Semiconductor storage device having a function to convert changes of an electric charge amount to a current amount Number:7,141,849 from the United States Patent and Trademark Office (PTO) owispatent In a semiconductor storage device, a gate insulating film and a gate electrode are laid on a first conductivity type semiconductor substrate, and charge holding portions are formed on both sides of the gate electrode. Second conductivity type first and second diffusion layer regions are formed in regions of the semiconductor substrate corresponding to the charge holding portions. The charge holding portions are each structured so as to change, in accordance with an electric charge amount held in the charge holding portions, a current amount flowing from one of the second conductivity type diffusion layer regions to the other of the diffusion layer regions through a channel region when voltage is applied to the gate electrode. Part of each charge holding portion is present below an interface of the gate insulating film and the channel region.<H Semiconductor, electric, Semiconductor, s, 7141849.html, Patent, pto, 7141849

Title: Semiconductor storage device having a function to convert changes of an electric charge amount to a current amount

Abstract: In a semiconductor storage device, a gate insulating film and a gate electrode are laid on a first conductivity type semiconductor substrate, and charge holding portions are formed on both sides of the gate electrode. Second conductivity type first and second diffusion layer regions are formed in regions of the semiconductor substrate corresponding to the charge holding portions. The charge holding portions are each structured so as to change, in accordance with an electric charge amount held in the charge holding portions, a current amount flowing from one of the second conductivity type diffusion layer regions to the other of the diffusion layer regions through a channel region when voltage is applied to the gate electrode. Part of each charge holding portion is present below an interface of the gate insulating film and the channel region.

Patent Number: 7,141,849 Issued on 11/28/2006 to Iwata, et al.

Inventors: Iwata; Hiroshi (Nara, JP), Shibata; Akihide (Nara, JP)
Assignee: Sharp Kabushiki Kaisha (Osaka, JP)

Appl. No.: 10/530,519

Filed: October 1, 2003

PCT Filed: October 01, 2003

PCT No.: PCT/JP03/12583

371(c)(1),(2),(4) Date: October 28, 2005

PCT Pub. No.: WO20/04/034474

PCT Pub. Date: April 22, 2004

Foreign Application Priority Data


Oct 09, 2002 [JP]			2002-296001
May 14, 2003 [JP]			2003-136453

Current U.S. Class: 257/324 ; 257/326

Current International Class: H01L 29/792 (20060101)

Field of Search: 257/314-316,326,324,623

References Cited [Referenced By]

U.S. Patent Documents


5880499	March 1999	Oyama
5998263	December 1999	Sekariapuram et al.
6388293	May 2002	Ogura et al.
6803624	October 2004	Rudeck et al.
6828621	December 2004	Kusumi et al.
2004/0245564	December 2004	Ogura et al.
2006/0145243	July 2006	Wong et al.

Foreign Patent Documents


5-81072	Nov., 1993	JP
9-116119	May., 1997	JP
2001-156188	Jun., 2001	JP
2001-230332	Aug., 2001	JP
2002-170891	Jun., 2002	JP
WO-99/07000	Feb., 1999	WO
WO-02/43158	May., 2002	WO

Primary Examiner: Pham; Hoai
Assistant Examiner: Hafiz; Mursalin B.
Attorney, Agent or Firm: Birch, Stewart, Kolasch & Birch, LLP

Claims

What is claimed is:

1. A semiconductor storage device comprising: a first conductivity type semiconductor part composed of any one of a semiconductor substrate a well region provided in a semiconductor substrate and a semiconductor film disposed on an insulator; a gate insulating film formed on the first conductivity type semiconductor part; a gate electrode formed on the gate insulating film; charge holding portions formed on opposite sides of the gate electrode; second conductivity type diffusion layer regions formed in regions of the first conductivity type semiconductor part corresponding to the respective charge holding portions; and a channel region below the gate electrode; wherein the charge holding portions are each constructed so as to change, in accordance with an electric charge amount held in the charge holding portions, a current amount flowing from one of the second conductivity type diffusion layer regions to the other of the diffusion layer regions through the channel region when a voltage is applied to the gate electrode; and part of each charge holding portion is present below an interface of the gate insulating film and the channel region.

2. The semiconductor storage device as defined in claim 1, wherein a distance (D) between the interface of the gate insulating film and the channel region and a plane approximately parallel to the interface and containing a bottom of each charge holding portion is 2 nm to 15 nm.

3. The semiconductor storage device as defined in claim 1, wherein the charge holding portions each include: a first insulator having a function of holding electric charges; and a second insulator having a function of preventing dissipation of the electric charges held in the first insulator by separating the first insulator from the gate electrode, the channel region and the corresponding diffusion layer region.

4. The semiconductor storage device as defined in claim 3, wherein the first insulator is silicon nitride, and the second insulator is silicon oxide.

5. The semiconductor storage device as defined in claim 1, wherein the charge holding portions each include a first insulator having a function of storing electric charges, and second and third insulators having a function of preventing dissipation of the electric charges held in the first insulator, and the first insulator is interposed between the second insulator and the third insulator.

6. The semiconductor storage device as defined in claim 5, wherein the first insulator is silicon nitride, and the second and third insulators are silicon oxide.

7. The semiconductor storage device as defined in claim 5, wherein the second insulator is provided in a manner so as to separate the first insulator from the gate electrode, the channel region and the corresponding diffusion layer region, and a thickness (T1) of the second insulator on the channel region is smaller than a thickness (T2) of the gate insulating film, but 0.8 nm or more.

8. The semiconductor storage device as defined in claim 5, wherein the second insulator is formed in a manner so as to separate the first insulator from the gate electrode, the channel region and the diffusion layer region, and a thickness (T1) of the second insulator on the channel region is larger than a thickness (T2) of the gate insulating film, but not more than 20 nm.

9. The semiconductor storage device as defined in claim 5, wherein the first insulator includes a portion opposed to a plane parallel to a top surface of the gate insulating film, with the second insulator disposed therebetween, the portion of the first insulator extending along the plane.

