CC BY-NC-ND
0Available online at www.sciencedirect.com
SciVerse ScienceDirect
Energy Procedía 37 (2013) 7091 - 7099
GHGT-11
The Calcium looping process for low CO2 emission cement
and power
Matteo C. Romanoa*, Maurizio Spinellia, Stefano Campanaria, Stefano Consonnia, Giovanni Cintib, Maurizio Marchib, Enrico Borgarellob
aPolitecnico di Milano, Energy Department, via Lambruschini 4, 20156 Milano, Italy _bC.T.G. - Italcementi Group, via Camozzi 124, 24121 Bergamo, Italy_
Abstract
Calcium looping appears as one of the most promising technologies for CO2 capture in short-medium term plants featuring the combustion of fossil fuels. Ca-looping (CaL) is a regenerative process which takes advantage of the capacity of Calcium Oxide-based sorbents in capturing the CO2 from combustion gases by means of sequential carbonation-calcination cycles. CaL technology appears very promising for CO2 capture from cement plants, since the CaO-rich purge stream which must be extracted from the process can be a valuable raw material for clinker production.
The aim of this study is to investigate from the technical and economic side the benefits arising from the integration between a coal-fired power plant with CaL process for CO2 capture and a cement plant using the CaL purge to substitute part of the raw meal. The main parameters affecting the CaL process are varied and the effects on both the plant performance and the final cost of clinker and electricityare discussed.
© 2013 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of GHGT
Keywords: Calcium looping, carbonator, cement, CO2 capture, techno-economic analysis
1. Introduction
Calcium looping is one of the most promising technologies for CO2 capture in future short-medium term plants featuring the combustion of fossil fuels. Ca-looping (CaL) is a regenerative process which takes advantage of the capacity of Calcium Oxide-based sorbents in capturing the CO2 from combustion
* Corresponding author. Tel.: +39-02-23993846; fax: +39-02-23993913. E-mail address: matteo.romano@polimi.it.
1876-6102 © 2013 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of GHGT doi: 10.1016/j.egypro.2013.06.645
gases by means of sequential carbonation-calcination cycles. This technology is now at a relatively advanced stage of development with the first MW-scale pilot installations [1,2]. Coal-fired power plants are considered as the main application for CaL process, for which interesting thermodynamic and economic advantages have been reported in the literature [3-5].
In addition to power plants, CaL technology appears very promising for CO2 capture from industrial sources [6], especially in the cement industry. Cement production is the largest industrial source of carbon dioxide emissions, responsible for about 7% of the total CO2 emission from large stationary sources [7]. In cement plants, CO2 emissions arise from both the calcination of the CaCO3 in the raw meal feed (about 60% of the total) and from fuel combustion, which is needed to provide heat for CaCO3 calcination and to heat the raw materials up to high temperatures to initiate the clinker formation reactions. CaL process can be integrated in stand-alone cement plants, removing the CO2 from the flue gases as proposed for example in [8]. Another promising option for emission reduction is to use the CaO-rich purge stream that must be extracted from the Ca-looping process of a power plant as a feed stream of the cement plant, assessed for example in [9]. In this way CO2 emissions can be strongly reduced with very limited modifications to the cement plant. The aim of this study is to investigate this second option, assessing the techno-economic aspects of the possible synergy between a power plant adopting the Ca-looping process and a cement plant.
Nomenclature
ASU Air separation unit
CFB Circulating fluidized bed
FCO2 Mole flow of CO2 in flue gas entering the carbonator [kmol/s]
F0 Mole flow of fresh makeup limestone [kmol/s]
Fr Mole flow of CaO cycling from the calciner to the carbonator [kmol/s]
SR Substitution rate of limestone with CaO as feed of the cement plant
Ws Solids inventory in the carbonator [kg]
2. Power plant with Ca-looping process
The proposed power plant is a greenfield plant similar to the one described in [3] and schematically shown in Fig. 1. It consists of a number of units: an ultra-supercritical Circulating Fluidized Bed (CFB) air-fired boiler, the CaL-process reactors, an oxygen separation unit, a CO2 purification island and an advanced steam cycle. The CFB boiler, which allows removing SO2 at high temperature inside the furnace, is particularly appropriate to this application because the SO2 in flue gases would otherwise react in the carbonator, reducing the activity of the sorbent towards CO2 capture. The fluidized bed operates at 850°C and uses part of the Ca-rich purge from the CaL unit as SO2 sorbent.
