2021 年 61 巻 4 号 p. 1333-1336
Components for hydrogen vessels and accessories in fuel cell vehicles (FCVs) are exposed to high pressure hydrogen gas environments. A limited range of materials that do not exhibit hydrogen embrittlement in such environments have been permitted for use as component materials. However, expansion of the range of usable materials is required for the widespread commercialization of FCVs. National projects have thus been promoted in the automobile field to expand the usable materials for high-pressure hydrogen environments. In these projects, the establishment of test methods to accurately evaluate the hydrogen compatibility of materials was one of the primary aims. The second aim was to standardize materials selection methods for high pressure hydrogen gas services. Two different test methods are proposed to evaluate the hydrogen compatibility of materials for FCVs. One is a methodology based on the notched specimen fatigue life test and the other is based on the slow strain rate technique (SSRT) and smooth specimen fatigue life test. These test methods are now under deliberation to be included in the Global Technical Regulation (HFCV-gtr) phase 2.
Hydrogen energy is attracting attention for the realization of a low-carbon society. 70 MPa-class fuel cell vehicles (FCVs) have already been commercialized in several countries. However, some load-bearing components in such systems are exposed to high pressure hydrogen gas environments. Only high Ni equivalent SUS316L and 6061-T6 aluminum alloy have been permitted for use in such components in Japan because of their good resistance to hydrogen embrittlement (HE). For such materials, the reduction of area (RA) measured from tensile tests multiplied by the relative reduction area (RRA) in the slow strain rate technique (SSRT) test must exceed the requirement by the material standard for austenitic stainless steels.1) In the General Ordinance, the materials selection method based on RRA is specified according to the proposed use. The RRA of austenitic stainless steels correlates with the Ni equivalent: therefore, the Japanese regulation allows the use of SUS316L with a Ni equivalent of >26.3%.
The global accessories in FCVs are designed based on the performance requirements (design by test). In the discussion of the Global Technical Regulation No. 13 (HFCV-gtr Phase 1), although the standard methods of material validation were examined, agreement among countries regarding the methods was not reached and leaving their determination to the HFCV-gtr Phase 2 was the preferred course of action. Figure 1 shows a conceptual flowchart for the evaluation of hydrogen compatibility of metallic material used for components of FCVs. It is desirable that as a performance requirement, the hydrogen compatibility of materials is evaluated by testing. In such a situation, under collaboration with relevant countries, research has been conducted on the standardization of hydrogen compatibility assessment of materials. National projects in Japan were conducted from 2013 to 2017 and 2018 to 2020 to achieve global harmonization of hydrogen compatibility testing in high pressure gaseous hydrogen. SSRT properties and fatigue life properties of austenitic stainless steels have been investigated in high pressure hydrogen gas and the effects of the test conditions on the hydrogen compatibility of the materials was clarified.
Rules for hydrogen compatibility of metallic materials. (Online version in color.)
This paper describes the concept of the hydrogen compatibility test, recommended test conditions and proposed performance based on test metrics consistent with the requirements of FCVs.
Only high Ni equivalent SUS316L and 6061-T6 aluminum alloy that do not exhibit HE in high pressure hydrogen gas environments are used for the components of high pressure hydrogen tanks for FCVs, whereas widely used austenitic stainless steels such as SUS304 are not permitted. However, an expansion of the range of usable materials is expected to reduce the cost of FCVs. For on-board accessories sufficient fatigue life with in-service-inspection is required over a long period of time. Therefore, the design stress is restricted below the fatigue limit and the components are designed based on infinite-life design. It is well-known that the tensile ductility and fatigue crack growth (FCG) rate of many metallic materials are degraded by hydrogen. The degree of degradation is significantly affected by the environmental conditions such as hydrogen gas pressure and temperature: therefore, it is difficult to capture the worst case of degradation. However, the fatigue limits of such materials are not significantly reduced by hydrogen, which suggests that under infinite-life design, various metallic materials with lower cost may be eligible for use as the components of FCVs without compromising safety and reliability. Figure 2 shows a schematic illustration of the hydrogen compatibility test concept. The maximum stress amplitude of the components is lower than the fatigue limit.
