A Study on the Effect of Hydrogen Embrittlement and Temperature of Welds in Steel Applied to Liquid Hydrogen Storage Systems

Tae-Wook Kim; Dong-Ha Lee; Young-Hwan Choi; Seok-Min Kim; Jeong-Hyeon Kim; Jae-Myung Lee

doi:10.5781/JWJ.2025.43.3.4

Abstract

The production of a system that stores and transports liquid hydrogen necessarily includes welds, and a large amount of heat is introduced through welding, which causes changes in material properties. At this time, there is a lack of research on the effect of the hydrogen environment on the welds. In particular, it is difficult to find a case that analyzed the effect up to the cryogenic temperature, which is the environment for storing liquid hydrogen. In this paper, among the steel candidates to be applied to liquid hydrogen storage containers, the effect of hydrogen embrittlement was analyzed for STS304L and STS316L, which are stainless steels that are currently used for liquefied natural gas storage containers due to their excellent cryogenic properties. Accordingly, in this study, hydrogen was injected into the welds of STS304L and STS316L, and the material behavior at room temperature and cryogenic temperatures was evaluated using a universal material testing machine, thereby comparing and verifying the applicability of each steel to liquid hydrogen storage containers. In all scenarios, the strength of the hydrogen charging specimens was lower than that of the non-hydrogen charging specimens, and the elongation was also lower in the hydrogen charing specimens than in the non-hydrogen charging specimens, except for two cases.

Key words: Liquid hydrogen, Stainless steel, Welds, Effects of hydrogen embrittlement, Effects of temperature, Evaluation of material behavior

1. Introduction

Hydrogen, characterized by an exceptionally high energy density per unit mass and producing only water upon combustion, is widely recognized as the ultimate clean energy source and a promising next-generation energy carrier¹⁾. However, due to its status as the lightest element, hydrogen has a relatively low volumetric energy density, which significantly affects its utilization efficiency depending on the storage method. Typically, hydrogen can be stored in gaseous, liquid, or solid (metal hydride) forms, each presenting distinct advantages and limitations in terms of cost, safety, and ease of handling²⁾.

Among these, liquid hydrogen offers superior volumetric energy density compared to other storage forms. Liquefaction of gaseous hydrogen at -253°C reduces its volume by approximately 800 times under atmospheric pressure, thereby enabling higher-density hydrogen storage than both compressed and solid forms³⁾. Owing to these advantages, liquid hydrogen has emerged as a key component in both stationary storage systems and hydrogen-powered mobility and transport infrastructure. This has elevated the importance of liquid hydrogen storage tanks capable of ensuring safe and stable storage under cryogenic conditions.

Designing cryogenic storage vessels requires careful consideration of low-temperature embrittlement in the steels used. In addition, hydrogen embrittlement (HE) must also be accounted for in systems handling hydrogen gas. HE is a phenomenon in which hydrogen atoms permeating the metal matrix significantly degrade its toughness and ductility, potentially leading to hydrogen delayed fracture. Several mechanisms have been proposed to explain HE, including hydride-induced embrittlement (HIE)⁴⁾, hydrogen-enhanced decohesion (HED)⁵⁾, and hydrogen-enhanced localized plasticity (HELP)⁶⁾. Nevertheless, because actual hydrogen service environments involve complex interactions among numerous variables, no single mechanism can fully explain the phenomenon. As such, hydrogen embrittlement remains a topic of ongoing investigation from diverse theoretical perspectives, with no unified theory established to date.

For hydrogen to be broadly adopted in industrial applications, ensuring the safety and reliability of its storage and transport infrastructure is paramount. This necessitates selecting materials with high resistance to hydrogen embrittlement and ensuring the quality of weld joints. International standards and guidelines for hydrogen storage and transport-such as ANSI/AIAA G-095 (Guide to Safety of Hydrogen and Hydrogen Systems), CGA G-5.6 (Hydrogen Pipeline Systems), and ASME B31.12 (Hydrogen Piping and Pipelines)-recommend the use of austenitic 300-series stainless steels and aluminum alloys, while the use of carbon steels and titanium alloys is restricted due to their susceptibility to hydrogen embrittlement⁷^-⁹⁾.

