ebook img

Cyclic Fatigue of Brittle Materials with an Indentation-Induced Flaw System PDF

10 Pages·1996·0.56 MB·English
by  ChoiSung R.
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Cyclic Fatigue of Brittle Materials with an Indentation-Induced Flaw System

-,Q _r_//_ _ _ # _/_ _ _ NASA-TM-I12475 Reprinted from Materials Science and Engineering A208 (1996) 126-130 Cyclic fatigue of brittle materials with an indentation-induced flaw system Sung R. Choi a'', Jonathan A. Salem b aCleveland State University, Cleveland, OH 44115, USA bNASA Lewis Research Center, Cleveland, OH 44135, USA Received 1May 1995; in revised form 10October 1995 ELSEVIER MATERIALS SCIENCE AND ENGINEERING A The journal provides an international medium for the publication of theoretical and experimental studies and reviews of the properties and behavior of a wide range of materials, related both to their structure and to their engineering application. The varied topics comprising materials science and engineering are viewed as appro- priate for publication: these include, but are not limited to, the properties and structure of crystalline and non-crystalline metals and ceramics, polymers and composite materials. Editor-in-Chief L. Priester (France) Professor H. Herman S. Sampath (USA) Associate Editors V. K. Sarin (USA) P. Shen (Taiwan) M. Koiwa (Japan) M. Suery (France) G. Kostorz (Switzerland) S. Suresh (USA) Editorial Board (MSE A} N. S. Stoloff (USA) J. _,gren (Sweden) M. Taya (USA) G. Ananthakrishna (India) A. K. Vasudevan (USA) R. J. Arsenault (USA) A. Vevecka (Albania) D. Brandon (Israel) B. Wilshire (UK) H. K. D. H. Bhadeshia (UK) M. Yamaguchi (Japan) J. Cadek (Czech Republic) T. S. Yen (China) J B. Cohen (USA) J. Driver (France) Print and Electronic Media Review Editor J. D Embury (Canada) A. H. King (USA) Y. Estrin (Australia) H. Fischmeister (Germany) Administrative Editor C. Garcia de Andros (Spain) Barbara Herman H. Gleiter (Germany) M. W. Grabski (Poland) Advisory Board (MSE A and B) M. Kato (Japan) H. Herman, Chairman (USA) Y. G. Kim (Korea) H. Curien (France) C Laird (USA) M. E. Fine (USA) J. Lendvai (Hungary) A. Kelly, FRS (UK) W. Mader (Germany) H. Mughrabi (Germany) M. McLean (UK) H. Rangu (Japan) Types of contributions Malaysia, Mexico, New Zealand, Pakistan, Singapore, South Original research work not already published; plenary lectures Africa, South Korea, Taiwan, Thailand and the USA. Airmail rates and or individual papers given at conferences; reviews of for other countnes are available on request. specialized topics within the scope of the journal; engineering For advertisang rates apply to the publishers. A specimen copy studies; letters to the editor will be sent on request US and Canadtan customers may obtain mformat_on from the Subscription Information 1996 following Volumes 205-221, each volume containing 2 Fssues, are scheduled for publication Prices are available from the ELSEVIER SCIENCE INC. publishers upon request Subscripttons are accepted on a pro Attn.: Journal Information Center, 655 Avenue of the Americas paid bas_s only Issues are sent by SAL (Surface Air Lifted) mail New York, NY 10010, USA. wherever this serwce =savailable. Airmail rates are available Telephone: (212) 633 3750 Telex: 420 643 AEP UI upon request Please address all requests regarding orders and Telefax: (212) 633 3764 subscnpt=on quenes to Abstracting and/or Indexing Services ELSEVIER SCIENCE SA PO. Box 564, 1001 Lausanne, Switzerland American Ceramic Sooety; Cambridge Scientific Abstracts; Telephone: (21) 3 20 73 81 Chemical Abstracts; Current Contents; Engineering Index; FIZ Telex: 450620 ELSA CH Telefax: (21)3 235 444 Karlsruhe; Fluid Abstracts; Fluidex; Glass Technology Abstracts; Inspec Physics Abstracts; Metals Abstracts; Pascal (Centre Issues are sent by surface mail after air delivery to Argentuna, National de la Recherche Scientifique); Physikalische Berichte; Australia, Brazil, Canada, China, Hong Kong, India, Israel, Japan, Research