NAS-NS-3025.txt

 

, " Wy R e Ty Wy
. !
e doa k.

 

it

i

National

Academy
of

Sciences | -.__

 

‘ -;lational Research Council

NUCLEAR SCIENCE SERIES

The Radiochemistry
of the Rare Gases

_— - '
o ——

— ——

T
Atomic

il

iifl

IR RY
Commission

I

L

 
COMMITTEE ON NUCLEAR SCIENCE

L. F. CURTIBSS, Chairman
National Bureau of Standards

ROBLEY D. EVANB, Vice Chairman
Massachusetts Institute of Technology

Jd. A. DaJUREN, Secrelary
Wenstinghouse Eleotric Corporation

C. J. BORKOWBKI
Oak Ridge Natonal Laboratory

ROBERT G. COCHRAN
Texas Agricultural and Mechanical

College

SAMUEL EPSTEIN
California Institute of Technology

U. FANO
Natonal Bureau of Standards

HERBERT GOLDSTEIN
Nuolear Development Corporation of
Amerioca

J. W, IRVINE, JR.
Magsachusetts Institute of Teohnology

E. D. KLEMA
Northwestern University

W. WAYNE MEINKE
University of Michigan

J. J. NICKBEON
Memorial Hospital, New York

ROBERT L. PLATZMAN .
Laboratoire de Chimie Physique

D. M. VAN PATTER
Bartol Research Foundation

LIAISON MEMBERS

PAUL C. AEBERSOLD
Atomic Energy Commission

J. HOWARD MoMITL1LEN
National Science Foundation

CHARLES K. REED
U. 8. Air Force

WILLIAM E. WRIGHT
Office of Naval Research

SUBCOMMITTEE ON RADIOCHEMISTRY

W. WAYNE MEINKE, Chairman
University of Michigan

GREGORY R. CHOPPIN
Florida State University

GEORQGE A. COWAN
Los Alamoa Bclentific Laboratory

ARTHUR W. FAIRHALL
University of Washington

JEROME HUDIS
Brookhaven National Laboratory

EARL HYDE
University of California (Berkeley)

HAROLD KIRBY
Mound Laboratory

GEORGE LEDDICOTTE
Osk Ridge National Laboratory

JULIAN NIELSEN
Hanford Laboratories

ELLIS P. STEINBERG
Argonne National Laboratory

PETER C. STEVENSON
University of California (Livermors)

LEO YAFFE
McGill University

CONSULTANTS
NATHAN BALLOU JAMES DeVOE
Centre d'Etude de 1"Energle Nucleaire University of Michigan
Mol-Donk, Belglum WILLIAM MARLOW

National Bureau of Standards
S
M133r

¢ .|

' The Radiochemistry of the Rare Gfises

By FLOYD F. MOMYER, JR.
Lasorence Radiation Laboratory

Universily of California
Livermore, California

October 1960

LOBS ALAMOS
SCIENTIFIC LABORATORY

APR - 6 1881
3 LIBRARIZS

— PROPERTY

=

=3

—

i=s

i

. Subcommittee on Radiochemistry
National Academy of Bciences —National Research Council

 

Printed in USA. Price $0.75. Available from the Office of Techmical
Ssrvices, Department of Commeroe, Washingtion 26, D. C.
FOREWORD

The Subcommittee on Radlochemlsgtry 1s one of a number of
subcommlttees workilng under the Committee on Nuclear Scilence
wlthin the Natlonal Academy of Sclences - Natlonal Research
Councll. 1Its members represent government, lndustrlial, and
unlverslty laboratorles 1n the areas of nuclear chemistry and
analytical chemistry :

The Subcommittee has concerned itself with those areas of
nuclear sclence whilch involve the chemlst, such as the collec-
tion and distributlon of radlochemlcal procedures, the estab-
lishment of specificatlons for radiochemlcally pure reagents,
avallabillity of cyclotron time for service irradlations, the
place of radiochemlstry in the undergraduate college program,
etc.

This serles of monographs has grown out of the need for
up-to-date compllations of radlochemlcal informetion and pro- -
cedures. The Subcommlttee has endeavored to present a serles
which will be of maximum use to the working sclentist and

which contalns the latest avallable Information. ‘Each mono-
graph collects 1n one volume the pertinent informatlon required
for radlochemical work with an individual element or a group of
closely related elemente.

An expert In the radlochemlstry of the particular element
has written the monograph, following a standard format developed
by the Subcommlttee. The Atomlc ‘Energy Commission has sponsored
the printing of the serles.

The Subcommittee 1ls confldent these publlcatlons willl be
useful not only to the radlochemist but also to the research
worker 1n other flelds such as physics, blochemlstry or mediclne
who wishes to use radlochemical technlques to solve a specifilc
problem.

W. Wayne Melnke, Chalrman
Subcommittee on Radlochemlstry

iii
INTRODUCTION

This report dealing with the radiochemistry of the rare gases was pre-
pared at the request of the Subcormmittee on Radiochemistry of the Committee
on Nuclear Science of the National Research Council a8 one of a series of
monographs on the radiochemistry of all the elements.

The early sections of this monograph are devoted to general reviews of
rare gas properties of interest to the radiochemist and to some general dis-
cussion of separation techniques for rare gases, The last three 6hapters are
respectively a discussion of the removal of rare gases from targets, a dis-
cuseion of techniques used for counting radioactive rare gases, and a collec-
tion of radiochemical prdcedures for rare gases.

The author would appreciate receiving from readers any additional in-
formation, published or unpublished, which should be included in such a report
on the radiochemistry of the rare gases. | _

The author takes this opportunity to acknowledge the able assiatance of
Mr. R. A. daRoza in the preparation of this monograph.

iv
CONTENTS

Foreword
Introduction
I. General References Pertinent to Rare Gas Radiochemistry

II. Table of Isotopes of He, Ne, A, Kr, Xe, and Rn )
III. Review of Features of Interest in Rare Gas Radiochemistry
IV. Sample Solution and Exchange with Carriers
VY. Counting Rare Gas Activities |
VI, Collection of Ra.di-ochemic'al Procedures for the Rare Gases

Procedure 1 — Removal of Kr and Xe from Air and
Their Subsequent Separation

Procedure 1 A — Removal and Separation of Kr and Xe
Fiassion Products from U235 Targets

Procedure 2 — The Extraction, Purification and
Industrial Uses of Krss

" Procedure 3 — Rapid Removal of Fiesion Product

Kr from U Foil

Procedure 4 — Recovery of Fission Product Xe
from U Metal .

Procedure 5 — Rapid Removal of Fission Product Xe

| | from UO, or UO,(NO,), - 6H,O Targets

Procedure 6 — Separation of Long-lived Fission Gases

Procedure 7 — Removal of Rn from Th Targets and its
Collection on the Cathode of a Discharge
Tube

Procedure 8§ — Determination of Active Gas Half-lives
by the Charged Wire Technique (II)

iii

iv

27 .
29

34

34
40
43
46
46
48
48
49

51
The Radiochemistry of the Rare Gases

FLOYD F. MOMYER, JR.
Lawrence Radiation Laboratory
University of California
Livermore, California

GENERAL REFERENCES PERTINENT TO RARE GAS RADIOCHEMISTRY

"H. Remy, Treatise on Inorganic Chemistry, Vol..I (translated by J. S.
Anderson), Elsevier Publishing Co., Few York (1956).

 

S. Dushman, Vacuum Technique, John Wiley and Sons, New York, 1949.

 

R. T. Sanderson, Vacuum Manipulation of Volatile Compounds, John
Wiley and Sons Inc., New York (1948).

 

R. E. Dodd and P. L. Robinson, Experimental Inorganic Chemistry
Chap. 2, Elsevier Publishing Co., New York (1954).

 

S. Brunauer, The Adsorption of Gases and Vapors, Vql. I, "Physical

 

Adsorption'", Princeton Universeity Press, Princeton, N.J. (1942),

A. I, M, Keulemans, Gas Chromatograzhx, Reinhold Publishing Co.,
New York, 1959, 2nd edition.

Lawrence A. Weller, '""The Adsorption of Krypton and Xenon on Activated
Charcoal -- A Bibliography," Mound Laboratory i'eport MLM-1092,
Miamisburg, Ohio, 1959.

A. Guthrie and R. K. 'Wakerling, editors, Vacuum Equipment a.fid Tech-
niques, National Nuclear Energy Series, Div. I, Vol. 1, McGraw-Hill
Book Co., Inc,, New York, 1949.

C. D. Coryell and N. Sugarmé.n, editors, Radiochemical Studies: The

 

Fission Products, National Nuclear Energy Series, Div. IV, Vol, 9,
Booke 2 and 3, McGraw-Hill Book Co., Inc., New York, 1951, Papers
64-70, 139, 141, 145-50, 154, and 311-17,
II. TABLE OF ISOTOPES OF HELIUM, NEON, ARGON
KRYPTON, XENON AND RADON. "

 

 

Isotope Half-life Type of Decay Method of Preparation
He3 Stable (abundance 1,3 X 10_4% atmos.) Natural
(abundance 1.7 X10™°% wells)
He4 Stable (abundance ~1007) Natural
He® ~0. 8 sec B Beg(n,a)
Ne 18 1.6 sec fl+ F_lg(P. 2n)
Nelg ~18 sec fi+ Flg(p,n)
Nezo Stable (abundance 90.92%) Natural
Nez'1 Stable (abundance 0.257%) Natural
Nez'z Stable {abundance 8. 82%) Natural
23 - 22 . 22
Ne ~40 sec B Ne "{n,y); Ne"" (4, p);
23 26
Na""(n,p); Mg~ (n,a)
Ne24 3.38 min g Nezz(t, p)
A35 ~1.8 sec fi+ S3z(u.,n); C135(p,n)
a3 Stable {abundance 0. 337%) Natural
37 . 34 37, ..
A -_35 day EC S™ "(a,n); C1™ (d, 2n};
c*’(p, n); X* (4, a);
Ca40(n. a)
A38 Stable (abundance 0. 063%) Natural
39 - 38 39
A ~265 yr B A7 (n,vy); K™ (n, p)
A0 Stable (abundance 99. 600%) Natural
41 . - 40 40
A ~110 min B A7(d,p); A" (n,¥)s
41
. K" (n,p)
A42 >3.5yr B A41(n,y); parent K42
Kr 0 9.7 hr EC - v8%(p, apall)
77 ; + T4
Kr ~1.2 hr EC~20%, B ~80% Se'“(a,n)
Kr'?8 Stable {(abundance 0, 3549%}) Natural
79 ' + 76 79
Kr 34,5 hr EC 95%, B 5% Se “(a,n); Br (4, 2n);
Br' ’(p,n); Kr ' (d,p);
Kr?B(n.v)
KrBO

Stable {abundance 2,27%)

 

Natural

x
Listed are thoee nuclides having an "A'" or "B'" classification in '"Table of
Isotopes'', D. Strominger, J. M., Hollander and G. T. Seaborg: Revs. Mod.

Phys. 30, No. 2, Part II, April 1958,

the original literature may be found in said article,

Further information and references to
TABLE OF ISOTOPES OF He, Ne, A, Kr, Xe,

and Rn,. {Cont'd)

 

‘Half-life

 

Isotope Type of Decay Method of Preparation

Kr81 ~10 sec IT Brsl(p,n); daughter Rbs-l

Krsl 2, l><105 yr EC Kreo(n,y)

K82 Stable (abundance 11.569%) Natural |

K™ 114 min IT 5e%%a,n); Kr®%(a,p);
Kraz(n.y)§ X-rays on

. Kr83-; fission U, daughter

Br®3; daughter RBS>

Kr83 Stable (abundance 11, 55%) Natural; fission U

Kr84 Stable (abundancé 56. 90%) Natural; fiseion U

Kr3®™ 4 36 hr p 78%, IT 22% se®?(a, n); Kr®4(4, p);
Kr84(n,y); Rb85(n,p);'
Sr88(n, a); fission U,
daughter Br35

I:(r85 10.3 yr B K:r84(n,y); fission U

Kr86 Stable (2bundance 17.37%) Natural; fission U

®Kr®7 78 min | 'y Kr28(a,v); kKe384, p);
Rb87(n, p): fission U

K98 ‘2.8 hr B Fission U, Th

Kr? 3 min 8 Fission U, Pu

Kz’ 33 gec - g Fission U, Pu

Krg1 9.8 sec B Fission U, Pu

Kr % 3.0 sec p Fission U, Pu, Th

K> 2.0 sec B Fiassion U, Pu

K:94 1.4 sec B Fission U

Kr?! l sec B Fission U, Pu

XelZI 40 min [3+ 112.7(p, n)

xel?2 20 hr EC 1127(p, 6n)

Xe123 2 hr EC, 'fi+ 1127(p,5n) .

Xe124 Stable {abundance 0, 096%) ‘Natural

Xe 2™ 55 gec IT (7) 1127(a, 6n)Cs?>; daugh-
ter Cs]'25

xe!?3 18,0 hr EC Tel?2(q, n) Xe 24 (n, v)

Xe126 Stable (abundance 0, 090%) Natural '

xel?7T 75 sec IT 1127(a, 4n)Cs '%7; daugh-
ter CSIZT '

(Table Continues on page 4)
TABLE OF ISOTOPES OF He, Ne, A, Kr, Xe, and Rn, (Cont'd)

 

 

Isotope Half-life Type of Decay Method of Preparation
xel?T  36.41 day EC Te'?%(a,n); 1'%7(q, 2n)
127(P.m); Xe 2 (r, y)
xe 128 ~ Stable {abundance 1.919%) Natural '
Xel?9™M g 0day IT xe'%8(n, y)
Xe129 Stable (abundance 26, 44%) Natural
Xe130 Stable (abundance 4. 08%) ' Natura.ll
xe!3!™ 12 0 day IT Xe'3l(n,n'); fission U
Xe131 Stable (abundance 21. 18%) Natural; fission U
Xe132 Stable (abundance 26. 89%) Natural; fission U
xe!33m 5 2 day IT , Xel3?(a,v); fission U
xe'3® 5,270 day B Xe132(n,v); Xe 1324, p);
.. xe 340, 20): Te 3%, n);
cs133 (. p); Be136(n a):
fission U
xe 134 Stable (abundance 10.44%) ‘Natural; fission U
Xe 113%™ 15 min IT - %e138(n, 2n); Xe!34m,v);
_ Ba138(n, a); figgsion U
xe!®® 9,13 hr 8 xe 3% n, v); xe 3¥(4, p);
36(n, 2n); Ba138(n, a);
fission U
Xe136 Stable {abundance B, 87%) Naturél' fission U
Xe137 3.9 min p 136(n v); fission U
Xe 138 17 min B Fission U
xe !9 41 sec B Fission U, Th
xe!?  -10 sec 8 Fission U, Th
xe 41 ~2 sec g Fission U
xe 143 1.0 sec B ' Fission U
xe 44 ~1 sec B Fission U
Rn?%® <6 min ' a 65%, EC 35% At (%, sn)
Rn?%7 <10 min EC 96%, a 4% Aat T4, an)
Rn?%® 22 min . EC~809%, a-20%  Spall Th; Pb(Clz, apall)
Rn?%? 30 min EC 83%, a17%  Spall Th; Pb(C'%, spall)
’a?'% 2.7h: ' a-96%, EC~4%  Spall Th; Pb(C %, spall)
RnZ!! 16 hr EC 74%, a26%  Spall Th; Ph(C'%, spall)
Rn?!'? 23 min a  Spall Th; Pb(C ', epall)
Rn% 1> ~10705ec {est.) a U??7chain from Th?32
(2, 9n)
TABLE OF ISOTOPES OF He, Ne, A, Kr, Xe, 'and Rn, (Cont'd)

 

Isotope Half-life Type of Decay Method of Preparation

 

Rn’ 16 ~10"%sec (est. ) a ~ u?28 hain from Th232
| (a.8n)
Rn? 17 1072 sec a - U%%%nain from Th?3?
. (a,7n)
Rn218 0.019 sec a U230chain from Th232
| (a, 6n)
' R.nz'19 3.92 sec a Member U2'35.deca.y
| chain
anzo 51.5 sec a Member Th232' decay
| chain
Rn?%! 25 min B ~80%, a~20%  Th%>%(p, spall)
anzz 3.8229 day a | Member Uz'38 decay
"chain

III. REVIEW OF FEATURES OF INTEREST
IN RARE GAS RADIOCHEMISTRY

The rare gases, Group O of the periodic table, are helium, neon, argon,
—'kry-pton, xenon, and radon. Helium and neonl possess no radioactive isotopes
of half-life long enough to permit radiochemical studies in the ordinary sense.
One can conceive situations in which the separation of the remaining four rare
gases from contaminants and from one another might be necessary. However,
the most common problem by far is the separation of krypton and xenon re-
sulting from fission of heavy elements from other fission producte and from
each other. The literature on rare gas radiochemistry is \largely concerned
with this problem, and most detailed discussion in this monograph will like-
wige center around krypton and xenon. As rare gas separations generally
depend on some property which varies greatly and in a regular manner as one
proceeds through the group from helium to radon, the further applicati'on of
the procedure to include other rare gasea will often be simple if required.

