WU e e e crion

DOCUMENT CCLLECTION
3 4456 0548369 7 ;

 

 

 

 

 

 

 

 

 

—_——as

ORNL-35%96
UC-4 — Chemistry
TID-4500 (30th ed.)

 

ADAPTATION OF THE FUSED-SALT
FLUORIDE-VOLATILITY PROCESS TO
THE RECOVERY OF URANIUM FROM
ALUMINUM-URANIUM ALLOY FUEL

M. R. Bennett
G. |. Cathers

 

 

 

OAK RIDGE NATIONAL LABORATORY

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

 

CENTRAL RESEARCH LIBRARY
DOCUMENT COLLECTION

LIBRARY LOAN copPY

DO NOT TRANSFER TO ANOTHER (3 &Te] )
: O LRV TV ARV T T Y TP else to see this
document, send in name with document
and the library will arrange a loan.

   
ORNL-3596

Contract No. W~ThO5-eng-26
CHEMICAL TECHNOLOGY DIVISION

Chemical Development Section B |

ADAPTATION OF THE FUSED-SALT FLUCRIDE-VOLATILITY PROCESS TO THE
RECOVERY OF URANIUM FROM ALUMINUM-URANIUM ALLCY FUEL

M. R. Bennett G. I. Cathers

JUNE 1964

 

OAK RIDGE NATIONAL TLABCRATORY
Oak Ridge, Tennessee
operated by
UNION CARBIDE CORPORATION

for the |
U.5. ATOMIC ENERGY COMMISSION

AK RIDGE NATIONAL LABORATORY LIBRARIES

(IRRRRARIN

 

 

 

 

 

Q.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3 4456 054839 7

 
il o

~1-

ADAPTATTION OF THE FUSED-SALT FLUCRIDE-VOLATILITY PROCESS TO THE
RECOVERY OF URANIUM FROM ALUMINUM-URANIUM ALLOY FUEL

M. R. Bennett G. I. Cathers

ABSTRACT

Experimental results are presented of studies conducted
to adapt the fused salt-fluoride volatility process to the
recovery of uranium from aluminum-~base enriched uranium fuel.
In this process, the uranium is recovered from spent aluminum-
base reactor fuels by dissolution in fused fluoride salt by
hydrofluorination, fluorination,and volatilization of uranium
as UFg, and, finally, purification of the product UFg in a
sorption-desorption step. The adaptation consists in use
of a fused salt system which has a high A1F3 solubility with-
in desirable temperature limitations, and which gives adequate
dissolution or hydrofluorination reaction rates. The KF—Zth-
AlF3

dissolution or hydrofluorination rate studies were made of

salt system was found to be satisfactory. Extensive

fuel in this salt on a laboratory scale to show the feasi-
bility of the process. Two types of flowsheet were investi-
gated. Process tests were also made to demonstrate the
completeness with which uranium is volatilized from the dis-
solution salt as UF6. Some of the chemistry involved as
side reactions in the dissolution step were studied. This
modified fluoride volatility process for aluminum-base

fuels works well and seems to be amenable to scaleup.

INTRODUCTION

The recovery of uranium from spent reactor fuels by a process based

on the volatility of UF6 offers the main alternative to aqueous dissolution-

solvent extraction processes for the recovery and decontamination of

 
 

-2~

uranium values. A fused-salt fluoride-volatility process under develop-

ment at Oak Ridge National Laboratory for the recovery of enriched

uranium has given satisfactory results on a pilot plant scale with ir-

radiated zirconium-uranium alloy reactor fuel.l’2 This process consists

of dissolution of fuel in fused fluoride salt by hydrofluorination,

volatilization of uranium as UF6 from the melt by direct fluorination,

followed by purification of the UF6 product from fission product con-

tamination through use of sorptive fluoride beds. Since fuels of the
aluminum-uranium alloy type are also used extensively for enriched fuel

elements it appeared desirable to adapt the process, if possible, to }
fuels of this type. This would make feasible the handling of both en- -
riched Zr-U and enriched Al-U alloy fuels in the same plant if economically

Jjustified.

