Quantifying Ultrafast Orbital Pumping and Nanometre-Scale Orbital Diffusion in Magnetic Heterostructures

arXiv:2608.29165 · cond-mat.mes-hall, physics.optics · Submitted 2026-08-29 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: We're looking at the paper: "Quantifying Ultrafast Orbital Pumping and Nanometre-Scale Orbital Diffusion in Magnetic Heterostructures".

Mira: Here's the gist of the paper. Spin and orbital pumping – the injection of spin and orbital angular momentum from a driven ferromagnet into an adjacent nonmagnetic layer – are fundamental processes underlying angular-momentum generation and transport in magnetic heterostructures.

Kai: Femtosecond optical excitation extends these phenomena into the ultrafast regime, where spintronic terahertz emission spectroscopy (STES) detects picosecond angularmomentum currents in a contact-free manner through spin-to-charge and orbital-tocharge conversion.

Mira: A central challenge is that spin and orbital currents are generated simultaneously and contribute additively to the same terahertz emission, preventing their quantitative separation.

Kai: Here, we overcome this limitation by combining STES with wedge-sample thickness control in heterostructures whose nonmagnetic layers (Ta, W, and Nb) have spin Hall and orbital Hall angles of opposite signs.

Paper summary: Kai: So this paper is titled "Quantifying Ultrafast Orbital Pumping and Nanometre-Scale Orbital Diffusion in Magnetic Heterostructures." It's about how spin and orbital angular momentum get pumped from a driven ferromagnet into a nearby nonmagnetic layer.

Mira: Exactly, it tackles the problem that when you excite this system with light in the ultrafast regime, both spin and orbital currents appear at the same time, making it hard to separate them.

Kai: The main thesis here is that they developed a way to quantitatively separate these two contributions using two different properties of the currents.

Mira: They use materials where the spin Hall angle and orbital Hall angle have opposite signs so that their resulting terahertz emissions have opposite polarities, which lets them tell them apart.

Lev: From an engineering standpoint, controlling the thickness of those nonmagnetic layers is key, using wedge-shaped samples to get continuous control over the layer thickness.

Kai: And they found that this method lets them decompose the signals based on both those different emission polarities and how the signal changes with that layer thickness.

Mira: What's really interesting from their systematic measurements on iron, cobalt, and nickel heterostructures is that the orbital pumping contribution actually gets progressively larger as you move toward nickel.

Lev: They found that for nickel, this orbital pumping contribution reaches several tens of percent of the spin current contribution, which is way higher than what theory predicted.

Kai: Even iron generates a non-negligible orbital current, especially when the magnetic layer is very thin, which suggests orbital effects matter even in small layers.

Mira: The results also show that the extracted orbital diffusion lengths are consistently shorter than the spin diffusion lengths and they get longer as the spin-orbit coupling strength of that nonmagnetic layer gets weaker.

Lev: That part about the diffusion lengths is important for real hardware because it tells us how far this orbital momentum can actually travel in these materials.

Kai: They looked at specific heterostructures like iron with tantalum, iron with tungsten, and iron with niobium to see different behaviors across those layers.

Mira: For instance, in iron with tantalum, the terahertz polarity stays negative from zero nanometers to twelve nanometers because the signal is dominated by spin-to-charge conversion in the tantalum.

Lev: But then you look at nickel with tantalum, and you see a polarity reversal happening around zero point three nanometers and it stays positive up to twelve nanometers, which they attribute to orbital-to-charge conversion.

Kai: That distinction between the spin-dominated behavior in iron with tantalum and the orbital-dominated behavior in nickel with tantalum is a clear way they're separating the two effects.

Mira: The theoretical benchmarking using their multi component terahertz emission model reproduces the thickness dependent amplitudes, including that threshold like behavior seen in iron with tantalum below zero point five nanometers.

Lev: The paper notes that for nickel based heterostructures, their estimated orbital current contribution can reach up to seventy percent of the total signal, which is much higher than what their microscopic theory predicted.

Kai: Their analysis of the orbital-to-spin current density ratio shows it's typically in the range of a few percent to about twenty percent for iron and cobalt, which lines up with microscopic theory for those metals.

Mira: But for nickel based heterostructures, that ratio can reach several tens of percent, up to seventy percent, which suggests a much stronger orbital pumping response than their theory accounted for.

