Module 2 CO2 capture: Post combustion flue gas separation

Original text: S. Wong, APEC Capacity Building in the APEC Region, Phase II Revised and updated by CO2CRC and ICF International


CO2 capture is the first and most expensive step in a CO2 capture and storage project. CO2 can be sourced from the waste gas from post-combustion or pre-combustion (decarbonization) and oxy-fuel combustion technologies.

Carbon dioxide emissions from burning fossil fuels for power and industry together contribute the largest proportion of CO2 emissions worldwide. There are three approaches to capturing CO2 from fossil fuels: post-combustion capture, pre-combustion capture and oxy-fuel combustion. Much of the current combustion process use air, and the resulting flue gas typically contains low concentrations of CO2 (<20%), which are more suitable for post-combustion capture technologies. This module focuses on approaches currently used or being developed to separate CO2 from post-combustion flue gases. Pre-combustion and oxy-fuel combustion technologies are covered in Module 3.

Learning objectives

By the end of this module you will:

  • Be able to identify sources of post combustion flue gas suited to CO2 capture;
  • Understand the various approaches that have been developed to separate CO2 from post combustion flue gas, including their advantages, disadvantages and commercial readiness; and
  • Be familiar with emerging research to improve CO2 separation and make it more economically viable.

Sources of post combustion flue gas suited to CO2 separation

CO2 can be sourced from the waste gas from post-combustion or pre-combustion (decarbonization) and oxy-fuel combustion technologies. A schematic showing the three approaches to fossil fuel combustion and capture is depicted in Figure 2.1.

Figure 2.1: A diagram of three approaches to capturing carbon dioxide. Note that the oxidation in IGCC can be through either air or oxygen from an air separation unit (courtesy of CO2CRC).

Carbon dioxide emissions from burning fossil fuels for power and industry together contribute the largest proportion of CO2 emissions worldwide (47.8% in 2006 (WEO, 2008)). Much of this combustion occurs in air, and the resulting flue gas typically contains low concentrations of CO2 (<20%).

Capture of CO2 is best carried out at large point sources of emissions, such as power stations, oil refineries, petrochemical and gas plants, steel works and large cement works. CO2 can be captured either from combustion flue gases or from process streams before combustion. Typical CO2 concentrations in process streams are shown in Table 2.1 below:

Table 2.1: Flue gas concentrations from various stationary sources (IPCC, 2005).

As of April 2009, EPRI has determined there are over 50 post-combustion CO2 capture concepts and six physical/chemical process types under development, which can be categorized as depicted below.4

Figure 2.2: Post-Combustion CO2 Capture Technology Groups. Source: EPRI 2009

Each of these different categories has different benefits and drawbacks, as well as applicability in different situations. These will be reviewed in the following sections. Other separation techniques include chemical looping which effectively removes oxygen from air prior to combustion.

Chemical and physical absorption

Chemical absorption

Chemical absorption involves one or more reversible chemical reactions between CO2 and an aqueous solution of an absorbent, such as an alkanolamine or potassium carbonate (explained below). Upon heating the product, the bond between the absorbent and CO2 can be broken, yielding a stream enriched in CO2. The chemical absorption process for separating CO2 from flue gas is borrowed from the gas processing industry. Amine based processes have been used commercially for the removal of acid gas impurities (CO2 and H2S) from process gas streams. It is therefore a proven and well-known technology.

Amine is a group of organic compounds, which can be considered as derived from ammonia (NH3) by replacement of one or more hydrogen molecules by organic radicals. Amines are classified according to the number of hydrogens of ammonia that have been replaced by radicals as follows:

  • Primary amine (RNH2) - one hydrogen molecule has been replaced;
  • Secondary amine (R2NH) - two hydrogen molecules have been replaced; and
  • Tertiary amine (R3N) – all three hydrogen molecules have been replaced.

The substitute groups (R) may be alkyl, aryl or aralykl. When the (R) is an alkyl, the amine is called alkanolamine. In general, it can be considered that a hydroxyl group serves to reduce the vapor pressure and increase the water solubility, while the amino group provides the necessary alkalinity in water solution to absorb the acid gases. Akanolamines remove CO2 from the waste gas streams through an exothermic reaction of CO2 with the amine functionality of the alkanolamine. The amines of commercial interest to capture CO2 are water-soluble.

A typical chemical absorption unit is shown in Figure 2.3. During the amine absorption operation the waste gas stream and liquid amine solution are contacted by countercurrent flow in an absorption tower (or absorber, see the left hand side of Figure 2.3).

Figure 2.3: Typical chemical absorption unit for CO2 recovery from flue gas.