Flue gases exiting the economizer of the boiler at 350°C are sent to the carbonator by a blower. The carbonator operates at 650°C, cooled by steam superheaters and reheaters. The CO2-lean flue gases exiting the carbonator are sent to further heat recovery steps and a Ljungstrom air preheater before being emitted to the stack at 130°C. CO2 capture efficiency mainly depends on three parameters: the make-up of fresh sorbent (F0/FCO2), solid recycle rate between the reactors (FR/FCO2) and solid inventory (Ws).
Fig. 1. Layout of the power plant and the cement plant assessed.
The carbonator model proposed in [10] has been utilized here to determine the CO2 capture efficiency as a function of these parameters. A compact reactor design has been considered, operating with a superficial gas velocity of 10 m/s and 20 m height. The solid inventory was fixed at 150 kg/(mgas3/s), corresponding to about 1500 kg per m2 of reactor cross section.
The calciner operates at 950°C and is assumed to be entirely refractory lined. The use of a low sulfur and low ash coal as power plant fuel allows avoiding excessive ash and CaSO4 buildup in the CaL loop. The CO2-rich exhausts leaving the calciner are cooled down to 350°C by generating superheated and reheated steam. A fraction of these gases is filtered in a hot ESP and recycled to the calciner as fluidizing gas and temperature moderator. The recycle flow rate is determined by imposing an oxygen concentration of 50% in the oxidant. Given the extremely high Ca/S ratio, it is assumed that SO2 removal is 100% in both the CaL reactors.
The power island is based on a state-of-the-art USC steam cycle, with single reheat and live steam conditions of 290/60 bar and 600/620°C [11]. Boiler feedwater is preheated up to 306°C in 9 regenerative heat exchangers. To take advantage of available low temperature heat (recovered from calciner gas cooling, CO2 intercooled compression and ASU compressors intercoolers), part of the water is heated outside the regeneration loop. The steam turbine expansion has been calculated with a stage-by-stage model, which estimates the efficiency of each turbine stage as a function of its specific speed, size parameter and moisture content in LP stages [12].
The oxygen used in the calciner is produced by a cryogenic Air Separation Unit (ASU) with the advanced layout proposed in [13], featuring a low pressure column with two reboilers, coupled with the condensers of two higher pressure columns. According to [13], such configuration allows generating O2 with 95% molar purity at a net electric consumption of 200 kWh per ton of pure O2.
The cooled CO2-rich stream from the calciner needs to be purified from the incondensable gases to reach the purity specifications required for transport and storage. Incondensable gases derive from nitrogen in the fuel burned in the calciner, from N2 and Ar contained in the 95% purity oxygen produced in the ASU and from air in-leakages. As for CO2 purification, the cryogenic purification process described in [14] was considered, which gives 97% pure CO2 at 150 bar.
The in-house code GS [15] has been used to calculate mass and energy balances of most of the units of the power plant. Manual iterations between GS, Aspen Plus [16] (used to calculate the CO2 compression and purification unit) and Matlab [17] (used to compute the carbonator efficiency) were made until convergence of the simulation.
3. Integration with cement plant
3.1. Reference cement plant
The reference cement plant produces 4050 tpd of clinker and is based on a state-of-the-art dry process, as shown in Fig. 1. The layout is representative of a large scale plant, including raw meal preheating, a pre-calciner, a rotary kiln and a clinker cooler. A 4.7% sulfur, 1.3% ash, 34.2 MJLHV/kg pet coke has been considered as fuel.
