Disentangling High Harmonic Generation from Surface and Bulk States of a Topological Insulator

This study demonstrates that by tuning the thickness of Bi$_2$Se$_3$ thin films and applying a quasi-static terahertz field, researchers can effectively isolate and distinguish the high harmonic generation contributions of topological surface states from bulk states, thereby resolving debates on extracting topological signatures in nonlinear optical responses.

The Gist

Imagine a topological insulator (specifically a material called Bi₂Se₃) as a very special kind of chocolate bar.

• The Inside (Bulk): The chocolate itself is a solid, insulating block. It doesn't conduct electricity well; it's like a wall that stops electrons from moving freely.
• The Wrapper (Surface): But this chocolate has a magical, conductive foil wrapper. On this surface, electrons can zip around freely, and they have a special "spin" property (like tiny spinning tops) that is locked to their direction of travel.

Scientists want to study this magical wrapper because it holds secrets about the future of electronics and quantum computing. However, there's a big problem: The wrapper is incredibly thin (less than 2 nanometers, which is thinner than a human hair by a million times), while the chocolate block is much thicker.

When scientists shine a powerful laser at this material to study it, the laser hits both the wrapper and the chocolate. The signal from the thick chocolate (the "bulk") is so loud that it drowns out the whisper of the magical wrapper (the "surface"). It's like trying to hear a single violin soloist in a stadium full of cheering fans.

This paper is about how the researchers finally managed to silence the crowd and hear the violin.

The Experiment: Shining a Light on the Problem

The researchers used a technique called High-Harmonic Generation (HHG). Think of this as hitting the material with a super-fast, intense laser pulse (a "MIR" pulse). This is like hitting a drum so hard that it doesn't just make a "thud," but starts singing back in much higher, shriller notes (harmonics).

By analyzing these high-pitched notes, scientists can learn about the material's internal structure. But, as mentioned, the "bulk" chocolate and the "surface" wrapper both sing, and they sound very similar.

The Solution 1: The "Thin Crust" Trick

The first way the team separated the signals was by changing the size of the chocolate bar.

• The Thick Bar (50 nm): When they used a thick sample, the "bulk" signal was dominant. The wrapper was there, but its voice was lost in the roar of the chocolate.
• The Ultra-Thin Crust (6 nm): They then shaved the chocolate down to an ultra-thin layer, almost like a cracker. Now, there is very little "bulk" chocolate left. The signal is almost entirely from the two surfaces (top and bottom).

The Result: By making the sample super thin, they amplified the "surface" voice. They found that the surface produces a unique type of "even-numbered" musical notes (harmonics) that the bulk cannot produce. In the thin sample, these unique notes were loud and clear.

The Solution 2: The "Push and Pull" (The THz Field)

Even with the thin sample, they wanted to be 100% sure they were looking at the surface and not just a tiny bit of bulk. So, they added a second tool: a Terahertz (THz) field.

Imagine the laser pulse is a strong wind blowing electrons around. The THz field is like a gentle, slow-moving hand that pushes the electrons in a specific direction before the wind hits them.

• The Bulk (The Chocolate): The inside of the material is perfectly symmetrical. If you push it left or right, it reacts the same way. The "hand" (THz field) doesn't change its song.
• The Surface (The Wrapper): The surface is not symmetrical. It has a built-in "compass" (called the Shift Vector and Berry Curvature). If the "hand" pushes in the same direction as the compass, the surface sings one way. If the hand pushes against the compass, it sings differently.

The Result: By toggling the direction of this gentle "hand," the researchers could see the surface signal change its tune, while the bulk signal stayed exactly the same. This allowed them to mathematically subtract the bulk noise and isolate the pure surface signal.

Why Does This Matter?

For years, scientists have been arguing: "Can we actually see the 'topological' magic of these materials using light?" Some said yes, others said the signals were just noise from the bulk.

This paper says: "Yes, we can, but you have to know how to listen."

By combining thin films (to minimize the bulk) and THz fields (to act as a switch that only the surface responds to), the team has built a new "microscope" for light. They can now:

1. Isolate the surface electrons from the bulk.
2. Measure the unique quantum properties (like the Berry Curvature) that make these materials special.
3. Prove that we can reliably extract "topological signatures" using light.

The Big Picture Analogy

Imagine you are trying to study the wind patterns on the roof of a skyscraper (the surface), but you are standing inside the building (the bulk). The wind inside is chaotic and loud, making it impossible to hear the wind on the roof.

• Method 1: You go to the roof (make the sample thin). Now you are close to the source, but you still hear the wind from the building's ventilation system (the bulk) mixing in.
• Method 2: You use a special fan (the THz field) that only affects the air outside the building because of how the roof is shaped. By turning the fan on and off, you can tell exactly which gusts of wind are coming from the roof and which are just noise from inside.

This paper provides the blueprint for that special fan, opening the door to a new era of studying quantum materials with light.

Technical Summary

1. Problem Statement

Three-dimensional topological insulators (TIs), such as $\text{Bi}_2\text{Se}_3$, possess insulating bulk states and conductive topological surface states (TSSs) with unique spin-momentum locking and non-trivial topology. High-order harmonic generation (HHG) has emerged as a powerful all-optical probe for condensed matter systems. However, a critical challenge in using HHG to study TIs is disentangling the signal from the TSSs from the overwhelming background of the bulk states.

