On relating quasi-static load threshold K_1scc to K_1c

1 Introduction

In any material, atoms are held together by a cohesive force (attractive or repulsive) to reach a minimum potential energy. As a result, cohesion between the atoms (or atomic bonds) becomes the key to the understanding of fracture. Fracture is defined as a separation of these atoms by an external force such as mechanical or chemical (or thermal). Such external forces affect repulsive/attractive forces between the atoms. Fracture then occurs when an external mechanical force of a certain magnitude is applied against the atomic bonding to result in the separation of a material into two pieces. Chemistry “force” can additionally help the mechanical force to bring the material to failure at a lower applied stress. In the absence of an external mechanical force, chemistry alone does not break these bonds. So, chemistry needs an external mechanical load to separate the atom bonds. The mechanical force can be either static or cyclic. In the present case, we consider the role of static load affecting fracture in chemical and inert environments.

The fracture of materials is preceded by a subcritical crack growth (in vacuum or in a chemical environment) originating from a surface or internal flaws contained in the materials. It is identified with the process of slow crack growth in materials subjected to rising static load. This phenomenon deals with a combined interaction of loading, material and environment that determines the overall crack growth resistance characteristics. In particular, for a given chemical environment, subcritical stress corrosion crack growth (SCCG) results from a stress-assisted chemical interaction between the material and moisture/aqueous environment. Since, subcritical crack growth at K_1scc precedes catastrophic fracture at K_IC (K_ISCC << K_IC) in an inert environment, a failure delay is often observed in components subjected to an applied static load. Subcritical crack growth also leads to a time dependence of the strength loss in a chemical environment, the slower the loading rate, the faster is the SCG. Slower loading allows more time for the environment to interact with the crack tip. The science of fracture mechanics provides a framework for analyzing the effect of subcritical crack growth on structural materials and for predicting their lifetime. In a plot such as crack velocity, da/dt, versus the applied stress intensity factor, K_appl, (da/dt − K_appl), one can assume subcritical process for the whole curve from the threshold K_1scc to K_1c. Figure 1 shows such a plot for 7075-T651 alloy in 3.5 % NaCl solution (Vasudevan and Sadananda 2011).

Figure 1:

Description of SCC data.

A schematic R-curve J_R–Δa (or K_R–Δa) in Figure 2 shows that there is subcritical crack extension from below the blunting line to the point where one defines J_1c. Here, threshold K_1scc has both mechanical and chemical contributions and K_1c is related to only the mechanical forces.

Figure 2:

Schematic diagram showing the J_R–Δa curve for typical ductile alloys.

It is generally “accepted” that there is little correlation between subcritical crack growth threshold (under static K_1scc) in a chemical environment and fracture toughness (K_1c) in an inert environment. This is because the mechanisms that lead to the fracture modes are very different. K_1c is for fracture where the sample is broken into two pieces. Whereas subcritical crack growth is not considered as fracture property like K_1c but may be related as a material crack driving force affecting the crack extension at low growth rates. Thus, they may be indirectly related through elastic properties as modulus. In general, SCCG is considered to be related to a time-dependent phenomenon. It can also be related to stress intensity K as a crack extension phenomenon, as in vacuum or in lab.

The diagram shows that there is a region below J_1c where the crack is growing slowly air where environmental contribution to crack growth is absent or small in static load applications. In this sense, K_1c can be “considered” as a “slow crack growth” behavior where there is finite crack extension below J_1c or K_1c in a R-curve (J–Δa), see schematic in Figure 2.

Figure 3 shows the comparison of the variation of K_1scc with YS and K_1c in a ductile 4340 steel (Brown and Beecham 1965) in NaCl and a brittle soda lime glass water vapor (Wiederhorn 1967).

Figure 3:

As YS increase both K_1c and K_1scc decrease in a ductile alloy and lack of plasticity in an elastic ceramic material gives the opposite trend!

