Jtam-A4.dvi JOURNAL OF THEORETICAL SHORT RESEARCH COMMUNICATION AND APPLIED MECHANICS 55, 2, pp. 741-745, Warsaw 2017 DOI: 10.15632/jtam-pl.55.2.741 ANISOTROPIC FRACTURE CRITERION OF TI6AL4V TITANIUM ALLOY UNDER WIDE RANGE OF STRAIN RATES Wojciech Moćko, Cezary Kostrzewski Motor Transport Institute, Warsaw, Poland e-mail: wojciech.mocko@its.waw.pl The new fracture criterion taking into account stress triaxiality, strain rate and anisotropy is introduced in this paper. The model is capable to predict the influence of the loading direction on the fracture strain. The equation is applied to estimate the fracture locus of Ti6Al4V titanium alloy under quasi-static and dynamic loading regimes. Keywords: titanium alloys, failure, fracture, anisotropy, Hopkinson bar 1. Introduction In complex structures the fracture behavior depends on themultiaxial stress statewhichmay be definedwith the use of the stress triaxiality coefficient. Stress triaxiality is defined as η=−p/q, where p is the pressure stress and q is the Mises equivalent stress. In the basic approach, the fracture criterion εf is independent of the stress triaxiality. However, Johnson and Cook (JC) (1985) introduced a newdefinition of the fracture criterion as amonotonic function of the stress triaxiality in the following form εf = [C1+C2exp(C3η)] ( 1+C4 ln ε̇ pl ε̇0 ) (1+C5T̂) (1.1) whereC1,C2 andC3 arematerial parameters,C4 – strain rate sensitivity,C5 – temperature sen- sitivity, ε̇0 – reference strain rate, T̂ – temperature function described in other papers (Johnson and Cook, 1985). The JC model given by Eq. (1.1) was applied for the analysis of fracture characteristics of OFHC copper, Armco iron, 7075 T-651 aluminium alloy and 4340 steel (Johnson and Cook, 1985). Further studies of Bao and Wierzbicki (BW) (2005) showed that in the case of some materials, like 2024-T351 aluminium alloy, fracture behavior is govern by two mechanisms, i.e. shear bands and voids formation. The BW fracture criterion taking into account both void formation due to tensile loadings and shear failuremay be expressed in the following form (Bao andWierzbicki, 2005) εf =    1 1+3η D1+D2 for −1/3<η< 0 D3η 2+D4η+D5 for 0<η< 0.4 D6exp(D7η) for 0.4<η< 0.95 (1.2) Comparison between the threementioned fracture criteria is shown inFig. 1. The curves present JC model calibrated using data obtained for 2024-T351 alloy (Bao and Wierzbicki, 2005) and BW fracture locus estimated for Ti6Al4V alloy (Giglio et al., 2012). It can be observed that the BW criterion reveals two local minimums, the first corresponding to shear loadings (stress 742 W.Moćko, C. Kostrzewski triaxiality equal to 0) and the second corresponding to pure tensile loadings (stress triaxiality higher than 1). Comparing the JC and BW criteria, it can be stated that for tensile loadings (stress triaxiality higher than 0.4) the fracture strain estimated using both equations gives a comparable predictions, whereas for shear loading conditions the results may be substantially overestimated by the JC fracture model. Depending on the specimen geometry and pre-notch radius, various range of stress traixialities may be obtained (Bao and Wierzbicki, 2004, 2005; Driemeier et al., 2010; Gruben et al., 2011). Fig. 1. (a) Comparison of the fracture criterion estimated on the basis of simplified, Johnson and Cook (1985) and Bao-Wierzbicki (2005) theory; (b) specimen geometry and dimensions applied to obtain various stress triaxiality; the number of geometry in figure (b) corresponds to triaxialitymarked in figure (a) Analysis of the ductile fracture locus of Ti6Al4V introduced by Giglio et al. (2012) proved that in the case of this grade of thematerial, theBWfracture criterion, taking into account both tensile and shear failure, must be considered. The results showed some discrepancies between force-elongation curves obtained experimentally and