Review—Polyaniline-Based Hydrogenochromic Sensor and Its Application for Visualizing Hydrogen Entry into Metals and Microstructure-Dependent Hydrogen Diffusion

Abstract

Structural metallic materials used in hydrogen gas or corrosive environments may suffer from loss of ductility owing to hydrogen atoms (hydrogen embrittlement). To design hydrogen-resistant metallic materials, it is crucial to elucidate the mechanism of hydrogen entry and diffusion. However, visualization of corrosion-induced hydrogen entry and microstructure-dependent hydrogen diffusion requires a highly sensitive hydrogen detection technique with high spatial and temporal resolutions. Hydrogen visualization techniques using polyaniline (PANI), which is a hydrogenochromic sensor, have recently been developed. The PANI layer reacts with atomic state hydrogen in a metal, changing its color from blue to yellow. Thus, the hydrogen distribution in the metal can be analyzed by observing the color distribution of the PANI layer using a digital camera. Owing to the high sensitivity and spatial resolution of hydrogenochromic sensors, corrosion-induced hydrogen entry and microstructure-dependent hydrogen diffusion have been successfully visualized in real time. In this paper, the principles of the sensor and representative application examples are introduced.

Fig. 20 Spatial and temporal resolutions of the HVIS and the other hydrogen analysis techniques. (online color)

1. Introduction

Hydrogen is a renewable energy source, and further use of hydrogen energy is expected to reduce greenhouse gas emissions. Infrastructure for producing and using hydrogen is essential to realize a society based on hydrogen energy (hydrogen energy society). Alloys with excellent mechanical properties have been extensively studied as structural metallic materials for a hydrogen energy society. However, certain alloys are susceptible to hydrogen embrittlement. Hydrogen embrittlement is the loss of ductility caused by hydrogen atoms entering a material [1–6]. The use of alloys that are susceptible to hydrogen embrittlement is often limited to hydrogen gas or corrosive environments where hydrogen atoms readily enter the metals. To develop hydrogen-resistant alloys, the prevention of hydrogen entry and embrittlement has been extensively investigated in recent years [7–10].

To inhibit hydrogen embrittlement of structural metallic materials, it is crucial to understand the mechanisms of hydrogen entry and diffusion. However, atomic state hydrogen, the smallest atom in the universe, is difficult to detect using conventional elemental analyzers. Various visualization techniques have been developed for the detection of hydrogen in metals. Hydrogen microprinting (HMT) [11–13], secondary ion mass spectrometry (SIMS) [14–16], Ag decoration (Ag) [14, 17, 18], scanning Kelvin probe force microscopy (SKPFM) [4, 14, 18–21], atom probe tomography (APT) [22–24], and scanning Kelvin probe (SKP) [25–29] can visualize hydrogen distribution in metals with high spatial resolution. In addition, Sugawara et al. developed a highly sensitive hydrogen detection technique using a metal oxide layer, which changes its color owing to the reaction with hydrogen [30, 31]. Ajito et al. utilized the gaschromic reaction of an Ir complex to visualize the hydrogen flux in a metal and developed a highly sensitive hydrogen visualization technique with high spatial resolution [32, 33].

Recently, hydrogenochromic sensors have been developed using polyaniline (PANI) to detect hydrogen at high spatial and temporal resolutions [34–40]. PANI is an inexpensive and easily polymerizable conductive polymer [41–44]. It has been revealed that a PANI layer reacts with atomic state hydrogen in metals, and the color of the PANI layer is brightened due to the hydrogenation reaction [34]. A highly sensitive hydrogenochromic sensor is developed using a thin PANI layer. The hydrogenochromic sensor is sufficiently sensitive to visualize corrosion-induced hydrogen entry and microstructure-dependent hydrogen diffusion in metals. This review discusses the principles and applications of hydrogen visualization using PANI-based hydrogenochromic sensors.

2. Hydrogenochromic Sensors

2.1 Principle of hydrogen visualization using PANI

Figure 1 shows a schematic of the hydrogenochromic sensor. When hydrogen is adsorbed onto a metal surface in an aqueous solution, some of the adsorbed hydrogen enters the metal. The hydrogen atoms diffuse to the other side of the specimen, where the PANI layer is formed. When PANI reacts with atomic state hydrogen in the metal, the quinoid structure is hydrogenated, resulting in a color change [34]. As the visible color of the PANI layer changes, the hydrogen distribution in the metal can be analyzed by observing the color distribution of the PANI layer.

Fig. 1

Schematic of the mechanism for visualizing the hydrogen distribution in metals using PANI [34]. (online color)

The visualization of macroscopic hydrogen diffusion in an Fe sheet has been attempted using a hydrogenochromic sensor [34]. Figure 2(a) shows the electrochemical cell used to visualize the macroscopic hydrogen distribution. An annealed Fe sheet (99.5%, thickness: 2 mm) is electrochemically polished in a mixed solution of perchloric acid and acetic acid to remove the deformed layer that inhibits hydrogen diffusion. A Ni layer with a thickness of 300–400 nm is formed on one side of the specimen via electroplating. This Ni layer acts as a catalyst for the hydrogenation reaction of PANI, thereby enhancing the sensitivity of the PANI layer to hydrogen atoms. The Ni layer is anodically polarized in a 0.5 M sulfuric acid–0.5 M aniline solution to form a PANI layer. The thickness of the PANI layer is approximately 300 nm. During the hydrogen visualization test, hydrogen is introduced into the Fe sheet via the cathodic hydrogen charging: the exposed Fe surface is immersed in a NaCl aqueous solution and cathodically polarized to accelerate the reduction of protons on the Fe surface. In this review, the specimen surface immersed in the aqueous solution to introduce hydrogen is referred to as the hydrogen entry side, and the specimen surface with the PANI layer is referred to as the hydrogen detection side. On the hydrogen entry side (Fig. 2(b)), the right half of the specimen is coated with resin, and hydrogen is charged only to the left half inside the O-ring; thus, the right half of the hydrogen detection side (Fig. 2(c)) corresponds to the area where hydrogen is charged (hydrogen-charged area).

Fig. 2

Schematic of the (a) electrochemical cell used for the hydrogen visualization test: (b) hydrogen entry and (c) detection sides [34]. (online color)

Figure 3 shows optical micrographs of the PANI layer before and during hydrogen charging. The areas on the left and right sides of the yellow dashed line correspond to the uncharged and charged areas of hydrogen, respectively. After 20 min, the PANI layer in the hydrogen-charged area becomes slightly brighter. Thereafter, the color of the hydrogen-charged area continues to brighten with time, while that of the hydrogen-uncharged area is barely changed. These results indicate that the color of the PANI layer is changed owing to the reaction with hydrogen, and the hydrogen flux distribution in the Fe sheet is visualized.

Fig. 3

Optical images of the PANI layer during hydrogen charging [34]. (online color)

Figure 4 shows the UV–visible light absorption spectra of the PANI layer measured using a UV–visible spectrophotometer. Compared to the absorption spectrum of the PANI layer after 1 h of polymerization, the spectrum is rarely changed after storage in air without hydrogen charging for 23 h (Fig. 4(a)), indicating that the optical properties of the PANI layer are stable under irradiation by the UV–visible light in air. Figure 4(b) shows the absorption spectra of the PANI layer before and after hydrogen charging. The peak wavelength before hydrogen charging is approximately 640 nm, which is roughly consistent with the absorption spectrum of the emeraldine base, an oxidized state of PANI [42, 45]. After 3 h of hydrogen charging, the absorbance is decreased in the range of 450–650 nm, and the peak wavelength shifts to 710 nm. When the emeraldine base is reduced in an aqueous solution to form a benzenoid structure, the absorbance at approximately 630 nm decreases, and the peak wavelength shifts from 630–660 nm to 780–860 nm [44, 46]. Therefore, the quinoid structure of the emeraldine base is considered to be reduced by atomic state hydrogen in the Fe sheet to form a benzenoid structure, thereby changing the optical properties of the PANI layer. Because the absorption spectrum of the PANI layer is not changed in a hydrogen gas atmosphere [34], the color change of the PANI layer shown in Fig. 4 is concluded to be caused by a reaction with hydrogen atoms permeating through the Fe sheet.

