U.S. patent application number 17/058529 was filed with the patent office on 2021-07-15 for real-time autofocus for maskless lithography on substrates.
The applicant listed for this patent is Applied Materials, Inc.. Invention is credited to Glen Alan GOMES, Zheng GU, Rendong LIN, Diana VALVERDE-PANIAGUA, Meenaradchagan VISHNU, Zhongchuan ZHANG.
Application Number | 20210216023 17/058529 |
Document ID | / |
Family ID | 1000005533414 |
Filed Date | 2021-07-15 |
United States Patent
Application |
20210216023 |
Kind Code |
A1 |
VALVERDE-PANIAGUA; Diana ;
et al. |
July 15, 2021 |
REAL-TIME AUTOFOCUS FOR MASKLESS LITHOGRAPHY ON SUBSTRATES
Abstract
Embodiments of the systems and methods discussed herein
autofocus an imaging apparatus by pre-processing image data
received via channels of the imaging system that include laser
beams and sensors configured to receive image data when laser beams
are applied across a substrate in a pixel-wise application across a
substrate. The substrate can comprise both a photoresist and
metallic material, and the images as-received by the sensors
include noise from the metallic material. During pre-processing of
the image data, a percentage of noise to remove from the image data
is determined, and the image data is filtered. A centroid of the
substrate is calculated for each channel and a focus deviation for
the exposure is determined. The centroids can be combined using one
or more filtering mechanisms, and the imaging system can be
autofocused in an exposure position by moving the stage and/or the
exposure source in one or more directions.
Inventors: |
VALVERDE-PANIAGUA; Diana;
(Sunnyvale, CA) ; ZHANG; Zhongchuan; (Fremont,
CA) ; LIN; Rendong; (San Jose, CA) ; GU;
Zheng; (Emeryville, CA) ; VISHNU; Meenaradchagan;
(San Jose, CA) ; GOMES; Glen Alan; (San Jose,
CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Applied Materials, Inc. |
Santa Clara |
CA |
US |
|
|
Family ID: |
1000005533414 |
Appl. No.: |
17/058529 |
Filed: |
April 9, 2019 |
PCT Filed: |
April 9, 2019 |
PCT NO: |
PCT/US2019/026445 |
371 Date: |
November 24, 2020 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
62692323 |
Jun 29, 2018 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G03F 7/70641 20130101;
G03F 7/70358 20130101; G03F 7/70716 20130101 |
International
Class: |
G03F 7/20 20060101
G03F007/20 |
Claims
1. A method of patterning a substrate, comprising: applying a
plurality of laser beams across a substrate while the substrate is
positioned on a stage in an imaging apparatus in an initial
position relative to an exposure source of the imaging apparatus,
the substrate comprising a photoresist and a metallic material;
receiving, by a plurality of sensors coupled to the imaging
apparatus, in response to applying the plurality of laser beams, a
plurality of images of each pixel of a plurality of pixels;
determining a percentage of noise in the plurality of images
received from each sensor for each pixel of the plurality of pixels
in response to the applying of the plurality of laser beams to the
metallic material of the substrate; filtering out the percentage of
noise from the plurality of images from each sensor for each pixel
of the plurality of pixels; subsequently, calculating a centroid of
the substrate based on the plurality of images; determining, based
on the calculating of the centroid for each sensor of the plurality
of sensors, a focus deviation; adjusting, the substrate to an
exposure position from the initial position, the exposure position
being different than the initial position, and the adjusting the
exposure source autofocusing the exposure source on the substrate;
and subsequent to adjusting the substrate, exposing the substrate
to the exposure source, wherein exposing the substrate forms a
plurality of features in the substrate.
2. The method of claim 1, further comprising: combining each
centroid for each sensor of the plurality of sensors, wherein the
combining comprises Kalman filtering,
proportional-integral-derivative filtering, or linear filtering;
and determining, based on the combining and the focus deviation,
the exposure position of the exposure source.
3. The method of claim 1, wherein the substrate comprises a
photoresist and a metallic material, wherein the percentage of
noise is noise in the plurality of images received in response to
the plurality of laser beams reflecting off of the metallic
material during the applying of the plurality of laser beams, and
wherein percentage of noise is from 1% to 30%.
4. The method of claim 1, further comprising determining the
centroid using a subset of the plurality of images.
5. A method of patterning a substrate, comprising: positioning, in
an imaging apparatus, a substrate comprising a photoresist and a
metallic material, the substrate being positioned on a stage in the
imaging apparatus in an initial position relative to an exposure
source of the imaging apparatus; determining, by an application
stored on a non-transitory memory of a server and executable by a
processor, an exposure position for the imaging apparatus, the
determining the exposure position comprising: applying a plurality
of laser beams to each pixel of the plurality of pixels; receiving,
by a plurality of sensors coupled to the imaging apparatus, a
plurality of images from each pixel of the plurality of pixels in
response to applying the plurality of laser beams; determining a
percentage of noise of the plurality of images for each sensor for
each pixel of the plurality of pixels; filtering out the percentage
of noise from the plurality of images from each sensor for each
pixel of the plurality of pixels; subsequently, calculating a
centroid of the substrate for each sensor of the plurality of
sensors; determining, based on the calculating of the centroid for
each sensor of the plurality of sensors, a focus deviation;
combining each centroid for each sensor; and determining, based on
the combining and the focus deviation, the exposure position of the
exposure source; and adjusting, by the application, at least one of
the stage or the exposure source from the initial position to the
exposure position, the exposure position being different than the
initial position, and the adjusting the exposure source
autofocusing the exposure source on the substrate.
6. The method of claim 5, further comprising, subsequent to
adjusting the stage, exposing the substrate to the exposure source,
wherein exposing the substrate forms a plurality of features in the
substrate.
7. The method of claim 5, wherein the combining comprises Kalman
filtering, proportional-integral-derivative filtering, or linear
filtering.
8. The method of claim 5, wherein the substrate comprises a
photoresist and a metallic material, wherein the percentage of
noise is noise in the plurality of images received in response to
the plurality of laser beams reflecting off of the metallic
material during the applying of the plurality of laser beams, and
wherein the percentage of noise is from 1% to 30%.
9. The method of claim 5, further comprising, adjusting, by the
application, a velocity of the substrate from a first velocity to a
second velocity based on the combining and the focus deviation.
10. A non-transitory computer-readable medium including
instructions that are configured to cause a computing system to:
determine an exposure position for an imaging apparatus, the
determining the exposure position comprising: apply a plurality of
laser beams to each pixel of a plurality of pixels of a substrate;
receive, by a plurality of sensors coupled to the imaging
apparatus, a plurality of images from each pixel of the plurality
of pixels in response to applying the plurality of laser beams;
determine a percentage of noise of the plurality of images for each
sensor for each pixel of the plurality of pixels; filter out the
percentage of noise from the plurality of images from each sensor
for each pixel of the plurality of pixels; subsequently, calculate
a centroid of the substrate for each sensor of the plurality of
sensors; determine, based on the calculation of the centroid for
each sensor of the plurality of sensors, a focus deviation; combine
each centroid for each sensor; and determine, based on the
combination and the focus deviation, the exposure position of an
exposure source; and adjust the substrate to the exposure position
from an initial position, the exposure position being different
than the initial position, and the adjustment of the exposure
source autofocuses the exposure source on the substrate.
