Semiconductor storage device
By adopting a combined layout structure of fork sheet FET and nanosheet FET in the mask ROM, the problems of reducing distance between nanosheets and reducing the area of storage devices are solved, and an efficient storage device design is achieved.
Patent Information
- Application Number
- CN202080087274.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-12-20
- Filing Date
- 2020-12-14
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2040-12-14
AI Technical Summary
In the prior art, the mask ROM layout structure using nanosheet FETs or fork FETs has not been studied, resulting in challenges in reducing the distance between nanosheets and realizing a small area storage device.
Using a nanosheet FET storage unit and a fork FET storage unit adjacent in the first direction, the nanosheet FET is arranged between the bit line and the ground power wiring, and the fork-shaped structure of the fork FET is used to reduce the nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's electrical circuit is arranged between the bit line and the ground power wiring, and the fork-shaped structure of the fork FET is used to reduce the nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's nanosheet's electrical circuit is used to reduce the nanosheet's nanosheet's nanosheet's electrical circuit's nanosheet's electrical circuit's nanosheet's electrical circuit's electrical circuit's nanosheet's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit's electrical circuit' The distance between the two realizes a small area layout structure.
The small-area layout structure of semiconductor memory devices is realized, the distance between nanosheets is reduced, the integration and efficiency of the memory devices are improved, and the working speed deviation between bit lines is suppressed.
Smart Images

Figure CN114868242B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to a semiconductor memory device including a nanosheet (nanowire) FET (Field Effect Transistor), and more particularly to a layout structure of a mask ROM (Read Only Memory) using the nanosheet FET. Background Art
[0002] The mask ROM includes memory cells arranged in an array, and the mask ROM is manufactured as follows: each memory cell is programmed to have a fixed data state. The transistor constituting the memory cell is arranged between the bit line and VSS, and the word line is connected to the gate. Depending on whether the source or drain is connected to the bit line or VSS, the bit data "1" / "0" is stored. The presence or absence of connection is achieved, for example, by the presence or absence of contacts or vias.
[0003] Transistors, the basic components of LSI, have achieved increased integration, lower operating voltage, and higher operating speed by reducing gate length (scaling). However, in recent years, the problem has arisen that excessive scaling will cause cutoff current, which in turn will cause a significant increase in power consumption. To solve this problem, people have begun to actively study three-dimensional transistors that change the transistor structure from the existing planar type to a three-dimensional type. Nanosheet (nanowire) FETs are one of them that have attracted much attention.
[0004] Non-Patent Documents 1 and 2 disclose a layout of an SRAM memory cell using a nanosheet FET having a forked gate electrode.
[0005] Non-patent literature 1: P. Weckx et al., “Stacked nanosheet fork architecture for SRAM design and device co-optimization toward 3nm”, 2017 IEEE International Electron Devices Meeting (IEDM), December 2017, IEDM 17-505~508
[0006] Non-patent literature 2: P.Weckx et al., "Novel forksheet device architecture as ultimate logic scaling device towards 2nm", 2019 IEEE International Electron Devices Meeting (IEDM), December 2019, IEDM19-871~874 Summary of the invention
[0007] -Technical problem to be solved by the invention-
[0008] In this specification, according to the description of Non-Patent Document 1, a nanosheet FET having a fork-shaped gate electrode is referred to as a fork sheet FET.
[0009] To date, no research has been conducted on the layout construction of mask ROM using nanosheet FETs or forksheet FETs.
[0010] The present disclosure aims to provide a layout configuration of a mask ROM using a nanosheet FET or a forksheet FET.
[0011] -Technical solutions to technical problems-
[0012] In a first aspect of the present disclosure, a semiconductor memory device includes a first ROM (Read Only Memory) memory cell and a second ROM memory cell adjacent to each other in a first direction, characterized in that the device includes: a word line extending in the first direction; a first bit line and a second bit line extending in a second direction perpendicular to the first direction; and a first ground power wiring and a second ground power wiring extending in the second direction, wherein the first ROM memory cell includes a first nanosheet FET (Field Effect Transistor). Transistor: field effect transistor), the first nanosheet FET is arranged between the first bit line and the first ground power wiring, and has a first nanosheet as a channel region, the second ROM storage unit includes a second nanosheet FET, the second nanosheet FET is arranged between the second bit line and the second ground power wiring, and has a second nanosheet as a channel region, the word line includes a first gate wiring and a second gate wiring, the first gate wiring surrounds the periphery of the first nanosheet in the first direction and the third direction, the third direction is perpendicular to the first direction and the second direction, the second gate wiring surrounds the periphery of the second nanosheet in the first direction and the third direction, the first ROM storage unit stores first data according to the connection relationship between the node of the first nanosheet FET and the first bit line and the first ground power wiring, the second ROM storage unit stores second data according to the connection relationship between the node of the second nanosheet FET and the second bit line and the second ground power wiring, the first nanosheet and the second nanosheet face each other in the first direction, and the surface of the first nanosheet close to the second nanosheet in the first direction is exposed from the first gate wiring, and the surface of the second nanosheet close to the first nanosheet in the first direction is exposed from the second gate wiring.
[0013] According to this aspect, the semiconductor memory device includes a first ROM memory cell and a second ROM memory cell adjacent to each other in a first direction. The first ROM memory cell includes a first nanosheet FET disposed between a first bit line and a first ground power wiring and having a first nanosheet as a channel region. The second ROM memory cell includes a second nanosheet FET disposed between a second bit line and a second ground power wiring and having a second nanosheet as a channel region. The first nanosheet and the second nanosheet face each other in the first direction, and a surface of the first nanosheet close to the second nanosheet in the first direction is exposed from the first gate wiring, and a surface of the second nanosheet close to the first nanosheet in the first direction is exposed from the second gate wiring. Thus, the distance between the first nanosheet and the second nanosheet can be reduced, and thus the area of the semiconductor memory device can be reduced.
[0014] In a second aspect of the present disclosure, a semiconductor memory device includes a ROM (Read 0nly Memory) memory cell, characterized in that it includes: a word line extending in a first direction; a bit line extending in a second direction perpendicular to the first direction; and a ground power wiring extending in the second direction, the ROM memory cell includes: a first nanosheet FET arranged between the bit line and the ground power wiring and having a first nanosheet as a channel region; and a second nanosheet FET arranged between the bit line and the ground power wiring and having a second nanosheet as a channel region, the first nanosheet FET and the second nanosheet FET are adjacent to each other in the first direction, the drains of the first nanosheet FET and the second nanosheet FET are connected to each other, the sources of the first nanosheet FET and the second nanosheet FET are connected to each other, the word line includes a first A gate wiring and a second gate wiring, wherein the first gate wiring surrounds the outer periphery of the first nanosheet in the first direction and the third direction, the third direction being perpendicular to the first direction and the second direction, the second gate wiring surrounds the outer periphery of the second nanosheet in the first direction and the third direction, the ROM storage unit stores data according to the connection relationship between the nodes of the first nanosheet FET and the second nanosheet FET and the bit line and the ground power wiring, the first nanosheet and the second nanosheet face each other in the first direction, and the surface of the first nanosheet close to the second nanosheet in the first direction is exposed from the first gate wiring, and the surface of the second nanosheet close to the first nanosheet in the first direction is exposed from the second gate wiring.
[0015] According to this aspect, the semiconductor memory device includes a ROM memory cell. The ROM memory cell includes: a first nanosheet FET disposed between a bit line and a ground power wiring and having a first nanosheet as a channel region; and a second nanosheet FET disposed between the bit line and the ground power wiring and having a second nanosheet as a channel region. The first nanosheet FET and the second nanosheet FET are adjacent in a first direction. The first nanosheet and the second nanosheet face each other in the first direction, and the surface of the first nanosheet close to the second nanosheet in the first direction is exposed from the first gate wiring, and the surface of the second nanosheet close to the first nanosheet in the first direction is exposed from the second gate wiring. As a result, the spacing between the first nanosheet and the second nanosheet can be reduced, so that the size of the ROM memory cell in the first direction can be reduced, and the semiconductor memory device can be reduced in area.
[0016] In a third aspect of the present disclosure, a semiconductor memory device includes a first ROM (Read Only Memory) memory cell and a second ROM memory cell adjacent to each other in a first direction, characterized in that: it includes: a word line extending in the first direction; a first bit line and a second bit line extending in a second direction perpendicular to the first direction; and a plurality of ground power wirings including a first ground power wiring and a second ground power wiring extending in the second direction, the first ROM memory cell includes a first nanosheet FET, the first nanosheet FET is arranged between the first bit line and the first ground power wiring, and has a first nanosheet as a channel region, the second ROM memory cell includes a second nanosheet FET, the second nanosheet FET is arranged between the second bit line and the second ground power wiring, and has a second nanosheet as a channel region, the word line includes a first gate wiring and a second gate wiring, the first gate wiring surrounds the outer periphery of the first nanosheet in the first direction and a third direction, the third direction being perpendicular to the first direction and the second direction, the second gate wiring surrounds the periphery of the second nanosheet in the first direction and the third direction, the first ROM storage unit stores first data according to the connection relationship between the node of the first nanosheet FET and the first bit line and the first ground power wiring, the second ROM storage unit stores second data according to the connection relationship between the node of the second nanosheet FET and the second bit line and the second ground power wiring, the distance between the first bit line and the ground power wiring adjacent to the first side of the first bit line in the first direction is equal to the distance between the second bit line and the ground power wiring adjacent to the first side of the second bit line in the first direction, the distance between the first bit line and the ground power wiring adjacent to the second side of the first bit line in the first direction is equal to the distance between the second bit line and the ground power wiring adjacent to the second side of the second bit line in the first direction, and the second side is the opposite side of the first side.