• The Challenge: The bulk band gap of TIs is small, allowing strong laser fields to easily excite bulk carriers. Furthermore, the penetration depth of the driving mid-infrared (MIR) light and the generated harmonics (tens to hundreds of nanometers) far exceeds the penetration depth of TSSs (<2 nm). Consequently, bulk contributions often dominate the harmonic emission, obscuring the topological signatures.
• The Debate: There is an ongoing scientific debate regarding whether HHG can reliably extract topological signatures. While some studies suggest non-integer harmonics or ellipticity dependence indicate topology, others argue these features can arise from trivial mechanisms (e.g., pulse chirp or bulk anisotropy). A definitive method to isolate surface responses is required to resolve this.

2. Methodology

The authors employed a combination of thin-film engineering and two-color field perturbation to isolate and characterize the HHG responses of $\text{Bi}_2\text{Se}_3$.

• Sample Preparation: High-quality $\text{Bi}_2\text{Se}_3$ thin films were grown via Molecular Beam Epitaxy (MBE) on $\text{Al}_2\text{O}_3$ substrates. Two specific thicknesses were compared:
  • Ultrathin (6 nm): Approximately two quintuple layers, minimizing bulk volume and maximizing the surface-to-bulk ratio.
  • Thick (50 nm): Represents a bulk-dominated sample where surface contributions are negligible relative to the volume.
• Experimental Setup:
  • Driving Fields: A strong, linearly polarized MIR pulse (3.6 $\mu$m, 60 fs, ~0.34 eV) drives the HHG process.
  • Perturbing Field: A weak, quasi-static, single-cycle Terahertz (THz) pulse (~0.5 THz) is applied simultaneously. The THz field acts as a symmetry-breaking knob.
  • Geometry: Transmission geometry with normal incidence. The crystal orientation was varied relative to the laser polarization (along $\Gamma M$ and $\Gamma K$ directions in reciprocal space).
• Theoretical Framework: The study utilized a tight-binding model for TSSs combined with gauge-invariant Semiconductor Bloch Equations (GI-SBEs) to simulate the dynamics, explicitly accounting for the shift vector ($R$) and Berry curvature ($\Omega$).

3. Key Contributions

The paper introduces two primary mechanisms to separate surface and bulk HHG contributions:

1. Thickness Control: Demonstrating that tuning film thickness alters the dominance of bulk vs. surface HHG.
2. Symmetry Breaking via THz Field: Using a quasi-static THz field to break dynamical symmetries differently for bulk (parity-invariant) and surface (parity-violating) states. This allows for the isolation of specific topological parameters ($R$ and $\Omega$).

4. Key Results

A. Thickness-Dependent HHG

• Total Yield: The total harmonic yield increases with film thickness, peaking around 20 nm due to absorption lengths and interference effects.
• Even-Order Harmonics: In the 6 nm film, even-order harmonics (which are forbidden in the inversion-symmetric bulk but allowed in the $C_{3v}$ surface) are enhanced by nearly two orders of magnitude compared to the 50 nm film.
• Mechanism: The enhancement in ultrathin films is attributed to the suppression of bulk-surface electron scattering. In thicker films, photoexcited bulk electrons scatter into the TSSs, creating hot electrons that increase scattering and suppress surface HHG efficiency. The 6 nm film minimizes this bulk volume, effectively isolating the surface response.

B. Disentangling via THz Perturbation (The "Knob")

The authors applied a THz field to break symmetries and probe the shift vector ($R$) and Berry curvature ($\Omega$).

• Probing the Shift Vector ($R$) along $\Gamma M$:

  • Theory: The bulk is parity-invariant ($R=0$), while TSSs are parity-violating ($R \neq 0$). The THz field couples with $R$ to induce a difference in harmonic yield depending on the relative orientation of the THz field and the crystal ($M^+$ vs. $M^-$).
  • Observation: In the 6 nm sample, a large contrast (yield difference) was observed between $M^+$ and $M^-$ orientations for both 6th and 7th harmonics, confirming a surface-dominated response. In the 50 nm sample, the contrast was negligible, confirming bulk dominance.
  • Significance: This proves that the THz field can selectively reveal the impact of the shift vector on odd-order harmonics, a feature previously thought to be insensitive to $R$ in single-color fields.

• Probing the Berry Curvature ($\Omega$) along $\Gamma K$:

  • Theory: Along $\Gamma K$, the shift vector vanishes, but the out-of-plane Berry curvature ($\Omega_z$) is non-zero. This induces perpendicular harmonic emission ($j_\perp \propto F \times \Omega$).
  • Observation: Without the THz field, only even-order perpendicular harmonics were generated. Applying the THz field suppressed even-order yields and enabled the generation of odd-order perpendicular harmonics.
  • Result: The 6 nm sample showed pronounced modulation of perpendicular harmonics across the spectrum, accurately matching simulations. The 50 nm sample showed minimal response, confirming that perpendicular odd-order generation is an exclusive signature of the TSSs.

5. Significance and Impact

• Resolution of the Debate: The study provides a robust, experimental method to isolate TSS responses from bulk backgrounds, addressing the skepticism regarding the reliability of HHG for detecting topological phases.
• New Probe for Topology: It establishes that the shift vector and Berry curvature have distinct, measurable impacts on HHG dynamics when probed with a two-color (MIR-THz) field.
• General Applicability: The strategy of varying the surface-to-bulk ratio via film thickness and exploiting intrinsic vs. field-induced symmetry breaking is broadly applicable to other topological materials and strong-field physics studies.
• Future Directions: The work opens pathways for exploring non-trivial topology in bulk states (e.g., band inversion signatures) and suggests that future experiments could compare topologically trivial vs. non-trivial surfaces (via doping) to further validate these topological signatures.

In summary, this paper successfully demonstrates that by combining nanoscale film engineering with symmetry-breaking THz perturbations, researchers can effectively isolate and characterize the unique topological contributions to high-harmonic generation in $\text{Bi}_2\text{Se}_3$, moving beyond the limitations of bulk-dominated signals.