A METHOD FOR THE SEPARATION OF GASEOUS ISOTOPE MIXTURES AND ITS APPLICATION TO THE ISOTOPES ON NEON

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP82-00039R000200050028-5
• Release Decision: RIPPUB
• Original Classification: R
• Document Page Count: 18
• Document Creation Date: December 22, 2016
• Document Release Date: April 20, 2012
• Sequence Number: 28
• Publication Date: June 17, 1952
• Content Type: REPORT

Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
A..Nethod for the Separation of Gaseous Isotope
t es and Its A lication to the Isoto es on Neon
by G. Hertz
Journal ffr Physik, 1932, pp 10$-]21; 700
(Berlin)
Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
STAT
 With lL Illustrations (1teceived on 30 August 1932)
The process utilizes diffusion through a porous wall into a
vacuum. This makes possible the simultaneous action of a great
number of diffusion cells and therewith an extensive separation of
an isotope mixture; in a sin ;1e operation. The coristruc'ted a.ppa.rstu.s,
consisting of 2L pumps and L8 clay pipes, yields a change in the
isotope ratio by almost a fac'bor of 8 in single application on the
isotopes of neon w Lthi n a few hours . Through repo ate d application
a more extensive seiaaratiof is possi.iale. Iv1.ass-spoctra and optical
:;ectra of nE;Un^15Qtgpe..rrT1.x'~"asps of different composition are re-
produced.
1rlic method subsequently described for the separation of
isotope mixtures utilizes diffusion through a porous wall into a
vacuum, as it was .first used for this purpose by Aston, and since
then in various ways by Harkins and his collaborators. (F. W.
Aston, Phil, Nag. 39, L1i~9, 1920; W. D. Har'dns and R. E. Hall,
Journ.A!ner, Chem. Soc. 38, 53, 1916; W. D. Harkins and C. E.
i3roeker, Nature 10, 230, 1920; J. N. Eronsted and G. v. Hevesy,
z3. f, phys. Chem, 99, 189, 1921; W. D. Harkins and A. Hayes,
Journ. Amer. Chem. Soc. L3, 1803, 1921; lt. S. Mulliken, same
1592, 1923; L D. Harkins and F. A. Jenkins, same 1.~8, 8, 1926;
D. I Harkins and D. Ilortimer, Phil. !'flag. 6, 601, 1928. )
A METHOD FOR THE SEPA1 ATION OF GASEOUS ISO'r'O?E
,~....,.....~,...,~.,,~.. ,.,
MIXTURES AND TTS APT'EICAT'ION TO THE ISOTOPES JF NEON
by G. Hertz, Berlin
1
Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5 q
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
Since the diffusion velocities for different uses behave themselves
as the roots from the molecular weights, a single diffusion, process
the case of isotopes wi11 yield only a very slight change in them
in
concentration ratio. Thus, an actual separation of the isotopes
has not been attained up to now, but only a relatively small change
of the atomic weight, which in the experiments of Harkins in the
case of chlorine attained the amount of O.1
In order to attain an extensive separation in a short period
of time, it was necessary to utilize a large number of diffusion
cells simultaneously. This was achieved by constructing the
apparatus from a series of separating units, of which each possesses
the property to decompose a gas mixture, introduced in a steady
stream, into two equal component currents of different composition.
Such a separating unit may in the simplest case consist of a pipe
with a porous wall (subsequently denoted as clay pipe) as repre-
sented in Figure 1. The gas mixture consisting of two components
enters at A. One part diffuses through the wall of the pipe and
is sucked off at B by a pump which maintains the vacuum in the
space surrounding the clay pipe. The remainder of the gas mixture
leaves at D* If care is taken to see that exactly one half of the
gas entering at A is sucked off at B, the entering gas stream will
be decomposed into two component currents, of which the one leaving
at B contains the lighter fraction, and the one at D the heavier
fraction. A disadvantage of this simple arrangement is the fact
that the mixture on the inside of the clay pipe already is so
strongly enriched with the heavy fraction, that the mixture diffus-
ing through the wall at the left end of the pipe has almost the same
Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
the clay pipe S and leaves as the second component current at ~+
The gas which diffuses ffuses through clay pipe S is sucked o.Ff at C by
A where it once again enters the current.
