Digital Camera Patent Abstract
The present invention provides a solid-state image sensor wherein
the color shading is decreased, and/or a solid-state image sensor
wherein the shading effect is outstanding. Moreover, the invention
provides a digital camera having the solid-state image sensor. The
solid-state image sensor of the present invention has color filters
and/or apertures of a light blocking layer, and the center of the
color filters and/or the center of the apertures of the light blocking
layer is offset with respect to the center of a light-receiving
part, in the direction to the center of a valid cell area. In each
photodetecting cell of a preferred solid-state image sensor, micro-lenses
are further placed on the light-receiving side of the solid-state
image sensor, and preferably, the center of the micro-lens is similarly
offset with respect to the center of the light-receiving part. Also,
the digital camera is mounted with an above-described solid-state
image sensor. Digital Camera Patent Claims
What is claimed is:
1. A solid-state image sensor comprising a valid cell area in which
a plurality of valid cells are placed, said valid cells each having
a light-receiving part and a color filter placed in an on-chip fashion
to correspond to said light-receiving part, and each outputting
a charge signal, wherein: center of said color filter placed in
the peripheral part of said valid cell area is offset in the direction
to the center of said valid cell area with respect to the center
of said light-receiving part; and the offset amount between each
center of each said color filter and each center of said each light-receiving
part is larger at the peripheral part of the valid cell area, and
smaller at the central part of the valid cell area.
2. The solid-state image sensor according to claim 1, wherein:
said valid cells are grouped into a plurality of concentric blocks;
and said offset amount between each center of said each color filter
and each center of said each light-receiving part is larger at the
peripheral part of said valid cell area, and smaller at the central
part of said valid cell area, in which photodetecting cells within
a same block have the same offset amount.
3. The solid-state image sensor according to claim 1, wherein:
said valid cell has a light blocking layer having an aperture corresponding
to said light-receiving part; center of said aperture of said light
blocking layer in said valid cell reserved in the peripheral part
of said valid cell area is offset in the direction to the center
of said valid cell area with respect to the center of said light-receiving
part; and said offset amount between the center of said aperture
and the center of said light-receiving part is larger at the peripheral
part of said valid cell area, and smaller at the central part of
said valid cell area.
4. The solid-state image sensor according to claim 1, wherein said
valid cell has a micro-lens placed in an on-chip fashion to correspond
to said light-receiving part.
5. solid-state image sensor according to claim 4, wherein; said
valid cells are grouped into a plurality of concentric blocks; and
said offset amount between a center of said aperture and a center
of said light-receiving part of said valid cells is larger at the
peripheral part of said valid cell area, and smaller at the central
part of said valid cell area, in which said offset amount is the
same within a same block.
6. A solid-state image sensor comprising: a valid cell area in
which a plurality of valid cells are placed, said valid cells each
having a light-receiving part and a color filter placed in an on-chip
fashion to correspond to said light-receiving part, and each outputting
a charge signal, wherein: center of said color filter placed in
the peripheral part of said valid cell area is offset in the direction
to the center of the valid cell area with respect to the center
of said light-receiving part; and said light-receiving parts and
said color filters are respectively placed in predetermined pitches,
the pitch said light-receiving parts are placed in being greater
than the pitch said color filters are placed in.
7. The solid-state image sensor according to claim 6, wherein:
said valid cell has a light blocking layer having an aperture corresponding
to said light-receiving part; center of said aperture of said light
blocking layer in said valid cell reserved at the peripheral part
of said valid cell area is offset in the direction to the center
of said valid cell area with respect to the center of said light-receiving
part; and each said aperture is reserved at a predetermined pitch
from another said aperture, the pitch being smaller than the pitch
said light-receiving parts are placed in and greater than the pitch
said color filters are placed in.
8. The solid-state image sensor according to claim 6, wherein:
said valid cells have micro-lenses placed in an on-chip fashion
in a predetermined pitch, to correspond with said each light-receiving
part; and the pitch said color filters are placed in is greater
than the pitch said micro-lenses are placed in.
9. The solid-state image sensor according to claim 8, wherein:
said valid cells each include said color filter and said light-receiving
part, in which the center of said color filter is offset with respect
to the center of said light-receiving part; and the configuration
thereof satisfies an expression of
10. A solid-state image sensor comprising a valid cell area in
which a plurality of valid cells are placed, said valid cells each
having a light-receiving part and a blocking layer having an aperture
reserved to correspond to said light-receiving part, and each outputting
a charge signal, wherein: center of said aperture of said light
blocking layer in said valid cell reserved at the peripheral part
of said valid cell area is offset to the direction of the center
of said valid cell area with respect to the center of said light-receiving
part; the offset amount between the center of said aperture and
the center of said light-receiving part is larger at the peripheral
part of said valid cell area, and smaller at the central part of
said valid cell area; and the configuration thereof satisfies the
expressions
and
11. The solid-state image sensor comprising a valid cell area in
which a plurality of valid cells are placed, said valid cells each
having a light-receiving part and a blocking layer having an aperture
reserved to correspond to said light-receiving part, and each outputting
a charge signal, wherein: center of said aperture of said light
blocking layer in said valid cell reserved at the peripheral part
of said valid cell area is offset to the direction of the center
of said valid cell area with respect to the center of said light-receiving
part; the offset amount between the center of said aperture and
the center of said light-receiving part is larger at the peripheral
part of said valid cell area, and smaller at the central part said
valid cell area; said valid cells are grouped into a plurality of
concentric blocks; and said offset amount between the center of
said aperture and the center of said light-receiving part of said
valid cells is larger at the peripheral part of said valid cell
area, and is smaller at the central part of said valid cell area,
in which said offset amount is same within a same block.
12. A digital camera comprising an optical system including an
iris and a solid-state image sensor, wherein: said solid-state image
sensor comprises a valid cell area in which a plurality of valid
cells are placed, each said valid cell having a light-receiving
part and a color filter placed in an on-chip fashion to correspond
to said light-receiving part, and each outputting a charge signal,
in which: said each valid cell further has a light blocking layer
having an aperture corresponding to said light-receiving part; center
of said color filter placed in the peripheral part of said valid
cell area is offset in the direction to the center with respect
to the center of said light-receiving part; each said light-receiving
part and each said color filter are placed in respective predetermined
pitches, the pitch said light-receiving parts are placed in being
greater than the pitch said color filters are placed in; center
of each said aperture of said light blocking layer in said valid
cell reserved at the peripheral part of said valid cell area is
offset in the direction to the center of said valid cell area with
respect to the center of said light-receiving part; each said aperture
is reserved at a predetermined pitch from another said aperture,
the pitch being smaller than the pitch said light-receiving parts
are placed in and greater than the pitch said color filters are
placed in; when the offset amount between the center of said light-receiving
part and the center of said micro-lens is Sm, the offset amount
between the center of said light-receiving part and the center of
said color filter placed in said valid cell is SOCF, the offset
amount between the center of said light-receiving part placed in
said valid cell and the center of said aperture reserved at said
valid cell is SOPN, thickness between said light-receiving part
and the layer on which said micro-lenses are positioned is d1, thickness
between said light-receiving part and said color filter is d2, and
thickness between said light-receiving part and said aperture is
d3, each said thickness fulfills the relation SOPN<SOCF<Sm;
and when the refractive index of the layers placed underneath said
micro-lens is n, said optical system has an eye-relief of l, and
the height of the image is p in said photodetecting cell of aforementioned
solid-state image sensor, at least one of
sin .theta.=p/[n.times.(p.sup.2 +1.sup.2).sup.1/2 ] is satisfied.
13. The solid-state image sensor comprising a valid cell area in
which a plurality of valid cells are placed, said valid cells each
having a light-receiving part and a blocking layer having an aperture
reserved to correspond to said light-receiving part, and each outputting
a charge signal, wherein: center of said aperture of said light
blocking layer in said valid cell reserved at the peripheral part
of said valid cell area is offset to the direction of the center
of said valid cell area with respect to the center of said light-receiving
part; the offset amount between the center of said aperture and
the center of said light-receiving part is larger at the peripheral
part of said valid cell area, and smaller at the central part of
said valid cell area; said valid cell has a micro-lens placed in
an on-chip fashion to correspond to said light-receiving part; said
valid cells each include said aperture and said light-receiving
part, in which the center of said aperture is offset with respect
to the center of said light-receiving part; and the configuration
thereof satisfies an expression of
14. A solid-state image sensor comprising a valid cell area in
which a plurality of valid cells are placed, said valid cells having
a light-receiving part placed and an aperture reserved to correspond
to said light-receiving part, and outputting a charge signal, wherein:
center of said aperture reserved at the peripheral part of said
valid cell area is offset in the direction to the center with respect
to the center of said light-receiving part; said light-receiving
parts are respectively placed and said apertures are respectively
reserved in predetermined pitches, the pitch said light-receiving
parts are placed in being greater than the pitch said aperture is
reserved at; and the configuration thereof satisfies the expressions
15. The solid-state image sensor comprising a valid cell area in
which a plurality of valid cells are placed, said valid cells having
a light-receiving part placed and an aperture reserved to correspond
to said light-receiving part, and outputting a charge signal, wherein:
center of said aperture reserved at the peripheral part of said
valid cell area is offset in the direction to the center with respect
to the center of said light-receiving part; said light-receiving
parts are respectively placed and said apertures are respectively
reserved in predetermined pitches, the pitch said light-receiving
parts are placed in being greater than the pitch said aperture is
reserved at; said valid cells have micro-lenses placed in an on-chip
fashion in a predetermined pitch, to correspond with said light-receiving
part, and the pitch said apertures are reserved at is greater than
the pitch said micro-lenses are placed in; said valid cells each
include said aperture and said light-receiving part, in which the
center of said aperture is offset with respect to the center of
said light-receiving part; and the configuration thereof satisfies
an expression of
Digital Camera Patent Description
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a solid-state image sensor having
a function of decreasing a shading amount, a production method for
the solid-state image sensor, and a digital camera using the solid-state
image sensor, and particularly to a solid-state image sensor wherein
micro-lenses are placed on the photodetecting cells belonging to
the incident side, the production method of the solid-state image
sensor, and a digital camera using the solid-state image sensor.
2. Description of the Related Art
Recently, video cameras and digital cameras have become wide-spread
in general. CCD-type or MOS-type solid-state image sensors are used
in these cameras. In such solid-state image sensors, a plurality
of photodetecting cells having a light-receiving part (a photoelectric
converter) are arranged to form a matrix. The energy of the light
incident to each photodetecting cell undergoes photoelectric conversion
in the light-receiving part, generating a signal charge. The generated
signal charge is outputted to the external parts, via a CCD and
a signal channel.
