Editor-in-Chief Harry Levinson outlines new guidelines on research involving patterning tools.

In lithography, optical proximity and process bias/effects need to be corrected to achieve the best wafer print. Efforts to correct for these effects started with a simple bias, adding a hammer head in line-ends to prevent line-end shortening. This first-generation correction was called rule-based optical proximity correction (OPC). Then, as chip feature sizes continued to shrink, OPC became more complicated and evolved to a model-based approach. Some extra patterns were added to masks, to improve the wafer process window, a measure of resilience to manufacturing variation. Around this time, the concept of inverse lithography technology (ILT), a mathematically rigorous inverse approach that determines the mask shapes that will produce the desired on-wafer results, was introduced. ILT has been explored and developed over the last three decades as the next generation of OPC, promising a solution to several challenges of advanced-node lithography, whether optical or extreme ultraviolet (EUV). Today, both OPC and ILT are part of an arsenal of lithography technologies called resolution enhancement technologies. Since OPC and ILT both involve computation, they are also considered as part of computational lithography. We explore the background and history of ILT and detail the significant milestones that have taken full-chip ILT from an academic concept to a practical production reality.

Guest Editors Martin Burkhardt and Vicky Philipsen summarize the papers of the Special Series on EUV Masks.

Background: Extreme ultraviolet (EUV) phase shifting masks (PSMs) may be introduced at the N3 node as alternative to double patterning or high-NA exposures. To be optically advantageous, PSMs may have to come with phase different from 180 deg and high transmission.

Aim: The main concern is how such exposures could be affected by stochastic side-lobe printing.

Approach: We present a simulation study where we compare stochastic distributions for contact layers with different mask approaches: binary mask, PSM with and without assist features, and TriTone masks. The purpose of this work is to evaluate whether PSMs in EUV lithography may become a valuable option to further extend single-exposure EUV at NA  =  0.33.

Results: We identify the best exposure conditions for lowest stochastic defectivity for each of the masks considered and perform a large-scale study at these conditions. We report CD and placement error distribution parameters, along with frequencies of observed defects.

Conclusions: We find that PSMs are more limited in range of CDs that can be printed without stochastic defects, compared with binary masks. When compared at target CD that is optimal for PSM, proposed phase shift masks did not exhibit a clear advantage, suggesting careful consideration of stochastic defects will be needed during future process development.

Background: An extreme ultraviolet (EUV)-transparent pellicle must be used during lithography to protect the photomask from fall-on particles. A pellicle made of free-standing carbon nanotube (CNT) films stops particles despite the presence of gaps while demonstrating high EUV transmission, mechanical stability, low EUV scattering and reflectivity, and DUV transmission that enables through-pellicle mask inspection. Aim: The CNT EUV pellicle properties can be tailored based on the diversity of CNT structures and tunability of their configuration within the CNT film (density, bundle size, composition, etc.) as shown in this work. A remaining challenge is extending the CNT EUV pellicle lifetime in the scanner environment of EUV-induced hydrogen-based plasma, and the effects on different CNT films are explored here. Approach: Optical and thermal properties of different CNT pellicles with respect to the CNT material type, density, composition, and bundle size were explored. The ability of uncoated CNT EUV pellicles to withstand high EUV powers in the hydrogen-based environment was tested. Transmission, spectroscopic, and chemical mapping of the exposed CNT membranes were performed to explore the material modifications under various exposure conditions. Results: Uncoated CNT pellicles withstand 600-W source power equivalent in the EUV scanner-like gas environment but exhibit structural changes with prolonged exposure. Multiwalled CNT pellicles exhibit less EUV transmission change as compared to single-walled CNT pellicles under the same exposure conditions. The protection of CNT material from structural degradation by means of coating was shown. Conclusions: These investigations add to the understanding of CNT EUV pellicle tunability for optimal performance and lifetime limiters of CNT pellicles under the influence of EUV radiation and plasma. We anticipate the need for coating the CNT pellicle to protect the CNT material against plasma damage for the current scanner conditions. Optimization of both the CNT membrane and its coating is in progress.

