CA2745987A1

CA2745987A1 - Sorbent pellets including oxygen carriers

Info

Publication number: CA2745987A1
Application number: CA2745987A
Authority: CA
Inventors: Vasilije Manovic; Edward J. Anthony
Original assignee: Canada Minister of Natural Resources
Current assignee: Canada Minister of Natural Resources
Priority date: 2011-07-04
Filing date: 2011-07-04
Publication date: 2013-01-04
Also published as: WO2013003948A1

Abstract

A reusable sorbent, and method of use, for carbon dioxide capture and purification. The sorbent comprises pelletized composites for use in a fluidized bed process, each pellet comprising calcium oxide; an aluminate binder; and at least one transition metal component. The aluminate binder is preferably calcium aluminate or bauxite, and the transition metal is preferably selected from copper, manganese and cobalt. For carbon dioxide capture, the transition metal is an oxide, which has an exothermic reduction reaction. For removal of oxygen from carbon dioxide, the transition metal in the pellets is non-oxidized. Reduction of the oxygen carrier can be performed by providing a flow of fuel gas, such as methane. The pelletized composites provide for improved carbon dioxide capture and longer use of the sorbents with significant mitigation of decay, and without the disadvantages of oxy-firing.

Description

SORBENT PELLETS INCLUDING OXYGEN CARRIERS
FIELD OF THE INVENTION

This invention relates to sorbents for use in industrial processes, and in particular to reusable sorbents. More particularly, the invention relates to pelletized composites for use in applications including fluidized bed reactors, moving or fixed beds, the composites comprising calcium oxide as a sorbent for carbon dioxide, together with an oxygen carrier.

BACKGROUND OF THE INVENTION

It is believed that climate change and global warming are mainly caused by the increased concentration of greenhouse gases in the atmosphere. It is also well known that carbon dioxide (C02) is a greenhouse gas of which a major proportion is released from fossil fuel combustion for the production of electricity. More than 30%
of anthropogenic carbon dioxide is released from large stationary energy sources such as fossil-fuel-fired power plants. In consequence, one of the most considered scenarios for mitigation of climate change, without losing the positive advantages of using fossil fuels, is carbon capture and sequestration (CCS). This means production of energy from fossil fuels with a concentrated stream of C02, as a by-product which can be used, or more likely liquefied, transported and stored in suitable geological formations.
Although it would be possible to directly store flue gas obtained from existing power plants, this option is not economically feasible, in that flue gas generally contains only about 15% CO2 and the direct compression, liquefaction, transport, and storage of flue gas would be very costly. However, new energy conversion technologies which are capable of directly producing concentrated CO2 streams cannot be easily integrated with existing power plants. Taking into account that the replacement of conventional power plants is not a viable option in the short term, developing new technologies for CO2 scrubbing from flue gas is considered as the preferred and most feasible short/medium-term solution. This scenario is known as post-combustion CO2 capture.
Concentrated I

CO2 streams can also be produced directly during combustion if oxygen is separated from air and pure oxygen is used for combustion with some recycle of flue gas.
This technology is known as oxy-fuel combustion, but current methods of separation of pure oxygen are expensive. Alternatively, pre-combustion CO2 capture can be used, i.e. by energy conversion processes where carbon is separated before the combustion step.
These processes typically begin with a fuel conversion step such as reforming of gaseous fuels (natural gas), or gasification of solid fuels (coal) followed by a shift reaction. Conventionally, after separation of CO2 from the conversion product stream, pure hydrogen is used for combustion.

Yet another possibility is chemical looping combustion (CLC), which is a combustion technology with inherent CO2 separation. Instead of gaseous oxygen a solid material is used as oxidizing agent, i.e. as an oxygen carrier. The main property of such materials is that they allow a reversible oxidation/reduction reaction, which enables oxidation of the reduced form of the carrier after reaction with fuel, so that the same sorbent particles can be used in thousands of oxidation/reduction cycles. The oxygen carrier is oxidized by air, which means that air and fuel are never in contact and the product of combustion is a concentrated CO2 stream.

Amine scrubbing is the post-combustion CO2 capture technology closest to commercialization. However, despite the fact that CO2 scrubbing by amines is already used in other industrial processes for CO2 separation, the costs for CO2 capture using this process for large power plants, and the associated efficiency losses, are still unacceptably high. Moreover, environmental and health hazards associated with the use of amines and the requirements for pre-cleaned flue gas (both 02, and acid gas removal) provide the incentive for the development of new classes of post-combustion technologies for CO2 scrubbing which are more environmentally friendly and provide less costly CO2 separation. These cycles employ calcium oxide (CaO) based sorbents for CO2 capture. These sorbents can then be regenerated, enabling a cyclical capture process, i.e., the sorbents function as CO2 carriers. Calcium looping cycles (CaL) are presently among the most promising new technologies. Together with CLC

2 they represent new classes of looping cycles called solid looping cycles for capture.

