PhysRev 77 841 2

LETTE RS TO THE E D I TOR
TABLE
I. Temperature
Temperature
Kelvin
300
195
dependence
P/Av
arbitrary
(T d, vp)/I P
556 +10
574 &15
32.1 +0.6
41.0&1.0
are summarized in Table I. In interpreting the data, it is assumed

that the line width is also proportional to the average component
of dipole moment P, of the colliding molecules along the angular
its
momentum axis, ~' As P, varies slightly with temperature,
ratio to the static dipole moment p, is also tabulated.
The experiment clearly shows that Av is inversely proportional
to the temperature and hence cr is inversely proportional to the
impact velocity, confirming Anderson's theory. These results
however should not be extrapolated indiscriminately to all types
of microwave collision broadening. They may be expected to hold
rigorously only for an inverse cube dependence of interaction
exemplified
energy on molecular separation
by the above
and
interactions.
symmetric top dipole-dipole
Experimental
theoretical papers on these and other interactions are now in
preparation.
* The research reported in this document has been made possible through
support and sponsorship extended by the Geophysical Research Directorate
of the Air Force Cambridge Research Laboratories under Contract No.
W19-122-ac-35. It is published for technical information only and does not
represent recommendations or conclusions of the sponsoring agency.
' B. Bleaney and R. P. Penrose, Proc. Phys. Soc. London LX, 540 (1948).
' P.
W. Anderson, Phys. Rev. 76, 647 (1949).
' Henry Margenau, Phys. Rev. 76, 121 (1949).
4 B. Bleaney and R. P. Penrose, Proc. Phys. Soc London LX, 83 (1948);
Robert L. Carter and William V. Smith, Phys. Rev. 73, 1053 (1948);
Robert Karplus, Phys. Rev. 73, 1120 (1948); Robert Karplus and Julian
Schwinger, Phys. Rev. 73, 1020 (1948).
Recoil Electron Spectrum of the

R. W.
PRINGLE,
Physics Department,
'HE
S.
STANDIL, AND
K" Gamma-Ray
K. I
RKC
26
(Tave )/(yP)
1&0.02
0.66 &0.02
0.54
0.516
of line widths.
ROULSTON
of Manitoba, Winnipeg, Canada

January 26, 1950
University
observation that the ratio of positron emission to

electron capture in the decay of K' is X~+jP '&0.005, leads
K40 mass difference must be less
to the conclusion that the A4p
than 1.6 Mev, if expressions for either third or fourth forbidden
beta-transitions and axial vector or tensor interactions are used. '
As considerations of the K40 P energy end point2 seem to indicate
that the gamma-ray from K'p be placed on the A4p side of the
decay scheme, it is remarkable that the energy which has been
observed' for the gamma-ray of 1.55~0.05 Mev should be so
K40 mass difference. It was
close to the maximum possible A40
therefore thought of interest to measure the energy of the gammaray more accurately, using the new scintillation spectrometer
which we have developed. 4
The crystal element of activated NaI, 1")&1"X1",was surrounded by approximately 90 g of KF, and the recoil electron
spectrum which was obtained after filtering of the radiation by
4 mm of Al is shown in Fig. 1. The shape of the spectrum was
later confirmed using another source of KCl. The pulse height
distribution for the background is given and also the recoil
electron spectrum of an uncollimated beam of Co" gamma-rays
for purposes of comparison and calibration. The Cosp scale is
greatly reduced and ordinates refer only to the K4p and background
spectra. In these experiments the spectrometer was enclosed in a
three-inch wall lead castle. The photo-cell noise becomes apparent
only below 20 kev on the pulse scale and the rise in the recoil
electron spectra at low energies is attributed to degenerate
radiation entering the crystal, as well as to the loss of recoil
electrons from the surfaces and possible low energy gamma-rays
in the case of K4P~Ca~. Any differences in the form of the two
spectra are due, in addition, to the nature of the extended source
in the case of K4p, and to the existence of two Co~ gamma-rays.
It would not be possible from these results to draw any conclusions
84i
X
l5
60
40
30
20
PULSE
HEIGHT
IH
60
50
VOLTS
k )G. 1. Differential pulse height distributions obtained with scintillation

spectrometer. The curves for K4' and Co60 have been corrected for background.
regarding a possible complexity of the K40 gamma-radiation.

Careful consideration has been given to the manner of approach
to the axis of the two spectra near the end point (Fig. 1) and, due
account being taken of resolving power, the form of the Compton
distribution, etc. , a value has been estimated for the gamma-ray
energy for K40 of 1.47~0.03 Mev in terms of the 1.33 Mev Coss
gamma-ray. This means that the energy available for electron
capture to the excited state of A' is less than 0.13~0.03 Mev.
Coincidence experiments are being carried out to elucidate further
the decay scheme.
The scintillation spectrometer has been used to measure the
possible gamma-activity of certain members of the isobaric pairs
of nuclei of neighboring Z. In some cases (notably cerium) activity
has been observed but measurements with the spectrometer have
indicated these activities to be due in the main to minute traces
of thorium. Attempts are being made to have the purest possible
materials prepared for this work.
' E. L. Fireman,
D. E. Alburger,
~
Phys. Rev. 75, 1447 (1949)

Phys. Rev. 75, 1442 (1949).
~
E. Gleditsch and T. Graf, Phys. Rev. 72, 640 (1947).

