In 2014, the Intergovernmental Panel on Climate Change (IPCC) reported1) that warming of the climate is unequivocal and that the largest contribution is a result of increasing atmospheric CO2 since 1750. In the previous year, the National Oceanic and Atmospheric Administration (NOAA) in Mauna Loa, Hawaii, observed that the concentration of atmospheric CO2 surpassed 400 ppm for the first time since measurements began in 1958. This value is approximately 120 ppm higher than that of the pre-industrial atmosphere (approximately 280 ppm).2) Global CO2 emissions from fuel combustion in 2012 reached a record of 31.7 gigatons (GtCO2) based on calculations performed by the International Energy Agency (IEA).3) On the other hand, there are many reports that CO2 is not an essential source for Climate Change.4–10) The experimental fact can only prove the answer of these discussions. Thus, techniques for reduction of CO2 must be globally important.
Current efforts for CO2 reduction include CO2 capture and storage (CCS)11) and artificial photosynthetic systems (APSs).12,13) CCS involves the capture of CO2 using chemical absorbent (i.e., monoethanolamine, MEA 1 etc.)14–20) in high-density areas, such as industrial facilities and power stations, and transporting CO2 to deep subsurface rock formations or the bottom of the ocean via pipelines. This is a useful and efficient technique that can prevent the release of large quantities of CO2. However, this technology does not provide any immediate economic benefit, and the captured CO2 is not chemically altered. In the projects of CCS, the liberation of CO2 gas from chemical absorbent requires heating at high temperature, which uses high amount of electricity. This means that introducing CCS plants spends one part of electric-generating capacity in power station. The limited number of sites is also problematic because CCS plants must be constructed near areas with high CO2 densities. In contrast, APSs use solar light and a metal catalyst. In this approach, CO2 is transformed into HCO2H12) or CO,13) whereby the products can be converted to sources of energy (such as methanol). The use of this system is more valuable than CCS from an economic perspective; however, APSs have been shown to work only when high concentrations of CO2 (>1 atm) are present. Planting trees and developing alternative energy resources (e.g., H2, biofuel, and solar power) are also ongoing activities. Unfortunately, planting trees does not offer an immediate economic benefit, and fossil fuels remain the cheapest energy source. Continuous efforts to establish a more promising method for CO2 reduction are needed. Clearly, the lack of an immediate economic advantage for CO2 reduction prevents positive action toward solving this issue. In this regard, an indirect economic approach called “carbon emission trading”21) has been introduced. This concept is currently one of the most effective tactics for reducing carbon dioxide production, although it is yet to limit CO2 production to an ideal level. To address these problems, we developed an “energyless CO2 fixation” method, or ECO2-fix. ECO2-fix is a novel atmospheric CO2 fixation reaction that does not require energy input, such as solar power, heating, cooling, or stirring. Moreover, ECO2-fix also has the capacity to produce high value-added materials, such as precursors of medicines, chemical products, and energy sources. ECO2-fix has four major advantages. First, the reaction absorbs CO2, which mimics processes that occur in plants. Second, the valuable products generated from CO2 possess financial value. Third, carbon credits can still be traded based on the quantity of CO2 that can be fixed. Finally, regardless of geographic location, CO2 is abundant in the atmosphere. Thus, the development of ECO2-fix will provide both ecological and economic benefits.
