Effect of constraint on fracture parameters of piping materials

Fracture behaviour under quasi-static tearing load has been studied through fracture toughness tests on SA333 Grade 6 and 20MnMoNi55 steels using single-edge notched bend (SENB) specimens. To understand the response of variation of constraint, J-R curves were obtained from specimens precracked to a/W ratio in the range of 0.25 to 0.75, in steps of approximately 0.1. Stretch zone widths (SZW) were measured on the fractured surfaces of broken specimens. The stretch zone dimensions that were determined have been used in conjunction with the experimentally derived J-R curve to obtain a value of the ductile fracture toughness parameter J _SZW. The initiation toughness, J _i, obtained at the intersection of the blunting line and the power-law fit to the J-R curve, and the critical toughness, J _c, determined following the procedure of the ASTM standard, were estimated. Comparisons of J _SZW with J _i and J _c have been made for 20MnMoNi55 and SA333 steels as a function of the a/W ratio. The J _i and J _c values are higher at lower a/W, i.e. under low constraint, and fall with increasing a/W. A loss of constraint thus enhances these parameters. It was observed that J _SZW does not effect by a/W, and that its value is lower bound to J _i and J _c values for both the materials investigated.     

Effects of microstructure on fracture toughness of a high-strength low-alloy steel

Fracture toughness and ductile-brittle transition behavior were measured for a copper-bearing HSLA steel. The value ofK _lc for cleavage failure was independent of heat treatment, whileJ _lc for ductile failure decreased monotonically with increasing strength level. With both failure modes, fracture appears to be controlled by cracking of sulfide inclusions. The decrease in ductile-failureJ _lc is caused by decreased work-hardening rates that suppress cleavage and facilitate void coalescence. Both higher austenitizing temperature and quenching rate after austenitization influence the ductile/brittle transition temperature, either through grain-size and precipitate refinement or through an increase in the resistance of the steel to shear failure. Formerly Graduate Student, The Ohio State University     

Machining of reproducible flaws corresponding to fatigue cracks for fracture mechanics tests of components and welded joints

In connection with the analysis and the development of failure concepts in fracture mechanics for quasistatic loaded components and elastic-plastic behaviour of the material, tests are also carried out with welded and/or complex shaped specimens or structures. Thereby the difficulty arises of generating reproducible flaws in the form of fatigue precracks in definite positions in the components, respectively in the welded joint. It is reported exemplarily about experiments on different CT25 and CCT specimens and on a pressure vessel which contained a fatigue pre-crack, a 0.2 mm saw cut or notches with notch root radius ≤ ≥ 0.1 mm as flaws. The comparison of the results with regard to J-integral at initiation of stable crack, J_i, and J_R curves shows that under certain conditions the 0.2 mm saw cut (notch root radius ≤ ≤ 0.02 mm) is a useful alternative, if reproducible generation of a fatigue pre-crack will not be successful or too expensive. The tests were carried out on StE 460 and on a welded joint of this steel at 25 ± 2°C.     

Combined mode I-mode III fracture of a high strength low-alloy steel

Special formulations of theJ integral were used to quantify the combined mode I-mode III fracture behavior of ASTM A710 Grade A steel, a tough material which displays elastic-plastic behavior at room temperature. Experimental fracture initiation loci for two heat-treated conditions were linear inJ _i-J_iii space, agreeing with analytical predictions based on linear elasticity. As commonly observed in mode I fracture behavior, large tearing modulus values accompanied largeJ values for mode I and mode III components, individually as well as for totalJ values. There was a pronounced tendency toward antiplane shear flow in the modified compact tension specimens tested. This may result from the concentration of mode III shear strains in the plane of the propagating crack as predicted by slip line theory. Shear fractures with the crack propagating at inclined angles to both load line and plate free surfaces are favored energetically over pure mode I cracks even though the total surface area formed by the latter is much smaller. formerly Graduate Research Assistant, The Ohio State University     

Fracture toughness and its development in high purity cast carbon and low alloy steels