As shown in FIG. 32, a CCD-type solid-state image sensor 10 of
the related art has a photodetecting cell 13 having a light-receiving
part 2, a vertical CCD 15 and a horizontal CCD 16 constituting the
light-receiving part 2 for transferring signal charges, and an output
amplifier 17.
Among the photodetecting cells placed in the light receiving area
(the area to which the light is incident) of the solid-state image
sensor 10, there are valid cells wherein the energy of the light,
incident to the light-receiving part 2, undergoes a photoelectric
conversion into a signal charge for outputting said signal charge,
and there are photodetecting cells for outputting dark currents
without photoelectric conversion.
A photodetecting cell which outputs dark currents is known, for
instance as a black dummy. Such a photodetecting cell has an incident
side which is shielded from the light, and is generally placed in
one row and/or one column surrounding the valid cell area wherein
a plurality of valid cells are arranged to form a matrix, or in
a row at the extremity of any side of the valid cell area where
a plurality of valid cells are arranged to form a matrix.
In addition, the solid-state image sensor 10 is equipped with a
light-blocking layer such that the light is only incident to the
light-receiving part 2 of the valid cells, and with signal driving
channels for applying voltage to CCD electrodes, not shown in FIG.
32.
In addition, color filters are placed above each light-receiving
part, for taking color pictures with the solid-state image sensor
10.
FIG. 33 is a layout view showing an example of an array of color
filters. The plurality of color filters together forms a layer.
R, G and B represent red, green and blue filters, respectively.
One of the R, G or B filters is placed above the light-receiving
part 2.
In addition, to improve the converging power, a micro-lens is sometimes
placed above each light-receiving part 2. FIG. 34 is a cross-sectional
view showing the structure of a photodetecting cell of the solid-state
image sensor 10 of the related art.
In the above-mentioned solid-state image sensor 10, the light-receiving
part 2 is formed on top of the semiconductor substrate 1 (for example
a silicon substrate), a light blocking layer 9 with apertures 8
is placed on the incident side of the light-receiving part 2.
Above each light-receiving part 2, one of the R, G and B color
filter 4 is placed in an on-chip fashion. In addition, a micro-lens
7 for improving the converging power is placed immediately above
the light-receiving part 2, via a flattening layer 6.
In fact, in such a solid-state image sensor 10, a phenomenon called
shading is known which gives rise to sensitivity fluctuations in
the valid cell area.
Shading originates from the fact that incident lights incident
to the peripheral part of the valid cell area, when compared to
incident lights incident to the central part of the valid cell area,
have incidences which are oblique. In other words, when the light
is obliquely incident, it generates eclipses and a degradation of
the photoelectric conversion rate at the level of the light-receiving
part 2.
In this case, since the quantity of incident lights in the central
part of the valid cell area is greater, for the same quantity of
incident lights, the output signal is greater for the photodetecting
cells of the central part when compared to the photodetecting cells
of the peripheral part. Therefore, a "sensitivity fluctuation"
is generated between the photodetecting cells of the central part
and the photodetecting cells of the peripheral part. Also, in this
document, the "sensitivity fluctuation (or the difference in
the output value)" is called the shading amount. The shading
amount increases when the number of photodetecting cells increases,
and the size of the valid cell area increases.
FIG. 35 shows an example of results obtained when the shading amount
is measured for a solid-state image sensor.
The measurements of the shading amount shown in FIG. 35 have been
obtained by measuring the output of a valid cell area, whose size
was 25.1 mm in the horizontal direction and 16.9 mm in the vertical
direction.
In the drawing, .DELTA. is the G output voltage of the central
part (sensitivity, equivalent to the actual aperture rate), .circle-solid.
is the calculated value for the latter, .times. is the actual measurement
of the G output voltage for the peripheral part, and .quadrature.
is the calculated value of the latter.
From this figure, it is clear that the shading amount between the
central part and the peripheral part depends on the F number of
the digital camera, the result of which is displayed as the difference
in the G output voltage (sensitivity).
A so-called "micro-lens positional offsetting" method,
wherein the center of micro-lens belonging to the peripheral part
is moved towards the central part of the valid cell area, taking
the center of the corresponding light-receiving part as the reference,
and a method wherein the aperture width of the light blocking layer
is larger the closer to the periphery it is, taking the center of
the corresponding light-receiving part as the reference, have been
proposed as methods for decreasing such shading amounts.
Of these, the "micro-lens positional offsetting" is publicly
known, for instance, as disclosed in Japanese patent No. 2600250.
In the "micro-lens positional offsetting", as shown in
FIG. 36, the center of a micro-lens 27 installed above a light-receiving
part (for example a photovoltaic such as a photodiode) 22, is matched
with the center of a light-receiving part 22 (double-broken lines
in the drawing) in case the light-receiving part belongs to the
central part 21A of the valid cell area 21, and offset by a specified
distance d1 towards the central part of the valid cell area 21,
in case the light-receiving part belongs to the peripheral part
21E of the valid cell area.
The specified distance d1 is defined so it becomes greater at a
constant rate, the further from the center 21X of the solid-state
image sensor 20. In addition, optimal values are determined for
the specified distance d1, taking into consideration the characteristics
of the camera lenses and the solid-state image sensor 20 actually
used. In addition, in the drawing, numeral 23 is an inter-level
isolation layer, numeral 24 is a color filter and numeral 26 is
a flattening layer.
The "micro-lens positional offsetting" shading countermeasure
mentioned above has been recognized to be effective to some extent,
but it is still not sufficient. The reasons are explained concretely
in the following.
First, the above-mentioned shading countermeasure has the problem
of not taking into account the solid-state image sensors wherein
color filters are placed on the incident side, thereby generating
color shading due to said color filters. Color shading designates
the offset of color balance between the central part and the peripheral
part.
Second, when applying the "micro-lens positional offsetting",
which offsets the position of the lenses, to actually-made solid-state
image sensors, it has not been possible to decrease the shading
to the same extent as the values calculated in simulations. In addition,
the aperture area of the light blocking layer mounted in the solid-state
image sensor has not been taken into consideration. In other words,
although when the aperture area of the light blocking layer is wider,
the light leak increases and leads to such a problem as cross-talk,
which is one of the effects due to shading, and a malfunction of
switch transistors, these phenomena have not been taken into account
in the "micro-lens positional offsetting".
Third, when designing in shading countermeasures, the characteristics
of the digital camera, wherein the corresponding solid-state image
sensor is applied, have not been taken into account. In other words,
the F number of the camera lens equipped in digital cameras, and/or
the actual F number change the angle of incidence of the light with
respect to the light-receiving part, and this F number dependency
of the incident angle influences the shading amount.
In the related art, taking into account this F number dependency
of the shading amount, a correction to increase the brightness of
the peripheral part by an image processing device installed in the
camera side, has been considered (a software shading correction).
This shading correction is performed as the step 1 of image processing
(FIG. 37) executed by the computer of a digital camera carrying
the solid-state image sensor 10. However, to execute the shading
correction program, it is normally necessary to equip the digital
camera side with a special control circuitry, which raises the costs.
In addition, when the shading amount is large, by executing this
shading correction, the efficiency of other processes requiring
faithful color reproduction degrades and causes the problem of the
image itself becoming unnatural. In addition, when the shading value
is large, depending on the performance of the computer loaded onto
the camera, rapid image processing can be difficult. The defect
becomes a larger problem for CCD-type solid-state image sensors
with an increased valid cell area.
In addition, the above-mentioned F number dependency of shading
becomes particularly problematic, in the case of exchangeable lens-type
digital still cameras, when the camera lens unit is substituted
(when exchanging lenses).
Furthermore, the decrease in the shading amount by "micro-lens
positional offsetting" is limited in the width of the offset,
since it is a method wherein correction is made by offsetting the
position of the micro-lens with respect to the position of the light-receiving
part (photovoltaics). This is particularly a problem for large-size
solid-state image sensors (film-size CCD-type solid-state image
sensors) wherein the degree of oblique incidence is extremely large,
and where not enough offset width can be maintained.
SUMMARY OF THE INVENTION
The present invention has been conceived in view of the above,
and its first object is to provide a solid-state image sensor which
can decrease even color shading, and/or a solid-state image sensor
with superior shading effects, and a digital camera using such solid-state
image sensor.
The second object of the present invention is to provide a solid-state
image sensor which decreases shading independent from the aperture
reserved at the light blocking layer, and/or a solid-state image
sensor with superior shading effects, and digital camera using such
solid-state image sensor.
In addition, the third object of the present invention is to provide
a solid-state image sensor wherein the shading amount of a solid-state
image sensor can be decreased while lowering the F number dependency,
in a simple structure.
In addition, the fourth object of the present invention is to provide
a solid-state image sensor wherein the shading amount of the solid-state
image sensor can be decreased accurately in response to an actual
environment of a digital camera taking pictures.
In order to achieve the above-mentioned objects, the solid-state
image sensor of the present invention comprises a valid cell area
wherein a plurality of light-receiving parts and a plurality of
valid cells having color filters placed in an on-chip fashion corresponding
to the light-receiving parts and outputting charge signals, are
arranged to form a matrix, the color filters placed in the peripheral
part of the valid cell area are offset with respect to the light-receiving
parts in the direction to the center of the valid cell area, and
the offset amounts between the color filters and aforementioned
light-receiving parts become gradually or continuously larger, the
further it is from the center and the closer it is to the periphery
of the valid cell area. The above configuration makes it possible
to appropriately offset the color filters with respect to the position
of the light-receiving parts, and color mixture is decreased even
when oblique incident light components are present.
Preferably, in the solid-state image sensor, the valid cell area
is divided into groups of a plurality of concentric blocks, with
the offset amount between the color filters and the light-receiving
parts being the same within each block, and increasing from the
central part to the peripheral part. The above configuration not
only allows a decrease in color mixture, but also allows use of
a relatively low-cost reticle as a reticle for making color filters
when producing solid-state image sensors, and also decreases production
costs.
In addition, another solid-state image sensor of the invention
comprises a valid cell area wherein a plurality of light-receiving
parts and a plurality of valid cells having color filters placed
in an on-chip fashion to correspond to the light-receiving parts
and outputting charge signals, are arranged to form a matrix, the
color filters placed in the peripheral part of the valid cell area
are offset with respect to the light-receiving parts in the direction
to the center of the valid cell area, the light-receiving parts
and the color filters are placed on a respective constant pitch,
with the pitch for the light-receiving parts being greater than
the pitch for the color filters. This configuration makes it possible
to make the difference of the offset amounts between the color filters
and the light-receiving parts continuous from the central part to
the peripheral part of the valid cell area. Images obtained from
such solid-state image sensors present more natural photographic
subjects.
