 |
Description  |
|
|
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method for inspecting a photomask
employed for replicating a circuit pattern onto a semiconductor wafer in
the process for producing a semiconductor integrated circuit, and more
particularly to a method for inspecting defects such as the presence of
foreign particles affecting the replication of the pattern.
2. Description of the Prior Art
Recent development of finer and denser integrated circuit pattern has
stimulated the widespread use of reduction projection exposure apparatus.
In such apparatus, in which a pattern on a photomask called reticle is
projected in a size reduced to 1/5 or 1/10 of the original size, foreign
particles such as dusts on the reticle are replicated in reduced size in
all the step-and-repeat exposures onto the wafer as long as they have a
certain size and a certain optical density. In order to avoid such
inconvenience, the resist image printed on the wafer has conventionally
been observed, in trial manner, under microscope by human eyes to detect
eventual replication of the foreign particles. This method however
requires an extremely tiring work for the eye, for 2 to 3 hours at maximum
per reticle, and may eventually overlook the foreign particles present.
These drawbacks have only recently been resolved commercially by an
apparatus which scans the reticle with a laser spot and detects the
scattered light to identify the presence and dimension of the foreign
particles, as disclosed in the Japanese Patent Application laid open No.
62543/1983, corresponding to the U.S. Pat. No. 4,468,120. Such apparatus
is capable of automatic inspection within a short time, and detecting
almost all foreign particles of a size large enough for replication.
However the inspecting method employed in said apparatus is designed to
satisfactorily distinguish foreign particles from the chromium layer, so
that foreign particles lying flat on the chromium layer, if large enough
in area, are sometimes mistaken as the chromium layer and not detected.
Also since this method does not directly inspect the replicated pattern,
all the foreign particles of a size that may be replicated have to be
detected. There has therefore been observed a tendency of detecting even
small particles which are in fact not replicated and thus requesting
excessive cleanness to the reticle.
Apart from the above-explained inspecting method, there is also considered
a totally different inspecting method of overlaying and replicating a
reticle to be inspected and another reticle, in which light and dark
patterns are inverted, in succession onto a photosensitive material, and
inspecting, after development thereof, the presence of spot patterns
resulting from lack of exposure. This method has however suffered from
insufficient precision in registration between a latent image formed by
the first exposure through the reticle to be inspected and a projected
image of the inverted pattern of the other reticle to be exposed next
time, mainly due to a fluctuation in the positioning of the reticle in the
exposure apparatus and to a drift in detecting the position of a stage
supporting the wafer. In case the precision of registration of the first
and second exposures is insufficient, there may be considered a method of
increasing the exposure to increase the width of overlapped exposure of
photoresist at the pattern edge, thus exposing the photoresist where the
pattern edges are positionally aberrated to apparently cancel such
aberration before the unexposed areas are inspected. In this method,
however, the foreign particles present on the pattern edge become harder
to be replicated and are often or totally overlooked at the inspection.
Besides, even if the positional registration is achieved with an optimum
precision, such satisfactory registration cannot be obtained over the
entire projection area if two reticles are exposed on different exposure
apparatus because of the slightly different image distortion between the
apparatus.
Finally, the detection of spot patterns formed on the photoresist layer has
to be made by the eyes through a microscope with an objective lens of a
high magnification and eventually through a television monitor, so that
this method is still associated with a drawback of requiring visual
observation in which the required labor is not significantly alleviated
since this method only provides a simpler form in the objects of
inspection. This method has not been applied to the practical use, since
the above-mentioned drawback has become more serious with the recent
development of finer circuit patterns.
SUMMARY OF THE INVENTION
A first object of the present invention is to provide a method for
inspecting the presence of foreign particles on a mask by precisely
registering a latent image formed by a first exposure with an image to be
formed in a second exposure, thereby enabling detection of replicated
images of even very small foreign particles.
