DESCRIPTION OF THE RELATIONSHIP BETWEEN PHYSICO- MECHANICAL PROPERTIES AND SIZE REDUCTION PROCESSING OF ROCKS

and standard test procedures such as ISRM (1981), TS699 (1976). For example, Bery et al. (1984) has performed a variety standard strength test on a laboratory type jaw crusher in order to relate its performance with strength of rocks. Moreover, Tang et al, (2001) have studied the particle breakage based on result of the indirect tensile strength and also discussed the relationship between energy consumption and rock strength properties. The purpose of this study is to discuss the affect of rock properties on the size reduction from a physico -mechanics point of view. Therefore, 10 types of rocks having different Internat ional Scient ific Confer ence of Moder n Management of Mine Pr oducing, Geology and Environment al Pr ot ect ion SGEM 200 6 6th International Multidisciplinary Scientific GeoConference SGEM2006 www.sgem.org Int er nat ional Confer ence SGEM 200 6 188 formations have been used in this study. All the physical, mechanical, and size reduction tests were performed in laboratory. The relationships between physico - mechanical properties of rocks and size reduction process (crushing, grinding) were described by using the statistical correlation analyses.

MATERIALS AND METHODS In this study, marble, granite, travertine and andesite samples having different petrographical, physical and mechanical properties were used.

1 Petrographical analysis methods Petrographic features of samples were analyzed by a petrography microscope and their detailed evaluations are given in Table 1. As seen in Table 1, four different types of rocks were used. Petrographical properties and mineralogical contents have important differences. Table 1. Description of petrographical and mineralogical properties of specimens (Unal and Kekec, 2006) Rock Types Petrographical Description Mineral Content Granite AG Generally, gray and dark gray coloured quartz, plagioclases feldspars are large enough to be seen by eyes. The biotites and amphiboles give the dark colour to the rock. The rock texture is a holocrystalline texture and hypidi omorph texture due to its crystallization level and grain condition respectively. There are observed oxidization and chloritization in biotites, sericitization and mirmecitic texture, albite twins zoned texture and semi pare -shaped grains in plagioclases, widespread carslbat twin and impute shaped pertitic texture grains in orthoclases and epidotization in amphiboles. Quartz (29%), plagioclase (31,5%), orthoclase (21%), amphibole (1,5%), biotite (7%), opac (3%), epidote (3%), muscovite (1,5%), apatite (1,5% ), sericite (1%) Magmatic rocks Intrusive rocks Granite KR Gray, greenish gray and pinkish coloured quartz, plagioclase, orthoclase, amphibole and biotites can be obzerved macroscopically. The rock texture is a holocrystalline texture and porphyric texture due to crystallization cond itions and grain condition respectively. There were seen sericitization and carbonization in plagioclases, oxidization and chloritization in biotites, pertitic texture in orthoclases. In a singular phase in the rock, there are met extremely coarse - grained orthoclase and plagioclase phenocrystals belonging to the same phase. Therefore, the same of the rock is defined as granite porphyry. Quartz (22%), plagioclase (34%), orthoclase (18%), amphibole (12%), biotite (10%), Opac (1%), Microclin (3%) SGEM 200 6 - Section III 189 Granite KC Gray and ligth gray coloured quartz, alkali feldspar, plagioclase, amphibole and biotites can be seen by eyes. However, there exists extremely widespread orthoclase in the rock. In the rock texture, there is observed holocrystalline texture and grainy texture dpending on the crystallization condition and grain condition respectively. Oxidization and chloritization in biotites, pertitic textures in orthoclase