Effect of Electrochemical Hydrogen Degradation on the Bond Microstructure of Explosively Welded Joints

Abstract

This study investigates hydrogen embrittlement mechanisms at the interfaces of explosively welded joints between 304L austenitic stainless steel and carbon/low-alloy steels (St41k, 15HM), focusing on the unique properties of local melting zones (LMZs) formed during joining. Advanced microstructural characterization, including scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDS), and microhardness testing, was combined with controlled electrochemical hydrogen charging. Results demonstrate that while base materials suffered substantial hydrogen-induced degradation—blistering in carbon steels and microcracking in stainless steel—the LMZ exhibited exceptional resistance to hydrogen damage. Compositional analyses revealed that the LMZ possessed intermediate chromium (4.8–8.8 wt.%) and nickel (1.7–3.6 wt.%) contents, reflecting mixing from both plates, and significantly higher microhardness compared to adjacent zones. The superior hydrogen resistance of the LMZ is attributed to their refined microstructure, increased density of hydrogen trapping sites, and non-equilibrium phase composition resulting from rapid solidification. These findings indicate that tailoring the process of the LMZ in clad steel joints can be an effective strategy to mitigate hydrogen embrittlement risks in critical hydrogen infrastructure.

1. Introduction

The global transition toward hydrogen-based energy systems has intensified the need for materials capable of withstanding hydrogen-containing environments without catastrophic failure [1,2]. Clad steels represent a particularly promising solution, combining the corrosion resistance of stainless steel with the structural economy of carbon steels [3,4]. However, the dissimilar metal interfaces created by this process introduce compositional gradients, residual stresses, and unique microstructural features that profoundly influence hydrogen embrittlement susceptibility. Understanding hydrogen behavior at these interfaces has become critical for ensuring the long-term integrity of hydrogen infrastructure, including storage tanks, pipelines, and pressure vessels operating under severe service conditions [5,6].

Despite being classified as a solid-state joining process, explosive welding produces localized melting at interfaces due to extreme energy dissipation during collision, creating regions with compositions intermediate between the parent materials and refined microstructures resulting from rapid solidification [7]. These LMZs represent a microstructural phenomenon that has received limited attention regarding their hydrogen embrittlement behavior, yet they may hold the key to understanding superior interface performance in hydrogen-containing environments [8].

Hydrogen embrittlement in steels remains one of the most complex and debated failure mechanisms in materials science, with susceptibility increasing dramatically in high-strength alloys above 1000 MPa tensile strength [6,9]. Two primary mechanisms have been identified: Hydrogen-Enhanced Decohesion (HEDE) [10], where hydrogen accumulation at grain boundaries or interfaces reduces cohesive strength and enables brittle, and Hydrogen-Enhanced Local Plasticity (HELP) [11], which involves hydrogen-facilitated dislocation nucleation and mobility leading to localized plastic failure with an apparently brittle fracture appearance. Recent advances by Djukic et al. [12] have proposed a unified HELP+HEDE model, demonstrating that both mechanisms operate synergistically depending on local hydrogen concentration, stress state, and microstructural features. At moderate hydrogen concentrations, HELP dominates through enhanced dislocation mobility, while at higher concentrations or at specific microstructural features such as grain boundaries, HEDE becomes significant, leading to catastrophic brittle fracture. This mechanistic complexity is further compounded in dissimilar metal joints where compositional gradients, phase transformations, and residual stresses create heterogeneous hydrogen distribution patterns and variable susceptibility across the interface region [13,14].

Recent advances in characterization techniques have enabled unprecedented insights into hydrogen behavior at microstructural length scales. Atom probe tomography has revealed hydrogen segregation to specific crystallographic features with near-atomic resolution [15], while in situ electrochemical nanoindentation has demonstrated real-time hydrogen effects on mechanical properties [3]. Thermal desorption spectroscopy continues to provide quantitative information about hydrogen trapping energies and site distributions [16]. These experimental capabilities, combined with computational modeling approaches [17], have transformed our understanding of hydrogen-microstructure interactions [14].

