⇦Chinese Physics Letters, 2020, Vol. 37, No. 5, Article code 056101 Bubble Formation in Apatite Structures by He-Ion Irradiation at High Temperature * Cai-Yu Wu (吴采宇)1, Ting-Ting Gao (高庭庭)2, Zhi-Wei Lin (林智威)1, Yue Zhang (张悦)3, Huan-Huan He (何环环)1, Jian Zhang (张建)1** Affiliations 1College of Energy, Xiamen University, Xiamen 361005 2Laboratory of Dielectric Materials, School of Materials Science and Engineering, Zhejiang University, Hangzhou 310027 3College of Physical Science and Technology, Xiamen University, Xiamen 361005 Received 23 February 2020, online 25 April 2020 *Supported by the Natural Science Foundation of Fujian Province, China (Grant No. 2017J01007) and partially by the Energy Development Foundation of Energy College (Grant No. 2018NYFZ01).
**Corresponding author. Email: zhangjian@xmu.edu.cn Citation Text: Wu C Y, Gao T T, Lin Z W, Zhang Y and He H H et al 2020 Chin. Phys. Lett. 37 056101    Abstract Apatite ceramics Ca$_{10}$(PO$_{4})_{6}$$X_{2}$ ($X=$F, OH) were prepared by the standard solid state sintering method and irradiated with He ions under a fluence of $5\times 10^{16}$ ions/cm$^{2}$ at 450 $^{\circ}\!$C. Irradiation induced formation and growth of the He bubbles were observed by a transmission electron microscope. Hydroxyapatite Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$ and fluoroapatite Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ with different He bubble morphologies indicate the influence of OH$^{-}$/F$^{-}$ substitution on the He-ion annealing efficiency, as well as the structure itself, which affects the process of He bubble evolution and formation. The grain boundaries also act as sinks to accumulate He bubbles. No obvious irradiation damage but slight intensity reduction and left shift of diffraction peaks were observed according to the grazing incidence x-ray diffraction and Raman spectra characterizations, indicating that defects of interstitials and vacancies were generated. DOI:10.1088/0256-307X/37/5/056101 PACS:61.72.U-, 61.80.Jh, 68.55.A-, 28.52.Fa © 2020 Chinese Physics Society Article Text The management of high level nuclear radioactive wastes, particularly the long-lived actinides such as Pu, Np, and Am, and various types of fission products (FP), is crucial to development of safe nuclear energy. As a promising candidate material to immobilize the high-level waste (HLW), apatite with a general formula of Ca$_{10}$(PO$_{4})_{6}$$X_{2}$ ($X=F$, Cl, OH) has attracted much scientific attention.[1–3] The crystal structure of apatite (space group $P6_{3}/m$) with two calcium positions allows substitution of a wide range of cations and anions.[4–6] The $\alpha$ particles and $\alpha$-recoil nuclei will greatly reduce the structure's stability and even cause the transition of crystalline to amorphization, which exhibits volume swelling in micro-structure and significant increasing in the water leach rate.[7] Besides the crystal structure damage induced by $\alpha$-recoil nuclei, He ions generated due to $\alpha$-decay also cannot be neglected, which will be apt to accumulate to a critical concentration for bubble nucleation in the host matrix, and may eventually lead to microscopic swelling and surface blistering. Miro et al.[8] studied the He diffusion constants in Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ and Ca$_{4}$Nd$_{6}$(SiO$_{4})_{6}$F$_{2}$ to determine the influence of the substitution (cationic/anionic) on He diffusion. Wang et al.[9] investigated the effects of different grain sizes on He bubble accumulation in T91. The effects of stress concentration on the evolution of He bubbles have been studied in He-irradiated aluminium.[10] Beaufort et al.