Semiconductor device and method for manufacturing the same

By adopting a charge storage layer with a multi-layer film structure in a nonvolatile memory cell, multiple trap levels are formed using insulating films of hafnium, silicon and oxygen and the insertion layer, combining the interface between the aluminum oxide film and the hafnium silicate film, the solution is solved The problem of charge leakage in the high dielectric constant film is improved, and the retention characteristics and reliability of the memory cell are improved.

CN110729301BActive Publication Date: 2024-11-08RENESAS ELECTRONICS CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN201910521429.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2018-07-17
Filing Date
2019-06-17
Publication Date
2024-11-08
Estimated Expiration
2039-06-17

Smart Images

  • Figure CN110729301B_ABST
    Figure CN110729301B_ABST
Patent Text Reader

Abstract

Embodiments of the present disclosure relate to semiconductor devices and methods for manufacturing the same. A memory cell as a nonvolatile memory cell includes a gate dielectric film having a charge storage layer capable of holding charge, and a memory gate electrode formed on the gate dielectric film. The charge storage layer includes an insulating film including hafnium, silicon, and an oxygen insertion layer formed on the insulating film and including aluminum, and an insulating film formed on the insertion layer and including hafnium, silicon, and oxygen.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] The disclosure of Japanese Patent Application No. 2018-134089 filed on Jul. 17, 2018 including the specification, drawings and abstract is incorporated herein by reference in its entirety. Technical Field

[0003] The present invention relates to a semiconductor device and a method for manufacturing the same, and more particularly, to a technology for a semiconductor device having a nonvolatile memory cell. Background Art

[0004] Flash memory or electrically erasable programmable read-only memory (EEPROM) is widely used as an electrically programmable and erasable non-volatile memory. Non-volatile memory cells include memory cells called metal oxide nitride oxide semiconductor (MONOS) type transistors, in which a charge storage layer having a trap insulating film such as a silicon nitride film is provided under a gate electrode. By injecting and releasing charges into and from the charge storage layer, the threshold of the transistor is shifted, thereby enabling the transistor to be used as a non-volatile memory cell. In recent years, non-volatile memory cells using a high dielectric constant film such as a hafnium oxide film instead of a silicon nitride film as a charge storage layer have been developed.

[0005] For example, Japanese Unexamined Patent Application Publication No. 2015 / 53474 discloses a nonvolatile memory cell using a hafnium silicate film as a charge storage layer. Summary of the Invention

[0006] In a nonvolatile memory cell using a high dielectric constant film, it is desired to improve reliability such as retention characteristics.

[0007] Other objects and novel features will become apparent from the description of the specification and the accompanying drawings.

[0008] Typical embodiments among the embodiments disclosed in this application are briefly described below.

[0009] A semiconductor device according to an embodiment includes a nonvolatile memory cell having a first gate dielectric film formed on a semiconductor substrate and including a charge storage layer capable of retaining charge, and a first gate electrode formed on the first gate dielectric film. The charge storage layer includes a first insulating film containing hafnium, silicon, and oxygen, a first insertion layer formed on the first insulating film and made of a material different from that of the first insulating film and containing aluminum, and a second insulating film formed on the first insertion layer and made of a material different from that of the first insertion layer and containing hafnium, silicon, and oxygen.

[0010] In addition, a semiconductor device according to an embodiment includes a nonvolatile memory cell having a charge storage layer capable of retaining charge. Herein, the charge storage layer includes a first insulating layer containing hafnium, silicon, and oxygen. At least one insertion layer containing a metal different from hafnium is formed in the first insulating film.

[0011] According to the embodiment, the reliability of a semiconductor device can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 is a cross-sectional view showing the semiconductor device of the first embodiment.

[0013] Figure 2 is an equivalent circuit diagram of the memory cell of the first embodiment.

[0014] Figure 3 is a table showing an example of conditions for applying a voltage to each portion of a selected memory cell during “write”, “erase”, and “read”.

[0015] Figure 4 is an enlarged cross-sectional view of a main portion of the semiconductor device of the first embodiment.

[0016] Figure 5 : is a graph showing the experimental results of the inventors of the present application.

[0017] Figure 6 is a schematic diagram showing the trap level existing in the charge storage layer.

[0018] Figure 7A and Figure 7B : is a graph showing the experimental results of the inventors of the present application.

[0019] Figure 8 : is a graph showing the experimental results of the inventors of the present application.

[0020] Figure 9 are cross-sectional views for describing the manufacturing process of the semiconductor device of the first embodiment.

[0021] Figure 10 It shows Figure 9 Cross-sectional view of the subsequent manufacturing process.

[0022] Figure 11 It shows Figure 10 Cross-sectional view of the subsequent manufacturing process.

[0023] Figure 12 It shows Figure 11 Cross-sectional view of the subsequent manufacturing process.

[0024] Figure 13 It shows Figure 12 Cross-sectional view of the subsequent manufacturing process.

[0025] Figure 14 is an enlarged cross-sectional view of a main portion of the semiconductor device of the second embodiment.

[0026] Figure 15 is an enlarged cross-sectional view of a main portion of the semiconductor device of the third embodiment.

[0027] Figure 16 yes Figure 15 An enlarged cross-sectional view of a portion of FIG.

[0028] Figure 17 is a cross-sectional view showing a semiconductor device according to a fourth embodiment.

[0029] Figure 18 is an equivalent circuit diagram of the memory cell of the fourth embodiment.

[0030] Figure 19 is a table showing an example of conditions for applying a voltage to each portion of a selected memory cell during “write”, “erase”, and “read”.

[0031] Figure 20 1 and 2 are cross-sectional views showing a manufacturing process of the semiconductor device of the fourth embodiment.

[0032] Figure 21 It shows Figure 20 Cross-sectional view of the subsequent manufacturing process.

[0033] Figure 22 It shows Figure 21 Cross-sectional view of the subsequent manufacturing process.

[0034] Figure 23 It shows Figure 22 Cross-sectional view of the subsequent manufacturing process.

[0035] Figure 24 It shows Figure 23 Cross-sectional view of the subsequent manufacturing process.

[0036] Figure 25 It shows Figure 24 Cross-sectional view of the subsequent manufacturing process.

[0037] Figure 26 is a cross-sectional view showing a modified semiconductor device.

[0038] Figure 27 is an enlarged cross-sectional view of a main portion of a semiconductor device according to a first comparative example.

[0039] Figure 28 is an enlarged cross-sectional view of a main portion of a semiconductor device according to a second comparative example.

[0040] Figure 29 is an enlarged cross-sectional view of a main portion of a semiconductor device according to a third comparative example. DETAILED DESCRIPTION

[0041] In the following embodiments, when for the sake of convenience, the description will be divided into multiple parts or embodiments, but unless otherwise specified, these parts are not independent of each other, and one is related to a modification, information, supplementary explanation, etc. of part or all of another. In the following embodiments, the number of elements, etc. (including the number of elements, numerical values, quantities, ranges, etc.) is not limited to a specific number, but may be not less than or equal to a specific number, except for the case where the number is specifically specified and is clearly limited to the specific number in principle. Further, in the following embodiments, it goes without saying that the constituent elements (including element steps, etc.) are not necessarily necessary, except for the case where they are specifically specified and the case where they are considered to be necessary in principle. Similarly, in the following embodiments, when it comes to the shape, positional relationship, etc. of the parts, etc., it is assumed that the shape, etc. are basically close to or similar to the shape, etc., except for the case where they are specifically specified and the case where they are considered to be obvious in principle, etc. The same applies to the above-mentioned numerical values ​​and ranges.

[0042] Hereinafter, embodiments are described in detail based on the accompanying drawings. In all drawings used to explain the embodiments, components having the same function are represented by the same reference numerals, and their repeated descriptions are omitted. In the following embodiments, unless particularly necessary, the description of the same or similar components will not be repeated in principle.

[0043] In the drawings used in the embodiments, hatching may be omitted to make the drawings easier to see.

[0044] First embodiment

[0045] A semiconductor device including the memory cell MC1 as the nonvolatile memory cell in the present embodiment is described with reference to the drawings. Figure 1 is a cross-sectional view of the semiconductor device of this embodiment, and Figure 2 yes Figure 1 Equivalent circuit diagram of a memory cell.

[0046] The memory cell MC1 of the present embodiment is an n-type transistor including a charge storage layer CSL capable of holding charges in a gate dielectric film MZ, and the charge storage layer CSL includes a trap insulating film.

