HAL Id: jpa-00226112
https://hal.archives-ouvertes.fr/jpa-00226112
Submitted on 1 Jan 1986
HAL 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.
KRYPTON XANES STUDIES IN IMPLANTED SYSTEMS
J. Budnick, D. Pease, M. Choi, Z. Tan, G. Hayes, F. Namavar, H. Hayden
To cite this version:
J. Budnick, D. Pease, M. Choi, Z. Tan, G. Hayes, et al.. KRYPTON XANES STUDIES IN IMPLANTED SYSTEMS. Journal de Physique Colloques, 1986, 47 (C8), pp.C8-1053-C8-1056.
�10.1051/jphyscol:19868204�. �jpa-00226112�
KRYPTON XANES STUDIES IN IMPLANTED SYSTEMS(')
J.I. BUDNICK, D.M. PEASE, M.H. CHOI, Z. TAN, G.H. HA YES(^), F. NAMAVAR and H.C. HAYDEN
Department of Physics and Institute of Materials Science, University of Connecticut, Storrs, CT 06268, U.S.A.
Resume.--On a mesure la dependance en temperature de l'absorption des rayons X pres du seuil K (XANES) pour le krypton implante dans du niobium, de l'aluminum et du grafoil. Pour les cibles d'aluminum et de grafoil on
~bserve une augmentation des "white line" mesurees avec des basses tenpera- tures. Apres les mesures avec XANES, nos echantillons on ete etudes par RBS.
Abstract.--Temperature dependent K edge XANES of krypton have been measured for krypton implanted into niobium, aluminum and grafoil. For the cases of aluminum and grafoil targets, there is a marked enhancement of the observed white line for the XANES measured at lower temperatures. In addition to the XANES studies, our samples are further characterized by Rutherford
backscattering spectroscopy.
We report some initial results of the study of K edge XANES for krypton implanted systems as part of our investigation of the local environments of rare gases in solids. There has been considerable recent interest in the structures produced when rare gas ions are implanted into metallic hosts, and it is evident that the implanted gases often accumulate into small clusters which, even at room temperature, are in the solid phase[l-31. Thus, Donnelly and Rossouw have shown from electron microscope studies that Xenon implanted into aluminum forms solid xenon precipitates at room temperature[l], and similar findings are presented by Evans and Mazey for the cases of krypton implanted into copper, nickel and gold[2], and by Birtcher and Jager for krypton implanted into aluminum[3]. We have implanted krypton into grafoil, aluminum and niobium, and have measured, in fluorescence, the krypton X-ray absorption near edge spectra (XANES) both at room temperature and near liquid nitrogen temperature. We have in addition characterized the samples by means of Rutherford back scattering (RBS) spectra, which show a maximum retained krypton dose, at roughly 1000 % penetration, of nearly 9 atomic percent for krypton in aluminum. The RBS spectrum of krypton in niobium is difficult to analyze, since niobium has a greater atomic mass than does krypton. Less krypton is retained in grafoil than aluminum targets, but by using multfple dose implanting, we were able to attain a retained maximum dose, at about 700 A penetration, of roughly 2 atomic percent. For the case of aluminum targets, at least, the retained krypton dose is comparable to that attained in other studies[l-31. Although the,retained krypton dose in grafoil targets was less than attained (by Evans and Mazey, for which solid krypton bubbles were formed in metal targets[2], solid krypton bubbles in aluminum were found by Birtcher and Jager[3] for even smaller krypton doses than we obtain in grafoil. In addition, there is evidence from the XANES that the atomic near neighbor of a typical krypton atom as implanted in our grafoil systems is primarily krypton rather than carbon. We will return to this point later. A typical rare gas solid bubble diameter obtained by Evans and Mazey[Z] and Donnelly and RossouwIl] is on the his work was performed a t beam l i n e XllA a t NSLS and i s supported by t h e DOE under Contract no. DE-ASOS-BOER 10742
("present Address : Optical Group Research, Perkin Elrner Corp., Danbury. CT 06810. U.S.A.
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:19868204
C8-1054 JOURNAL DE PHYSIQUE
order of 3x10-' meter. Structures this small are beyond the resolution limit of our scanning electron microscope (SEM) in the x-ray imaging mode. Nevertheless, we did investigate the implanted grafoil targets by means of an SEM and were able to resolve only a uniform lateral krypton distribution, as expected.
