JPH0472664A - Solid-state image sensing device - Google Patents

Abstract

PURPOSE:To photoelectrically convert light selectively without forming a filter by a method wherein impurity diffusion regions having different depths are formed in such a way that incident light having different wavelength regions in a photodetection part 2 is photoelectrically converted selectively. CONSTITUTION:A photodetection part 2 is formed so as to be surrounded by a p-type well region 5. It is provided with the following: an n<+> type impurity diffusion region 11 for photoelectric conversion use; and a p<+> type impurity diffusion region 11 for light-absorption and recombination use. The depth of the p<+> type impurity diffusion region 12 from the surface of a substrate is designated as LP2, and the depth of the n<+> type impurity diffusion region 11 is designated as XN2. In this case, LP2<XN2; the n<+> type impurity diffusion region 11 is situated in the lower part of the p<+> type impurity diffusion region 12. Since the photodetection part 2 is formed of the impurity diffusion regions whose depths are different as described above, it is possible to read out a color image signal without forming a color filter.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a solid-state image sensor such as a CCD, and particularly relates to a solid-state image sensor that can obtain color output signals [Summary of the Invention] In a solid-state image sensor in which an output signal is obtained by charging and converting in a plurality of formed light receiving sections,
By forming the light receiving portions at different depths so that incident light in different wavelength ranges can be selectively photoelectrically converted, color signals can be read out without a color filter.

[Conventional technology]

Solid-state image sensors installed in color cameras usually separate light before it reaches the silicon substrate of the solid-state image sensor, and a prism, color filter, etc. is used to separate the light. ing.

For example, prisms are often used in cameras for business use and broadcasting stations. In this case, the camera becomes large and requires adjustment of the registration shield, etc.

Further, in cameras for general consumers, an optical filter is often formed on a single-chip solid-state image sensor. FIG. 7 is a sectional view of an example of a conventional solid-state image sensor. Here, to briefly explain the structure of the conventional example, a p-type well region 101 is formed on an n-type silicon substrate 100, and in this p-type well region 101, an n-type impurity that becomes a photodiode is formed. A diffusion region 102 and a buried channel layer 103 for charge transfer are formed. A polysilicon layer 104 as a transfer electrode and an aluminum layer 1105 as a light shielding film are formed on the buried channel layer 103 via an insulating film.

In the region of the n-type impurity diffusion region 102 that functions as a sensor, a p-type hole accumulation layer 1 is formed on the surface side of the n-type impurity diffusion region 102.
06 and an insulating film are formed, and a passivation film 107 and a thick planarization film 108 are further laminated thereon.

A filter 109 for selectively transmitting light is formed on the thick flattening film 108 while being covered with a protective film 110, and this filter 109 is, for example, R (red).
, G (green), and B (blue), and transmit light according to the filter characteristics described below.

FIG. 8 shows the filter characteristics of the primary color system, where the vertical axis is the light transmittance and the horizontal axis is the wavelength. In the figure, curve F is the blue filter characteristic, curve FG is the green filter characteristic,
Curve F shows the red filter characteristics. When these three types of filters are formed on an element,
Incident light is separated according to the characteristics of each filter to generate signal charges.

[Problem to be solved by the invention]

However, in a solid-state image sensor in which a filter 109 is formed on a chip, after the unevenness formed by the oxide film, polysilicon layer 104, aluminum film 105, etc. is flattened,
It is necessary to stack the filters 109, which causes negative effects such as stress. In addition, the filter 109 performs a process of forming a flattening film and a dyeing base film in order to form a dyed layer.
A dyeing process, a process of forming the protective film 110, etc. are required, which increases the number of steps. Furthermore, filter 109
Because it is made of organic material, fading occurs.

SUMMARY OF THE INVENTION In view of the above-mentioned technical problems, the present invention aims to provide a solid-state imaging device that selectively converts light without forming a filter.

[Means for Solving the Problems] In order to achieve the above-mentioned object, the present invention provides a method in which a plurality of light receiving sections for converting incident light into signal charges are formed on a semiconductor substrate, and the signal charges generated in the light receiving sections are In the solid-state imaging device that can obtain an output signal according to shall be.

[Effect]

In general, semiconductor materials such as silicon have the function of absorbing light and photoelectrically converting it in visible light, and its light absorption characteristics change depending on the wavelength. FIG. 2 is a diagram showing the light absorption coefficient in silicon, where the horizontal axis is the wavelength of light and the vertical axis is the absorption coefficient. As shown in Figure 2,
Absorption increases exponentially on the short wavelength side. Therefore,
Light on the short wavelength side is photoelectrically converted in a relatively shallow region of the substrate, and light on the long wavelength side is photoelectrically converted in a relatively deep region of the substrate.

