Operando study of the evolution of peritectic structures in metal solidification by quasi-simultaneous synchrotron X-ray diffraction and tomography

By employing quasi-simultaneous synchrotron X-ray diffraction and tomography to analyze Al-Mn alloy solidification, this study elucidates the nucleation and co-growth dynamics of peritectic structures, revealing how Mn-rich diffusion layers govern phase transformations and how cooling rates can be tuned to control defect formation and morphology transitions.

The Gist

Imagine you are watching a high-speed, 3D movie of metal freezing, but instead of just seeing the outside shape, you can also see the invisible crystal structures and chemical ingredients moving inside. That is essentially what this paper does.

The researchers studied a specific metal mixture (Aluminum and Manganese) as it cooled down and turned from a liquid soup into a solid. They used a super-powerful "X-ray camera" (a synchrotron) to watch this happen in real-time, capturing a massive amount of data (about 30 terabytes!) to create a 4D map (3D space + time).

Here is the story of what they found, explained with some everyday analogies:

1. The Fast Lane vs. The Slow Lane (Anisotropic Growth)

When the metal started to cool, the first solid crystals to form were called Al4Mn. Think of these as the "pioneers."

• The Analogy: Imagine a pencil being sharpened. It grows very long and thin very quickly, but it gets wider very slowly.
• The Finding: These crystals grew about 70 times faster lengthwise (axially) than they did sideways (radially). They shot up like tall, thin towers or rods.
• Why? The atoms in the metal found it much easier to stack up in one direction (like adding books to a tall shelf) than to spread out sideways.

2. The Invisible "Moat" (The Diffusion Layer)

As these tall towers grew, they left a trail behind them.

• The Analogy: Imagine a construction crew building a wall. As they build, they leave a pile of extra bricks (Manganese atoms) right next to the wall, creating a thick, 5-micron-wide "moat" of concentrated material.
• The Finding: This "moat" is a layer where the Manganese concentration is very high. It acts as a barrier. It stops the tower from getting wider because the atoms get stuck in this pile, but it allows the tower to keep shooting upward.

3. The Second Wave (The Peritectic Reaction)

Once the temperature dropped a bit more, a second type of crystal (Al6Mn) started to form.

• The Analogy: Think of the first towers (Al4Mn) as a tree trunk. The second type of crystal (Al6Mn) grew like a thin, tight skin or shell wrapping around that trunk.
• The Connection: This new skin didn't just grow randomly; it grew in perfect alignment with the trunk underneath, like a glove fitting a hand. The researchers found a specific "handshake" rule between the two crystal structures that made them fit together perfectly.

4. The "Hollow Center" Mystery (Core Defects)

One of the most surprising discoveries was that these solid towers often had hollow tubes running right through their centers.

• The Analogy: Imagine a long, growing straw. As the straw gets longer, the liquid inside the very center gets "starved" of ingredients because the walls are growing so fast. Also, the heat generated by the freezing process gets trapped inside, melting the very center of the straw back into a tiny liquid channel.
• The Finding: Because the crystals grew so fast lengthwise, the center couldn't get enough Manganese to stay solid, and the trapped heat kept it liquid. This created a long, hollow tunnel inside the crystal. If the liquid ran out before the tunnel closed, it left a permanent hole or "core defect."

5. The Speed of Cooling Changes Everything

The researchers tested what happened if they cooled the metal at different speeds:

• Slow Cooling (The "Slow Cook"): The crystals had plenty of time to grow tall, thin, and perfect. They formed neat, faceted towers with long, hollow tunnels inside.
• Fast Cooling (The "Flash Freeze"): When they cooled the metal very quickly (like quenching hot metal in water), the "moat" of ingredients didn't have time to form.
  • The Result: The neat towers couldn't form. Instead, the metal turned into a messy, rugged, bush-like structure. The hollow tunnels disappeared because the freezing happened so fast that the "starvation" and "trapped heat" effects didn't have time to create the holes.

Summary

In simple terms, this paper shows that how metal freezes isn't random. It's a choreographed dance:

1. Tall towers grow first because they are the easiest path for atoms.
2. They leave a chemical barrier that stops them from getting wide.
3. A second layer wraps around them like a skin.
4. If they grow too fast, they leave hollow tunnels inside.
5. If you freeze it fast enough, you can stop the towers and tunnels from forming, resulting in a completely different, rougher shape.

This gives scientists a new "rulebook" for how to control the inside structure of metal alloys just by changing how fast they cool them down.