Fracture studies conducted using linear elastic fracture mechanics, crack opening displacement, and J-integral methods have shown that high room temperature toughness can be achieved in plain and low alloyed medium carbon cast steels. The high quality, carefully controlled, and heat treated steels prepared often yielded J_lc values which exceeded 0.1 MPa-m, or K_lc values exceeding 150 MPa-m 1/2 Correspondingly, crack tip opening displacement values were of the order of 0.24 to 0.26 mm. Of the several elements examined chromium is the most effective for developing optimum properties at medium carbon levels. However, in general, chemical composition, in terms of major alloying elements, was found to be only a minor variable in determining fracture toughness of the heat treated steel. Fractographic measurements revealed that a squared power relationship exists between the stretched zone width and relative toughness (K_lc/Φ_ys) at both low and high toughness levels. Formerly with McGill University     

Combined mode I-mode III fracture of a high strength low-alloy steel

Special formulations of theJ integral were used to quantify the combined mode I-mode III fracture behavior of ASTM A710 Grade A steel, a tough material which displays elastic-plastic behavior at room temperature. Experimental fracture initiation loci for two heat-treated conditions were linear inJ _i-J_iii space, agreeing with analytical predictions based on linear elasticity. As commonly observed in mode I fracture behavior, large tearing modulus values accompanied largeJ values for mode I and mode III components, individually as well as for totalJ values. There was a pronounced tendency toward antiplane shear flow in the modified compact tension specimens tested. This may result from the concentration of mode III shear strains in the plane of the propagating crack as predicted by slip line theory. Shear fractures with the crack propagating at inclined angles to both load line and plate free surfaces are favored energetically over pure mode I cracks even though the total surface area formed by the latter is much smaller.     

A study of crack tip blunting and the influence of blunting behavior on the fracture toughness of ultra high strength steels

Cross-sections of strained but not fractured compact tension J_IC speciments have been examined to investigate crack-tip blunting behavior as a function J level for four different microstructures. The microstructures were the as-quenched microstructures of HP9-4-20 and HO9-4-10 steels and the microstructures obtained by tempering these steels at 565°C. Smooth blunting was observed for the as-quenched microstructures while the fatigue cracks for tempered microstructures blunted to geometries characterized by two or three corners or vertices. The blunting geometries were clearly defined at J levels well below J_IC. For the case of smooth blunting voids tended to form directly ahead of the crack tip and crack extension by fracture occurred when the ligament between the blunting crack tip and the void directly ahead of the crack tip failed by shear fracture at an angle of about 45°C to the plane of the crack. Blunting to vertices was characterized by the growth of large voids very close to the corners or vertices of the blunting rack tip; it appears that the blunting geometry was maintained by the coalescence of these voids with the blunting crack tip. The results further suggest that if two microstructures have the same constrained ductility and identical inclusion distributions and one blunts smoothly and the other to vertices the microstructure which blunts to vertices can have substantially higher toughness.     

Corrosion Fatigue Crack Initiation in a Mode II Notch Specimen

The stress intensityK _II of a Mode II specimen was calculated using a finite element methodvia theJ integral. The site, direction, and the threshold value for crack initiation from the notch under cyclic Mode II loading in air, in water, and under dynamic charging with hydrogen were investigated. The results showed that the Mode II fatigue crack in a high strength steel initiated at or close to the site of the maximum principal stress, rather than at the site of the maximum shear stress, and the subsequent crack growth was oriented approximately normal to the direction of the maximum principal stress. The site and direction of crack initiation in water and under dynamic charging with hydrogen were similar but different from that in air. The threshold values for crack initiation in air, in water, and under dynamic charging were 28.8, 12.3, and 10.2 MPa m^1/2, respectively. The fracture surface of a corrosion fatigue crack in water and under dynamic charging consisted of intergranular facets at low ΔK _II values but of quasi-cleavage at higher ΔK _II values and were different from those in air.     

Some aspects of fracture mechanics research during the last 25 years

A review is given of the various concepts of fracture mechanics, furthermore of their application to assess the toughness of steels and to guarantee the safety of structures containing cracks. The three most important parameters are the stress-intensity factor K_I for linear elastic fracture mechanics (LEFM), the crack tip opening displacement CTOD and the J-integral for elastic-plastic fracture mechanics (EPFM). The ASTM designation E 1820-96 provides a common method for determining all applicable toughness parameters from a single test, including R-curves. Many investigations dealt with the numerous influences on these parameters, e.g. specimen geometry arid temperature and strain rate. Various failure concepts have been developed. The method of LEFM based on plane strain fracture toughness K_Ic is well established for high strength steels or low temperatures. For modern steels with high toughness the more complicated concepts of EPFM have to be applied. Because of the availability of commercial software for FE-calculations the application of the J-integral has become widespread in recent years. A number of approximative methods has been elaborated, e.g. CTOD-design-curve, BSI PD6493:1991, CEGB-R6-method, ETM, Eurocode 3 Annex C. Although the theory of fracture mechanics regards the material as an isotropic continuum the question is of great importance, how its materials parameters depend on the microstructure of the steels. Systematic investigations with the aid of the hot deformation simulator Wumsi showed the favourable material properties of fracture mechanics resulting from thermomechanical treatment. Many models were published for the quantitative correlation between microstructure and toughness parameters, mainly K_Ic. At the moment the modified Gurson model is in widespread use, which allows the prediction of J-R-curves.     

