Journal Information
Vol. 8. Issue 1.
Pages 447-456 (January - March 2019)
Share
Share
Download PDF
More article options
Visits
803
Vol. 8. Issue 1.
Pages 447-456 (January - March 2019)
Original Article
DOI: 10.1016/j.jmrt.2018.04.005
Open Access
Effects of the welding inclusion and notch on the fracture behaviors of low-alloy steel
Visits
803
Guowei Chena,b,c, Hongyun Luoa,b,c,
Corresponding author
Luo7128@163.com

Corresponding author.
, Haoyu Yanga,c, Zhiyuan Hana, Zhenying Lina, Zheng Zhanga,b,c, Yuqin Sua
a School of Materials Science and Engineering, Beihang University (BUAA), Beijing, China
b The Collaborative Innovation Center for Advanced Aero-Engine (CICAAE), Beihang University (BUAA), Beijing, China
c Beijing Key Laboratory of Advanced Nuclear Materials and Physics, Beihang University (BUAA), Beijing, China
This item has received
803
Visits

Under a Creative Commons license
Article information
Abstract
Full Text
Bibliography
Download PDF
Statistics
Figures (9)
Show moreShow less
Tables (1)
Table 1. Energy coefficients of the selected signals with wavelet analysis.
Additional material (4)
Abstract

Effects of the welding inclusions and notches on the deformation and fracture behaviors of low-alloy steels were investigated by combining the tensile performance testing and acoustic emission (AE). The micro-structures and fracture surfaces of the hot-rolled and welded samples were observed by OM and SEM. It was found out that the welding inclusions and notches would result in negative impacts on the mechanical properties of low-alloy steel. During the fracture process of the welds, there were four groups of the AE frequencies, representing three kinds of deformation and fracture behaviors: plastic deformation of the welding matrix, fracture of the welding matrix, and fracture or interface detachment of the welding inclusions. Plastic deformation of the welding matrix mainly took place in the plastic stage and the late hardening stage. Fracture of the welding matrix and inclusions were concentrated in the first half of the hardening stage.

Keywords:
Welding inclusions
Notch
Fracture behaviors
Acoustic emission
Low-alloy steel
Full Text
1Introduction

Low-alloy steels, like C-Mn steels, are widely used as structural engineering materials and trustworthily applied in vessels, pipelines, cranes and bridges. It is important to monitor and analyze the deformation damage of the welds and defects. Acoustic emission (AE) method had been intensively used in the researches of the deformation and fracture behaviors in many kinds of metal [1–6] and nonmetal materials [7,8], especially for the composite materials [9–13] lately, for its rapid and sensitive ability to detect the micro-plastic or fracture events.

Welding can quickly splice parts together or repair damages, but the welding inclusions are generally negative, seriously affecting the safety and reliability of the structural materials [14–21]. Some previous works had been focused on the acoustic characterization of the welding defects [3,22–27]. Ennaceur et al. [1] once were concentrated on the crack propagation on the welds of the C-Mn steel and found out that the signals have special distribution in different crack propagation stages. Venkitarishnan et al. [5] used AE in assessing the welding effects on the aluminum alloy, which discovered that the brittle fracture inside of the welds would produce large numbers of AE signals. However, few of the previous works mentioned the relationships between the welds, especially welding inclusions, and AE signals, and the deformation and fracture process were seldom reported.

Defects in the manufacturing and welding processes were nearly inevitable, which would cause stress concentration during the deformation process [3,6,16,28,29]. Qu et al. [30,31] thoroughly studied the facture criterions and notch effects on the strength of crystalline metals, ceramics and metallic glasses by experimental and mathematical methods. Na et al. [3] believed that the main AE source did not come from the initial cracks in the toughening dimples, but from the secondary cracking and shearing fractures. Han et al. [6] found that the notched specimens emitted higher AE energy than unnotched ones during the yield stage, and AE signals with higher amplitudes were observed in notched specimens during Luders band propagation. However, the dynamic notch effect was rarely reported [29], especially with the real-time characterization by AE.

Here, this work was mainly about the deformation and fracture processes of the hot-rolled and welded specimens, and the effects of welding inclusions and notch were investigated by AE method. Micro deformation and fracture mechanisms within the welds were studied by in-depth AE signals and mathematical analysis, and the fracturing process of the welds was proposed.

2Experimental details

The low-alloy steel studied herein was the hot-rolled Q345 steel (Chinese code), as received in the standard thermo-mechanical heat treatment condition in the form of 16-mm-thick plates. The welded samples were cut from the hot-rolled base metal, and then welded by 2-passes submerged arc welding (SAW) along the rolling direction. The microstructures of different samples were mechanically polished and observed by optical microscope (OM, Leica DM4000, Leica Microsystems, Germany).

