Transistor units, transistor circuits, sub-word line drivers, and semiconductor devices

By employing transistor cells with a common gate structure and sub-word line drivers in semiconductor memory devices, the problems of word line voltage speed delay and gate oxide layer cracking are solved, thereby improving high-voltage heat resistance and device stability.

CN114512486BActive Publication Date: 2026-11-17SAMSUNG ELECTRONICS CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202110975133.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2020-11-16
Filing Date
2021-08-24
Publication Date
2026-11-17
Estimated Expiration
2041-08-24

AI Technical Summary

Technical Problem

In semiconductor memory devices, as DRAM capacity increases, the number of memory cells connected to a word line increases, and the spacing between word lines decreases, resulting in word line voltage speed delay. Existing technologies have difficulty effectively solving this problem, and there is also a risk of gate oxide layer cracking.

Method used

The transistor cell employs a common gate structure, which separates the active region into two parts through a separation region and arranges an inverted pie-shaped gate above the active region to prevent the gate oxide layer from cracking. At the same time, a sub-word line driver is used to manage the defective drain through the common gate structure to ensure that the channel length remains long.

Benefits of technology

It effectively prevents the gate oxide layer from cracking, improves the heat resistance at high voltages, reduces the short-channel effect, and enhances the reliability and stability of semiconductor devices.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114512486B_ABST
    Figure CN114512486B_ABST
Patent Text Reader

Abstract

Transistor cells, transistor circuits, sub-word line drivers, and semiconductor devices are provided. A transistor circuit with a common gate structure includes an active region and a gate. The active region has a body extending on a substrate along a first direction and a protrusion extending from the central portion of the body in the first direction along a second direction perpendicular to the first direction. The gate is disposed above the active region to overlap with the channel region of the active region and has an inverted pie structure, which, in plan view, surrounds but does not cover portions of the active region including the two corner portions of the active region on three sides. The protrusion is divided by a separation region into a first portion and a second portion, which are separated into two sub-parts. Opposite ends of the body in the first direction correspond to two drain regions, the second portion of the protrusion corresponds to a common source region, and the two drain regions, the common source region, and the gate constitute two transistors, wherein the two transistors share a common gate.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] This application is based on and claims priority to Korean Patent Application No. 10-2020-0153077, filed on November 16, 2020, with the Korean Intellectual Property Office, the entire disclosure of which is incorporated herein by reference. Technical Field

[0002] The inventive concept relates to a semiconductor memory device, and more specifically, to a transistor cell having a common gate structure, as well as a sub-word line driver and semiconductor device based on the transistor cell. Background Technology

[0003] Driven by user demands for high performance, the capacity and speed of semiconductor memory devices used in various electronic systems are rapidly increasing. Semiconductor memory devices can be broadly classified into volatile memory devices and non-volatile memory devices. A representative example of a volatile memory device is Dynamic Random Access Memory (DRAM), and a representative example of a non-volatile memory device is Flash memory. As the capacity of DRAM increases, the number of memory cells connected to a word line increases, and the spacing between word lines (cell pitch) decreases. Speed ​​delay occurs when the word line voltage applied to the word line is supplied to more memory cells. To address the word line voltage speed delay, a method is used to divide a word line into multiple sub-word lines and drive each sub-word line with a sub-word line driver (SWD). Summary of the Invention

[0004] The inventive concept provides a transistor cell capable of isolating and managing the repair of defective drains while maintaining a long channel length and preventing the gate oxide layer from bursting, as well as a sub-word line driver and semiconductor device based on the transistor cell.

[0005] According to some embodiments, a transistor cell with a common gate structure includes an active region, a separation region, and a gate. The active region has a body extending along a first direction parallel to the top surface of a substrate and a protrusion extending along a second direction perpendicular to the first direction and parallel to the top surface of the substrate. The protrusion is connected to a central portion of the body that is centered relative to the first direction and extends away from the body in the second direction. The separation region extends through the body along the second direction to have a first end located in the protrusion and a second end located in the body and opposite to the first end in the second direction. The separation region separates the body into two parts and the first portion of the protrusion into two portions, one portion located on each side of the separation region. The gate is arranged above the active region in a third direction perpendicular to the first and second directions to cover the channel region of the active region, and has a structure at a channel avoidance region of the active region adjacent to the second end of the separation region that exposes the top of the active region relative to the gate. The active region is divided into a first active region and a second active region by the separation region, wherein the first active region is connected to the second active region via a second portion of the protrusion. The opposite ends of the main body in the first direction correspond to two drain regions, the second part corresponds to a common source region, and the two drain regions, the common source region and the gate constitute two transistors, wherein the two transistors share a common gate.

[0006] According to some embodiments, a transistor circuit having a common gate structure includes an active region and a gate. The active region has a body extending on a substrate along a first direction and a protrusion extending from a central portion of the body in the first direction along a second direction perpendicular to the first direction. The gate is disposed above the active region to be superimposed on the channel region of the active region and has an inverted piezoelectric point. Structure, inverted pie The structure, viewed in plan view, consists of a portion surrounding but not covering the active region on three sides, including the two corner portions of the active region. The active region is divided into a first active region and a second active region by a separation region extending along a second direction and separating the main body and a portion of the protrusion. The protrusion is divided into a first part and a second part. The first part is further divided into two sub-parts by the separation region, wherein the first part is located between the main body and the second part in the second direction. The opposite ends of the main body in the first direction correspond to two drain regions, the second part of the protrusion corresponds to a common source region, and the two drain regions, the common source region, and the gate constitute two transistors, wherein the two transistors share a common gate.

[0007] According to some embodiments, a semiconductor device includes a common gate structure. The semiconductor device may include a first active region portion extending along a first direction parallel to a top surface of a substrate and a protrusion projecting from the first active region portion and extending along a second direction perpendicular to the first direction and parallel to the top surface of the substrate. A separation region may extend from a first end to a second end along the second direction, dividing the first active region portion into a first active portion located on a first side of the separation region and a second active portion located on a second opposite side of the separation region, and dividing the first portion of the protrusion into a first protrusion located on a first side of the separation region and a second protrusion located on a second side of the separation region, wherein the first end of the separation region is located on the active region, and the second end of the separation region is located on the first portion of the protrusion. Two 3D corner portions of the active region may be adjacent to the first end of the separation region, and a gate may be disposed above the active region to cover the channel region of the active region while exposing the two 3D corner portions. In various embodiments, the separation region extends beyond a first edge of the gate in the second direction, and also extends beyond a second edge of the gate opposite to the first edge in the second direction. Furthermore, a portion of the gate extends beyond the separation region and beyond the active region in the second direction, and two 3D corner portions may be located between the first portion of the gate and the second portion of the gate in the first direction.