With growing interest in EUV attenuated phase-shift masks due to their superior image quality for applications such as dense contact and pillar arrays, it is becoming critical to model, measure, and monitor the intensity and relative phase of multilayer and absorber reflections. We present a solution based on physical modeling of reflectometry data, which can achieve single picometer phase precision and sensitivity to changes in average film thickness below one atomic monolayer. We measure absorber and multilayer reflectivity to determine thin-film parameters with a multidimensional optimization and then acquire a new measurement of either multilayer or absorber to determine perturbations in surface contamination thickness. While it is difficult to assess the accuracy of the first step, the simplicity of the second step allows us to characterize our sensitivity to changes in contamination thickness. We apply this analysis using an initial set of measurements and repeated measurements after a period of storage. For the multilayer, the total contamination growth was 1068 pm, which occurred almost exclusively during storage (1085 pm) and decreased very slightly during repeated measurements (−17  pm). For the absorber, the behavior was quite different, with a total growth of 126 pm, which occurred much less during storage (28 pm) and primarily during repeated measurements (98 pm). Ultimately, the change in relative phase (absorber minus multilayer) was −0.86  deg for the multilayer and −1.12  deg for the absorber. We estimate the precision of the surface contamination measurement to be 3σ  <  6  pm for measuring thickness and 3σ  <  0.2  deg for measuring phase.

Background: Repeated exposure of extreme ultraviolet (EUV) masks to UV radiation exacerbates the thermal deformation of the constituent materials. In particular, for systems with high numerical apertures (NAs) in next-generation EUV lithography machines, new structures and materials are needed to negate the shadow and three-dimensional effects of masks. However, because masks with such materials and structures have larger absorption and thermal expansion coefficients than conventional EUV masks, they may undergo new, unidentified thermal deformations.

Aim: This study aimed to evaluate the temperature accumulation and thermal deformation of EUV masks owing to radiation exposure.

Approach: The temperature accumulation and thermal deformation in EUV masks were compared and analyzed using various cooling conditions and patterns as well as various materials and structures.

Results: The thermal deformation increased with increasing pattern density. The results indicate that more careful control could be required regarding the pattern density and shape. Compared with conventional EUV masks, the high-NA EUV masks exhibited considerable thermomechanical deformations owing to their different multilayers and absorbers. However, the deformation could be reduced to the level of conventional EUV mask deformation through appropriate cooling.

Conclusions: The thermal deformation due to exposure is intensified depending on the mask structure, material, and pattern change; however, the thermal deformation can be alleviated with proper mask cooling.

Mask repair is an essential step in the manufacturing process of extreme ultraviolet (EUV) masks. Its key challenge is to continuously improve resolution and control to enable the repair of the ever-shrinking feature sizes on mask along the EUV roadmap. The state-of-the-art mask repair method is gas-assisted electron-beam (e-beam) lithography also referred to as focused electron-beam induced processing (FEBIP). We discuss the principles of the FEBIP repair process, along with the criteria to evaluate the repairs, and identify the major contributions determining the achievable resolution. As key results, we present several high-end repairs on EUV masks including a sub-10-nm extrusion achieved with the latest generation of e-beam-based mask repair tools, the MeRiT^® LE. Furthermore, we demonstrate the corresponding repair verification using at-wavelength (actinic) measurements.

Our work presents and investigates the effectiveness of a model-based proximity effect correction method for helium ion beam lithography (HIBL). This method iteratively modulates the shape of a pattern by a feedback compensation mechanism until the simulated patterning fidelity satisfies specific constraints. A point spread function (PSF) is utilized to account for all phenomena involved during the scattering events of incident ion beam particles in the resist. Patterning prediction for subsequent correction process is derived from the energy intensity distribution due to convolution between the PSF and the pattern, with an adequate cut-off threshold. The performance of this method for HIBL is examined through several designed layouts from 15- to 5-nm in half pitches, under specific process parameters, including acceleration voltage, resist thickness, and resist sensitivity. Preliminary results show its effectiveness in improving the patterning fidelity of HIBL.

Background: Fast advancing of microwave/terahertz communication constantly demands ultrashort T-shaped gates by innovative nanofabrication techniques. The emerging RE650 resist provides a promising alternative to existing resists owing to its outstanding performance in electron-beam (e-beam) lithography.

Aim: To report the exploration of the minimum foot-width achievable by e-beam lithography on RE650/UV5 bilayer resists through both numerical simulation and experimental tests.

Approach: Using Monte Carlo method, the spatial charge distributions by e-beam exposure in resists were calculated, followed by modeling the developing process using LAB software to obtain the lithography profiles in resists. By this way, the foot-width was compared for the RE650 thickness varying from 50 to 80 nm with various UV5 thickness fixed.

Results: As narrow as 15-nm foot-width has been achieved, but 30-nm foot-width can be routinely replicated in a stable exposure latitude, indicating that the developed process for ultra short T-shaped gates is reliable.

Conclusions: The bilayer of RE650/UV5 is capable of replicating 15-30-nm T-shape gates as the minimum width by e-beam lithography, which should give us high prospect for mass production of THz devices.

