Radiation imagery chemistry: process – composition – or product th – Radiation modifying product or process of making – Radiation mask
Reexamination Certificate
2000-02-07
2002-11-12
Rosasco, S. (Department: 1756)
Radiation imagery chemistry: process, composition, or product th
Radiation modifying product or process of making
Radiation mask
Reexamination Certificate
active
06479194
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention is in the field of fabricating semiconductor wafers. More particularly, the invention is in the field of photolithographic masks used to fabricate semiconductor wafers.
2. Background Art
Photolithography is used extensively in the semiconductor industry to form a desired pattern on the surface of a semiconductor wafer. Typically, the photolithographic process begins by coating the surface of a silicon wafer with photoresist. Conventionally, a “binary mask” having fully non-transmittive opaque regions made of chrome, and fully light transmittive transparent regions made of quartz, is then positioned over the surface of the photoresist coated wafer. Typically a lens system of a stepper is used to shine visible or ultra-violet light on the binary mask. This light passes through the transparent regions of the mask and exposes the corresponding regions of the underlying photoresist layer, and is blocked by the opaque regions of the mask, leaving the corresponding regions of the photoresist layer unexposed. The photoresist layer is then developed by chemical removal of the exposed or non-exposed regions (depending on whether a positive or a negative resist has been used). The end result is a silicon wafer and the desired pattern of photoresist thereon. This pattern can then be used for etching underlying regions of the silicon wafer.
A major shortcoming of the conventional binary mask is that it cannot be effectively used to pattern feature sizes which are substantially smaller than the exposure wavelength. For feature sizes smaller than the exposure wavelength, the use of the binary mask requires the maximum attainable numerical aperture of the lens system. However, as the numerical aperture is increased, the depth of field of the lens system is reduced. Since the surface of the silicon wafer is not optically flat, the reduction of depth of field of the lens system translates into poor focus in various regions of the silicon wafer. Thus, as the minimum feature size is reduced, the limits of the conventional binary mask are reached.
To overcome the limitations of the binary mask, one technique involves use of an “alternating phase shift mask.” An example of an alternating phase shift mask is mask
102
in FIG.
1
A. Referring to mask
102
in
FIG. 1A
, a fully opaque material such as chrome is used in region
106
of mask
102
. Chrome region
106
is flanked by transparent regions
104
and
108
. Transparent regions
104
and
108
are typically made of quartz. The thickness of quartz region
108
is approximately 2280 Angstroms less than that of quartz region
104
. The thickness of region
108
is calculated such that light passing through region
108
has a phase shift of 180 degrees relative to light passing through the thicker region
104
which has no phase shift, i.e. a phase shift of 0 degrees.
The 180 degree phase shift induced in light passing through quartz region
108
results in destructive interference between light passing through quartz region
108
and that passing through quartz region
104
. Graph
120
in
FIG. 1A
shows the approximate intensity of light under regions
104
,
108
, and
106
. The opposite phases of light passing through the 0 degree phase shift region
104
and light passing through the 180 degree phase shift region
108
cancel, resulting in a dark unexposed region
116
on the surface of the photoresist coated wafer. As shown in
FIG. 1A
, the dark unexposed region
116
falls on an area of the photoresist layer which is located below chrome region
106
.
Although small feature sizes can be patterned by using an alternating phase shift mask such as mask
102
in
FIG. 1A
, alternating phase shift masks have various shortcomings. For example, reference to graph
120
in
FIG. 1A
illustrates that the intensity of light passing through the 0 degree phase shift region
104
is greater (having peak
114
) than the intensity of light passing through the 180 degree phase shift region
108
(having peak
112
). The difference in intensity causes an unbalanced exposure of the photoresist layer such that areas of photoresist under mask region
108
are underexposed while areas of photoresist under mask region
104
are overexposed.
The above-described imbalance in intensity and image position of the conventional alternating phase shift mask is a great disadvantage and in an attempt to overcome this disadvantage a “dual trench” alternating phase shift mask has recently been devised. In the dual trench alternating phase shift mask, quartz from both regions
104
and
108
is etched to different depths so that, while maintaining the 180 degree phase shift of region
108
relative to region
104
, the intensity of light under both regions
104
and
108
is almost equalized. However, the dual trench results in a shift in the center of focus of light passing through one trench with respect to the center of focus of light passing through the other trench. Moreover, the dual trench alternating phase shift mask is more difficult to manufacture since quartz from both regions
104
and
108
has to be etched to precise, but different, depths in order to balance the intensity of light passing through both regions
104
and
108
.
Moreover, according to the dual trench alternating phase shift mask (and also the conventional alternating phase shift mask), the use of a first mask results in creation of artifacts at the boundaries between the 0 degree phase shift and the 180 degree phase shift regions. Accordingly, a second mask is required for a second exposure in order to erase the artifacts created by the first mask during the first exposure. The requirement that two masks (and two exposures) must be used increases the complexity of the photolithographic process, results in a need for an accurate alignment between the first and second exposures, reduces the throughput of processing the semiconductor wafer, and is more costly than a single mask, single exposure method.
A technique is disclosed in U.S. Pat. No. 5,858,580 to Wang et al utilizing a two-mask, two-exposure approach. A first “phase shift mask” consists of a “control chrome” that is flanked by 0 degree and 180 degree phase regions. Regions in the first mask other than the control chrome, 0 degree, and 180 degree phase regions are covered by a “phase shift mask chrome.” A dark unexposed area under the “control chrome” is created at the phase boundaries of the 0 degree and 180 degree phase regions. A second “structure mask” is used to expose selected parts of the underlying photoresist that were unexposed due to the existence of the phase shift mask chrome in the first mask.
Another technique is disclosed in U.S. Pat. No. 5,573,890 to Spence utilizing a single mask method in which transition regions are used to compensate for artifacts that would have been produced if there were 0 degree to 180 degree transitions in the mask. Spence also discloses a two-mask, two-exposure approach where a first “phase shift and structure mask” consisting of adjacent 0 degree and 180 degree phase regions is used to create dark unexposed areas on the underlying photoresist. A second “trim mask” is then used to erase artifacts produced by the first mask.
A recent variation to the alternating phase shift method contemplates use of four different phases of light, i.e. use of regions causing a 0 degree phase shift, 60 degree phase shift, 120 degree phase shift, and 180 degree phase shift so that artifacts are not produced after the first exposure. Although this type of alternating phase shift technique uses a single mask and a single exposure, the need for four different phase regions makes this type of alternating phase shift mask extremely expensive to manufacture.
Another recent attempt to overcome the limitations of the binary mask in patterning small feature sizes is the “attenuated phase shift mask.”
FIG. 1B
shows an attenuated phase shift mask
142
. Regions
144
and
148
are made of a transparent material such as quartz. Transparent region
Conexant Systems Inc.
Farjami & Farjami LLP
Rosasco S.
LandOfFree
Transparent phase shift mask for fabrication of small... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Transparent phase shift mask for fabrication of small..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Transparent phase shift mask for fabrication of small... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2920957