In addition, another solid-state image sensor of the present invention
comprises a valid cell area, having a plurality of valid cells,
each comprising a light receiving part and a light blocking layer
in which apertures are provided corresponding to the light-receiving
part, for outputting charge signals arranged to form a matrix, the
apertures reserved at the peripheral part of the valid cell area
are offset with respect to the light-receiving parts in the direction
of the center of the valid cell area, and the offset amounts between
the apertures and aforementioned light-receiving parts become gradually
or continuously larger, the further it is from the center and the
closer it is to the periphery of the valid cell area. The above
configuration makes it possible to appropriately offset the apertures
with respect to the position of the light-receiving parts, and the
shading amount can be decreased without generating eclipses, even
when oblique incident light components are present.
Preferably, the valid cells of the solid-state image sensor are
divided into groups of a plurality of concentric blocks, with the
offset amount between the apertures and the light-receiving parts
being the same within each block, and the offset amounts being larger
as it gets further from the central part and closer to the peripheral
part. The above configuration not only allows a decrease in the
shading amount, but it also allows use of a relatively low-cost
reticle for making the apertures of a light blocking layer, and
decreasing the production costs.
In addition, yet another solid-state image sensor of the present
invention is equipped with a valid cell area wherein a plurality
of light-receiving parts and a plurality of valid cells having a
light blocking layer in which apertures are reserved to correspond
to the light-receiving parts and outputting charge signals, are
arranged to form a matrix, the apertures reserved at the peripheral
part of the valid cell area are offset with respect to the light-receiving
parts in the direction to the center, the light-receiving parts
are placed and the apertures are reserved with a respective constant
pitch, with the pitch for the light-receiving parts being greater
than the pitch for the apertures. This configuration allows to make
the difference of the offset amounts between the apertures of the
light blocking layer and the light-receiving parts continuous from
the central part to the peripheral part of the valid cell area,
thus, the images obtained present more natural photographic subjects.
In addition, when considering the application of each of the above-mentioned
solid-state image sensors of the present invention to digital cameras,
the offset between the center of the light-receiving parts and the
center of the micro-lenses equipping the incident side, the offset
between the center of the light-receiving parts and the center of
the color filter, and the offset between the center of the light-receiving
parts and the center of the apertures can be determined based on
the total thickness of layers, the thickness of the layer between
the light-receiving parts and the layer equipped with the micro-lenses,
the thickness of the layer between the light-receiving parts and
the color filters, the thickness of the layer between the light-receiving
parts and the apertures, the refractive index of the layer placed
underneath the micro-lenses and the eye-relief of the optical system
equipped in the digital camera.
In addition, yet another solid-state image sensor of the present
invention has a valid cell area formed by a plurality of photodetecting
cells made of a plurality of photovoltaics arrayed on the main side
of a semiconductor substrate, in which, according to the position
of the photovoltaics inside the valid cell area, on the light receiving
side of the corresponding photovoltaic, a penetration adjusting
device is placed to adjust the optical penetrating amount of the
corresponding incident light. This allows proper reduction in the
shading amount which differs according to the position of the valid
cell area.
Preferably, the penetration adjusting device of the solid-state
image sensor is a layer made of organic materials, formed in the
upper part of the photovoltaics, which has different optical penetration
amount according to the position inside the valid cell area. This
allows, by only changing the optical penetrating rate of the layer
made of organic material, composition for an optimal decrease of
the shading amount without modification of other structures.
In addition, preferably, in the solid-state image sensor, micro-lenses
are configured in the layer made of organic material, formed on
the light receiving side of the photovoltaics, and have different
optical penetrating rates depending on the position from the peripheral
part to the central part of the valid cell area. This allows, by
only changing the optical penetrating rate of the micro-lenses,
in other words, without modification of other structures, composition
for an optimal decrease of the shading amount.
In addition, preferably, in the solid-state image sensor, micro-lenses
are placed on the light receiving side of the photovoltaics, with
the layer made of organic material serving as the flattening layer
formed between the photovoltaics and the micro-lenses. This allows,
by only changing the optical penetrating rate of the flattening
layer, in other words, without modification of other structures,
composition for an optimal decrease of the shading amount.
When producing such solid-state image sensor, a layer is formed,
wherein the optical penetrating rate changes according to the irradiation
amount due to ultra-violet rays, and the layer is selectively exposed
to ultra-violet rays in an amount which changes according to the
position of the valid cell areas. This allows, by only adjusting
the amount of ultra-violet irradiation occurring during the generally
performed "added exposure", production of a solid-state
image sensor with an optimal decrease of the shading amount.
In addition, preferably, a layer is formed wherein the optical
penetrating rate changes according to the temperature of heating,
and the layer is heat-treated, according to the position of the
valid cell areas. This allows, by only adding a simple production
process, production of a solid-state image sensor with an optimal
decrease of the shading amount.
In addition, preferably, the penetration adjusting device is placed
on the light receiving side of the valid cell area, and functions
as an optical penetrating rate controlling device capable of controlling
the optical penetrating rate. Fine control (decrease) of the shading
amount is possible using the optical penetrating rate controlling
device.
In addition, preferably, in the solid-state image sensor, the optical
penetrating rate controlling device controls the optical penetration
rate based on the signals from the brightness sensor mounted in
the surrounding of the valid cell area. This allows for a proper
and optimal decrease of the shading amount according to the actual
environment of a digital camera taking pictures.
Solid-state image sensors configured as mentioned above are mounted
into digital cameras. This allows for correction of the shading
according to the environment of a digital camera taking pictures.
BRIEF DESCRIPTION OF THE DRAWINGS
The nature, principle, and utility of the invention will become
more apparent from the following detailed description when read
in conjunction with the accompanying drawings in which like parts
are designated by identical reference numbers, in which:
FIG. 1 is a planar view of a CCD-type solid-state image sensor
100 of the present invention.
FIG. 2 is a cross-sectional view of the solid-state image sensor
100 in relation to a first embodiment of the present invention.
FIG. 3 shows the solid-state image sensor 100 in relation to a
variation example of the first embodiment of the present invention,
with (a) being a planar view and (b) being a cross-sectional view.
FIG. 4 is a table indicating the sizes and offset amounts for a
solid-state image sensor 110, a variation example of the first embodiment.
FIG. 5 is a cross-sectional view of a solid-state image sensor
120 in relation to a second embodiment of the present invention.
FIG. 6 is a cross-sectional view of a solid-state image sensor
150 in relation to a third embodiment of the present invention.
FIG. 7 is a measurement graph relating offset amount (Sm) between
a light-receiving part and a micro-lens, and the signal output from
the photodetecting cell, when the interval between the light-receiving
part and the micro-lens is varied.
FIG. 8 is a Figure for explaining the parameters required to calculate
the "appropriate offset amount", wherein (a) is a cross-sectional
view showing the configuration of a photodetecting cell of the solid-state
image sensor 150, (b) is a view showing the substantial part of
a digital camera loaded with the previous element, and (c) is a
magnified view of the surroundings of the micro-lens.
FIG. 9 is a cross-sectional view showing the light paths obtained
when simulating how the light converges at the level of each of
the photodetecting cells placed in the central part and in the peripheral
part of the valid cell area belonging to the solid-state image sensor
150.
FIG. 10 is a graph showing the relationship between the F number
of the optical system of a digital camera loaded with a solid-state
image sensor and the converging power.
FIG. 11 is a table showing each parameter of a digital camera in
relation to a fourth embodiment.
FIG. 12 is a cross-sectional view of a solid-state image sensor
200 of a fifth embodiment of the present invention.
FIG. 13 is a view showing the arrangement of a color filter 204
according to the Bayer array.
FIG. 14 is a planar view showing the layout of a block 210A, a
block 210B, etc., of the solid-state image sensor 200.
FIG. 15 is a cross-sectional view showing a production process
of the solid-state image sensor 200 of a fifth embodiment.
FIG. 16 is a planar view showing the layout of a mask area 250A,
mask area 250B, . . . , and mask area 250E, of a mask 250.
FIG. 17 is a planar view showing the patterns of micro-domains
in the mask, for differentiating the optical penetrating rate for
ultra violet rays of the mask area 250A, the mask area 250B, . .
. , and the mask area 250E.
FIG. 18 is a view showing the overall structure of a digital single-lens
reflex camera 300 loaded with the solid-state image sensor 200.
FIG. 19 is a graph showing the difference between the G output
voltage when the optical penetrating rate for ultra-violet rays
is 0% and the G output voltage when the optical penetrating rate
for ultra-violet rays is 100%, as a function of the F number.
FIG. 20 is a planar view showing another layout of the block 210A,
the block 210B, . . . of the solid-state image sensor 200.
FIG. 21 is a planar view showing yet another layout of the block
210A, the block 210B, . . . of the solid-state image sensor 200.
FIG. 22 is a cross-sectional view showing a production process
of the solid-state image sensor 200 of a sixth embodiment.
FIG. 23 is a view of the part of a production process of a seventh
embodiment of the solid-state image sensor 200 showing how the optical
penetrating rate is modified by heating a micro-lens 207.
FIG. 24 is a cross sectional-view of an eighth embodiment of a
solid-state image sensor 400.
FIG. 25 is a view of the part of a production process of the eighth
embodiment of the solid-state image sensor 400, showing how the
optical penetrating rate is modified by heating a flattening layer
403.
FIG. 26 is a graph showing the difference between the effective
aperture ratio when the optical penetrating rate of the flattening
layer 403 is 100% and the effective aperture ratio when the optical
penetrating rate of the flattening layer 403 is 70%, as a function
of the F number, in the solid-state image sensor 400.
FIG. 27 is a view showing the cross-sectional form and the planar
form of a ninth embodiment of a solid-state image sensor 500.
FIG. 28 is across-sectional view showing a magnification of a part
of an optical penetrating rate controlling layer (EC layer) 600.
FIG. 29 is a view showing the positioning pattern of a light sensor
561 in the periphery of a valid cell area 510 of the solid-state
image sensor 500.
FIG. 30 is a view showing the cross-sectional form and the planar
form of the solid-state image sensor 500 in relation to a variation
example of the ninth embodiment.
FIG. 31 is a view showing an optical penetrating rate controlling
layer (EC layer) 800 for use in a CCD of a tenth embodiment
FIG. 32 is a cross-sectional view of a CCD-type solid-state image
sensor 10 of the previous art.
