Optically induced thermal demagnetization and switching of antiferromagnetic domains in NiO and CoO thin films

This study demonstrates that all-optical thermal manipulation using laser pulses can induce demagnetization, controlled domain wall motion, and reversible 90° switching of antiferromagnetic domains in NiO/Pt and CoO/Pt thin films via thermal gradients, offering a current-free pathway for ultrafast antiferromagnetic spintronics and non-volatile memory applications.

The Gist

The Big Idea: Erasing and Rewriting Magnetic Memories with Light

Imagine you have a very special, invisible map drawn on a piece of glass. This map isn't made of ink, but of tiny, invisible magnets called antiferromagnetic domains. These are the building blocks of future super-fast computer memory.

The problem? These magnets are "invisible" to normal tools, and they are incredibly stubborn. They don't like to move. For a long time, scientists thought you needed heavy electric currents or strong magnets to move them around to write data.

This paper says: "No, we can do it with just a laser pointer."

The researchers (Maciej Dąbrowski and his team) discovered a way to use a laser beam to erase these magnetic maps and redraw them, all without using electricity.

---

The Characters in Our Story

1. The Antiferromagnets (NiO and CoO): Think of these as a crowd of people holding hands in pairs. In every pair, one person faces North, and the other faces South. Because they are perfectly balanced, the whole crowd looks like they aren't facing any direction at all. They are "fully compensated." This makes them invisible to standard magnetic detectors but makes them perfect for storing data because they are stable and don't interfere with each other.
2. The Platinum (Pt) Layer: The researchers put a thin layer of metal (Platinum) on top of the magnetic material. Think of this as a solar panel. The laser light can't pass through the magnetic material easily, but it hits the solar panel, gets absorbed, and turns into heat.
3. The Laser Beam: This is our "magic wand." It's not just a flashlight; it's a precise tool that can heat up tiny spots on the material.

---

Experiment 1: The "Melting" Effect (Static Beam)

The Analogy: Imagine a frozen pond covered in ice sculptures (the magnetic domains). If you shine a hot spotlight on one spot and keep it there, the ice melts. When the water refreezes, the sculptures are gone, and you are left with a random, messy puddle of ice.

What happened in the lab:
When the researchers shone a stationary laser beam on the sample, the heat from the platinum layer transferred to the magnetic layer. This heat "melted" the organized magnetic order. When it cooled down, the magnetic domains didn't return to their original shape. Instead, they broke into tiny, random pieces.

• Result: They successfully erased the magnetic data, turning a structured map into a random mess. This is called "thermal demagnetization."

---

Experiment 2: The "Snowplow" Effect (Moving Beam)

The Analogy: Now, imagine you have a snowplow (the laser beam) moving across a snowy field. If you just park the plow in one spot, it melts a hole in the snow. But if you drive the plow, it pushes the snow aside, clearing a path and creating a new shape.

What happened in the lab:
The researchers realized that if they just held the laser still, they could only erase things. But if they swept the laser beam across the sample, something magical happened.

1. The Temperature Gradient: As the laser moves, it creates a "hot zone" right behind it and a "cold zone" ahead of it.
2. The Ponderomotive Force: This is a fancy physics term for a "push" caused by energy differences. Think of it like a ball rolling down a hill. The magnetic domains want to move from the hot, high-energy area to the cold, low-energy area.
3. The Switch: As the laser sweeps, it pushes the boundaries between the magnetic domains (the "domain walls"). It pushes them just enough to flip them 90 degrees.

The Magic Trick:
The most amazing part? They could flip the domains one way by sweeping the laser left-to-right, and then flip them back to the original state by sweeping the laser right-to-left.

• Result: They achieved all-optical switching. They didn't need electric currents. They just needed to change the direction the laser was moving.

---

Why This Matters (The "So What?")

Think of your current computer hard drive. It uses electricity to move magnetic heads, which is slow and generates a lot of heat.

This new method is like using a laser pointer to write directly onto the memory chip:

• Faster: Lasers can move incredibly fast (ultrafast).
• Efficient: No electric currents are needed, meaning less energy waste.
• Denser: Because the magnetic material is so stable, you can pack more data into a smaller space.

The Takeaway

The team proved that you don't need heavy machinery or electric wires to control the most stubborn types of magnetic memory. You just need a laser, a thin layer of metal to catch the heat, and a steady hand to sweep the beam.

They turned a "frozen" magnetic landscape into a malleable one, allowing them to erase and rewrite data using nothing but light. This opens the door to a new generation of computers that are faster, smaller, and use much less power.

Technical Summary

1. Problem Statement

While all-optical switching (AOS) of magnetization has been successfully demonstrated in ferromagnetic and ferrimagnetic materials, controlling antiferromagnets (AFMs) remains a significant challenge.

• Detection Difficulty: AFMs lack net magnetization and stray fields, making their domains difficult to image and manipulate.
• Excitation Limitation: NiO and CoO are wide-bandgap insulators. Direct optical excitation of their spin systems is inefficient because standard ultrafast laser photon energies are often below the material's bandgap.
• Control Mechanism: Existing methods often rely on electric currents (spin-orbit torques) or magnetic fields. There is a need for a purely optical, current-free method to manipulate AFM domains for high-speed, energy-efficient spintronics and data storage.

