JP4621719B2 - Back-illuminated image sensor - Google Patents

Description

  The present invention relates to a backside illuminating type imaging device that irradiates light from the back surface side of a semiconductor substrate, reads out charges generated in the semiconductor substrate in response to light from the front surface side of the semiconductor substrate, and performs imaging.

  Currently, light is irradiated from the back side of the semiconductor substrate, and charges generated in the semiconductor substrate in response to this light are accumulated in a charge accumulation region formed on the front side of the semiconductor substrate, and in accordance with the charges accumulated here A backside-illuminated image sensor is proposed in which a signal is output to the outside by a CCD, a CMOS circuit, or the like for imaging. In the following, a general imaging device that is currently popular with respect to the backside illumination type imaging device is referred to as a front side illumination type imaging device. Examples of the back-illuminated image sensor include those shown in the following documents.

JP 2003-338615 A JP 2006-261372 A JP 2006-173351 A JP 2006-80457 A

  By the way, in the backside illuminating type imaging device, the silicon layer of the semiconductor substrate needs to have a thickness of at least about 10 μm so that the absorption rate of the light of the red component among the light irradiated from the backside is not lowered. However, in recent years, when the pixel size is further miniaturized with the miniaturization of the image sensor, the area of the pixel is reduced, and the photoelectric conversion region of each pixel needs to be deepened. It is technically difficult to manufacture an image pickup device having such a pixel structure, and when light enters the image pickup region from an inclined direction with respect to the image pickup region, the incident light is incident on other adjacent pixels. Entering the photoelectric conversion unit causes color mixing. On the other hand, when the depth of the photoelectric conversion region is shallower than that of the silicon layer, the surface of the semiconductor substrate irradiated with light before the electrons generated near the back surface of the semiconductor substrate reach the charge storage region on the surface. Diffuse in the horizontal direction, which also causes color mixing.

  In Patent Document 2, the p-type region is formed between the pixels up to the vicinity of the back surface, but it is technically difficult if the pixels are fine and the silicon layer is about 10 μm.

  In Patent Documents 1 and 3, a p-type impurity concentration profile is formed in the silicon layer, and electrons generated in the photoelectric conversion unit are drifted to the surface by the gradient of the potential or by applying an electric field from the outside. This is intended to prevent color mixing, but it does not fundamentally solve the problem of light entering from the tilt direction.

  In Patent Document 4, a wall-like silicon film is formed by etching the back surface, and a reflective film is formed on the silicon film. However, since the reflective film is not formed on the silicon layer in the photoelectric conversion region, It is not possible to prevent color mixing due to the entrance of light irradiated from the direction.

  In addition, although a pixel isolation region can be formed at the boundary between pixels on the surface of the semiconductor substrate, the thickness of the silicon layer is about 4 to 5 μm due to technical restrictions such as ion implantation. If it is this thickness, since all the light irradiated from the back surface cannot be absorbed, high sensitivity cannot be realized. If the irradiated light is not absorbed, the light that has not been absorbed enters the charge transfer section of the solid-state image sensor formed on the surface side, the amplifier or floating diffusion associated with each pixel in the CMOS image sensor, etc. There is a concern that it will generate electrons and cause defects such as smear. Furthermore, even if a light-shielding film is formed on the back side, it is not possible to prevent color mixing due to incident light in the tilt direction.

  The present invention has been made in view of the above circumstances, and an object thereof is to provide a back-illuminated image sensor that can prevent color mixing caused by incident light in an inclined direction.

The above object of the present invention is achieved by the following configurations.
(1) A backside illuminating type imaging device that irradiates light from the back side of a semiconductor substrate, reads out charges generated in the semiconductor substrate according to the light from the front side of the semiconductor substrate, and performs imaging.
A photoelectric conversion region including a plurality of impurity diffusion layers stacked such that the potential potential decreases from the back surface side to the front surface side in the semiconductor substrate;
A charge storage region for storing the signal charge generated by the photoelectric conversion region,
A plurality of pixels having the photoelectric conversion region, the charge storage region, and a reading unit for reading out signal charges stored in the charge storage region are formed, and light is applied to the photoelectric conversion region between the adjacent pixels. An embedded member made of a reflective material and extending from the back surface side to the front surface side of the semiconductor substrate is formed ,
An insulating film formed so as to cover the entire embedded member, and a voltage applying unit that applies a voltage to the embedded member during driving and depletes an impurity diffusion layer in the vicinity of the embedded member. A back-illuminated image sensor.
( 2 ) The backside illuminating type imaging device according to ( 1 ), wherein the embedded member is made of tungsten.

