Extraction and characterization of alumina nanopowders from ...

International Journal of Minerals, Metallurgy and Materials Volume 22, Number 4, April 2015, Page 429 DOI: 10.1007/s12613-015-1090-2

Extraction and characterization of alumina nanopowders from aluminum dross by acid dissolution process Md. Saifur Rahman Sarker1), Md. Zahangir Alam1), Md. Rakibul Qadir2), M A Gafur2), and Mohammad Moniruzzaman3) 1) Department of Applied Chemistry and Chemical Engineering University of Dhaka, Dhaka 1000, Bangladesh 2) Pilot Plant & Project Development Centre (PP& PDC), Bangladesh Council of Scientific and Industrial Research (BCSIR), Qudrat-I-Khuda Road, Dhanmandi, Dhaka 1205, Bangladesh 3) BCSIR Laboratories, Dhaka, Qudrat-I-Khuda Road, Dhanmandi, Dhaka 1205, Bangladesh (Received: 2 July 2014; revised: 23 September 2014; accepted: 25 September 2014)

Abstract: A significant amount of aluminum dross is available as a waste in foundry industries in Bangladesh. In this study, alumina was extracted from aluminum dross collected from two foundry industries situated in Dhamrai and Manikgang, near the capital city, Dhaka. Aluminum dross samples were found to approximately contain 75wt% Al2O3 and 12wt% SiO2. An acid dissolution process was used to recover the alumina value from the dross. The effects of various parameters, e.g., temperature, acid concentration, and leaching time, on the extraction of alumina were studied to optimize the dissolution process. First, Al(OH)3 was produced in the form of a gel. Calcination of the Al(OH)3 gel at 1000°C, 1200°C, and 1400°C for 2 h produced θ-Al2O3, (α+θ)-Al2O3, and α-alumina powder, respectively. Thermal characterization of the Al(OH)3 gel was performed by thermogravimetric/differential thermal analysis (TG/DTA) and differential scanning calorimetry (DSC). The phases and crystallite size of the alumina were determined by X-ray diffraction analysis. The dimensions of the alumina were found to be on the nano level. The chemical compositions of the aluminum dross and alumina were determined by X-ray fluorescence (XRF) spectroscopy. The microstructure and morphology of the alumina were studied with scanning electron microscopy. The purity of the alumina extracted in this study was found to be 99.0%. Thus, it is expected that the obtained alumina powders can be potentially utilized as biomaterials. Keywords: aluminum dross; alumina; nanoparticles; acid dissolution; calcination; X-ray fluorescence analysis

1. Introduction Extraction of alumina has been of interest for several of its inherent properties, such as relatively high hardness, high abrasion resistance, and bioinertness [1−4]. Currently most of the world’s commercial alumina is produced by the Bayer process using bauxite material [5]. Acidic and alkaline routes have also been reported for extraction of alumina. Acids such as H2SO4, H2SO3, and HNO3 have been used as leaching agents [6−7]. In this study HCl was used as the leaching agent because of its several advantages; it is mild in its action, easily available, and produced as a byproduct in many chemical industries.

In Bangladesh bauxite is not available, but a significant amount of aluminum dross is obtained as a waste in its foundry industries. Aluminum dross is a complex oxide material containing Al2O3, SiO2, Na2O, Fe2O3, and other compounds. It is formed when molten aluminum comes in contact with air at the outer surface of the melt. It is a hazardous waste material [8]; however, it contains a high percentage of aluminum oxide, so it may be an alternative source of alumina. Recently an increasing number of alternative methods to the Bayer process have been developed in countries lacking bauxite deposits [9−10]. Namely, alkaline and acid processes have been proposed to produce alumina from aluminous ores where aluminum is dissolved from the ore as the aluminate ion ( AlO −2 ). The Bayer process is quite

Corresponding author: Md. Zahangir Alam, E-mail: [email protected]; M A Gafur, [email protected] © University of Science and Technology Beijing and Springer-Verlag Berlin Heidelberg 2015

430

selective for aluminum, as iron is almost insoluble in alkaline solutions, but when treating aluminosilicates, silica can only be removed at the expense of extracted aluminum. On the other hand, when aluminum is taken into an aqueous solution as Al3+, silica remains substantially insoluble in the acid routes but the leaching operation is not selective for aluminum, as iron, titanium, potassium, sodium, magnesium, and calcium are generally co-dissolved. However, iron removal is usually required to achieve smelter-grade alumina [11−13]. The low iron content and high silica content of aluminum ash clearly indicate that, technically, an acid process could be more attractive than an alkaline process, as reported in a study on the economics of clay treatments [14]. Recently, the kinetics of alumina recovery from aluminum casting wastes and fly ash through fusion with sodium hydroxide was reported [15−16]. The present study investigates the extraction of alumina from aluminum dross by acid dissolution process using HCl as a leachant. It is expected that the alumina powder produced in this study will be at the nano level.

