The effect of disintegrated iron-ore pellet dust on deposit formation in a pilot-scale pulverized coal combustion furnace. Part I: Characterization of process gas particles and deposits

Highlights

• Addition of pellet dust changed the particle size distribution (PSD) slightly.

• A transition from a bimodal PSD to a trimodal PSD was observed.

• Coal-ash globules attached to the hematite particles suggested potential for slagging.

• Addition of pellet dust reduced the porosity of the deposits.

• Addition of pellet dust increased the slagging tendency.

Abstract

To initiate the elucidation of deposit formation during the iron-ore pelletization process, a comprehensive set of experiments was conducted in a 0.4 MW pilot-scale pulverized-coal-fired furnace where three different scenarios were considered as follows; Case 1 (reference case): Coal was combusted without the presence of pellet dust. Case 2: Natural gas was combusted together with simultaneous addition of pellet dust to the gas stream. Case 3: Coal was combusted together with the addition of pellet dust simulating the situation in the large-scale grate-kiln setup. Particles and deposits were sampled from 3 positions of different temperature via a water-cooled sampling probe. Three distinct fragmentation modes were identified based on the aerodynamic particle diameter (D_p). The fine mode: Particles with 0.03 < D_p < 0.06 μm. The first fragmentation mode: Particles with 1 < D_p < 10 μm. The second fragmentation mode: Coarse particles (cyclone particles, D_p > 10 μm). A transition from a bimodal PSD (particle size distribution) to a trimodal PSD was observed when pellet dust was added (Case 3) and consequently the elemental bulk composition of the abovementioned modes was changed. The most extensive interaction between pellet dust and coal-ash particles was observed in the coarse mode where a significant number of coal ash globules were found attached to the surface of the hematite particles. The morphology of the sharp-edged hematite particles was changed to smooth-edged round particles which proved that hematite particles must have interacted with the surrounding aluminosilicate glassy phase originating from the coal ash. The short-term deposits collected during coal combustion (Case 1) were highly porous in contrast to the high degree of sintering observed in the experiments with pellet dust addition (Case 3) which is attributed to the dissolution of hematite particles in the aluminosilicate glassy phase. The results suggest that pellet dust itself (Case 2) has low slagging tendency, independent of temperature. However, when coal-ash is present (Case 3), auxiliary phases are added such that tenacious particles are formed and slagging occurs.

Introduction

Iron ore is one of the most important natural resources with a reportedly annual world production of over 2 billion metric tons in 2017 [1]. Approximately 98% of the world's usable iron ore is consumed in the steelmaking industry [1], following which process the mined iron ore can be directly used as lumps or shaped into pellets to be reduced either in direct-reduction plants or in blast furnaces. During the pelletizing process, the iron ore is converted into a powder through a three-step crushing stage, then mixed with additives and a specific binder, and eventually rolled into green pellets in rolling drums. The 9–15 mm (in diameter) “green” pellets are then to be sintered in an induration machine [2,3]. The two most commonly-used processes for pelletizing are the travelling-grate process [commonly used for hematite (Fe₂O₃) mined ore], and the so-called grate-kiln process which is often used for magnetite (Fe₃O₄) mined ore. In the travelling-grate process, pellets are transported on a moving grate (in a stationary bed) undergoing drying, oxidation, sintering, and cooling which all occur on the very same grate. On the contrary, the grate-kiln process (which is the focus here) uses a shorter grate, and oxidation/sintering take place in a kiln which is a rotating cylindrical furnace (similar to that used in cement production) that allows a more homogeneous sintering of the pellets with the pellets' total residence time of approximately 30 min [4].

Fig. 1 shows a schematic of a commonly-used grate-kiln setup located in the 40 MWth pelletizing plant KK2 (LKAB No.2 installation in Kiruna, Sweden) owned by the Swedish mining company LKAB (Luossavaara-Kiirunavaara Aktiebolag). The machine consists of a 60 m long grate along which green pellets undergo drying, preheating, and oxidation, at the end of the grate there comes a rotating cylindrical furnace (kiln) wherein pellets are subjected to sintering at elevated temperatures while tumbling and descending along the kiln. The kiln is followed by an annular cooler where cold air is blown through the sintered finished products. The grate is divided into four distinct zones: the updraft drying (UDD), down draft drying (DDD), temperate preheat zone (TPH), and the preheat zone (PH). Hot flue gas (∼1200 °C) from the kiln and cooler together with the exothermic energy from the oxidation of the pellets (magnetite to hematite) heats up the bed. The partly oxidized pellets from the grate are discharged in the transfer chute (marked in Fig. 1) and then transported through the rotary kiln. Estimated temperatures in the kiln are ∼1200 °C at the inlet, while in the burner zone, the temperature reaches ∼1350 °C and the flame temperature reaches up to 1700–1800 °C. To reach an adequately high process temperature in the kiln which is essential for sintering, pulverized coal with a thermal power of 40 MW is used as the main fuel, however, fuel oil is also used when starting up the kiln and/or when encountering problems with the coal supply [5,6]. Preheated air (∼1200 °C) from the first zone of the cooler is used as combustion air. The air/fuel ratio corresponds to an excess air level of 16% O₂ in the gas stream leaving the kiln.