[0008] According to some embodiments, a sub-word line driver includes: a first sub-word line driver having a first transistor for supplying a negative voltage to a first word line in response to a drive signal; and a second sub-word line driver having a second transistor for supplying a negative voltage to a second word line in response to a drive signal, wherein the first transistor and the second transistor constitute a transistor unit, the transistor unit including a pi with an inverted piezoelectric element. The gate is shaped like a first transistor and the second transistor share the same gate.

[0009] According to some embodiments, a semiconductor device includes: a cell region wherein a plurality of memory cells are arranged in a two-dimensional array structure; and a core / peripheral region surrounding the cell region and including driving circuitry for driving the memory cells, wherein a sub-word line driver, as one of the driving circuitry, is located in the core / peripheral region and includes: a first sub-word line driver having a first transistor for supplying a negative voltage to a first word line in response to a driving signal; and a second sub-word line driver having a second transistor for supplying a negative voltage to a second word line in response to a driving signal, wherein the first transistor and the second transistor constitute a transistor cell, the transistor cell including a picometer with an inverted picometer. The gate is shaped like a first transistor and the second transistor share the same gate. Attached Figure Description

[0010] Embodiments of the inventive concept will become clearer from the following detailed description taken in conjunction with the accompanying drawings, in which:

[0011] Figure 1A and Figure 1B These are block diagrams and circuit diagrams of the sub-word line driver according to the embodiments;

[0012] Figure 2 This is a plan view of a portion of the transistor cell according to an embodiment;

[0013] Figures 3A to 3C They are along Figure 2 The cross-sectional view taken by line I-I', the plan view of the active region and the plan view of the gate;

[0014] Figures 4A to 5B These are the inverted pie numbers used to explain the transistor cells in some embodiments. Perspective view and SEM image of the advantages of the gate shape;

[0015] Figure 6 This is a graph showing the results of measuring time-dependent dielectric breakdown (TDDB) using a drive signal PXIB for transistor cells in some embodiments and comparative examples;

[0016] Figures 7A to 7F This is a plan view of a portion of the transistor cell according to an embodiment;

[0017] Figure 8A and Figure 8B This is a plan view of a portion of the transistor cell according to an embodiment;

[0018] Figure 9A and Figure 9B These are, respectively, a plan view and a SEM image of a pair of transistor units in which two transistor units according to an embodiment are arranged to share a common source region;

[0019] Figures 10A to 10C It is used for explanation Figure 9A A comparative example of the advantages of a pair of transistor cells; a plan view of a pair of transistor cells; and

[0020] Figure 11A and Figure 11B These are a plan view of a semiconductor device according to an embodiment and a block diagram showing the structure of a portion of the semiconductor device, respectively. Detailed Implementation

[0021] The present disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote the same elements, and therefore, repeated explanations of the same elements are omitted.

[0022] Figure 1A and Figure 1B These are block diagrams and circuit diagrams of the sub-word line driver according to the embodiments.

[0023] Reference Figure 1A and Figure 1B Each of the sub-word line drivers 200-1 (or referred to as "SWD1") and 200-2 (or referred to as "SWD2") according to this embodiment may include three transistors. For example, the first sub-word line driver 200-1 may include a first pull-up transistor PUT1, a first pull-down transistor PDT1, and a first transistor KT1, and the second sub-word line driver 200-2 may include a second pull-up transistor PUT2, a second pull-down transistor PDT2, and a second transistor KT2. It can be seen that ordinal numbers such as "first," "second," and "third" can be simply used as labels for certain elements, steps, etc., to distinguish them from each other. Terms not described using "first," "second," etc., in the specification may still be referred to as "first" or "second" in the claims. Furthermore, a term referenced with a specific ordinal number (e.g., "first" in a specific claim) may be described elsewhere with a different ordinal number (e.g., "second" in the specification or another claim).

[0024] The operation of the sub-word line drivers 200-1 and 200-2 according to this embodiment will be described below. The sub-word line drivers 200-1 and 200-2 can be driven from a drive voltage generator (see...) Figure 11B 400-1) Receive drive signal PXID <2> and PXIB <2> The first sub-word line driver 200-1 can respond to the word line enable signal NWEIB. <0> When activated, the second sub-word line driver 200-2 can respond to the word line enable signal NWEIB. <1> And thus activated. More specifically, when the word line enable signal NWEIB is provided at a low level. <0> At that time, the first sub-word line driver 200-1 can transmit the drive signal PXID. <2> The high voltage VPP provided is supplied to the word line WL <1> Additionally, when the word line enable signal NWEIB is provided at a higher level... <0> At that time, the first sub-word line driver 200-1 can block the drive signal PXID. <2> And use negative voltage VBB2 to the word line WL <1> Pre-charge is performed. Similarly, when the word line enable signal NWEIB is provided at a low level... <1> At that time, the second sub-word line driver 200-2 can transmit the drive signal PXID. <2> The high voltage VPP provided is supplied to the word line WL <5> Additionally, when the word line enable signal NWEIB is provided at a higher level... <1> At this time, the second sub-word line driver 200-2 can block the drive signal PXID. <2> And use negative voltage VBB2 to the word line WL <5> Perform pre-charging.

[0025] Sub-word line drivers 200-1 and 200-2 may include methods for pre-charging word lines WL <1> and WL <5> Transistors KT1 and KT2 are maintained at a negative voltage VBB2. Considering their functional properties, transistors KT1 and KT2 are also referred to as holding transistors. Transistors KT1 and KT2 can respond to the drive signal PXIB. <2> And the word line WL <1> and WL <5> It is fixed at the level of the negative voltage VBB2. Therefore, even the word line enable signal NWEIB <0> and NWEIB <1> Or, in the event of noise level fluctuations, the word line WL <1> and WL <5> It can also maintain a stable voltage value.

[0026] like Figure 1B As shown, transistors KT1 and KT2 of sub-word line drivers 200-1 and 200-2 can constitute transistor unit 100, also described as a transistor structure. Transistor unit 100 can be a transistor circuit and may include a common source electrode (or referred to as a "source contact") to which a negative voltage VBB2 is applied (see...). Figure 2 (140S) is the drive signal PXIB applied. <2> The common gate electrode (or "gate contact") (see) Figure 2 (140G) and connected to word lines WL respectively <1> and WL <5> The two drain electrodes (or "drain contacts") (see) Figure 2 (140D in the middle).