Int. J. Miner. Metall. Mater., Vol. 22, No. 4, Apr. 2015

was added to the solution for precipitation. Purification of the Al(OH)3 was performed by adding 5% NaOH solution and reprecipitating using 10% NH4Cl solution. The leached solutions were diluted and analyzed with an atomic absorption spectrophotometer (AA−7000F, Shimadzu). A schematic flow diagram showing the production of alumina from aluminum dross is also shown in Fig. 1, and possible chemical reactions that can occur during production of alumina are shown as follows: Al2O3 + 6HCl = 2AlCl3 + 3H2O; 2Al + 6HCl = 2AlCl3 + 3H2↑; Fe2O3 + 6HCl = 2FeCl3 + 3H2O; CaCO3 + 2HCl = CaCl2 + CO2 + H2O; AlCl3 + 3NH4OH = Al(OH)3 + 3NH4Cl; Al(OH)3 + NaOH = NaAlO2 + 2H2O; NaAlO2 + H2O + NH4Cl = Al(OH)3 + NaCl + NH3; 2Al(OH)3 = Al2O3 + 3H2O.

2. Experimental 2.1. Collection and processing of aluminum dross Aluminum dross samples were collected from two aluminum industries in Manikganj and Dhamrai. Two samples collected from Manikganj were fine and coarse in size and were designated as samples A and B, respectively. Another sample, collected from Dhamrai, was coarse and was denoted as sample C. Collected samples were filtered through a set of Taylor-series sieves with diameters of 4 mm, 2.5 mm, 1 mm, 500 μm, 250 μm, 125 μm, 63 μm, and 45 μm. The aluminum dross samples were blackish in color with a moisture content of about 3.5%. A planetary ball mill (Retsch, Germany) was used to grind the coarse aluminum ash sample to increase the surface area for reaction. The grinding time was 2 h and the speed was 300 r/min. Aluminum dross and extracted alumina powders were analyzed by XRF spectroscopy (Rigaku ZSX Primus) to determine their compositions. 2.2. Extraction of alumina A total of 25 g of ground aluminum dross was taken in a flat-bottomed glass reactor placed on a hot-plate magnetic stirrer (HS 33) and was leached using 500 mL of 1−6 mol/L HCl for different periods of time (15 to 240 min) at different leaching temperatures ranging from 298 to 373 K. Then, it was filtered to remove undissolved materials. After separating the solution from the solid residue, 25% NH3 solution

Fig.1. Flow diagram of producing alumina from aluminum dross.

2.3. Characterization of alumina powders XRD analyses were conducted using a D8Advance (BRUKER) diffractometer to identify the phases present in the powdered samples as well as to determine the crystallite size. The pattern was recorded with Cu Kα radiation (λ = 0.154 nm) with accelerating voltage of 40 kV, and scanned in the 2θ = 10° to 70° range with a scanning speed of 2°/min (0.02°/2θ step). The diffraction pattern was analyzed and the peaks were identified according to the powder diffraction files of the JCPDS-International Centre for Diffraction data. The crystallite sizes of the alumina were determined according to the following equation [17]. d = 0.9λ/(Bcosθ), where d is the crystallite size, nm; λ is the wavelength of X-ray (0.15406 nm); B is the full width at half maximum, rad; θ represents the X-ray peak position, (°) . Thermal gravimetric analysis (TGA) was performed using an Exstar 6000 TG/DTA 6300 instrument. The dry powder

Md. Saifur Rahman Sarker et al., Extraction and characterization of alumina nanopowders from aluminum dross by …

was heated up to 1000°C at a rate of 20°C⋅min−1, using pure alumina as reference. The phase transition behavior of aluminum hydroxide was investigated using a DSC (DSC−60) at a scanning rate of 10°C⋅min−1. The morphology of the alumina powder was studied using SEM (JSM 6490).

