Sales

Low temperature and grain size effects on threshold and fatigue crack propagation in a high strength low alloy steel

Description
Low temperature and grain size effects on threshold and fatigue crack propagation in a high strength low alloy steel
Categories
Published
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Transcript
  Materials Science and Engineering, 51 (1981) 203 - 212 203 Low Temperature and Grain Size Effects on Threshold and Fatigue Crack Propagation in a High Strength Low Alloy Steel J. P. LUCAS and W. W. GERBERICH Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, MN 55455 U.S.A.) (Received June 5, 1981;in revised form July 8, 1981) SUMMARY Threshold and fatigue crack propagation results were observed for a high strength low alloy steel at temperatures ranging from 123 to 300 K and grain sizes ranging from 10 to 123 pro. The threshold stress intensity was observed to increase as the grain size increased. Also a substantial increase in threshold was exhibited as the test temperatures were reduced from 300 to 123 K. The mode of fracture near and at threshold was ductile for all temperatures and grain sizes. However, at intermediate AK levels and low temperatures the mode of crack propagation was brittle. Superimposed on the river lines of the cleaved fracture surface were cyclic cleavage striations indicative of the local fracture mode. The fatigue crack propagation exponent n of the Paris law increased rather drastically as the temperature decreased and was determined to be directly associated with the brittle fracture process. Emphasis was placed on the subgrain cell structure that develops during cyclic deformation as being a very likely microstruc- tural unit controlling the fatigue threshold process. Thus the grain size influence of the threshold is partially interpreted in terms of the subgrain cell size dependence on grain size. The temperature effect on the threshold is discussed from a thermal activation analysis view of dislocation behavior and plastic flow at low temperatures. 1. INTRODUCTION The effects of various parameters on fatigue crack propagation (FCP) near the threshold stress intensity level AKTH [1 - 14] are still 0025-5416/81/0000-00001502.50 poorly understood and in some cases unspeci- fied. Considerable effort has been centered around describing the effects of microstructure on fatigue crack growth at threshold. Whereas many microstructural variables have been examined for their influence on threshold, grain size has been given considerable attention most recently [7 - 14]. In general, the thresh- old is shown to increase, at least for the class of low to medium strength steels, as the grain size increases [9, 10, 11 - 14]. However, there is some controversy over the exact isolated effect of grain size on threshold since, in most alloys, changes in the grain size are usually related to changes in the yield stress [15]. Nevertheless, large grain size effects on thresh- old are realized. In a ferritic steel, Masounave and Bailon [9, 10] showed that an increase in the grain size of nearly an order of magnitude resulted in an enhancement of the fatigue threshold by a factor of 2. Priddle [13] in- vestigating a stainless steel also found an increase in threshold with an increase in grain size;virtually no change in threshold was found when the grain size was kept constant while changing the yield stress by 40%. Attempting to isolate the grain size effects, Benson [14] investigated the influence of grain size on threshold in a low alloy steel, while maintain- ing a nearly constant yield stress, by control- ling the precipitation-hardening contribution to the yield stress. Again, the threshold was found to be higher for larger grain sizes [ 14]. Unlike other microstructural and environ- mental variables, low temperature i.e. below room temperature) effects of FCP in steels are not as prevalent in the literature. This is especially true of studies dealing with low temperature effects on FCP near threshold stress intensity levels. However, low tempera- © Elsevier Sequoia/Printed in The Netherlands  204 TABLE 1 The composition (wt.