Potentialities of low-temperature chlorination of aluminum and its alloys with uranium

Article abstract of DOI:10.1134/S1070427208110062

Potentialities of low-temperature (300-500°C) chlorination to remove the aluminum cladding of fuel elements and separate uranium and aluminum were studied.

Full text of DOI:10.1134/S1070427208110062

ISSN 1070-4272, Russian Journal of Applied Chemistry, 2008, Vol. 81, No. 11, pp. 1904–1908. © Pleiades Publishing, Ltd., 2008.
Original Russian Text © V.A. Lebedev, V.I. Sal’nikov, V.V. Tsvetov, A.V. Bychkov, Yu.P. Savochkin, M.V. Kormilitsyn, 2008, published in Zhurnal
Prikladnoi Khimii, 2008, Vol. 81, No. 11, pp. 1786–1790.
INORGANIC SYNTHESIS
AND INDUSTRIAL INORGANIC CHEMISTRY
Potentialities of Low-Temperature Chlorination of Aluminum
and Its Alloys with Uranium
V. A. Lebedev, V. I. Sal’nikov, V. V. Tsvetov, A. V. Bychkov,
Yu. P. Savochkin, and M. V. Kormilitsyn
Ural State Technical University, State Educational Institution for Higher Professional Education, Yekaterinburg, Russia
Research Institute of Nuclear Reactors, State Scientific Center, Federal State Unitary Enterprise,
Dimitrovgrad, Ul’yanovsk oblast, Russia
Received May 15, 2008
Abstract—Potentialities of low-temperature (300–500°C) chlorination to remove the aluminum cladding of
fuel elements and separate uranium and aluminum were studied.
DOI: 10.1134/S1070427208110062
Fuel elements with an aluminum cladding and
based on alloys with uranium and aluminum are
widely used in nuclear reactors of various types.
Water-based methods for processing of such fuel ele-
ments are multistage and produce a large amount of
radioactive waste [1, 2].
Al
2
O
3
+ 3Cl
2
= 2AlCl
3
+ 1.5O
2
,
(1)
calculations of changes in the Gibbs energy on the ba-
sis of the data from [3] for temperatures of 500, 1000,
and 1500 K give ΔG values of 156 808, 65883, and
r
–
1
–
19 694 J mol Al O , respectively.
2 3
It can be seen that reaction (1) becomes thermo-
dynamically possible only at temperatures close to
500 K. Therefore, effective chlorination of metallic
aluminum with gaseous chlorine is only possible upon
disintegration of the oxide film.
This study is concerned with the possibility of us-
ing low-temperature chlorination of aluminum (remov-
al of the cladding) and U–Al alloys (separation of ura-
nium and aluminum). As prerequisites served low sub-
limation points of aluminum trichloride (182.7°C) and
comparatively high boiling points of uranium tetra-
chloride (775°C) and especially uranium trichloride
1
The aim of the study was to determine the mini-
mum chlorination temperature of metallic aluminum,
chlorination rate, extent of chlorine utilization, and
completeness of uranium and aluminum separation in
chlorination.
(
1789°C).
The reaction of aluminum with chlorine 2Al +
Cl = 2AlCl is accompanied by a noticeable heat ef-
3
2
3
0
–1
fect (ΔH2 = –695.85 kJ mol ) and is characterized by
98
a substantial negative change in the Gibbs energy
EXPERIMENTAL
0
–1
(
ΔG2 = –637.65 kJ mol ) [3]. The thermodynamic
98
characteristics show that the chlorination of aluminum
must occur without any hindrance. However, alumi-
num is always covered, because of its high chemical
activity, with a dense film of an oxide whose crystal
lattice constants are very close to the lattice constants
of aluminum itself. This provides a rather high corro-
sion resistance of aluminum in a wide temperature
range. The reaction of chlorine with aluminum oxide
becomes noticeable only at sufficiently high tempera-
tures. For the reaction
The chlorination was carried out with aluminum of
chemically pure grade (≥99.9%) in the form of grains
~1 cm in diameter and 0.3–0.4 cm thick, or as shavings
or a foil, which had a substantially larger specific sur-
face area than grains. As an apparatus for chlorination
served a quartz or ceramic test tube tightly closed with
a cover through which chlorine was introduced and
volatile chlorination products, including unchanged
chlorine, were removed. The breakthrough of chlorine
was determined with an indicator, potassium iodide
1
904

