Despite the observation by a number of groups of a strong luminescence sensitization effect of erbium ions by excitation exchange from silicon nanoclusters, there is considerable experimental evidence that the fraction of Er ions excited by Si-nc is actually very low for much of the material reported. In this work, we examine the evidence and point out that Er excited state absorption is the likely cause. © 2006 American Institute of Physics. ͓DOI: 10.1063/1.2227637͔In recent years, silicon nanocluster-sensitized Er-doped materials have generated great interest, as it has been demonstrated that silicon nanoclusters ͑Si-nc͒ can transfer their energy very effectively to the Er ions surrounding them. The attraction of this indirect excitation is twofold: ͑i͒ the excitation cross section of the Si-nc is at least four orders of magnitude greater than that of rare earths in silica and ͑ii͒ the broadband absorption spectrum of the Si-nc allows pumping with sources such as light-emitting diodes ͑LEDs͒. The latter would dramatically lower the cost of Er-doped amplifiers, and allow pumping geometries such as top-down nonguided pumping, which would enable easy integration of amplifiers in planar optical circuits. However, no lasing action has been achieved to date. Indeed, there is cumulating evidence that, for much of the material that has been reported, the fraction of Er ions that can be excited by Si-nc is actually very low-of the order of a few percent or less. Clearly, this issue will need to be understood and resolved in order for viable lasers and amplifiers to be realized from this material system.There is now documented evidence from multiple groups, although it may not always have been explicitly recognized, that the fraction of Er ions that are excitable by Si-ncs is often quite low. 1,2 For example, for the data presented by Fujii et al., 1 the Er photoluminescence ͑PL͒ for two samples with the same Er concentration, one of which contained Si-nc ͓silicon-rich oxide ͑SRO͔͒ and the other without ͑SiO 2 ͒, was reported to differ by a factor of 30 under the same pumping conditions ͑see Fig. 1͒. It is straightforward to show by rate equation analysis that the Er excited state populations are given bywhere is the Er pump absorption cross section, the Er luminescence lifetime, P the pump flux, N Er the Er ion concentration, k the Si-nc to Er excitation transfer coefficient, and n b is the Si-nc excited state population. At the maximum pump power reported ͑1.5 W / cm 2 ͒, the SRO sample is in the saturation regime while the SiO 2 sample is still in the linear regime; therefore the ratio of the number of excited Er ions in the two samples can be simply estimated asTaking 488 nm ϳ 10 −20 cm 2 , = 10 ms, and P =4 ϫ 10 18 cm 2 s −1 ͑corresponding to 1.5 W / cm 2 ͒, this yields an Er concentration ratio of 2000. As the experimentally observed PL ratio is only 30, this indicates that just 1.5% of the available Er ions in the SRO sample have been excited. This conclusion seems surprising considering that, for the volume frac...