How a Single Layer of CrI3 Becomes a Magnet

This guide explains how a single atomic layer of chromium triiodide (CrI3) becomes a magnet, a phenomenon that was first demonstrated in 2017. You will learn the physical mechanism behind 2D magnetism, the key experimental steps to isolate and measure the magnetic properties of a monolayer, and the common pitfalls to avoid. No special equipment is needed to understand the concept, but if you are planning to reproduce the experiment, you will need access to an advanced materials science laboratory.

What You’ll Need

• Bulk CrI3 crystals (grown via chemical vapor transport)
• Scotch tape or a similar adhesive tape for mechanical exfoliation
• An optical microscope with a camera
• A silicon wafer with a 300 nm oxide layer (for substrate)
• An atomic force microscope (AFM) or a profilometer to confirm monolayer thickness
• A magneto-optical Kerr effect (MOKE) microscope or a superconducting quantum interference device (SQUID) magnetometer
• Inert atmosphere glovebox (to prevent degradation)
• Raman spectrometer (optional, for quality confirmation)
• Clean tweezers and a clean workspace

Step-by-Step

1. Obtain high-quality bulk CrI3 crystals. CrI3 is a layered material, similar to graphite, where individual layers are held together by weak van der Waals forces. The bulk crystal must be pure and free of defects; otherwise, the monolayer will not show intrinsic magnetism.
2. Prepare a clean substrate. Use a silicon wafer with a 300 nm thermal oxide layer. Clean it with acetone and isopropyl alcohol in an ultrasonic bath, then blow-dry with nitrogen. The surface must be dust-free to allow clean exfoliation.
3. Exfoliate the CrI3. Place a piece of bulk CrI3 on the sticky side of a piece of Scotch tape. Fold the tape over and peel it apart repeatedly, about 5-10 times, to thin the crystal. The goal is to produce flakes that are only a few atomic layers thick.
4. Transfer the flakes to the substrate. Press the tape with the thin flakes onto the cleaned silicon wafer. Gently rub the back of the tape with a soft tool (e.g., a plastic spatula) to ensure good contact, then slowly peel the tape away. Some flakes will remain on the substrate.
5. Identify a monolayer flake. Use an optical microscope to scan the substrate. Monolayer CrI3 appears as a very faint, semi-transparent flake with a slight purple/blue contrast on the 300 nm oxide. Compare the optical contrast with known images or use an AFM to measure the thickness. A monolayer should be about 0.7 nm thick.
6. Protect the flake (if needed). CrI3 degrades in air and under light. If you are not measuring immediately, store the sample in an inert atmosphere (argon or nitrogen glovebox) and keep it in the dark. For measurements, you can encapsulate the flake with a thin layer of hexagonal boron nitride (hBN) to prevent degradation.
7. Measure the magnetic properties. Place the sample in a MOKE microscope or a SQUID magnetometer. Cool the sample to below the Curie temperature (about 45 K for monolayer CrI3). Apply a magnetic field perpendicular to the plane of the flake and measure the Kerr rotation (MOKE) or the magnetization (SQUID) as a function of field.
8. Analyze the hysteresis loop. If the monolayer is ferromagnetic, you will observe a hysteresis loop in the magnetization vs. field curve. The presence of a nonzero remanent magnetization (magnetization at zero field) and a coercive field (the field needed to bring magnetization to zero) confirms that the layer is magnetically ordered.

Common Mistakes

• Confusing monolayer with bilayer. A bilayer CrI3 is antiferromagnetic (not ferromagnetic), so it shows no net magnetization. Always confirm the layer thickness with AFM or Raman spectroscopy before measuring.
• Oxidation of the sample. CrI3 is highly air-sensitive. If the flake is exposed to air for more than a few minutes, it will degrade and lose its magnetic properties. Work quickly and use a glovebox whenever possible.
• Using the wrong substrate. The optical contrast of a monolayer depends heavily on the substrate. A 300 nm oxide layer is standard, but if you use a different substrate, the flake may be invisible or hard to identify.
• Measuring at too high a temperature. The Curie temperature of monolayer CrI3 is around 45 K. If you measure above this temperature, the material is paramagnetic, and you will not see a hysteresis loop.
• Ignoring the magnetic field direction. CrI3 has strong magnetic anisotropy, meaning it prefers to magnetize along the out-of-plane direction. If you apply the field in-plane, you will not see a clear hysteresis loop.

