[en] The angular dependence of the Ge sputter yield is reported for primary beam energies between 270 and 3 keV. For energies above 1 keV no full oxidation occurs, whereas in the sub-keV energy range, full oxidation of the Ge is possible. The oxidation state and the ionization probability change very abruptly as function of the incidence angle. The length of the transient is angle independent at 3 keV, while there is a clear angular dependence of the transient for a primary beam of 500 eV O₂

[en] This paper investigates the effect of thickness, composition and anneal temperature in N₂ on the band gap of the mixed oxide layer and the band offset with the Si substrate for layers grown by atomic layer deposition. For 1:1 (Al/Zr or Al/Hf) mixed oxides, the band gap reaches the bulk value for a thickness larger than 2 nm. In both cases (Al_xZr_yO_z and Al_xHf_yO_z), the bulk values of the band gap show a linear dependence on the Al₂O₃ mol%, going from 5.6 (5) eV for pure ZrO₂ (HfO₂) to 6.7 eV for pure Al₂O₃. The effect of a post-deposition anneal on the band gap and the valence band alignment is also studied. Here, Al_xHf_yO_z and Al_xZr_yO_z seem to act differently: while the annealing temperature does not have any influence on the band gap of Al_xZr_yO_z mixed oxide, the annealing in N₂ at temperature from 900 deg. C generates an increase in the band gap value measured by XPS for Al_xHf_yO_z

[en] By combining in situ SIMS/RBS measurements, sputter yields variations and ionisation probability variations were measured in the transient region of a poly-Si_0.8Ge_0.2 layer bombarded with 12 keV O₂⁺ at normal incidence. The goal of the experiment is to investigate if the prepeak in the Ge⁺ SIMS signal could be explained by sputter yield variations. The result of the experiment shows a minimum in the sputter yield variation of Ge during the build-up of the altered layer, which does not coincide with the minimum of the Ge⁺-signal. This suggests that the prepeak is not only a sputter yield effect but also due to a change in ionisation probability

[en] Measuring HfO₂/Si stacks by secondary ion mass spectroscopy (SIMS) has become a common task in the semiconductor industry. However, when analyzing these stacks with an oxygen beam, a strong impact of the experimental conditions on the Hf⁺ signal as well on the Si signal can be observed. In order to understand these effects, we investigated the effect of oxygen on the depth profiling of these stacks using an Atomika 4500. In this paper, the ionization probability of Hf sputtered from a HfO₂/Si stack as well a Hf implantation in Si is measured for different beam incidence angles and energies. It is shown that basically two physical mechanisms are responsible for the Hf signal intensities. The first part of the signal can be correlated with the impact of oxygen on the Hf ionization degree, in line with the classical bond-breaking theory. The second mechanism is more difficult to define, but appears linked to the emission of Hf into a different (neutral, molecules?) emission channel, thereby reducing the available Hf for detection as Hf⁺. Finally, a very strong influence of Hf on the Si and Si-cluster ions is reported

[en] As emphasized in the ITRS roadmap, two-dimensional (2D) carrier profiling is one of the key elements in support of technology development. For CMOS silicon devices, scanning spreading resistance microscopy (SSRM) has demonstrated an attractive spatial resolution and concentration sensitivity. The automated construction of calibration curves allows for the fast semi-quantitative transformation of one-dimensional (1D) and 2D resistance profiles/images into resistivity/carrier profiles/images. However in order to arrive, at a reliable, fully quantitative analysis a new physical contact model involving a Schottky-like contact with tunneling and surface states has been proposed. The latter has been based on establishing a qualitative agreement with experimental data. The first aim of this work is to refine this contact model in order to achieve a quantitative agreement between device simulations (with ISE/DESSIS) and experimental 1D profiles on well-calibrated, junction isolated (carrier spilling affected), sub-micron CMOS structures. Among others, scanning spreading resistance spectroscopy (SSRS), i.e. collecting a full I-V curve at each data point, will be used. Furthermore, the impact of this new contact model on the deconvolution procedure from the measured resistance profiles/images towards resistivity/carrier profiles/images through an improved correction factor database will be discussed

