Bolhàr-Nordenkampf, J., Pröll, T., Kolbitsch, P., Hofbauer, H.

, 2009, in: Proceedings of the 4th European Combustion Meeting, Vienna, Austria, 2009. Concept Study for a 10MWth Chemical Looping Combustor for Power Generation
Johannes Bolhàr-Nordenkampf , Tobias Pröll, Philipp Kolbitsch, Hermann Hofbauer
Vienna University of Technology, Institute of Chemical Engineering Getreidemarkt 9, 1060 Vienna, Austria

Abstract A semi-commercial 10 MWth chemical looping combustion (CLC) plant for power production is proposed as a next scale demonstration plant. The design criteria for the CLC boiler are derived from the experience of the 120 kW CLC pilot rig at Vienna University of Technology. IPSEpro simulation environment is chosen for implementation of the process flow sheet of the CLC power plant. A single pressure steam cycle is suggested to suit the demonstration plant scale. Heat exchangers and a 5 stage steam turbine are arranged. By simulation, design parameters of the power plant are derived and discussed. Introduction To face the problem of increasing carbon dioxide concentration in the atmosphere the key mid-term strategy is carbon capture and storage [1]. One promising novel technology for carbon capture is chemical looping combustion (CLC). Chemical looping systems consist of two reaction zones in which different gas streams are in contact with circulating solids. The circulating solids are metal oxide particles which transport oxygen from air reactor (AR) to fuel reactor (FR), thus the fuel reactor exhaust gas consist of CO2 and H2O (see Fig. 1). After water condensation a highly concentrated CO2 stream ready for sequestration is obtained.
N2, O2 CO2, H2O

plant can be found in Kolbitsch et al. [8] and operation experience with different bed materials can be found in Pröll et al. [9], Kolbitsch et al. [10] and BolharNordenkampf et al. [11]. The next step in process development is to demonstrate this technology in an industrial size application. First basic design parameters of a CLC boiler in the scale of 10 MWth were calculated by Lyngfelt et al. [12] in 2001. The design criteria were chosen for an atmospheric CFB boiler with a bubbling bed as FR in the return loop of the CFB. The assumed boiler concept in the paper by Lyngfelt et al. [12] for demonstration of the CLC technology with gaseous fuel is used for heat production, district heating or production of industrial process steam. This paper gives basic design parameters for a 10 MWth CLC boiler for power generation based on the DCFB design of the 120 kW pilot rig. The DCFB boiler The chemical looping boiler assessed in the present work is a dual circulating fluidized bed (DCFB) system for gaseous fuels.
exhaust AR exhaust FR

MeO Air reactor Fuel reactor

Me

Air

Fuel
LS2

This technology was first introduced in 1954 [2] later in 1983 this process was presented by Richter and Knoche [3] and Ishida et al. [4] to increase the reversibility of combustion processes. In 2004 first reports from a 10 kW CLC unit operated with gaseous fuels at Chalmers University in Sweden [5], [6] were presented and later in 2004 Ryu et al. [7] reported from a 50 kW CLC unit from the Korean Institute of Energy Research. In January 2008 a 120 kW CLC pilot plant was put in operation at Vienna University of Technology. A detailed description of the

steam

air LS1 steam fuel/ steam steam

Fig. 2 DCFB boiler system

Corresponding author: johannes.bolhar-nordenkampf@tuwien.ac.at Proceedings of the European Combustion Meeting 2009 1

fuel reactor (FR)

air reactor (AR)

