Investigation of Impact Behavior of STS304L Steel Plate Under Cryogenic Temperature

Abstract

STS304L is widely used in liquefied natural gas cargo containment systems for cryogenic liquefied gas storage because of its excellent mechanical properties at low temperatures. However, unpredictable sloshing impacts can induce excessive plastic deformation, leading to phase transformation from austenite to martensite. This study investigated the impact resistance of STS304L under cryogenic conditions through drop-weight impact tests. Temperature sensitivity was analyzed using electron backscatter diffraction to quantify plastic deformation and phase fraction. The results showed that, as the temperature decreased, the energy absorption and stiffness increased, whereas the plastic deformation remained relatively constant. Energy absorption increased by 59.63% at −100 °C and 68.80% at −193 °C compared with that at 20 °C. The martensite fraction, measured at the end of the hemispherical impact region, increased from 19.26% at 20 °C to 77.85% at −100 °C and 96.87% at −193 °C, indicating significant strain-induced martensitic transformation at cryogenic temperatures.

1. Introduction

With the accelerating global transition to clean energy, interest in liquefied natural gas (LNG) continues to grow. LNG, primarily composed of methane, is stored and transported in liquid form at an extremely low temperature of approximately −163 °C to maximize efficiency and minimize its volume. However, maintaining LNG in its liquid state under such extreme cryogenic conditions requires advanced insulation and containment technologies to ensure safe and reliable transportation [1].

LNG carriers are equipped with specialized insulation systems designed to withstand the harsh conditions of maritime transport. These systems must endure loading stresses from wave-induced motions and environmental factors such as temperature fluctuations. Therefore, insulation systems play a crucial role in minimizing heat ingress, preserving the structural integrity of containment systems, and preventing LNG leakage [2,3].

Among the most widely used LNG containment technologies is the Mark III system, which ensures both thermal integrity and structural stability. This insulation system typically incorporates materials such as STS304L stainless steel, plywood, and reinforced polyurethane foam [4,5]. Each material plays a distinct role in maintaining the system’s thermal insulation performance, mechanical strength, and overall durability [6].

A key component of this insulation system is the primary barrier, which is in direct contact with the LNG cargo. Typically made of STS304L stainless steel, the primary barrier serves as the first line of defense against potential leaks while also ensuring the structural integrity of the containment system under various loading conditions [7]. Given its exposure to cryogenic temperatures, impact forces, and cyclic loadings, the primary barrier must exhibit a high mechanical reliability to withstand these challenges.

Despite its advantages, the structural integrity of STS304L can be compromised under repeated or sudden impact loadings. LNG carriers are subjected to sloshing loads, where the movement of the LNG inside the cargo tanks generates transient impact forces on the containment system. Over time, localized deformation, fracture initiation, and progressive damage accumulation can occur, weakening the primary barrier and potentially affecting long-term performance [8].

Owing to these concerns, extensive research has been conducted on the mechanical behavior and impact resistance of STS304L. Ibrahim et al. investigated the impact toughness of AISI 304L stainless steel at cryogenic temperatures, focusing on the effects of stacking fault energy and strain-induced martensitic transformation on fracture behavior [9]. Their study found that the impact energy of 304L stainless steel gradually decreased as the temperature dropped. Kim et al. investigated the failure mechanisms of STS304L primary barriers under cyclic impact loads, identifying localized necking, ductile failure, and shear failure as key damage modes [10]. Kim et al. examined the impact response of STS304L corrugated sheets in LNG cargo containment systems [11]. They demonstrated that the accumulated damage due to repeated impacts compromised structural integrity over time. Du et al. investigated the stress and strain distributions in membrane-type LNG cargo tanks under high-sea conditions, highlighting the structural vulnerabilities of STS304L under repeated loading and defective conditions [12]. Aljeaan and Mohamed evaluated the low-temperature impact behavior of AISI 304L stainless steel and compared its impact resistance at different temperatures [13]. The results showed that the impact energy decreased as the temperature dropped, with the lowest impact energy observed at −80 °C.

