CO2 capture-driven thermal and electric energy harvesting using amine-functionalized solvents with a poly(amidoamine) impregnated membrane

doi:10.21203/rs.3.rs-2119257/v1

Download PDF

Article

CO₂ capture-driven thermal and electric energy harvesting using amine-functionalized solvents with a poly(amidoamine) impregnated membrane

https://doi.org/10.21203/rs.3.rs-2119257/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The energy consumed for CO₂ capture in processes that capture, store, and utilize carbon frequently approaches 80% of the total energy consumed, and its reduction is a major area of focus. Herein, a system that captures CO₂ from flue gas to produce electricity is proposed. The exothermic reaction heat of amine-CO₂ in the point-source CO₂ capture process is harvested and converted into electricity using a thermoelectric (TE) device. The reaction heat of synthesized amine-functionalized mixtures (TEPA@SBA/MEA+PZ) is 52% higher compared with that obtained from the primary amine absorbent. The flue gas channel and solvent reservoir are separated using a membrane composed of tetraethylenepentamine-impregnated poly (ethylene glycol) dimethacrylate that selectively permeates CO₂, to minimize heat loss due to mass transfer. The CO₂ working capacity of TEPA@SBA/MEA+PZ is 6.6 mmol/g, which is defined as the capture capacity under steady-state conditions of 15 mol% CO₂ and 85 mol% N₂. Simultaneously, electricity can be harvested at a rate of 7.5 kJ/kg with a TE device composed of Bi₂Te₃. When put to industrial use, the proposed system can reduce the CO₂ emission of power plants from 0.8 ton/MWh to 0.25 ton/MWh, while simultaneously generating an additional 0.6% of the total electricity production.

In carbon capture, utilization, and storage processes, 80% of the energy is consumed by CO₂ capture. Thus, minimizing this energy consumption is of paramount importance¹. CO₂ capture system for power plants, which is primary sources of CO₂ emissions, can be divided into pre- and post-combustion capture systems based on the applied stage and conditions. In the pre-combustion CO₂ capture process, CO₂ is separated from syngas before being fed into the turbine. The gas contains 40 mol% CO₂ and is under ambient temperature (10–30 ^oC) and high pressure (> 40 bar). Considering this, a physical absorbent that absorbs CO₂ by virtue of the local concentration difference is utilized². Although the CO₂ selectivity of a physical absorbent is low, CO₂ separation is possible because the partial pressure of CO₂ in the flue gas is high, and the energy required for regeneration is small (when the temperature approaches the boiling point or the pressure is lowered, CO₂ is separated according to Henry’s law). The energy consumed by the physical absorption process using MeOH as the working fluid was 1.88 MJ/kg-CO₂³. The post-combustion capture process removes CO₂ after the gas is allowed to expand in the turbine. Flue gas contains 15 mol% CO₂ at room temperature (10–30 ^oC) and ambient pressure (1 bar)⁴. An amine-based chemical absorbent is used to capture CO₂ through a chemical reaction between the CO₂ molecule and solvent. Despite high selectivity at low CO₂ concentration, energy consumption approaches 3.8 MJ/kg-CO₂ because the chemical bond formed in the reaction must be broken during the regeneration process⁵.

Regarding this, the current methods are focused on the reduction of heat source temperature and energy consumption in the post-combustion capture process. The catalyst-aided amine regeneration was performed to reduce the regeneration energy load utilizing structure activity relations of zeolite catalyst for tri-blended monoethanolamine (MEA)/N-methyl diethanolamine (MDEA)/piperazine (PZ)⁶. The several catalysts has been synthesized including metal oxide⁷, Ag₂O-Ag₂CO₃⁸, metal modified metal-organic frameworks⁹, Ce(SO₄)2/ZrO₂¹⁰, Al₂O₃/HZSM5¹¹, SO₄₂/ZRO₂/SBA-15¹². Although the reduction in regeneration energy consumption presents 10–50% compared to the conventional system, their applications are limited in that the absorption kinetics decreases proportionally. The design and optimization of post-combustion systems are being discussed¹³. Hosseini-Ardali et al.¹⁴ optimized CO₂ capture system using MDEA and PZ bi-solvent as working fluid, and the regeneration energy consumption decreases to 2.68–2.76 MJ/kg-CO₂. The secondary amine-based system designed to replace monoethanolamine has a low CO₂ selectivity, and the system components become large. The system design that breaks away from the existing post-combustion CO₂ capture process has been attempted. Liu et al.¹⁵ reported that the highly polar branched polymers is an effective strategy for selective separation of CO₂ from the flue gas. Sutherland¹⁶ proposed the efficient carbon capture system based on magnetic induction swing adsorption in the sorption bed. Other studies have proposed dual function materials for on CO₂ capture and direct conversion. However, the feasibility of these methods is low under extreme driving conditions^{16, 17}.

In this study, a system that selectively captures CO₂ from flue gas containing 15 mol% CO₂ (post-combustion condition) and simultaneously harvests electrical energy is proposed. Previous studies have only attempted to minimize energy consumption of such systems. However, the present study seeks to alter the paradigm of gas capture applications by combining gas separation and energy harvesting. When applied to industrial energy power plants, zero-emission of CO₂ can be realized to reduce the adverse environmental impact of power plants. Furthermore, electricity is produced to obtain surplus energy.

Scheme 1 presents the industrial application of a point-source CO₂ capture-driven electric energy harvesting system. The batch cycle comprises two reactors that play the same role. Reactor 1 contained an unsaturated solvent that selectively captured CO₂ when flue gas was supplied. The exothermic reaction heat of amine-CO₂ was harvested as thermal energy, and the temperature of the reactor was maintained at 50–70 ^oC. It was converted into electricity using a thermoelectric (TE) device. In contrast, reactor 2 contained a saturated solvent, and external heat was supplied to break the bond of amine-CO₂. The CO₂ gas was regenerated and transported to the storage tank. Finally, the solvent in reactor 1 was saturated and that in reactor 2 was in an unsaturated state. Following this, the direction of flow was changed using the valve; flue gas flowed to reactor 2, and thermal energy was supplied to reactor 1 for regeneration. Thus, sustainable point-source CO₂ capture and electrical energy production are possible through a series of batch cycles. In the conventional post-combustion process, flue gas and liquid solvent flow in a counter-current direction in the tray column absorber, during which CO₂ is selectively collected. The saturated solvent is supplied to the desorber with a pump for regeneration, and returns to the absorber. In contrast, the present system is configured as a batch cycle to minimize the absorption heat loss by the convective flow of absorbent, which increases the thermal energy harvesting efficiency.

