KR20010061058A - Method for manufacturing image sensor - Google Patents

Abstract

PURPOSE: A method for manufacturing an image sensor is provided to improve quantum efficiency and sensitivity, by additionally depositing a dielectric layer of a low absorption rate between a photodiode and an initial interlayer dielectric to improve illuminance of incident light. CONSTITUTION: A photodiode region and a peripheral device are formed on a substrate(41). The first insulating layer in which an index of refraction, Nx is a square root of N0N and a thickness, Tx is (2n-1)lambda/4Nx(n=0,1,2,3,...), is formed on the photodiode region. The first insulating layer is eliminated to be left only on the photodiode region. The second insulating layer is formed on the resultant structure.

Description

Manufacturing Method of Image Sensor {METHOD FOR MANUFACTURING IMAGE SENSOR}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image sensor, and more particularly, to a CMOS image sensor which improves illuminance of incident light.

In general, a CCD (Charged Coupled Device) image sensor and a CMOS image sensor are used. A conventional CCD image sensor absorbs light captured by an external subject image and collects and accumulates photocharges. And a charge transport unit for transporting charges generated by the photoelectric conversion and charge accumulation unit, and a signal converter for outputting the photocharges transported from the charge transport unit as an electric signal.

The photoelectric conversion and charge storage unit mainly uses a photodiode (PD). The photodiode (PD) forms a potential well by using a PN junction and potential of the charge generated by light. It is an element that accumulates in a well. The charge generated in the charge accumulator is trapped in the potential well of the photodiode (PD). By moving the potential well, the charge can be transported to the required place. The signal converter also generates a voltage from the transferred charges. On the other hand, after the signal detection is finished, it is necessary to discharge the charge of the current potential well for the charge waiting for the next turn, for this purpose, the charge is removed by removing the barrier of the potential well of the signal conversion unit, which is called a reset.

As such, unlike the CMOS image sensor, the CCD image sensor detects signals by charge coupling rather than switching by a transistor. In addition, the photodiode (PD), which corresponds to a unit pixel and performs a photosensitive region, accumulates for a predetermined time instead of immediately extracting a photocurrent, and thus extracts the signal voltage so that the signal voltage can be increased by the accumulated time. While good, there is an advantage that can reduce noise, the driving method is complicated because the photoelectric charge must be transported continuously, high voltage of about 8-10V and high power of more than 1W is consumed.

On the other hand, CMOS image sensors are inferior in electro-optical characteristics to CCD image sensors, but CMOS image sensors are superior to CCD image sensors in terms of low power consumption and integration.

Among these electro-optic characteristics, external quantum efficiency and sensitivity are characteristics that determine the quality of the image sensor, and in particular, image frame and image quality that can be captured per unit time ( Image quality) is very important characteristic.

FIG. 1 is an equivalent circuit diagram illustrating a unit pixel of a conventional CMOS image sensor, and includes one photodiode PD and four NMOS transistors Tx, Rx, Sx, and Dx, and the four NMOS transistors Tx, Rx, Sx, and Dx are transfer transistors (Tx) for transporting the concentrated photocharges from the photodiode (PD) to the floating node (X) .The potentials of the nodes are set to the desired values and the charges are discharged. Reset transistor (Rx) to reset the node, drive transistor (Dx) that acts as a source follower buffer amplifier (Dx), and a select transistor that allows addressing as a switching role. (Sx).

2 is a view schematically showing a unit pixel manufacturing method of a CMOS image sensor according to the prior art.

As shown in FIG. 2, a unit pixel of a conventional CMOS image sensor includes a P well 12 formed locally on a P type semiconductor substrate 11 and then a field insulating film 13 for separation between unit pixels. ). Subsequently, PN junction layers 17a and 17b are formed inside the P-type semiconductor substrate 11 to form a photodiode PD, and the charge generated from the photodiode PD is transferred to the semiconductor substrate 11. A floating bonding layer 18a for storing is formed.

Subsequently, a gate electrode of a transfer transistor Tx for transferring the photocharges from the photodiode PD to the floating junction layer 18a is formed, and a reset transistor Rx for resetting the floating junction layer 18a. To form a gate electrode. Subsequently, a drive transistor Dx having a gate electrode electrically connected to the floating junction layer 18a and a select transistor Sx for receiving a signal for addressing as its gate electrode are formed. At this time, the reset transistor Rx and the drive transistor Dx have a common junction layer 18b, and the common junction layer 18b is formed at the boundary between the semiconductor substrate 11 and the P-type well 12. do. Next, a gate electrode of the select transistor Sx is formed on one side of the field insulating layer 13, and the impurity junction layer 19 of the lightly doped drain (LDD) structure is formed in the drive transistor Dx and the select transistor Sx. Is formed. The gate electrode is formed of a gate oxide film 14, polysilicon 15 and tungsten silicide 16, and sidewall spacers 20 are formed on the sidewalls of the gate electrode.