  The backside illumination type image pickup device according to the present invention includes an embedded member extending from the back surface side to the front surface side of the semiconductor substrate, and the embedded member is made of a light reflective material. When light is incident on each pixel in the tilt direction from the back side, the embedded member is provided in the photoelectric conversion region between the pixels, so the incident light is reflected on the surface of the embedded member while the surface of the semiconductor substrate is reflected. Is converted into a signal charge by a photoelectric conversion action. For this reason, the embedding member avoids entering the photoelectric conversion region of a pixel at a position different from the incident pixel, and color mixing can be prevented.

  ADVANTAGE OF THE INVENTION According to this invention, the back side illumination type image pick-up element which can prevent generation | occurrence | production of the color mixture resulting from the incident light of an inclination direction can be provided.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. The backside illuminating type imaging device according to the present invention irradiates light from the back side of the semiconductor substrate, and reads out the charge generated in the semiconductor substrate according to the light from the front side of the semiconductor substrate to perform imaging. In this specification, the surface on the light incident side of the semiconductor substrate is referred to as the back surface, and the opposite surface is referred to as the front surface.

  FIG. 1 is a plan view showing the arrangement of pixels of a backside illumination type imaging device according to the present invention. FIG. 1 is a diagram illustrating a positional relationship between a light receiving unit and an electrode in a state where the semiconductor substrate of the backside illumination type imaging device 10 is viewed in plan. On the back surface of the semiconductor substrate, a plurality of light receiving region PD are partitioned in a matrix. A readout electrode for applying a voltage to the surface of the semiconductor substrate to read out signal charges generated by photoelectrically converting light incident from each light receiving region PD in a photoelectric conversion region described later and accumulated in the charge accumulation region 6, 8 are formed. In FIG. 1, the readout electrode 6 is indicated by a one-dot chain line, and the readout electrode 8 is indicated by a two-dot chain line. In the present embodiment, the readout electrode 6 is made of first polysilicon, and the readout electrode 8 is made of second polysilicon.

  For each light receiving area PD, a signal charge is generated by a photoelectric conversion area formed in the thickness direction with respect to the semiconductor substrate, and a signal corresponding to the signal charge constitutes one pixel of image data. For this reason, in the present specification, the light receiving region PD is also referred to as a pixel.

  FIG. 2 shows a cross-sectional view taken in the direction of the arrow AA in FIG. The back-illuminated image sensor 10 includes an n-type silicon layer (hereinafter referred to as an n-layer) 16 having a low impurity concentration, and a p-type silicon layer having a low impurity concentration (which is laminated on the back side of the n-layer 16). The semiconductor substrate includes a p-type silicon layer (hereinafter referred to as a p + layer) 12 having a high impurity concentration and is stacked on the back surface side of the p− layer 14.

  A pixel isolation region 28 made of a p-type impurity diffusion layer is formed on the surface side of the n − layer 16. A p-type well region 34 having a low impurity concentration is formed below the pixel isolation region 28.

  A p-type silicon layer 22 having a high impurity concentration is formed on the front surface side of each light receiving region PD, and a silicon layer containing n-type impurities (hereinafter also referred to as an n layer) is formed on the back surface side of the silicon layer 22. .) 24 is formed. The silicon layer 24 functions as a charge accumulation region for accumulating signal charges. The silicon layer 22 has a function of preventing charges caused by dark current generated on the surface of the semiconductor substrate from being accumulated in the n layer 24.

  On the back side of the p + layer 12 of the semiconductor substrate, an insulating layer 42 that is transparent to incident light such as silicon oxide or silicon nitride is formed. In order to prevent reflection of light on the back surface of the semiconductor substrate due to the difference in refractive index between the insulating layer 42 and the semiconductor substrate, incident light such as silicon nitride or a diamond structure carbon film is formed on the light irradiation side surface of the insulating layer 42. In contrast, a transparent high refractive index transparent layer 44 is formed. The high refractive index transparent layer 44 is preferably a layer having a refractive index exceeding n = 1.46, such as amorphous silicon nitride which can be formed at a low temperature of 400 ° C. or lower such as plasma CVD or photo-CVD.

