sponsible for (i.e., to purchase) the mill 
scrap. In any event, prudence re- 
quires as short a half-life as possible. 
One year would seem to be a reasonable 
upper limit. 
All of these requirements are met 
very satisfactorily by the nuclide tan- 
talum-182. Its nuclear characteristics 
(1, 2) are 110-day half-life, 0.53-Mev 
betas, and 1.1-, 1.2-Mev and other 
gammas. One of the outstanding ad- 
vantages of tantalum-182 is that it is 
readily available in very high specific 
activities—up to 50 curies/gm. This 
means that entirely insignificant amounts 
of the element would have to be added. 
Experimental 
Simple experiments have been car- 
ried out to determine the lowest specific 
activity required for easy “‘tracing’’ of 
a steel and the dose rate to be expected 
in the neighborhood of a large accu- 
mulation of the tagged metal. 
Tracer preparation. Two pieces 
(60.6 and 111 mg, respectively) of 
25-mil tantalum wire were irradiated 
in the reactor at Brookhaven National 
Laboratory for an equivalent nvt of 
9 X 10!4 flux-seconds. The tantalum 
content was determined spectrographi- 
cally and found to be more than 98% 
Qualitative analysis showed no impuri- 
ties in quantity sufficient to produce ap- 
preciable radioactivity on irradiation. 
A few milligrams from each specimen 
were dissolved in hydrofluoric acid and 
the absolute beta activity determined 
by counting an aliquot of each of the 
diluted solutions. Since the decay 
scheme of Ta!*? is not known (1, 2), it 
was decided to use absolute beta count- 
ing for standardization rather than the 
less reliable gamma counting tech- 
niques. The aliquots were evaporated 
to dryness on gold-coated Pliofilm* 
* Rubber hydrochloride, sold by Reed 
Laboratories, Akron, Ohio. 
TABLE 1—Specific Activity Data* 
mounts (less than 0.7 mg/cm?), and the 
activity determined with 47 counters 
(3) of two different designs—both of 
which have been successfully used for 
several months at KAPL and checked 
against standard activities supplied by 
the National Bureau of Standards. 
The activation results are given in 
Table 1. 
With identical aliquots of these same 
solutions, the counting efficiency of the 
scintillation counter for Ta!®? was de- 
termined. This counter is a standard 
instrument in which the detector is a 
NalI(T1) crystal 14 in. thick and 2 in. 
in diameter. The background with 
lead shielding averages about 55 cpm. 
Accordingly, the Ta'*? counting effi- 
ciency under the experimental condi- 
tions is 6.5 + 0.1%. 
Preparation and sampling of billets. 
As stated previously, two cylindrical 
billets of mild steel were vacuum cast 
in graphite molds by induction heating. 
Dimensions of the as-cast billets were 
6.1-cm diameter, and 19.5-cm length. 
The tantalum wires were added before 
melting. Expected (calculated) spe- 
cific activities are given in Table 2. 
To be certain that the activity was 
uniformly distributed throughout the 
metal, chips were collected by machin- 
ing about 10 mils off the radius of each 
billet, and by drilling three 14-in.- 
diameter holes equally spaced along 
the axis of each billet. A 1-gm sample 
of each lot of chips was mounted on 
cardboard and counted with the scintil- 
lation counter. From the average of 
these determinations, the ‘observed 
specific activities’? were obtained and 
are tabulated in Table 1. 
In view of the considerable differ- 
ences between the chip mounts and the 
standard mount used to determine the 
counter efficiency, the agreement be- 
tween the calculated and observed 
specific activities is considered satis- 
factory. Further, the agreement be- 
tween chip samples from any one billet 
showed that the activity was dis- 
tributed uniformly to within 3% or 
better. 
The original billets were ‘‘counted”’ 
with a standard scintillation probe* 
and with a beta-gamma survey meter;t 
results are shown in Table 2. Also 
shown are the survey-meter counting 
rates observed with pieces of the billets 
weighing approximately 1 lb. Meas- 
urements were made with the detectors 
all but touching the sample. 
Since an estimation of the radiation 
dose rate requires a knowledge of the 
activity as a function of thickness of 
tagged metal, a series of disks of vary- 
ing thickness were cut from each of the 
billets. These disks were then counted 
in the scintillation counter, care being 
taken to maintain a fixed distance be- 
tween the top of the disk and the 
bottom of crystal can. 
The data for disks from billet No. 2 
are presented in Table 3. The “effi- 
ciencies”’ are the usual ratio of ob- 
served counts to total activity in the 
sample. The “relative counting effi- 
ciencies’’ are ratios of the observed 
counting rates to those expected if the 
sample were counted with the efficiency 
of the 50-mil disk. These efficiencies 
are then a measure of the self-absorp- 
tion of the radiation by the metal. 
The efficiencies are presented in 
graphical form in the illustration on 
page 18. The value for the 2.00-in. 
disk was obtained by extrapolation. 
Clearly this measurement is approxi- 
mate in that it takes no account of 
scattering by the walls of the counter 
* An example is the RCL scintillation 
head, mark 10, model 10, Radiation Coun- 
ter Laboratories, Chicago, III. 
+ For example, alpha-beta-gamma survey 
meter, model SU-5, Tracerlab, Inc., Boston, 
Mass. 
TABLE 2—Results of Survey of ‘‘Tagged’’ Billets 
Place counted and 
Scintillation Beta-gamma 
Weight physical arrangement probe (cpm) survey meter (cpm) 
of Ta Total 
wire activity Specific activity (curtes/gm) Backereund 3.550 ~200 
Billet (mg) (curies) Calculated Observed Bileteendron 6,150 
Billet 1—side 7,030 1,700 
Now 55.3 95.21 <10=* 1215) x 10-9 1.28 +10)..03 x 105? Billet 2—end on 12,100 
No.2 95.1 9.06 X 10-§ 2.00 x 10-9 2.44 + 0.07 X 10°° Billet 2—side 13,700 3,900 
529 gm of billet 1 — 800 
* All activities corrected for decay. 567 gm of billet 2 = 1,500 
163 