[0047] like Figure 1As shown, a p-type well region PW is formed in a semiconductor substrate SB, and a memory cell MC1 is formed on the well region PW. The semiconductor substrate SB is made of p-type single crystal silicon (Si) having a resistivity of, for example, 1 Ωcm to 10 Ωcm.

[0048] A gate dielectric film MZ is formed on the semiconductor substrate SB (i.e., on the well region PW1), and a memory gate electrode MG is formed on the gate dielectric film MZ. The memory gate electrode MG is, for example, a conductive film such as a polysilicon film into which n-type impurities have been introduced. Note that in order to ensure hole tunneling current during erasure, a polysilicon film into which p-type impurities have been introduced or a polysilicon film without impurities can be used for the memory gate electrode MG. The memory gate electrode MG can be, for example, a metal film such as a titanium nitride film, an aluminum film, or a tungsten film, or a stacked film of these metal films.

[0049] exist Figure 1 In the figure, the gate dielectric film MZ is shown as a single-layer insulating film to make it easier to see the drawings, but in fact, as will be described later, Figure 4 As described, the gate dielectric film MZ is a multilayer film having the insulating film BT, the insulating film HSO1 , the insertion layer AL1 , the insulating film HSO2 , and the insulating film TP.

[0050] The sidewall spacer SW is formed on the side surface of the memory gate electrode MG. The sidewall spacer SW is formed of, for example, a stacked film of a silicon oxide film and a silicon nitride film.

[0051] Extension region EXS and extension region EXD, which serve as low-concentration n-type impurity regions, are formed in well region PW below sidewall spacer SW. Diffusion region MS, an n-type impurity region with a higher concentration than extension region EXS, and diffusion region MD, an n-type impurity region with a higher concentration than extension region EXD, are formed in well region PW at a position aligned with sidewall spacer SW. Extension region EXS and diffusion region MS are connected to each other and each constitute a portion of the source region of memory cell MC1. Extension region EXD and diffusion region MD are connected to each other and each constitute a portion of the drain region of memory cell MC1.

[0052] A silicide layer SI is formed on the memory gate electrode MG, the diffusion region MS, and the diffusion region MD using, for example, cobalt silicide (CoSi2), nickel silicide (NiSi), or nickel silicide (NiPtSi). The silicide layer SI is formed mainly to reduce the contact resistance with the plug PG, which will be described later.

[0053] An interlayer insulating film IL1 is formed over memory cell MC1. Multiple contact holes are formed in interlayer insulating film IL1, and multiple plugs PG are formed in the multiple contact holes. Plugs PG are formed from a barrier metal film, for example, a titanium film, a titanium nitride film, or a stacked film thereof, and a conductive film primarily made of, for example, tungsten. Plugs PG are electrically connected to diffusion region MS or diffusion region MD via silicide layer SI. Although not shown, plugs PG electrically connected to memory gate electrode MG are also present in interlayer insulating film IL1.

[0054] An interlayer insulating film IL2 is formed over the plug PG and the interlayer insulating film IL1. A trench for wiring is formed in the interlayer insulating film IL2, and a wiring M1, for example, comprising a conductive film primarily made of copper, is embedded in the trench. The wiring M1 is electrically connected to the plug PG. This wiring has a so-called damascene structure. Multiple layers of wiring and an interlayer insulating film are also formed over the wiring M1, but illustration and description thereof are omitted herein.

[0055] Operation of memory cell MC1

[0056] See also Figure 2 and Figure 3 Hereinafter, an exemplary operation of the memory cell MC1 as a nonvolatile memory cell will be described. The memory cell MC1 described herein is a selected memory cell among a plurality of memory cells MC1 present in the semiconductor device.

[0057] Figure 2 is an equivalent circuit diagram of the memory cell MC1. Figure 3 is a table showing an example of conditions for applying voltages to respective portions of the memory cell MC1 at the time of "writing", "erasing", and "reading". Figure 3 In the table, the voltage Vd applied to the diffusion region MD serving as the drain region, the voltage Vmg applied to the memory gate electrode MG, the voltage Vs applied to the diffusion region MS serving as the source region, and the voltage Vb applied to the well region PW are described at each of "writing", "erasing" and "reading".

[0058] Notice, Figure 3 The contents shown in the table are suitable examples of voltage application conditions and are not limited thereto and can be variously changed as needed. In this embodiment, the injection of electrons from the well region PW into the charge storage layer CSL is defined as "writing", and the emission of electrons from the charge storage layer CSL into the well region PW is defined as "erasing".

[0059] The write operation is performed by the Fowler Nordheim tunneling (FN-tunneling) method. Figure 3The voltages shown in the "Write" column in the figure are applied to the various portions of memory cell MC1 to be written, and electrons are injected from the well region PW into the charge storage layer CSL of memory cell MC1 to perform writing. The injected electrons are trapped in the trap levels in the charge storage layer CSL, resulting in a rise in the threshold voltage of the memory transistor having the memory gate electrode MG. In other words, the memory transistor is in a write state.

[0060] The erase operation is performed by the FN tunneling method. Figure 3 The voltage shown in the "Erase" column in FIG is applied to the respective portions of the memory cell MC1 to be erased, and electrons in the charge storage layer CSL are emitted to the well region PW. As a result, the threshold voltage of the memory transistor is lowered. In other words, the memory transistor is in an erased state.

[0061] In a read operation, for example, Figure 3 The voltages shown in the "Read" column in FIG. 3 are applied to the respective portions of the memory cell MC1 to be read. By setting the voltage Vmg applied to the memory gate electrode MG during reading to a value between the threshold voltage of the memory transistor in the write state and the threshold voltage of the memory transistor in the erase state, the write state and the erase state can be distinguished.

[0062] Detailed structure of the gate dielectric film MZ

[0063] Figure 4 yes Figure 1 : is an enlarged cross-sectional view of a main portion of the semiconductor device shown in , and is a cross-sectional view showing a detailed structure of the gate dielectric film MZ.

[0064] The gate dielectric film MZ is interposed between the semiconductor substrate SB (well region PW) and the memory gate electrode MG, serving as the gate dielectric film of the memory cell MC1. It is a stacked film having a charge storage layer CSL therein. Specifically, the gate dielectric film MZ is formed of a stacked film comprising an insulating film BT formed on the semiconductor substrate SB, an insulating film HSO1 formed on the insulating film BT, an insertion layer AL1 formed on the insulating film HSO1, an insulating film HSO2 formed on the insertion layer AL1, and an insulating film TP formed on the insulating film HSO2.

[0065] The insulating film (bottom insulating film) BT is, for example, a silicon oxide film or a silicon oxynitride film, and has a thickness of, for example, 2 nm to 5 nm.

[0066] The insulating film HSO1 is a film having a dielectric constant greater than that of the silicon nitride film, is a metal oxide film, and has a thickness of, for example, 5 nm to 9 nm. Specifically, the insulating film HSO1 is a film containing hafnium (Hf), silicon (Si), and oxygen (O), and is preferably a hafnium silicate film such as Hf x Si 1-x O2 (0 < x < 1) film. In order to increase the density of trap levels in the insulating film HSO1, the composition ratio of the Hf x Si 1-x O2 film is preferably 0.6 < x < 1, and most preferably x = 0.8.

[0067] The insertion layer AL1 is a film containing a metal different from hafnium, is a film containing aluminum (Al), and has a thickness of, for example, 1 nm to 4 nm. Specifically, the insertion layer AL1 is an aluminum (Al) film, an aluminum nitride (AlN) film, an aluminum carbide (AlC) film, an aluminum silicate (AlSiO) film, or an aluminum oxide film. In this embodiment, a metal oxide film containing aluminum (Al) and oxygen (O) is preferably used as the insertion layer AL1, and most preferably an aluminum oxide film such as an Al2O3 film.

[0068] The insulating film HSO2 is a film made of a material different from that of the insertion layer AL1, is a film containing hafnium (Hf), silicon (Si), and oxygen (O), and has a thickness of, for example, 5 nm to 9 nm. The material of the insulating film HSO2 is preferably the same as that of the insulating film HSO1.

[0069] As described above, in this embodiment, a film containing aluminum (insertion layer AL1) is formed in the film containing hafnium, silicon, and oxygen (insulating film HSO1 and insulating film HSO2) in the charge storage layer CSL.