2.1. Strategy

Scheme 2a presents the reactor used for point-source CO₂ capture and electric energy harvesting. The inlet gas is composed of 15 mol% CO₂ and 85 mol% N₂, which is the same condition as the post-combustion capture process. The gas flows through the channel inside the reactor, and the CO₂ is selectively permeated by the surrounding membrane. The clean gas without CO₂ is discharged into the atmosphere. As CO₂ reacts with primary/secondary amines-functionalized solvents, thermal energy is harvested and converted into electricity by a TE device.

In CO₂ capture systems, if the absorbent reservoir and gas channels are not separated by the CO₂ membrane, the gas bubble and absorbent contact directly and generate a relative velocity. The consequent heat loss owing to forced convection is significant. In addition, the effective interfacial area is also enlarged. However, in the proposed CO₂ capture-driven electric energy harvesting system, the CO₂ membrane is inserted between the gas channel and absorbent reservoir. Thus, the heat loss can be minimized by increasing the thermal resistance. Specifically, the overall thermal resistance includes the natural convection of the absorbent, conduction inside the membrane, and forced convection of the membrane gas. Furthermore, the membrane is composed of a material with high CO₂ permeability which ensures that the CO₂ diffusion performance remains unaffected as presented in Scheme 2b.

2.2. Materials

Functionalized solvents have been developed to improve CO₂ capture performance (working capacity and kinetics) and thermal energy harvesting density, considering the conditions of flue gas. MEA, a primary amine absorbent, exhibits high CO₂ selectivity and large exothermic heat. However, the CO₂ capture capacity of MEA is low because of the bicarbonate reaction. Thus, it is supplemented with piperazine (PZ), a secondary amine absorbent. In the proposed system, an aqueous solution of MEA and PZ was prepared for use as the base fluid of the functionalized solvent. Tetraethylenepentamine (TEPA), which contains both primary and secondary amines, can promote thermal energy harvesting by inducing chain reactions. For this purpose, TEPA was functionalized on the surface of the mesoporous silica template and dispersed in the base fluid.

Santa Barbara Amorphous-15 (SBA-15), a mesoporous silica, was synthesized using hydrothermal self-assembly characteristics. As a manufacturing method, the pH was maintained at 2.5 by mixing deionized water and hydrochloride. 4 wt% Pluronic P123 copolymer with a molecular weight of 5,800 was added to it, and the solution was stirred for 24 h at 40 ^oC and 400 rpm. Subsequently, tetraethylorthosilicate was added and the mixture was stirred at 40 ^oC for 24 h. When the mixture was heated at 110 ^oC in N₂ gas environment for 24 h, silica growth was induced along the longitudinal direction with pores and a hexagonal cross section. After filtration with methanol and ethanol, vacuum was maintained to obtain solid-state SBA-15. SBA-15 was dispersed in methanol and 58 wt% TEPA (mass ratio of TEPA to SBA-15) was added. The successful functionalization of TEPA was ensured by stirring for 24 h in a N₂ gas environment, and centrifugation with methanol, ethanol, and water. Finally, solid state TEPA@SBA-15 was obtained after heating at 50 ^oC for 12 h.

The base fluid was a solution of 16.7 wt% PZ and 83.3 wt% MEA, which was stirred at 40 ^oC to fully dissolve PZ. After adding 2.0 wt% TEPA@SBA-15 to the base fluid, ultrasonication was performed for 30 min at 750 W and 20 kHz to prepare TEPA@SBA/MEA + PZ. Figure 1 presents the composition of TEPA@SBA/MEA + PZ and the CO₂ capture mechanism. The primary amine of MEA captures CO₂ and theoretically generates heat at a rate of 81 kJ/mol-CO₂. PZ forms the secondary amine-CO₂ bonds. TEPA@SBA-15 contains both primary and secondary amines, which can be used to induce additional reactions; for instance, the liquid reacts inside the bulk, and TEPA@SBA-15 reacts on the surface of the mesopores. The thermal energy production per unit mass of CO₂ was improved by the chain reaction.

Poly(ethylene glycol) dimethacrylate (PEGDMA) with molecular weight of 750, TEPA, and EtOH were stirred at a mass ratio of 6:4:5 for 24 hours. The required structure of membrane can be formed by adding 1-hydroxybenzotriazole (photoinitiator) (HCPK) to PEGDMA in a molar ratio of 1:60 and irradiating it with UV light with a wavelength of 345 nm¹⁸. As shown in Fig. 2, PEGDMA and HCPK are converted into a solid-state cross-linked resin, when exposed to UV light. Moreover, before the structure is formed, a bonding group of N⁺H₃—OCHN can be obtained between cross-links as TEPA is added. Subsequently, the CO₂ molecules in the flue gas migrate to the amine-rich phase owing to the difference in solubility. This is a transport mechanism that forms carbonate ion pairs of primary and secondary amines.

2.3. Characterization

Solid adsorbent (amine-functionalized mesoporous silica): Field emission scanning electron microscopy (FE-SEM) was performed using a Quanta FEG 240 from FEI. The sample was pretreated with a Pt target at 20 mA for 60 s. A Tecnai F20 G2 instrument from FEI was used for transmission electron microscope (TEM) measurements. Ultrasonication was conducted for 30 min at 20 kHz and 750 W to disperse the functionalized mesoporous silica in EtOH. The accelerating voltage, image resolution, and magnification was 150 kV, 0.23 nm, and 1,030,000, respectively. The bonding groups on the surface of the functionalized silica were evaluated by X-ray photoelectron spectroscopy (XPS) using a ULVAC-PHI's PHI 5000 VersaProbe that had a minimal spatial resolution of 10 µm, and resolution energy of 0.48 eV, and a monochromatized anode composed of Al Kα. The nitrogen adsorption curve was evaluated using a Micrometrics Tristar 3020. The bath temperature was − 195.9 ^oC; the equilibrium interval was 20 s; and the surface area was evaluated using the Brunauer–Emmett–Teller (BET) theory. The pore size distribution ranged from 1.7–300 nm, the adsorbate property factor was 0.953 nm, and the density conversion factor was 0.0015. Using the same equipment, the adsorption performance of CO₂ gas was evaluated and the adsorption heat in the isotherm was estimated. The experiment was conducted while injecting CO₂ at a temperature of 25 ^oC and a pressure of 10^− 4 kPa. Thermogravimetric analysis (TGA) of the CO₂ membranes was performed using a TGA Q500 instrument (TA Instruments), for which the temperature was raised to 530.00 ^oC at increments of 10.00 ^oC/min.

