Removal of chlorine from electric arc furnace dust by static washing with water

Introduction

The processing of metallurgical waste that is suitable for recycling remains a pressing global challenge. One such waste stream is electric arc furnace (EAF) dust, which is generated during steelmaking and captured by gas-cleaning systems. The primary value of EAF dust lies in its content of iron and zinc. Iron concentrations typically range from 20 to 50 %, while zinc contents vary between 2 and 25 % [1 – 3]. It is estimated that more than 30 million tons of EAF dust are currently accumulated in waste dumps at metallurgical plants across the Russian Federation. At present, however, the majority of this material remains unprocessed¹ [4; 5]. The processing of EAF dust is hindered by its complex and heterogeneous morphology, as well as by the presence of contaminating impurities, including lead, fluorine, and chlorine [6 – 8]. During electrolytic zinc production, these impurities cause several technological problems, such as difficulties in removing deposited zinc [9], corrosion of lead-based anodes, and increased electric power consumption. In the case of pyrometallurgical processing, careful control of operating conditions is required, as chlorine can promote the formation of highly toxic organic compounds, including dioxins and furans [10 – 12]. Among the available processing routes, hydrometallurgical processing is one of the main approaches used for EAF dust treatment. Most studies in this area focus on zinc extraction; however, under hydrometallurgical conditions, almost all iron becomes unsuitable for further processing, limiting its subsequent utilization [13; 14].

A number of studies have reported methods for chloride removal from EAF dust, including water washing [15 – 17]. These methods are generally effective provided that poorly soluble compounds such as PbOHCl and Pb₂CO₃Cl₂ are absent. In [15], water leaching of chloride ions from EAF dust with an initial chlorine content of 2.14 % was investigated. The authors demonstrated that 99.9 % of chlorides could be removed at a natural pH of 12 and a temperature of 25 °C. As a result, the chlorine content decreased to below 0.02 – 0.04 %, which is close to the detection limit of X-ray fluorescence (XRF) analysis.

The experiments were performed using distilled water at a solid-to-liquid ratio of 1:10, with washing times ranging from 5 to 60 min. Changes in the mineral composition of the dust and its behavior during leaching were systematically analyzed. The highest dechlorination rate was observed within the first 5 min, while most chlorides were removed within 40 min. Overall, water washing reduced the total chloride content from 70,200 to 17,500 mg/kg, corresponding to a removal efficiency of 75 %. Temperature was found to have a pronounced effect on the process. At 80 °C, chlorides were removed more rapidly than at room temperature, and the overall removal efficiency increased to 88 %. Nevertheless, the dechlorination efficiency reported in [18] remained lower than that achieved in [15], where values above 99 % were obtained. This discrepancy was attributed to the possible formation of water-insoluble compounds.

The aim of the present study is to investigate the possibility of chlorine removal from EAF dust generated at a major metallurgical enterprise using static washing with water, as well as to assess the feasibility of reusing waste water for repeated washing.

Materials and equipment

Dust obtained from one of the country’s major metallurgical enterprises was used in this study. The average chemical composition of the initial dust is summarized in Table 1.

Based on X-ray phase analysis and scanning electron microscopy, the predominant chlorine-containing phases in the dust were identified as sodium chloride and potassium chloride (NaCl and KCl). Published data indicate that both compounds are readily soluble in water. The solubility of sodium chloride is 35.6 g/100 mL at 0 °C, 35.9 g/100 mL at 25 °C, and 39.1 g/100 mL at 100 °C, while potassium chloride exhibits solubilities of 28.1 g/100 mL at 0 °C, 34.0 g/100 mL at 25 °C, and 56.7 g/100 mL at 100 °C [19]. Prior to experimentation, the key process parameters requiring optimization were identified, namely washing time, temperature, and solid-to-liquid ratio. Analytical measurements were performed using an OPTIMA 2100 DV inductively coupled plasma optical emission spectrometer (ICP-OES), with an analytical accuracy of 5 – 7 % of the element mass; a JEOL JSM-6460LV scanning electron microscope; an I-160-MI laboratory ion meter; and a chloride-selective ion-selective electrode (ELIS-131Cl).