[0070] The insulating film (top insulating film) TP is a film having a dielectric constant greater than that of the silicon nitride film, is a metal oxide film made of a material different from that of the insulating film HSO1 and the insulating film HSO2, and has a thickness of, for example, 5 nm to 12 nm. Specifically, the insulating film TP is a film containing aluminum (Al) and oxygen (O), and is preferably an aluminum oxide film, an aluminum oxynitride film, or an aluminum silicate film. As the insulating film TP, other metal oxide films can be used. For example, a metal oxide film made of an oxide of any one of titanium (Ti), zirconium (Zr), yttrium (Y), lanthanum (La), praseodymium (Pr), and lutetium (Lu) can be used as the insulating film TP1.

[0071] The insulating film TP is mainly used to prevent the charge stored in the charge storage layer CSL from escaping to the memory gate electrode MG. For this purpose, it is preferable that the thickness of the insulating film TP is greater than the thickness of the insertion layer AL1.

[0072] Incidentally, in a conventional non-volatile memory cell, an oxide-nitride-oxide (ONO) film in which a silicon oxide film, a silicon nitride film as a charge storage layer, and a silicon oxide film are stacked is known as a gate dielectric film having trap levels. When an ONO film is used, the dielectric constant is relatively low, so the equivalent oxide thickness (EOT) of the gate dielectric film increases. For this reason, it is feared that the operating voltage may be increased by increasing the EOT of the gate dielectric film. In addition, if the physical film thickness is reduced to reduce the EOT of the gate dielectric film, the retention characteristics (charge retention characteristics) may deteriorate due to charge leakage from the charge storage layer. These all reduce the reliability of the semiconductor device.

[0073] In this embodiment, the charge storage layer CSL is mainly formed of an insulating film HSO1 and an insulating film HSO2 as high-k dielectric films. The insulating film TP is also a high-k dielectric film. With these films, it is possible to increase the physical thickness of the gate dielectric film MZ while suppressing the EOT of the gate dielectric film MZ, thereby preventing deterioration of the retention characteristics due to leakage and improving the retention characteristics. In addition, since the EOT can be reduced while ensuring the physical thickness of the gate dielectric film MZ, the operating voltage and operating speed of the memory cell MC1 can be reduced while preventing deterioration of the retention characteristics due to leakage.

[0074] Herein, although there are many trap levels of the charge storage layer CSL of this embodiment in the insulating film HSO1 and the insulating film HSO2 which are Hf x Si 1-x O2 (0 < x < 1) films, there are more trap levels near the interfaces between the insulating film HSO1 and the insertion layer AL1 and between the insulating film HSO2 and the insertion layer AL1. Therefore, more trap levels can exist at positions closer to the center of the charge storage layer CSL. That is, a large number of deep trap levels can exist in the charge storage layer CSL at positions far from the lower surface of the insulating film HSO1 and far from the upper surface of the insulating film HSO2. As described later with reference to FIG. 7, many deep trap levels exist at the interface between the hafnium silicate film and the aluminum oxide film. Since such an interface can be formed in this embodiment, many deep trap levels can be formed. As a result, the charges stored in the charge storage layer CSL are less likely to escape from the charge storage layer CSL, and the retention characteristics of the memory cell MC1 can be improved.

[0075] Further, the insertion layer AL1 is a film for increasing the number of trap levels in the charge storage layer CSL. For example, when the insertion layer AL1 is an alumina film and the insulating films HSO1 and HSO2 are hafnium silicate films, the dielectric constant of the alumina film is lower than that of the hafnium silicate films. For this reason, the thickness of the insertion layer AL1 is preferably not greater than the required thickness, and the thickness of the insertion layer AL1 is preferably less than the thickness of the insulating film HSO1 and the thickness of the insulating film HSO2.

[0076] Comparison of the semiconductor devices of the first to third comparative examples with the semiconductor device of the present embodiment

[0077] Used below Figures 5 to 8 And Figures 27 to 29 The above improvements to the retention characteristics are described in detail.

[0078] Figures 27 to 29 Are enlarged cross-sectional views of the main parts of the semiconductor devices of the first to third comparative examples, Figure 6 Is a schematic diagram showing the trap levels existing inside the charge storage layer CSL, and Figure 5 Figure 7 and Figure 8 Is a graph showing the experimental results of the inventors of the present application.

[0079] In the semiconductor device of the first comparative example, as Figure 27 Shown, only the insulating film HSO1 is formed as the charge storage layer CSL, and the charge storage layer CSL does not include the insertion layer AL1 and the insulating film HSO2. The thickness of the insulating film HSO1 of the first comparative example is approximately the sum of the thickness of the insulating film HSO1, the thickness of the insertion layer AL1, and the thickness of the insulating film HSO2 in the present embodiment.

[0080] In the semiconductor device of the second comparative example, as Figure 28 Shown, the insulating film HO, the insertion layer AL1, and the insulating film HSO2 are formed as the charge storage layer CSL. The insulating film HO is not a Hf x Si 1-x O2 (0 < x < 1) film like the insulating film HSO1 of the present embodiment, but a hafnium oxide film (HfO film). The thickness of the insulating film HO of the second comparative example is substantially the same as the thickness of the insulating film HSO1 of the present embodiment.

[0081] In the semiconductor device of the third comparative example, as Figure 29 Shown, the insertion layer AL1 and the insulating film HSO2 are formed as the charge storage layer CSL, and the insulating film HSO1 is not formed. Therefore, the insertion layer AL1 of the third comparative example is in direct contact with the insulating film BT. The thickness of the insulating film HSO2 of the third comparative example is approximately the sum of the thickness of the insulating film HSO1 and the thickness of the insulating film HSO2 in the present embodiment.

[0082] Figure 5 The horizontal axis indicates a period during which the memory cell MC1 is at a high temperature of 150° C. after a write operation is performed on the memory cell MC1 . Figure 5 The vertical axis indicates a change ΔVfb in the flatband voltage, and specifically indicates a difference between the flatband voltage (Vfb) after a predetermined period of time has passed after the write operation and the flatband voltage (Vfbi) before the write operation. Figure 5 The value of ΔVfb in is a relative value of the change in the flat-band voltage. Here, measurements were performed using an HfxSi1-xO2 (x = 0.8) film applied to the insulating film HSO1 and a silicon oxynitride (SiON) film applied to the insulating film BT. Although the top insulating film TP is a single-layer film such as an aluminum oxide film, the ratio of the change ΔVfb in the flat-band voltage is the same even when the top insulating film TP is a stacked film such as the insulating films TP1-TP3, as in the third embodiment described later.

[0083] from Figure 5 As can be seen from the graphs, the amount of change ΔVfb decreases over time in the present embodiment (●), the first comparative example (□), and the second comparative example (▲), respectively. However, the amount of change ΔVfb is suppressed more significantly in the present embodiment (●) than in the first comparative example (□) and the second comparative example (▲). In other words, it can be seen that the retention characteristics are improved in the present embodiment (●) compared to the first comparative example (□) and the second comparative example (▲).

[0084] In addition, since the insertion layer AL1 is formed between the insulating film HSO1 and the insulating film HSO2 as a result of improving the retention characteristics in this embodiment (●), it can be inferred that the trap level within the charge storage layer CSL increases more in this embodiment (●) than in the first comparative example (□).

[0085] In addition, it can be inferred that by converting Hf x Si 1-x An O 2 film is applied to the insulating film HSO1 , and the trap level inside the charge storage layer CSL is increased more by applying an HfO film to the insulating film HSO1 as in the case of the second comparative example (▲).

[0086] Figure 6Schematic diagram showing the trap levels present in the charge storage layer CSL according to the structure of the first comparative example. Symbols (●) indicate deep trap levels with energies of 1.3 eV to 2.1 eV, and symbols (□) indicate shallow trap levels with energies of 0.8 eV to 1.3 eV. Distance Xa indicates the distance from the memory gate electrode MG to the vicinity of the interface between the insulating film TP and the insulating film HSO1. Distance Xb indicates the distance from the memory gate electrode MG to the interior of the insulating film HSO1. Distance Xc indicates the distance from the memory gate electrode MG to the vicinity of the interface between the insulating film HSO1 and the insulating film BT.

