ORNL-2530.txt

 
    
  
    
   
   
   
      
    

AT A R | j HEEARCH ) IBRARY b4/
AR : N /

ORNL-2530
Chemistry=-General

 

SOLUBILITY AND STABILITY OF PuF3 IN
FUSED ALKALI FLUORIDE-BERYLLIUM
FLUORIDE MIXTURES
C. J. Barton

W. R. Grimes
R. A, Strehlow

 

 

 

OAK RIDGE NATIONAL LABORATORY

operated by
UNION CARBIDE CORPORATION
for the
U.5. ATOMIC ENERGY COMMISSION

 

CENTRAL RESEARCH LIBRARY
DOCUMENT COLLECTION

LIBRARY LOAN COPY
DO NOT TRANSFER TO ANOTHER PERSON

If you wish someone else to see this
document, send in name with document
and the library will arrange a loan.
 

 

 

 

-1 -

ABSTRACT

The solubility of PuF; in one NaF-LiF-BeF, mixture, three
NaF-BeF; mixtures and four LiF-BeF, mixtures was determined at
tempefatures ranging from about 550 to 650°C. Solubility-
-compositiqn diagrams for the binary systems at 565°C show that
in the LiF-BeF, system a minimum solubility of 0.25 mole % PuF;
occurs at about 63 mole % LiF while in the NaF-BeF, system the
minimfim solubility of 0.18 mole % PuF; occurs at approximately
57 mole % NaF. In both systems, the solubility increases
rather slowly on adding BeF, to the composition exhibiting
minimum soiubility but increases quite rapidly on adding
alkali fluoride to this composition. The highest solubility
observed at 565°C was 1.2 mole % PuF; in the mixture NaF-LiF-
BeF2 (56.5-17.5-26 mole %). In fused LiF-BeF,-PuF, mixtures,
the only plutonium species observed with the help of a polar-
izing miéroscope was PuF,; but another compound believed to
have the formula NaPuF, was found in the fused NaF-BeF, -PuF;
mixture containing 64 mole % NaF and in the mixture resulting
from the addition of PuF; to the ternary solvent composition.
No evidence of disproportionation of PuF; was found in the

course of the solubility studies.
 

-2 -

SOLUBILITY AND STABILITY OF PuF; IN
FUSED ALKALI FLUORIDE-BERYLLIUM FLUORIDE MIXTURES

C. J. Barton, W. R. Grimes and R. A. Strehlow

INTRODUCTION

An extensive series of studies conducted by W. R. Grimes

|
§
|
2
|
|
|
|
a

and co-workers in the Chemistry Division of Oak Ridge National

A

SRR

e o

Laboratory demonstrated that UF, can be dissolved in a number of
fused fluoride solvents to provide fuel for fused salt reactors.
The feasibility of the circulating fused salt reactor concept

was established by the successful operation of the Aircraft

3

e Oty o S R o e U o 1

L D) 23 Y T

Reactor Experiment.z’ The only fissionable species that have

. . 35
been considered for a fused fluoride burner reactor are v Fy

5

235 3 3
and U F; but the possibility of a ThF,-U 33F4 fused salt

   

breeder reactor seems promising,l Some consideration has also

been given to the use of UZ35C14 or U?3°Cl, as the fissionable

material in a fast neutron fused salt reactor.l Thermodynamic
4
data  which indicate that PuF; is more stable than UF; and that

PuF, is much more corrosive than UF,; prompted the choice of PuF;

 

1. W. R. Grimes et al., Reactor Handbook, Vol. 2, Section 6,
AECD-3646, May 1955.

2. A. M. Weinberg and R. C. Briant, "Molten Fluorides as Power
Reactor Fuels," Nuclear Science and Engineering, 2, 797-803
(1957) . | T

3. E. S. Bettis,

 

W. B. Cottrell, E. R. Mann, J. L. Meem, and
G. D. Whitman, "The Aircraft Reactor Experiment--Operation,”
Nuclear Science and Engineering, 2, 841-853 (1957) .

4. Alvin Glassner, ''The Thermochemical Properties of the Oxides,
Fluorides, and Chlorides to 2500°K," ANL-5750.

|
|
|
g
%
g
@
%
%
]
!
it
fi
|
i

   
      
 
  
  
   

as the

 

examin:
materi:
seems
than U
reCOgn§
obtain;
prelimé
the sd
solvené
in the
intere
includ

tempef

 

of oth

is bei:

Equip@
shown
chambé

of neg

 
 
 
 
 
  
  

Isotop
pluton

micros
 

.;as the plutonium-bearing species for the first experimental
ziexamination of the possibility of using Pu,;, as the fissionable
fmaterial in a fused fluoride power reactor (see Appendix A). It
;seems reasonable to expect that PuF; will not be more corrosive
ithan UF,; under proposed power reactor conditions but it is
recognized that data to support this belief will not be easy_to_
obtain. This report gives the results obtained to date of a
preliminary investigation of one aspect of this problem, namely,
the solubility and stability of PuF; in suitable fused salt
solvents. Alkali fluoride-beryllium fluoride systems were used

;in these studies since they are the only solvents currently of

finterest in the fused salt power reactor program. The studies

included determination of the effect of solvent composition and
temperature on the solubility of PuF;. The effect of addition
~of other fluorides on the solubility of PuF; in fused LiF-BeF,

is being studied and will be reported at a later date.