10. The semiconductor storage device as defined in claim 9, wherein the first insulator includes a portion opposed to a side surface of the gate electrode, with the second insulator disposed therebetween, the portion of the first insulator extending along the side surface.

11. The semiconductor storage device as defined in claim 1, wherein the charge holding portions are arranged in a channel length direction such that at least part of each charge holding portion is laid on the corresponding diffusion layer region.

12. The semiconductor storage device as defined in claim 1, wherein a distance (B) between the diffusion layer regions is set longer than a gate electrode length (A) in a channel length direction, and the charge holding portions are arranged in the channel length direction such that only part of each charge holding portion is laid on the corresponding diffusion layer region.

Description

This application is the U.S. National Phase of International Application PCT/JP2003/012583, filed Oct. 1, 2003, which designated the U.S. PCT/JP2003/012583 claims priority to JP Patent Applications No. 2002-296001, filed Oct. 9, 2002 and No. 2003-136453, filed May 14, 2003. The entire contents of these applications are incorporated herein by reference.

TECHNICAL FIELD

This invention relates to a semiconductor storage device. In particular, it relates to a semiconductor storage device having a field-effect transistor having a function to convert changes of an electric charge amount to a current amount.

BACKGROUND ART

Conventionally, there has been a nonvolatile memory capable of storing two bits by one field-effect transistor (e.g., see JP-2001-512290 A, published Aug. 21, 2001). The structure of this nonvolatile memory and the principle of its write (program) operation will be described hereinbelow.

As shown in FIG. 25, this memory is composed of a gate electrode 909 formed on a P type well region 901 through a gate insulating film, and a first N type diffusion layer region 902 and a second N type diffusion layer region 903 formed at the surface of the P type well region 901. The gate insulating film is composed of a so-called ONO (Oxide Nitride Oxide) film in which a silicon nitride film 906 is interposed between silicon oxide films 904 and 905. In the silicon nitride film 906, there are formed memory holding portions 907, 908 in the vicinity of the edge portions of the first and second N type diffusion layer regions 902, 903. An electric charge amount in each of these memory holding portions 907, 908 is read as a drain current of the transistor so that two-bit information is stored in one transistor.

Next description will be given of a write operation method in this nonvolatile memory. The term "write (or program)" is used herein to refer to the action of injecting electrons into the memory holding portion 907, 908. In JP-2001-512290 A, there has been disclosed a method for injecting electrons into a right memory holding portion 908 by applying 5.5V to the second diffusion layer region 903 and 10V to the gate electrode 909. This makes it possible to write to a specific one of the two memory holding portions. There has been also disclosed a method for erasing and reading from a specific side. By combining these methods, two-bit operation is enabled.

In the above-stated nonvolatile memory, in order to provide the gate insulating film with the function of operating the transistor as well as the function as a memory film for storing electric charges, the gate insulating film is formed into three-layer structure with use of the ONO film. This makes it difficult to manufacture thinner gate insulating films. Also, in the above nonvolatile memory, as the channel length is shortened, the two memory holding portions 907, 908 in one transistor interfere with each other, which makes two-bit operation difficult. This obstructs further miniaturization of the devices.

SUMMARY OF THE INVENTION

In view of the above problems, it is an object of the present invention to provide a semiconductor storage device allowing further miniaturization while fulfilling two-bit or more memory holding operation in one transistor.

In order to accomplish the above object, a semiconductor storage device according to the present invention includes a first conductivity type (e.g., one of the P and N types) semiconductor part composed of any one of a semiconductor substrate, a well region provided in a semiconductor substrate and a semiconductor film disposed on an insulator; a gate insulating film formed on the first conductivity type semiconductor part; a gate electrode formed on the gate insulating film; charge holding portions formed on opposite sides of the gate electrode; second conductivity type (e.g., the other of the P and N types) diffusion layer regions formed in regions of the first conductivity type semiconductor part corresponding to the respective charge holding portions; and a channel region below the gate electrode. The charge holding portions are each constructed so as to change, in accordance with an electric charge amount held in the charge holding portions, a current amount flowing from one of the second conductivity type diffusion layer regions to the other of the diffusion layer regions through the channel region when a voltage is applied to the gate electrode. Also, part of each charge holding portion is present below an interface of the gate insulating film and the channel region. The charge holding portion herein refers to a component allowing the injection and pulling of electrons or holes.

According to the semiconductor storage device with the above constitution, a change in the electric charge amount in the charge holding portion is converted to a current amount whereby it is operated as a memory cell. Since the two charge holding portions formed on the opposite sides of the gate electrode are formed independently of the gate insulating film, a memory function implemented by the charge holding portions and a transistor function implemented by the gate insulating film are separated. Therefore, while maintaining the sufficient memory function, the gate insulating film is made thinner, so that short channel effect can be easily prevented.

Since the two charge holding portions formed on both sides of the gate electrode are separated by the gate electrode, interference in rewrite operation can effectively be prevented. In other words, the distance between the two charge holding portions can be shortened. Accordingly, a semiconductor storage device allowing further miniaturization while fulfilling the two-bit or more memory holding in one transistor can be realized.

Furthermore, since part of each charge holding portion is present below the interface of the gate insulating film and the channel region, it follows that the charge holding portions are present in a direction in which high-energy electric charges proceed during a write operation. Thus, the injection efficiency of electric charges into the charge holding portions is markedly improved, so that the speed of write operation can greatly be improved. It is also possible to reduce the power consumption in writing if the current in the writing operation is reduced.

In the semiconductor storage device, a distance between the interface of the gate insulating film and the channel region and a plane approximately parallel to the interface and containing a bottom of each charge holding portion may be 2 nm to 15 nm.

With the above constitution, the positional relationship between the charge holding portions and the gate insulating film is optimized, thus making it possible to keep the operation speed high enough.

