The fracture and fatigue FRACTURE AND FATIGUE BEHAVIOR OF

$Page 1: The fracture and fatigue FRACTURE AND FATIGUE BEHAVIOR OF$
Volume 50, Issue 4, 2014International Journal of Powder Metallurgy 11

INTRODUCTIONCemented carbides, also called hardmetals, are a group of powder metal-

lurgy (PM) liquid-phase-sintered materials consisting of brittle refractory car-bides of the transition metals (e.g., WC, TiC, TaC) embedded in a metallicmatrix.1 The success of this ceramic–metal composite resides in the combi-nation of the hardness and wear resistance of the ceramic particles with thetoughness of the metallic phase. It makes cemented carbides the materials ofchoice in several engineering and tooling applications, such as metal cutting,mining, rock drilling, metal forming, structural components, and wear-resist-ant parts.2

The exceptional toughness level exhibited by cemented carbides is related tohighly effective toughening by constrained deformation of the ductile phase.3–6

Ductile-phase (cobalt ligaments) reinforcement of a brittle matrix (tungstencarbide network) is a prime example of toughening mechanisms that act in thecrack wake to screen the crack tip from the far-field driving force.7 From afracture viewpoint, such toughening has proven to be a successful microstruc-tural design strategy in brittle-like materials because it implies the existenceof a crack-growth-resistance-curve (R-curve) behavior that imparts damagetolerance, and thus improved in-service reliability to the corresponding struc-tural components.8 Thus, the first step of crack propagation in cemented car-bides is the formation of a crack from a microstructural heterogeneity thatsubsequently advances continuously in the brittle phase and circumvents theductile regions leaving bridging ligaments across the crack faces.5 In this way,

The fracture and fatiguephenomena in WC-cobaltcemented carbides (hard-metals) have been subjectsextensively investigated inthe last 30 years. Fromthese studies, it is wellestablished that the metallicbinder phase plays a keyrole as the toughening andfatigue-susceptible agent inthese materials, as its effec-tive ductility is critical fordefining crack-growth resist-ance and cyclic-induceddegradation. However,experimental proof of therole of toughening andfatigue micromechanismshas usually been presentedon the basis of post-failurefractographic examination.In this work, the fractureand fatigue behavior of WC-cobalt is investigatedand a 3D characterization ofcrack–microstructure inter-action during stable crackgrowth in hardmetals is car-ried out in order to gain abetter understanding of thefailure processes in cement-ed carbides under monoton-ic and cyclic loads. In doingso, focused ion beam/fieldemission scanning electronmicroscopy (FIB/FESEM),3D tomography, and imag-ing reconstruction are combined with systematicmechanical and indentationprotocols for assessingcrack-extension behavior of cemented carbides.Experimental findings clearly highlight existing differences regarding failuremechanisms operativeunder monotonic and cyclicloads, and provide new andinteresting insights forunderstanding them.

BEST PAPERAWARD

*PhD Fellow, **Tenured Assistant Professor, ***PhD Fellow, ****MSc Engineer, ******Full Professor, CIEFMA, Universitat Politècnica deCatalunya, Barcelona, Spain, 08028; E-mail: [email protected], *****Senior Engineer in R&D/Technology, Sandvik Hyperion,Coventry, United Kingdom CV40XG

FRACTURE ANDFATIGUE BEHAVIOR OFCEMENTED CARBIDES:3D FOCUSED ION BEAM TOMOGRAPHYOF CRACK–MICROSTRUCTUREINTERACTIONSJose María Tarragó,* Emilio Jimenez-Piqué,** Miquel Turón-Viñas,*** Lorenzo Rivero,**** Ihsan Al-Dawery,*****Ludwig Schneider***** and Luis Llanes******


a multiligament zone (~2–4 ligaments) is developed inthe direction of crack propagation.4,5 As the crack prop-agates, the binder ligaments elongate and microcavitiesare formed inside the ligament to maintain volume con-stancy. Using finite element calculations it was demon-strated that binder sites exposed to a high local plasticstrain and stress triaxiality are preferred zones formicrovoid nucleation.9 First, a void will be followed byothers along the plane linking the adjacent cracks inthe carbide phase and finally the ligament fractures byvoid growth and coalescence. Fracture along the car-bide–binder interface proceeds in a similar manner tothat in the binder, i.e., by the nucleation, growth, andcoalescence of microcavities. However, shallow andclosely spaced microvoids are evidenced when the crackruns parallel and close to the binder–carbide interface.Fischmeister et al.9 attributed the formation of such afine dimple structure to the fulfillment of high-stresstriaxiality conditions in the binder zone close to the car-bide interface.