11. The non-transitory computer-readable medium of claim 10,
wherein the instructions are further configured to, subsequent to
adjustment of the substrate, expose the substrate to the exposure
source to form a plurality of features on the substrate.
12. The non-transitory computer-readable medium of claim 10,
wherein the instructions are further configured to combine each
centroid by Kalman filtering, proportional-integral-derivative
filtering, or linear filtering.
13. The non-transitory computer-readable medium of claim 10,
wherein the substrate comprises a photoresist and a metallic
material, wherein the percentage of noise is noise in the plurality
of images received in response to the plurality of laser beams
reflecting off of the metallic material during the applying of the
plurality of laser beams, and wherein the percentage of noise is
from 1% to 30%.
14. The non-transitory computer-readable medium of claim 10,
wherein the percentage of noise is from 5% to 20%.
15. The non-transitory computer-readable medium of claim 14,
wherein a height of a top surface across the substrate, as measured
perpendicular to the substrate, is not uniform.
Description
BACKGROUND
Field
[0001] Embodiments of the present disclosure generally relate to
apparatuses, systems and methods for processing one or more
substrates, and more specifically to apparatuses, systems and
methods for performing photolithography processes.
Description of the Related Art
[0002] Photolithography is widely used in the manufacturing of
semiconductor devices and display devices such as liquid crystal
displays (LCDs). Large area substrates are often utilized in the
manufacture of LCDs. LCDs, or flat panels, are commonly used for
active matrix displays, such as computers, touch panel devices,
personal digital assistants (PDAs), cell phones, television
monitors, and the like. Generally, flat panels include a layer of
liquid crystal material forming pixels sandwiched between two
plates. When power from a power supply is applied across the liquid
crystal material, an amount of light passing through the liquid
crystal material is controlled at pixel locations enabling images
to be generated.
[0003] Microlithography techniques have been employed to create
electrical features incorporated as part of the liquid crystal
material layer forming the pixels. According to these techniques, a
light-sensitive photoresist is applied to at least one surface of
the substrate. Then, a pattern generator exposes selected areas of
the light-sensitive photoresist as part of a pattern with light to
cause chemical changes to the photoresist in the selective areas to
prepare these selective areas for subsequent material removal
and/or material addition processes to create the electrical
features. The substrates patterned during microlithography can have
a variation in thickness across the substrate that is greater than
the degrees of freedom of the microlithography system. This can
lead to the system being out of focus with respect to the substrate
and results in a blurring of one or more pixels in the display
which can lead to mura or other undesirable effects in the
display.
[0004] In order to continue to provide display devices and other
devices at the prices demanded by consumers, new apparatuses and
approaches are needed to precisely and cost-effectively create
patterns on substrates, such as large area substrates, including
apparatuses and approaches for real-time autofocusing of the
imaging apparatus prior to exposing the substrate.
SUMMARY
[0005] In an embodiment, a method of patterning a substrate,
comprising: applying a plurality of laser beams across a substrate
while the substrate is positioned on a stage in an imaging
apparatus in an initial position relative to an exposure source of
the imaging apparatus, the substrate comprising a photoresist and a
metallic material; receiving, by a plurality of sensors coupled to
the imaging apparatus, in response to applying the plurality of
laser beams, a plurality of images of each pixel of the plurality
of pixels; and determining a percentage of noise in the plurality
of images received from each sensor for each pixel of the plurality
of pixels in response to the applying of the plurality of laser
beams to the metallic material of the substrate. Further in the
embodiment, the method comprises: filtering out the percentage of
noise from the plurality of images from each sensor for each pixel
of the plurality of pixels; subsequently, calculating a centroid of
the substrate based on the plurality of images; determining, based
on the calculating of the centroid for each sensor of the plurality
of sensors, a focus deviation; and adjusting, the substrate to the
exposure position from the initial position, the exposure position
being different than the initial position, and the adjusting the
exposure source autofocusing the exposure source on the
substrate.
[0006] In an alternate embodiment, a method of patterning a
substrate, comprising: positioning, in an imaging apparatus, a
substrate comprising a photoresist and a metallic material, the
substrate being positioned on a stage in the imaging apparatus in
an initial position relative to an exposure source of the imaging
apparatus; determining, by an application stored on a
non-transitory memory of a server and executable by a processor, an
exposure position for the imaging apparatus, the determining the
exposure position comprising: applying a plurality of laser beams
to each pixel of the plurality of pixels; receiving, by a plurality
of sensors coupled to the imaging apparatus, a plurality of images
from each pixel of the plurality of pixels in response to applying
the plurality of laser beams; determining a percentage of noise of
the plurality of images for each sensor for each pixel of the
plurality of pixels; and filtering out the percentage of noise from
the plurality of images from each sensor for each pixel of the
plurality of pixels; subsequently, calculating a centroid of the
substrate for each sensor of the plurality of sensors; determining,
based on the calculating of the centroid for each sensor of the
plurality of sensors, a focus deviation; combining each centroid
for each sensor; and determining, based on the combining and the
focus deviation, the exposure position of the exposure source.
Further in the embodiment, the method comprises adjusting, by the
application, at least one of the stage or the exposure source from
the initial position to the exposure position, the exposure
position being different than the initial position, and the
adjusting the exposure source autofocusing the exposure source on
the substrate.
[0007] In an embodiment, a non-transitory computer-readable medium
including instructions that are configured to cause a computing
system to: determine an exposure position for an imaging apparatus,
the determining the exposure position comprising: apply a plurality
of laser beams to each pixel of a plurality of pixels of a
substrate; receive, by a plurality of sensors coupled to the
imaging apparatus, a plurality of images from each pixel of the
plurality of pixels in response to applying the plurality of laser
beams; determine a percentage of noise of the plurality of images
for each sensor for each pixel of the plurality of pixels; filter
out the percentage of noise from the plurality of images from each
sensor for each pixel of the plurality of pixels; subsequently,
calculate a centroid of the substrate for each sensor of the
plurality of sensors; determine, based on the calculation of the
centroid for each sensor of the plurality of sensors, a focus
deviation; combine each centroid for each sensor; and determine,
based on the combination and the focus deviation, the exposure
position of the exposure source. Further in the embodiment, the
instructions are configured to adjust, by the application, the
substrate to the exposure position from the initial position, the
exposure position being different than the initial position, and
the adjustment of the exposure source autofocuses the exposure
source on the substrate.
BRIEF DESCRIPTION OF THE DRAWINGS
[0008] So that the manner in which the above recited features of
the present disclosure can be understood in detail, a more
particular description of the disclosure, briefly summarized above,
may be had by reference to embodiments, some of which are
illustrated in the appended drawings. It is to be noted, however,
that the appended drawings illustrate only exemplary embodiments
and are therefore not to be considered limiting of its scope, and
may admit to other equally effective embodiments.
[0009] FIG. 1A is a perspective view of a photolithography system
according to embodiments disclosed herein.
[0010] FIG. 1B is a perspective view of an alternate
photolithography system according to embodiments disclosed
herein.
[0011] FIG. 2A is a perspective schematic view of an image
projection apparatus according to embodiments disclosed herein.
[0012] FIG. 2B is an image projection apparatus according to
embodiments described herein.
