Strength and Gasification Reactivity of Coke Prepared by Blending a Ca/C Composite and Coal

Abstract

In this study, a calcium/carbon composite (Ca/C) was prepared from porous CaO by the water vapor swelling method; its pores were filled with tar-derived carbonaceous material to produce high-strength and highly reactive coke. The properties of coke prepared by blending the Ca/C composite and caking-coal were then investigated. The mesopores in the swelling Ca disappeared and the crushing strength was developed by filling tar-derived carbonaceous material into the porous Ca. Thus, a Ca/C composite could be produced by the abovementioned method, wherein the tar-derived carbonaceous material and Ca species were in close proximity. When the Ca/C composite was blended into the caking-coal, the strength of the coke obtained increased by up to 50% of Ca/C blending, which made it possible to produce high-strength coke. On examining the C structure of the prepared coke by X-ray diffraction, that of the original coke was found to be almost unchanged by adding the Ca/C composite. Conversely, the Ca/C blended-coke showed higher CO₂ gasification reactivity than the original coke prepared from caking-coal. Based on the CO₂ gasification reactivity test of the demineralized coke, it was clear that an increase in the gasification reactivity of the Ca/C blended-coke depended upon the catalytic effect of Ca. Thus, this method helped produce high-strength and highly reactive coke.

1. Introduction

The iron and steel industry in Japan accounts for 14% of domestic CO₂ emissions; in particular, the steelmaking process accounts for 70% of the total CO₂ emissions.¹⁾ Although Japan’s steel industry has made progress in reducing global CO₂ emissions, improving the efficiency of reactions in blast furnaces is critical toward further reducing CO₂ emissions; such an improvement could be considered an important technology for reducing these emissions by lowering the amount of reductant used. However, the gasification reactivity of coke with CO₂ is generally low. Therefore, one of the techniques to improve the reaction efficiency of blast furnaces is by increasing the gasification reactivity of coke; hence, there have been developments involving the production of highly reactive coke.²⁾ This is to substantially accelerate the gasification reactivity of coke with CO₂, thereby lowering the amount of generated CO required for iron ore reduction. A catalyst is used to promote gasification, while the production of catalyst-supported coke can be roughly divided into two categories: (1) the pre-mixing method, wherein a catalyst precursor is mixed with coal and then dry distilled, and (2) the coke reforming method, wherein a catalyst is added to the coke after carbonization.³⁾ Alkali, alkaline earth metals, and transition metal species are mainly used as catalysts.³⁾ Among them, Fe and Ca are known to have high catalytic reactivity toward carbonaceous materials.^{4,5,6,7,8,9,10)} Although gasification could be accelerated using either Fe or Ca by both methods abovementioned, use of Fe is considered to be more effective. In addition, pre-mixing method shows higher gasification performance than coke reforming method owing to its better dispersibility of catalyst.^{4,5,6,7,8,9,10)} However, the pre-addition of alkali, alkaline earth, and transition metals to coal reduces its fluidity during carbonization.^5,11) In other words, the addition of Fe and Ca to coal affects caking during carbonization; hence, uncertainty exists whether the increase in gasification reactivity is due to a catalytic effect or the underdevelopment of the coke substrate (C structure) due to reduced thermoplasticity. To elucidate this aspect, detailed studies have been conducted on coking properties, C structure, and gasification reactivity when Fe and Ca species are added to coal.^12,13) Among them, when Fe₂O₃ was added to caking-coal, the maximum fluidity (MF) at the gieseler examination value, drum index (DI), and coke strength after reaction (CSR) decreased with a decrease in particle size and an increase in the amount of the added Fe species. A decreasing MF, DI, and CRS, and an increasing coke reaction index (CRI) and amorphization of the C structure have also been reported by Fe addition to caking-coal.¹²⁾ Conversely, it was also found that the MF, DI, and CRS were unchanged, while the CRI increased on adding CaCO₃ as compared to when not adding CaCO₃ to coke, regardless of the size and amount of the Ca species added.¹²⁾ It was also reported that the gasification reactivity increased in both cases, on adding either Fe₂O₃ or CaCO₃ to caking-coal, with the former having a significant effect and the latter having almost no effect on the C structure during carbonization.¹³⁾ In other words, the addition of a Ca species to caking-coal helps produce coke with an improved CRI while maintaining the thermoplastic performance and C structure during carbonization, as compared to when an Fe species is added. Therefore, if a Ca species loaded with thermoplastic components could be prepared and carbonized, it could be developed to produce high-strength and highly reactive coke from coal that is non-caking or slightly-caking.

We have previously shown that high-strength coke could be produced by blending coal and C/C composites, which were prepared by biomass char impregnated biomass tar.¹⁴⁾ Here, tar-derived carbonaceous material was filled in the char pores of a composite. It is well-known that tar and its semi-carbonaceous material exhibit fluidity. By filling the pores of the char, which is non-fluid and porous, with tar-derived carbonaceous material, a composite could be prepared in which the char and thermoplastic material are brought into close proximity. In addition, when the composite was blended with the caking-coal, the extent of coal fluidity by char addition during carbonization decreased; hence, it was possible to produce high-strength coke. If a calcium/carbon composite (Ca/C) could be prepared by the same treatment as for the Ca species, which is effective for improving gasification reactivity (i.e., preparing a Ca/C composite by filling the pores of porous Ca species with thermoplastic carbonaceous material and producing coke by blending caking-coal and the composite), coke with high-strength and high gasification reactivity could be produced by maintaining/improving the fluidity of the caking-coal during carbonization by tar-derived carbonaceous material. In addition, the prepared Ca/C composite could be a raw fuel for sintered ore production because of its C content. Previous studies reported that coke made from iron oxide and C composites by filling the pores of low-grade iron ore with thermoplastic carbonaceous material had high-strength and was highly reactive.^15,16,17)

Therefore, here, CaO was first made porous by the water vapor swelling method; thereafter a Ca/C composite was prepared by filling tar-derived carbonaceous material into the pores of the swelling Ca. This study primarily aimed to thoroughly investigate the properties of coke prepared from a Ca/C composite blended with caking-coal.