Cemented carbides, despite the plasticity developedby the binder ligaments, exhibit brittle-like behavior.This fracture mode, such as in other brittle materials,is governed by unstable propagation of flaws that maybe inherent to processing, or induced in the formingprocess or under service conditions.10,11 Thus,strength and reliability of hardmetals are dependent onthe size, geometry, and distribution of existing flaws.Following this idea, linear elastic fracture mechanics(LEFM) is recognized as an acceptable theoretical basisto describe and rationalize the fracture behavior ofcemented carbides.12,13 On the other hand, a statisti-cal tool is required to rationalize the rupture strengthof this brittle material, as it is dependent on the sizeand geometry of the flaws. To this end, Weibull statis-tics are commonly employed to evaluate not only rup-ture strength, but also the reliability of the material.

Meanwhile, as is the case for other brittle-like com-posite systems (e.g., ceramic and intermetallic-basematerials) where crack-tip shielding mechanisms pre-vail,14 the susceptibility of hardmetals to mechanicaldegradation under cyclic loading is well estab-lished.15–19 Schleinkofer et al. documented a strongstrength degradation of cemented carbides undercyclic loads that predominantly occur in the ductilebinder phase.17,18,20 They established that subcriticalcrack growth is the controlling stage for fatigue failurein cemented carbides, not crack nucleation.17,20,21 Onthe other hand, taking into account this secondassumption and that fracture rupture behavior ofcemented carbides has been extensively rationalizedwithin the LEFM framework, Torres et al. proposed thefatigue-crack-growth threshold as the effective tough-ness under cyclic loading.19 Experimental validation

for such an approach was presented for a series of WC-cobalt hardmetal grades.22 Moreover, the results citeda strong influence of microstructure on the fatigue sen-sitivity of hardmetals, depending on the compromisingrole of the metallic binder, as both a toughening andfatigue-susceptible agent.23 These investigations alsopoint out a similarity to the Paris-Erdogan law forfatigue-crack-growth kinetics reported for structuralceramics and intermetallics. Binder fatigue degrada-tion, also observed in the experimental appliedstress–fatigue life (S–N) data published by Sailer etal.,24 has been rationalized on the basis of fatigue-induced accumulation of the FCC to HCP phase trans-formation within the cobalt binder.17 However, most ofthe experimental support validating the above-referredfailure scenarios are either indirect, through clear evi-dences of different fractographic features associatedwith fracture and fatigue,4,17,23 or rather limited (butdirect), by means of transmission electron microscopy(TEM) of slices (local and small areas) in regionsaround crack tips.17,21,25

Over the past few years, new advanced characteriza-tion techniques have been successfully implementedfor microstructure, surface modification and deforma-tion/damage characterization in various areas of mate-rials research. Among them, the focused ion beam(FIB) technique has shown to be extremely versatile forovercoming many of the experimental limitations/diffi-culties ascribed to more conventional approaches,such as scanning electron microscopy (SEM) and/orTEM, atomic force microscopy, and X-ray diffraction.Fruitful examples of its implementation for quantifyingmicrostructure, as well as evaluating tribological andmechanical phenomena in cemented carbides, arecited by Cairney et al.,26 Beste et al.,27 Mingard etal.,28 and Gant et al.29

It is the purpose of this study to assess the fractureand fatigue behavior of a WC-cobalt cemented carbideand to use the FIB technique for bringing new insightsinto the toughening and fatigue micromechanismsoperative in hardmetals when subjected to monotonic(toughness) and cyclic (fatigue) loading conditions. Indoing so, the main focus will be to document crack–microstructure interactions at regions close to the tipof cracks arrested after stable growth, and to discussresults on the basis of existing knowledge on fractureand fatigue phenomena for these materials.

EXPERIMENTAL ASPECTSThe material studied is a medium-grained WC-

cobalt cemented carbide supplied by SandvikHyperion. The key microstructural parameters (bindercontent (w/o), mean grain size (dWC), carbide contigui-ty (CWC), and binder mean free path (λbinder)) are listed

FRACTURE AND FATIGUE BEHAVIOR OF CEMENTED CARBIDES: 3D FOCUSED ION BEAM TOMOGRAPHY OF CRACK–MICROSTRUCTURE INTERACTIONS


in Table I. Mean grain size was measured following thelinear intercept method by means of FESEM, using aJEOL-7001F unit. Carbide contiguity and binder meanfree path were deduced from best-fit equations,attained after compilation and analysis of data pub-lished in the literature,3,5 on the basis of empiricalrelationships given by Roebuck and Almond,30 butextending them to include the influence of carbidesize.12,13