4 R. W. Pringle,
K. I. Roulston, and H. W. Taylor, Rev. Sci. Inst.
3
(in press l.
The Cation Distribution in Ferrites with

Spinel Structure
FRANK
G.
BRocKMAN
Phitips Laboratories, Inc. , lrvington-on-Hudson,

January 23, 1950
York
ROM studies of the cation distribution in spinels it was

concluded by Verwey and co-workers' that, for the ferrites
v ith this structure, zinc ferrite and cadmium ferrite are normal
spinels while nickel-, copper-, magnesium-, ferrous- and manganese-ferrite are inverse spinels. 2 In the normal spinel the divalent
ions (zinc or cadmium) occupy the tetrahedral position while the
trivalent iron occupies the octahedral position. In the inverse
spinel the tetrahedral position is occupied by one-half of the
trivalent iron ions while the octahedral positions are occupied by
a random distribution of the remainder of the iron ions together
with the divalent cations (e.g. , Cu~). In addition, it was indicated
that a solid solution of a normal-spinel ferrite with an inversespinel ferrite (for instance, zinc ferrite in solid solution with
copper ferrite) had a structure in which, in the example taken,
the tetrahedral positions were occupied preferentially by Zn~,
the remainder of these positions being 611ed by Fe+'; the octahedral positions were occupied by the remaining Fe+3 and the
Cu~.
Noel, ' however, from a consideration of the magnetic properties

of the ferrites concludes that the cation distribution is not as
"ordered" as in the description given. The distribution according
to Noel may vary so that, for instance, the distribution in copper
ferrite is not simply Fe LFeCu)04 but may be Fep. saCup. is
LETTERS Ta THE ED ITO
in the near future and will describe in more detail the phenomena
and their interpretation.
6OO
E
'Verwey, Haayman, and Heilmann, Philips Tech. Rev. 9, 185 (1947);

Verwey and Heilmann, J. Chem. Phys. 15, 174 (1947).
i Barth and Posnjak, Zeits. f. Krist. 82, 325 (1932).
i L. Neel, Ann. de Physique (12) 3, 137 (1948).
4 J. L. Snoek, Nne
Materials (Elsevier
DeveLopments in Ferromagnetic
Publishing Company, Inc. , New York, 1947).
600
CL
Q.
IZ
200-
0
)0
PO
40
50
60
70
Note on the CoefBcient of Eddy Viscosity in

Isotroyic Turbulence
80
90
DK6QEfs C
Frc. 1. EHect of quenching of copper-zinc ferrite on permeability vs.

temperature curves. Curve 1. Quenched in air from 550'C. Curve 2.
Quenched in air from 600'C. Curve 3. Quenched in air from 650'C. Curve 4.
Slow-cooled from 650'C to room temperature. Curve 5. Quenched in air
from 500'C. Curve 6. Quenched in air from 750'C.
T. D.
LEE
of Chicago, Williams Bay, Wisconsin

February 2, 1950
University
Yerkes Observatory,
T has been suggested by Heisenberg' that the inertial transfer

-of energy by eddies with wave number less than a particular k
can be pictured in terms of an eddy viscosity, vtdue to the
-
I Fej..jsCuo. s~104 for slowly cooled copper ferrite and even approximate the most random distribution, Fe2t3Cu~t3 LFe4~3Cug/3|04
when quenched, i.e. , when neither Fe+' nor Cu~ is located
preferentially in tetrahedral or octahedral positions.
This encourages us to report some experiments which were
performed prior to Noel's publication and which led to similar
conclusions as regards both single ferrites and mixed ferrites.
The experiments were conducted in order to establish whether
or not any kind of ordering occurred in ferrites. In mixed ferrites,
in particular, it is conceivable that several different kinds of
ordering could occur. Accordingly, test pieces of a copper-zincferrite (40 mole percent copper ferrite and 60 mole percent zinc
ferrite) were prepared and these were soaked for periods of eight
or more hours at elevated temperatures and then quenched quickly
in air to room temperature. The initial permeability of the sample
so treated was measured as a function of temperature from room
temperature to above the ferromagnetic Curie point. The results
for different quenching temperatures are shown in Fig. 1. The
numbers for each curve indicate the order in which the experiment
was performed; that is, the sample was quenched from 650'C as
the third treatment to which it was subjected and then for the
fourth treatment it was heated to 650'C and slowly cooled to
room temperature. After this it was subjected to the 6fth and
then the sixth treatment. It will be noted that the phenomenon
is a reversible one.
It has been demonstrated by Snoek4 that the Curie temperature
of a mixed crystal of a ferromagnetic ferrite with zinc ferrite
decreases with increasing zinc ferrite concentration. This fact,
together with the results given, suggested that quenching from
high temperatures brought about a condition which was equivalent
to a reduction of the zinc cation concentration. Since, as described,
the zinc ion was assumed to be in the tetrahedral position, it
appeared reasonable that the phenomenon was to be attributed
to a migration of zinc ions to the octahedral position with replacement of these ions by ferric ions. This is consistent with the fact
that the higher the quenching temperature the higher the Curie
point became, because it is to be expected that the most random
arrangement will occur at the highest temperature.
That no separation of phases occurred in this heat treating was
demonstrated by x-ray diffraction, and observation on the weight
of the test piece after each heating showed that no loss of material
was associated with the treatment.
In discussing these results and this interpretation with Professor
G. K. Uhlenbeck and Dr. 0. S. Duffendack they pointed out that,
if the interpretation is correct, then it should be possible to render
zinc ferrite and cadmium ferrite ferromagnetic by suitable heat
treatment. Such experiments were carried out and it was found
that if zinc ferrite is quenched from 1400'C it becomes slightly
ferromagnetic as indicated by its attraction to a permanent
magnet. Cadmium ferrite appears to be somewhat more magnetic
under the same treatment.
More extensive work along this line has been carried out in
the Kindhoven, Holland, laboratories of the Philips Company. A
communication by E. %'. Gorter of that laboratory will appear
eddies with wave numbers greater than k; he writes
vs=sf
Lt(k)]&k
(1)
&dk
where F(k) denotes the spectrum of turbulence and ~ is a numerical

