WO2013037128A1

WO2013037128A1 - Separation of gases

Info

Publication number: WO2013037128A1
Application number: PCT/CN2011/079759
Authority: WO
Inventors: Martin Philip Atkins; Nurida Mohd YUSOP; Guangwen CHEN; Quan Yuan; Chunbo YE; Yuchao Zhao; Yong Men; Hengqiang LI
Original assignee: Petroliam Nasional Berhad (Petronas); Dalian Institute Of Chemical Physics (Dicp)
Priority date: 2011-09-16
Filing date: 2011-09-16
Publication date: 2013-03-21

Abstract

Process for separating carbon dioxide from a gaseous feed stream containing carbon dioxide and at least one other gas is provided. The process comprises flowing the gaseous feed stream and a liquid absorbent through a microchannel and recovering a gaseous product stream having reduced carbon dioxide relative to the gaseous feed stream.

Description

Separation of Gases

The present invention relates to the separation of gases, and more specifically to processes for the separation of carbon dioxide from gaseous streams comprising carbon dioxide and at least one other gas using microchannel contactors.

Separation of carbon dioxide is of particular importance in the field of natural gas processing. Unprocessed natural gas comprises predominantly methane, but often contains significant quantities of carbon dioxide, along with other gaseous impurities. In some cases the carbon dioxide content of unprocessed natural gas may be 10 mol% or higher. Removal of carbon dioxide from natural gas is therefore important both to increase the energy content of the natural gas and to prevent corrosion due to the formation of carbonic acid (H₂CO₃). Commercial specifications typically require that processed natural gas as supplied to consumers contains less than 2 mol% of carbon dioxide, more usually less than 1 mol% of carbon dioxide.

The separation of carbon dioxide is also required in the processing of other types of gases. For instance, other sources of combustible hydrocarbon gases, such as biogas and Fischer-Tropsch reaction products often contain significant quantities of carbon dioxide which must be removed in order to increase the energy content of the gas.

Separation of carbon dioxide is further of importance in the treatment of flue gases so as to reduce emissions of carbon dioxide to the atmosphere. So-called "carbon capture and storage" (CCS) is an extremely active field of research, due to the increasing concern about global warming from the greenhouse effect, and the common belief that the build up of carbon dioxide in the atmosphere is a contributing factor.

The technologies used for the purpose of carbon dioxide separation can be divided into three groups: liquid absorbents, solid absorbents or adsorbents, and membranes. Liquid absorbents are by far the most commonly used of these, and can in turn be divided into physical and chemical absorbents. In the field of natural gas processing, liquid chemical absorbents are often preferred, since they generally have higher absorption capacities for carbon dioxide. In particular, carbon dioxide absorption using aqueous solutions of alkanolamines (such as monoethanolamine, diethanolamine, and methyldiethanolamine/piperazine) as chemical absorbents has been proposed. Carbon dioxide can subsequently be desorbed from the absorbent by pressure reduction, referred to as "pressure swing absorption" (PSA) or by temperature increase, referred to as "temperature swing absorption" (TSA). The energy efficiency of chemical absorption processes is determined largely by the energy demands of regenerating the absorbent by desorption of the carbon dioxide.

The reaction of carbon dioxide with an alkanolamine (R₂NH, where R represents an alkanolyl radical) is thought to involve at least the following steps:

- Formation of a zwitterion:

C0₂ + R₂NH <→ R₂NH⁺C0₂ ^"

- Removal of a proton to form a carbamate:

R₂NH⁺C0₂ ^" + R₂NH→ R₂NC0₂ ^" + R₂NH₂

R₂NH⁺C0₂ ^" + H₂0→ R₂NC0₂ ^" + H₃0⁺

R₂NH⁺C0₂ ^" + OH^"→ R₂NC0₂ ^" + H₂0

Overall reaction:

C0₂ + 2R₂NH <→ R₂NC0₂ ^" + R₂NH₂ ⁺

The removal of a proton from the zwitterion to form the carbamate (R₂NC0₂ ^~) is considered to take place instantaneously. The rate determining step of this process is therefore the formation of the zwitterion, the rate of this step being limited principally by the rate of mass transfer of carbon dioxide across the gas-liquid interface. The removal efficiency of carbon dioxide may thus be improved by limiting diffusion distances and increasing the area of the gas-liquid interface. Existing processes for the separation of carbon dioxide from gases using liquid absorbents have the disadvantage that the separation efficiency (i.e. the proportion of carbon dioxide absorbed) can be variable, particularly as gas throughput is increased. In conventional gas-liquid contactor apparatus, such as packed towers, spray columns and bubble columns, mass transfer may be significantly limited, reducing the efficiency and increasing the cost of the carbon dioxide removal process. The use of shear jet absorbers and vortex absorbers, which aim to increase turbulence and thus the interfacial area of the absorbent and the gas, has been proposed as a way of improving the mass transfer rates in gas separations. However, any increase in separation efficiency obtained by these techniques tends to be offset by additional power requirements of jet nozzles and vortex injectors.

The present invention provides improved processes for the separation of carbon dioxide from gaseous streams comprising carbon dioxide and at least one other gas using microchannel contactors. As used herein, the term microchannel refers to a flow channel having a hydraulic diameter of 5.0 mm or less, more preferably 4.0 mm or less, still more preferably 3.0 mm or less, and most preferably 2.0 mm or less. The hydraulic diameter is preferably at least 0.1 mm, more preferably at least 0.2 mm, still more preferably at least 0.3 mm, and most preferably at least 0.4 mm. In particularly preferred embodiments, the microchannel refers to a flow channel having a hydraulic diameter of from 0.5 to 1.5 mm, for instance, 0.5 mm, 0.6 mm, 0.7 mm, 0.8 mm, 0.9 mm, 1.0 mm, 1.1 mm, 1.2 mm, 1.3 mm, 1.4 mm or 1.5 mm. The small hydraulic diameter of the microchannels allows for very small diffusion distances and thus very high coefficients for mass transfer can be obtained. In some cases, the microchannel contactors may provide space-time efficiencies several hundred times greater than those obtainable with conventional reactors. Microchannels may be straight or may include one or more bends or angles so as to increase turbulence within the microchannel. In a preferred embodiment, the microchannel has one or more angled portions with an angle in the range of from 100° to 160°, preferably about 120°. In a particularly preferred embodiment, the microchannel may have a regular zigzag or sawtooth form.

The length of the microchannel is preferably in the range of from 10 mm to 500 mm, more preferably 10 mm to 400 mm, more preferably 10 to 200 mm, still more preferably 10 to 100 mm, and most preferably 20 to 80 mm. For instance, the microchannel may preferably have a length of 20, 30, 40, 50, 60, 70 or 80 mm. As used herein, references to the length of a microchannel relate to the length of the flow path from the microchannel inlet to the microchannel outlet. It will be appreciated that longer microchannels may be preferred in circumstances where it is desired to increase the contact time between a carbon dioxide containing gas and a liquid absorbent. Preferably, the length of the microchannel is from 10 to 500 times the hydraulic diameter of the microchannel, more preferably from 20 to 300 times the hydraulic diameter of the microchannel.

The use of microchannel contactors to date has mostly been in the context of single phase reactions, usually liquid phase reactions. There have been limited investigations into the use of microchannel contactors for multiphase reaction processes such as the catalytic hydrogenation of alkenes, direct fluorination processes, and the synthesis of hydrogen peroxide from hydrogen and oxygen.

The use of microchannels in the separation of gaseous mixtures generally is disclosed in WO 2009/017832, which is directed to processes using a temperature swing apparatus for the absorption and subsequent desorption of gases. However, the disclosure in this document in relation to the separation of carbon dioxide is very limited and no direction is provided as to suitable processing parameters for carbon dioxide separations.

Accordingly, there remains a need in the art for effective solutions to the problem of removing carbon dioxide from gases, which reduce the inefficiencies associated with the conventional procedures described above, and which optimise the operating parameters of microchannel contactors for the removal of carbon dioxide. The present inventors have surprisingly found that the separation of carbon dioxide from gaseous feed streams using known absorbents is highly variable and that carefully selected process parameters are required to separate carbon dioxide efficiently and thereby obtain a gaseous stream having low levels of residual carbon dioxide, for example such as are acceptable for commercial natural gas supplies. In particular it has been found that where the partial pressure of carbon dioxide in the gaseous feed stream is low, absorption of carbon dioxide in microchannel contactors using alkanolamine-based absorbents is in fact favoured by reducing the concentration of alkanolamine in the absorbent well below the levels used in convention alkanolamine- based separations of carbon dioxide. It has also surprisingly been found that, when optimized separation conditions are used, the efficiency of carbon dioxide separation is maintained even at very high flow rates, thus enabling very high gas throughput in the process and underlining the surprising commercial viability of such processes. In a first aspect, the present invention provides a process for separating carbon dioxide from a gaseous feed stream containing 20 mol% or less of carbon dioxide and at least one other gas, the process comprising flowing the gaseous feed stream and a liquid absorbent through a microchannel and recovering a gaseous product stream having reduced carbon dioxide relative to the gaseous feed stream, wherein the liquid absorbent comprises a mixture of water and an alkanolamine containing from 4 to 10 carbon atoms, and wherein the concentration of alkanolamine in the liquid absorbent is from 2 to 20 weight percent.

In accordance with this aspect of the invention, the alkanolamine is preferably selected from monoethanolamine (MEA), diethanolamine (DEA), triethanolamine (TEA), dimethylethanolamine (DMEA), methyldiethanolamine (MDEA), monoisopropanolamine (MIPA), diisopropanolamine (DIPA), triisopropanolamine (TIPA), dimethylisopropanolamine (DMIPA), methyldiisopropanolamine (MDIPA), and diglycolamine (DGA). Particularly preferred amines are MEA, DEA, and MDEA.