Schematic illustration of the hydrogen compatibility test concept for the components materials of FCVs.
The SSRT test is conducted firstly to verify that the material satisfies the criteria for tensile properties in a hydrogen environment. The fatigue life test is then conducted in a hydrogen environment at the stress amplitude equivalent to the fatigue limit in a reference gas environment. If the fatigue limit in hydrogen environment is equal to or greater than that in the reference environment, the material is judged to have hydrogen compatibility.
2.2. Test Conditions 2.2.1. Hydrogen GasThe maximum fueling pressure of a hydrogen tank is expected to not exceed 125% of the nominal working pressure (NWP), which is typically 70 MPa for vehicle storage. Therefore, the hydrogen pressure is set at greater than 87.5 MPa (125% NWP).
The purity of hydrogen should be appropriately controlled because it can have a significant influence on the test results. For example, it has been reported that a small amount of oxygen contained in hydrogen gas can retard the FCG rate in steels.2,3,4) The use of hydrogen gas that complies with CGA G3 Grade L (5N)5) is recommended in accordance with the Society of Automobile Engineers (SAE) standard. CGA G3 Grade L (5N) hydrogen allows impurities total of 10 mass ppm, but limits the amount of oxygen to 1 mass ppm or less.
The purity of hydrogen in the hydrogen gas chamber of the testing machine can be decreased due to the contaminants that adhere to the inner surface of the chamber, tubing and valves. Therefore, the hydrogen purity in the chamber is occasionally measured after testing. As a result, it has been confirmed that the purity of exhaust hydrogen gas was in accordance with the requirement by the standards such as ANSI/CSA CHMC-1(2014).6,7)
2.2.2. Test TemperatureThe effect of test conditions on HE in a hydrogen gas environment was investigated using verification tests with commercial austenitic stainless steels. The chemical composition of the tested materials is shown in Table 1.
| (mass%) | ||||||||
|---|---|---|---|---|---|---|---|---|
| C | Si | Mn | P | S | Ni | Cr | Mo | |
| SUS304 | 0.06 | 0.40 | 0.83 | 0.028 | 0.004 | 8.09 | 18.14 | – |
| SUS316 | 0.05 | 0.61 | 0.84 | 0.029 | 0.001 | 10.05 | 16.10 | 2.08 |
| SUS316L | 0.016 | 0.57 | 0.83 | 0.030 | 0.001 | 12.10 | 17.66 | 2.07 |
Figure 3 shows the elongation as a function of the test temperature for SUS304, SUS316 and SUS316L obtained by SSRT testing in 105 MPa hydrogen.8) Geometry of the SSRT test specimen is based on ASTM E89) with a diameter of 6 mm and a gauge length of 30 mm. The SSRT tests were performed at a strain rate of 5 × 10−5 s−1. Whereas strain-induced martensitic transformation (SIMT) in austenitic stainless steels is generally accelerated at low temperature, the hydrogen diffusion rate decreases with the temperature. As a result of the interaction between SIMT and hydrogen diffusion, the temperature of maximum hydrogen degradation (TMHD) is determined by the SSRT test. It has been reported that high Ni materials with stable austenitic microstructure exhibit low TMHD.1,10) Figure 3 shows that the lowest elongation for SUS304 was observed at 228 K. On the other hand, that for SUS316L is observed at 198 K; however, compared with SUS304 and SUS316, its value of 60% was still high. The temperature of the SSRT test was thus selected to be 228 K which was the TMHD for metastable austenitic stainless steel such as SUS304.
Effect of temperature on the elongation of austenitic stainless steels in a high pressure hydrogen gas environment.
