Modelling and optimisation by response surface technique for adsorption of carbon dioxide by aminated biosilica/alginate composite: Experiments, characterisation and regeneration studies

ABSTRACT Recently, there has been interest in designing recyclable and economical material for actual environmental remediation. In the present research work, amine-functionalised biosilica/alginate (NH2-SiO2/ALG) composite was evaluated for efficient capture of carbon dioxide (CO2). The morphological and textural properties of NH2-SiO2/ALG composite were comprehensively characterised by BET, SEM, XRD, TGA, and FTIR techniques. The impact of influential factors on the adsorption efficiency procedure was explored in a fixed-bed reactor and the corresponding breakthrough curves were plotted. The optimum value for CO2 removal efficiency (RE)% was predicted as 93.08% at 40°C temperature, 40 mL/min gas flow rate, and 5% moisture content. Under these conditions, the actual CO2 RE% was achieved 92.4%. The exothermic isosteric heat of adsorption (IHA) was found as 25.3–34.8 kJ/mol representing physical interactions between NH2-SiO2/ALG and CO2 molecules. Regeneration tests showed that the synthesised NH2-SiO2/ALG can be effectively recovered and used in multiple runs.


Introduction
Global warming has become a drastic environmental challenge and has created world community concern in recent years [1].Greenhouse gases (GHGs) are the primary concern behind this global issue.Carbon dioxide (CO 2 ), as a primary GHG, is believed to be the main contributor to the evolution of this crisis, contributing around 60% [2].Based on the evidence, the atmospheric levels of CO 2 have risen from about 270 ppm in the preindustrial revolution to 400 ppm today [3], and it is estimated that will reach to 760 ppm by 2100 [4].The growth trend of CO 2 emissions that related to the rising global population along with human activities and their energy demands [5] have caused mostly adverse effects in the last decades and would lead to diverse irreversible impacts for natural and human systems in the near future [6].Therefore, it is most important to mitigate atmospheric CO 2 concentrations using novel physicochemical techniques.Since power plants are the maximum consumer of fossil fuels and produced 40% of global emissions [7]; Hence, post-combustion CO 2 removal from the power plant flue gases can be a potential approach to reduce CO 2 emissions [8].
Currently, the most widely used technology for CO 2 reduction is absorption by liquid amines, but it has several major drawbacks such as the huge equipment requirements, highly corrosive nature of the amines, and desorption problems [9].Thus, the eco-friendly, simple, efficient, low energy consumption and low-cost alternative strategies must be considered.In this case, adsorption process is a befitting method that has the aforementioned features.To date, various types of solid materials have been tested for CO 2 adsorption/capture including zeolites [10], polymer composites [11,12], metal oxides [13], amine-functionalised mesoporous silica [14], and metal organic frameworks (MOFs), to name a few [15]; However, each of these sorbents faces certain privileges and critical defects.Also, their synthesis method, usage conditions, production cost, adsorption efficiency, regeneration capability, and toxicity for the environment and human health are considered as the key factors for choosing a suitable sorbent/for viable applications.With this in mind, of those, biopolymers have attracted the interest of researchers in recent years as a regenerable, efficient, and low cost material [12].Biopolymers prepared from natural sources, are often available, green, eco-friendly, low cytotoxic, and are industrially attractive materials [16].More recently, significant efforts have been made to improve the CO 2 capture/uptake capacity of the adsorbents by modifying them through impregnating or compounding the polymers (with each other or with other materials) because the adsorption process mainly depends on surface basicity, which is attributed the functional groups, specific surface area and porosity of adsorbent [9,17].
Sodium alginate is one of the widely applied biopolymers for synthesis of organic nanocomposites.It is a natural polysaccharide, which has been widely used in a variety of knowledge fields due to its superior features like nontoxicity, good biocompatibility, and biodegradability [18].Nevertheless, raw alginate has revealed some barriers such as poor mechanical strength, low chemical stability, and severe thermal degradation.To overcome this downsides, the researchers used different modifiers, among which, green/ natural base materials are desired [19].This reinforcement action has been revealed to be able to significantly elevate the stability of alginate and its adsorption capability towards pollutants [20,21].Biosilica (SiO 2 ) as a key member of the nature-based materials family, in this study was considered to be supporter for alginate due to its remarkable features, viz., high porosity, good thermal stability, cost-effectiveness, universal accessibility, and environmentally [22].Biosilica not only carries alginate powders but also upgrades its mechanical strength [23].Additionally, to enhance the specific adsorption ability of synthesised composite, we used (3-Aminopropyl)triethoxysilane (APTES) to surface functionalization, which has basic amino groups and serve as the adsorptive sites for acidic CO 2 molecules through a series of mechanisms of electrostatic interaction, hydrogen-bonding interaction, and acid-based interactions [24].It should be noted here that the alginate composites have been extensively investigated in aqueous environments [25,26], but records about its using in gas adsorption are limited.From the viewpoint of practical applications, it is necessary to investigate the adsorption performance experiments by using a continuous flow reactor.To achieve this goal, we have applied the fixed bed reactor (FBR) in our study.The major advantages of the FBR are its construction simplicity, operation easiness, and excellent gas-solid contactity.
To the best of our knowledge, upon the published literature, no relevant study has yet been reported on the utilising of NH 2 -SiO 2 /ALG composite for the capture of CO 2 ; the present research work, was aimed at exploring the prepared NH 2 -SiO 2 /ALG composite for efficient adsorption of CO 2 under the optimised influence of the process parameters including gas flow rate (40-100 mL/min), temperature (40-70°C), and moisture content (5-25%) based on the simulated flue gas conditions in a fixed-bed reactor.Besides, the structure and morphology of NH 2 -SiO 2 /ALG composite were systematically examined by several instrumental techniques, viz., BET, SEM, XRD, TGA and FTIR.In addition, equilibrium, thermodynamic, and mechanisms of adsorption process as well as the reusability of the prepared composite were thoroughly evaluated.

