Exciton radiative lifetimes in hexagonal diamond Ge and Si$_x$Ge$_{1-x}$ alloys

This study uses Bethe-Salpeter calculations to demonstrate that while pristine hexagonal diamond Ge exhibits extremely long excitonic radiative lifetimes inconsistent with reported strong room-temperature photoluminescence, alloying with Si or applying uniaxial strain can significantly reduce these lifetimes, though the ideal crystal alone cannot explain the experimental observations.

The Gist

The Big Mystery: The "Ghost" Light in Hexagonal Germanium

Imagine you have a new type of Germanium (Ge) crystal. It's shaped like a hexagon (think of a honeycomb) rather than the usual cube. Scientists recently looked at this material and shouted, "Wow! It glows incredibly bright at room temperature!" This was exciting because Germanium is usually a "bad" light emitter, and a bright light source made of pure Germanium could revolutionize computer chips and fiber optics.

However, the computer scientists (theoretical physicists) were confused. They ran simulations and said, "That's impossible. According to our math, this material should be almost invisible. It's like a ghost that shouldn't be able to cast a shadow."

The Conflict: The experimenters see a bright lightbulb; the theorists see a dark room.

The Investigation: Looking for the "Couple"

To solve this mystery, the authors of this paper decided to look closer at how electrons and "holes" (the empty spaces electrons leave behind) behave in this material.

Think of an electron and a hole as a dancing couple.

• In a normal light emitter (like Gallium Nitride): The couple is very energetic and loves to dance in the center of the room. When they stop dancing (recombine), they release a photon (a packet of light) instantly. It's a quick, efficient party.
• In this Hexagonal Germanium: The math suggested the couple is shy. They are stuck in a corner, and when they finally decide to dance, they do it so slowly and quietly that it takes them a long time to release a single photon.

The paper asks: Is the material actually a shy ghost, or is there something else going on?

The Three Experiments

The researchers tested three different scenarios to see how to make this "shy" material glow brighter.

1. The Pure Material (The Shy Ghost)

First, they looked at the pure, unaltered Hexagonal Germanium.

• The Result: They confirmed the theorists were right. The "dancing couple" (exciton) has a very weak connection to light.
• The Analogy: Imagine trying to get a shy person to sing on stage. They are there, but they whisper so quietly that no one hears them. The paper calculates that it would take over 100,000 seconds (more than a day) for this material to emit a single photon naturally.
• Conclusion: The bright light seen in experiments cannot be coming from the perfect crystal itself. Something else must be making it glow.

2. Mixing in Silicon (The Social Mixer)

Next, they mixed Germanium with Silicon (Si) to create an alloy.

• The Result: This helped a little bit. Mixing the atoms broke some of the "rules" that were keeping the couple shy.
• The Analogy: It's like inviting a few outgoing friends to the party. The shy couple feels a bit more comfortable and starts dancing a little faster.
• Outcome: The light emission got about 100 times faster, but it's still not fast enough to be a practical lightbulb. It's still in the "microsecond" range (very slow for a light source).

3. Stretching the Material (The Magic Strain)

Finally, they tried stretching the crystal along a specific axis (uniaxial strain).

• The Result: This was the game-changer. Stretching the material caused a "band crossover."
• The Analogy: Imagine the shy couple was stuck in a maze. Stretching the material is like removing the walls of the maze. Suddenly, the couple can run straight to the center of the room and dance wildly.
• Outcome: The light emission speed increased by 100,000 times (five orders of magnitude). The material went from taking a day to glow, to glowing in nanoseconds.
• Comparison: This stretched Germanium now glows almost as brightly and quickly as Gallium Nitride (GaN), which is the gold standard for blue LEDs.

The Verdict: What Does This Mean?

The paper solves the mystery with a two-part conclusion:

1. The "Ghost" is Real: The pure, perfect Hexagonal Germanium crystal is indeed a terrible light emitter. If you see a bright light coming from it in an experiment, it's not the crystal doing the work. It's likely caused by defects, weird shapes, or local stress in the material (extrinsic factors). The crystal itself is just a "dark" background.
2. The Solution is Stretching: If we want to use Hexagonal Germanium for real technology (like lasers or LEDs), we can't just grow it; we have to stretch it. By applying the right amount of tension (strain), we can turn this "shy ghost" into a "bright star."

Summary in One Sentence

This paper proves that pure Hexagonal Germanium is naturally too shy to glow, but if you stretch it like a rubber band, it becomes a super-efficient light source, explaining why previous experiments saw bright light (likely from accidental stretching or defects) and showing us exactly how to make it work for future technology.

Technical Summary

1. Problem Statement

Recent experimental reports (e.g., Fadaly et al., Nature 2020) observed strong room-temperature photoluminescence (PL) in hexagonal diamond (2H) germanium (Ge) and 2H-Si$_x$Ge$_{1-x}$ alloys. This finding was surprising because group-IV materials are traditionally indirect-bandgap semiconductors with weak optical emission.

However, theoretical predictions based on independent-particle (IP) approximations suggested that the fundamental band-edge transition in 2H-Ge is "pseudo-direct" with extremely weak oscillator strength, implying the material should be optically "dark." A critical gap existed in the literature: previous theoretical works neglected excitonic effects (electron-hole interactions), assuming that the exciton binding energy ($E_b$) in 2H-Ge would be small (similar to cubic Ge, $\sim$4 meV) and thus negligible at room temperature. Without a comprehensive excitonic description, the discrepancy between the observed strong PL and the predicted weak intrinsic emission remained unresolved.

2. Methodology

The authors performed a comprehensive ab initio investigation to resolve this discrepancy, moving beyond the independent-particle approximation to include many-body effects.

