Scholarly article on topic 'Construction and operation of a 10 kW CLC unit with circulation configuration enabling independent solid flow control'

Construction and operation of a 10 kW CLC unit with circulation configuration enabling independent solid flow control Academic research paper on "Chemical engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Energy Procedia
OECD Field of science
Keywords
{"Chemical-looping combustion" / NiO / "Fluidized bed" / L-valves}

Abstract of research paper on Chemical engineering, author of scientific article — Sébastien Rifflart, Ali Hoteit, Mohammad Mahdi Yazdanpanah, William Pelletant, Karine Surla

Abstract Chemical Looping combustion (CLC) is an oxy-combustion technology in which the oxygen required for combustion is supplied by metal oxides known as oxygen carriers (OC). The OC particles are employed to continuously transfer oxygen from an air reactor to a fuel reactor where oxygen is delivered to the fuel. This technology is regarded as one of the most promising CO2 capture technology in terms of efficiency penalty and cost reduction. Various CLC configurations have already been developed and tested in laboratory pilot plant scales. However more investigations are required to achieve optimized process operation. Among the different points to address, control of solid circulation rate between reactors is of highest importance regarding its effect on determination of oxygen transfer rate and solid oxidation state. A novel CLC configuration is proposed here where reactions are carried out in interconnected bubbling fluidized beds. Solid circulation rate control is achieved independent from gas flow rate in rectors through use of pneumatic non-mechanical valves as high temperature limits use of mechanical valves. Loop seals are also employed to minimize gas leakage problem. A 10 kW prototype CLC unit based on this configuration has been designed and built at IFP-Lyon in cooperation with TOTAL to operate with nickel-based oxygen-carrier particles. Successful operation has been obtained with CH4 as fuel resulting in more than 99% CH4 conversion to CO2. The effect of operating variables, reactor temperature and degree of reduction of particles on the gas conversion will be discussed in this paper.

Academic research paper on topic "Construction and operation of a 10 kW CLC unit with circulation configuration enabling independent solid flow control"

Available online at www.sciencedirect.com

m\' ScienceDirect

Energy Procedía 4 (2011) 333-340

Energy Procedía

www.elsevier.com/locate/procedia

GHGT-10

Construction and operation of a 10kW CLC unit with circulation configuration enabling independent solid flow control

Sébastien Rifflart a, Ali Hoteit b, Mohammad Mahdi Yazdanpanah b, William Pelletant b,

Karine Surla b

a TOTAL S.A., 2 Place Jean Millier, La défense 6, 92078 Paris la défense Cedex, France bIFP Energies nouvelles, Rond-point de l'échangeur de Solaize, BP3 - 69390 - Vernaison, France

Abstract

Chemical Looping combustion (CLC) is an oxy-combustion technology in which the oxygen required for combustion is supplied by metal oxides known as oxygen carriers (OC). The OC particles are employed to continuously transfer oxygen from an air reactor to a fuel reactor where oxygen is delivered to the fuel. This technology is regarded as one of the most promising CO2 capture technology in terms of efficiency penalty and cost reduction. Various CLC configurations have already been developed and tested in laboratory pilot plant scales. However more investigations are required to achieve optimized process operation. Among the different points to address, control of solid circulation rate between reactors is of highest importance regarding its effect on determination of oxygen transfer rate and solid oxidation state. A novel CLC configuration is proposed here where reactions are carried out in interconnected bubbling fluidized beds. Solid circulation rate control is achieved independent from gas flow rate in rectors through use of pneumatic non-mechanical valves as high temperature limits use of mechanical valves. Loop seals are also employed to minimize gas leakage problem. A 10 kW prototype CLC unit based on this configuration has been designed and built at IFP-Lyon in cooperation with TOTAL to operate with nickel-based oxygen-carrier particles. Successful operation has been obtained with CH4 as fuel resulting in more than 99% CH4 conversion to CO2. The effect of operating variables, reactor temperature and degree of reduction of particles on the gas conversion will be discussed in this paper.

© 2011 Published by Elsevier Ltd.

