Cost-Efficient Strategy for Sustainable Cross-Linked Microporous Carbon Bead with Satisfactory CO2 Capture Capacity.

Cross-linked microporous carbon beads (MCBs) were successfully synthesized via a green, convenient, and cost-efficient strategy derived from a renewable sugar source. Such an approach avoids the time-consuming procedure and the use of corrosive chemical activating agents and toxic solvents and only involves a simple carbonization process, which makes it to be applicable for rapid and large-scale industrial production of MCB materials. The obtained MCBs possessed well-defined microporous structure, narrow pore size, and high surface area. Particularly, the microporosity of the resultant MCBs could be easily tailored to arise primary pores of size 0.5-0.9 nm by adjusting the carbonization temperature and reaction time, which remarkably favor the CO2 capture. The optimal sample of MCBs-9-5 carbonized at 900 °C for 5 h was characterized by high microporosity (80% of the surface area from micropores), especially ultrahigh narrow microporosity (53% of pore volume from micropores of size <1 nm), which endowed it a great satisfactory CO2 uptake of 4.25 mmol g-1 at 25 °C and 1 bar. Significantly, a prominent CO2/N2 selectivity and superior recyclability of MCBs-9-5 were also achieved. Combined with the simple fabrication, the satisfactory adsorption capacity, and high selectivity, MCBs-9-5 should be a promising adsorbent for CO2 capture and separation.

Introduction

Carbon dioxide (CO2), as a primary greenhouse gas, has increased dramatically in the last recent decades owing to the large-scale burning of fossil fuels to meet the energy requirement of world, which results in global warming and anthropogenic climate change, such as the continuous rise of water level in sea, the increasing number of ocean storms, and floods.[1,2] Moreover, a high concentration of CO2 is harmful to humans, especially in the space-limited chambers such as submarines, space ships, and so on.[3] Thus, it is urgent and significant for reducing atmospheric CO2 concentrations to develop efficient CO2 capture and sequestration. Up to now, many technologies have been employed to separate and capture CO2, for example, chemical/physical adsorption, membrane separation, solution adsorption, and cryogenic separation. Among these strategies, adsorption using porous solid adsorbents is the most promising technology because of their unique advantages over other methods such as low cost, low energy consumption, easy handling, and ease of applications.

The key of the adsorption technology is developing a high-performance CO2 adsorbent. Porous solid materials used for the CO2 capture mainly involve porous carbons, zeolites, amine-modified mesoporous silica, metal–organic frameworks, porous organic polymers, graphene-based composite adsorbents, and so on.[4−8] Porous carbons, especially activated carbons, are regarded as one of the most potential candidates for CO2 adsorption owing to their high surface area, large pore volume, easy-to-tunable and developed porosity, low preparation cost, stable physicochemical properties, and low energy consumption for regeneration. However, traditional activated carbons exhibit a relatively low CO2 adsorption capacity of typically ca. 2–3 mmol g–1 under room temperature at 1 bar.[1,9] One efficient way of improving CO2 capture capacity is to introduce basic sites into the skeleton of carbon materials, as the basic sites are believed to promote the interaction between CO2 molecules and carbon surface. For this purpose, many efficient approaches have been developed to design nitrogen-doped porous carbons, including postmodification of anion on the carbon surface and incorporation of nitrogen groups into carbon skeleton.[10,11] Fan et al. synthesized nitrogen-doped microporous carbons by a K2CO3 chemical activation strategy using chitosan as a carbon precursor, which exhibited a high CO2 uptake of 3.86 mmol g–1 at 25 °C, 1 bar.[12] Wang et al. reported N-doped porous carbon hollow spheres with a CO2 adsorption capacity of 2.96 mmol g–1 (25 °C, 1 bar) by using a melamine–formaldehyde nanosphere as a hard template and a resorcinol–formaldehyde resin as a carbon precursor.[13] Unfortunately, designing nitrogen-doping materials usually involves poisonous, expensive, and corrosive raw materials, which can bring unexpected damage for the environment and equipment, restricting their practical applications.[14] Another way to enhance the CO2 capture capacity is to design microporous carbons especially containing rich and narrow micropores with a pore size of <1 nm.[15,16] Wahby et al. synthesized petroleum pitch-derived carbon molecular sieves with developed microporosity with a narrow micropore size of 0.35–0.7 nm by the KOH chemical activation method, which showed an excellent CO2 uptake of 8.6 mmol g–1 (0 °C, 1 bar).[17] Sevilla et al. prepared microporous biomass-based carbon materials via KOH chemical activation of hydrothermal carbons derived from mixtures of algae and glucose, which possessed a large number of narrow micropores (<1 nm) and exhibited a satisfactory CO2 adsorption capacity of 4.8 mmol g–1 (25 °C, 1 bar).[18] These results demonstrate that microporous carbon with a large proportion of narrow micropore size of <1 nm can efficiently improve the CO2 capture capacity. However, the synthesis of well-defined microporous carbons usually adopts the chemical activation route using KOH, NaOH, H3PO4, and ZnCl2 as chemical activating agents, which is a time-consuming, step-tedious, and costly process.[19−22] Meanwhile, these synthetic strategies suffer some other drawbacks, such as corrosion of the equipment, high energy consumption, and so on. Therefore, from the viewpoint of practical applications, it still remains a big challenge to design a more favorable ultramicroporous carbon sorbent for CO2 capture not only with high performance but also with environmental friendliness and cost-effectiveness.