[0013] FIG. 3 illustrates a schematic view of a plurality of image
projection apparatuses according to embodiments of the present
disclosure.
[0014] FIG. 4 is a partial schematic illustration of a solid state
emitter device according to embodiments of the present
disclosure.
[0015] FIG. 5 is a flow chart of a method 500 of autofocusing on a
substrate during a substrate patterning operation.
[0016] FIG. 6A is an example graph of reflected height intensity
data across a substrate.
[0017] FIG. 6B shows an example graph of pixel height data across a
substrate in an ideal case with a normal or near-normal Gaussian
distribution.
[0018] FIG. 6C illustrates a centroid calculation on an example
graph that results in a centroid value offset from a center of the
peak.
[0019] FIG. 6D illustrates an example graph of pixel height across
a substrate and a centroid calculation according to embodiments of
the present disclosure.
[0020] To facilitate understanding, identical reference numerals
have been used, where possible, to designate identical elements
that are common to the figures. It is contemplated that elements
and features of one embodiment may be beneficially incorporated in
other embodiments without further recitation.
DETAILED DESCRIPTION
[0021] Embodiments of the present disclosure provide improvements
in autofocusing during digital photolithography. Displays of
varying sizes can be fabricated using digital photolithography. The
digital photolithography systems discussed herein, which can be
referred to as imaging systems, use one or more exposure sources
comprising solid state emitter devices for operations including
patterning photoresists.
[0022] Various features of display-sized glass substrates, such as
substrate composition and size, make maintaining focus during
patterning difficult as compared to maintaining focus while
patterning substrates containing only silicon. For example, display
substrates can have dimensions of 1500 mm.times.1850 mm or larger,
and can have a highly non-uniform thickness across the substrate.
The thickness and the non-uniformity of the thickness across the
substrate can be further varied as layers are formed and patterned.
The non-uniformity is not able to be removed (smoothed) using
chemical mechanical polishing (CMP).
[0023] Digital photolithography systems have tight tolerances, for
example, +/-3 microns, for focus error due to the small
depth-of-focus lenses used for sub-micron printing resolution. The
digital photolithography systems tolerance is therefore less than
the variation in uniformity across a substrate, which can vary by
+/-10 microns or more. Therefore, as substrates increase in size
and vary in composition, it becomes harder to autofocus the imaging
system.
[0024] The systems and methods discussed herein are directed
towards autofocusing an imaging apparatus during patterning of a
substrate. The autofocusing discussed herein is achieved by
positioning the exposure source in an exposure position relative to
a substrate. As discussed herein, an "exposure position" is a
position of a stage in an imaging apparatus relative to an exposure
source of the imaging apparatus, and the stage is a component of an
imaging system that is configured to secure a substrate during
digital photolithography. The exposure position is determined in
real-time by a pixel-wise application. Pixel-wise image processing
as discussed herein includes processing a plurality of images. The
pixel-wise image processing is used for autofocusing herein, and
includes running a routine on a location on a substrate for data
received by a sensor and then re-running the same routine (e.g.,
the substrate height analysis discussed herein) on multiple
locations in real-time. In one example, the substrate comprises a
plurality of pixels, and the routine is run on each pixel. In this
pixel-wise autofocusing operation, a position of the substrate
relative to the position of the exposure source is determined using
imaging data, including a pre-processing of the imaging data to
remove extraneous reflections from the underlying layers, referred
to herein as "noise," and improve accuracy of the autofocusing. The
autofocus input data used herein is provided via a plurality of
linear imaging sensors such as complementary metal oxide
semiconductor (CMOS) sensors disposed in the imaging apparatus. The
sensors are coupled the imaging apparatus to capture the reflection
of the focus lasers off the substrate surface. In one example, a
surface of each pixel across a substrate is scanned in real time
while the substrate is positioned in the imaging apparatus prior to
patterning. On non-patterned metal or glass plates, a CMOS sensor
will capture a single Gaussian-like peak associated with the
substrate reflection, where the location of the peak indicates the
distance of the surface to the sensor. Over a patterned substrate,
however, such as a patterned substrate comprising metallic material
and a photoresist, metal reflections distort the photoresist
reflections in various undesirable ways, which negatively impacts
the ability of the system to focus, let alone allow for real-time
autofocusing.
[0025] The autofocusing discussed herein can be applied to
substrates comprising mixed materials such as photoresists and
metallic materials by capturing and pre-processing a plurality of
images of the substrate. The exposure position can be determined by
collecting a plurality of image data from each pixel across a
substrate and determining (1) a percentile filter to remove noise
from the images due to time-varying metal reflections; as well as
(2) a centroid computation to estimate the focus deviation; and (3)
a Kalman filter to combine the data from the multiple sensors taken
in real time across multiple pixels. This information is combined
and used as discussed below to determine an exposure position to
autofocus before patterning the substrate. When an exposure
position is determined, the imaging apparatus adjusts a stage on
which the substrate is positioned and/or the exposure source to the
exposure position prior to exposing and patterning the substrate.
Thus, autofocus-related mura, an undesirable effect discussed
below, is eliminated or reduced over challenging multi-layer
substrates using this pixel-wise application.
[0026] When an exposure or exposures are performed out of focus on
a substrate, an undesirable visual phenomenon called mura occurs.
Mura amplifies the microscopic effects of slight focus deviations,
leading to defective displays. This challenge is amplified for
display substrates with varying compositions, such as when the
substrate comprises a photoresist as well as metallic material.
When focus beams comprising laser beams are used to determine a
position of the exposure source, there will be a difference in the
reflection received by sensors in response to the reflection of the
laser beams off of the photoresist and the metallic materials. The
composition of the substrate, as discussed herein, includes a
metallic material and a photoresist, the photoresist is disposed
over the metallic material (and other materials that can be formed
on the substrate such as dielectric layers) prior to exposure.
Using the systems and methods discussed here, a substrate
positioned in an imaging apparatus is exposed to the exposure
source (i.e., the substrate is patterned). The substrate is exposed
when the exposure source is focused on the substrate and is in an
exposure position. The exposure position is determined based upon
deviations among and between pixel heights across the substrates.
Depending upon the embodiment, the stage, the exposure source, or
both, can be adjusted along one or more axis to an exposure
position subsequent to execution of the methods discussed herein in
order to autofocus the imaging system during substrate
patterning.
[0027] The autofocusing discussed herein allows for the continual
adjustment of an imaging system such as a digital lithography
system to produce an optimal, focused, exposure for the substrate
during one or more exposures during a patterning process that may
also be referred to as an etching process. The imaging apparatus
comprises a plurality of channels, each channel comprising a laser
beam and a sensor such as CMOS sensors. The images discussed herein
are comprised of a low-frequency photoresist signal and a
time-varying, high-frequency metal signal. Both signals are
positive and additive and taking merely the centroid of the signals
may not result in a focused exposure because extraneous reflections
of the metallic material may cause errors in the centroid
calculation. Over time spans from 0.1 ms to 5 ms, such as those
used to collect the images discussed herein, the signals from the
photoresist have less variation as compared to signals from the
metal, which have more variation and add noise to the images.
Additionally, the CMOS sensor noise is both positive and negative.
Therefore, when the methods discussed herein are used to
pre-process the images by removing the noise from the metallic
material, the noise from the sensor is not amplified. Thus, this
pixel-wise application autofocuses the imaging system prior to
exposure by determining an exposure position of the imaging
apparatus relative to the substrate based on the height of the
pixels across the substrate.
