Changes in Pore Structure of Metallurgical Cokes under Blast Furnace

Article pubs.acs.org/EF

Changes in Pore Structure of Metallurgical Cokes under Blast Furnace Conditions Xing Xing,*,† Harold Rogers,‡ Guangqing Zhang,§ Kim Hockings,∥ Paul Zulli,‡ and Oleg Ostrovski† †

School of Materials Science and Engineering, UNSW Australia, Kensington, New South Wales 2052, Australia BlueScope Steel, Port Kembla, New South Wales 2505, Australia § School of Mechanical, Materials & Mechatronic Engineering, University of Wollongong, Wollongong, New South Wales 2522, Australia ∥ BHP Billiton, Brisbane, Queensland 4000, Australia ‡

ABSTRACT: Metallurgical cokes were subjected to gasification by CO−CO2−N2 gas with blast-furnace-like composition− temperature profile to 1673 K (1400 °C) and annealing under N2 at temperature up to 2273 K (2000 °C). Pore structure of cokes was examined using image analysis. Porosity and pore size were both enlarged under gasification and annealing conditions. The pore structure change during gasification was mainly a result of the Boudouard reaction; the pore structure development upon annealing was attributed to the reactions of mineral matters with carbon and further devolatilization. Annealing and gasification caused a decrease in average pore roundness, an increase in the fraction of low roundness pores and the increase of coalescence points in the pore area. The degradation of coke strength following reaction and annealing was characterized using tensile testing. Both gasification and annealing decreased the mechanical strength of coke. Degradation of more reactive cokes (cokes C and D) by gasification at 1673 K (1400 °C) was visibly stronger in comparison with annealing at the same temperature. Increasing the amount of coalescence points and the degradation of coke microstrength during annealing and gasification were major factors in the degradation of cokes under the simulated blast furnace conditions.

1. INTRODUCTION Metallurgical coke quality is critical to all blast furnace (BF) ironmaking operations. The ultimate measurement of coke quality is its resistance to degradation under the BF operating conditions. From this perspective, coke integrity is a key parameter in the development of ironmaking technology to decrease coke and overall fuel consumption. Coke in a BF is subjected to significant mechanical stress being exposed to chemical reactions with gases (CO, CO2, H2) and heating to high temperature. Peak coke temperatures may approach 2273 K (2000 °C) (because typical raceway adiabatic flame temperatures are approximately 2473 K). The coke is required to maintain adequate strength upon heating and reaction, and minimize fine coke generation to secure the required burden permeability and uniformity of liquid and gas flows. The strength of highly porous and brittle materials like coke is dependent on both pore structure and the microstrength of coke wall components.1−6 Therefore, the degradation of coke in a BF should be related to the change of pore structure and microstrength when subjected to BF thermal and reaction conditions. A 3-D investigation of coke microstructure7,8 demonstrated that majority of pores are interconnected. Geometric irregularities, and coalescence points in the pore structure in particular, concentrate stress that causes the coke degradation.8,9 Yamamoto et al.10 found that the coarse powder generation was dependent on the coarse pore volume and the fine powder generation was related to the fine pore volume; tensile strength of coke was predominately influenced by the volume of pores larger than 100 μm. Pore structure of coke has been intensively studied with a focus on its development during the pyrolysis process.11−20 A © XXXX American Chemical Society

limited number of papers have examined the evolution of pore structure of carbonaceous materials upon heat treatment and reaction.21−25 Xing et al.21 conducted annealing of cokes in the temperature range of 973 to 1773 K (700 to 1500 °C) in an argon atmosphere; pore structure was determined using image analysis. The porosity of cokes was not affected by the annealing at temperatures below 1573 K (1300 °C) and slightly increased in the temperature range 1573−1773 K (1300−1500 °C). Pusz et al.23 studied gasification of cokes according to the standard CSR/CRI conditions (in 100 % CO2 atmosphere at 1373 K (1100 °C) for 2 h); the pore structure of coke prior and after reaction with CO2 was examined using helium gas densitometry and optical microscopy. The porosity and pore size were distinctly increased upon gasification; the shape of pores changed from approximately circular to highly irregular and the pore surfaces changed upon gasification to ragged outlines from originally smooth. Gudenau et al.24 studied the effect of char injection on the coke degradation upon gasification reaction through change of the gas composition, char powder with different concentration was blown into reactor with supplied gases (CO2/N2 = 1.2) to simulate the pulverized coal injection (PCI). They found that the presence of char powder affected the gasification mechanism as well as the pore size distribution after gasification: the volume of pores with size over 10 μm increased with increasing char concentration in the supplied gas flow. Shin et al.25 studied the effect of gasification with CO2 and H2O on porosity of Received: September 21, 2015 Revised: November 8, 2015

A

DOI: 10.1021/acs.energyfuels.5b02152 Energy Fuels XXXX, XXX, XXX−XXX

Article

Energy & Fuels

Samples were heated up to 1173 K (900 °C) in nitrogen, then CO− CO2−N2 gas mixture was introduced into the reactor to start the gasification. The composition of CO−CO2 with balance N2 gas mixture varied with temperature according to the atmosphere in different regions of the BF as shown in Figure 1. The gasification

metallurgical coke, and demonstrated that the macrostrength of cokes was strongly affected by the porosity in the coke center. However, the development of pore structure of metallurgical coke under the blast furnace thermal and reaction conditions have not been systemically studied, although it is significant for understanding of degradation of coke strength in the operation of a blast furnace. The aim of this paper is to quantify the change in the pore structure upon reaction under conditions simulating BF gas composition−temperature profiles to 1673 K (1400 °C) and annealing under N2 to 2273 K (2000 °C), and develop an understanding of the mechanism of coke degradation in the BF ironmaking.