The concentration of the alkanolamine in the liquid absorbent is preferably from 5 to 20 weight percent, more preferably 5 to 15 weight percent, still more preferably 8 to 12 weight percent, for example 8 to 10 weight percent or 10 to 12 weight percent. Most preferably, the concentration of the alkanolamine in the liquid absorbent is around 10 weight percent.

It has surprisingly been found that absorbents having a relatively low concentration of alkanolamine in these ranges are far more efficient absorbers of carbon dioxide from gaseous feed streams that contain low quantities of carbon dioxide, for example 20 mol% or less when using microchannel techniques. This surprising result is thought to be attributable to the gas to liquid ratio in the microchannel. Where the concentration of alkanolamine is high, the gas to liquid ratio at particular molar ratio of alkanolamine to carbon dioxide is also high. This is thought to reduce the gas liquid interfacial area, thus reducing mass transfer of carbon dioxide into the absorbent phase.

In accordance with this aspect of the invention, the gaseous feed stream preferably comprises less than 18 weight percent of carbon dioxide, more preferably less than 16 weight percent of carbon dioxide, more preferably less than 14 weight percent of carbon dioxide, more preferably less than 12 weight percent of carbon dioxide, and most preferably less than 10 weight percent of carbon dioxide. The gaseous feed stream preferably comprises more than 2 weight percent of carbon dioxide and more preferably more than 4 weight percent of carbon dioxide. For instance, in preferred embodiments, the gaseous feed stream may comprise from 4 to 6 weight percent of carbon dioxide, from 6 to 8 weight percent of carbon dioxide, or from 8 to 10 weight percent of carbon dioxide.

As noted above, the absorption of carbon dioxide by alkanolamines requires two moles of alkanolamine per mole of carbon dioxide absorbed. In accordance with this aspect of the invention, the molar ratio of alkanolamine to carbon dioxide is preferably at least 2.2:1 , more preferably at least 2.5:1 , still more preferably at least 2.7:1 , still more preferably at least 3.0:1 , still more preferably at least 3.2:1 , and most preferably at least 3.5:1. The molar ratio of alkanolamine is preferably less than 6.0:1 , more preferably less than 5.5:1 , still more preferably less than 5.0:1 , still more preferably less than 4.5:1 and most preferably less than 4.0:1.

In some cases, for instance where it is desired to obtain a gaseous product stream having a very low carbon dioxide content, an even higher ratio of alkanolamine to carbon dioxide may be used, for instance, at least 4.0:1 or at least 4.5:1 or even at least 5:1. It will of course be appreciated that the use of large excesses of alkanolamine leads to an increase in the regeneration cost of the absorbent, and thus, in general, the use of large excesses of alkanolamine should be avoided. The gaseous feed stream is preferably contacted with the absorbent at a pressure of from 100 to 10,000 kPa, more preferably from 500 to 8,000 kPa, still more preferably from 1 ,000 to 6,000 kPa, and most preferably from 2,000 to 6,000 kPa. Further preferred pressure ranges are from 2,000 to 3,000 kPa, from 3,000 to 4,000 kPa, from 4,000 to 5,000 kPa, and from 5,000 to 6,000 kPa.

It will be appreciated that increasing the operating pressure of the microchannel increases the mass transfer driving force of the absorption reaction. Where the molar ratio of alkanolamine to carbon dioxide is relatively high, for instance above 3:1 , then the mass transfer driving force is generally sufficient to obtain high conversion levels even at lower pressures, for example from 1 ,000 to 4,000 kPa. However, increased pressure may be of benefit where the molar ratio is lower, for instance below 3:1. In such cases, it may be appropriate to increase the pressure to at least 3,000 kPa, more preferably at least 4,000 kPa and most preferably at least 5,000 kPa to obtain optimum conversion.

The efficiency of carbon dioxide absorption is reduced at high temperatures, where the equilibrium favours desorption. However, if the reaction temperature is too low, then diffusion rates and reaction rates are reduced, thus a balance needs to be struck between the need to increase the absorption reaction rate and to avoid increasing the rate of the reverse reaction significantly. For these reasons, the reaction is preferably conducted at a temperature in the range of from 10 to 80 °C, more preferably 10 to 60 °C, still more preferably 10 to 40 °C, still more preferably 15 to 30 °C, and most preferably 20 to 30 °C. In accordance with this aspect of the invention, the gaseous feed stream is passed through the microchannel at a gas hourly space velocity (GHSV) in the range of 5,000 to 150,000, more preferably 10,000 to 120,000, still more preferably 20,000 to 100,000, and most preferably from 30,000 to 100,000. As used herein, the term gas hourly space velocity refers to the ratio of gas flux per hour under the reaction conditions to the total volume of the microchannel.

In practice, it has surprisingly been found that the highest level carbon dioxide separation efficiency is obtained when the GHSV is above 30,000 and that a comparable level of separation efficiency is maintained even on increasing the GHSV to as much as 100,000. This surprising result runs contrary to the expectation that carbon dioxide removal would be less efficient at high GHSV. Where the carbon dioxide content of the gaseous feed stream is low, the partial pressure of carbon dioxide and thus the mass transfer driving force for absorption is also low. Under these conditions it would be expected that increased residence time in the microchannel would lead to increased absorption of carbon dioxide. However, it has surprisingly been observed that low GHSV in fact provides the least efficient separation of carbon dioxide. The liquid absorbent is passed through the microchannel at a liquid hourly space velocity (LHSV, which refers to the ratio of liquid flux per hour to the total volume of the microchannel) which is appropriate to maintain the required molar ratio of carbon dioxide to alkanolamine, taking into account the concentration of alkanolamine in the liquid absorbent. Suitable LHSV values can be determined by the skilled person taking into account the concentration of carbon dioxide in the gaseous feed stream using the ideal gas equation (PV = nRT).

In a second aspect, the present invention provides a process for separating carbon dioxide from a gaseous feed stream containing greater than 20 mol% of carbon dioxide and at least one other gas, the process comprising flowing the gaseous feed stream and a liquid absorbent through a microchannel, and recovering a gaseous product stream having reduced carbon dioxide relative to the gaseous feed stream, wherein the liquid absorbent comprises a mixture of water and an alkanolamine containing from 4 to 10 carbon atoms, wherein the molar ratio of alkanolamine to carbon dioxide is at least 2.2:1 and wherein the operating pressure of the microchannel is at least 1 ,500 kPa.

In contrast with gaseous feed streams comprising low levels of carbon dioxide, as discussed above, it has been found that the microchannel separations of gaseous feed streams containing elevated amounts of carbon dioxide are surprisingly sensitive to the operating pressure of the microchannel contactor and the molar ratio of alkanolamine to carbon dioxide. Since the partial pressure of carbon dioxide is higher in these gaseous feed streams, it would be expected that mass transfer rates would be high and that there would therefore be little benefit from the use of excess absorbent and elevated pressures. In practice, however, it is surprisingly found that elevated operating pressure and increased molar ratio of absorbent is of significant benefit in obtaining high levels of carbon dioxide separation. The reasons for this surprising behaviour are not clear, but it is thought that the absorption of a larger amount of carbon dioxide entails a large drop in gas pressure and that this may leave an insufficient driving force for absorption of the residual carbon dioxide in the gas phase.

Thus, in accordance with this aspect of the invention, the molar ratio of alkanolamine to carbon dioxide is preferably at least 2.5:1 , more preferably at least 2.7:1 , still more preferably at least 3.0:1 , still more preferably at least 3.2:1 , and most preferably at least 3.5:1. The molar ratio of alkanolamine is preferably less than 6.0:1 , more preferably less than 5.5:1 , still more preferably less than 5.0:1 , still more preferably less than 4.5:1 and most preferably less than 4.0:1. As noted above, the use of large excesses of absorbent tends to increase the energy requirements for desorption of carbon dioxide when regenerating the absorbent.

The gaseous feed stream is preferably contacted with the absorbent at a pressure of from 2,000 to 10,000 kPa, more preferably from 3,000 to 8,000 kPa, still more preferably from 4,000 to 6,000 kPa, and most preferably around 5,000 kPa.

As above, the alkanolamine is preferably selected from monoethanolamine (MEA), diethanolamine (DEA), triethanolamine (TEA), dimethylethanolamine (DMEA), methyldiethanolamine (MDEA), monoisopropanolamine (MIPA), diisopropanolamine (DIPA), triisopropanolamine (TIPA), dimethylisopropanolamine (DMIPA), methyldiisopropanolamine (MDIPA), and diglycolamine (DGA). Particularly preferred amines are MEA, DEA, and MDEA.

The concentration of the alkanolamine in the liquid absorbent is preferably from 5 to 50 weight percent, more preferably 10 to 50 weight percent, still more preferably 20 to 45 weight percent, and most preferably, the concentration of the alkanolamine in the liquid absorbent from 25 to 35 weight percent. Unlike the gaseous feed streams containing low concentrations of carbon dioxide discussed above, it has been found that increased concentration of alkanolamine does not reduce the absorption rate. Indeed an increased alkanolamine concentration is required to provide adequate absorption capacity for the increased levels of carbon dioxide in the gaseous feed stream.

In accordance with this aspect of the invention, the gaseous feed stream preferably comprises at least 25 weight percent of carbon dioxide, more preferably at least 30 weight percent of carbon dioxide, more preferably at least 35 weight percent of carbon dioxide, more preferably at least 40 weight percent of carbon dioxide, and most preferably at least 45 weight percent of carbon dioxide. The gaseous feed stream preferably comprises less than 60 weight percent of carbon dioxide and more preferably less than 55 weight percent of carbon dioxide.

As discussed above, an increase in the operating pressure of the microchannel increases the mass transfer driving force of the absorption reaction. Where the molar ratio of alkanolamine to carbon dioxide is relatively high, for instance above 3:1 , then a pressure of, for instance, at least 1 ,500 kPa or 2,000 kPa is generally sufficient to obtain high conversion levels. However, increased pressure may be of benefit where the molar ratio is lower, for instance below 3:1. In such cases, it may be appropriate to increase the pressure to at least 3,000 kPa, more preferably at least 4,000 kPa and most preferably at least 5,000 kPa to obtain optimum separation of carbon dioxide.