The combustion flue gas coming out of the stack is hot (~ 232°C), and at atmospheric pressure. The flue gas needs to be cooled, before entering the absorber, which usually operates at less than 50°C. This is achieved by spraying cooling water in a direct contact cooler, as shown in bottom left hand side of Figure 2.3. A blower is installed to give the flue gas enough pressure for it to go through the absorption-desorption system. Conventionally, the waste gas to be scrubbed of the CO2 enters the absorber at the bottom, flows up, and leaves at the top, whereas the solvent enters the top of the absorber, flows down (contacting the gas), and emerges at the bottom. Dilution of the circulating amine with water is undertaken to reduce viscosity of the circulating fluid. A higher viscosity fluid would require more power to pump and provide circulation. The liquid amine solution containing the absorbed gas then flows to a regeneration unit (stripper) where it is heated and the acid gases liberated. The solvent regeneration can be carried out at low pressures to enhance desorption of CO2 from the liquid. Some amine solution is typically carried over in the acid gas stream from the regeneration step and the amine solution is recovered using a condenser, in order to avoid excessive solvent losses. The hot lean amine solution then flows through a heat exchanger where it is contacted with the rich amine solution from the contact tower and from there the lean amine solution is returned to the gas contact tower, i.e. absorber.

Figure 2.4: A solvent capture rig (background) retrofitted to International Power's Hazelwood coal-fired power plant in Australia to capture carbon dioxide from post-combustion flue gases (courtesy of CO2CRC).

Types of alkanolamines

There are three main groups of Alkanolamines: primary, secondary and tertiary amines.

Primary amines include monoethanol amine (MEA) and diglycolamine (DGA). MEA has been the traditional solvent of choice for carbon dioxide absorption and acid gas removal in general. MEA is the least expensive of the alkanolamines; its reaction kinetic is fast and it works well at low pressure, and low CO2 concentration. However, there are several disadvantages. First, it has a high heat of reaction with CO2, which means high level of energy has to be supplied in the regeneration step. Second, the absorptivity of MEA with CO2 is not great. In the case of primary and secondary alkanolamines the formation of carbamate (RNHCOO-) is the main reaction.

CO2 + 2RNH2 = RNHCOO- + RNH3+

In this reaction, two moles of MEA must be used to capture 1 mole of CO2. Third, the full upper absorption capacity of MEA is not realized in practice due to corrosion problems. The corrosion effect is due to dissolved CO2 and varies with the amines used. The concentration of MEA in the aqueous phase in the presence of O2 is limited to 20-wt% (weight percent). In addition, MEA has the highest vapor pressure of any of the alkanolamines and high solvent carryover can occur during CO2 removal from the gas stream and in the regeneration step. To reduce solvent losses, a water wash of the purified gas stream is usually required. In addition, MEA reacts irreversibly with minor impurities such as COS and CS2 resulting in solvent degradation. Foaming of the absorbing liquid MEA due to the build-up of impurities can also be a concern.

There is considerable industrial experience with MEA and most systems at present use an aqueous solution with only 15-25-wt% MEA, due to corrosion issues (GPSA, 1998). Corrosion inhibitors may be added to MEA solution, and this results in an increase in solution strength. The inhibitors are usually not disclosed, as these will distinguish one commercial MEA process from another. In the commercial Fluor Daniel ECONAMINE FG process, a concentration of MEA up to 30-wt% has been employed successfully to remove 80% - 90% of the carbon dioxide from the feed gas (Mariz, 1998). Another commercial process from ABB LUMMUS, which uses 20% MEA with inhibitors, is also offered for CO2 capture (Barchas, 1992).

For the current MEA absorber systems, the adsorption and desorption rates are reasonably high, hence good reaction kinetics. However, the packing in the absorber (contactors, to facilitate efficient mass transfer) represents a significant cost, and its energy consumption is also significant for CO2 capture from flue gas. In addition, the stripping temperature should not be too high (~ 150°C). Otherwise, dimerization of carbamate may take place, deteriorating the sorption capability of MEA.

Secondary amines include diethanolamine (DEA), di-isopropylamine (DIPA). Secondary amines have advantages over primary amines. Their heat of reaction with carbon dioxide is lower (360 calorie/g (650 BTU/lb) versus 455 calorie/g (820 BTU/lb) for primary amines). This means that the secondary amines require less heat in the regeneration step than primary amines. However, it has all the other problems of primary amines.

Tertiary amines include triethanolamine (TEA) and methyl-diethanolamine (MDEA). Tertiary amines react more slowly with carbon dioxide than primary and secondary amines thus require higher circulation rate of liquid to remove carbon dioxide compared to primary and secondary amines. This can be improved through the use of promoters. A major advantage of tertiary amines is their lower heat requirements for carbon dioxide liberation from the carbon dioxide containing solvent.

Tertiary amines show a lower tendency to form degradation products in use than primary and secondary amines, and are more easily regenerated. In addition, tertiary amines have lower corrosion rates compared to primary and secondary amines. The main drawback is its reaction rate is too slow.

Table 2.2 compares the heat of reaction between the three amine and carbon dioxide.

Table 2.2: Heat of reaction between three amines and carbon dioxide (Skinner et al, 1995).

Corrosion has been a serious issue in amine processes. In general, alkanolamines themselves are not corrosive to carbon steel; it is the dissolved CO2 that is the primary corroding agent. The alkanolamines indirectly influence the corrosion rate when they absorb CO2. The observed corrosivity of alkanolamines to carbon steel is generally from primary (most corrosive) to tertiary (least corrosive).

Table 2.3: Some commercial CO2 recovery plants worldwide.

