U.S. patent application number 16/069337 was filed with the patent office on 2019-01-17 for optical filter and optical device utilizing same.
This patent application is currently assigned to Samsung Electronics Co., Ltd.. The applicant listed for this patent is Samsung Electronics Co., Ltd.. Invention is credited to Gyu Weon HWANG, Doo Seok JEONG, In Ho KIM, Won Mok KIM, Kyeong Seok LEE, Wook Seong LEE.
Application Number | 20190016091 16/069337 |
Document ID | / |
Family ID | 61246201 |
Filed Date | 2019-01-17 |
View All Diagrams
United States Patent
Application |
20190016091 |
Kind Code |
A1 |
LEE; Kyeong Seok ; et
al. |
January 17, 2019 |
OPTICAL FILTER AND OPTICAL DEVICE UTILIZING SAME
Abstract
Provided is an optical filter including first and second
reflection layers separated from each other, a dielectric region
interposed between the first and second reflection layers and in
which two materials of which refractive indexes are different are
alternately disposed, and a buffer layer disposed between the
dielectric region and at least one of the first and second
reflection layers, wherein there are at least two filter regions in
which relative volume ratios of the two materials alternately
disposed are different.
Inventors: |
LEE; Kyeong Seok; (Seoul,
KR) ; HWANG; Gyu Weon; (Seoul, KR) ; KIM; Won
Mok; (Seoul, KR) ; KIM; In Ho; (Seoul, KR)
; LEE; Wook Seong; (Seoul, KR) ; JEONG; Doo
Seok; (Seoul, KR) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Samsung Electronics Co., Ltd. |
Suwon-si, Gyeonggi-do |
|
KR |
|
|
Assignee: |
Samsung Electronics Co.,
Ltd.
Suwon-si, Gyeonggi-do
KR
|
Family ID: |
61246201 |
Appl. No.: |
16/069337 |
Filed: |
July 31, 2017 |
PCT Filed: |
July 31, 2017 |
PCT NO: |
PCT/KR2017/008261 |
371 Date: |
July 11, 2018 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G02B 5/28 20130101; G02B
5/289 20130101; B32B 2307/418 20130101; G02B 1/11 20130101; G02B
5/201 20130101; B32B 2307/416 20130101; B32B 15/04 20130101 |
International
Class: |
B32B 15/04 20060101
B32B015/04; G02B 5/20 20060101 G02B005/20; G02B 5/28 20060101
G02B005/28; G02B 1/11 20060101 G02B001/11 |
Foreign Application Data
Date |
Code |
Application Number |
Aug 22, 2016 |
KR |
10-2016-0106416 |
Claims
1. An optical filter comprising: first and second reflection layers
separated from each other; a dielectric region interposed between
the first and second reflection layers and in which two materials
of which refractive indexes are different are alternately disposed;
and a buffer layer disposed between the dielectric region and at
least one of the first and second reflection layers, wherein there
are at least two filter regions in which relative volume ratios of
the two materials alternately disposed are different.
2. The optical filter of claim 1, wherein the two materials are
alternately disposed, and there are at least two filter regions in
which relative width ratios of the two materials are different.
3. The optical filter of claim 2, wherein a pair of the two
adjacent materials are allowed to respectively have same widths as
a pair of two adjacent other materials in one direction.
4. The optical filter of claim 1, wherein the two materials are
composed of a first material comprising in plurality and disposed
at a pre-determined interval, and a second material surrounding the
first material.
5. The optical filter of claim 2, wherein a width of a pair of the
two materials is smaller than a wavelength of a light passing
through the filter.
6. The optical filter of claim 1, wherein the filter regions, in
which relative volume ratios of the two materials are different
from each other, are in two or more directions in a plane.
7. The optical filter of claim 1, wherein the first and second
reflection layers are a metal layer or a dispersion Bragg reflector
(DBR).
8. The optical filter of claim 1, wherein the dielectric region
comprises three or more materials.
9. The optical filter of claim 1, wherein the buffer layer is
formed of the first or second material.
10. The optical filter of claim 1, wherein an upper part of the
second reflection layer comprises a wideband transmission filter
and/or anti-reflection coating.
11. The optical filter of claim 1, wherein the filter regions
respectively comprise different buffer layers.
12. An optical filter comprising: first and second reflection
layers separated from each other; a dielectric region interposed
between the first and second reflection layers and at least two
materials, of which refractive indexes are different from each
other, are alternately disposed; and at least two filter regions in
which relative volume ratios of the two materials alternately
disposed are different from each other, wherein an intermediate
reflection layer parallel to the reflection layers is added to a
central region of the dielectric region.
13. The optical filter of claim 12, further comprising: buffer
layers of which optical thicknesses are identical between the
intermediate reflection layer and the first and second reflection
layers.
14. An optical device comprising: filter regions in which relative
volume ratios of the two materials of claim 1 are different from
each other and through which different wavelengths are passed; and
photodetectors respectively corresponding to the filter
regions.
15. The optical device of claim 14, wherein the optical device is a
spectroscope, a CMOS image sensor, or a hyper-spectra imaging
device.
16. An optical device comprising: a transmissive substrate; and the
optical filter of claim 1 provided on an upper part of the
transmission substrate and integrated in a separate module type.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This U.S. non-provisional patent application claims priority
under 35 U.S.C. .sctn. 119 of Korean Patent Application No.
10-2016-0106416, filed on Aug. 22, 2016, the entire contents of
which are hereby incorporated by reference.
TECHNICAL FIELD
[0002] The present invention disclosed herein an optical filter,
and particularly to am optical filter for controlling a
transmission central wavelength with a relatively simple and small
structure and an optical device using the same.
BACKGROUND ART
[0003] The linear variable filter (LVF) kind of optical filter
having a Fabry-Perot resonator structure, and has a structure of
which the thickness of a cavity is linearly variable in the length
direction. In the LVF, a lower mirror layer and an upper mirror
layer are disposed with the dielectric cavity interposed
therebetween.
[0004] Such an LVF has a limitation in process reproducibility
because of a linear structure in which the thickness varies in the
length direction. Also, the resolution of spectrometer using a
typical LVF is determined by a height-to-length ratio of the LVF
and thus there is difficulty in minimizing a spectrometer element.
In particular, due to the linear structure, it is disadvantageous
in productivity, resulting from poor process compatibility with a
two-dimensional imaging sensor technology.
[0005] Since a per-position transmission spectrum of an LVF is
formed from an overlap of consecutive spectrums, and integration
between the LVF and a photodetector is not monolithic, the LVF is
spaced apart from a photodetector array, and due to a stray light
effect according thereto, a filter performance is lowered.
[0006] In addition, U.S. Pat. No. 5,726,805 discloses a planer
optical filter including a dielectric layer. According to the
patent, the optical filter includes a reflection layer and a
dielectric layer, and the dielectric layer has a period structure
formed therein and the period structure is provided with trenches
or grooves.
[0007] However, according to this scheme, substrates are attached
to each other, and thus a manufacturing process is difficult to be
understood and may be costly. A central wavelength may not also be
easily adjusted.
DISCLOSURE OF THE INVENTION TECHNICAL PROBLEM
[0008] The present invention provides an optical filter with a
miniaturized structure.
