looping combustion

Evaluation of novel ceria-supported materials as oxygen carriers for chemicallooping combustion Master’s Thesis in the Master Degree Program, Innovative and Sustainable Energy Engineering

ALI HEDAYATI

Department of Energy and Environment Division of Energy Technology CHALMERS UNIVERSITY OF TECHNOLOGY Götebory, Sweden 2011 Master’s Thesis T2011-354

REPORT NO. T2011/354

Evaluation of novel ceria-supported materials as oxygen carriers for chemical-looping combustion

ALI HEDAYATI

Department of Energy and Environment Division of Energy Technology CHALMERS UNIVERSITY OF TECHNOLOGY Göteborg, Sweden 2011

Evaluation of novel ceria-supported materials as oxygen carriers for chemical-looping combustion Ali Hedayati Supervisors: Dr. Henrik Leion, Prof. Abdul-Majeed Azad Examiner: Dr. Tobias Mattisson Department of Energy and Environment Division of Energy Technology Chalmers University of Technology

Abstract: According to the IPCC, increasing concentration of greenhouse gases in the atmosphere is the main reason of climate change and global warming. Resulting mainly from burning of fossil fuels, CO 2 has the most apparent global warming potential. Increasing rate of energy consumption by the society and high dependency of energy production on fossil fuels along with the obvious negative environmental consequences of climate change, oblige human to concern about the atmospheric CO2 concentration. Thus, quick and efficient techniques are necessary to be applied for CO 2 sequestration to prevent it released to the atmosphere. Chemical-looping combustion (CLC) is one new technology to capture CO2. CLC consists of two interconnected fluidized bed reactors i.e. air reactor and fuel reactor. In the fuel reactor, fuel reacts with an oxygen carrier in absence of nitrogen and is converted to CO2 and H2O. Then the reduced oxygen carrier is transferred to the air reactor to be re-oxidized back to its original oxidized state. So, the oxygen is carried from the air reactor to the fuel rector. After condensation of water vapor in the outflow gas coming from the fuel reactor, a highly pure stream of CO 2 is obtained. This thesis investigates the reactivity and performance of some oxygen carrier particles supported on ceria and gadolinia doped-ceria (GDC) for chemical-looping combustion. The oxygen carriers were oxides of copper, manganese and iron supported on ceria and GDC. Oxygen carriers were synthesized via extrusion technique and tested for successive oxidation and reduction cycles using methane and syngas (50% CO and 50% H2) as fuel. Tests were performed using fluidized bed batch reactors made of quartz. Reduction cycles were performed at 950°C for iron and manganese containing oxygen carriers and at 900 and 925°C for copper oxide. The reactivity during reduction and oxidation cycles, fluidization and agglomeration properties, oxygen release characteristics (CLOU effect) and phase analysis were evaluated for all the tested particles. The results showed that in general GDC supported particles were more reactive compared to ceria supported ones during reduction cycles. Methane was totally converted by copper oxide supported on GDC and the particles showed very high oxygen release potential as well, qualifying it as a CLOU material. Syngas was fully converted to CO2 and H2O by all the oxygen carriers synthesized and tested in this work. Very good fluidization properties and low attrition and agglomeration were observed for all the particles. Manganese oxide containing particles showed very low conversion during methane cycles that was somehow expected according to previous reports on this material. Keywords: carbon dioxide (CO2), chemical-looping combustion (CLC), oxygen carrier, ceria, gadolinia doped ceria (GDC), copper, iron, manganese, oxygen release

Table of contents: 1

2

3

4 5 6

Introduction 1.1 Greenhouse gases 1.2 Global warming 1.3 Concerns of high atmospheric CO2 concentration 1.4 Anthropogenic CO2 emission reduction 1.4.1 Pre-combustion carbon capture 1.4.2 Oxyfuel combustion carbon capture 1.4.3 Post-combustion carbon capture (PPC) 1.5 Chemical-Looping Combustion (CLC) 1.5.1 CLC with solid fuels 1.6 Oxygen carriers 1.7 Objective Experimental 2.1 Synthesis and fabrication of oxygen carriers 2.2 Experimental set up 2.3 Data analysis Results and discussion 3.1 Phase analysis of the oxygen carriers 3.2 Reactivity test results 3.2.1 Pure ceria 3.2.2 Fuel conversion 3.2.2.1 Methane conversion 3.2.2.2 Syngas conversion 3.2.2.3 Phase analysis 3.2.2.4 Fluidization and agglomeration characteristics 3.2.3 Oxygen release 3.2.4 Oxidation phase 3.2.5 Temperature variation during oxidation and reduction cycles Conclusion Acknowledgements References

1 1 1 2 2 2 3 3 3 4 5 7 8 8 10 12 13 13 13 13 14 15 19 19 20 21 22 24 26 27 28

0

1.

Introduction

Energy conversion is a key factor for the development of human society1. Living without energy supply – mainly electricity- is not possible in the modern world. Thus, as development continues with an aim to eradicate poverty and enhance quality of life, increase in energy production by all sectors is unavoidable2. A rapid shift toward efficient and cost effective sources of energy is obviously required. Apart from the energy supply issues, environmental concerns have also emerged. In a few decades, the energy demand may be double in comparison to today which presents challenge to the preservation of environment due to the concomitant increase in emissions caused by the power generation processes3. Consequently, clean technologies for energy production have attracted greater attention in order to reduce the emission of pollutants to the atmosphere, soil and water.

1.1 Greenhouse gases When sunlight passes through the atmosphere and reaches the earth surface, a part of this light is radiated back in longer wavelengths to the atmosphere. Certain atmospheric gases called greenhouse gases absorb these wavelengths and radiate them back to the surface of the earth resulting in an increase in ground temperature. This naturally occurring phenomenon is called the greenhouse effect4 caused by greenhouse gases where the most important ones are H2O, CO2, CH4, N2O and halocarbons. Water vapor is accounted for 60% of total greenhouse gas effect but human activities do not play any direct role in the balance of this substance in the atmosphere5. Moreover, the lifetime of water vapors in the atmosphere is about 9 days compared to hundreds of years for carbon dioxide. According to IPCC, human activities have today resulted in a significant increase in atmospheric concentrations of carbon dioxide, methane and nitrous oxide compared to 1750 6 . This increase has caused a change in the energy balance toward warming up the atmosphere. CO2 is the most important anthropogenic greenhouse gas which contributes most towards the rise of atmospheric temperature; its main source is fossil fuel burning7. Accordingly, CO2 is at the center of attention with regard to climate change and global warming concerns.