[0027] In transistor cell 100, the active regions constituting the source region, drain region, and channel region (S, D, C) (see...) Figure 2 110 in the middle) has a separation region (see Figure 2 The 130) separated structure allows the first transistor KT1 and the second transistor KT2 to have separate channel structures, and the repair can be separated and managed. Figure 1B In the diagram, the sawtooth shape indicates that the active region 110 has a channel separated by the separation region 130. Each of the first transistor KT1 and the second transistor KT2 has a "Γ"-shaped channel (also described as L-shaped), thereby maximizing the channel length, improving high-voltage thermal performance, and reducing short-channel effects. Furthermore, in the transistor cell 100, the gate above the active region (see...) Figure 2 In the case of 120), there is an inverted pi (pi, π). The shape effectively prevents the gate oxide layer from cracking due to fields such as the E-field. The gate is concentrated at the edge portion adjacent to the separation region 130. Figures 2 to 3C The structure of transistor unit 100 is described in more detail in the description.

[0028] like Figure 1BAs shown, sub-word line drivers 200-1 and 200-2 may each have a function for driving word lines WL. <1> and WL <5> The circuit structure. Specifically, transistors KT1 and KT2 of sub-word line drivers 200-1 and 200-2 can share a gate to construct transistor unit 100.

[0029] The first sub-word line driver 200-1 can receive the drive signal PXID from the drive voltage generator 400-1. <2> and PXIB <2> And it can be obtained from the line decoder (see Figure 11B The 300) receive word line enable signal NWEIB <0> As described above, the first sub-word line driver 200-1 may include a pull-up transistor PUT1, a pull-down transistor PDT1, and a first transistor KT1. The pull-up transistor PUT1 can respond to the word line enable signal NWEIB. <0> And the word line WL <1> Pull up to drive signal PXID <2> The level. On the other hand, pull-down transistor PDT1 can respond to the word line enable signal NWEIB. <0> And the word line WL <1> Pull down to negative voltage VBB2. When word line WL <1> When deactivated, the first transistor KT1 can activate the word line WL <1> The voltage is maintained at the level of the negative voltage VBB2. Therefore, the first transistor KT1 can respond to the drive signal PXID. <2> PXIB driving signals in a complementary relationship <2> The source, which is supplied with a negative voltage VBB2, is connected to the word line WL. <1> The switch between the drain and the electrode.

[0030] The second sub-word line driver 200-2 can receive the drive signal PXID from the drive voltage generator 400-1. <2> and PXIB <2> It can also receive the word line enable signal NWEIB from the line decoder 300. <1> As described above, the second sub-word line driver 200-2 may include a pull-up transistor PUT2, a pull-down transistor PDT2, and a second transistor KT2. The pull-up transistor PUT2 can respond to the word line enable signal NWEIB. <1> And the word line WL <5> Pull up to drive signal PXID <2> The level. On the other hand, pull-down transistor PDT2 can respond to the word line enable signal NWEIB. <1> And the word line WL <5> Pull down to the negative voltage level VBB2. When word line WL <5> When deactivated, the second transistor KT2 can activate the word line WL <5> The voltage remains negative VBB2. Therefore, the second transistor KT2 can respond to the drive signal PXID. <2> PXIB driving signals in a complementary relationship <2> The source, which is supplied with a negative voltage VBB2, is connected to the word line WL. <5> The switch between the drain and the electrode.

[0031] According to this embodiment, transistors KT1 and KT2 of sub-word line drivers 200-1 and 200-2 can receive a drive signal PXIB through a common gate. <2> They share the same gate voltage. Furthermore, the first transistor KT1 and the second transistor KT2 share the common source region 110S, allowing the same negative voltage VBB2 to be supplied to the common source region 110S.

[0032] In the sub-word line drivers 200-1 and 200-2 of this embodiment, each of the transistors KT1 and KT2 constituting the transistor unit 100 has a "Γ"-shaped channel, allowing the channel length to remain relatively long. This shape can also be described as an "L" shape or a 90° angle shape (e.g., including two straight sections arranged to intersect each other at a 90° angle, wherein one straight section is longer than the other). Furthermore, because the active region 110 has a structure separated by the separation region 130, the repair of transistors KT1 and KT2 can be separated and managed. Additionally, because the gate 120 above the active region has an inverted piezoelectric element... The shape effectively prevents the gate oxide layer from cracking due to field concentration at the edge portion adjacent to the separation region 130. Inverted pie The shape can also be described as a polygonal shape with three openings or a rectangular shape with three openings. The polygonal or rectangular shape with three openings has a first side and a second side, both extending substantially parallel to the direction along which the separation region 130 extends, and positioned opposite each other with the separation region 130 located therebetween; and a third side connecting the first and second sides, extending substantially perpendicular to the first and second sides, and intersecting the separation region 130. Further details of the gate 120 with the rectangular shape of the three openings will be described in more detail below. As used herein, terms such as “same,” “equal,” “planar,” “coplanar,” “parallel,” and “perpendicular” imply similarity or includes close similarity, for example, due to variations that may occur due to manufacturing processes. Unless the context or other statement otherwise indicates otherwise, the term “substantially” may be used herein to emphasize this meaning.

[0033] Figure 2 This is a plan view of a portion of the transistor cell according to an embodiment, and Figures 3A to 3C They are along Figure 2 The cross-sectional view taken along line I-I', the plan view of the active region, and the plan view of the gate are shown. (Refer to...) Figure 1A and Figure 1B Describe them together, and simply describe or omit them in Figure 1A and Figure 1B The content has already been described in the description.

[0034] Reference Figures 2 to 3CAccording to this embodiment, the transistor unit 100 may include an active region 110, a gate 120, and a separation region 130.

[0035] Active region 110 may be formed on substrate 101. Substrate 101 may include, for example, a semiconductor element (such as silicon or germanium (Ge)) or be formed of, for example, a semiconductor element (such as silicon or germanium (Ge)). Alternatively, substrate 101 may include a compound semiconductor (such as silicon carbide (SiC), gallium arsenide (GaAs), indium arsenide (InAs), or indium phosphide (InP)) or be formed of a compound semiconductor. Substrate 101 may have a silicon-on-insulator (SOI) structure. For example, substrate 101 may include a buried oxide (BOX) layer. Conductive regions (e.g., impurity-doped wells or impurity-doped semiconductor structures) may be formed in the upper portion of substrate 101. For example, a well or the upper portion of a well may constitute active region 110. Various device separation structures 150 (such as shallow trench separation (STI) structures), also described as isolation layers, may be formed on substrate 101.