Keywords : Chemical-looping combustion; NiO; Fluidized bed; L-Valves

1. Introduction

Chemical Looping Combustion (CLC) is an oxy-combustion technology in which the oxygen required for combustion is supplied by metal oxides known as oxygen carriers (OC). The OC particles are employed to continuously transfer oxygen from an air reactor to a fuel reactor where oxygen is delivered to the fuel. Consequently direct contact between air and fuel is prevented. Thus the combustion products CO2 and H2O are obtained without being diluted by the N2 coming from the air like in normal combustion. After reduction, the reduced oxygen carrier is then transported back to the air reactor for re-oxidation purpose, hence forming a chemical loop. Accordingly, CLC can provide an effective solution for clean combustion of fossil fuels with low cost for CO2 capture and maximum energy production efficiency comparable to conventional power generation stations. Most of works on CLC have been focused on oxygen carrier development and testing and, to the authors' knowledge, all

doi:10.1016/j.egypro.2011.01.059

these works concern oxygen carriers in particle form, albeit of varying particle diameters and production methods. In a fluidized bed system described above, the criteria for a good oxygen carrier for CLC are the following:

- Able to convert the fuel to CO2 and H2O to the highest degree possible (ideal 100%)

- High reactivity with fuel and oxygen

- Low fragmentation and attrition

- Low tendency for agglomeration

- Low production cost and preferably being environmentally sound.

With respect to the ability of the oxygen carrier to convert a fuel gas fully to CO2 and H2O, Mattisson and Lyngfelt investigated the thermodynamics of a few possible oxygen carriers and concluded that the metal oxide/metal (or metal oxide of lower oxidation state) systems of NiO/Ni, Fe2O3/Fe3O4, Cu2O/Cu, CoO/Co were feasible to use as oxygen carriers [1]. A comprehensive study was made by Jerndal et al where 27 different possible systems for chemical-looping combustion were investigated with respect to thermodynamics, melting points, oxygen ratio, fate of possible sulphur species in the fuel and carbon deposition [2]. For the often studied NiO/Ni system there is one slight disadvantage, the conversion of fuel to CO2 is not complete, although very high, 98.8% at 1000°C, and higher at lower temperatures. For CoO/Co the same problem exists, however with much less favorable thermodynamics, 93.0% conversion at 1000°C, and higher at lower temperatures. In practice it means that the CO2 will contain a small amount of combustible gases, i.e. CO and H2, if these systems are used. These can either be separated and recycled or oxidized by adding oxygen downstream of the fuel reactor. In addition to having good thermodynamic properties, the oxygen carrier must react at a sufficient rate. As the amount of oxygen carrier needed in the reactors is directly related to the reactivity of the oxygen carrier, a fast rate would mean less material and thus smaller reactor sizes, and so lower production costs. In relation to this, the oxygen carriers must also be able to transfer a sufficient amount of oxygen to the fuel to complete oxidation. This is directly related to the amount of active oxygen in the oxygen carrier and is dependent on the oxygen carrier used as well as the amount of inert material in the particle. This characteristic is called the oxygen transfer capacity, Ro(-), and is defined as the mass ratio of available oxygen over the mass of carrier in fully oxidized form.

Prior to the year 2001, limited information was published on how the reactors could be designed. Since then several cold-models and hot prototype units have been built and operated. These could be classified into two general categories, interconnected fluidized bed designs and innovative processes. In the first category, the most common design is based on the conventional circulating fluidized bed (CFB) systems with addition of a second reactor as fuel reactor (FR) [3-4]. In these systems a low velocity bubbling fluidized bed reactor is devoted to fuel reactor and a high velocity riser is considered as air reactor (AR). The advantageous of this design is the fact that due to its similarity with conventional CFB combustion process, it is best adopted with existing power generation units and its phenomena are well studied to date. Studies reveal the fact that solid flow control can be achieved in this configuration, however it is a function of gas flow in the reactors and solid inventory in each reactor, i.e. it could not be adjusted independently from other parameters of fluidization. In order to improve solid circulation control and achieve proper residence time in the reactors different modified designs are developed to date. Forero et al. [5] have added a bubbling fluidized bed as AR just below the riser. De Diego et al. [6] have also used a bubbling fluidized bed as AR while separating the riser and the AR using a loop-seal. Kolbitsch et al. [7] have developed concept of Dual Circulating Fluidized Bed (DCFB) where FR is in turbulent regime forming a second loop inside the overall system. They have accordingly achieved control of the solid flow rate just based on the aeration rate in AR, independent from the gas rate in FR. Keronberger B. et al. [8] have developed an alternative design, two compartment fluidized bed reactor, with simplified design resulting in smaller installations requirement and lower cost. Son et al. [9] have proposed an annular fluidized bed CLC reactor in order to provide sufficient reaction time and optimize heat transfer between the reactors. Ryu and Al. [10] have proposed a design using two fluidized bed interconnected via solid conveying lines starting directly from inside the bed with solid injection nozzles. They have achieved control of the solid flow rate through operating variables such as diameter and number of holes on the injection nozzle. More innovative design have been also proposed as: packed bed CLC [11], packed bed membrane assisted chemical looping reactor [12] and rotating bed reactor [13].