Lev: The uncertainty they have in the spin hall and orbital hall angle values from different calculations gets into the inferred ratio, and that uncertainty ends up being bigger than their experimental error.

Kai: They looked at how these diffusion lengths change across tantalum, tungsten, and niobium to get a better picture of the transport.

Mira: They found that across all those heterostructures with tantalum, tungsten, and niobium, the orbital diffusion length is consistently shorter than the spin diffusion length in agreement with other recent studies.

Lev: Interestingly, they noted that the orbital diffusion length is markedly longer in niobium at one point around one point five nanometers compared to tantalum and tungsten which are smaller at about zero point three to zero point eight nanometers.

Kai: That difference in the diffusion lengths suggests that orbital angular momentum can actually transport deeper into the niobium layer, while how that length varies across different nonmagnetic layers shows that orbital propagation might depend on the spin-orbit coupling strength of the nonmagnetic layer.

Mira: So, this paper establishes terahertz emission spectroscopy as a quantitative way to probe coupled spin and orbital pumping by resolving their separate contributions.

Lev: The analysis also suggests that the secondary orbital-to-spin conversion in materials with finite spin orbit coupling might not play a significant role in the terahertz emission they were measuring.

Kai: They conclude that while this method helps benchmark microscopic theories of orbital generation and transport, further theoretical and experimental work on nickel based heterostructures is needed to fully understand the orbital current generation and interfacial transmission.

Conclusion: Kai: So, this paper is about figuring out how much spin and orbital momentum gets pumped when you drive a ferromagnet into an adjacent nonmagnetic layer at ultrafast timescales.

Mira: Yeah, it tackles that problem where both spin and orbital currents show up at once during optical excitation, making them hard to separate.

Kai: The authors developed a way to do that by picking materials with opposite spin and orbital Hall angles so the resulting signals have opposite polarities.

Mira: And they control the thickness of those nonmagnetic layers using wedge samples, which lets them decompose everything based on polarity and thickness dependence.

Kai: What's the big picture here? They found that for nickel systems, this orbital pumping contribution gets really big—sometimes tens of percent—which is way more than their theories predicted.

Mira: Even iron generates a noticeable orbital current in very thin layers, which is something you need to account for if you're building real devices.

Kai: They also measured the diffusion lengths and found that the orbital diffusion length is consistently shorter than the spin diffusion length, but it gets longer when the nonmagnetic layer has weaker spin-orbit coupling.

Mira: That difference in how far these currents travel matters a lot for designing magnetic heterostructures.

Kai: This work establishes terahertz emission spectroscopy as a way to actually measure and separate these coupled spin and orbital effects.

Mira: It gives us a solid benchmark for the microscopic theories describing how orbital angular momentum is generated and transported in these structures.

Kai: The paper suggests that the secondary conversion from orbital to spin isn't playing a big role in the signal they measured, which simplifies things somewhat.

Mira: But they point out that for nickel based systems, you still need more theoretical and experimental work to fully understand how those orbital currents are generated at the interface.

Tongyang Guan, Jiahao Liu, Yuxiao Mo, Liangliang Zhu, Yizheng Wu, Zhensheng Tao

Fudan University

cond-mat.mes-hall, physics.optics

Submitted: 2026-08-29

Updated: 2026-10-31

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 87/100

The gist: Spin and orbital pumping – the injection of spin and orbital angular momentum from a driven ferromagnet into an adjacent nonmagnetic layer – are fundamental processes underlying angular-momentum…

Key concepts

Spin Hall (SH) Angle ($\gamma_{SH}$ NM)
This angle describes how an electric current flowing through a nonmagnetic layer causes a transverse spin current to be generated. The paper uses this property, along with the orbital Hall angle, to determine the relative contributions of spin and orbital pumping in heterostructures.
Orbital Hall (OH) Angle ($\gamma_{OH}$ NM)
Similar to the Spin Hall angle, this describes how an electric current induces a transverse orbital angular momentum current. By having opposite signs for $\gamma_{SH}$ and $\gamma_{OH}$, researchers could assign distinct polarities to the spin and orbital pumping signals.
Spin Diffusion Length ($\lambda_S$) vs. Orbital Diffusion Length ($\lambda_L$)
These lengths quantify how far spin or orbital angular momentum can travel before decaying. The findings consistently show that the orbital diffusion length is shorter than the spin diffusion length across all tested materials, suggesting OAM transport is more localized.
Multi-component Terahertz Emission Model (MC-TEM)
This mathematical model was used to fit experimental terahertz emission data from heterostructures. It allowed researchers to quantitatively decompose the total signal into separate contributions from spin, orbital, and bare material effects as a function of layer thickness.