[0036] When in such Figure 2 or Figure 3B When viewed in a plan view, the active area 110 may typically have a "T" shape (also written as "T-shape"), having a vertical portion including the separation area 130 and a horizontal portion including two suspended extensions. Alternatively, the active area 110 may primarily comprise a body 110B and a protrusion 110P. The body 110B may have a shape extending along a first direction (x-direction) and may be divided into a central body 110B1 and an extension 110B2. The extension 110B2 may extend from both ends of the central body 110B1 in the first direction (x-direction) along a second direction (y-direction). The protrusion 110P may extend from the central portion of the body 110B in the first direction (x-direction, e.g., the central portion is centered relative to the x-direction) away from the body 110B along the second direction (y-direction). The protrusion 110P may be divided into a first portion 110P1 separated by the separation area 130 and a second portion 110P2 not separated by the separation area 130. The body 110B forms a "T"-shaped top, and the protrusion 110P forms a "T"-shaped trunk. As can be seen in this embodiment and other embodiments, the gate 120 has an inverted pie shape. The shape is inverted relative to the "T" shape of the active region 110, such that when the active region is oriented to form a "T" shape with the front facing up, the gate 120 has an inverted shape relative to the "T" shape. Shape. Therefore, the separation region 130 extends through the body 110B along the second direction (y direction) to have a first end located in the protrusion 110P and a second end located in the body 110B and opposite to the first end in the second direction. The separation region 130 separates the body 110B into two parts and separates the first part of the protrusion 110P into two portions, one portion located on each side of the separation region 130.

[0037] A common-source region 110S and a drain region 110D, doped with a high concentration of impurity ions, can be formed in the upper portion of the active region 110. The common-source region 110S and the drain region 110D can be used... Figure 2 The implantation mask IIP, indicated by the dashed line, is formed by an ion implantation process. The common source region 110S and the drain region 110D can be formed by separate implantation processes. For example, the drain region 110D can be formed by an ion implantation process using a first type of implantation mask IIP that covers the common source region 110S, and the common source region 110S can be formed by an ion implantation process using a second type of implantation mask IIP that opens the common source region 110S, which is the opposite of the first type. Simultaneously, when forming the common source region 110S, impurity ions can also be doped into the channel avoidance region 110CA.

[0038] The separation region 130 may have a shape that extends from the central portion of the body 110B of the active region 110 in a first direction (x direction) to a second portion 110P that protrudes downward in a second direction (y direction). Furthermore, the upper and lower portions of the separation region 130 in the second direction (y direction) may not be covered by the gate 120 in the third direction (z direction). More specifically, due to the inverted pie of the gate 120... The shape of the separation region 130, the upper portion of which is in the second direction (y direction), can be connected to the channel avoidance region 110CA (see...). Figure 2 Together, they are exposed from the gate 120. Additionally, the lower portion of the separation region 130 in the second direction (y-direction) extends outside the gate 120; therefore, the lower portion of the separation region 130 in the second direction (y-direction) can be exposed from the gate 120 in a manner that protrudes from the lower surface of the gate 120. When viewed in plan view, the active region 110 is considered to be oriented to form a T-shape, an inverted pie. The two vertical legs of the gate 120 extend beyond the top of the T-shape, and the inverted pie The connecting portion of the shape overlaps with the middle vertical portion of the T-shape. In this way, the corner at the top of the T-shape where the separation region 130 is formed is not covered by the gate 120 in the third direction (z direction). The gate 120 is arranged above the active region 110 in the third direction (z direction) perpendicular to the first and second directions to cover the channel region of the active region 110. The gate 120 has a structure that exposes the top of the active region 110 relative to the gate 120 (when viewed in a plan view) at the channel avoidance region 110CA (also described as a channel avoidance area) adjacent to the second end of the separation region 130 of the active region 110.

[0039] If passed Figure 3B As can be seen, the active region 110 can be divided into a first active region 110-1A and a second active region 110-2A by the separation region 130. For example, the central body 110B1 of the body 110B and the first part 110P1 of the protrusion 110P can be divided into two parts by the separation region 130. Because the active region 110 is divided into the first active region 110-1A and the second active region 110-2A by the separation region 130, channel separation can be achieved. For example, the transistor cell 100 may include two channels divided into a channel facing the first active region 110-1A and a channel facing the second active region 110-2A.

[0040] For reference, when the channel is not separated or is separated indefinitely, even if a defect such as a word line bridge occurs only in one of the drains, it is usually impossible to distinguish the repair address, thus requiring the repair of two word lines. Therefore, repair resources are used twice, ultimately leading to a decrease in yield. On the other hand, when the channel separation is ensured, when a defect occurs in one of the drains, only the corresponding word line needs to be repaired, halving the use of repair resources and thus preventing a decrease in yield.

[0041] like Figure 3A As shown, the separation region 130 can have a depth sufficient to separate (e.g., electrically separate) the active region 110. For example, the separation region 130 can have a depth of several hundred nanometers or greater. However, the depth of the separation region 130 is not limited to this. Although not shown, the separation region 130 can be filled with an insulating layer (such as a silicon oxide film, a silicon nitride film, or a silicon oxynitride film). Additionally, Figure 2 The area outside the active region 110 and the gate 120 can also be filled with an insulating layer.

[0042] A gate 120 may be formed over the active region 110. The gate 120 may include a gate electrode 122 and a spacer 124. A gate oxide layer 125 may be located between the gate 120 and the active region 110. The gate oxide layer 125 may be formed as an interface layer and / or a high-dielectric layer. Figure 3AAs shown, the gate oxide layer 125 can be formed not only on the lower portion of the gate 120, but also on the upper surface of the active region 110 and the upper surface of the device separation structure 150. However, according to an embodiment, the gate oxide layer 125 can be formed only on the active region 110 below the gate 120.

[0043] The gate electrode 122 may include, or be formed of, a conductive material such as polysilicon or a metal. Additionally, the gate electrode 122 may be formed as a single layer or multiple layers. When the gate electrode 122 is formed as a multilayer and includes metal layers, the gate electrode 122 may include, for example, a first metal layer, a work function control layer, and a second metal layer. The first metal layer may be a lower metal layer including Ti, Ta, etc. The work function control layer is a metal layer including a work function material (such as Al), and may supply the work function material to the interface between the high-dielectric layer and the first metal layer. The second metal layer is an upper metal layer located on the work function control layer, and may include an n-type metal constituting the gate of an n-channel MOSFET (NMOS) or a p-type metal constituting the gate of a p-channel MOSFET (PMOS). According to an embodiment, the gate electrode 122 may also include a gap-filling metal layer.