First interlayer insulating films 21 and 22 and second interlayer insulating films 23, 24 and 25 for light transmission are formed on the four transistors Tx, Rx, Dx and Sx. Titanium / aluminum / titanium nitrides (26, 27, < RTI ID = 0.0 > 26) < / RTI > The first and second metal wirings formed of the laminated film of 28) are formed. In addition, a device protection film including an oxide film 29 and a nitride film 30 is formed on the second metal wiring to protect the device from moisture or scratches, and the unit is formed on the device protection film to realize a color image. A color filter array (CFA) process of a color filter 31 composed of red, green, blue, yellow, magenta, and cyan is performed on the pixel array. Subsequently, the flat layer 32 is formed on the array of the color filters 31, and then the microlens 33 is formed on the flat layer 32 to face the array of the color filters 31.

After all of the above processes are completed, the first and second interlayer insulating films 21, 22, 23, 24, and 25 for light transmission, the device protection layers 29 and 30, and the color filter 31 are disposed on the photodiode, which is a light sensing region. And only the microlens 33 is positioned.

As described above, in the related art, in order to improve external quantum efficiency and light sensitivity, the microlens 33 is formed for each unit pixel to focus incident light, but the incident light focused on the microlens 33 is a microlens 33. Since the incident angles of the center and the periphery of the) are different, even if the same number of photon fluxes are focused, the illuminance of the photons incident on the center and the periphery of the unit pixel is changed.

That is, if the illuminance of photons incident to the center of the unit pixel (θ = θ _c = 0) is i _c , the illuminance i _E of photons incident at the incident angle θ = θ _E around the unit pixel is i _c × cos ⁴ Since it is proportional to θ, it is reduced by cos ⁴ θ. Therefore, the angle of incidence of photons incident on the unit pixel must be reduced.

In order to improve the characteristics of the image sensor, the semiconductor substrate 11 has a refractive index (N _S ) of 4.72 to 3.48 in the same visible light region, and after forming the photodiode PD of a unit pixel, the surface of the photodiode PD The insulating layers (PMDs) 21 and 22 deposited for the first time typically have a refractive index (N ₀ ) of 1.48 to 1.46 in the visible region (450 to 635 nm). The microlens 33 is 7.5 mu m wide and 2.6 mu m high.

The initial angle of incidence of incident light incident on the unit pixel under such a condition is 0 (degree), but the incident angle of incident light at the moment of hitting the microlens 33 is determined by the medium difference of air from the air to the microlens 33 and of the microlens 33. It is changed by curvature. That is, when the incident light having an incident angle of 0 (degree) in the first air enters the center of the microlens 33, the incident angle becomes 0 (degree) and the incident angle when the incident light enters the edge of the microlens 33 is 33.5. (degree) In this way, incident light propagates through the photodiode PD of the unit pixel, and the incident angle of the incident light incident on the final photodiode PD determines the incident light intensity of the photodiode PD.

Therefore, the incident angle of the incident incident from the image sensor to the photodiode PD becomes 37 (degree). Therefore, the relative illuminance around the center that can be accepted by the photodiode (PD) of the unit pixel is cos ⁴ (37) = 0.488. In other words, when the illuminance at the center of the photodiode PD is 100%, the illuminance when the incident light incident at the edge of the microlens 33 is incident on the photodiode PD is 48,8%, and the photodiode PD At the periphery of), the amount of light of about 51.2% is lost.

The present invention has been made to solve the above problems, a method of manufacturing an image sensor suitable for improving the quantum efficiency and light sensitivity by improving the intensity of incident light by further depositing a dielectric layer having a low absorption rate between the photodiode and the first interlayer insulating film. The purpose is to provide.

