Influence of Imposed Strain Rate on Stress Corrosion Damage Mechanisms in Synthetic Seawater of Austenitic–Ferritic Stainless Steel X6CrNiMoCu25-6

Abstract

Austenitic–ferritic stainless steels are frequently selected for components in seawater-cooling pumps in power plants and nuclear stations due to their generally superior resistance to localized and stress corrosion cracking (SCC) in comparison with austenitic grades. However, during operation, these cast components are exposed to combined applied/residual stresses and chloride-rich seawater. This work addresses the research gap by investigating the SCC susceptibility of X6CrNiMoCu25-6 in the hyper-quenched state (45% ferrite–55% austenite) using slow strain-rate tensile (SSRT) tests in synthetic seawater at 70 °C at five imposed strain rates ranging from 2.38 × 10⁻⁷ s⁻¹ to 10⁻⁴ s⁻¹. The work is complemented by macrographic examination of crack initiation and SEM microfractographic analysis of the fracture surfaces. The findings indicate that the grade demonstrates only negligible susceptibility to SCC with respect to permissible stress, exhibiting a maximum reduction in tensile strength that is constrained to 5% relative to a neutral reference medium. However, a distinct critical strain rate of approximately 10⁻⁶ s⁻¹ has been identified. Further fractographic and microstructural analysis demonstrates that the susceptibility at this critical rate is governed by a localized mechanism. This mechanism is characterized by repeated pitting nucleation at slip-step emergence sites in the ferrite, selective ferrite dissolution, and progressive α/γ interfacial decohesion. These processes result in mixed intergranular/transgranular crack propagation, rather than by bulk mechanical softening. These findings address a particular lacuna in the duplex-steel SCC literature by providing a quantitative strain-rate/damage criterion, in conjunction with elongation-based rather than stress-based susceptibility indicators, for the assessment and management of the SCC risk of this cast duplex grade in real seawater-cooling pump service. From a fundamental standpoint, these findings indicate that the susceptibility of SCC in this duplex grade is governed by a competition between plastic-deformation kinetics and electrochemical dissolution kinetics. This competition reaches a maximum at an intermediate critical strain rate, thereby providing further experimental support for a general strain-rate/dissolution-competition model of SCC applicable to duplex steel.

1. Introduction

Stress corrosion cracking (SCC) of stainless steels has been shown to be associated with transgranular cracking of austenitic grades in chloride-containing environments under the action of applied or residual stresses [1,2]. This phenomenon has been comprehensively documented in numerous studies, including but not limited to: the impact of surface preparation on pitting resistance, residual stress and SCC in austenitic stainless steels [3,4,5], the influence of strain on the corrosion resistance of welded 304 stainless-steel joints [6], and the SCC behavior of sensitized AISI 304 in aqueous chloride solutions containing sulfur species [7,8]. It has been demonstrated that other classes of stainless steel, namely martensitic, ferritic and austenitic–ferritic (duplex), are also susceptible to SCC. This phenomenon is exemplified by the pitting corrosion of duplex stainless steels [9,10]. However, duplex austenitic–ferritic steels have been shown to outperform fully austenitic grades when exposed to chloride media. Consequently, manufacturers of circulating-pump components operating in aggressive environments, such as those with seawater, have increasingly opted for duplex steels. In service, such components are subjected to stresses of various origins [11]. The two primary categories of stress under consideration are as follows: The two categories of stress under consideration are as follows: (i) Residual stresses inherent to manufacturing processes such as casting and homogenization heat treatment, or introduced during repair operations involving reloading and grinding [11,12]. (ii) Service-related stresses associated with pump operation in circulating or cooling environments [13,14].

The duplex structure of these steels can be either fully homogenized or locally affected by thermal cycles, whether from short, high-temperature welding operations or from long-duration, lower-temperature aging, as encountered in nuclear circuits. It has been established that both routes can engender rapid local changes in chemical composition, particularly in the chromium and molybdenum content of the ferrite and austenite phases. This phenomenon has been reported in the context of welded, nitrogen-alloyed duplex steels [14] and for the phase transformations occurring under continuous cooling versus isothermal heat treatment [15,16]. Such alterations have the capacity to exert a substantial influence on the SCC resistance of the grade under assessment. It is evident that even slight local compositional changes resulting from manufacturing thermal effects, hyper-quench annealing, or in-service isothermal aging have the potential to render duplex grades susceptible to various forms of localized corrosion. This assertion is supported by the findings of research conducted on tungsten-containing duplex steels aged at 475 °C [17,18,19] and on the effect of thermal aging on the surface reactivity of duplex stainless steels [20,21,22,23]. Moreover, our preceding studies had previously demonstrated a correlation between the mechanisms of ferrite decomposition and the corrosion resistance of aged austenitic–ferritic stainless steels [24,25,26].