2. Methodology

The researchers employed a combination of advanced imaging and ultrafast laser manipulation techniques:

• Sample Fabrication: High-quality single-crystalline thin films of NiO and CoO were grown on MgO(001) substrates via Molecular Beam Epitaxy (MBE). Crucially, these films were capped with a thin (2 nm) Platinum (Pt) layer. The Pt layer acts as a metallic absorber for laser energy (since the AFMs are insulators) and protects against oxidation.
• Imaging Technique: Wide-field Kerr Microscopy (WFKM) utilizing the Magneto-Optical Birefringence (MOB) effect (Schäfer-Hubert effect).
  • A white-light LED source polarized at 45° relative to the Néel vector ($n$) was used.
  • Contrast was generated by taking the asymmetry of images recorded with the analyzer set symmetrically about extinction ($I_{as} = \frac{I(+\theta) - I(-\theta)}{I(+\theta) + I(-\theta)}$), enhancing sensitivity to the in-plane projection of the Néel vector while suppressing surface morphology artifacts.
• Optical Pumping:
  • A tunable 1035 nm laser was used as the pump source.
  • Static Beam: Focused to a 90 µm spot to induce thermal demagnetization.
  • Swept Beam: The laser beam was scanned across the sample surface at controlled velocities to induce domain wall motion.

3. Key Contributions

• All-Optical Control of Insulating AFMs: Demonstrated that insulating AFM domains (NiO, CoO) can be manipulated using only laser pulses, provided they are capped with a metallic layer to facilitate energy transfer via hot electrons.
• Identification of a New Switching Mechanism: Identified that thermal gradients generated by a moving laser beam exert a ponderomotive force on domain walls, driving their motion and enabling controlled switching.
• Reversible 90° Switching: Showed that the direction of the Néel vector can be toggled reversibly by simply reversing the direction of the laser beam sweep, without changing polarization or fluence.

4. Results

A. Thermal Demagnetization (Static Beam)

• Observation: Exposure to a static laser pulse (even a single pulse) causes thermal demagnetization.
• Effect: The original domain structure breaks down into smaller, randomly distributed domains. This is a relaxation toward an equilibrium state driven by magnetoelastic coupling to the substrate.
• Fluence Dependence: The size of the modified region scales with laser fluence. High fluence or multiple pulses can cause permanent sample damage, but lower fluence allows reversible modification.
• Material Differences: NiO films (higher Néel temperature, 500 K) showed smaller modified areas compared to CoO films (330 K) for the same fluence, indicating that the switching threshold is related to the material's magnetic ordering temperature.

B. Controlled Domain Switching (Swept Beam)

• Observation: When the laser beam is swept parallel to one of the easy axes of the Néel vector (e.g., [110] or [1$\bar{1}$0]), controlled 90° switching of AFM domains occurs.
• Mechanism: The moving beam creates a spatial temperature gradient. This gradient generates a spatial variation in magnetic energy density ($w_{mag} \propto M_s^2$).
• Ponderomotive Force: The gradient in magnetic energy exerts a force on the domain wall ($F_{pond} = -\nabla w_{mag}$), pushing the wall toward the colder region.
• Reversibility: By reversing the sweep direction, the thermal gradient reverses, allowing the domains to be switched back. This demonstrates true all-optical switching without electric currents.
• Velocity Dependence: The switching is highly dependent on the beam velocity.
  • Too slow/Static: The wall moves to a region of negligible force and stops (or is pinned).
  • Too fast: The wall enters a region where the force reverses direction, causing it to return to its initial position.
  • Optimal: The beam velocity must be comparable to the domain wall velocity to sustain motion and overcome pinning barriers.

C. Analytical Modeling

• The authors developed an analytical model using a Thiele-like equation for the domain wall center.
• The model incorporates a pinning force (magnetoelastic origin) and the ponderomotive force derived from the temperature gradient.
• The simulation successfully reproduced the experimental observation that steady motion and switching only occur at specific beam velocities, confirming the thermal nature of the mechanism and ruling out non-thermal effects like the inverse Faraday effect.

5. Significance

• Fundamental Physics: This work establishes a new physical mechanism for manipulating antiferromagnets: thermal pressure via ponderomotive forces. It proves that fully compensated insulating AFMs can be controlled optically.
• Technological Impact:
  • Spintronics: Offers a pathway for ultrafast, all-optical antiferromagnetic recording, potentially overcoming the speed and energy limitations of current-based spintronics.
  • Memory Technologies: The ability to switch domains without electric currents or magnetic fields suggests potential for high-density, non-volatile memory devices with low power consumption.
  • Scalability: The method works in continuous thin films and relies on standard laser scanning techniques, making it compatible with existing fabrication processes.

In conclusion, the paper demonstrates that by leveraging the thermal properties of a metallic capping layer and the physics of moving thermal gradients, researchers can achieve precise, reversible, and current-free control of antiferromagnetic domains in NiO and CoO.