The aim of this work was the development of a method for the generation of surface features on n-type silicon samples with deeply buried p-implants, in the form of heterogeneities aligned directly above the buried implants. This task was motivated by the realisation of a simpler process for the formation of superjunction transistors, which currently require the repeated creation of the same implantation structure over multiple steps of photolithography These lithography steps can be potentially replaced, if a suitable process for the self-alignment in accordance to the buried implants can be found. The work on this goal was separated into three parts: the analysis of samples for suitable surface properties, the generation of surface heterogeneities using such a property and the analysis of the mechanism for the used process of contrast generation.
Within this doctoral thesis, a before unseen method of selective etching on silicon was discovered and investigated. Hence, the overall aim of this work was successfully achieved.
• Samples containing buried p-implants inside a n-type silicon substrate were characterised with regard to various properties. Of these, the through-sample resistance showed a significant variation in accordance to the buried implants also through a homogeneous epitaxial layer.
• Various methods aimed at the usage of the resistance variation in order to generate a surface heterogeneity through electrodeposition failed to enable a suitable process. Instead, another method was found, which enables the replication of the implant structure via selective etching. This novel process enables the lithography free patterning of the substrates through a simple alkaline etch process performed under illumination. This results in a surface heterogeneity as an alteration of the sample topography combined with a material contrast due to the formation of an in-situ SiO2 etch mask. This material variation can also be used for the selective deposition of polymers, enabling further processing of the etched samples.
• For this new method, named Light Induced Selective Etching (LISE), a mechanism underlying the selectivity was proposed and through a number of experiments. In essence, the illumination during the etching process produces a flux of photogenerated electrons directed from the buried implants toward the surface, which increase the negative surface charge in the areas above these implants. The locally increased surface charge causes a local protection of the native silicon oxide layer against the alkaline etching, leading to the structuring of the substrate.
In essence, this novel method allows for the previously unreported self-adjusted structuring of silicon based on deeply buried implant structures. In general, even the characterisation of such implant structures is difficult, whereas this method allows for structuring with regard to such buried structures with a very simple setup of only an etchant solution and a suitable light source. With regard to the introduction and motivation of this thesis, this process can possibly be applied for the intended purpose of creating a self-aligned resist in order to replace repeating lithography steps. This is the case in particular in combination with polymer deposition, as shown in the last part of the results. Certain limitations, such as the resolution limit and dimensional size increase exist, but can be circumvented by appropriate device design and further optimisation of the process parameters. Furthermore, the LISE process appears applicable for the manufacturing of MEMS and MOEMS devices, as the typical feature sizes in these cases fit well to the achieved resolution of the LISE process. For devices needing a certain implant structure in combination with a corresponding topography, the new method allows for the elimination of at least one lithography step, including the necessary substeps such as alignment and measurement. Accordingly, LISE has the potential of simplifying the manufacturing process, enabling better and cheaper devices.
| Identifer | oai:union.ndltd.org:DRESDEN/oai:qucosa:de:qucosa:75226 |
| Date | 28 June 2021 |
| Creators | Schutzeichel, Christopher |
| Contributors | Voit, Brigitte, Mannsfeld, Stefan, Technische Universität Dresden |
| Source Sets | Hochschulschriftenserver (HSSS) der SLUB Dresden |
| Language | English |
| Detected Language | English |
| Type | info:eu-repo/semantics/publishedVersion, doc-type:doctoralThesis, info:eu-repo/semantics/doctoralThesis, doc-type:Text |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0017 seconds