Utilization of blast furnace sludge for the removal of zinc from steelmaking dusts using microwave heating

Utilization of blast furnace sludge for the removal of zinc from steelmaking dusts using microwave heating

Accepted Manuscript Utilization of blast furnace sludge for the removal of zinc from steelmaking dusts using microwave heating Mamdouh Omran, Timo Fab...

4MB Sizes 0 Downloads 5 Views

Accepted Manuscript Utilization of blast furnace sludge for the removal of zinc from steelmaking dusts using microwave heating Mamdouh Omran, Timo Fabritius PII: DOI: Reference:

S1383-5866(18)31171-7 https://doi.org/10.1016/j.seppur.2018.09.010 SEPPUR 14908

To appear in:

Separation and Purification Technology

Received Date: Revised Date: Accepted Date:

5 April 2018 4 September 2018 5 September 2018

Please cite this article as: M. Omran, T. Fabritius, Utilization of blast furnace sludge for the removal of zinc from steelmaking dusts using microwave heating, Separation and Purification Technology (2018), doi: https://doi.org/ 10.1016/j.seppur.2018.09.010

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Utilization of blast furnace sludge for the removal of zinc from steelmaking dusts using microwave heating

Mamdouh Omrana,b*, Timo Fabritiusa a

Process Metallurgy Research Group, Faculty of Technology, University of Oulu, Finland.

b

Mineral Processing and Agglomeration Lab, Central Metallurgical Research and Development Institute, Cairo, Egypt.

Abstract: Utilization of dusts generated by steelmaking industries prevents the disposal of wastes and enhances the use of secondary raw material. This work aimed to study the simultaneous recycling of steelmaking dust and blast furnace sludge (BFS) by using microwave heating. The possibility of utilizing BFS as a reducing agent to remove zinc from the dusts was studied. Three steelmaking dusts obtained from steel plants in Finland with varying zinc concentrations and zinc-containing phases were studied. Mixtures of the dusts and graphite, bulk BFS, >250 µm BFS, and >125–250 µm BFS reductants were heated for 20 min at the microwave power of 1100 W. The results indicated that the zinc removal efficiency depends on the amounts of iron oxide and calcium oxide in the mixtures. The highest zinc removal was observed for mixtures with high calcium oxide and low iron oxide contents. Zinc removal of 92.79–96.06% was achieved for mixtures with >125–250 µm BFS and >250 µm BFS reductants. Therefore, to achieve efficient zinc removal, coarser fractions (>125 µm) of BFS should be utilized. Owing to the excellent microwave absorption and high carbon content of the sludge, BFS can be applied effectively in the processing of steelmaking dusts.

Key words: steelmaking dust; blast furnace sludge; recycling; microwave heating.

Corresponding author: Mamdouh Omran Address: Process Metallurgy Research Group, Faculty of Technology, University of Oulu, Finland. P.O. Box: 4300. E-mail: [email protected]; [email protected]

1. Introduction Dusts and slugdes with different compositions are generated in iron and steelmaking processes in Finland. In Tornio steel plant, chromium converter (CRC) is used to treat liquid ferrochrome from ferrochrome melting shop, whereas an electric arc furnace (EAF) is used to melt the scrap [1, 2]. The CRC process generates approximately 4000 t a-1 of CRC dust [3]. In Ovako Imatra, steel is produced using steel scrap as the raw material in an EAF, generating approximately 3000 t a-1 of dust [3]. Dust is fine-grained material containing significant amounts of zinc and iron together with variable amounts of calcium, manganese, magnesium, silicon, and chromium [4]. Zinc represents the major constituent of the dust and its content varies between 7 and 40%, depending on the scrap used and the ratio of galvanized scrap utilized [5–7]. EAF dust is classified as an environmentally hazardous waste [8]. Moreover, secondary raw materials such as steelmaking dusts are considered to be a means of increasing iron and zinc supplies [9- 12]. Iron and zinc are important base metals necessary in many industrial applications [11, 12]. Blast furnace sludge (BFS) is generated during the purification of flue gas leaving the blast furnaces used in pig iron production [13-16]. BFS is a mixture of oxides whose major components are iron oxides and coke fines [15, 16]. Owing to the high C and Fe contents of sludge, it can be reused [17]. SSAB Raahe produces liquid steel using the blast furnace (BF)–basic oxygen furnace (BOF) production process. The amount of BF sludge and BOF sludge generated at the site are approximately 35000 and 40000 t a-1, respectively [18-19].

The direct recycling of dust back to steel production is hindered owing to the presence of zinc, which causes operational difficulties in the steelmaking processes [3]. For recycling wastes in the production of iron and steel, several pyrometallurgical and hydrometallurgical processes or a combination of both have been developed [5, 16, 20-26]. Hydrometallurgical leaching of zinc has been carried out using both acidic and alkaline solutions, such as sulfuric acid [22], and hydrochloric acid. The main difficulty of hydrometallurgical leaching is that, the majority of zinc is found as zinc ferrite, which is quite stable and insoluble in most acidic and alkaline solutions, and the unwanted dissolution of iron [5, 16, 20]. The Waelz process is the leading pyrometallurgical method for the recovery of zinc. This process is based on a carbothermic reduction of ZnO and ZnFe2O4 and volatilization of metallic zinc [27]. Although the pyrometallurgical process can yield a high zinc recovery, it still have some economic problems due to high-energy requirements [20, 27]. The objective of this work is to study the simultaneous recycling of steelmaking dust (CRC and EAF) and BFS by using microwave heating. Owing to the high economic value of Zn and Fe, this study aimed to separate Zn as a valuable element and reuse the remaining material in steelmaking processes. The present work proposes the possibility of utilizing BFS as a reducing agent to remove zinc from the steelmaking dusts, owing to the high carbon content of the sludge. Microwave irradiation is applied as a heating source; the main advantages of microwave heating are volumetric, selective, and fast heating [27-29]. The advantages of the proposed method are utilizing BFS as a reducing agent in the recycling of dusts and saving the cost of reducing agent, and simultaneously minimizing the environmental impact of BFS and saving landfill space.

