Coherent shear wave in crystalline Si upon sub-threshold femtosecond laser-induced excitation

arXiv:2503.10420 · cond-mat.mtrl-sci, physics.optics · Submitted 2025-03-13 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.

Kai: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "Coherent shear wave in crystalline Si upon sub-threshold femtosecond laser-induced excitation".

Paper summary: Kai: To set the stage, this paper explores ultrafast laser processing of silicon wafers, which is super relevant for electronics and photonics industry applications >

Mira: They use X-ray free electron lasers or XFELs as these powerful probes to reveal the ultrafast dynamics of changes induced by the laser on silicon >

Kai: The core idea here is that they present a new way to get snapshots of the strain field, both in time and in depth, when you excite bulk crystalline silicon with a laser pulse >

Mira: They found that at fluences below the melting threshold, these strong orthogonal lattice distortions propagate into the depth >

Lev: So they are looking for permanent or transient changes in the long-range order of matter, like strain or phonon oscillations, using XFELs for probing these ultrafast processes >

Kai: The paper is driven by applications in electronics and photonics industry because ultrafast laser processing offers reduced thermal load compared to nanosecond pulse and continuous wave processing >

Mira: They mention that this reduced thermal load lets us fabricate smaller and sharper feature sizes, even below the diffraction limit, which is a big deal for scaling up devices >

Lev: But they also have to contend with the fact that they are using bulk materials like wafers in real-world processing applications because of their dominance in the electronics industry >

Kai: They propose this technique because developing an experimental tool that measures fast and ultrafast laser-induced structural changes with sub-picosecond and depth resolution is something highly desired >

Mira: The paper touches on strategies to get a depth resolved mapping of strain fields, mentioning bright field x-ray microscopy and dark field x-ray microscopy as suitable methods >

Conclusion: Kai: The title, "Coherent shear wave in crystalline Si upon sub-threshold femtosecond laser-induced excitation," really captures the core finding, which is the discovery of these orthogonal shear waves propagating deep into silicon when you use femtosecond lasers below the melting point >

Mira: What this means for us is that we’re seeing strain dynamics that weren't predicted by previous models based only on longitudinal waves, and this could be key to understanding how silicon can transform from one phase to another >

Lev: It also points toward a mechanism for things like ultrafast melting or ablation at higher fluences because these orthogonal distortions show up when the laser power increases >

Kai: So, even in the sub-threshold regime, we're seeing more complex structural changes than just what we expected from simple longitudinal vibrations >

Mira: The authors also connect this finding to earlier work that showed similar propagation speeds, which is good because it confirms that these shear waves are real and have a consistent speed across different studies >

Lev: The limitation they mention is that the method relies on resolving fine structure inside the diffraction peaks, which is what allows them to see these variations in intensity and position >

Kai: So, while this gives us incredible depth resolution, we still have to figure out exactly what those specific orthogonal distortions are doing physically when they propagate through the bulk material >

Angel Rodríguez-Fernꇞz, Jan-Etienne Pudell, Roman Shayduk, James Wrigley, Alejandro Fraile-Gimeno, Wonhyuk Jo, Johannes Möller, Alexey Zozulya, Èorg Hallmann, Ánders Madsen

European X-ray Free-Electron Laser Facility · Lund University · Universitt Duisburg-Essen Department of Physics and Center for Nanointegration Duisburg-Essen, Carl Benz Street 199, 47057 Duisburg Germany · Warsaw University of Technology Faculty of Physics · Institute of Physics of the Polish Academy of Sciences

cond-mat.mtrl-sci, physics.optics

Submitted: 2025-03-13

Updated: 2026-11-18

DOI: 10.1103/jdf2-r5hr

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 83/100

The gist: Femtosecond laser processing of semiconductor wafers is driven by applications in the electronics and photonics industry The authors present a novel technique that provides time and depth resolved…

Key concepts

Orthogonal Lattice Distortions
These are specific types of lattice deformations that occur perpendicular to the direction of laser propagation or surface normal. The study found these distortions propagate into the silicon crystal at a speed slower than the sound waves traveling along the crystal's main axis, which is unexpected based on prior models.
Longitudinal Speed of Sound (LSS)
This refers to the speed at which sound waves travel parallel to the direction of excitation or propagation within a solid material like silicon. The experiment found that lattice distortions propagate slower than this speed, indicating a different type of wave motion is occurring during laser excitation.
Near Field X-ray Diffracted Microscopy (NFXDM)
This is an advanced microscopy technique used to record the fine structural details inside the diffraction peaks of a crystal. It allows researchers to capture snapshots of the crystal's internal deformation in real-time, providing sub-picosecond and depth resolution for studying ultrafast laser effects.

Terminology

Summary

Femtosecond laser processing of semiconductor wafers is driven by applications in the electronics and photonics industry The authors present a novel technique that provides time and depth resolved snapshots of the strain field upon laser excitation of bulk crystalline Si. p. 1 At fluences below the melting threshold, strong orthogonal lattice distortions were found to propagate into the depth. p. 1 Simulations support a propagation speed of 5. p. 18 km s−1, slower than the longitudinal speed of sound, 8. p. 14 km s−1.

