N
TOFI
ORNL P
1577
.
(
+
LO
-
EEEFEEEE
1.25 1.1.4 16
.
.
MICROCOPY RESOLUTION TEST CHART
NATIONAL BUREAU OF STANDARDS - 1963

1
.
.
.
LA
!
,
.
.
. ?
.
LEGAL NOTICE
This report was prepared as an account of
Government sponsored work. Neither the
United States, nor the Commission, nor any
person acting on behalf of the Commission:
A. Makes any warranty or representa-
tion, expressed or implied, with respect to
the accuracy, completeness, or usefulness
of the information contained in this report,
or that the use of any information, appa-
ratus, method, or process disclosed in this
report may not infringe privately owned
rights; or
..
..
e
B. Assumes any liabilities with respect
to the use of, or for damages resulting from
the use of any information, apparatus,
method, or process disclosed in this report.
As used in the above, “person acting on
behalf of the Commission" includes any em-
ployee or contractor of the Commission, or
employee of such contractor, to the extent
that such employee or contractor of the
Commission, or employee of such contractor
prepares, disseminates, or provides access
to, any information pursuant to his employ-
ment or contract with the Commission, or
his employment with such contractor.
:
...
..
ti
.
..
..
OPNr. Pt
1577
CONF 6509183
SEP 21 1965
BEHAVIOR OF THE TRANSPLUTONIUM ELEMENTS IN SOLVENT EXTRACTION SYSTEMS
R. E. Leuze, R. D. Baybarz, F. A. Kappelmann, Boyd Weaver
t
"
-
.
-
-
T
.
.
..
-
-
-
.
-
-
.
August 13, 1965
.
.
.
.
.
.
.
-
.
-
.
RELEASED FOR ANNOINCEMENT
IN NUCLEAR SCIENCE ABSTRACTS
OAK RIDGE NATIONAL LABORATORY
Oak Ridge, Tennessee
Operated by
UNION CARBIDE NUCLEAR CORPORATION
for the
U.S. ATOMIC ENERGY COMMISSION
*Research sponsored by the U. S. Atomic Energy Commission under contract with the Union
Carbide Corporation.
A paper to be presented at the International Conference on Chemistry of Solvent Ex-
traction of Metals.
.
..te
u
n..
...*.*
,....
.
.
...
.
...
.......
...
.
..
.
n
r
.
..
-..-
,*-*.....
..
.
.
.....
2
RHL AEC - OFFICIAL
Behavior of the Transplutonium Elements in Solvent Extraction Systems
..ORNI - AEC - OFFICIAL
R. E. Leuze, R. D. Baybarz, F. A. Kappelmann, Boyd Weaver
Very high activity levels were encountered during the separation of multi-
gram amounts of 24cm-243 Am and 242cm-241Am from Fission products by the
Tramex process, in which actinides are extracted into a tertiary amine from concena
trated lithium chloride solution. Rapid loss of acid and formation of a strong oxidant
.. in the lithium chloride feed solution were the principal effects of the high activity
level. Methanol was added to control acid loss and thus prevent precipitation of
americium and curium hydroxides. The presence of a strong oxidant makes it impos
sible to obtain good decontamination from cerium, since tetrevalent cerium that is
formed extracts with the actinides. Addition of stannous chloride to feed containing
170 curies of Cm per liter 15 w/liter) destroyed the oxidant, kept cerium in the
!
unextractable trivalent state, and resulted in good cerium decontamination. In solu-
tions at twice this activity level, stannous chloride was rapidly consumed, and poor
.
-
cerium decontamination was obtained.
Formation of complexes to preferentially suppress actinide extraction is the
basis of a proposed method for actinide-lanthanide group separation, TALSPEAK (an
acronym for Trivalent Actinide-Lanthanide Separation by Phosphorus Reagent Ex
traction from Aqueous Complexes). Group separation was not obtained by extraction
into monoacidic organo-phosphates or -phosphonates from mineral acid solutions.
ORNI - AEC - OFFICIAL
ORNL - AEC - OFFICIAL
ni
"
" x +
+ :
....…
.