Mechanical characterization includes hardness(HV30), flexural strength (σr), fracture toughness (KIc)and fatigue-crack-growth (FCG) parameters. Hardnesswas measured using a Vickers diamond pyramidalindenter under a load of 294N. In all the other cases,testing was conducted using a four-point bending fullyarticulated test jig with inner and outer spans of 20mm and 40 mm, respectively. Flexural strength testswere performed on an Instron 8511 servohydraulicmachine at room temperature. At least 15 specimens(45 mm × 4 mm × 3 mm) were tested per grade. Thesurface (which was later subjected to the maximumtensile loads) was polished to a mirror-like finish andthe edges were chamfered to reduce their effect asstress raisers. Results were analyzed using Weibullstatistics.

Fracture toughness and FCG parameters were deter-mined using 45 mm × 10 mm × 5 mm single-edge pre-cracked-notch beam (SEPNB) specimens with a notchlength-to-specimen width ratio of 0.3. Compressivecyclic loads were induced in the notched beams tonucleate a sharp crack; details are given elsewhere.31

The sides of the SEPNB specimens were polished to fol-low stable crack growth using a high-resolution confo-cal microscope. Fracture toughness was determined bytesting the SEPNB specimens to failure at stress-intensity-factor load rates ~2 MPa√m/s. FCG behaviorwas assessed for two different load-ratio (R) values, 0.1and 0.5, using a Rumul Testronic resonant testingmachine at load frequencies ~150 Hz.

Effective evaluation of crack–microstructure interac-tions requires a procedure for introducing sharp pre-cracks into specimens free of residual stresses. To thisend, a sharp indentation is suggested as a simple andpractical precracking technique, since hardmetals arebrittle enough.32 Tomography characterization imple-mented in this study involves automated serial section-ing using FIB and imaging of each milled surface with


FESEM. However, precise and detailed imaging ofregions close to the tips of cracks, just as they developafter unloading the indenter, may become a difficulttask. The main reason behind it is the intrinsic inden-tation residual stress field. Thus, as material isremoved through FIB milling, the effective stress stateat the crack tip changes and the latter gets displaced.This experimental inconvenience may be overcome bythe relief of indentation residual stresses before FIB.This was approached by controlled crack extensionunder cyclic loads. In this way much longer cracks canbe induced without any indentation residual stressesacting at their tips.

One of the purposes of this investigation is to docu-ment crack–microstructure interaction during stablecrack growth under both monotonic and cyclic loads. Itthen requires the arresting of cracks after stablegrowth under each of the loading conditions. Indeed,the residual stress-relief procedure could be taken asthe final step for assessing fatigue micromechanisms.However, to attain similar cracks under monotonicloading demands one further step: testing to failure ofspecimens containing multiple indentations.33

Strength-indentation tests were conducted on speci-mens (4 mm × 3 mm × 45 mm) under four-point bend-ing (outer and inner spans of 40 and 20 mm,respectively). Four controlled flaw patterns were pro-duced, 2 mm apart, in the center of the prospectivetensile surface of each flexure specimen by applying aload of 490N using a Vickers diamond pyramid. Carewas taken to orient one set of the correspondingPalmqvist cracks of each indentation flaw parallel tothe cross section of the specimen where the prospec-tive rupture would occur. Indentation residual stress-es for all the controlled flaws were relieved bysubjecting the specimens to tensile cyclic bending (loadratio of 0.1 and frequency 10 Hz) in order to inducestable crack growth. Finally, fracture of the crackedspecimens was induced in flexure, under either monot-onic or cyclic loads. All the specimens ruptured at oneof the controlled flaws. Stably grown and arrestedcracks remained for the other three indentations. FIBserial sectioning was then carried out in regions closeto the tips of these surviving cracks. A schematic of theprocess is shown in Figure 1.

In order to document crack–microstructure interac-tion during stable crack extension, combinedFIB/FESEM (Zeiss Neon 40) was used. Before ionmilling, a thin protective platinum layer was depositedon small areas of interest, corresponding to regionsclose to tips of arrested cracks, along surface crackpaths. U-shaped trenches with one cross-sectionalsurface (perpendicular to the crack path and to thespecimen surface) were produced by FIB (Figure 2(a)).

TABLE I. MICROSTRUCTURAL PARAMETERS

Binder dWC CWC λbinder(w/o) (μm) �(μm)

11% Co 1.12 ±0.71 0.38 ±0.07 0.42 ±0.28


agreement with values estimated from a direct imple-mentation of the basic LEFM equation relatingstrength, toughness, and critical flaw size (Table II), byconsidering defects as surface semicircular flaws. Thissupports the use of LEFM for rationalizing the fracturebehavior of the hardmetal grade studied.