EXPERIMENTAL
Equipment

Two glove boxes constructed for fused salt studies are
hown in Figures 1 and 2. They are connected by an interchange
fchamber and constitute an extension of an inter-connected series
;of negative-pressure glove boxes used by personnel of the
sotopes Division, Oak Ridge National Laboratory, for handling
fplutonium isotopes separated by use of the calutron. The

microscope glove box shown in Figure 1 has plywood back, bottom

 

 
, sealed
and sides while the front and top are made of clear plastic
_ used to
material for maximum visibility. The eyepiece of a Zeiss polar-
; : . i arrange
izing microscope protrudes through an opening in the slanting
at Los
front of the glove box which is sealed with an airtight bellows
the fro

  

arrangement permitting vertical movement of the eye piece. The

. . , which a
microscope support is constructed of transparent plastic mate-

o
rial to facilitate observation of ohjects beneath the microscope Scobe a
- OX ., -
which are illuminated with light from a bulb in the base of the box (
insid s
microscope. The microscope glove box is a modification of one h ©
of 3" I
used at Los Alamos Scientific Laboratory for microscopic exam-
HF from
ination of highly toxic materials. The stainless steel glove _m
_ ] ] ) _ in seri
box shown in Figure 2 is a modification of a muffle box built
_ . _ , vent ca
for plutonium isotope work. Some of the more interesting
_ _ L while t!
features of this glove box are: (1) A heating well consisting :
) _ plete ri
of a 1-1/2" length of 6" I.D. pipe welded around an opening in
Th
the bottom of the box and connected to an 8'" length of 4" I.D.
) . ] below tl
pipe closed at the bottom. The interior of the well is heated
) . Air £fil
to a maximum of approximately 800°C by means of a 5" I.D. 2700~ -
to the |
watt tube furnace mounted on a jack beneath the box so that it
imatel
can be lowered to facilitate rapid cooling of the well. The tmately
. . istu:
box is kept cool by water flowing through a copper coil soldered (mois
1
to the bottom of the box around the well. (2) A discharge chute on a pa
s . . argon, °
consisting of a 12" length of 4" I.D. pipe welded to the right N
. itl
side of the box at an angle of 60° from the vertical, loosely tus witl
sealed on the inside of the box by a sliding door and tightly supply
hydroger

 

 
 

Scope
: the

one

 

;ted
é?OOM
it

e
idered
;chute
ght
1y

5137

sealed at the other end by a long plastic sock which is also

used to encase materials removed from the box. This discharge
arrangement is similar to that used on a number of glove boxes
at Los Alamos Scientific Laboratory. (3) A small protrusion on

the front of the box at the left side having a glass top over

which a low-power long focal length (9X and 18X) binocular micro-

- scope and light is mounted for examination of objects inside the

box. (4) Sodium fluoride traps (not visible in Figure 2) mounted

- inside of the box on the back wall, consisting of two 30" lengths
~of 3" I.D. copper tubing filled with 1/8" NaF pellets to remove
- HF from exhaust gases. The first one of the traps connected

in series is heated to 100°C by means of a heating coil to pre-

vent caking of the NaF through formation of HF-rich complexes,

'.While the second trap is unheated in order to effect more com-

- plete removal of HF.

The pressure in the boxes is maintained 1/2 to 1" of water

:ibelow that of the laboratory by means of an exhaust system.

“Air filtered through a CWS filter can be alternately supplied
I;to the box from the room (50% RH), from an '"Lectrodryer" (approx-
fimately 15% RH) or from a drying tower packed with Drierite

- (moisture content not determined). Bellows-type valves mounted
;on a panel below the front face of the box control admission of
;_argony 95% argon-5% hydrogen mixture, HF, and vacuum to appara-
;fus within the box. The argon-hydrogen mixture was used to
fsupply a reducing atmosphere over the melts instead of pure

hydrogen because of safety considerations. The gas tanks and
vacuum pump are located outside of the laboratory.
Materials

Plutonium trifluoride used in these studies was high-purity
material supplied by Los Alamos Scientific Laboratory. Solvent
mixtures were purified prior to being introduced into the glove
box by treating fused mixtures with gaseous HF and H, at about
800°C and cooling under an inert atmosphere. The composition of
some of the purified mixtures was modified either by addition of

crystalline BeF, purified by the above-described procedure or by

addition of Harshaw optical grade LiF.
Apparatus

A picture of the filtration apparatus used in this inves=
tigation is shown in Figure 3. This apparatus is a modification f
of that used earlier in these laboratories for solubility
determinations,* the main change being a reduction in the diam-
eter of the sample container in order to permit smaller charges
to be used. The copper bellows permit the filter medium to be
kept out of contact with the charge material prior to the actual
filtration in order to minimize clogging of the filter medium.
The apparatus shown in Figure 3 was assembled by heliarc welding
in the Special Services Shop of the Y-12 Plant. An improved
version of the apparatus used for the last few experiments per-

formed was assembled in the Welding and Brazing Facility, ORNL,

 

* Designed by B. H. Clampitt, formerly a member of the
Chemistry Division, Oak Ridge National Laboratory.

