JP2000031466A - Manufacture of charge transfer device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce the manufacturing cost of a charge transfer device by decreasing the number of manufacturing steps to provide transfer gate electrodes of a single-layer structure and preventing generation of a potential pocket under between adjacent transfer gate electrodes, which causes signal charges which are not transferred. SOLUTION: A masking film 21 of a pattern, having openings 22 at the boundary between respective adjacent transfer gate electrodes 26t and 26s, is formed on a polycrystalline silicon layer 26 to be the transfer gate electrodes 26t and 26s on a gate insulating film 27, and a material insulating the polycrystalline silicon layer is implanted using the film 21 as a mask. Thereafter, an isolation insulating film 30 is formed through annealing between the respective adjacent transfer gate electrodes 26t and 26s, into which the material of the polycrystalline silicon layer 26 was ion implanted.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of manufacturing a charge transfer device, and more particularly to a method of manufacturing a charge transfer device having a single-layered transfer gate electrode.

[0002]

2. Description of the Related Art FIGS. 2A to 2C show a conventional example of a charge transfer device, in which FIG. 2A is a sectional view, FIG. 2B is a potential diagram for explaining a transfer operation, and FIG. FIG. 4 is a pulse diagram showing a driving clock pulse for transfer. The method of manufacturing the charge transfer device of this example is a method of manufacturing a charge transfer device of a two-phase transfer system, and is often used as a horizontal register of a CCD solid-state imaging device.

In the drawings, 1 is a P-type semiconductor substrate, 2 is an N-type channel layer formed on the surface of the semiconductor substrate 1,
Reference numerals 3, 3,... Denote N ⁻ -type transfer portions disposed at predetermined intervals along the transfer direction on the surface portion of the channel layer 2, and selectively deposit a P-type impurity in the N-type channel layer 2. It is formed by doping. Reference numerals 4, 4,... Denote areas located on the transfer destination side of each transfer section 3 on the surface of the channel layer 2 and constitute a storage section. However, the storage portions 4, 4,... Are portions of the N-type channel layer 2 where the P-type impurities are not doped, ie, the N-type channel layer 2
You can say that.

Reference numeral 5 denotes a gate insulating film formed on the semiconductor surface, 6s, 6t, 6s, 6t,..., Transfer gate electrodes made of polycrystalline silicon.
.. are composed of the first polysilicon layer and are located above the storage units 4, 4,. The transfer gate electrodes 6t, 6t,... Are formed of a second polysilicon layer, and are located on the transfer portions 3, 3,.
Incidentally, 7 is a gate insulating film, and 8 is an interlayer insulating film.

The principle of transfer will be described by focusing on, for example, φ1 of the two-phase drive clock pulses φ1 and φ2 shown in FIG. 2C and a pair of transfer gate electrodes 6t and 6s to which the pulse is applied. It is.

That is, the pulse φ1 is applied to the transfer gate electrode 6t on one transfer unit 3 and the transfer gate electrode 6s on the storage unit 4 on the transfer destination side, which forms a pair with the transfer gate electrode 6t. Since the transfer portion 3 has its N-type impurity concentration lower than that of the storage portion 4 by selectively ion-implanting a P-type impurity, its potential is lower than that of the storage portion 4 even when the same voltage is received. For example, a drop of about 2 to 3 V occurs, and the storage section 4 has a deeper potential. Therefore, the transfer charge always passes through the transfer unit 3 and the storage unit 4
Trying to get together.

At time t = t1, that is, when one of the driving clock pulses φ1 is at a high level voltage (5 in this example).
V), and when the other drive pulse φ2 is at a low voltage (0 V in this example), the area below the transfer gate electrodes 6s, 6s,... Made of the first polysilicon layer, that is, the transfer section 4 , 4,..., Signal charges are accumulated.

Thereafter, as time elapses, t = t2 and t = t3, the drive clock pulses φ1 and φ2 are inverted, and as shown in FIG. 2B, signal charges are transferred rightward in FIG. Is done.