[Example] Preferred embodiments of the present invention will be described with reference to the drawings.

This embodiment is an example of a CCD imager, in which light receiving portions of different depths are formed on a silicon substrate, and selective photoelectric conversion of incident light is performed.

First, its structure is as shown schematically in Figure 1.
A molded silicon substrate 1 has light receiving sections 2 at three different depths,
3.4 is formed.

The light receiving section 2 is formed surrounded by a P-type well region 5,
It has an n-type impurity diffusion region 11 for photoelectric conversion and a p-type impurity diffusion region 12 for light absorption and recombination. The depth of the p-type impurity diffusion region 12 from the substrate surface is I-pg, and the depth of the n-type impurity diffusion region 11 is XHz. Here, Lrz<XHz,
An n° type impurity diffusion region 11 is located below the p type impurity diffusion region w4 12. Further, the distance from the pn junction at the bottom of the n'-type impurity diffusion region 11 to the pn junction between the P-type well region 5 and the n-type silicon substrate l is assumed to be XN. This light receiving section 2 has the deepest impurity region depth among the light receiving sections having three different depths. Therefore, it is sensitive to light on the longer wavelength side of visible light, and can selectively generate signal charges for incident red light, for example.

The light receiving section 3 is surrounded by a P-type well region 5, and similarly receives light! an n-type impurity diffusion region 13 for conversion;
It has a p-type impurity diffusion region 14 for light absorption and recombination. The depth of the p-type impurity diffusion region 14 from the substrate surface is LP! The depth of the n° type impurity diffusion region 13 is XN3. Here, L and P3 are smaller than XNff, and the n'' type impurity diffusion region I3 is located below the p'' type impurity diffusion region 14. Further, the distance from the pn junction at the bottom of the n-type impurity diffusion region 13 to the pn junction between the p-type well region 5 and the n-type silicon substrate 1 is assumed to be Xo. This light receiving section 3 has an impurity region with a medium depth among the light receiving sections having three different depths.

The light receiving section 3 is sensitive to an intermediate wavelength range of visible light, and selectively generates signal charges for incident green light, for example.

Next, the light receiving section 4 is surrounded by a p-type well region 5 and an n-type impurity diffusion region 15 for converting 1X light.
and a relatively shallowly formed P' type impurity diffusion region 16 for light absorption and recombination. The depth of the p-type impurity diffusion region 16 from the substrate surface is LP4, and the depth of the n-type impurity diffusion region 15 is XN4. Here, LP4 is set to a small value, and the n' type impurity diffusion region 15 is located below the p' type impurity diffusion region 16.

Further, from the pn junction at the bottom of the n-type impurity diffusion region 15 to the p-n junction between the p-type well region 5 and the n-type silicon substrate l,
The distance to the n-junction is assumed to be XF. This light receiving section 4 is
It has an impurity region with the shallowest depth among the light receiving portions with three different depths. Therefore, the light receiving section 4 is sensitive to the short wavelength range of visible light, and can selectively generate signal charges for incident blue light, for example.

Here, to specifically explain the three light receiving sections 2 to 4, only the insulating film 6 is formed on these three light receiving sections 2 to 4, and there is no need to form a color filter. The CCD imager of this embodiment solves various problems associated with filter formation, such as stress, number of steps, and fading. In addition, to briefly explain the spectral sensitivity characteristics of the light receiving sections at different depths, the following formula %
Formula %) (where p=1.c=speed of light, λ: previous wavelength.

q: charge amount, h: blank constant, L, ;p-layer depth +
XN: depth of n layer, XP: depth of well)
It is expressed as

Figures 3 to 5 are diagrams showing the results of calculations using this 0th equation, where the horizontal axis represents the wavelength of light and the vertical axis represents the light! The flow rate is ■.

FIG. 3 is a characteristic diagram corresponding to the light receiving section 4, in which the depth XH4 of the n-type impurity diffusion region 15 is 1.0 μm.

These characteristics were obtained by setting the distance 11XP between the junctions to 4 μm and varying the depth L P4 of the p-type impurity diffusion region 16 between 0.2 and 1.0 μm. Each of the curves draws a curve in which the sensitivity in the short wavelength range becomes stronger. Therefore, by adjusting each parameter, it is possible to effectively extract the signal charge corresponding to the blue color.