1.2

Global warming

Svante Arrhenius was one of the first who mentioned global warming probability due to the increasing concentration of CO2 in the atmosphere7. He suggested that the mean temperature of the earth will probably increase due to emission of carbon dioxide originating from human activities. Now it is a fact that the atmospheric temperature has risen over the last decades and it is believed to be a result of increased concentration of greenhouse gases in the atmosphere contributing to an intensified greenhouse effect8. The mean temperature increase is estimated to between 0.4 and 0.8oC during the last century and this has resulted in having 10 of the warmest years among the past 15 past years9. Greenhouse gas emissions have risen by 70% between 1970 and 2004 of which the larger part has come from energy production10. CO2 has the strongest effect on global warming and 77% of total greenhouse gas emission in 2004 has been CO2.

1

1.3

Concerns of high atmospheric CO2 concentration

There are several natural reservoirs of carbon which interact with one another according to the carbon cycle. Because natural processes which can sequester carbon are rather slow, CO2 from human activities will accumulate and last for a long time in the atmosphere 11 . Atmospheric concentrations of CO2 have increased from natural mean value of 280 ppm to 379 ppm in 20056 and it is estimated to exceed 400 ppm by 203012 mainly due to the burning of fossil fuels. About 80% of world’s primary energy in 2004 came from fossil fuels which released 26.4 Gt CO2 to the atmosphere. It is expected that the worldwide energy consumption will increasing rapidly so that the yearly CO2 emissions from energy production will reach 33.8 Gt in 2020 and 42.4 Gt in 2035 13. These statistics show that carbon dioxide emission will have a severe effect in coming decades. Thus it is necessary to apply methods to prevent or at least decrease the emissions to the atmosphere otherwise harsh consequences of global warming would be inevitable.

1.4

Anthropogenic CO2 emission reduction

Quick and effective actions are needed to reduce the atmospheric concentration of CO2. Actions are necessary both in power production plants which are mainly based on fossil fuels and in commercial operations. There are some protocols and treaties like Kyoto protocol to decrease the emission of CO2 but these are not enough and the main question remains unanswered of how to reduce the emission14. The first solution would be the substitution of fossil fuel-based energy generation units by other technologies such as nuclear power plants, biomass, solar arrays and wind farms. But these technologies are not without challenges and limitations, such as the high cost per unit energy produced by these devices and nonfeasibility of rapid transition toward these technologies due to the required mammoth infrastructure changes15. So the tendency is toward faster and more reliable existing methods to mitigate the emission of CO2. One such method is carbon capture and storage. There are three main techniques to capture CO2 i.e. pre-combustion carbon capture, oxyfuel combustion and post-combustion carbon capture. Captured CO2 is then transferred to the storage sites where it is stored in deep geological formations or under sea beds.

1.4.1

Pre-combustion carbon capture (Pre-C3)

Pre-combustion carbon capture refers to chemical processes where a fuel – mainly solid fuel is converted to hydrogen and carbon dioxide and the latter is separated by physical (pressure swing adsorption, PSA) or chemical (amine adsorption) methods. Hence, the carbon in the fuel is completely removed. The process consists of two steps. First, the fuel reacts with oxygen/air/steam to produce carbon monoxide and hydrogen. Carbon monoxide undergoes a catalyzed water-gas-shift (WGS) reaction wherein carbon monoxide reacts with steam, generating more hydrogen plus carbon dioxide 16. The main advantage of this method is to produce a clean carbonless fuel which can be used in variety of industrial applications, but the drawback is the high operating cost of the process.

2

1.4.2

Oxyfuel combustion carbon capture (Oxy-C3)

In oxyfuel combustion, oxygen is used instead of air, resulting in a nitrogen-free combustion with high concentrations of water vapor and CO2 in the flue gas stream. The concentration of carbon dioxide in flue gases is nearly 80% which simplifies the separation processes 16. The flame temperature in the case of pure oxygen would be very high so a portion of the CO2-rich flue gases are recirculated both to control the temperature and to reach the required gas flow. Advantages of oxyfuel combustion are: (1) easy CO2 separation, (2) smaller flue gas volume, (3) better desulfurization and, (4) prevention of NOx formation. Disadvantages of the oxyfuel combustion are the high cost and electrical demands for oxygen production.

1.4.3

Post-combustion carbon capture (Post-C3)

Post combustion carbon capture is the most used technology for CO2 capture. The flue gases coming from the combustion of fossil fuels are treated and CO2 is separated mainly by a chemical sorbent 4. One of the disadvantages of post combustion carbon is large amount of flue gases resulting in low concentration of CO2 followed by high temperature of flue gases making it necessary to use powerful solvents16. There are different technologies for absorption of carbon dioxide like chemical and amine absorption, cryogenic purification, membrane separation and also algal bio-fixation11.

1.5

Chemical-Looping Combustion (CLC)

Chemical-looping combustion (CLC) is one of the emerging technologies which makes it possible to have a nitrogen free fuel conversion without the need for costly and energy consuming processes for oxygen production and purification of the exhaust. In CLC oxygen is provided by oxygen carrier particles so that direct contact between fuel and combustion air is avoided4, 11, 17. The main benefit of CLC is that a high concentration of CO2 mixed with water vapor is obtained. Water vapor is condensed and a highly pure stream of CO2 (nearly 100%) is ready for sequestration. Thus, there is no need for CO2 separation units. Besides, there is no NOx emission and the heat released from the combined oxidation and reduction process is equal to that in conventional combustion. One of the main challenges regarding CLC technology is the economical availability of oxygen carriers with required properties. Oxygen carrier particles, which are mainly solid oxides18, are discussed in details later in this thesis. Chemical looping combustion consists of two interconnected reactors, namely, the air reactor and the fuel reactor. In the air reactor, oxygen carries particles are exposed to an air flow and are oxidized according to reaction (1): O2 (g) + 2 MexOy-1 ↔ 2 MexOy

(1)

The fully oxidized particles are transported to the fuel rector. If the fuel is gaseous, it reacts with oxygen carrier particles, reducing them according to the reaction (2): (2n + m)MexOy + CnH2m ↔ (2n + m) MexOy-1 + mH2O + nCO2

(2)

The reduced particles are transferred back again to the air reactor and the cycle is repeated again. Thus, the fuel reacts with the oxygen in the carrier while no nitrogen exists in the fuel reactor. 3

Reaction 1 is highly exothermic while reaction 2 can be exothermic or endothermic depending on oxygen carrier characteristics and type of fuel. Flue gases are mainly composed of high concentration of carbon dioxide and some water vapor which is condensed and separated from gaseous CO2 4,17. In Figure 1 a scheme of CLC unit introduced by Lyngfelt et al.19 is shown.

Figure 1 – layout of a chemical-looping combustion process: 1) air reactor and riser, 2) cyclone and 3) fuel reactor 19.