The semiconductor storage device may be configured such that the charge holding portions each include a first insulator having a function of holding electric charges; and a second insulator having a function of preventing dissipation of the electric charges held in the first insulator by separating the first insulator from the gate electrode, the channel region and the corresponding diffusion layer region.

With the above constitution, dissipation of the electric charges stored in the first insulator can efficiently be prevented. Thus, a semiconductor storage device with good charge holding characteristics is provided.

The semiconductor storage device may also be configured such that the charge holding portions each include a first insulator having a function of storing electric charges, and second and third insulators having a function of preventing dissipation of the electric charges held in the first insulator, and that the first insulator is interposed between the second insulator and the third insulator.

With the above constitution, the electric charges injected into the charge holding portion are blocked by the second and third insulators and stored in the first insulator efficiently. Therefore, the injection efficiency of electric charges into the charge holding portion is improved, so that a high-speed rewrite operation (write and erase operations) is realized.

The first insulator may be silicon nitride, and the second and third insulators may be silicon oxide.

With the above constitution, since the silicon nitride film in which a number of levels for trapping electric charges are present is used as the first insulator, and the silicon oxide film having a large bandgap is used as the second and third insulators, a semiconductor storage device having a large hysteresis property and good holding characteristics is provided. Further, because both of the silicon nitride and the silicon oxide are materials normally used in LSI process, the semiconductor storage device is advantageously easy to produce.

The semiconductor storage device may be configured such that the second insulator is provided in a manner so as to separate the first insulator from the gate electrode, the channel region and the corresponding diffusion layer region, and a thickness of the second insulator on the channel region is smaller than a thickness of the gate insulating film, but 0.8 nm or more.

According to the above constitution, the thickness of the second insulator that separates the first insulator being a charge holding film from the channel region is smaller than that of the gate insulating film, but 0.8 nm or more. Thus, without deteriorating the voltage withstanding performance or electric strength of the memory, reduction of voltage in the write operation and erase operation or implementing a high-speed write operation and erase operation is enabled. This makes it possible to increase memory effect.

The "memory effect" means that when a voltage is applied to the gate electrode, a current amount that flows from one of the diffusion layer regions to the other through the channel region is changed in accordance with an electric charge amount held in the charge holding film (charge holding portion). The memory effect being large means that the change of the current amount is large.

The semiconductor storage device may be configured such that the second insulator is formed in a manner so as to separate the first insulator from the gate electrode, the channel region and the diffusion layer region, and that a thickness of the second insulator on the channel region is larger than a thickness of the gate insulating film, but not more than 20 nm.

According to the above constitution, the thickness of the insulating film that separates the charge holding film from the channel region is larger than that of the gate insulating film, but not more than 20 nm. This makes it possible to improve the holding characteristics without deteriorating the memory short channel effect.

The first insulator may include a portion opposed to a plane parallel to a top surface of the gate insulating film, with the second insulator disposed therebetween, the portion of the first insulator extending along the plane (for example, the first insulator may include a portion having a surface generally parallel to the top surface of the gate insulating film.)

According to the above constitution, the rewrite speed can be increased while preventing the deterioration of the holding characteristics of the semiconductor storage device.

The first insulator may include a portion opposed to a side surface of the gate electrode, with the second insulator disposed therebetween, the portion of the first insulator extending along the side surface (for example, the first insulator may include a portion having a surface generally parallel to the side surface of the gate electrode.)

According to the above constitution, the rewrite speed of the semiconductor storage device can be increased.

In the semiconductor storage device, the charge holding portions may be arranged in a channel length direction such that at least part of each charge holding portion is laid on the corresponding diffusion layer region.

Further, a distance between the diffusion layer regions may be set longer than a gate electrode length in a channel length direction, and the charge holding portions may be arranged in the channel length direction such that only part of each charge holding portion is laid on the corresponding diffusion layer region.

The above constitution increases the read or sense current of the semiconductor storage device so that the dispersion of read or sense current may be controlled. Therefore, the readout operation of the semiconductor storage device can be made at a higher speed.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic cross-sectional view of essential parts of one example of a semiconductor storage device according to Embodiment 1 of this invention;

FIG. 2 is an enlarged view of a part of FIG. 1;

FIG. 3 is a schematic cross-sectional view of essential parts of a modified example of the semiconductor storage device according to Embodiment 1 of this invention;

FIG. 4 is a schematic cross-sectional view of essential parts of another modified example of the semiconductor storage device according to Embodiment 1 of this invention;

FIG. 5 is a view for describing a method of writing to a second charge holding portion;

FIG. 6 is a view for describing a method of writing to a first charge holding portion;

FIG. 7 is a view for describing a positional relationship between a gate insulating film and charge holding portions in the semiconductor storage device;

FIG. 8 is a view for describing a positional relationship between a gate insulating film and charge holding portions in the semiconductor storage device;

FIG. 9 is a schematic cross-sectional view of essential parts showing a basic structure of a semiconductor storage device of this invention;

FIG. 10 is an enlarged view of part of FIG. 9;

FIG. 11 is a schematic cross-sectional view of essential parts of a semiconductor storage device according to Embodiment 2 of this invention;

FIG. 12 an enlarged view of a part of FIG. 11;

FIG. 13 is an enlarged view of a part of a modification to FIG. 11;

FIG. 14 is a graph showing electrical characteristics of a semiconductor storage device having the same construction as the semiconductor storage device of FIGS. 11 and 12 except that a bottom surface of the charge holding portion is at the same level as that of an interface between a gate insulating film and a semiconductor substrate;

FIG. 15 is a schematic cross-sectional view of essential parts of a modified example according to Embodiment 2 of this invention;

FIG. 16 is a schematic cross-sectional view of essential parts of a semiconductor storage device according to Embodiment 3 of this invention;

FIG. 17 is a schematic cross-sectional view of essential parts of a semiconductor storage device according to Embodiment 4 of this invention;

