Hydrogen-assisted decohesion and localized plasticity in dual-phase steel

TLDR: In this article, a high-resolution scanning electron microscopy-based damage quantification technique has been employed to identify strain regimes where damage nucleation and damage growth take place, both with and without hydrogen precharging.

About: This article is published in Acta Materialia.The article was published on 2014-05-15 and is currently open access. It has received 327 citations till now. The article focuses on the topics: Hydrogen embrittlement & Damage tolerance.

Figures

Fig. 4. Damage evolution curves plotted against local plastic strain with and without hydrogen charging: (a) damage area fraction, (b) number of damage incidents per area and (c) average crack size. The total elongations of the used specimens with and without hydrogen were 6% and 12%, respectively.

Fig. 5. In situ SEM observations during tensile testing without hydrogen charging in (a) the damage nucleation regime (11% macroscopic strain), (b) the damage growth regime (12% macroscopic strain) and (c) the late damage growth regime (at the vicinity of the fractured part). The strains are macroscopic strain, not local plastic strain. The dotted red lines in (c) indicate the fracture surface. According to Fig. 3, the local strain in the vicinity of fracture surface is speculated to be about 34% plastic strain. (a0–c0) The magnified images corresponding to the regions outlined in (a–c). The red dotted lines in (a0) indicates ferrite (F)/martensite (M) interfaces. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 10. (a) IQ, (b) RD IPF and (c) KAM maps at 7% local plastic strain of the sample with hydrogen precharging. (d–h) Schematics demonstrating how the crack shown in (a–c) nucleated, propagated and opened. F: ferrite, M: martensite, LS: localized slip.

Fig. 6. ECC images at two different magnifications showing damage evolution as a function of plastic strain in the fractured specimen with hydrogen precharging in (a, b) the end of the damage incubation regime (1–2% local plastic strain), (c, d) the early damage nucleation regime (4–5 local plastic strain), (e, f) the late damage nucleation regime (6–7% local plastic strain) and (g, h) the damage growth regime ( 10%). The magnified images correspond to the squares in the lower magnification image. Some of the cracks are indicated by the red arrows. The broken red line in (h) indicates prior austenite grain or ferrite/martensite phase boundaries around the crack. F and M indicate ferrite and martensite, respectively. The total elongations of the used specimens with and without hydrogen were 6% and 12%, respectively. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 9. (a) KAM maps at different plastic strain levels with and without hydrogen charging. The ferrite and martensite were separated in terms of grain average IQ value. (b) KAM value distribution of the specimen with and without hydrogen charging at 0% and 7% local plastic strains. The condition for the calculation of KAM values is the same with Fig. 1.

Table 1 Chemical composition of the present steel.

Frequently asked questions

Q1. What is the role of HEDE in the cracking of ferrite?

While the ferrite/martensite cracking is also attributed to decohesion, the ferrite/martensite boundary sliding and ferrite cracking mechanisms are associated with enhanced dislocation mobility, i.e. the HELP mechanism. 

Q2. What is the cause of the partial brittle fracture surface?

The martensite-related cracking associated with decohesion is considered to be the cause of the partial brittle fracture surface. 

Q3. How was the in situ bending test performed?

Since hydrogen desorption during observation is a critical issue for in situ SEM experiments, the in situ bending test was carried out under an optical microscope in a specimen that was precharged with hydrogen at a cathodic current density of 2 A mÿ2 for 1 h. 

Q4. What were the methods used to observe the microstructure of the specimens?

Ex situ microstructure observations were performed by optical microscopy, secondary electron imaging, EBSD and ECCI to observe the initial microstructure consisting of ferrite and martensite and the deformation-induced cracks with and without hydrogen charging. 

Q5. Why is the hydrogen charge in a ferrite steel higher than in a marten?

Since the effective diffusion coefficient of hydrogen is higher in ferrite than in martensite (i.e. due to the high density of dislocations in martensite acting as trapping sites, decreasing the diffusion speed), the diffusion coefficient of hydrogen in DP steels is considered to be higher than that in martensitic steels. 

Q6. How long was the deformed specimen in air?

The deformed specimen was left in air at ambient temperature for 10 days to desorb most of the diffusible hydrogen, then deformed again until fracture. 

Q7. What was the method used to measure the local strains on the tensile specimens?

Optical images, captured during the tensile tests, were postprocessed by digital image correlation (DIC) to measure the local strains on the tensile specimens [51,52]. 

Q8. What is the diffusivity of hydrogen in a DP steel?

Since the effective diffusion coefficient of hydrogen was reported to be 3.7 10ÿ11 m2 sÿ1 in an as-quenched low-alloy martensitic steel [45], the diffusivity of hydrogen in the DP steel is higher than the diffusivity of the martensitic steel. 

Q9. What mechanisms were found in previous studies on the damage evolution of DP steel without hydrogen charging?

Previous studies on the damage evolution of DP steel without hydrogen charging also showed such martensite and ferrite/martensite interface decohesion mechanisms [5,7,58]. 

Q10. What is the third regime of damage evolution in DP steels?

The third regime is characterized by the fact that, upon further straining, a critical strain for the onset of crack growth is reached, and crack propagation and opening take place, increasing the average crack size until occurrence of final fracture (crack growth regime). 

Q11. What is the main cause of the hydrogen-assisted martensite cracking?

This clearly suggests that the main cause of the hydrogen-assisted martensite cracking and ferrite/ martensite cracking is the influence of hydrogen not on the dislocation pile-ups but, rather, directly on the cohesive interface energy. 

Q12. Why are these more practical approaches not considered in the literature?

These more practical approaches are partially due to hydrogen-diffusion-related difficulties in experimentation, i.e. hydrogen can be released from the material, depending on the charging and environmental holding parameters. 

Q13. What was the damage evolution of the hydrogen-charged specimen?

In the case of the hydrogen-charged specimen, the authors observed that all three damage evolution stages were significantly reduced to much smaller strains by the hydrogen uptake. 

Q14. What is the reason for the brittle fracture regions of both types of specimens?

Since the size of each brittle region is larger than the grain size (shown in Fig. 1a), the brittle fracture regions of both types of specimens result from brittle crack propagation through multiple grains. 

Q15. What is the effect of hydrogen on the crack-arresting property of ferrite?

when considering the drop in the critical strain where the crack growth regime is initiated (Fig. 4c), the authors propose that hydrogen strongly decreases the crack-arresting property of ferrite.