Abstract

Support-free cross-linked polyethyleneimine sorbent (CPEI) for CO2 capture was evaluated as the regenerable sorbent. The total amines available for the CO2 capture on CPEI were determined by the polyethyleneimine/glutaraldehyde ratio for the synthesis of CPEI. The CO2 capacity of CPEI in the slurry bubble column reactor reached 4.92 mmol/g, which is 1.97 times higher than that obtained under anhydrous conditions. The adsorption kinetics of CPEI in the reactor were investigated in terms of the CPEI amount, the CO2 fraction, the gas flow rate, temperature, and the total amines available. The experimental breakthrough curves for the sorbent were well-fitted with a fractional-order kinetic model. The modeling analysis found the influence of diffusion resistance on the adsorption is more significant than that of the driving force. The CO2 capacity of CPEI remained almost constant during the temperature swing adsorption/desorption cycles.

1. Introduction

Anthropogenic activity is the main cause of the increased atmospheric concentrations of CO2 [1]. The long-term exposure to the CO2 level higher than 0.5% can cause severe health problems, such as fatigue, listlessness, malaise, mood changes, and headache [24]. Hence, the removal of CO2 from cabin atmosphere is a critical function of the life support system of submarine and space capsule [5].

The removal of CO2 from the ambient environment can be achieved by applying a variety of physical and chemical processes such as aqueous solutions of alkanolamine absorption [6], membrane separation [7], and biological capture [8]. Chemical absorption/adsorption is regarded as one of the most promising CO2 capture technologies. It has been widely used due to its high efficiency, low energy consumption, environmental friendliness, etc. [9, 10].

Amine can react with CO2 to form ammonium carbamates and carbonates reversibly at moderate temperature [11]. Chemical absorption using liquid amine has been commercially used in large-scale CO2 capture [12]. However, there are issues of corrosion and degradation in the applications of liquid amines for CO2 capture [13]. Besides, the solution state of CO2 capture makes the endothermic regeneration step (stripping) energy-intensive and costly [14].

Solid adsorbents are promising candidates for reduced regeneration energy, greater capacity, minimized corrosion issue, selectivity, ease of handling, etc. [15]. Amine-based solid sorbents are a group of materials or supports modified physically or chemically with amine groups for CO2 uptake in open and closed environments [1619]. Amine-based solid sorbents can be obtained by (1) impregnating amines into the pores of support, (2) chemically binding amines to support, known as grafting, and (3) polymerizing amine monomers in situ, resulting in polyamine structures tethered to the walls of support [20, 21]. Such hybrid amine-modified sorbents have been used successfully onboard the space vehicles for crew air scrubbing [22].

Among various amine-based solid sorbents, polyethyleneimine- (PEI-) impregnated supports like mesoporous silicas and mesoporous molecular sieve and resin have attracted extensive attention because of their high CO2 adsorption capacity, easy availability, and stability [2326]. The loading of amines onto the outer and interior surfaces of the supports by impregnation is simple and has been extensively studied [27, 28]. However, the physically impregnated PEI on support is structurally unstable and detachment from the support occurred easily in the absence of chemical bonds [29]. Amine-grafted sorbents exhibit more stability over many cycles owing to the formation of covalent bonds between the support surface and amines [30, 31]. However, the covalent tethering of PEI on the support is limited by the surface area, the number of accessible active groups [32]. The solid CO2 adsorbents may be further improved by eliminating the use of inert supports. Wang et al. prepared a support-free polyamine porous particle by cross-linking N-methyl-N-vinylformamide with di[2-(N-vinylformamido)ethyl] [33]. The CO2 capture capacity of the sorbent reached 2.3 mmol/g. Yoo et al. obtained a self-supported, branched poly(ethyleneimine) material with poly(ethylene glycol) diglycidyl ether as the cross-linker via an ice templating method [34]. The CO2 uptake of the material reached 5.5 mol CO2 per kg sorbent at 65% relative humidity. Other cross-linkers, such as glutaraldehyde [35] and diglycidyl ether [36], have also been applied for the cross-linking of polyethyleneimine (PEI) to synthesize support-free CO2 sorbent materials. The CO2 capacity of the support-free sorbent formed by the cross-linking PEI with diglycidyl ether displays a CO2 uptake of 4.43 mmol CO2/g (90°C) under anhydrous conditions and 5.34 mmol/g in the presence of humidity. A slurry bubble column reactor (SBCR) is frequently used where heterogeneous gas-liquid and gas-solid-liquid operations and reactions take place [2]. The reactor has been applied in CO2 capture using magnesium hydroxide or calcium hydroxide particles as sorbents [3739]. In the SBCR, the gas is dispersed in the liquid and/or solid phases in the shape of bubbles, which provides high contact areas. Furthermore, the small particles in the slurry can improve the surface renewal and turbulence in the liquid film at the gas-liquid interface increasing mass transfer coefficients [40]. The presence of fine particles in the liquid film can also hinder the coalescence behavior of bubbles, consequently increasing the gas-liquid interfacial area [40]. In the literature, the evaluation of CO2 capture on solid sorbents by a fixed bed system was often reported [41, 42]. The high particle attrition rate and the pressure drop built up in the fixed bed increase the blower power required to maintain flow and eventually require highly undesirable system maintenance [43].