The same considerations apply to deformation substructures, which arise from monotonic or cyclic loading and from their interaction with the environment. The resulting slip steps, striations and persistent slip bands have the capacity to locally eradicate the passive protective film. This hypothesis is supported by the results of studies on the effect of imposed strain on the corrosion resistance of stainless-steel welds [6,27,28] and on the SCC of advanced high-strength steels in marine environments [29,30,31,32]. The present studies demonstrated that the initiation of localized corrosion, manifesting as pitting, crevice attack or cracking, can be induced by slip steps, striations and persistent slip bands.

The investigation of a material’s susceptibility to stress corrosion can be facilitated through the utilization of either slow tensile tests or tests conducted at constant (imposed) strain rates. The latter method is referred to as the slow strain-rate tensile (SSRT) technique. The principal benefit of this approach is that it facilitates the evaluation of a material’s SCC resistance within a specified environment over an adequate testing duration, without the necessity for artificial exacerbation of the electrochemical conditions. As demonstrated in earlier studies, in a multitude of metal-environment systems, the imposed strain rate frequently functions as a more discerning parameter than stress alone [16,33]. Cracking has been observed to occur preferentially at strain rates ranging from 10⁻⁵ to 10⁻⁶ s⁻¹ in certain systems, a sensitivity that other test methods fail to reveal. In the specific context of duplex stainless steels, studies employing SSRT on the wrought 2205 grade in a 26% NaCl solution at 90 °C under controlled electrochemical potential have identified environmentally assisted cracking with a mixed intergranular/transgranular character and a strain-rate dependence of susceptibility [34,35]. Research conducted on austenitic AISI 316 stainless steel in concentrated NaCl solution at elevated temperatures has similarly reported a critical strain rate (of the order of 10⁻⁶ s⁻¹) at which susceptibility to SCC, expressed as a loss of tensile strength, is maximal (this reference is being verified by the authors for the final citation number). The present body of work establishes the strain rate as a pivotal controlling parameter of SCC in both austenitic and duplex stainless steels. However, it should be noted that this work has been obtained almost exclusively on wrought, welded or aged material under concentrated-chloride, controlled-potential conditions.

Notwithstanding the considerable body of extant literature on the subject, the majority of existing studies on SSRT in the field of duplex stainless steels have concentrated on wrought, aged, or welded grades. Examples include the extensively investigated 2205 duplex steel, which has been tested in concentrated chloride solutions (26% NaCl at 90 °C) under imposed, controlled electrochemical potential [35,36,37]. While these conditions are informative for mechanistic understanding, they are considerably more aggressive in terms of both chloride content and electrochemical driving force than the free-corrosion, moderate-chloride environment actually experienced by cast duplex components in seawater-cooling service. In particular, the cast, hyper-quenched, high-Mo/Cu grade X6CrNiMoCu25-6, used specifically for seawater-cooling pump castings in power plants and nuclear stations, has not previously been characterized in terms of the strain-rate dependence of its SCC susceptibility. Furthermore, the critical strain rate of this material under realistic, freely corroding synthetic seawater conditions at a service-representative temperature (70 °C) has remained unidentified.

The scientific objective is twofold. Firstly, the aim is to identify the strain-rate dependence of the stress corrosion cracking (SCC) susceptibility of the cast, hyper-quenched duplex stainless steel X6CrNiMoCu25-6 in synthetic seawater at 70 °C. Secondly, the underlying microstructural mechanism of crack initiation and propagation is to be elucidated. The latter comprises the following factors: slip-step emergence, localized passive-film rupture, selective ferrite dissolution and α/γ interfacial decohesion. The research utilizes a combination of slow strain-rate tensile (SSRT) testing at five imposed strain rates and macrographic and SEM microfractographic examination in order to achieve this objective. The overarching objective of this study is to translate these findings into a practical criterion for pump designers and operators. This objective will be accomplished by identifying a critical strain-rate/stress relationship that determines the imposed-strain-rate range for which SCC susceptibility of this grade is maximal. Furthermore, the permissible stress and ductility limits for this range are to be determined. The ultimate objective is to provide a framework for the selection of materials and assessment of the suitability for the service of components for seawater-cooling pumps manufactured from X6CrNiMoCu25-6.