The efficiency of carbon dioxide absorption is reduced at high temperatures, where the equilibrium favours desorption. However, if the reaction temperature is too low, then diffusion rates and reaction rates are reduced. Thus a balance needs to be struck between the need to increase the absorption reaction rate and to avoid increasing the rate of the reverse reaction significantly. For these reasons, the reaction is preferably conducted at a temperature in the range of from 10 to 80 °C, more preferably 10 to 60 °C, still more preferably 10 to 40 °C, still more preferably 15 to 30 °C, and most preferably 20 to 30 °C. In accordance with this aspect of the invention, the gaseous feed stream is passed through the microchannel at a gas hourly space velocity (GHSV) in the range of 5,000 to 150,000, more preferably 10,000 to 120,000, still more preferably 10,000 to 100,000, still more preferably 10,000 to 80,000 and most preferably from 20,000 to 50,000. In contrast to gaseous feed streams containing low concentrations of carbon dioxide, it has been found that the microchannel separation of gaseous feed streams containing greater than 20 mol% of carbon dioxide is very sensitive to the flow rate in the microchannel. In particular, it has been found that the absorption of carbon dioxide is reduced above a GHSV of 50,000 and particularly above a GHSV of 80,000. It is believed that this effect is also related to the significant drop in carbon dioxide partial pressure once a major proportion of the initial carbon dioxide content of the gaseous feed stream has been absorbed. As a result of the drop in pressure, it is believed that the mass transfer driving force for absorption of the residual carbon dioxide content of the gaseous stream is significantly reduced, and thus an increase in residence time is necessary to ensure high conversion.

As noted above, the liquid absorbent is passed through the microchannel at a liquid hourly space velocity (LHSV) which is appropriate to maintain the required molar ratio of carbon dioxide to alkanolamine, taking into account the concentration of alkanolamine in the liquid absorbent. Suitable LHSV values can be determined by the skilled person taking into account the concentration of carbon dioxide in the gaseous feed stream using the ideal gas equation (PV = nRT). In further embodiments, the present invention provides microchannel processes for the separation of carbon dioxide from gaseous streams using novel ionic liquid absorbents which show particular utility for microchannel separations.

Ionic liquids are a class of compounds which have been the subject of intense research over the past few decades. The term "ionic liquid" as used herein refers to a liquid that is capable of being produced by melting a solid, and when so produced consists solely of ions. The term "ionic liquid" includes both compounds having high melting temperature and compounds having low melting points, e.g. at or below room temperature (i.e. 15 to 30°C). The latter are often referred to as "room temperature ionic liquids" and are often derived from organic salts having pyridinium- and imidazolium-based cations. A feature of ionic liquids is that they have particularly low (essentially zero) vapour pressures. Many organic ionic liquids have low melting points, for example, less than 100°C, particularly less than 80°C, and around room temperature, e.g. 15 to 30°C, and some have melting points well below 0°C.

An ionic liquid may be formed from a homogeneous substance comprising one species of cation and one species of anion, or it can be composed of more than one species of cation and/or anion. Thus, an ionic liquid may be composed of more than one species of cation and one species of anion. An ionic liquid may further be composed of one species of cation, and more than one species of anion.

Ionic liquids generally exhibit a set of useful physicochemical characteristics that typically include extremely low vapour pressure, wide liquid range, non-degradability, non- flammability, good thermal stability and excellent ability to solubilise a large range of compounds. Due to the potential for controlling these properties of ionic liquids by judicious choice of the constituent ions, and the large variety of ions that can be combined to form low-melting salts, ionic liquids have been proposed for a broad range of applications.

Ionic liquids have been proposed as an alternative to chemical and physical acid gas absorbents for a number of reasons including: (i) the possibility of controlling their properties by the selection of the cation and anion components; (ii) the limited tendency of ionic liquids to crystallize under operating conditions; and (iii) the potential to prevent contamination of the gaseous streams by the absorbent due to the negligible vapour pressure of ionic liquids.

Anderson et al. (Accounts of Chemical Research, 2007, volume 40, pages 1208 to 1216) have reviewed the absorption of a number of different gases in pyridinium, imidazolium and ammonium ionic liquids. The molar enthalpies (ΔΗ) of gas dissolution were determined for the group of gases tested, and the low values observed indicate that only physical absorption takes place. In particular, carbon dioxide is said to interact with the ionic liquids by means of dispersion, dipole/induced dipole interactions and electrostatic effects.

The use of ionic liquids as chemical carbon dioxide absorbers has also been reported. Bates et al. (Journal of the American Chemical Society, 2002, volume 124, pages 926 to 927) have reported the use of a basic imidazolium ionic liquid having an amine functionality tethered to the imidazolium cation to sequester carbon dioxide as a carbamate. However, the high viscosity of these ionic liquids both before, and especially after, carbon dioxide sequestration is a serious limitation for their potential use in industrial processes.

Carvalho et al. (Journal of Physical Chemistry B, 2009, volume 1 13, pages 6803 to 6812) have reported the use of 1-butyl-3-methylimidazolium ionic liquids having acetate and trifluoroacetate anions as absorbents for carbon dioxide. This document teaches that purifying the ionic liquid by removal of water prior to use is essential to avoid a reduction in carbon dioxide absorbing capacity which is reported to take place when water is present in the ionic liquid. A number of prior art documents are cited by Carvalho et al., each of which support the deleterious effect of using wet ionic liquids for carbon dioxide absorption.

The absorption of carbon dioxide by ionic liquids containing imidazolium cations is also disclosed by Shiflett et al. (Journal of Physical Chemistry B, 2008, volume 1 12, pages 16654 to 16663). Again, the ionic liquids are purified by removing water under vacuum with heating for a period of 5 days, emphasizing the need for the ionic liquids to be dry. A single phosphonium ionic liquid (tetra-n-butylphosphonium formate) was also analysed, again in the absence of water, and shown to absorb modest amounts of carbon dioxide by a physical absorption mechanism.

Contrary to the teaching in the art to rigorously dry ionic liquids that are used to absorb carbon dioxide, it has been found that selected classes of ionic liquids demonstrate a marked improvement in carbon dioxide absorption capacity in the presence of water when compared to the ionic liquid alone in the absence of water. This is of particular benefit in the context of microchannel separations since many conventional ionic liquids are too viscous for microchannel processing using neat ionic liquids to be a viable option. It has been found that the addition of water to ionic liquids not only provides highly effective carbon dioxides absorbents, but also reduces the viscosity of the ionic liquids sufficiently to enable microchannel processing. Thus, in a further aspect, the present invention provides a process for separating carbon dioxide from a gaseous feed stream containing carbon dioxide and at least one other gas, the process comprising flowing the gaseous feed stream and a liquid absorbent through a microchannel and recovering a gaseous product stream having reduced carbon dioxide relative to the gaseous feed stream, wherein the liquid absorbent comprises a mixture of an ionic liquid and water in a molar ratio of from 10:90 to 90:10 and wherein the ionic liquid has the formula:

[Cat⁺][X ] wherein: [Cat⁺] represents one or more cationic species selected from tetrasubstituted phosphonium cations, tetrasubstituted ammonium cations, trisubstituted sulfonium cations, guanidinium cations and quinolinium cations; and [X ] represents one or more anionic species selected from conjugate bases of acids having a pKa of at least 3.6.

In accordance with this aspect of the invention, [Cat⁺] is preferably selected from tetrasubstituted phosphonium cations, tetrasubstituted ammonium cations, and trisubstituted sulfonium cations having the formulae:

[P(R^a)(R^b)(R^c)(R^d)]⁺, [N(R^a)(R^b)(R^c)(R^d)]⁺ and [S(R^b)(R^c)(R^d)]⁺ wherein R^a, R^b, R°, and R^d are each independently selected from a Ci to C₂₀ straight chain or branched alkyl group, a C₃ to C₈ cycloalkyl group, or a C₆ to do aryl group, or wherein any two of R^a, R^b, R°, and R^d may together form a saturated methylene chain of the formula -(CH₂)_q-, where q is an integer of from 4 to 7, or an oxyalkylene chain of the formula -(CH₂)₂-0-(CH₂)₂-, wherein said alkyl, cycloalkyl or aryl groups, said methylene chain, or said oxyalkylene chain are unsubstituted or may be substituted by one to three groups selected from: Ci to C₆ alkoxy, C₂ to Ci₂ alkoxyalkoxy, C₆ to Ci₀ aryl, -CN, -OH, -N0₂, -C0₂(Ci to C₆)alkyl, -OC(0)(Ci to C₆)alkyl, C₇ to C₃o aralkyl C₇ to C₃o alkaryl, and -N(R^Z)₂, where each R^z is independently selected from hydrogen, methyl, ethyl, n-propyl and iso-propyl, and wherein R^b may also be hydrogen. More preferably, [Cat⁺] is selected from tetrasubstituted phosphonium cations and tetrasubstituted ammonium cations having the formulae:

[P(R^a)(R^b)(R^c)(R^d)]⁺ and [N(R^a)(R^b)(R^c)(R^d)]⁺ wherein: R^a, R^b, R°, and R^d as defined above.

Still more preferably, [Cat⁺] is selected from tetrasubstituted phosphonium cations having the formula:

[P(R^a)(R^b)(R^c)(R^d)]⁺ wherein: R^a, R^b, R°, and R^d as defined above.

In the tetrasubstituted phosphonium cations, tetrasubstituted ammonium cations, trisubstituted sulfonium cations defined above, R^a, R^b, R°, and R^d (where present) are preferably each independently selected from a Ci to C16 straight chain or branched alkyl group, or any two of R^a, R^b, R°, and R^d may together form a methylene chain of the formula -(CH₂)_q-, where q is an integer of from 4 or 5.