In the 4340 alloy, both K_1scc and K_1c decrease as the YS increases. Relative difference between the two parameters decreases as YS increases. Plotting K_1scc versus K_1c shows the K_1scc increases with K_1c. The arrow indicates the direction of YS decreasing. Crack tip plasticity in 4340 seems to affect the relative chemical contribution to K_1scc. This trend is observed in Ti and Al alloys. On the other hand, lacking plasticity in soda lime glass, the K_1scc decreases with K_1c. These two contrasting examples show the role of crack tip plasticity on the environmental effect to crack initiation threshold K_1scc. Thus, showing that SCCG threshold K_1scc is related to K_1c and that crack tip plasticity is playing the key role in the degree of environmental reaction to affect the material resistance.

This article describes the relationship between the subcritical static threshold parameter K_1scc in NaCl aqueous solution to lab air K_1c under quasi-static load. Questions are as follows:

1. Is it possible to decouple mechanical and chemical forces?

2. Question is whether there is any relation between K_1scc and K_1c?

3. Relation and rational of K_1scc to the electrochemical driving force?

Alloys examples are based on (a) anodic dissolution in 5083 and Al–3Li and (b) HAC as in alloys of steel and titanium. We show that the mechanism for K_1scc in Al-alloys to be anodic dissolution and relate the calculated anodic overpotential η_AD to these critical K_1scc parameters. In order to address the above questions, we use the strain energy interpretation described below.

2 Materials and methods

Materials selected for discussion are (1) Al–3Li alloys (Tosca et al. 1989; Vasudevan and Doherty 1987, 1989; Vasudevan and Sadananda 2015; Vasudevan et al. 1985a,b); (2) 5083 alloy (Crane and Gangloff 2011; Gao and Quesnel 2011; Goswami et al. 2010; Goswami and Holtz 2013; Pao et al. 2011); and (3) commercial steels (Brown 1972; Lee et al. 2009, 2011), Ti-alloy (Brown 1972; Chakrabarti and Chesnutt 1967), and AL-alloys (Brown 1972; Sprowls et al. 1973). In the Al–Li alloys, the Li composition varies from 2.1 to 2.9 Li and is labeled as Al–3Li for convenience for discussion. The alloy data are for 3.5 % NaCl, at room temperature is for Al–3Li alloys and commercial steel, titanium, and aluminum alloys. It is 1 % NaCl for the 5083 alloys. We chose specific data from the literature that pertains to only NaCl environment using fracture mechanics CT and DCB samples to obtain the threshold data. The scatter in the data is a result of sample orientation and sample size used by different researchers.

3 Energy interpretation between K_ISCC and K_IC

This relation is rather through an energy balance between K_ISCC and K_IC. From the graph, K_ISCC = α K_IC where 0 < α < 1 is the slope of the line. Taking squares on both sides and dividing by E, we get the energy term:

( K 1 scc ) 2 E = a 2 ( K 1 c ) 2 E , which can be re-written as:

Noting that G_1scc has two components (from Figure 1): one related to the mechanical energy G_mech,1scc and the other chemical energy, the latter is related to corrosion G_chem,1scc, Equation (1) can then be written as:

Now, dividing Equation (2) by G_IC, we get:

or

Implications of Equation (3):

1. If G_chem,1scc << G_1c, Gchem,1sccG1c→0 and Gmech,1sccG1c≅α2→ the ratio is large and the damage is mostly due to mechanical;

2. If G_chem,1scc ∼ G_1c, Gchem,1sccG1c→1 and Gmech,1sccG1c≅α2→ the ratio is small and the contribution is largely due to chemistry.

This leads to address the two parts of the contribution to damage:

1. Provide an explanation in terms of energy consideration why there is a linear relation between K_1scc and K_1c and suggest an interpretation for the implication (a) above.

2. How to estimate the chemical energy contribution separate from mechanical, implication (b)? For this, we use West–Hall (Hall 2011; West 1973, 1980) approach to calculate the chemical energy G_chem,1scc.

4 Basic equations

The following discussion is a summary of Hall and West analysis (Hall 2011; West 1973, 1980) on the topic of mechano-electrochemical driving force for stress corrosion crack initiation and growth. We have modified and extended his model to apply to crack initiation under aqueous solutions assuming “anodic dissolution” mechanism along the high angle grain boundary or along slip bands.

