Materials Processing Technology ELSEVIER Journal of Materials Processing Technology 65(1997)302-304 Short comunication Machinability and brittleness of glass-ceramics A.R. Boccaccini I School of Metallurg and Materials. The Unitersity of Birmingham. Birmingham B15 2TT. UK Received 2 September 1995 Industrial summary The relationship between the machinability and the brittleness of glass-ceramic materials is investigated. a brittleness index (B) given by the ratio of the hardness to the fracture toughness, is proposed as a parameter for estimating the machinability. This approach is confirmed by considering experimental data from the literature on turning operations of mica-containing glass-ceram ics. It is shown that machinability parameters, such as the slope of the log-log plot of the specific cutting energy versus the cutting rate, or the specific cutting energy at low cutting rates, are in good agreement with the brittleness indices for seven different glass-ceramics. In order for a glass-ceramic to be machinable, it was found that the brittleness index of the material should be lower than B= 4.3 Hm-2.0 1997 Elsevier Science S.A Keywords: Gilass-ceramics: Brittleness; Machinability 1. Introduction its accurate quantitative measurement is difficult [2, 3 Various parameters have been suggested as the 'mea- Glass-ceramics are polycrystalline materials produced surement of the machinability, depending on the test by the controlled crystallization of glass. They have ing conditions employed, including tool wear, surface already found several applications as engineering mate- roughness, cutting force, cutting energy, drilling rate rials and new uses constantly appear [1]. An important etc [2-5]. These parameters will depend strongly on the group of these materials is that of the mica-containing microstructure and properties of the glass-ceramic. In glass-ceramics, due to their high machinability, which particular, fracture strength, hardness and fracture results in an increased versatility of the products and toughness have been considered for the prediction of numerous possibilities of industrial application. The he machinability of glass-ceramics [2-5] microstructural characteristics of these materials. con- In this contribution the use of a brittleness index(B) sisting of highly interlocked mica crystals embedded in is suggested for estimating the machinability of glas a glass matrix, facilitate microfracture along the weak ceramics. Brittleness is a measure of the relative suscep. mica-glass interfaces and mica basal planes, avoiding tibility of a material to deformation and fracture 16) macroscopic failure during machining, i.e. cutting, turn- ing or drilling. It has been shown that very high deformation, and toughnessK, e), which quantifies the resistance to fracture. on the basis of an indentation machinable precision in these materials can be achieved mechanics analysis, the ratio of hardness to toughness (+10 um)using regular high-speedtools [21 Although machinability can be simple assessed quali has been proposed as a simple index of brittleness tatively as the ease with which a given material is cut (B=H/Kle)and the general applicability of B for deter- mining the question of ductility(deformation)versus fracture has been suggested [6]. The value of B varies Present address: Fachgebiet Werkstofftechnik, Technische Uni- beween s0. I um -for steels and =17 um for Si 84 Ilmenau, Germany. Tel +49 3677 monocrystal, with values for glasses and ceramics vary- 692450;fax:+493677691597 teneral within the 3-9μm-12167 0924-013697S17.00@ 1997 Elsevier Science S.A. All rights reserved PS0924-0136(96)02275-1
Journal of Materials Processing Technology 65 (I 997) 302 - 304 Short Comrrunication Machinability and brittleness of A.R. Boccaccini ’ glass-ceramics Shol otJ Mt~rrrllrrrg~~ d Murrriuls, The Utriwrsiry (!f Birtttittgltotn. Birtnittpitart~ B /5 2TT. UK Received 2 September 1995 Industrial summary The relationship between the machinability and the brittleness of glass-ceramic materials is investigated. A brittleness index (B), given by the ratio of the hardness to the fracture toughness, is proposed as a parameter for estimating the machinability. This approach is confirmed by considering experimental data from the literature on turning operations of mica-containing glass-ceramics. It is shown that machinability parameters, such as the slope of the log- log plot of the specific cutting energy versus the cutting rate, or the specific cutting energy at low cutting rates, are in good agreement with the brittleness indices for seven different glass-ceramics. In order for a glass-ceramic to be machinable, it was found that the brittleness index of the material should be lower than E a 4.3 pm-’ ‘. 