IFP and TOTAL have started in 2008 a R&D collaboration project on CLC technology development. As a part of the project, a novel 10kWth prototype based on the interconnected fluidized bed concept has been constructed. One

of the objectives was to study CLC performances with various oxygen carriers with a flexible design under different experimental operating conditions: temperature, solids circulating rate and fuel inlet flow.

2. Materials and method

The 10 kWth prototype has been constructed with specific alloy HR-120 which provides excellent strength at elevated temperature up to 1095°C combined with very good resistance to carburizing and sulfidizing environments. The prototype (Figure 1) is composed of three interconnected fluidized bed reactors, one used as a FR with internal diameter of 0.13 m and two used as AR with i.d. of 0.1 m. The three reactors have a total height of 1 m. Bubbling beds are used to ensure sufficient contact time between solid and gas to achieve optimum reaction conversion. Beside, this will result in higher flexibility of the system and permits use of various oxygen carriers with different oxidation and reduction reaction rates during continuous operation. Gases are injected at the bottom of the reactors through staggered perforated plates through four identical hole resulting in a hole density of 509 orifice/m2. Reactors cross section increases at top in order to reduce gas velocity and trap the entrained solids back to the reactor. Pneumatic l-valves are used to control solid flow rate. Solids at the exit of the l-valves are transported through lift with 0.02 m i.d. and 2.25 m height. A blinded tee bend is employed to divert solid flow to a horizontal dilute phase conveying line. Solid gas separation is performed in a cyclone where separated solids are lead to a loop-seal located in the bottom of the cyclone dipleg. Loop-seal is used to ensure gas tightness between fuel reactor (R1) and air reactor (R2). Solid particles are then transferred to the second air reactor (R3) through an identical second circulation line. The same description is adapted to the circulation between R3 back to the first reactor R1. The entire reactor system is hanging on a scaffold and guided vertically using rails below the air and fuel reactors. The use of a third reactor add flexibility in term of experimental investigation but proposed circulating path could be applied to a 2 reactors configuration, one for FR and the other one for AR but with a larger i.d. compared to the one used here.

The reactors are heated by external electrical heaters regulated by different thermal regulators. The incoming gases to the different reactors can also preheated up to 500 °C. In the starting procedure, the heating reactors are heated up while air is used as fluidizing agent in the air reactors and nitrogen in the fuel reactor. When the temperatures in the reactors are high enough to begin fuel operation, nitrogen is switched to fuel (CH4). During stable operation, external heating is maintained to compensate heat losses from the system .

Nitrogen is by default used for fluidization of the particles seals. In an industrial system, steam or CO2 could be prefereded to avoid any dilution of the fuel reactor exit gases. Gas inflow for l-valves, loopseals and lifts are controlled using specific mass flow rate controllers connected to a computer for data collection. Pressure drops are measured by digital pressure transducers and are automatically registered on a computer with adjustable frequency. Several pressure taps are located all over the installation to measure pressure drop in different sections of the circulation loop. The pressure measurements are used to supervise the bed heights, to evaluate particle flows in the lifts, in the loop seals and in the L- valves as a function of operating conditions. A total of 21 thermocouples of type K monitor the temperatures in the unit. Four thermocouples

Figure 1 : Scheme of the 10

kWth prototype

Realtor 1 Reactor 2 Reactor 3

are placed in the fuel reactor, four also in each of the air reactors, ones in each lift, L-valve and loopseal. The gas flows leaving the air reactors and the cyclones are led through a heat exchanger device to promote cooling. After cooling, the flow from the cyclones and the reactors passes through filter bags, where entrained particles are trapped. The fuel reactor exit gas is led to a water seal, where most of the steam condenses. After cooling, a fraction of each gas leaving the air reactors and the fuel reactor is led first through a small filter, where any entrained fines are removed, and then to a cooling device where most of the remaining steam is condensed. Finally, the two gas streams enter the analyzers, CH4, CO and CO2 gas concentrations are measured by non-dispersive infrared analysis (NDIR), O2 concentration is measured by a paramagnetic analyzer and H2 concentration by gas conductivity. It must be noted that the two air reactors uses the same oxygen analyzer. Switching between the two reactors during operation allows to measure the different oxygen concentration in the gas outlet streams. Using a specific software, temperatures, differential pressures, incoming gas flows and gas concentrations are logged every five seconds. A supervision system allows unmanned operation of the prototype and permits operation at combustion conditions during long periods of time with continuous fuel feed.

