NAS-NS-3016.txt

National

 

Academy
of
Sciences

National Research Council

NUCLEAR SCIENCE SERIES

The Radiochemistry
of Protactinium

4BV LT
/' Commission 9

 
COMMITTEE ON NUCLEAR SCIENCE

L. F. CURTISS, Chairman
National Bureau of Standards

ROBLEY D. EVANS, Vice Chairman
Massachusetts Institute of Technology

J. A. DeJUREN, Secretary
Westinghouse Electric Corporation

C. J. BORKOWSKI
Oak Ridge National Laboratory

ROBERT G. COCHRAN
Texas Agricultural and Mechanical
College

SAMUEL EPSTEIN
California Institute of Technology

U. FANO
National Bureau of Standards

HERBERT GOLDSTEIN
Nuclear Development Corporation of
America

J. W. IRVINE, JR.
Massachusetts Institute of Technology

E. D. KLEMA
Northwestern University

W. WAYNE MEINKE
University of Michigan

J. J. NICKSON
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 McMILLEN
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

GEORGE A. COWAN
Los Alamos Scientific Laboratory

ARTHUR W. FAIRHALL
University of Washington

JEROME HUDIS
Brookhaven National Laboratory

EARL HYDE
University of California (Berkeley)

HAROLD KIRBY
Mound Laboratory

GEORGE LEDDICOTTE
Ozk Ridge National Laboratory

JULIAN NIELSEN
Hanford Laboratories

ELLIS P. STEINBERG
Argonne National Laboratory

PETER C. STEVENSON
University of California (Livermore)

LEO YAFFE
McGill University

CONSULTANTS

NATHAN BALLCU
Centre d’Etude de I’Energie Nucleaire
Mol~Donk, Belgium

JAMES DeVOE
University of Michigan

WILLIAM MARLOW
National Bureau of Standards
CHEMISTRY-—RADIATION AND RADIOCHEMISTRY

The Radiochemistry of Protactinium
By H. W. KIRBY

Mound Laboratory

Operated by
Monsanto Chemical Company

Miamisburg, Ohio

December 1959

Subcommittee on Radiochemistry
National Academy of Sciences —National Research Council

Printed in USA. Price $1.00. Available from the Office of Technical
SBervicea, Department of Commerce, Washington 25, D. C.
FOREWORD

The Subcommittee on Radlochemistry is one of a number of
subcommittees working under the Committee on Nuclear Science
within the National Academy of Sciences - Natlonal Research
Council. Its members represent government, industrial, and
‘university laboratories in the areas of nuclear chemistry
and analytical chemistry.

The Subcommittee has concerned itself with those areas of
nuclear science which involve the chemist, such as the col-
lection and distribution of radiochemical procedures, the
egtablishment of specificatlons for radiochemically pure
reagents, the problems of stockpiling uncontaminated
materials, the availability of cyclotron time for service
irradiations, the place of radiochemistry in the undergraduate
college program, etc.

This serles of monographs has grown out of the need for up-
to-date compilatlons of radiochemical information and pro-
cedures.. e Subcommittee has endeavored to present a
series which will be of maximum use to the working sclentist
and which contains the latest avallable information. Each
monograph collects in one volume the pertinent information
required for radiochemlcal work with an individual element
or a group of closely related elements.

An expert in the radiochemlstry of the particular element
has written the monograph, following a standard format
developed by the Subcommittee. The Atomlc Energy Commis-
sion has sponsored the printing of the seriles.

The Subcommittee is confident these publications will be
useful not only to the radiochemist but also to the research
worker in other fields such as physics, biochemlstry or
medicine who wlshes to use radiochemical techniques to solve
a specific problem.

W. Wayne Melnke, Chairman
Subcommittee. on Radiochemistry

111
INTRODUCTION

This volume which deals with the radiochemistry of grot-
actinium is one of a series of monographs on radiochemistry
of the elements. There is included a review of the nuclear
and chemical features of particular interest to the radio-
chemist; a dlscussion of problems of dissolution of a sample
and counting techniques, and finally, a collectlon of radio-
chemical procedures for the element as found in the literature.

The series of monographs will cover all elements for which
radiochemical procedures are pertinent. Plans include
revislon of the monograph perlodically as new techniques and
procedures warrant. The reader is therefore encouraged to
call to the attention of the author any published or unpub-
lished material on the radiochemistry of protactinium which
might be included in a revised version of the monograph.

Any new review on protactinium at this time has been rendered
largely superfluous by the recent§publication of the critical
article by Haissinsky and Bouissieres!. This excellent
comprehensive monograph covers the published (and much of the
unpublished)literature through November 1, 1957. Except for
translation from the French, 1t can hardly be ilmproved upon.
Such a translation has been made;, and it is hoped that per-
mission for its general distribution will be granted by the
publisher in the near future. : '

The present regort has the limited objective of acquainting
the reader with the brecad outllines of protactinium chemistry,
especially in relation to methods of preparation, separation,
and analysis. The literature survey gas been limiteg to
materlal reported since about 1950; for the older literature,
the author fias relled heavily on the Hailssinsky-Boulissieres
review. Critical comments, however, are those of this
writer. Patents, as such,; have been largely ignored, as in
the oplnion of the author, thelr significance is more legal
than sclentific. .

iv
I.

II.

I1I.

Iv.

CONTENTS

GENERAL REVIEWS OF THE CHEMISTRY OF

PRDTACTINIUM S0 wd o doee oot o tdygdode bt B IReben
ISOTOPES OF PROTACTINIUM ccevscvevvscncancas

CmSTRY OF PROTACTINIUM [ B BB B BN B B BN BN B BN BN BN BN BN BN BN
1. General

2. Metallic Protactinium

3. Soluble Salts of Protactinium

., Inasoluble Salts Useful in Separation
and Analysls

5. Coprecipication and Carrying of
Protactinium

6. Solvent Extraction of Protactinium
7. Ion Exchange Behavior of Protactinium -
8. Miscellany

Paper Chromatography

Electrochemistry

Spectrophotometry

Dry Chemistry

DISSOLUTION OF PROTACTINIUM SAMPLES ........

COWTIHG TECmIQUES o 9 8 & P & ® 09 3 8 P AaAOC S S ORI E " ES
l. Protactinium-233

2. Protactinium-231

10

12
13
7
23
23
26

26

28 -

28

30
30
32
VI.

VII..

DETAILED RADIOCHEMICAL PROCEDURES  FOR
PRDTACTINIIJM # 5 8 8 8 O F S0P P OF O ET SR A TS A e

A.

E.

Preparation of Carrler-Free Prot-
zgtinium-ij (Procedure 1 through

Determination of Protactinium
(Procedures 5 through 10)

Speclal Preparations of Protactinium
(Procedures 11 through 13)

Separations of Protactinium
(Procedures 14 through 15)

Urinalysis of Protactinium
(Procedures 16 through 17)

APPENDIK .......‘.II.I.'I.II...IO..II.lI.lIl.

Summary of the Protactinium Project at
Mound Laboratory

vi

37

37
41
54
57

59
65
The Radiochemistry of Protactinium*

By H. W. Kirby
Mound Isaboratory
Operated by Moneanto Chemical Company
Miamisburg, Ohlo
December 1959

I. GENERAL REVIEWS OF THE CHEMISTRY OF PROTACTINIUM

Hfizssinsky,'u. and Bouissffires, G., Protactinium, Nouveau

 

Traite de Chimie Mifiérale, XII, pp. 617-680, ed. by P. Pascal;
Masson ‘et Cie., Paris (1953). (This should be required read-
ing for anyone working with'protactinium. The bibliography,

containing 165 references, is complete to November 1, 1957.)

Salutsky, M. L., Protactinium, Comprehensive Analytical Chem-
istry, Vol. I, Chapter IV, Section 44, 11 pp.,-ed. by Cecil
L. Wilson; Elsevier Publishing Co., Amsterdam (Ifi'Press).

(Primarily devoted to analytical aspects. References, 49.)

* ' : - Permission for the use of copyrighted

material has been kindly granted by Pergamon Press, publishers
of the Journal of Inorganic and Nuclear Chemistry, and by
the editor of the Journal of the American Chemical Society.
Katzin, L. I., editor, Production and Separatlon of U233°
Collected Papers, U. S. Atomic Energy Comm., TID-5223, 728 pp.
in 2 vols. (1952). Avallable at $3.25 from Office of Tech-
nical Services, Dept. of Commerce, Washington 25, D. C. (79
papers devoted to thorium, protactinium, and uranium
chemistry, radlochemistry, separations, and nuclear charact-
eristics. Not a true review, but a valuable collection of |

research papers and data.)

Gmelins Handbuch der anorganischen Chemie, Protactinium und
Isotope, System Number 51, 99 pp., Verlag Chemle, G.m.b.H.,
Berlin (1942). (Reviews the literature to 1940.)

Elson, R. E., The Chemistry of Protactinium, The Actinide

. Elements, Chapter 5, pp. 103-129, National Nuclear'Energy
Series, Division IV, Plutonium Project Record, Vol. 1A, ed.
by G. T. Seaborg and J. J. Katz, MchEw,HiLl Book Co., New
York (1954). (Referefices, 69, the latest original reference

being dated 1951.)

Hyde, E. K., Radiochemical Separations of thé Actinide Elements,
Ibid., Chapter 15, pp. 542-95.

Literature Survey on: 1. The Chemistry of Actinium and Prota-
actinlum -- Especlally in Aqueous Solutions. 2. Determinatiom
of Actinium and Protactinium. 3. Technical Information on
Radium Industry Residues. Anon., Atomic Energy Commission,

Tel-Aviv, Israel, LS-6, 34 pp. (Sept., 1958).
IT. TABLE I

ISOTOPES OF PROTACTINIUMS

 

 

Mass Mode of Decay Half-Life Source

225 a 2.0 sec Th + d
226 a 1.8 min Th + a
227 a ~ 85% 38.3 min Th + d, U + a, daughter Np231

EC w 15%
228 EC ~ 98% 22 hrs Th + d, daughter Y228

aw 2%
229 EC - 99+% 1.5 days Th?30 4 4

a = 0.25%
230 EC ~ 85% 17.7 days Th + d, Pa23l 4+ 4, Th230 4 g

B~ ~15% ' '

a?, pt?
231 (Ppa) Q 32,500 yrs h230 . n, Th + n, descendigg

| | U

232 B" 1.31 days Th + d pa23l 4 n, Th + o
233 - 27.0 days Th +d, Th+n
234™(UXy) B~ - 99+% 1.175 min Descendant U238 (metastable)

IT - 0.63% -

- | 234M

234 (UZ) B 6.66 hrs Daughter Pa
235 8™ 23.7 min U+ p, U+ d, daughter ThZ>>
237 g~ 11 min U+ d
EC - electron capture P - proton
IT -~ 1iscmeric transitior n - neutron
d - deuteron

III. CHEMISTRY OF PROTACTINIUM

1. General

The only naturally occurring protactinium isotopes are prot-

actinium-231 and protactinium-234.

Because of their short
half-11ives, 1.2-minute protactinium-234 (UXp) and its 6.7-hour
isomer (UZ) fire of relatively little interest to radiochemists.
The 27-day protactinium-233 is readily produced by neutron |
irradiation of thorium and is of considerable value both as

a tracer and as the parent of fissile uranium-233. Signifi-
cantly, the v-p branching ratio of UXp was determined with the

aid of protactinium-233 as a tracer.9

Al though the natural sbundance of protactinium is almost as
great as that of radium, the known world supply of the iso-
lated element and its compounds did not exceed one or two

grams until the late 1950's.

Interest in thorium breeder reactors gave necessary lmpetus
to the recovery of gram quantities of protactinium for study

of its macrochemistry. .

Abundant source material was avallable as a result of the
accelerated prodfiction of uranium for nuclear reactors.
Chemical technology (solvent extraction) and Instrumentation
(scintillation spectrometers) had advanced to a point where
éeparations and analyses which were previously difficult or

impossible could be made with relative ease and rapidity.

Almost 100 grams of protactinium-231 was recently isolated
from uranium fefinery wastes In Great Britain, and groups at
Cambridge, Harwell, and elsewhere are actively engaged'in
studying the chemistry (as distinct from the radiochemistry)

of protactinium.
Dr. A. G. Maddock has kindly supplied this reviewer with an
unpublished report on the 1958 activities of the Cambridge
protaétinium grouplo, in which he describes the preparation,
in centigram quantities, of the penta- and tetrahalides of
protactinium and polarographlc studies of its oxalate,
chloride, fluoride, and sulfate solutions. These data will

be published shortly.

It is safe to predict that the 1960's will see much of the
mystery and wlitchcraft elimingted from protactinium chemistry.
Nevertheless, the greatest amount of information published to
date has come from work done with protactinium-233 on tracer

levels.

These data have not been consistently applicable to the
macrochemlstry of protactinium-231. To some extent, the
discrepancles are due to the origins of the isotopes. Prot-
actinium-233 1s likely to be contaminated 6n1y by thorlium.
Protactinium-231, on the other hand, may be contaminated
with any or all of the elements in groups IVa and Va of
the Period;c Table, as well as with phOSphate ion, which
tends to make its chemistry somewhat erratic. Furthermore,
radiochemical analysls of protactinium-231 is complicated

by thé presence of its own a-, B-, and y-active descendants

(Table I1) and those of uranium-238.

Protactinium-231 can also be produced by the neutron irradia-
tion of thorium-230 (ionium), in which case it will usually

also be dontaminated by thorium-232 and protactinium-233.

