- Email: [email protected]
Electronic excitations and defect creation in LiF crystals
a
Nuclear Instruments and Methodsin Physics Research B 107 ( 19%) 128- 132 __ __
Lk!IOM
B
Beam Interactions with Materials&Atoms
!I!!!
EISEVIER
Electronic excitations and defect creation in LiF crystals K. Schwartz Gesellschafrfir
Schwerionenforschung, Postfach 110552, D-64220 Darmstadt, Germany
Abstract A review of ion induced radiation damage processes in alkali halides is presented with special attention paid to Frenkel
defect creation by the decay of excitons. The exciton mechanism is analysed for radiation damage creation in LiF crystals under heavy ion (*@Bi (136.7 MeV/u) and 238U (11.4 MeV/u)) irradiation at fluences up to 5 X lo9 ions/cm*. In both cases F-and F,-centers are the dominant electronic defects in the track. Uranium tracks in LiF crystals can be etched, whereas after irradiation with Bi ions no track etching was observed.
1. Introduction
Under swift heavy ion irradiation with high electronic energy loss (d E/dx), in various solids, new radiation effects and processes have been observed [I -61. At such high excitation levels, collective electronic interaction processes are usually dominant. By strong electron-phonon coupling they lead to local heating (thermal spikes) with subsequent phase transitions, defect creation, and defect aggregation in the excited region of the heavy ion track [7,8]. However, in order to understand the formation of collective defects the creation of single point defects has to be taken into account. For such studies LiF and other alkali halides are privileged subjects due to their well-understood electronic structure, electron-phonon interaction, and defect creation processes under conventional irradiation (such as X-rays or electrons with an energy of - 1 MeV) [4,13,141. A review of radiation induced damage processes in alkali halides is presented. The creation of color centers is analysed and latent tracks of 238U ions (11.4 MeV/u) and 2WBi ions (136.7 MeV/u) are etched in LiF crystals.
2. Radiation damage processes in dielectrics Stimulated by the development of nuclear reactors, studies of radiation damage in solids began in the fifties [9]. At that time Varley proposed a mechanism for defect creation in ionic crystals by multiple ionization of anions followed by displacement from the regular lattice due to Coulomb interaction [lo]. Based on this mechanism, Fleischer et al. [l l] developed the Coulomb explosion model for defect formation in heavy ion tracks. However, further
investigations showed that the Varley mechanism in ionic crystals occurs only at the surface because the lifetime of multi-ionized anion states in the bulk is much shorter than the defect creation time determined by the lattice vibration period (- lo- r3 s; see Ref. [16]). In the sixties Frenkel defect creation by the decay of excited electronic states of the lattice (excitons, electrons and holes) was studied in detail [13,14]. Under conventional irradiation (e.g., X-rays, 1 MeV electrons) this mechanism is possible only in those dielectrics where the energy of the lowest exciton state E,,, exceeds the energy Ed needed to create a Frenkel pair. This takes place in alkali halides, but not in oxide materials (such as MgO, Al,Os and Y3Fe,0,,, and many other dielectrics) where under conventional irradiation, defects can be created only by direct electron-atom collisions as in metals 1151. The creation of defects in solids strongly depends on the irradiation temperature [12-15,23,24]. In some alkali halides at low temperature (7’~ 10-100 K) the efficiency of defect creation by the decay of excitons is extremely low, i.e. the defect creation energy is high [14]. At higher temperatures CT> lo-100 K) the efficiency of primary defect creation increases (F-H Frenkel pairs) as well as the mobility of self-trapped holes which strongly influence color center transformation and the probability of electron-hole recombination. Lithium colloid formation in LiF crystals is a complex process which depends on the irradiation dose, dose rate and irradiation temperature [ 12,16,22-241. Various lithium colloids with different absorption spectra and thermal stability are observed in irradiated LiF - from quasi-microscopic X-centers to large particles with a size of several micrometers [12,22,23]. A possible mechanism for the formation of lithium colloid centers is the aggregation of F-centers which usually takes place at higher temperatures
0168-583X/96/$15.00 0 1996 Elsevier Science B.V. All rights reserved SSDI 0168-583X(95)00846-2