.
" "
.
.
.
.
.
- ...
.1"
-:- '- --:1
--
*.' **?" "V'-'-
.iwwi bow in
, in
N
SE
.*...
*
mate ****
to
u
w
*
REA
***
"
r
..ORNL AEC - OFFICIAL
X1 AIC - OFFICIAL
" : 9
"! 17:55
RT
Even though conditions are shifted in favor of lanthanide extraction by the addi
.
..
tion of water soluble carboxylic acids and adjustment of pH, it is necessary to
.
.
.
decrease actinide extractability by the formation of strong complexes with poly-
aminocarboxylic acids in order to obtain conditions favorable for complete group
.
AY
separatior. Optimum results were obtained by extraction with di(2-ethylhexyl)
phosphoric acid from-lactic acid solutions containing diethylenetriamine pentaacetic
".
acid. Separations obtained for americium-curium from the lanthanides approach those
given by the Tramex process.
eed
Es-..
ORNI - AEC - OFFICIAL
LT
ORNL - AEC - OFFICIAL
BEHAVIOR OF THE TRANSPLUTONIUM ELEMENTS IN SOLVENT
EXTRACTION SYSTEMS
R. E. Leuze, R. D. Baybarz, F. A. Kappelmann, Boyd Weaver
A number of solvent extraction processes for isolation of transplutonium elements
· have been developed at Oak Ridge National Laboratory. This work is continuing and
culminate .
will Gulimate in the application of these processes in the Transuranium Processing
Facility for recovery of the transplutonium elements from targets containing 242pu,
243 Am, and 244cm irradiated in the High Flux Isotope Reactor? This paper is a progress
report covering two parts of the development program. The first part discusses irradiation
effects on the Clanex® and Tramex processes during high activity level processing, and
i
the second part presents an alternative method for group separation of trivalent actinide
and lanthanides which does not use the corrosive chloride solutions required in the
Tramex Process.
IRRADIATION EFFECTS ON CLANEX AND TRAMEX PROC
HIGH ACTIVITY LEVELS
TION EFFECTS ON CLANEX AND LISAMEX PRO
Both the Clanex and Tramex processes have been operated at high activity levels
during the isolation of gram amounts of 243 Am-244cm and 24: Am-242cm in the Curium
Recovery Facility at Oak Ridge National Laboratory. The Clanex process was used to
convert from a nitrate solution to a chioride solution by extracting trivalent actinides
and lanthanides into Alamine 336•HNO2 from a concentrated solution of Al(NO3)3–
LiNOZ, backwashing into hydrochloric acid, and scrubbing the last traces of nitrate
from the product solution with Alamine 336.HCI. The Tramex process was used to separate
americium-curium from lanthanides by selectively extracting the omericium-curium into
Alamine 336.HCI from concentrated lithium chloride solution. Discussion in this presenta-.
tion is limited to effects attributed to the high radiation levels encountered during this
processing, since a report of Curium Recovery Facility operations has been prepared for
publication elsewhere. 3
The Clanex process has been successfully operated with feed containing 440 curies
of
Cm/liter (16 w/liter). There were no detectable irradiation effects on either the
chemical or extraction behavior, except for a very slow loss of acid in the feed at the
rate of less than 0.01 inole per liter per day. In this processing, the solvent was loaded
to an activity level of ~5 watts or 150 curies per liter and received a total exposure of
less than 5 whr/liter. Since there were no adverse effects at this radiation level and;
since it has been previously shown that Alamine 336 exposures of 100 whr/liter can be
tolerated, it seems probable that the Clanex process can be operated at considerably
higher activity levels, perhaps as high as 1000 curies/liter.
Unlike the relative stability of the nitrate feeds used in Clanex processing, the
chloride feeds used in Tramex processing are adversely affected by high radiation levels.
Two important chemical changes in Tramex feed caused by radiolysis have been observed.
These are formation of strong oxidants, which convert cerium to the extractable tetra-
valent state and results in poor cerium decontamination, and a loss of acid, which increases
distribution coefficients but ultimately results in precipitation of metal hydroxides. By
---
-
-
-
--
-
-
-
-
-
modifying process conditions to minimize these difficulties, successful processing
runs have been made at feed activity levels up to 6 w/liter. This does not limit
throughput in the TRU facility since activity levels normally encountered will not
be this great; however, for 4Cm processing, production rates could be increased
if the feed was not limited to 6 w/liter or 50 mg 242cm/liter.
24.
Cerium Behavior in Tramex Processing. – The effectiveness of the Tramex
Process for separating actinides from lanthanides results from the fact that trivalent
actinides are much more extractable than the trivalent lanthanides into long chained
tertiary amines from concentrated lithium chloride solutions. Good rare earth decon-
tamination was demonstrated in laboratory scale tests made at low activity levels.
Distribution coefficients reported for trivalent actinides show that europium is the
most extractable and that if it is satisfactorily removed from the actinides, good de-
contamination from all other rare earths could be expected. However, during high .
activity level Tramex runs made in the Curium Recovery Facility considerable amounts
of cerium were found in the product. Unexpected cerium behavior was first noted
during a process run with feed containing only 20 curies of <** cm per liter (<1 watt/
liter) in which the cerium decontamination factor was only 8 compared to an europium
decontamination factor of >100. It appeared that this abnormal cerium behavior was a result of
attributed to partial oxidation to the extractable tetravalent state, Since the addi-
tion of 0.1 M SnCl, to the feed resulted in improved cerium decontamination, making
it greater than europium decontamination.
het
Since stannous chloride was effective during ce* cm processing, the teetmique
was the first runs made with irradiated americium-241 to isolate 24'Am-242 Cm were
made with feed containing 390 curies of
Cm per liter (14 w/liter) with 0.1 M
SnCl, added to keep cerium in the trivalent state. However, the stannous ion was
oxidized so rapidly at this activity level, that it was impossible to complete the
processing run before it was consumed. As a result, the overall cerium decontamination
factor was <2, compared to a decontamination factor of >500 for trivalent lanthanum-140.
It was possible to improve cerium decontamination by modifying the process só as
to limit irradiation exposure to 600 whr/equivalent of reductant. This was accomplished
in the following manner:
The material to be processed was stored as an acidic solution of 11 M LICI con-
taining 500 curies of 242cm per liter (18 w/liter). Immediately prior to processing, a
small batch of the stock solution was heated to distill out excess acid and oxidants which
had built up during storage de to this concentrate was added a solution of concentrated
lithium chloride solution containing sufficient Snci, to produce a Tramex feed solution
which contained 11 M LICI-0.3 M HC-0.1 M SnCly. This adjusted feed which had a
maximum activity level of 170 curies of Cm per liter (6 w/liter) was completely .
processed in less than 20 hr, thus limiting the exposure to <600 whr/equivalent of re-
ductant. The resulting cerium decontamination factor was 160, which is more than a

