Ca-based materials derived from calcined cigarette butts for CO2 capture and thermochemical energy storage

journal of environmental sciences 140 (2024) 230–241 Available online at www.sciencedirect.com w w w . e l s e v i e r . c o m / l o c a t e / j e s Research Article Ca-based materials derived from calcined cigarette butts for CO2 capture and thermochemical energy storage Nabil Amghar 1 , ∗, Virginia Moreno 1 , 2 , ∗,Pedro E. Sánchez-Jiménez 1 , 3 , Antonio Perejón 1 , 3 ,Luis A. Pérez-Maqueda 1 , ∗ 1 Instituto de Ciencia de Materiales de Sevilla, C. S. I. C. - Universidad de Sevilla, C. Américo Vespucio, 49, 41092 Sevilla, Spain 2 Instituto de Tecnología de Materiales (ITM), Universitat Politècnica de València (UPV), Plaza Ferrándiz y Carbonell 1, Alcoy, Alicante 03801, Spain 3 Departamento de Química Inorgánica, Facultad de Química, Universidad de Sevilla, 41012 Sevilla, Spain a r t i c l e i n f o Article history: Received 13 April 2023 Revised 18 July 2023 Accepted 19 July 2023 Available online 27 July 2023 Keywords: Cigarette butts Calcium-Looping CO2 Capture Thermochemical energy storage a b s t r a c t Cigarette butts (CBs) are one of the most common types of litter in the world. Due to the toxic substances they contain, the waste generated poses a harmful risk to the environment, and therefore there is an urgent need for alternative solutions to landfill storage. Thus, this work presents a possible revalorization of this waste material, which implies interesting environmental benefits. CBs were used as sacrificial templates for the preparation of CaO-based materials by impregnation with calcium and magnesium nitrates followed by flaming combustion. These materials presented enhanced porosity for their use in the Calcium Looping process applied either to thermochemical energy storage or CO2 capture applications. The influence of the concentration of Ca and Mg in the impregnating solutions on the multicycle reactivity of the samples was studied. An improved multicycle performance was obtained in terms of conversion for both applications. © 2023 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ) Introduction Cigarette butts (CBs) already constitute the largest share of litter in the world, with several trillions of them thrown away per ∗Corresponding authors. E-mails: na[email protected] (N. Amghar), vmorg[email protected].es (V. Moreno), [email protected] (L.A. Pérez-Maqueda) . year ( Slaughter et al., 2011 ). Smoked CBs are considered solid waste, generating secondary pollution and requiring costly treatments ( Wang et al., 2020b ). The contamination caused by CBs has generated enormous concern, making it mandatory to look for new alternatives and recycling ways to convert https://doi.org/10.1016/j.jes.2023.07.028 1001-0742/© 2023 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ) journal of environmental sciences 140 (2024) 230–241 231 CBs into value-added products ( Kurmus and Mohajerani, 2020 ; Marinello et al., 2020 ; Mohajerani et al., 2016 ; Torkashvand et al., 2020 ). The major fraction of CBs is cellulose acetate (CA), which fibers present limited photo degradability in sunlight ( Puls et al., 2011 ; Soltani et al., 2014 ). Moreover, CBs are composed of small fractions of harmful substances, unburned tobacco, ash and paper ( d’Heni Teixeira et al., 2017 ; Wang et al., 2020b ). The chemical properties of CA are unique, as it contains carboxyl groups in its structure that allow it to be reactive ( Mahalakshmi et al., 2019 ). Thus, properties such as solubility depend on the degree of substitution of the acetate groups ( Khoshnevisan et al., 2018 ). Moreover, CA can be converted into high-valued porous carbon materials suited to CO2 capture, catalysis and energy storage applications ( Chen et al., 2015 ; Jin et al., 2021 ; Koochaki et al., 2020 ; Testa et al., 2023 ). For instance, carbon obtained from CBs has been used as an alternative to conventional activated carbon to fabricate supercapacitor electrodes ( Lee et al., 2014 ; Li et al., 2020 ), materials for hydrogen storage ( Blankenship and Mokaya, 2017 ), and absorbents with high iodine uptake and CO2 adsorption capacity ( Sun et al., 2017 ). FeOx and CoOx materials with superior redox properties have been prepared by combustion of solutionactivated CBs ( Liu et al., 2021 ; Zhou et al., 2022 ). Moreover, CBs have been also used for the synthesis of CaO sorbents for CO2 capture ( Li et al., 2023a , 2023b ). The Calcium Looping (CaL) process, based on the reaction between CaO and CO2 , is used for post-combustion CO2 capture (CCS) ( Benhelal et al., 2021 ; Blamey et al., 2010 ; Guo et al., 2024 ; Imani et al., 2022 ; Sun et al., 2018 ; Yang et al., 2008 ). In this process, the CO2 present in the flue gases in a concentration in the range of 15-20 vol.