Journal of Materials Science & Technology  2020 , 42 (0): 75-84 https://doi.org/10.1016/j.jmst.2019.05.075

Orginal Article

Visualization of microstructural factors resisting the crack propagation in mesosegregated high-strength low-alloy steel

Shuxia Wanga, Chuanwei Lia*, Lizhan Hanb, Haozhang Zhonga, Jianfeng Gucd**

a Institute of Materials Modification and Modeling, School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai, 200240, China
b Shanghai Key Laboratory of Materials Laser Processing and Modification, Shanghai Jiao Tong University, Shanghai, 200240, China
c Collaborative Innovation Center for Advanced Ship and Deep Sea Exploration, Shanghai Jiao Tong University, Shanghai, 200240, China
d Materials Genome Initiative Center, Shanghai Jiao Tong University, Shanghai, 200240, China

Corresponding authors:   *Corresponding author. E-mail addresses: li-chuanwei@sjtu.edu.cn (C. Li).**Corresponding author. gujf@sjtu.edu.cn (J. Gu).

Received: 2019-02-12

Revised:  2019-05-3

Accepted:  2019-05-10

Online:  2020-04-01

Copyright:  2020 Editorial board of Journal of Materials Science & Technology Copyright reserved, Editorial board of Journal of Materials Science & Technology

More

Abstract

While relationship between fracture mechanism and homogeneous microstructures has been fully understood, relationship between fracture mechanism and inhomogeneous microstructures such as the mesosegregation receives less attention as it deserves. Fracture mechanism of the high-strength low-alloy (HSLA) steel considering the mesosegregation was investigated and its corresponding microstructure was characterized in this paper. Mesosegregation refers to the inhomogeneous distribution of alloy elements during casting solidification, and leads to the formation of positive segregation zones (PSZ) and negative segregation zones (NSZ) in ingots. The fracture surface of impact sample exhibits the quasi-cleavage fracture at -21 °C, and is divided into ductile and brittle fracture zone. Meanwhile, the PSZ and NSZ spread across ductile and brittle fracture zone randomly. In ductile fracture zone, micro-voids fracture mechanism covers the PSZ and NSZ, and higher deformation degree is shown in the PSZ. In brittle fracture zone, secondary cleavage cracks are observed in both PSZ and NSZ, but present bigger size and higher quantity in the NSZ. However, some regions of the PSZ still present micro-voids fracture mechanism in brittle fracture zone. It reveals that the microstructures in the PSZ exhibit a higher resistance ability to crack propagation than that in the NSZ. All observations above provide a better visualization of the microstructural factors that resist the crack propagation. It is important to map all information regarding the fracture mechanism and mesosegregation to allow for further acceptance and industrial use.

Keywords: High-strength low-alloy steel ; Heavy forgings ; Mesosegregation ; Inhomogeneous microstructures ; Fracture mechanism

0

PDF (11590KB) Metadata Metrics Related articles

Cite this article Export EndNote Ris Bibtex

Shuxia Wang, Chuanwei Li, Lizhan Han, Haozhang Zhong, Jianfeng Gu. Visualization of microstructural factors resisting the crack propagation in mesosegregated high-strength low-alloy steel[J]. Journal of Materials Science & Technology, 2020, 42(0): 75-84 https://doi.org/10.1016/j.jmst.2019.05.075

1. Introduction

Impact toughness of heavy structural components such as the high-strength low-alloy steel (HSLA) has received more and more attention, as the premature failure of them can cause the catastrophic accident and is unacceptable. Among the critical mechanical properties of heavy structural components, impact toughness is undoubtedly one of the most important factors. It plays a key role in mechanical properties of components, especially for the heavy forgings used in nuclear power plants, whose critical safety is always evaluated by the minimum impact value. At most of the time, Charpy V-notched (CVN) impact sample can be used to measure the impact toughness due to its smaller sample size and simple pendulum impact tester [1]. Generally speaking, characterization of fracture surface could reveal the fracture mechanism, but could not provide the detailed information of crystallographic and deformation during crack propagation [2]. Nevertheless, the characterization of deformed microstructures and cleavage cracks at cross-section surface in terms of the electron backscattering diffraction (EBSD) technic could provide better visualized evidences to the investigation of fracture mechanism [3,4].