For the case of static load condition, West (West 1973, 1980) and Hall (Hall 2011) have analyzed the electrochemical driving force affecting the mechanical driving force for a given alloy-environment system assuming (a) “anodic dissolution (AD)” in Al–Li and 5083 alloys and (b) “hydrogen-assisted cracking (HAC)” mechanisms in steels and titanium alloys. We present our analysis based on AD mechanism for an active path along grain boundary or slip band. HAC analysis requires more assumptions and hence not attempted.

Plane-stress fracture toughness for a material is the plastic work of fracture or mechanical energy. In an “inert” environment (like lab air or vacuum), it can be written as:

1. K²_1c/E = 2γ_p = G_1c

West (West 1973, 1980) and Hall (Hall 2011) have given an expression for a chemical crack driving force G_AD for the case of “active path anodic dissolution” process for a quasi-static load as:

1. G_AD = δ_corr. (ZF/2Ω). η_AD; where δ_corr = GB or slip band width, Z = valance, Ω = atomic volume, F=Faraday constant, and η_AD = anodic overpotential = electrochemical driving force.

2. (ZF/2Ω). η_AD = electrochemical energy for a given alloy-environment system.

3. G_AD,STATIC = (K²_1c − K²_1scc)/E = K²_CHEM/E is the chemical energy for static loading. It is noted that while K_1c is mostly related to the mechanical effect and K_1scc has both mechanical and chemical components. This subtraction is an “approximation” since there is some residual plasticity in the term K_1scc. The correction factor “k” for this residual plasticity is calculated by West (1980) and is listed in a table form. This difference is then the effective mechano-electrochemical crack driving force, which is that available to drive cracks in the environment.

Combining Equations(4) and (2), we can write:

1. (K²_1scc)/E = (K²_1c)/E − (1 + k). δ_corr,STATIC. (ZF/2Ω). η_AD,STATIC for static load; the correction factor (k) ∼ 3000 for Al-alloys with YS ∼ 300 MPa. This k is due to the residual plasticity correction for K_1scc. This equation suggests a linear variation of (K²_1c/E) versus (K²_{CHEM, STATIC}/E). It also suggests that electrochemical terms reduce the mechanical driving force. The chemical force is analogous to the mechanical resistance force and is dependent on the (δ).

2. [K²_CHEM/E]/[ δ_corr,STATIC. (ZF/2Ω)] = η_AD,STATIC; this relates K_CHEM to η_AD,STATIC. Here, we have can use δ_corr,STATIC = crack opening displacement COD or GB or slip band width for estimating η_AD,STATIC, dependent on the mechanism of choice.

3. West’s average “k” factor for ductile steels ∼2,000, ductile Ti-alloys∼12,000, and Al-alloys∼3000 of his data (West 1973, 1980).

4. We use these equations and the corrections to calculate η_AD for the Al-alloys. This can be done for any other alloys.

5 Anodic dissolution alloys

Figure 4a shows the variation in YS and K1c with aging heat treatment time at 200 °C. In inert (lab air) environment, matrix δ′ affects YS over the entire aging curve. Dislocations cut the coherent δ′ below PA (planar slip region) and the YS increases from AQ to PA. Past the PA, in the OA region, YS decreases due to dislocations bypass incoherent δ, Figure 4a GB-δ ppts strongly affect the static fracture toughness and K_Ic below PA (UA region), while in the OA region, K_Ic is independent of heat treatment time.

Figure 4:

Yield stress and fracture toughness variation with heat treatment time for (a) Al–2.9Li alloy and for (b) 5083 alloy.

This is due to weak δ/matrix interface resulting in GB void nucleation/growth. Static fractures in lab air show dimple fracture surface in AQ condition, slip line fracture with IG voids at intermediate aging times, and complete IG with voids in the OA region.

GB-β ppts in 5083 alloy do not affect YS as well as K_1c, in lab air. Mg in solid solution affects the YS and is constant with sensitizing time since the amount of Mg at the 175 °C temperature is constant, Figure 4b.

