Abstract

Ionic liquids (ILs) are proposed as potential “green” solvents with remarkable properties. Deep eutectic solvents (DESs) are a new type of ILs with additional properties, such as higher biodegradability and a lower price. ILs and DESs are “green” absorbents for various gas separations, such as CO2/N2, CO2/H2/CO, H2S/CH4, and N2O/N2. Due to their large number, the screening of ILs is crucial. Although ILs with high absorption capacities were screened using gas solubility and selectivity, it is important to consider the energy and solvents used in the process. In this paper, the absorbent amount and the energy consumption were used for screening absorbents for various gas separation processes. The results reveal that physical IL [Bmim][DCA] and chemical IL [Eeim][Ac] are screened for CO2/N2 and CO2/H2/CO separation, physical IL [Omim][PF6] for H2S/CH4 separation, and physical IL [P66614][eFAP] for NO/N2 separation. The screened ILs offer some advantages over commercial absorbents in terms of lower energy consumption or amount.

1. Introduction

Every year, large amounts of gases are emitted into the atmosphere, with CO2 being the primary greenhouse gas [1, 2], N2O having a potential impact on global warming that is 310 times greater than that of CO2 [3], and H2S being one of the highest sulfur-containing compounds. Excess emissions of CO2 and N2O result in global warming and climate change, while H2S emissions result in acid rain. Therefore, it is necessary to identify effective measures to mitigate the emission of CO2, N2O, and H2S from different sources.

Carbon capture and storage (CCS) can be used to reduce CO2 emissions. Meanwhile, biomass syngas from biomass gasification is used in transportation as biofuels. CO2 separation is required for CCS and biomass syngas purification. The current CO2 separation technologies have several shortcomings, such as high energy consumption [4], corrosion degradation, and/or large-scale operations, which affect removal efficiency.

The primary sources of anthropogenic N2O emissions are nylon production, nitric acid production, and vehicle exhaust emissions. Many N2O emission reduction technologies have been developed, such as thermal decomposition and catalytic decomposition. However, they may be constrained by the large energy consumption and increasing CO2 emissions.

There are several technologies for H2S separation, such as absorption, oxidation, and adsorption [5]. Absorption technology is extensively used for H2S separation, whereas aqueous alkanolamine solutions are currently employed in industrial natural gas treatment and sweetening plants [6]. However, the application of H2S absorption technology is impacted, similar to CO2 separation, by intensive energy consumption and/or large-scale operations.

Therefore, new gas separation processes for CO2, N2O, and H2S must be developed to curb climate change and protect the environment. Ionic liquids exhibit remarkable properties for gas separation, such as high solubility for CO2, N2O, and H2S. Although ILs have been used in studies related to gas separation [710], the large-scale commercial usage of ILs is limited by their high toxicity, poor biodegradability, and high cost. Deep eutectic solvents have low toxicity, biodegradability, and low cost and are proposed as promising CO2 absorbents [1114].

In our previous work, Gibbs free energy change (∆G) was employed to evaluate the performance of solvents [15, 16] for separating CO2 from biogas and NH3 from synthetic ammonia purge gas [17]. However, the effective absorbents for various gas streams with different conditions were not determined. In this study, thermodynamic analysis was utilized for gas separation using pure ILs and DESs. For CO2 streams, effluent gases, kiln gas, and biomass syngas were chosen. For the H2S stream, high-sulfur natural gas was chosen. For the N2O stream, adipic acid off-gas was chosen. For various gas streams, several absorbents were screened as potential gas absorbents. Furthermore, the relationship between the performances, the properties, and the critical properties of gas streams was investigated. All of these are employed in the development of new gas separation technologies.

2. Gas Separation and Thermodynamic Model

2.1. Gas Streams

In this work, effluent gases were chosen as CO2 streams due to their large emission amounts and low CO2 concentration, whereas kiln gas was chosen due to its high CO2 concentration. Biomass syngas is chosen as the CO2 stream for CO2/CO/H2. The high-sulfur natural gas is used as the H2S stream for H2S/CH4, and the adipic acid off-gas is used as the N2O stream for N2O/N2. Table 1 lists the typical conditions of different gas streams.