Liquid absorbent (functionalized solvent): Thermal conductivity was measured by a Flucon GmBH LAMBDA system using the transient hot-wire method. Viscosity was evaluated using A&D's SV-10 using the vibration method. The characteristics of functionalized solvents during the point-source CO₂ capture process were evaluated using the device shown in Figure S1. The inlet gas was composed of 15 mol% CO₂ and 85 mol% N₂ and was maintained at 25 ^oC and 1 bar. The mass flow rate of the gas was fixed at 0.01777 g/s. In the reactor, 200 g of the functionalized solvent selectively captured CO₂ from the flue gas and harvested thermal energy. The gas from which CO₂ was removed was discharged through a purifier. The composition ratio of the mixed gas was verified by AIRKOREA Co. (LTC), and the composition of the outlet gas was measured using a gas meter. The CO₂ capture capacity was calculated using Eq. (1), where m is mass (kg). The subscripts a, i, and o denote absorption, inlet, and outlet, respectively.

$${m}_{a, CO2}= \sum {\dot{m}}_{i, CO2}- \sum {\dot{m}}_{o, CO2}.$$

The flow rates of the mixed gas at the inlet and outlet are given by Eqs. (2) and (3), respectively, where ρ is the density (kg/m³), and $\dot{Q}$ is the volumetric flow rate (m³/s).

$${\dot{m}}_{i}= {\rho }_{CO2,}\bullet {\dot{Q}}_{CO2, i}+ {\rho }_{N2, i}\bullet {\dot{Q}}_{N2, i}$$

$${\dot{m}}_{o}= {\rho }_{CO2,}\bullet {\dot{Q}}_{CO2, o}+ {\rho }_{N2, o}\bullet {\dot{Q}}_{N2, o}$$

The thermal energy harvesting performance was characterized during the point-source CO₂ capture process, as shown in Figure S1, for which the reactor was set as a control volume. The energy conservation law was applied to the control volume, which is given by Eq. (4), where E is the total energy (J), Q is the heat transfer (J), W is the work (J), h is the specific enthalpy (J/kg), v is the velocity (m/s), g is the gravitational acceleration (m/s²), and z represents the height (m). The reactor was insulated, and $\dot{Q}$ was negligible. $\dot{W}$ was not considered as the volume remained constant.

$$\frac{dE}{dt}=\dot{Q} -\dot{W}+{\dot{m}}_{i}\left({h}_{i}+\frac{1}{2}{v}_{i}^{2}+g{\bullet z}_{i}\right)- {\dot{m}}_{o}\left({h}_{o}+\frac{1}{2}{v}_{o}^{2}+g{\bullet z}_{o}\right)+{\dot{Q}}_{g}$$

Eq. (5) can be obtained by integrating all the terms with respect to time (t), and the total energy can be reduced to the internal energy (U) change. ΔU was calculated as the sensible energy change in terms of the temperature change of the materials contained in the reactor. The energy change due to the mass transfer of gas at the inlet and outlet was evaluated based on the state (temperature and pressure) information obtained by the measuring instrument.

$$\varDelta U - {m}_{i}\left({h}_{i}+\frac{1}{2}*{v}_{i}^{2}+{g}_{z, i}\right)+ {m}_{o}\left({h}_{o}+\frac{1}{2}*{v}_{o}^{2}+{g}_{z, o}\right)={Q}_{g}$$

The thermal energy generated by the CO₂ capture reaction (Q_g) is given by Eq. (6), where q_g (J/kg-CO₂) is the specific reaction heat.

$${Q}_{g}={m}_{a, CO2}*{q}_{g}$$

Note that the specific reaction heat is assumed to be constant regardless of the CO₂ capture amount because the temperature increases linearly during the point-source CO₂ capture process. The design of the reactor ensures that the Biot number, i.e., the ratio of convection heat transfer within the fluid to conduction heat transfer within the solid, approaches 0.1, which minimizes the spatial temperature distribution.

2.4. Experiments

Figure 3a shows the experimental apparatus for point-source CO₂ capture and electric energy harvesting. A mixture of 15 mol% CO₂ and 85 mol% N₂ were supplied to the reactor at 20 ^oC and 3 bar. The flow rate of the inlet gas was varied in the range of 6.25–10.5∙10^− 5 kg/s. CO₂ was captured inside the reactor and the components of the outlet gas were measured using a gas meter. The reaction heat generated in this process was harvested and converted into electricity using the TE devices. The data reduction method for CO₂ capture was identical to that used for the process of characterization shown in Figure S1. The thermal energy harvesting performance within the reactor was separately evaluated considering that the TE devices were not attached to it, and that the exterior of the reactor was insulated. The thermal energy harvesting density (TEHD; kJ/kg) was calculated by dividing the total amount of energy produced by the mass of the solvent, as presented in Eq. (7). The Biot number of the membrane reactor reached 0.1, thereby enabling lumped system modeling. The temperature inside the reactor was measured over three parts: the membrane surface, middle of the solvent reservoir, and outer surface of the reactor.

$$TEHD=\frac{\varDelta E}{{m}_{solvent}}$$

The same experiment was repeated thrice to obtain an average value, and the standard deviation was calculated.

The electric energy harvesting performance was estimated by attaching the TE devices to the outer wall of the reactor. The TE devices, commercially manufactured by PCCTech, were composed of Bi₂Te₃, considering the working temperature range and stability. During the CO₂ capture process, the temperature of the hot side rose continuously owing to the heat of the reaction, and the temperature of the cold side was maintained constant by forced convection cooling provided by the attached fan. The durability of the materials was evaluated using cyclic absorption/desorption experiments.

Figure 3b presents the geometry of the reactor. The CO₂ membrane was attached to the frame structure in the form of a film, and the gas channel and absorbent reservoir were separated. Figure 3c shows the manufacturing method of the CO₂ membrane. TEPA-impregnated PEGDMA was cured on the frame of an aluminum plate. This process was repeated until the thickness of the membrane reached 8 µm.