2. Materials & Experimental 2.1. Materials Three steelmaking dusts and a BFS sample were used in this study: (1) Ferrochrome converter (CRC) and EAF stainless steel (EAFSS) dusts were obtained from Outokumpu Tornio stainless steel plant, Finland. (2) EAF carbon steel (EAFCS) dust was obtained from Ovako Imatra, Finland. (3) A BFS sample was obtained from SSAB Europe Oy, Raahe, Finland. Each dust is a representative sample of the average dust composition. The sludge sample was obtained from the discharge of a centrifuge. Tables 1 and 2 list the properties of CRC, EAFSS, and EAFCS samples. In this study, BFS was used as a reducing agent. Table 3 lists the carbon contents of different size fractions of BFS. The carbon content in bulk BFS was 24.9%. From the analysis, the carbon content increased as the particle size of the BFS increased. In the fine fraction (<32 μm), the carbon content was 13.3%, whereas in the coarse fraction (>250 μm), the carbon content was 62.2%. This indicates that coke was concentrated in the coarser fractions. Therefore, to investigate the effect of carbon content and particle size of BFS on the rate of zinc reduction, bulk BFS sample, >125–250 μm BFS, and >250 μm BFS fractions were used as reductants. Synthetic graphite (>99% purity, Alfa Aesar) was also used as a reducing agent. The percentage of reductants in the mixtures was calculated according to the stoichiometric amount of carbon required to reduce all the zinc oxide (as both zincite and zinc ferrite) in the sample. Therefore, the amounts of reductants added varied depending on the carbon contents of the reductants and the zinc contents of the dusts. The amount of carbon added was twice the

stoichiometric amount required to reduce all the zinc oxide (as zincite and zinc ferrite) to elemental zinc.

2.2. Microwave experimental set-up The diagram of the microwave experimental setup is shown in Fig. 1. The power supply of the microwave furnace was a magnetron with the frequency of 2.45 GHz and power of 1.1 kW, and it was cooled via water circulation. The temperature was measured using an IR probe. The IR probe provides feedback information to the control panel, which controls the power to the magnetron, thus controlling the temperature of the sample during the microwave heating process. Nitrogen was pumped into the microwave cavity to release air before placing the sample in the oven. Nitrogen was purged at a rate of 0.2 L/min to provide inert ambience until the end of the experiment. The vapors from the crucible were pulled by a pump and trapped in a collector. The samples were placed in a microwave transparent alumina crucible. At the end of the experiment, the residue remaining in the crucible was cooled to room temperature in the microwave oven. Subsequently, the residue was sent for chemical analysis to determine the chemical composition and rate of zinc removal. The removal of zinc from steelmaking dusts is based on a carbothermic reduction of ZnO/ZnFe 2O4 and volatilization of metallic zinc from the dust. The rate of zinc removal refers to the percentage of zinc removed (evaporated) from the feed. Zinc removal (R) was calculated according to the following equation [30]:

R%=

,

(1)

where C0 is the initial Zn concentration and C is the Zn concentration in the solid residue.

2.3. Characterization methods The mineralogical composition was measured using X-ray diffraction (XRD); Rigaku SmartLab, 9 kW). The measurements were obtained with cobalt tube in the 2θ range of 4 to 90°. The chemical composition of the raw samples were determined using Bruker AXS S4 Pioneer X-ray fluorescence (XRF) spectrometer. The zinc contents in the residues after the experiments were measured using atomic absorption spectroscopy (ASS). Perkin Elmer A Analyst 400 flame atomic absorption spectrometer +S10 autosampler was used. Zn was inserted into HNO 3 acid dilution 100x and autoclave at 121 °C for around 20 minutes. Single element lumina hollow cathode lamp (HCL) was used for the detection of elemental Zinc (Zn). The carbon contents of the raw materials and the residues after microwave experiments were determined using a LECO Carbon Analyzer. The microstructure and microanalyses of the raw samples and residues were investigated using a Zeiss ULTRA plus field-emission scanning electron microscope (FE-SEM), attached to an energydispersive X-ray spectroscopy (EDS) unit for chemical analysis. The particle diameters were determined via a laser diffraction and scattering method using a Beckman Coulter LS 13 320. The thermal behavior of steelmaking dusts was studied using differential scanning calorimetrythermogravimetry (DSC-TG) and mass spectrometry (MS). The tests were performed using a Netzsch STA409 PC Luxx under N2 atmosphere. The dielectric constant (ɛ') and dielectric loss factor (ɛ") of the materials were measured using an open-ended coaxial cavity resonator operating at 4.5 GHz. The details of the procedure are described elsewhere [31]. 3. Results 3.1. Material characterization

The chemical composition of the steelmaking dusts are listed in Table 1. The major elements of the dusts were Zn, Fe, Cr, Ca, Si, Mg, and Mn. The chemical analyses of the dusts revealed a significant variation in the elemental composition. The contents of zinc in CRC, EAFSS, and EAFCS dusts were 10.83, 19.84, and 35.76 wt.%, respectively, whereas the contents of chromium were 20.88, 3.19, and 0.47 wt.%, respectively (Table 1). In addition to iron, zinc, and chromium, the dusts were characterized by relatively high contents of calcium oxide and magnesium oxide, owing to the dolomitic lime added to the steelmaking furnace [32]. The CaO contents in CRC, EAFSS, and EAFCS dusts were 14.27, 11.91, and 5.93 wt.%, respectively (Table 1). The identified phases in CRC, EAFSS, and EAFCS dusts are shown in Fig. 2. In CRC dust, chromite (FeCr2O4) represented the main phase with a spinel structure, and zincite (ZnO) was the zinc-bearing phase. EAFSS and EAFCS dusts consisted mainly of franklinite (ZnFe2O4) and zincite (ZnO). In addition to the main phases, portlandite Ca(OH) 2, lime (CaO), and periclase (MgO) were identified. Fig. 3A shows the SEM image of CRC dust, wherein chromite exists as an aggregate of irregular particles. Zincite appeared as a monocrystalline sphere, whereas periclase (MgO) and portlandite Ca(OH)2 appeared as fine grains. EAFSS dust was characterized by encapsulation particles (Fig. 3B). Franklinite phases were enclosed in calcium-iron-silicate glass spheres, and encapsulation particles were often surrounded by fine-grained particles and spheres (Fig. 3B). The EAFCS dust was dominated by spherical form, which indicates the formation by ejection from the liquid metal [3]. Franklinite and zincite spheres were identified in the EAFCS dust (Fig. 3C). The particle size distribution for dusts (CRC, EAFSS, and EAFCS) showed that the dust consisted of very fine aggregates of particles (Table 1). The median (d 50) particle size for CRC, EAFSS, and EAFCS dusts were 3.15, 2.37, and 1.63 μm, respectively (Table 1). The chemical composition of BFS demonstrated that Fe and C were the dominant elements in BFS. The contents of Fe and C were 38.95 and 24.90 wt.%, respectively. The BFS also contained Si, Ca, and minor amounts of Zn (Table 2). The main crystalline phases of BFS were hematite (Fe2O3),

