been then assessed by evaluating the coefficient of regression, R2, and
[26,31,53]. the he hybrid fractional error function (HYBRID) as in Eq. (4).
Once computed the kinetic constant, the activation energy (Ea),
4. Adsorption kinetics defined as the minimum energy that must be overcome by the adsor-
bate molecules [64], has been calculated through the Arrhenius
Since the prediction of kinetic parameters of the adsorption equations [31].
process is typically complex, the most used approach consists in
fitting the experimental data to different available apparent ki- 5. Results and discussion
netic models and, then, select the one characterized by the best
fitting [54–57]. Among the kinetic models available in the litera- 5.1. Materials characterization
ture, pseudo-first order [58] and pseudo-second order [59] models
are the simplest and most extensively used ones [54–57]. It is The XRD patterns of the samples are shown in Fig. 2. The samples are
noteworthy to underline that, when used in the framework of ad- characterized by a rather high level of cristallinity, with characteristic
sorption kinetic studies, the prefix “pseudo” does not have its reflection peaks of chabazite located at 2θ = 20.45, 22.86, 24.61,
most classic connotation deriving from chemical kinetics (i.e. re- 30.40, and 30.73 [65], typically abundant in natural zeolites, together
ferring to systems in which one of the reactants is held sufficiently with other impurities (feldspar, quartz, calcite, etc.) [66]. From the com-
constant during the reaction so that it does not need to be in- parison of the XRD patterns (Fig. 2a and b) it can be inferred that the
cluded in the mathematical expression of the reaction rate) phase composition does not vary remarkably when the sample was con-
[60–63]. Indeed, it is used just to imply those rate laws for adsorp- ditioned at 250 °C, thus meaning that the material is quite stable in the
tion expressed in terms of adsorbed amounts (q) (i.e. occupied low and medium temperature range.
sites of adsorption) rather than in terms of concentration of the Detailed information regarding the morphology, chemical composi-
adsorbing species [60–63]. The main characteristic of these appar- tion and pore distribution of the tuff can be found elsewhere [26], also
ent models is that all the adsorption steps, including mass transfer including SEM and N2 adsorption characterization. Briefly, the material
resistances, such as external diffusion, pore diffusion, and surface is characterized by micronic cubic shaped structures, typical of the zeo-
adhesion, are grouped together [54–57]. Besides that, it is also as- lites, as well as glassy regions, and it is basically microporous
sumed that the difference between the equilibrium concentration (7 Å < d < 20 Å) with a unimodal pore size distribution, as typically re-
and the average solid phase concentration is the driving force for ported for zeolites.
adsorption.
The pseudo-first order kinetic model was first introduced by 5.2. Breakthrough curves
Lagergren in 1898 [58]. The main assumption of the pseudo-first order
kinetic model is that the rate of adsorption is proportional to the num- Figs. 3a and b report the breakthrough curves obtained at 25 °C for
ber of available free active sites on the adsorbent surface. Its rate expres- all the adopted CO2 inlet concentrations, and with an inlet CO2 concen-
sion in the differential form is expressed as [58]: tration of 10% vol. for all the adopted adsorption temperatures, respec-
tively. The adsorption results in terms of adsorption isotherms and
dqt adsorption parameters are shown in Fig. 4 and Table 2. It can be inferred
¼ kf ðqe −qt Þ ð6Þ
dt that temperature and pressure have different effects on the CO2 adsorp-
where qt (mmol g−1) is the amount of CO2 adsorbed per mass of sorbent tion process depending on whether the process is observed from the
at the time t, and kf (min−1) is the rate constant of first-order sorption. thermodynamic or the kinetic point of view.
Once integrated with the boundary conditions t = 0, qt = 0 and t = ∞, The CO2 adsorption capacity qe, i.e. the amount of CO2 adsorbed once
qt = qe it gives: the thermodynamic equilibrium is reached, is enhanced as PCO2 is in-
creased, which is in accordance with the fact that PCO2 is the adsorption
450 P. Ammendola et al. / Powder Technology 373 (2020) 446–458

Fig. 3. Breakthrough curves for CO2 adsorption on the natural tuff: a) at a fixed adsorption
Fig. 2. XRD patterns of the yellow tuff after drying at 100 °C (a) and after treatment at
temperature of 25 °C and varying the inlet CO2 concentration; b) at a fixed inlet CO2
250 °C (b).
concentration of 10%vol. and varying the adsorption temperature.

driving force from the thermodynamic point of view. On the contrary, qe

decreases as the temperature is increased (i.e. the adsorption isotherms pressures result in slower breakthrough curves due to the CO2 concen-
become less and less favorable), which is in agreement with the adsorp- tration front taking more time to reach the outlet of the bed [5,32,56].
tion exothermicity. Indeed, when the temperature is increased more Also this result is in agreement with several works available in the liter-
energy is conveyed to the CO2 molecules in the gaseous phase, which ature on CO2 adsorption on physical sorbents [5,32,56].
means, in turns, that CO2 diffusion takes place faster. On the other On the contrary, tb is negatively affected by both temperature and
hand, however, the more excited CO2 molecules are less prone to be PCO2, as it decreases as either of them is increased. As regards the ef-
trapped on the sorbent surface. As a matter of fact, at higher tempera- fect of temperature, this is due to the fact that when temperature is
tures, the adsorbed CO2 can more easily move back to the gaseous increased, the adsorption capacity is reduced and the adsorption rate
phase since they have enough energy to escape from the attraction is increased, which both lead to a faster saturation of the bed. A
forces exerted by the adsorbent surface and the CO2 molecules already more complex discussion must be devoted to the effect of PCO2,
in the gaseous phase tend to stay there [40]. which is the result of the combination of two opposed phenomena.
Looking at the process from the kinetic point of view, both temper- On the one hand, as discussed above, from a thermodynamic point
ature and PCO2 positively affect the CO2 adsorption rate (the process be- of view, the increased PCO2 results in an enhanced adsorption uptake
comes faster, i.e. the time needed to achieve the equilibrium becomes of the sorbent; as a consequence, tb is also increased. On the other
smaller), as clearly confirmed by the decrease of Δτ and the increase hand, higher values of PCO2 lead to a faster adsorption process,
of ψ as either temperature or PCO2 pressure is increased. This is also which in turns results in a faster saturation of the sorbent and, conse-
clearly shown by the breakthrough curves becoming steeper and quently, in a decrease of tb. Therefore all these things considered, the
steeper as either temperature or PCO2 are increased (Fig. 3). As regards latter aspect prevails on the first.
the adsorption temperature, this can be ascribed to the fact that faster
mass-transfer phenomena take place at higher temperature due to the 5.3. Adsorption thermodynamics
molecule mobility being enhanced [67]. Likewise, when PCO2 is in-
creased, the adsorption becomes faster, which is basically ascribable to 5.3.1. Adsorption isotherms
the enhancement of the mass transfer rate as a result of the higher ad- The fitting of the experimental data by Langmuir and Freundlich
sorption driving force [67]. As a matter of fact, lower CO2 partial models are reported in Figs. 4a and b, respectively. The marker points
 P. Ammendola et al. / Powder Technology 373 (2020) 446–458 451