AlertTM; Science Oration Index MATERIALS SCIENCE & ENGINEERING A ELSEV|ER Materials Science and Engineering A208 [19961126 131) Cyclic fatigue of brittle materials with an indentation-induced flaw system Sung R. Choi a'*, Jonathan A. Salem b "('h'vehmd State University, Ch'lehmd, OH 44115, USA hN,4SALe_is Research ('enter, ('leleland. OH 44135, USA Received I May 1995;in revised form 10October 1995 Abstract The ratio of static to cyclic fatigue life, or 'h ratio', was obtained numerically for an indentation flaw system subjected to sinusoidal loading conditions. Emphasis was placed on developing a simple, quick lifetime prediction tool. The solution for the h ratio was compared with experimental static and cyclic fatigue data obtained from as-indented 96wt.% alumina specimens tested in room-temperature distilled water. Keywords:Cyclic fatigue; Static fatigue; Indentation-induced flaw:,Ceramics I. Introduction The purpose of this study is to analyze cyclic fatigue of the indentation flaw system under sinusoidal loading For glass and ceramic materials which have slow crack conditions so that lifetime prediction from one fatigue growth (stress corrosion) as the unique, time-dependent condition to another is readily feasible. For this purpose, failure mechanism, it is possible to predict the fatigue life the complete solution of the ratio of static to cyclic under one loading condition from another. Prediction of fatigue life was obtained numerically in conjunction with cyclic fatigue lifetime from static lifetime for Griffith flaw fatigue parameter (n) and stress ratio iR ratio). The system can be done analytically by using the ratio of solution was compared with experimental data that were static fatigue to cyclic fatigue life, introduced by Evans obtained from static and cyclic fatigue testing of indented and Fuller [1]. In the case of the indentation flaw system, alumina specimens in room-temperature distilled water. the solution requires a numerical approach since an The analysis was carried out for material exhibiting a flat additional driving force appears in the net stress intensity R curve and a power-law crack propagation as a delayed factor. This term is attributed to residual contact stress failure mechanism. produced by the elastic/plastic deformation of indenta- tion [2]. Analyses of dynamic and static fatigue of the 2. Analysis indentation flaw system were carried out previously for both postthreshold [3,4] and subthreshold [5] indentation In many cases slow crack growth of glass and ceramic flaws. However, no general solution for the cyclic fatigue materials under Mode I loading conditions is described of such a flaw system has been found, although some by the following empirical, power-law relation [4] cyclic fatigue data on ceramic specimens containing indentation cracks exist [6 9]. da A_ K, ]" (I) t_= d7= LK,,,] where v, a, and t are crack velocity, crack size, and time, *Corresponding author. Addresscorrespondence to: NASA Senior Resident Research Associate, Lewis Research ('enter, Cleveland, OH respectively. A and n are the fatigue parameters which 44135, USA. depend on material and environment. K_ is the 0921-5093/96,:'$15.00,*_.1996 Elsevier Science S.A. Allrights reserved SSDI 0921-5093(95)10068-7 S.R. Choi, J.A. Salem _'Materials Science and Engineering A208 (1996) 126 130 127 Mode I stress intensity factor, and Kw is the Mode I .... I ...... I critical stress intensity factor or fracture toughness of the R=IO material with a flat R curve. 