2. Experimental Method

2.1. Samples

They were made of limestone (CaO, >99%, particle size of 3.3–4.0 mm and 60–90 μm), tar, and caking-coal (Coal A and Coal B). The CaO was provisionally calcinated at 800°C for 1 h under atmospheric conditions to remove CaCO₃. The MF values for Coal A (Ash: 12.6 wt%-dry and volatile matter (VM): 36.1 wt%-dry) and Coal B (Ash: 12.1 wt%-dry and VM: 31.1 wt%-dry) were 1.96 and 0.83 log (ddpm), respectively. To prevent weathering, these caking-coals were air-dried immediately before use, crushed to a particle size of −200 μm, and sieved for testing. While preparing coke from caking-coal, a mixture of Coal A and B (7:3) was used as described below. All the above samples were obtained from a Japanese steel manufacturer.

2.2. Vapor Swelling of CaO

The swelling of calcinated CaO was carried out using the following method:¹⁸⁾ the sample was placed in a cylindrical glass container, and the vessel was placed in a stainless steel container pre-filled with ion-exchange water. The container was sealed and then heat-treated for 1–24 h in a dryer at 110°C. The swelling-rate and yield were calculated using Eqs. (1) and (2):   

$$\mathrm{S}\mathrm{w}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{g}\text{-}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e}(\%)={\mathrm{V}}_{\mathrm{i}}/{\mathrm{V}}_0\times 100$$(1)

  

$$\mathrm{Y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}(\%)={\mathrm{W}}_{\mathrm{i}}/{\mathrm{W}}_0\times 100$$(2)

Here, V_i, V₀, W_i, and W₀ are sample volume after swelling (mL), sample volume before swelling (mL), sample weight after swelling (g), and sample weight before swelling (g), respectively. Swollen samples were dried overnight in a vacuum dryer at 108°C and used for the experiments described below.

2.3. Preparation of Ca/C Composite

The swollen samples prepared above and as-received tar were used here. Approximately 1.0 g of the swollen samples was physically mixed with the tar at room temperature in a weight ratio of 0.5:4.0 and then heat-treated to 400–500°C at 10°C/min in He using a flow-type quartz-made fixed-bed reactor.¹⁹⁾ Here, these samples are denoted as Ca/C_Temp, e.g., samples heated to 500°C are denoted as Ca/C_500. The yields of the prepared composites and tar-derived C yields were calculated using Eqs. (3) and (4), respectively. The prepared samples were stored in a N₂ atmosphere.   

$$\mathrm{C}\mathrm{a}/\mathrm{C}\mathrm{y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}\left(\mathrm{w}\mathrm{t}\%\right)={\mathrm{W}}_{\mathrm{p}}/{\mathrm{W}}_{\mathrm{s}}\times 100$$(3)

  

$$\mathrm{T}\mathrm{a}\mathrm{r}\text{-}\mathrm{d}\mathrm{e}\mathrm{r}\mathrm{i}\mathrm{v}\mathrm{e}\mathrm{d}\mathrm{C}\mathrm{y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}\left(\mathrm{w}\mathrm{t}\%\right)=\left({\mathrm{W}}_{\mathrm{p}}-{\mathrm{W}}_{\mathrm{s}}\right)/{\mathrm{W}}_{\mathrm{T}}\times 100$$(4)

Here, W_p, W_s, and W_T mean product weight after heating (g), swollen sample weight before heating (g), and tar weight before heating (g), respectively. Although the details are described later, the Ca morphology was confirmed to consist of Ca(OH)₂ before and after the composite preparation. Unless otherwise noted, the heating temperature was set to 500°C.

2.4. Preparation of Pelletized Coke

Co-carbonation of the prepared Ca/C composite and caking-coal was carried out as described here. The swollen CaO and Ca–C composite used here were prepared from CaO with particle diameters of 60–90 μm. The prepared composite or blended sample of the caking-coal and composite were uniaxially pressurized at 30 MPa to prepare the pelletized coal.²⁰⁾ Thereafter, the material thus formed was heated up to 900°C in He at 3°C/min and held for 30 min to obtain the formed coke. The apparent density of the pelletized coal and formed coke was calculated from caliper measurements; the coke strength was evaluated by the crushing strength test described in Section 2.5. The prepared coke was described by the name of the composition followed by the word “coke,” e.g., that prepared from the Ca/C_500 and the Ca/C_500 and caking-coal samples were designated as Ca/C_500_Coke and Ca/C_500/Coal_Coke, respectively.

2.5. Characterization

The Ca morphology and C structure in the prepared samples were evaluated by powder X-ray diffraction (XRD, Shimadzu XRD 6000). The measurements were obtained through Cu-Kα radiation and monochromatization with a Ni filter. The tube current, voltage, and scanning speed were 30 mA, 40 kV, and 1.0°/min (2θ), respectively. To study the extent of C crystallization, the ratio of amorphous-C (A-carbon) to turbostatic-C (T-carbon) was calculated by curve fitting the C(002) plane diffraction lines to those of a Gaussian type.¹⁹⁾ The XRD patterns obtained before the curve fitting were corrected by the Si peaks of the standards; thereafter, a background removal process was performed. As in previously studies, coke powder ground to −200 μm was treated in 48% HF solution at room temperature, because sharp quartz (SiO₂) peaks were detected in the prepared coke and blended-coke samples, which prevented peak deconvolution.¹⁹⁾ The demineralized samples were washed repeatedly with distilled water and dried overnight at 108°C under vacuum. Such a demineralization treatment is known to not affect the C structure.²¹⁾

The pore properties of the swollen samples and the prepared Ca/C composites were evaluated by the N₂ adsorption (Qunaterchrome, Nova1200e) method. The samples were vacuum dried at 108°C for 1 h; thereafter, the N₂ adsorption/desorption isothermal properties were obtained using liquid N₂ at −196°C. The specific surface area and pore size distribution were evaluated from the N₂ adsorption/desorption isotherms obtained using the BET and BJH methods.

