Ceramic Processing Research

Journal of Ceramic Processing Research. Vol. 13, No. 4, pp. 470~475 (2012) J O U R N A L O F Ceramic Processing Research The influence of process parameters on the structure an properties of micamechanically activate in an ultra centrifugal mill Ljubiša Anricº a, *, Zagorka Acºimovicº-Pavlovicº b, Milan Trumicº c an Milena Kostovicº a Institute for Technology of Nuclear an Other Mineral Raw Materials 86 Franchet Esperey Street, 11000 Belgrae, Serbia b University of Belgrae, Faculty of Technology an Metallurgy, Karnegijeva 4, 11120 Belgrae, Serbia c University of Belgrae, Technical Faculty in Bor, Vojske Jugoslavije12, 19210 Bor, Serbia University of Belgrae, Faculty of Mining an Geology, Ðušina 7, 11000 Belgrae, Serbia This paper presents the results of research on mechanical activation of ry mica in an ultra centrifugal mill with a peripheral comminution path. The variable parameters of the mill operation were: rate of rotor revolution (n o = 10.000 an n o = 20.000 r.p.m.), circle sieve mesh (80, 120, 200 an 500 µm) an the current intensity. The following technological parameters were stuie: mechanical activation time, rotor velocity, mill capacity an specific energy consumption. The mechanically activate power was examine by application of thermal an thermogravimetric analyses, analysis of the egree of mechanical activation an the specific surface area, as well as X-ray iffraction analysis. The optimal results of the mechanical activation were obtaine with a full mill loa, using a circle sieve mesh of 80 µm an nominal rate of rotor revolution of 20.000 r.p.m. It was shown that the mechanically activate mica obtaine employing these process parameters ha passe into the amorphous state. Key wors: Mica, Amorphous materials, Ceramics, Thermogravimetric analysis (TGA), Crystal structure. Introuction The quality requirements of mechanically activate mica are high for its application in the synthesis of contemporary materials. Due to their physical properties, micas have significant importance both, in conventional applications such as in plastics, pearlescent pigments, optical filters, stove winows, conensers, rheostats, fuses, insulators, rilling mus, asorbents, fire extinguishers, concretes, etc. an in nanotechnology, where they are extensively use as a substrate for eposition or selfassembly, as templates in the preparation of nanowires an in clay polymer nanocomposites [1]. Mica glass-ceramics are wiely use as mechanical, electrical, an biomeical materials ue to their unique machinability an goo electrical properties. By using a novel hot-pressing technique, it is possible for a new type of mica glass-ceramics to be prepare where the microstructure consists of preferentially aligne mica particlesg[2]. For many applications, it is necessary to reuce the particle size of the natural raw materials. An alternative route for ry mechanical activation of minerals of mica using knife-mills was foun. Comminution by cutting was shown as the most efficient technique to prouce *Corresponing author: Tel : +381-11-3691-722 Fax: +381-11-3691-583 E-mail: lj.anric@itnms.ac.rs material with a size below 100 mm without morphological or structural alterations [3]. Grining, either in the ry state or in the presence of water an chemical aitives, is the common proceure use for particle size reuction of these materials. Different grining proceures are use in inustry for micas: ry grining (yieling particles in the range from 1.2 mm to 150 m), wet grining (95-45 µm), or mechanical activation (< 53 µm) [4]. It is well known that grining not only reuces the particle size but also has various effects on the structure an laminar silicates properties, such as amorphization, aggregation or moification of the surface properties that are unesirable in general. In aition, particle size reuction of clay an mica minerals by a high-energy ball milling (HEBM) technique has been wiely stuie. Particle size reuction of clay an mica by a HEBM technique has a strong influence on their surface physical-chemical characteristics, i.e., the specific surface area (SSA), the cation exchange capacity (CEC), an the electrokinetic properties. Significant changes of the size, morphology an structure were followe by the change of the physical-chemical properties [5]. The mica power shoul consist of a fine particle size istribution, with strictly efine physical-mechanical, physical-chemical an mineralogical characteristics. Every eviation from the require characteristics exacerbates the structure an properties of mechanically activate proucts. Uner fine an ultra-fine grining, 470

