Effect of Coated Cow Dung on Fluidization Reduction of Fine Iron Ore particles

Abstract

The effects of reduction temperature, gas linear velocity, reduction pressure, reduction time, and reducing gas on the fluidized ironmaking process were studied for the fine iron Newman ore particles (0.154–0.178 mm) and the optimal experimental operating conditions were obtained. Under the optimal conditions, the effects of the coated cow dung on the reduction of fine iron ore particles were studied, and the inhibition mechanism of cow dung on particle adhesion in the fluidized ironmaking process was elucidated. The experimental results show that the optimal operating parameters are linear velocity of 0.6 m/s, reduction pressure of 0.2 MPa, reduction temperature of 1023 K, H₂ as the reducing gas, and reduction time of 60 min. Cow dung can react with oxide in the ore powder to form a high melting point substance that can form a certain isolation layer, inhibit the growth of iron whiskers, and improve the fluidization.

1. Introduction

The bonding of fine iron ore particles in the fluidized ironmaking process is complex and involves four bonding mechanisms, namely, the iron whisker bonding mechanism, the newly formed metal iron bonding mechanism, the low melting point eutectic bonding mechanism, and the van der Waals force and field bonding mechanism. The bonding mechanisms of iron whiskers and newly formed metal iron have been intensely investigated and are now well-understood. Many studies have been carried out on the fluidized direct reduction technology in a global research effort aiming to effectively inhibit particle adhesion. J.M. Pang et al. [1] conducted in-depth study on the reduction mechanism, reaction thermodynamics, and kinetics of fine iron ore particles. Zhong et al. [2] proposed a lattice gas simulation method for gas–solid two-phase flow, and used it to study and analyze the basic characteristics of gas–solid two-phase flow in a fluidized ironmaking reactor. Dang et al. [3] established a reduced unreacted core model and carried out numerical simulations. Bonalde et al. [4,5] calculated the apparent activation energy of the reduction of fine iron ore particles with different reducing gases, and Zou et al. [6,7] performed preliminary calculations of the kinetics of the reduction of fine iron ore particles and determined the rate control mechanism of the reduction reaction under different conditions. Chironea et al. [8,9,10] found that low-frequency and high-intensity acoustic and magnetic fields can induce ferromagnetic materials to form needle-like substances along the magnetic line of force in the bed, producing a crushing effect on the agglomeration force of newly precipitated iron agglomerates. This hinders the growth of the newly precipitated metallic iron and effectively improves the fluidization performance of the particles and reduces the interparticle adhesion. These results provide an important scientific basis for the industrial application and development of a fluidized direct reduction process.

In this work, to develop an environmentally friendly fluidized direct reduction with low energy consumption and simple and efficient production process, cow dung as a coating agent was added to the fine iron ore particles. The process parameters were optimized by varying the gas velocity, reduction pressure, reduction temperature, and reduction time. The self-designed hot visible pressurized fluidized bed was used to reduce Newman ore powder (Newman powder and block ore is hematite from Newman hill ore in East Pilbara, Australia, with good sintering performance. The grade of powder is about 62.5%, and that of the block is about 65%). Parameters for economic, convenient, and effective operation were obtained by optimizing the metallization rate, sticking ratio, and surface morphology of the reduced fine iron ore particles. The mechanism for the inhibition of the loss of adhesion of fine iron ore particles by cow dung was studied and elucidated, providing reference data and theoretical basis for industrial use of fluidized direct reduction.

2. Experiment

2.1. Experimental Materials

In this experiment, Australian Newman ore was used for the reduced fine iron ore particles, and its chemical composition and scanning electron microscopy (SEM) image are shown in

Table 1. Chemical composition (mass %) of Newman ore powder in Australia.

Composition	TFe	SiO2	CaO	MgO	Al2O3	S	P
Content	62.5	4.37	0.07	0.10	2.08	0.099	0.088

and Figure 1, respectively. The size range of the Newman ore particles is 0.154–0.178 mm. Cow dung was collected from Dongzheng animal husbandry Co., Ltd., Jiangxia District, Wuhan City, Hubei Province. After being dried and ground into powder, it was sieved through 100 mesh (0.154 mm) and sealed for storage. Cow dung with 5%, 10%, 15%, and 20% mass fraction was used as the coating agent. The chemical composition of cow manure is organic matter, and the specific composition is shown in

Table 2. Chemical composition of cow dung (wt %).