[0087] Figure 7A Shown are graphs obtained by deeply decomposing the trap level existing inside the charge storage layer CSL for each energy distribution using a thermally stimulated current-capacitance-voltage (TSC-CV) method. Figure 7A The vertical axis shows the change in flat-band voltage ΔVfb, while Figure 7A The horizontal axis of φ shows the electrical thickness of the gate dielectric film MZ. Here, the measurement results assuming that the thickness of the insulating film BT and the thickness of the insulating film TP are constant are shown.

[0088] Herein, the amount of change ΔVfb in the flat band voltage due to the charge Q stored in the charge storage layer CSL is expressed by the following equation (1), and this amount of change ΔVfb is proportional to the distance X from the memory gate electrode MG to the charge Q and inversely proportional to the dielectric constant k. Capacitance C is the capacitance between the charge Q and the memory gate electrode MG.

[0089] ΔVfb=Q / C=Q×X / k(1)

[0090] When the distance X is applied to Figure 6 When the distance Xa-Xc in the charge storage layer CSL is changed, the change ΔVfb is almost constant when X=Xa. When X=Xb, since the distance Xb changes inside the charge storage layer CSL film, the change ΔVfb is proportional to the integrated value of Xb (∫Xbdx). That is, the change ΔVfb is proportional to the square of Xb (Xb 2 When X=Xc, the change ΔVfb is proportional to Xc.

[0091] Therefore, if Figure 7A As shown by the dotted line in , the variation ΔVfb is substantially constant in the thickness direction in the deep trap level (●), and the variation ΔVfb substantially overlaps with the quadratic curve in the thickness direction in the shallow trap level (■).

[0092] Figure 7BThis graph shows the surface density of trap levels in the gate dielectric film MZ, obtained using the TSC-CV method. In Figure 7 , the horizontal axis represents the region near the interface between the insulating film TP and the insulating film HSO1, within the insulating film HSO1, and near the interface between the insulating film HSO1 and the insulating film BT. The vertical axis of Figure 7 represents the surface density of trap levels. Measurements were made using the structure of the first comparative example, where the insulating film BT is a silicon oxynitride film, the insulating film HSO1 is a hafnium silicate film, and the insulating film TP is an aluminum oxide film.

[0093] like Figure 7B As shown in FIG. 1 , it can be seen that a large number of deep trap levels exist at the interface between the insulating film TP as the aluminum oxide film and the insulating film HSO1 as the hafnium silicate film. Figure 7A and Figure 7B From the results, it can be seen that the distribution of trap levels inside the gate dielectric film MZ is as follows Figure 6 shown.

[0094] Applying this result, the inventors of the present application conceived that by inserting an aluminum oxide film into the hafnium silicate film, more interfaces are formed between the hafnium silicate film and the aluminum oxide film, thereby forming a large number of deep trap levels. That is, in this embodiment, since the insertion layer AL1 is formed between the insulating films HSO1 and HSO2, many deep trap levels can be present in the charge storage layer CSL.

[0095] As mentioned above, when Figure 6 The results in Figure 7 are similar to Figure 5 When the results of the present embodiment are combined and considered, although the thickness of the charge storage layer CSL of the present embodiment is substantially the same as that of the charge storage layer CSL of the first comparative example, the retention characteristics of the present embodiment are improved compared to the first comparative example. This is because in the present embodiment, by forming the insertion layer AL1 between the insulating film HSO1 and the insulating film HSO2, a greater number of deep trap levels can be caused to exist near the interface between the insulating film HSO1 and the insertion layer AL1 and near the interface between the insulating film HSO2 and the insertion layer AL1. In other words, the interface between the hafnium silicate film and the aluminum oxide film (that is, the interface where deep trap levels are easily formed) increases within the charge storage layer CSL. Therefore, a greater number of deep trap levels can exist at a position close to the center of the charge storage layer CSL. As a result, the retention characteristics can be improved, and thus the reliability of the semiconductor device can be improved.

[0096] Further, in the charge storage layer CSL of the present embodiment, the insulating film HSO1 is a film containing hafnium (Hf), silicon (Si), and oxygen (O), and is preferably a film such as Hf. x Si 1-xA hafnium silicate film such as an O2 film (0 < x < 1). As in the second comparative example, this makes it possible to increase the deep trap level inside the charge storage layer CSL and improve the retention characteristics compared to when a HfO film is applied to the insulating film HSO1.

[0097] Figure 8 The horizontal axis shows how long the memory cell MC1 has been at room temperature (26 °C) after a write operation is performed on the memory cell MC1. Similar to Figure 5 the vertical axis, Figure 8 the vertical axis indicates the change amount ΔVfb of the flat band voltage.

[0098] It can be seen from Figure 8 that in the present embodiment (●) and the third comparative example (▲) respectively, the change amount ΔVfb decreases over time, and the change amount ΔVfb is suppressed more in the present embodiment (●) than in the third comparative example (▲). In other words, it can be seen that the retention characteristics are improved in the present embodiment (●) compared to the third comparative example (▲). Additionally, although Figure 8 the graph is obtained from the memory cell MC1 at room temperature, it is obvious that when the memory cell MC1 is at a high temperature of 150 °C or higher for example, the difference between the change amount ΔVfb of the present embodiment (●) and the change amount ΔVfb of the third comparative example (▲) becomes larger.

[0099] That is, as in the third comparative example, when the insulating film HSO1 is not formed and the insertion layer AL1 is formed to be in direct contact with the insulating film BT, the retention characteristics deteriorate. Therefore, it is preferable that the insertion layer AL1 is formed between the insulating film HSO1 and the insulating film HSO2, and the insulating film HSO1 is in direct contact with the insulating film BT, such that the insertion layer AL1 is not in direct contact with the insulating film BT, as in the present embodiment. As a result, the retention characteristics can be improved.

[0100] Method for manufacturing the memory cell MC1 <0OO0312>The following describes the manufacturing method of the semiconductor device of the present embodiment with reference to Figures 9 to 13 which.

[0102] First, as Figure 9 shown, for example, boron (B) or boron difluoride (BF2) is introduced into the semiconductor substrate SB by lithography and ion implantation to form a p-type well region PW.

[0103] Figure 10 shows the step of forming the gate dielectric film MZ. In Figure 10In the following figures, the gate dielectric film MZ is shown as a single-layer film for clarity, but in fact, the gate dielectric film MZ is a multilayer film having an insulating film BT, an insulating film HSO1, an insertion layer AL1, an insulating film HSO2, and an insulating film TP. The multilayer film is shown in an enlarged view. Figure 10 The area enclosed by the dotted line.

[0104] First, an insulating film BT, such as a silicon oxide film, is formed on a semiconductor substrate SB by, for example, in-situ steam generation (ISSG) oxidation. The thickness of the insulating film BT is, for example, 2 nm to 5 nm. Thereafter, an NO treatment or plasma nitridation treatment may be performed to nitride the silicon oxide film to form a silicon nitride oxide film.

[0105] Next, a charge storage layer CSL having the insulating film HSO1 , the insertion layer AL1 , and the insulating film HSO2 is formed on the semiconductor substrate SB via the insulating film BT.

[0106] For example, an insulating film HSO1 containing hafnium (Hf), silicon (Si), and oxygen (O) is formed on the insulating film BT by low-pressure chemical vapor deposition (LPCVD) or atomic layer deposition (ALD). The thickness of the insulating film HSO1 is, for example, 5 nm to 9 nm. The film formation temperature during the formation of the insulating film HSO1 is, for example, 200° C. to 500° C.

[0107] Next, an insertion layer AL1 containing aluminum (Al), a metal different from hafnium, is formed on the insulating film HSO1 by LPCVD or ALD. The thickness of the insertion layer AL1 is, for example, 1 nm to 4 nm. The film formation temperature when forming the insertion layer AL1 is, for example, 200°C to 500°C.

[0108] Next, an insulating film HSO2 containing hafnium (Hf), silicon (Si), and oxygen (O) is formed on the insertion layer AL1 using the LPCVD method or the ALD method. The insulating film HSO2 is made of a material different from that of the insertion layer AL1 and the same material as that of the insulating film HSO1, and has a thickness of, for example, 5 nm to 9 nm. The film formation temperature when forming the insulating film HSO2 is, for example, 200°C to 500°C.

[0109] Next, an insulating film TP containing aluminum (Al) and oxygen (O) is formed on the charge storage layer CSL by an LPCVD method or an ALD method. The insulating film TP is made of a material different from that of the insulating film HSO2, preferably an aluminum oxide film, an aluminum oxynitride film, or an aluminum silicate film, and has a thickness of, for example, 5 nm to 10 nm. The film formation temperature when forming the insulating film TP is, for example, 200°C to 500°C.