Fig. 4

Absorption spectra of the PANI layer (a) stored in ambient air for 1 h and 24 h and (b) before and after hydrogen charging [34]. (online color)

2.2 Quantitative analysis of color change of the PANI layer

The main driving force of hydrogen diffusion in a metal is the hydrogen concentration gradient. Thus, the hydrogen atoms in the Fe sheet are expected to diffuse not only in the thickness direction but also in the horizontal direction. Hydrogen diffusion in the horizontal direction can be analyzed in detail by quantifying the color change of the PANI layer [34]. The R (red), G (green), and B (blue) values of each pixel on line A–B in the inset of Fig. 5 are extracted to calculate the brightness, Y, which is often used to quantify the brightness of the object [47]:

  

\begin{equation} Y = 0.229R + 0.587G + 0.114B \end{equation} (1)

Further, the difference in brightness, ΔY, was calculated as follows:

  

\begin{equation} \Delta Y = Y_{\text{t}} - Y_{0} \end{equation} (2)

Here, ΔY is the difference between the Y_t value at time t and the Y₀ value before hydrogen charging. Figure 5 shows the line profiles of the ΔY value on the line A–B. The center of line A–B (x = 0) corresponds to the boundary between the hydrogen-uncharged and charged areas. In the hydrogen-charged area (from x = 0 to point B), the ΔY value is increased with time. In the hydrogen-uncharged area, the ΔY value does not change in 5 min, but it is increased in the range from x = 0 to x = −3 after 30 min. The increase in the ΔY value in the hydrogen-uncharged area is larger near the hydrogen-charged area, indicating that hydrogen atoms diffuse in the horizontal direction of the Fe sheet. The hydrogen atoms in the hydrogen-charged area flow into the hydrogen-uncharged area. Consequently, the hydrogen flux in the hydrogen-uncharged area increased as the position approaches that of the hydrogen-charged area. Using the PANI-based hydrogenochromic sensor, the two-dimensional (2D) distribution of hydrogen flux in a metal can be observed with high sensitivity. Owing to the change in visible color of the PANI layer, the hydrogen distribution can be easily captured with a digital camera, and image analyses allow for a detailed understanding of the hydrogen diffusion behavior.

Fig. 5

Change in the line profiles of the positions indicated by the yellow line in the inset. The center of the yellow line (x = 0) indicates the boundary between the hydrogen-uncharged and hydrogen-charged areas [34]. (online color)

3. Visualization of Corrosion-Induced Hydrogen Entry

3.1 Hydrogen entry into an Fe sheet in a NaCl bulk solution

Hydrogen entry into a metal is promoted by corrosion reactions occurring on the metal surface [5, 48, 49]. The corrosion-induced hydrogen entry is thought to proceed locally [27, 29, 50], and hydrogen embrittlement is induced in areas where the hydrogen concentration is higher than a certain value in the metal [49, 51]. Therefore, elucidating the mechanism of corrosion-induced hydrogen entry is crucial for developing structural materials. Electrochemical hydrogen permeation tests using the Devanathan-Stachurski cell [52] are among the few methods that can measure corrosion-induced hydrogen entry in real time [5, 53]. However, revealing the hydrogen entry site with spatial resolution is difficult, because electrochemical hydrogen permeation tests measure the average hydrogen flux within an electrode area. In contrast, a hydrogenochromic sensor consisting of PANI and Ni layers can visualize corrosion-induced hydrogen entry in real time with high spatial resolution [36]. The white rectangle in the upper row of Fig. 6 shows an electrode area of 10 mm × 10 mm fabricated on an Fe sheet. The electrode area is naturally immersed in a 3 wt% NaCl aqueous solution. The lower row shows optical images of the PANI layer of the hydrogen detection side. The broken orange curve indicates the position of the electrode area on the hydrogen entry side. The optical images of the hydrogen entry side are flipped horizontally. Rust formation is observed in the electrode area after 2 h, whereas a color change begins after 8 h on the hydrogen detection side. The color of the PANI layer becomes brighter with time, indicating that hydrogen entry continues as corrosion proceeds. After 24 h, the rust-formed area was larger than the color-changed area of the hydrogen detection side. Figure 7(a) and 7(b) shows the surface appearance and profiles of the Fe sheet after the test. In the area indicated by the solid green curve, the Fe sheet is dissolved, exhibiting a metallic luster. As shown in Fig. 7(b), surface roughness is observed due to dissolution in the area. By contrast, rust deposition is confirmed in the area indicated by the green dashed curve, which rarely causes surface roughness. This indicates that corrosion proceeds little in the area indicated by the green dashed curve and that rust generated in the area indicated by the green solid curve diffuses and deposits in the area indicated by the green dashed curve. Figure 7(c) shows a contour map of the ΔY value of the PANI layer at 24 h. Hydrogen entry proceeds in the Fe dissolved area, whereas hydrogen is not detected in the area where rust deposits but no corrosion proceeds. The reduction of protons is thought to accelerate hydrogen entry owing to the decrease in pH and potential on the dissolving Fe surface.

Fig. 6

Optical images of the hydrogen entry (upper row) and hydrogen detection (lower row) sides of an Fe sheet immersed in a 3 wt% NaCl solution. The broken orange curves indicate the corresponding position of the electrode area of the hydrogen entry side [36]. (online color)

Fig. 7

(a) Surface appearance and (b) surface profiles of the Fe sheet after the hydrogen visualization test in Fig. 6. (c) Contour map of the ΔY value of the PANI layer at 24 h shown in Fig. 6 [36]. (online color)

3.2 Hydrogen entry under a NaCl droplet on an Fe sheet

High-strength steels used for automobiles and bridges are often used under atmospheric corrosion conditions. As a droplet of the NaCl solution is placed on the Fe surface, corrosion and hydrogen entry proceed under the droplet [54, 55]. Because the solution chemistry of the droplet changes kinetically due to evaporation and corrosion, the electrochemical reaction on the metal surface also changes with time. Therefore, it is believed that the hydrogen entry behavior changes kinetically under the droplet due to changes in the solution chemistry and surface state of the metal [48, 56, 57]. By fabricating a PANI-based hydrogenochromic sensor on one side of an Fe sheet and placing a droplet of NaCl solution on the other side, the corrosion-induced hydrogen entry behavior under the droplet can be visualized in situ [35, 39]. The upper and lower rows of Fig. 8 show an Fe sheet with a droplet of 3 wt% NaCl and a hydrogenochromic sensor formed on the hydrogen detection side, respectively. The white dashed curve on the hydrogen detection side indicates the initial position of the droplet on the hydrogen entry side. After 60 min, corrosion and rust formation are confirmed; however, the color of the PANI layer is not changed. As the rust-formed area grows, the color of the PANI layer starts to brighten in that area. This indicates that hydrogen entry is enhanced on the dissolving Fe surface. After 390 min, the droplets disappeared owing to evaporation, and the color of the PANI layer changed slightly. However, hydrogen entry was observed in the area indicated by the white arrow at 600 min, suggesting that hydrogen entry continues locally, even after the disappearance of the droplet.