[0044] When in such Figure 2 or Figure 3C When viewed in a planar diagram, gate 120 can have a substantially inverted pie shape. Shape. More specifically, the gate 120 may include a head 120H and two ears 120E. The head 120H may also be described as a body or beam portion. The ears 120E may also be described as legs or extensions extending from the body or beam portion. In one embodiment, the head 120H has a rectangular shape and may cover a portion of the body 110B and a portion of the first portion 110P1 of the protrusion 110P. Additionally, the head 120H may cover the portion of the separation region 130 corresponding to a portion of the body 110B and a portion of the first portion 110P1.

[0045] The two ears 120E may have an elongated rectangular shape extending upward from both ends of the head 120H in a first direction (x direction) along a second direction (y direction). The two ears 120E may cover a portion of the body 110B and extend upward in the second direction (y direction) to the exterior of the body 110B. Figure 3C The dashed line indicated by ACT in the diagram can correspond to the boundary of the active region 110. In one embodiment, the channel avoidance region 110CA can be exposed between the two ear portions 120E. That is, the channel avoidance region 110CA (see...) Figure 2 It is not covered by the gate 120. The space between the two ears 120E can have a rectangular first shape S1.

[0046] In the transistor cell 100 of this embodiment, since the gate 120 is not above the channel avoidance region 110CA, the channel avoidance region 110CA can avoid channel function. When there is a gate portion between the ear portion 120E, the portion of the active region 110 corresponding to the channel avoidance region 110CA can be used as a channel, and the E field can be applied to this portion of the active region 110 through the gate 120. Typically, the E field has the characteristic of being concentrated on the edge portion, so it can be concentrated on the corner portion (①) indicated by the dashed line. Therefore, a defect of gate oxide layer bursting on the corner portion (①) occurs. On the other hand, in the case of the transistor cell 100 of this embodiment, the gate 120 has an inverted pi The shape, and by exposing the channel avoidance region 110CA without covering it, the channel avoidance region 110CA is not used as a channel, and defects such as bursting of the gate oxide layer due to the concentration of the E field can be prevented.

[0047] Already referred to Figure 2 , Figure 3A and Figure 3C The shape of gate 120 is described, but not limited to. (The rest of the text is missing.) Figures 7A to 7F The description of the various types of gate 120 is more detailed.

[0048] Contact 140 can be formed on the active region 110 and the gate 120. More specifically, source contact 140S electrically connected to the common source region 110S can be formed, two drain contacts 140D electrically connected to the two drain regions 110D can be formed respectively, and gate contact 140G electrically connected to the gate 120 can be formed.

[0049] The transistor unit 100 in this embodiment may include two transistors. For example, the gate 120 and the common source region 110S and drain region 110D on the left side may constitute the first transistor (see...). Figure 1B In KT1), and the gate 120 and the common source region 110S and drain region 110D on the right can constitute a second transistor (see KT1). Figure 1B (KT2 in the first transistor KT1). On the other hand, each of the first transistor KT1 and the second transistor KT2, as shown by the two curved arrows, can have a "Γ" shaped channel (②), also described as an "L" shaped channel or a angular channel, so the length of the channel can remain relatively long.

[0050] The transistor cell 100 of this embodiment includes two transistors, each including a common source region 110S and sharing a common gate 120, thereby minimizing the size of the two transistors. Furthermore, because the channel of the active region 110 is separated by the separation region 130, the repair of the two transistors can be separated and effectively managed. Moreover, by maximizing the channel length of each of the two transistors in the form of a "Γ", the high-voltage heat resistance and high-voltage short-channel effect can be improved. Meanwhile, in the transistor cell 100 of this embodiment, because the gate 120 above the active region 110 has an inverted pi... The shape effectively prevents the gate oxide layer from bursting due to the concentration of the E-field at the edge of the active region 110 adjacent to the separation region 130.

[0051] Figures 4A to 5B These are the inverted pi used to explain the transistor unit in this embodiment. Perspective view and SEM image of the advantages of the gate shape, in which, Figure 4A and Figure 5A These are a perspective view and a SEM image of a transistor cell with a rectangular gate, respectively. Figure 4B and Figure 5B These are the inverted pies of this embodiment. Perspective view and SEM image of a transistor cell with a gate of the desired shape.

[0052] Reference Figures 4A to 5B ,like Figure 4A As shown, when the gate 120S has a rectangular shape and is above the active region 110, the E-field is concentrated at the field concentration point A indicated by the dashed circle. Field concentration point A can be, for example, a 3D edge portion (e.g., a corner portion) of the active region 110 adjacent to the separation region 130. Therefore, in Figure 5A In the case of "Gox P", a defect will occur where the gate oxide layer bursts at the field concentration point A.

[0053] On the other hand, such as in Figure 4B In the transistor cell 100 of this embodiment, when the gate 120 has an inverted piezoelectric point When the shape is such that the field E is above the active region 110, the field E may no longer be concentrated at the field concentration point A. Therefore, as Figure 5B As shown, the defect of gate oxide layer bursting at field concentration point A will not occur. Figure 4A and Figure 4B As shown, the separation region 130 can be filled by the separation insulation layer 160.

[0054] Figure 6This is a graph showing the results of measuring time-dependent dielectric breakdown (TDDB) using a drive signal PXIB for the transistor cell of this embodiment and the transistor cell of the comparative example, where SQGP represents the transistor cell of the comparative example including a rectangular gate, and iPGP represents the transistor cell including an inverted PXIB. The transistor cell of this embodiment has a gate of a specific shape. The x-axis represents lifetime in years, and the y-axis represents defects as a percentage.

[0055] Reference Figure 6 In the case of the transistor cell in the comparative example, as can be seen from portion B indicated by the ellipse of the solid line, a considerable number of defects appear early. On the other hand, the solid line corresponds to the fitted line of the graph of the transistor cell in the comparative example, and the dashed line corresponds to the fitted line of the graph of the transistor cell in this embodiment. Based on the fitted lines, it can be predicted that the lifetime of the transistor cell in this embodiment is increased by almost 15 times or more compared to the transistor cell in the comparative example.