In subwavelength lithography, printed patterns on the silicon wafer suffer from geometric distortions and differ from the original design. These nonrectangular patterns can seriously affect electrical characteristics and circuit performances. We extend the verification of location-dependent weighting method and further propose three single equivalent gate length (EGL) extraction methods for representing each nonrectangular gate (NRG) transistor with a single EGL model. These methods are applied to sub-20-nm fully depleted silicon on insulator (FDSOI) circuits to predict the postlithography performances. An in-house extreme ultraviolet lithography simulation tool is utilized for nonrectangular pattern simulation. Shape information is imported to TCAD to construct three-dimensional nonrectangular FDSOI transistor models. The accuracy of the location-dependent weighting method and EGL extraction methods is verified with TCAD circuit simulations. Preliminary simulation results indicate that weighting factors can improve the accuracy of electrical characteristics estimation, especially in leakage current analysis. On average, the EGLs extracted from off-state only data, and from data lumping both off- and on-states, respectively, can each predict SRAM electrical characteristics with overall error <1  %  , or a factor of 5 accuracy improvement over the EGLs extracted without the weightings. These methods could be used to simulate large-scale sub-20-nm FDSOI circuits with NRG transistors caused by nonideal optical effects.

In recent years, EUV lithography scanner systems have entered high-volume manufacturing for state-of-the-art integrated circuits, with critical dimensions down to 10 nm. This technology uses 13.5-nm EUV radiation, which is shaped and transmitted through a near-vacuum H₂ background gas. This gas is excited into a low-density H₂ plasma by the EUV radiation, as generated in pulsed mode operation by the laser-produced plasma in the EUV source. Thus, in the confinement created by the walls and mirrors within the scanner system, a reductive plasma environment is created that must be understood in detail to maximize mirror transmission over the lifetime and to minimize molecular and particle contamination in the scanner. In addition to the irradiated mirrors, reticle, and wafer, the plasma and radical load to the surrounding construction materials also must be considered. We provide an overview of the EUV-induced plasma in the scanner context. Special attention is given to the plasma parameters in a confined geometry, such as that found in the scanner area near the reticle. It is shown that plasma confinement and resulting contributions from secondary electron emission delay the formation of the plasma sheath and thereby reduce the peak ion energies to below the sputtering threshold for mirrors and construction materials. Furthermore, for a confined pulsed plasma with a pulse period shorter than the decay time of the plasma, the plasma consists of a quasi-steady-state cold background plasma and periodic transient peaks in ion energy and ion flux. In terms of modeling, this means that no assumptions can be made on the electron distribution functions and a (Monte-Carlo) particle-in-cell (PIC) model is needed. We present an extension of the PIC model approach to complex three-dimensional geometries and to multiple pulses using a hybrid PIC-diffusion approach. Also, the translation of these specific plasma parameters to off-line setups and the aspects that must be included to make a meaningful translation from off-line laboratory EUV setups to the scanner plasma are discussed.

In this work, we present a sound mathematical result that may correct the pixelization bias in the calculation of the line edge roughness (LER) even when the pixel size of SEM images is much larger than its rms value. This result can be useful in the LER metrology of current and future lithographic patterns where the desired rms value of LER is in the deep sub-nm range and smaller than the used pixel sizes. We computationally justify the predictions of the mathematical result and then demonstrate its application in synthesized SEM images where we can separate the impact of noise and focus on the pixelization effects in LER measurements. Furthermore, this result emphasizes the power of mathematical and computational methods to enhance the metrological output of SEM images in deep nanometer scale.

Background: In datasets for hotspot detection in physical verification, data are predominantly composed of non-hotspot samples with only a small percentage of hotspot ones; this leads to the class imbalance problem, which usually hinders the performance of classifiers.

Aim: We aim to enrich datasets by applying a data augmentation technique.

Approach: We propose a data augmentation flow-based generative adversarial network (GAN) to generate high-resolution hotspot samples.

Results: We evaluated our flow with the current state-of-the-art convolutional neural network hotspot classifier by comparison with conventional data augmentation techniques. Experimental results demonstrate that the accuracy improvement of our work can reach 3% at the same false alarm rate and the false alarm rate reduction can reach 5% at the same accuracy.

Conclusions: Our study demonstrates that rational hotspot classification can improve the efficiency of data. It also highlights the potential of GAN to generate complicated layout patterns.

Background: Extreme ultraviolet (EUV) pellicles are affected by heat deformation due to energy absorption during EUV exposure. Defects can accelerate thermal deformation, thereby shortening pellicle lifetime.

Aim: We compared the thermal stress and thermomechanical stability with different defects, focusing on the internal defects that occur during the fabrication of EUV pellicles.