constant From a comparison with the experimental results on
decay of turbulence, Heisenberg obtained for tc the value 0.85.
This determination depends chieQy on the region of the spectrum
where the ordinary kinematic viscosity v plays no role.
An alternative determination of k(: that depends sensitively on
the behavior of F(k} in the region of large k can be made by
results on vorticity by Batchelor and
using the experimental
Townsend. 2 They found that
~
((Bv,/)'}A/{(Bv,/}')A& = 0.39
S=
(2)
where v is the x component of the velocity of turbulent motion.

Now, from the Karman-Howarth equation3 the vorticity equation
can be written as
id
2dt
{{~&/)
)Av
7
-v{{&'~'.
=-{(./)'}A'
/')'}A. =7
In terms of the spectrum of turbulence, we have4

((Bv, /Bx)')A~=(2/15)1
F(k)k'dk
as
and (3) can be rewritten
(( /)'}A= --v
35
03
F(k) k'dk
2 ~OF,
k'dk.
35
Bt
{5)
Since S remains constant during the decay of turbulence, we can

use the equilibrium spectrum for stationary turbulence derived
from Heisenberg's theory, namelys
ko ~t3
k, 4t3
Li+(k/ke)4j
3'
3 t~
(Foko~ )~
(6)
Using (4), {5), (6) and neglecting the second member of the right
side of (5), we can evaluate S. Thus
9 15&
X't3dX/{1+X') "3]
~dd
-z = 1.a2v. .
i(1+ ')"'j)
From (2), we obtain for ~ the value 0.26. This differs from HeisenHowever, since the principal
kt = 0.85.
berg s determination
contribution to S comes from the transition region where F(k)
~)'
to the k law, the difference must be
changes from the k
traced to the inadequacy of the expression (1) for vz for all k and
can perhaps be interpreted in the following way: For a given k
it is to be expected that only those eddies with wave numbers
appreciably greater than k say ak(a&1) can effectively contribute to the eddy viscosity. We may expect a to be a function
of k but it must always be greater than one. If one neglects
completely the contribution by those eddies with wave numbers

PhysRev 77 841 2

Uploaded by

Copyright:

Available Formats

PhysRev 77 841 2

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PhysRev 77 841 2

Uploaded by

Copyright:

Available Formats

LETTE RS TO THE E D I TOR

are summarized in Table I. In interpreting the data, it is assumed

Recoil Electron Spectrum of the

of Manitoba, Winnipeg, Canada

observation that the ratio of positron emission to

k )G. 1. Differential pulse height distributions obtained with scintillation

regarding a possible complexity of the K40 gamma-radiation.

Phys. Rev. 75, 1447 (1949)

E. Gleditsch and T. Graf, Phys. Rev. 72, 640 (1947).

The Cation Distribution in Ferrites with

Phitips Laboratories, Inc. , lrvington-on-Hudson,

ROM studies of the cation distribution in spinels it was

Noel, ' however, from a consideration of the magnetic properties

LETTERS Ta THE ED ITO

'Verwey, Haayman, and Heilmann, Philips Tech. Rev. 9, 185 (1947);

Note on the CoefBcient of Eddy Viscosity in

Frc. 1. EHect of quenching of copper-zinc ferrite on permeability vs.

of Chicago, Williams Bay, Wisconsin

T has been suggested by Heisenberg' that the inertial transfer

eddies with wave numbers greater than k; he writes

where F(k) denotes the spectrum of turbulence and ~ is a numerical

where v is the x component of the velocity of turbulent motion.

In terms of the spectrum of turbulence, we have4

and (3) can be rewritten

Since S remains constant during the decay of turbulence, we can

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.