% is captured by the CaO in a carbonator reactor at about 650 °C, while the resulting CaCO3 is regenerated in a calciner reactor at high temperatures ( ∼900 °C) ( Blamey et al., 2010 ; de la Calle Martos et al., 2016 ; Valverde et al., 2014a ). The CaL process is also under consideration for thermochemical energy storage (TCES), due to the very high reaction enthalpy of carbonation ( Pardo et al., 2014 ; Sarrión et al., 2016 ). Under the Calcium-Looping process for thermochemical energy storage (CaL-TCES) concept, the heat generated by solar or any other appropriate renewable energy source is employed to drive the endothermic decomposition of CaCO3 ( Chacartegui et al., 2016 ). For the discharge stage, the reaction products are brought back together to carry out the reverse exothermic reaction, releasing the heat for on-demand energy production ( Chacartegui et al., 2016 ). This technology facilitates the possibility of long-term energy storage minimizing heat losses ( Ortiz et al., 2019 ). In both CCS and TCES system applications, one of the main inconveniences is the progressive loss of reactivity of CaO with the number of cycles ( Scaltsoyiannes and Lemonidou, 2021 ; Valverde et al., 2013 ; Yu et al., 2012 ). The deactivation is related to the substantial sintering due to the high temperatures required, which cause the maximum conversion attainable to drop sharply each cycle ( Benitez-Guerrero et al., 2017 ; Moreno et al., 2022 ; Sarrión et al., 2018b ; Sun et al., 2007 ). An effective strategy to minimize sintering is the modification of the textural properties of the active CaO material, to obtain a more porous, sintering-resistant morphology. This can be attained by using anti-sintering additives, stabilizing supports or the use of carbonaceous sacrificial templates. Sánchez Jiménez et al. (2019) prepared acicular calcium and magnesium acetate precursors using a simple technique that requires just the natural minerals and acetic acid. Upon thermal decomposition, the precursors yield a stable porous structure comprised of well-dispersed MgO nanoparticles coating the CaO grains, being resistant to pore-plugging and sintering. Antzara et al. (2015) developed mixed CaO- based CO2 sorbents with inert structural promoters prepared via sol-gel auto-combustion. In this case, Al2 O3 and ZrO2 resulted in the most stable sorbents due to the formation of the mixed phases Ca3 Al2 O6 and CaZrO3 . In another work, hollow, MgO-stabilized, CaO microspheres were prepared featuring highly porous multishelled morphologies ( Naeem et al., 2018 ). Ma et al. (2019) used a CaO/Ca12 Al14 O33 sorbent with a microtube-like structure that was prepared from carbide slag and Al(NO3 )3 ·9H2 O using paper fiber as a biotemplate. They concluded that the utilization of the biotemplate increases the cyclic CO2 capture reactivity of the synthetic sorbent. Another work used a citric acid-based carbon template. The carbon template was formed through in situ pyrolysis of citric acid in a simple heating process under nitrogen. Then, a stable porous MgO-stabilized nano-CaO powder was generated and achieved high long-term effective conversion due to its resistance to pore plugging and sintering ( Wang et al., 2020a ). Sarrión et al. (2018a) used natural CaO precursors (limestone and dolomite) and Ca-based composites (Ca3 Al2 O6 /CaCO3 and ZrO2 /CaCO3 mixtures) prepared by ball milling, which is a scalable technique for the production of massive amounts of materials at relatively low cost as required at the commercial scale. However, for a feasible upscaling, any devised strategy must rely on affordable materials and truly scalable synthesis methods. This makes attractive the use of waste materials or byproducts as stabilizers or structure-directing agents ( Benitez-Guerrero et al., 2018b ; Teixeira et al., 2022 ; Valverde et al., 2017 ). In this work, used CBs have been employed as sacrificial templates to support calcium and magnesium precursors in order to produce CaO-based materials with enhanced porosity. It is demonstrated that CBs can be used to fabricate good-performing materials for both CCS and TCES applications using a simple and cost-effective process. 