take place. The name of the rock is given as Syenite - Alkali Syenite. Quartz (3%), plagioclase (5%), orthoclase (75%), amphibole (6%), biotite (4%), opac (2%), epidote (2%), spfen (1%), muscovite (3%) Andesite A The plagioclase and amphibole phenocrystals having grayish, pink coloured bollowed appearance inside the rock can be seen by eyes. There are observed holocrystalline texture due to the crystallization level, porphyric texture dependent on grain condition, trachitic texture and holocrystalline porphyric texture due to the paste phase of the rock. In amphiboles, there exist apatit ization and chloritization. Microlite takes place inside the rock with an 80% percent. Plagioclase microlites (80%), opac (5%), amphibole (6%), plagioclase (9%) Volcanic rocks Dasite S Light pink, beige coloured quartz biotite and plagioclase phenocrystals are coarse enough to be seen by eyes inside the rock. There are observed hypocrystalline texture due to crystallization level, porphyric texture and vitrophyric porphyric texture dependent on the grain condition. Oxidation in biotite and amphiboles, zoning and sieve texture in plagioclase are seen. Volcanic glass inclusions are also seen in plagioclases. With the re-crystallization of large amount of volcanic glass by hydrothermal solitions, there are formed secondary quartzes in micron size. Plagioclase (35%),amphib ole (8%), biotite (12%), quartz (10%), volcanic glass (35%) Metamorphic rocks Marble A The general name of the rock is pelagic limestone. There are observed pelagic fossils inside the rock. The matrix is completely formed from micrite. The skeletal grains are observed in the micrite in sparitic. Here and there silicification are formed inside these sparitic infills. Sometimes chratization is seen widespread. In addition, secondery veins are taking place inside the rock. The vein infills are formed from spari-calcites. The fosil particles are generally composed of aragonite. However, during diagenesis the aragonite deteriorates and its place in filled with spari-calcite. Skeletal grains (38%), micrite (60%), Opac (2%) 6th International Multidisciplinary Scientific GeoConference SGEM2006 www.sgem.org Int er nat ional Confer ence SGEM 200 6 190 Beige Y The minerals in the cross -section are generally pure- unshaped. There were observed rhombohedra - shaped.dolomite minerals and zoning. The rock texture is xenotopical mosaic texture. The rock is named as dolomite. Dolomite (100 %) Beige B Lagune is a limestone precipi tated in its enviroment. During precipitation while the medium is motionless, these occurs channels on the ground with mini -flows formed from the changes in precipitation conditions and then ooid, intraclast and bioclast materials formed by the flows are precipitated inside these channels and connected with a microsparitical cement. Afterwards the development of secondery cracks due to tectonic movements and their infill with calcite were observed. The rock is named as oo-intra - bio micrite or Vache Stone. Ooid (2%), intraclast (13%), skeletal grains (25%), micrite (50%), sparite (10%) Travertine M The rock precipitated in a turbulent and shallower medium. The role of alges in the formation of the rock is great (for ex; they asist to the formation of micri ted calcites by holding the little carbonate pieces). It is observed that the empty sapaces formed from the deterioration of the organic part inside the rock are filled with sparitic calcite. The name of the rock is given as biointrasparite. Ooid (6%), intraclast (24%), skeletal grains (24%), sparite (34%), porosity (12%) Sedimentary rocks Lymra There are sparitic carbonete grains and micritic grains between 0,2-0,4 mm in the rock. Dolomite is observed as bond material. Depending on these characteristics of the rock is named as Dlomitic limestone. Intraclast (70%), skeletal grains (3%), dolomite (22%), Organic materials (5%)