Limitations of amine-based processes and technological advances

Much of the amine scrubbing technology in the past has focused on the removal of hydrogen sulfide for the natural gas sector. However, the requirements are different for the recovery of CO2 from power plant flue gas. One challenge is the low pressure of the flue gas for absorption of CO2. In addition, impurities in flue gas such as oxygen, sulfur oxides, nitrogen oxides, and particulate matter create special challenges during the separation process.

Low pressure – the greatest limitation for CO2 recovery from flue gas is the low pressure of the flue gas. CO2 is absorbed much more easily into solvents at high pressure. The only commercially available solvents that can absorb a reasonable amount of CO2 from dilute atmospheric pressure gas are primary and sterically hindered amines, such as MEA, DGA and the KS-1 series of solvents (Chapel et al., 1999). These solvents can absorb CO2 at low pressures because they have high reaction energies. This results in high-energy requirements to regenerate the rich solvent. However, energy costs may be reduced if the process can be fully integrated with a power plant where significant amount of low-grade heat may be available. (see Heat Integration)

Oxygen – most amine solvents degrade to varying degrees in oxidizing atmospheres. This leads to either high solvent losses or expensive reclaiming processes. Oxygen also causes corrosion problems in the process equipment, which can lead to failures or more expensive materials of construction. The use of inhibitors in the solvent to reduce degradation and corrosion appears to work well and produces very good results.

Sulphur oxides – (SO2, SO3) react with MEA to form heat-stable corrosive salts that cannot be reclaimed. Some commercial MEA processes require a sulphur oxides limit of less than 10 ppm level. It is generally accepted that installing a flue gas desulfurization unit before the absorber is the best way to alleviate the problem.

Nitrogen oxides – a typical flue gas contains some amount of NOx. NOx generally consists of NO and NO2 in a ratio of from 95:5 to 90:10. The main component NO performs as inert gas and will not affect the solvent. However, NO2 will partially lead to the formation of a heat stable salt. Generally some solvent degradation is acceptable in order to avoid the cost of removing the NO2.

Particulate matter – fly ash in the flue gas can cause foaming and degradation of the solvent, as well as plugging and scaling of the process equipment. A wash operation has been recommended to reduce the fly ash content to appropriate levels to abate the aforementioned problems.

Flue gas entering the absorber at high temperatures can lead to solvent degradation and decreased absorption efficiency. The flue gas must be cooled to a water dew point of 50°C, which can be accomplished in the desulfurization unit or with a direct contact water cooler.

In summary, the recovery of CO2 from combustion flue gas requires a significant amount of pre-treatment processing in order to avoid any foul-up in the solvent absorption step. This will add to the cost of CO2 capture. However, significant improvements can be made in the solvent absorption process in terms of optimizing the compositions of the absorbing amines and the gas-liquid contactors, in order to manage this.

Specialty amines

Considerable work is ongoing on improved amines and processes for the specific task of carbon dioxide capture. Specialty amines such as hindered amines are being developed to solve some of these issues. The idea behind hindered amines is based on attaching a bulky substitute to the nitrogen atom of the amine molecule. This molecular configuration plays an important role in process performance, by affecting the capacity of absorption and the desorption temperature. In the case of CO2 removal, the capacity of the solvent can be greatly enhanced if one of the intermediate reactions, such as the carbamate formation reaction, can be slowed down by providing steric hindrance to the reacting CO2. A 2006 study at a pilot plant in Osaka compared MEA, KS-1, a hindered amine and DEA found that the hindered amine performed well, with the highest effective CO2 loading of the solvents (Yagi et al, 2006).

In addition to slowing down the overall reaction, bulkier substitutes give rise to less stable carbamates. By making the amine carbamate unstable, one can theoretically double the capacity of the solvent (Chakma, 1994).

In this case the bicarbonate formation becomes the dominant reaction.

RNH2 + CO2 + H2O = RNH3+ + HCO3-

The advantage of sterically hindered amines over the alkanolamines is that in the bicarbonate reaction only 1 mole of hindered amine instead of 2 moles of alkanolamine is required to react with 1 mole of CO2. In addition, sterically hindered amine systems can have lower heats of absorption/regeneration as compared with MEA.

The International CO2 Capture Center in Regina, Saskatchewan, Canada also has developed a series of proprietary designer solvents designated as PSR solvents (Veawab et al., 2001). The PSR solvents have been designed specifically for the separation of CO2 from flue gas streams. The PSR solvents may be used at higher amine concentration than conventional MEA solvents and at a higher loading of CO2. The key features claimed for the PSR solvents are lower regeneration temperature, lower solvent circulation rate, lower solvent degeneration rate, and lower corrosion rate.

Further developments in amine solvents

Mistusbishi Heavy Industries (MHI) and KANSIA have a commercially developed propriety solvent KS-1, which is used to strip CO2 from flue gas from natural gas boilers and steam reformers. The plants operating with this technology are in Malaysia, Japan and India (two plants) (Kishimoto et al, 2009). There have been trials of the solvent with coal fired boilers in a 10 t/d slip steam from J-Power's 2x500MW units in Matushima, Japan. The CO2 recovery efficiency is 90%. The heat consumption was 730 – 820 kcal/kg CO2, but with improvements to the process, the steam consumption could be reduced by 15%.