Too much has been made of the apparently capricious chemlcal

behavior of protactinium, In the pentévalent state, its

5
TABLE II. URANIUM-ACTINIUM SERIES {(4n + 3)

 

. 1sotope Synonym Mode of Decay Energles, Mev. Half Life

 

 

y235 AcU a 204 4.6 ' 8
. . 8.8 x 10°Y
1 80% 4.4
Th231 1) p- 0.2 25.65 H
| oo
25% 4.9
pa23l . a 3% 4.8 92,500 Y
Lz 13% 4.7
Ac22777 [; - 0.02 -
' 22.0 Y
| 248 6.1
m227 | 2 o
i RdAe a 7% 5.9 18.6 D
! 25% 5.8
! 22% 5.7
Fre23e  aox B- 1.2 21 M
(555 2
Ra 223 AcK o 22 5:3 11.2 D
7% 5.
e
Rn21? An a 124 6.4 3.92 8
1 | 4% 6.2
a 7.4
Po?lo._  AcA [; 0.00183 S
1 ' 0.,0005% B~) (?)
| 20% 0.5
b2ll | AcB B- . 36.1 M
' 80% 1.4
1
At ¢ ... @ 8.0 10-¥ s
{ B4 6.6
a
B121l__  acc 16% 6.3 2,16 M
1 ; (0.32% B-) (?)
1
71207 | AcCn B" 1.5 4,76 M
. i
P°211 <  AcCt ) a 7.4 0.52 8
PbH207 AcD Stable ——— —

 
chemletry is similar to that of its homologues in graup'Va
of the Periodic Table. Like niobium and tantalum, prot-
actinium is almost completely insoluble in all the common
aqueous media except sulfuric and hydrofluoric acids. I*
18 readily precipitated by hydroxides and phosphates; and,
1n.trace quantities,.ia carried more or less quantitatively

by precipitates of a wide varlety of elements.

The well-publiclzed tendency of protactinium to deposlt on
the walls of glass vessels 1s primarily due to its insolubility;
this tendency 1s not apparent in appropriate concentrations of

gulfuric and hydrofluoric acids.

2, Metallic Protactinium

Metallic protactinium has been prepared by thermal decom-
position of 1ts halldes on a tungsten fllament and by
electron bombardment of the oxidell, by reduction of.the
tetrafluoride with barlum at 1400012. and by electrodeposi-.
tion on various metal cathodes from very dilute, slighflv
acid, fluoride solutionulB.

The metal is grey in color, malleable, and approximately as
hard as uranium. On exposure to air, it acquires a thin
pkin of PaO,

3. 8Soluble Salts of Protactinium

Nowhere 1s the literature of protactinium more confused or
ambiguous than in the refsrences to its solubility in common

mineral acids. The only asystematic study in this area is the
L4 with solution volumes of the

preliminary work of Thompson
order of 0,05 ml;_(Table I11). The avallablility of gram
quantities of protactinium makes this a potentially fertile

field for investlgation.

'The bast solvent for protactinium is hydrofluoric acid, which
rdndtly digsolves the ignited pentoxide and nearly all pre-
cipitates, forming the stable complex ion PaF7’.
Protactinium pentoxide disaolves slowly in hot concentrated
sulfuric acld, but the solubllity 1s low. Prolonged diges-
tion convarts the oxide to a sulfate, which dissolves on
dilution of the acid. |

A solution containing 17 mg./ml. of protactinium-231 in
approximately 7.7 N stou has been stable for over a year15,
and one contalning 36 mg./ml. in approximately 3 N D,S0, has
been stable for six monthslfi.

In the author's experience, only hydrofluoric and sulfurilec
acids permanently dissolve ‘appreciable quantities of prot-
actintuq. With all other mineral acilds, solutions are
unstable, resulting in precipitates or colloidal suspensions’

after periods ranging from a few hours to several weeks.

Solutiena containing 10~3 to 10'4 M protactinium in 6 M I-NOJ
(0.2 - 0.02 mg./ml.) hydrolyzed slowly, but, at concentra-
tions between 10-% and 10-7 M protactinium, the solutions
were sufficiently stable for 24 hours to ylald reproducible

extraction and ion exchange datal?.

Although the literature is prolific of. references to the

solvent extraction and ion exchange of protactinium from
 

 

TABLE III. SOLUBILITY OF Pa IN COMMON ACIDSlu*
Solubility Starting
Acids Normallity {g./1liter) Material
HC10,, 11.1 0.030 Hydroxide
7.1 0.0027 Dilution
HC1l 9.61 0.30 Hydroxide
4,90 0.01 Dilution
3.33 0.0085 Dilution
0.99 0.0015 Dilution
HNO, 15.3 4.2 Hydroxide
13.8 6.6 Evaporation
Q HNO3
9.44 5.5 Dilution
5.66 0.043 Dilution
1.88 0.0056 Hydroxide
1.17 0.0037 Hydroxide
H2304 32.5 0.093 Evaporation
21.9 1.8 Dilutlion
9.93 3.3 Dilution
0.92 0.78 Dilutlion
HF 0.05 3.9 1N HN03

¥ Reviewer'a Note:
the only one of its kind avallable.
recent experlence, however, the values, which were
based on volumes of the order of 0.05 ml., are highly
questionable.

This table is included because it is
In the light of

HC1 solutions, no data other than Thompson's are given to

indicate the limits of solubllity.

The insoiubility of prot-

actinium in 6-8fl HCl is useful as a method of separation from

decay products:

When the lodate preclpitate of protactinium

1s digested with concentrated HCl, the protactinium dissolves

temporarily, then reprecipitates quantitatively. After
digestion on a hot water bath and'centrifugation,;the'Sufier-
nate contains no detectable protactinium15°

Where an occaglonal reference occurs to solubllities of the
order of 1 mg./fil. of profactinium in 6 N HCllB, it must be
regarded as questionable. When elaborate precautions are
taken to eliminate organic complexlng agents and fluoride
ion, the subsequefit HC1l solutlons are unstable. A solution
containing 2 mg./ml. of protactinium, in 8 M HCLl was prepared
from a peroxlde precipitate, but, over a period of three

weeks, about 80 per cent of the protactinlum precipitfitedl9.

4. Insoluble Salts Useful in Separation and Analysis

The normal oxldation state of protactinlum is + 5, but prot-
actinium (V) probably does not exist in solution as a simple
cation. On reduction to protactinium (IV) [e.g., with zinc
amalgaml, the fluoride can be preclpitated and is insoluble
1n water and most acids. Tetravalent protactinium is slowly
oxidized 1n air to the pentavalent state, and the fluoride

redissolveszo.

Protactinium (V) can be precipitated from fluoride solution
by the addition of a stoichiometric amount of KF, which forms
an lnscluble double fluoride, KzPaF7. A double fluoride with

barium has also been reportele.

Alkall hydroxides and carbonates precipitate firotactinium in
both oxlidatlon states, and the precipltate 1ls not soluble in
excess 6f the reagent. Depending on the concentrations,
NH,0H mav fall to precipitate protactinium quantitatively

from fluoride solutionzz.

10
The phoéph#fie'and hypophosphate of profiactinium cant be pre-
cipitated from acid solutions. It has been reported that the
preclpitate will not redissclve even in strong mineral aclds.
However,:the author has regularly redissolved protactinium
precipitated by phosphate, whether in trace amounts carried

by titanium or niobium, or in carrier-free milligram quantities,
The precipitate, upon dilgestion with a sufficlent quantity of
warm 18 N H;504 either dissolves or becomes soluble on dilu-

tion with HCl to which some H,0, has been added.

The phosphate can alsc be precipitated from fluoride solution,
depending upon the relative concentratione of the anions. If
excess phosphate is separated by f£iltration or centrifugation,

the protactinium is soluble in dilute HF.

Iodates precipitate both protactinium (IV) and protactinium
(V) from moderately acid solution {e.g., 5 N H,S0y). The precip-

ltate 1is gelatinous and velumlnous, but becomes more dense on

standing, especially if it is warmed on a water bath. Iodate
precipitation from acid solution provides excellent separation

from phosphate.

Sodium phenylarsenate precipitates protectinium in either
oxldation state; the gelatinous precipitate 1s readlly soluble
in dilute HFZO.

Dilution of a sulfuric acid solution of protactinium(V)
produces a preclpitate which redissolves in ammonium sul-
fate. Potasslum sulfate, however, ylelds a crystalline pre-

clpitate, probably a double sulfate.

Tartaric and citric aclds dissolye the hydroxides of both

protactinium(V) and protactinium(IV), and the solution 1s

11
stable on the addition of NHuoH. Sodium hydroxide precipl-
tates protactinium from a citrate, but not from a tartrate

_solution.

With Hy,0, in large excess, protactinium(V) fo:ms a precip}tate

which is insoluble in NH OH and NaOH. The precipitate is

Iy
soluble in dilute stou only after decomposition of the per-

oxide on a water bath.

Freshly precipitated protactinium hydroxide dissolved rdpidly
in warm aqueous oxallc acidza. Addition of NHQOH to 3 M H2C204
containing 0.15 mg./ml. of protactinlum resulted in peféistent
turbldity at pH 5-6, but precipitation fias not éomplete until
pH 8-9 was reached. Addition of an equal volume of 1 N HC1

to the original oxalate solution had no effect in the cold,

but a white crystalline precipitate contalning approximately

90 per cent of the protactinium was formed in warm solutions.

This precipitate dissolved completely in 8 N HCl.

In general, protactinium(V) follows the chemistry of niobium,

while protactinium(IV) follows that of thorium.

5. Coprecipitation and Carrying of Protactinlum

Most methods for the recoverf of either protactinium-231 or
protactinium-233 rely, for an initial concentration step, on
the entrainment of protactinium by an insoluble carrier of
another element. From the foregoing discussion of insoluble
compounds of protacfinium, it follows that hydroxidé, carbon-
ate, or phosphate precipitateé of tantalum, zirconium, niobium,

hafnium, and titanium will carry protactinium quantitatively,
or nearly so. In addition, protactinlium is carried by most
other flocculent hydroxides (e.g., calcium and iron), probably

by adsorption rather than by isomorphous replacement.

Protactinium is carried by.MnOZ produced by the additlon of

KMnO, to a dilute HNO, solution containing Mn(NOB) In

3 2°
solutions of high lonlc strength the entrainment is not com-
pletely quantitative, but it becomes more so as the precipi-

tate 1s repeatedly redissolved and fractlionally precipitated.

Since titanium and zirconium are also carried, these
impurities, 1f present, can serve as carriers for the sep-

aration from manganeaezu.

6. Solvent Extractlon of Protactinium

At tracer levels (10'5 - 10'10

M) protactinium 1s extracted,
to some extent, from hydrochloric, nltric, sulfuric and even
perchloric acid solutions by a wide variety of unrelated
organic solvents. At the macro level (ca. 1 mg./ml.) the

roater of effective extractants is more exclusive.

In gefleral, protactinium is extracted readily by long-chain
alcohols and certain ketones, but poorly by the lower ethers.
Extractions from aqueous chloride media give better ylelds
and more reproducible results than those made from nitrate
solutions?®?. This would follow from the relative rates of

hydrolysis previously noted.

Maddock and his co-workers (18: 19, 21,.26, 27, 28) p,,e

studied the extraction of protactinium from hydrochloric acid
by various organic solvents (Figure 1l). For extraction of

macro amounts, they found diisopropyl ketone most satisfactory,

13
HT

%

Extraction,

100

 

60

1—." ’
@ Tributyl phosphate o :;(),r"‘
O ff-dichlorodiethyl ether : /

® nitrobenzene
@ 5% tributylamine in benzene / /
(not pre-equilibrated)

 

© di-isopropy! carbinol ° /
© acetophenone

® chlorobenzene

 

 

@ benzonitrile
/ ;
s
e .

 

 

 

 

 

 

 

 

 

/
/)
| e ° —‘“//1

 

 

 

 

 

1°0 2:0 30 4:0 50 60 7:0 80 90
Normality HCL

Fige 1o Percentage of protactinium extracted by an equal volume of the sclvent from
an aqueous hydrochloric acid sclution as a function of the acidity of the aqueous
phase. Initial concentration of protactimum~233 in the aqueous phase 4 to 6 x 10~10y,
(Goble, A. and Maddock, Ae Ge, Jo Inorg. Nucl. Chem. 7, 84 (1958).)

100
but  diisobutyl ketone was used because 1t was commercially

available.

Regardless of the organic solvent used, protactinium (V)
exhibits its highest distribution coefficients from strongly
acid aqueous media, consistent with its existence in aqueous
golution as a complex anion. In this respect, it resembles
most of the elements of groups IVa and Va of the Periodic
Table. Thorium and uranium, which can be extracted from
solutions of low acidity with the ald of s&lting agents,

can, therefore, be easily separafed.

In this laboratory, solvent extraction of protactinium has
been largely confined to diisobutyi carbinol diluted with
kerosene or benzene, primarily hecause of the extremely
high distribution coefflcients, capacity, and decontamina-
tion attainable wlth this solvent.

The author does not subscribe to the hypothesls of a soluble
but inéxtractable specles of protactiniung. In numerous
extractions of solutions containing both traces and milli-
grams of protactinium per milliliter, no such phenomenon

has been observed so long as the protacfiniufi was in true
solution and fluoride ion was absent. A transient specles,

preliminary to hydrolysis, remains a possibility.

The following conditions have ylelded apparently inextract-
able protactinium: (a) the "protactinium™ was actually
actinium-227 and its decay products; (b) £luoride ion was
present; (c) the solution was colloidal; (d) interfering
elements (e.g., nlobium and iron),were present in large

amounts and were preferentially extracted; (e) the organic

15
solvent was excesslvely soluble in the aqueous phase (usua-
lly due to insufficient diluent); and (f) there was insuf-
ficlent sulfuric acid or hydrochlorlc acid to ccmplex both

protactinium and the impurities.

It has been suggested that a polymer of protactinium alone

26,30 renders the protactinium

or of protactinium with niobium
inextractable or :educes the distribution coefficlent. Such
a suggestion is unténable unless the polymer is regarded as
the precursor of a colloid or a preclpitate. As has been
stated previously, aqueous solutions (other than sulfate or
fluoride solutions) contalning macro amounts of protactinium

invariably yleld precipltates after a period of time varying

from a few hours to several weeksl7! 19.-

It 1s significant to note that, whenever protactinium 1s
extracted from an acid phase containing only hydrochloric
or nitric acid, investlgators report that the exfiractions
are carried out soon after dissolution of the protactinium
[or after addition of fluoride-complexing catlons such as
aluminum (III) or boron (III)]. Aging of these solutions
increases the percentage of the "inextractable species", a
behavior consistent with slow hydrolysis and formation of

a colloid.

