' ’&%_-a 17‘ F

-
‘

I
i

M! w
IN

I

ll‘

‘

x
l

’

IIW

‘
‘

I

I

l

l

Ir
—s w W

 

AN EXPERBRflEE‘IfifiL EN‘t4”§$‘fEGA'f'§@M CEF THE
ALPHA-a ELECT ROE‘é .ENTERAC’Z iCN ACCOMWANYIHG
ALPHA @ECA‘I’

Thesis For: $416 (Emma: (2% {at S‘
MECHEGMQ STAT" 1' UMVEMET‘Y
Amara R; év‘izcsfl'tifian

€960

 

IHIIIINIINIIIHIIHIWlllllJllHJllllllllllllllllllllllllil

301764 0065

LIBRA R Y
Michigan Sta.
University

AN EXPERIMENTAL INVESTIGATION OF THE ALPHA-ELECTRON
INTERACTION ACCOMPANYING ALPHA DECAY

BY

Allan R. McMillan

AN ABSTRACT

Submitted to the College of Science and Arts of Michigan
State University in partial fulfillment of the
requirements for the degree of

MASTER OF SCIENCE

Department of Physic s

1960

5‘ ' \ 4 H" I
Approved \"\.I’._k."\‘-~‘. A”. _ 1...,

 

ABSTRACT

An experimental investigation of the alpha-electron interaction
accompanying alpha decay was accomplished by examining the K and L

210 using NaI (T1) scintillation

x -rays accompanying the alpha decay of P0
detectors. The probability of the ejection of an electron from the atom
was found by establishing the ratio of primary K and L shell vacancies
to alpha particles emitted. The K shell ratio of (2. 75 :l: O. 25) x 10'5
primary K shell vacancies per alpha particle emitted is, within experi-
mental error, equal to the value predicted by Migdal. The L shell
ratio of (8.4) x 10'4 primary L shell vacancies per alpha particle
emitted is almost an order of magnitude larger than Migdal‘s theory

predicts. Two experiments are suggested to establish the cause of

this large difference.

ii

AN EXPERIMENTAL INVESTIGATION OF THE ALPHA-ELECTRON
INTERACTION ACCOMPANYING ALPHA DECAY

BY

Allan R. McMillan

A THESIS

Submitted to the College of Science and Arts of Michigan
State University in partial fulfillment of the
requirements for the degree of

MASTER OF SCIENCE

Depa rtment of Physic s

1960

GIZHCg

H. '7- £0

ACKNOWLEDGMENT

The author wishes to thank Drs. G. B. Beard
and W. H. Kelly for suggesting this problem and their
valuable discussions throughout its execution.

***********

iv

TABLE OF CONTENTS

Page

I. INTRODUCTION ..................... 1
11. DESCRIPTION OF EXPERIMENTS ........... s 3
III. RESULTS ..... . .................. 8
K X-Luay to Gamma-ray Ratio Determination . . . 8

L X-‘ray to K X-Ir.ay Determination ......... 11
Results of the L to Alpha Experiment ........ 15

IV. COMPARISON OF RESULTS AND DISCUSSION ..... 18
BIBLIOGRAPHY .................. . . . . 24
APPENDIX ....................... . . 25

LIST OF FIGURES

FIGURE

m<©m$WN

210

. Decay Scheme of P0 ......... . ......
. Small Vacuum Chamber ...............

. Po210 800 Kev Gamma Ray Spectrum. . . . . . .

P0210 K X-‘ray Spectrum .......... . .

. P0210 L X-r.ay Spectrum . .......... .
. Po?‘lo K X-ray Spectrum .......... . .
. P0210 0. Spectrum ..............

. Log N vs. Layers of A1 Foil ......... . . .

vi

10
l3
14
16
26

I. INTRODUCTION

Evidence of K and L shell ionization accompanying the alpha
decay of P0210 has been advanced by Grace, Allen, West and Halban (l),
Curie and Joliot (2), Barber and Helm (3), and Rubinson and Bernstein
(4). Grace e__t a_.l. measured the number of K x- rays relative to the
number of alpha particles and explained the K x- rays as the result of
internal conversion of the 800 kev gamma ray accompanying the
polonium decay. Barber and Helm also examined the number of K
x- rays relative to the number of alpha particles and obtained results
in agreement with Grace gt a_._l. However Barber and Helm concluded
that the K 'x- rays were not due entirely to internal conversion, but
rather to ionization caused by the emission of the alpha particles.