[0017] According to this aspect, a semiconductor memory device includes a first ROM memory cell and a second ROM memory cell adjacent to each other in a first direction. The first ROM memory cell includes a first nanosheet FET disposed between a first bit line and a first ground power wiring and having a first nanosheet as a channel region. The second ROM memory cell includes a second nanosheet FET disposed between a second bit line and a second ground power wiring and having a second nanosheet as a channel region. The distance between the first bit line and the ground power wiring on both sides thereof is respectively equal to the distance between the second bit line and the ground power wiring on both sides thereof. Thus, since the load capacitance generated by the wiring inter-capacitance of the first bit line and the second bit line is consistent, it is possible to suppress performance deviations such as operating speed between the first bit line and the second bit line.
[0018] -Effects of the Invention-
[0019] According to the present disclosure, a semiconductor memory device using a nanosheet FET or a forksheet FET can achieve a small-area layout structure. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 is a circuit diagram showing the structure of a contact-type mask ROM as an example of a semiconductor memory device;
[0021] Figure 2 is a plan view showing an example of a layout structure of a semiconductor memory device according to the first embodiment;
[0022] Figure 3 (a) to (c) are Figure 2 A cross-sectional view of the layout structure;
[0023] Figure 4 is a plan view showing an example of a layout structure of a semiconductor memory device according to a first modification of the first embodiment;
[0024] Figure 5 is a plan view showing an example of a layout structure of a semiconductor memory device according to a first modification of the first embodiment;
[0025] Figure 6 is a plan view showing an example of a layout structure of a semiconductor memory device according to a second modification of the first embodiment;
[0026] Figure 7 is a plan view showing an example of a layout structure of a semiconductor storage device according to a third modification of the first embodiment;
[0027] Figure 8 (a) to (c) are Figure 7 A cross-sectional view of the layout structure;
[0028] Fig. 9 is a plan view showing an example of a layout structure of a semiconductor memory device according to a fourth modification of the first embodiment;
[0029] Fig.10 is a plan view showing an example of a layout structure of a semiconductor memory device according to a second embodiment;
[0030] Fig.11 (a) to (c) are Fig.10 A cross-sectional view of the layout structure;
[0031] Fig.12is a plan view showing an example of a layout structure of a semiconductor memory device according to a first modification of the second embodiment;
[0032] Fig.13 is a circuit diagram showing the structure of a contact-type mask ROM as an example of a semiconductor memory device;
[0033] Fig.14 is a plan view showing an example of a layout structure of a semiconductor memory device according to a third embodiment;
[0034] Fig.15 is a plan view showing an example of a layout structure of a semiconductor memory device according to a first modification of the third embodiment;
[0035] Fig.16 is a plan view showing an example of a layout structure of a semiconductor memory device according to a second variation of the third embodiment;
[0036] Fig.17 is a plan view showing an example of a layout structure of a semiconductor memory device according to a third modification of the third embodiment;
[0037] Fig.18 is a plan view showing an example of a layout structure of a semiconductor memory device according to a fourth embodiment;
[0038] Fig.19 2 are diagrams showing a basic structure of a chip FET, wherein (a) is a top view and (b) is a cross-sectional view. DETAILED DESCRIPTION
[0039] Hereinafter, the embodiments will be described with reference to the accompanying drawings. In the following embodiments, it is assumed that the semiconductor memory device includes a nanosheet FET (Field Effect Transistor). The nanosheet FET is a FET that uses a thin sheet (nanosheet) for current to flow. The nanosheet is formed of silicon, for example. In the semiconductor memory device, a portion of the nanosheet FET is a fork-shaped gate electrode. In the present disclosure, the semiconductor layer portion formed at both ends of the nanosheet and constituting a terminal that becomes a source or drain of the nanosheet FET is referred to as a "pad".
[0040] First, the basic structure of the fork FET is described.
[0041] Fig.19 1 is a diagram showing the basic structure of a fork-chip FET, (a) is a top view, and (b) is a cross-sectional view taken along line Y-Y' in (a). Fig.19In the basic structure, two transistors TR1 and TR2 are arranged in the Y direction with a spacing S therebetween. A gate wiring 531 serving as the gate of the transistor TR1 and a gate wiring 532 serving as the gate of the transistor TR2 both extend in the Y direction and are arranged at the same position in the X direction.
[0042] The channel portion 521 that becomes the channel region of the transistor TR1 and the channel portion 526 that becomes the channel region of the transistor TR2 are formed of nanosheets. Fig.19 In the embodiment, the channel portions 521 and 526 are respectively composed of nanosheets, and the nanosheets are formed of three thin sheet structures that overlap when viewed from above. On both sides of the channel portion 521 in the X direction, pads 522a and 522b that become the source region or drain region of the transistor TR1 are formed. On both sides of the channel portion 526 in the X direction, pads 527a and 527b that become the source region or drain region of the transistor TR2 are formed. The pads 522a and 522b are formed by epitaxial growth of the nanosheets that constitute the channel portion 521. The pads 527a and 527b are formed by epitaxial growth of the nanosheets that constitute the channel portion 526.
[0043] The gate wiring 531 surrounds the outer periphery of the channel portion 521 formed by the nanosheet in the Y direction and the Z direction via a gate insulating film (not shown). However, the surface of the nanosheet forming the channel portion 521 close to the transistor TR2 in the Y direction is not covered by the gate wiring 531 but is exposed from the gate wiring 531. Fig.19 In the cross-sectional view of (b), the gate wiring 531 does not cover the right side of the nanosheet constituting the channel portion 521 in the figure, but covers the upper side, the left side, and the lower side in the figure. The gate wiring 531 extends to the side opposite to the transistor TR2 in the Y direction relative to the nanosheet constituting the channel portion 521, and the extension amount is the length OL.
[0044] The gate wiring 532 surrounds the outer periphery of the channel portion 526 formed by the nanosheet in the Y direction and the Z direction via a gate insulating film (not shown). However, the surface of the nanosheet forming the channel portion 526 close to the transistor TR1 in the Y direction is not covered by the gate wiring 532 but is exposed from the gate wiring 532. Fig.19 In the cross-sectional view of (b), the gate wiring 532 does not cover the left side of the nanosheet constituting the channel portion 526, but covers the upper side, the right side, and the lower side of the nanosheet constituting the channel portion 526. The gate wiring 532 extends to the side opposite to the transistor TR1 in the Y direction with respect to the nanosheet constituting the channel portion 526, and the extension amount is a length 0L.
[0045] If the width (dimension in the Y direction) of each nanosheet is set to W and the height (dimension in the Z direction) is set to H, the effective gate width Weff is:
[0046] Weff=2×W+H.
[0047] Since the channel portion 521 of the transistor TR1 and the channel portion 526 of the transistor TR2 are both composed of three nanosheets, the effective gate widths of the transistors TR1 and TR2 are:
[0048] 3×(2×W+H).
[0049] according to Fig.19 In the structure, the gate wiring 531 does not extend toward the side close to the transistor TR2 in the Y direction relative to the nanosheet constituting the channel portion 521. The gate wiring 532 does not extend toward the side close to the transistor TR1 in the Y direction relative to the nanosheet constituting the channel portion 526. In this way, the transistors TR1 and TR2 can be brought closer together, thereby achieving a smaller area.
[0050] It should be noted that the number of nanosheets constituting the channel portion of the transistor is not limited to three. In other words, the nanosheet may be formed of a single sheet structure or may be formed of multiple sheet structures that overlap when viewed from above. Fig.19 In (b), the cross-sectional shape of the nanosheet is illustrated as a rectangle, but the cross-sectional shape of the nanosheet is not limited thereto. For example, the cross-sectional shape of the nanosheet may be a square, a circle, an ellipse, or the like.
[0051] Forksheet FETs and nanosheet FETs can be mixed in a semiconductor memory device, with the gate wiring of the nanosheet FET surrounding the entire circumference of the nanosheet.
[0052] In this specification, "VDD" and "VSS" refer to power supply voltage or power supply itself. In this specification, expressions such as "same wiring width" mean that the width is equal and include the range of manufacturing variation.
[0053] In this specification, the source region and the drain region of a transistor are appropriately referred to as “nodes” of the transistor. That is, one node of a transistor refers to the source or drain of the transistor, and two nodes of a transistor refer to the source and drain of the transistor.
