Role of fiber density of amine functionalized dendritic fibrous nanosilica on CO₂ capture capacity and kinetics

Synthesis of DFNS-amine with varying fiber densities

Five types of DFNS with varying fiber density from dense to moderate to light were prepared and then evaluated the effect of fiber density on the CO₂ capture capacity. These five DFNS with different surface areas, fiber density, and pore volume (Fig. 1, Table 1) were synthesized by our previously reported procedure [40]. Fig. 1 shows SEM images, particle size distribution, and corresponding TEM images of five different DFNS. From DFNS-1 to DFNS-5, the fiber density decreased, and qualitative analyses were made only based on TEM and SEM images. As the fiber density decreases, the surface area and pore volume increase until DFNS-3; however, DFNS-4 has a lower surface area compared to DFNS-3 & 5, but with almost equal pore volume (Table 1). This was attributed to the increase in the pores size distribution in the range of 5–20 nm.

Fig. 1:

SEM images, pore size distribution, and TEM images of DFNS-1 (a, b, c, d), DFNS-2 (e, f, g, h), DFNS-3 (I, j, k, l), DFNS-4 (m, n, o, p), DFNS-5 (q, r, s, t). Some images are reproduced from our own Ref. [23].

These DFNS were then loaded with tetraethylenepentamine (TEPA) using the physisorption method [49]. The textural properties of five DFNS and their corresponding amine sorbents are summarized in Table 1. There was around 84 % decreased in the surface area and 75 % decrease in the pore volume of all DFNS after amine loading. The textural properties, such as N₂ sorption isotherm and BJH pore sizes distribution before and after amine loading, are shown in Fig. S1. The DFNS before and after amine loading showed a type 4 isotherm (Fig. S1). The BJH desorption pore size distribution of DFNS (Fig. S1) showed two wide pore size distributions, narrow pore size distribution around 2.5–5 nm, and another wide pore distribution around 5–25 nm. As the fiber density decreased, 0.5–5 nm pores reduced and pores around 5–25 increased (Fig. S1). After amine loading, pores were still visible; however, with reduced numbers (Fig. S1), all five DFNS showed around 17 mmol/g N contents (Table 1). The stability and morphology after amine loading were studied using SEM (Fig. S2) and found that the morphology of DFNS remains unchanged after amine loading.

Role of DFNS fiber density on CO₂ adsorption and desorption

DFNS-Amine sorbents were evaluated based on the effect of amine loading on textural properties, CO₂ capture capacity, kinetics, the effect of temperature on CO₂ capture capacity, and recyclability under dry and humid conditions.

All DFNS-TEPA_ads sorbents were analyzed for CO₂ capture capacity at 75 °C according to the program given in Fig. 2 using TGA. The sorbents were first activated by removing the adsorbed gases and moisture at 100 °C under the flow of N₂ gas (150 mL/min). After 60 min at 100 °C, the sample temperature was stabilized at 75 °C for 30 min in N₂. After that, the gas flow was changed to 15 % CO₂ in N₂ (150 mL/min) at 75 °C, and flow was maintained for 60 min. CO₂ capture capacity with time was determined by weight gain in TGA. The desorption of gas was carried out at 110 °C under the flow of 150 mL N₂ for 15 min. Three subsequent adsorption cycles for each sorbent were carried out and Table 2 depicts the average values of CO₂ capture from three adsorption-desorption cycles. Change in CO₂ capture capacities with change in fiber density of DFNS clearly indicated the key role of fiber density on CO₂ capture performance. As the fiber density decreased from DFNS-1 to DFNS-5, CO₂ capture capacity increased. The surface area and pore volume of DFNS-5 lie in between the DFNS-3 and DFNS-4 the CO₂ capture capacity of DFNS-4 was found to be in good agreement with fiber density. This was due to more space between the fibers (pore volume) available to accommodate more CO₂ molecules which increased CO₂ capture capacity (Table 2, Figs. S3 and S4). Also, more amines were available for CO₂ capture due to an increase in their accessibility due to the separation between fiber at less dense DFNS. CO₂ capture is diffusion limited, and as the diffusion of CO₂ molecules increases in less dense DFNS, it directly results in more CO₂ capture capacity.

Fig. 2:

The schematic of the CO₂ adsorption-desorption using TGA.

Role DFNS fiber density vs. surface area and pore volume in CO₂ capture performance

The difference in CO₂ capture capacity could be due to the difference in surface area or pore volume and not fiber density (Fig. 3). To study this aspect, DFNS-3, and DFNS-4 were further studied to understand the precise effect of fiber density on CO₂ capture capacity. Both DFNS-3 and DFNS-4 have similar N contents i.e., 17.1 mmol/g and there was around similar reduction in surface area and pore volume, i.e., both having similar surface area and pore volume. Hence these two sorbents are ideal for the understanding role of fiber density. To optimize the TEPA loading on the DFNS-3 and DFNS-4, a series of amine-loaded sorbents were prepared and evaluated for CO₂ capture performance (Table 3). The CO₂ capture capacity of the DFNS-3 and DFNS-4 series first increased with an increase in TEPA loading and then decreased after the optimal concentration of amine loading. In the case of the DFNS-3 series, DFNS-3 (14.7 mmol/g) was found to be better than the 17.1 and 12.4 mmol/g N contents, respectively (Table 3 and Fig. S5). Whereas, in the DFNS-4 series, the DFNS-4 (17.4 mmol/g) was found best sorbent. DFNS-3 (14.7) even with lower N content was found better in terms of CO₂ capture capacity than DFNS-4 (17.4) with higher N content. These results clearly indicated that a combination of amine loading with optimized fiber density is key to maximizing the adsorption capacity of DFNS-based solid amine sorbents.

