ArF immersion lithography has opened the road towards increased optical resolution at the 193nm wavelength. Consequently, keeping the same 4X optical demagnification factor, the dimensions on the mask scale down to sub-wavelength values when we enter the 45nm node. At such dimensions, mask topography, mask type and materials as well as the polarization state of the light will influence the diffraction spectrum of a layout. As a result the image from high NA lithographic systems depends on the polarization state and intensities of the interfering orders. In general, with smaller features on the mask stronger polarization changes occur. Apart from the polarization changes in diffraction orders the total intensity in a diffraction order is also different from that predicted by standard scalar-Kirchhoff diffraction approximation used in present OPC packages. The difference in intensities of diffraction orders due to different mask materials and topography is the more dominating factor leading to through pitch CD errors when the scalar-Kirchhoff model is used for layout adjustment. Based on findings and classification of topography induced effects, a deviation-driver from scalar diffraction model was identified. This paper discusses a solution to compensate for topography effects while using the scalar diffraction model for reticle treatment. The area of applicability of such a scalar model, its advantages and limitations are illustrated with simulations and experiments.
Through ArF immersion lithography a road towards increased optical resolution at the 193nm wavelength has been opened. According to recently proposed roadmaps, ArF immersion lithography will be used for 65nm and 45nm technology nodes. Consequently, keeping the same 4x optical demagnification factor, the dimensions on mask scale down to wavelength values when entering these nodes. Moreover CD control becomes tighter and approaches values of 2-3nm. At such conditions, topography on mask, its type and materials cannot be ignored anymore while evaluating image formation either for design analysis or OPC adjustments. The objective of this paper is to analyze the influence of mask topography on imaging. The mask topography influences polarization state and diffraction efficiencies, which are determine further image formation. Therefore these parameters and their dependence on mask type, materials and pitches are of the major concern during the analyses. We analyze the process latitude and CD variations through pitch. The complete rigorous analysis shows improved process windows with the increase of feature aspect ratio and at the same time a large through pitch CD deviation compared to the conventional Kirchhoff diffraction model.
Assessment for introduction of immersion lithography into volume manufacturing has recently started, where one of the key focus areas includes defectivity. Particularly, the possible presence of bubbles in the immersion liquid could act as a defect source. The impact of bubbles strongly depends on their size and distance from the resist. This paper shows that a thick topcoat acts as a pellicle and suppresses the printability of the bubbles. A 1.5 μm thick topcoat has been developed especially for this purpose. A model experiment has been set to validate this approach and leads to a conclusion on the printability of defects depending on their size and distance from the resist. Both simulation and results from the model experiment are shown. In addition, a new method to detect very small bubbles will be introduced.
Today we see that 248nm lithography is pushed to the region of very low k1-factors. The first 193nm systems are now on the market, but the technology needs still needs to mature before its optimum performance can be reached. On the other hand, development of 157nm systems has been started in order to push optical lithography to the 100nm and 70nm nodes. In this paper simulations are used to show how far optical lithography could be extended assuming mature tools and resists. The simulations are performed using Prolith/2 and Solid-C in combination with Monte Carlo calculations to predict ED-windows and CD control at 193nm and 157nm illumination. Different resolution enhancement techniques are invested for dense and isolated lines and contact holes: off-axis illumination, phase shifting masks and high NA settings. Once the optimum NA-sigma combinations for maximum process windows are determined, CD control is calculated by taking into account variations in focus, dose, reticle CD and phase and lens aberrations. From these CD control calculations the most important contributions to CD variations for the different RET can be identified, showing also where restrictions have to be put to obtain sufficient CD control.
Lithography at 193nm is the first optical lithography technique that will be introduced for manufacturing of technology levels. where the required dimensions are smaller than the actual wavelength. This paper explores several techniques to extend 193nm to low k1 lithography. Most attention is given to binary mask solution in at 130nm dimensions, where k1 is 0.4. Various strong and Gaussian quadrupole illuminators were designed, manufactured and tested for this application. Strong quadrupoles show that largest DOF improvements. The drawback however, is that these strong quadrupoles are very duty cycle and dimensions specific, resulting in large proximity biases between different duty cycles. Due to their design, Gaussian quadrupoles sample much wider frequency ranges, resulting in less duty cycles specific DOF improvements and less proximity basis. At sub-130nm dimensions, strong phase shift masks provide significant latitude improvements, when compared to binary masks with quadrupole illumination. However, differences in dose to size for different duty cycles were up to 25 percent. For definition of contact holes, linewidth biasing through silylation, a key feature of the CARL bi-layer resist approach, demonstrated significant DOF latitude improvements compared to SLR at 140nm and 160nm contact holes.