2. EXPERIMENTAL SECTION 2.1. Materials. Three metallurgical cokes A, C, and D, were studied in this work. Coke A was produced from a medium volatile base blend of moderate inertinite content (36.9 vol %) with addition of 11 wt % semisoft coal. Cokes C and D were pilot oven cokes prepared from high rank and low rank coals, respectively. The proximate analyses are summarized in Table 1. Figure 1. Gas composition−temperature profiles for gasification test.

Table 1. Proximate and Partial Ash Analyses of Coke Samples moisture content, % (ad) volatile matter, % (db) ash, % (db) CSR, % CRI, % partial ash analysis, wt % SiO2 Al2O3 Fe2O3 CaO MgO Na2O K2O TiO2 P2O5 base/acida SiO2/Al2O3

coke A

coke C

coke D

0.4 1.4 12.0 70.2 20.7

0.5 1.5 12.1 62.7 24.6

0.9 0.3 11.9 31.9 46.7

57.1 29.8 4.7 1.9 0.7 0.4 0.9 1.5 0.8 0.097 1.92

55.1 25.7 9.2 3.2 0.6 0.5 0.1 2.3 1.6 0.164 2.14

51.4 23.1 14.3 3.7 0.8 0.4 1.4 1.1 1.7 0.272 2.23

started at 900 °C and stopped once temperature reached 1273 K (1000 °C, coke 6 (A6, C6, D6)), 1473 K (1200 °C, coke 7 (A7, C7, D7)), and 1673 K (1400 °C, coke 8 (A8, C8, D8)). For coke 9 (A9, C9, D9), the gasification was again followed to 1673 K (1400 °C) and then the sample was held at this final condition for an additional 2 h. Coke samples were quenched under N2 after the gasification reaction. The weight loss of sample in annealing and gasification was measured by weighting the mass of sample before and after annealing/ gasification, and calculated as

weight loss =

(Mbefore − Mafter) × 100 Mbefore

(1)

where Mbefore and Mafter are the mass of sample before and after annealing/gasification. 2.3. Porosity and Pore Geometry. A 3-D investigation of the coke structure showed that a large proportion of pores are interconnected,8 which allows gas escape during the coke making process. In this study, porosity and pore geometry were investigated by 2-D image analysis. The pore structure in 2-D images is presented as “individual pores” of different geometry. The coke porosity and structure change in the process of coke gasification and annealing, more interconnected channels were generated during these processes. Therefore, the chance of coalescence spots to be observed in 2-D images also increased, and the pore structure is shown as “connected pores”. In the analysis of the effect of the change in the coke structure on the mechanical strength, it is assumed that interconnected pore structure observed in 2-D images is a reasonable representation of the coke structure. In this study, the images for porosity and pore geometry analyses were taken in the center of the coke lumps as it has been confirmed that the tensile strength of coke is predominantly affected by the pores in the coke center.25 2.3.1. Porosity. A representative analysis of each tested coke was based on 60 images taken from 60 different lumps of this coke that were mounted in epoxy resin blocks (one image on each lump). Images were captured by a Nikon Model EPIPHOT 600 microscope with Nikon digital camera. Because the large pores were of most interest in connection to coke degradation, a low power objective lens (magnification ×5) was used. In this configuration, pores with diameter smaller than 12 μm were not resolved. The captured images were binarised using software ImageJ developed by NIH. After binarization, the pores and walls of cokes were represented by black and white areas, respectively. The porosity

a

Base/acid = [Fe2O3 + CaO + MgO + Na2O + K2O]/[SiO2 + Al2O3 + TiO2].