As to the influence factors to impact toughness, many investigations are focused on the homogeneous microstructures [2,[5], [6], [7], [8], [9]]. Impact toughness is determined by the morphology and size of grains and carbide particles. For example, fine lower bainitic microstructure with homogeneously distributed fine carbides provides better impact toughness, while coarse upper bainite and precipitation of coarse inter-lath carbides presents the inferior impact properties [5,7]. Besides, martensite has the positive effect on impact toughness, compared with the bainite, which is composed of smaller blocks and more high-angle grain boundaries and exhibits effective resistance to the crack propagation [2,6,8,9].

However, the influence factor such as inhomogeneous microstructures inside a sample has been discussed rarely and it lacks systematic investigation to the effects on impact toughness [[10], [11], [12]]. It is worth noting that the inhomogeneous microstructures, such as inclusion banding, different grain size or phase [13,14] and the segregated microstructures, would affect the impact toughness scatter in the ductile-to-brittle transition temperature (DBTT) range [15]. For example, mesosegregation, inducing inhomogeneous distribution of elements during casting solidification, is an inevitable phenomenon in heavy forgings. Because solubility of most elements in liquid phase is higher than that in solid phase, solute atoms such as Mn, Mo, and Ni etc. are expelled from solid phase to liquid phase during solidification [16]. Then continual enrichment of solutes in liquid phase and depletion in solid phase would lead to formation of positive segregation zones (PSZ) and negative segregation zones (NSZ), respectively [5].

It has been identified that mesosegregation can lead to deterioration of impact toughness in HSLA steel [5,10,[17], [18], [19], [20], [21], [22], [23]]. However, there are very limited studies on the relationship between mesosegregation and fracture mechanism. Mapping all information of them is significant, especially for detailed research about fracture mechanism discrepancies between PSZ and NSZ, which allows for further acceptance and industrial use. This paper presents a detailed microstructural characterization of crack propagation across the PSZ and NSZ, by scanning electron microscopy (SEM) and electron back-scattered diffraction (EBSD). Furthermore, in the viewpoint of visualization to the microstructural factor resisting the crack propagation, our findings underscore that the microstructures discrepancies between PSZ and NSZ would induce the resistance discrepancies of crack propagation significantly.

2. Experimental

The high-strength low-alloy steel used for experiment was taken from a 140 mm thick barrel of evaporator, and had not suffered the subsequent homogenization heat treatment. The sample was obtained at the 1/4 T position distance from the inner surface of barrel. Table 1 was the average chemical composition of high-strength low-alloy steel. The temperatures of Ac1 and Ac3 were measured as 730 °C and 805 °C by dilatometer with heating rate of 5 °C/s [18,24]. The as-received status steel was austenitized at 900 °C for 5 h and cooled to room temperature in air. Then it was tempered at 650 °C with a dwell time of 5 h and cooled to room temperature in furnace [24]. The standard CVN impact sample [25] was cut as Fig. 1(b) and tested at -21 °C. Testing temperature was controlled within ±2 °C via the mixture of liquid nitrogen and methanol in a medium container. Following the ASTM E23 [25] standard testing procedure, the sample was soaked in medium for 15 min to homogenize the temperature, and tested within 3 s after removed from the bath.

Table 1   Chemical composition (wt%) of the as-received HSLA steel.

CMnMoNiCrSiAlCuPFe
0.201.420.490.860.180.180.0140.030.003Bal.

New window

Fig. 1.   Three-dimensional characterization of mesosegregation in the ring shape heavy forging: (a) image of mesosegregation distribution in the view of three-dimensional; (b) sampling schematic diagram of CVN sample corresponding to (a); (c-e) mesosegregation distributions in the plane of XOZ, XOY and YOZ.

Microstructures in the PSZ and NSZ were characterized using the optical microscopy (OM; Zeiss Axio Observer A1), scanning electron microscopy (SEM; TESCAN-LYRA3) and electron back-scattered diffraction (EBSD; NordlysMax3, Oxford Instruments, United Kingdom). OM and SEM specimens were polished mechanically and etched by ethyl alcohol solution consisting of 4 % nital alcohol solution. EBSD specimen was polished using 50 nm colloidal silica particles (nano-chemical-mechanically Buehler-VibroMet polishing) for 2 h after mechanically polished to a surface roughness of 0.5 μm. EBSD measurements were obtained by field-emission SEM (TESCAN-LYRA3) equipped with an EBSD fast acquisition system operating at 20 kV/4 nA with the specimen tilted at 70°, working distance of 9 mm and at the step size of 0.2 μm. The captured data were post-processed using the Channel 5 software package (Oxford Instruments, United Kingdom). An electron-probe microanalysis (EPMA; EPMA-1720, Shimadzu, Japan) was conducted to detect any concentration discrepancies for the elements in the PSZ and NSZ.

