Method of Making Iron Ore Pellets with a Carbon Core by Disc Pelletizer

Abstract

The pellet making method was studied for production of iron ore pellets with a carbon core. The pellets are to be fired in a sintering machine to produce sintered ores containing carbonaceous material. This new raw material for the blast furnace will have advantages such as high reducibility and low slag generation. In this study, pellet making tests were performed using laboratory and pilot scale disc pelletizers with 3 to 5 mm coke breeze, concentrated iron ores and additives in order to construct a method for producing pellets with a carbon core. Scaling-up of the method was considered with four discs with diameters of 0.6, 1.2, 2.0 and 3.5 m. As a result, it was found that there were two strategies for producing pellets with a carbon core at high yield: supplying a suitable level of water with a small droplet size, and spraying water in a specific area where the coke breeze flows on the surface of the contents in the disc. This study demonstrated that iron ore pellets with a carbon core can be obtained at a yield of more than 90% by using the disc pelletizer with the 3.5 m diameter. It was also estimated that a production capacity of 70 t/h can be expected with a 7.5 m diameter disc for commercial scale operation. This achievement of constructing a method for producing pellets with a carbon core was a breakthrough which overcame one of the technical hurdles for the development of sintered ores containing carbonaceous material.

1. Introduction

In order to decrease the amount of reducing agents consumed in blast furnaces, new raw materials for the blast furnace have been developed with the aim of drastically improving reaction efficiency.^{1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19)} Among such developments, if the conventional sintering machine can be used for commercialization, mass production taking advantage of existing equipment becomes realistic. Otomo et al. conducted an elementary research about production of a new agglomerate, which was an iron ore pellet with an anthracite particle core. Kamijo et al.¹⁹⁾ named this new agglomerate as CIS (Carbon Included Sinter) and conducted pot tests to produce CIS. They also confirmed the effect of improving reducibility and reducing pressure loss in softening and melting tests. Iwase et al.²⁰⁾ designed the structure and composition of a pellet with a carbon core suitable for firing in a sintering machine. However, a mass production method for granulating iron ore pellets with a carbon core has not been established.

Therefore, in this study, the pelletizing method was studied for production of iron ore pellets with a carbon core for firing in a sintering machine to produce sintered ores containing carbonaceous material, with the aim of developing a blast furnace burden with high reducibility and low slag generation. Based on a pellet making test using a laboratory disc pelletizer, the authors investigated a methodology for producing pellets with a carbon core at high yield, focusing mainly on control of water spraying and the operational parameters of the disc pelletizer. In addition, upscaling of the method was considered with four different disc sizes, and the production capacity using a commercial-size disc was estimated.

2. Experimental Method

2.1. Pellet Making at Lab Scale

Disc pelletizers with inner diameters of 0.6 and 1.2 m were used to study the pellet making conditions in order to establish a continuous process for producing two-layer pellets consist of a coke breeze particle core and a covering layer of iron ore powder. Table 1 shows the chemical composition and particle size of the raw materials used in this test. Ore A is a Brazilian porous hematite pulverized in the laboratory to an extent suitable for pellet making, and ore B is a pellet feed of Brazilian specular hematite. The coke breeze for the carbon core was classified before use between 3 mm or more and 5 mm or less. As an additive for adjusting the basicity of the covering layer, a quicklime of 200 mesh or less was used.

Table 1. Chemical and granulometric analysis of iron ores.

	Fe	FeO	SiO2	Al2O3	P	Mn	LOI	−45 μm	Blaine (cm2/g)
Ore A	65.60	0.11	2.10	1.52	0.025	0.420	2.00	82.4	2127
Ore B	67.60	0.14	2.30	0.44	0.030	0.050	0.50	78.7	1842

Figure 1 shows a schematic diagram of the pellet making setup at the lab scale. The iron ore and quicklime were blended so that the basicity, CaO/SiO₂, was 2.0, and were mixed with an intensive mixer. The mixed ore was charged into a hopper and fed at a specified feed rate by a conveyor belt. The coke breeze was fed onto the conveyor at a specified feed rate by a vibrating feeder, and then fed into the disc pelletizer together with the mixed ore. While feeding, water was sprayed on the materials in the pelletizer using a flat spray nozzle with a spray angle of 65° at a water pressure of 0.3 MPa. In order to investigate the effect of the initial moisture level of the iron ore, pellet making tests were conducted with a moisture content of 0 mass% with ore A as well as 8.5, 10.0, and 11.0 mass% with ore B. Table 2 shows the operating parameters of the pelletizer. Here, the Froude number, Fr, which is a dimensionless number, was introduced so that comparisons can be made between the pellet making conditions with different disc diameters and rotational speeds. Fr consists of the ratio of inertial force to gravity, and is expressed by Eq. (1) using the inner diameter of the disc, D, the rotation speed, N, and the gravitational acceleration, g.   

$$Fr=\frac{D\cdot {\left(N/60\right)}^2}{g}$$(1)

Fig. 1.