[0110] Next, a heat treatment is performed at, for example, 800°C to 1050°C mainly to crystallize the insulating film HSO1, the insertion layer AL1, the insulating film HSO2, and the insulating film TP. Through this heat treatment, these insulating films change from amorphous films to polycrystalline films. Here, when the insulating film HSO1 and the insulating film HSO2 are Hf x Si 1-x O2 (0.9 ≤ x < 1) films, the temperature of the heat treatment is 800°C or higher and lower than 975°C, and when the insulating film HSO1 and the insulating film HSO2 are Hf x Si 1-x O2 (0 < x < 0.9) films, the temperature of the heat treatment is 975°C or higher and 1050°C or lower. By adjusting the temperature of the heat treatment in this way, the insulating film HSO1 and the insulating film HSO2 can be appropriately crystallized.

[0111] Figure 11 The steps of forming the memory gate electrode MG are shown.

[0112] First, a conductive film such as a polysilicon film is formed on the gate dielectric film MZ by, for example, the LPCVD method. Next, an n-type impurity is introduced into the polysilicon film by photolithography and ion implantation. The conductive film can be a polysilicon film into which a p-type impurity is introduced or a polysilicon film into which no impurity is introduced. The conductive film can be, for example, a metal film such as a titanium nitride film, an aluminum film, or a tungsten film, or a laminated film of these metal films.

[0113] Next, the conductive film is patterned by photolithography and etching to form the memory gate electrode MG. After that, the gate dielectric film MZ exposed from the memory gate electrode MG is removed by dry etching and wet etching.

[0114] Figure 12 The steps of forming the extension region EXS and the extension region EXD are shown.

[0115] For example, arsenic (As) or phosphorus (P) is introduced into the well region PW adjacent to the memory gate electrode MG by photolithography and ion implantation to form the extension region EXS and the extension region EXD as n-type impurity regions. The extension region EXS forms a part of the source region of the memory cell MC1, and the extension region EXD forms a part of the drain region of the memory cell MC1.

[0116] Figure 13 The steps of forming the sidewall spacer SW, the diffusion region MS, the diffusion region MD, and the silicide layer SI are shown. <0000First, an insulating film formed of, for example, a silicon nitride film is formed by, for example, an LPCVD method to cover the memory gate electrode MG. Next, the insulating film is anisotropically etched to form sidewall spacers SW on the side surfaces of the memory gate electrode MG. Note that the insulating film forming the sidewall spacers SW may be a silicon oxide film and a silicon nitride film formed on the silicon oxide film.

[0118] Next, arsenic (As) or phosphorus (P), for example, is introduced into well region PW by photolithography and ion implantation using sidewall spacers SW as a mask, thereby forming diffusion region MS and diffusion region MD, which serve as n-type impurity regions. Diffusion region MS has a higher impurity concentration than extension region EXS, is connected to extension region EXS, and forms part of the source region of memory cell MC1. Diffusion region MD has a higher impurity concentration than extension region EXD, is connected to extension region EXD, and forms part of the drain region of memory cell MC1.

[0119] Next, a low-resistance silicide layer SI is formed on the upper surface of each of the diffusion region MS, the diffusion region MS, and the memory gate electrode MG by a salicide (self-aligned polycide) technique.

[0120] Specifically, the silicide layer SI can be formed as follows. First, a metal film for forming the silicide layer SI is formed to cover the sidewall spacers SW, the diffusion region MS, the diffusion region MS, and the memory gate electrode MG. The metal film is made of, for example, cobalt, nickel, or a nickel-platinum alloy. Next, a first heat treatment is performed on the semiconductor substrate SB at, for example, 300°C to 500°C, and a second heat treatment is performed at, for example, 600°C to 700°C, so that the materials included in the diffusion region MS, the diffusion region MD, and the memory gate electrode MG react with the metal film. As a result, a silicide layer SI is formed on the upper surface of each of the diffusion region MS, the diffusion region MS, and the memory gate electrode MG. Thereafter, the unreacted metal film is removed.

[0121] As described above, the memory cell MC1 of this embodiment is formed.

[0122] exist Figure 13 After the step of forming the interlayer insulating film IL1, the plug PG, the interlayer insulating film IL2 and the wiring M1, thereby manufacturing Figure 1 The semiconductor device shown.

[0123] First, an interlayer insulating film IL1 is formed to cover the memory cell MC1. As the interlayer insulating film IL1, a single silicon oxide film, a stacked film of a silicon nitride film and a thick silicon oxide film formed thereon, or the like can be used. After the interlayer insulating film IL1 is formed, the upper surface of the interlayer insulating film IL1 can be polished by a chemical mechanical polishing (CMP) method as needed.

[0124] Next, contact holes are formed in the interlayer insulating film IL1 by photolithography, dry etching, or the like, and a conductive film primarily made of tungsten or the like is buried in the contact holes, thereby forming a plurality of plugs PG in the interlayer insulating film IL1. The plugs PG are connected to the diffusion regions MS and MD via the silicide layer SI. The memory gate electrode MG is also connected to the plugs PG, but this is not shown in this embodiment.

[0125] Next, an interlayer insulating film IL2 is formed on the interlayer insulating film IL1 in which the plug PG is embedded. After forming a trench for wiring in the interlayer insulating film IL2, a conductive film containing, for example, copper as a main component is embedded in the trench for wiring, thereby forming a wiring M1 connected to the plug PG in the interlayer insulating film IL2. The structure of the wiring M1 is called a damascene wiring structure.

[0126] Thereafter, a second wiring layer and subsequent wiring layers are formed by a dual damascene method or the like, but their description and illustration are omitted herein. The wiring layer above the wiring M1 and the wiring M1 are not limited to the damascene wiring structure and may be formed by patterning a tungsten film or an aluminum film, for example.

[0127] Second embodiment

[0128] References below Figure 14 A semiconductor device according to a second embodiment will be described. In the following description, differences from the first embodiment will be mainly explained. Figure 14 is an enlarged cross-sectional view of a main portion of the semiconductor device of the second embodiment, and is a cross-sectional view showing a detailed configuration of the gate dielectric film MZ.

[0129] In the first embodiment, a single aluminum oxide film (insulation layer AL1) is formed in the hafnium silicate film (insulation film HSO1 and insulation film HSO2) inside the charge storage layer CSL. That is, the insertion layer AL1 is formed between the insulation film HSO1 and insulation film HSO2.

[0130] like Figure 14 As shown, in the second embodiment, the charge storage layer CSL has the insulating films HSO1 to HSO3, the insertion layer AL1 between the insulating film HSO1 and the insulating film HSO2, and the insertion layer AL2 between the insulating film HSO2 and the insulating film HSO3.

[0131] The insulating film HSO3 is made of the same material as the insulating film HSO1 or the insulating film HSO2, and is a hafnium silicate film, etc. The insertion layer AL2 is a film containing a metal other than hafnium, a film made of the same material as the insertion layer AL1, an aluminum oxide film, etc. The method for forming the insulating film HSO3 is the same as the method for forming the insulating film HSO1, and the method for forming the insertion layer AL2 is the same as the method for forming the insertion layer AL1.

[0132] As described above, in the second embodiment, a film containing two aluminum layers (insulation layer AL1 and insertion layer AL2) is formed within a film containing hafnium, silicon, and oxygen (insulation films HSO1 to HSO3). Consequently, the interface between the hafnium silicate film and the aluminum oxide film is doubled compared to the first embodiment. Consequently, more trap levels can be present in the charge storage layer CSL. Consequently, the retention characteristics of the memory cell MC1 can be further improved.

[0133] In the second embodiment, a film including two aluminum layers (the insertion layer AL1 and the insertion layer AL2 ) is exemplified, however, a film including three or more aluminum layers may be formed.

[0134] For example, when the thickness of the insertion layer AL1 is set to 1 nm to 4 nm as in the second embodiment, the charge storage layer CSL has a maximum of four aluminum oxide films and a maximum of five hafnium silicate films. In this case, the number of interfaces between the hafnium silicate films and the aluminum oxide films in the charge storage layer CSL is eight.

