Properties of chemicals and solvents used in the experiment.
CO2 absorption into a binary solvent system was studied using a batch‐mode gas/liquid absorption apparatus. The binary system composed of potassium carbonate (K2CO3) and piperazine (PZ) showed a strong synergistic effect, whereby the binary solvent performed better than either of the individual solvents for CO2 absorption. The other pairs of solvents tested (K2CO3/monoethanolamine (MEA) and K2CO3/NaOH) showed no synergistic effects. The results indicate that this synergistic effect only occurs with specific pairs of solvents. The mechanism for the synergistic effect is postulated that the activated CO2 on PZ migrates to K2CO3, or a more reactive intermediate complex between K2CO3 and PZ is formed.
- carbon capture
- binary solvent
- synergy effect
- potassium carbonate
- CO2 absorption
There has been a growing concern over greenhouse gas emissions as they are considered to be the direct cause of global warming [1, 2]. Postcombustion capture technology is widely being studied for capturing CO2 produced in power generation plants [3–5]. Compared with other CO2 capture technologies such as oxy‐fuel combustion and integrated gasification combined cycle (IGCC), postcombustion capture is regarded as the most probable technology to be first employed when carbon capture becomes a reality in the near future in terms of technology readiness level, flexibility, and economics . Postcombustion capture technology uses liquid solvents to make efficient contact with CO2‐containing flue gas, during which CO2 interacts and reacts with the solvent and is removed from the flue gas stream. After absorption, the CO2‐laden solvent undergoes a regeneration operation, releasing pure CO2 which is then compressed, transported, and sequestered. The regenerated solvent, now at lean state, is returned to start the next cycle of CO2 capture. The whole operation is a continuous process. The same or similar technologies have been applied for decades for natural gas purification and syngas CO2 separation [7–9]. For greenhouse gas CO2 mitigation applications, commercial solvents such as amine, potassium carbonate, and methanol are currently being tested, however, improved solvents are required to reduce the cost and increase the efficiency of postcombustion capture systems. At the moment, solvents that are being developed for CO2 capture include nonconventional amines, aqueous ammonia, amino acids, ionic liquids, and mixtures of two or more solvents, i.e., hybrid systems [10, 11].
Potassium carbonate is known to be used in industrial CO2 separation processes, such as Benfield and Catcarb , as the main solvent with or without proprietary additives. It has advantages over amines: lower cost, lower heat of absorption, thermal stability, nonvolatile, less corrosiveness, low toxicity, and environmentally friendly. A major downside for using K2CO3, however, is its slow absorption rate and low CO2 absorption capacity, resulting in poor CO2 mass transfer rate relative to amines. The way to overcome the aforementioned shortcomings of K2CO3 is to add promoter, i.e., a hybrid solvent. Hybrid solvent systems have the potential to perform better than the individual components alone. Physicochemical properties of different solvents can supplement each other. Synergistic effect or cooperative effect of hybrid solvents has been found in applications in other areas such as extraction and coal swelling [13, 14]. The mechanisms of the synergistic effects are suggested to be engendered by thermodynamics and hydrogen bonding.
We have been studying CO2 absorption using an aqueous potassium carbonate solvent solution with the addition of other solvents in an attempt to improve CO2 absorption performance. In this chapter, we report results of a synergistic effect that became apparent during these studies. When small amount of piperazine (PZ) is added to the potassium carbonate solution, both CO2 absorption rate and capacity are significantly enhanced, exceeding the mathematical sum of the CO2 absorption rate and the capacity of the individual solvents.
Piperazine itself is an active absorbent for CO2 . For some engineering reasons, it has only been used as an additive or a promoter to other common CO2 capture amines . With amine solvents, piperazine has shown promotional effect. For instance, the CESAR‐1 solvent is an aqueous blend of AMP (2‐amino‐2‐methyl‐1‐propanol) and PZ which showed a reduction of about 20% in the regeneration energy and 45% in the solvent circulation rate compared to those of MEA‐based CO2 capture process under similar process condition .
There have been some reports on the promotional/synergistic effect on CO2 capture by K2CO3 and PZ . This study builds upon previous achievements and provides convincing experimental evidence of the synergistic effect.