0 1997 Elsevier Science S.A. Keyworrk Glass-ceramics; Brittleness; Machinability 1. Introduction Glass-ceramics are polycrystalline materials produced by the controlled crystallization of glass. They have already found several applications as engineering materials and new uses constantly appear [I]. An important group of these materials is that of the mica-containing glass-ceramics, due to their high machinability, which results in an increased versatility of the products and numerous possibilities of industrial application. The microstructural characteristics of these materials, consisting of highly interlocked mica crystals embedded in a glass matrix, facilitate microfracture along the weak mica-glass interfaces and mica basal planes, avoiding macroscopic failure during machining, i.e. cutting, turning or drilling. It has been shown that very high machinable precision in these materials can be achieved (+ 10 pm) using regular high-speedtools [2]. Although machinability can be simple assessed qualitatively as the ease with which a given material is cut, ’ Present address: Fachgebiet Werkstofftechnik. Technische UniVersitlt ilmenau, D-98684 Ilmenau, Germany. Tel.: +49 3677 692450; fax: +49 3677 691597. 0924-0136/97/S17.00 0 1997 Elsevier Science S.A. All rights reserved PII SO924-0136(96)02275-I its accurate quantitative measurement is difficult [2,3]. Various parameters have been suggested as the ‘measurement’ of the machinability, depending on the testing conditions employed, including tool wear, surface roughness, cutting force, cutting energy, drilling rates, etc. [2- 51. These parameters will depend strongly on the microstructure and properties of the glass-ceramic. In particular, fracture strength, hardness and fracture toughness have been considered for the prediction of the machinability of glass-ceramics [2-51. In this contribution the use of a brittleness index (B) is suggested f?r estimating the machinability of glassceramics. Brittleness is a measure of the relative susceptibility of a material to deformation and fracture [6], relating hardness (H), which quantifies the resistance to deformation, and toughness (K,,), which quantifies the resistance to fracture. On the basis of an indentationmechanics analysis, the ratio of hardness to toughness has been proposed as a simple index of brittleness (B = H/K,,) and the general applicability of B for determining the question of ductility (deformation) versus fracture has been suggested [6]. The value of B varies beween 20.1 pm-‘/* for steels and z 17 pm- I:* for Si monocrystal, with values for glasses and ceramics varying in general within the range 3-9 pm-“* [6,7]
R. Boc'caceini Jurnal of Mluteriuls Processing Techmefugu 65 (1997)302-304 Since machinability involves deformation and micro fracture, the brittleness index. comhining the response of the material to both of these phenomena, should be 025 a better parameter for its quantification than either hardness or fracture toughness taken separately. Under this premise. glass-ceramics with good machinability should behave in a less brittle manner. i. e. the index B should assume a low value. Literature results on the machinability of glass-ceramics are used in the next section to provide validation of this approach 2. Analysis of literature data Qualitative confirmation of the approach can be afforded by comparing the brittleness indices of Macor a well-known commercial machinable glass-ceramic [8]. -0,1 and of its base glass. From the data available in Ref. -0.15 [3], the brittleness indices for both materials can be calculated. Thus. whilst for the machinable glass Brittleness B(1/m) ceramic the brittleness index is very low (B=1.16 the brittleness index reaches a high value(B=8.05 mica glass-ceramics. Data fio Rel. 13/eness index(B)for seven Fig. I. Machinability parameter (n)vs um-2)that is typical for glasses [7] For quantitative verification of the proposed rela n=0.4570.106B tionship between machinability and brittleness, med- W, =0.23B.ss are required. Recent experimental data on turning tests Unfortunately, the values for both H and Kic.re of fluorophlogopite mica glass-ceramics with different quired for determining B, were available for ol y seven compositions and fabricated under different thermal glass-ceramic compositions in the original work [2]. For treatments can be used with this aim [2]. Turning