Calcium-based sorbents behaviour during sulphation at oxy-fuel fluidised bed combustion conditions

Sulphur capture by calcium-based sorbents is a process highly dependent on the temperature and CO₂ concentration. In oxy-fuel combustion in fluidised beds (FB), CO₂ concentration in the flue gas may be enriched up to 95%. Under so high CO₂ concentration, different from that in conventional coal combustion with air, the calcination and sulphation behaviour of the sorbent must be defined to determine the optimum operating temperature in the FB combustors.In this work, the SO₂ retention capacity of two different limestones was tested by thermogravimetric analysis at typical oxy-fuel conditions in FB combustors. The effect of the main operating variables affecting calcination and sulphation reactions, like CO₂ and SO₂ concentrations, temperature, and sorbent particle size, was analysed.It was observed a clear difference in the sulphation conversion reached by the sorbent whether the sulphation takes place under indirect or direct sulphation, being much higher under indirect sulphation. But, in spite of this difference, for a given condition and temperature, the CO₂ concentration did not affect to the sulphation conversion, being its major effect to delay the CaCO₃ decomposition to a higher temperature.For the typical operating conditions and sorbent particle sizes used in oxy-fuel FB combustors, the maximum sorbent sulphation conversions were reached at temperatures of about 900 °C. At these conditions, limestone sulphation took place in two steps. The first one was controlled by diffusion through porous system of the particles until pore plugging, and the second controlled by the diffusion through product layer. As a consequence, the maximum sulphation conversion increased with decreasing the particle size and increasing the SO₂ concentration.     

The effect of water on the sulphation of limestone

A series of tests was conducted in a thermogravimetric analyzer (TGA) to study the sulphation behaviour of limestone in the presence of water over the temperature range of 800-850 °C. Four different Canadian limestones, all with a particle size range of 75-425 μm, were sulphated using a synthetic flue gas with a composition of 15% CO₂, 3% O₂, 0% or 10% H₂O, 1750 ppm SO₂ and the balance N₂. Water was shown to have a significant promotional effect on sulphation, especially in the diffusion-controlled stage. However, the effect of water during the kinetic-controlled stage appeared to be much less pronounced. Based on these results, it is proposed that the presence of water leads to the transient formation of Ca(OH)₂ as an intermediate, which in turn reacts with SO₂ at a faster rate than CaO does. Alternatively stated, it appears that H₂O acts as catalyst for the sulphation reaction of CaO.     

Characteristics and reactivities of Ca(OH)2/silica fume sorbents for low-temperature flue gas desulfurization

Ca(OH)₂/silica fume sorbents were prepared with various Ca(OH)₂/silica fume weight ratios and slurrying times at 65°C and a water/solid ratio of 10/1. Dry sorbents prepared were characterized, and their reactivities toward SO₂ were measured in a differential fixed-bed reactor at the conditions similar to those in the bag filters in the dry and semidry flue gas desulfurization (FGD) processes. The reaction between Ca(OH)₂ and silica fume in the slurry was very fast. The formation of calcium silicate hydrates, which were mainly C-S-H(I), resulted in sorbent particles with a highly porous structure that seemed compressible under high pressures. The sorbents were mesoporous, and their specific surface areas and pore volumes were much larger than those of Ca(OH)₂ alone. The utilization of Ca of sorbent increased with increasing silica fume content mainly due to the increase in the specific surface area of sorbent. The sorbent with Ca(OH)₂ had the maximum SO₂ capture. Sorbents with Ca(OH)₂ contents less than and greater than would have a SO₂ capture greater than that of Ca(OH)₂ alone. Both the utilization of Ca and SO₂ capture per unit specific surface area of sorbent decreased in general with increasing specific surface area. At the same Ca(OH)₂ content, the utilization of Ca or SO₂ capture of the Ca(OH)₂/silica fume sorbent was greater than that of the Ca(OH)₂/fly ash sorbent; however, the amount of SO₂ captured per unit surface area of the former sorbent was smaller than that of the latter sorbent. The results of this study are useful to the preparation of silica-enhanced sorbents for use in the dry and semidry FGD processes.     