[11] investigated the evolution of bubble diameter with $5 \times 10^{16}$ ions/cm$^{2}$ He irradiation after pre-damaged by heavy ion Au irradiations. In this work, He bubble accumulation is studied in Ca$_{10}$(PO$_{4})_{6}$$X_{2}$ ($X=$F/OH) to deepen the understanding of the mechanism for He bubble formation and evolution in the $X$ site. The polycrystalline hexagonal apatite structure ceramics Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ were prepared using the standard solid state sintering from a stoichiometric mixture of Ca(H$_{2}$PO$_{4})_{2}$ (99.99%), CaF$_{2}$ (99.99%) and Ca(OH)$_{2}$ (99.99%) according to the following reaction: 3Ca(H$_{2}$PO$_{4})_{2}+$6Ca(OH)$_{2}+$CaF$_{2}\to$ Ca$_{10}$(PO$_{4})_{6}$F$_{2}+$12H$_{2}$O$\uparrow$ The Ca(H$_{2}$PO$_{4})_{2}$ and Ca(OH)$_{2}$ powders were preheated at 150 $^{\circ}\!$C for 8 h before weighting to remove hygroscopicity and other volatile impurities and then ball milled for 4 h. The resulting powders were compacted into pellets and subsequently sintered at 1100 $^{\circ}\!$C for 24 h. In order to attain a better homogeneity, the samples were ball milled again for 4 h and pressed into pellets, then sintered 48 h under 1300 $^{\circ}\!$C. Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$ was synthesized by Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$ power, and sintered under the same condition as Ca$_{10}$(PO$_{4})_{6}$F$_{2}$. The measured densities of both Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ and Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$ samples were above 95% of theoretical densities, respectively. The sample stage was heated to 450 $^{\circ}\!$C, and then the samples were implanted by 80 keV He ions to a dose of $5 \times 10^{16}$ ions/cm$^{2}$ with the ion flux kept at $2 \times 10^{13}$ ions/(cm$^{2}\cdot$s) (NEC-400 kV high current ion implanter, Xiamen University). The radiation dose was evaluated using the stopping and range of ions in matter (SRIM) code. The sintered pellets before and after irradiation were characterized by grazing incidence x-ray diffraction (GIXRD) (Rigaku Ultima IV Advanced x-ray diffractometer) with Cu $K_\alpha$ radiation in the 2$\theta$ range of 20–55$^{\circ}$ and transmission electron microscopy (TEM, FEI Tecnai F30) as well as Raman spectra (HR-800, HORIBA, Jobin Yvon) with 633 nm excitation. The incident angle of the GIXRD is fixed at 0.5$^{\circ}$. The samples for cross-sectional TEM specimens were prepared as follows: first cohering the irradiation surface onto a $\phi$3 mm copper-mesh TEM grid then polished the cross section down to a thickness of 20 µm or so. Then, a precise ion polishing system (PIPS, GATAN PIPS II 695) with Ar beam energy of 5 keV gradually decreasing to 3 keV and 1 keV was employed to obtain the proper thickness for TEM observation. TEM was conducted to observe the He bubble evolution in the irradiated sample operating at 300 keV. Figure 1 shows the result of SRIM calculated damage (dpa) and ion concentration profiles for Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ (similar results are obtained in Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$). According to the SRIM simulation, the calculated peak damage is about 1.5 dpa located at around 470 nm, the maximum concentration is at a depth of 530 nm and the ions are in the range of about 0–800 nm.
cpl-37-5-056101-fig1.png
Fig. 1. He-ion concentration and distribution of irradiation damage as a function of depth in the Ca$_{10}$(PO$_{4}$)$_{6}$F$_{2}$ by SRIM simulation.
cpl-37-5-056101-fig2.png
Fig. 2. GIXRD and Raman patterns of Ca$_{10}$(PO$_{4})_{6}$$X_{2}$ ($X=$F, OH) ceramics before and after He irradiation.