Experimental procedure

The concentrations of chlorides and other elements were determined in the resulting solutions. The dried solid residues were subsequently examined by scanning electron microscopy. To evaluate the effect of washing time, 50 g portions of the initial dust were placed in separate containers and mixed with 500 mL of water, corresponding to a solid-to-liquid ratio of 1:10. The suspensions were maintained at room temperature for predetermined time intervals. After washing, the liquid phase was decanted, and the solid residue was dried. To investigate the influence of the solid-to-liquid ratio, the samples were washed at room temperature for a fixed washing time of 1 h. The following solid-to-liquid ratios were examined: 1:1, 1:2, 1:5, 1:15, and 1:20. After washing, the solution was decanted and the solid residue was dried. To determine the effect of temperature, a suspension of the initial dust and water was heated in a drying oven to the required temperature. The washing was then carried out in separate flasks at a solid-to-liquid ratio of 1:10 and a washing time of 1 h. After washing, the solution was decanted and the solid residue was dried.

Results

The effects of washing time, temperature, and solid-to-liquid ratio, as well as the results of experiments on the reuse of waste water, are summarized in Tables 2 – 11.

Optimization of washing time

The experiments were performed at 20 °C and a solid-to-liquid ratio of 1:10. A single experiment was carried out for each selected washing time.

For scanning electron microscopy, five spectra were acquired for each sample, with three areas analyzed per spectrum. The average chlorine content in the dry residue, calculated from these measurements, is reported in Table 3.

The results demonstrated that the degree of chlorine transfer from the dust into the solution was essentially independent of washing time over the range investigated. Comparable chlorine removal was observed at both the shortest and longest washing durations. On this basis, the minimum washing time of 1 h was adopted for all subsequent experiments. In addition to chlorine, potassium, sodium, and calcium exhibited substantial transfer into the solution. The concurrent transfer of potassium and sodium indicates that chlorine in the dust is present predominantly in the form of NaCl and KCl. The pronounced transfer of calcium may, in turn, indicate the presence of calcium hydroxide (Ca(OH)₂) in the dust.

Effect of the solid-to-liquid ratio

The experiments were conducted at a temperature of 20 °C and a washing time of 1 h. One experiment was performed for each selected solid-to-liquid ratio. The results showed that the optimal solid-to-liquid ratios were 1:5 and 1:10, owing to ease of filtration and high reliability of the results. At a ratio of 1:1, the suspension was excessively dense, and dust losses reached up to 30 %. In addition, the large volume of material subjected to filtration led to filter rupture, requiring the use of three containers with two filters each. At a ratio of 1:2, the suspension viscosity decreased; however, dust losses remained high (up to 15 %), and filter rupture was still observed. In this case, two containers with one filter each were used. At a ratio of 1:5, the suspension was sufficiently fluid, and dust losses decreased to approximately 8 %. Filtration proceeded uniformly and did not require additional containers. At ratios of 1:15 and higher, no dust losses were observed; however, these conditions resulted in excessive water consumption, rendering such ratios impractical.

The average values obtained from three different measurement points are presented in Table 5. As follows from the experimental data, similar transfer behavior of the investigated elements was observed across the different solid-to-liquid ratios. Based on these considerations, all subsequent experiments were carried out at a solid-to-liquid ratio of 1:10, due to its operational convenience.

Static experiments on the effect of temperature

The static experiments were carried out at a washing time of 1 h and a solid-to-liquid ratio of 1:10. One experiment was performed at each selected temperature.

The results obtained at different temperatures showed that chlorine transfer was complete regardless of temperature, while the contents of the other elements remained essentially unchanged.

Experiments on reuse of waste water

To assess the feasibility of dechlorination of EAF dust using previously used washing water, the possibility of multiple reuse of the washing water was investigated. During these experiments, a white flocculent precipitate was observed on the surface of the used water and on the walls of the containers. The chemical composition of this precipitate was not determined at this stage, as it was not possible to collect a sufficient amount for analysis. Prior to repeated washing, the water was filtered to remove the precipitate. The washing was then conducted at a solid-to-liquid ratio of 1:10 and a washing time of 1 h, without heating; all experiments were performed at room temperature. For convenience, the repeated washing experiments were designated as “P2” and “P3”. Thus, different EAF dust samples were washed three times using the same water.