FIG. 33 is a layout view showing an example of arrangement of the
color filters of the solid-state image sensor 10 of the previous
art.
FIG. 34 is a cross-sectional view of the solid-state image sensor
10 of the previous art.
FIG. 35 is a graph showing the effects due to shading correction
in the solid-state image sensor 10 of the previous art, as a function
of the F number.
FIG. 36 is a cross-sectional view of a solid-state image sensor
20 wherein the "micro-lens positional offsetting" method
of the previous art has been applied.
FIG. 37 is a correction flow chart showing the image processing
performed on the digital camera side.
FIG. 38 shows the cross-sectional structure of the peripheral part
of valid cell area of a solid-state image sensor 30 of the previous
art, with (a) being a cross-sectional view for simplifying explanation
and (b) being a cross-sectional view of a case where a more homogeneous
control of the thickness of the color filter layer has been performed.
FIG. 39 is a cross-sectional view of the peripheral part of valid
cells belonging to a solid-state image sensor 40 of the previous
art.
FIG. 40 is a cross-sectional view of a magnified part of a solid-state
image sensor 50 wherein the "micro-lens positional offsetting"
method of the previous art has been applied.
DESCRIPTION OF THE PREFERRED EMBODIMENT
(First embodiment)
In the following, the first embodiment according to the invention
will be explained referring to the drawings.
To begin with, the reasons which led the inventors to the invention
of the first embodiment will be explained.
The inventors determined that the cause of the above-mentioned
color shading (first problem) was attributable to color mixing due
to the oblique incidence of the light. FIG. 38 is a cross-sectional
view of the peripheral part of a valid cell area (the region where
valid cells are arranged to form a matrix) of a solid-state image
sensor 30 wherein the "micro-lens positional offsetting"
has been applied.
In FIG. 38, (a) is a view for explaining and (b) is a view of a
case where a more homogeneous control of the thickness of the color
filter layer has been performed. In addition, here, the light blocking
layer is omitted.
A light-receiving part 32 is placed above a semiconductor substrate
(silicon substrate, for example) 31 and a micro-lens 36 is placed
by offsetting the position with respect to the light-receiving part
32, to decrease shading. The micro-lens 36 is formed above a flattening
layer 35. A CCD electrode 37 is placed between the light-receiving
part 32 and the light-receiving part 32. A silicon-oxide layer 39
is placed to isolate the CCD electrode 37 and protect the light-receiving
part 32.
A color filter 34 is formed above an inter-level isolation layer
33 by the spin-coating method, for each color. When produced as
above, the thickness of the filter layer differs depending on the
colors, and steps are generated. To give an example, in contrast
to a thicker 2.5 .mu.m thick color filter 34-1, a thinner color
filter 34-2 would be about 1.2 .mu.m thick.
When a step is generated between the color filters 34-1 and 34-2
which are different from each other, the light passes also through
the color filter of the adjacent photodetecting cell, following
the path indicated by numeral 38a.
This is the cause of color mixing. Color mixing is generated by
the oblique incidence of the light, as above. However, the oblique
incident light component decreases towards the central part of the
valid cell area.
Therefore, between the central part and the peripheral part of
the valid cell area, the degree of color mixing is different, and
accordingly, the color balance also differs between the central
part and the peripheral part of the valid cell area.
In addition, as in (b) of FIG. 38, for example, even if an ideal
homogeneity of the thickness of the layer could be controlled, with
the spin-coating method, the periphery of the filter formed afterwards
is raised. Therefore, it is still a cause of color mixing.
The solid-state image sensor 100 of the first embodiment is configured
such that the center of the color filters placed at the periphery
of the valid cell area is offset with respect to the center of the
light-receiving parts, in the direction of the center of the valid
cell area, as described in detail below. This allows decrease of
color shading due to the above causes.
A concrete explanation is given below.
FIG. 1 is a planar view of a CCD-type solid-state image sensor
100 of the previous art, and FIG. 2 is a cross-sectional view showing
the configuration of the photodetecting cell of the solid-state
image sensor 100 in relation to a first embodiment of the present
invention. In addition, in the solid-state image sensor 100 of FIG.
1, the pitch of a light-receiving part 102 is greater than the pitch
of a color filter 104, offsetting the center of the color filter
104 with respect to the center of the light-receiving part 102.
Furthermore, in FIG. 1 numeral 132 is a vertical CCD, numeral 133
is a horizontal CCD and numeral 134 is an output amplifier.
The light-receiving part 102 of the solid-state image sensor 100
is placed with a constant pitch at the top of a semiconductor substrate
101, as shown in FIG. 2. On top of these, via an inter-level isolation
layer 103, a color filter 104 is placed in an on-chip fashion. The
color filter 104 is also placed with a constant pitch, however,
it is smaller than the pitch of the light-receiving part 102.
In this case, in the central part of the valid cell area wherein
a plurality of valid cells are arranged to form a matrix, the center
of the light-receiving part 102 of a photodetecting cell coincides
with the center of the corresponding color filter 104. Then, at
the level of each photodetecting cell, the offset amounts between
the centers of the light-receiving parts 102 and the centers of
the color filters 104 become gradually larger, the further it is
from the central part, and the closer it is to the peripheral part.
Such a configuration allows decrease of color shading due to oblique
incident lights, in the solid-state image sensor 100. In addition,
the difference between the offset amounts between the center of
the color filter 104 and the center of the light-receiving part
102 is gradual, from the position closer to the central part, to
the position closer to the peripheral part of the valid cell area.
This allows for images obtained from the solid-state image sensor
100 to present more natural photographic subjects.
In addition, in this case, a micro-lens 107 is placed above a flattening
layer 106. The pitch of the micro-lens 107, when compared to the
pitch of the color filter 104, is smaller. Such a configuration
allows further decrease in the shading amount. In addition, the
converging power of each photodetecting cell in the solid-state
image sensor 100 is improved evenly.
In fact, FIG. 2 schematically shows the structure of the photodetecting
cells in the central part and the photodetecting cells in the peripheral
part of the valid cell area belonging to the solid-state image sensor
100. As shown in FIG. 2, in the peripheral part of the valid cell
area, the micro-lens 107, the color filter 104 and the light-receiving
part 102, drawn on top of the broken line, form one set corresponding
to a configuration of one photodetecting cell.
For example, in the solid-state image sensor 100 of the first embodiment,
the size of one side of the photodetecting cell is 10 .mu.m, the
size of the valid cell area is 24 mm.times.16 mm and the size of
the light-receiving part 102 is 8 .mu.m.times.4 .mu.m.
In addition, the offset amount for the color filter 104 in the
edge part of the valid cell area is designed to be 3 .mu.m in the
direction of the long side and 1 .mu.m in the direction of the short
side.
FIG. 3 is a figure for explaining the solid-state image sensor
100 in relation to a variation example of the first embodiment,
wherein (a) is a planar view and (b) is a cross-sectional view.
In FIG. 3(a), an A block, a B block and a C block represent the
valid cell area, and a D block represents an optically opaque part,
or a peripheral circuit part.
The valid cell area is divided into concentric groups, from the
central part to the peripheral part. In the central A block, the
center of a light-receiving part 112 coincides with the center of
a color filter 114 corresponding to the light-receiving part 112.
On the other hand, in the photodetecting cell of the other blocks,
the center of the color filter 114 is offset with respect to the
center of the light-receiving part 112, in the direction of the
center of the valid cell area. In addition, the offset amount (SOCF)
is constant within each block, and is greater in the peripheral
part. For example, it is greater in the B block than in the C block.
Such a configuration, as in the case of the solid-state image sensor
100 shown in FIG. 1 and FIG. 2, allows the offset amounts between
the centers of the light-receiving parts 112 and the centers of
the color filters 114 to become gradually larger, the further it
is from the photodetecting cells in the center and the closer it
is to the photodetecting cells in the periphery. This results in
the decrease of color shading due to oblique incident lights. In
this case, a micro-lens 117 is placed above a flattening layer 116.
In addition, in FIG. 3 (b), numeral 111 is a semiconductor substrate,
numeral 113 is an inter-level isolation layer.
In this case, the center of the micro-lens 117 in the B and C blocks,
as is the case for the color filter 114, is offset with respect
to the center of the light-receiving part 112, in the direction
of the center of the valid cell area. In addition, the offset amount
(Sm) is constant within each block. In addition, the offset amount
(Sm) is greater than SOCF, and the offset amount of the block closer
to the peripheral part is greater. This allows not only a further
decrease in the shading amount, but also improvement in the converging
power.
The appropriate value for a concrete "offset amount"
is different depending on the size of the cell, the thickness of
each layer, and so forth. The "appropriate offset amount"
is described below.
An example of the sizes of the solid-state image sensor 100 of
the current embodiment and offset amounts are shown in FIG. 4. The
offset amount is different between the X direction and the Y direction,
owing to a difference in the size of the valid cell area between
the X direction and the Y direction.
In fact, semiconductor devices having the microscopic patterns
that the solid-state image sensor 100 has, are produced using a
photolithography apparatus called a stepper. The photolithography
apparatus transfers the pattern of a reticle onto the top of a semiconductor
substrate.
The reticle is obtained by patterning a layer of metal such as
chrome on the top of a silicone glass plate. The cost is higher
for reticles having finer microscopic patterns formed in the metal
layer. A high-cost reticle is required to produce a solid-state
image sensor.
As described above, in the solid-state image sensor, differentiating
the pitch of the color filters or the pitch of the micro-lenses
from the pitch of the light-receiving parts (as in the case of the
solid-state image sensor of FIG. 2 ) involves changing SOCF or Sm
by a minute amount (in the order of 0.01 .mu.m) for each photodetecting
cell.
If a characteristic of the solid-state image sensor (for example,
color reproducibility) is important, the offset amount between the
color filter and the light-receiving part can be made so that the
difference in offset amounts according to different positions is
continuous from the central part to the peripheral part of the valid
cell area. In this case, however, an even higher cost reticle is
required.
In contrast, if the offset amount is different for each block as
in the case of the solid-state image sensor 110 shown in FIG. 3,
a comparatively rough accuracy of the pattern on the reticle is
acceptable. Therefore, a low-cost reticle can be used, which decreases
the production cost.
In either case, it is possible to decrease the color shading, with
the solid-state image sensor 100 or 110 of the embodiment. Either
configuration may be selected according to the objective for the
use of the solid-state image sensor or the degree of freedom in
the design and development.
(Second embodiment)
In the following, the second embodiment according to the invention
will be explained using FIG. 5.