CaO-based looping cycles (Equation I below) also offer a promising technology to separate CO2 from flue gas or syngas while producing a high-purity CO2 stream suitable for sequestration and, in the case of syngas applications, for sorption-enhanced H2 production.

CaO(,) + C02(g) = CaCO3(s) AH < 0 (1) Preliminary economic analyses suggest that such processes are economically attractive, and an important advantage of using CaO-based sorbent is that limestone (CaCO3) is inexpensive and widely available. It has been proposed that CO2 separation from flue gas is possible in a multi-cycle process in a dual fluidized bed combustion (FBC) reactor. This involves reaction of CaO with CO2 from flue gas/syngas in a carbonator and regeneration of sorbent and obtaining of a concentrated CO2 stream in a calciner. It is also known to operate a plurality of fluidized beds in a batch mode.

It should also be noted that the classification of technologies for CO2 capture is not necessarily strict; and sometimes they are interconnected. For example, CaL
can employ oxy-fuel combustion for regeneration of the sorbent; and other combinations of various aspects of the different processes have been suggested, such as integration of looping with sorption enhanced reforming for the production of hydrogen.

However, although natural limestone is a cheap and abundant CO2 carrier, the decay of its activity during CO2 capture cycles has led to the consideration of various methods of reducing or avoiding the decay. For example, CA 2,543,984 teaches the reactivation of lime-based sorbents by shocking with concentrated C02, and CA 2,543,990 teaches the use of hydration in such reactivation method. Concurrently, there have been various

3 proposals for research with the aim of obtaining sorbents with more stable CO2 capacity, i.e. by addressing the inherent composition of such sorbents, leading to the consideration of possible new materials.

It has been suggested in W02011033156A1 that CO2 can be recovered from gaseous streams by providing a second solid in addition to calcium oxide, such that an exothermic reaction generates sufficient heat which can be used to break down the CaCO3. The reference identifies the possibility of using a solid (suggested as being copper) or its oxide in a mixture with the calcium oxide in a fixed bed, but not fluidized beds, which it suggests are undesirable. The reference indicates that external heat may be necessary, and fails to recognize or address the disadvantage that as the exothermic heat is produced by the breakdown of an oxidized solid, this heat will not be available if the non-oxidized solid is used. The reference indicates that various benefits "may" be obtained, apparently based on speculation from process simulations. However, the reference fails to address the disadvantages of the suggested method, including the fact that fixed beds, unlike fluidized beds, are incompatible or problematic with many existing systems.

It has now been found that a calcium oxide sorbent can be provided in a composite form, as pellets, with an aluminate binder and morphology stabilizer, such as calcium aluminate; and that such pellets can be prepared in a manner which allows the addition of other materials, including oxygen carriers and catalysts. Such pellets containing oxygen carriers and/or catalysts are thus highly advantageous for processes such as CaL-CLC for post-combustion CO2 capture, SER-CLC for hydrogen production, and similar processes.

The pellets can be prepared by any suitable means, including but not limited to preparation by a pelletizer or by extrusion.

It has further been found advantageous to use certain transition metals to act as oxygen carriers in such composite materials, where the subsequent reduction comprises an

4 exothermic reaction which provides sufficient heat to perform the calcination step, i.e.
to drive the CO2 from the CaCO3. Suitable transition metals include, but are not limited to, copper, manganese and cobalt. Other metals provide less exothermic reaction so are less suitable for this purpose.

It has still further been found advantageous for some applications to include a catalyst in the composite. Nickel is particularly suitable for this purpose; ruthenium, rhodium and platinum are also suitable.

The composites and methods of the invention are particularly suitable for fluidized bed applications; but can also be used for various other processes, including but not limited to fixed or moving bed applications, and for sorption enhanced reforming processes.
Further, the composites and methods of the inventions can be used for capture of CO2 from flue gas streams, and for removal of oxygen from CO2 streams.

SUMMARY OF THE INVENTION

The invention therefore seeks to provide a reusable sorbent for carbon dioxide capture and purification, the sorbent comprising pelletized composites for use in a fluidized bed process, each pellet comprising:

(i) calcium oxide;

(ii) an aluminate binder; and (iii) at least one transition metal component.

Preferably, the transition metal is selected from at least one of copper, manganese and cobalt, and in some embodiments, at least some of the transition metal component comprises an oxide of the transition metal.

Preferably, the aluminate binder is selected from at least one of calcium aluminate , bauxite and kaolin, and where the aluminate binder is calcium aluminate, it is preferably provided as a cement.

Optionally, the reusable sorbent further comprises at least one catalyst, in which case preferably the catalyst comprises at least one of nickel, ruthenium, rhodium and platinum. Where the catalyst is nickel, it is generally provided as an oxide.

5 In some embodiments, the reusable sorbent has a core and shell construction, in which case preferably the core comprises calcium oxide and cupric oxide, and the shell comprises calcium oxide. In these embodiments, preferably the core comprises at least 25% parts by weight calcium oxide, and either the ratio of calcium oxide and cupric oxide in the entirety of the sorbent is substantially 1:1 parts by weight;
alternatively, the sorbent comprises about 50% cupric oxide, 40% calcium oxide and 10% of a cement binder.