Technical Summary

Technical Summary: Operando Study of Peritectic Structures in Al-Mn Solidification

Problem Statement
Peritectic reactions, where a primary solid phase reacts with remaining liquid to form a second solid phase, are common in alloy solidification. These reactions often produce complex, composite-type microstructures with convoluted 3D morphologies. While the final properties of these alloys depend heavily on the dynamic evolution of these structures, there is a significant lack of real-time, operando studies characterizing these processes in 4D space (3D + time). Specifically, there is a gap in understanding the nucleation and co-growth dynamics of peritectic systems involving a transition from a hexagonal close-packed (HCP) primary phase to an orthorhombic secondary phase, such as in the Al-Mn system. Traditional 2D metallographic techniques fail to reveal the true 3D complexity and dynamic evolution of these intermetallic phases.

Methodology
The authors employed a quasi-simultaneous synchrotron X-ray diffraction and tomography technique at the DIAD beamline (K11) of the Diamond Light Source.

• Sample: An Al-8wt.%Mn alloy was prepared and cast into a cylindrical bar.
• Experimental Setup: The sample was placed in a capillary quartz tube within a resistance furnace. The setup allowed for simultaneous data acquisition:
  • Diffraction: A monochromatic beam (24.85 keV) scanned a 5×3 grid on the sample to identify crystal phases and orientations.
  • Tomography: A "pink" beam (20 keV) performed 3D imaging (3000 projections) to capture morphological evolution.
• Conditions: The study covered solidification from the melt down to room temperature. Experiments were conducted at a baseline cooling rate of ~0.17 °C/s (10 °C/min), with additional comparative studies at ~2.0 °C/s and ~20 °C/s.
• Data Processing: Approximately 30 TB of 4D datasets were generated. Data were processed using DAWN, Fit2D, and SAVU pipelines for diffraction and tomography reconstruction, respectively. Post-processing included phase indexing, 3D rendering (Avizo), and EBSD/EDXS analysis on fully solidified samples to confirm crystallographic relationships.

Key Results

1. Anisotropic Growth of Primary Al4Mn: The primary Al4Mn phase (HCP structure) nucleated and grew with extreme kinetic anisotropy. Growth along the axial direction (<0001>) was approximately 70 times faster than in the radial direction (<10$\bar{1}$0>), resulting in long hexagonal prisms.
2. Solute Diffusion Layer: A ~5 µm Mn-rich diffusion layer formed at the liquid-solid interface of the growing Al4Mn. This layer created a sharp local solute gradient, with Mn enrichment in the solid and depletion in the adjacent liquid. This layer governed the subsequent peritectic transformation.
3. Peritectic Reaction and Orientation: The secondary Al6Mn phase (orthorhombic) nucleated epitaxially within the Mn-rich diffusion zone, forming a thin shell around the Al4Mn core. A specific orientation relationship was established: {10$\bar{1}$0}$_{HCP}$ // {110}$_{O}$ and [0001]$_{HCP}$ // [001]$_{O}$.
4. Core Defects: Both Al4Mn and Al6Mn phases developed distinct tubular core defects along their growth axes. These defects were attributed to anisotropic solute diffusion (depletion at the center of the growth front), mechanical strain from impingement with neighboring crystals, and the inability to dissipate latent heat from the inner crystal surfaces, leading to localized remelting.
5. Cooling Rate Effects:
   • At 0.17 °C/s, the system formed large, well-defined faceted prisms with continuous tubular defects.
   • Increasing the cooling rate to 2.0 °C/s disrupted the stability of the solute diffusion layer, leading to partly faceted, chain-like structures with fragmented defects.
   • At 20 °C/s, the solute diffusion zone was effectively suppressed. This forced independent nucleation of Al6Mn from the undercooled melt, resulting in refined, non-faceted, rugged dendritic structures with largely suppressed core defects.

Significance and Claims
The paper claims to provide the first systematic in-situ and operando study of peritectic reactions involving an HCP-to-orthorhombic transition in 4D space. By capturing ~30 TB of real-time data, the authors elucidate the deterministic sequential process that leads to complex peritectic microstructures, rather than viewing them as random occurrences.

The study establishes a new theoretical framework for tailoring peritectic structures in metallic alloys. It demonstrates that the final microstructure is controlled by the interplay between intrinsic crystallographic kinetics (anisotropy) and external processing parameters (cooling rate). Specifically, the research highlights that the ~5 µm solute-rich boundary layer is critical for initiating epitaxial growth and orientation relationships, while the cooling rate acts as a paramount parameter to switch the growth mode from faceted to non-faceted and to suppress or induce core defects. These findings offer a quantitative foundation for controlling the architecture and properties of peritectic alloys through precise solidification kinetics.