2.2. Annealing and Gasification of Cokes. 2.2.1. Annealing. A 200 g sample of coke with a particle size of 19 to 21 mm was heattreated in a graphite furnace for 2 h at temperatures of 1673, 1873, 2073, and 2273 K (1400, 1600, 1800, and 2000 °C) under an inert atmosphere. The heating rate to the nominated treatment temperature was fixed at 25 K/min (25 °C/min). The samples were contained in a graphite crucible, a continuous purge of 1 L/min of N2 (99.99 %) through a graphite ducting tube with the outlet touching the bottom of the crucible. Heat treatment time was commenced from the time the furnace temperature reached the designated value. The sample prior to any heat treatment was labeled as coke 1 (samples A1, C1, and D1). The four coke samples, after annealing at temperatures from 1673 to 2273 K (1400 to 2000 °C), were labeled sequentially as coke 2 (A2, C2, D2) to coke 5 (A5, C5, D5). 2.2.2. Gasification. Similarly, a 200 g sample of coke with a particle size of 19 to 21 mm, contained in a silicon carbide reaction vessel, was reacted under conditions simulating the BF gas composition− temperature profile from 1173 to 1673 K (900 to 1400 °C). In the current experiments, the total gas flow rate was fixed at 5 L/min. B


Article

Energy & Fuels

Figure 2. Pore structure of coke with different pore roundness. of cokes was calculated using software ImageJ as the area fraction of black area. 2.3.2. Pore Geometry. Parameters of pore geometry, including mean area, perimeter, and equivalent circle diameter, were determined by LAS Image Analysis software developed by Leica. Analysis of each coke was based on 10 images from 10 different lumps of this coke mounted in epoxy resin blocks (one image on each lump). Pores with a size smaller than 1000 μm2 were excluded from consideration. Pore roundness, R, was calculated using measured pore area and perimeter8 R=

4πS L2

(2)

where S and L are the area and perimeter of a pore, respectively. The maximum roundness of pores is 1 in the case of a circle; it decreases with increasing complexity of the pore shape.8 As an example, Figure 2 presents 2-D images of pore structure with roundness 0.7 and 0.2. In the 2-D cross section, the coke with high pore roundness tends to have simple pore structure with a small number of coalescence spots; the pore structure of coke with low pore roundness has pores with significant volume of coalescence spots (Figure 2). The area fraction of pores with roundness threshold ≤0.1 was calculated using the pore geometry parameters of each pore measured with LAS Image Analysis software. 2.4. X-ray Diffraction. Mineral phases in cokes were identified using X-ray diffraction (XRD) analysis. Cokes were crushed to passing 212 μm then ashed by heating in air at 1088 K (815 °C) to remove carbon. XRD spectra were obtained using a Philips X’Pert Multipurpose X-ray diffraction system (MPD). Copper Kα radiation (45 kV, 40 mA) was used as the X-ray source. Samples were scanned with 2θ in the range of 10 to 70° with a step size of 0.02° and 0.6 s scanning time at each step. Mineral phases were identified using X’Pert HighScore Plus software. 2.5. Tensile Strength. Tensile strength of cokes was measured with an Instron 1185 screw universal testing machine. The tensile strength, σ, of cylindrical coke pellets prepared for the test was calculated using eq 3:

σ=

2P πdl

Figure 3. Weight loss of cokes after annealing and gasification at different stages.

resulted from the loss of moisture, further devolatilization at high temperatures, and reduction of oxides in mineral matter by carbon. During the coke gasification by CO−CO2−N2 gas mixture, the solution loss reaction made a significant contribution to the weight loss; besides the Boudouard reaction, release of moisture and the carbon−mineral reactions at temperatures below 1673 K (1400 °C) also contributed to the weight loss.21,26 The weight losses of cokes A and C after gasification upon heating to 1273 K (1000 °C, coke 6) were 0.8 and 2.3%, respectively, which were lower than or equal to the sums of their individual moisture and volatile matter, which means that the solution loss reaction was insignificant at this stage. Coke D had a higher weight loss (3.2%) than A and C, although a sum of moisture and volatile matter in this coke was smaller (1.2%), indicating that coke D was more reactive to CO2. 3.2. Porosity of Cokes during Annealing and Gasification. The porosity and the relative standard deviation of the measurement of original cokes and the cokes treated under the simulated BF conditions is presented in Table 2. Originally, coke C had the highest porosity of 66.3 %, whereas the porosities of original cokes A and D were 51.1 and 57.9 %. Annealing in the temperature range of 1673−2273 K (1400− 2000 °C) caused the porosity evolution in all cokes. Annealing at 2273 K (2000 °C) increased the porosity of cokes A, C, and D by 16.5, 9.5, and 6.0 %, respectively. Comparison of porosity of coke 9 samples subjected to gasification to 1673 K (1400 °C) with the porosity of coke 2

(3)

where P is load at the sample failure and d and l are diameter (8 mm) and length (8 mm), respectively.

3. RESULTS AND DISCUSSION 3.1. Reactions of Cokes during Annealing and Gasification. A series of reactions took place when the coke samples were treated at high temperatures either in N2 or in a simulated BF gas atmosphere, which resulted in weight loss of samples. The weight loss of coke samples after annealing and gasification at different stages is shown in Figure 3. The weight loss of all the cokes subjected to annealing and gasification increased with increasing treatment temperature. The weight loss of cokes during annealing in a nitrogen atmosphere C


Article

Energy & Fuels

the gasification, coke D was more reactive to CO2 (CRI = 46.7 %) in comparison with cokes A and C; the porosity change of coke D was also more notable during gasification. 3.3. Change in the Pore Structure of Cokes during Annealing and Gasification. Change in the pore structure is of significant interest in relation to the coke mechanical properties. The pore mean area, perimeter, equivalent circle diameter (average pore diameter), average pore roundness, and area fraction of pores with roundness ≤0.1 are presented in Table 3.

