Abstract

Increasing concentrations of greenhouse gases (GHGs) such as CO2 in the atmosphere is a global warming. Human activities are a major cause of increased CO2 concentration in atmosphere, as in recent decade, two-third of greenhouse effect was caused by human activities. Carbon capture and storage (CCS) is a major strategy that can be used to reduce GHGs emission. There are three methods for CCS: pre-combustion capture, oxy-fuel process, and post-combustion capture. Among them, post-combustion capture is the most important one because it offers flexibility and it can be easily added to the operational units. Various technologies are used for CO2 capture, some of them include: absorption, adsorption, cryogenic distillation, and membrane separation. In this paper, various technologies for post-combustion are compared and the best condition for using each technology is identified.

1. Introduction

There are ten primary GHGs including water vapor (H2O), carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O) that are naturally occurring. Perfluorocarbons (CF4, C2F6), hydrofluorocarbons (CHF3, CF3CH2F, and CH3CHF2), and sulfur hexafluoride (SF6), are only present in the atmosphere due to industrial processes. Water vapor is the most important, abundant and dominant greenhouse gas, and CO2 is the second-most important one (Table 1). Concentration of water vapor depends on temperature and other meteorological conditions, and not directly upon human activities. So it was not indicated in Table 1 [13].

CO2 is the primary anthropogenic greenhouse gas, accounting for 77% of the human contribution to the greenhouse effect in recent decade (26 to 30 percent of all CO2 emissions). Main anthropogenic emissions of CO2 come from the combustion of fossil fuels. CO2 concentration in flue gases depends on the fuel such as coal (12–15 mol-% CO2) and natural gas (3-4 mol-% CO2). In petroleum and other industrial plants, CO2 concentration in exhaust stream depends on the process such as oil refining (8-9 mol% CO2) and production of cement (14–33 mol-% CO2) and iron and steel (20–44 mol-%). From 2004 to 2011, global CO2 emissions from energy uses were increased 26% (Figure 1) [410]. Figure 2 indicates that power plant (55% of global CO2 emissions), transportation (23%), and industry (19%) have highest share in the CO2 emission in USA. Cement and petrochemical plants are two major industries for CO2 emission, such that cement industry contributes about 5% to global anthropogenic CO2 emissions. Also, petrochemical industries are a large share of CO2 emission; for example, only in Iran, petrochemical industries emission was about 15 Mton CO2/year [1116].

The Kyoto Protocol is the first international agreement on emissions of GHGs. In this protocol, industrialized countries agreed to stabilize or reduce the GHGs emissions in the commitment period 2008–2012 by 5.2% on average (compared to their 1990 emissions level). Overall, the result of global CO2 emissions (Figure 1) shows the failure of Kyoto protocol; therefore, in 2011 Durban COP meeting, this protocol was extended until 2017. Several countries with high GHGs emission like China, India, Brazil, and even Iran have added to this Protocol. Intergovernmental Panel on Climate Change (IPCC) predicted the atmosphere may contain up to 570 ppmv CO2 by the year 2100, causing a rise of mean global temperature and sea level around 1.9°C and 38 m, respectively [15, 1720]. Given that the earth’s average temperature continues to rise, Intergovernmental Panel on Climate Change (IPCC) stated, global GHG emissions must be reduced by 50 to 80 percent by 2050 to avoid dramatic consequences of global warming [2123].

Carbon capture and storage (CCS) is the most indicated technology to decrease CO2 emission from fossil fuels sources to atmosphere. Also, CO2 separated from flue gases can be used in enhanced oil recovery (EOR) operations where CO2 is injected into oil reservoirs to increase mobility of oil and reservoir recovery [24, 25]. Pure CO2 has many applications in food/beverage and different chemical industries such as urea and fertilizer production, foam blowing, carbonation of beverages and dry ice production, or even in the supercritical state as supercritical solvent [2628].

From this definition, CCS consists of three basic stages: (a) separation of CO2, (b) transportation, and (c) storage. Operating costs of these stages have been estimated in 2008:(i)CO2 separation from exhausting gases: 24 to 52 €/ton-CO2,(ii)transportation to storage location: 1 to 6 €/ton-CO2 per 100 km,(iii)storage: −28 to 42 €/ton-CO2.

Therefore, CO2 separation is a major stage in CCS. The CCS total costs can vary from −3 to 106 €/ton-CO2 (negative values are expected for the injection of CO2 in EOR). There are three major approaches for CCS: pre-combustion capture, oxy-fuel process, and post-combustion capture (Figure 3) [25, 30, 31].

Pre-combustion capture involves reaction of a fuel with oxygen or air and in some cases steam to produce a gas mainly composed of carbon monoxide and hydrogen, which is known as synthesis gas (syngas) or fuel gas. The produced carbon monoxide is reacted with steam in a catalytic reactor, called shift converter, to give CO2 and more hydrogen. CO2 is then separated, usually by cryogenic distillation or chemical absorption process, resulting in a hydrogen-rich fuel that can be used in many applications, such as furnaces, gas turbines, engines and fuel cells [32, 33].

A main advantage of post-combustion is the higher CO2 concentration and pressure achieved in the output stream. The main disadvantage of pre-combustion capture is system needs long-term development in a number of enabling technical areas to achieve targeted efficiency towards a hydrogen economy. This disadvantage has limited application of this approach and increased investments costs of pre-combustion capture [34, 35].