represent the experimental data while the solid lines represent the re-
sults of isotherm models. Table 3 reports the values of the model param-
eters, R2 and HYBRID.
Fig. 4 clearly shows that both the models used in this work fitted
the experimental results quite well, in line with the results obtained
by other Authors with physical sorbents [37–45]. This was also con-
firmed by the fact that R2 is always larger than 0.99 and HYBRID is al-
ways lower than 0.5%. As a matter of fact, it is quite common that both
Langmuir and Freundlich isotherms are able to fairly fit the same set
of CO2 adsorption data at certain CO2 partial pressure ranges, at least
from a macroscopic point of view [68–70]. In particular, this is true es-
pecially when PCO2 is small (i.e. < 0.1 atm, as typical of post- combus-
tion capture applications) and qe is also small enough to have linear
shape of adsorption isotherm. As a confirmation, it clearly appears
from Fig. 4a that the Langmuir fitting is less accurate at lower temper-
atures, i.e. when qe is larger and the adsorption isotherms moves
away from the linearity (i.e. Freundlich is by far the best model to de-
scribe CO2 adsorption on the natural tuff at lower temperatures). This
can be ascribed to the fact that when a larger amount of CO2 is
adsorbed, it is most likely that multiple layers of CO2 molecules are
adsorbed on the sorbent surface; as a consequence, Freundlich
model is able to give a better fitting [48].
As regards the effect of temperature, Table 3 clearly shows that,
when temperature is increased, the predicted values of maximum
amount of CO2 adsorbed (qL) and adsorption isotherm constants (KL
and KF,) decrease. This is in line with the exothermicity of the CO2 ad-
sorption phenomenon on the tuff, i.e. it is thermodynamically favored
at low temperatures (this is also confirmed by an increase of tempera-
ture, about 4–5 °C, registered during the adsorption tests). Indeed,
when the temperature is reduced, more CO2 molecules are adsorbed
on the tuff surface; as a consequence, the competition among CO2 mol-
ecules for the limited number of still available adsorption sites is stron-
ger (also resulting from the stronger repulsion forces among the
molecules themselves), thus leading to a general decrease of adsorption
intensity [40].
Also regarding the results obtained from the Freundlich model,
n > 1 at all the investigated temperatures, which is an indication of a
high level of heterogeneity of the system as long as of good adsorption
Fig. 4. Adsorption isotherms of CO2 on the natural tuff at different adsorption
intensity [48,71,72]. Also, the dependence of n on the adsorption tem-
temperatures fitted by: a) Langmuir model; b) Freundlich model.
perature is a further confirmation of the fact that CO2 adsorption on
the tuff is favored at low temperatures, in accordance with its exother-
mic in nature.

Table 2
Results of CO2 adsorption tests on the natural tuff.

%CO2%vol. 25 °C 40 °C 70 °C 100 °C 150 °C

qe tb Δτ ψ qe tb Δτ ψ qe tb Δτ ψ qe tb Δτ ψ qe tb Δτ ψ
mmol g−1 s s % mmol g−1 s s % mmol g−1 s s % mmol g−1 s s % mmol g−1 s s %

1 0.216 332 225 51.972 0.164 280 160 59.170 0.089 183 92 68.373 0.047 66 69 71.255 0.028 40 42 72.570
3 0.347 222 110 67.149 0.265 201 83 69.371 0.166 131 36 71.499 0.077 47 27 73.461 0.046 27 25 75.471
5 0.426 190 76 70.472 0.348 165 52 73.352 0.196 84 32 75.786 0.111 44 24 76.281 0.067 26 23 77.446
10 0.561 144 43 72.999 0.491 128 31 75.617 0.331 76 23 76.912 0.200 41 19 77.302 0.124 24 17 78.696
15 0.643 111 29 74.513 0.593 106 24 76.627 0.404 68 18 77.400 0.256 39 17 78.222 0.167 23 15 79.393
20 0.710 104 23 76.889 0.659 94.998 21 78.187 0.453 64 17 78.995 0.306 37 16 79.606 0.198 21 14 80.479

Table 3
Langmuir and Freundlich adsorption equation parameters and fitting comparison.

T, °C Langmuir Freundlich

qL KL R2 HYBRID kF n 1/n R2 HYBRID

mmol g−1 atm−1 - % mmol g−1 atm-1/n - - - %

25 0.826 24.988 0.994 0.526 1.341 2.616 0.382 0.999 0.035

40 0.859 14.968 0.994 0.479 1.430 2.129 0.469 0.999 0.020
70 0.684 9.526 0.994 0.352 1.128 1.808 0.553 0.999 0.076
100 0.636 4.556 0.997 0.212 0.939 1.448 0.690 0.999 0.072
150 0.493 3.373 0.998 0.129 0.665 1.347 0.742 0.999 0.055
452 P. Ammendola et al. / Powder Technology 373 (2020) 446–458

Fig. 5. Separation factor as function of CO2 partial pressure. Fig. 7. Van't Hoff plot.

As regards the Langmuir model, Fig. 5 shows the calculated values of As a matter of fact, they can allow for a kinetically enhanced CO2-
the separation factor, RL, for different temperatures as a function of PCO2. over-CH4 selectivity [73–76]. This finding also confirms the results re-
RL is always < 1, which confirms that CO2 uptake is favorable using the ported for another natural tuff by Paolini et al. [29], reinforcing the
natural tuff sorbent. Besides, the decrease of RL with increasing values of idea to use the natural tuff as potential sorbent to be easily applied in
PCO2 shows that adsorption is less reversible at higher PCO2. From a prac- landfill gas purification, CO2 removal from natural gas and biogas
tical point of view, the CO2 uptake at low temperatures and PCO2 up to upgrading.
0.10 atm, i.e. reasonable conditions for post-combustion capture,
would have RL not lower than 0.75.
Fig. 6 shows the CH4 adsorption isotherm obtained at 25 °C; the CO2 5.3.2. Evaluation of thermodynamic properties
adsorption isotherm has also been reported for comparison. In line with Fig. 7 reports van't Hoff plot for the natural tuff in the temperature
data reported in literature for zeolites [34], CO2 is adsorbed in a much range of 25–150 °C. The values of ΔH0, ΔS0 and ΔG0, calculated from
larger amount than CH4, approaching a maximum selectivity factor of the slope (ΔH0/R) and intercept (ΔS0/R) of the plot and from Eq. (8)
8.7 at Pi = 0.01 atm that slightly decreases down to 7.7 at Pi = are listed in Table 4.
0.20 atm (see inset in Fig. 6). The selectivity factor results quite good, ΔG0 is always negative, thus indicating that the CO2 adsorption pro-
only lower to that of purposely synthetized sorbents, e.g. 13X zeolite, cess on the tuff occurs favorably and spontaneously at all the adopted
geopolymers, amine compounds [17]. This results can be explained con- adsorption temperatures. Besides that, the van't Hoff plot exhibits an in-
sidering that the tested tuff has been found to be characterized by a creasing trend; i.e. when temperature is increased, the magnitude of
rather high level of cristallinity, with characteristic reflection peaks of ΔG0 is decreased (ΔG0 becomes less negative), which is a confirmation
chabazite. Chabazite (structural formula (Ca2,Na4,K4)[(AlO2)4(SiO2)8] of the adsorption feasibility decreasing at higher temperatures. This re-
13H2O) has a three-dimensional framework delimited by eight- sult can be ascribed to the fact that, at higher temperatures, the physical
membered rings that create pore openings of 0.38 nm, i.e. very close bonds between CO2 molecules and adsorbent become weaker and the
to the kinetic dimension of CH4 (0.38 nm) [73–76]. Then, considering reverse of adsorption (i.e. desorption) is favored; namely there is a
that the kinetic dimension of CO2 (0.33 nm) is remarkably smaller shift of the equilibrium in the opposite direction, i.e. from adsorption
than CH4, chabazite and chabazite-like zeolites (and also other 8 rings to desorption [64].
zeolites) can be used as selective adsorbents for CO2 from natural gas. Besides that, lower temperatures favored also a better interaction of
CO2 with the tuff particles due to the lower kinetic energy of CO2 mole-
cules, thus favoring the multilayer adsorption of CO2 molecules [77,78].
This is also in line with the above reported adsorption isotherms.
The value obtained for ΔH0 is about −17 kJ mol−1. First of all, its
negative sign further confirms the exothermicity of the adsorption pro-
cess, i.e. energy in form of heat is generated into its surroundings during
the adsorptive process since new physical/chemical bonds are formed.
Besides that, its magnitude gives also further insight into the adsorption
mechanism. In particular, ΔH0 values are lower than 20 kJ mol−1, thus
confirming that CO2 adsorption on the tuff is physical in nature [50].
Finally, ΔS0 is also negative, meaning that the disorder and random-
ness of the system decrease during the adsorption process. This can be
ascribed to the fact that, when adsorbed on the tuff surface, the CO2