Similarly, where the sorbent has a construction other than core and shell, in some embodiments it preferably comprises about 50% cupric oxide, 40% calcium oxide and 10% of a cement binder.

The invention further seeks to provide a method of carbon dioxide capture and purification from a gas flow in a fluidized bed system, the gas flow containing at least one of carbon dioxide and oxygen, the method comprising the steps of (a) providing to the fluidized bed reusable sorbent pellets wherein each pellet comprises calcium oxide; an aluminate binder; and at least one transition metal component;

(b) passing the gas flow through the fluidized bed to combine any carbon dioxide with the calcium oxide and to combine any oxygen with the at least one transition metal component to produce an oxide of the transition metal;

(c) selectively regenerating the sorbent pellets in a stream of fuel gas to produce a contained stream of purified carbon dioxide and steam;

(d) condensing the steam to produce and remove a stream of water; and (e) removing the contained stream of purified carbon dioxide from the system.
Optionally, where the gas flow may not contain sufficient oxygen to oxidize all the transition metal component, the method preferably further comprises after step (b) the step of (b. 1) selectively providing a supply of additional oxygen to oxidize any non-oxidized portion of the at least one transition metal component.

In some embodiments, preferably the fuel gas is a flammable vapour, and in some embodiments comprises methane.

6 The composites for CaL-CLC processes should embody three main components: CO2 carrier, oxygen carrier, and binder. As noted above, the addition of catalysts is preferred when it is desirable that catalytic activity be enhanced, such as in reforming (SER-CLC) and similar processes.

It is widely accepted that CaO is preferable for high-temperature CO2 capture.
There are numerous methods for its modification with the main goal of enhancing its activity during long series of carbonation-calcination cycles. It has been established that hydrated sorbents after calcination have better performance during CO2 capture cycles;
and that aluminum compounds are desirable constituents in the CaO structure, to reduce the loss of sorbent activity. They also enhance mechanical strength of the sorbent, which is desirable for its use in fluidized bed combustor systems; and it has now been found that the CaO/A1203 matrix is a suitable solid support for Cu/CuO, a low-melting-temperature material particularly suitable for use for CLC.

The CaO/A1203 matrix is thus suitable for the preparation of the composites of the invention. As a preferred source of alumina, calcium aluminate cements are particularly suitable, because they are commercially available materials of high purity, which can be regarded as having less than I% impurities, in that CaO is not considered as an impurity for CaL-CLC composites. Bauxite is also a good candidate as a source of alumina for CaL-CLC composites, in that bauxite is a mixture of hydrated alumina compounds, which decompose at higher temperature to form A1203.

Oxygen carriers employed in CLC are typically oxides of transition metals such as iron, nickel, copper or manganese. The main property of an 02 carrier for the composites of the invention is that its oxidized form can react with a fuel (typically natural gas or syngas):

(2n+m)MeXOy + Cõ H2rri (2n+m)MeXOy_1 + nCO2 + mH2O (2)

7 In this reaction oxygen from the carrier is used for oxidation of the fuel instead of fuel oxidation by air, which means that the produced CO2 is a pure stream after condensation of H20-The depleted oxygen carrier can be regenerated by air:

McXOy.l + 1/202 -~ McXOy (3) These two reactions take place in two reactors, identified here as the fuel reactor (Reaction 2), and the air reactor (Reaction 3). The air reactor always provides an exothermic environment, whereas heat in the fuel reactor can be consumed or released depending on the particular oxygen carrier and fuel used.

It is known that some of the main criteria considered for the screening of oxygen carriers in CLC are oxygen transport capacity, reactivity, the reversibility of oxidation/reduction, selectivity in reaction with fuel, mechanical strength (to be used in fluidized bed combustors), and melting point. Finally, environmental and health aspects, and abundance of a material determine its practical application in CLC
systems. The heat of the oxidation and reduction reactions is considered by means of the air and fuel reactor heat balance. Hesse's law states that the net effect of combustion depends only on the heat value of the fuel used, i.e. the choice of a particular oxygen carrier determines the heat distribution between the air and fuel reactors. Strongly endothermic reactions in the fuel reactor are not desirable because this would require the supply of heat to the reactor to maintain temperatures necessary for the reaction to occur.
However, in the case of CaL-CLC and SER-CLC integrated systems, the main criterion that must be met by an oxygen carrier is its exothermic reduction reaction, in that the proposed process uses the heat of reduction for decomposition of CaCO3 in the calciner, which is, at the same time, the fuel reactor. Therefore, for the composites of the invention, suitable oxygen carriers are those having an exothermic reduction reaction.
The heat of oxidation becomes a less important parameter (but a lower value is desirable because it means more heat during reduction), and heat released during

8 reduction becomes the main parameter. However, it should be noted that an oxygen carrier still should meet other criteria required for classical CaL and CLC
systems.
Therefore, preferably oxygen carriers for CaL-CLC should be selected from among those already considered for CLC, as presented in Table 1 below.

