Journal of Environmental Management 143 (2014) 208e213
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Reduction behavior of zinc ferrite in EAF-dust recycling with CO gas as a reducing agent Chia-Cheng Wu a, Fang-Chih Chang b, *, W.-S. Chen c, Min-Shing Tsai d, Ya-Nang Wang b a
Resource R & D Department, China Hi-Ment Co., Kaohsiung, Taiwan The Experimental Forest, College of Bio-Resources and Agriculture, National Taiwan University, Nan-Tou 55750, Taiwan c Sustainable Environment Research Center, National Cheng Kung University, Tainan 709, Taiwan d Department of Resources Engineering, National Cheng Kung University, Tainan 701, Taiwan b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 November 2012 Received in revised form 18 March 2014 Accepted 11 April 2014 Available online
EAF-dust containing metal oxides can be regarded as an important source for zinc and iron. In this study, the reduction behavior of zinc ferrite with CO gas as a reducing agent under different temperatures was investigated to develop a new process for the recovery of zinc and iron from EAF-dust. The results of the phase studies with synthetic franklinite show that zinc substituted wustite, and spinel with low zinc content formed at lower temperatures from 450 to 850 C due to incomplete zinceiron-separation. Zinc ferrite was completely reduced to metallic zinc and iron at 950 C. After evaporation and condensation, metallic zinc was collected in the form of zinc powder while iron, the reduction residue, was obtained in the form of direct reduced iron (DRI). The mass balance indicates a high zinc recovery ratio of over 99%. The new treatment process by thermal reduction with CO gas as a reducing agent achieved higher recovery and metallization grade of both zinc and iron from EAF-dust at lower temperatures than other commercial processes. The metallic products can be used directly as semi-products or as raw materials for refinery. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: EAF-dust Reduction Zinc ferrite Zinc oxide
1. Introduction The production of electro steel by electric arc furnace (EAF) from scrap iron provides an efficient recycling method for iron resources in the economic system. One of the main byproducts from EAF is fly dust, also known as EAF-dust, collected in the offgas treatment devices. Because of the increasing galvanization of steel with zinc, EAF-dust contains large quantities of zinc and iron (Machado et al., 2006; Salihoglu et al., 2007; Chen et al., 2011; Machado et al., 2011). The zinc content is usually 20e30%, but in some cases it can be more than 40% (Havlık et al., 2006). The specific yield of EAF-dust is between 15 and 20 kg per ton raw steel production (Machado et al., 2006). In comparison with zinc blend ore, EAF-dust can be regarded as an excellent secondary source for zinc. The main mineral compounds in EAF-dust are zinc oxide and zinc ferrite (franklinite, ZnFe2O4) (Machado et al., 2006; Chen et al., 2011). The content of zinc oxide increases with increased zinc concentration (Hagni and Hagni, 1994). Moreover, alkali metal
* Corresponding author. Tel.: þ886 2 33664651; fax: þ886 2 33654520. E-mail address: [email protected] (F.-C. Chang). http://dx.doi.org/10.1016/j.jenvman.2014.04.005 0301-4797/Ó 2014 Elsevier Ltd. All rights reserved.
chlorides, lead oxide, zinc chloride and other heavy metal salts are also often included in EAF-Dust (Li and Tsai, 1993; Machado et al., 2006; Quijorna et al., 2014). Zinc ferrite has extremely low water solubility and this can be regarded as a guarantee of its stability in the environment. However, it adds to the difficulty of zinc and iron recovery. The methods developed for zinc recovery can be divided roughly into two groups: the pyrometallurgical and the hydrometallurgical processes. The pyrometallurgical processes that reduce zinc oxide using carbon, then distil the metallic zinc from the resulting mix in an atmosphere of carbon monoxide have become the most commonly used commercial techniques (Huaiwei and Xin, 2011). In the Waelz-Process, for example, EAF-dust is placed into a rotary kiln with a coke charge. The material is then treated at a temperature of over 1200 C. The diffusion of oxygen from the reaction atmosphere into the material bed is limited by the airflow control. Thus, the coke is partially oxidized to CO which reduces zinc oxide and zinc ferrite to metallic zinc (Galchinetskii et al., 2005; Huda et al., 2011). The direct reduction of zinc oxide by solid coke also occurs. Besides CO and CO2, the gas emitted from the material bed contains zinc vapor if the temperature exceeds the boiling point of zinc (906 C). Zinc vapor is then re-oxidized in the kiln atmosphere
C.-C. Wu et al. / Journal of Environmental Management 143 (2014) 208e213
so that the crude zinc oxide can be separated from the CO/CO2 enriched exhausted gas (Rao, 2006). The recovery ratio for zinc lies between 95 and 99% (Pickles, 2008). By adding quartz sand the iron-containing residue is turned into slag which is often used as foundation material in road construction. In the Rotary Hearth Furnace (RHF)-Process, EAF-dust is pelletized with coke and charged into a furnace with counter current airflow. However, instead of slag, direct reduced iron (DRI) is yielded at different temperatures controlled by a gas burner allocated on the hearth wall. The reaction temperature ranges between 1100 C and 1200 C. The recovery ratio for zinc is about 99% (Pickles, 2008). The main product is crude ZnO dust collected in the off-gas treatment devices. Besides, methods combining the pyrometallurgical and hydrometallurgical technologies have been suggested. In the Caron-ZincProcess (Nyirenda, 1990; Harvey, 2006; Suetens et al., 2014), reduction roasting is conducted in which the zinc ferrite-containing EAF-dust is partially reduced to zinc oxide and magnetite with a CO/CO2 gas mixture at the relative lower temperature of 750 C: 3ZnFe2O4 þ CO / 3ZnO þ 2Fe3O4 þ CO2
(1)
The roasted product is then leached with ammonium carbonate solution to recover zinc followed by electrolysis (Jha et al., 2001; Ruiz et al., 2007); the zinc recovery is about 80e85%. Dutra et al. (2006) and Ruiz et al. (2007) obtained similar results by means of acidic or alkaline leaching. Reduction roasting has not been commercialized because of its low zinc recovery rate. In addition, the magnetite-containing leaching residues, which have a zinc content of about 10%, cannot be charged into the conventional steel manufacturing process. The most crucial conditions of pyrometallurgical processes are the reducing agent and temperature; a temperature exceeding the sintering point of iron-containing slag would result in difficulty to separate iron from slag. The use of coke as reducing agent and fuel causes an oxidative condition in fumes and re-oxidation of zinc vapor. Using a reducing agent such as CO gas instead could solve these problems (Best and Pickles, 2001; Pickles, 2007). Besides, the Boudouard reaction (Eq. (2)) should also be noted: 2CO 4 CO2 þ C DH ¼ 172.58 kJ
(2)
Because of the strong exothermal reaction and the change of gas volume, the Boudouard reaction depends strongly on temperature. The volume of CO2 increases with decreasing temperature and is accompanied by the formation of fine carbon or graphite particle. In accordance with the reaction, a reduction gas, such as CO or CO/CO2 mixture, could lose its reduction effect at low temperatures so that the reduction could only be partially achieved. The phase diagram of the Boudouard reaction is shown in Fig. S1. The direct reduction with CO gas instead of coke has the following advantages: -
-
Reduction at lower temperature is possible because of the extremely low oxygen partial pressure. It not only reduces the energy consumption but also avoids the sintering of ironcontaining residue that could lead to congestion in the furnace. The tail gas remains reductive with a CO-surplus; thus the recovery of metallic zinc is possible.