factor of 10 less than rare earth decontamination factors demonstrated in very low
activity runs but is about a factor of 100 better than the cerium decontamination
factor for the run made at 14 w/liter. There are insufficient data to determine if
LU
..!:.
-
.
.
.
.
.
t.
-
13
NO
.
cerium decontamination varied throughout this run or if the cerium decontamination
factor may have been limited by oxidation or by aqueous entrainment during ex-
traction.
Radiolytic Acid Loss from Tramex Feed. – The second radiation effect on
Tramex feed solution (loss of acid) was first noted during laboratory testing at high
activity levels, and it was found that the addition of small amounts of methyl alcohol
greatly inhibited acid loss.
of
However, the use of methyl alcohol has not been as effective in Curium Re-
.
covery Facility operations as it was in the laboratory scale studies. The reasons for
this are obscure, but there are indications that the presence of hydrolyzable metal
ions interferes with the protective action of the alcohol. Other preliminary data,
obtained both in small scale experiments and in Curium Recovery Facility operations,
indicates that radiolytic acid loss can be inhibited more effectively by allowing the
radiolytic products to accumulate in Tramex feed for several hours before adding the
methyl alcohol.
Free acid concentration as a function of alpha irrudiation in watt-hours/liter
was determined on samples of Tramex feed prepared in the Curium Recovery Facility,
Addition of 5 vol % methyl alcohol to this feed had only a slight influence on the
radiolytic acid loss (Fig. 1), making it possible to store Tramex feed at 10 w/liter
for approximately one additional day before it became acid deficient. Since it was
thought that hydrolyzable metal ions present in concentrations of 10 to 100 mg/liter
1
.
0.4
ORNL Dwg 65-8556
• WITHOUT METHYL ALCOHOL
CONC. LICI: 10.1M
CONC. 242Cm: 0.096 g/lites
POWER LEVEL: 11.5 watt/liter
A 5 vol % METHYL ALCOHOL
CONC. of LiCl: 12.2 M
CONC. of 242Cm: 0.067 liter
POWER LEVEL: 8.0 watt/liter
UI.co.mod ostale
WC
CONCENTRATION OF FREE ACID, M
n.
be one by
0.16
.
PRECIPITATION OF
METAL HYDROXIDES
900
1000
1 000
1200
HOO1200

TOTAL XPOSURE OF FEED TO ALPHA IRRADIATION (watt-hour/liter)
Fig. 1. Effect of Methyl Alcohol on Radiolytic Acid Loss from Curium Recovery
Facility Tramex Feed.
might be responsible for destroying the effectiveness of methyl alcohol, a test run
methyl alcohol was much more effective in the synthetic feed which contained
essentially no hydrolyzable metal ion impurities except a trace of nickel. Further
evidence of the effect was demonstrated in a test made with synthetic feed con-
taining 0.1 M SnCig. The addition of this reagent also destroyed the effectiveness
of methyl alcohol for inhibiting acid loss.
An unexpected increase in the effectiveness of methyl alcohol was obtained
by allowing samples of Tramex feed to age several hours before methyl alcohol
addition. For example, a feed sample treated in this way with 8 vol % methyl
alcohol was still not acid deficient after an exposure to alpha irradiation totaling
2500 watt hr/liter, even through the acidity was only 0.15 M at the time of the
alcohol addition. This » cedure has been tested during recent *Cm processing
in the Curium Recovery Facility and appears to be effective. The Tramex feed was
adjusted and allowed to stand efore adding 5% methanol and 0.5% formalde- .
.
hyde. The feed acidity was 0.24 M at the time of the methyl alcohoi addition and
was 0.18 M after 5 days or a total alpha irradiation exposure of 950 watt-hr/liter.
Unless this or some other method of stabilizing acid can be perfected, it will
be necessary to frequently analyze high activity level feeds for food acid and to
replace the lost acid.
.
WE
1
***
India
antes tota la hora en el materia
l
en in
de winter is
i
n
the intensii....De arterie
.
A
.
ORNL Dwg 65-8557
Tom
SYNTHETIC FEED
CONC. LICI: 10 M LIC!
CONC. METHANOL: 5 vol %
CONC. FORMALDEHYDE: 0.2 wt %
CONC. 242cm: 7.132/lite:
POWER LEVEL: 15.8 watt/liter
CURIUM RECOVERY FACILITY
TRAMSX FEED
CONC. LiCl: 12.2 M
CONC. METHANOL: 5 voi %
CONC. 242Cm: 0.067 g/liter
POWER LEVEL: 8.0 watt/liter
CONCENTRATION OF FREE ACID, M
TT
o.fi
200
400
600
800
00
2200
1000 1200 1400 1600 1800 2000
TOTAL EXPOSURE OF FEED TO ALPHA IRRADIATION (watt-hour/liter)
2400
2600
2800-3000