[0054] (First Embodiment)
[0055] Figure 1 This is a circuit diagram showing the structure of a contact-type mask ROM as an example of a semiconductor memory device. The contact-type mask ROM is a mask ROM in which the stored data "0" or "1" is connected to or disconnected from a bit line via a contact corresponding to the drain of a memory cell transistor.
[0056] exist Figure 1In FIG. 1 , the mask ROM includes a memory cell array 3 , a column decoder 2 , and a sense amplifier 18 .
[0057] The memory cell array 3 is composed of N-type MOS transistors, namely memory cells Mij (i=0~m, j=0~n) arranged in a matrix. The gates of the memory cells Mij are connected to the word lines WLi in the row direction, and the sources of the memory cells Mij are connected to the ground potential VSS. Here, when the storage data is set to "0", the drain of the memory cell Mij is connected to the bit line BLj, and when the storage data is set to "1", the drain of the memory cell Mij is not connected to the bit line BLj.
[0058] The column decoder 2 is composed of N-type MOS transistors Cj. The drains of all the N-type MOS transistors Cj are connected in common, the gates are connected to the column selection signal lines CLj, and the sources are connected to the bit lines BLj.
[0059] The sense amplifier 18 includes a precharge P-type MOS transistor 5, an inverter 8 for determining the output data of the memory cell Mij, and an inverter 9 for buffering the output signal of the inverter 8. The precharge signal NPR is input to the gate of the P-type MOS transistor 5, the power supply voltage VDD is supplied to the source, and the drain is connected to the common drain of the N-type MOS transistor Cj. The inverter 8 receives the signal SIN of the common drain of the N-type MOS transistor Cj and determines the output data of the memory cell Mij. The inverter 9 receives the output signal SOUT of the inverter 8 and outputs the storage data of the memory cell Mij.
[0060] about Figure 1 The operation of the mask ROM will be described by taking the case of reading the data of the memory cell M00 as an example.
[0061] First, in the column selection signal line CLj, CL0 is set to a high level, and the other CL1 to CLn are set to a low level. As a result, C0 of the transistors constituting the column decoder 2 is turned on, and the other C1 to Cn are turned off. The word line WL0 is transitioned from a non-selected state, i.e., a low level, to a selected state, i.e., a high level.
[0062] Next, the precharge signal NPR is changed from a high level to a low level, and the precharge P-type MOS transistor 5 is turned on.
[0063] Here, when the drain of the memory cell M00 is connected to the bit line BL0, the current capability of the memory cell M00 is greater than that of the precharge P-type MOS transistor 5, so the input signal SIN of the inverter 8 becomes a voltage lower than the switching level of the inverter 8. Therefore, the output signal SOUT of the inverter 8 is maintained at a high level, and the output signal OUT of the inverter 9 is maintained at a low level.
[0064] On the other hand, when the drain of the memory cell M00 is not connected to the bit line BL0, the bit line BL0 is charged by the precharge P-type MOS transistor 5, and the input signal SIN of the inverter 8 becomes a voltage higher than the switching level of the inverter 8. Therefore, the output signal SOUT of the inverter 8 becomes a low level, and the output signal OUT of the inverter 9 becomes a high level.
[0065] That is, when the drain of the memory cell is connected to the bit line, a low level is output (data "0" is stored), and when the drain of the memory cell is not connected to the bit line, a high level is output (data "1" is stored).
[0066] In the mask ROM of the present disclosure, the value of each memory cell can be stored in two ways: setting by connecting / disconnecting the memory cell and the bit line, and setting by connecting / disconnecting the memory cell and VSS.
[0067] Figure 2 and Figure 3 FIG. 1 is a diagram showing an example of a layout structure of a mask ROM according to the first embodiment. Figure 2 is a top view of the memory cell array. Figure 3 (a) to (c) are Figure 2 A cross-sectional view of a memory cell array taken along a top-down direction. Figure 3 (a) is the section taken along line X1-X1'. Figure 3 (b) is the section along the line X2-X2'. Figure 3 (c) is a cross section taken along line X3-X3'.
[0068] It should be noted that in the following description, Figure 2 In the top view, the horizontal direction in the figure is set as the X direction (equivalent to the first direction), the vertical direction in the figure is set as the Y direction (equivalent to the second direction), and the direction perpendicular to the substrate surface is set as the Z direction (equivalent to the depth direction). Among them, the X direction is the direction in which the gate wiring and the word line extend, and the Y direction is the channel direction, which is the direction in which the bit line extends.
[0069] In each of the drawings of the first and second embodiments, a contact that determines a storage value of a memory cell is denoted by a letter “D”.
[0070] Figure 2 This corresponds to a (4×4) bit layout. The dotted line represents a one-bit storage unit frame. Figure 2 , a configuration is shown in which four storage cells are arranged in the X direction and four storage cells are arranged in the Y direction. In the Y direction, the storage cells are arranged so as to be reversed in the Y direction compared to the storage cells in the previous row. For example, the two storage cells from the left of the bottommost row in the figure are respectively equivalent to Figure 1 The memory cells M00 and M01 in the circuit diagram of FIG. 1 are shown in FIG.
[0071] like Figure 2 As shown, wirings 61 to 68 extending in the Y direction are formed on the M1 wiring layer. The M1 wirings 62, 64, 66, and 68 correspond to the bit lines BL0, BL1, BL2, and BL3, respectively. The M1 wirings 61, 63, 65, and 67 are supplied with VSS.
[0072] The memory cells M00 and M01 have the nanosheets 11 and 12 composed of three thin sheets as channel portions, respectively. That is, the memory cells M00 and M01 include nanosheet FETs.
[0073] exist Figure 2 In the figure, pads 21a and 22a are formed on the lower side of the nanosheets 11 and 12, respectively, and are composed of a semiconductor layer of an integral structure connected to three thin sheets. Pads 21b and 22b are formed on the upper side of the nanosheets 11 and 12, respectively, and are composed of a semiconductor layer of an integral structure connected to three thin sheets. Pads 21a and 21b become nodes of memory cell M00. Pads 22a and 22b become nodes of memory cell M01.
[0074] Gate wirings 31 and 32 extending in the X direction are formed. The gate wiring 31 surrounds the outer periphery of the nanosheet 11 of the memory cell M00 in the X direction and the Z direction via a gate insulating film (not shown). The gate wiring 31 becomes the gate of the memory cell M00. The gate wiring 32 surrounds the outer periphery of the nanosheet 12 of the memory cell M01 in the X direction and the Z direction via a gate insulating film (not shown). The gate wiring 32 becomes the gate of the memory cell M01. It should be noted that the gate wirings 31 and 32 are connected to other gate wirings arranged in a row in the X direction to form the word line WL0. That is, the word lines WL0 to WL3 each include gate wirings arranged in a row in the X direction.
[0075] Local wirings 41, 42, 43, and 44 extending in the X direction are formed. The local wiring 41 is connected to the pad 21a, and the local wiring 42 is connected to the pad 21b. The local wiring 43 is connected to the pad 22a, and the local wiring 44 is connected to the pad 22b. The local wiring 42 is connected to the M1 wiring 62 via the contact 51. The local wiring 44 is connected to the M1 wiring 64 via the contact 52.
[0076] The storage value of the memory cell is determined by the presence or absence of contacts 71 and 72. When contact 71 is formed, contact 71 connects local wiring 41 to M1 wiring 61 supplied with VSS. When contact 72 is formed, contact 72 connects local wiring 43 to M1 wiring 63 supplied with VSS.
[0077] exist Figure 2 In the embodiment, word lines WL0 to WL3 extend in the X direction. As described above, word lines WL0 to WL3 are respectively composed of gate wirings arranged in a row in the X direction. The drains of adjacent transistors are shared between word lines WL0 and WL1. The drains of adjacent transistors are shared between word lines WL2 and WL3. Dummy gate wirings 81, 82, and 83 for supplying VSS are formed to extend in the X direction.
[0078] Here, the surface of the nanosheet 11 close to the nanosheet 12 in the X direction is not covered by the gate wiring 31, but exposed from the gate wiring 31. The surface of the nanosheet 12 close to the nanosheet 11 in the X direction is not covered by the gate wiring 32, but exposed from the gate wiring 32. That is, the memory cells M00 and M01 include fork-sheet FETs. Among them, the gate wiring 31 and the gate wiring 32 are connected by a bridge portion 33 formed between the gate wiring 31 and the gate wiring 32. The bridge portion 33 is an example of a gate connection portion.
[0079] As a result, the space required between the nanosheet 11 and the nanosheet 12 is reduced, so that the distance d1 between the nanosheet 11 and the nanosheet 12 can be reduced (d1<d2). Therefore, the area of the semiconductor memory device having the fork-sheet FET can be reduced.