A batch‐mode liquid‐gas absorption apparatus was constructed in CanmetENERGY, Ottawa. A schematic and a photo of the apparatus are shown in Figure 1. All of the connections within the system are vacuum‐proof. The volume of the four‐neck flask is 690 ml. The solute gas used in the experiment is a mixture of CO2 and air (49 v% of CO2). CO2 absorption tests were carried out at 21℃ (room temperature). The flask was placed in a water bath to maintain a constant temperature (CO2 absorption is exothermic). First, the flask was purged by the solute gas for 10 min. Then all of the valves of the flask were closed, leaving the gas in the flask at ambient pressure. After this, 10 ml of solvent was introduced into the flask by opening the two valves of the funnel, and then closing them quickly so that the flask becomes a closed system with gaseous solute in contact with liquid solvent. The liquid was agitated by a magnetic stirrer at 350 rpm (there was no difference on the CO2 absorption results with rpm in the range of 300–400). When the CO2 was absorbed, the pressure in the flask decreased. This pressure was monitored with a solid state pressure sensor/transducer (PX209‐30V15GI) from Omega. A monotonous pressure declining curve was obtained, revealing the CO2 absorption kinetics (rate of decline) as well as capacity (the final level‐off of the decline).
The solvents used and their concentrations in aqueous solution are shown in Table 1. In the test, the primary solvent was aqueous potassium carbonate, K2CO3. Other solvents were used as secondary promoters to see if there was a synergistic effect between the primary and secondary solvents. The hybrid solvents were obtained by mixing the individual solvents (shown in Table 1) with certain ratio (quantity in ml). Water was added to adjust the effective concentration and the final volume in a test.
Three test series were completed, one for each of the secondary solvents. These included:
7 ml K2CO3/3 ml H2O (K2CO3 represents its solution in Table 1)
3 ml PZ/7 ml H2O (PZ represents its solution in Table 1)
7 ml K2CO3/3 ml PZ
7 ml K2CO3/3ml H2O
3 ml MEA/7 ml H2O (MEA represents its solution in Table 1)
7 ml K2CO3/3 ml MEA
7 ml K2CO3/3 ml H2O
3 ml NaOH/7 ml H2O (NaOH represents its solution in Table 1)
7 ml K2CO3/3 ml NaOH
3. Results and discussion
The CO2 absorption results for test series 1 are shown in Figure 2. After the solvent was introduced into the flask filled with CO2/air, the chemisorption occurred as demonstrated by the pressure decrease. From the results in Figure 2, it can be seen that K2CO3 showed a slow absorption rate and low absorption capacity. Piperazine's CO2 absorption rate was faster and had higher capacity. When the two solvents were mixed, the binary solvent system absorbed more CO2 at an even faster rate. The mathematical sum of the individual CO2 absorption curves of the K2CO3 and piperazine (the sum of the green curve and the light blue curve) is shown in Figure 2 as well (dark blue line). It is clear that the binary solvent system performed much better for CO2 absorption than the mathematical sum of the individual solvents. The two curves (orange and purple in Figure 2) showing the CO2 absorption results of the binary solvent system from two different tests under the same conditions indicate that the apparatus worked very well with a high degree of repeatability.
The test results of the binary solvent system of K2CO3 and MEA are shown in Figure 3. The component solvents of K2CO3 and MEA were of similar effectiveness for CO2 absorption. The binary solvent system showed only a slight synergistic effect.
In order to investigate the necessary and/or sufficient conditions for the synergistic effect of a stronger CO2 solvent with a milder solvent (e.g., PZ with K2CO3), the binary solvent system of K2CO3 with NaOH was tested (Figure 4). It can be seen from Figure 4 that, although NaOH is a much stronger CO2 solvent than K2CO3, the binary solvent system of K2CO3 and NaOH does not show any synergistic effect.
Therefore, it is a necessary but not a sufficient condition for a binary solvent system with different CO2 absorption capacities and kinetics to generate synergistic effect. Among the three pairs, only the binary solvent of K2CO3 and PZ showed a positive synergistic effect on CO2 absorption.
