Radiation imagery chemistry: process – composition – or product th – Radiation modifying product or process of making – Radiation mask
Reexamination Certificate
2000-01-07
2002-10-22
Baxter, Janet (Department: 1752)
Radiation imagery chemistry: process, composition, or product th
Radiation modifying product or process of making
Radiation mask
C438S479000, C438S480000, 43
Reexamination Certificate
active
06468700
ABSTRACT:
FIELD OF THE INVENTION
This invention pertains primarily to transfer mask blanks and transfer masks employed in microlithography apparatus and methods in which a charged particle beam of electrons, ions, or the like is used to transfer a pattern, defined on the transfer mask, to a suitable substrate such as a semiconductor wafer. Microlithography is used in the fabrication of, e.g., semiconductor integrated circuits and displays.
BACKGROUND OF THE INVENTION
For many years, optical microlithography (microlithography employing visible or ultraviolet light) has been the standard pattern transfer technology used in the manufacture of semiconductor integrated circuits. However, the resolution of optical microlithography systems is limited by the diffraction of light. In recent years the progressively decreasing size and increasing device density of integrated circuits has led to intensive efforts to develop a practical alternative microlithography apparatus that employs a beam of charged particles (e.g., electrons, ions, etc., hereinafter termed a “charged particle beam”) or a beam of X-rays to transfer a pattern from a transfer mask (or reticle) to a suitable substrate. Charged-particle-beam (CPB) microlithography systems offer prospects for better resolution, as compared to optical microlithography systems. Current CPB microlithography systems employ a scanning electron beam for exposing a pattern onto a substrate by “writing” the pattern feature-by-feature. By focusing the electron beam to a spot diameter of a few nanometers, such a system can form very fine pattern features, sized 1 &mgr;m or smaller.
However, conventional electron-beam exposure systems write only one line at a time, and the finer the pattern the more “focused” the electron beam must be for drawing (i.e., the smaller the “spot diameter,” or area illuminated by the beam at any instant). Hence, drawing time is increased and throughput correspondingly decreased. From the perspective of device production costs, electron-beam exposure systems that write the pattern feature-by-feature cannot be used practicably to expose wafers for mass production.
To increase throughput over that obtainable using CPB “writing” systems, CPB projection-transfer apparatus have been proposed. In such apparatus, an electron beam (as a representative charged particle beam) illuminates all or a portion of a pattern defined on a transfer mask. An image of the illuminated region of the transfer mask is demagnified as the image is projected onto a corresponding region of the wafer. Such projection is performed by passing the beam, after propagating through the transfer mask, through a CPB projection-optical system (projection lens) located between the transfer mask and the substrate.
For use in a “demagnifying” (or “reducing”) CPB projection-exposure apparatus, a transfer mask (also termed a “reticle”) is required upon which a circuit pattern is drawn (i.e., the mask “defines” the circuit pattern). Referring to FIGS.
3
(
a
)-
3
(
b
), types of transfer masks currently used with such apparatus include: (1) scattering-transmission masks
11
(FIG.
3
(
a
)), in which pattern features (or pattern “elements”) are defined by corresponding CPB-non-scattering regions
15
of a CPB-transmissive membrane
12
; and (2) scattering-stencil masks
21
(FIG.
3
(
b
)) in which pattern features are defined by a corresponding pattern of through-holes
24
defined in a membrane
22
. The membrane
22
in the scattering-stencil mask of FIG.
3
(
b
) is sufficiently thick to scatter an electron beam, whereas the membrane
12
in the scattering-transmission mask of FIG.
3
(
a
) transmits the electron beam with little to no scattering.
Each of the masks of FIG.
3
(
a
) and FIG.
3
(
b
) is typically divided into multiple small regions
12
a
and
22
a
, respectively, that are individually exposed. Each small region
12
a
,
22
a
is also termed an “exposure unit” or “subfield” in the art. Each exposure unit
12
a
,
22
a
defines a respective arrangement of pattern features or elements (corresponding to the respective portion of the overall pattern defined by the particular exposure unit) to be projection-transferred to a sensitive substrate. The exposure units
12
a
,
22
a
are divided from one another on the mask by boundary regions
13
a
,
23
a
, respectively, in which no pattern features are defined. The boundary regions
13
a
,
23
a
are typically the regions of the mask in which supporting struts
13
and
23
, respectively, are located.