   

by braz
hydroge:
use of
effecte

Thi
thermoc:
platinu
proport:
recoxrde:
the rec:
was alse
The rec;
2°c. T
colils wé

Th
controlj
mocouplé
to a poé

controlé

Experime

The
the gasf
the stai
Toom ten

tained,:

sisting

 
 

.purity
Slvent
%glove
1 bout
tion of
tion of

2 Oor by

AVES -

ication

diam~
harges

to be
dium.

ved
S per-—

ORNL,

 

 

 

éctual;

Welding:

by brazing the top part together with gold-nickel alloy in a

ydrogen atmosphere at about 1000°C. This procedure permitted
use of thin-wall tubing for the thermocouple well and also
effected remcval of oxide film from the nickel surfaces.

The temperature of the melt surrounding the tip of the

thermocouple well was measured and controlled by means of a

 

platinum-platinum 10% rhodium thermocouple connected to a Brown

roportional controller. It was recorded on a 300-1000°C

 

ecorder. A portable potentiometer provided an e.m.f. to keep

he recorder on scale while the temperature was below 300°C and

as also used to check the potential of the thermocouple junction.
The recorder and potentiometer readings generally agreed within
:OC, The temperature near the center of the furnace heating
coils was indicated on a Brown Protect-0O-Vane controller.

The temperature of the heated NaF trap was measured and

controlled by means of a chromel-alumel thermocouple, in a ther-
mocouple well extending from the center of one end of the trap

to a point near the middle, connected to a Brown Pyr-0O-~Vane

controller.

 

 

Experimental Procedure

 

The filtration apparatus (Figure 3) was first connected to
the gas tanks and vacuum pump through the manifold system below
the stainless steel glove box (Figure 2) and vacuum tested at
?oom temperature. After a vacuum of about 50 microns was Ob-

tained, the filter stick was removed and charge material con-

Sisting of 5.0 to 5.5 grams of solvent mixture and enough PuFj
to give 6.0 grams total charge was transferred from plastic vials

into the filter apparatus through a long-neck metal funnel. The
ratio of solvent to PuF,; was adjusted to provide at least-twice

the amount of'PuF3 that was expected to be dissolved. After the
filter stick was replaced, the filter apparatus was re-connected

and vacuum tested. It was then filled with argon-hydrogen mix-

ture introduced through the filter stick, which was positioned so

that its lower end was close to the fused salt mixture, providing'

reducing atmosphere to help keep the plutonium in the trivalent
state. Argon-hydrogen mixture was used in preference to pure
hydrogen because of safety considerations. Gaseous HF was mixed
with the argon-hydrogen mixture before the temperature of the
filter bottle reached 250°C in order to minimize hydrolysis
resulting from adsorbed water on the surface of the fused salt
and PuF; crystals or to recoavert any hydrolysis products to
fluorides.

The filter bottle and contents were held at the desired
filtration temperature ¥ 5°C for two hours with a slow flow of
the mixed gases over the surface of the liquid and then filtra-
tion was effected'by applying a vacuum to the filter stick and
argon pressure of 3 to 6 1lbs to the surface of the liquid while
the bellows was compressed to bring the filter stick into con-
tact with the liquid for a period of 5 to 10 minutes. The

filtrate and residue were allowed to cool slowly to room temper-

ature in an argon atmosphere. The filter stick was then removed

 

- and cu

of sol
filtra
the ab
peratu
comple
cut op
remove:

A
filtra:
Flexibj
leading
bottle.
filtrat
the fij
quite rx
300°0) .
of the
cooling
availafi
after fi
arrange
moving
are thé

the glo

 
The
wice
;r the
iected

mix-

1lent
ire
mixed

the

   

 

- yials

oned SO

sviding

 

    
 

  

and cut intoc sections for removal of the filtrate. A new charge

  

of sclvent and PuF, was added to the residue from the previous

 

  

 

filtration, a new filter stick was inserted in the apparatus and

 

  

the above filtration procedure was repeated at a different tem-

 

  

 

perature. After the required number of filtrations had been

  

  

completed with each solvent composition, the filter bottle was

  

cut open and the residue from the final filtration experiment was

 

removed for examination and analysis.

  

A variation of the above-described procedure was used in

 

  

filtration experiments with one solvent (71.3 LiF-28.7 BeF,).

   

Flexible hose connections were placed in the three gas lines

 

 
 

leading to the filtration apparatus. This permitted the filter

 

 
 

bottle to be withdrawn from the furnace at the conclusion of the

 

  

filtration period while argon gas was still being pulled through

  

the filter stick, causing the filter bottle and contents to cool

quite rapidly (approximately 100° per minute between 600 and

  

300°C). This procedure eliminated any possibility of a portion

 

0of the filtrate running through the filter medium during the

 

  

cooling period and it had the advantage of making the filtrate

  

available for examination at room temperature about 30 minutes

  

after the end of the filtration. It was also possible with this

 

 arrangement to agitate the contents of the filter bottle by
fmoving the apparatus up and down. The principal disadvantages
fare the hazards associated with the handling of hot objects in

jthe glove box and the extremely small crystal size of the

 

  

 

  

 

 
 

- 10 -

rapidly cooled mixtures which makes microscopic identification
of crystals difficult.
Results

Table I gives the composition of solvent mixtures which
were analyzed chemically. The first column shows the nominal or
intended composition of the mixtures while the last column shows
the composition calculated from analytical data. The composi-
tion of solvent mixtures not included in Table I was calculated.