Materials
Nano-powders of commercially available Sodium alginate (alginate, CAS no; 9005-38-3), and biosilica (SiO 2 ; 7631-86-9), the main materials of this research, were bought from Sigma Aldrich (USA).We also purchased (3-Aminopropyl) triethoxysilane from Sigma Aldrich (APTES, USA, CAS no; 9019-30-2).High purity calcium chloride (CaCl 2 , Molar Mass; 110.98 g/mol), methanol (CH 3 OH), ethanol (C 2 H 5 OH), toluene (C 6 H 5 CH 3 ), and epichlorohydrin along with other used epoxides were supplied from Merck (Germany) and employed without further purification.The compressed CO 2 and N 2 gas cylinders with purity greater than 99.99% were bought from Arian Gas Company.All the materials and reagents applied in the present work, were of analytical purity and used as supplied.Besides, in this research study, ultra-pure water (UPW) was freshly prepared with milli-Q Plus system (having 18.2 MΩ electrical resistance), and used.

Synthesis of SiO 2 /ALG
The SiO 2 /ALG composite was prepared based on the literature with minor alterations [27].Accordingly, 4 g of Na-alginate nan-powders was dissolved in 200 mL of distilled water with continuous stirring for 3 h under ambient conditions.Then, it was ultrasonicated for 30 min to get a homogeneous polymer solution.In the next step, 8 g of biosilica were put in the polymer solution and mixed for 2 h.At this point, to obtain bead form composites (SiO 2 /ALG), the desired volume of CaCl 2 solution (2%) was added to the system drop-wise under slow stirring.Then, these beads were kept in CaCl 2 for 24 h.

Functionalisation of synthesised NH 2 -SiO 2 /ALG
The obtained beads in the previous stage were refluxed in toluene-methanol solution (50% v/v) for 30 min, and then 10 mL of APTES reagent was added to the mixture dropwise and stirred vigorously for 30 min; subsequently, the system heated to 75°C for 1 h under reflux.At last, after being washed the amino-functionalised biosilica/alginate beads with ultrapure water and ethanol, they were dried at 80°C in vacuum for 12 h.The product obtained, which is called a NH 2 -SiO 2 /ALG composite, was applied as an adsorptive medium in fixed-bed column to evaluate its potential for adsorbing CO 2 .Furthermore, its capability as a catalyst for chemical fixation of CO 2 was also evaluated.

