Chemo-epitaxy Directed Self-assembly Research Articles

Self-Aligned Assembly of a Poly(2-vinylpyridine)-b-Polystyrene-b-Poly(2-vinylpyridine) Triblock Copolymer on Graphene Nanoribbons.

Directed self-assembly (DSA) of block copolymers is one of the most promising patterning techniques for patterning sub-10 nm features. However, at such small feature sizes, it is becoming increasingly difficult to fabricate the guiding pattern for the DSA process, and it is necessary to explore alternative guiding methods for DSA to achieve long-range ordered alignment. Here, we report the self-aligned assembly of a triblock copolymer, poly(2-vinylpyridine)-b-polystyrene-b-poly(2-vinylpyridine) (P2VP-b-PS-b-P2VP) on neutral graphene nanoribbons with the gap consisting of a P2VP-preferential silicon oxide (SiO2) substrate via solvent vapor annealing. The assembled P2VP-b-PS-b-P2VP demonstrated long-range, one-dimensional alignment on the graphene substrate in a direction perpendicular to the boundary of the graphene and substrate with a half-pitch size of 8 nm, which greatly alleviates the lithography resolution required for traditional chemoepitaxy DSA. A wide processing window is demonstrated with the gap between graphene stripes varying from 10 to 100 nm, overcoming the restriction on widths of guiding patterns to have commensurate domain spacing. When the gap was reduced to 10 nm, P2VP-b-PS-b-P2VP formed a straight-line pattern on both the graphene and the substrate. Monte Carlo simulations showed that the self-aligned assembly of the triblock copolymer on the graphene nanoribbons is guided at the boundary of parallel and perpendicular lamellae on graphene and SiO2, respectively. Simulations also indicate that the swelling of a system allows for rapid rearrangement of chains and quickly anneal any misaligned grains and defects. The effect of the interaction strength between SiO2 and P2VP on the self-assembly is systematically investigated in simulations.

Understanding Kinetics of Defect Annihilation in Chemoepitaxy-Directed Self-Assembly.

Directed self-assembly (DSA) of block copolymers (BCP) has attracted considerable interest from the semiconductor industry because it can achieve semiconductor-relevant structures with a relatively simple process and low cost. However, the self-assembling structures can become kinetically trapped into defective states, which greatly impedes the implementation of DSA in high-volume manufacturing. Understanding the kinetics of defect annihilation is crucial to optimizing the process and eventually eliminating defects in DSA. Such kinetic experiments, however, are not commonly available in academic laboratories. To address this challenge, we perform a kinetic study of chemoepitaxy DSA in a 300 mm wafer fab, where the complete defectivity information at various annealing conditions can be readily captured. Through extensive statistical analysis, we reveal the statistical model of defect annihilation in DSA for the first time. The annihilation kinetics can be well described by a power law model, indicating that all dislocations can be removed by sufficiently long annealing time. We further develop image analysis algorithms to analyze the distribution of dislocation size and configurations and discover that the distribution stays relatively constant over time. The defect distribution is determined by the role of the guiding stripe, which is found to stabilize the defects. Although this study is based on polystyrene-b-poly(methyl methacrylate) (PS-b-PMMA), we anticipate that these findings can be readily applied to other BCP platforms as well.

Block copolymer directed self-assembly defect modes induced by localized errors in chemoepitaxial guiding underlayers: A molecular simulation study