Tips

• Use fresh tape. The sticky residue from used tape can contaminate the flake and affect its properties.
• Look for uniform color. A good monolayer flake will have a uniform optical contrast. If you see patches of different contrast, the flake is likely non-uniform or has multiple layers.
• Encapsulate with hBN. Encapsulating the CrI3 flake with a few layers of hBN not only protects it from air but also reduces the influence of the substrate, which can slightly alter the magnetic properties.
• Use a low laser power in MOKE. The laser used in MOKE can heat the sample and potentially damage it. Keep the laser power low to avoid local heating that could shift the Curie temperature.
• Calibrate the MOKE signal. Before measuring the monolayer, measure a bulk CrI3 flake or a known ferromagnetic film to calibrate the Kerr rotation signal.

Troubleshooting

• No flake found on the substrate. This is common. Try exfoliating again with a fresh piece of tape and a new area of the bulk crystal. Press the tape onto the substrate with more pressure and for a longer time.
• The flake is too thick. If you see a flake but it is too thick, you can try to thin it further by pressing the tape onto the flake and peeling it off again. Alternatively, start the exfoliation process again with more peeling steps.
• No hysteresis loop observed. First, check the temperature. If it is above 45 K, cool down further. Then verify the layer thickness—if it is a bilayer, it will be antiferromagnetic and show no net magnetization. Also, ensure the magnetic field is applied perpendicular to the layer.
• Signal is too noisy. Increase the acquisition time of the MOKE or SQUID measurement. Also, ensure that the laser spot is focused on the flake and not on the substrate.
• The flake degrades during measurement. If the Kerr rotation signal decreases over time, the flake is likely degrading. Use a vacuum chamber or continuous inert gas flow over the sample during measurement.

Frequently Asked Questions

Why is CrI3 magnetic only when it is a single layer?

Bulk CrI3 is a ferromagnet, but the magnetic ordering is highly sensitive to layer thickness. In a bilayer, the interlayer coupling is antiferromagnetic, which cancels out the net magnetization. In a monolayer, there is no interlayer coupling, so the intralayer ferromagnetic order dominates, resulting in a net magnetic moment.

What is the Curie temperature of monolayer CrI3?

The Curie temperature of monolayer CrI3 is approximately 45 K (about -228 °C). This is well below room temperature, so the material must be cooled to observe ferromagnetism.

Can I make a monolayer of CrI3 at home?

No. The fabrication and measurement require advanced equipment such as an AFM, a glovebox, and a MOKE or SQUID system. The material is also highly air-sensitive, so it must be handled in an inert atmosphere.

What is the role of the substrate?

The substrate provides mechanical support and affects the optical contrast for identification. It can also influence the magnetic properties slightly through strain or charge transfer. Using a 300 nm SiO2 layer is standard for optical contrast, but other substrates like hBN are used for better isolation.

How do I know if I have a monolayer?

The most reliable method is atomic force microscopy (AFM) to measure the thickness. A monolayer of CrI3 has a thickness of about 0.7 nm. Optical contrast can also be used as a quick check, but it is not definitive.

Why is the magnetic field applied perpendicular to the layer?

CrI3 has strong magnetic anisotropy, meaning that the magnetic moments prefer to align along the out-of-plane direction. Applying the field in-plane would require a much larger field to rotate the moments, and the hysteresis loop would be less pronounced.

Related

• Google TV Default Launcher vs Lightweight Launcher: What Changes?
• Hyperscale Data Centers: Benefits & How They Work
• Cloudflare vs AWS Shield DDoS Protection Compared