For SEM-EDS analysis of the prepared samples, the pellets prepared by embedding the particles in the alloy were polished until the inside of the particles were exposed. The conditions were 15 keV and irradiation current of 1 nA. The obtained SEM images were subjected to particle surface analysis and line analysis.

The strength of the Ca/C composite and formed coke was analyzed based on the compression test method (JIS M8718). This test was conducted twenty times for each sample; the average value was taken as the crushing strength. The crushing strength of the formed coke was measured 2–4 times per sample, and the average measurements were considered. The indirect tensile strength was calculated from the measured crushing strength based on previous studies according to Eq. (5):²⁰⁾   

$$\mathrm{f}=2\cdot \mathrm{P}/\pi \cdot \mathrm{d}\cdot \mathrm{l}$$(5)

where f is the indirect tensile strength (MPa) and P is the crushing strength (N); d and l are the sample diameter (mm) and thickness (mm), respectively.

The combustion properties and CO₂ gasification reactivity of the prepared samples were examined using a thermobalance (Rigaku, TG-DTA8210). In the case of combustion test, one particle (20 mg) of each sample was held in a Pt pan and heated by 10°C/min up to 1500°C in air (200 mL/min). In the case of gasification test, crushed coke sample (particle size of −200 μm) was heated with 10°C/min up to 1500°C in CO₂ (200 mL/min). Weight loss curves were measured during combustion and gasification tests. Metallurgical coke with the same particle size was also used in the test for comparison. The demineralized coke (particle size of 200 μm) prepared above was also used in the CO₂ gasification experiments. The demineralization of samples was carried out with 48% HF solution at room temperature.

3. Results and Discussion

3.1. Swelling Behavior of Calcined CaO with Water Vapor

Figure 1(a) shows the swelling-rate and yield of CaO with respect to the swelling-time for particle sizes of 3.3–4.0 mm and 60–90 μm, respectively. The yield tended to increase with the swelling-time for both particle sizes, reaching a constant of almost 150% within 12 h. The degree of increase in the swelling-rate tended to be higher for large samples; however, the behavior of the swelling-rate was similar regardless of particle size, and was almost constant at 12 h when the increase in yield was completely achieved.

Fig. 1.

(a) Changes in the yield and swelling-rate of the calcined CaO with water vapor swelling-time, and (b) Relationship between yield and swelling-rate. (Online version in color.)

Figure 1(b) shows a plot of the relationship between yield and swelling-rate. The swelling-rate tended to increase with increasing yield for all particle sizes. When the swelling samples were subjected to an XRD analysis, a diffraction peak attributed to Ca(OH)₂ was observed at 1 h. The intensity of such a peak increased with time, and by 6 h of swelling, the peak attributed to CaO completely disappeared, while only that for Ca(OH)₂ was observed. Therefore, the reaction between CaO and water appeared to follow Eq. (6), which is well-known for slaked lime production.   

$$\mathrm{C}\mathrm{a}\mathrm{O}+{\mathrm{H}}_2\mathrm{O}\to \mathrm{C}\mathrm{a}{\left(\mathrm{O}\mathrm{H}\right)}_2$$(6)

3.2. Pore Structure of Calcined CaO with Swelling-time

Figure 2(a) shows the change in the pore size distribution of CaO with swelling-time for particle size of 3.3–4.0 mm. Broad peaks were observed at 3–4 nm and around 30 nm for the sample before the swelling treatment. At 2 h of swelling, the intensity of the peaks observed in the 3–4 nm range in the swollen sample increased, and pore development was observed in the 5–20 nm range. Conversely, the intensity of the peaks around 30 nm in the swollen sample remained unchanged, while that for the 3–4 nm range increased with time, becoming sharper above 12 h. A development of pore size in the 5–20 nm range was also observed to a small extent; in other words, it was found that the swelling treatment resulted in the development of nanometer-order pores. The surface area of CaO (10 m²/g) tended to increase with swelling-time, reaching 25 m²/g at 24 h (Fig. 2(b)). In accordance with previous reports on the change in pore properties of CaO during steam swelling using a mercury porosimeter,¹⁸⁾ the intensity of pore size peak around 50 nm observed in the swollen sample increased with steam swelling treatment. The differences between the reported and prevailing results is unclear; however, it could be due to the different analysis methods and sample compositions. These results suggest that CaO could be made porous by steam swelling.

Fig. 2.

Changes in (a) pore size distribution and (b) specific surface area with water vapor swelling-time. (Online version in color.)

3.3. Crushing Strength of the Prepared Ca/C Composite

Figure 3 shows the yields of the composites prepared using different tar/Ca mixing ratios at 500°C, and the yields of tar-derived carbonaceous material deposited (deposited-C) on the Ca particle surfaces. Samples with particle sizes of 3.3–4.0 mm after 12 h and 24 h of water vapor swelling-time were used. For all samples, the composite yields and yields of the deposited-C increased with an increasing tar mixture ratio. In particular, the degree of increase was significant for mixing ratios of 1.0–3.0, and was almost constant (180 wt%) at higher ratios; moreover, that for the yield tended to be higher for the swelling treatment time of 24 h. This is due to the development of pores with swelling treatment time. In other words, the pores of the sample treated for 24 h were more developed than of that treated for 12 h, which led to a difference in the behavior of the increasing yield due to tar filling in the pores. Tar is a raw material for chemical products; hence, the amount of tar used in the preparation of composites should be low. Here, the yield of deposited-C on CaO was around 20–30%, which was low, and the tar desorbed during pyrolysis could be recovered as tar.

