Radiation imagery chemistry: process – composition – or product th – Radiation modifying product or process of making – Radiation mask
Reexamination Certificate
2001-03-20
2002-08-06
Rosasco, S. (Department: 1756)
Radiation imagery chemistry: process, composition, or product th
Radiation modifying product or process of making
Radiation mask
C378S035000
Reexamination Certificate
active
06428939
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention is directed to high resolution lithography for the fabrication of nano-electronic devices and, more particularly, the manufacture and use of clear phase shifting masks for lithographic processes that enhance the bright-peak of the corresponding aerial image of the mask.
2. Description of the Related Art
Semiconductor computer chips typically include millions of devices fabricated on a single chip. These devices are defined at least in part by linear features or nodes that can be imaged on a wafer of the chip using a photolithographic process, typically comprising a number of steps for depositing multiple layers of materials. For example, trench features are often imaged on the wafer and then filled with metal to provide conductive traces on the chip.
The speed of semiconductor computer chips, such as the Intel Pentium® chip is dependent on the width of these linear features or nodes imaged on the associated wafer during production. As nodes are made smaller, more nodes can be placed on a chip and the processing speed increases. This technology has developed rapidly. For example, the Pentium® 3 chip has a nominal node size of approximately 180 nanometers (nm) and the Pentium® 4 chip has a nominal node size of approximately 120 nm.
A favored method for making these small nodes on chips includes placing a mask above a target (i.e., the chip) which comprises, for example, a silicon or GaAs substrate having a photoresist disposed thereon, and exposing the photoresist to light in a photolithographic technique. Notably, when writing features to a target in this fashion, the lithography process typically generates two types of images, an aerial image (i.e., the radiation intensity of the target surface), and a latent image (i.e., the image recorded in the target photoresist.)
With lithography techniques, the wavelength of the light utilized to form the image on the target photoresist imposes a fundamental limit on the achievable image definition or resolution, i.e., the minimum node width that can be imaged. One limiting factor in making smaller nodes has to do with diffraction, which is an unwanted side effect of the lithography production process. In particular, the image resolution is limited by diffraction of the light at the edges of the features of the masks through which the light is projected. To lessen the negative effects associated with diffraction, and thus allow the creation of smaller nano-electronic structures that are obtainable using visible or UV optical systems, sources that generate short-wave length light rays (e.g., X-rays) are being implemented.
Moreover, as the distance between the mask and the wafer increases, the diffraction of the light waves increases, so the feature produced on the chip is slightly larger than the feature on the mask. Therefore, typically, the smaller separation or gap between the mask and the photoresist, the less diffraction.
Ideally, you would eliminate diffraction by placing the mask directly on the target surface (a technique known as “contact printing”), but for several reasons this is not practical. For instance, the mask is fragile, and the photoresist may be sticky, so removing the mask from the surface after processing becomes difficult. Also, the additional time required in process (due to, for example, using extra care when coupling/separating the mask from the substrate) would decrease yield, as well as increase cost per part. In addition, the target and the mask both may have some amount of curvature, which makes it difficult to couple them and completely eliminate the gap.
Therefore, to further minimize the negative effects associated with diffraction, phase shifting masks (PSMs) may be used. PSMs enhance the imaging resolution of features to be printed on a wafer surface by exploiting a phase difference (e.g., a &pgr;-phase difference) between a phase shifting region and an open region both residing on the surface of a mask membrane. In particular, the phase difference causes interference between phase shifted and non-phase shifted radiation when writing features using the phase shifting mask. This interference occurs on both sides of an edge of the phase shifting regions, and can be observed by analyzing the corresponding aerial image. Typically, the image includes a “bright peak” and a “dark peak” associated with the edge.
In PSMs, two types of phase shift regions are typically employed, clear phase shift regions, and attenuating phase shift regions. Existing clear PSMs focus on compensating for diffraction by using “dark peak” interference waves to image the mask pattern onto the photoresist.
Note that a clear PSM is characterized by comprising phase shifting material that does not substantially attenuate the source radiation (e.g., X-rays) passing therethrough. To the contrary, attenuating PSMs typically comprise an attenuator that absorbs a significant portion of the radiation impinging thereon, thus generating the desired phase shift of that portion of the radiation passing therethrough. A suitable heavy metal absorber, such as Gold (Au) may be made of Tungsten (W) having an appropriate thickness. A drawback associated with attenuating PSMs is that they typically reduce the radiation intensity when imaging mask features. By providing more power to the wafer, clear PSMs require shorter exposure times, and thus facilitate greater wafer throughput.
A typical phase mask and its associated aerial image are shown in
FIGS. 1A and 1B
, respectively, for the case in which phase X-ray lithography (PXRL) is employed. A mask
10
includes a carrier or membrane
12
having generally planar top and bottom surfaces
14
,
16
, respectively. Membrane
12
supports a phase shift region
18
defining an edge
20
and is made of a suitable material such as polymethyl-methacrylate (PMMA) having a thickness of about 2.5 to 3 &mgr;m such that it provides the desired half wavelength (i.e., “&pgr;”) phase shift for a corresponding band of X-ray photons of approximately 1.0 nm in wavelength. Edge
20
defines a boundary line between phase shift region
18
and an open region
21
supported by membrane
12
. In this case, membrane
12
is preferably made of silicon nitride (Si
3
N
4
) which is sufficiently thin not to absorb greatly the incoming radiation flux (e.g.,
22
in FIG.
1
A). Further, the gap between the mask and the target resist (not shown) is typically about 15 &mgr;m or less.
An aerial image
24
produced by this arrangement is shown in FIG.
1
B and is plotted as Normalized Intensity versus Dimension (nm) on the target. Image
24
includes a bright peak
32
and a dark peak
26
. In known PXRL systems, dark peak interference is used to compensate for the diffraction at the mask edge, and thus it is the dark peak
26
that is used to image the feature. More particularly, when imaging a node to the target, destructive interference of the spatially coherent phase shifted and non-phase shifted X-rays results in their fields canceling near the boundary line
20
so that the intensity of X-ray energy absorbed by the target photoresist reaches a minimum at a point
26
(i.e., the dark peak) which lies generally on boundary line
20
. Preferably, the energy absorption represented by point
26
is below the exposure threshold energy level E
1
, of the resist; that is, those areas of the resist which receive photon energy below the level of the line E
1
illustrated in
FIG. 1B
will not, for a positive resist, be dissolvable by a developer solvent. The developer will, however, dissolve the regions of the target photoresist to the left of a point
28
at which the threshold E
1
intersects the exposure level curve
24
and to the right of a point
30
at which the threshold E
1
intersects curve
24
. For a more detailed analysis of this technique, see U.S. Pat. No. 5,187,726 issued on Feb. 16, 1993, to Wisconsin Alumni Research Foundation. This technique typically uses the dark peak
26
to image lines in positive-tone resists, or trenches in negative-tone res
Cerrina Francesco
Taylor James W.
Yang Lei
Boyle Fredrickson Newholm Stein & Gratz S.C.
Rosasco S.
Wisconsin Alumni Research Foundation
LandOfFree
Enhanced bright peak clear phase shifting mask and method of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Enhanced bright peak clear phase shifting mask and method of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Enhanced bright peak clear phase shifting mask and method of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2953599