Non-Aqueous Solvent Mixtures for CO₂ Capture: Choline Hydroxide-Based Deep Eutectic Solvents Absorbent Performance at Various Temperatures and Pressures

Abstract

Carbon dioxide (CO₂) absorption in a non-aqueous solution is a potential technology for reducing greenhouse gas emissions. In this study, a non-aqueous solvent, sulfolane and dimethylsulfoxide (DMSO), was functionalized with a deep eutectic solvent (DES) consisting of choline hydroxide and polyamines diethylenetriamine (DETA) and triethylenetetramine (TETA). The non-aqueous absorbents’ CO₂ absorption ability was investigated in a high-pressure absorption reactor with a variable absorption temperature (303.15–333.15 K) and pressure (350–1400 kPa). The results showed that 2M ChOH:TETA−DMSO solution had the highest CO₂ loading capacity when compared with other screened solutions, such as 2M ChOH:TETA−Sulfolane, 2M ChOH:DETA−DMSO and 2M ChOH:DETA−Sulfolane. It was also found that the absorption capacity increased with increasing pressure and decreased with temperature. The highest CO₂ absorption by 2M ChOH:TETA−DMSO was observed at a partial pressure of 1400 kPa at 303.15 K 1.2507 mol CO₂/mol DES. The use of a non-aqueous solvent in the mixture showed a phase separation phenomenon after the CO₂ absorption reaction due to the formation of insoluble carbamate salt, which was identified through FTIR analysis. These findings suggest that the use of a DES polyamine mixed with a non-aqueous solvent could be a promising solution for CO₂ capture.

1. Introduction

Among the primary pollutant acid gases, CO₂ is considered the most significant, as it contributes to around 77% of greenhouse gases leading to global warming. CO₂ emissions can be controlled by separating CO₂ from the gaseous streams in natural treatment plants or flue gas from industrial plants. Post-combustion CO₂ capture processes are currently one of the most common technologies used in experimental laboratories and industrial applications. Absorption, adsorption, membrane separation and cryogenic separation are the most commonly used techniques in post-combustion capture technology. According to the literature, the absorption process has been widely used in comparison to other separation techniques, particularly in the oil and gas sector, coal power plants, and the chemical industry [1,2]. This is mainly because of its dependable technology, vast processing capacity, and long history of industrial operation [3,4]. Traditionally, organic solvents are the most widely used in these applications. The first commercial solvent used in gas purification was triethanolamine (TEA), followed by monoethanolamine (MEA) and diethanolamine (DEA). The problem with organic solvents is that they are extremely reactive and cytotoxic, with a high regeneration cost, equipment corrosion, and oxidative degradation for some amines [5,6]. The high volatility of aqueous amine solutions results in increased regeneration energy requirements. Hence, nonvolatile ionic liquids were introduced to overcome this problem. As a result, the present concept of employing deep-eutectic ionic liquids (DES) is gaining attraction because of their low cost, simple production processes, low volatility, biodegradability, and non-toxicity [7,8]. Figure 1 shows the number of publications on DES for CO₂ absorption from 2014–2022. In recent years, researchers have synthesized DES by mixing hydrogen bond acceptors (HBA) and hydrogen bond donors (HBD). Abbott and colleagues studied and reported the first DES system in 2003, which comprised quaternary ammonium salt choline chloride (ChCl) and urea in the molar ratio of 1:2. A study by Leron and Li (2013) on the synthesis of choline chloride: ethylene glycol with a 1:2 molar ratio shows a high CO₂ absorption capacity at 3.126 mol/mol DES [9]. Another study by Ali et al. (2014), whereby they synthesized a 1:2.5 molar ratio of choline chloride:urea, recorded a CO₂ uptake of up to 0.0211 mol/mol DES [10]. According to Mahi et al. (2019), non-aqueous solvents could be a solution for CO₂ absorption [11]. In their work, 30% MEA was combined with choline chloride:ethylene glycol in a 1:2 molar ratio. The CO₂ absorption shows 0.8 mol/mol DES being absorbed in the solution.

