mediately following the heat treatment. 
Figure 6 presents the absorbancy- 
index values obtained as a function of 
time, and includes previous data where 
heat treatment was made one hour 
after irradiation. It is evident that 
immediately following heat treatment 
the absorbancy indexes drop to values 
approximately linear (dashed curve) 
with respect to the logarithm of the 
time, for all the time intervals between 
irradiation and heating (from 1 hour 
to 8 days) that were tested. 
However, subsequent to the initial 
measurement after the heat treatment 
the absorbancy indexes of a given group 
of glasses have, in general, a different 
linear relationship with respect to time. 
Fading following heat treatment pro- 
ceeds at a lower rate than that shown 
by the curve based on measurements 
taken immediately following heat treat- 
ment applied at different intervals. 
This would seem to suggest that the 
heat treatment not only eliminates 
some of the more unstable centers but 
alters the stability of some of the re- 
maining centers so as to reduce their 
probability of fading. This is verified 
by noting that the absorbancy indexes 
of glasses not given heat treatment, in 
several cases, fall below glasses given 
heat treatment. 
These results indicate that if heat 
treatment is to be applied at different 
time intervals following irradiation, 
one should not correct for the time 
interval by utilizing data on fading 
following heat treatment one hour 
after irradiation. Rather, it would be 
preferable to employ actual curves ob- 
tained by heating glasses after different 
time intervals. These curves appear 
to be fairly good linear plots with the 
logarithm of the time, so that measure- 
ments after two widely spaced time 
intervals such as one hour and one 
week will yield data sufficient for 
interpolations within this range. 
Absorbancy Index(cm~') 
6 8 
10 20 30 
7. Effect of freezing temperature 
during irradiation. Sometimes irradi- 
ation at temperatures below 0° C is a 
means of minimizing some of the un- 
desirable chemical and physical changes 
produced by radiation. Accordingly 
studies were initiated on the dosim- 
etry of materials frozen while being 
irradiated. 
Figure 7 shows the results of an ex- 
periment where two groups of glasses 
were given a dose of 3.66 X 10° rep, 
heated for 10 minutes at 130° C, then 
stored at room temperature. One 
group of glasses was frozen in water 
during the irradiation whereas the 
other was irradiated at room tempera- 
ture. Significant differences are ap- 
parent, with the glasses irradiated at 
the freezing temperature showing con- 
sistently lower readings. 
The conclusion is that the silver- 
activated phosphate glass dosimeter 
can be used for frozen materials if a 
correction factor is applied to calibra- 
tion charts constructed from results 
of irradiation at room temperature. 
Alternatively, calibration charts should 
be constructed from data obtained 
from glasses irradiated at freezing 
temperatures. 
High-temperature Erasure 
Preliminary experiments showed that 
exposed glasses may be ‘‘erased”’ (i.e., 
the coloration caused by irradiation 
removed) by application of 400- 
500° C temperatures. Further experi- 
ments were conducted to determine 
how the sensitivity of the glass varies 
with repeated erasures and how many 
repeated erasures are possible. 
Fifteen minutes at 450° C was found 
sufficient to erase these glasses after 
doses up to 5 X 108 rep. 
The results obtained after five suc- 
cessive exposure-erasure cycles, where 
the dose in each exposure was 1.24 X 
10° rep and the glasses were measured 
Room temp 
‘Freezing 
40 60 80 100 200 
Hours After End of Irradiation 
FIGS 7. 
Fading characteristics of glass given dose of 3.66 X 10° rep at two 
different temperatures, heat-treated for 10 min at 130° C, and stored at room 
temperature 
one hour after irradiation, are given in 
NYO-3345 (13). No definite increase 
or decrease in sensitivity of the glasses 
with repeated exposure-erasure cycles 
is observed. 
A series of ten successive exposure- 
erasure cycles was conducted in the 
high-dose region, doses being chosen 
randomly in the region of 2.5-50 X 105 
rep. The average values for a group 
of three glasses are plotted on curves 
determined for nonerased glass from 
the same lot (Fig. 8). The points on 
the curves are marked to show the se- 
quence of the exposure-erasure cycles. 
It is evident that the glass behavior re- 
mained essentially constant even after 
ten successive erasure-exposure cycles. 
Furthermore, the fading behavior ob- 
served in cycles 8, 9, and 10 remained 
unchanged from that of the original 
glass. 
It seems that these glasses can be re- 
used after erasing at high temperature; 
this reuse will add considerably to the 
value of the glass as a dosimeter. 
How Precise? 
Because of the large number of 
glasses whose optical densities were 
measured under identical conditions in 
this investigation, it has been possible 
to compile sufficient data for statistical 
analysis of the reproducibility charac- 
teristics of Corning 9761. 
The table on p. 35 presents probable 
percentage limits of error determined 
from a total of 48 glasses irradiated 
under nearly identical conditions* over 
a period of two months. These glasses 
were all from the same lot and all re- 
ceived a dose of about 1.30 X 10° rep. 
Optical density measurements on 
nineteen non-irradiated glasses from 
the same lot were also analyzed statis- 
tically. The ranges of values calcu- 
lated at a 99.7 % probability level were 
typically 0.045 + 0.007 at 4,000 A. 
It may be concluded that no ap- 
preciable increase in precision is ob- 
tained if individual glasses are meas- 
ured before irradiation with doses of 
10° rep or greater. The mean values of 
a few nonirradiated glasses from the 
same lot will suffice for the preirradia- 
tion absorption corrections. However, 
at the lower dose levels, where the pre- 
irradiation readings represent a larger 
percentage of the postirradiation read- 
* Slight differences in dose due to length 
of irradiation or source decay were com- 
pensated by correction of the optical read- 
ings before statistical analysis. 
