Band gap reduction in highly-strained silicon beams predicted by first-principles theory and validated using photoluminescence spectroscopy

TL;DR

This study combines first-principles calculations and photoluminescence spectroscopy to accurately predict and validate the reduction of silicon's band gap under tensile stress, demonstrating significant agreement between theory and experiment.

Contribution

It provides the first combined theoretical and experimental analysis of band gap reduction in highly-strained silicon beams using DFT and spectroscopy.

Findings

Band gap reduces by ~100 meV at 2% strain in silicon.

Experimental measurements confirm theoretical predictions within 5%.

Deformation along [110] direction yields the most significant band gap reduction.

Abstract

A theoretical study of the band gap reduction under tensile stress is performed and validated through experimental measurements. First-principles calculations based on density functional theory (DFT) are performed for uniaxial stress applied in the [001], [110] and [111] directions. The calculated band gap reductions are equal to 126, 240 and 100 meV at 2$\%$ strain, respectively. Photoluminescence spectroscopy experiments are performed by deformation applied in the [110] direction. Microfabricated specimens have been deformed using an on-chip tensile technique up to ~1$\%$ as confirmed by back-scattering Raman spectroscopy. A fitting correction based on the band gap fluctuation model has been used to eliminate the specimen interference signal and retrieve reliable values. Very good agreement is observed between first-principles theory and experimental results with a band gap reduction of, respectively, 93 and 91 meV when the silicon beam is deformed by 0.95$\%$ along the [110] direction.