2.1. Natural flaws 0 In static fatigue testing a constant applied stress (a) is employed. Noting that n > 10 for most glass and ceram- ics, one can derive the following static fatigue equation [10l 0.1 tfs= B S_' 2or " (2) where tfs is the time to failure, S, is the inert strength, and (S[NUSOIDAL WAVE) _'_ I / B is the parameter associated with fracture toughness, lO 100 crack geometry and fatigue parameters. FATIGUE PARAMETER, n In cyclic fatigue testing a time-varying, periodic stress (a(t)) is applied. The time to failure (trc) in cyclic fatigue Fig. I. Numerical solution of the h ratio as a function of fatigue parameter (n) for different levels of R ratios with the natural flaw can be expressed as [1] system. Each line represents a best-fit regression line. ! .... (3) is possible to obtain an approximate (but accurate) lfc = BSi 20 max_1 [f(t)]" dt relation between log h and log n for a given R ratio by T using the regression analysis where ar,ax is the maximum applied stress, f(t) is a periodic function defined as a(t) = O'marf(,/) with a range log h = c_log n + fl (6) of 0<f(t) < 1, and r is the period. The ratio of static where _ and fl are the regression coefficients. fatigue to cyclic fatigue life, h, with a condition of a in Table 1shows the regression coefficients _ and ft. For static loading equal to O'maxin cyclic loading (a = O'max) n> 10, the maximum error in h ratio, as compared with can be obtained from Eqs. (2) and (3) the exact solution, was 1.0% for R = 0 to 0.7, and 4.5 % h- tfs -- 1 ['_ [f(t)]" dt (4) for R = 0.8 to 0.9. Therefore, the table gives a convenient and accurate means of determining the h ratio for a /re r Jo natural flaw system for any given value of n, either integer Sinusoidal loading is the most common and popular or real. The complication of a lack of a simple analytical wave form used in cyclic fatigue testing. The periodic solution to Eq. (5) for real numbers can be eliminated function fit) for a sinusoidal wave is expressed as with Eq. (6) and Table 1. f(t) = [(1 + R)/2 + [(1 - R)/2]sincot], where R isthe stress (or load) ratio, defined as R = amin/_rmax with amin being 2.2. Indentation flaws the minimum applied stress, coisthe angular velocity. Eq. (4) can be solved either analytically or numerically for Because of the residual contact stress produced by the sinusoidal wave. The h ratio for any other periodic elastic/plastic indentation deformation [2], an addi- loading wave form such as trapezoidal, triangular or square can be solved analytically with a straightforward Table 1 procedure [1,11]. Eq. (4) was solved analytically for a Regression coefficients of the h ratio (Eq. (6)) for natural and sinusoidal wave by using the series expansion technique indentation flaw systems of Evans and Fuller [1]. Their analytical solution is Natural flaws Indentation flaws expressed as ...2. [ n, 1F 1-R 12k(l+R']" h=k_oL(n-2_)!(k!)2JLZ(l-+--Ri j \-5--/ (5) 0.0 - 0.4939 - 0.2607 -- 0.1 -0.4955 -0.2348 -0.4856 -0.1925 Eq. (5) is still complicated to solve and valid only for 0.2 -0.4976 -0.2051 -0.4885 -0.1612 integer values of n. Therefore, it is desirable to solve Eq. 0.3 -0.5005 -0.1703 -0.4925 -0.1240 (4) numerically to cover any integer or real value of n for 0.4 -0.5049 -0.1282 -0.4977 -0.0804 a full range of R = 0-1.0, even though some limited data 0.5 -0.5112 -0.0765 -0.5049 -0.0268 on the h ratio exist [1,12]. 0.6 -0.5195 -0.0115 -0.5126 -0.0383 0.7 -0.5291 0.0705 -0.5198 0.1172 Fig. 1 shows the results of the numerical solution of 0.8 -0.5353 0.1743 -0.5173 0.2075 Eq. (4) as a function of n for R ratios from R = 0 to 0.9 -0.5138 0.2977 -0.4743 0.2932 R = 1.0. If log h is treated as a linear function of log n, 1.0 0.0000 0.0000 0.0000 0.0000 coefficients of correlation of rcoer> 0.9966 result. Thus it 128 S.R. Choi, J.A. Salem / Materials Science and Engineering A208 (1996) 126 130 tional term appears in the net stress intensity factor and 10e ^S-I.DE.Ti':D 80''do ' an analytical solution of cyclic fatigue for the indenta- ._ 107 CRACKS 40 , 4 tion flaw system is not feasible. The solution needs to be done via numerical methods, as done previously for " 106 20 ', both dynamic and static fatigue analyses involving in- 105 10 -I dentation fracture mechanics [3,4]. As in the previous 04 .._. ":. \ ' studies [3,5], normalized variables are introduced here ...... ",'::.. %,. ,, I n=5 -.L.:.L.:.:.... "%:_.., '"%, as follows: ¢_ 102 r_ K*= --K,'l J = -A-t; or*-- O'ma x," C*=-- (./ (7) KIc am o"m am 101 =¢ 10° (cS,mtcuUeSOFI,Dla_owu,_v.E) q where K*, J, a* and C* are, respectively, normalized 0 --R= !.O(R'rAI"IC) z 10 -1 ..... R=o.s stress intensity factor, normalized time, normalized ........R..