numerically. One of the probable reasons for these differences may be related to material anisotropy, usually observed for the Ti6Al4V titanium alloy, which was not considered in the cited work (Giglio et al., 2012). Summarizing, the fracture criterion for majority of ductile materials may be expressed in form proposed by Johnson and Cook or Bao and Wierzbicki. However, in the case of materials with texture introduced by the fabrication process, further studies are required to obtain more accurate fracture models. A new approach should take into account especially the influence of the loading direction on the elasto-plastic and fracture behavior. This works extends earlier analysis (Giglio et al., 2012) of Ti6Al4V fracture criterion by introducing investigation of the titanium alloy anisotropy and deformation strain rate on the fracture locus. 2. Experimental methodology Thematerial was delivered in form of a hot rolled Ti6Al4V titanium alloy sheet of 3mm thick- ness. The specimens were cut along three orientations with respect to the rolling direction, that is, RD– along, 45D – 45 degree andTD– transverse to the rolling direction. Notched specimens with gauge length equal to 2mm, 5mm and 10mm were given various stress triaxiality coeffi- cients during tensile test (Fig. 1b).Additionally, shear specimensweredesignedand fabricated to obtain shear loading conditions (Fig. 1b). The same geometry was applied for both quasi-static and dynamic testing. The specimens were cut using electro-discharge machining (EDM). Ten- sile tests were carried out at quasi-static and dynamic loading regime using, respectively, with a servo-hydraulic testing machine and split Hopkinson tensile bar (Moćko et al., 2015, 2016). Simultaneously, plastic deformationwas recorded and analysed using a digital image correlation Anisotropic fracture criterion of Ti6Al4V titanium alloy... 743 software to determine strain distribution. Stress triaxiality during tensile tests was calculated usingFEMsimulation.Analysis was carried out usingABAQUSStandard software under quasi- static loading conditions. Digital models of specimens consists of 24554, 13070, 30555 and 10866 mesh elements, respectively, for type A, R1, R5 and shear samples. For geometry of type A, R1 and R5 C3D8R elements were applied, whereas for shear geometry C3D10M elements were used. Values presented in Fig. 2 are an average value calculated from themesh elements located near the fracture surface. Fig. 2. Comparison of experimental data with predictions of the proposedmodel at (a) quasi-static and (b) dynamic loading conditions 3. Results and discussion Fracture strain estimatedusing thedigital image correlationmethodat various stress triaxialities and strain rates is shown in Fig. 2. It can be observed that similarly to the BWmodel Ti6Al4V titaniumalloy loaded at thedirectionRDand45D is clear to the observedmaximumat the stress triaxiality equal to 0.5. At stress triaxialities higher than 0.5, the void formation mechanism is responsible for the fracture, whereas at stress triaxialities lower than 0.5, the failure of the material is governed by shear band formation or amixedmechanism. In the case of the loading force transverse (TD) to the rolling direction, the local maximum observed at 0.5 for other orientations is significantly diminished. An other observed phenomenon is the decreasing of the fracture strainwith an increase in the strain rate. Itmay be found thatRDand 45Dorientations are more sensitive to the strain rate effect than TD orientation. On the basis of experimental results, a new analytical model, including the strain rate and anisotropic effect, has been developed. It is based on the original model proposed by Bao and Wierzbicki (2005) taking into account only the stress state. In the original BW criterion, the fracture strain is expressed as a function of stress triaxiality