K. Schwartz/Nucl.
Insrr. and Meth. in Phys. Res. B 107 (1996) 128-132
(where the diffusion of F-centers is strong) and at high irradiation doses [ 16,22,23]. Therefore, in LiF crystals at high irradiation doses strong radiolysis takes place with local phase transitions in the cation (metal colloids) and anion sub-lattice (molecular anion products in the bulk and at the surface) [ 12,161. However, the ionic binding determines that no complete amorphization of the lattice takes place [ 171. Perez et al. [18] and Balanzat et al. [4] demonstrated that defect creation in LiF, NaCl, and KBr is also dominated by the exciton mechanism under heavy ion irradiation (up to lz9Xe (27 MeV/u)). At low temperatures (T - 15 K) the energy needed to create an F-center under heavy ion irradiation is approximately the same as for X-ray excitation [4,14]. Nevertheless, at higher (dE/d x), some non-linearity in F,-center creation was observed [4]. Microspectrophotometric measurements indicated that Fand F,-centers are the dominant electronic color centers in the track. The presence of Li colloids in heavy ion irradiated LiF crystals is an open question because of the small oscillator strength of the colloidal centers and the spectral overlapping of their absorption band with F-and F,-absorption [ 12,16,18,22,24]. The observed chemical etching of U fission products in LiF crystals by Young [20] can possibly be explained by the presence of Li colloids. The energy
129
loss for uranium fission products ((d E/d x), _ 1.6 keV/A) is larger than for Xe ions used in the experiment of Perez et al. ((d E/d x), - 1 keV/A) [ 181. At the higher value of (d E/d x)= the aggregation process could be more efficient. An open question of radiation damage creation is the superposition of various excitations in the solid. Ion induced defect creation is a complex process which occurs on a time scale (TV)much larger than the energy transfer time 7;r from the heavy ion to the electronic and vibronic systems of the solid (TV,- IO- ” s) [7]. In alkali halides F-center creation occurs within 7F lo-*’ s [l4] and the efficiency of F-center creation increases with temperature. Thus the energy of F-center creation (i.e., the primary electronic radiation products) can be used to estimate the radiation damage efficiency in the solid. It is a general rule that only a small part of the absorbed radiation energy in the solid is transformed into defects (only a few percent or less). The strong electronphonon interaction in most solids leads to efficient conversion of the absorbed energy into lattice vibrations (the radiation induced heating exceeds 90% of the absorbed energy). Therefore, radiation induced heating in solids is a general phenomenon which was successfully demonstrated in the thermal spike model by Toulemonde et al. [7,8]. However, the thermal spike model usually describes
Energy conversion of heavy ions in alkali halides ion
primary electrons
time scale
10-l’ _ 10-16 s
energy transfer +
excitation of the
Fig. 1. Energy conversion and defect creation in alkali halide crystals under heavy ion irradiation: e“, ey, free and self-trappedexcitons; c-, e+, free electrons and holes; e: , self-trapped holes; o~~,.~, wac,i. qi, optical, acoustical and local (point defect) vibration modes; AT, haiiation induced temperature increase; Tc,i,,critical temperature of phase transition; N, Nd, defect concentration and the critical defect concentration for phase transition.
II. ELECTRONIC ENERGY RELAXATION
130
K. Schwartz/Nucl. Instr. and Meth. in Phys. Res. B 107 (1996) 128-132
heavy ion track damage formation only as a result of local melting, not taking into account the possibility of a superposition of the local heating with the excited electronic states of the lattice. Due to the long’ lifetime of self-trapped excitons in alkali halides (for the triplet states several milliseconds and more [19]) a superposition of the exciton defect creation process with radiation induced heating is possible (Fig. 1). Radiation induced local heating (AT) can influence the efficiency of Frenkel pair creation as well as the aggregation of single defects, e.g. the formation of Li colloid centers, or the creation of molecular F2 (a phase transition is also possible at T < T,,, if the defect concentration Nd exceeds NC,,; Fig. 1) [13,16]. The superposition of the heavy ion induced electronic and thermal excitations can explain the influence of (dE/dx), on defect creation in solids under heavy ion irradiation. However, the understanding of such a superposition needs more detailed experimental and theoretical study (see the review [25]).
3. Experimental results on color center creation and chemical etching Chemical etching is widely used to reveal latent tracks [11,20]. This technique dissolves the ion track in the solid in a suitable solution, but usually it does not characterize the nature of the defects in the track. However, chemical etching is a very simple and sensitive method which gives evidence of the presence of latent tracks in the solid. Young showed that etched tracks of U fission fragments in LiF crystals were different from etched pits due to dislocation [20]. Therefore, it should be possible to correlate etching properties and the structure of color centers in ion tracks.