comparison, Fig 3 shows the variation of the machin- operations of rod samples without lubrication and us- ability parameter n with hardness(H) for the seven ing a Ti(C, N)-based cermet tool were carried out, and the cutting rate dependence of the specific cutting en- ergy determined [2]. Two parameters were proposed for evaluating the machinability; (i)n, the slope of the Dg-log plot of the specific cutting energy versus cut ting rate, and(ii)u, the specific cutting energy at low cutting rate, neglecting elastic impact effects. it was 03.5 shown that good machinability is related to positive values of n, whilst negative values of n represent th predominance of brittle fracture along the glass phase 2.5 delamination through cyrstals, resulting in poor machinability. In the paper alluded [2] to the machinability parameters n and u were fitted to the hardness of the material (H)to give the following relationships 刀=0.643-0.122H sccH=n 0.5 where k is a proportionality constant The correlation between n and H, and between u, ar were, however, poor [2]. Indeed a better result btained by plotting the machinability parameters Brittleness B (1/um) and u as functions of the brittleness index B, as shown Fig. 2. Machinability parameter(,)vsbrittleness index(B)for seven Figs. I and 2. The fitted equations in this case ai mica glass-ceramics. ( Data from Ref. [2]
Since machinability involves deformation and microfracture, the brittleness in of the material to both of a better parameter for its q~a~ti~cat~o~ than either rdness or fracture toughness t en separately. Under s premise, glass-ceramics wi good ~~acl~i~ability should behave in a less brittle manner. i.e. the index B should assume a low value. Literature results on t machinability of glass-ceramics are used in the next section to provide validation of this approach. Qualitative confirmation of the approach can be afforded by comparing the brittieness indices of acor, a well-known commercial machinable glass-ceramic [8], and of its base glass. From the data available in Ref. [3], the brittleness indices for both materials can be calculated. Thus, whilst for the machinable glassceramic the brittleness index is very low (B = 1. I6 pn~ -’ -)), for the base glass, which lacks machinability, the brittleness index reaches a high value (B = 8.05 pm-’ “) that is typical for glasses [7]. For quantitative verification of the proposed relationship between machinability and brittleness, measured values characterising the machinability behaviour are required. Recent experimental data on turning tests of fluorophlogopite mica glass-ceramics with different compositions and fabricated under different thermal treatments can be used with this aim [2]. Turning operations of rod samples without lubrication and using a Ti(C.N)-based cermet tool were carried out, and the cutting rate dependence of the specific cutting energy determined [2]. Two parameters were proposed for evaluating the machinability; (i) IZ, the slope of the log-log plot of the specific cutting energy versus cutting rate, and (ii) z(i, the specific cutting energy at low cutting rate, neglecting elastic impact effects. :t was shown that good machinability is related to positive values of n, whilst negative values of n represent the predominance of brittle fracture along the glass phase over the continuous delamination through mica cyrstals, resulting in poor machinability. In the paper alluded [2] to the machinability parameters FZ and U, were fitted to the hardness of the material (H) to give the following relationships: n = 0.643 - 0.122H U, = kHz.25 where k is a proportionality constant. (1) (2) The correlation between n and H, and between uI and H were, however, poor [2]. Indeed, a better result is obtained by plotting the machinability parameters ?I and t(i as functions of the brittleness index B, as shown in Figs. 1 and 2. The fitted equations in this case are: -0.15 1 2 3 4 5 6 7 Brittleness B ( l/pm)“2 Fig. 1. Machinability parameter 01) vs. brittleness index (B) tbr seven mica glass-ceramics. (Data from Ref. [2].1 99 = 0.457 - 0.106B (3) l,, = 0 73B’= ._ (4) Unfortunately, the values for both H and A’,,, required for determining B, were available for ot.Iy seven glass-ceramic compositions in the original work [2]. For comparison, Fig. 3 shows the variation of the machinability parameter r’z with hardness (H) for the seven ; 2.5 6 t s 2 .- p 5 1.5 0 0 1 2 3 4 5 Brittleness B tl/ym)“’ Fig. 2. Machinability parameter (u,) vs. brittleness index (B) for seven mica glass-ceramics. (Data from Ref. [2].)