[0028] The images collected as discussed herein are filtered using
adaptive percentile filtering. The scanning occurs in real-time,
and the auto-focusing is performed continuously during scanning.
The filtering removes a portion of the noise from the images
received by the sensors. The undesirable noise extraneous
reflections from the underlying layers of the substrate result from
the reflection of the laser beams off of the metallic material. The
filtering removes a plurality of the noise from the images and
results in enhanced accuracy for autofocusing, reducing the
instances of the formation of mura and other undesirable effects in
the display. By scanning over patterned substrates, an image is
generated of a low-frequency photoresist signal and of time-varying
high-frequency metal signals. Therefore, across scanning windows
that are larger than the feature size of the pattern, the
photoresist signal will be approximately constant, while the metal
signals will not. Further, the photoresist and metal signals are
positive and additive. Because the metal reflections are changing
rapidly, this means that, without sensor noise, taking the minimum
pixel value (e.g., pixel height) of a series of frames would result
in the closest approximation of the photoresist signal (i.e.
high-frequency metal-induced signals will be removed). Furthermore,
there is a small amount of both positive and negative sensor noise
generated during autofocusing. As such, by determining a
predetermined percentile of noise to remove from the plurality of
images, there is a more accurate, real-time autofocus determination
made for each substrate that balances eliminating the (larger)
metal noise without amplifying the smaller sensor noise. The
autofocus inputs are multidimensional, including x coordinates and
y coordinates over the substrates, the exposure source position,
the linear sensor pixel position, and a light intensity. As
discussed herein, either or both of a stage on which a substrate is
positioned in the imaging system and/or the exposure source can be
moved in the x direction, the y direction, or the z direction
during the autofocusing operation.
[0029] Conventional technology can determine a peak of a graph of
pixel amplitude (height) for each of n pixels across a substrate.
The peak may be a format in which the CMOS or other image sensors
of the imaging system provide the reflected light. However, using
the autofocus systems and methods discussed herein performs a
centroid calculation to determine a mean, e.g., not necessarily a
peak, of the pixel amplitude curves. This is discussed below and
shown in FIGS. 6A-6D. Each sensor of the plurality of sensors of
the imaging system produces at least one image for each pixel on
the substrate. The images are combined and analyzed among and
between substrates and a centroid is determined for each image
generated from each sensor of plurality of sensors. A centroid is a
center of mass measurement that multiplies each pixel by its
position (height) and dividing that number by a total number of
pixels scanned to determine a center measurement of pixel height
(e.g., an average). That is, each pixel has, for example, three
lasers applied to it, and each sensor is configured to capture an
image from each laser. Thus, each pixel produces 3 images, and the
images from each sensor are combined to form an image of the
substrate. The centroid discussed herein is a center of mass
measurement for the plurality of images received by each sensor and
is assumed to have a Gaussian distribution. The centroids are used
to determine a focus deviation, that is, how an initial position of
the exposure source relative to the substrate differs from a
second, exposure, position for the exposure source that will be
used to pattern the substrate. The centroid offset from the target
substrate and the current stage position is used to estimate a
height of the substrate. The centroids calculated for each sensor
as discussed herein are combined using a filter such as a Kalman
filter, a linear filter, or a PID filter as discussed below to
generate a model estimation of the exposure position. The
calculation and the model can be combined and/or compared to
determine the exposure position as discussed below.
[0030] FIG. 1A is a perspective view of a photolithography system
100A according to embodiments disclosed herein. The system 100A
includes a base frame 110, a slab 120, a stage 130, and a
processing apparatus 160. The base frame 110 rests on the floor of
a fabrication facility and supports the slab 120. A plurality of
passive air isolators 112 are positioned between the base frame 110
and the slab 120. In one embodiment, the slab 120 is a monolithic
piece of granite, and the stage 130 is disposed on the slab 120. A
substrate 140 is supported by the stage 130. A plurality of holes
(not shown) are formed in the stage 130 for allowing a plurality of
lift pins (not shown) to extend therethrough. In some embodiments,
the lift pins rise to an extended position to receive the substrate
140, such as from one or more transfer robots (not shown). The one
or more transfer robots are used to load and unload a substrate 140
from the stage 130.
[0031] The substrate 140 comprises any suitable material, for
example, quartz, is used as part of a flat panel display. In other
embodiments, the substrate 140 is made of other materials capable
of being used as a part of the flat panel display. In some
embodiments, the substrate 140 has a photoresist layer formed
thereon which is sensitive to radiation. A positive photoresist
includes portions of the photoresist, which, when exposed to
radiation, are respectively soluble to a photoresist developer
applied to the photoresist after the pattern is written into the
photoresist. A negative photoresist includes portions of the
photoresist, which, when exposed to radiation, will be respectively
insoluble to photoresist developer applied to the photoresist after
the pattern is written into the photoresist. The chemical
composition of the photoresist determines whether the photoresist
will be a positive photoresist or negative photoresist. Examples of
photoresists include, but are not limited to, at least one of
diazonaphthoquinone, a phenol formaldehyde resin, poly(methyl
methacrylate), poly(methyl glutarimide), and SU-8. In this manner,
the pattern is created on a surface of the substrate 140 to form
the electronic circuitry.
[0032] The system 100A includes a pair of supports 122 and a pair
of tracks 124. The pair of supports 122 are disposed on the slab
120, and the slab 120 and the pair of supports 122 are a single
piece of material. The pair of tracks 124 are supported by the pair
of the supports 122, and the stage 130 moves along the tracks 124
in the X-direction as indicated by the coordinate system shown in
FIG. 1A. In one embodiment, the pair of tracks 124 is a pair of
parallel magnetic channels. As shown, each track 124 of the pair of
tracks 124 is linear. In other embodiments, one or more track 124
is non-linear. An encoder 126 is coupled to the stage 130, the
encoder 126 is configured to provide location information to a
controller 190.
[0033] In an embodiment, the processing apparatus 160 includes a
support 162 and a processing unit 164. The support 162 is disposed
on the slab 120 and includes an opening 166 for the stage 130 to
pass under the processing unit 164. The processing unit 164 is
supported by the support 162. In one embodiment, the processing
unit 164 is a pattern generator configured to expose a photoresist
in a photolithography process. In some embodiments, the pattern
generator is configured to perform a maskless lithography process.
The processing unit 164 includes a plurality of image projection
apparatuses (shown in FIGS. 2A and 2B). In one embodiment, the
processing unit 164 contains as many as 84 image projection
apparatus. Each image projection apparatus is disposed in a case
165. The processing apparatus 160 is useful to perform maskless
direct patterning.
[0034] The controller 190 facilitates the control and automation of
the processing techniques described herein. The controller 190 is
coupled to or in communication with one or more of the processing
apparatus 160, the one or more stages 130, and the encoder 126. The
processing apparatus 160 and the one or more stages 130 provide
information to the controller regarding the substrate processing
and alignment. For example, the processing apparatus 160 provides
information to the controller 190 to alert the controller 190 that
substrate processing is complete. In another example, the
controller 190 is configured to execute a plurality of instructions
to determine an exposure position for the imaging apparatus in
real-time by scanning the substrate pixel-by-pixel to autofocus the
exposure source prior to exposure (patterning) of the substrate. In
one example, the controller 190 adjusts the one or more stages 130
to an initial position when a substrate is positioned on or in the
one or more stages. In still another example, the controller 190 is
configured to execute a plurality of instructions to determine a
velocity of the exposure scanning and an exposure position for the
imaging apparatus in real-time by scanning the substrate
pixel-by-pixel to autofocus the exposure source prior to exposure
(patterning) of the substrate.
