HAL Id: jpa-00216716
https://hal.archives-ouvertes.fr/jpa-00216716
Submitted on 1 Jan 1976HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
MÖSSBAUER EFFECT STUDIES OF XENON
IMPLANTED IN IRON
J. Odeurs, R. Coussement, J. de Bruyn, H. Pattyn, M. van Rossum
To cite this version:
JOURNAL DE PHYSIQUE Colloque C6, supplkment au no 12, Tome 37, Dicembre 1976, page C6-899
J. ODEURS, R. COUSSEMENT, J. DE BRUYN, H. PATTYN and M. VAN ROSSUM Leuven University, Instituut voor Kern- en Stralingsfysika, Celestijnenlaan 200 D,
3030 Heverlee, Belgium
R6sum6.
-
A l'aide de 1'Effet Mossbauer on a Btudi6 le xenon implante dans le fer, en fonction de la dose et en fonction du courant des ions.La fraction substitutionnelle des atomes de Xe diminue quand la dose est augmentke de
On a implant6 dans deux feuilles a une dose de 5 X 1013 atomesJcm2 tandis que le rapport des courants d'ions Btait de 75. La feuille dans laquelle on a implante avec le courant le plus Bleve donne la fraction substitutionnelle la plus importante.
Afin de dtcrire 1'6volution d'un systkme implant6 nous proposons un modele bash sur l'interaction des impuretes avec les ddfauts crkes pendant l'implantation. Le modkle est consistant, au moins qualitativement, avec les resultats experimentaux.
Abstract.
-
By means of the Mossbauer Effect a systematic dose and dose rate study has been performed on xenon implanted in iron.The substitutional fraction of the xenon nuclei decreases when the implantation dose is increased from 5 X 1011 atoms/cm2 to 1015 atoms/cmz.
Two samples were implanted both to a dose of 5 X 1013 atoms/cm* but with a dose rate ratio of 75. The sample with the highest dose rate shows the highest substitutional fraction.
To describe the evolution of an implanted system we propose a model based on the interaction of the impurities with the radiation damage during implantation. This model is consistent, at least qualitatively, with the experimental results.
--f
1. Introduction.
-
Xenon implanted in iron (XeFe) has been the object of several hyperfine interactions and channeling-backscattering measurements during the last years [l-81. General agreement exists now about the existence of different inequivalent sites for the impurities. The populations of these sites however are not unique but may be dependent on the implan- tation parameters, namely dose, dose rate and+
temperature. In this work the evolution of XeFe as a function of dose and dose rate has been s t u d i a by means of the Mossbauer effect (ME).
2. Experimental procedure.
-
All implantations were performed at room temperature in polycrystalline, high-purity (99.99%)
iron foils by the Leuven Isotope Separator at an energy of 75 keV. The ion current was measured at the target itself, the target being part of a Faraday cup, and the current was integrated continuously. The beam was swept in two directions to achieve maximum homogeneity. From measure- ments we claim that the error on the total dose is less than 15%.
For the investigation of the site populations, we have studied l z 9 " Z e ~ e . lZ9Xe has a suitable Moss- bauer level at 3 9 . 6 k e ~ with a natural linewidth
r,
= 3.4 mm/s. The activity was obtained from neutron irradiation of 1271 ; this procedure has beendescribed elsewhere [5]. The absorber consisted of sodium perxenate Na,Xe0,.2 H 2 0 (obtained from PCR Inc., Gainesville, Florida) and had a thickness of 20 mg Xe/cm2. The y-rays were detected in a Xe-filled proportional counter, with the window setting on the escape peaks of the 39.6 keV line.
Since the specific activity of the 129mXe reactor -+ sources is rather low, we used 133XeFe for the study of the dose rate dependence of the site populations. 133Xe (t,,, = 5.3 d) feeds throuih :h-decay the 81 keV
Mossbauer level of 133Cs. Because of the 512 -+ 712
transition, the Miissbauer spectra exhibit a compli- cated line pattern, but this is somewhat compensated by the narrow linewidth T o = 0.27mmls. The absorber consisted of CsCl powder and contained 300 mg 133Cs/~m2. A 3" X 3" NaI(T1) scintillator
was used to detect the 81 keV y-ray.
The sources were mounted on an electromechanical drive system, operating in the triangular velocity mode. All measurements took place in a liquid helium cryostat, where source and absorber were cooled by exchange gas.