Adsorption experiments
In the current research, adsorption process of CO 2 on NH 2 -SiO 2 /ALG composites was examined in a lab-scale continuous flow system.Adsorption capacity of the NH 2 -SiO 2 /ALG for CO 2 and the effects of operating parameters on this process were explored according to an experimental design developed using RSM.The experimental set-up of used system was schematically presented in Fig. S1 of Supplementary Material, in which a glass-made column of internal diameter 2 cm and height 20 cm was applied to the fixed bed experiments.Before inserting this column in the system, a known quantity of NH 2 -SiO 2 /ALG composites were loaded in the column and both ends of it were supported by steel mesh followed by glass wool to avoid the adsorbent losses and produce an even flow during the filtration process, respectively.
The procedure for performing adsorption experiments begins with adjusting the gas flow rates of CO 2 and N 2 to purpose flow rates using separately mass flow controllers, which its desired range (40-100 mL/min) was selected based on some preliminary tests.Thereafter, these two gases were mixed together in a mixing chamber.Subsequently, a water bath was employed to evaluate the effects of humidity on adsorption within a range of humidity from 5 to 25%, similar to combined-cycle power plant (CCPP) flue gas moisture, that its control was carried out using an indoor air quality metre.Henceforth, to explore the influence of temperature on CO 2 removal, an adjustable gas heater was used to maintain the temperature of gas at predefined degrees, the same temperature of the CCPP flue gas (40-70°C) after the desulphurisation process.A temperature sensor was utilised to measure the temperature.At this stage, after adjusting the system's pressure at 1 bar, the simulated flue gas passed upwardly through the adsorption column.And finally, the initial and residual concentrations of CO 2 were recorded at regular time intervals until the bed saturation.The uptake capacity of the bed is calculated by using Equations (1-2): Here q eq (mmol/g) signifies the bed capacity; Q (mL/min) is the volumetric flow rate; C 0 (vol%) and C (vol%) index the initial and residual concentrations of CO 2 , respectively; m (g) reports the adsorbent weight or the mass of the bed; and t total (min) shows the operation time.A (min) is related to the pollution concentration changes in the column over a period of time which its values can be obtained by the plot of C/C 0 versus time (the area above the breakthrough curve).
The mass transfer zone (MTZ, cm), as an important portion of fixed bed adsorption zones, represents a region inside the column where adsorption occurs and is calculated via Equation (3) [28]: In this equation, the z (cm) indicates the height of bed; t b (min) and t s (min) refer to breakthrough time at 5% of the initial CO 2 concentration and exhaustion/saturation time at 90% of the initial CO 2 concentration, respectively.Also, to determine the quantity of CO 2 adsorbed in the column (q total , mg), Equation ( 4) was used: In addition, the CO 2 % removal efficiency can be calculated by the following relationship: In Equation ( 5), all symbols are same as aforementioned.

Design generation, statistical analysis and optimisation
In the present study, experimental design, modelling, prediction and optimisation of the adsorption process were conducted by means of Central Composite Design (CCD) combining with RSM.A detailed modelling procedure of this method together with its advantages are well documented in our previous works [29,30].In brief, RSM approach can evaluate the individual and interactive influences of process parameters on the measured responses, as well as determining optimal conditions.Besides, this approach as a useful statistical tool not only leads to minimising the number of experiments but also decreases the consumable chemicals and costs for the experiments.Table S1 of Supplementary Material displays the actual values of the selected parameters and their coded levels.In total, the 17 experiments (including 9 central, 8 factorial/fractional, and 6 axillary points) were designed based on the CCD equation (N = 2 n + 2 n + C), in which N, n, and C, respectively, are attributed to the number of designed runs, the number of influential factors, and central set points.Three trial runs in the central point were executed randomly in duplicate to avoid systematic errors or evaluate experimental inaccuracy.
Herein, to achieve the response value for each run, the continuous adsorption of CO 2 experiments was carried out in the laboratory.After that, the obtained data were correlated with conventional RSM models (including the first-order response-surface, two-way interactions and the full second-order models).Then, the proper fitting model was selected based on four crucial statistic criteria namely, F-value, lack of fit, R 2 and adjusted R 2 , and p-value then applied to determine the optimum values of influential parameters.The Central Composite Design model, which applied to determine a mathematical relationship between the response and the different experimental design variables, shown by Equation ( 6): Where 'Y' demonstrates the response (CO 2 removal percentage), k the number of parameters, b 0 indicates the offset term, b i is attributed to the slope or the linear (first-order) influence, b ii indicates the squared (quadratic) influence, b ij indicates the interaction influence; X i 's correspond to numeric values of the variables and C is associated with the residual error.In addition, the significance level of influential factors and their interactions was affirmed by ANOVA test.Notably, all analyses were conducted using the open source R software.