Both of the base metal and welding steels for tensile test were machined as dumbbell-like plate specimens, as displayed in Supplemental Fig. S1(a). There were mainly three kinds of samples investigated here in this study, which were base metal (BM), across-welding-seam (ACW), and along-welding-seam (ALW) specimens (Supplemental Fig. S1(b)), respectively. Notches were introduced in the BM and ACW specimens (NBM, NACW) at the center of the gauge length. Tensile tests were performed on a mechanical test machine (SANS, MTS Industrial System Co. Ltd., China) at a nominal strain rate of 6.68×10−4s−1. Nominal strain was used to indicate the elongation at break.

The morphology of the fracture surfaces was observed in an electron scanning electron microscope (SEM, JSM-5800, Japan Electron Optics Laboratory Co. Ltd., Japan) with an accelerating voltage of 10kV. The chemical compositions of different parts were analyzed by an energy dispersive spectrometer (EDS) equipped within the SEM.

Acoustic emission (AE) signals were recorded and analyzed by a digital signal processor with an AEwin v2.19 AE system (Physical Acoustic Corporation, USA). Every data acquisition card (PCI-DSP, PAC, USA) has four channels, each with a sampling rate of 50Hz. Four broadband piezoelectric transducers with a resonant frequency of 375kHz, a preamplifier with 40dB gain and a compatible filter (10kHz–2MHz) were used to capture the AE signals. Vaseline was used at the interface between the transducers and the specimen surface to improve the signal transmission. AE detect schematic diagram can be seen in Supplemental Fig. S1(c).

3Results and discussions3.1Micro-structure and composition

The optical micrographs of the base metal and welded samples were displayed in Fig. 1. Their chemical compositions are listed in Supplemental Table S1. Typical ferrite and pearlite microstructures were observed in the base metal (Fig. 1a), where the ferrite grains had an average diameter of ∼15μm. While in the welded seams, there were more ferrite grains (with bigger sizes more than 20μm) and less pearlite observed, and there were also large numbers of spherical impurity inclusions introduced in Fig. 1b. Chemical compositions of the impurities were analyzed, which had distinct contents of O, Si and Mn elements (Fig. 1c). Compared to the base metal, there might be MnO2 and SiO2 inclusions introduced in the welding seams during the welding process.

Fig. 1.

Optical micrographs and chemical compositions of the Q345 steel. (a) Microstructures of base metal; (b) welding seam were adapted from Han et al. [27] and (c) chemical compositions of the welding impurities.

(0.57MB).
3.2Tensile performance

Different kinds of specimens had quite different tensile performances, in terms of strength and plasticity, as can be seen in Fig. 2. The BM specimens had slightly better yield strength than the ACW specimens, which were ∼380MPa, but far better than the ALW specimens, which were just around 290MPa. BM specimens had the largest strain (∼37%), and the ALW specimens were also more than 30%. The others (NBM, ACW and NACW specimens) were less than 20%, far below the former two, showing much worse plasticity than the BM and ALW specimens. However, the NBM samples shown higher strength than BM samples, which might be the effects of notch induced brittleness [30]. Additionally, there was an interesting distinguishing feature observed on the curves. The BM, NBM and ACW specimens had obvious yielding platforms, showing typical discontinues yielding processes [6]. While the fracture processes of NACW and ALW specimens indicated clearly continues yielding, obvious yielding platforms can hardly be found.

Fig. 2.

Stress–strain curves of the (notched) hot-rolled Q345 base metal and (notched) welding specimens. BM and NBM curves were adapted from Han et al. [6].

(0.09MB).
3.3Notch and welding inclusion effects on the fracture morphologies3.3.1Cracks and slippage bands on the profile

The tensile induced cracks in the necking area were influenced by the notch, which can be obtained by comparing the profile morphology displayed in Fig. 3a and b. Large numbers of cracks were observed equally scattering on the profile near the fracture surface of the BM samples, with a wide distribution area (Fig. 3a). While on the profile surface of the NBM samples, the cracks can only be found very close to the fracture surface, concentrated into a line along the fracture surface outline (Fig. 3b). The pre-notch caused severe stress concentration at the notch tip, so that the induced deformation and fracture would have narrower affected area. Similarly, on the profiles of the welded samples the induced cracks can only be observed close to the fracture surface (Fig. 3c). As compared Fig. 3c to a, it can be reasonably assumed that the concentrated distribution of the cracks was affected by the welding inclusions.

Fig. 3.

