Final Report Summary - UNIQUE (Integration of particulate abatement, removal of trace elements and tar reforming in one biomass steam gasification reactor yielding high purity syngas ...)
This project aims at a compact version of a gasifier by integrating the fluidized bed steam gasification of biomass and the hot gas cleaning and conditioning system into one reactor vessel. The main objective is to develop an innovative technology for the production of syngas with the specifications required for use in fuel cells in a cost-effective way, keeping high the thermal efficiency of the whole conversion process, as no cooling step is included.
A new catalyst for in-bed primary reduction of heavy hydrocarbons (tar) was prepared and characterized by increasing the iron content of olivine (a natural mineral substance). This catalyst was demonstrated effective to reduce tar content by 45 % and increase gas yield by 40 % (when compared to olivine), in bench and pilot scale (100 kWth) gasification tests. Catalyst production on large scale (1 ton) was also developed. This very low-cost material removes completely the problem of heavy metals in the ashes to be disposed.
Innovative sorbents to be added to the gasifier fluidized bed were synthesized and tested successfully: sulphur, chlorine and alkali were reduced below the threshold limits required to feed a solid oxide fuel cell (H2S and HCl below 1 ppmv; KCl below 100 ppbv). The SOFC performance obtained with the syngas from steam gasifier in Güssing and in Trisaia was comparable or better than that measured with a reference fuel.
Tar reforming catalytic filter elements were optimized to produce high-performance catalytic systems. Scale-up of the whole procedure allowed the manufacture of commercial-size catalytic candles, with guaranteed filtration properties (particle removal efficiency higher than 99.9 % of particulate present in the raw syngas). They were tested at bench scale (0.5 kg/h biomass feeding rate) and a reduction of up to 80 % of tar content in the producer gas was obtained by comparison with a blank test. Tar abatement increased up to 92 % when Fe/olivine catalyst and catalytic candle were combined together.
Catalytic filtration tests performed at Güssing industrial plant (8 MWth) did confirm high potential for hot gas cleaning, although further work is needed to prove the overall technical feasibility and the long term behavior remains an outstanding issue.
Finally, the hardware modifications needed to implement the UNIQUE hot gas cleaning and conditioning technology at the Trisaia pre-existing gasification rig were fully designed. The 1 MWth pilot plant in ENEA Research Centre was properly arranged to permit the insertion of filter candles inside the gasification reactor; however, because of delays in plant construction activities, significant tests were not carried out in the contractual period.
Project context and objectives:
In existing gasification installations, gas cleaning is normally done by filtration and scrubbing of the producer gas: in this way the clean gas is made available at temperatures close to ambient, and the most immediate option for power generation is gas engine. Such process configuration does not allow high electric conversion efficiencies: reported values are close to 25 % that is what is also obtainable with modern combustion plants coupled with steam turbines. This penalizes notably the overall economic balance of the plant, which would benefit of a higher share of electricity against heat production.
The project goal is a compact version of a gasifier integrating the fluidized bed steam gasification of biomass and the hot gas cleaning and conditioning system into one reactor vessel. This is obtained by placing a bundle of catalytic ceramic candles operating at a temperature as high as the gasification temperature (800 - 850 °C) in the gasifier freeboard; furthermore, by using a catalytically active mineral for primary tar reforming and sorbents into the bed for removal of detrimental trace elements.
A major gas contaminant is solid particulate entrained by the syngas. They can block gas passages and/or the anode surface of a fuel cell, and they are one of the major responsible of the local air pollution. The tolerance limit of high temperature fuel cell to solid particulate is of the order of a few mg/Nm3.
Tars (heavy hydrocarbons) also originate from gasification, and are contained in the producer gas. They condensate at temperatures lower than 400 °C, plugging gas passages and downstream equipment. The presence of tar among the products of gasification reduces gas yield and conversion efficiency. These contaminants are also responsible of carbon deposition that can plug the porous media of a fuel cell anode. The tolerance limit of high temperature fuel cell to tar is still not well defined in the literature (due to the novelty of this fuel cell application); a reasonable value is of the order of 100 ppmw.
Sulphur compounds are generated by the gasification process because of the presence of a limited amount of sulphur in the biomass. The most important is H2S, responsible of chemisorption on Ni surfaces that blocks active sites, in the catalytic gas conditioning system and in the fuel cell utilized for power generation. The tolerance limit of high temperature fuel cell to sulphur compounds is of the order of a few ppmv, to allow operation below 1000 °C.
Fuel bound chlorine in biomass is released as HCl during gasification. It is highly corrosive, especially against the interconnect material of the SOFC.
Alkali metal contaminants are released as chlorides during gasification. They can lead to ash softening and melting and thus plugging of the filter. By condensing in cooler parts, they can block gas passages and / or the anode surface of a fuel cell, and thus they are one of the major influencing factors of fouling.
Fuel-bound nitrogen in biomass is principally released into the gasification gas primarily as ammonia and only a small part as HCN. NH3 has no relevant effect to the fuel cell if it's lower than 1 % by volume (gasification raw gas usually contains a lower concentration of ammonia). When this gas is combusted, NH3 has the tendency to form nitrogen oxides (NOx ), which are pollutants difficult to remove, and precursors of 'acid rain'.
This project deals with the above mentioned problems by means of an integrated approach to hot gas cleaning and conditioning that includes a set of scientific and technical objectives itemized below.
An innovative catalytic system is developed for in-bed primary reduction of heavy hydrocarbons by means of a natural mineral substance (olivine) where the iron content is increased up to 20 % by impregnation. In comparison with previously obtained Ni/olivine catalysts, this very low-cost material removes completely the problem of heavy metals in the ashes to be disposed, and will assure similar reforming activity due to its higher content of iron.
Tar reforming catalytic filter elements are optimised by screening new catalyst supports at laboratory scale, with regard to the adjustment of a high BET surface area at high temperature, by studying the catalytic activation process, and producing a high-performance catalyst system.
The whole procedure is scaled-up to allow the manufacture of commercial-size catalytic candles, with characterised filtration properties (ceramic candles are able to guarantee particle removal efficiency higher than 99.9 % of particulate present in the raw syngas).
Synthetic sorbents to be added properly to the gasifier are chosen and characterized, to trap sulphur compounds and additional, detrimental trace elements in order to reduce their content in the hot gas to values compatible with downstream units, e.g. a high temperature fuel cell. Sulphur capture also reduces or eliminates deactivation by H2S of the catalytic filter.
The feasibility of the integrated arrangement proposed for the gasification reactor is checked experimentally, and the performance of the gas conditioning and cleaning system at real process conditions properly quantified, by means of a thorough test campaign at bench scale, in a bubbling fluidized bed gasifier with nominal load of 1 kg/h of biomass feedstock.
In addition, the novel catalyst and the synthetic sorbents are tested in the 100 kWth FICFB (dual fluidized bed steam blown biomass gasifier) pilot plant (Vienna University of Technology (TUV)) in comparison to olivine and Ni/olivine.
An industrial-scale benchmark of the effectiveness of the gas cleaning and conditioning system is obtained by housing a catalytic filter candle inside a commercial gasifier (Güssing plant, 8 MWth), to compare the producer gas quality in the slip stream passing through the innovative conditioning system, with that of the existing low temperature gas filtering and scrubbing system. Long duration tests to study aging or other time depending effects were also planned.
A pilot scale unit (1 MWth) designed according to the principles illustrated above is also operated. Its realisation is done by modifying the freeboard of an existing oxygen / steam, bubbling bed gasifier, in order to insert the candles needed for particulate filtration and tar reforming. The reactor is already equipped with a weir, over which the overload of fine and light particles, detached from the filter surface and floating on the bed surface, would flow to an adjacent chamber from which it may be withdrawn.
The feasibility of feeding the clean fuel gas finally obtained to a high efficient power generation system is tested by means of a bench scale, solid oxide fuel cell. As a result of laboratory work, a portable SOFC unit, suitable for tests with the producer gas on its anode side, is assembled and arranged to be connected to a slip stream of the commercial and pilot scale gasifier, respectively.
Modelling studies are carried out at different scales. The complex heterogeneous reacting system inside the filtering medium is simulated, at contact times typically shorter than one second: catalytic reactions are favoured by relatively low velocities in the filter and the influence of fine structure of the porous material for optimal operation needs to be studied in details, as well as that of a large pore size, characterising the filter material, which on one hand reduces mass transfer limitations, on the other hand reduces also the specific active surface. A CFD model of the reactor freeboard is also implemented, to optimise candles configuration in it. Process simulations of the whole energy conversion chain allow to characterise the process flow-sheet and its overall thermal and chemical efficiency.
Project results:
A summary description is provided of activities and major results obtained in the UNIQUE project.
'Catalyst and sorbents for in-bed primary reduction of tar and trace contaminants'
Characterisation and development at industrial scale of Fe / olivine catalysts was performed, to improve syngas yield and the hydrogen concentration, and to decrease tar production in the gasification of biomass in a fluidised bed reactor. This new bed material satisfies also attrition resistance, low cost, no toxicity and simplicity of the preparation procedure.