Fig. 8

Optical images of the hydrogen entry (upper row) and detection (lower row) sides of the Fe sheet. A droplet of a 3 wt% NaCl solution is placed on the hydrogen entry side. The broken white line in the image of the hydrogen detection side indicates the area initially covered with the droplet on the hydrogen entry side [35]. (online color)

Figure 9 shows the time variations of the rust-formed area (A_rust), color-changed area of the PANI layer (A_Y), and area of the droplet (A_drop) during the hydrogen visualization test shown in Fig. 8. The rust-formed area gradually increases after approximately 30 min, followed by a rapid increase after 90 min. Hydrogen entry barely proceeds in the initial stage of corrosion (first 90 min) and is accelerated by the rapid growth of the rust-formed area. Because hydrogen atoms enter the Fe sheet and diffuse to the other side in 1 min, the time lag between the corrosion initiation and detection of hydrogen implies the incubation time for hydrogen entry. It is believed that the pH and potential in the dissolved area gradually decrease as corrosion proceeds, and hydrogen entry is triggered when the pH under the rust becomes lower than 4 [39]. After the disappearance of the droplet, the growth of the rust-formed area ceased; however, the hydrogen entry site expanded slightly, indicating that corrosion continues locally under the rust layer even after the disappearance of the droplet, resulting in hydrogen entry.

Fig. 9

Temporal variation of the areas of the rust-formed area (A_rust), droplet (A_drop), and preferential hydrogen entry site (A_Y) during the hydrogen visualization test in Fig. 8 [35]. (online color)

As described above, a hydrogenochromic sensor consisting of PANI and Ni layers can be used for real-time visualization of the 2D distribution of corrosion-induced hydrogen entry into an Fe sheet in an aqueous solution or under a droplet. Based on the simultaneous observation of corrosion and the PANI layer, the relationship between the corrosion reaction and hydrogen entry, which changes significantly over time, can be analyzed in detail.

4. Microstructure-Dependent Hydrogen Diffusion in Metals

4.1 Hydrogen video imaging system

Hydrogen diffusion behavior in metals is thought to depend on the microstructure [58–61]. Grain boundaries (GBs) and crystallographic orientations are thought to contribute to hydrogen diffusion and trapping; however, microstructure-dependent hydrogen diffusion behavior remains unclear. For instance, it is still under discussion whether GBs act as hydrogen trapping sites that inhibit hydrogen diffusion or as a preferential diffusion path [62–66]. To elucidate the microstructure-dependent hydrogen diffusion behavior, a 2D visualization of the hydrogen flux distribution with micrometer-scale spatial and video-rate temporal resolutions is ideal. Furthermore, the observation view should be sufficiently large to cover the microstructural heterogeneity of the metal. In case of the PANI layer, the visible color is changed due to the reaction with hydrogen atoms. By observing the PANI layer with an optical microscope, the hydrogen flux distribution of a sub-millimeter-scale observation view can be analyzed with high spatial resolution. Figure 10 shows a schematic of the hydrogen video imaging system (HVIS) [37]. One side of the specimen was immersed in an aqueous solution, and hydrogen was introduced into the specimen via cathodic charging. On the other side of the specimen, the PANI layer is formed as a hydrogenochromic sensor, and the flux distribution of hydrogen atoms diffusing through the specimen is analyzed based on the color distribution of the sensor. The color distribution of the PANI layer can be observed in real time with sub-micrometer scale spatial resolution by observing the color change using an inverted optical microscope.

Fig. 10

Schematic of the hydrogen video imaging system [37]. (online color)

4.2 Preferential hydrogen diffusion at grain boundaries of pure Ni

Using the HVIS described above, the hydrogen diffusion behavior in pure polycrystalline Ni foil is analyzed [37]. A thin PANI layer can be formed on a pure Ni foil via constant voltage polarization in a mixed solution of sulfuric acid and aniline. The morphology of the PANI layer is confirmed to be independent of the microstructure of the Ni foil. Figure 11(a) shows optical micrographs of the PANI layer before and during the hydrogen visualization test obtained using HVIS. The white dot indicated by the yellow arrow is a mark to synchronize the area of multiple optical micrographs and not the color change of the PANI layer. The color of the PANI layer before hydrogen charging is uniformly purple. However, after 20 h of hydrogen charging, the color turns white locally. The whitening indicates that a hydrogen flux distribution exists in the Ni foil.

Fig. 11

(a) Optical micrograph of the PANI layer before and during hydrogen charging. A scratch, indicated by the yellow arrow, is made to synchronize the positions in multiple optical micrographs. (b) Enlarged views and (c) GBs of the pure Ni of the area indicated by the red rectangle in (a) [37]. (online color)

Figure 11(b) shows enlarged optical micrographs of the area indicated by the red square in Fig. 11(a). The PANI layer locally turned white after 16 h, and the number of white areas increased with time. Figure 11(c) shows a GB map of the Ni foil in the area corresponding to Fig. 11(b). As shown in Fig. 11(b) and 11(c), whitening occurs at some GBs, indicating that the hydrogen flux at the GBs is larger than that at the grain interior. In Fig. 11(c), the blue and orange curves indicate the corresponding site lattice (CSL) GBs (Σ3–Σ19) and random GBs, respectively. In the figure, the CSL GBs are characterized based on the misorientation angle and common rotation axis. Except for the CSL GBs, the GBs are shown as random GBs. The positions of the color-changed area of the PANI layer roughly corresponded to the random GBs of the Ni foil, suggesting that random GBs are the preferential hydrogen diffusion paths in pure Ni.

Figure 12(a) and 12(b) shows an optical micrograph of the PANI layer and the GB map of the Ni foil shown in Fig. 11, respectively. Areas 1–3 in Fig. 12(a) correspond to the areas containing the Σ3 GB, grain interior, and random GB, respectively. To quantitatively analyze the time variation of the color change in Areas 1–3, the R, G, and B values of each pixel in each area are extracted, and the difference in brightness, ΔY, is calculated. Figure 12(c) shows the time variations of the average ΔY value in each area. In Area 3, which contains a random GB, the ΔY value starts to increase rapidly after 16 h, while that in Areas 1 and 2 increases gradually after 18 h. The time variations of the ΔY value in Area 2 (grain interior) and Area 3 (Σ3 GB) are almost the same. It is found that the hydrogen flux at random GBs is larger than that at Σ3 GBs and that the hydrogen flux at the grain interior is almost the same as that at Σ3 GBs.

Fig. 12

(a) Optical micrograph of the PANI layer shown in Fig. 11(a), and (b) CSL and random GBs of the corresponding area obtained by EBSD. (c) Time variations of the ΔY value in Areas 1–4 indicated by the squares in (a) [37]. (online color)

In addition to GBs, other metallographic heterogeneities exist in metals, such as crystallographic orientations, inclusions, and dislocations. Therefore, statistical analysis is necessary to clarify the effect of material factors on hydrogen flux. Because the ΔY value of the PANI layer roughly corresponds to the logarithm of the integrated value of hydrogen flux, the hydrogen flux at each GB can be evaluated by analyzing the ΔY value. Then, the time when the ΔY value for each GB reaches 60 is defined as t_ΔY60. Figure 13 shows t_ΔY60 for GBs that are longer than 10 µm in the observation view shown in Fig. 11(b). As a reference, t_ΔY60 for grain interior is indicated by the blue dashed line. The GBs with the shortest t_ΔY60 are those with a misorientation angle of 30° ≤ θ < 40°, indicating that the hydrogen flux is highest at these GBs. It is found that the hydrogen flux at the low-angle and random GBs is larger than that of Σ3 GBs. Because the t_ΔY60 at the grain interior is ca. 90 h, the hydrogen flux at low-angle GBs (10° ≤ θ < 15°) is larger than that at the grain interior. These results indicate that the density of coincidence site lattice (Σ value) and misorientation angle, θ, at a particular corresponding GB are important factors governing the hydrogen flux at the GB. Zhou et al. reported that the GB of pure Ni is a preferential hydrogen diffusion path, based on first-principles calculations and kinetic Monte Carlo simulations [62]. Furthermore, they calculated the hydrogen diffusion coefficient of Ni GBs based on the Frank–Bilby model and suggested that the hydrogen diffusion coefficient depends on the misorientation angle. For instance, net defect density is known to increase monotonically with increasing θ in the range of θ < 36.87° and decrease monotonically in the range of 36.87° < θ < 90° in case a common rotation axis is ⟨001⟩ [67, 68]. This trend is in good agreement with the misorientation-dependence of hydrogen flux clarified using t_ΔY60, which is minimum in the range 30° ≤ θ < 40°. It was concluded that the geometric structure of the GBs plays a critical role in the hydrogen flux at the GBs of Ni foils.

