Strength and Ductility at High-speed Tensile Deformation of Low-carbon Steel with Ultrafine Grains
2017 Volume 58 Issue 10 Pages 1487-1492
Details
2017 Volume 58 Issue 10 Pages 1487-1492
Tensile properties at various deformation speeds up to ～1500 mm/sec of low-carbon steel with elongated ultrafine grains were examined, focusing on both strength and ductility. Fine-grained samples were prepared by producing severe plastic deformation by caliber rolling. The ultrafine-grained samples showed an increase of elongation while maintaining high strength at a higher deformation speed. At a deformation speed of ～1500 mm/s, the absorbed energy exhibited by the fine-grained samples was as large as that of the conventional-grained sample, regardless of the difference in strength.
Grain refinement is a well-known method of strengthening metallic materials1). Many studies have adopted severe plastic deformation2) and thermomechanical processes3) to obtain ultrafine-grained microstructures with mean grain sizes smaller than 1 μm. While previous research works4–7) reported that strengthening by ultra-grain refinement tends to result in low toughness and low tensile ductility at room temperature, a more recent report8) has clarified that the ultrafine lamellar boundary microstructure provides enhanced toughness to low-alloy steel. This report implies the possibility of further grain refinement to improve multiple mechanical properties simultaneously.
By the way, the plastic behavior at high-speed deformation is one of the most important properties since high-strength steel sheets are popularly used in the body structure of automobiles, to ensure passenger safety in vehicle collisions, as is highly desired. In addition, increased modern manufacturing speeds motivate the investigation as well. Previous works9,10) studied strength at the high-speed deformation of ultrafine-grained steels and discussed the strain-rate sensitivity of the strength from the viewpoint of thermal activation in the dislocation motion, whereas the ductility or toughness has not yet been described.
A possible reason for the lack of knowledge about ductility and toughness at the high-speed tensile deformation of ultrafine-grained steels lies with the miniaturization of tensile test pieces for the precise measurement of stress so as to avoid elastic waves11). In the previous research of the tensile test9–11), the dimension of the gauge length was less than 10 mm, and the test piece was cut in the shape of a sheet. On the other hand, it is well known that ductility is strongly dependent on the shape of the sample. Because of this dependence, the sample shape is best followed by a standard code, such as ASTM-8M. In addition, the round bar test piece is more desirable because the area reduction, which is another primitive index of the ductility, can be measured. Consequently, it should be worthwhile to evaluate ductility at high-speed deformation by the round bar test piece with an adequate sample size, even if a noiseless stress–strain curve is impossible to obtain. This study aims to clarify both the strength and ductility at high-speed deformation of ultrafine-grained steel.
Low-carbon steel with a chemical composition of 0.15 mass% C–0.3%Si–1.5%Mn–bal.Fe was studied. The as-received steel was a hot-rolled bar with a cross-sectional shape of 40 mm square. The as-received bar was heated to 900℃ and held for 3600 seconds for austenitization. Subsequently, the samples were air cooled or water quenched to obtain ferrite-pearlite or martensite structures. Hereafter, the air-cooled and water-quenched samples are denoted as A40 and W40, respectively.
The ultrafine-grained microstructure was evolved by caliber rolling at a warm temperature. This process can introduce a fine lamellar structure in bulky steel bars8,12). A40 and W40 were soaked at 500℃ for 3600 seconds, and they were subsequently subjected to a caliber-rolling simulator. The roll diameter and rolling speed of the simulator were 368 mm and 500 mm/s, respectively. The total reduction in area though caliber rolling was 88% in 13 passes. The rolling processes have been described in detail elsewhere12). The cross section of the rolled sample was 14 mm square. For comparison, the as-received sample was reheated to 1200℃ and then rolled to a cross section of 14 mm square, followed by air cooling, to prepare a sample with a conventional ferrite grain size. The microstructure was examined by scanning electron microscope (SEM) with an Electron Backscatter Diffraction (EBSD) system.