FIG. 18 is a schematic cross-sectional view of essential parts of a semiconductor storage device according to Embodiment 5 of this invention;

FIG. 19 is a schematic cross-sectional view of essential parts of a semiconductor storage device according to Embodiment 6 of this invention;

FIG. 20 is a schematic cross-sectional view of essential parts of a semiconductor storage device according to Embodiment 7 of this invention;

FIG. 21 is a schematic cross-sectional view of essential parts of a semiconductor storage device according to Embodiment 8 of this invention;

FIG. 22 is a graph showing electrical characteristics of a semiconductor storage device having the same construction as the semiconductor storage device of FIG. 21 except that a bottom surface of the charge holding portion is at the same level as that of an interface between a gate insulating film and a semiconductor substrate;

FIG. 23 is a graph showing electrical characteristics of a conventional flash memory;

FIG. 24 is a schematic cross-sectional view of essential parts of another embodiment of the semiconductor storage device of this invention; and

FIG. 25 is a schematic cross-sectional view of essential parts of a conventional semiconductor storage device.

DETAILED DESCRIPTION OF THE INVENTION

The semiconductor storage device of this invention will be described in detail using examples illustrated.

(Embodiment 1)

Before describing a memory device constituting a semiconductor storage device of the present invention, a basic structure that is a premise of a memory device to which this invention is applied will be described first with reference to FIGS. 9 and 10.

FIG. 9 is a schematic cross-sectional view of a memory device. In the memory device as a nonvolatile memory cell capable of storing two bits, as shown in FIG. 9, a gate electrode 13 having a gate length similar to those of normal transistors is formed on a semiconductor substrate 11 that is one example of the first conductivity type (e.g., one of the P type or N type) semiconductor part, with a gate insulating film 12 therebetween. Charge holding portions 10A, 10B in the shape of side wall spacer are formed on side surfaces, or side walls of the gate insulating film 12 and the gate electrode 13 that are stacked. That is, one of the charge holding portions 10A and 10B is disposed on one of the side walls, while the other of the charge holding portions 10A and 10B is formed on the other of the side walls.

In regions of the semiconductor substrate 11 corresponding to the charge holding portions 10A, 10B on the side opposite from the gate electrode 13, there are formed a second conductivity type (e.g., the other of P or N type) first diffusion layer region 17 and a second conductivity type second diffusion layer region 18. The first and second diffusion layer regions (source/drain regions) 17, 18 are offset from edge portions of the gate electrode 13 (from a region 41 where the gate electrode 13 is formed). Thus, an offset region 42 is formed between each edge portion of the gate electrode 13 and the first/second diffusion layer region 17, 18 in the semiconductor substrate.

Thus, the charge holding portions 10A, 10B of the memory device are formed independently of the gate insulating film 12. Consequently, a memory function implemented by the charge holding portions 10A, 10B and a transistor function implemented by the gate insulating film 12 are separated. Since the two charge holding portions 10A, 10B formed on both sides of the gate electrode 13 are separated by the gate electrode 13, possible interference in the rewrite operation is effectively controlled. Therefore, the memory device is capable of storing two-bit or more information and enables easy miniaturization.

Further, since the first and second diffusion layer regions 17, 18 are offset from the gate electrode 13, easiness of inversion of the offset region 42 under the charge holding portion 10A, 10B (i.e., a region in the semiconductor substrate 11 opposite from the charge holding portion 10A, 10B) when a voltage is applied to the gate electrode is largely changed by an electric charge amount stored in the charge holding portion 10A, 10B, which enables increase of the memory effect. The "memory effect" means that when a voltage is applied to the gate electrode 13, a current amount that flows from one of the diffusion layer regions to the other through the channel region is changed in accordance with an electric charge amount held in the charge holding film (charge holding portion 10A, 10B). That the memory effect is large means that the change of the current amount is large.

Further, because of the structure in which the first and second diffusion layer regions 17, 18 are offset from the gate electrode 13, the short channel effect can be strongly prevented, compared with normal logic transistors, which enables further reduction of the gate length. Also, since the memory device is structurally suitable for controlling the short channel effect, it becomes possible to adopt a gate insulating film with a larger film thickness, compared with logic transistors, thereby enabling increase of reliability.

FIG. 10 is an enlarged view of the charge holding portion 10B in the shape of sidewall spacer and its vicinity. As shown in FIG. 10, the charge holding portions 10A, 10B (only 10B is shown in FIG. 10) are each composed of a silicon nitride film 15 in the shape of side wall spacer that is an example of the first insulator and a silicon oxide film 14 that is an example of the second insulator for separating the silicon nitride film 15 from the gate electrode 13, the semiconductor substrate 11, and the first and second diffusion layer regions 17, 18. It is the silicon nitride film 15 that has a function of storing electric charges (electrons or holes), and the silicon oxide film 14 prevents the electric charges stored in the silicon nitride film 15 from leaking.

Next, examples of a memory device constituting the semiconductor storage device of this invention will be described below with reference to FIGS. 1 8. FIG. 1 and FIG. 2 are views showing one example of the memory device and a partly enlarged portion thereof, respectively. FIG. 3 shows a modified example in which the constitution in FIG. 1 is partly changed. FIG. 4 shows another modified example. FIGS. 5 and 6 are views for explaining a write operation in these examples, and FIGS. 7 and 8 are views for explaining a positional relationship between a gate insulating film and charge holding portions in the memory device.

FIG. 1 is a schematic cross-sectional view of a memory device as one embodiment of the semiconductor storage device of this invention. This memory device is different from the memory device shown in FIG. 9 in the positional relationship between the gate insulating film 12 and the charge holding portions 10A, 10B. That is, in the memory device shown in FIG. 1, part of each charge holding portion 10A, 10B is present below an interface between the gate insulating film 12 and the semiconductor substrate 11. A superficial portion of the semiconductor substrate 11 serves as a channel region.