LiF crystals (10 X 10 X 2 mm3) were irradiated with 209Bi (136.7 MeV/u; fluence IO9 ions/cm2) and 238U ions (11.4 MeV/u; 106-5 X lo9 ions/cm2) at GSI, Darmstadt. Samples for chemical etching were irradiated through a mask (Fig. 2a). For irradiated samples with a fluence 2 lo9 ions/cm2 absorption spectra in the range 200-700 nm were measured. To examine the dislocation structure and the existence of ion tracks all crystals were etched in a solution of HF (1 vol. part) and acetic acid (1 vol. part) saturated with FeF, before and after irradiation. This etchant was used earlier by Young to etch U fission tracks in LiF crystals [20]. The etching time was in the range from 10 to 30 min. The etched pits were observed by an optical and scanning electron microscope. Before inadiation all samples showed a dislocation density of several lo5 disl/cm2, which is typical for alkali halides grown from a melt. Due to the mask, irradiated and non-irradiated areas could easily be distinguished and allowed comparison of the etched pits of dislocations (usually larger) and the etched pits of ion latent tracks. Each
Fig. 2. Etched tracks in an irradiated LiF crystal (U ions, 11.4 MeV/u; fluence lo7 ions/cm’) after 30 min etching observed using a scanning electron microscope. The irradiation was performed through a mask. The density of etched pits is 10’ f 10% tracks/cm*. (a) Etched pits of U ions in the irmdiated area (upper part) and etched pits of dislocations (bottom) in the non-irradiated area (magnification: 200). (b) Etched pits in the irradiated area; the large etched pits are due to dislocations. The concentration of dislocations of about lo5 cm-* IS . not changed by ion irradiation (magnification: 800).
latent track of U ions was etched to a prismatic pit with a size from 1 to 3 p,rn (Fig. 2). In the case of Bi ions (136.7 MeV/u) no etched tracks were observed on the surface of irradiated LiF and the dislocation structure of the crystal was not changed. The absorption spectrum was measured only for crystals irmdiated with fluences above lo9 ions/cm2. In the absorption spectrum F-and F,-centers were dominant. However, for Bi ions the concentration of F-centers was two orders of magnitude smaller than in crystals irradiated with U ions. The concentration of F-centers NF was determined using the absorption coefficient pa, NF [cme3] = 9.48 X lOI& [cm-‘], according to the SmakulaDexter formula [ 181.The concentration of F-centers in LiF crystals irradiated with U at a fluence of 2 X IO9 ions/cm2 was N = 4 X 1017cm-3 which corresponds to an F-center creatioi energy A EF = h)Eabs/NF. track= 1300 + 150 eV.
K. Schwarrz/Nucl.
4. Discussion
Instr. and Meth. in Phys. Res. B 107 (1996) 128-132
and conclusion
After the etching of U fission tracks in LiF crystals by Young [20], there were no further investigations in this field. Our results demonstrate that U ions (11.4 MeV/u) with a specific energy loss (d E/dx), = 2.86 keV/i induce etchable tracks in LiF crystals, while irradiation with Bi ions (136.7 MeV/u) with (d E/d x& = 0.8 keV/A does not induce etchable tracks. Obviously, the value of (dE/dx) for Bi ions is below the threshold for track etching. Taking into account the results of Young ((d E/d x), N 1.6 keV/& the critical value for latent track etching in LiF is in the range 0.8 keV/W < (d E/d x), crit < 1.6 keV/W. However, an open question is the origin’of latent track etching in LiF crystals, i.e. what kind of defects are responsible for selective etching. Chemical etching of point defects in LiF or other alkali halides has never been observed after irradiation with electrons, X-rays, protons or o-particles [20,21]. Therefore, single point defects (F- and V-centers) and microscopic aggregates (Fz-, Fs-centers) are probably not sufficient. It is assumed that larger aggregates of F-centers, e.g. Li colloids, are responsible for the track etching. Such centers are created at high radiation dose and at irradiation temperatures exceeding the critical temperature for Li colloid formation, Tctit2 500 K. However, it is very difficult to observe these colloid centers by optical spectroscopy, since the oscillator strength of colloidal centers is much smaller than in the case of F-centers (usually the concentration of colloidal centers is low and an overlapping of the absorption spectrum with F-and F,-centers is possible). Perez et al. studied color center creation in heavy ion tracks in LiF up to Xe (27 MeV/u) with an energy loss of (dE/dx), N 1 keV/A, which is close to the situation of Bi (136.7 MeV/u, d E/d x), = 0.8 keV/.&, i.e. below the critical value for etching. For a more detailed analysis of the etching threshold and the origin of etching in LiF crystals (i.e., in order to determine which defects are
131
responsible for the etching) further experiments with various ions are necessary. Additional information on aggregate and colloid center creation can be obtained from the analysis of the energy needed A E, to create an F-center. The observed threshold value of (dE/d x), for track etching can be correlated with the dependence of A E, on the excitation density, i.e. the magnitude of (dE/dx). Under similar irradiation conditions the energy A E, must increase when strong aggregation takes place. Therefore, we analysed the F-center creation energy A E, for Ne, Ar, and Xe ions according to data of Ref. [ 181 and for U ions according to our results (under the same irradiation conditions of _ 10’ ions/cm’). The value of AE, increases with increasing energy loss (Fig. 3). This can be explained as an aggregation effect of F-centers to nF-centers and finally to Li colloids, or as a thermal annealing effect of F-centers. However, the value of A E, for Ne and Ar ions is smaller than A E, for X-ray excitation [ 141, which probably suggests that thermal annealing can be excluded. Therefore, the increase of A E, for U ions (Fig. 3) can be explained from the viewpoint of aggregate and colloidal center creation which determines the chemical etching. However, without additional experimental analysis of the track morphology this is only an assumption.