According to the present invention, the above-mentioned object is achieved
by a method which is featured by forming alignment marks on a wafer or the
like onto which the pattern is trially replicated, aligning projected
images by means of said alignment marks to form overlapped images of the
pattern of a mask to be inspected and the pattern of another mask in which
the light and dark areas are inverted on a photosensitive material on said
wafer, and inspecting the exposure status of the photosensitive material
on said wafer after development to identify defects such as foreign
particles attached to said mark to be inspected.
A second object of the present invention is to provide a method for
detecting the defects in the pattern such as replicated foreign particles
with a high sensitivity with an apparently improved aligning precision
even when a sufficient aligning precision cannot be obtained between the
patterns of the mask to be inspected and of the inverted mask due to a
distortion in the projection or an error in the magnification of the
projection at the overlapped replication.
According to the present invention, the above-mentioned object is achieved
by a method which is featured by forming, on a photosensitive surface,
plural replication areas each formed by overlapped exposures of a pair of
mutually inverted patterns, wherein said paired patterns are mutually
shifted by a small distance in different directions respectively in said
plural areas, and inspecting the exposure status of said plural
replication areas after development of said photosensitive surface to
detect the defects on the reticle to be inspected.
In an embodiment of the present invention, said inspection of the exposure
status is conducted in a replication area where the overlapped exposures
of the paired patterns are attained most precisely among said plural
areas.
In another embodiment of the present invention, each of plural replication
areas subjected to overlapped exposures is divided into plural partial
areas, and said inspection of the exposure status is conducted in a
partial area where the overlapped exposures are made most precisely among
mutually corresponding partial areas of said plural replication areas.
A third object of the present invention is to provide a method for
automatically detecting foreign particles on a mask with a high detection
sensitivity.
According to the present invention, the above-mentioned object is achieved
by a method which is featured by utilizing light scattering for example of
laser beam in inspecting the photosensitive surface after development in
the foregoing methods.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a schematic perspective view of a reduction projection exposure
apparatus adapted for use in a replicating step in each embodiment of the
present invention;
FIG. 2 is a plan view of a mark reticle;
FIG. 3 is a plan view of a reticle to be inspected, having a positive
pattern;
FIG. 4 is a plan view of an inverted reticle having a negative pattern;
FIGS. 5A to 5E are schematic views showing replicating steps in a first
embodiment of the present invention;
FIG. 6 is a plan view showing the arrangement of marks on a wafer;
FIG. 7A is a plan view of a reticle to be inspected, having a positive
pattern;
FIG. 7B is a plan view of an inverted reticle having a negative pattern;
FIG. 8A is a plan view of the reticle to be inspected, having a negative
pattern formed by the replication of the pattern of the reticle shown in
FIG. 7A;
FIG. 8B is a plan view of the inverted reticle having a positive pattern
formed by the replication of the pattern of the reticle shown in FIG. 7B;
FIG. 9 is a plan view showing a group of replicated patterns;
FIG. 10 is a schematic perspective view of a laser scanning inspecting
apparatus adapted for use in an inspecting step in the embodiments of the
present invention;
FIG. 11 is a schematic view showing the principle of wafer positioning in
the laser scanning inspecting apparatus;
FIGS. 12A to 12C are wave form charts showing photoelectric output signals
corresponding to laser beam scanning;
FIG. 13 is a block diagram showing an example of a circuit for processing
the photoelectric signals in the laser scanning inspecting apparatus;
FIGS. 14A to 14D are schematic views showing replicating steps in another
embodiment of the present invention;
FIG. 15 is a plan view showing the positional relationship of positive and
negative patterns replicated onto a wafer according to a second embodiment
of the present invention;
FIG. 16 is a block diagram showing another embodiment of the control system
in the apparatus shown in FIG. 10;
FIG. 17 is a plan view showing partial areas in a replicated area;
FIG. 18 is a plan view showing partial area selected in different
replicated areas; and
FIG. 19 is a plan view showing an imaginary replicated area.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
FIG. 1 shows a reduction projection exposure apparatus adapted for the
method of the present invention, wherein a mark reticle R1, to be employed
in a step for replicating marks in advance onto a wafer coated with a
photoresist, is positioned above a projecting lens 1. The mark reticle R1
is provided, as shown in a plan view in FIG. 2, with marks RM1 and RM2 for
reticle alignment in two positions.