In the present work, we synthesized DES by mixing an equimolar ratio of choline hydroxide (ChOH) as HBA with diethylenetriamine (DETA) and triethylenetetramine (TETA) as HBD to introduce mixtures of DES non-aqueous solvent. These solvents are gaining attention as energy-saving solvents that efficiently absorb CO₂ with a significant reduction in regeneration energy costs due to the absence of water [12,13]. Dimethylsulfoxide (DMSO) and sulfolane were chosen as the non-aqueous solvents in this study. DMSO can dissolve aromatic, unsaturated hydrocarbons and many inorganic salts, which makes DMSO a powerful and versatile physical solvent [14]. A current study reported that DMSO was used as a diluent not only to reduce energy consumption due to low vaporization energy and low heat capacity but also due to being a non-corrosive organic solvent [13,15,16]. On the other hand, sulfolane was chosen due to its maturity in acid gas removal processes [17]. There is a wide range of studies regarding the use of sulfolane as a promising non-volatile organic solvent with low corrosivity and exceptional solubility, which could potentially improve CO₂ absorption capability [18,19].

2. Materials and Methods

2.1. Materials

The following are the raw materials used in polyamine synthesis. DETA, TETA, choline chloride, 45% in methanol, sulfolane and DMSO were purchased from Sigma-Aldrich (M) Sdn. Bhd., Petaling Jaya, Malaysia. All products were used without further purifications with a high purity, as summarized in

Table 1. Table of provenance for the compound system.

Chemicals	CASRN	Source	Purity
45% Choline hydroxide in methanol	123-41-1	Sigma-Aldrich	99%
DETA	111-40-0	Merck	98%
TETA	112-24-3	Merck	98%
Carbon dioxide	MY 006-474	Alpha Gas Solution	99%
Sulfolane	126-33-0	Aldrich	99%
DMSO	67-68-5	Sigma-Aldrich	99%

.

2.2. Synthesis of DES

ChOH:TETA and ChOH:DETA were synthesized according to the literature [20]. An equimolar mixture of 45 wt% ChOH in methanol and polyamines (DETA, TETA) was mixed in a round bottom flask with heating and stirring at 353.15 K for 30 min. Then, the solvent removal took place in a rotary evaporator for 2 h under reduced pressure. The sample was then dried in a vacuum oven overnight with 70 mbar pressure to remove excess moisture, and the acquired DES was obtained and kept in tightly sealed bottles.

2.3. Sample Preparation

The components were weighed using a calibrated single pan digital balance (HR-250AZ, A&D Limited, Toshima, Japan) with an accuracy of 0.1 mg to make 2M DES non-aqueous mixtures. To reduce moisture absorption from the environment, all mixes were maintained in tightly sealed bottles.

2.4. CO₂ Solubility Measurements

The solubility of CO₂ was determined using a pressure drop method, as previously reported. Figure 2 depicts a schematic of the experimental setup, which includes a welded stirrer assembly, a thermocouple, an intake gas tube, and an air vent. Both integrated reactors were outfitted with a thermostat heater and a thermocouple. To begin the experiment, the system was purged with air by flushing it with CO₂. Next, purified CO₂ from a gas tank was introduced and heated to the desired temperature. A high-pressure reactor cell holding 20 mL of 2M DES mixtures was sealed, and CO₂ gas was continually supplied, while the sample was agitated until equilibrium was achieved. The pressure in the system gradually decreased, and equilibrium was achieved once the pressure in the reactor cell remained constant.

The amount of CO₂ in the gas phase was calculated using the known volume, pressure, and temperature parameters. The initial and final pressures (equilibrium pressure) that were previously recorded were used to calculate the mole differences during the absorption process. The CO₂ solubility was calculated from the modified Ideal Gas Law to accurately estimate the number of moles of CO₂ absorbed, as expressed in Equation (1) [21],

$$\chi =\frac{\left[\frac{\left(P_{Ti}-P_{Vi}\right)\times V_{gc}}{zRT}\right]-\left[\frac{\left(P_{Tf}-P_{Vf}\right)\times \left(V_{gc}+\left(V_{cell}-V_{sol}\right)\right)}{zRT}\right]}{n_{total}}$$

(1)

where χ is the CO₂ loading in mol of CO₂/total mol of DES. P_T is the total pressure (kPa), P_V is the vapor pressure, V_gc is the volume of the gas container, V_cell is the volume of the reactor cell, V_sol is the volume of the solution, i is the initial condition, and f is the final condition. Meanwhile, n_total is the summation of the moles of DES in the liquid phase.

2.5. Apparatus Validation

In order to evaluate the consistency of the equipment, an apparatus reliability study was performed, and the results were compared to those found in the previously reported data by Jalili et al. (2015) [22]. A series of runs were conducted at various pressures using sulfolane at a temperature of 303.15 K. The data obtained were compared to literature values, as depicted in Figure 3, and yielded an average absolute relative deviation (AARD) of 0.461%.