Table 2. Porosity of Cokes Subjected to Annealing (Cokes 2−5) and Gasification (Cokes 6−9) under the Simulated Blast Furnace Conditions, % coke A coke coke coke coke coke coke coke coke coke a

1 2 3 4 5 6 7 8 9

51.1 53.3 54.8 55.7 59.5 53.5 55.5 54.9 56.0

(0.12) (0.11) (0.14) (0.09) (0.12) (0.10) (0.12) (0.13) (0.11)

coke C a

66.3 69.0 70.5 71.3 72.6 67.2 68.4 68.7 69.3

(0.12) (0.09) (0.08) (0.08) (0.07) (0.10) (0.10) (0.09) (0.09)

coke D 57.9 58.7 59.8 59.2 61.4 58.3 60.6 63.1 63.6

(0.11) (0.14) (0.11) (0.11) (0.11) (0.15) (0.13) (0.10) (0.10)

Table 3. Parameters of the Pore Geometrya of Cokes Subjected to Annealing and Gasification under Simulated Blast Furnace Conditions

Relative standard deviation of measurement.

samples annealed under N2 at the same temperature shows that gasification had a greater effect on the porosity development, most notably in the case of coke D, which has high reactivity to CO2. The porosity of coke D after gasification at 1673 K (1400 °C) was 63.6 %, which is by 8.3 % higher than its porosity after annealing at the same temperature. Weight loss of cokes during gasification at temperatures below 1673 K (1400 °C) was caused by the Boudouard reaction, release of moisture, and carbon−minerals reactions. Further devolatilization at high temperatures as well as reactions of mineral matter with carbon were the main reasons of weight loss of cokes during annealing at 1673−2273 K (1400−2000 °C). It can be expected that all these reactions cause porosity development. Figure 4 shows the correlation of weight loss with porosity of cokes subjected to annealing and gasification. The porosity of cokes increased with increasing weight loss during both annealing and gasification. In the annealing tests at 1673 to 2273 K (1400 to 2000 °C), the relative increase in the weight loss of coke A with increasing annealing temperature was most significant. After annealing at 2273 K (2000 °C), the weight loss of coke A was 13.7 %, which was 4.5 times higher than its weight loss after annealing at 1673 K (1400 °C). Correspondingly, the porosity of coke A increased after annealing at 2273 K (2000 °C) by 11.7 % (relative precent) compared with its porosity after annealing at 1673 K (1400 °C), which was more significant than that of cokes C and D. In

coke A1 coke A2 coke A5 coke A9 coke C1 coke C2 coke C5 coke C9 coke D1 coke D2 coke D5 coke D9

perimeter, μm

equivalent circle diameter, μm

average pore roundness

area fraction of pores with roundness ≤0.1, %

16304

527

108

0.74

2.7

19457

605

115

0.67

8.9

25068

778

118

0.52

21.8

26095

675

126

0.72

12.2

16817

762

103

0.36

29.3

35423

1205

127

0.31

46.6

69871

2254

138

0.17

59.9

48601

1490

128

0.28

53.0

18760

599

110

0.66

13.0

19268

647

112

0.58

14.2

21175

750

115

0.47

26.2

35119

886

127

0.56

35.7

mean area, μm2

a Only pores with an area of more than 1000 μm2 are included in this table.

Figure 4. Correlation between porosity and weight loss during annealing and gasification. (a) Annealing; (b) gasification. D


Article

Energy & Fuels

Figure 5. Coke A before and after processing under simulated blast furnace conditions. (a) Original coke; (b) annealed at 1673 K (1400 °C) for 2 h (coke A2); (c) annealed at 2273 K (2000 °C) for 2 h (coke A5); (d) gasified at 1673 K (1400 °C) for 2 h (coke A9).

Figure 6. Coke C before and after processing under simulated blast furnace conditions. (a) Original coke; (b) annealed at 1673 K (1400 °C) for 2 h (coke C2); (c) annealed at 2273 K (2000 °C) for 2 h (coke C5); (d) gasified at 1673 K (1400 °C) for 2 h (coke C9).