2. Materials and Methods

The study of SCC was carried out using cast austenitic–ferritic stainless steel of the X6CrNiMoCu25-6 grade. This material casting used in this study was supplied by the Rades thermal power plant, located in Tunis, Tunisia. It is important to note that the test specimens were fabricated from a duplex stainless-steel casting for seawater-cooling pump components. The chemical composition (in percentage by weight) was determined by optical emission spectrometry (OES), (SOFOMECA, Ben Arous, Tunis, Tunisia) on the as-received casting, following standard practice for stainless steel grade certification, and mechanical properties (in terms of tensile strength, impact strength and hardness) are given in

Table 1. Chemical composition of X6CrNiMoCu25-6 stainless steel (% wt).

C	Cr	Ni	Mo	Cu	Mn	N	Si	S	P	Co	V	Fe
0.068	24.56	6.69	2.52	2.32	0.59	0.137	0.7	0.0008	0.024	0.109	0.11	Bal.

and

Table 2. Mechanical characteristics of X6CrNiMoCu25-6 stainless steel in the hyper-quenched state.

Tensile Properties				KCV J/cm2 at 20 °C	Microhardness		Hardness HV2
Re (MPa)	Rm (MPa)	At (%)	Z (%)		Austenite	Ferrite
293	790	31	61	192	260	325	280

.

The samples used in this study underwent hyper-quenching in air after being held at 1170 °C for two hours. Optical microscopic examination of polished and electrolytically etched samples in an acid solution revealed a two-phase structure consisting of ferrite and austenite in the following proportions: 45% ferrite and 55% austenite. These proportions were determined by image analysis using optical metallography. The austenitic phase is present as white platelets in a dark ferritic matrix (Figure 1a). High-magnification SEM analysis at the α/γ interfaces highlights a fine precipitation of mixed titanium and niobium carbides, known as ‘primary’, which were not put back into solution after the hyper-quenching treatment (Figure 1b).

Tensile tests were conducted at imposed strain rates in the conventional C.S.C. test medium, synthetic seawater heated to 70 °C, on standard TC6-type test specimens (Figure 2a). The M.T.S.-type universal tensile testing machine was utilized for this purpose, and was equipped with a data acquisition system connected to a microcomputer for data acquisition and processing (Figure 2b). The temperature of the solution is regulated by means of a contact thermometer that operates on a binary basis, i.e., either activated or deactivated. The thermometer is immersed in a heating tank, the contents of which are continuously supplied to the corrosion cell by a diaphragm pump. In the course of conducting the experimental protocol, the application of heat is initiated prior to the implementation of loading, and the potential of the test specimen is monitored over time using a recorder. This ensures the stability of the passive film in the given environment. Once the potential has been stabilized (i.e., no significant variation in the open-circuit potential has been observed), the specimen is loaded at a constant strain rate according to a well-defined programmer.

The test program involved monitoring the evolution of stress as a function of time at the following imposed strain rates: 2.38 × 10⁻⁷ s⁻¹, 5 × 10⁻⁷ s⁻¹, 2.38 × 10⁻⁶ s⁻¹, 2.38 × 10⁻⁵ s⁻¹ and 10⁻⁴ s⁻¹. These speeds are consistent with those commonly adopted for slow tensile stress cracking studies [34]. Slow tensile tests in a neutral silicone medium were also conducted under the same speed and temperature conditions to isolate the impact of the low-chloride medium (synthetic seawater). Susceptibility to SCC was verified by examining the specimen barrels under a microscope, and the crack initiation and propagation mechanisms were identified by examining the fracture surfaces under a scanning electron microscope (SEM). SEM microfractographic examinations were carried out on the fracture surfaces of TC6 specimens after the various tensile tests.

Figure 3 shows the flowchart that describes the research approach adopted in this study: starting material (cast, hyper-quenched X6CrNiMoCu25-6) → specimen preparation (TC6 specimens) → test matrix (5 imposed strain rates, in synthetic seawater and neutral silicone at 70 °C) → measured outputs (mechanical characterization, macrographic examination, SEM microfractography) → outcome (critical strain rate and SCC mechanism).

3. Results

3.1. Susceptibility to Cracking by SCC

3.1.1. Mechanical Characterization

The stress–strain records relating to the various slow tensile tests on specimens in the hyper-hardened state in the two-study media highlight the variation in total permissible strain with the imposed strain rate (Figure 4). The mechanical properties derived from conventional tensile curves at various strain rates in synthetic seawater at 70 °C are presented in

Table 3. Effect of strain rate on tensile mechanical properties in synthetic seawater at 70 °C.