1. Materials and methods 1.1. Materials Calcium nitrate tetrahydrate (Ca(NO3 )2 ·4H2 O, REF. 22399) and magnesium nitrate hexahydrate (Mg(NO3 )2 ·6H2 O, REF. A10329) were purchased from Alfa Aesar (Massachusetts, EE. UU). All aqueous solutions and analytical grade reagents were prepared using 18.2 M /cm water purified with a Mili-Q system (Millipore, Bedford, MA, USA). CBs were randomly collected from different locations on the street. 1.2. Sample preparation The first step consisted of peeling off the cigarette paper on the used CBs. Then, they were heated in distilled water at 75 °C under stirring for 30 min. This process was repeated three 232 journal of environmental sciences 140 (2024) 230–241 Fig. 1 –Schematic diagram of the preparation of treated cigarette butts (TCBs)-derived CaO materials. times to reduce filter waste. Finally, they were dried at 100 °C for 12 hr. To carry out the impregnation of the CBs, the methodology was based on previous works ( Liu et al., 2021 ; Zhou et al., 2020 ). Thus, it was found that 80 mL of water could be fully absorbed by 10 g of CBs. Considering this data, 24 g of CBs were impregnated in 192 mL of calcium nitrate tetrahydrate solution. Different molar concentrations of calcium nitrate tetrahydrate solution were investigated. Moreover, several relative amounts of magnesium nitrate hexahydrate with respect to calcium nitrate tetrahydrate were added since the MgO particles obtained after calcination are expected to preserve the long-term reactivity of the material ( Amghar et al., 2023 ; Daud et al., 2016 ; Kurlov et al., 2016 ; Sarrión et al., 2016 ). The nanometer-sized MgO particles, with high thermal resistance, act as spacers that hinder the agglomeration and sintering of CaO, widening the structure and enabling CO2 diffusion to facilitate the reaction with CaO. However, a welldistributed dispersion is required ( de la Calle Martos et al., 2016 ; Kurlov et al., 2016 ; Sánchez Jiménez et al., 2019 ), which can be reached by impregnation. CBs were maintained under impregnation for 2 days to ensure that the solution was completely absorbed by the fibers. Then, the impregnated CBs were freeze-dried for 24 hr to completely remove the water. Once the samples were completely dry, 8 g of the material was put into a tubular furnace in order to obtain the active materials by calcination. The flaming combustion process took place at a furnace temperature of 300 °C for 10 min to ensure the firing of the cellulose acetate support. A schematic diagram of the preparation of treated cigarette butts (TCBs)-derived CaO materials by flaming combustion synthesis is shown in Fig. 1 . The calcium nitrate-impregnated CBs were named TCBs XCa, being X the concentration of the calcium nitrate tetrahydrate solution in mol/L. In the case of materials prepared by simultaneous impregnation with calcium and magnesium nitrates, they were called TCBs XCaYMg, where Y is the molar percentage of magnesium with respect to calcium contained in the solution. Table 1 summarizes the sample names and the relative molar and mass amounts of CaO and MgO in the samples after combustion. 1.3. Characterization methods The calcination/carbonation tests were carried out in a thermogravimetric analyzer (TGA) Perkin Elmer 8000. This instrument is equipped with a high sensibility balance (0.1 μm) and a furnace that allows for reaching high heating rates as well as stable isotherms. Both, the calcination and carbonation steps, lasted 5 min to mimic the short residence times expected in the calciner and the carbonator. To reduce effects related to gas diffusion resistance through the samples, small amounts of sample (10 mg) were used in each experiment. Two kinds of operating schemes were explored, compatible with CaL-TCES and with Calcium-Looping for CO2 capture (CaL-CCS). The tests compatible with CaL-TCES were performed according to the closed CO2 CaL concept as devised by ( Sarrión et al., 2018a ). The tests started with calcination under a CO2 atmosphere, using a ramp at 100 °C/min from room temperature up to 950 °C. Then, the temperature was decreased to 800 °C under the same atmosphere to carry out the carbonation stage. Afterward, the temperature was raised again to 950 °C to start another cycle. These steps were repeated to complete 20 cycles. The entire experiment was performed under a CO2 atmosphere. These severe conditions yield lower multicyclic performance but they are more feasible for scaling up to the commercial scale as it has been stated in literature ( Ortiz et al., 2019 ). For CaL-CCS experiments, realistic conditions emulating those in real plants were selected. Thus, carbonation was performed at 650 °C for 5 min in an atmosphere of 15% CO2 /85% N2 (vol./vol.), and calcination at 900 °C for 