Fig. 1 Chemical looping combustion principle

Most oxides have to be supported by other inert materials to gain the necessary mechanical strength and attrition stability to be operated in a CFB. Loop seals between the reactors fluidized with superheated steam avoid mixing of AR and FR gases. 16. Flow sheet simulation for chemical looping processes with IPSEpro Due to the relative novelty of the CLC process. 3. It was designed in a modular structure. 18. For hydrocarbon fuels a high catalytic activity is beneficial (especially for methane conversion). Table 2 Oxygen carrier characterization Item Value oxygen carrier system Ni/NiO support materials Al2O3 + MgO NiO content 41. [21] presents the so-called advanced energy technology library (AETLib). The flow regime in the AR is fast fluidization and in the FR turbulent fluidization [8]. gas and solids are separated in cyclone separators. The important values are summarized in Table 1. Mn and Cd [13. A more detailed description on the IPSEpro Simulation Software and on the implemented CLC model can be found elsewhere [21]. Other possible metals besides Ni are: Cu. The basic setup can be seen in the IPSEpro flow sheet shown in Fig. The PSE is the main component of IPSEpro. data from the PSE (settings. TiO2 or yttriastabilized zirconium (YSZ) [12]. α-Al2O3 and MgO. equation orientated simulation software. 14. At this scale a simple heat recovery setup is proposed. because there the process is modeled and data is entered/read. More information on the OCs used can be found in the article by [20]. enables the user to edit the source code of the standard model as well as to create new models for special tasks. Table 1 Chosen or assumed parameter values Item Value Power (fuel) ~40MJ/kg 10 heat loss (reactor system) 2 AR cross section 1. The nominal power of the boiler is 10 MW with natural gas as the fuel. The basic assumption for the boiler parameters is based on the experience on the 120 kW pilot rig. Fe.). Bolhar-Nordenkampf et al.The boiler design is based on the Vienna 120 kW pilot rig which was designed with a direct focus on scalability to larger size. post combustor and filter) was used for parameter evaluation by means of reconciled calculation [21]. 17.75 FR temperature 900 FR pressure drop 14. The oxygen carrier characteristics are summarized in Table 2. In the present study. A flow sheet simulation of the whole pilot plant setup (reactor. The CLC model implemented in the IPSEpro environment was validated by empirical measurements of the 120 kW pilot rig. introduce new substances classes. The main structure consist of the user interface (process simulation environment.6 AR pressure drop 7 FR cross section 0. The model development kit. the model library and the equation solver (Kernel). highly active carriers Ni particles are used. The general setup is shown in Fig. Such support materials can be Al2O3. The OC is based on NiO.3 dp 90-210 Unit wt% µm 2 . PSE). (e. a user interface to change the model library. The steam cycle parameters (live steam parameters. The simulation software IPSEpro IPSEpro is a stationary. parameters) is sent to the equation solver (Kernel).g. etc. most commercial tools do not yet offer a possibility for treatment of solid streams or of gas-solid reactions as needed for CLC. The solver calls in data from a model library that contains all information about the inner structure of the apparatuses used in the process. The properties of 40 solid substances potentially involved in CLC/CLR processes have been collected from databases and implemented into the IPSEpro structure. In this paper the process model is used for predictive simulation of chemical looping combustion for power production.5 air/fuel ratio 1. Unit MW % m2 kPa m2 °C kPa % % Oxygen carrier characterization The basic requirements for oxygen carriers (OCs) are mechanical stability. The software IPSEpro offers an open structure for implementation of a new model library specially designed for chemical looping based processes. Co. Depending on the used fuel the OC has to fulfill other requirements too. gas cooler. 2. During the calculation. 19]. Ni-based carriers have a good catalytic activity and are suitable for methane combustion and reforming. lowest possible costs and high oxygen transport capacity. 15. The particle size of the OC is in the range of 90 210 µm. which is an extended version of a comprehensive model library previously built for gasification-based processes [22]. efficiencies) can be found in Table 3. Such a boiler is suitable for semi-commercial demonstration of CLC technology for gaseous fuels. Downstream of each reactor.1 solid conversion difference 10 mean AR oxidation state 50 The values of Table 1 and Table 2 together with the CLC reactor model implemented in the simulation software IPSEpro give the basis for further calculation.