Stress corrosion threshold is analyzed in 5083 and binary Al–3Li, both having anodic grain boundary precipitates. 5083 commercial alloy forms grain boundary Al₃Mg₂ (β-phase) while Al–3Li alloys have AlLi (δ-phase). GB-δ is discretely spaced and not continuous as in long time sensitized 5083 alloy. Area fraction of the GB-phases is measured using SEM (Vasudevan and Doherty 1989) and TEM (Goswami and Holtz 2013). These phases grow in size and area fraction with heat treatment time, and when exposed to the NaCl solution, they anodically dissolve and form grain boundary cracks under an applied stress. The crack initiation occurs when the appropriate local environmental conditions at the crack tip and stress state are established. The overall mechanism seems to be mostly anodic dissolution coupled with a minor contribution from hydrogen assisted cracking.

These two alloys have been discussed in their experimental detail (Vasudevan and Sadananda 2015), mechanical properties, and electrochemical description of the GB-phases reacting with NaCl solution. We present a brief discussion on the salient points.

5.1 Al–3Li alloys

In the case of Al–3Li alloy, GB-δ is incoherent oblate spheroid that grows in size and number (A_f) with aging time at 200 °C, Figure 5. In the early stage of aging where δ is very small in size, the deformation is along the planar slip giving a transgranular fracture.

Figure 5:

K_1c and K_1scc variation with A_f of GB-δ precipitates in Al–2.9Li alloys.

As δ grows in size and A_f with aging time, fracture increasingly becomes mixed mode (TG + IG) from severely UA to PA. This mix mode fracture comes partly from planer slip decohesion and GB-δ. With longer aging time past the PA condition, the fracture is totally along the GB with void nucleation and growth process at the interface of δ. Fracture surfaces at low and higher aging times are shown in Figure 6. In the UA condition, fracture is mixed with some slip and IG fracture, while in the OA case, it is completely IG fracture with micro voids and coalescence. The schematic crack tip illustration in Figure 6 gives the UA and PA fracture with the corresponding fracture surfaces showing mix mode in the UA and IG voids in PA and OA conditions.

Figure 6:

Fracture process in Al–3 Li alloys with GB δ ppts under inert lab air environment. In the UA, fracture surface is mixed TG + IG and in PA (and OA) it IG.

5.2 5083 alloy

Figure 7 shows the K_1scc decreasing with A_f of GB-β in 5083 alloy (Vasudevan and Sadananda 2015). There is an upper shelf plateau (here A_f of β is very low) and a lower shelf corresponds to when A_f has saturated. In between these limits K_Isccα (1/A_f). Even in 1 % NaCl solution at A_f < 0.03, the fracture is transgranular as in lab air. At A_f > 0.07, fracture is completely intergranular. The SEM fracture surfaces are shown in the insets on Figure 7. In between A_f ∼ 0.03 to 0.07, fracture starts off in mixed mode transitioning to intergranular mode with increasing A_f. This trend is related to the β dissolution in NaCl. At low A_f < 0.03, where the spacing between β is very large, the alloy is not sensitive to NaCl environment and K_Iscc remains same as K_Ic. When K_Ic ∼ K_Iscc at very low A_f of β transgranular mechanical fracture with dimples are observed due to applied stress. See inset on Figure 7 at K_Iscc ∼ 20  MPa m^1/2, showing IG with edges tearing. When A_f > 0.7, the fracture is completely IG and is due to just active path β dissolution and K_Iscc reaches a minimum value around 4 MPa m^1/2. Insets show the mode of fracture from TG to completely IG. Interestingly, as a comparison, 5083-NaCl K_Iscc ∼ 3 MPa m^1/2 data for heat treatment times longer than 300 h where A_f > 0.7. K_Iscc values at longer aging times are still higher than Griffith’s fracture values K (Griffith) ∼ 0.22 MPa m^1/2 for AL. Factors that can affect this difference could be oxide delaying the reaction at the crack tip and plasticity suppressing complete brittle fracture. Since β can dissolve without the need of an applied stress, then applied stress can just help in keeping the crack tip open to allow for the access of NaCl electrolyte to the crack tip to allow galvanic dissolution of β until its solubility limit is reached, a process similar to LME.

