JPH0724304B2 - Solid-state imaging device and manufacturing method - Google Patents

Description

TECHNICAL FIELD The present invention relates to a solid-state imaging device including a charge sweep device, and more particularly to a solid-state imaging device having an improved transfer electrode structure.

[Conventional example] The charge sweep device is a charge coupled device (CC
It is a kind of D), and a solid-state image pickup device composed of such a charge sweep device has a structure shown in FIG. 11. That is, the light receiving area A has a device configuration similar to a so-called interline transfer system, and the output side of the charge transfer line is provided with the storage section B for temporarily storing the signal charges for one column in the horizontal direction. Further, a charge transfer device (referred to as horizontal CCD) C for serially transferring the signal charges transferred from the storage section B in the horizontal direction is formed. Further 12th
The structure of the light receiving region A will be described with reference to the drawings. By forming a plurality of n-type impurity layers in a matrix in a P-type impurity layer (P well) formed on the front surface side of a semiconductor substrate, a plurality of photo-resistors can be formed. Diode Pd _{11 to} Pd _1n , Pd _{21 to} Pd _2n , Pd
_{31 to} Pd _3n are formed. These photodiodes (hereinafter referred to as Pd _ij , where i = 1 to m, j = 1 to n)
A plurality of charge transfer channels L ₁ , L ₂ , L ₃ ... (Hereinafter, referred to as L _j , where j = 1 to n) for transferring signal charges in the charge transfer direction are formed between Except for the photodiode Pd _ij and the charge transfer channel L _j , the shaded area surrounded by a dotted line in the figure is a channel stop.

On the upper surface of the charge transfer channel L _j, a plurality of transfer electrodes G ₁ , G ₂ , G ₃ , ... (Hereinafter referred to as G _i , where i = ₁ to _m) are formed of a polysilicon layer extending in the horizontal direction. Are installed side by side,
The signal charge output to the specific charge transfer channel L _j via the transfer gates T _g11 , T _g12 , ... T _gmn is transferred to the transfer electrode.
Drive signal φ applied to G ₁ , G ₂ , G ₃ , ...
Read according to _{1 to} φ ₄ . Charge sweep,
The difference between the device (hereinafter referred to as CSD) and the normal interline charge-coupled device (hereinafter referred to as ILT CCD) is that the width of each charge transfer line L _j of CSD is extremely narrower than that of the ILT CCD. It is designed (for example, 1.5 to 2 μm), and the aperture ratio of the photodiode Pd _ij is improved accordingly. In addition, when the width of the charge transfer line L _j is narrow in this way, the amount of signal charge that can be transferred in one charge transfer is smaller than the amount of signal charge generated in the photodiode Pd _ij , and thus charge transfer is performed a plurality of times. The signal is read out by repeating. That is, in FIG. 11, first, the signal charges of the photodiode group of the n-th column closest to the accumulation unit B are accumulated in each accumulation element of the accumulation unit B by repeatedly transferring several times, and all the signal charges of the n-th column are accumulated. When the storage of is completed, the signal is read out serially via the horizontal CCD C. Next, the signal charges of the photodiode group in the (n-1) th column are similarly transferred to the storage section B by a plurality of transfer operations and read out via the horizontal CCD C. This operation is further repeated, and the processing is completed by reading out the signal charges from the photodiode group on the first column.

As described above, although the structure is similar to that of a normal ILT CCD, the timing of the drive signals φ _{1 to} φ ₄ applied to the transfer electrode G _j is set so that the above transfer operation can be performed, and an excellent effect is obtained. ing.

[Problems to be Solved by the Invention] However, the solid-state imaging device including such a conventional charge sweep device has the following problems. Since it is necessary to repeat transfer multiple times for each column in order to read all the signal charges generated in the photodiode,
The transfer frequency must be extremely high. For example,
When performing reading according to the well-known NTSC method, it is necessary to increase the transfer frequency per horizontal scanning period (1H) by at least the number of times the transfer is repeated. Conventionally, in order to enable this high-speed operation, polysilicon is used. the metal wiring made of metal layer opposed to the normal transfer electrode group G _j, which is formed of a layer stacked by juxtaposing connected at appropriate places, the resistivity of the transfer electrode groups G _j of polysilicon layer To reduce
Attempts have been made to achieve high-speed operation by reducing the time constant of the transfer electrode group. That is, as shown in FIG. 13 showing a cross-sectional structure taken along line X ₁ -X _{1 in} FIG. 12, the second polysilicon layer is formed on the upper surface of the even-numbered transfer electrode G _{2 formed} of the first polysilicon layer. Of the odd-numbered transfer electrodes G ₃ are stacked, and further, metal wiring of the same pattern as the even-numbered transfer electrodes G ₂
Al ₂ is formed of the first metal layer, and the metal wiring Al ₃ having the same pattern as the odd-numbered transfer electrodes G ₃ is formed of the second metal layer at the uppermost portion. And the even numbered transfer electrodes
The G ₂ and the metal wiring Al ₂ are connected in parallel at an appropriate position, and the odd-numbered transfer electrodes G ₃ and the metal wiring Al ₃ are connected in parallel in the same manner to reduce the resistivity of the wiring.