The addition of sulfuric ficid, originally recommeénded by
Moore30 to break a "nonextractable complex of niobium-
protactinium oxalate"”, has been found to be necessary for
the complete extraction of protactinium by dlisobutyl
carbinol even in the abserce of niobium!?, The precise

conditions required have not yet been fully determined,

16
but an aqueous phase containing 9 N #5530, - 6 N HC1 has
been useful for all protactinium concentrations and degrees
of purity. Solutions containing up to 20 mg./ml. of prot-

actinium in dilsobutyl carbifiol have been prepared.

The extraction of protactinium-233 from niltric acid by
dilsobutyl carbinol has been examinéd, and high decon-
tamination factors for uranium, thorium, zirconium, nio-

bium, and rare earths are reported33.

Tributyl phosphate extraction of protactinlum from hydro-
chloric_acid29 and from nitric.acidl7 has been studied.
Dibutyl phosphate dlluted with dibutyl ether extracted
protactinium-233 quantitatively from an équal volume of
1M }INO3 contalning 2% H,C50y 34, Long-chain amines

extract protactinium from strong HCl solutions-l,

The effect of HF on the extraction of protactinium from
HC1 by diisopropyl cafbinol 1s shown in Figure-zzg. Since
the extraction of niobium is not inhibited by HF, asepara-
tion of protactinium is obtaimed if the niobium is extracted.
from 6 M H,50, containing 0.5 M WF. Protactinium remalns in

the aqueous phasejz.

7. Ion Exchange Behavior of Protactinium

A series of anion exchange studies by Kraus and Moore,
using protactinium-233 and Dowex-l1 resin, has provided
distribution coefficientsja, a separation from zirconlum,

nioblum, and tantal , a separation from iron36, and a

separation from thorium and uranium’’ . (Figures 3,4,5.)

17
100

 

 

 

g 6o

2

540

d

N 20
o
-4.4

LOG.HF NORMALITY

Fig. 2. Effect of HF on solvent extraotion of Pa from
" hydrochloric acld (shown for di-isopropylcarbinol and
5.79 ¥ K1), (Nairn, J. S., Collins, D. A., McKay,

He Ae Co; and Maddock, A. G., Second U. N. Intl. Conf,.
on Peaceful Uses of Atcmlc Energy, A/CONF.15/P/1158).

Typlcally, the feed solution is 9 M in HC1l, from which Pa(V)
is strongly adsorbed by Dowex-1, along with Fe, Ta, Nb, Zr,
and U(IV or VI). Thorium 1s only weakly adsorbed and

afipears in the feed effluent,

The elutriant is a mixture of HCl and HF, the concentrations
of each depending upon the separation required. Zirconium
(IV) and Pa(V) are eluted with 9 M HC1-0.004 M HF, with

the Zr preceding the Pa. Niobium(V) is eluted with

9 M HC1-0.18 M HF., Tantalum(V) is eluted with 1 M HF-4 M
NH,Cl. TIron(fII), U(IV), and U(VI) remain adsorbed when
Pa(V) 1s eluted with 9 M HCl-0,LMHF. They are subse-

quently eluted in dilute HC1,

18
 

 

 

 

 

 

 

10 T T rr——mrr-r- T Tr T T
Tr(l) .

5 s -
=
=3 I 4
1
'-:‘J Ta(X)
Lo | i
b
\C_Ef f

4 | 1
B i
=
> | i
_u-l
o 9 il P |
< M KC1- +BIHCI 0.18 N HMF -I"HNF-

0.004 N HF 4 NNHCI
] uum J ]
O .

 

0 40 80 120 160. 200
Volume (ml.) of eluent.

Fig. 3. Separation of zirconium(IV), protactiniwm(V),
nicbium(V) and tantalum(V) by anion exchange: 6-cm.
Dowex-1 columm, 0.32-8q.-cm. crfas-a_gtional area,
average flow rate 0.2 ml. min.™t cnm. (Eraus, K.
'_.A.- and Ioore, G‘. E., J. Am. Chem. Soc. 73 2900-2
(1951).)

Zirconlum and protactinium can be separated by elution with
HC1 alone38'39;-'Both elements are feebly adsorbed.from HC1
solutions below 5M, but strongly adsorbed above 9 M HCL.
From Amberlite IRA-400 resin, at least 95 per cent of the
Zr(IV) 15 eluted with 6-7 M HC1l.in about six-column volumes -
with not.moré than 0.1 per cent of the Pa(V). The latter
is eluted in small volume with HCl below 3 M. (Figure 6)

Relatively few elements are adsorbed on anion exchange
resins from strong HCl solufions, or if adsorbed, eluted

by strong HCl contalning small amounts of fluoride. Prot-

19
COUNTS/ mins/ml.

07— T T Y T — T T T ™7

 

T rrri
i1 1. 141])

Pa

T
A

 

 

   
 

 

 

 

% =
E A
5 -
10°= | =
- . 5
a .
0 — S - =
N -
L : | -
103 — —
F . | .
P _RVERAGE BAGKGROUND ™\

 

 

=9 M HCl—=fe——9 M HCI - 0.1 4 HF ——o}.0.5 # HCle
|02 ' 1 I 1 1 1 1 1 1 ] 1 1 -
O 15 30 45 60 75 90 105 120 135 150 165 180
Volume, ml. | :

 

Fig. 4. Separation of Pa(V) and Fe(IIT) with HC1-EF mixtures.
(5-cm. colum, flow rate 2.5 cm./min.). (Erens, K. A., and

- Moore, G. E., J. Am, Chem. Soc. 77, 1383 (1955).)

20
2

- CONCENTRATION (arbitrary units)

 

1 T T 1 l
IOMHCI  9MHGI, IMHF  O4 MHCI

_

A\ Th(m) (i Po@ | fl_U(m

1!)

 

 

 

 

 

 

 

 

 

LJ_x_S?mLJ—J___&-x-_x__I__J_J__—_ —————
T 2 3 4 5 6 -
NUMBER OF COLUMN VOLUMES.

Fig. 35. Separ&tion of Th(IV) PB(V)’ and U(V-I . (K:I."ID.B, Ko Ao, Hoore, G. '.E., and
Nelson, F., J. Am, Chem. Soc.’@, 2692=5 (1956;.)

 
log Kq —

 

 

 

 

 

 

 

 

 

® Pa2?33powex 1X%10
200-230 3 days (Ref.2).
0 Pa®®Dowex 2 _
| ~ .200-250 2 days (Ref.1)
2 .' . : - 0 Pa?3¥Amberiite 400

- 2days (Ref.1)
o Pa® Dowex 1x10
200-400 2hr

| | ® Zr°SAmberiite 400
J ‘ .. 2 days (Ref.1)

 

 

 

 

 

 

 

 

 

1 — .
i A Zr9pDowex 2
60-100-4 hr (Ref.3)
@ Zr®3Dowex 1X10 200~400
T L
0 2 4 6 8 10 12 ’

Normality of HCl =—
Flg. 6+ Relation between distribution coefficient K. and 'nol'ari{:y of

hydrochloric acid for Pa23l and 2r92 on different on~exchange resins.
(Kahn, S, and Hawkinson, D. E., J. Inorg. Nucl. Chem. 3, 155-6 (1956).)

actinium(V) in trace amounts can, therefore, be purified
to a considerable extent by this method. The protactinium
eluted in solutions containing HF can be resorbed if HBO'3'

of AlCl, 1s added to complex the fluoride'’.

Hardy and c.:a-v»;vt:u'.'l~7.e1.'|3]'7 have studied the ion exchange of
10'5 M protactinium from HNOB solutions in batch experi-
ments with ZeoKarb 225 cation resin and DeAcidi_te'FF anion

resin in the H' and NO,- forms, respectively (Figures 7

22
and 8). Equilibrium between the cation resin and 6 M HNO4
was reachea within 15 minutes, about ?3'per cent of the
protactinium being adsorbed. About 95 per cent of the
protactinifim was adsorbed on the anion resin after one to
two hours. The samé investigators found that, from a 105 M
protactinium solution in 0.0l M HF-6 M HNO3 16 per cent of
the protactinium was adsorbed on ZeoKarb 225 cation resin

and 75 per cent on DeAcidite FF anion resin.

On the other hand, the au_thor15 has found tfiat 10'5 M
protactinium-231 in 0.05 M HF-1 l.\_'ll-HNO3 or in 0.004 M HF-
0.04 M HNO3 passes freely thrfiugh a column of Dowex-50
cation'resin._ Whereas the decay products, actinium-227,
thorium-227, and radium-223; are qgantitativeiy adsorbed,
more than 99.5 per cent of thé protactinium-231 appears

in the effluept.

8. Miscellany

Paper Chromatography: The R value of protactinium(V)

 

inc:eases with increasing HCl concentration when the chro-
matogram 1s developed with a mixture'consisting of 90 parts
acetone and 10 parts HCl + H,0 41. Evidence for a soluble
form of'protactinium-23j in alkaline solution is based on
its movement on filter paper developed with 1 N KOH and

ite slow migration toward the anode in paper electrophoresis
with 1 N ROH as the electrolyteuz. Paper chromatographic
aeparations of protaétinium-from tantalum, niobium, titan-
ium, bismuth, 1ron, and polonium have been mdde by varying

the elution mixture, butanol-HF-HC1-H,0 with respect to

the HCl or HF concentrationuj.

23
10

 

[

o Protactinium 231
o Thorium 230 +232}ref.
A Urqnium 233

| T

 

(29)

 

 

N

 

 

/1

 

\

 

 

 

W

 

 

 

 

 

 

 

 

 

 

 

 

 

 

> 4 6

' Aqueous HNO3 _concentration > M

Fig. 7. Adsorption of thorim(IV), protactinium(V), and uranium
(VI), on DeAcidite FF resin. (Hardy, C. J., Scarglll, D., and
Fletcher, Jo Mo, Jo Inorg. Nucl. Chem, 7, 257-75 (1958).)

ol

8.
 

.103 ' : | I T I
o Thorium 230 + 232

° Protactinium 231(forward)
a Protactinium 231(strip)
* Uranium 233

 

 

 

10

 

T

 

 

 

 

A NL L

 

 

 

 

 

 

 

 

 

 

 

o 2 4 & 8 10 12
Aqueous HNO; concentration, M

 

Fig. 8. Adsorption of thorium(IV), prntactinifin(v) s and uranium
(VI), on ZeoKarb 225 resin. (Hardy, C. J., Scargill, '.D..s and
Fletcher, J. M., J. Inorge Nucl. Chem. 7, 257-75 (1958).

25
Electrochemistry: The spontaneous electrodeposition of

 

protactinium from HF and HZSOh'solutions on various metals

has been studieduu’ 45, 46. The critical potential for

cathodic deposition of protactinium by electrolysis of

neutral fluoride solutions is -1.20 volts with respect to

the hydrogen e1ectrodel3’u7.

Spectrophotometry: The absorption spectra of solutions
'containing 0.006 mg./ml, protactinium in varying concen-

trations of HCl have been examined by Maddock and his

| l. 1.ON HC! IMMEDIATELY AFTER PREPN.
7 2 I.ON HCI I5HRS. AFTER PREPN,
6\ 3 5ONHCI
- 4 7-ON HCI
5 '5.8.5N HClI
6. 9-5N HCI

O.2}

7 10-2N HCIi

o

OPTICAL DENSITY

   

ke

 

200 - 240 280 320 360 400
WAVELENGTH - IN ‘m .

Pig. 9. Absorption Spectra of protactinium in hydro-—
. chlorio aclid solutlons. (Naim, Je S., 'Co].'l.ins, D.

A., “cKay’ He A, G._, and l[addock, A. G., Second U.

N. Intl. Conf. on Peaceful Uses of Atcmic Energy,

A/CONF.15/P/1158).

26
MOLAR EXTINCTION COEFFIGIENT.

 

7000

~—— Pa(IV)-[Pa] = 1.1 X 103 M

6000 }—

5000 —

4000

3000
1600

1400

1200

1000

- 800

600

400

200

~ - Pa(V)~[Pa] = 43 X 10 M
-+« Ce{III)

 

Fig. 10,

Lo’ | 1 |

3400 3200 3000 2800 2600 2400 2200

X\ (4).

 

/

i
I
i
|

l

|

I

|

l
!

 

1

Absorption spectrum of Pa(IV) in 1 M HC1,

(Fried, S. and Hindman, J. Ce, J« Ame Chem. Scoc.

15,

4863~k (195L).)
19’280 They report that the onset of hydrolysis

co-workers
and disappearance of the solvent extractability of prot-
actinium coincide with the appeavance of an absorption band
with its maximum at 260 mu. (Figure 9}. In 2.4 - 11.8 M
HC10y; containing about 10-2 M protactinium(V), a weak peak
at 210 mu. disappeared in eight to ten days“’ga A pesk at
213 mu. was observed for 0.7 - & M H2304 containing

L ox 1075 M protactinium(V); in 6.5 M HéSOQ the peak was
displaced to 217.5 mu, and, in 9 - 18 M acid, it was
resclved into two components at about 212.5 and 220 mu.

The absorption spectrum of protactinium(IV) in 1 M HCL is

shown in Figure 10 %9 (Also see reference 50.)
Dry Chemistry: Protactinium metal, Pa0, Pal,, Pag@59 Pal4,
PaFa, PaCly,, and Pa0S have been prepared on a 50 - 100 micro-

gram scale and the compounds identified by X~Yay analysislze

Most of the compounds were found to be isostructural with

the analogous compounds of uranium.

The extraction of protactinium tracer from solid Th¥, by
fluorine and other gases was investigated under a variety

of conditionssl.

1V. DISSOLUTION OF PROTACTINIUM SAMPLES

Protactinium-233 or protactinium-231 made by neutron irrad-
iation of Th-232 or Th-230 offers no special problem other
than that of dissclving the thorium. Concentrated HNO3
containing 0.01 M HF will dissolve thorium without render-
ing it passive520 The HF also insures the solubllity of

protactinium.

28
In the case of'Pa231 from natural sources; the.variety of

these sburces defies any attempt to offer a general dissolu-
tion procedure. Typically, residues from processes for
recovering uranium and/or radium contain oxldes of Si, Fe,
Pb, Al, Mn, Ca, Mg, TL, and Zr, afid a random selgction of
trace elements which are usually more abundant than prot-
actinium. Virgln pitchblende or other uranium ores will

contain other elements as well, usually in a refractory

condition.