In oxy-fuel combustion, nearly pure oxygen is used for combustion instead of ambient air, and this results in a flue gas that is mainly CO2 and H2O, which are separated by condensing water. Three major advantages of this method are high CO2 concentration in output stream (above 80% v/v), high flame temperature, and easy separation of exhaust gases. The major disadvantages of oxy-fuel combustion are high capital cost and large electric power requirement to separate oxygen from air [3638].

The principle of post-combustion capture is CO2 separation from flue gas after combustion. Generally, the CO2 in flue gas is diluted (8–15% CO2) with inert gases such as nitrogen, argon, and water in addition to oxygen. Flue gases are normally at atmospheric pressure and high temperatures (between 320 K and 400 K) [3941]. Post-combustion capture does not require expensive technologies such as syngas separation, hydrogen turbine, fuel cell. Post-combustion capture is the most important to prevent CO2 emissions, because it offers flexibility and does not need to change combustion cycle. If the capture plant shuts down, the power plant can still operate [42, 43]. Major disadvantage of this method is unfavorable condition of flue gases.

Because of the importance in selecting suitable process for CO2 separation, in this research various technologies for this purpose have been focused.

2. CO2 Separation Technologies

Based on economical and environmental considerations, it is necessary to apply efficient and suitable technology for CO2 separation with low operating cost and energy consumption. Up to now, there are several gas separation technologies being investigated for post-combustion capture, namely, (a) absorption, (b) adsorption, (c) cryogenic distillation, and (d) membrane separation (Figure 4) [39, 44]. Although various new methods were suggested for CO2 separation, Granite and Brien [45] reviewed some of the most novel methods for carbon dioxide separation from flue and fuel gas streams, such as use of electrochemical pumps and chemical looping for CO2 separation.

2.1. Absorption

Absorption stripping is an important technology for CO2 capture from fuel gas; in this technology desired component in mixed gases are dissolved in a solvent (bulk phase) [46]. The general scheme of this process is depicted in Figure 5.

The flue gas (containing CO2) is cooled (between 318 and 323 K), and fed to the absorption column (scrubber) where the solvent absorbs CO2. The CO2-rich solution is fed into a heater to increase the temperature of solution, then to a stripper column to release the CO2. The released CO2 is compressed, and the regenerated absorbent solution is cooled and recycled to the absorber column [47, 48].

Energy required for post-combustion CO2 capture is an important issue. Thus, recent studies suggest that reduction of the cost of this capture could be achieved by finding suitable solvents that could process larger amounts of CO2 for a given mass and require less energy for stripping stage [49, 50].

2.1.1. Solvents

In absorption process, flue gas is contacted with a liquid “absorbent” (or “solvent”), and CO2 is absorbed by this solvent [21]. However, the absorbent should have a suitable capacity for CO2 absorption, high kinetic rate for CO2 absorption, negligible vapor pressure, and high chemical and thermal stability and should be harmless for labor persons [5153].

The solvents used for CO2 absorption can be divided into two categories: physical and chemical solvents. Physical solvent processes use organic solvents to physically absorb acid gas components rather than reacting chemically, but chemical absorption depends on acid-base neutralization reactions using alkaline solvents [54, 55]. In the recent years, many studies have compared the performance of different solvents as listed in Table 2.

(1) Alkanolamines. Between various solvent groups, alkanolamines group is the most important and more used for CO2 separation. A major problem in the usage of amines for CO2 absorption is equipment corrosion, so Albritton et al. [56] examined corrosion rate of various amine solvents and suggested corrosion rate could reduce in the following order: monoethanolamine (MEA) > 2-amino-2-methyl-1-propanol (AMP) > diethanolamine (DEA) > methyl diethanolamine (MDEA).

On the other way, MEA can react more quickly with CO2 than MDEA, but MDEA has higher CO2 absorption capacity and requires lower energy to regenerate CO2 [39, 57, 58]. Thus, it can be concluded that MEA is one of the best amine solvents for CO2 separation. Idem et al. [59] reported substantial reduction in energy requirements and modest reduction in circulation rates for amine blends relative to the corresponding single amine system of similar total amine concentration. Wang et al. [57] found that when MEA and MDEA are mixed at the appropriate ratio, the energy consumption for CO2 regeneration is reduced significantly. Dave et al. [28] compared the performance of several amine solvents and ammonia solutions at various concentrations. They showed that 30 wt% AMP based process has the lowest overall energy requirement among the solvents considered in their study (30% MEA, 30% MDEA, 2.5% NH3, and 5% NH3) [28, 60].

Knudsen et al. [61] studies showed that it is possible to run the post-combustion capture plant continuously while achieving roughly 90% CO2 separation levels and CASTOR-2 (blended amine solvents), operated in pilot scale with lower steam requirement and liquid-to-gas ratio (L/G) than the conventional MEA solvent.

Besides alkanolamines, carbonate-bicarbonate buffers and hindered amines are used in the bulk removal of CO2 owing to the low steam requirement for its regeneration. Mitsubishi Heavy Industries and Kansai Electric have employed other patented chemical solvents—strictly hindered amines called KS-1, KS-2, or KS-3. The regeneration heat of KS solvents is said to be ~3 GJ/t CO2, that is, 20% lower than that of MEA with ~3.7 GJ/t CO2 [60, 64, 77]. Generally, the overall cost of amine absorption/stripping technology for CO2 capture process is 52–77 US$/ton CO2 [71].