Only pores with a size over 1000 μm2 were considered, due to smaller pores do not have a significant effect on the

mechanical properties of metallurgical cokes.10 The average pore diameter of three cokes before high temperature treatment E


Article

Energy & Fuels

Figure 7. Coke D before and after processing under simulated blast furnace conditions. (a) Original coke; (b) annealed at 1673 K (1400 °C) for 2 h (coke D2); (c) annealed at 2273 K (2000 °C) for 2 h (coke D5); (d) gasified at 1673 K (1400 °C) for 2 h (coke D9).

was in the range of 103−110 μm; it was similar for all three cokes. Annealing at temperatures from 1673 to 2273 K (1400 to 2000 °C) enlarged the average pore size. The effect of annealing was more significant on coke C than on cokes A and D. Before treatment, coke C had the smallest pore diameter; it increased by approximately 34 % after annealing at 2273 K (2000 °C). Gasification of all cokes at 1673 K (1400 °C) had a stronger effect on the pore size than annealing at the same temperature. The effect was more evident for coke D, where the pore size after gasification at 1673 K (1400 °C) was 13 % larger than after annealing at the same temperature. Average pore roundness and area fraction of pores with low roundness were calculated based on the measurement of pore area and perimeter. Original coke A had the highest average pore roundness, 0.74, and the lowest area fraction of low roundness pores, 2.7 %. The pore structure of original coke A consisted less coalescence sports in 2-D scheme; the pores of original cokes C and D were of a more complicated shape. The pore structure of the three cokes before and after treatment are shown in Figures 5−7. During the annealing process, the average pore roundness of all cokes decreased and the volume of low roundness pore increased as the annealing temperature rose. Significant changes in average pore roundness and volume of low roundness pores were observed for cokes A and C (Figures 5 and 6). In comparison, the change of pore structure of coke D was less significant (Table 3 and Figure 7). Gasification of cokes at 1673 K (1400 °C) for 2 h increased the volume of low roundness pores of all three cokes and decreased their average pore roundness. These data indicate that more coalescence spots appeared in pore area and pore transform into a more complicated structure after gasification.8 The effect of gasification at 1673 K (1400 °C) on pore

structure of cokes was more significant than annealing at the same temperature but smaller than annealing at 2273 K (2000 °C) except for coke D. Gasification made considerable change in the pore structure of coke D due to its high reactivity (CRI of coke D is 46.7 %). After gasification at 1673 K (1400 °C) for 2 h, the area fraction of pores with roundness ≤0.1 of coke D was 1.5 times larger than that after annealing at 1673 K (1400 °C) and 36 % larger than that after annealing at 2273 K (2000 °C). 3.4. Tensile Strength. The tensile strengths and the relative deviation of the measurements of original cokes and cokes after annealing and gasification are presented in Table 4. Among the original coke samples, coke A had the highest tensile strength of 7.71 MPa, whereas the measured values of tensile strength of cokes C and D were 4.62 and 5.49 MPa, respectively. For all three cokes, the tensile strength decreased with increasing annealing temperature. A decrease by 38 % was Table 4. Tensile Strength of Cokes Subjected to Annealing and Gasification, MPa coke A coke coke coke coke coke coke coke coke coke a

F

1 2 3 4 5 6 7 8 9

7.71 7.10 6.38 5.51 4.95 7.07 6.91 6.92 6.99

(0.31) (0.30) (0.26) (0.25) (0.26) (0.24) (0.31) (0.34) (0.19)

coke C a

4.62 4.30 3.82 3.35 2.86 4.28 4.06 3.81 3.68

(0.23) (0.20) (0.24) (0.21) (0.26) (0.23) (0.25) (0.27) (0.30)

coke D 5.49 5.23 5.10 5.05 4.83 5.14 4.82 4.53 4.36

(0.33) (0.42) (0.32) (0.30) (0.34) (0.37) (0.32) (0.36) (0.38)

Relative standard deviation of measurement. DOI: 10.1021/acs.energyfuels.5b02152 Energy Fuels XXXX, XXX, XXX−XXX

Article

Energy & Fuels observed for cokes A and C after annealing at 2273 K (2000 °C); a smaller, 12 %, decrease was in the case of coke D. Gasification also negatively affected the tensile strength of cokes. Compared with annealing at 1673 K (1400 °C, coke 2), gasification at the same temperature (cokes 8 and 9) had a more significant effect on the strength degradation of the cokes with higher reactivity (cokes C and D). The tensile strength of coke A decreased by 8 % after gasification upon heating to 1273 K (1000 °C), and then changed marginally with a further increase in gasification temperature. The tensile strength of cokes C and D visibly decreased with increasing gasification temperature. Gasification at 1673 K (1400 °C) had a stronger effect on the tensile strength of coke D than annealing at 2273 K (2000 °C): the tensile strength of coke D after gasification (cokes D8 and D9) was lower than the tensile strength of the annealed sample, which was opposite to the case of cokes A and C. 3.5. Factors Affecting Pore Structure during Annealing and Gasification. The thermodynamic data for the calculations were taken from NIST-JANAF thermochemical tables.27 In the gasification process, change in the pore structure of coke was mainly caused by the Boudouard reaction: C(s) + CO2(g) = 2CO(g), ΔG° = 162.31 − 0.1691T (kJ)

Figure 8. XRD spectra of mineral matter of original coke D and after annealing. (4)

The equilibrium partial pressures of CO in reactions 5 and 6 are shown in Figure 9. The CO equilibrium partial pressures in