The aim of this study is to develop a new process for the recovery of zinc and iron from EAF-dust. In order to reduce synthetic zinc ferrite and EAF-dust, CO was used as reducing agent at different temperatures in this experimental research work. Reduction in relation to dependence on temperature was estimated
209
from the phase studies. All reduction solid residues with synthetic zinc ferrite were analyzed by X-ray power diffraction (XRD). Additionally, an estimation of mass balance is presented. 2. Materials and methods 2.1. Materials and preparations EAF-dust from an electro steel mill in southern Germany and synthetic zinc ferrite were used. For the preparation of the synthetic zinc ferrite, zinc oxide and hematite of reagent grade were first thoroughly mixed with molar ratio of 1:1 and pressed to tablets. The tablets were then sintered in a tube furnace with a quartz tube as the main reaction chamber at 750 C in Ar gas flow for 4 h (Akhtar et al., 2014). The schema of the furnace device is shown in Fig. S2. The formation of zinc ferrite can be described as follows: ZnO þ Fe2O3 / ZnFe2O4
(3)
After 4 h of sintering followed by quenching, the tablets were ground, remixed, and repressed to tablets. The second sintering was carried out under the same conditions. After 3 repeated procedures aimed at obtaining the best possible purification and crystallization, the product was kept for later determination and experiments. The EAF-dust sample was ground and then dried at 105 C for 12 h and kept for later analysis and experiments. 2.2. Reduction experiments The reduction experiments were carried out in the same furnace device shown in Fig. S2. Instead of Ar, CO was introduced into the reaction zone as reduction gas. At the exit, cooling water was used to quench the tail gas and condense the vapor. Quartz wool was used as filter to remove the condensed dust. The furnace was first preheated to the given temperatures of 450 C, 550 C, 650 C, 750 C, 850 C, and 950 C. Samples packed in a quartz boat were then sent into the hot zone. A constant CO gas flow rate of 833 ml/min and flow velocity of 0.19 m/sec were chosen. After 2 h of reaction, the quartz tube was shifted so that the boat was outside the furnace and could be quenched by cooling water. Both the solid residue and dust were ground and collected for later analysis. 2.3. Characteristics of materials All samples and products were characterized with X-ray diffraction (XRD, Philips X’Pert). The chemical compositions were quantitatively analyzed using flame atomic adsorption (Perkine Elmer, AA-100). The content of Fe2þ was determined by the titration method. The chemical composition of the single ferrite phase was also calculated. The measured d-spacing (d in A) by XRD was used to calculate the lattice constant (a in A) of different ferrite phases in accordance with Bragg’s equation (Eq. (4)):
l ¼ 2d$sin q
(4)
Schaefer/McCune and Popov/Ilinova came to the conclusion that the lattice constant of spinel phases ZnxFe3exO4 (with 0 x 1) shows a linear dependence on the zinc content represented by the parameter x (Popov and Ilinova, 1970; Schaefer and McCune, 1986). A change of slope from x ¼ 0.7 was also reported (Popov and Ilinova, 1970). On average, the lattice constant a of the ferrite phase lies between 8.392 A for magnetite Fe3O4 (x ¼ 0) and 8.441 A for franklinite
210
C.-C. Wu et al. / Journal of Environmental Management 143 (2014) 208e213
ZnFe2O4 (x ¼ 1); the following equations (5) and (6), gained from investigations with different synthetic zinc ferrites in previous research, were used for these calculations (according to Wu, 2003):
Table 1 Chemical compositions of the solid residues after the reduction of zinc ferrite by CO gas at various temperatures. Temperature ( C)
x ¼ 17:6991a 148:6142 for 0 x 0:7
(5)
x ¼ 30:7692a 258:8646 for 0:7 < x 1
(6)
All experiments were prepared and measured in triplicate samples to evaluate the reproducibility of the results.
450 550 650 750 850 950
Content (%)
Zinc evaporation rate
Zn
Fe
C
29.2 22.2 10.9 14.3 0.12 0.09
51.4 35.9 19.56 33.9 80.36 90.78
4.35 36.3 55.4 44.2 6.35 4.16
N.D. N.D. 4.80 27.93 99.79 99.86
3. Results and discussion 3.1. Characteristics of synthetic zinc ferrite Fig. S3 shows the XRD pattern of synthetic zinc ferrite. No other phases of zinceiron-oxides were identified in the product after 20 h of sintering at 750 C in Ar gas flow. According to equations (5) and (6), the calculated lattice constant of synthetic zinc ferrite is 8.4425 0.0001 A, which indicated the parameter x to be 0.9998. Moreover, the content of Fe2þ was under the detection limit. 3.2. Characteristics of EAF-dust The ground EAF-dust was of fine particle with an average size of 12.17 mm. The chemical composition is listed in Table S1 along with the average values from literature for comparison (Hagni and Hagni, 1994; Ruiz et al., 2007; Oustadakis et al., 2010). In addition to zinc and iron, the dust from carbon steel production also contained lead, manganese, alkali metals, chlorine and some trace elements such as Cd (0.08%), Cr (0.27%), Cu (0.46%) and Ni (0.09%). In comparison with the values reported in previous literature, the sample can be classified as high zinc-containing EAF-dust. Fig. S4 shows the XRD pattern of the EAF-dust sample. Zinc oxide and franklinite were identified as the main compounds. Judged by the 2q values and the calculation using Bragg’s equation, the lattice constant of the zinc ferrite phase was 8.4394 0.0018 A. This indicates a chemical formula of Zn0.8Fe2.2O4 and a Fe2þ/Fe3þ ratio of 0.05 according to equations (5) and (6). However, the Fe2þ/ Fe3þ ratio was determined to be 0.036 by chemical analysis. The difference in the ratio could be the consequence of the existence of other heavy metals such as manganese and nickel in the ferrite phase. The existence of manganese ferrite (Jacobsite, MnFe2O4) in particular, which has a similar XRD pattern as franklinite in EADdust, has often been reported (Machado et al., 2006; Holloway et al., 2007). Lead exists mainly as basic lead chloride PbCl2,Pb(OH)2 (Pickles, 2009), also known as laurionite which is a weathering product of lead slag under the influence of sea water. The other chlorides found were sodium chloride (NaCl) and potassium chloride (KCl). 3.3. Reduction of synthetic franklinite with CO gas Different grades of zinc vaporization were observed during the reduction of synthetic franklinite by CO gas at various temperatures. The compositions and the vaporization grade of zinc are listed in Table 1. The XRD patterns of solid residues are shown in Fig. 1. In general, the reduction of zinc ferrite and the recovery of zinc as metal dust increase with rising temperatures. This can be well explained by the loss of reduction effect of CO gas at low temperatures and the change of the reaction free energy as a function of temperature. At 450 C, the reduction effect of CO gas was so weak due to the low temperature and shifting of the Boudouard equilibrium to the side of CO2 that franklinite was only partially reduced. In addition
to metallic iron, the solid residue contained zinc ferrite and wustite (FeO). The calculation of the lattice constant and zinc content indicate the chemical formula of Zn0.7Fe2.3O4 for the remaining zinc ferrite phase. Based on the BaureGlaessnereDiagram, the coexistence of metallic iron and zinc ferrite can be explained by the direct reduction of magnetite to metallic iron at a temperature lower than 600 C (Varanda et al., 2002). The reduction effect of CO gas increased slightly at 550 C so that the bound zinc in ferrite was fully liberated to zinc oxide while iron was reduced to metallic iron and wustite. At 650 C, iron oxides were completely reduced to metallic iron. The phase identification shows the existence of iron carbide (cementite, Fe3C) and crystalline graphite after reduction at lower temperatures. The formation of iron carbide is the result of a series of reactions as follows: 2CO 4 C (graphite) þ CO2
(7)
C þ 3Fe / Fe3C(cementite)
(8)
As a consequence, the DRI was converted to cementite at lower temperatures (550 C). The reaction gas lost its reducing effect and the metal oxides were only partially reduced. The metallization grade for zinc, calculated by the evaporation rate, was less than 5%. With the increasing temperature, the vaporization of zinc rose dramatically. One of the reasons is that the Boudouard equilibrium shifted to the side of CO which maintained the reduction effect of the gas. The other reason is the decreasing free energy of the reduction of zinc ferrite and zinc oxide with increasing temperature. The carbon content in the iron-containing solid residue decreased gradually to less than 5%. At 950 C, the collected zinc dust contained nearly 100% of metallic zinc (99.9%). The results show that the temperature affects not only the reduction effect of CO gas but also the reaction free energy. The reactions at various temperatures based on the phase and elemental analysis are summarized in Table 2. It should be noted that zinc oxide and wustite do not exist in the pure phase form. Some authors have concluded that zinc oxide and iron(II) oxide dissolve in each other and form solid solutions at high temperatures (>800 C). The solubility of zinc oxide in wustite ranges from 14% to 23% while the solubility of wustite in zinc oxide ranges between 5% and 19.36% inside the temperature range of 750 Ce1000 C (Bates et al., 1966; Claude et al., 1978; Itoh and Azakami, 1995). Such oxides are known as zinc-substituted wustite and iron-substituted zinc oxide. However, the ratios of these two phases can only be estimated very roughly. The phase status for the reduction of zinc ferrite in relation to dependence on temperature can be estimated as shown in Fig. 2. In summary of the results and discussions above, the optimal reduction temperature to recover iron and zinc from franklinite with CO gas is 950 C.