-0.1 I 10 -2 , , , , , , ,, I maximum applied stress and normalized crack size. o-m 0.1 0.3 0.50 1 and am are, respectively, the strength and critical crack NORMALIZED MAX APPLIED STRESS, o size in the inert condition. Using these variables, the crack growth rate of Eq. (1) and net stress intensity Fig. 2. Typical example of numerically obtained normalized failure factor are expressed as follows: time (Jr) as a function of normalized maximum applied stress (o-*) for three different R ratios of R=0.1, 0.5 and 1.0 for the indentation dC* flaw system. dJ - [K*]" 3 , , I C, 32 which reduces to the relationship in static fatigue of K* = _oC + _ (8) indentation cracks [4]. This indicates that the relation- ship between the true (n) and apparent (n') fatigue , [-1 + R +--1--5-Rsln__)j o'* parameters is independent of R ratio. Based on the results as shown in Fig. 2, the h ratio Note that the normalized net stress intensity factor was determined for a given R using the relation h= Jfj consists of two terms: the first is a function of the Jfc ( = tf_/t,O, where J,-_and Jfc are the normalized time remote applied stress and the second is related to the to failure corresponding to static and cyclic fatigue, residual contact stress [2,3]. The differential form of Eq. respectively. Fig. 3 shows a summary of the h ratio as (8) was solved numerically using a fourth-order a function of n for R ratios from R = 0.1 to R = 1.0, Runge-Kutta method. The initial condition was C* = where log h was plotted against log n as for the natural 0.3967 at J=0 and the instability conditions were flaw system (Fig. 1). Again the excellent coefficients of K* = 1 and dK*/dC* > 0. The solution was initiated to correlation of rco_r->0.9911 allow one to obtain a rela- determine the normalized time to failure (Jr) as a func- tionship based on Eq. (6). The results of such regression tion of normalized maximum applied stress for the analysis for _ and fl are shown in Table 1. The maxi- selected values of n = 5-160. This procedure was con- tinued for the range of stress ratios from R = 0.! to 1.0. Since the solution is independent of frequency Or) as ' ' ' '1 ' .... I long as _am/A > 2_zf, any particular values of _am/A R=I.O satisfying ¢Oam/A _>2_ can be chosen. A value of (Oam/ 1 A = 100 was used in this analysis. Fig. 2 shows a typical example of the normalized time to failure as a function of normalized maximum o applied stress for stress ratios of R = 1.0, 0.5 and 0.1. n,, Fatigue susceptibility increases with increasing R ratio, .c yielding a maximum at R = 1.0 (static fatigue). Note 0.1 that regardless of R ratio the curves converge to a* = 1.0, in which the inert strength with no slow crack AS-INDENTED CRACKS "_ growth is defined. Similar to the previous static fatigue - (SINUSOIDAL WAVE) analysis [4], the slope in Fig. 2 is not representative of i ,A[ ..... f a 'true' fatigue parameter of n, due to the effect of 10 100 residual contact stress. The slope in Fig. 2, which is FATIGUE PARAMETER, n called 'apparent' fatigue parameter (n'), was found to have the following approximate relation Fig. 3. Numerical solution of the h ratio as a function of fatigue parameter (n) for different R ratios with the indentation flaw system. n = 4/3n' - 2/3 (9) Each line represents a best-fit regression line. S.R. Choi, J.A. Sah'm .;Materials Science and Engineering A208 (1996) 126 130 129 mum error associated in the h ratio was observed to be 107 ' I ' 967. ALUMINA about 4%, occurring at R=0.9 for n> 10. Hence, 106 _ (as-indented; lifetime prediction of an indentation flaw system with one 'W RT water) fatigue loading condition can be done quickly and 105 Static accurately from another by using Eq. (6) in conjunction 104 n'=35.4 with Table 1. ,..1 103 2.3. Evaluation oJfittigue parameters Jor indentation n'=32.9 _ Cyclic o 102 f[LIWS • cyclic B 101 {3 static O The "true" fatigue parameter n for both static and [3 cyclic fatigue can be determined using Eq. (9) once the 100 "apparent" fatigue parameter n' is determined from the 10-1 , , , , I slope of fatigue data. The fatigue parameter A in static 50 60 70 8090100 200 300 fatigue was determined previously [4] (MAX) APPLIED STRESS, (7[MPa] A = (2-_-']] _..a..,,zq1,-- (10) t.') Fig. 4. Results of static and cyclic fatigue testing obtained from indented 96 wt.