Strain Rate s−1	2.38 × 10−7	5 × 10−7	2.38 × 10−6	2.38 × 10−5	10−4
σn (MPa)	733	730	725	740	745
Ar (%)	20	24.7	15.8	22.2	26
At (%)	26	31.5	21.8	29.5	34
εr (%)	17	20	14.5	19	21.7
Z (%)	19	20	17	21	26
Tr (h)	300	169	25.5	3.5	2
W (106 J/m3)	119	142	94.6	136	160
Results of the macrographic analysis of the test specimen barrel	Slight cracking by SCC	Slight cracking by SCC	Cracking by SCC	Slight cracking by SCC	No cracking by SCC

. These results demonstrate that the change in permissible stress, or stress at fracture, with strain rate indicates low susceptibility to SCC for the studied grade in synthetic seawater at 70 °C. In fact, the reduction in tensile strength in synthetic seawater remains relatively small, reaching a maximum of 5% at a strain rate of approximately 10⁻⁶ s⁻¹. This phenomenon can be attributed to the observation that the mechanical response in the inert medium is essentially strain-rate independent over the range of tested parameters. In contrast, the permissible stress σ_n varies only mildly and non-systematically with strain rate, thus confirming the abstract’s conclusion that stress-based criteria are poorly discriminating for SCC detection in this grade, while ductility- and energy-based criteria (At%, Z%, W) are far more sensitive. This indicates that there is a speed or range of speeds at which the hyper-quenched state is most susceptible to SCC.

Superimposing the slow tensile curves at maximum susceptibility speed for corrosive synthetic seawater and neutral silicone at 70 °C shows that tensile elongation at break and allowable strain are better indicators of susceptibility to stress corrosion cracking (SCC) in austenitic–ferritic stainless-steel grades (Figure 5).

3.1.2. Microstructural Characterization

Material degradation by stress corrosion is well established to be the result of cracks initiating at the surface and propagating through the sub-layers under the combined effects of stress and the environment. This type of damage is caused by several elementary mechanisms that are linked to the material’s structure, the medium, the surface’s electrochemical reactivity, the levels of residual and applied stresses, and the strain rate. These mechanisms can be identified by analyzing the surfaces of specimens tested in SCC using micrographs (secondary cracks) and by studying the fracture surfaces (main cracks) using SEM (scanning electron microscopy). The rupture time (Tr) decreases monotonically and steeply with increasing strain rate (from 300 h at 2.38 × 10⁻⁷ s⁻¹ to 2 h at 10⁻⁴ s⁻¹), as would be expected kinematically. However, the fact that the SCC fracture fraction does not decrease monotonically alongside Tr suggests that exposure duration alone does not control the degree of SCC damage. Despite the exposure duration of 300 h being achieved at the lowest strain rate (2.38 × 10⁻⁷ s⁻¹), the SCC fraction at this point (47%) was found to be lower than at 2.38 × 10⁻⁶ s⁻¹ (60%, Tr = 25.5 h only). This finding confirms that the strain rate, as opposed to the cumulative duration in a corrosive environment, is the parameter that governs SCC severity. Permissible deformation decreases by 30% in synthetic seawater at 70 °C compared to the neutral silicone medium at the maximum susceptibility rate. Thus, based on this criterion (Ar%) or total elongation (At%), the rate of maximum SCC susceptibility of the studied grades in the two considered media is more clearly highlighted (Figure 6). This result is fairly consistent with the curve showing the evolution of Z% stress as a function of strain rate, which reveals a minimum stress associated with the lowest ductility at the same speeds (10⁻⁶ s⁻¹).

Metallographic analysis of the deformation substructures corresponding to various deformation rates imposed using SEM identifies the elementary deformation mechanisms operating under slow tensile conditions in each of the phases.

(a) Surfaces examinations:

Examination of the test specimen barrels using low-magnification SEM on either side of the maximum susceptibility speed (approximately 10⁻⁶ s⁻¹) revealed the onset of cracking at a speed of 2.38 × 10⁻⁷ s⁻¹ in synthetic seawater and no cracking at other speeds (Figure 7a,b). However, close to the critical cracking speed, a myriad of secondary cracks of varying sizes were revealed in the drums, aligned perpendicular to the loading axis and characteristic of SCC damage. Shafts tested by slow tension at the maximum susceptibility speed (2.38 × 10⁻⁶ s⁻¹) in neutral silicone at 70 °C did not exhibit any significant cracking in the stricture zone (Figure 8a,b).

High-magnification scanning electron microscopy (SEM) examination of the shafts of specimens fractured by slow tensile stress reveals the speed of maximum susceptibility to S.C.C.