Fig. 3:

Schematic diagram of DFNS with different fiber densities and effect on CO₂ capture capacity.

Why DFNS-3 (14.7) showed better CO ₂ capture capacity than DFNS-4 (17.4) even after low N contents? The reason is the better pore volume of DFNS-3; DFNS-3 with higher pore volume can accommodate more CO₂ molecules. Also, fiber density decides the diffusion time of CO₂ molecules within pores. To study this aspect, we measured the capture kinetics of all DFNS-TEPA sorbents (Fig. 4a). The difference in time required to achieve 90 % capture capacity of all five samples suggests the clear effect of fiber density on CO₂ capture capacity. From DFNS-1 to DFNS-5, the time to achieve 90 % CO₂ capture capacity decreased from DFNS-1 i.e., 10.38 min to DFNS-2 i.e., 6.57 min. In the case of DFNS-3 to DFNS-4, capture kinetics was found to be similar ∼6.4 min, while it decreased for DFNS-5 to 5.49 min. When we compared the two best sorbents from both series, DFNS-3 (14.7) and DFNS-4 (17.4) (Table 3, Figs. 4b and S6), even after low N contents and more CO₂ capture capacity in the case of DFNS-3 (14.7), it required nearly same time to achieve 90 % of the maximum CO₂ capture capacity, that of DFNS-4 (17.4). These results further confirm the key role of fiber density in controlling the diffusion of CO₂ molecules, as well as the accessibility of amines and hence in turn, CO₂ capture kinetics (Fig. 4b).

Fig. 4:

CO₂ capture performance. (a) Time required to achieve 90% of maximum capture capacity, (b) comparison between the two best sorbents DFNS-3 (14.7) and DFNS-4 (17.4).

Effect of adsorption temperature on CO₂ capture capacity

The adsorption temperature can affect both chemical adsorption and diffusion of CO₂ molecules. Hence, capture capacity and rate of adsorption can be affected by adsorption temperature. We evaluated the two best sorbents DFNS-3 (14.7), and DFNS-4 (17.4) for their CO₂ capture performance at four different temperatures. Both the sorbents showed an increase in CO₂ capture capacity with an increase in adsorption temperature and reached maximum CO₂ capture capacity at 75 °C (Figs. 5 and S7). The increase in CO₂ capture capacity can be attributed to increased temperature as it facilitated the chemical reaction between the amine and CO₂ molecules due to the exothermic nature of the reaction and as well as help in the diffusion of CO₂ molecule [13, 51]. With A further increase in adsorption temperature to 100 °C, the CO₂ capture capacity was decreased. The CO₂ adsorption and desorption processes are reversible at all temperatures. At temperatures up to 75 °C, the forward process is more favorable while at high temperatures (100 °C) the reverse reaction is more favorable, resulting into increase desorption of the adsorbed CO₂ [52].

Fig. 5:

CO₂ capture capacity at four different temperatures (a) DFNS-3 (14.7); (b) DFNS-4 (17.4). Effect of temperature on CO₂ capture capacity and adsorption lifetime (c) adsorption time of DFNS-3 (14.7), (d) adsorption time of DFNS-4 (17.4).

Kinetics of CO₂ adsorptions were still fast at all four temperatures and sorbents saturated within 5 min (Figs. 5, S8 and S9). However, as the adsorption temperature increased, the curve became more rounded around 6 min (Fig. 5) and the saturation time increased with an increase in the temperature. The increase of diffusion with an increase in the adsorption temperature can be seen by the adsorption kinetics for different adsorption capacities. For particular sorbents to calculate the time required to reach a particular adsorption capacity, we fixed two values 10.0 and 8.0 wt % to get an idea about how much time a sorbent took to reach at a fixed value at a controlled temperature. From these data, we observed that total time increased slightly as temperature increased from 30 to 50 °C but was constant for 75 °C and increased again at 100 °C (Fig. 5c and d). This is because at 100 °C, desorption of CO₂ molecules also started, and sorbents took more time to reach the maximum CO₂ capture capacity at 100 °C temperature.