In the present study, support-free cross-linked polyethyleneimine microspheres (CPEI) was proposed as the regenerable sorbent for CO2 capture. The CPEI is abundant in amino groups by avoiding the use of inert support as in the cases of amine-modified sorbents. The CO2 capacity of CPEI was evaluated by the thermogravimetric analysis (TGA) method and the SBCR method using mixing gases of CO2 and N2.

2. Materials and Methods

2.1. Materials

Polyethyleneimine ( Da, 50% aqueous solution), liquid paraffin (≥99.7%), Span-80 (chemically pure), Tween-80 (chemically pure), and glutaraldehyde solution (25 wt %) were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Pure CO2 gas (>99%) and pure N2 gas (>99%) were purchased from Wuxi Xinxiyi Technology Co., Ltd., China. The other reagents are used as received without further purification.

2.2. Synthesis of CPEI

In a 25 ml stoppered vial ( mm), 7 ml of 20wt % PEI aqueous solution (, adjusted with 12 mol/l HCl), 0.15 g Tween-80, and a certain amount of glutaraldehyde (GA, 25 wt %) were mixed. After the addition of 12 ml liquid paraffin solution of 0.7 g Span-80, the mixture was homogenized at 6000 rpm for 3 min at 20°C. Thereafter, 4 ml of 1.0 mol/l NaOH solution was added dropwise into the mixture at a speed of 2 ml/min. The reaction was carried out 4.5 h at 20°C under magnetic stirring (150 rpm). CPEI was obtained by centrifugation at 15000 rpm for 5 min. The sorbents were washed several times with absolute ethanol and anhydrous diethyl ether and dried under vacuum at room temperature. The influence of various PEI/GA ratio on the properties of CPEI was estimated at 2 : 1, 4 : 1, and 7 : 1 (g : g).

2.3. CO2 Adsorption Capacity of CPEI by the SBCR Method

Scheme 1 shows the experimental setup for CO2 capture using an SBCR as the mass transfer device. It consisted of a cylindrical vessel (120 cm high and 5 cm inner diameter) and a gas sparger at the bottom. The working volume of the reactor was 25 ml. The gas sparger installed in the middle of the bottom of the column was a stainless steel 316 l porous metal micro sparger with a pore size of 15 μm (Mott Corporation, U.S.). The gas was made by mixing pure CO2 gas with pure N2 gas at 101.3 kPa and 25°C. The flow rates of the two gas streams were controlled by mass flow controllers (thermal gas mass flow controller, Cole-Parmer Inc.; range: 0-5 l/min, accuracy: ±0.5% of full scale). The sorbents sat in the column. The temperature was set to the adsorption temperature to be evaluated. The mixed gas was then passed into the reactor. The influent and effluent gas streams flowed into a gas chromatograph (GC) equipped with a thermal conductivity detector (TCD) by an autosampling system. All operation pressure is atmospheric pressure. Before each run, the sorbents in the reactor drained of liquid were regenerated by purging with N2 at 100°C until no CO2 or water was detected in the exhaust gas.

The CO2 uptake capacity curves for CPEI were calculated by integrating the breakthrough curves according to Equation (1) where is the adsorption capacity of slurry (mmol/ml) at time , is the flow rate of the feed gas (mmol/min), is the volume of slurry in the bed (ml), is the voids of the reactor (ml), and is the adsorption time (min). is the pressure of the reactor (Pa). is the temperature of the slurry (K). and are the CO2 concentrations (mol/mol) in the feed gas and the exhaust gas at time , respectively. in Equation (2) is the adsorption capacity of CPEI (mmol/g) at time . is the sorbent fraction in the slurry (g/ml). is the adsorption capacity (mmol/ml) of blank water of slurry.

