Simulation study on carbone dioxide removal from Algerian natural gas by polystyrene membrane
https://doi-001.org/1025/17642242406416
Hammou Abdel Illah1*, Benguendouz Abdenour2, Bouterfa Asma3, Benzidane Abdel Kader 4
1 Laboratory of Beneficial Micro-Organisms, Functional Foods and Health (LMBAFS), Abdelhamid Ibn Badis University, Hocine Hamadou Street, Mostaganem 27000, Algeria
2,3 Laboratory of applied animal physiology, Abdelhamid Ibn Badis University, Mostaganem 27000, Algeria
4Mathematics department, faculty of exact sciences and computer science, University of Mostaganem 27000, Algeria
*Corresponding author: abdelillah.hammou@univ-mosta.dz
Received: 23/06/2025 ; Accepted: 24/10/2025
Abstract:
A simulation study was carried out on the use of tubular membrane for capturing CO2 from Algerian liquefied natural gas. A bed composed of dense microporous polystyrene membrane particles was considered. The Almeesoft simulator integrating the calculation of the Hysis membrane was used. A clogging time of 59.7 minutes with a maximum selectivity equal to 99.70 %, an adsorption capacity of 2327.16 mg/g and a transmembrane pressure of 2.76 bars were calculated. Additionally, selectivity, maximum gas pressure, maximum natural gas concentration, and adsorption models were considered. Particular focus was put on the permeability between the membrane skin and the Algerian natural gas.
Key Words: Simulation, decarbonation, Algerian natural gas, adsorption, PS membrane.
I. Introduction
Gas emissions emitted by different industries today constitute a great concern for both governments and international environmental protection organizations. The largest sources of gaseous pollutants remain the petroleum and associated petrochemical industries [1, 2]. To preserve the environment and improve public health, considerable efforts are constantly made, both at the legal and technical levels. The first makes it possible, via legal standards, to force polluters to include pollution issues in any market development, while the second studies possible scientific and technical approaches to reduce emissions and/or neutralize their negative effects. So far, many methods are considered for the treatment of gas emissions, including adsorption, condensation and absorption with or without chemical reactions [3].
Among these, although costly, the chemical reaction of toxic species to neutralize them from the gases emitted is the most effective method. Carbon dioxide is one of these toxic gases whose concentration must be reduced to the permitted level from natural gas. Governments and international organizations, including the United Nations, have strongly pushed for companies to reduce the carbone dioxide content of natural gas before its end use. Various limits have been set over the last two decades depending on the rules of the countries and areas where the gases are consumed, i.e. seas, land or cities.
A thorough review of the environmental literature reveals that the removal of carbone dioxide from natural gas is widely studied in the field of air pollution control. This gaseous contaminant becomes weakly acidic in the presence of traces of water. Consequently, this causes corrosion of equipment, pollutes the environment and has dramatic consequences on public health [4, 5]. The decarbonation of natural gas is a crucial step in gas processing industries. Note that a large part of the world’s natural gas reserves are not economically competitive because their high carbone dioxyde content exceeds 0.21% of carbon dioxide [6]. Many methods have been developed for the decarbonation of natural gas, including chemical neutralization and physical removal by adsorption onto a microporous bed of charcoal. The ultimate goal of these techniques is the fixation or isolation of a maximum amount of CO2 from natural gas. Membrane adsorption technology has progressed very quickly to become a promising alternative to the chemical reaction conventionally used to trap CO2 molecules from natural gas. Membranes based on inorganic and organic polymers have been used in the form of a bed of solid microporous particles [7].
In addition to the limited experimental research published on reducing the carbon dioxide content of natural gas, simulation has recently been potentially considered for the same purpose. This is primarily driven by the safety risks faced by experimental laboratory researchers investigating the transport properties of CO2. Huanghe Li et al. proposed a two-stage membrane-based process design, using a highly CO2-selective membrane (MB-I) for the first CO2 enrichment stage to 95% (dry base purity) and a highly CO2-permeable membrane (MB-II) for the second CO2 recycling stage, which targets economical carbon capture for coal-fired power plants when they achieve a CO2 capture rate of 90% with >95% CO2 product purity.
