Characterization of Amorphous and Microcrystalline Silicon using Raman Spectroscopy

Significance of the Topic

Silicon thin films combining amorphous and microcrystalline phases are central to high-efficiency, cost-effective photovoltaic cells. The ratio and spatial distribution of these phases directly impact cell performance, durability and manufacturing yield. Raman spectroscopy provides a rapid, non-destructive approach to quantify and map these silicon forms, enabling real-time quality control throughout production.

Objectives and Study Overview

This study aims to demonstrate how Raman spectroscopy can be used to:

• Quantify the relative proportions of amorphous and crystalline silicon in deposited films.
• Generate spatial maps that reveal phase distribution and uniformity across sample surfaces.
• Identify practical considerations, such as laser power and excitation wavelength, to ensure reliable and reproducible measurements.

Methodology and Instrumentation

Raman data were collected using a Thermo Scientific DXR Raman microscope equipped with:

• A 532 nm excitation laser and full-range grating
• Motorized XY stage for precise mapping
• OMNIC 8 software for spectral acquisition
• OMNIC™ Atlµs™ mapping suite for chemical imaging and data analysis
• An integrated laser power regulator to maintain consistent power at the sample and avoid phase conversion

Quantification employed a Beer’s Law approach, comparing the intensity of the crystalline Si peak at 521 cm⁻¹ with the amorphous Si band centered at 480 cm⁻¹. Line scans were performed at 2 µm steps over 30 µm, while two-dimensional maps used 25 µm intervals over areas up to 750 × 2 250 µm.

Main Results and Discussion

Distinct Raman signatures enable clear discrimination:

• Crystalline silicon: sharp peak at 521 cm⁻¹
• Amorphous silicon: broad band centered at 480 cm⁻¹

Mapping revealed localized microcrystalline regions within primarily amorphous films. Line-scan data identified a crystalline zone near the film midpoint, while 2D chemical images color-coded crystalline (red) and amorphous (blue) domains, with intermediate hues indicating mixed areas.

Laser power studies showed that excitation above ~4 mW can induce amorphous-to-crystalline transformation. The DXR power regulator ensured stable output and prevented unintended phase changes. Lower powers maintained film integrity.

Excitation wavelength considerations:

• Raman scattering efficiency scales as 1/λ⁴, favoring shorter wavelengths for stronger signals.
• 532 nm light penetrates ~0.10 µm in silicon, minimizing substrate contributions in films ≥100 nm thick.
• Fluorescence interference is reduced at 532 nm compared to 780 nm, improving signal-to-noise.

Benefits and Practical Applications

Raman spectroscopy offers:

• Non-destructive, label-free analysis of silicon phase composition
• High spatial resolution mapping for process control and uniformity assessment
• Rapid turnaround suitable for inline quality checks in photovoltaic manufacturing
• Reproducibility across instruments via regulated laser power

Future Trends and Potential Uses

Emerging developments may include:

• Integration of Raman mapping into automated inline production monitoring systems
• Advanced spectral unmixing with machine learning for more precise phase quantification
• Extension to other thin‐film semiconductor materials and multi‐layer PV architectures
• Higher spatial resolution through near-field or tip-enhanced Raman techniques
• Real-time monitoring of film growth and in-situ process feedback

Conclusion

Raman spectroscopy, when coupled with controlled laser power and appropriate excitation wavelength, delivers a robust, quantitative method for characterizing amorphous and microcrystalline silicon in photovoltaic films. Its mapping capabilities provide critical insights into film uniformity, supporting quality assurance and process optimization in solar cell manufacturing.

References

• Thermo Fisher Scientific, Application Note 51735: Characterization of Amorphous and Microcrystalline Silicon using Raman Spectroscopy, 2009.