Inverse orbital Hall effect induced terahertz emission enabled by a ferromagnet with quenched orbital moment in Fe/Pt/W trilayers

This episode covers a study on thin stacked layers of iron, platinum, and tungsten—called Fe/Pt/W trilayers—that produce strong terahertz emission through a process called the inverse orbital Hall effect. Terahertz waves are bursts of electromagnetic energy at frequencies between microwaves and infrared light, like very fast radio signals that can pass through materials for imaging or sensing. The puzzle at the heart of the paper is how iron, a material where the orbital part of its magnetism is mostly suppressed, suddenly drives these waves over long distances of about 100 nanometers in the tungsten layer—far beyond what usual spin currents can reach.

So the core problem is that typical setups with just iron and platinum or iron and tungsten fade out quickly, but adding all three layers changes everything?

Exactly. In standard iron/platinum or iron/tungsten bilayers, the terahertz signal drops sharply after just a few nanometers because spin currents—the flow of tiny magnetic twists from laser-excited electrons—only travel a short distance in heavy metals like platinum or tungsten before scattering and losing strength. That's due to the short spin diffusion length, often just a couple of nanometers. But in the trilayer, the signal persists much farther, with delays and broadening that point to a different kind of flow: orbital angular momentum, which carries a twisting motion around the electron's path, propagating ballistically through tungsten without quick scattering.

Orbital angular momentum sounds like it needs its own explanation—how does that even get started from iron, if its orbitals are suppressed?

Picture electrons in iron getting jolted by a quick laser pulse, creating a rush of spins that normally fizzle out fast. In the trilayer, platinum acts like a translator: its strong spin-orbit coupling—a interaction twisting electron spin and motion together—converts that short-range spin flow into a longer-lasting orbital flow. This orbital current then shoots through tungsten and turns into a charge current via the inverse orbital Hall effect, where the twisting orbital motion pushes charges sideways to make terahertz waves. The paper shows this with thickness tests: optimal two-nanometer platinum maximizes overlap between quick platinum effects and delayed tungsten ones.

Right, so platinum bridges the gap—turning a weak, short spin signal into something robust for the tungsten to use.

Yes, and that's notable because it sidesteps spin limits entirely. The evidence comes from signal patterns that don't match spin diffusion alone.

So those signal patterns—what exactly do they show about the bilayers versus the trilayer?

In iron/platinum and iron/tungsten bilayers, the terahertz waves come mainly from incoming spins pushing charges sideways in the heavy metal layer, thanks to the twisting force between spin and motion. Platinum and tungsten handle this differently: platinum sends charges one way while tungsten sends them the opposite way, because of their opposite spin Hall angles—a material trait that sets the direction and strength of that sideways push. You'd expect the trilayer to partly cancel out those opposing pushes, but instead the signal gets notably stronger.

Stronger than either bilayer alone? That goes against just adding them up.

Precisely. The paper compares waveforms: iron/platinum peaks higher than iron/tungsten due to platinum's stronger conversion, yet the trilayer's peak-to-peak amplitude surpasses both, with no cancellation. They tested this by varying heavy metal thicknesses in bilayers—signal strength maxes around two nanometers then drops sharply beyond fifteen nanometers. The timing of the waves stays fixed, without delay or spreading, confirming short-range spin flow limited by quick scattering.

Okay, so bilayers are stuck in that short zone. But the trilayer breaks through because platinum first flips spins to orbitals?

Yes—that conversion lets orbital flow reach farther into tungsten, adding constructively to the platinum's quick charge push. Time-domain signals show the trilayer's delayed broadening, matching long-range ballistic travel rather than diffusion. This cooperative boost is clearest at about two-nanometer platinum, aligning the fast and slow currents perfectly.

So that cooperative boost at two-nanometer platinum—does it hold up as you make the tungsten layer thicker?