Since the discovery of the famous Kolbe–Schmidt reaction22,23) and Grignard reaction24) with CO2, several CO2 fixation methods have been reported in the literature.25–47) However, no methods for CO2 fixation are similar to ECO2-fix, which is presumably due to the extremely low concentrations of CO2 (approximately 400 ppm, or 0.04%) in the ambient air. Air also contains other reactive species, including O2 (20%) and water (approximately 0.4%, depending on the environment), which may prevent the desired reactions. Therefore, we focused on monoethanolamine (MEA 1),14–20) which is a chemical absorbent that is capable of capturing CO2 from flue gas or other gaseous streams. Normal carbamic acid rapidly decarboxylates to form an amine while liberating CO2. The hydroxyethyl moiety of MEA 1 is crucial for trapping CO2, which is useful information for developing a method to fix CO2 from the atmosphere. Nevertheless, the ability of MEA 1 and its derivatives to absorb atmospheric CO2 has been undeveloped. Thus, we first tested CO2 absorption from ambient air. For this experiment, we prepared a battery-powered CO2 monitor and dessicator (dimensions of 465×290×265 mm, with a total volume of 35.7 L) to monitor CO2 absorption in an enclosed space. The results are shown in Figs. 1a and b. The initial CO2 concentration (437–573 ppm, 0.70–0.91 mmol CO2) was highly dependent on uncontrollable variables in the experimental room, including the time, season, number of people present, and air from ventilation. When 0.33 mL (5.0 mmol) of MEA 1 was incubated in the sealed box, the CO2 concentration immediately decreased from 551 ppm to less than 10 ppm within 17 h. When 10 mL of MEA 1 was incubated in the chamber, the concentration of CO2 reached 0 ppm after 6.5 h. These results suggest that a large excess of MEA 1 in the atmosphere could reduce the global CO2 concentration (400 ppm) to its pre-industrial level (280 ppm) within several hours. Next, we tested several MEA derivatives for their ability to absorb CO2 in the sealed box. The absorption ability of N-methylated 2-(methylamino)ethanol (MAE 4) was similar to that of MEA 1. Next, we tested α-substituted derivatives of MEA 1. However, 2-amino-2-methylpropan-1-ol (AMP 5) absorbed less CO2 than that of MEA 1, and phenylglycinol (PG 6) did not reduce the concentration of CO2 in the studied air. Derivatives of MEA having multiple hydroxyethyl functionalities were also tested and were found to be less effective than MEA 1. Specifically, the CO2 absorption ability of diethanolamine (DEA 2) was lower than that of MEA 1, and the absorption of CO2 by triethanolamine (TEA 3) was undetected. Thus, we concluded that MEA and certain derivatives with suitable functional groups are “active CO2 absorption species,” even under ambient pressures.
Next, we investigated the ratio of CO2 absorption in air using MEA 1. As shown in Fig. 2a, MEA 1 was placed on a scale exposed to air and its weight was measured over time. The average masses of MEA 1 over time from three replicates are shown in Fig. 2b. While MEA 1 is a volatile species, our results clearly indicate that the rate of CO2 absorption is faster than that of evaporation. In our study, CO2 absorption reached equilibrium after approximately 1.5 d. To determine the composition of the mixture, elemental analyses were conducted (Fig. 2a). Samples collected after 1 and 7 d were compared and analysed, revealing that the former had not yet reached equilibrium. In addition, the elemental analysis of the sample collected after 7 d suggested that the mixture x(CO2)·y(MEA)·z(H2O) 7, which was at equilibrium, contained CO2, MEA 1, and H2O at a ratio of approximately 1 : 3 : 3. Water would be derived from moisture in atmosphere. Judging from the values of IR (1480 cm−1) and 13C-NMR (δ 164.8), absorbed CO2 would be transformed into carbonate. The plausible structure of 7 was shown in Fig. 2c. The chemical yields of 7 from 1 showed between 85–95% yields.