%) of high strength low alloy steel C Nb Mn S P Si 0.06 0.03 0.35 0.01 0.01 0.03 Al Fe Grain size (pro) 0.01 Balance 10 ture investigations of FCP above threshold (region II) have been performed on several iron-based alloys [16 - 18]. In an Fe-Mo alloy system, Burck and Weertman [16] reported that crack growth rates were reduced as the temperature was lowered from 300 to 77 K. For a series of Fe-Ni and Fe-Si alloys, low temperatures had a very significant effect on the FCP rate. In fact, for these alloys, Gerberich and Moody [17] found that the FCP exponent n from the Paris relationship [19] da = C AK (1) dN increased from about 4 to 20 as the test temperatures were reduced from 298 to 123 K. Excessively high FCP exponents were observed to be associated with the brittle fracture modes which dominate during the fatigue process. Similar results were also observed for a high strength low alloy (HSLA) steel [20]. As acknowledged previously, very few studies exist on low temperature FCP and even fewer results have been reported on the effects of low temperatures on threshold stress intensity [18, 20 - 22]. In an HSLA steel, the thresholds increased as the temperature decreased [ 20], whereas Stonesifer [18] could find no effect of low temperatures on threshold even at 77 K. Tschegg and Stanzl [21], however, showed an increase in threshold as the temperature was changed from 300 to 77 K for a low carbon steel. In this investigation, low temperatures (300- 123 K) and grain size (10 - 123 pm) were simultaneously examined for an HSLA steel in order to ascertain their effects on threshold and near-threshold stress intensities for constant R ratio and test frequency. Some consideration is given to FCP above threshold at intermediate AK levels at low temperatures and to the predominate mode of fracture in this region. 2. MATERIALS AND EXPERIMENTAL PROCE- DURES The material tested in this investigation was an HSLA steel. Specimens were cut from 8.0 mm sheet stock oriented in the longitudi- nal-transverse position of the rolling direction. The major alloying elements are listed in Table 1, together with the as-received grain size. Grain sizes ranging from 10 to 123 pm were ob- tained by heat treating high in the austenitizing region at 1473 K for various periods of time in a vacuum of about 10 -5 Torr. All specimens were then allowed to furnace cool to room temperature. In Table 2 the heat treatment schedule is shown for the various grain sizes achieved. Test temperatures ranging from 300 to 123 K were controlled within an accuracy TABLE 2 Heat treatment and grain size Gram s~e(pm) Temperature K) Time h) 10 -- -- 65 1473 0.5 90 1473 3 123 1473 8 to + 1 K by utilizing an insulated chamber. A controlled flow of liquid nitrogen vapor was circulated over the specimen. Also a Nichrome heating element, controlled by a proportional controller in the specimen chamber, assisted in maintaining a constant temperature. The specimen temperature was monitored with a Cu-constantan thermocouple fixed to the specimen. Static testing to determine the tensile properties were carried out on an Instron test machine at a strain rate of 5.6 X 10 -4 s -1. Compact tension specimens 7.8 mm thick and surface ground to 600 grit were used for the FCP tests. Fatigue testing was performed on an MTS servohydraulic feedback test machine. A constant:amplitude sinusoidal load was applied to the specimen at a fre-  quency of 30 Hz. A fixed load ratio R equal to 0.1 was maintained for all tests. The fatigue crack growth rates were taken at constant Ap over a small incremental crack length Aa determined by compliance techniques [23]. The cyclic stress intensity range was calculated from Y AK = AP ~ (2a) BW1/2 where (a) z/2 \w][a 3/2 Y= 29.6 -- 185.5|--/ + +655.7 (2b) 205 Ap is the applied cyclic load, a is the crack length, B is the specimen thickness and W is the specimen width [23]. 3. EXPERIMENTAL RESULTS 3.1. Microstructural and tensile properties The as-received material consisted of a fine nearly equiaxed ferrite grain structure. Some small colonies of fine-lamellar pearlite could be found scattered throughout the predomi- nately ferritic grains (Fig. I(A)). After heat treating for grain growth, more fine-lamellar pearlite was observed along the grain boundary nodes in the 65, 90, and 123 tzm grain sizes. This is seen in Figs. I(B), I(C) and I(D) respec- (C) IOOl Fig. 1. Typical microstructures of the as-received and annealed material showing average grain sizes of (A) 10/,tm, (B) 65 ~tm, (C) 90 pm and (D) 123 ~m.  206 tively. The yield stress is given in Fig. 2 as a function of temperature at various grain sizes. There is little difference in the yield stress versus temperature curves for the 65, 90 and 123 pm grain sizes, and the reason for this is not clearly understood. However, the as- received 10/~m grain size has a much higher yield stress than the coarse grain specimens. This difference in the yield stresses is most probably due to a combination of the precipi- tation strengthening by Nb(CN) precipitates and the fine grain size in the 10 pm grain size material. 