In a scattering-stencil mask
21
(FIG.
3
(
b
)), the membrane
22
consists of a silicon membrane, about 2 &mgr;m thick, that defines apertures
24
for passage of the electron beam. The pattern of apertures
24
in the membrane
22
of an exposure unit
22
a
defines the respective portion of the pattern to be transferred to the sensitive substrate (normally situated downstream of the mask).
In either of the masks (FIGS.
3
(
a
)-
3
(
b
)), the exposure units
12
a
,
22
a
are typically small, each measuring, for example, about 1 mm square (i.e., 1 mm×1 mm). At any one instant, only one exposure unit
12
a
,
22
a
is illuminated by the electron beam, and the exposure units of the pattern are typically illuminated in an ordered manner (e.g., sequentially). Each exposure unit
12
a
,
22
a
defines a corresponding portion of the overall pattern defined by the mask for transfer to a chip region (“die”) on the sensitive substrate. The entire pattern to be exposed onto each die is defined on the respective mask
11
,
21
by a large number of exposure units
12
a
,
22
a
arrayed over a large area of the mask.
Either of the masks
11
,
21
can be used for projection exposure as shown in
FIG. 4. A
charged particle beam exposes each exposure unit
12
a
(
22
a
) in a sequential manner. The respective pattern portion corresponding to the respective arrangement of apertures (or non-scattering membrane regions) in each exposure unit
12
a
(
22
a
) is demagnified (reduced) and transferred to the sensitive substrate
17
by a projection-optical system (not shown). The images of the various exposure units
12
a
(
22
a
) are positioned relative to each other so as to be “stitched” together on the sensitive substrate
17
. As can be seen in
FIG. 4
, “demagnification” results in an image on the substrate
17
that is smaller than the corresponding region on the mask.
Conventional transfer mask blanks and transfer masks are manufactured by a process as illustrated in FIGS.
5
(
a
)-
5
(
f
). A silicon substrate
1
, having a (100) crystal-lattice orientation on its surface, is prepared or otherwise provided (FIG.
5
(
a
)). Boron is diffused into the silicon on one side of the substrate
1
, either by thermal diffusion or by ion implantation, forming an “activation layer”
2
. The activation layer
2
is destined to serve as an etch-stop layer during a later etching step (desirably wet, or anisotropic, etching), and ultimately becomes the silicon membrane of the mask. Regions of the silicon substrate
1
, other than the silicon activation layer
2
, are known as the “support silicon”
1
a
(FIG.
5
(
b
)).
Next, a thin film of silicon nitride
3
is formed over the entire surface of silicon substrate
1
(FIG.
5
(
c
)). Selected regions of the silicon nitride film
3
formed on the rear surface of the substrate
1
are removed to form windows
4
. The remaining silicon nitride film
3
forms a “wet-etch” mask
5
, for use in a subsequent wet-etching step (FIG.
5
(
d
)).
FIG.
5
(
d
) shows a window
4
in one location. But, in actual practice, a large number of windows
4
are formed as required to define corresponding exposure units on the mask.
After forming the wet-etch mask, the entire silicon substrate
1
, with wet-etch mask
5
formed thereon, is immersed in an etching solution such as a KOH solution. In the etching solution, anisotropic etching of the support silicon
1
a
proceeds depthwise into the thickness dimension of the silicon substrate
1
from each window
4
(the windows are regions not protected
Baxter Janet
Klarquist & Sparkman, LLP
Nikon Corporation
Walke Amanda C.
LandOfFree
Transfer mask blanks and transfer masks exhibiting reduced... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Transfer mask blanks and transfer masks exhibiting reduced..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Transfer mask blanks and transfer masks exhibiting reduced... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2935350