Curie and Joliot examined the lower energy L and M x- rays
using an ionization chamber connected to an electrosc0pe while
Rubinson and Bernstein investigated the L x- rays with prOportional
counters. In the course of these experiments Rubinson and Bernstein
demonstrated that the L x- rays are those of lead and not of polonium
excited by alpha bombardment of P0210 atoms in the source.

Migdal (5) first considered this problem theoretically. Making
use of the low velocity of the alpha particle compared to the velocity
of the atomic electrons he calculated the probability of a primary
electron vacancy in the final atom using time dependent adiabatic
perturbation theory. In these calculations Migdal assumed that the
alpha particle moves through the atom with a uniform velocity and
used non- relativistic hydrogenic wave functions.

Barber and Helm obtained a ratio of primary K shell vacancies

to the number of alpha particles emitted which was lower than the

ratio predicted by Migdal by a factor of about two. Rubinson and
Bernstein measured the ratio of L shell vacancies to alpha particles
emitted and obtained results an order of magnitude larger than the
value predicted by Migdal.

Levinger (6) re-examined the problem and introduced quadrupole
terms in his calculations whereas Migdal had used only the dipole
terms. Levinger's results predict a larger L shell yield than those
of Migdal but still account for less than twenty per cent of the L shell
x- ray yield observed by Rubinson and Bernstein. Levinger's pre-
dictions for the K shell ionization are an order of magnitude smaller
than the results of Migdal and an order of magnitude smaller than
the experimental results of Barber and Helm.

Schwartz (7) has considered the problem in a different manner
and has obtained results that are essentially equivalent to those of
Migdal. Schwartz invokes the relative smallness of the charge of the
alpha particle rather than its low velocity compared to the velocity of
the atomic electrons. He takes into account the recoil of the nucleus
and the change of charge that takes place at instant of alpha emission.

We are confronted with a large difference between the predictions
of the theory and the results of the experiments. Since the most
recent experiments were carried out in 1952 it was felt a new experi-
ment with P0210 and perhaps some other alpha emitter might yield

somewhat improved results with the use of better equipment.

11. DESCRIPTION OF EXPERIMENTS

In the measurement of the ratio of K x- rays to alpha particles
it is obvious that the use of the same source and geometry to count
K x- rays and alphas would greatly simplify the problem. Unfortu-
nately the low intensity of the K x- rays makes such a measurement
almost impossible. The existence of an 800 kev gamma ray in the
polonium decay scheme (fig. 1) makes possible an indirect measure-
ment of the K x- ray to alpha particle ratio.

A 1 me. source was prepared by evaporating to dryness a

*
21° in the form of polonium nitrate on a film of mylar

solution of P0
and attaching the mylar to a square of aluminum foil. Only the side

of the source covered with the mylar was used for counting in the subse-
quent experiments. The source was mounted 3 cm above a 1 3/4" by

2" cylindrical NaI (T1) crystal placed on an RCA 6342 photomultiplier
tube. The photomultiplier was mounted on a cathode follower preamp
and the entire counting arrangement was shielded on all sides with a
layer of lead bricks. Pulses from the preamp were fed into an RIDL
model A-61 amplifier and the amplified pulses were then analyzed

and recorded by an RIDL model 3301 256 channel pulse height analyzer.
In our experiment the spacing of the source above the detector is not
critical because this distance enters the calculations only in the
evaluation of the crystal efficiencies and essentially cancels out of

the calculations since we are interested only in ratios.

The determination of the K x- ray to alpha ratio was accomplished

by establishing the ratio of K x- rays to 800 kev gamma rays. This

 

:4:
Obtained from Atomic Energy of Canada Ltd. , Chalk River,
Ontario.

2+

0+

Figure l. Decay Scheme of P0

0+

210

 

 

0.8 Mev X

 

8?.

   

5.3 Mev ex

. szob

result is then combined with the ratio of 800 kev gamma ray to alpha
particles as given by Ascoli it a:_l. (8) to give an indirect determination
of the K x- ray to alpha ratio.