4 97. The proposed AR heat recovery boiler consists of a steam generator. In order to reduce the dew point to 25°C. The heat of the FR exhaust gas is withdrawn by a steam generator. A 10 MWth CLC boiler with heat recovery and a single pressure steam cycle is successfully simulated. some cooling water is required. 3 represents the entire CLC plant without CO2 compression. A single pressure steam cycle in natural circulation with no reheat and live steam parameters of 520°C at 60bar(a) is designed. 3 Basic setup of a chemical looping combustor for power production gas whereas the fuel is available at elevated pressure and ambient temperature. an economizer and an air preheater.1 89. the produced heat in the reactors is larger than the sensible heat in the exhaust gas streams. Some of the heat recovered in a FR exhaust gas condenser is used to preheat low temperature feed water. i. CO2 and H2O). The steam turbine has 5 stages with bled steam connection for feed-water heating and deaeration of the 3 Results and Discussion It is demonstrated that the IPSEpro environment is suitable for flow sheet simulation of CLC processes. Air is provided at ambient pressure and is preheated with heat from the AR exhaust condensed water. This FB-HE consists of two parts. The IPSEpro simulation covers the CLC boiler. Therefore it seems feasible to recover heat from the AR exhaust stream down to 60°C. The components of the AR exhaust gas are N2 and unreacted O2. however. only a gas compressor for air is needed. The fuel is taken from the natural gas grid and mainly consists of methane. condensed water and used bed material are the system boundaries of the plant. i. The input streams fuel. Different basic design parameters are derived from the flow sheet simulation and presented. FR exhaust (i. the heat recovery steam generator and the integrated steam cycle with a steam turbine for power generation. the risk of corrosion is very low.e.Table 3 Steam cycle parameters Item Steam temperature Steam turbine inlet pressure Condenser pressure total steam turbine efficiency generator efficiency pump efficiency motor drive efficiency Cooling water inlet temperature Cooling water outlet temperature Value 520 60 0. a super heater and a steam generator. Therefore a FB-HE has to be arranged.e. The assumed parameters are conservative and CLC reactor parameters are based on the experience of the 120 kW pilot plant [9.e. 11]. air and fresh bed material as well as the output streams AR exhaust. 10. The setup of the heat recovery steam generator includes the AR and FR heat recovery boiler and a fluidized bed heat exchanger (FB-HE). drum bed material fresh stgen sh stgen stgen AR FR eco aph air natural gas fwph turbine G depleted air cond pump pump CO2 to compression water Fig. The FB-HE is arranged in the back loop of the AR. The bed material holdup in the two reactors . At the operating parameters chosen (see Table 1).5 70 90 15 25 Unit °C bar(a) bar(a) % % % % °C °C The process flow diagram in Fig.

3 Unit Nm³/s Nm³/s m/s m/s kg kg kg/(m²s) kg/s kg/s kW kW % The Sankey diagram in Fig.3 710 1140 57. i. CLC features the principle of a chemical heat pump [3]. however.is calculated. The overall net electric efficiency of the 10 MWth CLC power plant is 36. the following two set points were determined.1 85. According to Smolders and Baeyens [24]. in other words. The main results of the process flow sheet simulation are summarized in Table 4.1 kg/m²s corresponds to the solid conversion difference of 10% and is in the typical range of CFB risers. This means. The turbulent regime in the FR enhances gas-solid contact and thus allows a reduction of the solids inventory [23]. the energy flow from the FR to AR consists only of chemical energy. Secondly.87 4. The energy consumption of all auxiliary units is in the range of 1% of the thermal input. then transport it to a second reactor where the chemical energy is released as heat at a higher temperature level.8 7. The solid circulation rate Gs of 57. The basic principle of a chemical heat pump is to bind chemical energy at a lower temperature to the bed material. The necessary mass in the FR per MW thermal input decreases from 510 kg/MW to 110 kg/MW [8].28 2. The AR has a solid inventory of 1140 kg and is in the same range as in the concept of Lyngfelt et al. 4 Energy flows in the 10 MWth CLC boiler 4 . the sensible heat of the bed material stream at 900°C is set to 0. GS can reach values up to 100 kg/m²s in CFB systems.e. Firstly. This sensible heat is partly contained in the exhaust gas of the reactor and partly in the returning bed material. Fig. With respect to the semi-commercial scale of the plant this efficiency value may be acceptable.3%.6 84. more work is needed to assess the potential.e. This is one of the advantages of a turbulent FR concept especially relevant at increased plant capacities. It is important to notice that the bed mass in the loop seal and the FB-HE is not considered. Table 4 Parameter values derived assumptions Item Fuel flow Air flow FR superficial velocity AR superficial velocity FR bed mass AR bed mass solid circulation rate GS solid flow from AR solid flow to AR air fan power pump power net electric efficiency from simulation and the basic Value 0. 4 shows this principle for the 10 MWth CLC boiler. i. that the sensible heat entering the fuel reactor is the difference in sensible heat between the solids stream entering and leaving the FR. the chemical energy in this energy stream is set to 0. For clear illustration of the energy transported by the circulating solids. [12]. the returning bed material to the fuel reactor consists only of sensible heat released in the exothermic AR. The total energy transported by the gas streams is based on specific sensible enthalpy and lower heating value.8 61 49 36.