2.5. Theoretical analysis

The thermal energy harvesting performance of the membrane reactor shown in Fig. 3b can be calculated numerically. The functionalized solvent and solid reactor were set as the boundary, and the energy conservation law expressed as Eq. (4) was applied. The work done is negligible, considering that the volume remained constant. However, the heat and mass transfer between the gas channel and solvent reservoir through the membrane are included, as summarized in Eq. (8).

$$\varDelta E= Q+{m}_{i, CO2}{\bullet h}_{CO2} +{Q}_{g}.$$

Finally, the temperature inside the reactor was estimated using Eqs. (9), where T is the temperature, and subscript i denotes the i^th iteration. where C denotes the specific heat capacity.

$${T}_{i+1}={T}_{i}+\frac{Q+ \varDelta {m}_{CO2}\bullet {h}_{CO2}+{Q}_{g}}{{m}_{tot}\bullet {C}_{tot}}$$

The thermal energy generation (Q_g) due to the amine-CO₂ reaction was considered using the data obtained from the characterization. The heat transfer rate was calculated based on the thermal resistances, including the forced convective heat transfer of gas-solid (flue gas – membrane), conduction heat transfer inside the solid, and natural convective heat transfer of the solid-liquid (membrane – solvent). The heat transfer coefficient was obtained using the Nusselt number correlation and the measured thermal conductivity. The permeability of CO₂ with respect to the membrane was 7.5∙10–12 m³/(m²∙s∙Pa)¹⁸.

3.1. Characterization

Amine-functionalized SBA-15 was synthesized to improve the reaction heat between CO₂ and the solvent. Because the template is mesoporous silica, the pore structure can be stably maintained under various conditions of temperature and pressure. In contrast, metal organic frameworks (MOFs) or covalent organic frameworks (COFs) form Armstrong-scale pores through a lattice structure. However, MOF and COF templates are unsuitable for amine functionalization, considering that the theoretical diameter of the amine groups is 0.935 nm. Moreover, they are vulnerable to moisture and cannot remain dispersion in aqueous amine solutions. An FE-SEM image of TEPA@SBA-15 is shown in Fig. 4a. The diameter is distributed in the range of 110–250 nm, the length is 500–750 nm, and the aspect ratio is large owing to pore structure growth during the self-assembly process. The structure of TEPA@SBA-15 dispersed in EtOH was studied using the TEM image as presented in Fig. 4b. The pore dimeters are distributed in the range of 3.7–5.8 nm. The cross-section is hexagonal, and the dimensions of SBA-15 observed by TEM and SEM are similar.

The bonding groups on the surface of SBA-15 were analyzed using XPS, as shown in Figure S2. It is mainly composed of silicon and oxygen, corresponding to amorphous silica; nitrogen is not detected. As shown in Figure S3, which presents the XPS spectrum of TEPA@SBA-15, the component ratio of silicon and oxygen is similar to that of SBA-15, and a large amount of nitrogen is observed with functionalized TEPA. Specifically, TEPA@SBA-15 contains 6.7% nitrogen, as shown in Table S1. The nitrogen content is proportional to the TEPA functionalization of NH₂-(CH₂)₂-[NH-(CH₂)₂]_n-NH₂, and it is significantly correlated with the CO₂ reaction kinetics. Figure S4a shows the N₂ adsorption curve of mesoporous silica. The surface area of SBA-15 reaches 619 m²/g, indicating the distributed presence of mesopores. As amine groups are synthesized in the pores, the surface of TEPA@SBA-15 contracts to 188 m²/g. As shown in Figure S4b, the average pore size of SBA-15 is 5.68 nm. TEPA@SBA-15 has a pore size of 5.54 nm, in addition to a significantly reduced peak value.

Figure S5 shows the reaction heat versus the CO₂ capture amount. The initial adsorption heat of SBA-15 was 37.1 kJ/mol-CO₂, which is consistent with the values reported in related literature¹⁹. This corresponds to physisorption, as the heat tends to decrease continuously in terms of the CO₂ capture amount. In the case of TEPA@SBA-15, the adsorption heat increases with CO₂ capture and presents a peak of 78.5 kJ/mol-CO₂. Gas adsorption is initially induced at the region with high surface energy. Following that, the reaction heat increases due to the chemisorption of amine-CO₂. The CO₂ capture capacity of TEPA@SBA-15 is improved by approximately 70% compared with that of SBA-15, as shown in Fig. 4c.

Table 1 compares TEPA@SBA-15 with the CO₂ adsorbents reported in related literature^{20, 21, 22, 23, 24, 25, 26, 27, 28}. Although certain adsorbents have a larger CO₂ adsorption capacity than TEPA@SBA-15^{21, 22, 23, 24, 25, 26, 27, 28}, the adsorption heat of TEPA@SBA-15 is significantly higher under similar conditions. Furthermore, considering that a large amount of thermal energy is harvested from a small amount of CO₂, the electric energy production can be improved. The CO₂ adsorption capacity of TEPA@SBA-15 is not an important parameter as the CO₂ capture capacity of TEPA@SBA/MEA + PZ is dominantly affected by the base fluid, while the adsorbents promote the exothermic reaction heat. The desorption of TEPA@SBA-15 can be performed using heat sources above 100 ^oC, which is identical to base fluids of MEA and PZ.

Figure 4d shows the specific reaction heats of the solvents. The reaction heat of MEA obtained in the experiment was 81.9 kJ/mol-CO₂, which deviates an error margin of 2.5% from the theoretical value of 84.0 kJ/mol-CO₂. The theoretical CO₂ capture capacity of MEA is 8.2 mmol-CO₂/g-solvent (0.5 mol-CO₂/mol-solvent) considering the bicarbonate reaction. However, the capture amount of MEA is 6.05 mmol/g in a fixed bed reactor, as presented in Fig. 4e. This can be attributed to the fact that CO₂ selectivity of the solvent is inversely proportional to the CO₂ concentration of flue gas. MEA + PZ induces a chain reaction between the primary and secondary amines, which increases the CO₂ capture capacity and reaction heat. In addition, TEPA@SBA-15 adsorbs CO₂ chemically on its surface. Finally, TEPA@SBA/MEA + PZ generates a reaction heat that is 52% higher than that generated by the primary amine. However, the CO₂ adsorption capacity of TEPA@SBA-15 is smaller than that of the base fluid and does not affect the capture capacity of the entire solvent.