Reviews of the '""chemical properties' of the rare gases will be found
in most texts on inorganic chemistry. . In a practical sense, the rare gases
are chemically inert. They will remain chemically unaltered in any reactions

chosen to quantitatively remove impurities other than rare gases from them. .
However, rare gas atoms do interact with other atoms,. molecules, or ions in
their neighborhood. Whether these interactions are properly considered
chemical or "van der Waal's' in nature is of no concern here,

The existence in discharges of species such as HeH" and He2+ has been
established. Hydrates for the four heavier members of the group are known
to exist, and solubilities of rare ga.ae.s in a number of solvents are relatively
high, Solubility in a given solvent generally increases with atomic number of
the rare gas._ A number of studies of rare gas solubility dependence on sol-
vent, temperature, and partial pressure of the rare gas (Henry's law is ap-
plicable over wide ré.nges) have appeared in the literature. One of the more
comprehensive studies also outlines a proposed system for recovering krypton
and xenon from gas streams by their absorption in a counter-current stream
of organic liquid (kerosene). 2

In handling rare gases, especially tracer amounts, one does well to
remember that co_nta.?:t with liquids or solids (including system walls) or
condensation from the gas phase of other liquids or solids may result in re-
moval of rare gases from the gas phase through solution, adaorption or phys-
ical occlusion. Thus in quantitative work rare gas carriers are usually added
for essentiallylthe same reasons that carriers are added for species to be
separated from solutions, . .

A series of substances known as clathrates has received considerable

3,4 Clathrates of argon, krypton and xenon have been prepared by

study.
crystallizing quinol under an atmosphere of the rare gas at high preasure.
Rare gas atoms are contained in "cages' within the reéulting crysfal, the
number of these cages setting an upper limit on the amount of rare gas in the
crystal (one rare gas atom for every three quinol molecules). The actual
concentration of rare gas in the crystal depends markedly on the conditions of
crystallization and is usually considerably less than the theoretical limit. '
Clathrates of fission-product krypton have recently been prepared for use as
sources of Kr85 activity. > The preparation described in the reference re-
sults in krypton concentrations equal to 25% of the theoretical limit, This
correEPOnds to the concentration in the gas phase at about 25 atmosphéres
pressure and results in 3 curies of Kr85' per gram of clathrate (using fission
product krypton. which is about 5% Kr85). The material may be ground to a
powder with no appreciable loss of activity and the loss on standing is only a
few parts per million. per day if the material is protected from water and
other subsfancea which dissolve quinol. '

The chemical operations involved in separating and purifying rare gases
are performed on elements or compounds other than rare gases. The Bepara-
tion of the rare gases from one another must be accomplished on the basis of

differences in some physical property (usually vapor pressure or extent of

6
adsorption). It may often happen that this physical separation will also sepa-
rate some or all of the impurities other than rare gases from a particular
rare gas. Thus it ie often neither necessary nor most convenient to purify
the rare gases chemically as a group as the first step or steps in a radiochem-
ical procedure. Depending upon the conta.ml:.nati.ng species and their amounts
it may be possible that no chernical operations whatever will be necessary.
Substances which may contaminate rare gases are limited to those which
may exist in appreciable concentratione in the gae phase at the temperature of
the experiment. At said temperature thie will include: 1) gases, the term
being used here to denote appreciable amounte of substances above their
critical temperature or having vapor pressures greater than 1 atmosphere;
2) unsaturated vapors, substances whose vapor presa-urés are less than one
atmosphere but which are present in amounts appreciable but small enough -
that their partial pressures are less than their vapor pressures; and 3) satu-
rated vapors in equilibrium with solids or liquids having appreciable vapor
pressures. The word "appreciable in the foregoing must of course be de-
fined in the context of the experiment. Although the above is a large category,
the number of contaminants usuélly encountered is quite small. The rare
gas'es must often be purified from the constituents of air, from volatile species
involved in target dissolution such as hydrogen, hydrogen halides and oxides
of nitrogen, and occasionally from small amounts of hydrocarbons and trace
amounts of elemental halogens. Methods of removal are listed below for each
of these. The list is by no means complete as regards methods which have
been or could be used, but it is hoped that it is representative enough to be

useful.

1) Nitrogen: Mole amounta may be qfia.ntitatively removed by reacting
with titanium sponge, 14-20 mesh at 1000-1100°C. Use of lower temperatures
" (ca. 850°C) has been reported. 6 Oxygen is also removed quantitatively,
Quartz or ceramic furnaces are necessary as these temperatures are above
the sbftening point of Pyrexl glasses,. The reaction is quite exothermic and
so the temperature must be monitored and flow rate of the gas controlled to
prevent destruction of the furnace. Calcium has often been used aa.tisfa.cltorily
to remove nitrogen (400-500°C). Calcium does, however, tend to become
passive through formation of surface filme, For small amounts of impurities
this problem is sometimes circumvented by conducting the reaction in the gas
phase with Ca vapor. 7 Clean uranium turnings at 800°C will aleo react with

nitrogen (and decompose hydrocarbOnsB).

2) Oxygen: Will be removed with nitrogen in the above reactions, Oxy-
gen also reacts rapidly with copper turnings above 350°C to give CuO.

3) Hydrogen, carbon monoxide, light paraffin hydrocarbons: Paesage
9

of the gas stream over CuO at 500°C will oxidize H, and CO to CO,

rapidly. Hydrocarbons will be oxidized in like manner to HZQ and COZ but a

to HZO

temperature of 900°C is necessary to achieve rapid reaction. Subsequent re-
moval of oxygen may be necessary (and is easily accomplished), if high enough
temperatures are used that the dissociation pressure of oxygen over CuQO becomes

appreciable. :
4) Water: In addition to condensation of water in cold traps (a method

often not specific enough), water may be removed by passage through any one
of a number of desiccants. PZOS' Mg(ClO4)Z,' and Ca.Clz are representative,
with equilibrium partial pressures of water over them at room temperature
-5
of 2 X 10

resent the extremes of efficiency in common desiccants. Magnesium per-

, 5%x10°%, and 0.2 mm Hg respectively. P,O; and CaCl, rep-

chlorate has achieved wide use as it is efficient enough for practically all pur-
poses and is easily regenerated by distilling the water off in vacuo at 220°C.
Where 'la.rge amounts of water are involved, gases are often driéd first with
C‘.=1C12 to remove most of the water and then with Mg(GlO4)2 to remove re-
maining traces of moisture,

5} Carbon dioxide: May be removed by passing the gas through a trap .
containing Ascarite, a granular commercial preparation which is esé_entia.lly
asbestos impregnated with sodium hydroxide. Sofnolite, or soda-lime, a
mixture of NaOH and Ca0 may also be used. Sweeping the gas through a
solution of alkali metal hydroxide is' also effective. Solutions are sometimes
less convenient in vacuum systems than solid materials, ‘however.

6) Hydrogen halides: May be removed in the same manner as carbon
dioxide. | |

7) Oxides of nitrogen: Dissolution of targets in nitric acid will result
in evolution of NZO' NO and NO2 ir varying proportions, NZO usually occurs
in relatively small amounts. Its removal may be accomplished by catalytic
reduction with hydrogen or by oxidation to NO by scrubbing the gas with.a
solution of strong oxidizing agént such as permanganate. NO and NO, may
be removed by sweeping the gas stream through a solution of sodium hydrox-
ide. Performing the solution under reflux will in fact wash a good portion of
the higher oxides back into the dissolver flask. If alternate methods of solu-
tion are available it is often most convenient to avoid the us e of nitric acid. |

8)- Halogens: Free halogens are all relatively volatile,. and active halo-
gens may often be present in trace amounts in tai-gets from which rare gases
are to be remdved. If the solution process leaves them partly or entirely in
the zero oxidation state, the gas stream will be contaminated.- Sweeping the
gas through a solution of sodium hydroxide will rapidly convert all the free
halogens to non-volatile species (halide and hypohalite). 10 The rapid ex-

change of the halogens with their corresponding halides in solution is another

8
useful means of decontamination from tracer halogens, 1,12

Iodine is quan-
titatively removed by contact with silver (silvered Raschig rings have been
used as a trap packing for this purpose).- Iodine is reduced to iodide in 1’\1’a.HSO3
solution. _

Where reaction rates are slow enough that purification may not be com-
pleted in one pass, it may be necessary to circulate fhe offgases from target
dissolution through a purification train. The classic, and still most generally
useful, device for circulation (or transfer) of gases is the Toepler pump,
with which rates of about a liter per minute may be obtained. Other circulat-
ing pumps which may be useful in particular instances have been designed.
These consist essentially of a chamber designed to permit only unidirectional
flow of gas and a means of producing periodic pressure changes within the
chamber. Alternate heating and cooling of the gas in the chambet has been
used to achieve flow rates of 0,5 liters per hour. 3 An iron piston enclosed
in glass and electromagnetically operated has produced rates of a few liters
per minute ,1 Pressure differentials de.vel.dped were. not noted in the reference
but are probably small. Another achieves aimilarly. high flow rates by pulsing
the chamber pressure by means of a fnyla.r diaphragm driven by high preasure
gas. 13 .

Ultimately the radiochemist must concern himself with the physical
properties of the rare gases. Table 1 summarizes some data which may be

of interest,

 

 

Table 1

He Ne A.r Kr Xe Rn
Atmospheric abundance 5.25 1,82 0.934 1.14 8.7 6x107'%
(Volume %) x107* x1073 x10"% x107°
Boiling Point (°C) -269 -246  -186 -153  -107 -65
Melting Point (°C) -272 -249  -189 -157  -112 -71

' (25 atm.)

Atomic diameter in 3.57 3.20 3,82 3.94 4.36 _——-

crystal (angstroms)

 

Distillation and adsorption techniques are those which first come to mind
for separation of the rare gases. The pure rare gases have of course been
successefully produced commercially by the fractional distillation of air, with
the exceptions of helium which is extracted from certain Texas natural gases

and radon which is obtained as 2 member of the 4n+2 natural radioactive decay
séries. Fractional distillation is discussed in most texts on physical chem-
istry and the ‘fractionation of rare gases from air is discussed in numerous
places. 17-20 Glueckauf used 2 procedure combifiing fractional distillation of
air and adsorption techniques in the detg_rmination-of krypton and xenon con-
tents of the atmosphere. 21 It is very unlikely that the radiochemist will have
need to use such low temperature fractionation columns in the laboratory so
there will be no d'iscussion. of the method here.

On the other hand, simple distillation and condensation processes will
be used in practically every experiment in transferring gases or effecting
rough separations. Thus a plot ( for low temperatures) of vapor pressures as
a_fu.nction of temperature for rare gases and other volatile species commonly
encountered will prove very handy to the radiochemist. The reader is refer-
red to the generally ava.fl.able Handbook .of Chemistry and Physuca22 for data
necessary for such a plot. Dushma.nZ (page 781 et seq.) also tabulates vapor
preasuré data for a number of substances at low temperatures. In order of
decreasing volatility (decreasing vapor pressure at a given temperature below
the critical temperature) the common gases are NZ' A and 02. Kr, Xe, Rn
and COZ-. and HZO' Gases listed in pairs have vapor pressure curves at low

temperatures which are very similar,

Efficiencies of processes such as the transfer of a gaseous species con-’
tained in a system to a cold trap attached to the system are determined by the
relation of the initial pa.rt-ia.l. pressure of the species in the system to the vapor
pressure of its solid or liquid after condensation in the cold trap, As an ex-
ample, the vapor pressure of solid Kr at -195°C ie 2-3 'mm Hg, so roughly
99%' of the Kr in a system at 20 cm Hg pressure may be collected in an attached
small cold trap at -195°C. As liquid nitroger is generally the coldest con-
veniently available refrigerant, Kr is often manipulated by condensation at
this temperature. The resultant losses decrease with increasing initial krypton
pressure and may be minimized by using- the emallest systeme and the [a.rgest
amounts of krypton carrier practicable -- usually they may be made small
‘enough that they are acceptable for the sake of epeed and convenience. If one
must of necessity collect krypton at pressures comparable to 2-3 mm Hg from
a system, he must resort to lower temperature refrigerants, refrigerated
adsorbents such as charcoal, or Toepler pumps. Presence of small amounts
of noncondensible gases may seriously lower the rates of transfer procegses |
such as the above krypton condensation, due to the relative slowness of gas-
eous -interdif_fusio_n processes. Thus one wishes a vacuum system with a base
pressure orders of magnitude lower than the lowest vapor pressures involved
in nganipulazions ‘he may wish to perform. Systems with base pressures of

10 to 10 © mm Hg are easily constructed and will suffice for the'manipula-

10
tions involved in most rare gas radiochemistry (assuming that macro rather
than tracer amounte are involved). |

Removal of only the least volatile species from a gas stream is feasible
when a temperature can be found at which its vapor pressure ie sufficiently
less than its partial pressure in the gas stream that condensation is essen-
tially complete, while the vapor preseuree of other constituents are higher
than their partial pressures over the t:rap'. Assuming equilibrium conditions
and insolubility of the other epecies in the condensed phase, passage of the
g3s stream through the cold trap effects the desired separation quite simply.
Iflc;ondensa.t'ion occurs under equilibrium conditiona, the fraction of species
lost (passing through the trap with the gas stream) is the ratio of its vapor
pressure at trap temperature to its partial pressure in the influent gases. To
' improve the approach to equilibrium conditions, flow rates are limited and
cold traps packed with some material to provide a large contact area and to
prevent mechanical blow-through of condensed materials. Glass wool, beads
or ringe are materials commonly used for this purpoae. Stainless steel bails
(3/16 in.) combined with suitable glase wool plugs have aleo proved very sat-
isfactory. Such a packing has the advantage that the steel may be warmed by
the induction of eddy currents in it to achieve rapid and efficient removal of
condensed materials. Time for thermal equilibrium to be reached in traps
must always be allowed before initiating gae flow. As cooling of trap packings
under vacuum may be quite slow, temporary introduction of some easily re-

movable gas such as helium to improve heat transfer is often useful.

In like fashion several condensed materials of widely differi.izlg volatil-
ities may often be separated by warming to a suitable tempéra.ture and distilling
the more volatile components into another cold trap. It is often necessary to
melt and refreeze the condensate remaifii.ng behind several times, each time
removing the volatile material evolved to the other cold trap, to effect com-
plete removal of occluded or dissolved traces of the volatile materials (e.g..
the distillation of CO, from CO,

The subject of trap refrigerants should be touched on briefly, Liquid

-ice condensate at dry-ice temperature),

nitrogen (-195.8°C}, dry ice (-78. 5°C), and ice (0°C) are commonly available
in most laboratories, -Fortuna.teljr temperatures lower than that of liquid
nitrogen will usually not be required in handling rare gases other than helium
and neon, It might be noted that pumping on liquid nitrogen will cool it by
evaporation to its triple point (-210, 9°C, 96.4 mm Hg). Témpera.tu.rea inter -
mediate to those of liquid nitrogen, dry ice, and ice are often necessary and
are commonly obtained by cooling a suitable liquid with liquid nitrogen or dry
ice to obtain a partially frozen '"slueh'" at the melting point. Slush or liquid

refrigerants provide better heat transfer from the trap than solid refrigerants,

11
thus dry ice is usually mixed with acetone to provide a slush refrigerant at
-78.5°C. Discussion of cold baths may be found in most texts where vacuum

24,25 The author has found the following baths par-

techniques are discussed.
ticularly useful in working with rare gases: acetone, -95°C, n-pentane, -130°C,
and isopentane, -160°C.

It is usually most convenient to refrigerate traps through which large
~ quantities of gas will flow with liquid nitrogen, dry ice, or ice since such
traps are easily replenished in place on the line. '"Slush-cooled' traps at
other temperatures may usually be reserved for 6pe1-a.tions where they need

take up only small amounts of heat.

Adsorption and Adsorption Techniques

By far the majority of separations of rare gases to be found in the litera-
ture have employed adsorption techniques. Of the large number of adsorbent
rnaterials available, a.ctivated- charcoal has most often been us.ed because it
has the largest surface area (largest adsorptive capacity) per unit mass and
because the rare gases adsorb on charcoal to widely differing extents.