Figure 7:

K_Ic and K_Iscc dependence on A_f of GB-β LME. Fractography at different A_f are shown. 7075-T651 alloy: K_Iscc (Hg) ∼ 1.5 MPa m, K_Iscc (Ga) ∼ 1.7 MPa m.

5.3 Comparing Al–3Li versus 5083 alloy

The following section brings out the key differences between Al–3Li and 5083 alloy with respect to their anodic GB-precipitates showing the effect of varying background plasticity and their variations with calculated anodic overpotential η_AD. η_AD can be viewed as the chemical driving force similar to the mechanical force G. While the alloy compositions are different with differing chemistry of the GB-precipitates, one can observe the variation in the degree of dissolution rates with respect to their background mechanical properties of YS and K_1c.

Figure 8 shows the contrasting differences between Al–3Li and 5083 alloys in their G_CHEM and η_AD parameters varying with A_f % GB-phases. The G_CHEM trend in both alloys track η_AD as a variation with A_f % GB-phases. In the Al–3Li alloy, it is a decreasing trend, while in the 5083 alloy, it is an increasing trend. Scales are different. In the Al–3Li alloy with A_f % > 0.2, plateau is due to GB-δ playing a role in K_1c and exposing GB-δ to NaCl to dissolve. As the A_f % GB-δ increases, the dissolution rate is decreased by the plasticity used to create voids at the δ/Al interface. In the 5083 alloy, the GB-β does not play contribute to K_1c and the dissolution rate increases with A_f % of β. The higher plateau in 5083 is due to the saturation on continuous GB-β. Interestingly, the calculated anodic η_AD values differ by an order of magnitude between the two alloys. This is probably related to the differences in the chemistry of the two GB-phases and the background plasticity affecting the rate of dissolution.

Figure 8:

Variation in G_CHEM and η_AD with A_f % of GB-precipitates in (a) Al–3Li alloy and (b) 5083 alloy.

Table 1 gives a summary of the comparison of fracture process between Al–3Li and 5083 alloys.

Table 1:

Possible mechanisms (G_slip < or > G_gb).

AL–3Li Alloys (GB-ppt δ)	5083 Alloy (GB-ppt β)
GB-ppts (δ) is anodic wrt matrix, they dissolve in NaCl solution (under no applied stress), & this reaction generates H Crack can be extend if Gslip > or < GGb crackinG; if Gslip < GGb slip is favored & if Gslip > Ggb, then GB decohesion is favored by reducing γgb Discrete GB-ppt dissolution do contribute to crack extension; dissolution rate ∼ H-diffusion rate, slip is favored to create microvoids around GB-ppts, and the cracktip comes in contact with GB-ppt to dissolve in NaCI Dislocations on the slip plane, creates microvoids around the GB-ppts before dissolution of GB-ppts starts leaving trace of voids	GB-ppt (β) is anodic wrt matrix & dissolves in NaCl solution (under not applied stress) generating H Dissolution rate of β ∼ H-diffusion rate; crack can extend if Gslip < Ggb cracking When β is spaced apart at low Af, slip is favored than β-dissolution, & the applied stress tears the space mechanically in between the ppts When β is continuous at high Af, dissolution in NaCl takes over suppressing slip. H effect is secondary

6 Other alloy systems

The trend of K_1scc threshold in NaCl solution increasing with lab air fracture toughness K_1c is observed in other alloys. The linear trend between K_1scc and K_1c. We did not go through the overpotential calculations as steels and titanium alloys are prone to HAC and AD depending on the composition in 3.5 % NaCl. We did classify the trends in terms of the susceptibility to HAC, and some compositions do not show any effect in NaCl solution. The collected data from the literature is shown in Figure 9a–c. All three alloy systems show a linear behavior representing K_1scc = α K_1c, where the slope a related to some mechanism.