C.-C. Wu et al. / Journal of Environmental Management 143 (2014) 208e213
1
1 Iron Fe 3 Zincite ZnO 5 Graphite C
950oC
5
22 2
2
211
2 Cohenite Fe3C 4 Wustite FeO 6 Franklinite ZnFe2O4 1
1
2 CO2
CO 850oC 1 5
22 2
3
1
1
2
750oC 1 5
3
3
3
3
3
22
2
3 2
3
3 1
3 3
1
3
3
33 2
1
3
3
33
1
3
1
Intensity
650oC
5
3
1 2 22 2 2 2 3 2
2
1
550oC
5
3
3
4 3 2 4
4
450oC
6 3
20
2 21 2 3
3 36
1 2
2
3
2 3
40
63 1
60
2 Theta
80
Fig. 1. XRD patterns of reduction solid residues with synthetic zinc ferrite at various temperatures.
Reduction of pelletized EAF-dust with CO gas at 950 C was carried out. The chemical compositions and the phase statuses of solid residue (the iron fraction) and dust collected in off-gas treatment devices (the zinc fraction) are listed in Table 3. Similar to the reduction experiment of synthetic franklinite at 950 C, the iron fraction contained mainly metallic iron and cementite Fe3C. The carbon content increased slightly from 4.16% to Table 2 Reduction reactions of zinc ferrite by CO gas at various temperatures. Temp. ( C)
Reactions
450e550
ZnFe2O4 þ 3CO 4 ZnO þ 2Fe þ 3CO2 ZnFe2O4 þ CO 4 ZnO þ 2FeO þ CO2 (3 x) ZnFe2O4 þ (1 x) CO 4 (3 3x) ZnO þ 2ZnxFe3xO4 þ (1 x) CO2 ZnFe2O4 þ 3CO 4 ZnO þ 2Fe þ 3CO2 ZnFe2O4 þ CO 4 ZnO þ 2FeO þ CO2 ZnFe2O4 þ 4CO 4 Zn(g) þ 2Fe þ 4CO2 ZnFe2O4 þ 3CO 4 ZnO þ 2Fe þ 3CO2 ZnFe2O4 þ 4CO 4 Zn(g) þ 2Fe þ 4CO2
550e650 650e950 >950
4.59% so that the austenite phase, a solid solution of g-iron and cementite, was also found. The reason for this can be explained by the evaporation of the chlorides, such as NaCl and KCl, which consumed a certain part of energy so that the energy utilization for 1 zinc oxide + iron substituted zinc oxide
zinc Mole of Metals
3.4. Reduction of EAF-dust with CO gas
metallic zinc vapor
0.08 0 spinel 0.24 zinc substituted wustite iron
1
2 450
metallic iron + cementite
550
650
750 Temp. ( oC)
850
950
Fig. 2. Phase diagram of zinc ferrite reduction in relation to dependence on temperature.
212
C.-C. Wu et al. / Journal of Environmental Management 143 (2014) 208e213
Table 3 Chemical compositions and the phase statuses of the solid residues and the collected dust from the reduction of EAF-dust by CO gas at 950 C. Element
Content (%)
Zn Fe Pb K Na Cl C Mn Phases
Solid residue
Zinc dust
0.065 89.9 0.257 0.193 0.180 0.012 4.59 4.50 Met. Fe, austenite Fe3C, graphite
80.3 0.068 9.51 1.93 1.23 7.32 0.49 n. d. Met. Zn, ZnO, met. Pb, PbO, KPb2Cl5, NaCl, KCl
the reduction itself was actually lower in comparison with the previous experiments. The purity of iron fraction, however, remained at 90%. The collected zinc dust, on the other hand, contained both metallic zinc and zinc oxide. One of the reasons is the reduction of energy utilization as described above. The content of zinc decreased to 80%. Lead existed as metals, oxides and chlorides (KPb2Cl5) in the collected dust. The concentrations of Na and K in the dust were 1.23% and 1.93%, respectively. The recovery ratio of zinc and lead were both over 99.8%. These results show that the reduction effect of CO gas at 950 C is capable of recovering zinc not only from the synthetic franklinite but also from EAF-dust containing franklinite and zinc oxide. The mass balance at 950 C is shown in Fig. 3. The recovery ratios of zinc and iron were 99.9% and 99.8%, respectively. The metallization grade was approximately 95% for iron and 90% for zinc. 4. Conclusions The investigation for recovering zinc and iron from EAF-dust in terms of a pyrometallurgical process using CO gas as reducing agent at various temperatures is described in this paper. The lower temperature led not only to the incomplete reduction of oxides but also the formation of carbide. Thus the purity of iron was reduced moderately. The optimal working temperature was determined to be 950 C. The reduction experiment with EAF-dust also demonstrated that this procedure was only slightly negatively influenced by the existence of other impurities such as alkali chlorides and lead compounds. After the treatment, two main products were obtained, namely direct reduced iron (DRI) and metallic zinc dust. With a purity of over 90% and an estimated metallization grade of over 95%, the iron sponge can be charged directly into the EAF process without any pretreatment. The zinc dust, on the other hand, can be used as raw material in the zinc refinery industry after certain treatments to remove chlorides, for example by distillation. EAF-Dust (g) 32.98 23.00 1.23
Zinc fraction (g) 32.96 1.20
→
4.10
1.27
1.15
0.55
Legend (g) Fraction
Reduction o at 950 C
CO gas
3.06
Zn
Fe
Pb
K
Na
Cl
Mn
C
Iron fraction (g)
0.03 4.033
1.26
0.017 22.97 0.066 0.008
3.02
2.01
0.029 0.003
0.01
1.14
1.17
Fig. 3. Mass balance of the process to recover zinc and iron from EAF dust.
