Evidence of Ultrashort Orbital Transport in Heavy Metals Revealed by Terahertz Emission Spectroscopy

Using terahertz emission spectroscopy on wedge-shaped heavy metal|Ni heterostructures, this study provides the first direct experimental evidence that orbital mean free paths in heavy metals are ultrashort (sub-nanometer scale) and shorter than spin counterparts, confirming that bulk inverse orbital Hall effect governs orbital-to-charge conversion.

The Gist

Here is an explanation of the paper using simple language, analogies, and metaphors.

The Big Idea: A New Kind of "Traffic" in Electronics

Imagine you are trying to build a super-fast, energy-efficient computer. For decades, engineers have used electron spin (a tiny magnetic property of electrons) to carry information. This field is called Spintronics. Think of spin as a tiny arrow that can point "Up" or "Down" to represent 1s and 0s.

But recently, scientists discovered another property called orbital angular momentum (or just "orbit"). If spin is like a spinning top, the orbit is like the top moving in a circle around a table. This "orbit" can carry more information and energy than spin, potentially making computers faster and cooler. This new field is called Orbitronics.

The Problem:
There was a huge argument in the scientific community.

• Team A (The Optimists): Said, "Orbits can travel very far! Like a marathon runner, they can go 50 or 80 nanometers (very far in the atomic world) before stopping."
• Team B (The Skeptics): Said, "No way! The crystal structure of metals is too messy. Orbits should stop almost immediately, within a single layer of atoms (less than 1 nanometer)."

This paper is the referee that finally settles the score.

---

The Experiment: The "Wedge" Slide

To solve this, the researchers from Fudan University built a special experiment.

The Analogy: The Sandcastle Ramp
Imagine you are building a sandcastle. Usually, you build a flat wall of sand. But to test how far water can soak through, you build a ramp (a wedge) that goes from very thin (a few grains of sand) to very thick (a whole bucket of sand) all in one piece.

• The Setup: They made a "sandcastle" of heavy metal (Tungsten) that gets thicker and thicker in a smooth slide.
• The Trigger: They hit this ramp with a super-fast laser pulse (like a tiny, high-speed hammer). This wakes up the electrons, sending their "spin" and "orbit" signals racing through the metal.
• The Detector: As these signals race through the metal, they create a burst of invisible "Terahertz" waves (like a radio signal). The researchers measured how strong this signal was at every point on the ramp.

The Discovery: The "Short-Lived" Orbit

When they looked at the data, they found something surprising that Team A (the Optimists) missed because they were only looking at the "thick" parts of the ramp.

The Metaphor: The Sprinter vs. The Marathoner

• Spin (The Marathoner): When the laser hit the metal, the "spin" signal traveled quite far. It kept going until the metal was about 2.2 nanometers thick. It was a steady, reliable runner.
• Orbit (The Sprinter): The "orbit" signal was different. It was incredibly fast but exhausted almost instantly. It stopped running after only 0.36 nanometers.

0.36 nanometers is tiny. It's roughly the size of one single layer of atoms.

The researchers found that in the very thin parts of the ramp (less than 3 nanometers), the "orbit" signal was actually doing something different than the "spin" signal. It was so short-lived that it couldn't travel across the metal like a marathon runner. It was more like a spark that fizzled out immediately.

Why Did People Get It Wrong Before?

The paper explains that previous studies were like looking at a forest from far away. You see the trees (the long-distance signals), but you miss the details of the ground right under your feet.

1. The "Interference" Effect: When the researchers looked closely at the thin metal, they saw that the "spin" and "orbit" signals were fighting each other. They had opposite directions (like two people pushing a car from opposite sides). In the thin layers, they canceled each other out, creating a weird dip in the signal. Previous studies missed this cancellation and thought the signal was just "long."
2. The Interface Trap: Some scientists thought the signal was coming from the surface of the metal (like a ripple on a pond). The researchers tested this by putting a "spacer" layer (like a piece of plastic) between the metal layers. If the signal was a surface ripple, the plastic would stop it. But the signal didn't stop. This proved the signal was coming from inside the metal, not the surface.

