PSI - Issue 54

Magdalena Eškinja et al. / Procedia Structural Integrity 54 (2024) 123–134 M. Eškinja et al. / Structural Integrity Procedia 00 (2019) 000 – 000

127

5

Fig. 2. Geometry of tensile specimen used for constant loading in gaseous hydrogen charging experiments. Prior to filling the chamber degassing procedure with argon is executed. For each condition, chamber is filled with different gas mixtures presented in test matrix (Table 3), and total filling pressure is defined to attain constant hydrogen partial pressure under all conditions. Additionally, some experiments were done by filling the chamber with electrolyte (3.5% NaCl) to simulate wet conditions. In case of electrolyte addition, the chamber is mounted on horizontal shafts and rotated during the test at 1 rpm. After 720 h, tensile specimen and cuboidal specimen are studied by TDS for hydrogen characterization and SEM for fractography.

Table 3. Test matrix of gaseous hydrogen charging experiments. Sample Temperature (°C) Load Electrolyte

Pressure H 2 (bar)

Pressure CO 2 (bar)

Pressure H 2 S (bar)

HMoS LMoS HMoS LMoS

RT RT RT RT

80/1.05% 80/1.05% 80/1.05% 80/1.05%

yes yes yes yes

120 120 120 120

- -

1 1 1 1

15 15

2.3.3. Hydrogen Permeation and Diffusion Coefficient Determination Diffusion coefficient of hydrogen was determined by electrochemical permeation test based on Devanthan and Stachurski method [11]. The specimens are square-like with geometry 40 × 40 × 1 mm. One side (exit side) of the specimen is coated with palladium layer employing physical vapor deposition to enable stable current and prevent formation of oxides. Two cell compartments are used as charging and oxidation cells, with the sample clamped between them. On charging side electrolyte was 3.5% NaCl solution with 1 g/L CH 4 N 2 S and on oxidation side 0.1M NaOH. Applied current was 1 mA/cm 2 , and potential on oxidation side was +540 mV against a standard hydrogen electrode. Measurements consist of two step permeation transients as result of charging and discharging cycles. Diffusion coefficient was evaluated from permeation transient using time-lag method according to Equation (1)[12]: = ∙ (1) where D eff is the effective diffusion coefficient, L is the sample thickness and t lag period between starting the charging and reaching 63% of the oxidation current plateau.