In an orthogonal coordinate system XY having the original point at the
center O of a replication area PA1 on the reticle R1, the mark RM1 is
positioned on the Y-axis while the mark RM2 is positioned on the X-axis.
The mark reticle R1 is also provided with three translucent marks MX, MY,
M.theta. in the peripheral part of an opaque area PA1. The marks MY and
M.theta. are positioned in two separate places in the system PA1 on a line
parallel to the X-axis of said coordinate system, while the mark MY is
positioned on a line parallel to the Y-axis. Each of said three marks MX,
MY, M.theta. has a grating structure containing grating elements diagonal
to the X and Y axes.
The mark RM1 of the mark reticle R1 is observed by a reticle alignment
microscope 2 provided with a mirror 2a, an objective lens 2b, a
half-mirror 2c etc. positioned at the side of an unrepresented light
source (above), while the mark RM2 is observed by a reticle alignment
microscope 3 composed of a mirror 3a, an objective lens 3b, a half-mirror
3c etc. The half-mirrors 2c, 3c are used for guiding the illuminating
light to the objective lenses 2b, 3b. These two microscopes 2, 3 are
respectively provided with references for aligning the observed images of
the marks RM1, RM2. A driving unit 4 for two-dimensionally moving the
reticle R1, including rotation thereof, so moves the reticle 1 that the
displacements of the marks RM1, RM2 and the references of two microscopes
2, 3 decrease. Thus the reticle R1 is aligned to the exposure apparatus,
whereby the optical axis l of the projecting lens 1 is so positioned to
pass through the center O of the reticle R1.
Immediately under the projecting lens 1 a wafer W is positioned so that the
pattern of the reticle is projected onto said wafer W. Said wafer W is
placed on a two-dimensionally movable stage 5 and is therefore rendered
movable in x- and y-directions of an orthogonal coordinate system xy by
means of a driving unit 6. The stage 5 is provided with a wafer holder 5a
capable of fixing the wafer W by suction and rotatable in the x-y plane
with respect to the stage 5. A coordinate position detector 7 measures the
two-dimensional position of the stage 5 with respect to a determined
original point, for example the optical axis l, for example by measuring
the amounts of displacement of the stage 5 in the x- and y-directions with
a laser interferometer. Around the projecting lens 1 there are provided
two wafer alignment microscopes 8, 9. The microscope 8 is so positioned
that the optical axis thereof is parallel to the optical axis l of the
projecting lens 1, and that said optical axis, or the center of
observation, crosses the x-axis of the coordinate system xy of which
original point is on the optical axis l.
On the other hand, the microscope 9 is so positioned that the optical axis
thereof is parallel to the optical axis l of the projecting lens 1 and
that said optical axis or the center of observation crosses the y-axis of
the coordinate system xy. The microscope 8 is used for detecting the wafer
position in the x-direction by observing the mark on said wafer, while the
microscope 9 is used for detecting the wafer position in the y-direction
by observing the mark on said wafer.
The reticle alignment microscopes 2, 3 and the wafer alignment microscopes
8, 9 have, in addition to the visual observation, a function of
photoelectric microscope by causing relative vibration between the
observed image of the mark and a slit and photoelectrically detecting the
displacement of the image of the mark with respect to the center of said
vibration, which is the aforementioned reference or the center of
observation.
Particularly the microscopes 8, 9 are preferably provided with a function
of so-called laser scanning microscope of causing a small vibration of a
laser beam spot on the wafer and photoelectrically detecting the light
scattered or diffracted by the mark on the wafer to identify the
displacement of the mark with respect to the center of vibration of the
laser beam spot.