The mass balance shows that this process is characterized by zero waste. Different from the conventional pyrometallurgical processes with solid coke, CO gas is suggested as reductant and indirect heating agent in this new process. With relative lower temperature, DRI can be yielded. Instead of reaction air that causes oxidation of zinc vapor, the CO gas guarantees a high metallization grade of zinc. In comparison with combined technologies such as the Caron-ZincProcess with CO/CO2 mixture, the reduction rate of zinc ferrite is raised to 100% so that the recovery ratio of zinc is also improved. Acknowledgments The financial supports by the Deutsche Forschungsgemeinschaft, Germany (project number Go 383/20-1) and the Ministry of Economic Affairs (project number 95-EC-17-A-10S1-0007) are gratefully acknowledged. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jenvman.2014.04.005. References Akhtar, M.N., Yahya, N., Sattar, A., Ahmad, M., Idrees, M., Asif, M.H., Khan, M.A., 2014. Investigations of structural and magnetic properties of nanostructured Ni0.5 þ xZn0.5xFe2O4 magnetic feeders for CSEM application. Int. J. Appl. Ceram. Technol. http://dx.doi.org/10.1111/ijac.12222. Bates, C.H., White, W.B., Roy, R., 1966. The solubility of transition metal oxides in zinc oxide and the reflectance spectra of Mn2þ and Fe2þ in tetrahedral fields. J. Inorg. Nucl. Chem. 28, 397e405. Best, T.E., Pickles, C.A., 2001. In-flight plasma reduction of electric arc furnace dust in carbon monoxide. Can. Metall. Q. 40, 61e78. Chen, W.S., Shen, Y.H., Tsai, M.S., Chang, F.C., 2011. Removal of chloride from electric arc furnace dust. J. Hazard. Mater. 190, 639e644. Claude, J.M., Zanne, M., Gleitzer, C., Aubry, J., 1978. Preparation and study of ZnxFe0.85xO (0.085 x 0.17). J. Solid State Chem. 24, 395e400. Dutra, A.J.B., Paiva, P.R.P., Tavares, L.M., 2006. Alkaline leaching of zinc from electric arc furnace steel dust. Min. Eng. 19, 478e485. Galchinetskii, L.P., Galkin, S.N., Ryzhikov, V.D., Rybalka, I.A., Voronkin, E.F., Lalayants, A.I., Starzhinskii, N.G., Silin, V.I., 2005. Chemical interaction in the ZnSe-ZnTe-Se-H2-C system. Inorg. Mater. 41, 934e938. Hagni, A.M., Hagni, R.D., 1994. Significance of Mineralogy of Electric Arc Furnace (EAF) Dust in Pyrometallurgical Treatment to Render the Dust Non-Hazardous. Extraction and Processing for the Treatment and Minimization of Wastes. TMSAIME, Warrendale, PA, pp. 1137e1148. Harvey, T.G., 2006. The hydrometallurgical extraction of zinc by ammonium carbonate: a review of the schnabel process. Min. Process. Extr. Metall. Rev. 27, 231e279. Havlık, T., Souza, B.V., Bernardes, A.M., Schneider, I.A.H., Miskufova, A., 2006. Hydrometallurgical processing of carbon steel EAF dust. J. Hazard. Mater. 135, 311e 318. Holloway, P.C., Etsell, T.H., Murland, A.L., 2007. Use of secondary additives to control the dissolution of iron during Na2CO3 roasting of La oroya zinc ferrite. Metall. Mater. Trans. B 38B, 793e808. Huaiwei, Z., Xin, H., 2011. An overview for the utilisation of wastes from stainless steel industries. Resour. Conserv. Recy 55, 745e754. Huda, N., Naser, J., Brooks, G., Reuter, M.A., Matusewicz, R.W., 2011. Computational fluid dynamic modeling of zinc slag fuming process in top-submerged lance smelting furnace. Metall. Mater. Trans. B 179H, 1e17. Itoh, S., Azakami, T., 1995. Thermodynamic study on the zinc-iron-oxygen system at 1200K. Mater. Trans. JIM 36, 1074e1080. Jha, M.K., Kumar, V., Singh, R.J., 2001. Review of hydrometallurgical recovery of zinc from industrial wastes. Resour. Conserv. Recycl 33, 1e22. Li, C.L., Tsai, M.S., 1993. A crystal phase study of zinc hydroxide chloride in electricarc-furnace dust. J. Mater. Sci. 28, 4562e4570. Machado, J.G.M.S., Brehm, F.A., Moraes, C.A.M., dos Santos, C.A., Vilela, A.C.F., da Cunha, J.B.M., 2006. Chemical, physical, structural and morphological characterization of the electric arc furnace dust. J. Hazard. Mater. 136, 953e960. Machado, A.T., Valenzuela-Diaz, F.R., de Souza, C.A.C., de Andrade Lima, L.R.P., 2011. Structural ceramics made with clay and steel dust pollutants. Appl. Clay Sci. 51, 503e506. Nyirenda, R.L., 1990. An appraisal of the caron zinc process when zinc ferrite is reduced to a magnetite containing product. Min. Eng. 3, 319e329. Oustadakis, P., Tsakiridis, P.E., Katsiapi, A., Agatzini-Leonardou, S., 2010. Hydrometallurgical process for zinc recovery from electric arc furnace dust (EAFD)
C.-C. Wu et al. / Journal of Environmental Management 143 (2014) 208e213 part I: characterization and leaching by diluted sulphuric acid. J. Hazard. Mater. 179, 1e7. Pickles, C.A., 2007. Thermodynamic analysis of the selective reduction of electric arc furnace dust by carbon monoxide. High. Temp. Mater. Process 26, 79e91. Pickles, C.A., 2008. Thermodynamic analysis of the separation of zinc and lead from electric arc furnace dust by selective reduction with metallic iron. Sep. Purif. Technol. 59, 115e128. Pickles, C.A., 2009. Thermodynamic analysis of the selective chlorination of electric arc furnace dust. J. Hazard. Mater. 166, 1030e1042. Popov, G.P., Ilinova, G.N., 1970. Deviation of certain ferrite solid solutions from Vegard’s law. Russ. J. Phys. Chem. 45, 299e301. Quijorna, N., de Pedro, M., Romero, M., Andrés, A., 2014. Characterisation of the sintering behaviour of Waelz slag from electric arc furnace (EAF) dust recycling for use in the clay ceramics industry. J. Environ. Manage. 132, 278e286. Rao, S.R., 2006. Resource Recovery and Recycling from Metallurgical Wastes. Elsevier Waste Management series v. 7, Amsterdam, pp. 269e327. Ruiz, O., Clemente, C., Alonso, M., Alguacil, F.J., 2007. Recycling of an electric arc furnace flue dust to obtain high grade ZnO. J. Hazard. Mater. 141, 33e36.
213
Salihoglu, G., Pinarli, V., Salihoglu, N.K., Kar, G., 2007. Properties of steel foundry electric arc furnace dust solidified/stabilized with Portland cement. J. Environ. Manage. 85, 190e197. Schaefer, S.C., McCune, R.A., 1986. Electrochemical determination of thermodynamic properties and an X-ray diffraction investigation of the Fe3O4-ZnFe2O4 system. Metall. Mater. Trans. B 17, 515e521. Suetens, T., Klaasen, B., Acker, K.V., Blanpain, B., 2014. Comparison of electric arc furnace dust treatment technologies using exergy efficiency. J. Clean. Product. 65, 152e167. Varanda, L.C., Jafelicci, M., Tartaj, P., O’Grady, K., González-Carreño, T., Morales, M.P., Muñoz, T., Serna, C.J., 2002. Structural and magnetic transformation of monodispersed iron oxide particles in a reducing atmosphere. J. Appl. Phys. 92, 2079e2085. Wu, C.C., 2003. Phase Investigation in the Reduction of Franklinite for the Zinc Extraction from Steel Plant. Ph.D. thesis. Institute for Treatment and Disposal Technology, Technical University of Clausthal, Clausthal-Z., Germany.