F. Design of a chemical looping combustor using a dual circulating fluidized bed (DCFB) reactor system. Ryu. Energy 12 (2) (1987) 147–154. Ch. J. Adanez. T. A five-stage steam turbine is designed with conservative efficiency estimation. Operating experience with chemical looping combustion in a 120 kW dual circulating fluidized bed (DCFB) unit. 2005. P. International Energy Agency (IEA). K. Gayan. Abad. Lewis. Demonstration of inherent CO2 separation and no NOx emission in a 50 kW chemical-looping combustor: Continuous reduction and oxidation experiment. Lyngfelt.. For the conversion of the fuel some additional heat is needed as the reaction is slightly endothermic (Ni-based particles). Canada. Leckner. The major part is bound to the bed material. [11] J. accepted for publication in Energy Procedia. Elsevier Science. [8] P.). T. The GRACE project: Development of oxygen carrier particles for chemical-looping combustion. Jin. velocities. Pröll. Hofbauer. the chemically bound energy in the solids stream to the AR is only the difference to the chemically bound energy present in the returning solids stream. Bolhar-Nordenkampf. Natural minerals as oxygen carriers for chemical looping combustion in a dual circulating fluidized bed system. Design and operation of a 10 kW chemical-looping combustor. J. Lyngfelt. Vancouver. T. Celaya. Canada. F. Hofbauer.972 (1954). Kronberger. Performance of a NiO-based oxygen carrier for chemical looping combustion and reforming in a 120 kW unit. The smaller part is released with the exhaust gas stream. A single pressure steam cycle with live steam parameters of 520°C and 60 bar is suggested to suit the demonstration plant scale. Yi. J. Mayer. K. 019800). G. K. J. H. On the basis of successful implementation of the required unit operations. Adanez. To fulfill the energy balance of the system. 5 . Construction and 100h of operational experience of a 10-kW chemical looping combustor. accepted for publication in Energy Procedia. T. Hofbauer. Morin. Bolhar-Nordenkampf. Chemical Engineering Science 56 (10) (2001) 3101–3113. [10] P. [2] W. J. F. H. P. H. [12] A. Chemical Engineering Science 62 (1-2) (2007) 533–549. Chemical Engineering and Technology 32 (3) (2009) 398–403. volume flows.665. T. Lyngfelt.3% is feasible for the scale of the plant. Pröll. Kolbitsch. a first basic design for a semi-commercial 10 MWth CLC plant for power production is proposed. Heat exchangers are arranged to recover heat from the bed material and exhaust gas streams. Birol. 2004. 31. Pröll. in: 7th Conference on Greenhouse Gas Control Technologies (GHGT7). Garcia-Labiano. in: Carbon Dioxide Capture for Storage in Deep Geologic Formations. Evaluation of a chemical-looping-combustion power-generation system by graphic exergy analysis. J. J. The simulation provides additional design parameters (mass flow. [5] A. ACS Symposium Series 235 (1983) 71–85. Lyngfelt. Mattisson. Amsterdam. Mattisson. Fe-. [9] T. C. Hurst. Knoche. in: 7th Conference on Greenhouse Gas Control Technologies. 2004.-X. coordinated by Chalmers University of Technology. Kolbitsch. pp. B. In the FR the energy input is bound primarily to the outgoing bed material as chemical energy. T. P..S. Pröll. and Ni-based oxygen carriers in chemical-looping combustion. Ishida.g.. Production of pure carbon dioxide. Patent Office. [6] A. 2006. World Energy Outlook 2006. The bed material transports the chemical energy to the AR where in interaction with the preheated air it is transformed to sensible heat at a higher temperature level (e. de Diego. The CLC boiler represents the DCFB concept of the 120 kW pilot rig. Mapping of the range of operational conditions for Cu-. H. 980°C). However. This heat is than released in the exhaust gas stream and in the returning bed material stream. The IPSEpro simulation environment is chosen for implementation of the CLC reactor model which was validated by empirical measurements. The main energy input in the FR is the natural gas. pp. Vancouver. Akehata. Gilliland. Kolbitsch. The design criteria for the CLC boiler are based on the experience from a 120 kW CLC pilot rig operated since January 2008. Richter. Reversibility of combustion processes. more work is needed to assess the potential of such a concept before any final conclusion can be made. This heat is transported into the FR by returning bed material from the AR. heat has to be withdrawn by a FB-HE out of the bed material. E. J. etc. The project was also part of Phase II of CCP (CO2 Capture Project) through Shell. Bolhar-Nordenkampf. [7] H. Acknowledgements The model library for CLC processes was developed in the EU financed project CLC GAS POWER (FP6 Contract No. [3] H.Accordingly. Hofbauer. D. Thunman. [4] M. Number 2. L. P. application of chemical-looping combustion. accepted for publication in Energy Procedia. Bolhar-Nordenkampf. Kolbitsch. A fluidizedbed combustion process with inherent CO2 separation. –. Zheng. Conclusions Based on a continuous development of the CLC technology from laboratory scale to pilot scale started in 2000. B. References [1] F. 625–645. [13] A. H. The net electric efficiency of 36. R. U. the process flow sheet of a whole power plant has been set up. The loop is closed by the returning energy flow to the FR. A.