2 Physical and mechanical tests Physical and mechanical properties of the rock samples used in this work were defined by the tests complying with the ISRM standards (1981). For the rock samples used hardness, porosity, density, uniaxial compressive strength and point loading strength were determined and results of these tests are in Table 2. SGEM 200 6 - Section III 191 Table 2. Results of the physical and mechanica l tests (Kekec, 2005). g: Mineral grain density P: Porosity d: Hardness values  c: Uniaxial compressive strength  t: Indirect tensile strength Is50: The point load strength

3 Crushing and grinding experiments Crushing and grinding degree of rock samples was defined by the type of breakage, sieve analysis and shape measurements. Sample blocks and pieces which are selected to represent the sample rocks were broken into pieces to the dimensions of (-64 + 32 mm) with a hummer in order to feed the crusher. Crushing works were achieved by a laboratory type, Sigmoid, a V shape jaw crusher with an outlet gap of 2 cm in a dry basis while grinding conditions of the experiment were carried out in a laboratory ball mill complying with a standard S and B technique (Austin et al., 1984). The experiments have been performed at a ball load of

% of the mill volume filled with the ball bed and a powder load corresponding to a formal interstitia l filling of the void spaces of the ball bed 0.5. In the grinding tests, a laboratory size ceramic ball mill was used. The samples -1,0+0,5 mm feed size fraction of 200 g were dry ground in the mill at a grinding time of 30 min. For this purpose, graphs formed to ascertain the relationship between particle size and weight percentage of classified samples and cumulative quantity passing under the sieve were used to calculate the sieve size (d50) through which 50 % of the crushed samples passed. The grindabil ity rate of the samples was determined by calculating the Rock types g (g/cm 3) P (%) d  c (MPa)  t (MPa) Is50 (MPa) Granite - 2,67 5,62 99,10 ± 6,18 166,713 ± 10,657 4,041 ± 0,248 8,040 ± 1,460 Granite - 2,81 6,76 89,30 ± 3,51 82,572 ± 6,772 3,114 ± 0,127 8,120 ± 0,710 Granite - 2,58 1,16 85,20 ± 5,75 91,594 ± 9,258 3,156 ± 0,158 9,778 ± 0,750 Andesite - 2,61 20,69 76,60 ± 5,40 75,118 ± 7,215 2,165 ± 0,235 5,360 ± 0,610 Dasite -S 2,60 13,08 66,80 ± 5,29 29,224 ± 8,210 0,481 ± 0,101 1,435 ± 0,236 Marble -A 2,73 1,83 59,90 ± 2,18 73,844 ± 7,321 1,417 ± 0,230 3,755 ± 0,485 Beige -Y 2,88 3,47 70,20 ± 2,72 46,385 ± 6,123 2,124 ± 0,421 5,793 ± 0,679 Beige -B 2,71 1,48 80,50 ± 1,83 55,310 ± 5,356 2,491 ± 0,322 6,168 ± 1,011 Travertine - 2,68 12,31 77,10 ± 3,80 61,193 ± 8,224 1,455 ± 0,153 4,958 ± 0,586 Lymra 2,69 11,90 58,10 ± 4,32 50,602 ± 4,263 1,182 ± 0,241 1,482 ± 0,223 6th International Multidisciplinary Scientific GeoConference SGEM2006 www.sgem.org Int er nat ional Confer ence SGEM 200 6 192 percentage of the samples passing under 212. The results obtained from these experiments are summarized in Table 3. Table 3. Grinding and crushing values of rocks (Kekec, 2005). Rock types d50 g (%) Granite - AG 20,946 38,36 Granite - KR 16,216 40,31 Granite - KC 19,324 39,25 Marble -A 19,527 96,78 Travertine -M 16,860 89,98 Andesite - A 18,986 45,59 Dasite -S 14,392 65,92 Beige -Y 18,514 89,36 Beige -B 16,892 38,12 Lymra 18,243 100 d50: Sieve s paces where the 50 % of broken samples were passed (mm) g : The degree of grindability (%)

ANALYSES AND EVALUATIONS To define relationships among these (physical, mechanical, crushing and grinding) parameters, the approach of correlation coefficient was used. These relations were given in Figure 1-2, as the graph of relations. In this evaluation, type of rocks was not considered. It was considered as a whole instead. Associations among these parameters were appraised in three stages. The relation havi ng a correlation coefficient over 0.7 were accepted as a good and if the correlation coefficient between 0.7 and 0.6 were also accepted as possible while they were counted as probable when around 0.5. Crushing and grinding works were applied to three different types of rocks such as magmatic, metamorphic and sedimentary rocks. Correlation coefficients for the physical and mechanical parameters effective on definition of crushing and grinding features were described in Figure 1 and Figure 2.

1. Descriptio n of relation between physical and communition parameters A relation between the shore hardness of rock samples and d50 values that determine with crushing characteristics of rocks has not determined (Figure 1b), although we can mention that there is a relation between hardness and the grindability (Figure 1a). When graphs of the relation between the mineral grain density and crushing, grindability properties (Figure 1c and Figure 1d) are examined, it is determined that the mineral grain density is not an effective parameter on crushing and SGEM 200 6 - Section III 193 grinding properties. In the same way it is not determined that there is an evident relation when we compare the grindability, crushing values and porosity values (Figure 1e and Figure 1f). (a) (b) 6th International Multidisciplinary Scientific GeoConference SGEM2006 www.sgem.org Int er nat ional Confer ence SGEM 200 6 194 (c) (d) SGEM 200 6 - Section III 195 Figure 1. The relation graphs between physical and communition parameters