Adaptation of the fused-salt volatility process to aluminum fuel
mainly involves modification of the head-end or dissolution step of the
process. Other parts of the process, namely those involving volatili-
zation and purification of the UF6 product remain essentially the same
as those used in processing of Zr-U alloy fuel. The laboratory work
described here was directed primarily towards defining the chemistry
and dissclution rates prevailing in the dissolution step. The salt
compositions used in these process studies were based on recommendations
made by R. E. Thoma and B. J. Sturm of the Reactor Chemistry Division
as a result of a cooperative effort at determining the phase diagrams
of possibly useful salt Systems.3 Further testing of the process for
aluminum-base fuel is being conducted by R. W. Horton et al., in the «
Unit Operations Section of the Chemical Technology Division. Eventually,

pilot plant tests with fully irradiated (ORR) fuel are planned.

ACKNOJLEDGEMENT

Acknowledgement is gratefully made of many process flowsheet sug-
gestions by C. E. Guthrie and J. W. Ullmann of the Chemical Technology
Division. Acknowledgement is also made for the helpful work of people
in the Analytical Division under the supervision of W. R. Laing,

L. J. Brady, and C. Feldman.

 
-3-
FLOWSHEET DISCUSSION

The three steps of the fused-salt volatility process, namely, dis-
solution by hydrofluorination, UF6 volatilization by fluorination, and,
finally, UF6 purification by use of sorption-desorption on NaF would re-
main essentially the same for Al-U alloy fuel process as for Zr-U alloy

processing (Fig. 1).

To attain dissolution of Al-U fuel by hydrofluorination, use of
the KF-ZrFu-AlF3
capacity, salt cost, and dissolution rates. Establishment of the

salt system was satisfactory in terms of process

phase-equilibrium diagram for this system (Fig. 2) was required before

proceeding with the development work directed towards flowsheet studies.

Two preferred process flowsheets are summarized in Table 1, although
other variants could be proposed. The recycle process appears optimum
from the viewpoint of simplicity and ease in obtaining the correct batch
volumes relative to equipment size and weight of the fuel element. The
disadvantage of the step process, on the other hand, is that dif-
ferent salt volumes and amounts of fuel to be processed are involved
in each part. In the recycle dissolution flowsheet the solvent salt
composition ie cycled between 67-21-12 and 51.8-16.2-32 mole % KF-ZxF) -
AlF3 with inexpensive KF and 2KF'ZrFu salts used as makeup material,

The step process, 1in contrast, involves four separate salt compositions,

with some ZrFu being reguired in addition to KF and 2KF'ZrFu.

An important factor to be considered in choice of a flowsheet is
the cost of the salt. The commercial availability of ZKF-ZrF) at
$0.50 per pound or $0.67 per pound of contained ZrF) vas a factor in

choice of the KF-ZrFu-AlF system. The main advantage of the recycle

3

type of flowsheet over the step process is the fact that only this in-
expensive form of ZrFM is required in the former, vhereas, in the
latter, the use of some ZrF), per se at about $L.00 per pound is
necessary to increase the Zth content of the commercial salt from

33 mole % up to about 40%. Using the $0.50 per pound cost for 2KF-

Zth.and $0.37 per pound for KF, the salt cost for the recycle flow-

sheet is $0.35 ver gram of 235y processed.
METALLIC
FUEL

FEED SALT GASEQUS HF

 

DISSOLUTION IN
MOLTEN SALT BY
HYDROFLUORINATION
500-600°C

 

UNCLASSIFIED
ORNL DWG. 64-3747

 

 

 

 

 

 

U-BEARING
MOLTEN SALT

 

Fig. 1.

Schematic Flowsheet of Fused Salt Volatility Process.

 

 

 

 

 

 

GASEOUS F2
l UF,, excess Fy, fission-product fluorides (<1%)
UF, VOLATILIZATION NaF SORPTION
P BY FLUORINATION FOR UFg
500-600°C PURIFICATION
WASTE SALT UFe PRODUCT

(>99% fission products)
 

UNCLASSIFIED
ORNL -DWG 63-{685R

AlF3~4400

‘/\
KF-AIFy \/l ?1,0/7/7)-
P-575— 3

£-560—

 

3KF-AIFy —
996
£-850— 850
80
KF l \/
856 & &
N ~ O
¢ R

 

3"
-6~

A third flowsheet variant in addition to the two described consists
of adding only K¥ to allow the dissolution of more Al-U alloy. This
could be done with the waste salt from either of the two flowsheets
described. Theoretically, this adding of KF could be done indefinitely,
with the ZrFu content continually decreasing. The only requirement
would be to keep within the phase-diagram area defined by the two 600°C
liquidus lines of importance. It appears doubtful however that use of
this type of flowsheet would be justified in view of the operational
complexity.