The inventors determined the cause of the problem (the second problem,
i.e. the eclipse) wherein the shading amount differs depending on
the shape of the aperture reserved at the light blocking layer.
This is explained below. FIG. 39 is a cross-sectional view of the
peripheral part of the valid cell area belonging to a solid-state
image sensor 40 of the previous art. Also, the color filter is omitted.
The "micro-lens positional offsetting" is applied to
the solid-state image sensor 40, so that the center of a micro-lens
46 is placed to be offset with respect to the center of a light-receiving
part 42 taken as the reference. In addition, in the drawing, numeral
41 is a semiconductor substrate.
As it is clear from the FIGure, even if the center of the micro-lens
46 is offset with respect to the center of the light-receiving part
42, if an aperture 49 of a light blocking layer 47 is reserved immediately
above the light-receiving part 42, an oblique incident light 45b
is generated, blocked by the light blocking layer 47. This is why,
it was not possible to decrease the shading amount as effectively
as the simulations.
In a solid-state image sensor 120 of the second embodiment, the
center of the aperture in the peripheral part of the valid cell
area is offset with respect to the center of the light-receiving
part, in the direction of the center of the valid cells. This configuration
makes it possible to decrease the shading amount due to the above-mentioned
cause.
A concrete explanation is given below.
FIG. 5 is a cross-sectional view showing the structure of the solid-state
image sensor 120 in relation to the second embodiment of the present
invention.
In the solid-state image sensor 120, by increasing the pitch of
a light-receiving part 122 to be greater than the pitch of an aperture
129 of a light blocking layer 126 (simply called aperture, hereafter),
the center of the aperture 129 is offset with respect to the center
of the light-receiving part 122.
In other words, the light-receiving part 122 is placed with a constant
pitch on top of a semiconductor substrate 121. Above these, the
light blocking layer 126 containing the aperture 129 is placed via
an inter-level isolation layer 128. The aperture 129 is also reserved
with a given constant pitch which is smaller than the pitch of the
light-receiving part 122. In addition, in the drawing, numeral 127
is a micro-lens.
In this case, in a photodetecting cell belonging to the center
part belonging of the valid cell area, the center of the light-receiving
part 122 coincides with the center of the corresponding aperture
129. This allows the offset amounts between the apertures 129 and
the light-receiving parts 122 to gradually become larger, the further
it is from the photodetecting cells of the central part and the
closer it is to the photodetecting cells of the peripheral part.
This results in a further decrease of the shading amount due to
oblique incident lights. In addition, the offset amounts between
the center parts of apertures 129 and the centers of the light-receiving
parts 122 are continuously different from the central part to the
peripheral part of the valid cell area, allowing for obtained images
to present photographic subjects more naturally.
In addition, in the embodiment, the micro-lens 127 is placed on
above the inter-level isolation layer 128. Also, the pitch of the
micro-lens 127, when compared to the pitch of the aperture 129,
is smaller. As a result of this, not only is the shading amount
further decreased, but the converging power is also improved. In
addition, the micro-lens 127, the aperture 129 and the light-receiving
part 122, drawn on top of a broken line in FIG. 5, form one set
corresponding to the configuration of one photodetecting cell.
In addition, in the embodiment, by changing the pitch of the light-receiving
part 122 and the pitch of the aperture 129, the shading amount is
decreased. However, the above does not constitute a limitation,
and it is also acceptable to divide the valid cell area into groups,
as shown in the variation example of the first embodiment, offset
the aperture 129 with respect to the light-receiving part 122 in
the direction of the center of the valid cell area, and make the
offset amount (SOPN) constant for each block. In this case, it suffices
to make the offset amount between the center of the light-receiving
part and the center of the aperture, within the block of the peripheral
part, greater than the offset amount in the central part. If the
offset amount is defined block by block as above, when forming the
aperture of the light blocking layer, a low-cost reticle can be
used, which decreases the production cost for the solid-state image
sensor.
(Third embodiment)
In the following, the third embodiment according to the invention
will be explained.
FIG. 6 is a cross-sectional view showing the configuration of a
solid-state image sensor 150 in relation to a third embodiment of
the present invention.
In addition, the planar layout of the solid-state image sensor
150 is almost identical to the layout shown in FIG. 3 (a). In other
words, the valid cell area is divided into an A block, a B block
and a C block (a D block represents an optically opaque part, or
a peripheral circuit part).
In this case, each photodetecting cell of the solid-state image
sensor, forms a light-receiving part 152 on top of a semiconductor
substrate (a silicone plate) 151, and an aperture 159 of a light
blocking layer 156, a color filter 154 and a micro-lens 157 are
placed, corresponding to the light-receiving part 152.
In the central A block, the center of the light-receiving part
152 coincides with each of the center of the aperture 159, the center
of the color filter 154 and the center of the micro-lens 157, corresponding
to the light-receiving part 152.
On the other hand, at the level of the photodetecting cells of
the other blocks, each of the center of the aperture 159, the center
of the color filter 154, and the center of the micro-lens 157 is
offset with respect to the center of the light-receiving part 152,
in the direction of the central part of the valid cell area.
The offset amounts (sequentially, SOPN, SOCF and Sm) for each photodetecting
cell, are constant for the photodetecting cells contained in the
same block, and are greater for photodetecting cell of a block close
to the peripheral part. In addition, each offset amount for a photodetecting
cell, which is constant for each block, is related by SOPN SOCF
Sm.
The above makes it possible for the offset amounts to become gradually
larger with respect to the light-receiving parts 152 of each of
the centers of the apertures 159, the centers of the color filters
154 and the centers of the micro-lenses 157, in photodetecting cells,
the further they are from the central part and the closer they are
to the peripheral part. Therefore, the shading and the shading amounts
due to oblique incident lights are decreased.
Here, a method for determining the "appropriate offset amount"
with respect to a light-receiving part, for the aperture, the color
filter and the micro-lens, will be explained referring to FIG. 7
and FIG. 8.
FIG. 7 is a measurement graph relating the "offset amount
(Sm)" between a light-receiving part and a micro-lens, and
the signal output from the photodetecting cell, when the interval
between the center of the light-receiving part and the center of
the micro-lens is varied.
The horizontal axis is the "offset amount" of the micro-lens
with respect to the corresponding light-receiving part, and the
vertical axis is the output voltage. In addition, in this case,
the position of the aperture of the light blocking layer coincides
with position of the color filter, and the "offset amount"
of the micro-lens as well as the thickness of the flattening layer
placed immediately underneath the micro-lens are varied. In other
words, the distance between the light-receiving part and the micro-lens
is varied.
The thickness of the flattening layer is larger in the order of
161, 162 and 163.
As is clear from the figure, an appropriate value (or range) to
decrease the shading exists for the "offset amount" of
a micro-lens, and it is revealed that the "appropriate offset
amount" is larger corresponding to the distance between the
micro-lens and the light-receiving part being larger. FIG. 7 is
the data on the offset amount of the micro-lens.
However, if an appropriate value exists for the "offset amount"
of a micro-lens, similarly, an "appropriate offset amount"
should also exist for the aperture of a light blocking layer, or
a color filter.
Therefore, based on the above-mentioned results, a number of assumptions
were made, leading to each "appropriate offset amount".
FIG. 8 is a drawing for explaining the parameters required to calculate
the "appropriate offset amount", wherein (a) is a cross-sectional
view of the solid-state image sensor 150 of the invention, (b) is
a view showing the substantial part of a digital camera 160 loaded
with the solid-state image sensor 150, and (c) is a magnified view
of the surroundings of the micro-lens.
In FIG. 8, SOPN, SOCF and Sm are the "offset amounts"
with respect to the light-receiving part 152 for each of the center
of the aperture 159, the center of the color filter 154 and the
center of the micro-lens 157, respectively.
In addition, d3, d2 and d1 are the distances from the upper side
of the light-receiving part 152 to each of the upper side of the
aperture 159, the upper side of the color filter 154, and the upper
side of the micro-lens 157, respectively.
In addition, as shown in FIG. 8(b), the digital camera 160, wherein
the above-mentioned solid-state image sensor 150 is applied, has
an optical system 166, wherein a lens 167 and an iris 168 are placed.
In addition, an optical system is shown in the drawing, having a
two lens set of numeral 167-1 and numeral 167-2, but this does not
constitute a limitation. The light becomes incident to the solid-state
image sensor 150, via the optical system 166.
Incident lights 169 have an almost perpendicular incidence in the
central part of the valid cell area. On the other hand, in the peripheral
part, incident lights 20 have an incidence other than a right angle.
In this case, P, indicated in FIG. 8(b), is the distance between
the center of the valid cell area and the photodetecting cell for
which the "appropriate offset amount" is calculated (generally
called "height of the image"), and l is the distance between
the light receiving side of the solid-state image sensor 150 and
the iris 168 (generally called "eye-relief").
The "appropriate offset amount" is sought with a number
of assumptions.
First, the "appropriate offset amount" is calculated
taking the rays incident to the center of each micro-lens (169a,
170a; to simplify, will be called .theta.0 rays, hereafter), as
the reference.
The light passing through the optical system 166 of the digital
camera 160, actually has a certain spread centered around the incident
light which passed through the center of the micro-lens at the level
of the height of the image P, which is clear from FIG. 8(b). The
light having the aforementioned spread, becomes incident to the
micro-lens of the solid-state image sensor, and exits to the light-receiving
part. Therefore, by simulating each condition for taking a picture,
for example, the converging condition by micro-lenses, of a light
flux wherein the spread depends on the F number of the camera lens
system, an accurate calculation of an "optimal offset amount"
is possible, according to the characteristics of actually mounted
micro-lenses and color filters, and actually reserved apertures
of light blocking layers.
Here .theta.0 rays are used as the reference. This makes it possible
to consider the camera lens and easily calculate the "optimal
offset amount" according to the characteristics of the micro-lens,
color filter or aperture of a light blocking layer.
Second, each type of layer between the micro-lens and the light-receiving
part is assumed to have the same refractive index as the refractive
index n of the layer immediately below the micro-lens. Then, the
incident angle .theta.0 and the exit angle .theta. of the micro-lens
are determined by the angle of the first refraction after the .theta.0
rays have arrived on the solid-state image sensor, as shown in FIG.
8(c).
