Growth of Bi2Sr2Cu1Ox films by laser ablation

Growth of Bi2Sr2Cu1Ox films by laser ablation

Physica C 311 Ž1999. 231–238 Growth of Bi 2 Sr2 Cu 1O x films by laser ablation J. Perriere ` a a,) , R.M. Defourneau a , A. Laurent a , M. Morcret...

175KB Sizes 3 Downloads 47 Views

Physica C 311 Ž1999. 231–238

Growth of Bi 2 Sr2 Cu 1O x films by laser ablation J. Perriere ` a


, R.M. Defourneau a , A. Laurent a , M. Morcrette a , W. Seiler


Groupe de Physique des Solides, CNRS UMR 7588, UniÕersites ´ Paris VII et Paris VI, Tour 23, 2, Place Jussieu, 75251 Paris Cedex 05, France b Laboratoire Microstructure et Mecanique des Materiaux, Ecole Nationale Superieure des Arts et Metiers, 151 BouleÕard de l’Hopital, ´ ´ ´ ´ ˆ 75013 Paris, France Received 4 August 1998; revised 21 October 1998; accepted 22 October 1998

Abstract The formation and the properties of thin BiSrCuO films grown on MgO single crystal substrates by the pulsed laser ablation of a Bi 2 Sr2 Cu 1O6 target have been studied. The precise influence of the oxygen pressure and substrate temperature on the atomic composition, nature and structure of the grown phases, crystalline quality and superconductivity of the films has been analyzed. In the 600–7508C temperature range and 0.05–0.5 mbar oxygen pressure range, highly textured films of the 2201 phase ŽBi 2 Sr2 Cu 1O6 phase. were formed despite large composition deviations ŽBi enrichment and Sr depletion. with respect to the ideal composition. A high crystalline quality was evidenced for films grown at low and intermediate pressure and high temperature. xmin values in the films deduced from channeling experiments were similar to the case of single crystal material. Despite this high crystalline quality, incomplete or very low Tc superconducting transitions were observed in the resistivity measurements. q 1999 Published by Elsevier Science B.V. All rights reserved. PACS: 81.15.Fg; 68.55.N; 74.76.Bz Keywords: Thin films; Laser ablation; Crystalline phase; Superconductivity

1. Introduction In the high Tc Bi 2 Sr2 Ca ny1Cu nO y family, the three most stable phases with n s 1 Ž2201 phase., 2 Ž2212 phase. and 3 Ž2223 phase. have only slightly different free energies of formation between each other. Moreover, these three distinct phases which present a high anisotropic two-dimensional character, are practically only structurally distinguishable by the number of CuO planes per unit cell. As a result, their c-axis length is different, but they do not )

Corresponding author. Tel.: q33-1-44-27-61-29; Fax: q331-43-54-28-78; E-mail: [email protected]

present significant differences in the a–b plane, i.e., a similar lattice parameter roughly equal to 0.54 nm is observed for the these phases. Thus, it is difficult to grow single phase films of the Bi compound, the phase intergrowth being generally observed w1–4x: each elemental cell in the c direction will grow independantly of the nature and composition of the preceding cell, leading to the random stacking along the c direction of 2212, 2201 and 2223 elemental cells. The dependance of the phase formation in in situ grown Bi oxide films prepared by laser ablation has been studied as a function of substrate temperature and oxygen pressure w5x. However, in this work, the

0921-4534r99r$ - see front matter q 1999 Published by Elsevier Science B.V. All rights reserved. PII: S 0 9 2 1 - 4 5 3 4 Ž 9 8 . 0 0 6 2 8 - 5