The determination of the L x- ray to alpha ratio was obtained by
establishing the ratio of L x- rays to K x- rays and combining the
result with the experiment described above. The first of the L x- ray
to K x- ray determinations was accomplished by placing the 1 me source
4. 2 mm above a 1 1/2” by 2 mm cylindrical NaI(T1) crystal with a
4 mil beryllium window. A second experiment was performed with the
same source 0. 5 cm above an unmounted 1 1/2" by 1/4" cylindrical
NaI(T1) crystal. The electronic arrangement for these experiments
was the same as described above.

The reasons for using the thin crystals were to reduce the back-
ground and Compton background from the higher-energy gamma and
because the available thicker crystals were mounted in aluminum
housings which attenuate the low energy L x- rays to a negligible
intensity.

A direct determination of the L x- ray to alpha ratio was also
performed. The difficulties of such an experiment are twofold. First,
the use of a vacuum chamber is required for the alpha counting. The
use of the vacuum chamber causes a great deal of scattering from the
walls of the chamber and adds a low energy tail on the alpha peak
(see fig. 7). This problem can be reduced by the use of baffles but we
were unable to completely eradicate the problem. The second problem
is that a source must be used which is strong enough that the L x- ray
is well resolved from the background. Such a source will be so strong
an alpha source that the detector must be partially shielded to prevent
the alphas from saturating the amplifier and analyzer. The exposed
area of the detector is critical in the determination of the efficiency of

the detector. It is then very important to know the geometry of the

experiment as accurately as possible. In our experiments we knew
the diameter of the detector to i O. 000]" and the height to :l: 0. 05".

The direct determination of the L x- ray to alpha ratio was
performed using the small vacuum chamber shown in Fig. 2. A new
source of ~0.1 mc was prepared by evaporating to dryness the
.polonium nitrate solution on a strip of aluminized mylar. The source
was then centered over a 5/16" circular hole in a circular aluminum
disk of thickness 1/32" and the other side of the hole was covered
with a thin film of zapon. The purpose of the zapon film was to avoid
contamination from recoiling clusters of P0210. A 1 1/2" by 2 mm
unmounted NaI(T1) crystal was placed on an EMI9578B photomultiplier
and joined with the vacuum chamber. For the measurement of the
alpha intensity it was necessary to use a small window over the face
of the crystal to effectively reduce the size of the crystal. An aluminum
plate with a 3/16" hole in the center was fitted over the face of the
crystal. The aluminum plate was tapered such that the edge of the
hole was reduced to a thickness of a few thousandths of an inch.

This was done to reduce the alpha scattering from the edge of the
window. For the measurements of the L x- ray intensity the plate was
removed and replaced by a thick aluminum ring of inside diameter

1 1/4" to insure accurate geometry.

A sheet of aluminized mylar of thickness 1. 9 :1: 30% mg/cmz
was stretched over the crystal for the alpha counting to improve the
optical qualities of the crystal. The mylar was replaced by a layer
of aluminum foil of thickness 4. 3 mg/cmz in the L x- ray measurement.
The aluminum foil aided reflection in the crystal and stopped the alpha

particles.

Figure 2. - Small Vacuum Chamber

5%” ALUMINUM ZAPON "FILM

.50URCE HOLDER

-SOURCE ON ‘ALUMINIZED
MYLAR 'BACKlNG

3);” ALUMINUM BAFFLE

rll/IZ

”11% —.._, m/wrn.
\= _

 

L\\\

' LUMINUM FO|L
ALUMINUM OR MYLAR
KEEPER RING u
Ifixzmm
TYPE EMI 95788 NA: CRYSTAL
PHOTOMULTIPLI ER -

 

 

III. RESULTS

K X-ray to Gammauray Ratio Determination

The determination of K x- ray to gammawray ratio resulted
in the spectra shown in Figs. 3 and 4. The dotted line in the gamma-
ray spectrum shows the background with no source present. The
dotted line in the K x- ray Spectrum shows the assumed noise and
Compton background.

The total number of quanta emitted by the source per unit time
is obtained from the expression

N
T(E) P(E) A, t

 

No:

where N is the number of quanta detected under the peak

T(E) is the efficiency of the detector
P(E) is the peak to total ratio
Ai is any correction or attenuation factor

t is the duration of the measurement.

The NaI(T1) efficiencies and peak to total ratios given by Heath

fit all. (9) are

T(81 kev) 0. 0949 a: 0. 0005
T(800 kev) 0. 0370 i 0. 0003
P(81 kev) 1. 000

P(800 kev) 0. 342 i 0. 004

Counts

 

P0210 800 Kev Gamma Ray Spectrum
100 Min. January 13.