2.2. Gas Separation Process

The gas separation process using liquid absorbents is shown in Figure 1. The absorbent quantity (mabs, g abs·g X−1) can be calculated using the following equation:where Mabs and Mx are the molecular weights of the absorbent and X gas, respectively, and g·mol−1. xa and xs are the molar ratios of the X gas in the absorption tower and desorber, respectively.

The energy consumption required in the gas separation process can be calculated as shown inwhere nx denotes the molecular weight of the X gas in moles. Hx denotes the Henry constant of X gas in the absorbents in bar units. Pa and P1 denote the pressure of the absorption tower and the initial pressure, respectively, in bar units. Cp, abs represents the isobaric heat capacities of the absorbents in J·mol−1·K−1. Ta and Ts denote the temperatures of the absorption tower and desorber, respectively.

2.3. Theory

In thermodynamics, ∆G was used as the evaluation criteria to determine if the isothermal reversible process was spontaneous. As shown in Figure 2, the separation process is nonspontaneous, and the Gibbs free energy change for System 1 is above zero (∆G1 > 0). Six reversible processes were designed, and the summation is represented as ∆G2 for Surrounding 2. System 3 is composed of System 1 and Surrounding 2. Moreover, the addition of energy and absorbents causes ∆G2 to be negative. In Figure 2, ∆Gcomp represents the isothermal reversible compression process, ∆Gabs represents the reversible absorption process of X absorption, ∆GT represents the reversible adiabatic expansion and reversible temperature increasing process of the solution, ∆Gdes represents the reversible X gas desorption process, ∆Gexp represents the isothermal reversible expansion of the other gas, and ∆GT’ represents the reversible adiabatic expansion and the reversible isothermal compression of the X gas. The optimal operating conditions are achieved when ∆G3 is equal to 0, and the performance of different absorbents can be evaluated. Therefore, using the thermodynamic analysis as proposed in our previous work, the number of absorbents and the energy consumption are combined by Gibbs free energy change [12, 13].

G for different systems and processes can be calculated using the equations given in equations (3)–(10).where n is the number of moles in the gas stream, and yi represents the molar fraction of component i in the stream.where and are the G values of the gaseous component i at (T, P) and (Ta, Pa), respectively, and they are calculated using the formula G = H − TS. The NIST standard reference database provides the values of H and S for the gas components.where Hx denotes Henry’s constant and Kx denotes the chemical reaction constant of X gas in absorbents.where ΔfHj, 298.15 K represents the standard enthalpy change in the formation of j in the absorbents at 298.15 K; Sj (298.15 K) represents the standard molar entropy of j at 298.15 K; and Cp,j represents the isobaric heat capacity of j in the absorbents.where yx is the mole fraction of X gas in the gas stream, (Ta,yxPa) is the G value of the gaseous component i at (Ta, yxPa).where is the G value of the gaseous component i at (Ts, Ps).

3. The Properties of Absorbents

The properties of liquid absorbents, including Henry’s law constants of different gases in ILs/DESs, reaction equilibrium constants, the density of ILs/DESs at 298.15 K, and the heat capacity of the ILs/DESs, have been collected and listed in Table 2. In conventional ILs/DESs, the uncertainty in Henry’s law constants of CO2, H2S, and N2O was estimated to be ±8 bar, ±0.4 bar, and ±0.07 bar, respectively. Henry’s law constants and reaction equilibrium constants were correlated for some physical ILs and the chemical ILs/DESs as per the CO2 solubility data described in our previous work [11]. The uncertainties in the densities and the heat capacities of the conventional ILs/DESs were ±63 J·mol−1·K−1 and ±13 g·cm−3, respectively, based on the experimental data presented in Table 2. The names and molar weights of absorbents are listed in Appendix A (Supplementary data available here).

4. Results and Discussion

The desorption temperatures of absorbents in this study are identical to those in our previous work [12], which are 299.15–323.15 K for physical absorbents, 299.15–345.15 K, and 299.15–353.15 K for chemical DES and ILs, respectively. The absorption pressure is iterated until ∆G3 = 0, and the optimal conditions, absorbent amount, and energy consumption are obtained.