FIG. 3 is a plan view of an example of the reticle R2 to be inspected,
having a true pattern to be replicated onto the wafer W. The dimension of
a replication area PA2 of the reticle R2 to be inspected is same as that
of the replication area PA1 of the mark reticle R1, and said reticle R2 is
provided with marks RM1, RM2 in the same manner and in the same positions
as the marks RM1, RM2 of the reticle R1. In FIG. 3, the hatched area is
composed of an evaporated opaque chromium layer, while the translucent
glass plate remains in the other area. The pattern in the replication area
PA2 will hereinafter be called positive pattern for the ease of
explanation.
On the other hand, FIG. 4 is a plan view of an inverted reticle R3 prepared
for executing the method of the present invention. The inverted reticle R3
is provided with marks RM1, RM2 in a similar manner and in the same
positions as in the reticles R1 and R2. The dimension of a replication
area PA3 is same as that of the area PA2 in the reticle R2. The pattern in
the area PA3 corresponds to the positive pattern in the area PA2 and is
formed as a negative pattern in which the translucent and opaque areas are
inverted.
The reticle R2 to be inspected and the inverted reticle R3 are prepared by
a knwon process of exposing a photoresist layer provided on a
chromium-evaporated transparent substrate to electron beam irradiation.
For the purpose of preparing the reticle R2, design data for forming the
true pattern are given to an electron beam irradiating apparatus. On the
other hand, at the preparation of the reticle R3, inverted data are
prepared by inverting said design data and are given to the electron beam
irradiating apparatus.
In the present invention, not only a reticle or a photomask is regarded as
a pattern-bearing transparent substrate, but also such substrate provided
with a dust-preventive thin layer is called reticle or photomask.
FIGS. 5A to 5E show replication steps in a first embodiment of the
inspecting method of the present invention.
At first a positive photosensitive material is coated on the wafer W to
form a uniform positive resist layer 10, which is subsequently baked. Said
wafer is placed on the wafer holder 5a of the exposure apparatus as shown
in FIG. 1. In this state the wafer is positioned, by means of a peripheral
notch, called orientation flat, to the wafer holder 5a. In the present
embodiment the direction of said orientation flat is selected parallel to
the x-axis moving direction of the movable stage 5.
On the other hand, the mark reticle R1 shown in FIG. 2 is positioned to the
exposure apparatus as shown in FIG. 1. Then there is repeated a step of
moving the stage 5 in the x- and y-directions by a determined pitch
through the driving unit 6 in response to the measurement of the detector
7 and illuminating the area PA1 of the reticle R1 for a determined period,
thus guiding light beams Lm from the marks MX, MY, M.theta. of the reticle
R1 to the positive resist layer 10 on the wafer, thereby forming reduced
latent images of the marks MX, MY, M.theta. in said layer.
Then the wafer is removed from the wafer holder 5a and the resist layer 10
is developed, whereby the resist corresponding to the latent images of the
marks MX, MY, M.theta. is removed to obtain recessed marks A corresponding
to the marks MX, MY, M.theta. as shown in FIG. 5B. On the wafer, as shown
in FIG. 6, said marks A are formed in the replication areas as the marks
AX, AY, A.theta. respectively corresponding to the marks MX, MY, M.theta..
Naturally said marks AX, AY, A.theta. are provided with diagonal lattice
structure.
Then the mark reticle R1 is removed from the exposure apparatus, and the
inverted reticle R3 shown in FIG. 4 is positioned to the exposure
apparatus by means of the marks RM1, RM2.
On the other hand, the wafer shown in FIG. 6 is again placed on the wafer
holder 5a and is positioned thereto by means of the orientation flat. The
photoresist material constituting the positive resist layer should
preferably retain the photosensitive property after the image development.
If the resist material loses the photosensitive property by the image
development, a resist layer must be formed again on the wafer shown in
FIG. 6.
When the wafer is fixed on the wafer holder 5a by vacuum suction, the
position of said wafer to the exposure apparatus is usually aberrated
slightly from the wafer position at the replication of the reticle R1.