Figure 3 .

 

 

 

I03 5
o
\ 1
I
J
\ O
\
\
~ Io” °\ --
“h
°\
K
s
: I I I I
2.0 40 60 50 \00

Channel Number

. Counts

Figure 4. Po210 K X-ray Spectrum
100 Min. January 13.

 

 

 

Channel Number

3
.. 2X )0
I03 .
. os-
« ox IO q
0
6‘79
e
{I 9 \
‘ ‘.
\
\
\
\
\
\
\
\
\ ______________
2'5 42) 50 50 I00

10

11

The total number of counts under the K x- ray peak including
the escape peak and excluding the background and Compton background*
was (33. 3 :1: 0. 24) x 103 per 100 minutes. The total number of counts
under the 800 kev gamma ray peak excluding background was
(20. 3 :L- 0.19) x 103 per 100 minutes. Then the ratio of the K x-rays
to the 800 kev gamma rays is 0° 218 :t 0.005 (A7 /Ak), where A7 /Ak
is the correction due to the relative absorption of the aluminum housing
of the crystal mounting, calculated to be 1. 016 :1: 0. 008, and the
fluorescence yield, 1. 05 :t 0. 01 (10). With these corrections the ratio

of primary K shell vacancies to gamma rays is

__ = 0.23 $0.01

Using calculated internal conversion coefficients (11) and the
gamma ray to alpha particle ratio determined by Ascole at 2:1. (8)
the final value of primary K shell vacancies caused by the alpha

emission per alpha particle emitted is determined to be

—— = (2.75 i: 0.25) x 10'"6
N

L X-Ray to K X-Ray Determination

The first determination of the L x- ray to K x- ray ratio was
made using the 1 1/2" by 2 mm NaI crystal with the Be window.

The result of this determination was

 

ac:
An investigation of the Compton background was made using

the 840 kev gamma ray of Mn“ with the same experimental arrangement.

This was found to be constant in the area corresponding to the K x- ray
of lead.

12

:1}:- 76 a 12
Nk

The main source of the large error in this experiment resulted from
the evaluation of the detector efficiency using Simpson's rule for
numerical integration and the uncertainty of the actual detector
geometry. The numerical integration was found to be reliable to

only 10% and the effective crystal size was less than the given diameter
due to mounting material around the edge of the window. These errors
were in different directions but would have been extremely difficult

to evaluate.

In the subsequent experiments the L x- ray was not as well
separated from the noise as in the experiment above. It was possible
to use the L x- ray peak in the first experiment to analyze the peaks
obtained in subsequent experiments. The shape and the resolution of
the Spectra in the later experiments were the same as those from the
first experiment. We defined the peak position to be at 0. 464 of the
half maximum width measured from the low energy side of the line.
Then the total number of counts under the peak was 2. 316 :i: 0. 053
multiplied by the high energy portion of the peak, i. e. , that portion
of the peak above 0. 464 of the half maximum width.

The second determination of the ratio of L x- rays to K x- rays
was made using a 1 1/4” by 1/4” NaI crystal as previously described.
The K and L x- ray spectra obtained are shown in Figs 5 and 6. The
determination of the shape of the K peak was made by fitting a
Gaussian curve to the high energy edge of the L x- ray using the method
of least squares and the Compton background was subtracted as in
the K x- ray to gamma-ray determination. These corrections are
shown in dotted lines in Fig. 6. The L x- ray was analyzed as outlined

above.

C ount s

 

13

Figure 5. P0210 L X—ray Spectrum

2 Min. Feb. 2

/ \

 

 

//
/ '~\...Ha1£ Maximum Width
\
fi I03
0.
~L I01
I I I I I I
20 4 o e 0 so I 00 I2 0

Channel Numbe r

C ount s

Iaon3

 

Figure 6. Po210 K X-ray Spectrum
130 Min. Feb. 2

so I30 I75

Channel Numbe r

270

250

14

15

The ratio of the detector efficiencies for the K and L x-rays

was

T(80 kev)

1. a: .. 1
T(lZ kev) 0° 0 0 3

 