There are two fluidized bed reactors connected with each other through loop-seals. Fluidization results in very effective and close mixing of particles and air or fuel gas, respectively; this design is very similar to fluidized bed systems designed for solid fuels. Oxygen carrier particles are separated from the air stream in the cyclone and drop down to the fuel reactor by gravity. There are also particle locks in place to prevent mixing of air from the air reactor and gases from the fuel reactor4. 1.5.1

CLC with solid fuels

It is beneficial to utilize solid fuels in CLC systems since they are cheaper and more abundant compared to the gaseous fuels like natural gas. It is possible to gasify the solid fuel mainly via gasification process in the presence of steam or CO2 followed by converting the gasified products namely, CO, H2 and CH4. To avoid the slow gasification reactions, the oxygen carrier must be capable of releasing molecular oxygen so that the solid fuel reacts directly with gas phase oxygen as in the case of normal combustion. This strategy is called chemicallooping combustion with oxygen uncoupling (CLOU) and the oxygen release behavior of the oxygen carrier is known as the CLOU effect 17. The CLOU reaction during which oxygen is released in the gas phase can be represented as follow: MexOy ↔ MexOy-2 + O2 (g)

(3)

The reduced oxygen carrier is re-oxidized back to its original state in the air reactor via the following reaction: MexOy-2 + O2 ↔ MexOy

(4)

4

As an example, copper oxide is well known for its CLOU effect17. This effect will be shown and discussed later in this work. Copper oxide releases gas phase oxygen through the following decomposition reaction: 4 CuO → 2 Cu2O + O2

(5)

The molecular oxygen released reacts with the solid fuel. However, the focus of this thesis would be on gaseous fuels i.e. methane and syngas.

1.6

Oxygen carriers

Oxygen carrier particles are a central part of the CLC system and play the main role of transporting oxygen from the air to the fuel reactor. Therefore, the properties of these particles are important for investigation and improvement of the chemical-looping combustion technology 4, 17. According to Jerndal et al. relevant properties of particles are: sufficient rate of oxidation and reduction, ability to perform high conversion of fuel to CO2 and H2O, resistance against attrition and fragmentation, and, being cheap and environmentally sound 20 . In addition, there are other criteria that are of relevance in the selection of suitable carrier, such as melting temperature of the oxygen carrier, oxygen ratio which is the maximum transported mass of oxygen for a given mass flow of particles 20. The most commonly used active materials in oxygen carriers are the oxides of nickel, copper, manganese or iron. According to Leion, a general comparison of the reactivity of metal oxides with methane shows their propensity in the following descending order: NiO>CuO>Mn2O3>Fe2O317. Copper oxide, iron oxide and manganese oxide react with methane as follows21: CuO (at 800°C): CH4 +4 CuO  CO2 +2 H2O + 4 Cu (fuel oxidation)

(6)

4 Cu + 2 O2  4 CuO (carrier regeneration)

(7)

Fe2O3 (at 800°C): CH4 + 12 Fe2O3  CO2 + 2 H2O + 8 Fe3O4 (fuel oxidation)

(8)

8 Fe3O4 + 2O2  12 Fe2O3 (carrier regeneration)

(9)

CH4 + 4Fe2O3  CO2 + 2 H2O + 8 FeO (fuel oxidation)

(10)

8 FeO + 2 O2  4 Fe2O3 (carrier regeneration)

(11)

Mn3O4 (at 950°C): CH4 + 4 Mn3O4  CO2 + 2 H2O + 12 MnO (fuel oxidation)

(12)

12 MnO + 2 O2  4 Mn3O4 (carrier regeneration)

(13)

Reduction and oxidation (regeneration) reactions are continuously done in the fuel reactor and air reactor, respectively.

5

Copper oxide is a very promising oxygen carrier with advantages 21, such as, being very reactive during oxidation and reduction cycles, full conversion of gaseous hydrocarbon fuels like methane, exothermic nature of the oxidation and reduction reactions and, reasonable price of the material. The challenges of copper oxide are its decomposition (it can be advantageous in terms of oxygen release via CLOU effect) at rather low temperature as well as the low melting point of elemental copper. Even then, due to its obvious advantages, copper oxide has been investigated a great deal as a potential oxygen carrier. For example, Gayan et al. investigated the effect of different supports on the behavior of copper oxidebased oxygen carriers22. Iron oxide has also been extensively investigated as an oxygen carrier. Its natural abundance together with favorable thermodynamic properties makes it quite attractive for CLC applications. Activity of iron oxide supported by alumina, silica and titanium dioxide has been tested by various researchers and it has been shown that transformation of hematite to magnetite is the main chemical reaction during the process21. Abad et al. tested iron oxide supported on alumina at different temperatures and reported 10-94% conversion of methane23. Johansson et al. have investigated the reactivity of manganese oxides produced by different methods and reported poor reactivity and evidence of agglomerations24. Literature shows that manganese oxides react with the supports made up of Al2O3, SiO2 and TiO2 resulting in lower activities. Johansson et al. investigated manganese oxides on ZrO2 supports stabilized with CaO, MgO or CeO2 and reported good activities during the reduction cycles with methane 24. Several other oxygen carrier systems have also been studied in the literature21, 25. Oxygen carriers are generally supported on inert materials. The supports are usually porous materials used for maintaining the mechanical structure of particle during the process and porosity increases the surface area of particle and the reactivity as well4. A number of supports have been used. However, Al2O3, SiO2, TiO2, ZrO2, NiAl2O4, and MgAl2O4 are among the most tested and reported supports26. A comprehensive literature survey of various oxygen carriers was done by Lyngfelt et al. 27. Abad et al. developed a model to investigate the behaviour of oxygen carriers in the fuel reactor and they used CuO-based oxygen carries as a validation model28. As mentioned above, the major efforts have concentrated on investigating various oxygen carriers supported on inert materials. So, it would be innovative to utilize supports that are participating and thus can act as a minor but additional oxygen carrier or as a facilitating oxidizing catalyst during CLC operation. Thereby exploiting the synergy of the composites made up of the support and the oxygen carrier. One of these materials is cerium dioxide (CeO2) also known as ceria, which is in use extensively in three-way catalysts (TWC) for oxidizing carbon monoxide and unburnt hydrocarbons and reducing nitrogen oxides in the exhaust stream of automobiles before they are released to the environment. Ceria as an oxygen carrier supported on alumina has been investigated by Wei et al. for partial oxidation of methane 29 . They investigated different compositions of ceria and alumina at different temperatures and showed that 10% ceria on alumina had the highest methane conversion up to 80% at 925°C. Xing et al. prepared ceria supported on Fe2O3 to reform methane to hydrogen and syngas in the presence of steam at 850°C30. They found out that CeO2-Fe2O3 is a suitable oxygen carrier for methane conversion and pure hydrogen production, where they reported that CeFeO3 was formed under harsh reductive environment of the test that could help the durability and performance of the oxygen carrier during successive cycles. 6

These data show that investigation of ceria for reforming applications is not a new concept. However, utilization of ceria as a support for common oxygen carriers for direct application in CLC systems for combustion of carbonaceous fuels is a new and promising idea in the light of the known behavior of ceria. The favorable properties of ceria supported oxygen carriers would be higher activity and lower cost of production due to larger fraction of the active materials in terms of mass and volume.

