merarils
Upcoming SlideShare
Loading in...5
×
 

merarils

on

  • 295 views

 

Statistics

Views

Total Views
295
Views on SlideShare
295
Embed Views
0

Actions

Likes
0
Downloads
2
Comments
0

0 Embeds 0

No embeds

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

merarils merarils Document Transcript

  • Journal of Welcome Comments Help MATERIALS RESEARCH A new method for hardness determination from depth sensing indentation tests J. Gubicza, A. Juh´ sz, and J. Lendvai a Department of General Physics, E¨ tv¨ s University, Budapest, H-1088 M´ o o uzeum krt. 6-8, Budapest, Hungary (Received 18 December 1995; accepted 30 July 1996) A new semiempirical formula is developed for the hardness determination of the materials from depth sensing indentation tests. The indentation works measured both during loading and unloading periods are used in the evaluation. The values of the Meyer hardness calculated in this way agree well with those obtained by conventional optical observation, where this latter is possible. While the new hardness formula characterizes well the behavior of the conventional hardness number even for the ideally elastic material, the mean contact pressure generally used in hardness determination differs significantly from the conventional hardness number when the ideally elastic limiting case is being approached. Hardness testing with sharp indenters is generally the same velocity is applied in the unloading periodconsidered as a simple method for characterizing the when the pyramid moves backwards. In the course ofmechanical strength of materials. Recently the depth the test, the load is registered as a function of thesensing indentation (DSI) test became a widely used penetration depth. The measurements were carried outmethod of hardness determination.1–8 In the DSI tests in the macrohardness region (Pm 100N) on the fol-the applied load is registered as a function of indentation lowing materials: metals (99.99% pure Al and Cu), sodadepth both during loading and unloading. A schematic lime silica glass, sodium chloride, polypropylene, andload versus penetration depth curve is shown in Fig. 1. Si3 N4 ceramics of two compositions sintered to differentThe most frequently used DSI method was developed by densities. Compositions and densities of ceramic samplesOliver and Pharr3 by which the hardness number can be shown in Table I.determined without optical observation. In this method The conventional definition of the Meyer hardnessthe mean contact pressure at the maximum depth is used for any sharp indenter geometry is the following10 :to characterize the plastic properties of materials. If, Phowever, a considerable part of the deformation during H , (1)the indentation is elastic, this pressure deviates signif- Aicantly from the conventional hardness number which where P is the applied load and A is the residualis determined by optical observation of the indentation projected area of the hardness impression after unload-trace. For example, in the limiting case of the ideally ing. The hardness measurement was originally devel-elastic material, the conventional hardness number tends oped for testing metals in which the deformation isto infinity while the mean contact pressure gives finite mostly plastic (sy E is small where sy is the flowvalue since an elastic contact surface is developing stress and E is Young’s modulus); therefore, there isbetween the indenter tip and the ideally elastic material.2 practically no elastic recovery under unloading, and The paper is a continuation of a recently pub- the projected area at the maximum depth equals thelished work9 in which a new semiempirical formula residual projected area after unloading. Consequently,has been developed for the determination of the Meyer the mean contact pressure at the maximum penetrationhardness10 of materials. With the formula proposed, even depth (defined as the indentation load divided by thethe limiting case of the ideally elastic materials can be projected contact area) equals the conventional hardnesscorrectly described. The main results of our recent paper number (H) determined after unloading. This is alsoare briefly summarized below to give a basis for the the case when the elastic portion of the deformationsubsequent comparison with another evaluation method.3 during the indentation is negligible and the projected area Hardness measurements were carried out on dif- before unloading agrees well with that measured afterferent materials by the DSI method using a computer- unloading, because—although the elastic recovery maycontrolled hydraulic mechanical testing machine with a be significant for the indentation depth— the character-Vickers indenter. During the loading period the Vickers istic surface dimensions exhibit only minor recovery.1,10pyramid penetrates the sample at constant velocity, and On the other hand, if the deformation is mostly elastic2964 J. Mater. Res., Vol. 11, No. 12, Dec 1996 © 1996 Materials Research Society