Figure 9:

Variation of K_1scc with K_1c in NaCl for commercial (a) steels (Nibur et al. 2010), (b) titanium alloys, and (c) aluminum alloys.

Figure 9a represents the K_1scc variation with K_1c for steels. There are two α-slopes, one with α = 1.3 close to α = 1 and the other slope α = 0.17. The first high slope shows that these alloys are not affected by NaCl while the lower slope of α = 0.17 is probably due to HAC. We have included Nibur et al. (1972) pressurized hydrogen data on a few steels fall on the NaCl data. In addition, Aermet-100 internal H-data (Thomas et al. 2003) is also shown. These results fall, within the scatter, on the low slope (= 0.17) data, suggesting that the low slope is related to HAC. The slope = 1 may be due to (1) viscous oxide consuming the water content and reducing the Cl⁻ concentration, (2) increase in K_1c is giving crack blunting, and (3) as the slope approaches unity, at high K_1c the steel is immune to NaCl damage.

Figure 9b shows the result of titanium alloys, similar to that of steels with two slopes. The higher slope α = 0.89 is unaffected by NaCl and K_1scc varied directly with K_1c. The second slope with α = 0.25 probably represents HAC. Figure 9c gives a summary of results from Al-alloys on a log-log scale. Here, the trend in the results shows two slopes. The resistant alloys have low slope (∼0.1) and the susceptible ones have higher slope (∼0.25). A few data points from LME for the 7075 alloy in Hg and in Ga show that these fall in the range of NaCl. The higher slope can be due to the different rate of dissolution between the Al-alloys.

7 Comments on HAC

Most of the steels and titanium alloys have been prone to HAC in aqueous solutions and in H-atmospheres. Nelson (1972) performed a unique experiment on many alloys to observe which alloy exhibit failure by HAC. Figure 10 gives the summary of his results. His experiment consisted of placing a heated filament at the sharp notch tip of a three-point bend fracture toughness sample in the presence of molecular hydrogen. The evacuated vacuum chamber was first filled with molecular H that was converted to nascent H by locally heating with a W-filament close to the notch tip of the three-point bend samples to measure the K-values. At the room temperature, at a very low displacement rate of 1.48 × 10⁻⁹ m/s, slow crack growth toughness or threshold K_SCG was measured in vacuum at 5 × 10⁻⁷torr (∼5 × 10⁻⁴Pa), of molecular H₂ at 90.6 N/m², and of atomic-molecular H-mixtures at a pressure of 1.07 N/m². The K-values were plotted for the two H-atmospheres selected versus the vacuum results (K_1c). Slopes are indicative of the degree of H-effect on fracture. It is observed that alloys of Al, Cu, and Mg do not show gaseous HAC as the results fall on slope = 1. Others fall on slope 1/3 and 2/3.

Figure 10:

Plot of K_SCG in H versus K_1c in vacuum for many alloys. Slopes indicate the magnitude of HAC on fracture.

The second example is on a 7075-aluminum alloy heat treated to give various YS from severely UA to PA condition, shown in Figure 11. 7075-T651 data are shown for comparison. Here, the comparison is made between the samples tested in 3.5%NaCl (Speidel 1975) and in LME Hg (Wheeler et al. 1989). The results show that the two environments give the same result suggesting that the role of HAC in 7075 alloy in NaCl need not be related to HAC but may be related to variation in the degree of anodic dissolution process, since LME is related only to dissolution. Lynch (1988) has commented on such comparison between LME and HAC on several alloys in static and cyclic failures.

Figure 11:

Variation of K_1scc with YS of 7075 aluminum alloy heat treated to give various YS. For reference peak, aged T651 data are indicated.