Fig. 13

Permeation time (t_ΔY60) of each GB with a length of more than 10 µm in the area shown in Fig. 12(a) [37]. (online color)

Figures 11–13 show that the hydrogen flux at the Σ3 GB is almost the same as that at the grain interior. However, t_ΔY60 of Σ3 GBs in Fig. 13 scatters: some Σ3 GBs show hydrogen flux as high as that of random GBs. In the GB maps shown in Figs. 11 and 12, the Σ3 GBs are defined based on the misorientation angle and common rotation axis, but there are coherent and incoherent Σ3 GBs. Because incoherent Σ3 GBs have a larger geometric space than the coherent Σ3 GBs parallel to the (111) plane of adjacent grains [69], preferential hydrogen diffusion may occur as in the case of random GBs. Thus, the effect of coherency of the Σ3 GBs on hydrogen flux has been investigated [40]. The deviation of the Σ3 GBs alignment from the ideal coherent twin boundary is often two-dimensionally analyzed using surface traces of target Σ3 GBs and the corresponding coherent twin boundary plane, and the coherency of Σ3 GBs is evaluated based on the trace tolerance [70–72]. In this review, the deviation angles of a Σ3 GB from the (111) surface trace of adjacent grains are ω₁ and ω₂, and the average values are defined as the deviation, ω_ave, of the Σ3 GB. Thus, the completely coherent Σ3 GB exhibits deviation, ω_ave = 0. Here, a Σ3 GB with a deviation less than 5° is referred to as a coherent Σ3 GB, and the others are referred to as incoherent Σ3 GBs, considering the measurement error of the EBSD method. Figure 14(a) shows an optical micrograph of a hydrogen visualization test with a pure Ni foil (thickness: 90 µm) using the same HVIS as in Fig. 11. Figure 14(b) shows the backscattered electron (BSE) image of the same area taken after the removal of the PANI layer. The green lines in Fig. 14(b) show the traces of the {111} planes of each grain. The white dashed lines in Fig. 14(b) are both coherent Σ3 GBs with a deviation of 2°; the crystal structure of coherent Σ3 GBs is almost the same as that of the grain interior, so preferential hydrogen diffusion is not observed. The color of the PANI layer at coherent Σ3 GBs is the same as that of the grain interior. Meanwhile, the Σ3 GBs indicated by the yellow arrows show enhanced hydrogen diffusion. One of the GBs is an incoherent Σ3 GB with a deviation angle of 24°, while the other is a coherent Σ3 GB. Although the Σ3 GBs can be coherent at the microscale, they often consist of incoherent facets that result in the formation of an incoherent portion [64, 65, 67]. Figure 14(c) and 14(d) show enlarged views of the GBs indicated by the yellow arrows in Fig. 14(a) and 14(b), respectively. Figure 14(e) shows the contour map of the ΔY value in Fig. 14(c). The dashed lines and arrows in Fig. 14(e) indicate the positions of GBs and facets, respectively. The green dashed lines in Fig. 14(d) and 14(e) indicate incoherent GBs with a deviation of approximately 16, and the white dashed lines indicate coherent GBs with a deviation of less than 5. Figure 14(e) shows that incoherent and coherent portions are formed due to the facets and that preferential hydrogen diffusion occurred at the incoherent portions.

Fig. 14

(a) Optical micrograph of a PANI layer on pure Ni foil and (b) BSE image of the corresponding area of the Ni foil. The traces of the {111} planes are shown in green. The deviation angle (Δω) of each boundary is shown in degrees. (c) Enlarged optical micrograph, (d) BSE image, and (e) the ΔY contour map of the GB indicated by the yellow arrows in (b). The green and white dashed lines in (d) and (e) indicate the incoherent and coherent portions of the Σ3 GB, respectively [40]. (online color)

The relationship between the GB characteristics and hydrogen flux is statistically analyzed using time variations in the ΔY value. The curves in Fig. 15 show the time variations of the average ΔY values at the GBs. The colored shadows indicate the range of statistical data. The average ΔY value at the grain interior is also shown as a reference. The average hydrogen fluxes at the GBs and grain interior are found to be in the following order: high-angle random > incoherent Σ3 > low-angle > coherent Σ3 and grain interior. This indicates that as the free volume of the grain boundary grows, so does the hydrogen flux. Previously, it has been challenging to experimentally and statistically examine hydrogen diffusion at the GBs of pure Ni, but HVIS can visualize hydrogen diffusion at thousands of GBs with high spatial and temporal resolution. This paves the way for the experimental elucidation of the relationship between hydrogen diffusion and GB characteristics. It is envisioned that future research will clarify not only the GB characteristics, but also various material factors, such as segregated components and precipitates at the GBs.

Fig. 15

Time variations of the average ΔY value of high-angle random (Random), incoherent Σ3 (Incoherent Σ3), low-angle (Low-angle), coherent Σ3 (Coherent Σ3) GBs, and grain interior (Grain interior) [40]. (online color)

4.3 Microscopic hydrogen flux distribution in super duplex stainless steel

The hydrogenochromic sensor and HVIS can be applied not only to pure metals but also to practical alloys, such as stainless steel. Super duplex stainless steels are composed of ferrite (α) and austenite (γ) phases and are often used in harsh corrosive environments owing to their excellent strength and corrosion resistance [73]. However, hydrogen embrittlement may occur when hydrogen enters super duplex stainless steel [4, 74–76]. HVIS has been used to analyze the microstructure-dependent hydrogen diffusion behavior in super duplex stainless steels [38]. Figure 16(a)–(h) shows optical micrographs of the hydrogenochromic sensor formed on super duplex stainless steel. The color of the PANI layer is changed locally, and the color change proceeds with time. Figure 16(i) shows the phase map of the super duplex stainless steel in the same observation view as the optical micrographs. A comparison of Fig. 16(d) and 16(i) confirms that the hydrogen flux in the α phases is greater than that in the γ phases. Hydrogen diffusion coefficients tend to be larger in the body-centered cubic lattice than in the face-centered cubic lattice [77, 78]. Although the hydrogen diffusion coefficient is also affected by the solid solution elements [79, 80], the difference in concentration of the solid solution elements in each phase has a negligible influence on hydrogen diffusion. Thus, the difference in hydrogen flux in each phase is attributed to the crystal structure.