[0135] Note that the charge storage layer CSL of the second embodiment has more films than the first embodiment. However, if the total thickness of the charge storage layer CSL of the stacked films becomes too large, the EOT of the gate dielectric film MZ increases. Therefore, it is preferable to adjust the thickness of each of the insulating films HSO1 to HSO3, the insertion layer AL1, and the insertion layer AL2 so that the thickness of the charge storage layer CSL of the second embodiment is approximately the same as that of the charge storage layer CSL of the first embodiment.

[0136] Third embodiment

[0137] See below Figure 15 and Figure 16 A semiconductor device according to a third embodiment will be described. In the following description, differences from the first embodiment will be mainly explained. Figure 15 is an enlarged cross-sectional view of a main portion of the semiconductor device of the third embodiment, and is a cross-sectional view showing a detailed configuration of the gate dielectric film MZ. Figure 16 4 is a cross-sectional view showing a detailed configuration of the insulating film TP by enlarging a portion of the gate dielectric film MZ.

[0138] In the first embodiment, the top insulating film TP is a single-layer film formed of an aluminum oxide film or the like.

[0139] like Figure 15 As shown, in the third embodiment, the top insulating film TP is a stacked film having an insulating film TP1 formed on the charge storage layer CSL, an insulating film TP2 formed on the insulating film TP1, and an insulating film TP3 formed on the insulating film TP2.

[0140] The insulating film TP1 is a film having a dielectric constant greater than that of the silicon nitride film, and is a metal oxide film made of a material different from that of the insulating film HSO2, and has a thickness of, for example, 2 nm to 5 nm. Specifically, the insulating film TP1 is a film containing aluminum (Al) and oxygen (O), preferably an aluminum oxide film, an aluminum oxynitride film, or an aluminum silicate film, more preferably an Al2O3 film. As the insulating film TP1, other metal oxide films can be used. For example, a metal oxide film made of an oxide of any of titanium (Ti), zirconium (Zr), yttrium (Y), lanthanum (La), praseodymium (Pr), and lutetium (Lu) can be used as the insulating film TP1.

[0141] The insulating film TP2 is a film made of a material different from that of the insulating film TP1, and is a silicon oxide film, a silicon nitride oxide film, or a silicon nitride film, and has a thickness of, for example, 1 nm to 2 nm. This insulating film TP2 can be formed by, for example, LPCVD or ALD.

[0142] The insulating film TP3 is a film made of the same material as that of the insulating film TP1 and has a thickness of, for example, 2 nm to 5 nm. The insulating film TP1 and the insulating film TP3 can be formed in the same manner as the insulating film TP of the first embodiment.

[0143] The insulating film TP1 and the insulating film TP3 are mainly used to prevent the charge stored in the charge storage layer CSL from escaping to the memory gate electrode MG. Therefore, the insulating film TP1 and the insulating film TP3 are preferably insulating films having a band gap larger than the band gap of the insulating film constituting the insulating film HSO2, and preferably have a thickness larger than that of the insertion layer AL1.

[0144] In the third embodiment, the insulating film TP2 is formed between the insulating film TP1 and the insulating film TP3. Therefore, the charge stored in the charge storage layer CSL can be prevented from easily escaping to the memory gate electrode MG via the insulating film TP, and the retention characteristics of the memory cell MC1 can be improved. The reason for this will be explained below.

[0145] In the first embodiment, a single-layer film made of an aluminum oxide film, such as the insulating film TP, is formed between the charge storage layer CSL and the memory gate electrode MG. If large crystal grains form within the insulating film TP, grain boundaries form around the grains, connecting the charge storage layer CSL and the memory gate electrode MG. Consequently, the grain boundaries become leakage paths, and charge stored in the charge storage layer CSL may leak to the memory gate electrode MG.

[0146] The insulating film TP2 is mainly provided to divide the leakage path. Figure 16 As shown, the plurality of crystal grains GR1 constituting the insulating film TP1 and the plurality of crystal grains GR2 constituting the insulating film TP3 are separated by the insulating film TP. Since the insulating film TP1 and the insulating film TP3 are formed separately from each other, the position of the grain boundary GB1 of the insulating film TP1 and the position of the grain boundary GB2 of the insulating film TP3 can be offset, and the grain boundary GB1 of the insulating film TP1 and the grain boundary GB2 of the insulating film TP3 can be separated.

[0147] In addition, in the first embodiment, by Figure 10 In the step of performing heat treatment, the insulating film TP is crystallized. By performing the same heat treatment as in the third embodiment, the insulating film TP1 and the insulating film TP3, which are amorphous films, are crystallized to form polycrystalline films, but the insulating film TP2 can be left as an amorphous film. Since the insulating film TP2 is an amorphous film, the possibility of the grain boundary GB1 of the insulating film TP1 and the grain boundary GB2 of the insulating film TP3 being connected to each other via the insulating film TP2 can be more reliably prevented.

[0148] As described above, in the third embodiment, leakage caused by the grain boundary of the insulating film TP between the charge storage layer CSL and the memory gate electrode MG can be suppressed. Therefore, the retention characteristics of the memory cell MC1 can be further improved, and the reliability of the semiconductor device can be further improved.

[0149] The technology disclosed in the third embodiment can also be used in combination with the second embodiment.

[0150] Fourth embodiment

[0151] See below Figures 17 to 19 The semiconductor device of the fourth embodiment will be described with reference to Figures 20 to 25 , a method for manufacturing a semiconductor device according to a fourth embodiment will be described. In the following description, differences from the first embodiment will be mainly explained.

[0152] The memory cell MC1 of the first embodiment is a single-gate memory cell having a memory gate electrode MG.

[0153] The memory cell MC2 of the fourth embodiment is a split-gate memory cell having not only a memory gate electrode MG but also a control gate electrode CG at a position adjacent to the memory gate electrode MG. Hereinafter, a semiconductor device including the memory cell MC2 as the nonvolatile memory cell of the fourth embodiment will be described. Figure 17 A cross-sectional view of the memory cell MC2 is shown.

[0154] Structure of memory cell MC2

[0155] like Figure 17 As shown, a p-type well region PW is formed in the semiconductor substrate SB. A gate dielectric film GF is formed on the well region PW, and a control gate electrode CG is formed on the gate dielectric film GF. The gate dielectric film GF is, for example, a silicon oxide film and has a thickness of, for example, 2 nm to 5 nm. Instead of the silicon oxide film, the gate dielectric film GF may be a high dielectric constant film such as a metal oxide film (such as a hafnium oxide film). The control gate electrode CG is, for example, a polycrystalline silicon film into which n-type impurities have been introduced. The control gate electrode CG may be, for example, a metal film such as a titanium nitride film, an aluminum film, or a tungsten film, or a stacked film of these metal films.

[0156] The gate dielectric film MZ is formed on the well region PW and one side surface of the control gate electrode CG. The gate dielectric film MZ of the fourth embodiment is the same as the gate dielectric film MZ of the first embodiment. Figure 17 In the figure, for the sake of clarity, the gate dielectric film MZ is shown as a single-layer film, but in fact, the gate dielectric film MZ is a multi-layer film having an insulating film BT, an insulating film HSO1, an insertion layer AL1, an insulating film HSO2 and an insulating film TP, which is shown in the enlarged view. Figure 17 The area enclosed by the dotted line.

[0157] The memory gate electrode MG is formed on one side surface of the control gate electrode CG via a gate dielectric film MZ. That is, an insulating film such as the gate dielectric film MZ is formed between the control gate electrode CG and the memory gate electrode MG, and the control gate electrode CG is insulated and separated from the memory gate electrode MG. Sidewall spacers SW are formed on the side surface of the memory gate electrode MG opposite the control gate electrode CG, on both side surfaces of the memory gate electrode MG, and on the other side surface of the control gate electrode CG.

[0158] An extension region EXS, which is an n-type impurity region, is formed in the well region PW below the sidewall spacer SW on the memory gate electrode MG side, and an extension region EXD, which is an n-type impurity region, is formed in the well region PW below the sidewall spacer SW on the control gate electrode CG side. The extension region EXS constitutes a portion of the source region of the memory cell MC2, and the extension region EXD constitutes a portion of the drain region of the memory cell MC2.

[0159] A diffusion region MS, serving as an n-type impurity region, is formed in the well region PW at a position aligned with the sidewall spacers SW on the memory gate electrode MG side, and a diffusion region MD, serving as an n-type impurity region, is formed in the well region PW at a position aligned with the sidewall spacers SW on the control gate electrode CG side. Diffusion region MS has an impurity concentration higher than that of the extension region EXS, is connected to the extension region EXS, and forms a portion of the source region of memory cell MC2. Diffusion region MD has an impurity concentration higher than that of the extension region EXD, is connected to the extension region EXD, and forms a portion of the drain region of memory cell MC2.

