Ion Irradiation Effects in Synthetic Garnets Incorporating Actinides ...

Mat. Res. Soc. Symp. Proc. Vol. 713 © 2002 Materials Research Society
Ion Irradiation Effects in Synthetic Garnets Incorporating Actinides
Satoshi Utsunomiya 1 , Lu-Min Wang 1 , Sergey Yudintsev 2 and Rodney C. Ewing 1,3
1 Department of Nuclear Engineering and Radiological Sciences, 3 Department of Geological
Sciences, The University of Michigan, Ann Arbor, Michigan 48109-2104, U.S.A.
2 Institute of Geology of Ore Deposits, Russian Academy of Sciences, Staromonetny 35, 109017
Moscow, RUSSIA.
ABSTRACT
Radiation durability of garnet [A 3 B 2 (XO 4 ) 3 ; I a3d ; Z=8] has been examined by 1.0 MeV Kr 2+
irradiation with in situ transmission electron microscopy over the temperature range of 50 to
1070 K. The targets were five synthetic garnets incorporating various contents of actinides and
andradite, Ca 3 Fe 2 Si 3 O 12 . The synthetic garnets were silicates (N series) and ferrate-aluminate
series (G series).
The critical amorphization temperatures (T c ), above which amorphization does not occur,
were determined to be 1050 K for N77, 1130 K for N56, 1100 K for G3, 890 K for G4 and 1030
K for andradite. T c of the synthetic garnets increased as the average atomic mass of the garnet
increased. The maximum transferred energy by ballistic interaction was positively correlated to
the atomic mass. The larger cascade size that formed due to the larger E max might lead to the
higher T c .
INTRODUCTION
Incorporation of actinides into ceramics is an important issue for the immobilization of
radionuclides in the high level waste (HLW) form, and the radiation durability of potential host
materials for HLW form requires careful consideration [1, 2]. Because the long term radiation
effects due to radioactive decay can be simulated in short term with heavy ion-irradiation[3],
many irradiation experiments using heavy ions have been completed in the potential ceramics for
HLW, which are summarized in [4, 5]. Garnet is one of the candidates for HLW host because it
can incorporate actinides, Zr and rare earth elements into its crystal structure. Garnet based
ceramics have been synthesized recently for the immobilization of actinide elements [6-8].
Wang et al. [9] first investigated ion-irradiation effects in complex silicates with in situ
transmission electron microscopy (TEM). Eby et al. [10] then examined the relationship between
physical-chemical properties and the critical amorphization dose of 25 complex silicates at room
temperature. Almandine, andradite, grossular and spessartine were included in the Eby et al.
study [10]. But the temperature dependence of the critical amorphization dose of the garnet was
not investigated in this first study. A few data for pyrope at T < 673 K were measured in a
systematic ion irradiation study of phase in the MgO-Al 2 O 3 -SiO 2 system [11]. Synthetic garnets
incorporating rare earth elements (REE) are also important for other industrial applications,
particularly yttrium aluminum garnet (YAG), Y 3 Al 5 O 12 and YAG doped by other ions for use in
laser systems and in digital display systems [12]. However the radiation susceptibility of the
garnet structure incorporating actinides, REE, etc. to the amorphization has never been
investigated. In the present study, ion irradiation (1.0 MeV Kr 2+ ) experiments are completed on
five synthetic garnets incorporating different compositions of actinides, Zr, rare earth elements in
addition to andradite [13].
JJ11.31.1

Figure 1. Schematic figure of garnet structure from the view along [001].
Garnet (A 3 B 2 (XO 4 ) 3 ; I a3d ; Z = 2) consists of a distorted cubic close-packed array of oxygens
with isometric symmetry. Fig. 1 shows the structure of andradite viewed along the [001]
direction. Six-coordinated BO 6 octahedra and XO 4 tetrahedra establish a framework structure
alternately sharing their corners. The eight-coordinated A-site cation forms AO 8 dodecahedra.
For synthetic garnet, both the A-site and B-site can incorporate cations such as actinides and
REE depending on the charge balance. The X-site is replaceable by Fe 3+ , Al 3+ , Ga 3+ , Ge 4+ and
V 5+ . These are ferrates, aluminates, gallates, germinates and vanadates, respectively.
EXPERIMENTAL METHODS
The chemical compositions of the phases in each sample were determined by electron
microprobe analysis, EMPA (Cameca, CAMEBAX). The samples were analyzed by a focused
beam spot with a beam current of 20 nA and an accelerating voltage of 20 keV. The Cameca
PAP correction routine (modified ZAF) was used for data reduction. Interferences from some
overlapping rare earth elements were checked before the analysis.
Because there were several phases besides garnets in the samples, back-scattered electron
imaging (BEI) and semi-quantitative analyses were performed by the field emission scanning
electron microscopy (FE-SEM, Philips XL30). Then, transmission electron microscopy and
analytical electron microscopy (TEM and AEM, with a JEOL 2010F) were used for the initial
identification of the phases in each sample. TEM specimens were prepared by mechanical
polishing to a thickness of a few tens of µm, followed by ion milling using 4.0 keV Ar + . Before
ion-irradiation, all TEM specimens were observed by BEI and the maps of BEI were made for all
specimens to make sure the exact positions of garnets in the irradiated section.
All samples were irradiated with in situ TEM observation using 1.0 MeV Kr 2+ ions in the
IVEM (intermediate-voltage electron microscope) at Tandem Facility of Argonne National
Laboratory. Dose rate was varied from 12.5 x 10 10 to 50 x 10 10 ions/cm 2 /sec. The specimen
temperature during irradiation varied from 50 K to 1073 K. Selected area electron diffraction
(SAED) was used to monitor the amorphization process during intervals of increasing
irradiation. Subsequent observation was carried out by high resolution TEM (HRTEM). The ion
dose for complete amorphization, D c (ions/cm 2 ) was converted to displacement per atom (dpa)
JJ11.31.2