Herein, we develop a green and cost-efficient strategy for preparing cross-linking microporous carbon bead with large numbers of narrow micropores (<1 nm). Compared with other reported methods, this strategy only involves a simple heat treatment procedure and avoids the use of chemical activating agents and toxic organic solvents; meanwhile, it chooses inexpensive and renewable starch sugar as a carbon source, which exhibits an important potential for fast and large-scale preparation of microporous carbons to meet practical industrial applications. The evolution of microporosity with different calcination temperatures and reaction times was investigated. The optimal sample with a high surface area of 1755.4 m2 g–1 and a large micropore surface area of 1444.9 m2 g–1 was obtained by directly carbonizing at 900 °C for 5 h. As a CO2 adsorbent, the CO2 capture capacity of the resultant materials is evaluated and the optimal sample exhibits considerable CO2 adsorption capacities of 6.15 and 4.25 mmol g–1 under 0 and 25 °C at 1 bar, respectively. Furthermore, such a CO2 adsorbent also shows an outstanding recyclability and CO2/N2 selectivity adsorption property.

Results and Discussion

Material Characterizations

Cross-linked microporous carbon beads (MCBs) were synthesized via an air-assisted activated strategy without adding chemical activating agents and toxic organic solvents, and the possible fabrication mechanism of MCB materials is displayed in Scheme . Glucose-based carbon sphere precursors were the colloidal complex mixture of organic compounds with an extremely low carbonization degree. In the beginning stage, these CS precursors were easy to pyrolysis at high temperature to bring a microporous structure with the release of small molecules.[24] Meanwhile, the colloidal glucose-based precursors were converted into cross-linked carbon beads, and the diameter of the bead reduced as a result of the structural shrinkage.[25] Subsequently, under the assistance of air, some of the carbons in the skeleton were burned by O2 in the air to generate a promotive porosity and meanwhile released CO2 during the process of carbon burning escaped through the spherical carbon substrate, which also produced more micropores and even enlarged pores. Moreover, some carbon substrates could react with CO2, yielding CO whilst generate some fine micropores.[26−28] Thus, the microporosity of MCB materials is greatly dependent on the calcination temperature and reaction time. In brief, the porosity of the resulting MCB materials mainly derives from the incomplete burning of carbon taken in the reaction process. Different annealing temperatures and reaction times bring different degrees of carbon burning, resulting in the variation of porosity.

Scheme 1

Schematic Illustration of Probable Mechanism for the Formation of Cross-Linked MCBs

The influences of carbonization temperature and reaction time on the porosity of MCB materials were characterized by the N2 sorption technology. Figure a,b shows the relationship of the porosity of MCB samples with carbonization temperature. It is clear that all samples present a typical type I isotherm of the microporous structure (Figure a), and these micropores should originate from the elimination of surface O- and H-groups during the carbonization process.[29] Notably, although the isotherm of MCBs-10-3 can still be attributed to the type I curve, the adsorbed N2 volume increases in a stepwise manner to the high-pressure region, indicating the increase in the percentage of mesopores in the total porosity of MCBs-10-3, which should be ascribed to the elimination of more volatile substance under high heat treatment temperature, resulting in the coalescence of some narrow micropores to form some larger micropores and even small mesopores (2–3 nm). Such results can be demonstrated by the pore size distribution in Figure b. For the MCBs-7-3 sample, the micropores mainly center in 0.60 and 1.25 nm, and micropores further develop to generate a multiscale microporosity in MCBs-8-3 and MCBs-9-3 samples with an extra micropore of ∼0.83 nm, and MCBs-9-3 should possess a superior microporosity than that of MCBs-8-3, which could be testified from the pore volume value. With continuously increasing annealing temperature to 1000 °C, most micropores of MCBs-10-3 enlarge to 1.25–1.58 nm and some micropores even further enlarge to mesopores of 2.05 nm, bringing the remarkable enhancement of mesoporosity (Table ), which is consistent with the analytic result of isotherm. As shown in Figure S1a (Supporting Information), the proportion of micropores progressively enhances with the rise of temperature from 700 to 900 °C, in spite of the degree of growth in the micropore proportion was slow, whereas when the activation temperature rose to 1000 °C, the proportion of mesopores in the MCBs-10-3 sample sharply increased and accordingly the proportion of micropores in total pores noticeably declined. On the basis of these results, it was easy to draw such a conclusion that the carbonization temperature had a significant impact on the evolution of microporosity and pore proportion in MCB materials.