Incidentally, the transfer gate electrodes 6s, 6t, 6
One of the reasons why s, 6t,... have a multilayer structure (in this example, a two-layer structure, but also a three-layer structure) as shown in FIG. This is to make the adjacent transfer gate electrodes 6s and 6t overlap each other so that a potential pocket is not formed between the insulating films. That is, the transfer gate electrodes 6s, 6t, 6s, 6t,.
Is formed, the interval between the adjacent transfer gate electrodes 6s and 6t becomes large as shown in FIG. Then, a place where the effect of the electric field by the transfer electrode is reduced is formed between the adjacent transfer gate electrodes 6s and 6t, and FIG.
Potential pockets as shown in (B) are generated between adjacent transfer gate electrodes. Since this potential pocket can serve as a reservoir for signal charges, it causes transfer residue.

Therefore, conventionally, the transfer gate electrodes have a two-layer structure so that adjacent transfer gate electrodes overlap each other. FIG. 4 (A),
(B) is an explanatory view of the reason why the transfer gate electrode has a two-layer structure, (A) is a sectional view, and (B) is a potential diagram.

That is, in the case of the two-layer structure, the adjacent transfer gate electrodes can overlap each other, and the reduction of the electric field effect on the channel between the adjacent transfer gate electrodes is reduced. As shown in (1), the potential pocket is also small. Therefore, the transfer residue of the signal charge in the potential pocket is reduced.

FIGS. 5A to 5G are sectional views showing an example of a method of forming a transfer gate electrode having a multilayer structure in the order of steps.

(A) As shown in FIG. 5A, after forming a gate insulating film 7 on the surface of the channel 2, a first polycrystalline silicon layer 6a is formed on the gate insulating film 7. .

(B) Next, as shown in FIG. 5B, a resist film 9 is selectively formed on the polycrystalline silicon layer 6a. This resist film 9 is formed so as to cover the storage section not shown in FIG.

(C) Next, using the resist film 9 as a mask, the polycrystalline silicon layer 6a is etched by, for example, RIE, thereby transferring the transfer gate electrodes 6s, 6s,.
Is formed, and then the polycrystalline silicon layer 6a is removed.
FIG. 5C shows a state after removing the polycrystalline silicon layer 6a. The exposed surface of the gate insulating film 7 is slightly etched by RIE.

(D) Next, as shown in FIG. 5D, the surface of the polycrystalline silicon layer 6a is heated and oxidized,
An interlayer insulating film 8 is formed. At this time, since the silicon oxide film also grows on the gate insulating film 7, the gate insulating film 7 thinned in the previous step (C) returns to a substantially predetermined thickness.

(E) Next, as shown in FIG. 5E, a second polycrystalline silicon layer 6b is formed by CVD.

(F) Next, as shown in FIG. 5 (F), a resist film 10 is selectively formed on the polycrystalline silicon layer 6b. More specifically, the resist film 10 is used as a transfer gate electrode 6 for driving a transfer section (not shown in FIG. 5).
It is formed so as to cover the portion where t is to be formed.

(G) Next, using the resist film 10 as a mask, the polycrystalline silicon layer 6b is etched by, for example, RIE, thereby transferring the transfer gate electrodes 6t, 6t,.
Is formed, and then the resist film 10 is removed. FIG.
(G) shows a state after the resist film 10 is removed. Thus, a transfer gate electrode having a multilayer structure is completed.

[0020]

According to the conventional method of manufacturing a charge transfer device, a polycrystalline silicon layer is formed and patterned by etching using a resist film selectively formed as a mask. The process of forming the transfer gate electrode must be repeated a plurality of times, and a step of forming an interlayer insulating film by heat oxidation on the surface of the transfer gate electrode made of polycrystalline silicon is also required. However, there is a problem that the manufacturing cost is high.

In addition, as the number of manufacturing steps increases, the yield necessarily decreases, which also increases manufacturing costs.