Although extensive research exists, studies on the drop-weight impact testing of STS304L under extreme cryogenic conditions remain limited. Unlike cyclic impact testing, which simulates long-term fatigue, drop-weight impact testing is crucial for directly evaluating single-impact energy absorption and fracture behavior relevant to the transient impacts caused by sloshing in LNG cargo containment systems.

Additionally, austenitic steels, such as STS304L, undergo strain-induced martensitic transformation at low temperatures [14,15]. This phenomenon involves a phase transition in the material’s crystal structure, changing from austenite (a face-centered cubic (FCC) structure) to martensite (a body-centered cubic (BCC) structure). The formation of strain-induced martensite significantly alters the steel’s mechanical properties, including its strength, ductility, and fracture resistance, thereby influencing its overall impact performance [16].

To address this research gap, this study investigated the impact resistance of STS304L at various temperatures (20 °C, −100 °C, and −193 °C) using drop-weight impact tests. The impact energy was varied from 50 J up to the crack initiation point to assess the correlation between energy absorption and deformation behavior. Additionally, electron backscatter diffraction (EBSD) analysis was conducted to examine plastic deformation and strain-induced martensitic transformation in the impact zone.

2. Experiment Details

2.1. Material and Specimen

This study examines changes in the mechanical behavior of STS304L due to temperature-dependent strain-induced martensitic transformation. The resistance of metastable stainless steel to cold embrittlement varies depending on its chemical composition, with a higher nickel content acting as an austenite stabilizer and improving resistance to cold embrittlement [17]. The chemical composition of the STS304L used in this study is presented in

Table 1. Chemical composition (wt%) of STS304L.

C	Si	Mn	P	S	Ni	Cr	Mo
0.018	0.4	1.63	0.03	0.001	8.2	18.1	0.22

. The specimens were fabricated in accordance with ASTM A240/A240M-17 standards [18] and cut using a water jet. The prepared samples had a square shape with dimensions of 50 mm (L) × 50 mm (W) × 1 mm (T), as shown in Figure 1a. Figure 1b presents the initial microstructure of STS304L, which is primarily composed of the austenite (γ) phase, with characteristic features observed within the grains and along the grain boundaries.

2.2. Drop-Weight Impact Test

To analyze the temperature sensitivity of the mechanical behavior of STS304L under impact loading, a drop-weight impact test apparatus (Instron CEAST 9340, Instron, Pianezza, Italy) was used. The total impact weight was 32.546 kg, comprising the impactor weight (0.046 kg) and an additional weight (32.5 kg). The drop height was determined based on the required impact energy and total weight, with energy levels ranging from 50 J up to the point where cracks appeared. A schematic of the drop-weight impact test apparatus is illustrated in Figure 2a. The impact force signal was recorded at 500 Hz using a force sensor and data acquisition system. To ensure a single impact per test, an anti-rebound system was employed, preventing multiple strikes on the specimens.

Because insufficient impactor hardness can alter the mechanical response, the impactor was fabricated from high-hardness SKD11 steel and reinforced with hard chrome plating to further enhance hardness and reduce friction.

The impactor shape plays a crucial role in drop-weight impact tests. Previous studies have indicated that sharper impactors do not significantly contribute to energy absorption through strain-induced martensitic transformation, whereas rounded impactors enhance energy absorption owing to this transformation [19]. As this study focuses on the relationship between mechanical behavior and strain-induced martensitic transformation at different temperatures, a hemispherical impactor was selected, as depicted in Figure 2b. All tests were performed in triplicate, and the average values were used for analysis to ensure reliability and reproducibility.

2.3. Environmental Chamber and Temperature Control

To maintain stable testing conditions, the specimen was securely clamped using bolts and frames to minimize disturbances during the impact test, as shown in Figure 3. Additionally, low-thermal-conductivity glass wool was placed between the specimen and jig to minimize temperature rise due to heat conduction from the jig.