Schematic of pellet making process for pellets with a carbon core at lab scale. (Online version in color.)

Table 2. Dimensions and operational parameters of disc pelletizers.

	Lab scale		Pilot scale
Inner dia. (m)	0.6	1.2	2.0	3.5
Depth (mm)	150	200	270	300
Inclination (deg.)	45	45	45	47
Rotation speed (rpm)	17	14	11	7.5
Froude No. (x 103, -)	4.9	6.7	6.9	5.5

Table 3 shows the feed rate of each raw material and the water supply condition. In order to investigate the effect of the initial moisture level of iron ore on the yield of pellets with a carbon core, a pellet making study was also performed with the initial moisture of iron ore at 0, 8.5, 10.0 and 11.0%. Also, to investigate the effect of the droplet size for water spraying, spray nozzles with different droplet size of 200, 1000 and 2000 μm were applied. The target size of green pellets was 8–16 mm and produced pellets were sampled as needed to investigate the yield of pellets with a carbon core and to measure the moisture level. For the yield of pellets with a carbon core, 100 pellets were randomly sampled from every 10 kg of pellet making, and the presence or absence of coke breeze as the core particle was investigated. The moisture level of the pellets was obtained by sampling 5 pellets at the above same frequency and measuring the weight loss with a balance equipped with an infrared heating furnace.

Table 3. Raw materials and water supply conditions for pellet making tests with the disc pelletizer of 0.6 m diameter.

	Ore A	Ore B
Initial moisture of iron ore (mass%)	0	8.5	10.0	11.0
Moisture after mixing with quicklime (mass%)	0	6.5	7.9	9.0
Feeding rate (dry-kg/min.)	2.0	2.0	2.0	2.0
Coke rate (kg/min.)	0.04	0.04	0.04	0.04
Coke/mixed ore (mass%, dry basis)	2.0	2.0	2.0	2.0
Water spraying rate (L/min.)	0.20	0.06	0.02	0

2.2. Pellet Making at Pilot Scale

In order to investigate the upscaling and production capacity of the process for continuous production of pellets with a carbon core, pilot-scale pellet making tests were conducted using disc pelletizers with inner diameters of 2.0 and 3.5 m. Figure 2 shows the setup when using the 2.0 and 3.5 m pelletizer, and photo of the 3.5 m pelletizer during production of pellets with a carbon core. The iron ore and quicklime were fed at a specified feed rate by a screw feeder so that the basicity, CaO/SiO₂, was 2.0, and were charged into an intensive mixer. The holding amount of the ores in the mixer was controlled so that the residence time was 2 min. The coke breeze was fed onto the conveyor at a specified feed rate by a vibrating feeder and then fed into a pelletizer together with the mixed ore. Water was sprayed in both the mixer and pelletizer. Ore A was used as a raw material for pelletizers of 1.2 and 2.0 m diameter and ore B was used for the pelletizer of 3.5 m diameter. Table 2 shows the operating parameters of each pelletizer, and Table 4 shows the feed rate of each raw material and water. The target size of green pellets was 8–16 mm. Kamijo et al.²¹⁾ discussed that green pellets granulated by the disc pelletizer with 2.5 m diameter became larger than pellets by the pelletizer with 0.58 and 0.80 m diameter, because green balls could collapse and be re-granulated in a larger pelletizer. In this test, the operating parameters were adjusted so that even if the pelletizer was scaled up, pellets with the targeted size could be produced.

Fig. 2.

a) Schematic of process flow for pellets with a carbon core at pilot scale, and b) photo of the disc with 3.5 m diameter during production of pellets with a carbon core. (Online version in color.)

Table 4. Raw materials and water supply conditions for pellet making tests with the disc pelletizers of 1.2, 2.0 and 3.5 m diameter.