More preferably, R^a (where present), R^b, R°, and R^d are preferably each independently selected from a Ci to C16 straight chain or branched alkyl group. Examples of preferred alkyl groups include: methyl, ethyl, n-propyl, n-butyl, n-pentyl, n-hexyl, n-heptyl, n-octyl, n- nonyl, n-decyl, n-undecyl, n-dodecyl, n-tridecyl, and n-tetradecyl.

Still more preferably, R^a (where present), R^b and R° are preferably each independently selected from a Ci to Cs straight chain or branched alkyl group, and still more preferably a Ci to C₄ straight chain or branched alkyl group, and R^d is preferably a Ci to C16 straight chain or branched alkyl group, and still more preferably a Ci to Cs straight chain or branched alkyl group. Still more preferably, R^a (where present), R^b and R° are each the same Ci to Cs straight chain or branched alkyl group and most preferably the same Ci to C₄ straight chain or branched alkyl group, and R^d is preferably a Ci to C16 straight chain or branched alkyl group, and most preferably a Ci to C₈ straight chain or branched alkyl group.

Preferably, R^d is different from each of R^a (where present), R^b and R°.

In a further preferred embodiment, two of R^a (where present), R^b, R°, and R^d taken together form a saturated methylene chain of the formula -(CH₂)_q-, where q is an integer of from 4 to 7, or an oxyalkylene chain of the formula -(CH₂)₂-0-(CH₂)₂-. Preferably, q is an integer of 4 or 5.

Examples of preferred tetrasubstituted phosphonium cations and tetrasubstituted ammonium cations and trisubstituted sulfonium cations in accordance with the present invention, include those where R^a (where present), R^b and R° are each the same alkyl group selected from ethyl, n-butyl and n-hexyl, and where R^d is selected from methyl, ethyl, n-butyl, n-hexyl, n-octyl, n-decyl, n-dodecyl, and n-tetradecyl.

In particularly preferred embodiments, the tetrasubstituted phosphonium cations and tetrasubstituted ammonium cations used in accordance with the present invention are non-symmetrical. As used herein, the term non-symmetrical means that at least one of R^a, R^b, R°, and R^d is different from each of the other three of R^a, R^b, R°, and R^d. For example, preferred non-symmetrical cations include those in which R^d is different from each of R^a, R^b, and R°, wherein R^a, R^b, and R° may be the same or different.

Specific examples of phosphonium cations that may be used in accordance with the present invention include n-butyl-triethylphosphonium, n-hexyl-triethylphosphonium, n- octyl-triethylphosphonium, tetra-n-butylphosphonium, n-hexyl-tri-n-butylphosphonium, n- octyl-tri-n-butylphosphonium, n-decyl-tri-n-butylphosphonium, n-dodecyl-tri-n- butylphosphonium, n-octyl-tri-n-hexylphosphonium, n-decyl-tri-n-hexylphosphonium, n- dodecyl-tri-n-hexylphosphonium, and n-tetradecyl-tri-n-hexylphosphonium. Particularly preferred phosphonium cations include tetra-n-butylphosphonium and n- octyl-tri-n-butylphosphonium.

Specific examples of ammonium cations that may be used in accordance with the present invention include tetraethylammonium, n-butyl-triethylammonium, n-hexyl- triethylammonium, n-octyl-triethylammonium, methyl-tri-n-butylammonium, tetra-n- butylammonium, n-hexyl-tri-n-butylammonium, n-octyl-tri-n-butylammonium, n-decyl-tri-n- butylammonium, n-dodecyl-tri-n-butylammonium, n-octyl-tri-n-hexylammonium, n-decyl- tri-n-hexylammonium, n-dodecyl-tri-n-hexylammonium, n-tetradecyl-tri-n- hexylammonium, choline.

Particularly preferred ammonium cations include tetraethylammonium and methyl-tri-n- butylammonium. Further examples of cyclic ammonium cations include those wherein two of R^a (where present), R^b, R°, and R^d taken together form a saturated methylene chain of the formula - (CH₂)₄- (pyrrolidinium), or of the formula -(CH₂)₅- (piperidinium), or an oxyalkylene chain of the formula -(CH₂)₂-0-(CH₂)₂- (morpholinium), and wherein the other two of R^a (where present), R^b, R°, and R^d are as defined above.

Where, [Cat⁺] is a quinolinium cation, it preferably has the formula:

wherein: R^a, R^b, R°, R^d, R^e, R^f, R⁹, R^h and R' are each independently selected from hydrogen, a Ci to C₂₀ straight chain or branched alkyl group, a C₃ to C₈ cycloalkyl group, or a C₆ to do aryl group, or any two of R^b, R°, R^d, R^e, R^f, R^h and R' attached to adjacent carbon atoms may form a saturated methylene chain -(CH₂)_q- wherein q is from 3 to 6, and wherein said alkyl, cycloalkyl or aryl groups, or said methylene chain are unsubstituted or may be substituted by one to three groups selected from: Ci to C₆ alkoxy, C₂ to Ci₂ alkoxyalkoxy, C₆ to Cio aryl, -CN, -OH, -N0₂, -C0₂(Ci to C₆)alkyl, -OC(0)(Ci to C₆)alkyl, C₇ to C₃o aralkyl C₇ to C₃o alkaryl, and -N(R^Z)₂, where each R^z is independently selected from hydrogen, methyl, ethyl, n-propyl and iso-propyl. In the above quinolinium cations, R^a is preferably selected from Ci to C₂₀ linear or branched alkyl, more preferably C₂ to C₂₀ linear or branched alkyl, still more preferably C₂ to C-I6 linear or branched alkyl, and most preferably C₄ to do linear or branched alkyl. Examples of suitable R^a groups include ethyl, butyl, hexyl, octyl, nonyl, decyl, undecyl, dodecyl, tridecyl, tetradecyl, pentadecyl, hexadecyl, heptadecyl and octadecyl.

In the above quinolinium cations, R^b, R°, R^d, R^e, R^f, R^h and R' are preferably independently selected from hydrogen and Ci to C5 linear or branched alkyl, and more preferably R^b, R°, R^d, R^e, and R^f are hydrogen. Examples of preferred quinolinium and cations which may be used in accordance with the present invention include: A/-(C₈-Ci₈)alkyl-quinolinium, and A/-(C₈-Ci₈)alkyl-6- methylquinolinium.

Where, [Cat ] is a guanidinium cation, it preferably has the formula:

wherein: R^a, R^b, R°, R^d, R^e, and R^f are each independently selected from a

Ci to C₂₀ straight chain or branched alkyl group, a C₃ to C₈ cycloalkyl group, or a C₆ to do aryl group, or any two of R^b, R°, R^d, R^e, R^f, R^h and R' attached to adjacent carbon atoms may form a saturated methylene chain -(CH₂)_q- wherein q is from 3 to 6, and wherein said alkyl, cycloalkyl or aryl groups, or said methylene chain are unsubstituted or may be substituted by one to three groups selected from: Ci to C₆ alkoxy, C₂ to Ci₂ alkoxyalkoxy, C₆ to do aryl, -CN, -OH, -N0₂, -C0₂(d to C₆)alkyl, -OC(0)(d to C₆)alkyl, C₇ to C₃o aralkyl C₇ to C₃o alkaryl, and -N(R^Z)₂, where each R^z is independently selected from hydrogen, methyl, ethyl, n-propyl and iso-propyl.

In the above guanidinium cations, R^a is preferably selected from Ci to C₂₀ linear or branched alkyl, more preferably C₂ to C₂₀ linear or branched alkyl, still more preferably C₂ to C-I6 linear or branched alkyl, and most preferably C₄ to do linear or branched alkyl. Examples of suitable R^a groups include ethyl, butyl, hexyl, octyl, nonyl, decyl, undecyl, dodecyl, tridecyl, tetradecyl, pentadecyl, hexadecyl, heptadecyl and octadecyl.

In the above guanidinium cations, R^b, R°, R^d, R^e and R^f are preferably selected from Ci to do linear or branched alkyl, more preferably, Ci to d linear or branched alkyl, and most preferably R^b, R°, R^d, R^e and R^f are each a methyl group.

As noted above, [X ] represents an anionic species which is a conjugate base of an acid having a pKa of 3.6 or more. More preferably, [X ] represents an anionic species which is a conjugate base of an acid having a pKa of 4.0 or more, still more preferably 5.0 or more, still more preferably 6.0 or more, and most preferably 7.0 or more.

Still more preferably, [X ] represents an anionic species which is a conjugate base of an acid having a pKa of 15.0 or less, more preferably 14.0 or less, more preferably 13.0 or less, still more preferably 12.0 or less, still more preferably 1 1.0 or less and most preferably 10.0 or less.

In one preferred embodiment, [X ] is selected from conjugate bases of carboxylic acids. More preferably, [X] is selected from anions having the formula [R^xC0₂]^~, wherein R^x is selected from hydrogen, a Ci to do straight chain or branched alkyl group, a C₃ to C₈ cycloalkyi group, or a C₆ to do aryl group, wherein said alkyl, cycloalkyi or aryl groups are unsubstituted or may be substituted by one or more groups selected from -F, -CI, - OH, -CN, -NO₂, -SH, -CO₂H, and =0.

More preferably, R^x is selected from hydrogen, or a Ci to C₁0 straight chain or branched alkyl group, wherein said alkyl group is optionally substituted by one or more groups selected from -F and -OH. Examples of anions which may be used in accordance with the present invention include: formate, acetate, hydroxyacetate, propanoate, lactate, butanoate, isobutanoate, pivalate, pyruvate, thiolactate, benzoate, oxalate monoanion, tartrate monoanion, malonate monoanion, adipate monoanion, heptanedioic acid monoanion, and octanedioic acid monoanion.

More preferably, the anion is selected from: formate, acetate, hydroxyacetate, propanoate, lactate, butanoate, isobutanoate and pivalate.

