Photoinduced phase heterogeneity and charge localization in SnSe

Using time-resolved multi-terahertz spectroscopy, this study reveals that photoexcitation in SnSe induces a non-thermal, ultrafast nucleation of higher-symmetry semi-metallic domains within 200 fs, leading to phase heterogeneity, charge localization, and a suppression of long-range transport at intermediate pump fluences.

The Gist

The Big Picture: Shaking the Crystal

Imagine a block of Tin Selenide (SnSe) as a crowded dance floor. In its normal, calm state (at room temperature), the dancers (atoms) are arranged in a specific, slightly wobbly pattern called the Pnma phase. They are moving to a specific rhythm, but they aren't moving very fast, and the "electricity" (the flow of dancers) has a hard time getting through the crowd.

The scientists in this paper wanted to see what happens if they hit this dance floor with a super-fast, powerful flash of light (a laser pulse). They wanted to know: Does the whole floor change its dance style instantly, or does it turn into a chaotic mix of different dance styles?

The Experiment: The Ultra-Fast Camera

To watch this happen, they used a special tool called Terahertz (THz) spectroscopy. Think of this as a super-fast camera that can take pictures of the dance floor not just once, but thousands of times in a single second.

They shined a laser pulse (the "pump") to excite the dancers, and then used the THz camera (the "probe") to watch how the electricity flowed and how the atoms vibrated over time. They did this at different "loudness" levels (fluences) of the laser flash.

What They Found: The "Traffic Jam" and the "New Dance"

1. The Traffic Jam (Charge Localization)

When they used a low-power laser flash, the dancers (electrons) could still move around freely, like cars on an open highway. The electricity flowed well.

But when they turned up the power to a medium level, something strange happened. The electricity flow suddenly stopped, even though there were plenty of excited dancers.

• The Analogy: Imagine the laser flash causes the dance floor to suddenly sprout thousands of tiny, invisible walls. The dancers are still moving, but they are trapped in small, isolated rooms. They can't get from one side of the floor to the other.
• The Science: The researchers call this charge localization. The laser created a "phase heterogeneity," meaning the material became a patchwork quilt of different states. Some parts were still the old "wobbly" style, and new parts were trying to be a different style. The boundaries between these patches acted like walls, blocking the long-distance flow of electricity.

2. The New Dance Move (Phase Transition)

While the electricity was getting stuck, the scientists looked closely at how the atoms were vibrating (the "phonons").

• The Observation: One specific vibration (a "B2u mode") started to get sharper and faster (blueshifted). Also, a brand new vibration appeared that didn't exist before.
• The Analogy: It's like the dancers in the middle of the floor suddenly stopped doing the "wobbly" dance and started doing a completely different, more symmetrical, and efficient dance (the Immm phase).
• The Twist: The whole floor didn't change at once. Instead, small islands of this "new dance" formed inside the "old dance" crowd. This is why the electricity got stuck—the dancers in the new islands couldn't easily talk to the dancers in the old islands.

3. The "Plasmonic Peak" (The Crowd's Pulse)

At the highest energy levels, they saw a giant spike in the data that looked like a collective pulse of the crowd.

• The Analogy: This is like a "stadium wave." Even though the dancers are in different groups (some old dance, some new dance), the whole crowd starts swaying together in a specific rhythm. This "wave" (a plasmonic resonance) slowly changed its pitch over time, showing that the new dance islands were growing and then slowly relaxing back to the original state.

The "Threshold" Moment

The most exciting discovery was finding a tipping point.

• Below a certain laser power, nothing much happened.
• Right around 3.1 mJ/cm², the material suddenly switched behaviors. The "traffic jam" appeared, and the "new dance" started.
• This suggests that you don't need to melt the whole crystal to change its properties; you just need enough energy to nucleate (start) these new, high-performance islands within the old material.

Why Does This Matter?

Tin Selenide is famous for being a great thermoelectric material (it turns heat into electricity). However, it usually only works well at high temperatures.

• The Hope: If we can use light to force the material into this "new dance" state (the Immm phase) at room temperature, we might be able to make super-efficient energy converters that work all the time, not just when it's hot.
• The Catch: The new state is unstable. It's like a sandcastle; it forms quickly when you hit it with water (the laser), but then the waves (heat and time) wash it away, and it returns to being a pile of sand (the old phase) within a few billionths of a second.

Summary

The scientists used a super-fast laser to "poke" a crystal of Tin Selenide. They discovered that if you poke it hard enough, the crystal doesn't just get hotter; it breaks into a patchwork of different internal structures. This creates a "traffic jam" for electricity and forces the atoms to try out a new, more symmetrical dance. This happens so fast (in less than a trillionth of a second) that it's like watching a movie in slow motion, revealing that the secret to better energy materials might lie in creating these temporary, mixed-up dance floors.

Technical Summary

1. Problem Statement

Tin selenide (SnSe) is a record-setting thermoelectric material, particularly in its high-temperature $Cmcm$ phase. However, its room-temperature $Pnma$ phase also exhibits strong electron-phonon coupling and anharmonicity. While previous studies have suggested that optical excitation can induce ultrafast transitions to higher-symmetry phases (such as $Immm$ or $Fm\bar{3}m$), the exact mechanism, the nature of the phase coexistence (heterogeneity), and the impact on charge transport remain unclear. Specifically, there is a lack of direct evidence regarding:

• How photoexcitation alters long-range carrier transport in the presence of potential phase domains.
• The dynamics of lattice modes during non-equilibrium transitions.
• Whether a complete phase transition occurs or if a heterogeneous mix of phases (nucleation) is formed.

2. Methodology

The authors employed a combination of ultrafast time-resolved multi-terahertz (THz) spectroscopy and density-functional theory (DFT) calculations.