Therefore, the wafer W is positioned again to the exposure apparatus
through the use of the microscopes 8, 9. The positioning procedure will be
briefly explained herein as it was detailedly disclosed in the U.S. Pat.
No. 4,385,838.
At first, among plural marks AY, A.theta. formed on a particular line in
the x-direction parallel to the orientation flat, a pair of marks AY,
A.theta. positioned closest to the periphery of the wafer is used for
detecting the rotational error and the position in y-direction of the
wafer. For this purpose the stage 5 is moved in such a manner that the
mark AY at the left-hand end of the wafer coincides with the center of
observation in the y-direction of the microscope 9. Then the stage 5 is
moved in the x-direction, and the mark A.theta. at the right-hand end of
the wafer is observed by the microscope 9. Then, from this position, the
stage 5 is moved in the y-direction until the mark A.theta. coincides with
the center of observation in the y-direction of the microscope 9. The
amounts of stage movements in the x- and y-directions in the
above-mentioned steps are detected by the detector 7 to determined the
rotational error of the wafer to the coordinate system xy. This rotational
error can be corrected within a certain limit by the rotation of the wafer
holder 5a, but a more precise correction is achieved, as disclosed in the
aforementioned U.S. Pat. No. 4,385,838, by forming a coordinate system
which is rotated from the coordinate system xy by a small amount
corresponding to the slight rotational error of the wafer and positioning
the stage according to thus formed coordinate system at the step movement
of the stage 5 for exposure.
After the amount of rotation of the wafer to the coordinate system xy is
determined in this manner, the y-coordinate value is determined when the
mark AY at the left-hand end of the wafer coincides with the center of
observation in the y-direction of the microscope 9, and then the
y-coordinate value is determined when a particular mark AX on the wafer
coincides with the center of observation in the x-direction of the
microscope 8. In this manner the positional relationship of the wafer in
the x- and y-directions with respect to the optical axis l of the
projecting lens 1.
Then there is conducted repeated a step of stepwise moving the stage 5 with
a same pitch as in the exposures of the mark reticle R1 by means of the
driving unit 6 and the detector 7 following the already replicated marks
AX, AY, A.theta., and illuminating the replication area PA3 of the
inverted reticle R3, thereby replicating the negative pattern in
succession onto the wafer. In this step the wafer is so positioned that
the marks AX, AY, A.theta. are present around the projected image of the
area PA3, in order to avoid the exposure of said marks AX, AY, A.theta..
FIG. 5C shows the step of negative pattern exposure, wherein a light beam
Lm through the transparent area of the reticle R3 is guided to the resist
layer 10 to form a reduced latent image of the negative pattern in said
resist layer 10. In said stepwise exposures of the negative pattern. One
of plural replication areas on the wafer having the marks AX, AY, A.theta.
is excluded from the exposure of the negative pattern.
Then the positive pattern of the reticle R2 to be inspected is replicated
onto the latent image formed in the resist layer 10 of said wafer.
For this purpose the reticle R3 is removed from the exposure apparatus, and
the reticle R2 is set and positioned on the exposure apparatus by means of
the marks RM1, RM2. In this state, therefore, the reticle R2 is aligned
with the wafer on the stage 5 indirectly through the exposure apparatus,
taking the marks AX, AY, A.theta. as references.
Then there is repeated a step of moving the stage 5 to a position same as
that for the stepwise exposure of the negative pattern in such a manner
that the projected image of the positive pattern of the reticle R2 to be
inspected is superposed on the previously replicated latent image of the
negative pattern of the inverted reticle R3, and illuminating the area PA2
of the reticle R2. Also in this case the marks AX, AY, A.theta. are
excluded from the exposure. FIG. 5D shows the state of this exposure. If
the transparent area of the reticle R2 is free from foreign particles, a
light beam Lm irradiates the entire resist layer 10 except the area
irradiated by the light Lm in FIG. 5C, thereby forming a reduced latent
image of the positive pattern. However, a foreign particle, if present,
intercepts the illuminating light to form a dark area Dm corresponding to
the foreign particle in the light beam Lm as shown in FIG. 5D. The resist
layer 10 is not exposed in said dark area Dm. Thus the resist layer 10
forms an unexposed area due only to the foreign particles adhered to the
reticle R2. In this manner the negative and positive patterns are
replicated, in mutually superposed manner, onto the resist layer 10. The
area of the wafer not exposed to the negative pattern is exposed only to
the positive pattern of the reticle R2.