Disc dia. (m)	1.2	2.0	3.5
Initial moisture of iron ore (mass%)	0	0	9.6
Moisture after mixing with quicklime (mass%)	0	4.5	8.5
Feeding rate (dry-kg/min.)	12	50	167
Coke rate (kg/min.)	0.24	1.00	3.33
Coke/mixed ore (mass%, dry basis)	2.0	2.0	2.0
Water spraying rate (L/min.)	1.4*	5.2	–

*  Droplet size of water spray: 200, 1000, 2000 μm

3. Results and Discussions

3.1. Control of Water Supply

Iveson et al.²²⁾ presented an approach to understanding the wet granulation operation by dividing the operation into three processes, namely, (i) wetting & nucleation, (ii) consolidation & growth and (iii) attrition & breakage. If this approach is applied to pellet making with a disc pelletizer, it can be considered as shown in Fig. 3. Raw feed materials and aggregates in the way of growing are wetted by spraying with water while sliding down from the top of the disc, and nuclei are generated by the liquid-bridging force consisting of the capillary force and viscosity of water (wetting & nucleation). Immediately after this, they receive dropping impact at the bottom of the disc, and the nuclei collide with each other or collide with the raw feed materials and consolidate. Subsequently, while they pass under the product-sized pellets by the rotation of the disc, they are compacted and grown by the load from the materials on them (consolidation & growth). On the other hand, there is also a possibility of attrition or breakage for the nuclei and their agglomerates when dropping impact or load acts on them (attrition & breakage). In short, “consolidation & growth” and “attrition & breakage” are competitive processes. Overall, if consolidation & growth prevails, the nuclei and their agglomerates will grow into pellets, and if attrition & breakage prevails, they will not grow into product pellets even after the residence time, and will discharged as-is from the disc. The borderline between these two conditions depends on whether or not a sufficient amount of water has been supplied. If a sufficient amount of water is supplied, the nuclei and their agglomerates will have enough liquid-bridging force to prevent attrition and breakage even if subjected to dropping impact or load, and thus will continue growing. However, if the water supply is excessive, agglomerates and product-sized pellets will coalesce by liquid-bridging, making stable pellet production impossible. This means that the lower limit of the water supply for stable pellet production is the amount necessary for the nuclei and their agglomerates to have the liquid-bridging force required to withstand attrition and breakage, and the upper limit is the maximum moisture level in the range where the product-sized pellets do not coalesce with each other. This suitable moisture range for stable pellet making is generally not particularly wide, and it is difficult to predict in advance from information such as the raw material properties and disc operating parameters. Therefore, in a continuous pellet making process, it is crucial to first determine the suitable moisture range in which pellets can be stably produced, and then control the water supply rate to fill the gap between the target moisture level and the initial moisture content of raw materials. In the lab-scale tests, the suitable moisture range for the mixed ores was 9.5±0.5 mass% for ore A and 8.5±0.5 mass% for ore B. Ore A required more water due to the following factors: As shown in Table 1, ore A had more particles with a size of −45 μm than ore B and had a larger Blaine specific surface area (BSA), so more water was required to wet the surface of each particle. From the viewpoint of the ore structure, ore A was porous and ore B was dense, so it is estimated that the porous ore A had to be supplied with a larger amount of water than ore B to fulfill pores inside the ore particles.

Fig. 3.

Schematic of modern approach for granulation process²²⁾ and its application to iron ore pellet making. (Online version in color.)

3.2. Control of Nucleation Rate

In order to produce pellets stably, while it is important to control the amount of water supplied within a suitable range, the size of the water droplets sprayed from the spray nozzle is also a key factor that affects the size of the product pellets. Moreover, if the purpose is to produce pellets with a carbon core as in this test, not only the product size but also the pellet yield would be affected. Figure 4 shows the effect of the size of the water droplets sprayed from the water nozzle on the yield of pellets with a carbon core. Compared to nozzles with the droplet size of 1000 μm or more, a nozzle with a droplet size of 200 μm improved yield up to 90%. The reason why the droplet size affects yield is considered as follows.

Fig. 4.

Yield of pellets with a carbon core vs. droplet size of water spraying in a pellet making test with a disc pelletizer of 1.2 m diameter. (Online version in color.)