Table 4
Thermodynamic properties for CO2 adsorption on the natural tuff.

ΔH0 ΔS0 ΔG0

kJ mol−1 kJ mol−1 kJ mol−1

18 °C 40 °C 70 °C 100 °C 150 °C
Fig. 6. Adsorption isotherms of CH4 and CO2 on the natural tuff obtained at 25 °C. Pi =
−17.04 −0.031 −7.709 −7.239 −6.300 −5.360 −3.794
partial pressure of CH4 and CO2 in the CH4/N2 and CO2/N2 mixtures, respectively.
 P. Ammendola et al. / Powder Technology 373 (2020) 446–458 453

adsorbent and also on the energetic heterogeneity of the solid surface

[46]. The obtained values of Qst were plotted as a function of the equilib-
rium surface loading in Fig. 9. First of all, the yellow tuff exhibits values
of Qst ranging between 15 and 45 kJ mol−1 that are typical of a purely
physical interaction (< 80 kJ mol−1) [40].
Then, it is clear that Qst, i.e. the slope of the adsorption isosteres, de-
creases with increasing CO2 surface loading, which means that the tuff is
characterized by a high level of heterogeneity and that there is a varia-
tion in adsorbate-adsorbent and adsorbate-adsorbate interactions [79].
In general, Qst is associated with the interaction between CO2 molecules
and the adsorbent surface through different kinds of adsorptive molec-
ular forces, such as Van der Waals forces, hydrophobicity, ligand ex-
change and chemical bond [80]. At the initial stages of adsorption,
large numbers of free adsorption sites are available on the tuff surface.
As a consequence, the CO2 molecules may easily come into direct con-
tact with adsorbent surface, more intense forces being present between
CO2 molecules and the adsorbent; as a consequence, the activation en-
ergy is low and the heat of adsorption is high. As the adsorption process
goes further, i.e. the surface coverage is increased, less and less adsorp-
tion sites are still available and, as a consequence, the adsorption of fur-
ther CO2 molecules becomes more and more difficult. Besides that, with

Fig. 8. Adsorption isosteres of CO2 on the natural tuff. Adsorption temperature range:
25–150 °C; CO2 loading: 0.05–1.00 mmol g−1.

molecules lose the freedom of movement they have in the gaseous

phase; in other words, CO2 molecules pass from a random state (gas)
to an ordered state (on the tuff surface) [40].
Fig. 8 shows the adsorption isosteres of the yellow tuff, i.e. plots of ln
(PCO2) as a function of reciprocal absolute temperature at constant load-
ing, 0.05–1 mmol g−1, in the investigated temperature range 25–150 °C.
Then, applying the Clausius–Clapeyron equation, Qst was evaluated
from the slopes (Qst/R) of the above-mentioned isosteres. All the
isosteres exhibit negative slopes, which is consistent with the CO2 ad-
sorption being exothermic.
Then, the magnitude and change of Qst during CO2 adsorption (i.e.
with the surface loading increasing) can provide information on the
molecular-scale interactions between the CO2 molecules and the

Fig. 9. Variation of the isosteric heat of adsorption with the CO2 loading. Fig. 10. a) CO2 transient adsorption uptakes; b) CO2 fractional adsorption uptakes.
454 P. Ammendola et al. / Powder Technology 373 (2020) 446–458

Fig. 11. Experimental (dotted line) CO2 uptake on the natural tuff and corresponding fit to the pseudo-first (solid line) and pseudo-second (dashed line) order kinetic models at different
adsorption temperatures.

the surface coverage increasing, also adsorbate–adsorbate interac- 5.4. Adsorption kinetics
tions start to occur. Therefore, it is most likely that some kind of lat-
eral interactions in the adsorbed layers take place in the form of Fig. 10a reports the CO2 transient adsorption uptakes on the tuff at
attractive/repulsive London dispersion intermolecular forces between different adsorption temperatures (at 10%vol. of CO2 inlet concentra-
adsorbed CO2 [79,81,82]. As a result, as the adsorption process pro- tion), as evaluated from the experimental breakthrough curves. The
ceeds, the activation energy is increased and the heat of adsorption same curves have been also plotted as fractional uptake (qt/qe)
is decreased [83,84]. (Fig. 10b) in order to better highlight the effect of the temperature.
 P. Ammendola et al. / Powder Technology 373 (2020) 446–458 455

Table 5
Values of the pseudo-first order and pseudo-second kinetic model parameters and fitting comparison.

T, °C Pseudo-first order Pseudo-second order

qe kf R2 HYBRID qe ks R2 HYBRID
mmol g−1 min−1 - % mmol g−1 g mmol−1 min−1 - %

25 0.605 0.403 0.995 0.0672 0.734 0.516 0.987 0.111

40 0.513 0.476 0.997 0.0636 0.607 0.766 0.988 0.109
70 0.331 0.745 0.998 0.0432 0.372 2.173 0.989 0.084
100 0.197 1.295 0.999 0.0294 0.212 7.372 0.995 0.067
130 0.122 2.108 0.999 0.0244 0.129 21.052 0.996 0.047