1.7

Objective

The objective of this study is to fabricate oxygen carriers containing oxides of copper, iron and manganese supported either on pure ceria or gadolinia-doped ceria (GDC, Gd0.1Ce0.9O1.9) and investigate their performance and activity in chemical-looping combustion processes using the fuel of syngas and methane in a fluidized bed batch reactor.

7

2.

Experimental

2.1

Synthesis and fabrication of oxygen carriers

CuO, Mn2O3 and Fe2O3 were chosen as active phases due to their known properties as CLC materials and also for the sake of comparison of the synthesis method adopted in this work with spray drying and freeze granulation techniques. The selected metal oxides were mixed with ceria or GDC for the production of oxygen carriers. In each case, a given metal oxide was mixed with ceria or GDC in the weight ratio of 60:40 to make 170g batch. The dry powders were transferred to a pear-shaped distillation flask; 400g of water was added and the mixture was homogenized using a Buchi R-110 rotary evaporator equipped with a Buchi vacuum pump, pressure controller and a chiller. The water bath was maintained at 60°C. After 2/3rd of the water was removed by distillation, the thoroughly homogenized mixture (now reduced to a thick slurry) was dried in an air oven at 150oC. The rotary evaporator setup used for this purpose is shown in Figure 2.

Figure 2 – rotary evaporator set-up used for oxygen carrier synthesis

The ideal formulation of the ceramic dough that is ready for extrusion is dependent on several factors, such as the particle size and particle size distribution, viscosity and rheology of the mix as well as the choice of solvent, dispersant, binder and the plasticizer, each of which plays an important role to impart the right property to the dough that needs to be extruded. In our case, the following components were employed:     

Quaternary ammonium compound as dispersant Ammonium hydroxide as peptizing agent PVA as a binder Water soluble starch as auxiliary binder Water both as binder/solvent

Dispersant is added to improve separation of particles and to prevent settling or clamping; its role is akin to that of a surfactant (Surface Active reagent). Plasticizers are additives with low molecular weight that reduce the deformation temperature of the binder to room temperature or lower. It acts as an internal lubricant to aid in densification. The choice of plasticizer is important, the most important criterion being that the plasticizer must be soluble in the binder. PEG is an effective plasticizer for PVA. The relative concentration of the two must be selected and adjusted properly by experimental trials for optimum results. In our case, use of PEG was not required.

8

A binder glues together the particles of a ceramic body to give it strength after forming. PVA is a classic binder. In the case of aqueous solutions (using water as solvent), using PVA as binder is advantageous because of its good binding property, low viscosity, appreciable pseudoplasticity and easy removal during burnout. In some cases, other high-polymer compounds such as cellulose or polysaccharides (sugars) can act both as plasticizers as well as binders, in high-shear forming techniques (with viscous mixtures). After obtaining a suitable viscosity of blended materials, they were extruded using a handheld single-screw manual extruder. Extrudates were dried on a stainless steel of aluminium plate at 220°C overnight. Calcination of the dry extrudates was carried at 950°C or 1050°C for 6 or 12 hours depending on the materials, using a well-conceived heating schedule for the binder burn-out and consolidation of the carrier material; the firing schedule is shown in Figure 3:

T°C/xh

500°C/2h

350°C/2h

1°/min.

5°/min.

5°/min.

½°/min. RT

RT

Figure 3 - Binder burn-out and calcination schedule adopted in this work for the calcinations of the extruded oxygen carriers

Calcined oxygen carriers were sieved into size ranges of 125-180 μm and 180-250 μm. In the case of relatively hard materials, occasional dry ball-milling (using alumina jar and alumina milling balls) was resorted to. In every case, crushing strength and apparent density of the oxygen carriers were measured before the fluidized bed experiment. To perform the phase analysis, X-Ray diffraction (XRD) was done for both fully oxidized and fully reduced particles of all the tested oxygen carriers. Results of XRD tests will be discussed in a later section. Production data and properties of the tested oxygen carriers are present in Table 1.

9

System tested

Composition ratio (wt%)

ID

CuO-CeO2 CuO-CeO2 CuO-CeO2* CuO-GDC* CuO-GDC Fe2O3- CeO2 Fe2O3- CeO2 Fe2O3-GDC Mn2O3- CeO2

60-40 60-40 60-40 60-40 60-40 60-40 60-40 60-40 60-40

COC-ÅF COC1 COC2 COGDC1 COGDC2 FOC1 FOC2 FOGDC MOC

Calcination temperature (°C)/duration (h) 950/6h 950/6h 950/12h 950/6h 950/12h 950/6h 950/12h 950/6h 1100/6h

Size range μm

Crushing strength (N)

Apparent density (kg/m3)

180-250 125-180 125-180 125-180 180-250 125-180 180-250 125-180 180-250

0.33 0.42 0.99 is presented in table 4. Oxygen carrier Temperature (°C) Methane conversion (%)

COC

FOC1

FOC2

MOC

FOGDC

COGDC

900

925

950

950

950

950

900

925

88

95

68

74

< 10

88

100

100

Table 4 – Summary of methane conversion by the oxygen carriers for ω>0.99

Gas yield of methane versus degree of mass-based conversion for the ceria-based oxygen carriers is plotted in Figure 7.