Figure 1

N2 adsorption–desorption isotherms and pore size distribution of all as-obtained MCBs materials: (a,b) MCBs-x-3 and (c,d) MCBs-9-y.

Table 1

Textural Characteristics and Yields of All of the As-Prepared MCB Samples

sample	SBETa (m2 g–1)	Slangmuirb (m2 g–1)	Smicroc (m2 g–1)	Sultramicrod (m2 g–1)	Smesoe (m2 g–1)	Vtotalf (cm3g–1)	Vmicrog (cm3g–1)	Vultramicroh (cm3g–1)	DHKi (nm)
MCBs-7-3	621.7	710.8	542.8	356.8	78.9	0.27	0.22	0.12	0.66
MCBs-8-3	959.4	1088.1	856.1	381.1	103.3	0.40	0.33	0.16	0.70
MCBs-9-1	859.7	970.2	771.8	251.2	87.9	0.39	0.34	0.08	0.70
MCBs-9-2	1084.2	1211.7	979.6	301.1	104.6	0.45	0.38	0.10	0.72
MCBs-9-3	1177.7	1313.1	1052.9	338.3	124.8	0.48	0.41	0.15	0.74
MCBs-9-4	1526.2	1736.2	1326.7	309.5	199.5	0.64	0.52	0.11	0.78
MCBs-9-5	1755.4	1957.8	1411.9	271.4	343.5	0.76	0.57	0.09	0.80
MCBs-9-6	2018.1	2490.9	1445.3	186.1	572.8	0.91	0.58	0.06	0.84
MCBs-10-3	2201.8	2838.1	1292.9	216.9	908.9	1.01	0.52	0.10	0.85
MCBs-9-5-r	492.8	555.2	439.5	375.5	53.3	0.21	0.17	0.12	0.67

BET surface area.

Langmuir surface area.

Micropore surface area calculated using the V–t plot method.

Ultramicropore (<0.7 nm) surface area calculated from CO2 adsorption at 0 °C using the DFT method.

Mesopore surface area calculated using the V–t plot method.

The total pore volume calculated by single point adsorption at P/P0 = 0.9945.

The micropore volume calculated using the V–t plot method.

Ultramicropore (<0.7 nm) volume calculated from CO2 adsorption at 0 °C using the DFT method.

The median micropore size by the HK method.

Activation time is another critical parameter for developing a superior porosity, and the results are displayed in Figure c,d. Obviously, all N2 adsorption–desorption isotherms exhibit a typical type I isotherm, representing a typical microporous structure (Figure c). The increase of N2 adsorbed quantity indicates the larger and larger surface area and pore volume, and the adsorbed N2 volume gradually increases at the high-pressure region with the prolonging of activation time, suggesting the generation of enlarged micropores and even some mesopores. Figure d depicts the pore distribution curves of all MCBs-9-y samples, and it clearly shows the evolution process of the micropore size with the increase of activation time. All MCBs-9-y samples exhibit hierarchical narrow micropore size distributions of ∼0.48, 0.68, and 0.85 nm as well as a relatively large micropore of ∼1.25 nm. Noticeably, these hierarchical micropores further develop and optimize with the prolonging of reaction time, and as the reaction time progressively increases to 4 h, some new micropores of ∼1.48 nm begin to generate. When the reaction time continuously prolongs to 6 h, such micropores of ∼1.25 and 1.48 nm dominates the pore size, and even larger micropore of ∼1.86 nm are generated in the MCBs-9-6 sample. Figure S1b analyzes the proportion of micro-/mesopores from the surface area and pore volume. Apparently, with the increase of reaction time from 1 to 3 h, the proportion of micropores retains a similar value of ∼89% and then sharply declines as the reaction time increased from 4 to 6 h. A similar result can be drawn from the values of the micropore surface area and micropore volume (Table ), and when the reaction time exceeds 4 h, Smicro and Vmicro extremely slow increase and even reach a saturation value, whereas Smeso and Vmeso gradually enhance, which brings the increase of the Brunauer–Emmett–Teller (BET) surface area and total pore volume. Figure S2 shows the comparison of pore structures of MCBs-9-5 and MCBs-9-5-r. In the absence of air, the MCBs-9-5-r sample has a BET surface area of 492.8 m2 g–1 mainly contributed by the micropore surface area of 439.5 m2 g–1 and these pores should be entirely originated from the pyrolysis of the CS precursor. From the comparison of the pore size distribution (Figure S2b), the MCBs-9-5-r sample owns an indistinct pore size distribution, indicative of the existence of small amount of pores, which is much poorer than the perfect multiscale micropore size distribution of the MCBs-9-5 sample. Such result further testifies the vital effect of air assisted on the generation of superior microporosity. Thus, these results suggest the formation of a multiscale microporous structure in the following stages: (i) the high-temperature pyrolysis of the carbonaceous precursor generated abundant micropores; (ii) with the assistant of air, these micropores further developed; (iii) some ultramicropores were enlarged and even produced mesopores by the combination of the partial carbon skeleton. Figure depicts the micropore size distribution of all as-obtained MCB materials measured by CO2 adsorption at 0 °C using the density functional theory (DFT) method. Obviously, a multiscale micropore can be found in all samples, especially a superior hierarchical narrow micropore size (<0.7 nm) distribution. With the increase of calcination temperature, micropore sizes are gradually developed and enlarged and MCBs-9-3 exhibits a best micropore size in all MCBs-x-3 samples, especially the outstanding narrow micropore size distribution. Similarly, the micropore size shows a similar evolution with the prolonging of calcination time, and micropore sizes are gradually widened from narrow micropores (<0.7 nm) to fine micropores (0.7–1.0 nm) and even large micropores (>1.0 nm), which should be related to the further pyrolysis of the glucose-based precursor, resulting in the release of more volatile substance.

