Direct observation of ultrafast defect-bound and free exciton dynamics in defect-engineered WS$_2$ monolayers

This study utilizes ultrafast optical spectroscopy to directly observe and characterize the sub-picosecond formation, trapping, and coherent interconversion dynamics between free and defect-bound excitons in alkali metal halide-assisted CVD-grown monolayer WS$_2$ with high defect densities.

The Gist

Imagine a tiny, ultra-thin sheet of material called WS₂ (Tungsten Disulfide). Think of this sheet as a perfectly smooth, high-speed highway for tiny particles of light and energy called excitons. In a perfect world, these particles zip along this highway effortlessly, making the material great for future electronics and quantum computers.

However, real-world materials aren't perfect. They have defects—tiny potholes, missing bricks, or debris on the road. In the past, scientists thought these defects were just bad news: they slowed down traffic and caused energy to get lost.

But this new research discovered something surprising: Defects aren't just potholes; they are also secret underground tunnels and high-speed express lanes.

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

1. Creating the "Defect-Rich" Highway

The researchers wanted to study these defects, but they are usually rare and scattered randomly, like finding a single specific grain of sand on a beach. To solve this, they used a special cooking method (called "alkali metal halide-assisted CVD") to bake the WS₂ sheets.

Think of this like adding a special spice (Sodium Bromide) to the recipe. This spice didn't just flavor the dish; it created a high concentration of specific defects (missing sulfur atoms) right in the center of the crystal. This gave them a "laboratory" where the defects were so dense they could be easily studied, rather than hidden.

2. The Two Types of Traffic

Once they had their defect-rich sheet, they shined a laser on it to watch how the energy moved. They saw two types of traffic:

• Free Excitons (The Highway Runners): These are the energy particles zooming freely along the perfect parts of the sheet.
• Defect-Bound Excitons (The Tunnel Dwellers): These are particles that get trapped in the "potholes" (defects).

Usually, scientists thought it took a while for the "Runners" to get caught in the "Tunnels." But the researchers used a super-fast camera (ultrafast spectroscopy) that takes pictures in femtoseconds (one quadrillionth of a second).

The Surprise: They found that the "Runners" don't just wander into the tunnels slowly. Instead, they get trapped almost instantly (within 300 femtoseconds). It's as if the moment the car hits the road, it immediately knows exactly where the secret tunnels are and dives in.

3. The Magic "Teleportation" (Up-Conversion)

The most mind-blowing discovery happened when they shined a laser with low energy (a dim light) specifically at the "Tunnel Dwellers."

Normally, if you give a ball a small push, it can't jump over a high wall. In physics terms, you can't turn low-energy light into high-energy light easily. It's like trying to fill a bucket with a teaspoon when you need a firehose.

However, the researchers saw that the "Tunnel Dwellers" were somehow teleporting back onto the "Highway" and becoming "Free Runners" with more energy than they started with!

• The Analogy: Imagine a person sitting in a deep hole (the defect). You give them a tiny nudge, and instead of staying down, they suddenly pop up into the air, flying higher than the top of the hole, ready to run fast again.
• How? The scientists ruled out normal physics (like bouncing off a wall or using heat). They concluded that the "Tunnel Dwellers" and "Highway Runners" are coherently coupled.
• The Metaphor: Think of them as two dancers holding hands. Even if one is in a deep pit and the other is on a stage, they are so tightly connected that when one moves, the other moves instantly. They aren't just bouncing off each other; they are sharing a single, synchronized rhythm. This allows energy to jump from the low-energy state to the high-energy state almost instantly (in about 150 femtoseconds).

Why Does This Matter?

This discovery changes how we think about "broken" materials.

• Old View: Defects are trash that ruins electronics.
• New View: Defects are features. We can engineer them to act as quantum switches or energy boosters.

Because these "teleportation" events happen so fast and are so efficient, we might be able to build:

1. Super-fast quantum computers that use these defects to store information.
2. Better solar cells that capture more energy by using these "tunnels" to boost weak light into strong electricity.
3. New types of lasers that work at room temperature.

The Bottom Line

The scientists took a material with "flaws," turned those flaws into a super-highway of defects, and discovered that energy can zip between different states faster than we ever thought possible. It turns out that in the quantum world, a little bit of brokenness can actually make things work better and faster.

Technical Summary

1. Problem Statement

Two-dimensional transition metal dichalcogenides (TMDCs), such as WS₂, exhibit unique optical and electronic properties, but their performance is heavily influenced by intrinsic point defects (e.g., sulfur vacancies). While defects can degrade carrier mobility and quantum yield, they also enable emergent phenomena like single-photon emission and long valley lifetimes, which are crucial for quantum technologies.

Despite the importance of these defect-bound excitons (DBEs), their ultrafast dynamics (formation, trapping, and interconversion with free excitons) have remained elusive due to:

• Weak Oscillator Strength: DBEs have localized wavefunctions, resulting in weak transient optical absorption signals that are difficult to detect.
• Random Defect Distribution: Conventional synthesis methods produce random defect distributions, making it impossible to isolate and study specific defect-rich regions with high spatial resolution.
• Indirect Inference: Previous studies relied on indirect interpretations of free exciton dynamics or theoretical models, lacking direct experimental evidence of DBE formation timescales and coupling mechanisms.