  A color filter layer formed by arranging a plurality of color filters 46 in the horizontal direction is formed on the light incident side surface of the high refractive index transparent layer 44. The plurality of color filters 46 are a plurality of types of color filters that transmit light in different wavelength ranges, and are arranged based on a predetermined color pattern such as a Bayer array. For example, the color filter layer includes an R color filter that transmits light in a red wavelength range, a G color filter that transmits light in a green wavelength range, and a B color filter that transmits light in a blue wavelength range. It has an arranged configuration. The color filter 46 is formed on the back side of each charge storage region corresponding to a plurality of charge storage regions, and one color filter 46 is provided corresponding to each charge storage region.

  A light shielding member 48 for preventing color mixing is formed between adjacent color filters 46. The light blocking member 48 may be any member having a function of not transmitting light, and a metal having a low visible light transmittance such as tungsten (W), molybdenum (Mo), and aluminum, or a black filter can be used.

  On the light incident side of each color filter 46, a microlens 52 is formed that collects incident light and guides it to the back side of the semiconductor substrate. The shape of the micro lens 52 is determined so that incident light refracted on the surface thereof passes through the approximate center in the horizontal direction (left and right direction in FIG. 2) of each light receiving region PD without being blocked by the light blocking member 48. ing.

  A read gate 32 is formed adjacent to one side of the silicon layer 22 in the horizontal direction on the surface side of the semiconductor substrate. In addition, a charge transfer channel 26 made of an n-type impurity diffusion layer is formed so as to be adjacent to the silicon layer 22 via a read gate 32.

  A gate insulating film 2 made of a silicon oxide film, an ONO film or the like is formed on the surface of the semiconductor substrate, and a voltage is supplied to the charge transfer channel 26 through the gate insulating film 2 to control the charge transfer operation. A read electrode 6 and a read electrode 8 (see FIG. 1) are formed. An insulating film 14 such as silicon oxide is formed around the readout electrode 6 and the readout electrode 8. A transparent insulating layer 1 is formed on the gate insulating film 2 and the insulating film 14.

  The backside illumination type imaging device 10 of the present embodiment has a CCD type charge readout configuration by the charge transfer channel 26 and the readout electrodes 6 and 8.

When a driving voltage is applied, the semiconductor substrate photoelectrically converts light incident from the light receiving region PD and moves signal charges to the n layer 24 on the front surface side, so that the potential from the back surface side to the front surface side is increased. In this configuration, the p + layer 12, the p− layer 14, the n− layer 16, and the n layer 24 are sequentially stacked so that the potential is lowered. FIG. 3 is a diagram showing a potential profile of the semiconductor substrate of the backside illumination type imaging device 10 of the present embodiment. The concentration of the plurality of impurity diffusion layers is not particularly limited. For example, the p + layer 22 is 1 × 10 ¹⁹ / cm ² , the p well region 34 is 1 × 10 ¹⁶ / cm ² , and the p− layer 14 is from the surface side. 1 × 10 ¹⁷ / cm ² , the n− layer 16 is about 5 × 10 ¹⁶ / cm ² , and the p + layer 22 is about 1 × 10 ¹⁹ / cm ² . The plurality of impurity diffusion layers function as a photoelectric conversion region.

  In the backside illumination type imaging device 10 of the present embodiment, the thickness of the semiconductor substrate is preferably about 10 μm or more so as to be able to absorb all visible light according to the difference in the light absorption coefficient for each wavelength. .

  In the back-illuminated image sensor 10 of the present embodiment, during driving, light incident on one microlens 52 is incident on the color filter 46 above the microlens 52, and light transmitted through the color filter 46 is positioned at that position. Is incident on the corresponding n− layer 16. Then, signal charges are accumulated in the n layer 24 while being photoelectrically converted by the photoelectric conversion region of the semiconductor substrate. The signal charge accumulated in the n layer 24 is read and transferred to the charge transfer channel 26, converted into a voltage signal by an output amplifier (not shown), and output to the outside.

  In the backside illumination type imaging device 10 of the present embodiment, an embedded member 18 made of a light reflective material such as tungsten is formed inside a semiconductor substrate so as to partition adjacent pixels. FIG. 4 is a diagram illustrating the position of the embedded member with respect to the light receiving region in a state where the semiconductor substrate is viewed in plan. The embedding member 18 is formed in a lattice shape so as to surround the periphery of the light receiving portion region PD.