% alumina flexure specimens tested in room-tempera- where 5_[is the intercept of static fatigue data, expressed lure distilled water. The solid lines represent the best-ill lines in a plot as tr_a'"= 2_. Since the following relation for the inden- of log t_vs. log a,,,,_. tation flaw system holds perature distilled water are depicted in Fig. 4. Consis- Jl+ _ tf+ z"+o- ,,, -)-_=h (ll) tent with the results shown in Fig. 3, the alumina is slightly more susceptible to fatigue in static loading where 2'< is the intercept of cyclic fatigue data as than in cyclic loading within the experimental range trio-n.'... = z"_¢. Hence, from Eqs. (10) and (11), the parame- used. The "apparent" fatigue parameter was obtained ter A in cyclic fatigue is obtained from the data as n' = 32.92_+ 6.01 and 35.37 +4.79, t/2_'_12 ,, I respectively, for static and cyclic fatigue. The corre- A = tp_7 ) O'm(/m _,ch (12) sponding "true" fatigue parameter was obtained from Eq. (9) as n=43.23 _+8.01 and 46.49_+ 6.39, respec- 3. Experimental procedure tively, for static and cyclic fatigue• The prediction of cyclic fatigue from static fatigue data can be done using Static and cyclic fatigue tests of indented alumina (96 Eq. (6) together with Table 1.The prediction thus made wt.%, ALSIMAG 614, General Electrical Ceramics) is presented in Fig. 5, where the cyclic fatigue data flexure specimens were carried out in room-temperature obtained from the experiments are included for com- distilled water using a four-point bend fixture of 6.05 mm parison. The prediction somewhat overestimates (less inner and 19.05 mm outer spans. The nominal dimen- sions of the test specimens were 4 mm by 5 mm by 25 mm, respectively, in height, width, and length. The center 107 (5 mm side) of each specimen was indented in air for I 96X ALUMINA about 20 s using a Vickers microhardness indenter 106 \_ (as-indented; _, RTwater) (Zwick, model 3212, Germany) with an indentation load 105 Best fit_ \\"_ of 49 N. Static fatigue testing was conducted by loading of .static \_'_ Best fitof indented specimens in a lever-arm creep machine (ATS) _e_d 104 fahgue _-_"" cyclic fatigue with constant stress levels of 95 to 133 MPa. Cyclic 103 fatigue testing of indented specimens was conducted by .< i_. _ Predicted sinusoidal loading with a stress ratio of R = 0.5 and a o 102 _ _".."_(from static. A,'_ fatigue data) frequency of f= 5 Hz with a servohydraulic testing 10 _ machine (|nstron, Model 8562). The maximum applied F • cyclic data stress in cyclic fatigue ranged from 100 to 130 MPa. The 10 o sinusoidal wave shape was frequently checked and ver- 10 -1 , _ _ , I ified by a digital storage oscilloscope. 50 60 70 80 90100 200 300 (MAX) APPLIED STRESS, (7[MPa] 4. Results and discussion Fig. 5. Prediction of cyclic fatigue lifetime from static fatigue data for The results of the static and cyclic fatigue testing of indented 96 wt.% alumina specimens tested in room-temperature the indented 96 wt.% alumina specimens in room-tem- distilled water. 130 S,R. Choi, J.A. Salem / Materials Science and Engineering A208 (1996) 126 130 than one order of magnitude) the actual cyclic time to compared with static and cyclic fatigue data obtained failure. from indented 96 wt.