Figure 4 shows an example of fatigue life test result for SUS304 in a H2 pressure of 115 MPa at 293, 228 and 193 K under a stress ratio of −1 and a test frequency of 1 Hz.8) The round bar specimens with a diameter of 6 mm were used for the fatigue life test. The fatigue limit in a hydrogen gas environment is determined from fatigue life test at 1 × 106 cycles. The knee point of the S-N curve for SUS304 tested in air was observed at 1.5 × 105 cycles therefore, this method for determination of the fatigue limit in hydrogen gas seems to be acceptable.
Fatigue life test results for SUS304 in a high pressure hydrogen gas environment.
The low cycle fatigue life in a hydrogen gas environment is obviously shorter than that in the air, however, the fatigue limit in a hydrogen gas environment is not degraded by hydrogen, but is the same as that in air at room temperature (293 K). The specimen did not fail at low temperatures of 228 K and 193 K, even though the applied stress was 35 MPa higher than the fatigue limit at room temperature. In the fatigue life test, the amount of SIMT was much less than that in the SSRT testing, because the applied stress was sufficiently lower than the yield strength. It is also well-known that the yield strength of austenitic stainless steels is increased with a decrease in temperature within the low temperature range. This temperature dependency of the fatigue limit was confirmed to also be consistent for the 300-series austenitic stainless steels in hydrogen gas environments. Therefore, the test temperature for fatigue life tests was set at 293 K, at which the steels exhibit a lower fatigue limit, compared with that at lower temperatures.
2.3. SSRT TestThe objective of the SSRT test is to verify whether the tensile and ductile properties of materials satisfy the minimum requirements in a worst case environment. In Japan, smooth round-bar specimens are used for the SSRT test conducted in hydrogen gas environments. The geometry and machining procedure of the SSRT test specimen are described in ASTM G142 or ASTM E8.9,11) The specimen diameter ranges from 3 mm to 8 mm, in principle.12,13) The specimens shall have been manufactured with a minimum involvement of cold working.
An appropriate strain rate should be selected for the SSRT testing, because it can have a significant effect on the SSRT properties (e.g. ductility) in high pressure hydrogen gas environments. Some researchers investigated the effect of strain rate on the SSRT properties in a hydrogen gas environment and reported that there was no significant difference in the RA within the strain rate range of 10−4 to 3 × 10−6 s−1.14,15) The ASTM standard G142 recommends a nominal strain rate of 7.0 × 10−5 s−1 for SSRT testing in a hydrogen gas environment.11) Taking these experimental results and standards into consideration, a strain rate of 5.0 × 10−5 s−1 or lower was selected for the hydrogen compatibility test.
2.4. Fatigue Life TestThe main objective of the fatigue life test is to verify that the fatigue limit is not degraded in hydrogen gas compared with that in the reference gas (e.g. air). Load-controlled tension-compression fatigue tests of a smooth round-bar specimen were performed with a sinusoidal waveform at a stress ratio of −1. In the case of a notched specimen, a stress ratio of 0.1 was selected. The conditions for machining and surface finishing have a significant influence on fatigue strength. Therefore, to minimize these effects, the machining and surface finishing methods followed ASTM E466.16)
The hydrogen compatibility test method has been discussed by the material expert team in the SAE Fuel Cell Safety Task Force. Two different test methods have been proposed by this team. Option 1 is the notched specimen methodology and Option 2 is a methodology based on the SSRT and smooth specimen fatigue. Figure 5 shows the flow of the hydrogen compatibility test method. If one of two options is satisfactory, then the material is authorized to use in the hydrogen tank components of FCVs.
Flowchart for evaluation of the hydrogen compatibility of materials.
Force-controlled fatigue life tests shall be performed at a test frequency of 1 Hz in a high pressure hydrogen gas environment at 293 ± 5 K in Option 1. The circumferentially notched specimens shall have an elastic concentration factor (Kt) of greater than or equal to 3. A minimum of three tests shall be conducted in the high pressure hydrogen gas environment. The stress ratio of the notched fatigue life test shall be 0.1. The maximum net stress shall be greater than or equal to 1/3 of the tensile strength measured in air at room temperature. The minimum number of fatigue cycles shall be greater than 1 × 105. The SSRT test is not required in Option 1.