After analysis of the existing literature. University of Strasbourg (ULP) decided to focus on an iron salt impregnated on the olivine support. The key of the success was the conditioning of the material (calcination at high temperature and in-situ reduction) to develop a strong interaction between the iron metal or the iron oxides particles and the olivine support. The best choice for the formulation has been a olivine based catalytic material containing 10 wt% of iron, calcined between 900 and 1100 °C and reduced in situ by the gasification gas.
The catalysts have been fully characterized by several techniques of bulk and surface: X-ray diffraction (XRD), Mössbauer spectroscopy, X-ray photoelectrons spectroscopy (XPS), scanning electron microscopy (SEM) and temperature programmed reduction (TPR). These techniques showed that olivine structure was maintained whatever the thermal treatment. The different oxidation states of iron and the environment of iron particles were determined at different calcination conditions. Diffusion of iron oxides inside the grains has been evidenced at high calcination temperatures. At the surface, the amount of iron III is also related to the calcination temperature. The reduction temperature and the amount of reducible iron (iron reducibility) have been quantitatively determined.
The 10 % Fe/olivine catalyst has been tested at different operating conditions in order to identify the active sites and to optimize its activity in toluene (T) and 1 methylnaphthalene (1-MN) reforming. The results show clearly the greater efficiency of iron / olivine catalyst compared to olivine, and a good activity compared to Ni/olivine.
In addition, ULP studied the scaling up of the catalyst preparation at the scale of kg then ton. Finally, 300 kg and 700 kg have been produced for pilot plant tests at Vienna University of Technology (TUV) and ENEA Trisaia Research Centre (Italy), respectively.
Additional research activities were addressed to the choice and characterization of synthetic sorbents to be properly added to the gasifier to trap detrimental trace elements like H2S, HCl, alkali metals and heavy metals in order to reduce their content in the hot produced gas to levels compatible with a high temperature fuel cell. The interactions between ashes, sorbents, and filter candle materials was also investigated.
Sorbents for alkali and sour gas (i.e. H2S and HCl) removal were tested regarding their sorption efficiency and capacity as fixed-bed in lab-scale furnaces. Furthermore, the capability of different sour gas and alkali sorbents for co-sorption of heavy metals, i.e. Zn, Cd, and Pb was investigated. For the sorption experiments a gas stream was passed through a sorbent fill with a length of 50 mm which corresponded to a weight of about 25 g. The temperature of the sorbent bed ranged from 700-900 °C. In order to determine the composition of the cleaned gas leaving the sorbent fill the tube furnace was connected with a molecular beam mass spectrometer (MBMS).
Bauxite, kaolinite and n.o. zeolite are suitable for KCl reduction to values below 100 ppbv. Slag lime, as the best tested 'conventional' sulphur sorbent only reduced the H2S concentration to 50 ppmv at 800 °C. A new stabilised Ba-based sorbent developed at Forschungszentrum Jülich (FZJ) achieves H2S concentrations below 1 ppmv at temperatures higher than 760 °C. This should be sufficient to prevent poisoning of fuel cell materials. Furthermore, the stabilised Ba-based sorbent is suitable to reduce HCl below 1 ppmv. Zink was best absorbed by aluminosilicates which are already used for alkali sorption limiting the Zn concentration to 400 ppb. Lead was also best absorbed by aluminosilicates with kaoline limiting the Pb concentration to 200 ppb. Cadmium could not effectively be removed by any of the tested sour gas and alkali sorbents at an inlet concentration of 1-2 ppm. Furthermore, the fact that neither the alkali nor the sour gas removal is kinetically limited at a space velocity of 9800 h-1 shows the suitability of these sorbents for the product gas stream of the Güssing gasifier.
In-bed tar reforming catalyst (primary tar reduction) and the synthetic sorbents were tested at process conditions representing the industrial application of dual fluidized bed (DFB) steam gasification. Experimental investigations highlighting the effect of Fe-olivine, sulphur and alkali sorbent in the DFB biomass gasification system were carried out at bench scale (10 kWth fuel input) and pilot plant scale (100 kWth). A comprehensive process parameter study was accomplished to emphasise the effect of the in-bed catalyst resp. sorbents on the gasification process performance. Further, reference tests with silica sand and olivine were carried out. Thus, an extensive mapping of the process performance was developed including the scale up effects. Furthermore, the influence of oxygen transport on tar conversion caused by cyclic oxidation-reduction of the Fe was experimentally investigated and assessed at a dual circulating fluidized bed reactor (DCFB) system at pilot rig scale (100 kWth fuel power).
The bench scale unit at the Consejo Superior de Investigaciones Científicas (CSIC) was designed and operated at conditions as similar as possible to the 100 kWth dual fluidized bed gasification plant located at TUV. This unit consists of two interconnected fluidized beds (gasifier and combustor) with solid material circulating between them. The unit has a nominal capacity of about 300 g/h. Pine wood delivered by TUV, after crushing and sieving, was used as biomass for the gasification tests.
The bench scale tests have been useful to determine the effect of the main operating conditions, such as gasification temperature and steam / fuel ratio, on the tar primary reduction.
Silica sand, olivine and Fe-olivine were applied as bed material at the pilot scale unit. Table 3 highlights a summary and comparison, respectively for gasification temperature of 850 °C. The specified tar content is related to GC/MS measurement. The results for Fe-olivine are related to the long term experiment carried out with used Fe-olivine. Application of silica sand is taken as bench mark value for comparison as this material is suggested to be inert in terms of catalytic activity. Beside the differences in composition of the product gas, the tar content is strongly varying. Decreasing tar content is found in the order Silica sand > olivine > Fe-olivine. Thus, Fe-olivine is identified as best promoting the tar reduction by its use as in-bed catalyst for primary tar reduction. Further, the experiments at the DCFB reactor system revealed the oxygen transport capacity of Fe-olivine due to cyclic reduction and oxidation. This characteristic is also developed in the DFB system. The oxygen transport causes partially oxidation of product gas in the gasifier and contributes to the heat demand in the gasifier.
'Catalytic filter candles'
Nickel containing tar reforming catalytic filter elements for hot gas cleaning and conditioning were developed from laboratory to commercial scale, by Pall Filtersystems GmbH Werk Schumacher (PALL) in collaboration with the University of Strasbourg (ULP) and the Consejo Superior de Investigaciones Científicas (CSIC). Validation of the most catalytically active catalytic filter element was performed in the freeboard of a bench-scale gasifier (University of L'Aquila (UNIVAQ)), before testing a catalytic filter candle of commercial scale in the freeboard of the Güssing gasifier (Biomasse Kraftwerk Güssing GmbH & Co KG - BKG).
A screening of activity of several catalytic layer systems led to the development of an active catalytic layer system consisting of a MgO-Al2O3 supported Ni catalyst. The catalyst was examined on its catalytic activity under model gas conditions using toluene and 1-methyl-naphthalene as model tars at ULP as well as under real gas conditions at UNIVAQ.
After this screening phase an alternative catalytic filter design was examined to increase the catalytic performance. In this fixed catalyst bed design the catalyst is integrated in grain form into the annular space between the filter candle and a porous inner tube. Appropriate samples were tested at CSIC for real gas validation.
A catalytic filter candle segment of 400 mm length was tested at UNIVAQ integrated in the freeboard of the bench-scale gasifier, with 80 % average tar conversion, over 22 hours on test. Scale-up of the catalytic filter candle of fixed bed design was performed to provide a 1.5 m long catalytic filter candle prototype for real gas testing.
With respect to operating temperatures 850 °C in the freeboard of the Güssing gasifier, a new Al2O3 based catalytic filter candle was developed from laboratory to commercial scale (DeTarCat CL) by transfering the most catalytically active layer on this new filter material.
CSIC has tested the activity behaviour of different catalytic filter elements delivered by PALL (s. Figure 4).
In a test campaign at UNIVAQ with simultaneous utilization of Fe / olivine catalyst in the fluidized bed of the gasifier and the catalytic filter candle in its freeboard (0.5 kg/h biomass feed rate), the gas yield increased on average by 75 % and the hydrogen yield by 152 %. Correspondingly, the methane and tar content in the gas was reduced by 30 % and 92 %, respectively, and tar production per kg of dry ash free (daf) biomass by 86 %.
The filter candles were directly integrated into the freeboard of the Güssing gasifier. Thus, industrial scale performance is investigated since the unpurified raw gas is charged to the filter candle.
A filter candle test module was erected including the direct mounting of the filter candle into the freeboard of the gasifier. The test module comprises the main units:
- fixture unit of the filter candle with the freeboard construction;
- back pulsing/dedusting unit (pressure vessel, dedusting valve);
- pipe system equipped with trace heating;
- control unit (for back pulsing, volume flow, pipe heating, pump control, temperature, pressure);
- pump, filter and condensate trap unit;
- gas sampling.