Despite extensive research on hydrogen embrittlement in steels and welded joints, the specific behavior of local melting zones (LMZs) at explosive welding interfaces remains poorly understood. While previous studies have documented the microstructural characteristics of LMZs [18], including their intermediate chemical composition and refined grain structure, systematic investigations of their hydrogen embrittlement resistance are notably absent from the literature. The few studies that have examined hydrogen effects in explosively welded joints report conflicting results, with some identifying interfaces as preferential failure sites under hydrogen charging while others observe superior performance [19,20]. These discrepancies likely stem from variations in welding parameters, parent material combinations, hydrogen charging conditions, and the presence or absence of well-developed melting zones. Furthermore, the unique solidification conditions in LMZs—characterized by cooling rates of 10⁶–10⁸ K/s and mixing of dissimilar metals—create microstructures fundamentally different from those produced by conventional welding or heat treatment processes, suggesting that existing models of hydrogen behavior may not directly apply.

The role of microstructural refinement in hydrogen embrittlement resistance has been extensively debated [14]. While grain boundaries engineering through severe plastic deformation has shown promise in some systems, the introduction of high-density interfaces can also create fast diffusion pathways and preferential cracking sites [21]. The ultrafine microstructures produced by rapid solidification in explosive welding represent a unique opportunity to investigate these competing effects under conditions that are difficult to replicate through conventional processing routes.

This study addresses critical knowledge gaps regarding hydrogen embrittlement at explosive welding interfaces through comprehensive experimental investigation of 304L austenitic stainless steel welded to carbon/low-alloy steels (St41k, 15HM). Specifically, we aim to: (1) characterize the compositional and microstructural features of local melting zones compared to parent materials and directly bonded regions, (2) quantify hydrogen-induced damage in different zones through controlled electrochemical charging, (3) correlate microhardness measurements with hydrogen resistance, and (4) elucidate the mechanisms responsible for the observed differences in hydrogen embrittlement susceptibility. By combining scanning electron microscopy, energy-dispersive spectroscopy, and microhardness profiling with systematic hydrogen charging experiments, this study provides new insights into the role of LMZs as potential hydrogen-resistant barrier layers in clad steel structures intended for hydrogen service applications.

2. Materials and Methods

2.1. Material Preparation and Explosive Welding Process

The explosive welding process was conducted using AISI 304L austenitic stainless steel (flyer plate, 8 mm thickness) and high-strength carbon and low-alloy steel, separately (base plate, 20 mm thickness, both). The chemical compositions of materials are presented in

Table 1. The chemical composition (wt%) of the studied materials.

Material	Chemical Composition [% wt]
	C	Mn	Si	P	S	Cr	Ni	Cu	Fe	Ti
Steel 304L	0.023	1.979	0.313	0.039	0.0011	18.08	8.3	----	R	---
Steel St41k	0.155	1.35	0.3	0.012	0.005	0.039	0.057	0.007	R	0.02
Steel 15HM	0.15	0.59	0.25	---	0.0015	0.83	0.093	---	R	---

. Prior to joining, all surfaces were mechanically ground to remove oxide layers and cleaned with acetone to ensure optimal bonding conditions.

The explosive welding setup utilized a parallel configuration with a standoff distance of 15 mm. ANFO (Ammonium Nitrate/Fuel Oil mixture) explosive with a detonation velocity of approximately 2800 m/s was employed at a loading ratio of 1.8 kg/m². The collision angle and velocity were calculated to fall within the welding window necessary for wave formation without excessive melting or insufficient bonding. Post-welding, the clad plates were visually inspected for surface defects and subjected to ultrasonic testing to verify bond continuity.

2.2. Microstructural Characterization

Specimens for microstructural analysis were sectioned perpendicular to the welding interface using wire electrical discharge machining (WEDM) to minimize mechanical deformation. Metallographic preparation involved grinding with SiC papers (240 to 2400 grit), followed by polishing with diamond suspensions (6 μm, 3 μm, 1 μm) and final polishing with colloidal silica (0.05 μm).

Microstructure observations were performed on samples before and after hydrogen charging using scanning electron microscopes (SEM, Hitachi 3500N and Hitachi SU-70, Tokyo, Japan, operated at 30 kV) equipped with energy-dispersive X-ray spectroscopy (EDS). For microstructural examination, samples were chemically etched with 4% Nital solution or electrolytically etched in oxalic acid solution. Microhardness profiles were obtained using a Vickers microhardness tester to characterize the deformation effect in the joint and melted zones.