Subsequently the wafer W is removed from the exposure apparatus and is
subjected to the development of the resist layer 10, whereby the areas
exposed to the positive and negative patterns are removed the unexposed
area resulting from the adhered foreign particle subsists on the wafer
surface as a remaining resist P as shown in FIG. 5E.
As explained in the foregoing, minute foreign particles present in the
transparent area of the reticle to be inspected are clearly marked on the
wafer as indicia, which can be detected and of which positions can be
determined by a wafer inspecting apparatus to be explained later. In the
above-described method the foreign particles present in the transparent
area of the inverted reticle are similarly marked on the wafer, but the
positions of the indicia on the wafer allow to identify whether the
foreign particles are present on the reticle to be inspected or on the
inverted reticle. However, a defect not present in the transparent area of
either reticle, for example a pinhole present in the opaque area of the
reticle or a partial lack at the edge or corner of the chromium pattern
does not form a remaining resist on the wafer. Consequently the defect in
the opaque area cannot be detected by the superposed replication of the
reticles R2, R3 shown in FIGS. 3 and 4.
FIGS. 7 and 8 show another embodiment enabling the detection of foreign
particles present in the transparent area of reticles and of defects
present in the opaque area.
A reticle R20 to be inspected, shown in FIG. 7A, is provided with four
alignment marks RM20 around a replication area PA20. Each alignment mark
is composed of a pair of marks, namely a transparent mark M1 surrounded by
an opaque area and an opaque mark M2 surrounded by a transparent area. The
pattern and the marks of said reticle R20 can be prepared, as in the case
of FIG. 3, by electron beam exposure according to design data.
An inverted reticle R30 shown in FIG. 7B is provided with alignment marks
RM30 and a pattern in a replication area PA30, both inverted from the
reticle shown in FIG. 7B, and can be prepared for example by electron beam
exposure according to inverted design data.
The pattern of said paired reticles R20, R30 are replicated on a wafer W in
superposed manner through steps similar to those shown in FIGS. 5A to 5E
to detect the foreign particles present in the transparent area. The marks
RM20, RM30 are used for alignment between the reticles and the wafer. When
the reticle R20 is set on the exposure apparatus shown in FIG. 1, the
marks M1 are used to enable observation of images A' (represented by chain
lines in FIG. 7A) of reference marks of the wafer projected on the
reticle, while the other marks M2 are used when the reticle 30 is set on
said apparatus.
A reticle R22 shown in FIG. 8A is provided with a negative pattern
laterally inverted from the pattern of the reticle R22 to be inspected
shown in FIG. 7A. The negative pattern of the reticle R22 is prepared by
replicating, either by contact or proximity printing, the entire pattern
of the reticle R20 onto a negative photoresist coated on a substrate. In
this step the pattern bearing face of the reticle R20 is so positioned as
to face the substrate to be exposed and the illumination is made from the
rear face of said reticle R20, so that a laterally inverted negative
pattern is formed in the replication area PA22 of the reticle R22.
Different from the negative pattern of the inverted reticle R30 shown in
FIG. 7B, the negative pattern prepared as explained above also contains
the defects of the reticle R20 to be inspected.
A reticle R33 shown in FIG. 8B has a positive pattern laterally inverted
from the pattern of the inverted reticle R30 shown in FIG. 7B. The
positive pattern of the reticle R33 is prepared from the inverted reticle
R30 in a similar manner as the reticle R22, and therefore contains the
defects present on the inverted reticle R30.