[0035] The controller 190 includes a central processing unit (CPU)
192, memory 194, and support circuits (or I/O) 196. The CPU 192 is
one of any form of computer processors used in industrial settings
for controlling various processes and hardware (e.g., pattern
generators, motors, and other hardware) and/or monitoring the
processes (e.g., processing time and substrate position). The
memory 194 is connected to the CPU 192. The memory 194 is one or
more of a readily available memory, such as random access memory
(RAM), read only memory (ROM), floppy disk, hard disk, or any other
form of digital storage, local or remote. Software instructions and
data are coded and stored within the memory 194 for instructing the
CPU 192. The support circuits 196 are also connected to the CPU for
supporting the processor in a conventional manner. The support
circuits include conventional cache, power supplies, clock
circuits, input/output circuitry, subsystems, and the like. A
program (or computer instructions), which may be referred to as an
imaging program, is readable by the controller determines which
tasks are performable on a substrate. The program is software
readable by the controller and includes code to monitor and
control, for example, the processing time and substrate
position.
[0036] During operation, the stage 130 moves in the X-direction at
a predetermined velocity from a loading (first) position, as shown
in FIG. 1A, to a processing (second) position. The processing
position is one or more positions of the stage 130 as the stage 130
passes under the processing unit 164. During operation, the stage
130 is be lifted by a plurality of air bearings (not shown) and the
stage 130 moves while lifted along the pair of tracks 124 from the
loading position to the processing position. A plurality of
vertical guide air bearings (not shown) are coupled to the stage
130 and positioned adjacent an inner wall 128 of each support 122
in order to stabilize the movement of the stage 130. The stage 130
also moves in the Y-direction by moving along a track 150 for
processing and/or indexing the substrate 140. The stage 130 is
capable of independent operation and can scan a substrate 140 in
one direction and step in the other direction.
[0037] A metrology system (not shown) measures the X and Y lateral
position coordinates of each of the stage 130 in real-time so that
each of the plurality of image projection apparatuses can
accurately locate the patterns being written in a
photoresist-covered substrate 140. The metrology system also
provides a real-time measurement of an angular position of the
stage 130 about the vertical or Z-axis. The angular position
measurement may be used to hold the angular position of the stage
130 constant during scanning by means of a servo mechanism (not
shown), or the angular position measurement may be used to apply
corrections to the positions of the patterns being written on the
substrate 140 by the image projection apparatus 200A and 200B,
shown in FIGS. 2A-2B and discussed below. In some examples, these
techniques may be used in combination.
[0038] FIG. 1B is a perspective view of a photolithography system
100B according to embodiments disclosed herein. The system 100B is
similar to the system 100A; however, the system 100B includes a
first stage 130A and a second stage 130B. Each of the two stages
130A and 130B is capable of independent operation such that the
system can scan a substrate 140A or 140B in a first direction and
step in a second, other direction. In some embodiments, when one of
the two stages 130A or 130B is scanning a substrate 140, the other
of the two stages 130A or 130B is unloading an exposed substrate
and loading the next substrate to be exposed. In one example, the
first stage 130A scans the substrate 140A, and the second stage
130B unloads an exposed substrate 140B and loads the next substrate
(not shown) to be exposed. Also shown in FIG. 1B are the first and
second encoders 126A and 126B, respectively co-located with stages
130A and 130B.
[0039] While FIGS. 1A-1B each depict an embodiment of a
photolithography system (100A and 100B, respectively), other
systems and configurations are also contemplated herein. For
example, photolithography systems including any suitable number of
stages (e.g., 3 or more stages 130) are also contemplated.
[0040] FIG. 2A is a perspective schematic view of an image
projection apparatus 200A according to one embodiment. The image
projection apparatus 200A may be useful for a photolithography
system, such as system 100A or 100B in FIGS. 1A and 1B. The image
projection apparatus 200A includes one or more spatial light
modulators 202, an alignment and inspection system 208 including a
focus sensor 204 and a camera 206, and a plurality of projection
optics 210. In some examples, the image projection apparatus 200A
can be referred to as a "channel." The components of the image
projection apparatus 200A may vary depending on a type of spatial
light modulator 202 being used. Spatial light modulators 202
include, but are not limited to, solid state emitters, digital
micromirror devices (DMDs), vertical-cavity surface-emitting lasers
(VCSELs), liquid crystal displays (LCDs), and other solid state
emitters of electromagnetic radiation. Solid state emitters emit
electromagnetic radiation and are microscopic, for example, less
than about 100 .mu.m at the largest dimension. In some examples,
light emitting diodes such as microLEDs may be employed. Various
examples of solid state emitter devices can be arranged in arrays
in an exposure source of an imaging system used for digital
photolithography. Each solid state emitter may be associated with
at least one pixel of the display. In some embodiments, a plurality
of sub-pixels may be associated with each pixel. The solid state
emitters discussed herein emit electromagnetic radiation and may be
used in light-emitting devices (LEDs), uLEDs, vertical-cavity
surface-emitting lasers (VCSEL), and other devices. Various
voltages may be used, depending upon the type of solid state
emitter used. In examples where a uLED is employed, a voltage
applied may be from 100 mV to 7 V.
[0041] In operation, the spatial light modulator 202 is used to
emit light and/or modulate one or more properties of the light,
such as an amplitude, a phase, or a polarization of the light,
which is projected through the image projection apparatus 200A and
to a substrate, such as the substrate 140 in FIG. 1A. The alignment
and inspection system 208 is used for alignment and inspection of
the components of the image projection apparatus 200A or 200B. In
one embodiment, the focus sensor 204 includes a plurality of lasers
which are directed through the lens of the camera 206 and then back
through the lens of the camera 206 and imaged onto sensors to
detect whether the image projection apparatus 200A is in focus. The
camera 206 is used to image the substrate, such as substrate 140,
including to ensure the alignment of the image projection apparatus
200A and photolithography system 100A or 100B is correct or within
an predetermined tolerance. The projection optics 210, such as one
or more lenses, is used to project the (modulated or unmodulated)
light onto the substrate, such as the substrate 140.
[0042] FIG. 2B is an image projection apparatus 200B according to
embodiments described herein. In the embodiment shown in FIG. 2B,
the image projection apparatus 200B, which can also be referred to
as a "channel," similarly to the image projection apparatus 200A,
includes a solid state emitter device 212 as the spatial light
modulator(s), a focus sensor 204 and inspection system 208 and
projection optics 210. In one embodiment, the image projection
apparatus 200B further includes a beamsplitter (not shown). The
autofocusing discussed herein employs the image projection
apparatuses of FIGS. 2A and 2B, utilizing the image data collected
to autofocus the projection optics 210 prior to patterning the
substrate to avoid the formation of mura.