pump P and returned to
Such r ~epa ; ratin r units may be connected in parallel in any
~
flows, for further concentration with the heavier f'racta.an~ t h rong
b ~ r exe~ore, a sepaxat:~ng un~.t was useda
composa.t~,on as the original. 1~ ~ted in Figure 2, Here the inix~~uxe Flows
as schematically represented
cla ~.pes . Only the gas which diffuses thevely wall of through the f z two y p
'xst clay pipe is sucked off at ~ as the
lighter fraction comp anent current. The rema.ining gas ~.xture
h
desirable number.
units the two component currents, into which
to the two add acent
This is done in such a way that each separating
, ~ th -i.ts adjacent units, and gives off again
unit receives gas from ao
Figure 3 shoals the diagram of an
apparatus onsi s tang of four separating units. Separating Unit
3 Separating Unit
2 Separating Unit
1 Separating Unit
It also shows in 110 of the actual size a picture of the arrange
meat and the dimensions of the separating units as used in the
Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
se s,xatng un~.~Gs ? The ~,ndiv~.dua1
apparatus cans~.sta,ng off' 2~ p ,
saparating units are marked off 'ram each other by dotted 1:>.nes.
The setters used in the third unit correspond to those of 2.
is
thid unit
the clay pipe R of ther, The gas entex:~ng at A into
fraction, which is sucked aff at B by
decomposed into a light r
ff
through pipe S which is sucked o
pump P~, in.ty a part da.ff usa.ng
at C and returned by pup m, P3 to A, and a.nty a heavier fraction
the second unit at the left. regulatory
Wh1G11 HAWS of at D to
installations are uirnecessarY since under the action of the pumps
tself so, that each of these component
the current rc ul ate s i
e~third of the gas current entering into
currents 1s almost on
' a s tationaxv state the same quantity of
pipe R at A. Since in
arating unit to rhe next, just as in the
gas must flout from one Sep
he as quantity at A, originating from the
reversed direction, t g
unit, and which flows to the third unit, must
fourth separating be equal to the quantity sucked off at B; Since the pressure,
isoto)e mixtures also the composition, change
and in the case of Z
very little separating unit to the next, the quantity
from one
to the second unit, must also be very nearly
flowing off at B
equal to the c;uan~ t= t sucked off at B. But this quantitY, be-
cause of the equal y clay pipe lengths, is again equal to the
~
qL1a11ti tY sucked off at C and returned to A. Thus each of the
three component currents receives 1/3 of the gas entering into
pipe R at A.
Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
:Both ends of the apparatus are schemat:.caliy represented
Since no further adjacent units are available to
in Figure 3.
the end units, the gas which actually should be given off to the
next adjacent, is here returned to the end unit through a storage
~
container. This makes possible the formation of a stationary
which is essential for the process. The clay pipe to the
state
left of the first separating unit does not act as a separating
cell but only causes the gas to circulate through the end cones
a
tamer with the proper velocity.
We may y completely neglect further details concernhng the
operation of the apparatus, and consider only the fact that each
separating unit decomposes the gas which it receives into two
components, of which the one is the lighter and the other the
heavier fraction of the mixture. As will be shown later, these
two constituents differ in the concentration ratio of their com-
constant factor q, which for example has a value of
ponents by a
1.092, in the case of neon. In Figures L and ~, the individual
separating units are schematically represented by means of squares.