Industrial & Engineering Chemistry Research 45 (17) (2006) 5911–5919. Effect of pressure on the behavior of copper-.. Baeyens. Pröll. Chemical Engineering and Technology 32 (3) (2009) 410– 417. Industrial & Engineering Chemistry Research 45 (8) (2006) 2689–2696.. Bolhar-Nordenkampf. Mattisson. Garcia-Labiano. L. Pröll. Fuel 86 (7-8) (2007) 1021–1035. Johansson. S. [17] S. Fuel 85 (5-6) (2006) 736–747. Smolders. Grace. D.. E. A. T. Hofbauer. Lyngfelt. Snijkers. and nickel-based oxygen carriers for chemical-looping combustion. Hofbauer. P. T.. J. Jerndal. F. [15] M. J. F. H. M. Chemical Engineering Science 45 (8) (1990) 1953–1966. accepted for publication in Energy Procedia. M. I. Powder Technology 119 (2-3) (2001) 269–291. Mattisson. A. Mattisson. [19] B. Adanez. Son. M. T. Johansson. Development and application of a simulation tool for biomass gasification based processes.com/ijcre/vol6/A89 J. Kolbitsch.. R. J. Palacios. T. A. K. Johansson. M. 6 . T.. de Diego. The use of iron oxide as oxygen carrier in a chemical-looping reactor. Fuel 86 (1-2) (2007) 113–122. J. Energy & Fuels 20 (1) (2006) 26–33. Gas fluidized beds operating at high velocities: a critical review of occurring regimes. Mattisson. Titania-supported iron oxide as oxygen carrier for chemical-looping [20] [21] [22] [23] [24] combustion of methane. http://www. NiO particles with Ca and Mg based additives produced by spray-drying as oxygen carriers for chemical-looping combustion. Abad. Lyngfelt. iron-.[14] F. [18] A. Gayan. High-velocity fluidized bed reactors. International Journal of Chemical Reactor Engineering 6 (2008) –. H. [16] T. Abad. A. Corbella. Lyngfelt. Use of NiO/NiAl2O4 particles in a 10 kW chemicallooping combustor.bepress. A. P. R. Thijs. Chemical-looping combustion with NiO and Fe2O3 in a thermobalance and circulating fluidized bed reactor with double loops. The use of NiO as an oxygen carrier in chemical-looping combustion. Comprehensive modeling tool for chemical looping bas processes. Kim. Lyngfelt.

Sign up to vote on this title
UsefulNot useful