Further, according to the above-described conventional method of manufacturing a charge transfer device, the transfer gate electrode 6s formed of the first polycrystalline silicon layer and the transfer gate electrode formed of the second polycrystalline silicon layer are formed. 6 t below the thickness T of the gate insulating film 7
1. There was also a problem that a difference occurred in T2. this is,
This is an unfavorable problem because the electric field effect caused by the same gate voltage differs between the storage unit and the transfer unit. The problem that arises is that
After the gate insulating film 7 is formed, the thickness of the exposed portion is reduced by the etching in the step (C), and thereafter, the film is heated and oxidized in the next step (D) to grow a silicon oxide film on the exposed portion. , The thickness in that part becomes thicker and RIE
This is because it is very difficult to make the thickness of the thinner portion equal to the thickness of the portion increased by the thermal oxidation.

In order to solve this problem, a search has been made, and the L length of the transfer gate electrode has become extremely short with the rapid miniaturization and the increase in the number of pixels of the CCD type solid-state imaging device in recent years.
It has been found that since the silicon oxide film thickness between the transfer gate electrodes is very thin, the effect of reducing the remaining transfer by overlapping the adjacent transfer gate electrodes is considerably reduced. Specifically, in the charge transfer device as shown in FIG.
When the thickness of the inter-electrode silicon oxide film was 100 nm, it was found that there was almost no difference in the effect of preventing transfer residue even if the overlap structure was not used.

Therefore, the inventor of the present application can separate adjacent transfer gate electrodes with a thin insulating film in order to hardly cause the problem of transfer residue while having a single-layer structure, that is, a structure that does not overlap. He realized that this was the case, and sought an extremely thin insulating film between adjacent transfer gate electrodes, and as a result of repeated thoughts, came to the present invention.

That is, the present invention reduces the number of manufacturing steps by making the transfer gate electrode of the charge transfer device a single layer structure and preventing a potential pocket which causes transfer of signal charges from being left between adjacent transfer gate electrodes. And
Accordingly, it is an object to reduce the manufacturing cost of the charge transfer device.

[0026]

According to a first aspect of the present invention, there is provided a method for manufacturing a charge transfer device, wherein an opening is formed in a portion corresponding to a boundary between adjacent transfer gate electrodes on a polycrystalline silicon layer serving as a transfer gate electrode on a gate insulating film. A mask film having a pattern having the following pattern, ion-implanting a substance for converting the polycrystalline silicon layer into an insulator using the mask film as a mask, and then annealing the polycrystalline silicon layer to ion-implant the substance. An isolation insulating film is formed in a portion corresponding to a region between adjacent transfer gate electrodes.

According to the method of manufacturing a charge transfer device of the first aspect, the material is turned into a polycrystalline silicon insulator by using a mask film having an opening on a portion of the polycrystalline silicon layer corresponding to a boundary between adjacent transfer gate electrodes as a mask. Is ion-implanted and then annealed, so that an isolation insulating film can be formed in a portion corresponding to between each adjacent transfer gate electrode,
Insulation separation between the electrodes can be performed by the separation insulating film.

By forming the opening of the mask film narrow using the current photolithography technology, the insulating film for separating the adjacent transfer gate electrodes can be made thin and narrow. Therefore, it is possible to make the transfer gate electrode one-layer structure and to make almost no potential pockets occur, thereby significantly reducing the number of manufacturing steps of the charge transfer device.

[0029]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention basically relates to a method for manufacturing a charge transfer device, in which a boundary between adjacent transfer gate electrodes is formed on a polycrystalline silicon layer serving as a transfer gate electrode on a gate insulating film. A mask film having a pattern having an opening in a portion is formed, a material for converting the polycrystalline silicon layer into an insulator is ion-implanted using the mask film as a mask, and thereafter, the material of the polycrystalline silicon layer is ionized by annealing. A separation insulating film is formed at a portion between the implanted adjacent transfer gate electrodes. Oxygen ions and nitrogen ions are typical examples of the above-described ion implanted materials.