After obtaining quantitative information regarding the ability of MEA 1 to absorb CO2, our efforts shifted from the absorption of CO2 to the fixation of CO2 without consuming external energy. For this purpose, a reaction in which urethanation of propargylamines forms oxazolidinone derivatives using a metal catalyst39,41,47) was selected. Many types of pharmacologically active compounds containing oxazolidinone frameworks have been reported.48,49) When propargylamine and its derivatives bearing alkyl or phenyl groups 8 were added to a Au or Ag catalyst in MeOH under atmospheric conditions, no desired product 9 was observed (Fig. 3, Eq. 1). The addition of MEA 1 or (CO2)·3(MEA)·3(H2O) 7 also failed to produce 9. These results suggested that the carboxylation of the amine to carbamic acid requires activation or high CO2 concentrations. Thus, we prepared N-hydroxyethyl propargylamine (HEPA) 10. Because MEA 1 absorbs atmospheric CO2, the N-hydroxyethyl group of HEPA 10 should assist in the absorption of CO2, and intramolecular cyclization can occur more easily via an intermediate 10′. Indeed, the incubation of HEPA 10 and 2 mol% Au(Xantphos)Cl in MeOH at ambient temperature provided the desired product 1150) (Eq. 2). Notably, a lack of reproducibility regarding yields (2–20%) occurred, which is presumably due to variations in the CO2 concentration (300–700 ppm) in the studied air. No improvements were observed when additional MEA 1 or (CO2)·3(MEA)·3(H2O) 7 was added. Although other derivatives of 12 that contained an alkyl or phenyl group (R1) or hydroxyethyl group (R2) at the alkyne terminus provided the product 13 with up to 41% yields; these results were dependent on uncontrollable variables of the environment rather than the reactants (Eq. 3). The addition of the N-hydroxyethyl group was necessary to observe metal-catalysed urethanation in this study. However, its reactivity was highly affected by variable CO2 concentrations and other atmospheric variables.
Based on the reactions shown in Fig. 3, we inferred that more effective fixation would require higher and more stable CO2 concentrations. During several experiments using (CO2)·3(MEA)·3(H2O) 7, we found that the addition of 10% HCl effectively liberated gaseous CO2 at room temperature (Fig. 4a). The formation of ammonium ions may prompt CO2 disassociation. Thus, we devised a new CO2 generator using atmospheric CO2. The system is shown in Fig. 4b. The generation of gaseous CO2 by adding 10% HCl to (CO2)·3(MEA)·3(H2O) 7 in a closed flask enabled us to create an environment for the reaction with high CO2 levels. Using this improved technique, the reaction of HEPA 10 (2 mol% Au(Xantphos)Cl, MeOH, rt) provided the product 11 with a good yield (73%) (Fig. 4c, entry 1). Moreover, the other selected substrates 12a–g, which incorporated various substituents, produced the desired products 13a–g with acceptable yields and reliable reproducibilities (entries 2–9). We further demonstrated that the intramolecular assistance of the hydroxyethyl group in these substrates was not essential. Using propargylamine (8a) or other derivatives (8b–d, i.e., no hydroxylethyl groups) provided the carboxylated products 9a–d with high yields (entries 10–14). Notably, such reactions do not require stirring. After generating CO2 and incubating the substrate and catalyst in MeOH by shaking by hand for several seconds, the reaction of 12a provided the product 13a with a comparable yield (72%, entry 3). Similar results were observed for the reaction of 8a (85% yield, entry 11). Thus, fixation proceeded without external energy inputs.
In summary, we found that MEA 1 and its derivatives are very capable of absorbing CO2 at atmospheric temperature and pressure. Importantly, MEA 1 completely absorbed atmospheric CO2 (549 ppm) in a closed system within several hours. Although photosynthesis by plants works well under natural sunlight (approximately 12 h), MEA 1 can absorb atmospheric CO2 regardless of the ambient sunlight conditions and without external energy inputs in the form of heating, cooling, or stirring. We also developed CO2 fixation reactions using oxazolidinone derivatives. An N-hydroxyethyl moiety was essential for the absorption of CO2 and its subsequent fixation. Furthermore, the construction of a CO2 generator that uses atmospheric CO2 enabled us to test not only N-hydroxyethyl substrates but also propargylamine derivatives (that did not require N-hydroxyethyl groups). In contrast to photosynthesis, which produces inexpensive sugar, our system can provide high value-added material from atmospheric CO2. Thus, we succeeded in establishing a procedure for acquiring this carbon resource from the atmosphere. We anticipate that the methods described herein will provide an alternative to other CO2 fixation methods, including artificial photosynthesis, the Kolbe–Schmitt reaction, and the Grignard reaction. Further improvements (e.g., recycling of MEA) and applications (e.g., other reactions using a CO2 generator) are currently under investigation.