[0160] The silicide layer SI is formed on the memory gate electrode MG, the control gate electrode CG, the diffusion region MS, and the diffusion region MD.

[0161] Although the interlayer insulating film IL1 , the plug PG, the interlayer insulating film IL2 , the wiring M1 , and the like are formed over the memory cell MC2 as in the first embodiment, description thereof is omitted herein.

[0162] Operation of memory cell MC2

[0163] Next, refer to Figure 18 and Figure 19 An exemplary operation of the memory cell MC2 as a nonvolatile memory cell is described.The memory cell MC2 described herein is a selected memory cell among a plurality of memory cells MC2 present in the semiconductor device.

[0164] Figure 18 1 is an equivalent circuit diagram of the memory cell MC2 of the nonvolatile memory. Figure 19 : is a table showing an example of conditions for applying voltages to respective parts of the memory cell MC2 at the time of "writing", "erasing" and "reading". Figure 19 In the table, the voltage Vd applied to the diffusion region MD serving as the drain region, the voltage Vcg applied to the control gate electrode CG, the voltage Vmg applied to the memory gate electrode MG, the voltage Vs applied to the diffusion region MS serving as the source region, and the voltage Vb applied to the well region PW are described in the "write", "erase" and "read" states respectively.

[0165] Notice, Figure 19 The contents shown in the table are suitable examples of voltage application conditions and are not limited thereto and can be variously changed as needed. In the fourth embodiment, injecting electrons into the charge storage layer CSL is defined as "writing", and injecting holes into the charge storage layer CSL is defined as "erasing".

[0166] The writing operation is performed by a writing method using hot electron injection of source side injection, which is called source side injection (SSI) method. Figure 19 The voltage shown in the “Write” column of is applied to the respective portions of the memory cell MC2 to be written, and electrons are injected into the charge storage layer CSL to perform writing.

[0167] At this time, hot electrons are generated in the channel region covered by the memory gate electrode MG and the control gate electrode CG and injected into the charge storage layer CSL below the memory gate electrode MG. The injected hot electrons are trapped in the trap level in the charge storage layer CSL, resulting in a rise in the threshold voltage of the memory transistor having the memory gate electrode MG. In other words, the memory transistor is in a write state.

[0168] The erase operation is performed by an erase method using hot hole injection of BTBT, which is called a band-to-band tunneling (BTBT) method. That is, holes generated by BTBT are injected into the charge storage layer CSL to perform erase. For example, Figure 19 The voltage shown in the "Erase" column in the figure is applied to the various parts of the memory cell MC2 to be erased. Holes are generated by the BTBT phenomenon and injected into the charge storage layer CSL through electric field acceleration. As a result, the threshold voltage of the memory transistor is lowered. In other words, the memory transistor is in the erased state.

[0169] In a read operation, for example, Figure 19 The voltages shown in the "Read" column in FIG. 3 are applied to the respective portions of the memory cell MC2 to be read. By setting the voltage Vmg applied to the memory gate electrode MG during reading to a value between the threshold voltage of the memory transistor in the write state and the threshold voltage of the memory transistor in the erase state, the write state and the erase state can be distinguished.

[0170] In the memory cell MC2 of the fourth embodiment, the gate dielectric film MZ is a multilayer film including the insulating film BT, the insulating film HSO1, the insertion layer AL1, the insulating film HSO2, and the insulating film TP, as in the memory cell MC1 of the first embodiment. Therefore, even in the fourth embodiment, the retention characteristics of the memory cell MC2 can be improved, and the reliability of the semiconductor device can be improved.

[0171] Method for manufacturing memory cell MC2

[0172] See below Figures 20 to 25 A method for manufacturing the semiconductor device of the fourth embodiment will be described.

[0173] First, if Figure 20 As shown, a p-type well region PW is formed in the semiconductor substrate SB by photolithography and ion implantation.

[0174] Next, a gate dielectric film GF made of, for example, silicon oxide is formed by, for example, a thermal oxidation method or an ISSG oxidation method. Thereafter, an NO treatment or a plasma nitridation treatment may be performed to nitride the silicon oxide film to form a silicon nitride oxide film. As the gate dielectric film GF, a metal oxide film such as a hafnium oxide film may be formed by, for example, an ALD method.

[0175] Next, a conductive film made of, for example, a polysilicon film is deposited on the gate dielectric film GF by, for example, CVD. The conductive film can be, for example, a metal film such as a titanium nitride film, an aluminum film, or a tungsten film, or a stacked film of these metal films. Next, the conductive film is patterned by photolithography and dry etching. Thus, the conductive film is processed to form the control gate electrode CG. Next, the gate dielectric film GF exposed from the control gate electrode CG is removed, leaving the gate dielectric film GF below the control gate electrode CG.

[0176] Figure 21 The step of forming the gate dielectric film MZ is shown.

[0177] The insulating film MZ is formed on the well region PW and the upper and side surfaces of the control gate electrode CG. As described above, the gate dielectric film MZ includes the insulating film BT, the insulating film HSO1, the insertion layer AL1, the insulating film HSO2, and the insulating film TP, and the method of forming these insulating films is the same as that of the first embodiment.

[0178] Figure 22 There is shown a step of forming the memory gate electrode MG.

[0179] First, a conductive film made of, for example, polysilicon is deposited on the gate dielectric film MZ by, for example, a CVD method. The conductive film can be, for example, a metal film such as a titanium nitride film, an aluminum film, or a tungsten film, or a stacked film of these metal films. Next, anisotropic etching is performed to process the conductive film into a sidewall shape, thereby forming a memory gate electrode MG formed of the conductive film on both side surfaces of the control gate electrode CG through the gate dielectric film MZ.

[0180] Figure 23 There is shown a step of removing a portion of the gate dielectric film MZ and a portion of the memory gate electrode MG.

[0181] First, a resist pattern is formed to cover the memory gate electrode MG formed on one side surface of the control gate electrode CG. Next, dry etching and wet etching are performed using the resist pattern as a mask to remove the gate dielectric film MZ and the memory gate electrode MG that are not covered by the resist pattern. As a result, the gate dielectric film MZ and the memory gate electrode MG on the drain region side of the memory cell MC2 are removed, leaving the gate dielectric film MZ and the memory gate electrode MG on the source region side of the memory cell MC2.

[0182] Figure 24 The steps of forming the extension region EXD and the extension region EXS are shown.

[0183] The n-type extension region EXD and the n-type extension region EXS are formed by introducing arsenic (As) or phosphorus (P) into the well region PW by, for example, photolithography and ion implantation. The extension region EXD and the extension region EXS are formed to be self-aligned with the control gate electrode CG and the memory gate electrode MG.

[0184] Figure 25 The steps of forming the sidewall spacers SW, the diffusion regions MD, and the diffusion regions MS are shown.

[0185] First, an insulating film made of, for example, silicon nitride is formed by, for example, CVD to cover the memory cell MC. Next, anisotropic dry etching is performed on the insulating film to form sidewall spacers SW on the side surfaces of the control gate electrode CG and the memory gate electrode MG. Note that the insulating film forming the sidewall spacers SW may be a silicon oxide film and a silicon nitride film formed on the silicon oxide film.

[0186] Next, by photolithography and ion implantation, using the sidewall spacers SW as masks, arsenic (As) or phosphorus (P), for example, is introduced into the well region PW, thereby forming the n-type diffusion region MD and the n-type diffusion region MS.

[0187] Thereafter, a silicide layer SI is formed on the memory gate electrode MG, the control gate electrode CG, the diffusion region MS, and the diffusion region MD in the same manner as in the first embodiment, thereby manufacturing Figure 17 The memory cell MC2 is shown.

[0188] Furthermore, the techniques of the second embodiment and the third embodiment may be combined and applied to the memory cell MC2 of the fourth embodiment.

[0189] Revise

[0190] See below Figure 26 A semiconductor device of a modification of the fourth embodiment is described. In the following description, differences from the fourth embodiment are mainly explained.

[0191] Similar to the memory cell MC2 of the fourth embodiment, the memory cell MC3 of this modification is a split-gate memory cell having a memory gate electrode MG and a control gate electrode CG. In the fourth embodiment, the gate dielectric film GF and the control gate electrode CG are formed first, and then the gate dielectric film MZ and the memory gate electrode MG are formed. However, in this modification, the order of forming them is reversed.