Although various studies have explored hydrogen embrittlement in steels under cryogenic conditions¹⁰^-¹³⁾, real-world structures are typically fabricated through welding, making it essential to investigate the embrittlement characteristics of welded joints. Despite this, comprehensive studies on hydrogen embrittlement of welds under cryogenic environments remain limited. Therefore, the present study aims to elucidate the effects of hydrogen embrittlement on the welded joints of austenitic 300-series stainless steel in cryogenic environments. The focus is on assessing the mechanical degradation of welds produced using the gas tungsten arc welding (GTAW) process, which is widely employed in cryogenic liquid hydrogen systems that require joints between high- and low-temperature zones. Mechanical performance tests were conducted according to ASTM E8 - Standard Test Methods for Tension Testing of Metallic Materials¹⁴⁾, and hydrogen was introduced into the steel via an electrochemical charging method to quantify the reduction in elongation caused by hydrogen. This method, which generates hydrogen through the electrolysis of an electrolyte solution and diffuses it into the metal, is considered safer and more controllable than using high-pressure hydrogen environments and is therefore widely adopted¹⁵⁾. Additionally, tensile tests were carried out under varying strain rate conditions to analyze the fracture behavior of specimens with respect to deformation rate.

2. Materials

2.1 Materials

During the fabrication of liquid hydrogen storage vessel structures, the welding process introduces significant heat input, which can lead to material deformation. Accordingly, it is essential to assess the susceptibility of weld zones to hydrogen embrittlement. In particular, there is a lack of studies that have investigated this susceptibility under cryogenic conditions typical of liquid hydrogen storage environments, highlighting the need for further research into the low-temperature properties of welded joints. Among candidate materials for liquid hydrogen storage tanks, STS316L stainless steel has been traditionally used for the inner vessel due to its excellent resistance to hydrogen embrittlement and its ability to maintain high ductility and strength at low temperatures. Numerous standards in countries such as the United States and Japan also recommend the use of STS316L. Additionally, to reduce material costs, recent studies-including those by Kawasaki Heavy Industries- have explored the use of STS304L, which is gaining traction as a viable alternative. Based on these trends, STS304L and STS316L were selected in this study as candidate structural steels for liquid hydrogen storage tanks¹⁶⁾.

To evaluate the effects of hydrogen embrittlement and temperature on welded joints, tensile specimens incorporating welds were fabricated using STS304L and STS316L plates. The welds were produced using an automated gas tungsten arc welding (GTAW) system. The welding parameters used in the hydrogen embrittlement test specimens are provided in Table 1.

Table 1

Welding conditions for hydrogen charing specimens

Item	Contents
Welding robot	YASKAWA HP-20
Welding machine	WORLDWEL longrun 1000PT3
Base metal	STS 304L, STS 316L
Filler metal	STS 308L
Filler cross section area (mm²)	1.13
Filler density (g/cm²)	8
Shield gas	Ar 93% + H27%, 20ℓ/min
Electrode angle	Travel 10o, Work 0^o
Feeding angle	20^o ~ 40^o
Arc length (mm)	5
Deposition area (mm²)	3.0
Deposition rate (kg/h)	1.13
Feeding rate (cm/min)	208.3

3. Experimental Methods

3.1 Experimental Methods

Hydrogen charging of the welded steel specimens was performed using the cathodic electrolysis method in accordance with ISO 16573, thereby simulating a hydrogen embrittlement environment. The hydrogen charging system setup is shown in Fig. 1, and specimens were charged with hydrogen at a current density of 10 mA/cm^{2 17)}. Tensile specimens of the welded steels were prepared according to KS B 0801, and the dimensions and photographs of the specimens are presented in Fig. 2,¹⁸⁾.

Fig. 1

Cathode electrolysis hydrogen charging system

Fig. 2

Specifications and photos of tensile test specimens of steel welds

Mechanical property evaluations were conducted using a 500 kN universal testing machine (UTM). A custom-designed tensile jig was employed to facilitate secure and precise testing of the weld specimens. A cryogenic chamber was used for low-temperature testing. Liquid nitrogen gas was injected into the chamber, and temperature control was maintained using an automated system. To ensure the thermal equilibrium of the specimens, a pre-cooling period of 60 minutes was applied prior to each test. The overall configuration of the experimental setup is illustrated in Fig. 3.

Fig. 3

Configuration of a test system for performance evaluation of steel welds

3.2 Test Scenarios

To evaluate hydrogen embrittlement susceptibility and characterize the cryogenic performance of welded joints based on steel weld properties, quasi-static tensile tests were conducted in accordance with ASTM E8.¹⁴⁾ Tests were performed at strain rates of 10^-1/s and 10^-3/s, and at three different temperatures: 20°C, -70°C, and -196°C, thereby enabling the construction of a comprehensive performance database for each material.

The complete test matrix used in this study is summarized in Table 2. To ensure the reliability of results, five replicate tests were conducted for each case, and the data point closest to the mean value was selected for comparative analysis.