2.4. Kinetic Model

The fractional-order kinetic model (Equation (3)) proposed by Heydari-Gorji and Sayari has been proved to be suitable for describing the CO2 adsorption on amine-modified sorbents [5, 44]. The fractional-order kinetic model represents the complexity of the reaction mechanism that may involve more than one reaction pathway [45]. In this model, the adsorption rate is proportional to the th power of the driving force and the th power of the adsorption time: where and are the sorption capacity at adsorption equilibrium and at time ; , , and are the model constants. is regarded as the adsorption rate constant of the kinetic model. The value of reflects the pseudo-order of reaction with respect to the driving force. For , Equation (3) turns to the Avrami equation; for and or 2, Equation (3) turns to the pseudo-first-order or pseudo-second-order kinetic model, respectively [44].

The parameters of the fractional-order kinetic model (Equation (3)) were obtained by regression analysis of the experimental data of this work using the Levenberg-Marquardt optimization algorithm in the MATLAB toolbox.

2.5. CO2 Adsorption Capacity of CPEI under Anhydrous Conditions

A thermogravimetric analyzer (Mettler-Toledo International Inc., Zürich, Switzerland) was used to assess the CO2 capacity as a function of time. In a typical adsorption run, about 10 mg of the sorbent was placed in a platinum pan. 100% CO2 gas was used for the adsorption measurements at 101.3 kPa. To drive off the preadsorbed gases, moisture, or any other impurities, CPEI was first degassed at 100°C for about 20 min with N2 at the flow rate of 100 ml/min. Then, the temperature was decreased to the desired adsorption temperature. CO2 capture measurements were carried out by exposing the samples to CO2 with a flow rate of 100 ml/min.

In order to test the reversibility and stability of CPEI, repeated adsorption/desorption operations were performed for 20 cycles to examine the regeneration characteristics of each sample. Regeneration of sorbent was carried out on the same experimental TGA setup. The saturated sorbent of CO2 was purged with purified N2 gas stream at 100°C for about 20 min in order to be generated for the next cycle step.

2.6. Characterization

Fourier transformation infrared spectra (FT-IR) were obtained using a Tensor 27 spectrometer (Bruker Corporation, Ettlingen, Germany). Thermogravimetry analysis (TGA) was performed on a TGA/SDTA851e thermogravimetric analyzer (Mettler-Toledo International Inc., Zürich, Switzerland). The morphological structure of CPEI was examined by an S-4800 scanning electron microscope (Hitachi Ltd., Tokyo, Japan). The diameter distribution of CPEI was determined by the method described by Cohen et al. [46]. A 2,2-bicinchoninic acid (BCA) assay method was used to determine the quantity of the primary and secondary amine group of the sorbent [47]. The tertiary amino groups were determined by titration with perchloric acid [48]. The nitrogen sorption isotherms of CPEI at 77 K were determined by a Micromeritics 3Flex surface characterization analyzer (USA). Before the measurement, the microspheres were degassed at 373 K for 12 hours. The BET surface area and pore volume were determined by Brunauer-Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH) methods.

3. Results and Discussion

3.1. Characterization of CPEI

The CPEI sorbents were synthesized in 12 ml liquid paraffin solution under the conditions of 0.7 g Span-8, 7 ml of 20 wt % PEI solution, 0.15 g Tween-80, and GA (25 wt %) with homogenization at 6000 rpm for 3 min at 20°C. Three types of CPEI, CPEI-1, CPEI-2, and CPEI-3, were obtained at the PEI/GA ratio of 2 : 1, 4 : 1, and 7 : 1 (g : g), respectively. Images of CPEI obtained by scanning electron microscopy (SEM) are displayed in Figure 1. The microsphere of CPEI is regular in shape with a PDI value (the polydispersity index) of 1.04. The diameter distribution of the sorbents was in the range of 1.42-4.45 μm.