The release of moisture, and carbon−mineral reactions at temperatures below 1673 K (1400 °C) also contributed to the change of pore structure during gasification.21 During the annealing, the changes of pore size and structure of coke were predominantly attributed to the reactions of minerals with carbon and further devolatilization at elevated temperature. XRD spectra of the ash of coke D annealed under the BF conditions are presented in Figure 8. Ashes of cokes A and C have similar XRD spectra. Hematite Fe2O3 in the remnant mineral matter of annealed coke was formed by the reoxidation of iron in the ashing process. The XRD pattern showed that the predominant (crystalline) phases of mineral matter in the coke before treatment were quartz and aluminosilicate, which is consistent with observations in previous works.28−30 After annealing at 1673 K (1400 °C), silicon carbide SiC and a small amount of silicon nitride Si3N4 were formed. The standard Gibbs free energy changes for overall reactions of SiC and Si3N4 formation were calculated using data from NIST-JANAF thermochemical tables,27 and presented in reactions 5 and 6.

Figure 9. Equilibrium partial pressure of CO for reactions 5 and 6.

SiO2(s) + 3C(s) = SiC(s) + 2CO(g), ΔG° = 590.4 − 0.3301T (kJ)

reactions 5 PCO(5) and (6) PCO(6), at 1673 K (1400 °C) are 0.254 and 0.294 atm, respectively. Partial pressure of nitrogen in reaction 6 was 1 atm. During the annealing of cokes in flowing N2, the formed CO was continuously removed from the system; therefore the actual partial pressure of CO in the system was much lower than the CO equilibrium partial pressure for reactions 5 and 6, which make reactions 5 and 6 thermodynamically feasible at 1673 K (1400 °C). Both SiC and Si3N4 were formed at 1673 K (1400 °C) under the given experimental conditions. At 2273 K (2000 °C), the equilibrium partial pressures of CO in reactions 5 and 6 increase to 68.8 and 15.9 atm, respectively, which indicates a strong thermodynamic potential for the reactions to occur. However, silicon nitride Si3N4 is not thermodynamically stable and decomposes at 2273 K (2000

(5)

3SiO2(s) + 6C(s) + 2N2(s) = Si3N4(s) + 6CO(g), ΔG° = 1260.1 − 0.6923T (kJ)

(6)

After annealing at 2273 K (2000 °C), silicon carbide (SiC) was the major inorganic phase in the coke, aluminum nitride (AlN) was also observed in the XRD pattern. Silicon nitride was not observed in this sample. The standard Gibbs free energy change for reactions 5 and 6 are zero at 1789 K (1516 °C) and 1820 K (1547 °C). However, under experimental conditions, CO partial pressure is lower than the standard pressure 1 atm; therefore, both SiC and Si3N4 can be synthesized at 1673 K (1400 °C). G


Article

Energy & Fuels

Figure 10. Effect of the pore structure on the coke mechanical strength: (a) area fraction of pores with roundness threshold ≤0.1; (b) average pore roundness.

were found as +25.71, +0.25, −0.66, +0.69, and +0.31 correspondingly. The correlation between tensile strength, microstrength, and pore structure of coke is presented by eq 10:

°C), so only silicon carbide SiC was detected in the ash of cokes annealed at this temperature. 3.6. Factors Affecting Coke Strength under the Simulated Blast Furnace Conditions. Coke strength is related to the strength of coke matrix, porosity, and pore structure. The empirical relationship between coke strength, its pore structure and matrix properties established in previous works1,2,6 is based on eq 7 derived by Knudsen:31 S = S0 × exp( −b × P)

σ = 25.71 × (1 − P) × K1c 0.25 × exp{−0.66 × [P′0.69 + (P − P′)0.31]}

(10)

A correlation between calculated and experimental data is shown in Figure 11. The coke tensile strength increases with

(7)

where S is strength of porous materials, S0 is parameter related to the matrix strength, b is constant, and P is porosity. The microstrength of cokes in this study determined by ultramicro indentation are reported elsewhere.32 A relationship between coke porosity, microstrength and coke tensile strength is presented by eq 8: σ = A × (1 − P) × K1c B × exp(C × P)

(8)

where σ is coke tensile strength (MPa), K1c is coke microstrength (MPa·m1/2), and P is the porosity of coke (%), A−C are constants. The pore structure of coke also affects the coke strength. The effects of the area fraction of pores with roundness threshold ≤0.1 and average roundness on the mechanical strength of coke are presented in Figure 10. Tensile strength of cokes decreased with increasing volume of low roundness pores and decreasing average pore roundness. This indicates that the coke strength decreased when the pore structure changed to a more complicated network with an increased number of coalescence points. It follows from plots in Figure 10 that the connectivity of the pores has a significant effect on the coke strength. To take into account the pore structure, eq 8 can be modified as

Figure 11. Correlation of tensile strength with coke microstrength, porosity, and pore structure.

microstrength, but it is inversely affected by porosity of both “connected pores” and “individual pores”, the effect of porosity of “connected pores” (P′) is stronger in comparison with the effect of porosity of “individual pores” (P−P′). Cokes with a large proportion of “connected pores” have a high level of defects that cause stress concentration. The significantly higher tensile strength of the original coke A in comparison with coke C (up to 40 %), can be attributed to the difference in their pore structure: the pore structure of coke C included a significantly higher amount of coalescence points