For a given adsorbate gas and adsorbent material the amount adsorbed
per unit mass of adsorbent, called v and usually expressed as cc STP
(Standard Temperature and Pressure,i.e., 0°C and 760 mm Hg) of gas per
gram adsorbent, is a function of temperature of the adsorbent and partial
pressure of the gas over the adsorbent. Three functional relationships among
these variables are of interest: 1) Isotherms showing the dependence of v
on P at constant temperature; 2) Isosteres relating P and T at constant v;
and 3) Isobars relating v and T at constant P. In experimental studies of
adsorption the isotherms are most usually determined at a series of tempera-
tures. From these data the other relationships may be obtained if desired.
The Clausius-Clapeyron equation also relates heats of adsorption and the pres-
sure change with temperature (shape of the isostere),

A number of equations have been used to exp-resa adsorption isotherms,
most of which are empirica.l in nature. Langmuir first derived the hyperbolic
isotherm for monolayer adsorption from theoretical considerations. 26
Brunauer, Emmett, and Teller27 later .derived equations for multilayer ad-
sorption which reduce to Langmuir's equation in the case of a monolayer. The
hyperbolic isotherm may be written v = vsbP/(l +bP), P = v/b(vs—v) , or
P/v = l/b\a"3 + P/vs » where Vg is the cc STP of adsorbate requiret_:l to
form a monolayer per gram of adsorbent, and b is a constant dependent on
temperature and the nature of the adsorbate and adsorbent. The last form is
particularly useful for the extension of data by interpolation since the plot of

P/v versus P is linear. At low pressures, or perhaps more definitiw;rely low

12
v , it is found that adsorption obeys Henry's law (v = kP), and interpolation .on
a linear plot of P- versus v is possible. It will be noted that for P much less
than 1/b, or v much less than Vg the hyperbolic isotherm reduces to Henry's
law with k equal to ‘m_rB . Freundlich's parabolic equation (semi-empirical in
nature) is often useful for linear interpolation on a log-log plot in cases where
it is obeyed: v = KP /n . Peters and Weil28 studied the adsorption of argon,
krytpon and xenon on charcoal and calculated the constants in the Freundlich
equation from their data at several temperatures, These are given in Table

2 for handy reference. Adsorption data for a number of gases on charcoal are
tabulated in Chapter 8 of Dushman. 23 The adsorption process releases heat
and as expected from LeChatelier's principle adsorption decreases with in~
creasing temperature, other factors remaining constant. Adsorption increas-
es with increasing pressure. Adsorption of different gases tends to increase
with decreasing volatility (extent of adsorption tends to increase in the same

order as the boiling points of the gases),

Table 2. Adsorption constants for argon, krypton

and xenon: v = KPl/n

 

 

T°C K 1/n
Argon -80 0.500 0.950

-18 .0.0764 1.0

0 0.0581 1.0
Krypton -80 : 2.927 0.711
-18 - 0.497 . 0.885

) 0.340 1.0
Xenon _ -80 15.99 0.574
' -18 2.458 0.692

0 1.583 0.77

 

Theoretical treatments discuss adsorption in terms of: 1) formation of
a monolayer of adsorbate on the surface, 2) formation of multilayers on plane
surfaces, and 3) capillary condensation in small pores. Adsorption of a gas
is considered the result of van der Waal's forces of the same type involved in
its condensation. The magnitude of the interactions resulting in adsorption is
indicated by the fact that heats of adsorption are generally comparable to the
heat of condensation for a given gas, Experimental isotherme may conform
over wide ranges to theoretically derived equatione (as the hyperbolic iso-
therm) for suitable values of the constants in the equation. However, values
of the constants thus obtained are often not easily related to the physical
models used in the derivations. For discussion of adsorption in general and
a key to the literature on the subject the reader is referred to Chapters 7 and

29

8 in Dushman’> or to §. Brunaues's book on physical adsorption.

13
Some comments on the nature of activated charcoal may be worth while.

30,31 There are numer-

These are mostly from two recent review articles.
ous materials from which activated charcoal may be prepared and many meth-
ods of preparation, 32 For example, one might start with coconut shells and
heat in the absence of air to remove the greater part of the elements other.
than carbon as volatile materials. The resuiting material is then activated
with steam at 700-1000°C. The large specific surface area of charcoal (of
the order of 1000 square meters per gram) is due to its high porosity, as
opposed to the nonporous adsorbents such as l.amp. black whose large specific
surface area is the result of extremely small particle size. Thus the surface
per gram of lamp black is a mmarked function of the average particle size,
whereas the specific surface of activated charcoal is almost independent of
particle size. Furthermore, most of the surface area in charcoal apparently
Tesides in pores with diameters comparable to molecular dimensions. It
follows that one might expect principally monolayer adsorptidn on charcoal.
Thus adsorption on charcoal might be expected to obey the hyperbolic iso-
therm, and does so over wide ranges. The potential pore structure of 2 char-
coal is dependent on both the starting material and the conditions of pyrolysis.
Charcoale from hardwoods,; nutshells and certain grades of coal generally
give the highest porosity. The material after pyrolysis (though porous) still
has low adsorptive capacity and the process of activation consists essentially
of burning out more material in order to either open new pores or to enlarge
existing pores to diameters which will a.c.cor.nmpdate adsorbate molecules
(perhaps both processes occur), Steam is widely used for activation as the
reaction is endothermic and better temperature control is possible. One ex-
pects then that the propérties of the material he receives in the laboratory
depend not only on ‘materials used in preparation but also very markedly on
the history of its pyrolysis and activation. Excessive activation can in fact -
remove more surface area than it creates, the adsorptive capacity going
through a maximum as the activation proceeds. Since the distribution of pores
as to size includes m#ny with diameters comparable to molecular dimensions,
a "molecular sieve' effect may be expected to some extent, Thus some pores
will accommodate a given molecule but are too small for a slightly larger
molecule. During activation the adsorptive capacity for various species may
therefore not be increasing in the same ratio. Charcoal has in the past been
referred to as an allbtro;pe of carbon, but elemental analyses indicate atom
percents of carbon always less than 100% and sometimes as low as 75%. The
other elements present are principally hydrogen and oxygen, but appreciable
amounts of nitrogen, sulfur, chlorine, and inorganic ash may also Be present,
It appears likely that the surfaces on which adsorption occurs are not carbon

surfaces but hydrocarbon surfaces modified by the presence of oxygen and

1k
traces of other elemente. This implies that all adsorptiofi sites are very
likely not identical. | | _

Activated charcoal is thus a class of substances whose chemical and -
physical properties may vary within rather wide limits. Let us hasten to add
that this does not really impair ite usefulness as an adsorbent. The adsorp-
tive properties of charcoals in separation systerns are usually not so critical
that even variations like factors of two or more will result in loss of sépara-
tion, though such variations certainly affect the procedure in detail. The
intent is to point out that the experimenter sheuld not be surprised if_the ad-
sorptive properties of his charcoal differ from those he sees quoted in the
literature. In one study of the adeorption of xenon on seven charcoals at room
temperature, the amount per gram of charcoal adsorbed at a g.iven pressure
varied as much as a factor of five, 3.3 These charcoals were of course chosen
on the basis of differences in materials and preparations expected to prbduce
a range of adsorption characteristics. In order that early experience may be
used to predict with fair accuracy the course of separations in systems built
at later dates, it is probably wise to originally purchase a quantity of char-
coal sufficient for many years' needs.

Before use and between successive uses it is necessary to outgas char-
coal, both to prevent contamination of samples with small amounts of residual
gases from previous runs and to remove adsorbed materiale (such as water)
whose accumulation fna.y_ adversely affect the adsorptive properties of the
charcoal. This process should not be s0 e:lctrern_e as to change the material
through further activation. A routine which has proved satisfactory is out-
gassing in vacuo at 350-400°C. At Lawrence ‘Radiation Laboratory the out-
gaseing is routinely continued until the pressure is 10_4 mm Hg or less as
judged by a thermocouple vacuum gauge in the vacuum system manifold.

One may operate in various ways to achieve separations of.gases through
differences in the adeorption on charcoal (or any other adsorbent). It may
prove useful to point out that such separations are basically similar to sep-
arations of species in solution using ion exchange resins, The many.radio—
chemists who are familiar with separations using such resins will note close
analogies even in the various techniques used. |

Separation procedures must be designed with kinetic cofisiderationa in
mind -- adsorption from a stream of gas ie not an instantaneous process, It
was stated earlier that the adsorptive capacity of charcoals does not depend
on particle size, but it is expected that rates of adsorption will be so depend-
ent. A reduction in the average particle diameter increases the exterior

surface (per gram adsorbent) presented to the gas phase and shortens paths
through cracks and pores along which adsorbate molecules must diffuse in

order to reach adsorption sites in the particle interior. Approach to equi-

15
librium conditions should thus improve with smaller particle size and in-
creased contact time with the adsorbent (decreased flow rate), A compromise
is necessary as too small particle size will result in high resistance to gas
flow. In flow systems low temperature adsorbent beds may not be maintained
in thermal equilibrium with the refrigerant if the gas enters the bed at ele-
vated tempera.tufe or if the rate of heat release from the adsorption processes
occurring ie high., Flow rates and conditions affecting the rate of heat trans-
fer are thus important in this respect. Although it is certainly of great inter-
est and long-range usefulness to understand in detail the kinetics of these
processes, the more practical and faster approach to design is empirical.
Deeign factors are chosen to improve the kinetics insofar as possible and the
experiment is tried, If the experiment succeeds, factors affecting kinetics
.are usually nof varied to deterrnine the extreme conditions under which the
procedure is satisfactory. Thus in plé.ces such as the Detailed Procedures
where flow rates and charcoal particle sizes may be quoted it should be under-
stood that these are sufficient for the separation, but may or may not be

necessary should other considerations make changes desirable,

_ Most rare gas separations will commence with total adsorption of all
the gases of interest in an adsorbent trap. In thie manner one may minimize
pfbblenis arising from non-uniform introduction of sample, temperature
variations, and adsorption kinetics by ueing lower temperatures and more
adsorbent than should really be necessary. By treating all adsorption steps
in this manner one tends to reduce losses, since in subsequent desorption
steps slow processes may affect the history of the separation in detail but

will not result in actual loes of sample.

In their early work on rare gases, Peters and Weilz8 showed that argon,
krypton and xenon could be separated by total é._ds orption followed by fractional
desorption in vacuum from activated charcoal. They adsorbed the mixture
on charcoal at -190°C (roughly 10cc STP of each gas was used on 38 grams
of charcoal) and then raised the trap tb a suitable temperature and desorbed
the gases using a mercury diffusion pump backed by a Toepler pump to trans-
fer them to a gas buret for measurement. Temperatures at which each gas
may be desorbed '"pure" of the others are somewhat a compromise between
speed and separation factor -- removal of argon at -93°C, krypton at -80°C
and xenon at 0°C would be typilcal. " From knowledge of the isotherms, amounts
of adsorbent and of each adsorbate initially, and the volume with which the
adsorbent bed is equilibrated before each lift of the Toepler pump one could
calculate in a straightforward but laborious fashion the composi;‘.ion of the
gas in the adsorbed and gaseous phases after each stroke, assuming uniform

distribution of the gases on'the adsorbent. Even cursory examination of the

16
isotherms suggests that while separations performed in this manner may pro-
duce gases which are around 99% pure, separation factors of about 100 are
the limit obtainable and would not be éatisfa.ctory in most radiochemistry.

It might be noted that Glueckau.f34 analyzed mathematically the 0peratiofi of

a system in which rare gases are successively adsorbed and desorbed on a
series of charcoal traps and used such a system for separating helium and
neon as the final step in the determination of their abundance in 2ir. A simi-
lar system was later used35 for the separation of argon, krypton and xenon
in an experiment to determine the relative yields of krypton and xenon isotopes
from uranium fission by volume measurements on the separated gases and
later mass spectrographic analysis. In these procedures a series of many
traps was required.

A sidelight in the work of Peters and Weil28 is worth mentioning to
point out a matter of safety. They performed an experiment in which it wae
shown that anzz adsorbed quantitatively from liquid air onto silica gel. As
they point out, silica gel was used because of the explosive potentialities of
charcoal and liquid oxygen. Because of this hazard it seems advisable to use
liquid nitrogen, rather than liquid air or oxygen, to refrigerate charcoal
traps as a safeguard in case of breakage, and to operate at reduced pressure
when introducing air to a charcoal trap cooled in liquid nitrogen to avoid con-

densation of liquid air in the trap itself.

The techniques of Peters and Weil and of Glueckauf mentioned above
have today been replaced by what may be generally termed chromatographic
techniques. These 'u.su.a.lly involve adsorption of a .sa'mple on one end of an
adsorbent column, and subsequent passage of some eluent gas through the
column to achieve separation. I.n fact, Gl.uecka.u.f21 in his later determination
of the Kr and Xe contents of the atmosphere separated the rare gases by elu-
tion from a charcoal column ra.fher than by the method used in the He and Ne
determination already mentioned. Keulemans' recent book36 will serve as
"~ an introduction to the subject of gas chromatography and as an up-to-date
source of literature references. Only a few remarks will be made here and
some journal references cited. Chromatographic seeparations involve a sta-
tionary bed of liquid or solid adsorbent through which a stream of liquid or
gas moves. Rare gas separations on charcoal involve a stream of gas moving
through a packed column of solid and thus would be classified as gas-solid
chromatography. Methods of operating chromatographic columns will gener-
ally fall into one of three categories: 1) Elution development in which a
sample of the gases to be separated is adsorbed at one end of the column and
a flow of slightly adesorbed carrier gas (eluent) set up through the column.

2) Frontal analysis in which a stream of gas containing the sample mixture

17
is continuously passed through the colufnn. 3) Displacement development in
- which the sample mixture is adsorbed at one end of the colfimn and a contin-
uous flow initiated of gas more strongly adsorbed than a.fiy of the gases in the
sample. Elution development results in the eventual emergence from the
column of bands or "peaks'' of the sample components- (diluted of course with
eluent) in the order of their increasing adsorption. If the components differ
enough in their adsorption it ise possible under pfoper conditions to obtain
each component in pure form as a peak separated by pure eluent from adja-
cent peaks. If the stream of gas consists of an ""unadsorbed' carrier .gas in
addition to the mixture of sample gases, frontal analysis results first in the
emérgence of pure carrier ‘'and then of the least adsorbed aampie component,
diluted with carrier. Eventually the component which is next to the least ad-
sorbed also breaks through, and so on until the column is completely saturat-
ed with all components and the influent and effluent gases have the same com-
position. Only a sample of the least adsorbed component may be 6bta.ined
pure of other sample components. In displacement developmefit, bands of
the components separated by zones containing a mixture of the species in
adjacent bands emei-ge in order of increasing adsorption. Eventually the pure

dieplacer emerges. Partial recovery of fairly pure componenta is possible.

Theoretical predictions as to the details of chromatogi—aphic separations
vary as the adsorption isotherm is linear or nonlinear, and the process 'ideal"
(thermodynamically reversible) or '"non-ideal'". Linear-ideal and nonlinear
ideal chromatography were treated by ‘l!.’ilson31r and by de Vault. 38 Linear

' 39-46

non-ideal chromatography has been treated by a number of investigators,
and nonlinear non-ideal chromatography by Klinkenberg and Sjenitzer. 47 If
very large amounts of gas are involved, adsorption isotherms will usually be
nonlinear. Gas-solid chromatography is also likely to be non-ideal as the
attainment of equilibrium between the phases offen cannot be considered in-
stantaneous, among other things. Thus the radiochemist may well be con-
cerned in his separations with the nonlinear non-ideal case, the least acces-
. Bible to accurate theoretical treatment. The théory of the simplest case of
linear-ideal chromatography for a single adsorbate provides useful qualitative
insight into the processes occurring, however, Keulemans3 treats this case
on page 112 et seq. of his book. A brief account of the reasoning will be
given here. _

Consider an adsorbent uniformly packed into a cylindrical column of
constant diameter and at constant temperature throughout. If an adsorbate
is present in some small cylindrical element of the column taken in cross-
section, it will dietribute itself so that a fraction g of the total ie in the gas

phase and a fraction (1 --¢) on the adsorbent, As a fraction g of the adsorbate

18
is in the gas phase, a given adsorbate molecule resides in and thus moves

with the gas phage g of the time on the average. It then follows that the rate

of movement of adsorbate through the element is just ¢ times the linear flow
rate F of eluent gas. Assuming the perfect gas laws, one may calculate g

and F, Let us call Ge the volume available to the moving gas and Ae the
weight of adsorbent in the element e. Pa. is the partial pressure of adeorbate
and v the amount adsorbed in cc STP per gram of adsorbent. The factor for -
converting gas volumes in cc-mm of Hg at column temperature to standard
conditions may be called C = 273 "K/(T.,K X 760 mm Hg). The fraction of
adsorbate in the gas phase will then be ¢ = CP_G_/(CP, G, +vA)=(l+va_/
CPa. Ge)-1 . As the column is uniformly packed the ratio of Ae/Ge will be

the same in any element of column volume and equal to the ratio of total weight
A of adsorbent to total void space G in the column. Furthermore, v/Pa =

k will be constant for all v if the isotherm is linear, in which case k may.

be referred to as the Henry'e law constant. In this case, ¢ = (1 + k.A/C(I‘:)-1
and is the .sa.me at constant temperature everywhere in the column regardless
of adsorbate concentration. The linear flow rate of eluent will be F =

qL/C GP, .
length in centimeters, G is the cc of total column void space, Ptot is total
gas pressure in mmm Hg, and C is the consetant defined above, The rate of
movement of adsorbate is then OF = (1 + xA/C G)_1 {(gL/C GPtot) = qL/Ptot
(CG + kA). In most instances ¢ << 1, or (kA/C G) >> 1, so that this expres-
sion simplifies to OF = qL/kA P, In either case the rate of adsorbate

» where q is the gas flow in cc STP per minute, L is the column

movement is constant everywhere along the colurhn regardless of the concen-
_tration of adsorbate. It follows that the rate of movement of all points in a
band is the same, and the band shape does not change during its travel along
the column.