Table 1. Potential 02 Carriers for Composites in CaO-based CO2 sorbents 1251 02 Carrier Fuel Reduction Reaction OH` [kJ/moll Qs [kJ/91a 1000 c CuO/Cu CH4 4CuO + CH4 = 4Cu + CO2 +2H20 -211.6 0.66 CO CuO + CO = Cu + C02 -133.5 1.67 H2 CuO + H2 = Cu + H2O -101.3 1.27 CuO/Cu20 CH4 8CuO + CH4 = 4Cu2O + CO2 +2H2O -283.3 0.45 CO 2CuO + CO = Cu20 + CO2 -151.4 0.95 H2 2CuO + H2 = Cu20 + H2O -119.2 0.75 Cu20/Cu CH4 4Cu2O + CH4 = 8Cu + CO2 +21120 -139.9 0.24 CO Cu20 + CO = 2Cu + C02 -115.6 0.81 H2 Cu20 + H2 = 2Cu + H2O -83.5 0.58 NiO/Ni CO NiO + CO = Cu + CO, -47.2 0.63 H2 NiO + H2 = Cu + H2O -15.0 0.20 Co304/CoO C114 4Co3O4 + CH4 = 12CoO + CO2 + 21120 -30.3 0.03 CO Co304 + CO = 3CoO + CO2 -88.2 0.37 H2 Co304 + H2 = 3CoO + H2O -56.0 0.23 C03O4/Co CO '/4C03O4 + CO = 3/4Co + CO2 -58.1 0.97 H2 '/4C03O4 + H2 = 3/4Co + H2O -25.9 0.43 CoO/Co CO COO + CO = Co + CO2 -48.0 0.64 H2 COO+H2=Co+H2O -15.8 0.21

9 Table 1 Continued 02 Carrier Fuel Reduction Reaction AH1`000 C [kJ/mol] Qs [kJ/g Mn203/Mn3O4 CH4 12Mn2O3 + CH4 = 8Mn3O4 + CO2 + 21-1,0 -446.3 0.24 CO 3Mn2O3 + CO = 2Mn304 + CO2 -192.2 0.41 H, 3Mn2O3 + H2 = 2Mn3O4 + H2O -160.0 0.34 Mn203/MnO CH4 4Mn2O3 + CH4 = 8MnO + CO2 + 2H20 -85.2 0.13 CO Mn2O3 + CO = 2MnO + CO2 -101.9 0.65 H2 Mn203 + H2 = 2MnO + H2O -69.7 0.44 Mn304/MnO CO Mn304 + CO = 3MnO + CO2 -56.8 0.25 H2 Mn304 + H2 = 3MnO + H2O -24.6 0.11 Fe2O3/Fe304 CO Fe2O3 + CO = 2Fe3O4 + CO2 -42.0 0.26 H2 Fe203 + H2 = 2Fe3O4 + H2O -9.9 0.06 a Qs = AH1000 c/nox-Mox (4H - enthalpy of reduction, n,,, - stoichiometric coefficient for the oxidized form of oxygen carrier, and M-0,, - molar mass of the oxidized form of oxygen carrier) This table also identifies exothermic reactions and corresponding reaction enthalpies, and specific heat (Qs), defined as heat released during reduction of 1 g of the metal oxide (in oxidized form). This is a suitable parameter, which more closely indicates the potential of an 02 carrier to be used in composites with CaO. Some oxides such as PbO, CdO, and SnO are less suitable for the composites of the invention, despite the fact that they release heat during reduction, because the specific heat of reduction is negligible, and/or they have a very low melting point, and/or they have significant toxicity associated with their use.

It can be seen from Table 1 that the reactions with CH4 are typically less exothermic, and the highest specific heat of reduction is in the case of the CuO/Cu carrier (0.66 kJ/g). On the other hand, heat released during reduction of Co3O4 is negligible (0.03 kJ/g); but manganese oxides can be suitable as oxygen carriers in the composite materials for CaL-CLC. It appears that reduction by CO and H2 are more promising, given the released heat, and the highest Qs values are for CuO/Cu (1.67 and 1.27 kJ/g, respectively), suggesting the use of syngas for the reduction/calcination step for the composite materials.