3. Results and discussion

3.1. Microstructures characterization

The steel with mesosegregation is taken from the evaporator barrel and cut as the cubic sample. Fig. 1(a) displays the three dimensional image of the sample after polishing and etching of three faces with the common vertex. To facilitate the analysis, a spatial rectangular coordinates is built in the cubic sample, and the direction of OX, OY, and OZ stand for the tangential direction (TD), radial direction (RD), and axial direction (AD) of the evaporator barrel, respectively. As shown in Fig. 1(c, d), mesosegregation in XOZ and XOY plane exhibits parallel stripes along with the direction of OX (TD). Fig. 1(e) presents cloud-shape mesosegregation in YOZ plane. Fig. 1(b) schematically displays the sampling diagram of CVN impact sample. Hence the fracture surface of sample is parallel to YOZ plane, and the cross-section surface is parallel to XOY plane.

Fig. 2(a) shows the OM image of mesosegregation distribution under low magnification and the EPMA surface scanning analysis of elements to the corresponding locations. Obviously, elements of C, Mn, Mo, Ni are enriched in the dark regions (PSZ) and depleted in the light regions (NSZ). Fig. 2(b, c) displays the OM images of the PSZ and NSZ selected in Fig. 2(a). Generally speaking, under the identical cooling speed, higher element concentration in the PSZ would lead to better hardenability during quenching. Therefore, as shown in Fig. 2(b), the black and dark grey regions in the PSZ are the mixed microstructures of tempered martensite (M) and tempered lower bainite (BL). Besides, the white small bulk-shape Bainite ferrite (BF) with few carbides could also be observed and is distributed around the M and BL. Actually, it is not easy to clarify M and BL after tempering, especially under OM observation, because small and uniform carbides are precipitated simultaneously from the ferrite matrix both in M and BL during tempering, which agree with the similar distribution [24,26]. Fig. 2(d) presents the further distinguishable SEM observation of them, which displays different lath width of ferrite matrix. That is, the ferrite lath of M is narrower and longer, and the shape of parallel needle is observed, while the ferrite lath of BL is wider and shorter, and the shape of parallel feather is observed. In addition, as shown in Fig. 2(f), the distinguishable discrepancies of M and BL in the PSZ could also be identified by the EBSD analyzed inverse pole figures (IPF) map. It is obvious that the misorientation between the laths of M is higher than the laths of BL. That is, the lath boundary density of M is higher than BL, and it also indicates that more energy is stored in M.

Fig. 2.   Microstructural characterization: (a) OM image of mesosegregation and corresponding EPMA surface scanning analysis results of elements; (b, c) OM images in the PSZ and NSZ; (d, e) SEM images in the PSZ and NSZ; (f, g) EBSD analyzed inverse pole figures (IPF) maps in the PSZ and NSZ.

On the contrary, the lower element concentration in the NSZ results in poorer hardenability. As shown in Fig. 2(c) and (e), microstructures in the NSZ are tempered upper bainite (BU) and some allotriomorphic ferrite (αa). Carbides are precipitated along with the laths boundaries of BU and coarsened during tempering [24,26]. Furthermore, the black carbides stacked at the boundary of primary austenite grains are decomposed from the M/A islands during tempering [24,26]. As shown in Fig. 2(g), IPF map presents the misorientation and effective grain size of microstructures in the NSZ clearly. The laths of BU are wider and the misorientations between them are smaller than that in the PSZ.

3.2. Fractograph

SEM observations of fracture surface are shown in Fig. 3(a), which consists of the fibrous region, radial region and shear lip. As shown in Fig. 3(b), microstructures in fibrous region are dimples and correspond to ductile fracture. Hence the fibrous region could also be regarded as ductile fracture zone. The random dimples with different size and depth also means different micro-voids nucleation sites during ductile fracture. Meanwhile, Fig. 3(c) shows the microstructures in radial region, which display the large areas of cleavage surface and several rough stripes along with the crack propagation direction. Hence the radial region could also be called brittle fracture zone. The river patterns in cleavage surface are presented in Fig. 3(d), with tear ridge and indicate the typical quasi-cleavage fracture. A small secondary cleavage crack cuts off the river pattern and extends into the underneath of fracture surface, which could probably be observed in cross-section surface. Fig. 3(e) amplifies the microstructures in rough stripes, which are small, uniform and shallow dimples. Reasons for the appearance of rough stripes are unclear. It may be related to the microstructural heterogeneity and need the further research.