Table 1. Process Flowsheet Alternatives for Al-U Fuel Alloy

 

Step Process: (1) Dissolution of fuel in 60-40 mole % KF-ZrF) at
600°C until the following composition is attained; 51-34-15 mole &

KF-Z1F) -AlF (neglecting UF), concentration); (2) addition of KF to

3
change composition to 63.5-25.2-11.3 mole %; (3) dissolution of
second batch of fuel until 51.5-20.5-28 mole % composition is reached;

(4) fluorination at 600°C to volatilize and recover UFg.

Recycle Process: (1) Dissolution of fuel in 67-21-12 mole % KF-
ZI‘FI+-A]_F3
16.2-32 mole %; (2) fluorination at 600°C to volatilize and recover
UF6. Partial reuse of this waste salt with added KF and 2KF-ZrFu

 

at 600°C until the following composition is reached: 51.8-

provides salt for a new dissolution cycle.

 

LABORATORY DEVELOPMENT WORK

Tests on a small laboratory scale were made with aluminum and
uranium-aluminum alloy to study dissolution rates, hydrogen evolution,
the volatilization of UF6 by fluorination, and other factors to be

considered in development of an aluminum fused salt volatility process

flowsheet. These tests are described below.

 
-T~

Dissolution Rate Studies

An extensive series of dissolution or hydrofluorination tests were
made at 600°C using as a basis the KF-ZrFu~A1F3 phase diagram (Table 2).
In some of the initial work made in an effort to develop the step proc-
ess, the starting salt contained no AlF3, with the ZrFu content varying
from 37 to 44 mole % (runs 1-12). Extension of this range up to a higher
Zerr content was possible with the initial use of some AlF3 in the salt
in order to stay below a liquidus temperature of 600°C. The dissolution
rates for aluminum in this series of runs was apparently quite dependent
on composition, with the results varying from 7 up to 50 mils/hr. A
plot of these data versus salt composition (considered as the binary
KF—Zerr system) showed that the maximum attack rates were obtained at
approximately 60 and 67 mole % KF, corresponding to the compounds
3KF-QZrFu and 2KF'Zth (Fig. 3). The two peaks exhibited in this plot
indicated that there was probably more than one type of reaction in-
volved, and that at compositions of 60 and 67 mole % KF a reaction

giving a fast dissolution rate predominated.

There appeared to be no outstanding or consistent correlation
of the dissolution rate with the buildup of the A1F3 concentration in

this group of tests, This was true particularly in run 6-10, where

the final AlF3 content varied from 5.5 to 14.5 mole %.

The second main group of dissolution tests were made 1n support
of the recycle-type flowsheet (runs 19-24), Starting with a fused
salt composed of 67-21-12 mole % KF-Zth—AlF3, the dissolution rate
increases from about 7 up to over 40 mils/hr as the AlF3 content

increased up to 35 mole % (Fig. &4).

Formation of Black Precipitate and Probable Chemical Reactions
in Dissolution Step
Large amounts of a black material were seen in many of the aluminum
dissolution tests. On continued hydrofluorination this material re-
dissolved, leaving a clear molten salt or white salt on solidiflcation.
Attempts to identify the material failed. Chemical analysis of the

black material was not conclusive because it was excessively diluted
 

Table 2. Summary of Aluminum Dissolution Tests

~3-5 g Al in 50-70 g salt in a l-in. dism, 8-in. height nickel reactor at 600°C.
HF sparge rate 100 ml/min (STP)

 

 

 

 

Amount of
Salt Composition Run Sample Dissolution Rate

Run (mole o KF—Zth-AlF3) Duration Dissolved (mils/hr)