In ideal chromatography, frontal analysis results in the gradual eiten-
sion of a band of adsorbate at.consté.nt concentration down the column, assum-
ing constant column temperature and constant partial pressure of adsorbate
in the gas entering the column. By equating total adsorbate introduced to the
column to the amount present in the gas phase plus adsorbent phase in a band
of length I on the column, one arrives at £ =({(q L/Pfot\) (CG + vA/Pa)_lt ,
where t is total time of flow in minutes, In the case of a linear isotherm
this is just !a = 0Ft, as one intuitively expects. Even if the isotherm is
nonlinear, (v/Pa) is constant everywhere in the band, for Pa. constant, This
formula is thus a useful approximation in this case if P_ is known and con-
stant in the gas entering the column, and the isotherm is known at column
temperature. All the above discussion is strictly for a single adsorbate, as
appreciable concentrations of a second adsorbate may affect the adsorption

of the first.

19
In many experiments, the sa.rriple introduction is accomplished in a man-
ner corr’espofiding to frontal analysis, with subsequent separation of the com-
ponents by elution development. Detailed Procedure | (see last chapter) in
which air containing Kr and Xe is admitted to a charcoal column over a long
period is an example. After the admission of the sample is complete, the Kr
and Xe are separated and recovered by elution into a suitable. receptacle with
helium at suitable column temperatures. In such ‘cases a minimal requirement
for satisfactory operation is that the length of the band of least adsorbed sub-
stance be less at the end of sample introduction than the column length. Thus
in Procedure 1, the final length of the band of krypton must be less than the
column length. In the subsequent separation by elution the rates of movement

for two adsorbates are & F and csz, go that times of emergence can be

“1
calculated. For separation to be achieved the rear of the first band must

overtake the front of the second band. Thus where T is the time for emer-

gence of the rear of the first band, T o F =To,F + 1, =L, where [, is the

band width of the more highly adsorbed component. Thus L = Izkz/(kz _kl)

is the minimum column length in which separation of two components may be
achieved, in the linear-idedl case,

If, as is usually the case with charcoal, the isotherm is hyperbolic,
~then V/Pa = b(vs—v). -Thusl_l as v increases toward Vg (that is, the fraction
of the surface covered with adsorbate becomes appreciablezs), V/Pa decreas -
es. It will be seen that ¢ thus incre-éses as v increases, and the more con-
centrated regions in a band move faster. Assuming ideal conditions the re-
sult is an emergent band with a very sharp front and a2 long tail. In a prac-
tical sense, if one has used a value of k obtained from the linear low pressure
region of the isotherm to estimate the emergence time, the peak frofit will
emerge earlier than calculated. In the latter portions of the peak tail where
concentrations lie in the linear region of the isotherm, rate of movement will
approximate that calculated. '

Departures from '"ideality'" such ae those due to appreciable rates of
longitudinal diffusion in the gas or non-instantaneous mass transfer between
phases result in broadening of bands during their travel, usually in a nearly
symmetrical fashion. Superposition of these effects can give useful qualita-
tive pictures.of the bands and their motion in the various cases. Thus non-
linear, non-ideal conditions lead to asymmetric bands emerging earlier than
under linear -ideal conditions. As to peak shape, one might ekpect the emer-
gence first of a region in which adsorbate concentration increases slowly,
then a eteep main peak front, a less steep main peak back, and finally a long
tail in which adsorbate concentration slowly approaches zero.

There are two general approaches to more rigorous theoretical treat-

ment of chromatography.

20
The "'plate' theory treats the column ase a series of fuxtaposed "equiv-
alent theoretical plates! in each of fivhich a separation factor between two
species corresponding to the ratio of their distribution coefficients is achieved.
The distribution coefficient is a ratio of concentrations in the two phases for
a substance,.

The ""rate'" theory takes into account the physical processes such as
convection and diffusion, and séts up partial differential equations to des¢ribe
them. The solution of these equations (in the adsorbate concentrations and
their derivatives, with time and axial distance along the column as independent

variables) is then essayed.

Both approaches involve various simpl.-i.fy-lng assumptions. The journal
references to theory given earlier include accounts of both treatments. It
might be apecifically noted that one article by (_Trlueclcau.f48 deals with the
effects of heating on movement and shapes of peaks when the adsorbate is a
highly radioactive gas such as'Kr85. Peaks are sharpened and separations
improved. One might say qualitatively that as the latter portions of the band
move through regione warmed by passage of the active gas, ¢ is larger there,
The tail would thus tend to move faster and ''catch up" with the main peak,

In the event that the radiochemist has need to use charcoal after exposure to
high radiation fielda, iiL might be noted that the adsorption of krypton and
xenon on charcoal was found to be unaffected by previous irradiation with 1-
Mev electrons (doses up to 1350 watt-hr per gram).

The "rate" theory predicts the nature of the dependence of column effi-
ciency (i.e., the '"height of an equivalent theoretical plate' ) on various fac-
tors such as eluent flow rate, adsorbent particle size, column inlet and outlet
pressure, nature of the eluent gas, etc. Although the development of the
theory becomes mathematically fairly complicated, application of the results
to practice is often relatively simple. The radiochemist who plane to make
much use of gaé chromatography will certainly find -study of the theory inter-

esting and profitable.

However, one actually needs only a bare acquaintance with the theory to
build a system to separate rare gases. The following paragraphe are written
for the radiochemist who meets a problem involving rare gas separations, and
who wishes to solve it with a2 minimum expenditure of time and effort. For
this purpose the equations given earlier for the linear-ideal case may prove
useful. The assumption of linear-ideal conditions is of course a coneiderable
departure from reality in most cases, and the treatment recorded earlier is
overly simplified in some respects even for these conditions. Unless one
actually determines the adsorptive properties of his charcoal, he must also

keep in mind that these may vary by factors of two or more from quoted values

21
in the literature. The author is in fact unaware of adsorption data in the lit-
erature for Kr and Xe at temperatures less than -80°C. Isotherms at lower
temperatures (such as -195°C) must thus be estimated by rather dubious ex-

trapolations.

‘The distribution of rare gases will generally be heavily in favor of the
adsorbent phase in separation systems so that rate of band movement may be
written OF = q L/kA%o_b. Not all of these quantities may be varied at will with-
out impairing column efficiency. Thus theory predicts an optimum value for
the linear flow rate of gas: th:rough the column, Linear :Eldw rates of the order
of 10 em/sec might be considered typical, though this parameter is not par-
ticularly critical and may be varied by factors of two or three without serious
effect. Column efficiency di'ops less rapidly near the optimum value with
increased linear flow than with decreased linear flow, Efficiency is also
impaired at very large column diameters, Column diameters are generally
around 1 cm or less, Again, efficiency is not a rapidly varying function of
diameter, and columnse of several centimeters diameter have been success-
fully used (e.g., Detailed Procedure 1, given in last chapter). As the bulk
density of charcoal generally liee in the range 0,5 = 0,2 gm/cms. it follows
that extreme values of L/A, column length per gram of charcoal, found in
various satisfactory columns will differ by only about a factor of 10. 'In
addition, L/A is not conveniently varied once one has constructed his system,
Columne are most often operated with an inlet pressure of eluent sufficiently
above atmospheric pressure to sustain the desired flow, and with atmospheric
pressure at the outlet. Thus %otis-nearly always close to 760 mm Hg. One
wishes the linear flow rate of eluent to be as nearly constant as possible
along the length of the column, and it ie thus desirable that the ratio of inlet
to outlet pressure be as near unity as possible. As a certain difference
between inlet out outlet pressure is required to maintain the desired flow in
a given column, roughly speaking, it follows that higher operating pfessures
are preferable in this respect. Band velocities for a given adsorbate a.fe
most conveniently varied over wide ranges by adjusting the column temper-
ature, i.e., by changing the ratio v/Pa = k which in the case of a linear
- isotherm is a constant. Adsorption generally increases rapidly with decreas-
ing témperature. and band velocities will often vary several orders of
magnitude over accessible temperature ranges. Thus for Kr at 1 mm Hg
partial pressure over charcoal, the values of v in cc STP per gram are
about 0.1 at 25°C, 3 at -80°C and 100 ( a rough guéss) at -195°C. Thus for
a2 typical chromatographic column one niay generally find some temperature
at which the band velocity of an adsorbate of interest is auch as to give a

suitable retention time.
In addition to adjusting band velocity, one may adjust column length to
achieve initial sample retention. Increases in column length wil} usually
improve separations of several components also, since retention times are
proportional to L, the column length, while band broadening increases

roughly as Ll/z.

A few examples of prior calculations which one might perform in es-
timating the p'erformance of a proposed chromatographic column are given
below. If the calculations indicate that the proposed operation is unsatisfac-
tory or borderline, then various column parameters and/or operating con-
ditions may be adjusted in the indicated directions before construction of the
system. If the operation appears to be satisfactory with a suitable margin of
safety, particularly as regards initial retention of sample. in a band of width
small with respect to column length, the system might be tried as i:roposéd.
In the latter case, further changes which may prove neceassary can probably

be made on the basis of a very few trial runs.

The length of an adsorbate band was given earlier as ‘Ea = (th/Ptot)
(CG + vA/Pa)_l, and for the usual case where CG << vA/Pa this may be re-
' written to a close approximation as £, =@qP_t/P,_ NvA/L)-1. In this form
it is obvious that the band length is simply the total cc STP of adsorbate in-
troduced divided by the cc STP on the adsorbent per unit column length. If
the partial pressure of adsorbate is constant during sample introduction, v
may be determined if the isotherm is known at column temperature. Values
of other quantities required for the calculation of band length are e:k-perimen—
tally accessible, Assume that 100 cc STP of Kr in air is introduced to a char-
coal column at -195°C, and that the pa.rtié.l pressure of Kr in the influent gas
is constant at 0.02 mm Hg. The column containe 5 grams of charcoal per

centimeter length. If the adsorption of Kr under these conditions is 10 cc STP
er gram of charcoal, one arrives at f = - 100 cc STP Kr '

per g ' ' Kr (10 cc STP Kr/gm)(5 gm/cm)

2 cm at the end of sample introduction. The value of v used above is strictly an

 

estimate as the adsorption isotherm is not experimentally known at this.tem-
perature. The foregoing corresponds to a typical sample introduction in
Detailed Procedure 1, however, and using a p-y survey meter on bands of
radioactive Kr, widths of a very few centimeters are noted. The actual col-
umn length used in Detailad Procedure 1 is about 25 cm, a considerable
margin of safety thus being allowed for uncertainties in the calculations and
variations from run to run in such factors as the partial pressure of Kr in
the gas entering the column. Le.t us assume that the above sample also con-
taine 100 cc STP of Xe. Under the same conditions of temperature and rare
gas partial pressure, values of v for neighboring rare gases are typically

about a factor of 10 greater for the heavier of the two. Thus it ie safe to say

23
that Xe will be initially retained‘ in a band narrower than the band of Kr,
Beyond this, no statements as to the actual disposition of Kr and Xe on the
column will be attempted. The values of v being discussed here are compar-
able to v, the cc STP required to form a monolayer, which is of the order of
100 cc STP per gram of charcoal. The portions of the isotherms involved are
unknown, but definitely nonlinear. A more serious complication is the fact
that at these concentrations on the adsorbent the adsorption of each rare gas
will certaifily be affected by the preeence of the other, and by the adsorption

on the column of large quantities of air,

Having retained the Kr and Xe in suitably small portions of the column,
their separation by elution development may be attempted. As the adsorptivi-
ties, and thus the band velocities, differ by about an order of magnitude under
most conditions, sufficient separations are easily obtained, even allowing for
band '"tailing''. However, because of this tailing the purity of the Kr with
respect to Xe may be expected to be better in general than the purity of the
Xe with respect to Kr. The purity of both fractions will usually be sufficient
for most purposes. Although the equation given earlier for mifiim_um column
length to achieve separation, L = 12'k2/(k2—kl) where ''2'" refers to the
more and "1" to the less adsorbed component, may be useful in other con-
nections, it is of little use with rare gases in the mode of operation under
diecussion. If kl/kl is gi'ea.ter than 2, the ca.lcul.ated minimum length will
be less than the band width of the less adsorbed gas at end of sample intro-
duction. That is, in the linear-ideal case, krypton could be eluted off a col-
umn only a few percent longer than the band width of xenon, leaving pure xenon,
However, all but a few percent of the krypton would have left the column be-
fore the end of sample introduction. Elution of Kr from a column would take
an unnecessarily long time at -195°C, so one raises the column temperature
in order to bring the Kr off in a few minutes. A sample calculation of eati-
mated retention time during the elution developrflent appears in the second
example below, - :

Adams, Browning, and Ackley50 recently reported on a éystem for re-
moving Kr and Xe fission products from homogeneous-reactor off-gases and
retaining thern until radioactive species have decayed sufficiently to permit
release to the atmosphere. It was of course necessary in this case ‘that the
system be designed as accurately as possible before trial. Some discussion -
of the theory used in designing the system is given in the article and a refer-
ence to a more detailed treatment of the theory is noted. 5.1 While the linear-
ideal approximation would not be satisfactory for use in design in this case, it
nevertheless provides a useful illustration of the calculations and a comparison

of the rough predictions with experiment. Fission gases contained in a stream

2
of oxygen flowing at a rate of 2 liters per minute (at 1 atmosphere preasufe
and 25°C) were passed through two parallel series of i:ipes containing charcoal.
Each series of pipes contained a total of 520 lb of activated charcoal and con-
sisted of 40 ft of 0. 5-in. pipe, 40 ft of 1-in. pipe, and 60 ft of 6-in. i.d. pipe.
The charcoal was at 25°C. From the total internal volume of the pipes and the
weight of charcoal one arrives at an-average bulk density of the charcoal equal
to 0.69 grn/cm3. Assuming uniform packing the ratio L/A in each section of
pipe is 1/(bulk density)(cross-section in cmz). Values of L/A in cmn/gm are
thus 1.14 for 0,5-in, pipe, 0,285 for 1-in. pipe, and 0.00792 for 6-in. pipe.
In the 0.5-in. pipe oF = qL/k-APtot _ 1000 em3 /min. X I, 14 cm/gm

_ 760 mm Hg X k
1.50/k'in cm/min. Similarly OF = 0.375/k in the l-in. pipe and 0.0104/k

 

in the 6-in. pipe. As there is roughly 1 cm3 of void space per gram of char-
coal in the column, one may calculate that only a few percent of the adsorbate
in the column resides in the gas phase. Use of the simpl.er formula above for
8F is thus justified. Retention time in each section equals the length of pipe

o . . . . _ {1220 1220 .
divided by band velocity, Total retention time is Fhus Ty = 150 t 5375t

fi% k = 1.80 X 105 X k(min.) = 125 X k (days). The experimental values
of k given in the article are about 1.3 for Xe and 0.08B for Kr (in cc STP of
adsorbate per gram charcoal per mm Hg adsorbate pressure). The calculated
retention times are thue 10 days for Kr and 163 days for Xe. Observed re-
tention times were 10 days and 60 days respectively. This article also con-
tains interesting experimental data on variations of Henry's law constant with
temperature, rmoisture content of the adsorbent, nature of carrier, partial
pressure of krypton, and the adsorbent used. |

Koch and Gra.ndyEz Tecently reported studies of Kr and Xe separations
on charcoal using tracer Kr85 and }(e133. The tracer rare gases were de-
posited at one end of a column of charcoal and eluted off using columns of ebout
l-cm diameter and lengths of 5, 16 or 43 cm, helium eluent flow of 0.6 or
1.6 liters /min., and operating temperatures of 25°C, 0°C and -80°C. Emer-
gent peaks were observed by measuring 'ra.dioactivity in the exit stream, and
times for breakthrough, peak maximum and 1% of peak activity in the tail
recorded. Satisfactory separations were obtained in all the trials, ‘To cal-
culate the linear-ideal approximations for retention times in the case of 1.6
liter/min. helium flow, 43-cm.column, and 25°C column temperature, we
may write Tp = L/gF = kAZ%D_b/q- In the article the effective (bulk) density
of the 40 to 50 mesh charcoal is estimated at about 1 gm/crn3. Column vol-
ume is estimated geometrically to .be 43 cm3. Thus TR = W =
20.4 k (in min,). Using k as 0.08 for Kr and 1.3 for Xe
from the last example, the calculated Kr retention time is 1. 6 min. and the
Xe retention time 26 min, Observed times for emergence of the peak max-

ima were 2.1 and 20 min., respectively.

25
The emergence of various components from a column may be detected
by means of the corresponding change in some property of the effluent gas.
Keulemans discusses various methods of detection. The most widely used
technique involves a thermal conductivity cell (ké.thérometer) which detects
changes in the thermal conductivity of the effluent gases passing through it.53’ >4
For radioactive gases another method is tb pass the gas leaving the column
through an ionization chamber or near a counter of some sort55 to detect the
activities of the various componeéents as they leave the column.