The first stage uses highly CO2-selective membranes to enrich the gas to >95% purity, while the second stage uses a highly CO2-permeable membrane to capture remaining CO2 to reach a total capture rate of 90%. This process design, optimized using Aspen Plus simulation, found that a critical CO2/N2 selectivity of 300–400 for the highly selective membranes is key for achieving economic and energy-efficient results [8].
Mahsa Bagi et al. investigated the essential factors of using hollow fiber membranes that affect CO2 removal efficiency. In the simulation, a finite element model for a ten-fiber membrane was used.
Each fiber is 175mm long with an inner radius of 0.75mm and an outer radius of 1.5mm. The liquid and gas flow rates were set between 100 and 800 ml min-1 and the adsorbent concentration was adjusted in the range of 200-1500 mol m-3 for monoethanolamine, piperazine (PZ) and ethylenediamine sorbents. Increasing the liquid flow rate, gas flow rate, and absorbent concentration results in an increase, decrease, and increase in efficiency, respectively. Thus, using PZ as an absorbent with a liquid concentration of 1080 mol m−3 and gas flow rates of 400 and 180 mL min−1, respectively, showed CO2 removal efficiency of >95% for an effective membrane length of 0.3 m [9].
In this work, a simulation work on the elimination of a maximum amount of carbone dioxide from natural gas using tubular polystyrene membrane. Almeesoft software-Hysis simulator was used to carry out the calculations. The approach was based on mono film theory using the Langmuir and Elovich models.
II. Results and discussion
The characteristics of the membrane, its module and the working conditions used for the calculations carried out are listed in table 1.
Table 1. Characteristics of the membrane, its module and the working conditions used.
From a PS polymer having a density of 1.04 g/cm3, the calculated mass (m) and volume (Vm) of a membrane having a thickness of 0.5 mm are 443.76 g and 426 cm3 respectively. A volume of gas (Vg) of 11.2 liters required to saturate the volume of the internal membrane was calculated from the product of the internal surface area and the length of the tubular membrane.
II.1. Adsorption Kinetics
Adsorption kinetics makes it possible to determine the contact time before equilibrium is reached as well as the average fluid speed. Indeed, by fixing the volumetric flow and neglecting any parasitic flow the variation of the pressure as a function of time is according to S. Cassette [10]:
P0=Pexp[-(Dv.t)/Vg] (1)
Where t is the contact time, Vg is the gas volume, Dv is the gas volumetric flow, P is the gas feed pressure and P0 is the gas permeate pressure.
Figure 1. Effect of contact time on the selectivity of the PS membrane.
At room temperature and under a low pressure permeate of 0.7 atm. The adsorption process progressed significantly and rapidly as indicated by the selectivity versus time (Fig. 1).
A selective CO2 removal of 98.32% is achieved after 40.25 min of contact time. This rapid adsorption process was expected because the membrane pores initially empty and are gradually saturated. This is indicated by the slow increase in selectivity in the time range from 40.25 to 59.7 min.
Contact times greater than 59.7 minutes showed no change in selectivity. It has stabilized at a value of 99.70%. This can be explained by the expected saturation of the pores with CO2 molecules. If the feed pressure is greater than the permeate pressure, an average fluid velocity of approximately 0.013 m h-1 is obtained from Ʋm = Dv/A; Where Ʋm is the average fluid velocity, Dv is the volumetric flow rate and A is the membrane surface area.
II.2. Adsorption Isotherm
Below a supply pressure of 13.79 atm. the selectivity increased significantly to reach a value of 99.32%. It slows down to reach a plateau at 99.70% corresponding to a supply pressure of 58.62 atm. At this stage all the pores were occupied and the membrane no longer retains CO2 molecules (fig.1).
If we consider the gas mixture to be an ideal gas, then we can write the concentration of CO2 as follows:
C = P/(R.T) (2)
The equations 2 and 7 were used to calculate the selectivity as a function of feed, permeate and retentate concentrations, respectively. From figure 2 we deduce at equilibrium a selectivity of 99.70%, a feed concentration of 59.145 g/l, a permeate concentration of 58.967 g/l and a retentate concentration as small as 0.177 g/l.
Figure.2. Effect of concentrations on the PS membrane selectivity.