Stable filtration of raw product gas was not accomplished for the type DeTarCAT FB and DeTarCAT CL due to attack of SiC structure caused by temperature peaks. A test run with the type DeTarCAT CL-Al was successfully accomplished with a stable filtration period of about 28 hours.
Based on these results, a preliminary functional and operability analysis allows highlighting that the filter candle positioned inside the gasifier freeboard shows high potential for hot gas cleaning, while further work is needed to prove the overall technical feasibility, and the long term behavior remains an outstanding issue. As a matter of fact, complications were experienced in matching the requirements of the UNIQUE experimentation with the shut-down time intervals of a plant operating commercially. This resulted in limitations for the number of affordable tests, although some important results were obtained as pointed out below. UNIQUE partners recognize the need of further testing activity, more precisely about the characterization of interactions between the candles and the gasifier environment, from the point of view of materials, average temperature, temperature hot spots, etc. However this type of study would require an experimental plant (not a commercial one).
In addition, focussing on the gasifier temperature and the temperature control is partially correct. In particular the temperature control is meant for levelling the heat balance in the gasifier. Especially, during the start-up and in case of particular requirements (charging of wet biomass), the temperature in the gasifier might exceed the range of temperature (of 800-850 °C as indicated in the DoW). The feasibility and functionality of the original envisaged filter candles material (SiC) has been proven at the lab-scale device, whereby moderate temperatures and steady temperature levels are applied. However, it has been shown that the requirements at industrial scale (fluctuations in temperature, impact of real process gas with dust and char load, and mechanical impacts, e.g. vibrations) differ. The combination of temperature and present atmosphere which surrounds the filter candle, have revealed that other materials have to be applied to resist and to preserve the chemical stability. Thus, the application of Al2O3 as suitable filter candle material was the main finding gathered during the test runs and in consultation with the evaluation of the applied filter candle material. In line with this outcome, the tests were an important part for the development of properly operating filter candles.
The 28 h hot gas filtration test can be regarded as the key test for an assessment of the practicability of the developed catalytic filter on industrial scale and in part of the UNIQUE concept, even if the full demonstration is still to be realized. Additional results e.g. the tar reforming activity and tar analysis will be the objective of a future publication by TUV in collaboration with PALL, so that key results about combined particle separation and catalytic tar reforming under gasifier freeboard conditions with commercial-sized catalytic filter candles can be taken into consideration for all developers active in this field.
'The UNIQUE prototype pilot plant'
The 1 MWth pilot plant in ENEA Trisaia Research Centre was properly modified to permit the insertion of filter candles inside the gasification reactor.
On the basis of the input data given by ENEA about the expected gas flow rate (400 Nm3/h) and the original freeboard geometry, PALL proposed the preliminary tube sheet design. The tube sheet was sized in order to accommodate 75 filter candles, distributed in 5 rows of 15 candles, with 70 mm outer diameter and 1520 mm length. ENEA prepared the layout concerning all hardware modifications: the head of the reactor, the nearby pipeline, and the baffle welded to the freeboard wall.
In addition, ENEA analysed the gas emission risks at the plant. The assessment was carried out according to the current Italian standard (CEI 3135, CEI 3130) and the European directives (94/9/EC, 99/92/EC). Twelve main sources of gas emissions were identified and all hazardous areas were classified as zones of low risk (zone 2). A detailed description of the risk analysis was provided.
Due to possible risks of pore blocking of the filter candles during initial operation of the plant, PALL planned to first deliver 75 non-catalytic filter candles, and provide the catalytic ones after the proper and reliable functioning of the modified pilot plant was proved by means of preliminary gasification tests.
Regarding the realisation of the pilot plant, some of the previously described reactor modifications were implemented during the 3th and 4th semester. The length of the head was increased to 54 cm, while the baffle, originally welded to the reactor wall, was made removable so as to allow the proper housing of the candles. Additional sampling points were arranged.
Some strength calculations were undertaken in ENEA. In particular, for the tube sheet, a structural linear / nonlinear and thermal analysis was done (using ANSYS® software). It was necessary to choose a tube sheet 50 mm thick, made of A310S steel, a high temperature resistant steel.
Because of substantial delays in the prototype construction phase, the original planning of the gasification test campaign could not find adequate time in the contractual period, and most of it will be carried out by ENEA after the end of the project. The first phase of the experimental campaign will be addressed to test the performance of the steam / O2 gasification process when using the bed catalytic material (10 % Fe/Olivine) and the sorbents, however without the hot catalytic filter system. After this first campaign, the new filter system, including catalytic and non-catalytic candles, will be installed in the gasifier and the long running test, devoted to evaluate the performance of the whole UNIQUE technology, will be carried out.
To partially overcome these difficulties and to gather experimental evidence useful for operation of the 1MWth gasifier, bench scale experimental activities (10 kWth ICBFB gasifier available in Trisaia) were included in the original project programme, with a contract amendment procedure. The design of this reactor is similar to that of the UNIQUE prototype, and it also operates with oxygen and steam gasification agents.
Gasification tests were carried out by using 10 % Fe / Olivine and bauxite. Changes in process conditions induced by key parameters, such as ER and S/B ratios, were also evaluated. The raw syngas was characterised in terms of gas composition, organic & inorganic contaminants, particulate content and process efficiency (gas yield, carbon and water conversion, LHV). In order to assess the efficacy of the new active phases, towards the removal of tar and alkali halides, the results were then compared to those obtained by using natural olivine as bed material. A significant reduction of about 40 %wt was observed for the tar load while, due to the greater brittleness of the new catalytic material, an increase of the entrained particulate was observed, which became twice as much.
As far as the alkali halides removal is concerned, the positive effect of bauxite was confirmed, mainly in terms of KCl removal, although less extended than expected on the basis of the lab tests.
'Operation of a solid oxide fuel cell (SOFC) with biomass gas'
A portable SOFC test station was designed and assembled to allow fuel stream tests with both pilot scale and commercial scale gasifiers using cleaned product gas from the biomass gasifier as a fuel supply. It consists of air supply and filtering system, fuel humidification system, electric furnace, air and fuel pre-heating systems, instrumentation and data acquisition system, performance test equipment (electronic load) and SOFC assembly. The gas manifolding and valve system allows for different gases to be supplied to fuel manifolds during the start-up, operation and shut-down of the test station. In the standard operating mode, test station is fuelled with syngas supplied from the biomass gasification unit.
Tests of the SOFC station components have been completed to evaluate gas leaks, performance, measurement and control range accuracy for the fuel pump, mass flow rate controllers, mass flow meters, gas cooler, gas humidifier, air and fuel heaters and electric furnace. After validation of the component units' performance and construction of the portable SOFC station, preliminary tests of the unit have been completed. The test station was operated in a full range of fuel flows. Temperatures of the gas streams at the selected process check-points have been consistent with expectations. Onsite test requirements, for the operation of portable test station with the gasifier, are limited to availability of utilities (electric, technical gases, water, and ventilation system).
Preliminary off-site tests with the SOFC cells fuelled with the BKG (Guessing, Austria) syngas compressed in gas cylinders have been completed at the Institute of Power Engineering (Warsaw, Poland). Onsite tests have been performed at the ENEA Research Centre ( Trisaia, Italy) and BKG.
During the first series of tests at ENEA Research Centre, after initial polarization measurements, the cell was operated under electric load of 4 Amps. Lower than expected cell performance was observed. After cell disassembly, heavy carbon deposits were observed on the current collector surface. Anode surface remained mostly carbon free. The results have been attributed to humidification temperature of 55 ºC, which was too low for carbon-free cell operation. In the second series of tests, syngas humidification was kept at 65 ºC to avoid carbon deposition. Performance of the SOFC cell was much better than in the first series of tests and post-test SOFC cell inspection did not show any cell cracks. Minor carbon deposits were visible on the current collector surface facing the gas supply side. Following performance tests, cell was operated under electric load.
Similar test were performed at BKG biomass gasification plant in Guessing. During the 26 hour test, no cell performance decline was observed.
The results of tests performed at IEn, ENEA and BKG indicate that cleaned syngas from biomass gasification reactor is a feasible fuel for the anode supported SOFC cell. Cell performance on cleaned syngas is comparable or better than cell performance on 47.5 % H2+47.5 % N2+3 % H2O gas mixture. Proper humidification of the gas stream, necessary to avoid carbon deposition, is critical for the long term cell performance (carbon deposition). The presence of larger amounts of methane and higher hydrocarbons may lead to SOFC cell cracking due to large temperature gradients resulting from internal reforming process.
The product gas from gasifier contains impurities (H2S, HCl, particulates, tars, alkali metals, etc.). This syngas stream must be cleaned before it can be used to fuel SOFC. Experimental setup to investigate the effect contaminants in the fuel on the SOFC cell performance was designed, assembled and tested. In the experiment, fuel components (H2, N2) were supplied from gas cylinders, mixed, humidified and supplied to SOFC cell. The effect of contaminants, containing chlorine in form of HCl(g) and sulfur in form of H2S, on the SOFC performance has been investigated at 7500 C and 8000 C. The cell performance was evaluated using constant current load at the varying contaminant levels. The effect of contaminants is reflected by the lower cell voltage.