(2) Amino Acid. Amino acids have the same functional groups as alkanolamines and can be expected to behave similarly towards CO2 but do not deteriorate in the presence of oxygen. Based on the results of tests, the aqueous potassium salts (composed of sarcosine and proline) are considered to be the most promising solvents. The most common amino acids used in the gas treating solvents are glycine, alanine, dimethyl glycine, diethyl glycine, and a number of sterically hindered amino acids [65, 67, 68].

Results of many research groups showed that these solvents are suitable for application in membrane gas absorption units, because these solvents have better performance and degradation resistance than other chemical solvents [78]. Amino acid salts formed by neutralization of amino acids with an organic base such as amine showed better CO2 absorption potential than amino acid salts from neutralization of amino acid salts from an inorganic base such as potassium hydroxide [79, 80]. Aronu et al. [69] studied the performance of amino acids neutralized with 3-(methylamino)propylamine (MAPA), glycine, β-alanine, and sarcosine. Their results indicated that sarcosine neutralized with MAPA has the best CO2 absorption performance. Its performance is also enhanced by promoting with excess MAPA [69].

(3) Ammonia. Since ammonia is a toxic gas, prevention of ammonia “slip” to the atmosphere is a necessity. Despite this disadvantage, chilled ammonia process (CAP) was used for CO2 separation (Figure 6). In the CAP, CO2 is absorbed in an ammoniated solution at a lower absorption temperature (275–283 K) that reduced ammonia emissions from the CAP absorber. Ammonium carbonate solution resulted in approximately 38% carbon regeneration compared to MEA solution [70, 81, 82].

(4) Aqueous Piperazine (PZ). Piperazine (PZ) is as an additive used for amine systems to improve kinetics of CO2 absorption, such as MDEA/PZ or MEA/PZ blends. Because PZ solubility in water is low, concentration of PZ is between 0.5 and 2.5 M. As indicated in Table 2, increasing the concentration of PZ in solution allows for increased solvent capacity and faster kinetic. The presence of potassium in solution increases the concentration of / in solution; therefore, solution has buffering property. These competing effects yield a maximum fraction of reactive species at potassium to piperazine ratio of 2 : 1 [75, 83, 84].

2.2. Adsorption

Adsorption operation can reduce energy and cost of the capture or separation of CO2 in post-combustion capture. To achieve this goal, it is necessary to find adsorbents with suitable properties. In general, CO2 adsorbent must have high selectivity and adsorption capacity and adequate adsorption/desorption kinetics, remain stable after several adsorption/desorption cycles, and possess good thermal and mechanical stability [51, 8588]. The adsorbents used for CO2 separation placed into two main categories: physical and chemical adsorbents.

2.2.1. Chemical Adsorption

Chemisorption is a subclass of adsorption, driven by a chemical reaction occurring at the exposed surface. Adsorption capacities of different chemical adsorbents are summarized in Table 3.

A wide range of metals have been studied including [89](i)metal oxides: CaO, MgO,(ii)metal salts from alkali metal compounds: lithium silicate, lithium zirconate to alkaline earth metal compounds (i.e., magnesium oxide and calcium oxide),(iii)hydrotalcites and double salts.

In general, one mole of metal compound can react with one mole of CO2 with a reversible reaction. The process consists of a series of cycles where metal oxides (such as CaO) at 923 K are transformed into metal carbonates form (such as CaCO3) at 1123 K in a carbonation reactor to regenerate the sorbent and produce a concentrated stream of CO2 suitable for storage [90, 91].

Considerable attention was paid to calcium oxide (CaO) as it has a high CO2 adsorption capacity and high raw material availability (e.g., limestone) at a low cost. Lithium salts was recorded a good performance in CO2 adsorption, but it gained less focus due to its high production cost. Although double salts can be easily regenerated due to low energy requirement, their stability has not been investigated [93, 96].

The reaction of CO2 adsorption with Li2ZrO3 is reversible in the temperature range of 723–863 K. The capacity of lithium silicate (8.2 moL CO2/kg sorbent at 993 K) is larger than that of lithium zirconate (4.85 moL/kg sorbent) [17].

Hydrotalcite (HT) contains layered structure with positively charged cations balanced by negatively charged anions [97, 98]. Adsorption and final capacity of different adsorption/desorption cycles are listed in Table 3.

One way for improving CO2 adsorption efficiency is application of nanomaterials. Different nano-materials can be used for CO2 separation (Table 3). However, nanomaterials always have high production cost with complicated synthesis process such as carbon nanotubes and graphite nanoplatelets [99, 100].

The main disadvantage of chemical adsorbents is difficult regeneration process, and application of these adsorbents needs more studies for finding new adsorbents [88, 95].

2.2.2. Physical Adsorption

Physisorption, also called physical adsorption, is a process in which the electronic structure of the atom or molecule is barely perturbed upon adsorption. If the CO2 adsorption capacity of solid adsorbents reaches 3 mmoL/g, the required energy for adsorption will be less than 30–50% energy for absorption with optimum aqueous MEA [101]. The major physical adsorbents suggested for CO2 adsorption include activated carbons and inorganic porous materials such as zeolites [102, 103]. The adsorption capacities of various physical adsorbents are summarized in Table 4.

