THE DETERMINATION OF TERBIURI(II1) IN BORATE' GLASSES"
The development of lasers has greatly increased interest both in the luminescence of rare earths in glasses, and in the analytical determination of their composition. Moreover, the increasing use of inorganic glasses as matrices for trapped radicals and ions of unusual valenciesr requires the development of non-destructive analytical methods for the determination of impurities. While excitation and emission of rare eartlls is similar in glasses and crystalsz, much higher concentrations of rare earths can be introcluced into glasses. The catlmclc-ray-escitcd emission method for the determination of terbiuma was ineffective in the detection of terbium in glasses, probably because of dissipation of the energy of the esciting electrons by the traps in glass". Tlicreforc, a metliocl based on optical escitation was developed.
Initially, the optical conditions for the analysis were established. This study included measurements of excitation spectra, emission spectra and decay times of fluorescence. The concentration dependence of fluoresccncc was then measured. This teclmiclue of measurement can be adapted as a general procedure for the estimation of rare eartlls in various glasses.
The spectrofluorimeter used for tile measurement of excitation and fluorescence emission spectra and decay times w-as constructed in this laboratory". Briefly, it consists of a 500-w xenon arc (Mazda) lamp, interchangeable with a mercury mediumpressure 100-w lamp, an escitation monochromator (Bausch C% Lomb, 250 mm focal length), a sample compartment, an analysing monochromator (Bausch R: Lomb, 500 mm focal length), an EM1 9558 Q/13 photomultiplier connected through a 16 Mti resistor to an EIL Vibron electrometer, and a Kipp & Zenon recorder. Slow drifts in light intensity were corrected by frequent measurements of a reference glass which contained I o/o (w/w) of terbium(II1).
Fast fluctuations were filtered out. The fluorescence was viewed from the front (illuminated) side of the glass at an angle of 35", which was found to be the optimum position for signal-to-noise ratio. The photomultiplier was cooled to 20~ for all measurements, thus decreasing the dark current by about two orders of magnitude.
The decay times were measured by the technique described earlierG. The absorbance of the glasses in the visible and UV regions was measured with a Perkin C% Elmer 137 UV spectroplrotometer ve7s7cs pure borax. * This work was pcrformccl under NBS contract (G) 83. Amd. Clrirn. Ada, 50 (1970) 24g-25dl Af#I+ Ofi/., 2 (1963) 675. .4md. Chira. Acta, so (~970) zdfg---254