Fig. 16

Optical micrograph of the PANI layer (a) before and (b)–(h) during hydrogen charging and (i) phase map of super duplex stainless steel [38]. (online color)

Figure 17 shows enlarged views of an area in the optical micrograph in Fig. 16. The yellow dashed curves indicate the boundary between the α and γ phases (phase interface). After 30 h, hydrogen is detected only in the α phase, followed by detection in the γ phase. Figure 18 shows the line profile of the ΔY value on the line A–B shown in Fig. 17. Shaded areas correspond to the α phase. Hydrogen is detected only in the α phase until 26 h. However, after 30 h, hydrogen is detected in the γ phase near the phase interface. Furthermore, the hydrogen flux in the α phase decreased as it approaches the phase interface, suggesting that hydrogen atoms diffuse from the α to γ phases. The HVIS clarified that the hydrogen flux in super duplex stainless steels is phase-dependent and that the microscale distribution of hydrogen is caused by the diffusion of hydrogen in the α phase, which is the preferential diffusion path, into the γ phase. As most of the hydrogen in the γ phase diffuses from the α phase, the hydrogen distribution in the γ phase strongly depends on the distance from the phase interface. Figure 19 shows the relationship between the ΔY value and distance from the phase interface at the center of the γ and α phases. The smaller the γ phase, the greater the hydrogen flux because the distance between the center of the γ phase and the phase interface is smaller. In contrast, for the α phase, it was found that the larger the phase, the larger the hydrogen flux.

Fig. 17

Enlarged view of the optical micrograph shown in Fig. 16 [38]. (online color)

Fig. 18

Line profiles of the ΔY value at line A–B in Fig. 17 [38]. (online color)

Fig. 19

The ΔY value of (a) the γ phases at 80 h and (b) the α phases at 30 h during hydrogen charging as a function of the grain size of each phase [38]. (online color)

Figure 20 shows the spatial and temporal resolutions of hydrogen visualization techniques. HVIS simultaneously achieves high spatial and temporal resolutions, which is a trade-off for conventional hydrogen visualization techniques. While HVIS is a simple and inexpensive method for visualizing hydrogen, it can provide detailed information about the microstructure-dependent hydrogen diffusion behavior. In the future, HVIS is expected to be used not only on steels and Ni alloys, but also on Al and Pd alloys [81–85], and to become a versatile technique for analyzing hydrogen diffusion in metals.

Fig. 20

Spatial and temporal resolutions of the HVIS and the other hydrogen analysis techniques. (online color)

5. Conclusion

Hydrogen entry and diffusion behaviors in materials are crucial for elucidating hydrogen embrittlement mechanisms and developing techniques for preventing hydrogen entry. This paper reviewed a recently developed hydrogenochromic sensor using polyaniline (PANI) and its applications. The PANI layer is hydrogenated (reduced) by atomic state hydrogen in a metal, resulting in a color change. The hydrogenation reaction is accelerated by the Ni layer between the PANI layer and the metal; thus, the hydrogenochromic sensor consisting of PANI and Ni layers shows high sensitivity. A highly sensitive hydrogenochromic sensor can be used to visualize corrosion-induced hydrogen entry. It is expected that the simultaneous observation of corrosion and hydrogen entry will contribute to the further clarification of the corrosion-induced hydrogen entry mechanism and development of inhibitors and surface treatments to prevent hydrogen entry. Additionally, the hydrogen video imaging system (HVIS) has been developed using a hydrogenochromic sensor and optical microscope. The HVIS can visualize hydrogen diffusion behavior at a sub-millimeter-scale observation view with sub-micrometer scale spatial and video-rate temporal resolutions, enabling a comprehensive analysis of the relationship between material factors and hydrogen diffusion. This simple and versatile approach enables the application of HVIS not only to pure metals but also to practical alloys, such as stainless steels. In future work, the role of material factors, such as dislocation, grain boundary segregation, and inclusion, in hydrogen diffusion is expected to be clarified, contributing to the establishment of material design guidelines for alloys that are highly resistant to hydrogen embrittlement.

Acknowledgments

This work was supported by JSPS KAKENHI Grant number JP25K01541.