Experimental Methodology

The experiment utilized Near field x-ray diffracted microscopy (NFXDM) to record the fine structure inside the Borrmann fan for a Si single crystal in Laue diffraction geometry at the European XFEL. The researchers investigated the laser-induced transient changes of the BF signal as a function of laser fluence and pump-probe time delay in an excitation regime below the melting threshold. p. 3 The setup involved an optical laser with a central wavelength of 800 nm, a pulse duration of 25 fs and a maximum fluence of 300 mJ/cm2. p. 4

Key Findings on Strain Wave Propagation

The study revealed that good agreement with the experimental data is only achieved by introducing strong lattice distortions orthogonal to the surface normal propagating at slower velocities than the longitudinal speed of sound (LSS), 8.4 km s−1. This observation challenges current understanding where only longitudinal strain waves are expected due to the quasi 1D excitation geometry. p. 3 The calculated speed of this wave of approximately 5. p. 88 km s−1 matches the TSS. p. 8

Modeling and Interpretation

The authors retrieved time and depth information from the transient signal together with numerical calculations that combine UDD simulations with a 3D version of the model of Thomsen et al. [45] to describe the generation and propagation of ps strain pulses. p. 3 The analytical Thomsen model describes the lattice displacement in the perpendicular direction as a function of depth z and delay time t using equation (1). p. 5

Conclusions and Implications

The results challenge previous x-ray work using monochromatic Laue geometry diffraction where no orthogonal deformation was observed after laser excitation. The presence of the orthogonal distortion agrees with work performed at higher fluences. p. 9 p. 9 This finding could be key to understanding the starting mechanism of the phase transformation from diamond Si-I cell to Si-II phase. p. 9 The method could help unveil ultrafast processes at higher fluences, such as ultrafast melting, ablation and shock wave generation. p. 2

Pruning redundant reasoning tokens

The authors treated the laser absorption depth as a parameter to account for non-linear absorption effects (i.e. free-carrier and multi-photon absorption) of the 800 nm ultrashort laser pulses in silicon. p. 6 The effective absorption depth of 300 nm gave the best match to the experimental data for a fluence of 50 mJ/cm2. p. 6 A progressive intensity reduction around the position 50 µm of the signal was observed experimentally. p. 6 The simulation showed a stronger suppression than observed experimentally for some fluences. p. 6 The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 p. 9 A waterfall representation shows the experimental data features near-surface signal depression whose amplitude increases with time and whose position moves along the crystal depth. p. 7 The chosen Laue reflection is sensitive only to the orthogonal lattice components with respect to the wave propagation direction. p. 9 The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8 p. 8

How the training signal is built

The UDD process involves generating multiple diffracted beams, so-called echoes, at the surface of a single crystal. The spatial distribution of the echoes at the exit surface is the Fourier transform of the reflectivity curve for a particular moment in time of the crystal lattice along the depth. p. 2 The UDD signal can be used as a stroboscopic probe of the crystal lattice deformation for a defined time delay, which represents a deformation snapshot. p. 2

Where the gains come from

The capability of resolving the fine structure inside the diffraction peaks is what allows this work to resolve variations in intensity and position of the maxima. The use of NFXDM allows recording in single shot mode the fine structure inside the BF for a Si single crystal. p. 3 This method provides sub-picosecond and depth resolution for measuring fast and ultrafast laser-induced structural changes and strain fields. p. 2 The research could be key to understand the starting mechanism of the phase transformation from diamond Si-I cell to Si-II phase. p. 9 The findings also agree in the speed of propagation presented in the work by Smith et al. p. 9 [8].

The authors attribute the difference with respect to previous x-ray monochromatic studies to their capability of resolving the fine structure in the diffraction peak shape. The presence of the orthogonal distortion agrees with work performed at higher fluences. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental data collected at a laser fluence of 50 mJ/cm2, using an effective absorption depth of 300 nm as a single fit parameter. The overall good match of the simulations with the experimental data suggests that femtosecond laser excitation of Si - a cubic material - leads to the generation of a transverse strain wave with an amplitude comparable to the expected bi-polar longitudinal wave. The model fits reasonably well the experimental d… p. 9

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 AI systems can perform time and depth resolved snapshots of strain field upon laser excitation of bulk crystalline Si, as demonstrated by using X-ray free-electron lasers to probe ultrafast dynamics.

  2. AI systems can simulate transient changes and strain fields by combining UDD simulations with a 3D version of the model of Thomsen et al. to describe the generation and propagation of ps strain pulses in materials like Si.

  3. Improved systems can use a waterfall representation of experimental profiles as a function of X-detector and pump-probe time delay to unravel strain wave propagation dynamics, such as distinguishing between longitudinal sound waves (LSS) and transverse sound waves (TSS).

  4. AI systems can predict the position in depth of the strain wave for the first instants after laser excitation by comparing pristine signals to distorted signals using trigonometric relations derived from experimental data.

  5. The AI system can incorporate non-linear absorption effects, treating the laser absorption depth as a parameter to account for free-carrier and multi-photon absorption during simulations, which improves matching with experimental profiles.

Related papers