The ratio of the L x- rays to the K x- rays after correction for two
=I<
layers of aluminum foil, used to insure no alpha particles were

counted, was
""" = 79. O :l: 2.1

Combining this result with the result of the K x- ray to alpha particle

ratio we obtain

= (2. 2 a 0. 33) x 10-4

No

as the result of the indirect measurement of the ratio of L x- rays

detected to the number of alpha particles emitted.
Results of the L to Alpha Experiment

The determination of the ratio of the L x- ray to alpha particles
by direct comparison of the L x- ray and alpha intensities was
accomplished using the small vacuum chamber previously described.
The alpha spectrum obtained is shown in‘Fig. 7. The L x- ray spectrum
was essentially the same as that obtained in the previous experiment.
Determination of the NaI crystal efficiencies for these measure-
ments consists merely of evaluating the solid angle subtended by the

detector at the source since every alpha particle and L x- ray that

 

*Appendix .

C ount s

- 0

q- '0

d)- '0

 

Figure 7, Po210 Alpha Spectrum
2 Min. April 6

40 so Izo '.60

200

 

240

16

 

Channel Number

l7

strike the NaI crystal will contribute a count to the peak. The
efficiencies of the detector corrected for the finite size of the source
were

T(alpha) : 0. 0080 d: 0. 00056

T(L x-ray) = 0.181 a: 0.009

The L x- ray intensity was calculated as in the previous experi-
ment and the alpha intensity was calculated by correcting the counts
under the alpha peak for background, efficiency and time. The result
was

is.

= (2.49 a 0.23) x 10-4 AL
0.

where AL is the attenuation factor due to one layer of aluminum foil
placed over the face of the NaI crystal to stop the alpha particles in
the L x- ray measurement and equal to l. 12 i 0. 01. * Then the ratio
of the number of L x- rays emitted by the source to the number of
alpha particles emitted is

N

L -4
—— = (2.80 i0.24)x10

Na

It will be noted that there is no definite cutoff of the alpha peak
on the low energy end. This is due to scattering of alpha particles
from the vacuum chamber and straggling. In the analysis of this peak
we counted as the peak that region between channels thirty and 240.
The area from channel thirty to channel ninety contributed less than
one percent to the total area under the peak. The choice of channel
thirty as the low energy end of the peak was arbitrary but does not

appreciably affect the analysis of the peak.

 

*Appendix.

IV. COMPARISON OF RESULTS AND DISCUSSION

Barber and Helm have previously obtained results for the
K x- ray to alpha ratio by a direct measurement of the K x- ray and
alpha intensities. A comparison of their results and our results is

given below.

Barber and Helm

(K/o.) (2. 2 a 0.42) x 10-6

Present experiments

(K/o.) (2.75 a: 0.25) x 10-6

These ratios are the ratio of primary K shell vacancies to alpha
particles emitted.

Barber and Helm have obtained a K x- ray Spectrum that is very
similar to the K x- ray spectra which we obtained (see Figs. 4 and 6).
In our spectra the escape peak is evident. The spectrum of Barber
and Helm gives no indication of a resolved escape peak although they
made corrections for it. We feel that they have left out another
portion of the peak that should be included. We have investigated the
Compton background under the K peak and concluded that this back-
ground is very nearly constant with respect to energy in this energy
region. Barber and Helm have apparently assumed that this Compton
background is strictly decreasing under the K peak. We believe that
they have excluded some of the K x- rays in this analysis and that
their results are thus low by perhaps ten percent. With this correction
their results are in fair agreement with ours.

The direct K to alpha determination that Barber and Helm

carried out was accomplished by reducing the effective efficiency of

18

19

the detector by means of a shutter between the source and the
detector similar to the method we used for the direct NL/N determin-
ation. . Although the diameter of the hole in their shutter is given as
0. 033 cm no statement is made as to the accuracy with which the
area was determined. As was mentioned before the efficiency of
the detector is sensitive to the area of such a small aperture.
Rubinson and Bernstein have obtained results for the L x— ray to
alpha ratio which differ somewhat from our indirect ratio obtained by
combining the L to K ratio with the K to alpha ratio. , Their ratio agrees
to within experimental errors with the direct L x- ray to alpha ratio
we have obtained. A comparison of the results of Rubinson and

Bernstein and our results is given below.