Assessment of ettringite from hydrated FBC residues as a sorbent for fluidized bed desulphurization

The performance of synthetic ettringite as a sorbent in fluidized bed desulphurization has been assessed and compared with that of a commercial limestone. Experiments have been carried out in a bench scale fluidized bed reactor under simulated desulphurizing (steadily oxidizing) combustion conditions. Sorbent performance has been characterized in terms of desulphurization rate, maximum sulphur uptake and attrition propensity. Fluidized bed sulphation experiments have been complemented by microstructural characterization of solid samples, accomplished via X-ray diffraction analysis, scanning electron microscopy and sulphur mapping of cross-sections of particles embedded in epoxy resin.

Experimental results show that both the rate and the maximum extent of sulphur uptake by ettringite significantly exceed those of the limestone. Maximum degree of free calcium utilization is 0.58 for ettringite compared with 0.27 for the limestone. Sulphation tests also indicate that attrition propensity of ettringite is larger than that correspondingly observed for the limestone. Microstructural characterization indicates that sulphation of ettringite takes place evenly throughout the particle cross-section, whereas sulphation of limestone mostly conforms to a core-shell pattern.

Along a parallel pathway, the rate and yield of ettringite formation by hydration of fly ash from a utility fluidized bed boiler have been assessed. Formation of ettringite in these experiments appears to be quantitative upon curing of ash at 70 °C for times up to 4 days.     

Sulphation and carbonation properties of hydrated sorbents from a fluidized bed CO2 looping cycle reactor

Sulphation and carbonation have been performed on hydrated spent residues from a 75 kW_th dual fluidized bed combustion (FBC) pilot plant operating as a CO₂ looping cycle unit. The sulphation and carbonation tests were done in an atmospheric pressure thermogravimetric analyzer (TGA), with the sulphation performed using synthetic flue gas (0.45% SO₂, 3% O₂, 15% CO₂ and N₂ balance). Additional tests were carried out in a tube furnace (TF) with a higher SO₂ concentration (1%) and conversions were determined by quantitative X-ray diffraction (QXRD) analyses. The morphology of the sulphated samples from the TF was examined by scanning electron microscopy (SEM). Sulphation tests were performed at 850 °C for 150 min and carbonation tests at 750 °C, 10 cycles for 15 min (7.5 min calcination + 7.5 min carbonation). Sulphation conversions obtained for the hydrated samples depended on sample type: in the TGA, they were ～75–85% (higher values were obtained for samples from the carbonator); and in the TF, values around 90% and 70% for sample from carbonator and calciner, respectively, were achieved, in comparison to the 40% conversion seen with the original sample. The SEM analyses showed significant residual porosity that can increase total conversion with longer sulphation time. The carbonation tests showed a smaller influence of the sample type and typical conversions after 10 cycles were 50% – about 10% higher than that for the original sample. The influence of hydration duration, in the range of 15–60 min, is not apparent, indicating that samples are ready for use for either SO₂ retention, or further CO₂ capture after at most 15 min using saturated steam. The present results show that, upon hydration, spent residues from FBC CO₂ capture cycles are good sorbents for both SO₂ retention and additional CO₂ capture.     

Sorption-enhanced water gas shift reaction by sodium-promoted calcium oxides

The water gas shift reaction was evaluated in the presence of novel carbon dioxide (CO₂) capture sorbents, both alone and with catalyst, at moderate reaction conditions (i.e., 300-600 °C and 1-11.2 atm). Experimental results showed significant improvements to carbon monoxide (CO) conversions and production of hydrogen (H₂) when CO₂ sorbents are incorporated into the water gas shift reaction. Results suggested that the performance of the sorbent is linked to the presence of a Ca(OH)₂ phase within the sorbent. Promoting calcium oxide (CaO) sorbents with sodium hydroxide (NaOH) as well as pre-treating the CaO sorbent with steam appeared to lead to formation of Ca(OH)₂, which improved CO₂ sorption capacity and WGS performance. Results suggest that an optimum amount of NaOH exists as too much leads to a lower capture capacity of the resultant sorbent. During capture, the NaOH-promoted sorbents displayed a high capture efficiency (nearly 100%) at temperatures of 300-600 °C. Results also suggest that the CaO sorbents possess catalytic properties which may augment the WGS reactivity even post-breakthrough. Furthermore, promotion of CaO by NaOH significantly reduces the regeneration temperature of the former.     