-

The presence of α/γ interfacial decohesion is evident in the proximity of both the primary fracture and its immediate vicinity (Figure 9a,b). However, as the fracture is approached, the severity of this decohesion increases. Decohesions at the austenite-ferrite interfaces in synthetic seawater heated to 70 °C (Figure 9a). These decohesions are the precursors of the final stress corrosion cracking (SCC) fracture of duplex steel (Figure 9b).

-

There is also a mainly intergranular failure by slow traction, which appears to be accelerated by the selective dissolution of ferrite, particularly in zones close to the α/γ interfaces. As demonstrated in Figure 10, ferrite dissolution is essentially complete in the immediate vicinity of the fracture, while austenite exhibits only slight pitting. This confirms the phase-selectivity of the attack. This is related to traces of plastic deformation (sliding step). This dissolution weakens as one move away from the main fracture zone where the plastic deformation of the ferrite is less significant (Figure 11).

The predominant elementary mechanism of ferrite dissolution appears to result from interactions between plastic deformation by slow traction and localized corrosion involving repeated sprouting and pitting growth (Figure 12a,b). Pitting initiation was observed at the emergence of slip tracks within ferritic grains (Figure 12c) and at their termination at α/γ interfaces (Figure 12). Dissolution appears to be supported by multiple corrosion pitting initiations on slip steps, as well as their growth and coalescence.

As demonstrated in Figure 13, even in circumstances where two slip systems are concurrently activated in the austenite 2 mm from the fracture, the austenite itself remains predominantly free from localized attack. This is in marked contrast to the adjacent, clearly delineated, and evidently eroded α/γ interfaces. This finding serves to further substantiate the conclusion that the austenite performs an essentially mechanical role, namely a load-bearing and ductile one, whilst the ferrite and the α/γ interface are the electrochemically active sites of damage. Multiple pitting initiations on the slip plane emergence in ferrite, and their growth, lead to degradation by ferrite dissolution. Despite being deformed at a rate that maximizes susceptibility to SCC, the austenite remains intact apart from a few pits, which are probably associated with initial inclusions and appear to be independent of sliding traces (Figure 14).

At speeds below or above the speed at which maximum susceptibility to SCC occurs, the two phases (α and γ) do not reactivate and show no signs of localized attack (pitting, etc.). The longitudinal-section observations (Figure 15) demonstrate that this surface-nucleated, distance-graded damage pattern manifests as a tangible sub-surface crack network: cracks emerge at the surface and branch as they propagate inward. This explains the near-total absence of SCC cracks on the shafts and why the tensile fracture occurs at the level of the widely extended stress zone.

(b) Examination of fractured specimen sections:

Optical microscope examination of a longitudinal section of the specimen stressed at a strain rate of 2.38 × 10⁻⁶ s⁻¹, corresponding to the maximum susceptibility rate with respect to SCC, confirms the initiation of cracks from the surface and their branching as they propagate inside the specimen barrel (Figure 15a). Electrolytic etching reveals mainly intergranular propagation (Figure 15b). It has been established that the crack is influenced by the alternating dissolution (i.e., intergranular at the α/γ interfaces) and mechanical contributions (transgranular through grains).

(c) Examination of fracture surfaces:

Low-magnification SEM examination of the general appearance of the fracture surfaces enables a clear distinction to be made between fractions associated with fracture by SCC (net effect of the medium) and fractions associated with fracture by slow tension (mechanical effect) at different strain rates in the two media studied.

The SCC (brittle, F) fraction is largest (60%) at the critical strain rate (Figure 16a) and smallest at the strain-rate extremes (Figure 16b,c), mirroring exactly the Z% and W minima identified in Section 3.1.1. As demonstrated in Figure 16, the detailed fractographic sequence of the SCC fraction reveals the presence of an initiation zone, a clearly resolved transition zone to the ductile-tensile fracture region, and dimpled ductile zones (Figure 17b–d). Conversely, the reference fracture surface obtained in neutral silicone (Figure 18) demonstrates uniform ductility and dimpling, devoid of any selective dissolution or crack initiation. This provides an unambiguous, medium-only baseline against which the sea-water-induced damage described above can be attributed with confidence to the corrosive environment rather than to the mechanical environment. This shows that the fracture fraction corresponding to slow traction (mechanical effect) at different speeds in the two media studied is lowest at the speed of maximum susceptibility to SCC. It assumes greater proportions under slow traction conditions at deformation speeds on either side of this speed. The results of these observations are shown quantitatively in

Table 4. Influence of corrosion environment and strain rate on the resulting fracture fractions of SCC and slow tensile.