Although oxygen carriers usually have some catalytic activity in the reforming reactions, it is known that to enhance the activity and selectivity more specific catalysts are generally desirable in composite materials. It is also known that various precious metals such as Ru, Rh, and Pt provide high reforming activity and improved resistance towards carbon deposition. However, despite the problems of coke formation and lower activity compared with these or other noble metals, Ni or NiO-based catalysts are still commercially used due to their relatively low cost. Taking into account that NiO is inexpensive and has shown a high catalytic activity for SER, both as a hybrid material (NiO-CaO-Ca12A114O33) and as an A1203-supported catalyst mixed with CaO, it is also a suitable catalyst in the CaO/CuO/A1203 composites of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will now be described with reference to the drawings, in which Figure 1 is a graphical representation showing the reduction and oxidation cycles of manganese oxides at temperatures of interest for CaL-CLC cycles;

Figure 2 is a graphical representation showing the temperature-programmed reduction and oxidation of manganese oxides;

Figure 3 is a graphical representation showing the relation between CaO
content (wt%
CaO + wt% oxidized 02 carrier = 100 wt%) in the composite materials and specific heat of 02 carrier reduction, which in the ideal case (quantitative oxidation/reduction, no heat losses) enables calcination of CaCO3 formed in previous carbonation step with a given conversion (curves present 10-40% conversion);

Figure 4 is a graphical representation showing the reduction/calcination/oxidation/carbonation cycles with CaO/CuO-based pellets (66%
CO + 34% H2 during reduction/calcination; 20% CO2 during carbonation);

Figure 5 is a graphical representation showing the reduction of CuO powder in TGA by CH4, and 66% CO + 34% H2 gas mixture;

Figure 6 is a graphical representation showing the effect of catalytic activity of NiO in CaO/CuO/A1203 composite detected as CaO conversion during sorption-enhanced methane reforming. Conditions: (i) reduction/calcination (CuO/CaCO3-*Cu/CaO) in 80% CH4 + 20% H20(g) at 800 C, (ii) methane reforming (Cu/CaO-*Cu/CaCO3), 80%
CH4 + 20% H20(g) at 600 C - CaO conversion presented in the diagram, and (iii) oxidation (Cu/CaCO3-CuO/CaCO3) in 80% air + 20% H2O(g) at 600 C; and Figure 7a and 7b are graphical representations of the calcinations/reduction/oxidation/carbonation cycle of CaO-CuO based pellets;
Figure 7a shows the process as separated steps, and Figure 7b shows simultaneous calcinations-reduction and oxidation-carbonation.

DETAILED DESCRIPTION OF THE DRAWINGS

The composites of the invention, and their constituent materials, were successfully tested, in the experimental work described in detail below, with reference to the drawings.

Materials The powder sample (<5 m) of Mn02 has been used as a precursor of manganese oxides to test their behaviour during oxidation/reduction cycles under conditions characteristic for Cal, cycles. The same CaO/CuO/A1203 pellets used in a previous study by the present inventors were tested in CaL-CLC cycles, but a mixture of 66%

CO and 34% H2 (simulated syngas) was used in this work as a fuel during the reduction/calcination stage. Moreover, the CaO/CuO/A1203 pellets in this study were prepared using bauxite instead of calcium aluminate cement, but with the same CaO:CuO:A1203 ratio (45:45:10). Finally, the pellets obtained using cement were impregnated with NiO in order to test catalytic activity during the sorption-enhanced steam-methane reforming reaction.

Cadomin (CD) limestone from Canada, already used for preparation of pellets, was also used in this work. A powder of CuO, 98% of particles < 5 gm, produced by Aldrich, was used as a CuO-containing material. A commercial calcium aluminate cement, CA-14, (71 % A1203 and 28% CaO, > 80% of the particles < 45 gm), produced by Almatis Inc., was used as a binder. Powdered refractory grade bauxite sample, 90%
purity, was used as another binder and source of alumina. Ni(N03)2.12H2O was used to prepare a saturated Ni(N03)2 solution for pellet impregnation.

The pellets were prepared using calcined powdered limestone (lime), the powder of CuO, and the aluminate cement or bauxite to obtain the final CaO/CuO/A1203 ratio of 45:45:10 in the pellets. The limestone was calcined at 850 C for 2 h before hydration, and weighed amounts of lime, CuO, and cement were mixed in a glass beaker. The paste obtained was extruded through a 1.0 mm sieve to obtain uniform pellet diameters, and the resulting pellets were then air dried for 24 h. Some of the air-cured pellets prepared using cement were immersed in a saturated solution of Ni(N03)2, removed from the solution after 2 h, and then air dried for 24 h.