Removal of SO2 in Semi‐Dry Flue Gas Desulfurization Process with a Powder‐Particle Spouted Bed

Semi‐dry flue gas desulfurization was investigated with several kinds of SO₂ sorbents, such as slaked lime, limestone, Mg(OH)₂ and concrete pile sludge, in a powder‐particle spouted bed. Slurry droplets including sorbent fine particles were fed to a spouted bed of coarse inert particles spouted with hot gas containing SO₂. SO₂ removal efficiency was strongly affected by the approach to saturation temperature, Ca/S molar ratio and particle size of sorbent. Slaked lime showed the highest desulfurization efficiency. In this process, despite very short gas residence time, more than 90% SO₂ removal was easily achieved by choosing appropriate conditions.     

SO2 retention at low temperatures by Ca(OH)2-derived CaO: a model for CaO regeneration

The retention of SO₂ by Ca(OH)₂-derived CaO has been studied at 573 K. The thermal treatment of Ca(OH)₂ forms CaO sorbents with high activity to retain SO₂. At the temperature of 573 K, it has been proved that SO₂ interacts with CaO to form a surface CaSO₃ specie.The regeneration process by thermal decomposition of CaSO₃, giving SO₂ and CaO, has been examined using techniques such as in situ FTIR and in situ XRD. In addition, CO₂ chemisorption and SO₂ sorption and sorbent regeneration was evaluated by thermogravimetry.The thermal regeneration process used to decompose CaSO₃ causes an important loss in the SO₂ retention capacity of the sorbent. Using several experimental procedures, a series of reactions have been analysed (Ca(OH)₂ decomposition, formation of superficial CaSO₃, change from surface to bulk CaSO₃ and CaSO₃ decomposition) and an insight into the regeneration process has been obtained.The decrease in SO₂ retention, following the thermal regeneration step, is due to both an increase in the particle size and the conversion of CaSO₃ into CaSO₄ and CaS (CaSO₃ disproportionation) upon heat treatment. The increase in the particle size is responsible for about 88% of the loss of activity, whereas CaSO₃ disproportionation explains a 12% of loss.     

Preparation of diatomite/Ca(OH)2 sorbents and modelling their sulphation reaction

Mixtures of Ca(OH)₂ and diatomite were hydrated at different conditions to produce reactive SO₂ sorbents. Two different hydration techniques were used; namely, atmospheric and pressure hydration. The effect of the hydration temperature, time and diatomite/Ca(OH)₂ weight ratio on the physical properties of the activated sorbents were investigated. In atmospheric hydration, it was found that increasing the temperature and hydration time caused an increase in the total surface area of the sorbents. However, surface area values of the sorbents prepared from mixtures which have different diatomite/Ca(OH)₂ weight ratio were generally not changed significantly. In pressure hydration, the surface area of the activated sorbents was positively affected from the hydration temperature and pressure. Finally, Ca(OH)₂ and two diatomite/Ca(OH)₂ sorbents were sulphated at constant temperature using a synthetic gaseous mixture consisting of 5% O₂, 10% CO₂, and the balance of nitrogen with a 55% relative humidity. The sulphation reaction of these sorbents were investigated and modelled. The unreacted shrinking core model was chosen to describe this non-catalytic solid/gas (hydrated sorbent/SO₂) reaction mechanism. The experimental results were found to be correlated successfully by this model.     