Figure 3 shows the selectivity as a function of low permeate pressure. The selectivity decreases monotonically as the low pressure of the permeate increases due to saturation of the membrane pores. This clogging phenomenon restricts the mobility of CO2 molecules in the membrane.
Figure3. Effect of permeate low pressure on the selectivity of the PS membrane.
The effect of temperature on selectivity is carried out at a supply pressure of 30 atm. and a low permeate pressure of 0.7 atm. The selectivity decreases linearly with temperature. This behavior was expected as the pores enlarge due to heat treatment. In practice, this phenomenon has been attributed to the breakdown of secondary chemical bonds within the membrane (fig.4) [11].
Figure 4. Effect of temperature on the PS membrane selectivity.
II.3. Energies and the Adsorption Process
The working conditions used for the calculations carried out are presented in Table 2.
Table 2. The working conditions used.
| T (°C) | Permeate low Pressure P0 (atm) | Feed Pressure P (atm) |
| 0 | 0.7 | 30 |
| 25 | 0.7 | 30 |
| 50 | 0.7 | 30 |
| 75 | 0.7 | 30 |
The distribution coefficient Kd characterizing the affinity of the solute for the adsorbent can be translated into an equation by:
Kd=(Ci/Ce).(Vg/m) (3)
Where Ci and Ce are the initial and residual concentrations of solute in (mg/l) at equilibrium, m (gr) is the adsorbent mass and Vg (ml) is the volume of gas [12].
The free energy ΔG, enthalpy ΔH and entropy ΔS of CO2 adsorption on the PVC membrane can be calculated using equation 4 derived from equation 3.
ln Kd=(ΔS /R)-(ΔH/RT) (4)
Where R is the molar gas constant and T (k) is the temperature [13].
Figure 5. Effect of temperature on the sorption of CO2 by the PS membrane.
Plotting ln (Kd) versus 1000/T gives a straight line with the slope and origin corresponding to (-ΔH/R) and (ΔS/R), respectively (Fig.5). Table 3 lists the thermodynamic parameters calculated using the free enthalpy ΔG=ΔH-TΔS [13, 14].
Table 3. Calculated thermodynamic parameters.
| ∆G0 (KJ/mol) | |||||
| ΔS0 (J/mol.K) | ∆H0 (KJ/mol) | 273.15 K | 298.15 K | 323.15 K | 348.15 K |
| 68.780 | -1.026 | -19.813 | -21.533 | -23.522 | -24.971 |
• The negative value of ∆G° indicates that the process of elimination of CO2 molecules is spontaneous and that their adsorption by the membrane is favorable.
•The negative ∆H° reveals that the process is exothermic.
• ΔH°<40KJ/mol reveals that the process of CO2 fixation to the membrane is probably a physisorption phenomenon.
II.4. Adsorption Isotherm Models
Many authors have proposed theoretical or empirical models to describe the relationship between the adsorbate mass fixed at equilibrium (qe) and the corresponding concentration (Ce).
The quantity of solute adsorbed at equilibrium is given by:
qe = (Ci-Ce).(Vg/m)=ʂ.Ce (5)
Where ʂ is the selectivity, Vg (L) is the gas, m (g) is the mass of the adsorbate and Ce (mg/L) is the concentration of the adsorbate at equilibrium [15].
Figure 6. Langmuir adsorption isotherm of CO2 by the PS membrane.
Figure 7. Isotherm according to Elovich model.
In Figures 6 and 7 the adsorption isotherms are plotted according to the Langmuir and Evolich models, respectively.The calculated data used are listed in Table 4. The results confirm that the Langmuir model is the most appropriate to adapt to our study of adsorption of CO2 molecules. A regression of 0.92 with an adsorption capacity of 2327.16 mg/g was found.
Table 4. Parameters calculated from the Langmuir and Elovich models applied to the sorption of CO2 molecules by the PS membrane.
| Langmuir | Elovich | |||||
| qmax (mg/g) | k (g/mg) | R2 | α (mg.g-1.min-1) | Β ( g.mg-1) | R2 | |
| CO2 | 2327.16 | -0.0061 | 0.92 | 0.025 105 | 1.408 10-5 | 0.74 |
II.5. Intraparticle diffusion process
The kinetics of intraparticle diffusion is generally presented by: [16]
qt =Kid.t1/2+C (6)
Where Kid is the intraparticle diffusion rate constant and C is a constant. A plot of qt =ƒ (t1/2) is shown in Figure 8.