The Conclusion: What Does This Mean?

The Verdict:
Orbits in heavy metals (like Tungsten) are not long-distance runners. They are local phenomena. They don't travel across the material; they stay right where they are born, within a single atomic layer.

The Takeaway:

• For Engineers: If you want to use "orbit" to build new computers, you can't rely on it traveling long distances like spin does. You have to design your devices to be incredibly thin and precise, working with the "local" nature of the orbit.
• For Science: This paper clears up a major confusion. It shows that while "spin" can travel far, "orbit" is a short-range, high-intensity burst.

In a Nutshell:
The researchers used a special "ramp" of metal to catch electrons in the act. They discovered that while electron "spin" is a marathon runner that can go the distance, electron "orbit" is a sprinter who stops after a single step. This changes how we think about building the next generation of ultra-fast electronics.

Technical Summary

Here is a detailed technical summary of the paper "Evidences of Sub-Nanometer Orbital Diffusion Lengths in Heavy Metals using Terahertz Emission Spectroscopy."

1. Problem Statement

The field of orbitronics aims to utilize electron orbital angular momentum ($L$) for information processing, offering potential advantages over spintronics due to larger angular momentum. However, a significant controversy exists regarding the orbital diffusion length ($\lambda_L$) in heavy metals (HMs):

• Experimental Claims: Recent studies using terahertz (THz) emission spectroscopy suggest long-range orbital transport, with diffusion lengths ranging from 9 nm to 80 nm. These studies attribute THz generation to interfacial orbital-to-charge conversion mechanisms, specifically the Inverse Orbital Rashba-Edelstein Effect (IOREE).
• Theoretical Predictions: Conversely, recent first-principles calculations suggest that orbital propagation is constrained by crystal symmetry to sub-atomic-layer scales (sub-nanometer), contradicting the experimental claims of long-range transport.
• The Gap: Existing experimental techniques struggle to disentangle spin and orbital contributions in thin films, and previous measurements often lack the sub-nanometer resolution required to probe the critical thickness regime (<3 nm) where orbital signatures are expected to dominate or vanish.

2. Methodology

To resolve this contradiction, the authors employed a high-resolution experimental approach combined with a multi-component theoretical model:

• Wedge-Sample Platform: The researchers fabricated wedge-shaped heterostructures (HM|FM, where HM = W, Ta, Pt and FM = Ni, Fe) on $Al_2O_3$ substrates.
  • The Heavy Metal (HM) layer thickness ($d_{HM}$) varies continuously along the in-plane direction with a gradient of ~1.6 nm/mm.
  • The Ferromagnetic (FM) layer thickness is fixed at 5 nm.
  • Resolution: By scanning the sample with a 50 $\mu$m laser spot in 0.1 mm steps, they achieved a **sub-nanometer thickness resolution (0.16 nm)**, allowing continuous scanning from 0 to 15 nm.
• Terahertz Emission Spectroscopy: Upon femtosecond laser excitation, ultrafast spin ($j_S$) and orbital ($j_L$) currents are generated in the FM layer and injected into the HM layer. These are converted into transverse charge currents via the Inverse Spin Hall Effect (ISHE) and Inverse Orbital Hall Effect (IOHE), emitting THz radiation.
• Material Selection:
  • Ni vs. Fe: Ni is chosen for its high orbital current generation efficiency, while Fe is used as a reference for dominant spin current generation.
  • Heavy Metals: W, Ta, and Pt were selected due to their distinct signs of spin ($\gamma_{SH}$) and orbital ($\gamma_{OH}$) Hall angles, allowing for the separation of contributions based on polarity.
• Control Experiments: To rule out interfacial mechanisms (IOREE/ISREE), 1-nm and 2-nm Cu or Al spacer layers were inserted at the HM/FM and substrate/HM interfaces.
• Multi-Component Terahertz Emission Model (MC-TEM): The authors developed a model treating the total THz emission as a superposition of three components:
  1. Spin contribution from HM ($A_S$).
  2. Orbital contribution from HM ($A_L$).
  3. Intrinsic contribution from the FM (Ni) layer ($A_{Ni}$).
     The model incorporates optical absorption, electrical shunting, and interfacial reflections, using a diffusive transport formalism for both spin and orbital currents.