In the memory device shown in FIG. 1, the same constituent parts as those of the memory device shown in FIG. 9 are designated by the same numerals and their description is omitted.

FIG. 2 is an enlarged view of the charge holding portion 10B in the shape of sidewall spacer and its vicinity. As shown in FIG. 2, the charge holding portions 10A, 10B (only 10B is shown in FIG. 2) is composed of a silicon nitride film 15 in the shape of side wall spacer and a silicon oxide film 14 for separating the silicon nitride film 15 from the gate electrode 13, the semiconductor substrate 11, and the first/second diffusion layer region 17, 18 (source/drain regions). It is the silicon nitride film 15 that has a function of storing electric charges (electrons or holes), and the silicon oxide film 14 prevents the electric charges stored in the silicon nitride film 15 from leaking.

FIG. 3 shows a schematic cross-sectional view of a memory device as a modified example of the memory device (semiconductor storage device) of this invention. The constitution of this memory device is different from that of the memory device shown in FIGS. 1 and 2. That is, as shown in FIG. 3, charge holding portions 20A, 20B each have a structure in which a silicon nitride film 22 as an example of the first insulator having a function of trapping electric charges is sandwiched between silicon oxide films 21, 23 as examples of second insulators.

In the memory device shown in FIG. 3, the same constituent parts as those of the memory device shown in FIG. 9 are designated by the same numerals and their description is omitted.

FIG. 4 shows a schematic cross-sectional view of a memory device as another modified example of the memory device (semiconductor storage device) of this invention. The constitution of this memory device is different from that shown in FIG. 3 in the substrate. That is, as shown in FIG. 4, here is used a substrate in which a semiconductor layer 72 is formed on an insulating layer 71. Examples of such a substrate include SOI substrates (the insulating layer 71 and the semiconductor layer 72 correspond to a buried oxide layer and an SOI layer respectively), and substrates formed by adopting wafer bonding technology.

In the memory device shown in FIG. 4, the same constituent parts as those of the memory device shown in FIG. 9 are designated by similar numerals and their description is omitted.

The memory device of the second modified example (FIG. 4) has the same effect as the memory device of the first modified example (FIG. 3).

The memory devices (FIGS. 1 4) are each characterized in that part of the charge holding portion 10A, 10B (20A, 20B) is present below an interface between the gate insulating film 12 and the semiconductor substrate 11 (semiconductor layer 72). The effect obtained by disposing the gate insulating film 12 and the charge holding portions 10A, 10B (20A, 20B) in this manner will be described below.

According to the above memory devices, high-speed write operation is enabled. It is noted that the term "write operation" refers to the action of injecting electrons into the charge holding portion when the memory device is of N channel type, and to the action of injecting holes into the charge holding portion when the memory device is of P channel type.

Next, a method of writing to the memory device will be described using FIGS. 5 and 6. The method of writing is common to the memory devices of the embodiments and even memory devices having the basic structure described above. The writing is performed by injecting electrons accelerated by drain electric fields into the charge holding portion.

First, in order to inject electrons (write) into the second charge holding portion 20B, as shown in FIG. 5, the first diffusion layer region 17 is used as a source electrode and the second diffusion layer region 18 is used as a drain electrode as shown in FIG. 5. For example, a 0V is applied to the first diffusion layer region 17 and the semiconductor substrate 11, +5V to the second diffusion layer region 18, and +5V to the gate electrode 13. Under these voltage conditions, an inversion layer 31 extends from the first diffusion layer region 17 (source electrode) but fails to reach the second diffusion layer region 18 (drain electrode), resulting in generation of a pinchoff point. Electrons are accelerated by drain electric fields from the pinchoff point to the second diffusion layer region 18 (drain electrode) and injected into the second charge holding portion 20B (more precisely the silicon nitride film 22 in the second charge holding portion 20B). The write operation is executed in this manner.

It is noted that in the vicinity of the first charge holding portion 20A, electrons accelerated by drain electric fields are not generated and therefore writing is not executed. The voltage for write operation is not limited to the above voltages. For example, when a 0V was applied to the first diffusion layer region 17 and the semiconductor substrate 11, +10V to the second diffusion layer region 18, and +5V to the gate electrode 13, hot electrons (thermoelectrons) were injected into the charge holding portion 20B, so that a write operation is implemented.

In this manner, electrons are injected into the second charge holding portion 20B so as to enable a write operation.

In the memory devices according to the first to third examples, it is possible to implement write operations extremely efficiently as compared with the memory device with the basic structure shown in FIGS. 9, 10. That is, in the memory device shown in FIGS. 9 and 10, among electrons that have been accelerated directed from the pinchoff point to the second diffusion region 18, only a small part of which are scattered upward and injected into the second charge holding portion 10B. On the other hand, in the memory devices of the above embodiments (FIGS. 1 4), electrons move toward a direction of arrow 32 in FIG. 5, a large part of which are injected as such into the silicon nitride film 22 in the charge holding portion 20B. In other words, because most of the electrons that have been accelerated from the pinchoff point have a large momentum in the direction of arrow 32, the number of electrons that pass the silicon oxide film 21 to be injected into the silicon nitride film 22 remarkably increases.

In this way, in the memory devices of the above embodiments, since the efficiency of write operation is remarkably improved, the speed of write operation can remarkably be improved. Or, in the case where the current in writing is small, the power consumption in writing to the semiconductor storage device can be reduced.

In order to inject electrons (write) into the first charge holding portion 20A, as shown in FIG. 6, the second diffusion layer region 18 is used as a source electrode, and the first diffusion layer region 17 is used as a drain electrode. For example, a 0V is applied to the second diffusion layer region 18 and the semiconductor substrate 11, +5V to the first diffusion layer region 17, and +5V to the gate electrode 13. In this case, electrons move toward a direction of arrow 33 and injected into the silicon nitride film 22 in the charge holding portion 20A. By thus switching the source and drain regions from the case of injecting electrons into the second charge holding portion 20B, electrons are injected into the first charge holding portion 20A for enabling write operation.