An EU project, CASTOR, involves trials of novel blended amine solvents CASTOR- 1 and CASTOR-2 at a 24t/d absorption pilot plant at Esbjerg power station, Denmark. Knudsen et al report that CASTOR -2 has a lower steam demand than MEA and is more chemically stable (Knudsen et al, 2009).

Chilled ammonia

The CO2 in the flue gas is cooled before entering the absorber where it reacts with ammonium carbonate to form ammonium bicarbonate. Ammonia is released as a gas from the solvent solution when the CO2 is absorbed, and the temperature is kept low to minimize this. Gases exiting the absorber pass through a water wash to remove ammonia.

The ammonium bicarbonate is heated in the regenerator, separating the CO2. The ammonium carbonate solvent is returned to the absorber. Water and ammonia are removed from the CO2 stream exiting the stripper column.

Alstom's Chilled Ammonia Process has been piloted at the We Energies Pleasant Prairie Power Station. A larger demonstration (100,000 tonnes per annum) including sequestration in a saline formation is planned for the AEP Mountaineer coal-fired Power Station in the US.

The chilled ammonia-based process is very similar to that of amine capture process, but has a reduced efficiency penalty. The greater efficiency of the chilled ammonia carbon capture process can be attributed to its reduced heat of reaction energy needs (60% lower than MEA), its ability to regenerate without stripping steam, and its greater CO2 absorptive capacity.5 Other advantages include:

Lower energy requirement for solvent regeneration (steam consumption is low) (Gal 2006);

Reduced sensitivity of the solvent to acid gases such as SOx and Nox;

Resistance to oxidative and thermal degradation;

Ability to clean the flue gases of Nox, Sox and other pollutants;

Cheap solvent; and

Low corrosion impact.

The main hurdles, at present, towards large scale demonstration of ammonia-based capture are:

Ammonia losses as vapour phase in the discharged flue gases, due to its small molecular weight and large vapour pressure-direct discharge into the atmosphere is not preferable;

Absorption must be carried at low temperature to minimise ammonia loss (hence, 'chilled' ammonia);

High refrigeration load requirements;

Technology maturity and confidence;

Stripper fouling; and

Increased capital costs due to its need for several absorber vessels used for minimizing loss of ammonia and maintaining water balance.

More broadly, the lack of commercial-scale process experience at this time is limiting the deployment of this technology.6

Chemical absorption with potassium carbonate

Many alkaline salt-based processes have been developed for carbon dioxide removal. These utilize the alkali salts of various weak acids. The most popular salts in the industry have been sodium carbonate and potassium carbonate. Low cost and minimal degradation of the solvent are the primary reasons.

The major commercial processes that have been developed for H2S and CO2 absorption are aqueous solutions of sodium or potassium compounds. Potassium carbonate can absorb CO2 at high temperatures, an advantage over amine-based solvents. The principal technologies employed are processes based on hot potassium carbonate (K2CO3) solutions that are used for the removal of CO2 from high-pressure gas streams, among other applications.

The hot potassium carbonate process is used in many ammonia, hydrogen, ethylene oxide and natural gas plants. Potassium carbonate has a low rate of reaction. To improve CO2 absorption mass transfer promoters such as piperazine, diethanolamine and arsenic trioxide have been used. Less toxic promoters such as borate are currently being investigated for used with flue gas streams (Ghosh et al, 2009). To limit corrosion, inhibitors are added. These systems are known as activated hot potassium carbonate systems.

Studies indicated that presence of flue gas impurities SOx and NOx reduces the operational efficiency of the potassium carbonate as a solvent. SO2 and NO2 are not able to be released from the solvent under industrial conditions. Current research is investigating the selective precipitation of the impurity salts formed by SOx and NOx so that they can be removed (Smith et al, 2009).

Some licensed hot, activated potassium carbonate systems are the Benfield and the Catacarb process. The processes are designed for bulk CO2 removal from high-pressure streams, but also produce high-purity CO2.

Contactors in solvent systems

Various column packings are used to increase mass transfer rates. Examples include Pall rings (a random packing), Mellapak (a structured packing). Novel packings, such as SMR, are being trialled and significant improvements in mass transfer have been reported (Smith et al 2009).

Capture plant process optimisation and heat integration

Careful design of the post combustion capture plant and of the capture plant within a power plant can reduce the energy cost of adding a capture unit to a power plant. Tests by MHI have shown that using lean solvent and steam condensate heat for regeneration inside the stripper can achieve a 15% reduction in the heat required for capture (Kishimoto et al, 2009). In addition, integration of heat between the power plant and the capture plant has been studied, and improvements in design show that further reductions in the energy required are possible.

Studies of a lignite power station have shown that through proper integration of CCS into the plant, particularly through the use of energy in the flue gas, the energy penalty of the plant can be reduced to approximately 15%. Further reductions are possible if the coal is pre-dried (Harkin et al, 2009).