Acknowledgements
The author is grateful to Prof. A. Scharmann (University of Giessen) for the LiF crystals and to Dr. C. Trautmann (GSI) for many fruitful discussions and assistance in the experiment.
References
[II E. Balanzat, Radiat. Eff. and Def. in Solids 126 (1993) 97. Dl Ming-dong Hou, S. Kiaumlinzer and G. Schumacher, Phys. Rev. B 41 (1990) 1144.
[31 A. Dunlop, D. Lesueur, P. Legrand and H. Danmak, Nucl. Instr. and Meth. B 90 (1994) 330.
[41 E. Balanzat, S. Bouffard, A. Cassimi, E. Doorhyee, L. Protin,
4001 2
3
4
5
6
7
8
9
10
In @%I@ Fig. 3. F-center creation energy under irradiation with various heavy ions: Ne (40 MeV/u), Ar (60 MeV/u), Xe (27 MeV/u) after Ref. [ 181; U (11.4 MeV/u) - our results. For all measurements the fluence was in the range of a few 10’ ions/cm’.
J. Grandin, J.L. Doualan and J. Margerie, Nucl. Instr. and Meth. B 91 (1994) 134. [51 C. Trautmann, R. Spohr and M. Toulemonde, Nucl. Instr. and Me&t. B 83 (1993) 513. [61 E. Balanzat, S. Bouffard, A. Le Mabel and N. Betz, Nucl. Ins&. and Meth. B 91 (1994) 140. [71 M. Toulemonde, S. Bouffard and F. Studer, Nucl. Instr. and Meth. B 91 (1994) 108. [81 Z.G. Wang, Ch. Dufour, E. Paumier and M. Toulemonde, J. Phys. Condensed Matter 6 (1994) 6733. [91 F. Seitz and E.P. Wigner, Sci. Am. 8 (1956) 76. [lOI J.H.O. Varley, Nature 174 (1954) 886. [Ill R.F. Fleischer, P.B. Price and R.M. Walker, J. Appl. Phys. 36 (1965) 3645.
II. ELECTRONIC ENERGY RELAXATION
132
K. Schwariz/Nucl.
Instr. and Meth. in Phys. Res. B 107 (1996) 128-132
[12] N. Seifett, S. Vijayalakshmi, Q. Yan, A. Barnes, R. AI-
bridge, H. Ye. and W. Husinsky, Radiat. Eff. and Def. in Solids 128 (1994) 15. 1131 M. Klinger, Ch.B. Lushchik, T.V. Mashovets, GA. Kholodar, M.K. Sheikman and M.A. Elango, Sov. Phys. Usp. 28 (1985) 994. [ 141 N. Itoh and K. Tanimura, J. Phys. Chem. Solids 51 (1990) 717. [ 151 F. Seitz and J.S. Koehler, Solid State Physics, vol. 2, eds. F. Seitz and D. Thumbull (Academic Press, New York, 1956) p. 307. [ 161 K. Schwartz and Yu. Ekmanis, Radiation Damage and Radiation Sensitivity of Dielectric Materials (Zinatne, Riga, 1989). in Russian.
[I71 L. Hobbs, Nucl. Instr. and Meth. B 91 (1994) 30. [I81 A. Perez, E. Balanzat and J. Dural, Phys. Rev. B 41 (1990) 3943. [I91 R.T. Williams and K.S. Song, J. Phys. Chem. Solids 51 (1990) 679. [20] D.A. Young, Nature 183 (1958) 375. 1211A. Sigrist and R. Balzer, Helv. Phys. Acta 50 (1977) 49. 1221 A.E. Hughes and S.C. Jain, Adv. Phys. 28 (1979) 717. [23] P.D. Townsend, Rep. Prog. Phys. 50 (1987) 501. [24] A.T. Davidson, J.D. Comins, T.E. Deny and F.S. Khumalo, Radiat. Eff. 98 (1986) 305. [25] L. Protin, E. Balanzat, S. Bouffard, A. Cassini, E. Doorhyee, J.L. Doualan, C. Dufour, J.P. Grandin, J. Margetie, E. Pannier and M. Toulemonde, Radiat. Eff., in press (1995).