2. Methodology

The authors employed a multi-modal approach combining advanced synthesis, nanoscale characterization, and ultrafast spectroscopy:

• Sample Synthesis:
  • Control Sample (#1): WS₂ monolayers grown via standard Chemical Vapor Deposition (CVD).
  • Defect-Engineered Sample (#2): WS₂ monolayers grown using alkali metal halide (NaBr)-assisted CVD. This method intentionally introduced a high density of defects, specifically creating spatially heterogeneous defect distributions (high density at the crystal center, lower at edges).
• Characterization:
  • Photoluminescence (PL) & Raman Mapping: Used to correlate defect density with optical emission. The ratio of LA to A'₁ Raman modes served as a proxy for defect density.
  • Kelvin Probe Force Microscopy (KPFM): Mapped work function variations to confirm defect-induced Fermi level shifts.
  • STEM-HAADF: Atomic-resolution imaging to identify specific defect types (mono-sulfur vacancies $V_S$ and W-site defect complexes $S_WV_S$) and quantify their spatial distribution.
• Ultrafast Spectroscopy:
  • Pump-Probe Microscopy: Utilized a broadband white-light continuum probe and tunable pump wavelengths (400 nm for above-band-edge; 595–720 nm for band-edge).
  • Measurement: Transient reflectivity ($\Delta R/R_0$) was measured to track the population dynamics of Free Excitons (X) and Defect-Bound Excitons (D) with femtosecond temporal resolution (~100–150 fs).

3. Key Contributions & Results

A. Defect Characterization and Spatial Mapping

• Defect Identification: STEM analysis confirmed that the NaBr-assisted growth produced a high density of mono-sulfur vacancies ($V_S$) and $S_WV_S$ complexes.
• Spatial Correlation: The center of the crystal exhibited 70% higher defect density (specifically $V_S$) compared to the edges. This spatial variation directly correlated with a strong, low-energy PL emission peak (1.88 eV) in the center, attributed to DBEs, while the edges showed dominant free exciton emission (~1.96 eV).
• Energy Separation: The energy difference between free excitons (X) and defect-bound excitons (D) was measured at ~80 meV.

B. Ultrafast Dynamics under Above-Band-Edge Excitation (400 nm)

• Simultaneous Formation: Upon hot carrier relaxation, both free (X) and defect-bound (D) excitons formed simultaneously within ~300 fs. This contradicts previous assumptions that defect trapping is a slower, sequential process (1–100 ps).
• Population Imbalance: While formation is simultaneous, the D state has a significantly shorter lifetime than the X state. This leads to a population difference and dynamic trapping of free excitons into defect states within the 1–100 ps window.

C. Ultrafast Interconversion and Up-Conversion

The study investigated the coupling between X and D states using band-edge excitation:

• X $\to$ D Conversion: Exciting the free exciton band (595 nm) resulted in the rapid formation of defect-bound excitons with a rise time of ~142 fs. This is much faster than phonon-mediated trapping models predict.
• D $\to$ X Up-Conversion: Exciting the defect-bound state (665 nm) resulted in the population of the free exciton state (612 nm) with a rise time of ~155 fs.
  • Energy Barrier: This process involves an energy gain of ~80 meV, which is energetically unfavorable via thermal activation at room temperature ($k_BT \approx 26$ meV).
  • Wavelength Independence: Remarkably, up-conversion was observed even when the pump was detuned by ~300 meV below the free exciton resonance (up to 720 nm pump), with the rise time remaining constant (~150 fs).
  • Mechanism Ruling Out:
    • Two-Photon Absorption (TPA): Ruled out because the signal intensity scaled sublinearly ($F^{0.48}$) with pump fluence, whereas TPA would scale quadratically ($F^2$).
    • Phonon-Mediated Scattering: Ruled out because the timescale (~150 fs) is orders of magnitude faster than the picosecond timescales predicted by Arrhenius equations for phonon-assisted processes.
  • Proposed Mechanism: The authors attribute this ultrafast, wavelength-independent interconversion to coherent coupling (Dexter-like interaction) between the X and D states, mediated by valley-spin polarizations. This allows for instantaneous population transfer without the need for thermal activation or multi-phonon scattering.

4. Significance

• Direct Observation: This work provides the first direct experimental evidence of ultrafast defect-bound exciton formation and interconversion, moving beyond indirect inferences.
• Timescale Revision: It establishes that defect trapping and interconversion occur on sub-150 fs timescales, significantly faster than the previously accepted 1–100 ps range.
• Coherent Coupling: The discovery of coherent coupling between free and defect-bound excitons suggests that defects in TMDCs are not merely passive traps but active participants in quantum coherent dynamics.
• Applications: These findings open new avenues for:
  • Quantum Photonics: Engineering defect states for efficient single-photon sources.
  • Valleytronics: Utilizing spin-valley polarization in defect states for information processing.
  • Optoelectronics: Designing devices that leverage efficient up-conversion and energy transfer between defect and free exciton states.

In summary, the paper demonstrates that by engineering defect densities in WS₂, researchers can directly observe and manipulate the ultrafast quantum dynamics of excitons, revealing a regime of coherent coupling that challenges traditional phonon-scattering models.