In the memory device shown in FIGS. 1 and 2, the silicon nitride film 15 having a function of storing electric charges is separated from the gate electrode 13, the semiconductor substrate 11, and the diffusion layer region 17, 18 by the silicon oxide film 14. The silicon oxide film 14 prevents dissipation of electric charges stored in the silicon nitride film 15. The silicon oxide film 14 preferably has a thickness of at least 2 nm, because if the thickness of the silicon oxide film 14 is less than 2 nm, tunneling effect of electric charges becomes outstanding, with the result that the retention time of the memory device becomes short.

In the case of the memory device shown in FIG. 3 or 4, the charge holding portions 20A, 20B each have a structure in which the silicon nitride film 22 as the first insulator having a function of trapping electric charges is interposed between the silicon oxide films 21, 23 as the second insulators. Therefore, electric charges injected into the charge holding portions 20A, 20B are blocked by the silicon oxide film 23 and stored in the silicon nitride film 22 efficiently. As described above, since the charge holding portion 20A, 20B has the structure in which the silicon nitride film 22 is interposed between the silicon oxide films 21, 23, the injection efficiency of electric charges into the charge holding portion 20A, 20B is improved, so that a high-speed rewrite operation (write and erase operations) is realized.

The structure of the charge holding portion is not limited to the above examples (FIGS. 1 to 4), and the charge holding portion, therefore, may contain nanometer-sized quantum dots having a function of storing electric charges. Also, the charge holding portion does not need to have a side wall shape, as long as the charge holding portion is in both sides of the gate electrode and part thereof is in contact with the semiconductor substrate 11 and the source/drain regions 17, 18. However, if the charge holding portion is formed to have a side wall shape, it can be formed using a self-alignment process in the same manner as in forming side walls of gate electrodes of transistors with a conventional structure. In this case, it becomes possible to easily form an LSI having both logic transistors and memory transistors by forming gate electrode side walls common to the logic and memory transistors.

Next, a preferred positional relationship between the gate insulating film 12 and the charge holding portions 20A, 20B will be described using FIGS. 7 and 8. The distance between an interface of the gate insulating film 12 and the semiconductor substrate 11 (first plane), and a plane containing a lower face of the charge holding portions 20A, 20B (second plane) is denoted by D. The thickness of the silicon oxide film 21 that separates the silicon nitride film 22 from the diffusion layer region 17, 18 is denoted by T. In the case of the memory device with the structure shown in FIG. 1, the thickness of the silicon oxide film 14 that separates the silicon nitride film 15 from the diffusion layer regions 17, 18 can be denoted by T. At this time, the distance D between the first plane and the second plane is preferably within the range of from 2 nm to 15 nm. The reason for that will be described below.

As described above, the thickness T of the silicon oxide film 21 (corresponding to the silicon oxide film 14 in FIG. 1) is preferably 2 nm or more. In the case where the thickness T of the silicon oxide film is 2 nm or more, but the distance D is less than 2 nm, the silicon nitride film 22 is not present below the interface (first plane) of the gate insulating film 12 and the semiconductor substrate 11. Therefore, electrons accelerated by the drain electric fields during the write operation are not directly injected into the silicon nitride film 22, resulting in deterioration of the efficiency of write operation. Accordingly, the distance D is preferably 2 nm or more.

On the other hand, in the case where the distance D is 15 nm or more, regions where inversion layers are hardly formed (corresponding to regions indicated by 81, 81 in FIG. 7) because of not being affected by gate electric fields become larger, so that the drive current of the transistor markedly decreases. This mainly causes an increased time of readout operation. Thus, the distance D is preferably not more than 15 nm.

Because of the reasons as above, the distance D between the interface (first plane) of the gate insulating film 12 and the semiconductor substrate 11, and the place (second plane) including the lower surface of the charge holding portions 20A, 20B having the memory function is preferably in the range of between 2 nm and 15 nm. Thereby, the positional relationship between the charge holding portions 20A, 20B and the gate insulating film 12 is optimized, which makes it possible to keep the operation speed of the memory device high enough.

Incidentally, the silicon oxide film 21 has a clearly-angled L shape in FIG. 7, but it may have a round shape as shown in 8. In this case, the distance D is defined as a distance between the interface (first plane) of the gate insulating film 12 and the semiconductor substrate 11, and a plane (second plane) approximately parallel to that interface and in contact with the lowermost surface portion of the charge holding portions 10A, 10B. The rounded silicon oxide film 21 improves the device characteristics as well as the reliability thereof by avoiding the concentration of the electric fields which would otherwise occur in an angled portion.

As described above, in any one of the above examples of the memory device as the semiconductor storage device, the charge holding portions of the memory device are formed independently of the gate insulating film and disposed on both sides of the gate electrode. Therefore, the memory storage operation of two-bit or more information is enabled. Further, because the charge holding portions are separated from each other by the gate electrode, interference therebetween in writing can effectively be suppressed. Furthermore, the memory function implemented by the charge holding portions and the transistor operation function implemented by the gate insulating film are separated. Thus, the thickness of the gate insulating film is reducible, so that the short channel effect can be suppressed. Accordingly, miniaturization of the device is facilitated.

In any one of the above examples of the memory device as the semiconductor storage device, since the charge holding portions are located in a direction in which high-energy electric charges proceed in the write operations, the injection efficiency of electric charges into the charge holding portions is markedly improved, so that the writing speed can greatly be improved. It is also possible to reduce a writing current to thereby reduce the power consumption in write operations.

In the following embodiments, various preferred configurations of a memory device constituting the semiconductor storage device of the present invention will be described.

