This work describes the advantages, tolerances and integration issues of using Pixelated Phase Masks for patterning
logic interconnect layers. Pixelated Phase Masks (PPMs) can act as variable high-transmission attenuated phase shift
masks where the pixelated phase configuration simultaneously optimizes OPC and SRAF generation. Thick mask
effects help enable PPMs by allowing larger minimum pixel sizes and phase designs with near equal sized zero and piphase
regions. PPMs with a 3-tone pixel mask (un-etched glass, etched glass, chrome) offer more flexible patterning
capability compared to 2-tone pixel mask (no chrome) style but at the detriment of a more complex mask making
process. We describe the issues and opportunities associated with using PPMs for patterning a 65nm generation first
level metal layer of a micro-processor.
In June 2007, Intel announced a new pixelated mask technology. This technology was created to address the problem
caused by the growing gap between the lithography wavelength and the feature sizes patterned with it. As this gap has
increased, the quality of the image has deteriorated. About a decade ago, Optical Proximity Correction (OPC) was
introduced to bridge this gap, but as this gap continued to increase, one could not rely on the same basic set of
techniques to maintain image quality. The computational lithography group at Intel sought to alleviate this problem by
experimenting with additional degrees of freedom within the mask. This paper describes the resulting pixelated mask
technology, and some of the computational methods used to create it. The first key element of this technology is a thick
mask model. We realized very early in the development that, unlike traditional OPC methods, the pixelated mask would
require a very accurate thick mask model. Whereas in the traditional methods, one can use the relatively coarse
approximations such as the boundary layer method, use of such techniques resulted not just in incorrect sizing of parts of
the pattern, but in whole features missing. We built on top of previously published domain decomposition methods, and
incorporated limitations of the mask manufacturing process, to create an accurate thick mask model. Several additional
computational techniques were invoked to substantially increase the speed of this method to a point that it was feasible
for full chip tapeout. A second key element of the computational scheme was the comprehension of mask
manufacturability, including the vital issue of the number of colors in the mask. While it is obvious that use of three or
more colors will give the best image, one has to be practical about projecting mask manufacturing capabilities for such a
complex mask. To circumvent this serious issue, we eventually settled on a two color mask - comprising plain glass and
etched glass. In addition, there were several smaller manufacturability concerns, for example a "1X1" glass pillar (an
isolated 0 phase pixel) were susceptible to collapse under the stress of mask processing, and therefore these had to be
constrained out of the final configuration. A third key element was defining the objective function. We experimented
with a large number of choices and eventually settled on a form that allows us to trade-off fidelity and contrast. A fourth
key element was the optimization algorithm. The number of possible configurations for a trillion pixels present on our
final product mask is greater than the number of total elementary particles in the known universe, so finding the
proverbial needle in this haystack was difficult to say the least. We chose a mixture of stochastic and direct descent
algorithms to find an arrangement that meets the demands. While we have not proved we are close to the absolute global
minimum, we conducted several experiments to suggest this is the case. A fifth key element, and a large one at that, was
scaling up our software system from micron length scale to centimeter length scale required for full chip tapeout. This
software, in turn, has several key components - hierarchy handling, the non-trivial handling of pixelated domain
boundaries, repair of regions not converged in terms of image quality, and verification of the entire assembled database.
All elements described above were validated through the tapeout of an actual mask to pattern the most complex metal
layer for the leading 65nm node microprocessor in high volume manufacturing. This very first experimental tapeout
resulted in wafer parts yield comparable to yields on mass produced wafers made with production 65nm technology.
While condenser aberrations under Koehler illumination were previously treated in the literature their mathematical derivation did not take conservation of radiance into consideration. Here we make use of a more rigorous derivation of the mutual intensity where the source deformation term is treated in the context of radiance conservation. The derivation predicts that condenser aberrations lead to radiance invariance while aberrations have a direct bearing on illumination uniformity and the angular extent of the local effective source. This result significantly contrasts with the previously established conclusion in the literature that condenser aberrations lead to a modification of the source radiance but preserves irradiance in the reticle plane. Source aberrations of first and third order are derived and then systematically explored both analytically and numerically. Aberration impact on linewidth control are further considered and quantified from the aerial image perspective. It is shown that third order coma has the most significant impact on CD control as a result of the asymmetry in the deformation of the source shape. Similarly coma also significantly impacts overall mask illumination uniformity.