The influence of process parameters on the structure an properties of micamechanically activate in an ultra centrifugal mill 471 the mechano-chemical alterations in the structure of the particles of the material occur initially in the surface layer. The rate of alteration in the initial phases of mechanical activation is etermine by the rate of increase in the specific surface area, which epens on the physical properties of the material an the possibilities of the apparatus employe for mechanical activation to efficiently transfer a major amount of the mechanical energy onto the material [6-9]. The process of mechanical activation of mineral raw materials, incluing micas, is not a simple process an has been stuie wiely uring the last few ecaes. Two methos of mechanical activation (ball milling an oscillatory milling) have been stuie intensively. Mechanical activation using a ball mill was foun to be a slow process where ry grining prove to have some effect compare to wet grining. In aition, mechanical activation in an oscillatory mill coul prouce almost complete estruction. These finings referre to clay minerals [10-14]. The results of the kinetics an mechanisms of the mechanical activation process of these minerals in a centrifugal mill presente in this paper represent a contribution to the investigation an etermination of the occurrences an processes occurring in the material uner the effects of mechanical forces uring ispersion. The changes in the structure an properties, as well as of the energy state of the materials, were stuie. Variations in the parameters of the grining process were realize with the goal of optimization an automation of the grining processes in a centrifugal mill. The results obtaine serve as a base for etermining the correlation of the parameters of the grining technology with the structure an esire properties of the resulting mica powers, as well as for forecasting their application in the synthesis of contemporary materials. Experimental The investigations of the mechanical activation of ry micas were realize in five series of experiments. As the initial material, a flotation mica concentrate obtaine by the technological processing of white granites from the be Samoljica, Bujanovac, Serbia, was use, the series esignate M 0. The grinings were realize in a centrifugal mill with a peripheral comminution path. The mill coul be operate at nominal rotor spees of n o = 10.000 an n o = 20.000 r.p.m. In aition, it was possible to choose a circle sieve of ifferent mesh sizes (80, 120, 200 an 500 µm) an vary the intensity of the current. The series performe with the mesh sizes 80, 120, 200 an 500 were esignate M1, M 2, M 3 an M 4, respectively. The process of kinetics was analyze through the changes in the particle size istribution, respectively the specific surface area, with the time of mechanical activation. The analyses of the mechanical activation proucts were realize using ifferent instrumental techniques. For etaile power characterization (etermination the particle size istribution, the mean iameter an specific surface area of the grains) a Coulter-Electronics-Coulter Multisizer, which possesses software for all important correction factors for etermination of the physical characteristics of powers, was use. To efine the thermal an thermogravimetric characteristics of the samples, a simultaneous thermal analyzer STA-409 EP was use. X-Ray structural analysis was utilize for the etermination an observation of the phase composition