TGA tests The reduction/calcination/oxidation/carbonation cycles were performed in a Perkin Elmer TGA-7 thermogravimetric analyzer apparatus using typically -3 mg samples suspended in a quartz tube (i.d. 20 mm) on a platinum pan (i.d. 5 mm). The gas flow rate was 0.04 dm3/min. The temperature and gas used were controlled by Pyris software. Data on sample mass during the experiments were monitored and conversions were calculated on the basis of mass change, assuming that mass change occurs only due to formation/decomposition of CaCO3 and reduction/oxidation of oxygen carrier. In the case of manganese oxides, pellets were not tested because it was easier to explore the transition between the oxides with no mass changes due to carbonation/calcination of CaO/CaCO3. The reduction/calcination steps were typically done during heating from 600 C to 800 C (and continued at 800 C) in an atmosphere of CH4 or syngas (66%
CO + 34% H2). The oxidation steps were performed during cooling from 800 C to 600 C, and carbonation continued for 12 min at 600 C. The heating/cooling rate was 50 C/min. These conditions were typical, but some variation was incorporated in particular runs in order to separate the calcination and reduction steps, as well as the oxidation and carbonation steps. Finally, some specific temperature-programmed reduction and oxidation tests were performed with both manganese oxides and CuO.
Results and Discussion Manganese oxides may be considered as potential oxygen carriers for CaL-CLC
cycles because some of their reduction reactions are exothermic and potentially can provide heat for calcination of CaCO3. Due to several possible transformation changes among different oxidation stages the tests were started with pure Mn02 (not CaO/MnO2 composites) because some mass changes may overlap with those due to carbonation/calcination. The reduction/oxidation cycles are presented in Figure 1. The results on mass changes are calculated and presented as the change of the O/Mn molar ratio, and horizontal lines correspond to the particular manganese oxides. As can be seen, MnO is always obtained after reduction by CH4, which is consistent with the literature. The first mass increase started during the cooling step in N2, which most likely can be attributed to the oxidation by 02 contained as an impurity in the N2 used.
At 600 C N2 was switched with air and MnO was quickly oxidized to Mn304.
During the next reduction step, CH4 was switched directly with air at 600 C to avoid oxidation by the 02 impurity in N2. As can be seen the mass increase again corresponds to oxidation to Mn304. Some more reduction/oxidation steps were carried out, changing conditions (mainly temperature) slightly during the oxidation step, but the MnO-Mn304-MnO transformation was always seen. This means that under typical CaL-CLC
conditions (reduction/calcination at 800-900 C; carbonation/oxidation at -600 C) only Mn304/MnO cycles can be reached. Unfortunately, CH4 cannot be used in this case for the reduction step because the reaction is endothermic (Table 1 presents only exothermic reactions).

This prompted the exploring of conditions that might enable the exothermic reduction reactions with CH4 (Mn203---*Mn3O4, and Mn203-MnO, Table 1). In that order, the temperature-programmed oxidation and reduction tests were performed and the results are presented in Figure 2. Although reduction of Mn02 has two slightly separated peaks, MnO is the only species present at 650 C. During further temperature increase above 800 C a slight mass increase is noticed, which should not be attributed to formation of a manganese species. Instead, it is supposed that this represents formation of carbon due to decomposition of CH4:

CH4 = C + 2H2 AH1000 C=90.29 kJ/mol (4) This reaction is endothermic and, therefore, favoured at higher temperatures, and leads to carbon deposition, which is undesirable because it reduces the efficiency of CLC
processes. Carbon deposition was also visually noted in the experiment presented in Figure 1, at the edges of the platinum pan. However, when CH4 was switched to N2 the deposited carbon appeared to be oxidized by traces of 02 present in the N2.
This impurity of N2 most likely also caused oxidation of MnO, which started at -750 C and continued to 350 C. Therefore, before the programmed oxidation step started, some MnO had already been oxidized. With the increase in temperature, oxidation by air became faster, and the maximum sample mass was reached at -650 C and further temperature increase did not cause significant mass changes. As can be seen after this oxidation step, Mn203 is obtained, which is, under these conditions (02 partial pressure), the most stable phase at temperatures below 550 C. However, once air was switched with N2, Mn203 quickly decomposed to Mn304. The further reduction by showed that the reduction of Mn304 started at 550 C and finished at 650 C.
During the next cycle the transformations of manganese oxides were confirmed.

The main result of this test is that both oxidation and reduction steps should be at temperatures <550 C in order to employ the most exothermic reduction by methane, Mn203 --f Mn304. Consequently, this means that this reaction cannot be used for regeneration of CaCO3 because it requires significantly higher temperatures (>
800 C, even when CO2 in the calciner is diluted, for example by steam).

In the case of the Mn2O3-+MnO reduction with CH4, the specific heat is only 0.13 kJ/g Mn203. This is a very low Qs value, which necessitates a higher content of 02 carrier in the composite material. The relation between the 02 carrier/CaO ratio in the composite material and Qs is presented in Figure 3. It can be seen that in the case of the Mn2O3--->MnO reduction by CH4 this ratio in the ideal case (quantitative reactions with no heat loss) should be 70:30 to enable calcination of 10% carbonated sorbent.
To reach full calcination of 30% carbonated material it should contain at least 88%
Mn203, but under real conditions with heat losses its content should be even greater than 90%.
Moreover, it should be noted that CO2 carrying capacity of those materials would be very small because of a low CaO content, and even in the case of 30%
conversion it would be only few weight percent (g CO2 per 100 g composite material).
Consequently, it most likely does not allow their competitive use (when compared with CaO/CuO
composites) in CaL-CLC cycles for post-combustion CO2 scrubbing from fossil-fuel-firing power plants. Another issue which would reduce the efficiency of Mn203/MnO
cycles is that before the oxidation step manganese-based material should be cooled below 550 C. Thus, due to both an inappropriate temperature range of reactions and small specific heat released the use of manganese oxides was not explored in composite materials for CaL-CLC cycles in more detail.