4.1. Analysis of Gas Absorbents
4.1.1. Absorption Pressure

As shown in Figures 3 and 4, [BmPy][FAP] represents the minimum Pa, which is 16.16–87.77 bar, 8.75–45.65 bar, and 7.66–44.98 bar in 299.15–323.15 K for effluent gases, kiln gas, and biomass syngas, respectively. Among the four chemical absorbents that were analyzed, [Eeim][Ac] exhibits the lowest absorption pressures for the three CO2 streams, with values of 70.60–137.96 bar, 36.00–74.44 bar, and 41.98–123.37 bar within 315.15–345.15 K, respectively. Among the ten physical absorbents that were analyzed for N2O/N2 separation, [Omim][PF6] exhibits the lowest value of Pa in the temperature range of 299.15–323.15 K with pressure values in the range of 3.97–13.49 bar. [Bmim][FAP] exhibits the lowest Pa within 299.15–323.15 K with values of 4.40–22.25 bar among the eight physical ILs that were analyzed for H2S/CH4 separation.

As illustrated in Figure 3, the absorption pressures of different CO2 absorbents in physical absorbents are greater than those of chemical ILs/DES for CO2 streams. For the same IL [Bmim][BF4], the absorption pressures for different gas streams are as follows: effluent gases ( = 0.12) > kiln gas  = 0.25) > adipic acid off-gas ( = 0.45) > biomass syngas ( = 0.3) > biogas ( = 0.4) > high-sulfur natural gas ( = 0.32) > synthetic ammonia purge gas ( = 0.45). The key factors influencing the absorption pressure are gas concentration and gas solubility in absorbents. The absorption pressures increase with increasing gas concentrations in the gas stream and decreasing gas solubilities. Meanwhile, the absorption pressure shows an increasing trend as gas solubility decreases. Due to the low solubility of N2O in [Bmim][BF4], adipic acid off-gas exhibits a higher absorption pressure than biomass syngas. However, due to the high concentration of H2S, high-sulfur natural gas shows a lower absorption pressure than biogas.

4.1.2. Amount of Absorbent

Figures 5 and 6 show the absorbent amounts. As the desorption temperature rises, the amount of absorbents decreases, especially at low desorption temperatures.

The physical [Bmim][DCA] exhibits the lowest amount with values of 44.22–11.15 g·gCO2−1, 36.60–9.62 g·gCO2−1, and 32.26–6.84 g·gCO2−1 within the range of 300.15–323.15 K for effluent gases, kiln gas, and biomass syngas, respectively, in the 30 physical absorbents. The chemical Ch[Pro]/PEG200 (1 : 2) represents the minimum values of mabs, which are 13.39–5.62 g·gCO2−1, 14.60–4.63 g·gCO2−1, and 9.35–1.44 g·gCO2−1 in 315.15–345.15 K, respectively. [P66614][FAP] and [Omim][PF6] exhibited the lowest amount with values of 104.59–6.71 g·gN2O−1 and 4.72–14.74 g·gH2S−1 for adipic acid off-gas and high-sulfur natural gas, respectively, within the range of 300.15–323.15 K.

Due to differences in CO2 solubility, the amounts of CO2 absorbents follow the order of chemical absorbents < physical absorbents. Generally, the solubility of CO2 in chemical absorbents is greater than that in physical absorbents. For different CO2 streams with the same absorbents, the quantity has the following order: high-sulfur natural gas ( = 0.32) > synthetic ammonia purge gas ( = 0.45) > effluent gases ( = 0.12) > kiln gas ( = 0.25) > biomass syngas ( = 0.3) > biogas ( = 0.4) > adipic acid off-gas ( = 0.45). The major reason for the significant amount of absorbents is the high gas solubility in high-sulfur natural gas and synthetic ammonia purge gas.