In reality, in the solid-state image sensor, between the micro-lens
and the light receiving part, a plurality of layers are formed such
as the flattening layer, the color filter, the insulation layer,
the oxide layer. Therefore, strictly, since the refraction index
of each layer is different, the incident light is refracted in a
complex manner at the level of each layer. However, the difference
in the indices being generally small, they are assumed to be constant
(refraction index n), as described above. Following this assumption,
the relationship between the incident angle .theta.0, the exit angle
.theta. and the refraction index n, is expressed in the following
equation, according to Snell's law:
Third, although the incident lights output from the camera lens
system are to first become incident to the micro-lenses, whether
there are micro-lenses or not, the .theta.0 rays are assumed to
arrive on the light-receiving part the via the same route. Therefore,
the "appropriate offset amount" calculated here is applicable
to solid-state image sensor without micro-lenses.
In addition, measurement of the exit angle .theta. from the micro-lens
is difficult in reality. Therefore, the incident angle .theta.0
at the micro-lens is defined by the following equation, using the
height of the image P, the eye-relief 1, the equation have being
an approximation, obtained by taking into consideration geometrical
considerations.
From equations (1) and (2), the exit angle is approximated by sin
.theta.=P/[n.times.(P.sup.2 +1.sup.2).sup.1/2 ] and the exit angle
is defined here by the previous equation.
With the above-described approximation, the optimal values of Sm,
SOCF and SOPN, i.e., the offset amounts with respect to the light-receiving
part for micro-lens, color filter and aperture, are calculated with
the following equations, as a function of the distance from the
light-receiving part and the exit angle .theta..
It is mostly preferred to use the above-mentioned equations in
order to calculate all of the "appropriate offset amounts"
Sm, SOCF and SOPN, of the photodetecting cell of the valid cell
area belonging to the solid-state image sensor. However, even if
not all of the Sm, SOCF and SOPN, are set to the values of the above
equations, in other words, even if one of the offset amounts Sm,
SOCF and SOPN, is set to a value calculated using the above equation,
there will be an effect on the decrease in the shading amount.
Also, these "appropriate offset amounts" are theoretical
values obtained by approximation, based on the above-mentioned equation.
Because of errors in the accuracy of the semiconductor manufacturing
technique, actually produced solid-state image sensors are not always
produced exactly by offsetting micro-lenses, color filters and apertures
only by the "appropriate offset amounts" obtained from
the equations (3), (4) and (5). Therefore, "appropriate offset
amounts" having the optimal values of the above equations provided
with a certain width, are preferred. According to the experiments,
it is possible to provide the "appropriate offset amounts"
with a width of .+-.30%. Therefore, the "appropriate offset
amounts" are calculated using the equations (6), (7) and (8),
with the condition of satisfying the relation SOCF<SOPN <Sm.
Even if not all of the Sm, SOCF and SOPN, are not set to values
in the ranges of the above equations, in other words, even if only
one of the offset amounts Sm, SOCF and SOPN, is set to a value in
the ranges of the above equation, there will be an effect on the
decrease in the shading amount.
In addition, if the "offset amounts" are to be changed
for each block as in the case of the solid-state image sensor 110
shown in FIG. 3, it suffices for the mean value of the offset amounts
of each block to fall within the ranges of the optimal values obtained
with the above equations. This also makes it possible to achieve
a decrease of the shading.
Here, it is clear that, focussing on equation (3) and equation
(4), the optimal ratio between Sm and SOCF is given by the ratio
between d1 and d2, and focussing on equation (3) and equation (4),
the optimal ratio between Sm and SOPN is given by the ratio between
d1 and d3. The equations for calculating the optimal ratio are shown
in (9) and (10).
In addition, if a width of .+-.30% is to be provided, the appropriate
ranges of ratio expressed by the above equations (9) and (10) are
calculated with the equations (11) and (12) below, provided with
the width of .+-.30%.
Also in this case, even if the actually manufactured solid-state
image sensor does not satisfy the conditions of both equations (11)
and (12), if the ratio of at least one of the above equations is
satisfied, there will be an effect on the decrease in the shading
amount.
In addition, if the offset amounts are to be changed for each block
as in the case of the solid-state image sensor 110 shown in FIG.
3, it suffices for the mean value of the offset amounts of each
block to fall within the ranges of the optimal values obtained with
the above equations.
FIG. 9 is a cross-sectional view obtained by simulation, showing
how the light converges in a photodetecting cell of the central
part, and how the light converges in a photodetecting cell of the
peripheral part, of the valid cell area belonging to a solid-state
image sensor.
The upper part is the light path of the incident light when the
F number of the lens of the optical system placed on the digital
camera is 1.4, and the lower part is the light path of the incident
light when the F number of the lens of the optical system placed
on the digital camera is 11. In the drawing, (a) and (e) are photodetecting
cells wherein the micro-lens and the aperture have been offset by
the "appropriate offset amounts" for photodetecting cells
of the peripheral part, given by the above-mentioned equations (6)
and (8). In addition, (b) and (f) in the drawing are photodetecting
cells in which only the micro-lens has been similarly offset by
the "appropriate offset amount", (c) and (g) in the drawing
are comparative examples wherein neither have been offset, and (d)
and (h) in the drawing are photodetecting cells of the center.
In the above simulation, the size of the photodetecting cell was
10 .mu.m.times.10 .mu.m, the thickness of the micro-lenses was 2.5
.mu.m the size of the aperture of the light blocking layer and the
size of the light-receiving part was 8 .mu.m in the horizontal direction
and 4 .mu.m in the vertical direction. In addition, the values for
d1, d2 and d3 (Ref. FIG. 8(a)) were 7 .mu.m, 5.5 .mu.m and 2.5 .mu.m,
respectively.
In addition, the offset amount for the micro-lenses of the photodetecting
cells of the peripheral part was 0.8 .mu.m in the horizontal direction
and 0.6 .mu.m in the vertical direction, in the direction of the
center of the image-sensing side. In the photodetecting cells of
the peripheral part, the center of the aperture of the light blocking
layer was offset by only 0.4 .mu.m (offset amount) in the horizontal
direction and 0.3 .mu.m (offset amount) in the vertical direction,
in the direction of the center of the valid cell area.
When the micro-lenses are offset with respect to the light-receiving
part, since the light converges in the center of the light-receiving
part, there is an effect on the decrease in the shading amount (FIG.
9(b) and (f)).
In addition, the existence of light components eclipsed by the
light blocking layer can be seen. When the light blocking layer
is further offset, the incident lights are adequately incident to
the light-receiving part, and the shading amount is further decreased
(FIG. 9(a) and (e)).
In other words, also from this simulation, it can be seen that,
in the photodetecting cells of the peripheral part, by offsetting
the position of the micro-lens in the direction of the center of
the valid cell area, and by further offsetting the aperture of the
light blocking layer adequately with respect to the above offset
micro-lens, obliquely incident lights pass through the center of
the aperture of the light blocking layer and converge efficiently
onto the light-receiving part.
FIG. 10 is a graph showing the relationship between the F number
of the lens used in the optical system of a digital camera, and
the converging power of a solid-state image sensor.
The relationship between the converging power and the F number
is shown by a dotted line 171 for the photodetecting cells of the
central part of a valid cell area (FIG. 9(d) and (h)), by a solid
line 172 for the photodetecting cells in the peripheral part wherein
the micro-lens and the aperture have been offset to their respective
appropriate position ((a) and (e)), by a solid line 173 for the
photodetecting cell of the peripheral part wherein only the micro-lens
has been offset to the appropriate position, and by a solid line
174 for the photodetecting cells of the peripheral part, wherein
neither the micro-lens nor the aperture have been offset.
As can be seen also from this graph, the shading is improved even
when "positional offsetting" with respect to the light-receiving
part is performed only for the micro-lens, but the shading amount
is further decreased, by adequately performing a further "positional
offsetting" of the aperture of the light blocking layer.
In addition, here, a configuration wherein the micro-lens and the
aperture of the light blocking layer are offset with respect to
the light-receiving part have been explained as an example. However,
equivalent simulation results have been obtained when the "positional
offsetting" was performed not only for the micro-lenses and
the apertures, but also for the color filters. In addition, with
this configuration, a further decrease in the color shading can
be achieved.
(Fourth Embodiment)
The fourth embodiment according to the invention will be explained
in the following. FIG. 11 is a table showing each parameter of a
digital camera in relation to a fourth embodiment according to the
present invention.
Not shown in the drawing, in the solid-state image sensor loaded
on the embodiment, similarly to the solid-state image sensor 100
of the first embodiment, the pitch of the light-receiving part,
and the pitch of each of the micro-lens, the color filter, and the
aperture are varied.
The pitch of the light-receiving part is set uniquely by the number
of photodetecting cells which is dictated by the resolution power
required for the solid-state image sensor and the size of the solid-state
image sensor.
For example, if the size of the valid cell area is 24 mm in the
X direction, and 1000 photodetecting cells are aligned in the X
direction, the pitch of the light receiving part is 24 .mu.m.
In the embodiment, in order to obtain the pitch for each micro-lens,
color filter and aperture, the "offset amounts" for the
outermost part of the valid cell area are first obtained based on
the previously explained equations (3), (4) and (5). Then, the pitch
for the aperture is calculated from these "offset amounts".
In fact, the "optimal offset amounts" obtained from equations
(3), (4) and (5) do not necessarily change linearly from the central
part to the peripheral part of the valid cell area. In such a case
where the change is not linear, with the configuration of the embodiment,
a deviation from the "appropriate offset amount" may occur
for photodetecting cells in an area other than the outermost part.
In FIG. 11, the "optimal offset amounts" calculated from
equations (3), (4) and (5) are listed together with the actual offset
amounts arising from the above-mentioned pitch. In addition, the
actual offset amount was obtained from the equation below.
From FIG. 11, it is clear that both "offset amounts"
are almost the same. However, in this case, the calculation is performed
with p Max=15 mm.
The reason for the above-mentioned match is due to the fact that,
as the height of the image decreases, the inclination .theta. of
incident lights inside the solid-state image sensor decreases, and
the approximation sin .theta..apprxeq.tan .theta..apprxeq.p/l becomes
applicable.
(Fifth Embodiment)
In the following, the fifth embodiment according to the invention
will be explained referring to FIG. 12 to FIG. 21.
The inventors determined that, in the solid-state image sensor,
even when the center of a micro-lens 57 of a peripheral part 51E
of the valid cell area is placed offset only by a specified distance
d1 from the center of a light-receiving part 52 by applying the
"micro-lens positional offsetting", a portion of the incident
lights converged onto the light-receiving part 52 by a micro-lens
57 of the peripheral part 51E, can not become incident to the light-receiving
part 52, depending on their incidence angle, as shown in FIG. 40.
This is because in the solid-state image sensor, the specified distance
d1, which is the "offset amount" for the micro-lens 57,
is determined by taking a predetermined incidence angle (a value
defined by the F number) as the reference.