Coal is one of the adsorbents being suggested for CO2 separation. The total amount of CO2 that can be adsorbed in coal depends on its porosity, ash, and affinity for this molecule [111, 112]. Sakurovs et al. [113] showed that the ratio of maximum sorption capacity between CO2 and methane decreases with increasing carbon content. The average CO2/CH4 sorption ratio is higher for moisture-equilibrated coal and decreases with increasing coal rank (1.4 for high rank coals to 2.2 for low rank coals) [114116].

Activated carbon (AC) has a number of attractive characteristics, such as its high adsorption capacity, high hydrophobicity, low cost, and low energy requirement for regeneration [117119]. Activated carbons are inexpensive, insensitive to moisture, and easy for regeneration. These adsorbents have well developed micro- and mesopore structures that are suitable for high CO2 adsorption capacity at ambient pressure [120122].

However, activated carbon CO2/N2 selectivities (ca. 10) are relatively low; zeolitic materials offer CO2/N2 selectivities 5–10 times greater than those of carbonaceous materials. The adsorption capacity and selectivity of zeolites are largely affected by their size, porous diameter, charge density, and chemical composition of cations in their porous structures. The average value of heat adsorption on zeolites (36 kJ/moL) is larger than for activated carbon (30 kJ/moL), confirming the mentioned affirmation. Moreover, activated carbon can be regenerated easily and completely. Also its capacity did not decay after 10 consecutive processes cycles [122124].

Due to the increase in cost of raw materials, growing research interest has been focused on producing AC from agricultural wastes. Some of the agricultural wastes include the shells and stones of fruits, wastes resulting from the production of cereals, bagasse, and coir pith [100]. Rosas et al. [125] prepared hemp-derived AC monolith by phosphoric acid activation. The activated carbons from hemp stem are microporous materials and therefore suitable ones for hydrogen storage and CO2 capture [126].

Siriwardane et al. [127] studied CO2 adsorption on the molecular sieve 13X, 4A and activated carbon. The molecular sieve 13X showed better CO2 separation than molecular sieve 4A. At lower pressures (<50 psi), activated carbon had a lower CO2 separation than the molecular sieves, but adsorption was higher for activated carbon than molecular sieves at higher pressures [127, 128].

Liu et al. [129] indicated that zeolite 5A has higher volumetric capacities and less severe heat effect of the zeolite 13X. Chabazite zeolites were prepared and exchanged with alkali cations: Li, Na, K and alkaline-earth cations: Mg, Ca, Ba. Zhang et al. [130] studied the potential of these zeolites for CO2 separation from flue gas by vacuum swing adsorption. It was found that NaCHA and CaCHA hold comparative advantages for high temperature CO2 separation whilst NaX showed superior performance at relatively low temperatures [130]. In physical adsorption, the size and volume of the pores are important. Micropores are defined as pores, 2 nm in size, mesopores between 2 and 50 nm, and macropores, 50 nm in size. The micropores make better selective adsorption of CO2 over CH4 [131, 132].

Carbon nanotubes (CNTs) are the most famous among nano-hollow structured materials and their dimension ranges from 1 to 10 nm in diameter and from 200 to 500 nm in length [133]. Cinke et al. [134] indicated that purified single-walled carbon nanotubes (SWNTs) adsorbed CO2 better than unpurified SWNT. In addition, multiwalled carbon nanotubes (MWNTs) showed stability for 20 cycles of adsorption and regeneration [135].

More recently, nanosystems researchers have synthesized and screened a large number of zeolitic-type materials known as zeolitic imidazolate frameworks (ZIFs). CO2 capacities of the ZIFs are high, and selectivity against CO and N2 is good [136, 137]. The results of researchers (Burchell and Judkins [138], Dave et al. [28], and Yong et al. [139]) indicated that the CO2 adsorption efficiency of the honeycomb monolith is twice than activated carbon and 1.5 times greater than ZIF material [29]. Results of Kimber et al. [140] showed that CO2 selectivity of honeycomb monolithic composite decreased with increasing in burn-off.

Graphite nanoplatelets (GNP) were prepared by acid intercalation followed by thermal exfoliation of natural graphite. Functionalized graphite nanoplatelets (f-GNP) were prepared by further treatment of GNP in acidic medium. Palladium (Pd) nanoparticles were decorated over f-GNP surface by chemical method [109, 141, 142]. Adsorption capacity of this adsorbent is presented in Table 4.

The presence of several impurity gases (//H2O) greatly complicates the CO2 separation processes. Therefore, conventional adsorption-based CO2 separation processes rely on using a pretreatment stage to remove water, , and , which adds considerably to the overall cost. Also this prelayer can be used before the amine absorption column [143, 144]. Deng et al. [145] showed that the adsorption capacities follows the order SO2 > CO2 > NO > N2 on both zeolites (5A and 13X). Comparing two different adsorbents, the better separation efficiency can be achieved by 5A zeolite [145].

Zhang et al. [130] focused on the effect of water vapour on the pressure/vacuum swing adsorption process. The selected adsorbents in this study were CDX (an alumina/zeolite blend), alumina, and 13X zeolite as these adsorbents are either the prelayer for water adsorption or the main CO2 adsorption layer in the packed bed [130].

Metal-organic framework (MOF) materials are crystalline with two- or three-dimensional porous structures that can be synthesised with many of the functional capabilities of zeolites. Several MOFs have been proposed as adsorbents for CO2 separation processes, and among these Cu-BTC [polymeric copper (II) benzene-1,3,5-tricarboxylate] has proved to be dedicated with CO2 adsorption performances that are higher than those of typical adsorbents such as 13X zeolite [105, 107, 146, 147].