In a preferred embodiment, [X ] is acetate.

In another preferred embodiment, [X ] is formate.

The ionic liquids which may be used in accordance with this aspect of the invention also include those in which [X ] is selected from phosphate dianions and phosphonate dianions.

More preferably, [X] is selected from phosphate anions having the formula [R^xOP(0)0₂]^2" and phosphonate anions having the formula [R^XP(0)0₂]^2", wherein R^x is selected from hydrogen, a Ci to C₁0 straight chain or branched alkyl group, a C₃ to C₈ cycloalkyi group, a C₆ to C₁0 aryl group, a C₆ to Ci aralkyi group, or a C₆ to Ci alkaryl group, wherein said alkyl, cycloalkyi, aryl, aralkyi, or alkaryl groups are unsubstituted or may be substituted by one or more groups selected from -F, -CI, -Br, -I , -OH, -CN, -N0₂, -SH, and =0. More preferably, R^x is selected from hydrogen or a Ci to do straight chain or branched alkyl group, wherein said alkyl group is optionally substituted by one or more groups selected from -F, -CI, -Br, -I , and -OH. Still more preferably, R^x is selected from hydrogen or a Ci to d straight chain or branched alkyl group, wherein said alkyl group is optionally substituted by one or more groups selected from -F, -CI, -Br, -I , and -OH.

Examples of suitable R^x groups include hydrogen, methyl, ethyl, n-propyl, n-butyl, n- pentyl, n-hexyl, n-heptyl, n-octyl, fluoromethyl, chloromethyl, bromomethyl, iodomethyl, and hydroxymethyl.

Preferably, [X ] is selected from phosphate anions having the formula [R^xOP(0)0₂]^2~, wherein R^x is as defined above.

Examples of particularly preferred phosphate anions include [HOP(0)0₂]^2~, [MeOP(0)0₂]^2~, [EtOP(0)0₂]^2", [n-PrOP(0)0₂]^2", and [N-BuOP(0)0₂]^2".

Still more preferably, [X ] is [HOP(0)0₂]^2" (also referred to herein as [HP0₄]^2" or hydrogen phosphate).

The ionic liquids which may be used in accordance with this aspect of the invention further include those in which [X ] is selected from dicarboxylate dianions having the formula [0₂C-R^y-C0₂]^2", wherein R^y represents a Ci to straight chain or branched alkylene or alkenylene chain, a Ci to C₆ cycloalkylene group, or a C₆ arylene group, wherein said alkylene, alkenylene, cycloalkylene or arylene groups are unsubstituted or may be substituted with one or more groups selected from -F, -CI, -Br, -I , -OH, -CN, -N0₂, -SH, -C0₂H and =0. Preferably said alkylene, alkenylene, cycloalkylene or arylene groups are unsubstituted or substituted with one or more groups selected from -OH, -C0₂H and -SH. More preferably, [X] may be one or more dicarboxylate dianions selected from oxalate dianion, malonate dianion, succinate dianion, glutarate dianion, adipate dianion, pimelate dianion, methylmalonate dianion, fumarate dianion, maleate dianion, methyl succinate dianion, malate dianion, tartrate dianion, citrate dianion, itaconate dianion, mesaconate dianion, o-phthalate dianion, m-phthalate dianion, p-phthalate dianion, aspartate dianion, glutamate dianion, octanedioic acid dianion and heptanedioic acid dianion.

Still more preferably, [X ] may be one or more dicarboxylate dianions selected from glutarate dianion, adipate dianion, pimelate dianion, methylmalonate dianion, fumarate dianion, maleate dianion, methyl succinate dianion, malate dianion, itaconate dianion, and mesaconate dianion.

In a further preferred embodiment, [X ] may be selected from ascorbate anion and urate anion.

In view of the foregoing disclosure, it will be appreciated that the present invention is not limited to ionic liquids comprising cations and anions having only a single charge. Thus, the formula [Cat⁺][X^~] is intended to encompass ionic liquids comprising, for example, doubly, triply and quadruply charged cations and/or anions. The relative stoichiometric amounts of [Cat⁺] and [X ] in the ionic liquid are therefore not fixed, but can vary to take account of cations and anions with multiple charges. For example, the formula [Cat⁺][X^~] should be understood to include ionic liquid species having the formulae [Cat⁺]₂[X² ]; [Cat²⁺][X^"]₂; [Cat²⁺][X² ]; [Cat⁺]₃[X³ ]; [Cat³⁺][X^"]₃ and so on. The ionic liquids used in accordance with the present invention preferably have a melting point of 200 °C or less, more preferably 150 °C or less, and most preferably 100 °C or less. However, ionic liquids with melting points falling outside this range may also be used, provided that the mixture of an ionic liquid and water is liquid at the operating temperature of the process.

Thus, in a preferred embodiment of the invention, the mixture of an ionic liquid and water has a melting point of 100 °C or less, more preferably 80 °C or less, more preferably 50 °C or less, still more preferably 30 °C or less, and most preferably 25 °C or less. The molar ratio of ionic liquid to water is preferably in the range of from 80:20 to 20:80, more preferably 70:30 to 30:70, still more preferably in the range of from 60:40 to 40:60, still more preferably in the range of from 55:45 to 45:55, and most preferably the molar ratio of ionic liquid to water is around 50:50.

In accordance with this aspect of the invention, the gaseous feed stream is preferably contacted with the carbon dioxide absorbent at a temperature of from 10 to 80 °C, more preferably from 10 to 50 °C and most preferably from 20 to 30 °C. For example, the gaseous feed stream may be contacted with the carbon dioxide absorbent at a temperature at or around 25 °C.

The gaseous feed stream is preferably contacted with the absorbent at a pressure of from 100 to 10,000 kPa, more preferably from 200 to 8,000 kPa, still more preferably from 500 to 5,000 kPa, and most preferably from 1000 to 5,000 kPa. Further preferred pressure ranges are from 1000 to 2000 kPa, from 2000 to 3000 kPa, from 3000 to 4000 kPa, and from 4000 to 5000 kPa.

In accordance with this aspect of the invention, the gaseous feed stream is passed through the microchannel at a gas hourly space velocity (GHSV) in the range of 5,000 to 150,000, more preferably 10,000 to 120,000, still more preferably 10,000 to 100,000, and most preferably from 10,000 to 50,000. As used herein, the term gas hourly space velocity refers to the total volume of the microchannel. The processes of the present invention may be used to separate carbon dioxide from a number of different gaseous feed streams that comprise carbon dioxide together with at least one other gas having lower affinity than carbon dioxide for the absorbents defined above. Examples of other gases which may be present in the feed stream include N₂, 0₂, H₂, He, Ne, Ar, Xe, hydrocarbon gases. In preferred embodiments the gaseous feed stream comprises carbon dioxide and at least one hydrocarbon gas, which is preferably methane or ethane, and is most preferably methane. In particularly preferred embodiments, the gaseous feed stream comprises or consists of a natural gas, or a natural gas derivative. In other preferred embodiments, the gaseous feed stream comprises or consists of a biogas or a hydrocarbon gas derived from a Fischer-Tropsch process.

The processes of the present invention may be operated using a plurality of microchannels operated in parallel and/or in series. The processes of the present invention are fully scalable and the use of multiple microchannels allows the process of the invention to be scaled up to process large quantities of feed gas. In preferred embodiments, the processes of the present invention are operated using a plurality of microchannels operated in parallel, for instance 10 or more microchannels, more preferably 20 or more microchannels, still more preferably 50 or more microchannels and most preferably 100 or more microchannels. In some cases, the number of microchannels in parallel may be 200 or more, 500 or more, 1000 or more, 2000 or more, 5000 or more, or even 10,000 or more. It will be appreciated that modern microchannel contactor design readily enables a consistent distribution of liquid and gaseous phases to be provided across multiple microchannels so as to ensure consistent contact ratios in the microchannels.

In further embodiments, the processes of the present invention may be operated in series. For instance, an ionic liquid process may be used in series with an alkanolamine process. In another embodiment, an alkanolamine process for gaseous feed streams containing greater than 20 mol % carbon dioxide, as defined above may be used to remove carbon dioxide from a gaseous feed stream, and the gaseous product stream obtained may be passed to a further alkanolamine process for gaseous feed streams containing 20 mol% or less of carbon dioxide. Preferably, where the processes of the present invention are conducted in series, each stage of the series comprises a plurality of microchannels operated in parallel as described above.

The present invention will now be described by way of Example and with reference to the following Figures, in which:

Figure 1 shows a pilot apparatus for the separation of carbon dioxide from gaseous feed streams using microchannels. Figures 2 to 16 show results for the separation of carbon dioxide according to the processes of the invention as outlined in Examples 2 to 9.

EXAMPLES

Example 1 - Pilot apparatus A schematic diagram of a pilot apparatus for examining the absorption of carbon dioxide using microchannel contactors is shown in Figure 1. The apparatus shown in Figure 1 was used to obtain the results in each of the following Examples.

A gaseous feed stream 105 comprising carbon dioxide and nitrogen is supplied from gas cylinder 100 and fed to the microchannel contactor 125 via reducing valve 110, mass flow controller 115 and check valve 120. An aqueous liquid absorbent 135 is fed to the microchannel contactor 125 via pump 140 and check valve 145. The microchannel contactor 125 has a gas inlet 150, a liquid inlet 155 and a mixed fluid outlet 165 in fluid communication with n parallel microchannels 160.

For convenience, only four microchannels are shown in Figure 1. However, unless stated otherwise, the Examples below were conducted with a microchannel contactor having 8 microchannels. The microchannels used in the Examples had a length of 60 mm and a hydraulic diameter of 0.45 mm.