Fig. 3.   SEM images showing the fracture surface morphology of CVN impact sample: (a) entire fracture surface morphology in low-magnification; (b) dimples in fibrous region shown in (a); (c) microstructures in radial region shown in (a); (d) river patterns in cleavage facets of crack radial region shown in (c); (e) dimples in rough stripes of crack radial region shown in (c).

As shown in Fig. 4(a,c), the demarcation line between ductile and brittle fracture zone is marked in facture surface. Then the cross-section surface in Fig. 4(c) could be divided into matching fracture zones according to this marked line. Fig. 4(c) presents the distribution of PSZ and NSZ in cross-section surface, which are perpendicular to the fracture surface and spread across ductile and brittle fracture zone randomly. Fig. 4(d) displays the magnified OM image of ductile fracture zone in cross-section surface, which covers the PSZ and NSZ. Fig. 4(e) provides the deformed microstructures of the PSZ and NSZ in brittle fracture zone. Furthermore, a string of secondary cleavage cracks located in the NSZ of brittle fracture zone are amplified in Fig. 4(f).

Fig. 4.   SEM (a, b) and OM (c-f) images showing the fracture surface morphology of sample: (a) schematic showing the method used to cut the CVN impact sample; (b) determination of demarcation line between ductile and brittle fracture zone in fracture surface; (c) distribution characteristic of mesosegregation among ductile and brittle fracture zone in cross-section surface; (d) deformation characteristic of PSZ and NSZ in ductile fracture zone shown in (c); (e) deformation characteristic of PSZ and NSZ in brittle fracture zone shown in (c); (f) second crack in the NSZ distributed in brittle fracture zone shown in (c).

3.3. Deformation analysis along the crack propagation

3.3.1. Deformation analysis in ductile fracture zone

The deformed microstructures at the tearing tip of ductile fracture zone are shown in Fig. 5(a), including both the PSZ and NSZ. Fig. 5(c, d) contrasts the SEM images of deformed microstructures in the PSZ and NSZ selected in Fig. 5(a), and both displays the micro-voids fracture mechanism. As shown in Fig. 5(c), small and uniform micro-voids are scattered among the matrix of M. It indicates that the dispersed tiny carbides are nucleation sites of micro-voids. Besides, bigger and ununiform voids are distributed at the boundaries of M and BL. It reveals that the stacked carbides at the boundaries play the role of voids nucleation sites as well. These voids are stretched along with tearing direction of microstructures and are connected together end to end. Identically, as shown in Fig. 5(d), the BU in the NSZ also displays the micro-voids nucleation mechanism. The deformed voids distribute both at the lath boundaries of BU and the stacked carbides. It should be mentioned that the stacked carbides are decomposed from the M/A islands during tempering [24,26]. However, the size and quantity of voids in the NSZ are smaller than that in the PSZ, which means lower deformation degree in the NSZ. The identical conclusion can also be demonstrated by EBSD analyzed strain contouring map in Fig. 5(b). It is obvious that the strain in the PSZ is higher than that in the NSZ. Furthermore, the strain in the little cusp of tearing tip is extremely high, even if in the NSZ. That is, the deformed regions are close to the vicinity of shear facture surface and caused by the accumulation of plastic deformation heat under the rapid loading condition [27,28].

Fig. 5.   Deformation analysis results of ductile fracture zone in cross-section surface: (a) OM microstructural image of ductile fracture zone; (b) EBSD analyzed strain contouring map corresponding to (a); (c, d) SEM images of deformed microstructures in the PSZ and NSZ selected in (a); (e, f) EBSD analyzed local misorientation (LM) maps of deformed microstructures in the PSZ and NSZ selected in (b); (g) statistical chart of data extracted from the LM maps in (e) and (f).

Local misorientation (LM) maps for selected areas of PSZ and NSZ in Fig. 5(b) are presented in Fig. 5(e, f). As shown in Fig. 5(g), the data obtained during EBSD collection and extracted from the LM maps are drawn as the statistical chart. It is observed that local misorientation in the PSZ shows more dispersive distribution and has higher relative frequency when below 0.5°. It also indicates that microstructural deformation in the PSZ is more radical and homogeneous. However, the distribution of local misorientation in the NSZ is narrower and mainly concentrates around 1.2°. As shown in Fig. 5(d), microstructural deformation in the NSZ is ununiform, and the micro-voids are focused on the lath boundaries of BU.