REFERENCES

• 1)   I.M.  Robertson,  P.  Sofronis,  A.  Nagao,  M.L.  Martin,  S.  Wang,  D.W.  Gross and  K.E.  Nygren: Hydrogen Embrittlement Understood, Metall. Mater. Trans. A 46 (2015) 2323–2341. doi:10.1007/s11661-015-2836-1
• 2)   M.  Koyama,  K.  Tomatsu,  H.  Kakinuma,  S.  Ajito and  E.  Akiyama: Hydrogen Effects on Hall–Petch Coefficient and Fracture Strength in Pure Ni and Ni-20Cr Alloy: A Macroscopic Insight of the Hydrogen Embrittlement Mechanism, Metall. Mater. Trans. A 56 (2025) 771–777. doi:10.1007/s11661-025-07688-6
• 3)   M.  Koyama,  E.  Akiyama,  Y.-K.  Lee,  D.  Raabe and  K.  Tsuzaki: Overview of hydrogen embrittlement in high-Mn steels, Int. J. Hydrogen Energ. 42 (2017) 12706–12723. doi:10.1016/j.ijhydene.2017.02.214
• 4)   C.  Örnek,  P.  Reccagni,  U.  Kivisäkk,  E.  Bettini,  D.L.  Engelberg and  J.  Pan: Hydrogen embrittlement of super duplex stainless steel – Towards understanding the effects of microstructure and strain, Int. J. Hydrogen Energ. 43 (2018) 12543–12555. doi:10.1016/j.ijhydene.2018.05.028
• 5)   M.  Kawamori and  F.  Yuse: In-situ measurement of hydrogen entry and hydrogen embrittlement of steel by atmospheric corrosion, Corros. Sci. 219 (2023) 111212. doi:10.1016/j.corsci.2023.111212
• 6)   T.  Depover,  E.  Wallaert and  K.  Verbeken: Fractographic analysis of the role of hydrogen diffusion on the hydrogen embrittlement susceptibility of DP steel, Mater. Sci. Eng. A 649 (2016) 201–208. doi:10.1016/j.msea.2015.09.124
• 7)   N.-E.  Laadel,  M.  El Mansori,  N.  Kang,  S.  Marlin and  Y.  Boussant-Roux: Permeation barriers for hydrogen embrittlement prevention in metals – A review on mechanisms, materials suitability and efficiency, Int. J. Hydrogen Energ. 47 (2022) 32707–32731. doi:10.1016/j.ijhydene.2022.07.164
• 8)   M.  Wetegrove,  M.J.  Duarte,  K.  Taube,  M.  Rohloff,  H.  Gopalan,  C.  Scheu,  G.  Dehm and  A.  Kruth: Preventing Hydrogen Embrittlement: The Role of Barrier Coatings for the Hydrogen Economy, Hydrogen 4 (2023) 307–322. doi:10.3390/hydrogen4020022
• 9)   M.  Kawamori,  W.  Urushihara and  S.  Yabu: Improved Hydrogen Embrittlement Resistance of Steel by Shot Peening and Subsequent Low-temperature Annealing, ISIJ Int. 61 (2021) 1159–1169. doi:10.2355/isijinternational.ISIJINT-2020-463
• 10)   Y.-S.  Chen,  C.  Huang,  P.-Y.  Liu,  H.-W.  Yen,  R.  Niu,  P.  Burr,  K.L.  Moore,  E.  Martínez-Pañeda,  A.  Atrens and  J.M.  Cairney: Hydrogen trapping and embrittlement in metals – A review, Int. J. Hydrogen Energ. (2025) 789–821. doi:10.1016/j.ijhydene.2024.04.076
• 11)   J.  Ovejero-García: Hydrogen microprint technique in the study of hydrogen in steels, J. Mater. Sci. 20 (1985) 2623–2629. doi:10.1007/BF00556094
• 12)   K.  Ichitani,  S.  Kuramoto and  M.  Kanno: Quantitative evaluation of detection efficiency of the hydrogen microprint technique applied to steel, Corros. Sci. 45 (2003) 1227–1241. doi:10.1016/S0010-938X(02)00218-4
• 13)   K.  Ichitani and  M.  Kanno: Visualization of hydrogen diffusion in steels by high sensitivity hydrogen microprint technique, Sci. Technol. Adv. Mater. 4 (2003) 545–551. doi:10.1016/j.stam.2003.12.006
• 14)   Z.  Tarzimoghadam,  M.  Rohwerder,  S.V.  Merzlikin,  A.  Bashir,  L.  Yedra,  S.  Eswara,  D.  Ponge and  D.  Raabe: Multi-scale and spatially resolved hydrogen mapping in a Ni–Nb model alloy reveals the role of the δ phase in hydrogen embrittlement of alloy 718, Acta Mater. 109 (2016) 69–81. doi:10.1016/j.actamat.2016.02.053
• 15)   M.  Koyama,  M.  Rohwerder,  C.C.  Tasan,  A.  Bashir,  E.  Akiyama,  K.  Takai,  D.  Raabe and  K.  Tsuzaki: Recent progress in microstructural hydrogen mapping in steels: Quantification, kinetic analysis, and multi-scale characterisation, Mater. Sci. Technol. 33 (2017) 1481–1496. doi:10.1080/02670836.2017.1299276
• 16)   K.  Takai,  Y.  Chiba,  K.  Noguchi and  A.  Nozue: Visualization of the hydrogen desorption process from ferrite, pearlite, and graphite by secondary ion mass spectrometry, Metall. Mater. Trans. A 33 (2002) 2659–2665. doi:10.1007/s11661-002-0387-8
• 17)   E.  Akiyama and  S.  Matsuoka: Hydrogen Visualization in Steels Using Ag Decoration Method, Mater. Trans. 56 (2015) 793–797. doi:10.2320/matertrans.M2014431
• 18)   T.  Nagashima,  M.  Koyama,  A.  Bashir,  M.  Rohwerder,  C.C.  Tasan,  E.  Akiyama,  D.  Raabe and  K.  Tsuzaki: Interfacial hydrogen localization in austenite/martensite dual-phase steel visualized through optimized silver decoration and scanning Kelvin probe force microscopy, Mater. Corros. 68 (2017) 306–310. doi:10.1002/maco.201609104
• 19)   M.  Koyama,  A.  Bashir,  M.  Rohwerder,  S.V.  Merzlikin,  E.  Akiyama,  K.  Tsuzaki and  D.  Raabe: Spatially and Kinetically Resolved Mapping of Hydrogen in a Twinning-Induced Plasticity Steel by Use of Scanning Kelvin Probe Force Microscopy, J. Electrochem. Soc. 162 (2015) C638–C647. doi:10.1149/2.0131512jes
• 20)   S.  Evers,  C.  Senöz and  M.  Rohwerder: Hydrogen detection in metals: a review and introduction of a Kelvin probe approach, Sci. Technol. Adv. Mater. 14 (2013) 014201. doi:10.1088/1468-6996/14/1/014201
• 21)   S.  Evers,  C.  Senöz and  M.  Rohwerder: Spatially resolved high sensitive measurement of hydrogen permeation by scanning Kelvin probe microscopy, Electrochim. Acta 110 (2013) 534–538. doi:10.1016/j.electacta.2013.04.171
• 22)   B.  Gault et al.: Towards Establishing Best Practice in the Analysis of Hydrogen and Deuterium by Atom Probe Tomography, Microsc. Microanal. 30 (2024) 1205–1220. doi:10.1093/mam/ozae081
• 23)   J.  Takahashi,  K.  Kawakami,  Y.  Kobayashi and  T.  Tarui: The first direct observation of hydrogen trapping sites in TiC precipitation-hardening steel through atom probe tomography, Scr. Mater. 63 (2010) 261–264. doi:10.1016/j.scriptamat.2010.03.012
• 24)   Y.-S.  Chen,  P.-Y.  Liu,  R.  Niu,  A.  Devaraj,  H.-W.  Yen,  R.K.W.  Marceau and  J.M.  Cairney: Atom Probe Tomography for the Observation of Hydrogen in Materials: A Review, Microsc. Microanal. 29 (2023) 1–15. doi:10.1093/micmic/ozac005
• 25)   H.  Katayama,  T.  Katsumura,  T.  Akashi and  Y.  Tsutsumi: Hydrogen Entry Behavior on Steel Materials Exposed to Wet-dry Cyclic Corrosive Environment Using Surface Potential Measurement, ISIJ Int. 61 (2021) 1215–1221. doi:10.2355/isijinternational.ISIJINT-2020-552
• 26)   A.  Nazarov,  F.  Vucko and  D.  Thierry: Scanning Kelvin Probe for detection of the hydrogen induced by atmospheric corrosion of ultra-high strength steel, Electrochim. Acta 216 (2016) 130–139. doi:10.1016/j.electacta.2016.08.122
• 27)   R.F.  Schaller and  J.R.  Scully: Measurement of effective hydrogen diffusivity using the Scanning Kelvin Probe, Electrochem. Commun. 40 (2014) 42–44. doi:10.1016/j.elecom.2013.12.025