The gaseous feed stream and the liquid absorbent enter the n parallel microchannels 160 via gas inlet 150 and liquid inlet 155, respectively. The gas and liquid phases are contacted in the microchannels 160 and carbon dioxide is absorbed from the gas phase into the liquid phase. The entire microchannel contactor 125 is immersed in a water bath at constant temperature so as to obtain uniform reaction conditions and the system pressure is controlled by a back pressure valve 170. Gas hourly space velocity (GHSV) and liquid hourly space velocity (LHSV) is calculated based on the values of the inlet temperature and system pressure. The gas-liquid mixture obtained from the microchannel outlet 165 is passed to a gas- liquid separator 175 via a PTFE eduction tube with an inner diameter of 6 mm. The gas and liquid phases are separated, and the gas stream 180 is passed to a cold trap 185 to remove water vapour. The residual carbon dioxide content of the gas stream 190 is analyzed by a gas component analyzer 195 such as a gas chromatograph (GC) or a carbon dioxide sensor. The liquid stream 200 is passed to spent absorbent tank 205, where it is collected for subsequent regeneration and reuse. The carbon dioxide absorption was calculated to evaluate the performance of the microchannel contactor under each set of conditions examined. The absorption was calculated by the following equation. n(C0₂ , in) - n(C0₂ , out)

n(C0₂ , in) wherein n(C0₂, in) and n (CO 2, out) indicate the molar flow rates of carbon dioxide in the gaseous feed stream 105 and the gaseous product stream 190, respectively.

Example 2 - Effect of GHSV and molar ratio of alkanolamine to carbon dioxide

The effects of the gas hourly space velocity (GHSV) and the molar ratio of DEA to carbon dioxide at a DEA concentration of 30 wt% have been examined for gaseous feed streams having initial carbon dioxide concentrations of 36 vol% and 49.2 vol% (16 microchannels). The results are shown in Figures 2 and 3 respectively.

It can be seen the conversion increases with increased molar ratio of DEA to carbon dioxide. As discussed above, it is believed that the formation of zwitterions is a rate- determining step. As the molar ratio of DEA to carbon dioxide increases, the chance of collision between carbon dioxide and DEA molecules also increases, thereby increasing the rate of mass transfer of carbon dioxide into the liquid phase. Due to the large gas- liquid interfacial area in the microchannels, highly effective separation of carbon dioxide is observed.

It can also be seen that optimum carbon dioxide separation is obtained when the GHSV is in the range of from 10,000 to 80,000, although use of increased molar ratios of DEA to carbon dioxide enable very high separation efficiencies to be obtained even outside of this GHSV range. Thus, separation efficiency of greater than 95% can be obtained even at very high GHSV by the use of a molar ratio of DEA to carbon dioxide of 3.70.

As discussed above, the effective gas-liquid interfacial area usually increases with an increase in superficial gas velocity. It has been observed that due to the fast reaction of DEA with carbon dioxide and fast mass transfer rates in the microchannel contactor, carbon dioxide may be efficiently separated even at very short residence times.

Example 3 - Effect of system pressure

The effect of system pressure on carbon dioxide separation is shown in Figure 4. It is observed that the separation increases with the increased of system pressure, although only a relatively small effect is observed when the pressure is greater than 1.5 MPa. Example 4 - Effect of reaction temperature

The effect of reaction temperature on carbon dioxide separation is shown in Figure 5.

It can be seen that the conversion is increased as the reaction temperature is reduced. As the reaction temperature increases, it is believed that both the diffusivity of carbon dioxide in the aqueous solution and the reaction rate constant of carbon dioxide with DEA increase. These are beneficial to increase the overall separation. However, the Henry's coefficient of carbon dioxide in the aqueous DEA solution also decreases with an increase in temperature. The reduced Henry's coefficient is believed to decrease in the mass transfer driving force, and this is considered to outweigh the beneficial effects of the higher diffusivity and reaction rate constant at higher temperatures.

Similarly, at very low temperatures, it is expected that the reduction in the diffusivity of carbon dioxide in the aqueous solution and the reaction rate constant of carbon dioxide with DEA at very low temperatures will outweigh any increase in the Henry's coefficient of carbon dioxide in the aqueous DEA solution. Thus separation performance is also expected to be reduced at very low temperatures. Example 5 - Effect of GHSV on the separation of carbon dioxide from gaseous feed stream having low carbon dioxide content

The effect of the GHSV on the separation of carbon dioxide from a gaseous feed stream comprising 3.74 vol% of carbon dioxide is shown in Figure 6. The experiment was conducted under two different pressures: 3.0 MPa using 16 microchannels and 5.0 MPa using 8 microchannels.

It is observed that where the feed gas stream has low carbon dioxide content, a counterintuitive result is obtained in which separation of carbon dioxide increases as the GHSV increases. At a GHSV of 91285 and a pressure of 5.0 MPa the residual carbon dioxide content of the gaseous product stream could be reduced to as little as 392 ppm

Example 6 - Effect of DEA concentration on the separation of carbon dioxide from a gaseous feed stream having low carbon dioxide content

The effect of the DEA concentration on the separation of carbon dioxide comprising 4.45 vol% of carbon dioxide using aqueous DEA at a concentration of 30 wt% and 16 microchannels is shown in Figure 7.

It is observed that at a molar ratio of DEA to carbon dioxide of 2.28 the separation performance is very poor with less than 50 mol% of the carbon dioxide from the gaseous feed stream absorbed. To obtain separation of greater than 90 mol% of the carbon dioxide from the gaseous feed stream it is necessary to increase he molar ratio of DEA to carbon dioxide to 7.74. As noted above, the use of such a high ratio of DEA to carbon dioxide is uneconomical due to the increased energy required to regenerate the absorbent. It will be appreciated that when using a lower molar ratio of DEA to carbon dioxide where the DEA is at elevated concentration, the gas-liquid ratio is very high in the case of feed gases containing low amounts of carbon dioxide. It is believed that the reduction in gas-liquid interfacial area entails a reduction in mass transfer rates of carbon dioxide into the liquid phase. Consequently, where the carbon dioxide content of the feed gas is low, it has been found that it is far preferable to use DEA at comparatively low concentration.

Figure 8 shows a corresponding process using a molar ratio of DEA to carbon dioxide where the concentration of DEA is 5 wt% or 10 wt%, again with 16 microchannels. In view of the fast rate of formation of the zwitterion, it has been found that the reduction in DEA concentration has little effect on the rate of carbon dioxide absorption, compared to the large improvement in carbon dioxide absorption obtained by the decreased gas to liquid ratio obtained at lower DEA concentrations.

It is believed that the reason why the results of DEA of 5 wt% concentration providing lower carbon dioxide absorption than DEA of 10 wt% concentration is mainly due to a lower mass transfer driving force. Considering the energy required by the regeneration and the transport systems as well as the above analysis, it has surprisingly been found that DEA with concentration of 10wt% provides optimum results for the separation of carbon dioxide from a gaseous feed stream of low initial carbon dioxide concentration. Example 7 - Effect of pressure and molar ratio of DEA to carbon dioxide on the separation of carbon dioxide from a gaseous feed stream having low carbon dioxide content

Figure 9 shows the effect of the microchannel operating pressure and the molar ratio of DEA to carbon dioxide on the separation of carbon dioxide from a gaseous feed stream having a carbon dioxide concentration of 4.45 vol% when using DEA of 10 wt% concentration using 16 microchannels.

It can be seen that at the molar ratio of 3.72, the conversion reaches almost 100% almost independently of whether the microchannel operating pressure is 3.0 MPa or 5.0 MPa. It is believed that the mass transfer driving force is close to a maximum even at a pressure of 3.0 MPa, and thus a further increase in pressure has negligible effect on the separation efficiency. Where the molar ratio is 2.28, however, the reduction in driving force is evident, and a clear improvement is obtained on increasing the microchannel operating pressure from 3.0 MPa to 5.0 MPa.

Figure 10 shows the residual carbon dioxide content of the gaseous product stream obtained from the process shown in Figure 9. It is observed that, where the molar ratio of DEA to carbon dioxide is 3.72, the carbon dioxide content of the gaseous product stream is always below 0.2 vol% regardless of whether the pressure is 3.0 MPa or 5.0 MPa. On the other hand, where the molar ratio is 2.28, the carbon dioxide content of the gaseous product stream can be as low as 0.179 vol% when system pressure is 5MPa, while at 3MPa system pressure, the carbon dioxide content of the outlet gas may be slightly higher than 0.2vol% due to the limited mass transfer driving force. Example 8 - Effect of molar ratio of ionic liquid to carbon dioxide

The effect of the molar ratio of the ionic liquid tributylmethylphosphonium formate to carbon dioxide on the absorption of carbon dioxide is shown in Figures 1 1 , 12 and 13 at temperatures of 25 °C, 35 °C and 45 °C, respectively, and using 16 microchannels. In each case, the ionic liquid was used in an aqueous solution at a concentration of 40 wt% and the gaseous feed stream has a concentration of 17.2 vol%.

As shown in each of these Figures, an improvement in carbon dioxide absorption is obtained using a higher molar ratio of ionic liquid to carbon dioxide, although at higher temperatures this effect is reversed.

At higher temperatures overall carbon dioxide absorption is reduced. As discussed above, this effect is thought to be due to the reduced Henry's coefficient of carbon dioxide in the ionic liquid-water mixture predominating over any increase in reaction rate and mass transfer. Example 9 - Effect of temperature on separation of carbon dioxide from a gaseous feed stream having carbon dioxide content of 17.2 vol%

Figures 14 and 15 show the effect of temperature on the separation of carbon dioxide from a gaseous feed stream having an initial carbon dioxide content of 17.2 mol% using the ionic liquid tributylmethylphosphonium formate in an aqueous solution at a concentration of 40 wt% with 16 microchannels. The molar ratio is 2.27 and 3.72 for Figures 14 and 15 respectively. In Figure 14, it can be seen that the absorption pattern initially increases, then reduces and then increases again as the GHSV increases, although this pattern is not obvious at 25 °C.

