Would you like to add an event to this list? Simply register your event using this form.
Directed self-assembly of block copolymers for complementing extreme ultraviolet lithography
Category
Ph D Defense
Date
2026-03-23 17:00
Venue
KU Leuven, Aula Rosalind Franklin, 01.01 - Celestijnenlaan 200H
3001 Leuven, België
3001 Leuven, België
Promovendus/a: Julie Van Bel
Promotor(en): Prof. dr. Stefan De Gendt, Dr. Hyo Seon Suh, Dr. Lander Verstraete
Further downscaling the most critical features in microchip semiconductor devices requires the industry to switch from 193 nm immersion lithography to extreme ultraviolet (EUV) lithography with a wavelength of 13.5 nm. However, this large energy increase is severely limiting the number of photons produced by a given source power leading to non-uniformity, roughness and defectivity with conventional chemically amplified resist patterning. Block copolymer directed self-assembly (DSA) could provide a solution. It is a complementary patterning approach to photolithography and has proven to be promising for pattern rectification. By phase-separation at the nanometre scale, block copolymers like PS-b-PMMA can self-assemble into line or hole patterns with domain sizes ranging from 3 to 80 nm. The photolithography pattern underneath serves as chemical guide pattern to impose translational and long-range order, but the final pattern quality is defined by the block copolymer structure. In this PhD study, the patterning performance of EUV + DSA is examined. The compatibility of EUV lithography and DSA has been proven in the literature; however, at the start of this thesis, a flow combining the two had not been developed in a high-volume manufacturing compatible fab. We successfully set up an EUV + DSA process flow creating pitch 28-nm line space patterns. Although the high-volume manufacturing targets for roughness and defectivity could not be reached, our systematic study learns that the line edge roughness can be significantly improved by optimizing DSA material and process parameters. In particular, the most critical parameters are the BCP periodicity and BCP film thickness. Regarding defectivity, a combination of optical inspection and electron beam review pointed out that the BCP annealing temperature, as well as the chemistry and geometry of the guide pattern, play a key role. To conclude, a comparative study was conducted between our EUV + DSA process flow (1X DSA) and the previous ArFi + DSA process flow (3X DSA), both generating patterns with equal resolution. It reveals that the all-guided pattern of 1X DSA leads to more uniform and lower line roughness, as well as faster assembly kinetics. As a consequence of the fast assembly kinetics, DSA-specific dislocation defects could be eliminated from EUV + DSA patterning. Furthermore, the pattern rectification ability of EUV + DSA with respect to the EUV lithography guide pattern was studied by examining the EUV defect healing probability of DSA. Artificial EUV defects provided a practical study case. DSA can effectively restore the line space pattern for all EUV stochastic issues and other EUV defects up to three periods in size within three minutes. Contributing significantly to the healing of EUV defects and thus limiting defects in the BCP pattern are similar BCP and guide pattern pitch, higher BCP annealing temperatures and increased BCP guiding strength. Another critical parameter is again the BCP film thickness. Increasing the BCP film thickness eliminates all investigated artificial EUV defects in the BCP pattern. For EUV + DSA to be viable for implementation in high-volume manufacturing, the throughput of this process flow must still be increased. Resist-assisted DSA appears promising, and one successful guide pattern for PS-b-PMMA DSA patterning is using a thin layer of patterned EUV Multi-Trigger Resist, neutral toward both blocks, on top of a cross-linked polystyrene underlayer. Overall, while the specifications for high-volume manufacturing have not yet been achieved, several pathways for improvement of the patterning performance and throughput are presented. These learnings can be applied to the design of EUV + DSA for future patterning applications.All Dates
- 2026-03-23 17:00
Powered by iCagenda