Fig. 3.

Changes in the yield of prepared composites and deposited-C (derived from tar on porous CaO) with the mixing ratio of tar/Ca. (Online version in color.)

Figure 4 shows the strength of the composites prepared at different tar/Ca mixing ratios at 500°C. In both swollen samples, the strength remained unchanged up to a tar/Ca mixing ratio of 1.0. At a swelling-time of 12 h, the strength tended to increase from a ratio of 2.0, and reached 40 daN at a ratio of 3.0. For the sample with a swelling-time of 24 h, the strength increased with an increase in the mixing ratio from 1.5 to 3.0, remaining almost constant (60 daN) thereafter. The strength of the composite prepared from a swelling-time of 24 h was greater than that prepared from a swelling-time of 12 h. This is because the samples with a swelling-time of 24 h had more pores, which were easily filled with tar-derived carbonaceous materials while preparing the composites, as compared to those with a swelling-time of 12 h. The Ca(OH)₂ peak in both composites was attributed to the XRD measurement. As the decomposition temperature of Ca(OH)₂ is ~450°C,²²⁾ Ca could have existed as CaO in Ca/C_500. Only swollen CaO (Ca(OH)₂) was subjected to a thermal analysis, and a peak in the weight loss rate was observed at ~450°C. Therefore, the decomposition of Ca(OH)₂ could have shifted to higher temperatures because of its composition with tar-derived carbonaceous materials. Thus, it was determined that the tar-derived carbonaceous material could be filled into the pores of the Ca compound by heat treatment after mixing the porous Ca and tar; hence, a Ca/C composite with high-strength could be prepared by this method. A swelling-time of 24 h and a tar/Ca mixing ratio of 3.0 were used to prepare the composite for further studies.

Fig. 4.

Change in crushing strength of the composites prepared from porous CaO with the mixing ratio of tar/Ca. (Online version in color.)

When the samples prepared above were subjected to N₂ adsorption analysis, the peaks at 3–4 nm and 30 nm observed in Fig. 2(a) decreased and disappeared as the mixing ratio of tar increased. The surface area (25 m²/g) and pore volume (0.12 cm³/g) of swollen Ca were < 5 m²/g and < 0.01 cm³/g, respectively, at a mixing ratio of 3.0, where the strength was constant. This result indicates that the pores are filled with tar-derived carbonaceous material. According to previous reports, the gaseous-tar generated during the heating treatment for mixture of the porous material and tar preferentially diffuses and precipitates in the pores, and then the tar-derived carbonaceous material also deposits on the particle surface.¹⁴⁾

Figure 5 shows photographs and SEM analysis results of the inside of swollen 3.3–4.0 mm CaO particles before and after composite preparation. The treatment temperature was 400°C. The inside of the swollen Ca particles was white, while the Ca/C particles were black (Figs. 5(a), 5(g)). When the interior of the particles was subjected to plane and line analysis, it was found that carbon was present inside and outside the particles in Ca/C, whereas it was not found in the swollen Ca particles (Figs. 5(c), 5(i)). (Figs. 5(f), 5(l)). From the surface analysis results, the mass percentages of C, O, and Ca in the swollen Ca and Ca/C were determined to be 2.5 and 40 wt%-C, 39 and 35 wt%-O, and 59 and 25 wt%-Ca, respectively, indicating that the Ca/C contained 40 wt% carbon. In other words, it was found that the present method can be used to load tar-derived carbonaceous materials inside and outside of swollen Ca particles.

Fig. 5.

Cross-section image (a, b, g, h), elemental mapping (c–e, i–k) and line analyses (f, l) of porous CaO (a–f) or Ca/C_400 (g–l). (Online version in color.)

3.4. Thermal Reactivity of Prepared Composite

Figure 6(a) shows the weight loss behavior of Ca/C_500 prepared from CaO with a particle size of 3.3–4.0 nm during combustion. For comparison, the results of metallurgical coke combustion are also shown in Fig. 6(a). Weight loss was observed from ~500°C, occurring in two stages: 500–650°C and 700–800°C. Ca(OH)₂ is known to decompose between 400°C and 500°C.²²⁾ The weight loss in the first stage (500–650°C) could be due to the decomposition of Ca(OH)₂ and the combustion of tar-derived carbonaceous materials. The decomposition of CaCO₃ was observed above 600°C.^22,23) Therefore, the weight loss in the second stage could be attributed to the decomposition of CaCO₃ formed by the reaction of Ca species with CO₂ generated by the combustion of tar-derived carbonaceous material in the first stage. The burning rate of Ca/C_500 was higher than that of the metallurgical coke particles. This result suggested that the composite is unsuitable as a sintered-ore fuel.

Fig. 6.

Thermogravimetric curves of Ca/C_500 and coke in (a) air and (b) CO₂. (Online version in color.)