Lister et al.²³⁾ conducted an experiment in which water droplets of different sizes were sprayed onto a powder bed flow, and showed quantitatively that smaller droplet diameters resulted in the formation of smaller aggregates, and conversely, larger droplet diameters produced larger aggregates. This means that in granulation with a disc pelletizer with a constant water supply rate, if the droplet diameter is smaller, many small aggregates will form in the pan, and if the droplet diameter is larger, smaller amount of large aggregates will form. When the aggregates grow beyond a certain size (e.g., size of 1 mm or more in the case of iron ore pellet making), they become nuclei for granulation and eventually grow into product pellets. The formation rate of the nuclei that reach a certain size or more, that is, the nucleation rate, determines the final number and size of the product pellets. In ordinary pellet making, the raw material feed rate is constant and the amount of water sprayed is controlled within a suitable range. Under this condition, if the nucleation rate is too high, more of the supplied raw material than necessary will be consumed for the generation of new nuclei and will not be supplied sufficiently for nuclei growth, resulting in the production of a large number of small pellets. On the other hand, if the nucleation rate is too low, more raw material than necessary will be supplied to the nuclei, resulting in the production of pellets larger than the target size. An appropriate nucleation rate produces an appropriate amount of nuclei that will later become pellets while also supplying the necessary amount of raw material for nuclei growth, leading to the production of pellets of the target size. Since the purpose of this research is to produce two-layer pellets with a carbon core, the ideal nucleation rate is the feed rate of the coke breeze. If only the mixed ore agglomerates and grows to form nuclei, the yield of pellets with a carbon core decreases, so nucleation of the mixed ore should be minimized. In Fig. 4, the yield of pellets with a carbon core decreased when using water spraying nozzles with the droplet diameters of 1000 or 2000 μm because the coarse droplets produced a large number of agglomerated nuclei of the mixed ore, and these nuclei grew into product pellets without incorporating coke breeze. Therefore, nozzles were selected to be capable of spraying as small a droplet as possible so as to suppress the formation of nuclei consisting only of the mixed ore, while also ensuring the necessary amount of water supply. A nozzle with the droplet size of 150 μm was used for the disc pelletizer with diameter of 0.6 m and a nozzle with a 200 μm droplet was used for the disc with the 1.2 m diameter.

3.3. Operation Parameters of Disc Pelletizer

There are two strategies for producing pellets with a carbon core at high yield: supplying a suitable level of water with a small droplet size, and spraying the water at a specific spot where coke breeze flows on the surface of the material flow in the disc. Since the suitable level of water supply and the droplet size have been discussed in the Chapter 3.1 and 3.2, respectively, the specific spot to be sprayed and the operating parameters of the disc which define that part are discussed below. When the coke breeze which is to become the core for granulation is wetted by water spraying and covered with the mixed ore and then grows, the desired two-layer pellets with a carbon core are formed. If the mixed ore is not directly sprayed with water, it is possible to prevent the formation of agglomerated nuclei which do not contain coke breeze due to the liquid-bridging force, and to suppress the production of pellets having no carbon core. In order to spray water only on the coke breeze, it is necessary to determine the area where it flows in the disc pelletizer. Figure 5 shows a photo of the inside of the disc pelletizer with the 1.2 m inner diameter after suspending the rotation of the disc during pellet making. It was found that there were three types of contents in the disc: (i) not-granulated feed material of the mixed ore, (ii) thinly coated coke nuclei and (iii) product pellets. However, it was difficult to distinguish the area where the coke breeze flows from the difference in the color of the coke and mixed ore since the coke breeze is thinly covered with the mixed ore immediately after being fed into the disc. Therefore, in order to visualize the area where the coke breeze flows in the disc pelletizer, the above three types of contents (i) to (iii) were represented by alumina spheres of different colors, as shown in Fig. 6. The not-granulated feed material of the mixed ore (i) was represented by white alumina spheres with a size of 1 mm, the thinly coated coke nuclei (ii) were represented by black spheres 5 mm in diameter, and the product pellets (iii) were represented by red spheres 15 mm in diameter, and the effect of the disc rotation speed on the movement of these three types of alumina spheres was visualized. When the rotation speed of the disc was too low, as in Fig. 6(a), it was difficult to spray water only on the part where the black spheres representing the coke nuclei flowed on the surface of the contents inside the disc because the three types of spheres move in a narrow area in the disc. As the disc rotation speed increased, the flow areas of the red, black and white spheres expanded. In the condition shown in Fig. 6(b), the area where the black spheres flowed on the surface of the contents inside the disc was relatively wide, and it was possible to spray water selectively on only the black spheres. However, when the rotation speed was too high, as shown in Fig. 6(c), the area where the red spheres representing product pellets flowed also expanded excessively, while the area where the black spheres flowed became narrower. When the rotation speed was further increased, as shown in Fig. 6(d), a so-called doughnut-like state of movement occurred, in which a void space was generated in the middle of the disc, and stable pellet making was not considered to be possible.

