1. Introduction

When a glass laser is operated, thermal effects produce optical distortions which virtually preclude diffraction-limited operation even if the materials themselves are of diffraction-limited quality [1]. The changes in optical pathlength induced by optical pumping of neodymium-doped glass rods have been studied by several investigators [2-8]. This overall change is caused by changes in certain parameters; thermal expansion alters the physical dimensions of the cavity, and the refractive index varies as a function of the local temperature and stress. Data on these properties are needed to calculate the corrections for the distortion of the wavefront.

Several mechanisms have been proposed in the literature to account for the damage observed in laser materials operated at high power levels [9-12]. These can be properly assessed only when data on the material properties are known.

Compilations of data on the properties of laser materials have been published [13-14], but these are not complete. For the purposes mentioned above, data are presented in this paper on thermal expansion, the elastic constants and the change in refractive index as a function of temperature and stress. Included also are original data on the important laser

'The work described in this report was sponsored by the Advance Research Projects Agency, Department of Defense.

'Figures in brackets indicate the literature references at the end of this paper.

properties; optical quality, transmittance, refractive index, thermal conductivity, hardness, density, and chemical composition. Measurements were made on specimens of five commercially made neodymiumdoped laser glasses, and, for purposes of identification, these have been designed as glasses A, B, C, D, and E.

2. Optical Quality

Each specimen was examined to be sure that the glass contained no gross inhomogeneities, and that the specimen was well annealed. Inclusions and seeds were found by illuminating with a beam of light from the side and viewing normally against a dark background [15]. The optical homogeneity was determined from shadowgraphs and interferograms [16]. The state of anneal was ascertained by measuring the strain birefringence with a sensitive tint plate when the specimen was placed between crossed polarizers [17].

Specimen A was in the form of a cylinder 8 cm in diameter and 2.9 cm thick. Three inclusions that appeared to be metallic and two seeds were found. The specimen was annealed, and the birefringence was reduced to a negligible value except for the very edge of the specimen where a residual path difference of 2 nm/cm remained. The interferogram and shadowgraph are shown in figures la and lb, respectively.

Specimen B, which was cut from a finished oscillator rod, had no seeds or inclusions. It was 5.5 cm

Specimens C and D were in the form of rectangular blocks about 4.5 cm X1 cm X1 cm. No seeds or inclusions were observed, and the glasses appeared to be well-annealed. The specimens were too small for meaningful interferograms so that only shadowgraphs were taken. The shadowgraph of specimen C is shown in figure 3a and that of specimen D is shown in figure 3b.

Specimen E was in the form of a cylinder 15.2 cm long and 2.5 cm in diameter. There were no seeds or inclusions, and the specimen was well-annealed. The interferogram and shadowgraph are shown in figure 4a and b, respectively.

From the striae that can be observed in the five shadowgraphs, it was concluded that all of the glasses could be regarded as Grade B,2 or better, according

to the military specification for optical glass [15]. The absence of abrupt changes in the fringe pattern of the interferogram was a confirmation that there

In reference [15], Grade B glass is defined as glass which contains only striae that are light and scattered when viewed in the direction of maximum visibility when tested by the specified methods.

FIGURE 2. (a) Twyman-Green interferogram of laser glass specimen B; (b) shadowgraph of laser glass specimen B.

were no gross inhomogeneities present. From speci mens A through E smaller samples were cut for the various studies of this report.

FIGURE 6. (a) Schematic of spectrometer used for visible-region refractometry. A, source; B, divided circle; C, prism table; D, collimator; E, telescope; (b) schematic of spectrometer used for nonvisible-region refractometry. A, source; B, divided circle; C, prism table; F, collimating mirror (fixed): G, movable mirror; H, detector.

erves only to simplify our adjustment of the paramter C which enters into the formula nonlinearly. The term drops out of the final equation for relatively mall sets of data, as is the case in these experiments, where the wavelength range is primarily over the isible and near-infrared regions of the spectrum. The average absolute residuals of index for these lasses were from 5 to 10 × 10-6. The maximum reidual, as much as 3 x 10-5, was obtained at λ = 0.5876 m where there is observational uncertainty because of

the proximity of this line to a strong absorption band of neodymium. The computed values of index are listed in table 2 and plotted as a function of wavelength in figure 8 (glasses designated B and D are represented by the same curve for sake of clarity). The refractive index at any specific wavelength intermediate to those listed in the table for each glass may be interpolated to five decimal places by means of the dispersion equations.