4.1.3. Energy Consumption

In Figures 7 and 8, the energy consumption is displayed. The energy consumption increases with an increase in the desorption temperature. The physical [Hmpy][NTf2] displays the lowest Qtot for the CO2 separation process in the 30 physical absorbents, with values of 1.94–2.87 GJ·ton CO2−1, 1.02–1.49 GJ·ton CO2−1 and 0.89–1.21 GJ·ton CO2−1 for effluent gases, kiln gas, and biomass syngas, respectively, within the range of 299.15–323.15 K. The chemical [Eeim][Ac] exhibits the lowest energy consumption in the four chemical absorbents, with values of 2.48–3.70 GJ·ton CO2−1, 1.40–2.27 GJ·ton CO2−1, and 1.24–1.92 GJ·ton CO2−1 respectively, within 303.15–345.15 K. The physical IL [Omim][PF6] exhibits the lowest energy consumption for the separation of H2S from high-sulfur natural gas in the 10 physical ILs, with values of 1.05–1.44 GJ·ton H2S−1 within 299.15–323.15 K for high-sulfur natural gas. [P66614][FAP] shows the lowest energy consumption for N2O separation in the 8 physical ILs, with values of 0.65–0.95 GJ·ton H2S−1 within 299.15–323.15 K for adipic acid off-gas.

Due to the high heat of chemical absorbents, the energy consumption of chemical absorbents is larger than that of physical absorbents. The CO2 concentrations have an impact on the energy consumption for various CO2 streams. The energy consumption sequence is as follows: effluent gases ( = 0.12) > synthetic ammonia purge gas ( = 0.45) > kiln gas ( = 0.25) > biomass syngas ( = 0.3) > high-sulfur natural gas ( = 0.32) > biogas ( = 0.4) > adipic acid off-gas ( = 0.45). The energy consumption for separating CO2 from effluent gases (CO2/N2) is the largest due to the low CO2 molar ratio, and the energy consumption for separating N2O from adipic acid off-gas is the lowest due to the highest N2O solubility and a high amount of absorbents.

4.2. Screening Absorbents

The screening criteria for absorbents include both absorbent amounts (mabs) and energy consumption (Qtot). The physical [Bmim][DCA] was screened between 299.15–323.15 K with low amounts of the absorbent (<190 g·gCO2−1 for effluent gases, <150 g·gCO2−1 for kiln gas, <140 g·gCO2−1 for biomass syngas) and low energy consumption (<2 GJ·tonCO2−1 for effluent gases, <1 GJ·tonCO2−1 for kiln gas, <0.9 GJ·tonCO2−1 for biomass syngas). The chemical [Eeim][Ac] was screened with a lower amount of ILs (<43 g·gCO2−1 for effluent gases at 303.15–345.15 K, <142 g·gCO2−1 for kiln gas at 299.15–345.15 K, <68 g·gCO2−1 for biomass syngas at 300.15–345.15 K). A lower amount of ILs (<293 g·gH2S−1) was used to screen [Omim][PF6] for high-sulfur natural gas at 299.15–323.15 K. A lower amounts of ILs (<108 g·gN2O−1) was used to screen [P66614][FAP] for adipic acid off-gas at 299.15–323.15 K. The uncertainties of Henry’s law constant, heat capacity, and density were taken into consideration while estimating the uncertainties in the iterated results. The uncertainties in the absorption pressure, the required amount of ILs, and the energy expended for CO2 separation were estimated to be ±3.97 bar, ±2.47 g·gCO2−1, and ±0.09 GJ·tonCO2−1, respectively. The uncertainties in the absorption pressure, the required amount of ILs, and the energy expended for H2S separation were estimated to be ±0.41 bar, ±0.37 g·gH2S−1, and ±0.003 GJ·tonH2S−1, respectively. The uncertainties in the absorption pressure, the required amount of ILs, and the energy used for N2O separation were estimated to be ±0.12 bar, ±0.17 g·gN2O−1, and ±0.04 GJ·tonN2O−1, respectively. Table 3 lists the screened absorbents and their values of Ts, Pa, mabs, and Qtot.

4.3. Comparison with Commercial Absorbents

The superiority of the screened absorbents is demonstrated by comparisons with commercial absorbents. For CO2 separation, the chemical [Eeim][Ac] in this study is compared with 30% MEA and 30% MDEA. [Bmim][DCA] is compared with DEPG for CO2 separation. Furthermore, for H2S separation, [Omim][PF6] is compared with water. The findings of Pa, mabs, and Qtot of commercial absorbents for gas separation from effluent gases, kiln gas and biomass syngas, biogas, high-sulfur natural gas, and adipic acid off-gas are displayed in Table 4 with the set desorption temperature.