Nuclear Instruments and Methodsin Physics Research B 107 ( 19%) 128- 132 __ __
Lk!IOM
B
Beam Interactions with Materials&Atoms
!I!!!
EISEVIER
Electronic excitations and defect creation in LiF crystals K. Schwartz Gesellschafrfir
Schwerionenforschung, Postfach 110552, D-64220 Darmstadt, Germany
Abstract A review of ion induced radiation damage processes in alkali halides is presented with special attention paid to Frenkel
defect creation by the decay of excitons. The exciton mechanism is analysed for radiation damage creation in LiF crystals under heavy ion (*@Bi (136.7 MeV/u) and 238U (11.4 MeV/u)) irradiation at fluences up to 5 X lo9 ions/cm*. In both cases F-and F,-centers are the dominant electronic defects in the track. Uranium tracks in LiF crystals can be etched, whereas after irradiation with Bi ions no track etching was observed.
1. Introduction
Under swift heavy ion irradiation with high electronic energy loss (d E/dx), in various solids, new radiation effects and processes have been observed [I -61. At such high excitation levels, collective electronic interaction processes are usually dominant. By strong electron-phonon coupling they lead to local heating (thermal spikes) with subsequent phase transitions, defect creation, and defect aggregation in the excited region of the heavy ion track [7,8]. However, in order to understand the formation of collective defects the creation of single point defects has to be taken into account. For such studies LiF and other alkali halides are privileged subjects due to their well-understood electronic structure, electron-phonon interaction, and defect creation processes under conventional irradiation (such as X-rays or electrons with an energy of - 1 MeV) [4,13,141. A review of radiation induced damage processes in alkali halides is presented. The creation of color centers is analysed and latent tracks of 238U ions (11.4 MeV/u) and 2WBi ions (136.7 MeV/u) are etched in LiF crystals.
2. Radiation damage processes in dielectrics Stimulated by the development of nuclear reactors, studies of radiation damage in solids began in the fifties [9]. At that time Varley proposed a mechanism for defect creation in ionic crystals by multiple ionization of anions followed by displacement from the regular lattice due to Coulomb interaction [lo]. Based on this mechanism, Fleischer et al. [l l] developed the Coulomb explosion model for defect formation in heavy ion tracks. However, further
investigations showed that the Varley mechanism in ionic crystals occurs only at the surface because the lifetime of multi-ionized anion states in the bulk is much shorter than the defect creation time determined by the lattice vibration period (- lo- r3 s; see Ref. [16]). In the sixties Frenkel defect creation by the decay of excited electronic states of the lattice (excitons, electrons and holes) was studied in detail [13,14]. Under conventional irradiation (e.g., X-rays, 1 MeV electrons) this mechanism is possible only in those dielectrics where the energy of the lowest exciton state E,,, exceeds the energy Ed needed to create a Frenkel pair. This takes place in alkali halides, but not in oxide materials (such as MgO, Al,Os and Y3Fe,0,,, and many other dielectrics) where under conventional irradiation, defects can be created only by direct electron-atom collisions as in metals 1151. The creation of defects in solids strongly depends on the irradiation temperature [12-15,23,24]. In some alkali halides at low temperature (7’~ 10-100 K) the efficiency of defect creation by the decay of excitons is extremely low, i.e. the defect creation energy is high [14]. At higher temperatures CT> lo-100 K) the efficiency of primary defect creation increases (F-H Frenkel pairs) as well as the mobility of self-trapped holes which strongly influence color center transformation and the probability of electron-hole recombination. Lithium colloid formation in LiF crystals is a complex process which depends on the irradiation dose, dose rate and irradiation temperature [ 12,16,22-241. Various lithium colloids with different absorption spectra and thermal stability are observed in irradiated LiF - from quasi-microscopic X-centers to large particles with a size of several micrometers [12,22,23]. A possible mechanism for the formation of lithium colloid centers is the aggregation of F-centers which usually takes place at higher temperatures
0168-583X/96/$15.00 0 1996 Elsevier Science B.V. All rights reserved SSDI 0168-583X(95)00846-2