Physical absorption

For physical absorption, CO2 is physically absorbed in a solvent according to Henry's Law. The absorption capacity of organic or inorganic solvents for CO2 increases with increasing pressure and with decreasing temperatures. Absorption of CO2 occurs at high partial pressures of CO2 and low temperatures. The solvents are then regenerated by either heating or pressure reduction. The advantage of this method is that it requires relatively little energy; but the CO2 must be at high partial pressure. Hence, it is suitable for recovering CO2 from "Integrated Gasification Combined Cycle" (a pre-combustion technology – see Module 3), where the exhaust CO2 would leave the gasifier at elevated pressures. Some physical solvent processes are the Selexol process (dimethylether of polyethylene glycol), the Rectisol process (cold methanol), the Fluor Solvent process (Propylene Carbonate) and Purisol process (N-Methyl-2-Pyrrolidone). The Purisol and Selexol processes have a high selectivity for H2S, while the Flour Solvent is best used with feed gases with low levels of H2S.

Selexol has been used since 1969 to sweeten natural gas, both for bulk CO2 removal and H2S removal. Absorption takes place at low temperature (0 - 5°C). Desorption of the rich Selexol solvent can be accomplished either by letting down the pressure (CO2 removal) or by stripping with air, inert gas or steam. Hydrocarbons, COS, CS2 and mercaptans are also removed by the solvent. Additionally, the low absorption temperature used requires that the lean solvent be returned to the absorber via a refrigeration unit. The Exxon gas plant at La Barge, Wyoming, USA uses two Selexol processes in series, one for removing H2S and other for removing CO2.

Rectisol has mainly been used to treat synthesis gas, hydrogen and town gas streams and removes most impurities. The Great Plains Synfuels Plant in North Dakota, USA - a coal gasification plant - uses a Rectisol process to separate CO2 from a mixture of H2, CO and CO2. More than 3Mt/CO2 per year is captured at this plant. Chilled methanol is used in the North Dakota plant, however, in general, other solvents are also available for special applications.

Because the partial pressure of CO2 in the combustion flue gas is low and the temperature is relatively high, the physical absorption approach does not appear competitive compared to chemical absorption for post-combustion capture.

Solid physical adsorption

An adsorption process consists of two major steps: adsorption and desorption. The technical feasibility of a process is dictated by the adsorption step, whereas the desorption step controls its economic viability. Adsorption requires a strong affinity between an adsorbent and the component to be removed from a gas mixture (in this case, CO2). However, the stronger the affinity, the more difficult it is to desorb the CO2 and the higher the energy consumed in regenerating the adsorbent for reuse in the next cycle. The desorption step, therefore, has to be very carefully balanced against the adsorption step for the overall process to be successful.

Figure 2.5: The principle of adsorption capture (courtesy of CO2CRC).

The main advantage of physical adsorption over chemical absorption is its simple and energy efficient operation and regeneration, which can be achieved with a pressure swing or temperature swing cycle (a swing in pressure or temperature as the process goes through an absorption-desorption cycle in order to achieve separation). These separations processes are known as pressure swing adsorption (PSA) and temperature swing adsorption (TSA). Where the regeneration of the adsorbent bed is achieved through a pressure reduction to near-vacuum pressure, it is known as vacuum swing adsorption (VSA). Pressure swing adsorption is a commercial process for hydrogen separation from H2 and CO2 mixtures in hydrogen production.

There have been significant advances in the development of adsorbents for CO2 removal from flue gases. The primary adsorption material used has been zeolites. Zeolites are more effective for CO2 separation from species which are less polar than CO2, so the presence of water and SOx in flue gas streams poses a problem (Ram Reddy et al, 2008).

New adsorbents have been considered and developed such as carbons, mesoporous silico-aluminates (eg zeolitic imidazolate frameworks, ZIFs) and metal organic frameworks (MOFs). Carbon-based adsorbents have the potential to be regenerated by applying a electrical voltage, (electrical swing adsorption), or ESA. New materials being investigated include layered double hydroxide derivatives (LDHs and LDOs). Other advances include functionalising the pores of the adsorbent material by incorporating amines to increase CO2 loading. In this case, the CO2 is separated through a chemisorption process (Chaffee et al, 2006).

New processes are being developed for dealing with high humidity flue gas streams and impurities. These include multilayered adsorbent beds. Multilayered beds enable the use of adsorbents with high CO2 selectivity but which degrade significantly in the presence of water.

Figure 2.6: Multilayered adsorption beds for separating carbon dioxide (courtesy of CO2CRC).

Low temperature distillation (cryogenic separation)

Low temperature distillation, or cryogenic separation, is a commercial process commonly used to liquefy and purify CO2 from relatively high purity (> 90%) sources. It involves cooling the gases to a very low temperature so that the CO2 can be liquefied and separated.

Distillation generally has good economies of scale. This method is worth considering where there is a high concentration of CO2 in the waste gas. The advantage is that it produces a liquid CO2 ready for transportation by pipeline. The major disadvantages of this process are the amount of energy required to provide the refrigeration and the necessary removal of components that have freezing points above normal operating temperatures to avoid freezing and eventual blockage of process equipment.