[0009] The present invention also provides an optical filter
structure capable of increasing productivity by enhancing process
reproducibility.
[0010] The present invention also provides am optical filter
structure for making monolithicity and integration with a
photodetector easy, preventing a stray light effect by minimizing a
distance between the filter and detector, and enhancing a
wavelength variable range and performance such as out-of-band
rejection performance.
Technical Solution
[0011] An embodiment of the present invention is to provide an
optical filter including: first and second reflection layers
separated from each other; a dielectric region interposed between
the first and second reflection layers and in which two materials
of which refractive indexes are different are alternately disposed;
and a buffer layer disposed between the dielectric region and at
least one of the first and second reflection layers, wherein there
are at least two filter regions in which relative volume ratios of
the two materials alternately disposed are different.
[0012] In an embodiment, the two materials may be alternately
disposed, and there may be at least two filter regions in which
relative width ratios of the two materials are different. A width
of a pair of the two materials may be smaller than a wavelength of
a light passing through the filter.
[0013] In an embodiment, a pair of the two adjacent materials may
be allowed to respectively have same widths as a pair of two
adjacent other materials in one direction, or the filter regions,
in which relative volume ratios of the two materials are different
from each other, may be in two or more directions in a plane.
[0014] In an embodiment, an intermediate reflection layer parallel
to the reflection layers may be further added to a central region
of the dielectric region. In this case, an optical filter has a
structure provided with two double resonance cavities of an upper
structure and a lower structure on the basis of the intermediate
reflection layer. In this case, each of the upper structure and
lower structure is possible or is not possible to include a buffer
layer. Namely, the buffer layer may be further included between the
dielectric region and at least one of the first reflection layer,
the second reflection layer, and the intermediate reflection
layer.
[0015] An embodiment of the present invention is to provide an
optical device including: first a second reflection layers
separated from each other; a dielectric region interposed between
the first and second reflection layers and at least two materials,
of which refractive indexes are different from each other, are
alternately disposed; and a buffer layer disposed between the
dielectric region and at least one of the first and second
reflection layers, wherein at least two filter regions in which
relative volume ratios of the two materials alternately disposed
are different from each other, and photodetectors are respectively
provided in correspondence to the filter regions. The optical
device may be a spectroscope, a CMOS image sensor, or a
hyper-spectra imaging device.
[0016] An embodiment of present invention is to provide an optical
device including: a transmissive substrate; and the optical filter
of the foregoing provided on an upper part of the transmission
substrate and integrated in a separate module type.
Advantageous Effects
[0017] An optical filter structure of the present invention may be
miniaturized by including a dielectric region for allowing at least
two regions where relative volume ratios of two materials are
different to be present.
[0018] In addition, the optical filter structure of the present
invention may increase productivity by improving process
reproducibility.
[0019] On the other hand, the optical filter structure of the
present invention allows monolithicity and integration with a
photodetector to be easy by including first and second reflection
plates parallel to each other and a dielectric region interposed
therebetween and including two materials, and minimizing a distance
between a filter and a detector array to prevent a stray light
effect and enhance performance.
BRIEF DESCRIPTION OF THE DRAWINGS
[0020] FIG. 1 is a longitude-sectional view of an optical filter
according to an embodiment of the present invention;
[0021] FIG. 2 is a plan view of the dielectric region of FIG.
1;
[0022] FIGS. 3 and 7 illustrate process charts illustrating a
method of manufacturing an optical filter according to an
embodiment of the present invention;
[0023] FIGS. 8 and 9 are plan views of a dielectric region of a
two-dimensional optical filter according to an embodiment of the
present invention;
[0024] FIG. 10 illustrates cross-sectional views of other optical
filters according to an embodiment of the present invention;
[0025] FIG. 11 illustrates cross-sectional views of other optical
filters according to an embodiment of the present invention;
[0026] FIGS. 12 to 19 illustrate simulation results of an optical
filter of one-dimensional structure, and FIGS. 20 and 22 illustrate
optical filer structures of which a planar structure is
two-dimensional;
[0027] FIG. 24 illustrates cross-sectional views of other optical
filters according to an embodiment of the present invention;
[0028] FIGS. 25 and 26 are simulation documents for comparing a
single resonance cavity structure with a double resonance cavity
structure of the present invention, and FIG. 27 are simulation
documents for comparing double resonance cavity structures
including a dispersion Bragg reflector (DBR);
[0029] FIG. 28 illustrates a cross-sectional view of other optical
filters according to an embodiment of the present invention;
[0030] FIGS. 29 and 30 are simulation documents for comparing a
structure in which a buffer is added to a double resonance cavity
structure of the present invention;
[0031] FIG. 31 illustrates a cross-sectional view of other optical
filters according to an embodiment of the present invention;
[0032] FIG. 32 illustrates a cross-sectional view of other optical
filters according to an embodiment of the present invention;
[0033] FIG. 33 is a cross-section view of another optical filter
according to an embodiment of the present invention;
[0034] FIG. 34 illustrates an example of an optical device
according to an embodiment of the present invention;
[0035] FIG. 35 is a conceptual diagram of a one-dimensional
spectroscope according to an embodiment of the present
invention;
[0036] FIG. 36 illustrated a calculation example of an L-curve
analysis method for determining an optimal a value in an L-curve
analysis method, FIG. 37 illustrates an L-curve (left) and a
curvature (right) of an L-curve function, and FIG. 38 is graphs
showing calculation results showing an Savitzky-Golay filter
application for reducing a system noise in a filter system and a
signal recovery performance according thereto.
[0037] FIG. 39 is a conceptual diagram illustrating an overlap
situation of a filter function of each filter in a spectroscope,
FIG. 40 shows a graph of simulation in which an overlap function is
calculated according to a full width at half maximum (FWHM) of a
filter function, and FIG. 41 is a graph of simulation in which an
error value (%) according to an overlap of the filter function is
calculated in a spectroscope according to an embodiment of the
present invention.
[0038] FIG. 42 illustrates an ORIGINAL signal and a RECOVERED
signal for each wavelength according to an FWHM in a spectroscope
according to an embodiment of the present invention;
[0039] FIG. 43 is a conceptual diagram of a two-dimensional
spectroscope according to an embodiment of the present
invention;
[0040] FIGS. 44 and 45 are conceptual diagrams of image sensors
according to an embodiment of the present invention; and
[0041] FIG. 46 is a conceptual diagram of a filter array including
a hyper spectral image sensor according to an embodiment of the
present invention.
MODE FOR CARRYING OUT THE INVENTION
[0042] Description will now be given in detail according to
exemplary embodiments disclosed herein, with reference to the
accompanying drawings. For the sake of brief description with
reference to the drawings, the same or equivalent components may be
provided with the same or similar reference numbers, and
description thereof will not be repeated. In general, a suffix such
as "module" and "unit" may be used to refer to elements or
components. Use of such a suffix herein is merely intended to
facilitate description of the specification, and the suffix itself
is not intended to give any special meaning or function. In the
present invention, that which is well-known to one of ordinary
skill in the relevant art has generally been omitted for the sake
of brevity. The accompanying drawings are used to help easily
understand various technical features and it should be understood
that the embodiments presented herein are not limited by the
accompanying drawings. As such, the present invention should be
construed to extend to any alterations, equivalents and substitutes
in addition to those which are particularly set out in the
accompanying drawings.
