Refine
Document Type
Keywords
- 3h-Auger (1)
- Bandlücke (1)
- Ga2O3 (1)
- Metalloxide (1)
- Organik (1)
- Spektroskopie (1)
- resonant photoelectron spectroscopy (1)
Institute
BTU
Im Rahmen der vorliegenden Arbeit wurde mit Synchrotron-basierenden, optischen und Elektronen-basierenden Methoden die elektronischen Eigenschaften zweier unterschiedlicher Materialklassen untersucht. Der Fokus lag dabei im besonderen auf der Bestimmung der Größe der Bandlücke/HOMO-LUMO-Lücke und ihre Abhängigkeit der verwendeten Methode.
Die zwei unterschiedliche Materialklassen, welche untersucht wurden sind folgende:
Erstens die organischen Systeme Cobalt-Phthalocyanin und Fulleren. An Ihnen wird der Einfluss der verwendeten Messmethoden aufgezeigt und die sich draus ergebenen HOMO- LUMO-Lücken (elektronisch, Volumen, Oberfläche, optisch) diskutiert. Es zeigt sich, dass die optische Methode (Ellipsometrie) generell den kleinsten Wert und die elektronischen Methoden (XPS, XAS) den größten Wert ergeben. Die Gründe dafür werden ausführlich diskutiert. Durch die Kombination von XPS- , XAS- und Ellipsometrie-Messungen war es möglich die Orientierung der aufgedampften CoPc-Moleküle auf den SiO2-Substrat zu bestimmen.
Zweitens wurden die Metalloxide HfO2, Al2O3, In2O3 und Ga2O3 aufgrund ihrer großen technologischen Bedeutung untersucht. Mit Hilfe von ALD Zyklus-für-Zyklus Experimenten konnten die elektronischen Eigenschaften von HfO2 in Abhängigkeit der ALD- Zyklenzahl und das Wachstum von HfO2 in den ersten Zyklen studiert werden. Auf Basis der XPS- und EELS-Ergebnisse wurde ein Wachstumsmodell entwickelt. Es wird ebenfalls der Einfluss einer Dipolschicht an der Grenzfläche HfO2/SiO2 auf die Größe der Bandlücke in den EELS-Messungen diskutiert. Mittels resonanten Photoemission wurde ein neuer Auger-Mechanismus (3-Loch-Auger-Prozess) in den Oxiden gefunden. Dieser hat seine Ursache in Defekten. Defekte führen auch zu einer Fano-Anti-Resonanz, welche an der O1s-Absorptions-Kante beobachtet und diskutiert wird.
We report on spectroscopic investigations on β-Ga2O3single crystals. We
focus on the detailed analysis of the O1s resonance profile by resonant photoelectron
spectroscopy. We analyze the electronic structure and assign the O2p to build the valence band and both, O2p and Ga4s4p states to contribute to the conduction band. We determine the partial density of states for the valence and conduction bands and find a strong hybridization of O2p and Ga4s states. This is deduced from constant final state spectra on the O-KLL-Auger over the O K-edge and Ga L2,3-edge and a comparison to the corresponding X-ray absorption spectroscopy data. We also identify several types of defects. A broad band of oxygen derived defects is identified that extends through-out the gap. Small polarons are attributed to cause an anti-resonance in the valence states around the O1s threshold. In addition, a separate Auger decay at resonance indicates the existence of localized charge transfer states which involves localized Ga4sp
states.
We show recent results of our in-situ investigations where the role of substrate chemistry in ALD is outlined. While the usual strategy to develop new properties of ALD films is to find new precursors or new procedures, the influence of substrate chemistry on the growth properties of ALD films has been often underestimated. This has principally a technical reason, as the usual characterization methods (QMS, FTIR, ellipsometry) are only weakly sensitive to the substrate, and a characterization of substrates before ALD is often not possible. Thanks to the use of in-situ characterization methods, including photoemission and X-ray absorption spectroscopy with synchrotron radiation, we are able to determine chemical properties of substrates before ALD and after either half or complete ALD cycles. The substrate chemistry influences the standard Al2O3 ALD with TMA and water [1,2], as well as the TiO2 ALD with TTIP and either water, O2 or O2-plasma.
[1] M. Tallarida, K. Kukli, M. Michling, M. Ritala, M. Leskelä and D. Schmeisser, Chem. Mater. 23, 3159 (2011);
[2] M. Tallarida, C. Adelmann, A. Delabie, S. van Elshocht, M. Caymax, and D. Schmeisser, Appl. Phys. Lett. 92, 042906 (2011).
In this contribution we report on our in-situ2 cycle-by-cycle (up to the 25 cycles) investigation of the HfO2 atomic layer deposition (ALD) process using the method of x-ray photoelectron spectroscopy (XPS) and electron energy loss spectroscopy (EELS). We used Tetrakis-Dimethyl-Amino-Hafnium (TDMA-Hf) and H2O as precursors and p-type Si wafer with native oxide as a substrate. The XPS measurements were carried out at BESSY II in Berlin with primary energies of 150 eV and 640 eV and the EELS measurements were done with a primary energy of 50 eV. We measured the O1s, Si2p core level and the valence band including the Hf 4f core level. From the Hf4f to Si2p ratios taken at each energy we developed a growth model for the first monolayer and for the following layers too. From our data we conclude, that in the first monolayer up to the fourth ALD cycle an island growth occurs. The height of these islands is about 0,5nm. After the first monolayer is completed, a layer- by-layer growth can be expected. In order to proof this observation we have simulated the Hf/Si ratio for different excitation energies and found a very good agreement with our measurement data. The EELS data especially the evaluation of the loss function onset confirm our growth model.
We present a study of the initial growth of TiO2 on Si(111) by atomic layer deposition (ALD). The Si substrate was etched with NH4F before ALD to remove the native oxide film and to produce a Si-H termination. In−situ experiments by means of photoemission and X-ray absorption spectroscopy were conducted with synchrotron radiation on Ti-oxide films produced using Ti-tetra-iso-propoxide (TTIP) and water as precursors. O1s, Ti2p, C1s, and Si2p core level, and O1s and Ti2p absorption edges show the transition of the Ti-oxide properties during the first layers. The growth starts with a very small growth rate (0.03nm/cycle) due to the growth inhibition of the Si-H termination and proceeds with higher growth rate (0.1nm/cycle) after 1.5nm Ti-oxide has been deposited.