3. Results and discussion 3.1. Size analysis of aluminum dross The sizes of aluminum dross samples were analyzed us-

431

ing a set of Taylor-series sieves and the results are tabulated in Table 1. From Table 1 it is observed that sample A contains the highest percent (33.60%) of 125 μm ≥ d > 63 μm particle size, sample B contains the highest percent (48.09%) of 1 mm ≥ d > 500 μm particle size, and sample C contains the highest percent (41.01%) of 2.5 mm ≥ d > 1 mm particle size of the raw sample. After grinding, Samples B and C contained 38.75% and 37.90% of 250 μm ≥ d > 125 μm particle size, respectively.

Table 1. Sieve analysis results of raw and ground aluminum ash samples

wt%

Before grinding

Particle size

Sample A

Sample B

After grinding Sample C

Sample B

Sample C

4 mm ≥ d > 2.5 mm

0.00

4.72

5.98

2.40

2.46

2.5 mm ≥ d > 1 mm

0.06

30.09

41.01

5.54

7.03

1 mm ≥ d > 500 μm

5.51

48.09

36.05

11.95

12.24

500 μm ≥ d > 250 μm

25.64

8.61

7.51

12.99

12.13

250 μm ≥ d > 125 μm

23.55

4.33

4.16

38.75

37.90

125 μm ≥ d > 63 μm

33.60

3.53

4.12

25.32

24.97

63 μm ≥ d > 45 μm

10.03

0.51

1.06

2.50

2.45

45 μm ≥ d

1.71

0.07

0.09

0.65

0.82

3.2. Compositional analysis of aluminum dross The XRF spectroscopic data given in Table 2 reveal that the compositions of all samples are very similar and that alumina could be extracted from all the samples. However,

sample A contains the lowest percentage of alumina and its silica content is the highest. The alumina and silica contents of samples B and C were almost identical (about 75wt% alumina and 12wt% silica). Therefore, in this study samples B and C were selected for extraction of alumina.

Table 2. Chemical composition of aluminum ash samples Sample

Na2O

MgO

Al2O3

SiO2

CaO

wt% Fe2O3

Cl

A

2.51

2.77

72.85

13.23

0.76

2.31

3.06

B

2.29

2.03

75.25

11.83

0.51

1.76

3.97

C

1.95

1.54

75.19

11.90

0.78

2.44

3.97

3.3. Effects of various parameters on alumina extraction 3.3.1. Effect of leaching time on alumina extraction Leaching of aluminum dross was performed for a period of 15 to 240 min. Fig. 2 shows the effect of leaching time on alumina extraction for different acid concentrations. Extraction of alumina increased with time. The initial rate of extraction was higher, and as time elapsed it decreased and became almost constant after 1 h. 3.3.2. Effect of acid concentration on alumina extraction Fig. 2 also indicates the percentage of alumina extracted as a function of acid concentration. Extraction of alumina was found to increase with an increase in acid concentration up to 4 mol/L HCl. A maximum of 51% alumina was ex-

tracted by leaching aluminum dross with 4 mol/L HCl, and then the percentage of alumina extracted decreased. It is assumed that the diffusion rates of Al3+ ions from the solid to the solution increase with an increase in concentration and diffusion of hydronium ions. It is also assumed that Al3+ species block H+ diffusion at a concentration of HCl at and above 6 mol/L. This phenomenon was earlier reported by several research groups [18−19]. 3.3.3. Effect of leaching temperature The results of the effects of acid concentration on the percent extraction of alumina reveal that 4 mol/L was the optimum concentration of HCl for the leaching process of the aluminum dross samples. Therefore, the acid concentration was selected to be 4 mol/L for determining the optimal

432


temperature. Using 4 mol/L HCl at different temperatures (298, 323, 348, and 373 K), the leaching reaction was conducted. The results of the effect of varying temperatures on the dissolution of alumina, illustrated in Fig. 3, indicate that the percent extraction of aluminum oxide increases largely with increasing leaching temperature. For 298, 323, 348, and 373 K, the highest percent extraction was found to be about 51%, 55%, 62%, and 71%, respectively. The maximum percent extraction (about 71%) was obtained at 373 K.

by XRD analyses. Fig. 4 shows the XRD pattern of the aluminum dross while Figs. 5(a), 5(b), 5(c), and 5(d) show diffraction patterns of extracted aluminum hydroxide dried at 100°C and alumina samples obtained by calcination at 1000, 1200, and 1400°C, respectively. The figures reveal the identified phases as well as their corresponding planes. XRD analysis of the aluminum dross confirmed the presence of mainly aluminum (ICDD 004−0787), alumina (ICDD 046−1212), and quartz (ICDD 046−1045) as shown in Table 3.