γ

γ vs. ω for ceria-supported oxygen carriers with CH4 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

COC at T=925 COC at T=900 FOC2 at T=950 FOC1 at T=950 MOC at T=950 1

0.99

0.98

0.97

ω Figure 7 - Gas yield () vs. degree of mass-based conversion () for ceria-supported carriers using methane as fuel

As seen from Figure 7, the reactivity of COC at both 900 and 925°C is very high. Also, with time, the conversion of methane is slightly increased, up to about 95% at 900oC and 99% at 925oC. Thus, as expected, methane conversion is higher at higher temperature, such that at 925oC when the conversion is nearly complete. in order to maintain good fluidization conditions and prevent agglomeration of materials in the reactor bed considerably higher flow rate of methane (900 ml/min) was necessary during the test due to the high density of COC (and COGDC as well) formulation. However, the high flow rate of the fuel seems to be the reason why methane was not fully converted by the copper oxide containing oxygen carrier. The somewhat lower conversion of methane at 900°C compared to that at 925°C could be a function of temperature as well as the high flow rate. In summary, it could be concluded that ceria- based copper oxide is a promising and favourable oxygen carrier. In the case of FOC particles, methane conversion was complete in the very beginning of the methane cycle. With time, however, the fuel conversion decreased which could be due to the depletion of oxygen in the carrier particles. Generally, iron particles are not expected to cause full conversion of methane with this setup. 16

Cho et al. tested iron oxides particles supported on alumina in the experimental setup and conditions similar to this work and have reported similar results32. Very high conversion of methane was observed at the beginning, followed by a decrease due possibly to the oxygen depletion. Abad et al. investigated, in detail, the use of iron oxide for its application in CLC systems23. They used both syngas and natural gas in a 300 W fluidized bed unit to study the fuel conversion performance of iron oxides. They reported low conversion (~70%) when methane was used at 800°C; the conversion increased up to 94% at 850°C by lowering the fuel flow rate. In the case of MOC, very low methane conversion was observed at 950°C. In order to investigate the reason for this apparent anomaly, MOC was tested at 800, 850 and 900oCas well. The Ellingham diagram for the Mn-O system is shown in Figure 8. The test results showed that at none of the temperatures employed for testing (viz., 800-950°C), methane conversion improved for MOC samples. The XRD signatures collected on fresh and fully oxidized samples showed no trace of Mn2O3; all samples contained hausmannite (Mn3O4) and ceria (CeO2) only; in the fully reduced samples (reduced by syngas) the phases detected were Mn3O4, MnO and CeO2. This can be explained by the predicted equilibrium of this system, as seen in the Ellingham diagram in Figure 8. At 5% O2 it is clear that the phase change between Mn2O3 and Mn3O4 occurs below 800oC, hence Mn3O4 will be the stable phase in the experiments conducted here. During the reducing phase the equilibrium product will be MnO.

Figure 8 – Ellingham diagram of manganese-oxygen system33

Zhu et al. investigated the cerium-manganese mixed oxides in details for the oxidation of methane and n-butane 34 . They presumably synthesized and used Mn-substituted ceria (Ce0.5Mn0.5O1.75 and Ce0.8Mn0.2O1.9) but observed the presence of separate phases of manganese oxide during the test. They mentioned the clear observation of transmission from MnO to Mn3O4 during the oxidation cycle. It is worth pointing out that when they tried to oxidize Mn3O4 to Mn2O3 in pure oxygen at 700°C, it was not successful. Also, the partial pressure of oxygen for the MnO-Mn3O4 equilibrium was shifted to lower values due to the interaction between ceria and manganese oxide. It may be possible that a similar interaction mechanism is operative in our testing protocol with MOC. 17

From the foregoing discussion it appears that the Mn2O3-CeO2 system requires a thorough, systematic and comprehensive investigation. According to Johansson et al. similar results have been reported by many researchers24 but Adanez et al. reported methane conversions higher than 80% using a TGA system25. Gas yield of methane versus degree of mass-based conversion for the GDC-based oxygen carriers is plotted in Figure 9.

γ vs. ω for GDC-supported oxygen carriers with CH4 1.2 1

γ

0.8

FOGDC

0.6

COGDC T=925

0.4

COGDC T=900

0.2

FOGDC T=950

0 1

0.99

0.98

0.97

ω Figure 9 – Gas yield () vs. degree of mass-based conversion () for GDC-supported carriers using methane as fuel

As shown in Figure 9, COGDC converted methane fully (100%) both at 900 and 925°C and the conversion remained stable with time. The performance is comparable to the results obtained with the COC system, discussed above. For both particles, tests were done at 900 and 925oC with a 900 ml/min flow rate of methane. As discussed above, some unconverted methane was detected in the outflow stream of the COC tests, while it was fully converted in the case of COGDC under identical experimental conditions. It, therefore, appears that the contribution of pure ceria towards the reactivity of copper oxide is somewhat lower than that of GDC. It should be noted that the conversion rate of methane by COC increased with time (Figure 7), even though methane was not purged for more than 12s, because further reduction of copper(I) oxide to elemental copper was not desired. The approximate duration of methane purge was calculated based on the flow rate so that the formation of metallic copper was prevented, although from a thermodynamic point of view, the formation of some small amounts of copper is inevitable. In comparison to the reactivity of FOC particles (discussed above), FOGDC oxygen carrier interestingly showed higher conversion of methane, lasting for longer time, with higher activity as the test progressed. Cho et al. reported a decreasing conversion of methane with time for iron oxides32. Also, Johansson et al. tested some iron oxides sintered at different temperatures and supported on various materials like silica, alumina, zirconia and magnesium aluminate 35 . Large amount of unconverted methane was detected in the flue gas stream, signifying rather low reactivity with respect to conversion of methane in the presence of 50% 18

steam. These results indicate that FOGDC could be a promising oxygen carrier for large scale CLC applications. The preliminary test results show that the contribution of GDC towards the reactivity of oxygen carriers examined in this work is higher and more favourable than ceria. Full conversion of methane by COGDC and around 90% conversion by FOGDC materials confirm the potential of GDC as an effective participating support and warrant the need for more systematic investigation of the reactivity and performance of these materials in the future, to unequivocally establish their benign contribution.

3.2.2.2

Syngas conversion

Syngas conversion was investigated to assess the performance of the developed oxygen carriers with the gaseous products derived from the solid fuel gasification process. The syngas conversion for all the materials tested in this work was over 99%. The gas yield (γ) vs. the degree of mass-based conversion (ω) is plotted in Figure 10. For the case of MOC, the conversion for most of the process is over 99% and the decreasing rate is probably due to the oxygen depletion in the materials.

γ vs. ω for ceria-supported oxygen carriers with syngas at 950oC 1.005

FOC1

γ

1

FOC2

0.995 0.99 0.985

FOC2

MOC

0.98

FOC1

0.975

MOC

0.97 1

0.99

0.98

0.97

0.96

ω

Figure 10 - Gas yield () vs. degree of mass-based conversion () for ceria-supported carriers using syngas as fuel

It is concluded that ceria and GDC supported oxygen carriers can fully convert the syngas.