Cracks on the profiles near the fracture surfaces of (a) BM; (b) NBM; and (c) ALW specimens.

(0.48MB).

Large numbers of slippage bands could be observed on the profile surfaces of the fractured hot-rolled and welded specimens (Fig. 4), where the slippage bands had different orientations varied from one grain to another. It can be reasonably assumed that most of the slippage bands originated from the grain boundary and then propagated throughout the grains, as marked in the white circles in Fig. 4(a). The formation and movement of the slippage bands might be related to the activated dislocations at the grain boundaries. While on the fractured profile surfaces of welded samples (Fig. 4b and c), the slippage bands have bigger sizes in both of length and width, since the average grain size in welded specimens was bigger than in the BM. However, the interactions between the slippage bands and grain boundaries could hardly be observed in the welded samples.

Fig. 4.

Slippage bands, welding inclusions and micro-cracks on the side surface of the fractured (a) BM, (b) ACW; and (c) ALW specimens.

(0.37MB).

Additionally, large amount of welding inclusions and particle pits were mainly observed embedded or wrapped within the welding seams on the fractured profile surfaces of the ACW and ALW specimens (Fig. 4b and c), which had varied diameters from 1 to 18μm. These welding impurities would cause stress concentration around the inclusion particles, which might further interact with the slippage bands and grain boundary [20]. Interestingly, ALW specimens had higher strain than the ACW, and there were lots of micro-cracks induced by the welding inclusions at the slippage bands or grain boundaries in the severe deformation process. The interactions between welding inclusions and cracks can be reflected in Fig. 4(c), where some inclusions or pits were linked together, forming a few micro cracks.

3.3.2Fracture dimples and tearing ridges

Specimens with and without notches had significantly different morphologies on their fracture surfaces. The fracture surfaces of BM samples were found full of equiaxed dimples with different sizes and depths, as displayed in Fig. 5(a). While on the NBM samples near the notch, there were numbers of shearing dimples and wide tearing ridges observed (Fig. 5b). The flat bottom and the straight outline of certain dimples indicated that there might be some intergranular fractures during the fracture process near the notch.

Fig. 5.

Morphologies of the fracture surfaces. (a) BM; (b) NBM, near the notch; (c) ACW; (d) NACW, near the notch; and (e, f) ALW with different magnifications.

(1.63MB).

There were also equiaxed dimples conquering the fracture surfaces of the welded specimens, for both ACW and ALW (Fig. 5c and e). Dimples there had varied sizes and depths, with large numbers of spherical inclusions embed at the bottom (Fig. 5c and f). The fracture surface of the NACW had less dimples near the notch, but a few narrow tearing ridges, as can be seen in Fig. 5(d). There were also some introduced welding impurity particles uncovered by the violent tearing process. The relatively flat fracture surface and narrow tearing ridges implied that the NACW had unsatisfactory plasticity, which was consistent with the corresponding stress–strain curve displayed in Fig. 2.

Compared to the BM, the dimples on the fracture surface of welded samples (ACW and ALW) had smaller and more regular sizes, as shown in Fig. 5(a, c and e). Most of the dimples were obviously filled with welding inclusions, and the impurity particles have diameters from 1 to 8μm. The welding compositions resulted negative impacts on both of the strength and plasticity of the welded specimens, as comparing the BM to the ALW samples (Fig. 2). The unsatisfactory tensile performance of the welded specimens could be ascribed to the ferrite coarse grains and the impurity inclusions in the welding seams. Welding inclusions here might be the stress concentration points, cutting down the plasticity of the ALW samples. The ACW has higher strength but lower plasticity than ALW samples, due to the constraint from the base metal or the triaxial stresses at the heterogeneous interfaces, where the stress concentration would short the hardening ability and bring forward the fracture. Additionally, the less strong welds were constrained by the strong base metal around, so that the inclusions-contained welding seam will be quickly fractured. In the ALW specimens, the hardening process would be undertaken by the whole sample, and inclusions-nucleation induced dimples could have enough time to grow. Thus, more deep dimples were found on the fracture surface (Fig. 5e), and ALWs showed competitive plasticity but poor strength. The growing process of the dimples would cost more energy, so that the ALW samples had better tensile plasticity performance than the ACW samples.

Notched specimens had significantly different fracture surfaces from the unnotched counterparts (Fig. 5), which indicated that there were different fracture modes due to the existing of notches. BM and ACW samples had been through a numbers of toughness polymerized fractures, as can be reflected from the countless varying sized dimples (Fig. 5a and c). As notches were introduced in, dimples there would be nucleated and generated around the notches in the early deformation stage. Since there are excessive shearing forces near the notches, the ligaments around the newborn dimples would be hardening quickly, and soon be tore up violently (Fig. 2), especially for the welding samples. Shearing dimples and tearing ridges could be observed on the fracture surface of the NBM and NACW specimens (Fig. 5b and d).