[0080] As described above, according to the present embodiment, the semiconductor memory device includes ROM memory cells M00 and M01 adjacent to each other in the X direction. The ROM memory cell M00 includes a nanosheet FET, which is arranged between the bit line 62 and the ground power wiring 61 and has a nanosheet 11 as a channel region. The ROM memory cell M01 includes a nanosheet FET, which is arranged between the bit line 64 and the ground power wiring 63 and has a nanosheet 12 as a channel region. The nanosheets 11 and 12 face each other in the X direction, and the surface of the nanosheet 11 close to the nanosheet 12 in the X direction is exposed from the gate wiring 31, and the surface of the nanosheet 12 close to the nanosheet 11 in the X direction is exposed from the gate wiring 32. As a result, the distance between the nanosheets 11 and 12 can be reduced, so that the area of the semiconductor memory device can be reduced.
[0081] The M1 wirings 61 to 68 extending in the Y direction are arranged at equal intervals. Furthermore, the M1 wirings 62, 64, 66, and 68 corresponding to the bit lines BL0 to BL3 are sandwiched by the M1 wirings 61, 63, 65, and 67 supplied with VSS. By sandwiching the bit lines BL between the VSS wirings, crosstalk between the bit lines BL can be prevented, thereby suppressing malfunctions. Since the distances from the bit lines BL to the VSS wirings on both sides are equal for any bit line BL, the load capacitance caused by the wiring capacitance of the bit lines BL is consistent. Therefore, performance deviations such as operating speed between the bit lines BL are suppressed.
[0082] It should be noted that the M1 wirings 61 to 68 extending in the Y direction may not be arranged at equal intervals. Even in this case, if the distance from the bit line BL to the VSS wirings on both sides is equal for any bit line BL, the load capacitance caused by the wiring capacitance of the bit line BL is consistent, so the performance deviation such as the operating speed between the bit lines BL is suppressed. For example, even if the M1 wirings 61, 63, 65, and 67 supplying VSS are all offset by the same distance in the X direction, the load capacitance caused by the wiring capacitance of each bit line BL is consistent, so the performance deviation such as the operating speed between the bit lines BL is suppressed.
[0083] <Modification 1>
[0084] Figure 4 It is a plan view of a memory cell array according to Modification 1 of the first embodiment. Figure 4 The layout structure is basically the same as Figure 2 The layout structure is the same. However, in Figure 4In the example, the M1 wirings 62, 64, 66, and 68 corresponding to the bit lines BL0 to BL3 are located in the same position relative to the nanosheet. The M1 wirings 61, 63, 65, and 67 supplying VSS are also located in the same position relative to the nanosheet. As the layout positions of the M1 wirings 61 to 68 are changed, the positions of the contacts and the length of the local wirings are changed. Figure 2 different.
[0085] In this embodiment, a fork-shaped FET is used in the memory cell, and in the X direction, a portion where the spacing between transistors is wide and a portion where the spacing between transistors is narrow appear alternately. Therefore, if the bit lines BL are arranged at equal intervals, the positions of the bit lines BL relative to the nanosheets will be inconsistent. In contrast, in this variation, the arrangement positions of the bit lines BL relative to the nanosheets are consistent. As a result, the characteristics of each bit line BL can be made consistent.
[0086] exist Figure 4 In the space created by aligning the arrangement positions of the bit lines BL with respect to the nanosheets, M1 wirings 69a, 69b, and 69c for supplying VSS are further arranged. As a result, for any bit line BL, the distance to the VSS wirings on both sides can be made consistent, so the load capacitance for the bit line BL can be made consistent. It should be noted that M1 wirings 69a, 69b, and 69c for supplying VSS may not be additionally arranged.
[0087] like Figure 5 As shown, thick M1 wirings 61A, 65A, and 69A may be arranged instead of the additionally arranged M1 wirings 69A, 69b, and 69c for supplying VSS.
[0088] <Modification 2>
[0089] Figure 6 It is a top view of a memory cell array according to Modification 2 of the first embodiment. Figure 6 The layout structure and Figure 4 The layout structure is roughly the same. However, in Figure 6 In the embodiment, the positions of M1 wirings 61 and 62 are reversed, and the positions of M1 wirings 65 and 66 are reversed. As a result, the positions of the bit lines BL are consistent with respect to the portion of the fork exposed from the gate wiring. That is, the M1 wirings 62, 64, 66, and 68 corresponding to the bit lines BL0 to BL3 are all located on the opposite side of the fork exposed from the gate wiring. As a result, the characteristics of each bit line BL can be made more consistent.
[0090] <Variation 3>
[0091] Figure 7 is a top view of a memory cell array according to a third variation of the first embodiment. Figure 8 (a) to (c) are Figure 7A cross-sectional view of a memory cell array taken along a top-down direction. Figure 8 (a) is the section along the line X1-X1'. Figure 8 (b) is the section taken along line X2-X2'. Figure 8 (c) is a cross section taken along line X3-X3'.
[0092] In this variation, the transistor of each bit of the memory cell is composed of two fork-shaped FETs arranged in the X direction and connected in parallel. The structure of the memory cell will be described below by taking the memory cells M00 and M01 as an example.
[0093] The memory cell M00 has three nanosheets 111 and 112 as a channel portion, and the memory cell M01 has three nanosheets 113 and 114 as a channel portion.
[0094] exist Figure 7 In the figure, pads 121a and 122a are formed on the lower side of the nanosheets 111 and 112, respectively, and are composed of a semiconductor layer of an integral structure connected to three sheets. Pads 121b and 122b are formed on the upper side of the nanosheets 111 and 112, respectively, and are composed of a semiconductor layer of an integral structure connected to three sheets. Pads 121a, 121b, 122a, and 122b become nodes of the memory cell M00. Pads 123a and 124a are formed on the lower side of the nanosheets 113 and 114, respectively, and are composed of a semiconductor layer of an integral structure connected to three sheets. Pads 123b and 124b are formed on the upper side of the nanosheets 113 and 114, respectively, and are composed of a semiconductor layer of an integral structure connected to three sheets. The pads 123a, 123b, 124a, and 124b serve as nodes of the memory cell M01.
[0095] Gate wirings 131, 132, 133, and 134 extending in the X direction are formed. The gate wiring 131 surrounds the outer periphery of the nanosheet 111 of the memory cell M00 in the X direction and the Z direction via a gate insulating film (not shown). The gate wiring 132 surrounds the outer periphery of the nanosheet 112 of the memory cell M00 in the X direction and the Z direction via a gate insulating film (not shown). The gate wirings 131 and 132 serve as the gate of the memory cell M00. The gate wiring 133 surrounds the outer periphery of the nanosheet 113 of the memory cell M01 in the X direction and the Z direction via a gate insulating film (not shown). The gate wiring 134 surrounds the outer periphery of the nanosheet 114 of the memory cell M01 in the X direction and the Z direction via a gate insulating film (not shown). The gate wirings 133 and 134 serve as the gate of the memory cell M01.
[0096] Here, the surface of the nanosheet 111 in the X direction close to the nanosheet 112 is not covered by the gate wiring 131, but exposed from the gate wiring 131. The surface of the nanosheet 112 in the X direction close to the nanosheet 111 is not covered by the gate wiring 132, but exposed from the gate wiring 132. The surface of the nanosheet 113 in the X direction close to the nanosheet 114 is not covered by the gate wiring 133, but exposed from the gate wiring 133. The surface of the nanosheet 114 in the X direction close to the nanosheet 113 is not covered by the gate wiring 134, but exposed from the gate wiring 134.
[0097] Thus, the space required between nanosheets 111 and 112 is reduced, and the distance between nanosheets 111 and 112 can be reduced. Similarly, the space required between nanosheets 113 and 114 is reduced, and the distance between nanosheets 113 and 114 can be reduced.
[0098] The gate wiring 131 and the gate wiring 132 are connected by a bridge 135 formed between the gate wiring 131 and the gate wiring 132. The gate wiring 133 and the gate wiring 134 are connected by a bridge 136 formed between the gate wiring 133 and the gate wiring 134. The bridges 135 and 136 are examples of gate connecting portions.
[0099] Local wirings 141, 142, 143, and 144 extending in the X direction are formed. Local wiring 141 is connected to pads 121a and 122a, and local wiring 142 is connected to pads 121b and 122b. That is, the drains of the two nanosheet FETs of the memory cell M00 are connected to each other, and the sources are connected to each other. Local wiring 143 is connected to pads 123a and 124a, and local wiring 144 is connected to pads 123b and 124b. That is, the drains of the two nanosheet FETs of the memory cell M01 are connected to each other, and the sources are connected to each other. Local wiring 142 is connected to M1 wiring 62 via contact 151. Local wiring 144 is connected to M1 wiring 64 via contact 152.
[0100] The storage value of the memory cell is determined by the presence or absence of contacts 171 and 172. When contact 171 is formed, contact 171 connects local wiring 141 to M1 wiring 61 supplied with VSS. When contact 172 is formed, contact 172 connects local wiring 143 to M1 wiring 63 supplied with VSS.