The MCM-41 material is one of the mesoporous products which was prepared by the hydrothermal method from mobil composition of matter (MCM) powders. Lu et al. [148] showed that mesoporous silica spherical particles (MSPs) can be synthesized using low-cost Na2SiO3 thus they can be cost-effective adsorbents for CO2 separation from flue gas [149, 150].

Layered double hydroxides (LDHs) have general formula with typically in the range between 0.10 and 0.33. These materials can be readily and inexpensively synthesized with the desired characteristics for a particular application such as CO2 adsorption [108, 151].

2.2.3. Adsorbent Modification

The role of CO2 as a weak Lewis acid is well established. Because of the nature of CO2, the surface of the physical adsorbents can be modified by adding basic groups, such as amine groups and metal oxides to improve CO2 adsorption capacity or selectivity [152154]. Three different methods for the production of these adsorbents were investigated: activation with CO2, heat treatment with ammonia gas (amination and ammoxidation), and heat treatment with polyethylenimine (PEI). However, it has been suggested that amine modification can produce better and cheaper CO2 adsorbents [24, 104, 155, 156].

Xu et al. [157, 158] designed selective “molecular basket” by grafting polyethylenimine (PEI) uniformly on MCM-41. CO2 adsorption capacity of the adsorbent was 24 times higher than MCM-41 and 2 times higher than PEI [93]. The addition of ammonium hydroxide resulted in the Zr-MOF with a slight lower adsorption of CO2 and CH4; however, the selectivity of CO2/CH4 is significantly enhanced [159, 160]. Results of Abid et al. [107] showed that the selectivity of CO2/CH4 on Zr-MOF is between 2.2 and 3.8, while for Zr-MOF-NH4 selectivity is between 2.6 and 4.3.

A nitrogen-rich carbon with a hierarchical micro-mesopore structure exhibited a high CO2 adsorption capacity (141 mg/g at 298 K, 1 atm), excellent separation efficiency (CO2/N2 selectivity is ca. 32), and excellent stability [161]. Plaza et al. [162] results showed that CO2 adsorption capacity of the DETA-impregnated alumina (≥2.3 mmoL/g) exhibited is the highest.

Amine modified layered double hydroxides (LDHs) have been prepared by several different methods. Park et al. [163] used dodecyl sulfate (DS) intercalated LDH as precursor and added (3-aminopropyl) triethoxysilane (APTS) together with N-cetyl-N,N,N-trimethylammonium bromide (CTAB) [164]. The highest adsorption capacity of amine modified LDHs for CO2 was achieved at 1.75 mmoL/g by MgAl N3 at 353 K and 1 bar. According to data in Table 4, this adsorbent has high CO2 capacity at high temperature; therefore, this adsorbent is suitable for post-combustion CO2 capture [108].

Wang et al. [114] reported that porous carbons with well-developed pore structures were directly prepared from a weak acid cation exchange resin (CER) by the carbonization of a mixture with Mg acetate in different ratios [108]. The main parameters of this adsorbent (such as CO2 capacity) are indicated in Table 4.

Shafeeyan et al. [165] prepared different adsorbents based on the central composite design (CCD) with three independent variables (i.e., amination temperature, amination time, and the use of preheat treated (HTA) or preoxidized (OXA) sorbent as the starting material). They demonstrated that the optimum condition for obtaining an efficient CO2 adsorbent is using a preoxidized sorbent and amination at 698 K for 2.1 h [165].

Table 4 compares CO2 adsorption capacities and stability of different absorbents, which were studied for post-combustion CO2 capture.

2.2.4. Different Cycles for CO2 Adsorption

Five different regeneration strategies were demonstrated in a single-bed CO2 adsorption unit: pressure swing adsorption (PSA), temperature swing adsorption (TSA), vacuum swing adsorption (VSA), electric swing adsorption (ESA), and a combination of vacuum and temperature swing adsorption (VTSA). The difference between these technologies is based on the strategy for regeneration of adsorbent after the adsorption step (Figure 7). In PSA applications, the pressure of the bed is reduced. VSA is preferred to the special PSA application where the desorption pressure is below atmospheric, whereas in TSA, the temperature is raised while pressure is maintained approximately constant, and in ESA the solid is heated by the Joule effect [166169].

For the single-bed cycle configurations, the productivity and CO2 recovery followed the sequence:

The performances of PSA, TSA, VSA, VTSA, and ESA processes for CO2 separation are reported in Table 5. Since application of adsorption process for CO2 capture in industrial scale is very important, in recent years some researches have been focused on this area; for example, Lucas et al. [170] studied the scale-up CO2 adsorption with activated carbon.

2.3. Cryogenic Distillation

Cryogenic method utilized low temperatures for condensation, separation, and purification of CO2 from flue gases (freezing point of pure CO2 is 195.5 K at atmospheric pressure). Therefore, under the cryogenic separation process, the components can be separated by a series of compression, cooling, and expansion steps. It enables direct production of liquid CO2 that can be stored or sequestered at high pressure via liquid pumping [171173].

The advantages of this technology can be summarized as follows [6, 8, 174].(1)Liquid CO2 is directly produced, thus making it relatively easy to store or send for enhanced oil recovery.(2)This technology is relatively straightforward, involving no solvents or other components.(3)The cryogenic separation can be easy scaled-up to industrial-scale utilization.