Microcrack initiation and acoustic emission during fracture toughness tests of A533B steel

The initiation of microcracks at MnS inclusions during fracture toughness tests of ASTM A533B steel compact tension specimen was detected by an 8-channel acoustic emission recording system. The microcracks were located as far as 7 mm ahead of the precrack front with a large spread of 5 mm above and below the plane of the main crack. Most of the microcracks were found to form before they were engulfed by the plastic zone which was determined by a finite element analysis. Assuming that the material is homogeneous and elastic outside the plastic zone, we estimated that microcracks were formed when the normal tensile stress σ_z at each inclusion is in the range of 400 to 800 MPa. The estimated stress value exceeds the uniaxial yield strength, σ_ys = 477 MPa of the material because σ_z near the crack tip could be as large as3σ _ys due to the stress triaxiality under plane strain conditions. Formerly Visiting Scientist in the Department of Theoretical and Applied Mechanics, Cornell University (on leave from Institute of Industrial Science, University of Tokyo, Tokyo, Japan).     

Effect of Ta Additions upon in situ Prepared Nb3Sn-Cu Superconducting Wire

The effect of Ta additions on the critical currents,J _c , and upper critical fields,H _C2 , ofin situ prepared Nb₃Sn-Cu superconducting wires has been examined. It was found that the Ta segregates virtually completely to the Nb dendrites and causes no observable change in the Nb₃Sn filament morphology. Both high fieldJ _c andH* _c2 values increase to a maximum at 10 wt pct Ta in the Nb-Ta phase and then drop for higher Ta contents. The improvement inJ _c in the 12 to 16T range, however, is small. TheJ _c values ofin situ wires are equivalent to bronze processed wires up to12T, but above that value they become significantly lower. Formerly a Graduate Assistant at Ames Laboratory     

Fatigue crack growth behavior of an oxide dispersion strengthened MA 956 alloy

Fatigue crack growth behavior of oxide dispersion strengthened ferritic MA 956 alloy was studied at 25 °C and 1000 °C in air at 0.17 Hz. The growth rates were analyzed using the linear elastic parameter ΔK and the elastic-plastic parameter ΔJ. Crack growth, although transgranular at both temperatures, increased by nearly three orders of magnitude with increase in temperature from 25 to 1000 °C. The growth rates were essentially the same in terms of either ΔK or ΔJ parameters indicating that plasticity effects are small even at 1000 °C. Detailed fractographic analysis revealed the presence of ductile striations in the ΔK range of 25 to 40 MPa√m at 25 °C and in a much narrower range at 1000 °C. Presence of voids could be detected at 1000 °C. Using the measured load-displacement hysteresis energies for a unit increment in crack length, crack growth rates were calculated using cumulative damage models and were compared with the experimental data. At 1000 °C the predicted and the experimental values agree within a factor of two and it is concluded that the growth occurs essentially by a damage accumulation process except in a narrow range of ΔK where the plastic blunting process is superimposed, resulting in ductile striations that were observed. At 25 °C the predicted and the experimental value reasonably agree for ΔK values greater than 40 MPa√m, and below this value the two diverge with predicted values being much lower. This divergence is related to occurrence of the plastic blunting process in this ΔK range as confirmed by fractographic evidence. The cumulative damage process at 1000 °C was related to the environmentally assisted void formation at dispersoid-matrix interfaces. At 25 °C the damage is related to the formation of microcracks ahead of the crack tip. These results and interrelation between alloy microstructure and fatigue fracture path are discussed in detail.     