calcite (CaCO3), and quartz (SiO2) (Fig. 4). The SEM image shows that BFS was composed mainly of hematite and carbon, in addition to fractions of silicate and carbonate materials (Fig. 3D). Hematite was present as aggregates of particles of irregular shape (< 20 μm). Coke was present as elongated crystals of larger particle size (>125 μm). The particle size distribution of BFS showed that BFS had different size fraction ranges owing to the grade efficiency of the first separation stage [19]. The average particle size (d50) of BFS was 18.42 μm. 3.2. Thermal analysis The behavior of CRC, EAFSS, and EAFCS dusts as a function of temperature is shown in Figs. 5 and 6. The DSC-TG curves showed that CRC, EAFSS, and EAFCS dusts exhibited similar thermal behavior. The first peak was observed at 190 °C, which is the reaction corresponding to the evaporation of physically adsorbed water (Fig. 5). There were two main reactions at 415–418 °C and 622–645 °C, related to the dehydroxylation of calcium hydroxide and the decomposition of calcium carbonate, respectively, according to the following reactions [3, 33, 34]. Ca(OH)2 →CaO + H2O

(2)

CaCO3 → CaO + CO2

(3)

Both Ca(OH)2 and CaCO3 existed originally as free lime in the dusts, which, in turn, may have hydrated and carbonated on exposure to ambient moisture and CO 2 [35, 3]. There were apparent mass losses of the dusts at the temperature above 900 °C (Fig. 6), although there was no corresponding gas release detected by MS. This phenomenon is related to the volatilization of Zn and Pb [22, 33]. The carbon contained in the dust could react with zinc, which evaporated into zinc vapor [3]. TG curves show that the mass losses of CRC, EAFSS, and EAFCS dusts were 0.5, 4.8, and 8.3%, respectively, depending on the carbon contents of the samples (Fig. 6). The behavior of the dusts with a reducing agent is shown in Figs. 7–9. Carbon is assumed to be converted into carbon monoxide (CO), according to equation (4).

The behavior of CRC under reducing conditions is shown in Fig. 7. There were two reactions at the temperatures of approximately 955–1100 °C and 1222–1297 °C, associated with two mass loss stages. The reduction of ZnO began at the temperature of 955 °C. When the temperature reached approximately 1100 °C, Zn was completely reduced and vaporized [36], (Fig. 7). The TG curve demonstrates the mass loss owing to the evaporation of zinc and graphite gasification (which causes the formation of the CO/CO2 atmosphere). Zinc oxide reacted with carbon as follows [31, 37]: ZnO (s) + C (s) = Zn (g) + CO (g)

(4)

This reaction is a combination of the following reactions: ZnO (s) + CO (g) = Zn (g) + CO2 (g)

(5)

CO2 (g) + C (s) = 2CO (g)

(6)

The reaction at the high temperatures of 1222–1297 °C is attributed to the formation of calcium ferrite. At high temperature, the calcium carbonate contained in the sample began to decompose and the free calcium oxide combined with the free iron oxide to form calcium ferrites [36]. The behavior of EAFSS and EAFCS with a reducing agent is shown in Figs. 8 and 9. The DSC curves show a series of peaks at 807, 865, 925, and 951 °C and at 895, 943, and 1000 °C, for EAFCS and EAFSS, respectively. These reactions were due to the decomposition of zinc ferrite and the reduction of zinc oxide to zinc vapor. The decomposition of zinc ferrite is believed to occur as follows [36]: C + ZnFe2O4 = ZnO + 2FeO + CO(g) and

(7)

CO(g) + ZnFe2O4= ZnO + 2FeO + CO2(g)

(8)

First, ZnFe2O4 decomposes to ZnO and FeO at 807–895 °C; subsequently, the reduction of ZnO and FeO occurs. Above 925–1000 °C, zinc oxide is reduced to elemental zinc vapor, according to equation (4). CO/CO2 atmosphere is produced by the reduction processes and Boudouard mechanism (reaction (6)). Gas/solid reactions between CO and ZnO/FeO occur, and the zinc oxide

and iron oxide are reduced to zinc vapor and iron metal, respectively. The mass loss in the temperature from 807 to 1000 °C is related to the evaporation of zinc and carbon gasification (Figs. 8 and 9). In the case of EAFSS dust, at the high temperature of 1235–1391 °C, a deep valley was observed in the DSC (Fig. 9). This reaction was due to the formation of calcium ferrite. 3.3. Heating behavior with microwave irradiation

ɛ'

ɛ"

ɛ'

ɛ" ɛ"

ɛ'

δ ɛ'

ɛ"

δ







°

3.4. Microwave treatment (Zn removal) 3.4.1. CRC The main results of microwave heating of the mixtures of CRC at the power of 1100 W are presented in Table 4. When a mixture of CRC dust and graphite was heated for 20 min, the zinc content was reduced from 10.29 to 0.34 wt.% (96.86% Zn removal). The zinc contents were reduced from 10.29 to 1.25 wt.% (88.45% Zn removal) and 0.78 wt.% (92.79% Zn removal), after heating the mixtures of CRC:>125–250 µm BFS and CRC:>250 µm BFS, respectively, for 20 min.