Figure 2

Micropore (<1 nm) size distribution of all as-obtained MCB materials measured by CO2 adsorption at 0 °C using the DFT method: (a) MCBs-x-3 and (b) MCBs-9-y.

Figure S3a,b presents the wide-angle X-ray diffraction (XRD) patterns of all MCB materials with different calcination temperatures and times. All of the samples exhibited two similar broad and weak diffraction peaks at about 2θ = 21.8 and 43.5 °, which correspond to (002) and (100) diffractions of the amorphous carbon skeleton, respectively, suggesting the low crystallinity of the resultant MCB materials.[30] In addition, the intensity of diffraction peaks in XRD patterns gradually weakened from MCBs-9-1 to MCBs-9-6, which could be related to the gradual decrease in the microcrystallite zones of graphite in the MCB samples as the increase of reaction time.[31] The Raman spectra of all MCBs-x-3 materials are displayed in Figure S4. Clearly, two strong peaks can be seen in these materials, which center at ∼1334 and 1587 cm–1. The peak located at about 1334 cm–1 is assigned to the D-band, which should be ascribed to the vibration of carbon atoms with dangling bonds in planar terminations of the disordered graphite-like framework.[32] The other peak centered at about 1587 cm–1 is referred to the G-band, corresponding to the ideal graphite in-plane vibrations with E2g symmetry.[33] Besides, the degree of graphitization can be further characterized by the ratio of relative intensities of D- and G-band peaks (ID/IG). The values of ID/IG for MCBs-7-3, MCBs-8-3, MCBs-9-3, and MCBs-10-3 are 1.08, 0.99, 0.98, and 0.96, respectively, which indicate that the degree of graphitization enhances with the rise of carbonization temperature.

To evaluate the variation of surface functional groups, Fourier transform infrared (FTIR) spectra of the resultant samples are presented in Figure S5. In the CS precursor, the bands at 1710 and 1620 cm–1 should be attributed to C=O and C=C stretching vibrations, respectively, and the results support the concept of aromatization of glucose during hydrothermal treatment.[34] The bands centered at 1000–1300 cm–1 are related to the C–OH stretching vibration and −OH bending vibration, which imply the existence of large numbers of residual hydroxy groups.[35] The other adsorption bands at 1410–1510 and 2840–2930 cm–1 correspond to the C–H bending vibration and stretching vibration, respectively.[36] The broad adsorption band at 3450 cm–1 should be ascribed to −OH bending vibration in the H2O molecule. Such results suggest that the CS precursor should possess abundant oxygen-containing and hydrogen-containing functional groups. Obviously, a similar FTIR spectrum can be found in MCB samples, in which only the C=C stretching vibration at ca. 1620 cm–1 can be clearly found in MCB samples, and the characteristic adsorption bands of C=O and C–O disappear, which should be ascribed to the pyrolysis of precursors at high temperature, resulting in the elimination of surface oxygen-containing groups. In addition, it is clear that the adsorption bands of C–H (at 2840–2930 cm–1) and −OH (at 3450 cm–1) become indistinct and even disappeared with the rise of activation temperature, indicating that the activation temperature has a significant impact on the surface chemical property of MCB materials.