Fig. 2. Comparison of Radiolytic Acid Loss from Curium Recovery Facility Tramex
Feed and from Synthetic Tramex Feed both Containing Methyl Alcohol.
.
.
..
..
...
.
-
-
.
-
-
-
-
Summary of Clanex and Tramex High Activity Level Processing Experience. -
The Clanex process has been successfully operated at activity levels up to 440
curies of
Cm/liter and there are no obvious reasons why this cannot be increased
to as much as 1000 curies/liter.
The Tramex process is more susceptible to deleterious irradiation effects, and
difficulties with cerium decontamination arise even at 40 curies/liter unless cor-
rective measures are taken to destroy oxidants formed by radiolysis in the feed.
Stannous chloride can be used for this purpose, but in order to be effective the
feed activity level must be limited to 6 w/liter and the accumulated dosage must
be less than 600 whr/equivalent of reductant. This is adequate for processing in
the TRU facility, since activity levels will normally be less than this. Although
242
*** Cm can be satisfactorily processed at this activity level, it would be desirable
to operate at considerably higher activity levels since 6 w/liter represents only
50 mg *Cm/liter. Acid loss in Tramex feed continues to be a problem and unless
a reliable method for inhibiting acid loss is developed, it will be necessary to in-
termittently replace acid in the feed. Again, this problem is much more serious for
242
***Cm production than for TRU operations because of the higher activity levels
involved.
.
.
.
de
at
m ente en tommin brillant annet
TALSPEAK: AN ALTERNATIVE PROCESS FOR ACTINIDE-LANTHANIDE
GROUP SEPARATION
continua
in same
to the team
s
The Tramex Process appears to be adequate for separating transplutoniurn
elements from trivalent rare earth fission products. It is the method that has been
selected for processing irradiated High Flux Isotope Reactor (HFIR) targets in the
commen
oratione
Awwis
Transuranium Processing Plant (TRU). However, the corrossiveness of the chloride
solutions requires the use of special equipment fabricated from unusual and expensive
materials. For example, equipment and piping in the TRU facility are fabricated from
tantalum and Zircaloy-2. Instrument lines, process water lines, and the cell floor;
pans are made from Hastelloy. Laboratory scale studies at Oak Ridge National
Laboratory have been continued in a search for other separation methods which may
be used in extraction facilities fabricated from stainless steel which already exist
for processing nuclear reactor fuels. It has been found that the extraction of acti-
nides and rare earths into monoacidic organo-phosphates or -phosphonates can be
greatly altered by the addition of carboxylic acids and/or polyaminocarboxylic
acids to the aqueous phase. Proper adjustment of conditions to form complexes
which preferentially suppress actinide extraction is the basis of a proposed new.
.
.
method for actinide-lanthanide group separation, TALSPEAK (an acronym for
Trivalent Actinide-Lanthanide Separation by Phosphorus Reagent Extraction from
Aqueous Complexes).
Flowsheet Description and Results of Laboratory Tests. There are many com-
binations of different carboxylic acids, polyaminocarboxylic acids, extractants, and
.
.
.
RT
L
INES, AVEN.
diluents that can be used to give actinide-lanthanide group separation. However,
the best overall results have been obtained by extraction with di(2-ethylhexyl)
phosphoric acid (HDEHP) from lactic acid solutions containing pentasodium diethylene-
triaminepentaacetate (Na, DTPA). Typical conditions that will give good separation
by extracting the rare earths and leaving the trivalent actinides in the aqueous solution
are given in Fig. 3.
A second extraction cycle is also shown, in which conditions are adjusted so
· the trivalent actinides will extract. The purpose of this cycle is to remove the
actinides from any contaminants which were not extracted in the first cycle. For
this process, feed containing actinides and contaminants are adjusted to 1.0 M lactic
acid at a pH of 1.8. Trivalent actinides are removed in a compound extraction con-
tactor. For I volume of feed, 3 volumes of 0.5 M HDEHP in DIPB diluent and 2
volumes of 1.0 M lactic acid-0.1 M DTPA at pH 3.0 are used. The aqueous phase
is adjusted to pH 1.3 and used as feed for the second cycle in which the actinides
are extracted in a compound contactor. For 3 volumes of feed, 2 volumes of 1.0 M
HDEHP in Amsco 125-82 diluent and I volume of 1.0 M lactic acid-0.05 M DTPA
at pH 2.0 are used. Actinides can be readily backwashed into 6 to 8 M HCl or
HNOZ: Laboratory scale tests of this flowsheet have been made in mixer-settlers
with 8 scrub and 8 extraction stages in each contactor. Runs have been made with
247
tracer amounts of irradiated 'Am spiked into the feed to demonstrate the behavior
of <*%Cm (~10 mg/liter) and fission products. A rare earth decontamination factor
of ~100 was obtained in the first cycle, and <0.1% of the curium and only small
amounts of ruthenium and zirconium were extracted. In the second cycle, 99.96%
www:
LA
Dewan
D
'honi.it
ORNL Dwg 65-8555
1.0 M LACTIC ACID
ÕI M DIPA
pH 3.0
1.0 M LACTIC ACID
0.05 M DTPA
PH 2.0
ym.
:
WASTE SOLVENT
<0.1% Com
<0.5% Ru
~2% Zr
>99.9% RE
PRODUCT
SOLVENT
99.96% Cm
~25% Zr
<0.1% Ru
SECOND EXTRACTION CYCLE
FIRST EXTRACTION CYCLE
FEED
1.0 M LACTIC ACID=0
PH 1.8
Cm - F.P.'s
ADJUST TO PH 1.3
0.5 M HDEHP
in DIPB
WASTE
~75% Zr
~99% Ru

