Nanoscale Periodic Modulation of Doping over Large Areas with Block Copolymer Lithography and Ion Implantation

Abstract

Whenever periodic patterning at the nanoscale over a large surface is required, block-copolymers (BCPs) represent an extremely attractive alternative for lithography1 and nano-templating,2 because of the low cost, if compared to conventional photolithography,3 and the high throughput, if compared to serial lithographic processes, such as electron beam lithography (EBL).4 Despite the relatively simple structure of BCPs, two different linear homopolymers linked to each other at one end by a covalent bond, when annealed above the glass transition temperature, they spontaneously microphase separate generating a variety of periodic nanostructures. The periodicity (L0 ) of the microdomains can be varied in the 10 – 100 nm range by properly adjusting the molecular weight and the interaction parameter of the two blocks.5 In this work, the combination of BCP lithography and ultra-low energy implantation of phosphorus ions at high fluences is investigated to promote a periodic modulation at the nanoscale of the concentration of dopant impurities over the near-surface layer of a silicon substrate.For the fabrication of the masks, two different poly(styrene-b-methyl methacrylate) (PS-b-PMMA) BCPs were used, to obtain two different geometries: the first with out of plane hexagonally packed PMMA cylinders, with an average diameter of 22 nm and center-to-center distance of 35 nm (Figure 1A) , and the second with out of plane lamellae having a periodicity of 28 nm (Figure 1B). The thickness of the BCP films was 35 nm thick. In addition, a graphoepitaxy protocol was implemented to obtain different sets of perfectly aligned 20 μm long parallel lamellae. A mesoporous soft mask can be readily obtianed by selective removal of PMMA cylinders with UV light and acetic acid. Conversely, a different process is required to avoid the collapse of the PS lamellae. Using Sequential Infiltration Synthesis (SIS),6 Al2O3 is incorporate into the PMMA phase of the BCP film. Upon the removal of the organic material by a mild O2 plasma cleaning, a nanostructured Al2O3 layer, that perfectly matches the PMMA pattern, remains on the substrate (Figure 1B).Phosphorus ions were implanted through the soft masks into the Si substrate operating at ultra-low energy (3 keV) to prevent the degradation of the polymeric film and reduce P penetration (10 nm) into the Si substrate. Additionally, high implantation doses are considered, to induce local amorphization of the silicon substrate. In this way Solid Phase Epitaxial Regrowth (SPER) can be exploited to recover the crystallinity of the substrate by thermal treatments at relatively low temperatures.7 During the SPER process the phosphorus atoms are substitutionally incorporated into the silicon lattice preserving their spatial confinement.AFM morphological analysis and ToF-SIMS depth profiling of the mask confirm the capability of the PS and alumina matrix to properly shield the low energy phosphorus ions in the dose range under investigation. AFM and ToF-SIMS characterization of the samples upon removal of the mesoporous template demonstrated that the P ions were effectively implanted into the Si substrate through the mask, leading to localized implantation (Figure 1C - 1E). Raman spectra suggested the presence of a thin layer of amorphous Si in the implanted samples and subsequent recrystallization after the annealing. By specific SPM measurements, it was possible to compare the morphology, the map of surface potential and the map of conductivity of the samples. The analysis of the Si surface after implantation through the mask with cylindrical pores and activation of the P is reported (Figure 1F, 1G). Small variations of the surface potential that perfectly match the implanted regions were obtained, suggesting a local modification of the electrical properties of Si. This result is also supported by conductivity measurements and finite element method (FEM) simulations. Further results will be presented, to investigate the electrical properties of the samples as a function of the implanted P dose and of the geometry of the mask.(1) Feng, H. et al., Nat. Mater. 21, 1426–1433 (2022).(2) Chai, J. & Buriak, J. M., ACS Nano 2, 489–501 (2008).(3) Levinson, H. J., Jpn. J. Appl. Phys. 61, (2022).(4) Yang, X. M. et al., ACS Nano 3, 1844–1858 (2009).(5) Seguini, G. et al. Soft Matter 16, 5525–5533 (2020).(6) Tseng, Y. C. et al. J. Phys. Chem. C 115, 17725–17729 (2011).(7) Luce, F. P. et al., Nucl. Instruments Methods Phys. Res. Sect. B Beam Interact. with Mater. Atoms 370, 14–18 (2016). Figure 1