(Embodiment 2)

In a memory device (semiconductor memory storage device) of Embodiment 2, as shown in FIG. 11, each charge holding portion 161, 162 is composed of a region for holding electric charges and a region for obstructing release or escape of electric charges. The region for holding electric charges is a region that stores electric charges, which may be a film having a function of holding electric charges. The region for obstructing release of electric charges may be given by a film having a function of obstructing release of electric charges.

The charge holding portion 161, 162 has, for example, an ONO (Oxide Nitride Oxide) structure. More specifically, the charge holding portions 161, 162 are each structured in the state that a silicon nitride film 142 is interposed between a silicon oxide film 141 and a silicon oxide film 143. Here, the silicon nitride film 142 implements a function of holding electric charges. The silicon oxide films 141, 143 implement a function of obstructing release of the electric charges stored in the silicon nitride film.

Also, the regions (silicon nitride films 142) for holding electric charges in the charge holding portions 161, 162 overlap with diffusion layer regions 112, 113, respectively, in a channel length direction. Herein, the term "overlap" is used to refer to the state that at least part of the region (silicon nitride film 142) for holding electric charges lies on at least part of the diffusion layer region 112, 113. Reference numeral 171 indicates offset regions 171 generated by offsetting the diffusion layer regions 112, 113 outwardly in the channel length direction relative to a gate electrode 117. Though not shown in the drawing, an uppermost surface area of a semiconductor substrate 111 under a gate insulating film 114 (a portion in the semiconductor substrate 111 opposite to the gate insulating film 114) serves as a channel region.

Next, description will be given of an effect of overlapping of the region 142 for holding electric charges in the charge holding portions 161, 162 and the diffusion layer regions 112, 113.

FIG. 12 is an enlarged view showing the vicinity of the charge holding portion 162 that is one of the charge holding portions 161, 162 shown in FIG. 11. In the figure, reference symbol W1 denotes an offset amount between the gate insulating film 114 and the diffusion layer region 113. Also, reference symbol W2 denotes the width of the charge holding portion 162 in the channel length direction of the gate electrode. In the memory device shown in the figure, since an edge of the silicon nitride film 142 on the side away from the gate electrode 117 in the charge holding portion 162 is aligned with an edge of the charge holding portion 162 on the side away from the gate electrode 117, the width of the charge holding portion 162 is defined as W2. An overlap amount between the charge holding portion 162 and the diffusion layer region 113 is represented by an expression of W2-W1. What is particularly important is that the silicon nitride film 142 in the charge holding portion 162 overlaps with the diffusion layer region 113, that is, the silicon nitride film 142 is configured such that the relation of W2>W1 is satisfied.

In the case where the edge of the silicon nitride film 142a on the side away from the gate electrode 117a in the charge holding portion 162a does not coincide with the edge of the charge holding portion 162a on the side away from the gate electrode 117a as shown in FIG. 13, W2 may be defined as the width from the edge of the gate electrode 117a to the edge of the silicon nitride film 142a on the side away from the gate electrode 117a. In FIG. 13, the portions of the memory device are denoted by adding a character "a" to those numerals denoting the corresponding portions in FIG. 12.

FIG. 14 shows a drain current Id in a semiconductor storage device having the same structure as the semiconductor storage device shown in FIGS. 11, 12, except that a lower face of each charge holding portion is at the same level as an interface of the gate insulating film and the semiconductor substrate, with the width W2 of the charge holding portion 162 being fixed to 100 nm and with the offset amount W1 being varied. Herein, the drain current was obtained by device simulation performed under the conditions that the charge holding portion 162 is in an erased state (a state in which holes are stored), and that the diffusion layer regions 112, 113 are used as a source electrode and a drain electrode, respectively. It has been confirmed that the electrical characteristics in various embodiments of this invention are similar to those shown in FIG. 14. Thus, it should be understood that the following description can apply to all the embodiments of this invention.

As is apparent from FIG. 14, with W1 being 100 nm or more (i.e., when the silicon nitride film 142 and the diffusion layer region 113 do not overlap), the drain current shows rapid reduction. Since a drain current value is almost in proportion to a read operation speed, memory performance rapidly deteriorates when W1 is 100 nm or more. In the range where the silicon nitride film 142 and the diffusion layer region 113 overlap, the drain current shows mild reduction. Therefore, it is preferable that the silicon nitride film 142 that is a film having a function of holding electric charges at least partially overlaps with the corresponding diffusion layer region (source/drain region) 112, 113.

Based on the above-described result of the device simulation, memory cell arrays were manufactured with W2 being fixed to a design value of 100 nm, and W1 being set to 60 nm and 100 nm as design values. When W1 is 60 nm, the silicon nitride film 142 overlaps with the diffusion layer region 112, 113 by 40 nm as a design value, and when W1 is 100 nm, there is no overlap as a design value.

As a result of measuring read times of these memory cell arrays and comparing them in worst cases considering dispersion or variations, it was found out that the case where W1 was 60 nm as a design value was 100 times faster in readout access time. From a practical standpoint, it is preferable that the read access time is 100 nanoseconds or less per bit. It was found out, however, that this condition is never satisfied in the case of W1=W2. It was also found out that W2-W1>10 nm is more preferable in consideration of the manufacturing variation.

For reading information stored in the charge holding portion 161 (region 181), it is preferable to set the diffusion layer region 112 as a source electrode and the diffusion layer region 113 as a drain region, as in Embodiment 1, and to form a pinchoff point in the channel region on the side closer to the drain region. More specifically, in reading information stored in one of the two charge holding portions 161, 162, the pinchoff point is preferably formed in a region closer to the other charge holding portion of the channel region. This makes it possible to detect stored information in one charge holding portion 161, for example, with good sensitivity regardless of the storage condition of the other charge holding portion 162, resulting in large contribution to implementation of two-bit operation.

In the case of storing information only in one of the two charge holding portions 161, 162, or in the case of using these two charge holding portions 161, 162 in the same storing condition, a pinchoff point does not necessarily need to be formed in read operations.