3.4Acoustic emission signals during the fracture process

Similar to the reinforcement in composites, the welding inclusions might have interface detachment or be fractured during the deformation and fracture process of the welding matrix, which can be in situ detected by AE.

3.4.1Signal waveforms and amplitude distributions

The AE signal waveforms were analyzed and there were mainly two types in the fracture processes of all specimens, type A and type B, as displayed in Fig. 6, which were burst signals and continuous-burst signals, respectively. The former type of signals were produced by dislocation multiplication and unpinning from Cottrell atmospheres or dislocation tangling, and the latter type was attributed to the simultaneous motion of high-density dislocations during Luders band propagation [6].

Fig. 6.

AE signal waveforms detected during the fracture process of the low-alloy steel. (a) Burst-type signal and (b) continuous-burst type signal, adapted from [6].

(0.18MB).

The signal types were found with different distributions within the BM and ACW specimens, as shown in Fig. 7. As can be seen in Fig. 7(a) and (b), there were large numbers of type B signals only generated in the yield stages of BM and NBM specimens, and the type A signals were produced mainly in the elastoplastic stage, and small proportions in the hardening stage and necking fracture stage. Obviously, the type B signals had lower amplitude (30–50dB) than type A signals (40–100dB). A clear declination happened to the type A signals toward the late hardening stages, which can be reasonably assumed that the dislocations were blocked in the hardening stage [6].

Fig. 7.

Stress–strain curves and the AE amplitude distributions of the (a) BM and (b) NBM; (c) ACW; (d) NACW; (e) ALW specimens; and (f) AE accumulative energy of the hot-rolled and welded specimens (1, plasticity stage; 2, yielding stage; 3 – hardening stage).

(0.58MB).

Interestingly, in the fracture processes of the welded samples, as can be seen in Fig. 7(c) and (d), the signals types had special distributions different from the hot-rolled specimens. ACW specimens had narrow yield platform, and there were less type B signals appeared in the yield stage than in the BM. NACW specimens had none obvious yielding during the deformation and fracture process, and there were only type A signals detected in the whole fracture stages. Additionally, the hardening stages of the welded samples (ACW and NACW) had more AE signals (type A) than the hot-rolled samples, and the signals had their amplitudes concentrated in 45–80dB, higher than the hot-rolled counterparts (40–65dB).

As displayed in Fig. 7(e), the amplitude distributions along the fracture process of the ALW specimens were quite different. A small period of yield platform was found, but there were no type B signals detected. Compared to the BM and ACW samples, more type A signals were found gathered at the first half of the hardening stage, where there were also numbers of special signals with higher amplitudes more than 80dB. These high amplitude signals diminished in the second half of the hardening stage, as well as the lower ones.

3.4.2AE energy of the fracture process

For the further information of the AE signals of whole fracture process of the above specimens, the accumulative AE energy was used for showing fracture mechanisms and behaviors. As can be seen in Fig. 7(f), different specimens had big gaps on the accumulative AE energy in different fracture stages.

In the plasticity stages, specimens had similar AE energy released, which might imply the same deformation mechanism within. Specimens with obvious yield stages (BM, NBM and ACW) had higher AE energy than the NACW and ALW specimens in the stage 2, because they had no or just little yield platform, and there was big quantity gap on the signal numbers. Tough the welded samples (ACW) had a yield stage, their AE energy were still obviously lower than the hot-rolled (BM, NBM) samples because of the lack of type B signals. The NBM had far more AE energy released in the yield stage than BM, due to the notch effect, and there were more type B signals detected during the yielding process. NBM had the lowest energy in the hardening stage, because the type A signals might be blocked and reduced by severe stress concentration caused by the notch. ACW and NACW specimens shared the similar AE energy in their hardening stages, a little lower than BM, but certainly with different deformation and fracture mechanisms, affected by the notch-induced stress concentration and the welding inclusions. ALW samples had the highest accumulative AE energy in the hardening stage, where there were the most signals detected, which should be attributed to the countless welding inclusions.

3.4.3Signal frequency within the welds

As noted above, the fracture process of the ACW specimens were constrained by the base metal and the NACW samples were deeply affected by the pre-notch. Pure fracture behaviors within the welds can only be detected and recognized by AE method in the fracture process of ALW samples.