The analysis of Fig. 10 clearly shows that adsorption kinetics and form [31]. Fig. 12 shows the plots of the linearized Arrhenius equation.
thermodynamics are differently affected by the temperature. Indeed, Ea > 0, meaning that when the temperature is increased, the reaction
when temperature is increased, the CO2 adsorption becomes faster rate is increased. More specifically, the value obtained (14.4 kJ mol−1)
due to enhanced kinetics and, on the contrary, the amount of CO2 is in line with data available in other works focused on zeolites [85].
adsorbed at equilibrium is decreased due to thermodynamics limitation. This quite small value confirms again the physical adsorption with the
In other words, the equilibrium CO2 adsorption capacity decreases with relatively weak attraction forces.
the temperature increasing (i.e. the asymptotic values reached by the
CO2 uptake curves in Fig. 10a has a decreasing trend as temperatures in-
creased), which is in line with the exothermicity of the adsorption. On
the contrary, higher temperatures positively affect adsorption kinetics
6. Assessment of the process performance
(i.e. the time needed to achieve the equilibrium becomes smaller), as
it is clearly shown by the curves becoming steeper (Fig. 10b) with in-
The results obtained with the yellow tuff trigger some speculations
creasing temperatures, in accordance with the fastened mass-transfer
about possible application in a real process for CO2 capture in a post-
and diffusion phenomena at higher values of adsorption temperatures.
combustion scheme.
Then, the pseudo-first order and pseudo-second order models have
First of all the purity of the separated carbon dioxide is dictated by the
been used to fit the experimental data. Fig. 11 reports the comparison be-
ratio between the useful volume VCO2 for selective CO2 adsorption and
tween the experimental and theoretical curves, the points representing
“dead volume” Vd, where the original gas mixture may remain
the experimental values whereas the dashed lines the amounts predicted
entrapped. Apart from intrinsic dead volumes in the plant setup (e.g.
by the kinetic models. Table 5 reports the values of the kinetic constants
valves, tubing, manifolds), the inter-particle voids and, in some extent,
with the relative correlation coefficients and errors.
the particle macro-porosity represent major source of “dead volume”.
Fig. 11 clearly shows that for each investigated temperature, the
For a bed of granular sorbent, the volume of the separated CO2 is given by:
pseudo-first order kinetic model fits the experimental data better than
the pseudo-second order kinetic model does (this shows larger devia-
tions from the experimental data). More specifically, the pseudo- RT
second order model tends to underestimate the CO2 uptake in the first V CO2 ¼ qe V r ρe ð1−εÞ ð10Þ
P
few minutes (t < 5 min); in contrast, this is remarkably overestimated
in the last stage of the adsorption process, namely as the equilibrium
is approaching. As a consequence, the pseudo-second order model sig- Where, Vr, ρe and ε are the vessel volume, the envelope or particle
nificantly overestimates the equilibrium CO2 uptake (qe) (Table 5). In density and the bed voidage, respectively. Assuming only the contribu-
contrast to this, the pseudo-first kinetic model is capable of adequately tion of the inter-particle voids, the dead volume in the adsorption equip-
fitting the experimental points during the entire time evolution of the ment is Vd = εVr. Thus, the volumetric concentration CCO2 of the
adsorption. This better quality is clearly confirmed by the data reported separated CO2 can be calculated from the species conservation equation:
in Table 5, in terms of higher values of R2 and lower values of HYB (%).
These results can be ascribed to the fact that that the pseudo-first
order can fairly describe adsorption processes characterized by revers-
ible adsorbate/adsorbent surface interactions, i.e. purely physical ad-
sorption. On the contrary, the second-order kinetic model can better
describe chemisorption processes, i.e. it is a better choice when it is nec-
essary to account for the occurrence of chemical bonding between the
adsorbate and the adsorbent surface. Therefore, these results are in
line with those obtained from the thermodynamic study, showing that
the CO2 adsorption on the tuff is purely physisorption. This is also in ac-
cordance with several works available in literature, reporting that the
pseudo-first order kinetic model is more suitable for adsorption on
physical sorbents at post-combustion operating conditions, i.e. at low
CO2 partial pressure [55].
It can also be observed that, whatever the kinetic model, the
kinetic constant (kf and ks) increase with increasing temperatures,
which is in line with high temperatures favoring the adsorption process
(Table 5).
Considering the suitability of the pseudo-first order kinetic model,
the corresponding kinetic coefficients at various temperatures have
been used to evaluate the activation energy and pre-exponential factor Fig. 12. Arrhenius plot: logarithm of the kinetic constant versus inverse of absolute
of an Arrhenius type equation by a linear regression on its linearized temperature.
456 P. Ammendola et al. / Powder Technology 373 (2020) 446–458

Table 6 and that captured in the regenerated sorbent, also taking into account
Sorbent rate and energy penalties for CO2 capture. the partial utilization of this latter. Eq. 16 expresses the virtual mass
:
1 2 3 4 flow rate of the sorbent ms that linearly depends on the carbon content
T uptake, °C 25 40 70 25 in the fuel (XC), and the inverse of sorbent capacity qe and utilization ψ,
qe, mol kg−1 0.643 0.593 0.404 1.286 28 MJ/kg and 0.012 kg mol−1 being the coal heating value and the car-
a) TS - 105 °C
bon atomic weight, respectively:
Sorbent rate kg s−1 5.967 6.586 10.835 2.702
Penalty Ht MJt kg−1 6.311 5.700 5.093 3.069 : 1 1
Penalty We MJe/kg 0.000 0.000 0.000 0.000 ms ¼ XC ð16Þ
28 0:012 ψ qe
b) VS - 0.05 atm
Sorbent rate kg s−1 14.630 12.958 15.263 3.691
Penalty Ht MJt kg−1 0.041 0.041 0.041 0.041 It is worth noting that in real cases the sorbent is kept in static con-
Penalty We MJe kg−1 −0.288 −0.302 −0.331 −0.288
ditions. Thus it only virtually circulates between CO2 capture and regen-
c) TS & VS - 70 °C & 0.10 atm
eration reactors; indeed these steps are accomplished by alternatively
Sorbent rate kg s−1 10.175 12.117 43.488 3.324
switching the gaseous streams in coupled vessels.
Penalty Ht MJt kg−1 6.069 4.898 0.387 2.244
Penalty We MJe kg−1 −0.249 −0.241 −0.223 −0.249 The results of calculations are reported in Table 6, assuming a con-
centration of CO2 in flue gas equal to 15% vol that is typical of a coal
fueled power plant.
Four different cases are reported in the table at different adsorption
V d C 0CO2 þ 100 V CO2
C CO2 ¼ ð11Þ temperature (cols. 1–3), and for a sorbent with double capacity with re-
V d þ V CO2
spect to yellow tuff (col. 4). Three regeneration strategies are compared:
a) temperature swing, b) vacuum swing and c) temperature and vac-
where, C0CO2 is the inlet CO2 concentration. The following expression,
uum swing in cascade.
giving the theoretical purity of the separated carbon dioxide, can be ob-
It appears that in case of TS (a) the sorbent rate is the lowest, thanks
tained by substitution:
to the higher regeneration temperature (105 °C) that better evacuates
0
C CO2 εC CO2 =100 þ qe RPT ρe ð1−ε Þ the sorbent. In contrast, the energy penalty Ht is very high and obviously
¼ ð12Þ We is null. For the vacuum swing method (b) the enthalpy penalty Ht is
100 ε þ qe RPT ρe ð1−εÞ
only the latent heat for CO2 desorption, whilst negative work We is re-
quired because of the gas compression from 0.05 to 1.00 atm. The sor-
The purity increases at lower voidage, higher inlet concentration,
bent rates are higher than in case (a). The case (c) is a cascade of TS at
and lower temperature because of the non-linear dependence of qe on T.
70 °C followed by VS from 0.10 to 1.00 atm. In this case, the work re-
For the case of the yellow tuff, assuming the values of the properties
quest is lower than for (b), being the CO2 evacuated under less severe
reported in Tables 1 and 2 at atmospheric pressure, the purity turns out
condition. In particular, the results obtained for the case (c) at an ad-
to be equal to 0.975 and 0.966, at temperature of 25 and 70 °C, respec-
sorption temperature of 40 °C denote a good compromise between tem-
tively. It is worth noting that in a CCS process the minimum purity should
perature and pressure swing for CO2 capture with yellow tuff.
be 0.96, in consideration of the constrain for gas delivery and in presence
When the adsorption capacity of the sorbent is increased (cols. 1 vs.
of non-condensable gases in the separated stream [86]. Thus, the studied
4), obviously the comparison turns out favorable for the more
sorbent complies with the criterion of the CO2 purity for CCS.
performing sorbent, whose rate largely decreased. However, in the
As second speculation, the energy required for the sorbent regener-
case of VS only (b) the same enthalpy penalty and work are foreseen
ation has been taken into account. To this aim both temperature swing
(i.e. Ht = 0.041 MJ/kg, We = −0.288 MJ/kg), since these depend on
(TS) and/or vacuum swing (VS) procedures have been considered. In
the captured carbon dioxide only. Altogether, a sorbent with limited
the case of TS, the enthalpy penalty Ht, per unit mass of separated car-
CO2 capture capacity would be more conveniently employed in VS or
bon dioxide, has been computed (Eq. 13) as the heat required for
combined TS and VS modes, in order to minimize the whole energy
heating the sorbent up to the regeneration temperature Tr plus the la-
penalties.
tent heat of the adsorbed CO2, λ being the molar enthalpy of adsorption:

C ðT r −T Þ 7. Conclusions
Ht ¼ þ 0:044 λ ð13Þ
0:044 qe
In this work CO2 adsorption on a natural tuff has been studied focus-
In the case of VS, two different penalties are present, the first being ing on the thermodynamics and kinetics of the process. Dynamic break-
again Ht (Eq. 13), the second being the work We for evacuating the sor- through experiments have been performed in a lab-scale fixed bed
bent from CO2. In particular, We has been calculated assuming the con- reactor.
servative case of isentropic compression from the absolute vacuum As regards thermodynamics, it is oppositely affected by temperature
level Pv to atmospheric pressure according to: and pressure. Indeed, when CO2 partial pressure is increased, the equi-
librium CO2 adsorption uptake increases, whereas it decreases when
R ðT−T ad Þ temperature is increased. The equilibrium experimental data have
We ¼ ð14Þ
γ−1 0:044 been fitted by Langmuir and Freundlich models. The Freundlich model
provides the better results in terms of fitting quality, which suggests
where, Tad and γ are the adiabatic temperature and isentropic expan- that CO2 adsorption on the tuff is multilayer and happens with a hetero-
sion factor for poly-atomic gas, respectively. In turn, Tad has been com- geneous surface binding (the adsorption energy exhibits by a non-
puted as: uniform distribution). Then, the study of the thermodynamic parame-
 ters has shown that CO2 adsorption on the tuff is spontaneous, exother-
γ−1
1 mic and physical in nature. As the surface loading increases the isosteric
T ad ¼ T ð15Þ
Pv heat of adsorption decreases, which is a further confirmation of the fact
that the tuff has an energetically heterogeneous surface and that some
For a case study of 1 MWt power plant fired with coal, the sorbent kind of lateral interactions take place among the CO2 molecules
rate has been evaluated on the basis of the emitted CO2 in the flue gas adsorbed on the tuff surface.
 P. Ammendola et al. / Powder Technology 373 (2020) 446–458 457