Lower zinc removal yield (72.57%) was obtained after heating the mixture of (CRC:BFS) and the zinc content was reduced from 10.29 to 2.97 wt.%. The chemical compositions of residues demonstrated that the residues of the mixtures with bulk BFS reductant were rich in Fe content (29 wt.%) owing to the high Fe content in raw BFS, whereas the residues of other mixtures contained significant chromium. The overall lead contents in all the residues was reduced from 0.11 to 0.025–0.029 wt.%. The chloride level was reduced from 0.29 to 0.05 wt.%. The alkaline elements Na and K were significantly reduced from 1.25 and 0.61 wt.% to 0.48 and 0.20 wt.%, respectively. A comparison between the XRD patterns of raw CRC dust and the residues showed that ZnO peaks disappeared completely after microwave heating of the mixtures of CRC:>125–250 µm BFS and CRC:>250 µm BFS for 20 min (Fig. 11). In the case of the CRC:BFS mixture, ZnO peaks were detected in the residue, and the amounts of wustite and metallic iron increased owing to the reduction of hematite present in the BFS (Fig. 11 D). The main phases in the residues were chromite, iron phases, calcium magnesium silicate, and calcium ferrite. The SEM images of the residues obtained after heating the mixtures of CRC:graphite and CRC:>250 µm BFS are shown in Fig. 12. The residues consisted mainly of chromite spheroids surrounded by Ca-Mg silicates phases (Figs. 12B&D). The morphology of the treated CRC dust was different from that of untreated CRC. The chromite particles changed from fragments to spheroidal particles owing to chromite sintering during microwave heating. No zinc-bearing phases were detected in the residues. The SEM image of the CRC:BFS residue after microwave heating shows that calcium ferrite was formed in the residue Fig. 13. Calcium oxide combined with free iron oxide in the mixture to form calcium ferrite (Fig. 13 A&B), as indicated by DSC and XRD results. Iron oxide present in the BFS was reduced to iron metal (Fig. 13 C&D). Fig. 14 shows the EDX mapping of the CRC:>250 µm BFS residue after microwave heating. Cr was present with Fe, which indicates the presence of chromite phase. Mn and Mg appeared to be forming phases with Ca

and Si elements, which was also indicated by XRD results. The EDX distribution map of Zn indicated low concentration of Zn in the residue.

3.4.2. EAFSS The main results of microwave heating of the mixtures of EAFSS for 20 min at the power of 1100 W are presented in Table 5. With the mixture of EAFSS:Graphite, the zinc content was reduced from 19.84 to 1.22 wt.% (93.85% Zn removal). In the case of the mixture of EAFSS:>250 µm BFS, zinc was reduced from 19.84 to 0.78 wt.% (96.06% Zn removal). Zinc removal rate of 93.64% was obtained with the mixture of EAFSS:>125–250 µm BFS. Under the same conditions, with the mixture of EAFSS:BFS, zinc in the residue was reduced from 19.84 to 3.12 wt.% (84.27% Zn removal). The residues of all the mixtures were rich in Fe contents (about 35–39 wt.%), in addition to calcium, magnesium, and silica. The lead contents in all the residues were significantly reduced from 0.60 to 0.10 wt.%. The chloride level was reduced from 1.25 to 0.07 wt.%. The alkaline elements Na and K were reduced from 3.67 and 1.23 wt.% to 0.62 and 0.20 wt.%, respectively. The XRD of the solid residues of the EAFSS mixtures after microwave heating are shown in Fig. 15. The diffraction peaks of ZnO and ZnFe2O4 almost disappeared in the residues of EAFSS:>250 µm BFS and EAFSS:Graphite mixtures. Fig. 15 B shows that wustite resulted from the decomposition of franklinite. The main phases in the residues were iron, calcium silicate, and calcium ferrite. Fig. 16 shows the SEM images of the EAFSS:>250 µm BFS residue after microwave heating. The residue appeared as an agglomeration of spherical and elongated grains. The EDX analysis indicated that these agglomerations were composed mainly of calcium silicate and calcium ferrite (Fig. 16 D), in addition to chromite and iron spheres (Fig. 16 C). Metallic iron was formed in the

residue of the EAFSS:BFS mixture, owing to the reduction of iron oxide present in BFS (Fig. 17). Fig. 17A shows that the iron particles were partly sintered. Approximately 2–3 wt.% of Ni was observed in most metallic iron particles. Fig. 17C shows that calcium ferrite and calcium silicate grains were observed in the residue. The EDX mapping of the EAFSS:>250 µm BFS residue is shown in Fig. 18. Ca appeared in high amounts when Si and O also appeared. Fe was distributed in regions where Cr and Ca were present. This suggests the presence of calcium silicate, calcium ferrite, and chromite phases. 3.4.3. EAFCS The main results of microwave heating of the mixtures of EAFCS are presented in Table 6. The residue of the EAFCS:Graphite mixture show that the zinc was reduced from 35.76 to 5.29 wt.% (85.20% Zn removal) after microwave heating for 20 min at 1100 W. Under the same conditions, with a mixture of EAFCS:BFS, the zinc in the residue was reduced from 35.76 to 5.98 wt.% (83.27% Zn removal). In the case of the mixtures of EAFCS:>125–250 µm BFS and EAFCS:>250 µm BFS, the zinc contents in the residues were reduced from 35.76 to 4.53 wt.% (87.33% Zn removal) and 3.42 wt.% (90.43% Zn removal), respectively. Chemical analysis showed that the residues in all the mixtures were rich in Fe (approximately 37– 42 wt.%) and calcium (6–12 wt.%) contents. The lead content in all the residues was reduced from 2.15 to 0.16 wt.%. The chloride level was reduced from 1.69 to 0.05 wt.%. The alkaline elements Na and K were reduced from 6.36 and 2.66 wt.% to 2.01 and 1.21 wt.%, respectively. XRD indicated that the diffraction peaks of ZnO disappeared after microwave heating in all the mixtures (Fig. 19). In the case of the mixture of EAFCS:>125–250 µm BFS, the peaks of ZnO completely disappeared and the peaks of ZnFe2O4 were weakened. In the residue of the EAFCS:BFS mixture, the peaks of ZnFe2O4 were still observed. The phases present in the residues were iron metal, wustite, calcium ferrite, and calcium silicate.

Fig. 20 shows the SEM image of the EAFCS:>250 µm BFS residue after microwave heating. The residue was composed mainly of the agglomeration of calcium ferrite, calcium silicate, and iron. EDX analysis showed that calcium ferrite agglomerations were rich in Mn (Fig. 20 C). The residue obtained after heating the EAFCS:BFS mixture was rich in iron and calcium ferrite grains (Fig. 21 B&C). Fig. 21 A shows that the particles were partly sintered. Fig. 22 shows the EDX mapping for the EAFCS:>250 µm BFS residue. The distribution of Ca, Si, Fe, and O suggests the existence of calcium silicate and calcium ferrite phases. Moreover, areas with high Fe suggest the presence of iron metal.