  • Communications the entire loading-unloading period (Fig. 1): Wd Wt 2 We . (4) The work performed during loading (Wt ) and that re- gained during unloading (We ) can be calculated by the integration of Eqs. (2) and (3), respectively. It was found that in spite of the linear terms appear- ing in Eqs. (2) and (3) for a broad variety of materials and in a wide load range, the following relationship is with good accuracy valid (see Fig. 2): r We c3 . (5) Wt c3 The parameter c3 characterizes the resistance of the material against the elastic-plastic deformation.11 In the case of ideally plastic materials, the load-depth function FIG. 1. Schematic picture of an indentation cycle. is purely quadratic: P c3 h2 ,1,12 and in this case there is no elastic relaxation, the d 7 ? hm equation betweenTABLE I. Compositions and densities of Si3 N4 ceramic samples. the diagonal d and the maximum indentation depth hm 90 wt. %, Si3 N4 , 4 wt. % 90.9 wt. %, Si3 N4 , 3 wt. % which is the consequence of the geometry of the Vickers Al2 O3 , 6 wt. % Y2 O3 Al2 O3 , 6.1 wt. % Y2 O3 pyramid is exactly satisfied [Fig. 3(a)]. Consequently the Meyer hardness of ideally plastic materials can be givenSample Density (g cm3 ) Sample Density (g cm3 ) in the following form: 1 2.03 6 2.697 2 2.11 7 2.823 P P H 2 a1 ? c3 (6) 3 2.34 8 2.935 d2 2 24.5hm 4 2.54 9 2.954 5 2.70 10 3.032 with a1 0.0408 for the Vickers geometry. 11 3.115 If the material is not ideally plastic then with in- 12 3.161 creasing elastic contribution, the elastic deflection under the indenter is increasing [Fig. 3(b)]. Consequently 7hm becomes increasingly larger than d, and as a result of(sy E is large) then a significant portion of the contact this a1 ? c3 will be less than H. This can be taken intoarea at maximum depth is due to elastic deformation. account introducing a relationship between H and a1 ? c3Consequently, the residual projected area is smaller than of the form:the projected contact area at maximum penetration, and Wtthe conventional Meyer hardness is larger than the mean H a1 ? c3 ? . (7)contact pressure. Wd The load-penetration depth function can be de-scribed with quadratic polinoms (Fig. 1): P c2 h 1 c3 h2 , (2) P c2 h 2 h0 1 c3 h 2 h0 2 , (3)both in the loading and in the unloading periods, respec-tively, where P is the load, h is the penetration depth,and h0 is the residual indentation depth after removingthe punch; c2 , c3 , c2 , and c3 are fitting parameters.The total indentation work, Wt , is the integral of theload versus the indentation depth, i.e., the area underthe load-penetration depth curve during the unloadingperiod. Upon unloading a part of this work, We , can beregained; it equals to the area under the load-indentationdepth curve for this latter period. The difference of these FIG. 2. The ratio of the elastic and total work versus the parameterstwo quantities, Wd , gives the net work expended during of the indentation curves. J. Mater. Res., Vol. 11, No. 12, Dec 1996 2965
  • Communications Sneddon’s elastic theory13 and empirical results of Oliver and Pharr,3 the contact depth can be given as: Pm hc hm 2 e , (9) S where S is the slope of the initial part of the unloading curve (Fig. 1) and e 0.75 for the case of Vickers indenters. To compare the new hardness formula (7) with the mean contact pressure as expressed in Eqs. (8) and (9), Pm and hc are expressed with the c3 and c3 parameters and the indentation works. If only the quadratic terms existed in Eqs. (2) and (3), Pm and hc could be expressed easily with the parameters c3 and c3 . Because of the existence of the linear terms, some approximations areFIG. 3. Schematic picture showing the behavior of various materials used in the considerations.(a –c) during Vickers indentation. According to Eq. (3) S can be given as dP ÅIt is obvious that this new definition of H gives back S c2 1 2c3 hm 2 h0 . (10)Eq. (6) for ideally plastic materials and it increases dh hmwith increasing elasticity, and in the limiting case of The second term in Eq. (10) may be expressed as aan ideally elastic material it becomes infinite, because in fraction of S:this case there is no residual deformation after unloading[Fig. 3(c)]. Figure 4 shows that the H values determined kS 2c3 hm 2 h0 (11)from DSI measurements according to Eq. (7) agree wellwithin the experimental errors with the conventionally and similarly the quadratic term of the load-depth func-determined hardness. tion of the unloading curve as a fraction of the maximum The most frequently used hardness determination load:method from DSI measurements introduced by Oliver k Pm c3 hm 2 h0 2 . (12)and Pharr3 is based on the mean contact pressure atthe maximum indentation depth with the following With Eqs. (11) and (12) Pm S can be given asexpression: Pm Pm Pm Pm k k hm 2 h0 Hm a1 , (8) . (13) A 2 24.5hc 2 hc S k k 2 From Eqs. (3) and (10) –(13), hc can be given in thewhere A is the projected contact area and hc is the following form:contact depth at the maximum load [Fig. 3(b)]. Using 2 e hc hm 2 h m 2 h0 . (14) 11k 2 Expressing the quadratic term in Eq. (2) as a fraction of the maximum load: 2 kPm c3 hm , (15) the Hm mean contact pressure in Eq. (8) can be written as follows: 1 Hm a 1 c3 ≥ p p q ¥. 