A custom-built environmental chamber was installed to maintain consistent low-temperature conditions. As illustrated in Figure 4, the chamber consisted of spray polyurethane foam, plywood, and expandable polystyrene, which provided insulation and prevented external heat ingress. Liquefied nitrogen was injected through an injection valve until the interior reached the desired test temperature. A digital control system regulated the nitrogen flow, while expandable polystyrene was used to seal the chamber and prevent nitrogen leakage. Cryogenic silicone (HARDEX 650 °F RED Gasket Maker, Hardex Corporation Sdn Bhd, Kuantan, Malaysia) was applied to seal the gaps between the jig and environmental chamber, further preventing nitrogen leakage.

The temperature–time history was recorded using a T-type thermocouple mounted behind the specimen to verify thermal stability. As shown in Figure 5, the time required to achieve thermal equilibrium was 1640 s at −100 °C and 2281 s at −193 °C. Once equilibrium was reached, the temperature remained stable throughout the test, confirming the effectiveness of the chamber in maintaining cryogenic conditions.

2.4. Test Procedure and EBSD Analysis

The test scenarios are listed in

Table 2. Summary of test scenarios.

Impact Weight [kg]	Temperature [°C]	Impact Energy [J]	Impact Velocity [m/s]
32.546	20	50	1.824
		70	2.159
		…	…
		Crack initiation energy	$\sqrt{2gh}$
	−100	50	1.824
		70	2.159
		…	…
		Crack initiation energy	$\sqrt{2gh}$
	−193	50	1.824
		70	2.159
		…	…
		Crack initiation energy	$\sqrt{2gh}$

. The impact energy was varied from 50 J up to the point where cracks appeared to evaluate the temperature sensitivity of STS304L’s mechanical behavior and energy absorption capacity under drop-weight impacts. The tests were conducted at room temperature (20 °C), −100 °C, and −193 °C. In this study, the drop weight was fixed, while the drop height was adjusted to control the impact velocity. The velocity at impact was calculated using the following free-fall equation:

$$v=\sqrt{2gh}$$

(1)

where $v$ is the impact velocity, $g$ is the gravitational acceleration (9.81 m/s²), and $h$ is the drop height. The test parameters, including the drop heights and corresponding velocities, are summarized in Table 2.

EBSD analysis was conducted to investigate the correlation between plastic deformation and strain-induced martensitic transformation at different temperatures by measuring the martensite fraction in the radial direction. These results were used to assess the influence of temperature on the mechanical behavior of STS304L.

To ensure proper alignment with the impact direction (thickness direction), the specimens were vertically hot-mounted. Surface preparation involved sequential grinding with SiC paper (400 to 2000 grit, 1 min per step), followed by fine polishing using diamond suspensions (3 μm and 1 μm) and colloidal silica (0.04 μm) to eliminate surface scratches. Prior to analysis, the specimens were ultrasonically cleaned in water and ethanol. EBSD measurements were performed using a TESCAN MIRA 3 LMH In-Beam Detector (TESCAN ORSAY HOLDING a.s., Brno, Czech Republic) under the following conditions: 20 kV accelerating voltage, ×500 magnification, and 1 μm step size.

3. Results and Discussion

3.1. Mechanical Behavior

In this study, the impactor was driven by gravity to exert a load on the specimen, and the resulting force was plotted against the displacement between the impactor and specimen. The displacement of the specimen was defined as the distance traveled by the impactor after contacting the upper surface of the specimen. The force–displacement relationship was measured throughout the impact energy absorption process to evaluate the mechanical behavior of the specimens.

When the impact load was first applied, a fluctuation region appeared. As the strain-hardening effect occurred, both the displacement and force increased linearly until they reached their peak values. Subsequently, necking was observed, where the displacement increased without a corresponding increase in force. The mechanical behavior of STS304L at all test temperatures exhibited similar trends, as depicted in Figure 6a–c.