of the refractory rivers. For this Table 1. The characteristics of the initial mica sample (M 0 ) Particle size istribution Mineralogical composition Chemical composition Size class Mass M Cumu. Over. R Cumu. uner. D Content, Wt% Mineral Cont. Wt% Comp. (mm) (%) (%) (%) Size class 0.589 + 0.104 mm -0.833 + 0.589 00.10 0 0.10 100.00 SiO 2 56.60-0.589 + 0.417 04.40 04.50 099.90 K- muskovite 74.11 Al 2 O 3 24.50-0.417 + 0.295 22.50 27.00 095.50 CaO 00.10-0.295 + 0.208 29.00 56.00 073.00 MgO 00.60-0.208 + 0.147 23.00 79.00 044.00 Namuskovite 10.13 Na 2 O 01.90-0.147 + 0.104 10.50 89.50 021.00 K 2 O 10.59-0.104 + 0.074 06.00 95.50 010.50 Fe 2 O 3 01.00 Quarz 7.19-0.074 + 0.063 01.40 96.90 004.50 MnO 00.03-0.063 + 0.053 01.10 98.00 003.10 TiO 2 00.18-0.053 + 0.040 00.82 98.82 002.00 Felspar 3.74 P 00.40-0.040 + 0.000 0 1.18 100.00 001.18 S 0 0.40 Total 100.00 - - Total 95.17 I.L. 0 3.70

472 Ljubiša Anricº, Zagorka Acºimovicº-Pavlovicº, Milan Trumicº an Milena Kostovicº Table 2. Process parameters relate to the ultra centrifugal mill Series Test No. Parameters relate to the ultra centrifugal mill Rotor revolutions (r.p.m.) Actual rot. rev. (r.p.m.) Sieve mesh (µm) Intensity of current (A) M 1 M 12 20,000 19,314.29 080 2.80 M 11 10,000 10,824.26 080 2.30 M 2 M 22 20,000 20,626.48 120 2.50 M 21 10,000 13,360.13 120 1.60 M 3 M 32 20,000 22,213.92 200 3.20 M 31 10,000 14,435.96 200 1.40 M 4 M 41 10,000 11,738.24 500 2.80 M 41 10,000 11,738.24 500 2.80 Fig. 1. Diffractogram of the initial mica sample (M 0 ) Table 3. Technological parameters of mechanical activation Series Test No. Technological parameters of mechanical activation Process uration (min) Rotor spee (m/s) Mill capacity (kg/h) Specific energy consumption (kwh/t) M 1 M 12 12.00 100.24 0.375 1642.66 M 11 30.00 074.40 0.150 3080.00 M 2 M 22 06.00 114.70 0.500 0968.00 M 21 06.00 074.40 0.500 0616.00 M 3 M 32 06.00 114.70 0.500 0968.00 M 31 04.50 074.40 1.000 0461.77 M 4 M 42 04.00 114.70 0.750 0645.33 M 41 04.00 081.63 0.750 0352.00 purpose, an X-ray iffractometer PW-1710 was use. The results of the particle size istribution, chemical an minerological composition of the initial sample (M 0 ) are given in Table 1. During all experiments in all series, the process parameters of ry mechanical activation were observe, i.e., mechanical activation time (t), rotor spee (v), mill capacity (Q) an the specific energy consumption (e). For the characterization of the mechanical activation proucts, the following relevant parameters were stuie: ( 1 ) an ( 2 ) - the sieve mesh sizes through which the samples were passe through (µm); (R 1 ) an (R 2 )- the cumulative screen outs, (%); the ' parameter- epens on the particle size istribution of the sample, which gives information concerning the massiveness of a sample an represents the sieve mesh size where R = 36.79%; n- a irection coefficient which epens on the particle size istribution of the sample; 95 - the sieve mesh through which 95% of the mechanically activate prouct passes (µm); S 1 - theoretical specific surface area (m 2 /kg); S r - the real specific surface area (m 2 /kg). The results of the measurements of the process parameters relate to the mill operation an the parameters connecte to the Fig. 2. DTA an TGA curves of the initial mica sample (M 0 ) mechanical activation process are given in Tables 2 an 3, respectively. Results an iscussion The results of the X-ray an thermal analysis of the initial mica sample, series (M 0 ), are shown in Figs. 1 an 2, respectively. In accorance with the ata obtaine by chemical analysis (Table 1.), it can be seen in Fig. 1 that the initial sample (M 0 ) was crystalline mica (muscovite), which was also registere by the enothermic effect at 575 ο C in Fig. 2. In aition, in the same Figure, the TGA results show that the sample graually lost mass on heating to 1000 ο C ; the total mass loss being 3.47%. The values of the parameters measure to observe the mechanical activation process an evaluate the quality of the mechanical activation proucts in series M 1, M 2, M 3 an M 4 are given in Table 4. It was note that the mechanical activation rate, as the main characteristic of mechanical activation kinetics, increases with an increase in the circle sieve mesh from 80 to 500 µm an with an increase in the mill loa. The rate maximum is at the nominal mill loa an at the highest rate of mill rotor revolutions, 20.000 r.p.m. Accoring to the results of the particle size istribution