Terminology

Summary

Spin and orbital pumping – the injection of spin and orbital angular momentum from a driven ferromagnet into an adjacent nonmagnetic layer – are fundamental processes underlying angular-momentum generation and transport in magnetic heterostructures. p. 0

Where the paper says it

cn; zhenshengtao@fudan. Abstract Spin and orbital pumping – the injection of spin and orbital angular momentum from a driven ferromagnet into an adjacent nonmagnetic layer – are fundamental processes underlying angular-momentum generation and transport in magnetic heterostructures.

Femtosecond optical excitation extends these phenomena into the ultrafast regime, where spintronic terahertz emission spectroscopy (STES) detects picosecond angularmomentum currents in a contact-free manner through spin-to-charge and orbital-tocharge conversion. p. 0
Where the paper says it

Abstract Spin and orbital pumping – the injection of spin and orbital angular momentum from a driven ferromagnet into an adjacent nonmagnetic layer – are fundamental processes underlying angular-momentum generation and transport in magnetic heterostructures. Femtosecond optical excitation extends these phenomena into the ultrafast regime, where spintronic terahertz emission spectroscopy (STES) detects picosecond angularmomentum currents in a contact-free manner through spin-to-charge and orbital-tocharge conversion.

A central challenge is that spin and orbital currents are generated simultaneously and contribute additively to the same terahertz emission, preventing their quantitative separation. p. 0
Where the paper says it

Microscopic theory predicts that orbital-pumping efficiency increases from Fe to Ni across the 3d series, yet whether these predictions hold under ultrafast excitation remains unclear. A central challenge is that spin and orbital currents are generated simultaneously and contribute additively to the same terahertz emission, preventing their quantitative separation.

Here, we overcome this limitation by combining STES with wedge-sample thickness control in heterostructures whose nonmagnetic layers (Ta, W, and Nb) have spin Hall and orbital Hall angles of opposite signs. p. 0
Where the paper says it

A central challenge is that spin and orbital currents are generated simultaneously and contribute additively to the same terahertz emission, preventing their quantitative separation. Here, we overcome this limitation by combining STES with wedge-sample thickness control in heterostructures whose nonmagnetic layers (Ta, W, and Nb) have spin Hall and orbital Hall angles of opposite signs.

The extracted orbital diffusion lengths are consistently shorter than spin diffusion lengths and increase with decreasing spin-orbit coupling strength of the nonmagnetic layer. p. 0
Where the paper says it

Even Fe generates a non-negligible orbital current that becomes essential in the thinnonmagnetic-layer regime. The extracted orbital diffusion lengths are consistently shorter than spin diffusion lengths and increase with decreasing spin-orbit coupling strength of the nonmagnetic layer.

p. 0
Where the paper says it

Even Fe generates a non-negligible orbital current that becomes essential in the thinnonmagnetic-layer regime. The extracted orbital diffusion lengths are consistently shorter than spin diffusion lengths and increase with decreasing spin-orbit coupling strength of the nonmagnetic layer.

Quantitative Disentanglement Strategy

The authors developed a method to quantitatively separate spin and orbital pumping contributions by exploiting two complementary properties of the currents. First, they select NMs whose spin Hall (gammaSH NM) and orbital Hall (gammaOH NM) angles have opposite signs, so that the spin- and orbital-current-induced emissions carry opposite polarities. p. 2

Where the paper says it

Here, we exploit two complementary properties of spin and orbital currents to disentangle their contributions within a single STES platform. First, because the polarity of the terahertz emission follows the sign of the corresponding Hall angle, we select NMs whose spin Hall (𝛾SH NM) and orbital Hall (𝛾OH NM) angles have opposite signs, so that the spin- and orbital-current-induced emissions carry opposite polarities.

Second, they use wedge-shaped samples to enable continuous control of the nonmagnetic layer thickness. p. 2
Where the paper says it

Second, because the two currents decay over different characteristic lengths in the NM, highprecision thickness-dependent measurements separate their contributions quantitatively. The latter is realized using wedge-shaped samples, which enable continuous control of

This approach allows for the quantitative decomposition of the contributions based on distinct emission polarities and thickness dependences. p. 0
Where the paper says it

Here, we overcome this limitation by combining STES with wedge-sample thickness control in heterostructures whose nonmagnetic layers (Ta, W, and Nb) have spin Hall and orbital Hall angles of opposite signs. The two angular-momentum channels therefore exhibit distinct emission polarities and thickness dependences, enabling their quantitative decomposition.

