Experimental studies of ash film fractions based on measurement of cenospheres geometry in pulverized coal combustion

Abstract

In pulverized coal particle combustion, part of the ash forms the ash film and exerts an inhibitory influence on combustion by impeding the diffusion of oxygen to the encapsulated char core, while part of the ash diffuses toward the char core. Despite the considerable ash effects on combustion, the fraction of ash film still remains unclear. However, the research of the properties of cenospheres can be an appropriate choice for the fraction determination, being aware that the formation of cenospheres is based on the model of coal particles with the visco-plastic ash film and a solid core. The fraction of ash film X is the ratio of the measuring mass of ash film and the total ash in coal particle. In this paper, the Huangling bituminous coal with different sizes was burnt in a drop-tube furnace at 1273, 1473, and 1673 K with air as oxidizer. A scanning electron microscope (SEM) and cross-section analysis have been used to study the geometry of the collected cenospheres and the effects of combustion parameters on the fraction of ash film. The results show that the ash film fraction increases with increasing temperature and carbon conversion ratio but decreases with larger sizes of coal particles. The high fraction of ash film provides a reasonable explanation for the extinction event at the late burnout stage. The varied values of ash film fractions under different conditions during the dynamic combustion process are necessary for further development of kinetic models.

Keywords

ash film fraction / cenospheres / coal combustion / fly ash

Introduction

The concern about high efficiency and clean utilization of pulverized coal combustion has motivated the investigation on coal combustion characteristics. In the complete process of particle combustion, the char oxidation reaction is considered as the slowest step [¹,²], especially in the burnout stage with a carbon conversion ratio of more than 70%. The observation of the reactivity loss and extinction phenomena in the burnout stage has been widely reported by Hurt et al. [³] and Murphy and Shaddix [⁴]. Most kinetic models focusing on the main proportion of carbon conversion provide a good prediction at early times (0%–70%), but a poor description at the later burnout stage [⁵,⁶]. The deviation from experimental measurements of char burnout is due to the incomplete treatment of ash effects [³,⁷]. The ash inhibition effect increases the oxygen diffusion resistance to the char particle when the particle is encapsulated by ash, especially at the char burnout stage.

Ash minerals show complex transformation behaviors. The dispersed mineral matters are liberated in combustion, part of which act as a layer on the encapsulated coal particle and pose an additional resistance to oxygen transport to the reacting surface [⁷–¹⁰], while the rest of which penetrate back into the carbon-rich core [⁴,⁸,¹⁰]. The ash surrounding the carbon-rich core is an explanation for the char conversion rate reduction at the char burnout stage and is the so-called ash film. The diffusion of liberated ash back into the core is modeled as the ash dilution process, which is reported to have a significant effect on the apparent reaction order in char combustion [⁴].

The considerable effect of ash film has been proposed by Chen and Kojima [¹¹] that the combustion of char particle with high ash content is mainly ash film diffusion controlled. Hurt et al. [³] have developed a carbon burnout kinetic model (CBK) to predict the extinction event at the burnout stage. The results suggest that the ash inhibition effect is the primary reason. Recently, taking the combination of the ash effect into account, an intrinsic kinetic model has been developed to quantitatively reveal the complex ash behavior [⁸,¹²,¹³]. The relative deviations of burnout time obtained by comparing simulation with no ash film formation are found between -32% and+13% under different combustion conditions [¹⁰]. However, the detailed information on the ash fraction distributed as the ash film still remains obscure. For example, in CBK, all the ash liberated is assumed to form an ash film and the fraction value is constant, which is inappropriate considering the above-mentioned dilution effect. Besides, the experimental measurement of the fractions of ash film is difficult due to the irregular ash structure and the high combustion temperature.

The research on the properties of cenospheres can be an appropriate choice for fraction determination, since the formation of cenospheres is based on the model of the coal particle with the visco-plastic ash film and a solid core [¹⁴,¹⁵], and its regular morphological form enables a simple measurement method. Approximately 1 wt%–2 wt% of the fly ash are cenospheres [¹⁶]. The coarsely dispersed minerals in the coal particle melt, coalescence, and form a visco-plastic shell cover [¹⁷]. The core is formed of less easy-to-melt components and unburnt carbon. Decomposition of the components in the closed ash film would lead to the swelling of the particle under the increasing internal gas pressure. The swelling can result in the hollow cenosphere after the cooling stage. However, when the gas pressure exceeds the critical value, the hollow sphere blows up into many fragments [¹⁸,¹⁹]. In some cases, the carbon residual prevents the coalescence of encapsulated ash particles inside the fused ash envelopment, which results in the formation of plerosphere, as concluded by Raask [²⁰]. The formation of plerosphere can be explained by the delay in the melting of different layers of ash [²¹].