Figure 4f shows the thermal resistance with respect to the reactor shape. The overall thermal resistance of a fixed bed reactor can be estimated by the forced convective heat transfer between the gas (gas mixture) and liquid (solvent). The bubble volume and relative velocity are obtained depending on the porous nozzle diameter, and the heat transfer coefficient is evaluated based on the Nusselt number correlation. In the case of the membrane reactor, the natural convective heat transfer coefficient is obtained by considering the length scale and thermal conductivity of the solvent. In addition, the forced convective heat transfer coefficient is estimated for the gas channel. A comparison between the two reactor shapes show that the overall thermal resistance of the membrane reactor is 3.5 times higher than that of the fixed bed reactor. Specifically, CO₂ membrane (thermal barrier) is extremely thin (8 µm), resulting in negligible thermal conduction. However, separating the gas channel and absorbent reservoir significantly increases the thermal resistance. Figure 5a presents the TGA of the CO₂ membrane, and the values of 0.1, 0.3, 0.5 represent the weight fraction of the impregnated TEPA. Weight loss occurs at 70–72 ^oC regardless of the composition, because EtOH was included for uniform dispersion of TEPA. Evaporation of EtOH is independent of thermal stability of the membrane, and it can sufficiently maintain a stable structure at 20–150 ^oC, which is the operating range of CO₂ absorption/regeneration. In Fig. 5b, the membrane can be manufactured in the form of a thin film. CO₂ molecules can easily permeate the thin membrane by the solubility difference of amine-phase due to TEPA-impregnated structure. Moreover, the cross-linked resin structure of PEGDMA functions as a thermal barrier between gas and liquid phases by maintaining its solid state. CO₂ selectivity and thermal energy harvesting performance can be controlled by varying the thickness of the membrane layer.

3.2. Point-source CO₂ capture and thermal energy harvesting

The CO₂ capture and thermal energy harvesting performance were evaluated for a CO₂ membrane reactor (TEPA@PEGDMA) using a functionalized solvent (TEPA@SBA/MEA + PZ). Figure 6 shows the parameter variations over time during the CO₂ capture process. As the inlet pressure was fixed at 3 bar in all cases, CO₂ penetrates the membrane with a partial pressure difference and induces a CO₂-amine reaction. Figure 6a shows that the initial pressure of the solvent reservoir was 1.75 bar, and the driving force is 7.5 kN, which is sufficient for the permeation of CO₂. Here, the driving force is calculated by considering the pressure difference and surface area of the membrane. The steady section is defined as the working region (it is marked in Fig. 6(a)). The amount of CO₂ captured in the working region is calculated as the working capacity, as presented in Fig. 6b. As the solvent begins to saturate, the pressure in the reservoir increases and the absorption rate decreases. As the inlet pressure increases to a value greater than 3 bar, the CO₂ capture kinetics and capacity remain unaffected. This is because the permeation amount of CO₂ does not depend on the partial pressure difference, and the CO₂ selectivity of the functionalized solvent is decided by the temperature. However, it must be noted that a stable working region can only be obtained when the minimum inlet pressure is maintained.

Figure 6c presents the temperature variation during the point-source CO₂ capture process, for which the measurement locations are marked. The temperatures at the exterior of the reactor and solvent reservoir are similar; the percentage of error between the two values is smaller than 0.2%, indicating that lumped system modeling is reasonable. Fluctuation of the solvent temperature occurs as natural convective motion is promoted by the temperature rise. The average temperature of the membrane surface is 4.5% higher than that of the solvent. In addition, the temperature variation in the longitudinal direction is negligible as CO₂ gas is uniformly permeated by the same driving force, and the natural convective motion due to density difference maintains thermal equilibrium within the absorbent reservoir. Furthermore, the inlet gas temperature is maintained at 20 ^oC. In contrast, the temperature of the outlet gas gradually increases, as a result of the heat loss owing to mass transfer. For MEA in the membrane reactor, the maximum temperature reaches 54.2 ^oC at an inlet gas flow rate of 5.0 10^− 5 kg/s. This is 28.4% higher than the maximum temperature of the MEA (42.2 ^oC) in the fixed bed reactor.

Figure 7a summarizes the working capacity of the point-source CO₂ capture process. Although the working capacity of the membrane reactor is lower than that of the fixed bed reactor (CO₂ capture capacity of materials in Fig. 4e), the performance trends are similar in terms of materials. Specifically, TEPA@SBA/MEA + PZ has a higher working capacity than MEA. The CO₂ working capacity remains unaffected by the mass flow rate of the inlet gas. Furthermore, the large fluctuation in terms of the time step in the transient analysis can be attributed to the pressure instability caused by membrane penetration. However, the error in reproducibility of the performance remains within 5%, as demonstrated by the results of repeated experiments.

Figure 7b shows the variation in the maximum temperature during the CO₂ capture process. The maximum temperature increases in proportion to the mass flow rate of the flue gas because the heat loss decreases as the saturation time is shortened. The forced convective heat transfer increases with the flow rate of the inlet gas, and the thermal resistance of the gas channel decreases. However, the effect of the gas channel is small because the thermal resistance of the solvent reservoir (natural convective heat transfer) is the dominant factor impacting the overall thermal resistance. As the flow rate increases, the heat loss over time is not significantly affected, and the reaction time is shortened to maximize the harvesting of thermal energy. The maximum temperature of TEPA@SBA/MEA + PZ reaches 67 ^oC at a gas flow rate of 8.33 10^− 5 kg/s; the corresponding temperature rise in the fixed bed reactor is 47.5 ^oC. Figure 7c shows the TEHD calculated based on the temperature variation; the TEHD of TEPA@SBA/MEA + PZ is 15% higher than that of MEA. The TEHD of TEPA@SBA/MEA + PZ in the fixed bed reactor is approximately 200 kJ/kg. In the case of the membrane reactor, the TEHD is 445 kJ/kg at an inlet mass flow rate of 8.33 10^− 5 kg/s, which indicates that heat loss was reduced by 245 kJ/kg. As shown in Fig. 7d, the CO₂ capture performance of MEA and TEPA@SBA/MEA + PZ remains constant even after repeated absorption/desorption cycles.

3.3. Theoretical performance analysis

The numerical analysis was performed to analyze the working mechanisms of CO₂ capture and thermal energy harvesting process in the membrane reactor of TEPA@PEGDMA. Furthermore, if the CO₂ capture-driven energy harvesting process can be predicted, the system can be optimized through the parametric analysis. In Fig. 8a, the results of the temperature variation of MEA + PZ deviate from the experimental results by an error of 0.3%. As the CO₂ capture kinetics in the working region are steady, the performance is predicted by considering the material characteristics. However, after that, the numerical value deviates from the experimental results. Thus, the data was collected only up to the temperature peak, and the subsequent section were not considered.