[0101] As described above, according to this variation, the semiconductor memory device includes a ROM memory cell M00. The ROM memory cell M00 includes: a first nanosheet FET, which is arranged between the bit line 62 and the ground power wiring 61 and has a nanosheet 111 as a channel region; and a second nanosheet FET, which is arranged between the bit line 62 and the ground power wiring 61 and has a nanosheet 112 as a channel region. The nanosheets 111 and 112 face each other in the X direction, and the surface of the nanosheet 111 close to the nanosheet 112 in the X direction is exposed from the gate wiring 131, and the surface of the nanosheet 112 close to the nanosheet 111 in the X direction is exposed from the gate wiring 132. As a result, the distance between the nanosheets 111 and the nanosheets 112 can be reduced, so the size of the ROM memory cell M00 in the X direction can be reduced, and the semiconductor memory device can be reduced in area. On the other hand, in the X direction, the distance between the ROM memory cells can be widened, so it is easy to separate the local wiring, and the manufacturing ease of the semiconductor memory device is improved.
[0102] By sandwiching the bit line BL between the VSS wiring, crosstalk between the bit lines BL can be prevented, thereby suppressing malfunction. Since the distance from the bit line BL to the VSS wiring on both sides is equal for any bit line BL, the load capacitance caused by the wiring capacitance of the bit line BL is consistent. Therefore, performance deviations such as operating speed between the bit lines BL are suppressed. Furthermore, the layout position of the bit line BL relative to the nanosheet is consistent, so the characteristics of each bit line BL can be made consistent.
[0103] <Variation 4>
[0104] Fig. 9 This is a top view of a memory cell array according to variant 4 of the first embodiment. In this variant, the transistor of each memory cell is composed of two fork-piece FETs arranged adjacent to each other in the Y direction and sharing a source. The following describes the structure of the memory cell by taking memory cells M00 and M01 as examples. It should be noted that the cross-sectional structure is the same as that of the first embodiment, and the cross-sectional view is omitted here.
[0105] The memory cell M00 has three nanosheets 211 and 212 as a channel portion, and the memory cell M01 has three nanosheets 213 and 214 as a channel portion.
[0106] On the lower side of the nanosheet 211 in the figure, a pad 221a is formed of a semiconductor layer of an integral structure connected to three thin sheets. Between the nanosheets 211 and 212, a pad 221b is formed of a semiconductor layer of an integral structure connected to three thin sheets. On the upper side of the nanosheet 212 in the figure, a pad 221c is formed of a semiconductor layer of an integral structure connected to three thin sheets. The pads 221a and 221c become the drain region of the memory cell M00. The pad 221b becomes the source region of the memory cell M00.
[0107] A pad 222a consisting of a semiconductor layer connected to three sheets is formed on the lower side of the nanosheet 213 in the figure. A pad 222b consisting of a semiconductor layer connected to three sheets is formed between the nanosheets 213 and 214. A pad 222c consisting of a semiconductor layer connected to three sheets is formed on the upper side of the nanosheet 214 in the figure. The pads 222a and 222c become the drain region of the memory cell M01. The pad 222b becomes the source region of the memory cell M01.
[0108] Local wirings 241, 242, 243, 244, 245, and 246 extending in the X direction are formed. The local wiring 241 is connected to the pad 221a, the local wiring 242 is connected to the pad 221b, and the local wiring 243 is connected to the pad 221c. The local wiring 244 is connected to the pad 222a, the local wiring 245 is connected to the pad 222b, and the local wiring 246 is connected to the pad 222c. The local wiring 241 is connected to the M1 wiring 62 via the contact 251, the local wiring 243 is connected to the M1 wiring 62 via the contact 252. The local wiring 244 is connected to the M1 wiring 64 via the contact 253, and the local wiring 246 is connected to the M1 wiring 64 via the contact 254.
[0109] The storage value of the memory cell is determined by the presence or absence of the contacts 271 and 272. When the contact 271 is formed, the contact 271 connects the local wiring 242 to the M1 wiring 61 supplied with VSS. When the contact 272 is formed, the contact 272 connects the local wiring 245 to the M1 wiring 63 supplied with VSS.
[0110] The surface of the nanosheet 211 in the X direction close to the nanosheet 213 is not covered by the gate wiring, but exposed from the gate wiring. The surface of the nanosheet 213 in the X direction close to the nanosheet 211 is not covered by the gate wiring, but exposed from the gate wiring. The surface of the nanosheet 212 in the X direction close to the nanosheet 214 is not covered by the gate wiring, but exposed from the gate wiring. The surface of the nanosheet 214 in the X direction close to the nanosheet 212 is not covered by the gate wiring, but exposed from the gate wiring.
[0111] As a result, the space required between the nanosheets 211 and 212 and the nanosheets 213 and 214 is reduced, so that the distance between the nanosheets 211 and 212 and the nanosheets 213 and 214 can be reduced. Therefore, the area of the semiconductor memory device having the fork-sheet FET can be reduced.
[0112] Regarding the bit line BL, the same effects as those of the first embodiment can be obtained.
[0113] (Second Embodiment)
[0114] In the second embodiment, it is indicated Figure 1 Another example of the layout structure of the mask ROM. In this embodiment, instead of the fork sheet FET, a nanosheet FET is used.
[0115] Fig.10 and Fig.11 FIG. 1 is a diagram showing an example of a layout structure of a mask ROM according to the second embodiment. Fig.10 is a top view of the memory cell array. Fig.11 (a) to (c) are Fig.10 A cross-sectional view of a memory cell array taken along a top-down direction. Fig.11 (a) is the section taken along line X1-X1'. Fig.11 (b) is the section taken along line X2-X2'. Fig.11 (c) is a cross section taken along line X3-X3'.
[0116] The same as shown in the first embodiment Figure 2 same, Fig.10 This corresponds to a (4×4)-bit layout. The dotted line represents a one-bit storage unit frame. Fig.10 , a configuration is shown in which four storage cells are arranged in the X direction and four storage cells are arranged in the Y direction. In the Y direction, the storage cells are arranged so as to be reversed in the Y direction compared to the storage cells in the previous row. For example, the two storage cells from the left of the bottommost row in the figure are respectively equivalent to Figure 1 Storage units M00 and M01 in the circuit diagram.
[0117] exist Fig.10 and Fig.11 In, with Figure 2 and Figure 3 Common component annotation and Figure 2 and Figure 3 The same symbols are sometimes omitted in detailed description.
[0118] The memory cells M00 and M01 each have a nanosheet 11A and 12A composed of three thin sheets as a channel portion. That is, the memory cells M00 and M01 include a nanosheet FET. Both sides of the surface of the nanosheet 11A in the X direction are covered by the gate wiring 31A. Both sides of the surface of the nanosheet 12A in the X direction are covered by the gate wiring 32A. Therefore, the spacing between the nanosheets 11A and 12A is equal to the spacing between other nanosheets (d3).
[0119] The M1 wirings 61 to 68 extending in the Y direction are arranged at equal intervals. Furthermore, the M1 wirings 62, 64, 66, and 68 corresponding to the bit lines BL0 to BL3 are sandwiched by the M1 wirings 61, 63, 65, and 67 supplied with VSS. By sandwiching the bit lines BL between the VSS wirings, crosstalk between the bit lines BL can be prevented, thereby suppressing malfunctions. Since the distances from the bit lines BL to the VSS wirings on both sides are equal for any bit line BL, the load capacitance caused by the wiring capacitance of the bit lines BL is consistent. Therefore, performance deviations such as operating speed between the bit lines BL are suppressed.
[0120] Furthermore, the layout positions of the bit lines BL with respect to the nanosheets are aligned, thereby making it possible to make the characteristics of the bit lines BL uniform.
[0121] <Modification 1>
[0122] Fig.12 1 is a top view of a memory cell array according to Modification 1 of Embodiment 2. In this modification, similar to Modification 4 of Embodiment 1, the transistor of each bit of the memory cell is composed of two nanosheet FETs arranged in the Y direction and connected in parallel.
[0123] Here, detailed description of the same configuration as that of Modification 4 of the first embodiment may be omitted.
[0124] The memory cell M00 has nanosheets 211A and 212A composed of three thin sheets as the channel portion. The memory cell M01 has nanosheets 213A and 214A composed of three thin sheets as the channel portion. The surfaces on both sides of the nanosheets 211A, 212A, 213A, and 214A in the X direction are covered by gate wiring. Therefore, the spacing between the nanosheets 211A and 213A and the spacing between the nanosheets 212A and 214A are the same as the spacing between other nanosheets.
[0125] Regarding the bit line BL, the same effects as those of the second embodiment can be obtained.
[0126] (Third Embodiment)
[0127] Fig.13This is a circuit diagram showing the structure of a mask ROM as an example of a semiconductor memory device. Fig.13 The mask ROM makes the stored data "1" or "0" correspond to whether the source and drain of the memory cell transistor are connected to the same line of the bit line and the ground power wiring or to different lines.
[0128] exist Fig.13 In FIG. 1 , the mask ROM includes a memory cell array 3A, a column decoder 2 , and a sense amplifier 18 .
[0129] The memory cell array 3A is composed of N-type MOS transistors, namely memory cells Mij (i=0~m, j=0~n), arranged in a matrix. The gates of the memory cells Mij are connected to the word lines WLi in the row direction. The source and drain of the memory cell Mij are connected to the bit line BLj or to the ground power wiring VSS. Here, when the storage data of the memory cell Mij is set to "0", one of the source and the drain is connected to the bit line BLj, and the other is connected to the ground power wiring VSS. On the other hand, when the storage data of the memory cell Mij is set to "1", both the source and the drain are connected to the bit line BLj or the ground power wiring VSS.