The krypton implanted XANES spectra, together with XANES spectrum for gaseous krypton, are shown in Fig. la-d. None of the spectra observed for implanted krypton is characteristic of krypton in the gaseous state. A pronounced "white line" is observed near threshold for all implanted systems. A large, reversible enhancement of the white line structure is observed for the XANES of krypton in grafoil taken at liquid nitrogen temperatures, and a smaller enhancement for krypton in aluminum case.
Any low temperature enhancement in the niobium samples is small, and barely detecta- ble with our present data. Also, at present we do not detect any significant
- -10 0 10 20 30
( a ) Energy, eV
I
-10 0 10 20 30
( c ) Energy, eV
. .
-10 0 1'0 2 0 30 (b) Energy, eV
Fig. 1. Norma+!zed K d e XANES of krypton implanted with a total dose of
2x10 rons/em5 :t 150 keV into (a) grafoil, (b) A1 and ( c ) Nb. NT and BT refer to liquid nitrogen temperature and room temperature data. Curves are arbitrarily aligned so that initial inflection points fall on E=O.
(d) K edge XANES of gaseous krypton.
Although the temperature dependence in EXAFS is well studied and understood, it is most uncommon to find significant temperature dependence in XANES, for energies down to the initial absorption threshold. One previously studied example is in the K edge XANES of lithium metal[4], in which it was necessary to take phonon broadening effects into account in order to compare experiments to threshold singularity theories[5]. Also, Bouldin and Stern have recently studied krypton EXAFS of krypton absorbed onto grafoill61. Although only EXAFS is discussed, comparison of figures in the Bouldin-Stern paper show significant enhancement of the XANES upon going from 100° K to 10' K, as well as enhancement of the EXAFS.
It is, however, interesting to compare the temperature dependence of our implanted Kr-graphoil XANES with the results of Bouldin and Stern on adsorbed KrI61. In the study by Bouldin and Stern, the spectra taken with the polarization vector parallel to the grafoil surface contain signals corresponding to Kr-Kr pairs, whereas the spectra taken with the polarization vector perpendicular to the surface contain primarily signals from carbon backscatterers. Inspection of the threshold region of the Bouldin-Stern data reveals that the perpendicular polarization data shows an enhancement of the second XANES oscillations, upon going from loOD to lo0 K, which is even greater than the enhancement of the first oscillation, and at lo0 K the second oscillation has the higher amplitude. On the other hand, the parallel polarization curve for 10' K shows a first XANES peak, or white line, that is much more pronounced than all other features in the signal.
The parallel polarization data of Bouldin and Stern is therefore more like the XANES of solid krypton as observed by Kutzler, et a1.[8] than is the perpendicular polarization data, which is primarily influenced by carbon backscatterers. Our XANES data for implanted krypton into grafoil shows that the first XANES oscillation enhances much more than the second upon going from room temperature to approx. 77" K and therefore it is more likely, from the above comparison, that our spectra show effects which arise from Kr rather than from carbon near neighbors.
Soules and Shaw obtain, for the K edge XANES of solid krypton, two threshold peaks separated by approx. 10 eV[7]. Kutzler, et al., report initial peak separations of 12.6 eV[8], with much better apparent resolution than was obtained by Soules and Shaw[7]. Soules and Shaw show a high energy threshold shift of approx. 4 eV relative to the XANES of gaseous krypton, and Kutzler, et al., detect only an approx. 1 eV shift[7,8]. The peak separations we obtain in our spectra are of the order of 10 - 13 eV and in this regard are consistent with reported results on bulk solid krypton at helium temperatures, though the amplitude of our white line varies greatly with temperature and in no case approaches that observed
for solid krypton at 4 . 2 O K by Kutzler, et a1.[8].
There is more than one theoretical framework in which one might wish to discuss these temperature dependent effects. Thus, Kutzler, et al., have calculated the solid krypton K edge XANES using a cluster calculation, and describe the XANES in terms of continuum state scattering of the ejected electron wave functions[8], whereas there are other theories of K edge white lines in which core hole virtual bound states are important[9]. A scattering formalism would imply that some low energy extension of an EXAFS type thermal disorder treatment would be appropriate[5], whereas it should be pointed out that thermal broadening effects on threshold XANES of light elements have previously been described in terms of core hole-phonon interactions[lO].