Examples of the charge transfer device include a horizontal transfer register of a CCD solid-state image pickup device. Other than the above, for example, a vertical transfer register, a transfer register of a linear sensor type CCD solid-state image pickup device, and a transfer register not for a solid-state image pickup device. For example, the present invention can be applied to various charge transfer devices such as a delay charge transfer device.

[0031]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail with reference to the illustrated embodiments. 1A to 1C are sectional views showing a first embodiment of a method of manufacturing a charge transfer device according to the present invention in the order of steps.

(A) After a gate insulating film 27 is formed on the surface of the semiconductor substrate 20, a polycrystalline silicon layer 26 is formed on the gate insulating film 27, and then the polycrystalline silicon layer 2 is formed.
6, a resist film 21 is formed, and the resist film 21 is exposed and developed to form a pattern in which openings 22, 22,... Are located on portions to be boundaries between adjacent transfer gate electrodes. Is patterned.

Next, by exposing and developing the resist film 21 on the gate insulating film 27, openings 22, 22,... Are formed on portions to be boundaries between adjacent transfer gate electrodes.
・ Pattern into a pattern where is located. The smaller the openings 22, 22,..., The thinner the silicon oxide film (28) for insulating between transfer gate electrodes to be formed later, and the smaller the potential pockets, the longer the insulation. It can be said that it is preferable to reduce the size as long as an insulating film necessary for the minimum separation can be formed. By the way, if the widths of the openings 22, 22... Can be made 0.5 μm or less, it is possible to sufficiently prevent the potential pockets from becoming large enough to cause the transfer residue, and then, the openings 22, 22,. It is sufficiently possible to make the width of 0.5 μm or less according to the current photolithography technology. FIG. 1A shows the openings 22, 22,
.. Shows the state after formation.

(B) Next, oxygen ions are implanted into the polycrystalline silicon layer 26 using the resist film 21 as a mask. FIG. 1B shows a state after the resist film 21 is removed. Due to this ion implantation, oxygen is forcibly implanted into portions of the polycrystalline silicon layer 26 exposed to the openings 22, 22,... Of the resist film 21. Numeral 28 is a portion where the oxygen is implanted.

At the time of this ion implantation, oxygen is implanted into a region narrower than the width of the opening 22 by adjusting the thickness of the resist film 21 and the ion implantation angle, so that oxygen implantation becomes an insulating film for interphase insulation separation. The width of the region 28 can be made smaller than the width of the opening 22, and the potential pocket can be made smaller.

(C) Next, the resist film 21 is removed.
Thereafter, the semiconductor substrate 20 is heated and oxidized to form a silicon oxide film 30 for isolation between different phases between the adjacent transfer gate electrodes 26t and 26s. That is, the oxygen implanted region 28 becomes the silicon oxide film 30. Thereafter, the polycrystalline silicon films 26t, 26s, 26t, 26
oxidizes the surface of FIG. 1C shows a state after the oxidation, and 29 is an oxide film on the surface.

According to such a method of manufacturing a charge transfer device, a resist film 21 having a pattern having an opening 22 at a portion corresponding to a boundary between adjacent transfer gate electrodes 16t and 16s is formed on a polycrystalline silicon layer 16. , The resist film 2
1 is used as a mask, oxygen is ion-implanted into the polycrystalline silicon layer 16 and then annealed, so that the silicon oxide film 30 for separating the different phases is exposed at the portions of the polycrystalline silicon layer 16 exposed to the openings 22, 22,. Can be formed. Therefore, the transfer gate electrodes 26s, 26t, 26s, 26t,... Which are substantially or completely separated and independent from each other can be formed.