Fig. 2. Effects of acid concentration and leaching time on alumina extraction from aluminum dross at 298 K.

3.4. Characterization of aluminum dross and extracted alumina The aluminum dross and alumina were also characterized

Fig. 3. Effects of temperature and leaching time on extraction of alumina from aluminum dross using 4 mol/L HCl.

Fig. 4. XRD pattern of the aluminum dross. Table 3. Phases present in aluminum ash samples Name of the phase Aluminum Corundum (aluminum oxide) Quartz (silicon oxide)

Chemical Formula Crystal system Al

Face-centered cubic

α-Al2O3

Rhombohedral

SiO2

Hexagonal

Fig. 5 shows the XRD pattern of extracted alumina powders at different temperatures. The phases, crystal systems and crystallite sizes of produced aluminum hydroxide and alumina sample are presented in Table 4. A poor crystalline gel of aluminum hydroxide was obtained by heating at 100°C for 24 h, which is confirmed by the low intensity of some peaks in the XRD spectra. The observed phases were


bayerite (ICDD 077−0250) and gibbsite (ICDD076−1782). Calcination of aluminum hydroxide at 1000°C for 2 h (Fig. 5(b)) led to the formation of θ-alumina (ICDD 023−1009) while a mixture of α-alumina (ICDD 046−12123) and θ-alumina (ICDD 023−1009) was obtained on calcination at 1200°C. From Fig. 5(d) it is clear that upon calcination at

433

1400°C, 100% α-Al2O3 was obtained. This is the most stable form of alumina and is our desired product, which can potentially be utilized as a biomaterial. As observed, the crystallite sizes of alumina were in the nano level and were found to increase with an increase in calcination temperature, as reported by Rogojan et al. [1].

Fig. 5. XRD patterns of aluminum hydroxide powders dried at 100°C (a) and alumina obtained by calcination of aluminum hydroxide powders at 1000°C (b), 1200°C (c), and 1400°C (d).

3.5. Compositional analysis of extracted alumina The composition of extracted alumina obtained by XRF

analysis is given in Table 5. It is observed that the extracted alumina was 99.0% pure. Other minor constituents found in the alumina were SiO2, Fe2O3, K2O, CaO, and PbO.

434

Int. J. Miner. Metall. Mater., Vol. 22, No. 4, Apr. 2015 Table 4. Phases present of produced aluminum hydroxide and alumina samples Temperature / °C

Phase name

Chemical formula

Bayerite

α-Al(OH)3

Hexagonal

ND

Gibbsite

γ-Al(OH)3

Monoclinic

ND

Theta alumina

θ-Al2O3

Monoclinic

9.48

Alpha alumina (73.0%)

α-Al2O3

Rhombohedral

24.60

Theta alumina (27.0%)

θ-Al2O3

Monoclinic

14.37

Alpha alumina

α-Al2O3

Rhombohedral

30.83

100 1000 1200 1400

Crystal system

Crystallite size / nm

Note: ND―not detected Table 5. Composition of extracted alumina

wt%

Al2O3

SiO2

Fe2O3

K2O

CaO

PbO

99.00

0.52

0.27

0.06

0.08

0.07

3.6. Thermal analysis Fig. 6 shows the TG, DTA, and differential thermogravimetric (DTG) curves of extracted aluminum hydroxide powders dried at 100°C for 24 h. The TG curve shows only 0.1% initial mass loss due to loss of moisture. After 257°C the mass loss increased rapidly up to 284°C and then decreased. Above 311°C the mass loss decreased slowly. The mass loss became constant at 683°C. The total mass loss

was 31%, which shows a good agreement with the stoichiometry of the dehydration of aluminum hydroxide. The DTA curve shows three endothermic peaks at 242°C, 293°C, and 548°C, respectively. The first and second can be attributed to hydroxyl loss from the decomposition of aluminum hydroxide powders and the third may be attributed to the transformation of polymorphous enantiotropic transformation. The DTG curve shows that the maximum decomposition rate (0.571 mg/min) was obtained at 288°C. The maximum slope of TG (284°C), second peak of DTA (293°C), and maximum decomposition rate temperature (288°C) are closely related to the decomposition of aluminum hydroxide powders.