Table 1. Chemical composition (wt%) of the synthetic garnet normalized by 24 oxygen.
and to the kinetic energy transferred to each target atom through nuclear collision (E n ) using
SRIM2000 [14]. In the calculation, the displacement threshold energy, E d , was assumed to be 23
eV for Si, 47 eV for O and 79 eV for the other cations in the dodecahedral site. These values
correspond to E d value used in calculations for zircon [5], and 20 eV was used for octahedral
cation-site in garnet. The critical amorphization temperature (T c ), above which completely
amorphization does not occur, can be obtain by fitting the dose-temperature data based on the
model by Weber [3, 5].
RESULTS AND DISCUSSION
Chemical compositions by EMPA are given in Table 1. The synthetic garnets in this study
are classified to two types: silicate garnet (N series) and ferrate-aluminate garnet (G series). The
radiation-induced transformation of the G3 garnet is shown in the sequence of SAD patterns (Fig.
2). The sequence of SAED patterns reveals that the strong contrast of the defuse halo
(amorphous ring) appears during irradiation (Fig. 1b).
HRTEM images of G3 garnet during irradiation at room temperature are shown in Fig. 3.
Disordered or amorphous domains (~3nm) form in the collision cascades that appear in the
periodic structure of the garnet and the amorphous domains develop by cascade overlap until
isolated domains of crystals remain in an amorphous matrix (Fig. 3b) before complete
amorphization.
(a) (b) (c)
Figure 2. Selected area electron diffraction (SAED) patterns of transition for the G3 garnet
irradiated by 1.0 MeV Kr ++ at room temperature. (a) initial stage, (b) 0.14 dpa and (d) 0.22
dpa.
JJ11.31.3

(a)
(b)
5 nm [111] 5 nm
[001]
Figure 3. HRTEM images of G3 garnet irradiated at room temperature showing formation
of amorphous domains. (a) initial stage, 0 dpa and (b) 0.14 dpa.
The temperature dependences of D c for the synthetic garnets (Fig. 4) show a typical behavior,
that is an increase in amorphization dose at higher temperature. The G2 garnet does not show a
dramatic increase of D c in the present temperature range (50-1073 K). T c can be obtained for the
synthetic garnets except for G2; 1100 K for G3, 890 K for G4, 1130 K for N56 and 1050 K for
N77. T c of G2 can be above 1073 K. In the aluminate-ferrate series (G-series), the T c varies
largely. It may be ascribed to the fact that the G-series garnets are solid-solution of aluminate
and ferrate. Activation energies for radiation-enhanced defect annealing, E a (eV) [3], were also
calculated to be 0.46, 0.35, 0.36 and 0.31, respectively.
Figure 4. Temperature dependence of amorphization dose (dpa) for the synthetic garnets.
JJ11.31.4