[0056] Figures 7A to 7F This is a plan view of a portion of a transistor cell according to an embodiment. (The description may be simplified or omitted.) Figures 1A to 3C The content has already been described in the description. In all Figures 7A to 7F In the previously discussed embodiments, gates 120, 120a, etc., do not cover the 3D corner portion of the active region 110 adjacent to the separation region 130. Additionally, in all these embodiments, the separation region 130 extends beyond the first edge of the gates 120, 120a, etc., in a second direction parallel to the surface of the substrate 101 (e.g., the y-direction), and also extends beyond the second edge of the gate opposite to the first edge in the second direction. Furthermore, in each of these embodiments, a portion of the gates 120, 120a, etc., extends beyond the separation region 130 and beyond the active region 110 in the second direction.

[0057] Reference Figure 7A In this embodiment, the transistor unit 100a can be shaped similarly to the gate 120a. Figure 2 The transistor cell 100 is different. More specifically, in the transistor cell 100a of this embodiment, the gate 120a may have a gate formed in the ear (see...). Figure 3C The tab 120T is located at the end of the tab 120E. For example, the gate 120a may include a tab 120T extending from the end of the tab 120E in a second direction (y direction) along a first direction (x direction). The space between the two tabs 120E may have the following characteristics: Figure 2 The first rectangular shape S1 in the transistor unit 100.

[0058] In the transistor cell 100a of this embodiment, because the gate 120a includes a fin 120T at the end of the ear portion 120E, the width of the ear portion 120E at the end in the first direction (x direction) can be greater than the width of other portions of the ear portion 120E. In this way, because the end of the ear portion 120E of the gate 120a is formed to be wide, it can be ensured that the length of the channel in the portion of the active region 110 adjacent to the upward boundary in the second direction (y direction) is to a certain extent. For example, when the ear portion 120E is formed to be short in the second direction (y direction) or the end of the ear portion 120E outside the active region 110 is formed to be thin, due to the lack of allowance in the manufacturing process, the channel will not be formed in the portion of the active region 110 adjacent to the upward boundary in the second direction (y direction), or a defect will occur because the channel is formed to be very short. On the other hand, when the width of the end of the ear portion 120E outside the active region 110 is formed to be wide from the beginning, sufficient channel length can be ensured in the portion of the active region 110 adjacent to the boundary upward in the second direction (y direction) due to sufficient allowance in the manufacturing process.

[0059] Reference Figure 7B In this embodiment, the transistor unit 100b can be shaped similarly to the gate 120b. Figure 2 The transistor cell 100 is different. More specifically, in the transistor cell 100b of this embodiment, the gate 120b may include a gate extension 120P formed at the end of the ear portion 120E. In addition, the space between the two ear portions 120E may have a first similar shape S1' similar to a rectangle.

[0060] In the transistor cell 100b of this embodiment, the gate extension 120P can have a rectangular shape with a width greater than the width of other portions of the ear portion 120E in the first direction (X direction). Because the gate extension 120P is formed wide, similar to... Figure 7A The gate 120a of the transistor cell 100a includes a fin 120T, so the channel length can be ensured in the portion of the active region 110 adjacent to the boundary upward in the second direction (y direction). Meanwhile, according to the embodiment, the gate extension 120P can be formed in a curved shape at the corner, rather than a rectangular shape.

[0061] Reference Figure 7C In this embodiment, the transistor unit 100c can be shaped similarly to the gate 120c. Figure 2 The transistor unit 100 is different. Figure 2In the transistor cell 100, the space between the two ears 120E of the gate 120 has a first rectangular shape S1. However, in the transistor cell 100c of this embodiment, the space between the two ears 120E of the gate 120c can have a second shape S2, which can be triangular (e.g., inverted triangle). Due to the shape of the space between the two ears 120E, the ears 120E of the gate 120c can have a shape in which the width in the first direction (x direction) decreases as the ears 120E move upward in the second direction (y direction).

[0062] In the transistor cell 100c of this embodiment, the size of the channel avoidance region 110CA can be minimized. For example, the gate 120c can expose only the smallest portion of the 3D edge portion of the location where the gate oxide layer of the active region 110 is most likely to burst. From a plan view, the portion of the active region 110 not covered by the gate 120c can have a triangular shape, for example, a triangular shape. Therefore, due to the shape of the gate 120c, the transistor cell 100c of this embodiment can ensure, to a certain extent, the channel length in the portion of the active region 110 adjacent to the boundary upward in the second direction (y direction).

[0063] Reference Figure 7D In this embodiment, the transistor unit 100d can be shaped similarly to the gate 120d. Figure 2 The transistor unit 100 is different. Figure 2 In the transistor cell 100, the space between the two ears 120E of the gate 120 has a first rectangular shape S1. However, in the transistor cell 100d of this embodiment, the space between the two ears 120E of the gate 120d can have a third shape S3 similar to a semicircle, a semi-ellipse, or a semi-oval. Furthermore, in the transistor cell 100d of this embodiment, the width of the ears 120E of the gate 120d in the first direction (x direction) can decrease as the ears 120E move upward in the second direction (y direction). Additionally, the gate 120d can expose only the smallest portion of the 3D edge portion of the active region 110 where the gate oxide layer is most likely to burst. From a top view, the portion of the active region 110 not covered by the gate 120d can have a triangular shape, for example, having two straight lines connecting at the vertices and a curve joining these two straight lines. Therefore, due to the shape of the gate 120d, the transistor cell 100d of this embodiment can ensure, to a certain extent, the channel length in the portion of the active region 110 adjacent to the boundary upward in the second direction (y direction).

[0064] Reference Figure 7E In this embodiment, the transistor unit 100e can be shaped similarly to the gate 120e. Figure 2 The transistor unit 100 is different. Figure 2 In the transistor cell 100, the space between the two ears 120E of the gate 120 has a rectangular first shape S1. However, in the transistor cell 100e of this embodiment, the space between the two ears 120e of the gate 120e may have a fourth shape S4, which includes a semi-circular shape at the lower portion in the second direction (y direction) and a straight line shape parallel to each other at the upper portion in the second direction (y direction).

