1 | P a g e

Newsletter2011

DECARBIT progress ultimo 2011DECARBit responds to the urgent need for further research and development inadvanced pre-combustion capture techniques to substantially reduce emissions ofgreenhouse gases from fossil fuel power plants. The project will accelerate thetechnology development and contribute to the deployment of large-scale carboncapture and storage (CCS) plants, in line with the adopted European policies foremission reductions. The project focus is to pursue the search for improved andnew pre-combustion technologies. DECARBit is designed as a Collaborative Large-scale Integrating Project.

Prolongation: Due to reduced availability of a suitable test slot for high pressuretesting of Pilot 4 at DLR in Cologne, ALSTOM CH has requested a prolongation ofthe final date of their high-pressure hydrogen combustion activity organized underSP5 – Pilot 4. Based on this the EC granted a prolongation of the full project untilmonth 54 (June 2012).

2010-2012 -> The PILOT phase

DECARBit has now reached the end of Y4and has already achieved many of its goals. Thecurrent newsletter sum up some of the results, of which, DECARBit is proud to givepublicity to.

Newsletter Contents page

Advanced pre-combustion separation (SP2) 2

Advanced oxygen separation technologies (SP3) 3

Enabling technologies for pre-combustion (SP4) 4

CAESAR progress 5

CESAR progress 6

2 | P a g e

Advanced pre-combustion separation (SP2)

The work undertaken in DECARBit WP2.1 has focusedon 3 types of membranes.

Polymer/ceramic membranes consisting of inorganicnanoparticles that show a large absorption for CO2

compared to H2 and a polymeric “binder” that has alow permeability for gases and some selectivity for CO2

have been developed by TNO. As polymer matrix,polyacrylonitril (PAN) was selected, while LayeredDouble Hydroxides (LDHs) were selected as absorbingparticles. The thermal stability was demonstrated to begood enough to allow the use in the 100 – 200 °C range.The pore structure of the LDH particles was modified toimprove CO2 adsorption and transport. Membrane filmformation studies with mixtures with different LDHcontent showed that during film formation LDHaggregates are formed. Only at low LDH contenthomogeneous defect free films were formed.Permeability measurement at different temperaturesshowed that the properties of the PAN film weredominant as only little LDH was present and thediffusion of the gases took place predominantly throughthe PAN matrix. To reduce the aggregate formationallowing for a higher LDH loading, the particle-polymerinteraction needs to be improved for better degree ofdispersion, thereby avoiding mesopores at the particle-poly.

NTNU has developed carbon membranes forapplication at medium to high temperature. Inconclusion, using poly (furfuryl alcohol) (PFA) aspolymeric precursor, both selective surface flow (SSF)carbon membranes and carbon molecular sieve (CMS)membranes were obtain The coating methods and aseries of carbonization parameters, includingcarbonization temperature and carbonizationatmosphere as well as heating rate, were investigated.The thin selective carbon membranes with thickness 5-10 µm were obtained from polymer precursor PFA.Sealing methods at medium temperature (<250°C) andhigh temperature (200-350°C) were developed by usingepoxy resin and metallization-brazing technique,respectively. CMS membranes showed selectivity ofH2/CO2 at 25 at 180 °C. SSF carbon membranes showedselectivity of CO2/H2 at ambient temperature, butinversed selectivity above 150 °C. Efforts to improve theselectivity of CO2/H2 were not successful, including the

addition of PVP additives to the polymer precursor andthe incorporation of magnesium oxide in the carbonmembranes.

At SINTEF, new inorganic dual-phase membranes aredeveloped for high temperature CO2 separation. Themembranes consist of selected binary or ternary metal-carbonates with Li+, Na+ and K+ cations embedded in aceramic matrix of CeO2 or Sm-doped CeO2. Themembranes are prepared as dense disks by firstlyproducing a porous matrix with tuneable pore size, poredistribution and porosity using various pyrolyzablefillers. The molten salts are then impregnated at hightemperature in air to fill in the connected pores of thematrix. The resulting dual-phase membranes are gas-tight at room temperature, as demonstrated withhelium permeance. Several sealing procedures wereinvestigated to enable flux measurements of theproduced membranes. The membranes were testedusing multi-component gas streams on the feed sidewith various partial pressures of CO2, O2 and steamapplying Ar as sweep gas. O2 and steam were alsoapplied in the sweep gas mixture in order to investigatetransport properties of the membranes. The results ofthis work enabled to demonstrate high CO2 selectivityand permeance of these novel membranes. Amembrane of about 1 mm thickness exhibits a CO2

selectivity of 1074 (CO2 over He) and a permeance of2.1•10-8 mol•m-2•s-1•Pa-1, respectively at 700 °C.These experiments were also used to further investigatethe transport mechanisms as a function of theprocessing parameters and the composition of themembranes.

