Evaluation test of structural materials used for liquid
hydrogen storage and transportation system
(Study on low temperature materials used in WE-NET 1)

Takao Horiya*, Norihisa Yamamoto, Tadashi Iida, Akio Yamamoto,
Hideji Okaguchi, Noriaki Yaegashi, Takenori Doko,
Masahiro Saito, Kiyoshi Yokogawa, Toshio Ogata
*Japan Research and Development Center for Metals
17 Mori Bldg. 6F, 1-26-5, Toranomon, Minato-ku, Tokyo 105, Japan
Executive members of Subtask 6 in the WE-NET project
17 Mori Bldg. 6F, 1-26-5, Toranomon, Minato-ku, Tokyo 105, Japan

ABSTRACT

For evaluation of low temperature embrittlement and hydrogen embrittement of structural materials used for liquid hydrogen storage tanks or tankers, three commercial materials, SUS304L, SUS316L (austenitic stainless steels) and A5083 (aluminum alloy), were selected. Evaluation test, such as tensile test, Charpy impact test, fracture toughness test and fatigue test were conducted both at liquid He temperature (4K), instead of liquid hydrogen temperature (20K), and at room temperature. Welded joints of each plates were made by TIG welding for the stainless steels and MIG welding for the aluminum alloy and test specimens were taken from base metal and weld metal and heat-affected zone of the each plates. Some of specimens were hydrogen charged in the autoclave before the test for evaluating susceptibility to hydrogen embrittlement.
Test results indicated that the weld metals were most susceptible to both low temperature and hydrogen embrittlement, irrespective of the materials. The values of tensile elasticity, Charpy absorbed energy and fracture toughness for each of the materials were always the lowest in the weld metal at 4K after the hydrogen charging.


1. INTRODUCTION

For assuring the safety of large structures in storage and transportation system of liquid hydrogen (LH2), it is of great importance to understand characteristics of low temperature embrittlement and hydrogen embrittlement of structural materials in liquid hydrogen circumstances. However, there have been few materials data of structural materials at cryogenic temperatures, especially fracture toughness and fatigue properties of weld metals below liquid hydrogen temperature (20K). For determining suitable structural materials used for liquid hydrogen storage and transportation system, it is necessary at first to know low temperature properties and hydrogen embrittlement behavior at service condition. Purpose of this investigation is to evaluate low temperature embrittlement and hydrogen embrittlement of commercial structural materials at cryogenic temperature. In this paper, outline of the evaluation test results conducted by Subtask 6 of the WE-NET project was shown and detail of the results, particularly microstructural examination and fracture surface observation of the test specimens were described elsewhere.

2. Experimental method

Three commercial materials were selected for the evaluation test. The materials were two austenitic stainless steels, SUS304L (S304L) and SUS316L (S316L), which have been already used at liquid helium temperature(4K), and non-heat-treatable aluminum alloy, A5083-0, which have been used as structural materials in LNG tankers. Welded joints of each materials were made and test specimens were taken form base metal, weld metal and heat-affected-zone (HAZ) of the welded joints. Some of specimen were hydrogen charged before the testing for evaluating susceptibility to hydrogen embrittlement. In addition, tensile and fatigue test in high pressure hydrogen gas atmosphere were conducted at room temperature (RT).
25mm-thick plates of the three materials were used. Chemical compositions of the plate were shown Table 1. Welded joints were made by Tungsten Inert-Gas arc welding (TIG) for the two stainless steels and Metal Inert-Gas arc welding (MIG) for the aluminum alloy. Hydrogen charging were performed in the autoclave under the conditions of 400degree 9.8MPa for the stainless steels and 150degree 9.8MPa for the aluminum alloy. Hydrogen contents after the charging were about 30ppm for the stainless steels and 0.08ppm for the aluminum alloy. Tensile, charpy impact, fracture toughness tests were conducted at 4K, instead of 20K, and RT. High cycle fatigue test was also conducted at RT both in the air and in the high pressure (1.1MPa and 19.7MPa) hydrogen gas.

3.TEST RESULTS

3.1 Before the charging

Table 2 shows the summary of change in values obtained when the position of test specimens varied from base metal to weld metal or the testing temperature varied from RT to 4K.

(1) Tensile property
Tensile strength at 4K for the base metals of S304L and S316L were larger than those at RT, but their elongation and reduction of area of the weld metal at 4K were decreased fairly. Judging from the fact that the lowest value of the elongation, that of the weld metal of S316L at 4K, is still 27% and the morphology of fracture surfaces were completely elastic, it is considered that tensile properties of S304L and S316L were enough for cryogenic service. In A5083 alloy, tensile strength of the weld metal was equal or slightly lower than those of base metal. At 4K, elongation of the weld metal decreased to about 60% of that of the base metal.