With reference to Figure 15, it can be seen that the conversion shows an increasing trend at 25 °C and a downward trend at 45 °C with increasing GHSV. At 35 °C, an initial increase in absorption with GHSV is followed by a decrease as GHSV exceeds around 25,000.

The reasons for this complex behaviour are not fully understood, but are believed to be attributable to a number of competing factors such as viscosity, surface tension, Henry's coefficient and equilibrium constant, all of which are significantly influenced by temperature. In particular, it is believed that high viscosity of the ionic liquid solution may be a significant factor in reducing the rate of carbon dioxide absorption. Example 10 - Use of multiple microchannels

Figure 16 shows the effect of using multiple microchannels for the separation of carbon dioxide from a gaseous feed stream containing 36.2 vol% carbon dioxide using a molar ratio of DEA to carbon dioxide of 2.88 and a DEA concentration of 30 wt%. It is observed that the excellent separation of carbon dioxide obtained using 8 microchannels is maintained when the number of microchannels is increased to 16. It is believed that the use of well-designed fluid distributors to maintain a constant gas-liquid distribution in each of the microchannels makes the separation efficiency virtually independent of the number of microchannels used.

Claims

1. A process for separating carbon dioxide from a gaseous feed stream containing 20 mol% or less of carbon dioxide and at least one other gas, the process comprising flowing the gaseous feed stream and a liquid absorbent through a microchannel and recovering a gaseous product stream having reduced carbon dioxide relative to the gaseous feed stream, wherein the liquid absorbent comprises a mixture of water and an alkanolamine containing from 4 to 10 carbon atoms, and wherein the concentration of alkanolamine in the liquid absorbent is from 2 to 20 weight percent.

2. A process according to Claim 1 , wherein the concentration of the alkanolamine in the liquid absorbent is from 5 to 20 weight percent. 3. A process according to Claim 2, wherein the concentration of the alkanolamine in the liquid absorbent is from 5 to 15 weight percent.

4. A process according to any one of the preceding claims, wherein the gaseous feed stream comprises less than 18 weight percent of carbon dioxide, more preferably less than 16 weight percent of carbon dioxide, more preferably less than 14 weight percent of carbon dioxide, more preferably less than 12 weight percent of carbon dioxide, and most preferably less than 10 weight percent of carbon dioxide. 5. A process according to any one of the preceding claims, wherein the gaseous feed stream comprises more than 2 weight percent of carbon dioxide and more preferably more than 4 weight percent of carbon dioxide.

6. A process according to any one of the preceding claims, wherein the molar ratio of alkanolamine to carbon dioxide is at least 2.2:1 , more preferably at least 2.5:1 , still more preferably at least 2.7:1 , still more preferably at least 3.0:1 , still more preferably at least 3.2:1 , and most preferably at least 3.5:1.

7. A process according to any one of the preceding claims, wherein the molar ratio of alkanolamine is less than 6.0:1 , more preferably less than 5.5:1 , still more preferably less than 5.0:1 , still more preferably less than 4.5:1 and most preferably less than 4.0:1.

8. A process according to Claim 6, wherein the molar ratio of alkanolamine to carbon dioxide is at least 4.0:1 or at least 4.5:1 or at least 5:1.

9. A process according to any one of the preceding claims, wherein the gaseous feed stream is contacted with the absorbent at a pressure of from 100 to 10,000 kPa, more preferably from 500 to 8,000 kPa, still more preferably from 1 ,000 to 6,000 kPa, and most preferably from 2,000 to 6,000 kPa.

10. A process according to any one of the preceding claims, wherein the gaseous feed stream is contacted with the absorbent at a temperature in the range of from

10 to 80 °C, more preferably 10 to 60 °C, still more preferably 10 to 40 °C, still more preferably 15 to 30 °C, and most preferably 20 to 30 °C.

1 1. A process according to any one of the preceding claims, wherein the gaseous feed stream is passed through the microchannel at a gas hourly space velocity

(GHSV) in the range of 5,000 to 150,000, more preferably 10,000 to 120,000, still more preferably 20,000 to 100,000, and most preferably from 30,000 to 100,000.

12. A process for separating carbon dioxide from a gaseous feed stream containing greater than 20 mol% of carbon dioxide and at least one other gas, the process comprising flowing the gaseous feed stream and a liquid absorbent through a microchannel, and recovering a gaseous stream product having reduced carbon dioxide relative to the gaseous feed stream, wherein the liquid absorbent comprises a mixture of water and an alkanolamine containing from 4 to 10 carbon atoms, wherein the molar ratio of alkanolamine to carbon dioxide is at least 2.2:1 and wherein the operating pressure of the microchannel is at least 1 ,500 kPa.

13. A process according to Claim 12, wherein the molar ratio of alkanolamine to carbon dioxide is at least 2.5:1 , more preferably at least 2.7:1 , still more preferably at least 3.0:1 , still more preferably at least 3.2:1 , and most preferably at least 3.5:1.

14. A process according to Claim 12 or Claim 13, wherein the molar ratio of alkanolamine is preferably less than 6.0:1 , more preferably less than 5.5:1 , still more preferably less than 5.0:1 , still more preferably less than 4.5:1 and most preferably less than 4.0:1.

15. A process according to any one of Claims 12 to 14, wherein the gaseous feed stream is contacted with the absorbent at a pressure of from 2,000 to 10,000 kPa, more preferably from 3,000 to 8,000 kPa, still more preferably from 4,000 to 6,000 kPa, and most preferably around 5,000 kPa.

16. A process according to any one of Claims 12 to 15, wherein the concentration of the alkanolamine in the liquid absorbent is from 5 to 50 weight percent, more preferably 10 to 50 weight percent, still more preferably 20 to 45 weight percent, and most preferably from 25 to 35 weight percent.

17. A process according to any one of Claims 12 to 16, wherein the gaseous feed stream comprises at least 25 weight percent of carbon dioxide, more preferably at least 30 weight percent of carbon dioxide, more preferably at least 35 weight percent of carbon dioxide, more preferably at least 40 weight percent of carbon dioxide, and most preferably at least 45 weight percent of carbon dioxide.

18. A process according to any one of Claims 12 to 17, wherein the gaseous feed stream is contacted with the absorbent at a temperature in the range of from 10 to 80 °C, more preferably 10 to 60 °C, still more preferably 10 to 40 °C, still more preferably 15 to 30 °C, and most preferably 20 to 30 °C.

19. A process according to any one of Claims 12 to 17, wherein the gaseous feed stream is passed through the microchannel at a gas hourly space velocity (GHSV) in the range of 5,000 to 150,000, more preferably 10,000 to 120,000, still more preferably 10,000 to 100,000, still more preferably 10,000 to 80,000 and most preferably from 20,000 to 50,000.

A process according to any one of the preceding claims, wherein the alkanolamine is selected from monoethanolamine (MEA), diethanolamine (DEA), triethanolamine (TEA), dimethylethanolamine (DMEA), methyldiethanolamine (MDEA), monoisopropanolamine (MIPA), diisopropanolamine (DIPA), triisopropanolamine (TIPA), dimethylisopropanolamine (DMIPA), methyldiisopropanolamine (MDIPA), and diglycolamine (DGA).

A process according to Claim 20, wherein the alkanolamine is selected from monoethanolamine, diethanolamine and methyldiethanolamine.

A process for removing carbon dioxide from a gaseous feed stream containing carbon dioxide and at least one other gas, the process comprising flowing the gaseous feed stream and a liquid absorbent through a microchannel and recovering a gaseous product stream having reduced carbon dioxide relative to the gaseous feed stream, wherein the liquid absorbent comprises a mixture of an ionic liquid and water in a molar ratio of from 10:90 to 90:10 and wherein the ionic liquid has the formula:

[Cat⁺][X ] wherein: [Cat⁺] represents one or more cationic species selected from tetrasubstituted phosphonium cations, tetrasubstituted ammonium cations, trisubstituted sulfonium cations, guanidinium cations and quinolinium cations; and

[X ] represents one or more anionic species selected from conjugate bases of acids having a pKa of at least 3.6.

A process according to Claim 22, wherein [Cat⁺] is selected from tetrasubstituted phosphonium cations, tetrasubstituted ammonium cations, and trisubstituted sulfonium cations having the formulae:

[P(R^a)(R^b)(R^c)(R^d)]⁺, [N(R^a)(R^b)(R^c)(R^d)]⁺ and [S(R^b)(R^c)(R^d)]⁺ wherein R^a, R^b, R^c, and R^d are each independently selected from a Ci to C₂₀ straight chain or branched alkyl group, a C₃ to C₈ cycloalkyl group, or a C₆ to Cio aryl group, or wherein any two of R^a, R^b, R°, and R^d may together form a saturated methylene chain of the formula -(CH₂)_q-, where q is an integer of from 4 to 7, or an oxyalkylene chain of the formula -(CH₂)₂-0-(CH₂)₂-, wherein said alkyl, cycloalkyl or aryl groups, said methylene chain, or said oxyalkylene chain are unsubstituted or may be substituted by one to three groups selected from: Ci to C₆ alkoxy, C₂ to Ci₂ alkoxyalkoxy, C₆ to Ci₀ aryl, -CN, -OH, -N0₂, -C0₂(Ci to C₆)alkyl, -OC(0)(Ci to C₆)alkyl, C₇ to C₃o aralkyl C₇ to C₃o alkaryl, and -N(R^Z)₂, where each R^z is independently selected from hydrogen, methyl, ethyl, n-propyl and iso-propyl, and wherein R^b may also be hydrogen.

A process according to Claim 23, wherein [Cat⁺] is selected from tetrasubstituted phosphonium cations and tetrasubstituted ammonium cations having the formulae:

[P(R^a)(R^b)(R^c)(R^d)]⁺ and [N(R^a)(R^b)(R^c)(R^d)] wherein: R^a, R^b, R°, and R^d as defined in Clai

A process according to Claim 24, wherein [Cat⁺] is selected from tetrasubstituted phosphonium cations having the formula:

[P(R^a)(R^b)(R^c)(R^d)]⁺ wherein: R^a, R^b, R°, and R as defined in Claim 23. 26. A process according to Claim 23 or Claim 24, wherein [Cat⁺] is non-symmetrical.