Block copolymer (BCP) directed self-assembly (DSA) has been presented as a potential economically attractive enhancement to extend the capabilities of optical lithography for semiconductor manufacturing. One concern in DSA is the level of defectivity that can be achieved in such a process. Although entropic effects will always lead to some degree of defectivity, highly ordered structures with a low theoretical equilibrium defect density can be produced by guiding the ordering and placement of the BCP domains using a patterned underlayer. Recent experimental studies have shown that while DSA processes can significantly reduce the observed defect density, defectivity levels are generally still higher than allowable for high-volume manufacturing and higher than what would be anticipated from free energy estimates of the observed defect modes. In particular, bridge defects are one of the most commonly observed defect modes in experimental DSA studies. A number of hypotheses have been proposed to explain the origins of these defects. One hypothesis is that so-called affinity defects present in the underlayer can spawn bridge defects in the overlying BCP film. The goal of the work reported here was to investigate the extent to which bridge defects can be generated or further reinforced in lamellae-forming block copolymer films due to affinity defects in the underlayer pattern. Coarse-grained molecular dynamics simulations were used to simulate the chemoepitaxial DSA of monodisperse block copolymer films atop underlayers with varying affinity defect sizes. Affinity defects were simulated by creating circular regions of a single polymer block type (which is the opposite block type of that used to pattern the underlayer guiding stripes) in the nominally neutral background region of the underlayer. These affinity defects were positioned in regions of the underlayer where they were the incorrect type to match the overlying block copolymer pattern. It was observed that the presence of an affinity defect in the neutral region of the underlayer caused the energetically preferential polymer block to wet the affinity defect, thus creating the nucleus of what could potentially become a bridge defect—even when the affinity defects were very small. As the radius of the underlayer affinity defect (RoD) increased, the amount of block copolymer of incorrect type (with respect to a perfectly assembled copolymer pattern) that assembled above the affinity defect increased; but, in general, the thickness of the wetting layer in contact with the affinity defect was only roughly one polymer chain thick. These data suggest that an affinity defect in the underlayer alone is unlikely to be noticeably enhanced by significant bridge defect formation in a monodisperse block copolymer film.

Strategies for Increasing the Rate of Defect Annihilation in the Directed Self-Assembly of High-χ Block Copolymers.

Directed self-assembly (DSA) of high-χ block copolymer thin films is a promising approach for nanofabrication of features with length scale below 10 nm. Recent work has highlighted that kinetics are of crucial importance in determining whether a block copolymer film can self-assemble into a defect-free ordered state. In this work, different strategies for improving the rate of defect annihilation in the DSA of a silicon-containing, high-χ block copolymer film were explored. Chemo-epitaxial DSA of poly(4-methoxystyrene-block-4-trimethylsilylstyrene) with 5× density multiplication was implemented on 300 mm wafers by using production level nanofabrication tools, and the influence of different processes and material parameters on dislocation defect density was studied. It was observed that only at sufficiently low χN can the block copolymer assemble into well-aligned patterns within a practical time frame. In addition, there is a clear correlation between the rate of the lamellar grain coarsening in unguided self-assembly and the rate of dislocation annihilation in DSA. For a fixed chemical pattern, the density of kinetically trapped dislocation defects can be predicted by measuring the correlation length of the unguided self-assembly under the same process conditions. This learning enables more efficient screening of block copolymers and annealing conditions by rapid analysis of block copolymer films that were allowed to self-assemble into unguided (commonly termed fingerprint) patterns.

Path to Move Beyond the Resolution Limit with Directed Self-Assembly.

Directed self-assembly (DSA) of block copolymers (BCPs) has long been viewed as a powerful alternative to extend the resolution of optical lithography. For full-area patterning applications, despite significant progress, the two most prominent DSA methods (chemoepitaxy and graphoepitaxy) are facing a scalability challenge: the critical dimension (CD) of the guiding patterns will need to be continuously scaled down to closely match the dimension of the BCP microdomain, a task that not only contravenes some of the resolution gains achieved by density multiplication but that will also become particularly difficult below 10 nm. To avoid this conundrum, we propose here a synergistic integration of graphoepitaxy and chemoepitaxy through self-registered self-assembly (SRSA) to enable the simultaneous realization of feature density multiplication and CD shrinkage resolution gains. We report nearly perfect DSA on prepatterns with highdensity multiplication factors and CD of several multiples of the BCP microdomain size. A prepattern consisting of alternating stripes of a relatively thicker neutral mat and a thinner neutral brush with preferential wetting sidewalls serves as a topographic pattern to guide an ultrathin BCP blend film inside the trenches. As the oriented BCP pattern assembles, the blend film deploys a layer of chemical markers on the bottom surface through SRSA generating 1:1 chemical contrast patterns inside the trenches. After thorough removal of the blend film, the newly formed self-registered chemical patterns interpolated by the remaining neutral mat strips serve as the guiding patterns for a second chemoepitaxial DSA step to achieve full-area, defect-free DSA of thick BCP films.

Engineering the Kinetics of Directed Self-Assembly of Block Copolymers toward Fast and Defect-Free Assembly.