[en] This study is dedicated to a better understanding of the processes occurring under ion bombardment of ultra-thin ZrO₂/SiO₂/Si gate dielectric stacks. Complex-shaped depth profiles were obtained by using TOF-SIMS with dual beam (500 eV for sputtering and 10 keV for analysis) Ar⁺ ions. The SIMS intensities of all the elements depend critically on the amount of oxygen at any moment of the sputtering process. Increased intensity is observed at the surface and at the ZrO₂/SiO₂ interface. A long tail of the Zr signal is present in the Si substrate, even after the second (SiO₂/Si) interface, and a double bump structure in the ⁹⁰Zr and ZrO dimer is observed, which is more pronounced with increasing thickness of the interfacial SiO₂ layer. Computer simulations using the dynamic Monte Carlo code (TRIDYN) are performed in order to distinguish the ion-bombardment-induced effects from changes in the ionization degree. The original code is extended with simple models for the ionization mechanism and for the molecular yield during sputtering. Oxygen preferential sputtering at the surface and ballistic transport of Zr towards and through the interface are clearly demonstrated, but there is also evidence that due to recoil implantation oxygen gets piled-up near the ZrO₂/SiO₂ interface

[en] Using Cs⁺ in the TOF-SIMS dual beam mode offers a semi-quantitative solution to depth profiling. Specifically, the use of these alkali ions strongly increases negative ion yields, decreases the positive ones and allows the formation of MCs⁺ and MCs₂⁺ clusters. Recently, Niehuis and Grehl [Proceedings SIMS XII (2000) 49] developed a new approach consisting of co-sputtering Xe and Cs in order to control the Cs surface concentration, thus allowing the optimization of elemental and cluster ion yields. We applied that technique on different well-defined samples (e.g. Si, SiO₂ and Al₂O₃) and we monitored positive ions (e.g. Si⁺, Al⁺, CsSi⁺, CsAl⁺, CsO⁺, Cs₂O⁺, Cs₂Si⁺, etc.) as a function of the sputtering beam Cs concentration. First, we observed the decrease of the elemental ions due to the work function lowering, as is predicted by the tunneling model. We then studied the behavior of the MCs⁺ and the MCs₂⁺ clusters. The MCs⁺ yield exhibits a maximum at a given Cs/Xe beam concentration ratio, depending on the studied element M and also on its chemical environment (e.g. Si and SiO₂), and on the energy of the Cs beam. In other words, it is hypothesized that this yield maximum is a consequence of the competition between the varying surface Cs coverage (direct concentration effect) and the decreasing ionization probability due to that varying Cs [Phys. Rev. Lett. 50 (1983) 127; Phys. Rev. B 29 (1984) 2311; K. Wittmaack, Proceedings SIMS VIII, (1992) 91]. Simple models based on the tunneling model were applied to interpret our results. The MCs₂⁺ signal behaves in a very different way. As shown by Gao [Y. Gao, Y. Marie, F. Saldi, H.N. Migeon, Proc. SIMS IX, (1994) 382], these clusters are predominant for electronegative elements and increase in a monotonous way with Cs beam concentration

[en] Using a 3 keV O₂-beam in an in situ SIMS/RBS set-up, changes in sputter yield and ionization probability were measured in order to come to a better understanding of the behavior of the Si⁺ signal during the build up of the altered layer. Based on the RBS-spectra the evolution of the sputter yield within the transient region can be determined. When combined with complementary results from a low energy ion scattering (LEIS) experiment [T. Janssens, C. Huyghebaert, W. Vandervorst, A. Gildenpfennig, H.H. Brongersma, in these proceedings], the ionization probability enhancement as a function of the oxygen concentration at the surface can be calculated

[en] We analyzed ultra-thin ZrO₂/SiO₂/Si gate dielectrics under post-deposition anneals in dry O₂ at temperatures from 500 to 700 deg. C. TOF-SIMS profiling of ZrO₂/SiO₂/Si stacks is hampered by many sputter induced artifacts. The depletion of oxygen leads to a decrease in SIMS intensities. However, preferential sputtering is accompanied by transport of the depleted species towards the surface. Due to recoil implantation oxygen gets piled-up near the ZrO₂/SiO₂ interface. Either normal or radiation-enhanced diffusion transports oxygen back to the surface. Simultaneously also segregation of zirconium towards and through the interface is observed, resulting in a large zirconium tail in the underlying silicon substrate

[en] As SiGe becomes a building block of advanced devices, depth profiling of SiGe gains in importance, requiring an understanding of its basic behavior under oxygen bombardment. In this study, the sputter yield and ionization probability of Si and Ge sputtered from SiGe-substrates are studied as a function of the incidence angle. The altered layer formation and the element redistribution are studied with in situ sputter/RBS and the oxidation of the bombarded SiGe surface is analyzed with XPS. The results show that the altered layer formation in the SiGe system is much more complex than in the Si case and strongly dependent of the incident angle/primary energy. In addition to sputter related effects, thermodynamics and volatility of the Ge-oxide are important mechanisms