POTENTIALITIES OF LOW-TEMPERATURE CHLORINATION OF ALUMINUM
T, °C
1905
solution, whose reaction with chlorine
2
KI + CI
2
= 2KCl + I
2
(2)
yields elemental iodine making the indicator paper
brown. A weighed portion of aluminum was 5–15 g,
and the charge bed height, 5–10 cm.
Chlorine was produced by electrolysis of lead chlo-
ride in a chlorine-producing unit. After being purified
to remove trace amounts of moisture (bubbling
through concentrated sulfuric acid) and solid dust sub-
limates (filtration through glass wool), chlorine was
delivered via a ceramic tube into the experimental cell.
The rate of chlorine supply was controlled by varying
the current strength through the chlorine unit. The
chlorination process was carried out with an excess of
chlorine, determined from its breakthrough with the
indicator, which enabled estimation of the chlorination
rate.
ΔT
Thermogram of aluminum chlorination.
relative to the amount consumed at the actual chlorina-
tion rate in all the experiments. The chlorination dura-
tion was 1 to 5 h. Data on the conditions and results of
chlorination without preliminary treatment of the grain
surface are summarized in Table 1.
The minimum temperature at which aluminum
chlorination begins was determined using a differential
thermocouple that measured the temperature difference
between the exterior and interior of the chlorinator.
Because the heating elements of the furnace were out-
side the chlorinator, the temperature of a heated chlori-
nator remained 5–10°C lower. The reaction of alumi-
num with chlorine, accompanied by a strong heat ef-
fect, results in that the negative difference between the
internal and external temperatures first decreases in
magnitude and then, after the chlorination is intensified
and a considerable amount of heat is released, becomes
positive. This was methodologically achieved by grad-
ual heating of the furnace at a constant chlorine deliv-
ery into the chlorinator. The temperature difference ΔT
was recorded with a RECORDER self-recorder. The
thermogram obtained is plotted in the ΔT–T coordi-
nates in the figure. As can be seen in the thermogram,
a noticeable heat release in aluminum chlorination is
observed at a temperature close to 250°C, and just this
temperature is the minimum chlorination temperature.
As a rule, chlorination at 200–300°C fails to yield
positive results: aluminum is either not chlorinated
(
Table 1, run nos. 1 and 2) or the chlorination rate is
–
2
–1
low and does not exceed 2 mg cm h (run nos. 10,
1). The best result is obtained in chlorination of
1
freshly prepared shavings. An increase in temperature
to 350–500°C raises the chlorination rate of aluminum
–
2
–1
to the maximum value of 9–12 mg cm h . However,
no clearly pronounced temperature dependence of the
chlorination rate was observed. The maximum value of
–
2
–1
1
3
2 mg cm h (Table 1, run no. 4) was observed at
50°C, whereas at 450°C (run no. 8), the chlorination
–
2
–1
rate was lower, 8.2 mg cm h . The chlorination rate
did not exhibit any pronounced dependence on the
process duration, either. The best result was obtained
at 350°C (Table 1, run no. 4) in a 1-h-long experiment
–
2
–1
(
12 mg cm h ), and at a duration of 4–5 h (run
nos. 6, 9), the chlorination rate was two times lower.
The chlorination was mostly performed using
The chlorination rate is presumably low because the
oxide film is not effectively removed from the alumi-
num surface at the process parameters (temperature,
duration) used. At the same time, an interesting type of
behavior was observed: in all the runs with the highest
chlorination rates, some grains (1 to 2–3 grains) lost 50
to 90% of mass, whereas parameters of all the others
(15–20 grains) remained almost unchanged. Thus, the
average chlorination rate is determined by chlorination
of only a part of aluminum grains (~10%), with the
granulated aluminum with an average grain mass of
2
0
.4–0.6 g and surface area of about 2.5–3 cm . To have
a larger specific surface area, aluminum foil (Table 1,
run no. 10) or freshly prepared shavings (run no. 11)
were used in separate experiments.
The chlorination temperature was varied from 200
to 500°C. The flow rate of chlorine delivery (in terms
of the current through the chlorine unit) was within the
range 2–3 A, which provided an excess of chlorine
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 81 No. 11 2008