The arrows indicate the direction of the current. Each separating
member receives gas from both sides, and gives off to the two
adjacent units the separated constituents. At the ends, the gas
flows through the storage containers. Two cases are readily
visualized namely the initial state when the apparatus is started,
and the stationary end state. In the initial state (Figure L),
the same mixture is present everywhere. As a result thereof, the
same processes take place in all the separating units. Each
separating unit gives off lighter mixture to its right adjacent
Declassified in Part - Sanitized Coy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
unit and heavier to its left. In the figure the lighter mixture
is marked by a dotted line and the heavier by a soiid line, One
recognizes that through the entire apparatus two currents flow
total current now consists of a series of adjacent circular
towards each other, of which the one predominantly moves the
lighter constituent toward the right, and the other predominantly
the heavier one toward the left. This state is disturbed due to
the change in concentrations in the end containers. Intermediate
states, which are difficult to overlook, result in a stationary
end state in an asymptotic fashion. In this stationary state the
transport from separating unit to separating unit for each indi
vidual constituent of the mixture must be equal to zero, that is9
the same mixture must flow from one separating unit to the next as
in the reversed direction. Since the condition remains that the
two component currents which leave one separating unit differ by
a certain factor q in their concentration ratio, the state repre-
sented in Figure S now results. In this figure the concentration
ratio is indicated by the relationship of the dotted line and inter-
mediate space between the lines representing the current. The
currents, and the concentration ratio changes from circular
differ by the factor qm where in is the number of separation units,
stationary end state, gas mixtures, whose concentration ratios
current to circular current by the factor q. Therefore, in. the
are found in the two end containers Vs and V.
The proper selection of clay pipes was essential for the
porosity must fit the output of the pumps that were used.
practical performance of bhe process. Their surface area and
Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
fle?t(~r must not be too great to avoid
~rthrmore, their a.nsa.de da.aa
on the inside of the pipe, and not too
a radial concentration drop
h
pressure drop as a result of the current ?thxoug
s-nal~. s o that the pre -~
eat. Pipes made from 'tQ-~ mass's by the
the pipe will not be too gr .
A..G, proved to be extremely well suited.
S?tea~,:~.~-I'~agnesa.a
massIt is a kaolin-rich, fine fireproof clay. (I express
~-~~_5
the Steatitwl~agnesia A. -G for the use of
rr>y sincere thanks to
the required pipes.) The pipes used were 30 centimeters long,
of millimeters and a wall thickness of
had an inside disaneter
1 millimeter. The mass is very f'lne-porous and has the very
that the pipes may be sealed into ordinary glass
great advantage
ith the help of sealing glass. Mercury vapor ejectors Model I
w
y Hanf f and ]duest were used as pumps. These pumps are mounted
b
on a wooden frame. The diffusion apparatus is connected to the
pumps by pa. ' vein-cemented ground sections and is held by the punps
in the manner depicted in Figure 3. Suitable cock connections
enable the insertion of selective containers of
at ?the ends
different size into to end circles for Vs and V1. In addition,
for evacuating the apparatus and to draw
prova.sa.ons are made
off the obtained gas 'into containers with the help of a vapor
ejector and a rl'opler pump.
After preliminary experiments with neon-helium mixtures
in an apparatus of four separating-units were successful, the
bus with 2L separating units was set up in two parallel
appara
of 12 units each, so that the two ends are close together.
raves
The pumps are heated with gas. The water coolings of six pumps
each are connected in series. A safety device, as described
Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
recently by w. pp, disconnects gas and water independently as
soon as something goes wrong with the cooling water supply or if
the pressure apparatus rases above 20 millimeters Hg. (W?
in the ~
Pu.pp, Phys. GS. 33, X30, 1932.) The pressure in the apparatus
normally is 10 millimeters Hg. in the end container Vl. At
higher pressure the pump velocity of individual pumps is no
suff'icienb. As a result of the pressure drop in the pipes,
longer
a pressure difference of 2.5 millimeters Hg. exists between V1
and Vs
Experiments with neon were made to test the apparatus.
Normal neon-helium mixture was used as the original material.
eon-helilun mixture was provided for me by the Linde -Corporation
(N
and b the Griesheim-Auta;enMSales Corporation for which T express
sincere thanks.) Since in the stationary state a definite ratio
my
of the isotope relationships iii the two end containers occurs, a
mixture will be found in V1, with equal size of the containers,
which is filled with the liglter isotope in comparison to the
original gas, and in Vs one which filled with the heavier one.