8 Implications

The discussion in this article indicates that static corrosion threshold K_1scc is related to lab air fracture toughness K_1c, in several commercial alloys. The binary alloys of Al–3Li and 5083 were selected to indicate that plasticity plays a key role in SCC, particularly for a given mechanism like anodic dissolution. What is not clearly understood is the local crack tip mechanism of chemical reaction in the presence of plasticity (YS, K_1c). There are some indirect experimental observations on this relationship of how plasticity affects K_1scc. This is shown in Figure 12 for a 4,340 commercial steel with YS = 1530 MPa, K_1c = 53  MPam^1/2, K_1scc = 9.3 MPam^1/2. By varying notch radii, one can modify the plasticity (Hirose and Mura 1984; Pao unpublished data; Vasudevan 2013) and measure the fracture toughness (apparent) in vacuum and in 0.1 % H₂SO₄ solution. The apparent fracture toughness properties vary similarly for the two environments with different magnitude. As the notch gets to be blunter at higher radii, the K values tend to saturate, suggesting the reaction rate getting slower as the stress gradient becomes lower. This result is replotted as K_1scc variation with K_1c, giving a linear behavior with a large deviation from the slope = 1. This graph gives the linear relationship between K_1scc and K_1c as in the previous slides.

Figure 12:

Comparison of notch K_1scc and K_1c for 4340 steel (a) notch radii variation and (b) K_1scc versus K_1c.

For structural applications in marine environments, both a high K_1scc and a high K_1c are beneficial. These parameters can be measured and assessed using the existing methods. High K_1c is also useful during load excursion like overloads and underloads during service environments. The analysis presented here can aid in the life prediction modeling in which chemical contribution can be added to the mechanical component.

9 Key points

The following are some salient points of the SCC discussion in this article at room temperature, NaCl environments:

General observations:

1. K_1scc is linearly related to K_1c with a proportionality constant “α.” When α → 1, crack extension is mostly “mechanical,” and when α < 1, it is mostly due to chemistry. There seem to be two slopes in the K_1scc–K_1c behavior where “α” falls between 1 and 0.

2. K_CHEM = (K²_1c/E) − (K²_1scc)/E) = (G_1c − G_1scc) is also directly related to electrochemical driving force “η”; chemistry helps to reduce K_1c. Mechanism could be AD or HAC or both.

3. This analysis can help in inserting an electrochemical corrosion parameter [δ. (ZF/2W)] into a Life Prediction Model.

Alloy observations:

Binary AL–3Li planar slip alloy:

1. Stress corrosion fracture initiation K_iscc is mostly due to anodic dissolution of δ with a weak dependence on HAC. In the AQ condition, fracture surface is masked by Li-oxide showing strong reaction with NaCl. At very low A_f, the fracture mostly consists of some slip fracture and IG, and at high A_f, the SCC fracture is completely IG and the fracture surface seems to maintain the dimples without δ, which suggests that deformation preceded before δ dissolution. NaCl has large effect on K_Iscc when the A_f is very and this environmental effect decreases as A_f is increasing with aging time. These alloys being chemically very active is prone to pitting.

Commercial 5083 solid solution (serrated yielding) alloy:

1. In NaCl environment, there are three regions of SCC behavior. Environmental effect on K_1scc increases, with increasing A_f, to reach a minimum plateau value of about 3–4 MPa m^1/2, related to the dissolution of β.

2. At very low A_f of β, near unsensitized condition, SCC is negligible since K_Ic = K_Iscc. This also suggests that HAC is negligible and the failure is mechanical with minimal effect from NaCl.

3. At low sensitization times from about 5 h to 300 h, where β is discontinuously precipitated, failure is mixed with mechanical tearing (in the β free regions) and dissolution of β as observed on the fracture surface. In this region, K_Isccα (1/A_f).

4. At longer heat treatments times greater than 300 h, the failure is mostly due to anodic dissolution of GB-β and K_Iscc is independent of A_f, A_f of GB ppts δ & β show an inverse relationship with K_Iscc = C/A_f.

   1. The large differences in the background fracture behavior in lab air & dissolution (in NaCl) kinetics between the two alloys is affected by the interface strength of δ & β and their electrochemical behavior.

   2. The overall behavior analyzed in the binary alloy system is also observed in commercial steels, titanium alloys, and aluminum alloys.

   3. The rate of dissolution is affected by the background plasticity in the alloy and the reason for this has not been understood.