J. Perriere ` et al.r Physica C 311 (1999) 231–238

use of a BiPbSrCaCuO target leads to a rather complex situation in the oxygen pressure–temperature Ž P–T . diagram, which may be due to the variation in cationic composition in the films. Moreover, both 2223 and 2201 intergrowth defects have been observed in thin films with the ideal 2212 composition w6x. This means that even without any composition deviations, 2212 films can contain intergrowth defects. In that sense, the 2201 phase appears as a more favorable case, according to the fact that it is not possible a priori to have both intergrowth of 2212 and 2223 defects and a perfect 2201 composition in the films. As a result, it seems that the 2201 phase can be obtained as a single phase in thin films w7x, and therefore this phase could be a good subject to investigate the precise effects of the growth conditions on the structural and physical properties of the films. In this paper, we report on the study of the influence of oxygen pressure and substrate temperature on the phase formation, structure, crystalline quality and transport properties in the BiSrCuO films in situ grown by the laser ablation method. A Bi 2 Sr2 Cu 1O6 ceramic target was used in this work in order to grow pure 2201 phase films.

2. Experimental The BiSrCuO thin films were grown by the pulsed laser ablation method using a frequency quadrupled Ž266 nm. Nd:YAG laser ŽB.M. Industries.. A Bi 2 Sr2 Cu 1O6 target was thus irradiated by 200 mJ pulses of 7 ns duration at 1–5 Hz repetition rate. The laser beam was focused by a quartz lens onto the rotating bulk target at a 458 oblique incidence, to give power density around 100 MWrcm2 . The various species emitted by the BiSrCuO target during the laser irradiations were deposited onto MgO Ž100. single crystal substrates located at a 4 cm distance. During deposition, the substrate was heated Žin the 500–7508C range., and the in situ growth of the films occurs under partial oxygen atmosphere Žup to 1 mbar.. After deposition the films were cooled down Žin 2 h. under the oxygen pressure used for the deposition. Rutherford backscattering spectrometry ŽRBS. was used to determine the film composition and

thickness, with 4 Heq ion beams in the MeV energy range. The precise depth distribution of the various elements was obtained by the use of the RUMP simulation program w8x. Moreover, in the special geometry of channeling this RBS technique was also used to characterize the crystalline quality of the in situ grown films, by the measurements of the ratio of aligned to random contribution in the RBS spectra, i.e., the minimum yield xmin w9,10x. X-ray diffraction ŽXRD. in the Bragg–Brentano mode Žwith a Cu K a radiation. was used to study the crystalline structure of the films, to determine the nature of the phase and to measure the lattice parameters. The measurements of rocking curves were also carried out to determine the mosaic spread normal to the substrate surface. The resistivity of the films as a function of the temperature was determined by the standard dc four probe measurements.

3. Results and interpretation The variations of the BirSr composition ratio in the films Ždeduced from RBS measurements., as a function of the oxygen pressure and substrate temperature during laser ablation are respectively presented in Figs. 1 and 2. The plotted values correspond to the central part of the films, i.e., to the maximum of film thickness. Lateral nonuniformities in composition were not studied in this work. Figs. 1 and 2 evidence strong composition deviations with

Fig. 1. BirSr concentration ratio in the laser deposited films as a function of the oxygen pressure, for a fixed substrate temperature Ž7208C..

J. Perriere ` et al.r Physica C 311 (1999) 231–238

Fig. 2. BirSr concentration ratio in the laser deposited films as a function of the substrate temperature, for a fixed oxygen pressure Ž0.2 mbar..

respect to the ideal 2201 target composition Ži.e., BirSrs 1.. In fact for a fixed substrate temperature Ž7208C., a maximum in the BirSr ratio is observed around 0.1 mbar. At decreasing pressure, a large Bi deficiency is observed, and under vacuum Ž10y6 mbar. Bi free films are deposited, while at higher pressure Žup to 0.5 mbar. a stationary regime is observed with a BirSr ratio of about 1.15. This value is roughly constant with the substrate temperature ŽFig. 2. for a fixed oxygen pressure Ž0.2 mbar., except for the lower and higher temperatures. The BirCu ratio Žnot presented here. also shows a Bi excess ŽBirCus 2.1–2.3., which is practically constant over both the oxygen pressure and temperature range, except for the highest Ž0.5 mbar, 7808C. and lowest Ž0.03, 5008C. growth conditions. It is generally admitted that laser ablation yields films with a composition close to that of the target. This being due to the very high heating rates during laser irradiations Žabout 10 10 Krs.: the target is evaporated without any change in composition. The composition deviations observed for the Bi 2 Sr2 Cu 1O6 target can be explained by the phenomena taking place in the gas phase Žduring the expansion of the plasma., and at the gas phase–film interface. First, in the gas phase, the scattering of the species by the oxygen atmosphere substantially reduce their velocity and broadens their angular distribution, with a broadening function of their mass w11x. Moreover, the gas phase–film surface plays also an important role via the sticking coefficient. This sticking coeffi-