2. Description of relation between mechanical and communition parameters It is possible to mention that there is a relation when we examine the relations between grindab ility, d50 values and indirect tensile strength (Figure 2a and Figure 2b). In the same way, it is observed that there is also a similar relation between uniaxial compressive strength and d50 values, grindability (Figure 2c and Figure 2d). When the (e) (f) 6th International Multidisciplinary Scientific GeoConference SGEM2006 www.sgem.org Int er nat ional Confer ence SGEM 200 6 196 point load strength and grindability properties of used rock are examined, a relation is shown (Figure 2e). However there is not an effective relation between the point load strength and d50 values of rocks (Figure 2f). (a) (b) SGEM 200 6 - Section III 197 (c) (d) 6th International Multidisciplinary Scientific GeoConference SGEM2006 www.sgem.org Int er nat ional Confer ence SGEM 200 6 198 Figure 2. The relation graphs between mechanical properties and grinding, crushing properties of sample rocks. (e) (f) SGEM 200 6 - Section III 199

CONCLUSIONS When an evaluation of works carried out is made with the help of quotations subtracted from literature, it can be concluded there is a relation between physico - mechanical properties and communition process of the rocks. Results obtained from present data and evaluations were:  It is determined that there is an inverse proportion between hardness values and percentage grindability of rocks.  It is observed that there is an inverse proportion between percentage grindability and some mechanical properties like the point load strength , uniaxial compressive strength and indirect tensile (Brazilian) strength.  It is possible to mention that there is a direct proportion between mechanical properties and d50 values which determines the properties of crushing.  It is observed that some physical features like density, hardness, porosity are not effective parameters on the properties of crushing and grinding.  It is thought that mechanical properties are more effective parameters than physical properties.  There are many parameters affecting crushing and grinding properties. However, it seems impossible to define their effectiveness due to their heterogeneous structure. ACKNOWLEDGEMENT This work was financially supported by the Office of Scientific Research (Bu çalışma Selçuk Üniversitesi Bilimsel Araştırma Projeleri (BAP) tarafından desteklenmiştir). Project number: 2003/122; Selcuk University, Turkey.

REFERENCES Austin L. G., Klimpel R. R. & Luckie P. T., 1984, Process Engineering of Size Reduction: Ball Milling, New York, USA

Bearman, R.A., Barley, R.W. & Hitchcock, A., 1991, Prediction of Power Consumption and Product Size in Cone Crushing, Minerals Eng., 4,No. 12, 1243 -1256

Bearman, R.A., Briggs, C.A. and Kojovıc, T., 1997, The Applıcatıon Of Rock Mech anıcs Parameters To The Predıctıon Of Commınutıon Behavıour Minerals

Engineering, Vol. 10, No. 3, Pp. 255-264

Berry, P., Dantini, E.M. & Massacci, P., 1984, Influence of Mechanical Characteristics of Rock on Size Reduction Processing, Proc. Mineral Proces sing and Extractive

Metallurgy, Beijing, IMM, 15-26

Briggs, C.A. & Bearman, R.A., 1996, An Investigation of Rock Breakage and Damage in Comminution Equipment, Minerals Eng., 9,No. 5,489 -497 D.J. Gates And M. Westcott, 2000, Generatıng A Rock Breakage Pro cess From Sieve

Data, Minerals Engineering, Vol. 13, No. 14-15, Pp. 1603 -1614, 2000 ISRM, 1981, Rock Characterization Testing and Monitoring, Suggesting Methods, Oxford 6th International Multidisciplinary Scientific GeoConference SGEM2006 www.sgem.org Int er nat ional Confer ence SGEM 200 6 200

Kekec, B., 2005, Investigation of the texture, physical and mechanical properties of the rocks used as building stone, MSc Thesis, Selcuk University, Graduate School of Natural and Applied Sciences, Konya, Turkey

Tang, C.A., Xu, X.H., Kou, S.Q., Lindqvist, P.A., Liu, H.Y., 2001, Numerical investigation of particle breakage as applied to mechanical crushing -Part I: Single - particle breakage, Int. J. Of Rock Mech., 1147 -1162

Tanga, C.A., Liua, H.Y., Zhua, W.C., Yanga, T.H., Lib, W.H., Songb, L., Linc, P., 2004, Numerical approach to particle breakage under different loading conditions, Powder Technology 130–143 TS 699, 1987 “Methods of testing for natural building stone”, Institute of Turks

Standarts, Ankara.

Unal, M., Kekec, B., 2006, Investigation of the relation between texture and mechanical properties of rocks, Research Project (2003/122) , Selcuk University, Office of Scientific Research, Konya, Turkey 6th International Multidisciplinary Scientific GeoConference SGEM2006 www.sgem.org