2.3. Microhardness Measurements

Microhardness measurements were performed using a Vickers indenter (HV0.1 standard) with a load of 100 g and a dwell time of 15 s. Individual indentations produced characteristic diagonal impressions of approximately 20–30 μm, establishing the spatial resolution of the measurement. Successive measurements were taken at intervals of 50 μm across the interface region, ensuring adequate spacing between indentations to avoid overlapping stress fields and stress-induced artifacts. A minimum of 5 individual indentations were recorded in each distinct zone (parent materials, deformation-affected regions, and melting zones) to establish representative hardness values.

2.4. Hydrogen Charging and Thermal Desorption Analysis

The susceptibility of the materials to hydrogen embrittlement was evaluated by cathodic hydrogen charging. Samples with carefully polished cross-sections exposing the bonding interface were charged using a current density of 0.05 A/cm² for 24 h at ambient temperature (approximately 20 °C), with a platinum anode in 0.5 M H₂SO₄ solution containing 1 mg/L As₂O₃ as a hydrogen recombination poison. This current density was selected to achieve substantial hydrogen uptake while avoiding excessive hydrogen evolution and surface damage. The As₂O₃ poison inhibits the surface recombination of atomic hydrogen to molecular H₂, thereby maximizing hydrogen absorption into the material rather than gas evolution. The experimental setup is illustrated in Figure 1.

Following hydrogen charging, samples were immediately examined by optical and scanning electron microscopy to document hydrogen-induced damages. The charging conditions used in this study represent an accelerated test that produces hydrogen concentrations and damage patterns relevant to long-term service in hydrogen-containing environments.

3. Results

3.1. Interfacial Microstructure Characterization

3.1.1. Macroscopic Interface Morphology

The explosive welding process successfully produced metallurgically bonded interfaces between the AISI 304L stainless steel and both St41k or15HM steels, exhibiting the characteristic wavy morphology typical of explosive welds (Figure 2A,C). The wavelength ranged from 800 to 1200 μm with amplitudes of 150–250 μm, consistent with the collision parameters employed. Visual inspection and optical microscopy revealed no unbonded regions or gross defects along the examined interface length, indicating successful welding under the selected parameters.

Localized melting zones appeared as resolidified material concentrated at wave crests and troughs, where collision energy concentration was maximum (Figure 2B,C). These melting zones exhibited dimensions ranging from 50 to 400 μm in the largest dimension, with highly irregular morphologies conforming to the interstitial spaces between colliding materials. The melting zones displayed a distinct etching response compared to adjacent parent materials (Figure 2B,C), indicating significant compositional and microstructural differences. In the vicinity of the bonding interface, elongated and refined grains were observed (Figure 2D), providing evidence of severe plastic deformation during the high-velocity collision process.

3.1.2. Compositional Analysis Across the Interface

EDS point analyses and elements mappings across the interface revealed sharp compositional transitions between the 304L stainless steel and carbon or low-alloy steel parent materials in directly bonded regions. The melting zones, however, exhibited intermediate compositions consistent with mixing of molten materials from both sides during the brief liquid state.

For the 304L/St41k joint, chromium and nickel concentrations in the melting zones (Cr: 4.78–5.28 wt%, Ni: 1.74–2.04 wt%, based on points 2–4 in Figure 3A) fell between the 304L steel (Cr: 19.58 wt%, Ni: 7.37 wt%, point 1) and St41k steel (Cr: 0.17–0.23 wt%, Ni: 0.12–0.20 wt%, points 5–6) compositions. For the 304L/15HM joint, the melting zones showed higher chromium and nickel content (Cr: 8.67–8.75 wt%, Ni: 3.39–3.58 wt%, points 2–4 in Figure 3B) due to the slightly higher chromium content in the 15HM base material. Iron concentration remained relatively constant at 81–92 wt% throughout the melting zones in both joints.

Element distribution maps (Figure 4) confirmed mutual diffusion between elements at all bonding interfaces, with clear evidence of metallurgical mixing in the melting zones. The distribution of chromium and nickel at each bonding interface shows gradual variations across the LMZ width, suggesting that the melting zones consist of metastable phases formed during rapid solidification rather than simple mechanical mixing.