Modelling of adsorption isotherms
To design adsorption systems, the equilibrium adsorption isotherm should be established based on the factors governing the distribution of pollutant molecules across the gas/ solid phase [31,32].Therefore, in this study, adsorption isotherms were expressed in terms of adsorbent capacity, surface properties, and affinity.In this framework, the isotherm experimental runs were carried out under optimal conditions and different CO 2 pressures (0.5-5 bar).The operation method was the same as presented in section 2.4.Then, the experimental isotherm data were simulated with four of the traditional isotherm models (i.e. the Langmuir, Freundlich, Toth and Sips models).The nonlinear expressions of the isotherm models are described by the equations that are given below: The Langmuir isotherm is essentially based on cover of adsorbent surface with a complete monolayer (homogeneous), adsorbent layers of uniform and independent adsorption of a molecule to a given site.In the Langmuir formula, q m,L (mg/g) indexes the maximum adsorption capacity, P (bar) designates the pressure, b L (L/mg) is the Langmuir coefficient representing the free energy and q L (mg/g) defines the amount of adsorbed CO 2 .Furthermore, to better characterise the reversible and/or irreversible nature and favourability of adsorption mechanism, a non-dimensional parameter, R L , was applied (R L = 1/[1+ bC 0 ]).The values of R L = 0, 0 < R L < 1, R L = 1, and R L > 1 reflect irreversible, favourable, linear, and unfavourable sorption, respectively.On the other hand, the Freundlich isotherm is based on the multi-molecular layer adsorption (heterogeneous), uneven adsorption, and lack of uniform distribution of energy.In the Freundlich formula, q F (mg/g) defines the amount of adsorbed CO 2 ; k F (mg/g) (L/mg) 1/n symbolises the Freundlich capacity variable, and n F (dimensionless) is the Freundlich exponent and indicates the adsorption strength/intensity.The values of n F > 1, n F < 1, and 1 < n F < 10 signify physical, chemical and favourable (linear) sorption, respectively.The Toth model can also be used to explore non-ideal adsorption on heterogeneous surfaces.In the Toth formula, q T (mg/g) denotes the maximum adsorption capacity and q m,T (mg/g) refers to the amount of adsorbed CO 2 ; m T is the Toth model constants (known as heterogeneity parameter); and k t (L/g) shows the Toth coefficient that associated with Henry's law.But, the Sips model is a combination of Langmuir and Freundlich isotherms that predicts the multi-layer adsorption systems.In the Sips formula, q m,s (mg/g) q s (mg/g), respectively, indicate theoretical max adsorption uptake of the Sips model and the amount of adsorbed CO 2 ; k s (L/mg) is the Sips model constant and m s is the Sips exponent.

Nonlinear regression analysis
We considered three terms to select the best fitting isotherm model for prediction of the adsorption process: the greater regression coefficient (R 2 ), the lower average relative error (ARE %), and the smaller residual root mean square error (RMSE).The mentioned terms' values were calculated via the following Equations (7-9) [12,33]: Where, q exp (mg/g) and q modl (mg/g) are the experimental and calculated adsorption capacity, respectively, and n is the number of experiments.

The reusability of NH 2 -SiO 2 /ALG composite beads
In this study, while employing NH 2 -SiO 2 /ALG as adsorbent, a continuous adsorption experiment was carried out and related breakthrough curve was constructed.Then, exhausted bed was regenerated to evacuate the guest molecules and physically adsorbed moisture using continuous N 2 purge at temperature of 100°C for 1.5 h; and reused to the following experiment.The mentioned process was performed for seven consecutive times.Finally, based on breakthrough curves, the adsorption uptakes (mg/g) were calculated by using Equation (4).