Composition	C	N	H	S	O
Content	36.21	2.39	5.14	1.32	54.95

. The reduction gas consists of H₂, CO, and the mixture of H₂ and CO. N₂ with a purity of 99.99% was used as the protective gas. T_Fe is the mass fraction of iron in Newman ore.

2.2. Experimental Equipment

A pressurized visible fluidized bed is the main equipment used in this study (Figure 2 and Figure 3). The double stainless steel tube is used as the reactor, and the inner tube is the fluidized bed. The gas enters the fluidized bed after preheating in the middle layer between the outer tube and the inner tube, due to the heating cabinet contained in the outer tube. The different flow rates of CO, H₂, and N₂ are regulated and controlled by multiple flowmeters so that the gas entering the fluidized bed meets the composition and linear velocity requirements of the experiment. The gas mass flow controller is used to control the inlet flow of the fluidized bed through the reduction of the fine iron ore particles in the inner tube. The linear velocity at the inlet of the fluidized bed is regulated by the gas mass flow controller, and the reduction temperature of the fluidized bed is measured by a thermocouple. Prior to the experiment, inert gas was introduced to check the air tightness of the device, and after each experiment, inert gas was introduced to protect the ore powder from oxidation after reduction. During the experiment, the reduction status of fine iron ore particles is monitored through the observation window in real time. After the experiment, the bond ratio is measured to evaluate the bonding condition.

2.3. Experimental Scheme and Steps

Prior to the experiment, N₂ was introduced into the fluidized bed, and after the air was evacuated, the outlet valve was closed to increase the pressure of the fluidized bed to 0.5–0.6 MPa. The air tightness of the equipment was checked, and then the temperature of the fluidized bed was raised. When the reduction temperature reaches the target experimental temperature, the N₂ valve is closed and the reducing gas used in the experiment is injected. In this process, the pressure is controlled by the tightness of the valve, and the experiment starts when the pressure is stabilized at the set pressure. In the experiment, fine iron ore particles (20 g) and prepared dried coating materials were weighed each time, and then the two were mixed evenly in the mixer, only one additive was used in each experiment. The experiment was stopped after reaching the set reduction time. After cooling to room temperature, the bonded and unbonded mineral powder was removed from the fluidized bed and the bond ratio was calculated. The bond ratio is the ratio of the binding mass of the reducing powder to the total mass of the reducing powder. A smaller bonding ratio corresponds to a better fluidization state. Then, the metallic iron (M_Fe) and total iron (T_Fe) contents were determined by the potassium dichromate volumetric method and the ferric chloride titration method, and the metallization rate η was calculated. A higher metallization rate corresponds to better quality of the reduced ore powder, and lower sticking ratio corresponds to better fluidization of the reduction process. Therefore, the metallization rate and bond ratio were selected as the indices for the determination of the effect of fluidization reduction. The metallization rate is calculated as follows:

$${\mathsf{\eta }=\mathrm{M}}_{{\mathrm{F}}_{\mathrm{e}}}/{\mathrm{T}}_{\mathrm{Fe}}$$

(1)

where η is metallization rate; M_Fe is metallic iron, g; and T_Fe is total iron, g.

The sticking ratio is calculated as follows:

Ω = M_sticking/M_total

(2)

In the formula, M_total is the total mass of reduced materials, which is used to evaluate the quality of total materials after reduction of fine iron ore, g; Ω is the binding ratio; and M_sticking is the quality of adhesion material after reduction (refers to the lower part of the fluidized bed gas distribution plate that part of the ore powder), which is used to evaluate the quality of adherent material after reduction of fine iron ore, g;

The total mass of the material after reduction can be calculated as follows:

M_total = M_sticking + M_unsticking

(3)

In order to reduce the experimental error, we made three measurements and took the average value as the final experimental result. Because the bond loss of fine iron ore powder is affected by reduction temperature, reduction time, reduction pressure, linear velocity of reducing gas, and type of reducing gas, these factors are investigated from three aspects. The five factors are represented by letters A, B, C, D, and E, where A represents reduction temperature, with A₁, A₂, and A₃ representing 923 K, 1023 K, and 1123 K respectively; B represents reduction time, with B₁, B₂ and B₃ representing 20 min, 40 min, and 60 min, respectively; C represents reduction pressure, with C₁, C₂, and C₃ representing atmospheric pressure of 0.1 MPa, 0.2 MPa, and 0.3 MPa, respectively; D represents linear velocity of reduction gas, with D₁, D₂, and D₃ representing 0.4 m/s, 0.6 m/s, and 0.8 m/s, respectively; E represents the type of reducing gas, with E₁, E₂, and E₃ representing H₂, CO and H₂, and CO mixture, respectively (see

Table 3. Factor levels.