Yes. When researchers increase tungsten from one to 100 nanometers, the terahertz signal weakens steadily but stays detectable even at 100 nanometers—far beyond the few nanometers spins typically travel. They fit the drop-off with a decay length that stretches longest around two-nanometer platinum. Time delays grow straight-line with tungsten thickness, giving transport speeds of about 0.3 to 0.6 nanometers per femtosecond—slower than in iron-tungsten bilayers, which fits ultrafast spins but not here.

So the slowing speed and that long reach point straight to orbital flow making it through tungsten.

Precisely. That combination—long propagation, thickness-tied delays, and broadening—lines up with orbital Hall transport dominating in these trilayers. But orbital alone doesn't max the output; the real gain comes from timing the quick charge push in platinum with the delayed one from tungsten. Those pushes have matching directions, so they add up coherently instead of canceling—like syncing two waves to build higher.

Right, so for thin tungsten around 0.75 nanometers, two-nanometer platinum gives the strongest lift over what you'd get from platinum or tungsten alone.

Exactly. They define enhancement as the trilayer signal divided by the sum of bilayer signals; it tops one over a range of platinum thicknesses for thin tungsten. This rules in spin-to-orbital handover via platinum's spin-orbit coupling, with same-polarity interference driving the extra strength. The paper's waveforms and fits make a clear case for this paired mechanism.

Okay, so two-nanometer platinum hits the sweet spot for syncing those currents. But what happens if you make the platinum thicker, say past three nanometers?

The study shows orbital transport into tungsten persists, as delays keep building and pulses continue broadening with thicker tungsten. But the incoming spin flow from iron mostly fades inside the thicker platinum before reaching the platinum-tungsten edge, weakening the quick charge push from platinum. That makes the paired action less effective, so the terahertz output doesn't grow further.

So the orbital part hangs on, but the teamwork falls apart without enough spin making it through.

Precisely. They also tested flipping the platinum and tungsten order—putting tungsten next to iron instead of platinum. That kills both the extra strength and the long-distance flow.

Why would the order matter that much?

Platinum's strong spin-to-motion twisting is key for first turning short spins into the longer orbital flow. Tungsten excels at converting that orbital twist into sideways charge motion, but lacks platinum's conversion step. So only platinum-first lets both layers contribute without loss.

Right—like platinum has to handle the handoff before tungsten can play its part.

Exactly. The paper ties this to a carefully matched teamwork: platinum's quick sideways charge from spins pairs with tungsten's delayed one from orbitals, adding up in both position and timing around two-nanometer platinum. This setup revives iron's weak orbital side through layer tuning.

So bilayers stay spin-short, trilayers go long-range orbital, and the gain is this non-adding-up teamwork?

Yes—the bilayers match pure short-spin expectations with no orbital signs, while trilayers show clear long-range traits like steady delays and spread pulses over tens of nanometers in tungsten. Overall, it points to interface tweaks and thickness choices as a solid way to boost these terahertz signals and tap orbital effects in everyday ferromagnets like iron. The evidence from thickness sweeps and order tests supports this.

So to pull this together, the trilayers get their edge from that precise handoff and sync-up, but it all hinges on getting the thicknesses just right—like two nanometers for platinum.

Yes. The paper notes a key limit: that boost demands exact nanometer-scale control of the layers, as small changes disrupt the overlap of those fast and slow currents. Reversing to iron-tungsten-platinum wipes out the long-range flow and extra strength entirely.

Right—and you mentioned the stacking order matters too. What if designers can't nail that precision every time?

The study highlights this as a practical hurdle, suggesting future work might explore other metals or interfaces to broaden options. Still, the evidence from waveforms and thickness sweeps makes a clear case for the mechanism here—cooperative inverse spin Hall and inverse orbital Hall effects delivering stronger, longer-reaching terahertz output from quenched iron via platinum's bridge to tungsten.

Well put, Sam. It's a thoughtful look at how layers can team up to push signals farther. Thanks for breaking it down—that's it for this episode of ResearchPod on terahertz emission in Fe/Pt/W trilayers. Thanks for listening.