• Experimental Setup:

  • Sample: Exfoliated single-crystal SnSe ($2.5 \times 2.5 \times 0.5 \, \mu m$) mounted on a diamond substrate.
  • Excitation: 35 fs, 800 nm optical pump pulses with fluences ranging from $0.1$ to $7.5 \, mJ/cm^2$.
  • Probe: Air-plasma generated THz pulses covering a broadband spectral range of 0.5 – 11 THz (2 – 45 meV) with $\sim 40$ fs temporal resolution.
  • Conditions: Measurements performed at 80 K to isolate non-thermal responses.
  • Analysis: The complex THz transmission was converted to optical conductivity ($\tilde{\sigma}(\omega)$). The spectra were globally fitted using a Drude-Smith model (for free carrier transport and localization) and a series of Lorentzian oscillators (for optical phonons and high-frequency resonances).

• Theoretical Framework:

  • DFT/cDFT: Calculations were performed using Quantum ESPRESSO with constrained density-functional theory (cDFT) to model photoexcited electron-hole plasmas.
  • Phonon Calculations: Stochastic self-consistent harmonic approximation (SSCHA) was used to include anharmonic effects and simulate spectral functions for ground-state $Pnma$, photoexcited $Pnma$, photoexcited $Immm$, and $Fm\bar{3}m$ phases.

3. Key Contributions

• Observation of Phase Heterogeneity: The study provides direct spectroscopic evidence for the non-thermal nucleation of higher-symmetry domains ($Immm$) within the $Pnma$ background, rather than a uniform, complete phase transition.
• Charge Localization Mechanism: It demonstrates that photoexcitation induces a transition from long-range transport to charge localization due to phase disorder, quantified via the Drude-Smith backscattering parameter.
• Identification of a New Phonon Mode: The discovery of a transient phonon mode at $\sim 3.0$ THz, which is absent in the equilibrium $Pnma$ phase, serves as a fingerprint for the mixed-phase state.
• Threshold Fluence: The identification of a critical fluence threshold ($\approx 3.1 \, mJ/cm^2$) required to trigger these non-equilibrium structural and electronic changes.

4. Key Results

A. Charge Transport Dynamics (Low Frequency: 0.5 – 2 THz)

• Suppression of DC Conductivity: At fluences below $3 \, mJ/cm^2$, a standard Drude response is observed. However, as fluence increases, the low-frequency (DC) photoconductivity is suppressed, and the conductivity peak blueshifts away from $\omega = 0$.
• Localization: This suppression is modeled by the Drude-Smith parameter $c \to -1$, indicating strong carrier backscattering and localization. This suggests the formation of insulating/metallic domain boundaries that interrupt long-range percolation.
• Threshold Behavior: The localization effect is most pronounced at $3.1 \, mJ/cm^2$. At the highest fluence ($7.5 \, mJ/cm^2$), the parameter $c$ slightly recovers, suggesting a possible percolative transition where domains become dense enough to reconnect.

B. Lattice Dynamics (Mid Frequency: 2 – 4.5 THz)

• Phonon Narrowing and Shifting: The $B^2_{1u}$ optical phonon mode (normally at $\sim 3.7$ THz) exhibits dynamic narrowing and a blueshift with increasing fluence. This is counter-intuitive (as heating usually broadens modes) and indicates a reduction in phonon anharmonicity, consistent with a transition toward a higher-symmetry structure.
• Emergence of a New Mode: A new phonon feature appears at $\sim 3.0$ THz at fluences $\ge 3.1 \, mJ/cm^2$. This mode does not exist in the pure $Pnma$ or $Cmcm$ phases.
• Theoretical Correlation: DFT calculations confirm that a mixture of $Pnma$ and photoexcited $Immm$ phases produces a new mode in this frequency range, supporting the experimental observation of phase heterogeneity.

C. High-Frequency Response (4.5 – 11 THz)

• Plasmonic Resonance: A high-frequency Lorentzian peak appears, which redshifts over time (on a $\sim 90$ ps scale). This is attributed to phase heterogeneity, where local metallic domains create depolarization fields (effective medium behavior).
• Dirac-like Signatures: At early times ($< 2$ ps) and $3.1 \, mJ/cm^2$, residuals in the conductivity fit resemble interband transitions in Dirac materials (similar to graphene), suggesting the transient formation of Dirac-like electronic states within the nucleated domains.

5. Significance and Conclusions

• Mechanism of Transition: The paper establishes that photoexcitation in SnSe does not immediately create a bulk uniform phase. Instead, it triggers the nucleation of semi-metallic, higher-symmetry domains ($Immm$) within the $Pnma$ matrix within 200 fs.
• Timescales: The system evolves through distinct stages:
  1. < 2 ps: Nucleation of domains, carrier localization, and emergence of Dirac-like states.
  2. 2 – 5 ps: Maximum phase heterogeneity, phonon narrowing, and peak plasmonic response.
  3. > 200 ps: Relaxation back to the homogeneous $Pnma$ semiconducting phase.
• Implications for Thermoelectrics: The findings suggest that the high thermoelectric performance of SnSe is intimately linked to its structural anharmonicity and that this can be dynamically tuned via light. The ability to induce a metastable, higher-symmetry phase (potentially a topological crystalline insulator) offers a pathway for controlling electronic and thermal transport properties on ultrafast timescales.
• Future Directions: The authors propose that spatially resolved techniques (e.g., THz-Scanning Tunneling Microscopy) are needed to directly image the nucleated domains and confirm the spatial distribution of the phase transition.

In summary, this work provides a comprehensive picture of how light drives SnSe into a complex, heterogeneous non-equilibrium state, characterized by charge localization and the coexistence of multiple structural phases, fundamentally altering its transport properties.