Of importance in the present discussion, however, is the fact that scattering treatments of thermal disorder, as applied to solid krypton, would predict a decrease in the temperature dependence of the vibrational disorder as pressure increased. Thus, in anharmonically vibrating systems, there will be a close coupling between the thermal coefficient of expansion and the thermal dependence of the mean square atomic displacement from equilibrium[ll]. Moreover, for rare gas solids, the thermal coefficient of expansion will decrease with pressure;
C8-1056 JOURNAL DE PHYSIQUE
thus, inspection of the molar volume versus pressure isotherms of solid xenon shows that the partial derivative of the molar volume with respect to temperature, holding pressure constant, will decrease with increases in pressure[l2]. The situation is more complex as far as core hole-phonon broadening is concerned, but the widths resulting from this mechanism appear to be, at maximum, of the order of 0.2 eV, even for the case of lithium metal[4]. We therefore believe the temperature dependent effects we observe are more likely to be associated with EXAFS like scattering phenomenon, and that the thermal smearing of solid krypton should, from the above arguments, decrease with pressure. Another possibility would be that some mixture of gaseous and solid phases occurs; but in this picture as well the fractional amount of gaseous phase and consequent temperature dependence of the white line, should decrease with internal pressure.
From the above, one interpretation of our data, is that changes in the temperature dependence of the observed XANES, upon going from system to system, correlate with changes in the pressure influencing the local krypton environment.
Thus, Evans and Mazey, by measuring lattice constants of as-planted solid krypton bubbles and referring to an equation of state, have measured average pressures of solid krypton bubbles. They find that the resulting pressures tend to vary according to the shear moduli of the implanted metal[2]. In the case of our systems, the shear modulus of aluminum is 0.26 megabars, and that of Nb is 0.36 to 0.39 megabars [l3]. We have been unable to find specific values of shear module for grafoil, but generally soft materials such as grafoil have very small elastic moduli[l3]. Thus, upon going from grafoil, with the smallest shear modulus, to aluminum, and then to niobium, we find large, systematic decreases in the temperature dependence of the XANES. This result should be compared to the results of Evans and Mazey[2], who show that for krypton as-implanted into gold, copper, and nickel the target shear moduli were in the ratio of 1 to 1.7 to 3.1 while the as-implanted average krypton pressures, as determined from lattice constants, were in the ratio of 1 to 2.6 to 5.
Of course, the hypothesis of a correlation, between the internal pressure on as-implanted solid rare gas bubbles, and the temperature dependence of the XANES is at present speculative, and must be tested by further studies.
References
[I] Donnelly, S. E. and Rossouw, C. J., Science 230 (1985) 1272.
[2] Evans, J. H. and Mazey, D. J., J. Phys. F: Met. Phys 15 (1985) L1.
[3] Birtcher, R. C. and Jager, W., Nucl. Inst. and Meth. B15 (1986) 435;
Petersen, H., Phys. Rev. Lett. 35 (1975) 1363.
[4] Callcott, T. A., Arakawa, E. T. and Ederer, D. L., Phys. Rev. B16 (1977) 5185; Baer, Y., Citrin, P. H. and Wertheim, G. K., Phys. Rev. Lett. 37 (1976) 49.
[5] Stahan, G. r)., Phys. Rev. 163 (1967) 612; Nozieres, P. and DeDominicLs, C. T., Phys. Rev. 178 (1969) 1097.
[6] Bouldin, C. and Stern, E. A., Phys. Rev. B25 (1982) 3462.
[7] Soules, J. A. and Shaw, C. H., Phys. Rev. 113 (1959) 420.
[8] Kutzler, F. W., Ellis, D. E., Morrison, T. I., Shenoy, G. K., Viccaro, P. J., Montano, P. A., Appelman, E. H., Stein, L., Pellin, M. J. and Gruen, D. M., Solid State Comm. 46 (1983) 803.
[9] Holland, B. W., Pendry, J. B., Pettifer, R. F. and Bordas, J., J. Phys. C11 (1978) 633; Cauchois, Y. and Mott, N. F., Phil. Mag. 40 (1949) 1260.
[lo] Parratt, L. G., Rev. Mod. Phys. 31 (1959) 616: See footnote on pg. 642 giving a phonon broadening formula attributed to A. 8. Overhauser; Flynn, C. P., Phvs. Rev. Lett. 37 (1976) 1445; Citrin, P. H. and Hamann, D. R., Phys. Rev.
~ 1 5 (1977) 2923.
Kittel, C., Introduction to Solid State Physics, 3rd edition. John Wiley and Sons, Inc., N.Y., pp. 184-185.
Anderson, M. S. and Swensen, C. A., J. Phys. Chem. Solids 36 (1975) 145.
Simmons, G. and Wang, H., Single Crystal Elastic Constants and Calculated Aggregate Properties, 2nd edition, M.I.T. Press, Cambridge (1971) pp. 153- 298.