Figure 8b shows that TEHD increases in proportion to the mass flow rate as heat loss decreases. The effect of the mass flow rate can be discussed based on the heat transfer mechanisms. The Nusselt number (Nu) correlation for gas channel is given by Eq. (10).

$$\text{N}\text{u}= 2+1.8·{\text{R}\text{e}}^{0.5}·{\text{P}\text{r}}^{0.33}$$

The Prandtl number (Pr; defined as the ratio of momentum diffusivity to thermal diffusivity) for flue gas is 0.72 that is similar to ambient air. The Reynolds number (Re; defined as the ratio of inertial force to viscous force) of the gas channel is smaller than 1,000, and corresponds to laminar flow. When the inlet gas flow rate increases from 5 to 8.33 10^− 5 kg/s, the heat transfer coefficient changes from 17.9 to 20.5 W/m² K (thermal resistance changes from 0.12 to 0.095 K/W). However, the overall thermal resistance decreases only by 3.7%, and the heat loss over time is not significantly affected. In addition, the saturation time for CO₂ capture process is reduced by 66.6% that corresponds to a heat loss saving of 61.3 kJ/kg.

To summarize, as the inlet mass flow rate increases, a larger amount of thermal energy can be harvested. In contrast, the CO₂ concentration at the outlet increases considering that the driving force of CO₂ passing through the membrane is constant. For instance, the inlet gas requires a flow rate of 8.33 10^− 5 kg/s or less to maintain the outlet concentration at 3 vol%, which is the design condition of the conventional chemical absorption process. An optimal design is required based on the target outlet CO₂ concentration.

3.4. Electric energy conversion

Thermal energy was converted to electricity using TE devices composed of Bi₂Te₃. Figure 9a shows the thermopower for a closed circuit by connecting resistors (the optimal resistors vary in terms of the working temperature range). The temperature of the hot side was varied, while that of the cold side was kept almost constant at 20 ^oC. The thermopower in working temperature range of 20–70 ^oC is presented in Fig. 9b, and optimal performance was obtained at a resistance of 5 Ω. Under these conditions, the system generated 0.12–0.94 W of thermopower, which is 44% higher than that generated by a resistance of 1 Ω for the same temperature difference.

Figure 9c shows the thermopower curve for a heat generation system that simulates the thermal energy harvesting process in point-source CO₂ capture. As the solvent temperature fluctuates severely during the CO₂ capture process, a constant heat generation system is considered to reduce the uncertainty. Thermopower is generated when the temperature difference is greater than 5 ^oC. The TEPA@SBA/MEA + PZ mixture generated a maximum power of 1.29 W. The average efficiency of thermoelectric conversion is 1.69%, although it varies in terms of materials. Figure 9d shows that the electric energy harvesting density (EEHD) of TEPA@SBA/MEA + PZ is 7.5 kJ/kg, i.e., TEPA@SBA/MEA + PZ can generate 7.5 J/g of electricity while capturing 6.6 mmol/g (0.29 g/g) of CO₂. When only thermal energy harvesting is considered instead of thermo-electric conversion, TEHD is 445 kJ/kg-solvent. Furthermore, CO₂ regeneration energy of amine absorbent is 4.2 kJ/g-CO₂. Considering that 0.29 kg of CO₂ is captured per 1 kg of solvent (working capacity), the energy consumption for CO₂ regeneration is 1,218 kJ/kg-solvent. Then, the TEHD can cover 36% of thermal energy consumption in the post-combustion CO₂ capture process.

3.5. Industrial applications

The industrial applications of point-source CO₂ capture-driven electric energy harvesting system were considered. The data were obtained from thermal power plants units 1–7 of Korea Western Power Co., Ltd., located in Taean, South Korea. The CO₂ emission of the reference system is 0.8 ton/MWh. Figure 10a shows that the fixed bed reactor using MEA as a working fluid can reduce CO₂ emission to 0.18 ton/MWh. The mass transfer area of the membrane reactor is lower than that of the fixed bed reactor, and the outlet CO₂ concentration is marginally higher. The CO₂ capture capacity of MEA + PZ is higher than that of MEA owing to the effect of the secondary amine. However, the outlet CO₂ concentration is also higher owing to the low reaction rate. The CO₂ outlet concentration of TEPA@SBA/MEA + PZ is 7.4% lower than that of MEA + PZ. Because amine-functionalized silica is dispersed in the fluid, the CO₂ working capacity remains unaffected, and the capture kinetics and selectivity are improved. Note that, the emission per unit energy generation is estimated based on the outlet concentration value of lab-scale results.

Figure 10b shows the contribution of electric energy harvesting to the industrial systems. The fixed bed reactor with MEA does not produce electricity, and the energy consumption for CO₂ desorption reaches 4.2 kJ/g-CO₂. In other words, from the perspective of energy index, 0.723 MWh of thermal energy is consumed to produce 1 MWh of electricity, which considerably reduces the efficiency of energy production. In contrast, a membrane reactor using MEA as the working fluid can harvest 23 MJ of electricity per 1 MWh of electricity production. In addition, the working fluid also significantly impacts the performance. For instance, TEPA@SBA/MEA + PZ improves the specific reaction heat by 52% compared with MEA and harvests 12.2% more electricity. This corresponds to 0.6% of the total electricity production in a power plant. Furthermore, in typical CO₂ capture applications reported in recent studies³, the energy consumption of 4.2 kJ/g is only partially reduced (i.e., 10–20%). However, the system proposed in this study produces electricity, which is highly feasible and significantly increases the energy efficiency.