cient, which depends upon thermodynamical conditions and kinetic energy of the species reaching the substrate, can be different for Bi, Sr and Cu leading to large composition deviations in the grown layers. At the lower pressures, the Bi losses seem mainly be due to the low sticking coefficient of Bi species. For these low pressures and at low temperatures Ž- 3008C. such composition deviations are not observed, but at T ) 5008C Bi species are easily reevaporated from the substrate, leading to a large Bi deficit in the films. In the high pressure domain ŽFig. 1., all the emitted species are thermalized by collisions with oxygen molecules and the BirSr Žor BirCu. ratio can be considered as a measure of the ratio of the sticking coefficient of the Bi and Sr Žor Cu. species. In the intermediate region Žaround 0.1 mbar. the rapid variation in BirSr may be related to the plasma expansion. Actually, it was found w12x that the expansion velocity of the various species ejected from a polycationic oxide target remains roughly constant from vacuum up to an oxygen pressure of 0.1 mbar, and decreases rapidly beyond this pressure. So, 0.1 mbar appears as a threshold between a diffusive and a collisionnal plasma expansion. In this last regime, the broadening of the angular distribution of the species will be function of their mass w11x: the lighter elements having the larger broadening. According to their mass, a Bi enrichment and a Sr depletion in the central part of the film can be expected. Despite these composition deviations, the crystalline 2201 phase was found to be formed by laser ablation. The in situ crystallisation of the films begins around 5008C, but the peaks in the diffraction diagram cannot be ascribed to the 2201 phase alone, other oxide phases being present and a large part of the film being amorphous. Then, for higher temperature, the crystallisation of the Bi compound occurs in the domain where it is thermodynamically stable, i.e., in the domain defined by the stability line observed in the oxygen pressure vs. temperature Ž P, T . diagram showing the decomposition conditions of this compound previously described w13x. In this domain, the phases depend upon the conditions. For T ) 6808C, whatever the oxygen pressure higher than 0.05 mbar, the pure 2201 phase is grown as shown in Fig. 3 by typical XRD patterns for a film grown at 7208C and 0.2 mbar oxygen. In this diagram, all the


J. Perriere ` et al.r Physica C 311 (1999) 231–238

Fig. 3. Typical XRD u –2 u patterns in a log scale, of a BiSrCuO film deposited on Ž001. MgO under 0.1 mbar oxygen pressure at 7208C.

peaks can be identified as the Ž001. diffraction lines characteristic of the 2201 orthorhombic phase. Moreover, the peak positions indicate the absence of any effects Žslight shifts of the peaks. which could be associated with an intergrowth effect: pure 2201 films are grown. This means that the composition deviations are not accompanied by the phase intergrowth observed in the laser ablation of a 2212 target w1,4x. In contrary to the case of bulk material, whatever the growth conditions, we never observed in the films the insulating phase with a monoclinic cell and composition close to Bi 2 Sr2 Cu 1O y w14–16x which is often present in the synthesis of the Bi 2201. In the Bi oxide compound, the 2201 phase is known to form a solid solution with a rather wide range in cationic composition w14–16x. For 2201 single crystals, a Sr deficiency was observed, compensated by an excess of Bi with a significant partial occupation of Sr sites by Bi atoms w15x. Figs. 1 and 2 show that except the specific case of the very low pressure, all the growth conditions lead to BirSr ratio higher than unity, i.e., a ratio favorable to the synthesis of the 2201 phase. Therefore it is not surprising that the XRD patterns indicate the presence of the 2201 phase despite severe composition deviations in some cases. Fig. 3 indicates the growth of highly textured films, with their c-axis normal to the substrate surface. The lattice parameter along the c-axis was thus