The analysis in this study indicated that CuO-based composite materials were preferred for CaL-CLC and SER-CLC processes. As shown in Figure 3, the pellets with the 50:50 CaO/CuO ratio can provide heat for the calcination of -20% carbonated sorbent.
It should be noted here that sorbents with higher carbonation conversion can be only partially calcined. However, the partial calcination may provide more stable carrying capacity during longer series of captures cycles, and vice versa: the operation regimes can be maintained under conditions of partial carbonation, which has been proven as an option for higher integral CO2 capture capacity during a longer series of CO2 capture cycles.

However, the interest in composites with higher CO2 carrying activity still remains, leading to the exploring of more exothermic reduction reactions employing syngas, which is also an attractive fuel obtained during steam gasification of solid fuels such as coal. As can be seen from Table 1, the CuO/Cu oxygen carrier is again the most attractive. The specific heat of CuO->Cu reduction by syngas is in the range 1.27-1.67 kJ/g, dependent on syngas composition.

The CaL-CLC cycles with CaO/CuO/A1203 pellets performed using a mixture of 66%
CO and 34% H2 are presented in Figure 4. These cycles are similar to those performed with CH4. The main difference is a two-step reduction/calcinations; namely, during reduction by CH4, the reduction/calcination step occurred simultaneously, with a monotonic mass decrease. It appears here that reduction of CuO by syngas is a much faster reaction, which takes place at lower temperatures. This can also be seen during the first calcination of fresh pellets, where the first intensive mass decrease took place by -150 C. To confirm that the initial mass loss during the calcination stage is due to reduction of CuO, calcination during the last (3rd) cycle was done in a N2 atmosphere.
It can be seen that in this case the reaction (calcination) started at - 700 C
and it was the first mass loss in this cycle. Calcination finished at 800 C and further slower mass reduction is attributed to decomposition of CuO. After the introduction of syngas, the mass quickly dropped to the value that corresponds to the production of elemental Cu.
Another significant difference between the use of syngas and CH4 is the formation of carbon deposits. Although this formation was noted during the cycles with CH4, in the case of syngas it was more pronounced. The reaction of carbon formation (Boudouard reaction) is exothermic:

2CO = C + CO2 AH,000,c=-167.74 kJ/mol (5) This means that the direct reaction is favoured at lower temperatures, and the first mass increase which appeared just after the CuO reduction (at temperatures < 500 C) is mainly due to carbon deposition. This can be confirmed by the fact that after the first calcination, the sample mass was higher than it was after the 2nd and 3rd calcination.
This can be explained by the fact that during heating the sample from room temperature to 800 C, the sample was exposed to CO for a longer time at lower temperatures. It may be concluded that syngas has a greater tendency to form carbon deposition under experimental conditions employed here than does CH4. However, it should be noted that carbon deposition might be less pronounced under more realistic conditions when steam is present.

To explore in more detail the noticed difference in reduction of CuO by syngas and CH4, temperature-programmed reduction of CuO powder was performed with both gases, and the results are presented in Figure 5. The results confirmed that reduction by syngas started even below 150 C and finished at 250 C. This is in close agreement with the results on CuO reduction by H2. The rate of CuO reduction by CO is not considerable at temperatures below 300 C. On the other hand, reduction by CH4 also starts at lower temperatures but it is still not finished at 600 C (Figure 5).
These experiments show that CO and H2 are even more reactive with CuO than CH4, which together with the higher heat of reduction, recommends syngas to be used as a fuel in CaL-CLC processes.

These preliminary tests showed that CaO/CuO/A1203-based pellets prepared using refractory-grade bauxite have similar CO2 capture activity as pellets prepared using calcium aluminate cements. It should be noted that refractory bauxite is a significantly cheaper and widely commercially available material. The main compounds in these bauxites are hydrated alumina minerals, with relatively small amounts of Si02 (1.5-5.5%). Therefore, they are predisposed to form mayenite (Ca12A114O33) during calcination with Ca(OH)2, which is used for preparation of CaO-based pellets.
Mayenite is desired for both pellet activity in longer series of CO2 capture cycles and mechanical strength.

Finally, CaO/CuO/A1203 pellets were impregnated with a solution of Ni(N03)2 in order to enhance their catalytic activity in the sorption-enhanced methane reforming reaction:
CH4 + CaO + 2H20 = CaCO3 + 4H2 (6) Catalytic activity can be detected by mass changes, i.e. CaO conversion, because enhanced catalytic activity of pellets implies higher CO2 concentration in/around particles and consequently higher conversion rate. It should be noted that catalytic activities measured in this manner should only be used in relative comparisons with and without nickel in the pellets because it can be expected that metal surfaces in the TGA
must also catalyze the methane reforming reaction. However, the strongest effect on carbonation should be CO2 formed in pellet particles because it determines the rate of carbonation. The reforming steps were performed using CH4 with 20% steam at three temperatures: 500, 600, and 700 C. This is followed with oxidation in air at 600 C, and reduction/calcination using CH4 during heating to 800 C (20% steam present at all times).