5 min in 70% CO2 /30% N2 (vol./vol.) ( Arcenegui- Troya et al., 2020 ). Heating and cooling rates were also set at 100 °C/min. X-Ray diffraction (XRD) measurements were acquired in a Rigaku Miniflex diffractometer with a Ni filtered Cu K αradiation ( λ= 1.5406 ˚ A) working at 40 kV and 15 mA in the 2 θrange from 20 °to 60 °. The elemental composition of the samples was analyzed by X-Ray microfluorescence (XRF) using an Eagle III Micro XRF instrument (EDAX, New Jersey, USA) equipped with an X-Ray anticathode 50 W rhodium tube and an energy-dispersive X- Ray detector. journal of environmental sciences 140 (2024) 230–241 233 Table 1 –Molar and mass percentage of CaO and MgO in the TCBs-derived CaO materials prepared in this work. Sample Molar percentage Mass percentage CaO MgO CaO MgO TCBs 0.05Ca 100% - 100% - TCBs 0.1Ca 100% - 100% - TCBs 0.5Ca 100% - 100% - TCBs 1Ca 100% - 100% - TCBs 0.1Ca5Mg 95% 5% 96.4% 3.6% TCBs 0.1Ca10Mg 90% 10% 92.6% 7.4% TCBs 0.1Ca20Mg 80% 20% 84.8% 15.2% Table 2 –XRF analysis of untreated CBs and the samples after the flaming combustion synthesis. Sample Elemental composition (wt.%) Al Ca K Mg Si Ti C.L. Untreated CBs 0.37 0.55 0.13 0.15 0.84 0.20 97.8 TCBs 0.1Ca 0.16 20.51 0.03 0.07 0.59 0.06 78.51 TCBs 0.5Ca 0.11 35.62 0.02 0.11 0.45 0.03 63.62 TCBs 1Ca 0.08 49.27 0.02 0.12 0.34 0.04 50.08 TCBs 0.1Ca10Mg 0.21 18.19 0.05 1.45 0.91 0.12 79.07 C.L.: Combustion Losses. The morphological properties of the samples were studied by means of a scanning electron microscope (SEM) Hitachi S4800. To facilitate the imaging, the samples were first sputtered coated with a thin layer of gold employing an Emitech K550 Telstar sputter-coating equipment. The surface area analysis of the samples was performed using a TriStar II (Micromeritics) instrument. The adsorptiondesorption isotherms were quantified by BET analysis using physical gas adsorption of nitrogen gas at an analysis temperature of 77 K. In order to eliminate any residual carbon left from the flaming combustion, the samples were completely calcined in CO2 at 950 °C for 30 min, and then switched to N2 during cooling to avoid the carbonation of the sample in the furnace. In this way, the BET evaluation would correspond to the CaO obtained after the calcination stage. Moreover, before the analysis, the samples were degassed at 400 °C for 2 hr. A Telstar LyoQuest equipped with a vacuum pump system Telstar 2G-6 was used to remove water from the samples. Prior to the process, the samples were frozen with liquid N2 and lyophilized at 0.003 mbar and -55 °C. Particle size distribution was measured by laser diffraction using a Mastersizer 2000 from Malvern. The samples were sonicated for 30 min and dispersed in distilled water to avoid aggregation. 2. Results and discussion 2.1. Characterization of the TCBs-derived CaO materials XRD diffractograms of samples obtained after the flaming combustion process are shown in Appendix A Fig. S1. In order to better observe the phases that are in smaller proportion, the intensity is represented in square root. As may be seen, the main phase in all samples is the calcite polymorph of CaCO3 . Nevertheless, the sample prepared using the solution with the higher calcium nitrate concentration (1 mol/L) shows a significant amount of CaO, probably due to incomplete conversion into CaCO3 during the combustion process. For those samples impregnated with different amounts of calcium and magnesium salts, calcite was also the main phase while reflections corresponding to MgO can also be observed at 43 °, but with low intensity due to the low molar amount of Mg(NO3 )2 employed in the impregnation ( Table 1 ). The elemental compositions of different samples after the flaming combustion synthesis, including untreated CBs are included in Table 2 . As can be seen, the results indicate that untreated CBs contain a number of metals due to the chemicals trapped in the filters ( Aeslina and Mohajerani, 2012 ; Roselli et al., 2021 ). Moreover, as expected, the fraction of CaO increases with the concentration of calcium nitrate in the impregnation solution ranging from 20.51% for the sample TCBs 0.1Ca to 49.27% for the sample TCBs 1Ca. Examples of the N2 adsorption-desorption isotherms and the pore volumes corresponding to the samples TCBs 0.1Ca, TCBs 0.5Ca and TCBs 0.1Ca10Mg are shown in Appendix A Fig. S2. These results indicate a system with unlimited monolayermultilayer adsorption ( Abebe et al., 2018 ), where, first, the adsorption volume increases rapidly at low relative pressures due to the contact of the adsorbate molecules with the more