[0130] The column decoder 2 is composed of N-type MOS transistors Cj. The drains of all the N-type MOS transistors Cj are connected in common, the gates are connected to the column selection signal lines CLj, and the sources are connected to the bit lines BLj.
[0131] The sense amplifier 18 includes a precharge P-type MOS transistor 5, an inverter 8 for determining the output data of the memory cell Mij, and an inverter 9 for buffering the output signal of the inverter 8. The precharge signal NPR is input to the gate of the P-type MOS transistor 5, the power supply voltage VDD is supplied to the source, and the drain is connected to the common drain of the N-type MOS transistor Cj. The inverter 8 receives the signal SIN of the common drain of the N-type MOS transistor Cj and determines the output data of the memory cell Mij. The inverter 9 receives the output signal SOUT of the inverter 8 and outputs the storage data of the memory cell Mij.
[0132] right Fig.13 Here, the operation of the mask ROM is described. Here, the case of reading the data of the memory cells M00 and M10 is described as an example.
[0133] First, in the column selection signal line CLj, CL0 is set to a high level, and the other CL1 to CLn are set to a low level. As a result, C0 of the transistors constituting the column decoder 2 is turned on, and the other C1 to Cn are turned off. The word line WL0 is transitioned from a non-selected state, i.e., a low level, to a selected state, i.e., a high level.
[0134] Next, the precharge signal NPR is changed from a high level to a low level, so that the precharge P-type MOS transistor is turned on.
[0135] One of the source and drain of the memory cell M00 is connected to the bit line BL0, and the other is connected to the ground power wiring VSS. Therefore, the current flows from the bit line BL0 to the ground power wiring VSS via the memory cell M00, so the input signal SIN of the inverter 8 becomes a voltage lower than the switching level of the inverter 8. Therefore, the output signal SOUT of the inverter 8 is maintained at a high level, and the output signal OUT of the inverter 9 is maintained at a low level.
[0136] When data of the memory cell M10 is read, the word line WL1 is transitioned from a low level in a non-selected state to a high level in a selected state.
[0137] In the memory cell M10, both the source and the drain are connected to the bit line BL0. Therefore, no current flows in the bit line BL0, so the input signal SIN of the inverter 8 becomes a voltage higher than the switching level of the inverter 8. Therefore, the output signal SOUT of the inverter 8 becomes a low level, and the output signal OUT of the inverter 9 becomes a high level.
[0138] That is, when one of the source and drain of the memory cell is connected to the bit line and the other is connected to the ground power wiring, a low level is output (storage data "0"), and when both the source and drain of the memory cell are connected to the bit line or the ground power wiring, a high level is output (storage data "1").
[0139] Fig.14 It is a diagram showing an example of the layout structure of the mask ROM according to the third embodiment and is a plan view of a memory cell array. Fig.14 This corresponds to a (4×4)-bit layout. The dotted line represents a one-bit storage unit frame. Fig.14 , a structure in which four storage cells are arranged in the X direction and four storage cells are arranged in the Y direction is shown. For example, the two storage cells from the left of the bottommost column in the figure are respectively equivalent to Fig.13 Storage units M00 and M01 in the circuit diagram.
[0140] Fig.14 The structure is basically the same as that shown in the first embodiment. Figure 2 The structure is the same as that of the first embodiment. However, due to the difference in the circuit structure of the mask ROM, the position of the contact and the like are also different. No dummy gate wiring for supplying VSS is arranged. Hereinafter, the description of the same structure as that of the first embodiment is sometimes omitted. Figure 3 Similarly, the cross-sectional view is omitted here.
[0141] The memory cells M00 and M01 each have a nanosheet 311 and 312 composed of three thin sheets as a channel portion. That is, the memory cells M00 and M01 include a nanosheet FET. The nanosheets 311 and 312 face each other in the X direction.
[0142] exist Fig.14 In the figure, pads 321a and 322a are formed on the lower side of the nanosheets 311 and 312, respectively, and are composed of a semiconductor layer of an integral structure connected to three thin sheets. Pads 321b and 322b are formed on the upper side of the nanosheets 311 and 312, respectively, and are composed of a semiconductor layer of an integral structure connected to three thin sheets. Pads 321a and 321b become nodes of memory cell M00. Pads 322a and 322b become nodes of memory cell M01.
[0143] Gate wirings 331 and 332 extending in the X direction are formed. The gate wiring 331 surrounds the outer periphery of the nanosheet 311 of the memory cell M00 in the X direction and the Z direction through the gate insulating film. The gate wiring 331 becomes the gate of the memory cell M00. The gate wiring 332 surrounds the outer periphery of the nanosheet 312 of the memory cell M01 in the X direction and the Z direction through the gate insulating film. The gate wiring 332 becomes the gate of the memory cell M01. It should be noted that the gate wirings 331 and 332 are connected to other gate wirings arranged in a row in the X direction to form the word line WL0. That is, the word lines WL0 to WL3 each include gate wirings arranged in a row in the X direction.
[0144] Local wirings 341, 342, 343, and 344 extending in the X direction are formed. The local wiring 341 is connected to the pad 321a, and the local wiring 342 is connected to the pad 321b. The local wiring 343 is connected to the pad 322a, and the local wiring 344 is connected to the pad 322b.
[0145] Contacts 371, 372, 373, and 374 determine the storage value of the memory cell according to their positions. That is, when the two nodes of the memory cells M00 and M01 are connected to VSS via the contacts, or when the two nodes are connected to the bit line BL via the contacts, the memory cells M00 and M01 have storage data "1". On the other hand, when one node of the memory cells M00 and M01 is connected to VSS via the contacts, and the other node is connected to the bit line BL via the contacts, the memory cells M00 and M01 have storage data "0".
[0146] exist Fig.14In the structure of , one node of the memory cell M00 is connected to the M1 wiring 61 supplied with VSS via the contact 371, and the other node is connected to the M1 wiring 62 equivalent to the bit line BL0 via the contact 372, and thus has the stored data "0". One node of the memory cell M01 is connected to the M1 wiring 63 supplied with VSS via the contact 373, and the other node is connected to the M1 wiring 64 equivalent to the bit line BL1 via the contact 374, and thus has the stored data "0".
[0147] Here, the surface of the nanosheet 311 close to the nanosheet 312 in the X direction is not covered by the gate wiring 331, but exposed from the gate wiring 331. The surface of the nanosheet 312 close to the nanosheet 311 in the X direction is not covered by the gate wiring 332, but exposed from the gate wiring 332. That is, the memory cells M00 and M01 are fork-sheet FETs. Among them, the gate wiring 331 and the gate wiring 332 are connected by a bridge 333 formed between the gate wiring 331 and the gate wiring 332. The bridge 333 is an example of a gate connection portion.
[0148] As a result, the space required between the nanosheet 311 and the nanosheet 312 is reduced, so that the distance between the nanosheet 311 and the nanosheet 312 can be reduced. Therefore, the area of the semiconductor memory device having the fork-sheet FET can be reduced.
[0149] As described above, according to the present embodiment, the semiconductor memory device includes ROM memory cells M00 and M01 adjacent to each other in the X direction. The ROM memory cell M00 includes a nanosheet FET, which is arranged between the bit line 62 and the ground power wiring 61 and has a nanosheet 311 as a channel region. The ROM memory cell M01 includes a nanosheet FET, which is arranged between the bit line 64 and the ground power wiring 63 and has a nanosheet 312 as a channel region. The nanosheets 311 and 312 face each other in the X direction, and the surface of the nanosheet 311 close to the nanosheet 312 in the X direction is exposed from the gate wiring 331, and the surface of the nanosheet 312 close to the nanosheet 311 in the X direction is exposed from the gate wiring 332. As a result, the distance between the nanosheet 311 and the nanosheet 312 can be reduced, so that the area of the semiconductor memory device can be reduced.
[0150] The M1 wirings 61 to 68 extending in the Y direction are arranged at equal intervals. Furthermore, the M1 wirings 62, 64, 66, and 68 corresponding to the bit lines BL0 to BL3 are sandwiched by the M1 wirings 61, 63, 65, and 67 supplied with VSS. By sandwiching the bit lines BL between the VSS wirings, crosstalk between the bit lines BL can be prevented, thereby suppressing malfunctions. Since the distances from the bit lines BL to the VSS wirings on both sides are equal for any bit line BL, the load capacitance caused by the wiring capacitance of the bit lines BL is consistent. Therefore, performance deviations such as operating speed between the bit lines BL are suppressed.