A Temperature/K			B Reduction Time/min			C The Reducing Pressure/MPa			D Linear Velocity/m/s			E Type of Reducing Gas
A1	A2	A3	B1	B2	B3	C1	C2	C3	D1	D2	D3	E1	E2	E3
923	1023	1123	20	40	60	0.1	0.2	0.3	0.4	0.6	0.8	H2	CO	Mixture

). The optimal operation parameters were obtained by orthogonal experiment. Under the optimal operation parameters, the bonding mechanism of fine iron ore and the influence of cow dung on bond loss of Newman ore were analyzed.

3. Experimental Results and Discussion

3.1. Optimum Operating Parameters

The metallization rate and bonding ratio measured in the experiment are shown in

Table 4. Orthogonal experimental scheme and results.

	Reduction Temperature/K	Reduction Time/min	Reduction Pressure/MPa	Linear Velocity/m/s	Type of Reducing Gas	Metallization Rate/%	Sticking Ratio /%
1	923	20	0.1	0.4	H2	77.67	27.98
2	923	40	0.2	0.6	CO	51.17	20.12
3	923	60	0.3	0.8	H2 + CO	73.58	29.36
4	1023	20	0.1	0.6	CO	56.87	26.36
5	1023	40	0.2	0.8	H2 + CO	69.04	32.47
6	1023	60	0.3	0.4	H2	83.44	48.63
7	1123	20	0.2	0.4	H2 + CO	65.34	33.67
8	1123	40	0.3	0.6	H2	91.1	45.73
9	1123	60	0.1	0.8	CO	49.2	38.92
10	923	20	0.3	0.8	CO	40.29	17.98
11	923	40	0.1	0.4	H2 + CO	63.15	25.58
12	923	60	0.2	0.6	H2	89.61	35.59
13	1023	20	0.2	0.8	H2	89.7	35.15
14	1023	40	0.3	0.4	CO	60.46	24.97
15	1023	60	0.1	0.6	H2 + CO	67.69	34.25
16	1123	20	0.3	0.6	H2 + CO	67.98	34.53
17	1123	40	0.1	0.8	H2	87.45	46.64
18	1123	60	0.2	0.4	CO	62.87	37.16

,

Table 5. Result analysis of metallization rate.

Factor	Indicators	A	B	C	D	E
Metallization Rate %	K1	395.47	397.85	402.03	412.93	518.97
	K2	427.2	422.37	427.73	424.42	320.86
	K3	423.94	426.39	416.85	409.26	406.78
	k1	65.91	66.31	67.01	68.82	86.50
	k2	71.20	70.40	71.29	70.74	53.48
	k3	70.66	71.07	69.48	68.21	67.80
	R	5.29	4.76	4.28	2.53	33.02
	Primary and Secondary Factors EABCD Optimization Scheme E1A2B3C2D2

and

Table 6. Result analysis of sticking ratio.

Factor	Indicators	A	B	C	D	E
Sticking Ratio %	K1	156.61	175.67	199.73	197.99	239.72
	K2	201.83	195.51	194.16	196.58	165.51
	K3	236.65	223.91	201.2	200.52	189.86
	k1	26.10	29.28	33.29	33.00	39.95
	k2	33.64	32.59	32.36	32.76	27.59
	k3	39.44	37.32	33.53	33.42	31.64
	R	13.34	8.04	1.17	0.66	12.37
	Primary and Secondary Factors AEBCD Optimization Scheme A1E2B1C2D2

Note: Ki represents the sum of the test results for the corresponding level number in each column; ki = Ki/s where s is the number of times the level appears on any column, s = 6 in this table; the range R indicates that it affects the Newman ore and the primary and secondary of powder reduction are the maximum value of each factor level minus the minimum value, R = k_max − k_min. The larger the R value, the more obvious the reduction effect of this factor on Newman ore.

.