FT-IR spectra of CPEI-1, CPEI-2, and CPEI-3 were found to be similar. Using as an instance, the spectrum of CPEI-3 is displayed in Figure 2. The characteristic peaks are at 3534-3422 cm-1 (-N-H stretching), 2930-2855 cm-1 (-C-H stretching), 1653 cm-1 (-C=N stretching), 1569 cm-1 (-N-H in-plane bending), 1446 cm-1 (-C-H bending), 1341-1031 cm-1 (-C-N stretching), and 944 cm-1 (-N-H out-of-plane bending). With CO2 adsorbed on CPEI, the peak at 1569 cm-1 can also be assigned to the N-H deformation in carbamates [49].

According to Figure 3, CPEI-1, CPEI-2, and CPEI-3 showed two common weight losses in the 150-500°C range, with total weight losses of 91%. The two-stage weight loss was accomplished when samples were heated to 278°C and 475°C, respectively. In the first weight-loss event, the sorbent lost 5% of its weight, due to the released CO2 and other small molecules. The CO2 released is previously adsorbed by CPEI from ambient air. In the second stage of weight loss, the total weight loss in 278-475°C was about 85% of the initial weight owing to the decomposition of the sorbent. It can be also found that CPEI decomposed faster with an increasing ratio of PEI/GA (Figure 3).

Glutaraldehyde (GA), epichlorohydrin, N-hydroxysuccinimide esters, and imidoesters have been used to cross-link polymers such as chitosan and protein, through Schiff reactions with amino groups [50]. GA has the advantages of commercial availability, low cost, and high reactivity. Excess glutaraldehyde waste after reaction can be easily detoxified by mixing with glycine [51]. Thus, GA was used in this work to cross-link PEI through emulsion cross-linking polymerization for obtaining CPEI sorbents. According to the results displayed in Table 1, an increase of the value of total amines of CPEI was observed when the PEI/GA ratio used for the synthesis of CPEI increased from 2 : 1 to 7 : 1 (g : g). The further increase of the PEI/GA ratio resulted in the collapsed or broken microsphere (Figure S1). This may be due to the decrease in the strength of the microsphere wall formed by reducing the amount of the cross-linker GA. Therefore, the following studies were conducted with the CPEI sorbents obtained at the PEI/GA ratio of 2 : 1, 4 : 1, and 7 : 1.

According to the data listed in Table 1, the fraction of primary and secondary amines of CPEI increased with the increase of the PEI/GA ratio used for the synthesis of sorbent. The triple ratio of primary amines to secondary amines to tertiary amines (mol : mol : mol) increased from 3.74 : 2.59 : 1 (CPEI-1) and 3.86 : 2.61 : 1 (CPEI-2) to 4.36 : 3.02 : 1 (CPEI-3). Primary amines and secondary amines react rapidly with CO2 to form carbamates (Equations (4) and (5)). According to the mechanism proposed by Caplow [52], the reaction between CO2 and the primary and secondary amine involves the formation of a zwitterion and the deprotonation of the zwitterion by a basic amine, resulting in the formation of carbamate:

The zwitterion can react easily with water via the formation of bicarbonates [18, 53, 54]. In this case, only one amino group is necessary for reaction with one CO2 molecule (Equations (6) and (7)) instead of two in the case of carbamate formation under the anhydrous conditions (Equations (4) and (5)).

Tertiary amines are hindered amines which possess no hydrogen atom attached to the nitrogen atom, as in the case of primary and secondary amines. Thus, tertiary amines do not react directly with carbon dioxide [55]. Instead, tertiary amines facilitate the CO2 hydrolysis reaction forming bicarbonates (Equation (8)) [55]. A base-catalysis reaction mechanism involving a very instable zwitterion was suggested by Yu et al. for tertiary amines [56].

Recently, the use of mixed amine absorbents, such as blends of primary and tertiary amines or secondary and tertiary amines, aroused great interest, due to the combination of the higher equilibrium capacity of the tertiary amine with the higher reaction rate of primary or secondary amine [57]. In order to achieve both fast absorption rates and low regeneration energy consumption, the sorbents such as CPEI of mixed amines becomes more and more attractive [58].

Figure 4 displays the typical N2 adsorption/desorption isotherms of CPEI prepared at the PEI/GA ratio (g : g) of 2 : 1 (a), 4 : 1 (b), and 7 : 1 (c), respectively. As seen from Figure 4, CPEI sorbents show typical type III isotherm. The BET surface area of CPEI sorbent is 7.44, 5.63, and 4.19 m2/g at the PEI/GA ratio of 2 : 1, 4 : 1, and 7 : 1 (g : g), respectively. Correspondingly, the pore volume was determined as 0.022 cm3/g (2 : 1), 0.004 cm3/g (4 : 1), and 0.022 cm3/g (7 : 1), respectively.