3.2.2.3

Phase analysis

XRD signatures were collected on all the oxygen carriers tested in this work, in order to examine the phase change(s), if any. The X-ray diffraction patterns were obtained on each sample in fully oxidized as well as in fully reduced state. Particles in fully oxidized state were obtained by flowing 5% O2 in a N2 stream after the last reduction phase. Enough time was 19

allowed for the particles to get fully oxidized. The samples were cooled to room temperature in the oxidizing environment. Samples in fully reduced state were obtained by cooling the carrier in the fuel reactor in a dynamic flow of high purity nitrogen gas after the last reduction cycle was completed. The X-ray diffraction results are summarized in Table 5. The reducing gas in the case of CuO-based carriers was methane, while syngas was the fuel for the Fe2O3and Mn2O3-based particles due to better and complete reactions. Oxygen carrier Reducing gas

COC

FOC1

Methane Syngas

FOC2

MOC

Syngas Syngas

FOGDC

COGDC

Syngas

Methane

Phases identified in reduced sample

Cu2O CuO CeO2

Fe3O4 CeO2

Fe3O4 CeO2

MnO Mn3O4 CeO2

Fe3O4 Ce0.9Gd0.1O1.9-x

Cu2O Cu Ce0.9Gd0.1O1.9-x

Phases identified in fully oxidized sample

CuO CeO2

Fe2O3 CeO2

Fe2O3 CeO2

Mn3O4 CeO2

Fe2O3 Ce0.9Gd0.1O1.9

CuO Ce0.9Gd0.1O1.9

Table 5 – Phase analysis summary in reduced and oxidized oxygen carriers after testing

The XRD results presented in Table 5 conform to the phases expected in these carriers in both oxidized and reduced conditions. Moreover, there was no evidence of the formation of new compound or solid solutions between the carriers and the support in any of these cases. As stated earlier, Mn3O4 was seen rather than Mn2O3 even in the fully oxidized sample of MOC which is reasonable according to the phase diagram in figure 8. However, the behavior of FOC and FODGC carriers was as expected and the phases in the XRD patterns are as expected. In these cases, according to the phase analysis, further reduction of Fe3O4 to FeO did not occur, thereby satisfying the theoretical phase equilibria in the iron-based systems during successive redox cycles, which makes both these materials promising as oxygen carriers, especially FOGDC with high methane conversion. Copper oxide particles supported on ceria and GDC both showed excellent performance, detection of elemental copper in the case of COGDC particles is a result of longer reduction period. Existence of CuO in the XRD analysis of COC shows the presence of unreacted CuO particles even in the reduced sample, due possibly to the fact that there is 60 wt% CuO in the sample. Since testing with ceria (discussed above) showed that ceria has the propensity to oxidize the fuel at its own, a similar behavior could be speculated (tests underway) with DGC as well. However, XRD is not the right tool to reveal if nonstochiometric ceria (CeO2-x; x < 0.5) or GDC (Ce0.9Gd0.1O2-x; x < 0.5) were formed. X-ray photoelectron spectroscopy (XPS) or extended X-ray absorption fine structure (EXAFS) technique would be adequate tools for such investigation in future studies.

3.2.2.4

Fluidization and agglomeration characteristics

Defluidization incidence was monitored by pressure difference measurements over the bed of the reactor. All the oxygen carriers tested in this study showed very good fluidization properties during the successive oxidation and reduction cycles. No defluidization was seen during the tests except for FOC1 particles after reduction with syngas. Defluidization after reduction is not unusual and is expected for oxygen carriers with high reactivity during a fuel 20

cycle. This defluidization could be due to the formation of a more reduced phase during reduction, in the case of iron oxide-based carriers this could be FeO (wustite). Agglomeration was seen in the case of FOC2 sample (6040 Fe2O4-CeO2 calcined at 950°C/12h) after the test. Around 15% of the sample was found agglomerated on the quartz bed; however, the agglomerated particles were very soft and could be broken down easily by slight tapping of the reactor. Small agglomeration was observed in the case of COGDC particles as well. According to the XRD results in Table 5, metallic copper exists in the reduced COGDC sample which could be the reason for the observed agglomeration. Transformation of copper oxide to metallic copper makes particles stick together and thus agglomerate at high temperature. For other oxygen carriers, no sign of agglomeration was observed. This is a distinct advantage for the carrier materials supported on ceria or GDC, whereas severe cases of agglomerations have been reported in the case of other supports such as alumina and zirconia. In the case of COC particles (made by freeze-granulation and calcined at 950°C/6h), large amount of dust was seen on the top part and along the walls of the reactor. This is a sign of high attrition under the test conditions in the fluidized bed reactor. Around 10% of the materials in the bed got stuck to the top part of walls of the reactor and could not be recovered after the test. This could be attributed to the disintegration of the granulated (somewhat hollow) particles under the combined force of fluidization and reaction.

3.2.3

Oxygen release

Oxygen release aspect of oxygen carriers is one the interesting properties to investigate. If a particle can release oxygen to the gas phase in an inert atmosphere- such as nitrogen- it is deemed a promising carrier material for application in solid fuel conversions via chemicallooping with oxygen uncoupling31. Solid fuels cannot penetrate to the surface of the oxygen carriers so it is necessary that oxygen be available on the surface to react with the solids. In this work, the only particles that released oxygen during inert cycles were COC and COGDC. This is not surprising, since CuO is well known to decompose to Cu2O at these temperatures, with subsequent release of gas phase oxygen. Spontaneous oxygen release was not seen for either the iron- or manganese-based particles. The behaviour of ceria and GDC-supported CuO particles during inert cycles is shown in Figure 11. Pure nitrogen was purged for 360s at constant temperature.

21

oxygen release characteristics of ceria-based carriers 6 oxygen concentration %

5 4

MOC T=900

3

FOC1 T=900

2

FOC2 T=900 COC T=875

1

COC T=875 COC

0 0

100

a

200 300 time (sec)

400

oxygen concentration %

oxygen release characteristics of GDC-based carriers 6 5 4 3 2 1 0

COGDC T=900 COGDC T=925 FOGDC at T=900 0

b

100

200

300

time (sec)

Figure 11 – oxygen release of: (a) ceria- and (b) GDC-based oxygen carriers during inert cycles.

3.2.4

Oxidation phase

The reactivity of oxygen carriers during oxidation - especially after reduction cycles by fuelis sometimes followed in order to estimate the residence time needed by the reduced particles in the air reactor to get fully oxidized. Figure 12 shows the oxidation behaviour of ceria and GDC-based oxygen carriers after reduction by methane or syngas.