Hahn and Meitner53 mixed a siliceous residue with Ta205
and fused the mixture with NaHS30, . After washing the

cooled melt with water, they dissolved the Pa and Ta in HF.

Pitchblende was digested with a mixture of HNO, and H,S0,,

- and the digestion liquor was treated with Nazcgjsun The
protactinium in the carbonate precipitate'remained lnsoluble
when the carbonates were dissolved in excess HNO3° When

the washed residue from the HNOj ;reatment was digested
with a mixture of HF and H,50, at an elevated temperature,
over 90% of the protactinium went into solution, while tne

bulk of the residue remalned insoluble.

A slliceous material consistingimainly of lead, barium, and
calcium sulfatea was digested wi;h 60% oleum, and the sul-
fates dissolved. The siliceous fesidue, containing the
protactinlum, was separated and dissolved in HF. Alterna-
tively, the original materlial was attacked with 404 HF,
dissolving the Pa and leaving the heavy-metal sulfates as

residueZI.

Depending upon the amount of material to be processed,
alkaline fusions are sometimes useful in separating large

amounts of silica, etc., while leaving the Pa insoluble.

29
In géneral, the followlng treatments can be recommended, in

the order given:
1. Digest the materlal wlth strong HCl to remove
- Fe.
2. Digest the residue from the HCl treatment with

hot, concentrated stou.

3. Digest the residue from the H,S0, treatment
with 25-48% HF or a mixture of HF and H,S50,.

4. Digest the residue with a hot mixture of HNO3
5. Digest the residue with hot 40-50% NaOH.
6. Fuse the residue with NaZCOJ, NaOH, XOH,

or KHSO4.

Throughout the dissolution, the fragtipns should be examined
by gamma-ray spectrometry, as the 27 kev peak of protactinium-
231 is unique and diastinctive (Figure 11).

While Pa233 has sometimes beén used as a tracer to follow
' the course of the Pa23l dissolutionZI, it is doubtful that
isotopic exchange takes place to a great extent between a

reffactory solid and a tracer-contalning dolufiion.

V. COUNTING TECHNIQUES
1. Protactinium-233

Protactinium-233 decays by beta emission, with a half-life
of 27.0 days, to uranium-233, an alpha emitter with a half-

30
 

INTENSITY (ARBITRARY UNITS)

 

 

 

 

1 1| ]
100 130 200 250 300 350 400 450 500

KEV

Fig. 11. Gamma spectrum of protactinium-231.
(Kitby, He W., unpublished.) -

ME] 4 c/mEC

 

Fig., 12. Camma spectrmm of protactinium-233. (Heath, R. L.,
U. Se Atomic Energy Comm. Repgrt IDO-16408, July 1, 1957)

31
life of 1.62 x 107 yearsu7. The princlpal beta energies
in mev are: 0.15 (37 per cent), 0.257 (58 per cent), and
0.568 (five per cent)®5, The gamma spectrum is shown in

Figure_1256.

In most work, only relativé values of protactinium-233 are
needed, hence the counting problems are essentially'the
game as with other relatively weak beta-emitters; sélf-
absorption, geometry, an& backscattering must be carefully
controlled:. For the absolute determination of pure prot-
actinium-233, a method based on 1fis growth from neptunium-
23? has been suggested57. Here, é highly purified sample
of néptunium-ZB? 1s .electrodeposited and allowed to decay.
The beta-counting rate of protactinium-233 is compared with
the alpha-counting rate of neptunium—Zj?Iand the counting
efficiency calculated from the tfieoretical growth curve.
Altefnatively, the neptunium-237 may be permitted to decay
to secular equilibrium, when the disintegration rateé of

the two lsotopes are equal.

A simpler and more versatile method58 consists of standard-

1zing a scintillation gamma counter with-é solution of
purified protactinium-233 whose disintegration rate has been
determined in a 4T beta counter. A scintillation spectro-
meter is useful for identifying and determining protactinium-
233 in the presence of other activities. Correction for
interference by Compton conversion electrons must be made

if more energetic gamma rays are present.-

2. Protactinium-231

Protactinium-231 decays by alpha emlssion to actinium-227,
also an alpha emitter (Table 1I). The half-life has been

32
reported fo be 34,300 years59, and 32,000 yeafszz. A Tecent
calorimetric measurement on approximately 0.5 gram of prot-

actinium pentoxide15 yielded a value of 32,480 * 260 years.

Protactinium-231 has ten groups of élphfi particles, of which
43 per cent have energles below 5.0 mev. Thié_fact, combined
"with its low specific activity (about 56 mc/g);.makes prot-
actinium-231 alpha counting especially susceptlble to self-
absorptidn. Care must-be taken to keep the radioactive
deposit as thin as possible. A higher degree of counting
precision is attainable with a proportional alpha counter
than with either a parallel-plate alr-ionization counter or

a zinc sulfide scintillation counter. .

Ideally, the alpha plateau should be determined 1ndividually
for each samplé, but this is a tedious and time-consuming
procedure. The following mounting and counting techniqge
has been found to give high alpha-counting'precision15:

Trfinsfer one ml. or less of a fluoride solution
of protactinium-231 to a platifium or gold plate,
or to a stainless steel disk coated-with a thin
plastic film (a clear plaétic spray coatlng or a
~ dilute collodion solution is convenient for this
lpurpose). Evaporate the solution to dryness
under an infrared lamp, tilting the plate as
necessary to retain the solution in the center
of the plate. Cool the plfite and cover the resi-
due with one ml. of 0,1 N HNOB. Add one drop of
concentrated NHuoH and again gvaporate;the solu-

tion to dryness. Lower the lamp sufficiently to
drive off all the NHuNOB, and-ignite the plate

33
over a flame until the organic coating has burned
off (or, 1f gold or platinum was used, ignite to

just below red heat). .

Determine the alpha platéau-of the proportional
counter by counting a standard alpha source
(Pu239, radium D-E-F, or Po210) at 50-volt inter-
vals. At the high-voltage end of the alpha
plateau, count a sample of Sr90/Y90 at 25-volt
iniervals,-and find the voltage below which only

0.01 per cent of the betas are counted. The
protactinium-231 sample can be counted at this

point with goo& precision and no significent

interference from beta emitters.

A gamm#-ray scintillation spectrometer is virtually a
necegssity in modern work with protactinium-zjl. The 27,

| 95, and 300 kev photopeaks (Figure 11) are characteristic,
and the 27 kev peak in particular is unique in the gamma
spectra of the naturally occurring radioisotopes. (Compare

Figures 13 through 16.)

The sensitivity of this method of detectlon 1s shown in
Figure 17, where 0.2 ppm of protactinium-231 is positively
.1dentified in a uranium refinery residue by the presence

of the distinctlve pho;opeak at 27-kev.

The high susceptibility of the 27 kev peék to absorption
precludes its use for the quantitative determination of
protactinium-231, but it 1s particularly valuable in the
preliminary dissolutién of a raw materlal. The 300 kev peak
js useful for quantitative work if cofrgétion i1s made for
the contribution of decay products and other gamma-active

materials.
 

-
1

b
|

 

o
1

INTENSITY (ARBITRARY UNITS)

~n
T

 

 

] I | L L1 ] I ]
50 100 150 200 250 300 350 400 450 500

 

KEV

Fig. 13, CGamma spectrum of actinium-—227 in equilibrium,
(Xirby, H. W., unpublished.)

 

;9'

INTENSITY (ARBITRARY UNITS)

 

 

 

 

5C 100 150 200 250 300 350 400 430 300
KEV

Tige. 1. Gazma spectrmm of thoriwm-230 (lonium).
(Xirby, He W., unpublished.)

35
INTENSITY (ARBITRARY UNITS)

 

 

 

 

 

l l l l l ] | |

 

50 100 (50 200 250 300 350 450 450 500
KEV

Fige 15, Camma spectrum of radium=-226 in equilibrium.
(Kirby, H, W., unp'l]blishedo)

INTENSITY (ARBITRARY UNITS)

9

 

 

 

I I I I | | | | I

 

50 100 150 200 250 300 350 400 450 500
KEV

. f‘:l.g. 16. Gamma spectrum of uraniuwm-235/238,
(Kirby, He W., unpublished.)

36
 

INTENSITY (ARBITRARY UNITS)

 

 

 

] | | | l | | L
S0 100 150 200 250 300 350 400 450 500

KEV

¥ig. 17. OCamag spectrum of raw material (ursanium refinery residuoe.)
(Eirby, H. W., unpublished.)

 

VI. DETAILED RADIOCHEMICAL PROCEDURES FOR PRDTACIINIUfi

Procedure 1

"Preparation of carrler-free protactinium-233", according to
J. Golden and A. G. Maddock, J. Inorg. Nucl. Chem. 2, 46 (1956)

Protactinium-233 was prepared by the neutron irradiation of
samples of acid-insoluble thorlum carbonate. Thls material,
of uncertain composltion, ls reasonably thermally stable, but
completely soluble in dilute HND, and HCl. After irradiationm
the material was dissolved in 8 N HCl and extracted wlth diiso-
propyl ketone (DIPK), The extract was treated with 2  HCL

when the Pa passed back to the aquecus layer. The aqueous

37T
Procedure 1 (Continued)
extract was then made 8 N again and the extraction repeatéd.
The protactinium was kept in solution in 8 N HCL in a poly-
ethylene vessel, It was observed that losses by adsorption
quickly took place on the walls of glass containers even with

golutions of the chloride complex in DIPK.

Better recoveries were obtained if the irradiated carbonate

was dissolved in 8 N HF, Thié solution was nearly saturated
with A1C13 before the first solvent extraction. -The subsequent
separation followed the first procedure. Neither product con-

tained detectable amounts of thorium.

(Reviewer's Note: - Irradiation of one gram of Th232 at

1013n/cm2-sec. for one day willl produce 0,25 curie of P3233.)

Procedure 2

Preparation of carrier-free protactinium-233, accerding to
F. Hagemann, M., H. Studier, and A. Ghiorso,
U. S. Atomic Energy Comm. Report CF-3796 (1947),
as quoted by Hyde (&)

The bombarded Th metal was dissolved in concentrated HNOg,
using a small amount of F~ as a catalyst. The aqueous solu-
tion was salted with Ca(NO3), to give solutions of 2.5 M

Ca(NO

1 M HNO The bulk of the U233

3)2’ 3 4°
was removed by ether extraction, following which Pa233 yas iso-

» and 0.42 M Th(NO,)

lated by dilsopropyl ketone extraction. Further purification
wag obtained by three Mn0, cycles in which 1 mg/ml of MnO, was
successively precipitated from HNO3 solution and redissolved

in HNO5 in the presence of NaNOz. After the third cycle the

38
Procedure 2 (Continued})
solution was made 3.5 M Ca(NO3), and 1 M HNO; and was extracted

with ethyl ether again to effect complete removal of uranium.
The protactinium was again coprecipitated on MnO;, dissolved
in concentrated HCl, and diluted to an acidity of 0.05 M; the
protactinium was extracted into é 0.15 M solution of thenoyl-

trifluorocacetone in benzene.

Procedure 3
-Prepafation of carrier-free protactinium-233, according to
W. W. Meinke, U. S. Atomic Energy Comm. Reports
AECD-2738 and AECD-2750 (1949) as quoted by Hyde (6)
The bombarded thorium-metal target is dissolved in concentrated
HNO;, using 0.01 M (NH4)231F6 as a catalyst. The solution is
diluted to approximstely 4 N HNO; and a thorium concentration
less than 0.65 M. Then Mo'' in excess and KMnO; are added to
precipitate 1.5 mg/ml MnO, to carry the protactinium. The Mn02
is dissolved in a small amount of 4 M NH,OH, The MnO, is
reprecipltated and redissolved three times to reduce the final
solution volume to a few milliliters. After it is made 6 M
HC1 or HNO,, it 1s extracted with two to three volumes of diiso-
propyl ketone. The ketone phase is washed three times with 1 M
HNOy and 3 M NH,NO; wash solution. The protaétinium is finally
stripped out into 0.1 M HNOj. The protactinium left behind is
recovered by using fresh ketome to repeat the extraction cycle,
All 0.1 M HNOj re-extraction solutions are combined, made 6 M
HNO3, and contacted with an equal volume of 0.4 M thenoyltri-
fluoroacetone (TTA) in benzene. The benzene solution of prot-

actinium-TTA complex is washed with 6 M HNO4 once.

39
Procedure 4
Preparation of carrier-free protactinium-233, according to
Max. W. Hill (Thesis), U. S. Atomic Energy Comm. Report
UCRL-8423 (August, 1958)
The targets were either Th metal or ThCl, powder, some of

which was converted to ThO, during the bombardments.

Concentrated HC1-0.01 M HF was used for dissolving relatively
small amounts of powder. Thorium metal was dissolved in con-
centrated HC1 - 0.2 M HF., The flfioride was complexed by the
addition of borax (Na,B,07-10 H,0) or AlCly béfore the solu-

tion was passed through the columm.

The colum was 3 mm. in diameter and was filled to a.height
of 55 mm. with Dowex-1 anion-exchange resin. The colum
volume, defined as the number of drops required for a band to

traverse the length of the resin column, was 5-6 drops.

In 10 M HC1l solutions, Pa(V), Zr(IV), and Nb(V) stick to the
resin, while Th(IV) passes through with the other alpha

emltters below Pa in the periodic table.

With 6 M HCl as the eluting agent, Zr(IV) is rapidly stripped
off in a few column volumes, without loss of Pa(V) or Nb(V).
The Pa is then eluted in 9.0 M HC1-0.1 M HF. The Nb(V) is

eluted in 1-4 M HCL.

The 9.0 M HC1 - 0.1 M HF solution containing the Pa was con-
tacted with an equal volume of diisopropyl ketone (DIPK).
Under these conditions, such species as Fe(1II) and Po(1V)

extract quantitatively into the DIPK along with appreclable

40
Procedure 4 (Continued)

amounts of Sn(IV), Nb(V), and others. Protactinium(V) ex-

tracts to the extent of less than 0.4%.