In this study, the Huangling bituminous coal with different sizes was burnt in a drop-tube furnace at 1273, 1473, and 1673K in air atmosphere. The fly ash collected was mixed with resin and underwent fine polishing, scanning electron microscope (SEM), and cross-section analysis to distinguish the cenospheres from other ash particles and study the geometry of the cenospheres. The ash film fraction was then calculated by the self-derived approach. The aim of this paper is to provide calculated values of ash film fraction under different conditions for further development of kinetic models, by means of the effects of temperature, particle size and carbon conversion ratio.

Experimental methods

Experimental setup

Huangling bituminous coal, a common coal in Chinese power plants, was dried for 24 h in an oven at 105°C before being sieved in three size fractions of 61–75, 75–90, and 90–125 mm for the following combustion. The particle size distribution was measured by a particle-size analyzer (OMEC LS-909), whose results are provided in Fig. 1 while the coal properties were summarized in Table 1. The coal has an ash content of 13.63 wt%, which is appropriate for ash film measurement, and a comparatively high volatile content.

After drying, the coal particles with three size fractions were burnt at 1273 K, 1473 K, and 1673 K under air conditions in a drop-tube furnace, as shown in Fig. 2. The pulverized coal particles were carried by 300 mL/min of N₂ from the micro feeder, with a feeding rate of 70 mg/min, before the gas-solid mixture was injected into the water-cooled feeding probe at the top of the furnace. The 3 L/min gas flow of 21 vol.% O₂ within N₂ was introduced into the furnace to assist the combustion process. The residual ash passed through a quenched collecting probe at the bottom of the furnace, and was collected by a collector, where a glass fiber filter with a pore size of 0.4 mm was used. To reveal the change of ash film fraction with carbon conversion (85%–90%, 90%–95%,>95%), various residence times were employed by locating the relative positions between the feeding probe and the collecting probe. The carbon conversion ratios of the particles collected at different residence times were tested by a thermogravimetric analyzer (TGA) (STA409PC, NETZSCH, Germany).

Sample characterization using SEM

The cross-section study of the cenospheres was conducted from the SEM images to analyze their physical structure parameters, such as morphology, particle diameter and thickness. A series of steps were performed before SEM [²²]. The ash particles collected were mixed with resin at 4 wt% before pouring into a cylindrical mold, and was left to solidify for 24 h. After that, the resin mixture was polished by abrasive papers from coarse to fine (400, 800, and 1200 grade). At the last stage of polishing, a small amount of micropolish gel (grain size of 0.5 mm) was poured onto the paper for the fine and smooth polishing of the sample. The samples polished were prepared to identify their cross-section and distinguish the cenospheres from other ash particles, since the polished section of cenosphere was hollow while that of fly ash was solid. The mass fraction of the cenosphere particle in the total ash particles is 1 wt%–2 wt%, but its number fraction should be much higher due to the low cenosphere density.

Noteworthy is that the random cross-section analysis makes it impossible to determine the true size of a cenosphere particle, since polishing does not always intercept the particle center. Therefore, stereological corrections are employed according to Ngu et al. [²³], which are essential for the further study of the ash film fraction. For a cenosphere particle with a measured size smaller than the observed average size, it is likely to be intercepted by polishing at the top or bottom part, as illustrated in Fig. 3(a). Therefore, the measuring results lead to an underestimated outer diameter but an overestimated wall thickness, making it impossible to accurately calculate the ash film fraction.

To avoid this, in the corrections, if the outer diameter of a particle is smaller than the average outer diameter D_out, its true outer diameter is assumed to be the average value. The average diameter for different combustion conditions is obtained from SEM images by the measurement of piles of ash collected from the corresponding condition. The true inner diameter is derived as

(1)D_{\mathrm{in}}=\sqrt{D_{\mathrm{o}\mathrm{u}\mathrm{t}}^2-M_{\mathrm{o}\mathrm{u}\mathrm{t}}^2+M_{\mathrm{i}\mathrm{n}}^2},

where M_out and M_in are the measured outer and inner diameters of the ash cenosphere particle, as depicted in Fig. 3(a). The correction results can be divided into overcorrection, correction, and non-correction. When the true outer diameter of the particle is smaller than the average diameter, its true outer diameter is still assumed to be the average value, which apparently causes correction errors as overcorrection. The calculated errors of cenospheres with a mean size of 47 mm produced from 75 to 90 mm coal particles were plotted in Fig. 3(b). The results show that the errors remain below 20% for ash cenospheres with sizes of 41–64 mm. The average wall thickness t of the same particle is calculated as

(2)t={\displaystyle \frac{D_{\mathrm{out}}-D_{\mathrm{in}}}{\mathrm{2}}}.