For post combustion flue gases, the waste streams contain water and other trace combustion by-products such as NOx and SOx several of which must be removed before the stream is introduced to the low temperature section. Moreover, these by-products are usually generated near atmospheric pressure. These tend to make cryogenic process less economical than others in separating CO2 from flue gas. However, it is a serious contender for high-pressure gases such as in pre-combustion decarbonization processes. (see Module 3).

Membrane separation

Separation membranes are thin barriers that allow selective permeation of certain gases. They are predominately based on polymeric materials. Membranes for gas separation are usually formed as hollow fibers arranged in the tube-and-shell configuration, or as flat sheets, which are typically packaged as spiral-wound modules (Figure 2.7).

Figure 2.7: A spiral wound module showing the separation of carbon dioxide from other gases (courtesy of CO2CRC).

The membrane process has been widely used on a commercial scale for hydrogen recovery from purge gases in ammonia synthesis, refinery and natural gas dehydration, sour gas removal from natural gas, and nitrogen production from air.

The advantages of the membrane process are:

  • It does not require a separating agent, thus no regeneration is required;
  • The systems are compact and lightweight, and can be positioned either horizontally or vertically, which is especially suitable for retrofitting applications;
  • Modular design allows optimization of process arrangement by using multi-stage operation; and
  • Low maintenance requirements because there are no moving parts in the membrane unit.

A number of solid polymer membranes are commercially available for the separation of CO2 from gas streams, primarily for natural gas sweetening. These membranes selectively transmit CO2 versus CH4. The driving force for the separation is pressure differential across the membrane. As such, compression is required for the feed gas in order to provide the driving force for permeation. In addition, the separated CO2 is at low pressure and requires additional compression to meet pipeline pressure requirements. The energy required for gas compression is significant when a very high pressure is required.

The commercial membranes for CO2 separation are mainly prepared from cellulose acetate, polysulfone, and polyimide. These membranes are primarily tailor-made for natural gas processing. Condensable components in a flue gas, particularly water, will reduce the CO2 permeability and affect the selectivity of the membrane (by making a glassy polymer more rubbery). Laboratory trials mimicking real flue gas have been carried out at CO2CRC's laboratories, and trials on post-combustion and pre-combustion flue gas streams are being undertaken at pilot plants (Scholes et al, 2009).

The selectivity of CO2/N2 of these membranes is generally in the range of 1:20 ~ 1:40 depending on the operating temperature. Because of the specific characteristics of flue gas composition, and the specific features of the separation (i.e. large volumetric flow rate, low source pressure, high temperature, and the relatively low commodity value of CO2), further development is being undertaken for economically capturing CO2 from flue gas on a large scale. These developments aim to improve the selectivity of CO2/N2 and the permeability of the membranes. New composite polymeric membranes have been developed. Some of these combine membranes with high selectivity with membranes with high permeability. Others improve separation performance by blending polymides.

Membrane research includes work on inorganic membranes which can operate at high temperatures (eg for H2/CO2 separation in pre-combustion gas streams – see Module 3). Inorganic membranes can contain functionalized pores which increase the selectivity above what can be achieved by molecular sieving alone. Research is being undertaken into mixed matrix membranes (inorganic particles in a polymeric matrix) and facilitated transport membranes (where a chemical reaction occurs between the gas and the membrane). Facilitated transport membranes rely on a reversible reaction occurring with the membrane to transport the CO2 through the membrane. A pressure difference is still required to drive separation. (Scholes et al, 2008)

Membrane gas absorption – a hybrid membrane/chemical absorption process

Some efficiencies could be realized by developing amine and membrane technologies in tandem, thereby forming a hybrid process to capture CO2 from flue gas. Micro-porous hollow fiber membranes are evolving as a new technology for CO2 separation using amine-based chemical absorption processes. Micro-porous membranes are used in the gas-liquid unit where the amine solution is contacted with the CO2 containing flue gas. The principle advantage of the micro-porous membrane is the reduction in the physical size and weight of the gas-liquid contacting unit.

Figure 2.8: Membrane gas absorption: hollow fibre module (courtesy of CO2CRC).

Unlike conventional membrane separation, the micro-porous hollow fiber membrane separation is based on reversible chemical reaction, and mass transfer occurs by diffusion of the gas through the gas/liquid interface just as in the traditional contacting columns.

The hollow fiber membrane itself does not contribute to the separation but instead acts as a contacting medium between the gases and the liquid. There are a number of advantages to using the gas-liquid membrane contactors, including:

  • High gas/liquid contact area due to the high packing density of the hollow fibers (500 to 1,500 m2/m3, versus 100~250 m2/m3 for a conventional column);
  • Foaming is eliminated since the gas flow does not impact the solvent and there is no convective dispersion of gas in the liquid;
  • The membrane acts as a partition between the gas and liquid, and the gas/liquid flow rate ratio may vary widely without causing flooding problems;
  • The available gas/liquid contact area is not disturbed by variations in flow rates. This means the process can tolerate a wider range of process condition variations;
  • Solvent degradation is minimized as oxygen (a degradation agent to amines) is prevented from intimate contact with the solvents; and
  • Unlike the absorption column that can only be operated vertically, the hollow fiber membrane contactor may be operated in any orientation to suit the overall plant layout.