Fatigue BehaviorFCG rates are plotted against the range and the

maximum applied stress intensity factor, ∆K (Figure4(a)) and Kmax (Figure 4(b)), respectively, for the twoload ratios studied. As previously reported for WC-cobalt cemented carbides,19,23 the hardmetalgrade under consideration exhibits: (i) a large-powerdependence of FCG rates on ∆K, and (ii) subcriticalcrack growth at ∆K values much lower than the frac-ture toughness. Also, a pronounced load-ratio effect isobserved in the dependence of FCG on ∆K. However, as


Subsequently a series of crack–microstructure interac-tions was obtained by periodic removal of the materialby FIB, within the U-shaped crater parallel to thecross-sectional surface, using automated software(Figure 2(b)). A 15 µm × 15 µm × 10 µm3 volume wasion milled and ~600 images were obtained.

The 3D reconstruction of the images was carried outusing commercial Avizo software. The images werealigned by applying image processing techniques inorder to correct, equalize, and differentiate the greylevel of the three phases (WC, cobalt, and cracks). Thenext step consisted of segmenting the three phases toreconstruct the 3D volume.

RESULTS AND DISCUSSIONHardness, Strength, and Toughness

Hardness, flexural strength, Weibull modulus, andfracture toughness of the material are listed in Table II.The dispersion in flexural strength is relatively smalland, accordingly, the corresponding Weibull analysisyields high values, indicative of high reliability from astructural viewpoint (Figure 3(a)). Fractographic exam-ination reveals that critical defects are abnormalcoarse grains and carbide agglomerates (Figure 3(b))with an equivalent diameter (2acr) ~20–30 μm. It is in

Figure 1. Schematic representation of process to generate stable cracks for subsequent sectioning by FIB/FESEM

Figure 2. FESEM images of trenches generated by FIB: (a) previous to sequential milling and (b) after completion of sequential milling, around the regions of interest

(a) (b)

TABLE II: HARDNESS, STRENGTH, AND FRACTURE MECHANICSPARAMETERS FOR WC-COBALT HARDMETAL

HV30 Flexural Strength Weibull KIc Estimated 2acr(GPa) (MPa) Modulus (MPa�√m) (μm)

12.8 ±0.2 3,101 ±102 36 13.9 ±0.3 31 ±2


observed for other brittle materials, R effects are large-ly reduced when plotting FCG against Kmax. This is anindication of the predominance of static failure overcyclic failure modes.16,23

The FCG–fatigue life relationship proposed and vali-dated by Torres et al. for WC-cobalt cemented car-bides19 has been extended for the hardmetal studied.In doing so, a classical approach on the basis of fatiguelimit, within an infinite-life framework, and FCGthreshold (Kth) is implemented by defining the latter asthe effective toughness under fatigue for a given criti-cal flaw size. Thus, the fatigue limit (σf ) is deducedfrom the stress-intensity factor threshold of a smallnon-propagating crack emanating from a defect of crit-ical size (2acr), according to a relationship of the form:

Kthσf = Y ____ (1)√acr

Hence, the fundamental LEFM correlation betweenstrength, stress-intensity factor, and defect size alsoapplies to natural defects in cemented carbides. Thisassertion may be made considering that: (i) the size of

the critical flaws is larger than both dWC and √binder; (ii)plasticity is confined to process zone ahead of the cracktip; and (iii) process zones governing fracture (multiliga-ment zones behind the crack tip) extends over a rela-tively short distance (~five ligaments).19,34 Thus,fatigue-limit values can be estimated from the relationgiven by equation (2), under the assumption that flawscontrolling strength have the same size, geometry, anddistribution under monotonic and cyclic loading.

Kthσf = σr____ (2)KIc

In an attempt to validate the estimated fatigue limit,an experimental study was conducted using 15 sam-ples, following an up-and-down (staircase) loadingregime for fatigue testing (Figure 5(a)), and defining“infinite fatigue life” at 106 cycles. Predicted and exper-imentally determined fatigue limits are listed in TableIII. The FCG threshold–fatigue limit correlation is vali-dated by the excellent agreement attained betweenthem. Furthermore, the fractographic examinationconducted on failed specimens reveals that size, geom-


Figure 4. da/dN behavior vs. (a) ∆K, and (b) Kmax, for the two load ratios studied

(a) (b)

Figure 3. (a) Fracture Weibulldistribution and (b) agglomerateof carbides acting as a criticalflaw for fracture

(a) (b)


etry, and the nature of the initial critical defects aresimilar under both monotonic- and cyclic-loading con-ditions (Figure 5(b)). However, under the application ofcyclic loads, the defects grow until they reach a criticalsize and fracture is unstable.