Kinetics of the reaction of iron blast furance slag/hydrated lime sorbents with SO2 at low temperatures: effects of sorbent preparation conditions

Sorbents highly reactive towards SO₂ have been prepared from iron blast furnace slag and hydrated lime under different hydration conditions. The reaction of the dry sorbents with SO₂ has been studied under the conditions similar to those in the bag filters in the spray-drying flue gas desulfurization system. The reaction was well described by a modified surface coverage model which assumes the reaction rate being controlled by chemical reaction on sorbent grain surface and takes into account the effect of sorbent Ca molar content and the surface coverage by product. The effects of sorbent preparation conditions on sorbent reactivity were entirely represented by the effects of the initial specific surface area (S_g0) and the Ca molar content (M⁻¹) of sorbent. The initial conversion rate of sorbent increased linearly with increasing S_g0, and the ultimate conversion increased linearly with increasing S_g0M⁻¹. The initial conversion rate and ultimate conversion of sorbent increased significantly with increasing relative humidity of the gas. Temperature and SO₂ concentration had mild effects on the initial conversion rate and negligible effects on the ultimate conversion.     

Characteristics and so2 capture capacities of sorbents prepared from products of spray-drying flue gas desulfurization

The characteristics and the SO₂ capture capacities of sorbents prepared from products of spray-drying flue gas desulfurization (FGD) have been studied. Sorbents were prepared by first slurrying Ca(OH)₂ and CaSO₃ and/or CaSO₄ with and without the addition of fly ash and then drying. Compared to the use of pure Ca(OH)₂, the SO₂ capture and Ca(OH)₂ utilization decreased for sorbents prepared without fly ash and increased for sorbents with fly ash. Flakelike ill-crystallized tobermorites were observed for all the sorbents containing fly ash. In addition, significant amounts of needle-shape Ca₄Al₂(OH)₁₂SO₄ . 6H₂O and Ca₆Al₂ (OH)₁₂(SO₄)₃ . 26H₂O (ettringite) were also observed for the sorbents containing CaSO₃ and/or CaSO₄. These newly formed compounds dissociated into CaSO₃ . 0.5 H₂O and inert precursors upon sulfation, and were responsible for the high SO₂ capture capacities and Ca(OH)₂ utilizations of the sorbents prepared with fly ash.     

Sequential SO2/CO2 capture enhanced by steam reactivation of a CaO-based sorbent

The steam hydration reactivation characteristics of three limestone samples after multiple CO₂ looping cycles are presented here. The CO₂ cycles were performed in a tube furnace (TF) and the resulting samples were hydrated by steam in a pressure reactor (PR). The reactivation was performed with spent samples after carbonation and calcination stages. The reactivation tests were done with a saturated steam pressure at 200 °C and also at atmospheric pressure and 100 °C. The characteristics of the reactivation samples were examined using BET and BJH pore characterization (for the original and spent samples, and samples reactivated under different conditions) and also by means of a thermogravimetric analyzer (TGA). The levels of hydration achieved by the reactivated samples were determined as well as the conversions during sulphation and multiple carbonation cycles. It was found that the presence of a CaCO₃ layer strongly hinders sorbent hydration and adversely affects the properties of the reactivated sorbent with regard to its behavior in sulphation and multiple carbonation cycles. Here, hydration of calcined samples under pressure is the most effective method to produce superior sulphur sorbents. However, reactivation of calcined samples under atmospheric conditions also produces sorbents with significantly better properties in comparison to those of the original sorbents. These results show that separate CO₂ capture and SO₂ retention in fluidized bed systems enhanced by steam reactivation is promising even for atmospheric conditions if the material for hydration is taken from the calciner.     

Breakthrough and Sulfate Conversion Analysis during Removal of Sulfur Dioxide by Calcium Oxide Sorbents

Sorption of sulfur dioxide (SO₂) was carried out on calcium‐based sorbents under dynamic conditions in a fixed bed. The experimental conditions were reaction temperature (700 to 1000°C), SO₂ concentration (1000‐10 000 ppm), sorbent particles size (1 to 2 mm) and the types of sorbents (hydroxide or carbonate). The sorption process was found to be effective at low concentration levels (less than 10 000 ppm) as the breakthrough time significantly decreased with increase in concentration. The maximum removal of SO₂ was observed at a reaction temperature of 950°C. The hydroxide‐based sorbents of relatively smaller particle size were found to exhibit superior sorption performance in terms of longer breakthrough time and higher sulfate conversion. A mathematical model developed, assuming a porous structure of the sorbent materials, attributed the low sulfation conversion during SO₂ sorption due to a pore diffusion mechanism.     