Effect of specimen geometry, gage length, and width measurement locations on plastic strain ratio (R-value) in sheet metals

The effects of gage length, width measurement locations, and specimen geometry on plastic strain ratio have been investigated for the AA8011 aluminum alloy and interstitial-free (IF) steel sheets. The specimens were ASTM E 517 to 92a subsize, type A, and type A alternative with, respectively, 37.5, 76, and 57 mm reduced parallel sections and with different fillet radii. In each specimen type, there were differences in axial strains of gage marks and changes in width strain over the reduced parallel section, which depend on the applied axial strain, reduced parallel section legth, and fillet radius. Therefore, the magnitudes of the calculated R-values depend upon gage length, width measurement location, axial strain, and specimen geometry. These dependencies were more pronounced in the high R-value IF steel sheet relative to the low R-value AA8011 aluminum alloy sheet. The dependencies of R-value on gage length and width measurement location are negligible in all AA8011 specimens, while in IF specimens, these dependencies can be neglected only for type A specimens with 12.5-mm fillet radius. It is concluded that the observed differences in the measured R-values for specimens with different geometries can be attributed to the constraints imposed by the shoulders, which affect the width strain measurements and the resulting R-values.     

Combined mode I-mode III fracture toughness of a spherodized 1090 steel

The aim of this investigation was to determine the fracture behavior of a spherodized 1090 steel under combined mode I-mode III loading conditions. Suitably defined formulations of the J integral denoted J_ic and J_iiic were used to characterize the elastic-plastic fracture of this steel. As the mode III component in the system is increased, the resolved mode I J integral at initiation decreases, its mode III counterpart increases and the total J value remains nearly a constant. This implies a constant energy requirement for fracture initiation under mixed mode loading. As the crack plane becomes less inclined to the load line, the slopes of the mode I and total J resistance curves increase from their pure mode I values until a crack inclination angle of about 65° is reached. Somewhere in the region of 65-55°, a maximum in these values is reached and they fall off rapidly for larger mode III components. This drop is accompanied by the breakup of the crack front into mode I and mode III steps, which is shown to be an energetically more favorable process for this steel.     

Fracture and toughening mechanisms in an α2 titanium aluminide alloy

The deformation and fracture behaviors of the Ti-24Al-11Nb alloy with an equiaxed α₂ + β microstructure have been characterized as a function of temperature by performing uniaxial tension andJ _IC fracture toughness tests. The micromechanisms of crack initiation and growth have been studied bypost mortem fractographic and metallographic examinations of fractured specimens, as well as byin situ observation of the fracture events in a scanning electron microscope (SEM) equipped with a high-temperature loading stage. The results indicate that quasistatic crack growth in the Ti-24Al-11Nb alloy occurs by nucleation and linkage of the microcracks with the main crack, with the latter frequently bridged by ductile β ligaments. Three microcrack initiation mechanisms have been identified: (1) decohesion of planar slipbands in the α₂ matrix, (2) formation of voids and microcracks in β, and (3) cracking at or near the α₂ + β interface due to strain incompatibility resulting from impinging planar slip originated in α₂. The sources of fracture toughness in the 25 °C to 450 °C range have been attributed to crack tip blunting, crack deflection, and a bridging mechanism provided by the ductile β phase. At 600 °C, a change of toughening mechanisms leads to a lowering of the initiation toughness (theK _IC value) but a drastic increase in the crack growth toughness and the tearing modulus.     

Effect of loading mode on the fracture toughness of a ferritic/martensitic stainless steel

The critical J integrals of mode I (J_IC), mixed-mode I/III (J_TC), and mode III (J_IIIC) were examined for a ferritic stainless steel (F-82H) at ambient temperature. A determination of J_TC was made using modified compact-tension specimens. Different ratios of tension/shear stress were achieved by varying the principal axis of the crack plane between 0 and 55 deg from the load line. The value for J_IC was determined by means of specially designed specimens. The results showed that F-82H steel had high fracture toughness. Both J_IC and J_IIIC were about 500 kJ/m₂, and the mode I tearing modulus J_Ida) was about 360 (kJ/m²)/mm. However, J_TC and mixed-mode tearing modulus (dJ_T/da) values varied with the crack angles and were lower than their mode I and mode III counterparts. Both the minimum J_TC and dJ_T/da values occurred at a crack angle between 40 and 50 deg, at which the load ratio of σ_iii/σ, was 0.84 to 1.2. The J_min was 240 kJ/m², and ratios of J_lC/J_min and J_IIICJ_min were 2.1 and 1.9, respectively. The morphology of the fracture surfaces was consistent with the change in J_TC and dJ_T/da values. While the upper shelf-fracture toughness of F-82H depended on loading mode, the J_min value remained high. Other important considerations include the effect of mixed-mode loading on the ductile-brittle-transition temperature and effects of hydrogen and irradiation on J^.     