The force–displacement history characterizes the dynamic response, including the peak force, absorption energy, and stiffness, as summarized in

Table 3. Summary of drop-weight impact test results at crack initiation.

Test Temperature [°C]	Crack Initiation Energy [J]	Peak Force [N]	Stiffness [N/mm]
20	110	19,137	2064.8
−100	175	27,561	2673.8
−193	185	32,607	3018.2

. The slope of the force–displacement curve, which represents the section where the displacement and force increase linearly, is steeper at lower temperatures. This indicates that a greater force was required to induce displacement, suggesting that stiffness increased with a decreasing temperature, enhancing resistance to deformation. Across all test conditions, increasing the impact energy led to a greater displacement and peak force. However, beyond a certain impact energy threshold, the force curve decreases sharply without a corresponding increase in displacement, accompanied by fluctuations due to the formation of a hole at the impact site. This indicates that the specimen lost its resistance to impact loading owing to crack initiation [20,21].

Figure 7 shows a schematic of the force–displacement history at cracking, illustrating the influence of temperature on the mechanical behavior. At 20 °C, no crack initiation was observed until 109 J, while at 110 J, cracks began to form. Similarly, the impact energy absorption capacity was 174 J at −100 °C and 184 J at −193 °C. This corresponds to a 59.63% increase at −100 °C and 68.80% increase at −193 °C compared with that at 20 °C, indicating that STS304L exhibited a negative temperature sensitivity in impact energy absorption.

As temperature decreases, dislocation mobility becomes limited, resulting in a higher dislocation density than that at higher temperatures [22,23]. Because greater free energy is required to move dislocations, more impact energy is necessary to achieve the same deformation owing to an increasing stiffness. Therefore, temperature plays a crucial role in determining the impact of energy absorption capacity.

3.2. Impact Energy Absorption Mechanism

The Charpy V-notch test has long been widely used to determine the ductile-to-brittle transition temperature (DBTT) as a criterion for evaluating the temperature sensitivity of energy absorption capacity. Previous studies revealed that austenitic stainless steel exhibited a 47.2% decrease in energy absorption capacity at −196 °C compared with that at 20 °C [24]. However, this result did not define a specific DBTT point. In contrast, this study found that the energy absorption capacity at −193 °C increased by 68.80% compared with that at 20 °C, primarily due to the specimen geometry, particularly thickness and notch effects, which influence stress states and energy absorption mechanisms. Moreover, defining a specific DBTT point in austenitic stainless steel is challenging because of its unique mechanical properties at cryogenic temperatures. These results indicate that austenitic stainless steels exhibit an excellent impact resistance in cryogenic environments.

Total energy can be classified as elastic deformation energy and permanent deformation energy. Permanent deformation energy can be further subdivided into plastic deformation energy, martensitic transformation energy, and heat dissipation energy. Therefore, the total impact energy absorption of the specimen can be represented as follows:

$$E_{total}=E_{el}+E_{pe}=E_{el}+(E_{pl}+E_T+E_H)$$

(2)

where $E_{el}$ is the elastic deformation energy of the specimen, $E_{pe}$ is the permanent deformation energy of the specimen, $E_{pl}$ is the plastic deformation energy of the specimen, $E_T$ is martensitic transformation energy of the specimen, and $E_H$ is the energy dissipated as heat.

According to previous studies, elastic deformation energy and heat dissipation are negligible compared with other energy absorption factors [25]. Additionally, strain-induced martensitic transformation consumes part of the plastic deformation energy [26], making plastic deformation energy the dominant contributor to the total impact energy absorption. The shape of the impactor also plays a key role in plastic deformation and energy absorption capacity in drop-weight impact tests.

The hemispherical impactor used in this study accelerates circumferential necking due to friction between the specimen and impactor surface [27]. The maximum friction effect occurs immediately before crack initiation, influencing the energy absorption mechanism. Therefore, further evaluation of plastic deformation in drop-weight impact tests is required to better understand its dominant role in total impact energy absorption and its dependence on the impactor shape and deformation mechanisms.