In a membrane reactor (TEPA@PEGDMA), TEPA@SBA/MEA + PZ absorbs 6.6 mmol/g of CO₂ and simultaneously produce 7.5 kJ/kg of electricity. However, the feasibility of the present system is examined for industrial applications by considering additional energy consumption compared to a fixed bed reactor. The flue gas discharged from power plants is at ambient pressure (1 bar). In contrast, the inlet flue gas of the present system must be maintained at 3 bar to permeate the membrane. The work required for gas compression was obtained by applying the isotherm ideal gas equation to $\int PdV$ where P is pressure and V is volume. When the inlet flow rate is 8.33 10^− 5 kg/s, the gas compression work corresponds to 0.87 kJ. The temperature of cold side of the TE devices was maintained by forced convection of air. The work required to operate the fan until the temperature peak is 0.35 kJ (the heat transfer coefficient of air to maintain the temperature of the cold side was designed, and the air velocity was obtained from Nusselt number correlation. Based on the generated kinetic energy of air, isentropic work for operating fan was estimated). The net EEHD reaches 6.28 kJ/kg considering the additional work for present system compared to the fixed bed reactor. Note that, the reported thermo-electric conversion efficiency is 4% in the working temperature range (25–70 ^oC)²⁹, and the EEHD can be improved by more than 2 times

A breakthrough system that selectively captures CO₂ from flue gas discharged from power plants and simultaneously harvests electricity is proposed. The primary objective of this system is to transform the CO₂ capture system into an energy producer instead of merely minimizing its energy consumption. It captures CO₂ using functionalized solvents, harvests thermal energy through the exothermic reaction of amine-CO₂, and converts it into electricity using a thermoelectric (TE) device. A functionalized solvent in which tetraethylenepentamine (TEPA)-functionalized mesoporous silica (SBA-15) was dispersed in amine absorbents (MEA, monoethanolamine + PZ, piperazine) (TEPA@SBA/MEA+PZ) was synthesized. A CO₂ membrane composed of TEPA-impregnated poly(ethylene glycol) dimethacrylate (PEGDMA) was manufactured to minimize heat loss due to the mass transfer of gas-liquid, which increases the thermal energy harvesting performance.

The specific reaction heat of TEPA@SBA/MEA+PZ is 124.6 kJ/mol-CO₂, which is 52% higher than that of the primary amine solvent (MEA). Moreover, the CO₂ capture capacity of TEPA@SBA/MEA+PZ reaches 7.1 mmol-CO₂/g-solvent. The proposed membrane reactor increases the thermal resistance by a factor of more than three (0.706 K/W) compared with the existing fixed bed reactor and minimizes the heat loss during thermal energy harvesting. In the membrane reactor, CO₂ working capacity of TEPA@SBA/MEA+PZ in the region presenting steady performance is 6.6 mmol-CO₂/g-solvent from flue gas containing 15 mol% CO₂ and 85 mol% N₂, and the thermal energy harvesting density is 445 kJ-thermal/kg-solvent. In addition, TEPA@SBA/MEA+PZ exhibits constant performance over repeated absorption/regeneration cycles. The thermoelectric conversion efficiency of the TE device composed of Bi₂Te₃ reaches 1.69% in the working temperature range, corresponding to 7.5 kJ-electric/kg-solvent of electricity harvesting. The industrial application of CO₂ capture-driven electric energy harvesting was also discussed based on the thermal power plants located in Taean, South Korea. The data from the power plants show that the CO₂ emission is reduced from 0.8 ton/MWh to 0.25 ton/MWh, and 25.8 MJ/MWh of electric energy is harvested, which is 0.6% of the total electricity output of power plant.

Bui, M., et al. Carbon capture and storage (CCS): the way forward. Energy & Environmental Science 11, 1062–1176 (2018).
Lee, J.W., Kim, S., Pineda, I.T., Kang, Y.T. Review of nanoabsorbents for capture enhancement of CO₂ and its industrial applications with design criteria. Renewable and Sustainable Energy Reviews 138, 110524 (2021).
Kim, S., Xu, R., Lee, W., Lim, H.S., Kang, Y.T. Energy-efficiency analysis of industrial CO₂ removal system using nanoabsorbents. Journal of Cleaner Production 289, 125153 (2021).
Hu, X.E., et al. A review of N-functionalized solid adsorbents for post-combustion CO₂ capture. Applied Energy 260, 114244 (2020).
Luis, P. Use of monoethanolamine (MEA) for CO₂ capture in a global scenario: Consequences and alternatives. Desalination 380, 93–99 (2016).
Zhang, X., Huang, Y., Gao, H., Luo, X., Liang, Z., Tontiwachwuthikul, P. Zeolite catalyst-aided tri-solvent blend amine regeneration: An alternative pathway to reduce the energy consumption in amine-based CO₂ capture process. Applied Energy 240, 827–841 (2019).
Bhatti, U.H., Nam, S., Park, S., Baek, I.H. Performance and mechanism of metal oxide catalyst-aided amine solvent regeneration. ACS Sustainable Chemistry & Engineering 6, 12079–12087 (2018).
Bhatti, U.H., Sivanesan, D., Nam, S., Park, S.Y., Baek, I.H. Efficient Ag₂O–Ag₂CO₃ catalytic cycle and its role in minimizing the energy requirement of amine solvent regeneration for CO₂ capture. ACS Sustainable Chemistry & Engineering 7, 10234–10240 (2019).
Zhang, X., et al. Amine-based CO₂ capture aided by acid-basic bifunctional catalyst: Advancement of amine regeneration using metal modified MCM-41. Chemical Engineering Journal 383, 123077 (2020).
Prasongthum, N., Natewong, P., Reubroycharoen, P., Idem, R. Solvent regeneration of a CO₂-loaded BEA–AMP Bi-blend amine solvent with the aid of a solid Brønsted Ce (SO₄)₂/ZrO₂ superacid catalyst. Energy & Fuels 33, 1334–1343 (2019).
Zhang, X., et al. Reducing energy consumption of CO₂ desorption in CO₂-loaded aqueous amine solution using Al₂O₃/HZSM-5 bifunctional catalysts. Applied Energy 229, 562–576 (2018).
Gao, H., et al. Catalytic performance and mechanism of SO₄₂–/ZrO₂/SBA-15 catalyst for CO₂ desorption in CO₂-loaded monoethanolamine solution. Applied Energy 259, 114179 (2020).
Chao, C., Deng, Y., Dewil, R., Baeyens, J., Fan, X. Post-combustion carbon capture. Renewable and Sustainable Energy Reviews 138, 110490 (2021).
Hosseini-Ardali, S.M., Hazrati-Kalbibaki, M., Fattahi, M., Lezsovits, F. Multi-objective optimization of post combustion CO₂ capture using methyldiethanolamine (MDEA) and piperazine (PZ) bi-solvent. Energy 211, 119035 (2020).
Liu, J., et al. Highly polar but amorphous polymers with robust membrane CO₂/N₂ separation performance. Joule 3, 1881–1894 (2019).
Xia, Y., et al. Highly selective CO₂ capture and its direct photochemical conversion on ordered 2D/1D heterojunctions. Joule 3, 2792–2805 (2019).
Jeong-Potter, C., Farrauto, R. Feasibility study of combining direct air capture of CO₂ and methanation at isothermal conditions with dual function materials. Applied Catalysis B: Environmental 282, 119416 (2021).
Taniguchi, I., Kinugasa, K., Toyoda, M., Minezaki, K. Effect of amine structure on CO₂ capture by polymeric membranes. Science and Technology of advanced MaTerialS 18, 950–958 (2017).
Yoo, C.J., Lee, L.C., Jones, C.W. Probing intramolecular versus intermolecular CO₂ adsorption on amine-grafted SBA-15. Langmuir 31, 13350–13360 (2015).
Park, J.M., Yoo, D.K., Jhung, S.H. Selective CO₂ adsorption over functionalized Zr-based metal organic framework under atmospheric or lower pressure: Contribution of functional groups to adsorption. Chemical Engineering Journal 402, 126254 (2020).
Shin, S., Yoo, D.K., Bae, Y.S., Jhung, S.H. Polyvinylamine-loaded metal–organic framework MIL-101 for effective and selective CO₂ adsorption under atmospheric or lower pressure. Chemical Engineering Journal 389, 123429 (2020).
Yoo, D.K., Yoon, T.U., Bae, Y.S., Jhung, S.H. Metal-organic framework MIL-101 loaded with polymethacrylamide with or without further reduction: effective and selective CO₂ adsorption with amino or amide functionality. Chemical Engineering Journal 380, 122496 (2020).
Ben-Mansour, R., Qasem, N.A. An efficient temperature swing adsorption (TSA) process for separating CO₂ from CO₂/N₂ mixture using Mg-MOF-74. Energy Conversion and Management 156, 10–24 (2018).
Guo, Y., Tan, C., Sun, J., Li, W., Zhang, J., Zhao, C. Porous activated carbons derived from waste sugarcane bagasse for CO₂ adsorption. Chemical Engineering Journal 381, 122736 (2020).
Jang, E., Choi, S.W., Lee, K.B. Effect of carbonization temperature on the physical properties and CO₂ adsorption behavior of petroleum coke-derived porous carbon. Fuel 248, 85–92 (2019).
Choi, S.W., Tang, J., Pol, V.G., Lee, K.B. Pollen-derived porous carbon by KOH activation: Effect of physicochemical structure on CO₂ adsorption. Journal of CO₂ Utilization 29, 146–155 (2019).
Rehman, A., Park, S.J. Tunable nitrogen-doped microporous carbons: Delineating the role of optimum pore size for enhanced CO₂ adsorption. Chemical Engineering Journal 362, 731–742 (2019).
Wang, X., et al. Polyetheramine improves the CO₂ adsorption behavior of tetraethylenepentamine-functionalized sorbents. Chemical Engineering Journal 364, 475–484 (2019).
Maye, S., Girault, H.H., Peljo, P. Thermally regenerative copper nanoslurry flow batteries for heat-to-power conversion with low-grade thermal energy. Energy & Environmental Science 13, 2191–2199 (2020).