As shown in Figs. 2(a) and 2(b) after ion irradiation for Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ and Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$, slight intensity reduction and left shift of diffraction peaks were observed in both apatite samples. Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ and Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$ adopt the $P6_{3}/m$ space group, and the lattice parameters calculated from the GIXRD data by the least-squares method are as follows: $a=b=9.3744$(33) Å, $c=6.8846$(24) Å, $V_{\rm cell}= 524.0(3)$ Å$^{3}$ and $a=b=9.3816$(12) Å, $c=6.8903$(49) Å, $V_{\rm cell}= 525.2(4)$ Å$^{3}$ are for Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ before and after irradiation, respectively, while the corresponding lattice parameters for Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$ are $a=b=9.4114$(51) Å, $c=6.8991$(36) Å, $V_{\rm cell}= 529.2(5)$ Å$^{3}$ and $a=b=9.4164$(48) Å, $c=6.9048$(30) Å, $V_{\rm cell}= 530.2(4)$ Å$^{3}$, respectively. Both lattice parameters $a$ and $c$ increase slightly about 0.05%–0.08% after He-ion irradiation in Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ and Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$ samples, and the lattice expansion is caused by the defects of He ions and self-interstitials after He-ion irradiation.[12] In the present work, the Raman signal contains the response of the damaged and undamaged layer underneath due to the deeper visible light penetration depth than the damaged region. Figures 2(c) and 2(d) give the Raman spectra of Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ and Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$, and four asymmetric stretching vibrations are observed below 1200 cm$^{-1}$, which arise from $\upsilon_{2}$, $\upsilon_{4}$, $\upsilon_{1}$ and $\upsilon_{3}$ for ${\mathrm{PO}}_{4}^{3-}$ group in sequence.[13] The two broad bands around 1350 and 1600 cm$^{-1}$ have been attributed to fluorescence emission band that may be caused by rare earth impurities.[14] The intensity reduction of GIXRD diffraction peaks and the obvious intensity drop of Raman modes indicates the ion irradiation induced amorphization of grains with ${\mathrm{PO}}_{4}^{3-}$ tetrahedra distortion to some extent.[15,16,17] However, the crystalline structure remains to be dominant under the present experimental condition since sharp GIXRD and Raman peaks can be clearly observed after He-ion irradiation.[18] In addition, the defects introduced by He-ion irradiation may reduce the transparency of the present apatite ceramics, in turn lead to the intensity reduction of the Raman modes. Furthermore, the absence of these two broad bands around 1350 and 1600 cm$^{-1}$ may be caused by the defects induced impurity-fluorescence emission vanishing. To observe the defects induced by irradiation as well as the formation and growth of He bubbles, cross sectional TEM is carried out to investigate the micro-structure process. The He bubbles are detected in over-focused and under-focused conditions to highlight the bubble edges. Figure 3(a) shows the TEM micrographs of numerous near-spherical He nano-bubbles (white dots) at the depth about 800 nm, and which are almost densely and homogeneously distributed in the Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$ samples. The diameter distribution of He bubbles is calculated in Fig. 3(b) with Gauss fit. According to the Gauss fit, the mean diameter of about 8.1$\pm$0.1 nm for He bubbles agrees well with the TEM obtained from Fig. 3(a). Figure 3(c) shows the TEM image of the He bubble distribution in a Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ sample at the depth about 800 nm, the He bubbles are not near-spherical as we observed in the Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$ sample, but cluster together and connect to other bubbles randomly. Thus it is difficult to recognize single He bubble and count the size and quantity. The TEM micrograph shows the process of development and merging of He bubbles, arrow A (marked in Fig. 3(c)) reveals that the He bubble is going to be absorbed by the neighboring large bubbles and becomes a part of connected bubble chains, while arrow B (marked in Fig. 3(c)) exhibits an isolated He bubble on account of the surrounding bubbles are integrated into a complex labyrinth of chains. According to the TEM images, the He bubble distribution shows great difference in two samples. The difference between the two samples may originate from the following two main reasons. Firstly, the OH$^{-}$/F$^{-}$ substitution has a strong influence on the He-ion annealing efficiency as well as the structure itself,[19] and the migration energy of F$^{-}$ ions is smaller than that of OH$^{-}$ ions in either lattices, namely, $E_{\rm m}$($F^{-}) < E_{\rm m}$(OH$^{-})$.[20] The lower $E_{\rm m}$(F$^{-}$) leads to an easier volume diffusion mechanisms[21] of He bubbles in Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ compared to Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$. Secondly, Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ may have higher ionicity than Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$. Higher ionicity results in higher resistance to amorphization and the competition between the long-range and short-range forces originated from the ionic and covalent contributions to bonding can define the efficiency of damage recovery and resistance to amorphization.[22] Considering these two reasons, it is reasonable that the migration energy of He bubbles in Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ is smaller than that in Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$, leading to the significant accumulation of He bubbles in Ca$_{10}$(PO$_{4})_{6}$F$_{2}$.
cpl-37-5-056101-fig3.png
Fig. 3. Cross sectional TEM images for samples implanted with 80 keV He ion with the fluence of $5 \times 10^{16}$ cm$^{-2}$ at 450 $^{\circ}\!$C. (a) Typical morphology of He bubbles in Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$ sample at about 800 nm. (b) Statistic of the diameter of He bubbles and fitted Gaussian line for (a). (c) He bubble distribution in Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ sample at about 800 nm. [(d), (e)] He bubble chain formed by He bubbles and migrate along the grain boundary or towards to surface in Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ samples.