Microhardness measurements across the explosive-welded interfaces revealed three distinct zones with characteristic hardness values (Figure 5). For both joints, the parent and flyer plates at distances greater than 300 μm from the interface exhibited baseline hardness values representative of the undeformed materials: approximately 150–170 HV for St41k, 180–200 HV for 15 HM, and 180–200 HV for 304L.

The melting zones themselves exhibited dimensions ranging from 50 to 400 μm in the largest dimension. However, the microhardness gradient extending approximately 300 μm on each side of the interface reflects plastic deformation of the parent materials during the high-velocity collision process, rather than a direct consequence of the melting zones. The deformation-affected zone, distinguished from the melting zone proper, is characterized by increased hardness compared to the undeformed base materials.

Within approximately 300 μm of the interface, hardness increased significantly in both the parent and flyer plates due to plastic deformation during the high-velocity collision. This deformation-affected zone showed hardness values of 200–250 HV in the carbon steels and 220–280 HV in the 304L, representing increases of approximately 30–60% compared to the base material values.

The melting zones exhibited the highest hardness values 280 HV, significantly exceeding both the parent materials and the deformation-affected zones. This elevated hardness can be attributed to two factors: (1) the refined microstructure resulting from rapid solidification (cooling rates of 10⁶–10⁸ K/s), which produces fine grain sizes with effective Hall-Petch strengthening, and (2) the intermediate chemical composition with potentially favorable phase distribution. The microhardness gradient extending approximately 300 μm on each side of the interface provides evidence of the extent of plastic deformation during the explosive welding process.

Cathodic hydrogen charging at 0.05 A/cm² for 24 h produced significantly different degradation in the various materials and zones examined (Figure 6 and Figure 7).

Carbon Steel Base Materials:

In both St41k and 15HM carbon steels hydrogen-induced blisters formed on the surface, with dimensions typically ranging from 100 to 500 μm in diameter (Figure 6A,B). Many of these blisters exhibited associated cracks propagating from the blister periphery into the surrounding material. These blisters form through the following mechanism: atomic hydrogen diffuses into the steel and accumulates at internal interfaces such as inclusions, laminations, or pre-existing microvoids. The hydrogen atoms recombine to form molecular H₂, creating high internal pressure that exceeds the local yield strength of the material, resulting in blister nucleation and growth. The subsequent cracking indicates that the internal pressure exceeded the fracture toughness of the hydrogen-charged material.

While both St41k and 15HM carbon steels exhibited hydrogen-induced blistering, the different inclusion populations and microstructural characteristics warrant explanation of observed variations.

St41k steel contains very low chromium (0.039 wt%) and negligible molybdenum, producing a typical ferritic-pearlitic microstructure with predominantly oxide-based inclusions characteristic of killed steels. The 15HM steel, by contrast, contains moderate chromium (0.83 wt%) and higher manganese (0.59 wt%), promoting formation of different inclusion morphologies and enabling precipitation hardening through formation of chromium nitrides and other phases.

The variations in hydrogen-induced blistering severity between these steels reflect fundamental differences in the following:

(1) Inclusion Characteristics. Size, morphology, and distribution of nonmetallic inclusions differ between compositions. Inclusions serve as preferential hydrogen accumulation sites and blister nucleation locations. Larger or more numerous inclusions facilitate more severe blistering at lower hydrogen concentrations.

(2) Matrix Strength and Yield Strength. The elevated chromium and molybdenum content of 15HM provides increased precipitation hardening compared to St41k, resulting in somewhat higher yield strength. This affects the critical hydrogen pressure required for blister nucleation (P_crit ∝ σ_y), potentially explaining variations in observed blister sizes and distribution patterns.

(3) Dislocation Density and Substructure. The different thermomechanical processing histories result in different dislocation densities and subgrain structures. Such substructure differences affect hydrogen diffusivity and the spatial distribution of potential hydrogen accumulation sites.

Despite these differences in detailed damage mechanisms, both materials exhibit significantly greater hydrogen-induced degradation compared to the LMZ regions, reinforcing the conclusion that explosive welding creates uniquely hydrogen-resistant interfaces through formation of well-developed melting zones.