[0151] It should be noted that the M1 wirings 61 to 68 extending in the Y direction may not be arranged at equal intervals. Even in this case, if the distance from the bit line BL to the VSS wirings on both sides is equal for any bit line BL, the load capacitance caused by the wiring capacitance of the bit line BL is consistent, so the performance deviation such as the operating speed between the bit lines BL is suppressed. For example, even if the M1 wirings 61, 63, 65, and 67 supplying VSS are all offset by the same distance in the X direction, the load capacitance caused by the wiring capacitance of each bit line BL is consistent, so the performance deviation such as the operating speed between the bit lines BL is suppressed.
[0152] <Modification 1>
[0153] Fig.15 It is a plan view of a memory cell array according to Modification 1 of the third embodiment. Fig.15 The layout structure is basically the same as Fig.14 The layout structure is the same. However, in Fig.15 In the embodiment, as in the first variant of the first embodiment, the positions of the M1 wirings 62, 64, 66, and 68 corresponding to the bit lines BL0 to BL3 are consistent with each other relative to the nanosheet. The positions of the M1 wirings 61, 63, 65, and 67 supplying VSS are also consistent with each other relative to the nanosheet. As the layout positions of the M1 wirings 61 to 68 are changed, the positions of the contacts and the lengths of the local wirings are changed. Fig.14 different.
[0154] In the space created by aligning the arrangement positions of the bit lines BL with respect to the nanosheets, M1 wirings 69a, 69b, and 69c for supplying VSS are further arranged. As a result, for any bit line BL, the distance to the VSS wirings on both sides can be made consistent, so the load capacitance for the bit line BL can be made consistent. It should be noted that M1 wirings 69a, 69b, and 69c for supplying VSS may not be additionally arranged.
[0155] and Figure 5 Similarly to the structure of FIG. 1 , a thicker M1 wiring may be arranged instead of two adjacent M1 wirings for supplying VSS.
[0156] <Modification 2>
[0157] Fig.16 It is a plan view of a memory cell array according to Modification 2 of the third embodiment. Fig.16 The layout structure and Fig.15 The layout structure is roughly the same. However, in Fig.16 In the embodiment, as in the second variation of the first embodiment, the positions of the M1 wirings 61 and 62 are reversed, and the positions of the M1 wirings 65 and 66 are reversed. As a result, the positions of the bit lines BL are consistent with respect to the portion of the fork exposed from the gate wiring. That is, the M1 wirings 62, 64, 66, and 68 corresponding to the bit lines BL0 to BL3 are all located on the opposite side of the fork exposed from the gate wiring. As a result, the characteristics of each bit line BL can be made more consistent.
[0158] <Variation 3>
[0159] Fig.17 1 is a top view of a memory cell array according to a third variant of the third embodiment. In this variant, as in the third variant of the first embodiment, the transistor of each bit of the memory cell is composed of two fork FETs arranged in the X direction and connected in parallel. The following describes the structure of the memory cell by taking the memory cells M00 and M01 as examples.
[0160] The memory cell M00 has three nanosheets 411 and 412 as a channel portion, and the memory cell M01 has three nanosheets 413 and 414 as a channel portion.
[0161] exist Fig.17 In the figure, pads 421a and 422a are formed on the lower side of the nanosheets 411 and 412, respectively, and are composed of a semiconductor layer of an integral structure connected to three sheets. Pads 421b and 422b are formed on the upper side of the nanosheets 411 and 412, respectively, and are composed of a semiconductor layer of an integral structure connected to three sheets. Pads 421a, 421b, 422a, and 422b become nodes of the memory cell M00. Pads 423a and 424a are formed on the lower side of the nanosheets 413 and 414, respectively, and are composed of a semiconductor layer of an integral structure connected to three sheets. Pads 423b and 424b are formed on the upper side of the nanosheets 413 and 414, respectively, and are composed of a semiconductor layer of an integral structure connected to three sheets. The pads 423a, 423b, 424a, and 424b serve as nodes of the memory cell M01.
[0162] Gate wirings 431, 432, 433, and 434 extending in the X direction are formed. The gate wiring 431 surrounds the outer periphery of the nanosheet 411 of the memory cell M00 in the X direction and the Z direction via a gate insulating film (not shown). The gate wiring 432 surrounds the outer periphery of the nanosheet 412 of the memory cell M00 in the X direction and the Z direction via a gate insulating film (not shown). The gate wirings 431 and 432 serve as the gate of the memory cell M00. The gate wiring 433 surrounds the outer periphery of the nanosheet 413 of the memory cell M01 in the X direction and the Z direction via a gate insulating film (not shown). The gate wiring 434 surrounds the outer periphery of the nanosheet 414 of the memory cell M01 in the X direction and the Z direction via a gate insulating film (not shown). The gate wirings 433 and 434 serve as the gate of the memory cell M01.
[0163] Here, the surface of the nanosheet 411 in the X direction close to the nanosheet 412 is not covered by the gate wiring 431, but exposed from the gate wiring 431. The surface of the nanosheet 412 in the X direction close to the nanosheet 411 is not covered by the gate wiring 432, but exposed from the gate wiring 432. The surface of the nanosheet 413 in the X direction close to the nanosheet 414 is not covered by the gate wiring 433, but exposed from the gate wiring 433. The surface of the nanosheet 414 in the X direction close to the nanosheet 413 is not covered by the gate wiring 434, but exposed from the gate wiring 434.
[0164] Thus, the space required between nanosheet 411 and nanosheet 412 is reduced, and the distance between nanosheet 411 and nanosheet 412 can be reduced. Similarly, the space required between nanosheet 413 and nanosheet 414 is reduced, and the distance between nanosheet 413 and nanosheet 414 can be reduced.
[0165] The gate wiring 431 and the gate wiring 432 are connected by a bridge 435 formed between the gate wiring 431 and the gate wiring 432. The gate wiring 433 and the gate wiring 434 are connected by a bridge 436 formed between the gate wiring 433 and the gate wiring 434. The bridges 435 and 436 are examples of gate connecting portions.
[0166] Local wirings 441, 442, 443, and 444 extending in the X direction are formed. Local wiring 441 is connected to pads 421a and 422a, and local wiring 442 is connected to pads 421b and 422b. That is, the drains of the two nanosheet FETs of the memory cell M00 are connected to each other, and the sources are connected to each other. Local wiring 443 is connected to pads 423a and 424a, and local wiring 444 is connected to pads 423b and 424b. That is, the drains of the two nanosheet FETs of the memory cell M01 are connected to each other, and the sources are connected to each other.
[0167] Contacts 471, 472, 473, and 474 determine the storage value of the memory cell according to their positions. That is, when the two nodes of the memory cells M00 and M01 are connected to VSS via the contacts, or when the two nodes are connected to the bit line BL via the contacts, the memory cells M00 and M01 have storage data "1". On the other hand, when one node of the memory cells M00 and M01 is connected to VSS via the contacts, and the other node is connected to the bit line BL via the contacts, the memory cells M00 and M01 have storage data "0".
[0168] exist Fig.17 In the structure of , one node of the memory cell M00 is connected to the M1 wiring 61 supplied with VSS via the contact 471, and the other node is connected to the M1 wiring 62 equivalent to the bit line BL0 via the contact 472, so the memory cell M00 has the storage data "0". One node of the memory cell M01 is connected to the M1 wiring 63 supplied with VSS via the contact 473, and the other node is connected to the M1 wiring 64 equivalent to the bit line BL1 via the contact 474, so the memory cell M01 has the storage data "0".
[0169] As described above, according to this variation, the semiconductor memory device includes a ROM memory cell M00. The ROM memory cell M00 includes: a first nanosheet FET provided between the bit line 62 and the ground power wiring 61 and having a nanosheet 411 as a channel region; and a second nanosheet FET provided between the bit line 62 and the ground power wiring 61 and having a nanosheet 412 as a channel region. The nanosheets 411 and 412 face each other in the X direction, and the surface of the nanosheet 411 close to the nanosheet 412 in the X direction is exposed from the gate wiring 431, and the surface of the nanosheet 412 close to the nanosheet 411 in the X direction is exposed from the gate wiring 432. Thus, the distance between the nanosheets 411 and 412 can be reduced, so that the size of the ROM memory cell M00 in the X direction can be reduced, and the area of the semiconductor memory device can be reduced. On the other hand, in the X direction, the distance between the ROM memory cells can be widened, so that the local wiring can be easily separated, and the manufacturing ease of the semiconductor memory device is improved.
[0170] By sandwiching the bit line BL between the VSS wiring, crosstalk between the bit lines BL can be prevented, thereby suppressing malfunction. Since the distance from the bit line BL to the VSS wiring on both sides is equal for any bit line BL, the load capacitance caused by the wiring capacitance of the bit line BL is consistent. Therefore, performance deviations such as operating speed between the bit lines BL are suppressed. Furthermore, the layout position of the bit line BL relative to the nanosheet is consistent, so the characteristics of each bit line BL can be made consistent.
[0171] (Fourth Embodiment)
[0172] In the fourth embodiment, it is indicated Fig.13 Another example of the layout structure of the mask ROM. In this embodiment, instead of the fork sheet FET, a nanosheet FET is used.