Fig. 6. Thermal analysis curves of aluminum hydroxide powders obtained by the acid dissolution process from aluminum dross, dried at 100°C for 24 h.

Thermal phase transition behaviors of Al(OH)3 were studied from 30°C to 600°C using DSC. Fig. 7 shows the phase transition behavior of the extracted aluminum hydroxide (a mixture of bayerite and gibbsite) obtained by DSC. As can be observed from Fig. 7, the DSC thermogram of Al(OH)3 exhibited three significant endothermic peaks. The first endothermic peak, at 109°C, was due to moisture

content in the sample; the second, at 283°C, may have been due to decomposition of aluminum hydroxide to γ (gamma)-Al2O3; and the third, at 418°C, corresponded to the phase transition of γ (gamma)-Al2O3 to δ (delta)-Al2O3. Other phases of alumina, namely θ-Al2O3 and α-Al2O3, obtained at 1000°C and 1400°C, respectively, were confirmed by XRD analysis.


435

Fig. 7. DSC thermogram of aluminum hydroxide.

3.7. SEM analysis Information on morphological and textural characteristics of the powders obtained by the acid dissolution process from the aluminum dross was gathered using SEM analysis. The SEM images of aluminum hydroxide and alumina (heat treated at 1000°C and 1200°C) are given in Fig. 8. The SEM image of aluminum hydroxide shows that the major component of the powders consisted of a porous, spongy plate and irregular shapes in different sizes, ranging between 5 and 15 µm (Fig. 8(a)). Fine particles with colloidal dimen-

sion show a strong tendency towards agglomeration, being kept close together. It is obviously revealed that the agglomerate consists of very fine particles that can be converted to obtain fine alumina. The SEM images of alumina obtained by the acid dissolution process from the aluminum dross and heat treated at 1000°C for 2 h reveal a fine particulate matter. It can be observed in Fig. 8(b) that the powders form aggregates of different geometries with rounded edges. The particles have irregular outlines and are agglomerates of small round particles, presenting a parallel orientation of strings of particles observed in thinner particles. In

Fig. 8. SEM images of aluminum hydroxide (a), alumina obtained by calcination at 1000°C (b), and alumina obtained by calcination at 1200°C (c).

436


the scanning micrographs, the particles appear to be packets of layers. The SEM images of alumina powders obtained by heat treating at 1200°C show similar morphology to that of alumina heat treated at 1000°C, with observations that the particles are more homogenous and their dimensions are more reduced (Fig. 8(c)).

4. Conclusion Alumina was successfully recovered from aluminum dross without any pretreatment by leaching in hydrochloric acid, a process that is more attractive than the Bayer process for extraction of alumina from high-silica raw materials. The resulting alumina was characterized by XRF, XRD, TG-DTA, DSC, and SEM. Extraction of alumina was found to increase with an increase in acid concentration, leaching time, and temperature. A maximum 71% of alumina was extracted from aluminum dross at optimum conditions of 4 mol/L HCl, 120 min leaching time, and a temperature of 100°C. XRD analyses of the dry gel revealed that a mixture of bayerite and gibbsite was obtained. θ-alumina, a mixture of (α+θ)-alumina, and α-alumina were obtained by heat treatment of the dry gel at 1000°C, 1200°C, and 1400°C, respectively. All forms of alumina extracted in this study were in the nano scale, as observed by XRD analysis. The purity of the extracted alumina (99.0%) makes it a potential candidate as a biomaterial for various applications. Moreover, extraction of alumina from aluminum dross also reduces environmental hazards produced by foundry industries in Bangladesh.

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

Acknowledgements The fellowship from the University Grants Commission of Bangladesh is gratefully acknowledged to perform the research. The authors would also like to thank BCSIR for providing the facilities for this research.