Therefore, even in the case where the same camera lens is used,
if the F number is changed by changing the aperture, a portion of
the incident lights do not become incident to the light-receiving
part 52 in the peripheral part 51E, and the amount of converging
light is decreased when compared to the central part of the valid
cell area, leading to shading occurring, as shown by the broken
lines in FIG. 40. In addition, in the drawing, numeral 53 is an
inter-level isolation layer, numeral 54 is a color filter and numeral
56 is a flattening layer.
The solid-state image sensor 200 of the fifth embodiment decreases
the shading independently from the effective F number of the camera
lens.
A concrete explanation is given below.
A solid-state image sensor 200 of the fifth embodiment is a charge-coupled
device-type (CCD-type) image sensor. In a valid cell area 210, wherein
valid cells are arranged to form a matrix, a light-receiving part
202 formed at the top of a semiconductor substrate 201, a flattening
layer 203, a color filter layer 204, a micro-lens stabilization
layer 206, a micro-lens 207 are formed as shown in FIG. 12. In this
case, the light-receiving part 202 and the micro-lens 207 are installed
on each photodetecting cell of the valid cell area 210.
In this case, the main components of the flattening layer 203 are
propylene glycol monomethyl ether acetate (PGMEA) and propylene
glycol monoethyl ether acetate (PGEEA).
In addition, for the color filter layer 204, pigments corresponding
to each color (red, green or blue) are dispersed in propylene glycol
monomethyl ether acetate (PGMEA) and propylene glycol monoethyl
ether acetate (PGEEA).
In addition, the main components of the micro-lens stabilization
layer 206 are methyl 3-metoxy propionic acid (MMP) and acrylic resin.
In addition, the main components of the micro-lens 207 are PGEEA,
ethyl lactate (EL) and phenolic resin.
In addition, as will be detailed later, the micro-lens 207 is made
to have different optical penetrating rates for the incident lights
(optical penetrating rate) for each micro-lens in each area from
a central part 210A to the peripheral part 210E of the valid cell
area 210 (FIG. 14).
In this solid-state image sensor 200, a vertical transfer electrode,
a horizontal transfer electrode and an amplifier for reading the
signal charge are installed, in the vicinity of the light-receiving
part 202 on the semiconductor substrate 201. In addition, in the
peripheral part of the solid-state image sensor 200, other circuits
such as correlated double sampling circuits (omitted from the Figure)
are installed on top of the same semiconductor substrate 201. In
addition, the planar structure of the solid-state image sensor 200
is almost identical to the solid-state image sensor 100 of the first
embodiment and FIG. 12 corresponds to the section along X-X' of
FIG. 1.
In addition, for each light-receiving part 202, a predetermined
color is selected and stains the color filter 202 of the solid-state
image sensor 200.
For example, in the case of bayer array, green (G), blue (B) and
red (R) pigments are implanted following the pattern indicated in
FIG. 13. In this case, the electric signal from the photodetecting
cell (light-receiving part 202) in which a green (G) filter is positioned
is used as the signal indicating brightness (G output voltage).
The micro-lens 207 is configured to have different optical penetrating
rates between the central part (block 210A) and the peripheral part
(block 210E) of the valid cell area 210. The optical penetrating
rate of the blocks 210B, 210C and 210D (FIG. 14) between the two
areas change in a stepwise manner. In other words, the optical penetrating
rate of each photodetecting cell is adjusted in a stepwise manner
according to its position inside the valid cell area 210.
By determining the optical penetrating rate for each of these blocks
210A, 210B, . . . , according to, for instance, the F number of
the camera lens actually used in a digital camera (refer to FIG.
18), the shading amount can be decreased for each of the block 210A,
210B, . . .
In addition, in FIG. 14, the number of blocks (number of divisions),
being 5, is small, but if the optical penetrating rate of the incident
lights is varied more finely by dividing into a larger number of
groups, the difference between the output from each area is smaller,
and a fine adjustment of the shading (shading correction) becomes
possible, allowing for the contrast steps of the image taken with
the solid-state image sensor 200 to be less visible.
In the following, a method for producing the solid-state image
sensor 200 of the above configuration will be explained using FIG.
15 through FIG. 17.
FIG. 15 is a cross-sectional view showing a production process
of the solid-state image sensor 200.
To produce the solid-state image sensor 200, first, at the top
of the semiconductor substrate 210, a diffusion zone 232 constituting
the light-receiving part 202, or other diffusion zones constituting,
for instance, transistors, and signal lines are formed. Then, on
the top side of these, the flattening layer 203, a color filter
layer 204, a micro-lens stabilization layer 206 are formed (FIG.
15(a)). In addition, diffusion zones and signal lines not related
to the present invention are omitted from the Figure.
Then, a resin having propylene glycol monoethyl ether acetate (PGEEA),
ethyl lactate (EL) and phenolic resin as the main components is
used to spin-coat the top side of the semiconductor substrate 201,
prior to patterning into the required shape with a publicly known
lithography method, to form a rectangular micro-lens base 237 as
shown in FIG. 5(b).
Then, a mask 250 is used, allowing the ultraviolet optical penetrating
rate for ultraviolet lights to differ in a stepwise manner for the
blocks 210A, 210B, . . . , 210E of the valid cell area 210, to expose
the micro-lens base 237 with ultraviolet light. This exposure to
ultraviolet light is performed to render the micro-lens base 237
forming the micro-lens 207 transparent (back exposure), and the
optical penetrating rate of the micro-lens base 237 is decreased
by decreasing the ultraviolet exposure dose (hereafter, simply called
"back exposure dose").
In addition, the optimum value for the back exposure dose depends
on conditions (for instance the output of the light source) on the
side of the actual exposing apparatus used (omitted from the drawing).
However, it is considered to be about 3 times the exposure dose
used to expose the resist used for the patterning of the micro-lens
base 237 (for instance, when using NSR150G4D (trademark) made by
Nikon, it is about 5 seconds).
Up to this stage of the process, in the central part 210A of the
valid cell area 210, a rectangular micro-lens base 237A having a
low optical penetrating rate is formed, in the peripheral part 210E,
a rectangular micro-lens base 237AE having a high optical penetrating
rate (more transparent) is formed, as shown in FIG. 15.
After the micro-lens bases 237A, . . . , 237E having different
optical penetrating rates between the central part 210A and the
peripheral part 210E have been formed, a heat treatment is performed
(140.degree. C. to 220.degree. C.) on the semiconductor substrate
201, using a hot plate. With this heat treatment, the micro-lens
bases 237A, . . . , 237E are re-flow soldered to become hemispheric.
As a result, in the solid-state image sensor 200, a micro-lens
207A having a low optical penetrating rate is formed in the central
part 210A, and a micro-lens 207E having a high optical penetrating
rate is formed in the peripheral part 210E.
Here, the mask 250 used when exposing the micro-lens base 237 to
ultraviolet light will be explained.
As mentioned above, the optical penetrating rate of the micro-lens
207 of the valid cell area 210 is defined according to the blocks
210A, 210B, . . . (FIG. 14). Therefore, the mask 250 is also divided
into a plurality of mask areas (5 in this case) 250A, 250B, . .
. , 250E having different ultraviolet optical penetrating rates
according to these blocks 210A, 210B, . . . as shown in FIG. 16.
In each mask area 250A, 250B, . . . , 250E, the ultraviolet optical
penetrating rate is adjusted by microscopic areas (a) through (e),
where two types of microscopic squares are arranged to form a mosaic,
one type of microscopic square transmitting the ultraviolet light
(white cutout in the drawing) and one type of microscopic square
not transmitting the ultraviolet light (hatched part), shown in
FIG. 17. This microscopic area (10 .mu.m.times.10 .mu.m size photodetecting
cell) is divided into microscopic squares of 5.times.5.
In other words, the microscopic area (a) is formed in the above-mentioned
mask area 250A, the microscopic area (b) is formed in the above-mentioned
mask area 250B, the microscopic area (c) is formed in the above-mentioned
mask area 250C, the microscopic area (d) is formed in the above-mentioned
mask area 250D, and the microscopic area (e) is formed in the above-mentioned
mask area 250E.
In this case, the ultraviolet optical penetrating rate is 0% in
the mask area 250A, the ultraviolet optical penetrating rate is
24% in the mask area 250B, the ultraviolet optical penetrating rate
is 52% in the mask area 250C, the ultraviolet optical penetrating
rate is 76% in the mask area 250D, and the ultraviolet optical penetrating
rate is 100% in the mask area 250E.
In addition, the larger the number of divisions into microscopic
areas as shown in FIGS. 17(a) to (e), the smaller the irregularities
due to back exposure using the mask 250, and therefore, the more
preferable.
In addition, to adjust the ultraviolet optical penetrating rate
in each mask area 250A, 250B, . . . , 250E, it is possible to have
different area ratio between the two types of microscopic square.
In addition, it is acceptable to adjust the optical penetrating
rate in each mask area 250A, 250B, . . . , 250E by pasting metal
films having different transparencies for each of the mask area
250A, 250B, . . . , 250E.
In the following, a single-lens reflex digital camera 300 loaded
with the solid-state image sensor 200 of the embodiment will be
explained.
As shown in FIG. 18, the single-lens reflex digital camera 300
consists of a camera body 310, a viewfinder 320 and an exchangeable
lens 330.
In this case, a photographic lens 331 and an iris 332 are built
into the exchangeable lens 330 which is flexibly fastened to the
camera body 310.
In addition, a quick-turn mirror 311, a focus detector 312 and
a shutter 313 are installed in the camera body 310. In addition,
the solid-state image sensor 200 is placed behind the shutter 313.
In addition, a finder mat 321, a pendaprism 322, an eye-piece 323,
a prism 324, an imaging lens 325 and a white balance sensor 326
are installed in the viewfinder 320.
In a single-lens reflex digital camera 300 configured as mentioned
above, the light L30 from the subject is incident to the camera
body 310, through the exchangeable lens 330.
In this case, before release, the quick-turn mirror 311 is in the
position shown by the broken lines in the drawing, and therefore,
a portion of the light L30 from the subject reflected by the quick-turn
mirror 311 is directed into the viewfinder 320 and is imaged onto
the finder mat 321. A portion of the subject image taken at this
point is directed via the pendaprism 322 to the eye-piece 323, another
portion becomes incident to the white balance sensor 326 via the
prism 324 and the imaging lens 325. The white balance sensor 326
detects the color temperature of the subject image. In addition,
at this point, a portion of the light L30 from the subject is reflected
by the auxiliary mirror 311A affixed to the quick-turn mirror 311
and imaged onto the focus detector 312.
