Measurement of the absolute Raman cross section of the optical phonon in silicon

R. L. Aggarwal, L. W. Farrar, Semion K. Saikin, Alán Aspuru‐Guzik, M. Stopa, D.l. PollaSolid 2011 · DOI

TL;DR

This paper precisely measures the absolute Raman scattering cross-section of silicon's optical phonon across different laser excitation wavelengths. By employing a novel blackbody calibration method, they provide more accurate values than previously available, offering crucial data for understanding light-matter interactions in silicon.

What Problem Does This Paper Address?

Raman spectroscopy is a powerful tool for probing the vibrational properties of materials, with silicon being a cornerstone of modern electronics. The "Raman cross-section" quantifies how likely a material is to scatter light via Raman processes. Accurately knowing this value for silicon's fundamental optical phonon is essential for:

  • Quantitative Raman analysis: Precisely determining material properties (like temperature, stress, or doping) from Raman signals.
  • Device modeling: Simulating and predicting the behavior of optoelectronic devices that utilize Raman scattering.
  • Fundamental understanding: Deepening our knowledge of light-phonon interactions in a ubiquitous semiconductor.

While previous studies have measured or calculated this cross-section, there have been discrepancies and a need for more precise, absolute measurements across a range of excitation wavelengths. This paper aims to fill that gap by providing highly accurate, absolute measurements of the Raman cross-section for silicon's zone-center optical phonon.

Methods

The core of this work lies in its meticulous experimental setup and calibration.

  • Sample: A high-quality, 0.63-mm thick [001] silicon wafer was used.
  • Optical Setup: A standard Raman spectroscopy configuration was employed, featuring a focusing lens to direct the pump laser onto the sample and collect the scattered light. A beamsplitter was used to align the pump beam.
  • Excitation Lasers: Three different continuous-wave (CW) lasers were used for excitation: 785 nm (1.58 eV), 1064 nm (1.17 eV), and 1535 nm (0.81 eV). The pump beam was polarized along the [100] axis of the silicon.
  • Raman Spectrometers: Specialized Raman spectrometers were used for each laser wavelength, equipped with either silicon CCD or InGaAs cameras depending on the excitation energy.
  • Novel Calibration: The key innovation is the use of a temperature-controlled blackbody as a calibrated source for the Raman spectrometers. This provides a more accurate and absolute calibration of the signal intensity compared to previous methods that relied on relative measurements or less precise calibration sources. The blackbody's aperture was matched to the focused pump beam size.
  • Data Analysis: The Raman cross-section ($\sigma_{RS}$) was calculated using a formula that relates the integrated Raman signal intensity to the pump power, atomic concentration, and optical properties of silicon. The Raman scattering efficiency ($S$) and Raman polarizability ($|d|$) were then derived from $\sigma_{RS}$.

Datasets

Not discussed in the paper. The paper focuses on experimental measurements using a silicon sample and a calibrated blackbody source.

Key Insights and Ideas

  • Blackbody Calibration is Key: The use of a temperature-controlled blackbody for absolute signal calibration is a significant methodological advancement, leading to more reliable and accurate Raman cross-section values.
  • Wavelength Dependence Matters: The Raman cross-section and polarizability show a clear dependence on the excitation laser wavelength. This is crucial for selecting optimal excitation wavelengths for specific applications and for understanding resonance effects.
  • Agreement with Theory (with caveats): The measured Raman polarizability values generally agree with theoretical predictions (e.g., from Wendel's model and DFT calculations), especially in the static limit. However, deviations at higher excitation energies suggest the influence of indirect band gap resonance effects.
  • Photoluminescence Challenge: At longer wavelengths (like 1064 nm), strong photoluminescence from the silicon sample can interfere with Raman measurements at room temperature. Cooling the sample to ~80 K was necessary to overcome this issue.

Key Results

The paper reports the following absolute Raman cross-section ($\sigma_{RS}$) values for the 519 cm⁻¹ optical phonon in silicon:

  • 785 nm (1.58 eV): $\sigma_{RS} = 1.0 \pm 0.2 \times 10^{-27} \text{ cm}^2$
  • 1064 nm (1.17 eV): $\sigma_{RS} = 3.6 \pm 0.7 \times 10^{-28} \text{ cm}^2$ (measured at ~80 K due to photoluminescence)
  • 1535 nm (0.81 eV): $\sigma_{RS} = 1.1 \pm 0.2 \times 10^{-29} \text{ cm}^2$

Corresponding Raman polarizability ($|d|$) values were also determined:

  • 785 nm: $|d| = 4.4 \pm 0.4 \times 10^{-15} \text{ cm}^2$
  • 1064 nm: $|d| = 5.1 \pm 0.5 \times 10^{-15} \text{ cm}^2$
  • 1535 nm: $|d| = 1.9 \pm 0.2 \times 10^{-15} \text{ cm}^2$

These measured values for $|d|$ at 785 nm and 1064 nm are approximately 1.3 and 2.0 times larger, respectively, than previously calculated values.

Limitations and Future Work

  • Photoluminescence at Longer Wavelengths: As noted, photoluminescence can be a significant issue for Raman measurements in silicon at longer excitation wavelengths, necessitating low-temperature measurements for some cases.
  • Comparison with Theory: While agreement with theoretical models is generally good, further theoretical work might be needed to fully explain the observed wavelength dependence, particularly concerning resonance effects.
  • Future Work: The authors suggest that their precise measurements provide a solid foundation for future studies, including investigations into the temperature dependence of the Raman cross-section and its application in more complex silicon-based nanostructures. The paper also highlights the potential for using their blackbody calibration method in other Raman spectroscopy applications.

Summary updated

BibTeX
@inproceedings{aggarwal2011measurement,
  author = {R. L. Aggarwal and L. W. Farrar and Semion K. Saikin and Alán Aspuru‐Guzik and M. Stopa and D.L. Polla},
  title = {{Measurement of the absolute Raman cross section of the optical phonon in silicon}},
  booktitle = {Solid},
  year = {2011},
  doi = {10.1016/j.ssc.2011.01.011},
}
Figure