To reveal the evolution of the morphology and pore structure, the as-obtained MCB samples were characterized by electron microscopy. As displayed in Figure a, glucose-based CS precursors present a cross-linked bead structure with an average diameter of ca. 400 nm. Obviously, such cross-linked beads are well-maintained after carbonizing the CS precursor in air atmosphere at different temperatures for 3 h (Figure ), suggesting that the pyrolysis and air-assisted activation process do not destroy the initial structure of the CS precursor. However, the diameter size of cross-linked carbon beads reduces as a result of structural shrinkage, and MCBs-9-3 exhibits an average diameter of ∼200 nm, and even MCBs-10-3 possesses a smaller size of ∼150 nm because of a higher carbonization temperature. Figure S6 shows the scanning electron microscopy (SEM) images of as-prepared MCBs carbonized at 900 °C with different activation times. Similarly, all MCBs-9-y samples also retain the cross-linked bead structure, and importantly, these cross-linked carbon beads retain an almost identical diameter of ∼200 nm, testifying that prolonging the activation time has no influence on the morphological structure. The evolutions of the morphology and structure are further verified by transmission electron microscopy (TEM) images. As shown in Figure , MCBs-x-3 materials clearly exhibit the progressive variation of the morphology and porous pore structure with the change of the carbonization temperature. Consistent with the SEM observation, all MCBs-x-3 materials present the cross-linked bead structure. Furthermore, from the high-magnification TEM images (Figure e–h), large numbers of nanopores are easily discernible on the surface of carbon beads. Meanwhile, it could be easily observed that these pores in MCBs-x-3 samples gradually enlarge as the increasing activation temperature. Similar results can be found from the TEM images of MCBs-9-y samples (Figure S7). Such results further confirm that the carbonization temperature and reaction time play a crucial role in the development of porosity of MCB materials.

Figure 3

SEM images of as-prepared materials: (a) CS precursor; (b) MCBs-7-3; (c) MCBs-8-3; (d) MCBs-9-3; and (e) MCBs-10-3.

Figure 4

TEM images: (a,e) MCBs-7-3; (b,f) MCBs-8-3; (c,g) MCBs-9-3; and (d,h) MCBs-10-3.

CO2 Capture Performance

The CO2 adsorption capacities of all as-obtained materials are investigated at two representative temperatures (0 and 25 °C), and the adsorption isotherms are presented in Figure , and the related uptake capacities are listed in Table . As shown in Figure a, it can be observed that the MCBs-7-3 sample has a CO2 adsorption capacity of 3.79 mmol g–1 at 0 °C and 1 bar. With the increasing activation temperature, the obtained MCBs-8-3 and MCBs-9-3 samples exhibit the enhanced CO2 capture capacities of 4.74 and 5.64 mmol g–1, respectively; however, the further increase in the annealing temperature brings a negative effect on the CO2 adsorption capacity, and a declined CO2 uptake value of 5.39 mmol g–1 is obtained over the MCBs-10-3 sample. Figure b displays the CO2 adsorption isotherms of MCBs-9-y series materials; it can be seen that CO2 uptakes are in the range of 4.85–6.15 mmol g–1, and the CO2 adsorption capacities of MCBs-9-y materials are improved and then declined with the progressive prolonging of reaction time. Similarly, the same variation tendency of CO2 adsorption capacity over MCBs-x-3 and MCBs-9-y series materials tested at 25 °C can also be observed in Figure c,d, respectively. For all the as-prepared materials, the CO2 uptake enhances with the increase of CO2 pressure. Noticeably, at relatively low pressure, the CO2 uptakes of both MCBs-x-3 and MCBs-9-y series materials, especially MCBs-10-3 and MCBs-9-6 samples, are a little lower at higher activation temperature and longer reaction time, which should be related to the size effect[37] because the micropore size of the obtained MCB materials further developed and enlarged gradually as the temperature and time increased, resulting in the decrease of the proportion of micropores (<1 nm).

Figure 5

CO2 adsorption isotherms of all as-obtained MCB samples at 0 °C (a,b) and 25 °C (c,d).