3. Results and Discussion

3.1. Chemical Equilibria

The chemical route between polyamine and CO₂ is quite complicated, as each of a polyamine molecule’s amine moieties can interact with CO₂ individually. In both aqueous and organic solvents, primary and secondary amine groups (denoted as R₁R₂H, with R₁ = H for primary amines) engage in the primary reaction pathway, resulting in the creation of the matching amine carbamate (R₁R₂NCO₂⁻) species. Aqueous solutions, on the other hand, produce bicarbonate and carbonate products. In aqueous environments, tertiary amine functionality (R₁R₂R₃N) reacts with CO₂ to produce bicarbonate (Equation (7)), but it is unable to form carbamate (Equation (2)) due to the absence of hydrogen atoms [23].

2 R₁R₂NH + CO₂ ⇄ R₁R₂NCO₂⁻ + R₁R₂NH₂⁺

(2)

R₁R₂NCO₂⁻ + CO₂ + 2 H₂O ⇄ 2 HCO₃⁻ + R₁R₂NH₂⁺

(3)

R₁R₂NH + CO₂ + H₂O ⇄ HCO₃⁻ + R₁R₂NH₂⁺

(4)

HCO₃⁻ + R₁R₂NH ⇄ CO₃²⁻ + R₁R₂NH₂⁺

(5)

CO₂ + CO₃²⁻ + H₂O ⇄ 2 HCO₃⁻

(6)

R₁R₂R₃N + CO₂ + H₂O ⇄ HCO₃⁻ + R₁R₂R₃NH⁺

(7)

Polycarbamate species can occur when two or more amino functionalities of a polyamine experience simultaneous chemical reactions. As a result, polyamines with primary and secondary functionalities can form polycarbamate compounds via the following reaction between a carbamate and CO₂, as shown in Equations (8)–(11), where H₂N–R–NH–R–NH₂ stands for a generic triamine with two primary amino groups and a singular secondary group [24].

H₂N–R–NH–R–NH₂ + CO₂ ⇄ ⁺H₃N–R–NH–R–NHCO₂⁻

(8)

2 H₂N–R–NH–R–NH₂+ CO₂ ⇄ H₂N–R–NH–R–NHCO₂⁻ + H₂N–R–NH–R–NH₃⁺

(9)

H₂N–R–NH–R–NHCO₂⁻ + H₂N–R–NH–R–NH₂+ CO₂ ⇄
NH–(R–(NHCO₂⁻))₂ + H₂N–R–NH–R–NH₃⁺

(10)

NH–(R–(NHCO₂⁻))₂ + H₂N–R–NH–R–NH₂+ CO₂ ⇄
NCO₂⁻–(R–(NHCO₂⁻))₂ + H₂N–R–NH–R–NH₃⁺

(11)

3.2. CO₂ Absorption of 2M DES in Non-Aqueous Solutions

Figure 4a–d illustrates the solubility data for CO₂ in 2M DES non-aqueous mixtures measured experimentally. The equilibrium exhibits a typical pattern in which the CO₂ loading decreases with an increasing temperature and increases with an increasing pressure. Each combination in the figures shows a decreasing trend with rising temperatures from 303.15 to 333.15 K, demonstrating a high loading of CO₂ at lower temperatures at a constant pressure. The observed behaviors can be explained by a rise in the vapor pressure with the temperature, which results in a decrease in CO₂ loading. At a constant temperature, however, a rise in pressure has a positive influence on CO₂ solubility. This phenomenon may be described by the direct relation between gas diffusion into liquid and gas pressure above the solution’s surface. In correlation with this study, the phenomenon could be seen in most of the DES non-aqueous mixtures, except for DES that contains DETA, where a high gas solubility could be seen at 313.15 K. A recent study by Tzirakis and coworkers reported that the highest CO₂ absorption was observed at 313.15 K with 1.68 mol/mol [25]. Another study by Lua and team found that the CO₂ absorption capacity of the DETA mixture was the highest at 313.15 K with 1.70 mol/mol [26].