Mo, Initial o Final (hr) (%) Gravimetric Micrometer
1 56.0-Ll,C-0 5L, 8-L3,0-2.2 1.0 13 8.6 8.5-9.5
2 58.0-42.0-0 56.8-11.1-2,1 1.0 12 7.9 8-9
3 58.0=42,0-0 56.,0-40,6-3.3 1.0 18 11.7 10-12
L 60.0-40,0-0 5L.6-36.L-8,9 1.0 7 L6, b L5-51
5 €0.0-L0.0-0 52.8-35.3-11.9 1.0 8 51.2 li6-50
6 £1,0-39.0-0 57.6-36.8-5.5 0.5 30 39.6 -
7 £1.0-39.0-0 5).,2-3L,7-11.1 1.0 66 2.3 -
8 61.0-39.0-0 54,0-3%,5-11.5 1.0 70 45.8 Lo-L7
9 61.0-39.0-0 52.6-33,6-13.7 1.5 88 38.0 ———-

10 61..0-39,0-0 52,1=23,3=14k.5 1.5 8l 3.7 h1-L43

1] £2.0-33.0-0 58.7-36.0-5.L 1.0 33 21.2 18-21

12 63.0~37.0-0 £1..8-26.3-2.0 1.0 13 8.8 8-10

13 62.4-33.6-4.0 60.9-32,8-6.3 1.0 1L 9.4 9-10

14 62.4-33.6-1,0 60.6-32.6-6.7 1.0 17 11.1 9-10

15 £2.7-32.3=5.0 59.0-30.4-10.6 1.0 37 21.7 18-22

16 62.3-30,7-7.0 £0.7-29.9-9.L 1.0 58 37.2 -

17 62.3-30.7-7.0 58.L-28,8-12.8 1,0 40 25.6 2].-2L

18 63.0-27.0-10.0 61.7-26.5-11.8 1.0 14 9.1 8-9

19 67.0-21.0-12.0 66.0-20,7-13.3 1.0 12 7.8 -

20 67.0-21,0-12.0 65.7-20.4-13,9 1.0 11 7.4 ——-

21 60.A-19.4-20.0 58,0-18.2-23.8 1.0 27 17.8 -

22 59.5-18,5-22.0 57.2=17.9-2L,9 1.0 36 2.8 _——

23 57.0-17.0-26.0 52,0-~15.8-32.2 1.0 55 36.2 ——

2l 54,5-16.5-29.0 49,.5-15.0-35.5 1.0 66 Lho.6 ——

25 54.5-0-45.5 51.8-0-48.2 1.0 31 20.3 -

-8_

 
UNCLASSIFIED

ORNL DWG. 64-3756

 

 

 

 

 

[ l
INITIAL AlIF3 CONTENT
_ INDICATED IN MOLE % WHENEVER PRESENT L
S
E
O
o 7.0 —
s o
<
<
[a"4
O -
Ly
< o
[a"
7.0
Z 50 @
O ° —
—
D
|
O
v
g : 10.0
@ — ]
.\_/
| I I
55 60 65 70

KF CONTENT (mole % in KF-ZrF4 binary system)

Fig. 3.
System at 600°C.,

Dependence of Aluminum Dissolution Rate on Fused Salt Composition Using the KF-Zerr-AJ_F

3
-10-

UNCL ASSIFIED
ORNL DWG. 64-3754

 

20 —

DISSOLUTION RATE, GRAVIMETRIC {mils/hr)

10 —

 

0 | | |

‘/

 

 

12 15 20 25
AlF3 CONCENTRATION (mole %)

Fig. L. Dependence of Dissolution Rate for Aluminum at 600°C on ALlF

Concentration when the Recycle Type of Salt was Used.

3
-11~-

by the salt matrix. Lack of an x-ray diffraction pattern indicated
that it was amorphous. Chemical tests showed that it was a strong
reducing agent, liberating hydrogen from water, and nitrogen oxides

from nitric acid.

Although observation of the black material was not always at-
tempted it was generally found in the first group of tests, namely
runs 1-18; it was definitely never observed in the tests made according
to the recycle flowsheet, namely, runs 19-25. The largest amount of
black material seemed to occur at those compositions where a fast

rate was obtained (Fig. 3).