To obtain more features from the AE signals, Fast Fourier Transformation (FFT) was used to change the waveform from time domain to the frequency domain. Here, the peak frequencies were extracted from the waveforms for the fracture mechanism analysis, since certain kinds of fractures had characteristic acoustic peak frequencies [9,32,33]. It was found out that the peak FFT frequency of the collected signals had clustering distribution, and the AE signals could be obviously divided into four groups, as can be seen in Fig. 8. From 100 to 150kHz, the first group was named type 1 signals, which had the least amount. The second group, type 2 signals, was varied from 200 to 300kHz, and had an intensive distribution in the short yielding stage and the early half of the hardening stage. The type 3 and type 4 signals were closely stacked, ranging in 330–385kHz and 400–460kHz, respectively, and they were concentrated in the early hardening stage.

Fig. 8.

FFT frequency distribution in the fracture process of the ALW specimen.

(0.15MB).

Signals with different peak frequencies might be related to certain fracture behaviors and mechanisms [7]. To separate the AE signals and find out the underlying mechanisms, eight signals with different peak frequencies were randomly selected from the four frequency bands for the further mathematical analysis. Wavelet was used to decompose and reconstruct the frequencies [23]. The wavelet function φ(x) can be transformed to a group of sequence after a series of telescopic and translation transformations:

where a is the scale factor of the telescopic degree, and the b is the time factor of the translation mount. The continuous transformation of wavelet sequence should follow the law of:

The wavelet decomposition and reconstruction could be computed by the db8 wavelet function with excellent frequency resolution:

The energy coefficients of the selected signal points are displayed in Table 1. The coefficient details of peak frequency and the corresponding spectrum can be seen in Supplemental Fig. S2.

Table 1.

Energy coefficients of the selected signals with wavelet analysis.

No.  Frequency (kHz)  DGGGGG
105  0.0781  0.1195  0.4912  45.0864  24.9504  29.2744 
118  0.2551  0.2066  0.7067  49.4040  28.5886  20.8390 
216  0.3463  0.2621  0.7299  2.2544  27.3394  69.0679 
272  0.5104  0.3322  0.3003  3.1655  24.0950  71.5966 
351  0.4246  0.2122  0.5809  1.6305  13.4867  83.6652 
385  0.2055  0.2627  0.4719  2.3826  14.8325  81.8447 
410  0.2348  0.2920  0.4235  1.1581  16.0698  81.8219 
445  0.5832  0.3489  0.5771  3.5460  14.3992  80.5456 

The bold numbers can be defined as dominant energy coefficients.

There are clear differences on the proportion of the spectrum coefficient, which could help to separate the AE signals into three categories. The first category, with G3, G4 and G5 taking most of the proportion, representing type 1 signals with peak frequencies ranging from 100 to 150kHz, might come from the plastic deformation of the coarse ferrites. The second category, with G4 and G5 accounting for the largest, standing for the type 2 signals with peak frequencies varied in 200–300kHz, may be caused by the fracture of the welding matrix (coarse ferrites and peralites). The third category of signals, including type 3 signals with frequencies ranging from 330 to 385kHz and type 4 from 400 to 460kHz, as G5 taking the similar highest percentage, could be resulted by the fracture or interface detachment of the hard inclusion particles [34,35] (e.g. SiO2 and MnO2). More discussions can be found in Appendices.

3.5Welding and notch effects on damage mechanisms and fracture behaviors3.5.1Damage mechanisms in the yielding stage

Low amplitude type B signals were related to the collective movements of the dislocations, only found in the yielding stages (Fig. 7a–c). There were much less type B signals in the ACW, as compared to the BM samples, which indicated that there were less dislocation movements in the welded samples. Besides, there were no type B signals in the NACW and ALW specimens, which all had their welded parts to sustain the tensile strains. This was consistent with stress–strain curves of the NACW and ALW in Fig. 2, where there was no or little yielding during their fracture processes. The slippage bands observed on the side surfaces could also provide some trails (Fig. 4). According to a previous work [8], the Luders band propagation is mainly related to the reactions between dislocations and grain boundaries. BM had finer grains, as can be seen in Fig. 4(a), which confirmed that there were plenty of opportunities and spaces for the dislocations’ pop in and out at the grain boundaries. Thus, large numbers of type B signals were generated in the yield stage of the BM and NBM samples (Fig. 7a and b). While as displayed in Fig. 4(b), there were bigger grains and less grain boundaries, and the slippage bands were wide and long. Therefore, there would be less Luders band movement at the coarse ferrite grain boundaries, so that the type B signals will be reduced and even disappeared [32], as shown in Fig. 7(c–e).