According to the above discussion, ash cenospheres are formed by various layers of ash. The regular morphology of cenosphere ash particle makes it possible for simple calculation on the fraction of ash film. Therefore, the ash film fraction X is the ratio of the measuring mass of ash film and the total ash in coal particle, according to

(3)X={\displaystyle \frac{(D_{\mathrm{o}\mathrm{u}\mathrm{t}}^3-D_{i\mathrm{n}}^3){\rho }_{\mathrm{ash}}(\mathrm{1}-{\theta }_{\mathrm{ash}})}{D_{\mathrm{c}}^3{\rho }_{\mathrm{c}}NY}},

where q_ash is the porosity of ash film, D_c is the coal diameter, N is the carbon conversion ratio, and Y is the ash content (13.63 wt% from Table 1). The true density of solid ash for cenospheres is 2.0–2.4 g/cm³ [²⁰]. Therefore, the ash film density r_ash takes values of 2.1. The coal density r_c is 1.3 g/cm³, and q_ash takes the value of 0.3 (from the suggested porosity range of 0.3–0.5 [³]).

Results and discussions

Morphology and ash film thickness of cenospheres

To find out the outer and inner diameters of ash cenospheres, the cross-section was studied from the SEM images of samples polished, as demonstrated in Fig. 4. The outer and inner diameters of the cenospheres selected from each combustion condition were analyzed by Image-Pro Plus 6.0. Considering sometimes the non-uniform film thickness, the outer and inner diameters were determined based on the covering area of the cenosphere and the hollow respectively. The calculated data were illustrated in Fig. 5. The diameters obtained were corrected by the method proposed and the mean values were calculated for the determination of ash film fraction.

Figure 5 exhibits the relationship between the ash film thickness and the diameter of ash cenospheres. Although the particle size of cenospheres spans widely, the wall thickness has a relatively narrow range from 2 to 10 mm. The cenosphere wall thickness broadly increases with an increase in cenosphere diameter. The wall thickness is closely related to the surface tension forces in the shell and further swelling can lead to the shell destruction. At the same time, the mean diameter of cenospheres appears to increase with the size of coal particles, as indicated in Fig. 5. The coarsely dispersed minerals on the surface of the coal particle melt and coalescence. The larger coal particle forms an intact shell with a greater size, and consequently an ash cenosphere with a larger diameter after cooling stage.

Results of ash film fraction

Based on the measurement of the true outer and inner diameters of ash cenospheres, the ash film fraction could be calculated according to Eq. (3). However, the complex formation progress of cenospheres inevitably leads to the fact that the ash film fraction largely depends on several factors, including the chemical composition of ash [²¹] and the combustion conditions.

Effect of temperature on ash film fraction

The ash film fraction of ash cenospheres obtained in air atmosphere at different temperatures is illustrated in Fig. 6. For the coal particles with the sizes of 61–75 mm, the ash film fraction slightly increases with increasing temperature, while for those of 75–90 and 90–125 mm, the ash film fractions upsurge from 0.53/0.52 to 0.96/0.88 with the temperature increasing from 1273 K to 1673 K. The high ash film fraction at 1673 K indicates that the majority of ash is distributed as ash film at char burnout stage with a carbon conversion ratio of more than 95%. However, the low fraction values suggest that the assumption in CBK that all the liberated ash forms the ash film, i.e., X = 1, is inappropriate. The error bars are relatively wide because sometimes the coal particle size of the cenosphere measured is not equal to the mean coal particle size. Besides, some ash cenospheres possess irregular shape during the combustion, which leads to the occurrence of measuring error.

The formation of ash film has been explained by the melting of different ash layers, and the ash melting points depend on the chemical composition [²⁴]. The film is mainly formed from easy-to-melt components. The high-ferrous silicate and ferritic melts at about 1000°C–1200°C under reduction conditions, which corresponds to complex eutectics of FeO-SiO₂ (±CaO, Al₂O₃), FeO-CaO-Al₂O₃-MgO-SiO₂, and K₂O-Al₂O₃-SiO₂ (±FeO, CaO) systems [¹⁸]. However, with a further increase in temperature, the non-molten quartz particles are scavenged into the molten system and partially dissolved [¹⁹], causing an increase of ash film fractions.