2 k e c3 (16) k2 11k 2 c3 According to our measurements for the different materials investigated with different loads, k and k vary pFIG. 4. The conventionally determined hardness number versus the between 0.64 and 1. If k > 0.64 then 2 k 11quantity calculated on the basis of Eq. (7). k > 0.98; therefore, this quantity can be taken as 1.2966 J. Mater. Res., Vol. 11, No. 12, Dec 1996
  • CommunicationsUsing Eq. (5) the following relationship is obtained for A new formula is proposed for the characterizationHm : of the hardness of materials from DSI tests. The val- 1 ues of the hardness calculated by this equation agree Hm a 1 c3 ≥ p ¥ . (17) well with those measured by the conventional method e We 2 k2 2 Wt for a broad variety of materials. Comparing the new hardness formula with that based on the mean contactThis can be compared with the new hardness formula pressure, the former correctly describes the behavior ofEq. (7), which using Eq. (4) can be written as: the conventional hardness both in the ideally plastic and 1 ideally elastic limiting cases, while the latter deviates H a 1 c3 We . (18) from the conventional hardness number in the ideally 12 Wt elastic limit. At the same time in the range where optical The difference between H and Hm can be seen in measurements can also be applied (0 < We Wt < 0.7),Fig. 5 in which the two hardness numbers divided by the difference between the two DSI evaluation methodsa1 c3 are shown as a function of We Wt . For Hm a1 c3 p is within the experimental error. A technical advantagethree curves are shown as k equal to 0.8 or 0.9 or of the new evaluation method is that the parameters used1 (k 0.64 or 0.81 or 1). We Wt equals zero for the in the hardness formula can be determined at a highideally plastic limiting case; it increases with increasing accuracy from the registered indentation curves. Furtherelastic deformation in the indentation, and it is 1 for the measurements are planned on materials and for loadingideally elastic material. By the new hardness formula conditions falling in the region where the new hardness[Eqs. (7) or (18)], the behavior of the conventional hard- number is rapidly increasing (0.8 < We Wt < 1). Thisness number (and consequently the plastic properties of regime, however, is not easy to obtain experimentally.materials) is better characterized than by the mean con-tact pressure, because while the former tends to infinity ACKNOWLEDGMENTSthe latter gives a finite value in the ideally elastic limitingcase. As it can be seen in Fig. 5 in the region 0 < The authors are grateful to P. Arat´ for providing oWe Wt < 0.7, there is no significant difference between the Si3 N4 ceramic samples. This work was supportedthe two hardness numbers. The We Wt values for the ma- by the Hungarian National Scientific Fund in Contractterials investigated here are in this region; consequently Nos. T-017637 and T-017639.the hardness numbers obtained by both methods agreewell with the values of the Meyer hardness obtained by REFERENCESoptical observation after unloading (Fig. 5). 1. M. Sakai, Acta Metall. Mater. 41, 1751 (1993). 2. G. M. Pharr, W. C. Oliver, and F. R. Brotzen, J. Mater. Res. 7, 613 (1992). 3. W. C. Oliver and G. M. Pharr, J. Mater. Res. 7, 1564 (1992). 4. G. M. Pharr and W. C. Oliver, MRS Bull. 17, 28 (1992). 5. A. Juh´ sz, G. V¨ r¨ s, P. Tasn´ di, I. Kov´ cs, I. Somogyi, and a oo a a J. Sz¨ ll¨ si, Colloque C7, suppl´ ment au J. de Phys. 3, 1485 o o e (1993). 6. F. R. Brotzen, Int. Mat. Rev. 39, 24 (1994). 7. A. Juh´ sz, M. Dimitrova-Luk´ cs, G. V¨ r¨ s, J. Gubicza, a a oo P. Tasn´ di, P. Luk´ cs, and A. Kele, Fortchrittsberichte der a a Deutschen Keramischen Gesellschaft 9, 87 (1994). 8. J. Gubicza, Key Eng. Mater. 103, 217 (1995). 9. J. Gubicza, A. Juh´ sz, P. Arat´ P. Tasn´ di, and G. V¨ r¨ s, a o, a oo J. Mater. Sci. 31, 3109 (1996). 10. D. Tabor, Hardness of Metals (Clarendon Press, Oxford, 1951). 11. F. Fr¨ hlich, P. Grau, and W. Grellmann, Phys. Status Solidi (a) o 42, 79 (1977).FIG. 5. The hardness numbers divided by a1 c3 as a function of 12. B. R. Lawn and V. R. Howes, J. Mater. Sci. 16, 2745 (1981).We Wt . 13. I. N. Sneddon, Int. J. Engng. Sci. 3, 47 (1965). J. Mater. Res., Vol. 11, No. 12, Dec 1996 2967