Challenges ahead for post-combustion capture

Cost and scale are two significant issues for CO2 capture. Advances are being made to reduce the cost of capturing CO2, but these advances may be overrun by higher plant construction costs. The focus should be on reducing the energy penalty for solvents, including adapting current solvents and developing new solvents

While there are new pilot plants for post-combustion capture, demonstration of the technology for these applications at the scale of 2-4 Mt CO2 year on a coal-fired power plant has not yet taken place.

Other issues are water usage, environmental impact, and the feasibility of retrofitting capture plants.


Capture of CO2 is best carried out at large point sources of emissions, such as power stations, oil refineries, petrochemical and gas plants, steel works and large cement works. CO2 can be captured either from combustion flue gases or from process streams before combustion. CO2 in combustion flue gas has low partial pressure. CO2 is absorbed much more easily into solvents at high pressure. Commercially available solvents that can absorb a reasonable amount of CO2 from dilute atmospheric pressure gas are primary and sterically hindered amines. These solvents have high reaction energies. This results in high-energy requirements to regenerate the rich solvent.

CO2 scrubbing with amines has been borrowed from the natural gas processing industry. Much of the amine scrubbing technology in the past has focused on the removal of hydrogen sulfide; however, for the recovery of carbon dioxide from power plant flue gas, the requirements are different. Other impurities such as oxygen, sulfur oxides, nitrogen oxides, and particulate matter are present in flue gas and these substances must be removed before the chemical absorption step. These add to the cost of CO2 separation.

In the short term, significant improvements can be made to the chemical absorption process in optimizing the compositions of the absorbing amines; and improving the performance of the gas-liquid contactors and the overall heat utilization by better heat integration with the CO2 source plant. With further improvements, cost is expected to fall.

Hybrid membrane/amine process is promising. Micro-porous hollow fiber membranes are evolving as a new technology for CO2 separation using amine-based chemical absorption processes. Other separation processes such as physical adsorption and low temperature distillation are being developed. The advantages of physical adsorption over chemical adsorption is that it is simpler and more energy efficient. Low temperature distillation is a commercial process commonly used to liquefy and purify CO2 from relatively pure sources. Distillation is suited to situations where there is a high concentration of CO2 in the waste gas. It creates transport-ready liquid CO2, but draws a high amount of energy.


Aboudheir, A., Tontiwachwuthikul, P., and Chakma A. CO2-MEA Absorption in Packed Columns: Comprehensive Experimental Data and Modeling Results, in Proceedings of the Fifth International Conference on Greenhouse Gas Control Technologies, pp. 217-222, Cairns, Australia, 2001.

Al-Juaied, M. A. and Whitmore, A., "Realistic Costs of Carbon Capture" Discussion Paper 2009-08, Cambridge, Mass.: Belfer Center for Science and International Affairs, July 2009.

Barchas, R., and Davis, R. The Kerr-McGee/ABB Lummus Crest Technology for the Recovery of CO2 from Stack Gases, Energy Convers. Mgmt., Vol. 33, No. 5-8, pp 333-340, 1992.

Chaffee, A, Knowles, G, Liang, Z, Delaney, S, Graham, J, Zhang, J, Xiao, P and Webley, P. CO2 capture by adsorption: materials and process development. Oral presentation given at the Eighth International Conference on Greenhouse Gas Control Technologies, Trondheim, Norway, 19-22 June, 2006.

Chakma, A. Separation of Acid Gases from Power Plant Flue Gas Streams by Formulated Amines, in Gas Separation, Edited by Vansant, E.F., Elsevier Science Publishers, pp. 727-737, 1994.

Chakma, A. and Tontiwachwuthikul, P. Economics and Cost Studies of Designer Solvents for Energy Efficient CO2 Separation from Flue Gas Streams, in Proceedings of Fifth International Conference on Greenhouse Gas Control Technologies, Cairns, Australia, 2001.

Chapel, D., Ernst, J., Mariz, C. Recovery of CO2 from Flue Gases: Commercial Trends, presented at the Canadian Society of Chemical Engineers Annual Meeting, Saskatoon, Saskatchewan, Canada, October 4-6, 1999.

EPRI. Status of CO2 Capture Technology, 2009.

Falk-Pedersen, O., Dannstrom, H., Gronvold, M., Stuksrud D-B., and Ronning, O., Gas Treatment Using Membrane Gas/Liquid Contactors, in Proceedings of the Fifth International Conference on Greenhouse Gas Control Technologies, pp.115-120, Cairns, Australia, 2001.

Friedman, B.M., Wissbaum, R.J., and Anderson, S.P. Various Recovery ProcessesSupply CO2 for EOR Projects, Oil and Gas Journal, pp 37-43, August 23, 2004.

Gas Processors Suppliers Association, (GSPA), Engineering Data Book, Volume II, Section 21, Tulsa, Oklahoma 1998.

Ghosh, U.K., Kentish, S. E. and Stevens, G. absorption of carbon dioxide into aqueous potassium carbonate promoted by boric acid Energy Procedia 1(1), pp 1075-1081, 2009.