Then, the pseudo-first and pseudo-second kinetic model have been [14] R. Girimonte, B. Formisani, F. Testa, Adsorption of CO2 on a confined fluidized bed of
pelletized 13X zeolite, Powder Technol. 311 (2017) 9–17, https://doi.org/10.1016/j.
adopted to fit the experimental data. The pseudo-first order kinetic powtec.2017.01.033.
model is the best one, being able to provide a better quality of the fitting [15] S.C. Lee, Y.M. Kwon, C.Y. Ryu, H.J. Chae, D. Ragupathy, S.Y. Jung, J.B. Lee, C.K. Ryu, J.C.
all the investigated adsorption temperatures. The values of activation Kim, Development of new alumina-modified sorbents for CO2 sorption and regen-
eration at temperatures below 200°C, Fuel. (2010) 2–7, https://doi.org/10.1016/j.
energy, evaluated from the obtained kinetic coefficients, further con-
fuel.2010.11.006.
firmed that the CO2 adsorption process is physical in nature. [16] K.B. Lee, M.G. Beaver, H.S. Caram, S. Sircar, Reversible chemisorbents for carbon di-
The yellow tuff is also characterized by quite good CO2/CH4 equilib- oxide and their potential applications, Ind. Eng. Chem. Res. 47 (2008) 8048–8062,
rium selectivity, thus being potentially applied also in landfill gas puri- https://doi.org/10.1021/ie800795y.
[17] M. Minelli, E. Papa, V. Medri, F. Miccio, P. Benito, F. Doghieri, E. Landi, Characteriza-
fication, CO2 removal from natural gas and biogas upgrading, where
tion of novel geopolymer – zeolite composites as solid adsorbents for CO2 capture,
larger concentration of carbon dioxide occurs. Chem. Eng. J. 341 (2018) 505–515, https://doi.org/10.1016/j.cej.2018.02.050.
In the framework of a possible application in a real process for post- [18] M. Minelli, V. Medri, E. Papa, F. Miccio, E. Landi, F. Doghieri, Geopolymers as solid ad-
combustion CO2 capture, it can be speculated that the natural tuff, i.e. a sorbent for CO2 capture, Chem. Eng. Sci. 148 (2016) 267–274, https://doi.org/10.
1016/j.ces.2016.04.013.
sorbent with limited CO2 capture capacity with respect to other more [19] M.G. Plaza, C. Pevida, J.J. Pis, F. Rubiera, Evaluation of the cyclic capacity of low-cost
performing materials, could be more conveniently employed in vacuum carbon adsorbents for post-combustion CO2 capture, Energy Procedia 4 (2011)
swing or mixed mode, thus minimizing the energy penalty of the 1228–1234, https://doi.org/10.1016/j.egypro.2011.01.178.
process. [20] M. Sevilla, J.B. Parra, A.B. Fuertes, Assessment of the role of micropore size and N-
doping in CO2 capture by porous carbons, ACS Appl. Mater. Interfaces 5 (2013)
6360–6368, https://doi.org/10.1021/am401423b.
Declaration of Competing Interest [21] F. Raganati, V. Gargiulo, P. Ammendola, M. Alfe, R. Chirone, CO2 capture perfor-
mance of HKUST-1 in a sound assisted fluidized bed, Chem. Eng. J. 239 (2014)
75–86, https://doi.org/10.1016/j.cej.2013.11.005.
The Authors, whose names are listed immediately below, certify
[22] V. Gargiulo, M. Alfè, F. Raganati, L. Lisi, R. Chirone, P. Ammendola, BTC-based metal-
that: organic frameworks: correlation between relevant structural features and CO2 ad-
sorption performances, Fuel. 222 (2018) 319–326, https://doi.org/10.1016/j.fuel.
• they have NO affiliations with or involvement in any organization or 2018.02.093.
entity with any financial interest, or non-financial interest in the sub- [23] R.V. Siriwardane, M.-S. Shen, E.P. Fisher, J.A. Poston, Adsorption of CO2 on molecular
ject matter or materials discussed in this manuscript sieves and activated carbon, Energy Fuel 15 (2001) 279–284, https://doi.org/10.
1021/ef000241s.
• no funding was received for this work
[24] N. Gargiulo, K. Shibata, A. Peluso, P. Aprea, T. Valente, G. Pezzotti, T. Shiono, D.
Caputo, Reinventing rice husk ash: derived NaX zeolite as a high-performing CO2
Acknowledgement adsorbent, Int. J. Environ. Sci. Technol. 15 (2018) 1543–1550, https://doi.org/10.
1007/s13762-017-1534-5.
[25] M. Ackley, R. Rege, H. Saxena, Application of natural zeolites in the purification and
Mr. A. Piancastelli is gratefully acknowledged for MIP and XRD anal- separation of gases, Microporous Mesoporous Mater. 61 (2003) 25–42, https://doi.
yses on tuff samples. org/10.1016/S1387-1811(03)00353-6.
[26] P. Ammendola, F. Raganati, R. Chirone, F. Miccio, Preliminary assessment of tuff as
References CO2 sorbent, Energy Procedia 114 (2017) 46–52, https://doi.org/10.1016/j.egypro.
2017.03.1145.
[1] P. Ammendola, R. Chirone, G. Ruoppolo, G. Russo, Production of hydrogen from [27] P. Misaelides, Application of natural zeolites in environmental remediation: a short
thermo-catalytic decomposition of methane in a fluidized bed reactor, Chem. Eng. review, Microporous Mesoporous Mater. 144 (2011) 15–18, https://doi.org/10.
J. 154 (2009) 287–294, https://doi.org/10.1016/j.cej.2009.03.048. 1016/j.micromeso.2011.03.024.
[2] D.Y.C. Leung, G. Caramanna, M.M. Maroto-Valer, An overview of current status of [28] C. Colella, Applications of Natural Zeolites, in: Handb. Porous Solids, Wiley-VCH
carbon dioxide capture and storage technologies, Renew. Sust. Energ. Rev. 39 Verlag GmbH, Weinheim, Germany, 2008 1156–1189, https://doi.org/10.1002/
(2014) 426–443, https://doi.org/10.1016/j.rser.2014.07.093. 9783527618286.ch18i.
[3] A. Samanta, A. Zhao, G.K.H. Shimizu, P. Sarkar, R. Gupta, Post-combustion CO2 cap- [29] V. Paolini, F. Petracchini, E. Guerriero, A. Bencini, S. Drigo, Biogas cleaning and
ture using solid sorbents: a review, Ind. Eng. Chem. Res. 51 (2012) 1438–1463, upgrading with natural zeolites from tuffs, Environ. Technol. 37 (2016)
https://doi.org/10.1021/ie200686q. 1418–1427, https://doi.org/10.1080/09593330.2015.1118557.
[4] M. Karl, R.F. Wright, T.F. Berglen, B. Denby, Worst case scenario study to assess the [30] A. Alonso-Vicario, J.R. Ochoa-Gómez, S. Gil-Río, O. Gómez-Jiménez-Aberasturi, C.A.
environmental impact of amine emissions from a CO2 capture plant, Int. J. Greenh. Ramírez-López, J. Torrecilla-Soria, A. Domínguez, Purification and upgrading of bio-
Gas Control. 5 (2011) 439–447, https://doi.org/10.1016/j.ijggc.2010.11.001. gas by pressure swing adsorption on synthetic and natural zeolites, Microporous
[5] M.G. Plaza, S. García, F. Rubiera, J.J. Pis, C. Pevida, Post-combustion CO2 capture with Mesoporous Mater. 134 (2010) 100–107, https://doi.org/10.1016/j.micromeso.
a commercial activated carbon: comparison of different regeneration strategies, 2010.05.014.
Chem. Eng. J. 163 (2010) 41–47, https://doi.org/10.1016/j.cej.2010.07.030. [31] P. Ammendola, F. Raganati, R. Chirone, CO2 adsorption on a fine activated carbon in
[6] M. Alfe, P. Ammendola, V. Gargiulo, F. Raganati, R. Chirone, Magnetite loaded carbon a sound assisted fluidized bed: thermodynamics and kinetics, Chem. Eng. J. 322
fine particles as low-cost CO2 adsorbent in a sound assisted fluidized bed, Proc. (2017) 302–313, https://doi.org/10.1016/j.cej.2017.04.037.
Combust. Inst. 35 (2015) 2801–2809, https://doi.org/10.1016/j.proci.2014.06.037. [32] F. Raganati, M. Alfe, V. Gargiulo, R. Chirone, P. Ammendola, Isotherms and thermo-
[7] P. Ammendola, F. Raganati, R. Chirone, Effect of operating conditions on the CO2 re- dynamics of CO2 adsorption on a novel carbon-magnetite composite sorbent,
covery from a fine activated carbon by means of TSA in a fluidized bed assisted by Chem. Eng. Res. Des. 134 (2018) 540–552, https://doi.org/10.1016/j.cherd.2018.04.
acoustic fields, Fuel Process. Technol. 134 (2015) 494–501, https://doi.org/10. 037.
1016/j.fuproc.2015.03.010. [33] D. Geldart, Gas Fluidization Technology, Wiley & Son, Chichester; New York, 1986.
[8] F. Raganati, P. Ammendola, R. Chirone, On improving the CO2 recovery efficiency of [34] N.K. Jensen, T.E. Rufford, G. Watson, D.K. Zhang, K.I. Chan, E.F. May, Screening zeo-
a conventional TSA process in a sound assisted fluidized bed by separating heating lites for gas separation applications involving methane, nitrogen, and carbon diox-
and purging, Sep. Purif. Technol. 167 (2016) 24–31, https://doi.org/10.1016/j. ide, J. Chem. Eng. Data 57 (2012) 106–113, https://doi.org/10.1021/je200817w.
seppur.2016.05.001. [35] M.V. Gil, N. Álvarez-Gutiérrez, M. Martínez, F. Rubiera, C. Pevida, A. Morán, Carbon
[9] F. Raganati, P. Ammendola, R. Chirone, Effect of acoustic field on CO2 desorption in a adsorbents for CO2 capture from bio-hydrogen and biogas streams: breakthrough
fluidized bed of fine activated carbon, Particuology. 23 (2015) 8–15, https://doi.org/ adsorption study, Chem. Eng. J. 269 (2015) 148–158, https://doi.org/10.1016/j.cej.
10.1016/j.partic.2015.02.001. 2015.01.100.
[10] F. Raganati, P. Ammendola, R. Chirone, Role of acoustic fields in promoting the gas- [36] F. Raganati, M. Alfe, V. Gargiulo, R. Chirone, P. Ammendola, Kinetic study and break-
solid contact in a fluidized bed of fine particles, KONA Powder Part. J. 32 (2015) through analysis of the hybrid physical/chemical CO2 adsorption/desorption behav-
23–40, https://doi.org/10.14356/kona.2015006. ior of a magnetite-based sorbent, Chem. Eng. J. 372 (2019) 526–535, https://doi.org/
[11] V. Gargiulo, M. Alfè, P. Ammendola, F. Raganati, R. Chirone, CO2 sorption on surface- 10.1016/j.cej.2019.04.165.
modified carbonaceous support: probing the influence of the carbon black micropo- [37] B. Guo, L. Chang, K. Xie, Adsorption of carbon dioxide on activated carbon, J. Nat. Gas
rosity and surface polarity, Appl. Surf. Sci. 360 (2016) 329–337, https://doi.org/10. Chem. 15 (2006) 223–229, https://doi.org/10.1016/S1003-9953(06)60030-3.
1016/j.apsusc.2015.11.026. [38] B. Selen, A.˘. Glayan, A.E. Aksoylu, CO2 adsorption behavior and kinetics on chemi-
[12] M. Hemmat, A. Rahbar-Kelishami, M.H. Vakili, Preparation of carbon molecular cally modified activated carbons, Turk. J. Chem. 40 (2016) 576–587, https://doi.
sieves and its impregnation with co and Ni for CO2/N2 separation, Int. J. Environ. org/10.3906/kim-1507-95.
Sci. Technol. 15 (2018) 2213–2228, https://doi.org/10.1007/s13762-017-1526-5. [39] J. Sreńscek-Nazzal, U. Narkiewicz, A.W. Morawski, R. Wróbel, A. Gęsikiewicz-
[13] F. Raganati, P. Ammendola, R. Chirone, CO2 capture performances of fine solid sor- Puchalska, B. Michalkiewicz, Modification of commercial activated carbons for CO
bents in a sound-assisted fluidized bed, Powder Technol. 268 (2014) 347–356, 2 adsorption, Acta Phys. Pol. A 129 (2016) 394–401, https://doi.org/10.12693/
https://doi.org/10.1016/j.powtec.2014.08.062. APhysPolA.129.394.
458 P. Ammendola et al. / Powder Technology 373 (2020) 446–458