DFNS-TEPA sorbents stability and recyclability

We also studied how long a sorbent DFNS-3 (14.7) can hold captured CO₂ at different temperatures; these experiments were carried out to find out the best storage/desorption temperature. We studied CO₂-saturated sample stability at three temperatures 30, 50, and 75 °C for 15 h without a flow of any gas. For these experiments, first, we captured the CO₂ at 75 °C and after 1 h exposure to CO₂, we hold samples at different temperatures (30, 50, and 75 °C) to check the thermal stability of amine-CO₂ adduct for 24 h. We maintained the temperature for 24 h without a flow of any gas. As the holding temperature increased, desorption kinetics increased (Fig. 6). At 30 °C, the loss of CO₂ was around 3.8 % in 23.4 h; however, as the temperature increased from 30 to 100 °C the rate of CO₂ desorption increased. It was 13.1 % in 22.3 h at 50 °C, 100 % loss in 4.8 h at 75 °C, and 100 % in 3.3 h at 100 °C.

Fig. 6:

Desorption study of sorbent DFNS-3 (14.7) at different temperatures (a) 30 °C, (b) 50 °C, (c) 75 °C, and (d) 100 °C.

Good sorbents must show stability under moist conditions. For this study, 21 CO₂ adsorption-desorption cycles were conducted on DFNS-3 (14.7) and DFNS-4 (17.4) under dry and humid conditions (Fig. 7). For each step, the adsorption of CO₂ gas at 75 °C followed by the desorption of CO₂ gas at 110 °C under N₂ flow. During the recyclability study, both sorbents showed an increase in CO₂ capture capacity after 1st cycle. DFNS-3 (14.7) showed a CO₂ capture capacity increase from 18.8 to 23.4 wt % from the first to the second cycle, which was maintained up to 10th cycle under dry conditions. After 10th cycle of CO₂ capture, there was a slow reduction in CO₂ capture capacity, and in the 21st cycle, the DFNS-3 (14.7) still showed better CO₂ capture capacity than its first cycle. DFNS-4 (17.4) (Fig. 7a) also showed similar behavior in the recyclability study.

Fig. 7:

Sorbent recyclability under (a) dry, (b) humid conditions. (c) Total CO₂ capture capacity in 21 cycles under dry and humid conditions at 75 °C using 15 % CO₂ in N₂.

However, DFNS-4 (17.4) showed more reduction in CO₂ capture capacity compared to DFNS-3 (14.7) in 21 cycles. Nevertheless, after 21 cycles under dry conditions, DFNS-4 (17.4) still showed CO₂ capture capacity similar to DFNS-3 (14.7). There is always a loss of amines during every adsorption-desorption cycle, but even with a decrease in N contents, the sorbents still showed similar CO₂ capture capacity to its first cycle. This indicates excellent amine accessibility and diffusion of CO₂ molecules in the DFNS-TEPA sorbents. With the number of adsorption-desorption cycles, amine accessibility increase, and diffusion of CO₂ molecules also increase, which lead to maintaining capture capacity for a long time. These results were also indicating about the role of fiber density in the accessibility of amines and efficient diffusion of CO₂ molecules.

The recyclability of sorbents under humid conditions is the real parameter to mimic the flue gas condition. The effect of humidity on CO₂ capture was studied using the pre-humidified CO₂ gas. The humid CO₂ was generated at room temperature by passing CO₂ gas through the two water bubblers. There was approximately 1 wt % increase in CO₂ capture capacity. Notably, under humid conditions, DFNS-3 (14.7) showed much better recyclability than DFNS-4 (17.4) (Fig. 7b). For a comparison under humid conditions, DFNS-3 (14.7) has 18.6 wt % in the 1st cycle and 20.8 in 21st cycle, indicating no decrease in CO₂ capture capacity in 21 cycles. DFNS-4 (17.4) showed a decrease in CO₂ capture capacity, which was 17.4 in 1st cycle and 15.9 wt % in the 21st cycle, a 8.6 % decrease. We observed DFNS-3 (14.7) was more stable and recyclable as compared to DFNS-4 (17.4). From recyclability data, it was revealed that fiber density not only affects the CO₂ capture capacity but also affects the stability of sorbents during the adsorption-desorption steps.

We also calculated the total CO₂ capture capacity in 21 cycles under dry and humid conditions. We found that DFNS-3 (14.7) captured 108.5 and 105 mmol/g (Fig. 7c, Tables S1 and S2) total CO₂ in 21 cycles under humid and dry conditions, respectively. DFNS-4 (17.4) captured 90.6 and 88.2 mmol/g (Fig. 7c, Tables S1 and S2) of total CO₂ under dry and humid conditions, respectively. Their result further confirmed the fiber density effect of DFNS on CO₂ capture capacity.

After the recyclability study, both sorbents were then re-characterized to estimate amine loss after 21 cycles during humid and dry conditions (Table 4 and S10). DFNS-3 (14.7) showed 4 mmol/g nitrogen loss in 21 cycles, whereas DFNS-4 (17.4) showed 3 mmol/g N loss. This further indicates the role of fiber density in stabilizing the amine molecules. The selectivity of the sorbents toward the different gases is also a very important parameter in deciding the industrial application of sorbents for CO₂ capture. The selectivity study was carried out under similar conditions of CO₂ capture in TGA. Both sorbents showed 100 % selectivity for CO₂ over N₂ and O₂ (Fig. S11). Overall comparisons are summarized in Table 5.