Two reticles R22, R33 thus prepared are aligned to the wafer by means of
the marks M2 of the marks RM22 and the marks M1 of the marks RM33 and are
replicated in succession on the wafer to form, on said wafer, indicia of
the foreign particles present in the transparent area of the reticles R22,
R33 corresponding to the defects in the opaque area of the reticles R20,
R30.
Consequently a complete inspection of the reticle to be inspected is
achieved by detection and positional determination of the remaining
resists on the first wafer prepared by superposed replication of the
reticles R20, R30 and on the second wafer prepared by superposed
replication of the reticles R22, R33.
In the following there will be given an explanation on an embodiment of the
wafer inspecting apparatus. The following description will deal with the
inspection of the first wafer only, since the first and second wafers can
be inspected in the same manner.
In a first embodiment of wafer inspection, a group of patterns as shown in
FIG. 9 is formed on the wafer W after image development. In FIG. 9 there
are only shown representative six replication areas S0-S5 arranged in
matrix form. It is assumed that the positive pattern of the reticle R2 to
be inspected is alone replicated in the area S0, while the positive and
negative patterns are superposedly replicated in other areas S1-S5. It is
further assumed that the six areas S0-S5 are positioned at a determined
pitch in relation to an orthogonal coordinate system .alpha..beta. defined
on the wafer.
At the wafer inspection, one compares said areas S1-S5 and determines
whether the remaining resists P1-P5 or N1-N5 exist in the corresponding
positions of said areas. Remaining resists which do not appear in the
corresponding positions but in random positions can be considered to have
resulted from dusts present on the photoresist of the wafer prior to the
exposure, or from deposition of fine particles of the photoresist at the
image development step. In such case it is therefore possible to conclude
that the foreign particle is not present at the corresponding position on
the reticle R2.
After confirming the presence of remaining resists P1-P5, N1-N5 in the
corresponding positions of the areas S1-S5, one determines whether similar
remaining resists exist in the corresponding positions in the area S0. In
the illustrated example, a remaining resist P0 exists, in the positive
pattern of the area S0, in a position corresponding to that of the
remaining resists P1-P5, whereas no resist is found in a position
corresponding to that of the remaining resists N1-N5. Consequently the
remaining resistors P0-P5 are identified to have resulted from a foreign
particle adhered to the transparent area of the reticle R2, while the
remaining resists N1-N5 are identified to have resulted from a foreign
particle adhered to the transparent area of the inverted reticle R3. It is
therefore rendered possible to conclude that a foreign particle, affecting
the production yield in the lithographic process, is present in a position
of the reticle R2 corresponding to the position of the remaining resist P0
in the area S0.
FIG. 10 shows an embodiment of the wafer inspecting apparatus suitable for
such inspection.
The wafer W after image development is placed on a two-dimensionally
movable table 20, which can be displaced in mutually orthogonal directions
x, y. As in the exposure apparatus shown in FIG. 1, the table 20 is
provided with a wafer holder which is rotatable to said table 20 and is
driven by an unrepresented driving unit. The wafer is in fact placed on
said wafer holder and is fixed by suction. The positions of said table 20
in the x- and y-directions are detected as coordinate values by position
detectors 21, 22 utilizing for example laser interferometers or optical
linear encoders. On the crossing point of the mutually orthogoanl
measuring lines of the position detectors 21, 22, there is provided an
XY-microscope 23 in such a manner that the optical axis or the center or
observation coincides with said crossing point. Said XY-microscope 23 is
provided with a function of photoelectric or laser scanning microscope for
detecting the marks AX, AY (or A.theta.) on the wafer, and is utilized to
achieve wafer alignment in the x- and y-directions.