3. Experimental results and their discussion.
-
3.1 HYPERFINE FIELDS AND THE NATURE OF THE DIFFERENT SITES. - Evidence for various implantation sites of Xe in Fe has already been reported for 133Xe [6]
C6-900 J. ODEURS, R. COUSSEMENT, J. DE BRUYN, H. PATTYN AND M. VAN ROSSUM
as well as for 12'Xe [5]. Since these early results, the accumulation of data has clarified the analytical procedure by showing that a four sites model (includ- ing high-field, intermediate-field, low-field and zero- field component) is required in order to get a reasonable fitting for all experimental spectra. If one accepts this hypothesis, one obtains the following mean values for the hyperfine fields [g] :
H, = 1 485 & 55 kG
Hi
= 1100 f 100kG H, = 400 f 150 kG H, 150 kG.Apart from the magnetic field, each site is charac- terized by a specific Debye temperature 8, leading to different recoilless fractions f at LHe temperature. In order to obtain this information, measurements were done in which the source was kept at various temperatures between 15 K and 115 K. From the inten- sity variation of the various components, the follo- wing values were obtained for the Debye-tempera- tures andf-factors at 0 K. Component 0, f (T = 0 K) - - h 341
+
30 0.72+
0.04 i 275 $. 30 0.66 f 0.03 l 269 & 30 0.65t
0.03 o 185 $. 20 0.54+
0.03 The problem arises now about the nature of these field sites. Combination of channeling-backscatter- ing [3] and hyperfine interaction measurements [6] leads to the identification of the high field site with the substitutional one. The results for hyperfine fields and Debye temperature, together with the calculations of Sondhi [l01 lead to the conclusion that the i-field site belongs to xenon atoms with one associated vacancy, the I-field site to xenon atoms with two associated vacancies and the o-field site to xenon atoms with three or more associated vacan- cies (see also ref. [2] and [ l l]).3.2 DOSE DEPENDENCE OF THE SITE POPULATIONS. -
Several nuclear orientation (N. 0.) experiments have
+
been performed previously on XeFe [4, 81. The analysis of the results was based on atwo-site model :
a fraction a of the implanted atoms feels the full hyperfine field, the remaining fraction feels such a small field that no measurable anisotropy comes from these impurities. With *this analysis the full field fraction decreases continuously when the dose was increased from
As the N. 0. experiments could not differentiate further between the different low field fractions, a
systematic dose study with the Mossbauer effect technique has been done. Figure 1 to 4 indicate that the population of the different field sites is not con- stant, but varies when the implantation dose is varied.
FIG. 1. - Mossbauer spectrum of 129Xe following implanta- tion of 129mXe in iron at a dose of 5 X 1011 at/cm? The fit includes highfield, intermediate-field, low-field sites and a near-
zero-field site. The separate components are also shown.
FIG. 2.
-
Mossbauer spectrum of 129Xe following implanta- tion of 129mXe in iron at a dose of 5 X 1012 atlcmz.FIG. 4.
-
Mijssbauer spectrum of 129Xe following implantation of 12gmXe in iron at a dose of 1 0 1 5 atlcm2.The parameters p j ( j = h, i, I, o) describing the popula- tions are defined as :
Ij being the fitted intensity and fj the $factor of the
jth component.
The results are summarised in table 11.
Dose P m (at/cm2) P h (Pi
+
P L ) PO 7 - - 1 X 10'' 0.35+
0.04 0.33+
0.03 0.32+
0.03 5 X 1013 0.44+
0.04 0.31 f 0.03 0.25+
0.03 5 X 1012 0.41+
0.04 0.33 f 0.03 0.26 f 0.03 5 X 10'' 0.51+
0.05 0.23+
0.03 0.26+
0.03The populations of site i and I have been added since the fitting procedure does not allow a strict decorrelation between the two intensity parameters. A possible explanation for this dose dependence of the different site populations will be given in
8
4. 3.3 DOSE RATE DEPENDENCE OF THE SITE POPULATION.-
A series of ME-measurements were done to inves- tigate the influence of dose rate (atoms/cm2 S) on the different site populations.In a first experiment we have implanted 133Xe in two different iron foils up to the same dose (5 X 1012 atoms/cm2) but with a dose rate ratio of
50. There was not any significant difference between the two spectra. When we use a dose of 5 X 1013
atoms/cm2 and a dose rate ratio of 75, both spectra are clearly different (Figs. 5 and 6). Performing a three-component fit [6] resulted in the intensities of table 111.
13'xe (Fe) 5 . 1 0 ' h t /cm2 HIGH CURRENT
FIG. 5. - Mossbauer spectrum of I33Cs fed by the decay of 133Xe implanted in iron with a dose rate of 5 X 1011
atoms/cmz S.
"X= (Fe) 5.10'~ ~t /cm2 LOW CURRENT 53390
0 - 5
v(rnmls)
-
FIG. 6. - Mossbauer spectrum of 133Cs fed by the decay of I33Xe implanted in iron with a dose rate of 7 X 1 0 8
atoms/cmz S.
Dose rate
-
- I h -Zi
- 1,high 0.78
+
0.05 0.16 +_ 0.04 0.06+
0.02low 0.64 f 0.04 0.25
+
0.03 0.11 f 0.02 The ratios between the intensities are dose rate dependent. Absolute populations cannot be given. Indeed Reintsema pt al. reveal the existence of afourth component which is not visible in the Moss- bauer spectra due to its zero-recoilless fraction [6]. Since we have not measured the fourth component we cannot take it into account for the calculation of the absolute site populations. In any case, the dose rate dependence of site populations is obvious. The high field site population is favoured in the samples which are prepared with the high dose rate. A possible explanation of this phenomenon will be given in 4.