If one wishes to obtain a possible large change of the isotope
ratio in comparison bo the original one, then either Vs or V1
must be chosen possibly large and the other container small. The
composition will then change only little in the larger container,
and a change of the isotope ratio by almost the entire separation
factor of to apparatus will be obtained in the smaller one. In
cases where the one isotope is present in only extremely small
such as with oxygen or even to a greater degree with
quantities,
hydrogen, the normal composition in the one end container must be
.
Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
maintained by Pen:~tta. 'ng fresh gas from the outside to circulate
r~
continuously through the end volume,
erimen'ts with neon a container composed
wring the first exp
f sa.x glass spheres with a capacity of 30 liters was used for Vs,
.
o
similar one of S liters, After the apparatus was
and for V1 a
-helium mixture, pure helium first collected in
filled with neon
Vl. After the greatest part of the helium was removed in this
manner, the 30 liter container was inserted as Vl, and a volume
ters3 used as Vs? After the apparatus was
of about x.00 centime
in operation in this connection for 8 hours, the contents of Vs
pumped or for examination, (Subsequent experiments showed
1vEc?
that the end state with this size of Vs is already attained
after hours). In order to obtain a fraction enriched with
~.
the lighter isotope, the two containers were exchanged. After
the heliwa which was stall collecting in Vl was pumped off, the
apparatus was put into operation for 8 hours again, and then the
content of Vi removed,
A mass spectrograph based on the Thomson parabola method
to examine the composition of the fractions obtained by
was used
the above method. (The mass spectrograph was constructed by
Lukanow as part of his diploma fulfillment,)
the mass spectra of the fractions obtained by a
Figure 6 shows
single separation in the manner described above, 'On the spectriun
fraction, the isotope 22 is recognized with extreme
of the lighter
difficulty, while the one of the heavier fraction shows both isotopes in ap~roximately equal intensity. For comparison,
p
Declassified in Part- Sanitized Cop Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
are reproduced of the spectra of different fractions, taken with
Figure 7 shows the mass spectrum of normal neon.
Figure 7: Normal Neon
The mass spectra are of little value for a quantitative
determination of the isotope ratio, because of the unfavorable
form of their characteristic film curve. In the case of neon
it is more convenient to observe the optical spectrum, in which
the isotopic fine structure was found by G. Hansen. (G. Hansen?.
Naturwissenschaften 15, 163, 1927.) In Figures 8 to 11 plates
the help of a Perot.Fabry~,standard instrument at 4L. millimeters
plate distance (61113 to 6402 Angstroms). (I an obliged to Dr.
F. Houtermuns for the taking of these spectra.)
Figure 8: Normal I"de on
10 one of a mixture which contains the two isotopes in exactly
in which the isotope 22 is no longer optically detectable; Figure
of the heavier lines. Figure 9 shows the spectrum of a fraction
cator, which is identifiable on the photograph only in the case
Figure 8 shows the spectrum of norrnal neon. The isotopic
fine structure is noticeable by means of a weak short waved indi~
same plate. The two last fractions were obtained by using the
densities with those of an intensity scale photographed on the
Ne20: Ne22~ 1:2.5. This ratio is estimated by comparison of the
equal quantity; Figure 11 one of a mixture with the isotope ratio
Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
process twice, since only about the ratio 10;4 ..s obtained if the
process is used once.
to follow the periodic course of the separation
In order
process, suitable spectral tubes were corrn.ected to the two end
containers for the observation of the fine structure, (An inter"
late generously made available by the 1lalle Compm y was
erence p u
T
serving the fine structure.) By use of a volume of
used for ob .~
ters3 for Vs, the equilibriUnl state was attained after
~,UQ cent~,me
Since the gas pressure in Vs is a maximum of 7.; mi.11i
L hours, meters Hg, a quantity of ) centimeters3 is obtained from atmos-
pheric pressure during this time.
A parabola, corresponding to the mass 23, is always
present on the mass spectra, of the heavy fractions, in addition
to the isotope 21, when sufficient lighting is available. Ex-
periments with further concentration will have to decide whether
another neon isotope is involved.
