Effect of Annealing on Al Diffusion and its Impact on the Properties of Ga$_2$O$_3$ Thin Films Deposited on c-plane Sapphire by RF Sputtering

This study demonstrates that thermal annealing of RF-sputtered Ga$_2$O$_3$ thin films on c-plane sapphire induces aluminum diffusion to form a tunable $\beta$-(Al$_x$Ga$_{1-x}$)$_2$O$_3$ alloy, enabling the bandgap to be adjusted between 4.85 and 5.30 eV by varying the annealing temperature.

The Gist

Imagine you have a very special, transparent glass window made of Gallium Oxide ($Ga_2O_3$). This isn't just any window; it's a high-tech material used to make super-fast electronics and sensors that can "see" invisible ultraviolet light (like the kind that causes sunburns).

Now, imagine you place this glass window on top of a very hard, blue-tinted ceramic tile made of Sapphire (which is actually Aluminum Oxide, $Al_2O_3$).

The scientists in this study asked a simple question: What happens if we bake this sandwich in an oven?

Here is the story of their experiment, explained simply:

1. The Setup: A Cold Sandwich

First, they sprayed a thin layer of the Gallium Oxide onto the Sapphire tile. At this stage, the layer was a bit messy and "amorphous" (like a pile of sand rather than a structured brick wall). It wasn't very strong or clear yet.

2. The Oven: The Magic of Heat

They took this sandwich and put it in a furnace, heating it up to temperatures ranging from a warm 550°C to a scorching 1300°C (that's hotter than a pizza oven!).

What happened inside the oven?
Think of the atoms in the Gallium layer and the atoms in the Sapphire tile as two groups of people at a party.

• Before heating: They are standing in their own separate rooms.
• After heating: The heat gives them energy to dance around. The Aluminum atoms from the Sapphire tile start sneaking into the Gallium room, and the Gallium atoms start sneaking into the Sapphire room.

This is called interdiffusion. It's like mixing blue and yellow paint; eventually, you get green. In this case, mixing Gallium and Aluminum creates a new, tunable material called $\beta-(Al_xGa_{1-x})_2O_3$.

3. The Results: What Changed?

The scientists used several "super-microscopes" to see what happened. Here are the three big discoveries:

A. The Surface Got Rougher (The "Popcorn" Effect)

When the heat was turned on, the tiny grains of the material started to grow and merge, like popcorn kernels popping and sticking together.

• The Result: The surface became much rougher. It went from being as smooth as a calm lake to being as bumpy as a cobblestone street. While this sounds bad, it actually means the material is becoming more solid and crystalline (like ice forming from water).

B. The Structure Got Stronger (The "Brick Wall" Effect)

Before baking, the material was a messy pile of sand. After baking, the atoms lined up perfectly to form a strong, organized brick wall.

• The Result: The material became much clearer and more ordered. However, if they baked it too hot (1300°C), the perfect alignment started to get a little jumbled again, like a brick wall that got shaken too hard.

C. The "Magic" Color Shift (The Sunglasses Effect)

This is the most important part. The scientists wanted to know: Can we change what this material does just by changing the temperature?

• The Answer: Yes! By mixing in more Aluminum (by baking it hotter), they changed the material's "bandgap."
• The Analogy: Imagine the material is a pair of sunglasses.
  • Low heat: The sunglasses let in a certain amount of light (like regular sunglasses).
  • High heat: Because more Aluminum sneaked in, the sunglasses became "darker" for certain types of light. They blocked more energy, shifting the material's ability to handle electricity and light.
  • The Shift: They managed to tune the material's energy barrier from 4.85 eV to 5.30 eV. This is a huge deal because it means they can customize the material to be perfect for specific jobs, like detecting deep ultraviolet light for security cameras or solar cells.

Why Does This Matter?

Usually, making high-tech materials requires expensive, complex machines. This study shows that you can take a simple, cheap method (spraying the material) and use a simple oven to upgrade it.

By just controlling the temperature, they can:

1. Make the material stronger and more organized.
2. Mix in Aluminum to change its properties without needing to buy new raw materials.
3. Create custom "super-glass" for future electronics that can survive in harsh environments (like space or nuclear reactors).

In a nutshell: They took a simple layer of material, baked it on a sapphire tile, and watched the atoms dance around to create a stronger, smarter, and more customizable material just by turning up the heat.

Technical Summary

1. Problem Statement

Gallium oxide ($\text{Ga}_2\text{O}_3$) is a promising ultra-wide bandgap semiconductor for high-power electronics and deep-ultraviolet (DUV) optoelectronics. While Radio-Frequency (RF) sputtering is a cost-effective method for depositing high-quality thin films, films deposited at room temperature are typically amorphous and require post-deposition thermal annealing to crystallize.

A critical, yet often overlooked, issue arises when $\text{Ga}_2\text{O}_3$ films are deposited on sapphire ($\text{Al}_2\text{O}_3$) substrates and subjected to high-temperature annealing. There is a potential for interdiffusion between the film and the substrate: Aluminum (Al) from the sapphire may diffuse into the $\text{Ga}_2\text{O}_3$ film, while Gallium (Ga) may diffuse into the substrate. This process forms a $\beta\text{-(Al}_x\text{Ga}_{1-x})_2\text{O}_3$ alloy, which fundamentally alters the film's stoichiometry, structural quality, and optical properties (specifically the bandgap). Understanding the extent of this diffusion and its impact on material properties is essential for device engineering, particularly for tuning the bandgap for DUV applications.