Tensile testing was conducted with a Hopkinson-Kolsky bar11), consisting of a 600-mm-long incident bar and a 700-mm-long transmitted bar. Both bars were made of carbon steel, and their diameters were 24 mm. Displacement was applied by a hydraulic cylinder at various applied velocities ranging from 0.02 mm/s to ～1500 mm/s. Usually, in the Hopkinson-Kolsky bar method, the applied strain is evaluated from the elastic strain conditions of both the incident and the transmitted bars11), so that the applied strain is difficult to measure when the elastic waves appear dominantly. To avoid this difficulty, in this work, displacement was measured by an electro-optical extensometer when the deformation speed is higher than 100 mm/s. The electro-optical extensometer (Model-200XH, H.-D. RUDOLPH GmbH) is a contactless device that produces an analog-output signal proportional to the strain of a test piece. At a strain rate smaller than 100 mm/s, displacement was measured by a conventional extensometer. The load was evaluated from the elastic strain on the transmitted bar at any strain rate. The tensile direction was parallel to the rolling direction (RD). The shape of the tensile test is shown in Fig. 1. The gauge length and gauge diameter were 20 mm and 4 mm, respectively. This dimension follows the smallest size test piece written in the standard ASTM-8M. The sample has small rids on both ends of the gauge length to attach the marker plate for the extensometers.
Tensile test piece.
Figure 2 shows orientation color maps (a), (b), (c) and SEM images (d), (e), (f) of the hot-rolled sample (a), (d), A40 (b), (e), and W40 (c), (f). The horizontal and vertical directions of these images are parallel to the RD and the normal direction (ND), respectively. The color maps (a), (b), (c) indicate the crystallographic orientation parallel to the RD with the coloring showing at the top of this figure, and the black line indicates the high-angle boundaries. The hot-rolled sample has equiaxed ferrite grains whose mean grain size is 6.4 μm with a pearlite structure, whereas A40 and W40 exhibit elongated lamellar grains. The mean lamellar spacing of A40 is 1.7 μm, which is a little larger than that of W40 (0.6 μm). The fine carbide in A40 distributes more heterogeneously than in W40. This is because the starting microstructure before the rolling of W40 is martensite, where carbon is more uniformly distributed as solute atoms and fine carbides than in a ferrite-pearlite structure12). These results clarify that both A40 and W40 samples have ultrafine-grained microstructures. The orientation color maps of both A40 and W40 show mainly green, indicating the strong <110>//RD texture.
Orientation color maps (a), (b), (c) and SEM images (d), (e), (f) of the hot-rolled steel (a), (d), A40 (b), (e), and W40 (c), (f).
Figure 3 shows stress-strain curves at a quasi-static deformation (0.02 mm/s). Both A40 and W40 show preferable strengths higher than 700 MPa, while these exhibit less elongation than does the hot-rolled sample. The limited elongation of ultrafine-grained steel has been reported previously1). The strength of A40 is a little less than that of W40 because of these mean grain sizes (Fig. 2).
Nominal stress–nominal strain curves of the hot-rolled steel, A40, and W40 at a quasi-static deformation speed.
Figure 4 shows changes of displacement (a) and load (b) during the tensile test of W40 at a speed of 604 mm/s or 1660 mm/s. The nominal stress–nominal strain curves evaluated with the displacement and load at various deformation speeds are shown in Fig. 5. The displacement increases monotonically at both low and high speeds. However, the load shows a large oscillation. This oscillation was observed when a deformation speed was higher than 100 mm/s. It was reported that the elastic wave is accelerated when the gauge length of the test piece becomes larger11). Because of this oscillation, the flow stress at a certain strain is difficult to evaluate, so an alternative value for flow stress was evaluated in this work.
Displacement (a) and load (b) of W40.
Nominal stress–nominal strain curves of the hot-rolled steel, A40, and W40 at ～150 mm/sec (a), ～700 mm/sec (b) or ～1600 mm/s (c).
In order to compare the flow stress for the full range of the deformation speed, average stress, save was calculated according to the following equation:
| \[s_{ave} = \frac{1}{S_0 \Delta L} \int_0^{\Delta L}Pdx,\] | (1) |
Yield stress, tensile strength, and average stress of the hot-rolled steel (a), A40 (b), and W40 (c).