[0043] FIG. 1 is a longitude-sectional view of one dimensional
optical filter according to an embodiment of the present invention,
and FIG. 2 is a plan view of a dielectric region of FIG. 1.
[0044] Referring to FIGS. 1 and 2, a description will be provided
about a structure of an optical filter 100 according to an
embodiment of the present invention.
[0045] The optical filter 100 of the present invention includes
first and second reflection layers 110 and 120, a dielectric layer
130 and a buffer layer 140. The first and second reflection layers
110 and 120 may form both side surfaces of the optical filter 100.
For example, the first and second reflection layers 110 and 120 may
be disposed in one-dimensional type so as to be disposed parallel
to one direction, and possibly form a two-dimensional optical
filter 100.
[0046] The buffer layer 140 is disposed between the dielectric
layer 130 and at least one of the first and second reflection
layers 110 and 120. In other words, the buffer layer 140 is
disposed on at least one of an upper side and a lower side of the
dielectric region 130, and the dielectric region 130 and the buffer
layer 140 are interposed between the first and second reflection
layers 110 and 120.
[0047] The buffer layer 140 together with the dielectric region 130
operates as an optical resonance cavity. The presence of the buffer
layer 140 increases the effective thickness of the optical
resonance cavity, and thus a central wavelength of a transmission
band of the optical filter 100 is moved to a long wavelength region
while the thickness of the dielectric region layer 130 is
maintained small. When the optical resonance cavity is configured
only from the dielectric region layer 130, an aspect ratio is
excessively increased to compensate for any one material in the
dielectric region layer 130 that is composed of a combination of
two materials that have different refractive indexes and a
sufficiently smaller period than an operation wavelength, which
results difficulty on process. Accordingly, the presence of the
buffer layer 140 is advantageous in that the thickness of the
dielectric region layer 130 is maintained small and a wavelength
variable range is effectively increased.
[0048] The first and second reflection layers 110 and 120 may be
respectively a metal thin film having semi-transmission property
and a distributed Bragg reflector (DBR) formed of a periodic
multilayer structure of a high refractive index dielectric region
layer and a low refractive index dielectric region thin layer.
[0049] Here, the dielectric region 130 and the buffer layer 140
will be described.
[0050] The dielectric region 130 is disposed between the first and
second reflection layers 110 and 120, and at least two materials
134 and 137 with different refractive indexes are disposed. There
are at least two regions in which relative volume ratios of the two
materials 134 and 137 forming the dielectric region 130 are
different.
[0051] The two materials 134 and 137 forming the dielectric region
130 may be alternately disposed. On the other hand, the dielectric
region 130 may be formed along one direction such that there are at
least two regions in which relative width ratios of the two
materials 134 and 137 are different. Such a structure becomes
one-dimensional optical filter structure (see FIG. 2).
[0052] Referring to FIG. 1, a region in which a central wavelength
of a transmission light varies is illustrated as three regions of
a, b, and c. In this case, the two different materials 134 and 137
are alternately disposed in the three regions of a, b, and c, at an
identical period, and the relative width ratios of the two
materials 134 and 137 are different. Due to such a structure, the
three regions of a, b, and c have different light transmission
wavelengths between the first and second reflection layers 110 and
120.
[0053] In detail, one direction of FIG. 2 means a horizontal
direction to which the length direction of the dielectric region
130 is extended laterally. At least two materials 134 and 137 of
which refractive indexes in the dielectric region 130 are disposed
in the one direction.
[0054] On the other hand, the dielectric region 130 may be formed
such that there are at least two regions in which relative volume
ratios of the two materials 134 and 137 are different. For example,
in a certain region, two materials may be alternately disposed in a
pre-determined width, and in another certain region, the two
materials may be alternately disposed in a differently
pre-determined width. Even when the two materials have different
widths in different regions, widths of the two adjacent materials
in an identical filter that transmits an identical wavelength is
made constant. Accordingly, a to c regions are shown, and in each
region, relative volume ratios of the two materials 134 and 137 are
constant. As exemplarily shown in the drawing, the relative volume
ratios of the two materials 134 and 137 are different in adjacent
regions and widths of the adjacent to materials 134 and 137 are
constant in all three regions a to c.
[0055] The two materials 134 and 137 of the dielectric region 130
may be named as, for example, first and second materials 134 and
137. The first and second materials 134 and 137 may be dielectrics
having different refractive indexes. In addition, the first
material 134 may be a dielectric material with a relatively low
refractive index and the second material 137 may be a dielectric
material with a relatively high refractive index, but the present
invention is not limited thereto. In other words, the first
material 134 may be a material of a relatively high refractive
index, and the second material 137 may be a material of a
relatively low refractive index. The dielectric of the low
refractive index may be, for example, a fluorine ultraviolet ray
resin, a spin on glass, hydrogen silsesquioxane (HSQ), magnesium
fluoride (MgF.sub.2), calcium fluoride (CaF.sub.2), or silicon
oxide (SiO.sub.2). The dielectric of high refractive index may be,
for example, a metal-oxide such as titanium oxide (TiO.sub.2).
[0056] In other words, the dielectric region 130 of the present
invention may be a binary dielectric region 130 that is a kind of
an effective medium of a combination of two dielectric regions 130
having different refractive indexes, and a relative ratio of the
first and second materials 134 and 137 may become gradually
different in one direction.
[0057] Relative ratios between the first and second materials 134
and 137 of the present invention may be defined as a duty cycle or
a fill factor. The duty cycle of fill factor in the present
invention is a relative volume ratio that a component of the first
material 134 between the first and second materials 134 and 137
occupies on the basis of the first material 134. The refractive
index of the dielectric region 130 is made variable in one
direction by gradually changing a mutual duty cycle or fill factor
between the first and second materials 134 and 137 in the
dielectric region 130.
[0058] A wavelength of a light passing the optical filter 100 is
controlled by the optical thickness of the dielectric region. The
optical thickness may be defined as a value of the physical
thickness multiplied by a refractive index, and a determinant of a
transmission band wavelength of the filter of the present invention
is the optical thickness of the dielectric region. Therefore, a
central wavelength may also be controlled by a refractive index
change besides the physical thickness.
[0059] On the other hand, a width of a pair of the first and second
materials 134 and 137 has a relation with a wavelength of a filter
to be transmitted. For example, the width of the pair of the first
and second materials 134 and 137 may be made sufficiently small,
and in this case, a light does not discriminate the two materials
as individual materials and recognizes the same as one effective
medium defined by a specific effective dielectric constant. At this
point, an optical constant of the effective medium is determined by
a geometrical distribution of the two materials and a relative
volume fraction. For the dielectric region 130 of which an
imaginary part of the dielectric constant is close to 0, the
optical constant of the effective medium has an arbitrary value
between optical constants of the two components.
[0060] FIG. 1 illustrates an example in which the buffer layer 140
is disposed on an upper part of the dielectric region 130. However,
the present invention is not limited thereto, and the buffer layer
140 may be disposed in any one of the upper part and lower part. An
operation wavelength control ability of the filter 100 is enlarged
by the buffer layer 140. The buffer layer 140 may be, for example,
a dielectric spacer.