From the observations, it can be noted that, at the highest measured pressure, the 2M ChOH:TETA−DMSO mixture had the highest CO₂ absorption capacity, while the 2M ChOH:DETA−Sulfolane mixture had the lowest CO₂ absorption capacity among the mixtures. In general, the trend of CO₂ uptake is as follows: 2M ChOH:TETA−DMSO > 2M ChOH:TETA−Sulfolane > 2M ChOH:DETA−DMSO > 2M ChOH:DETA−Sulfolane. Moreover, the results of this work showed that the CO₂ absorption capacity was reduced at elevated temperatures, which agrees with the findings of the literature. For example, at 303.15 K and 1400 kPa, the CO₂ solubility of 2M ChOH:TETA−DMSO was 1.251 mol CO₂/mol DES and was reduced to 0.7566 mol/mol at a 333.15 K temperature.

Generally, non-aqueous organic solvents have significant advantages of low energy penalty and reduced corrosiveness as well as less amine degradation compared to aqueous solvents [27]. As reported in the literature, an amine-based non-aqueous solvent resulted in a reduction of energy consumption by up to 1.73 MJ/kg, which is 55% lower than that of an amine-based aqueous solvent [28,29,30]. Figure 5 shows a comparison between 2M ChOH:DETA-Sulfolane and 2M ChOH:DETA−DMSO at 313.15 K and various pressures. The results show that the 2M ChOH:DETA−DMSO mixture absorbed up to 1.137 mol CO₂/mol DES, 12.5% higher compared to 2M ChOH:DETA−Sulfolane. Similarly, Figure 5 shows a CO₂ absorption capacity of 2M ChOH:DETA and 2M ChOH:TETA mixed in sulfolane at 303.15 K and at the pressure range measured. From both observations on solubility in non-aqueous solvents, DMSO gives a better CO₂ absorption capacity compared to sulfolane. This is due to a higher physical solubility of CO₂ in DMSO as compared to sulfolane. A similar observation was reported in the measurement of CO₂ solubility in the binary solvent of aqueous DMSO and sulfolane with MDEA, which clearly showed that the CO₂ solubility in aqueous DMSO + MDEA was higher than sulfolane + MDEA [17]. Based on these observations, the CO₂ solubility of 2M ChOH:TETA−Sulfolane reached up to 1.2507 mol CO₂/mol DES, 12.5% higher compared to 2M ChOH:DETA−Sulfolane. This is attributed to the presence of a higher number of amine moieties in the TETA polyamine carbon chain compared to DETA polyamine, which was able to react and bond more CO₂ to the molecule. A similar observation was reported in the literature, which confirmed that increasing the number of amine groups from 2 amine functionalized (EDA) to 5 amine functionalized (TEPA) drastically increased the CO₂ loading capacity [31]. Due to an increase in the number of amine active sites, the CO₂ cyclic capacity and reaction rate of amine solutions increased, allowing for the formation of numerous species, such as carbamate and bicarbonate, under equilibrium conditions [32].

Furthermore, apart from 2M ChOH:DETA−DMSO, all other samples produced biphasic layers/phase separation after CO₂ absorption in a mixture with polyamine DES in the composition. Figure 6 demonstrates this phase separation. The formation of insoluble polyamine-carbamate in the non-aqueous mixture might be the source of the phase separation phenomenon. Polyamine-carbamate solubility in polar aprotic solvents such as DMSO and sulfolane is hampered by the presence of strong ionic and hydrogen bonding interactions. A similar observation was made for the CO₂ capture procedure utilizing a sulfolane combination of monoethanolamine and 1-butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide [21]. The formation of this phase separation could contribute to more energy efficiency during the solvent regeneration step by only regenerating the CO₂-rich portion of the solution, reducing the amount of transported material to be regenerated [33].

3.3. Fourier Transform Infrared Analysis

Table 2. FTIR peak assignment for the studied systems.

Assignment	2M ChOH:DETA-DMSO	2M ChOH:DETA-Sulfolane	2M ChOH:TETA-DMSO	2M ChOH:TETA-Sulfolane
	Frequency (cm−1)
−OH, −NH stretching	3419	3337	3315	3536
−NH bending	698, 663	581	656, 699	673
−CH2−OH bending	1572	1552	1562	1550
−C−N− bending	1230	1267	1276	1264
−C−O− bending	1023	-	1020	1035
Carbamates band	1306	1410	1313, 1348	1416
−CH2−, −CH3 stretching	1476, 1410	1469	1403, 1468	1461
stretching of NCOO−	1654	1641		1651

