WP1: Long-term Pilot Testing (TUD)

This WP focused on long-term tests in an upgraded 1 MWth pilot plant. Four comprehensive test campaigns of 4 weeks each for hard coal and lignite were conducted to investigate the long-term behavior of the process focused on sorbent stability and reactivity. In-furnace measurements (gas extraction probe for in-bed gas analysis, capacitive probe for measuring solid load and velocity) were carried out to further validate the developed models with these experimental data. Solid samples extracted while testing were analyzed to complete the set of data. Additionally, a modified solids flow measuring system was tested to create a reliable method for measuring solids mass flow at the high temperatures of the process.

Experimental setup

A semi-industrial scale CCL pilot plant consisting of two interconnected CFB reactors and a combustion chamber with a thermal capacity of 1 MWth each is located at Technische Universität Darmstadt. In previous tests, the main focus was proof of operation in 1 MWth scale. Therefore, CCL test were accomplished decarbonizing synthetic flue gas, a mixture of air and CO2. The calciner was operated with oxygen enriched air in these tests not representing realistic operating conditions in terms of calcination, i.e. CO2 partial pressure in the calciner. To bring the pilot plant closer to real conditions, it was extensively upgraded to operate with coal originated flue gas from the furnace and oxy-combustion in the calciner as well as other upgrades to enhance operability and additional measurement equipment was installed. The heat release of the carbonation reaction can be extracted directly from the carbonator bed by means of five axially arranged internal cooling tubes. The flue gases are cooled down by means of two-pass heat exchangers. The entrained solid particles are removed from flue gases in bag filters. The inventory of solids in the carbonator and calciner is continuously determined by the measurement of the reactor pressure drop between the plane above the distributor and the exit of the reactor. The average composition of the solids entering and leaving the carbonator leaving is frequently determined from solid sampling using sampling ports in the loop seals and the screw conveyor. Also samples from the heat exchanger and the filters are taken.

Figure 1: Scheme of upgraded 1 MWth CCL pilot plant at Technische Universität Darmstadt

Long-term pilot testing

Four comprehensive long-term CaL test campaigns of four weeks each were conducted using hard coal either in pulverized and coarse form as well as pulverized or grained lignite, respectively. During these test campaigns, two different sorbents from western and southern Germany with different chemical compositions and particle size distributions were utilized. Semi-industrial pilot testing of the CCL process during 16 weeks with more than 1,200 hours of stable CO2 capture showed the maturity of the process realizing high CO2 capture rates in steady-state operation realizing operating points up to 60 hours. During steady-state operation, CO2 absorption rates in the carbonator higher than 90 % and overall CO2 capture rates higher than 95 % were proven under a wide range of parameters, e.g. fuel characteristics in the calciner, solid circulation rates, make-up rate etc.

A challenge of the CCL technology addressed in the pilot tests is the decreasing reactivity with increasing number of cycles of carbonation and calcination of the sorbent. Hence, a continuous feed of limestone as make-up is required in order to maintain a certain capacity of the sorbent to absorb CO2. Consequently, a realistic assessment of sorbent performance can only be provided by a homogeneous mixing of existing and continuous make-up feed. Therefore, extensive long term operating is required to achieve steady conditions of the sorbent phase. To investigate the sorbent performance, particular focus during operation was set on achieving steady state operating points, i.e. stable conditions of gas and sorbent phase. Thereby, the mean residence time of the sorbent was used to estimate the required period to achieve steady-state conditions in the solid phase. Three times the residence time of the sorbent, i.e. around 30-70 hours, ensure that at least 95 % of the inventory is exchanged by the make-up flow. Based on the screening, stable operating conditions based on the chemical composition of the sorbent samples were assessed. Extensive screening on selected solid samples has been performed based on chemical composition determined by X-ray fluorescence (XRF), the so-called loss on ignition (LoI) to quantify the amount of residual water, carbon and CO2 as well as particle size distribution (PSD). More than 400 sorbent samples collected during the different test campaigns have been analyzed and steady-state operation points have been identified for evaluation.