Systematic Findings across Materials

Systematic measurements on Fe, Co, and Ni heterostructures reveal that the orbitalpumping contribution increases progressively toward Ni, reaching several tens of percent of the spin-current contribution – far exceeding theoretical predictions. Even Fe generates a non-negligible orbital current that becomes essential in the thinnonmagnetic-layer regime. p. 0

Where the paper says it

Systematic measurements on Fe, Co, and Ni heterostructures reveal that the orbitalpumping contribution increases progressively toward Ni, reaching several tens of percent of the spin-current contribution – far exceeding theoretical predictions. Even Fe generates a non-negligible orbital current that becomes essential in the thinnonmagnetic-layer regime.

The extracted orbital diffusion lengths are consistently shorter than spin diffusion lengths and increase with decreasing spin-orbit coupling strength of the nonmagnetic layer.

Analysis of Specific Heterostructures

The analysis involved examining FMTa, FMW, and FMNb heterostructures to observe distinct behavior. For FeTa, the terahertz polarity remains negative from dTa=0 to 12 nm because the signal is dominated by ISHE-mediated spin-current conversion in Ta. p. 4

Where the paper says it

The Fe|Ta signal is therefore dominated by ISHE-mediated spin-current conversion in Ta. Because the ISHE-induced signal and the bare-Fe contribution have the same polarity, no polarity reversal is observed as dTa increases.

In contrast, NiTa exhibits a polarity reversal from negative to positive at dTa≈0. p. 5
Where the paper says it

6 By contrast, Ni|Ta exhibits a polarity reversal from negative to positive at dTa≈0. 3 nm and remain positive up to 12 nm (Fig.

3 nm and remains positive up to 12 nm, which is naturally assigned to IOHEmediated orbital-to-charge conversion. p. 5
Where the paper says it

This positive Ta-mediated signal cannot arise from ISHE-dominated conversion. Instead, it is naturally assigned to IOHEmediated orbital-to-charge conversion, because 𝛾OH Ta and 𝛾SH Ta possess opposite signs [15,19].

Theoretical Benchmarking and Limitations

The multi-component terahertz-emission model (MC-TEM) was adopted to reproduce the thickness-dependent amplitudes, including the threshold-like behavior observed in FeTa below dTa≈0.5 nm. The orbital-to-spin current-density ratio is calculated as jL/jS = ÃL/S ∙ sigmaSH NM / σOH NM, which is typically in the range of a few percent to 20% for Fe and Co, broadly consistent with microscopic theory. p. 3

Where the paper says it

First, the orbital-pumping contribution increases progressively from Fe to Co to Ni. After accounting for the reported range of NM spin Hall and orbital Hall coefficients, Fe and Co show orbital-current contributions of a few percent to ~20% of the spin-current contribution, whereas Ni reaches several tens of percent, exceeding current theoretical predictions [13].

However, for Ni-based heterostructures, the estimated orbitalcurrent contribution can reach several tens of percent, up to 70%, indicating a substantially stronger orbital-pumping response than predicted. p. 0
Where the paper says it

Systematic measurements on Fe, Co, and Ni heterostructures reveal that the orbitalpumping contribution increases progressively toward Ni, reaching several tens of percent of the spin-current contribution – far exceeding theoretical predictions. Even Fe generates a non-negligible orbital current that becomes essential in the thinnonmagnetic-layer regime.

The uncertainty in sigmaSH NM and sigmaOH NM values between different ab-initio methods propagates directly into the inferred jL/jS, whose uncertainty far exceeds the experimental error. p. 9
Where the paper says it

For a given FM layer, the extracted 𝐴̃ 𝐿 𝐴̃ ⁄ 𝑆 ratios are generally consistent across different NMs, indicating that they reflect the intrinsic spin- and orbital-pumping efficiencies of the FM source layer rather than NM-dependent conversion processes. The sole exception is Ni|Nb, where the spin-induced signal is negligible, yielding a substantially larger fitted 𝐴̃ 𝐿 𝐴̃ ⁄ 𝑆 ratio with increased uncertainty (see Table 1).