3.5.2Damage mechanisms in the hardening stage

AE signals in the hardening stage come from the energy released by further dislocations movement and local fractures, and the signal strength was related to the plastic deformation volume in the dislocations movement [6,36]. BM samples had much better plasticity and higher AE energy than the NBM, because the stress concentration near the notch led to the deformation volume declination. Compared to the BM, ACW had lower AE energy released, because the hardening process was mainly endured by the welds, which had smaller deformation volume under the constraint of the base metal, though the welds had inclusions inside that will provide extra AE sources. The NACW had smaller deformation volume than the ACW because of the notch, but there was no yielding in the NACW. In addition, the AE signals of the ACW and NACW had different sources. The AE energy of the ACW was released by the brittle fracture of welding inclusions, while the later was mainly come from the severe stress concentration and shearing fractures, where the welding inclusions around the notch would seriously intensify the stress concentration. Therefore, the AE signals in the hardening stage indicated that the NACW samples suffered continuous hardening and fracture processes. The AE signals in the late half hardening stage of the NACW samples were alien to the other samples, as shown in Fig. 7(d), which had a continuous distribution till the final fracture. It was assumed to be the combined effects by the notch and welding inclusions, as compared to the NBM sample displayed in Fig. 7(b).

ALW had the highest AE energy in the hardening stage. On one hand, the ALW had competitive plasticity, which insured that there might be a large plastic deformation volume during the hardening of the ALW specimens. On the other hand, the welding inclusion fractures, inclusions-matrix separations, and the local fast ligament fractures cannot be ignored. Large numbers of type A signals were generated within the first half of the hardening stage, as can be seen in Fig. 7(e and f), which had much more than the other samples and contributed most of the AE energy.

3.5.3Fracture process and behaviors within the welds

There are three kinds of deformation and fracture behaviors during the dynamic facture process of the ALW specimens. They are plastic deformation of the welding matrix (type 1), fracture of the welding matrix (type 2) and fracture or interface detachment of the welding inclusions (type 3 and type 4).

The fracture process was shown in Fig. 9. In the plasticity stage of the fracture process, there were mainly fracture and plastic deformation of the welding matrix (ferrite and pearlite). A small mount of welding inclusions would be broken during the gentle plastic deformation of the welds matrix. As in the short yielding stage, there were only the type 2 signals were detected, which indicated that there was only the fracture of the welding matrix. Large numbers of toughening dimples would be nucleated around the inclusions, and a few local dimple ligaments will be fractured during the dimple consolidation and micro-crack propagations. The local matrix fracture of the first two stages (dimple ligaments fracture) would trigger more severe damage in the beginning of the hardening stage, wherefore large numbers of matrix fracture signals were detected. In the mean while, the welding inclusions within would be deformed and fractured along with the welding matrix, and that is why the fracture signals were almost emerged in the first half of the hardening stage. In the late half hardening stage close to the necking fracture, dimples were deepened and there were only plastic deformation signals were detected, until the last necking fracture came with two fracture signals.

Fig. 9.

Dynamic fracture behaviors and process of the ALW specimen. (a) AE amplitude distribution of different signal types and (b) signal numbers of the signal types in the fracture stages (1, plastic stage; 2, yielding stage; 3, hardening stage; 4, necking fracture).

(0.18MB).
4Conclusions

Welds and notch would result negative effects on the mechanical performance of the low-alloy steel. Notches would bring more damages to the welded samples, and there was no Luders band propagation in the welds during the yielding stage. During the fracture process of the welds, there were four groups of the characteristic frequencies, representing three kinds of deformation and fracture behaviors: plastic deformation of the welding matrix, fracture of the welding matrix and fracture or interface detachment of the welding inclusions. Plastic deformation of the welding matrix mainly took place in the plastic stage and the late hardening stage. Fracture of the welding matrix and fracture of the welding inclusions were concentrated in the first half of the hardening stage.

Conflicts of interest

The authors declare no conflicts of interest.

Acknowledgements

This work was supported by National Key Research and Development Program of China (Nos. 2017YFF0210002, 2016YFF0203301 and 2016YFC0801903) and the National Natural Science Foundations of China (Nos. 51175023 and U1537212).

Appendix A
Supplementary data

The following are the supplementary data to this article:

Appendix B
Supplementary data

Similar conclusions can be obtained by comparing the fracture process of NBM and NACW samples, as shown in Supplemental Fig. S3. The fracture process of the NBM samples were detected with two groups of signals, 100–150kHz and 200–280kHz, respectively, representing the plastic deformation and fracture behaviors of the ferrites and pearlites (Supplemental Fig. S3(a)). The NACW samples had similar FFT frequency distributions with the ALWs, but much less in the signal numbers (Supplemental Fig. S3(b) and Fig. 8). As welded specimens, they all had three kinds of deformation and fracture behaviors in the fracture processes. Because of the existence of welding inclusions, they had higher frequency signals recorded during the fracture process, which strongly implied that the signals with higher frequencies were released by the fracture behaviors of welding inclusions.