However, as is apparent from the structure shown in FIG. 12, it is difficult in area to connect the metal wiring and the transfer electrode in parallel in the light receiving region where the photodiode Pd _ij is formed. Could not reduce the resistivity. Attempts have also been made to form contacts on the charge transfer lines L _j to make connections, but during the formation of the contacts, impurities intrude into the charge transfer lines L _j , and noise such as mixing in the signal charges may occur. It was not an effective means because it caused problems.

[Means for Solving Problems] The present invention has been made in view of such problems.
It is an object of the present invention to provide a solid-state imaging device including a charge sweep device, which has a structure capable of reducing the resistivity of a transfer electrode and performing a high-speed charge transfer operation.

In order to achieve this object, the present invention provides a plurality of light receiving elements such as photodiodes, a charge transfer channel extending between the light receiving elements in a predetermined charge transfer direction, and a charge transfer channel of the charge transfer channel. A plurality of transfer electrodes, which are arranged in parallel on the upper surface so as not to overlap each other along the charge transfer direction and to which a predetermined clock signal is applied, and isolation for electrically separating the respective light receiving elements and the charge transfer channels from each other are provided. In a solid-state imaging device including a charge sweep device, a first metal wiring group wired along the charge transfer channel on an upper surface of the transfer electrode, and the first metal wiring group crossing each other and A second metal wiring group electrically insulated from each other and wired along the upper surface of the isolation is provided. A pair of end portions formed so as to cover the upper surfaces of the pair of charge transfer channels located on both sides of the child, and an intermediate portion connecting each end portion through the upper surface of the isolation, It is characterized in that it is connected to a specific metal wiring of the metal wiring group through a predetermined contact provided in the section.

Further, in order to form such a solid-state image pickup device, a first polysilicon layer and a nitride film are sequentially formed on a top surface of a charge transfer channel formed in a semiconductor substrate for forming the solid-state image pickup device via a silicon oxide film. A first manufacturing step of stacking and further forming a second polysilicon layer having a shape matching the shape of the transfer electrode to be formed on the upper surface of the nitride film, and a poly oxide film on the surface of the second polysilicon layer. And a second manufacturing step of exposing the upper end surface of the polyoxide film and covering the other surface with a photoresist layer, the upper end part of the polyoxide film and the second polysilicon. A fourth manufacturing step in which the layer is removed by etching and the side end of the polyoxide film is left on the surface of the nitride film, and a portion covered by the side end of the polyoxide film by anisotropic etching. Manufacturing process for removing the nitride film except for And a sixth manufacturing step of forming a silicon oxide film on the surface of the first polysilicon layer excluding the portion covered with the remaining nitride film, and removing the remaining nitride film and polyoxide film by etching Then, a seventh manufacturing step of exposing a part of the first polysilicon layer covered with the nitride film and removing an exposed part of the first polysilicon layer by anisotropic etching are performed. The transfer electrode group is formed through the respective manufacturing steps of the eighth manufacturing step for separating one polysilicon layer into a plurality of transfer electrodes.

Embodiment An embodiment of the present invention will be described below with reference to the drawings.

First, the structure will be described with reference to FIG. 1, which shows a part of the light receiving region corresponding to FIG. 12. By forming a plurality of n-type impurity layers in a matrix in the P well, A plurality of photodiodes Pd _ij (i = 1 to m, j =
1 to n) are formed, and a plurality of charge transfer channels L _j (j = 1 to n) for vertically transferring the signal charges are formed between the respective photodiodes Pd _ij . Then, the hatched portion surrounded by the dotted line in the figure is isolation, and the signal charges generated in the respective photodiodes Pd _ij are transferred to the transfer gates Tg _ij (i = 1 to m, j =
1 to n) and output to a predetermined charge transfer channel L _j .