An exemplary operation period of 85 hours is shown in Figure 2 when two steady-state operating points were achieved. The first steady state period was reached after 40 hours leading to the absorp-tion efficiency in the carbonator of 70 %. In order to investigate the effect of an increased make-up rate, the make-up flow F0 was increased and the carbonator efficiency significantly rises until a new steady-state was reached absorbing 90 % in the carbonator. Solid analyses show a significant in-crease of the CO2 carrying capacity (Xcarb) between both steady-state periods as well as the fraction of impurities, especially sulphated material (Xsulphur) decreased. Consequently, more CO2 can be absorbed at the same operating conditions.

Figure 2: Exemplary steady-state operation from long-term pilot testing

Another focus of the investigation was the influence of the calciner fuel particle size and the fuel type on the composition of the sorbent circulation between the reactors and thus, the performance to absorb CO2 from the flue gas lead in the carbonator since an increased amount of impurities represent an inactive fraction decreasing the ability of the sorbent of absorbing CO2. Figure 3 shows the effect of the fuel type and particle size on the sorbent circulating during CaL pilot tests with the same feed ratio of make-up. Obviously, significant difference in the amount of inactive material (gypsum, ash) can be observed both depending on the particle size and on the type of fuel. Using hard coal with the same composition and different particle size distribution, it can be observed that the share of gypsum is similar but ash differs. With coarse hard coal, more ash accumulates in the circulating solid stream compared to firing with pulverized coal (see Figure 3a). The particle size of the coal feed influences the ash separation efficiency in the calciner cyclone. The bigger the ash particles, the more ash is separated and accumulates in the circulating material. A significant difference in gypsum amount can be recognized comparing the composition of the samples for pulverized hard coal as well as lignite. A rather low of share gypsum accumulates in solid circulation firing lignite compared to hard coal. The low sulphur content of the lignite with 1.8 gsulphur/MJ compared to the hard coal with 5.6 gash/MJ has a positive effect on accumulation of inert material in the process.

Figure 3: Solid composition of circulating sorbent with the same make-up rate using different fuel particle sizes (a) and different types of fuel with the same particle size (b) in the calciner

In-furnace measurements

TUD constructed and installed two probes for in-furnace measurements of solids load and velocity as well as gas composition. A capacitance probe was used for particle flow pattern measurements, and a gas extraction probe combined with a mobile gas analysis system was used for gas composition measurements. Detailed flow profiles could be recorded by horizontally traversing the probe and measuring at up to 30 positions along the reactor cross section.

An exemplary profile measurement of particle concentration and velocity profiles is shown in Figure 4. For all shown profiles, an increase of particle load towards the reactor wall is detectable in the entire reactor. There exists a clear core-annular flow structure with up-flowing particles in the reactor center and down-falling particles near the reactor wall.  The results show the prevalence of a strong core-annular flow structure at almost every measured height inside the reactors. These results were very helpful for the validation of 3D CFD tools.

Figure 4: Particle concentration profiles (a) and particle velocity profiles (b) at different heights in the carbonator

Main conclusions

Long-term pilot testing of the CCL process in 1 MWth scale was successfully conducted under a wide range of parameters under realistic operating conditions. The results show that high CO2 absorption rates were achieved during steady-state operation, which will be the basis for validation of scale-up tools. Carbonator absorption rates of 80-90 % corresponding to total CO2 capture efficiencies of 90-95 % were accomplished for periods up to 70 h in steady-state operation. Thus, the long-term pilot testing provides the experimental experience and data for reliable scale-up of the Calcium Carbonate Looping process. A broad data basis is provided for validating the developed scale-up tools in WP2, which were applied for a 20 MWth pilot plant in WP3. Crucial process parameters, e.g. make-up and sorbent looping ratio as well as fuel particle size in the calciner and temperatures in the reactors, strongly influence the engineering and could be identified. The gained experience of realistic long-time sorbent properties enables scale-up on a solid foundation the results were used to  establish the design heat and mass balance of the scaled-up plant.  The operating conditions favorable for high performance and interesting for further investigation in 20 MWth scale were based on the pilot experience. The experiences from the 1 MWth pilot plant in various fields, such as upgrade and reactor design (e.g. coal preparation and residence time in the calciner to avoid insufficient char burnout and transfer to carbonator), security reviews, measurement equipment, operating procedures etc. were transferred to WP3 as a basis for the measurement planning, the operating procedures as well as the health, safety and risk and the technical risk assessment.