tabulates the major peaks and their assigned groups, while Figure 7a–d illustrate the FTIR spectra for 2M ChOH:TETA−DMSO, 2M ChOH:TETA−Sulfolane, 2M ChOH:DETA−DMSO and 2M ChOH:DETA−Sulfolane before and after CO₂ absorption. The presence of carbamate could be observed through the formation of the new peaks starting from 1100 to 1700 cm⁻¹. The spectra show the presence of carbonate peaks reflecting the formation of carbamate salt after absorption of CO₂, 1313, 1416, 1306, 1401 cm⁻¹, respectively. The presence of these peaks can be attributed to the nitrogen atom bonded to two imide groups, indicating the occurrence of a reaction between carbon dioxide (CO₂) and the nitrogen atom linked to the carbonyl group (C=O) within the amide group [34]. Moreover, the establishment of a new peak formation at 1738, 1651, 1654 and 1641 cm⁻¹, respectively, is due to the skeletal vibration of NCOO⁻. The formation of protonated amine may also contribute to the mentioned peak. It is worth mentioning that the individual amine groups present in TETA and DETA possess the potential to function as bases or participate in the formation of carbamates, encompassing monocarbamates, bicarbamates, tricarbamates, and tetracarbamates. All spectra show peaks that were found in the respective singular component; however, some peaks maybe be overlapped. An accurate assignment of all absorption bands in infrared spectra is challenging due to the extensive presence of numerous absorbing species, resulting in bands that are closely proximate, significantly broad, and often exhibiting a strong overlap.

3.4. Comparison with Another Study

Figure 8 shows a comparison between the CO₂ loading capacity of this study and that of the work by Mohammad et al. (2019), using 2M ChOH:TETA−DMSO and 2M GLY−DMAPA, respectively, at 303.15 K. For both samples, the CO₂ loading increases with an increasing pressure, and this is mostly in agreement with other published literature. At the highest measured pressure, the results indicated that the CO₂ absorption capacity of 2M ChOH:TETA−DMSO (1.2507 mol CO₂/mol absorbent) was higher than 2M DMAPA−GLY (0.720 mol CO₂/mol absorbent) [34]. The high CO₂ loading observed in 2M ChOH:TETA−DMSO solution is mainly due to the presence of two primary and two secondary amino groups in its chemical structure, which provide an active site for capturing CO₂, making TETA a high CO₂ loading species [35]. Moreover, the increasing trend in CO₂ absorption at a high pressure is attributed to the presence of more CO₂ molecules at elevated pressures of CO₂ as well as a higher kinetic energy, which resulted in a higher probability of collisions between CO₂ molecules and 2M ChOH:TETA−DMSO. The observed differences in solubility can be explained by the variations in the physical absorbent components, notably DMSO and glycerine.

4. Conclusions

A non-aqueous mixture of choline hydroxide-based deep eutectic solvent has been developed with an enhanced CO₂ absorption capacity. The implementation of DES as a viable alternative for commercial CO₂ absorbents appears advantageous due to its cost-effectiveness resulting from the utilization of inexpensive raw materials and a straightforward synthesis process. Furthermore, the incorporation of polyamine has demonstrated a significant enhancement in its capacity to capture CO₂, with the presence of multiple amine groups in the structure. The results demonstrated that 2M ChOH:TETA−DMSO exhibits a superior CO₂ loading capacity (mol CO₂/mol DES) compared to 2M ChOH:TETA−Sulfolane, 2M ChOH:DETA−DMSO and 2M ChOH:DETA−Sulfolane. This is due to the presence of a high number of amine groups in the TETA molecule compared to DETA and a higher physical solubility of CO₂ in DMSO compared to sulfolane. We also observed that the absorption capacity increased with an increasing pressure and decreased temperature. The highest CO₂ absorption by 2M ChOH:TETA−DMSO was observed at a partial pressure of 1400 kPa at 303.15 K. Furthermore, a phase separation was observed in most of the solutions post-CO₂ absorption process. This is due to the formation of insoluble carbamate salt, which was identified through FTIR analysis. The formation of a phase separation by the CO₂-rich layer of carbamate salt could contribute to a more energy-efficient regeneration process. This is possible via selectively transporting the CO₂-rich layer, thus reducing the amount of material transported during the regeneration process. By only regenerating the CO₂-rich phase of the absorbent, the total energy required for the regeneration process could be reduced. Moreover, with the incorporation of polyamine, a higher number of CO₂ was absorbed. However, future studies are necessary to evaluate the absorbent’s performance in real-world applications and to determine the optimal operating conditions.