Effect of coal size on ash film fraction

As displayed in Fig. 6, the size of coal particle exerts a negative effect on ash film fraction, which decreases with increasing coal particle size. Compared to the coarse coal particle, the particle with a smaller size is more combustible at a higher combustion temperature [⁸]. The refractory mineral components are easier to melt and coalesce into ash film, leading to an increase in ash film fraction. At the same time, according to the calculation method, the larger coal particle is assumed to contain more ash due to the average ash content in coal particle, and it is hard for the amount of ash increased inside the particle to completely transit to ash film.

Change of ash film fraction with carbon conversion

Based on the formation process of ash cenospheres, the ash film fraction obtained inevitably changes with carbon conversion. The effect of carbon conversion ratio on ash film fraction is distinct at varied combustion temperatures, as illustrated in Fig. 7. The ash film fraction increases from 0.48 to 0.53 at 1273 K and from 0.91 to 0.96 at 1673 K with carbon conversion. Two possible reasons are proposed. One is that an incomplete transition of components in the core continues in the char burnout stage. The core is formed by more refractory minerals and unburned carbon [¹⁴], whose further transformation involves the interrelated process with ash film. The other reason is that with carbon conversion, the refractory minerals, such as quartz particles, have more time to melt and capture by the ash film. More minerals are distributed on particle surface as ash film, consequently causing higher ash film fractions.

At the same time, it is noted that the ash film fractions at 1673 K almost double those at 1273 K, but their changes with particle size and carbon conversion are relatively small. This indicates that temperature is the primary factor influencing the ash film fraction. The different ash film fraction indicates that the fraction value needs to be changed with carbon conversion under various combustion conditions for kinetic model development.

Figure 8 presents the change of ash film fraction under different conditions in the char burnout stage. The ash is considered to uniformly distribute in the coal particle and shows evidence of partial fusion under SEM, being aware that many ash particles show glassy rather than granular surfaces. The thick ash film on the char particle surface would prevent the oxygen diffusion toward the char core and thus increases the burnout time, particularly in the late combustion stage. When the temperature increases, the ash film is thickened by the addition of molten refractory minerals and thus a higher ash film fraction is yielded. However, for the coal particle with an increased size but unchanged temperature, its ash film fraction decreases due to the possible decreased combustion temperature, despite of the possession of larger amount of ash. With carbon conversion the ash film fractions increase, but the coal particle with a small size at high temperatures still enjoys the relatively high ash film fraction in dynamic combustion process.

Conclusions

In this paper, the Huangling bituminous coal of different sizes (61–75, 75–90 and 90–125 mm) was burnt in a drop-tube furnace at 1273, 1473, and 1673 K in air atmosphere. To reveal the change of ash film fraction with carbon conversion (85%–90%, 90%–95%,>95%), various residence times were employed. The fly ash collected was mixed with resin and underwent fine polishing, SEM, and cross-section analysis, to distinguish the cenospheres from other ash particles and study their geometry. Stereological corrections were employed for further calculation of ash film fraction by using the self-derived approach. Furthermore, the effects of temperature, particle size, and carbon conversion ratio on ash film fraction were investigated.

The study of the SEM images provides essential information on physical structure of cenospheres. The cenosphere wall thickness broadly increases with an increase in cenosphere diameter. At the same time, the mean diameter of cenosphere appears to increase with the size of coal particle.

The ash film fraction exceeds 0.85 at a high temperature (1673 K), which indicates that the majority of ash is distributed as ash film in the char burnout stage with a carbon conversion ratio of more than 95%. However, the ash film fraction from coal particles 75–90 and 90–125 mm is about 0.53 at 1273 K, far less than that at 1673 K. The ash film fraction increases with increasing temperature and carbon conversion ratio but decreases with coal particles of larger sizes. The different ash film fraction indicates that the fraction value needs to be changed with carbon conversion under various combustion conditions for kinetic model development. The high fraction of ash film provides a reasonable explanation for the extinction event in the late burnout stage.