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Reduction behavior of zinc ferrite in EAF-dust recycling with CO gas as a reducing agent Chia-Cheng Wu a, Fang-Chih Chang b, *, W.-S. Chen c, Min-Shing Tsai d, Ya-Nang Wang b a
Resource R & D Department, China Hi-Ment Co., Kaohsiung, Taiwan The Experimental Forest, College of Bio-Resources and Agriculture, National Taiwan University, Nan-Tou 55750, Taiwan c Sustainable Environment Research Center, National Cheng Kung University, Tainan 709, Taiwan d Department of Resources Engineering, National Cheng Kung University, Tainan 701, Taiwan b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 November 2012 Received in revised form 18 March 2014 Accepted 11 April 2014 Available online
EAF-dust containing metal oxides can be regarded as an important source for zinc and iron. In this study, the reduction behavior of zinc ferrite with CO gas as a reducing agent under different temperatures was investigated to develop a new process for the recovery of zinc and iron from EAF-dust. The results of the phase studies with synthetic franklinite show that zinc substituted wustite, and spinel with low zinc content formed at lower temperatures from 450 to 850 C due to incomplete zinceiron-separation. Zinc ferrite was completely reduced to metallic zinc and iron at 950 C. After evaporation and condensation, metallic zinc was collected in the form of zinc powder while iron, the reduction residue, was obtained in the form of direct reduced iron (DRI). The mass balance indicates a high zinc recovery ratio of over 99%. The new treatment process by thermal reduction with CO gas as a reducing agent achieved higher recovery and metallization grade of both zinc and iron from EAF-dust at lower temperatures than other commercial processes. The metallic products can be used directly as semi-products or as raw materials for refinery. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: EAF-dust Reduction Zinc ferrite Zinc oxide
1. Introduction The production of electro steel by electric arc furnace (EAF) from scrap iron provides an efficient recycling method for iron resources in the economic system. One of the main byproducts from EAF is fly dust, also known as EAF-dust, collected in the offgas treatment devices. Because of the increasing galvanization of steel with zinc, EAF-dust contains large quantities of zinc and iron (Machado et al., 2006; Salihoglu et al., 2007; Chen et al., 2011; Machado et al., 2011). The zinc content is usually 20e30%, but in some cases it can be more than 40% (Havlık et al., 2006). The specific yield of EAF-dust is between 15 and 20 kg per ton raw steel production (Machado et al., 2006). In comparison with zinc blend ore, EAF-dust can be regarded as an excellent secondary source for zinc. The main mineral compounds in EAF-dust are zinc oxide and zinc ferrite (franklinite, ZnFe2O4) (Machado et al., 2006; Chen et al., 2011). The content of zinc oxide increases with increased zinc concentration (Hagni and Hagni, 1994). Moreover, alkali metal
* Corresponding author. Tel.: þ886 2 33664651; fax: þ886 2 33654520. E-mail address: [email protected] (F.-C. Chang). http://dx.doi.org/10.1016/j.jenvman.2014.04.005 0301-4797/Ó 2014 Elsevier Ltd. All rights reserved.
chlorides, lead oxide, zinc chloride and other heavy metal salts are also often included in EAF-Dust (Li and Tsai, 1993; Machado et al., 2006; Quijorna et al., 2014). Zinc ferrite has extremely low water solubility and this can be regarded as a guarantee of its stability in the environment. However, it adds to the difficulty of zinc and iron recovery. The methods developed for zinc recovery can be divided roughly into two groups: the pyrometallurgical and the hydrometallurgical processes. The pyrometallurgical processes that reduce zinc oxide using carbon, then distil the metallic zinc from the resulting mix in an atmosphere of carbon monoxide have become the most commonly used commercial techniques (Huaiwei and Xin, 2011). In the Waelz-Process, for example, EAF-dust is placed into a rotary kiln with a coke charge. The material is then treated at a temperature of over 1200 C. The diffusion of oxygen from the reaction atmosphere into the material bed is limited by the airflow control. Thus, the coke is partially oxidized to CO which reduces zinc oxide and zinc ferrite to metallic zinc (Galchinetskii et al., 2005; Huda et al., 2011). The direct reduction of zinc oxide by solid coke also occurs. Besides CO and CO2, the gas emitted from the material bed contains zinc vapor if the temperature exceeds the boiling point of zinc (906 C). Zinc vapor is then re-oxidized in the kiln atmosphere
C.-C. Wu et al. / Journal of Environmental Management 143 (2014) 208e213
so that the crude zinc oxide can be separated from the CO/CO2 enriched exhausted gas (Rao, 2006). The recovery ratio for zinc lies between 95 and 99% (Pickles, 2008). By adding quartz sand the iron-containing residue is turned into slag which is often used as foundation material in road construction. In the Rotary Hearth Furnace (RHF)-Process, EAF-dust is pelletized with coke and charged into a furnace with counter current airflow. However, instead of slag, direct reduced iron (DRI) is yielded at different temperatures controlled by a gas burner allocated on the hearth wall. The reaction temperature ranges between 1100 C and 1200 C. The recovery ratio for zinc is about 99% (Pickles, 2008). The main product is crude ZnO dust collected in the off-gas treatment devices. Besides, methods combining the pyrometallurgical and hydrometallurgical technologies have been suggested. In the Caron-ZincProcess (Nyirenda, 1990; Harvey, 2006; Suetens et al., 2014), reduction roasting is conducted in which the zinc ferrite-containing EAF-dust is partially reduced to zinc oxide and magnetite with a CO/CO2 gas mixture at the relative lower temperature of 750 C: 3ZnFe2O4 þ CO / 3ZnO þ 2Fe3O4 þ CO2
(1)
The roasted product is then leached with ammonium carbonate solution to recover zinc followed by electrolysis (Jha et al., 2001; Ruiz et al., 2007); the zinc recovery is about 80e85%. Dutra et al. (2006) and Ruiz et al. (2007) obtained similar results by means of acidic or alkaline leaching. Reduction roasting has not been commercialized because of its low zinc recovery rate. In addition, the magnetite-containing leaching residues, which have a zinc content of about 10%, cannot be charged into the conventional steel manufacturing process. The most crucial conditions of pyrometallurgical processes are the reducing agent and temperature; a temperature exceeding the sintering point of iron-containing slag would result in difficulty to separate iron from slag. The use of coke as reducing agent and fuel causes an oxidative condition in fumes and re-oxidation of zinc vapor. Using a reducing agent such as CO gas instead could solve these problems (Best and Pickles, 2001; Pickles, 2007). Besides, the Boudouard reaction (Eq. (2)) should also be noted: 2CO 4 CO2 þ C DH ¼ 172.58 kJ
(2)
Because of the strong exothermal reaction and the change of gas volume, the Boudouard reaction depends strongly on temperature. The volume of CO2 increases with decreasing temperature and is accompanied by the formation of fine carbon or graphite particle. In accordance with the reaction, a reduction gas, such as CO or CO/CO2 mixture, could lose its reduction effect at low temperatures so that the reduction could only be partially achieved. The phase diagram of the Boudouard reaction is shown in Fig. S1. The direct reduction with CO gas instead of coke has the following advantages: -
-
Reduction at lower temperature is possible because of the extremely low oxygen partial pressure. It not only reduces the energy consumption but also avoids the sintering of ironcontaining residue that could lead to congestion in the furnace. The tail gas remains reductive with a CO-surplus; thus the recovery of metallic zinc is possible.