3. Key Contributions

• Sub-Nanometer Resolution: The study is the first to systematically map THz emission in the 0–3 nm thickness regime with sub-nanometer precision, a range previously inaccessible to standard discrete thin-film samples.
• Disentanglement of Spin and Orbital Dynamics: By combining the wedge-platform with the MC-TEM and utilizing materials with opposing Hall angle signs, the authors successfully separated the orbital contribution from the spin contribution without relying on X-ray or electron microscopy.
• Exclusion of Interfacial Mechanisms: Through rigorous spacer-layer control experiments, the study definitively rules out the Inverse Orbital Rashba-Edelstein Effect (IOREE) as the dominant mechanism for the observed THz signals in the thin-film regime.
• Unified Model: The establishment of the MC-TEM provides a robust framework for semi-quantitatively extracting orbital diffusion lengths ($\lambda_L$) and distinguishing them from spin diffusion lengths ($\lambda_S$).

4. Key Results

• Sub-Nanometer Orbital Diffusion Lengths:
  • In W|Ni heterostructures, the orbital diffusion length was extracted to be $\lambda_L \approx 0.36$ nm, which is comparable to a single unit cell.
  • This is drastically shorter than the spin diffusion length in W, measured at $\lambda_S \approx 2.20$ nm.
  • Similar results were found for Ta ($\lambda_L \approx 0.84$ nm) and Pt ($\lambda_L \approx 0.94$ nm), where $\lambda_L < \lambda_S$ in all cases.
• Non-Monotonic Amplitude and Polarity Reversal:
  • In W|Ni, the THz amplitude exhibited a non-monotonic behavior: a peak at ~0.8 nm, a minimum at ~2.8 nm, and a subsequent rise.
  • A polarity reversal was observed at $d_W \approx 2.8$ nm. For $d_W < 2.8$ nm, the polarity was opposite to that of the spin-dominated W|Fe reference, indicating a dominant orbital contribution (IOHE). For $d_W > 2.8$ nm, the polarity aligned with W|Fe, indicating a crossover to spin-dominated (ISHE) emission.
• Destructive Interference: The intensity minimum at ~2.8 nm corresponds to the point where the spin and orbital contributions destructively interfere ($A_S + A_L \approx 0$), confirming the coexistence and opposite signs of the two mechanisms.
• Ruling Out Interfacial IOREE: The insertion of Cu/Al spacers did not suppress the characteristic amplitude peak at ~0.8 nm or alter the polarity trends. This proves that the observed signals originate from bulk conversion (IOHE) rather than interfacial Rashba effects.
• Phase Offsets: The study observed peak-time delays (~250 fs) only in the thickness range where spin and orbital contributions interfere, supporting the interpretation of these delays as arising from the interference of bulk currents rather than delayed interfacial transport.

5. Significance

• Resolving a Fundamental Controversy: The findings provide strong experimental evidence supporting theoretical predictions that orbital transport in heavy metals is confined to sub-nanometer scales, challenging the prevailing view of long-range orbital currents.
• Reinterpretation of Orbitronics: The results suggest that in the THz regime, the orbital response in HMs is better described as a localized orbital polarization rather than a propagating ballistic current.
• Methodological Advancement: The wedge-sample platform combined with THz spectroscopy offers a powerful, non-invasive tool for disentangling spin and orbital dynamics at the nanoscale, applicable to future studies of low-dimensional materials.
• Implications for Device Design: The extremely short orbital diffusion lengths imply that orbitronic devices relying on long-range orbital transport may need to be redesigned to operate at atomic-scale interfaces or utilize materials with different orbital transport characteristics (e.g., weak SOC materials like Ti or Cr).

In conclusion, this work fundamentally shifts the understanding of orbital transport in heavy metals, demonstrating that orbital diffusion is highly localized and that previous interpretations of long-range transport likely conflated spin and orbital effects or misattributed bulk phenomena to interfacial mechanisms.