Cm,
1.0 M HDEHP
IN AMSCO 125-82
.
Zr, Ru
it
Fig. 3. Typical Talspeak Flowsheet.
lla,
10
of the curium was extracted along with about 25% of the zirconium. There was no
detectable ruthenium in the solvent, which corresponds to a ruthenium decontami-
nation factor of 210°. Distribution coefficients have been determined, and counter-
current extraction tests have been made to study the behavior of other contaminants.
In general, those contaminants that did not extract in the first cycle did not extract
in the second cycle; thus, most impurities were separated from the americium-curium.
The first cycle will give good decontamination from strontium in addition to good
decontamination from lanthanides. The second cycle can be expected to give good
decontamination from cadmium, cobalt, chromium, copper, nickel, iron, molybdenum,
and lead. Barium and zinc removal was poor in the demonstration tests. If solvent
is recycled, a method of washing aluminum from the solvent will be required since
some aluminium was extracted and tightly held by the solvent. Both zirconium and
silver are insoluble in lactic acid solution containing DTPA and will precipitate in
the extraction contactor unless they are removed from the feed. :
Use of Carboxylic and Polyaminocarboxylic Acids to Enhance Actinide-Lanthanide
Group Separation. - A number of studies have been reported on the extraction of acti-
nides and lanthanides from mineral acid or salt solutions into monoacidic organo-
.
5,112167
phosphates and -phosphonates. <-Extraction from dilute hydrochloric acid into 1.0 M
2-ethylhexyl (phenylphosphonic) acid (2-EH(OP)A) in diethylenebenzene (DEB) diluent
has been selected as the method for use in the TRU facility to separate transcurium
*5,15

elements from americium and curium, but it is obvious from the distribution coefficients
Gr.
(over a range of several decades by using other monoacidic phosphates, diluents,
25 50
Nay.7 iki mbi
commerc al distribution Co-
efficients shown in Fig. 41. that actinidé-Ianthanide group separation cannot be
obtained with this system. It is possible to change these distribution coefficients
.
ORNL-LR-DWG 74486.R
1000
هع: فرز من نمره يعنننننمنننهننس لانضمتت سقارمنستستجد

IM HEHØP)-DEB VS 2 M HCI
100
جميعا لننمننمسیسمس-
LANTHANIDES
IC
DISTRIBUTION COEFFICIENTS.
- ACTINIDES I
-0.0014
Am Cm Bk Cf Es Fm
Fig. 4. Distribution of Actinides and Lanthanides between 1 M HEHØP) and
2 M HCI.
..
.
..
..
...
..
.
.
..
...,
maar
?
.
.
ORNL-DWG 63–7677 R1 ,
SEPARATION FACTORS.

in
om N
APROXIMATE LANTHANIDE - AMERICIUM
,1 HNO3
2 GLYCINE · HNO3
3 TARTARIC ACID
4 GLYCOLIC ACID
5 DIGLYCOLIC ACID
.-
.
-
.-
-
.
La
Ce
Pr
Sm
Eu
,
Nd Pm
LANTHANIDES
SWR
Fig. 5. Lanthanide-Americium Separation Factors Obtained by Extraction into
HDEHP are Shifted by Presence of Carboxylic Acids.
.
.
A
S
.
.
.
W
......
-.
.
Y
.
-
..
.
!r
*OW
acidities, and concentrations; however, none of these variables have an appreciable
effect on the relative extractabilities of the actinides and lanthanides. When sim-.
.
ple carboxylic acids are substituted for mineral acid, relative extractabilities can
be changed considerably. A summary of approximate lanthanide-americium separa-
tion factors for HDEHP extraction from dilute nitric acid and from several different
carboxylic acid solutions is given in Fig. 5. Cerium is less extractable than americium
from dilute nitric acid. Cerium and americium have the same extractability from
glycine.HNOZ. The cerium-americium separation factor becomes progressively
greater for extraction from tartaric, glycolic, and diglycolic acids. It can also be
noted that there is a decrease in the extractability of the heavier lanthanide. relative
to the lighter lanthani des, and in some cases there is a decrease in the extractability
of the heavier lanthanides relative to americium. This can be at least partly explained
by the fact that americium (and other actinides) form stronger complexes than the light
.
.
.
-.
.
lanthanides. The complexes formed are either not extractable or are much less exc
tractable than the uncomplexed metal ion.
Since complex formation is important, the solution pH, which greatly affects the
amount of complex formed, also has a marked effect on the extraction system. As the
pH of glycolic acid and lactic acid solutions is increased, the distribution coefficients
increase only slightly (Fig. 6), and are not inversely proportional to the cube of the
hydrogen ion concentration as is the case for extraction from mineral acids. The in-
creased extraction of the uncomplexed metal ion which results from decreasing the
acidity is apparently almost counterbalanced by the increased formation of unextractable
complex.
-
.
--
ORNL-DWG 63–7678 R2