Approach: Pellicles resistant to high thermal stress can be fabricated based on mechanical properties. To evaluate this, we compared the mechanical stability based on robustness against thermal stress during exposure.

Results: Our results show that external contaminants had a greater contribution than internal defects on the mechanical stability of pellicles. However, pellicles with internal defects exhibited increased thermal stress and decreased mechanical stability compared with defect-free pellicles during EUV exposure.

Conclusions: Thermal stress is used as an indicator in most studies to predict pellicle lifetime, but evaluation of thermomechanical stability including thermal stress and mechanical durability is required because the material can break under low thermal stress depending on the mechanical properties, and pellicle defects can be a major cause of shortening the lifetime of pellicles.

Background: As target feature sizes for EUV lithography shrink, it is becoming ever more important to understand the intricate details of photoresist materials, including the role of the “third dimension”—the dimension perpendicular to the wafer. With resist thicknesses shrinking toward the single-digit nanometer scale alongside target linewidths, small changes in resist performance in this dimension will have a greater overall effect on pattern quality.

Aim: To use modeling to understand the effect that the third dimension has on resist performance, in particular the interplay between the third dimension and resist stochastics.

Approach: We developed a three-dimensional version of the multivariate Poisson propagation model, a stochastic resist simulator. As a test case for the model, we explore the role of acid diffusion in the so-called third dimension by simulating 10⁵ vias at a series of z-blur conditions.

Results: Our model suggests that increased z-blur yields an improvement in both dose to size and pattern uniformity without sacrificing resolution.

Conclusions: We have developed a 3D resist model that can simulate large numbers of contacts. Early results from the 3D model show improved patterning performance can be achieved by increasing the z-blur.

Background: Polyphthalaldehyde (PPA)-based systems can be interesting candidates for extreme ultraviolet (EUV) lithography as dry development resists with simple chemistry.

Aim: We present EUV-induced mechanistic and contrast curve studies for the end-capped PPA.

Approach: Fourier transform infrared (FTIR) spectroscopy and desorption studies were conducted to understand the EUV-induced mechanistic pathway. Although, contrast curve analysis was used to check its feasibility for EUV-patterning.

Results: First, FTIR and desorption studies confirmed that the EUV-radiation is capable to remove the end-capping group and induce unzipping (direct depolymerization) reaction in the PPA polymer chain. Second, contrast curve analysis showed a gradual decrease in the film thickness with respect to the EUV dose, which is due to the low EUV sensitivity of the PPA polymer. A post exposure bake step is important to improve the contrast curve, as it helps to depolymerize any residual polymer remaining after exposure. Further, even though the polymer sublimates when exposed to the EUV radiation, eliminating the need to apply a wet development step, crosslinking at high doses causes deposition of residual film onto the wafer surface. Therefore, a thin film (below 20 nm) and a wet development process might be important to get clean patterns.

Conclusion: This study confirms that the end-capped PPA, with a simple patterning mechanism, can be a useful system for the development of new dry development resists for EUV lithography.

The success in the miniaturization of the electronic device constituents depends mostly on the photolithographic techniques. Recently, to achieve patterning at the sub-10-nm node, extreme ultraviolet (EUV) lithography has been introduced into high volume production. Continued scaling of EUV via increased numerical aperture to achieve nodes at 3-nm and below requires the development of fundamentally new patterning materials and new characterization methods. Current EUV-resist film thicknesses are in the 20- to 40-nm range, and further thickness reduction is required for the next generation. Therefore, interfaces become exceedingly important, and the properties of the resist film would be dominated by top and bottom interfacial effects. X-ray photoelectron spectroscopy (XPS) combined with standing-wave excitation (SW-XPS), a fairly new method in the EUV lithography field, previously had been largely applied in multilayers and superlattices for characterizing the composition and electronic structure of buried layers and interfaces as a function of depth. We applied the SW-XPS method to organic/inorganic photoresists to provide depth-selective information on their structural and chemical conditions of as a function of temperature, EUV exposure, different underlayers, and other fundamental parameters. As a first attempt, we perform an SW-XPS feasibility study on self-assembled monolayer (SAM) films after exposure to an electron beam. By SW-XPS, we determined that the interface between the Al₂O₃ underlayer and the SAMs is smooth, with a mean roughness of about 0.2 nm. Moreover, we determined that the SAM chains are, on average, tilted by ∼30  deg off the sample normal. The SW-XPS results also suggest that the SAM is not a perfectly aligned and uniform monolayer, with some areas having thickness higher than a single monolayer. We demonstrated that SW-XPS can provide useful information on ultrathin materials with high potential for being used as a characterization method of organic/inorganic photoresists.