[0192] In the following, reference Figure 26 A semiconductor device including the memory cell MC3 of the present modification is described.

[0193] like Figure 26 As shown, a gate dielectric film MZ is formed on the well region PW (semiconductor substrate SB), and a memory gate electrode MG is formed on the gate dielectric film MZ. An insulating film IF1 is formed on one side surface of the memory gate electrode MG. The insulating film IF1 is made of, for example, silicon nitride or silicon oxide. The gate dielectric film GF is formed on the well region PW and the insulating film IF1. When the insulating film IF1 is a silicon nitride film, as shown in FIG. Figure 26 As shown in FIG. 1 , the gate dielectric film GF formed by the ISSG oxide method is also formed on the insulating film IF1. When the insulating film IF1 is a silicon oxide film, the gate dielectric film GF is not formed on the insulating film IF1. When the gate dielectric film GF is formed of a high dielectric constant film such as a hafnium oxide film by the ALD method, the gate dielectric film GF is formed on the insulating film IF1 regardless of whether the insulating film IF1 is a silicon nitride film or a silicon oxide film.

[0194] On one side surface of the memory gate electrode MG, the control gate electrode CG is formed via the insulating film IF1 and the gate dielectric film GF. That is, an insulating film such as at least the insulating film IF1 is formed between the control gate electrode CG and the memory gate electrode MG, and the control gate electrode CG is insulated and separated from the memory gate electrode MG.

[0195] The structure and manufacturing method of each of the gate dielectric film MZ, the memory gate electrode MG, the gate dielectric film GF, and the control gate electrode CG are the same as those of the fourth embodiment.

[0196] Sidewall spacers SW are formed on the side surface of the control gate electrode CG opposite the memory gate electrode MG, on both side surfaces of the control gate electrode CG, and on the other side surface of the memory gate electrode MG. An extension region EXS, which is an n-type impurity region, is formed in the well region PW below the sidewall spacers SW on the memory gate electrode MG side, and an extension region EXD, which is an n-type impurity region, is formed in the well region PW below the sidewall spacers SW on the control gate electrode CG side. Diffusion region MS, which is an n-type impurity region, is formed in the well region PW at a position aligned with the sidewall spacers SW on the memory gate electrode MG side; and diffusion region MD, which is an n-type impurity region, is formed in the well region PW at a position aligned with the sidewall spacers SW on the control gate electrode CG side. A silicide layer SI is formed over the memory gate electrode MG, the control gate electrode CG, the diffusion region MS, and the diffusion region MD.

[0197] Incidentally, the equivalent circuit diagram of the memory cell MC3 and the operating voltages for "write", "erase" and "read" are shown in FIG. Figure 18 and Figure 19 The same ones in .

[0198] Furthermore, in the present modification, the retention characteristic of the memory cell MC3 can be improved, and the reliability of the semiconductor device can be improved.

[0199] Although the invention made by the present inventors has been specifically described based on the embodiments, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the gist thereof.

[0200] For example, in the above-described embodiment, the memory cells MC1 to MC3 are formed on a flat semiconductor substrate SB, but the memory cells MC1 to MC3 may be fin structures. That is, the memory cells MC1 to MC3 may be provided by processing a portion of the semiconductor substrate SB into a convex shape to form a protruding portion, and forming a gate dielectric film MZ to cover the upper surface and side surfaces of the protruding portion.

Claims

1. A semiconductor device, comprising: A non-volatile memory cell having: a first gate dielectric film formed on a semiconductor substrate and including a charge storage layer capable of retaining charges; as well as a first gate electrode formed on the first gate dielectric film, The charge storage layer comprises: a first insulating film formed on the semiconductor substrate, wherein the first insulating film is made of a hafnium silicate film; a first insertion layer formed on the first insulating film, the first insertion layer being made of an aluminum nitride film, an aluminum carbide film, an aluminum silicate film or an aluminum oxide film, and a second insulating film formed on the first insertion layer, the second insulating film being made of the hafnium silicate film, The thickness of the first insertion layer is smaller than the thickness of the first insulating film and the thickness of the second insulating film.

2. The semiconductor device according to claim 1, wherein the first gate dielectric film includes a third insulating film between the semiconductor substrate and the charge storage layer, wherein the third insulating film is made of a silicon oxide film or a silicon nitride oxide film, and The first insulating film is in direct contact with the third insulating film.

3. The semiconductor device according to claim 2, wherein the charge storage layer comprises: a second insertion layer formed on the second insulating film, the second insertion layer being made of the same film as that of the first insertion layer, and A fourth insulating film is formed on the second insertion layer, and the fourth insulating film is made of the hafnium silicate film.

4. The semiconductor device according to claim 1, wherein the first gate dielectric film includes a third insulating film between the charge storage layer and the first gate electrode, wherein the third insulating film is made of the aluminum oxide film, the aluminum oxynitride film or the aluminum silicate film, and, wherein the third insulating film has a thickness greater than a thickness of the first insertion layer.

5. The semiconductor device according to claim 1, wherein the first gate dielectric film includes a third insulating film between the charge storage layer and the first gate electrode, The third insulating film comprises: a fourth insulating film formed on the charge storage layer, the fourth insulating film being made of the aluminum oxide film, the aluminum oxynitride film or the aluminum silicate film, a fifth insulating film formed on the fourth insulating film, the fifth insulating film being made of a silicon oxide film, a silicon nitride oxide film, or a silicon nitride film, and a sixth insulating film formed on the fifth insulating film, the sixth insulating film being made of the aluminum oxide film, the aluminum oxynitride film or the aluminum silicate film, and The thickness of the first insertion layer is smaller than the thickness of the fourth insulating film and the thickness of the sixth insulating film. 6 . The semiconductor device according to claim 5 , wherein the fourth insulating film including a plurality of first crystal grains and the sixth insulating film including a plurality of second crystal grains are separated by the fifth insulating film.

7. The semiconductor device according to claim 6, wherein the fourth insulating film is a first polycrystalline film, wherein the sixth insulating film is a second polycrystalline film, and The fifth insulating film is an amorphous film.

8. The semiconductor device according to claim 1, The non-volatile memory unit comprises: A second gate dielectric film, formed on the semiconductor substrate, and A second gate electrode, formed on the second gate dielectric film, and Wherein the first gate electrode is insulated from and separated from the second gate electrode.

9. The semiconductor device according to claim 1, wherein the first insertion layer is the alumina layer.

10. A method of manufacturing a semiconductor device, the method comprising: (a) Forming a first gate dielectric film on a semiconductor substrate, the first gate dielectric film including a charge storage layer capable of holding charges; And (b) Forming a first gate electrode on the first gate dielectric film; Wherein, in (a), forming the charge storage layer includes: (a1) Forming a first insulating film on the semiconductor substrate, the first insulating film being a hafnium silicate film; (a2) Forming a first insertion layer on the first insulating film, the first insertion layer being an aluminum nitride film, an aluminum carbide film, an aluminum silicate film or an alumina film; and (a3) Forming a second insulating film on the first insertion layer, the second insulating film being the hafnium silicate film, Wherein the thickness of the first insertion layer is less than the thickness of the first insulating film and the thickness of the second insulating film.

11. The method according to claim 10, including performing a heat treatment after (a3), wherein the hafnium silicate film of each of the first insulating film and the second insulating film is an Hf x Si 1-x O2 (0 < x < 1) film Wherein when 0.9 ≤ x < 1, the temperature of the heat treatment is set to 800 °C or higher and lower than 975 °C, and Wherein when 0 < x < 0.9, the temperature of the heat treatment is set to 975 °C or higher and 1050 °C or lower.

12. The method according to claim 10, Wherein in (a), forming the charge storage layer includes: (a4) Forming a second insertion layer on the second insulating film, the second insertion layer having a thickness less than the thickness of the second insulating film, the second insertion layer being the aluminum nitride film, the aluminum carbide film, the aluminum silicate film or the alumina film; And (a5) Forming a third insulating film on the second insertion layer, the third insulating film having a thickness greater than the thickness of the second insertion layer, the third insulating film being the hafnium silicate film, Wherein the thickness of the first insertion layer is less than the thickness of the first insulating film, and Wherein the thickness of the second insulating film is greater than the thickness of the first insertion layer.