[0061] According to a preferred embodiment, the buffer layer 140
may be the first material 134 or the second material 137. For
example, when the buffer layer 140 is the second material 137, it
is advantageous in manufacturing process. For example, first, in a
situation where the first material is patterned, the second
material 137 is spread on the entire patterned region of the first
material 134 to fill a gap of the first material 134. For example,
when a resin is used as the second material 137, the gap of the
first material 134 is filled with the resin by spin coating, and
the resin may also be spread widely on the top surface of the
second material 137. In this case, a planarization work may be
separately performed.
[0062] FIGS. 4 and 7 illustrate process charts illustrating a
method of manufacturing an optical filter according to an
embodiment of the present invention.
[0063] Referring to FIG. 4, a first reflection layer 120 is formed
on a substrate 101. As described above, the first reflection layer
120 may be formed of a thin film of such as Ag, Au, Al, Cr, or Mo,
which has semi-transmission property, or a DBR formed of periodic
multilayer structure of a high refractive index dielectric region
layer and a low refractive index dielectric region layer. As the
substrate 101, a transmissive substrate may be adopted, and the
transmissive substrate may be a glass or a polymer. For example, a
transmissive film may be configured from a transparent or a
semi-transparent polymer having an appropriate adhesive power or
impact absorption. A specific example may be non-restrictively
Polystyrene (PS), Expandable Polystyrene (EPS), Polyvinyl Chloride
(PVC), Styrene Acrylonitrile Copolymer (SAN), Polyurethane (PU),
Polyamide (PA), Polycarbonate (PC), Modified Polycarbonate,
Polyvinyl butyral, Polyvinyl acetate, Acrylic Resin, Epoxy Resin
(ER), Silicone Resin, Unsaturated Polyester (UP), polyimide,
polyethylene naphtalate, or polyethylene terephtalate, etc., and
these may be used individually or by mixing two or more.
[0064] Referring to FIGS. 4 and 5, the first material 134 is
deposited on the top surface of the first reflection layer 120 and
then patterned. Such works may be performed by a photolithography,
e-beam lithography, or nanoimprint lithography process. When the
photolithograpy process is adopted, as FIG. 3B, the first material
134 is entirely deposited, a photosensitive photoresist is
deposited thereon, and the photosensitive photoresist is
selectively opened by a light. Then, the first material 134 is
patterned by an etching process. When the nanoimprint process is
adopted, after the first material 134 is deposited, a resin layer
is spread thereon and a pattern is imprinted using an imprint
device, and then an etching process is conducted to pattern the
first material 134. On the other hand, without directly conducting
the etching process for the first material, it is also possible
that a first material layer is deposited on a pre-patterned
photoresist or resin layer and then the first material is patterned
through a lift-off process.
[0065] Referring to FIG. 6, the second material 137 is formed on
the entire surface including a patterned first material 134 and the
first reflection layer 120. The second material may be formed by
spin coating or vapor deposition.
[0066] After forming the second material 137, in order to make the
formation of the second reflection layer 110 easy or enhance the
property of an optical filter, a process of planarizing the top
surface of the second material 137 may be further included.
[0067] Referring to FIG. 7, the optical filter is completed by
forming the second reflection layer 110 on the entire structure in
which the second material 137 is formed. Like the first reflection
layer 120, the second reflection layer 110 may also be formed of a
thin film of such as Ag, Au, Al, Cr, or Mo, which has
semi-transmission property, or a DBR formed of a periodic
multilayer structure including high refractive index dielectric
regions and low refractive index dielectric regions.
[0068] FIGS. 8 and 9 are plan views of a dielectric region of a
two-dimensional optical filter according to an embodiment of the
present invention.
[0069] For convenience of explanation, a difference with FIG. 1
will be mainly described. The optical filter structure of FIG. 1 is
configured to have at least two spots in which relative width
ratios of the two materials 134 and 137 are different along one
direction, in a state where the two materials 134 and 137 forming
the dielectric region 130 are alternately disposed. In comparison,
the two-dimensional optical filters of FIGS. 8 and 9 have a
structure in which two materials 234 and 237; 334 and 337 forming
the dielectric regions 230 and 330 vary in two-dimensional
directions. FIG. 8 shows that the first material 234 between the
two materials 234 and 237 has a circular shape and the second
material 237 has a shape that surrounds the circular shape. FIG. 9
shows that the first material 334 between the two materials 334 and
337 has a rectangular shape and the second material 337 has a shape
that surrounds the rectangular shape.
[0070] For the optical filters of FIGS. 8 and 9, a wavelength
control for forming a variable filter is achieved by forming at
least two spots in which relative volume ratios of the two
materials, which are alternately disposed, are different. FIGS. 8
and 9 exemplarily show that the first materials have respectively
the circular and rectangular shapes, but the shape may also be
circular, rectangular, hexagonal, or octagonal. In other words, if
provided only at least two spots in which the relative volume
ratios between the first and second materials 234 and 237 are
different, the shape is not particularly limited and various shapes
may be available. Additionally, various shapes such as a cross
shape and a polygonal shape is adoptable.
[0071] On the other hand, a planar disposal of the first material
is performed in a periodic lattice structure, and various lattice
structures may be available besides a hexagonal lattice shown in
FIG. 8 or a rectangular lattice shown in FIG. 9.
[0072] FIG. 10 illustrates a cross-sectional view of other optical
filters according to an embodiment of the present invention.
[0073] For convenience of explanation, a difference with FIG. 1
will be mainly described. FIG. 1 shows a case where the buffer
layer 140 formed on the top surface of the dielectric region 130 is
formed of the same material as the first or second material, but
FIG. 9 shows a case where the buffer layer 140 adopts a different
material from the first or second material. An operation wavelength
control ability of the filter 100 is enlarged by the buffer layer
140. The buffer layer 140 may be, for example, a dielectric
spacer.
[0074] On the other hand, FIG. 10 illustrates that the buffer layer
140 is formed of a completely different material from the first or
second material, but in actual implementation, a portion of the
first or second material remains as a buffer layer and a different
material from the first or second material may form the buffer
layer 140 thereon. According to the configuration of FIG. 10, there
is a large variety of selecting the buffer layer 140.
[0075] FIG. 11 illustrates cross-sectional views of other optical
filters according to an embodiment of the present invention.
[0076] For convenience of explanation, a difference with FIG. 10
will be mainly described. FIG. 6 shows a case where the buffer
layer 140 formed on the upper part of the dielectric region 130
exists, but FIG. 11 shows a case where the buffer layer 140 exists
on a lower part of the dielectric region 130 as well as the upper
part. The buffer layer existing on the upper part and the buffer
layer existing on the lower part may be formed of the same material
or different materials. In addition, each of the buffer layers
existing on the upper part and the lower part may be formed of the
same material as or a different material from the first or second
material. When the buffer layers 140 exist on the upper and lower
parts, the operation wavelength control ability is enlarged.
[0077] Hereinafter, graphs showing results of simulating an
operation of each filter will be described. FIGS. 12 to 19
illustrate simulation results of an optical filter of
one-dimensional structure, and FIGS. 12 and 13 illustrate optical
filter structures of which a planar structure is
two-dimensional.
[0078] FIG. 12 illustrates a structure of an optical filter on
which the present simulation is performed, and graphs of FIGS. 13
to 15 show transmission band simulation results calculated for a
linear lattice structure in order to show a filter operation of the
present invention.