Austenitic Stainless Steel:

The 304L austenitic stainless steel exhibited a different damage mode, characterized by networks of fine microcracks propagating across the surface (Figure 6C,D). These cracks, typically less than 50 μm in length and forming interconnected networks, are consistent with the hydrogen-enhanced localized plasticity (HELP) mechanism. Under cathodic charging, hydrogen facilitates localized slip band formation, and subsequent crack nucleation occurs at stress concentrations created by dislocation pile-ups. The network pattern suggests multiple sites of hydrogen-assisted crack initiation distributed across the charged surface.

Local Melting Zones:

In contrast to both parent materials, the local melting zones at the bonding interfaces remained free of visible hydrogen-induced damages after identical charging conditions (Figure 7). Despite exposure to the same hydrogen environment that caused blistering in carbon steels and microcracking in austenitic steel, no blisters, cracks, or other degradation features were observed in the LMZ regions. The transition from damaged parent material to undamaged LMZ was sharp and clearly delineated along the interface, as shown in Figure 7a where blisters in the parent steel and cracks in the austenitic stainless steel are visible on either side of the intact melting zone.

This observation provides direct experimental evidence that local melting zones possess exceptional hydrogen embrittlement resistance under accelerated test conditions representative of hydrogen exposure. The mechanisms responsible for this performance are discussed later.

4. Discussion

The experimental observations reveal a significant difference in hydrogen embrittlement susceptibility between local melting zones (LMZs) and parent materials at explosive welding interfaces. While both carbon steel and 304L austenitic stainless steel exhibited significant hydrogen-induced degradation—manifested as blistering and cracking in carbon steel and net of microcracks in stainless steel—the LMZ remained free of visible damage under identical hydrogen charging conditions. This exceptional hydrogen resistance arises from a combination of microstructural refinement, compositional effects, and mechanical constraints, creating a robust multi-barrier defense against hydrogen-induced failure.

The microstructure of the LMZ, formed during rapid solidification at cooling rates of 10⁶–10⁸ K/s, is markedly refined compared to parent materials. This ultrafine grain structure, confirmed by its distinct etching response and elevated microhardness, provides a high density of grain boundaries that serve as effective hydrogen traps. High-angle grain boundaries, known to bind hydrogen with energies of 45–60 kJ/mol, immobilize hydrogen atoms and prevent their migration to critical regions. Studies, such as those by Islam et al. [5], have shown that reducing grain size from 88 μm to 15 μm can more than double total trap density. Although grain size was not directly measured in this work, the observed hardness increase indicates significant refinement. The “dilution effect” associated with fine-grained materials further contributes to hydrogen resistance, as the distributed network of trap sites reduces local hydrogen concentration at any given interface, suppressing crack initiation via the hydrogen-enhanced decohesion (HEDE) mechanism. This phenomenon has been documented in ultrafine-grained steels, where even high total hydrogen uptake does not result in embrittlement due to effective distribution among numerous trapping sites [22].

In addition to grain boundary trapping, explosive welding induces plastic deformation near the interface, as reflected by elevated hardness extending roughly 300 μm into both parent materials. This deformation introduces high dislocation densities that act as additional hydrogen traps with binding energies of 25–35 kJ/mol. Although these traps are weaker than grain boundaries, their abundance significantly enhances total hydrogen capture capacity. The absence of cracks or blisters in LMZ after hydrogen exposure implies that hydrogen was effectively immobilized within these trap sites, preventing its accumulation at damage-prone locations.

Compositional gradients within LMZs further contribute to hydrogen resistance. Energy dispersive spectroscopy (EDS) revealed that LMZ compositions lie between those of the parent materials, containing intermediate levels of chromium (5–9 wt%) and nickel (1.7–3.6 wt%). These elements play a crucial role in mitigating hydrogen effects. Chromium reduces hydrogen diffusivity, stabilizes protective surface oxides, and forms carbides that act as deep hydrogen traps with high binding energies (75–95 kJ/mol) [23]. Nickel, on the other hand, stabilizes austenitic or mixed austenite-ferrite microstructures that exhibit high hydrogen solubility but low diffusivity, effectively reducing hydrogen migration to critical zones. Rapid solidification of such intermediate compositions likely promotes the formation of metastable and fine-grained microstructures containing numerous interfaces, precipitates, and segregation patterns that further enhance hydrogen trapping efficiency [24].