Diffusion Length Dependence

The diffusion lengths extracted from the MC-TEM fitting show that across all FMNM heterostructures with Ta, W, and Nb, the orbital diffusion length, λL, is consistently shorter than the corresponding spin diffusion length. Notably, λL is markedly longer in Nb (1. p. 3

Where the paper says it

Second, even Fe generates a non-negligible orbital current that becomes particularly important in the thin-NM regime, resulting in a threshold-like behavior in the terahertz emission. Third, the extracted orbital diffusion lengths are consistently shorter than the corresponding spin diffusion lengths across all NMs studied, but markedly longer in Nb than those in the heavier 5d metals Ta and W, suggesting that orbital propagation may depend on the SOC strength of the NM.

5 nm) than those in the heavier 5d metals Ta and W (0. p. 3
Where the paper says it

Second, even Fe generates a non-negligible orbital current that becomes particularly important in the thin-NM regime, resulting in a threshold-like behavior in the terahertz emission. Third, the extracted orbital diffusion lengths are consistently shorter than the corresponding spin diffusion lengths across all NMs studied, but markedly longer in Nb than those in the heavier 5d metals Ta and W, suggesting that orbital propagation may depend on the SOC strength of the NM.

3 - 0.8 nm). This extended λL in Nb suggests that OAM can indeed transport deep into Nb, while the variation of λL across NMs indicates that orbital propagation may depend on the SOC strength of the NM. p. 3
Where the paper says it

Third, the extracted orbital diffusion lengths are consistently shorter than the corresponding spin diffusion lengths across all NMs studied, but markedly longer in Nb than those in the heavier 5d metals Ta and W, suggesting that orbital propagation may depend on the SOC strength of the NM. Together, these findings establish a quantitative framework for investigating ultrafast orbital pumping and open a route to engineering spin and orbital angular-momentum generation and transport in magnetic heterostructures.

Conclusion and Future Directions

The results establish STES as a quantitative probe of coupled spin and orbital pumping in magnetic heterostructures By resolving their coexisting contributions, this work provides a basis for benchmarking microscopic theories of OAM generation and transport. The analysis suggests that the secondary orbital-to-spin conversion is unlikely to play a significant role in the present analysis. p. 3

Where the paper says it

Together, these findings establish a quantitative framework for investigating ultrafast orbital pumping and open a route to engineering spin and orbital angular-momentum generation and transport in magnetic heterostructures. Analysis of Results in FM|Ta Heterostructures Figure 1a illustrates the general concept of our experiments.

Orbital-to-spin conversion in NMs with finite SOC can generate secondary spin currents that contribute to terahertz emission via the ISHE, but this mechanism was not found to be significant. p. 1
Where the paper says it

Their conversion into transient charge currents through the inverse spin Hall effect (ISHE) gives rise to broadband terahertz emission, enabling spintronic terahertz emission spectroscopy (STES) as a contact-free, ultrafast probe of spin-current pumping and transport [9–12]. Beyond spin, magnetization dynamics can also generate orbital currents – flows of orbital angular momentum (OAM) – through orbital pumping, the orbital analogue of spin pumping [13,14].

The paper concludes that orbital-current generation and interfacial transmission in Ni-based heterostructures require further theoretical and experimental investigations. p. 0
Where the paper says it

Systematic measurements on Fe, Co, and Ni heterostructures reveal that the orbitalpumping contribution increases progressively toward Ni, reaching several tens of percent of the spin-current contribution – far exceeding theoretical predictions. Even Fe generates a non-negligible orbital current that becomes essential in the thinnonmagnetic-layer regime.

Table 1

The table summarizes the orbital-to-spin amplitude-coefficient ratios, ÃL/S, obtained from the FMTa, FMW, and FMNb series. The ratio increases systematically from Fe to Co to Ni, revealing a progressive enhancement of the orbital contribution across the 3d FM series. p. 9

Where the paper says it

The sole exception is Ni|Nb, where the spin-induced signal is negligible, yielding a substantially larger fitted 𝐴̃ 𝐿 𝐴̃ ⁄ 𝑆 ratio with increased uncertainty (see Table 1). Despite this caveat, 𝐴̃ 𝐿 𝐴̃ ⁄ 𝑆 increases systematically from Fe to Co to Ni, revealing a progressive enhancement of the orbital contribution across the 3d FM series, consistent