The aim of this study is to develop a new process for the recovery of zinc and iron from EAF-dust. In order to reduce synthetic zinc ferrite and EAF-dust, CO was used as reducing agent at different temperatures in this experimental research work. Reduction in relation to dependence on temperature was estimated
209
from the phase studies. All reduction solid residues with synthetic zinc ferrite were analyzed by X-ray power diffraction (XRD). Additionally, an estimation of mass balance is presented. 2. Materials and methods 2.1. Materials and preparations EAF-dust from an electro steel mill in southern Germany and synthetic zinc ferrite were used. For the preparation of the synthetic zinc ferrite, zinc oxide and hematite of reagent grade were first thoroughly mixed with molar ratio of 1:1 and pressed to tablets. The tablets were then sintered in a tube furnace with a quartz tube as the main reaction chamber at 750 C in Ar gas flow for 4 h (Akhtar et al., 2014). The schema of the furnace device is shown in Fig. S2. The formation of zinc ferrite can be described as follows: ZnO þ Fe2O3 / ZnFe2O4
(3)
After 4 h of sintering followed by quenching, the tablets were ground, remixed, and repressed to tablets. The second sintering was carried out under the same conditions. After 3 repeated procedures aimed at obtaining the best possible purification and crystallization, the product was kept for later determination and experiments. The EAF-dust sample was ground and then dried at 105 C for 12 h and kept for later analysis and experiments. 2.2. Reduction experiments The reduction experiments were carried out in the same furnace device shown in Fig. S2. Instead of Ar, CO was introduced into the reaction zone as reduction gas. At the exit, cooling water was used to quench the tail gas and condense the vapor. Quartz wool was used as filter to remove the condensed dust. The furnace was first preheated to the given temperatures of 450 C, 550 C, 650 C, 750 C, 850 C, and 950 C. Samples packed in a quartz boat were then sent into the hot zone. A constant CO gas flow rate of 833 ml/min and flow velocity of 0.19 m/sec were chosen. After 2 h of reaction, the quartz tube was shifted so that the boat was outside the furnace and could be quenched by cooling water. Both the solid residue and dust were ground and collected for later analysis. 2.3. Characteristics of materials All samples and products were characterized with X-ray diffraction (XRD, Philips X’Pert). The chemical compositions were quantitatively analyzed using flame atomic adsorption (Perkine Elmer, AA-100). The content of Fe2þ was determined by the titration method. The chemical composition of the single ferrite phase was also calculated. The measured d-spacing (d in A) by XRD was used to calculate the lattice constant (a in A) of different ferrite phases in accordance with Bragg’s equation (Eq. (4)):
l ¼ 2d$sin q
(4)
Schaefer/McCune and Popov/Ilinova came to the conclusion that the lattice constant of spinel phases ZnxFe3exO4 (with 0 x 1) shows a linear dependence on the zinc content represented by the parameter x (Popov and Ilinova, 1970; Schaefer and McCune, 1986). A change of slope from x ¼ 0.7 was also reported (Popov and Ilinova, 1970). On average, the lattice constant a of the ferrite phase lies between 8.392 A for magnetite Fe3O4 (x ¼ 0) and 8.441 A for franklinite
210
C.-C. Wu et al. / Journal of Environmental Management 143 (2014) 208e213
ZnFe2O4 (x ¼ 1); the following equations (5) and (6), gained from investigations with different synthetic zinc ferrites in previous research, were used for these calculations (according to Wu, 2003):
Table 1 Chemical compositions of the solid residues after the reduction of zinc ferrite by CO gas at various temperatures. Temperature ( C)
x ¼ 17:6991a 148:6142 for 0 x 0:7
(5)
x ¼ 30:7692a 258:8646 for 0:7 < x 1
(6)
All experiments were prepared and measured in triplicate samples to evaluate the reproducibility of the results.
450 550 650 750 850 950
Content (%)
Zinc evaporation rate
Zn
Fe
C
29.2 22.2 10.9 14.3 0.12 0.09
51.4 35.9 19.56 33.9 80.36 90.78
4.35 36.3 55.4 44.2 6.35 4.16
N.D. N.D. 4.80 27.93 99.79 99.86
3. Results and discussion 3.1. Characteristics of synthetic zinc ferrite Fig. S3 shows the XRD pattern of synthetic zinc ferrite. No other phases of zinceiron-oxides were identified in the product after 20 h of sintering at 750 C in Ar gas flow. According to equations (5) and (6), the calculated lattice constant of synthetic zinc ferrite is 8.4425 0.0001 A, which indicated the parameter x to be 0.9998. Moreover, the content of Fe2þ was under the detection limit. 3.2. Characteristics of EAF-dust The ground EAF-dust was of fine particle with an average size of 12.17 mm. The chemical composition is listed in Table S1 along with the average values from literature for comparison (Hagni and Hagni, 1994; Ruiz et al., 2007; Oustadakis et al., 2010). In addition to zinc and iron, the dust from carbon steel production also contained lead, manganese, alkali metals, chlorine and some trace elements such as Cd (0.08%), Cr (0.27%), Cu (0.46%) and Ni (0.09%). In comparison with the values reported in previous literature, the sample can be classified as high zinc-containing EAF-dust. Fig. S4 shows the XRD pattern of the EAF-dust sample. Zinc oxide and franklinite were identified as the main compounds. Judged by the 2q values and the calculation using Bragg’s equation, the lattice constant of the zinc ferrite phase was 8.4394 0.0018 A. This indicates a chemical formula of Zn0.8Fe2.2O4 and a Fe2þ/Fe3þ ratio of 0.05 according to equations (5) and (6). However, the Fe2þ/ Fe3þ ratio was determined to be 0.036 by chemical analysis. The difference in the ratio could be the consequence of the existence of other heavy metals such as manganese and nickel in the ferrite phase. The existence of manganese ferrite (Jacobsite, MnFe2O4) in particular, which has a similar XRD pattern as franklinite in EADdust, has often been reported (Machado et al., 2006; Holloway et al., 2007). Lead exists mainly as basic lead chloride PbCl2,Pb(OH)2 (Pickles, 2009), also known as laurionite which is a weathering product of lead slag under the influence of sea water. The other chlorides found were sodium chloride (NaCl) and potassium chloride (KCl). 3.3. Reduction of synthetic franklinite with CO gas Different grades of zinc vaporization were observed during the reduction of synthetic franklinite by CO gas at various temperatures. The compositions and the vaporization grade of zinc are listed in Table 1. The XRD patterns of solid residues are shown in Fig. 1. In general, the reduction of zinc ferrite and the recovery of zinc as metal dust increase with rising temperatures. This can be well explained by the loss of reduction effect of CO gas at low temperatures and the change of the reaction free energy as a function of temperature. At 450 C, the reduction effect of CO gas was so weak due to the low temperature and shifting of the Boudouard equilibrium to the side of CO2 that franklinite was only partially reduced. In addition
to metallic iron, the solid residue contained zinc ferrite and wustite (FeO). The calculation of the lattice constant and zinc content indicate the chemical formula of Zn0.7Fe2.3O4 for the remaining zinc ferrite phase. Based on the BaureGlaessnereDiagram, the coexistence of metallic iron and zinc ferrite can be explained by the direct reduction of magnetite to metallic iron at a temperature lower than 600 C (Varanda et al., 2002). The reduction effect of CO gas increased slightly at 550 C so that the bound zinc in ferrite was fully liberated to zinc oxide while iron was reduced to metallic iron and wustite. At 650 C, iron oxides were completely reduced to metallic iron. The phase identification shows the existence of iron carbide (cementite, Fe3C) and crystalline graphite after reduction at lower temperatures. The formation of iron carbide is the result of a series of reactions as follows: 2CO 4 C (graphite) þ CO2
(7)
C þ 3Fe / Fe3C(cementite)
(8)
As a consequence, the DRI was converted to cementite at lower temperatures (550 C). The reaction gas lost its reducing effect and the metal oxides were only partially reduced. The metallization grade for zinc, calculated by the evaporation rate, was less than 5%. With the increasing temperature, the vaporization of zinc rose dramatically. One of the reasons is that the Boudouard equilibrium shifted to the side of CO which maintained the reduction effect of the gas. The other reason is the decreasing free energy of the reduction of zinc ferrite and zinc oxide with increasing temperature. The carbon content in the iron-containing solid residue decreased gradually to less than 5%. At 950 C, the collected zinc dust contained nearly 100% of metallic zinc (99.9%). The results show that the temperature affects not only the reduction effect of CO gas but also the reaction free energy. The reactions at various temperatures based on the phase and elemental analysis are summarized in Table 2. It should be noted that zinc oxide and wustite do not exist in the pure phase form. Some authors have concluded that zinc oxide and iron(II) oxide dissolve in each other and form solid solutions at high temperatures (>800 C). The solubility of zinc oxide in wustite ranges from 14% to 23% while the solubility of wustite in zinc oxide ranges between 5% and 19.36% inside the temperature range of 750 Ce1000 C (Bates et al., 1966; Claude et al., 1978; Itoh and Azakami, 1995). Such oxides are known as zinc-substituted wustite and iron-substituted zinc oxide. However, the ratios of these two phases can only be estimated very roughly. The phase status for the reduction of zinc ferrite in relation to dependence on temperature can be estimated as shown in Fig. 2. In summary of the results and discussions above, the optimal reduction temperature to recover iron and zinc from franklinite with CO gas is 950 C.