[0043] FIG. 3 illustrates a schematic view 300 of a plurality of
image projection apparatuses 200B. It is to be understood that
while image projection apparatus 200B are shown, it is contemplated
that image projection apparatus 200A may be used either in place of
or in addition to the image projection apparatus 200B. The
plurality of image projection apparatuses 200B may be used in a
photolithography system, such as systems 100A and 100B. In
operation, each image projection apparatus 200B produces a
plurality of write beams 302 onto a surface 304 of the substrate
140. As the substrate 140 moves in the X-direction and/or the
Y-direction, as indicated by the coordinate system in FIG. 3, the
surface 304 (e.g., an entire surface across a substrate 140 from
edge to edge) is patterned by the write beams 302. In various
embodiments, the number of image projection apparatuses 200B
included in the system varies based on factors such as the size of
the substrate 140 and/or the speed of the one or more stages 130
(as shown in FIGS. 1A and 1B).
[0044] Turning to FIG. 4, an example 400 of a solid state emitter
device 212 is shown. As shown in FIG. 4, the solid state emitter
device 212 includes a plurality of solid state emitters 402 in an
array. Since solid state emitters are self-emitting, an outside
light source is not included in the image projection apparatus 200B
that illustrates the solid state emitter device 212. Additionally,
unlike DMD micromirrors which are configurable between an "on" and
"off" state, each solid state emitter has a variable intensity
which enables an enhanced grayscale controllability. The solid
state emitter device 212 may comprise 4 sides, which may be
referred to as "edges" herein. The "edge" is a physical boundary of
the substrate, and each solid state emitter device 212 may comprise
four edges, a first edge 408A, a second edge 408B, a third edge
408C, and a fourth edge 408D. In some examples, the first edge 408A
may be referred to as a "north" edge, the second edge 408B may be
referred to as a "west" edge, the third edge 408C may be referred
to as a "south" edge, and the fourth edge 408D may be referred to
as an "east" edge. Thus, references to the "north-south"
adjustability or directionality references a direction 410 from
408A to 408C (and 408C to 408A), and references to "east-west"
adjustability of directionality references a direction 412 from
408B to 408D (and 408D to 408B). The east-west adjustability
discussed herein refers to the tunability of a voltage that is
applied to the solid state emitter device 212 as well as the
tunability of a gate bias (which may be set to 0 or to greater than
0). In one example, as substrate moves from the fourth edge 408D to
the second edge 408B, and both the substrate and solid state
emitter device 212 are substantially the same shape if not the same
size, a plurality of "stripes" are formed when the voltage is
received by some of the emitters from the solid state emitter
device 212. In some examples, the substrate is larger than the
solid state emitter device 212 and a plurality of passes of the
solid state emitter device 212 are made over the substrate.
[0045] The solid state emitters 402 are arranged in an array
comprising a plurality of rows 404A-404F and a plurality of columns
406A-406G. While six rows and seven columns are shown in FIG. 4 as
an example, in other embodiments, a solid state emitter device 212
may include millions of solid state emitters 402, each of which
corresponds to a pixel in a display. For example, one embodiment of
a solid state emitter device has a size of 1024 by 2048 pixels.
Various pulses can be applied to the plurality of rows 404A-404F
and/or columns 406A-406G, or to individual solid state emitters
402, or to groups of solid state emitters 402 other than the rows
and columns discussed herein. The application of pulses to the
solid state emitter device 212 across various regions of a
substrate in either direction 410 and/or 412 thereby forms a
plurality of patterned features on the substrate of a desired
(target) feature size. In alternate embodiments, pulses can be
applied to individual solid state emitters 402, or to groups of
solid state emitters other than the rows and columns illustrated in
FIG. 4.
[0046] Autofocusing the imaging apparatus includes capturing images
using three focus (laser) beams, which can be part of imaging
projection apparatuses referred to herein as channels. Each laser
beam is applied to each pixel across a substrate in real-time. The
image pre-processing discussed herein takes the images received by
CMOS sensors of the imaging apparatus and removes a percentage of
noise to account for the metal reflection noise, without amplifying
the sensor noise. This enables the imaging system to capture
multiple images of a substrate in a pixel-wise application to later
sort an "n" number of pixels to eliminate a predetermined
percentage of noise from the images. Subsequently, this filtered
data is distilled to a single-point positional/height measurement
via an estimation, then the centroids are determined for each
sensor's images. This information is put into a filter such as a
Kalman filter to track the substrate height over time in real-time
and combine the centroid images to determine an exposure position,
which the stage and/or the exposure source is adjusted to prior to
exposure.
[0047] FIG. 5 is a flow chart of a method 500 of autofocusing on a
substrate during a substrate patterning operation. In the method
500, at operation 502, a substrate is disposed in an imaging
apparatus, such as the imaging apparatuses 102A in FIG. 1A or 102B
in FIG. 1B. At operation 502, the substrate can be positioned on a
stage in the imaging apparatus, such as the stage 130 in FIG. 1A or
the stages 130A or 130B in FIG. 1B. The imaging apparatus is
configured in an initial position at operation 502 with respect to
a position of the stage relative to the exposure source.
[0048] At operation 504, a plurality of laser beams is applied to
each pixel across the substrate. Each laser beam of the plurality
of laser beams is configured to contact each pixel at a different
angle. Subsequently, at operation 506, in response to the laser
application at operation 504, a plurality of sensors such as the
focus sensor 204 in FIG. 2 receive a plurality of images from each
pixel across the substrate. The images received at operation 506
can be distorted due to noise because the reflection of the laser
beams off of the metallic material can cause noise in the images
that may inhibit autofocusing. In an embodiment, at operation 506,
an initial (current) position of the exposure source is determined
(e.g., in microns) and an offset value is determined in pixels. The
images received at operation 506 are shifted by adding the offset
determined to the image pointer. This can be a manner of taring the
imaging system before determining the exposure position. Shifted
images can be stored in a fixed-size priority queue for each pixel,
such that for each pixel across a substrate, there is a buffer
comprising a height value of that pixel, so the first pixel is
frame N, the second pixel is frame N+1, and so on.
[0049] As such, at operation 508, the images are analyzed to
determine a percentage of noise to filter. The percentage of noise
to be filtered can depend on a plurality of factors, including an
intensity of the laser beams and whether the laser beams are
hitting photoresist material or metallic material under the
photoresist. In an embodiment, the percentage of noise to filter
from the received images is from 1% to 30%, or from 5% to 25%, or
from 10% to 20%, or other ranges as appropriate. The filtered noise
discussed herein is non-Gaussian noise, that is, noise that has a
probability density function is outside of a normal statistical
distribution (Gaussian distribution).
[0050] At operation 510, the images are filtered to remove the
percentage of noise determined at operation 508. Filtering the
noise through the pre-processing including operations 508 and 510
aids in autofocusing the imaging system because filtering the noise
removes a portion of the reflection of the metallic content that
can cause the exposure position measurements to be incorrect, thus
resulting in mura. The percentage of noise filtered at operation
510 can be from 0% to 30% or greater.