3.3. Plastic Deformation

In the drop-weight impact test, the specimen deformed because of multiple forces acting simultaneously, including a radial stretching force as the specimen moved with the impactor, a shear force in the thickness direction at the hemispherical radius point, and a global bending force applied across the entire specimen. These forces induced bulging and dishing deformations, as illustrated in Figure 8.

Upon impact, the impactor exerted a compressive force, generating a plastic wave that propagated radially, resulting in bulging deformation. When the radius of the bulging deformation matched that of the hemispherical impactor, dishing deformation occurred outside the contact area [28]. Figure 9 shows a specimen after impact, clearly displaying the bulging and dishing deformations caused by this mechanism.

Because the clamped region was fixed by bolts, deformation was limited to the free region of the specimen. The impact area, which was directly affected by the impactor, exhibited bulging deformation, whereas dishing deformation occurred in the surrounding area owing to bending forces. Figure 10a–c display deformation measurements under different conditions. Across all conditions, the maximum deformation occurred at the impactor–specimen contact region, and the deformation patterns remained consistent regardless of the test temperature. The maximum deformations recorded for STS304L were 10.89 mm at 20 °C, 11.02 mm at −100 °C, and 11.12 mm at −193 °C.

Figure 11 presents the relationship between the impact energy absorption capacity, plate deflection, and temperature. The difference in plate deflection across temperatures was insignificant, suggesting that cold embrittlement effects did not occur in STS304L at cryogenic temperatures. The increase in stiffness at lower temperatures contributed to a greater impact energy absorption. However, because the plastic deformation differences were minimal, deformation alone cannot fully explain the changes in energy absorption capacity.

A portion of the plastic deformation energy is consumed by strain-induced martensitic transformation, which exhibits temperature sensitivity [26]. Therefore, to fully understand the temperature sensitivity of impact energy absorption, it is necessary to analyze the fraction of transformed martensite.

3.4. EBSD Analysis: Volume Fractions of Strain-Induced Martensite

The EBSD analysis region is depicted in Figure 12, and the martensite fraction results at various temperatures are presented in Figure 13 and

Table 4. Results of EBSD analysis.

Temperature [°C]	Fraction	Region 1 [%]	Region 2 [%]	Region 3 [%]	Region 4 [%]
20	Martensite	13.94	19.26	16.99	23.78
	Austenite	86.06	80.74	83.01	76.22
−100	Martensite	76.29	77.85	46.54	24.61
	Austenite	23.71	22.15	53.46	75.39
−193	Martensite	89.56	96.87	60.65	26.97
	Austenite	10.44	3.13	39.35	73.03

. Strain-induced martensitic transformation under impact loading affects both energy absorption capacity and stiffness. Additionally, strain-induced martensite formation accelerates at lower temperatures [29]. Therefore, when evaluating the temperature sensitivity of STS304L, both strain-induced martensitic transformation and mechanical properties must be considered.

At 20 °C, the martensite fraction remained relatively constant across all analyzed regions. However, in the cryogenic environment (−100 °C and −193 °C), the martensite fraction increased significantly with deformation owing to the acceleration of strain-induced martensitic transformation. Region 2, located at the end of the hemispherical radius, exhibited the highest martensite fraction, as follows: 77.85% at −100 °C and 96.87% at −193 °C. This region experienced both axial deformation and significant thickness deformation because it was subjected to localized shear stress.

Except for Region 4, all regions exhibited the same trend—strain-induced martensitic transformation increased as temperature decreased. This aligns with previous studies, which showed that lower temperatures accelerate strain-induced martensitic transformation when deformation exceeds a critical threshold [29]. The higher dislocation density at lower temperatures plays a key role in this accelerated transformation.

Because strain-induced martensite is harder and stronger than austenite, it resists deformation when impact energy is transferred to STS304L, ultimately increasing stiffness and impact resistance.