The goal of this paper is to define a 'state of the art' of the lithographic performance obtained with an advanced 193 nm single layer resist process, for 150 nm technology generation specification and below. Even if the goal of the paper is not to propose a process ready to be implemented in pilot lines, the resist used should be commercially available and exhibit good performance. The Sumitomo PAR101 A4 meets both criteria and is selected for the study. In order to get the best performance from the resist, all evaluation work is completed using a BARC. Both inorganic and organic materials are first considered; their processes are optimized to obtain the best reflectivity control and chemical compatibility with the resist. Then, using G0, conventional illumination and a binary reticle, the process performance is evaluated in terms of linearity, depth of focus, energy latitude and proximity effects for 150 nm and 130 nm lines, and depth of focus and energy latitude for 170 nm contact holes. Different optical extension techniques are then compared for increasing the 130 nm lines process latitudes: off-axis illumination, and alternating phase shift masks.
In this paper, (sub) 0.18 micrometers KrF DUV processes are optimized for logic Front-End-Of-Line (FEOL) CMOS applications. A commercial DUV resist is used without resolution enhancement techniques such as phase-shift masks and off-axis illumination. The full patterning process is considered, i.e., in the final optimized process account is taken of the etch process. Statistical data shows that a stable process was obtained. However, due to minimal process windows at gate level after poly-etch, 0.18 micrometers FEOL cannot be realized in production with KrF DUV.
Isolated to dense linewidth offsets, also known as proximity bias, can consume a significant portion of the CD budget. As a result, it has received great attention over the recent years. It is demonstrated that proximity bias shows a cyclic swing behavior on reflective substrates with respect to resist thickness variations. The amplitude of proximity bias swing was found to be influenced by coherence, substrate reflectivity, feature dimension and pitch. Proximity bias swing is caused by differences in optical path lengths of light passing through the resist film. Due to different diffraction angles for different pitches, the incoupling positions for different pitches vary. The offset in CD swing incoupling positions for different pitches results in proximity bias swing. At low coherence however, an averaging effect on diffraction angles from different pitches takes place due to the wide range of angles of light passing through the mask. In addition, the impact of exposure margin variations on resolution and proximity bias was demonstrated. Low exposure margins offer high resolution. As a consequence, high proximity bias is observed. Furthermore, different line to space ratios were studied to identify the transition point between dense and isolated features with respect to proximity bias swing. At high coherence ((sigma) equals 0.35) it was observed that for 0.25 micrometers features with pitches smaller than 0.65 micrometers , proximity bias swing is larger than the +/- 0.5% CD budget, which makes it impossible to do effective application of proximity bias correction schemes. At low coherence, only limited proximity bias swing was found. Through variation of bake conditions it was demonstrated that these process variations had no measurable effect on proximity bias swing. Optical settings, in combination with substrate reflectivity, are the main contributors to (eliminate) proximity bias swing.
Chemically amplified resists are notoriously sensitive to substrate contaminations. Such substrates include Si3N4, SOG, SiO2 and TiN. Contaminations can result in deactivation of the acid catalyst, leading to resist footing for positive tone deep UV resists. In this paper we have tested several state of the art deep UV resists on TiN. Through cross-sectional inspections, it was seen that several of the most advanced resists available still suffer from footing on TiN. By varying the process parameters of the TiN PVD process, TiN layers with various ratios of Ti:N were obtained. Variations in TiN composition result in changes of deactivation of acid catalyst. In addition, optical properties of the TiN layer are changed as well, resulting in different resist performances. For resists suffering from footing on TiN, it was demonstrated that footing is related to the nitrogen saturation of the TiN layer. However, for ARCH2 resist series, no resist footing was seen on different TiN layers. For the optimization of deep UV patterning of 0.25 micrometers CMOS metal layers using standard TiN layers, we have tested several resists of the ARCH2 resist series. The series of resists are based on the same resist chemistry. The difference between the formulations is in their absorbance, being 0.21/micrometer, 0.28/micrometer and 0.44/micrometer for ARCH214, ARCH212, and ARCH200, respectively. It was seen that with transparent resists notching can occur due to substrate roughnesses. By increasing the resist thickness and/or the resist absorbance, notching was minimized.