The major disadvantages of this process are the large amount of energy required to provide the refrigeration and the CO2 solidification under a low temperature, which causes several operational problems [176178]. Therefore, more studies are required for reducing the cost of cryogenic separation.

Clodic et al. [179] indicated that the energy requirement for cryogenic process was in the range of 541–1119 kJ/kg CO2. Zanganeh et al. [6] have constructed a pilot-scale CO2 capture and compression unit (CO2 CCU) that can separate CO2 as liquid phase from the flue gas of oxy-fuel combustion. Their results showed that cryogenic is the most cost effective when the feed gas is available at high pressure. Therefore, cryogenic is not suitable for post-combustion and it is well effective for separation stream with high CO2 concentration such as oxy-fuel combustion. Amann et al. [180] reported that conversion of O2/CO2 cycle was more efficient than amine scrubbing but more difficult to implement because of the specific gas turbine.

Xu et al. [175] studied a novel CO2 cryogenic liquefaction and separation system (Figure 8). In this system, two-stage compression, two-stage refrigeration, two-stage separation, and sufficient recovery of cryogenic energy were adopted. The energy consumption for CO2 recovery is only 0.395 MJ/kg CO2. Furthermore, this CO2 cryogenic separation system is more suitable for gas mixtures with high initial pressure and high CO2 concentration [175].

Song et al. [181] developed a novel cryogenic CO2 capture system based on Stirling coolers (SC). The operation of Stirling cooler contains four processes: isothermal expansion, refrigeration under a constant volume, isothermal compression, and heating under a constant volume condition. This novel cryogenic system can condense and separate H2O and CO2 from flue gas. Their results showed that under the optimal temperature and flow rate, CO2 recovery of the cryogenic process can reach 96% with 1.5 MJ/kg CO2 energy consumption.

Tuinier et al. [182] exploited a novel cryogenic CO2 capture process using dynamically operated packed beds (Figure 9). By the developed process, above 99% of CO2 could be recovered from a flue gas containing 10 vol.% CO2 and 1 vol.% H2O with 1.8 MJ/kg CO2 energy consumption [181].

Chiesa et al. [183] proposed an advanced cycle that a molten carbonate fuel cell (MCFC) was used to separate the CO2 from the gas turbine exhaust of a natural gas fired combined cycle power plant. In this cycle, gas turbine flue gases actually are used as cathode feeding for MCFC. While CO2 is moved from the cathode to anode side, concentrate CO2 in the anode exhaust. Then the CO2 is concentrated on the anode side of MCFC allowing to easily treat this spent fuel stream in a cryogenic process to split combustible species (routed back to gas turbine combustor) from the CO2 addressed to storage (Figure 10) [183].

2.4. Membrane Separation

The membrane separation method is a continuous, steady-state, clean and simple process, and ideal as an energy-saving method for CO2 recovery. Gas separation using membranes is a pressure-driven process. Due to the low pressure of flue gases, driving force is too low for membrane processes in post-combustion (low pressure and low CO2 concentration). Membrane processes offer increased separation performances when CO2 concentration in the feed mixture increases [184186].

Membrane separation processes have several advantages over other CO2 separation technologies. The required process equipment is very simple, compact, relatively easy to operate and control, clear process and easy to scale up [187, 188].

The energy required for the recovery of CO2 by membrane processes depends on the target purity, flue gas composition, and membrane selectivity for CO2. Howevre membrane processes require too much energy for post-combustion CO2 capture; therefore, low partial pressure of CO2 in the flue gas is a possible disadvantage for the application of membranes. Another disadvantage of membrane process is that the membrane selectivity for the separation of CO2 from and is very low. Membrane process is not useful for high flow rate applications [189191].

Therefore, the useful membrane for post-combustion CO2 capture should have some specification such as [192, 193](i)high CO2 permeability,(ii)high selectivity for CO2 separatation from flue gases,(iii)high thermal and chemical stability,(iv)resistant to plasticisation,(v)resistant to aging,(vi)cost effective,(vii)low production cost for different membrane modules.

Many efforts have been made to find new material with suitable properties (Table 6).

Various groups of materials have been already proposed and experimentally investigated for post-combustion CO2 capture with membrane process. By modifying membrane their properties can be improved. For example, when amine functional groups are randomly dispersed in the silica matrix, this membrane can separate CO2 with high selectivity. On the other hand, membrane structure can be modified by adding arginine salts [194196].

2.4.1. Inorganic Membranes

Based on structure, inorganic membranes can be classified into two categories: porous and dense. In porous inorganic membranes, a porous thin top layer is supported on a porous metal or ceramic support. Zeolite, silicon carbide, carbon, glass, zirconia, titania, and alumina membranes are mainly used as porous inorganic membranes supported on different substrates, such as -alumina, -alumina, zirconia, zeolite, or porous stainless steel [17, 199, 209, 210].

Zeolite membrane is the most important group of inorganic membranes. Zeolite membranes are considered more expensive than polymeric membranes, and therefore their unique properties of size selectivity and thermal and chemical stability should be exploited for successful application [211213].

The dense inorganic membranes (nonporous material) consist of a thin layer of metal, such as palladium and its alloys (metallic membrane), or solid electrolytes, such as zirconia. Another form of inorganic membrane is the liquid-immobilized membrane, where the pores of a membrane are completely filled with a liquid, which is permselective for certain compounds. Recently, attempts have been made to develop dense molten carbonate selective membranes for CO2 separation at high temperatures (>723 K). The inorganic membranes have high thermal stability for CO2 separation, but their selectivity and permeability are very low [200, 214, 215].