Anode supported SOFC cells were operated for a total of around 500 hours at 800 °C with H2S levels ranging from 0.1 to 2 ppm. Performance tests indicate negligible short term effect (around 200 hrs.) of up to 1 ppm H2S in the fuel at 800 °C. Significant voltage decline has been recorded at 2 ppm H2S. Another series of tests were performed at 750 °C with H2S concentrations ranging from 0.3 ppm to 1.5 ppm. The effect of H2S on the cell performance is evident for 1.5 ppm H2S concentration in the fuel at 750 C. The voltage decline of 150 mV has been recorded after 100 hours of operation with the 47.5 % H2 + 47.5 % N2 + 3 % H2O + 1.5 ppm H2S fuel. Two distinctive log-linear regions of voltage decline are visible after H2S is introduced to the fuel stream: faster and shorter region, followed by the slower voltage decline region.
Following the H2S tests, HCl(g) effect on the SOFC cell performance was examined in a series of experiments. No significant effect was observed during the tests for HCl(g) concentrations ranging from 1 ppm up to 10 ppm at 750 ºC for both dry fuel and fuel humidified at 25 ºC. The voltage decline of 16 mV has been recorded after over 60 hrs of cell operation with 10 ppm HCl contamination. The temperature effect of HCl on the anode performance is less significant than H2S effect while cell response to HCl contamination is fast relatively to H2S.
'Accompanying modelling activities at different scales'
Modelling and numerical simulation activities, carried out in strict connection with the experimental investigations, are conceived as a means to better understand either potentiality or limits of the developed systems when applied to a wide range of conditions. The calculation tools used inside UNIQUE are useful for design and optimization of the gas cleaning and conditioning system here proposed, as well as for the optimization of the integrated energy conversion chain.
Modelling activities inside UNIQUE were developed at different scales (micro and macro systems) and related to different expertises (thermodynamics, chemical reaction kinetics, fluid dynamics, plant design, and energy systems).
The thermodynamic process model, designed using SimuSageTM (GTT-Technologies), allows calculation of the type and amount of gaseous and condensed trace elements in parts of the UNIQUE plant downstream the hot gas cleaning. The results of the thermodynamic process model were obtained by Gibbs free energy minimisation. The variable parameters of the model are the inlet streams like biomass, steam and oxygen, the gasifier temperature and the sorbents.
The processes of solids filtration and tar reforming taking place inside the catalytic filter candle were also modelled. A computational fluid dynamic (CFD) tool was developed and validated against experimental findings, able to simulate the fluid dynamic behavior of gas and solid particles inside the fluidized bed freeboard of the UNIQUE prototype, where filtration candles are inserted to integrate the steam gasification of biomass and the hot gas cleaning system into one reactor vessel.
Regarding the tar clean-up process, the catalytic filter has been modeled as a fixed bed at cylindrical coordinates considering both the tar and hydrocarbon reforming reactions. For a given input data, the model is able to predict the longitudinal profile of tar and gases inside the filter, and obviously the tar and gas composition at the outlet of the filter. The kinetic model, together with the kinetic parameters determined, has been used to simulate the behavior of the filter under different operating conditions (face velocities or tar inlet content). In addition, the model is also useful to design new filters including other configurations or sizes.
A modelling tool based on the commercial software IPSEpro was set up for process simulation of the energy conversion system comprising gasification and application of syngas by SOFC. The model allows calculation of the mass- and energy balance regarding the gasification unit. Further, IEn has developed semi-empirical model of the solid oxide fuel cell unit. The SOFC process unit model predicts stack performance based on the inlet reactant streams, geometry of the unit, operating stack temperature and fuel utilization. Finally, a comprehensive simulation tool based on process flow sheet calculation was set up describing the gasification process. This modular tool is focused on the items actually necessary for process designers and technology users. Further, an economic calculation tool was developed for economic assessment of the gasification process. The tools are on-line available (via link through Unique homepage) with open access for potentially users. The process flow sheet tool involves the case of dual fluidized bed gasification (DFB) and bubbling fluidized bed process (BFB), the latter is available for the economic calculator. These tools are available at the following links:
- http://www.processweb.net/demos/TU-Wien/PSWeb_Flow_Sheet_UNIQUE.php
- http://www.processweb.net/demos/UNIQUE/PSWeb_UNIQUE_GE.php
- http://www.processweb.net/demos/UNIQUE/PSWeb_UNIQUE_GE_ORC.php.
Potential impact:
The webpage covers publicly available information on the project, partners, and publications. The project flyer summarizing objectives, activities and key data of the UNIQUE project is also available from the webpage and was used by the partners, e.g. in the workshop in Timisoara.
All partners produced short videos presenting their work in UNIQUE. The videos have been integrated into a PowerPoint presentation. The videos are available on the UNIQUE web page: http://www.uniqueproject.eu/video_pro.asp
On 2 July 2010, a workshop was held in Timisoara (Romania) titled 'Fluidised bed biomass gasification technologies for syngas and power generation' with the aim to disseminate the UNIQUE technology to the potential users. UNIQUE partners gave presentations on the general project, catalyst for tar reduction, chemical hot gas cleaning, biomass gasification and hot gas cleaning in one reactor vessel, the CHP plant in Guessing, dual fluidized bed steam gasification, and economic evaluation of UNIQUE technology. Despite the economical problems the region is confronting, 55 participants from research institutes (7), academia (30, including PhD students), departmental and governmental agencies (6) and companies (12) have attended the workshop. During the open discussions, 2 companies revealed their interest in applying UNIQUE fluidized bed gasification technology. The importance of a long term and stable support policies in terms of legislation and financial support have been underlined as essential for the future of biomass energy production.
Thirty-six scientific paper and proceedings relating to the foreground of the project have been published and further papers and oral presentations on conferences have already been accepted or submitted. In addition to the dissemination activities, two patents, FR0953376 PCT extension FR2010/050962 and US 2010/0223848 A1, have been developed and secured during the project.
UNIQUE computational tool has been developed to help potential users (engineers, project planners, academia) to evaluate technical and economic advantages of the UNIQUE technology. As far as industrial competitiveness is concerned, the following main UNIQUE achievements are identified and briefly summaries in the table about 'exploitable foreground' that is part of this report.
Due to the variety and complexity of the problems, capabilities and know-how, either scientific or technological, which were necessary for implementation of the technology, the project aims have been fulfilled by the active collaboration of all the members of this consortium.
Project website: http://www.uniqueproject.eu
Contacts:
UNIQUE project coordinator (UNIVAQ)
Foscolo Prof. Pier Ugo
Department of Chemistry, Chemical Engineering and Materials, Faculty of Engineering
Address: Monteluco di Roio,L'Aquila -I 67040 Italy
Phone: +39-086-2434001, +39-334-6488206
Fax: +39-086-2434203
E-mail: info@uniqueproject.eu
BKG
Reinhard Koch
Biomasse Kraftwerk Güssing GmbH & Co KG
Address: Europastrasse 1 A-7540 Güssing Austria
Phone: +43-332-2901085031
E-mail: r.koch@eee-info.net
CSIC
Juan Adanez
Carboquimica-Dept. Energy & Environment
Address: M. Luesma Castán 4, ZARAGOZA 50018, Spain
Phone: +34-976-733977
Fax: +34-976-733318
E-mail: jadanez@icb.csic.es
ENEA
Braccio Dr. Giacobbe
Ter-EneBio
Address: SS Jonica 106, Km 419 + 500 - Rotondella (Mt) - I-75026 Italy
Phone: +39-083-5974387, +39-083-5974387
Fax: +39-083-5974210
E-mail: giacobbe.braccio@trisaia.enea.it
FZJ
Michael Mueller
Institute for Energy Research, IEF-2
Address: Leo-Brandt-Strasse D - 52425 Juelich German
Phone: +49-246-1616812
Fax: +49-246-1613699
E-mail: mic.mueller@fz-juelich.de
IEN
Janusz Jewulski
Thermal Processes Department
Address: ul. Augustowka 36 - Warszawa 02-981 Poland
Phone: +48-223-451410, +48-604-355472
Fax: +48-226-428378
E-mail: janusz.jewulski@ien.com.pl
PALL
Steffen Heidenreich
Werk Schumacher Crailsheim
Address: Zur Flügelau 70 - Crailsheim 74564 - Germany
Phone: +49-795-1302172
E-mail: steffen.heidenreich@europe.pall.com
TUV
Christoph Pfeifer
Institute of Chemical Engineering
Address: Getreidemarkt 9/166 A-1060 Vienna Austria
Phone: +43-158-80115951
E-mail: cpfeifer@mail.zserv.tuwien.ac.at
ULP
Alain Kiennemann
Laboratoire Materiaux,Surface, Procédés pour la Catalyse
Address: Becquerel 25 - Strasbourg 67087, France
Phone: +33-390-242766
Fax: +33-390-242768
E-mail: kiennemann@chimie.u-strasbg.fr
UPT
Teodor Todinca
Applied Chemistry, Organic and Natural Compounds Engineering
Address: Victoriei Square 2 - Timisoara 300006 Romania
Phone: +40-256-403078, +40-746-394811
Fax: +40-256-403060
E-mail: teodor.todinca@chim.upt.ro
A new catalyst for in-bed primary reduction of heavy hydrocarbons (tar) was prepared and characterized by increasing the iron content of olivine (a natural mineral substance). This catalyst was demonstrated effective to reduce tar content by 45 % and increase gas yield by 40 % (when compared to olivine), in bench and pilot scale (100 kWth) gasification tests. Catalyst production on large scale (1 ton) was also developed. This very low-cost material removes completely the problem of heavy metals in the ashes to be disposed.