1 is an equivalent circuit diagram illustrating a unit pixel of a general CMOS image sensor;

2 is a cross-sectional view of a manufacturing process showing a method of manufacturing a unit pixel of an image sensor of the related art;

3A to 3F are cross-sectional views illustrating a manufacturing method of a unit pixel manufacturing method of an image sensor according to an exemplary embodiment of the present invention;

* Description of the symbols for the main parts of the drawings *

41 semiconductor substrate 47 photodiode

51: buffer dielectric layer 53: PMD

57: first metal wiring 60: interlayer insulating film

63 element protection film 64 color filter

66: microlens

According to an aspect of the present invention, there is provided a method of manufacturing an image sensor, the method including forming a photodiode region and a peripheral device on a substrate, and having a refractive index on the photodiode region. And thickness And a third step of forming a first insulating layer, a third step of removing the first insulating layer so that only the upper portion of the photodiode region remains, and a fourth step of forming a second insulating layer on the resultant product. It is characterized by.

Hereinafter, the most preferred embodiments of the present invention will be described with reference to the accompanying drawings so that those skilled in the art can easily implement the technical idea of the present invention. .

3A through 3F are cross-sectional views illustrating a method of manufacturing a unit pixel of an image sensor according to an exemplary embodiment of the present invention, wherein the unit pixel of the image sensor of the present invention has the same structure as a conventional unit pixel.

As shown in FIG. 3A, the silicon substrate 41 is subjected to a mask process for forming a well, implanted boron ions, and then thermally processed to form a P-type well 42 by Lateral diffusion. Form. The P type well 42 is a region in which the select transistor Sx and the drive transistor Dx are formed.

Subsequently, after forming the separator field insulating layer 43 between the unit pixels, the gate oxide layer 44 is formed and the polysilicon 45 and the tungsten silicide layer are formed on the gate oxide layer 44 to form gate electrodes of the transistors in the unit pixel. 46 is formed continuously, and a plurality of gate electrodes are formed through a mask and an etching process. The plurality of gate electrodes are used as gate electrodes of a transfer transistor (Tx), a reset transistor (Rx), a drive transistor (Dx), and a select transistor (Sx), respectively.

Subsequently, a mask process for forming a photodiode (PD) is performed on the resultant, and N ^- ion implantation and P ⁰ ion implantation for forming an N ^- doped region and a P ⁰ doped region are performed through the gate electrode of the transfer transistor (Tx). Self-alignment is performed on one side. In this case, the P-type substrate 41, the N ⁻ doped region 47a and the P ⁰ doped region 47b form a photodiode 47 having a PNP structure.

Subsequently, a mask pattern for implanting N-type LDD (Lightly Doped Drain) ions of the drive transistor Dx and the select transistor Sx is formed, and boron ions are implanted to form the LDD region 49a.

In order to form source / drain regions of the transistors, an oxide spacer (LP-Oxide) is deposited using low pressure chemical vapor deposition (LPCVD) and etched to form an entire surface, and then etched to a sidewall of the gate electrode. ), And then a mask pattern for high concentration N-type impurity ions is formed to perform N-type impurity ion implantation to form impurity junction layers 49b.

As shown in FIG. 3B, a buffer dielectric layer 51 is formed on the resultant product, particularly on the photodiode PD.

At this time, the buffer dielectric layer 51 contains a dielectric layer having a refractive index of 2.25 to 2.65 in the visible light region of 450 to 635 nm, a dielectric layer having a thickness of 425 to 2113 s, or a material having no absorption of visible light, for example, silicon. One of a dielectric layer including a silicon-rich oxide, an oxynitride, or a silicon nitride. The buffer dielectric layer 51 increases the illuminance of incident light incident on the photodiode PD.

Subsequently, a photosensitive film is coated on the buffer dielectric layer 51 and patterned by exposure and development. Then, the buffer dielectric layer 51 except the photodiode region is wet or dry-etched using the patterned photosensitive film as a mask.

As shown in FIG. 3C, after depositing a first TEOS (TetraEthylOrthoSilicate) film 52a by low pressure chemical vapor deposition (LPCVD) on the resultant, an atmospheric pressure chemical vapor deposition (Atmospheric Pressure) method is applied on the first TEOS film 52a. The BPSG film 52b is deposited using CVD; APCVD. Then, heat treatment is performed to reflow the BPSG film 52b. Here, the first TEOS film 52a and the BPSG film 52b are referred to as a PMD (Premetal Dielectric Layer) 53.

Subsequently, an isotropic etching is performed on the PMD 53 using a buffered oxide film etching solution (BOE) through a mask process, and anisotropic etching is performed using a plasma etching process to form contact holes on the surfaces of the bonding layers 49b. Form. Subsequently, titanium / aluminum / titanium nitride (Ti / Al / TiN) (54, 55, 56) is deposited on the entire surface including the contact hole, and then the first metal wiring 57 is formed by performing a mask process and an etching process. Form.