2.4.2. Polymeric Membranes

In polymeric membranes, the selective layer is generally a nonporous film that transports gases across by the solution-diffusion mechanism. Polyacetylenes, polyaniline, polyarylene ethers, polyarylates, polycarbonates, polyetherimides, polyethylene oxide, polyimides, polyphenylene oxides, polypyrroles, polysulfones, and amino groups such as polyethyleneimine blends, polymethacrylates are examples of polymeric membranes used for CO2 separation [17, 216218].

Selective polymeric membranes can be divided into two basic categories: glassy and rubbery. Almost glassy polymeric membranes are more suitable than rubbery polymeric membranes for CO2 separation because of their high gas selectivity and good mechanical properties. On the other hand, rubbery membranes are flexible and soft and they have a high permeability but a low selectivity, whereas glassy polymers exhibit a low permeability but a high selectivity [206, 219221].

Several advantages of polymeric membranes are (i) low cost of production; (ii) high performance separation; (iii) ease of synthesis; and (iv) mechanical stability. Although the polymeric membranes have high selectivity and permeability for CO2 separation, but their thermal stability is very low, and these membrane may be plasticized with influence of CO2 in membrane. Therefore, application of these membranes for post-combustion capture is limited, and flue gas must first be cooled down to 313–333 K for membrane process [184, 222, 223].

Ren et al. [205] prepared polymeric membranes with block copolymers; the balance of the hard and soft blocks can provide a good CO2 separation performance without loss of its permeability.

Improved polymeric membrane materials with superior separation performance can be obtained by synthesizing new polymers or modification or blending existing commercial polymers with organic or inorganic compounds [208, 224].

Due to high operating cost of membrane processes, it is necessary to perform more researches and studies about preparation of suitable membranes.

2.4.3. Mixed Matrix Membranes

Zeolites, carbon molecular sieves (CMS), and many polymeric materials offer attractive transport properties for CO2 separation. By mixing membrane material, excellent membrane with high performance for CO2 separation (selectivities of CO2/) can be prepared [200]. A group of scientists proposed the use of membrane based on polymer/immobilized liquid system especially polymerized ionic liquid membrane (PILM) or gelled ionic liquid membrane. ILMs consisting of aqueous solutions of 20% DEA immobilized in 25.4 μm microporous polypropylene supports have low permeability and suitable selectivity (974 barrer, 276, resp.) in 2 atm at 298 K [225228].

2.4.4. Hollow Fiber Membrane

Most industrially important membranes for gas separations are hollow fiber ones. Asymmetric hollow fiber membranes (such as polyvinylidene difluoride (PVDF)) with inner skinless structures are favourable for CO2 separation and absorption in gas-liquid membrane by low mass-transfer resistance and high permeability. In addition, this process can achieve significantly high adsorption efficiencies due to the much larger surface area for gas-liquid interface than conventional gas absorption processes [206, 229232].

According to data in Table 6, inorganic membranes have high permeability (about 150000 barrer) and low selectivity (about 15). Of course, some of inorganic membranes such as Y (FAU) with -A12O3 support and chitosan group are highly selective for CO2/N2 separation (selectivity () 100–800). Among polymeric membranes, polyamines have high permeability and selectivity (106 (barrer) and 980, resp.), and the second FSCM membranes have high permeability and fine selectivity (105 (barrer), 230, resp.). Other polymeric membrane groups are not selective for CO2/N2 separation, and maximum selectivity of these membranes is about 30.

2.5. Novel CO2 Capture Technologies

These methods include electrochemical pumps and chemical looping approaches to CO2 separation. The molten carbonate and aqueous alkaline fuel cells have been studied for use in separating CO2 from both air and flue gases. Electrochemical pumps discussed include carbonate and proton conductors. Molten carbonate is nearly 100% selective for CO2 separation, but major disadvantage in the application of molten carbonate electrochemical cells for CO2 separation is that this process is not repeatedly. Other disadvantages of these technologies are: corrosion, difficult operating condition ( K), and sensitivity to the presence of   [45, 233].

In chemical looping combustion, the oxygen for combustion of the fuel is provided by a regenerable metal oxide catalyst. The chemical looping scheme can be presented in the general form [45]:

Nickel oxide is one main candidate for the chemical looping combustion of methane, as low as 673 K, because it is extensive and effective for the chemical looping combustion [45].

2.6. Discussion

Various technologies such as absorption, adsorption, cryogenic distillation, and membrane have been suggested for CO2 separation from flue gases (Table 7). In this paper, various technologies for different feed conditions were investigated. Absorption is an important technology for CO2 separation. Although physical solvents required low energy for regeneration, they have low absorption capacity and selectivity for CO2 separation. Selexol is the best physical solvent and suitable for sweetening natural gas. However, physical absorption is not economical for flue gas streams with CO2 concentration lower than 15 vol% (95 US$/ton CO2 [234]).