energetic section, followed by the interaction with the less energetic section. It is known that sorption capacity depends on the degree of porosity development, surface area and pore volume ( Daud et al., 2016 ). Moreover, the N2 adsorptiondesorption curves correspond to a type IV which is indicative of mesoporous systems ( Rouquerol et al., 2012 ). This is confirmed in Appendix A Fig. S2b, which displays the pore volume showing that the greatest pore volume corresponds to a pore 234 journal of environmental sciences 140 (2024) 230–241 Table 3 –BET surface area, pore volume and pore size of different samples. Sample BET surface area (m2 /g) Pore volume (cm3 /g) Pore size (nm) Untreated CBs 431.7 0.240 2.2 TCBs 0.1Ca 8.1 0.022 11.0 TCBs 0.5Ca 11.0 0.035 12.7 TCBs 0.1Ca10Mg 10.1 0.036 13.8 Fig. 2 –SEM images of (a) CBs before flaming combustion, (b) TCBs 0.1Ca as produced after flaming combustion, and (c) particle size distribution of TCBs 0.1Ca after flaming combustion. width of 20 to 50 nm, confirming the mesoporous structure of the TCBs samples. The curves for N2 adsorption-desorption (Appendix A Fig. S2a) and pore volume (Appendix A Fig. S2b) have the same shape for all samples indicating that the porosity is obtained from the flaming combustion process and has no dependence of the calcium content. The values of the BET specific surface area, pore volume and pore size for untreated CBs, TCBs 0.1Ca, TCBs 0.5Ca and TCBs 0.1Ca10Mg samples are shown in Table 3 . The impregnated TCBs exhibit largely similar BET surface values, within the 8-11 m2 /g range, which is consistent with the elimination of the porous organic support (CA). Such values are slightly higher than those often reported for CaO derived from limestone and dolomite (3-8 m2 /g), calcined under pure CO2 ( Benitez-Guerrero et al., 2018a ; Manovic et al., 2009 ; Valverde et al., 2015 ). Moreover, the addition of MgO particles (TCBs 0.1Ca10Mg) results in a small increase in the surface area, pore volume and pore size as compared to TCBs 0.1Ca. Allegedly, the insertion of MgO particles allows to space up the CaO grains in the framework as occurs in natural dolomite or synthetic mixtures of calcium and magnesium oxides ( Albrecht et al., 2008 ; de la Calle Martos et al., 2016 ; Kurlov et al., 2016 ; Sultan et al., 2010 ). Fig. 2 includes SEM micrographs of untreated CBs before the flaming combustion, and TCBs 0.1Ca after the flaming combustion. The images show the untreated CBs ( Fig. 2 a) as a hank of cellulose acetate fibers of considerable size ( > 300 μm). The morphology of TCBs 0.1Ca after flaming combustion synthesis ( Fig. 2 b) proves that the acetate fiber structure is lost during combustion, producing a relatively porous structure formed by agglomerated particles. The particle size distribution of TCBs 0.1Ca after flaming combustion is shown in Fig. 2 c. As can be seen, together with the SEM image, the combustion yields small nanometer sized particles that strongly form micron-sized aggregates of a wide range of sizes, with Dv (50) of about 7 μm and Dv (90) as high as 124 μm. 2.2. Evaluation of TCBs performance under a CaL-TCES operating scheme The multicycle activity of TCBs-derived CaO was evaluated under CaL-TCES operating conditions, consisting of carbonations at 800 °C and calcinations at 950 °C under a CO2 atmosphere, as explained in the methods section. A representative selection of thermogravimetric curves of different tested samples (TCBs 0.1Ca, TCBs 0.5Ca, TCBs 1Ca and TCBs 0.1Ca10Mg) is included in Fig. 3 . The multicycle performance of the TCBsderived materials is evaluated in terms of effective conversion. Thus, the conversion was calculated from the mass changes observed in the thermogravimetric curves according to the following equation: Xeff =( mcarb ( t) −m ) m ×WCaO WCO2 =mCO2 ( t) m ×WCaO WCO2 (1) where, m is the mass of the calcined sample (thus including all CaO and inert solids present in the sample), mcarb (t) is the sample mass at the time ( t), and WCaO and WCO2 are the molar masses of CaO and CO2 , respectively. Thus, the difference mcarb (t) −m corresponds to the CO2 uptake during carbonation (mCO2 (t)) . The conversion thereby expressed takes into consideration the presence of non-reactive compounds in the material such as MgO. The thermograms in Fig. 3 clearly show the progressive loss of conversion with the cycle number typical of CaL. The low Tammann temperature of CaCO3 results in intense grain growth and sintering of the surface CaCO3 grains, leading to a progressive reduction of the surface area available for carbonation