The optical properties of gadolinium gallium aluminum garnet, Gd3(Ga,Al)5O12, doped with Ce(3+) are investigated as a function of the Ga/Al ratio, aimed at an improved understanding of the energy flow and luminescence quenching in these materials. A decrease of both the crystal field strength and band gap with increasing content of Ga(3+) is observed and explained by the geometrical influence of Ga(3+) on the crystal field splitting of the 5d level in line with theoretical work of Muñoz-García et al. ( uñoz-García, A. B.; Seijo, L. Phys. Rev. B 2010, 82, 184118 ). Thermal quenching results in shorter decay times as well as reduced emission intensities for all samples in the temperature range from 100 to 500 K. An activation energy for emission quenching is calculated from the data. The band gap of the host is measured upon Ga substitution and the decrease in band gap is related to Ga(3+) substitution into tetrahedral sites after all octahedral sites are occupied in the garnet material. Based on the change in band gap and crystal field splitting, band diagrams can be constructed explaining the low thermal quenching temperatures in the samples with high Ga content. The highest luminescence intensity is found for Gd3(Ga,Al)5O12 with 40% of Al(3+) replaced by Ga(3+).

The d-f emission from Ce 3+ and Pr 3+ in garnets is attracting considerable attention, especially in relation to application in white light LEDs and scintillators. An important aspect is the luminescence quenching temperature T Q . It is not trivial to determine T Q and to unravel the quenching mechanism. In this paper the T Q of d-f emission for Ce 3+ and Pr 3+ are determined by temperature dependent lifetime measurements. The results show a T Q for Pr 3+ of 340 K for Y 3 Al 5 O 12 :Pr 3+ (YAG:Pr) and 680 K for Lu 3 Al 5 O 12 :Pr 3+ (LuAG:Pr). For Ce 3+ the T Q is too high to measure. An onset of quenching above 600 K (YAG:Ce) or 700 K (LuAG:Ce) is observed. The differences in T Q between YAG and LuAG are explained by a smaller Stokes shift for the d-f emission in LuAG (∼2300 cm −1 ) compared to YAG (∼2750 cm −1 ) derived from low temperature luminescence spectra. The large difference in T Q between Ce 3+ and Pr 3+ is related to the smaller energy difference between the lowest energetic fd state of Pr 3+ and the next lower 4f 2 state ( 3 P 2 ) compared to the 5d -4f 1 ( 2 F 7/2 ) energy difference for Ce 3+ . Both observations are consistent with luminescence temperature quenching by non-radiative relaxation from the 5d state to the 4f state described by a configurational coordinate diagram and not by thermally induced photoionization.

Lu(3)Al(5)O(12) (LuAG) doped with Ce(3+) is a promising scintillator material with a high density and a fast response time. The light output under X-ray or γ-ray excitation is, however, well below the theoretical limit. In this paper the influence of codoping with Tb(3+) is investigated with the aim to increase the light output. High resolution spectra of singly doped LuAG (with Ce(3+) or Tb(3+)) are reported and provide insight into the energy level structure of the two ions in LuAG. For Ce(3+) zero-phonon lines and vibronic structure are observed for the two lowest energy 5d bands and the Stokes' shift (2 350 cm(-1)) and Huang-Rhys coupling parameter (S = 9) have been determined. Tb(3+) 4f-5d transitions to the high spin (HS) and low spin (LS) states are observed (including a zero-phonon line and vibrational structure for the high spin state). The HS-LS splitting of 5400 cm(-1) is smaller than usually observed and is explained by a reduction of the 5d-4f exchange coupling parameter J by covalency. Upon replacing the smaller Lu(3+) ion with the larger Tb(3+) ion, the crystal field splitting for the lowest 5d states increases, causing the lowest 5d state to shift below the (5)D(4) state of Tb(3+) and allowing for efficient energy transfer from Tb(3+) to Ce(3+) down to the lowest temperatures. Luminescence decay measurements confirm efficient energy transfer from Tb(3+) to Ce(3+) and provide a qualitative understanding of the energy transfer process. Co-doping with Tb(3+) does not result in the desired increase in light output, and an explanation based on electron trapping in defects is discussed.

No abstract