found around 2.46 nm. This measured d values show a rather large scattering, between 2.45 and 2.465 nm. This scattering may be due to the fact that the c-axis parameter is a function of the BirSr ratio and of the oxygen content or distribution in the cell. As Fig. 1 shows that similar BirSr ratios are observed for different oxygen pressures Žand thus oxygen content., differences in c-axis parameters are expected. Any way, despite this scattering, the c-axis parameter shows a slight contraction Ž- 0.01 nm. with increasing BirSr values. This slight contraction being due to the replacing of divalent Sr 2q by smaller trivalent Bi 3q cations in the Sr sites w16x. These results are comparable to the variation Ž0.025 nm. observed in the case of bulk material for a wider range in BirSr ratio, from 1 to 1.5 w14x. For T - 6808C, a region exists in the Ž P, T . diagram, very close to the thermodynamical stability line w13x, in which the presence of both 2212 and 2201 phases can be observed in the films. As a matter of fact, Fig. 4 represents the XRD patterns for a film grown at 6608C and 0.03 mbar. First, a large broadening of the peaks is observed as compared to Fig. 3, this broadening being characteristic of a decrease in the crystalline quality of the films. The diffraction peaks in Fig. 4 are not only due to the 2201 phase, some other peaks are present which may be characteristic of the 2212 phase. Actually, the presence of the 2212 phase was first evidenced in the

Fig. 4. XRD u –2 u patterns of a BiSrCuO film deposited on Ž001. MgO under 0.03 mbar oxygen pressure at 6608C.

J. Perriere ` et al.r Physica C 311 (1999) 231–238

Fig. 5. Random Žy. and aligned Ž`. RBS spectra for a BiSrCuO thin film grown under 0.1 mbar oxygen pressure at 7208C.

resistivity measurements through its superconductivity behaviour Žsee below.. In Fig. 4, the peaks not characteristic of the 2201 phase, do not correspond exactly to those of the 2212 phase. This may be due to the intergrowth process, i.e., some crystallites in the film contain stacking faults. The formation of the 2212 phase in bulk material without any Ca, has only been reported in a brief report w17x. The growth of crystallites containing the 2212 phase in the films grown by laser ablation of a 2201 target Žwithout Ca. is explained by the fact that in the 2212 structure, Sr atoms can be easily substituted to Ca atoms Žup to 40% of the Ca content. without large structural modifications, except a slight change in the interplanar distances w6x. For the specific conditions Žlow temperature and oxygen pressure. leading to the synthesis of multiphased Ž2201 and 2212. films as presented in Fig. 5, the composition ŽBi 1.05 Sr1.55Cu 1O. indicates a large enrichment in Sr and Cu with respect to the target composition. Such BirSr


Ž- 0.7. and BirCu Žf 1. ratio can favored the formation of a phase structurally similar to the 2212 one, but without any Ca. The crystalline quality of the 2201 films was studied using RBS in channeling geometry w9,10x. Fig. 5 represents typical aligned and random backscattering spectra recorded on a pure 2201 phase film. From these spectra we deduce the xmin value for each element. In the best case the following values were measured xminŽBi. s 10%, xminŽSr. s 6% and xminŽCu. s 5%, while a xminŽO. of about 50% was estimated. Channeling experiments in 2201 thin films or single crystals have not been reported until now. Thus we can just compare the best xmin value for Bi in this work Ž10%., with that reported for the Bi 2 Sr2 Ca 1Cu 2 O x compound. For thin films, the best results for the growth by laser ablation of 2212 films on MgO is a xmin value around 25% w18x. Our results are significantly better, and this could be due to the formation of pure 2201 phase in the films, without any phase intergrowth Žas observed for the 2212 films.. For single crystal xmin values for Bi in the 7–10% range have been reported w9x, and a very good agreement is observed with our measurements in thin films. In this aspect our 2201 films behave as single crystal. The variation of the xmin values was studied as a function of oxygen pressure during laser ablation. Table 1 reports the best xmin values for each element, and 0.1 mbar appears as the optimum condition for the crystalline quality of the films. Moreover, rocking curve measurements were also performed to evaluate the misalignment of the c-axis of the textured 2201 films with the normal to the substrate. The FWHM values of the rocking curves obtained from the 2201 Ž008. reflections are thus presented in Table 1. An optimum in the oxygen pressure Žaround 0.1 mbar. is also observed leading to a FWHM value close to the instrumental resolu-