4. Model for the site population and conclusions.
-
G6-902 J. ODEURS, R. COUSSEMENT, J. DE BRUYN, H. PATTYN AND M. VAN ROSSUM
An impurity may interact with its own damage cascade where P, is the probability for landing directly substi- and with the damage coming from other cascades. tutionally.
If there was only an interaction of the impurity with I. e. in the low dose region one can look at the its own damage cascade there would not be any dose landing dynamics of an impurity.
dependence. This is contradicted by our experiments.
lim a = v , ,
So every model trying to describe the evolution of n - t m an implanted system has to be constructed on the
basis of the interaction of impurities with defects from its own collision cascade and with defects from the other cascades.
In an extensive treatment we have developped a model by which an implanted system may be des- cribed (*). The rest of this paragraph will be a short summary of this model.
During implantation each inequivaknt site is fed with a certain probability. Transitions between diffe- rent sites are possible due to trapping, detrapping and annihilation of defects. It means that all transi- tions are proportional to the defect density N,. An
expression is obtained for N, in the case of dynamical equilibrium of the defects :
with Q, the dose rate.
A set of simultaneous differential equations may be written down, the solution of which gives the evolution of the different site populations as a func- tion of time, which may be converted to a function of dose n.
E. g. the substitutional fraction or goes as :
N is the number of different sites, Q, is the dose rate,
ui, m , v, and W , are constants, only dependent on
the specific combination impurity-host. Some features of this curve are :
lim a = P , ,
n-rO
(*) ODEURS, J., COUSSEMENT, R. and PATTYN, H., to be published.
i. e. the substitutional fraction goes to a saturation value which is not dependent on dose rate Q.
In the high dose region the trapping and detrapping mechanisms are dominant and this may give, more fundamentally, an insight into the different defect potentials near an impurity. The experimental dose dependence is qualitatively in agreement with the produced curve. From (2) it is clear that the substi- tutional fraction a would be independent on the dose rate Q, if and only if y = 1. Since it has been shown [l21 that y = 1 means correlated annihilation of defects and y = 2 uncorrelated annihilation of defects our dose rate dependence proves that the transitions between the different sites are, at least partly, produced
bv free. uncorrelated defects. with the substitution
(2) may be rewritten
An important result is that all data concerning the population of one particular site, on one implanted system have to be fitted with a single curve which we call the reduced curve (4).
From these considerations we may conclude that one has to take into account the implantation condi- tions as dose and dose rate when one compares different experiments on the same combination host-impurity.
Acknowledgements.
-
We thank G. Brijs andR. Vanautgaerden for the implantations performed at the Leuven Isotope Separator.
We also thank the I. I. K. W. for their financial aid.
References [l] DE WAARD, H. and DRENTJE, S. A., Proc. R. SOC. A311
(1969) 139.
[2] DE WAARD, H., MiiSSbai~er Spectra and its Applications,
(IAEA, Vienna), (1972) 123.
[3] FELDMAN, L. C. and MURNICK, D. E., Phys. Rev. B5
(1972) 1.
[4] PATTYN, H., COUSSEMENT, R., DUMONT, G., SCHOETERS,
SILVERANS, R. E. and VANNESTE, L., Phys. Lett. 45 A
(1973) 131.
1.51 VAN ROSSUM, M., LANGOUCHE, G., PATTYN, H., DU-
MONT, G., ODEURS, J., MEYKENS, A., COUSSEMENT, R. and BOOLCHAND, P., J. Physique Colloq. 35 (1974) C6-301.
[6] REINTSEMA, S. R., DRENTJE, S. A., SCHURER, P. and DE WAARD, H., Rad. Efl 24 (1975) 145.
[7] ODEURS, J., COUSSEMENT, R. and PATTYN, H., Proc. Int. Conf. on Fundamental Aspects of Radiation Damage in Metals, Gatlinburg (1975) (to be published).
[8] ODEURS, J., COUSSEMENT, R., PATTYN, H., DUMONT, G., SCHOETERS, E., SILVERANS, R. E. and VANNESTE, L., (to be published in Hyperfine Interactions).
[g] VAN ROSSUM, M., DE BRUYN, J., LANGOUCHE, G. and Cous- SEMENT, R., Proc. Int. Meeting on Hyperfine Interac- tions, Leuven (1975) (to be published in Hyperfine Interactions).
[l01 SONDHI, I., J. Chem. Phys. 82 (1975) 1385.
[l11 DE WAARD, H., COHEN, R. L., REINTSEMA, S. R. and DRENTJE, S. A., Phys. Rev. B10 (1974) 3740.
[l21 Vacancies and Interstitials in Metals, SEEGER, A., SCHU-