Mechanical constraint effects arising from the geometric configuration of LMZs also play a key role. The LMZs are embedded within wavy interfaces between high-strength materials, providing compressive restraint that suppresses hydrogen blister formation even under internal pressure buildup. This constraint, combined with the fine-grained microstructure, may also arrest microcrack propagation at very small scales. Additionally, rapid solidification and differential thermal contraction likely generate compressive residual stresses within LMZs, which further oppose crack initiation and propagation, although direct measurements are still required to confirm this hypothesis.

Overall, the superior hydrogen resistance of LMZs could be a product of synergistic interactions between refined microstructure, beneficial compositional gradients, and mechanical constraints—a multi-barrier approach that prevents visible hydrogen-induced degradation. Within the framework of the HELP + HEDE unified model proposed by Djukic [12], this can be interpreted as the suppression of both the hydrogen-enhanced localized plasticity (HELP) and decohesion (HEDE) mechanisms due to the effective reduction in mobile hydrogen concentration. In contrast to the parent materials, where sufficient mobile hydrogen activates either HELP (in 304L) or HEDE (in carbon steel), the LMZ maintains hydrogen concentrations below the critical threshold for either mechanism, leading to their remarkable stability.

The intermediate chromium and nickel content of the LMZ (5–9 wt% Cr, 1.7–3.6 wt% Ni), combined with cooling rates of 10⁶–10⁸ K/s, strongly suggests formation of non-equilibrium phase mixtures during rapid solidification. In the Fe-Cr-Ni ternary system, intermediate compositions in this range can produce:

(1)

Mixed austenite–ferrite microstructures (duplex-type character), which combine austenite’s high hydrogen solubility with ferrite’s lower diffusivity,

(2)

Retained metastable austenite with high dislocation densities,

(3)

Potentially acicular ferrite or low-carbon martensite depending on precise cooling conditions and carbon redistribution.

The ultrafine grain size observed in these zones further increases the density of interfaces and promotes hydrogen trapping. While X-ray diffraction (XRD) analysis was not performed in the current study due to the small and geometrically irregular nature of the LMZ regions (50–400 μm), the elevated microhardness (280 HV) and distinct etching behavior strongly support the presence of refined, multi-phase microstructures. Future work employing electron backscatter diffraction (EBSD) or transmission electron microscopy (TEM) would provide definitive phase identification and detailed crystallographic characterization of the LMZ regions.

Microstructural evidence combined with literature data [14] suggests that several complementary hydrogen-trapping mechanisms operate concurrently within the local melting zones (LMZs). The ultrafine grain structure—indicated by elevated hardness (~280 HV) and distinct etching behavior—introduces a high density of high-angle grain boundaries that strongly trap hydrogen (45–60 kJ/mol) [25]. Severe deformation during explosive welding further generates high dislocation densities, providing additional, though weaker, trapping sites (25–35 kJ/mol) [26]. The chromium-enriched solidification chemistry of the LMZ (5–9 wt% Cr) also favors the formation of chromium carbides, which act as deep traps with binding energies of 75–95 kJ/mol [27]. Moreover, the rapid solidification conditions likely produce mixed austenite–ferrite microstructures and metastable phases, whose interfaces and austenitic regions contribute additional hydrogen sequestration due to their high solubility and low diffusivity. The combined action of these trapping mechanisms distributes hydrogen across a wide spectrum of trap strengths, thereby maintaining the mobile hydrogen concentration below the levels needed to activate HELP or HEDE [28]. This multi-mechanism trapping behavior provides a plausible microstructural basis for the superior hydrogen resistance of the LMZ compared with the parent materials. A remaining limitation is the absence of Thermal Desorption Spectroscopy (TDS) data, which would enable quantitative differentiation of the various trap types.

In the context of the existing literature, hydrogen embrittlement in explosive-welded structures has been sparsely investigated, with most studies focusing on mechanical strength or corrosion resistance rather than hydrogen effects. Yao et al. [29] reported hydrogen-related degradation in clad steels but did not differentiate between specific interfacial regions. Similarly, Chen et al. [3] demonstrated hydrogen sensitivity through ultrasonic diagnostics without mechanistic interpretation. Mazancová et al. [13] observed that hydrogen-induced cracks in 304 SS/Ti explosive welds occurred exclusively in intermetallic zones, leaving other interfacial areas intact—an observation consistent with the present finding that LMZ remain undamaged despite adjacent materials showing embrittlement. The present study is thus the first to systematically document, with direct visual and microstructural evidence, the superior hydrogen resistance of LMZ compared to their parent materials.