[0065] Reference Figure 7F In this embodiment, the transistor unit 100f can be shaped similarly to the gate 120f. Figure 2 The transistor unit 100 is different. Figure 2 In the transistor cell 100, the space between the two ears 120E of the gate 120 has a rectangular first shape S1. However, in the transistor cell 100f of this embodiment, the structure between the two ears 120E of the gate 120f can have a fifth shape S5, which includes a polygonal (e.g., octagonal) partial shape pointing downwards in the second direction (y-direction) and a straight line shape parallel to each other pointing upwards in the second direction (y-direction). In this way, as in Figure 7E and Figure 7F As can be seen, the portions of the two ears 120E furthest from the protrusion 110P of the active region 110 in the second direction (y direction) are straight lines parallel to each other, and the portions of the two ears 120E closest to the protrusion 110P of the active region 110 in the second direction (y direction) have edges forming a part of a circle or a part of a polygonal shape. Furthermore, as can be seen in the various examples above, the gate disposed above the active region can cover the channel region of the active region and can have, for example, an inverted pie shape. The structure of the pie, the inverted pie From the plan view, the structure consists of a portion surrounding but not covering the active region on three sides, including the two corner sections of the active region.

[0066] Figure 8A and Figure 8B This is a plan view of a portion of the transistor cell according to an embodiment. (The description may be simplified or omitted.) Figures 1A to 3C and Figures 7A to 7F The content has already been described in the description.

[0067] Reference Figure 8A Although gate 120 is shown as having a bonding Figure 2 The shape is described, but may optionally have a combination. Figures 7A to 7F The shape of any gate described. Further references. Figure 8AIn this embodiment, the transistor unit 100h can be structurally similar to the active region 110a (specifically, the drain region 110D1). Figure 2 The transistor cell 100 is different. More specifically, in the transistor cell 100h of this embodiment, an extension may not be formed in the main body 110B of the active region 110a (in FIG. 3). Therefore, the main body 110B of the active region 110a may only include an extension extending along the first direction (x direction). Figure 3B The central body 110B1 is part of the body without any extension.

[0068] Reference Figure 8B In this embodiment, the transistor unit 100i can be structurally similar to the active region 110b (specifically, the drain region 110D2). Figure 2 The transistor cell 100 is different. More specifically, in the transistor cell 100i of this embodiment, the main body 110B of the active region 110b may include an extension 110B2 formed symmetrically in the second direction (y direction). For example, the main body 110B of the active region 110b may include a central body 110B1 and extensions 110B2 extending upward and downward in the second direction (y direction) at both ends of the central body 110B1 in the first direction (x direction).

[0069] Figure 9A and Figure 9B These are, respectively, a plan view and a SEM image of a pair of transistor cells arranged according to an embodiment, where two transistor cells share a common source region. The description will be simplified or omitted in... Figures 1A to 3C and Figures 7A to 8B The content has already been described in the description.

[0070] Reference Figure 9A and Figure 9B According to this embodiment, the transistor unit pair 100P may include a first transistor unit 100-1 and a second transistor unit 100-2. Each of the first transistor unit 100-1 and the second transistor unit 100-2 may correspond to Figure 2 The transistor unit 100. Additionally, each of the first transistor unit 100-1 and the second transistor unit 100-2 may have... Figures 7A to 8B Transistor units 100a to 100i (replacing) Figure 2 The structure of any one of the transistor units 100.

[0071] like Figure 9A As shown, the first transistor unit 100-1 and the second transistor unit 100-2 can share the common source region 110S. Therefore, the common source region 110S can be formed by the first transistor unit 100-1 ( Figure 1BThe first transistor unit 100-1 can be shared by two transistors KT1 and KT2, and can also be shared by two transistors of the second transistor unit 100-2. On the other hand, in terms of the sub-word line driver concept, the first transistor unit 100-1 can correspond to a shared structure of the first transistor KT1 of the first sub-word line driver 200-1 and the second transistor KT2 of the second sub-word line driver 200-2, and the second transistor unit 100-2 can correspond to a third sub-word line driver (see...). Figure 11B The third transistor and fourth sub-word line driver (see 200-3) Figure 11B The shared structure of the fourth transistor (200-4) in the middle.

[0072] In the transistor cell pair 100P of this embodiment, the common source region 110S and drain region 110D of the first transistor cell 100-1 and the second transistor cell 100-2 can be formed by ion implantation using an implantation mask IIP' indicated by a dashed line. The common source region 110S and drain region 110D can be formed by separate implantation processes. For example, the drain region 110D of the first transistor cell 100-1 and the second transistor cell 100-2 can be formed by ion implantation using a first type of implantation mask IIP' covering the common source region 110S, and the common source region 110S of the first transistor cell 100-1 and the second transistor cell 100-2 can be formed by ion implantation using a second type of implantation mask IIP' that opens the common source region 110S, opposite to the first type. Simultaneously, when forming the common source region 110S, the channel avoidance region 110CA of the first transistor cell 100-1 and the second transistor cell 100-2 can also be included. Figure 2 (in the middle) doped with impurity ions.

[0073] If passed Figure 9B As can be seen from the arrows in the SEM image, in the transistor cell pair 100P of this embodiment, the first gate 120-1 of the first transistor cell 100-1 and the second gate 120-2 of the second transistor cell 100-2 can expose the channel avoidance regions 110CA-1 and 110CA-2 of the corresponding active regions 110-1 and 110-2 without covering them. Therefore, the problem of gate oxide layer cracking due to fields (such as E-fields) concentrated in the edge portions adjacent to the separation regions 130-1 and 130-2 can be effectively prevented. Figure 2 Similarly, in Figure 9A In this configuration, the first contact 140-1 can be formed on the active region 110-1 and the first gate 120-1, and the second contact 140-2 can be formed on the active region 110-2 and the second gate 120-2.

[0074] Figures 10A to 10C It is used for explanation Figure 9A A comparative example of the advantages of a pair of transistor cells is shown in a plan view of a pair of transistor cells, wherein... Figure 10A A pair of transistor cells are shown in the first comparative example. Figure 10B A pair of transistor cells is shown in the second comparative example. Figure 10C A pair of transistor cells is shown as a third comparative example.

[0075] Reference Figure 10A In the transistor cell pair 100P1 according to the first comparative example, the first gate 120P1-1 of the first transistor cell 100P1-1 and the second gate 120P1-2 of the second transistor cell 100P1-2 may only have a header 120H (in Figure 3C (In the middle) without ears. Thus, when the gates 120P1-1 and 120P1-2 have a shape that only includes the head, in the ion implantation process for the common source region 110S and the drain region 110D, because a small-sized first implantation mask IIP1 is used as indicated by the dashed line, the process margin will be insufficient. Therefore, in the ion implantation process for the common source region 110S and the drain region 110D, there is a problem of increased defect rate or the process precision needs to be increased to minimize the defect rate. At the same time, the first implantation mask IIP1 can also have a first type that covers the corresponding region and a second type that opens the corresponding region.