and also a reduction of reactivity ( Han et al., 2022 ; Valverde et al., 2014b ). Moreover, pore-plugging is an additional limitation mechanism, substantially promoted when carbonation takes place under high CO2 concentration. Under such conditions, the very rapid carbonation creates a thick journal of environmental sciences 140 (2024) 230–241 235 Fig. 3 – Time evolution of temperature and effective conversion for multicycle CaL-TCES tests of (a) TCBs 0.1Ca, (b) TCBs 0.5Ca, (c) TCBs 1Ca, and (d) TCBs 0.1Ca10Mg. The experiments consisted of 5 min calcination stages at 950 °C and 5 min carbonations at 800 °C. blocking layer of CaCO3 that hinders the diffusion of CO2 to the inner pores of the CaO particles, resulting in grains with heavily sintered surfaces and relatively porous inner cores composed of unreacted CaO ( Benitez-Guerrero et al., 2017 ; Duran- Martin et al., 2020 ). As the material is expected to undergo a large number of cycles before it needs to be replaced, the residual conversion constitutes a relevant parameter to compare the performance of different samples. The multicycle conversion data can be generally well fitted following the semiempirical equation ( Valverde et al., 2013 ): Xeff,N =Xeff, 1 k( N −1 ) +1 −Xr /Xeff, 1 −1 (2) where, Xeff,1 is the effective conversion at the first cycle, N is the cycle number, k is the deactivation rate constant and Xr is the residual conversion towards which conversion converges after a large number of cycles. In Fig. 4 a, the conversion attained at the end of each carbonation stage cycle, Xeff,N , is compared for the studied materials, with different concentrations of Ca(NO3 )2 in the impregnation solution. The samples impregnated with solutions of the higher calcium concentrations (TCBs 0.5Ca and TCBs 1Ca) yield higher initial conversion values ( Xeff ,1 = 0.717 and Xeff ,1 = 0.674, respectively). However, the deactivation rate is clearly more pronounced in these samples. Arguably, a larger amount of CaO deposited on the CA fibers after flaming combustion results in a structure more prone to agglomeration and sintering, as it is noticed in the rapid decay in conversion with the cycle number. Thus, during the 20 calcination and carbonation cycles, the conversion of TCBs 0.05Ca drops from 0.440 to 0.198, as compared to the 0.674 to 0.176 decrease obtained for TCBs 1Ca. The highest residual conversion is obtained for TCBs 0.1Ca, which exhibits Xeff,20 = 0.234. Ty pical values of conversion after 20 carbonation/calcination cycles of CaO-derived materials tested under similar operating conditions lie within the range from 0.23 to 0.14 ( Duran-Martin et al., 2020 ; He et al., 2017 ; Sarrión et al., 2018b ; Sun et al., 2013 ). Fig. 4 b illustrates the effect of the simultaneous impregnation with magnesium and calcium precursors on the multicycle conversion. The addition of MgO, obtained after the flaming combustion process, improves the conversion of CaO in all cases. Thus, the optimum load is 10 mol% of MgO, leading to an increase over the residual conversion value obtained for the sample TCBs 0.1Ca. However, the residual conversion decreases slightly when 20 mol% of MgO is added. Arguably, increasing the amount of MgO facilitates its eventual segregation ( Sarrión et al., 2016 ). The values of effective conversion are higher than those reported for other CaO-based materials tested under identical CaL-TCES conditions ( Arcenegui- Troya et al., 2020 ). Table 4 lists, in order to facilitate comparisons, the conversion at the first and the twentieth carbonation stages. Moreover, the residual conversion, deactivation constant and R2 values obtained from the fittings to Eq. (2) are also included. The capability of MgO to minimize the decay in conversion 236 journal of environmental sciences 140 (2024) 230–241 Fig. 4 –Multicycle effective conversion of impregnated TCBs samples. (a) with calcium nitrate and (b) with 0.1 mol/L of calcium nitrate (TCBs 0.1Ca) and different magnesium nitrate concentrations (5 mol%, 10 mol% and 20 mol%) with respect to the molar amount of calcium. TCBs 0.1Ca is presented in both graphics for comparison (red circles). Table 4 –CaO conversion at the first and twentieth cycles, and residual conversion ( Xr ), deactivation constant ( k ) and R -squared ( R2 ) values obtained from fitting to Eq. (2) the conversion values presented in Fig. 4 . Sample Xeff,1 Xeff, 20 Xr k R2 TCBs 0.05Ca 0.440 0.198 0.089 0.199 0.997 TCBs 0.1Ca 0.542 0.234 0.166 0.327 0.999 TCBs 0.5Ca 0.717 0.199 0.075 0.243 0.999 TCBs 1Ca 0.674 0.176 0.090 0.327 0.999 TCBs 0.1Ca5Mg 0.550 