• 28)   R.F.  Schaller and  J.R.  Scully: Spatial determination of diffusible hydrogen concentrations proximate to pits in a Fe–Cr–Ni–Mo steel using the Scanning Kelvin Probe, Electrochem. Commun. 63 (2016) 5–9. doi:10.1016/j.elecom.2015.12.002
• 29)   D.  Rudomilova,  T.  Prošek and  M.  Ström: Hydrogen Entry into Steel Under Corrosion Products, Corrosion 77 (2021) 427–432. doi:10.5006/3675
• 30)   Y.  Sugawara and  H.  Saito: Improved Responsivity and Sensitivity of Hydrogen Mapping Technique in Pure Iron Using WO₃ Thin Film by Control of Pd Intermediate Layer, ISIJ Int. 61 (2021) 1201–1208. doi:10.2355/isijinternational.ISIJINT-2020-563
• 31)   Y.  Sugawara and  T.  Sato: Real-time and highly responsive hydrogen mapping in pure Fe using TiO₂ thin films, Int. J. Hydrogen Energ. 72 (2024) 237–246. doi:10.1016/j.ijhydene.2024.05.376
• 32)   S.  Ajito,  T.  Hojo,  M.  Koyama,  K.  Fujita and  E.  Akiyama: Application of an iridium complex for detecting hydrogen permeation through pure iron, Int. J. Hydrogen Energ. 45 (2020) 25580–25586. doi:10.1016/j.ijhydene.2020.06.113
• 33)   S.  Ajito,  H.  Kakinuma,  M.  Koyama and  E.  Akiyama: Visualization of Microscopic Hydrogen Diffusion in Polycrystalline Pure Ni Using an Ir Complex, Mater. Trans. 65 (2024) 1436–1441. doi:10.2320/matertrans.MT-C2024008
• 34)   H.  Kakinuma,  S.  Ajito,  T.  Hojo,  M.  Koyama and  E.  Akiyama: Real-Time Visualization of Hydrogen Distribution in Metals Using Polyaniline: An Ultrasensitive Hydrogenochromic Sensor, Adv. Mater. Interfaces 9 (2022) 2101984. doi:10.1002/admi.202101984
• 35)   H.  Kakinuma,  S.  Ajito,  T.  Hojo,  M.  Koyama,  S.  Hiromoto and  E.  Akiyama: In situ 2D mapping of hydrogen entry into an Fe sheet under a droplet of NaCl solution using a hydrogenochromic sensor, Int. J. Hydrogen Energ. 47 (2022) 38468–38476. doi:10.1016/j.ijhydene.2022.09.006
• 36)   H.  Kakinuma,  S.  Ajito,  T.  Hojo,  M.  Koyama,  S.  Hiromoto and  E.  Akiyama: Simultaneous observations of the corrosion behavior of an Fe sheet and the associated hydrogen distribution therein employing a hydrogenochromic sensor, Corros. Sci. 206 (2022) 110534. doi:10.1016/j.corsci.2022.110534
• 37)   H.  Kakinuma,  S.  Ajito,  T.  Hojo,  M.  Koyama and  E.  Akiyama: In situ visualization of misorientation-dependent hydrogen diffusion at grain boundaries of pure polycrystalline Ni using a hydrogen video imaging system, Acta Mater. 263 (2024) 119536. doi:10.1016/j.actamat.2023.119536
• 38)   H.  Kakinuma,  S.  Ajito,  M.  Koyama and  E.  Akiyama: Microscopic visualization of hydrogen diffusion in UNS S32750 super duplex stainless steel: Roles of crystal structure and grain size, Int. J. Hydrogen Energ. 59 (2024) 866–873. doi:10.1016/j.ijhydene.2024.02.069
• 39)   H.  Kakinuma,  S.  Hiromoto,  T.  Hojo,  S.  Ajito,  M.  Koyama and  E.  Akiyama: Mechanism of local hydrogen entry into Fe sheets induced by atmospheric corrosion: Significance of potential, pH, and rust layer thickness, Corros. Sci. 232 (2024) 112043. doi:10.1016/j.corsci.2024.112043
• 40)   H.  Kakinuma,  M.  Koyama,  R.S.  Varanasi,  S.  Ajito and  E.  Akiyama: Semi-quantitative characterisation of coherency-dependent hydrogen diffusion at Σ3 grain boundaries in pure Ni, Mater. Res. Lett. 13 (2025) 357–364. doi:10.1080/21663831.2025.2457369
• 41)   M.  Akhtar,  H.A.  Weakliem,  R.M.  Paiste and  K.  Gaughan: Polyaniline thin film electrochromic devices, Synth. Met. 26 (1988) 203–208. doi:10.1016/0379-6779(88)90237-8
• 42)   W.S.  Huang and  A.G.  MacDiarmid: Optical properties of polyaniline, Polymer 34 (1993) 1833–1845. doi:10.1016/0032-3861(93)90424-9
• 43)   T.  Kobayashi,  H.  Yoneyama and  H.  Tamura: Polyaniline film-coated electrodes as electrochromic display devices, J. Electroanal. Chem. 161 (1984) 419–423. doi:10.1016/S0022-0728(84)80201-6
• 44)   A.  Ray,  G.E.  Asturias,  D.L.  Kershner,  A.F.  Richter,  A.G.  MacDiarmid and  A.J.  Epstein: Polyaniline: Doping, structure and derivatives, Synth. Met. 29 (1989) 141–150. doi:10.1016/0379-6779(89)90289-0
• 45)   J.A.  Osaheni,  S.A.  Jenekhe,  H.  Vanherzeele,  J.S.  Meth,  Y.  Sun and  A.G.  MacDiarmid: Nonlinear optical properties of polyanilines and derivatives, J. Phys. Chem. 96 (1992) 2830–2836. doi:10.1021/j100186a010
• 46)   G.E.  Asturias and  A.G.  Macdiarmid: The oxidation state of “emeraldine” base, Synth. Met. 29 (1989) 157–162. doi:10.1016/0379-6779(89)90291-9
• 47)  H.-C. Lee: Color Imaging Science, (Cambridge University Press, Cambridge, 2005).
• 48)   S.  Ajito,  E.  Tada,  A.  Ooi and  A.  Nishikata: Hydrogen Absorption Behavior of Pre-Rusted Steels under an NaCl Droplet, J. Electrochem. Soc. 166 (2019) C243–C249. doi:10.1149/2.0591910jes
• 49)   E.  Akiyama,  S.  Li,  T.  Shinohara,  Z.  Zhang and  K.  Tsuzaki: Hydrogen entry into Fe and high strength steels under simulated atmospheric corrosion, Electrochim. Acta 56 (2011) 1799–1805. doi:10.1016/j.electacta.2010.09.043
• 50)   H.  Katayama: Visualization of Hydrogen Permeation Behavior through Steel Materials by Surface Potential Measurement, ISIJ Int. 56 (2016) 478–482. doi:10.2355/isijinternational.ISIJINT-2015-314
• 51)   E.  Akiyama,  M.  Wang,  S.  Li,  Z.  Zhang,  Y.  Kimura,  N.  Uno and  K.  Tsuzaki: Studies of Evaluation of Hydrogen Embrittlement Property of High-Strength Steels with Consideration of the Effect of Atmospheric Corrosion, Metall. Mater. Trans. A 44 (2013) 1290–1300. doi:10.1007/s11661-012-1403-2
• 52)   M.A.V.  Devanathan and  Z.  Stachurski: The adsorption and diffusion of electrolytic hydrogen in palladium, Proc. R. Soc. London, Ser. A 270 (1962) 90–102. doi:10.1098/rspa.1962.0205
• 53)   T.  Omura,  T.  Kudo and  S.  Fujimoto: Environmental Factors Affecting Hydrogen Entry into High Strength Steel due to Atmospheric Corrosion, Mater. Trans. 47 (2006) 2956–2962. doi:10.2320/matertrans.47.2956
• 54)   S.  Ajito,  E.  Tada,  A.  Ooi and  A.  Nishikata: Measurement of pH in a Thin Electrolyte Droplet Using the Kelvin Probe Technique, Mater. Trans. 60 (2019) 531–537. doi:10.2320/matertrans.M2018385
• 55)   S.  Ajito,  E.  Tada,  A.  Ooi and  A.  Nishikata: Simultaneous Measurements of Polarization Resistance and Hydrogen Permeation Current of Iron in an Aqueous NaCl Droplet, ISIJ Int. 61 (2021) 1222–1228. doi:10.2355/isijinternational.ISIJINT-2020-583
• 56)   S.  Ajito,  E.  Tada,  A.  Ooi and  A.  Nishikata: Simultaneous Measurements of Corrosion Potential and Hydrogen Permeation Current in Atmospheric Corrosion of Steel, ISIJ Int. 59 (2019) 1659–1666. doi:10.2355/isijinternational.ISIJINT-2019-113