Reinvestigation of hydration/reactivation characteristics of two long-term sulphated limestones which previously showed uniformly sulphating behaviour

Two Canadian limestones, Calpo and Luscar, were fully sulphated, and the residues were hydrated with liquid water and steam and then re-sulphated with synthetic flue gas in a thermogravimetric analyzer (TGA). Both sulphated limestones were previously classified as showing uniform sulphation patterns, and it was expected that they would not demonstrate significant reactivation by hydration. However, the current work demonstrates that both spent sorbents can be reactivated by steam hydration, while one of them, Luscar limestone, can also be reactivated by hydration with liquid water, whereas water hydration is less effective with Calpo limestone. Long-term sulphation was employed on the fresh limestones to ensure that the sorbents were fully sulphated to levels typical of full-scale units during the first sulphation. No evidence was found that the SO₂ concentration for first sulphation influenced the degree of reactivation, indicating that these sulphation times are sufficient. Total calcium utilization after re-sulphation was markedly improved – up to 90% by hydration. Possible explanations for the failure to reactivate these limestones by previous workers may well be that they chose unsuitable hydration conditions and/or that there are wide variations in limestone properties between different batches, even from the same supplier. It is also evident that it may be premature to categorize the sulphation patterns of a given limestone on the basis of limited tests.     

Mineralogical and engineering characteristics of dry flue gas desulfurization products

Fifty-nine coal combustion products were collected from coal-fired power plants using various dry flue gas desulfurization (FGD) processes to remove SO₂. X-ray diffraction analyses revealed duct injection and spray dryer processes created products that primarily contained Ca(OH)₂ (portlandite) and CaSO₃·0.5H₂O (hannebachite). Most samples from the lime injection multistage burners process contained significant amounts of CaO (lime), CaSO₄ (anhydrite), and CaCO₃ (calcite). Bed ashes from the fluidized bed process were often dominated by CaSO₄ but also contained CaCO₃, CaO (lime), and MgO (periclase). Cyclone ashes were similar in composition to the bed ashes but contained more unspent sorbent and CaSO₄ and less MgO. Fly ash in all samples ranged from 10 to 79 wt%. Samples usually exhibited two distinct swelling episodes. One occurred immediately after water was applied due to hydration reactions, especially the conversion of CaO to Ca(OH)₂ and CaSO₄ to CaSO₄·2H₂O (gypsum). The second began between 10 and 50 d later and involved formation of the mineral ettringite (Ca₆[Al (OH)₆](SO₄)₃·26H₂O). The final pH after 112 d ranged from 10.0 to 12.1. If samples are incubated under ‘closed’ (i.e. incomplete recarbonation with atmospheric CO₂) and alkaline weathering conditions, gypsum and portlandite are initially formed followed by the conversion of the gypsum to ettringite. Closed, alkaline conditions typically can occur when FGD products are placed in confined settings such as a road embankment or buried as a discrete layer as occurs in some surface mine reclamation projects.     

The effects of limestone type on the sulphur capture of slaked lime

This study examines the effect of the chemical composition and origin on the performance of two calcitic and two dolomitic limestones from different sources in South Africa. The experiments were carried out in a fixed bed reactor maintained at 80 °C. The raw sorbent materials were calcined at 900 °C and the resulting quicklime hydrated to produce the relevant hydrates which were used in the tests. Results obtained show that the maximum temperature rise during the hydration of the samples varied from 5 to 65 °C depending on the chemical composition of the sorbent. Sorbents with higher temperature rise resulted in products with a more porous structure and a better performance in the sulphur capture. The maximum sorbent conversion in terms of mol of SO₂ per mol of sorbent varied from 0.0274 for dolomitic limestones to 0.1823 for the calcitic limestones. The presence of Fe₂O₃ in small quantities was observed to have a positive effect on the performance of the sorbent.     