Based on these encouraging results of the dual-phasemembrane concept, and the risks/results of the othermembranes developed, the piloting work in the finalyear has focused on up-scaling the dual-phasemembrane concept from lab-scale flat pellet toasymmetric tubular membranes. In this configuration, athin dense membrane layer (molten carbonateinfiltrated ceria layer) is typically applied on amechanically strong porous ceramic supports.

In addition to new membranes, a number of potentialadsorbent materials have been evaluated for the actualPSA application. More detailed testing has been carriedout with four different materials: A commercial AC(AP3-60, Chemviron, Germany), the meso-porous silica

3 | P a g e

(MCM-41), and two different MOFs (UiO-67 and USO-2-Ni). All these adsorbents have significant specificsurface areas (between 1000 to 2000 m2/g) and lowadsorption energy of CO2 (20-25 kJ/mole). Specialemphasis has been put on preparing particulates of theadsorbents giving high bulk density in the PSA columns.This is important to help minimizing the void fractioninside the columns as well as the total volume of thecolumns needed for the full PSA process.

In the following chart, from left to right: Adsorptionisotherms of CO2 on activated carbon with experimentalvalues (symbols) and model (lines); breakthroughresults at 25 °C and 15 bar with an equimolar feed ofCO2 and H2; evaluation of the process performance atdifferent CO2 desorption pressures using a paretorepresentation to show the trade-off between CO2

purity and capture rate.

Additional work was performed on the development ofnovel solvent systems for CO2-H2 separation. In thiswork package, the joined research activities of fourdifferent institutes, TU Delft, SINTEF, TIPS, and TNO arecombined. A concept to regenerate absorption liquidsfor acid gas removal has been developed. The keyaspect is a membrane contactor that is being used forthe regeneration at elevated pressures of chemicalabsorption liquids loaded with CO2. Experiments wereconducted to obtain some Vapor-Liquid Equilibrium(VLE) data for a series of absorption liquids selectedtemperatures, i.e. at 80, 100 and 120 °C. Theabsorption liquids studied included TEA (triethanolamine), DEA (diethanol amine), AMP (2-amino-2-methyl-1-propanol) and MDEA (methyl-diethanolamine). The results show that MDEA and AMP are themost promising solvents at a temperature of 100 °C.Both MDEA and AMP show potential for a loading up toabout 0.4 or 0.5, for MDEA and AMP respectively. TEAexhibits unfavorable VLE data for all isotherms, havingonly a limited loading as a function of the CO2 partialpressure.

Different properties of membrane materials have beenevaluated for the separation of CO2-H2 mixtures in amembrane gas desorption unit. The main objective isthe selection of a suited combination of membranematerials and suited solvents for CO2 removal from ahydrogen rich stream at high operating pressures. Itwas concluded that amine solvents with a polyolefinmembrane materials was the most suited solvent -membrane combination. General guidelines are givento combine absorption liquids with membranecontactors and different configurations have beenevaluated. Focus is on the application of an absorptionstep with regeneration of the solvent by a membranegas desorption process. The main feature is that theabsorption liquids remains pressurized during theprocess of absorption and desorption of CO2.

The results from SP3 were passed on to SP1 for thetechno-economic evaluation.

Advanced oxygen separation technologies(SP3)

SP3 pursued the development of advanced oxygenseparation technologies that could allow significantreduction in energy consumption for oxygen productionfrom air. In the DECARBit project Oxygen TransportMembranes (OTM) are considered for implementationinto an IGCC power plant as an alternative to oxygenproduction by cryogenic distillation. The perovskitematerial CaTi0.9Fe0.1O3 (CTF) is used in this work. CTF isan attractive candidate for OTM applications due to itshigh stability over a large range of oxygen partialpressures at high temperature and in the presence ofCO2. Tubular symmetric (dense) membranes and poroussupports of up to 60 cm length were prepared asillustrated below.

By tailoring parameters of the coating procedure, densemembrane layers with thickness of < 10 µm wereachieved. Further investigation of the sealing materials

4 | P a g e

and procedures are pursued within Pilot 2 of DECARBitproject where a goal is to test the prepared asymmetricmembranes.

The production of oxygen by air separation using hightemperature O2 selective sorbents was alsoinvestigated, specifically the CAR (Ceramic AutothermalRecovery) process with high purity perovskite materialsas sorbents operating at high temperatures (~600°C).The best performances were found with the spinels,one of which reached a 4.9 wt% OTC which represents a6-fold gain compared to the reference material.However, such performance levels were found only atoperating temperatures above 800°C which lies wellabove the target temperature of 600°C. Overall, no newusable and reliable material was identified, but verypromising prospects leading to further investigationhave been opened.