Figure 6(b) shows the temperature dependence of the gasification rate of Ca/C_500 during CO₂ gasification. During the gasification of Ca/C_500, an increase in weight due to the transformation of Ca(OH)₂ in the composite into CaCO₃ and a decrease in weight due to the decomposition of CaCO₃ were observed (not shown in Fig. 6(b)). A difference in the decomposition temperature between CaCO₃ independently and CaCO₃ in Ca/C_500 in CO₂, i.e., a shift in the CaCO₃ decomposition temperature to a lower temperature for the Ca–C composite, was also observed. Based on previous studies, a CaCO₃–CaO cycle has been proposed for CO₂ gasification, wherein CaO reacts with CO₂ to form CaCO₃, which then reacts with C and returns to CaO, which occurs mainly at 700–900°C.²³⁾ Therefore, the difference in the decomposition temperature is due to the above cyclical mechanism. First, a temperature correction was made for the weight loss behavior of Ca(OH)₂ in CO₂, which was measured beforehand from the weight loss behavior of Ca/C_500 to describe the gasification rate of Ca/C_500 on a dacf (dry ash catalyst free) basis. Then, the conversion of C by gasification was determined by correcting the weight loss due to CaCO₃ decomposition, considering the difference. For comparison, the results of the high-strength (DI87_coke) coke (denoted as metallurgical coke in Fig. 6(b)) with the same particle size are also presented. The gasification of the coke samples began around 1050°C. Conversely, the gasification of Ca/C_500 began at ~800°C and was completed at 1000°C. These results suggested that the gasification rate of the composite was higher than that of coke.

3.5. Strength of Coke Prepared by Blending Ca/C Composites and Caking-coal

Figure 7 shows the density, yield, and indirect tensile strength of the coke prepared by blending caking-coal and swollen CaO (Ca(OH)₂). Swollen samples prepared from CaO with particle sizes of 60–90 μm were used in the examination. As the amount of Ca(OH)₂ (swelling CaO) added to the coal increased, the density of the pelletized coal tended to increase, while that of the formed coke tended to decrease (Fig. 7(a)). The coke yield tended to increase slightly with an increasing amount of added Ca. The coke strength decreased with an increasing amount of added Ca, becoming almost zero at a 10% addition. These results suggested that the addition of Ca(OH)₂ decreased the coke strength; this was caused by the impeded fluidity of the caking-coal due to the addition of non-fluid substances, similar to previous reports.^5,11)

Fig. 7.

Changes in (a) bulk density and coke yield and (b) indirect tensile strength of coke prepared with different blending ratios of Ca(OH)₂ and coal. (Online version in color.)

Figure 8 shows the characteristics of coke prepared by adding Ca/C_400 to coal. The yields of Ca/C_400 and Ca/C_500 were 260 wt% and 180 wt%, respectively, which were higher in the former case. This indicates that more tar-derived carbonaceous material remained in Ca/C_400 as compared to Ca/C_500. In Ca/C_400, the pyrolysis of tar was incomplete, with many persistent tar-derived thermoplastic components. This was supported by the fact that the yields of carbonaceous material at 400°C and 500°C were 50 wt% and 35 wt%, respectively, when only tar was pyrolyzed under the same conditions as the Ca/C composite. Considering Ca/C_400, the density of the pelletized coal increased with increasing amounts of Ca/C_400. A similar trend was observed for the coke formed after carbonization. Conversely, the coke yield tended to decrease with an increase in the addition of Ca/C_400 (Fig. 8(a)). The strength of coke formed without the addition of a composite (5 MPa) increased with an increase in the Ca/C_400 dosage up to 50 wt%, reaching 15 MPa; however, the coke strength decreased as the amount of Ca/C_400 added increased (>50 wt%).

Fig. 8.

Changes in (a) bulk density and coke yield, and (b) indirect tensile strength of coke prepared with different blending ratios of Ca/C_400 and coal. (Online version in color.)

Figure 9 shows the results of coke prepared by adding Ca/C_500 to coal. The density of pelletized coal and the formed coke tended to increase with increasing amounts of Ca/C_500 (Fig. 9(a)). Conversely, the coke yield was almost constant regardless of the amount of the composites (Fig. 9(a)). The coke strength increased with the addition of Ca/C_500, reaching 13 MPa at a 50 wt% addition; however, it decreased with the addition amount. Such a trend in the change in strength was similar to that achieved by addition of Ca/C_400.

Fig. 9.

Changes in (a) bulk density and coke yield, and (b) indirect tensile strength of coke prepared with different blending ratios of Ca/C_500 and coal. (Online version in color.)