εf = f(η) (3.1) The experimental results show that in the case of a hot rolled titanium alloy sheet, the fracture model should additionally take into account the effect of strain rate g(ε̇ pl ) anisotropy h(Θ) and temperature T̂(T) as follows εf = f(η)g(ε̇ pl )h(Θ)T̂(T) (3.2) The effect of strain rate may be expressed in form introduced by Johnson and Cook (1985) g(ε̇ pl )= 1+A1 ln ε̇ pl ε̇0 (3.3) 744 W.Moćko, C. Kostrzewski Anisotropic characteristics of Ti6Al4V titanium alloy are governed by two aspects: firstly, asym- metric geometry of the HCP crystallographic structure and its slipping plane and, secondly, texturing due to cold rolling processing. Results of material anisotropy on the fracture strain may be determined using the following equation h(Θ) = 1+B1exp ( 1− B2 Θ ) for Θ> 0 (3.4) where the loading angleΘ is estimated as an angle between the rolling direction and the loading force direction. Finally, the new fracture criterion taking into account stress triaxiality, strain rate and loading direction with respect to rolling directions takes the form εf =    ( 1 1+3η D1+D2 ) A for − 1 3 <η< 0 (D3η 2+D4η+D5)A for 0<η< 0.4 D6exp(D7η)A for 0.4<η< 0.95 (3.5) where A= ( 1+C1 ln ε̇ pl ε̇0 )[ 1+B1exp ( 1− B2 Θ )] (1+C2T̂) In order to calibrate Eq. (3.5), initially coefficients of the BW fracture criterion D1-D7 are estimated using the last square method. Subsequently, the strain rate sensitivity factor C for the alloy loaded in the rolling direction has been determined. In the final stage of calibration the parameters B1 andB2 describing anisotropic properties at quasi-static loading conditions have been calculated. Values of the particular coefficients obtained using the mentioned procedure are shown in Table 1. Comparison between the experimental data and predictions of the new model are presented in Fig. 2. The influence of the anisotropy coefficients B1 and B2 on the fracture strain is shown in Fig. 3. It can be observed that a good agreement between them has been obtained. Table 1.Coefficients of the new fracture criterion B1 B2 C ε̇0 [1/s] D1 D2 D3 D4 D5 D6 D7 200 1.7 −0.18 10−4 0.164 0.18 1.5 −0.052 0.34 8 −5 Fig. 3. Influence of the anisotropy coefficientsB1 andB2 on the fracture strain Acknowledgements This work was supported by theMotor Transport Institute (grant No. CBM/6506). Anisotropic fracture criterion of Ti6Al4V titanium alloy... 745 References 1. Bao Y., Wierzbicki T., 2004, On fracture locus in the equivalent strain and stress triaxiality space, International Journal of Mechanical Sciences, 46, 81-98 2. BaoY.,WierzbickiT., 2005,On the cut-offvalue of negative triaxiality for fracture,Engineering Fracture Mechanics, 72, 1049–1069 3. Driemeier L., Brünig M., Micheli G., Alves M., 2010, Experiments on stress-triaxiality dependence of material behavior of aluminum alloys,Mechanics of Materials, 42, 207-217 4. Giglio M., Manes A., Vigan F., 2012, Ductile fracture locus of Ti-6Al-4V titanium alloy, International Journal of Mechanical Sciences, 54, 121-135 5. Gruben G., Fagerholt E., Hopperstad O.S., Børvik T., 2011, Fracture characteristics of a cold-rolled dual-phase steel,European Journal of Mechanics A/Solids, 30, 204-218 6. Johnson G.R., Cook W.H., 1985, Fracture characteristics of three metals subjected to various strains, strain rates, temperatures and pressures,Engineering Fracture Mechanics, 21, 31-48 7. Moćko W., Brodecki A., Kruszka L., 2016,Mechanical response of dual phase steel at quasi- static and dynamic tensile loadings after initial fatigue loading,Mechanics of Materials, 92, 18-27 8. Moćko W., Brodecki A., Radziejewska J., 2015, Effects of pre-fatigue on the strain locali- zation during tensile tests of DP 500 steel at low and high strain rates, Journal of Strain Analysis for Engineering Design, 50, 571-583 Manuscript received November 21, 2016; accepted for print December 18, 2016