K. Schwartz/Nucl.
Insrr. and Meth. in Phys. Res. B 107 (1996) 128-132
(where the diffusion of F-centers is strong) and at high irradiation doses [ 16,22,23]. Therefore, in LiF crystals at high irradiation doses strong radiolysis takes place with local phase transitions in the cation (metal colloids) and anion sub-lattice (molecular anion products in the bulk and at the surface) [ 12,161. However, the ionic binding determines that no complete amorphization of the lattice takes place [ 171. Perez et al. [18] and Balanzat et al. [4] demonstrated that defect creation in LiF, NaCl, and KBr is also dominated by the exciton mechanism under heavy ion irradiation (up to lz9Xe (27 MeV/u)). At low temperatures (T - 15 K) the energy needed to create an F-center under heavy ion irradiation is approximately the same as for X-ray excitation [4,14]. Nevertheless, at higher (dE/d x), some non-linearity in F,-center creation was observed [4]. Microspectrophotometric measurements indicated that Fand F,-centers are the dominant electronic color centers in the track. The presence of Li colloids in heavy ion irradiated LiF crystals is an open question because of the small oscillator strength of the colloidal centers and the spectral overlapping of their absorption band with F-and F,-absorption [ 12,16,18,22,24]. The observed chemical etching of U fission products in LiF crystals by Young [20] can possibly be explained by the presence of Li colloids. The energy
129
loss for uranium fission products ((d E/d x), _ 1.6 keV/A) is larger than for Xe ions used in the experiment of Perez et al. ((d E/d x), - 1 keV/A) [ 181. At the higher value of (d E/d x)= the aggregation process could be more efficient. An open question of radiation damage creation is the superposition of various excitations in the solid. Ion induced defect creation is a complex process which occurs on a time scale (TV)much larger than the energy transfer time 7;r from the heavy ion to the electronic and vibronic systems of the solid (TV,- IO- ” s) [7]. In alkali halides F-center creation occurs within 7F lo-*’ s [l4] and the efficiency of F-center creation increases with temperature. Thus the energy of F-center creation (i.e., the primary electronic radiation products) can be used to estimate the radiation damage efficiency in the solid. It is a general rule that only a small part of the absorbed radiation energy in the solid is transformed into defects (only a few percent or less). The strong electronphonon interaction in most solids leads to efficient conversion of the absorbed energy into lattice vibrations (the radiation induced heating exceeds 90% of the absorbed energy). Therefore, radiation induced heating in solids is a general phenomenon which was successfully demonstrated in the thermal spike model by Toulemonde et al. [7,8]. However, the thermal spike model usually describes
Energy conversion of heavy ions in alkali halides ion
primary electrons
time scale
10-l’ _ 10-16 s
energy transfer +
excitation of the
Fig. 1. Energy conversion and defect creation in alkali halide crystals under heavy ion irradiation: e“, ey, free and self-trappedexcitons; c-, e+, free electrons and holes; e: , self-trapped holes; o~~,.~, wac,i. qi, optical, acoustical and local (point defect) vibration modes; AT, haiiation induced temperature increase; Tc,i,,critical temperature of phase transition; N, Nd, defect concentration and the critical defect concentration for phase transition.
II. ELECTRONIC ENERGY RELAXATION
130
K. Schwartz/Nucl. Instr. and Meth. in Phys. Res. B 107 (1996) 128-132
heavy ion track damage formation only as a result of local melting, not taking into account the possibility of a superposition of the local heating with the excited electronic states of the lattice. Due to the long’ lifetime of self-trapped excitons in alkali halides (for the triplet states several milliseconds and more [19]) a superposition of the exciton defect creation process with radiation induced heating is possible (Fig. 1). Radiation induced local heating (AT) can influence the efficiency of Frenkel pair creation as well as the aggregation of single defects, e.g. the formation of Li colloid centers, or the creation of molecular F2 (a phase transition is also possible at T < T,,, if the defect concentration Nd exceeds NC,,; Fig. 1) [13,16]. The superposition of the heavy ion induced electronic and thermal excitations can explain the influence of (dE/dx), on defect creation in solids under heavy ion irradiation. However, the understanding of such a superposition needs more detailed experimental and theoretical study (see the review [25]).
3. Experimental results on color center creation and chemical etching Chemical etching is widely used to reveal latent tracks [11,20]. This technique dissolves the ion track in the solid in a suitable solution, but usually it does not characterize the nature of the defects in the track. However, chemical etching is a very simple and sensitive method which gives evidence of the presence of latent tracks in the solid. Young showed that etched tracks of U fission fragments in LiF crystals were different from etched pits due to dislocation [20]. Therefore, it should be possible to correlate etching properties and the structure of color centers in ion tracks.