Sulphur dioxide removal using South African limestone/siliceous materials

This study presents an investigation into the desulfurization effect of sorbent derived from South African calcined limestone conditioned with fly ash. The main aim was to examine the effect of chemical composition and structural properties of the sorbent with regard to SO₂ removal in dry-type flue gas desulfurization (FGD) process. South African fly ash and CaO obtained from calcination of limestone in a laboratory kiln at a temperature of 900 °C were used to synthesize CaO/ash sorbent by atmospheric hydration process. The sorbent was prepared under different hydration conditions: CaO/fly ash weight ratio, hydration temperature (55-75 °C) and hydration period (4-10 h). Desulfurization experiments were done in the fixed bed reactor at 87 °C and relative humidity of 50%. The chemical composition of both the fly ash and calcined limestone had relatively high Fe₂O₃ and oxides of other transitional elements which provided catalytic ability during the sorbent sorption process. Generally the sorbents had higher SO₂ absorption capacity in terms of mol of SO₂ per mol of sorbent (0.1403-0.3336) compared to hydrated lime alone (maximum 0.1823). The sorbents were also found to consist of mesoporous structure with larger pore volume and BET specific surface area than both CaO and fly ash. X-ray diffraction (XRD) analysis showed the presence of complex compounds containing calcium silicate hydrate in the sorbents.     

Characterization of the reactivity of limestones with SO2 in a fluidized bed reactor

The reaction between SO₂ and calcined limestone particles has been studied in a fluidized bed combustor. Measurements of sorbent reactivity with SO₂ were made for small batches of limestone injected into the combustor. Simultaneous continuous combustion of bituminous coal provided conditions like those of a boiler for study of the sulphation reaction. A semi-empirical rate model of the CaO-SO₂ reaction has been developed. External mass transfer of SO₂, diffusion within the particles and chemical reaction are taken into account. The limestone reactivity with SO₂ is characterized by two parameters which are dependent on the temperature and sorbent particle size. A model for predicting the limestone requirements in a fluidized bed boiler has been developed. Parameters from the batch experiments are included. The predictions for sulfur retention agree with the experimental results. In addition, effects of operating conditions (gas velocity, recycle, limestone particle size) on the retention of SO₂ were simulated using the model.     

HYDROTHERMAL REACTION OF FLY ASH/HYDRATED LIME: CHARACTERIZATION OF THE REACTION PRODUCTS

Class F coal fly ash was slurried with hydrated lime at 90°C in 1/3, 5/3, 9/3, and 15/3 weight ratios and for 3, 5, 7, and 9 hours of hydration, in a process to prepare sorbents for SO₂ removal. The amounts of aluminum, silicon, and calcium in the product of the pozzolanic reaction were determined in order to study the evolution of product composition with the initial raw materials ratio and hydration time and to relate this composition to the desulfurization capability of the material. Al, Si, and Ca were present in the solid product for any raw materials ratio and hydration time, showing that calcium silicates, calcium aluminates, and/or calcium aluminum silicates were obtained simultaneously. The products formed show a nearly constant molar ratio of Al₂O₃/SiO₂ independent of the experimental conditions tested and similar to the Al₂O₃/SiO₂ ratio in the fly ash. The SiO₂/CaO molar ratio in the products decreased as the initial fly ash/Ca(OH)₂ ratio decreased, being approximately constant for each ratio with respect to hydration time after 5 hours of hydration. The maximum moles of CaO, SiO₂, and Al₂O₃ per gram of sorbent in the reaction product were found for any hydration time for the 5/3 sorbents, meaning that at this initial ratio the pozzolanic reaction takes place at the highest rate. The capacity of the sorbent for SO₂ removal depends not only on the amount of products produced by the pozzolanic reaction but also on the specific surface area of the sorbent.