σ = a × (1 − P) × K1c b × exp{c × [P′d + (P − P′)e ]} (9)

where σ is coke tensile strength, K1c is coke microstrength, P is the total porosity of coke, P′ is porosity of “connected pores”, which is the area fraction of pores with roundness ≤0.1 (Table 3), (P−P′) is the porosity of “individual pores”, a−e are constants. The constants a−e in eq 9 were determined by the regression analysis using Excel, to fit calculated tensile strength to the experimental data with maximum R2. The constants a−e H


Article

Energy & Fuels

(4) Sato, H.; Patrick, J. W.; Walker, A. Effect of coal properties and porous structure on tensile strength of metallurgical coke. Fuel 1998, 77, 1203−1208. (5) Andriopoulos, N.; Loo, C. E.; Dukino, R.; McGuire, S. J. Microproperties of Australian coking coals. ISIJ Int. 2003, 43, 1528−1537. (6) Kim, S. Y.; Sasaki, Y. Simulation of effect of pore structure on coke strength using 3-dimensional discrete element method. ISIJ Int. 2010, 50, 813−821. (7) Saito, K.; Kunitomo, K.; Fukuda, K.; Katoh, K.; Komaki, I. Analysis of chemical reaction in the voids of cokes using NMR gas imaging. Tetsu to Hagane 2002, 88, 651−657. (8) Kubota, Y.; Nomura, S.; Arima, T.; Kato, K. Quantitative evaluation of relationship between coke strength and pore structure. ISIJ Int. 2011, 51, 1800−1808. (9) Ueoka, K.; Ogata, T.; Morozumi, Y.; Aoki, H.; Miura, T.; Uebo, K.; Fukuda, K. Evaluation of coke strength considering pore shapes by using a homogenization method. Tetsu to Hagane 2006, 92, 184−190. (10) Yamamoto, T.; Hanaoka, K.; Sakamoto, S.; Shimoyama, I.; Igawa, K.; Takeda, K. Effect of Coke Pore Structure on Coke Tensile Strength before/after CO2 Reaction and Surface-breakage Strength. Tetsu to Hagane 2006, 92, 206−212. (11) Hays, D.; Patrick, J. W.; Walker, A. Pore structure development during coal carbonization. 1. Behaviour of single coals. Fuel 1976, 55, 297−302. (12) Tomeczek, J.; Gil, S. Volatiles release and porosity evolution during high pressure coal pyrolysis. Fuel 2003, 82, 285−292. (13) Singla, P. K.; Miura, S.; Hudgins, R. R.; Silveston, P. L. Pore development during carbonization of coals. Fuel 1983, 62, 645−648. (14) Strugala, A. Changes of porosity during carbonization of bituminous coals: effects due to pores with radii less than 2500 nm. Fuel 2002, 81, 1119−1130. (15) Nomura, S.; Thomas, K. M. The effect of swelling pressure during coal carbonization on coke porosity. Fuel 1996, 75, 187−194. (16) Gray, R. J.; Champagne, P. E. Petrographic characteristics impacting the coal to coke transformation. In 47th Ironmaking Conference, Toronto, April 17−20, 1988; Iron and Steel Society: New York, 1988; pp 313−324. (17) Iwakiri, H.; Kamijo, T.; Kobayashi, I.; Kitamura, M. A fundamental study in new carbonization process at medium temperature for metallurgical coke. In 51st Ironmaking Conference, Toronto, April 5−8, 1992; Iron and Steel Society: New York, 1992; pp 581− 586. (18) Graham, J. P.; Wilkinson, H. C. Coal properties, charge preparation and their influence on coke quality. In Ironmaking Conference, 1978; Iron and Steel Society: New York, 1978; pp 421− 436. (19) Vander, T.; Alvarez, R.; Ferraro, M.; Fohl, J.; Hofherr, K.; Huart, J. M.; Mattila, E.; Propson, R.; Willmers, R.; Vdvelden, B. Coke quality improvement.Possibilities and limitations. In 3rd International Ironmaking Congress Proceedings, Gent, Belgium, September 16−18, 1996; Iron and Steel Society: New York, 1996; pp 16−18. (20) Simons, G. A. Coal pyrolysis I. Pore evolution theory. Combust. Flame 1983, 53, 83−92. (21) Xing, X.; Zhang, G.; Dell’Amico, M.; Ciezki, G.; Meng, Q.; Ostrovski, O. Effect of annealing on properties of carbonaceous materials. Part II: porosity and pore geometry. Metall. Mater. Trans. B 2013, 44, 862−869. (22) Senneca, O.; Salatino, P.; Masi, S. Microstructural changes and loss of gasification reactivity of chars upon heat treatment. Fuel 1998, 77, 1483−1493. (23) Pusz, S.; Krzesińska, M.; Smędowski, Ł.; Majewska, J.; Pilawa, B.; Kwiecińska, B. Changes in a coke structure due to reaction with carbon dioxide. Int. J. Coal Geol. 2010, 81, 287−292. (24) Gudenau, H.; Senk, D.; Fukada, K.; Babich, A.; Froehling, C. Coke Behavior in the Lower Part of BF with High Injection Rate. In International BF Lower Zone Symposium, Wollongong, Australia, 2002; pp 1−12.

between pores; whereas the pore structure of original coke A had a smaller coalescence area between pores.