Crack–Microstructure InteractionsAfter failure, the fracture surfaces were examined

using FESEM. Clear differences are evidenced whencomparing fractographic features corresponding to sta-ble crack growth (Figure 6(a)) and unstable crack growth

(Figure 6(b)). While in the former, “step-like” fatiguedamage features are discerned within the metal binder,in the latter well-defined dimples are evident, suggestinga ductile fracture mechanism. Fracture under cyclicloading in the cobalt binder appears to follow a faceted,crystallographic fracture mode, as can be appreciated bythe sharp angular facets localized within broken binderregions. With the purpose of documenting and under-standing the crack-growth processes under monotonicand cyclic loads, a tomographic reconstruction of theprocess zone at the crack tip has been performed bymeans of the FIB/FESEM technique.

Stable Crack Growth Under Monotonic LoadingFigure 7 shows crack-microstructure details along

the path of an arrested crack, after stable growthunder monotonic loading. From these preliminary find-ings, many of the conclusions drawn by Sigl, Exnerand Fischmeister4,5 are validated, namely, (1) continu-


TABLE III: FATIGUE-CRACK-GROWTH THRESHOLD ANDPREDICTED AND EXPERIMENTALLY DETERMINED FATIGUE-LIMITVALUES IN TERMS OF MAXIMUM APPLIED STRESS

FCG Threshold Predicted Fatigue Limit Experimental Fatigue(MPa�m1/2) (MPa) Limit (MPa)

7.6 ±0.2 1,696 ±80 1,632 ±130

Figure 5. (a) Up-and-down (staircase) fatigue test to determine the mean fatigue limit for the WC-nickel cemented carbide, and (b) example of critical defectgrowth under cyclic loading

(a) (b)

Figure 6. FESEM images corresponding to (a) stable crack growth under cyclic loads (R = 0.1) and (b) unstable crack growth under monotonic loading for WC-cobalt hardmetal. Fatigue facets in (a) and ductile dimples in (b) are evident in the metallic phase

(a) (b)


ous cracks enclose cobalt regions (ligaments) that elon-gate during crack opening by void formation with local-ized plastic deformation at the bridges between thevoids, (2) ligaments finally fail by void coalescence, (3)the mean depth of the plastic zone is always smallerthan the mean-free path in the binder, and (4) partitionof crack paths between binder and carbide does notcorrespond to the volume fraction but significantlyfavors the ductile binder. Additionally, local bluntingeffects as crack fronts end within the binder (asinvoked by McVeigh and Liu,35 although at a higher-length scale) and the role of carbide corners as stressrisers are clearly discerned.

In conclusion, the multiligament zone may be unam-biguously established as the primary foundation for


Figure 7. Stable crack growth under monotonic loading: FESEM micrographs showing crack–microstructure interactions. Binder ligaments are clearly discerned in these images, corresponding to serial sections obtained by FIB/FESEM tomography

Figure 8. 3D reconstruction of process zone for stable crack growth under monotonic loading. Light blue and dark blue correspond to the WC skeleton and thecobalt phase, respectively; red represents the cracked region. A volume of 6 μm (X axis) x 10 μm (Y axis) x 7.5 μm (Z axis) was reconstructed

Figure 9. Microvoid formation and growth within a cobalt ligament

rationalizing R-curve behavior and exceptional fracturetoughness of cemented carbides. Within this frame-work, a description and understanding of microstruc-tural effects on fracture toughness, R-curvecharacteristics, strength variability, and damage toler-ance for hardmetals are also validated. In Figure 8 the3D reconstruction of the microstructure containing thecrack propagated under monotonic loading is shown.The light blue corresponds to the WC carbides, thedark blue corresponds to the cobalt binder, and the redarea corresponds to the crack path. An examination ofthe reconstruction permits a clear visualization andunderstanding of the fracture process (microvoidnucleation and growth). A clear example of the processis presented in Figure 9.