Rubinson and Bernstein

(L/a) (2.93 4 0.44) x 10“

Present experiments
‘(L/a) indirectly (2. 2 :l: 0. 33) x 10"
(L/a) directly (2.8 a 0.24) x 1074

Comparing our results with the results of Rubinson and Bernstein
we believe the higher ratio to be the more accurate. The indirect
ratio has been obtained by combining three ratios and treating the
errors as random errors. The ratios given above are not corrected
for fluorescence yield.

. It is interesting to compare the above results with the theoretical
predictions of Migdal (5) and Schwartz (7). The probability, wnl’

for ejection of an n1 electron is, using the notation of Migdal
Wnl = (V/Ve)z Cnl/ZZ

where v is the velocity of the alpha particle

20

VB is the velocity of the atomic electron,

Cnl is a constant that depends on n and 1.

The C have been calculated by Migdal and are given as

n1

C10 3 Z. 2
C20 + CZI 1' 20. 5
C30 '1' C314” C32 3 68. 6

Using this expression for the K electrons we obtain the result

w = 2.5x10‘6

K
and for the L electrons

wL = 1.13 x10"4

In the calculation of the L probability there is some question as to
what to use for Z the atomic number. Making use of the internal
screening constants given by Sommerfeld and Wentzel (12) we have
used Z = 80 for the 2, 0 electrons and Z = 78. 5 for the 2, l electrons
in obtaining the results given above.

The experimental result we have obtained for the K shell ioni-
zation agrees quite well with the theoretical result. The L shell
results do not agree at all however. Using the values for the fluores-
cence yield for the L shell calculated by Kinnsey (13), we obtain a
ratio of the primary L shell vacancies per alpha particle emitted of
8.4 x 10". The theoretical result accounts for less than twenty
percent of the observed result. An examination of the approximation
used in the calculations of Migdal does not appear to yield the reason
for the large difference in the predicted results and the observed

results for the L shell. These approximations are:

21

1) The charge of the alpha particle is small compared to the
charge of the daughter nucleus. For Po210 the ratio of the
charge of the alpha particle to the. charge of the daughter

nucleus is 0. 025.

2) The velocity of the alpha particle is much less than that of
the electron. For the K shell electrons the ratio of the
alpha velocity to the electron velocity is 0. 1 and for the

L shell electrons the ratio is 0. 2.

3) The use of nonrelativistic hydrogenic wave functions. As
mentioned. by Rubinson and Bernstein this approximation

should be valid for the L shell electrons.

One possible solution to the problem of the large difference in
the predicted L shell yield and the observed result could be the
existence of a low lying level of the Pb206. This level would have to
have an energy less than that of the K shell binding energy and greater
than that of the L shell in order that it would not affect the K shell
yield and still be L shell converted. The gamma transition would have
to be almost completely internally converted such that the gamma
itself would not be observed. We have investigated the possibility of
such an excited level. The 800 kev gamma cannot be involved to any
appreciable extent because its intensity is a factor of ten lower than
the effect we are attempting to explain. We conclude that the low lying
level, if it exists, must be reached through an alpha transition from
the P0210.

We have examined the hinderance due to the difference in energy
of the alpha that would be required for the above transition. - Using the
empirical formula from Perlman and Rasmussen (14) relating the

half life to the disintegration energy of alpha decay

22

A
1T = +B
°g4 «T?—

where T1 is the half life in seconds
Q is the disintegration energy in Mev
Z is the charge of the parent nucleus

A and B are constants (For Po A = 129. 35 B = 49. 9229)

we find a hinderance due to the difference in energy of less than ten
is expected. The spin of the state may also add to the hinderance.
Blatt and Weisskopf (15) quotecalculated hinderances due to Spin
which indicate that Spins of less than 4 would give an effect that would
be at least a hundred times too large. The higher Spin states, _>_ 4,
are hindered enough and would be very highly converted in the L shell
(11). It appears that a high Spin, low energy excited state of lead
could cause the large L shell yield. A Spin of 4 would give rise to
an E-4 transition which should have a lifetime of ~109 seconds (16).

21° and examin-

By chemically separating the lead from a sample of P0
ing the lead for L x- rays it would be possible to investigate the
possibility of such a high spin, low energy state.

However lead is a magic number nucleus and such low lying
states are not to be expected. An investigation of the even-even lead
isotOpes shows that none have an excited level of energy less than
570 kev. That a low energy excited level of Pb?‘06 is causing the high
L shell yield appears unlikely.