22

oxygen concentration %

oxygen uptake by the reduced ceria-based oxygen carriers 5.5 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 0

a

100

200

300 400 time (sec)

500

600

700

MOC after syngas FOC1 after methane FOC1 after syngas FOC2 after methane FOC2 after syngas COC T=900

oxygen concentration %

oxygen uptake by the reduced GDC-based oxygen carriers 5.5 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0

COGDC after methane T=900 COGDC after methane T=925 FOGDC after methane T=950 0

b

100 200 300 400 500 600 700 800 900 1000 time (sec)

FOGDC after syngas T=950

Figure 12 – Oxygen uptake characteristics of the: (a) ceria-based and (b) GDC-based oxygen carriers during oxidation in 5% O2-N2 stream after their reduction by fuel

It appears that the behavior could be divided into two distinct patterns. They either oxidized gradually over a long period of time in oxidizing stream, or they consumed the available oxygen quickly after a short induction period. FOC2 and FOGDC both consumed oxygen quickly and fully to reach the original chemical state. MOC and FOC1, on the contrary, took longer and were gradual in reaching the original fully oxidized state. This oxidation behaviour after reduction by syngas demonstrates the higher activity of these oxygen carriers in syngas compared to methane. Generally, in the case of iron- and manganese-based oxygen carriers, the oxidation phase subsequent to the syngas reduction cycle is longer than that for the methane cycle due to the higher reactivity of the carrier with syngas, which led to higher conversion as well. In the case of FOGDC on the other hand, the particle was oxidized for more than 300s after reduction by methane which signifies higher reactivity leading to higher degree of phase change to a reduced form of the carriers. Copper oxide particles changed to the fully oxidized state gradually which is commensurate with the theoretically predicted thermodynamic considerations at and above 900°C. In summary, all the particles got oxidized to the desired forms after the reduction periods.

23

3.2.5

Temperature variation during oxidation and reduction cycles

As stated in the introduction section, oxidation reactions are exothermic and reduction reactions could be either exothermic or endothermic. For example, for copper oxide-based carriers, it is advantageous that both the reduction and oxidation reactions are exothermic. In Tables 6 and 7, changes in temperature during the oxidation and reduction reactions for the tested oxygen carriers are presented. The positive and negative signs refer to exothermic and endothermic reactions, respectively. However, this is only the observed temperature change which should not be confused with the actual enthalpy changes. But it gives a relative direction and size of the enthalpy change.

FOC1 FOC2 MOC FOGDC

Reduction by methane

Oxidation after reduction by methane

Reduction by syngas

Oxidation after reduction by syngas

-14 -14 +2 -14

+18 +18 +10 +15

+2 +2 +18 +2

+22 +22 +25 +18

Table 6 – Temperature variation for the Fe2O3- and Mn2O3-based oxygen carriers.

COC COGDC

Reduction by methane at T=900OC

Oxidation after reduction by methane at T=900OC

Reduction by methane at T=925OC

Oxidation after reduction by methane at T=925OC

+22 +27

+6 +9

+18 +25

+4 +7

Table 7 – Temperature variation for the CuO-based oxygen carriers.

As can be seen, all the Fe2O3-based oxygen carriers showed nearly identical behaviour with regard to temperature variations during the fuel and the oxidation cycles after reduction, though the oxidation of the FOC series is a little more exothermic than for FOGDC but the difference is only minimal. Also, the reduction reaction with syngas is exothermic while it is endothermic with methane for the iron-based particles. On the other hand, the copper oxidebased particles showed an exothermic trend for the reduction as well as the oxidation reactions during the all tests. The trend in the temperature variation for the copper oxidebased carriers is interesting. For example, the exothermicity of COGDC is higher than that of the COC particles. This could be attributed to the slightly higher reactivity of COGDC carriers than of COC. The temperature variation data presented in Tables 6 and 7 shows the superiority of the GDC-supported materials over those supported on ceria. This could be interrelated to the better oxygen transport capability of the GDC support due to the oxygen ion vacancies in it by virtue of doping. Again, the behavior of MOC is different. First, both reduction and oxidation reactions are exothermic by a large amount (in terms of temperature changes). It is worth mentioning that methane conversion with MOC was below 10%, but the temperature increased by 2° during methane cycle and by 10° during the post-reduction oxidation phase purge. Also, the temperature increase is significantly higher than those observed with iron oxide-based samples during syngas and oxidation cycles.

24

It may be recalled that the XRD results showed the presence of MnO produced via Mn3O4 ↔ MnO and not Mn2O3 (Mn2O3 ↔ Mn3O4). The theoretical oxidation enthalpies of MnO and Mn3O4 at 950oC are as below21, 34: 6 MnO + O2  2 Mn3O4

ΔH = -451.4 kJ/mol

(19)

4 Mn3O4 + O2  6 Mn2O3

ΔH = -189.5 kJ/mol

(20)

The enthalpy change for the Mn3O4 ↔ MnO reaction is much higher than that for the Mn2O3 ↔ Mn3O4 reaction, thus, a large temperature increase is expected during the oxidation process. Thus, by virtue of the presence of MnO in the reduced sample of MOC, the observed temperature increase is not unexpected or abnormal. Same argument could be made for the high temperature increase seen in the case of reaction of MOC with syngas.

25

4

Conclusion

Samples of copper oxide, iron oxide and manganese oxide oxygen carriers supported on ceria and gadolinia-doped ceria (GDC) were fabricated by extrusion and their behavior was investigated by fluidized bed reactivity tests. Relevant parameter of significance to the CLC process, such as fuel conversion, oxygen release measurement, fluidization properties and, temperature variations during fuel and oxidation cycles, were examined for all the oxygen carriers made in this work. In the light of the obtained results, it is concluded that ceria and GDC-supported oxygen carriers hold promise for CLC applications. All the oxygen carriers showed very good fluidization properties during the tests without any agglomeration. Copper oxide-based oxygen carriers showed nearly full conversion of methane with high oxygen release, high temperature increase during oxidation and no sign of defluidization. FOGDC showed very high and improved conversion of methane together with favourable reactivity during oxidation periods after the fuel cycles. The performance of FOGDC and COGDC materials was the most promising in terms of their reactivity behavior. However, the behaviour of MOC was somewhat different. Some explanation has been offered in this thesis for the observed behavior of MOC in the light of thermodynamic consideration of the phases involved and the XRD results. Nevertheless, MOC system needs to be more comprehensively and systematically investigated in the future. Based on the preliminary results obtained in this work, there are opportunities to make great improvements in the performance of copper oxide and iron oxide-based carriers with ceria and GDC supports. Finally, the GDC supported oxygen carriers showed better performance than their ceria counterparts.

26

5.

Acknowledgements

I would like to express my greatest thanks to my supervisors Henrik Leion and Abdul-Majeed Azad for their close, friendly and precise supervision and help during my work. I am thankful to Anders Lyngfelt for providing me with the opportunity to work with the CLC group. I also owe lots of my knowledge and work to Abdul-Majeed Azad who taught me a lot both scientifically and ethically. I thank the members of the CLC group in the chemical engineering department – Erik Jerndal, Golnar Azimi, Peter Hallberg, Dazheng Jing and Mehdi Arjmand – who helped me a lot and were my teachers as well. Special thanks to my examiners Tobias Mattisson and Magnus Ryden for very informative and useful discussions. Last but not least, I owe my gratitude to my parents who supported me emotionally and financially from thousands of miles away.