• Electronic Structure:
  • Calculations were based on Density Functional Theory (DFT) using the Quantum ESPRESSO code.
  • To correct the bandgap underestimation typical of standard DFT (which predicts a negative gap for 2H-Ge) without the prohibitive cost of hybrid functionals, they employed a DFT+J scheme. A $J$-parameter correction was applied to the Ge 4p orbitals (and Si 3p for alloys) to reproduce HSE06 hybrid functional bandgaps and dispersions.
  • Spin-orbit coupling (SOC) was included in all calculations.
• Excitonic Properties:
  • Optical properties were calculated by solving the Bethe-Salpeter Equation (BSE) using the YAMBO code.
  • The study covered:
    • Pristine 2H-Ge.
    • 2H-Ge under 2% uniaxial strain along the $c$-axis (a strategy known to induce band crossover).
    • 2H-Si$_x$Ge$_{1-x}$ alloys with compositions $x = 1/6, 1/4, 1/2$, modeled using Special Quasirandom Structures (SQS) in a $24$-atom supercell.
    • Wurtzite GaN was used as a benchmark for an efficient wide-bandgap emitter.
• Key Metrics Calculated:
  • Exciton binding energies ($E_b$).
  • Exciton dipole moments ($\mu$) and oscillator strengths ($f$) for both in-plane ($\perp c$) and out-of-plane ($\parallel c$) polarizations.
  • Radiative lifetimes ($\tau$): Computed using a thermal average over excitonic states, accounting for the Maxwell-Boltzmann distribution of exciton center-of-mass momenta (CMM) and the light-cone constraint.

3. Key Contributions

• First Comprehensive Excitonic Analysis: This is the first study to explicitly calculate exciton binding energies, dipole moments, and radiative lifetimes for 2H-Ge and its alloys, proving that excitonic effects are non-negligible even at room temperature.
• Resolution of the PL Paradox: The study provides a microscopic explanation for why pristine 2H-Ge is intrinsically a poor light emitter, despite experimental reports of strong PL.
• Strain Engineering Validation: It quantifies the dramatic impact of uniaxial strain on optical properties, identifying it as a viable route to enhance intrinsic emission.
• Alloying Effects: It characterizes how Si alloying modifies symmetry and optical selection rules, offering a comparative analysis of composition-dependent radiative efficiency.

4. Key Results

A. Exciton Binding Energies

• Contrary to the assumption that 2H-Ge behaves like cubic Ge, the study finds a sizable exciton binding energy of $\sim$30 meV in pristine 2H-Ge.
• This large $E_b$ (due to reduced electron-hole mass along the $c$-axis and lower dielectric constants) confirms that excitonic effects dominate the optical response up to room temperature, necessitating the BSE approach.

B. Pristine 2H-Ge: Intrinsically "Dark"

• Dipole Moments: The lowest-energy exciton (four-fold degenerate) has extremely small dipole moments ($10^{-10}$ to $10^{-7} a_0^2$).
• Radiative Lifetime: The intrinsic radiative lifetime is exceptionally long, exceeding $10^{-4}$ seconds (hundreds of microseconds) at low temperatures.
• Conclusion: The strong PL observed experimentally in pristine samples cannot originate from the ideal crystal. It must be driven by extrinsic mechanisms such as defects, morphology, or local strain fields.

C. Effect of Si Alloying

• Symmetry Breaking: Random alloying breaks the crystal symmetry, lifting the suppression of the out-of-plane ($\parallel c$) dipole component.
• Lifetime Reduction: Alloying reduces the radiative lifetime by two orders of magnitude (down to the microsecond range, $\sim 10^{-6}$ s) compared to pristine Ge.
• Limitation: While improved, the efficiency remains far below that of efficient emitters like GaN.

D. Effect of Uniaxial Strain (The Breakthrough)

• Band Crossover: Applying 2% uniaxial strain along the $c$-axis induces a crossover between the conduction band states ($\Gamma^-_7$ and $\Gamma^-_8$).
• Enhanced Dipole: This crossover activates strong in-plane optical transitions. The dipole moment increases by five orders of magnitude, reaching values comparable to GaN ($\sim 10^{-1} a_0^2$).
• Radiative Lifetime: The lifetime drops dramatically to the nanosecond scale ($\sim 10^{-9}$ s) at low temperatures.
• Comparison to GaN: While strained 2H-Ge approaches the radiative efficiency of GaN, it remains about one order of magnitude slower due to GaN's significantly higher exciton energy (3.3 eV vs. 0.3 eV), which amplifies the oscillator strength.

E. Temperature Dependence

• All systems exhibit a characteristic $T^{3/2}$ scaling of the radiative lifetime. This arises because the thermal average is dominated by the four lowest-energy degenerate excitons, and the lifetime is governed by the thermal population of the light cone.

5. Significance and Conclusion

This work fundamentally clarifies the optical nature of hexagonal diamond Ge:

1. Intrinsic Limit: It establishes that pristine 2H-Ge is an intrinsically inefficient light emitter with a "dark" band edge, explaining the theoretical/experimental mismatch. The reported strong PL in unstrained samples is likely extrinsic.
2. Strain Engineering: It identifies uniaxial strain as a highly effective, viable strategy to transform 2H-Ge into a bright, nanosecond-scale emitter, potentially enabling group-IV based optoelectronics (lasers, LEDs) without relying on III-V materials.
3. Methodological Benchmark: By providing a rigorous excitonic description, the paper sets a new standard for analyzing polytypic group-IV materials, moving beyond independent-particle approximations that fail to capture the true radiative dynamics.

The authors conclude that while the ideal crystal is dark, the material holds immense potential for integrated photonics if strain engineering can be successfully implemented in nanostructures (e.g., nanowires) where such deformations are physically sustainable.