The interplay between thermomigration and stress-driven hydrogen transport in metals

This paper presents a thermodynamically consistent finite element framework to quantify the competing and synergistic effects of thermomigration and stress-driven transport on hydrogen redistribution in metals, demonstrating that while thermomigration often dominates in heat-carrying components, stress-driven transport becomes decisive near sharp stress concentrators, thereby offering practical guidance for assessing embrittlement risks.

The Gist

Imagine you are trying to move a crowd of people (hydrogen atoms) through a crowded city (a metal structure). Usually, people move away from crowded areas toward empty ones. But in the world of metals, there are two invisible "wind forces" that can push these people around, often fighting against each other.

This paper is about figuring out which wind is stronger: the Heat Wind or the Stress Wind.

The Two Invisible Winds

1. The Stress Wind (The "Crowded Room" Effect):
   Imagine a room where the floor is uneven. People naturally want to move to the flat, comfortable spots. In a metal, "stress" is like that uneven floor. If a part of the metal is being squeezed or pulled hard (high stress), hydrogen atoms usually want to gather there or move away from it, depending on the material. This is well-known to engineers.

2. The Heat Wind (The "Thermomigration" Effect):
   This is the new, less understood force. Imagine a hallway where one end is freezing cold and the other is a hot sauna.

   • In some metals (like Iron and Nickel), the hydrogen atoms act like shy introverts who hate the heat. They run away from the hot side and hide in the cold side.
   • In other metals (like Zirconium, used in nuclear reactors), the hydrogen atoms act like sunbathers. They love the heat and run toward the hot side.
   • This movement caused purely by temperature differences is called Thermomigration.

The Big Question

For decades, engineers mostly worried about the Stress Wind. They thought, "If there's a crack or a sharp corner causing high stress, that's where the hydrogen will go and cause the metal to break."

But this paper asks: What if the Heat Wind is actually stronger?

The Experiments (The Case Studies)

The researchers built digital models of two real-world scenarios to see which wind wins:

Scenario A: The Heat Exchanger (Iron and Nickel)
Think of a heat exchanger as a metal sandwich. One side is blasted with hot air, the other with cold hydrogen gas.

• The Setup: The metal wants to expand on the hot side and stay small on the cold side. Because it's stuck together, this creates a lot of internal stress (the Stress Wind).
• The Result: Even though the stress was high, the Heat Wind was the boss.
  • In Iron and Nickel, the hydrogen atoms ran away from the hot side and piled up on the cold side.
  • The Surprise: The hydrogen didn't go to the "danger zones" (high stress areas) as expected. Instead, the temperature gradient swept them all to the cold side. This means the metal might be safer from stress-induced breaking in those specific spots, but the cold side might become dangerously saturated with hydrogen.

Scenario B: The Nuclear Fuel Rod (Zirconium)
Nuclear fuel rods are thin tubes of Zirconium. They get incredibly hot on the inside and cooler on the outside.

• The Setup: Zirconium is special because its hydrogen atoms love the heat (they run toward the hot center).
• The Twist: The researchers added a "notch" (a tiny scratch or crack) to the tube.
• The Result:
  • Smooth Tube: The Heat Wind won. Hydrogen moved toward the hot center.
  • Notched Tube: Near the sharp scratch, the Stress Wind became incredibly strong. It overpowered the Heat Wind. The hydrogen got sucked into the crack tip.
  • Why it matters: This explains why nuclear rods sometimes crack (Delayed Hydride Cracking). The stress at the crack tip is so intense that it pulls the hydrogen right into the danger zone, causing the crack to grow.

The "Traffic Light" Tool

The most practical part of this paper is a new Graphical Method (a simple chart).

Imagine you are an engineer designing a new engine part. You don't want to run a super-complex, days-long computer simulation just to check if hydrogen will move.