References
[1]
C. Ennaceur, A. Laksimi, C. Herve, M. Cherfaoui.
Monitoring crack growth in pressure vessel steels by the acoustic emission technique and the method of potential difference.
Int J Press Ves Pip, 83 (2006), pp. 197-204
[2]
H.N.G. Wadley, C.B. Scruby.
Tempering effect on acoustic emission-microstructural relationships in ferritic steels.
J Mater Sci, 28 (1993), pp. 2501-2516
[3]
E.G. Na, K. Ono, D.W. Lee.
Evaluation of fracture behavior of SA-516 steel welds using acoustic emission analysis.
J Mech Sci Technol, 20 (2006), pp. 107-204
[4]
J. Kumar, S. Punnose, M.M. Kumar, T. Jayakumar, V. Kumar.
Acoustic emission study on effect of stress triaxiality on damage micromechanisms in near alpha titanium alloy.
J Nondestruct Eval, 28 (2009), pp. 85-90
[5]
P.V. Venkitakrishnan, P.P. Sinha, R. Krishnamurthy.
Study and analysis of effect of various thermal processes in AA 2219 annealed sheet using acoustic emissions.
Mater Des, 27 (2006), pp. 770-775
[6]
Z. Han, H. Luo, H. Wang.
Effects of strain rate and notch on acoustic emission during the tensile deformation of a discontinuous yielding material.
Mater Sci Eng A, 528 (2011), pp. 4372-4380
[7]
G. Chen, H. Luo, S. Luo, Z. Lin, Y. Ma.
Vertically oriented structure and its fracture behavior of the Indonesia white-pearl oyster.
J Mech Behav Biomed Mater, 66 (2017), pp. 211-223
[8]
K.N. Tandon, K. Tangri.
Acoustic emissions during deformation of polycrystalline silicon-iron.
Mater Sci Eng, 20 (1975), pp. 47-54
[9]
C.R. Ramirez-Jimenez, N. Papadakis, N. Reynolds, T.H. Gan, P. Purnell, M. Pharaoh.
Identification of failure modes in glass/polypropylene composites by means of the primary frequency content of the acoustic emission event.
Compos Sci Technol, 64 (2004), pp. 1819-1827
[10]
S. Suresh, N. Shenbaga Vinayaga Moorthi, S.C. Vettivel, N. Selvakumar, G.R. Jinu.
Effect of graphite addition on mechanical behavior of Al6061/TiB2 hybrid composite using acoustic emission.
Mater Sci Eng A, 612 (2014), pp. 16-27
[11]
C.M. Anand Partheeban, M. Rajendran, S.C. Vettivel, S. Suresh, N.S.V. Moorthi.
Mechanical behavior and failure analysis using online acoustic emission on nano-graphite reinforced Al6061-10TiB2 hybrid composite using powder metallurgy.
Mater Sci Eng A, 632 (2015), pp. 1-13
[12]
G. Chen, H. Luo, H. Yang, T. Zhang, S. Li.
Water effects on the deformation and fracture behaviors of the multi-scaled cellular fibrous bamboo.
Acta Biomater, 65 (2018), pp. 203-215
[13]
G. Chen, H. Luo, S. Wu, J. Guan, J. Luo, T. Zhao.
Flexural deformation and fracture behaviors of bamboo with gradient hierarchical fibrous structure and water content.
Compos Sci Technol, 157 (2018), pp. 126-133
[14]
X. Li, J. Zhang, S. Shen, Y. Wang, X. Song.
Effect of tempering temperature and inclusions on hydrogen-assisted fracture behaviors of a low alloy steel.
Mater Sci Eng A, 682 (2017), pp. 359-369
[15]
Q. Sun, H. Di, J. Li, B. Wu, R.D.K. Misra.
A comparative study of the microstructure and properties of 800MPa microalloyed C-Mn steel welded joints by laser and gas metal arc welding.
Mater Sci Eng A, 669 (2016), pp. 150-158
[16]
W.W. Bose Filho, A.L.M. Carvalho, R. Bowen.
Micromechanisms of cleavage fracture initiation from inclusions in ferritic welds – Part I. Quantification of local fracture behaviour observed in notched test pieces.
Mater Sci Eng A, 460 (2007), pp. 436-452
[17]
W.W. Bose Filho, A.L.M. Carvalho, R. Bowen.
Micromechanisms of cleavage fracture initiation from inclusions in ferritic welds – Part II. Quantification of local fracture behaviour observed in fatigue pre-cracked test pieces.