[14]

[15]

References [1]

[2]

[3]

R. Rogojan, E. Andronescu, C. Ghiţulică, and B.S. Vasile, Synthesis and characterization of alumina nano-powder obtained by sol-gel method, UPB Sci. Bull. B, 73(2011), No. 2, p. 67. R. Cordingley, L. Kohan, B. Ben-Nissan, and G. Pezzoti, Alumina as an orthopaedic biomaterial: characteristics, properties, performance and applications, J. Australas. Ceram. Soc., 39(2003), No. 1 p. 20. T.R. Yoon, SM. Rowe, S.T. Jung, K.J. Seon, and W.J. Maloney, Osteolysis in association with a total hip arthroplasty with ceramic bearing surfaces, J. Bone Joint Surg. Am., 80(1998), No. 10, p. 1459.

[16]

[17] [18]

[19]

T.V. Thamaraiselvi and S. Rajeswari, Biological evaluation of bioceramic materials: a review, Trends Biomater. Artif. Organs, 18(2004), No. 1, p. 9. A. Adamson, E. Bloore, and A. Carr, Essential Readings in Light Metals, Alumina and Bauxite, Edited by D. Donaldson and B.E. Raahauge, WILEY, Hoboken, New Jersey, 2013, p.100. W.H. Tingle, J. Petit, and W.B. Frank, Handbook of Extractive Metallurgy, Edited by F. Habashi, WILEY-VCH, Weinheim, Germany, 1997, p. 1039. A.D. Kelmers, R.M. Canon, and B.Z. Egan, Chemistry of the direct acid leach, calsinter, and pressure digestion-acid leach methods for the recovery of alumina from fly ash, Resour. Conserv., 9(1982), p.271. E. David and J. Kopac, Aluminum recovery as a product with high added value using aluminum hazardous waste, J. Hazard. Mater., 261(2013), p. 316. D. Panias, C. Skoufadis, and I. Paspaliaris, The boehmite process: an innovative variation of the Bayer process, [in] Proceedings of EMC, Friedrichshafen, 2001, p. 371. B. Dash, B.R. Das, B.C. Tripathy, I.N. Bhattacharya, and S.C. Das, Acid dissolution of alumina from waste aluminium dross, Hydrometallurgy, 92(2008), No. 1-2, p. 48. F.A. Peters, P.W. Johnson, and R.C. Kirby, Methods for producing alumina from clay: an evaluation of five hydrochloric acid process, Investigation Report of U.S. Bureau of Mines, 1962, No. 6133, p. 68. C.V. Phillips and K.J. Wills, A laboratory study of the extraction of alumina of smelter grade from China clay micaceous residues by a nitric acid route, Hydrometallurgy, 9(1982), No. 1, p. 15. P. Mahi and N.T. Bailey, The use of coal spoils as feed materials for alumina recovery by acid-leaching routes: 4. The extraction of iron from aluminiferous solutions with amines, in particular alamine 336, Hydrometallurgy, 13(1985), No. 3, p. 293. F.A. Peters and P.W. Johnson, Revised and Updated Cost Estimates for Producing Alumina from Domestic Raw Materials, Information Circular (8648), U.S. Department of the Interior, Wahington DC, 1974. U.K. Sultana, F. Gulshan, M.A. Gafur, and A.W.S. Kurny, Kinetics of recovery of alumina from aluminium casting waste through fusion with sodium hydroxide, Am. J. Mater. Eng. Technol., 1(2013), No. 3, p. 30. M.A. Rahaman, M.A. Gafur, and A.S.W. Kurny, Kinetics of recovery of alumina from coal fly ash through fusion with sodium hydroxide, Am. J. Mater. Eng. Technol., 1(2013), No. 3, p. 54. B.D. Cullity, Elements of X-Ray Diffraction, Edition 2, Addison Wesley Publishers, Reading, Massachussets, 1978, p. 398. A.A. Al-Zahrani and M.H. Abdul-Majid, Extraction of alumina from local clays by hydrochloric acid process, JKAU Eng. Sci., 20(2009), No. 2, p. 29. J.A. Eisele, D.J. Bauer, and D.E. Shanks, Bench-scale studies to recover alumina from clay by a hydrochloric acid process, Ind. Eng. Chem. Prod. Res. Dev. 22(1983), No.1, p. 105.