Unstable Crack Growth Under Cyclic LoadingCrack–microstructure interaction during stable

crack growth under cyclic loading is shown in Figure10. Different from the failure scenario discerned undermonotonic loading, fatigue micromechanisms incemented carbides are less defined and understood.Taking this into consideration, images attainedthrough FIB “slice and view” shed further light for pro-posing and/or validating specific failure micromech-anisms: (1) subcritical fatigue-crack growth is morepredominantly located in the ductile binder phase thanunder monotonic loading;9 (2) fatigue-crack extensionfollows crystallographic-like paths (steps); and (3)crack-arrest phenomena within binder ligamentsappear to be normal, particularly in regions far from

microstructure-related stress risers. Although thecrystallographic nature of the steps is evident, theirintrinsic origin as related to shear bands, stackingfaults and/or twins resulting from FCC-to-HCP phasetransformation is not clear. The fact that step-likemarkings on fatigue-fracture surfaces have beenobserved not only in cobalt-base hardmetals but alsoin nickel- and cobalt/nickel-base grades,36,37 makesclear that fatigue susceptibility of cemented carbidesgoes beyond changes in deformation mode induced bythe FCC-to-HCP transformation, well established inWC-cobalt grades.38 As in monotonic loading, the 3Dreconstructed images of the whole volume and theindividual phases is presented in Figure 11. It is inter-esting to note that under cyclic loading the crack prop-


Figure 10. Stable crack growth under cyclic loading: FESEM micrographs showing crack–microstructure interactions. A crystallographic-like fracture path with-in the binder is evidenced in these images, corresponding to serial sections obtained by FIB/FESEM tomography

Figure 12. Crack propagation within a binder pool under the application of cyclic loads

Figure 11. 3D reconstruction of process zone for stable crack growth under monotonic loading. Light and dark blue correspond to the WC skeleton and thecobalt phase, respectively; red represents the cracked region. A volume of 10 μm (X axis) x 10 μm (Y axis) x 8 μm (Z axis) was reconstructed


agates through the ductile metallic phase and bridgingligaments are not formed. Figure 12 shows the fatiguefailure process in a binder pool.

CONCLUSIONSThe fracture and fatigue behavior of a WC-cobalt

cemented carbide has been investigated. A detailedstudy of crack–microstructure interactions during sta-ble crack growth under monotonic and cyclic loads hasalso been conducted by the complementary use ofmechanical and indentation testing protocols andFIB/FESEM tomography. From the results the follow-ing conclusions may be drawn:1. The fracture behavior of the WC-cobalt cemented

carbide has been successfully rationalized usingLEFM, on the basis of a reliable assessment of frac-ture toughness as well as the size and geometry ofcritical flaws.

2. As previously reported for cemented carbides, theWC-cobalt grade studied exhibited a large-powerdependence of FCG rates on ∆K, subcritical crackgrowth at ∆K values lower than the fracture tough-ness, and pronounced load-ratio effects in FCG–∆Kcurves. R effects are largely reduced when plottingFCG against Kmax. This indicates the predominanceof static over dynamic failure modes.

3. Fatigue behavior (“infinite fatigue life”) can berationalized by a fatigue mechanics approach con-sidering the initiation of subcritical crack growth asthe control parameter in fatigue failure (Kth is theeffective toughness under cyclic loads).

4. The key role played by the metallic binder phase asthe toughening and fatigue susceptible constituentin cemented carbides is validated. Unequivocalproof of the multiligament zone as the foundationfor understanding toughness and R-curve behaviorin hardmetals is provided. Fatigue susceptibility ofthese materials is associated with inhibition ofsuch toughening mechanism under cyclic loading.In this case, stable crack growth follows a crystal-lographic path through the metallic binder, distinctfrom the typical ductile failure mode found undermonotonic loading.

ACKNOWLEDGEMENTSThis work was supported financially by the Spanish

Ministerio de Economía y Competitividad (GrantMAT2012-34602). The authors are grateful to T.Trifonov (CRnE) for his technical assistance duringFIB/FESEM research. Additionally, J.M. Tarragóacknowledges a PhD scholarship received from the col-laborative Industry–University program betweenSandvik Hyperion and the Universitat Politècnica deCatalunya.

REFERENCES1. H.E. Exner, “Physical and Chemical Nature of Cemented

Carbides”, Int. Met. Rev., 1979, vol. 24, pp. 149–173.2. L.J. Prakash, “Fundamentals and General Applications of

Hardmetals”, Comprehensive Hard Materials, edited by V.K.Sarin, D. Mari and L. Llanes, Elsevier, 2014; vol. 1, ch. 1.02,pp. 29–90.

3. R.K. Viswanadham, T.S. Sun, E.F. Drake and J. Peck,“Quantitative Fractography of WC-Co Cermets by AugerSpectroscopy”, J. Mat. Sci., 1981, vol. 16, no. 4, pp.1,029–1,038.