Figure 8. Intraparticle diffusion of CO2 molecules by the PS membrane.
A Kid intraparticle diffusion rate constant of 2.32 · 104 mg.g-1.min-1/2 and a C constant equal to -12.62 104 mg.g-1 were derived from the slope and intercept, respectively.
II.6. Clogging of membranes
Clogging is a process of filtration and permeation on membranes. The formation of material in dispersed form on the surface of the membrane leads to a reduction in the permeation process. This is due to saturation of the membrane pores. In Figure 9, the gas flow crossing the membrane is plotted as a function of the transmembrane pressure (TMP):
TMP = (Pfeed+Pretentate)/2-(Ppermeate) (7)
Figure 9. Effect of transmembrane pressure on the stationary flow of natural gas.
The filtration flow increases linearly as a function of the transmembrane pressure up to 2.76 bars. Then, it is stabilized at a constant equilibrium flow due to a clogging phenomenon which restricts the flow [17-18]. A solid layer (cake) forms on the surface of the membrane, which affects the flow of natural gas through the pores of the membrane. The permeation flow is linked to the transmembrane pressure according to:
J=Q/S=Lp.TMP (8)
Where J is the volumetric permeation flux density, Q is the gas permeation rate, S is the membrane surface area, and Lp is the permeability of the membrane skin. Using the results in Figure 9, it is possible to deduce the permeability of the membrane skin as Lp= 0.469 m3.m-2.h-1.bars-1.
III. Conclusion
The presence of carbon dioxide (CO2) in Algerian natural gas is corrosive to processing facilities, harmful to public health and devastating to the environment after marketing and consumption. Algerian natural gas contains 0.21% carbon dioxide (CO2). However, to be competitive, this content must be reduced to an acceptable limit. The present investigation was carried out as part of research into reducing the quantity of CO2 in Algerian natural gas by simulating the adsorption process on polystyrene membrane. The kinetic calculation revealed an optimal contact time of 59.7 min, which resulted in a maximum selectivity of 99.70% at a feed pressure of 58.62 bar. In terms of concentration 58.97 g/l of CO2 were permeated before clogging. Longer times did not improve the selectivity because the pores of the polymer membrane (polystyrene) were saturated. In practice, this information could be very useful for setting up a procedure for either changing or passing the saturated membrane with a virgin. Regarding temperature, careful measures should be taken to avoid heat generation during the separation process, as this reduces selectivity. To avoid clogging of the membrane, it is recommended to work with a transmembrane pressure lower than 2.76 bars and a skin permeability of the membrane not exceeding 0.469 m3.m-2.h-1.bars-1.
IV. Acknoledgments
I would like to express my sincere gratitude to my co-autors, Abdenour BENGUENDOUZ, Asma BOUTERFA and BENZIDANE Abdel Kader for their valuable contribution and assistance in the completion of this work.
V. References
1. Weiwang Lim, Wen He, Ji Ma, Shabi Ul Hassan, Jingcheng Du, Qian Sun, Dong Cao, Jian Guan, Hongjun Zhang, Jiangtao Liu a, Membranes with hollow bowl-shaped window for CO2 removal from natural gas, Advanced Membranes, 5, 2025, 100129.
2.Mohamed Loucif, Mourad Bidaoui, Ahmed Fedag, Mohammed Amin Chemrak, Yassine Chaker, Meriem Bouabib, Asma Deba, Abdallah Ben Rhaiem, Abderrazek Oueslati, Abdelfattah Mahmoud, Frédéric Boschini,Fixed-bed natural gas CO2 adsorption performance of ZSM-5 zeolites impregnated with amines, Studies in Engineering and Exact Sciences, 5(2), 2024, 01-28.
3. Lubna Ghalib, Chapter 1 – Impact of presence of CO2 in natural gas streams, Nanomaterials for Carbon Dioxide Capture and Conversion Technologies, 2023, ISBN:978-0-323-89851-5, 1-24.