Various methods for the application of phase-shift mask (PSM) technology have been discussed, and for non-periodic features such as isolated contact holes, the "rim-shifter" and "out-rigger" methods show particular promise. While both approaches can improve process latitude, they introduce a new complication in the form of secondary illumination intensity lobes which can degrade lithographic performance. The present work specifically addresses whether an optimal "high contrast" resist process for conventional lithography will also be optimal for processes using rim-shifters and subresolution out-rigger shifters in the production of isolated trenches and contact holes. Simulations using SAMPLE and SPLAT show that for a given mask design, high contrast processes can amplify the secondary lobes and therefore may not be optimal. The reason for the enhanced printing of the secondary lobes is traceable to the higher Exposure Margin (defined as the ratio dose-to-size/dose-to-clear) associated with high contrast processes. Such processes require high exposures (relative to the clearing dose) to achieve the target developed dimension and so the secondary lobes are, in a sense, overexposed. Because the preferred mask designs are still evolving, it is uncertain whether secondary lobe printing will be an important factor in process optimization; however, the present work suggests that the problem is minimized by using positive resist systems with high surface inhibition and high transparency at the exposing wavelength. These are the same qualities which maximize profile-related defocus latitude with conventional masks. Lithographic results are presented showing superior performance of a photoresist formulated with these characteristics over conventional materials.
Chromeless phase-shifting is a novel concept that completely avoids the use of chrome for pattern formation in optical lithography. This scheme uses 180 degree(s) phase-shifters on transparent glass to define patterns. The method relies on the destructive interference between phase-shifters and clear areas at the edges of the phase-shifters to define dark or opaque areas on the mask. Gratings sufficiently small (named dark-field gratings) will produce sufficient interference to completely inhibit the transmission of light. The combination of these effects makes it possible to form a wide range of patterns, from line-space patterns to isolated bright or dark areas. The lithography simulators SPLAT and SAMPLE were used to understand the principles behind this new scheme, and to verify various pattern designs. Simulation and experimental results are presented to demonstrate the concept.
A 3-D optical lithography simulator has been developed based on a new ray-string algorithm for dissolution etch-front advancement. This simulator, SAMPLE-3D, integrates a number of process simulators on a workstation while also providing display and print capabilities. SAMPLE-3D has been used to look at 3D resist profiles from 2-dimensional mask patterns, including isolated contacts, isolated islands, and elbow patterns. Simulations have been performed on both positive and negative photoresists, and the effects of resist contrast and surface rate retardation were explored. The correlation between the 2D aerial image and the 3D developed resist profile has been investigated. This includes applications to the printability of defects where the nonvertical resist dissolution effects play a strong role.
Simulation has been used to systematically investigate the effects of phase-shifters on dark- field patterns (openings in a dark-field mask), and to determine the phase-shifter configurations that are most effective for different mask patterns. This study has resulted in a design methodology based on the distance between the centers of clear features and the surrounding phase-shifters. A key verification is that isolated phase-shifted patterns print best when the distance from the center of the phase-shifter to the center of the feature is approximately 0.7 (lambda) /NA. At this optimal spacing, the peak image intensity, image slope and resist wall-angle of the printed pattern is maximized. Optimally-aligned phase- shifters will also have the best focus-exposure behavior of all the different dark-field phase- shifter configurations. However, dark-field phase-shifters will only provide a resolution increase on the order of 0.05 0.10 (lambda) /NA. Different dark-field mask configurations will print with different amounts of bias; the amount of print bias is dependent on the width of the phase-shifters and the distance between the phase-shifters and the feature. Response curves of resist opening as a function of feature size and phase-shifter/feature separation can be used to keep track of the amount of bias required in a given phase-shifter configuration.
This paper introduces a novel concept, 'chromeless phase-shifting', that eliminates the need for the use of chrome to form patterns in optical lithography. Chromeless phase-shifting uses 180 degree(s) phase-shifters on transparent glass to define patterns. The method relies ont eh destructive interference between phase-shifters and clear areas at the edges of the phase-shifters to define dark or opaque areas on the mask. Gratings sufficiently small will produce sufficient interference to completely inhibit the transmission of light (these gratings are thus named dark-field gratings). The combination of these effects makes is possible to form a wide range of patterns, from line-space patterns to isolated bright or dark areas. In this study, the lithography simulators SPLAT and SAMPLE were used to understand the principles behind this new scheme, and to verify various pattern designs. Simulation and experimental results are presented to demonstrate the concept.