Solubility data obtained with LiF-BeF, mixtures are given

in Table II; NaF-BeF, solubility data are contained in Table III; |

and the solubility of PuF; in one NaF-LiF-BeF., mixture is shown
in Table IV. The data are shown graphically in Figures 4 and 5,

omitting doubtful results, as a plot of the log of the molar

concentration of PuF; versus 1/T (PK). Figure 6 shows solubility'i

data at three temperatures as a function of solvent composition
for LiF-BeF, mixtures while Figure 7 shows a similar plot for
NaF-BeF, compositions. The data obtained with the ternary mix-
ture (Table IV) were used to extend the data in Figure 7 by
considering 56.5% NaF and 17.5% LiF as equivalent to 74 mole %
NaF. A comparison of the two systems, using interpolated solu-
bility values at 565°C is shown in Figure 8, again using data on
the ternary system to extend the NaF-BeF, curve.

Dissolved portions of seven samples were examined spectro-

photometrically for Put* content. No Pu™ was detected in any

of the samples. The limit of detection varied from 0.2 to 3.0%,

 

 
 

 

 

 

  

 

                                 

"SOTJI01BIOQR] OM] AQ PSBUTIRIQO SIINS8JI JO 98BIDAY 4

 

TT 68°¢

 

 

 

 

‘qed 0°92=dTT §°LT-dBN G°96 %86 °§ BN 98°0¢ cdeg @Z=ATT 9T=dEN 99
¢qed L e%~dTT €°96 *2T "T11 86 °0T tqeg v¥—dTT 9§
M ‘asd ¢°8%—dTT L 164 0° 2T 06°6 ‘qed 0S=dTI1 06
J ‘qod 6°T¢~dTT T1°89 x08 ° 8 LY FT tded ¢ree~dATT L°99
‘ded 9°9g=deN ¥°¢9 *x8% ° L 21 ¢¢ ‘god 9¢=deN %9
‘god 0°¢H—dEN 0°LS *¥9°8 Y1°62 ‘deog ¢h-deN LG
tged £°0s=deN L 6¥ x90°0T 25°G¢ ‘dad 0g~deN 0%
(6 oTOW) % “IM) % “IM) (0 o1owU)
uorylirsodwo) palrernore)d ag T JO BN uotlrsodwo) TeUTWON

 

SOJIN}XTN JUSATOS JOo uoritrsodwo)

I 9T19®eL

for

  

\ shows
yle IIT

 any
0 3.0%,

?ctro~

sition
data on

7 mix—
Dy

316 Yo
‘solu-

 
 

- 12 -

Table 11

Solubility of PuF, in LiF-BeF, Mixtures

 

Solvent Filtration ' _ | _ - Cfi
Composition Temperature Plutonium in Filtrate T
Mole % Li (°c.) Wt. % Pu Mole % PuF,
51,7 - 463 | 1.02 | 0.16
" 549 | 2,44 0.38
" 599 . 2.96% 0.47%
" 654 5,76 0.93
56,3 494 1,04%* 0,15
i 560 | 1,89%x 0.28
" 602 3.15%% 0.48
" 653  5.47%% 0.86
" 550 1.98 0.30
i 599 2.04% 0.31%
" 649 6.24% 0.98%
65. 4 532 - 1,15 0.16
" 600 o 1.78% 0.27%
" 643 4.30 - 0.63
71.3 546 | ,' 4,00 0.56
" 597 | 5.90 0.85 *
" 650 8.48 1.26

* Doubtful result, excluded from Figure 4. /
** Results obtained with a pre-fused mixture. Other results
with this composition were obtained with mixtures pre- "
pared by mixing LiF, Li,BeF, and PuF, in the filter bottle

 

 
 

27%
s
56

85

 

26

iresults
- pre-
er bottl

 

Table III

Solubility of PuF, in NaF-BeF, Mixtures

Solvent Filtration
Composition Temperature . Plutonium in Filtrate
Mole % NaF (°C) Wt. % Pu Mole % PuF,
49.7 552 1.18 0.22
" 600 1.73 0.33
" 600 1.79 0.34
" 651 2.72 0,52
57.0 538 1.17% 0,22+
" 552 1.63% 0.31%
" | 600 1.35 0.26
" 600 1.26 0.24
" 609 1.26 0,24
" 650 1.48% 0,28%*
" 652 2.21 0.42
m 706 3.40 0.66
63.4 550 1.54 0.29
" 598 2,43 0.46
" 600 2.00% 0,38%
" 650 4,40 - 0,85

* Doubtful result, excluded from Figure 5.