Corrosion Medium	Speed	Fraction of Facies Resulting from Fracture by SCC (%)	Fraction of Facies Resulting from Slow Tensile Fracture (%)
Synthetic seawater at 70 °C	2.38 × 10−7	47	53
	5 × 10−7	50	50
	2.38 × 10−6	60	40
	2.38 × 10−5	10	90
	10−4	0	100

.

Observation of the SCC fracture zone (Figure 16a–c) confirms the results of micrographic examinations of the barrel surfaces, i.e., clear selective dissolution of the ferrite by the same mechanism described above. On the other hand, the final fracture surface by slow traction is characterized by cup-shaped zones indicating significant plastic deformation prior to fracture (Figure 17d). The fracture surface of a specimen fractured in silicone at 70 °C at a maximum susceptibility speed of 2.38 × 10⁻⁶ s⁻¹ (Figure 18a) reveals a completely different appearance, free from any cracks initiated at the surface and any selective dissolution, with a ductile character typical of the slow tensile fracture surfaces of duplex steel (Figure 18b).

3.1.3. Slow Tensile Deformation Mechanism

The deformation mechanisms operating in the constituent phases (α and β) of the alloy under slow (ε = 2.38 × 10⁻⁶ s⁻¹) and conventional (ε = 5 × 10⁻³ s⁻¹) tensile conditions in open air were identified using SEM analysis of slip traces and their evolution during deformation. This analysis is based on metallographic examinations of the surfaces of flat specimens that have been polished and electrolytically etched, coupled with Vickers microhardness measurements (HV_0.05). The following findings emerged from these analyses:

-

Plastic deformation begins in the austenitic phase, which is lower in hardness (260 HV_0.05) than the ferritic phase (325 HV_0.05). In fact, traces of activation of a slip system are revealed in the austenitic islands at a total deformation rate of less than or equal to 1%, regardless of the deformation rate (slow or conventional) (Figure 19a).

-

When the deformation rate increases to greater than or equal to 3%, other slip systems are activated in the austenite. The spacing between the slip tracks is 2 µm or less, which testifies to the high deformability of this phase. This phase is reputed to be more ductile than ferrite (Figure 19b). This deformation is partially transmitted to the ferrite via the undamaged α/γ interface (Figure 19a).

This reveals the activation of a single slip system in the ferritic islands, with larger spacing of 3 µm or more between slip planes (Figure 20b). The emergence of slip planes in the ferrite leads to the formation of larger steps, as demonstrated by the intense contrast of the slip lines compared to the neighboring austenite (Figure 20c).

The hardness-evolution data presented in Figure 21 provides direct and quantitative confirmation of the aforementioned asymmetry between the two phases. The austenite hardness exhibited a steep and continuous increase with tensile strain, from approximately 305 HV_0.05 in the undeformed state to approximately 360–368 HV_0.05 at 20–30% strain. This is indicative of the significant strain hardening associated with its widespread, multiple slip, as previously described. In contrast, the ferrite hardness commences from an elevated initial value (approximately 345 HV_0.05) and undergoes a marginal increase, reaching approximately 350–355 HV_0.05 within the same strain range (an increase of approximately 3%). This outcome is consistent with its weak hardening and simple, single-slip-system deformation, which is confined to intense, sparsely distributed bands. This phenomenon can be attributed to the localized deformation, which occurs within a limited number of intensely active slip bands. In contrast to austenite, where deformation is distributed across numerous systems, the deformation in this case is confined to these specific bands.

Above a certain strain rate (6%), interfacial decohesion (α/γ) progressively isolates the austenite islands. These then participate less and less in plastic deformation. In this case, deformation of the alloy continues in the ferrite and remains confined to the first activated slip system (deformation highways). This explains the presence of bands of intense or persistent deformation with a sharper contrast in SEM images (Figure 22).

4. Discussion

The results of this study confirm that austenitic–ferritic steels are less susceptible to stress corrosion cracking in chloride environments than austenitic grades [18,21,38,39]. Slow tensile tests reveal a maximum susceptibility rate of around 10⁻⁶ s⁻¹ for the X6CrNiMoCu25-6 grade, which is fairly close to that of austenitic grades (5 × 10⁻⁷ s⁻¹). However, in terms of permissible stress at the maximum susceptibility rate, austenitic–ferritic grades demonstrate virtually unchanged strength in synthetic seawater at 70 °C (with only a 5% loss compared to the neutral silicone medium at 70 °C), i.e., 545 MPa compared to 745 MPa in the neutral silicone medium at 70 °C.