The solvent phase was discarded, and borax or anhydrous AlCl3
was added to the aqueous phase. (Although AlCl, seemed-to have
. 6lightly better complexing characteristics, it appeared to be
appreciably soluble in DIPK, and excess quantities in solution

gave voluminous precipitates.)

The aqueous phase was then contacted with an equal volume of
fresh DIPK, Under these conditions, Pa(V) is extracted
quantitatively by the solvent phase, sefiarating it from Th(IV),
Ti(1v), v(V), Zr(IV), U(VI), and other species. The Pa was
then re-extracted from the solvent phase into an equal volume

of 2.0 M HCI.

To remove all extraneous mass and reduce the volume to a very
few drops, the golution containing the Pa was then made apprpxi-
mately 10 M in HCl1l and p&ssed through a small anion resin
colum (3 mm. in diameter, 12 mm. in height). After being
washed with 10 M HCl, the Pa was eluted with 2.7 M HCl. The
third through sixth drops contained more than 90 fier cent of

the protactinium,

Procedure 5

Determination of protactinium-233, according to
F. L. Moore and S. A. Reynolds
Anal. Chem. 29, 1956 (1957)
1. The sample drawn from the process should immediately

be adjusted to 6 M HCl or greater. Add an aliquot of suitable

L1
Procedure 5 (Continued)

counting rate to a separatory funnel or 50-ml. Lusteroid tube,
Adjust the aqueous phase to 6 M HCl-47% HpCp04. (If the original
sample contains Th, omit the H,C,0, in the original aqueous
phase and perform.tfie extraction from 6 M HCl. Wash the di=-
isobutyl carbinol (DIBC) phase (see next step) for 1 to 2
minutes with an equal volume of 6 I{ HC1l before beginning the
three scrubé of 6 M HCl-4% H,C,0,).

2. Extract for 5 minutes with an equal volume of DIBC
(previously treated for 5 minutes with an equal volume of
6 M HC1). | |

3. After the phases havé disengaged, draw off and
discard the aqueous phase. Scrub the organic phase for
5 minutes with- an equal volume of wash solution (6 Y HCl -
4% HoC90s). Repeat with two additional scrubs of

the organic phase. Draw the organic phase into a

50-ml, Lusteroid tube,-centrifuge for 1 minute, and draw off
an aqueous phase which appéérs in the bottom of the tube,
being careful.not tollose any of the organic phase. (In many
samples, an aiiquot of the organic phase may be taken at this
stage for counting. Only if a substantial activity of Nb, Sb,
or free I, is present is it necessary to-strip the organic
phasé.)

4. Strip the organic éhase by extracting for 3 minutes
with an equal volume of 6 M H,S0, -6 If HF. Allow the phases
to disengage, centrifuge for 1l minute, and draw off most of

the organic phase, being careful not to lose any of the aqueous

ho
Procedure 5 (Continued)
~ phase. Add an equal volume of DIBC and extract for 3 minutes.
Centrifuge ffir 1l minute and draw off most of the organic phase,
being careful not to lose any of the aqueous phase. Centrifuge
for 1 minute.

5. Pipet suitable aliquots of the aqueous phase for
233

Pa counting. The expected yield is 97-98%.

The follofiing table is given by the authors:

Per Cent Extracted by DIBC from 6 M HCl-47% 1-120204

2 g mdie 2 gos gt
99.5 0.4  0.07 0.003 0.01 9.2 <0.04
99.6 0.5 0.13 0.003 0.04 0.3  <0.04

0.3 0.12 0.005 0.0L 0.1 <0.01
0.3 0.1l 0.003 0.02 0.2

0.06
* Oxallc acid omitted from original aqueous phase, values
given for Th represent lower limit of detection of analytical

" method used.

(Reviewer's Comment: - The procedure is basically sound; but

the degree of separation is affected by the concentrations of
Pa and Nb, A DIBC solution containing 2.5 mg. Pa23l and 20
mg. Nb per ml. plus a stoichiometric amount of phosphate was
scrubbed with an equal volume of 6 M HC1-5% H,C,0,. The
aqueous phase contained 1.5% of the Pa and 75.3% of the Nb,

When the organic phaee was scrubbed a second time with fresh

L3
Procedure 5 (Continued)

aqueous, the aqueous phase contained 92.1% of the residual

Nb and 2.3% of the Pa.

The presence of iron also interferes significantly, presumably

because of the formation of a strong oxalate complex. .The

concentration of oxalic acid is probably less important than
the total amount relative to the amount of nicbium and other

oxalate-complexing cations.)

Procedure 6
Determination of protactinium in uranium residues,
according to J. Golden and A. G. Maddock,
J. Inorg. Nucl. Chem. 2, 48-59 (1956)

The siliceous material consisted mainly of the sulfates of

Pb, Ba, and Ca.

Two methods of opening up were used. In the first, one-gram
samples were treated with 10 ml. of 60% 6leum, and a known

233 ,dded to the mixture. The Pb, Ba, and €a

amount of Pa
sulfates were dissolved, and the siliceous residue, containing

the Pa, was separated and dissclved in 5 ml, of 40% HF,

| Alternatively, it was found that the original material could
be attacked with 40% HF containing the pa233 tracer, leaving

the heavy-metal sulfates as residue and dissolving both Pa231

and Pa233. (Reviewer's Comment: - It is highly questionable

that isotopic exchange was complete.)

Either solution was diluted to 100 ml., and an excess of

BaCi2 was added. The Ban precipitéte was washed until no

bl
Procedure 6 (Continued)

obvious decrease in bulk took place. The remaining precipitate

consisted presumably of fluosilicates and was dissolved in
1M AL(NO4)4-6 N Hfl03.. The Pa was then carried down on a
MnO, precipitate, sufficient MnCl, and KMnO, solutions being
added to produce about 10 mg. of precipitate per ml. of
original solution. The mixtfire was digested at 100°C. for

half an hour.

The resulting precipitate was separated, washed, and dissolved
in 20 ml. of 7 N HCl with a trace of NaNOz. Alternatively,

the precipitate was leached with 10 ml. of 1 N HF,

In some analyses, the MnO2 stage was omitted entirely, and
the fluosilicate precipitate was dissolved in 7 N HCl-1 M
AlCly. The HCl solution from either of these three variations

was extracted with an equal volume of diisopropyl ketone.

Separation from most of the polonium can be effected by re-

extraction from the solvent into 7-8 N HCl-0.5 N HF.-

Procedure 7
Determination of protactinium in uranyl solutiom,
according to J. Golden and A. G. Maddock,
J. Inorg. Nucl. Chem. 2, 46-59 {(1956)
The solution was made 8 N in HCl and 0.5 N in HF and the

tracer protactinium=-233 added.

Direct extraction with diisopropyl ketone removed many impuri-
ties, e.g., iron and polonium. Extraction after adding AlCl3

to complex the fluoride present brought the protactinium into

L5
Procedure 7 (Continued)

the solvent layer. Washing this layer with 8 N HCl removed
those elements whose extraction had been enhanced by the

presence of AlCl;, e.g., uranium and zirconium.

Finally, the protactinium was back-extracted with 6-8 N HCI -

0.5 N HF,

The next solvent cycle avoided the use of AlCl,, repeated
NH,OH precipitation and HCl solution, or continued evaporation

with HCl, being used to remove fluoride.

ln either case, the efficiency of a cycle was 98% or more,
without repetition of any step. At the most two of these

cycles were found to be sufficient to achleve radiochemical

purity.

Procedure 8
Determination of protactinium by gamma spectrometry,
according to M. L. Salutsky, M. L. Curtis, K. Shaver,
A, Elmlinger, and R, A. Miller,
Anal, Chem. 29, 373 (1957)
l. Welgh about 5 g. of uranium residue into a vial
suitable for counting in a well-type gamma scintillation

spectrometer. Determine the counting rate at 300 kev.

2. Transfer the sample to # beaker with a small amount
of HZO’ add 100 ml. of 9 N HCl and 1-2 ml. of 48% HF. Heat
until the sample dilissolves. Add more HF, if fiecessary.

3. Cool to room temperature. Add slowly, with stirring,

10 ml. 1 N HCl containing 10 mg. Th. Allow the mixture to

46
Procedure 8 (Continued)

stand 5 minutes. Add a second 10 mg. of Th, stir, and allow
the mixture to stand 5 minutes. Filter the ThF,.

4. Trafisfer the filter baper and precipitate to the vilal
used for the original sample. Determine-the gamma counting
rate at 300 kev. Make a blank determinatlion for the activity-
of the Th carrier used.

5. Obtain the fa gamna counting rate by difference.
Compare this counting rate wlth that of a standard profactinium
gample at 300 kev, and calculate the concentration of Fa in the

uranium residue.

(Reviewer's Comment: - This procedure contains a number of
fundamental errors, but, with appropriate modifications, may
be useful for obtalining a rough estimate of the Pa concentra-

tlon in certain types of samples.

The basic assumption of this procedure is that, under the
conditions given, all of the protactinium will remain in
solution, while all other radioisotopes which contribute to
the 300 kev gamma-counting rate will be carried on ThF,.

The first part of this assumption is subject to considerablé
doubt; Moore and Reynolds (cited in Procedure 1) found wide
variations in the behavior of protéctinium in the presénce of
fluoride precipitates. The second part of the assumption
lgnores the substantial contribution of radium-223 and other
radium isotopes to the 300 kev counting rate. It 1s unlikely
that a large percentage of the radium would be carried on ThF,

under the conditions recommended.

47
Procedure 8 (Continued)

Finally, no correction was made for the contribution to the

300 kev gamma-counting'rate by.decay products in the "“standard

-gample",

The following modifications seem indicated:.

1. Determine the yield by adding protactinium-233 to
a separate sample at Step 2. (The yield determination must
be:fiade separately because of the 310 ke§ gamma ray of
protactinium-233.)

2. Add barium carrier after Step 4,.and precipiltate
barium chloride by the addition of diethyl ether. (Moore
and Reynolds found that both BaCl2 and BaS0, carried sub-
stantial afiounts of protéctinium-233 in the absence of fluoride,
bgt did not test them in fluoride solutions. The yield
correction allows for_thé-possibility that some protactinium
is carried by the barium precipitate.)

3. Correct the gamma-count rate of the standard sample
for growth of decay products, or prepare a fresh standard

sample free of decay products (See Procedure 9). )

Procedure 9

Determination of protactinium by differential gamma spectrometry,
according to H. W, Kirby and P. E. Figgins. Unpublished.

(Reviewer's Note: - This procedure is limited to analysis of
samples containing only protactinium-231 and the actinifim-227

chain.)
1.

Procedure 9 (Continued)

Obtain a sample of actinium-227 in equilibrium with

its decay products (AEM). The actinium should preferably have

been prepared by neutron irradiation of radium-226 to avoid

possible Pa

2.

231 contamination.

231

Prepare a radibchemically pure Pa standard by

one of two methods.

8.

If the protactinium is in a dry state; dissolve
it in hot concentrated H2504 and dilute the
solution to 18 N H,50,. Add an equal volume of
12 N HC1 to the cooled solution and one or two
drops of 30% Hy0,. (If the prétactinium is in
aqueous solution, adjust the concentration to

9 N H,S0, -6 N HCl. Fluoride must be absent or
complexed with Al+3 or HSBO3. It may also be

removed by evaporation to fumes with H SDA.)

2
Extract the aqueous solution with 2 ml. of
diisobutyl carbinol (DIBC) diluted to 50% with
benzene, Amsco'kerosene, or other inert diluent,
Scrub the organic phase with fresh 9 N HZSO4’

6 N HCl. The organic phase will contain radio-

chemically pure Pa231.

" 1f the PaZ3l ig in DIBC or other organic solution,

scrub the organic solution with two or three
volumes of 9 N H2504-6 N HC1 to remove decay
products. The Pa should be repurified at least

once a week.

L9
Procedure 9 (Continued)

3. Determine the location of the peaks of the unknown
sample in the regilons of 90 and 300 kev. Count the AEM, the
unknown, and standard Pa in that order at the peak of the
unknown in the 90 kev ‘region. Repeat the procedure in the
300 kev region. (Background in each region should be de-
termined before and after each series of three counts, and
weighted if there is significafit variation with time.)

4, Calculate the ratio of the 90 kev count to the 300
kev count in each sample.

6. Determine the alpha counting rates of Fhe Pa and
AEM standard samples either by transferring them to alpha-
counting plates or by counting aliquots of the same solutions.

6. Bj simultaneous equations, calculate the contribu-
tion of protactinium and.of AEM to the gamma-counting rate
of the unknown at its peak in the 300 kev region. Correlate

these gamma counts of AEM and protactinium.
Example - Determination of Pa by differential gamma spectrometry.

Gamma Cts/Min at Base

AEM std. 3001.5 343.4  8.741 127,367
Pa + AEM 2395.4 1044.9 2,292 332,112
Pa?3l gea. 1233.5 885.2 1.393 267,124

2395.4 = 8.741 AEM + 1,393 Pa

Simultaneous equations:
1044.9 = AEM + Pa
Procedure 9 (Continued)

127.9 cts/min; Pa = 917.0 cts/min (Gamma
cts/min at B.L. - 295)

Solution: AEM

47,438 cts/min; Pa = 276,723 cts/min

AEM
- (Alpha in unknown)

Total alpha calculated: 324,161 cts/min (97.6%)

(Reviewer's Comment: - The results are invalid 1f the

actinium-227 chain (thorium-227, radium-223) has been
recently broken. Both the AEM standard and the Pa unknown

should be at least six months old.

'The procedure needs to be evaluated with synthetic mixtures
of Pa and AEM;.but it appears to work well with samples of
‘known age, in which the decay product growth can be

calculated.)