4. L.S. Sigl and H.E. Exner, “Experimental Study of theMechanics of Fracture in WC-Co Alloys”, Metall. Mat. Trans. A,1987, vol. 18, no. 7, pp. 1,299–1,308.

5. L.S. Sigl and H.F. Fischmeister, “On the Fracture Toughness ofCemented Carbides”, Acta Metall. Mater., 1988, vol. 36, no. 4,pp. 887–897.

6. P.A. Mataga, “Deformation of Crack-bridging DuctileReinforcements in Toughened Brittle Materials”, Acta Mater.,1989, vol. 37, no. 12, pp. 3,349–3,359.

7. R.O. Ritchie, “Mechanisms of Fatigue Crack Propagation inMetals, Ceramics and Composites: Role of Crack Tip Shielding”,Mater. Sci. Eng. A, 1988, vol. 103, no. 1, pp. 15–28.

8. A.G. Evans, “Perspective on the Development of High-Toughness Ceramics”, J. Am. Ceram. Soc., 1990, vol. 73, no. 2,pp. 187–206.

9. H.F. Fischmeister, S. Schmauder and L.S. Sigl, “Finite ElementModeling of Crack Propagation in WC-Co Hard Metals”, Mat.Sci. Eng. A, 1988, vol. 105/106, part 2, pp. 305–311.

10. H. Suzuki and K. Hayashi, “The Strength of WC-Co CementedCarbide in Relation to Structural Defects”, Trans. Japan Ins.Met., 1975, vol. 16, no. 6, pp. 353–360.

11. B. Casas, Y. Torres and L. Llanes, “Fracture and FatigueBehavior of Electrical-discharge Machined Cemented Carbides”,Int. J. Refract. Met. Hard Mater., 2006, vol. 24, no. 1–2, pp.162–167.

12. Y. Torres, “Comportamiento a Fractura y Fatiga de CarburosCementados WC-Co”, 2002, PhD Thesis, Universitat Politècnicade Catalunya, Catalunya, Spain.

13. D. Coureaux, “Comportamiento Mecánico de CarburosCementados WC-Co: Influencia de la Microestructura en laResistencia a la Fractura, la Sensibilidad a la Fatiga y laTolerancia al Daño Inducido Bajo Solicitaciones de Contacto”,2012, PhD Thesis, Universitat Politècnica de Catalunya,Catalunya, Spain.

14. R.O. Ritchie, “Mechanisms of Fatigue-crack Propagation inDuctile and Brittle Solids”, Inter. J. of Fract., 1999, vol. 100, no.1, pp. 55–83.

15. W. Dawihl, “Die Wissenschaftlichen und TechnischenGrudlagen der Pulvermetallurgic und IihrerAnwendungsbereiche”, Stahl u. Eisen, 1941, vol. 61, pp.909–919.

16. P.R. Fry and G.G. Garret, “Fatigue Crack Growth Behaviour ofTungsten Carbide-cobalt Hardmetals”, J. Mater. Sci., 1988, vol.23, no. 7, pp. 2,325–2,338.

17. U. Schleinkofer, H.G. Sockel, K. Görting and W. Heinrich,“Microstructural Processes During Subcritical Crack Growth inHard Metals and Cermets Under Cyclic Loads”, Mater. Sci. Eng.A, 1996, vol. 209, no. 1–2, pp. 103–110.

18. U. Schleinkofer, H.G. Sockel, K. Görting and W. Heinrich,“Fatigue of Hard Metals and Cermets”, Mater. Sci. Eng. A, 1996,vol. 209, no. 1–2, pp. 313–317.

19. Y. Torres, M. Anglada and L. Llanes, “Fatigue Mechanics ofWC-Co Cemented Carbides”, Int. J. Refract. Met. Hard Mater.,



2001, vol. 19, no. 4–6, pp. 341–348.20. U. Schleinkofer, H.G. Sockel, K. Görting and W. Heinrich,

“Fatigue of Hard Metals and Cermets—New Results and aBetter Understanding”, Int. J. Refract. Met. Hard Mater., 1997,vol. 15, no. 1–3, pp. 103–112.

21. S. Kursawe, P. Pott, H.G. Sockel, W. Heinrich and M. Wolf, “Onthe Influence of Binder Content and Binder Composition on theMechanical Properties of Hardmetals”, Int. J. Refract. Met. HardMater., 2001, vol. 19, no. 4–6, pp. 335–340.

22. Y. Torres, M. Anglada and L. Llanes, “Fatigue Limit—FatigueCrack Growth Threshold Correlation for Hardmetals: Influenceof Microstructure”, Proc. 8th Int. Fatigue Congress, compiled byA.F. Blom, EMAS Publishing, Stockholm, Sweden, 2002, pp.1,171–1,178.