As shown by our experiment (Figure 2) and others , PZ is a stronger and faster CO2 solvent than K2CO3. When the ratio of K2CO3 and PZ was varied, the CO2 absorption curves shifted from the curve of K2CO3 to the curve of PZ, as shown in Figure 5. The binary solvent systems between the two pure solvents exhibit synergistic effect. Illustrated in Figure 6 is the synergistic performance of the binary solvent as well as the relationships with the two pure solvents (this is only a general illustration).
PZ is an expensive solvent. Whether or not it is suitable, alone, as a CO2 capture solvent is still being explored in terms of thermal stability, corrosiveness and cost, etc. . As shown by this study, it is promising to apply a binary solvent of K2CO3 and PZ at a ratio that maximizes the synergistic effect on CO2 capture. Savings from operating at this condition could be realized in terms of solvent cost, reduction of the absorber and regenerator sizes due to the improved CO2 absorption rate and capacity. More effective solvents would require smaller absorbers and regenerators, leading to lower capital costs.
J. Tim Cullinane and Gary T. Rochelle have reported the promotional effect of K2CO3 and PZ by kinetics . They concluded that the promotional effect comes from the kinetics of the two individual solvents and that the two solvents absorb CO2 independently. These cannot explain the observations of this study. The promotional or synergistic effect of PZ to K2CO3 has been suggested to occur through an intermediate formed between CO2 (aq) and PZ [20–22]. This hypothesis, however, still needs to be verified experimentally. Our results indicate that there may be a more interactive mechanism affecting the hybrid solvent performance. Having a binary solvent system with one solvent more effective than the other is a necessary condition for the synergistic effect (the pairs of K2CO3 and PZ, K2CO3, and NaOH), but not a sufficient condition (K2CO3 and NaOH). There must be other reasons behind the synergistic effect. Here we postulate two mechanisms:
CO2 transition (or spill over or migration): CO2 is reactivated by solvent B forming a labile state [[B] · [CO2](aq)], then transfers or migrates to solvent A to finish CO2 absorption (Figure 6). Likely hydrogen bonding is involved.
Reactive complex intermediate structure between the two solvents: in the CO2 absorption system, there occur some kind of interactions between the two solvents by hydrogen bonding or local ionic attraction, forming a more reactive intermediate complex [A·B] with improved CO2 absorption ability.
The factors of electron donner strength, dielectric constants, solubility parameters of the individual absorbent, and hydrogen‐bonding/nonhydrogen‐bonding may influence the degree of synergistic effects. There needs more research work to capture and characterize the reactive intermediate complex or transition state, to prove or disprove these postulated mechanisms.
The idea of combining solvents to improve absorption is effective for piperazine and K2CO3. These two solvents interact together and generate a greater absorption than each of the individual solvents. The other solvents, i.e., MEA and NaOH, when mixed with K2CO3 did not improve CO2 absorption, implying that the synergistic effect only occurs selectively between specific pairs of solvents. The solution of 3 ml piperazine with 7 ml potassium carbonate is the optimal ratio that increases CO2 absorption using the least amount of piperazine. The results of these tests show the possibility of using piperazine and K2CO3 solution at an industrial scale. If correctly implemented, it would result in savings in capital by reducing the absorber size compared to use K2CO3 alone. The next step for this project is to apply these results within a larger system. The major conclusions from the tests conducted are summarized below:
A synergistic effect between K2CO3 and piperazine was observed.
This synergistic effect only happens between this specific pair of solvents and is not universal. Other than the thermodynamic reasons behind the effect, there seems to be some additional mechanism that enhances the reaction (potentially a labile [CO2] formation followed by migration or some more reactive intermediate complex structure formed between the two solvent molecules).
3 ml piperazine/7 ml K2CO3 ratio is the most effective (faster absorption rate and higher absorption capacity).
The project was financially supported by the Canadian Federal Government ecoEII Program. Thanks goes to Mr. Zlatko Lovrenovic, coop student from University of Ottawa, for his contribution to the project. An extra financial support from AirScience Technologies Inc., Montreal, Canada, is gratefully acknowledged.