Characteristics
Texture, surface morphology and functional groups of produced NH 2 -SiO 2 /ALG composite beads were characterised systematically with BET, XRD, TGA, SEM, and FTIR analysis.The N 2 adsorption/desorption technique was used to inspect specific areas and pore volume of pure alginate and NH 2 -SiO 2 /ALG composite.Figure 1(a, b) displays N 2 -sorption isotherms of them.As seen, both native alginate and NH 2 -SiO 2 /ALG have a type IV isotherm model implying that they could be considered as the mesoporous material based on the IUPAC classification [34].Their isotherm also exhibits H3 hysteresis loop at high pressure indicating the hierarchical pore structures and further confirms the mesoporous nature of the fabricated composite [35].Based on the results of BET analysis, BET surface area was obtained as 178.2 m 2 /g for pure alginate, and 165.5 m 2 /g for NH 2 -SiO 2 /ALG.And according to obtained results of BJH method (Figure 1(c, d)), the total pore volume values for alginate and NH 2 -SiO 2 /ALG were estimated to be 0.17 and 0.551 cm 3 /g, respectively.
To identify the surface morphology and structure arrangements of the studied material, SEM technique was employed.Representative SEM images of pure alginate, pure biosilica, and aminated-SiO 2 /Alginate composite are illustrated in Figure 2. As seen in Figure 2(a, b), the nano-sized alginate surface is smooth/dense, whereas the biosilica nanoparticles have a honeycomb-like texture with circular holes.Meanwhile, the composite displays an uneven and rough surface, indicating that it has greater porous morphology than sodium alginate and biosilica, which is suitable for adsorption (Figure 2(c)).Based on Figure 2(c), the biosilica nanoparticles appropriately immobilised within the alginate matrix.Also, as clearly seen, honeycomb-like network structure of biosilica is entrapped by alginate biopolymer.The crystallographic structure of the products was systematically analysed by XRD.X-ray diffractograms of biosilica, alginate polymer and as-fabricated composite can be seen in Figure 3(a).The diffractogram of biosilica exhibits diffraction peaks at 22.1° and 27.8° at 2θ scale, indicating amorphous nature of pure silica.These results were quite consistent not only with that of other researchers [36] but also with the database in JCPDS (00-001-0647).In native alginate, the intense characteristic peak emerged at 2θ = 11.5°,24.9°, and 38.1°, demonstrating its appropriate crystallinity.Noteworthy, in the diffractogram of NH 2 -SiO 2 /ALG, the peak from alginate 2θ = 11.5°,38.1° and the peaks from biosilica at 2θ = 27.8° were not detected, which associated with the extensive hydrogen bonding between them and resulting the formation of the polyelectrolyte complex [37].
In order to examine the thermal stability of pristine alginate and NH 2 -SiO 2 /ALG composite and provide the appropriate processing temperature for NH 2 -SiO 2 /ALG adsorbent manufacturing, thermogravimetric analysis (TGA) was conducted by heating samples under nitrogen environment up to 800°C at a steady heating rate of 10°C/min.Figure 3(b) shows the TGA thermograms of alginate and NH 2 -biosilica/ALG composite.In the TGA thermogram of alginate, three steps of mass loss were observed.In the first step (at 87.95°C), the occurred mass decrease was about 15.02%, which attributed to the vaporisation of the physically absorbed moisture.In the second step, at temperature 204.15°C, about 5.16% of alginate mass is lost which is presumed to be the breakdown of the saccharide rings (volatile fraction of alginate); And in the third step (at 301.6°C), about 34.41% mass loss occurred, which may be correlated to the complete breakdown of alginate residues [37].The TGA curve phase transition of NH 2 -SiO 2 /ALG resembles with TGA of native alginate, both having three steps of mass losses.At an early step, a mass loss of 6.63% up to 119.24°C was accounted for the loss of hydration water.In the subsequent step, mass decrease of 3.43% that happened up to 221.02°C represents the breakdown of linked APTES molecules and volatile fraction of alginate.In the final step, the weight loss of 12.22% at 311.25°C was related to the decomposition of residual fraction of alginate and APTES molecules.Accordingly, NH 2 -SiO 2 /ALG composite can be regarded as a thermally stable material and could potentially be applied to environmental remediation objectives in the future at high temperatures (below 250°C).
The FT-IR spectra of tested sample prior and post capture of CO 2 were scrutinised and shown in Figure 3(c).Considering the NH 2 -SiO 2 /ALG spectra, the 861 cm −1 broad band relates to the out of plane bending of Si-O and two bands (480 and 1050 cm −1 ) relate to the Si-O-Si stretch.In this spectra, the band located at 1261 cm −1 belongs to the stretching vibration of -CH bonds.The band at 1457 cm −1 is in-plane -CH bending vibration of alkene.In spectrum of 1762 cm −1 , NH 2 -SiO 2 /ALG has a medium peak associated with C = O stretching of the carboxylic groups.Also, it displays presence of one band at 2926 cm −1 which assigns to -CH.The strong and broad bands at 1640 and 3421 cm −1 are related to the O-H and amino groups.Note that due to the monolayer loading of APTES on the beads, some bands are not visible [24].Taken together, these observations affirmed that biosilica and alginate compounded with each other and the aminosilane agent successfully imprinted on the beads surface.After CO 2 capture, some bands have been changed (shifted, disappeared, or emerged), which confirms the chemical nature of the adsorption process and involvement of the composite functional groups in the sorption process.

