Radiation imagery chemistry: process – composition – or product th – Radiation modifying product or process of making – Radiation mask
Reexamination Certificate
2000-06-19
2003-05-27
McPherson, John A. (Department: 1756)
Radiation imagery chemistry: process, composition, or product th
Radiation modifying product or process of making
Radiation mask
C430S296000, C430S323000
Reexamination Certificate
active
06569575
ABSTRACT:
TECHNICAL FIELD
The present invention concerns a new scheme for optical lithography that allows extension of optical lithography beyond the conventional resolution limits imposed by the wavelength of the exposing light.
BACKGROUND OF THE INVENTION
The resolution of conventional optical lithography schemes is mainly limited by the wavelength of the light used for the transfer of a mask pattern onto a resist. The wavelength of the exposing radiation is a main determinant of pattern resolution W, given by the Rayleigh equation W=k
j
&lgr;/NA, where &lgr; is the wavelength of the exposing light, NA is the numerical aperture of the optical lithography tool, and k
j
is a constant for a specific lithography process. In other words, the resolution W is proportional to the wavelength &lgr; of the exposing light. Cutting edge production today creates features that are 250 nm wide using 248 nm illumination. Currently, the implementing schemes based on light are the bottleneck when trying to obtain structures of a feature size below 200 nm.State-of-the-art optical lithography system for making current DRAMs, for example, are quite expensive. Alternative processes become attractive when moving on to smaller feature sizes, but the required investments are huge. Thus techniques that maintain compatibility with much of the existing processes are inherently valuable.
Trends in both integrated circuit and flat-panel display manufacturing technologies require improvements in small scale lithography. In these and other fields, there is an increasing demand for a cost-effective lithographic technique that can produce large fields (to approximately 45 cm diagonal for displays) of nanoscale structures. The semiconductor industry road map calls for leading-edge manufacturing at 180 nm in the year 2001 and 70 nm in the year 2011.
One well known form of optical lithography is the so-called hard contact lithography, where the mask is moved directly into contact with the substrate targeted for patterning. Features on a mask, comprising alternatively translucent and opaque regions in a well-defined pattern, are printed into photo resist in a 1:1 relation to their area on the source. Hard contact lithography can, in principal, make structures with sizes below the wavelength of illumination. But the contact used to place the mask on the substrate compromises the integrity of the process as the possibility of confounding material on the surface of the mask and mask damage greatly limit (compared to projection lithography) the useful number of prints it can form. Cost is particularly worrisome as the feature scale shrinks and the expense of mask fabrication skyrockets with the increase in the density of its features. Contact masks are also generally much more expensive than those used in optical projection lithography since the critical dimensions in the former need to be smaller than those in the latter, for equivalent resolution, by the reduction factor used in a projection system. Dust particles and other physical impediments to the substrate are catastrophic in hard contact lithography as they lift the mask away from the surface, blurring the pattern. Such defects appear over an area much larger than the obscuring particle because the mask is unable to conform around their presence; this problem is compounded as the feature scale shrinks such that even a 200 nm particles can be harmful. In addition, resist can get stuck to the mask. Hard contact lithography has thus not found a significant role in manufacturing of small scale integrated circuits.
There are many approaches known, that improve conventional lithography systems in that filters, projection lens, or appropriately modified masks are employed. These approaches get more and more complicated and expensive with reducing feature scale. One example here is the so-called optical projection lithography. The optical lithography based on projection is undoubtedly the most successful and widely employed means of making features down to ~200 nm. Here, a pattern of intensity variations in the far field results when light is shown through a mask like that used in contact lithography. The light propagates through air and is focused by a lens to form an image of the desired pattern on a resist covered substrate, often demagnified by a factor of 5-10 from its size on the mask. Projection lithography is largely limited to features sizes at, and larger than, the wavelength &lgr; of light, however. Its implementation becomes increasingly difficult, in addition, as the scale shrinks towards, and below, 200 nm, where very complicated systems of lenses and materials are required to carry out existing and proposed schemes. The area over which uniform illumination can be achieved is particularly problematic. The maximum current field sizes in the best 248 nm exposure tools is now only ~20×20 mm. The useful area of exposure will continue to shrink dramatically with the wavelength of illumination, principally because of the materials and engineering challenges in forming uniform exposures through complex lenses based on silicates.
It is then generally a disadvantage of most of these approaches that they are getting more and more complicated and expensive when trying to obtain smaller feature sizes. Furthermore, there is a tradeoff between maximum resolution, depth of focus and achievable field image which comes from the use of a lens to focus the light.
The resolution of standard photolithography systems can be increased and feature size decreased by using masks that manipulate the phase (referred to as phase masks, phase shifting masks or PSMs) instead of the amplitude of the light used for exposure. Two examples of phase shift-based approaches are described by D.M. Tennant et al. in “Phase Grating Masks for Photonic Integrated Circuits Fabricated by E-Beam Writing and Dry Etching: Challenges to Commercial Applications”, Microelectronic Engineering, Vol. 27, 1995, pp. 427-434, and by J. A. Rogers et al. in “Using an elastomeric phase mask for sub-100 nm photolithography in the optical near field”, Appl. Phys. Lett., No. 70, Vol. 20, May 19, 1997, pp. 2658-2660.
Tennant et al. propose the use of hard contact masks, whereas Rogers et al. favor elastomeric masks
10
(see
FIG. 1
) for the formation of high density sub-wavelength features
17
. In these methods the pattern on the mask results in interference in the illumination arising in the near field from contact between a photo resist
11
and the structured mask
10
. Light passes everywhere through the mask
10
which is completely translucent but has a pattern of surface reliefs
14
that vary in a well-defined manner. Light traveling through such a structured mask
10
experiences a comparatively longer or shorter path depending on the place of its exit. This change in the effective path length through the structured mask
10
contributes to phase (and only phase) differences in the propagating light. These phase differences result in sub-wavelength nodes in intensity of the exposing radiation at the surface of the resist
11
. If these masks
10
are designed and made appropriately, there are nodes at the mask/resist interface
15
with a relative minimum in intensity.
Rogers et al. showed that using a phase approach with an elastomeric mask allowed them to make sub-wavelength features
18
in a photo resist layer
11
while avoiding the problems associated with brittle contact masks (like Tennant et al.), as illustrated in FIG.
1
. These features
18
can then be transferred into a substrate
16
by means of dry etching, or wet chemical dissolution of the substrate, as is well known in the art. The features
17
formed in the substrate have about the same lateral dimensions as the features
18
formed in the photo resist
11
. The problem with the aforementioned approaches to lithography based on phase shifting of light through a mask
10
is that, while small features
17
(sub-wavelength) can be generated, these features
17
are constrained to a one dimensional geometry (lines) or low density on
Biebuyck Hans
Michel Bruno
Schmid Heinz
McPherson John A.
Morris Daniel P.
LandOfFree
Optical lithography beyond conventional resolution limits does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Optical lithography beyond conventional resolution limits, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Optical lithography beyond conventional resolution limits will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3072409