4. Discussion

zinc removal from steelmaking dusts was evaluated for different mixtures (Tables 4–6). The results indicated the effectiveness of zinc removal was influenced by the amounts of iron oxide and calcium oxide in the mixtures [36, 44]. The amounts of iron oxide phases in the mixtures were observed to have a strong influence on the zinc removal rate. The results revealed that the zinc removal in the case of mixtures with >125–250

µm BFS and >250 µm BFS reductants was higher than that in the case of mixtures with bulk BFS reductant (Tables 4–6). The low Zn removal efficiency in the mixtures with bulk BFS reductant was due to its high Fe content (38.95 wt.%) compared to (<10 wt.%) in the case of >250 µm BFS reductant. An increase in the amount of iron oxide, e.g., hematite and magnetite, would result in an increased consumption of the reducing agent. Therefore, the amount of reducing agent was insufficient to reduce all the zincite or franklinite [44]. The amounts of calcium oxide in the mixtures can influence the efficiency of zinc removal. The highest zinc removal was observed in mixtures with high CaO contents. The free calcium oxide in the mixtures can combine with the free iron oxide to form calcium ferrites (Figs. 13 & 17). The increase in the amount of calcium oxide resulted in a significant increase in the amount of calcium ferrite, and also a corresponding decrease in the amount of iron oxide [36, 44]. Therefore, the amount of iron oxides (hematite and/or magnetite) reduced by the carbon decreased and thus, more reducing agent was available for the ZnO/ZnFe2 O4 reduction and this results in an improved zinc removal ratio. The higher efficiencies of the zinc removal in the case of CRC and EAFSS dusts compared to the EAFCS dust were due to the high CaO contents in the raw CRC and EAFSS dusts (Tables 4–6). The DSC curves showed that calcium ferrites were formed in the CRC and EAFSS dusts (Figs. 7 and 9). The highest zinc removal in the range of 92.79–96.06% was achieved for the mixtures of CRC and EAFSS with >125–250 µm BFS and >250 µm BFS reductants. The particle size, carbon content, and type of the reductant can also affect the Zn removal reaction [37, 45]. The higher efficiency of Zn removal and heating in the cases of >125–250 µm BFS and >250 µm BFS reductants was due to the high C contents of 47.5 and 62.2%, respectively, compared to 24.29% in the bulk BFS reductant. The larger particle size of >125–250 µm BFS and >250 µm BFS reductants compared to d50=18.41 in the bulk BFS reductant improved the heating efficiency. Microwave heating was more effective for coarser fractions compared to finer fractions [46]. The type of carbon also influenced the microwave heating efficiency. In the BFS, the reductant is coke;

the carbon in coke is in microcrystalline form and the electrons of carbon are freer to move than those of pure crystalline graphite (Fig. 10). Therefore, coke exhibits the best coupling with microwave irradiation [47]. The residual carbon content for the tests is shown in Table 7. The residual carbon for the mixtures with bulk BFS reductant are insignificant owing to the high Fe content in bulk BFS, which resulted in increased consumption of the reducing agent. The residual carbon for the mixtures with other reductants are not negligible. The carbon remaining in the sample can be recycled with the residue back to Steel furnace. Moreover, the results demonstrated a significant reduction in the contents of lead and alkaline elements after microwave heating of all the mixtures (Tables 4–6). The chemical compositions of the residues indicated that lead was almost vaporized and removed from the mixtures after microwave treatment. Owing to the lower thermodynamic stability of lead oxide in comparison to zinc oxide, high lead recoveries could be achieved at lower temperatures and reactant ratios [9, 36]. In summary, the efficiencies of zinc removal in the case of mixtures with >125–250 µm BFS and >250 µm BFS reductants were due to low iron oxide and high coke and calcium oxide contents in these fractions, as indicated by the investigation of different size fractions of BFS. Therefore, in order to achieve efficient zinc removal, coarser fractions (>125 µm) of BFS should be used.

5. Conclusions The present work aimed to study the possibility of utilizing BFS to remove zinc from steelmaking dusts by using microwave heating.

The results indicated that the addition of BFS to the steelmaking dusts significantly improved the microwave heating efficiencies of the dusts. The results indicated that zinc removal efficiency was influenced by the amounts of iron oxide and calcium oxide present in the mixtures. The highest zinc removal percentages in the case of mixtures with >125–250 µm BFS and >250 µm BFS reductants were due to low iron oxide and high coke and calcium oxide contents; therefore, more reducing agent was available to reduce all the zinc oxide. In order to achieve an efficient zinc reduction rate, coarser fractions (>125 µm) of BFS should be used.

The results indicated that BFS can be applied effectively in the removal of zinc from steelmaking dusts. The proposed method has economic and environmental advantages such as utilizing BFS as a reductant and saving the cost of reducing agent, and reducing the environmental impact of sludge landfilling.

Acknowledgments The authors gratefully acknowledge the financial support of Tekes (6905/31/2016). The authors acknowledge Outokumpu Tornio plant, Ovako Imatra, and SSAB Raahe for supplying industrial samples.

References [1] H.T. Makkonen, A. Kekki, E-P. Heikkinen, T. Fabritius, J. Aromaa, O. Forsen, Characterization and Sulfuric Acid Leaching of Ferrochrome Converter (CRC) Dust, steel research int. 86 (2015) 19. [2] E-P. Heikkinen, T. Fabritius, Modelling of the Refining Processes in the Production of Ferrochrome and Stainless Steel. In: Nusheh, M., Ahuett, G.H., Arrambide, A. (eds.). Recent Researches in Metallurgical Engineering - From Extraction to Forming, In Tech. 186 (2012) 65-88. [3] M. Omran, T. Fabritius, Effect of steelmaking dust characteristics on suitable recycling process determining: Ferrochrome converter (CRC) and electric arc furnace (EAF) dusts, Powder Technol. 308 (2017) 47–60. [4] T. Sofilić, A. Rastovčan-Mioč, Š. Cerjan-Stefanović, V. Novosel-Radović, M. Jenko, Characterization of steel mill electric-arc furnace dust, J. Hazard. Mater. 109 (2004) 59-70. [5] G. Orhan, Leaching and cementation of heavy metals from electric arc furnace dust in alkaline medium, Hydrometallurgy 78 (2005) 236-245. [6] G. Salihoglu, V. Pinarli, Steel foundry electric arc furnace dust management: stabilization by using lime and Portland cement, J. Hazard. Mater. 153 (2008) 1110-1116. [7] C.F. Pereira, Y.L. Galiano, M.A. Rodriguez-Piñero, J.V. Parapar, Long and short term performance of a stabilized/solidified electric arc furnace dust, J. Hazard. Mater. 148 (2007) 701707.

