Radiation resistance and chemical stability of yttrium aluminum garnet

Article abstract of DOI:10.1134/S1066362211020123

The radiation resistance and chemical stability of yttrium aluminum garnet doped with short-lived ²⁴⁴Cm for accelerated accumulation of radiation damages was studied. Two garnet samples of the stoichiometry Y _2.8853Cm_0.1024

Full text of DOI:10.1134/S1066362211020123

ISSN 1066-3622, Radiochemistry, 2011, Vol. 53, No. 2, pp. 186–190. © Pleiades Publishing, Inc., 2011.
Original Russian Text © S.V. Tomilin, A.A. Lizin, A.N. Lukinykh, T.S. Livshits, 2011, published in Radiokhimiya, 2011, Vol. 53, No. 2, pp. 162–165.
Radiation Resistance and Chemical Stability
of Yttrium Aluminum Garnet
a
a
a
b
S. V. Tomilin , A. A. Lizin* , A. N. Lukinykh* , and T. S. Livshits
a
Research Institute of Nuclear Reactors, State Scientific Center, Public Joint-Stock Company,
Dimitrovgrad-10, Ulyanovsk oblast, Russia; * e-mail: lukinykh@niiar.ru
b
Institute of Geology of Ore Deposits, Petrography, Mineralogy, and Geochemistry,
Russian Academy of Sciences, Moscow, Russia
Received February 26, 2010
Abstract—The radiation resistance and chemical stability of yttrium aluminum garnet doped with short-lived
2
44
Cm for accelerated accumulation of radiation damages was studied. Two garnet samples of the stoichiometry
244
Y
5
2.8853Cm0.1024Pu0.0092Al O12, containing ~4 wt % Cm with 70% content of Cm, were synthesized. The ce-
ramics consist of the target phase of the garnet structure and a small amount of corundum. The chemical
stability of the freshly prepared sample in water was studied (МCС-1 test, 90°С). The Cm leach rate is
–
2
–2
–1
–3
–2
–1
~
10 g m day , and that of Al and Y, ~10 g m day . The garnet became fully X-ray amorphous owing to
18
self-irradiation in 530 days at the accumulated dose of 4.0 × 10 α-disintegrations per gram, or 0.3 displace-
ment per atom. After the destruction of the garnet structure, the Cs leach rate from the matrix 14 days after the
start of the experiment increased by a factor of 10, and that of Y, by a factor of 60 relative to the freshly pre-
pared ceramics.
Keywords: yttrium aluminum garnet, chemical stability, radiation resistance
DOI: 10.1134/S1066362211020123
Artificial compounds of the garnet structural type
actinides, primarily Am [3]. For substantiated choice
of the host material, it is necessary to study compre-
hensively its physicochemical and isolation properties,
such as thermal conductivity, heat and radiation resis-
tance, chemical stability, and isomorphic capacity.
VIII
VI
IV
(
A B X O , space group Ia3d, Z = 8) are con-
3 2 3 12
sidered as potential host materials for radioactive
waste immobilization [1, 2]. The presence of three
cationic positions of different sizes in the lattice of
such phases creates prerequisites for wide isomor-
phism, which is important for solidification of multi-
In this study we examined the radiation resistance
and chemical stability of yttrium aluminum garnet
VIII
component wastes. The A position can be occupied
244
doped with Cm (Т1/2 = 18.1 years). The isomorphic
by bi- (Ca, Mn, Mg, Fe, Co, Cd), tri- (Y, REE), and
capacity of yttrium aluminum garnet for such a large
tetravalent (Ce, An) cations with the ionic radii r =
3+ VIII
cation as Cm ( r = 1.08 Å) is unknown. To evaluate
VI
0
.9–1.1 Å. The octahedral position B is usually occu-
the possibility of incorporation and determine the solu-
bility limits of Cm in the yttrium aluminum garnet
structure, we performed preliminary experiments with
inactive materials using Sm and Nd as crystal-
chemical analogs of Cm. On the basis of data obtained
with these imitators, we determined the optimal formu-
lations for preparing the Cm-containing aluminum gar-
net. We monitored variation of the crystal structure of
pied by smaller tri- (Fe, Al, Ga, Gr, Mn, In, Sc) and
tetravalent (Zr, Ti, Sn) cations with the ionic radii r =
.5–0.8 Å. The tetrahedral position is suitable for ac-
commodation of tri- (Al, Ga, Fe), tetra- (Ge, Si), or
0
pentavalent (V, As) cations (r = 0.3–0.5 Å).