DISTRIBUTION COEFFICIENTS
Amo
--- LACTIC ACID
GLYCOLIC ACID
1.5
.
3.5
2.0
2.5
3.0
PH OF AQUEOUS PHASE
Fig. 6. Effects of pH Adjustment on Extraction of Americium, Cerium, and
Europium. Extractant: 0.3 M HDEHP in DIPB. Aqueous Phase: ' M Carboxylic Acid.
..
a
t
,
W.
,
.
.-
..
.
. .
.
..
.
YA
RALLY
ه شمية
12
.
ستنقلتستمتهن لعدم ننننمننسا منہ بن نعمان بن محمد
بن نعمانشست
Because the actinides form much stronger complexes with polyamino-polycarboxylic
acids than do the lighter lanthanides," addition of polyamino-polycarboxylic acids to
the aqueous phase greatly enhances the separation of light lanthanides from actinides.
Although many aminocarboxylic acids show this behavior, the best results have been
obtained with diethylenetriaminepentaacetic acid. At optimum pH and DTPA concen-
tration in mineral salt solutions, separation factors between europium, one of the less
extractable lanthanides, and americium are about 100." However, solutions of 0.1 M
DTPA in 0.5 M NaNO3 are at best metastable and crystals of HZDTPA frequently
form on aging. Extraction kinetics for some lanthanides were very slow, e.g., equilibrium
for europium extraction usually required more than 30 min, while less than 5 min was
needed for equilibrium in the extraction of cerium. The presence of carboxylic acids
improved extraction kinetics, provided a highly buffered'aqueous phase, and increased
the solubility of DTPA.
The progressive improvement of lanthanide-americium separation factors as the
aqueous phase is changed from HNO3 to 1 M glycolic acid at pH 1.8, 1 M glycolic
acid at pH 3.5, and then to 1 M glycolic acid plus 0.1 M Na, DTPA is shown in Fig.
Distribution coefficients for trivalent actinides, most of the light lanthanides,
thulium, and ythium have been determined for extraction into 0.3 M HDEHP in di-
isopropylbenzene (DIPB) diluent from 1 M lactic acid containing various amounts of
Na DTPA (Fig. 8). Distribution coefficients decrease as the Na DTPA concentration
fond pH) increases because larger fractions of the metal ions are held in the aqueous
ORNL-OWG 63-4711 R9

*
,..
+
(
x
-
S.."
IM GLYCOLIC ACID,
-0.1 M NASDTPA
*+
-
-
-
++
+'
M
•
•
•
•
•, ..
.
-
:
1M GLYCOLIC ACID, DH 1.8
HNO3
4. •
•
•
•
•
•
•
SEPARATION FACTORS
M GLYCOLIC ACID, PH 3.5
APPROXIMATE LANTHANIDE - AMERICIUM
0.1
La
Ce Pr Nd Pm Sm Eu
Fig. 7. Lanthanide-Americium Separation Factors Obtained by Extraction into
HDEHP in DIPB from Various Aqueous Solutions.
.
161
.. ORNL-DWG 64-4316 RI
PH
1192.3 2.5 2.6
3.0
3.5
Eu a
• Tm
Pm
Ce
Lor
- Amo
Credo
-La
DISTRIBUTION COEFFICIENTS
-
Sm
- Ce
Eu
* Pm
Nd
Am
-
0.02
0.04
0.06
0.08
0.1
CONCENTRATION OF NO OTPA (M)
Fig. 8. Americium and Lanthanide Distribution Coefficients as a function of
DTPA Concentration. Extractant: 0.3 M HDEHP in DIPB. Aqueous Phase: 1 M Lactic
Acid, Variable Na OTPA Concentration.
n
a
-.......
.
.
.
.
-
-

--
-
13
white man die wind their abili
phase as inextractable complexes. The initial increase in lanthanum and cerium
to manager
distribution coefficients is apparently a result of an increase in uncomplexed ion
stor rolle vicinan .
extraction with increasing pH under conditions at which little of the cerium or
lanthanum is held as inextractable complexes.
Actinide and lanthanide distribution coefficients for the system 0.3 M
HDEHP in DIPB versus 1 M lactic acid containing 0.05 M Na DTPA at pH 3.0
are given in Fig. 9. These data show that it should be possible to separate all
the rare earth Fission products from all the trivalent actinides, americium through
fermium. This system is especially good for separating americium-curium from the
rare earth fission products. The separation factor between neodymium, the least
extractable rare earth, and americium-curium is about 50.
Effect of pH, Extractant Concentration and Alpha Irradiation Dose on Ex-
.
.
.
traction. – Distribution coefficients vary markedly with the pH of the aqueous
phase as shown by the data for americium in Fig. 10. Maximum americium ex-
traction into 1 M HDEHP in n-dodecane diluent from either 0.5 M or 1 M lactic
acid containing 0.01 M NA DTPA was obtained at a pH of about 1.5. There is
essentially no difference in extraction from 0.5 M and 1.0 M lactic acid if they
are adjusted to the same pH. It has also been found that the addition of up to -
, at least 2 M NaNO, in the aqueous phase had no appreciable effect on americium
distribution coefficients.
.
.
.
. .
.
.
.
-
- -
.
..
imao ....
.......-;-
..........
...
"
..
pat peny
erapan tot
wartym
ORNL DWG 65-6162 R1