 

 

 
 

 

- 14 -

Table IV

Solubility of PuF; in NaF-LiF-BeF, (56.5-17.5-26 mole %)

Filtration Temperature

(°C)

 

500
554
565
600
634

655

Plutonium in Filtrate

Wt. % Pu Mole % PuF,
2.92 0.51
7.68 1.43
2,61% 0.,46%
7.59 1.41

13.0 2.58
6.45% 1.18%*

* Doubtful result, excluded from Figure 5,

 

depend.
size o
a care:
micros
sampléa
chemi c:
examins
graphig
Appendi

S ]
have b{

or by t

resulti
ed witt

similar

 

were ifi
nickel |
such mé
total fi
NaF, Be
contacfi

6500C 1

two sam

 
 

   

snding upon the plutonium concentration in the samples and the

  
 
  
   
  
   
   
  
  
  
   
  
  
  
  
  
  
  
  

» of the sample analyzed. Since these results, coupled with
féful examination of each sample by means of the petrographic
iéscope, demonstrated that all of the plutonium in these

_lés wfis in the trivalent form, other samples were analyzed

mically only for total plutonium content but microscopic

mination of the samples continued. A summary of the petro-

fphic observations made by R. A. Strehlow is contained in
pendix B.

Since Pu™™ produced by disproportionation of PuF; could

e been reduced by hydrogen in the atmosphere above the melt
;fby the nickel walls of the container, it seemed advisable to
ok for other evidence of the absence of disproportionation of
uF; . In similar studies with UF;-containing melts, U metal
_E$ulting’from disproportionation of UF; was found to have alloy-
'fijwith the container wall. In order to determine whether a
’milar ailoying effect occurred when PuF;-containing melts

re in contact with nickel, the bottom section of two welded
hickel filter bottles (Figure 3) that had been in contact with
Such melts were scraped as clean as possible and analyzed for

otal plutonium content. One sample had contacted a mixture of

 

NaF, BeF, and PuF; at 600°C for two hours while the other had

contacted a similar mixture at temperatures varying from 550 to

6509C for approximately eight hours. The plutonium found in the

two samples, amounting to 0.05 and 0.17 mg respectively, could

 

 

 

 

 

 

 

 
 

 

 

 

be accounted for by the trace of fused salt remaining on the
nickel. ©No evidence of disproportionation of PuF; in fused

alkali fluoride-~beryllium fluoride melts under the conditions

maintained in these solubility studies has been observed to date.

Discussion

Straight lines correlate the data in Figures 4 and 5 with-
in the limits of reproducibility of the results. Changes in
composition of the solvent mixtures seem to have little effect
on the slope of the lines. It appears, therefore, that the heat
of solution of PuF; in these solvents is essentially constant
for the temperature and composition ranges studied.

The solubility-composition plots, Figures 6, 7, and 8,
show that the solubility of PuF; in the melts goes through a
minimum with a sharp increase on the alkali fluoride side and a
gradual increase on the BeF, side of the minimum. This result
is not unexpected since the BeF,-PuF; eutectic probably contains
a very low concentration of PuF; while the alkali fluoride-PuF;
eutectics almost certainly contain a much higher PuF; concentra-
tion.

Inspection of Figure 8 and the phase diagrams of the binary

systems LiFmBeFZS and NaF—BeFZ6 shows that a higher solubility

 

5. D. M. Roy, R. Roy, and E. F. Osborn, J. Ceram. Soc. 37, 300
(1954); Novoselova, Simanov, and Yarembash, J. Phys. Chemn.
USSR, 26, 1244 (1952).

6. D. M. Roy, R. Roy, and E. F. Osborn, J. Ceram. Soc., 33,

85 (1950); Ibid, 36, 185 (1953); E. Thilo and H. Schroder,
Z. physik. chem., 197, .41 (1951).

 

of F
the
temg
molé
molé
melt
mole
PuF,
temy

sSOlv

Tab]
were
the$
of
equ]
des]
men;
£1ue
sib:
dat:
sol
med

of

 
 

 

', 300
?hem.

3,
der,

of PuF, at 565°C can be achieved in the LiF-BeF, system than in

the NaF-BeF, system for solvent mixtures having the same liquidus
temperature. For example, the LiF-BeF, mixture containing 69
mole % LiF which is liquid at 520°C, will dissolve about 0.50
mole % of PuF; at 565° while the comparable NaF-BeF, mixture
melting at 520°C contains 62.5 mole % NaF and will hold 0.26
mole % PuF,; in solution at 565°. If much more than 0.5 mole %
PuF; is needed for a circulating fuel reactor having a minimum
temperature of 5652, it may be necessary to resort to a ternary
solvent in order to keep the liquidus temperature of the sclvent
at an acceptably low figure.

Approximation 25% of the filtration data recorded in
Tables II, III, and IV (those results marked with asterisks)
were excluded from Figures 4 and 5 and some of the points in
these graphs do not lie very close to the lines. This scatter
of data reflects to some extent the difficulty in obtaining
equilibrium data with small amounts of fused salts. It was
desirable to use small amounts of materials for these experi-
ments because of neutron production in plutonium~beryllium
fluoride melts. However, an analysis of factors having a pos-
sible influence on the results may be of help in evaluating the
data. Low results could be due to failure to saturate the
solvent because of inadequate equilibration time, absence of a
means of agitation (see page 9), or lack of a sufficient amount
of PuF; in the mixture. There are two fairly obvious explana-

tions of high results. A temperature gradient is known to
 

 

 

 