[0079] The results are calculated under assumption that a lattice
period is 200 nm, and the thickness of the dielectric region layer
is 60 nm in FIG. 13, 100 nm in FIGS. 14, and 150 nm in FIG. 15. It
is also assumed that a refractive index of a low refractive index
material between the two materials is 1. 38 and a high refractive
index material is 2.7. As the upper and lower reflection layers, 20
nm thick Ag layers having a semi-transmission property are adopted.
It may be known that as the thickness of the dielectric region
layer increases, an entire wavelength band with a transmission band
formed therein moves toward a long wavelength region.
[0080] The calculation results show a variation aspect of the
transmission band in a state where the period is fixed to 200 nm
and the width of a low refractive index nanostructure is increased
from 50 nm to 150 nm at an interval of 20 nm. It may be known that
as a fill factor of the high refractive index nanostructure
increases, a central wavelength of the transmission band moves
toward a long wavelength region.
[0081] Since the calculation is conducted by using a metal mirror
layer having a semi-transmission property, a full width at half
maximum (FWHM) of the transmission band is wide and a
transmissivity shows also a certain limit or lower. But when a DBR
is alternatively adopted, a very narrow FWHM and high
transmissivity may be achieved identically to a typical linear
variable filter technology in which the thickness varies in a
length direction.
[0082] Next, FIGS. 16 and 17 are simulation results for a structure
in which a lower buffer layer is inserted to a basic structure of
FIG. 12. FIG. 16 shows transmission band curves, when calculation
is conducted in a state where a buffer layer having a refractive
index of 1.38 and thickness of 50 nm is inserted to a lower part of
the dielectric region layer in the structure of FIG. 12 of which
the thickness of the dielectric region layer is 60 nm. FIG. 17
shows a variation in transmission band curve, when only the
thickness of a lower buffer layer is increased to 100 nm under the
condition of FIG. 18. It may be known that as the buffer layer is
inserted, the transmission band entirely moves toward a long
wavelength region and a movement degree is proportional to the
thickness of the buffer layer.
[0083] FIG. 18 shows a simulation result for a structure in which
upper and lower buffer layers are inserted to a basic structure of
FIG. 12. FIG. 18 shows a transmission band spectrum, when buffers
having a refractive index of 1.38 and the thickness of 50 nm are
respectively applied to the structure of FIG. 12. It may be known
that the central wavelength of the transmission band moves further
to the long wavelength region by a symmetric insertion of the upper
and lower buffer layers in comparison to the case of FIG. 17 where
only the lower buffer layer is increased to 10 nm, and there is an
effect of improving a out-of-band rejection property.
[0084] FIG. 19 shows a simulation result for a structure in which
an upper buffer layer is inserted to a basic structure of FIG. 12.
FIG. 19 shows a transmission band spectrum, when a buffer having a
refractive index of 2.7 and the thickness of 50 nm is applied to
the upper part of the dielectric region in the structure of FIG.
12. Red-shift of the entire transmission band occurs even by the
insertion of the upper buffer layer.
[0085] FIGS. 20 and 22 illustrate optical device structures of
which a planar structure is two-dimensional. First, FIG. 20 shows a
calculation result for a filter structure in which a low refractive
index nanostructure forms a hexagonal lattice structure and a high
refractive index material encloses the surrounding. Here, FIG. 20
shows a transmission band variation aspect, when the diameter of
the low refractive index nanostructure, which is in a spherical
pillar type, is increased from 50 nm to 150 nm in a state where the
thickness of the dielectric region 130 is 100 nm and the period of
the hexagonal lattice structure is fixed to 200 nm. FIG. 21 shows a
spectrum when only the thickness of the dielectric region is
increased to 150 nm in the structure of FIG. 20. Similar to
one-dimensional structure, it may be known that as the volume
fraction of the high refractive index material increases among
materials forming the dielectric region layer, the transmission
band is red-shifted and a control range thereof is easily enlarged
by insertion of the buffer layer.
[0086] FIG. 22 shows a transmission band spectrum when a dielectric
layer of which a refractive index is 1.38 and the thickness of 50
nm is inserted as the lower buffer layer in the structure of FIG.
20. Also, FIG. 23 shows a calculation result when a dielectric
layer of which a refractive index is 2.7 and the thickness of 50 nm
is inserted as the upper buffer layer in the filter structure of
FIG. 20.
[0087] FIG. 24 illustrates a cross-sectional view of other optical
filters according to an embodiment of the present invention.
[0088] Referring to FIG. 24, an optical filter of FIG. 24 has a
structure in which a separate reflection layer is additionally
inserted between dielectric regions. Through this, the optical
filter of FIG. 12 includes a lower structure including reflection
layers and a dielectric region therebetween, and an upper structure
including reflection layers and a dielectric region. In other
words, the optical filter of FIG. 12 has a double resonance cavity
structure or an induced transmittance filter structure. Since each
layer is identical to the above-described configurations, a
description thereabout will be omitted so as to avoid repetition.
For example, each reflection layer is possibly formed of a DBR. The
optical filter of the present invention has a structure including
first and second reflection layers 710 and 620, dielectric regions
730 and 630, and a separate intermediate reflection layer 610
inserted thereto. a, b, and c denote filter regions.
[0089] FIGS. 25 and 26 are simulation documents for comparing a
single resonance cavity structure and a double resonance cavity
structure of the present invention. In the present simulation, the
upper and lower reflection layers respectively adopt 30 nm thick Ag
layers, and a dielectric cavity is formed to have the thickness of
100 nm. In the double resonance cavity structure, the upper and
lower dielectric regions respectively have the thicknesses of 100
nm, and the intermediate reflection layer is formed of a 70 nm
thick Ag layer. According to the simulation documents, by
introduction of a double cavity structure of a
metal-dielectric-metal layer, a line width of the transmission band
becomes smaller, band squareness is improved, and an out-of-band
rejection property is also improved.
[0090] FIG. 27 shows a simulation result of the induced
transmittance effect for a structure in which a metal layer is
inserted to a middle part of the dielectric region in a Fabry-Perot
filter having a DBR. For the induced transmittance effect of a
specific band, a dielectric-metal-dielectric layer of a central
part, from which upper and lower DBRs are excluded which are
composed of a high refractive index layer and a low refractive
index layer of a quarter-wave condition, is designed to have the
thickness by which optical impedance is matched in order to
minimize reflection in a corresponding band.
[0091] In FIG. 27, the intermediate reflection layer is formed of
80 nm thick Ag, and both side reflection layers are formed of DBR
layers. The dielectric region layer has an upper and lower
symmetric structure, and the thickness thereof is determined to be
100 nm. Thereby, the induced transmittance effect is revealed, a
very sharp and narrow transmission band may be formed, band
squareness is improved, and an excellent out-of-band rejection
property may be realized. Therefore, it may be known that the
central wavelength of the transmission band is consecutively
red-shifted according to a control of refractive indexes of the
upper and lower dielectric region layers adjacent to the
intermediate layer.
[0092] FIG. 28 illustrates a cross-sectional view of other optical
filters according to an embodiment of the present invention.