40 Time (s)

Figure 2 : Pressure drop evolution in reactors when circulation is stopped between R2 and R3

In the present study, NiO/NiAl2O4 (60/40) oxygen carrier is investigated. This Ni-based oxygen carrier has been manufactured by Marion Technologies (France) and prepared by precipitation. Its average arithmetic and harmonic diameters are 171 and 125 |j.m, respectively. The bulk density of the particles is 3200 kg/m3. The BET surface area of the particles is 7 m2/g. The theoretical oxygen transfer capacity, R0, is 12.8%, corresponding to reduction of NiO to Ni. This value will be used for the calculation of oxidation degrees, even if we have to noticed that some authors, have measured oxygen transfer capacity up to 14.5 wt% according to thermogravimatric tests, [14], due to the slower reduction of NiAl2O4 to Ni and Al2O3.

3. Results

3.1. Analysis of a typical run

The high flexibility of the unit, given by the independent control of solid circulation and gas flow rates, allows stable operation with fuel flow rate from 2kWth, up to 10 kWth. In this run, CH4 was injected into R1 during 220 min with a constant inlet flow rate of 0.75 Nm3/h corresponding to thermal input power of 7.5 kWth based on low calorific value. On the air reactor sides, R2 was fed with a constant air flow rate of 4 Nm3/h and R3 with a constant air flow rate of 3.5 Nm3/h. All l-valves, loopseals and lifts were fed with nitrogen.

The circulating solid flow rate was set to 135 kg/h which correspond to a use of 12.35% of the oxygen transfer capacity of the solids. This circulating flow rate is controlled by the aeration of the l-valve placed before the lift between 2 reactors and is evaluated in operation via the measurement of the pressure drop in the lift. Actually, it has been determined on a cold flow model of the unit that a linear relation exists between solid flow rate and lift pressure drop [15]. However, at the end of each run, the solid circulating flow rate is controlled with the following procedure: while maintaining constant and equal gas flow in the three reactors and in the three lifts, the aeration of the l-valve between R2 and R3 is stopped, resulting in the break of the circulation between R2 and R3 while the circulation is maintained from R1 to R2 and from R3 to R1. As a consequence, the inventory of R2 is increasing while the inventory of R3 is decreasing and the one of R1 is maintained constant. Figure 2 represent the pressure

Figure 3 : Dry gas concentration at the outlet of fuel reactor (R1)

Figure 4 : Gas concentrations at the outlet of the air reactors (R2 and R3)

drop evolution in both reactors during this procedure. The solid flow rate can be calculated using the slop of the curve using the following relation:

AR dP,

In this case, the calculated solid flow rate through this procedure is 137 kg/h ± 15%. It can be observed that the pressure drop through R1, and so the inventory of R1, is constant confirming that the solid flow rate from R1 to R2 and from R3 to R1 are equals. The total inventory of solid is 33 kg, during normal operation the bed heights of reactors are equal and are maintained constant.

Figure 3 gives the dry gas composition at the outlet of the fuel reactor, R1, during the run. The average composition is 85.9% CO2, 1.3% H2, 0.4% CO and 0.4% CH4, the remaining gas composition is nitrogen coming from loopseal placed at the inlet of R1 for the solid coming from R3. This dilution of the exhaust gases enables us to calculate that a flow rate of 0.1 Nm3/h of N2 is entrained with the particles to R1 which correspond exactly to the N2 feed flowrate to the downstream part of the loopseal. The gas yield, yred, is defined as the fraction of methane that has been fully oxidized to CO2 through the formula :

+ Ych4 + Y

Here, the gas yield is equal to 99.15%, which is close to the thermodynamical limit at the operating temperature of the fuel reactor which was regulated at 800°C [16]. Once corrected from N2 dilution due to the loopseal, the average composition is 97.7% CO2, 1.5% H2, 0.4% CO and 0.4% CH4.