[8] C. Sikalidis, M. Mitraka, R. Tsitouridou, Immobilization of electric arc furnace dust toxic elements within the matrix of concrete based products, Global NESTJ. 12 (4) (2010) 368–373. [9] M. Grabda, S. Oleszek, E. Shibata, T. Nakamura, Study on simultaneous recycling of EAF dust and plastic waste containing TBBPA, J. Hazard. Mater. 278 (2014) 25–33 [10] T. Havlik, B.V. Souza, A.M. Bernardes, I.A.H. Schneider, A. Miskufova, Hydrometallurgical processing of carbon steel EAF dust, J. Hazard. Mater. B135 (2006) 311−318. [11] M.H. Morcali, O. Yucel, A. Aydin, B. Derin, Carbothermic reduction of electric arc furnace dust and calcination of Waelz oxide by semiPilot scale rotary furnace, J. Min. Metall. Sect. BMetall. 48 (2) B (2012) 173 – 184. [12] C.A. Pickles, Thermodynamic analysis of the selective chlorination of electric arc furnace dust, J. Hazard. Mater. 166 (2009) 1030–1042. [13] J.G. Machado, F.A. Brehm, C.A. Moraes, C.A. Santos, A.C. Vilela, J.B. Cunha, Chemical, Physical, Structural and Morphological characterization of the electric arc furnace dust, J. Hazard. Mater. 136 (3) (2006) 953 - 960. [14] T. Mansfeldt, R. Dohrmann, Chemical and mineralogical characterization of blast-furnace sludge from an abandoned landfill, Environ. Sci. Technol. 38 (22) (2004) 5977-5984. [15] P. Van Herck, C. Vandecasteele, R. Swennen, R. Mortier, Zinc and lead removal from blast furnace sludge with a hydrometallurgical process, Environ. Sci. Technol. 34 (17) (2000) 3802−3808. [16] Z.H. Trung, F. Kukurugya, Z. Takacova, D. Orac, M. Laubertova, A. Miskufova, T. Havlik, Acidic leaching both of zinc and iron from basic oxygen furnace sludge, J. Hazard. Mater. 192 (3) (2011) 1100-1107. [17] B. Das, S. Prakash, P.S.R. Reddy, V.N. Misra, An overview of utilization of slag and sludge from steel industries, Resource Conservation & Recycling, 50 (1) (2007) 40−57. [18] C. Wang, R. Jennes, O. Mattila, T. Paananen, J. Lilja, M. Larsson, Investigation of applying OxyCup® process for an integrated steel plant from a Nordic country, Düsseldorf, 15 – 19 June 2015. [19] M. Omran, T. Fabritius, T. Paananen, Effect of Blast Furnace Sludge (BFS) Characteristics on Suitable Recycling Process Determining, JMMCE, 5 (2017) 185-197. [20] M.K. Jha, V. Kumar, R.J. Singh, Review of hydrometallurgical recovery of zinc from industrial wastes, Resources, Conservation & Recycling, 33 (2001) 1-22. [21] N. Leclerc, E. Meux, J.-M. Lecuire, Hydrometallurgical extraction of zinc fromzinc ferrites, Hydrometallurgy, 70 (2003) 175-183. [22] P. Oustadakis, P.E. Tsakiridis, A. Katsiapi, S. Agatzini-Leonardou, Hydrometallurgical process for zinc recovery from electric arc furnace dust (EAFD) Part I: Characterization and leaching by diluted sulphuric acid, J. Hazard. Mater. 179 (2010) 1-7. [23] H. Matsuura, T. Hamano, F. Tsukihashi, Removal of Zn and Pb fromFe2O3-ZnFe2O4-ZnOPbO mixtures by selective chlorination and evaporation reactions, ISIJ Int. 46 (8) (2006) 11131119. [24] H. Matsuura, F. Tsukihashi, Chlorination of ZnO with Ar-Cl2-O2 gas and the effect of oxychloride formation, Metall. Mater. Trans. B 37 (2006) 413-420.

[25] Y. Masuda, T. Uda, O. Terakado, M. Hirasawa, Pyrolysis study of poly(vinylchloride)-metal oxide mixtures: quantitative product analysis and the chlorine fixing ability of metal oxides, J. Anal. Appl. Pyrolysis, 77 (2006) 159-168. [26] L.G. Seung, S.Y. Jun, Recycling EAF dust by heat treatment with PVC, Mater. Eng. 20 (2007) 739–746. [27] M. Al-harahsheh, S. Kingman, L. Al-Makhadmah, I.E. Hamilton, Microwave treatment of electric arc furnace dust with PVC: Dielectric characterization and pyrolysis-leaching. J. Hazard. Mater. 274 (2014) 87-97. [28] K.E. Haque, Microwave energy for mineral treatment processes—a brief review, Int. J. Miner. Process. 57 (1) (1999) 1-24. [29] M. Omran, T. Fabritius, R. Mattila, Thermally assisted liberation of high phosphorus oolitic iron ore: A comparison between microwave and conventional furnaces, Powder Technol. 269 (2015) 7-14. [30] P.P. Falciglia, D. Malarbì, R. Maddalena, V. Greco, F.G.A. Vagliasindi, Remediation of Hgcontaminated marine sediments by simultaneous application of enhancing agents and microwave heating (MWH), Chemical Engineering Journal, 321 (2017) 1-10. [31] M. Omran, T. Fabritius, E-P. Heikkinen, G. Chen, Dielectric properties and carbothermic reduction of zinc oxide and zinc ferrite by microwave heating. R. Soc. open sci. 4 (2017) 170710. [32] D. Mombelli, C. Di Cecca, C. Mapelli, S. Barella, E. Bondi, Experimental analysis on the use of BF-sludge for the reduction of BOF-powders to direct reduced iron (DRI) production, Process Saf. Environ. Protect. 102 (2016) 410-420. [33] S.A. Mikhail, A-M. Turcotte, Thermal reduction of steel-making secondary materials I. Basicoxygen-furnace dust, Thermochimica Acta 311 (1998) 113-119. [34] S.A. Mikhail, A-M. Turcotte, J. Aota, Thermoanalytical study of EAF dust and its vitrification product, Thermochim. Acta. 287 (1996) 71-79. [35] C. Navarro, M. Díaz, M. Villa-García, Physico-Chemical Characterization of Steel Slag. Study of its Behavior under Simulated Environmental Conditions, Environ. Sci. Technol. 44 (2010) 53835388. [36] C.A. Pickles, Thermodynamic analysis of the selective carbothermic reduction of electric arc furnace dust, J. Hazard. Mater. 150 (2008) 265-278. [37] B. Kim, J. Yoo, J. Park, J. Lee, A Kinetic Study of the Carbothermic Reduction of Zinc Oxide with Various Additives, Mater. Trans. 47(9) (2006) 2421-2426. [38] M. Lovas, M. Kovacõva, G. Dimitrakis, S. čuvanova, I. Znamenaćkova, Š. Jakabsky´, Modeling of microwave heating of andesite and minerals, Int. J. Heat and Mass Transfer. 53 (2010) 3387-3393. [39] M. Al-Harahsheh, S. Kingman, A. Saeid, J. Robinson, G. Dimitrakis, H. Alnawafleh, Dielectric properties of Jordanian oil shales, Fuel Processing Technol. 90 (2009) 1259-1264. [40] M. Omran, T. Fabritius, A. Elmahdy, N.A. Abdel-Khalek, M. El-Aref, A. Elmanawi, Effect of microwave pre-treatment on the magnetic properties of iron ore and its implications on magnetic separation, Separation Purif. Technol. 136 (2014) 223-232.