Next, the difference from the conventional example shown in FIG. 12 will be described. A transfer electrode group composed of a polysilicon layer formed in a concave shape when viewed from the upper surface side is provided. The upper surface of these transfer electrodes, second extending along the charge transfer channel _{L 1, L 2, L 3} ... L n 1
A plurality of metal wires a ₁ made of metal layer, a _2, ... a _2n and lateral are formed along the isolation upper surface of these metal wires a _1, a _2, ..., a second crossing the a _n a plurality of metal wires b _1, b ₂ made of metal layers, ... b _m are laminated. Since these transfer electrodes are regularly arranged in the same pattern with respect to the photodiode array, the transfer electrodes g ₁ , g ₂ , g ₃ , g ₄ will be described as a representative. Both ends of the transfer electrode g ₁ are arranged on the upper surfaces of the charge transfer channels L ₁ and L ₂ , and the intermediate portion connecting these both ends is the photodiode P 1.
It is arranged on the upper surface of the isolation that separates d ₁₂ and Pd ₂₂ . Both ends of the transfer electrode g ₂ are charge transfer electrode channels
An intermediate portion located on the upper surfaces of L ₂ and L ₃ and connecting both ends of these is arranged on the upper surface of the isolation that separates the photodiodes Pd ₁₃ and Pd ₂₃ . Further, the transfer electrode g ₃ is similar to the transfer electrode g ₁ between the photodiodes Pd ₂₂ and Pd ₃₂ ,
The transfer electrode g ₄ is similar to the transfer electrode g _{2 in} that the photodiode P
It is located between d ₂₃ and Pd ₃₃ . _One ends of the transfer electrodes g _{1 to} g ₄ on the charge transfer channel L ₂ have an extremely narrow interval ΔW (about
0.2 μm) electrically separated. And
Transfer electrodes g ₁ is metal wire b ₁ via a contact C _1, the transfer electrodes g ₂ are metal wires a ₂ via a contact C _2, transfer electrodes
The g ₃ is connected to the metal wiring b ₂ via the contact C ₃ , and the transfer electrode g ₄ is connected to the metal wiring a ₃ via the contact C ₄ . The clock signals φ _{1 to} φ ₄ are applied to the other transfer electrodes by the same connection, and so-called potential wells are formed in the respective charge transfer channels L _j to transfer the signal charges.

FIG. ₂ shows a cross section taken along line X ₂ -X _{2 in} FIG. 1, in which the surface of the charge transfer channel L _j formed in the P well of the semiconductor substrate is covered with a silicon oxide film (SiO ₂ ) 1. , A polysilicon layer 2 constituting the transfer electrodes g ₁ , g ₂ , ... Is laminated on the surface thereof. Further, the upper surface thereof is covered with a silicon oxide film (SiO ₂ ) 3 and the predetermined polysilicon layer 2 and the metal wirings a ₁ , a ₃ ... Are partially connected via a penetrating portion provided for the contact C ₄ . Has been done. Further, the upper surface is further covered with a layer 4 of silicon oxide film.
And the other transfer electrode formed of the polysilicon layer 2 is connected via a contact C ₃ . Here, since the interval ΔW between the transfer electrodes adjacent to each other is extremely narrow as about 0.2 μm as described above, a potential barrier is not generated between the transfer electrodes, and the signal charges can be transferred stably.

Next, a manufacturing process of the solid-state imaging device having such a structure will be described with reference to FIGS. 3 to 10. Incidentally, these figures schematically show the sectional structure corresponding to FIG. 2, and these figures (b) are plan views showing the vicinity of the portion corresponding to FIG. 2 from the top.

In the first step shown in FIGS. 3A and 3B, after the P well, the n-type impurity layer for the photodiode, the isolation and the like are formed on the surface side of the semiconductor substrate, the silicon oxide film 1 and the first layer are formed. Polysilicon layer 2 and nitride film (Si ₃ N ₄ ) of
5 is laminated. These layers 2 and 5 are formed so as to cover almost the entire light receiving region to be formed, and the first polysilicon layer 2 is processed so that the transfer electrode g _i (i =
1 to m) are formed.

Further, a second polysilicon layer 6 having substantially the same shape as the concave transfer electrode is formed on the surface of the nitride film 5. Here, each of the polysilicon layers 6 has transfer electrodes located at discrete positions, for example,
Formed according to the shape of g ₁ , g ₃ , g ₅ …, and the other transfer electrode
The portions corresponding to g ₂ , g ₄ , g ₆ , ... Are opened as a gap as shown by the dotted line in FIG. 3 (b). Further, a particularly important point in the present invention is that the transfer electrodes on the charge transfer channels are arranged in parallel at an extremely narrow interval ΔW, so that the hatched portion in FIG. The dimensional accuracy of the portion where the polysilicon layer 6 and the gap 7 are adjacent to each other is designed in advance sufficiently in advance. The other parts are designed in appropriate shapes.