Through the range comparison of five factors, the optimum process parameters of metallization rate are determined as E₁, A₂, B₃, C₂, D₂, namely: reduction gas, H₂; reduction temperature, 1023 K; reduction time, 60 min; reduction pressure, 0.2 MPa; and gas linear velocity, 0.6 m/s. The main factor affecting metallization rate is the type of reducing gas. The results show that the optimum process parameters are A₁, E₂, B₁, C₂, and D₂, namely: reduction temperature, 1023 K; reduction gas, CO; reduction time, 20 min; reduction pressure, 0.2 MPa; and gas linear velocity, 0.6 m/s. The main factor affecting the bond ratio is reduction temperature.

Using the above-described experiments, we have identified the optimal operating conditions as follows: reduction temperature of 1023 K, linear velocity of 0.6 m/s, reduction time of 60 min, reduction pressure of 0.2 MPa, and H₂ as the reduction gas. Using these conditions, we first performed a set of preliminary experiments without any coating composition. We obtained the metallization rate of Newman ore and the bond ratio of its Newman ore of 60.97% and 45.39%, respectively.

3.2. Study of Inhibition of Bond Loss by Cow Dung

3.2.1. Effect of Coated Cow Dung on the Metallization Rate and Sticking Ratio

In this work, by controlling the temperature, pressure, gas linear velocity, and other variables, the effect of different coating materials on the reduction effect of fine iron ore particles and the influence of the coating material content on the reduction effect of the fine iron ore particles were studied.

To better study the effect of cow dung on the fluidization quality of fine iron ore particles, MgO (density is 2940 kg/m³, with porous morphology), plastic particles (PP, PE, etc.) and carbon (powder) were selected to compare with cow dung and their contents were set to 5%, 10%, 15%, and 20%, respectively, in order to explore the influence of content on fluidization quality. The results of these experiments are shown in

Table 7. Metallization rate of Newman ore with different coating materials and compositions.

	Content	5%	10%	15%	20%
Component
MgO		63.11	68.22	75.53	70.58
Plastic Particles		65.16	72.56	82.95	79.42
Carbon		60.34	65.88	70.89	65.54
Cow dung		72.28	76.95	89.98	80.78

and

Table 8. Sticking ratio of Newman ore with different coating materials and compositions.

	Content	5%	10%	15%	20%
Component
MgO		39.05	30.54	25.26	35.78
Plastic particles		35.58	28.46	23.41	32.64
Carbon		40.55	36.74	28.04	38.89
Cow dung		32.05	26.83	20.78	28.62

.

To more clearly demonstrate the effect of the coating composition and content on the metallization rate and sticking ratio, we use the data in Table 7 and Table 8 to plot the histograms shown in Figure 4 and Figure 5. It is observed from Figure 4 and Figure 5 that when the mass fraction of the coating is 5–15%, the metallization rate will gradually increase and the sticking ratio will gradually decrease when the flow loss occurs. When the coating content is in the 15–20% range, the metallization rate will decrease and the sticking ratio will increase. Meanwhile, two different coating materials with the same mass fraction show quite different effects of improving the metallization rate at a flow loss. The best effect is obtained by coating cow dung, and in particular with 15% cow dung coating, the metallization rate reaches 89.98%, and the sticking ratio is only 20.78%, greatly improving the fluidization.

To further analyze the influence of cow dung on the fluidization, the metallization rate and sticking ratio of cow dung coated were plotted versus cow dung content, as shown in Figure 6. It was observed that when the content of cow dung is 5–10%, the metallization rate increases slowly, indicating that the reduction effect of fine iron ore particles is not pronounced due to the small coating content. However, when the content of cow dung is between 10% and 15%, the metallization rate first increases strongly and then decreases with an increase in coated cow dung content. When the cow dung content is between 5% and 15%, the sticking ratio and the cow dung content show a linear relationship, with the sticking ratio decreasing with increasing coating content, and reaching the lowest values for the coating content of approximately 15%. Therefore, it is concluded that 15% coated cow dung is optimal and obtains the best reduction effect on Newman ore.

Through the above experiments, we can know that 15% cow dung coating has the best reduction effect on Newman ore. In order to determine the best coating amount of cow dung, we complete experiments for cow dung content between 10% and 20% with small increments (2%). The experimental results are shown in

Table 9. Effect of cow dung content on metallization rate and sticking ratio.