[0051] At operation 512, subsequent to filtering out the noise at
operation 510 and using the filtered images, a centroid of the
reflection at a best-focus position is taken to determine a target
centroid value to use for adjustment prior to exposure. That is,
one centroid is calculated for each sensor's images. The centroid
calculation at operation 512 comprises a background brightness
measurement and a brightness threshold, each of which are
subtracted from the image prior to calculating the centroid. When
the exposure source is autofocused, the laser beams (focus beams)
reflect on to specific locations on the CMOS sensor(s) associated
with each beam. The magnitude and direction of the focus deviation,
determined subsequently at operation 514, can be determined by
converting the difference between the target centroid value and an
actual centroid value to a z-axis deviation measured in microns,
using a calibrated ratio of a z-height change to a sensor pixel
shift.
[0052] At operation 514, based upon the determination of the
centroids at operation 512, a focus deviation is determined. The
focus deviation is determined because the position on the sensors
that receive the images represent a relative distance of the
exposure source and the substrate, e.g., based on a measurement of
each pixel of the substrate. The focus deviation is a measurement
of how the initial position of the exposure source relative to the
substrate differs from a second, exposure position for the exposure
source. The exposure position is the position the exposure source
will be in during the exposure discussed below at operation
522.
[0053] At operation 516, the centroids calculated at operation 514
are combined to generate a model estimation of an exposure
position. The combination of centroids is executed at operation 516
in order to combine the centroids as determined from each of the
images from each of the sensors. The combination of centroids is
performed at operation 516 using Kalman filtering, a
proportional-integral-derivative filtering, or linear filtering.
The Kalman filter combines multiple measurements from the sensors
taken in real time, over time, to provide an estimate of the
exposure position. Kalman filters can also be employed at operation
516 to weight measurement quality among and between each
sensor.
[0054] In one example, a Kalman filter is used at operation 516 to
combine the centroid calculations and determine an exposure
position for the exposure source. A Kalman filter is a Bayesian
statistical filter which combines a series of measurements over
time using a system model and a measurement model to generate a
single optimal estimate for the state of a system. In an
embodiment, the Kalman filter is used at operation 516 to combine
the measurements of the centroids from all three sensors, producing
a single estimate of the substrate height. The use of the Kalman
filter enables a model estimation that can be performed in
real-time and knowledge such as the substrate composition,
substrate size, and historical imaging data can be used in the
Kalman filter to generate the model at operation 516. By combining
measurements from all the three sensors, a model exposure position
for in-focus exposure is determined at operation 516.
[0055] The exposure position, as discussed above, is a position of
the exposure source relative to a position of the substrate, and
can be defined in one or more of the x, y, and z directions.
Depending upon the embodiment, and dependent in part upon an
initial position of the substrate and the exposure source (e.g.,
210 or 212 from FIGS. 2A and 2B), the exposure position and can be
obtained by moving one or both of the exposure source or the stage
(e.g., the stage 130 from FIG. 1). The Kalman filter takes "clean"
data, the image data that has been filtered to remove a
predetermined percentage of noise, from the images captured by the
plurality of sensors, this data is viewed as an independent
Gaussian distribution from each sensor and can be modeled. The
measurement data taken via the laser beams is Gaussian noise and
includes unknown measurement noises from various surface factors
including composition (alloy, polymer, etc.) and composition of the
substrate. The Kalman filtering at operation 516 produces a model
estimation used to determine the exposure position along with the
centroid calculation at operation 514.
[0056] At operation 518, based upon the combination of the
centroids at operation 516 and the focus deviation determined at
operation 514, an exposure position is determined. This
determination can comprise comparing an offset determined at
operation 504 and the focus deviation determined at operation 512
to the model generated using the Kalman filter at operation 516. In
alternate embodiments, operation 518 can comprise further analysis
to determine the exposure position. The pre-processing of the image
data, including the removal of the noise at operation 508, provides
for an autofocused exposure more easily and more reliably than
conventional methods without compromising cycle times for the
exposure operation. Further in the method 500, at operation 520,
either the stage on which the substrate is positioned is moved or
the exposure apparatus is moved (adjusted) to the exposure position
determined at operation 518.
[0057] Depending upon the embodiment, when an exposure position is
determined, the stage and/or an exposure source (e.g., solid state
emitters such as the projection optics of FIG. 2A or the solid
state emitter device 212 of FIG. 2B) can be moved in various
directions to autofocus the exposure source on the substrate. The
adjustment of, for example, the stage, to the exposure position
comprises one or more of moving the stage along the x-axis, the
y-axis, or the z-axis, and, in some embodiments, can comprise
adjusting one side or one corner of the stage such that the stage
is not parallel to the x-axis. At operation 522, subsequent to
establishing the exposure position within the imaging apparatus,
the substrate is exposed and patterned. In some examples,
patterning the substrate comprises one or more exposures at
operation 522, and some or all of operations 504-520 can be
executed to determine an exposure position of subsequent exposures
at operation 522. In an embodiment, operations 504-520 are executed
in real-time in a time period from 0.3 ms to 5 ms, or from 0.5 ms
to 3 ms, or from 0.8 ms to 1.2 ms, or other periods of time as
appropriate for a predetermined number of pixels. The predetermined
number of pixels can be from 2 to 100, 10 to 50, 5 to 10, or other
ranges or targets of pixel counts as appropriate for the size and
composition of the substrate. The one or more exposures at
operation 522 are performed in-focus, resulting in a reduced
likelihood and occurrence of mura.
[0058] FIG. 6A is an example graph of pixel amplitudes measured for
a plurality of pixels across a substrate. FIG. 6A shows the pixel
amplitude (height) vs. a number of pixels associated with that
height. FIG. 6A shows how many pixels are at a particular height,
as measured for a plurality of pixels across a substrate. As one or
more exposures can be performed to pattern a substrate using
digital photolithography, the exposures discussed herein can be for
an entire substrate or for portions of a substrate since the
analysis and exposure position adjustments discussed herein are
performed in real-time during substrate patterning. As noted above,
conventional technology can determine a peak of the graph in 6A.
However, the centroid of the substrate may not be the center of the
peak, for example, if the peak is irregularly shaped. Using the
autofocus systems and methods discussed herein performs a centroid
calculation to determine a mean, e.g., not necessarily a peak, of
the pixel amplitude curves is discussed.
[0059] FIG. 6B shows an example graph of pixel height data across a
substrate in an ideal case with a normal or near-normal Gaussian
distribution. FIG. 6B is an ideal case, so the center of the peak
is about the mean of the data, labeled as the centroid line 602 in
FIG. 6B. However, the image data obtained herein may not be ideal,
for example, because of noise and/or because of the variation in
height among and between pixels across a substrate that can be
caused by various operations during device fabrication.
[0060] FIG. 6C illustrates a centroid calculation on an example
graph that results in a centroid value offset from a center of the
peak. FIG. 6C illustrates the use of an averaging operation across
the full set of image capture data results across a substrate to
determine a centroid 604, which is located outside of the peak.
[0061] FIG. 6D illustrates an example graph of pixel height across
a substrate and a centroid calculation according to embodiments of
the present disclosure. FIG. 6D is the graph of FIG. 6C that shows
the centroid 604 calculated using conventional methods and a second
centroid 606 determined according to embodiments calculation for
the centroid is performed on a subset of data indicated by the
shaded box 608. FIG. 6D further illustrates removal of background
noise below a threshold 610. In order to obtain a more accurate
centroid value, using the filtered data discussed herein where a
predetermined percentage of background noise has been removed from
the data set, a subset of the image capture data is used for the
centroid calculation. The background noise shown in FIGS. 6C and 6D
and discussed with respect to FIGS. 6C and 6D can include noise
from the metallic material as discussed above, and can be
determined and subsequently filtered as discussed in the method 500
in FIG. 5. In alternate embodiments, the background noise indicated
by 610 can include metallic material reflection and/or sensor noise
or other background noise.