[0093] Referring to FIG. 28, an optical filter of FIG. 28 has a
structure in which a separate buffer layer is added to the optical
filter structure of FIG. 24. Buffer layers 840 and 940 may be added
between the reflection layer and the dielectric region. As
described above, the buffer layer is effective in moving the
wavelength variable range. At this point, it is effective to apply
the buffer layer in order to identically maintain an optical
thickness (i.e., refractive index multiplied by physical thickness)
in each of the upper and lower structures, as shown in FIG. 16.
When the optical thickness is identical, similar to FIG. 16, the
buffer layer may be identically disposed in an upper stage of the
upper and lower dielectric region layers, or on the contrary, in a
lower stage of the upper and lower dielectric region layers. In
addition, it is possible to dispose the buffer layer to be
mirror-symmetric with the intermediate reflection layer 810 at the
center. For example, the buffer layer may be possibly disposed
between the intermediate reflection layer 810 and the dielectric
region 930 in the upper structure and/or between the dielectric
region 830 and the reflection layer 820 in the lower structure. In
addition, the buffer layers may be disposed in an upper part and
lower part of the dielectric regions of the upper and lower
structures.
[0094] FIGS. 29 and 30 are simulation documents for comparing a
structure in which a buffer layer is added to a double resonance
cavity structure that adopts the upper and lower metal reflective
layers of the present invention. FIG. 29 shows a transmission band
spectrum when dielectric buffer layers having a refractive index of
1.5 and the thickness of 50 nm are respectively inserted to the
upper parts of the upper and lower dielectric regions in the
structure of FIG. 26. It may be known that due to interposition of
the buffer layer, the central wavelength of the transmission band
entirely moves to a long wavelength region. In addition, as the
refractive index of the dielectric layer increases from 1.5 to 2.0,
the central wavelength of the transmission band is also
consecutively red-shifted.
[0095] FIG. 30 shows a simulation result in which transmission
spectrum shapes are compared for a single resonance cavity
structure with the buffer layer inserted thereto and a double
resonance cavity structure. The insertion of the buffer layer has
an effect of moving the central wavelength of the transmission
band, and similarly to the case without the buffer layer, a FWHM is
largely narrowed and squareness is improved.
[0096] FIG. 31 illustrates a cross-sectional view of other optical
filters according to an embodiment of the present invention.
[0097] When mainly describing a difference with FIG. 1, in relation
to FIG. 1, the case where a material forming the dielectric region
130 is formed of two 134 and 137 has been described, but in FIG.
18, the material is formed by three or more. In FIG. 18, as one
case of them, the first material 134 is formed of two materials
134a and 134b vertically laminated. In actual manufacturing, two
materials 134a and 134b are consecutively deposited on the
reflection layer 120 and in a subsequent process, the two layers
are selectively etched, or are consecutively deposited on a
photoresist pore structure formed through prepatterning and then a
lift-off process is conducted. In this way, when three or more
dielectric regions are formed, it is effective in that more
flexible control for an operation wavelength is possible. A
wavelength of a light passing the optical filter 100 of the present
invention is controlled by the optical thickness of the dielectric
region layer formed of the first and second materials 134 and 137.
At this point, an effective refractive index responsible for a
variable control of the optical thickness of the dielectric region
layer is determined by geometrical distributions and relative
volume fractions between the first and second materials 134 and
137. Here, when the first material is formed of materials of
different refractive indexes, a control range of an effective
refractive index may be enlarged and consequently, the control of
the optical thickness of the dielectric region layer becomes
further convenient.
[0098] On the other hand, FIG. 31 illustrates a case where the
first material 134 is configured from two vertically laminated
materials 134a and 134b, but it is also possible that the first
material 134 is vertically laminated with at least two layers or is
horizontally composed of two or more materials. But in a case where
multiple layers are formed, a pattern may be finer at the time of
forming a nanopattern. Therefore, it may be effective that such a
limitation is minimized.
[0099] In addition, for the second material 137, it is possible to
vertically laminate two or more layers, or to be horizontally
formed of two or more materials.
[0100] FIG. 32 illustrates a cross-sectional view of other optical
filters according to an embodiment of the present invention.
[0101] The optical filter of the present invention may include an
anti-reflection coating layer 153 and/or a broadband transmission
band filter 152. In addition, although not illustrated in FIG. 19,
it is also possible to additionally dispose an adhesion layer or a
diffusion barrier layer between the first and second reflection
layers 110 and 120, and the first and second materials 134 and
137.
[0102] The anti-reflection coating layer 153 is a component
adoptable for reducing an amount of a light that is incident to the
optical filter from the outside and then is reflected to disappear
to the outside. The broadband transmission band filter 152 is an
effective component capable of adjusting a necessary wavelength
band of a light incident to the optical filter.
[0103] On the other hand, the anti-reflection coating layer may be
formed on a surface other than a surface on which the optical
filter is formed on a separate transparent substrate on which the
optical filter of the present invention is formed.
[0104] FIG. 33 is a cross-section view of another optical filter
according to an embodiment of the present invention.
[0105] FIG. 33 illustrates an example that three optical filter
regions 100e, 100f, and 100g are combined, wherein the optical
filter region 100e has a type that the buffer layer 140 is disposed
in a lower part of the dielectric region 130, the optical filter
region 100f has a type that the buffer layers 140 are respectively
disposed in an upper and lower parts of the dielectric region 130,
and the optical filter region 100g has a type that like the optical
filter region 100f, the buffer layers 140 are respectively disposed
in the upper and lower parts, and a part of the buffer layers 140
may be formed to have difference in thickness. Here, it also
possible to have only difference in thickness of the upper buffer
layer without the lower buffer layer 140. In FIG. 20, a, b, and c
regions, in which different wavelength regions of the optical
filter region 100e are filtered, are illustrated to have the buffer
layers 140 in a lower part of the dielectric region 130, but the a,
b and c regions also have different buffer layers like the optical
filter regions 110e, 100f, and 100g.
[0106] According to configuration of FIG. 33, it is possible to
overcome a limitation in variable wavelength band coverable by an
individual optical filter region, enlarge an entire operation
wavelength range of the present optical filter, and make wavelength
adjustment easy.
[0107] FIG. 34 illustrates an example of an optical device
according to an embodiment of the present invention.
[0108] The present optical device is provided with filter regions
a, b and c, and photo-detectors PD1, PD2, and PD3 respectively
corresponding thereto. Although FIG. 34 illustrates a type that a
structure such as a transmissive substrate is not inserted between
the filter regions a, b, and c, and the photodetectors PD1, Pd2,
and PD3, the structure such as the transmission substrate may be
formed. Each of the photodetectors performs a function of detecting
a light corresponding to a wavelength band from each light
transmission region, and performs a well-known function such as
conversion of an amount of a light detected through various
electric circuit units and electrodes, etc., to an electrical
signal.
[0109] Hereinafter, another optical device according to an
embodiment of the present invention will be described. The optical
filter of the present invention may be formed on the top surface of
the transmissive substrate to be manufactured as a separate optical
filter module (see FIG. 7). Any material only to be transparent at
an operation wavelength may be used as the transmissive substrate,
and a glass or a polymer may be used. Such an optical filter module
may be manufactured in a type to which an optical detector is not
integrated. Therefore, in actual use, the optical detector may be
attached or when manufacturing a complete product, an optical
filter module may be attached to the optical detector. In the case
where the optical filter module is actually attached to the optical
filter and used, it is also possible, for example, to couple the
optical filter module to the front of a camera lens.