The influence of process parameters on the structure an properties of micamechanically activate in an ultra centrifugal mill Parameters of the mechanical activation proucts Parameters of the mechanical activation proucts Series Test No. δ 1 2 1 2 95 n /%/ /%/ M11 040.00 92.66 03.60 20.55 1.79 037.20 M1 M11 030.00 90.02 05.10 16.40 1.86 028.94 M21 063.00 95.70 04.00 27.47 1.80 049.25 M2 M22 050.00 94.00 0 21.99 1.84 038.93 M31 063.00 95.20 04.20 25.50 1.82 045.42 M3 M32 069.00 94.00 00.90 22.48 1.82 040.04 M41 147.00 99.43 04.00 79.15 1.86 139.29 M4 147.00 99.36 13.00 71.31 1.88 124.74 M42 473 Table 4. Fig. 3. R R Comparative DTA results for (M0), b) (M11), c) (M12) analysis (Table 4.), it can be observe that changing the circle sieve mesh from 80 to 500 µm an ecreasing the nominal rate of rotor revolution from 20.000 r.p.m. to 10.000 r.p.m. le to increase massiveness in the particles of the mechanically activate prouct. The finest prouct in terms of its massiveness was obtaine with a circle sieve mesh of 80 µm an a nominal rate of rotor revolution of 20.000 r.p.m. Therefore, it coul be conclue that with an increase in the circle sieve mesh, the mechanical activation rate, as the main characteristic of mechanical activation kinetics, also increases. Basically, the circle sieve mesh is not irectly connecte to mechanical activation kinetics, it is more relate to fineness; actually, the esire fineness of mechanical activation ictates which circle sieve an which mesh shoul be use. For a etaile structure an properties analysis of mechanically activate mica, samples obtaine in the ultra centrifugal mill with a sieve mesh of 80 µm at a nominal rate of mill rotor revolutions of 20.000 r.p.m. were use, respectively 10.000 r.p.m., samples of the series (M1). Comparative DTA an TGA results are shown in Figs. 3 an 4 for: a) the initial sample, (M0), b) a mica sample mechanically activate at a nominal Fig. 4. (M12) St /m2 kg 1 181.38 218.19 134.92 164.55 143.60 162.89 045.21 049.63 Sr /m2 kg 1 544.14 654.57 404.76 493.65 430.80 488.67 135.63 148.89 Comparative TGA results for samples (M0), b) (M11), c) rotor rate of 10.000 r.p.m., (M11) an c) a mica sample mechanically activate at a nominal rotor rate of 20.000 r.p.m. (M12). It can be seen from the comparative iagrams that, at the beginning of heating up to 200 οc, respectively 250 οc, a change in the baseline of the curves of all samples occurre, by even up to 20 µv. This change was similar for all three samples. A ifference in the behavior of the mechanically activate micas samples arose uring further heating, while there was no change in that of the initial sample. For the mechanically activate samples, it can be seen that the DTA curves slowly return towar the initial line, at a nominal revolution rate of the mill rotor of 20.000 r.p.m., sample (M12), as a consequence of amorphization of the mica. The X-ray analysis results of the mechanically activate mica samples of the series M1, obtaine in the mill with a circle sieve mesh of 80 µm an nominal rotor rate of 10.000 r.p.m., (M11), respectively the samples from the series (M12), at a nominal rotor rate 20.000 r.p.m., are shown in Figs. 5 an 6. The X-ray analyses (Figs. 5 an 6.) inicate that changes in the structure of the samples appeare as a consequence of mechano-activation in the mechanical