It is intended to increase the apparatus in order to
increase the degree of separation with a single application of
the process to such an extent, that the separation or concen-
tration of the isotopes of hydrogen, oxygen (in form of water
vapor), nitrogen (NH3), and chlorine (HC1) may be undertaken
with the possibilitY of success. It should also be investigated
whether a change in the structure of the apparatus would also
increase its efficiency. To get a picture of the quantitative
relationships, a single separating unit will be considered as
represented in Figure 12. The glass pipe between the two clay
Declassified in Part - Sanitized Co
py Approved for Release 2012J04/20 :CIA-RDP82-000398000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
pipes R and S has been omitted, so that the two clay pipes are com-
bined into one, of which however the parts R and S are located in
separate, vacuucrl spaces. In contrast to the existing installation,
the diagram is generalized insofar as pipes h and S are no longer
of equal length. The letters in the diagram correspond to those
in Figures 2 and 3. Here also the gas quantity flowing from B
to the right separating unit per unit of time, will be equal to
the one flowing from 1) to the left unit, as a result of the
linkage with the neighbor units. while, however, in the existing
arrangement the gas entering at A left, at B and D in the quantity
of 1/3 at each point, here it will be a different fraction which
we will call f. As is readily seen, f depends only on the ratio
of the lengths of the two clay pipes. Thus 1 = is + 2, where lr
f it
and is are the lengths of the two pipes.
In order to estimate the maximum attainable degree of
separation and the velocity of the separation, we will calculate
the composition of the component currents which leave at B and D,
with a given composition of the gas entering at A. To simplify
the calculation, we limit ourselves to the special case of an
is present in only a small quantity. From the lighter isotope,
isotope mixture of only two components, of which the heavier one
no molecules will enter into the clay pipe at A in the unit time,
and from the heavier one Vo c,, no, where the concentration cC
0
molecules of the heavier components will pass point E and enter
off through the wall of R, in the unit time n,.nl no (1 - f)
is assumed as being small near 1. Since the fraction f is pumped
Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
pipe S. Of these n2 no 'f will reach the left end of S and will
leeve at I). Since the heavier molecules will diffuse slower through
the wall, corresponding to the ratio of the roots of the molecules
weights, the concentration c within the clay pipe will increase
from right to left. If we assume a host of molecules of the
mixture which enter at A at the time f = 0, the number of molecules
belonging to this host for each of the two components, will de-
crease exponentially during passage through the clay pipe, since
the number of molecules which leave in the unit time by diffusion
through the pipe wall is proportional to the number present. II'
the number of the lighter molecules decreases according to the law
a
then
c o A
will be valid for the heavier ones, where
nil
.., .~.. and m for
/?Y
ms are the molecular weights of the lighter or heavier component.
For concentration c this will yield
\-'
(I n
L. - f Al
rect ion of the pipe, we may substitute N/Ido by n/no. The con-
if we neglect the diffusion in the gas in the longitudinal
centrations;
c,=~ (/-/)'j
result for points F and D, and therefore for the number of molecules
of the heavier component flowing past in the unit time.
The number of molecules of the heavy isotope which are
sucked off at B in the unit time is
Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
p/A ~m
\ (1_c)M
The number of those leaving at D iii the unit time is:
Since equal quantities of the lighter molecules leave at D and D
.
under the assump p t 't~.on C< wich was made, we obtain for the
ratio of the concentration of the heavier component lxi he two
component currents, which leave the separating unit under con'
sideration, the value:
the total apparatus has m separating members, then
If
f
there will result for the separating factor, that is or the
ratio of the concentrations in the two end containers in the
stationary state, the value:
elds Q (1.092) 2L~ ~?~!? The actual attained separation
In the apparatus used up to now, f = 1/3, u = 0.953 for
the neon isotopes, and m = 2L. Substitubaon of these values
,
degree has not yet been measured exactly. In any case, it
be more than 20 percent less than the calculated value.
well not
In addition to the separation factor Q, the magnitude :
I ( ) I C h
a i
-0
as of significance for the practical application of the process,
Declassified in Part- Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
since it indicate the vaiocIty wLth which at the start the heavy
isotope is transported through the apparatus. It is proportional
to the concentration, the intensity of the gas current, and the
magnitude
which may be denoted as the relative separation velocity.