[0173] Fig.18 It is a diagram showing an example of the layout structure of the mask ROM according to the fourth embodiment, and is a plan view of a memory cell array. Fig.18 This corresponds to a (4×4)-bit layout. The dotted line represents a one-bit storage unit frame. Fig.18 , a structure in which four storage cells are arranged in the X direction and four storage cells are arranged in the Y direction is shown. For example, the two storage cells from the left of the bottommost column in the figure are respectively equivalent to Fig.13 Storage units M00 and M01 in the circuit diagram.
[0174] The memory cells M00 and M01 each have a nanosheet 311A and 312A composed of three thin sheets as a channel portion. That is, the memory cells M00 and M01 include a nanosheet FET. Both sides of the surface of the nanosheet 311A in the X direction are covered by the gate wiring 331A. Both sides of the surface of the nanosheet 312A in the X direction are covered by the gate wiring 332A. Therefore, the spacing between the nanosheets 311A and 312A is equal to the spacing between other nanosheets.
[0175] The M1 wirings 61 to 68 extending in the Y direction are arranged at equal intervals. Furthermore, the M1 wirings 62, 64, 66, and 68 corresponding to the bit lines BL0 to BL3 are sandwiched by the M1 wirings 61, 63, 65, and 67 supplied with VSS. By sandwiching the bit lines BL between the VSS wirings, crosstalk between the bit lines BL can be prevented, thereby suppressing malfunctions. Since the distances from the bit lines BL to the VSS wirings on both sides are equal for any bit line BL, the load capacitance caused by the wiring capacitance of the bit lines BL is consistent. Therefore, performance deviations such as operating speed between the bit lines BL are suppressed.
[0176] Furthermore, the layout positions of the bit lines BL with respect to the nanosheets are aligned, thereby making it possible to make the characteristics of the bit lines BL uniform.
[0177] It should be noted that in the above embodiments and variations, each nanosheet is composed of three sheet structures, but the invention is not limited thereto, and a part of the nanosheets or all of the nanosheets may also include one, two or more than four sheet structures.
[0178] In the above embodiments, the cross-sectional shape of the nanosheet is a rectangle, but is not limited thereto and may be, for example, a square, a circle, an ellipse, or the like.
[0179] -Industrial Applicability-
[0180] In the present disclosure, a semiconductor memory device using on-chip FETs can realize a layout structure with a small area, and thus is useful for, for example, miniaturization of semiconductor chips and improvement of integration.
[0181] -Explanation of symbols-
[0182] 11.12 Nanosheets
[0183] 11A, 12A Nanosheets
[0184] 21a, 21b, 22a, 22b pads (nodes)
[0185] 31, 32 Gate wiring
[0186] 31A, 32A Gate wiring
[0187] 33 Gate connection
[0188] 61, 63, 65, 67 M1 wiring (ground power wiring)
[0189] 62, 64, 66, 68 M1 wiring (bit line)
[0190] 111, 112, 113, 114 Nanosheets
[0191] 121a, 121b, 122a, 122b, 123a, 123b, 124a, 124b pads (nodes)
[0192] 131, 132, 133, 134 Gate wiring
[0193] 135, 136 Gate connection portion
[0194] 211, 212, 213, 214 Nanosheets
[0195] 211A, 212A, 213A, 214A Nanosheets
[0196] 221a, 221b, 221c, 222a, 222b, 222c pads (nodes)
[0197] 311, 312 Nanosheets
[0198] 311A, 312A Nanosheets
[0199] 331, 332 Gate wiring
[0200] 331A, 332A Gate wiring
[0201] 333 Gate connection
[0202] 411, 412, 413, 414 Nanosheets
[0203] 421a, 421b, 422a, 422b, 423a, 423b, 424a, 424b pads (nodes)
[0204] 431, 432, 433, 434 Gate wiring
[0205] 435, 436 Gate connection portion
[0206] BL bit line
[0207] M00, M01 ROM storage unit
[0208] WL Word Line
Claims
1. A semiconductor memory device comprising a first ROM (Read Only Memory) memory cell and a second ROM memory cell adjacent to each other in a first direction, characterized in that: include: word lines extending in the first direction; a first bit line and a second bit line extending in a second direction perpendicular to the first direction; as well as a first ground power wiring and a second ground power wiring extending in the second direction, The first ROM storage unit includes a first nanosheet FET (Field Effect Transistor), The first nanosheet FET is provided between the first bit line and the first ground power wiring and has a first nanosheet as a channel region, The second ROM storage cell comprises a second nanosheet FET, The second nanosheet FET is provided between the second bit line and the second ground power wiring and has a second nanosheet as a channel region, The word line includes a first gate wiring and a second gate wiring, the first gate wiring surrounds the outer periphery of the first nanosheet in the first direction and a third direction, the third direction being perpendicular to the first direction and the second direction, and the second gate wiring surrounds the outer periphery of the second nanosheet in the first direction and the third direction, The first ROM storage unit stores first data according to the connection relationship between the node of the first nanosheet FET and the first bit line and the first ground power wiring, and the second ROM storage unit stores second data according to the connection relationship between the node of the second nanosheet FET and the second bit line and the second ground power wiring. The first nanosheet and the second nanosheet face each other in the first direction, and a surface of the first nanosheet close to the second nanosheet in the first direction is exposed from the first gate wiring, and a surface of the second nanosheet close to the first nanosheet in the first direction is exposed from the second gate wiring.
2. The semiconductor memory device according to claim 1, wherein: The first bit lines and the second bit lines, and the first ground power wiring and the second ground power wiring are arranged at equal intervals in the first direction.
3. The semiconductor memory device according to claim 1, wherein: The word line includes a gate connection portion formed between the first gate wiring and the second gate wiring and connecting the first gate wiring and the second gate wiring.
4. The semiconductor memory device according to claim 1, wherein: The first nanosheet FET is composed of two nanosheet FETs arranged adjacent to each other in the second direction and sharing a source. The second nanosheet FET is composed of two nanosheet FETs that are adjacently arranged in the second direction and share a source.
5. The semiconductor memory device according to claim 1, wherein: The first nanosheet and the second nanosheet are composed of a single sheet structure or a plurality of sheet structures that overlap when viewed from above.
6. The semiconductor memory device according to claim 1, wherein: The first ROM storage unit stores first data according to whether the drain of the first nanosheet FET is connected to the first bit line or whether the source of the first nanosheet FET is connected to the first ground power wiring. The second ROM storage unit stores second data according to the presence or absence of connection between the drain of the second nanosheet FET and the second bit line or the presence or absence of connection between the source of the second nanosheet FET and the second ground power wiring.
7. The semiconductor memory device according to claim 1, wherein: The first ROM storage unit stores first data according to whether the connection destinations of the source and the drain of the first nanosheet FET are the same line or different lines in the first bit line and the first ground power wiring, The second ROM storage unit stores second data according to whether the connection destinations of the source and the drain of the second nanosheet FET are the same line or different lines among the second bit line and the second ground power wiring.
8. A semiconductor storage device comprising a ROM (Read Only Memory) storage unit, characterized in that: include: word lines extending in a first direction; a bit line extending in a second direction perpendicular to the first direction; as well as a ground power supply wiring extending in the second direction, The ROM storage unit comprises: a first nanosheet FET disposed between the bit line and the ground power wiring and having a first nanosheet as a channel region; as well as a second nanosheet FET provided between the bit line and the ground power wiring and having a second nanosheet as a channel region, The first nanosheet FET and the second nanosheet FET are adjacent to each other in the first direction, the drains of the first nanosheet FET and the second nanosheet FET are connected to each other, and the sources of the first nanosheet FET and the second nanosheet FET are connected to each other, The word line includes a first gate wiring and a second gate wiring, the first gate wiring surrounds the outer periphery of the first nanosheet in the first direction and a third direction, the third direction being perpendicular to the first direction and the second direction, and the second gate wiring surrounds the outer periphery of the second nanosheet in the first direction and the third direction, The ROM storage unit stores data according to the connection relationship between the nodes of the first nanosheet FET and the second nanosheet FET and the bit line and the ground power wiring. The first nanosheet and the second nanosheet face each other in the first direction, and a surface of the first nanosheet close to the second nanosheet in the first direction is exposed from the first gate wiring, and a surface of the second nanosheet close to the first nanosheet in the first direction is exposed from the second gate wiring.
9. The semiconductor memory device according to claim 8, wherein: The word line includes a gate connection portion formed between the first gate wiring and the second gate wiring and connecting the first gate wiring and the second gate wiring.
10. The semiconductor memory device according to claim 8, wherein: The first nanosheet and the second nanosheet are composed of a single sheet structure or a plurality of sheet structures that overlap when viewed from above.
11. The semiconductor memory device according to claim 8, wherein: The ROM storage unit stores data according to whether the drains of the first nanosheet FET and the second nanosheet FET are connected to the bit line or whether the sources of the first nanosheet FET and the second nanosheet FET are connected to the ground power wiring.
12. The semiconductor memory device according to claim 8, wherein: The ROM storage unit stores data depending on whether the connection destinations of the source and the drain of the first nanosheet FET and the second nanosheet FET are the same line or different lines among the bit line and the ground power wiring.