Effect of the experimental parameters and contour plots
The independent and interactive effects of the process parameters on the recovery/removal of CO 2 were investigated by contour plotting.The effects of temperature (50-70°C) and gas flow rate (40-100 mL/min) on the recovery performance were simultaneously monitored.Changes in the flue gas temperature can exert significant control on various conditions, including the sorption capacity of CO 2 molecules with a changing surface energy, carbonation reaction, and molecular diffusion rate.As portrayed in Figure 4(a), the adsorption rate at lower temperature is high and gets diminished with increasing the temperature value; thus, temperature = 40°C, which was near the temperature of CCPP flue gas, was selected as an optimum temperature.Also, it is worth of notice that at higher temperatures (> 80°C) the carbonation reaction might stop [38].The reduction of removal percentage with increasing temperature is ascribed to the reversible and exothermic nature of the adsorption process.With increasing temperature, the vibration of CO 2 molecules gets intensive and decreases the contact between adsorbent/adsorbate, as a result removal percentage decreased [39].In addition, it can be observed from the breakthrough curves (Fig. S3 of Supplementary Material) that all curves were shifted towards the left with raising temperature and the adsorption capacity declined.As presented in Table 1, by increase of gas temperature from 46°C to 64°C, the adsorption capacity declines from 330.79 mg/g to 291.61 mg/g.With a deep look, investigation of temperature effect on adsorbent performance is important, as it can supply valuable facts about the thermal naturality, spontaneity and feasibility of the adsorption reaction.The above results suggested that the adsorption reaction is exothermic; in addition the decreasing of uptake capacities also indicated the antagonistic effect of temperature on the 'favorability'.In other studies, similar outcomes have been observed [40].
Figure 4(b) portrays the combined influences between gas flow rate and humidity/ moisture content on the CO 2 removal at optimal temperature.As presented in Figure 4(b), the removal of CO 2 reduced from 88.5 to 74% when the gas flow rate enhanced from  52 mL/min to 88 mL/min and according to evidence, the best removal occurred as the flow rate was 40 mL/min.Furthermore, the breakthrough period decreased at higher flow rates (Fig. S3 of Supplementary Material).The reduction of uptake capacity with increasing CO 2 flow rate is ascribed to the decrease of the contact and retention times of CO 2 molecules with sorbent, which leads to a decrease in the quantity of CO 2 adsorbed in the column [6].
We also studied the effect of the interaction between moisture and temperature on the removal efficiency (Figure 4(c)).As clearly observed, the role of moisture content on the removal efficiency is inverse.For example, the removal uptake decreased from 92.5% to 81% when the moisture enhanced from 10% to 20%, even with the temperature was 46°C and 64°C, respectively.This implies the fact that the influence of humidity on the CO 2 removal uptake is more significant at low temperature, as reported previously by Zhao et al. [41].Also, adsorption effectiveness dropped from 91% to 80% when the humidity enhanced from 5% to 25%.At moisture content = 5%, which occurred the excellent efficiency, was selected as an optimum humidity.At lower humidity, the predominant CO 2 species is bicarbonate; in these conditions, there is a high ratio of surface binding sites which makes it easy to adsorb CO 2 molecules whereas, increase in humidity causes the formation of carbonate ions, which leads to saturation of the sorbent active sites with them, as a result removal percentage decreased.Furthermore, based on the breakthrough curves (Fig. S3 of Supplementary Material), with a decline in moisture content from 25% to 15%, the adsorption capacity improved from 247.79 mg/g to 258.22 mg/g.

The adsorption process optimisation
As aforementioned, the reduced model due to satisfactory agreement with the coded data was selected as a well-fitted model.Therefore, optimisation process was done based on the designed model (reduced model); therefore, this model was employed for the optimisation process.Stationary points of response surface are some confirmatory data near optimal conditions.Based on stationary point in original units obtained from the full reduced model, the maximum removal performance (93.08%) with involve all parameters simultaneously, were achieved at temperature 40°C; gas flow rate 40 mL/min and moisture content 5%.Under these conditions, a series of breakthrough experiments were carried out and average amount of the actual removal efficiency was attained approximately 92.4%.The closeness of theoretical removal performance (93.08%) to the actual removal efficiency (92.4%) testified the reliability/high adequacy of the selected model.Note that the aims of optimisation study were segregated as i) to determine the optimum values of influential parameters to get the maximum performance of adsorbent, ii) to provide required CO 2 removal efficiencies from Industrial Flue Gases, which their outlet concentrations change regard to local environmental legislation requirements, and iii) to minimise operational costs through control of practical conditions.