3.2. CO2 Capacity of CPEI under Anhydrous Conditions

Thermogravimetric analysis is the common approach used in the literature for determining the CO2 capacity of sorbent under anhydrous conditions [18]. Herein, the CO2 capacity of CPEI-1, CPEI-2, and CPEI-3 was measured at 50°C in the pure CO2 stream by the TGA method, respectively, and the results are shown in Figure 5. The CO2 capacities rise and flatten out in 60 min from the start of adsorption. With the increase of the PEI/GA ratio, the equilibrium CO2 capacity on CPEI sorbents increases and reaches the maximal capacity using CPEI-3. According to the results listed in Table 1, the value of total amines of CPEI-3 is the highest among CPEIs tested, which may account for the highest CO2 capacity on CPEI-3.

Adsorption temperature is a momentous factor for CO2 adsorption kinetics. The CO2 capture curves of CPEI-3 plotted at different isothermal conditions (30, 40, 50, 60, and 70°C) are presented in Figure 6. It shows that the CO2 capacity of the sorbent increased with the increase of adsorption temperature, reaching about 2.5 mmol/g at 50°C. A sharp decrease in the CO2 capacity was observed when the adsorption temperature was higher than 50°C. Based on the mechanism proposed by Song’s group [59], the CO2 adsorption in the cross-linked PEI matrix of CPEI-3 may occur in two stages: the rapid sorption by amines in the exposed outer layer, followed by the diffusion of CO2 into the microsphere to react the interior amines. At low temperature, the interior amines may not be accessible due to diffusion limitation. As the adsorption temperature was increased, the diffusion of CO2 in the cross-linked PEI matrix was improved. The interior amines would be accessible to the CO2, resulting in improved CO2 capacity. However, as the adsorption temperature was increased to be higher than 50°C, the CO2 capacity reduced sharply because the enhancement of the exothermic effect of the reaction between CO2 and amines caused the adsorption equilibrium to shift to desorption, resulting in the decrease of CO2 capacity [60]. A similar result was also reported by Xu et al., Chen et al., and Monazam et al. using the amine-modified sorbents [6163].

3.3. CO2 Capacity of CPEI Sorbents in an SBCR

A slurry bubble column reactor (SBCR) has been used in CO2 removal utilizing fine particles such as magnesium hydroxide or calcium hydroxide as sorbents [37, 39]. In this work, the application of regenerable CPEI in the SBCR system was evaluated for CO2 capture.

The CO2 capacities of CPEI-1, CPEI-2, and CPEI-3 were assayed under the conditions of 0.025 g/ml CPEI, 5 mol % CO2/N2, and 100 ml/min gas flow rate at 20°C (Figures 7(a) and 7(b)). The CO2 capacity elevated with the increase of the total amine value of CPEI (Table 1). Besides, the lesser cross-linked materials using less cross-linker are more permeable to CO2, therefore allowing more amines to react with CO2 [36]. According to Figure 5, the highest value of the CO2 capacity of CPEI is achieved using CPEI-3.

The influence of change of adsorption temperature (from 20°C to 50°C) on the CO2 capacity of CPEI-3 is shown in Figures 7(c) and 7(d). The CO2 capacity reached the maximal value at 30°C under the conditions of 0.025 g/ml CPEI-3 fraction, 5 mol % CO2/N2, and 100 ml/min the gas flow rate under 101.3 kPa. When the temperature was increased higher than 30°C, the CO2 capacity reduced. This may be owing to the negative impact of high temperature on the exothermic reaction of adsorption.

The investigation of the influences of various CO2/N2 mixtures (from 2.5 to 15 mol %) on the CO2 capacity of CPEI-3 was conducted at 0.025 g/ml of the sorbent fraction, 100 ml/min the gas flow rate and at 20°C and 101.3 kPa. The results depicted in Figures 7(e) and 7(f) show the adsorption leveled off in a shorter time when the ratio of CO2 to N2 in flow increased. That is due to enhanced driving force promoting the diffusion rate of CO2 through the increase of CO2 concentration difference between the gas flow and the slurry. It was also found the CO2 capacity was increased slightly, by around 9.8%, when the CO2 fraction in N2 was increased sixfold (from 2.5 to 15 mol %). That implies the CPEI sorbent had almost reached its adsorption saturation at 2.5 mol % of CO2/N2.

