Photocopying – Projection printing and copying cameras – Step and repeat
Reexamination Certificate
2000-11-15
2002-05-14
Adams, Russell (Department: 2851)
Photocopying
Projection printing and copying cameras
Step and repeat
C355S055000, C355S067000, C355S071000, C355S077000, C430S004000, C430S005000, C430S311000, C430S312000, C250S492200, C250S492220
Reexamination Certificate
active
06388736
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a method of microlithography using an optical exposure system to project mask images for use in the fabrication of semiconductor devices.
BACKGROUND OF THE INVENTION
Optical lithography involves the creation of relief image patterns through the projection of radiation within or near the UV visible portion of the electromagnetic spectrum. Techniques of optical microlithography have been used for decades in the making of microcircuit patterns for semiconductor devices. Early techniques of contact or proximity photolithography were refined to allow circuit resolution on the order of 3 to 5 &mgr;m. More modern projection techniques minimize some of the problems encountered with proximity lithography and have lead to the development of tools that currently allow resolution below 0.15 &mgr;m.
Semiconductor device features are generally on the order of the wavelength of the ultraviolet (UV) radiation used to pattern them. Currently, exposure wavelengths are on the order of 150 to 450 nm and more specifically 157 nm, 193 nm, 248 nm, 365 nm, and 436 nm. The most challenging lithographic features are those which fall near or below sizes corresponding to 0.5 &lgr;/NA, where &lgr; is the exposing wavelength and NA is the objective lens numerical aperature of the exposure tool. As an example, for a 248 nm wavelength exposure system incorporating a 0.60NA objective lens, the imaging of features at or below 0.18 micrometers is considered state of the art.
FIG. 1
shows the configuration of a projection exposure system. Such an exposure system can be used in a step-and-repeat mode (referred to a stepper tool) or in a step-and-scan mode (referred to as a scanner tool). A UV or vacuum ultraviolet (VUV) source
1
is used to pass radiation through the illumination system
2
using a condenser lens system
3
and a fly's eye microlens array
4
. An aperture
5
shapes the illumination profile to a defined area and radiation is relected from a mirror
6
to pass through an illumination lens
7
to illuminate a photolithographic mask
8
. Upon illumination of the photomask
8
, a diffraction field
11
distributed as spatial frequency detail of the photomask
8
is directed through the objective lens
9
to be imaged onto the photoresist coated semiconductor substrate
10
. Such an exposure system forms an image by collecting at least more than the 0
th
-order of the diffraction field from the photomask
8
with the objective lens
9
. The absolute limitation to the smallest feature that can be imaged in any optical system corresponds to 0.25 &lgr;/NA. Furthermore, the depth of focus (DOF) for such an exposure tool can be defined as +/−k
2
&lgr;/NA
2
where k
2
is a process factor that generally takes on a value near 0.5.
As geometry sizes continue to shrink below 0.5 &lgr;/NA, methods of resolution enhancement are being required to ensure imaging with adequate fidelity and depth of focus. Such methods of resolution enhancement developed over recent years can allow for improvement in addition to those made possible with shorter exposing wavelengths and larger numerical apertures. Off-axis illumination (OAI) and phase-shift masking (PSM) are current examples of resolution enhancement techniques.
By using OAI in a projection imaging system, image refinement is carried out by considering illumination apertures which are not necessarily circular. In a system where illumination is obliquely incident on the mask at angles so that the zeroth and first diffraction orders are distributed on alternative sides of the optical axis, two diffraction orders are sufficient for imaging. An illumination angle can be chosen using two uniquely placed circular poles (dipoles) for a given wavelength, NA, and feature size. This is shown for example in the prior art of
FIG. 2
, where the normalized angular distribution of illumination (sin &thgr;/NA)is represented. This illumination angle resulting from the two apertures,
20
, is can be chosen for dense features as sin &thgr;=(0.5 &lgr;/p) where p is the feature pitch. The most significant impact of this dipole off axis illumination is realized when considering focal depth. In this case, the zeroth and first diffraction orders travel a more similar path length compared to conventional illumination as defocus is considered.
Off axis illumination using dipole illumination, oriented in the direction of mask geometry, can offer the most significant enhancement to imaging performance. This is because only oblique illumination at an optimized illumination angle can be designed to allow projection of mask diffraction energy at the outermost edges of an objective lens pupil. Frequency doubling is made possible (at the limit where pole size approaches zero and point source behavior occurs) and extreme focal depth can be achieved (since radial usage of the objective lens pupil in limited to a narrow region near the outside edge). The problem with dipole illumination arises when geometry of both X and Y (or horizontal and vertical) nature is considered. In practice, by limiting illumination to allow for one narrow beam or pair of beams leads to zero intensity. Also, imaging is limited features oriented along one direction in an X-Y plane. To overcome this, an annular or ring distribution has been historically employed which delivers illumination at angles needed with a finite ring width to allow for some finite intensity [see for instance H. H. Hopkins, Proc. Royal Soc. A, Vol. 217, 408-432 (1953)]. The resulting focal depth is less than that for the ideal case but improvement over a full circular aperture can be achieved.
For most integrated circuit applicatons, features are limited to X and Y orientations only and a four pole or quadrupole configuration can be more suitable (see for instance U.S. Pat. No. 5,305,054). In these cases, a quadrupole type illumination is required to accommodate the two orthogonal orientations of mask features. Solutions for such quadrupole illumination is where poles are at diagonal positions oriented 45 degrees to X and Y mask features. This is shown in the prior art of FIG.
3
. Here, each illumination pole,
30
, is off axis to all mask features and image improvement for X and Y oriented features occurs. The maximum angle for this quadrupole illumination is limited compared to a dipole illumination because of the placement of poles on diagonal axes. The maximum illumination angle is smaller than that for the dipole configuration by a factor of the square root of two. Resolution or imaging potential is also reduced as compared to dipole type off-axis illumination by this factor of the square root of two. This diagonally oriented quadrupole approach to off-axis illumination, and variation on this approach including weak Gaussian-pole designs, has been used for optical microlithography applications for several years now. Imaging below 0.4 &lgr;/NA (for 1:1 line to space ratio geometry) has not been demonstrated using this approach however.
Phase shift masking has been use for several years to improve lithographic imaging [see for instance Levenson et ale, “Improving Resolution in Photolithography with a Phase-Shifting Mask”, IEEE Transactions on Electron Devices, vol. ED-29, No. 12, p. 1828-, December 1982]. With conventional binary masking, only the control of the amplitude of a mask function is considered and phase information is assumed to be non-varying additional manipulation of phase information at the mask can allow for improvement of imaging performance. For coherent illumination, when a &pgr; “phase shifter” is added at alternating mask openings in a mask, an objective lens pupil has a 50% decrease in required numerical aperture required to capture required diffraction orders. Alternatively, for a given lens numerical aperture, a mask which utilizes such alternating aperture phase shifters could image features one-half the size that is possible using a conventional binary mask. As partial coherence is considered, the impact of this phase shi
Petersen John S.
Smith Bruce W.
Adams Russell
ASM Lithography B.V.
Brown Khaled
Pillsbury & Winthrop LLP
LandOfFree
Imaging method using phase boundary masking with modified... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Imaging method using phase boundary masking with modified..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Imaging method using phase boundary masking with modified... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2828154