Table 2

CO2 Uptake of the As-Obtained MCB Materials at Different Temperatures and Pressures

sample	0 °C, 0.15 bar	0 °C, 1 bar	25 °C, 0.15 bar	25 °C, 1 bar
	CO2 uptake per mmol g–1 (mg g–1)
MCBs-7-3	1.75 (77)	3.79 (166.76)	0.89 (39.16)	2.51 (110.44)
MCBs-8-3	1.86 (81.84)	4.74 (208.56)	0.95 (41.8)	3.05 (134.2)
MCBs-9-1	1.94 (85.36)	4.85 (213.4)	1.01 (44.44)	3.15 (138.6)
MCBs-9-2	1.89 (83.16)	5.07 (223.08)	0.99 (43.56)	3.29 (144.76)
MCBs-9-3	1.96 (86.24)	5.64 (248.16)	1.03 (45.32)	3.58 (157.52)
MCBs-9-4	1.84 (80.96)	5.85 (257.4)	1.28 (56.32)	3.92 (172.48)
MCBs-9-5	1.78 (78.32)	6.15 (270.6)	1.35 (59.4)	4.25 (187)
MCBs-9-6	1.72 (75.68)	5.32 (234.08)	0.97 (42.68)	3.35 (147.4)
MCBs-10-3	1.33 (58.52)	5.39 (237.16)	0.98 (43.12)	3.38 (148.72)
MCBs-9-5-r	1.86 (81.84)	3.72 (163.68)	1.09 (47.96)	2.79 (122.76)

The micropore surface area and micropore pore volume, especially micropores of size <1 nm, play a significant role in CO2 adsorption, and the relevant analysis results are displayed in Figure and Figure S8, respectively. Apparently, even though MCBs-9-6 and MCBs-10-3 possess higher total surface area and larger total pore volume compared with those of MCBs-9-3, MCBs-9-4, and MCBs-9-5 materials, their CO2 uptakes are much lower than those of MCBs-9-3, MCBs-9-4, and MCBs-9-5, which should be attributed to their lower micropore (<1 nm) surface areas and micropore pore volumes. As displayed in Figures and S7, the CO2 uptake (at 0 and 25 °C, 1 bar) is not determined by the total surface area and pore volume, but by the pore size, especially narrow micropores of <1 nm, which are more efficient than larger micropores and mesopores, creating stronger interactions between CO2 molecules and carbon adsorbents. Such results are consistent with the previous reported works.[38−40]

Figure 6

Dependence of CO2 capture capacity (at 25 °C and 1 bar) on surface area/micropore (<1 nm) surface area (a) and pore volume/micropore (<1 nm) volume (b) for as-prepared MCBs-9-y materials.

On the basis of the above results, it can be clearly found that MCB materials display a good property for CO2 capture, exhibiting an excellent potential as CO2 adsorbents. The optimal sample of MCBs-9-5 shows the highest CO2 capture values of 6.15 mmol g–1 (at 0 °C, 1 bar) and 4.25 mmol g–1 (at 25 °C, 1 bar), which should mainly be related to the presence of large numbers of narrow micropores (<1 nm). To examine the influence of air activation, the MCBs-9-5 and MCBs-9-5-r samples were compared with regard to their CO2 adsorption capacities (Figure S9). Obviously, the CO2 uptakes of MCBs-9-5-r are only 3.72 mmol g–1 (at 0 °C, 1 bar) and 2.79 mmol g–1 (at 25 °C, 1 bar), which are much lower than those of the MCBs-9-5 sample. Table lists the CO2 adsorption capacities of a variety of reported carbon-based adsorbents. Apparently, the CO2 adsorption capacity of the MCBs-9-5 material is comparable and even higher than other similar carbons and even some of the reported nitrogen-doped porous carbons and hierarchical porous carbons derived by chemical activation. For example, Wang et al. prepared a nitrogen-doped porous carbon hollow sphere with an ultrahigh nitrogen content of 15.9 wt %, which showed a CO2 uptake of 4.42 mmol g–1 at 0 °C and 1 bar.[13] Cai et al. reported poly(vinylidene chloride)-based carbon with ultrahigh microporosity prepared by KOH activation, which exhibited a lower CO2 adsorption capacity of 3.64 mmol g–1 at 25 °C and 1 bar.[41] Therefore, from these comparisons, it is clear that the resultant MCBs-9-5 material via this green, cost-efficient, and free-chemical activating agent strategy achieves a satisfactory CO2 capture capacity, which should be attributed to high surface area, optimal micro-/mesopore proportion, and developed microporosity, especially large microporosity dominated by pores of 0.5–0.9 nm.

Table 3

Comparison of the CO2 Adsorption Capacity Over Different Carbon-Based Adsorbents

sample	CO2 uptake (mmol g–1)	T (°C)a	pressure (bar)	selectivityb	refs
AC-2-635	5.90/3.86	0/25	1	21	(9)
RFL-500	3.13	25	1	NA	(11)
N-TC-EMC	4.0	25	1	14	(35)
CP-Z-700	5.9/3.1	0/25	1	NA	(45)
TB-MOP	4.05/2.57	0/25	1	50.6	(46)
MC-200D8H	2.73	25	1	21.3	(51)
AC	2.8/1.8	0/25	1	17	(52)
Om-ph-MR	2.5/1.77	0/25	1	NA	(53)
PTEB	3.47	0	1	25.9	(54)
NSC	4.8/3.1	0/25	1	NA	(55)
commercial ACs	2–3	25	1	NA
MCBs-9-5	6.15/4.25	0/25	1	61	this work

Adsorption temperature.