The raw meal, containing the raw materials for clinker production (namely CaCO3, SiO2, Al2O3, Fe2O3 and other minor compounds), is fed to the preheater after milling and drying. Preheating is carried out in a 5-stage preheater, where solids and gas are sequentially contacted in risers and separated in cyclones. While gas and solids flow co-currently in each riser, the preheating process is basically counter-current, since the solids entering at the top of the system descend through the preheater being heated up, while the hot gas stream flows upwards and releases heat to the solids.
Preheated solids enter the pre-calciner, where most of the limestone is decomposed at high temperature (860°C) into CaO and CO2. Heat for the endothermic calcination reaction is provided by the combustion of 60-65% of the fuel consumed by the plant. In the pre-calciner, most of the CO2 is also generated, due to fuel combustion and, mainly, due to calcination. As a matter of fact, about 60% of total CO2 emissions are associated to limestone decomposition.
The hot pre-calcined solids then enter the rotary kiln, where they are further heated to initiate the slightly exothermic reactions which lead to the formation of the clinker phases at 1400-1450°C. The kiln slope and its rotation around the axis allow the solids moving towards the hot end of the kiln, where the remaining fuel (35-40% of the thermal input) is burned to heat the solid materials. The combustion gases and some CO2 originating from residual calcination exit the kiln from the cold end, enter the pre-calciner and participate to raw meal heating. The hot and partially melted clinker exits from the hot side of the rotary kiln and it is finally sent to the clinker cooler, where it is cooled down quickly, in order to improve the properties of the final product. Ambient air is used here as cooling media; most of this air reaches very high temperatures (900-1000°C) and is later used as oxidant in the kiln and the pre-calciner burners. A smaller flow rate, heated to lower temperatures, is used for fuel and raw meal drying. This efficient heat recovery allows increasing the thermal efficiency of the whole process.
3.2. Integration: use of CaL process purge as feed for cement plant
Replacing part of the cement plant raw meal with the purge from a CaL process generates beneficial reductions of fuel consumption and CO2 emission. The positive effects are basically related to the partial substitution of CaCO3 with CaO in the raw material, which will not be decomposed generating CO2 and will not need any heat input for calcination. The level of integration between the power plant and the cement plant can hence be quantified with an index named Substitution Rate (SR), defined as the molar ratio between the moles of CaO fed to the plant with the CaL purge and the total moles of Ca fed to the plant.
A number of cases with SR between 20% and 100% have been simulated. The fuel to be burned in the kiln and the pre-calciner was estimated so to maintain the same gas and solids temperatures at the kiln
ends and at the pre-calciner outlet. Despite largely in excess with respect to the stoichiometric air for combustion in the pre-calciner, especially at high SR, the hot air stream available from the clinker cooler is used as tertiary air in the pre-calciner and the suspension preheater, in order to maximize the thermal efficiency of the process.
Our estimates indicate that complete substitution (SR=100%) gives a reduction of up to 70-75% of the thermal input and up to 85% of CO2 emissions, with respect to the reference plant. For lower SR, a nearly linear trend has been obtained. Hence, the potential of primary energy and CO2 emission savings from the integration between a cement plant and a CaL power plant are very relevant. In addition, limited variations have been calculated for the temperature profile and the gas flow rate in the preheater, remaining in the range of variability typical of part load operations; therefore, the integration discussed here can be carried out simply by retrofitting existing cement plants, without the need for significant modifications of the existing equipment.
The purge from a real CaL process contains relevant amounts of species other than CaO like sulfur and ash from the fuel burned in the calciner; in an integrated plant, such species eventually end up in the clinker. Given that the composition of the clinker must be kept within a rather narrow range, the Substitution Rate may have to be limited to values significantly lower than 100%. As a general rule, the higher the make-up (F0/FCO2) and the lower the recycle rate (FR/FCO2), the purer the purge stream and the higher the allowed maximum Substitution Rate. Also the composition of the fuel used in the CaL calciner has an important effect on the purge purity. For the plants assessed here fed with low sulfur and low ash coal (where ash is 67% SiO2 and 33% Al2O3), CaSO4 always results as the limiting species.