Although not shown in FIG. 11, a well region (P type well in the case of N-channel devices) is preferably formed at the surface of the semiconductor substrate 111. Forming the well region facilitates control of electrical characteristics (withstand voltage, junction capacitance, and short channel effect) while maintaining the impurity concentration of the channel region optimum for memory operations (rewrite operation and read operation).

From the viewpoint of improving the memory holding characteristic, the charge holding portion 161, 162 preferably incorporates a charge holding film having a function of holding electric charges, and an insulating film. This embodiment uses the silicon nitride film 142 as a charge holding film having levels for trapping electric charges, and the silicon oxide films 141, 143 as insulating films having a function of preventing the electric charges stored in the charge holding film from dissipating. The charge holding portion 161, 162 having the charge holding film and the insulating film makes it possible to prevent electric charges from dissipating and to thereby improve the holding characteristics. Further, compared with the charge holding portion 161, 162 composed of only a charge holding film, it becomes possible to appropriately decrease the volume of the charge holding film. Appropriate decrease of the volume of the charge holding film makes it possible to restrain the movement of electric charges in the charge holding film and thereby to control occurrence of changes in characteristics due to the movement of electric charges during the memory holding.

Also, it is preferable that the charge holding portion 161, 162 contains a charge holding film having a portion disposed approximately parallel to the top surface of the gate insulating film 114. In other words, it is preferable that an upper face of the charge holding film in the charge holding portion 161, 162 is disposed so as to be at a constant distance from an upper face of the gate insulating film 114.

More particularly, as shown in FIG. 15, a charge holding film 142b (e.g., silicon nitride film) in the charge holding portion 162 has a face approximately parallel to the surface of the gate insulating film 114. In other words, the charge holding film 142b is preferably formed to have an equal distance from the level corresponding to the surface of the gate insulating film 114. The constitution of the charge holding film 142b corresponds to the region 181 that is a portion having a face approximately parallel to the gate insulating film 114 in the silicon nitride film 142 shown in FIG. 12.

The presence of the charge holding film 142b approximately parallel to the surface of the gate insulating film 114 in the charge holding portion 162 makes it possible to effectively control formation of an inversion layer in the offset region 171 with use of an electric charge amount stored in the charge holding film 142b, thereby enabling increase of memory effect. Also, by placing the charge holding film 142b approximately parallel to the surface of the gate insulating film 114, change of the memory effect is kept relatively small even when there are variations in the offset amount (W1), thus enabling restraint of memory effect variation. In addition, movement of electric charges toward the upper side of the charge holding film 142b is controlled, and therefore characteristics change due to the movement of electric charges during memory holding can be restrained.

Furthermore, the charge holding portion 161, 162 preferably contains an insulating film (e.g., a portion of the silicon oxide film 144 on the offset region 171) that separates the charge holding film 142b approximately parallel to the surface of the gate insulating film 114 from the channel region (or the well region). This insulating film corresponds to portions in the silicon oxide films 141, 143 approximately parallel to the surface of the gate insulating film 114 in the memory device shown in FIG. 11. This insulating film restrains dissipation of the electric charges stored in the charge holding film 142b, thereby contributing to obtaining a memory device (semiconductor storage device) with good holding characteristics.

It is noted that controlling the film thickness of the charge holding film 142b as well as controlling the film thickness of the insulating film under the charge holding film 142b (a portion of the silicon oxide film 144 on the offset region 171) to be constant makes it possible to keep the distance from the surface of the semiconductor substrate to the electric charges stored in the charge holding film 142b approximately constant. More particularly, the distance from the surface of the semiconductor substrate 111 to the electric charges stored in the charge holding film 142b can be controlled to be within the range of from a minimum film thickness value of the insulating film under the charge holding film 142b to the sum of a maximum film thickness of the insulating film under the charge holding film 142b and a maximum film thickness of the charge holding film 142b. Consequently, the density of electric lines of force generated by the electric charges stored in the charge holding film 142b becomes roughly controllable, and therefore variation in the degree of memory effect of the memory devices can be minimized.

(Embodiment 3)

In a memory device of Embodiment 3, a silicon nitride film (charge holding film) 142 in the charge holding portion 161, 162 has an approximately uniform film thickness as shown in FIG. 16. Further, the charge holding film 142 includes a portion (region 181) disposed approximately parallel to a top surface of the gate insulating film 114 and a portion (region 182) approximately parallel to a side surface of the gate electrode 117.

When a positive voltage is applied to the gate electrode 117, an electric line of force in the charge holding portion 162 passes the silicon nitride film 142 totally twice (through the first portion 181 and the second portion 182) as shown with arrow 183. It is noted that when a negative voltage is applied to the gate electrode 117, the direction of the electric line of force is reversed.

Herein, a relative permittivity, or dielectric constant of the silicon nitride film 142 is approx. 6, while a dielectric constant of silicon oxide films 141, 143 is approx. 4. Eventually, in the charge holding portion 161, 162, an effective dielectric constant in the direction of electric line of force 183 becomes larger than that in the case where only the charge holding film 142a corresponding to the region 181 is present, which makes it possible to decrease the potential difference between both edges of the electric line of force. More specifically, a large part of the voltage applied to the gate electrode 117 is used to reinforce electric fields in the offset region 171.

In the memory device, electric charges are injected into the silicon nitride film 142 in rewrite operations because generated electric charges are pulled by electric fields in the offset region 171. As a consequence of the charge holding film 142 including the region 182, increased electric charges are injected into the charge holding portion 162 in rewrite operations, thereby increasing a rewrite speed.

In the case where the silicon oxide film 143 is replaced with a silicon nitride film, more specifically, in the case where the upper surface of the silicon nitride film (charge holding film) is not at a constant height relative to the surface of the gate insulating film 114, movement of electric charges toward upper side of the silicon nitride film becomes outstanding, and holding characteristics are deteriorated.

Instead of silicon nitride, the charge holding film is more preferably formed from high-dielectric substances such as hafnium oxide havin