Innovative sorbents to be added to the gasifier fluidized bed were synthesized and tested successfully: sulphur, chlorine and alkali were reduced below the threshold limits required to feed a solid oxide fuel cell (H2S and HCl below 1 ppmv; KCl below 100 ppbv). The SOFC performance obtained with the syngas from steam gasifier in Güssing and in Trisaia was comparable or better than that measured with a reference fuel.
Tar reforming catalytic filter elements were optimized to produce high-performance catalytic systems. Scale-up of the whole procedure allowed the manufacture of commercial-size catalytic candles, with guaranteed filtration properties (particle removal efficiency higher than 99.9 % of particulate present in the raw syngas). They were tested at bench scale (0.5 kg/h biomass feeding rate) and a reduction of up to 80 % of tar content in the producer gas was obtained by comparison with a blank test. Tar abatement increased up to 92 % when Fe/olivine catalyst and catalytic candle were combined together.
Catalytic filtration tests performed at Güssing industrial plant (8 MWth) did confirm high potential for hot gas cleaning, although further work is needed to prove the overall technical feasibility and the long term behavior remains an outstanding issue.
Finally, the hardware modifications needed to implement the UNIQUE hot gas cleaning and conditioning technology at the Trisaia pre-existing gasification rig were fully designed. The 1 MWth pilot plant in ENEA Research Centre was properly arranged to permit the insertion of filter candles inside the gasification reactor; however, because of delays in plant construction activities, significant tests were not carried out in the contractual period.
Project context and objectives:
In existing gasification installations, gas cleaning is normally done by filtration and scrubbing of the producer gas: in this way the clean gas is made available at temperatures close to ambient, and the most immediate option for power generation is gas engine. Such process configuration does not allow high electric conversion efficiencies: reported values are close to 25 % that is what is also obtainable with modern combustion plants coupled with steam turbines. This penalizes notably the overall economic balance of the plant, which would benefit of a higher share of electricity against heat production.
The project goal is a compact version of a gasifier integrating the fluidized bed steam gasification of biomass and the hot gas cleaning and conditioning system into one reactor vessel. This is obtained by placing a bundle of catalytic ceramic candles operating at a temperature as high as the gasification temperature (800 - 850 °C) in the gasifier freeboard; furthermore, by using a catalytically active mineral for primary tar reforming and sorbents into the bed for removal of detrimental trace elements.
A major gas contaminant is solid particulate entrained by the syngas. They can block gas passages and/or the anode surface of a fuel cell, and they are one of the major responsible of the local air pollution. The tolerance limit of high temperature fuel cell to solid particulate is of the order of a few mg/Nm3.
Tars (heavy hydrocarbons) also originate from gasification, and are contained in the producer gas. They condensate at temperatures lower than 400 °C, plugging gas passages and downstream equipment. The presence of tar among the products of gasification reduces gas yield and conversion efficiency. These contaminants are also responsible of carbon deposition that can plug the porous media of a fuel cell anode. The tolerance limit of high temperature fuel cell to tar is still not well defined in the literature (due to the novelty of this fuel cell application); a reasonable value is of the order of 100 ppmw.
Sulphur compounds are generated by the gasification process because of the presence of a limited amount of sulphur in the biomass. The most important is H2S, responsible of chemisorption on Ni surfaces that blocks active sites, in the catalytic gas conditioning system and in the fuel cell utilized for power generation. The tolerance limit of high temperature fuel cell to sulphur compounds is of the order of a few ppmv, to allow operation below 1000 °C.
Fuel bound chlorine in biomass is released as HCl during gasification. It is highly corrosive, especially against the interconnect material of the SOFC.
Alkali metal contaminants are released as chlorides during gasification. They can lead to ash softening and melting and thus plugging of the filter. By condensing in cooler parts, they can block gas passages and / or the anode surface of a fuel cell, and thus they are one of the major influencing factors of fouling.
Fuel-bound nitrogen in biomass is principally released into the gasification gas primarily as ammonia and only a small part as HCN. NH3 has no relevant effect to the fuel cell if it's lower than 1 % by volume (gasification raw gas usually contains a lower concentration of ammonia). When this gas is combusted, NH3 has the tendency to form nitrogen oxides (NOx ), which are pollutants difficult to remove, and precursors of 'acid rain'.
This project deals with the above mentioned problems by means of an integrated approach to hot gas cleaning and conditioning that includes a set of scientific and technical objectives itemized below.
An innovative catalytic system is developed for in-bed primary reduction of heavy hydrocarbons by means of a natural mineral substance (olivine) where the iron content is increased up to 20 % by impregnation. In comparison with previously obtained Ni/olivine catalysts, this very low-cost material removes completely the problem of heavy metals in the ashes to be disposed, and will assure similar reforming activity due to its higher content of iron.
Tar reforming catalytic filter elements are optimised by screening new catalyst supports at laboratory scale, with regard to the adjustment of a high BET surface area at high temperature, by studying the catalytic activation process, and producing a high-performance catalyst system.
The whole procedure is scaled-up to allow the manufacture of commercial-size catalytic candles, with characterised filtration properties (ceramic candles are able to guarantee particle removal efficiency higher than 99.9 % of particulate present in the raw syngas).
Synthetic sorbents to be added properly to the gasifier are chosen and characterized, to trap sulphur compounds and additional, detrimental trace elements in order to reduce their content in the hot gas to values compatible with downstream units, e.g. a high temperature fuel cell. Sulphur capture also reduces or eliminates deactivation by H2S of the catalytic filter.
The feasibility of the integrated arrangement proposed for the gasification reactor is checked experimentally, and the performance of the gas conditioning and cleaning system at real process conditions properly quantified, by means of a thorough test campaign at bench scale, in a bubbling fluidized bed gasifier with nominal load of 1 kg/h of biomass feedstock.
In addition, the novel catalyst and the synthetic sorbents are tested in the 100 kWth FICFB (dual fluidized bed steam blown biomass gasifier) pilot plant (Vienna University of Technology (TUV)) in comparison to olivine and Ni/olivine.
An industrial-scale benchmark of the effectiveness of the gas cleaning and conditioning system is obtained by housing a catalytic filter candle inside a commercial gasifier (Güssing plant, 8 MWth), to compare the producer gas quality in the slip stream passing through the innovative conditioning system, with that of the existing low temperature gas filtering and scrubbing system. Long duration tests to study aging or other time depending effects were also planned.
A pilot scale unit (1 MWth) designed according to the principles illustrated above is also operated. Its realisation is done by modifying the freeboard of an existing oxygen / steam, bubbling bed gasifier, in order to insert the candles needed for particulate filtration and tar reforming. The reactor is already equipped with a weir, over which the overload of fine and light particles, detached from the filter surface and floating on the bed surface, would flow to an adjacent chamber from which it may be withdrawn.
The feasibility of feeding the clean fuel gas finally obtained to a high efficient power generation system is tested by means of a bench scale, solid oxide fuel cell. As a result of laboratory work, a portable SOFC unit, suitable for tests with the producer gas on its anode side, is assembled and arranged to be connected to a slip stream of the commercial and pilot scale gasifier, respectively.
Modelling studies are carried out at different scales. The complex heterogeneous reacting system inside the filtering medium is simulated, at contact times typically shorter than one second: catalytic reactions are favoured by relatively low velocities in the filter and the influence of fine structure of the porous material for optimal operation needs to be studied in details, as well as that of a large pore size, characterising the filter material, which on one hand reduces mass transfer limitations, on the other hand reduces also the specific active surface. A CFD model of the reactor freeboard is also implemented, to optimise candles configuration in it. Process simulations of the whole energy conversion chain allow to characterise the process flow-sheet and its overall thermal and chemical efficiency.
Project results:
A summary description is provided of activities and major results obtained in the UNIQUE project.
'Catalyst and sorbents for in-bed primary reduction of tar and trace contaminants'
Characterisation and development at industrial scale of Fe / olivine catalysts was performed, to improve syngas yield and the hydrogen concentration, and to decrease tar production in the gasification of biomass in a fluidised bed reactor. This new bed material satisfies also attrition resistance, low cost, no toxicity and simplicity of the preparation procedure.