A.R. Boccaccini Journal of Materials Processing Technology 65(1997)302-304 the best correlation with the machinability parameter n for the glass-ceramics investigated, as expected 3. Conclusions The experimental data available for the machinability behaviour of glass-ceramics are shown to have a good correlation with the brittleness indices of the materials 005 This correlation is better than that between machinabil- ity and hardness or between machinability and fracture 0 toughness. The brittleness index B, therefore, should be preferred over hardness or toughness in order to predict -005 he machinability of glass-ceramics. If the criterion by Baik et al. [2]is valid, then good machinability occurs for n>0, and, from eq (3), good machinability occurs when the brittleness index of the material is lower than B=4.3 Hm-12. The validity of this result for other brittle materials has still to be investigated Hardness H(GPa) Fig 3. Machinability parameter(n)vs hardness(H)for seven mica glass-ccram ta from Ref. [2)-) References A.E. McHale, Engineering properties of glass-ceramics, in Engi eered Materials Handbook Vol. 4: Ceramics and Glasses, ASM International, Metals Park, OH, 1991, pp. 870-878 glass-ceramics considered. An indication of the accu- [2]DS Baik, K.S. No, J.S. Chun, Y.J. Yoon and H.Y. Cho,. Mater. sci,30(1995)1801-1806. racy of a linear fit can be given by the simple dimen-[3]HHKxu andS. Jahanmir, J. Am. Ceram Soc. 78(1995)497-500 sionless correlation coefficient, r[]: better correlation 4]M. Taira and M. Yamaki, J Mater. Sci. Lett, 13(1994)480-482 between two variables exists for r values closer to unity. [5]M. Reise and G. Muller, Strengthening of mica glass-ceramics, in For the linear function fitted to the n versus h data shown in Fig 3 the correlation coefficient is r=0.7, (6)BR. Lawn and D.B. Marshall, J. Am. Ceram Soc.62(1979) compared to r=0.95 found for the n versus b linear fit 347-350. given by Eq. 3)and shown in Fig. I. Moreover, the [7J Sehgal,Y. Nakao, H. Takahashi and S Ito, JMater. Sci. Lett (linear) relationship between n and fracture toughness 14(1995)167-169 (Klc), not shown here, also results in a poorer correla- [8 G.H. Beall, US Patent No. 3 689 293(1972). tion, with r=0. 8. Thus, the brittleness index B provides 91 W. Gohler, Hohere Mathematik, VEB Deutscher Verlag, Leipzig, 1989.p.I14
304 A. R. Boccaccini / Journal oj’ Materials Processing Teclmology 65 (1997) 302-304 1 2 3 4 5 6 7 Hardness H (GPa) Fig. 3. Machinability parameter (n) vs. hardness (H) for seven mica glass-ceramics. (Data from Ref. [2].) glass-ceramics considered. An indication of the accuracy of a linear fit can be given by the simple dimensionless correlation coefficient, r [9]: better correlation between two variables exists for r values closer to unity. For the linear function fitted to the n versus H data shown in Fig. 3 the correlation coefficient is r = 0.7, compared to r = 0.95 found for the n versus B linear fit given by Eq. (3) and shown in Fig. 1. Moreover, the (linear) relationship between n and fracture toughness (K,,), not shown here, also results in a poorer correlation, with r = 0.8. Thus, the brittleness index B provides the best correlation with the machinability parameter n for the glass-ceramics investigated, as expected. 3. Conclusions The experimental data available for the machinability behaviour of glass-ceramics are shown to have a good correlation with the brittleness indices of the materials. This correlation is better than that between machinability and hardness or between machinability and fracture toughness. The brittleness index B, therefore, should be preferred over hardness or toughness in order to predict the machinability of glass-ceramics. If the criterion by Baik et al. [2] is valid, then good machinability occurs for n > 0, and, from Eq. (3), good machinability occurs when the brittleness index of the material is lower than -B z 4.3 pm-li2. The vahdity of this result for other brittle materials has still t’o be investigated. References [l] A.E. McHale, Engineering properties of glass-ceramics, in Engineered Materials Handbook, Vol. 4: Ceramics and Glasses, ASM International, Metals Park, OH, 1991, pp. 870-878. [2] D.S. Baik, KS. No, J.S. Chun, Y.J. Yoon and H.Y. Cho, J. Mater. Sci., 30 (1995) 1801-1806. [3] H.H.K. XuandS. Jahanmir, J. Am. Ceram. Sot., 78(1995)497-500. [4] M. Taira and M. Yamaki. J. Mater. Sri. Left.. 13 (1994) 480-482. [5] M. Reise and G. Miiller, Strengthening of mica glass-ceramics, in P. Duran and J.F. Fernandex (eds.), Third Eworceramics, I/. 2, Faenza Editrice IbCrica. Madri’d, Spain, 1993. pp. I ISI- 1156. [6] B.R. Lawn and D.B. Marshall, J. Am. Ceram. Sot.. 62 (1979) 347-350. 171 J. Sehgal, Y. Nakao, H. Takahashi anld S. Ito, J. Marer. Sci. Lett., 14 (1995) 167-169. [8] G.I-I. Beall, US Patent No. 3 689 293 (1972). [9] W. Giihler, H&ere Matirematik, VE& Deutscher Verlag, Leipzig, 1989, p. 114