474 Ljubiša Anricº, Zagorka Acºimovicº-Pavlovicº, Milan Trumicº an Milena Kostovicº From the quantitative aspect, the process is relatively simple, bearing in min that the only process occurring is the transformation crystal phase amorphous phase. Conclusions Fig. 5. Diffractogram of sample M 11 Fig. 6. Diffractogram of sample M 12 activation process. They are noticeable both qualitatively an quantitatively. The qualitative alterations of the phases present in the initial sample (mica, quartz, felspar) functionally epen on the rate of rotor revolution an they are shown through the appearance of amorphous layers at the grains of the phases present. This amorphization process of the surface of the grains of the mechanically activate minerals epens also on their properties; hence for the mica (γ = 2.7 g/ cm 3 ), it is very rapi in interval of n = 0-10.000 r.p.m., while for the har minerals, quartz (γ = 7.0 g/ cm 3 ) an felspar (γ = 6.00 g/cm 3 ), the amorphization process is retare. This appearance can be explaine by the mechanical activation of the mica grains an the release the quartz an felspar inclusions in them, which naturally increase their concentration in the sample (M 11 ) at 10.000 r.p.m. In the case of the mechanically activate mica obtaine at a nominal rate of the mill rotor of 20.000 r.p.m., sample (M 12 ), the effects of the process effects were slightly ifferent. On the iffractogram of sample M 12, the intensities of characteristic reflections of quartz an felspar were reuce, which inicates the commencement of their amorphization also. Unlike in the sample M 11, amorphous materials of quartz an felspar composition were present in the sample M 12. Accoring to the results obtaine in this investigation of the kinetics of mechanical activation of mica in an ultra centrifugal mill, where the mechanical activation is realize firstly by impact uner conitions of high mill rotor spees, the best results were obtaine at full mill loa. The mechanical activation rate, as the main characteristic of mechanical activation kinetics, increase with increasing loa an rate of revolution of the centrifugal mill rotor. The finest prouct in terms of massiveness was obtaine with a circle sieve mesh of 80 µm an nominal rotor rate of 20.000 r.p.m. X-Ray an thermal analyses results showe that uner these conitions of mechanical activation, the transformation process of a crystal phase into an amorphous phase occurre. The results obtaine in this investigation of mechanical activation represent a contribution to the interpretation of appearances an processes arising uring mica mechanical activation in this type of mechanoactivator an may be useful for the introuction of these processes into practice. Acknowlegements These investigations were conucte uner the Project 33007 an 34006 fune by the Ministry of Eucation an Science of the Republic of Serbia. References 1. J.L. Pérez-Roríguez, A. Wiewióra, J. Drapala an L.A. Pérez-Maquea, Ultrason. Sonochem. 13 (2006) 61-66. 2. K. Cheng, J. Wan an K. Liang, Mater. Lett. 39 (1999) 350-359. 3. S.F. Santos, S.C.A. França an T. Ogasawara, Min. Sci. an Tech. (China), 21 (2011) 7-12. 4. S.G. Barlow an A.C. Manning, Brit. Cer. Trans. 3 (1999) 122-126. N. Vovicº, I. Jurina, S.D. Škapin an I. Soni, App. Clay. Sci. 48 (2010) 575-581. 5. S. Janjicº an P. Risticº, in Mineralogy (Naucºna Knjiga Press, 1997) p.g147. 6. S. Miloševicº, in Domacºe nemetalicºne mineralne sirovine za komericijalnu upotrebu (Institute for Technology of Nucelar an other Mineral Raw Materials in Belgrae Press 2005) p. 54. 7. M.M. Risticº, in Principles of Materials Science (Institute for Technology of Nucelar an other Mineral Raw Materials in Belgrae Press, 1995) p. 78. 8. Lj. Anriæ, in Liskuni-Priprema i Primena (Institute for Technology of Nuclear an other Mineral Raw Materials in Belgrae Press, 2006) p.g66. 9. G. Suraj, C.S.P. Iyer, S. Rugmini an M. Lalithambika,

The influence of process parameters on the structure an properties of micamechanically activate in an ultra centrifugal mill 475 App. Clay. Sci. 12 (1997) 111-117. 10. Y. Lee, H. Cho an H. Lee, Int. J. Miner. Process. 78 (2006) 250-259. 11. F. Del Rey-Perez-Caballero an G. Poncelet, J. of Microp. Mesop. Mater. 41 (2000) 169-175. 12. G. Bauet, V. Perrotel, A. Seron an M. Stellatelli, Power Technol. 105 (1999) 125-135. 13. H. Lee, H. Cho an J. Kwon, Power Technol. 198 (2010) 364-371.