C.-C. Wu et al. / Journal of Environmental Management 143 (2014) 208e213
1
1 Iron Fe 3 Zincite ZnO 5 Graphite C
950oC
5
22 2
2
211
2 Cohenite Fe3C 4 Wustite FeO 6 Franklinite ZnFe2O4 1
1
2 CO2
CO 850oC 1 5
22 2
3
1
1
2
750oC 1 5
3
3
3
3
3
22
2
3 2
3
3 1
3 3
1
3
3
33 2
1
3
3
33
1
3
1
Intensity
650oC
5
3
1 2 22 2 2 2 3 2
2
1
550oC
5
3
3
4 3 2 4
4
450oC
6 3
20
2 21 2 3
3 36
1 2
2
3
2 3
40
63 1
60
2 Theta
80
Fig. 1. XRD patterns of reduction solid residues with synthetic zinc ferrite at various temperatures.
Reduction of pelletized EAF-dust with CO gas at 950 C was carried out. The chemical compositions and the phase statuses of solid residue (the iron fraction) and dust collected in off-gas treatment devices (the zinc fraction) are listed in Table 3. Similar to the reduction experiment of synthetic franklinite at 950 C, the iron fraction contained mainly metallic iron and cementite Fe3C. The carbon content increased slightly from 4.16% to Table 2 Reduction reactions of zinc ferrite by CO gas at various temperatures. Temp. ( C)
Reactions
450e550
ZnFe2O4 þ 3CO 4 ZnO þ 2Fe þ 3CO2 ZnFe2O4 þ CO 4 ZnO þ 2FeO þ CO2 (3 x) ZnFe2O4 þ (1 x) CO 4 (3 3x) ZnO þ 2ZnxFe3xO4 þ (1 x) CO2 ZnFe2O4 þ 3CO 4 ZnO þ 2Fe þ 3CO2 ZnFe2O4 þ CO 4 ZnO þ 2FeO þ CO2 ZnFe2O4 þ 4CO 4 Zn(g) þ 2Fe þ 4CO2 ZnFe2O4 þ 3CO 4 ZnO þ 2Fe þ 3CO2 ZnFe2O4 þ 4CO 4 Zn(g) þ 2Fe þ 4CO2
550e650 650e950 >950
4.59% so that the austenite phase, a solid solution of g-iron and cementite, was also found. The reason for this can be explained by the evaporation of the chlorides, such as NaCl and KCl, which consumed a certain part of energy so that the energy utilization for 1 zinc oxide + iron substituted zinc oxide
zinc Mole of Metals
3.4. Reduction of EAF-dust with CO gas
metallic zinc vapor
0.08 0 spinel 0.24 zinc substituted wustite iron
1
2 450
metallic iron + cementite
550
650
750 Temp. ( oC)
850
950
Fig. 2. Phase diagram of zinc ferrite reduction in relation to dependence on temperature.
212
C.-C. Wu et al. / Journal of Environmental Management 143 (2014) 208e213
Table 3 Chemical compositions and the phase statuses of the solid residues and the collected dust from the reduction of EAF-dust by CO gas at 950 C. Element
Content (%)
Zn Fe Pb K Na Cl C Mn Phases
Solid residue
Zinc dust
0.065 89.9 0.257 0.193 0.180 0.012 4.59 4.50 Met. Fe, austenite Fe3C, graphite
80.3 0.068 9.51 1.93 1.23 7.32 0.49 n. d. Met. Zn, ZnO, met. Pb, PbO, KPb2Cl5, NaCl, KCl
the reduction itself was actually lower in comparison with the previous experiments. The purity of iron fraction, however, remained at 90%. The collected zinc dust, on the other hand, contained both metallic zinc and zinc oxide. One of the reasons is the reduction of energy utilization as described above. The content of zinc decreased to 80%. Lead existed as metals, oxides and chlorides (KPb2Cl5) in the collected dust. The concentrations of Na and K in the dust were 1.23% and 1.93%, respectively. The recovery ratio of zinc and lead were both over 99.8%. These results show that the reduction effect of CO gas at 950 C is capable of recovering zinc not only from the synthetic franklinite but also from EAF-dust containing franklinite and zinc oxide. The mass balance at 950 C is shown in Fig. 3. The recovery ratios of zinc and iron were 99.9% and 99.8%, respectively. The metallization grade was approximately 95% for iron and 90% for zinc. 4. Conclusions The investigation for recovering zinc and iron from EAF-dust in terms of a pyrometallurgical process using CO gas as reducing agent at various temperatures is described in this paper. The lower temperature led not only to the incomplete reduction of oxides but also the formation of carbide. Thus the purity of iron was reduced moderately. The optimal working temperature was determined to be 950 C. The reduction experiment with EAF-dust also demonstrated that this procedure was only slightly negatively influenced by the existence of other impurities such as alkali chlorides and lead compounds. After the treatment, two main products were obtained, namely direct reduced iron (DRI) and metallic zinc dust. With a purity of over 90% and an estimated metallization grade of over 95%, the iron sponge can be charged directly into the EAF process without any pretreatment. The zinc dust, on the other hand, can be used as raw material in the zinc refinery industry after certain treatments to remove chlorides, for example by distillation. EAF-Dust (g) 32.98 23.00 1.23
Zinc fraction (g) 32.96 1.20
→
4.10
1.27
1.15
0.55
Legend (g) Fraction
Reduction o at 950 C
CO gas
3.06
Zn
Fe
Pb
K
Na
Cl
Mn
C
Iron fraction (g)
0.03 4.033
1.26
0.017 22.97 0.066 0.008
3.02
2.01
0.029 0.003
0.01
1.14
1.17
Fig. 3. Mass balance of the process to recover zinc and iron from EAF dust.