Option 2 requires both SSRT and fatigue life tests. A minimum of three SSRT tests shall be conducted in a high pressure hydrogen gas environment at 228 ± 5 K. For comparison, a minimum of three SSRT tests shall be conducted in air at the same temperature. The strain rate shall be 5 × 10−5 s−1 or slower. The yield point in the hydrogen environment shall be greater than or equal to that measured in air at the same temperature. However, there is a variation of the test results in high pressure hydrogen gas environments caused by the testing equipment and the machined position of specimens, e.g., sliding friction at the sealing position, and effect of hydrogen on the inner load cell. An error of 20% at maximum was observed in the yield strength from the austenitic stainless test data in the hydrogen environment and in air environment.17) Therefore, the criterion for the yield strength in hydrogen gas environments shall be greater than 80% of the yield strength (YS) measured in the air at the same temperature. The materials should also practically deform to some extent beyond the yield point. This means that no brittle fracture should occur before reaching the yield point.
Force-controlled fatigue life tests of smooth round-bar specimens shall be performed at a test frequency of 1 Hz in the high pressure hydrogen gas environment at 293 ± 5 K. The stress ratio of the smooth fatigue life test shall be −1. The maximum stress amplitude shall be greater than or equal to 1/3 of the tensile strength measured in air at room temperature. For annealed austenitic stainless steels, this stress amplitude corresponds to the lower bound of the fatigue limit in air at room temperature.18) The minimum number of fatigue cycles shall be greater than 2 × 105 which corresponds to that at the knee point of the S-N curve for the smooth specimen of a common annealed austenitic stainless steel as shown in Fig. 4. If the materials pass these criteria, then the fatigue limit is confirmed to not be reduced by the presence of hydrogen. The test conditions and criteria of Options 1 and 2 are summarized in Table 2.
| Test and requirements | Option 1 Notched method | Option 2 Smooth method | |
|---|---|---|---|
| SSRT test | Test conditions | Not required | H2 pressure: 1.25 NWP Temperature: 228 ± 5 K Strain rate: 5 × 10−5 s−1 |
| Number of specimens | 3 | ||
| Criteria | YS(H2) > 0.8 × YS(air) | ||
| Fatigue life test | Test conditions | H2 pressure: 1.25 NWP Temperature: 293 ± 5 K Stress: 1/3 × S* Frequency: 1 Hz | H2 pressure: 1.25 NWP Temperature: 293 ± 5 K Stress: 1/3 × S* Frequency: 1 Hz |
| Number of specimens | 3 | 3 | |
| Criteria | N > 105 | N > 2 × 105 | |
This paper described the methods for material testing in a high pressure hydrogen gas environment and evaluation of the hydrogen compatibility of metallic materials for application in FCVs. The proposed test methods are summarized as follows:
(1) The effects of the test conditions on the SSRT test and fatigue life test were clarified and test conditions for hydrogen compatibility tests were determined.
(2) Two different test methods were proposed. Option 1 is the notched fatigue life specimen methodology and Option 2 is the SSRT and smooth fatigue life specimen methodology.
These test methods are now under deliberation to be included in HFCV-gtr phase 2
This work was supported by the New Energy and Industrial Technology Development Organization (NEDO), Research and Development of Technology for Hydrogen Utilization/Research and Development for Optimization of Domestic Regulation, International Harmonization and International Standardization for FCV and Hydrogen Infrastructure/Research and Development on Improvement and International Harmonization of compressed hydrogen container regulations for FCVs (FY2013–FY2017) and Development of Technologies for Hydrogen Refueling Stations/Research and Development for International Deployment and Standardization/Research and Development on International Harmonization of regulations and standards for FCVs (FY2018-FY2020).