The results of mabs and Qtot for [Bmim][DCA] and DEPG are displayed in Figures 9 and 10. Figure 11 displays the predicted results of mabs and Qtot for chemical [Eeim][Ac] at 338.15 K, 30 wt% MEA at 393.15 K, and 30 wt% MDEA at 393.15 K.

Based on the comparisons of the performances of physical absorbents for CO2 separation shown in Figures 9 and 10, both energy consumption and the amount required for DEPG are lower than those of [Bmim][DCA] when Pa is 12.5 bar and 25 bar, respectively. This suggests that the performance of the screened ILs is not superior to that of commercial physical absorbents for DEPG. Based on the comparison of the performances of chemical absorbents for CO2 streams in Figure 11, [Eeim][Ac] displays a larger absorbent amount and lower energy consumption than those of 30 wt% MDEA and 30 wt% MEA. Assuming that the cost of [Eeim][Ac] is high, the screening of chemical ILs for CO2 separation must be further researched. However, the volatility of [Eeim][Ac] can be ignored.

From the comparison of the performances of physical absorbents for the H2S stream in Figure 12, [Omim][BF4] exhibited higher energy consumption and a lower amount of absorbents. The comparison of [Omim][BF4] and DEPG shows that the performance of physical IL is not superior to that of DEPG. The screened ILs have the advantage of low energy consumption and nonvolatility, although they have a higher amount than chemical CO2 absorbents (30 wt% MDEA and 30 wt% MEA). The physical H2S absorbent consumes less energy than water.

The cost of conventional ILs is significantly higher than that of commercial CO2 absorbents when operational costs and solvent costs are considered. For instance, the estimated costs of water, DEPG, MEA, and MDEA were estimated to be 0.3 × 10−3 US$·kg−1 [134], 3.5 US$·kg−1 [135], 1.25–2.25 US$·kg−1 [136, 137], and 20 US$·kg−1 [138], respectively, while the cost of the used commercial ILs was estimated as 6–13.5 US$·kg−1 [136]. Both the price and the amount needed of conventional ILs are higher than those of commercial gas absorbents; therefore, the advantages of ILs are located at nonvolatility, low energy use for chemical absorbents, and nondegradability. It is needed to develop process simulation with consideration of solvent degradation to compare conventional ILs and commercial gas absorbents. Furthermore, with technological advancement, the cost of ILs will drop, and the cost difference between conventional ILs and commercial absorbents will reduce.

5. Conclusions

In this study, green absorbents were analyzed based on ∆G to screen the absorbents for CO2 separation from effluent gases, kiln gas, and biomass syngas; H2S separation from high-sulfur natural gas (H2S/CH4); and N2O separation from adipic acid off-gas (N2O/N2).

[Bmim][DCA] and [Eeim][Ac] were screened for CO2 separation using mabs and Qtot as criteria. [Omim][PF6] and [P66614][FAP] were screened for H2S separation and N2O separation, respectively. According to comparisons between the screened absorbents and the DEPG, the physical IL has a higher energy consumption and higher mabs than the DEPG. When compared with 30 wt% MDEA and 30 wt% MEA, the chemical IL has a lower Qtot and is involatile. Lower energy consumption and a higher amount of H2S separation were observed when screened [Omim][BF4] was compared to H2O.

Due to the differences in gas concentration in gas streams and gas solubility in absorbents, the amount of absorbents has the following order: high-sulfur natural gas > synthetic ammonia purge gas > effluent gases > kiln gas > biomass syngas > biogas > adipic acid off-gas. The order of energy consumption is as follows: effluent gases > synthetic ammonia purge gas > kiln gas > biomass syngas > high-sulfur natural gas > biogas > adipic acid off-gas.

Data Availability

The data used to support the findings of this study are from previously reported studies and datasets, which are cited in the article. The enthalpy and entropy of gas components are available and can be found at NIST Standard Reference Data (https://webbook.nist.gov/chemistry/fluid/).

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was supported by the Scientific and Technological Key Project in Henan Province (grant no. 22A530004), Ph.D. Research Startup Foundation of the Zhengzhou University of Light Industry (grant nos. 2017BSJJ029 and 2020BSJJ018), and the National College Students’ Innovation and Entrepreneurship Training Program (grant no. 202210462074).

Supplementary Materials

Appendix A: supplementary data associated with this article. (Supplementary Materials)