Fig. 5.

Three contents inside a disc of 1.2 m diameter for pellet making with a carbon core: (i) not-granulated feed material of mixed ore, (ii) thinly coated coke nuclei and (iii) product pellets. (Online version in color.)

Fig. 6.

Photos of physical simulation using three types of alumina balls at four different rotation speeds for disc with inner diameter of 600 mm (e.g. white & 1 mm: represents not-granulated feed material, black & 5 mm: represents thinly coated coke nuclei and red & 15 mm: represents product pellets). (Online version in color.)

Figure 7 shows the relationship between the ratio of the area where the black spheres flowed on the surface to the bottom area of the disc, S_coke ratio, and the Froude number, Fr, which is related to disc size and rotation. When Fr (x 10³) was in the range of 3 to 7, the S_coke ratio was 0.3 or more, which made it relatively feasible to spray the black spheres representing coke breeze. In this test, as shown in Table 2, the rotation speed of the discs with the inner diameters of 0.6 m, 1.2 m, 2.0 m and 3.5 m were 17 rpm (Fr = 4.9), 14 rpm (Fr = 6.7), 11 rpm (Fr = 6.9) and 7.5 rpm (Fr = 5.5), respectively. As described above, in order to obtain a high yield of pellets with a carbon core, it is necessary to create a state in which the coke breeze in the disc can be sprayed selectively with water. In other words, it is necessary to adjust the disc rotation speed so that (i) not-granulated feed material of the mixed ore, (ii) thinly coated coke nuclei and (iii) product pellets move and flow in the disc in a well-classified manner.

Fig. 7.

S_coke ratio vs. Froude number for disc rotation (S_coke ratio is the area fraction in which coke breeze flow on surface of contents inside disc). (Online version in color.)

3.4. Water Spraying Spot

The water spraying spot for supplying water to the coke breeze was determined based on the consideration in Chapter 3.3 and taking into account the configuration of other equipment such as the scrapers and conveyor feeder of the raw materials. Figure 8 shows the water spraying spot and the three material streams of (i) not-granulated feed material of the mixed ore, (ii) thinly coated coke nuclei and (iii) product pellets during pellet making using a disc pelletizer with an inner diameter of 0.6 m. In conventional granulation of iron ore pellets without a carbon core, as shown in Fig. 3, the norm is to supply an appropriate amount of water on the area where the not-granulated feed material flows in the disc and control the nucleation rate so that the product reaches the target size. However, in granulation of two-layer structured pellets with a carbon core, the nuclei particles are supplied from outside the disc as coke breeze, so it is crucial to limit the water spraying spot to the area where (ii) thinly coated coke nuclei flow on the surface of the material flow in the disc so as to avoid generating excessive nuclei made of (i) not-granulated feed material of the mixed ore as much as possible.

Fig. 8.

Illustration of water spraying spot and material movement of three kinds of content overlaid on photo of disc pelletizer making pellets with a carbon core (e.g., (i) not-granulated feed material of mixed ore, (ii) thinly coated coke nuclei and (iii) product pellets). (Online version in color.)

3.5. Yield of Pellets with a Carbon Core at Lab Scale

Based on the study discussed in Chapters 3.1 to 3.4, pellet making operation was conducted in line with the operation parameters of the disc pelletizer and the water supplying conditions verified in the previous chapters. Figure 9 shows the yield of pellets with a carbon core and the moisture level of the pellets when using ore A with an initial moisture of 0 mass%. The yield of pellets with a carbon core remained at above 90%, and after pelletizing a total of 100 kg, pellets with a carbon core were produced at an average yield of 97%. In almost all cases, a single coke breeze particle was present as a core at the center of the pellets. This is because even if multiple coke breeze particles adhered to each other due to the liquid bridge effect of water, they were forcibly separated by the dynamic movement in the pelletizer so that the mixed ore coated a single coke breeze. The moisture level of the pellets was almost constant at 9 mass%.

Fig. 9.

Yield of pellets with a carbon core and moisture of pellets in lab scale pellet making test. (Online version in color.)