Yttrium aluminum garnet Y Al (AlO ) occupies an
3
2
4 3
important place in the series of synthetic garnets. It
exhibits valuable optical properties and find use in la-
ser engineering and electronics; also, it is considered
as a promising host material for transmutation of minor
244
garnet, caused by the Cm decay, up to complete
amorphization. We compared the chemical stability of
the sample before and after the garnet amorphization.
1
86

RADIATION RESISTANCE AND CHEMICAL STABILITY
187
244
EXPERIMENTAL
Al(NO
3
)
3
·6H
2
O,
Y O ,
CmO2,
sample
2
3
reagent
reagent
Model samples of garnet ceramics containing Sm
and Nd as Cm imitators were prepared by cold press-
ing and sintering (CPS). The gross composition of the
two synthesized samples with Sm was described by
the formulas Y2.8218Sm0.1782Al O and YSm Al O ,
Heat treatment,
50°С, 3 h
Dissolution
in 3 M HNO
Dissolution
in 3 M HNO
3
3
3
5
12
2
5
12
and that of the Nd-containing sample, by the formula
Y NdAl O . A study of these samples by X-ray dif-
fraction (DRON-7 diffractometer) and scanning elec-
tron microscopy with energy-dispersive X-ray analyzer
Drying, 150°С, 3 h;
heat treatment, 400°С, 2 h
Nitric acid
2
5
12
244
3+
Cm solution
(
FE-SEM complex, Philips XL30; JSM-5610LV +
2 3 2
Impregnation of Y O and Al O
244 3+
3
with nitric acid Cm solution
JED-2300) showed that the solubility of Sm and Nd in
the garnet reaches 15 wt %.
The Cm-containing garnet was prepared by the
scheme shown in Fig. 1. The oxide Y O was dis-
Drying and precipitation of
hydroxides, 100–200°С
Aqueous 25%
ammonia solution
2
3
solved in HNO (3 M), aluminum nitrate was calcined
3
at 350°С with the subsequent drying and denitration of
the precipitate at 400°С, and the Cm sample in the
2
44
Denitration, 500°С, 1 h
Calcination, 800°С, 2 h
form of dioxide was also dissolved in HNO (3 M).
3
The isotope composition of the Cm sample was as fol-
2
44
240
245
246
lows, %: Cm 67.28, Pu 8.23, Cm 15.18, Cm
247
248
8
.65, Cm 0.48, and Cm 0.18%. The daughter iso-
tope, Pu, accumulated in the course of 3-year stor-
age of the sample. The Cm concentration in the so-
Pressing in pellets,
240
P ≈ 200 MPa
2
44
lution before the synthesis was refined by α-ray spec-
2
44
–1
trometry and amounted to 1.39 (mg Cm) ml .
Annealing (3 4-h steps):
Pellet crushing
and grinding (2 steps)
1
450, 1475, 1450°С
The specially prepared samples of yttrium and alu-
minum oxides, calcined at low temperatures, were im-
pregnated in portions with the nitric acid solution of
curium and dried at 100–120°С. After heating the re-
sulting slurry to these temperatures, the yttrium and
aluminum oxides fully dissolved in the nitric acid solu-
tion of curium. On heating above 130°С, the thick
transparent mass formed underwent vigorous foaming
because of fast and simultaneously occurring processes
of dehydration and melting of the nitrates and start of
their decomposition with the evolution of nitrogen ox-
ides. Further heating was stopped, the thick mass was
washed with nitric acid from the test tube walls, and a
large excess of 25% aqueous ammonia was added for
the solution neutralization and subsequent precipitation
of the hydroxides. The resulting precipitate of the hy-
droxides was again dried at 120°С to the anhydrous
state with slow further heating (5 h) to 200°С. The fi-
nal steps of the process involved denitration (500°С,
Fig. 1. Scheme of preparation of garnet-based Cm-con-
taining ceramic.
were then subjected to threefold annealing with inter-
mediate crushing, grinding, and pelletizing. The pellets
were heat-treated in succession at 1450, 1475, and
1
450°C (4 h at each temperature).