1
906
LEBEDEV et al.
Table 1. Chlorination of aluminum without pretreatment of its surface
Run
no.
Cl feed rate,
Initial Al
Al mass loss, g
Al surface Cl utilization Chlorination rate,
2
T, °C
–1
Time, h
2
–2 –1
A g
mass, g experiment calculation area, cm efficiency, %
mg cm h
1
2
3
4
5
6
7
8
9
200
300
350
350
350
350
400
450
500
300
300
2.5
2.5
2.5
3.0
3.0
2.0
2.0
3.0
2.0
3.0
2.0
1
13.040
17.924
13.040
11.472
10.872
11.865
10.977
10.764
11.849
2.306
–
0.838
1.678
1.678
1.007
1.007
3.389
0.678
1.007
2.916
1.007
2.034
60
70
–
–
–
2
–
–
2
1.112
0.600
0.333
1.420
0.213
0.409
1.280
0.062
1.210
60
66
60
32
43
31
41
44
6
9.2
1
50
12
1
50
6.7
5.7
4.2
8.2
5.9
0.1
2
5
50
1
50
1
50
4.3
1
50
1
1
0
1
600
200
3
7.515
60
a
Table 2. Chlorination of aluminum with pretreatment of its surface
Run
no.
Cl feed rate,
Initial Al
Al mass loss, g
Al surface Cl utilization Chlorination rate,
2
T, °C
–1
Time, h
2
–2 –1
A g
mass, g experiment calculation area, cm efficiency, %
mg cm h
1
2
3
4
5
6
7
8
9
300
350
300
300
300
300
370
400
300
2.0
2.0
3.0
3.0
3.0
2.0
2.0
2.0
1.3
1
12.524
11.215
12.390
12.116
11.585
8.339
–
0.671
2.74
60
40
55
55
55
40
40
35
60
–
–
7.8
6.2
9.4
12.2
1.3
7
4.08
0.75
1
1.254
0.256
0.519
0.674
0.055
0.364
0.570
0.352
45.6
36
53
67
8
0.755
1.007
1.007
0.671
0.671
0.873
1.678
1
1
1
8.284
54
66
21
1.3
3.8
7.920
12.5
1.5
13.422
a
4 3 2 2
Treating solution: run no. 1, NaOH; run no. 2, ZnSO + NaOH; run nos. 3–5, Hg(NO ) ; run nos. 6–8, CuCl ; run no. 9, HCl.
chlorination rate of these grains exceeding the average
value by at least an order of magnitude, i.e., it is not
The following solutions were tested in decomposi-
tion of the protective film: 2 N NaOH solution, alka-
line solution of zinc sulfate, and dilute (0.05 N) solu-
tions of mercury nitrate and copper(II) chloride. The
results obtained are listed in Table 2.
–
2
–1
lower than 100 mg cm g . Interestingly, no chlorine
breakthrough through the cell was observed in these
experiments, i.e., as the intense chlorination of sepa-
rate grains begins, the whole amount of passed chlo-
rine is consumed by these grains to give aluminum
chloride, which suggests that an even higher chlorina-
tion rate is possible.
The stripping of the aluminum surface in alkaline
solutions is based on the reaction
Al
2
O
3
+ 2NaOH = 2NaAlO
2
+ H
2
O,
(3)
The observed behavior indicates that the effective
chlorination of metallic aluminum is limited by the
stage of disintegration of the protective oxide film.
which gives sodium aluminate well soluble in water.
Prior to chlorination, aluminum grains were placed for
several minutes in a 2 N NaOH solution and then
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 81 No. 11 2008

POTENTIALITIES OF LOW-TEMPERATURE CHLORINATION OF ALUMINUM
1907
charged into the chlorinator. The chlorination was per-
formed at 300°C. No changes in the mass of aluminum
were observed. Presumably, aluminum oxide is not
removed from the aluminum surface under the condi-
tions specified.
Table 3. Conditions and results of the low-temperature
350°C) chlorination of uranium–aluminum alloys
(
Passed quantity
of electricity, A h
Parameter
1.0
1.4
Mass of alloy, g:
initial
after chlorination:
before washing
The highest efficiency was observed in the case of
triple short-time (0.5–1 min) wetting of aluminum
grains with dilute solutions of mercury nitrate and cop-
per(II) chloride.
1.088
0.182
1.042
0.900
0.121
0.045
after washing
Content of uranium U in alloy, %:
initial
after washing
Mass loss, g:
U
Al
Quantity of electricity consumed for
chlorination, A h:
U
Al
The grains were treated with a 0.05 N Hg(NO3)2
solution three times for 0.5–1 min. Chlorination was
performed after each treatment. As can be seen in Ta-
ble 2, amalgamation produces a positive effect. At
39.5
34.0
31.8
31.8
0.124
0.053
3
00°C, the chlorination occurs at an acceptable rate,
0.064
0.072
with each subsequent treatment of the aluminum sur-
face leading to an increase in the chlorine utilization
efficiency and in the chlorination intensity. The chlori-
0.042
0.191
0.018
0.215
–2
–1
nation rate increased from 6.2 to 12.2 mg cm h , and
the chlorine utilization efficiency, from 36 to 67%.
Cl
2
utilization efficiency, %
23.3
23.3
Content in fumes:
The advisability of using mercury and its salts is
limited by their high toxicity. It was of interest to try
other activating agents. A triple treatment of aluminum
grains with a dilute copper chloride solution yielded a
positive result. Upon a triple treatment, the chlorina-
tion rate at 300°C and the chlorine utilization effi-
U, mg
U, %
Al, mg
Al, %
Mass of solid residue, g:
found from ΔP
found from difference
25.4
20
57.6
70
5.3
10
65.3
92
0.142
0.143
0.076
0.069
–
2
–1
ciency increased from 1.3 to 12.5 mg cm
from 8 to 66%, respectively.
h
and
The contents of uranium in the starting and final,
washed alloy and the differences of their masses were
used to find the loss of mass by uranium and alumi-
Presence of uranium in aluminum alloys must
markedly accelerate the chlorination process for two
reasons. First, the fact that uranium has several stable
valence states substantially accelerates delivery of the
chlorinating agent to the metal surface. Uranium tri-
chloride is stable in contact with aluminum, and ura-
nium tetrachloride and even pentachloride, in contact
with gaseous chlorine, and these uranium compounds
are effective transporters of chlorine to the surface of
aluminum and its alloys. Second, the protective prop-
erties of the surface oxide film are considerably im-
paired in the presence of uranium in aluminum.
num in the course of chlorination (Δm ); the quantities
i
of electricity required for this purposes are given by
the equation Q = Δm /q , where q and q are 0.3355
i
i
i
Al
U
–
1
–1
and 2.96 g A h , respectively; the chlorine utilization
efficiency η = (Q + Q )/Q, where Q is the total quan-
U
Al
tity of electricity passed.
An analysis of the sublimates demonstrated that
they mostly contain aluminum chloride (≥90%) and 10
to 30% uranium chlorides. The mass of the salt residue
after the chlorination, found from difference of the
masses of the alloy before and after its washing, almost
coincides with that calculated from the difference of
masses after the chlorination and sublimation of ura-
nium (in terms of its trichloride), which additionally
confirms that most of aluminum chloride is removed
into sublimates.
To verify these assumptions, experiments on low-
temperature chlorination of uranium–aluminum alloys
were carried out. The alloys were chlorinated at 350°C
in a small Alundum test tube. After chlorination for
approximately 1 h, the cooled solid residue was
weighed before and after the salts were removed by
washing. An analysis of the resulting solutions demon-
strated that they are nearly fully composed of uranium
The rate of low-temperature chlorination of ura-
–2 –1
chlorides (mostly UCl ). The conditions and results of
nium–aluminum alloys (200–300 mgcm h ) substan-
tially exceeds (by factors of tens and hundreds) the
3
chlorination are listed in Table 3.
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 81 No. 11 2008