Figure 4 gives the dry gas composition at the outlet of the air reactors. As previously mentioned, only one oxygen analyzer is used for the two air reactors. During operation, the O2 concentration at the outlet of the first air reactor, R2, is null due to high kinetic of the oxidation reaction and good contact in the fluidized bed reactor. So most of the time, the analyzer is used to measure the outlet of the second air reactor, R3, were the oxygen excess can be measured. For short periods, the analyser is switch to R2 in order to verify the complete conversion of O2 in this reactor, as it can be seen on the figure 4 for t comprise between 120 and 130min. At the outlet of R3, the average O2 concentration measured is 2.6%. Considering the air inlet flowrate of R2 and R3, respectively 4 and 3.5 Nm3/h, this

1 1 ' *- u

'* * ^reii

Q-o y*H, .

„_-E -

800 T(°C>

Figure 5 : Gas yield , yred , and concentration of H2 and CO as a function of solid conversion difference (AX)

Figure 6 : Gas yield , yred, and concentration of H2 and CO as a function of temperature

value is in accordance with the global air ratio of 1.05 defined as the quantity of air injected over the stoechiometric quantity of air needed. Finally, no CO or CO2 can be detected at the outlet of R2, which indicated that no coke is formed during reduction, and as a consequence 100% of the CO2 produced is captured.

3.2. Effect of solid circulation flow rate

To evaluate the effect of the solid flow rate on the methane conversion, four experiment have been realized with a CH4 inlet flow rate of 0.5 Nm3/h but with different solid circulations flow rate : 33, 56, 95 and 130 kg/h which correspond to AX = 33.7%, 19.9%, 11.0% and 8.6% respectively, where AX is defined as the difference in solid conversion between the outlet of R3 and the outlet of R1. During these 4 experiments, the temperature of R1 was regulated to 800°C and the total air flow rate (R2 +R3) was maintained to 5 Nm3/h.

Results are presented on figure 5 : the gas conversion is kept over 99% for the whole range of AX from 8 to 33% revealing the large flexibility of the process using NiO particles. The gas concentrations on the figure have been corrected from N2 dilution due to the loopseal. H2 and CO concentration are constant for the four tests and respectively around 1.5% and 0.4%. CH4 has not been represented as the measured value for these tests were under 0.1%.

3.3. Effect of temperature

Three tests have also been conducted with a CH4 inlet flow rate of 0.5 Nm3/h but with different regulation temperature in R1: 750°C, 800°C and 850°C. For these tests, the solid circulation flow rate was set to 56 kg/h and the total air flow rate (R2 +R3) was maintained to 5 Nm3/h.

Results are presented on figure 6 : the gas conversion is kept over 99% for the whole temperature range comprise between 750°C and 850°C. This result is in accordance with the thermodynamical equilibrium prediction given by Mattisson [15] who expects more that 99% CH4 conversion for temperature under 900°C. Regarding H2 and CO concentration, the values on the figure have been corrected from N2 dilution and are respectively around 1.5% and 0.4% with a slight tendency to increase with the temperature as could be expected from thermodynamical equilibrium prediction. CH4 has not been represented as the measured value for these tests were under 0.1%.

4. Conclusions

A 10 kW CLC pilot has been constructed at IFP-Lyon in cooperation with TOTAL with a design based on the use of pneumatic non-mechanical valves, also called l-valves, to enable the control of the solid circulation rate independently from the gas flow rate in the reactors. This pilot has been successfully operated with nickel-based oxygen-carrier particles and CH4 as fuel. Stable operations have been achieved with more than 99% CH4 conversion to CO2 in a wide range of operating conditions. Some H2 and CO are also produced in the fuel reactor, resulting in concentration in the CO2 rich gas of 1.5% and 0.4% respectively. In industrial configuration, these gases could be separated and recycled or oxidized by adding a small amount of oxygen downstream of the fuel reactor. The same level of conversion has been obtained for the range of solid circulation rate explored, equivalent to a range in solid conversion AX from 8 to 33%, revealing the large flexibility of the CLC process using NiO particles. No effects of fuel reactor temperature have either been noticed on the CH4 conversion from 750°C to 850°C. Finally, the air ratio has been adjusted to 1.05, i.e. 5% of additional air injected compared to the stoechiometric quantity of air needed, with no CO or CO2 detection at the outlet of the air reactor. This indicates that no coke is formed during reduction, and as a consequence 100% of the CO2 produced is captured.

Nomenclature

Yred Gas yield, -

Ar Cross area of the reactor, m2

MSR Mass of solid inventory in the reactor, kg

PR Pressure drop through reactor, Pa

X Solid oxidation degree ( =1 if fully oxidized and 0 if fully reduced), -

WS Solid flow rate, kg/s

Yi gas fraction of the specie i, -

References

[1] Mattisson, T. and Lyngfelt, A. Capture of CO2 using chemical-looping combustion. in Scandinavian-Nordic Section of Combustion Institute. 2001. Göteborg, 163-8.