In the second step shown in FIG. 4, a polyoxide film 7 is deposited on the surface of the second polysilicon layer 6 by vapor phase epitaxy (CVD) or the like, and a photoresist layer 8 is further laminated on the upper surface. Here, as an example, the silicon oxide film 1 is designed to have a thickness of 250 A, the first polysilicon layer 2 to 5000 A, the nitride film 5 to 1500 A, the second polysilicon layer 6 to 5000 A, and the polysilicon layer 7 to a thickness of about 0.2 μm. .

In the third step shown in FIGS. 5A and 5B, a region of the photoresist layer 8 having a width approximately equal to the width of each second polysilicon layer 6 is removed by etching, and the poly oxide film 7 is exposed. The end face is exposed.

In the fourth step shown in FIGS. 6A and 6B, the upper end of the poly oxide film 7 is removed by etching, and the second polysilicon layer 6 is also removed by etching to expose the nitride film 5. As a result, the side edge of the poly oxide film 7 is about 0.2 μm.
Is left on the upper surface of the nitride film 5.

In the fifth step shown in FIG. 7, the nitride film 5 is anisotropically etched using a gas composed of CF ₄ + 10% oxygen.
Remove the exposed part of the. As a result, the nitride film 5 is left under the polyoxide film 7 with a width of about 0.2 μm.

In the sixth step shown in FIG. 8, the first step is performed by wet oxidation or the like.
A silicon oxide film (SiO ₂ ) 9 is formed on the surface of the polysilicon layer 2. At this time, the silicon oxide film 9 is not formed in the portion covered with the nitride film 5.

In the seventh step shown in FIG. 9, the nitride film 5 is formed by etching.
And the poly oxide film 7 is removed. At this time, the surface of the silicon oxide film 9 is slightly etched, but remains as shown in the drawing, and the portion of the first polysilicon layer 2 not covered with the silicon oxide film 9 is exposed.

In the eighth step shown in FIG. 10, the exposed portion of the first polysilicon layer 2 is removed by anisotropic etching to obtain about 0.2
A groove 10 having a width ΔW of μm is formed. By forming the groove portion 10, mutual transfer electrodes g ₁ , g ₂ , g ₃ , ... Are mutually separated. Then, after the silicon oxide film 9 is removed, unnecessary portions of the first polysilicon layer 2 remaining on the upper surface of the photodiode or the like are removed by etching to remove the first portion.
A transfer electrode group having a concave shape as shown in the figure is formed.

In the next final step, after covering the upper surface of the transfer electrode group with a silicon oxide film, metal wirings and contacts as shown in FIG. 2 are formed, and clock signals φ _{1 to} φ ₄ are applied to the respective predetermined transfer electrodes. So that it can be applied.

As described above, according to this embodiment, since the transfer electrodes can be formed by separating the single polysilicon layer 2 with the extremely narrow interval ΔW, the conventional overlapping portion of the transfer electrodes becomes unnecessary, It is possible to improve yield and achieve high integration.

Further, the structure is such that the contact connecting the metal wiring group and the transfer electrode group can be provided on the isolation region, and for example, the contact is not provided on the charge transfer channel, so that stable high-speed transfer operation is possible. Can be performed and the yield can be improved.

In this embodiment, in the first step (see FIG. 3), first, the second polysilicon layer 6 is provided so as to correspond to the transfer electrodes g ₁ , g ₃ , g ₅ ,. The other transfer electrode g ₂ , g ₄ ,
Even if the second polysilicon layer 6 is provided so as to correspond to g ₆ , ..., A similar transfer electrode group can be formed.

EFFECTS OF THE INVENTION As described above, according to the present invention, it is possible to form a plurality of transfer electrodes by separating the polysilicon layers arranged on the same plane at an extremely narrow interval. In addition, since there is no overlapping portion as in the conventional case, the problem of breakdown voltage is solved and the yield can be improved. Furthermore, since the transfer electrode and the metal wiring can be connected by the contact on the upper surface of the isolation, the resistivity of the transfer electrode can be reduced and high-speed charge transfer can be performed without adversely affecting the charge transfer line.