As mentioned above, the Ca/C composite contained tar-derived carbonaceous materials, and at a composite preparation temperature of 400–500°C, thermal decomposition of the tar in the composite was complete. Therefore, it is considered that the density of the pelletized coal and formed coke increased with an increase in the amount of composite added owing to the binding effect of tar or its derivatives loaded into the Ca/C composite. The characteristics of coke prepared by adding carbonaceous materials (tar_C_400 and tar_C_500) to the caking-coal, which were prepared by pyrolyzing only tar at 400°C and 500°C, respectively, were thoroughly investigated; Fig. 10 demonstrates the outcome. The density of coke was in the range of 0.95–1.0 g/cm³, and the yield was 65–70 wt%. The tar_C_400 contained more volatiles than tar_C_500; however, the degree of decrease in the yield was similar to that of tar_C_500/Coal. Thus, co-carbonization could have occurred between tar_C and the caking-coal. The coke strength of tar_C_500 reached the same level as that of the 50%-Ca/C_500/Coal (12.5 MPa) at a blending ratio of 20–25 wt%. Conversely, the strength of tar_C_400 decreased at a blending ratio of 25 wt%. The formation of a foamed structure was observed in the cross-section after the compression test of the coke preparation; thus, the decrease in strength was due to excessive melting. The proportion of tar-derived carbonaceous material in the blended samples at blending ratios of 50% of Ca/C_400 and Ca/C_500 was ~30 wt% and 25 wt%, respectively. Therefore, the increase in strength of the Ca/C mixture is attributable to the tar-derived carbonaceous material. Excessive melting was observed in the 25 wt% tar_C_400 blend, but not in the 25 wt% Ca/C_400 blend. This could be due to the adverse effect of Ca species on the fluidity of the caking-coal by blending Ca/C_400. The thermoplasticity of coal could be explained by the continuous self-dissolution model (the thermal mobility of components with low molecular weights in the coal helps dissolve those with heavier molecular weights).²⁴⁾ According to this model, the softening of coal could be attributed to its components having low-temperature dissolution (mobile components with low molecular weight). Generally, coal becomes fluid at ~400°C and re-solidifies around 500°C after it attains MF at ~450°C. Ca/C_400 and Ca/C_500 used here were subjected to temperatures of 400°C and 500°C, respectively, during the preparation process. Hence, most of the tar loaded into the composites up to these temperatures could volatilize, but Ca/C_400 contained more components with low molecular weights than Ca/C_500 (Fig. 3). Therefore, it is considered that the mobile components with low molecular weights in Ca/C_400 along with those in the caking-coal dissolved the heavy components in the caking-coal and Ca/C_400. Thus, the thermoplasticity of the caking-coal and Ca/C_400 could be improved, and high-strength coke could be prepared with Ca/C_400. Conversely, Ca/C_500 contained fewer mobile components with low molecular weights, which could have contributed to the initial softening and melting of the caking-coal; however, it is presumed that heavy components are included. Mobile components with low molecular weights derived from the caking-coal could have contributed to the thermoplasticity of the caking-coal by dissolving the heavy components in Ca/C_500, resulting in high-strength coke. According to our previous studies on the properties of coke prepared from coal that is non-caking/slightly-caking, chemically modified with thermoplastic components, the modification process helps improve the thermoplasticity of coal that is non-caking/slightly-caking. In addition, the contact between the inert material and the melt at the cross-section of coke is improved.²⁵⁾ Hence, the strength of the components of the Ca/C composite could have increased due to better contact of the Ca/C interface with the caking-coal during carbonization. In other words, it can be considered that the placement of tar-derived material as a binder on the surface of the Ca species, which does not melt, was effective in increasing the coke strength when Ca/C prepared from small particle size was blended as described above. In that case, the effect of carbonaceous material filling in the Ca/C particles is questionable. As described in Section 3.3, in this method, the pores of the porous material are preferentially filled with carbonaceous material, and then carbonaceous material is deposited on the particle surface as well. Therefore, it is impossible to load carbonaceous materials only on the particle surface. On the other hand, the effect of carbonaceous material filling in Ca/C particles on coke properties can be considered as follows: (i) Ca and C in Ca/C are in close proximity, which contributes to the speeding up of the initial gasification reaction due to the catalytic effect of Ca and the subsequent generation of gasification gas, thereby increasing the gasification rate of the coke bulk. (ii) Ca(OH)₂ (2.21 g/cm³) present in Ca/C changes to CaCO₃ (2.71 g/cm³) and CaO (3.35 g/cm³) upon heating, resulting in density increase (shrinkage). The filling of porous Ca with carbonaceous material affects the shrinkage behavior of this Ca species during heating and contributes to maintaining the strength of the coke bulk. Microstructural observation of the prepared blended coke or investigation of shrinking behavior of blending sample (Ca/C-coal) is an issue to be addressed in the future.

Fig. 10.

Changes in (a) bulk density and coke yield, and (b) indirect tensile strength of coke prepared with different blending ratios of tar_C_400 and coal, or tar_C_500 and coal. (Online version in color.)

3.6. Effect of Preparing a Composite (by Blending Ca/C and Coal) on Coke Strength

The properties of coke prepared from samples containing tar_C, Ca(OH)₂, and caking-coal were investigated. Figure 11 shows the results, which help show the effect of preparing a composite on coke properties. The blending ratio of the above carbonaceous material (tar_C_400/ tar_C_500) and swollen CaO (Ca(OH)₂) to the caking-coal was equal in both, 50%-Ca/C_400/Coal and 50%-Ca/C_500/Coal. The density, coke density, and coke yield were in the ranges of 1.15–1.20 g/cm³, 1.0–1.15 g/cm³, and 70–80 wt%, respectively for all pelletized samples. Figure 11 shows the coke strength. The sequence of coke strength for the various composites was Ca(OH)₂/tar_C_500/Coal < Ca(OH)₂/tar_C_400/Coal < 50%-Ca/C_500/Coal < 50%-Ca/C_400/Coal; which was lesser in the physically mixed samples. In other words, the strength was improved by preparing composites. This was due to the minimal contact between the Ca(OH)₂ particles and the thermoplastic tar-derived components in the physical mixture. The former inhibited the thermoplasticity of the caking-coal (in the case of Ca(OH)₂-tar_C_500/Coal; in particular, the coke strength was lower than that of the caking-coal by 4.5 MPa). Similar results were observed in our previous studies,^14,26,27) suggesting that the reduction in the strength of Ca(OH)₂/tar_C_400/Coal was relatively less due to the high residual softening and melting capacity of tar_C_400. From these results, it was determined that high-strength coke could be produced by preparing a composite of non-fluid materials with soft-melting materials, i.e., composites having mesoscale proximity and surface coating.

Fig. 11.

Effect of compositing on the indirect tensile strength of prepared coke. (Online version in color.)

3.7. CO₂ Gasification Reactivity of Coke Prepared from a Blend of Ca/C and Coal

Figure 12 shows the temperature variation of the gasification rates of 50%-Ca/C_500/Coal_Coke in a CO₂ atmosphere. Coke, 100%-Ca/C_500/Coke, and 25%-tar_C/Coal (called Coke, 100%-Ca/C_500_Coke, and 25%-tar_C/Coal_Coke, respectively) are also shown for comparison. The gasification of coke began at ~900°C, which progressed to 1200°C with increasing temperature. Conversely, the gasification of 50%-Ca/C_500/Coal_Coke began at ~800°C and ended at 1200°C. The gasification rate of 25 wt% tar_C500/Coal_Coke was lower than that of the coke over 1100°C, which occurred due to the addition of tar_C_500. This was possible as the addition of tar_C_500 could have increased the coke strength and contributed to the absence of the non-fluid material, Ca species, making it easier for the mobile component of tar_C_500 to contribute to the caking of the caking-coal. This could result in the development of a coke structure and the slowing down of the gasification rate in the high-temperature region. The gasification of 100%-Ca/C_500_Coke occurred at 800–1000°C and the gasification rate was the highest among the tested samples. The gasification behavior of 50%-Ca/C_500-Coal_Coke was calculated from that of 100%-Ca/C_500_Coke and the coke; it was almost the same as the measured value, as shown by the dashed-line in the figure. This result suggested that the Ca/C composite did not affect the caking of the caking-coal (development of a C structure on gasification reactivity) during carbonization. These results suggest that the gasification reactivity of the Ca/C blended-coke was greater than that of the caking-coal coke.