Adsorption selectivity of CO2/N2 calculated by IAST at 25 °C; NA represents not available.

CO2/N2 Selectivity

From the viewpoint of practical applications, superior adsorbents should possess a high selectivity against other gases in addition to high CO2 uptake. To estimate the CO2 separation performance of the optimal sample of MCBs-9-5 and the reference sample of MCBs-9-5-r, their CO2 and N2 isotherms at 0 and 25 °C are displayed in Figure a,b, respectively. Clearly, the adsorption capacity of N2 is much lower than that of CO2 at the same condition for both MCBs-9-5 and MCBs-9-5-r samples. At 25 °C and 1 bar, the MCBs-9-5 sample has a CO2 uptake of 4.25 mmol g–1 and a N2 adsorption capacity of 0.24 mmol g–1, which translates to an equilibrium CO2/N2 adsorption ratio of 18, and this adsorption ratio is far higher than the CO2/N2 adsorption ratio of 8 for the MCBs-9-5-r sample, suggesting the better selectivity of MCBs-9-5 for CO2 from N2 than that of MCBs-9-5-r. More importantly, the CO2/N2 adsorption ratio of 18 for MCBs-9-5 is much higher than that of conventional porous carbons (typical 5–11) and even comparable to those of nitrogen-doped carbons and microporous organic polymer-derived carbons.[42−45]

Figure 7

CO2 and N2 adsorption isotherms of MCBs-9-5 (a) and MCBs-9-5-r (b) at 0 and 25 °C; IAST selectivity of CO2/N2 on MCBs-9-5 (c) and MCBs-9-5-r (d) at 0 and 25 °C.

We also employ the ideal adsorbed solution theory (IAST) model to further evaluate the selectivity for CO2 adsorption from the simulated postcombustion flue gas, which is an often used method to estimate the selectivity of solid adsorbents for any two gases in a binary gas mixture. The CO2/N2 ratio is 15/85 in the calculation, which represents the typical composition of the flue gas. The selectivity for CO2 adsorption from the IAST model was derived from the following equationwhere S is the selectivity for CO2, p is the uptake of adsorbed CO2/N2, q(CO2) is 0.15, and q(N2) is 0.85. The calculation results of IAST selectivity are presented in Figure c,d. MCBs-9-5 exhibits the CO2 adsorption selectivity of 49 and 61 at 0 and 25 °C under 1 bar, respectively, which are higher than the CO2 adsorption selectivity of 29 and 41 for MCBs-9-5-r. Importantly, the CO2/N2 selectivity of MCBs-9-5 is remarkably high compared with that of the other porous carbon-based sorbents and even comparable with that of nitrogen-rich porous carbons,[46−49] as shown in Table . The good selectivity in combination with the satisfactory adsorption capacity makes the MCBs-9-5 material highly promising in the selective capture of CO2 from gas mixtures containing N2.

CO2 Isosteric Heat of Adsorption (Qst) and Regeneration of the Adsorbent

To further reveal the interaction between the CO2 molecule/adsorbents and evaluate the energetic heterogeneity of surfaces of adsorbents, the isosteric heat of adsorption (Qst) was calculated from the CO2 adsorption isotherms measured at 0 and 25 °C. As depicted in Figure , the Qst values for all as-obtained MCBs at low surface coverage are ranged from 23 to 38 kJ mol–1. Obviously, the heat of adsorption gradually declines at the low CO2 uptake and then reaches a near plateau with the increasing CO2 uptake, manifesting the heterogeneity of interaction between CO2 molecules and the surface of adsorbents, which could be ascribed to the continuous occupation of adsorption active sites with the growing CO2 uptake. Moreover, it is worth of notice that the Qst values of MCBs-x-3 samples gradually decline with the increase of activation temperature, indicating the weaker and weaker interactions between adsorbent surfaces and CO2 molecule as the elevated activation temperature.[50] Such result should be related to the poorer and poorer microporosity, especially the decreased narrow microporosity (<1 nm) with the rise of activation temperature, resulting in the weaker interaction. Similarly, the Qst values of MCBs-9-y present a roughly similar variation trend, and the Qst values gradually decrease with the prolonging of activation time, which could also be attributed to the variation of micropores’ proportion and size. For the MCBs-9-5-r sample, even though it has a higher Qst value than those of other MCBs-9-y materials, its Qst value continuously declines with the enhanced CO2 uptake, suggesting the nonuniform and unstable CO2 adsorption on the MCBs-9-5-r adsorbent surface, resulting in the poor regeneration and reversibility, which seriously restricts it to be an efficient CO2 adsorbent.