Uptake isotherms investigations
To characterise the experimental data of CO 2 adsorption on fabricated composite, adsorption results were analysed and fitted onto the aforementioned isotherm models (section 2.6), and the optimum values of the models' parameters were achieved by nonlinear fitting algorithms with Solver in Excel.The (q e )-P curves for the models are plotted in Figure 5.As represented in Figure 5, the role of gas pressure on the removal efficiency is positive.Besides, isotherm parameters along with their corresponding R 2 , RMSE and ARE% values for each model are presented in Table 2.According to models outputs, all of the models are valid within the range of experimental conditions.Nevertheless, the Toth and Freundlich models with greater R 2 , lower RMSE and smaller ARE %, indicated good affinity between the sorbent and sorbate among the studied models.These findings declare that adsorption of CO 2 onto NH 2 -SiO 2 /ALG involves a heterogeneous adsorption behaviour with uneven adsorption, and lack of uniform distribution of energy, is also reaffirming the good agreement between characterisation results and experimental outcomes.
According to Table 2, the maximum adsorption capacity of CO 2 per mass unit of NH 2 -SiO 2 /ALG was determined as 346.08 mg/g for Langmuir model, which enlightened that  the NH 2 -SiO 2 /ALG, due to its grafting APTES, can be applied as a potential sorbent to the capture of CO 2 and other acidic gases.Comparison of the maximum uptake between current composite and previous adsorbents for the capture of CO 2 is depicted in Table 3.
The low b L value implies that the sorption process is exothermic, which is in accordance with our findings in the RSM and related contour plotting results.Also, value of R L (0.92) achieved from the Langmuir model, and n F (5.37) achieved from the Freundlich model indicated that adsorption of CO 2 onto NH 2 -SiO 2 /ALG is favourable and physical type, respectively [42].Furthermore, a suitable adsorption capacity achieved from the Sips model (q m,S = 338.47mg/g), which is close to the achieved value from Langmuir model (q m,L ).The high CO 2 efficient capture of NH 2 -SiO 2 /ALG composite is attributed to its porosity as well as electron-rich N species in the amine groups, which increased the basicity of its surfaces.Moreover, the congruence between the outputs of Toth (m T < 1) and Sips (m S > 1) models emphasised that NH 2 -SiO 2 /ALG composite surface has a multilayer structure and CO 2 chemisorption might be a heterogeneous process [12].

Thermodynamic properties
Investigation of temperature effect on the adsorbent performance is important, as it can supply valuable facts about the thermal naturality, spontaneity and feasibility of the adsorption reaction.Thermodynamic studies were investigated in temperature range 313-348 °K under the optimal circumstances (moisture content = 5%, flow rate = 40 mL/min, and pressure = 5 bar).The rate of CO 2 diffusion onto the NH 2 -SiO 2 /ALG was found to decrease linearly with increasing temperature.This can be related to the intensive vibration of CO 2 molecules and decelerated useful contact between gas/solid phase resulting in heighten the temperature [39].Actually, with increasing temperature, the bonds of CO 2 molecules and NH 2 -SiO 2 /ALG disintegrate and as a result adsorbed molecules are separated from composite and return to the flow, as a result the removal percentage decreased.To calculate the major thermodynamic parameters (ΔG°, ΔΗ°, and ΔS°), the following equations were used: Where, q e (mg/g) is related to the amount of adsorbed CO 2 and C e (mg/l) is the concentration of CO 2 at the saturation time, T (°K) denotes the flow temperature and R (8.314 J/ mol.K) refers to the ideal gas constant as well as, ΔG° (kJ/mol) corresponds to Gibbs free energy change, ΔS° (J/mol.K) indicates entropy change, and ΔH° (kJ/mol) represents standard enthalpy change.In brief, K D was obtained from Equation (10), followed by based on Equation (11) the values of ΔG° calculated, then by using Van't Hoff Equation (13), the values of ΔH° and ΔS° calculated.The values of ΔH° and ΔS° were obtained from the slope and the intercept of van't Hoff plot (Figure 6) and tabulated in Table 4.The negative values of ΔG revealed the feasibility and spontaneity of CO 2 molecules sorption on the NH 2 -SiO 2 /ALG composite.The negative value of ΔH° proved that the adsorption of CO 2 molecules onto the NH 2 -SiO 2 /ALG has an exothermic nature [47].Additionally, the enthalpy value demonstrated that the interactions were dominantly from contribution of weak intermolecular forces and physisorption is the main process.The negative value of entropy indicated a consequently decrease in randomness as a result of CO 2 sorption on the surface of NH 2 -SiO 2 /ALG composite [48].Furthermore, the isosteric heat of adsorption (IHA) as another important thermodynamics parameters is conveniently applied to recognise the energetic heterogeneity of composite surfaces.This parameter can provide a lot of information about deep understanding of the strength of interaction between NH 2 -SiO 2 /ALG and CO 2 .In this framework, to determine the value of IHA, adsorption isotherm experiments were performed at various temperatures.Then, its values were calculated with ln(P) versus (1/T) plot and Clausis-Clapeyron equation that can be epitomised as: In Equation ( 14), all symbols are same as aforementioned and θ indicates the thickness of adsorption layer.The calculated IHA values are summarised in Table 4.As seen, the values of IHA are within the range of 25.3-34.8kJ/mol, which authenticate a physical interaction between adsorbate and adsorbent [49].Besides, the negative values of IHA further affirmed the exothermic process of adsorption reaction.In other studies, analogue results refer to the obtained values of IHA and type of contributed interaction in the adsorption reaction of CO 2 on various similar adsorbents were reported [50,51].Overall, small adsorption energy of NH 2 -SiO 2 /ALG for CO 2 along with its great adsorption performance assures its cost-effectivity owing to the low energy requirement for regeneration.

