The Human Plutonium
Injection Experiments
192
Los Alamos Science Number 23 1995
Met Lab procedure to analyze urine
samples of four Los Alamos workers
who had already experienced instances
of high readings from their nose swipes
failed to detect concentrations of pluto-
nium alpha activity consistent with the
high nose-count records.
As it turned out, one problem with the
Chicago procedure was that running a
complete 24-hour urine sample (1 to 2
liters) through the column overloaded
the resin with organic material. A drop
in resin performance altered results and
nullified the expected increases in sen-
sitivity. The Chicago method worked
well with 100-milliliter aliquots at the
activity level of excreted plutonium-
239 expected for 5-microgram body
burdens. But detection of body bur-
dens of 1-microgram or less would re-
quire an analytical procedure that used
a 24-hour urine sample and eliminated
the organic material and urine salts.
Concerns were heightened by an acci-
dent in August in which part of a plu-
tonium-chloride solution sprayed into
the mouth of Don Mastick, a young
chemist (see “A Swallow of Plutoni-
um”). How much of the plutonium had
been absorbed by his gastrointestinal
tract? What fraction of a serious dose
did the absorbed plutonium represent?
Was it safe for him to go back to work
at his old job and possibly be exposed
again? In fact, to avoid further expo-
sures, Mastick was transferred tem-
porarily to Hempelmann’s group “to
work on the problem of detection of
plutonium in the excreta.”
The research team at Los Alamos that
attacked the problem of detection meth-
ods included Perley, who continued to
investigate the Chicago procedure,
Robert Fryxell, who studied a method
of separating plutonium from urine that
used cupferron as the main complexing
agent, and Mastick, who investigated
various ether extractions. The analyti-
cal procedure for isolating plutonium
from one liter of urine (a 24-hour sam-
ple) was outlined by Arthur Wahl. In
September, Roger Kleinschmidt joined
the team to investigate methods of iso-
lating plutonium from urine ash samples
using a lanthanum-fluoride carrier to
precipitate plutonium from the dissolved
ash. He would also direct the plating
and measurement of the final precipitate
with a goal of 90-per-cent chemical re-
covery of spiked urine samples.
Fryxell consulted with Wright Lang-
ham on the cupferron technique for
plutonium isolation. Langham was a
biochemist who had been transferred to
Los Alamos in July 1944. Previously,
he had spent a short period at the Met
Lab in the analytical chemistry group
where he’d been involved in plutonium
purification research. Before long,
Wright Langham would become one of
the major names associated with the
detection, analysis, and evaluation of
plutonium in humans.
By late 1944,
Hempelmann’s team had devised a sat-
isfactory technique, using cupferron ex-
traction, for analysis of urine contain-
ing tenths of a nanogram of plutonium.
After collection, the samples underwent
a multistep preparation that included
evaporation to dryness, treatement with
acid and peroxide to remove organic
matter, and the cupferron extraction
step. Eventually, the plutonium was
carried out of solution as a co-precipi-
tate with lanthanum fluoride, and this
final precipitate was transferred to a
platinum disc. The activity of the plat-
ed sample was measured by placing the
disc in an alpha counter.
However, analyzing spiked urine sam-
ples—or even samples taken from ani-
mals—in a laboratory environment was
one thing. Analyzing samples from
people working with plutonium on a
daily basis was another thing entirely.
Early assays of workers yielded surpris-
ingly high results, indicating that if the
0.01-per-cent-per-day excretion rate de-
rived from the animal data were applic-
able to humans, then these workers had
significant levels (greater than micro-
gram amounts) of deposited plutonium.
Sample contamination. An analysis
technique sensitive enough to detect
tenths of nanograms would easily de-
tect tiny particles of plutonium dust or
contaminated skin that, say, dropped
from a worker’s hand into the sampling
flask. As a result, a collection proce-
dure was set up in which the worker to
Estimates of the Detection Regime
Plutonium-239 has a specific activity of 0.06 curies per gram, which means
that a nanogram of the substance undergoes about 130 disintegrations per
minute ((0.06 Ci/g) (10
-9
g/ng) (3.7 x 10
10
d/s/Ci) (60 s/min)
<
130 d/min/ng).
However, the Hanford “product” contained small quantities of other plutonium
isotopes (at the time, it was commonly referred to as 239-240 Pu), and ac-
counting for such impurities increases the rate to about 140 disintegrations
per minute per nanogram. If we want to detect a tolerance limit of 5 micro-
grams of “product” in the body and only 0.01 per cent of the plutonium is
being excreted per day (several weeks after the initial exposure), then a 1-
liter, 24-hour sample of urine will contain 0.5 nanograms of plutonium. If
only 100 milliliters (10 per cent) is analyzed, the test must be capable of de-
tecting 0.05 nanograms of plutonium. A sample at this level emits about 7
alpha particles per minute (0.05 ng
3
140 d/m/ng), which, in an alpha counter
with 50 per cent efficiency, corresponds to a reading of 3 or 4 counts per
minute. If we want to detect a lower tolerance limit of 1 microgram—one-fifth
as large—the counting rate drops to less than 1 count per minute.