Generally, in the pendular to funicular states, i.e., a three-phase packing structure of the solid, liquid and gas phases, which is usually aimed at in pellet making, the liquid-bridging force increases with the water content, which increases the probability that raw material particles will agglomerate. In the steel industry, iron ores are usually stored in stockpiles in outdoor yards and have some initial moisture. If the initial moisture of the iron ore is high, a large number of agglomerated nuclei with a size of 1 mm or more, which are agglomerates of only ore particles, may form and grow, and in this case, the ratio of pellets without a carbon core will increase. In order to investigate the effect of the initial moisture content of the iron ore on the making of pellets with a carbon core, the iron ore B was collected after arrival at a stock yard, and two cases were studied: use of ore B in the pellet making tests without drying, and addition of more moisture to simulate rainfall. As shown in Table 3, the initial moisture levels of the iron ores were 8.5, 10.0 and 11.0 mass%, which declined to 6.5, 7.9 and 8.8 mass%, respectively, after mixing with quicklime. The main reason for the decline of the moisture level was attributed to consumption of the water in the mixed ore by the hydration reaction with quicklime, together with slight evaporation due to heat generation by the exothermic reaction. A sufficient amount of water was supplied in the pelletizer to fill the gap between the target moisture of the pellets and the moisture level of the mixed ore after mixing with the quicklime. Figure 10 shows the effect of the initial moisture level of the iron ore on the yield of pellets with a carbon core. The yield was 90% or more up to the initial moisture of 10 mass%, but deteriorated to 55% when the initial moisture was 11 mass%. If the initial moisture is as high as 10 mass%, even if it exceeds the target level of the product, it declines as a result of mixing with quicklime, and this reduces the liquid-bridge force among mixed ore particles that causes agglomeration. However, if the initial moisture exceeds a certain threshold, addition of quicklime cannot prevent agglomeration of the mixed ore particles. This promotes the formation and growth of agglomerated nuclei in which only mixed ore particles are agglomerated, and greatly reduces the yield of pellets with a carbon core.

Fig. 10.

Yield of pellets with a carbon core vs. initial moisture of iron ore in lab scale pellet making test. (Online version in color.)

For production of pellets with a carbon core with higher productivity, scale-up tests were carried out using the disc pelletizers with 1.2 m and 2.0 m inner diameters. Figure 11 shows the relationship between the yield of pellets with a carbon core and the inner diameter of the disc pelletizers. Even when the pelletizers with inner diameters of 1.2 m and 2.0 m were used, pellets with a carbon core could be produced with a yield of more than 90% by considering the water spraying method and operating parameters of the disc pelletizer, as detailed in the following discussion.

Fig. 11.

Yield of pellets with a carbon core vs. diameter of disc pelletizers. (Online version in color.)

3.6. Scale-up for Making Pellets with a Carbon Core

Figure 12 shows the production capacity of the four pelletizers investigated in this test in comparison with the production capacity of conventional iron ore pellets without a carbon core. In the case of the pellets with a carbon core, the rate of increase in the production capacity, i.e. the slope in Fig. 12, on a scale-up of the disc was almost the same, but the production capacity was lower than that of conventional pellets when using a pelletizer of the same size. Based on this comparison, the production capacity of about 70 t/h per unit can be extrapolated when two-layer structured pellets with a carbon core are produced by the largest unit, which has an inner diameter of 7.5 m. In the case of conventional pellets without a carbon core, the production capacity is about 130 t/h at the same scale, so when producing pellets with a carbon core, the production capacity is estimated to be 54% of the conventional level. One of the main reasons for the lower capacity can be attributed to the difference in the method of supplying water. In conventional pellet production, it is sufficient simply to supply water to the part where the not-granulated feed material flows, and this area usually becomes wider as the disc becomes larger. Thus, there are few constraints on the spraying area, and the only limitation is to control the nucleation rate so as to obtain pellets of the target size. Specifically, spraying nozzles should be selected so that the water droplet size does not become excessively large, and the number of spraying spots should be multiplied so as not to cause locally excess moisture. Alternatively, it is also possible to pre-moisturize the feed material to the suitable level for pellet making before feeding it to the disk. However, when producing pellets with a carbon core, the water spraying spot is limited to the area in the disc where the coke breeze flows on the surface of the contents inside the disc. When attempting to increase production capacity, the amount of water supply to a disc will increase, but the area that can be sprayed will be relatively narrow. Spraying water in a limited area causes excess moisture locally, which results in a large number of nuclei formed only from the mixed ore and coalescence of pellets in the middle of the growth process. This will not only reduces the yield of pellets with a carbon core, but also hinders stable pellet making. Thus, when producing pellets with a carbon core, the limited area for spraying water becomes a constraint, and the rate of supplying water and the Froude number cannot be increased as much as when producing conventional pellets without a carbon core. As a result, the feed rate of the raw materials is also restricted, and the production capacity cannot be increased as much as the optimum level discussed by Suzuki et al.²⁴⁾ The residence time of the fed material in a disc was about 3 min for conventional pellets, while it was 5–6 min for pellets with a carbon core. Regarding this constraint, there is room for consideration to improve the production capacity by pre-moisturizing the feed material of the mixed ore as much as possible before it is fed into the disc pelletizer and reducing the amount of water sprayed in the disc.