Procedure 10

Determination of.protactinium in uranium residues,
according to H. W. Kirby (Unpublished)

1. To five gramé of the residue in' a 50-ml. centrifuge
tube add 25 ml. of 12 N HCl, mix thoroughly, and warm gently
until the initial evolution of_gés subsides. Bring the
temperature of the water bath to 85-90°C. and heat the mi#ture
in the bath for one hdur,'stirring occasionally. | |

2. Without waiting for thé mixture to cool,.centrifuge
and decant to a second centrifuge tube.

3. To the residue in the first tuBe, cautiously add
3 mi. concentratéd H,50,, and digest on the water bath 15

minutes with occasional stirring. Cool to room temperature,

o1
Procedurs 10 (Continued) o
and cautiously add 3 ml. H,0. Mix. Add 6.ml. 12 N HCl and

3-4 drops of 30% H,0,. Extract the slurry. three times with
two ml. diisobutyl carbinol (DIBC) diluted to 50 per cent
with benzene. Separare the.phases (ceqtrifuging if necessary)
and retain the organic. Centrifuge the elurry and discard.
the aqueous supernate. Repeat the'digestion of ‘the residue
“with fresh HZSD , adjust with HC], and H2 2. and extract as °
before. Dlscard the aqueous phase.

4. Transfer the resldue as a slurry in 10 ml. H20 to a
test tube of a size suitable for use in a well-type gamma
acintillation_epectremeter; Centrifuge the residue:tp'fhe
bottoe of rhe test tube and pour or draw off the'sepernate.

To the residue add 10 ml. of a saturated solution of the
tetraaodium salt of ethylene dismine tetraacetic acid (Na4EDTA)
adJueted to pH 12. Digeat the mixture on the hot water bath
 for 15 minutes, stirring occasionally. Cool, centrifuge, and
discard the aqueous supernate.. Examine.the gamma spectrum
of the residue’ in the region of 27 kev If there is no _peak
| at 27 kev diacard the resldue. If there 1s a poasibility that
a 27 kev peak exists, but it :Le obacured by excessive gamma
" radiation in. the 90 kev region, repeat the digestion with the’
NasEDTA., If a definite 27 kev peak is found, proceed to
: Step-S. -

| 5. Digest the residue with 10 ml. 40%Z NaOH for 15 minutes
on the hot water bath, stirring occasionally. Centrifuge and

' discard the supernate. . Wash the_reeidue with 10 ml. H,0,
centrifuge, and transfer the wash to the HCl solution from.

52
. Procedure 10 (Continued)

Step 2. Treat the residue as in Steps 3 and 4. Retain the
organic extracts and discard the aqueous phase and the
residue. (We have not found it necessary to go beyond this

- point to recover protactinium from the HCl insoluble residue.) =

(1f the fesidue-should.prove more refractory than those withl
which we have had experienée, a repetition of Sfiep 5 is- |
. Tecommended. If that should fail, the use of HCl would be
consideredf).-

6.. To the.HCI solution in the second centrifuge tube
(Step 2). add 10 mg. Ti as TiCly and mix thoroughly. Add
0.5 ml, of 85% HaPO, and two or three drops_of concentfgted
HN03. Mix and heaf on the water bath at 85-90°C for one
hour, stirring occasionally. Cool, centrifuge and decant the
sfiperfiate to a third centriffige tube, |

7. Treat the precipitate as in Step 3. (The Ti preci-
pitate usually dissolves completely after the addition of |
HCl and H,0,. However, any insoluble residue may be treated
as in Steps 4 and 5, if necessary.)

8. To the HCl golution in the tfiird cenfirifuge tube
(Step 6), add 10 mg. Ti, two or three drops of concentrated
HN03, and enough 12 N HC1l to restore the volume to 30;35 ml.

Treat this solution as in Steps 6 and 7.

9, Combine all the organic extracts in a single 50-m1. 
centrifuge'tube §r a separatory funnel and add two ml. of
1 H'HNO3-r0.05 N HF. Mix the phaseé thoroughly for ten

minutes, and draw off the aqueous phase with a transfer pipet.

53
_ Procedurs 10 (Cfintiuuadi)
'_ Rape#t the strip of the organic.phaae with fresh HNO,-HF.
Discard the organic phase, combine the aqueous phases and
evaporate the solution to two ml. in a vial or test tube
sultable for ulé_in the well of the gamma scintiila;ifin.
counter, _

10. . Compare the 'gma. count at l300 kev with that of a
standard sample of protacfiinium—231 (See Procedure 9). |

Procedure 1l
Preparatlion of tetravalent protactinium, according to
M. Halssinsky and G. Bouissiéres, Bull. Soc. Chim. France
1931, 146-8, No, 37, (From the translation
_ by Mae Sitney, AEC-tr-1878) '
A 1-3 E,HZSO4_or'Hc1 golution of pure protactinium or prot-
actinium mixed with its carriers 1s placed in contact with

solid zin¢ amalgam in a plexiglas-colufin.

The column is joine& to a Buchner funnel, also of plexiglas.
The latter 1s closed at the bottom by clamped rubber tubing,

80 that the liquid does not run out through the filter paper.

Hydrogen gas ls constantly circulated through.the various
parts of the apparatus, and at the game time, the solution

of protactlnium_is rfip throfigh the Buchner. It slowly forms
a.precipitate congisting either of the fluoride sait of
reduted protactinium or of a mixturé with'LaF3 if a lanfhanum

halt 1a used as a carriér.

When the precipitate is collected, it 1s separated rapidly

by flltration, accomplished by increasing the pressure of

o4
Procedure 1i (Continued}
hydrogen and opening the clamp on the Buchner. Eventually,

it can be washed with Hy0 which is passed over the amalgam,

'The reduction can also be carried out in the Buchner, in
‘which the amalgam is placéd in contact with the HF solution
of protactinium. The precipitation of the fluoride profiuced
in this case is proportional to the reduction. The procedure
may be advantageous for separating protactinium from tantalum,
zirconium and titanium, which can.be dissoived much more

easily in HF than in other mineral acids.

If the precipitate (6f PaFQ) is collected on filter paper, it
can be stored for 12 to 15 hours without being combletely

reoxldized.

Procedure 12
Preparation of solutions of prOtEQtiniufi(V) in alkali
according te Z. Jakovac and M. Lederer,

J. Chromatography 2, 411-17 (1959)
 Protactinium=233 tracer in- 6 N HCl was evaporated in a micro-
beaker, a few pelletg of NaOH or KOH were added and fused
- over a naked flame for a few minutes, cooled and diluted with
HZO to yield a solution SIE with respect to alkali. Such

solutions usually contain a soluble fraction but also an in-

goluble activity.

1f the solution in HCl ig taken to dryness and moistened with
concentrated HCl and again evaporated,-and this process

- repeated three times, the insoluble compound does not form.

55
Procedure 12 (Continued)

It seems that during evaporation with 6 N HCl some radio-
colloid is formed which does not - react readily with NaOH,
When evaporated repeatedly with concentrated HCl this seems
to he inhibited and presumably the protactinium(v) is left
in the beaker as a very thin layer'on ‘the surface, which then

reacts readily.with fused NaOH.

When solutione which have been evaporated three timeg with
concentrated HGCl are created wlth aqueous 6 N KOH, some
transformation into a soluble form was also noted. without

this pretreatment no soluble fraction is obtained.

Procedure 13

Preparation of solutions of protactinium in nitric acid,
according to C. J. Hardy, D. Scargill, and J. M. Fletcher,
J. Imorg. Nucl. Chem. 7, 257 75 (1958)

Milligram amounts of protectinium-23i were available in a

- HC1-HF solution at a concentration of 0.3 mg./ml. Stock solu-
tions (of the order of 10-3 to 10“4_fl_Pa in 6,&-HN03)were‘_
prepared from this by evaporating almost to dryness with HNO, .-
several times, followed by three precipitations by NH40H; to
assist in_the decontamination from fluoride, with re-solution

each time in cold 6 M HN03.

It was found necessary to dissolve the hydroxide precipitate
shortly (<5 minutes) after its formation to prevent aging to.

HNO4=insoluble compounds of protactinium.

%
. Procedure 13 (Continued)
The solutiohs'so-prepared~always contained smali émounts of
alpfia activity (approximatelj five per cent of the total)
inextractable by tributyl phosphate, Prolonged centrifuging
~ and heating to 100°C, in sealed tubes reduced the alpha-active
inextractable materiallto one per cent; this-was-due to

_daughters of pxbtactinium-ZBl;

Aqueous stock solutions Wére obtained free from inextractable
alpha activity by a preliminary solvent extractlon cycle, |
e.g., by extraction from 10 M HNO3 for two minutes with 50
per cent tributyl phosphate, stripping with. 2 M HNO5, and

scrubbing with benzene to remove traces of tributyl phosphéte.

Procedure 14
Separation of protactinium and niobium
according to F. L. Moore,
Anal. Chem. 27, 70-72 (1955)
Polyethylene bottles were used in all the extractions. The
original aqueous phases contained 1 mg./ml. of Nb carrier
(dissolved in 0.18 M H,C,0,) and a total radioactivity of -

95

1,46 x 106 arma counts per minute or Nb 'radioactivity of
& P &

4 gamma counts per minute. Separate extractions were

3.4 x 10
done under the same conditions for Pa233 tracer and Nb%>

tracer.

Three-minute extractions were peffdrmed with equal volumes
(9 ml.) of diiéobutyicarbinbl that had been pretreated for

three minutes with HF of the same concentration as the

57
| Procedfire 14 (Continued)
original aqueous phase. The organic.phasés were separated,
centrifuged, and washed for ome minute with an equa1 volume
of a solufiion of the same HF-HZSO4 concentration as the

original aqueofis phase.

0 -- <0.1
0.5 <0.01 87.8
| 1.0 . <0.02 92.5
2.0 . <0.02 9.8
4.0 <0.02 98. 4

6.0 | <0.01 | 98.2°

a Each aqueous phase was 6 Y in H,50,.

b A second extraction left no detectable Nb in the
aqueous phase.

IProcedure 15
Paper chromatography of protactinium,
- according to Jacques Vernois,
J. Chromatography 1, 52-61 (1958)

quyéthylene was used throughout. Chromatograms were developed
by ascending elution, Separations were carried ofit-in a
cylinder which was 17 cm. in diameter and 27 em. high. Sfieets
of Whatman No. 1 chromatographic paper (22 X 22.cm.) were
rolled into cylinders and placed in the.bofitom of the reservoir.
-Solutions.were depositéd on_the-paper with the aid of a poly-

ethylene microPipét. Chromatograms were developed over a

59
Procedure 15 (Contlnued)

period of about ten hoyrs, after which the paper was removed

and ailr-dried.

l. Separation of Pa-Ta-Nb with a solvent mixture con-
sisting of 25 ml. of 12 N HCl, 50 ml, butanol, one ml. 20 N
HF, 24 ml. H20. The Rf values were: Pa - 0.50; Nb - 0,.82;

Ta - 1. | |

| | 2, Separation of Pa-Ti-Bi with the mixture: 25 ml. of
12 N HC1l, 5 ml, 20 N HF, 50 ml. butan&l, 20 wl. Hy0. The Rg¢
values were: Pa - 0.45; Ti - 0.66; Bi - 0.66.

3, Separation bf Pa-Fe with the mixture: 33 ml., of 12 
N HC1l, omne ml; 20 N HF; 50 ml. butanol made up to 100 ml..with.
HZO' The Rg valfies were: Pa - 0.46; Fe = 1.

4. Separation of Pa-Po: an unspecified mixture of HCl-
.HF-butanol-HZO (similar to those above) easily separated Pa
from Po. The.Pa moves about half-way, while the Po moves with

the solvent front.

Procedure 16
Determination of protactinium in urine,
.according to E, R. Russell,
U, S, Atomic Energy Comm. Report AECD-2516 (1958).
1. Transfer 400 ml. of urine specimen to a beaker,
wash the container with 100 ml. of concentrated HNO, and
add the wash to the beaker; Evéporate to dryness, and ash

to a white solld with alternate treatments of concentrated

HNO3 and superoxol.

59
Procedure 16 (Continued)

| 2, Treat the white residue witfi 50 ml, concentratedfl
HC1 and evaporéte to near dryness on a low-temperature hot
plate. | |
3. Add 60 ml. of 10 M HCL and heat 10-15 minutes with
occasibnal stirring.
4. Let the.salts settle, and decant the hot solution
to a 125-ml. separatory funnel. .
" 5. Wash the remaining salts with 20 ml. hot 10 M HCl
"and decant to the se-paratory-l funnel.
6. Add 25 ml. of diisopropyl ketone to the hot solution
and shake the mixture 8-10 minutes.' |
7. Digcard the aqueous layer and collect the ketome
in a beaker. Wash the ffinnel with 5 ml. of ketone, and add
the wash to the beaker. \
8. .Evaporate the.ketone in a d:ying oven at.100-110°C.
Avoid higher femperatufes. | | | |
9. When the ketone is ccmpletely evaporated, ignite.the'
beaker in a muffle-furnace at 250-30090 for 5-10 minutes.
10, After the beaker is cooled, take up the resldue in
warm concentrated flNOs, and evaporate éliqudts on counting

~dishes. |

Average-fecovery'in-eight samfiles analyzed was 8416 per cent.
(Reviewer's Comfient: - We have not evaluated this procedure
in the laboratory, but it is easy to agree with its author
tha; “Becausé‘of=the tendency'of Pa to undéfgo hfdrolfsls

and to become colloidal, it is difficult to obtain reproducible

60
' Procedure 16 (Continued)
results. The urine procedure used in this laboratory is
basically the same as that used for all pther.aétinides (See

Procedure 17).

In this reviewer's opinion, urinalysis for protactinium is
an exercise in futility. Ingested protactinium is far more

likely to be found in the feces.)

Procedufe'17

Determination of protactinium in urine,
according to H, W. Kirby and W. E. Sheehan
(Unpublished -- based on U. 5. Atomic Energy
Comm. Report MLM-1003, August, 1954)

'Collection_Procedfire: Personnel are.requested to collect
every bladder discharge'in one fu11.244hou: day, beginning
with the first voiding in the morning either on Saturday or
Sunday. As an alternate, they are permitted to collect the
first voiding in the morning and the last voiding before
retifing on both Saturday and Sunday of the weekend.the sample

"1s to be collected. The first procedure is preferred.

1. Transfer the urine specimen to a 2000-ml. graduated
cyllnder.
2, Dilute the urine if necessary in the graduate to a

.volume of 1800 ml. with H,O,

2 _
3. Transfer to a 3000-ml. beaker and add 25 ml. of
concentratedZNH40Ha. Stlr the sample in the beaker for ten

minutes.