After analysis of the existing literature. University of Strasbourg (ULP) decided to focus on an iron salt impregnated on the olivine support. The key of the success was the conditioning of the material (calcination at high temperature and in-situ reduction) to develop a strong interaction between the iron metal or the iron oxides particles and the olivine support. The best choice for the formulation has been a olivine based catalytic material containing 10 wt% of iron, calcined between 900 and 1100 °C and reduced in situ by the gasification gas.
The catalysts have been fully characterized by several techniques of bulk and surface: X-ray diffraction (XRD), Mössbauer spectroscopy, X-ray photoelectrons spectroscopy (XPS), scanning electron microscopy (SEM) and temperature programmed reduction (TPR). These techniques showed that olivine structure was maintained whatever the thermal treatment. The different oxidation states of iron and the environment of iron particles were determined at different calcination conditions. Diffusion of iron oxides inside the grains has been evidenced at high calcination temperatures. At the surface, the amount of iron III is also related to the calcination temperature. The reduction temperature and the amount of reducible iron (iron reducibility) have been quantitatively determined.
The 10 % Fe/olivine catalyst has been tested at different operating conditions in order to identify the active sites and to optimize its activity in toluene (T) and 1 methylnaphthalene (1-MN) reforming. The results show clearly the greater efficiency of iron / olivine catalyst compared to olivine, and a good activity compared to Ni/olivine.
In addition, ULP studied the scaling up of the catalyst preparation at the scale of kg then ton. Finally, 300 kg and 700 kg have been produced for pilot plant tests at Vienna University of Technology (TUV) and ENEA Trisaia Research Centre (Italy), respectively.
Additional research activities were addressed to the choice and characterization of synthetic sorbents to be properly added to the gasifier to trap detrimental trace elements like H2S, HCl, alkali metals and heavy metals in order to reduce their content in the hot produced gas to levels compatible with a high temperature fuel cell. The interactions between ashes, sorbents, and filter candle materials was also investigated.
Sorbents for alkali and sour gas (i.e. H2S and HCl) removal were tested regarding their sorption efficiency and capacity as fixed-bed in lab-scale furnaces. Furthermore, the capability of different sour gas and alkali sorbents for co-sorption of heavy metals, i.e. Zn, Cd, and Pb was investigated. For the sorption experiments a gas stream was passed through a sorbent fill with a length of 50 mm which corresponded to a weight of about 25 g. The temperature of the sorbent bed ranged from 700-900 °C. In order to determine the composition of the cleaned gas leaving the sorbent fill the tube furnace was connected with a molecular beam mass spectrometer (MBMS).
Bauxite, kaolinite and n.o. zeolite are suitable for KCl reduction to values below 100 ppbv. Slag lime, as the best tested 'conventional' sulphur sorbent only reduced the H2S concentration to 50 ppmv at 800 °C. A new stabilised Ba-based sorbent developed at Forschungszentrum Jülich (FZJ) achieves H2S concentrations below 1 ppmv at temperatures higher than 760 °C. This should be sufficient to prevent poisoning of fuel cell materials. Furthermore, the stabilised Ba-based sorbent is suitable to reduce HCl below 1 ppmv. Zink was best absorbed by aluminosilicates which are already used for alkali sorption limiting the Zn concentration to 400 ppb. Lead was also best absorbed by aluminosilicates with kaoline limiting the Pb concentration to 200 ppb. Cadmium could not effectively be removed by any of the tested sour gas and alkali sorbents at an inlet concentration of 1-2 ppm. Furthermore, the fact that neither the alkali nor the sour gas removal is kinetically limited at a space velocity of 9800 h-1 shows the suitability of these sorbents for the product gas stream of the Güssing gasifier.
In-bed tar reforming catalyst (primary tar reduction) and the synthetic sorbents were tested at process conditions representing the industrial application of dual fluidized bed (DFB) steam gasification. Experimental investigations highlighting the effect of Fe-olivine, sulphur and alkali sorbent in the DFB biomass gasification system were carried out at bench scale (10 kWth fuel input) and pilot plant scale (100 kWth). A comprehensive process parameter study was accomplished to emphasise the effect of the in-bed catalyst resp. sorbents on the gasification process performance. Further, reference tests with silica sand and olivine were carried out. Thus, an extensive mapping of the process performance was developed including the scale up effects. Furthermore, the influence of oxygen transport on tar conversion caused by cyclic oxidation-reduction of the Fe was experimentally investigated and assessed at a dual circulating fluidized bed reactor (DCFB) system at pilot rig scale (100 kWth fuel power).
The bench scale unit at the Consejo Superior de Investigaciones Científicas (CSIC) was designed and operated at conditions as similar as possible to the 100 kWth dual fluidized bed gasification plant located at TUV. This unit consists of two interconnected fluidized beds (gasifier and combustor) with solid material circulating between them. The unit has a nominal capacity of about 300 g/h. Pine wood delivered by TUV, after crushing and sieving, was used as biomass for the gasification tests.
The bench scale tests have been useful to determine the effect of the main operating conditions, such as gasification temperature and steam / fuel ratio, on the tar primary reduction.
Silica sand, olivine and Fe-olivine were applied as bed material at the pilot scale unit. Table 3 highlights a summary and comparison, respectively for gasification temperature of 850 °C. The specified tar content is related to GC/MS measurement. The results for Fe-olivine are related to the long term experiment carried out with used Fe-olivine. Application of silica sand is taken as bench mark value for comparison as this material is suggested to be inert in terms of catalytic activity. Beside the differences in composition of the product gas, the tar content is strongly varying. Decreasing tar content is found in the order Silica sand > olivine > Fe-olivine. Thus, Fe-olivine is identified as best promoting the tar reduction by its use as in-bed catalyst for primary tar reduction. Further, the experiments at the DCFB reactor system revealed the oxygen transport capacity of Fe-olivine due to cyclic reduction and oxidation. This characteristic is also developed in the DFB system. The oxygen transport causes partially oxidation of product gas in the gasifier and contributes to the heat demand in the gasifier.
'Catalytic filter candles'
Nickel containing tar reforming catalytic filter elements for hot gas cleaning and conditioning were developed from laboratory to commercial scale, by Pall Filtersystems GmbH Werk Schumacher (PALL) in collaboration with the University of Strasbourg (ULP) and the Consejo Superior de Investigaciones Científicas (CSIC). Validation of the most catalytically active catalytic filter element was performed in the freeboard of a bench-scale gasifier (University of L'Aquila (UNIVAQ)), before testing a catalytic filter candle of commercial scale in the freeboard of the Güssing gasifier (Biomasse Kraftwerk Güssing GmbH & Co KG - BKG).
A screening of activity of several catalytic layer systems led to the development of an active catalytic layer system consisting of a MgO-Al2O3 supported Ni catalyst. The catalyst was examined on its catalytic activity under model gas conditions using toluene and 1-methyl-naphthalene as model tars at ULP as well as under real gas conditions at UNIVAQ.
After this screening phase an alternative catalytic filter design was examined to increase the catalytic performance. In this fixed catalyst bed design the catalyst is integrated in grain form into the annular space between the filter candle and a porous inner tube. Appropriate samples were tested at CSIC for real gas validation.
A catalytic filter candle segment of 400 mm length was tested at UNIVAQ integrated in the freeboard of the bench-scale gasifier, with 80 % average tar conversion, over 22 hours on test. Scale-up of the catalytic filter candle of fixed bed design was performed to provide a 1.5 m long catalytic filter candle prototype for real gas testing.
With respect to operating temperatures 850 °C in the freeboard of the Güssing gasifier, a new Al2O3 based catalytic filter candle was developed from laboratory to commercial scale (DeTarCat CL) by transfering the most catalytically active layer on this new filter material.
CSIC has tested the activity behaviour of different catalytic filter elements delivered by PALL (s. Figure 4).
In a test campaign at UNIVAQ with simultaneous utilization of Fe / olivine catalyst in the fluidized bed of the gasifier and the catalytic filter candle in its freeboard (0.5 kg/h biomass feed rate), the gas yield increased on average by 75 % and the hydrogen yield by 152 %. Correspondingly, the methane and tar content in the gas was reduced by 30 % and 92 %, respectively, and tar production per kg of dry ash free (daf) biomass by 86 %.
The filter candles were directly integrated into the freeboard of the Güssing gasifier. Thus, industrial scale performance is investigated since the unpurified raw gas is charged to the filter candle.
A filter candle test module was erected including the direct mounting of the filter candle into the freeboard of the gasifier. The test module comprises the main units:
- fixture unit of the filter candle with the freeboard construction;
- back pulsing/dedusting unit (pressure vessel, dedusting valve);
- pipe system equipped with trace heating;
- control unit (for back pulsing, volume flow, pipe heating, pump control, temperature, pressure);
- pump, filter and condensate trap unit;
- gas sampling.
Stable filtration of raw product gas was not accomplished for the type DeTarCAT FB and DeTarCAT CL due to attack of SiC structure caused by temperature peaks. A test run with the type DeTarCAT CL-Al was successfully accomplished with a stable filtration period of about 28 hours.
