Phonon dynamics of a bulk WSe$_2$ crystal excited by ultrashort near-infrared pulses

Pump-probe reflectivity measurements on a bulk WSe$_2$ crystal excited by ultrashort near-infrared pulses reveal phonon dynamics characterized by intense oscillations around 7.5 THz and smaller peaks at 4.0 and 11.5 THz, which are well reproduced by simulations superimposing three distinct oscillatory modes.

The Gist

Imagine a crystal of Tungsten Diselenide (WSe₂) not as a hard, static rock, but as a giant, intricate trampoline made of atoms. Inside this trampoline, the atoms are constantly jiggling and vibrating, much like a crowd of people doing a synchronized dance.

This paper is about a team of scientists who decided to "wake up" this crystal and watch how it dances using an incredibly fast camera.

Here is the story of what they found, explained simply:

1. The Setup: The Ultra-Fast Camera Flash

Usually, if you want to see something move, you take a photo. But atoms vibrate so fast that a normal camera would just see a blur. To see them clearly, the scientists used ultrashort pulses of light (near-infrared lasers).

Think of these laser pulses as camera flashes that last only 20 femtoseconds. To put that in perspective: a femtosecond is to a second what a second is to about 31.7 million years. It's so fast that the light pulse hits the crystal and bounces off before the atoms have even had time to take a single step in their dance.

2. The Experiment: The "Pump and Probe"

The scientists used a two-step process:

• The Pump: They hit the crystal with a strong laser flash. This is like kicking the trampoline. It wakes up the atoms and makes them start vibrating in unison.
• The Probe: A split-second later, they sent a weaker laser flash to check how the trampoline was moving. By waiting a tiny bit longer for each subsequent flash, they could build a movie of the vibration.

3. The Discovery: A Complex Dance Routine

When they looked at the data, they expected to see the atoms vibrating at one main speed. In the world of physics, this main speed is about 7.5 Terahertz (which is 7.5 trillion vibrations per second).

However, the data was more interesting than a simple vibration. It looked like the crystal was trying to do three different dances at the same time, but they were slightly out of sync.

• The Main Dancers: There were two very similar vibrations happening at 7.45 THz and 7.49 THz.
• The Soloist: There was a third, slightly faster vibration at 7.7 THz.

The "Rise" Mystery:
Usually, when you kick a trampoline, it jumps immediately. But here, the scientists noticed something strange: the vibration didn't start strong immediately. Instead, it took about 1 picosecond (a trillionth of a second) to build up to its full strength before fading away.

The Analogy:
Imagine three drummers starting to play a beat.

• Drummer A and Drummer B start playing a rhythm that is almost the same, but they are slightly out of step with each other.
• Drummer C starts playing a slightly faster rhythm, but they are playing the opposite beat (when A and B hit the drum, C hits the space between).
• At first, their beats cancel each other out, so the sound is quiet.
• But as they keep playing, their different phases line up perfectly, and suddenly the sound gets louder and louder for a moment before they all start to get tired and stop.

The scientists realized that the "rise" in the signal wasn't a single vibration getting stronger; it was the perfect timing of these three different vibrations overlapping to create a bigger wave.

4. The Hidden Notes

Besides the main 7.5 THz "song," the scientists also heard two quieter "notes" in the background:

• A low note at 4.0 THz.
• A high note at 11.5 THz.

These were so quiet that they were invisible in the raw "movie" of the vibration. You only saw them when the scientists used a mathematical tool (called a Fourier Transform) to separate the sounds, like using a prism to split white light into a rainbow. These extra notes confirmed that the crystal has a complex internal structure, vibrating in multiple ways simultaneously.

Why Does This Matter?

Tungsten Diselenide is a "super-material" being studied for future electronics and solar cells. Understanding exactly how its atoms vibrate is like understanding the engine of a car. If we know how the atoms dance, we can design better, faster, and more efficient devices.

In a nutshell:
The scientists used a super-fast laser to kick a crystal and watch it vibrate. They discovered that the crystal doesn't just vibrate at one speed; it performs a complex, three-part dance where different vibrations combine to create a temporary "loud" moment before fading away. This helps us understand the hidden mechanics of materials that could power our future technology.

Technical Summary

1. Problem Statement

While the optical and electrical properties of Transition Metal Dichalcogenides (TMDCs) like Tungsten Diselenide (WSe₂) are well-studied, the ultrafast dynamics of their coherent phonons remain partially understood, particularly regarding high-frequency modes and complex temporal behaviors.