4. Hongfang Guo, Jing Wei, Yulei Ma , Jing Deng, Shouliang Yi, Bangda Wang, Liyuan Deng, Xia Jiang a, Zhongde Dai,Facilitated transport membranes for CO2/CH4 separation – State of the art, Advanced Membranes, 2, 2022, 100040.
5. Yick Eu Chew, Zulfan Adi Putra, Dominic C.Y. Foo, Process simulation and optimisation for acid gas removal system in natural gas processing, Journal of Natural Gas Science and Engineering, 107,2022, 104764.
6. Kangyin Dong, Renjin Sun, Gal Hochman, Xiangang Zeng, Hui Li, Hongdian Jiang, Impact of natural gas consumption on CO2 emissions: Panel data evidence from China’s provinces, Journal of Cleaner Production, 162, 2017, 400-410.
7. Kangyin Dong, Renjin Sun, Gal Hochman, Xiangang Zeng, Hui Li, Hongdian Jiang, Impact of natural gas consumption on CO2 emissions: Panel data evidence from China’s provinces, Journal of Cleaner Production,162, 2017, 400-410.
8. Huanghe Lie, Fan wang, Shigang Li, Miao Yu, Two-stage membrane-based process utilizing highly CO2-selective membranes for cost and energy efficient carbon capture from coal flue gas: A process simulation study, Journal of Membrane Science, 669, 2023, 121259.
9. Mahsa bagi, Mohammad Vakili Razlighi, Mina Shanbedi,Alamgir Karim, A Comprehensive Parametric Study on CO2 Removal from Natural Gas by Hollow FiberMembrane Contactor: A Computational Fluid Dynamics Approch, Chemical Engeneering Technology, 47(4), 2024, 732-738.
10. Simone, C.; ‘Techniques du vide’, Report; Polytechnique Paris, 2011, pp102.
11. Stephen Okiemute Akpasi, Yusuf Makarfi Isa, Effect of operating variables on CO2 adsorption capacity of activated carbon, kaolinite, and activated carbon – kaolinite composite adsorbent, Water-Energy Nexus, 5, 2022, 21-28.
12. Lifang Liu, Jiaqi Zhang, Gening Bi, Yufei Liang and Runping Han, Adsorption of 2,4-Dichlorophenol onto Cationic Surfactant-Modified Wheat Straw in Batch Mode, Asia-Pacific Engineering and Technology Conference, APETC, 2017, 754-759, doi: 10.12783/dtetr/apetc2017/11078.
13. Tianxiang Guo, Zeng Fan, Yarong Du, Junpeng Xu, Lingfeng Kong, Yuanfeng Pan, Huining Xiao, Qing Xie,Thermodynamics of CO2 adsorption on cellulose-derived biochar prepared in ionic liquid, The Canadian Journal of Chemical Engeneering, 99(9), 2021, 1940-1961.
14 . Major Mabuza, Kasturie Premlall, Michael O. Daramola, Modelling and thermodynamic properties of pure CO2 and fue gas sorption data on South African coals using Langmuir, Freundlich, Temkin, and extended Langmuir isotherm models, International Journal of Coal Science & Technology, 2022, doi.org/10.1007/s40789-022-00515-y
15. Hammou A.I., Harrats C., Djennad M., Simulation study on sulfur removal from algerian natural gas using adsorption on high density polyethylene membrane, Algerian J. Env. Sc.Technology, 6 (1),2020, 1200-1206.
16. Cininta Nareswari, Utari Zulfiani, Afdhal Junaidi, Triyanda Gunawan, Nurul Widiastuti, Adsorption of CO2 and H2 on the polymer-based membrane from High-densit Polyethylene (HDPE) Plastic, South African Journal of Chemical Engineering, 49, 2024, 105-113.
17. M. Dencheva-Zarkova, J. Genova, I. Tsibranska, Effect of pressure and cross-flow velocity on membrane behaviou in red wine nanofiltration, Journal of Physics, 2023, 1-5.
18. Sari, M.; John, H.; Marianne, N. Critical flux in ultrafiltration of myoglobin and baker’s yeast. Journal of Membrane Science 196(1) (2002) 13-25.