To assess reproducibility, several identical 50 g samples of EAF dust were repeatedly washed using waste water.

The results of the repeated washing experiments showed that chlorine continued to transfer into the water, although inconsistent trends were observed for some of the other elements. The water samples P2 (1), P2 (2), and P2 (3) were then combined, and this same water was used for a third washing of three new 50 g dust samples. The corresponding results are presented in Table 10.

Following the third washing with the same water, it can be concluded that chlorine continues to transfer into the reused water, even after two prior washing cycles.

Discussion

In all experiments involving static washing of EAF dust with water, a minor mass loss of approximately 3 – 5 g was observed. This loss can be attributed to the high dispersity of the dust and to material losses resulting from adhesion of fine particles to the walls of the glass vessels.

An identical degree of chlorine transfer from EAF dust into the solution was observed at both the minimum and maximum washing times investigated. Consequently, a washing time of 1 h was adopted for all subsequent experiments. In addition to chlorine, K⁺, Na⁺, and Ca²⁺ were found to transfer actively into the solution, as evidenced by the data presented in Tables 2, 3, and 5. The transfer of potassium and sodium indicates that chlorine in the dust is present predominantly in the form of NaCl and KCl. The pronounced transfer of calcium, in turn, may indicate the presence of calcium hydroxide, Ca(OH)₂, in the dust.

The experiments also showed that the optimal solid-to-liquid ratios fall within the range of 1:5 to 1:10, which provides both operational convenience and the highest reliability of the results.

The average values obtained from three different measurement points are presented in Tables 4 and 5. As follows from the experimental data, similar transfer behavior of the investigated elements was observed across the different solid-to-liquid ratios.

The experiments performed at different temperatures demonstrated that chlorine transfer occurs completely regardless of temperature, while the contents of the other elements remain essentially unchanged. At the same time, it should be noted that higher temperatures promote increased dissolution of heavy metals into the solution and lead to an acceleration of the reaction kinetics.

After repeated washing, it was established that chlorine continues to transfer into the water, and the degree of dechlorination of new EAF dust samples was approximately 86 %, while no zinc losses were detected. The results of the repeated washing experiments are presented in Tables 8 – 11. In the solution obtained after the first use of water for washing, the chloride ion concentration was 3.7 mg/L. After the second use of the same water (experiments P2), the average chloride ion concentration increased to 6.5 mg/L. Following the third use of the same water (experiments P3), the chloride ion concentration reached 7.2 mg/L. At the same time, the chlorine content in each EAF dust sample washed with the same water did not exceed 0.2 wt. %.

Conclusions

Water washing can therefore be considered a highly effective method for dechlorination of EAF dust, provided that poorly soluble compounds such as PbOHCl and Pb₂CO₃Cl₂ are absent. The static experiments made it possible to establish the following parameters for laboratory conditions: an optimal washing time of 1 h to transfer chlorine into the aqueous phase and a solid-to-liquid ratio in the range of 1:5 – 1:10. It was also established that, in the material under study, the dominant chlorine-containing phase is NaCl, as indicated by electron microscopy data (Tables 3 and 4) and by the absence of sodium in the dry residues. To a lesser extent, chlorine is present in the form of KCl, which is confirmed by the presence of potassium in the dry residue. The high transfer of calcium into solution suggests that calcium is partially present in the form of Ca(OH)₂, which is readily soluble in water. The experiments demonstrated the feasibility of multiple reuse of the same waste water for washing EAF dust to remove chlorine. Under the conditions investigated, complete saturation of the solution with chlorine was not achieved.

In all experiments, a stable transfer of chloride ions into the aqueous phase was observed, with the degree of chlorine transfer reaching up to 98 %. As a result of static water washing, the chlorine content in the initial dust decreased from approximately 2 to 0.2 wt. %.