As shown in FIG. 3D, a second TEOS oxide layer 58 is deposited on the first metal interconnection using plasma enhanced CVD (PECVD), followed by a spin on glass (SOG) oxide layer 59. Is formed by coating twice and planarized by heat treatment and full surface etching using plasma. Subsequently, an interlayer insulating layer 60 is deposited on the planarized SOG oxide layer 59 by using plasma chemical vapor deposition (PECVD).

As shown in FIG. 3E, the interlayer insulating layer 60 is isotropically etched with a buffered oxide etching solution (BOE) through a mask process, and anisotropic etching using plasma is performed to form a contact hole for the second metal wiring. . Subsequently, titanium / aluminum / titanium nitrides 54a, 55a, and 56a are deposited on the resultant including the contact hole in order, and then a second metal wiring 61 is formed by performing a mask process and an etching process. Next, an oxide layer 62a and a nitride layer 62b are deposited on the second metal interconnection 61 as a device protection layer 63 by using plasma chemical vapor deposition (PECVD).

As shown in FIG. 3F, the nitride layer 62b and the oxide layer 62a are selectively removed to expose a predetermined surface of the second metal wiring 61 for pad open. Then, a color material (Dyed photoresistor) (hereinafter referred to as a color photosensitive film) is coated on the pad open area and the unit pixel, and the color filter 64 is formed by a developing process. Subsequently, the microlens flattening layer 65 is formed to uniformly form the microlens on the color filter 64, and then the microlens 66 is formed on the microlens flattening layer 65.

As described above, in the present invention, a dielectric layer having a refractive index of 2.25 to 2.65 and a thickness of 425 to 2113 에 can be formed on the photodiode PD before forming the PMD to increase the illuminance of incident light incident on the photodiode. That is, the illuminance of incident light depends on the refractive index and the thickness of the dielectric layer.

The dielectric layer refractive index (Nx),

In addition, the thickness (Tx) of the dielectric layer,

to be. In this case, N ₀ is the refractive index (= 1.48 to 1.46) of the PMD formed in a subsequent step, and N _S is the refractive index (= 4.72 to 3.48) of the substrate 41.

In this way, the deposition of the low-absorption buffer dielectric layer may improve the relative illuminance of the peripheral portion relative to the center of the photodiode in the visible ray region of the incident light by about 50%. For example, the buffer dielectric layer deposited under the conditions of λ = 635 nm, Nx = 2.261, and Tx = 700 Å can significantly improve the illuminance of incident light with a relative illuminance of 71.9% relative to the center.

Although the technical idea of the present invention has been described in detail according to the above preferred embodiment, it should be noted that the above-described embodiment is for the purpose of description and not of limitation. In addition, those skilled in the art will understand that various embodiments are possible within the scope of the technical idea of the present invention.

The present invention described above applies a conventional CMOS and CCD image sensor manufacturing process and forms a buffer dielectric layer on top of the photodiode to increase the relative illuminance of the periphery of the photodiode with respect to incident light, thereby increasing the quantum efficiency of the image sensor. There is an effect that can improve the light sensitivity.

Claims (7)

In the image sensor manufacturing method, Forming a photodiode region and a peripheral device on the substrate; Refractive index on the photodiode region And thickness A second step of forming a first insulating layer; A third step of removing the first insulating layer so that only the upper portion of the photodiode region remains; And A fourth step of forming a second insulating layer on the resultant product Image sensor manufacturing method characterized in that it comprises a. The method of claim 1, After the fourth step, And a fifth step of forming a metal wiring and a device protection film on the resultant, And a sixth step of forming a color filter and a microlens on the resultant. The method of claim 1, In the second step, And the first insulating layer has a thickness of 425-2113Å. The method of claim 1, In the second step, And the first insulating layer has a refractive index of 2.25 to 2.65. The method of claim 1, In the second step, The method of claim 1, wherein the anti-reflective layer is used as the first insulating layer. The method of claim 1, In the second step, The N ₀ is the refractive index of the second insulating layer, the N _S is a refractive index of the substrate, the manufacturing method of the image sensor. The method of claim 1, In the second step, The first insulating layer is a method of manufacturing an image sensor, characterized in that using any one of a silicon-containing oxide film, oxynitride or silicon nitride.