In separations by charcoal chromatography impurities other than rare -
gases will interfere in certain cases. Thus while argon, oxygen and nitrogen
adsorb to slightly different extents, the separation of argon from large amounts
of oxygen and nitrogen is best accomplished chemically. If only krypton and

xenon are of interest it is not necessary to remove air before the chromatographic
separation. Methane is very difficult to separate from krypton by adsorption

on charcoal or by distillation, thus light hydrocarbons must usually be separ-
ated chemically from krypton. Certain materials should always be removed
from the gas before its introduction to charcoal since they are difficult to
remove from charcoal and their accumulation tends to 'poison' it (e.g.,
water, 50 oxides of nitrogen, HCl). Most impurities may either be removed
from the mixture of rare gases before their adsorption or from the separated
rare gas fractions after their elution. This choice may be made as a matter
of convenience in the particular experiment.

Though charcoal is the most generally useful adsorbent it is by no means
the only possible one. "Adsorption is a very general property of surfaces.
Silica gel has specific areas less than but of the same order of magnitude as
those of activated charcoal (its use in the adsorption of radon has already
been mentioned). 28 There is in partiéular_ one other class of adeorbents
which should be mentioned. These are the so-called "molecular sieve'' ma-

56,57 now commercially available from Linde Air Products Company,

terials,
Tonawanda, New York. Actually these are zeolites and as such their adsorp-
tion properties are discussed to some extent by Dushman. 23 Thus, the 5A
Linde _.Molecular Sieve 18 CaO. A1203.SiO'2. and the 4A Sieve is a sodium
alumino-silicate. Their peculiar property is that removal of the original
water of hydration (activation) occurs with little change in the lattice structure,
Thus a highly. porous medium results in which the pores are of remarkably
uniform (and molecular) size. The 4A and 5A materials have pores of roughly
4 and 5 angatréms diameter, respectively. Specific surface areas as great

as those of many activated charcoals have been reported. The extent of ad-
"sorption for a given gas is strongly dependent on the degree of dehydration of
the a.cisorbefit and in fact goes through a maximum at nearly complete dehydra-

tion. Because the pores in which adsorption occurse are of molecular dimen-

26
sions, the size of the adsorbate molecule strongly influences the extent of
adsorption. Thus the adéorptiOna of two species may be quite different rela-
tive to one another from the adsorptions on charcoal, and separations impos-
gible on charcoal may be feasible. A recent article gives an example which
will illustrate the point, 58 The mixture studied was nitrogen-krypton -meth-
ane., Omn a charcoal column (20°C), the nitrogen came off first and the krypton
and methane considerably later but together. On 5A Molecular Sieve (2%
water by weight) at 20°C, the nitrogen and kryptofi came oif together, followed
considerably later by the methane. It was then found that the separation of all
three gases could be achieved by using a mixed adsorbent of "molecular sieve"
and charcoal. It was further found that the relative positions of the peaks were
quite close to those calculated by using the weight percents of the two adsorb-
ents. in the mixture to obtain a weighted average of the peak positions usi:ng
the pure adsorbent. The possibility that "‘molecular sieves" or mixed adsorb-
ents could be usefu.l in the separation of rare gases from samples is worth
keeping in mind. The author is unaware of any such radiochemical procedures

in the literature to date.

IvV. SAMPI.E SOLUTION AND EXCHANGE WITH CARRIERS

The radiochemist may have need to recover rare gases from a variety
of target materials, though most usually from heavy elements in which fission
“has occurred. Any method suitable for bringing the target material into a
clear solution may be used in principle. Since the recovery of the rare gases
must be effected from the volatile products of solution, solution procedures
which produce the more tractable gdseous products are to be desired. Meth-
ods reported in the literature for the solution of spent fuel elements and re-
covery of the rare gases therefrom55 certainly exemplify the most extreme
problems likely to be met as regards volatile products of dissolution. With
the emall targets that the radiocheémist usually encountere the feasibility of
complete initial containment of the evolved gases will often simplify matters.

Solution of metal targets is usually done with acid. If the choice is
available the products from solution with hydrochloric acid are simpler to
handle than those with nitric acid. Thus uranium is easily dissolved in con-
centrated HCIl with a few drops of 3079 HZOZ added and the HCl furnes simply
removed in a trap containing caustic. Solution of uranium in warm 92%
H3P04 has also been found satis;‘fa.cf:orys.9 and hydrogen is the only gaseous
product. Aluminum dissolves nicely in 6N NaOH, again with no gaseous prod-
ucts by hydrogen., This method of solution may be useful for aluminum catch-
er foils or as a preliminary to the solution of small targets clad in relatively -

large amounts of aluminum,

27
Solution must be carried out in a closed system or a system in which
the rare gases are removed from the effluent gas stream prior to 'venting..
The former is preferable in quantitative work as introduction of carriers and.
their exchange with the active rare gases may be effected before any chemical
or adsorption operations are performed on the gas phase, With the rare gases
one has no problems with rates of exchange between chemical states. The
problem is simply one of achieving complete physical mixing of the carrier
and active gases., As diffusion processes are relatively slow, thorough ex-
change in the gas phase is usually insured by convective mixing (as by alter-
nate cooling and heating of parts of the system). One must also consider the
complete removal of the tracer active gases from the target solution, or at
least the complete exchange of dissolved gases with carrier. Removal -;f rare.
gases from solutions by bubbling through inert carrier gases such as helium
has been studied theoretically (starting from Fick's la.w of diffusion) and ex-
perimentally. 60 It was found as expected theoretica.ll.-jr that the depletion of
tracer xenon in solution is exponential with time, and half-times for the re-
moval are of the order of a few minutes. Thus gaseous -specieé with half-
lives of more than a few minutes are often swept out of solution with helium,
Boiling the clear solutions is also effective. Efficiencies of various other
methods for removing rare gases from targets (specifically xenon from uran-
ium) have been studied. 60 Loss of xenon from uranium on the formation of
the hydride at 300°C was found to be 1% or less. Thus uranium metal may be
converted firat to the hydridé and the xenon then removed from the hydride by
some method. Amalgamation of uranium from freshly decomposed uranium
hydride and boiling, or solution of uranium hydride in aqueous AgC lO4 and
boiling are satisfactory. Uranium dissolves in saturated potassium cupri-
chloride with the evolution of a small amount of hydrogen.

The diffusion of rare gases from metals at low temperatures is gener-
ally negligible. Thus dissolver systems may generally be evacuated with the
target in place before solution is comrnenced-if thie is desirable. From uran-
ium metal containing éaseous fission products there is no detectable loss at
lesr than 1000°C, 2 At higher temperafures evolution of rare gases begins,
accompanied by swelling and cracking of the sample. Negligible diffusion .
from most metals might be expected at room temperature as the atomic di-

ameters of the heavier rare gases are often larger than interatomic distances
in metal lattices, Movement of rare gases must thus be mainly through

"microcracks" in the lattice.

A particularly neat method for removing rare gases from targets has
been reported recently. It is based on the fact that long-chain fatty acid-
salts of heavy metals emanate rare gases quantitatively and very rapidly

{more rapidly than gases can be swept fromn solutions). Removal of more

28
than 999 of the 3. 9-sec an 19

Btearate targets and barium stearate catchers were used in studies of inde-

was achieved from barium stearate. Uranyl

pendent yields in fission chains containing rare gases.

There are perhaps a few comments ta be made concerning the addition
of carriers. Amounts of carriers added for rare ga.sés of interest are usually
determined by consideration of how much gas one will need to perform sub-
sequent counting operations accurately and conveniently. Actual percentage
yields are of no primary interest assuming complete interchange of carrier
and activity. Rather, ones needs are expressed in terms of so many cc STP
yield of gas. The yields required may vary considerably from experiment to
experiment and one point in particular éhdu,ld be kept in mind when one finds
it desirable to reduce the amount of carrier to something less than amounts
previously used with success. It often happens that most of the losses in a
procedure are of the type which may be expreased in terms of a certain resid-
ual pressure left behind in a given vol-u.me in a transefer by condensation, or a
certain partial pressure of a gas in the exit stream from a trap in which the
gas is condensed, These are losses in terms of cc STP of gas and bear no
relation to the amount of carrier gas added. Thuse if one obtains 90% yield
in an experiment with 100 cc STP of carrier, and in a later experiment wish-
es to increase his specific activity by using only 20 cc STP of carrier he may
find that his yield is only.50% (10 rather than 18 cc STP). There ie essentially
a lower limit to the amount of carrier one may use without modification of the
procedure to reduce losses of the type mentioned (where these are the prin-

cipal losses),

V. COUNTING RARE GAS ACTIVITIES

Methods which have been used ta count rare gas activities may be con-
veniently divided into several categories. The choice of method will be de-
termined by the usual considerations of kind and energy of radiation, level of
activity, half-life of activity, and type of information desired,

1) Incorporation of the rare gas to be counted in the {filling of a Geiger-

 

Muller or proportional counter: The use of this technique involves one in all

 

the problems of counter construction and filling. The experimenter faced for
the first time with these problems in detail must be referred to standard texta

on the subject. 64,65

Only a few comments will be made here. This is of
course the most efficient means of counting gaseous activities as a result of
favorable geometry and absence of absorption losges,. It is thus desirable
and often necessary in the assay of low levels of activity or of active species
which emit radiations of short range (soft beta~emitters, or alpha-emitters

such as radon isotopes). Both Geiger and proportional counting have been

29
used, each having advantages and disadvantages. The electronic equipment
associated with Geiger counters is gimpler and less expensive, However,
proportional counteres do have a g-r.eater range of usefulness if one may some-
times wish to count fairly high levels of activity, since dead-times and the
resulting coincidence losses are smaller. Since performing counts in this
manner is relatively slow in any case, time s'avings are important if very
many samples must be run. The counter fillings containing the rare gas activ-
ity will of course vary in composition and some method must be found for ex-
tablishing reproducible counting conditions if the best accuracy is to be obtain-
ed. Counting conditions can be reproduced at least as accurately by using the
pulse height observed with some suitable monoenergetic radiation (as Fe
x-rays) to set high voltages and discriminator levels as by running more time
consuming plateaus. This method is applicable to both Geiger and proportion-
al countera, but is more accurate with the latter as gas: rnu;tiplications vary
more rapidly with voltage in the proportional région than do pulse heights with
voltage in the Geiger region. As usual it is much easier to calibrate this type
of counter with samples whose assay is known and to reproduce the same
counting conditions in assaying unknowns, than to attempt absolute counting.

If absolute counting must be attempted, corrections are necessary for end
effects (electric field distortion at the tube extremities) and for wal] effects
depending on the-specific ionization of the radiation in the counter filling and
the minimurmn number of ion pairs required for detection, - Counters may be
constructed with active volumes over which the field is radial by installation
of field tubes concentric with the wire and held at the proper voltage. 66, 67
General discussion of end effects and wall effects, including an approxima.te
theoretical derivation of wall effect corrections may be found in Rossi and
Staub65 (page 234, Appendix 9). End effects are generally determined empir-
ically by counting aliquots of the same sample in two tubes of identical con-
struction except for different lengths, 68 and wall effecta by a similar tech-
nique using tubes of different diameters (remémbering that the average ion-
ization produced by the radiation per unit path length, and thus the wall effect,
depends on pressure and composition of the filling).

Counter fillings may.be rmade using any of the rare gases and a suitable
quench. Positions and shapes of plateaus depend on so many factors as to
preclude brief diecussion here. Some examples appear in the literature,
but the researcher will probably find some experimentation necessary as op-
timum fillings will be determined by his electronics as well as by his counter
construction (particularly in proportional counting). Purity of materials,
especially from highly electronegative species such as oxygen which tend to
form negative ions, is important, A large variety of polyatomic gaseous

quench materials may be used. Light saturated hydrocarbons are among the

.30
most common. It might be noted that krypton and xenon purified according to
Detailed Procedure 1 are used in proportional counter fillings with no further
purification, At Lawrence Radiation Laboratory tech.fiical-gra.de is opefitane
(roughly 20 cm Hg pressure) is used as quench after drying with sodium wire.
Argon, krypton, and xenon pressures up to 20 crmn Hg have been used in con-
junction with this quench. Plateaus are geflerally better at the smaller pro-
portions of rare gas, but several cm Hg pressure is desirable to allow accu-
rate and convenient measurement of the aliquot counted. )

When low levels of Kr85 are to be counted it should be borne in mind
that the atmospheric level of Kr85, is now appreciable, In the Paris region
this level in 1954 was 1200 disintegrations /min. and in 1958 4200 d/m per
liter of krypton removed from the atmosphere. 69 One should assay Kr
in carriers used and correct for this background where necessary. In very
low level work it may be necessary to use for carrier krypton which was re-
moved from the atmosphere. at an early date. |

2) External counting in fixed or reproducible geometry: For higher

 

levels of activity of gamma-emitting or energetic beta-emitting rare gases,
the gas may be counted in 2 thin-walled container whose geometry with respect
to a2 thin-walled counter is accurately reproducible. 70 Practically such con-
tainers are limited to fairly small volumes, and overall counting efficiencies
are generally lower than with the '"fixed geometry' tubes to be described next.
As will be seen below the main advantage of this method over the use of "fixed
geometry'" tubes is that a series of samples may be counted in rapid succes-
sion on one counter (8ay by constructing containers to fit the shelves in a
counter ordinarily used for plated solid samples). 71 The effort which rn@st

be expended in calibration is thus considerably reduced. _

For best accuracy and counting efficiency the container for the gas may
be affixed permanently to 2 thin-walled counting tube, as in commercially
available thin-walled counting tubes with concentric glass jackets (R.a.dia.tioh
Counter Laboratories, Skokie, Illinois). Re.ynolda']r2 used this type of tube
for counting i(rss. Combined geometry and absorption factors resulted in an
estimated counting efficiency of 5-209% with jacket volumes of 3-10 cc and 30
rng/cm2 counter wall. Thus Kr85 samples with specific activities as low as
100 to 200 disintegrations /minute per cc STP could be accurately assayed.
These counters were used in the Geiger region in the aforementioned work.
The counter was sealed permanently into a vacuum system and provisions for
reducing the aliquot of the sample being counted by a series of expansions and
dilutions with inert carrier are described for cases where the activity '1.5 in-
itially too great,

At Lawrence Radiation Laboratory these same tubes are used for count-

ing and following the decay of a variety of rare gas activities. They are suit-

31
ably filled and used as proportional counters to reduce coincidence losses and
permit decay curves to be taken from initially higher activities. With the
particular electronics used, filling with 10 cm Hg pressure of a mixture con-
taining equal volumes of argon and isopentane gives satisfactory counter char-
acteristics,. .Absorptior_x losses vary with the energy of the beta radiation be-
ing counted, but overall efficiencies for various activities are still in the range
quoted by Reynolds. Each tube is fitted with a base which may be plugged into
the counting circuitry, and the connection to the jacket is fitted with a stop-

cock and ground-glass ball joint to allow easy attachment to and removal from
vacuum systems. To facilitate the counting of many samples in ré.pid succes-

sion, a panel into which a number of tubes may be plugged is provided, and a

switching arrangement allows rapid selection of the tube to be counted. As
regolution of fission product 2. 80-hour Kr'88 from 4.35-hour Kr85m is diffi-
cult and tedious, a few tubes were purchased with special 200 mg/cmz glass

walls to remove the softer Kr85m betas by absorption and permit simpler

accurate determination of KrBB. This brings up one special point concerning:

the counting of KrBB. It was noticed that decay curves of Kr88 exhibited scat-
tering definitely beyond statistical expectation. As the scattering became

roughly statistical when the decay reached the KrBSrn tail, and Xe133 and

Xe135 decay curves behaved statistically in the same counter it was assumed
that the behavior of the Kr88 decay curves was associated with the 18, O-min.
R.b88 daughter, As the Rb88 is expected to be ionized immediately after its

73,74

formation it was hypothesized that accumulation of local charges on the

glass walls of the jacket containing the gas was resulting.in irregular depoegi-
tion of the Rl_)a,8 on the walls. It was found that silvering the internal walls of
the jacket to provide a conducting surface encloeing the gas sample resulted in
decay curves for Krss conforming to statistics. The matter was pursued no
farther but conducfing walls are always provided for containers in which Kr88
is to be counted in high geometry.

It was statéd above that the efficiency of these counters was of the order
of 10%, but it should be noted that no effort is ordinarily made to determine
the actual efficiency accura.te.ly. The countere are calibrated for each nuclide
of interest in some suitable manner and the same counting conditions are re-
produced in subsequent counting of unknown samples. Such counting proce- .
dures are undoubtedly familiar to most radiochemists in other contexts and
are essentially no different for rare gaéea._ Thus, if one counts the KrB8 from
a U235 target in which a known number of fissions has been induced by_therm.al
neutrons, he can later determine the number of such fissions which occurred
in a U235 target irradiated with thermal neutrons by counting the KrBB (using
the same tube and counting conditions). The calculation of the calibration fac-

tor k in this case involves the number of fissions f, the activity A°, of Kr88

32
extrapolated back to a suitable zero time(the midpoint of the irradiation for
sufficiently ehort irradiations), and a suitable aliquot factor, Thus, f = kl x
A° x SCSTP carzier  _ 4y 4° X ¢c STP carrier/P

—TP._R'—WTG— ooc.' The actual volume
0°C _ .