Table 1. Comparison of CO₂ adsorbents based on capture capacity and reaction heat.

Sorbent	CO₂ capture (mmol/g)	Adsorption heat (kJ/mol)	Method	Conditions	Ref.
SBA-15	0.79	37	TSA	25 ^oC, 1 bar	Present
TEPA@SBA-15	1.34	79	TSA	25 ^oC, 1 bar	Present
M-808-EDTA-ED-R	1.62	48	TSA	25 ^oC, 1 bar	²⁰
PVAm@MIL-101	3.3	49	VSA	25 ^oC, 1 bar	²¹
R-PM24@MOF	3.6	50	VSA	25 ^oC, 1 bar	²²
Mg-MOF-74	6.8	42	TSA	25 ^oC, 1 bar	²³
CAC-S	4.3	39	TSA	25 ^oC, 1 bar	²⁴
C500-K	4.2	33	TSA	25 ^oC, 1 bar	²⁵
K3-PDC1	3.4	40	TSA	25 ^oC, 1 bar	²⁶
GMK-3-800	3.7	42	TSA	25 ^oC, 1 bar	²⁷
MEM-41-TEPA	3.6	13	TSA	30 ^oC, 1 bar	²⁸

Schemes 1-2 are available in the Supplementary Files section.

There is NO Competing Interest.

NatComm222SI.docx
Scheme1.png
Scheme 1. Industrial system of a point-source CO₂ capture-driven electric energy harvesting.
Scheme2.png
Scheme 2. Strategy for point-source CO₂ capture and electric energy harvesting using functionalized solvents with thermal barrier; a) Reactor and b) Membrane

Download PDF

Version 1

posted

You are reading this latest preprint version

CO₂ capture-driven thermal and electric energy harvesting using amine-functionalized solvents with a poly(amidoamine) impregnated membrane

Status:

Version 1

Abstract

Figures

1. Introduction

2. Methods

2.1. Strategy

2.2. Materials

2.3. Characterization

2.4. Experiments

2.5. Theoretical analysis

3. Results And Discussion

3.1. Characterization

3.2. Point-source CO₂ capture and thermal energy harvesting

3.3. Theoretical performance analysis

3.4. Electric energy conversion

3.5. Industrial applications

4. Conclusions

References

Table

Schemes

Additional Declarations

Supplementary Files

Status:

Version 1

CO2 capture-driven thermal and electric energy harvesting using amine-functionalized solvents with a poly(amidoamine) impregnated membrane

Status:

Version 1

Abstract

Figures

1. Introduction

2. Methods

2.1. Strategy

2.2. Materials

2.3. Characterization

2.4. Experiments

2.5. Theoretical analysis

3. Results And Discussion

3.1. Characterization

3.2. Point-source CO2 capture and thermal energy harvesting

3.3. Theoretical performance analysis

3.4. Electric energy conversion

3.5. Industrial applications

4. Conclusions

References

Table

Schemes

Additional Declarations

Supplementary Files

Status:

Version 1

CO₂ capture-driven thermal and electric energy harvesting using amine-functionalized solvents with a poly(amidoamine) impregnated membrane

3.2. Point-source CO₂ capture and thermal energy harvesting