27. A process according to Claim 22, wherein [Cat⁺] is selected from quinolinium cations having the formula:

wherein: R^a, R^b, R°, R^d, R^e, R^f, R⁹, R^h and R' are each independently selected from hydrogen, a Ci to C₂₀ straight chain or branched alkyl group, a C₃ to C₈ cycloalkyi group, or a C₆ to Ci₀ aryl group, or any two of R^b, R°, R^d, R^e, R^f, R^h and R' attached to adjacent carbon atoms may form a saturated methylene chain -(CH₂)_q- wherein q is from 3 to 6, and wherein said alkyl, cycloalkyi or aryl groups, or said methylene chain are unsubstituted or may be substituted by one to three groups selected from: Ci to C₆ alkoxy, C₂ to Ci₂ alkoxyalkoxy, C₆ to Cio aryl, -CN, -OH, -N0₂, -C0₂(Ci to C₆)alkyl, -OC(0)(Ci to C₆)alkyl, C₇ to C₃o aralkyl C₇ to C₃o alkaryl, and -N(R^Z)₂, where each R^z is independently selected from hydrogen, methyl, ethyl, n-propyl and iso-propyl.

28. A process according to Claim 22, wherein [Cat⁺] is selected from guanidinium cations having the formula:

Ci to C₂₀ straight chain or branched alkyl group, a C₃ to C₈ cycloalkyl group, or a C₆ to do aryl group, or any two of R^b, R°, R^d,

R^e, R^f, R^h and R¹ attached to adjacent carbon atoms may form a saturated methylene chain -(CH₂)_q- wherein q is from 3 to 6, and wherein said alkyl, cycloalkyl or aryl groups, or said methylene 5 chain are unsubstituted or may be substituted by one to three groups selected from: Ci to C₆ alkoxy, C₂ to Ci₂ alkoxyalkoxy, C₆ to do aryl, -CN, -OH, -N0₂, -C0₂(d to C₆)alkyl, -OC(0)(d to C₆)alkyl, C₇ to C₃o aralkyl C₇ to C₃o alkaryl, and -N(R^Z)₂, where each R^z is independently selected from hydrogen, methyl, ethyl, n-propyl and 10 iso-propyl.

29. A process according to any one of Claims 22 to 28, wherein [X ] represents an anionic species which is a conjugate base of an acid having a pKa of 4.0 or more, more preferably 5.0 or more, still more preferably 6.0 or more, and most

15 preferably 7.0 or more.

30. A process according to Claim 29, wherein [X ] represents an anionic species which is a conjugate base of an acid having a pKa of 15.0 or less, more preferably 14.0 or less, more preferably 13.0 or less, still more preferably 12.0 or less, still

20 more preferably 1 1.0 or less and most preferably 10.0 or less.

31. A process according to any one of Claims 22 to 28, wherein [X ] is selected from conjugate bases of carboxylic acids.

25 32. A process according to Claim 31 , wherein [X ] is selected from anions having the formula [R^xC0₂]^~, wherein R^x is selected from hydrogen, a Ci to do straight chain or branched alkyl group, a C₃ to C₈ cycloalkyl group, or a C₆ to Cio aryl group, wherein said alkyl, cycloalkyl or aryl groups are unsubstituted or may be substituted by one or more groups selected from -F, -CI, -OH, -CN, -N0₂, -SH,

30 -C0₂H and =0.

33. A process according to Claim 32, wherein [X ] is selected from: formate, acetate, hydroxyacetate, propanoate, lactate, butanoate, isobutanoate, pivalate, pyruvate, thiolactate, benzoate, oxalate monoanion, tartrate monoanion, malonate monoanion, adipate monoanion, heptanedioic acid monoanion, and octanedioic acid monoanion.

5 34. A process according to Claim 33, wherein [X ] is acetate or formate.

35. A process according to any one of Claims 22 to 28, wherein [X ] is selected from phosphate dianions and phosphonate dianions.

10 36. A process according to Claim 35, wherein [X ] is selected from phosphate anions having the formula [R^xOP(0)0₂]^2" and phosphonate anions having the formula [R^xP(0)0₂]^2~, wherein R^x is selected from hydrogen, a Ci to do straight chain or branched alkyl group, a C₃ to C₈ cycloalkyl group, a C₆ to Ci₀ aryl group, a C₆ to Ci aralkyl group, or a C₆ to Ci alkaryl group, wherein said alkyl, cycloalkyl, aryl,

15 aralkyl, or alkaryl groups are unsubstituted or may be substituted by one or more groups selected from -F, -CI, -Br, -I , -OH, -CN, -N0₂, -SH, and =0.

37. A process according to Claim 36, wherein [X ] is selected from [HOP(0)0₂]^2~, [MeOP(0)0₂]^2~, [EtOP(0)0₂]^2", [n-PrOP(0)0₂]^2", and [N-BuOP(0)0₂]^2".

20

38. A process according to any one of Claims 22 to 28, wherein [X ] is selected from dicarboxylate dianions having the formula [0₂C-R^y-C0₂]^2", wherein R^y represents a Ci to C6 straight chain or branched alkylene or alkenylene chain, a Ci to C6 cycloalkylene group, or a C₆ arylene group, wherein said alkylene, alkenylene,

25 cycloalkylene or arylene groups are unsubstituted or may be substituted with one or more groups selected from -F, -CI, -Br, -I , -OH, -CN, -N0₂, -SH, -C0₂H and =0.

39. A process according to Claim 38, wherein [X ] is selected from: oxalate dianion, 30 malonate dianion, succinate dianion, glutarate dianion, adipate dianion, pimelate dianion, methylmalonate dianion, fumarate dianion, maleate dianion, methyl succinate dianion, malate dianion, tartrate dianion, citrate dianion, itaconate dianion, mesaconate dianion, o-phthalate dianion, m-phthalate dianion, p- phthalate dianion, aspartate dianion, glutamate dianion, octanedioic acid dianion and heptanedioic acid dianion.

A process according to any one of Claims 22 to 28, wherein [X ] is selected from ascorbate anion and urate anion.

A process according to any one of Claims 22 to 40, wherein the ionic liquid has a melting point of 200 °C or less, more preferably 150 °C or less, and most preferably 100 °C or less.

A process according to any one of Claims 22 to 41 , wherein the molar ratio of ionic liquid to water in the absorbent is in the range of from 80:20 to 20:80, more preferably 70:30 to 30:70, still more preferably in the range of from 60:40 to 40:60, and still more preferably in the range of from 55:45 to 45:55.

A process according to any one of Claims 22 to 42, wherein the gaseous feed stream is contacted with the liquid absorbent at a temperature of from 10 to 80 °C, more preferably from 10 to 50 °C and most preferably from 20 to 30 °C.

A process according to any one of Claims 22 to 43, wherein the gaseous feed stream is contacted with the liquid absorbent at a pressure of from 100 to 10,000 kPa, more preferably from 200 to 8,000 kPa, still more preferably from 500 to 5,000 kPa, and most preferably from 1000 to 5,000 kPa.

A process according to any one of Claims 22 to 44, wherein the gaseous feed stream is passed through the microchannel at a gas hourly space velocity (GHSV) in the range of 5,000 to 150,000, more preferably 10,000 to 120,000, still more preferably 10,000 to 100,000, and most preferably from 10,000 to 50,000.

A process according to any one of the preceding claims, wherein the microchannel has a hydraulic diameter of 5.0 mm or less, more preferably 4.0 mm or less, still more preferably 3.0 mm or less, and most preferably 2.0 mm or less. A process according to any one of the preceding claims, wherein the microchannel has a length in the range of from 10 mm to 500 mm, more preferably 10 mm to 400 mm, more preferably 10 to 200 mm, still more preferably 10 to 100 mm, and most preferably 20 to 80 mm. 48. A process according to any one of the preceding claims, wherein the gaseous feed stream comprises carbon dioxide and at least one other gas selected from N₂, 0₂, H₂, He, Ne, Ar, Xe, and hydrocarbon gases.

49. A process according to Claim 48, wherein the gaseous feed stream comprises carbon dioxide and at least one hydrocarbon gas selected from methane and ethane.

A process according to Claim 49, wherein the gaseous feed stream comprises or consists of a natural gas or a natural gas derivative.

A process according to any one of the preceding claims when operated in parallel or series using a plurality of microchannels.

A process for removing carbon dioxide from a gaseous feed stream containing carbon dioxide and at least one other gas, the process comprising passing the gaseous feed stream to a process as defined in any one of Claims 12 to 19 to obtain a first gaseous product stream having reduced carbon dioxide relative to the gaseous feed stream, and passing the first gaseous product stream to a process as defined in any one of Claims 1 to 1 1 to obtain a second gaseous product stream having reduced carbon dioxide content relative to the first gaseous product stream.

A process for removing carbon dioxide from a gaseous feed stream containing carbon dioxide and at least one other gas, the process comprising passing the gaseous feed stream to a process as defined in any one of Claims 12 to 19 to obtain a first gaseous product stream having reduced carbon dioxide relative to the gaseous feed stream, and passing the first gaseous product stream to a process as defined in any one of Claims 22 to 45 to obtain a second gaseous product stream having reduced carbon dioxide content relative to the first gaseous product stream.

A process for removing carbon dioxide from a gaseous feed stream containing carbon dioxide and at least one other gas, the process comprising passing the gaseous feed stream to a process as defined in any one of Claims 22 to 45 to obtain a first gaseous product stream having reduced carbon dioxide relative to the gaseous feed stream, and passing the first gaseous product stream to a process as defined in any one of Claims 1 to 1 1 to obtain a second gaseous product stream having reduced carbon dioxide content relative to the first gaseous product stream.