Fig. 12.

Production capacity vs. disc diameter for comparison of productivity of conventional iron ore pellets (reference) and double-layered pellets with a carbon core (this experiment). (Online version in color.)

4. Conclusions

A pellet making method for production of pellets with a carbon core was studied for the development of sintered ores containing carbonaceous material. Pellet making tests were performed using laboratory- and pilot-scale disc pelletizers, and the following findings were obtained.

(1) For stable pellet making and product size control, it was crucial to control the amount of water supplied and the nucleation rate.

(2) It was also important to adjust the rotation speed of the disc so that the not-granulated feed material, coke breezes and product pellets flowed in the disc in a well-classified manner.

(3) Continuous production of pellets with a carbon core at a yield of more than 90% was achieved by spraying water in a suitable amount as small droplets in a specific area where the coke breeze flows on the surface of the contents in the disc pelletizer.

(4) The watering rate was a constraint due to the limited water spraying area, resulting in a production capacity of about 54% compared to conventional pellets.

Nomenclature

D: Inner diameter of a disc pelletizer (m)

Fr: Froude number for rotation of a disc (-)

g: Gravitational acceleration (m/s²)

N: Rotation speed of a disc (rpm)

References

• 1)   S.  Nomura,  H.  Terashima,  E.  Sato and  M.  Naito: Tetsu-to-Hagané, 92 (2006), 849 (in Japanese). https://doi.org/10.2355%2Ftetsutohagane1955.92.12_849
• 2)   T.  Yamamoto,  T.  Sato,  H.  Fujimoto,  T.  Anyashiki,  K.  Fukada,  M.  Sato,  K.  Takeda and  T.  Ariyama: Tetsu-to-Hagané, 97 (2011), 501 (in Japanese). https://doi.org/10.2355%2Ftetsutohagane.97.501
• 3)   T.  Kon,  S.  Natsui,  S.  Ueda,  R.  Inoue,  S.  Enoeda and  T.  Ariyama: Tetsu-to-Hagané, 100 (2014), 593 (in Japanese). https://doi.org/10.2355%2Ftetsutohagane.100.593
• 4)   T.  Anyashiki,  H.  Fujimoto,  T.  Yamamoto,  T.  Sato,  H.  Matsuno,  M.  Sato and  K.  Takeda: Tetsu-to-Hagané, 101 (2015), 515 (in Japanese). https://doi.org/10.2355%2Ftetsutohagane.TETSU-2015-038
• 5)   M.  Ando,  Y.  Iwai,  H.  Fujimoto,  T.  Sato,  Y.  Hashimoto,  T.  Anyashiki and  M.  Sato: Tetsu-to-Hagané, 104 (2018), 543 (in Japanese). https://doi.org/10.2355%2Ftetsutohagane.TETSU-2018-042
• 6)   S.  Nomura,  K.  Higuchi,  K.  Kunitomo and  M.  Naito: ISIJ Int., 50 (2010), 1388. https://doi.org/10.2355%2Fisijinternational.50.1388
• 7)   K.  Higuchi,  S.  Nomura,  K.  Kunitomo,  H.  Yokoyama and  M.  Naito: ISIJ Int., 51 (2011), 1308. https://doi.org/10.2355%2Fisijinternational.51.1308
• 8)   K.  Higuchi,  S.  Nomura,  K.  Kunitomo,  H.  Yokoyama and  M.  Naito: Tetsu-to-Hagané, 98 (2012), 517 (in Japanese). https://doi.org/10.2355%2Ftetsutohagane.98.517