The CaO conversions obtained during the sorption-enhanced reforming steps at are presented in Figure 6. The results for the other temperatures are not presented because there were no noticeable mass changes at 700 C, and only slight mass changes were monitored at 500 C. The explanation for this is that, due to mass transfer between the pellet particles and the surrounding gas stream, CO2 formed was diluted and the effective CO2 concentration was below that required for carbonation (determined by the chemical equilibrium at 700 C). At the lower temperature (500 C) carbonation took place, but due to both low temperature and diluted C02, the carbonation rate was slow.
However, the main result presented in Figure 6 is that catalytic activity of pellets is enhanced by the presence of Ni. The final conversions are doubled with only 3.0% NiO
(which was determined by EDX analysis of the pellet surface). Conversions in both cases decrease with the cycle number, but this can be attributed to loss of CaO activity due to sintering rather than to the loss of catalytic activity. It should also be noted that there is some evidence of carbon deposition, which is characteristic for NiO-based catalysts, but it was significantly less pronounced than that seen in the experiments when steam was not used and/or with syngas (Figures 2, 4, and 5). These results confirm that the CaO/CuO/A1203/NiO composites are good candidates for processes that integrate sorption-enhanced reforming and chemical looping combustion.

Referring now to Figure 7, composites of the pellets of the invention were successfully made and tested in a process of CO2 capture with subsequent regeneration of the sorbent, employing the heat obtained by reduction of the CuO oxygen carrier.
It was found that both the reduction of the CuO and the oxidation of Cu are fast and quantitative reactions, thus providing a significant advantage over prior methods. It was also found that CaO reached carbonation conversion of about 65% in the first cycle, with the potential for a further increase.

Claims

1. A reusable sorbent for carbon dioxide capture and purification, the sorbent comprising pelletized composites for use in a fluidized bed process, each pellet comprising:

(i) calcium oxide;

(ii) an aluminate binder; and (iii) at least one transition metal component.

2. A reusable sorbent according to Claim 1, wherein the transition metal is selected from at least one of copper, manganese and cobalt.

3. A reusable sorbent according to Claim 1 or Claim 2, wherein at least some of the transition metal component comprises an oxide of the transition metal.

4. A reusable sorbent according to any one of Claims 1 to 3, wherein the aluminate binder is selected from at least one of calcium aluminate, bauxite and kaolin.

5. A reusable sorbent according to Claim 4, wherein the aluminate binder is calcium aluminate and is a cement.

6. A reusable sorbent according to any one of Claims 1 to 5, further comprising at least one catalyst.

7. A reusable sorbent according to Claim 6, wherein the catalyst comprises at least one of nickel, ruthenium, rhodium and platinum.

8. A reusable sorbent according to Claim 7, wherein the catalyst is nickel and is provided as an oxide.

9. A reusable sorbent according to any one of Claims 1 to 8, wherein the reusable sorbent has a core and shell construction.

10. A reusable sorbent according to Claim 9, wherein the core comprises calcium oxide and cupric oxide, and the shell comprises calcium oxide.

11. A reusable sorbent according to Claim 10, wherein the core comprises at least 25% parts by weight calcium oxide.

12. A reusable sorbent according to Claim 11, wherein a ratio of calcium oxide and cupric oxide in the sorbent is substantially 1:1 parts by weight.

13. A reusable sorbent according to Claim 10, comprising about 50% cupric oxide, 40% calcium oxide and 10% of a cement binder.

14. A reusable sorbent according to any one of Claims 1 to 8, comprising about 50%
cupric oxide, 40% calcium oxide and 10% of a cement binder.

15. A method of carbon dioxide capture and purification from a gas flow in a fluidized bed system, the gas flow containing at least one of carbon dioxide and oxygen, the method comprising the steps of (a) providing to the fluidized bed reusable sorbent pellets wherein each pellet comprises calcium oxide; an aluminate binder; and at least one transition metal component;

(b) passing the gas flow through the fluidized bed to combine any carbon dioxide with the calcium oxide and to combine any oxygen with the at least one transition metal component to produce an oxide of the transition metal;

(c) selectively regenerating the sorbent pellets in a stream of fuel gas to produce a contained stream of purified carbon dioxide and steam;

(d) condensing the steam to produce and remove a stream of water; and (e) removing the contained stream of purified carbon dioxide from the system.

16. A method according to Claim 15, further comprising after step (b) the step of (b. 1) selectively providing a supply of additional oxygen to oxidize any non-oxidized portion of the at least one transition metal component.

17. A method according to Claim 15, wherein step (a) comprises providing reusable sorbent pellets according to any one of Claims 1 to 14.

18. A method according to Claim 15, wherein the transition metal component comprises a non-oxidized transition metal, and step (b) comprises passing a flow of carbon dioxide through the fluidized bed to remove any oxygen from the carbon dioxide flow.

19. A method according to any one of Claims 15 to 18, wherein the fuel gas is a flammable vapour.

20. A method according to Claim 19, wherein the fuel gas comprises methane.