[0110] FIG. 35 is a conceptual diagram of a one-dimensional
spectroscope according to an embodiment of the present
invention.
[0111] FIG. 35 illustrates an example that a spectroscopic device
10000a is configured through one-dimensional linear array coupling
between a filter array 1000a and a photodetector 2000a. In a
schematic diagram, the spectroscopic device 10000a is illustrated
which includes the filter array 1000a composed of M filter regions
100 and M photodetecting units 500. A period of each optical filter
may be determined to be matched with a period of the
photo-detecting unit 500 of the coupled one-dimensional linear
array photodetector, or matched to the size of a group of a
plurality of photo-detecting unit 500. In other words, coupling
between the filter and pixels of the photodetector may be
one-to-one or one-to-multiple.
[0112] An array photodetector coupled to the filter array operates
as a spectroscopic device through a mathematical digital signal
processing algorithm. When assuming an ideal filter having a delta
function property, the resolution simply becomes equal to a value
obtained by dividing an operation wavelength region of the
spectroscopic device by the number of filters. Accordingly, there
is a limitation in that the number of filters required for an high
resolution operation increases proportionally thereto.
[0113] The optical filter according to the present invention is a
non-ideal filter defined by a Lorentzian function, and has a
property that an FWHM of a transmission band is determined
according to a design of a lower reflection layer. A case of using
a metal reflection layer is more typical than a case of using a
DBR, and as reflectance of the metal reflection layer is lower or
as the period number of unit combination of a low refractive index
layer and a high refractive index layer that configure the DBR is
smaller, the FWHM increases.
[0114] A signal recovery principle in a spectrometer based on the
non-identical filter array may be explained using FIG. 35. When a
spectrum of an object to be analyzed is s(.lamda.), a transmission
function of an individual filter is f.sub.i(.lamda.), and a
sensitivity function of a photodetector is d.sub.i(.lamda.), a
detection signal r.sub.i generated when an object spectrum passes
the filter and reaches the photodetector is represented as the
following Equation (1), and is deployable in a matrix equation of
Equation (2), when represented by a discrete model.
r i = .intg. .lamda. f i ( .lamda. i ) d i ( .lamda. i ) s i (
.lamda. i ) d .lamda. i = .intg. .lamda. D i ( .lamda. i ) s i (
.lamda. i ) d .lamda. i ( 1 ) [ r 1 M r i M r M ] = [ D 1 ( .lamda.
1 ) .LAMBDA. D 1 ( .lamda. N ) M M M D i ( .lamda. 1 ) .LAMBDA. D i
( .lamda. N ) M M M D M ( .lamda. 1 ) .LAMBDA. D Mi ( .lamda. N ) ]
[ s ( .lamda. 1 ) s ( .lamda. 2 ) M M s ( .lamda. N ) ] + [ n 1 n 2
M M n M ] ( 2 ) ##EQU00001##
[0115] Since the number M of filters is typically smaller than the
number N of wavelength samplings, the linear algebra equation of
Equation (2) comes down to an ill-posed problem. Since an explicit
inverse matrix of D(.lamda.) having the size of M.times.N (M<N)
does not exist, a spectrum signal may be recovered using a pseudo
inverse matrix, but is vulnerable to a small fluctuation or system
noise and shows an unstable result.
[0116] As a measure for obtaining a more effective and numerically
stable solution, a regularization scheme is being used. The most
representative scheme may be a Tikhonov regularization scheme. This
scheme recovers a spectrum of an analysis target object by
determining a solution to minimize a sum of a residual norm and a
side constraint norm as Equation (3). Here, .alpha. is a
regularization factor for determining a weight for minimization of
the residual norm in contrast to minimization of the side
constraint norm, and there exists an optimal value of .alpha. so as
to obtain a robust solution. When using singular value
decomposition (SVD) and L-curve analysis, the method adapts to a
system and determines an optical regularization factor for itself
to enable spectrum recovery in real time.
s.sub..alpha.=arg
min|Ds-r.parallel..sub.2.sup.2+.alpha..sup.2.parallel.L(s-s*).parallel..s-
ub.2.sup.2 (3)
[0117] When using such a regulation scheme, it is advantageous in
that a spectrum may be recovered with a relatively high resolution,
while a non-identical filter array having a wide FWHM is used. A
signal recovery algorithm is not limited to the exemplified
regularization scheme, but various schemes may be available.
[0118] On the other hand, the L-curve analysis is a method in which
a solution of a Tikhonov regularization equation is obtained when a
value is gradually increased and substituted, the obtained solution
is substituted again to the residual norm
.parallel.Ds-r.parallel..sub.2.sup.2 and solution norm
.parallel.L(s-s*).parallel..sub.2.sup.2, and resultant values are
represented on log-scaled coordinate axes. Then an L-curve shaped
graph is obtained and a corner value of the L-curve is adopted as
an optimal value .alpha.. A scheme for obtaining the corner value
is to take log-scaled values of the residual norm and solution norm
as variables and to determine .alpha. having the smallest radius of
curvature. The value obtained in this way is substituted again to
the Tikhonov regularization to obtain S.sub.a and recover the
object's spectrum. FIG. 36 illustrates an example of calculating
the L-curve analysis performed to determine an optimal .alpha.
using the L-curve analysis, and FIG. 23b illustrates a L-curve
(left) and a curvature of the L-curve function (right). A corner
value is determined as .alpha. value (here, .lamda.) having the
maximum curvature.
[0119] Next, for a presence of a system noise, there is a case
where a digital signal recovery process by regularization does not
properly operate and an unstable solution is output. In order to
reduce such a problem, a per-unit cell strength distribution curve
of a photodetector, in which a filter array is integrated, is
applied to a Savitzky-Golay smoothing algorithm that is effective
in noise filtering, and then an influence to signal recovery may be
evaluated. A Savitzky-Golay filter is one of smoothing schemes for
making waveforms of a data sequence including noises to smooth
waveforms from which the noises are excluded while original signal
disposition is not largely damaged, and is a filter for obtaining
k-th order polynomials for fitting surrounding points best at an
individual point by a least square method, and determining a data
value at that point. The Savitzky-Golay filter relatively well
conserves a maximum, minimum, or peak/valley value by applying a
moving average in a scheme that data near a data disposition is
more weighted and distant data is less weighted. There occurs a
situation where when a noise is mixed in a per-unit cell strength
distribution curve that is measured from a photodetector array,
even when a Tikhonov regularization scheme is applied, an original
spectrum may not be properly recovered. But when the Savitzky-Golay
filter is applied, it has been checked that signal recovery
performance by digital-signal-processing is excessively improved.
FIG. 23c shows graphs of an application of the Savitzky-Golay
filter for reducing a system noise in a filter system, and a
calculation result showing improvement in signal recovery
performance according thereto.
[0120] On the other hand, the inventors of the present invention
find a fact that for spectrum recovery, an error ratio may be
reduced when a transmission spectrum of each filter, namely, filter
functions form a proper overlap. A description thereabout will be
provided.