4. CONCLUSIONS The degradation of cokes was studied under simulated blast furnace conditions that included annealing under N2 at temperatures up to 2273 K (2000 °C) and gasification with simulated blast furnace gas composition−temperature profiles to 1673 K (1400 °C). The major findings can be summarized as follows: (1) Both gasification and annealing decreased the mechanical strength of coke. Compared with annealing at 1673 K (1400 °C), gasification at the same temperature caused larger degradation for more reactive cokes. (2) Both gasification to 1673 K (1400 °C) and annealing in the temperature range of 1673−2273 K (1400−2000 °C) increased the porosity and enlarged the pore size of cokes. Compared with annealing at 1673 K (1400 °C), gasification at the same temperature had a stronger effect on the coke porosity and size. The porosity development and pore enlargement of the most reactive coke, coke D, upon gasification at 1673 K (1400 °C) was slightly higher than after annealing at 2273 K (2000 °C). (3) After annealing and gasification, all cokes showed a decrease in average pore roundness and an increase in the area fraction of low roundness pores. The effect of gasification at 1673 K (1400 °C) on the change of the pore structure of cokes A and C was more significant than annealing at the same temperature but smaller than annealing at 2273 K (2000 °C). Effect of gasification at 1673 K (1400 °C) on the pore structure of coke D with had high reactivity was stronger than effect of annealing at 2273 K (2000 °C). (4) The change of the coke pore structure during gasification was mainly as a result of the Boudouard reaction; release of moisture and carbon−mineral reactions occurring at temperatures below 1673 K (1400 °C) also contributed to the change of pore structure during gasification. The pore structure development during annealing was caused by the reactions of mineral matter with the carbon and further devolatilization. (5) The degradation of cokes upon gasification and annealing under the blast furnace conditions could be attributed to the modification of the pore structure with increasing number of coalescence points between pores, which increases the defects for stress concentration and the degradation of strength on coke matrix.

■

AUTHOR INFORMATION

Corresponding Author

*X. Xing. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This project was supported by BlueScope Steel, BHP Billiton and Australian Research Council (ARC Linkage Project LP130100701).

■

REFERENCES

(1) Xing, X.; Zhang, G.; Dell’Amico, M.; Ciezki, G.; Meng, Q.; Ostrovski, O. Effect of Annealing on Properties of Carbonaceous Materials. Part III: Macro and Microstrengths. Metall. Mater. Trans. B 2013, 44, 870−877. (2) Patrick, J. W.; Walker, A. Macroporosity in cokes: Its significance, measurement, and control. Carbon 1989, 27, 117−123. (3) Grant, M. G. K.; Chaklader, A. C. D.; Price, J. T. Factors affecting the strength of blast furnace coke. Fuel 1991, 70, 181−188. I


Article

Energy & Fuels (25) Shin, S. M.; Jung, S. M. Gasification Effect of Metallurgical Coke with CO2 and H2O on the Porosity and Macro-strength in the Temperature Range of 1100 to 1500 °C. Energy Fuels 2015, 29, 6849. (26) Ye, Z.; Gupta, S.; Kerkkonen, O.; Kanniala, R.; Sahajwalla, V. SiC and Ferro-silicides Formation in Tuyere Cokes. ISIJ Int. 2013, 53, 181−183. (27) Chase, M. W. NIST-JANAF Thermochemical Tables; American Chemical Society: Washington, DC, 1998; pp 550, 641, 643, 650, 1600, 1633, 1753. (28) Gupta, S.; Dubikova, M.; French, D.; Sahajwalla, V. Characterization of the origin and distribution of the minerals and phases in metallurgical cokes. Energy Fuels 2007, 21, 303−313. (29) Grigore, M.; Sakurovs, R.; French, D.; Sahajwalla, V. Influence of mineral matter on coke reactivity with carbon dioxide. ISIJ Int. 2006, 46, 503−512. (30) Grigore, M.; Sakurovs, R.; French, D.; Sahajwalla, V. Mineral reactions during coke gasification with carbon dioxide. Int. J. Coal Geol. 2008, 75, 213−224. (31) Knudsen, F. Dependence of mechanical strength of brittle polycrystalline specimens on porosity and grain size. J. Am. Ceram. Soc. 1959, 42, 376−387. (32) Xing, X.; Rogers, H.; Zhang, G.; Hockings, K.; Zulli, P.; Ostrovski, O. Coke Degradation under Simulated Blast Furnace Conditions. 2015, submitted.

J


Changes in Pore Structure of Metallurgical Cokes under Blast Furnace

Recommend Documents