Figure 8 shows the influences of different CPEI-3 fractions and various gas flow rates of CO2/N2 mixture on the CO2 capture of an SBCR. After introducing the flow to the SBCR, fast adsorptions of CO2 were observed. Thereafter, the sorbent gradually got saturated and a rise in the CO2 concentration in the outlet gas was detected until complete saturation. Figure 8(b) displays the increase of CPEI-3 fraction in the slurry resulted in the increased CO2 capacity but prolonged the time when the CO2 capacity plateaued. The CO2 capacity reached the maximal value as the CPEI-3 fraction was 0.05 g/ml. The further increase of CPEI-3 fraction resulted in the reduced capacity. That is due to the increased viscosity of the slurry which has negative influences on the diffusion of CO2 in the slurry reducing the capacity of CPEI-3 [64].

The influences of the gas flow rate on the CO2 capture in the SBCR were investigated under the conditions of 5 mol % CO2/N2 mixture, 0.025 g/ml of CPEI-3 fraction at 20°C and 101.3 kPa. According to the results demonstrated in Figures 8(c) and 8(d), with the increase of the gas flow rate, the time for the CO2 capacity to reach the equilibrium decreased. That is because the slurry is agitated more severely with the increase of gas flow rate, promoting the diffusion and the dissolution of CO2 in the slurry [19]. However, when the gas flow rate is higher than a certain critical value, 150 ml/min in this study, the CO2 capacity reduced due to the insufficient contact between the slurry and CO2 at the too rapid gas flow rate.

Above all, the highest CO2 capacity can be obtained with CPEI-3 as the sorbent in the SBCR under the conditions of 0.05 g/ml of the sorbent fraction, 15 mol % of CO2/N2, 100 ml. The CO2 capacity of CPEI sorbent determined by the TGA method is obviously lower than that obtained in the SBCR. That depends on the presence or absence of water in the two adsorption methods. Yoo et al. [34] and Hamdy et al. also found the CO2 capacity of poly(ethylene glycol)-diglycidyl-ether-cross-linked poly(ethyleneimine) materials and diglycidyl-ether-cross-linked polyethyleneimine materials were increased by 1.95 times [34] and 1.2 times [36] in the presence of water, respectively. According to Equations (3)–(5), the presence of water can theoretically double the CO2 capacity due to the formation of bicarbonates during the chemical interaction between the amines and CO2. Therefore, the CO2 capacity of the CPEI sorbent determined by the SBCR method is higher than that by the TGA method. The enhancements in the presence of water observed for CPEI sorbents were within the theoretical limit.

3.4. Kinetic Modeling Analysis

The kinetic of adsorption is important to assess the performance of adsorbents for CO2 removal [21]. To date, the literature provides a number of kinetic models describing the kinetics of the solute adsorption at the solid/solution interface. Several studies have been conducted on the kinetics of CO2 adsorption on sorbents [44]. The fractional-order kinetic model (Equation (3)) developed by Heydari-Gorji et al. [44] demonstrated that two kinds of physical and chemical adsorption took place and was used to fit the data of the CO2 adsorption on amine-modified sorbents [65]. In this work, this model was used to describe the CO2 adsorption on CPEI-3 (Figures 7(b), 7(d), 7(f), 8(b), and 8(d)). The values of the kinetic constants and parameters are listed in Table 2. According to the extremely high values of , the model agrees with the experimental data very well. is regarded as the adsorption rate constant of the kinetic model. The highest value of was obtained at 200 ml/min of gas flow rate under the conditions of 5 mol % of CO2/N2, 0.025 g/ml of the sorbent fraction and at 20°C and 101.3 kPa, using CPEI-3. However, the equilibrium adsorption capacity () did not increase with the increase of the gas flow rate. That is because the increase of the gas flow rate will reduce the gas-liquid contact time [66], resulting in the reduced values.