	Content	10%	12%	14%	16%	18%	20%
Index
Metallization Rate		76.95	80.84	85.78	91.35	84.12	80.78
Sticking Ratio		26.83	24.32	21.63	20.02	24.68	28.62

. From Table 9, we can observe that 16% cow dung has the best reduction effect on Newman ore.

3.2.2. Effect of Coated Cow Dung on Phase Structure

The results of the X-ray diffraction (XRD, instrument model: D8ADVANCE, Manufacturer: German Brooke company. Main technical specifications: target material: Cu Target; power: 3 KW; scanning mode of goniometer: θ/θ scanning range: −3–150 degrees; accuracy of goniometer: 0.0001 degree; 2θ angle accuracy: ≤0.02 degree) analysis of different samples before and after reduction are shown in Figure 7. It was found that the main phases of the fine iron ore particles (sample a) are Fe₃O₄ and Fe₂O₃ transformed into FeO and Fe (Figure 7c); the phase of the fine iron ore particles coated with 15% cow dung shows little change from the original sample, and the main phase is Fe₃O₄ and other gangue phases. However, after fluidized reduction, the iron phase is mainly composed of Fe₃C, metal Fe, and some FeO (Figure 7d).

Figure 8 shows the SEM images of the morphology of the original Newman ore before reduction and after reduction with cow dung coating. It is observed from Figure 8a that the surfaces of the original fine iron ore particles are relatively dense, with a small amount of small gangue particles and fine iron ore particles. It is observed from Figure 8b that a large number of dense iron whiskers are present before reduction of fine iron ore particles. According to Figure 8c, a large number of colloidal substances are precipitated on the surface, possibly due to the reaction between the oxide in the cow dung and the ore powder at high temperature after coating the cow dung. By comparing Figure 8b with Figure 8d, we found that the whiskers were significantly reduced and the distribution was sparse after covering cow dung. Therefore, we conclude that the Newman ore is prone to bond loss most likely due to the formation of iron whiskers on its surface. After reduction with cow dung, cattle dung reacts with the oxide in the fine iron ore particles to form new substances that inhibit the formation of iron whiskers to a certain extent, and therefore effectively inhibits the bond loss.

4. Analysis of the Mechanism of Fine Iron Ore Particles Bond Loss

4.1. Characteristics of Fe₂O₃ Particles Bond Loss Behavior in Fluidization Reduction

The bond loss of the fine iron ore particles in the process of fluidized reduction is not only related to the fluid properties of gas and ore powder, but also to the physical and chemical properties of the ore powder surface. The change in the physical and chemical properties of the ore powder surface is mainly caused by the reduction reaction. Therefore, the bond loss of flow is quite a complex problem. The bonding forms of fine iron ore particles can be divided into three types: (1) the iron whiskers produced by reduction are interconnected to create bonding between the particles; (2) the surface energy of the freshly precipitated iron is high, favoring the bonding of the particles to each other; and (3) low melting point compounds appear in the ore powder, causing the particles to soften, or the liquid phase appears locally, causing the particles to bond. These conclusions are based on the studies that examined the changes of the ore particles before and after the reduction in order to infer the bonding mechanism. However, the bonding occurs during the reduction process, and there have been few studies of the bonding mechanism through the examination of the evolution of mineral powder particles in the reduction process. In addition, both iron whiskers and fresh iron are metallic iron precipitated by the reduction, but the relationship between bonding and the properties of the precipitated iron has not been studied.

The development process of bond loss can be observed through the observation window installed on the side of the fluidized bed. Upon reaching a certain degree of reduction, the particles at the bottom of the material layer first agglomerate and stop fluidization. As the reduction continues, the height of the fixed bed in the bonding area increases gradually until fluidization is stopped for the whole material layer. According to the classification of the fluidized particles by Geldart [11], the particles used in this experiment are class B particles that present the fluidized state of a bubbling bed in the process of gas-solid fluidization. The static pressure in the bed is directly proportional to the depth and density of the material layer. Therefore, the bottom of the bed is subjected to the maximum pressure and has the maximum density of the material layer. The stirring effect of the gas on the bed is the lowest at the bottom of the bed, so that the bonding is initiated at the bottom of the bed. As shown in Figure 9b, the particles with adhesive surfaces are sintered with the sieve plate. These particles lead to the increase in the gas resistance and the uneven distribution of gas flow, promoting the occurrence of bonding. Some vertical pipes with a diameter of 3–5 mm appear in the bonded fixed bed, as shown in Figure 9a. When the gas flows through these pipes, it is subjected to almost no resistance from the fluidized particles, so that the bed pressure drops sharply [12]. In some cases, the bed surface shows normal fluidization, but some large agglomerated particles have been formed inside, as shown in Figure 9c.