• This paper gives you a Traffic Light Chart.
• You just plug in your material's properties and your temperature/stress numbers.
• Green Light: The Heat Wind wins. (Hydrogen moves based on temperature).
• Red Light: The Stress Wind wins. (Hydrogen moves based on cracks and stress).
• Yellow Zone: They are fighting, and you need a closer look.

Why Should You Care?

This research changes how we design things that get hot and cold, like:

• Green Energy: Future hydrogen-powered airplanes will need to store liquid hydrogen (very cold) and heat it up to burn it. This paper helps engineers predict where hydrogen will hide in the pipes so they don't break.
• Nuclear Safety: It helps predict exactly when and where nuclear fuel rods might crack, making reactors safer.
• General Safety: It tells us that sometimes, the "obvious" danger (stress) isn't the problem; the "invisible" danger (temperature) is actually moving the dangerous atoms around.

The Bottom Line

For a long time, we thought stress was the only thing that moved hydrogen in metals. This paper says, "Wait a minute! Temperature is often the stronger force."

However, if you have a sharp crack or a notch, stress takes over again. By using their new "Traffic Light" chart, engineers can quickly figure out which force is winning and design safer, longer-lasting machines.

Technical Summary

1. Problem Statement

Hydrogen embrittlement (HE) is a critical failure mechanism in structural metals, driven by the redistribution and trapping of hydrogen (H) at microstructural defects (e.g., dislocations, grain boundaries). While stress-driven hydrogen migration is well-understood, thermomigration (hydrogen transport driven by temperature gradients) remains underexplored due to a lack of mechanistic models and experimental data.

In modern engineering applications—such as hydrogen-powered aviation heat exchangers and nuclear fuel cladding—components operate under complex thermomechanical loading. These environments feature simultaneous temperature gradients and thermal incompatibility stresses. The central challenge is determining which driving force (stress gradients vs. temperature gradients) dominates hydrogen redistribution, as this dictates where hydrogen accumulates and where embrittlement risk is highest. Current models often rely on empirical data or fail to account for the temperature dependence of the "heat of transport" ($Q^*$).

2. Methodology

The authors developed a thermodynamically consistent framework for hydrogen transport that integrates a mechanistic model for thermomigration into a finite element (FE) simulation environment.

• Thermodynamic Framework: The study utilizes a non-equilibrium thermodynamic approach based on Onsager reciprocal relations. Instead of relying solely on the chemical potential gradient ($\nabla\mu$), the model incorporates the heat of transport ($Q^*$) to define an effective chemical potential ($\mu_{eff}$).
  • The hydrogen flux ($\bar{J}_H$) is expressed as: $\bar{J}_H = -D_L \bar{C}_L \nabla(\mu_{eff}/T)$.
  • $\mu_{eff}$ combines contributions from stress ($\sigma_H$), lattice occupancy ($\theta_L$), and a temperature-dependent function ($f$) derived from $Q^*$.
• Mechanistic Heat of Transport Model: The paper employs a previously developed model where $Q^*$ is the sum of three contributions:
  1. Intrinsic ($Q^*_{int}$): Related to vibrational enthalpy differences.
  2. Electrostatic ($Q^*_{es}$): Interaction between H and thermoelectric fields.
  3. Electron-wind ($Q^*_{ele}$): Momentum transfer from electron scattering.
  • Crucially, the model accounts for the strong temperature dependence of $Q^*$, fitting it to a second-order polynomial for implementation.
• Finite Element Implementation: The authors adapted the Abaqus UMATHT subroutine (typically used for thermal transport) to solve for hydrogen transport. By treating $\mu_{eff}/T$ as the primary degree of freedom (analogous to temperature in heat transfer), they efficiently modeled coupled stress-temperature effects without requiring complex second-order displacement derivatives.
• Graphical Analysis Tool: A simplified analytical criterion was derived to rapidly identify the dominant transport mechanism without full simulation:
  $$\left| \frac{\nabla \sigma_H}{\nabla T} \right| \approx \left| \frac{Q^*}{V_L T} \right|$$
  If the ratio of stress gradient to temperature gradient exceeds the material-specific threshold ($Q^*/V_L T$), stress-driven transport dominates; otherwise, thermomigration dominates.