Possible mechanisms of CO 2 sorption
The Lewis acid−base reaction between the reactant molecules (NH 2 -SiO 2 /ALG, H 2 O and CO 2 ) can be predicted from the ionic charge of APTES amine groups (as basic groups) and acidic CO 2 molecules, resulting in the formation of surface ammonium carbamate [52].Furthermore, dipole-quadrupole interaction between electron-rich N species in the amine (-NH 2 ) functional group of APTES and electron-deficient carbon atom of CO 2 molecules, to produce bicarbonates, can be another main mechanism in adsorption process.In addition, the adsorption isotherm and thermodynamic data confirmed that the physisorption, in which pore diffusion and weak hydrogen bonds play dominant role in the adsorption process, involved in the interaction between CO 2 and NH 2 -SiO 2 /ALG.In connection with this, based on the BJH analysis, the tested composite is a mesoporous material with small pores that can result in better porewall interaction of CO 2 , which thereby increases the adsorption efficiency.Besides, the hydrogen bonding between hydrogen containing functionalities of the composite (viz., carboxylic, hydroxyl, and amino groups) and lone pair containing oxygen atoms of CO 2 may be another dominant physical mechanism involved in this process [53,54].Moreover, the (-OH) and (-COOH) groups owing to the higher electron density around (O) could enhance the CO 2 adsorption rate of NH 2 -SiO 2 /ALG composite; which is an agreement with the previous reports [55].All these results revealed that the NH 2 -SiO 2 /ALG, due to its high porosity and large surface area as well as its adsorptive functional groups, has a promising application potential for fast and efficient treatment of CO 2containing gases.

Reusability investigations
The material regeneration and recyclability study in practical applications is important because it helps to better understand the stability and cost-effectivity of adsorbent as well as its environmental-friendly nature.Figure 7 portrays the regeneration of NH 2 -SiO 2 /ALG for CO 2 capture from simulated flue gas.As observed in Figure 7(a), the performance of the tested composite is fairly stable.Also, in Figure 7(b), the sorption capacity was still high (above 250 mg/g) after six cycles, but after the sixth cycle, its performance drastically decreased.This may be due to the low quantity of CO 2 irreversibly chemisorbed in the adsorbent.The result indicated that NH 2 -SiO 2 /ALG is stable to hold of CO 2 and can be reused for at least 6 consecutive times to the adsorptive removal of CO 2 .This affirms a remarkable balance between ecological impact of NH 2 -SiO 2 /ALG and its affordability.

Conclusions
The purpose of this study was to optimize/predict the adsorption potential of NH 2 -SiO 2 /ALG adsorbent towards CO 2 capture by deploying RSM optimisation technique.The physicochemical characteristics of the synthesised composite were comprehensively evaluated by various analytical methods.The influence of operating parameters on the absorbent performance was examined in a fixed-bed reactor and their breakthrough curves were constructed.The obtained data from breakthrough studies were entered to RSM and its outputs suggested gas flow rate as the dominant factor affecting the capture of CO 2 .As well as the modelling findings demonstrated that the adsorption reaction is favourable and exothermic.Regarding optimisation outcomes, the maximum CO 2 RE% was found as 93.08% at 40°C temperature, 40 mL/min gas flow rate, and 5% moisture content.Under the aforementioned provisions, the actual CO 2 RE% was achieved 92.4%.The adsorption isotherm data were well fitted by the Freundlich and Toth models indicating that adsorption of CO 2 molecules occurred on a multilayered surface which supported the data obtained from characteristic analyses.Also, the monolayer adsorption capacity (q m ) reached 346.08 mg/g.Excellent adsorptive performance, well-regenerability, availability, low-cost, and environmentally friendly properties of NH 2 -SiO 2 /ALG composite make it a candidate material towards CO 2 capture from flue gas.

Figure 4 .
Figure 4. Corresponding contour plots indexing the composition effect of (a) temperature versus gas flow rate, (b) moisture content versus gas flow rate, and (c) temperature versus moisture content.

Table 1 .
CCD matrix with uncoded values of operating parameters and results of CO 2 adsorption.

Table 2 .
Adsorption isotherm constants together with determination coefficients.

Table 3 .
Comparison of q m of NH 2 -SiO 2 /ALG composite towards CO 2 capture with other materials.

Table 4 .
Thermodynamic parameters for the adsorption of CO 2 molecules onto the NH 2 -SiO 2 /ALG adsorbent.