4. Benefits generated by the integration
CO2 emissions resulting from the simulation of the integrated system with different make-up flows in the CaL process and with the maximum substitution rate SRmax allowed by the purge composition are shown in Fig. 2 for a specific plant featuring specific recycle rates FR/FCO2=6. Calculations have been carried out for systems where the reference cement plant with capacity 4050 tpd of clinker is integrated with a power plant giving the purge flow required for a maximum Substitution Rate allowed by the composition of the CaL purge. The resulting size of the power plants range between 407 and 664 MWe, increasing with Fo/FCO2 due to the higher SR allowed, reaching at F0/FCO2=0.25 an integration level of 97%. For this case, 12.9 kg/s are emitted in total from the two plants, 93% less than the reference plants without CO2 capture. For the intermediate case with F0/FCO2 = 0.1, an overall emission reduction of 75% has been calculated, due to the lower allowed substitution rate of 31%. The emissions avoided in the cement plant are however as large as the residual emissions from the power plant. Therefore, this limited level of integration is essentially equivalent to a system composed by a cement plant without CO2 capture and a zero-emission 550 MWe power plant.
In order to quantify the benefits which can be obtained by integrating the two plants, the following equivalent electric efficiency and equivalent CO2 specific emission from power generation have been defined:
Vel,eq
QlHV.tot ~ ™clk ' Rclk.ref — otnn ^C02,tot ~ mclk ' Ec02,clk,ref
EC02,el,eq ~ 3600 ■
where QLHVltot and Ec02tot are total heat input (in MWLHV) and the total CO2 emission (in kg/s) of the complete integrated system, mclk is the mass flow rate of clinker produced by the cement plant (47 kg/s) and qcit,ref and Ec02 clk re^ are the specific heat consumption (3230 kJ/kgclk) and the specific CO2 emission (855 g/kgclk) of the reference cement plant without CO2 capture. In this way, the power plant performance indexes account for credits from the reduced fuel consumption and the reduced CO2 emission in the cement plant.
■ Emitted C02, cement plant Avoided C02, cement plant
■ Emitted C02, power plant I Avoided C02, power plant
60 40 20 0 20 40 60 80 100 120 140 160 C02 emissions, kg/s
Fig. 2. CO2 emission from the complete power and cement plant system, for for FR/FCo2=6 and different F0/FCo2.
Calculation results are reported in Fig. 3, where the equivalent performance indexes are compared with the corresponding indexes for a stand-alone power plant. Credits deriving from the integration with the cement plant lead to higher equivalent efficiencies (up to 2% points) and lower specific emission, which may become negative when the emissions avoided in the cement plant exceed the actual emissions of the power plant. In general higher benefits are obtained with higher make-up flows thanks to the higher level of integration with the cement plant.
4.1. Economic analysis
The cost of power plant equipment has been estimated with exponential method functions [18], regressed against data reported in [19-22]. The capital cost of Ca-looping process reactors have been estimated according to Eq.3, as a function of both the reactors volume and the heat input in the calciner, where the scale factors have been assumed equal to 0.67 for the volume (SF, V) and 0.9 for the heat input (SF, Q). Based on data regression for oxyfuel CFBCs reported in [23], the parameter a, representing the relative weight of heat transfer surfaces on the total cost of a cooled CFB reactor, has been assumed equal to 0.85. Note that relevant heat transfer surfaces are needed to extract heat from the carbonator and from the flue gases exiting the carbonator and the calciner. However, the total thermal power extracted is
FD/FC V O2=0.25 64 MW„
FD/FC Pe=! 02=0.15 99 MWe
FD/FC Pe=! 02=0.10 51 MWe
FD/FC Pe=4 02=0.06 80 MWe
Fd/FC Pe=4 02=0.04 07MWe
basically proportional to the heat input in the calciner: for this reason, this quantity (QLHV,calc) has been used as scaling parameter.
(Qlhv,co.Ic\ , (Vcalc\SF'V , , (Vcarb\S
{-%-) +(1-a)i—) +(1-a)i—)
As far as the other assumptions for the economic analysis are concerned, the methodology reported in [11] has been used, which includes a cost of 7 €/tcO2 for transport and storage. Due to the different percentage of CO2 captured from the system, the cost of the electricity has been calculated by assuming different values of the carbon tax, instead of calculating the cost of the CO2 avoided, which can provide misleading results for low CO2 capture cases.