4.2. Mechanism of the Inhibition of the Loss of Fine Iron Ore by Cow Dung Coating

4.2.1. Effects of Cow Dung on Thermodynamics and Kinetics

Under high-temperature conditions, cow dung will undergo pyrolysis or gasification reaction, generating CO, CO₂, CH₄, H₂, and other gaseous small-molecule C_nH_m hydrocarbons, and at the same time also generating heavy hydrocarbons such as tar and coke. The thermogravimetric analysis (TG) and differential thermogravimetric analysis (DTG) curves of cow dung are shown in Figure 10.

With increasing reaction temperature, the pyrolysis process of cow dung can be divided into three stages. The first stage is the dehydration stage and occurs from room temperature to 105 °C. In this stage, the TG curve decreases gently, and the DTG curve is stable at a high value; this stage involves the rapid volatilization of free water with the increasing temperature. In the second stage, between 105 and 190 °C, the TG curve is almost horizontal and the DTG curve decreases slightly, which is less in cow dung. In this stage, the bound water and a small amount of volatile fat soluble substances are evaporated by heating. The main volatiles of the pyrolysis stage occur from 190 °C to 600 °C, and the TG curve once again decreases rapidly, and then again changes to a slow decrease. The DTG curve presents a large reaction valley. In this stage, the fat soluble matter, hemicellulose, cellulose, and lignin in the original cow dung will decompose successively [13]. Finally, in the slow pyrolysis and carbonization stage that occurs from 600 °C until the end of the reaction, the TG curve decreased slowly and the DTG curve shows a small reaction valley in the temperature range of 620–750 °C. In this stage, the changes are mainly due to the decomposition reaction of some minerals in cow dung and the carbonization reaction of some pyrolysis products. The main chemical reactions of cow dung under high temperature are as follows:

biomass→H₂ + CO₂ + C_nH_m + tar + char

(4)

C + CO₂→2CO

(5)

C + H₂O→H₂ + CO

(6)

C + 2H₂O→2H₂ + CO₂

(7)

CO + H₂O→H₂ + CO

(8)

C + 2H₂→CH₄

(9)

CH₄ + H₂O→3H₂ + CO

(10)

CH₄ + 2H₂O→4H₂ + CO₂

(11)

Tar + H₂O→H₂ + CO₂ + CO + C_nH_m

(12)

C_nH_m + H₂O→H₂ + CO₂ + CO

(13)

When the temperature rises to approximately 200 °C, the thermal decomposition reaction of cow dung begins in which the organic functional groups break and recombine, releasing a large amount of volatile matter. The reaction is mainly carried out according to Equation (4). When the pyrolysis temperature is lower than 600 °C, the main components of the gas are CO and CO₂, and are mainly generated by the decomposition of hemicellulose and cellulose. Hemicellulose is a heterogeneous polymer composed of several different types of monosaccharides such as pentose and hexasaccharide, while cellulose is a macromolecular polysaccharide composed of glucose [14]. CO and CO₂ may originate from the breakage of glycosyl linkage, i.e., the glycosidic bond [15]. With the increase in the temperature and the extension of reaction time, the residual moisture of raw materials is further removed in the low-temperature carbonization stage and exists in the free state, while the solid products will further react to form CO and H₂ according to Equations (5) and (6), and then H₂, CO₂, and CH₄ are generated according to Equations (8) and (9). When the gasification temperature is 400 °C, the volume fraction of CO₂ is still large, indicating that this stage is mainly due to the breaking of the glycosidic bond in hemicellulose and cellulose. This breaking is due to the reaction of water vapor with CO₂ and CO to form CH₄ and H₂ (Equations (7), (9), and (10)). With the increase in the temperature and residence time, the reaction is intensified. The contents of CH₄ and C_nH_m decrease gradually due to the reaction described by Equations (10), (11), and (13). When the gasification temperature is higher than 500 °C, the volume fraction of H₂ increases rapidly, and the gas quality is improved at high temperature (approximately 700 °C). However, when the temperature rises from 800 °C to 900 °C, the volume fraction of CH₄ and C_nH_m decreases gradually. The rate of this increase is low, and decreases slightly with increasing steam temperature. The above experiments show that cow dung will decompose to produce a large amount of H₂ at high temperature, and this added H₂ will act as the reducing gas. Hydrogen reduction of iron oxide is an endothermic reaction that greatly reduces the energy required for the growth of iron whiskers, inhibits their growth, and improves the fluidization quality.