[40] X. Zhou, H. Yi, X. Tang, H. Deng, H. Liu, Thermodynamics for the adsorption of SO2, [66] R. Leyva-Ramos, J.E. Monsivais-Rocha, A. Aragon-Piña, M.S. Berber-Mendoza, R.M.
NO and CO2 from flue gas on activated carbon fiber, Chem. Eng. J. 200–202 (2012) Guerrero-Coronado, P. Alonso-Davila, J. Mendoza-Barron, Removal of ammonium
399–404, https://doi.org/10.1016/j.cej.2012.06.013. from aqueous solution by ion exchange on natural and modified chabazite, J. Envi-
[41] A. Golchoobi, H. Pahlavanzadeh, Molecular simulation, experiments and modelling ron. Manag. 91 (2010) 2662–2668, https://doi.org/10.1016/j.jenvman.2010.07.035.
of single adsorption capacity of 4A molecular sieve for CO 2 –CH 4 separation, Sep. [67] D.M. Ruthven, Principles of Adsorption and Adsorption Processes, John Wiley &
Sci. Technol. 51 (2016) 2318–2325, https://doi.org/10.1080/01496395.2016. Sons, 1984.
1206571. [68] E.R. Monazam, L.J. Shadle, D.C. Miller, H.W. Pennline, D.J. Fauth, J.S. Hoffman, M.L.
[42] J.-S. Lee, J.-H. Kim, J.-T. Kim, J.-K. Suh, J.-M. Lee, C.-H. Lee, Adsorption equilibria of CO Gray, Equilibrium and kinetics analysis of carbon dioxide capture using immobilized
2 on zeolite 13X and zeolite X/activated carbon composite, J. Chem. Eng. Data 47 amine on a mesoporous silica, AICHE J. 59 (2013) 923–935, https://doi.org/10.1002/
(2002) 1237–1242, https://doi.org/10.1021/je020050e. aic.13870.
[43] X. Xu, X. Zhao, L. Sun, X. Liu, Adsorption separation of carbon dioxide, methane, and [69] J. Sreńscek-Nazzal, U. Narkiewicz, A.W. Morawski, R.J. Wróbel, B. Michalkiewicz,
nitrogen on Hβ and Na-exchanged β-zeolite, J. Nat. Gas Chem. 17 (2008) 391–396, Comparison of optimized isotherm models and error functions for carbon dioxide
https://doi.org/10.1016/S1003-9953(09)60015-3. adsorption on activated carbon, J. Chem. Eng. Data 60 (2015) 3148–3158, https://
[44] M.A. Hernandez, A. Pestryakov, R. Portillo, M.A. Salgado, F. Rojas, E. Rubio, S. Ruiz, V. doi.org/10.1021/acs.jced.5b00294.
Petranovskii, CO2 sequestration by natural zeolite for greenhouse effect control, [70] H. Yi, H. Deng, X. Tang, Q. Yu, X. Zhou, H. Liu, Adsorption equilibrium and kinetics for
Procedia Chem. 15 (2015) 33–41, https://doi.org/10.1016/j.proche.2015.10.006. SO2, NO, CO2 on zeolites FAU and LTA, J. Hazard. Mater. 203–204 (2012) 111–117,
[45] L. Hauchhum, P. Mahanta, Carbon dioxide adsorption on zeolites and activated car- https://doi.org/10.1016/j.jhazmat.2011.11.091.
bon by pressure swing adsorption in a fixed bed, Int. J. Energy Environ. Eng. 5 [71] R. Sips, On the structure of a catalyst surface, J. Chem. Phys. 16 (1948) 490–495,
(2014) 349–356, https://doi.org/10.1007/s40095-014-0131-3. https://doi.org/10.1063/1.1746922.
[46] M. Saleh Shafeeyan, W. Mohd Ashri, A. Wan Daud, N. Aghamohammadi Shamiri, Ad- [72] J. Toth, State equations of the solid-gas Interface layers, Acta Chim, Acad. Sci.
sorption equilibrium of carbon dioxide on ammonia-modified activated carbon, Hungaricae. 69 (1971) 311–328.
Chem. Eng. Res. Des. 104 (2015) 42–52, https://doi.org/10.1016/j.cherd.2015.07. [73] J. Shang, G. Li, R. Singh, Q. Gu, K.M. Nairn, T.J. Bastow, N. Medhekar, C.M. Doherty, A.J.
018. Hill, J.Z. Liu, P.A. Webley, Discriminative separation of gases by a “molecular trap-
[47] I. Langmuir, The constitution and fundamental properties of solids and liquids. Part I. door” mechanism in Chabazite zeolites, J. Am. Chem. Soc. 134 (2012) 29, https://
solids, J. Am. Chem. Soc. 38 (1916) 2221–2295, https://doi.org/10.1021/ doi.org/10.1021/ja309274y.
ja02268a002. [74] T. De Baerdemaeker, D. De Vos, Trapdoors in zeolites, Nat. Chem. 5 (2013) 89–90,
[48] M.F. Freundlich, Over the adsorption in solution, J. Phys. Chem. 57 (1906) 355–471. https://doi.org/10.1038/nchem.1560.
[49] D. Tiwari, C. Goel, H. Bhunia, P.K. Bajpai, Dynamic CO2 capture by carbon adsor- [75] G.C. Watson, N.K. Jensen, T.E. Rufford, K.I. Chan, E.F. May, Volumetric adsorption
bents: kinetics, isotherm and thermodynamic studies, Sep. Purif. Technol. 181 measurements of N 2 , CO 2 , CH 4 , and a CO 2 + CH 4 mixture on a natural
(2017) 107–122, https://doi.org/10.1016/j.seppur.2017.03.014. Chabazite from (5 to 3000) kPa, J. Chem. Eng. Data 57 (2012) 93–101, https://doi.
[50] A.A. Adelodun, J.C. Ngila, D.-G. Kim, Y.-M. Jo, Isotherm, thermodynamic and kinetic org/10.1021/je200812y.
studies of selective CO2 adsorption on chemically modified carbon surfaces, Aerosol [76] Z. Bacsik, O. Cheung, P. Vasiliev, N. Hedin, Selective separation of CO2 and CH4 for
Air Qual. Res. 16 (2016) 3312–3329, https://doi.org/10.4209/aaqr.2016.01.0014. biogas upgrading on zeolite NaKA and SAPO-56, Appl. Energy 162 (2016)
613–621, https://doi.org/10.1016/j.apenergy.2015.10.109.
[51] D. Iruretagoyena Ferrer, Supported Layered Double Hydroxides as CO2 Adsorbents
[77] A.K. Mishra, S. Ramaprabhu, Magnetite decorated graphite nanoplatelets as cost ef-
for Sorption-Enhanced H2 Production, Springer International Publishing, Cham,
fective CO2 adsorbent, J. Mater. Chem. 21 (2011) 7467, https://doi.org/10.1039/
2016https://doi.org/10.1007/978-3-319-41276-4.
c1jm10996k.
[52] J.a. Mason, K. Sumida, Z.R. Herm, R. Krishna, J.R. Long, Evaluating metal–organic
[78] A.K. Mishra, S. Ramaprabhu, Nano magnetite decorated multiwalled carbon nano-
frameworks for post-combustion carbon dioxide capture via temperature swing ad-