Strain-induced martensitic transformation is also influenced by strain rate. In the drop-weight impact tests, the strain rate was not constant, because the impactor velocity decreased after contact. The average strain rate can be defined as follows:

$$\dot{\epsilon }=V/L$$

(3)

where $\dot{\epsilon }$ is the average strain rate, $V$ is the axial velocity upon impact, and $L$ is the axial deformation. The calculated average strain rates were 225/s at 20 °C, 256/s at −100 °C, and 262/s at −193 °C, all of which fell within the high strain rate range.

At high strain rates, adiabatic heating (localized temperature increase) is a critical factor affecting mechanical behavior [30]. In metastable austenitic stainless steels such as STS304L, the mechanical properties are governed by the austenite phase stability. If adiabatic heating suppresses strain-induced martensitic transformation, it can alter the material’s mechanical response.

A previous study concluded that at a strain rate of 200/s, adiabatic heating significantly reduced ductility by preventing strain-induced martensitic transformation [30]. Given that the strain rate in this study was higher than 200/s at 20 °C, the potential effects of adiabatic heating must be considered.

When a high strain rate load (~1000/s) is applied to metastable stainless steel, martensite forms rapidly at low strain levels. However, at high strain levels, adiabatic heating suppresses martensitic transformation. Despite similar deformation patterns across all test temperatures, at 20 °C, the martensite fraction remained nearly constant across all regions (1 to 4). This suggests that adiabatic heating occurred at 20 °C, preventing further strain-induced martensitic transformation at a strain rate of 225/s.

In this study, the lower impact energy absorption ability of STS304L at 20 °C can be attributed to the suppression of strain-induced martensitic transformation due to adiabatic heating. This means that the difference in the impact energy absorption ability across test temperatures was a result of reduced stored energy from phase transformation at 20 °C.

4. Conclusions

This study investigated the temperature sensitivity of the mechanical behavior of STS304L under impact loading using drop-weight impact tests. As a metastable austenitic stainless steel, STS304L exhibits a negative temperature sensitivity in terms of its stiffness and impact energy absorption capacity. To determine the cause of this phenomenon, deformation measurements and martensite fraction analysis were conducted. The key findings are summarized as follows.

• The stiffness of STS304L increased with a decreasing temperature, rising by 59.63% at −100 °C and 68.80% at −193 °C compared with that at 20 °C. The impact energy absorption capacity also increased, reaching 109 J at 20 °C, 174 J at −100 °C, and 184 J at −193 °C. The impact energy was stored in STS304L in the form of elastic deformation, plastic deformation, heat dissipation, and phase transformation (austenite to martensite).
• The deformation mechanism did not exhibit temperature sensitivity under hemispherical impact conditions. The measured plastic deformations were 10.89 mm at 20 °C, 11.02 mm at −100 °C, and 11.12 mm at −193 °C, showing no significant variation. However, the negative temperature sensitivity of the stiffness indicated an increase in stored energy within STS304L.
• Strain-induced martensitic transformation increased as the temperature decreased, as shown in the EBSD analysis. The martensite fraction increased progressively from Region 1 to Region 4, confirming that strain-induced martensitic transformation contributed to the negative temperature sensitivity of STS304L by increasing the stored energy in the form of phase transformation.
• Adiabatic heating due to high strain rates was the primary cause of reduced energy absorption at 20 °C. Localized temperature increases suppressed strain-induced martensitic transformation, leading to premature cracking owing to softening effects.
• The negative temperature sensitivity in stiffness and strain-induced martensitic transformation caused distinct differences in the mechanical behavior of STS304L between cryogenic and 20 °C conditions. Notably, adiabatic heating at 20 °C resulted in significantly altered mechanical behavior compared with that at cryogenic conditions.

This study provides insights into the relationship between temperature and mechanical behavior under large plastic deformation induced by a hemispherical impactor. However, the hemispherical impactor does not fully replicate the sloshing impacts experienced in actual applications. Therefore, further evaluation of impact resistance using various impactor shapes is necessary to accurately simulate sloshing and assess temperature sensitivity in real-world environments.