4.2.2. Model of Inhibition of Fine Iron Ore Particles Bond Loss by Cow Dung

The surface modification mechanism of mineral powder coated with antibinder can be summarized as consisting of physical adsorption and chemical adsorption. Physical adsorption originates from the van der Waals force [16], capillary force, and magnetic force. These results [16] show that the van der Waals force on alumina particles with a diameter of 1 μm is 10⁷ times greater than the force of gravity on the alumina matrix. The van der Waals force on the particles on a general surface as shown in Figure 11a can be expressed quantitatively by Equation (14). When the coating agent particles such as cow dung and Fe₂O₃ particles adhere to the matrix at high temperature, deformation may occur on their contact surface, as shown in Figure 11c. At this time, the van der Waals force increases rapidly, and its quantitative description changes into the form of Equation (15). In addition, when the particles are at high temperature and do not cover any components, the stress on the particles is shown in Figure 11d. When the particles are coated with a layer of cow dung, the surface of the matrix is loose and porous at high temperature, as observed for the surface of Fe₂O₃ particles after reduction. The coating agent particles are adsorbed in these holes as observed from the experimental results presented in Figure 11e. The form is shown in Figure 11f. The van der Waals force between the particles and the matrix will further increase due to the increase in the contact surface:

$$F_{vdw}=\frac{hr}{8\pi s^2}$$

(14)

$$F_{vdw}=\frac{hr}{8\pi s^2}+\frac{h{\delta }^2}{8\pi s^3}$$

(15)

where: F_vdw—van der Waals force, N; r—diameter of micro particles, m; h—Liftshitz–van der Waals force constant, fluctuating between 0.6 and 9.0, ev; s—distance between microparticles and substrate surface, m; and γ—radius of bonding area, m.

Chemisorption is due to the chemical reaction or the generation of a solid solution between the microparticles and matrix. The adhesion force of chemical adsorption is much higher than that of physical adsorption. Cow dung can react with Fe₂O₃, Al₂O₃, and SiO₂ on the surface of mineral powder particles at high temperature to realize the chemical adsorption of the coating agent.

The contact between cow dung and Fe₂O₃ particles in the briquetting method is stronger than that in the slurry method. When Fe₂O₃ particles are reduced to the critical point of bond loss, the viscosity of Fe₂O₃ surface also reaches the maximum value. At this time, due to the reduction reaction, many loose pores appear on the surface, so that it is expected to obtain a good coating effect by the use of an additional coating agent at this time.

The powder method used in this experiment is the most simple and cost-effective of the four coating methods. Similar to the other three coating methods, when the cow dung coating amount is 15%, the powder method can also be carried out. The on-line method shows a better inhibition effect than the powder method, and this method can be used in the actual production by installing a powder spray gun in the fluidized bed. Compared to the powder method, the agglomeration sintering crushing method and the slurry method do not show a pronounced inhibition of bond loss. Additionally, the operation of these two coating methods is complex and their implementation costs are higher than that of the powder method.

In the previous research on the mechanism of the inhibition of the bond loss of fine iron ore particles by cow dung coating [17,18,19,20,21], it was convincingly demonstrated that cow dung reacts with Fe₂O₃ at high temperature to generate substances with high melting points. These substances improve the softening temperature of the surface material of the ore powder particles and increase the metal iron content, thus reducing the bonding of the highly reduced ore powder particles.

Figure 12 shows the energy spectrum of cow dung with and without cow dung. It is observed from Figure 12a that smaller contents of carbon and iron are obtained in the absence of cow dung addition; however, it is observed from Figure 12b that the contents of carbon and iron are significantly higher than those for the results presented in Figure 12a. This is because a large amount of free C is generated and deposited on the particle surface under the catalysis of metallic iron in H₂ atmosphere after the addition of cow dung, and then the free C reacts with metallic iron. In addition, at high temperature, organic matter in cow dung reacts with oxides in the fine iron ore particles such as Fe₂O₃ to form iron compounds with high melting points and adheres to the surface of the particles. Figure 13 shows the SEM image of the fine iron ore particles coated with cow dung after reduction. It is observed that colloidal high melting point substances are formed on the surface of coated particles, thus forming a certain isolation layer that reduces the likelihood of direct contact of reduced iron in the ore powder particles and effectively inhibits the loss of adhesion.