1
908
LEBEDEV et al.
chlorination rate of pure aluminum. Despite that the
weighed portions of the alloys were small, the chlorine
utilization efficiency was considerable (>20%).
(3) The highest efficiency in disintegration of the
oxide film on aluminum upon its wetting before chlori-
nation was observed for an alkaline solution of zinc
sulfate, 0.05 N solution of mercury nitrate, and 0.05 N
solution of copper(II) chloride.
The considerable loss of uranium with the fumes is
presumably observed because a part of uranium is
chlorinated to its tetrachloride, which has a noticeable
vapor pressure. Therefore, by passing the vapor
through a reducing packing (e.g., aluminum shavings),
it is possible to deposit uranium trichloride reduced
from the tetrachloride on this packing, with loss of ura-
nium with the fumes precluded. It is recommended to
condense aluminum chloride vapor at 240–260°C on
lumps or ingots of sodium chloride to obtain molten
NaCl–AlCl mixtures containing up to 60 mol % AlCl
(4) In chlorination of solid uranium–aluminum al-
loys (39.5% uranium) at 350°C, the presence of ura-
nium in an alloy provides high chlorination rates (up to
–
2
–1
300 mg cm h ) and high chlorine utilization efficien-
cies. Almost the whole amount of chlorinated alumi-
num and part of uranium are removed into sublimates.
To preclude the loss of uranium, the gas flow should
be passed through aluminum shavings reducing the
volatile uranium tetrachloride to the considerably less
volatile trichloride.
3
3
[
4].
CONCLUSIONS
1) The onset temperature of chlorination of pure
REFERENCES
1
2
. Matrin, F.S. and Miles, G.L., Chemical Processing of
Nuclear Fuels, London: Butterworth, 1958.
. Stoller, S.M. and Richards, R.B., Reactor Handbook,
vol. 2: Fuel Reproceissing, New York: Interscience,
(
aluminum by gaseous chlorine is close to 250°C.
(
2) The existence of a dense oxide film on the metal
–2
–1
surface results in low chlorination rates (<10 mg cm h )
and chlorine utilization efficiencies (<20%). Upon dis-
integration of the oxide film at least on a part of the
sample surface, the chlorination rate increases by more
than an order of magnitude, with the chlorine utiliza-
tion efficiency reaching 60%.
1
961, 2nd ed.
3
. Termodinamicheskie svoistva neorganicheskikh ve-
shchestv: Spravochnik (Thermodynamic Properties of
Inorganic Substances: Reference Book), Zefirov, A.P.,
Ed., Moscow: Atomizdat, 1965.
4. USSR Inventor’s Certificate no. 683 188.
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 81 No. 11 2008