GHGT9 Conference Summary. A summary of the 9th International Conference on Greenhouse Gas Control Technologies, Washington, 16-20 November,2008. Available online at [accessed 19/08/2009]

Gunter, W.D., Wong, S., Cheel, D.B. and Sjostrom, G. Large CO2 Sinks: Their Role in the Mitigation of Greenhouse Gases from an International, National (Canada) and Provincial (Alberta) Perspective, Applied Energy 61, 209-227, 1998.

Harkin, T., Hoadley, A. and Hooper, B. Redesigning the cold end of a lignite power station for CO2 capture. In: Proceedings of the Fourth International Conference on Clean Coal Technologies (CCT2009), Dresden, Germany, 18-21 May2009.

Hill, G. and Moore, R. A Collaborative Project to Develop Technology to capture and Store CO2 from Large Combustion Sources, presented at First National Conference on Carbon Sequestration, Washington, DC, May 15-17, 2001.

Iijima, M. A Feasible New Flue gas CO2 Recovery Technology for Enhanced Oil Recovery, SPE paper 39686, Presented to 1998 SPE/DOE Improved Oil Recovery Symposium, Tulsa, Oklahoma, April 19-22, 1998.

Kishimoto, S., Hirata, T., Iijima, M., Ohishi, T, Higake, K., and Mitchell, R., Current Statis of MHI's CO2 Recovery Technology and Opitmistaion of CO2 Recovery Plant with a PC Fired Power Plant, Energy Procedia 1(1), pp1091-1098, 2009.

Knudsen, J., Jensen, J., Vilhelmsen, P. and Biede, O. Experience with CO2 Capture from coal flue gas in pilot-scale: Testing of different amine solvents, Energy Procedia 1 (1), pp783-790, 2009.

Mariz, C.L. Carbon Dioxide Recovery: Large Scale Design Trends, Journal of Canadian Petroleum Technology, Vol. 37, 7, July 1998.

National Energy Technology Laboratory, Carbon Sequestration Project Portfolio, FY 2004. USDOE Publication, July 20, 2004.

Ram Reddy, M. K., Xu, Z.P., Lu G. Q. (Max) and Diniz da Costa, J. C., Influence of water on high-temperature CO2 capture using layered double hydroxide derivatives, Ind. Eng. Chem, Res, 47, pp2630-2635, 2008.

Rangwala, H.A., Absorption of Carbon Dioxide into Aqueous Solutions using Hollow Fiber Membrane Contactors, Journal of Membrane Science, 112, 229-240, 1996.

Scholes, C, Kentish, S and Stevens, G. Carbon Dioxide Separation through Polymeric Membranes. Recent Patents on Chemical Engineering, vol. 1, pp. 52-66, 2008.

Scholes, C, Kentish, S and Stevens, G, 2008. The effect of condensable minor components on the gas separation performance of polymeric membranes for carbon dioxide capture. In: Proceedings of GHGT-9 -the 9th International Conference on Greenhouse Gas Control Technologies, Washington DC, USA, 16-20 November. Elsevier, Science Direct and Energy Procedia, published online at

Skinner, F.D., McIntush, K.E., and Murff, M.C., Amine-based Gas Sweetening and Claus Sulfur Recovery Process Chemistry and Waste Stream Survey – Technical Report, Gas Research Institute, December 1995.

Smith, K., Ghosh, U., Khan, A., Simioni, M., Endo, K, Zhao, X, Kentish, S, Qader, A, Hooper, B and Stevens, G,. Recent developments in solvent absorption technologies at the CO2CRC in Australia. Greenhouse Gas Control Technologies 9, Proceedings of the 9th International Conference on Greenhouse Gas Control Technologies (GHGT-9), vol. 1 (1), pp. 1549-1555, 2009.

Thambimuthu, K., Davison, J. and Gupta, M., CO2 Capture and Reuse, in Proceedings of a IPCC Workshop on Carbon Dioxide Capture and Storage, Regina, Canada, November 18-21, 2002.

Veawab, A., Aroonwilas, A., Chakma, A. and Tontiwachwuthikul, P. Solvent Formulation for CO2 Separation from Flue Gas Streams. Presented at First National Conference on Carbon Sequestration, Washington, DC, May 15-17, 2001.

White, C.M., Strazisar, B.R., Granite, E.J., Hoffman, J.S. and Pennline, H.W. Separation and Capture of CO2 from Large Stationary Sources and Sequestration in Geological Formations –Coalbeds and Deep Saline Aquifers, I Air & Waste manage. Assoc. 33:645-715, June 2003.

Wong, S., Gunter, W.D. and Bachu, S. Geological Storage of CO2: Options for Alberta, Combustion and Global Climate Change Conference, Calgary, Alberta, May 26-28, 1999.

Yagi, y, Mimura, T, Yonekawa, T and Yoshiyama, R.. : Development and improvement of CO2 capture system, presented at GHGT-8 2006, Trondheim, Norway. Available from [accessed 29 July, 2009]


Overview of capture technologies

4 EPRI. CO2 Capture Status. 2009.

5 EPRI. CO2 Capture Status. 2009.

6 Justin, Zachary, PhD, and Sara Titus. "CO2 Capture and Sequestration Options – Impact on Turbomachinery Design". 2008.