V of the counter jacket provides no useful information in most cases and its
inclusion in the counter calibration factor without actual determination saves
considerable work. POOC is the measured préssure of the carrier in the
counter corrected to 0°C from the temperature of measurement. It is often
possible to determine the number of fissions f by assaying the target solution
for some fission product for which counters have already been calibrated

99

(e.g., Mo’ ’). Ultimately, of course, experiments involving a direct deter-
mination of fiseions in targets must have been performed. While one can
achieve accuracy sufficient for practically all purposes, the use of these count-
ers has the obvious drawback that each counter must be separately calibrated
and that counting a number of samples in a short time requires a tube for each
sample due to the time required for accurate ma.nii:ula.tion of gases.

3) Depositionl of rare gases on the cathode in a glow disecharge: Trace

 

amounts of rare gases may be deposited on wires or foils used as the cathode
in a gldw discharge chamber, Tracer rare gas activities are collected in

a glow discharge tube along with a few hundred microns pressure of some gas
to support the discharge (nitrogen, air, helium). A few hundred volts dc
potential is applied to cause a glow discharge and the discharge maintained -
for a few minutes. Deposition and retention of the gaseous activities on the
cathode is surprisingly high (in the percent range for RnZZZ)-. Heating re-
moves the activity, but it is quantitatively retained for days at room temper-
ature, The tecfi.nique has proved useful in studies of the half-lives and radia-

tions of rare gas activities, 76,77

 

4) Indirect study of rare gas activities through their daughters:
Although not a method of counting rare gases themselves, cognizance should
be taken of the feasibility of studying rare gases through counting their daugh-
ter activities.. Thus some early experiments were performed in which rare
gas fission product daughters were collected on a negatively charged wire

. coaxial with 2 moving gas stream!: 78,79

Rough half-lives for the rare gases
could be obtained from flow rates and the daughter activity as a function of
distance along the wire. Fractional chain yields have also been studied through
the relative amounts of daughters of rare gas activities found in heavy metal

' stearates (from which rare gases emanate very rapidly) and on container walls
where the decay products of rare gases leaving the stearate 'depoéit. 63 Such.
techniques may be useful where gas counting is inconvenient or where the rare
gases themselves are too short-lived to count, but the immediate or subsequent

decay pfoducts are of reasonable half-life for radiochemical assay.

33
5) Rare gas scintillation counting: It is interesting to note that the rare
80, 81, 82 The

 

gases have been found to be satisfactory scintillants in counters. _
main usefulness of such scintillants promises to be in the counting of heavy

particles such as fission fragments. They have the highly desirable property
that light output is dependent only on total energy expended {and not on ioniza-

tion density).

6) The feasibility in specific instances of counting rare gas activities.
which have grown into solid samples of their parent activities, plated for

counting, should be borne in mind. For example Xe135 and Xe133 were

studied in this manner. 83, 84

Fission product iodine was separated and plated
out for counting as Ag I or Pdi,. Radioxenon growing into the plated eamples

was retained quantitatively for weeks, even in vacuo.

VI. COLLECTION OF RADIOCHEMICAL PROCEDURES
FOR THE RARE GASES '

The largeet collection of papers dealing with rare gaa radiochemistry
of which the author is aware ia found in the last one of the nine General Ref-
erences given at the beginning of this monograph.

The following procedures were chosefi for description with a view to
calling attention to as many useful techniques and operations as possible. The
author would appreciate information concerning other procedures which should

be included in such a collection.

PROCEDURE 1 — REMOVAL OF Kr AND Xe FROM AIR
AND THEIR SUBSEQUENT SEPARATION

-Source — F. F. Momyer, Jr., et al. unpublished work at

Lawrence Radiation Laboratory, Livermore, California

Figure 1 shows a photograph of the system used in the separation. A
standard pumping arrangement consisting of a trap refrigerated in liquid
nitrogen, a mercury diffusion pump, and a mechanical forepump is used to-

® 0 107% mm Hg. Proceeding from left to right

obtain a base pressure of 10~
along the separation train let us designate and label for purposes of later
discuseion the following: 1) a flowmeter; 2) T,, a trap packed with 3/16 in.
stainless steel balls and suitable glass wool plugs to retain fine particles of
condenpates and preceded by a glass coil; 3) Cl’ a t.ra.p similar to T, but

containing 100 grams of activated charcoal; 4) CZ’ C3. and C4, three small-

34
 

in Procedures 1 and 1A

ion system used

1 —Separat

ig.

F

35
PROCEDURE ‘1 (Continued)

er traps each containing 10 grams of activated charcoal; 5) T,, an unpacked
trap similar to the preceding three; 6) a 200-cc bulb; 7) an outlet for sample
removal; and 8) a mercury manometer. Helium ie admitted to the line
through the charcoal trap at the lower left. This trap is cooled in liquid nitro-
gen to remove trace impurities from the helium. Partially obscured by this
charcoal trap is another manometer between T1 and C1 which may be cloesed

" off from the system with 2 stopcock, It is certainly not obvious from the
photograph but connections and stopcocks have been so arranged that helium
flow may be directed through any charcoal trap, through 2n ionization chamber
(fi.n’seen behind the drape), into any subsequent charcoal trap, and thence to

the atmosphere. The current through the ionization chamber is amplified by

a vibrating reed electrometer whose output drives the recorder partially seen
at the right of the bench. Thus a continuous record of activity in the ionization
chamber during a separation is obtained. To the left of the recorder is the
thermocouple vacuum gauge povirer supply and meter. The large traps a;re'
constructed of 41-mm and the small traps of 22-mm o.d. glass tubing, and
their respective lengths are about 25 and 20 cm. The charcoal is Columbia
Grade-G activated charcoal (6-14 mesh), ground in a ball mill and sieved to
vield 20-50 mesh material.

The system'is used for the separation of rare gases from various types
of samples, and the procedure described below admittedly does not treat a
problem often encountered. It was chosen to represent the extremes of
sample size, flow rates and partial pressures of rare gases for which separa-
tion has been successfully achieved with the system, and to afford a frame-
work within which the lfirgest number of operationa could be described.

The sample is 100 cubic feet (STP) of air compressed into.a bottle con-
taining 100 cc STP of active krypton and 75 cc STP of active xenon. Average
yields are about 90% for krypton and 95% for xenon. The time required for
the separation is about 8 hours of which 3 hours are spent in bleeding the
sample into the system. Before the run all charcoal traps have been heated
to 350°C and pumped on until the thermocouple vacuum gauge in the ma.nifo.ld.

is '"pinned" (probably less than 1074 mm Hg).

1. Gallon dewars of liquid nitrogen are pla.ced on T1 and C 1° including
the glass coils. The refrigerant level should be above the connection from

coil to trap.

2. Through a pressure gauge and reducing valve the sample is connect-
ed to the line below the flowmeter. Sample is bled into T, and C1 until equi-

librium pressure over C, is about 35 cm Hg. Wait a few minutes until ther-

36
PROCEDURE 1 (Continued)

mal equilibrium is reached as evidenced by steady pressure and decreased

boiling of the liquid nitrogen.

3. Usinga mechanical pump on the exit from C1 and suitable valving,

establish sample flow through T, and G, of 15 liters STP per minute at about

35 cm Hg pressure. Reduced prlessure is necessary to avoid condensation of
liquid air in the system, An NRC-4S5S Rotary Gas Ballast Pump* (4 cfm) is
used to handle the necessary large amounts of air over prolonged periods.
Keeping the liquid nitrogen replenished, bleed the sample in at the above rate
until the pressure in the bottle is about one atmosphere. Pumping for a few
minutes with all valves open completely will reduce the pressure in the bottle
to about 10 cm Hg. This represents suificiently quantitative removal of the
rare gases from the bottle.

4, Close the manometer by C, and the stopcock at the exit side of Tl’
place liquid nitrogen on CZ' C3 and C4, and introduce helium at 1-2 psi gauge
to all the charcoal traps. Thermal and pressure equilibrium in the traps will

be evidenced by cessation of helium flow into the system.

5. Set up helium flow of 1250 cc STP per minute through C,, the ion-
ization chamber, C2 in liquid nitrogen, and thence to the air. When steady"
flow ig established remove the liquid nitrogen from C 1+ Allow C, to warm
elowly in air for 5-10 minutes. (This avoids high pressures in the system
from too _rapi& desorption of the roughly 30 liters STP of air adsorbed on Cl).
After 10 minutes immerse C,in 10-20°C water. Krypton activity will be ob-
served in the ionization chamber in about 5 minutes, the activity rising rapidly
to a miximum and then dropping off more slowly (full-width at half-maximum
about 2 minutes). When the activity is lees than 19% of the peak value (10-15

minutes), the elution is halted.

6. Set up helium flow of 750 cc STP per minute through C, in liquid
nitrogen, into the ionization chamber, and thence to the air., Remove the
liquid nitrogen from G, and after 2 couple of minutes immerse C, in dry ice-
acetone slush (-78,5°C). Elute approximately ten minutes to remove air
from CZ' If activity appears in the ionization chamber before this period is

up, stop the elution immediately and proceed with the next step.

7. Direct the helium flow (750 cc/min) so that the sequence is C, in

 

* )
National Research Corporation EqQuipment Division, Newton Highlands

61, Massachusetts.

37
PROCEDURE 1 (Continued)

dry ice-acetone, the ionization chamber, C3 in liquid nitrogen, and out to the
air. Remove the dry ice-acetone from Cz and replace with 10-20°C water.
.The krypton activity will appear in 1-2 minutes in the ionization chamber,
Full-width at half-maximum is about 30 seconds. Elute until activity in the
ionization chamber is less than 1% of the peak value (should take less than ten

minutes), and stop the elution at that point.

8. Repeat Step 7 to elute krypton from (_‘,3 to C4;

9. With C4 still in liquid nitrogen, pump until the pressure on the ther-
mocouple vacuum gauge in the manifold is less than one micron. This should
require less than five minutes and results in the removal of helium from C4
with no loss of krypton. If the trap doee not pump down to one micron in less

than five minutes, discontinue pumping and proceed to the next step.

10. Place T, in liquid nitrogen. At this point C4 containe the krypton
and some residual air. The removal of air is essentially accomplished by
warming C4 elowly, passing the evolved gases through T2 in liquid nitrogen
to condense any krypton evolved, and every time the pressure reaches about
1/2 atmosphere in the éystern beyond T-2 pumping the gas off briefly by rotat-
ing the stopcock to the vacuum manifold through the open position. It should
be borne in mind that once krypton is condensed in Tz the partial pressure of
krypton in the system beyond is 2-3 mm Hg (the vapor pressure of krypton at
liquid nitrogen temperature) and that one thus wishes to keep the total time
which one pumpe on T2 to a minimum to reduce losses. Remove the liquid
nitrogen from C4 and allow the trap to warm in air a few minutes until evolu-
tion of gas begins, use the pump in the manner described above, and each
 time -evolution of gas slows down speed the procese by heating the trap to
successively higher temperatures (dry ice—é.cetone. room témperature water,
and finally warm water -- about 40°C). When the trap is at 40°C and the
pressure in the system is steady, pump off the last vestiges of air, leaving
' 2-3 mm Hg pressure of kryptan in the sysfern. (The distillation of krypton
from C4 to T2 will be seriOuély slowed by any observable pressure of residual
air}). Allow 5-10 minutes for the complete distillation of krypton. If the vol-
‘ume of the systermn has been measured beforehand, the stopcock between C4

and T, may be closed, T, warmed to room temperatfire, and the pressure of

2 2
krypton measured to get a rough determination of the yield. Yield should be
about 909%. Distill the krypton into an evacuated bulb attached to the system

outlet and cooled in liquid nitrogen, and remove from the line.

11, Place boiling water on G, and C4 and pump down to base pressure

38
PROCEDURE ) {(Continued)

to remove traces of krypton. Cool the traps to room temperature, then place

in liquid nitrogen, and admit helium to speed the cooling.

12, Set up helium flow of 1250 cc STP per minute through G, at room
temperature, the ionization chamber, C2 in liquid nitrogen and thence to the
air. The xenon will be eluted to C, fri:.u'n'C]L in boiling water in about an hour.
The time may be reduced to about 30 min. by heating C1 more rapidly in a
furnace set to reach an eventual temperature of 350°C., Elute to less than 1%

of peak activity,

13, Set up helium flow of 750 cc STP per minute through C2 in liquid

nitrogen, the ionization chamber,. C4 in liquid nitrogen, and thence to air.

Warm C, to about 40°C and throw away any amall krypton peak which appears
(less than five minutes). Elute xenon to C4 using either boiling water or a

furnace as in Step 12 (15 to 30 min. ).

14, TZ is cooled in .liquid nitrogen, C4 warmed to 100°C, residual air
pumped off if any is present, and the xenon distilled to T2 in 10-15 minutes.
Xenon losses are slight in this process as the vapor pressure at liquid nitro-
gen temperature is only about 50 microns. Measure the xenonl yield (should

be 95% or greater) and distill to an evacuated bulb for removal from the line.

Comments

l. For smaller samples in which the rare gas partial pressures are
higher over T1 the poseibility that a large fraction of the xenon may originally
be condensed in T, at -195 °C should be noted. (It seldom happens that this
occurs with krypton since its vapor pressure is higher (2-3 mm Hg) at -195°C,
but partial preesure of krypton over the trap should also be considered with
each sample). If this occurs and one wighes quantitative recovery of the xenon
one could distill xenon to ('J1 before starting the elution from C,. Itis usually
more convenient to recover the xenon from Tl and C, separately however,

The fastest way to accomplish the recovery from T1 is as follows: Cool the

' glase coil before T. in liquid nitrogen. Use an induction coil to produce eddy

currents in the ateell balls and thus heat them rapidly to well above room tem-
perature.. Pump off residual air which may have been occluded in the conden-
sate through the glass coil and then distill the xenon, carbon dioxide, and .
water into the coil. Distill xenon and carbon dioxide from the coil at -78.5°C
into a bulb attached to the outlet at the upper left of the rack and cooled in
liquid nitrogen. Thaw and refreeze the wat.er several times until occluded or
dissolved gases are removed. On another rack remove the CO, from the

~ xenon by passage through Ascarite or by distilling the xenon from a bath at

39
PROCEDURE 1 (Continued)

abour -150°C into liquid nitrogen. The vapor pressure of xenon at i1sopentane
“slush" remperature (-160°C) is only a few mm Hg so liquid isopentane at a
few °C higher temperature 18 usually used to speed the distillanon. One may
of course use dry ice-acetone on 1'1 10 remove water without condensing Kr

and Xe. Most of the carbon dioxide will then pass onto Cl'

Z. Cross-contaminanon of the krypton and xenon fractions has never
been observed in terms of their radioactivity. In one run with inactive krypton
and active xenon, the krypron was distilled directly off of C, afrer the elution
from C:l and no xenon activity was observed in th3e krypton. - A lower limat for
the separation facror in this one elunion of 5 X 10° was calculated. It thus
appears quite likely that the two subsequent elutions could be eliminated i1n

many experiments,

3. The inclusion of a thermal conducrtivity cell in series with the iomza-
tion chamber to allow detecton of nonradioactive species eluted from the char-
coal has proved to be a valuable aid in many operanons. As an example, the
elution of air from the charcoal traps contaiming Kr can be performed more
efficiently if one can observe the emergence of the air "peak'. Also, with
rare gas samples not active enough to register in the ionization chamber, one
can observe the elution peaks of the carriers in the thermal conductivity cell
rather than operating purely on the basis of previoualy observed retention

imes,

4. In a recent modificarion which promises 10 1mprove the versatility
of the system, the second small charcoal trap, Cy, was replaced with 60 grams
of 30-60 mesh 5A Linde Molecular Sieve in a coiled glass column about 1 cm
in diameter and 120 cm long. Separations of methane-krypton mixtures (typ-
ically abour 50 cc STP of each) have been performed very satisfacrtorily on
this column. Eluring the mixture from C, to the top of this column at -195°C,
and then raising the temperature 10 -36°C (ethylene dichloride *'slush-), the
following elution data with He ilow of 600 cc STP per minute were found in
several runs: Kr, peak rerention rime 18 mun., full-width at half-maximum

3 min, ; CH4, peak rerention time 35 min,, full-widrh at half-maximum 8 min.

PROCEDURE 1A — REMOVAL AND SEPARATION OF Kr AND Xe
FISSION PRODUGTS FROM U23° TARGETS

This procedure is used for U235 targets of the order of 100 mg 1n weight

and clad in a few hundred milligrams of 25 aluminum,

Even though only one of the rare gases is of interesr it 18 the pracrice to

Lo
PROCEDURE 1A (Continued)

add carriers for both to circumvent the vagaries of tracer behavior. The
amount of carrier added is determined by the usual considerations concerning
the counting operations to be performed later, but 100 cc STP each of krypton
and xenon would be typical,

The system as shown in Figure | may be used for the separation. Safety

precautions such as shielding and containers to catch active solutions in case
of breakage must of course be provided. The dissolver flask é.t the extreme
left is a 250-ml round bottom flask fitted with a 34/45 inner-tapered ground-
glaés joint, fitting an outer 34/45 joint into the top of which a separatory
funnel has been sealed. The outlet for evolved gases is through an arm at
the top fitted with an 18/9 ground-glass ball joint. The trap to the right of
the flask is packed with Ascarite.