Table 1 xmin values and FWHM of the rocking curves for films grown at various pressures Oxygen pressure wmbarx

xmin ŽBi. w%x

xmin ŽSr. w%x

xmin ŽCu. w%x

FWHM w8x

0.05 0.1 0.2 0.4

25 10 12 20

17 6 8 14

15 5 6 12

0.41 0.09 0.09 0.16


J. Perriere ` et al.r Physica C 311 (1999) 231–238

Fig. 6. Aligned backscattering spectrum of a BiSrCuO film grown under 0.15 mbar oxygen pressure at 7208C. The hatched areas indicate the surface peak yields for the Bi, Sr and Cu contributions.

tion Ž0.068.. Let us notice that rocking curve measurements give information on the orientation of the 2201 crystallites, while channeling experiments check the presence of point or extended defects in the crystallites. The best results for both rocking curves and channeling experiments being obtained in the same growth conditions, this indicates that the kinetic energy of the incident species plays a major role in the crystalline quality of the 2201 films. Table 1 also shows that differences exist between the xmin values of the various constituents: xminŽBi. ) xminŽSr. f xminŽCu.. Such differences which do not change with the growth conditions have to be related to some specific properties of the Bi films. First the high xminŽBi. value can be due to the excess of Bi in the films, which leads to a partial occupation of the Sr sites by Bi which can change the potential along the atomic rows for the channeled He ions. Moreover, this can be also related to the variation in atomic location associated with the modulation. In fact, the large Žabout 0.03 nm. static atomic displacement due to the incommensurate modulation induced dechanneling effects in the Bi compound w9x, and this effect can be enhanced for the Bi species as their static displacements are important in the Bi plane. In channeling experiments, the analysis of the surface peak areas as observed for the aligned spec-

tra provides interesting results about the disorder in the near surface region of the films w19x. In particular, in the case of Fig. 6, the Bi, Sr and Cu elements are well separated in energy, and the surface peak can be analyzed for each element. As a result, we always found a Bi and Sr enrichment in the surface peak with respect to the film composition. For example, in the case of Fig. 6, the cationic compositions Bi 2.18 Sr1.8 Cu 1 and Bi 3.5 Sr2.28 Cu 1 were respectively measured for the film and the surface peak. A change in the film composition in the near surface region could be assumed to explain this result. However, thin films with a thickness equal to that of the surface peak Žabout 20 = 10 15 at.rcm 2 . did not show such a Bi and Sr enrichment. Therefore, this result characterizes a different behaviour of the Bi, Sr and Cu atoms in the surface disordered region. In fact, such an analysis provides a measure of the number of atoms per row exposed to the incident ion beam on the surface w19x. The analyses of the surface peaks for films grown under various conditions lead to an average value of about 2 Bi atoms per row, 0.9 Sr atoms per row and 0.5 atoms per row. Due to the fact that the Bi compound growth occurs half unit cell by half unit cell w20x, the surface of the film is a Bi plane, while the Sr and Cu planes are underneath. Therefore the respective numbers of Bi, Sr and Cu atoms per row in the surface peak can be explained by the presence of a disordered half unit cell of the compound in the surface of the films.

Fig. 7. Resistivity as a function of temperature for various laser deposited BiSrCuO thin films grown at 7208C under various oxygen pressures: 0.3 mbar Ža., 0.2 mbar Žb., 0.05 mbar Žc. and 0.04 mbar Žd..