Chemical solvents are classified in main groups such as alkanolamines, ammonia, aqueous piperazine (PZ), and amino acids. Chemical absorbents such as monoethanolamine (MEA) have high absorption capacities and are very flexible for CO2 separation; therefore, these solvents are usually preferred to physical solvents. Chemical absorption with alkanolamines is the only technology that is used in an industrial scale for post-combustion capture. Amines react rapidly, selectively, and reversibly with CO2 and are relatively nonvolatile and inexpensive solvents. In this process, the corrosion is the main problem; therefore, in recent studies, new amines and various mixtures of them were proposed and compared with previous ones to find suitable solvents. Suitable solvents for CO2 separation must have high CO2 absorption capacity, less corrosion, less viscosity and less regeneration energy. These studies showed that CASTOR 1 and 2, which are blended amine solvents (MEA/MDEA), are the best chemical adsorbents so far proposed for post-combustion CO2 capture. Experimental results indicated that amine amino acid salts (AAAS) have better performance than MEA of the same concentration for CO2 absorption, but do not deteriorate in the presence of oxygen. However, absorption has several disadvantages such as it requires high energy to regenerate solvents (3.0 GJ/ton CO2 for absorption with 40%wt MEA in 210 kPa [235]), therefore need more efforts in the future to reduce energy consumption in post-combustion CO2 capture with chemical absorption.

Adsorption is the one effective technology that can reduce energy and cost of the capture or separation of CO2 in post-combustion capture. Adsorption is suitable for separating CO2 from dilute and low flow rate stream, but flue gases conditions are the main problem against industrialization adsorption process. The CaO-MgAl2O4 and nano CaO/Al2O3 are the best chemical adsorbents. Although, the chemical adsorbents have high capacity and selectivity, but their regeneration is difficult. Physical adsorption is the most suitable for CO2 capture at high pressures and low temperatures. At higher pressure (above 4 bar) activated carbons are more efficient than zeolites. The energy and cost of adsorption for activated carbons are nearly half of that of zeolites. On the other hand, zeolites (particularly 13X and 5A) have high selectivity for CO2 separation. Generally, zeolite 5A may have better adsorption efficiency at co-adsorption of SO2, NO and CO2 than zeolite 13X.

In order to achieve more selective CO2 separation from flue gases, the modified adsorbent surface was considered. New adsorbents such as honeyncomb monolith, MOFs, CHAs (NaCHA and CaCHA), PMO (MCM and SBA) and MSPs (Na2SiO4) are suitable adsorbents for selective CO2 separation but they require more researches and studies. However, the development of suitable adsorbents with high CO2 adsorption capacity, which can be replaced absorption with chemical adsorbent, is still demanded.

Cryogenic distillation separation can be used for CO2 separation but its major disadvantage is the large amount of energy required to provide the refrigeration. Many new processes have been proposed for using cryogenic, but generally this technology is not suitable for post-combustion capture and is appropriate for oxy-fuel combustion method and CO2 separation from exhaust of cement industry (stream with high CO2 concentration).

The membrane separation method is a continuous, steady-state, clean and simple process for CO2 recovery. Since the pressure drop is driving force for membrane process, the flue gas stream must compress. Since compressing flue gas is very difficult and expensive, membrane separation is suitable for high pressure stream with high concentration (>10 vol%). Inorganic membrane have high thermal and chemical stability but their selectivity is lower than polymeric membranes. Although Y (FAU) with -Al2O3 support and arginine salt chitosan are the best inorganic membrane, zeolite mambranes are suitable ones for CO2 separation. Polymeric membranes are very selective for CO2 separation but they have low thermal stability. Therefore, polymeric membranes are suitable for application in pre-combustion processes. Glassy polymeric membranes have higher selectivity, while the rubbery polymeric membranes have higher thermal stability. Perfect membrane with high performance for CO2 separation (selectivities of CO2/) can be prepared by mixing various membranes.

Because of operating problems and high cost of compressing, membrane separation is not suitable for post-combustion capture, but membrane technology is suitable for producing oxygen-enriched streams from air, in oxy-fuel combustion systems.

Electrochemical pumps and chemical looping are two new technologies suggested for CO2 capture. Now these technologies are not effective in comparison with other technologies. Therefore, application of electrochemical pumps and chemical looping in CCS needs more research.

3. Conclusion

Because of economical and environmental incentives, researchers have mainly focused on CO2 separation from different process streams, especially from the flue gases. In recent years, post-combustion capture has been the topic of many researches, because it is more flexible and can be easily added to the fossil fuel power plants.

Based on above findings, it can be concluded that flue gases properties (mainly concentration of CO2, temperature and pressure) are the most effective factors for selection of suitable process for CO2 separation.

Since flue gases have high temperature (about 373 K), low pressure, and low CO2 concentration (1 atm and 10–15% moL), bulk absorption and adsorption processes may be the best suitable process for CO2 separation from these streams. Due to simplicity of absorption process, this process has been applied in industrial plants, although many researches have been focused on preparation of adsorbents with high selectivity and capacity, in recent years. For industrial application, more studies about adsorbents are necessary. Cryogenic distillation and membrane processes are efficient for gas streams with high CO2 concentration. Therefore, these process are economically efficient for pre-combustion capture. In recent years, different studies have been performed to optimize cryogenic cycles and preparation of suitable membrane for CO2 separation from post-combustion flue gases.

By the result of this study, future research direction on the scale-up and industrialization of adsorption (with modified adsorbent), and membrane process for CO2 separation is suggested. Therefore more studies must be focused on modeling and simulation of these processes (membrane and adsorption), although research for finding new adsorbent, suitable mambrane (with mixed or modified present membrane) and blending amine solvents can reduce CCS cost.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.