There is further provided a .theta.-microscope 24 for detecting the mark
A.theta. (or AY) on the wafer to achieve wafer alignment in the
y-direction, and the optical axis or the center of observation is
positioned on the measuring line in the x-direction of the position
detector 21. The distance between the optical axes or the centers of
observation of the XY-microscope 23 and .theta.-microscope 24 is so
selected as to enable simultaneous observation of two distant marks, for
example the marks AY and A.theta., on the wafer.
Furthermore there is provided a microscope 25 for visually observing the
wafer surface and confirming the resists remaining thereon.
On the other hand there is provided a laser scanning device for detecting
the remaining resists. A laser beam 31 emitted from laser uhit 30 enters a
polygonal mirror 32 rotated anticlockwise at a determined speed, and is
focused into a spot by means of an unrepresented focusing lens, thus
performing a linear scanning motion at a constant speed in the x-direction
along a trajectory L. The scanning trajectory L is placed at a
predetermined position to the microscopes 23, 24, 25. There are also
provided a photosensor 33 for timing detection at the start of the
scanning trajectory L, and a photoelectric detector 34 for detecting the
light scattered from the edges of minute surface irregularities such as
remaining resists illuminated by the laser beam spot when the table 20 is
so moved that the scanning trajectory L crosses the wafer. Said
photoelectric detector 34 is provided with a condenser lens of a
determined steric entrance angle for receiving the light scattered from
any part of the scanning trajectory L. The optical axis of the detector 34
diagonally looks at the center of the scanning trajectory L. The laser
beam 31 enters the wafer surface preferably in substantially perpendicular
direction, in order to precisely determine the incident position of the
laser beam.
The detection of the remaining resists on the wafer is achieved by raster
scanning the wafer with the laser beam spot, by displacing the table 20 in
the y-direction while causing the laser beam spot to scan in the
x-direction.
The operation of remaining resist detection is achieved by the
above-described wafer inspecting apparatus in the following manner.
At first a wafer W as shown in FIG. 6 is positioned, by means of the
orientation flat thereof, on the wafer holder on the table 20. The
orientation flat is positioned to be parallel to the x-direction as shown
in FIG. 10.
Then the table 20 is so moved that the wafer is positioned under the
microscopes 23, 24 as shown in FIG. 11, and one of the marks AY on the
wafer is observed by the XY-microscope 23 while one of the marks A.theta.
on the wafer is observed by the .theta.-microscope.
For the ease of following description, it is assumed that said mark AY is
the mark AY3 in the area S3 in FIG. 9, and said mark A.theta. is the mark
A.theta.5 in the area S5. Consequently the distance between the marks AY3
and A.theta.5 is selected equal to the distance Dx.theta. between the
microscopes 23 and 24. In FIG. 9, a coordinate axis passing through the
marks AY3 and A.theta.5 will be called .alpha.1.
Then the table 20 is so displaced that the mark AY3 coincides with a
reference 23b in the y-direction of the XY-microscope 23, and the wafer
holder is so rotated that the mark A.theta.5 coincides with a reference
24a in the y-direction of the .theta.-microscope 24. In this manner the
wafer is aligned in rotational direction.
In this state, in a counter (hereinafter called .beta.-counter) of the
position detector 22, there is preset a value corresponding to a distance
Dy from the reference 23b of the XY-microscope 23 to the scanning
trajectory L of the laser beam spot.
Subsequently the table 20 is so displaced in the x-direction that one of
the marks AX on the wafer, i.e. the mark AX3 in the area S3 in the
illustrated case, coincides with the reference 23a in the x-direction of
the XY-microscope 23. Then a counter (hereinafter called .alpha.-counter)
of the position detector 21 is preset for example to zero.
The above-described procedure allows the wafer to be positioned with
respect to the scanning trajectory L of the laser beam spot. More
specifically, the scanning trajectory L coincides with the coordinate axis
.alpha.1 of the coordinate system .alpha..sub.1 .beta. on the wafer when
the .beta.-counter is brought to zero by the displacement of the table 20,
and the center C of the scanning trajectory L coincides with the
coordinate axis .beta. when the .alpha.-counter is brought to zer | | |