A simple probabilistic approach to the derivation of minimum yield strength values of steels at elevated temperatures

For the derivation of minimum yield strength values of steels at elevated temperatures, the formula R_p0.2/T/min = R_p0.2/20/min (1.1·f_av/T ? 0.1) is proposed, where f_av/T is the interpolated average trend value at a specified temperature, obtained by polynomial interpolation of the individual ratio values f_T = R_p0.2/T/R_p0.2/20 at all the test temperatures. The values of f_av/T characterize the trend of the R_p0.2-mean value line in function of the temperature. They are dependent on the material and on the temperature. The factor 1.1 and the reduction value 0.1 are interpreted in terms of a simple probability model. The model is based on the assumption that the total scatter of the test results is composed of two parts. One of these two parts is the correlation induced scatter due to the correlation between R_p0.2/T? and R_p0.2/20-values. The other one is the residual scatter. By evaluation of a number of data groups it is demonstrated that, for engineering purposes, the standard deviation of the residual scatter may be taken as about 5% of R_p0.2/20/min independent of the temperature. The validity of the formula is confirmed for ferritic and austenitic steels with minimum yield strength values at room temperature between 200 and 800 MPa and for f_av/T-values between 1 and 0.3. The connections with the ISO-method set down in ENV 22605-1 and ENV 22605-2, with the trend curve method set down in ENV 22605-3 and with the recently proposed modified trend curve method are explained.     

Influence of microstructure on intrinsic and extrinsic toughening in an alpha-two titanium aluminide alloy

The toughening mechanisms in the Ti-24A1-11Nb (Ti-24-11) alloy have been identified previously to include crack-tip blunting, bridging, and deflection by the ductileβ phase, while the fracture mechanisms involve the nucleation, growth, and linkage of microcracks with the main crack. By performing appropriate theoretical analyses and critical experiments, the relative contributions of intrinsic and extrinsic toughening mechanisms, including microcrack shielding, crack-tip blunting, bridging, and deflection by theβ phase, to the initiation and crack growth toughness values of the Ti-24-11 alloy have been studied for three microstructures. The results indicate that the microstructure affects not only the amount of toughness enhancement, but also the type of toughening mechanisms present in the Ti-24-11 alloy. The initiation toughness in Ti-24-11 arises from the matrix toughness, crack-tip blunting, and, occasionally, from crack deflection by the ductile phase. As a result, theK _IC values increase with the volume fraction of the ductile phase. In contrast, the resistance curve behavior originates from (1) a change of crack-tip singularity, which occurs when the blunted crack extends into the plastic zone, (2) crack bridging by ductile phase and shear ligaments, and (3) microcrack shielding, which occurs mostly at elevated temperatures.     

Fatigue Strength and Crack Initiation Mechanism of Very-High-Cycle Fatigue for Low Alloy Steels

The fatigue strength and crack initiation mechanisms of very-high-cycle fatigue (VHCF) for two low alloy steels were investigated. Rotary bending tests at 52.5?Hz with hour-glass type specimens were carried out to obtain the fatigue propensity of the test steels, for which the failure occurred up to the VHCF regime of 10⁸ cycles with the S-N curves of stepwise tendency. Fractography observations show that the crack initiation of VHCF is at subsurface inclusion with ??fish-eye?? pattern. The fish-eye is of equiaxed shape and tends to tangent the specimen surface. The size of the fish-eye becomes large with the increasing depth of related inclusion from the surface. The fish-eye crack grows faster outward to the specimen surface than inward. The values of the stress intensity factor (K _I ) at different regions of fracture surface were calculated, indicating that the K _I value of fish-eye crack is close to the value of relevant fatigue threshold (??K _th ). A new parameter was proposed to interpret the competition mechanism of fatigue crack initiation at the specimen surface or at the subsurface. The simulation results indicate that large inclusion size, small grain size, and high strength of material will promote fatigue crack initiation at the specimen subsurface, which are in agreement with experimental observations.