3.2. Fatigue crack propagation results FCP curves for the 10 ~m grain size are shown at four temperatures in Fig. 3. On examination of the fatigue threshold results at about 10 -1° m cycle -1 , it is seen that at 300 K the threshold stress intensity AKTH is 8.2 MPa m 1/2. The threshold increased as the test temperature was lowered from 300 to 123 K. At the lowest test temperature the threshold is raised by 5.3 MPa m z/2 at constant grain size. The threshold is enhanced because of low temperature effects by a factor greater than 1.60. The profile of the FCP curves showed a drastic change at intermediate AK levels above threshold, especially at 233, 173 and 123 K. The abrupt change in crack growth rate occurred at a da/dN value of approxi- mately 10 -s m cycle 1 . Such drastic increases in the FCP rate were related to a change in the fracture mode of the propagating crack. l O 759 .~'o~8 ..... 100 9 0 621 580 O 70 483 50 345 40 300 510 i 150 00 200 250 500 550 4()0 207 TEMPERATURE, K Fig. 2. Yield stress vs test temperature for various grain sizes (HSLA 2): o, 10 pro; o, 65/Jm; A, 90 pro; O, 123 pro. The threshold results for the 65 pm grain size can be seen in Fig. 4. Similarly, the threshold increased substantially as the temperature was lowered. At 300 K for the 65 pm grain size, the threshold stress intensity is 9.7 MPa m 1/2. Hence the grain size effect on threshold becomes apparent. There is a 20% increase in d id 7 E I0 to id ° 8 10 12 15 20 25 30 AK, MPa m ~ Fig. 3. Threshold and FCP curves for the 10/Im grain size s mples t 123 K m), 173 K D), 233 K o) and 300 K e) HSLA 2;R = 0.10; frequency, 30 Hz). d iv' 10' [] I , ,/,f,lf,,~,l, ,I,L 10 12 15 20 25 Z~, MPo m L~ Fig. 4. Threshold and FCP curves for the 65/.tm grain size samples at 123 K m), 173 K D), 233 K o) and 300 K o) HSLA 2;R = 0.10; frequency, 30 Hz).  threshold as the grain size is increased from 10 to 65/am. Again, if we start at da/dN 10 -s m cycle 1 , higher crack growth rates are evident with increasing AK values for the 65/am grain size. just as was found for the 10/am grain size in Fig. 3. The FCP curves for the largest grain size, 123 pm, are depicted in Fig. 5. The same trend for da/dN versus AK is exemplified as before with the 10 and 65/am grain sizes. Once more, a noticeable increase in threshold was witnessed as the grain size in- creased and as the temperature was decreased. It is worth while to note the combined grain size and temperature effects on the threshold stress intensity. For the grain size-temperature conditions 10/am, 300 K, and 123/am, 123 K, a difference in threshold of 10 MPa m 1/2 was observed. An increase in threshold of this magnitude would not be predicted by theoret- ical models using a dislocation emissary approach [ 24, 25]. Perhaps, the extent of the combined temperature and grain size effect on threshold is better illustrated in Fig. 6 where AKTH is plotted against temperature for the various grain sizes. Fractographs were taken of failed fatigue specimens in an attempt to discern the type of fracture involved during the threshold fatigue process. The mode of failure at and near threshold was predominantly ductile, although occasional cleaved facets were found ,d iO 7 ~lO a id a io e . . 10 12 15 20 25 N:? Z~K, M~ m ~'~ Fig. 5. Threshold and FCP curves for the 123/am grain size samples at 123 K (1), 173 K (D), 233 K (o) and 300 K (o) (HSLA 2;R -- 0.10; frequency, 30 Hz). 207 while scanning the breadth of the specimen at threshold [ 20]. The ductile nature at thresh- old is seen in Fig. 7 for the fine grain size material at 300 K. Surprisingly, the mode of fracture at threshold for a low temperature (123 K) and a coarse grain size (123/am) was also ductile. The fact that we would obtain extensive plasticity at threshold for a coarse grain size material at very low temperatures was not anticipated. Figure 8 is typical of the fracture surface observed at threshold at 123 K for the 123/am grain size. The failure appeared to be a ductile transcrystalline type. At intermediate AK levels the shape of the FCP curve change was very obvious especially for 10 and 65/am grain sizes. In this region the FCP exponent n increases rapidly as the temperature decreases [ 17, 26 ]. This accelera- tion in growth rate is emphasized in Fig. 9. 20 i i i ~8 16 10 B I i o 50 Ioo 400 IEMPERATURE, K Fig. 6. Threshold vs. temperature for 10 #m (o), 65/am ([]) and 123/am (o) grain sizes (HSLA 2; R = 0.10; frequency, 30 Hz). Fig. 7. A fractograph of fine grain (10 Pro) material observed at threshold at 300 K.
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x