LiF crystals (10 X 10 X 2 mm3) were irradiated with 209Bi (136.7 MeV/u; fluence IO9 ions/cm2) and 238U ions (11.4 MeV/u; 106-5 X lo9 ions/cm2) at GSI, Darmstadt. Samples for chemical etching were irradiated through a mask (Fig. 2a). For irradiated samples with a fluence 2 lo9 ions/cm2 absorption spectra in the range 200-700 nm were measured. To examine the dislocation structure and the existence of ion tracks all crystals were etched in a solution of HF (1 vol. part) and acetic acid (1 vol. part) saturated with FeF, before and after irradiation. This etchant was used earlier by Young to etch U fission tracks in LiF crystals [20]. The etching time was in the range from 10 to 30 min. The etched pits were observed by an optical and scanning electron microscope. Before inadiation all samples showed a dislocation density of several lo5 disl/cm2, which is typical for alkali halides grown from a melt. Due to the mask, irradiated and non-irradiated areas could easily be distinguished and allowed comparison of the etched pits of dislocations (usually larger) and the etched pits of ion latent tracks. Each
Fig. 2. Etched tracks in an irradiated LiF crystal (U ions, 11.4 MeV/u; fluence lo7 ions/cm’) after 30 min etching observed using a scanning electron microscope. The irradiation was performed through a mask. The density of etched pits is 10’ f 10% tracks/cm*. (a) Etched pits of U ions in the irmdiated area (upper part) and etched pits of dislocations (bottom) in the non-irradiated area (magnification: 200). (b) Etched pits in the irradiated area; the large etched pits are due to dislocations. The concentration of dislocations of about lo5 cm-* IS . not changed by ion irradiation (magnification: 800).
latent track of U ions was etched to a prismatic pit with a size from 1 to 3 p,rn (Fig. 2). In the case of Bi ions (136.7 MeV/u) no etched tracks were observed on the surface of irradiated LiF and the dislocation structure of the crystal was not changed. The absorption spectrum was measured only for crystals irmdiated with fluences above lo9 ions/cm2. In the absorption spectrum F-and F,-centers were dominant. However, for Bi ions the concentration of F-centers was two orders of magnitude smaller than in crystals irradiated with U ions. The concentration of F-centers NF was determined using the absorption coefficient pa, NF [cme3] = 9.48 X lOI& [cm-‘], according to the SmakulaDexter formula [ 181.The concentration of F-centers in LiF crystals irradiated with U at a fluence of 2 X IO9 ions/cm2 was N = 4 X 1017cm-3 which corresponds to an F-center creatioi energy A EF = h)Eabs/NF. track= 1300 + 150 eV.
K. Schwarrz/Nucl.
4. Discussion
Instr. and Meth. in Phys. Res. B 107 (1996) 128-132
and conclusion
After the etching of U fission tracks in LiF crystals by Young [20], there were no further investigations in this field. Our results demonstrate that U ions (11.4 MeV/u) with a specific energy loss (d E/dx), = 2.86 keV/i induce etchable tracks in LiF crystals, while irradiation with Bi ions (136.7 MeV/u) with (d E/d x& = 0.8 keV/A does not induce etchable tracks. Obviously, the value of (dE/dx) for Bi ions is below the threshold for track etching. Taking into account the results of Young ((d E/d x), N 1.6 keV/& the critical value for latent track etching in LiF is in the range 0.8 keV/W < (d E/d x), crit < 1.6 keV/W. However, an open question is the origin’of latent track etching in LiF crystals, i.e. what kind of defects are responsible for selective etching. Chemical etching of point defects in LiF or other alkali halides has never been observed after irradiation with electrons, X-rays, protons or o-particles [20,21]. Therefore, single point defects (F- and V-centers) and microscopic aggregates (Fz-, Fs-centers) are probably not sufficient. It is assumed that larger aggregates of F-centers, e.g. Li colloids, are responsible for the track etching. Such centers are created at high radiation dose and at irradiation temperatures exceeding the critical temperature for Li colloid formation, Tctit2 500 K. However, it is very difficult to observe these colloid centers by optical spectroscopy, since the oscillator strength of colloidal centers is much smaller than in the case of F-centers (usually the concentration of colloidal centers is low and an overlapping of the absorption spectrum with F-and F,-centers is possible). Perez et al. studied color center creation in heavy ion tracks in LiF up to Xe (27 MeV/u) with an energy loss of (dE/dx), N 1 keV/A, which is close to the situation of Bi (136.7 MeV/u, d E/d x), = 0.8 keV/.&, i.e. below the critical value for etching. For a more detailed analysis of the etching threshold and the origin of etching in LiF crystals (i.e., in order to determine which defects are
131
responsible for the etching) further experiments with various ions are necessary. Additional information on aggregate and colloid center creation can be obtained from the analysis of the energy needed A E, to create an F-center. The observed threshold value of (dE/d x), for track etching can be correlated with the dependence of A E, on the excitation density, i.e. the magnitude of (dE/dx). Under similar irradiation conditions the energy A E, must increase when strong aggregation takes place. Therefore, we analysed the F-center creation energy A E, for Ne, Ar, and Xe ions according to data of Ref. [ 181 and for U ions according to our results (under the same irradiation conditions of _ 10’ ions/cm’). The value of AE, increases with increasing energy loss (Fig. 3). This can be explained as an aggregation effect of F-centers to nF-centers and finally to Li colloids, or as a thermal annealing effect of F-centers. However, the value of A E, for Ne and Ar ions is smaller than A E, for X-ray excitation [ 141, which probably suggests that thermal annealing can be excluded. Therefore, the increase of A E, for U ions (Fig. 3) can be explained from the viewpoint of aggregate and colloidal center creation which determines the chemical etching. However, without additional experimental analysis of the track morphology this is only an assumption.