• 9)   A.  Kasai,  Y.  Matsui,  F.  Noma,  H.  Iwakiri and  M.  Shimizu: Tetsu-to-Hagané, 87 (2001), 313 (in Japanese). https://doi.org/10.2355%2Ftetsutohagane1955.87.5_313
• 10)   A.  Kasai,  H.  Toyota,  K.  Nozawa and  S.  Kitayama: ISIJ Int., 51 (2011), 1333. https://doi.org/10.2355%2Fisijinternational.51.1333
• 11)   A.  Kasai,  M.  Naito,  Y.  Matsui and  Y.  Yamagata: Tetsu-to-Hagané, 89 (2003), 1212 (in Japanese). https://doi.org/10.2355%2Ftetsutohagane1955.89.12_1212
• 12)   A.  Uchida,  T.  Kanai,  Y.  Yamazaki,  K.  Hiraki,  Y.  Saito,  H.  Aoki,  N.  Komatsu,  N.  Okuyama and  M.  Hamaguchi: ISIJ Int., 53 (2013), 403. https://doi.org/10.2355%2Fisijinternational.53.403
• 13)   A.  Uchida,  Y.  Yamazaki,  S.  Matsuo,  Y.  Saito,  Y.  Matsushita,  H.  Aoki and  M.  Hamaguchi: ISIJ Int., 56 (2016), 2132. https://doi.org/10.2355%2Fisijinternational.ISIJINT-2016-339
• 14)   A.  Uchida,  Y.  Yamazaki,  S.  Matsuo,  Y.  Saito,  Y.  Matsushita,  H.  Aoki and  M.  Hamaguchi: ISIJ Int., 57 (2017), 1524. https://doi.org/10.2355%2Fisijinternational.ISIJINT-2016-339
• 15)   H.  Go,  Y.  Kuwano,  K.  Suzuki,  T.  Chang,  M.  Matsuzaki,  S.  Nakamura,  E.  Tsuzi and  M.  Tate: Tetsu-to-Hagané, 66 (1980), 1870 (in Japanese). https://doi.org/10.2355%2Ftetsutohagane1955.66.13_1870
• 16)   H.  Yokoyama,  K.  Higuchi,  T.  Ito and  A.  Oshio: ISIJ Int., 52 (2012), 2000. https://doi.org/10.2355%2Fisijinternational.52.2000
• 17)   K.  Higuchi,  H.  Yokoyama,  H.  Sato,  M.  Chiba and  S.  Nomura: ISIJ Int., 57 (2017), 55. https://doi.org/10.2355%2Fisijinternational.ISIJINT-2016-418
• 18)   H.  Sato,  S.  Machida,  K.  Nushiro,  K.  Ichikawa,  M.  Sato,  T.  Ariyama and  K.  Takeda: Tetsu-to-Hagané, 92 (2006), 815 (in Japanese). https://doi.org/10.2355%2Ftetsutohagane1955.92.12_815
• 19)   C.  Kamijo,  M.  Matsumura and  T.  Kawaguchi: ISIJ Int., 49 (2009), 1498. https://doi.org/10.2355%2Fisijinternational.49.1498
• 20)   K.  Iwase,  T.  Higuchi,  T.  Yamamoto and  T.  Murakami: ISIJ Int., 62 (2022), 1178. https://doi.org/10.2355/isijinternational.ISIJINT-2021-590
• 21)   C.  Kamijo,  M.  Hara,  Y.  Yamaguchi,  M.  Yoshikawa,  J.  Kano,  M.  Matsumura and  T.  Kawaguchi: ISIJ Int., 53 (2013), 1497. https://doi.org/10.2355/isijinternational.53.1497
• 22)   S. M.  Iveson,  J. D.  Litster,  K.  Hapgood and  B. J.  Ennis: Powder Technol., 117 (2001), 3. https://doi.org/10.1016/S0032-5910(01)00313-8
• 23)   J. D.  Litster,  K. P.  Hapgood,  J. N.  Michaels,  A.  Sims,  M.  Roberts,  S. K.  Kameneni and  T.  Hsu: Powder Technol., 114 (2001), 32. https://doi.org/10.1016/S0032-5910(00)00259-X
• 24)   S.  Suzuki,  K.  Sato and  M.  Fujimoto: Tetsu-to-Hagané, 73 (1987), 1932 (in Japanese). https://doi.org/10.2355/tetsutohagane1955.73.15_1932