After release, the quick-turn mirror 311 rotates clockwise in the
drawing (shown by solid lines in the Figure), and the light L30
from the subject becomes incident to the shutter 313.
Therefore, when taking a picture, first, after the convergence
of the focal point has been detected by the focus detector 312,
the shutter 313 opens. With the opening movement of the shutter
313, the light L30 from the subject becomes incident to the solid-state
image sensor 200 and is imaged onto its light-receiving side.
The solid-state image sensor 200 which received the light L30 from
the subject, generates the electric signal corresponding to the
light L30 from the subject, and at the same time performs various
image signal processing on the electric signal (see FIG. 37), such
as a white-balance correction based on signals from the white balance
sensor 326, and outputs the processed image signals (RGB data) to
the buffer memory (omitted from the figure).
In the above image signal processing, the shading correction is
performed according to actual shading amounts. Therefore, in the
digital camera loaded with the solid-state image sensor 200 of the
invention, the influence of the shading is already removed from
the signal charge outputted from the solid-state image sensor 200,
and it is possible to omit the shading correction step in the image
signal processing.
In addition, if enough shading correction could not be performed
in the solid-state image sensor 200 due to, for instance, a variation
in the F number, it is possible to execute a shading correction
at the image signal processing stage. In this case, the difference
or the color difference being small, the system load due to shading
correction is small.
In the following, the results of measurements to assess how much
the shading amount can be decreased by adjusting the optical penetrating
rate of the micro-lens 207 will be described in detailed.
In this case, using a mask having the minimum ultraviolet optical
penetrating rate (0%) and a mask having the maximum ultraviolet
optical penetrating rate (100%), a back exposure of micro-lens was
performed for each (0% back exposure and 100% back exposure) and
the extent of the change for the optical penetrating rate of the
micro-lens was measured. In addition, because it is not possible
to directly measure the optical penetrating rate of a single micro-lens,
the "G output voltage (equivalent to sensitivity etc.,)"
obtained for each case have been compared.
To prevent any influence due to other factors from appearing in
the results of both measurements, the single-lens reflex digital
camera loaded with a solid-state image sensor with 0% back exposure
and the single-lens reflex digital camera loaded with a solid-state
image sensor with 100% back exposure were placed against each other
and positioned to align the central axis of each camera lens, a
picture was taken under the same predetermined condition, and the
"G output voltage" of the event was measured. The camera
lens used was a NIKKOR 50 mm F1.4S (trademark), and the subject
was a uniformly bright picture without any pattern.
The "G output voltage" obtained in the above-mentioned
condition is shown in FIG. 19.
In the drawing, the broken line is the "G output voltage"
of the solid-state image sensor with 0% back exposure and the solid
line is the "G output voltage" of the solid-state image
sensor with 100% back exposure.
As is clear from the drawing, there is a difference of about 10%
between the "G output voltage". In addition, it can be
established that the F number dependency of the difference in the
"G output voltage" is extremely low.
By adjusting the ultraviolet optical penetrating rate of the mask
250 between 0% and 100% as described above, the "G output voltage"
of the solid-state image sensor can be varied by about 10%.
Therefore, when producing one solid-state image sensor 200, to
perform the back exposure on the micro-lens 207, by using a mask
250 in which the ultraviolet optical penetrating rate differs in
each of the mask areas 250A, 250B, . . . , 250E, the optical penetrating
rate of the blocks 210A, 210B, . . . , of the solid-state image
sensor 200 can be converted into "G output voltage" and
freely adjusted within a range of about 10%.
In addition, the division of the valid cell area 210 into blocks
210A, 210B, 210C, . . . is not limited to the pattern shown in FIG.
14, and, depending on the characteristics of the camera, or the
user's idea, for example, as shown in FIG. 20, blocks 210A, 210B,
210C . . . can be formed by concentric circles from the center of
the valid cell area 210, or as shown in FIG. 21, blocks 210A, 210B,
210C . . . can be formed as stripes in the vertical direction from
the center of the valid cell area 210. In this case, the longer
direction of the valid cell area 210 is divided into blocks 210A,
210B . . . since the converging characteristics in the direction
of the short axis (left-right direction of the FIG. 21) of the light-receiving
part (photoelectric converter) 202 of each photodetecting cell decreases
when approaching the peripheral part (left and right edges, in the
drawing) of the valid cell area 210 and leading to a prominent shading,
and by simply assigning block 210A, block 210B, block 210C . . .
as shown in FIG. 21, a sufficient shading correction effect can
be obtained.
In addition, without dividing the valid cell area 210 into blocks
210A, block 210B, block 210C, . . . , an equivalent effect can be
obtained by gradually changing the optical penetrating rate, for
one photodetecting cell (one light-receiving part 202) at a time,
or for a plurality of photodetecting cells at a time.
In fact, in the embodiment, for each mask area 250A, 250 B, . .
. of the mask 250, the ultraviolet optical penetrating rate is advantageously
set to 0%, 24%, 52%, 76% and 100%. The "G output voltage"
actually obtained in this case, when the ultraviolet optical penetrating
rate of the mask 250 falls into the 100% to 52% range, showed an
increase following the order 100%>76%>52%. However, when the
ultraviolet optical penetrating rates were 52%, 24%, 0%, no noticeable
difference could be identified. As shown previously, there is an
appropriate range for the ultraviolet exposure rate of the mask
250 for displaying an effect on the "G output voltage (corresponding
to the optical penetrating rate)". Therefore, by setting the
ultraviolet optical penetrating rate within the aforementioned appropriate
range, a desired optical penetrating rate can be easily realized
for a valid cell area at a desired position of the solid-state image
sensor.
(Sixth Embodiment)
In the following, the sixth embodiment according to the invention
will be explained using FIG. 22.
The sixth embodiment is the solid-state image sensor 200 wherein
the micro-lens 207 is formed by a so-called "etch back method".
Also in the sixth embodiment, first, at the top of a semiconductor
substrate 201 the diffusion zone 232 constituting the light-receiving
part 202 is formed, on top of these, the flattening layer 203, the
color filter layer 204, the micro-lens stabilization layer 206 are
formed (FIG. 22(a))
Then a uniform micro-lens layer 261 is formed by spin-coating (FIG.
22(b)).
A photoresist layer 262 then coats the top side of the micro-lens
layer 261 formed, to transfer the shape of the micro-lens 207, prior
to patterning the photoresist layer with the desired shape by the
photolithography technique (FIG. 22(c))
A heat treatment is performed on the photoresist layer 262 patterned
with the desired shape, and re-flow soldered to form a hemispheric
photoresist layer 263 (FIG. 22(d)).
Next, dry etching is performed on the hemispheric resist layer
263 which is then etched back to transfer the hemispheres from the
resist layer 263 onto the micro-lens layer 261. The result is the
formation of a hemispheric micro-lens 267 whose transparency is
not sufficient (FIG. 22(e)).
Finally, "back exposure" is performed on the hemispheric
micro-lens layer 267 to render it transparent. The conditions for
the "back exposure" are the same as in the fifth embodiment,
and the details are omitted here (FIG. 22(f)).
By performing the above "back exposure", a micro-lens
267 (267A, 267E) is obtained, wherein the optical penetrating rate
is different between the central part 210A and the peripheral part
210E (FIG. 22(g)).
In addition, in the sixth embodiment, the "back exposure"
is performed onto a hemispheric micro-lens 267 formed by the etch
back method. However, this does not constitute a limitation, and
for instance, immediately after coating the micro-lens layer 261,
"back exposure" can be performed on the micro-lens layer
261, and the optical penetrating rate (transparency) of the micro-lens
267 changed, at this stage. Then, in this case, a hemispheric micro-lens
267 is formed by the above-mentioned etchback method.
(Seventh Embodiment)
In the following, the seventh embodiment in relation to the method
of production of the solid-state image sensor 200 will be explained
using FIG. 23.
In the above-mentioned fifth and sixth embodiments, to obtain a
different optical penetrating rate for the microlens 207, the irradiation
amount during "back exposure" of the micro-lens 207 or
the micro-lens layer 261 has been differentiated, in the seventh
embodiment the temperature distribution is differentiated when performing
the heat treatment on the hemispheric micro-lens 207 to obtain a
micro-lens 207 in which the optical penetrating rate follows the
temperature distribution.
The embodiment exploits the characteristics of the micro-lens layer
261 having propylene glycol monoethyl ether acetate (PGEEA), ethyl
lactate (EL) and phenolic resin as the main components, whose optical
penetrating rate becomes lower when heated with high temperatures.
In the following, the heat treatment of hemispheric micro-lens
207 formed on the micro-lens stabilization layer 206 will be explained.
In addition, except for the fact that "back exposure"
is not performed, the other production processes are identical to
the fifth and sixth embodiments.
To heat the micro-lens 207 installed on the valid cell area 210
of the solid-state image sensor 200, with a different temperature
for each of the block 210A, block 210B, etc . . . , a hot plate
280 shown in FIG. 23(a) is used.
Protruding arc-shaped parts 281, 281, . . . are formed on the surface
of the hot plate 280. On the surface of the hot plate 280, a semiconductor
wafer W, in which a plurality of solid-state image sensor are formed,
is placed face-down.
In this case, in the central part 210A of each solid-state image
sensor 200, the micro-lens 207 is separated by a slight gap d41
and faces the apex parts 282, 282, . . . of each protruding part
281, 281, . . . of the hot plate 280, (FIG. 23(b)). At this point,
the central part 210A is heated approximately at the temperature
set for the hot plate 280 (for example 220.degree. C.).
On the other hand, in the peripheral part 210E of the solid-state
image sensor 200, the micro-lens 207 is separated by a predetermined
distance d42 (for instance 1 mm to 5 mm) and faces the bottom part
283, 283, . . . , of the hot plate 280 (FIG. 23(b)). The peripheral
part 210E is heated at a temperature lower than the temperature
set for the hot plate 280 (220.degree. C.) due to the predetermined
distance d42.
As described above, by setting the temperature of the hot plate
280 and the predetermined distance d42 to desired values, the optical
penetrating rate of the micro-lens 207 (207A) of the central part
210A and the optical penetrating rate of the micro-lens 207 (207E)
of the peripheral part 210E of the solid-state image sensor 200
can be adjusted to different values.
In this case, an alignment mark 209A as shown in FIG. 23(a) is
required to superimpose the hot plate 280 and the semiconductor
wafer W, and when the hot plate 280 and the semiconducto |