Effect of hydrogen passivation on luminescence-center-mediated Er excitation in Si-rich ${\text{SiO}}_2$ with and without Si nanocrystals

The influence of hydrogen passivation on luminescence-center-mediated excitation of ${\text{Er}}^{3+}$ in Er-doped Si-rich ${\text{SiO}}_2$ films with significantly different microstructures is studied. Photoluminescence measurements are presented for samples containing no detectable silicon nanocrystals (annealed at $600°\text{C}$) and for samples containing silicon nanocrystals (annealed at $1100°\text{C}$) as a function of hydrogen passivation temperature. Passivation is found to have little effect on the ${\text{Er}}^{3+}$ photoluminescence intensity at 1535 nm in the samples that do not contain nanocrystals. In contrast, a pronounced increase in the ${\text{Er}}^{3+}$ photoluminescence intensity is observed in the samples containing Si nanocrystals, which is accompanied by a similar increase in the nanocrystal photoluminescence intensity and a gradual increase in the Si nanocrystal emission lifetime. This observation is attributed to two interrelated effects, namely, (a) an increase in the density of fully passivated optically active nanocrystals due to the passivation-induced removal of silicon dangling bonds and (b) a concurrent reduction in nonradiative ${\text{Er}}^{3+}$ relaxation from levels above the ${{}^{4}I}_{13/2}$ level due to a direct interaction of excited ${\text{Er}}^{3+}$ ions with silicon dangling bonds. In addition, the observed counterintuitive gradual increase in the nanocrystal photoluminescence decay time upon passivation is successfully explained taking into account a passivation-induced change in the concentration of optically active nanocrystals with different sizes and the inhomogeneous nature of the nanocrystal-related emission band. It is shown that the combination of luminescence-center-mediated ${\text{Er}}^{3+}$ excitation and silicon-dangling-bond-induced ${\text{Er}}^{3+}$ de-excitation can explain at least 14 experimental observations reported by independent authors.

Figure 1

(Color online) Photoluminescence spectra in the visible–near-infrared region of the LTA samples passivated in the temperature range $300–600°\text{C}$. The PL spectrum of an unpassivated sample is included as a reference. Inset: PL intensity of the luminescence center emission of the LTA samples at 750 nm as a function of passivation temperature.

Figure 2

(Color online) Photoluminescence spectra in the near-infrared region of LTA samples passivated in the temperature range of $300–600°\text{C}$. The PL spectrum of an unpassivated sample is included as a reference. Inset: PL intensity of the ${\text{Er}}^{3+}$-related emission of the LTA samples at 1535 nm as a function of passivation temperature.

Figure 3

(Color online) Photoluminescence spectra in the visible–near-infrared region of the HTA samples passivated in the temperature range of $300–800°\text{C}$. The PL spectrum of an unpassivated sample is included as a reference. Inset: PL intensity of nanocrystal-related emission of the HTA samples at 900 nm as a function of passivation temperature.

Figure 4

(Color online) Photoluminescence spectra in the near-infrared region of the HTA samples passivated in the temperature range of $300–800°\text{C}$. The PL spectrum of an unpassivated sample is included as a reference. Inset: PL intensity of ${\text{Er}}^{3+}$-related emission of the HTA samples at 1535 nm as a function of passivation temperature.

Figure 5

Measured and deduced photoluminescence parameters of the LTA (solid symbols) and HTA samples (open symbols) as a function of passivation temperature: (a) ${\text{Er}}^{3+}$ PL decay times at 1535 nm of the LTA and HTA samples (${\tau }_{\text{dec},\text{Er}}^{\text{LTA}}$ and ${\tau }_{\text{dec},\text{Er}}^{\text{HTA}}$, respectively) and silicon nanocrystal PL decay times at 800 and 900 nm in the HTA samples $\left({\tau }_{\text{dec},\text{NC}}^{\text{HTA}}\right)$; (b) the effective absorption cross sections of the ${\text{Er}}^{3+}$ PL at 1535 nm in the LTA and HTA samples (${\sigma }_{\text{abs},\text{Er}}^{\text{LTA}}$ and ${\sigma }_{\text{abs},\text{Er}}^{\text{HTA}}$ respectively), and of the silicon nanocrystal PL at 900 nm in the HTA samples $\left({\sigma }_{\text{abs},\text{NC}}^{\text{HTA}}\right)$; (c) luminescence center PL intensity at 750 nm in the LTA samples $\left(I_{\text{PL},\text{LC}}^{\text{LTA}}\right)$, the relative density of sensitized optically active ${\text{Er}}^{3+}$ ions in the LTA and HTA samples ($N_{\text{Er}}^{\text{LTA}}$ and $N_{\text{Er}}^{\text{HTA}}$, respectively), and the relative density of optically active silicon nanocrystals contributing to the emission at 900 nm $\left(N_{\text{NC}}^{\text{HTA}}\right)$. The arbitrary units of the relative density of sensitized optically active ${\text{Er}}^{3+}$ ions and optically active silicon nanocrystals are not related.

Figure 6

(Color online) Proposed model of the effect of hydrogen passivation on the Si nanocrystal and ${\text{Er}}^{3+}$ emission at three different passivation temperatures: (a) 300, (b) 500, and (c) $600°\text{C}$. The large dark and bright solid circles represent Si nanocrystals with low and high emission efficiencies, respectively. The crosses represent dangling bonds, the dashed circles schematically indicate the interaction range of a dangling bond, and the letter H represents a hydrogen terminated dangling bond. The small solid and open circles represent ${\text{Er}}^{3+}$ ions that are affected or unaffected by a dangling bond, respectively.

Figure 7

(Color online) Measured and deduced photoluminescence parameters of the silicon-nanocrystal-related emission in the HTA samples as a function of passivation temperature: the ratio of the silicon nanocrystal PL intensity at 900 nm to that at 850 nm $\left(I_{\text{PL},900\text{nm}}^{\text{HTA}}/I_{\text{PL},850\text{nm}}^{\text{HTA}}\right)$, silicon nanocrystal PL decay times of the HTA samples at 900 nm $\left({\tau }_{\text{dec},\text{NC}}^{\text{HTA}}\right)$, silicon nanocrystal PL decay time dispersion factor of the HTA samples at 900 nm $\left({\beta }_{\text{dec},\text{NC}}^{\text{HTA}}\right)$. Insets: Measured silicon nanocrystal emission spectra (thick solid line) for an unpassivated HTA sample (left) and for a HTA sample passivated at $700°\text{C}$ (right), normalized at a wavelength of 900 nm. The Gaussian curves underneath the PL spectrum schematically depict the individual PL contributions of silicon nanocrystals with different sizes.