The atomic mass of the target material is a possible parameter which effects on the
susceptibility to amorphization. The mass of the target will affect the nuclear cross section or the
maximum transferable energy for ballistic interaction, which in turn effects the size of the
displacement cascade.
The maximum transferable energy, E max , for an elastic interaction is:
E
max
=
4M
1M
E
( M + M
1
2 0
2
2
)
(4)
where, M 1 , M 2 , Z 1 and Z 2 represent atomic mass and atomic number of the ion and the target. E 0
is the initial energy of incident ion. The E max calculated by the equation is also given in Table 2.
In both of the silicate and the aluminate-ferrate garnet series, T c increases as E max increases. If
the transferred energy by collision is large, the size of the cascade becomes large. The cascade
size is correlated with T c in the model by Wang et al. [14] based on direct impact amorphization
within the displacement cascade.
T
c
~ T
m
( T
−
m
− T ) R
g
rB
cryst
(5)
where T m , T g , R cryst , r and B are the melting temperature, the glass-transition temperature, the
crystallization rate, the size of the cascade radius and a constant related to heat diffusivity,
respectively. Although some parameters are related to the target besides the cascade size, T c has
a positive correlation with the cascade size. Therefore the E max is a reasonable parameter to
explain the present results. The relationship between T c -mass is consistent with previously
proposed T c -mass relationships for A-site cations of the zircon-structure phosphate APO 4 with
various A-site cations [15].
CONCLUSION
Radiation experiments with 1.0 MeV Kr 2+ were completed in five synthetic garnets
incorporating various compositions of actinides and andradite. All these phases except for G4 are
susceptible to radiation induced amorphization below 1000K. T c of the silicate garnet series,
N56, N77 and andradite, are 1130 K, 1050 K and 1030 K, respectively, while the T c of ferratealuminate
garnet series, G3 and G4, are 1100 K and 890 K, respectively. Both series show a
trend of increasing T c as the atomic mass of the garnet increases. The larger atomic mass of the
Table 2. The summary of T c (K), the molecular weight normalized by 12 oxygen and the
maximum transferable energy by elastic collision, E max (keV)
silicate series
aluminate-ferrate series
andradite N56 N77 G2 G3 G4
molecular weight 508.2 585.6 506.1 791.9 807.5 729.5
T c (K) 1030 1140 1050 --- 1100 890
E max (keV) 714 769 729 877 840
JJ11.31.5

target leads to the larger energy transfer by collision and eventually to a larger cascade size.
Because T c has positive correlation with the cascade size, heavier target material has a higher T c .
The T c -mass relationship is consistent with previously reported T c -mass relationships for A-site
cations in zircon-structure phosphate.
ACKNOWLEDGEMENTS
The authors thank the staff of the HVEM/IVEM-Tandem Facility at Argonne National
Laboratory for assistance during the ion irradiation. S.U. thanks the staff of the Electron
Microbeam Analysis Laboratory at University of Michigan and Chris Palenik for helping with
EMPA. This work was supported by US DOE, Office of Basic Energy Sciences under grant DE-
FGO2-97ER45656.
REFERENCES
1. A. E. Ringwood, S. E. Kesson, N. G. Ware, W. Hibberson, A. Major, Nature 278, 219
(1979).
2. R. C. Ewing, Proc. Natl. Acad. Sci. USA 96, 3432 (1999).
3. W. J. Weber, R. C. Ewing, L. M. Wang, J. Mater. Res. 9, 688 (1994).
4. R. C. Ewing, A. Meldrum, L. M. Wang, S. X. Wang, Rev. in Miner. Gechem. 39 ,319 (2000).
5. W. J. Weber, R. C. Ewing, C. R. A. Catlow, T. Diaz de la Rubia, L. W. Hobbs, C. Kinoshita,
Hj. Matzke, A. T. Motta, M. Nastasi, E. K. H. Salje, E. R. Vance, S. J. Zinkle, J. Mater. Res.
13, 1434 (1998).
6. B. E. Burakov, E. B. Anderson, D. A. Knecht, M. A. Zamoryanskaya, E. E. Strykanova, M.
A. Yagovkina, Mater. Res. Soc. Symp. Proc. 556, 55 (1999).
7. B. E. Burakov, E. B. Anderson, D. A. Knecht, M. A. Zamoryanskaya, M. A. Petrova, Mater.
Res. Soc. Symp. Proc. 608, 419 (2000).
8. S. V. Yudintsev, M. I. Lapina, A. G. Ptashkin, T. V. Ioudintseva, S. Utsunomiya, L. M.
Wang, R. C. Ewing, Mater. Res. Soc. Symp. Proc. (2001) Submitted.
9. L. M. Wang, R. K. Eby, J. Janeczek, R. C. Ewing, Nucl. Instrum. and Meth. B 59/60,395
(1991).
10. R. K. Eby, R. C. Ewing, R. C. Birtcher, J. Mater. Res. 7, 3080 (1992).
11. S. X. Wang, L. M. Wang, R. C. Ewing, R. H. Doremus, J. Non-Cryst. Solids 238, 198
(1998).
12. Y. C. Kang, I. W. Lenggoro, S. B. Park, K. Okuyama, Mater. Res. Bull. 35, 789 (2000).
13. S. Utsunomiya, L. M. Wang, R. C. Ewing, Nucl. Instrum. and Meth. B (Submitted)
14. S. X. Wang, L. M. Wang, R. C. Ewing, Phys. Rev. B 63, 024105-1 (2001).
15. Meldrum, S. J. Zinkle, L. A. Boatner, R. C. Ewing, Phys. Rev. B 56, 13,805 (1997).
JJ11.31.6