[2] E. Jerndal, T. Mattisson, A. Lyngfelt, Thermal Analysis of Chemical-Looping Combustion Chemical Engineering Research and Design, Volume 84, Issue 9, September 2006, Pages 795-806.

[3] Lyngfelt Anders, Thunman Hilmer, Construction and 100 h of Operational Experience of A 10-kW Chemical Looping Combustor, in: Thomas D.C., Benson S.M. (Eds.), Carbon Dioxide Capture for Storage in Deep Geologic Formations, Elsevier Ltd., 2005, pp. 625-645.

[4] Ryu H.J., Jin G.T., Yi C.K., Demonstration of Inherent CO2 Separation and No NOx Emission in a 50kW Chemical-Looping Combustor: Continuous Reduction and Oxidation Experiment, in: Morris T., Gale J., Thambimuthu K. (Eds.), Elsevier Ltd., 2005, pp. 1907-1910.

[5] Forero C.R., Gayan P., de Diego L.F., Abad A, Garcia-Labiano F., .Adanez J, Syngas combustion in a 500 Wth Chemical-Looping Combustion system using an impregnated Cu-based oxygen carrier, Fuel Processing Technology, 90 (2009) 1471-1779.

[6] De Diego L.F., Garcia-Labiano F., Gayan P., Celaya J., Palacios J.M., Adanez J., Operation of a 10 kWth chemical-looping combustor during 200 h with a CuO.Al2O3 oxygen carrier, Fuel, 86 (2007) 1036-1045.

[7] Kolbitsch P., Pröll T., Bolhar-Nordenkampf J., Hofbauer H., Design of a Chemical Looping Combustor using a Dual Circulating Fluidized Bed Reactor System, Chem. Eng. Technol., 32 (2009) 398-403.

[8] Kronberger B., Johansson E., Löffler G., Mattisson T., Lyngfelt A., Hofbauer H., A Two-Compartment Fluidized Bed Reactor for CO2 Capture by Chemical-Looping Combustion, Chemical Engineering & Technology, 27 (2004) 1318-1326.

[9] Real Son Sung, Kim Sang Done, Lee Jea-Keun. Continuous and Semi-Continuous Operations of Chemical-Looping Combustion in an Annular Fluidized Bed Reactor with Solids Circulation. Berrruti Franco, Bi Xiaotao, and Pugsley Tod. 2007 ECI Conference on The 12th International Conference on Fluidization - New Horizons in Fluidization Engineering. 2007. Vancouver, Canada. - Ref Type: Conference Proceeding

[10] Noorman S., van Sint Annaland M., Kuipers H., Packed Bed Reactor Technology for Chemical-Looping Combustion, Ind. Eng. Chem. Res., 46 (2009) 4212-4220.

[11] Ryu H.J., Park Y.C, Jo S.H. and Park M.H., Development of novel two interconnected fluidized bed system, Korean Journal of Chemical Engineering, 25 (2008) 1178-1183

[12] Noorman S. Packed Bed Reactor Technology for Chemical Looping Combustion. International Oxy-Combustion Research Network. 2007. 25-1-2007. Ref Type: Conference Proceeding

[13] Dahl Ivar M., Bakken Egil, Larring Yngve, Spjelkavik Aud I., Hakonsen Silje Fosse, Blom Richard, On the development of novel reactor concepts for chemical looping combustion, Energy Procedia, 1 (2009) 1513-1519.

[14] M. Ishida, M. Yamamoto and T. Ohba, Experimental results of chemical-looping combustion with NiO/NiAl2O4 particle circulation at 1200 C, Energ Convers Manage 43 (2002), pp. 1469-1478

[15] M. Yazdanpanah, A. Hoteit, A. Forret, A.Delebarre and Th.Gauthier, A novel chemical looping combustion process : pressure drop and solid circulation rate modeling, Proceedings of Fluidisation XIII conference, Edited by S.D.Kim, Y.Kang, J.K.Lee and Y.C.Seo, Engineering Conference International, 2010, pp 359-366

[16] T. Mattisson, M. Johansson and A. Lyngfelt, The use of NiO as an oxygen carrier in chemical-looping combustion, Fuel, 85 (2006) 736-747