Based on these results, we infer the schematic diagram of agglomeration and inhibition of fluidized reduction of Newman ore particles, as shown in Figure 14. Figure 14a shows the original particles M and N of fine iron ore particles. After direct reduction by fluidization, the particles undergo a step-by-step reduction reaction: Fe₃O₄ in the particles is rapidly reduced to FeO, and the surface FeO is further reduced to metallic iron. Then, the metallic iron covering the particle surface acts as an adhesive at high temperature. With the increase in the metallization rate, the amount of iron on the surface of the particles increases gradually. When the generated viscous force exceeds the drag force of the fluidizing gas, the particles will agglomerate (as shown in Figure 14b,c) until the complete loss of flow.

Figure 14d shows particles X and Y of fine iron ore particles coated with a certain amount of cow dung. A certain gap exists between the cow dung crystals attached to the surface of the mineral powder particles by the powder method, and the cow dung crystals separated by drying after coating by the solution method cover the surface of the particles with many submicron pores. Therefore, the reducing gas can still contact the original particles through diffusion to form a reaction interface, and the reaction front gradually proceeds toward the center of the particles. The outer layer of fine iron ore particles is first reduced to FeO, and then will further continue the reduction to metallic iron and will accumulate continuously. When the metallization rate is the same as that in Figure 14c (as shown in Figure 14f, the adhesive force between the particles decreases due to the addition of the cow dung isolation layer, effectively prolonging the time for maintaining fluidization, so that the degree of viscosity reduction is determined by the density uniformity of the isolation layer. For greater initial coating amount, the separation effect is better and the viscosity of the particles is reduced significantly, so that the time of maintaining fluidization is relatively long. With the prolongation of the fluidization reduction time, carbon evolution reaction occurs in H₂ atmosphere under the catalysis of metallic iron, and a large amount of free C is generated and deposited on the particle surfaces. Then, C reacts with metallic iron to form Fe₃C with low viscosity. In addition, at high temperature, cow dung will react with oxides in fine iron ore particles to form high melting point compounds that are attached to the particle surface and inhibit the growth of iron whiskers (as shown in Figure 14g). Since the last two reactions in Figure 14a,b are carried out after the formation of metallic iron, the particles without coating oxide surface modification usually suffer flow loss before the latter two reactions occur when the metallic iron is formed. However, the surface viscosity of the modified particles is effectively reduced due to the isolation effect of the oxide, providing time for the last two reactions, thus increasing the extent of reduction. The dynamic force of particle viscosity prolongs the time of fluidization, so that the metallization rate is clearly improved.

5. Conclusions

In this work, to solve the problem of adhesion and loss of flow in the process of fluidized reduction ironmaking, the optimal experimental operating parameters were obtained by orthogonal experiments. The influence of coated cow dung on the reduction of fine iron ore particles and the mechanism of the inhibition of the fine iron ore particles by cow dung were studied. Using the orthogonal analysis method, the microscale morphology and mechanical properties of the fine iron ore particles in the process of fluidized reduction were explored, and the mechanisms of adhesion and loss of flow were elucidated. This provides reference data and theoretical basis for the industrialization of fluidized direct reduction process. The main conclusions are as follows.

• The addition of a coating component increases the metallization rate, reduces the sticking ratio, and improves the fluidization state.
• The optimal operating parameters for the addition of cow dung are as follows: reduction temperature of 923–1023 K, reducing gas velocity of 0.6 m/s, reduction time of 40–60 min, reduction pressure of 0.2 MPa, H₂ as the reducing gas, and cow dung content of 16%.
• At high temperature, cow dung decomposes a large amount of hydrogen that reacts with iron oxide, absorbs a high amount of energy and inhibits the growth of iron whiskers, thus inhibiting the loss of adhesion and improving the fluidization quality.
• At high temperature, cow dung reacts with the oxide in fine iron ore particles to form a high melting point substance, thus forming a certain isolation layer that reduces the likelihood of direct contact of the reduced iron between the ore powder particles and effectively inhibits the bond loss.