Two possible extreme approaches have been considered to estimate the cost of the electricity in an integrated layout, since the cost of the electricity will depend on the income from the selling of cement. In the first approach (COE-I), it is assumed that the selling price of cement increases by the same amount of the cost associated to the carbon tax in a cement plant without CO2 capture (i.e. for a carbon tax of 40 €/tCO2, the selling price of cement increases by 34.2 €/tclk). In the second approach (COE-II), it is assumed that the selling price of cement is unaffected by the carbon tax. As a result, the additional cost of the CO2 emitted from the stack of the cement plant is entirely charged on the cost of the electricity. Generally speaking, it is reasonable to expect that any real economic scenario would be intermediate between these two boundary situations.
0.0 0.1 0.2 0.3 0.4 0.5
Solids make-up, kgCaC03/kgc
Fig. 3 Equivalent efficiency and equivalent CO2 emission for different recycle rates and make-ups of the integrated process (light blue lines), compared with electric efficiency and CO2 emission of a stand-alone power plant (dark blue lines).
The results of this analysis are reported in Fig. 4, where the cost of electricity of the integrated cases is compared with the cost of electricity of the reference power plant without CO2 capture and with that of the stand-alone Ca-looping power plant, for a carbon tax of 40 €/tCO2. First, it can be noted that, differently from the non-integrated CaL cases, the cost of electricity decreases by increasing the make-up
flow in the integrated cases. This is due again to the increasing substitution rate, the decreasing CO2 emission from the cement plant stack and the lower operation cost due to the carbon tax. Second, in the most favorable economic scenario for the power plant, where the maximum increase of the clinker selling price is assumed, COE-I is always lower than in the case of the stand-alone CaL power plant. Conversely, in the other economic scenario COE-II is higher than that of the non-integrated case in most of the makeup range considered. For low make-ups, the cost of electricity is even higher than in the case without CO2 capture, since the cost of a large amount of CO2 emitted by the cement plant, operating with low SR, is charged to the electricity cost. However, it is worth underlying that for make-ups higher than 0.30.4 kgCaCO3/kgcoai, the cost of electricity is competitive with a reference plant without CO2 capture, even by keeping a clinker selling price unaffected by the carbon tax.
If the value of the carbon tax is reduced, the economic competitiveness of the low emission cases obviously decreases. For values around 27 €/tcO2 the minimum cost of electricity obtained for the first economic scenario (COE-I) is equal to the cost of electricity without CO2 capture. Thus, 27 €/tCO2 is the minimum cost of CO2 avoided expected for the system.
Integrated plants: COE-I
FR/FC02=4
- FR/FC02=6
— FR/FC02=10
reference power plant w/o C02 capture
0.2 0.4 0.6
Solids make-up, kgcacos/kgcoai
Fig. 4. Cost of the electricity for the different cases assessed, for a carbon tax of 40 €/tCO2
5. Conclusions
A techno-economic assessment of an integrated cement and power plant featuring the Ca-looping process for CO2 capture has been presented. The substitution of the cement plant raw meal with the CaO-rich purge from the CaL reactors would allow reducing fuel consumption and CO2 emission up to 75% and 85% respectively, without major modifications on the cement plant. By considering the maximum substitution rates of the cement plant raw meal, which may be limited by the presence of ash and CaSO4 in the purge, the equivalent electric efficiency and specific emissions have been calculated for different values of CaL process operating parameters, showing improved performance with respect to a stand-alone power plant, especially for high sorbent make-ups. The results of the economic analysis are also very
promising, showing economic competitiveness for carbon taxes higher than 27 €/tCO2, suggesting the interest for further investigations on this process.
References
[1] Sanchez-Biezma A, Ballesteros JC, Diaz L, De Zarraga E, Alvarez FJ, Lopez J et al. Postcombustion CO2 capture with CaO. Status of the technology and next steps towards large scale demonstration. Energy Procedia 2011 ;4:852-9.