Coal ash particles together with disintegrated iron-ore pellet dust particles accumulate on the refractory walls, resulting in the build-up of scaffold deposits, most drastically in the hot areas (e.g., fireside slagging closer to the burner, transfer chute, inside the rotary kiln, and the beginning of the cooler) [6]. Consequently, the build-up of these sintered deposits disturbs the flow of gas and pellets and eventually results in now-and-again unscheduled stoppages for mechanical removal of the deposited layer. It has also been observed that the accumulated deposits in rotary kilns of iron-ore pelletizing plants, not only cause mechanical strains but also degrade the refractory lining over time due to high temperature corrosion [7]. Furthermore, scaffolds adversely affect the process efficiency, and quality of the product.

Effects of coal minerals on ash formation in pulverized-coal-fired boilers have been extensively studied [[8], [9], [10], [11], [12]], on the contrary, there is a limited understanding of ash deposition phenomena in iron-ore pelletizing rotary kilns which explains why there is very limited information in this context in the literature [[13], [14], [15], [16], [17], [18], [19]]. In coal combustion fly ashes form through two main mechanisms giving rise to a bimodal particle size distribution. The bulk of the mineral matter disintegrates into relatively large ash particles ranging from 0.5 to 20 μm in diameter through the fusion of mineral matter on the surface of the burning char particles. However, smaller particles (fume particles, <0.5 μm in diameter) form through volatilization and subsequent re-condensation of a smaller fraction of the mineral matter and constitute the sub-micron mode of the particle size distribution [[20], [21], [22], [23], [24], [25]]. Thereafter, deposits build up from the fly ash particles through a combination of the following mechanisms; inertial impaction, diffusion of fume, thermophoresis, condensation, and chemical reactions [26,27].

Compared to pulverized coal fired boilers, there are similarities along with differences regarding ash deposition phenomena in the rotary kilns of iron ore pelletizing plants. In some respects, ash transformation mechanisms in the great-kiln process are relatively similar to those occurring in pulverized-coal-fired boilers, specifically in the high temperature regions closer to the flame. However, once the ash particles are formed, their transportation to the surfaces and subsequent deposit formation mechanisms can be significantly different than that in pulverized coal fired boilers. Several governing factors influence the ash-deposition phenomena in pulverized coal combustion such as the coal type (ash composition, melting temperature of the ash, and distribution of the mineral matter), not to mention the reaction atmosphere, the temperature of the fly ash particles, flow dynamics, the temperature of the surface onto which ash-deposition takes place and so forth. Several reviews addressing ash-deposition characteristics have already been reported in the literature [28].

In contrast to pulverized coal fired boilers, a grate-kiln process is characterized by a longer residence time, a highly oxidizing atmosphere, and the presence of recirculating alkalis and disintegrated iron-ore pellet dust in the flue gas. Given the foregoing, ash deposition in a grate-kiln process is a much more complex phenomenon compared to ordinary pulverized coal-fired boilers and is affected by several factors including the process conditions, the chemical properties of iron-ore pellets and coal-ash and their potential interactions with one another and with the refractory walls. Concisely, the interaction between fly ash particles from the combustion of coal and the inevitably entrained Fe-rich particles (arising from the disintegration of iron-ore pellets) makes it a cumbersome task to address the respective ash transformation phenomena. This study was initiated with the objective of investigating the effect of disintegrated pellet dust particles on melt formation (liquid slag) and, thus, on deposit formation in a grate-kiln process. Furthermore, this study was inspired by our investigations and findings from our previously conducted pilot-scale [6,29] and full-scale measurement campaigns [15,16] which proved there was a considerable difference in the rate of deposition and the properties of the resulting deposits when iron ore pellet dust was present. In light of this, several experiments were performed in a pilot-scale (0.4 MW) Experimental Combustion Furnace (ECF) with the objective of addressing the effect of disintegrated iron-ore pellet dust on deposit formation in a pilot-scale pulverized coal combustion furnace.

The results can expectantly be used by iron-ore pelletizing plants to mitigate the formation of deposits and increase the production rate of pelletized iron ore. Characterization of fly ash particles and deposits is presented in this paper whereas thermochemical considerations and viscosity estimations are discussed in part II of this work.