% alumina specimens tested in It has been reported that, for certain ceramic materi- room-temperature distilled water. als, damage accumulation and/or fatigue synergisms can be active in cyclic fatigue, resulting in more fatigue susceptibility in cyclic than in static loading [8,13-16]. Acknowledgements If such synergies exist or if creep at elevated tempera- tures is simultaneous with cyclic fatigue, the analysis The authors are grateful to R. Pawlik for the experi- given in this paper may not be valid. However, the mental work. This work was sponsored in part by the analysis may still provide, by comparing with experi- Ceramic Technology Project, DOE Office of Trans- mental data and by using fractographic analysis, some portation Technologies, under contract DE-AC05- clues with which the prevailing mechanism associated 84OR21400 with Martin Marietta Energy System, Inc. with failure can be pinpointed. It should be noted that most of glass and ceramic materials under static fatigue loading conditions in a room-temperature moisture en- References vironment are subjected to one failure mechanism, stress corrosion. Therefore, in view of the reasonable [1] A.G. Evans and E.R. Fuller, Crack propagation in ceramic agreement between the static and cyclic fatigue data materials under cyclic loading conditions, Metall. Trans., 5 shown here, it can be stated that stress corrosion is a (1974) 27-33. governing failure mechanism, either in static or in cyclic [2] B.R. Lawn, A.G. Evans and D.B. Marshall, Elastic/plastic in- loading conditions. This is supported by the fact that dentation damage in ceramics: the median/radial crack system, J. the material contains a large amount of glassy phase, Am. Ceram. Soc., 63 (1980) 574-581. [3] B.R. Lawn, D.B. Marshall, G.R. Anstis and T.P. Dabbs, Fa- which is highly susceptible to stress corrosion in an tigue analysis of brittle materials using indentation flaws, Part I. aqueous environment. The fatigue parameters for this General theory, J. Mater. Sci., 16 (1981) 2846 2854. material system, therefore, can be obtained by using [4] E.R. Fuller, B.R. Lawn and R.F. Cook, Theory of fatigue for either static or cyclic fatigue testing. In terms of testing brittle flaws originating from residual stress concentrations, J. economy, however, static fatigue is preferred because of Am. Ceram. Soe., 66 (1983) 314 321. [5] S.R. Choi, J.E. Ritter and K. Jakus, Failure of glass with the much higher testing costs associated with cyclic subthreshold flaws, J. Am. Ceram. Soc., 73 (1990) 268-274. fatigue testing. [6] C.-K.J. Lin and D.F. Socie, Static and cyclic fatigue of alumina As mentioned before, the analytical solution of the h at high temperatures, J. ,4m. Ceram. Soe., 74 (1991) 1511 1518. ratio for other loading functions such as trapezoidal, [7] S. Lathabai, Y.-W. Mai and BR. Lawn, Cyclic fatigue behavior triangular and square wave forms can be easily ob- of an alumina ceramic with crack-resistance characteristics, J. Am. Ceram. Soc., 72 (1989) 1760 1763. tained for the natural flaw system because of their [8] S. Horibe and R. Hirahara, Cyclic fatigue of ceramic materials: mathematical simplicity [1,11]. However, the conven- influence of crack path and fatigue mechanisms, Acta Metall. tional numerical solution for these loading functions Mater., 39 (1991) 1309-1317. may not be simple for the indentation flaw system, [9] R.H. Dauskardt, R.R. James, J.R. Porter and R.O. Ritchie, since the functions, unlike the sinusoidal wave, are not Cyclic fatigue-crack growth in SiC whisker-reinforced alumina ceramic composite: long- and small-crack behavior, J. Am. Ce- continuous but discontinuous in nature. In this case, ram. Soc., 75 (1992) 759-771. each discontinuous periodic function should be con- [10] (a) J.E. Ritter, P.B. Oates, ER. Fuller and S.M. Wiederhorn, verted into a continuous periodic function so that nu- Proof testing of ceramics, Part I Experiment, J. Mater. Sci., 15 merical solution is feasible. This may be done by using (1980) 2275-2281. (b) E.R. Fuller, S.M. Wiederhorn, J.E. Ritter the Fourier analysis from which a discontinuous func- and P.B. Oates, Proof