The chemical composition of the resulting garnet
(
based on the empirical formula Y2.8853Cm ·
Pu0.0092
0.1024
2
44
Al O ) was as follows, wt %: Y 42.01, Cm 3.00,
5
12
2
45–248
240
Cm 1.10, Pu 0.36, Al 22.09, and O 31.44. The
geometric parameters of ceramic pellet nos. 1 and 2
after the third, final annealing were as follows: diame-
ter 5.61 and 5.63 mm, height 1.70 and 1.97 mm, vol-
3
ume 0.042 and 0.049 cm , surface area 0.794 and
2
0.846 cm , respectively. Pellet no. 1 was subjected to
1
h) and calcination (800°С, 2 h). The powders thus
chemical stability tests, and pellet no. 2 was used for
X-ray diffraction monitoring of the destruction of the
prepared were then thoroughly dispersed.
2
44
garnet structure due to Cm decay.
From the resulting calcinate, by pressing at
2
00 MPa and 25°С we prepared two pellets which
From pellet no. 2, we filed off a small amount of
RADIOCHEMISTRY Vol. 53 No. 2 2011

1
88
TOMILIN et al.
prepared and amorphized owing to self-irradiation)
was evaluated by the procedure similar to MCC-1 test
90°С). The procedure is described in our papers, e.g.,
in [4], but in our case the ratio of the sample surface
(
–
1
area to the solution volume was 3.1 m (according to
–
1
MCC-1, the optimal ratio is 10 m ). The solution was
taken in a larger volume than prescribed by the stan-
dard because of small size of the samples and necessity
for taking leachate samples for the analysis. The initial
double-distilled water was analyzed for the Al content.
The Cm and Pu concentrations in the solutions after
leaching were determined by α-ray spectrometry with
Fig. 2. Diffraction pattern of a freshly prepared pellet of
yttrium aluminum garnet with Cm. (D) Diamond, (T) Tef-
lon, and (C) corundum; the other peaks refer to garnet.
1
0–12 and 20% uncertainty, respectively; the Al and Y
concentrations were determined by emussion spectrum
analysis with 20–25% uncertainty. The pH values was
estimated by coloration of universal indicator paper.
RESULTS AND DISCUSSION
The diffraction pattern of the freshly prepared ce-
ramics in the angle range 2θ = 20°–120° exhibits well-
defined reflections of the garnet phase with the unit
cell parameter (UCP) of 12.0140(7) Å (Fig. 2). Very
weak corundum reflections are also present. The garnet
α-doublets at large and medium angles are well re-
solved and have small width, suggesting good crystal-
linity of the garnet structure. Data of the photographic
method fully agree with the results of sample examina-
tion with a diffractometer. The chemical composition
of the garnet in the ceramic was not determined
because of high activity of the sample. The results of
X-ray diffraction studies suggest that the garnet
composition is close to that preset by the formula
Y2.8853Cm0.1024Pu0.0092Al O .
5
12
Structural changes in the ceramic were monitored
for approximately 18 months. Diffractometric analysis
showed a typical pattern of the deterioration of the
crystal structure with accumulation of radiation defects
(Fig. 3). In the diffraction pattern recorded ~200 days
after the sample synthesis, in the angle range 2θ = 20°–
Fig. 3. Variation of the diffraction pattern of the ceramic
based on yttrium aluminum garnet with accumulation of
1
8
the damaging dose (in units of 10 α-disintegrations per
gram): (1) 0.05, (2) 0.8, (3) 1.58, (4) 2.46, and (5) 4.0.
(G) Garnet, (T) Teflon, and (D) diamond.
1
20° only approximately 15 garnet reflections remain
powder for a photographic study in a sealed quartz
capillary (RKU-114M camera). After that, the sample
was placed in a special aluminum cell for studies with
an X-ray diffractometer. The pellet was isolated from
the environment with a thin Teflon film (d ≈ 10 μm).
A diamond reference was used. Photographic study
was preformed monthly, and diffractometric study,
once in 3 months.
of approximately 45 reflections observed immediately
after the synthesis. The intensity of the major peak
with hkl = 420 decreased in this period by a factor of 7.
With the dose accumulation, weak corundum reflec-
tions disappear also, which is apparently associated
either with overall deterioration of X-ray diffraction
patterns due to garnet amorphization, or with internal
(
isomorphic incorporation of macroamounts of Cm in
The chemical stability of pellet no. 1 (both freshly
corundum) or external α-radiation. Because corundum
RADIOCHEMISTRY Vol. 53 No. 2 2011

RADIATION RESISTANCE AND CHEMICAL STABILITY
189
grains are apparently surrounded by the bulk of garnet,
the corundum structure can also be subjected to radia-
tion damage.
