Technique to separate dose-induced vs. focus-induced CD or...

Details Technique to separate dose-induced vs. focus-induced CD or... Technique to separate dose-induced vs. focus-induced CD or...

1999-08-31

2002-07-02

Nguyen, Kiet T. (Department: 2881)

Radiant energy

Irradiation of objects or material

Irradiation of semiconductor devices

C359S637000, C355S053000

Reexamination Certificate

active

06414326

ABSTRACT:

FIELD OF THE INVENTION
The present invention generally relates to optical aligner lithography tools and methods for using such tools, and more particularly relates to a method for separating dose-induced from focus-induced critical dimension or linewidth variations of features across an image field in a lithographic printing process.
BACKGROUND OF THE INVENTION
Lithography in semiconductor processing relates generally to the process of transferring patterns which correspond to desired circuit components onto one or more thin films which overlie a substrate. One important step within the field of lithography involves optical tools and methods for transferring the patterns to the films which overlie the semiconductor wafer. Patterns are transferred to a film by imaging various circuit patterns onto a photoresist layer which overlies the film on the wafer. This imaging process is often referred to as “exposing” the photoresist layer. The benefit of the exposure process and subsequent processing allows for the generation of the desired patterns onto the film on the semiconductor wafer, as illustrated in prior art
FIGS. 1
a
-
1
f.
Prior art
FIG. 1
a
illustrates a photoresist layer
10
deposited by, for example, spin-coating, on a thin film
11
such as, for example, silicon dioxide (SiO
2
) which overlies a substrate
12
such as silicon. The photoresist layer
10
is then selectively exposed to radiation
13
(e.g., ultraviolet (UV) light) via a photomask
14
(hereinafter referred to as a “mask”) to generate one or more exposed regions
16
in the photoresist layer
10
, as illustrated in prior art
FIG. 1
b
. Depending on the type of photoresist material utilized for the photoresist layer
10
, the exposed regions
16
become soluble or insoluble in a specific solvent which is subsequently applied across the wafer (this solvent is often referred to as a developer).
The exposed regions
16
are made either soluble or insoluble in the developer. When the exposed regions
16
are made soluble, a positive image of the mask
14
is produced in the photoresist layer
10
, as illustrated in prior art
FIG. 1
c
, and the photoresist material is therefore referred to as a “positive photoresist”. The exposed underlying areas
18
in the film
11
may then be subjected to further processing (e.g., etching) to thereby transfer the desired pattern from the mask
14
to the film
11
, as illustrated in prior art
FIG. 1
d
(wherein the photoresist layer
10
has been removed). Conversely, when the exposed regions
16
are mode insoluble, a negative image of the mask
14
is produced in the photoresist
10
layer, as illustrated in prior art
FIG. 1
e
, and the photoresist material is therefore referred to as a “negative photoresist”. In a similar manner, the exposed underlying areas
20
in the film
11
may then be subjected to further processing (e.g., etching) to thereby transfer the desired pattern from the mask
14
to the film
11
, as illustrated in prior art
FIG. 1
f.
The transfer of patterns to the photoresist layer
10
as discussed above involves the use of optical aligners. Optical aligners are machines which contain a variety of subsystems that work together to form the imaging function. Such optical aligners include: (1) an illumination source which provides the optical energy (UV light in the above example) for transforming the photoresist via exposure, (2) an optical subsystem that focuses the circuit patterns onto the photoresist surface and allows for controlled exposure times, and (3) and a movable stage that holds the wafer being exposed.
Historically, three primary methods have been used to optically transfer a mask pattern to a photoresist covered film. These methods are: contact printing, proximity printing and projection printing and are illustrated in simplified form in prior art
FIGS. 2
a
-
2
d
, respectively. Contact printing
100
, as illustrated in prior art
FIG. 2
a
, was the earliest method used to produce patterns. Contact printing
100
involves a light source
112
, an optical system
114
, a mask
116
and a photoresist layer
118
overlying a thin film
119
(not shown) which, in turn, overlies a semiconductor wafer
120
. The mask
116
, which contains the desired circuit patterns for transfer to the photoresist layer
118
, is positioned (aligned) relative to any existing patterns that already existed on the wafer
120
. The mask
116
is then clamped down to the photoresist layer
118
, thereby making physical contact with the photoresist layer
118
, and exposed with ultraviolet (UV) light from the light source
112
. This method provides for an excellent image transfer and good resolution (i.e., good minimum linewidth spacing).
Contact printing, however, suffers from the direct contact made between the mask
116
and the photoresist layer
118
. The repeated contact made between the mask
116
and the photoresist layer
118
in the process results in defects generated in the mask
116
which are then reflected in the transfer made on subsequently processed wafers. To prevent this problem, the masks
116
must disadvantageously be inspected and cleaned regularly. In addition, small particles may be caught between the mask
116
and the photoresist layer
118
when affixing the two elements, thereby preventing the desired direct contact between the mask
116
and the photoresist layer
118
. This particulate contamination results in reduced resolution in the area local to the foreign particle. Consequently, contact printing is not common in VLSI semiconductor manufacturing.
Proximity printing
122
, as illustrated in prior art
FIG. 2
b
, involves placing the mask
116
near the wafer
120
(which is covered with the photoresist
118
) during exposure, however, the mask
116
and the wafer
120
do not make contact. By introducing a gap
124
between the mask
116
and the wafer
120
, the defect problem of contact printing is substantially avoided. Unfortunately, as the gap
124
increases, the resolution of the proximity printing system
122
rapidly deteriorates. For example, a 10 &mgr;m gap with a 400 nm exposure (the wavelength of the light source
112
) results in a minimum resolution of about 3 &mgr;m. In addition, proximity printing
122
requires extremely flat masks
116
and wafers
120
in order to prevent gap variations spatially about the wafer
120
. Since many VLSI semiconductor circuits today require features of 0.25 &mgr;m or less, proximity printing
122
is not considered adequate for many VLSI semiconductor manufacturing operations.
Projection printing is a generic term that encompasses various pattern transfer techniques. These techniques, for example, include: (a) projection scanning systems, (b) reduction (e.g., 4× or 10×) step-and-repeat projection systems and (c) reduction step-and-scan systems. In each system, lens elements or mirrors are used to focus the mask image on the wafer surface (containing the photoresist).
Projection scanning systems (often called scanning projection aligners), use a reflective spherical mirror (reflective optics) to project an image onto the wafer surface, as illustrated, for example, in prior art
FIG. 2
c
. The system
126
includes a primary mirror
128
and a secondary mirror
129
which are arranged with the mask
116
and the wafer
120
to image the mask pattern onto the photoresist layer
118
which overlies the film
119
on the wafer
120
(the photoresist layer
118
and the thin film
119
are not shown in
FIG. 2
c
for simplicity). A narrow arc of radiation is imaged from the mask
116
to the wafer
120
with light that travels an optical path that reflects the light multiple times. The mask
116
and the wafer
120
are scanned through the arc of radiation by means of a continuous scanning mechanism (not shown). The scanning technique minimizes mirror distortions and aberrations by keeping the imaging illumination in the “sweet spot” of the imaging system
128
and
129
.
Reduction step-and-repeat systems
130
(also called reduction steppers) use refractive opt

Affiliated with

Also associated with

Rating

Technique to separate dose-induced vs. focus-induced CD or... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Technique to separate dose-induced vs. focus-induced CD or..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Technique to separate dose-induced vs. focus-induced CD or... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2868908