Design And Operation Optimisation Of A MEA-based CO2 .

Design and Operation Optimisation of a MEA-based CO2Capture UnitArtur Andradeartur.andrade@ist.utl.ptInstituto Superior Técnico, Lisbon, PortugalSupervisors: Prof. Dr. Carla Pinheiro; Dr. Javier RodriguezNovember 2014AbstractThe present thesis has the objective of analysing the cost reduction obtained through a rigorousmodel-based optimisation of a post-combustion CO2 capture plant for carbon capture and storageapplications. Presently, capture technology is mainly based on chemical absorption with alkanolamines.Even though this is a well-known technology, its application in power plants presents high costs, thus limitingits implementation.This way, a full scale capture plant model was developed considering MEA as the solvent. This modelis based on a conventional flowsheet and was implemented in gPROMS , using the gCCS libraries. Itcomprises an absorption section, in which CO2 is dissolved by reacting with the amine, and a regenerationsection where it is stripped from the solvent. After its validation, the model was optimised by modifying thedesign and operation parameters.A cost estimation model was applied to the plant model, in order to determine capital and operationalexpenditures. From the total cost obtained, 69% is due to the steam required in the regeneration section. Asfor the equipment cost, the absorber packing is the most relevant fraction.Considering typical values for the capture rate, CO2 purity and MEA concentration in the solvent asconstraints, the plant’s model optimisation led to a reduction of 15% in the specific total cost. Withoutimposing these typical values, the total cost was further reduced. These results clearly show the potential ofmodel-based optimisation in the reduction of the cost associated with CO 2 capture, thus contributing to itseffective implementation in power plants.Keywords: Carbon capture and storage, post-combustion capture, chemical absorption, MEA, costoptimisation, gPROMSCO2 obtained in power plants, in the preparationof natural gas, or in other chemical industries [2].1. IntroductionFor a power plant, the capture cost is theCO2 is a naturally-occurring gas with amaincomponentof the overall CCS costs, mainlymajor influence on Earth’s surface temperature.duetotheprocessenergy requirements. ThisHowever, due to human invention andimpliesanextraconsumptionof steam andindustrialisation, its concentration in theelectricity,leadingtoareductioninnet efficiencyatmosphere has greatly increased. According toandconsequentlyanincreaseintheelectricitythe 2013 IEA report [1], between the years ofcosts[2].2001 and 2011, the worldwide CO2 emissionsThe technology employed in the capturefrom fossil fuel combustion increased by 31%,processapplied in power plants depends on thereaching a value of 31 billion tonnes per year.characteristicsof the gas mixture being treated.From this value, approximately 42% are due toThesearerelatedto the kind of fuel andthe electricity and heat production sector.technologyusedintheproduction of electricity.According to the Global CCS Institute,Thisway,thecaptureprocessescan be dividedcarbon capture and storage (CCS) has thein:post-combustioncapture(PCC), prepotential to significantly reduce the CO2combustioncaptureandoxycombustion.emissions. The CCS process chain comprisesFrom the available technologies for CO2several technologies involved in removal ofcapture,absorption is at the present the only onecarbon dioxide from a gaseous stream, on and final sequestration in site awayNevertheless,adsorption,membranesandfrom the atmosphere. This can be applied to thecryogenic separation are currently being tested1