 

exist in the heated zone occupiéd by the filter bottle during
filtration experimenté, with the temperature decreasing in going
from the bottom to the top of the chamber. If the thermocouple
junction was above the surface of the liquid in the%filter
bottle, the temperature indicated by the thermocouple was prob-
ably lower than that of the liquid. Some of the filter bottles
used in these studies had thermocouple wells too short to reach
the surface of the liquid and it is also possible that éfrors in
the determination of the‘temperature of the melts resulted from
the thermocouple being inadvertently pulled part way out of the
thermocouple well. The other possibility is that a portion of
the solvent or a low-melting ternary eutectic having a lower
plutonium content than that of the original filtrate ran back
through the filter after partial solidification of the filtrate
occurred. (See Experimental Procedure, p. 7)

Confidence in plutonium values reported by the analytical
laboratory was strengthened by having a number of samples
reanalyzed under different sample designations. The results,
shown in Appendix C, indicate that satisfactory reprdducibility
was obtained although the differences for some samples were
greater than expected from the precision of the potentiometric
plutonium determination. It is possible that some of the early

samples which were ground but not sieved were not entirely homo-

geneous .

 

 
 

ng
going

uple

brOobh-
fitles
reach
:rs in
from

f the

 

er
ack

irate

ical

 

ility

&ric
early

~ homo-

Acknowledgments

 

The writers are grateful to F. N. Case of the Isotopes
Division of ORNL, for help in planning the glove boxes used
in this investigation; to R. J. Sheil, Chemistry Division,
ORNL, for designing the gas manifold system; to H. Insley,
Consultant, ORNL, for help in designing the microscope glove
box; to personnel of the Analytical Division of ORNL under
the direction of L. T. Corbin, J. H. Cooper and W. R. Laing
for performing the numerous analyses required in these
studies; and to personnel of the Los Alamos Scientific
Laboratory including R. D. Baker, W, J. Maraman, Charles Metz
and K. W, R. Johnson for helpful advice during the planning
stage of the project and for furnishing the high purity PuF,

used in this investigation.

 

 

 

 

 

 

 

 

 

 

 
 

 

 

 

 

- 20 -

Appendix A
Thermodynamic Calculations for Reactions at IOOOOK.

AF for Reaction AF per Gm. Atom

 

Reaction as written (k Cal) of F, k Cal
4UF, — 3UF, + U° + 12 + 1.0
4PuF, — 3PuF, + pu’ + 214 + 18
2UF, + Cr° — 2UF, + CrF, + 31 + 3.9
2UF, + 3Cr® — 20° + 3CrF, + 120 - + 20
2PuF, + Cr® — 2PuF, + CrF, -~ 80 - 10
2PuF,; + 3Cr° — 2pu® + 3CrF, + 194 + 32

 

1
fc
ST
Un
ap
mi

of

 
 

“Atom
Cal

 

)

 

Optical Properties of Fluoride Mixtures Containing PuF,

and residues obtained in the study with a principal objective of
identifying the plutonium containing phase.
preparations for the work were made in a glove-port plexi- .
glas box (Figure 1)

scope.

- 21 -

Appendix B

R. A. Strehlow

Petrographic examinations were carried out on filtrates

The optical character of the various plutonium

All of the slide

in which was mounted a polarizing micro-=

compounds permitted distinction to be made from the crystals

of the various solidified solvents.
contained one or more of the compounds:

2 . 1
NaBeF,, LiBeF,

1

refractive indices of less than 1.4.

fractive indices an initial observation was made of the growth

form of all phases of higher index by examining the crushed

Li,BeF, , '

Na,BeF,,

and BeF, . All of these compounds have

Inasmuch as the plutonium phases had much higher re-

specimen in an oil with a refractive index of about 1.35.

The solvents generally

2

Under these conditions all phases of appreciably higher index

appear quite heavily bordered and are easily located in the

microscope field.

of the phase indicated which of the possible compounds con=

 

1.

2.

D. M. Roy,
300 (1954)
D. M. Roy,
85 (1950).

R. Roy and E. F. Osborn,

R. Roy and E. F. Osborn,

J. Ceram.

J, Ceram,

Soc.,

Soc.

37

33

r—

7
g

Observation of the color and birefringence
 

 

 

 

 

 

 

 

- 22 -

taining plutonium was present: Pu(III) fluorides are blue

and Pu(1V) fluorides tan. (No PuF, was observed in the fil-
trates or residues in this study.) The maximum birefringence
of PuF; observed in this study is about 0.003. The inter-~
ference color (observed with crossed polarizing discs) is
approximately that of the crystal observed in "white"™ light,
pale blue. The compound NaPuF, appears reddish blue and has

a maximum birefringence of about 0,02. The optical properties
of these compounds are compared in Table B-1 with those of some
Ce*3, U*®, and Ut4 compounds.

The Pu(III) ion is about the same size as the Ce(III)
and U(III) ions and differs from them in its polarizability
which is, of course, shown by the differences in refractive
indices.

The most common growth form of PuF, as crystallized from
the solvents used in the study was that of thin sheets which
could be observed with their maximum birefringence only when
they existed as platy inclusions in the»solvent crystalso.
Under these conditions they exhibited their maximum bire-
fringence and parallel extinction. No compound was found
which could be considered a LiF-PuF, binary compound in LiF=
BeF,~PuF, melts and no change in the refractive index of PuF;
was seen which would indicate solid solution of LiF in the

PuF, .