Simulation tools were also developed to represent bothfixed bed reactors as well as rotating reactor units. Theabove mentioned data from various materials wereused to generate isotherm models (ETHZ) forimplementation into the simulator. Both the simulatorand experimental data indicated that wash-coatedmonoliths offer the best type of bed both on a sorbentweight and bed volume basis. Such supports fromSINTEF were selected for an ageing test assessed over a100 hour, 12000 cycle period at 600°C by cyclingbetween streams of air and steam. The experimentsand the post characterization analysis indicated no signsof degradation. A rotating device was considered.However, the required pressure ratio for properoperation would induce excessive leakage across themoving parts due to sealing difficulties. Classicalpressure swing absorption (PSA) processes can toleratethe required large pressure ratios and therefore appearto be better suited for the use of sorbents in oxygenproduction.

Additional work was performed on the development ofadvanced cryogenic air separation technologies toreduce overall energy consumption for oxygenproduction. A model for air distillation columns ofdifferent kind of configurations was established and anoptimization model for minimization of entropyproduction in distillation columns was developed. Apilot concept based on the work performed(heated/cooled distillation stages) in the task has been

established. This concept was planned tested in a pilotlaboratory plant. Unfortunately, due to delays in theconstruction, only very few preliminary part-tests wereperformed. The rig itself is visualized below.

Enabling technologies for pre-combustion(SP4)

In the numerical part of SP4 the focus has been onthree different topics related to numerical simulationsof hydrogen combustion. The first topic is onsimulations for showing how hydrogen mix andcombust under gas turbine conditions by using verydetailed Direct Numerical Simulations (DNS). Here anopen source DNS code has been developed and severalsimulations have been performed. Secondly we havealso implemented and started verifying how a LinearEddy Model (LEM) can simulate mixing of hydrogen bythe use of significantly less CPU resources than a DNS.Furthermore the reaction mechanisms for hydrogencombustion have been studied in detail, and a preferredmechanism has been chosen.

A novel sub-grid model, denoted LEM3D, has beenformulated in physical space to capture turbulentmixing and reaction at all scales of turbulent reactive

5 | P a g e

flows. LEM3D is a 3D construction based on the LinearEddy Model (LEM), which is a mixing model for scalarsoriginally formulated on a 1D domain. In this work,LEM3D is coupled to the Reynolds-Average Navier-Stokes (RANS) flow solver of the commercial softwareANSYS FLUENT. The LEM3D-RANS coupling has thedesirable benefit of imparting spatially and temporallywell-resolved unsteadiness to steady-state RANSsolutions. In addition, the modeling approach resolvesall scales of turbulent reactive flows at acomputationally affordable cost compared to acorresponding Direct Numerical Simulation (DNS). Inthe project, the coupled model LEM3D-RANS has beenfurther developed and verified against experimentaldata. Results from simulations were passed on to aidthe construction of the new pilot H2 GT burner.

Siemens AG was engaged in SP4 for the design of anoptimized fuel system concept to guarantee reliablefuel conditioning for stable gas turbine operation andoptimized interactions with the connected systems likeair separation unit, gas treatment and water steamcycle. The evaluation of ignition limits and auto-oxidation effects in lab-scale experiments andsimulations confirmed that there will be no risk of self-ignition in the fuel system, for typical syngas pressureand composition, if the syngas temperature is lowerthan 400°C. Experiments in a plug-flow reactor andmodeling of different auto-oxidation reaction pathwayscame to the conclusion that only 3% of the residualoxygen concentration – which is very small (<< 1 vol%)in normal operation - will be consumed by auto-oxidation for residence times of ca. 4 s in a realistic fuelsystem, mainly caused by oxidation of H2. The materialselection for the fuel system found that carbon steel inthe dry section and austenitic stainless steel in the wetpart (downstream the water saturation process) aresafe concerning corrosion and H2 embrittlement. Forthe welding of dissimilar (ferrite/austenite) materials,low carbon steels were suggested to avoid the risk of H2

embrittlement and the formation of carbide seams. Themain result of this study is that converting existingnatural gas pipelines to deliver pure hydrogen mayrequire substantial modifications or be impossible,whereas converting natural gas pipelines to carry ablend of natural gas and hydrogen (up to about 20 vol%hydrogen) may require only modest modifications tothe pipeline and could be feasible with moderate costs.