Table 2

The table summarizes the spin and orbital diffusion lengths for Ta, W, and Nb. The orbital diffusion length λL is consistently shorter than the corresponding spin diffusion length λS in agreement with recent studies. p. 3

Where the paper says it

Second, even Fe generates a non-negligible orbital current that becomes particularly important in the thin-NM regime, resulting in a threshold-like behavior in the terahertz emission. Third, the extracted orbital diffusion lengths are consistently shorter than the corresponding spin diffusion lengths across all NMs studied, but markedly longer in Nb than those in the heavier 5d metals Ta and W, suggesting that orbital propagation may depend on the SOC strength of the NM.

p. 3
Where the paper says it

Second, even Fe generates a non-negligible orbital current that becomes particularly important in the thin-NM regime, resulting in a threshold-like behavior in the terahertz emission. Third, the extracted orbital diffusion lengths are consistently shorter than the corresponding spin diffusion lengths across all NMs studied, but markedly longer in Nb than those in the heavier 5d metals Ta and W, suggesting that orbital propagation may depend on the SOC strength of the NM.

Figure 1

Figure 1 illustrates the general concept of our experiments. Figures 3a-c show the W-thicknessdependent terahertz waveforms from Fe|W, Ni|W, and Co|W, respectively. Figure 4ac presents the terahertz waveforms as a function of Nb thickness, dNb, in Fe|Nb, Ni|Nb, and Co|Nb, respectively.

Figure 2

Figure 2 presents the MC-TEM decomposition of terahertz signals from the FM|Ta heterostructures. Figures 3d and Supplementary Section S8 show that the thickness-dependent amplitudes at t=0 and waveform evolution are well reproduced by the MC-TEM model. p. 7

Where the paper says it

The thickness-dependent amplitudes at t=0 and waveform evolution are well reproduced by the MC-TEM model, as shown in Fig. 3d and Supplementary Section S8.

Figure 4d shows that the terahertz amplitudes extracted at t=0 and waveform evolution are well reproduced by the MC-TEM model for FM|Nb heterostructures.

Figure 3

Figure 3 shows the terahertz emission from the FM|W heterostructures. Figure 4e shows a threshold-like behavior in Fe|Nb, which is resolved by comparing measurements from Fe|Nb and Ni|Nb regions that share a common Nb wedge.

Figure 4

Figure 4 shows the terahertz emission from FM|Nb heterostructures. Figure 4f shows the extracted AS, AL, and ACo contributions in Fe|Nb as a function of dNb.

Figure 5

Figure 5 shows the decomposed amplitudes of spin (AS), orbital (AL), and bare-Fe (AFe) contributions in FeTa for dTa<3 nm. The decomposition reveals that the thresholdlike behavior originates from near cancellation between AS and AL, with AS+AL≈0 below dTa≈0. p. 7

Where the paper says it

8 Figure 2f shows the decomposed amplitudes of spin (AS), orbital (AL), and bare-Fe (AFe) contributions in Fe|Ta for dTa<3 nm. The decomposition reveals that the thresholdlike behavior originates from near cancellation between AS and AL, with AS+AL≈0 below dTa≈0.

p. 7
Where the paper says it

8 Figure 2f shows the decomposed amplitudes of spin (AS), orbital (AL), and bare-Fe (AFe) contributions in Fe|Ta for dTa<3 nm. The decomposition reveals that the thresholdlike behavior originates from near cancellation between AS and AL, with AS+AL≈0 below dTa≈0.

p. 7
Where the paper says it

8 Figure 2f shows the decomposed amplitudes of spin (AS), orbital (AL), and bare-Fe (AFe) contributions in Fe|Ta for dTa<3 nm. The decomposition reveals that the thresholdlike behavior originates from near cancellation between AS and AL, with AS+AL≈0 below dTa≈0.

5 nm.

Figure 6

Figure 6 shows the decomposed amplitudes of spin (AS), orbital (AL), and bare-Fe (AFe) contributions in FeTa for dTa<3 nm. The decomposition reveals that the thresholdlike behavior originates from near cancellation between AS and AL, with AS+AL≈0 below dTa≈0.5 nm.

Figure 7

Figure 7 shows the extracted spin (AS), orbital (AL), and bare-Fe (AFe) contributions in FeTa for dTa<3 nm. The decomposition reveals that the thresholdlike behavior originates from near cancellation between AS and AL, with AS+AL≈0 below dTa≈0.5 nm.