The growth form of the PuF, is strongly dependent on the

 

 
Lue
filw
Lgence
s
ght,

l'hés

 

 

 

erties

of some

 

Table B-1

Refractive Indices and Optical Character of Some

 

Plutonium, Cerium, and Uranium Fluorides

 

Me N

 

 

 

 

 

 

 

 

 

Description
PuF, 1.683 1.685 Uniaxial (-); blue*
UF, ~2 Isotropic; red
Cel, 1.607 1.613 Uniaxial (-); colorless
Ty |  ”7 Description
PuF, 1.577 "16629 2v = =65°
pink green-brown
UF, 1.550 . 1.594 2v = -57°
blue-green  green-brown
Mgy | Ne Description
NaUF, 1.552 1.564 Uniaxial (+)
NaPuF, 1.534 | 19552 Uniaxial (+) (length slow)
NaCeF, 1.494 S 1.512 Uniaxial (+)
* The sign ofPuF; as observed in this work does not agree

with that listed by E. Staritsky and A. L. Truitt in

"The Actinide Elements,' NNES, IV-14A, McGraw-Hill Book Co.,

 

 

 

 

 

 

 
 

 

 

 

  
    
   
   
  
  
 
 
  
   
   
   
  
   
   
  
 

cooling rate and is described in Table B-2,

Table B-2

The Effect of Extremes in Cooling Rates

 

On Appearance of PuF; in LiF-BeF, Mixtures

Rate Appearance

Rapid Asterisms <5l across frequently imbedded in
glassed solvent. Occasional identifiable
crystals near 3-84 in diameter.

Slow Large plates (>50n across, 5 thick) gifiing

the uniaxial (~) interference figure; solvent
well crystallized.

A sample prepared by mixing LiF-NaF-PuF; (50-33.3-16.7
mole %) was examined petrographically after fusing. A 1arge
amount of a plutonium containing phase, LiF crystals, and
NaF-LiF eutectic aggregates were observed. Since the plutonium:
compound was found only in mixtures containing sodium fluoride
and since the optical properties were similar to those for the
known compounds NaCeF, and NaU¥,, this plutonium compound was
designated as NaPuF, .

It is interesting to note that NaPuF, was observed as the
sole plutonium containing phase in the filtrate from the mixture
NaF-LiF-BeF, (56.5-17.5-26 mole %). Some NaPuF, was seen in the
residue from the mixture containing 64 mole % NaF but not;in ’

the filtrate. This indicates that the isotherms as shown in

 

    

Figure 7 probably cross the NaPuF, -PuF,; primary phase field

 
 

in

ing
lvent

 

6.7
;rge
?tonium
?oride
Qr the

d was

és the

mixture

in the
t in -
1 1in

2 1d

—~ 25 -

boundary at a relative composition of very near 64-36 mole
NaF per mole BeF,. PuF,; was observed in both filtrates and
residues of other NaF content.

A comparison was made of transmission spectra by means
of a Leitz pupillary spectroscope., The principal difference
between NaPuF, and PuF,; is an absorption band between 6500 2
and 6650 K for PuF; which exists at 6660 K to 6800 & for
NaPuF, . For the PuF; there is also more transmission between
5200 fi and 6000 ® which presumably accounts for the difference

in color between NaPuF, and PuF,.

 

 
 

S
oVt
e
g
o
T
i

Appendix C

 

 

Analysis of Re~submitted Samples

 

 

 

 

 

 

Original Sample % Pu Re-Check % Pu
A5 1.36 G2 1.33
B5 1.41 E8-C9 1.15-1.22
A7 2.16 G3 2.25
B3 1.37 Gé6 1.60
D6 1.77 ' G7 1.68
B7 1.82 H3 1.74

ET : 2.88 H8 3.03

 

 

 

 
 

02867 OLOHd
A313185y 10NN

 

 

 

 

1.22

 

 
 

 
 
 

 

 

 

 
 

/@% =

.

 
PR U U R s

 

 

 

SYZ0t OLOHd
a3a1dISSYTIONN

m wwm el . 3 /M/%&wa L X
T

1

NCL ASSIF)
PHOTO 298

 
       
   

650

TEMPERATU
600

RE (°C)
550

UNCLASSIFIED
ORNL—LR—DWG 29605

500

 

   
 

\//74.3 LiF-2

8.7 B(-'aF2

|

 

   

N\

 

    

N

 

v ONONL

AN

 

N

N,

 

N\
NN

o,

~

 

0.5

N\
A

~

~

 

PuF, (mole % )

N

N
A

NN

 

 

65.4 LiF-34.6 BeF,”

®

N\

N\

= 51.7

LiF-48.3Bef, _|

 

 

0.2

 

 

 

 

56.3LiF-4

 

| |

3.7BeF,

 

|

 

 

 

 

0.1

 

10.5

  

  
    

11.0

11.5

2.0 1
10%/°K

2.5

13.0

Fig. 4. Solubility of PuF3 as a Function of Temperature for LiF-BeFy Solvents.