[41] J.A. Aguilar, I. Gomez, Microwaves applied to carbothermic reduction of iron ore pellets, Int. Microw. Power Inst. 32(2) (1997) 67-73. [42] N. Standish, W. Huang, Microwave application in carbothermic reduction of iron ores, ISIJ Int., 31(3) (1991) 241-245. [43] D.A. Jones, T.P. Lelyveld, S.D. Mavrofidis, S.W. Kingman, N.J. Miles, Microwave heating applications in environmental engineering- a review, Conserv. Recycl. 34 (2002) 75-90. [44] C.A. Pickles, Thermodynamic analysis of the separation of zinc and lead from electric arc furnace dust by selective reduction with metallic iron, Sep. Purif. Technol. 59 (2008) 115-128. [45] H.K. Chen, C.Y. Yang, Effects of additives on carbothermic reduction of zincoxide, J. Mater. Sci. Lett. 18 (1999) 1399-1400. [46] S.W. Kingman, W. Vorster, N.A. Rowson, The influence of mineralogy on microwave assisted grinding, Miner. Eng. 13 (3) (2000) 313-327. [47] X. Sun, J-Y. Hwang, X. Huang, The Microwave Processing of Electric Arc Furnace Dust, JOM, 60(10) (2008) 35-39.

List of figures Fig. 1. Schematic diagram of the microwave experimental set-up. Fig. 2. X-ray diffraction (XRD) patterns of CRC, EAFSS, and EAFCS dusts. Fig. 3. SEM images of the studied materials. (A) CRC; (B) EAFSS; (C) EAFCS; (D) BFS. Frank = Franklinite; Ch= Chromite; Zn=Zincite; He= Hematite.

Fig. 4. X-ray diffraction pattern of BFS. Fig. 5. DSC curves for CRC, EAFSS, and EAFCS dusts. Fig. 6. TG curves for CRC, EAFSS, and EAFCS dusts. Fig. 7. TG-DSC curves for CRC dust. Fig. 8. TG-DSC curves for EAFCS dust.

Fig. 9. TG-DSC curves for EAFSS dust. Fig. 10. Fig.11. Fig. 12.

Fig. 13.

Fig. 14. Fig. 15. Fig. 16.

Fig. 17. Fig. 18. Fig. 19. Fig. 20.

Fig. 21. Fig. 22.

Table (1) Chemical, mineralogical and physical properties of steel making dusts. Analyses

CRC

Chemical composition Element C Fe Zn Cr CaO MgO MnO SiO2 K2O

0.3 18.74 10.83 20.88 14.27 9.76 1.56 9.99 0.74

Mineralogical composition

EAFSS Concentration (wt.%) 0.5 23.70 19.84 3.19 11.91 7.21 5.82 8.75 1.49

EAFCS

1.5 23.50 35.76 0.47 5.93 1.07 3.99 3.13 3.21

Measuring device

LECO

XRF

Phase Franklinite (ZnFe2O4) Zincite (ZnO) Chromite (FeCr2O4) Lime (CaO) Periclase (MgO)

-xx xxxx x x

Particle morphology Franklinite (ZnFe2O4) Zincite (ZnO) Chromite (FeCr2O4)

-Sphere Irregular particles

Abundance xxxx xx x x x

Encapsulation particle Sphere Enclosed in glass sphere

xxxx xxxx -x --

XRD

Sphere Sphere --

Dielectric properties at room temperature (22 °C) and frequency of 2423 MHz Value Dielectric constant (ɛ') 1.78 1.80 1.77 Loss factor (ɛ") 0.028 0.041 0.022 δ 0.016 0.022 0.012

SEM-EDX

Open-ended coaxial probe

Particle size analysis Median (d50) d25 d75 d100

3.15 0.84 4.84 15.42

Value (μm) 2.37 0.31 3.94 13.00

1.63 0.48 2.96 14.62

10 0.22

Value 10-11 0.30

10-11 0.54

Laser diffraction

Physical properties pH Moisture content (%)

Table (2) Chemical, mineralogical and physical properties of blast furnace sludge. Chemical composition Element

Concentration (wt.%)

C content Fe CaO SiO2 MgO Al2O3 Na2O K2O Zn

24.90 38.95 7.76 7.07 1.57 2.21 0.52 0.42 0.45

Mineralogical composition Phase Hematite (Fe2O3)

Abundance xxxx

Measuring device LECO

XRF

XRD

x x

Quartz (SiO2) Calcite (CaCO3) Particle morphology Hematite (Fe2O3) Coke (C) Calcite (CaCO3) Thermal analyses Temperature (°C) 160 to 360 540, 660, and 760

Irregular particles (below 20 µm) Elongated large crystals (>125 µm) Coarse grains (125 - 250 μm)

SEM-EDX

Reaction Evaporation of free and chemically combined water Coke gasification CO/ CO2

TG - DSC

Dielectric properties at room temperature (22 °C) and frequency of 2423 MHz

Dielectric constant (ɛ') Loss factor (ɛ") δ

Value 8.96 3.13 0.35

Open-ended coaxial probe

Particle analysis Median (d50) d25 d75 d100

Value (μm) 18.41 6.97 58.37 228.6

Laser diffraction

Physical properties pH Moisture content (%)

Value 8-9 32

Table (3) Total carbon content of BFS at different size fractions Size fraction (µm)

Total carbon (%)

SD (%)

Raw BFS < 32 +45-65 +90-125

24.9 13.3 22.5 36.3

0.08 0.05 0.37 0.25

+125-250 >250

47.5 62.2

0.79 0.78

Table (4) Chemical analysis of CRC residues after microwave heating.