testing of ceramics, Part 2 Theory, J. Mater. Sci., 15 (1980) 2282-2295. tion can be approximated into a continuous function. [11] J.A. Salem and S.R. Choi, Bimonthly Progress Report, April/ May 1994, Ceramic Technology Project, Oak Ridge National Laboratory, Oak Ridge, TN. [12] T. Kawakubo and K. Komeya, Static and cyclic fatigue behavior 4. Conclusions of a sintered silicon nitride at room temperature, Metall. Trans., 70 (1987) 400-405. The ratio of static to cyclic fatigue life, or h ratio, [13] D.A. Krohn and D.P.H. Hasselman, Static and cyclic fatigue behavior of a polycrystalline alumina, J. Am. Ceram. Soc., 55 was obtained numerically with an emphasis on the (1972) 208-21 I. indentation flaw system subjected to sinusoidal loading [14] L. Ewart and S. Suresh, Crack propagation in ceramics under conditions. The h ratio decreases with increasing n and cyclic loads, J. Mater. Sei. Lett., 22 (1987) 1173-1192. decreasing R ratio. The solution provides a simple and [15] M.J. Reece, F. Guiu and M.F.R. Sammur, Cyclic fatigue crack quick methodology to predict the lifetime of one fatigue propagation in alumina under direct tension-compression load- ing, J. Am. Ceram. Soc., 72(1989) 348-352. condition from another for indentation flaw systems (as [16] R.H. Dauskardt, D.B. Marshall and R.O. Ritchie, Cyclic fa- well as the natural flaw system) when the governing tigue-crack propagation in magnesia-partially-stabilized zirconia failure mechanism is stress corrosion. The solution was ceramics, J. Am. Ceram. Soc., 73 (1990) 893-903. Instructions for Authors SUBMISSION OF PAPERS References must be listed on separate sheet(s) at the end of the paper Every reference appeanng in the text should be quoted in Manuscnpts for the mapn part of the journal and for the Letters the reference list, and wce versa When reference _smade to a Sect,on should be submitted as follows publication written by more than two authors _t_spreferable to For authors in Europe gwe only the first author's name =nthe text followed by "et aL" However, in the list of references the names and in_bals of all Editor in Chief authors must be gwen Professor Herbert Herman Three sets of figures should be submitted One set of line Department of Materia(s Science and Engineenng drawings should be tn aform suitable for reproduction, drawn tn State University of New York at Stony Brook Indian ink on drawing or tracing paper (letter hetght, 3 5mm) Long Island, NY 11794-2275, USA Alternatwely, such =llustrat_ons may be suppl_ed as high contrast, Fax: +1 {516) 632 8052 black-andwh_te glossy pnnts Duplicate original micrographs or should be provided wherever possible to facilitate the refereeing process Magndications should be _nd_cated by a ruled scale bar Professor Gernot Kostorz on the m_crograph Capttons to rllustrat_ons should be typed _n ETH Zunch sequence on aseparate page Institut fur Angewandte Physik All abbreviated terms must be defined when hrst used (both _n CH-8093 Zurich, Switzerland the abstract and in the text) and authors must express all Fax: +41 {1633) 1105 quant_ttes nnSI units, with other units =nparentheses ifdesired Authors in Japan please note that _nformat_on about how to have For authors rnJapan the Enghsh of your paper checked, corrected and _mproved Professor Masahiro Koiwa (before submission)is available from: Elsevier ScLence (Japan), Department of Metal Science and Technology 20-12 Yushima 3-chome, Bunkyo-ku, Tokyo 113; Tel (03) 3833- Faculty of Engineering 3821; Fax: /03) 3836-3064 Kyoto University Further information Yoshida-Honmachi, Sakyo-ku Kyoto 606-01, Japan All queshons arising after the acceptance of manuscripts, Fax: +81 (75)751 7844 especially those relating to proofs, should be directed to Elsevier For authors in North and South America and the rest of the Editorial Services, Mayfield House, 256 Banbury Road, Oxford OX2 7DH, UK (tel +44 1865 314900: lax +44 1865 314990) world Submission of electronic text Professor Herbert Herman USA The