13.5 14.0

     
     
 
 

 
 

05

 

 

0.5

PuF; (mole %)

0.2

O

-31-

UNCLASSIFIED

ORNL—LR—DWG 29606

TEMPERATURE (°C)

650 600 550 500

 

| 1 | |
NJ{_56.5 NoF-17.5 LiF- 26 BeF,

l

 

 

 

 

 

 

 

/
/

7/
'./
o /|

 

 

 

 

 

\\
\\
| 57.0NaF-43.0BeF~
~
~
~

 

 

 

 

 

 

 

~
\\
\+

49.7NaF-50.3Be

 

 

R

 

 

10.0 105 10 1.5 120 125
10%/°K

13.0

13.5

Fig. 5. Solubility of PuF3 as a Function of Temperature for NaF-BeFy and NaF-LiF-
BeFy Solvents,
-32-

UNCLASSIFIED
ORNL-LR-DWG 29607

 

e

Pl
o
\M/

i

 

 

 

 

 

650°C

600°C

550°C

 

 

 

 

 

 

 

 

 

 

 

 

1.4

1.2

1.0

D
O

(% ®jow)

©
O

€4ng

0.4

0.2

60 65 70 75
LiF (mole %)

55

50

 

. Solubility of PuF3 as a Function of Solvent Composition for Li -BeFy

Fig. 6

Mixtures.

 
 

Fig. 7.
Mixtures.

2.5

2.0

1.5

PuF3(moIe %)

0.5

0

Solubility of PuF3 as a Function of Solvent Composition for NaF-BeF,

-33-

 

UNCLASSIFIED

ORNL-LR-DWG 29608

 

 

 

 

 

 

 

/

~.

 

/

N

 

650°C

600°C_~

yd

/

 

/

 

56.5 NaF — {7.5 LiF ————a——\

 

 

 

 

\ 550°C |
50 5

 

5 6

0 6

5 70
NaF (mole %)}

 
-34-

 

UNCLASSIFIED
ORNL-LR-DWG 29609

 

1.2

 

 

o8 32.

 

 

0.6 ;)

NaF

 

 

 

PuFz (mole %)

 

0.4 NLE
N

N0

47.

56.5 NaF =17.5 LiF —1————3m—
£
<

 

 

 

 

 

 

 

 

 

 

 

 

50 55 60 65 70 75 27
NaF OR LiF (mole %) 59.

Ei
L‘
L
A
L
J
]
E
E
S
M
GE
J
R
H
F
A
A
M
44, K
T
A
c
c
D
D
F
D
J
M
W
G
C
J
¥
E
. G
Fig. 8. Solubility of PuF3 in LiF-BeFy and NaF-BeFp at 565°C, 62, S
R

E

F

¥

E

I‘

£

 

 

 
 

C. E. Center

Biology Library

Health Physics Library
Central Research Library
Reactor Experimental
Engineering Library

Laboratory Records Department
. Laboratory Records, ORNL R.C.

M. Weinberg
Emlet (K-25)
Murray (Y-12)
Swartout
Taylor
Shipley
Lind

Nelson
Keim

Frye, Jr.
Livingston
MacPherson -
. Culler

. Snell
Hollaender
Kelley
Morgan
Linceln

. Householder
Harrill
Winters
Cardwell -
Phillips
VonderLage
Billington
Lane
Skinner.
Joxrdan
Boyd

. Barton

. Blakely
Blander

F. Blankenship
M. Blood
‘Cantor

B. Ewvans

A. Friedman

A. Gilbert

R. Grimes
Insley

Kertesz

Langer

THEouDwE oo me g e

SEHwLmPE NS

CHEEREAVOAMECOD IR G RUD OO P RARPPAERLOROEEGL DR
RN -V P

INTERNAL DISTRIBUTION

70.
71,
72.
73.
74.
75.
76.
/7.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94,
95.
96.
97.
98.
99.
10G.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.

 

ORNL-2530
Chemistry-General
TID-4500 (13th ed., Rev.)
February 15, 1958

R. Moore

Nessle

Newton

Overholser

Redman

Shaffer

Sheil

Smith

Strehlow

Sturm

Thoma

Ward

Watson

Weaver

. Bredig

Datz

Minturn

Dworkin

Smith

Robinson

Miller

Alexander

Dickison

Keilholtsz

Manly

Fraas

Mann

Bruce

Bettis

Milford

Charpie

Ergen

Lindauer

White

Mevyer

Vaughn

Ferguson

BRlanco

Long

Cathers

Carr, Jr.

Goeller

Blomeke

Ricci (consultant)

E. Larson (consultant)
H. Emmett (consultant)
Eyring (consultant)

T. Seaborg (consultant)
ORNL - Y-12 Technical Library,
Document Reference Section

>1flg§%?1g?-§umcomum

o

CEPOLUEZOLAYI P UFERAR AN P EORE P ROPFINRO0NERIFN IR L LD DO
pgngfifqhagig:fifagfipd?:w*msm;wtnPg:jzzzfo g - M
    
       
    
    

EXTERNAL DISTRIBUTION

 

119. Division of Research and Development, AEC, ORO
120-674. Given distribution as shown in TID-4500 (13th ed., Rev., Feb. 15, 1958)
under Chemistry-General category (75 copies - 0OTS)