0.258 0.178 0.507 0.975 TCBs 0.1Ca10Mg 0.649 0.326 0.181 0.216 0.993 TCBs 0.1Ca20Mg 0.587 0.302 0.176 0.188 0.988 is highlighted in Table 4 . Thus, samples with 5 mol%, 10 mol% and 20 mol% of MgO attenuate the decay as compared with TCBs 0.1Ca, while the sample TCBs 0.1Ca10Mg yields both the higher residual conversion and the minimum decay ( Xr = 0.181, Xeff,20 = 0.326). Consequently, the magnesium-added TCBs samples, especially TCBs 0.1Ca10Mg, improve the residual conversion obtained for limestone tested under similar conditions. SEM images of TCBs 0.1Ca, TCBs 0.5Ca and TCBs 0.1Ca10Mg after 20 calcination/carbonation cycles ending in calcination are included in Appendix A Fig. S3. The sample TCBs 0.5Ca (Appendix A Fig. S3c and d) exhibits a much more sintered morphology as compared to TCBs 0.1Ca (Appendix A Fig. S3a and b), with and without the addition of MgO. In these samples, a puzzle-like microstructure is evident, with spaced CaO grains. Furthermore, TCBs 0.1Ca10Mg in Appendix A Fig. S3e and f shows well-dispersed, small MgO quasi-spherical particles ( Sarrión et al., 2016 , 2018a ). Fig. 5 displays compositional maps obtained by EDS of TCBs 0.1Ca10Mg after being subjected to 1 and 20 calcination/carbonation cycles, ending in calcination. Fig. 5 a-c, corresponding to the sample after the first calcination, show that MgO grains are homogenously distributed throughout the composite material. After 20 cycles, as observed in Fig. 5 d-f, Fig. 5 –Secondary electron micrographs (SEM at 2 and 20 kV) and compositional mapping of CaO derived from TCBs 0.1Ca10Mg showing Ca and Mg after 1 cycle (a-c) and 20 cycles (d-f). MgO eventually segregates from the CaO framework, which limits its stabilizing effect ( Sarrión et al., 2016 ). The semiquantitative analysis obtained by EDS shows that the molar content of Mg in this sample is 8% which is in reasonable agreement with the expected value. 2.3. Energy storage capacity of TCBs-derived CaO The energy storage capacity of the materials, which corresponds to the heat of reaction recovered when CaO is carbonated, has been calculated according to the following equation: Dm,N = H0 R ·mCO2 ( t) m (3) where. H0 R is the reaction enthalpy (4045.5 kJ/kg CO2 ). The energy storage density ( Dv , GJ/m3 ) of the calcined materials has been calculated as follows: Dv,N =Dm,N ·ρ(4) The theoretical densities of the materials were estimated using the rule of mixtures considering the theoretical densities of CaO and MgO. Thus, the densities of the calcined mate- journal of environmental sciences 140 (2024) 230–241 237 Fig. 6 – Accumulated storage energy (ASE) of each sample considering 1000 calcination/carbonation cycles. rials were 3340 kg/m3 for the samples that only contain CaO, and 3348, 3357 and 3376 kg/m3 for the samples containing 5 mol%, 10 mol% and 20 mol% of MgO with respect to the amount of CaO. From the values of Dv,N , the accumulated storage energy (ASE), in GJ/m3 , after a number N of cycles can be defined as: ASE =Dv,N ·N =Xr ·WCO2 WCaO ·H0 R ·ρ·N(5) In this way, the ASE has been calculated up to 1000 cycles using the values of residual conversion displayed in Table 4 as an estimation. As can be seen in Fig. 6 , the values of ASE are higher for the materials prepared from calcined CBs in which MgO was added, since they present greater values of residual conversion ( Table 4 ). This makes CBs, a waste material that is useless and harmful to the environment, an interesting alternative for preparing materials for CaL-TCES applications. 2.4. Evaluation of TCBs performance under a CaL-CCS operating scheme The operating conditions for CO2 capture are quite different from those needed for TCES, as they are imposed by the nature of the post-combustion flue gas. The synthesized CaO- based materials have been tested according to an operating scheme composed of 5 min calcination stages in 70% CO2 /30% N2 (vol./vol.) at 900 °C, and 5 min carbonation stages in 15% CO2 /85% N2 (vol./vol.) at 650 °C ( Benitez-Guerrero et al., 2018a ; Naeem et al., 2018 ; Valverde et al., 2014a ). Fig. 7 displays the time evolution of the temperature and effective conversion for different samples: TCBs 0.1Ca, TCBs 0.5Ca, TCBs 1Ca and TCBs 0.1Ca10Mg. The calcination temperature of 900 °C is too close to the thermodynamic equilibrium to allow for full calcination in just 5 min, which explains the incomplete decomposition in the first few cycles ( Sánchez-Jiménez et al., 2016 ). However, as the material is repeatedly cycled, calcination kinetics fastens, thereby allowing full calcination at the 3rd or 4th cycle ( Sánchez-Jiménez