Page 255 - Chalcogenide Glasses for Infrared Optics

230 Cha pte r Ni ne

When AMI had a rod 1.5 to 2 mm in diameter and 8 cm long tested at
Quantronix, 50 percent of the energy was lost by scatter due to the
lack of a smooth surface. As evidence of that fact, transmission varied
greatly with angle of incidence of the beam relative to the face of the
rod. Rods were etched to reduce diameter and smooth their outside
surfaces. The quality of the etched surface was affected by seed
orientation. Even so, rods in this diameter range had, on average, a
bend-to-break radius of about 1.75 in. There was wide variation
depending on the seed orientation. The silicon fibers were not flexible.
Absorption and resistance to the intense radiation test were passed.
Figure 9.13 shows typical results for single-crystal grown silicon rods.

9.7 Gallium Arsenide as an Infrared Optical Material
The effort to develop the compound semiconductor gallium arsenide
(GaAs) began in the late 1950s at TI. As mentioned earlier, George
Cronin was deeply involved in the effort from the beginning. The
idea was to replace silicon for high-frequency devices and enjoy
greater temperature stability due to a larger bandgap than that in
silicon. Production of GaAs in purity and quality as a device material
that could be fabricated went on for years. The technological difficul-
ties associated with crystal, growth, preparation, and device fabrica-
tion had been greatly underestimated. The author was only slightly
involved as part of his assignment in the analytical chemistry group
as head of physical methods applied to semiconductor materials, e.g.,
measuring the photoconductivity associated with impurities as a
function of wavelength in GaAs crystals. The author measured and
characterized, using the Perkin Elmer 13 instrument, the light output
from the first TI gallium arsenide light emitters. From the measure-
ments, conversion efficiency of current to light was calculated for the
first time. Several TI GaAs transistor task force teams were organized
and operated over several years, trying to solve all the production
problems.
The author first became interested in GaAs as an infrared optical
material during the high-energy laser window programs 12,13 of the
1970s when GaAs became considered as an optical material rather
than used only for electronic devices. On the periodic chart, Ga is on
one side of Ge and As is on the other side. GaAs is said to be isoelec-
tronic with Ge. Many of the physical properties of the two materials
are almost identical. As one of the candidate window materials, GaAs
14
suffered from not being produced in large enough size, high enough
purity, low enough cost and was shown to be subject to high-speed
rain erosion damage. Over silicon and germanium it had an upper
use temperature of 400°C. At TI in the 1960s, George Cronin and his
15
colleague Bob Haisty developed a method based on chromium
doping to insure gallium arsenide would be semi-insulating. Cronin