ИЗВЕСТИЯ РАН. СЕРИЯ ФИЗИЧЕСКАЯ, 2010, том 74, № 2, с. 212-220
УДК 378.374
AL-K-AUGER ENERGY SPECTRA: PROBING THE ELECTRON DYNAMICS IN ION-SOLID INTERACTIONS
© 2010 г. G. Schiwietz1, M. Roth1, R. Hellhammer1, K. Czerski1*, F. Staufenbiel1, R. C. Fadanelli2, J. Morais2, P. L. Grande2
E-mail: schiwietz@helmholtz-berlin.de
K-Auger electron emission has been investigated for incident electrons and for different types of heavy ions interacting with mono-crystalline aluminum (100) targets at specific kinetic energies of 3 to 5 MeV/u. In an effort to gain a profound knowledge about the ionization and vacancy-decay dynamics for the K-shell in Al, spectra have been measured with different energy resolutions and angular distributions have been taken as well. Here we concentrate on the energy spectra — we identify the measured peak structures and we investigate different line intensities and mean target charge-states quantitatively, in comparison with theoretical results.
INTRODUCTION
Energy and angular distributions of Auger electrons serve as sensitive probes of the atomic and electronic solid-state structure, of the excitation dynamics during the interaction of fast ions with solid matter and of the subsequent vacancy decay dynamics, involving also the atomic and electronic material properties. In this work, we will concentrate on the short-time dynamics with less attention on the geometrical target structure.
The interaction of light fast ions with matter gives rise to a linear response of the target, corresponding to individual excitations separated in space and time. This is not the case for fast heavy and especially highly-charged ions at velocities close to their stopping-power maximum [1—3]. Such ions yield so high energy-deposition densities that atoms close to the ion path are multiply ionized [3—7], largely consistent with the general atomic physics knowledge [8—9]. The pathways of the energy relaxation inside solids and the formation of ion tracks, however, are subject of a longstanding discussion [10—11]. Specifically, the influence of Coulomb explosion (a mutual repulsion of highly ionized target atoms) or of the electronic thermal spike (atomic motion due to electron-atom collisions and electron-phonon couplings) on the atomic rearrangement processes is often a matter of debate. In the case of fast heavy projectile ions, electron spec-troscopy has mainly been performed for amorphous carbon targets [12—14] and it turns out that convoy electrons (running parallel to the projectile and being attracted by the electrostatic projectile field) as well as Auger electrons resulting from the decay of target inner-shell electrons are sensitive to the central ion-
1 Helmholtz-Zentrum Berlin f. Materialien und Energie GmbH, Germany.
Inst. de Física, Universidade Federal Porto Alegre, Brazil
3 presently at Inst. of Physics, Uni. of Szczecin, Poland.
track region close to the ion path. Contrary to most lattice studies, such investigations allow for a more clear distinction of track-formation mechanisms.
For simple organic insulators such as polypropylene and Mylar, reduced convoy-electron energies [15] have been observed in comparison to the expectations for neutral atomic systems [16]. The corresponding (small) energy reductions point to a considerable fraction of nuclear-track guided electrons emitted close to the convoy-electron velocity. The high ionization density in the solid leads to an attractive potential, the ion-track potential. It guides electrons from deep inside the solid, where they suffer enhanced energy losses. Furthermore, the observation of decelerated Auger electrons for these materials has shown that the nuclear-track lasts for at least 10-14 sec [17]. Auger energy shifts of up to 42 eV are consistent with complete suppression of electron neutralization and at least with a partial Coulomb explosion of the hydrogen atoms in these polymers [18]. This might also influence the size and shape of craters on polymer surfaces [19].
For the semiconductor Si, small but significant Auger-energy shifts have been found (note that the value < 2 eV for amorphous Si is too low for a strong Coulomb explosion) [3, 20]. For semi-metals (C and Be) and conductors (metals and metallic glasses), however, no significant energy shift has been found [3, 5—6, 21—22]. Thus, it follows that crystalline Si as well as semi-metals and conductors allow for a faster neutralization of the track, thereby excluding the Coulombexplosion mechanism. For these materials, however, a broadening of the high-energy side of the Auger electron-peaks corresponding to high electron temperatures has been found [3, 5—6, 21—23]. Electron temperatures of up to 100 000 K have been extracted (for Be) by comparison of ion induced Auger lines with the corresponding shapes obtained for incident electrons [22] and the observed electron temperature seems to be closely related to an effective band gap around the
Fermi level. A new track related effect, namely an unusual emission-angle dependent intensity-variation has been found for Be K-VV and Al L-VV Auger electrons [24]. This suppression of up to about 40% is most likely related to an enhanced absorption of 100-eV electrons when they move along the hot ion track. In order to determine and to understand the Al K-Auger electron properties including the related target charge-states after electron or ion impact, we will concentrate on Auger-electron energy-distributions in this work.