Figure 8

Figure 8 shows the extracted AS, AL, and ACo contributions in Co|W as a function of dW. The grey shaded region corresponds to the threshold-like regime shown in e. p. 6

Where the paper says it

It also captures the threshold-like behavior of Fe|Ta below dTa≈0. 5 nm, as shown in Fig.

p. 6
Where the paper says it

It also captures the threshold-like behavior of Fe|Ta below dTa≈0. 5 nm, as shown in Fig.

, where the spin and orbital contributions nearly compensate with each other. p. 2
Where the paper says it

Resolving these quantities is essential for benchmarking microscopic theories of ultrafast orbital pumping. Here, we exploit two complementary properties of spin and orbital currents to disentangle their contributions within a single STES platform.

p. 2
Where the paper says it

Resolving these quantities is essential for benchmarking microscopic theories of ultrafast orbital pumping. Here, we exploit two complementary properties of spin and orbital currents to disentangle their contributions within a single STES platform.

Figure 9

Figure 9 shows the extracted AS, AL, and ACo contributions in Fe|Nb as a function of dNb. The grey shaded region corresponds to the threshold-like regime shown in e., where the spin and orbital contributions nearly compensate with each other.

Table 1 summarizes the orbital-to-spin amplitude-coefficient ratios, ÃL/S, obtained from the FMTa, FMW, and FMNb series. The ratio increases systematically from Fe to Co to Ni. p. 9

Where the paper says it

The sole exception is Ni|Nb, where the spin-induced signal is negligible, yielding a substantially larger fitted 𝐴̃ 𝐿 𝐴̃ ⁄ 𝑆 ratio with increased uncertainty (see Table 1). Despite this caveat, 𝐴̃ 𝐿 𝐴̃ ⁄ 𝑆 increases systematically from Fe to Co to Ni, revealing a progressive enhancement of the orbital contribution across the 3d FM series, consistent

Table 2 summarizes the spin and orbital diffusion lengths for Ta, W, and Nb. The orbital diffusion length λL is consistently shorter than the corresponding spin diffusion length λS in agreement with recent studies.

Figure 1 shows the experimental scheme and terahertz emission from the FM|Ta heterostructures. Figures 3a-c show the W-thicknessdependent terahertz waveforms from Fe|W, Ni|W, and Co|W, respectively. Figure 4ac presents the terahertz waveforms as a function of Nb thickness, dNb, in Fe|Nb, Ni|Nb, and Co|Nb, respectively.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved materials design for spintronic devices by utilizing specific nonmagnetic layers like Nb to enhance orbital current transport, as the variation of lambdaL across NMs suggests that orbital propagation may depend on the SOC strength of the NM.

  2. Development of ultrafast spin and orbital pumping characterization tools capable of quantitatively separating spin and orbital contributions in terahertz emission using a multi-component model (MC-TEM). This allows researchers to resolve their coexisting contributions which was previously impossible.

  3. Enhanced predictive modeling for angular momentum generation by incorporating femtosecond laser excitation into microscopic theory, specifically addressing the challenge that spin and orbital currents are generated simultaneously and contribute additively to the same detected terahertz emission.

  4. Creation of a quantitative framework to benchmark theoretical predictions for orbital pumping efficiency across different 3d transition metals (Fe, Co, Ni) by extracting orbital-to-spin amplitude-coefficient ratios (A L/A S).

  5. Design of novel magnetic heterostructures where the spin Hall and orbital Hall angles of nonmagnetic layers have opposite signs to ensure that the spin- and orbital-current-induced emissions carry opposite polarities.

Abstract

Quantifying ultrafast orbital pumping and transport remains challenging because spin and orbital currents contribute coherently to terahertz emission. Here, we combine spintronic terahertz emission spectroscopy with wedge-sample thickness control to disentangle these contributions in Fe, Co and Ni heterostructures with Ta, W and Nb. Their opposite spin Hall and orbital Hall polarities enable quantitative determination of relative pumping strengths and propagation lengths. We find that orbital pumping increases from Fe to Ni, with inferred orbital-to-spin current ratios reaching several tens of percent in Ni, exceeding current microscopic predictions. Remarkably, Nb exhibits an orbital diffusion length of ~1.5 nm, substantially longer than in Ta and W and beyond the sub-nanometre scale predicted by recent orbital-relaxation calculations. These findings provide quantitative experimental benchmarks for ultrafast orbital pumping and transport.

Related papers