J. Perriere ` et al.r Physica C 311 (1999) 231–238

Fig. 8. Resistivity as a function of temperature for two BiSrCuO thin films grown under same conditions Ž7008C and 0.05 mbar., but cooled under vacuum Žcurve a., or under oxygen Žcurve b..

Finally, the superconducting properties of the films were studied via resistivity measurements. Fig. 7 represents some RŽT . curves which show that the films grown at the higher pressures behave as semiconductors, while decreasing pressures lead to films presenting a rather complex behaviour: metallic at high temperature followed by a semiconducting behaviour at low temperature, superconducting transitions being not observed for these growth conditions. A superconducting transition begins to be observed at 0.05 mbar, and the variation of the substrate temperature at this pressure does not lead to large effects on this transition, i.e., Tc higher than 5 K were not observed in this work. Thus, despite the crystalline quality of the 2201 films grown by laser ablation, incomplete or very low temperature superconducting transitions are only observed. It is worth noting that single crystals with a BiSrCuO composition showed a comparable behaviour w15x, i.e., the dependence of resistivity on temperature shows a linear decrease from room temperature down to 100 K, where a semiconducting like behaviour starts to be observed. This means that the formation of the crystalline 2201 phase does not determine alone the transport properties of this material. One important factor for getting high Tc values is the composition of the BiSrCuO compound as shown previously w21x, and the variation in the BirSr ratio observed in this work can be the origin of the transport properties measured in these films. It has been also pointed out that the superconducting transition is strongly influenced by the cooling process


immediately after deposition, meaning that the oxygen content must be optimized very carefully in this material. As a matter of fact superconducting films with Tc up to 18 K have been obtained by RF sputtering of a BiPbSrCuO target with an optimal oxygen content Žoptimal hole concentration. in the films w7x determined by the cooling procedure. Let us note that our work was focused on the influence of growth conditions on crystalline quality of the films, and accordingly we have not tried to systematically study the influence of the cooling procedure on the superconducting properties. However, the influence of this step of film formation can be deduced from Fig. 8 which represents the RŽT . curves for 2201 films grown under similar conditions Ž7008C, 0.05 mbar., the cooling process being different. For the film cooled under oxygen Ž0.05 mbar. a superconducting transition is observed, while a semiconducting behaviour is measured for the film cooled under vacuum. Whatever the precise origin of this effect Žchange in oxygen content or in oxygen distribution in the cell., it leads to the conclusion that it will be possible to vary the Tc value of such single phase 2201 films by additional thermal treatment under reducing or oxidizing atmosphere. The transport properties of films grown at low temperatures Ž650 to 6808C. and low pressures Ž0.02 or 0.03 mbar. were also studied, and a typical RŽT . curve is shown in Fig. 9. Two transitions are observed around 80 K and 20 K Žindicated by the arrows.. The first transition at 80 K can be ascribed to the 2212 crystallites, while the second is certainly

Fig. 9. Resistivity as a function of temperature for a multiphased Ž2201 and 2212. laser deposited BiSrCuO thin film Žcurve a., and for a film grown by laser ablation of a 2212 target Žcurve b..


J. Perriere ` et al.r Physica C 311 (1999) 231–238

related to 2201 crystallites in the films. A very similar behaviour can be observed Žcurve b. for a film grown by laser ablation of a Bi 2 Sr2 Ca 1Cu 2 O 8 target, the cationic composition of the film being: Bi 2.3 Sr1.65 Ca 1Cu 1.6 , and XRD showed the presence of both 2212 and 2201 phases in such a film. These results evidence the fact that crystallites of the 2212 phase can be grown by laser ablation of a pure 2201 target Žwithout Ca., and lead to transport properties characteristic of the presence of this phase.