  As shown in FIG. 2, the embedded member 18 has a shape extending in parallel to the thickness direction of the semiconductor substrate from below the p well region 34 to the surface of the p + layer 12. An insulating film 19 such as silicon oxide is formed between the embedded member 18 and the surrounding silicon layer. The length of the embedded member 18 in the thickness direction of the semiconductor substrate is preferably 5 μm or more.

  The p + layer 12 and the embedded member 18 are electrically connected to voltage application means (not shown) through wirings, respectively, and can apply a negative voltage during driving. Further, the pixel separation region 28 is connected to GND. At the time of driving, a negative voltage is applied to the p + layer 12 and the embedding member 18 by voltage applying means, and the p + layer 12, the p− layer 14 and the n− layer 16 on the back surface side of the semiconductor substrate, and the embedding are performed. A depletion layer is formed in the silicon layer around the member 18. By doing so, the depletion layer can be expanded from the front surface side and the back surface side of the semiconductor substrate, so that there is no n− region not depleted in the n− layer 16. Electrons generated in the photoelectric conversion region are accumulated in the n layer 24, and holes are collected in the p + layer 12 on the back surface side and discharged to the outside.

  The back-illuminated image pickup device 10 has the embedded member 18 provided in the photoelectric conversion region between the pixels when light is incident on each pixel in the tilt direction from the back surface side of the semiconductor substrate. The incident light is reflected and converted to signal charges by the photoelectric conversion action while being guided to the surface side of the semiconductor substrate. For this reason, the embedded member 18 avoids entering the photoelectric conversion region of a pixel at a position different from the incident pixel, and color mixing can be prevented.

  The embedded member is not limited to tungsten, and other refractory metals generally used in semiconductors (for example, nickel, cobalt, platinum, molybdenum) and compounds of refractory metals and silicon can be used. is there. Moreover, it is also possible to have a laminated structure in which a laminated film of the above materials or a side wall is made of the above material and the inside is made of polysilicon.

  As a method for forming the embedded member 18, a method can be used in which an epitaxially grown layer is formed by epitaxially growing an impurity diffusion layer, a groove is formed in the epitaxially grown layer, the embedded member 18 is embedded, and further epitaxially grown. There is also a method of performing epitaxial growth after forming the pattern of the embedded member 18 before epitaxial growth. A technique for embedding by forming a pattern such as an embedding member and then performing epitaxial growth is known, and examples thereof include those disclosed in Japanese Patent Application Laid-Open Nos. 5-114563 and 5-123117.

It is a top view which shows arrangement | positioning of the pixel of the backside illumination type imaging device concerning this invention. FIG. 2 shows a cross-sectional view seen in the direction of arrow AA in FIG. 1. 3 is a diagram illustrating a potential profile of a semiconductor substrate of the backside illumination type imaging device 10. FIG. It is a figure which shows the position of the embedding member with respect to a light-receiving part area | region in the state which planarly viewed the semiconductor substrate.

Explanation of symbols

10 Back-illuminated image sensor 18 Embedded member PD Light receiving area

Claims (2)

A backside-illuminated imaging device that irradiates light from the back side of a semiconductor substrate, reads out charges generated in the semiconductor substrate in response to the light from the front side of the semiconductor substrate, and performs imaging,
A photoelectric conversion region including a plurality of impurity diffusion layers stacked such that the potential potential decreases from the back surface side to the front surface side in the semiconductor substrate;
A charge storage region for storing the signal charge generated by the photoelectric conversion region,
A plurality of pixels having the photoelectric conversion region, the charge storage region, and a reading unit for reading out signal charges stored in the charge storage region are formed, and light is applied to the photoelectric conversion region between the adjacent pixels. An embedded member made of a reflective material and extending from the back surface side to the front surface side of the semiconductor substrate is formed ,
An insulating film formed so as to cover the entire embedded member, and a voltage applying unit that applies a voltage to the embedded member during driving and depletes an impurity diffusion layer in the vicinity of the embedded member. A back-illuminated image sensor. The backside illumination type image pickup device according to claim 1 , wherein the embedded member is made of tungsten.

Images