The irradiation damage in both samples is more serious at the depth of 700 nm to 1000 nm during TEM observation, exceeding the SRIM simulation ion range of 800 nm and according to Fig. 1. The main reason is that the mobility of He interstitial and vacancy caused by irradiation is increased at high irradiation temperature but the SRIM code does not take into account the effect of evaluated temperature. In addition, the average grain size is about 1.5 µm, which is larger than the ion range. Thus the channeling effect may aggravate the deep range to some extent. Channeling effect on ion implantation has a significant effect for low energy ion implantation, and could transport the ion to further distance through the channel. The deviation of SRIM simulation and experimental value has also been reported. Zhang et al.[23] reported about 25% deviation between SRIM simulation and TEM observation in Au$^{+}$ ion irradiation in the GaN sample. Li et al.[14] observed that the damage peak was 23% deviation from SRIM simulation and owed the large discrepancies between SRIM simulation and experimental data to the overestimation of electronic stopping power of the medium mass ions. Grain boundary also tends to promote the nucleation and formation of He bubbles. As shown in Figs. 3(d) and 3(e), it can be observed that the average size and density of He bubbles nucleated at the grain boundary are larger and denser than those in the intragranular regions.[24,25] Grain boundary works as efficient sites for both interstitials and vacancies.[9,26] He ions have higher mobility and greater quantity than the vacancies, so the most He interstitials are maintained and accumulated along grain boundaries. Owing to higher mobility of He interstitial and vacancy at grain boundary, He bubbles accumulate and migrate deeper along the grain boundary. Figure 3(e) reveals that He bubbles also have a tendency of migration towards the surface by forming bubble chains. In summary, we have investigated the process of bubble evolution and formation under 80 keV He irradiation at 450 $^{\circ}\!$C in apatite Ca$_{10}$(PO$_{4})_{6}$$X_{2}$ ($X=$F and OH). Spherical He bubbles with the average diameter of 8.1 nm are formed in apatite Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$, while He bubbles cluster together and connect with other bubbles randomly in apatite Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ possibly due to the smaller migration energy and higher ionicity in Ca$_{10}$(PO$_{4})_{6}$F$_{2}$ than those of Ca$_{10}$(PO$_{4})_{6}$(OH)$_{2}$. Grain boundary also works as an efficient site to accelerate the nucleation and formation of He bubbles. References Thermal diffusion of molybdenum in apatiteThermal analysis of apatite structureNew constraints on electron-beam induced halogen migration in apatiteMechanisms involved in thermal diffusion of rare earth elements in apatiteFabrication and chemical durability of hot-pressed Na-bearing fluorapatite-type Ca8Sm1Na1(PO4)6F2 ceramic for immobilization of trivalent minor actinideSynthesis and microstructure of fluorapatite-type Ca10-2Sm Na (PO4)6F2 solid solutions for immobilization of trivalent minor actinideRadiation effects in crystalline ceramics for the immobilization of high-level nuclear waste and plutoniumEffect of composition on helium diffusion in fluoroapatites investigated with nuclear reaction analysisGrain size effects on He bubbles distribution and evolutionEvolution of helium bubbles induced by stress concentration in He-irradiated aluminiumIn-situ evolution of helium bubbles in SiC under irradiationEffect of He ion irradiation on the microstructure of t′ phase yttria-stabilized