Fig. 12.

CO₂ gasification conversion of prepared coke samples. (Online version in color.)

However, it is unclear whether this increase in reactivity is due to the catalytic effect of Ca or the difference in C structure due to the addition of tar-derived carbonaceous material in the Ca/C composite. Figure 13 shows the outcome of the XRD of each coke sample prepared by demineralization to clarify this aspect. The XRD analysis of the demineralized samples of the coke, 100%-Ca/C_500_Coke, 25%-tar_C/Coal_Coke, and 50%-Ca/C_500/Coal_Coke (Figs. 13(a)–13(d)) showed no clear differences in the C structure of either sample. The gasification reactivity of coke is known to increase with a decreasing crystallite size on the (002) surface.^28,29,30) Therefore, the crystallite size of C calculated from the (002) band using the Debye-Scherrer equation was found to be in the range of 28.5–30.3 Å, and the C structure was found to be insignificantly altered by preparing a composite, tar-C addition, and the presence of Ca species. The result of curve-fitting against the XRD profiles of the demineralized coke are shown in Figs. 13(e)–13(h). Both coke samples contained 79–87% A-C and 13–21% T-C. This suggested that the treatment method used here had little effect on the C structure of the prepared coke. Therefore, the effect of C structure on the difference in gasification reactivity could be overlooked.

Fig. 13.

XRD patterns and the curve-fitting of Coke_Dem (a, e), 100%-Ca/C_Coke_Dem (b, f), 25%-tar_C/Coal_Coke_Dem (c, g), and 50%-Ca/C_500/Coal_Coke_Dem (d, h). (Online version in color.)

To investigate whether the increase in reactivity was due to the catalytic effect of Ca, the CO₂ gasification behavior of the abovementioned demineralized coke is shown in Fig. 14. No difference in the gasification reactivity was observed in coke, regardless of the presence/absence of demineralized coke. In the case of 100%-Ca/C_500/Coal_Coke, the demineralized coke caused a shift in the gasification temperature, higher than that of 100%-Ca/C_500/Coal_Coke. On the other hand, the gasification reactivity of 100%-Ca/C_500/Coal_Coke was significantly lower than that of 100%-Ca/C_500/Coal_Coke, regardless of the presence of demineralized coke. In 50%-Ca/C_500/Coal_Coke, demineralization caused a reduction in the gasification rate below 1000°C, but the gasification behavior above 1000°C was almost consistent with that of the non-demineralized sample. This implied that Ca helped catalyze the gasification reaction below 1000°C. Considering previous studies, a CaCO₃–CaO cycle has been proposed for CO₂ gasification, wherein CaO reacts with CO₂ to form CaCO₃, which then reacts with C, thereafter returning to CaO.²³⁾ Studies on the mechanism of the action of a Ca catalyst using the elevated temperature desorption method have also proposed a mechanism to diffuse CO₂ through the catalyst and react with C at the catalyst-C interface.^26,27) As the above CaCO₃–CaO cycling mechanism occurs at 700–900°C, the increased gasification rate of 50%-Ca/C-500/Coal_Coke below 800–1000°C could have occurred due to the CaCO₃–CaO cycling mechanism, as shown in Fig. 12. As the measured and calculated values are relatively close, the CaCO₃–CaO cycle mechanism could have occurred in the vicinity of the Ca composite particles. The gasification proceeded at the interface between Ca in the Ca/C composite and tar-derived carbonaceous material, which affected the gasification reaction rate for the entire coke sample. Conversely, the gasification rate at 1000–1100°C was consistent with 50%-Ca/C_500 considering the demineralized sample, because Ca/C is unlikely to have any catalytic effect on the coke substrate derived from caking-coal. From these results, it was determined that a Ca/C composite prepared from porous Ca could be blended with caking-coal to produce high-strength and highly reactive coke. The blast furnace process is known to help discharge gangue and ashes from the input material as slag. The amount of impurities in the feedstock should be minimal to reduce the operating cost and CO₂ emissions. Producing high-strength and highly reactive coke by blending a small amount of Ca/C is a future challenge.

Fig. 14.

CO₂ gasification conversion of prepared coke samples and demineralized coke. (Online version in color.)

4. Conclusions

In this study, a Ca species calcium composite (Ca/C) was prepared by the steam swelling method to produce high-strength and highly reactive coke with porous CaO, and its pores were filled with tar-derived carbonaceous material. Thereafter, the properties of the coke prepared from the Ca/C composite blended with the caking-coal were thoroughly investigated; the following findings were obtained:

(1) The mesopores disappeared on filling the porous Ca with tar-derived carbonaceous material; the Ca was made porous by steam swelling, resulting in crushing strength.

(2) High-strength coke could be produced by adding up to 50% of the prepared Ca/C composite in the caking-coal.

(3) The Ca/C blended-coke demonstrated higher gasification reactivity than the original caking-coal.

(4) The C structure of the Ca/C blended-coke was unaffected.

(5) An increase in the gasification reactivity of the Ca/C blended-coke depended on the catalytic effect of Ca.

Acknowledgements

This work was partly supported by a grant from the Iron and Steel Institute of Japan.