final text may be submitted on a 35 in or 525 _ndiskette (in addition to a hard copy with orig(nal figures) Double density (DD) or high density (HD) diskettes are acceptable, but must be Manuscripts formatted to their capacity before the files are copied on to them Three coptes should be submitted to the Editor, in double-spaced The mare text, list of references, tables and figure legends should typing on pages of A4 s_ze and with w=de margins (Letters be stored bnseparate text files with clearly identifiable file names should not exceed 2000 words and amaximum of 5 figures) All The format of these Ides depends on the word processor used tables and dlustrat_ons should bear a title or legend WordPerfect 51 is the most preferable but for other formats An abstract should accompany rewews, onginal papers and Let- please refer to the Instruchons to Authors booklet It _sessential ters It should present (preferably in 100-150 words; 50 words that the name and version of the wordprocessbng program, type of or less for Letters} a bnef and factual account of the contents computer on which the text was prepared, and format of the text and conclusions of the paper, and an ind_cabon of the relevance files are clearly indicated. of new material The final manuscript may contain last minute corrections which are References should be indicated by numerals in square brackets, not included in the electronic text but such corrections must be introduced consecutively and appropriately _nthe text clearly marked on the hard copy ©1996 Elsevier Science All rights reserved 0921-5093 96 $1500 No part of th=s publication may be reproduced, stored _n a retrieval system or transmitted _nany form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the pnor written permission of the pubhsher, Elsewer Science SA, PC Box 564, 1001 Lausanne, Switzerland Submission of an article for publication =mphes the transfer of the copyright from the author(s) to the pubhsher and entails the author(s) _rrevocable and exclusive authorization of the pubhsher to collect any sums or considerations for copying or reproduction payable by third part=es Upon acceptance of an article by the journal, tile author(s) wifl be asked to transfer copybght of the article to the pubhsher Th_stransfer w=ll ensure the widest possible d_ssem_nat_on of information For Material Subject to US Copyright Law Special regulations for readers _nthe USA This iournal has been registered withlhe Copyr_ghlClearance Center, Inc Consent _sgwen for copymgofart_clesfor pe_sonalorinternal use, or for the personal use of spec_hc chents Th_s consent _sgwen on the condition that ]he cop_e_ pavsthrough the Cen_er the per-copy fee stated _nthe code on the first page of each article for copying beyond that permqted by Sections 107 or 108 of tile US Copyright Law The appropriate fee should be forwarded _,,,_thacopy of the first page of the ar,icle to the Copyright Clearance Center, Inc, 222 Rosewood Dnve, Danvers, MA01923, USA If nocode appears _rla_]art_cle, theat_thor has not gwen broad consent to copy and perm_ss_on to copy must be oblainedd_rectly from the author The fee mdicatedonttle first page of anart_cle_nthis issue _vil!applyretroactwelytoai!arhcles published _nthe iournal, regardless of the year of pubhcat_on Th,s consent does not e×tend to other k_nds of copying, suciq as for general d_stribut_on, resale, advertising and promot_orr purposes, or for creattng new collectwe works Specia written perrmss_on must be obtatned from the pubhsher for such copying No respons_bihty _sassumed by the Pubhsher fo_ a'w _r]jury and of dart,age to persons or property as a matter of products hab_hty, negl_gence or otb,erw_se, or from any use or operation otany n_e':hods, products, H]strucbons or ,(leas contained H_the material here_n The paper used ,n th,s pub!,cat,on meets the requ,rements of ANSI NISO 239 48 1992 ,Permanence of Pape:l Pr,nted ,nThe Netherlands

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.