Further investigation of the L shell yield for other isotopes
would aid in the solution of this problem. Unfortunately the re are
very few alpha emitters which are adapted to this experiment. The
only other isotope which appears promising is Sml‘" which is a pure

alpha emitter. The long half life (~1011 years) and the low energy

(~5 kev) of the L x- ray of Nd143 to which the samarium decays make

23

it very difficult to examine using the scintillation method. We have
been able to observe what appears to be the L x- ray accompaning this
decay but were not able to obtain accurate results due to a lack of
resolution and a high noise level. We hope in the near future to obtain
accurate results.

The theory of Migdal predicts for the samarium 5 x 10“ primary
L shell vacancies per alpha particle emitted. The L x- ray yield would
be about 1/3 of the number of primary vacancies. An experimental
result Significantly larger than this predicted result would indicate
that the theory is inadequate in its description of the process in the
case of the L shell. A result in agreement with the predicted value

210

would indicate that the decay scheme of P0 is probably inaccurate.

BIBLIOGRAPHY
(1) Grace, Allen, West, and Halban, Proc. Phys. Soc. (London)
.Pfl’ 493 (1951).
(2)1. Curie and F. Joliot, phys. et radium_2_, 20 (1931).
(3) W. C. Barber and R. H. Helm, Phys. Rev. 89, 275 (1952).
(4) W. Rubinson and W. Bernstein, Phys. Rev. _8_6_, 545 (1952).
(5) A. Migdal, J. Phys. (USSR) 4_, 449 (1941).
(6) J. S. Levinger, Phys. Rev. 20, 11 (1953).
(7) H. M. Schwartz, Phys. Rev. L02, 135 (1955).

(8) A. Ascoli, M. Asdente, and E. Germagnoli, Nuovo Cimento i,
946 (1956).

(9) R. L. Heath, L. L. Marsden, and S. H. Vegors, Calculated
Efficiencies of Cylindrical Radiation Detectors (AEC Research
and Development Report IDO- 16370, 1958).

 

 

(10) I. Bergstr'om, Beta- and Gamma—Ray Spectroscopy, edited by
Kai Siegbahn (Interscience Publishers Inc. , New York, 1955),
p. 624.

 

(11) M. E. Rose, Internal Conversion Coefficients (Interscience
Publishers Inc. , New York, 1958).

(12) See H. E. White, Introduction to Atomic Spectra (McGraw-Hill
Book Company, Inc., New York, 1934), p. 318.

 

 

(13) B. B. Kinsey, Can. J. Research 26A, 404 (1948).
(14) I. Perlman and J. O. Rasmussen, UCRL-3424 (1956), p. 33.

(15) J. M. Blatt and V. F. Weisskopf, Theoretical Nuclear Physics
(John Wiley and Sons, Inc., New York, 1952), p. 577.

 

 

(16) D. H. Wilkinson, Nuclear Spectroscopy, Part B, edited by
Fay Ajzenberg-Selove, (Academic Press, 1960).

 

24

APPENDIX

In the course of these experiments it became necessary to
evaluate the absorption of the L x- ray by the thin Al foil (4. 3 mg/cmz).
The 1 1/2" by 2 mm NaI (Tl) crystal with a 4 mil Be window mounted
on an RCA 6342 photomultiplier tube was used to evaluate this
absorption.

Spectra with one to seven layers of the Al foil between the
source and the crystal were recorded and analyzed. A least squares
fit of the logarithm of the L x- ray counts under the peak vs. the

number of layers of Al foil was made with the result (Fig. 8)
Log N = 5.18836 - 0. 04948x
where x is the number of layers of Al foil.

A1 1.12 i 0. 011 (one layer of Al foil)
A2 1. 26 :I: 0. 013 (two layers of Al foil)

There is no significant difference between having the Al foil
adjacent to the source or the detector. The results of the experiment
are usable in any Situation where the geometry is Similar to that of
this experiment.

The absorption of the K x- ray due to one layer of Al foil was

found to be negligible.

25

LOG N

5.2 -

5J1

5.0 r

4.9 —

4.8

Figure 8.

Log N vs. Layers of Al Foil

 

26

 

 

’d

3" 4 5

Layers of Aluminum Foil

NICHIGRN STQTE UNIV. LIBRQRIE

IIlll"LIIIIIllllIllIllllIIIIllllllllllllllllllllll"III“HI

293017640065