Regarding the experimental part, the concept ofsequential combustion has been implemented withinAlstom’s GT24/GT26 family of gas turbines, offeringenhanced operational flexibility, low emissions and ahigh efficiency in a very compact arrangement. Themain flow passes through the first combustion stage,comprising - EV burners, wherein a part of the fuel isburnt. After its expansion in the high-pressure turbinestage, the remaining fuel is added in the SEVcombustor. Both combustion stages comprise premixedburners, as NOx emissions depend on the quality of thefuel and oxidant mixing. Two different combustionsetups were tested; one with a high burner velocitywhich gives a low residence time in the premixingchamber, and the other one with a lower velocity of thefuel inlet yielding a long residence time. The former iscapable of dealing with shorter ignition delay times, buthas higher burner pressure drop.

CAESAR progressThe CAESAR project has now reached its end. The mostimportant results were presented at the at the thirdEuropean Conference on CCS Research, Developmentand Demonstration, 24 - 26 May 2011, London, UK

The main findings of CAESAR were presented at theconference. The presentations can be found here:

http://caesar.ecn.nl/presentations/

6 | P a g e

CESAR progressThe CESAR project was finalized in 2011. The mainresults are the work performed in the EuropeanBenchmarking Task Force (EBTF). The final EBTF reportis the result of a joint effort of a team of members ofthe CAESAR, CESAR and DECARBit FP7 projects – Itpresents a compilation of the contents of two previousreports of the EBTF – assumptions and parameters forCarbon Capture projects from the Common FrameworkDefinition Document and three technical study cases ofpower plants without and with CO2 capture – and itincludes new material related to the costs andeconomics of carbon capture. The performance of newcycles proposed within the three projects, incorporatinginnovative capture technologies, should be comparedand referred to the performance of these three cases.The three cases are: an Advanced SupercriticalPulverized Coal plant, an Integrated GasificationCombined Cycle and a Natural Gas Combined Cycle. Foreach case, a general description of the case ispresented, followed by the specification of the processstreams, operational characteristics and operationalperformance. The final part of the report is dedicated tothe economics of these three cycles. This report is thusself sufficient and does not require the reader to knowthe two previous reports.

A high degree of collaboration and interaction amongthe members of the European Benchmarking Task Forcewas required for the completion of its mission. Theauthors believe that this requirement was fully satisfied,allowing the achievement of the EBTF objectives.Opinions, suggestions and contributions given by manycolleagues not directly engaged in the task force aregratefully acknowledged. As members of the three

7 | P a g e

projects funded by the FP7 program, the authors areindebted to the European Commission for theopportunity of carrying out such a difficult but verynecessary task. They confidently expect that their workwill be useful to the carbon capture research anddevelopment community.

The table below gives results for the AdvancedSupercritical Pulverized Coal (ASC) test case. The highsimilarity of the parameters and assumptionsconsidered in the two projects has led to a remarkablyhigh similarity of results. The gross electricityproductions and the efficiencies are practically thesame in the two projects. Also the emissions in thecases without capture are practically the same.

The second table shows results for the IntegratedGasification Combined Cycle (IGCC) test case. Thenumbers from CAESAR shown in the table have beenobtained under some assumptions defined in theCAESAR project, not the same as the correspondingones defined in DECARBit. The consequence is that thegross electricity output is not in as good an agreementas the other results for the case.

For the results shown in the third table different plantconfigurations have been considered by CESAR andCAESAR. The gross electricity output is hugely different

but easily explained. The efficiencies, specific emissionsand CO2 removal percentages are, nevertheless, in verygood agreement.

The results shown in the EBTF report allow other teamsof other current or future projects to evaluate their owntechnology propositions in a consistent and welljustified way, using the same sets of assumptions andparameters in the report. Advantages or disadvantagesof a technology over another can thus be crediblydemonstrated to a good approximation.

The report can be found on the CESAR web-site:

http://www.co2cesar.eu/site/en/downloads.php

8 | P a g e

CArbon-free Electricity by SEWGS: Advanced materials,Reactor-, and process designStarting date 1 January 2008Duration 48 monthsBudget 3.1 million €EU-contribution 2.3 million €Coordinator Energy Research Centre of The Netherlands

CESAR Enhanced separation & recoveryStarting date 1 February 2008Duration 48 monthsBudget 6 million €EU-contribution 4 million €Coordinator TNO Science and Industry

DECARBit Enabling advanced pre-combustion capturetechniques and plantsStarting date 1 January 2008Duration 48 monthsBudget 15.5 million €EU-contribution 10.2 million €Coordinator SINTEF Energy Research

Contact: Nils A. Røkke - nils.a.rokke@sintef.noEditor: Marie Bysveen - marie.bysveen@sintef.noPublished by: SINTEF Energi AS (SINTEF Energy Research) - NO-7465 Trondheim - Phone: + 47 73 59 72 00 - www.sintef.no/energy