Most well-known litho-simulation packages work with Manhattan unit cells to exploit the advantages of fast Fourier transforms. Simple 1D features with arbitrary orientations may lead to the fact that quite a large 2D area is needed to describe those features. The obvious consequences are that the feature is approximated by a staircase and that the pitches are related to the angle of rotation. The latter leads often to an increase in computation time, especially when the topography of the mask is taken into account. An obvious way of avoiding that is to rotate the features and all pupil functions or lens properties to x- or y-axes. However, for anamorphic systems, where the demagnifications are fixed in absolute space, this is not straightforward and several actions have to be taken. We described how to do this exactly. It will be shown that the aerial images from a full 2D simulation compare well with that of an equivalent 1D version.

Alternative patterning solutions, such as spacer-based pitch splitting, have been a cornerstone of advanced technology nodes to enable device scaling. The greatest utility comes from the ability to self-align a pitch splitting process; however, traditional spacer-based patterning techniques require the deposition and etch of multiple materials, which reduce throughput and increase manufacturing costs. Anti-spacer technology, on the other hand, enables both self-aligned pitch splitting and high throughput via a single pass track-based process. We will describe the advancement of 193i anti-spacer technology to pattern trench dimensions beyond the critical dimension resolution of single-print extreme ultraviolet lithography and the utility of combining anti-spacer patterning with litho-freeze-litho-etch to enable the formation of sub-20-nm slot contact features for a minimum tip-to-tip (T2T) cut, with a roadmap to achieve sub-12 nm. A through process performance evaluation was conducted to further the understanding of fundamental process parameters and their associated effects on anti-spacer roughness and critical dimension uniformity. Such variables include photoresist, developer optimization, and overcoat dissolution. At pitches varying from 50 to 80 nm, we have demonstrated narrow trench widths down to 11.8 nm, which corresponds to the critical T2T dimension. Through hardmask etch transfer, we observe a 56% improvement in unbiased space width roughness and pitch-walking below 0.3 nm at 60-nm pitch.

With high-numerical aperture extreme ultraviolet (EUV) exposure tools soon to be available for utilization by chip makers, the factors that could limit the useable resolution of these systems are considered, to address these potential limitations and thereby enable patterning close to the theoretical optical resolution limit. Extended volumes of solubility-switched resist around the absorption sites of individual photons will limit resolution and lead to large line-edge roughness. If the volumes of chemically converted resist are small, there need to be many such volumes to provide continuous paths for development. Hence, the matters of exposure dose and molecular considerations are not fully separable. New resist architectures will be needed to avoid extremely high exposure doses to pattern features below 10-nm half-pitch. More energy-efficient EUV light sources will be needed. Computational lithography must account for a multiplicity of issues, necessitating increased automation of mask pattern optimization.

High-NA extreme ultraviolet lithography (EUVL) at 0.55 NA provides the resolution for single-exposure patterning of devices at 2-nm technology node and beyond. In a single exposure, the feature sizes in the 4× direction on the high-NA masks are tighter than the minimum dimensions imageable at 0.33 NA. We present a general discussion of the need for actinic inspections and then a detailed description of actinic pattern mask inspection (APMI) and actinic blank inspection (ABI), two critical capabilities for the EUV mask infrastructure. The advantages of actinic inspection are to detect all types of defects on blanks and masks with and without a pellicle. The recently developed APMI capabilities for high-NA EUV mask inspections offer higher resolution, image contrast, and defect sensitivity. The high-NA APMI is significantly upgraded from the platform with optics, EUV light source, and detector. The high-NA APMI has been released in 2023, and we presented the inspection results. Equally important is the ABI capability for high-NA mask blanks. The upgrade from the current platform includes a Schwarzschild objective with higher magnification that enhances defect signals in the dark field and improved optics with better imaging in the bright field for defect review. The new ABI provides not only higher sensitivity to multilayer phase defects but also an improved defect location accuracy, both are required for blank inspection in high-NA EUVL applications.

To achieve higher resolution extreme ultraviolet lithography (EUVL) targeted toward sub-10 nm, reflective projection scanner image numerical apertures (NAi) are being increased beyond the current value of 0.33 to 0.55 and upward of 0.75 as a desirable target. Bragg reflectors using alternating silicon and molybdenum that have heretofore been coated as periodic multilayers cannot achieve desired reflected amplitudes as corresponding 0.25× mask numerical apertures (NAm) are accordingly increased. In addition, transverse magnetic-polarized image modulation decreases with NA, which becomes significant at 0.55 and above. We present here the optimization of non-regular alternating, or aperiodic, silicon-molybdenum multilayer reflective coatings that can achieve improved amplitude and polarization performance through angle as higher-NA EUVL lithography is pursued. Through the use of rigorous EM computation paired with a genetic optimization method, we show that amplitude apodization can be recovered to 60% peak reflectance for NAm values up to 0.2 (corresponding to NAi values of 0.8) while at the same time achieving a transverse electric degree of polarization exceeding 40%. In addition, aperiodic multilayers optimized for spectral bandwidth using refractory metals also show improvement over periodic designs.

At this moment, extreme ultraviolet (EUV) systems equipped with a 0.33 numerical aperture (NA) have proven themselves and are successfully applied in high-volume manufacturing. The next step is 0.55 NA and is ready to enter mass production. This so-called high NA scanner, targeting an ultimate resolution of 8 nm half-pitch, will bring multiple benefits to the semiconductor market such as reduction of process complexity, yield improvement, higher resolution enabling printability of smaller features at increased density, and cost of technology reduction. It will extend Moore’s law for at least another decade. A lens design, capable of providing the required NA, has been identified; this so-called anamorphic lens will provide 8 nm resolution in all orientations. Paired with new, faster stages, and more accurate sensors providing the tight focus and overlay control, it enables future nodes. The first 0.55 NA scanner is located in the so-called high NA Lab in Veldhoven where it is interfaced with a track and operated in cooperation with Imec, Leuven. It also allows for early customer access. We will provide the backgrounds of the architecture of the high NA tool. Next to this, an update will be given on the status of the imaging and overlay performance of this exposure tool.

We investigate the benefit of sub-resolution assist features (SRAF) in improving the performance of potential depth of focus (DOF)-limited logic-like layers in 0.55 numerical aperture (NA) lithography. Our study involves a systematic evaluation of generic lines and spaces through pitch, as well as larger clips using the current Tachyon source mask optimization tool in both horizontal and vertical orientations for a darkfield tantalum boron nitride (TaBN) mask. We assess the robustness of the SRAF solution with respect to global processes and local mask variations. The impact of the variations is quantified using key lithography metrics, such as DOF and overlapping process window and process variation bands. We conclude that the presence of SRAFs reduces the impact of process and mask variability, and the benefit is orientation-dependent.