61
Procedure 17 (Contimued)

4. Transfer the solution back.to-the_ZOOO-ml. graduate -
éhd allow the precipitate to settle fof two'hoursb.,

5. Siphon off the supernate to within 106 ml, of the
precipitate in the bottom of fhe cylinder and discard the
supernate.-

:6. Transfer.the precipitafe to fwb 250=ml, centrifuge
bottles and centrifuge for 15 minu;es; -

7. Siphon off the pupernate from the bottles and
discard. | |

8. Dissolve the precipitate in one of the bottles with
five ml. of C6ncentratéd HHO3 and combine this solution with
the precipitate in:;he other bottle, -

9. Wash the emptf centrifuge:bottle with Héo afid add
" to the second centrifuge bottle containing the precipitate,
Dllute the precipitate to 100 ml. with'Hzo; |

--10. Stir the solution in the éentrifuée bottle until
all the precipitate_is dissolved and then add ten ml. of
concentrated,NHaofi. Stir ffir five mifiutesf- -

11. Centrifuge for 15 minutes, siphon off,'énd'discard
the supernate, B |

12. Redissolve the precipitate in 25 ml. of concentréted
HN03. When thelprecibitate has dissolved, transfer the solu-
tion to a 100-ml. beaker. Rinse the qentrifuge bott}e with
‘two 3-ml. portions of HNO3 and add to the beaker. "

13, Evaporate fhe HN03 éolfition'on a hot-platg to 3-5

c
ml. . Allow to cool.
Procedure 17 (Continued)

14. Add 2 mg. of Ce as Ce(NO4),; to a 50-ml. centrifuge tube.

15. Dilute the solution in the 100-ml. beaker with
approximately 5 ml. HZO, swiri, and transfer to the 50-ml.
centrifuge-tube containing the Ce.

16. Rinse the 100-ml. beakér with two lO-mi. portions
of Hy0 and add to the centrifuge fube.' The totai volume in
‘the centrifugé tube should not exceed 30 ml. |

17.  Add two drops of methyl orange indicator'and,'while
stirring, adjust to pH 4-5 wilth NH40Hd. Stir for 15-minutes.

18. 'Centrifuge five minutes and discafd the supernate, )

19. Rinse the sides of the centrifuge tube.with 20 ml.
of 2 N HCi. Any precipitate clinging to the stifring rod
used in Step 17 ghould be rinsed off with 2 N HCl into the
centrifuge tube.

20. Rinse the sides of the centrifuge tube with five
ml. of H,0 gfid add one drop of methyl orange_indicétor.

2l. Adjust to pH 4~5 vhile stixting,by the dropwlse addi-
tion of concentrated HN4OH. Stir for 15 mifiutes.

22. Cefltrifuge fotjfive minutes.and_discard the supernate.

23. Cover fihe precipitate with 10 mi; of 1% NH,Ho PO,
and stif until the slurry is homogeneous;__Cent:ifuge the
precipitaté and discard.;he wash. | |

2@. _flbunte the précipitate; as a slurry in approximgtelf
2 ml. of Hy0, on a stainless steel disk. Ignite the samfile
for 15 seconds over a Meker burner and count in_a:low back=-

ground alpha counter.

63
Procedure 17 {Continued) °

Footnotes to Procedure

A - ApH of 9.0 or greater 1is desiréd. 1f necessary,
add 5-ml. portifins of concentrated NH,OH untill u?ine is at
pH 9-10. | | o

b - If at the end of two hours the precipitate 1e gredter
than 350 mlfi; éllow it to settle until'fhat volume is reached.
Occdsionall&,.the preéipitate may setflé to less than 100
fii. When this is the case, retu;fi the suspension to the
53060-m;. beaker, add concentrated HNO, until the solufiipn
1ls clear, then add iOO mg. of da as Ca(N03)27 _Re?eat the
addition of Nfladfl, gtir, and return the suspension to the
2000-ml, graduated cyiinder. Allow the precipitate to.
settle for two hours. -
| c - If the solution boile dry, fediasolve in 5 ml, of
concentrated HN03 . |

d - It has been_ffiund that Pa will co-precipitaté with
Ce Better from solutions of high salt concentrations. :The
high salt:cohéentrétlon at tfiis_point necessltates a certain
amount of caution on the part of the operator. The adjust=-
ment of the pH with NH;OH gust be dohé.slowly,.gr large
amountQ of Ca salts may.be precipitated along with the Ce. -
It 15 recomnended that the following procedure be used in
adjusting the pH: While stirring vigorously, add the con=-
cénfrafed NH,OH one drop .at a time from a dropping bottle,
_allowing the particles forméd to diaaipgté-béfore makiné

the next addition. Add 10-15 drops in this manner. :Continug

6
Procedure 17 (Continued)

adjusting the pH by adding 3 N NH,OH dropwise until the end-
point is near. Complete the pli adjustment with 1 N NH,OH
and, if neceséary, 1 N HNOs3. |

e - Samples are-mounted on. stainless steel digks 1-7/8
inches in diameter and 0.018 inch thick. Beforé use, the
disks are washed in a solutioq'of a déte:gent, rinsed in
wéter, and dried. Since aquéous solutions-ao not spread
fiell on bright staifiless steel, the disk is held in the flame
of a Meker burner until the surface is slightly oxidized
- (color of brass). The disk is allowed to cool énd a ring ofl
-Zapon lacquer or collodion, 2-5 mfi. wide, is applied with a
brush to the outer edge. The lacquer is dried under an infra-
red lamp, and the disk is allowed to cool.- The CePb4 slurry
is then transferred with the aid of a transfer pipefi'to_the
area within fhe'lacquer ring. As many as three'one-ml. water

washes may be added without danger of overflowing.

fl(ngigugzlg_yg;gf - The éerium used in this urinalysis_pro-
:éedure should have a radiochemical purity of one count pér
hour per milligram; At Mound Laboratory the cerlum islpuri-
fied by a solveqt-extraction method. Yields of greater fhan

‘90 per cent are exfiected.)

 VIL. ' APPENDIX

Summary of the Protactinium Project at Mound Laboratory

The production of approximately one gram of protactinium-231

at Mound Laboratory is dwarfed in significance by the fact,

65
noted elsewhere in this report, that approximately 100 grams

~of this nuclide'waé_recently isolated in Great Britain.

Nevertheiess,_it seems worthwhile to describe the operatipn at
this LabdratOry for its historical value as well as for the
.additional chemical insight aff&rded by the various steps in
'tbe process. It 1ls especially nO;eworthy that.at no point in
the.proceas was it found necesséry, or even desirable, tb use

fluoride for the solubilization of the protactinium.

The protactinium project waé initiated in-1954 at the ;eqfiest
of Oak Ridge National Laboratory for tfie.pu:pOSe,of obtaining.
a supply of pfétactinifim which could be used to study its
macrochemical properties. |

The best afgilable source was a resifiué which could be filtered
from the aqueous raffinate resulting from the diethyl ether
extraction of uranium,. The residue, whose_mdjqr constituents
were iron.'aluminum, caicium? magnesium, cbbélt, and copper,
‘contained approximately 0.1 to 0.2 ppm of protactinium. A
process was developed and réported6o which gave good recoveries
“and purity on a microgram scale, The process consisted of
dissolution in 1 N HCl, saturation with NaCl, and boiling to
coagulate a small precipitate, which'consisted principally of
calcium and silica, and which carried protactinium quantita-
tively. The precipitate was digested with NaOH, to remové
silica, and the hydroxide reaidfie was dissolved in 9 [ HCIL,

The solution was passed through an anion exchanger, leaving
protactinium on the resin. The protaétinium was eluted with

a mixture of HC1 and HF.
On the basis of the apparently simple and inexpensive process,
20 tons of raw material was obtalned from the uranium refineryw
Inasmuch as the filtratlon of the precipitate from the aqueous
waste stream was not a normal part of the treatment of the.
aqueous waste, no stockpile was available, and the raw material

was obtained from the then-current refinery operations.

When received at this Laboratory (in 80 steel drums) it was
obvliously inhomogeneous; the color of the material ranged

from a light tan to the dark reddish-brown of ferric hydroxide.
Spectrographic analysis confirmed that the material was sig-
nificantly different in composifiion from the samples previously
received for analysis and process development. Typically,

iron was a major constituent, and the aluminum and calcium
previously found were either absefit or reduced to minor con-
stituents. No two drums contained raw material of identical

composition.

A single drum was selected which appeared to be intermediate
in composition, and optimum conditions for the dissolution and
precipitétion of the protactinium were deve10ped61. Plant

operations were then begun.

The material in the selected drum behaved exactly as predicted,
as did one or two other batches. However, the inhomogeneity
of the raw material quickly became evident wfien the protac-
tinium either failed to dissolve quantitatively in the
selected acid (2 N HCl) or, after dissolving, failed to pre-

cipitate when NaCl was added and the solution boiled.

67
Since of the 20 tons of raw matefial available, only about
half would Be needed to fulfill the commitment for one gram

of protactinium, batches were selected on the basié of the
solubility of protactinium in 2 H HCl. Simultaneous analyti-
cal and process development fSée Procedure 10) showed that

the precipiltate which carried protactinium from the NaCl
solutlon was nét, as préviously thought, a calcium silicate,
but T10,+xPO;, and that the coprecipitation could be made
quantitative by addition of TiClj to the HCl solution.

(The raw material contained sufficient nitrate ion to oxidize
the Ti*3, Precipitatiofi of Ti'3 is incomplete and excessively
slow in the presence of ferric ion.) Therefore, failure of
the protactinium to coprecipitate quantitatively was corrected
by the expedient of adding TiCl; to the process solution.

(It was subsequently found that those batches in which prot-
actinium failed to dissolve easily were phosphate-deficient
and probably required much higher acidities to solubilize the
iron. In addition, the protactinium was prbbably held in an

insoluble condition by the more refractory Ti0, and Nb,0s.)

The NaOH metathesis of the protactiniferous precipitate, which
was called for by the original process, produced a residue
which was only partially soluble in 9 § HCl, and the protac-
tinium which did dissolve.hydrolyied-élowly and reprec¢lpitated.
1t was,_therefqre, not suitable for use as an anion exchange
feed solution. Dissolution in HF and separatioh of the iron
on an anion exchangérso was found to be impractical because of

the excessively large quantities of iron which had to be removed.

68
However, a slurry of the hydroxide residue in 6 N BC1l could
be extracted with diisobutyl carbinol (DIBC) diluted to 50
per cent with Amsco kerosené° If the dissolution and extrac-
tlon were carried out shortly after the NaOH treatment, about
80 per cent of the protactinium could be extracted before
hydrolysis produced "inextractable" protactinium. It is note-
worthy that the presence of large amounts of iron decreases

the rate of hydrolysis of protactinium in HCI.

After extraction by DIBC, the aqueous raffinate was allowed
to settle, and the supernate was discarded. The insoluble
residue was retained for further processing with H,80,-HC1

(Procedure 10, Step 3).

The crganlc phase was strippéd with two successlve portions

of one-tenth its volume of water.

Although considerable iron was separated by the coprecipita-
tion.of protactinium on Tl from HCl solution, the iron which
remained was extracted by DIBC and stripped along with the
protactinium. When attempts were made to concentrate the
protactinlum by recycling through DIBC, it was found that, in
high concentration, iron was extracted preferentially and
limited tfie degree of protactinium-concentration which could

be achieved.

The strip solution was made 5-6 N in HCl and extracted with
isopropyl ether, which removed a large, but undetermined,

percentage of the iron with a loss of oniy about one per cent

€9
of the protactinium. Higher HCl concentrations drove more
iron into the organic phase, but the loss of protactinium
became prohibitive. The aqueous phase was adjusted with H,50,

and HCl, and the protactinium was extracted in pIBCO2,

The organic phase was stripped with one-tenth its volume of
30% Hy0,. The strip solution was adjusted to 9 N HZSO4 -

6 N HCl, and the protactinium was re-extracted in DIBC, After
four cycles of extraction and stripping, appro#tmately one
gram of protactinium was concentrated in fractions totaling

less than one gallon of aqueous and organic solutions.

The work was interrupted at this point and was not resumed
for more than a year. By that time, all of the protactinium
had hydrolyzed, affording further separation from iron. Dis-
solution in H580, and extraction in DIBC yielded 660 ml. of
organic solution containing approximately 900 milligrams of
protactinium. A spectrographic analysis of the first concen-

trate is glven in Table IV¥*,

It was apparent that the principal impurities were irom,
phosphate; and niobium. The presence of bismuth was highly
questlonable, and no specific effort was made to separate

it or any of the other reported impurities. Analysis of the

final product confirmed fhe validity of the decision.

* = The spectrographic analyses were made by Mr. D, L, Roesch
of Mound Laboratory.

TO
Pa

S1

Mg
Fe
Al

T1

~ Be

Cu

TABLE 1IV.

Aud

Pt

a4 = :]:50%.

b - By gamma counting.

First Concentrate
Per Cent

1.7
0.02
0.42
4.3
0.02
1.97
0.02
0.06
6.58
0.002
0.06
0.26

4,02

Possible Fe interference.

Probable Nb interference.

SPECTROGRAPHIC ANALYSTS OF PROTACTINIUM2

Final Product
—Fexr Cent

65

0.008

0.15

Development of purification methods proceeded concurrently

wlth the purification itself, so that, by the time a method

had been developed which could be used routinely, nearly the

entire batch had been partially purified.

It was found that
methods which were effectlve at one level of purity were
elther partially or totally ineffective at thé next higher
level, so that no one method was ever developed which could,
with cdmpleté confidence, be used at all stages of the

purification.

For example, in a typical series of experiments, 25 ml, of
DIBC containing Fe, Nb, and about 35 mg. of Pa was stripped
with an aqueous 501ution which was 4 N HCl and 10% Hy05.