LANTHANIDES
DISTRIBUTION COEFFICIENTS
öö ē
ACTINIDES
La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
Am Cm Bk Cf Es Fm
Fig. 9. Distribution Coefficients of Actinides and Lanthanides Between 0.3 M
HDEHP in DIPB and 1 M Lactic Acid Containing 0.05 M DTPA at pH 3.0.
- SYYS
.
:
+
TIM.
'.
" "
"
Lirir
.
ORNL-DWG 63–768
..
.
..
1.
6
.
DISTRIBUTION COEFFICIENTS
-
• 0.5 M LACTIC ACID,
0.01 M Nag DTPA
01.0M LACTIC ACID,
0.01 M Nag DTPA
w
0.2
2.0
3.0

1.0
PH OF AQUEOUS PHASE
Fig. 10. Americium Distribution Coefficients as a function of pH. Extractant:
1 M HDEHP in n-dodecane. Aqueous Phase: Lactic Acid, 0.01 M Na, DTPA, PH
Adjusted with HNOZ.
_
.
-
.-
.
...
The effect of extractant concentration for the TALSPEAK process has not
been thoroughly studied, but a simple relationship does not exist. For extraction
into HDEHP in DIPB diluent from 1 M lactic acid containing 0.05 M Na DTPA,
the americium, europium, and cerium distribution coefficients are proportional to
:
the 2.65, 2.80, and 2.85 powers of the extractant concentration, respectively
(Fig. 11). This same relationship does not hold for other extractants and diluents
.
and appears to be different with other carboxylic acids.
weinen
The effects of high activity levels on this system were investigated briefly
by determining distribution coefficients for lanthanum and curium between 0.3 M
HDEHP in DIPB and 1 M glycolic acid containing 0.1 M NA DTPA, tracer 140la,
and sufficient 242 Cm' to give a dose rate of 10 watts/liter. Distribution coefficients
were determined as a function of alpha irradiation dose by intermittently stirring,
-
allowing the phases to settle, sampling each phase, and analyzing each phase for
both lanthanium and curium. There were no increased difficulties in phase separa-
tions for irradiation exposures up to 440 whr/liter. The only effect noted was a
gradual increase in distribution coefficients (Fig. 12). Since the curium distribution
coefficient increased more rapidly than that for lanthanum, the lanthanum-curium
separation factor decreased from an initial value of about 50 to only 30 at an ex-
posure of 440 whr/liter. This radiation damage appears to be destruction of the
complexing agent, N DTPA, which results in the increased extraction.
ORNL-DWG 63–7686 R1.

02
.
..
.... B
SLOPE:
42.85
70 2.80
· DISTRIBUTION COEFFICIENTS
0 2.65
!
10-3
L
. Amon
0.05 0.1
0.5 1
2
CONCENTRATION OF HDEHP (M)
Fig. 11. Americium, Çerium, and Europium Distribution Coefficients as a Function
of Extractant Concentration. Extractant: HDEHP in DIPB. Aqueous Phase: 1 M Lactic
Acid Containing 0.05 M DTPA at pH 3.0.
;
..........
PT
time to
morning
and st
r
umente
ORNL-DWG 63-7675 R1

Mil..
.,
.,
I
<
DISTRIBUTION COEFFICIENTS
cm
o
100
200
300
400
:
ALPHA IRRADIATION DOSE ( whr/liter)
Fig. 12. Change of Lanthanum and Curium Distribution Coefficients with Alpha
Irradiation Exposure. Extractant: 0.3 M HDEHP in DIPB. Aqueous Phase: 1 M Glycolic
Acid Containing Various Amounts of NoDTPA.
.
. . .
.
.
.
.
.
.
. .
.
..'
15
:
,
Behavior of Lanthanum at High Concentration. – In order for the TALSPEAK
process to be practical, it must be capable of handling rare earth fission products
''
at concentrations up to 10 or 20 g/liter. Scouting tests with several different
carboxylic acids have shown that lanthanum solubility is greatest in lactic acid
solution. In these tests, lanthanum hydroxide was dissolved in various carboxylic
acid solutions. Lanthanum solubilities in 1 M acids were: less than 2 g/liter in ..
tartaric acid, less than 3 g/liter in citric acid, and less than 6 g/liter in glycolic
acid. Crystals of lanthanum lactate monohydrate have formed from solutions con-
taining 25 to 30 g of lanthanum per liter. However, stable solutions containing
60 g La/liter have been obtained by dissolving lanthanum oxide in 1 M lactic acid.
Since the equivalent weight of lanthanum is only slightly more than 46, it must be
present in these solutions as a basic salt.
...".
.
..
In order to determine lanthanum extraction behavior at higher concentrations,
.
...
a series of experiments were run in which lanthanum was extracted into 0.3 M
.
... ... ..
HDEHP in DIPB from 1 M lactic acid containing various amounts of Na DTPA. In
:
...
........
each test, the organic phase was contacted many times with fresh aqueous phase,
..
...
and lanthanum distribution was determined after each of these contacts. The results
.
...
plotted in Fig. 13 show the peculiar extraction behavior obtained. Not only is there
an abrupt change in extraction behavior at a lanthanum concentration in the solvent
of 5 to 7 g/liter but also chemical equivalents of the lanthanum extracted by one
mole of HDEHP was not 1.0 as expected but was 1.5.
TTTTT
TTTÖ
JORNL DWG AS-PLE
ORNL DWG 65-911R1
.
TTTTT
Bestellen
anddoaolo
goma
AL
7*
amooooooo
0.025 M
NO DTPA
· LANTHANUM CONCENTRATION IN EXTRACTANT. 6/liter)
0
ayo
05
M
0.2 mm
0.05 M
S
....
.....
...
..
.
........
0.01
.
0.1
10
.
100
2
.
.
..............

