De rol van argon ets plasma samenstelling op ultradunne extreme ultraviolet resists

State-of-the-art microchips can contain over a billion transistors with up to a hundred intricate layers forming complex circuits. The goal of advancing semiconductor fabrication is to produce increasingly smaller transistors, as this can improve the performance and reduce the power consumption of these chips. This fabrication process of these chips is highly complex, involving many steps starting from a silicon substrate. In this project, we focus on two steps in this fabrication process flow: photolithography (or lithography) and plasma etching. These are involved in making the intricate patterns in a chip. In lithography, a pattern from a template, known as a mask, is transferred using light to a photosensitive layer called photoresist (or resist), which is coated on top of the layer to be patterned. One of the main enablers for the miniaturization of transistors has been the improved resolution of patterning during lithography. By reducing the wavelength of light used during lithography, the diffraction-limited resolution can be improved, allowing the patterning of small features. Currently, the state-of-the-art lithography tools use light with a wavelength of 13.5 nm , referred to as extreme ultraviolet (EUV) light. With EUV lithography, sub-10 nm features can be printed in the resist. The next step after patterning the resist is to transfer it from the resist to the layer below. This is done via plasma etching. As the name states, plasma is used to remove (or etch) the regions of the layer that are not protected by the patterned resist creating the pattern transfer. During etching, it is important for the patterned resist to remain intact under plasma exposure for a successful pattern transfer. Typical plasmas used for etching contain ions, vacuum ultraviolet (VUV) photons, radicals, and electrons. One of the challenges with EUV lithography is that the resists used are ultrathin (<50 nm) and can undergo significant modification by these plasma species during etching. This can, in turn, complicate the pattern transfer process. In this context, it is valuable to study the interactions of plasma species with ultrathin resists, as this can help in resist design and etch process development. The main goal of this project is to study the impact of argon plasma on ultrathin resists. Argon gas is commonly used in various etch chemistries for plasma etching. Here, we focus on the impact VUV photons and Ar+ ions, which are the main etching species present in an argon plasma, on ultrathin resists. To do this, we process the resist films under three different conditions: exposure to only VUV photons using a deuterium lamp, exposure to only Ar+ ions using an ion beam etch tool, and exposure to Ar plasma using an inductively coupled plasma etch tool. We then use different characterization techniques such as atomic force microscopy, ellipsometry, and infrared spectroscopy to study the physical and chemical changes occurring in these films after different exposures. In the first part of this project, we studied the impact of argon plasma species on a 40 nm thick polymethylmethacrylate (PMMA) film. PMMA is a simple homopolymer and has been widely studied as a model photoresist. This made it an ideal first candidate to validate our testing methodology. We observed that VUV photons and Ar+ ions interact quite differently with PMMA. VUV photons penetrate the entire film, causing significant side chain degradation throughout the bulk. Ar+ ions mainly interact at the top of the film, creating a carbon-rich ion-modified layer while leaving the bulk pristine-like. For the plasma-exposed PMMA, both bulk modification from VUV photons and the formation of a top ion-modified layer are observed. These interactions result in additional synergistic effects, leading to substantial surface modification. Next, we extended this methodology to a popular type of organic EUV resist called chemically amplified resists (CARs). Here, we investigated three different formulations of negative tone EUV CARs, differing in their polymer structure, along with PMMA as a reference resist. The etch rate of the resists was dependent on both their chemistry and starting thickness. Resists with aromatic groups and a high carbon to oxygen ratio showed higher etch resistance. Thinner PMMA shower lower etch rate compared to its thicker counterpart. This behavior was attributed to the attractive interaction between PMMA and the underlayer, which affects the film properties and further influences the etch rate. Similar to PMMA in the previous study, the CARs also showed the presence of an ion-modified carbonized layer at the top. The surface roughness evolution of the resists when exposed to full Ar plasma was complex. Multiple factors such as plasma heating, substrate effect, resist chemistry, and the initial thickness of the resist, are expected to influence this process. We also tested the validity of the Ohnishi number and ring parameter etch rate models for the resists and processing conditions used. Both models could predict the etch rate trends of the EUV CARs reasonably well. In the final part of this project, we investigated the impact of Ar plasma species on a zinc-containing organometallic EUV resist called ZONE. Metal-containing hybrid resists are promising for EUV lithography due to their high EUV absorption cross-section, small molecular size, and good etch resistance. ZONE is composed of zinc-oximate molecules, and similar to the previous studies, we exposed it to VUV photons, Ar+ ions, and Ar plasma. VUV photons mainly affected the oxime ligand, causing degradation of the organic moieties in the resist and an increase in the refractive index. Ar+ ions, similar to previous results on PMMA and CARs, still appeared to modify only the top of the film. However, unlike in PMMA and CARs, no indication of carbonization was observed. The lack of carbonization was attributed to the absence of carbon-rich polymeric structures in ZONE. The thermal stability of ZONE was also tested by baking it on a hot plate to above 100 C. The thickness of ZONE decreased with baking, indicating that heating is also an important pathway for degradation. The plasma-exposed ZONE showed a significant loss of the ligand IR peaks, a substantial increase in the refractive index, and a distinct grainy surface. These findings strongly pointed to the formation of zinc oxide grains at the surface of the plasma-exposed ZONE. This formation of these grains was attributed to significant modification of the film surface due to the combined effects of Ar+ ions, VUV photons, and plasma heating. In conclusion, over the course of multiple studies in this project, we have investigated the impact of VUV photons and Ar+ ions on ultrathin EUV resists during argon plasma etching. Our findings highlight the role of VUV photons in modifying the bulk properties of the resists, while Ar+ ions predominantly affect the surface. The complex interaction of these two modifications, along with additional factors like plasma heating, give rise to unique combined effects in Ar plasma, especially at the surface of the resist films. In addition to this, the importance of the starting thickness of these ultrathin films and the interfacial interactions with the underlayer have also been highlighted. The results from this study serve as an initial point for further exploration of plasma-resist interactions in EUV resists.