WORLDWIDE EFFECTS OF ATOMIC WEAPONS
SAMPLING AND RESEARCH PROGRAM
75 per cent. It should be possible to do this on counters small enough to
ensure the detection of the few disintegrations per minute expected in the
average sample for the worldwide assay. Typical counters that have been
used in other connections have sensitive areas of 200 to 400 cm? with wall
thicknesses of 2 or 3 mg/cm* and background counting rates of six or
eight counts per minute. This means that up to 1 or 2 gm of yttrium
sample can be mounted in thin layers under conditions of maximum common sensitivity, thereby making it possible to measure disintegration rates
Since both strontium and its yttrium daughter are pure f-ray emitters,
one immediately chooses a Geiger counter as the detection instrument to
be used, since it is the only instrument that will detect a single thermal
energy electron with apparently 100 per cent efficiency. It is, therefore,
clearly the most sensitive of all instruments for the detection of ionization
as such. Having selected the Geiger counter, one seeks to mount the
strontium or yttrium sample in the position that will ensure the maximum
ratio of sample activity to background activity. It is clear that the counting
geometry of the Geiger counter should be as high as practicable. Several
different types of counters might be considered, such as (1) the 47 type,
(2) the conventional thin-window bell type, (3) the windowless flow-type,
and (4) the thin-walled cylindrical type.’ The choice of a particular type
of counter is somewhat arbitrary and will depend on the samples to be
analyzed and on the availability of the equipment.
ft is further clear that one should not interpose between the counter gas
and sample any more solid material than is necessary, though both the
strontium and yttrium radiations are quite penetrating (Sr°° range, 180
62
of one or two, or to be conservative, five, disintegrattons per minute with
some degree of accuracy. The techniques employ the principle of anticoincident shielding for the reduction of the counter background. The
construction of counters of clean materials and the general techniques
employed in other low-level counting applications should suffice. The
particular apparatus has been described by several workers—e.g., the
apparatus used in natural radiocarbon assay."
The problem of detection and measurement of weak radioactive substances is an ancient one, around which considerable lore and artistry
have been built. In the case of radiostrontium at the levels likely to be
found, one needs to use some of the more sensitive techniques known.
On the other hand, it is essential that the procedure be as simple and as
reliable as possible. There are two obvious ways in which to proceed in
the low-level strontium-measurement problem. One is to measure purified
strontium, as such, in equiltbrium with its yttrium daughter-——the equilib-
rium being ensuredby allowing the purified strontium to stand for at least
a week before measurement. The other is to separate the yttrium daughter
and to measure it alone. The first technique has the advantage of giving
more radiation for measurement, but is obviously inapplicable to samples
that contain a large bulk of strontium and are thus more subject to radioactive contamination errors. The second technique, though it involves
sacrificing half of the radiation, affords the opportunity of measuring a
large butk of matertal by treating the yttrium daughter, which in itself
ant
Guar ¥et
ee ee
1
at
PLU
+
UP Grr
Lari
1}
rE
eERGhU Cr
c
set
1
Ls
c*
Orea
provided the sample has previously been purified for bulk yttrium and
other rare earths. In order to demonstrate the desirability of the second
procedure in these instances, we will give some calculations below that
indicate the regions of applicability of the two procedures.
63
mg/cm’; Y"" range, 1065 mg/cm’; half-thicknesses, 14 and 134 mg/cm’,
respectively). The solid material in the sample itself must also be con-
sidered, For an isotropic source, if X represents the thickness of the sample
in mean free paths (X = 0.693 T/r, where T is the thickness and is the
half-thickness in milligrams per square centimeter), then the flux through
the surface of the sample is given by
F(X) eh ~ er f° 2Ei(—z) dz,
(1)
where N is the number of particles originating in unit time in a unit
volume of the sample, » is the absorption coefficient, E7(—z) is the expo-
nential integral {° (e*/x) dx, x is the depth of the volume element below
the surface, and z== px. The activity from a sample of thickness T
me /cm? is then
A=Adsis)! on (2) HC)
where A, = 14 times specific activity in d/m iimes area in square centi-