Acknowledgements
The author is grateful to Prof. A. Scharmann (University of Giessen) for the LiF crystals and to Dr. C. Trautmann (GSI) for many fruitful discussions and assistance in the experiment.
References
[II E. Balanzat, Radiat. Eff. and Def. in Solids 126 (1993) 97. Dl Ming-dong Hou, S. Kiaumlinzer and G. Schumacher, Phys. Rev. B 41 (1990) 1144.
[31 A. Dunlop, D. Lesueur, P. Legrand and H. Danmak, Nucl. Instr. and Meth. B 90 (1994) 330.
[41 E. Balanzat, S. Bouffard, A. Cassimi, E. Doorhyee, L. Protin,
4001 2
3
4
5
6
7
8
9
10
In @%I@ Fig. 3. F-center creation energy under irradiation with various heavy ions: Ne (40 MeV/u), Ar (60 MeV/u), Xe (27 MeV/u) after Ref. [ 181; U (11.4 MeV/u) - our results. For all measurements the fluence was in the range of a few 10’ ions/cm’.
J. Grandin, J.L. Doualan and J. Margerie, Nucl. Instr. and Meth. B 91 (1994) 134. [51 C. Trautmann, R. Spohr and M. Toulemonde, Nucl. Instr. and Me&t. B 83 (1993) 513. [61 E. Balanzat, S. Bouffard, A. Le Mabel and N. Betz, Nucl. Ins&. and Meth. B 91 (1994) 140. [71 M. Toulemonde, S. Bouffard and F. Studer, Nucl. Instr. and Meth. B 91 (1994) 108. [81 Z.G. Wang, Ch. Dufour, E. Paumier and M. Toulemonde, J. Phys. Condensed Matter 6 (1994) 6733. [91 F. Seitz and E.P. Wigner, Sci. Am. 8 (1956) 76. [lOI J.H.O. Varley, Nature 174 (1954) 886. [Ill R.F. Fleischer, P.B. Price and R.M. Walker, J. Appl. Phys. 36 (1965) 3645.
II. ELECTRONIC ENERGY RELAXATION
132
K. Schwariz/Nucl.
Instr. and Meth. in Phys. Res. B 107 (1996) 128-132
[12] N. Seifett, S. Vijayalakshmi, Q. Yan, A. Barnes, R. AI-
bridge, H. Ye. and W. Husinsky, Radiat. Eff. and Def. in Solids 128 (1994) 15. 1131 M. Klinger, Ch.B. Lushchik, T.V. Mashovets, GA. Kholodar, M.K. Sheikman and M.A. Elango, Sov. Phys. Usp. 28 (1985) 994. [ 141 N. Itoh and K. Tanimura, J. Phys. Chem. Solids 51 (1990) 717. [ 151 F. Seitz and J.S. Koehler, Solid State Physics, vol. 2, eds. F. Seitz and D. Thumbull (Academic Press, New York, 1956) p. 307. [ 161 K. Schwartz and Yu. Ekmanis, Radiation Damage and Radiation Sensitivity of Dielectric Materials (Zinatne, Riga, 1989). in Russian.
[I71 L. Hobbs, Nucl. Instr. and Meth. B 91 (1994) 30. [I81 A. Perez, E. Balanzat and J. Dural, Phys. Rev. B 41 (1990) 3943. [I91 R.T. Williams and K.S. Song, J. Phys. Chem. Solids 51 (1990) 679. [20] D.A. Young, Nature 183 (1958) 375. 1211A. Sigrist and R. Balzer, Helv. Phys. Acta 50 (1977) 49. 1221 A.E. Hughes and S.C. Jain, Adv. Phys. 28 (1979) 717. [23] P.D. Townsend, Rep. Prog. Phys. 50 (1987) 501. [24] A.T. Davidson, J.D. Comins, T.E. Deny and F.S. Khumalo, Radiat. Eff. 98 (1986) 305. [25] L. Protin, E. Balanzat, S. Bouffard, A. Cassini, E. Doorhyee, J.L. Doualan, C. Dufour, J.P. Grandin, J. Margetie, E. Pannier and M. Toulemonde, Radiat. Eff., in press (1995).