Directed self-assembly (DSA) of block copolymers (BCPs) can achieve perfectly aligned structures at thermodynamic equilibrium, but the self-assembling morphology can become kinetically trapped in defective states. Understanding and optimizing the kinetic pathway toward domain alignment is crucial for enhancing process throughput and lowering defectivity to levels required for semiconductor manufacturing, but there is a dearth of experimental, three-dimensional studies of the kinetic pathways in DSA. Here, we combined arrested annealing and TEM tomography to probe the kinetics and structural evolution in the chemoepitaxy DSA of PS- b-PMMA with density multiplication. During the initial stages of annealing, BCP domains developed independently at first, with aligned structures at the template interface and randomly oriented domains at the top surface. As the grains coarsened, the assembly became cooperative throughout the film thickness, and a metastable stitch morphology was formed, representing a kinetic barrier. The stitch morphology had a three-dimensional structure consisting of both perpendicular and parallel lamellae. On the basis of the mechanistic information, we studied the effect of key design parameters on the kinetics and evolution of structures in DSA. Three types of structural evolutions were observed at different film thicknesses: (1) immediate alignment and fast assembly when thickness < L0 ( L0 = BCP natural periodicity); (2) formation of stitch morphology for 1.25-1.45 L0; (3) fingerprint formation when thickness >1.64 L0. We found that the DSA kinetics can be significantly improved by avoiding the formation of the metastable stitch morphology. Increasing template topography also enhanced the kinetics by increasing the PMMA guiding surface area. A combination of 0.75 L0 BCP thickness and 0.50 L0 template topography achieved perfect alignment over 100 times faster than the baseline process. This research demonstrates that an improved understanding of the evolution of structures during DSA can significantly improve the DSA process.

Defect annihilation in chemo-epitaxial directed self-assembly: Computer simulation and Self-Consistent Field Theory

ABSTRACTExcept at the order-disorder transition, defects in lamella-forming block copolymers have a free energy of several hundreds kBT where kBT denotes the thermal energy scale. Thus, they cannot be conceived as equilibrium fluctuations around a perfectly ordered state. Instead, defects, which are observed in experiments, are formed in the course of self-assembly. Their behavior is dictated by the kinetics of structure formation, in particular, the kinetic pathways of defect motion and annihilation.Computational modeling can contribute to optimize materials parameters such as film thickness, interaction between copolymer blocks and substrate, geometry of confinement, in order to avoid the formation of defects in the early stages of structure formation or facilitate defect annihilation. Computations also provide fundamental insights into the universal physical mechanisms of directing the self-assembly, addressing the equilibrium structure and thermodynamics as well as the kinetics of self-assembly.We present computer simulation of highly coarse-grained particle-based models and self-consistent field calculations that allow us to access the long time and large length scales associated with self-assembly. These calculations provide information about the free-energy landscape and mechanisms of defect annihilation in thin films. Additionally, opportunities for directing the kinetics of self-assembly by temporal changes of thermodynamic conditions are discussed.

28 nm pitch of line/space pattern transfer into silicon substrates with chemo-epitaxy Directed Self-Assembly (DSA) process flow

Directed Self-Assembly (DSA) using block-copolymer (BCP) materials is considered as an alternative to Extreme Ultra-Violet (EUV) lithography for sub-14nm half pitch technology. A BCP with natural periodicity of 28nm was assembled on 84nm pitch pre-pattern. 3X feature multiplication can be achieved with the chemo-epitaxy DSA flow. The BCP thickness should be thin in order to reduce the self-assembly defectivity. As a consequence, plasma etch process to transfer such thin polymer material to subsequent underlayer becomes a challenging task. The main purpose of this paper is to study the DSA pattern transfer capability of 20–25nm BCP layer aiming at silicon fin patterning through the Si3N4 hardmask. Alternative Ar/O2 and Ar plasmas were developed to increase the selectivity of PMMA to PS, and as well as reducing hydrogen passivation plasma for neutral brush layer etch. A SiCl4 encapsulation process was introduced to increase the selectivity towards Polystyrene during the Si3N4 hardmask patterning. Finally, DSA full pattern transfer to create 14nm half-pitch line trenches with the silicon depth of 67nm was achieved. Line width roughness (LWR) and line edge roughness (LER) of 2.56nm and 2.64nm respectively were achieved after PMMA removal. This DSA-based 14nm half pitch line and space enable the building of fin structures meeting future 7nm-technology node (N7) design rules.