• Limitations of Previous Studies: Prior investigations using transient reflection or transmission with tens-of-femtosecond pulses have primarily identified a dominant coherent phonon oscillation at ~7.5 THz (250 cm⁻¹).
• The Gap: These studies often fail to resolve higher-frequency phonon modes or explain anomalous temporal behaviors, such as the delayed rise in oscillation amplitude observed immediately after pump excitation. Furthermore, the limited number of observed peaks in time-domain measurements contrasts with the richer spectra obtained via Raman spectroscopy.
• Objective: The authors aim to investigate coherent phonon dynamics in bulk WSe₂ using ultrashort (~20 fs) near-infrared pulses to excite and resolve higher-frequency phonons and to elucidate the mechanism behind the non-instantaneous rise of the oscillation amplitude.

2. Methodology

The study employed a femtosecond pump-probe transient reflectivity technique combined with spectral analysis and numerical simulation.

• Experimental Setup:
  • Source: A Ti:sapphire oscillator (FEMTO-LASER Rainbow) generating near-infrared pulses.
  • Pulse Characteristics: Group-velocity dispersion was compensated using chirped mirrors, resulting in a pulse width of ~20 fs.
  • Configuration: The pulse was split (3:1 ratio) into pump and probe beams. The pump power was set to ~90 mW.
  • Detection: A balanced photodetector system measured the differential reflectivity ($\Delta R/R$) as a function of pump-probe delay ($\tau$).
  • Sample: A commercially available bulk single crystal of WSe₂.
• Complementary Measurements:
  • Raman Spectroscopy: Performed using a 784.7 nm continuous-wave laser to establish a baseline for phonon frequencies and validate the presence of specific modes.
• Data Analysis:
  • Fourier Transform (FT): Applied to the transient reflectivity signals (delay range 0.2–4.0 ps) to extract frequency components.
  • Short-Time Fourier Transform (STFT): Used to analyze the time-evolution of the frequency components.
  • Numerical Simulation: The experimental oscillation data was modeled by superimposing three damped harmonic oscillations with distinct frequencies, amplitudes, decay times, and phases.

3. Key Results

• Transient Reflectivity Dynamics:
  • The signal showed a strong positive response at $\tau=0$ with a superimposed oscillation.
  • Anomalous Rise: Unlike typical coherent phonon generation (which occurs within the pulse duration), the oscillation amplitude in WSe₂ gradually increased for approximately 1 ps (covering ~10 oscillation cycles) before decaying exponentially.
• Frequency Spectrum (FT Analysis):
  • Main Peak: An intense peak at 7.48 THz (corresponding to the Raman active $A_{1g}$ and overlapping $E_{2g}$ modes).
  • Secondary Peaks: Weak but distinct peaks were identified at 4.00 THz (low-frequency) and 11.56 THz (high-frequency). These were too weak to be visually resolved in the raw time-domain signal but were confirmed by the FT spectrum and Raman data.
• Simulation Success:
  • The complex temporal behavior (specifically the 1 ps rise) was successfully reproduced by simulating the superposition of three damped oscillations:
    1. 7.45 THz (Long decay, $\tau \approx 10$ ps)
    2. 7.49 THz (Medium decay, $\tau \approx 2$ ps)
    3. 7.70 THz (Short decay, $\tau \approx 1$ ps)
  • Phase Interference: The simulation revealed that the 7.7 THz component has a phase difference of $\pi$ relative to the 7.45 and 7.49 THz components. This destructive/constructive interference pattern explains the initial suppression and subsequent rise of the net amplitude.

4. Key Contributions

1. Detection of High-Frequency Modes: By utilizing ultrashort (~20 fs) pulses, the authors successfully excited and detected the 11.5 THz phonon mode, which is often missed with longer pulse durations due to bandwidth limitations.
2. Mechanism of Amplitude Rise: The paper provides a physical explanation for the delayed rise in coherent phonon amplitude. It demonstrates that this is not a single-mode phenomenon but the result of interference between multiple closely spaced phonon modes (7.45, 7.49, and 7.7 THz) with different phases and decay rates.
3. Correlation with Raman Data: The study bridges the gap between time-domain pump-probe measurements and frequency-domain Raman spectroscopy, confirming that the weak peaks observed in the FT spectrum (4.0 and 11.5 THz) correspond to coupled modes observed in Raman spectra.

5. Significance

• Fundamental Understanding: The work refines the understanding of electron-phonon coupling and phonon dynamics in bulk TMDCs, showing that "single" peaks in time-domain data often represent complex superpositions of multiple modes.
• Methodological Advancement: It highlights the critical importance of pulse duration in phonon spectroscopy; shorter pulses are essential for accessing the high-frequency end of the phonon spectrum.
• Material Characterization: The findings offer a more comprehensive view of the vibrational landscape of WSe₂, which is crucial for optimizing its performance in optoelectronic devices, superconductors, and topological insulators where phonon interactions play a dominant role.
• Model Validation: The successful simulation of the transient response using a three-component model provides a robust framework for analyzing complex coherent phonon dynamics in other layered materials.