Stress corrosion cracking behavior of EV31A and Mg-Mn magnesium alloys
2017-05-15T04:50:24Z (GMT) by
Magnesium alloys are the lightest metallic engineering materials, and hence they are quite attractive for light weight applications, such as automotive and aerospace industries. The primary motive of alloy development has been to impart high strength and creep resistance. However, all Mg-alloys are vulnerable to various forms of corrosion including stress corrosion cracking (SCC) due to high thermodynamic driving force. So research relating to various form of corrosion including SCC is important. In this work mechanism(s) of SCC of a sand cast alloy EV31A (Mg-0.28Zn-0.6Zr-2.9Nd-1.5Gd-0.14 other rare earth (RE) (each <0.01%)-<0.001Ag-<0.0001Cu-0.003Fe-<0.0001Ni) and a wrought alloy Mg-Mn hot rolled (Mg-0.1Al-0.3Zn-1.75Mn-0.25Ce-0.3(other rare earths)) was also studied. AZ91E, a well known alloy was studied as a reference. Several complementary techniques were employed. Optical micrography and X-ray diffraction (XRD) were used for phase analysis. Slow strain rate test (SSRT) and constant load test (CLT) were used for SCC studies. Smooth and notched specimens were employed; the later were used for fracture mechanics investigation. Circumferential notched tensile (CNT) specimens were used to obtain threshold stress intensity for SCC (KISCC) even for thin plate (12.2 mm). SCC studies were done in 0.01, 0.1 M NaCl solution saturated with Mg(OH)2. EV31A, AZ91E and Mg-Mn to have 50 ± 5 μm, 320 ± 10 μm and 12 ± 3 μm size grains respectively. EV31A contained Mg, Mg12Nd, Mg5Gd, and Zr3Zn2 phases, while AZ91E had Mg, Mg17Al12 and Al8Mn5 phases and Mg-Mn alloy had Mg, (Mg, Mn)12Ce and Mn phases. SSRT and CLT study showed EV31A more resistant to SCC than AZ91E in distilled water as well as chloride solutions saturated with Mg(OH)2. For example, under CLT conditions alloy AZ91E in 0.1 M NaCl solution saturated with Mg(OH)2 shows threshold stress (σSCC) to be 102 MPa (which is 60% of yield stress measured at 0.2% offset strain) However, EV31A did not fracture during 1008 h CLT even when the applied stresses were considerably greater than the yield strength. Susceptibility to SCC and hydrogen evolution increases with chloride contents in the test solutions. SCC crack initiated at locations where the alloys suffered localized attack such as pitting. In EV31A pitting was near the large grain boundary precipitates in SSRT specimens and intergranular cracking along bulky precipitates at grain boundry. In the case of AZ91E, the pits were found to occur even within grain. EV31A alloy was found to be more resistant to pitting than AZ91E, which may be the reason for former exhibiting higher SCC resistance than the later. Threshold stress intensity factor for SCC (KISCC) determined using the plot of stress intensity, KI against time to fail was in agreement to that obtained using the plot of log crack growth rate against applied stress intensity at crack growth rate of 10-10 m s-1 in the test environment. The KISCC in distilled water, and Mg(OH)2 saturated solution of 0.1 M NaCl were determined for EV31A to be 10 and 8 MPa m½ , and for AZ91E it turns out to be 8 and 6 MPa m½, indicating better SCC resistance of EV31A than AZ91E. The slope, m of stage I regime increases and KISCC decreases with the aggressiveness of the environment. AZ91E shows lower stage II crack growth rate than EV31A in both the test environments, which is attributed to crack blunting. However, the same alloy exhibited an increase in stage II crack growth rate in chloride containing environment than in distilled water. Fractographic feature in SCC region of EV31A are mostly transgranular when tested in distilled water and mixed of intergranular and transgranular in 0.1 M NaCl solution saturated with Mg(OH)2 under SSRT condition and CLT conditions using compact tension, C(T) specimens. For alloy AZ91E, the fracture features of SCC region are mostly transgranular in all SSRT, CLT and C(T) tested specimens. SCC region showing parallel facets in TGSCC are the evidence of hydrogen assisted cracking and discontinuous crack growth in all the test alloys and IGSCC involving anodic dissolution near grain boundary precipitates. In the case of Mg-Mn alloy (wrought condition), the specimens having the long transverse alignment were more susceptible to SCC than the longitudinal alignment in each SCC test environment. However, Mg-Mn alloy was considerably less susceptible to SCC in distilled water. Pitting near precipitates containing Ce and crack initiation through pit was consistent with Mg-Mn alloy. KISCC obtained using circumferential notch tensile (CNT) was 10 MPa m1/2 for Mg-Mn wrought alloy in Mg(OH)2 saturated solution of 0.1 M NaCl. Pitting was the main precursor event in the SCC initiation process. Their nucleation and growth are influenced by chlorides and applied stress. However, the main role of pitting was in enhancing hydrogen evolution kinetics rather than to act as stress raisers. The main role of chloride was to damage the passive film (pitting) to create a bare surface, and thereby, enhancing hydrogen evolution kinetics. A phenomological model was proposed to explain the interrelation among chloride-pitting-hydrogen evolution and hydrogen embrittlement. It is believed that enhancing the pitting resistance can improve the resistance of Mg-alloy to hydrogen-embrittling environments. This thesis is divided into six chapters. Chapter 1 is devoted to describing motivation for the present work, and frame work of the thesis. Chapter 2 surveys literature pertaining to the present work and outlines motivations, objectives and scope for the present work. Chapter 3 describes, in detail, the materials used, experimental procedures in specimens preparation, designing experimental tools and test rigs for SCC testing and the techniques used for various characterization. Chapter 4 concerns with the stress corrosion cracking studies of sand cast alloys EV31A and AZ91E using smooth tensile cylindrical specimens, and C(T) specimens. In chapter 5 stress corrosion cracking behaviour of wrought Mg-Mn alloy examined using smooth tensile cylindrical specimens, C(T) specimens, and CNT specimens are presented and discussed. Chapter 6 summarizes overall outcome of the present work and suggests work for future research. Thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy of the Indian Institute of Technology Bombay, India and Monash University, Australia.