In Figure 13, the magnitudes Q and G are represented as
a function of the magnitude f, which is depended on the length
ratio of the pipes, for an apparatus of 2Li. similar separating
units and for the case using neon isotopes. The small crosses
denote the points f' = 1/3 corresponding to the apparatus used
up to now. It is seen that by decreasing f, that is by short-
ening clay pipe 1Z and correspondingly lengthening 5, a subs
stantial increase of the separation factor could oee attained
at the expense of the separating velocity. The increase in
the separation factor, which is possible in this manner, is
limited in practice by the fact that with a decreasing value
of f the current velocity in the left end of S will decrease.
At a certain lindt, the assumption that the diffusion in the
gas in the longitudinal direction of the pipe might be neglected,
which was made in the calculation, no longer will hold. Below
this limit, in the case being considered, the separation caused
by the diffusion through the clay pipe will be more or less
cancelled again as a result of the diffusion of the heavier
isotope compared to the current of the lighter one. Where this
Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
limit is located is to be determined by experiments. It will
also be attemped to lessen this, influence of the longitudinal
diffusion by giving the left end of S a smaller diameter.
Of special interest is the liiniting value f w 0. It
leads to an arrangement as represented in Figure i1~ for the case
of three separating units. Since here the velocity on the left
side of each separating unit is zero, the diffusion in comparison
to the gas current plays the decisive role. The diffusion limits
the separation of the mixture, which during its passage through
the clay pipe loses continually more of the lighter than of the
heavier part. On the other hand, the diffusion also causes the
separation velocity not to become zero, as it WOU1CI be according
to equation (3), since now a gas exchange between neighboring
separating units results by diffusion. A preliminary experiment
with such an apparatus, consisting of 6 separating units, has
already shown that it is at least equivalent to the former one
with respect to the degree o? separation. The separation velocity,
on the other hared, will be smaller. Despite this, ..t is not im-
possible that this arrangement may be used advantageously in
certain cases 'because of its simplicity and its smaller volume.
Regarding the velocity with which the separation occurs
in an apparatus consisting of a larger number of separating
units, the following should be noted, According to equation
(2), the velocity with which a component is transported is pro-
portional to its concentration. As a result thereof, bhe
periocac progress of the process in the entire separation apparatus
will 'be determined by the location of the least concentration, thus
Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
 Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5
aratir1; ur~it. For the last separating units one
by the last sep
would thus select such, which possess a relatively large separat-
a.ng velocity ty with a moderate separation factor. For those
separating units in which in a stationary state, higher concen~
tzata.ons already exist, such should be selected whose separation
factor is increased at the expense of the separating velocity.
The execution of the experiments was made possible with
the Aid Society of German Science, for which I
funds granted by
w express my gratitude. W. Schu.tze, professional engineer,
~.s h t o provided me with valuable assistance during the construction and
operation of the apparatus.
Berlin, Physical Institute of the `technical College,
CORRECTION
To the paper y G. Hertz: A Nethod For The Separation Of
Isotope I'dxtures and its Application to the Isotopes of Neon.
~
(ZS. f. Phys. 79, 108-121, 1932-)
1. On pages 11LG and 11 two photographs have been inter-
changed, so that the figures do not correspond to the text and to
the captions. rt'he right photograph of Figure 6 actually shows the
mass spectrum of formal neon. The mass spectrum printed as Figure
7 is that of a mixture obtained as a heavier fraction by a single
application of the separation process, and which contains the two
isotopes in approximately equal quantity.
2. On page 118 the last word of the 19th line, namely,
ttheavier'I, is to be replaced
n.li ghtert'.
Declassified in Part - Sanitized Copy Approved for Release 2012/04/20 : CIA-RDP82-00039R000200050028-5