et al., 2016 ). Nevertheless, in order to minimize sintering and better preserve long-term capture capability, 900 °C has been selected for calcination. Fig. 7 – Time evolution of temperature and effective conversion obtained in multicycle tests performed according to CaL-CCS operating conditions: calcination stages at 900 °C in 70% CO2 /30% N2 (vol./vol.) and carbonation stages at 650 °C in 15% CO2 /85% N2 (vol./vol.). (a) TCBs 0.1Ca, (b) TCBs 0.5Ca, (c) TCBs 1Ca and (d) TCBs 0.1Ca10Mg. 238 journal of environmental sciences 140 (2024) 230–241 Fig. 8 –Multicycle evolution of the CO2 uptake of the TCBs samples impregnated (a) with different calcium nitrate concentrations and (b) with 0.1 mol/L of calcium nitrate (TCBs 0.1Ca) and different magnesium nitrate concentrations (5 mol%, 10 mol% and 20 mol%) with respect to the molar amount of calcium. TCBs 0.1Ca is presented in both graphics for comparison (red circles). The performance of the CO2 capture materials is presented in this work in terms of the CO2 uptake, calculated from the effective conversion ( Eq. (1) ), according to Eq. (6) : CO2 uptake =mCO2 m = Xeff ×WCO2 WCaO (6) where, m is the mass of the fully calcined sample (thus including all CaO and any inert solids present in the sample) and mCO2 is the gained mass at the end of each carbonation stage. The residual value of CO2 uptake after a large number of cycles can be obtained, in the same way as in the previous section, using the semiempirical expression: C O2 uptake ( N) =CO2 uptake ( 1) k( N −1) +( 1 −C O2 upta ke ( r) /C O2 upta ke ( 1) ) −1 (7) where, CO2 uptake ( 1 ) is the CO2 uptake at the first cycle, N is the cycle number, k is the deactivation rate constant and CO2 uptake (r) is the residual CO2 uptake which, in practice, is close to the value after hundreds of cycles. Fig. 8 shows the evolution of the CO2 uptake (g CO2 /g calcined material) with the cycle number for the samples tested in this work. The plots in Fig. 8 a suggest the influence of the concentration of the precursor solution is different from what was observed in CaL-TCES operating conditions. Under CaL- CCS conditions, the CO2 uptake is greater for the samples prepared using 0.5 mol/L and 1 mol/L solutions of calcium nitrate (TCBs 0.5Ca and TCBs 1Ca, respectively), thereby those containing a larger amount of CaO. The anomalies observed at the first 3-4 cycles are due to incomplete calcination. As it occurred for the tests performed under CaL-TCES conditions, the addition of MgO benefits the multicycle performance of the materials. The CO2 uptake plots in Fig. 8 b show that TCBs 0.1Ca5Mg and TCBs 0.1Ca10Mg improve the performance of TCBs 0.1Ca by 34% and 61% at the twentieth cycle, respectively. Conversely, TCBs 0.1Ca20Mg presents a similar performance as the non-Mg-added sample TCBs 0.1Ca. This might be attributed to an early segregation of MgO due to the larger content. Table 5 summarizes the CO2 uptake at the first and twentieth cycles, as well as the residual CO2 uptake, deactivation constant and R-squared values. As may be seen, the best residual values for the samples containing only CaO are obtained for TCBs 0.5Ca and TCBs 1Ca. The addition of 5 mol% and 10 mol% of MgO with respect to the molar amount of CaO improves the residual CO2 uptake of TCBs 0.1Ca by 91% and 82%, respectively, proving that a small addition of MgO enhances the CaO-derived TCBs performance, making them a highly promising material for CO2 capture applications. SEM micrographs of the samples after 20 calcination/carbonation cycles ending in calcination are shown in Appendix A Fig. S4. These pictures show a porous and fractured microstructure, which helps CO2 uptake. Moreover, agglomeration and sintering are evident but to a lower extent than in samples subjected to CaL-TCES conditions (Appendix A Fig. S3). This is noticeable in the samples containing only CaO (Appendix A Fig. S4a-d). Similarly, Appendix A Fig. S4e and f show the TCBs 0.1Ca10Mg sample, where it can be appreciated the inert MgO particles covering the surface of the CaO grains. As reported, calcination at high temperatures in a CO2 - enriched environment leads to sintering, generating a decrease in the surface area that remains available for the carbonation process ( Borgwardt, 1989 ). In this work, the highly porous template gives rise to CaO-based materials composed of smaller grains and more resistant to sintering, leading to improved performance. This makes CBs also a great alternative for the preparation of materials for CO2 capture ( Arcenegui-Troya et al., 2020 ; Naeem et al., 2018 ).