These results also align with broader research on grain refinement and hydrogen embrittlement. Studies by Mine et al. [22] and Islam et al. [5] highlight that reducing grain size and increasing trap site density enhance hydrogen tolerance by lowering effective diffusivity and preventing hydrogen accumulation. However, the grain refinement achieved through explosive welding’s rapid solidification offers additional advantages over plastic deformation techniques—namely, optimized intermediate compositions, unique non-equilibrium phases, and mechanically constrained configurations that collectively enhance hydrogen resistance beyond what can be achieved by mechanical processing alone.

From an application standpoint, these findings hold significant implications for the design of clad steel structures and hybrid joints intended for hydrogen environments. The demonstrated hydrogen resilience of LMZ suggests that controlled formation of such zones during processing could be intentionally leveraged to improve component durability in hydrogen pipelines, pressure vessels, and energy storage systems. By harnessing the multi-barrier defense strategy observed in LMZ—combining microstructural refinement, compositional optimization, and mechanical constraints—future material designs can achieve enhanced performance and safety in demanding hydrogen service conditions.

The observation that local melting zones (LMZs) remain unaffected by hydrogen exposure, whereas both parent alloys degrade, suggests that LMZs serve as effective hydrogen-resistant interfaces capable of arresting or diverting crack growth. This implies that incorporating well-controlled LMZs at structurally critical regions of explosive-welded components could substantially improve their durability in hydrogen-rich environments. While this study focuses on hydrogen charging at 20 °C, practical applications such as hydrogen pipelines and storage tanks typically operate at 100–300 °C. Consequently, the thermal stability of LMZs under these elevated temperatures becomes a key design consideration. The LMZs examined here exhibit non-equilibrium microstructures produced by extremely rapid solidification (10⁶–10⁸ K s⁻¹), resulting in ultrafine grains and metastable phases. Prior studies on rapidly solidified systems indicate that such microstructures remain thermally stable up to approximately 400 °C.

This study provides direct experimental evidence of exceptional hydrogen resistance in local melting zones (LMZs) but has several limitations. It lacks thermal desorption spectroscopy for quantitative hydrogen measurements, direct grain size characterization by EBSD or TEM, testing at elevated temperatures relevant to real hydrogen service, and uses accelerated cathodic charging conditions that do not mimic gradual hydrogen uptake in practice. Additionally, only one stainless steel and two carbon/low-alloy steels were examined, limiting generality. Despite these limitations, consistent observations across materials and complementary microstructural data support the findings. Addressing these gaps in future research would strengthen the mechanistic understanding and applicability of the results.

5. Conclusions

This study provides a comprehensive investigation of hydrogen embrittlement behavior at explosive welding interfaces between 304L austenitic stainless steel and carbon/low-alloy steels (St41k and 15 HM). The key conclusions are as follows:

• Explosive welding produced metallurgically bonded interfaces characterized by distinct wavy morphology and localized melting zones (LMZs) with dimensions ranging from 50 to 400 μm.
• Compositional analysis revealed that LMZs possess intermediate chromium (4.8–8.8 wt%) and nickel (1.7–3.6 wt%) contents, reflecting metallurgical mixing during rapid solidification.
• Microhardness measurements indicate that LMZs exhibit significantly higher hardness than the parent materials, attributable to refined microstructure and plastic deformation.
• Hydrogen charging at 0.05 A/cm² for 24 h induced blisters with cracks in carbon steels and networks of microcracks in 304L stainless steel, whereas LMZs remained free of visible hydrogen-induced damage.
• The superior hydrogen embrittlement resistance of LMZs is attributed to their refined microstructure that provides a high density of hydrogen trapping sites, optimized chemical composition, and possibly beneficial compressive residual stresses from rapid solidification.

These findings highlight the role of local melting zones as effective barriers against hydrogen-assisted failure propagation in clad steel structures, suggesting that controlling LMZ formation during explosive welding can enhance durability in hydrogen service environments.