Figure 8

Isosteric heat of CO2 adsorption on all as-obtained MCBs-x-3 (a) and MCBs-9-y (b) materials calculated from the adsorption isotherms at 0 and 25 °C.

Apart from the high CO2 uptake capacity, the remarkable CO2 adsorbents should possess easy regeneration ability for practical utilization. The regeneration test of the MCBs-9-5 material was conducted for five consecutive cycles at 0 and 25 °C. As shown in Figure S10, the CO2 uptake remains almost constant in five cycles, which testifies that the obtained MCBs-9-5 material has satisfactory recyclability with relatively low energy requirement for regeneration.

Conclusions

In summary, a simple and cost-efficient strategy was successfully developed for the green synthesis of cross-linked MCBs derived from sustainable starch sugar. Inimitably, this approach avoids the use of chemical activating agents and only involves a simple curing process in an air atmosphere, which makes it convenient and rapid to achieve the preparation of MCBs in a large-scale industrial production. By tuning the carbonization temperature and reaction time, the microporosity, especially ultramicroporosity, of the resultant materials could be easily tailored to arise primarily from pores of size 0.4–0.9 nm, which greatly favors the postcombustion CO2 adsorption. The optimal sample of MCBs-9-5 carbonized at 900 °C for 5 h under air atmosphere exhibits a satisfactory CO2 uptake of 4.25 mmol g–1 at room temperature under 1 bar, which should be related to its high proportion of micropores of <1 nm, high narrow micropore surface area, and large micropore volume. More importantly, the MCBs-9-5 material also possesses a superior recyclability for CO2 capture and a high CO2 adsorption selectivity against N2 for CO2/N2 separation.

Experimental Section

Preparation of MCBs

The carbon bead precursor was synthesized by a convenient and controllable hydrothermal synthetic route using glucose as a carbon source.[23] Typically, 6 g of glucose was dissolved in 60 mL of deionized water to obtain a clear solution, and then the solution was transferred into a Teflon-sealed autoclave and maintained at 180 °C for 8 h. The puce products were obtained by filtration, washed repeatedly by deionized water, and dried at 80 °C for more than 8 h. The obtained glucose-based carbon bead precursors were defined as the CS precursor.

A 1 g mass of dried CS precursors was placed in a 10 mL corundum crucible with a cover and then directly transferred into a muffle furnace at high temperatures (700, 800, 900, and 1000 °C) using crucible tongs. The samples were calcined for regular times (1, 2, 3, 4, 5, and 6 h) and then were taken out to cool to room temperature for obtaining cross-linked MCBs. The resultant samples were denoted as MCBs-x-y, where x = 7, 8, 9, and 10, referring to the calcination temperature of 700, 800, 900, and 1000 °C, respectively; y = 1, 2, 3, 4, 5, and 6, representing calcination time of 1–6 h, respectively. In the synthesis process, the air, especially O2 in the air, can be used in the calcination process of material preparation.

For comparison, the CS precursor was directly carbonized at 900 °C for 5 h under a N2 atmosphere, which was designated as MCBs-9-5-r.

Characterizations

XRD patterns were monitored by a Bruker D8 diffractometer using Cu Kα radiation (λ = 0.15418 nm) as an X-ray source. Nitrogen adsorption–desorption isotherms were carried out at −196 °C using a micromeritics ASAP 2020 HD88 analyzer. Before adsorption, the samples were out-gassed at 200 °C for 10 h. The specific surface area (SBET) was evaluated using the BET method, and the hierarchical pore size distributions were calculated according to the DFT method. The micropore was analyzed using the t-plot method, and the ratio of micro-/mesopore was calculated according to the t-plot method. The micropore size distributions were also analyzed by fitting the CO2 adsorption isotherm at 0 °C using the DFT model. The morphology was observed from a FEI Tecnai G2 20 TEM with an accelerating voltage of 200 kV and a SEM (Quanta 250 FEG). FTIR spectroscopy spectra of a sample in KBr wafer were recorded on a Nicolet Avatar 370 spectrometer.

Gas Adsorption Tests

Gas adsorption isotherms of CO2 and N2 were measured using a Micromeritics ASAP 2020 HD88 instrument. Highly pure gases CO2 (99.999%) and N2 (99.999%) were employed for the measures. The isotherms of CO2 and N2 at 0 and 25 °C were conducted in an ice-water bath and a water bath, respectively. Prior to each gas uptake measurement, the samples were degassed at 200 °C for 10 h to remove the guest molecules from pores. For the regeneration experiment, the recovered adsorbents were evacuated at room temperature for 10 min and then reused for next adsorption.