Based on these results, a preliminary functional and operability analysis allows highlighting that the filter candle positioned inside the gasifier freeboard shows high potential for hot gas cleaning, while further work is needed to prove the overall technical feasibility, and the long term behavior remains an outstanding issue. As a matter of fact, complications were experienced in matching the requirements of the UNIQUE experimentation with the shut-down time intervals of a plant operating commercially. This resulted in limitations for the number of affordable tests, although some important results were obtained as pointed out below. UNIQUE partners recognize the need of further testing activity, more precisely about the characterization of interactions between the candles and the gasifier environment, from the point of view of materials, average temperature, temperature hot spots, etc. However this type of study would require an experimental plant (not a commercial one).
In addition, focussing on the gasifier temperature and the temperature control is partially correct. In particular the temperature control is meant for levelling the heat balance in the gasifier. Especially, during the start-up and in case of particular requirements (charging of wet biomass), the temperature in the gasifier might exceed the range of temperature (of 800-850 °C as indicated in the DoW). The feasibility and functionality of the original envisaged filter candles material (SiC) has been proven at the lab-scale device, whereby moderate temperatures and steady temperature levels are applied. However, it has been shown that the requirements at industrial scale (fluctuations in temperature, impact of real process gas with dust and char load, and mechanical impacts, e.g. vibrations) differ. The combination of temperature and present atmosphere which surrounds the filter candle, have revealed that other materials have to be applied to resist and to preserve the chemical stability. Thus, the application of Al2O3 as suitable filter candle material was the main finding gathered during the test runs and in consultation with the evaluation of the applied filter candle material. In line with this outcome, the tests were an important part for the development of properly operating filter candles.
The 28 h hot gas filtration test can be regarded as the key test for an assessment of the practicability of the developed catalytic filter on industrial scale and in part of the UNIQUE concept, even if the full demonstration is still to be realized. Additional results e.g. the tar reforming activity and tar analysis will be the objective of a future publication by TUV in collaboration with PALL, so that key results about combined particle separation and catalytic tar reforming under gasifier freeboard conditions with commercial-sized catalytic filter candles can be taken into consideration for all developers active in this field.
'The UNIQUE prototype pilot plant'
The 1 MWth pilot plant in ENEA Trisaia Research Centre was properly modified to permit the insertion of filter candles inside the gasification reactor.
On the basis of the input data given by ENEA about the expected gas flow rate (400 Nm3/h) and the original freeboard geometry, PALL proposed the preliminary tube sheet design. The tube sheet was sized in order to accommodate 75 filter candles, distributed in 5 rows of 15 candles, with 70 mm outer diameter and 1520 mm length. ENEA prepared the layout concerning all hardware modifications: the head of the reactor, the nearby pipeline, and the baffle welded to the freeboard wall.
In addition, ENEA analysed the gas emission risks at the plant. The assessment was carried out according to the current Italian standard (CEI 3135, CEI 3130) and the European directives (94/9/EC, 99/92/EC). Twelve main sources of gas emissions were identified and all hazardous areas were classified as zones of low risk (zone 2). A detailed description of the risk analysis was provided.
Due to possible risks of pore blocking of the filter candles during initial operation of the plant, PALL planned to first deliver 75 non-catalytic filter candles, and provide the catalytic ones after the proper and reliable functioning of the modified pilot plant was proved by means of preliminary gasification tests.
Regarding the realisation of the pilot plant, some of the previously described reactor modifications were implemented during the 3th and 4th semester. The length of the head was increased to 54 cm, while the baffle, originally welded to the reactor wall, was made removable so as to allow the proper housing of the candles. Additional sampling points were arranged.
Some strength calculations were undertaken in ENEA. In particular, for the tube sheet, a structural linear / nonlinear and thermal analysis was done (using ANSYS® software). It was necessary to choose a tube sheet 50 mm thick, made of A310S steel, a high temperature resistant steel.
Because of substantial delays in the prototype construction phase, the original planning of the gasification test campaign could not find adequate time in the contractual period, and most of it will be carried out by ENEA after the end of the project. The first phase of the experimental campaign will be addressed to test the performance of the steam / O2 gasification process when using the bed catalytic material (10 % Fe/Olivine) and the sorbents, however without the hot catalytic filter system. After this first campaign, the new filter system, including catalytic and non-catalytic candles, will be installed in the gasifier and the long running test, devoted to evaluate the performance of the whole UNIQUE technology, will be carried out.
To partially overcome these difficulties and to gather experimental evidence useful for operation of the 1MWth gasifier, bench scale experimental activities (10 kWth ICBFB gasifier available in Trisaia) were included in the original project programme, with a contract amendment procedure. The design of this reactor is similar to that of the UNIQUE prototype, and it also operates with oxygen and steam gasification agents.
Gasification tests were carried out by using 10 % Fe / Olivine and bauxite. Changes in process conditions induced by key parameters, such as ER and S/B ratios, were also evaluated. The raw syngas was characterised in terms of gas composition, organic & inorganic contaminants, particulate content and process efficiency (gas yield, carbon and water conversion, LHV). In order to assess the efficacy of the new active phases, towards the removal of tar and alkali halides, the results were then compared to those obtained by using natural olivine as bed material. A significant reduction of about 40 %wt was observed for the tar load while, due to the greater brittleness of the new catalytic material, an increase of the entrained particulate was observed, which became twice as much.
As far as the alkali halides removal is concerned, the positive effect of bauxite was confirmed, mainly in terms of KCl removal, although less extended than expected on the basis of the lab tests.
'Operation of a solid oxide fuel cell (SOFC) with biomass gas'
A portable SOFC test station was designed and assembled to allow fuel stream tests with both pilot scale and commercial scale gasifiers using cleaned product gas from the biomass gasifier as a fuel supply. It consists of air supply and filtering system, fuel humidification system, electric furnace, air and fuel pre-heating systems, instrumentation and data acquisition system, performance test equipment (electronic load) and SOFC assembly. The gas manifolding and valve system allows for different gases to be supplied to fuel manifolds during the start-up, operation and shut-down of the test station. In the standard operating mode, test station is fuelled with syngas supplied from the biomass gasification unit.
Tests of the SOFC station components have been completed to evaluate gas leaks, performance, measurement and control range accuracy for the fuel pump, mass flow rate controllers, mass flow meters, gas cooler, gas humidifier, air and fuel heaters and electric furnace. After validation of the component units' performance and construction of the portable SOFC station, preliminary tests of the unit have been completed. The test station was operated in a full range of fuel flows. Temperatures of the gas streams at the selected process check-points have been consistent with expectations. Onsite test requirements, for the operation of portable test station with the gasifier, are limited to availability of utilities (electric, technical gases, water, and ventilation system).
Preliminary off-site tests with the SOFC cells fuelled with the BKG (Guessing, Austria) syngas compressed in gas cylinders have been completed at the Institute of Power Engineering (Warsaw, Poland). Onsite tests have been performed at the ENEA Research Centre ( Trisaia, Italy) and BKG.
During the first series of tests at ENEA Research Centre, after initial polarization measurements, the cell was operated under electric load of 4 Amps. Lower than expected cell performance was observed. After cell disassembly, heavy carbon deposits were observed on the current collector surface. Anode surface remained mostly carbon free. The results have been attributed to humidification temperature of 55 ºC, which was too low for carbon-free cell operation. In the second series of tests, syngas humidification was kept at 65 ºC to avoid carbon deposition. Performance of the SOFC cell was much better than in the first series of tests and post-test SOFC cell inspection did not show any cell cracks. Minor carbon deposits were visible on the current collector surface facing the gas supply side. Following performance tests, cell was operated under electric load.
Similar test were performed at BKG biomass gasification plant in Guessing. During the 26 hour test, no cell performance decline was observed.
The results of tests performed at IEn, ENEA and BKG indicate that cleaned syngas from biomass gasification reactor is a feasible fuel for the anode supported SOFC cell. Cell performance on cleaned syngas is comparable or better than cell performance on 47.5 % H2+47.5 % N2+3 % H2O gas mixture. Proper humidification of the gas stream, necessary to avoid carbon deposition, is critical for the long term cell performance (carbon deposition). The presence of larger amounts of methane and higher hydrocarbons may lead to SOFC cell cracking due to large temperature gradients resulting from internal reforming process.
The product gas from gasifier contains impurities (H2S, HCl, particulates, tars, alkali metals, etc.). This syngas stream must be cleaned before it can be used to fuel SOFC. Experimental setup to investigate the effect contaminants in the fuel on the SOFC cell performance was designed, assembled and tested. In the experiment, fuel components (H2, N2) were supplied from gas cylinders, mixed, humidified and supplied to SOFC cell. The effect of contaminants, containing chlorine in form of HCl(g) and sulfur in form of H2S, on the SOFC performance has been investigated at 7500 C and 8000 C. The cell performance was evaluated using constant current load at the varying contaminant levels. The effect of contaminants is reflected by the lower cell voltage.