Mater Sci Eng A, 452 (2007), pp. 401-410
[18]
G.Z. Wang, Y.G. Liu, J.H. Chen.
Investigation of cleavage fracture initiation in notched specimens of a C-Mn steel with carbides and inclusions.
Mater Sci Eng A, 369 (2004), pp. 181-191
[19]
K.C. Hsieh, S.S. Babu, J.M. Vitek, S.A. David.
Calculation of inclusion formation in low-alloy-steel welds.
Mater Sci Eng A, 215 (1996), pp. 84-91
[20]
A. Sánchez Osio, S. Liu, D.L. Olson.
The effect of solidification on the formation and growth of inclusions in low carbon steel welds.
Mater Sci Eng A, 221 (1996), pp. 122-133
[21]
H. Jing, L. Huo, Y. Zhang.
Simulation of stable crack growth for welded joints including strength mismatching.
J Mater Sci Technol, 15 (1999), pp. 567-570
[22]
D.R. Vincenzo, B. Bruno, C. France.
Defect detection in stainless steel ‘Uranus 45’ TIG welded joints by acoustic emission.
11th world conference on nondestructive testing, pp. 85-90
[23]
U. Kumar, I. Yadav, S. Kumari, K. Kumari, N. Ranjan, R.K. Kesharwani, et al.
Defect identification in friction stir welding using discrete wavelet analysis.
Adv Eng Softw, 85 (2015), pp. 43-50
[24]
X. Long, G. Cai, L. Svensson.
Investigation of fracture and determination of fracture toughness of modified 9Cr-1Mo steel weld metals using AE technique.
Mater Sci Eng A, 270 (1999), pp. 260-266
[25]
D. Racks.
Acoustic emission from welds as indicator of cracking.
Mater Sci Tech Lond, 3 (1987), pp. 1062-1066
[26]
X. Liu, D. Xiao, Y. Shan, Q. Pan, T. He, Y. Gao.
Solder joint failure localization of welded joint based on acoustic emission beam forming.
Ultrasonics, 74 (2017), pp. 221-232
[27]
Z. Han, H. Luo, J. Cao, H. Wang.
Acoustic emission during fatigue crack propagation in a micro-alloyed steel and welds.
Mater Sci Eng A, 528 (2011), pp. 7751-7756
[28]
C.K. Mukhopadhyay, T. Jayakumar, B. Raj, K.K. Ray.
The influence of notch on the acoustic emission generated during tensile testing of nuclear grade AISI type 304 stainless steel.
Mater Sci Eng A, 276 (2000), pp. 83-90
[29]
S.A.S. Vanini, M. Abolghasemzadeh, N. Kordani, M. Sarmadian.
Dynamic notch fracture toughness J(C)(V) assessment of functionally austenitic–martensitic graded steels.
Mater Sci Eng A, 546 (2012), pp. 198-206
[30]
R. Qu, P. Zhang, Z. Zhang.
Notch effect of materials: strengthening or weakening?.
J Mater Sci Technol, 30 (2014), pp. 599-608
[31]
R.T. Qu, Z.F. Zhang.
A universal fracture criterion for high-strength materials.
[32]
R. Gutkin, C.J. Green, S. Vangrattanachai, S.T. Pinho, P. Robinson, P.T. Curtis.
On acoustic emission for failure investigation in CFRP: pattern recognition and peak frequency analyses.
Mech Syst Signal Process, 25 (2011), pp. 1393-1407
[33]
P.J.D. Groot, P.A.M. Wijnen, R.B.F. Janssen.
Real-time frequency determination of acoustic emission for different fracture mechanisms in carbon/epoxy composites.
Compos Sci Technol, 55 (1995), pp. 405-412
[34]
W. Mao, I. Towhata.
Monitoring of single-particle fragmentation process under static loading using acoustic emission.
Appl Acoust, 94 (2015), pp. 39-45
[35]
A. Tranchot, A. Etiernble, P.X. Thivel, H. Idrissi, L. Roue.
In-situ acoustic emission study of Si-based electrodes for Li-ion batteries.
J Power Sources, 279 (2015), pp. 259-266
[36]
Z. Han, H. Luo, Y. Zhang, J. Cao.
Effects of micro-structure on fatigue crack propagation and acoustic emission behaviors in a micro-alloyed steel.
Mater Sci Eng A, 559 (2013), pp. 534-542
Journal of Materials Research and Technology

Subscribe to our newsletter

Article options
Tools
Supplemental materials
Cookies policy
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.