Sample code Untreated CRC CRC:Graphite CRC:BFS CRC:BFS CRC:BFS CRC:125µm BFS CRC:250 µm BFS

Reductant Percentage

(wt. %) 10 25 25 25 12.5 10

Composition (wt. %)

Treatment time (min)

Zn

Cr

Fe

Ca

Mg

Si

Mn

Ni

K

Na

Al

Cl

Pb

Ni

20 5 10 20 20 20

10.83 0.34 8.62 7.77 2.97 1.25 0.78

20.88 22.88 13.71 13.47 14.09 18.40 20.60

18.74 18.18 28.28 28.16 29.83 24.37 22.80

10.19 8.01 6.62 7.29 7.93 7.69 8.25

5.88 5.59 3.17 2.41 3.79 4.25 4.54

4.66 6.23 4.16 4.74 4.41 5.81 5.69

1.20 1.41 0.95 0.87 1.12 1.19 1.12

0.25 0.34 0.22 0.22 0.30 0.22 0.27

0.61 0.19 0.43 0.20 0.22 0.17 0.15

1.25 0.48 0.66 0.52 0.32 0.44 0.54

0.42 0.13 0.48 1.37 0.31 0.57 0.47

0.29 0.045 0.23 0.047 0.05 0.053 0.06

0.11 0.029 0.028 0.025 -

0.25 0.34 0.22 0.22 0.30 0.22 0.27

Zn removal (%) 96.86 20.40 28.25 72.57 88.45 92.79

Sample code Untreated EAFSS EAFSS:Graphite EAFSS:BFS EAFSS:BFS EAFSS:BFS EAFSS:125 µm BFS EAFSS:250 µm BFS

Reductant percentage (wt. %)

Treatment time (min)

12.5 47 47 47 25 19

20 5 10 20 20 20

Zn removal (%)

Composition (wt.%) Zn 19.84 1.22 2.65 7.92 3.12 1.26 0.78

Cr 3.19 3.06 2.16 2.26 2.12 2.96 1.89

Fe 23.70 35.12 33.11 35.53 39.53 36.04 34.80

Ca 8.51 10.62 8.14 8.33 9.43 9.87 10.12

Mg 4.34 4.12 2.73 2.79 3.79 4.02 4.23

Si 4.09 5.12 3.47 4.34 4.14 4.02 4.86

K 1.23 0.14 0.60 0.12 0.12 0.16 0.20

Mn 4.50 4.12 2.89 2.99 3.99 4.28 4.16

Na 3.67 0.62 1.37 1.12 0.62 0.66 0.54

Al 0.46 1.12 0.90 1.49 1.22 1.13 1.22

Table (5) Chemical analysis of EAFSS residues after microwave heating.

Cl 1.25 0.07 0.70 0.059 0.059 0.07 0.06

Pb 0.62 0.10 0.10 -

Ni 0.37 0.33 0.29 0.32 0.32 0.36 0.28

Cu 0.30 0.22 0.19 0.18 0.18 0.29 0.28

93.85 36.23 60.08 84.27 93.64 96.06

Sample code Untreated EAFCS EAFCS:Graphite EAFCS:BFS EAFCS:BFS EAFCS :125µm BFS EAFCS:250 µm BFS

Reductant percentage

(wt. %) 15 59 59 31 23

Treatment time (min) 20 10 20 20 20

Composition (wt.%) Zn

Cr

Fe

Ca

Mg

Si

Mn

K

Na

Al

Cl

S

Pb

Cu

35.76 5.29 13.09 5.98 4.53 3.42

0.47 0.78 0.48 0.37 0.52 0.59

23.50 38.02 37.62 42.58 38.39 38.62

4.23 9.8 5.54 5.48 9.26 11.54

0.64 1.18 0.58 0.95 1.61 1.58

1.46 2.35 2.23 2.09 3.29 3.23

3.09 4.46 2.12 1.96 3.13 2.86

2.66 0.91 1.11 0.49 0.63 0.82

6.36 2.50 2.04 1.90 2.26 2.04

0.14 0.59 1.79 4.68 1.17 0.79

1.69 0.05 -

1.22 1.10 0.85 0.48 0.98 0.85

2.15 0.16 0.14 -

0.25 0.36 0.29 0.20 0.29

alysis of EAFCS residues after microwave heating.

Zn removal (%) 85.20 63.39 83.27 87.33 90.43

Ta ble (6) Ch em ical an

Table (7) Residual Carbon Content (wt.%) after microwave heating for 20 min at the microwave power of 1100 W Mixture code

Carbon (%)

SD (%)

EAFCS:BFS

3.85 0.559

0.17 0.03

EAFCS :125µm BFS

2.16

0.07

EAFCS:250 µm BFS

2.78 2.63 0.627 2.39 2.25 3.17 0.453 1.81 1.21

0.16 0.13 0.01 0.30 0.22 0.40 0.08 0.32 0.33

EAFCS: Graphite

EAFSS: Graphite EAFSS:BFS EAFSS :125µm BFS EAFSS:250 µm BFS CRC:Graphite CRC:BFS CRC:125µm BFS CRC:250 µm BFS

Highlights:     

Simultaneous recycling of steelmaking dust and BFS by microwave heating was studied. BFS was utilized as a reducing agent to remove zinc from the steelmaking dusts. Zinc removal rate depends on the amounts of FeO and CaO in the mixtures. Highest zinc removal was achieved for mixtures with high CaO and low FeO contents. The results showed that BFS can be applied to the processing of steelmaking dusts.