Anode supported SOFC cells were operated for a total of around 500 hours at 800 °C with H2S levels ranging from 0.1 to 2 ppm. Performance tests indicate negligible short term effect (around 200 hrs.) of up to 1 ppm H2S in the fuel at 800 °C. Significant voltage decline has been recorded at 2 ppm H2S. Another series of tests were performed at 750 °C with H2S concentrations ranging from 0.3 ppm to 1.5 ppm. The effect of H2S on the cell performance is evident for 1.5 ppm H2S concentration in the fuel at 750 C. The voltage decline of 150 mV has been recorded after 100 hours of operation with the 47.5 % H2 + 47.5 % N2 + 3 % H2O + 1.5 ppm H2S fuel. Two distinctive log-linear regions of voltage decline are visible after H2S is introduced to the fuel stream: faster and shorter region, followed by the slower voltage decline region.
Following the H2S tests, HCl(g) effect on the SOFC cell performance was examined in a series of experiments. No significant effect was observed during the tests for HCl(g) concentrations ranging from 1 ppm up to 10 ppm at 750 ºC for both dry fuel and fuel humidified at 25 ºC. The voltage decline of 16 mV has been recorded after over 60 hrs of cell operation with 10 ppm HCl contamination. The temperature effect of HCl on the anode performance is less significant than H2S effect while cell response to HCl contamination is fast relatively to H2S.
'Accompanying modelling activities at different scales'
Modelling and numerical simulation activities, carried out in strict connection with the experimental investigations, are conceived as a means to better understand either potentiality or limits of the developed systems when applied to a wide range of conditions. The calculation tools used inside UNIQUE are useful for design and optimization of the gas cleaning and conditioning system here proposed, as well as for the optimization of the integrated energy conversion chain.
Modelling activities inside UNIQUE were developed at different scales (micro and macro systems) and related to different expertises (thermodynamics, chemical reaction kinetics, fluid dynamics, plant design, and energy systems).
The thermodynamic process model, designed using SimuSageTM (GTT-Technologies), allows calculation of the type and amount of gaseous and condensed trace elements in parts of the UNIQUE plant downstream the hot gas cleaning. The results of the thermodynamic process model were obtained by Gibbs free energy minimisation. The variable parameters of the model are the inlet streams like biomass, steam and oxygen, the gasifier temperature and the sorbents.
The processes of solids filtration and tar reforming taking place inside the catalytic filter candle were also modelled. A computational fluid dynamic (CFD) tool was developed and validated against experimental findings, able to simulate the fluid dynamic behavior of gas and solid particles inside the fluidized bed freeboard of the UNIQUE prototype, where filtration candles are inserted to integrate the steam gasification of biomass and the hot gas cleaning system into one reactor vessel.
Regarding the tar clean-up process, the catalytic filter has been modeled as a fixed bed at cylindrical coordinates considering both the tar and hydrocarbon reforming reactions. For a given input data, the model is able to predict the longitudinal profile of tar and gases inside the filter, and obviously the tar and gas composition at the outlet of the filter. The kinetic model, together with the kinetic parameters determined, has been used to simulate the behavior of the filter under different operating conditions (face velocities or tar inlet content). In addition, the model is also useful to design new filters including other configurations or sizes.
A modelling tool based on the commercial software IPSEpro was set up for process simulation of the energy conversion system comprising gasification and application of syngas by SOFC. The model allows calculation of the mass- and energy balance regarding the gasification unit. Further, IEn has developed semi-empirical model of the solid oxide fuel cell unit. The SOFC process unit model predicts stack performance based on the inlet reactant streams, geometry of the unit, operating stack temperature and fuel utilization. Finally, a comprehensive simulation tool based on process flow sheet calculation was set up describing the gasification process. This modular tool is focused on the items actually necessary for process designers and technology users. Further, an economic calculation tool was developed for economic assessment of the gasification process. The tools are on-line available (via link through Unique homepage) with open access for potentially users. The process flow sheet tool involves the case of dual fluidized bed gasification (DFB) and bubbling fluidized bed process (BFB), the latter is available for the economic calculator. These tools are available at the following links:
- http://www.processweb.net/demos/TU-Wien/PSWeb_Flow_Sheet_UNIQUE.php
- http://www.processweb.net/demos/UNIQUE/PSWeb_UNIQUE_GE.php
- http://www.processweb.net/demos/UNIQUE/PSWeb_UNIQUE_GE_ORC.php.
Potential impact:
The webpage covers publicly available information on the project, partners, and publications. The project flyer summarizing objectives, activities and key data of the UNIQUE project is also available from the webpage and was used by the partners, e.g. in the workshop in Timisoara.
All partners produced short videos presenting their work in UNIQUE. The videos have been integrated into a PowerPoint presentation. The videos are available on the UNIQUE web page: http://www.uniqueproject.eu/video_pro.asp
On 2 July 2010, a workshop was held in Timisoara (Romania) titled 'Fluidised bed biomass gasification technologies for syngas and power generation' with the aim to disseminate the UNIQUE technology to the potential users. UNIQUE partners gave presentations on the general project, catalyst for tar reduction, chemical hot gas cleaning, biomass gasification and hot gas cleaning in one reactor vessel, the CHP plant in Guessing, dual fluidized bed steam gasification, and economic evaluation of UNIQUE technology. Despite the economical problems the region is confronting, 55 participants from research institutes (7), academia (30, including PhD students), departmental and governmental agencies (6) and companies (12) have attended the workshop. During the open discussions, 2 companies revealed their interest in applying UNIQUE fluidized bed gasification technology. The importance of a long term and stable support policies in terms of legislation and financial support have been underlined as essential for the future of biomass energy production.
Thirty-six scientific paper and proceedings relating to the foreground of the project have been published and further papers and oral presentations on conferences have already been accepted or submitted. In addition to the dissemination activities, two patents, FR0953376 PCT extension FR2010/050962 and US 2010/0223848 A1, have been developed and secured during the project.
UNIQUE computational tool has been developed to help potential users (engineers, project planners, academia) to evaluate technical and economic advantages of the UNIQUE technology. As far as industrial competitiveness is concerned, the following main UNIQUE achievements are identified and briefly summaries in the table about 'exploitable foreground' that is part of this report.
Due to the variety and complexity of the problems, capabilities and know-how, either scientific or technological, which were necessary for implementation of the technology, the project aims have been fulfilled by the active collaboration of all the members of this consortium.
Project website: http://www.uniqueproject.eu
Contacts:
UNIQUE project coordinator (UNIVAQ)
Foscolo Prof. Pier Ugo
Department of Chemistry, Chemical Engineering and Materials, Faculty of Engineering
Address: Monteluco di Roio,L'Aquila -I 67040 Italy
Phone: +39-086-2434001, +39-334-6488206
Fax: +39-086-2434203
E-mail: info@uniqueproject.eu
BKG
Reinhard Koch
Biomasse Kraftwerk Güssing GmbH & Co KG
Address: Europastrasse 1 A-7540 Güssing Austria
Phone: +43-332-2901085031
E-mail: r.koch@eee-info.net
CSIC
Juan Adanez
Carboquimica-Dept. Energy & Environment
Address: M. Luesma Castán 4, ZARAGOZA 50018, Spain
Phone: +34-976-733977
Fax: +34-976-733318
E-mail: jadanez@icb.csic.es
ENEA
Braccio Dr. Giacobbe
Ter-EneBio
Address: SS Jonica 106, Km 419 + 500 - Rotondella (Mt) - I-75026 Italy
Phone: +39-083-5974387, +39-083-5974387
Fax: +39-083-5974210
E-mail: giacobbe.braccio@trisaia.enea.it
FZJ
Michael Mueller
Institute for Energy Research, IEF-2
Address: Leo-Brandt-Strasse D - 52425 Juelich German
Phone: +49-246-1616812
Fax: +49-246-1613699
E-mail: mic.mueller@fz-juelich.de
IEN
Janusz Jewulski
Thermal Processes Department
Address: ul. Augustowka 36 - Warszawa 02-981 Poland
Phone: +48-223-451410, +48-604-355472
Fax: +48-226-428378
E-mail: janusz.jewulski@ien.com.pl
PALL
Steffen Heidenreich
Werk Schumacher Crailsheim
Address: Zur Flügelau 70 - Crailsheim 74564 - Germany
Phone: +49-795-1302172
E-mail: steffen.heidenreich@europe.pall.com
TUV
Christoph Pfeifer
Institute of Chemical Engineering
Address: Getreidemarkt 9/166 A-1060 Vienna Austria
Phone: +43-158-80115951
E-mail: cpfeifer@mail.zserv.tuwien.ac.at
ULP
Alain Kiennemann
Laboratoire Materiaux,Surface, Procédés pour la Catalyse
Address: Becquerel 25 - Strasbourg 67087, France
Phone: +33-390-242766
Fax: +33-390-242768
E-mail: kiennemann@chimie.u-strasbg.fr
UPT
Teodor Todinca
Applied Chemistry, Organic and Natural Compounds Engineering
Address: Victoriei Square 2 - Timisoara 300006 Romania
Phone: +40-256-403078, +40-746-394811
Fax: +40-256-403060
E-mail: teodor.todinca@chim.upt.ro