for application in the CCS chain [3]. Even thoughthe absorption technology has been applied inseveral other industries, due to the considerablecosts associated with the application of thecapture process to such high flow rates, itsapplication to the flue gas obtained in a largescale power plant fuelled by coal or natural gas iscurrently on a demonstration phase [4]. The onlyfull scale capture unit already operating is locatedon the Boundary Dam Power Station (Canada)and was designed for the capture of 1 MtCO2/year.Besides this operating capture unit, there areseveral others planned to start operating in thenear future, all of them based on CO2 capture byabsorption [5]. The improvement of the currentlyavailable processes should allow a reduction ofits costs by 20 to 30%, being the option with morepotential for the reduction of CO2 emissions in thenear future [2].A literature review was conducted in orderto understand how the absorption is applied toCO2 capture from a power plant, whichtechnologies are commercially available, andwhich are the current trends to reduce the capturecost.With the reduction of costs associated withthe CO2 capture as the main objective, aconventional capture plant model was built in thegPROMS ModelBuilder 3.7.1 environment usingthe recently developed gCCS libraries. After themodel validation, the optimisation features ofgPROMS were used in order to minimise thecosts associated with the unit’s design andoperation.2. Post-combustion CaptureTechnologyPost-combustion capture is applied inpower plants based on the combustion of a fossilfuel (coal, natural gas or oil), where a capture unitis used to remove the CO2 present in the flue gas.This flue gas is mainly composed of N2, CO2, H2Oand O2, and also of reduced amounts of SOx, NOxand ash [6]. The flue gas being treated have aconsiderably low CO2 content and is atatmospheric pressure, leading to a reduced CO2partial pressure. Therefore, the chosen solventhas to be able to ensure acceptable loadings andkinetics these conditions. Due to this, chemicalabsorption poses a better option in PCCprocesses, when compared with physicalsolvents [2].The processes by chemical absorption arebased on the CO2 chemical dissolution in analkaline solvent, through its selective reactionwith one or several of the solvent components.Examples of these solvents, and the availableprocesses in which there are applied are shownin Table 1. As can be observed, amines, moreprecisely alkanolamines, are the chemicalsolvents most used for CO2 capture, due itsreactivity and adequate basicity. From these,monoethanolamine (MEA) is the one typicallyconsidered. This compound have a primaryamine group, which confers a high reactivity tothe molecule, allowing a strong reaction with CO2at elevated rates.Due to this reactivity, MEA is highly proneto oxidative and thermal degradation, which limitsthe solvent concentration. The application ofoxidation inhibitors allows the increase of MEAconcentration to the typical value of 30 wt%, andeven 40 wt% [7]. In order to cope with this highreactivity this amines can also be blended withsterically hindered or tertiary amines which areless reactive [8].Concerning the thermal degradation, forMEA it is negligible if the regenerationtemperature is kept below 110 to 120 C [9].The capture process is based on the CO2dissolution, conducted in a packed column(absorber). The solvent loaded with CO2 (usuallycalled rich solvent) is pumped to the regenerationsection. Before entering the regenerator, the richsolvent is heated through an integrated heatexchange with the regenerated solvent (alsocalled lean solvent). The regeneration processtypically occurs through stripping aboveatmospheric pressure. The heat demanded bythis process is provided by a reboiler, whichconstitutes the main energy requirement of thecapture unit. The lean solvent is pumped back tothe absorber after being regenerated. Since thesolvent is subject to losses during the process, amake-up is required before re-entering theabsorber [2].Table 1 – Commonly used chemical solvents and processes in which they are applied.SolventTypeMEAPrimary amine ProcessesABB Lummus CressFluor’s Econamine FG PlusSMKS1TMHindered amine MHI’s KM-CDRMDEATertiary amineNH3AmmoniaHot PotassiumCarbonate Praxair’s AmineShell’s Cansolv CO2 Capture SystemAlstom’s Chilled AmmoniaUOP’s BenfieldGiammarco-Vetrocoke’s Low EnergyK2CO3Typical compositionUp to 30 wt% with corrosioninhibitorsBlend of sterically hinderedaminesUp to 40 wt%, blended with aprimary amineUp to 28 wt%Blend of potassium carbonate andamine promoters

In these processes, it is typical to achievecapture rates between 80 and 90%, and a purityas high as 99.9% (in volume) for the recoveredCO2 [2]. The inclusion of a PCC unit in a coalfuelled power plant leads to an increase of 29%in the energy input to achieve the same output,while for a natural gas fuelled power plant thisincrease is of 16% [4].4. Models ValidationThe validation of the models used in thesimulation of a capture unit was achieved throughthe comparison of the simulation results with theexperimental data publically available. For thatpurpose, were considered the papers byTobiesen et al., [12], and by Notz et al., [13],whose flowsheets were implemented ingPROMS .Tobiensen et al. article [12] presents theexperimental data for 20 non-equal runs in pilotscale absorber. From the presented data, it waspossible to conclude that between Onda andBillet & Schultes correlations for the calculation ofmass transfer coefficients, the first one is moreaccurate. Using this correlation were obtaineddeviations from -13% to 26%, comparing thesimulated and experimental flow of captured CO2,as can be observed in Figure 1.3. Materials and MethodsSimulated AbsorbedCO2 (kg/h)In the present work, gPROMS ModelBuilder 3.7.1 is the simulation platformused for flowsheet model simulation andoptimisation. Starting from the existing gCCS model library, it is possible to assemble aflowsheet, in which are also included any otherauxiliary equations and custom sub modelsrequired.AnotherfeatureofgPROMS ModelBuilder is the optimisation tool, which canbe used to optimise the steady state behaviour ofa continuous flowsheet, considering both designand operation properties. Since the modelsconsidered in this thesis are non-linear, theoptimisation problem constitutes a nonlinearprogrammingproblem(NLP).gPROMS ModelBuilder 3.7.1 uses the SRQPD solver in thesolution of NLP problems, with an “ImprovedEstimation Based” convergence criterion [10].The models included in the gCCS libraryuse gSAFT as physical properties package. Thispackage is based on the Statistical AssociatingFluid Theory (SAFT) equation of state, anadvanced molecular thermodynamic method,based in physically-realistic models of moleculesand their interactions with other molecules. Forcarbon capture, gSAFT presents a modellingalternative to phase and chemical equilibrium inthe CO2-MEA-H2O system, since the chemicalbound between CO2 and MEA can beincorporated as a short-range association, beingincluded in the molecular model. This way, theinvolved reactions are treated implicitly, thusgreatly reducing the complexity of the model, thusincreasing its robustness [11].10864200246810Experimental Absorbed CO2 (kg/h)Figure 1 – Parity diagram of the absorbed amount of CO2(using Billet & Schultes correlation).These deviations are deemed acceptablewhen considering that predictive models that donot consider any fitting to the experimental datawere applied. It was also verified that thisdeviation tends to increase with the decrease ofthe solvent lean loading (Figure 2).Deviation in AbsorbedCO2 (%)26166-4 0.150.250.350.45-14Lean Loading (molCO2/molMEA)Figure 2 – Deviation between experimental and simulatedabsorbed CO2 with the considered lean solvent loading.Table 2 – Experimental and simulation results for the process key parameters and respective variationExample 1Key parameterCO2 capture rate(%)Specific heatrequirement(GJ/tCO2)3Example 83.985.2732