Fig. 6(a) presents the OM image of another selected region in ductile fracture zone, and the corresponding strain contouring map is shown in Fig. 6(b). Identically, the microstructures in the PSZ displays higher strain than those in the NSZ. Fig. 6(c, d) contrasts the SEM images of fracture surface corresponding to the PSZ and NSZ, respectively. It is obvious that the dimples in fracture surface are smaller and deeper corresponding to the PSZ, but bigger and shallower corresponding to the NSZ. According to the micro-voids fracture mechanism, deeper and smaller dimples mean higher strain and severer deformation degree. Hence the size and depth of dimples in fracture surface could also offer an evidence that the deformation degree in the PSZ is higher than that in the NSZ. In cross-section surface, as selected in Fig. 6(c, d), the corresponding deformed microstructures close to the edge of fracture surface are shown in Fig. 6(e, f). It is obvious that size and quantity of micro-voids in the PSZ are higher than those in the NSZ, which also reflects higher deformation degree in the PSZ.

Fig. 6.   Analysis results of the adiabatic shear bands (ASBs) adjacent to the ductile fracture cracks in cross-section surface: (a) OM image of the ASBs; (b) EBSD analyzed strain contouring map corresponding to (a); (c, d) dimples corresponding to the PSZ and NSZ in fracture surface perpendicular to the cross-section surface shown in (a); (e, f) SEM images of the deformed microstructures in the PSZ and NSZ at cross-section surface near to the edge of fracture surface.

3.3.2. Deformation analysis in brittle fracture zone

Fig. 7(a) displays the amplified OM image of Fig. 4(e), and the corresponding strain contouring map is shown in Fig. 7(b). It is observed that the selected regions (Ⅰ) and (Ⅱ) in the PSZ present two fracture mechanisms in brittle fracture zone. Fig. 7(c-e) summarizes the SEM image, LM and IPF maps of region (Ⅰ), and it proves that micro-voids fracture mechanism occurs not only in ductile fracture zone normally, but also in brittle fracture zone unexpectedly. It may be related to the rough stripes in fracture surface, as shown in Fig. 3(c), whose microstructures are observed as the small, uniform and shallow dimples. However, there are very limited evidences for this speculation and it needs further experimental verification.

Fig. 7.   Deformation analysis results of brittle fracture zone in cross-section surface: (a) OM microstructural image of brittle fracture zone; (b) EBSD analyzed strain contouring map corresponding to (a); (c, f) SEM microstructural images of the deformed PSZ, according to regions (Ⅰ) and (Ⅱ) shown in (a), respectively; (d, g) EBSD analyzed LM maps of the deformed PSZ, according to regions (Ⅰ) and (Ⅱ) shown in (a), respectively; (e, h) EBSD analyzed IPF maps of the deformed PSZ, according to regions (Ⅰ) and (Ⅱ) shown in (a), respectively.

The analysis of two short secondary cleavage cracks in region (Ⅱ) is shown in Fig. 7(f-h). The length of them is measured as 13 and 8 μm, respectively. Furthermore, it is observed that both two cracks are opened and ended within grains. The microstructure between two cracks is martensite, and it indicates good resistance ability to crack propagation in the PSZ. Fig. 7(g) shows matching LM map of cracks, and the significant plastic work occur around the cracks. It is reported that the microstructural factors arresting microcrack contain high-angle boundaries [29,30], second phase [31] and so on. Furthermore, plastic work associated with crack opening and resistance at boundaries is demonstrated. Potential energy of system is consumed and converted into the crack surface energy via the plastic work to open the crack [32]. Because the plastic work could be visualized by LM map, the concentrated plastic work in yellow at point (ⅰ) indicates that it is the location of crack opening. Furthermore, point (ⅲ) also displays higher plastic work in yellow and indicates as the stopped point of crack. As shown in Fig. 7(h), points (ⅰ) and (ⅲ) are both located at grain boundaries, which means that the plastic work is associated with the larger resistance at grain boundaries as well. In addition, Fig. 7(g) also displays a small change of color at point (ⅱ), which could be observed as a small jag in Fig. 7(f). It is expected that a small resistance would be encountered at point (ⅱ) during crack propagation path from point (ⅰ) to (ⅲ). Actually, as expected from the IPF map in Fig. 7(h), the appearance of lath boundaries at point (ⅱ) increases the resistance slightly and induces a small change to crack propagation.