[0121] FIG. 37 is a conceptual diagram illustrating an overlap
situation of a filter function of each filter in spectroscope
according to an embodiment of the present invention, FIG. 40 is a
graph showing that in a spectroscope according to an embodiment of
the present invention, under assumption of a Gaussian function as a
filter function, an overlap factor according to an FWHM thereof is
calculated, and FIG. 41 is a simulation graph showing that in a
spectroscope according to an embodiment of the present invention, a
signal recovery error value (%) according to an overlap of filter
functions is calculated.
[0122] For quantitative finding, the overlap factor is defined as a
value obtained by dividing a transmissivity or reflectance value at
a point, at which spectrums of two adjacent filter functions cross
each other, by a maximum transmissivity or reflectance value of the
filter functions, and a variation in signal recovery error value
according thereto is evaluated. It may be known that when an
overlap degree of the two adjacent filters is lowered to a certain
value or lower, the signal recovery error value largely
increases.
[0123] FIG. 42 illustrates an object spectrum according to an FWHM
of a filter function and a spectrum recovered through digital
signal processing, when it is assumed that a Gaussian filter is
used in a spectroscope according to an embodiment of the present
invention. In this case, the graph shows an exemplary result that a
degree of spectrum recovery is improved, when the overlap factor is
increased while the FWHM of the filter is changed from 20 nm to 30
nm in a state where a distance between filter central wavelengths
is fixed to 30 nm.
[0124] For parameters for simulations of FIGS. 40 and 41, the
number of filters is 20, an interval between the filters is 30 nm,
the FWHM of original spectrum is 100 nm, a Peak-to-Peak value is
150 nm. In addition, the overlap value is changed while increasing
the FWHM (1 nm to 100 nm, pitch of 1 nm) of the filter function.
The overlap value is defined as Ip/Imax. Imax denotes a maximum
intensity of the filter function, and Ip defines the strength of
the filter function at a point at which two filter functions cross
each other (see FIG. 39). The error value ERROR is defined as the
following Equation, where norm( ) denotes a function related to a
magnitude of a vector in linear algebra.
ERROR=100*(norm(recovered signal-Original Signal)/norm(Original
signal)
[0125] Referring to FIGS. 40 and 41, when the error value is 10%,
it may be checked that the overlap value is about 0.4. At this
point, the FWHM is about 25 nm. Referring to FIG. 25, it may be
checked that the error value is sharply changed around .+-.5 nm
from a point at which the FWHM is 25 nm.
[0126] On the other hand, a description will be provided about a
proper overlap range in realization of a spectrum of the present
spectroscope. As checked in FIGS. 40 and 41, the smaller the error
value, the better. According to FIG. 41, as the overlap increases
to reach about 1.0, the error value approaches 0. However,
excessively many filter functions may be required to cover a
certain wavelength band, since the overlap is excessively large but
an interval between filter functions becomes smaller.
[0127] Therefore, a preferred lower limit value of the overlap will
be discussed. A range of the error value may be preferably smaller
than about 30%, and more preferably, smaller than 10%. Accordingly,
when it is converted based on the overlap, a preferable overlap is
0.2 or greater and a more preferable overlap is 0.4 or greater.
Next, a preferred upper limit value will be discussed. The overlap
is preferred to be as high as possible, when a spectrum of adjacent
filters exceeds a noise signal level, is not overlapped and is
distinguishable. At this point, the number of filters may
excessively increase, a structural factor difference between
adjacent filters is minute, and thus there may be a process limit.
Accordingly, it is preferred that the structural factor difference
between adjacent filters is not set to 1 nm or smaller, and there
may be an overlap upper limit according thereto.
[0128] On the other hand, when filter functions of filters form
proper overlaps, an error ratio may be reduced. This aspect may be
variously applied without being limited to shapes or kinds of the
filters. For example, it may be applied to FIG. 1 or other
embodiments of the present invention, and is also applicable to a
typically known plasmonic filter, etc. In other words, it is
related to a technology for reducing an error ratio and securing
the proper number of filters, when overlaps of filters are in a
certain range and when filters of each wavelength band are
introduced and the filter are integrated to analyze a wider band
spectrum.
[0129] FIG. 43 is a conceptual diagram of a two-dimensional
spectroscope of the present invention. An example of a spectroscope
10000b is illustrated which is configured through a combination in
which a filter array 1000b and a photodetector 2000b are arranged
in two-dimension. In comparison to the spectroscope 10000a of
one-dimensional combination, it is advantageous in integration, and
advantageous in combination with an existing CMOS image sensor,
etc. The spectroscopes 10000a and 10000b enable a light to pass and
intensity information is enabled to be output from a light in a
specific wavelength band of which a central wavelength is slightly
moved for each position in one direction of a filter.
[0130] Accordingly, the spectroscopes operate as a spectrometer
that enables conversion to a strength distribution according to a
light wavelength, and a spectroscope based on a filter array may be
realized.
[0131] FIGS. 44 and 45 are conceptual diagrams of image sensors
according to an embodiment of the present invention. FIGS. 44 and
45 respectively correspond to CMOS image sensor manufacturing
structures in an FIS scheme and BIS scheme. In these image sensor
structures, color filters such as R, G, B are essentially added.
The optical filter of the present invention may be applied to R, G,
B of the CMOS image sensors.
[0132] Referring to FIG. 44, a separation region 405 is interposed
between photodetection regions 406, and various electrode lines 403
are formed inside an intermediate dielectric region 402. And
thereon, R, G, B filter regions are formed, and thereon,
micro-lenses 401 are respectively formed in correspondence to the
R, G, B filter regions.
[0133] Referring to FIG. 45, various electrode lines 508 are formed
inside the dielectric region 507. Thereon, the separation regions
405 exist between the photodetection regions 406. And on the photo
detection regions 406, R, G, B filter region is formed and thereon,
the micro-lenses 501 are respectively formed in correspondence to
the R, G, B filter regions.
[0134] FIG. 46 is a conceptual diagram of a filter array including
a hyper-spectral image sensor according to an embodiment of the
present invention. FIG. 46 is a drawing for extracting a part of a
unit pixel of the CMOS image sensor and a unit pixel of the hyper
spectral image sensor.
[0135] The hyper spectral image sensor is an element configured to
sense several (relatively narrow) wavelength parts or a wavelength
band of an entire hyper spectrum emitted from or absorbed by an
object.
[0136] As illustrated in FIG. 46, according to one embodiment, the
hyper spectral image sensor may be used in a type of being coupled
to the CMOS sensor. In detail, a region referred to R, G, and B
denotes that of color filters configured to filter R, G, B of the
CMOS image sensor, and a region referred to H denotes a filter
region configured to sense a hyper spectral image. In filter region
H, a partially narrow region in the infrared ray region is
designated, and in this region, it is possible to secure hyper
spectral data different from R, G, B.
[0137] An optical filter described above are not limited to the
configuration and the method in the embodiment described above, and
the embodiment may have a configuration in which all or a part of
each embodiment is selectively combined such that various
modifications can be made.
[0138] The present invention may be carried out in other specific
ways than those set forth herein without departing from the spirit
and essential characteristics of the present invention. Therefore,
the above embodiments should be construed in all aspects as
illustrative and not restrictive. The scope of the invention should
be determined by the appended claims and their legal equivalents,
and all changes coming within the meaning and equivalency range of
the appended claims are intended to be embraced therein.
* * * * *