27

6. 1

References

United Nations Development Program, Energy for Development, Human development report, 2007

2

United Nations Development Program, Human Development Report, 20th Anniversary Edition, 2010 3

IPCC, technical paper I: technologies, policies and measures for mitigating climate change, 1996

4

Erik Jerndal, investigation of nickel and iron based oxygen carriers for chemical looping combustion, Doctoral Thesis, Chalmers university of technology, Gothenburg, Sweden, 2010 J. T. Kiehl and Kevin E. Trenberth, Earth’s Annual Global Mean Energy Budget, National Center for Atmospheric Research, Boulder, Colorado, 1997 5

6

IPCC, Summary for Policymakers, Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, 2007 7

Vante Arrhenius, On the Influence of Carbonic Acid in the Air upon the Temperature of the Ground, Philosophical Magazine, 1896, Vol 41, 237-276 8

Bo Nordell, Thermal pollution causes global warming, Global and Planetary Change 38, 305– 312, 2003 9

EPA, United States Environmental Protection Agency, 2001, available at www.epa.gov/globalwarming, as accessed 2 February 2011 10

IPCC, Summary for Policymakers. In: Climate Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, 2007 11

Golnar Azimi, Experimental evaluation and modeling of steam gasification and hydrogen inhibition in CLC with solid fuel, master of Science thesis, Chalmers University of Technology, Gothenburg, Sweden, 2010 12

IPCC, Implications of Proposed CO2 Emissions Limitations, IPCC Technical Paper 4, 1997

13

EIA, International Energy Outlook 2010, U.S. Department of Energy DOE, 2010

14

Sujata Gupta and Preety M Bhandari, An effective allocation criterion for CO 2 emissions, Energy Policy, Vol 27, 727-736, 1999 15

Kourosh E. Zanganeh and Ahmed Shafeen, A novel process integration, optimization and design approach for large-scale implementation of oxy-fired coal power plants with CO2 capture, International Journal of Greenhouse gas Control, Vol 1, 4 7 – 54, 2007 16

Abass A. Olajire, CO2 capture and separation technologies for end-of-pipe applications - A review, Energy, Vol 35, 2610-2628, 2010 17

Henrik Leion, capture of CO2 from solid fuels using chemical-looping combustion and chemical looping combustion uncoupling, Doctoral thesis, Chalmers University of Technology, Gothenburg, Sweden, 2008 18

Tobias Pröll et al, Natural minerals as oxygen carriers for chemical looping combustion in a dual circulating fluidized bed system, energy Procedia, Vol 1, 27-34, 2009 19

Anders Lyngfelt et al, A Fluidized-bed combustion process with inherent CO2 separation; application of chemical-looping combustion, Chemical Engineering Science, Vol 56, 3101–3113, 2001 20

E. Jerndal, T. Mattisson and A. Lyngfelt, Thermal analysis of chemical looping combustion, chemical engineering research and design, Vol 84, 795-806, 2006 21

Mohammad M. Hossain, Hugo I. de lasa, Chemical-looping combustion (CLC) for inherent CO2 separation-a Review, Chemical Engineering Science, Vol 63, pp 4433-4451, 2008 28

22

Pilar Gay, Carmen R. Forero, Alberto Abad, Luis F. de Diego, Francisco Garc-Labiano, and Juan Adnez, Effect of support on the behavior of Cu-based oxygen carriers during long-term CLC operation at temperatures above 1073 K, Energy and Fuel, Vol 25, pp 1316–1326, 2011 23

A. Abad, T. Mattisson, A. Lyngfelt, M. Johansson, The use of iron oxide as oxygen carrier in a chemical-looping reactor, Fuel, Vol 86, pp 1021-1035, 2007 24

M. Johansson, T. Mattisson and A. Lyngfelt, Investigation of Mn 3O4 with stabilized ZrO2 for chemical-looping combustion, ICHEME, 2006 J. Ada´nez, L. F. de Diego, F. Garcı´a-Labiano, P. Gaya´n, and A. Abad, Selection of Oxygen Carriers for Chemical-Looping Combustion, Energy & Fuels, Vol 18, pp371-37, 2004 25

26

He Fang, Li Haibin, and Zhao Zengli, Advancements in Development of Chemical-Looping Combustion: A Review, International Journal of Chemical Engineering, Volume 2009, 2009 27

A. Lyngfelt, M. Johansson, and T. Mattisson, Chemical-looping combustion -Status of development, 9th International Conference on Circulating Fluidized Beds (CFB-9), 2008, Hamburg, Germany 28

Alberto Abad, Juan Adanez, Francisco Garcia-Labiano, Luis F. de Diego, Pilar Gayan, Modeling of the chemical-looping combustion of methane using a Cu-based oxygen-carrier, Combustion and Flame, Vol 157, pp602–615, 2010 29

Yonggang Wei, Hua Wang, Fang He, Xianquan Ao, and Chiyuan Zhang, CeO 2 as the Oxygen Carrier for Partial Oxidation of Methane to Synthesis Gas in Molten Salts: Thermodynamic Analysis and Experimental Investigation, Journal of Natural Gas Chemistry, Vol 16, pp6-11, 2007 30

ZHU Xing, WANG Hua, WEI Yonggang, LI Kongzhai, CHENG Xianming, Hydrogen and syngas production from two-step steam reforming of methane over CeO2-Fe2O3 oxygen carrier, Journal of rare eraths, Vol. 28, 2010 31

Tobias Mattisson, Anders Lyngfelt and Henrik Leion, Chemical-looping with oxygen uncoupling for combustion of solid fuels, International Journal of Greenhouse Gas Control, Vol3, pp11-19, 2008 32

P. Cho, T. Mattisson, A. Lyngfelt, Defluidization Conditions for Fluidized-Bed of Iron, Nickel, and Manganese oxide-Containing Oxygen-Carriers for Chemical-Looping Combustion. Industrial and Engineering Chemistry Research 2006, 45, (3), 968-977 33

Collaboration: Authors and editors of the volumes III/17G-41D: Mn2O3: phase diagram, crystal structure, lattice parameters of high temperature phase. Madelung, O., Rössler, U., Schulz, M. (ed.). SpringerMaterials - The Landolt-Börnstein Database (http://www.springermaterials.com). DOI: 10.1007/10681735_467 34

Gong Zhu, Parag R. Shah, Raymont J. Gorte, A Study of Cerium–Manganese Mixed Oxides for Oxidation Catalysis, Catalysis Letters, Vol 120, pp191-197 35

Marcus Johansson, Tobias Mattisson, Anders Lyngfelt, comparison of oxygen carriers for chemical looping combustion, Thermal science, Vol10, pp93-107, 2006

29