(2) Charpy impact property
Absorbed energy at 4K for all of the materials was reduced significantly, regardless of specimen position. The most embrittled position of the specimen at 4K was weld metal. In S316L, absorbed energy at 4K of the weld metal reduced about 85% of that at RT(Fig.1). It is considered that there was a close relationship between low temperature embrittlement of the weld metal and formation of d-ferrite phase in the weld metal, because at 4K cracks often propagated along the d-ferrite phase or the d-ferrite / austenite phase boundaries. In A5083, the impact energy of base metal at 4K was 1/2 of that at RT. The absorbed energy at RT of the weld metal was about 1/2 of that of the base metal. At 4K the absorbed energy of the weld metal was 1/4 of that at RT.

(3) Fracture toughness
All of test specimens of the materials but the weld metal of S316L showed ductile fracture even at 4K. KIC converted from JIC (values of crack initiation for ductile fracture), of S304L and S316L was rather high at RT, but KIC of the weld metal at 4K for the stainless steels were reduced greatly and their value were 132MPaăm for S304L and 110Mpaăm for S316L. Decrease of KIC of the weld metal was attributed to presence of d-ferrite phase in the weld metal as in Charpy impact test. At 4K, cracks were found to propagate preferential along the boundaries of d-ferrite phase of the weld metal. In A5083 alloy, KIC of the base metal did not decrease so much at 4K, but KIC of the weld metal at 4K was reduced to about 40% of that at RT (Fig.2).

3.2 After hydrogen charging

Table 3 shows the summary of evaluation on susceptibility to hydrogen embrittlement by comparing testing values of the specimens after the hydrogen charging with those before the charging.

(1) Tensile Property
In the stainless steels, S304L had a little higher susceptibility to hydrogen embrittlement than S316L and showed the large reduction of elasticity for the weld metal at 4K. The quasi-cleavage fracture surfaces were observed partly on the fracture surfaces at 4K of weld metal in SUS304L. In A5083, difference in tensile property by the hydrogen charging was relatively small both at RT and at 4K.

(2) Charpy impact property
In S304L, the weld metal at 4K was most embrittled and its absorbed energy was about 34% lower than that of the uncharged one. In S316L, the absorbed energy of weld metal was reduced by about 50% after the hydrogen charging both at RT and 4K. In A5083, the absorbed energy at RT of base metal and weld metal change little after the charging and the effect of the hydrogen charging on the absorbed energy was not clear at 4K.

(3) Fracture toughness
In S304L, KIC (J) of the weld metal decreased more than 20% both at RT and 4K after the hydrogen charging. And KIC of the weld metal at 4K after the charging was about 100MPaăm. In S316L, KIC did not change significantly for the base metal both at RT and at 4K. However, the hydrogen charging reduced KIC of the weld metal at RT as well as at 4K. d-ferrite / austenite phase boundaries in the weld metal were found to become the preferential path of the cracks by the hydrogen charging and this was assumed to lead to the large reduction of KIC of the weld metal. In A5083, KIC values before the hydrogen charging were almost the same as that after the charging, indicating that influence of hydrogen charging on KIC was not evident both at RT and at 4K.

(4) Fatigue property
Only in S304L, hydrogen charging resulted in a decrease of fatigue life. Although in S316L, difference in the morphology of fracture surface of the fractured specimens were observed after the hydrogen charging, there ware no evident change in high cycle fatigue life. In A5083, no clear change in the fatigue life at RT by the hydrogen charging was observed.

3.3 Tensile and fatigue properties in high pressure hydrogen gas.

(1) Tensile property
In S304L and S316L, hydrogen reduced the elongation and the reduction of area. In both steels, hydrogen also showed a marked effect on the fracture surface. The quasi-cleavage fracture was observed on the austenitic matrix and cleavage fracture was also observed on d-ferrite phase in the fractured specimens. The susceptibility to hydrogen embrittlement of the base metal was a little larger than that of the weld metal. In A5083, hydrogen showed little effect on the tensile properties both of the base metal and the weld metal.

(2) Fatigue property
In S304L, hydrogen reduced the fatigue life of the specimens for the base metal and weld metal. Hydrogen also showed a evident effect on the morphology of the fracture surface in the fatigue test. In S316L and A5083, hydrogen showed little effect on the number of cycles to failure for the base metal and the weld metal.

4. Conclusions

The evaluation tests were conducted at 4K and RT for evaluating susceptibility both to low temperature embrittlement and to hydrogen embrittlement of S304L, S316L and A5083. Test result was summarized as follows: 1) All of the materials used have a susceptibility to both low temperature and hydrogen embrittlement in tensile elasticity, Charpy absorbed energy and fracture toughness. Specimens of the weld metal at 4K showed the most embrittled values, regardless of the materials. 2) By the hydrogen charging, Charpy absorbed energy and fracture toughness for the weld metal of the two stainless steels were remarkably decreased especially at 4K. 3) In the stainless steels, presence of d-ferrite phase in the weld metal was considered to be attributed to the large embrittlement of the weld metal.

Acknowledgement

This study was conducted as a part of research activities of Subtask 6 in WE-NET project which is supported by NEDO (New Energy and Industrial Technology Development Organization) and MITI (Ministry of International Trade and Industry) in Japan.