As described in the Introduction, the strength increase with the increasing of the strain rate is an important property for an engineering application. Therefore, it is important to compare the present results with the data reported in the literature. In our study, stress increased per the unit log of the strain rate. ∆s/∆log (dx/dt) of both the W40 and A40 is 20 MPa/decade; whereas, that of ultrafine-grained steel with a ferrite grain size and carbide similar to the sample annealed at 655℃ for 120 sec reported by Okitsu et al.14) is 30 MPa/decade. The differences between the present result and those previously can be considered as those of texture and the strengthening mechanism. According to the theory of the thermal activation of the dislocation, the strength increase per unit log of the strain rate is proportional to the product of the activation volume13) and the Taylor factor15). As shown in Fig. 2, the present samples have the strong texture of <110> parallel to the tensile axis. This direction is considered to have a larger Taylor factor than any other orientation14). On the other hand, the sample reported by Okitsu et al. has a relatively random texture. Therefore, it could be predicted that the strength increase per unit log of the strain rate of both A40 and W40 would be larger. However, the actual result is the opposite. This means that the strengthening mechanisms change. In the present work, both A40 and W40 were just in the as-rolled state, in which the samples contain higher densities of dislocation than does the sample prepared by annealing. This dislocation provides a short-range obstacle against the dislocation motion, while the grain boundary is regarded as the long-range obstacle. The short-range obstacle can possibly change the activation volume, so that the difference in the dislocation densities should be the reason for the strength increase per unit log of the strain rate. It should be noted, however, that the strengthening mechanism of ultrafine-grained materials is a topic still needing to be discussed, and more experimental data must be accumulated along with further discussion to solve it.
Figure 7 shows the elongation (a), the area reduction (b), and the absorbed energy, saveS0∆L (c), at various deformation speeds. In all three samples, the elongation and the absorbed energy increased, and the area reduction decreased when the speed was increased. This reciprocal relationship between elongation and area reduction is different from the usual tendency of ductility, that is, the increase in elongation is normally accompanied by area reduction16). It should be noted that the absorbed energies of the fine-grained samples increase to that of the hot-rolled sample at high-speed deformation. The absorbed energy can be regarded as an index for toughness, so that the present result indicates the superiority of ultrafine-grained carbon steel at high-speed deformation.
Elongation (a), area reduction (b), and absorbed energy (c) of the hot-rolled steel, A40, and W40.
Figure 8 shows the SEM images of the fracture surface of the hot-rolled steel (a), (b), A40 (c), (d), and W40 (e), (f) at a deformation speed of 0.02 mm/sec (a), (c), (e) or 1.5 × 103 mm/s (b), (d), (f). All samples showed typical dimple patterns, indicating that, in this work, all samples were produced with a ductile fracture manner at any deformation speed. The size of the dimple on the fracture surface of the hot-rolled sample appears larger than those of the ultrafine-grained steels. This seems to be due to differences in strength.
SEM images of fracture surfaces of the hot-rolled steel (a), (b), A40 (c), (d), and W40 (e), (f) at a deformation speed of 0.2 mm/sec (a), (c), (e) or 1.5 × 103 mm/sec (b), (d), (f).
The most significant finding of this study is the preferable tensile properties of ultrafine-grained steel at high-speed deformation, as shown in Fig. 7. One of the characteristics of high-speed deformation is the temperature increase of the test piece17). Measuring the temperature increase is significantly difficult, especially at high-speed deformation, due to the response delay of thermal censors, such as thermocouples. According to the literature17), at least, the temperature seems to reach more than 100℃. The actual maximum temperature should be much higher, due to strain localizations, preferably occurring in the ultrafine-grained sample18). Although assessing the temperature is difficult, this work's significant finding of ductility implies how it increases: concerning carbon steel, both the increased elongation and decreased area reduction occur when the tensile temperature lies in the range of blue brittleness, around 500℃, in some kinds of steels, such as 0.3%C cast steel19). This fact indicates that high-speed deformation generates abundant heat—sufficient for blue brittleness locally. Blue brittleness is a well-known phenomenon that is mainly due to dynamic strain aging. Previous research20) clarified the strain-rate dependence and reported that the temperature range that shows the inverse temperature dependence of strength becomes higher when the strain rate is increased. Strain-rate dependence prevents the precise prediction of whether blue brittleness will occur or not. The present work clarified, at least, that the ductility of ultrafine-grained microstructures at high-speed deformation maintains ductility, even if the fracture behavior is similar to the manner of blue brittleness. The increased elongation by the large heat generation increases the absorbed energy due to the high strength by ultragrain refinement.
This study clarifies the preferable ductility of low-carbon steels with elongated ultrafine grains at high-speed tensile deformation. Both fine-grained and conventional-grained steels show increased elongation with decreased area reduction, indicating a temperature increase during tensile testing. As a result, the absorbed energy of fine-grained steel increases to that of conventional steel. This indicates the merit in manufacturing ultrafine-grained carbon steel at high-speed forming.
The authors thank the Amada Foundation (AF-2016019), Japan, for its financial support.