LANTHANUM CONCENTRATION IN AQUEOUS PHASE (liter)
Fig. 13. Amount of Lanthanum Extracted as a function of Lanthanum Concen-
tration in the Aqueous Phase. Extractant: 0.3 M HDEHP in DIPB. Aqueous Phase:
IM Lactic Acid Containing Various Amounts of NaDIPA.
..
.
...
-
-
-
lyt
tas.
16
.
.
Acidi
Another unusual behavior in this extraction is shown in Fig. 14. No detect-
able amount of lactate was extracted into HDEHP when lanthanum was absent.
..-
.
?
CRY
However, lactate always extracted along with the lanthanum, but the ratio of
lactate to lanthanum extracted varied. When the amount of lanthanum present
was small, such that 0.5 g/liter was in the extractant, 4 moles of lactate was ex-
tracted for each gram atom of lanthanum extracted. At high concentrations of
· lanthanum, this ratio decreased to a minimum of about 0.5 at 10 g La/liter solvent
and then increased to about 1 at saturation. At saturation, each mole of HDEHP
had extracted 0.5 mole of lactate and 0.5 gram atom of lanthanum.
These results indicate the complexity of the system and point out the need
for additional studies to determine the nature of the extraction mechanism.
Status of the TALSPEAK Process. — The experimental results of laboratory
tests show that TALSPEAK is potentially an effective new method for the separation
of americium and curium from the lanthanide elements. The primary advantage that
i
this method may be used in existing solvent extraction plants arises from the fact
that non-corrosive solutions are used which can be satisfactorily contained in con-
ventional stainless steel equipment. The large number of variables involved makes
development of an optimum process tedious but at the same time promises versatility
so that conditions can be tailored to meet a variety of objectives. This process
should be useful for recovering heavy elements from high burnup power reactor fuels.
Additional development is required, primarily to fix the detailed conditions
for a specific recovery program and to test and demonstrate the process in actual
processing runs.
'.
'.'..
.
--
1
u.
1
1
.-
.
' - 24
ORNL DWG 65-6161R 2
.
LACTATE CONCENTRATION IN SOLVENT (moles/I)
• MOLES LACTATE/G ATOM LANTHANUM,
0.5 1 2 . 5 10 .20 50 100
LANTHANUM CONCENTRATION IN SOLVENT (9/1)
Fig. 14. Amount of Lactate Extracted as a function of the Amount of Lanthanum
Extracted. Extractont: 1 M HDEHP in DIPB. Aqueous Phase: ! M Lactate (Acid +
Lanthanum Salt).

REFERENCES
1. Leuze, R. E., Nucl. Sci. Eng., 17, 448 (193).
2. Burch, W. D., Arnold, E. D., and Chetham-Strode, A., Nucl. Sci. Eng., 17,
438 (1963).
3. Winters, C. E., Nucl. Sci. Eng., 17, 443 (1963).
4. Culler, F. L., Report ORNL-3452, p. 100 (1963) and Report ORNL-3627, p. 136
(1964).
5. Leuze, R. E., Baybarz, R. D., and Weaver, B., Nucl. Sci. Eng., 17, 252 (1963).
6. Baybarz, R. D., Weaver, B. S., and Kinser, H. B., Nucl. Sci. Eng., 17, 457 (1963).
7. Culler, F. L., Report ORNL-3627, p. 144 (1964).
8. Vaughen, V. C. A., and Baybarz, R. D., paper presented at ACS meeting, Atlantic
City, Sept. 12-16, 1965 (to be submitted for publication in Inorg. Eng. Chem.)
9. Baybarz, R. D., J. Inorg. Nucl. Chem., 27, 725 (1965).
10. Weaver, B., and Kappelmann, F. A. Report ORNL-3559 (1964).
11. Peppard, D. F., Moline, S. W., and Mason, G. W., J. Inorg. Nucl. Chem., 12,
141 (1959).
12. Peppard, D. F., Mason, G. W., Driscoll, W. J., and McCarty, S., J. Inorg. Nucl.
Chem., 12, 141 (1959).
13. Dyrssen, D., and Hay, L. D., Acta Chem. Scand., 14, 1091 (1960).
14. Baes, C. F., Jr., J. Inorg. Nucl. Chem., 24, 707 (1962).
15. Baybarz, R. D., Nucl. Sci. Eng., 17, 463 (1963).
16. Maxon, Ģ. W., Lewey, S., and Peppard, D. F., J. Inorg. Nucl. Chem., 26, 2271 (1964).
17. Baykarz, R. D., accepted for publication in J. Inorg. Nucl. Chem.

END
WID
LLLL
LLL
.
DATE FILMED
10/14/65