Fig. 8(a) presents a string of secondary cleavage cracks in the NSZ of brittle fracture zone, which cross BU and αa in a transcrystalline fracture mode. However, the cracks in the NSZ have larger size and higher quantity than that in the PSZ. The crack propagation path is studied by strain contouring map in Fig. 8(b). It is obvious that the strain is concentrated at the deflection of cleavage crack path. As shown in Fig. 8(c-h), two selected regions of deflective cracks in Fig. 8(a) are analyzed by LM and IPF maps. This study aims to understand the role of boundaries types, which can induce the path deflection of crack propagation. Besides, EBSD analysis of crystallographic orientation in the vicinity of these secondary cracks could shed light on the mechanism of crack arrest [[33], [34], [35]].

Fig. 8.   Secondary cleavage cracks’s analyses in the NSZ of brittle fracture zone: (a) SEM image of secondary cleavage cracks; (b) EBSD analyzed strain contouring map corresponding to (a); (c, f) SEM image of secondary cleavage cracks according to regions (Ⅰ) and (Ⅱ) shown in (a), respectively; (d, g) LM maps of secondary cleavage cracks according to regions (Ⅰ) and (Ⅱ) shown in (a), respectively; (e, h) IPF maps of secondary cleavage cracks, according to regions (Ⅰ) and (Ⅱ) shown in (a), respectively.

The crack propagation path in region (Ⅰ) is summarized in Fig. 8(c-e), which has small deviation between EBSD map and SEM image. Because a deposition layer has been generated during EBSD scanning before the SEM in situ observation. It needs slight polishing to remove this layer and further microstructural etching for subsequent SEM observation. The slight depth loss of specimen surface would result in small deviation of crack path, which nearly has no influence to relationship between SEM microstructures and EBSD maps. As shown in Fig. 8(d, e), LM and IPF maps display that the crack gets deflected when it runs into the high angle grain boundaries (black lines in the LM and with the misorientation higher than 10°). From the depth direction of specimen surface, the SEM image displayed in Fig. 8(c) reveals two discontinuous cracks, which are underneath a coherent crack presented in Fig. 8(d, e). As shown in Fig. 8(c), it also indicates the grain boundaries resistance to crack A and the reinitiation of new crack B.

The crack deflection in region (Ⅱ) is described in Fig. 8(f-h). It reveals that the crack encounters with both high angle grain boundaries (misorientation higher than 10°) and low angle sub-boundaries (misorientation from 2° to 10°) in propagation path. Firstly, the crack encounters a grain boundary indicated as point (ⅰ), where it gets deflected for the first time. Then the adjacent grain is divided into two parts by this deflected crack. As shown in Fig. 8(g), the high LM value distributed in crack neighbour area is shown as the red sub-boundaries. Next, this crack propagates across the third grain and gets deflected for the second time at point (ⅱ). Then the crack extends into the third grain and stops at point (ⅲ), where it displays the sub-boundaries of bainite-ferrite lath. Meanwhile, another new crack in the vicinity of bainite-ferrite lath inside the same grain is reinitiated and continues to extend. However, this new crack ends at grain boundaries again, and then the third crack is reinitiated again inside this grain. All deflections of cracks above reveal that high angle grain boundaries significantly resist the crack propagation other than the low angle sub-boundaries.

4. Conclusions

The important information regarding fracture mechanism and mesosegregation is mapped by SEM and EBSD analysis, which have provided a better method for the visualization of microstructural factor that resists deformation and crack propagation. Present investigations visualize the discrepancies of fracture mechanism in the PSZ and NSZ, where the following conclusions can be obtained.

(1) For the high-strength low-alloy steel with mesosegregation, CVN impact sample tested at -21 °C displays the combination of ductile and brittle fracture modes.

(2) In ductile fracture zone, both the PSZ and NSZ manifest as a micro-void related fracture mechanism, while the deformation degree in the PSZ is higher than that in the NSZ.

(3) In brittle fracture zone, the NSZ manifests as a secondary crack propagation fracture mechanism. However, both micro-voids fracture and cleavage fracture mechanism occur in the PSZ. But the size and quantity of secondary cleavage cracks in the PSZ are smaller than that in the NSZ.

(4) The discrepancies of element concentration resulted from mesosegregation would induce microstructural discrepancies in the PSZ and NSZ. Then the discrepancies of resistance ability to crack propagation during rapid impact test are affected by these microstructural discrepancies.

Acknowledgement

This work was financially supported by the National Natural Science Foundation of China (No. 51801126).


/