Table 2

Scenario for hydrogen embrittlement testing of steel welds

Specimen	Temperature(°C)	Strain rate(/s)	Hydrogen charging
STS304L	20	0.1	O
		0.1	X
		0.001	O
		0.001	X
	-70	0.1	O
		0.1	X
		0.001	O
		0.001	X
	-196	0.1	O
		0.1	X
		0.001	O
		0.001	X
STS316L	20	0.1	O
		0.1	X
		0.001	O
		0.001	X
	-70	0.1	O
		0.1	X
		0.001	O
		0.001	X
	-196	0.1	O
		0.1	X
		0.001	O
		0.001	X

4. Results and Discussion

4.1 Results and Discussion

Figs. 4-6 present the stress-strain curves analyzing the effects of hydrogen embrittlement on the weld zones of STS304L and STS316L, which are candidate materials for use in liquid hydrogen storage tanks, under varying temperatures and strain rates. Across all test conditions, the specimens charged with hydrogen exhibited lower strength than those without hydrogen charging. Similarly, except for two cases, the elongation of hydrogencharged specimens was also lower compared to the uncharged counterparts. These results confirm that hydrogen embrittlement significantly reduces both the strength and ductility of the welded regions of the tested materials.

Fig. 4

Stress-strain curve according to strain rate considering the effect of hydrogen embrittlement at 20°C

Fig. 5

Stress-strain curve according to strain rate considering the effect of hydrogen embrittlement at -70°C

Fig. 6

Stress-strain curve according to strain rate considering the effect of hydrogen embrittlement at -196°C

When comparing the two materials, STS304L generally exhibited higher strength than STS316L within the same test group, regardless of the presence or absence of hydrogen embrittlement. In terms of elongation, STS316L tended to show greater ductility when hydrogen was not present. However, in hydrogencharged conditions, this trend was inconsistent, suggesting that hydrogen embrittlement has a more pronounced and less predictable impact on ductility than on strength. Notably, this impact is not only quantitative but also qualitatively more significant in elongation than in strength reduction.

Temperature-wise, the typical trend observed in structural steels was reaffirmed: strength increased as temperature decreased. For tests conducted at a strain rate of 10^-3/s, elongation decreased with decreasing temperature, regardless of hydrogen charging. A similar trend was observed at 10^-1/s, but an anomalous behavior was noted in STS316L with hydrogen charging, where the elongation at 20°C was lower than at the colder temperatures. This deviation will be further investigated through fractographic analysis and additional follow-up experiments.

A numerical comparison of the strength values between hydrogen-charged and uncharged specimens across all test cases is summarized in Table 3. Under a strain rate of 10^-1/s, STS316L exhibited a smaller strength reduction ratio due to hydrogen embrittlement than STS304L across all temperatures. Conversely, under a strain rate of 10^-3/s, STS316L showed a greater reduction in strength compared to STS304L. These results suggest that when subjected to relatively fast loading conditions, STS316L may allow for more flexible design margins than STS304L. Therefore, the choice of structural material may need to vary depending on the expected strain rate and loading characteristics of the component in actual hydrogen systems.

Table 3

Comparison of strength ratios considering the effect of hydrogen embrittlement of STS304L and STS316L

Specimen	Strain rate(/s)	Temperature-dependent strength ratio(hydrogen charging O/X, %)
Specimen	Strain rate(/s)	20°C	-70°C	-196°C
STS304L	0.1	90.6	82.8	89.5
STS316L	0.1	91.2	94.6	95.3
STS304L	0.001	98.9	98.0	99.9
STS316L	0.001	98.0	95.4	94.5

5. Conclusion

In this study, STS304L and STS316L welded joints were subjected to hydrogen charging, and their mechanical behavior was evaluated under ambient and cryogenic conditions using a universal testing machine. This allowed for a comparative assessment of the applicability of these steels for use in liquid hydrogen storage vessels. The key findings of the study are summarized as follows:

1) In all test scenarios, the hydrogen-charged specimens exhibited lower tensile strength compared to their uncharged counterparts. Similarly, elongation was also lower in the hydrogen-charged specimens, except in two test cases.
2) In material-to-material comparisons, STS316L consistently showed greater elongation than STS304L in the absence of hydrogen embrittlement. However, in the presence of hydrogen, this trend became less consistent. Additionally, consistent with general trends observed in structural steels, tensile strength increased as temperature decreased, even in welded specimens.
3) At a strain rate of 10^-1/s, STS316L exhibited a smaller reduction in strength due to hydrogen embrittlement compared to STS304L across all temperature conditions. Conversely, at a strain rate of 10^-3/s, STS316L showed a larger strength reduction than STS304L.

These findings indicate that STS316L, when subjected to rapid strain rates, may allow for more flexible structural design compared to STS304L. Consequently, the choice of steel material may need to be optimized based on the strain rate and loading conditions expected in hydrogen-related infrastructure. The results of this study are expected to contribute to the safe and reliable structural design of components in the liquid hydrogen industry.