The mass balance shows that this process is characterized by zero waste. Different from the conventional pyrometallurgical processes with solid coke, CO gas is suggested as reductant and indirect heating agent in this new process. With relative lower temperature, DRI can be yielded. Instead of reaction air that causes oxidation of zinc vapor, the CO gas guarantees a high metallization grade of zinc. In comparison with combined technologies such as the Caron-ZincProcess with CO/CO2 mixture, the reduction rate of zinc ferrite is raised to 100% so that the recovery ratio of zinc is also improved. Acknowledgments The financial supports by the Deutsche Forschungsgemeinschaft, Germany (project number Go 383/20-1) and the Ministry of Economic Affairs (project number 95-EC-17-A-10S1-0007) are gratefully acknowledged. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jenvman.2014.04.005. References Akhtar, M.N., Yahya, N., Sattar, A., Ahmad, M., Idrees, M., Asif, M.H., Khan, M.A., 2014. Investigations of structural and magnetic properties of nanostructured Ni0.5 þ xZn0.5xFe2O4 magnetic feeders for CSEM application. Int. J. Appl. Ceram. Technol. http://dx.doi.org/10.1111/ijac.12222. Bates, C.H., White, W.B., Roy, R., 1966. The solubility of transition metal oxides in zinc oxide and the reflectance spectra of Mn2þ and Fe2þ in tetrahedral fields. J. Inorg. Nucl. Chem. 28, 397e405. Best, T.E., Pickles, C.A., 2001. In-flight plasma reduction of electric arc furnace dust in carbon monoxide. Can. Metall. Q. 40, 61e78. Chen, W.S., Shen, Y.H., Tsai, M.S., Chang, F.C., 2011. Removal of chloride from electric arc furnace dust. J. Hazard. Mater. 190, 639e644. Claude, J.M., Zanne, M., Gleitzer, C., Aubry, J., 1978. Preparation and study of ZnxFe0.85xO (0.085 x 0.17). J. Solid State Chem. 24, 395e400. Dutra, A.J.B., Paiva, P.R.P., Tavares, L.M., 2006. Alkaline leaching of zinc from electric arc furnace steel dust. Min. Eng. 19, 478e485. Galchinetskii, L.P., Galkin, S.N., Ryzhikov, V.D., Rybalka, I.A., Voronkin, E.F., Lalayants, A.I., Starzhinskii, N.G., Silin, V.I., 2005. Chemical interaction in the ZnSe-ZnTe-Se-H2-C system. Inorg. Mater. 41, 934e938. Hagni, A.M., Hagni, R.D., 1994. Significance of Mineralogy of Electric Arc Furnace (EAF) Dust in Pyrometallurgical Treatment to Render the Dust Non-Hazardous. Extraction and Processing for the Treatment and Minimization of Wastes. TMSAIME, Warrendale, PA, pp. 1137e1148. Harvey, T.G., 2006. The hydrometallurgical extraction of zinc by ammonium carbonate: a review of the schnabel process. Min. Process. Extr. Metall. Rev. 27, 231e279. Havlık, T., Souza, B.V., Bernardes, A.M., Schneider, I.A.H., Miskufova, A., 2006. Hydrometallurgical processing of carbon steel EAF dust. J. Hazard. Mater. 135, 311e 318. Holloway, P.C., Etsell, T.H., Murland, A.L., 2007. Use of secondary additives to control the dissolution of iron during Na2CO3 roasting of La oroya zinc ferrite. Metall. Mater. Trans. B 38B, 793e808. Huaiwei, Z., Xin, H., 2011. An overview for the utilisation of wastes from stainless steel industries. Resour. Conserv. Recy 55, 745e754. Huda, N., Naser, J., Brooks, G., Reuter, M.A., Matusewicz, R.W., 2011. Computational fluid dynamic modeling of zinc slag fuming process in top-submerged lance smelting furnace. Metall. Mater. Trans. B 179H, 1e17. Itoh, S., Azakami, T., 1995. Thermodynamic study on the zinc-iron-oxygen system at 1200K. Mater. Trans. JIM 36, 1074e1080. Jha, M.K., Kumar, V., Singh, R.J., 2001. Review of hydrometallurgical recovery of zinc from industrial wastes. Resour. Conserv. Recycl 33, 1e22. Li, C.L., Tsai, M.S., 1993. A crystal phase study of zinc hydroxide chloride in electricarc-furnace dust. J. Mater. Sci. 28, 4562e4570. Machado, J.G.M.S., Brehm, F.A., Moraes, C.A.M., dos Santos, C.A., Vilela, A.C.F., da Cunha, J.B.M., 2006. Chemical, physical, structural and morphological characterization of the electric arc furnace dust. J. Hazard. Mater. 136, 953e960. Machado, A.T., Valenzuela-Diaz, F.R., de Souza, C.A.C., de Andrade Lima, L.R.P., 2011. Structural ceramics made with clay and steel dust pollutants. Appl. Clay Sci. 51, 503e506. Nyirenda, R.L., 1990. An appraisal of the caron zinc process when zinc ferrite is reduced to a magnetite containing product. Min. Eng. 3, 319e329. Oustadakis, P., Tsakiridis, P.E., Katsiapi, A., Agatzini-Leonardou, S., 2010. Hydrometallurgical process for zinc recovery from electric arc furnace dust (EAFD)
C.-C. Wu et al. / Journal of Environmental Management 143 (2014) 208e213 part I: characterization and leaching by diluted sulphuric acid. J. Hazard. Mater. 179, 1e7. Pickles, C.A., 2007. Thermodynamic analysis of the selective reduction of electric arc furnace dust by carbon monoxide. High. Temp. Mater. Process 26, 79e91. Pickles, C.A., 2008. Thermodynamic analysis of the separation of zinc and lead from electric arc furnace dust by selective reduction with metallic iron. Sep. Purif. Technol. 59, 115e128. Pickles, C.A., 2009. Thermodynamic analysis of the selective chlorination of electric arc furnace dust. J. Hazard. Mater. 166, 1030e1042. Popov, G.P., Ilinova, G.N., 1970. Deviation of certain ferrite solid solutions from Vegard’s law. Russ. J. Phys. Chem. 45, 299e301. Quijorna, N., de Pedro, M., Romero, M., Andrés, A., 2014. Characterisation of the sintering behaviour of Waelz slag from electric arc furnace (EAF) dust recycling for use in the clay ceramics industry. J. Environ. Manage. 132, 278e286. Rao, S.R., 2006. Resource Recovery and Recycling from Metallurgical Wastes. Elsevier Waste Management series v. 7, Amsterdam, pp. 269e327. Ruiz, O., Clemente, C., Alonso, M., Alguacil, F.J., 2007. Recycling of an electric arc furnace flue dust to obtain high grade ZnO. J. Hazard. Mater. 141, 33e36.
213
Salihoglu, G., Pinarli, V., Salihoglu, N.K., Kar, G., 2007. Properties of steel foundry electric arc furnace dust solidified/stabilized with Portland cement. J. Environ. Manage. 85, 190e197. Schaefer, S.C., McCune, R.A., 1986. Electrochemical determination of thermodynamic properties and an X-ray diffraction investigation of the Fe3O4-ZnFe2O4 system. Metall. Mater. Trans. B 17, 515e521. Suetens, T., Klaasen, B., Acker, K.V., Blanpain, B., 2014. Comparison of electric arc furnace dust treatment technologies using exergy efficiency. J. Clean. Product. 65, 152e167. Varanda, L.C., Jafelicci, M., Tartaj, P., O’Grady, K., González-Carreño, T., Morales, M.P., Muñoz, T., Serna, C.J., 2002. Structural and magnetic transformation of monodispersed iron oxide particles in a reducing atmosphere. J. Appl. Phys. 92, 2079e2085. Wu, C.C., 2003. Phase Investigation in the Reduction of Franklinite for the Zinc Extraction from Steel Plant. Ph.D. thesis. Institute for Treatment and Disposal Technology, Technical University of Clausthal, Clausthal-Z., Germany.