[Brief description of drawings]

FIG. 1 is a plan view showing the structure of one embodiment of a solid-state image pickup device according to the present invention from the upper surface side, FIG. 2 is a sectional view taken along the line X ₂ -X _{2 of} FIG. 1, and FIGS. FIG. 11 is a cross-sectional view corresponding to the cross-sectional view in FIG. 2 for explaining the manufacturing process of the solid-state imaging device shown in FIG. FIG. 1 is a block diagram schematically showing the structure of a conventional solid-state imaging device having a CSD structure, FIG. 12 is a plan view showing the structure of a conventional solid-state imaging device from the top side, and FIG.
12 is a sectional view taken along the line X ₁ -X ₁ of FIG. Pd ₁₁ , Pd ₁₂ , Pd ₁₃ , ...: Photodiodes g ₁ , g ₂ , g ₃ , g ₄ , ...: Transfer electrodes L ₁ , L ₂ , L ₃ , ...: Charge transfer channels a ₁ , a ₂ ,, a ₃ ,…, b ₁ , b ₂ , b ₃ , b ₄ ,…: Metal wiring C ₁ , C ₂ , C ₃ , C ₄ ,…: Contact ₁ , ₃ , 9: Silicon oxide film 2: First polysilicon Layer 4: Wiring layer 5: Nitride film 6: Second polysilicon layer 7: Polyoxide film 8: Photoresist layer 10: Groove part

Claims (2)

[Claims] 1. A plurality of light receiving elements, a charge transfer channel extending between the light receiving elements in a predetermined charge transfer direction, and an upper surface of the charge transfer channel along the charge transfer direction. Solid-state imaging consisting of a plurality of transfer electrodes that are arranged side by side so as not to overlap with each other and to which a predetermined clock signal is applied, and a charge sweep device having isolation that electrically separates the respective light receiving elements and the charge transfer channels In the device, a first metal wiring group wired along the charge transfer channel on an upper surface of the transfer electrode, and a first metal wiring group that intersects with the first metal wiring group and is electrically insulated from each other. A second metal wiring group wired along the upper surface of each of the transfer electrodes, the transfer electrodes of the pair of charge transfer channels located on both sides of each of the light receiving elements. A pair of end portions formed so as to cover the surface and an intermediate portion connecting the respective end portions through the upper surface of the isolation, and the first contact is provided through a predetermined contact provided in each intermediate portion. A solid-state imaging device characterized in that the solid-state imaging device is connected to specific metal wirings of the first and second metal wiring groups. 2. A plurality of light receiving elements, a charge transfer channel extending between the light receiving elements in a predetermined charge transfer direction, and an upper surface of the charge transfer channel along the charge transfer direction. A plurality of transfer electrodes that are arranged side by side so as not to overlap with each other and to which a predetermined clock signal is applied, an isolation that electrically separates the respective light receiving elements and the charge transfer channels, and the respective transfer electrodes are A pair of end portions formed so as to cover the upper surfaces of the pair of charge transfer channels located on both sides of the light receiving element, and intermediate portions connecting the respective end portions to the upper surface of the isolation, and respective intermediate portions. A charge sweep connected to a specific metal wire of the above-mentioned metal wire group via a predetermined contact provided on
In a solid-state imaging device including a device, a first polysilicon layer and a nitride film are sequentially laminated on a top surface of a charge transfer channel formed in a semiconductor substrate for forming the solid-state imaging device with a silicon oxide film interposed therebetween, Furthermore,
A first manufacturing step of forming a second polysilicon layer having a shape matching the shape of the transfer electrode to be formed on the upper surface of the nitride film, and a second manufacturing step of forming a polyoxide film on the surface of the second polysilicon layer. And a third manufacturing step in which the upper end surface of the polyoxide film is exposed and the other surface is covered with a photoresist layer, and the upper end portion of the polyoxide film and the second polysilicon layer are etched. A fourth manufacturing step of removing the side edge of the poly oxide film and leaving the side edge of the poly oxide film on the surface of the nitride film, and removing the nitride film except the portion covered by the side edge of the poly oxide film by anisotropic etching. Fifth manufacturing step of removing, a sixth manufacturing step of forming a silicon oxide film on the surface of the first polysilicon layer excluding the portion covered with the remaining nitride film, and the remaining nitride film and The poly oxide film is removed by etching, and the nitride film A seventh manufacturing step for exposing a part of the covered first polysilicon layer, and a plurality of transfer of the first polysilicon layer by removing the exposed part of the first polysilicon layer by anisotropic etching. A method for manufacturing a solid-state imaging device, characterized by being formed by an eighth manufacturing step of separating into electrodes.