References

[1] Chen J Y, Mann A P, Kent J H. Computational modelling of pulverised fuel burnout in tangentially fired furnaces. Symposium (International) on Combustion, 1992, 24(1): 1381–1389

[2] Hottel H C, Stewart I M. Space requirement for the combustion of pulverized coal. Industrial & Engineering Chemistry, 1940, 32(5): 719–730

[3] Hurt R, Sun J K, Lunden M. A kinetic model of carbon burnout in pulverized coal combustion. Combustion and Flame, 1998, 113(1–2): 181–197

[4] Murphy J J, Shaddix C R. Effect of reactivity loss on apparent reaction order of burning char particles. Combustion and Flame, 2010, 157(3): 535–539 doi:10.1016/j.combustflame.2009.09.013

[5] Field M A. Measurements of the effect of rank on combustion rates of pulverized coal. Combustion and Flame, 1970, 14(2): 237–248

[6] Hurt R H, Mitchell R E. Unified high-temperature char combustion kinetics for a suite of coals of various rank. Symposium (International) on Combustion, 1992, 24(1): 1243–1250

[7] Cloke M, Wu T, Barranco R, Lester E. Char characterisation and its application in a coal burnout model. Fuel, 2003, 82(15–17): 1989–2000

[8] Niu Y, Shaddix C R. A sophisticated model to predict ash inhibition during combustion of pulverized char particles. Proceedings of the Combustion Institute, 2015, 35(1): 561–569

[9] Lunden M M, Yang N Y C, Headley T J, Shaddix C R. Mineral-char interactions during char combustion of a high-volatile coal. Symposium (International) on Combustion, 1998, 27(2): 1695–1702

[10] Niu Y, Liu S, Shaddix C R, Hui S. An intrinsic kinetics model to predict complex ash effects (ash film, dilution, and vaporization) on pulverized coal char burnout in air (O₂/N₂) and oxy-fuel (O₂/CO₂) atmospheres. Proceedings of the Combustion Institute, 2019, 37(3): 2781–2790

[11] Chen C, Kojima T. Single char particle combustion at moderate temperature: effects of ash. Fuel Processing Technology, 1996, 47(3): 215–232

[12] Niu Y, Wang S, Shaddix C R, Hui S. Kinetic modeling of the formation and growth of inorganic nano-particles during pulverized coal char combustion in O₂/N₂ and O₂/CO₂ atmospheres. Combustion and Flame, 2016, 173: 195–207

[13] Niu Y, Liu X, Wang S, Hui S, Shaddix C R. A numerical investigation of the effect of flue gas recirculation on the evolution of ultra-fine ash particles during pulverized coal char combustion. Combustion and Flame, 2017, 184: 1–10

[14] Fenelonov V B, Mel’gunov M S, Parmon V N. The properties of cenospheres and the mechanism of their formation during high-temperature coal combustion at thermal power plans. Kona Powder and Particle Journal, 2010, 28(0): 189–208

[15] Żyrkowski M, Neto R C, Santos L F, Witkowski K. Characterization of fly-ash cenospheres from coal-fired power plant unit. Fuel, 2016, 174: 49–53

[16] Yu J, Li X, Fleming E, Meng Z, Wang D, Tahmasebi A. Analysis on characteristics of fly ash from coal fired power stations. Energy Procedia, 2012, 17: 3–9

[17] Vassilev S V, Vassileva C G. Mineralogy of combustion wastes from coal-fired power stations. Fuel Processing Technology, 1996, 47(3): 261–280

[18] Li Y, Wu H. Ash cenosphere from solid fuels combustion. Part 1: an investigation into its formation mechanism using pyrite as a model fuel. Energy & Fuels, 2012, 26(1): 130–137

[19] Li Y, Gao X, Wu H. Further investigation into the formation mechanism of ash cenospheres from an Australian coal-fired power station. Energy & Fuels, 2013, 27(2): 811–815

[20] Raask E. Mineral Impurities in Coal Combustion: Behavior, Problems, and Remedial Measures. New York: Hemisphere, 1985

[21] Ranjbar N, Kuenzel C. Cenospheres: a review. Fuel, 2017, 207: 1–12

[22] Kolay P K, Bhusal S. Recovery of hollow spherical particles with two different densities from coal fly ash and their characterization. Fuel, 2014, 117(Part A): 118–124

[23] Ngu L L, Wu H, Zhang D K. Characterization of ash cenospheres in fly ash from Australian power stations. Energy & Fuels, 2007, 21(6): 3437–3445

[24] Niu Y, Gong Y, Zhang X, Liang Y, Wang D, Hui S. Effects of leaching and additives on the ash fusion characteristics of high-Na/Ca Zhundong coal. Journal of the Energy Institute, 2019, 92(4): 1115–1122

RIGHTS & PERMISSIONS

Higher Education Press