The hyper-quenched condition employed in this study (45% ferrite–55% austenite, with fine undissolved primary Ti/Nb carbides at the α/γ interfaces) differs from the solution-annealed or controlled-potential-tested condition typically used for 2205 duplex steel in the cited literature. The phase fraction and interface carbide distribution resulting from these different conditions modify the density and connectivity of the preferential dissolution/decohesion sites. Consequently, this modifies the strain-rate range over which these sites are activated. Thirdly, the test temperature is of significance. The present series of tests were conducted at 70 °C, a temperature which is representative of the actual pump-circuit service. In contrast, the AISI 316 and 2205 literature data were obtained at 90 °C. Since both anodic dissolution kinetics and repassivation kinetics are thermally activated, it is hypothesized that the lower test temperature used here will shift the critical strain-rate window and reduce the overall severity of attack in comparison with the sodium chloride solution at 90 °C, resulting in a 20% loss of mechanical strength [38].

This result can be explained by the low tensile consolidation of these materials (n ≈ 0.2) across a range of slow tensile test speeds (9 × 10⁻⁸ s⁻¹ ≤ v ≤ 10⁻⁴ s⁻¹) in either a chloride (synthetic seawater at 70 °C) or a neutral (silicone at 70 °C) environment. Therefore, it is expected that materials with low consolidation (n ≤ 0.2), such as austenitic–ferritic grades, will not exhibit significantly different critical stress thresholds for S.C.C. non-cracking in chloride or neutral environments. Conversely, crack initiation and propagation under tensile loading considerably limits permissible deformation (or distributed elongation at break) at maximum susceptibility rates. This limitation depends on the configuration of the network of initiated cracks, the propagation mechanism (intergranular or intragranular) and interactions with electrochemical reactivity in the test environment under consideration. Consequently, elongation at fracture or energy consumed per unit volume are the best indicators of stress corrosion damage in austenitic–ferritic stainless-steel grades.

This is evidenced by the limited stress loss (≤5%) and the moderate, rather than severe, ductility loss observed in this study, relative to the more aggressive literature conditions. The combination of these four factors: composition, heat treatment/microstructure fraction, test temperature and chloride concentration, provides a coherent, mechanistically grounded explanation for the comparable critical-strain-rate phenomenology exhibited by X6CrNiMoCu25-6, as compared to previously studied austenitic and duplex grades. Furthermore, the present study demonstrates that X6CrNiMoCu25-6 exhibits a markedly lower overall SCC severity under the free-corrosion, service-representative conditions investigated herein [40,41,42].

Furthermore, an analysis of crack initiation and propagation mechanisms explains the low susceptibility of austenitic–ferritic stainless-steel grades in the hyper-hardened state to CSC cracking, compared with austenitic grades. This analysis reveals an interaction between plastic deformation and localized pitting attack that accelerates damage to the duplex structure at maximum susceptibility rates, thereby limiting its deformability through predominantly intergranular fracture. In this case, selective dissolution by the germination and growth of corrosion pits in the ferritic phase at sliding plane emergence and α/γ interfaces favors interfacial decohesion propagation. This leads to premature failure and explains why the elongation at break (Ar%) decreases significantly without a corresponding decrease in stress. These results corroborate those of Wen-Ta Tsai et al. [35], who studied the 22.6Cr-4.8Ni-3.02Mo-0.11N-0.023C grade in a 26% NaCl corrosion environment at 90 °C with a controlled potential.

The ensuing comparative

Table 5. Comparative study of different stainless-steel grades.

Study (Material)	Environment/Temperature/Cl− Content	Strain Rate (Tested/Critical)	ΔRm (%)	Dominant Fracture Mode	Distinguishing Feature vs. Present Study
Present study—X6CrNiMoCu25-6 (cast, hyper-quenched duplex, 45% α–55% γ)	Synthetic seawater, 70 °C,	5 rates: 2.38 × 10−7–10−4 s−1; critical ≈ 10−6 s−1	≤5%	Mixed inter-/transgranular; ferrite-selective dissolution + α/γ decohesion	— (reference case)
AISI 316 austenitic Stainless steel [35]	26% NaCl solution, 90 °C	Critical strain rate ≈ 4.1 × 10−6 s−1	≈20%	Predominantly transgranular (typical of austenitic SCC)	Wrought, single-phase austenite markedly higher susceptibility than present duplex grade
2205 duplex stainless steel [34,37]	26% NaCl solution, 90 °C, imposed controlled electrochemical potential	SSRT under controlled potential (strain-rate matrix not directly comparable)	not directly comparable	Mixed inter-/transgranular; interface-related cracking	Wrought (not cast); imposed potential (not free corrosion); higher Cl−/T
General duplex stainless steel [7,19,26]	Various chloride often aged/welded	Strain rate not systematically varied	Not announced	Localized/pitting-assisted cracking associated with phase decomposition	Establishes that duplex grades generally outperform austenitic grades in Cl− media, but does not quantify strain-rate dependence

is to be situated subsequent to the prevailing qualitative comparison with AISI 316 and 2205 duplex steel. The following table will summarize the key distinguishing features, originality and novelty of the present work in relation to previously published SCC studies.