Appearance of the black material led to the conclusion that the

dissolution step was not simply one of hydrofluorination, as follows,

2Al + 6HF —>» AlFy + 31, , (1)

but that a competitive ZrFu reductive effect was also present:

hal + 37xF), —> ll-AlF3 + 37T . (2)

Although the free energy change for the last reaction is not
favorable (AF® = +9 kilocalorie per mole of aluminum at 600°C), it is
possible that activities in the salt solution are sufficlently anomalous
to overcome this. The literature, moreover, mentions the possibility

of obtaining amorphous zirconium and also an Al-Zr alloy in this manneru.

Hydrogen Evolution

Measurements of hydrogen evolution in many dissolution tests were
made in determining how the reaction proceeded and vhether it would be
of use as a means of monitoring fuel dissolution on a large scale. In
several of the first group of runs, hydrogen evolution continued after
removal of the fuel sample to indicate that something in the salt con-
tinued to liberate hydrogen gas (Fig. 5). In all cases, however, the
postdigestion hydrofluorination period eventually resulted in almost
total recovery of the theoretical quantity of hydrogen gas. At the
time of sample removal, however, the hydrogen gas was usually about

20 to 30% when the 60-40 mole % KF-ZrF) salt was used

 
UNCL ASSIFIED
ORNL DWG. 64-3753

 

 

 

 

  

 

 

3 100 — —
2 Run 6
o
o
=
S
&
O
[UN)
>
—
9 S
u:‘ 50 — <>t { Time -
L g Initial Salt Sample
L Run Composition Removed
; No.  (mole % KF-ZrF4-AlF3) (min)
l 6 61,0-39.0-0 30
7/ 61.0-39.0-0 60
? 61.0-392.0-0 90
16 62.3 - 30.7 - 7.0 &0
0 | | 1 |
0 50 100 150 200

TIME (min)

Fig. 5. Hydrogen Evoluticn in Aluminum Dissolution Tests where HF Sparging was Continued After

Sample Removal.

-a‘[.—
-13-

The plot of hydrogen gas recovery (at sample removal time) and salt
composition showed in an inverse manner the same correlation as in the
rate-vs-composition plot (Figs. 3 and 6). The fact that the highest dis-
solution rates were obtained with the lowest hydrogen recoveries was
indicative that the ZrF) reaction (reaction 2) rather than the reaction
caused by ordinary hydrofluorination was the effective mechanism in

achieving a fast attack rate.

No delay in hydrogen gas evolution was observed in tests of the
recycle flowsheet (consistent with the absence of black material)}. A
hydrogen gas evolution plot of a complete process test using the recycle
salt compositions showed agaln that complete hydrogen gas recovery was

obtained (Fig. 7).

Material Balance Test

A complete inventory of aluminum and HF useage, together with
hydrogen gas, provided the basis in dissolution tests with 61-39 mole
% KF-ZrFu salt for conclusions regarding the magnitude of effects other
than that of the hydrofluoriration reaction (Table 3). This test in-
dicated that more than half of the aluminum was dissolved by ZrFu reaction.
Of the theoretical amount of hydrogen gas generated (from HF useage) U43.29
failed to evolve. This is suspected to have been due to the formation of
Zer, vhere x was calculated to be 1l.34, a value guite consistent with
the known Zr-H2 isotherm. Although much amorphous zirconium was produced,

its concentration in the salt was only 4.8%.

In two other material balance tests in salt representing the recycle
flowsheet there was little evidence of the ZrFu reduction effect. In
these tests (runs 20-21, Table 2), the hydrogen recovery was about 90

and 92%, respectively.

Aluminum Melt-Down in Dissolution

Tests with intentional overheating during the fused salt dissolution

step showed that no catastrophic effects occurred, although complete and

total reaction of the aluminum with ZrFu present in the salt could be

 

 
UNCLASSIFIED
ORNL DWG. 64-3757

 

 

 

 

 

 

3
= 50 — —
@
5
P
o
o 40 — ® —
0o
Ll
>
O
2 30— ® ]
T °®
7.0
20 |— @ —
°
10 — INITIAL AIF3 CONTENT INDICATED IN |
MOLE % WHENEVER PRESENT
0 I l |
55 60 65 70

Fig. 6. Extent of Hydrogen Evolution in One-Hour Partial Aluminum Dissolution Tests at 600°C

Using the KF-ZrFu-AlF

3

Salt System.