[2] Galloy A, Strohle J, Epple B. Design and operation of a 1 MWth carbonate and chemical looping CCS test rig. VGB PowerTech 2011;91:64-8.
[3] Romano M. Coal-fired power plant with calcium oxide carbonation for postcombustion CO2 capture. Energy Procedia 2009;1:1099-106.
[4] Romeo LM, Lara Y, Lisbona P, Escosa JM. Optimizing make-up flow in a CO2 capture system using CaO. Chem Eng J 2009;147:252-8.
[5] Martinez I, Murillo R, Grasa G, Carlos Abanades J. Integration of a Ca looping system for CO2 capture in existing power plants. AIChE J 2011;57:2599-607.
[6] Romano MC, Anantharaman R, Arasto A, Ozcan DC, Ahn H, Dijkstra JW et al. Application of advanced technologies for CO2 capture from industrial sources. GHGT-11 conference, Kyoto, Japan 2012.
[7] IPCC. IPCC Special Report on Carbon Dioxide Capture and Storage. Prepared by Working Group III of the Intergovernmental Panel on Climate Change. : [Metz, B., O. Davidson, H. C. de Coninck, M. Loos, and L. A. Meyer (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2005.
[8] Rodriguez N, Murillo R, Abanades JC. CO2 capture from cement plants using oxyfired precalcination and/ or calcium looping. Environmental Science and Technology 2012;46:2460-6.
[9] Romeo LM, Catalina D, Lisbona P, Lara Y, Martinez A. Reduction of greenhouse gas emissions by integration of cement plants, power plants, and CO2 capture systems. Greenhouse Gases: Science and Technology 2011;1:72-8.
[10] Romano MC. Modeling the carbonator of a Ca-looping process for CO2 capture from power plant flue gas. Chemical Engineering Science 2012;69:257-69.
[11] EBTF. European best practice guidelines for assessment of CO2 capture technologies. www.energia.polimi it/news/D%204_9%20best%20practice%20guide.pdf
[12] Lozza G. Bottoming Steam Cycles for Combined Gas Steam Power Plants: A Theoretical Estimation of Steam Turbine Performance and Cycle Analysis. ASME Cogen-Turbo, 27-29th August 1990, New Orleans, LA 1990.
[13] IEA Greenhouse Gas R&D Programme. Oxy Combustion Processes for CO2 Capture from Power Plant. 2005;IEA report 2005/9.
[14] Chiesa P, Campanari S, Manzolini G. CO2 cryogenic separation from combined cycles integrated with molten carbonate fuel cells. Int J Hydrogen Energy 2011;36:10355-6.
[15] GS Software. www.gecos.polimi.it/software/gs.html.
[16] aspentech. Aspen Plus. ;version 7.3.
[17] The MathWorks Inc. MATLAB. 2009;version 7.8.0.
[18] Perry RH, Green DW. Perry's Chemical Engineers' Handbook. 8th ed. : McGraw-Hill, 2007.
[19] ALSTOM POWER INC. Greenhouse Gas Emissions Control By Oxygen Firing In Circulating Fluidized Bed Boilers: Phase 1 - A Preliminary Systems Evaluation. 2003;PPL REPORT NO. PPL-03-CT-09.
[20] DOE/NETL. Pulverized Coal Oxycombustion Power Plants. Volume 1: Bituminous Coal to Electricity. Final Report. 2008;DOE/NETL-2007/1291.
[21] DOE/NETL. Cost and Performance Baseline for Fossil Energy Plants - Volume 1: Bituminous Coal and Natural Gas to Electricity. Revision 2. 2010;DOE/NETL-2010/1397.
[22] DOE/NETL. Cost and Performance for Low-Rank Pulverized Coal Oxycombustion Energy Plants. Final Report. 2010;DOE/NETL-401/093010.
[23] Jantti T, Eriksson T, Hotta A, Hyppanen T, Nuortimo K. Circulating fluidized-bed technology - toward zero CO2 emissions. Power Gen Europe 2007.