zirconia ceramic coatingsX-ray structure refinement, vibrational spectroscopy, and ionic conductivity of Na0.11Ca9.64Sm0.23(PO4)6F2Vibrational study and fluorescence bands in the FT–Raman spectra of Ca10−xPbx(PO4)6(OH)2 compoundsPreparation and the effects of ion irradiation on bulk SiOC ceramicsIrradiation damage in xenon-irradiated α-Al2O3 before and after annealingRaman spectroscopy of apatite irradiated with swift heavy ions with and without simultaneous exertion of high pressureStructural Distortion and Defects in Ti 3 AlC 2 irradiated by Fe and He IonsSimulation of the α-annealing effect in apatitic structures by He-ion irradiation: Influence of the silicate/phosphate ratio and of the OH−/F− substitutionThe migration of F−, OH− and O2− ions in apatitesModeling of He-bubble migration in bcc FeDamage and microstructure evolution in GaN under Au ion irradiationNucleation and growth of helium bubble at (110) twist grain boundaries in tungsten studied by molecular dynamicsEffects of helium implantation on mechanical properties of (Al 0.31 Cr 0.20 Fe 0.14 Ni 0.35 )O high entropy oxide filmsEfficient Annealing of Radiation Damage Near Grain Boundaries via Interstitial Emission
[1] Gaillard C, Chevarier N, Millard-Pinard N, Delichère P and Sainsot P 2000 Nucl. Instrum. Methods Phys. Res. Sect. B 161–163 646
[2] Tõnsuaadu K, Peld M and Bender V 2003 J. Therm. Anal. Calorim. 72 363
[3] Stock M J, Humphreys M C S, Smith V C, Johnson R D and Pyle D M 2014 Am. Mineral. 100 281
[4] Martin P, Carlot G, Chevarier A, Den-Auwer C and Panczerc G 1999 J. Nucl. Mater. 275 268
[5] Gao X N, Huang Y, Teng Y C, Yan M H, Zhang H B, Tuo X G and Peng S M 2018 J. Nucl. Mater. 507 297
[6] Huang Y, Zhang H, Zhou X and Peng S 2017 J. Nucl. Mater. 485 105
[7] Weber W J, Ewing R C, Catlow C R A, de la Rubia T D, Hobbs L W, Kinoshita C, Matzke H, Motta A T, Nastasi M, Salje E K H, Vance E R and Zinkle S J 1998 J. Mater. Res. 13 1434
[8] Miro S, Studer F, Costantini J M, Haussy J, Trouslard P and Grob J J 2006 J. Nucl. Mater. 355 1
[9] Wang J, Gao X, Gao N, Wang Z G, Cui M H, Wei K F, Yao C F, Sun J R, Li B S, Zhu Y B, Pang L L, Li Y F, Wang D and Xie E Q 2015 J. Nucl. Mater. 457 182
[10] Chen H, Liu W, Cheng Y, Peng J, Liang X, Hu L and Wang G 2019 Mater. Lett. 247 11
[11] Beaufort M F, Vallet M, Nicolaï J, Oliviero E and Barbot J F 2015 J. Appl. Phys. 118 205904
[12] Liu B, Zou J, Lin L, Wang S, Bai Y, Liu C and Li Q 2018 Nucl. Instrum. Methods Phys. Res. Sect. B 431 67
[13] Boujelbene M and Mhiri T 2014 Ionics 20 1267
[14] Hadrich A, Lautié A and Mhiri T 2001 Spectrochim. Acta Part A 57 1673
[15] Zhao Z, Niu M, Wang H, Gao H, Peng K, Zang H and Ma M 2019 J. Eur. Ceram. Soc. 39 832
[16] Li B, Liu H, Kang L, Zhang T, Xu L and Xiong A 2019 J. Eur. Ceram. Soc. 39 4307
[17] Liu J, Glasmacher U A, Lang M, Trautmann C, Voss K O, Neumann R, Wagner G A and Miletich R 2008 Appl. Phys. A 91 17
[18] Pang L L, Li B S, Shen T L et al 2018 Chin. Phys. Lett. 35 026102
[19] Soulet S, Carpen J, Chaumont J, Kaitasov O, Ruault M O and Krupa J C 2001 Nucl. Instrum. Methods Phys. Res. Sect. B 184 383
[20] Tse C, Welch D O and Royce B S H 1973 Calc. Tiss. Res. 13 47
[21] Morishita K, Sugano R 2007 Nucl. Instrum. Methods Phys. Res. Sect. B 255 52
[22]Trachenko K 2004 J. Phys.: Condens. Matter 16 1491
[23] Zhang Y, Ishimaru M, Jagielski J, Zhang W, Zhu Z, Saraf L V, Jiang W, Thome L and Weber W J 2010 J. Phys. D 43 085303
[24] Li F B, Ran G, Gao N, Zhao S Q and Li N 2019 Chin. Phys. B 28 085203
[25] Yang Z M, Zhang K, Qiu N, Zhang H B, Wang Y and Chen J 2019 Chin. Phys. B 28 46201
[26] Bai X M, Voter A F, Hoagland R G, Nastasi M and Uberuaga B P 2010 Science 327 1631