Variation of the UCP of garnet could be monitored
1
8
up to a dose of 2.0 × 10 α-disintegrations per gram.
In the process, the garnet cell size increased from the
initial value ot 12.0140(7) to 12.20(1) Å. Thus, linear
swelling was 0.186 Å or 1.55%, and the volume swell-
ing was 4.6% (Fig. 4). The garnet phase becomes
X-ray amorphous on accumulation of a dose of 4.0 ×
Fig. 4. Variation of the unit cell parameter of yttrium alu-
minum garnet with the dose accumulation.
18
1
0
α-disintegrations per gram (0.3 displacement per
atom).
can be assumed that the whole amount of Pu present in
the sample is incorporated in the garnet. High value of
the measured Pu leach rate is apparently caused by low
accuracy of its determination because of low contribu-
tion of this element to the total measured activity.
We studied the corrosion resistance of the Cm-
containing ceramic in water before and after amorphi-
zation of the garnet phase. The normalized leach rates
of elements from the yttrium aluminum garnet are
given in the table. The solutions after leaching were
transparent and contained no suspended material and
precipitates. The pH value of the leachates was in the
range 4–4.5, which is caused by water radiolysis under
intense α-irradiation. The results of α-ray spectromet-
ric measurements indicated that the solution activity
We compared the chemical stability of the alumi-
nate garnet and previously studied ferrite garnet of the
composition Ca Gd0.908Cm0.092Th ZrFe O [5]. The
1
.5
0.5
4
12
Cm leach rate from aluminate and ferrite of the garnet
structure, determined immediately after their synthesis,
–
2
–2
–1
is different and is of the order of 10 g m day for
2
44
–3
–2
–1
after leaching was due to
Cm (99.1–99.8%) and
aluminate and 10 g m day for ferrite garnet. At
the same time, the major structure-forming elements,
240
Pu (0.9–0.2%). Destruction of the garnet structure
244
–3
–2
–1
leads to an increase in the rate of transfer of Cm
and Y from the matrix to the solution. As compared to
the freshly prepared sample, the intensity of the Cm
leaching from the ceramic with the amorphized garnet
Al and Y (leach rates ~10 g m day ), are retained
–
2
–2
–1
better than Fe and Gd (~10 g m day ) in ferrite
garnet.
The destruction of the garnet phase structure affects
the chemical stability of the ceramic with aluminate
–1
–2
–1
increased by a factor of 10 (to 3.6 × 10 g m day ),
and that of Y, by a factor of 60 (to 8.1 ×
–2
–2
–1
1
0 g m day ) on the 14th day after the start of the
experiment. The Al leach rate from the garnet, how-
ever, did not change noticeably and was in the range
Leach rates of elements from the Cm-containing sample
before and after amorphization of the yttrium aluminum
garnet structure
–3
–2
–1
(
5–6) × 10 g m day .
It is known that, in synthesis of aluminate garnets, a
–2
–1
Leach rate, g m day
Ele- Leach time,
temperature as high as 1600°C is required for the
phase formation completion. We were able to prepare
immediately after
synthesis
ment
days
after amorphization
2
44
the garnet with Cm at 1450–1475°С with twofold
intermediate grinding and compaction. The resulting
ceramics appeared to be practically single-phase, with
UCP of the garnet [12.0140(7) Å] close to that of ref-
erence (“pure”) yttrium aluminum garnet (12.00–
–2
–1
3
7
5.6 × 10
1.6 × 10
1.5 × 10
3.6 × 10
244
–2
–1
Cm
3.6 × 10–
2
–1
1
1
1
1
4
3.5 × 10
–2
–3
3
7
4
1.9 × 10
7.6 × 10
1.1 × 10
5.8 × 10
–2
–2
Al
Y
1.2 × 10–
3
–3
1
2.02 Å).