The parameters and reflect the effect of the driving force and the diffusion resistance [44]. Based on the data listed in Table 2, the values of are higher compared to the values of indicating the diffusion resistance had a greater influence on the rate of CO2 adsorption on CPEI sorbents. Within the temperature range of the study, the value of and reached the extremum, respectively, at 30°C under the conditions of 0.025 g/ml of the sorbent fraction, 5 mol % of CO2/N2, and 100 ml/min of the gas flow rate under 101.3 kPa, using CPEI-3. The CO2 adsorption of CPEI in the SBCR probably occurs in three stages: first, dissolution of CO2 in water; in the second stage, diffusion of CO2 across the water film surrounding each sorbent; finally, the adsorption of CO2 to the sorbent by reacting amines. When the adsorption temperature increased from 20 to 30°C, the value dropped to the minimum, indicating the dissolution and the diffusion rate of CO2 in the water of the slurry reached the maximum. When the adsorption temperature was increased over 30°C, the desorption rate of CO2 from slurry become dominant increasing the value of . On the other hand, the negative effect of high temperature on the thermodynamics of the exothermic reaction between CO2 and the amines of sorbent also resulted in the reduced value of .

The elevated value of with the increase of sorbent fraction is due to the increase in slurry viscosity caused by a strong interaction between particles [67]. Increasing the CO2 fraction can increase both the driving force and the diffusion resistance (Table 2). Therefore, the CO2 capacity did not increase accordingly with the increase of CO2 fraction. The value of increases with the elevation of the PEI/GA ratio since the number of amines available for CO2 adsorption increased.

3.5. Cyclic Adsorption/Desorption

Reusability of adsorbents determines the efficiency and feasibility of a CO2 capture technology [68]. The practical industrial applications require that the sorbents do not simply have a high CO2 adsorption working capacity but also are stable for long-term operation. Ideally, CO2 adsorbents should be regenerable, meaning that they should be able to undergo numerous adsorption/desorption cycles without noticeable loss of adsorption capacity. To study the multicycle behavior of CPEI sorbents in the SBCR for CO2 adsorption, ten consecutive cycles of repeated adsorption/desorption operation were conducted under the conditions of 0.025 g/ml of the sorbent fraction, 5 mol % of CO2/N2, and 100 ml/min of the gas flow rate at 30°C and 101.3 kPa, using CPEI-3. The dynamic curves of the CO2 capture in the SBCR in consecutive cycles of ten runs are displayed in Figure 9. After draining the water out of the reactor, the regeneration of CPEI-3 was accomplished by purging N2 at 100°C until no CO2 or water was detected in the exhaust gas. The results show that the CO2 capacity of CPEI slurry decreased by about after ten cycles of operations.

The regenerability of CPEI-3 was also examined by the TGA method. The CPEI sorbents previously saturated with CO2 were submitted to a temperature swing desorption of the adsorbed gas in the purified N2 gas stream at 100°C for about 20 min. The sorbent was then used again for CO2 adsorption on the same TGA setup. The repeated adsorption/desorption operation was performed for twenty cycles (Figure 10). The CO2 capacity decreased by about 1% after 20 cycles of adsorption/desorption operation. Hence, CPEI sorbent exhibited good regenerability.

4. Conclusions

The CO2 adsorption on the sorbent of CPEI was evaluated using an SBCR and under anhydrous condition by TGA. The CPEI-3 sorbent prepared at the PEI/GA ratio of 7 : 1 demonstrated the highest CO2 adsorption capacity. The CO2 capacity of the sorbent reached 4.92 mmol/g in the SBCR, which is 1.97 times higher than that obtained by TGA (2.5 mmol/g). The CO2 adsorption kinetics of the CPEI slurry in the SBCR were well fitted by the fractional-order kinetic model. This indicates the CO2 adsorption involved more than one reaction pathway. The modeling also demonstrated the diffusion resistance played the main role in the CO2 adsorption by the CPEI sorbent. The CO2 capacity decreased by about after 10 cycles of adsorption/desorption in the SBCR. The promoting effect of H2O on the CO2 adsorption on the CPEI sorbent implies the potential application of the sorbent in the purification of the gas stream in the fermentation plant, the food factory, the thermal power station, or the submarine where water is available.

Data Availability

All data generated or analyzed during this study are included in this published article (and its supplementary information files).

Conflicts of Interest

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Acknowledgments

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: this work was supported by the Postgraduate Research and Practice Innovation Program of Jiangsu Province (SJCX18-0618).

Supplementary Materials

The supplementary material is available free of charge at https://SEM images of CPEI sorbent prepared at different ratio of PEI to GA (g : g) (word). (Supplementary Materials)