Next, FIG. 4 is a characteristic diagram corresponding to the light receiving section 3, and n°
The depth XN3 of the type impurity diffusion region 13 is 1°0 μm, p
The depth Lo of the type impurity diffusion region 14 is 0.3 am.
These are the characteristics obtained by changing the distance HX P 3 between the junctions between 10 and 50 μm. Each curve has sensitivity in the medium wavelength region.

Finally, FIG. 5 is a characteristic diagram corresponding to the light receiving section 2, and n
The depth XSZ of the type impurity diffusion region 11 is set to '2. ．． 0μ
The depth L of the m, p' type impurity diffusion region 12 is 0.3μ.
m, and the characteristics were obtained by changing the distance XP2 between the junctions between 10 and 50 μm. Depth xN of n゛-type impurity diffusion region 11! As the depth increases, the peak of sensitivity softens toward longer wavelengths, making it possible to effectively extract signal charges corresponding to long wavelength light rays such as red.

The light receiving sections 2 to 4 having three different depths as described above are arranged on a plane, for example, as shown in FIG. 6. In Fig. 6, R represents a pixel having a light receiving section 2 for red (long wavelength), G represents a pixel having a light receiving section 3 for green (medium wavelength), and B represents a pixel having a light receiving section 3 for blue (short wavelength). Pixels (G) corresponding to the light receiving section 3 are arranged in each vertical column representing pixels having the light receiving section 4, and pixels (R) corresponding to the light receiving section 2 and the light receiving section 4 are arranged in each vertical column, two pixels each. Pixels (B) corresponding to are arranged alternately.

As shown in FIG. 1, signal charges are transferred to each of the light receiving sections 2 to 4 arranged in a plane via a vertical charge transfer section 7 in the vicinity thereof. These vertical charge transfer sections 7 are directly or indirectly electrically connected to the horizontal charge transfer section,
Color image signals are taken out from the horizontal charge transfer section.

The vertical charge transfer section 7 includes a buried channel layer 8 made of an n-type impurity diffusion region formed on the substrate surface;
It has a polysilicon layer 9 formed on the buried channel insulating layer 8 with an insulating film 6 interposed therebetween, and an aluminum film 10 covering the vertical charge transfer section 7. Polysilicon layer 9 is supplied with a necessary transfer signal and functions as a transfer electrode.

Aluminum film 10 is formed on an insulating film covering polysilicon layer 9, and functions as a light shielding film. The space between the buried channel layer 8 and each of the light receiving sections 2 to 4 is used as a readout gate, but the buried channel layer 8 may be formed deep to correspond to the deepened light receiving section, or a groove may be formed from the surface to form a readout gate. A buried channel layer 8 can also be formed at the bottom. Further, the impurity concentration in the channel region of the readout gate may be adjusted or the applied voltage during readout may be increased to effectively extract signal charges from the light receiving sections 2 to 4 at different depths (6). [Effects] As described above, the solid-state image sensor of the present invention has a light-receiving section in which impurity diffusion regions with different depths are formed, so that it is possible to read out color image signals without forming a color filter. Therefore, all the problems associated with the formation of color filters, such as an increase in the number of steps, fading, and film stress, can be solved.

FIG. 7 is a schematic diagram showing the arrangement; FIG. 7 is a cross-sectional view showing an example of a conventional solid-state image sensor; and FIG.

1...Silicon substrate 2 to 4... Light receiving section 5...well area 6...Insulating film Patent applicant: Sony Corporation Representative patent attorney Akira Koba (and 2 others)

[Brief explanation of drawings]

FIG. 1 is a schematic cross-sectional view of an example of the solid-state imaging device of the present invention, FIG. 2 is a characteristic diagram showing the light absorption characteristics of silicon, and FIGS. FIG. 3 shows the short wavelength type, FIG. 4 shows the medium wavelength type, and FIG. 5 shows the long wavelength type. FIG. 6 shows the solid-state imaging of the present invention. Planar photocurrent of a pixel as an example of an element! (uA/uW) Hikari 111LI Mata Nara 1 Da Ni Busy m-1) Masanaga (nm) Figure 4 return & (nm) Figure 5 Figure 6 Figure 7 Figure 8

Claims (1)

[Scope of Claims] A solid-state imaging device in which a plurality of light receiving sections for converting incident light into signal charges are formed on a semiconductor substrate, and an output signal is obtained in accordance with the signal charges generated in the light receiving sections, comprising: 1. A solid-state imaging device characterized in that impurity diffusion regions are formed at different depths so that incident light in different wavelength ranges is selectively photoelectrically converted.