Then, the opening 22 of the resist film 21
Are formed to be narrow, for example, 0.5 μm or less by utilizing the current photolithography technology, so that silicon oxide films 30, 30,... For insulation between adjacent transfer gate electrodes are formed. Can be reduced, and the resulting potential pocket can be made smaller than before. In particular, at the time of this ion implantation, oxygen is implanted into a region narrower than the width of the opening 22 by adjusting the thickness of the resist film 21 and the ion implantation angle, and the oxygen implantation region 28 which becomes the insulating film 30 for interphase insulation separation is formed. Can be made smaller than the width of the opening 22, and the potential pocket can be made smaller.

Accordingly, it is possible to make the transfer gate electrode one layer structure and to make almost no potential pocket occur, and it is possible to remarkably reduce the number of manufacturing steps of the charge transfer device.

In the above embodiment, oxygen ions are ion-implanted as a material for converting the polycrystalline silicon layer into an insulator, thereby forming the silicon oxide film as an insulating film for phase separation. However, the present invention is not limited to this. However, the present invention is not limited to this. For example, nitrogen ions may be implanted to form the silicon nitride film as the phase separation insulating film.

[0041]

According to the method of manufacturing a charge transfer device of the present invention, a material which is converted into a polycrystalline silicon insulator by using a mask film having an opening on a portion of a polycrystalline silicon layer at a boundary between adjacent transfer gate electrodes as a mask. Is ion-implanted and then annealed, so that an insulating film can be formed between each adjacent transfer gate electrode, thereby enabling insulation separation between the electrodes.

By forming the opening of the mask film narrow using the current photolithography technique, the insulating film for separating the adjacent transfer gate electrodes can be made thin and narrow. Therefore, it is possible to make the transfer gate electrode one-layer structure and to make almost no potential pockets occur, thereby significantly reducing the number of manufacturing steps of the charge transfer device.

[Brief description of the drawings]

FIGS. 1A to 1C are sectional views showing a first embodiment of a method for manufacturing a charge transfer device according to the present invention in the order of steps.

FIGS. 2A to 2C show a charge transfer device for explaining the technical background of the present invention, wherein FIG.
(B) is a potential diagram, and (C) is a pulse diagram of a driving clock pulse.

3 (A) and 3 (B) show a problem when a transfer gate electrode is formed of a single polycrystalline silicon layer by a conventional technique, wherein FIG. 3 (A) is a cross-sectional view and FIG. It is a potential diagram.

FIGS. 4A and 4B are for explaining the reason why a transfer gate electrode has a multilayer structure in the related art.
(A) is a sectional view, and (B) is a potential diagram.

5A to 5G are cross-sectional views showing a conventional example in the order of steps.

[Explanation of symbols]

20: semiconductor substrate, 26: polycrystalline silicon layer,
26s, 26t: transfer gate electrode, 21: mask film (resist film), 22: opening, 26: polycrystalline silicon layer, 26t, 26s: transfer gate electrode,
27: gate insulating film, 28: ion-implanted region of a substance to be converted into an insulating material, 30: insulating film for insulating separation.

Claims (3)

[Claims] A step of forming a polycrystalline silicon layer serving as a transfer gate electrode on a gate insulating film formed on a surface of a semiconductor forming a transfer channel; and forming each adjacent transfer to be formed on the polycrystalline silicon layer. A step of forming a mask film having a pattern having an opening at a portion corresponding to a boundary between gate electrodes, an ion implantation step of ion-implanting a substance for insulating the polycrystalline silicon layer using the mask film as a mask, and annealing. Forming a portion of the polycrystalline silicon layer, into which the material for insulating the polycrystalline silicon is implanted, as an insulator for separating between adjacent transfer gate electrodes. 2. The substance to be ion-implanted is oxygen,
2. The charge transfer device according to claim 1, wherein the substance obtained by converting the polycrystalline silicon layer into an insulating film is a silicon oxide film. 3. The ion implanting substance is nitrogen,
2. The charge transfer device according to claim 1, wherein the substance in which the polycrystalline silicon layer is turned into an insulating film is a silicon nitride film.