6.9 × 10
–
3
3
3
–2
Because the system contains a small amount of Pu
0.36 wt %), a question of its incorporation in the gar-
3
7
4
1.2 × 10–
2.7 × 10
3.4 × 10
8.1 × 10
–2
(
1.6 × 10–
1.3 × 10
–2
net structure arises. The solubility of Pu in yttrium alu-
minum garnet was studied by Burakov et al. [9]. They
showed that the capacity of the garnet structure reaches
3
7
4
1.5
9.1 × 10
5.6 × 10
4.5
1.9
2
40
–1
–1
Pu
3+
4+
–1
9.6 × 10
0
.3 wt % for Pu and 5.3 wt % for Pu . Therefore, it
RADIOCHEMISTRY Vol. 53 No. 2 2011

1
90
TOMILIN et al.
garnet more strongly than that of the ferrite garnet ce-
ramic. For example, 14 days after the start of the ex-
periments, the Cm leach rate from amorphized ferrite
garnet increased by a factor of 4, and from yttrium alu-
minum garnet, by a factor of 10. In addition, after the
amorphization of the yttrium aluminum garnet, the Y
leach rates increased by a factor of 60, reaching 8.1 ×
the crystalline ceramic: 14 days after the start of the
leaching experiments, the Cm leach rate increased by a
factor of 10, and that of Y, by a factor of 60. Under
real conditions of underground HLW repositories with
weakly alkaline hydrogeochemical medium (рН 8–9),
the actinide leach rate from the matrix will be lower by
orders of magnitude.
–2
–2
–1
1
0 g m day . At the same time, because of water
radiolysis under intense α-irradiation, the measured pH
of the solutions after leaching was 4. Increased solu-
tion acidity is one of the factors decreasing the resis-
tance of the matrices to leaching. The suggested sites
of underground repositories will be characterized by
reductive weakly alkaline hydrogeochemical condi-
tions, favorable for the chemical stability of the matrix
in solutions.
ACKNOWLEDGMENTS
The study was financially supported by the Russian
Foundation for Basic Research (project nos. 08-05-
0024-а and 13500 ofi_ts).
0
REFERENCES
1
. Yudintsev, S.V., Geol. Rudn. Mestorozhd., 2003,
vol. 45, no. 2, pp. 172–187.
The doses required for amorphization of the ferrite
[
0
5] and yttrium aluminum garnets somewhat differ:
.17 and 0.3 displacement per atom, respectively. Yt-
2. Burakov, B.E., Anderson, E.B., Knecht, D.A., et al.,
Mater. Res. Sos. Symp. Proc., 1999, vol. 556, pp. 55–62.
3. Gorskii, V.V., At. Tekh. Rubezhom, 2000, no. 10,
pp. 3–8.
4. Lukinykh, A.N., Tomilin, S.V., Lizin, A.A., et al., Ra-
diokhimiya, 2008, vol. 50, no. 5, pp. 469–474.
5. Lukinykh, A.N., Tomilin, S.V., Livshits, T.S., et al., in
Proc. 11th Int. Conf. on Environmental Remediation
and Radioactive Waste Management, Bruges (Belgium),
September 2–6, 2007, CD-version.
trium aluminum garnet appeared to be more resistant
to self-irradiation. The amorphization doses for natural
silicate garnets and artificial phases with U, Th, Ce,
and Gd, determined by their ion bombardment, vary
from 0.15 to 0.30 displacement per atom [6, 7]. Thus,
the radiation resistance of phases of the garnet struc-
ture at 25°С depends on the composition. At the same
time, this difference is not so significant as for matri-
ces of the pyrochlore structure whose radiation resis-
tance strongly depends on the gross chemical composi-
tion, in particular, on replacement of Ti in the lattice
by Zr, Hf, or Sn [8].
6
7
8
. Utsunomiya, S., Wang, L.M., Yudintsev, S., and Ew-
ing, R.C., J. Nucl. Mater., 2002, vol. 303, p. 177.
. Utsunomiya, S., Yudintsev, S., and Ewing, R.C.,
J. Nucl. Mater., 2005, vol. 306, pp. 251–260.
. Yudintsev, S.V., Lapina, M.I., Ptashkin, A.G., et al.,
Mater. Res. Soc. Symp. Proc., 2002, vol. 713, CD-
ROM, paper JJ11.28.
Thus, we have shown that yttrium aluminum garnet
244
doped with Cm is more resistant to internal irradia-
tion compared to ferrite garnet. The chemical resis-
tance of the amorphized ceramic is lower than that of
9
. Zamoryanskaya, M.V. and Burakov, B.E., Mater. Res.
Soc. Symp. Proc., 2000, vol. 608, pp. 437–432.
RADIOCHEMISTRY Vol. 53 No. 2 2011