References

• 1)   T.  Anyashiki,  K.  Fukada and  H.  Fujimoto: JFE Giho, 22 (2008), 20.
• 2)   M.  Naito,  A.  Okamoto,  K.  Yamaguchi,  T.  Yamaguchi and  Y.  Inoue: Tetsu-to-Hagané, 87 (2001), 357 (in Japanese).
• 3)   S.  Nomura,  S.  Matsuzaki,  M.  Naito,  H.  Ayukawa,  S.  Koizumi,  T.  Abe,  H.  Kitaguchi and  T.  Tahara: Shinnittesu Giho, 384 (2006), 100.
• 4)   Y.  Ohtsuka,  Y.  Kuroda,  Y.  Tamai and  A.  Tomita: Fuel, 65 (1986), 1476.
• 5)   S.  Nomura,  H.  Terashima,  E.  Sato and  M.  Naito: ISIJ Int., 47 (2007), 823.
• 6)   S.  Nomura,  H.  Kitaguchi,  K.  Yamaguchi and  M.  Naito: ISIJ Int., 47 (2007), 245.
• 7)   G.  Hermann and  K. J.  Hüttinger: Fuel, 65 (1986), 1410.
• 8)   T.  Kyotani,  S.  Karasawa and  A.  Tomita: Fuel, 65 (1986), 1466.
• 9)   Y.  Kashiwaya,  S.  Nakaya and  K.  Ishii: Tetsu-to-Hagané, 77 (1991), 759 (in Japanese).
• 10)   M.  Kawakami,  H.  Taga,  T.  Takenaka and  S.  Yokoyama: ISIJ Int., 44 (2004), 2018.
• 11)   K.  Uebo,  S.  Suyama and  T.  Chikada: CAMP-ISIJ, 11 (1998), 705.
• 12)   N.  Kannari,  K.  Takakusagi,  J.-i.  Ozaki,  Y.  Harada,  D.  Anraku and  H.  Fujimoto: Tetsu-to-Hagané, 96 (2010), 249 (in Japanese).
• 13)   A.  Sharma,  K.  Uebo and  Y.  Kubota: Tetsu-to-Hagané, 96 (2010), 280 (in Japanese).
• 14)   Y.  Mochizuki and  N.  Tsubouchi: Fuel Process. Technol., 193 (2019), 328.
• 15)   K.  Miura,  K.  Miyabayashi,  M.  Kawanari and  R.  Ashida: ISIJ Int., 51 (2011), 1234.
• 16)   M.  Kawanari,  A.  Matsumoto,  R.  Ashida and  K.  Miura: ISIJ Int., 51 (2011), 1227.
• 17)   K.  Miura,  J.  Yan,  H.  Ohgaki,  R.  Ashida and  K.  Ichikawa: ISIJ Int., 59 (2019), 2182.
• 18)   S.  Wu,  N.  Sumie,  C.  Su,  E.  Sasaoka and  M. A.  Uddin: Ind. Eng. Chem. Res., 41 (2002), 1352.
• 19)   Y.  Mochizuki,  Y.  Kubota,  K.  Uebo and  N.  Tsubouchi: Powder Technol., 355 (2019), 782.
• 20)   N.  Tsubouchi,  R.  Naganuma,  Y.  Mochizuki,  H.  Hayashizaki and  T.  Shishido: ISIJ Int., 59 (2019), 1427.
• 21)   N.  Yoshizawa,  K.  Maruyama,  Y.  Yamada,  Y.  Takahashi,  G.  Katagiri,  Y.  Shimane and  M.  Harada: J. Jpn. Inst. Energy, 80 (2001), 343.
• 22)   M.  Kato,  T.  Hari,  S.  Saito and  M.  Shibukawa: Tetsu-to-Hagané, 100 (2014), 340 (in Japanese).
• 23)   D. W.  McKee: Fuel, 59 (1980), 308.
• 24)   T.  Takanohashi,  T.  Yoshida,  M.  Iino,  K.  Kato,  K.  Fukada and  H.  Kumagai: Structure and Thermoplasticity of Coal, ed. by I. Komaki et al., Nova Science Publishers, New York, (2005), 55.
• 25)   N.  Tsubouchi,  R.  Naganuma,  Y.  Mochizuki,  H.  Hayashizaki,  T.  Shishido and  A.  Sharma: ISIJ Int., 59 (2019), 1396.
• 26)   N.  Tsubouchi,  R.  Naganuma,  Y.  Mochizuki,  H.  Hayashizaki and  T.  Shishido: ISIJ Int., 59 (2019), 1427.
• 27)   N.  Tsubouchi,  R.  Naganuma,  Y.  Mochizuki,  H.  Hayashizaki and  T.  Shishido: ISIJ Int., 59 (2019), 1419.
• 28)   A.  Sharma,  H.  Kadooka,  T.  Kyotani and  A.  Tomita: Energy Fuels, 16 (2002), 54.
• 29)   H.  Orikasa,  H.  Nishihara,  Q. H.  Yang,  K.  Matsuoka,  T.  Kyotani,  A.  Sharma,  K.  Matsui,  H.  Fujimoto,  K.  Fukada and  K.  Kato: Tetsu-to-Hagané, 92 (2006), 137 (in Japanese).
• 30)   A.  Sharma,  T.  Kyotani and  A.  Tomita: Carbon, 38 (2000), 1977.
• 31)   D.  Cazorla-Amoros,  A.  Linares-Solano,  A. F.  Marcilla Gomis and  C.  Salinas-Martinez de Lecea: Energy Fuels, 5 (1991), 796.
• 32)   D.  Cazorla-Amoros,  A.  Linares-Solano,  C.  Salinas-Martinez de Lecea and  J. P.  Poly: Energy Fuels, 6 (1992), 287.