1. A bulb containing the carriers is attached at the upper left of the
rack., The target is placed in the flask and the tapered joints greased and
fitted together. Air is evacuated from the entire dissolver eyatem. This is
permissible with most targets as the diffusion of the rare gases from uranium
at room temperature is negligible. The other alternative is to pull the air in
the s_y'stern onto Cl at -195°C. T1 is cooled in dry ice-acetone and C1 in

liquid nitrogen.

2. With the 5t0pcolck béfore Cll closeci. open the carrier bulb to allow
carriers to expand through the rest of the system. Start the solution by drip-
ping in concentrated HCl with a few drops of 30% HZOZ added. Rate of reac-
tion may be controlled by warming or cooling the flask. A wet towel contain-
ing ice may be wrapped around the neck of the flask to reflux most of the water
and acid fumes back into the solution. Inclusion of a reflux condenser cooled
with ice water ébove the dissolver is more esthetically pleasing and probably
more efficient. If hydrogen pressure approaches atmospheric it may be nec-

essary to transfer some to C. during the solution. (See Note 4 below). Dilute

1
the final few cc of solution to 20-30 cc with cold water and cool in an ice bath,
3. Close the carrier bulb to retain some carrier for flushing. Puyll
other gases onto C, to a residual few mm Hg of hydi'ogen. The solution should

boil -- warm slightly if neces_aa.fy. It was originally cooled to avoid too vig-

orous boiling and splashing during this operation.

4. Close the stopcock to C1 and open the carrier bulb. Again pull all

gases onto C 1

5. Close the stopcock to C, and admit a few cm Hg of air to the system

1
through the separatory funnel. Pull the air onto CI' {Provisions for the use

of helium for thie final flush of the system would be desirable).

41
PROCEDURE 1A (Continued)

6. Krypton, xenon, hydrogenla.nd air are now on C1 at liquid nitrogen
temperature, and one follows the previous procedure starting from Step 4.
The final steps will. be simpler as the sample contains very little air, and the
hydrogen is nearly quantitatively removed by the elution from dry ice-acetone.

7. Radiochemical analyses necessary for the purpose of the experiment

are performed on the target solution, As an example, in calibrating Krsa

99

counters one might asgay Mo’ ’ to determine the number of fissions in the
target, assuming calibration factors for this fiesion product to have been pre-

‘viously determined.

Notes

1. The system has given results in calibration runs which are indistin-
guishable (the same to within 2-3%) from those obtained with an earlier sys-
tem in which a portion of the carriers was admitted to the system prior to
‘solution, the rest bubbled through the clear solution, and the entire system

finally fluehed out through C, with helium bubbled in through the solution.

1

2. In the absence of air the system has been used to separate argon,
krypton and xenon. The argon is adsorbed along with the krypton and xenon at
liquid nitrogen temperature, and eluted off the charcoal at -78.5°C. . The ar-
gon will not be separated from nitrogen and oxygen if air is present; and in
this case éhemical means are used to pui:ify the argon.

3. In experiments involving Xe135 separations within a few hours of

135

‘irradiation, no evidence of I reaching the charcoal has ever been observed.

Resuits of the experiments performed were generally not very sensitive to
1135 contamination, however. In experiments where results would be greatly
affected by slight iodine contamination of the xenon fraction, this point should
be checked more closely. Detailed Procedures 4 and 5 contain examples of

other effective means of iodine removal.

4... In building a system for the solution of varioug types and sizes of
targets, the author would recommend in’clusiofi ahead of T1 of a trap packed-
with CuO and heated to 500°C for conversion of hydrogen to water, followed
by a trap filled with a desiccant such as magnesium perchlorate. The size
target which may be dissolved in a systerfi of convenient volume is then not
limited by the volume of hydrogen evolved, and the fact that the system contains
only the rare gases and a small amount of unburned hydrogen after dissolution
is complete simplifies the total interchange of carrier and activity as well as
the ultimé.te quantitative transfer of the carriers to C'l' Such a system. has

been used successfully for the diesolution of targete weighing up to a few grams.

4o
PROCEDURE 2 — THE EXTRACTION, PURIFICATION
85

AND INDUSTRIAL USES OF Kr
Source — E. J. Wilson, GC. Evane, J. Chadwick, J. Eakins,
K. J. Taylor, report AERE I/R 2216 (1957).

The procedure described in this paper is for the recovery of fission
product krypton from kilogram amounts of irradiated uranium- metal. A brief
description of th_e operations involved is given below to point out methods which

may be applicable to emaller scale experiments.

1, Three 2.3-kilogram uranium slugs containing fission product krypton
are dissolved in a 70-liter stainless steel container, Dissolution is accom--
plished with roughly 12 liters of 10N I-INO3

ried out "furhelessly' with respect to oxides of nitrogen bjr admitting oxygen

at 104°C. The dissolution is car-

over the solution (See Note 1). Approximately 6 liters of water are used to
recombine acid fumes and wash them back into the vessel during solution,
Solution is complete in about 20 hours and the final result is a solution of uranyl
nitrate in approximately 3N H__N03. A slight excess of oxygen is used to main-
tain flow to the purification train during dissolution and the dissolver system

is finally flushed with helium to effect complete removal of fission gases to

the pui'ification train.

2. Gases leaving the dissolver are circulated around an apparatus in
which they receive the following treatment; 1) rough drying with Ca._ClZ. 2)
carbon dioxide removed with Sofnolite {an NaOH-CaO mixture), 3) final dry-
ing with magnesium perchlorate, 4) removal of oxygen by reaction with cop-
per turnings at 600°C, 5) any water from oxidation of hydrogen or hydrocar-
bons removed with Sofnolite, "6) traces of hydrocarbons and gaseous compounds
of nitrogen removed by reaction with clea.n’uranium turnings at about 800°C,
and 7) rare gases adsorbed on charcoal at liquid nitrogen temperature. From

the charcoal trap, remaining gases (mainly helium) return to the point of entry

from the dissolver. The system is inltially filldwith one atmosphere of
helium. A loosely inflated balloon attached to the system maintaine the pres-
sure at one atmosphere durling sudden bursts of evolved gases. Circulation of
about one liter per hour is maintained by two identical parallel pumps consist-
ing of a chamber permitting unidirectional flow within which the pressure is

varied by periodic heating and cooling.

3. Upon completion of the dissolving and flushing the dissolver with
helium the charcoal trap will contain all the fission product krypton and xenon. .
Argon present as an impurity in the oxygen ueed for dissolving will also be

adsorbed on the charcoal and will in fact constitute the bulk of the adsorbed

43
PROCEDURE 2 (Continued)

gases (several hundred cc STP). The charcoal trap is removed from the line
and the rare gases desorbed by warming the charcoal in boiling water and

~ evacuating with a Toepler pump. The desorbed gases may be transferred di-
rectly to the final purification apparatus with the Toepler pump_, or if the total
volume of the rare gas fraction is very large, it may be most convenient to
transfer the gases first to a series of storage bulbs. The smaller aliquots of

rare gases contained in these bulbs may then be purified singly as convenient.

4. The final purification apparatus consists of a train of five charcoal
traps. All are heated and outgassed to a vacuum before the run and then all
but the second are cooled in liquid nitrogen. The second trap is the chromato-
graphic column. It contains 10 grams of charcoal in a column about 20 cm
long and is cooled to -50°C with a dry ice-chlorobenzene mixture. The sample
of rare gases to be separated is transferred to the first charcoal trap cooled
in liquid nitrogen. Helium is introduced to saturate all charcoals at atmos-
pheric pressure and a flow of 3.5 liters per hour of helium set up through the
train of charcoal traps. Gradual lowering of the dewar around the first trap
transfers first the bulk of the argon to the chromatographic column, and
finally upon complete removal of the dewar the remaining gases deposit in

diffuse bands at the head of the column.

5. Radioactivity in the gas leaving the column is monitored with a GM
counter suitably placed near the exit line from the charcoal. Until activity is
noted in the effluefit gas the flow is directed from the column to the last char-
coal trap (a precaution in case of an error in judgment). As soon as activity
is noted the flow is changed to pass through all three charcoal traps subsequent
to the column. Activity appears in about an hour and approaches background
in 2-3 more hours. The elution is stopfed at this point with most of the kryp-
ton removed from the column and all the xenon still adsorbed. Helium is

pumped off the charcoal trape while atill cooled in liquid nitrogen.

6. The krypton may at this point contain a few volume percent of hydro-
carbons previously undecomposed by uranium and unseparated from krypton
by charcoal chromatography. These are removed by warming the krypton-
containing charcoals in boiling water and circulating the evolved gasés over
uranium turnings at 800°C to decompose hydrocarbons, and then over pyro-
phoric uranium at room terfiperatqre to remove any hydrogefi resulting. Puri-
fication is considered complete when the preessure in the system reaches a

steady state.

7. Krypton is transferred to suitable storage bulbs with 2 Toepler pump,

3y
! PROCEDURE 2 (Continued)

or by adsorption on charcoal contained in the storage bulbs and cooled to liquid

nitrogen temperature,

Notes

1. The dissolution of uranium in nitric acid has been studied in t:lei:a.il.85

The dissolution in 11.7M HNO3 proceeds according to the following equation:
U +4.5 HNO, —~ UOZ(NOE;)Z +1.57 NO +0.84 NO, +0.0005 N,O +0.043 N, +2.25 H,

In the presence of oxygen, however, the equation for the reaction is U + 2 I-INO3 +

0.

'3/2 OZ. - UOZ(NO‘;})2 + HZO’ so that oxides of nitrogen are not present in the

products.

2. Xenon could of course be recovered from the chromatographic column
at 100°C if desired. In this particular procedure the uranium is dissolved
after xenon activities have largely decayed to avoid working with high levels of
radioactivity. There is a report describing the early recovery of multi-curie
amounts of Xel33 from uranium. 8 Aside from problems of shielding and
remote control introduced by the high levels of radioactivity, the procedure is
. very similar. The same dissolving and preliminary purification apparatus is
in fact used. As xenon is 8o much more strongly adsorbed than argon, the
argon is removed by pumping on the charcoal containing the rare gases with
the trap at -90°C, rather than by elution from a charcoal column. The final
xenon fraction contains some argon and most of the fission product krypton
"present in the sample. As the final step in the procedure (after all fhe steps
in the Kr =~ purification) trace impuritiés other than rare gases are removed
with a calcium getter. 7 The apparatus for this step consista esaentially of a
stainless steel crucible containing calcium chips and heated to 550-600°C. A
"mirror'" of calcium is deposited on an adjacent cold pyrex glass surface.
Cleanup is followed by taking periodic pressure measurements with a McLeod
gauge and is considered complete when no further pressure decrease occurs.

To obtain smaller amounts of Xe 133 (of the order of 50 millicuries),
solution of 10-20 gram amounts of irradiated uranium was carried out with
nitric acid and oxygen in a smaller system. ‘Effluent gases were passed
through a frap packed with glass rings and cooled to -80°C to remove water
vapor and NOZ' Complete removal of xenon was again effected by sweeping
the solution with helium, and the helium stream was passed through a charcoal
trap at -50°C to adsorb the xenon. Subsequent purifica.tidn of the xenon follow-
ed the same pattérn except that 0.03 cc STP of inactive xenon carrier was
added.

L5
PROCEDURE 3 — RAPID REMOVAL OF FISSION
PRODUCT Kr FROM U FOIL

Source — G. D. O'Kelley, N. H. Léza.r, and E. Eichler,
Phys. Rev, 102, 223-227 (1956)

This procedure was used to obtain samples of the 15-min, Rb'89 for
study and the operations are briefly described below as given for this purpose.
The technique for removal of krypton from the target may prove useful in other
contexts.

The target is a uranium film (0.3 rng/cmz) deposited on.a 1. 6-cm-diam- .
eter Pt disk. It is mounted at one end of a vacuum-tight recoil chamber with '
a3.12 mg/cmz' Al foil between the uranium and the recoil chamber. This
amount of aluminum absorber should stop Xe fission fragments but permit
passage of the krypton fission fragments into the recoil chamber. A rubber
serum bottle cap is used to cover and seal the exit line at the opposite end of
the recoil chamber from the ta.rget.'

' 1.- Target and recoil chamber are placed in a pilé and irradiated with
neutrons for 3 minutes. (The half-life of Kr89 is 2.6 min. )

2. The tar-get is removed rapidly from the pile via a pneumatic tube
and krypton isotopes shorter-lived than Kr89 allowed to decay out for 3 min.

3. The rubber cap sealing the recoil chamber is punctured with a
double-walled hypodermic needle, Air is flushed down the center tube of the
hypo, and air containing the krypton in the recoil chamber exits through the
coaxial outer annulus of the needle. The effluent gases are passed through a
trap packed with glas®s wool and cooled in dry ice-acetone and thence into an
evacuated cell. The Rbag daughter of ZEC:'89
charged strip of Al foil for 3 min. The foil is removed and Rb

is collected on a negatively
89 counted.
Due to straggling some xenon recoils penetrate to the recoil chamber, and a

few percent by activity of Cs 13.8 is found on the foil,

PROCEDURE 4 — RECOVERY OF FISSION PRODUCT
Xe FROM U METAL

Source — E. J. Hoagland and N. Sugarman, Paper 147:
Radiochemical Studies: The Fission Products,

 

‘National Nuclear Energy Series, Div. IV, Vol. 9,
Book 2, McGraw-Hill, New York (1?51) pp. 1035,

This procedure effects rapid dissolution of the target and separation of

iodine and xenon fission products. It was designed for studies of the indepen-

dent fission yield of Xe135.
PROCEDURE 4 (Continued)

1. The uranium target is placed in the dissolver flask, and the flask
attached to the vacuum line in which the separation is to be performed. Air
is evacuated from the entire apparatus and heliumn is then admitted to about

one half atmosphere pressure.

2. The target is dissolved in concentrated HCl containing a small
amount of I for '"holdback'' carrier., After the metal is dissolved any resid-
ual oxide or carbide may be dissclved by the addition of a small amount of
conc, HNOs. It was found that iodide ion was not oxidized during diesolution

if present originally in low concentration,

3. The liberated gases are passed in succession through a cold trap at
-85°C a conc. NaOH solution, an NaHSO,

an activated charcoal trap at -85°C, and an activated charcoal trap at -195°C.

solution, two cold traps at -85°C,

When dissolution is complete the solution is swept with about 300 cc of helium
at 10 cmn Hg over a period of 3-4 min. (See Note 1, below). The procedure
to this point can be completed. within 10 min. after the end of an irradiation.

The dissolver flask is immediately closed off.

4. All the xenon and eome of the krypton are now on the charcoal trap
at -85 °C. The remainder of the krypton is on the second charcoal trap at
-195°C. In this procedure the first charcoal trap is heated to 250°C and the
desorbed gases (containing all the xenon) transferred By’ a Toepler pump to an
evacuated storage bulb. Aliquots of xenon for counting are té.ken by filling
thin-window gas-tight containers from the storage bulb.

5. The amount of Xe135 growing into the target solution from 1135 may

be determined by repeating the helium sweeping at a2 suitable known later time
after irradiation and counting an aliquot of the xenon fraction so obtained.
Early experiments indicated no I135 reached the firét cold trap, NaOH solu-
tion, or NaHSO, solution. Thue the helium sweep containing the xenon activ-
ity may be transferred directly to the storage bulb, and aliquots for counting

taken.

Note

 

1. The double-ended dissolver flask used in this procedure is illustrat-
ed in the article. The closed end ie used for solution of the target and storage.
For sweeping with helium the flask may be rotated through 180° about the |
horizontal axis of the sidearm qpnnecting it to the system through a tapered
glass joint, The solution then rests on a sintered glass disk through which

helium may be introduced. A emall bulb between two stopcocks below the

b7
PROCEDURE 4 (Continued)

sintered glass disk is used to measure out aliquots of helium for sweeping.

Efficiency of xenon removal by the helium sweep wasa found to be about 95%.

PROCEDURE 5 — RAPID REMOVAL OF FISSION PRODUCT
Xe FROM UO, OR UO,(NO,), - 6H,0 TARGETS

Source — F. Brown and L. Yaffe, Can. J. Chem. él, 242 (1953).

This procedure was also designed for studies of the independent yield
of Xe135. The targets used were 200-300 mg of U0, or U0, (NO,), - 6H20
enclosed in thin-walled aluminum capsules. _

The target is placed in a flask containing about 50 mg of KI as holdback
carrier, attached to the vacuum system, and the apparatus evacuated. H2 is
started bubbling through the solution and hot 1:1 sulfuric acid admitted to
dissolvej the capsule and contents rapidly. The solution is boiled while hydro-
gen is bubbled through. Effluent gases pase through a reflux water condenser
for rough drying, a trap cooled to dry-ice temperature, a trap containing
NaQH pellets, and two charcoal traps at dry-ice temperature in successioén.

)

traps are evacuated to less than 5-microns pressure and sealed off with cool-

bubbling is continued until 30 min. after the end of irradiation. Charcoal

ant still in place. Kr and Xe activities are on the first trap, and Kr but no
Xe activity on the second. The xenon activity.is coufited by placing the char-
coal trap in a well-type ionization chamber. Later samples of xenon may be
obtained for determination of the xenon activity growing into the solution by

repeating the hydrogen sw'eep.