23. L. Llanes, Y. Torres and M. Anglada, “On the Fatigue CrackGrowth Behavior of WC–Co Cemented Carbides: KineticsDescription, Microstructural Effects and Fatigue Sensitivity”,Acta Mater., 2002, vol. 50, no. 9, pp. 2,381–2,393.

24. T. Sailer, M. Herr, H.G. Sockel, R. Schulte, H. Feld and L.J.Prakash, “Microstructure and Mechanical Properties ofUltrafine-grained Hardmetals”, Int. J. Refract. Met. Hard Mater.,2001, vol. 19, no. 4–6, pp. 553–559.

25. G. Erling, S. Kursawe, S. Luyck and H.G. Sockel, “Stable andUnstable Fracture Surface Features in WC-Co”, J. Mat. Sci.Lett., 2000, vol. 19, no. 5, pp. 437–438.

26. J.M. Cairney, P.R. Munroe and J.H. Schneibel, “Examination ofFracture Surfaces Using Focused Ion Beam Milling”, ActaMetall., 2000, vol. 42, no. 5, pp. 473–478.

27. U. Beste, E. Coronel and S. Jacobson, “Wear Induced MaterialModifications of Cemented Carbide Rock Drill Buttons”, Int. J.Refract. Met. Hard Mater., 2006, vol. 24, no. 1–2, pp. 168–176.

28. K.P. Mingard, H.G. Jones, M.G. Gee, B. Roebuck, A. Gholinia,B. Winiarski and P. Withers, “3D Imaging of Structures in Bulkand Surface Modified WC-Co Hardmetals”, Proc. EuroPM2012,EPMA, Basel, Switzerland, 2012, vol.2, pp. 155–160.

29. A.J. Gant, M.G. Gee, D.D. Gohil, H.G. Jones and L.P. Orkney,

“Use of FIB/SEM to Assess the Tribo-corrosion of WC/CoHardmetals in Model Single Point Abrasion Experiments”,Tribol. Int., 2013, vol. 68, pp. 56–66.

30. B. Roebuck and E.A. Almond, “Deformation and FractureProcesses and the Physical Metallurgy of WC-Co Hardmetals”,Int. Mater. Reviews, 1988, vol. 33, no. 2, pp. 90–110.

31. Y. Torres, D. Casellas, M. Anglada and L. Llanes, “FractureToughness Evaluation of Hardmetals: Influence of TestingProcedure”, Int. J. Refract. Met. Hard Mater., 2001, vol. 19, no.1, pp. 27–34.

32. D.K. Shetty, I.G. Wright, P.N. Mincer and A.H. Clauer,“Indentation Fracture of WC-Co Cermets,” J. Mat. Sci., 1985,vol. 20, no. 5, pp. 1,863–1,882.

33. R.F. Krause, “Rising Fracture Toughness from the BendingStrength of Indented Alumina Beams”, J. Am. Ceram. Soc.,1988, vol. 71, no. 5, pp. 338–343.

34. Y. Torres, R. Bermejo, L. Llanes and M. Anglada, “Influence ofNotch Radius and R-curve Behaviour on the FractureToughness Evaluation of WC–Co Cemented Carbides,” Eng.Fract. Mech., 2008, vol. 75, no. 15, pp. 4,422–4,430.

35. C. McVeigh and W. K. Liu, “Multiresolution Modeling of DuctileReinforced Brittle Composites,” J. Mech. Phys. Solids, 2009, vol.57, no. 2, pp. 244–267.

36. J.M. Tarragó, C. Ferrari, B. Reig, D. Coureaux, L. Schneiderand L. Llanes, “Mechanics and Mechanisms of Fatigue in a WC-Ni Hardmetal and a Comparative Study with Respect to WC-CoHardmetals”, Int. J. Fatigue, 2014, in press, DOI: 10.1016/j.ijfatigue.2014.09.011.

37. J.M. Tarragó, J.J. Roa, V. Valle, J.M. Marshall and L. Llanes,“Fracture and Fatigue Behavior of WC–Co and WC–CoNiCemented Carbides”, Int. J. Refract. Met. Hard Mater., 2014, inpress, DOI: 10.1016/j.ijrmhm.2014.07.027.

38. V. K. Sarin and T. Johannesson, “On the Deformation of WC-Co Cemented Carbides”, Met. Sci., vol. 9, no. 1, 1975, pp.472–476.


ijpm

Documents

The fracture and fatigue FRACTURE AND FATIGUE BEHAVIOR OF