[0062] Embodiments and examples of the present disclosure can be
combined in various combinations. In an example method of
patterning a substrate, the method includes: applying a plurality
of laser beams across a substrate while the substrate is positioned
on a stage in an imaging apparatus in an initial position relative
to an exposure source of the imaging apparatus, the substrate
comprising a photoresist and a metallic material; receiving, by a
plurality of sensors coupled to the imaging apparatus, in response
to applying the plurality of laser beams, a plurality of images of
each pixel of the plurality of pixels; and determining a percentage
of noise in the plurality of images received from each sensor for
each pixel of the plurality of pixels in response to the applying
of the plurality of laser beams to the metallic material of the
substrate. In this example, the method further includes filtering
out the percentage of noise from the plurality of images from each
sensor for each pixel of the plurality of pixels; subsequently,
calculating a centroid of the substrate based on the plurality of
images; determining, based on the calculating of the centroid for
each sensor of the plurality of sensors, a focus deviation; and
adjusting, the substrate to an exposure position from the initial
position, the exposure position being different than the initial
position, and the adjusting the exposure source autofocusing the
exposure source on the substrate.
[0063] In some examples, which can be combined herein with other
examples, the method further includes, subsequent to adjusting the
substrate, exposing the substrate to the exposure source, wherein
exposing the substrate forms a plurality of features in the
substrate. Furthermore the method can include combining each
centroid for each sensor of the plurality of sensors, wherein the
combining comprises Kalman filtering,
proportional-integral-derivative filtering, or linear filtering;
and determining, based on the combining and the focus deviation,
the exposure position of the exposure source.
[0064] In an example of the method, the substrate comprises a
photoresist and a metallic material, and wherein the percentage of
noise is noise in the plurality of images received in response to
the plurality of laser beams reflecting off of the metallic
material during the applying of the plurality of laser beams. In
another example of the method, wherein percentage of noise is from
1% to 30%. In yet another example of the method, the method
includes determining the centroid using a subset of the plurality
of images.
[0065] In another example, a method of patterning a substrate
includes: positioning, in an imaging apparatus, a substrate
comprising a photoresist and a metallic material, the substrate
being positioned on a stage in the imaging apparatus in an initial
position relative to an exposure source of the imaging apparatus;
and determining, by an application stored on a non-transitory
memory of a server and executable by a processor, an exposure
position for the imaging apparatus. The determining the exposure
position includes applying a plurality of laser beams to each pixel
of the plurality of pixels; receiving, by a plurality of sensors
coupled to the imaging apparatus, a plurality of images from each
pixel of the plurality of pixels in response to applying the
plurality of laser beams; determining a percentage of noise of the
plurality of images for each sensor for each pixel of the plurality
of pixels; and filtering out the percentage of noise from the
plurality of images from each sensor for each pixel of the
plurality of pixels. The method can further include, subsequent to
the filtering, calculating a centroid of the substrate for each
sensor of the plurality of sensors; determining, based on the
calculating of the centroid for each sensor of the plurality of
sensors, a focus deviation; combining each centroid for each
sensor; and determining, based on the combining and the focus
deviation, the exposure position of the exposure source. The method
can further include adjusting, by the application, at least one of
the stage or the exposure source from the initial position to the
exposure position, the exposure position being different than the
initial position, and the adjusting the exposure source
autofocusing the exposure source on the substrate.
[0066] In some examples, this method further includes, subsequent
to adjusting the stage, exposing the substrate to the exposure
source, wherein exposing the substrate forms a plurality of
features in the substrate. In other examples, the combining
includes Kalman filtering, proportional-integral-derivative
filtering, or linear filtering. The substrate has a height across
the substrate surface, as measured perpendicular to the substrate,
is not uniform. The substrate includes a photoresist and a metallic
material, and wherein the percentage of noise is noise in the
plurality of images received in response to the plurality of laser
beams reflecting off of the metallic material during the applying
of the plurality of laser beams. The percentage of noise can be
from 1% to 30%. The method further includes adjusting, by the
application, a velocity of the substrate from a first velocity to a
second velocity based on the combining and the focus deviation.
[0067] In an example, a non-transitory computer-readable medium
includes instructions that are configured to cause a computing
system to: determine an exposure position for an imaging apparatus.
The determination of the exposure position includes causing the
computing system to: apply a plurality of laser beams to each pixel
of a plurality of pixels of a substrate; receive, by a plurality of
sensors coupled to the imaging apparatus, a plurality of images
from each pixel of the plurality of pixels in response to applying
the plurality of laser beams; determine a percentage of noise of
the plurality of images for each sensor for each pixel of the
plurality of pixels; filter out the percentage of noise from the
plurality of images from each sensor for each pixel of the
plurality of pixels; subsequently, calculate a centroid of the
substrate for each sensor of the plurality of sensors; determine,
based on the calculation of the centroid for each sensor of the
plurality of sensors, a focus deviation; combine each centroid for
each sensor; and determine, based on the combination and the focus
deviation, the exposure position of an exposure source. The
computing system can be further configured to adjust the substrate
to the exposure position from an initial position, the exposure
position being different than the initial position, and the
adjustment of the exposure source autofocuses the exposure source
on the substrate. The instructions are further configured to,
subsequent to adjustment of the substrate, expose the substrate to
the exposure source to form a plurality of features on the
substrate. In some examples, the instructions are further
configured to combine each centroid by Kalman filtering,
proportional-integral-derivative filtering, or linear filtering.
The substrate a height across the substrate surface, as measured
perpendicular to the substrate, is not uniform. The substrate
includes a photoresist and a metallic material, and wherein the
percentage of noise is noise in the plurality of images received in
response to the plurality of laser beams reflecting off of the
metallic material during the applying of the plurality of laser
beams. In some examples, the percentage of noise is from 1% to 30%,
and, in other examples, the percentage of noise is from 5% to
20%.
[0068] Using the systems and methods discussed herein, digital
lithography imaging systems are autofocused on maskless substrates
in real-time by performing a pixel-wise analysis of height
variation of pixels across a substrate that comprises a photoresist
as well as metallic material. The analysis is performed using three
laser beams and three sensors, each of which receives images from
each pixel across the substrate when the substrate is positioned on
a stage in an initial position relative to an exposure source of
the imaging system. A predetermined percentage of noise is filtered
out from the images collected, and the filtered images are analyzed
to determine a centroid. The centroids are used to determine an
offset from the initial position that will contribute towards
autofocusing of the imaging system during exposure. The filtered
data is analyzed to determine an exposure position, including
determining centroids for each sensor and combining the centroids
using, for example, a Kalman filter. When the imaging apparatus is
adjusted to the exposure position, the exposure source is
autofocused on the substrate as to reduce or eliminate the
occurrence of mura and other undesirable features resulting from an
unfocused exposure.
[0069] While the foregoing is directed to embodiments of the
present disclosure, other and further embodiments of the disclosure
may be devised without departing from the basic scope thereof, and
the scope thereof is determined by the claims that follow.
* * * * *