3. Key Contributions

1. Unified Modeling Framework: A robust, thermodynamically consistent FE implementation that couples stress-driven and thermomigration effects using an effective chemical potential.
2. Mechanistic $Q^*$ Model: Integration of a temperature-dependent, mechanistic heat of transport model (incorporating intrinsic, electrostatic, and electron-wind effects) rather than relying on single-point empirical measurements.
3. Engineering Screening Tool: Introduction of a computationally efficient graphical method to predict the dominant hydrogen transport regime based on material properties ($E, \alpha_{th}, Q^*$) and loading conditions.
4. Decoupled Analysis Strategy: A practical workflow where thermal and mechanical fields are solved separately and mapped to the hydrogen transport model, significantly reducing computational cost while maintaining accuracy for redistribution studies.

4. Results

The framework was applied to three case studies: Iron and Nickel heat exchangers, and Zirconium alloy nuclear fuel cladding.

• Iron and Nickel Heat Exchangers:

  • Dominance of Thermomigration: In both materials, thermomigration was the dominant driving force for hydrogen redistribution, even in the presence of significant thermal incompatibility stresses.
  • Counter-Intuitive Redistribution: In Iron (negative $Q^*$), hydrogen migrated from cold to hot regions, accumulating in the hot channel and depleting the cold channel. This occurred despite the presence of high tensile stresses in the cold region (which would typically attract hydrogen via stress-driven migration).
  • Material Differences: Nickel showed less redistribution than Iron due to a lower thermal expansion coefficient and lower partial molar volume of hydrogen, but thermomigration remained dominant.
  • Graphical Validation: The graphical method correctly predicted that the stress-to-temperature gradient ratios in these geometries fell well within the "temperature-dominated" regime.

• Zirconium Alloy Fuel Cladding:

  • Synergistic Effects: In smooth cladding sections, both stress gradients (due to thermal expansion) and thermomigration (positive $Q^*$ in Zr) drove hydrogen outward toward the cooler outer surface, acting synergistically.
  • Stress Concentrators: When a notch (representing hydride blistering or damage) was introduced, the local hydrostatic stress gradient increased dramatically. The graphical analysis showed that near the notch root, the system shifted from temperature-dominated to stress-dominated transport.
  • Implication for DHC: This confirms that while thermomigration governs bulk redistribution, stress-driven transport becomes decisive at sharp stress concentrators, which is critical for understanding Delayed Hydride Cracking (DHC).

5. Significance and Implications

• Risk Assessment: The study reveals that in heat-carrying components, thermomigration can override stress-driven accumulation. This means hydrogen may be drawn away from high-stress tensile regions (reducing local embrittlement risk) but simultaneously drawn toward hot regions, potentially increasing total hydrogen uptake from the environment.
• Design Guidance: The proposed graphical method allows engineers to rapidly assess whether a component's hydrogen redistribution will be governed by thermal gradients or stress concentrations without running expensive, fully coupled simulations.
• Failure Mechanisms: The findings highlight that while thermomigration dominates in smooth geometries, stress-driven transport is the critical factor near defects (cracks, notches). This distinction is vital for predicting the initiation and growth of cracks in hydrogen-exposed infrastructure.
• Future Directions: The authors note that while isotropic assumptions were sufficient for this study, future work must address crystallographic anisotropy (particularly in textured Zirconium alloys) and the coupling of internal transport with surface absorption/desorption boundary conditions.

In summary, this paper provides a critical advancement in understanding hydrogen behavior in non-isothermal environments, demonstrating that thermomigration is often the primary driver of hydrogen redistribution in structural components, except in the immediate vicinity of severe stress concentrators.