As demonstrated in the table, the present study is distinguished from prior SSRT/SCC literature on stainless steels by three combined features that, to the authors’ knowledge, have not been previously reported together: In order to proceed with this study, three criteria must firstly be met. Firstly, a cast (as opposed to wrought) duplex microstructure in the hyper-quenched condition must be demonstrated. Secondly, testing under free-corrosion conditions in a moderate-chloride, service-representative synthetic seawater at 70 °C is required, rather than in concentrated NaCl solutions or under imposed electrochemical potential. Thirdly, the explicit identification and mechanistic explanation of a critical strain rate (≈10⁻⁶ s⁻¹) governing SCC susceptibility in this specific grade is necessary. This combination constitutes the original and novel contribution of the present work to the SCC literature on duplex stainless steels.

5. Conclusions

1. The primary scientific accomplishment of this article in the realm of materials degradation science is the elucidation that, within a cast, hyper-quenched austenitic–ferritic (duplex) stainless steel, the susceptibility to SCC is governed by a competition between the kinetics of plastic deformation (slip-step emergence at the surface and at α/γ interfaces) and the kinetics of localized electrochemical dissolution (repeated pit nucleation and selective ferrite dissolution). The competition passes through a maximum at a critical strain rate of the order of 10⁻⁶ s⁻¹, at which passive-film rupture and dissolution are optimally synchronized, thereby maximizing pit coalescence, α/γ interfacial decohesion, and mixed inter-/transgranular crack propagation. This strain-rate/dissolution-competition mechanism serves to extend the extant theoretical framework of strain-assisted localized corrosion, which was previously established principally for single-phase austenitic alloys, to cast two-phase duplex microstructures.

2. It has been demonstrated that X6CrNiMoCu25-6 is an appropriate material for the fabrication of seawater-cooling pump castings. This is evidenced by its ability to retain a ductile, largely transgranular-slip-dominated behavior outside the critical strain-rate range. Quantitatively, the permissible reduction in fracture stress in synthetic seawater at 70 °C remains limited to ≤5%, even at the most damaging strain rate (approximately 10⁻⁶ s⁻¹), demonstrating that this grade retains most of its load-bearing capacity in service. However, it has been demonstrated that ductility (the degree to which a material can be elongated or reduced in area) can decrease by up to 30% at this same critical rate, in conjunction with a marked increase in surface SCC crack density. For engineers and operators, this translates into a practical design and inspection criterion. Components that have been subjected to slow, sustained strain (residual stresses resulting from casting or weld repair, or long-term service loading) should be assessed using elongation- or ductility-based acceptance criteria, rather than stress-based criteria alone. It is imperative to direct particular attention to the strain-rate range of 10⁻⁶ s⁻¹.

3. The experimental design of the study was such that testing was restricted to a single temperature (70 °C) and a single chloride environment (synthetic seawater). The impact of temperature on the critical strain rate was not explored, nor was the impact of chloride concentration. Each strain rate was represented by a single specimen, without replicate testing or statistical treatment of scatter. Consequently, the reported 5% stress reduction and the ductility loss at the critical rate have not yet been validated against normal SSRT experimental variability. It is imperative to acknowledge that only specimens devoid of notches were employed in this study. This methodological decision was taken to avoid any potential bias in results, as the effect of pre-existing casting defects, porosity, or weld-related microstructural and residual-stress heterogeneities encountered in service cannot be captured by using such specimens.

4. Prospective avenues for future research and the potential for further development within this field:

-

The strain-rate/dissolution-competition framework established for X6CrNiMoCu25-6 is expected to be transferable to other cast or wrought duplex and super-duplex stainless steels, and potentially to other two-phase or multiphase corrosion-resistant alloys exposed to chloride-containing environments under sustained or cyclic straining. This is due to the fact that the critical strain rates and dissolution kinetics are grade-specific.

-

In the longer term, the coupling of this experimental strain-rate/dissolution-competition description with physics-based or data-driven predictive models could allow the critical strain rate and SCC risk of a given duplex microstructure and environment to be predicted a priori. This would support the development of updated design codes and inspection standard for cast duplex components in seawater-cooling and other chloride-service applications, and guide future research on SCC of other stainless-steel families under similarly aggressive corrosive conditions.