• 57)   T.  Tsuru,  Y.  Huang,  M.R.  Ali and  A.  Nishikata: Hydrogen entry into steel during atmospheric corrosion process, Corros. Sci. 47 (2005) 2431–2440. doi:10.1016/j.corsci.2004.10.006
• 58)   C.  Hempel,  M.  Mandel,  C.  Schröder,  C.  Quitzke,  C.  Schimpf,  M.  Wendler,  O.  Volkova and  L.  Krüger: Influence of microstructure on hydrogen trapping and diffusion in a pre-deformed TRIP steel, Int. J. Hydrogen Energ. 48 (2023) 4906–4920. doi:10.1016/j.ijhydene.2022.11.017
• 59)   A.J.  Haq,  K.  Muzaka,  D.P.  Dunne,  A.  Calka and  E.V.  Pereloma: Effect of microstructure and composition on hydrogen permeation in X70 pipeline steels, Int. J. Hydrogen Energ. 38 (2013) 2544–2556. doi:10.1016/j.ijhydene.2012.11.127
• 60)   T.  Das,  R.  Chakrabarty,  J.  Song and  S.  Yue: Understanding microstructural influences on hydrogen diffusion characteristics in martensitic steels using finite element analysis (FEA), Int. J. Hydrogen Energ. 47 (2022) 1343–1357. doi:10.1016/j.ijhydene.2021.10.048
• 61)   H.  Iwaoka,  M.  Arita and  Z.  Horita: Hydrogen diffusion in ultrafine-grained palladium: Roles of dislocations and grain boundaries, Acta Mater. 107 (2016) 168–177. doi:10.1016/j.actamat.2016.01.069
• 62)   X.  Zhou,  N.  Mousseau and  J.  Song: Is Hydrogen Diffusion along Grain Boundaries Fast or Slow? Atomistic Origin and Mechanistic Modeling, Phys. Rev. Lett. 122 (2019) 215501. doi:10.1103/PhysRevLett.122.215501
• 63)   T.  Tsuru and  R.M.  Latanision: Grain boundary transport of hydrogen in nickel, Scr. Metall. 16 (1982) 575–578. doi:10.1016/0036-9748(82)90273-3
• 64)   A.  Oudriss,  J.  Creus,  J.  Bouhattate,  E.  Conforto,  C.  Berziou,  C.  Savall and  X.  Feaugas: Grain size and grain-boundary effects on diffusion and trapping of hydrogen in pure nickel, Acta Mater. 60 (2012) 6814–6828. doi:10.1016/j.actamat.2012.09.004
• 65)   A.M.  Brass and  A.  Chanfreau: Accelerated diffusion of hydrogen along grain boundaries in nickel, Acta Mater. 44 (1996) 3823–3831. doi:10.1016/1359-6454(95)00446-7
• 66)   Z.  Ma,  X.  Xiong and  Y.  Su: Study on hydrogen segregation at individual grain boundaries in pure nickel by scanning Kelvin probe force microscopy, Mater. Lett. 303 (2021) 130528. doi:10.1016/j.matlet.2021.130528
• 67)   R.  Dingreville and  S.  Berbenni: On the interaction of solutes with grain boundaries, Acta Mater. 104 (2016) 237–249. doi:10.1016/j.actamat.2015.11.017
• 68)   P.R.M.  van Beers,  V.G.  Kouznetsova,  M.G.D.  Geers,  M.A.  Tschopp and  D.L.  McDowell: A multiscale model of grain boundary structure and energy: From atomistics to a continuum description, Acta Mater. 82 (2015) 513–529. doi:10.1016/j.actamat.2014.08.045
• 69)   J.  Li,  A.  Hallil,  A.  Metsue,  A.  Oudriss,  J.  Bouhattate and  X.  Feaugas: Antagonist effects of grain boundaries between the trapping process and the fast diffusion path in nickel bicrystals, Sci. Rep. 11 (2021) 15533. doi:10.1038/s41598-021-94107-6
• 70)   M.  Sennour,  S.  Lartigue-Korinek,  Y.  Champion and  M.J.  Hÿtch: HRTEM study of defects in twin boundaries of ultra-fine grained copper, Philos. Mag. 87 (2007) 1465–1486. doi:10.1080/14786430601021611
• 71)   S.I.  Wright and  R.J.  Larsen: Extracting twins from orientation imaging microscopy scan data, J. Microsc. 205 (2002) 245–252. doi:10.1046/j.1365-2818.2002.00992.x
• 72)   D.G.  Brandon: The structure of high-angle grain boundaries, Acta Metall. 14 (1966) 1479–1484. doi:10.1016/0001-6160(66)90168-4
• 73)   S.  Amatsuka,  M.  Nishimoto,  I.  Muto,  M.  Kawamori,  Y.  Takara and  Y.  Sugawara: Micro-electrochemical insights into pit initiation site on aged UNS S32750 super duplex stainless steel, npj Mater. Degrad. 7 (2023) 15. doi:10.1038/s41529-023-00335-8
• 74)   X.Z.  Liang,  G.H.  Zhao,  M.F.  Dodge,  T.L.  Lee,  H.B.  Dong and  P.E.J.  Rivera-Díaz-del-Castillo: Hydrogen embrittlement in super duplex stainless steels, Materialia 9 (2020) 100524. doi:10.1016/j.mtla.2019.100524
• 75)   A.  Świerczyńska,  D.  Fydrych,  M.  Landowski,  G.  Rogalski and  J.  Łabanowski: Hydrogen embrittlement of X2CrNiMoCuN25-6-3 super duplex stainless steel welded joints under cathodic protection, Constr. Build. Mater. 238 (2020) 117697. doi:10.1016/j.conbuildmat.2019.117697
• 76)   Y.  Matsuo,  M.  Hatakeyama,  R.  Tsugawa,  K.  Makizaki and  S.  Sunada: Effects of σ Phase on the SCC Susceptibility and the Hydrogen Embrittlement Susceptibility of Super Duplex Stainless Steel F55, Mater. Trans. 66 (2025) 236–245. doi:10.2320/matertrans.MT-M2024134
• 77)   K.  Hirata,  S.  Iikubo,  M.  Koyama,  K.  Tsuzaki and  H.  Ohtani: First-Principles Study on Hydrogen Diffusivity in BCC, FCC, and HCP Iron, Metall. Mater. Trans. A 49 (2018) 5015–5022. doi:10.1007/s11661-018-4815-9
• 78)   Y.  He,  Y.  Li,  C.  Chen and  H.  Yu: Diffusion coefficient of hydrogen interstitial atom in α-Fe, γ-Fe and ε-Fe crystals by first-principle calculations, Int. J. Hydrogen Energ. 42 (2017) 27438–27445. doi:10.1016/j.ijhydene.2017.08.212
• 79)   T.  Omura,  H.  Sawada,  K.  Kobayashi and  Y.  Arai: Effects of Alloying Elements on Hydrogen Diffusion in Iron, ISIJ Int. 61 (2021) 1287–1293. doi:10.2355/isijinternational.ISIJINT-2020-301
• 80)   T.  Kushida and  T.  Kudo: Hydrogen diffusion and its effect on hydrogen embrittlement in martensitic steels, Defect Diffus. Forum 95–98 (1993) 359–364. doi:10.4028/www.scientific.net/DDF.95-98.359
• 81)   H.  Kakinuma,  I.  Muto,  Y.  Oya,  T.  Momii,  Y.  Sugawara and  N.  Hara: Improving the Pitting Corrosion Resistance of AA1050 Aluminum by Removing Intermetallic Particles during Conversion Treatments, Mater. Trans. 62 (2021) 1160–1167. doi:10.2320/matertrans.MT-M2021071
• 82)   K.  Kawaue,  M.  Hino,  R.  Kuwano,  Y.  Oda,  K.  Horikawa and  T.  Kanadani: Evaluation of Hydrogen Embrittlement of Three Aluminum Alloys by Three-Point Bending Test Affected by Three Types of Plating, Mater. Trans. 66 (2025) 165–170. doi:10.2320/matertrans.MT-L2024009
• 83)   A.  Suzuki,  H.  Yukawa,  T.  Nambu,  Y.  Matsumoto and  Y.  Murata: Anomalous Temperature Dependence of Hydrogen Permeability through Palladium–Silver Binary Alloy Membrane and Its Analysis Based on Hydrogen Chemical Potential, Mater. Trans. 57 (2016) 695–702. doi:10.2320/matertrans.M2016020
• 84)   J.  Okazaki,  D.A.P.  Tanaka,  M.A.L.  Tanco,  Y.  Wakui,  T.  Ikeda,  F.  Mizukami and  T.M.  Suzuki: Preparation and Hydrogen Permeation Properties of Thin Pd-Au Alloy Membranes Supported on Porous α-Alumina Tube, Mater. Trans. 49 (2008) 449–452. doi:10.2320/matertrans.MBW200720
• 85)   N.  Takata,  T.  Mizuguchi,  K.  Ikeda and  H.  Nakashima: Atomic and Electronic Structure of ⟨110⟩ Symmetric Tilt Boundaries in Palladium, Mater. Trans. 45 (2004) 2099–2105. doi:10.2320/matertrans.45.2099