KF CONTENT (mole % in KF-2rFy binary system)

—-]-[T_
Hy EVOLVED (% of theoretical)

 

 

 

 

UNCL ASSIFIED

ORNL DWG. 64-3755
100 —
80 — —
60 | — _
40 }— —_
20 |— —

l | | | | I
0 30 60 90 120 150 180 210
TIME (min)

Fig. 7. Hydrogen Evolution

in Dissolution Step of Recycle Process (Run

G, Table 4).

_gT_
=16~

 

Table 3. Material Balance in an Aluminum Dissolution Run
Showing Reduction of ZrH)
+

 

67.9 g of 61-39 mole % KF-ZrF) in a l-in.-diam nickel reactor sparged
vwith HF at rate of 100 ml/min (sTP) for 1 hr; no digestion period
after cdissolution

 

 

Grams Moles
Aluminum in test 3.254 --
Aluminum actually dissolved ir ) hr 2.286 0.0848
Hydrogen evolved -- 0.0315
HF used from weighed source T.739 --
HF recovered in cold trap 5.516 --
HF used in reaction 2.223 0.111

Results Calculated from Above Data

 

Fraction of aluminum reacted through reduction of ZrF) - 56.49
Fraction of theoretical hydrogen evolved - 56.8

Moles of reduced Zr produced - 0.0358

Moles of H2 produced but not evolved - 0.02L40

Zirconium hydride average composition = ZrH X = 1.3k

Concentration of amorphous Zr in salt - 4.8%

 

i expected. However, the Al-ZrFu reaction does not proceed below the

1 melting point of aluminum with He sparging of the salt. Sparging with
hydrogen fluoride was required to initiate and sustain the ZrFu reaction,
possibly indicating that a protective layer of zirconium or Zr-Al alloy
forms on the solid surface and inhibits further reaction with the salt.
In both of the tests, 3-g aluminum specimens in 63-37 mole % KF-ZrF) salt
were raised slowly to the melting point of aluminum, 658°C. At this point,
when either He or HF sparging was being used, the temperature rose rapidly
(about 35 to 40°C) in less than 10 sec, then decreased. Examination of

the specimens showed complete conversion to black nonmetallic material.
p

 
-17-

PROCESS FLUCRINATION TESTS

A series of fluorination tests with both the step and recycle process
flowsheets indicated that there was no difficulty in achieving almost
total UF, volatilization (Table 4). Some of the tests were of the com-
plete process using 3.6% U-Al alloy fuel; in others, the salt was "spiked"

with UF,_L to simulate uranium concentrations expected in the process.

Table 4. Summary of Fluorination Results in Process Testing

50-T70 g salt in a l-in.-diam nickel reactor; fluorine sparging rate,
approximately 100 ml/min (STP)

 

 

Fluorination
Temp. mime, U Couc. (poa)
Run Type Flowsheet (°c) (hr) Init, Final
A Step process - lst pha.seb 575 0.5 1530 1.2
~— 1.0 ——— 0.8
Step process - lst phase™ 600 1.0 1570 7
C Step process - lst pha.seb 625 0.5 1480 15
——— 1.0 - l.2
D Complete step processb 575 0.5 3200 1.2
o 1.0 - 1.h
E Complete step process™ 600 1.0 5300 30
Complete step process? 625 0.5 3500 2.0
- l.o - - 008
G Recycle processa 600 1.0 3900 .
H Recycle process: 600 1.0 2300 20

 

®Salt prepared by dissolution of 3.6 wt % U-Al alloy.
bSalt prepared by addition of UFu.

CONCLUSIONS AND DISCUSSION

The processing of aluminum-base fuel by use of the KF-Zth-AlF3 salt
system in a modified fused-salt fluoride-volatility process has been
demonstrated to be feasible from the viewpoint of dissolution rates,

AlF3 solubility limits, and completeness of uranium volatilization. Two

process flowsheets, the step and recycle processes, have been described

and tested. Other flowsheet variations appear possible.