tubes: a robust nanomaterial for enhanced carbon dioxide adsorption, Energy Envi-
sorption, Energy Environ. Sci. 4 (2011) 3030–3040, https://doi.org/10.1039/
ron. Sci. 4 (2011) 889–895, https://doi.org/10.1039/C0EE00076K.
c1ee01720a.
[79] V.K. Singh, E. Anil Kumar, Measurement and analysis of adsorption isotherms of CO2
[53] S.-C. Hsu, C. Lu, F. Su, W. Zeng, W. Chen, Thermodynamics and regeneration studies
on activated carbon, Appl. Therm. Eng. 97 (2016) 77–86, https://doi.org/10.1016/j.
of CO2 adsorption on multiwalled carbon nanotubes, Chem. Eng. Sci. 65 (2010)
applthermaleng.2015.10.052.
1354–1361, https://doi.org/10.1016/j.ces.2009.10.005.
[80] A. Heydari-Gorji, A. Sayari, CO2 capture on polyethylenimine-impregnated hydro-
[54] S. Loganathan, M. Tikmani, S. Edubilli, A. Mishra, A.K. Ghoshal, CO2 adsorption kinet-
phobic mesoporous silica: experimental and kinetic modeling, Chem. Eng. J. 173
ics on mesoporous silica under wide range of pressure and temperature, Chem. Eng.
(2011) 72–79, https://doi.org/10.1016/j.cej.2011.07.038.
J. 256 (2014) 1–8, https://doi.org/10.1016/j.cej.2014.06.091.
[81] Y. Belmabkhout, A. Sayari, Effect of pore expansion and amine functionalization of
[55] R. Serna-Guerrero, A. Sayari, Modeling adsorption of CO2 on amine-functionalized mesoporous silica on CO2 adsorption over a wide range of conditions, Adsorption.
mesoporous silica. 2: kinetics and breakthrough curves, Chem. Eng. J. 161 (2010) 15 (2009) 318–328, https://doi.org/10.1007/s10450-009-9185-6.
182–190, https://doi.org/10.1016/j.cej.2010.04.042.
[82] A. Koide, T. Kihara, Intermolecular forces for D2, N2, O2, F2 and CO2, Chem. Phys.
[56] N. Álvarez-Gutiérrez, M.V. Gil, F. Rubiera, C. Pevida, Kinetics of CO2 adsorption on (1974) https://doi.org/10.1016/0301-0104(74)80004-2.
cherry stone-based carbons in CO2/CH4 separations, Chem. Eng. J. 307 (2017) [83] S. Himeno, T. Tomita, K. Suzuki, S. Yoshida, Characterization and selectivity for
249–257, https://doi.org/10.1016/j.cej.2016.08.077. methane and carbon dioxide adsorption on the all-silica DD3R zeolite, Microporous
[57] C. Goel, H. Kaur, H. Bhunia, P.K. Bajpai, Carbon dioxide adsorption on nitrogen Mesoporous Mater. 98 (2007) 62–69, https://doi.org/10.1016/j.micromeso.2006.05.
enriched carbon adsorbents: experimental, kinetics, isothermal and thermody- 018.
namic studies, J. CO2 Util. 16 (2016) 50–63, https://doi.org/10.1016/j.jcou.2016.06. [84] P. Ning, F. Li, H. Yi, X. Tang, J. Peng, Y. Li, D. He, H. Deng, Adsorption equilibrium of
002. methane and carbon dioxide on microwave-activated carbon, Sep. Purif. Technol.
[58] S.Y. Lagergren, Zur Theorie der sogenannten Adsorption gelöster Stoffe, K. Sven. 98 (2012) 321–326, https://doi.org/10.1016/j.seppur.2012.07.001.
Vetenskapsakad. Handl. 24 (1898) 1–39. [85] V.K. Singh, E.A. Kumar, Comparative studies on CO2 adsorption kinetics by solid ad-
[59] G. Blanchard, M. Maunaye, G. Martin, Removal of heavy metals from waters by sorbents, Energy Procedia 90 (2016) 316–325, https://doi.org/10.1016/j.egypro.
means of natural zeolites, Water Res. 18 (1984) 1501–1507, https://doi.org/10. 2016.11.199.
1016/0043-1354(84)90124-6. [86] E. de Visser, C. Hendriks, M. Barrio, M.J. Mølnvik, G. de Koeijer, S. Liljemark, Y. Le
[60] Y.S. Ho, G. McKay, Kinetic model for lead(II) sorption on to peat, Adsorpt. Sci. Gallo, Dynamis CO2 quality recommendations, Int. J. Greenh. Gas Control. 2
Technol. (1998) https://doi.org/10.1177/026361749801600401. (2008) 478–484, https://doi.org/10.1016/j.ijggc.2008.04.006.
[61] Y.S. Ho, G. McKay, A comparison of chemisorption kinetic models applied to pollut- [87] F Raganati, R Chirone, P Ammendola, Gas–solid fluidization of cohesive powders,
ant removal on various sorbents, Process. Saf. Environ. Prot. 76 (1998) 332–340, Chem Eng Res Des 133 (2018) 347–387, https://doi.org/10.1016/j.cherd.2018.03.
https://doi.org/10.1205/095758298529696. 034.
[62] M.A. Hubbe, S. Azizian, S. Douven, Implications of apparent Pseudo-second-order [88] F Raganati, R Chirone, P Ammendola, Effect of Temperature on Fluidization of
adsorption kinetics onto cellulosic materials: a review, BioResource. 14 (2019) Geldart’s Group A and C Powders: Role of Interparticle Forces, Ind Eng Chem Res
7582–7626. 56 (2017) 12811–12821, https://doi.org/10.1021/acs.iecr.7b03270.
[63] G. Song, X. Zhu, R. Chen, Q. Liao, Y.-D. Ding, L. Chen, An investigation of CO2 adsorp- [89] A Viscusi, P Ammendola, A Astarita, F Raganati, F Scherillo, A Squillace, et al., Alumi-
tion kinetics on porous magnesium oxide, Chem. Eng. J. 283 (2016) 175–183, num foam made via a new method based on cold gas dynamic sprayed powders
https://doi.org/10.1016/j.cej.2015.07.055. mixed through sound assisted fluidization technique, J Mater Process Technol 231
[64] H.N. Tran, S.-J. You, H.-P. Chao, Thermodynamic parameters of cadmium adsorption (2016) 265–276, https://doi.org/10.1016/j.jmatprotec.2015.12.030.
onto orange peel calculated from various methods: a comparison study, J. Environ. [90] F Raganati, R Chirone, P Ammendola, CO 2 Capture by Temperature Swing Adsorp-
Chem. Eng. 4 (2016) 2671–2682, https://doi.org/10.1016/j.jece.2016.05.009. tion: Working Capacity As Affected by Temperature and CO 2 Partial Pressure, Ind
[65] M.M.J. Treacy, J.B. Higgins, Collection of Simulated XRD Powder Patterns for Zeolites, Eng Chem Res 59 (2020) 3593–3605, https://doi.org/10.1021/acs.iecr.9b04901.
Elsevier, 2001.