The volume ratio of the phases was varied, and the course of
the iron separation was followed with the aid of Fe’? added
as a tracer. The percentage of Pa stripped rose sharply with
increasing aqueous volume and reached a plateau of 98% at an
aqueoug-organic ratio of about 0.15, The amount of Fe stripped
also increased, but less sharply, so that, at a volume ratio
of 0.14, 95.67% of the Pa was stripped with 2.0% of the Fe.
However, when the organic phase was stripped a second time
with fresh strip solution in the same ratio, the aqueous

phase contained almost 20% of the Fe.

1t is evident that, in solutions of relatively high concentra-
tions, the degree of separation by solvent extraction is
governed by the relative solubilities of the metal ilon com-
plexes in the aqueous and organic solutionms, ratfier than by
their individual distribution coefficients. This contrasts

with their behavior in more nearly ideal solutions.

The HCl-H,0, strip gave no separation from Nb. However, after

addition of HZSO4 to the strip solution, the Pa was re-

T2
extracted with DIBC, and apprqximatelj half of the Nb remained

in the aqueous phase with less than 6% of the Pa.

Spectrographi¢ analysis showed that 75% of the phosphorus had |
been removed. The intermediate product consisted of 835 mg.
of protactinium in 150 ml. of DIBC. The iron-laden organic
and the niobium-laden aquebus raffinates were retained, and

the protactinium was subsequently recovered and purified.

A second, ‘and far more effective, Fe separation was discovered
at a later stage in the purification: When 10 ml. of DIBC
containing Pa, Fe, and Nb was stripped with a mixture of 4.5
ml. of 18 N H,80, and 0.5 ml. of 30% H,0,, the organic phase
retailned 627% of the Nb and 1007 of the Fe, whlle 1007 of the
Pa passed into the aqueous phase., The Pa was reextracted

into DIBC after addition of HCl to the aqueous phase.

Attempts to separate Nb from Pa by stripping the organic solu-

tion with an HCl/H20204 mix'ture63

were unsuccessful so long

as both elements were present in relatively high concentrations.
With the bulk of the Nb removed, the efficlency of separation
was lmproved, but was still not completely satisfactory: 1In

a single pass with equal volumes of organic and aqueous phases,
a mixture of 5% H,C,0, and 6 N HCl stripped out about 70% of
the Nb and about 57 of the Pa. When Fe(III) was present to a
eignificant degree, the H2C204/HCI strip was almost totally

ineffective.

A batch countercurrent method was developed which stripped

the Nb from each batch of organic solution with three separate

T3
portions of the HyCp04/HCl mixture. About 957% of the Nb was

removed in the aqueous phases, with about 5% of the Pa.

The aqueous phase was diluted with an equal volume of H,0

and heated for 10 minutes at 90°C. Protactinium precipitated,
leéving Nb and Fe quantitatively in the supernmatant solution
(ef. Ref, 23). The preclpitate was centrifuged, and the Pa

was redissolved in H,S0, and extracted with DIBC.

Another separation of Nb from Pa, based on the hydrolytic
precipitation of Pa from acid oxalate solution, comnsisted of
stripping the DIBC solution with 10% H20204 which was 1L N in
HCl, A precipifiate irmediately appeared in the aqueous phase,
and was found to contain all of the Pa and 207% of the Nb.

The precipitate, on being heated with fresh strip solutionm,
dissolved almost completely, but a new precipitate formed in
a few minutes. The new precipitate contained all of the Pa,
while Nb remained quantitatively in solution. The precipita-
tion of Pa was somewhat erfatic under these conditions, but
could be made quantitative by the addition of one drop of 857%

H3P04 to 10 ml, of the oxalate solution.

Eliminatioq of phosphate ion was achleved by precipitation

of protactinium iodate from dilute H2504. The Pa was stripped
from DIBC with 5 N H2504, and the solution was diluted with
H,0. The solution, containing 120 mg. of Pa in 35 ml. of 1.4
N HyS0,, was warmed on a water bath. Protactinium lodate was
precipifated by the dropwise addition of 15% HIO3. The pre-

cipitate was cooled, centrifuged, and separated from the

Th
supernatant solution. The solution was gamma-counted and
contained the equivalent éf between 0.1 and Q.S mg. of Pa,
However, gamma-ray spectrometry showed that the gamma activity
came almost entirely from actinium~227 and its decay products;
the 27-kev gamma peak of protactinium-231 was completely

absent.

The protactinium iodate was digested with concéntrated HC1
and wafmed gently until gas evolution ceased. The Pa dis-
solved initially, but reappeared shortly as a flocculent
precipitate, which was centrifuged, washed, and dried at-110°C.
A spectrographic analysis of the hydrous oxide is given in

Table 1V.

Approximately 0.5 g. of Pa was ignited to the pentoxide at
675°C. An x-ray powder pattern showed that the crystal struc-

ture corresponded to that of cuble Pa205'12*°

The heat output of the Pa205 was measured 1ln an isothermal
calorimeter**, and the half-lite of Pa23l yas found to be
32,4801260 years, after correction ffir chemical impurities
and decay products. Details of the half-life measurement

will be reported in the near future.

As a result of chemical operations at Mound Laboratory thus

far, approximately 700 milligrams of chemically and radio-

* - The crystallographic patterns were analyzed by C. R.
Hudgens of Mound Laboratory.

*% - The calorimetric measurements were made by K. C. Jordan
of Mound Laboratory.

13
chemically pure Pa?3l has been produced. An additional 250
milligrams has been recovered from residues, raffinates, and

other process wastes and is now being purified.

T6
REFERENCES

Ha1551nsky, M. and Boulsgleres, .y Erotactinium, Nouveau
Traité de Chimie Minfrale, XI1, pp. 617-680, ed. by P.
Pascal; Masson et Cie., Paris (1958) (This should be
required reading for anyone working with protactinium.
The bibliography, containing 165 references, is complete
to November 1, 1957.)

Salutsky, M. L., Protactinium, Comprehensive Analytical
Chemistry. Vol. I, Chapter 1V, Section 44, 11 pp., ed. by
Cecil L, Wilson; Elsevier Publishing Co., Amsterdam (In
Press). (Primarily devoted to analytical aspects. &9
references. )

Katzin, L. I., editor, Production and Separation of U233,
Collected Papers, U. S. Atomic Energy Comm. TID«5223, 728

PP. in 2 vols. (1952). Available at 33.25 from Office of
Technical Services, Dept. of Commerce, Washington 24,

D. C. (79 papers devoted to thorium, protactinium, and
uranium chemistry, radiochemistry, separations, and nuclear
characteristics. Not a true review, but a valuable collec-
tion of research papers and data.)

Gmelins Bandbuch der ancrganischen Chemie, Frotgectinium

und Isotope, System Number 31, 99 pp., Verlag Chemie,
G.m.b.H., Berlin (1942). (Reviews the literature to 1940,)

Elson, R, E., Thg Chemigtrv of Protactiniium, The Actinide
Elements, Chapter 5, pp. 103-129, National Nuclear Energy
Serieg, Divigion IV, Plutonium Progevc Record, vol., l44A,
ed. by G, T. Seaborg and J. J, Katz, McGraw=Hill Bocok Coe,
New York (1954). {69 references, the latest original
reference being dated 1931.)

Hyde, E. K., Radio; dcsi Separationz of the Actinide
Elements. Ibid., Chapter 15, ppa 542=95,

therature Survey on: 1. The Chemistyy of Actinium and
Protactinium -= Especiglly in Aqueous Solutiong.

 

  

7T
10.
11,

12,

13.

14,

15.
16.
17.

18,

19.

20.

21.

22.
23.

24,

2. DflL::mina;iQn_9i_Asnininm_flndhzznnassinium 3. Iech-
. Anon.,
Atomic Energy Commigsion, Tel-Aviv, Israel, LS-6, 34 pp.

Strominger, D., Hollander, J. M., and Seaborg, G. T.,
Revs, Mod Phys. 30, 585=904, April, 1958.

Zijp, W. L., Tom. Sj., and Sizoo, G. J., Physica 20,
727-35 (1954),

Maddock, A. G., Private Communication, September, 1959.

Grosse, A. V. and Agruss, M., J. Am. Chem. Soc. 3§,
2200 (1934). _

Sellers, P. A.,, Fried, S., Elson, R. E;, and Zachariasen,
W. H,, J. Am. Chem. Soc. 78, 5935 (1954).

Ferradini, C., J. Chim. Phys. 53, 714 (1956).

Thompson, Roy, U. S. Atomic Energy Commission Report
AECD-1897 (1948). (Also reported in TID-5223, p. 294 -
See Section I., General Reviews.)

Kirby, H. W., Unpublished Work.

Burgus, W. H., Private Commmication, October, 1959,

Hardy, C. J., Scargill, D., and Fletcher, J. M., J. Inorg.
Nucl, Chem. 7, 257-275 (1958).

Goble, A., Golden, J., and Maddock, A. G., Can. J. Chem.
34, 284-292 (1956),.

Casey, A, T. and Maddock, A. G., J. Inorg. Nucl. Chem.
10, 58-68 (1959).

Haissinsky, M. and Bouissidres, G., Bull. Soc. Chim.
France, 1951, 146-148, No. 37 (AEC-tr-1878).

Golden, J. and Maddock, A. G,, J. Inorg. Nucl. Chem. 2,
46-59 (1956). :

Grosse, A. V., J. Am. Chem. Soc. 52, 1742-47 (1930).

Bouissiéres, G. and Odiot, S., Bull. Soc. Chim. France,
1951, 918-19, (5) 18.

Katzin, L. I. and Stoughton, R. W., J. Inorg. Nucl. Chem.
3, 229-32 (1956).
25.

26.

27.

28.

29.

30.
31.
32.
33.

34.

35.
3é6.
37.

38.

39.

40.
41.
42.
43.

Kraus, K. A. and Van Winkle, Q., U. S. Atomic Energy
Commission Report ORNL-239. (See also TID=5223, paper

6.11, p. 296.)

Goble, A. and Maddock, A. G., J. Inorg. Nucl. Chem. 7,
84 (1958).

Goble, A., Golden, J., Maddock, A. G., and Toms, D. J.,
Progresg in Nuclear Epergy --ILI, FProcess Chemlstry 2
pp. 86-98, Pergamon Press, London (1958).

Nairn, J. S., Collins, D. A., McKay, H. A. C., and
Maddock, A. G., 2nd U. N, Intl. Conf. on Peaceful Uses
of Atomic Energy, A/CONF.15/P/1458 (May, 1958).

Peppard, D. F., Mason, G, W., and Gergel, M. V., J. Inorg.
Nucl. Chem. 3, 370-78 (1957).

Moore, F. L., Anal. Chem. 27, 70-72 (1955).
Moore, F. L., Ibid. 29, 1660-62 (1957).
Fudge, A. J. and Woodhead, J. L., Chem. and Ind., 1122 (1957).

Oliver, J. R., Meriwether, J. R., and Rainey, R. H.
U. S. Atomic Energy Commission Report ORNL-2668 (Aprll 1959).

Kraus, K. A. and Moore, G. E., J. Am. Chem. Soc. 72,
4293-94 (1950).

Kraus, K. A. and Moore, G. E., Ibid. 73, 2900-2 (1951).
Kraus, K. A. and Moore, G. E., Ibid. 27, 1383 (1955).

Kraus, K. A.,, Moore, G. E., and Nelson, F., 1lbid. 78,
2692-95 (1956).

Maddock, A. G. and Pugh, W., J. Inorg. Nucl. Chem. 2,
114=17 (1956)

Kahn, S, and Hawkinson, D. E., J. Inorg. Nucl. Chem. 3,
155-56 (1956).

Barnett, M. K., J. Inorg. Nuecl. Chem. 4, 358-66 (1957).
Lederer, M., J. Chromatography 1, 172-78 (1958).
Jakovac, Z. and Lederer, M., Ibid. 2, 411-17 (1959).
Vernois, J., J. Chromatography 1, 52-61 (1958).

Camarcat, M,, Boulssidres, G., and Haissinsky, M., J.
Chim. Phys. 46, 153-57 (1949).

79
45.
46.
47.
48.

49.
50.
51.
52.

53.

54.

55,

56.

57.

58.

59.

60.

6l.
62.
63.

Ferradini, G. and Haissinsky, M 1., Ibid. 533, 722- 25 (1956).
Ferradini, C., Ibid. 50, 589-90 (1953).
Danon, J. and Ferradini, C., Compt., rend. 234, 1361-63 (1952).

Guillaumont, R., Muxart, R., Bouissidres, G., and -
Haissinsky, M., Compt. rend. 248, 3298-3400 (1959).

Fried, S. and Hindman, J. C., J. Am. Chem, Soc. 76,
4863-64 (1954).

Brown, D., Smith, A. J., and Wilkins, R. G., J. Chem. Soc.
(1959), 1463-66.

Miles, F. T. Heus, R. J., and Wiswall, R. H., Jr., U. S.
Atomic Energy Commission Report BNL 482(T-109), (Nov., 1954).

Hyde, E., K., U, S. Atomic Energy Commission Report UCRL-'
8703, p. 6 (1959),.

Hahn, O, and Meitner, L., Ber. 32, 1812 (1919).

Kraus, K. A, and Van Winkle, Q., U. S, Atomic Energy
Commission Report CC-3365X (June 30, 1945). (Also see
TID-5223, paper 6.2, p. 259 (1952).

Ong Ping Hok and Sizoo, G. J., Physica 20, 77 (1954).

Heath, R. L., U. S. Atomic Energy Commission Report
IDO-16408 (July, 1957).

Mc Isaac, L. D., and Freiling, E. C., Nucleonics l4%,
No. 10, 65-7 (1956).

Fudge, A. J. and Woodhead, J. L., Analyst 81, 417-26 (1956),

Van Winkle, Q., Larson, R. G., and Katzin, L. I., J. Am.
Chem. Soc. 71, 2585-86 (1949). -

Salutsky, M, L., Shaver, K., Elmlinger, A., and Curtis,
M. L., J. Inorg. and Nucl. Chem. 3, 289 (1956).

Barnett, M. K. and Elmlinger, A., Unpublished Work.
Jones, L. V., Unpublished work.

Moore, F. L. and Reynolds, S. A., Anal. Chem. 29,
(1957). (Also see Procedure 1l4.)

Bibliography closed December, 1959,

80 USAEC Offica of Technlo! Information Exterslan, Ock Ridge, Tennames