PROCEDURE 6 — SEPARATION OF LONG-LIVED FISSION GASES

Source —J. A. Ayres and I. B. Johns, Paper 311:

Radiochernical Studies: The Fission Products,

 

National Nuclear Energy Series, Div. IV, Vol.9,
Book 3, McGraw-Hill, New York (1951), pp. 1763.

This procedure was used in studies of the efficiencies of various methods
of xefion removal from urapium metal, and gave results reproducible to 2-3%,

The uranium sample containing fission product xenon is placed in a dis-
tilling flagk, and the flask connected to the purification train. Air is removed
by flushing the entire apparatus with COZ' The uranium is then diseolved by

running in 92% H,PO,, the rate of reaction being controlled by warming or

48
PROCEDURE 6 (Continued)

cooling the flask as retju.ired. Dissolution is very fast at the boiling point of
the acid. Xenon, 'hy"drog.e-n. and some fission product iodine and tellurium
enter the pufification train ‘Bubbling the gases through a saturated NaH'SO3
solution removes I and Te Pa.saa.ge over hot CuO ox.ldlzes the H, to HZO
which is c0ndensed in a subsequent flask, The active xenon and small amounts
of other gases are then collected in a eudiometer over KOH. After the uran-
ium is dissolved, any xenon remaining behind is swept from the solution into
the eud:Lometer with CO After the CO2 is removed by reactlon with KOH,
air may be swept in if neceesary to prov1de an easily handled volume of gas.
The sample is withdrawn to an evacuated bulb connected to a counting cha.mber
into which an Al finger-type GM counter is sealed. The aliquot of the xenon
being countéd may be calculated from knowledge of the volume of the counting

chamber and of the rest of the system containing the sample,

PROCEDURE 7 — REMOVAL OF Rn FROM Th TARGETS AND ITS
COLLECTION ON THE CATHODE OF A DISCHARGE TUBE

Source —F. F. Momyer and E. K.  Hyde,"
Journal of Inorganic and Nuclear Chemistry,
1, 274-95 (1955).

The overall yield of this procedure is no more than a few percent, It
has proved useful, however, for the collection and retention of rare gases on
wires or foils in studies of decay characteristice of rare gas n_L_lclide_s._.

. The thorium metal target, irradiated with 340-Mev prpto.ns_ to fn-oduc_e .
radon isotopes by spallation, is placed in the disSQIQer flagk, the eystern
evacuated, and hot co'ncentrated HCI! containing ammqnium__ fl_qoe_ilicate ca.taLyst
dropped in to dissolve the thorium. Evolved gaees er_e bfil_:bl_e‘d_ through sodium
hydroxide solution, passed through a trep_copled in dry i,ce/a.cet_one, and t_h_en-
paseed through 3 successive trape cooled in liqnid nitrogen. After soluticn
is complete the dissolver and sodl.urn hydroxide solution are closed off and the
4 cold traps pumped down to 10 - 10 -5 mm Hg. The traps a.:_.-e_shu__t off from
the pump and gases condensed in the traps are distilled to the last trap by .
keeping it immersed in liquid nitrogen and warzning the others. The contents
of this trap are finally distilled into a tip on the glow discharge tube which is
cooled in liquid nitrogen. The discharge tube is then closed off, the tip
warmed, and a glow diecharge run for a few minutes to collect radon on the
cathode, This technique has been discussed earlier in this monograi:h under

counting procedures. Usually, small amounts of condensed impurities pro-

49
PROCEDURE 7 (Continued)

vide a few hundred micflrons pressure to support the 'discha..r.ge. If this is not
the case, one may add some gas such as air, nitrogen or helium to achieve
the necessary pressui-e in the discharge tube. Any radon not deposited on the
cathode may be frozen back into the last trap at liquid nitrogen temperature

for use in later experiments.

Notes

1, Stoner and Hyde76

studied isotopes of radon produced by bombard-
ment of gold with nitrogen ions in similar fashion., The gold was dissolved in
mercury while being heated by induction to remove the radon from the target.

2., Mathur and Hyde77

by bombardment of KI with 100-Mev protons using a similar procedure. The

studied neutron deficient xenon isotopes produced

KI was dissolved with water in a closed system and the xenon removed from
solution by pumping through the same series of traps used in the radon pro-

cedure. The NaOH scrubbing was omitted.

3. The observed retention of trace amounts of rare gases in traps at
-195°C should be commented upon, It was also observed in the radon experi-
ments that fission product rare gases and their interfering P-y activity could
be largely removed from radon by warming the trap containing all the gases
briefly in air and distilling the fiasion products.into another trap at -195°C.
Distillation was monitored with a B-y survey meter and continued until the
activity in the trap ceased to drop rapidly (about 30 sec). Well over 90% of
the rare gas fission product activity in the radon could be removed thus., The
vapor pressures of the rare gasea at -195°C certainly exceed the partial pres-
sures of the tracer amounts of rare gases in the above experiments. Retention
in the cold traps is thus not due to condensation of the pure solids. Speculation
as to the exact mechanism involved seems'pointlesé without further experimen-
tal data. It seems worth calling attention to the fact that the retention is not
a result of processes involving the rare gases alone (that is, édsorption on or
condensation with other substances must be involved), so that circumstances
under which trace xenon, for example, might not be held up inla portion of a

system at -195°C are poesible.
PROCEDURE 8 —- DETERMINATION OF ACTIVE GAS HALF -LIVES
BY THE CHARGED WIRE TECHNIQUE (O)

Source — C. R, Dillard, R. M. Adams, H. Finston, and A. Turkevich,

Paper 68: Radiochemical Studies: The Fission Products,

 

National Nuclear Energy Series, Div. IV, Vol.9, Book 2,
McGraw-Hill, New York (1951}, pp. 624.

A number of the earlier experiments to determine the half-lives of
short-lived rare gases were performed using this interesting technique.

A solution of uranyl nitrate is contained in a small polystyrene vessel -
over a porous graphite disk through which nitrogen may be bubbled. The
vessel is connected through a baffle system, to prevent spray carry-over, to
a 10-ft aluminum tube of 1-in, inner diameter. A fine copper wire down the
center of the tube is maintained at -1000 volte to effect collection of the daugh-
ters of gaseous activities passing through the tube. The entire apparatus is
placed in a reactor and the solution irradiated with neutrons while maintaining
a constant flow of nitrogen through the system, From experimental data the
linear flow rate of gas through the aluminum tube may be calculated. Thus by
cutting the collector wire into a number of equal lengths and determining the
relative amounts of daughter activity on each piece, the half-life of a short-
lived rare gas with a daughter of appreciable hali-life may be determined with

fair accuracy.

51
REFERENCES

lH Remy, Treatise on Inorganic Chemistry, Vol. (tra.nala.ted by J. S.
AnderBOn), Elsev-ler Publishing Co., New York, (1956)

2 M. Steinberg and B. Manowitz, Ind, Eng. Chem., 51, 47-50 (1959),
34, M. Powell, J. Chem Soc. London, p. 298, 300, 468 (1950).
“H. M. Powell, J. Chem. Soc. London, Part 2, 2658 (1954).

5D.I.J Chleck and C, A. Ziegler, Nucleomcs 17, No. 9,.130 (1959).
E’_s.'Katcofl and W. Rubinson, Phys. Rev, 91, 1458-69 (1953).

"F. Soddy, Proc. Roy. Soc. A78, 429 (1906).

BA'. S. Newton, AEC report series M.D.D.C. 724. : .
-9W. D. Bond and W. E. Clark, "Reduction. of CuO by Hydrogen. II,

. Conversion of Hydrogen to Water over Fixed Beds." OQOak Ridge National
Laboratory report ORNL-2816 (1960).
10F. R. Bruce, J. M. Fletcher, H. H. Hyman, and J. J. Katz, editors,

‘Process Chemistry, Progress in Nuclear Energy, Series III, McGraw-

 

Hill, New York, or Pergamon Press, London, (1956), p. 345.

1. Brown and L. Yaffe, Can. I. Chem. 31, 242-9 (1953),

12_0. E. Myers and R. J. Prestwood, in Radioactivity Applied to Chemistry,
A. C. Wahl and N. Bonner, Editors, John Wiley and Sons, Inc. New York,
(1951, pp. 28-29.

13E. 3. Wilson, Vacuum 3, 287-88 (1953).

YN, s. Corney, et al., AERE-C/M-364, Feb, 1959.

15 A. Sayres and G. 5. Wu, Rev. Sci. Instr. 28, 758-64 (1957).

16N. V. Sidgwick, The Chemical Elements and Their Compounds, Vol. I,
Oxford, Clarendon Press (1950), p. 1.

17F. 7. Allen and P. B. Moore, JAGCS 53, 2512-22 (1931).

185, J. Metzger, Ind. Eng. Chem. 27, 112-16 (1935). | .

1‘;M. Ruhemann, Separation of Gases, 2nd Edition, Oxford, Clarendon

Press (1949).
OG. H. Ca.dy and H. P. Cady, Ind. Eng. Chem., Anal. Ed. 1__7, 760-6 (1945).

52
21E. Glueckauf and G, P. Kitt, Proc. Roy. Soc. (London) A234, 557-65

{1956}. _ , .

Handbook of Chemwtand Phys1cs 39th Edition, _1957'-5_8,i Ghernicql
Rubber Publishing Co., Cleveland, Ohio. '

3S Dushman, Vacuum Technlque, .]'oh.n W1ley a.nd SDnS, New York, (1949)

24R E. Dodd and P. L. Robinson, "Exmanmental In0j:1c Chemlstry,”

Elsevier Publishing Co., New York, (1954).

R. S. Sanderson, Vacuum Manlgulatlon of Volatile Compou.nds. John

Wiley and Sons, New York,(1948). ' ‘ .

I. Langmuir, JACS 40, 1361, (1918); also Chem. Reva. 13, 147 (1933),

S ‘Brunauer, P, H. Emmett and E, Teller, JACS 60, 309 (1938).

28K Peters and K. Weil. Z 'Physik, Chem 148(A ), 1-26 (1930) available

ag AEC-tr-2470 (35 pa.ges)

S. Brunauer, The Adsorption of Gases and Vapors, Vol. 1, "Physical

22

 

 

 

25

_26

29

 

Adsorption, " Pnnceton, Pr1nceton Umvermty Press Pr1nceton. 'N.J. .
(1943). | | | o
OW. F. Wolff, J. Phys. Chem. 62, 829-33 (1958).
31'.'! J. .Kiplling,. Q'ua.rterly Revie'v'rs (Chefi Soc. London) 10, 1- 26 (1956),
32V R. Deitz, "Bibliograpry of Solid Adsorbents, 1900 1942 " National
Bureau of Sta.ndarda, Washington, D.C., 1944.
33R A, Ladd and T. F. Young, Paper 317: Rad1ochem1ca.1 Stud:l.es The
Fission Products, Na.tlonal Nuclear Energy Serles, D1v IV Vol.9,
Book 3, McGraw-Hill, New York (1951) pp. 1833, _
E. Glueckauf, Proc. Roy, Soc. (_Lond_on) Al185, 98 (1946).
W. J. Arrol, K. F. Chackett, and §. Epstein, Gan. J. Res. 27B, 757-63
(1949), | | S | o :
A, I. M. Keulemé.ns, Gas Chro‘maf’ogi;a;ph)r_, 2nd Edition, .Reinhold Pub-
lishing Corp., New York (1959). o
J. N. Wilson, JACS 62, 1583 (1940).
D. de Vault, JACS 65, 532 (1943).
A J. P. Martm a.ndR L M Synge, B1ochem J. (London) 35, 1358
(1941)
405, W. Mayer and E. R. 'rompkma, JACS 69, 2866 (1947).
41E Glueckauf, Trans. Far. Soc. 51, 34 (1955),
L Lapidus and N, R. Amundson, J. Phys. Chem 56, 984(1952)
43, E. Glueckauf, "Principles of Operatmn of Ion Excha.nge Column, ' paper
read at the Conference on Ion Exchange, London, 5-7th Apnl 1954.
44 N. N. Tunitskii, Doklady Akad. Nauk. §.5.5.R. (Compt. Tend. acad.
sci. URSS) 99, 577 (1954). ' _
J. C. Giddings and H. Eyring, J. Phys. Chem. 59, 416 (1955).

 

34
35

36

 

37
38
39_

45

53
46.'.[. J. van Deemter, F. J. Zuiderweg, and A, Klinkenberg, Chem. Eng,

Sci, 5, 271 (1956).

A. Klinkenberg and F. Sjenitzer, Chem, Eng. Sci. 5, 258 {1956).

E. Glueckauf, AERE C/R 2435 (4) ('1958).

. *9C. B. Amphlett and B. F. Greenfield, AERE-G/R-2632 (July 1958).

50R' E. Adams, W. E. Browning, Jr. and R, D. Ackley, Ind. Eng. Chem.
51, 1467-70 (1959).

47
48

51W E. Browning and C. C. Bolta, ORNL-2116 (1956),

2R. C. Koch and G. F. Grandy, Nucleonics 18, No. 7, 76-80 {1960).

>3 A. L. Lindsay and L. A. Bromley, Ind. Eng. Chem. 42, 1508 (1950).
5";'E R. Weaver, Physical Methods in Chemical Analysis, W. G._.\Berl, ed.,

 

Vol. II, Academic Press, Inc., New York (1951), pp. 387-437,

E. J. Wilson, C. Evans, J. Chadwick, J. Eakins and K, J. Taylor,
AERE I/R 2216 (1957).

R. M. Barrer and A, B. Robins, Trans. Far. Soc. 49, 807 (1953}
D. W. Breck, W. G. Eversole, R. M Milton, T. B. Reed and T. L.
Thomas, JACS 78, 5963-77 (1956). '

M. Krejci, K. Tesarik and J Janak, 2nd Biannual International Gas

55

56
57

58

Chromatography Symposium (1959) Vol. 2, p. 105, Instrument Society
-0f America, Pittsburgh, Pennsylvania,
59.1 A. Ayres and I, B. Johns, Paper 311: Radiochemical Studies: The
Fission Products. National Nuclear Energy Series, Div, IV, Vol. 9.
Book 3, McGraw-Hill, New York (1951), pp. 1763.
60Reference 59, Paper 316, pp. 1823,
61W. J. Arrol, K. F. Chackett and S. Epstein, Can,. J, Res. 27B, 757-63 _
(1949). | |
M. B. Reynolds, Nuc. Sci. and Eng. 1, 374-84 (1956).
63A. C. Wahl, J, Inorg. and Nuc. Chem. 6, 263-77 (1958).
645. A. Korff, Electron and Nuclear Counters, Van Nostrand Inc. New York
(1946). . '
5B. 'B. Rossi and H. H. Staub, lonization Chambers and Counters, National
Nuclear Energy Series, Div. V, Vol. 2, McGraw-Hill, New York (1949).
C. P. Sikkema, Nuclear Instr. 1, 148-51 {1957).
A, L, Cockcroft and S. C, Curran, Rev. Sci. Instr. _Z_E. 37 (1951).
W. B. Mann and G. B. Parkinson, Rev. Sci. Instr. 20, 41 (1949).
G. Delibrias and C. Jehanno, Bull. inform. sci. et tech. (Pa.ria)'No'. 30,
14-15(1959) (in French) _ ‘
OF D. Rosen and W, Davis, Jr. K-678 (Nov, 1950).
7l Reference 59, Book 2, Paper 150, pp. 1052,
72 M. B. Reynolds, KAPL-1201 (Sept. 1954).

 

62

 

66
67
68
69

Sh
13 A. Wattenberg and W. H. McCorkle, editors, Phyeics progress report
for Jan., Feb., and March 1950, Argonne National Laboratory, ANL-4437.
74 o, H, Snell and F. Pleasonton, Phys. Rev. 107,-740-5(1957). |
7517'. F. Momyer, Jr. and E. K, Hyde, J. Inorg.and Nuc. Chem 1, 274-95
(1955). _ '
76A. W. Stoner and E, K. Hyde, I. Inorg. and Nuc. Chem 4, 77-83 {1957).
"TH. B. Mathur and E. K. Hyde, Phys. Rev. 96, 126 (1954).

78R. Overstreet and L. Jacobson, Paper 67: Radiochemical Studies: The

 

Fiseion Products, National Nuclear Energy Series, Div. IV, Vol. 9,

Book 2, McGraw-Hill, New York (1951), pp. 621.

Ibid, Paper 68, pp. 624.

J. A, Northrop and Judith C. Gureky, Nuc. Instr. 3, 207-12 (1958).

81J. A. Northrop, J. C. Gursky and A. E. Johnsrud, IRE Trans. Nuc. Sci.,
NS-5, 81-7 (1958).

82 A. Sayres and C. S. Wu, Rev. Sci. Instr. 28, No. 10, 758-64 (1957).

83 Reference 78, Paper 139, pp. 984.

841'bid, Paper 141, pp. 1005,

' _85 Reference 10, pages 85, 97, and 356,

86y, p. Dibbs, J. D. Eakins and E. J. Wilson AERE I/R 2224 (1957).

79
80

USAEC OFfica of Technlcal Information Extersian, Qok Kidge, Tenmenes 55