Atomic layer deposition assisted pattern transfer technology for ultra-thin block copolymer films
Source abstract
Abstract
Abstract As an emerging developing technique for next-generation lithography, directed self-assembly (DSA) of block copolymer (BCP) has attracted numerous attention and has been a potential alternative to supplement the intrinsic limitations of conventional photolithography. In this work, the self-assembling properties of a lamellar diblock copolymer poly(styrene-b-methylmethacrylate) (PS-b-PMMA, 22k-b-22k, L 0 = 25 nm) on Si substrate and an atomic layer deposition (ALD)-assisted pattern transfer technology for the application of DSA beyond 16/14 nm complementary metal oxide semiconductor (CMOS) technology nodes, were investigated. Firstly, two key processing parameters of DSA, i.e . annealing temperatures and durations of BCP films, were optimized to achieve low defect density and high productivity. After phase separation of BCP films, self-assembling patterns of low defect density should be transferred to the substrate. However, due to the nano-scale thickness and the weak resistance of BCP films to dry etching, it is nearly impossible to transfer the BCP patterns directly to the substrate. Therefore, an ALD-based technology was explored in this work, in which deposited Al 2 O 3 selectively reacts with PMMA blocks thus hardening the PMMA patterns. After removing PS blocks by plasma etching, hardened PMMA patterns were left and transferred to underneath SiO 2 hard mask layer. Using this patterned hard mask, nanowire array of 25 nm pitch were realized on Si substrate. From this work, a high-throughput DSA baseline flow and related ALD-assisted pattern transfer technique were developed and proved to have good capability with the mainstream CMOS technology.
Source: openaire · Quality: candidate unverified · Source record ↗
Research context
OpenAlex topics: Block Copolymer Self-Assembly · Ferroelectric and Negative Capacitance Devices · Advancements in Photolithography Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References
Reference metadata was not supplied for this record.