4. Summary The epitaxial growth of BiSrCuO thin films by pulsed laser deposition has been investigated. The main parameters governing the in situ growth of pure 2201 films are the oxygen pressure and substrate temperature. Whatever these conditions the pure 2201 composition was never observed, a Bi enrichment being always present in the films. Despite these composition deviations, the pure 2201 structural phase was observed for a wide range of experimental growth conditions. By a careful optimisation of these conditions, high crystalline quality 2201 films were grown at 700–7208C around 0.1 mbar oxygen pressure. But as for 2201 single crystalline 2201 material, these films do not exhibit high quality transport properties. A semiconducting behaviour being observed except for films grown at the lower oxygen pressure Ž- 0.04 mbar.. To obtain high crystalline films with good superconducting properties, the oxygen content and distribution in the elemental 2201 cell has to be carefully optimized.

Acknowledgements The authors wish to thank G. Cario for his help. This work was supported by the CNRS ŽGDR no.

86., and the EEC under contract ERBCHRXCT 930325. References w1x L. Ranno, D. Martinez-Garcia, J. Perriere, ` P. Barboux, Phys. Rev. B 48 Ž1993. 13945. w2x H. Ota, K. Sakai, Z. Mori, R. Aoki, Appl. Phys. Lett. 70 Ž1997. 1471. w3x M. Ohkubo, J. Gerek, G. Linker, O. Meyer, Appl. Phys. Lett. 67 Ž1995. 2403. w4x G. Poullain, R. Desfeux, H. Murray, Physica C 214 Ž1993. 195. w5x W.T. Lin, K.C. Wu, J. Appl. Phys. 72 Ž1992. 4812. w6x A. Brazdeikis, A. Vailionis, A.S. Flodstrom, ¨ C. Traeholt, Physica C 253 Ž1995. 383. w7x Z.Z. Li, H. Riffi, A. Vaures, S. Metgert, H. Raffy, Physica C 206 Ž1993. 367. w8x L.R. Doolittle, Nucl. Instrum. Meth. B 9 Ž1985. 344. w9x L.E. Rehn, R.P. Sharma, P.M. Baldo, Y.C. Chang, P.Z. Jiang, Phys. Rev. B 42 Ž1990. 4175. w10x L. Ranno, J. Perriere, ` J.P. Enard, F. Kerherve, ´ A. Laurent, R. Perez-Casero, Solid State Commun. 83 Ž1992. 67. w11x J. Gonzalo, R. Gomez-san Roman, J. Perriere, ` C.N. Afonso, R. Perez-Casero, Appl. Phys. A 66 Ž1998. 487. w12x C. Girault, D. Damiani, C. Champeaux, P. Marchet, J.P. Mercurio, J. Aubreton, A. Catherinot, Appl. Phys. Lett. 56 Ž1990. 1472. w13x L.M. Rubin, T.P. Orlando, J.B. Van der Sande, G. Gorman, R. Savoy, R. Swope, R. Beyers, Appl. Phys. Lett. 61 Ž1992. 1977. w14x R.M. Fleming, S.A. Sunshine, L.F. Schneemeyer, R.B. Van Dover, R.J. Cava, P.M. Marsh, J.V. Waszczak, S.H. Glarum, S.M. Zahurak, F.J. DiSalvo, Physica C 173 Ž1991. 37. w15x H. Leligny, S. Durcok, P. Labbe, M. Delesert, B. Raveau, Acta Crystallogr. B 48 Ž1992. 407. w16x N.R. Khasanova, E.V. Antipov, Physica C 246 Ž1995. 241. w17x H. Adachi, Y. Ichikawa, K. Hirochi, T. Matsushima, K. Setsune, K. Wasa, Jpn. J. Appl. Phys. 29 Ž1990. L81. w18x J. Arnold, A. Pfuch, J. Borck, K. Zach, P. Seidel, Physica C 213 Ž1993. 71. w19x O. Meyer, J. Geerk, Q. Li, G. Linker, X.X. Xi, Nucl. Instrum. Meth. B 45 Ž1990. 483. w20x C. Marechal, E. Lacaze, A. Laurent, J. Perriere, ´ ` W. Seiler, Physica C 294 Ž1998. 23. w21x M. Inoue, H. Matsushita, H. Hayakawa, K. Ohbayashi, Phys. Rev. B 51 Ž1995. 15448.