 
-18-

The presence of side reactions in the dissolution step merits dis-
cussion from the viewpoint of process operability and possible hazards
or difficulties. This is particularly the case in the step process. 1In
this process (but not the recycle process) the delay in hydrogen evolu-
tion due to the reduction of ZrFu makes measurement of this gas not a
true measure of dissolution rate; however, it still provides a means of
determining when complete dissolution has been achieved, that is, when
the reduced zirconium metal as well as all of the aluminum-uranium fuel
has been dissolved. A postdissolution or digestion period was required
in laboratory hydrofluorinations for complete dissolution of all metal
and complete hydrogen recovery. However, in the case of engineering-scale
tests, where greater flow rates are practicable, this postdissolution

period may not be as noticeable or as important.

The presence of reduced zirconium metal as an amorphous sludge pos-
sibly presents some hazard in engineering operations. In the test pre-
sented in Table 3, it was calculated that the sludge had a concentration
of 4.8% in the salt. This does not appear high, but practical experience
with other sludges in fused salts indicates that this 4.8% quantity of
material might lead to difficulty if not kept in the dispersed state.

The heat release involved when the temperature exceeds the melting
point of aluminum and complete reaction occurs was shown not to be
important. This is due to the fact that the net enthalpy change for
the ZrFu reduction reaction is only the difference between the large
enthalpy changes involved in the hydrofluar ination of aluminum and
zirconium. The principal hazard in allowing overheating and melting
of aluminum fuel is the high corrosivity of molten aluminum on nickel

or high-nickel alloys.

 
-19-

REFERENCES

F. L. Culler, Chemical Technology Division Annual Progress Report,
May 31, 1963, ORNL-3L52.

 

R. C. Vogel et al., "Fluoride Volatility Processes for the Recovery
of Fission Material from Irradiated Reactor Fuels," Proceedings of
Third International Conference on the Peaceful Uses of Atomic Energy,
Geneva, August 31-September 9, 196L, (to be published).

 

 

B. J. Sturm, R. E. Thoma, and E. H. Guinn, Molten-Salt Solvents for
Fluoride-Volatility Processing of Aluminum-Matrix Nuclear Fuel Elements,
ORNL-3594 (in preparation).

 

 

W. B. Blumenthal, The Chemical Behavior of Zirconium, Chapters 1-3,
J. Van Nostrand Co., New York, 1958.

 
107

ORNL-3596
UC-4 — Chemistry
TID-4500 (30th ed.)

INTERNAL DISTRIBUTION

1. Bioclogy Library 55. Calvin Lamb
2-4. Central Research Library 56. C. E. Larson
5. Reactor Division Library 57. H. F. McDuffie
6-7. ORNL — Y-12 Technical Library 58. R. P. Milford
Document Reference Section 59. M. J. Skinner
8-42. Laboratory Records Department 60. S. H. Smiley (K-25)
43. Laboratory Reccrds, ORNL R.C. 61l. B. J. Sturm
44. M. R. Bennett 62. R. E. Thoma
45. R. E. Blanco 63. J. A. Swartout
46. J. C. Bresee 64. J. W. Ullmann
47. J. E. Bigelow 65. A. M. Weinberg
48. G. E. Boyd 66. M. E. Whatley
49. W. H. Carr 67. Lloyd Youngblood
50. W. L. Carter 68. P. H. Emmett (consultant)
51. G. I. Cathers 69. J. J. Katz (consultant)
52. F. L. Culler 70. T. H. Pigford (consultant)
53. C. E. Guthrie 71. C. E. Winters (consultant)
54. R. W. Horton

EXTERNAT, DISTRIBUTION

72. E. L. Anderson, Atomic Energy Commission, Washington
73. 0. T. Roth, Atomic Energy Commission, Washington

74. Harry Schneider, Atomic Energy Commission, Washington
75. R. C. Vogel, Argonne National Labcratory

76. A. Jonke, Argonne National Laboratory

77. N. Levitz, Argonne National Laboratory

78. J. Fischer, Argonne National Laboratory

79. L. P. Hatch, Brookhaven National Laboratory
0. V. J. Reilly, Brookhaven National Laboratory

81. Research and Development Divisicn, AEC, ORO
82-673. Given distribution as shown in TID-4500 (30th ed.) under
Chemistry category (75 copies — OTS)