2. Methodology

The researchers employed a comprehensive multi-technique characterization approach to analyze $\text{Ga}_2\text{O}_3$ films before and after thermal treatment.

• Sample Preparation:
  • Deposition: $\text{Ga}_2\text{O}_3$ thin films (~118 nm thick) were deposited on c-plane sapphire substrates using RF magnetron sputtering at room temperature (60 W, 6 mTorr, Ar flow).
  • Annealing: The samples were cut into pieces and annealed in ambient air at temperatures ranging from 550 °C to 1300 °C (in 150 °C steps) for 1 hour each.
• Characterization Techniques:
  • Rutherford Backscattering Spectrometry (RBS): Used to determine elemental composition depth profiles and quantify the interdiffusion of Al and Ga.
  • X-ray Diffraction (XRD): Used to analyze crystalline structure, phase identification, texture, grain size (via Scherrer equation), and microstrain (via Wilson equation).
  • Atomic Force Microscopy (AFM): Used to measure surface morphology and root-mean-square (RMS) roughness.
  • Raman Spectroscopy: Used to confirm phase purity and detect shifts in vibrational modes indicative of Al incorporation.
  • Optical Transmission (OT): Used to calculate the optical bandgap ($E_g$) using Tauc's method.
  • Composition Calculation: Al content ($x$) was estimated using Vegard's law based on the measured bandgap and lattice parameters.

3. Key Contributions

• Quantification of Interdiffusion: The study provides direct experimental evidence (via RBS) of bidirectional interdiffusion between $\text{Ga}_2\text{O}_3$ films and sapphire substrates, identifying the onset temperature of diffusion.
• Correlation of Structure and Composition: The paper establishes a clear correlation between annealing temperature, Al concentration, crystalline quality, and surface roughness.
• Bandgap Tuning Mechanism: It demonstrates that the increase in bandgap is not solely due to crystallization but is primarily driven by the formation of a $\beta\text{-(Al}_x\text{Ga}_{1-x})_2\text{O}_3$ alloy via Al diffusion.
• Multi-Technique Validation: The results from RBS, XRD, and Optical Transmission were cross-validated, showing consistent trends in Al concentration despite minor discrepancies attributed to depth gradients.

4. Key Results

A. Compositional Changes (Interdiffusion)

• Onset Temperature: Significant interdiffusion begins at approximately 700–850 °C.
• Al Content: As annealing temperature increases, Al content in the film rises significantly. At 1300 °C, the Al content reaches 68.5% (determined by RBS), forming a high-Al alloy.
• Ga Diffusion: Simultaneously, Ga diffuses into the sapphire substrate.
• Depth Profile: RBS analysis reveals a compositional gradient, indicating that the Al concentration is not uniform throughout the film depth.

B. Structural and Morphological Evolution

• Crystallization: Films remain amorphous below ~700 °C. Above this threshold, they crystallize into the monoclinic $\beta$-phase.
• Texture: Films initially show a strong $\bar{2}01$ preferential orientation. However, at 1300 °C, this texture is lost, and peaks associated with the $\{100\}$ family of planes become prominent.
• Grain Size & Strain: Crystal domain size increases with temperature, while microstrain decreases, indicating improved crystalline quality.
• Surface Roughness: RMS roughness increases dramatically from 0.5 nm (as-deposited) to 8.3 nm (at 1300 °C). This is attributed to grain growth/coalescence and the incorporation of Al.

C. Optical Properties

• Bandgap Tuning: The optical bandgap increases from ~4.85 eV (pure $\text{Ga}_2\text{O}_3$) to ~5.30 eV as the annealing temperature rises to 1150 °C.
• Mechanism: This widening of the bandgap is attributed to the alloying effect (increasing Al content) rather than quantum confinement (which would typically narrow the gap in larger grains).
• Raman Shift: Raman spectra show a shift of the $A_g^{(3)}$ peak to higher wavenumbers with increasing temperature, confirming Al incorporation.

5. Significance and Implications

• Device Engineering: The study highlights that using sapphire substrates for $\text{Ga}_2\text{O}_3$ devices requires careful thermal budget management. Unintended Al diffusion can alter the designed bandgap, which is critical for DUV photodetectors and solar-blind applications.
• Material Design: Conversely, this diffusion mechanism can be intentionally exploited to create tunable $\beta\text{-(Al}_x\text{Ga}_{1-x})_2\text{O}_3$ alloys without complex co-deposition processes. By controlling annealing temperature, researchers can tune the bandgap between 4.85 eV and 5.30 eV.
• Process Optimization: The findings suggest that while high-temperature annealing improves crystallinity, it degrades surface smoothness and alters stoichiometry. Future processes must balance crystalline quality with the preservation of the intended film composition.

In conclusion, the paper demonstrates that thermal annealing of RF-sputtered $\text{Ga}_2\text{O}_3$ on sapphire is a double-edged sword: it is necessary for crystallization but inevitably leads to significant Al diffusion, fundamentally transforming the material into a tunable alloy with modified optical and morphological properties.