EXPERIMENTAL DETAILS
Experiments have been performed with heavy ions at energies of 3 to 5 AMeV corresponding to velocities of roughly 10% the speed of light and electrons at only slightly higher velocities (5 and 7 keV). Stable ion beams of 60 MeV 12C6+, 100 MeV 20Ne10+, 200 MeV 40Ar16+, 645 MeV 129Xe40+, and 592 MeV 197Au46+ with electrical currents of a few hundred nA have been delivered for this experiment by the ISL RFQ-cyclotron combination of the Hahn-Meitner Institute in Berlin. The above given high projectile charge-states are average values, calculated using a simple charge-state fit formula [3], but measured in the case of Au (a mean projectile charge-state of qmean = 46.3+ [25] was found for incident 592 MeV 197Au30+ ions). These high projectile charge-states were obtained using a thin carbon stripper foil about 1 m before the target. Thus, chargestate equilibrium was approximately reached in front of the target surface, however, with a slightly reduced component of excited projectile states [26—27]. Hence, target excitation densities are nearly independent of penetration depth.
Although fast ions penetrate deeply into the target, Auger electrons at energies around 1400 eV stem only from the first few atomic surface layers because inelastic mean-free-paths are typically 27 A corresponding to about 7 atomic layers (NIST standard reference (inelastic-mean-free-path) database 71, version 1.1, National Institute of Standards and Technology, USA (2001)). This means, the effective target volume is a thin layer close to the surface and attention has to be paid to the preparation of clean surfaces. Furthermore, electron-collision cascades of fast 5-electrons (secondary electrons generated directly due to the interaction of projectiles with target atoms) close to the surface may produce inner-shell vacancies and contribute indirectly to the Auger signal [18, 28]. This unwanted fraction of the Auger signal is reduced for electron detection at backward ejection angles as well as for light targets. Because of the high K-shell binding energy of 1559 eV for Al and because of the low elec-tron-backscattering cross-sections, we estimate that this component will be relatively small (■ 10%) for Auger-electron detection at all backward angles (ejection from the ion-entrance surface).
All experiments have been performed using the UHV chamber presented and discussed in ref. [20]. The vacuum pressure was in the low 10-10 mbar during the experimental runs and the targets have undergone sputtering/heating cycles before the ion irradiation so that the surfaces stayed atomically clean during one or two days of beam time. In order to enable precise energy and angle scans, the surrounding magnetic field is reduced by a factor exceeding 100 (down to about 0.5 ^T) using a double magnetic shielding made of metal. An electron spectrometer with improved focusing properties of the electrostatic deflector (using a bathtub-like configuration of the electrostatic potential) has been used for all experiments (patent pending). The spectrometer could be moved around the target inside the magnetic shielding to determine also angular distributions. For all measurements performed in this work, the incident beam, the target-surface normal and the emitted electron direction lie in the same scattering plane. The manipulator tilt-axis was used to adjust the target surface-normal at an angle of 45° with respect to the electron or ion beam-direction, respectively (except for the 129Xe40+ data points, with anti-parallel directions of ions and surface normal).
ENERGY SPECTRA AND PEAK ASSIGNMENT
Figure 1 displays electron induced K-Auger energy spectra of Al for two different experimental situations (different incident energies, energy resolutions, and emission angles). The spectrum in the lower plot (induced by 7 keV electrons) is an average of 6 spectra that have been taken in between the ion-induced spectra during cyclotron beam times and thus, it serves as a reference for these experimental runs. This spectrum corresponds to emission along the surface normal, at an angle of 45° with respect to the incident beam. It involves an energy resolution of 14 eV, corresponding to the normal (non-deceleration) mode of the spectrometer. Because of the large energy range, we have already subtracted the continuous background due to 5-electrons for the lower graph. The solid curve shows a fit through the data points that allows separating the main lines by a
Для дальнейшего прочтения статьи необходимо приобрести полный текст. Статьи высылаются в формате PDF на указанную при оплате почту. Время доставки составляет менее 10 минут. Стоимость одной статьи — 150 рублей.