[0076] Reference Figure 10B In the transistor cell pair 100P2 according to the second comparative example, the shapes of the gates 120P2-1 and 120P2-2 of transistor cells 100P2-1 and 100P2-2 can be substantially the same as the shapes of the gates 120P1-1 and 120P1-2 of transistor cells 100P1-1 and 100P1-2 in the transistor cell pair 100P1 according to the first comparative example. However, the shape of the second implantation mask IIP2 used in the ion implantation process for the common source region 110S and the drain region 110D can be different from the shape of the first implantation mask IIP1. More specifically, the second implantation mask IIP2 is not limited to the common source region 110S and the gates 120P2-1 and 120P2-2, but can extend in the second direction (y direction) to the outside of the active regions 110-1 and 110-2.

[0077] In ion implantation processes used for the common source region 110S and the drain region 110D, when using the second implantation mask IIP2, which is relatively larger in size than the first implantation mask IIP1, the issue of ensuring process margin can be addressed to some extent. However, in ion implantation processes used for the common source region, since impurity ions are also implanted into the channel avoidance region, there is a risk of source / drain non-separation. This leads to degradation of hot carrier characteristics, resulting in a shortened lifetime.

[0078] Reference Figure 10C The transistor cell pair 100P3 according to the third comparative example is shown in one form, wherein the problems of ensuring process margins for the transistor cell pair 100P1 according to the first comparative example and the source / drain non-dissociation problem for the transistor cell pair 100P2 according to the second comparative example are solved, while only the headers of the gates 120P3-1 and 120P3-2 of the transistor cells 100P3-1 and 100P3-2 are maintained. (As shown by...) Figure 10C As can be seen, in order to form a third implantation mask IIP3 approximately the size of the second implantation mask IIP2, while ensuring process margin and preventing doping of impurity ions in the channel avoidance region, the height of the common source region 110S' of the active regions 110P3-1 and 110P3-2, and the height of the gates 120P3-1 and 120P3-2 in the second direction (y-direction) need to be increased. Therefore, with Figure 9A Compared to the transistor cell pair 100P in this embodiment and the transistor cell pairs 100P1 and 100P2 according to the second and third comparison examples, the transistor cell pair 100P3 according to the third comparison example has an increased height in the second direction (y direction). Therefore, the problem of increasing chip size arises.

[0079] Conversely, in the transistor cell pair 100P of this embodiment, the first gate 120-1 of the first transistor cell 100-1 and the second gate 120-2 of the second transistor cell 100-2 have a structure including a head 120H and two ears 120E, which allows for minimization of size while addressing the issues of ensuring margin and source / drain non-separation during ion implantation processes. As a result, the transistor cell pair 100P of this embodiment can help improve the reliability of semiconductor devices including the transistor cell pair 100P and reduce chip size.

[0080] Figure 11A and Figure 11B These are a plan view of a semiconductor device according to an embodiment and a block diagram showing the structure of a portion of the semiconductor device, respectively. They will be referred to together. Figures 1A to 3C describe Figure 11A and Figure 11B And will simply describe or omit in Figures 1A to 10C The content has already been described in the description.

[0081] Reference Figure 11A and Figure 11B The semiconductor device 1000 (hereinafter simply referred to as "semiconductor device") based on the transistor cell of this embodiment may include a cell region CA and a core / peripheral region CO / PA. The term "semiconductor device" may refer to a semiconductor chip formed on a bare die from a wafer, or it may refer to a semiconductor package including one or more semiconductor chips formed on a package substrate. One or more of various memory devices may be arranged in the cell region CA. For example, various memory devices (such as flash memory, dynamic random access memory (DRAM), static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), phase-change random access memory (PRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FeRAM), and resistive random access memory (RRAM)) may be arranged in the cell region CA.

[0082] Peripheral circuitry for operating memory cells and / or core circuitry for operating them can be arranged in a core / peripheral region CO / PA. For example, the entire region excluding cell region CA can be referred to as the core / peripheral region CO / PA. However, according to embodiments, the core / peripheral region CO / PA can be divided into a peripheral circuitry region PA in which peripheral circuitry is arranged and a core circuitry region CO in which core circuitry is arranged. Furthermore, typically, the peripheral circuitry region can be arranged on the outer or central portion of the chip, and the core circuitry region can be located between cell regions CA. Hereinafter, the peripheral circuitry region and the core circuitry region are not distinguished and are collectively referred to as the core / peripheral region CO / PA.

[0083] In the semiconductor device 1000 of this embodiment, the specific name of the semiconductor device 1000 can vary depending on the type of memory device included in the cell region CA. For example, when a flash memory device is included in the cell region CA, the semiconductor device 1000 may correspond to a flash memory semiconductor device. Similarly, when a DRAM device is included in the cell region CA, the semiconductor device 1000 may correspond to a DRAM semiconductor device. In the following description, for ease of explanation, a DRAM semiconductor device will be primarily described. However, the semiconductor device 1000 of this embodiment is not limited to a DRAM semiconductor device.

[0084] The DRAM semiconductor device corresponding to the semiconductor device 1000 in this embodiment may include a row decoder 300, pre-decoders 310 and 320, drive voltage generators 400-1, 400-2, 400-3 and 400-4, sub-word line drivers 200-1, 200-2, 200-3 and 200-4, sense amplifier blocks (SAs) 500, a cell array 600, and a connection circuit 700. The cell array 600 may be pre-arranged in the cell region CA, and the remaining row decoders 300, pre-decoders 310 and 320, drive voltage generators 400-1, 400-2, 400-3 and 400-4, sub-word line drivers 200-1, 200-2, 200-3 and 200-4, sense amplifier blocks 500, and connection circuit 700 may be arranged in the core / peripheral region CO / PA. Because embodiments of the inventive concept involve subword line drivers, a description of the column selection structure will be omitted.

[0085] The line decoder 300 can select the word line of the memory cell to be accessed in response to the input line address RADD. The line decoder 300 can generate a word line enable signal NWEIB to enable the corresponding word line by decoding the input line address RADD. <n>(n is an integer greater than or equal to 0). The word line enable signal NWEIB of the row decoder 300 can be activated for a selected memory cell in either write or read operation mode. <n>Additionally, in self-refresh operation mode, the row decoder 300 can enable the corresponding word line by decoding the row address RADD generated from the address counter (not shown).

[0086] Predecoders 310 and 320 can generate a predecode signal PXI in response to the row address RADD. <0> PXI <1> PXI <2> PXI <3> For example, predecoders 310 and 320 can generate a predecode signal PXI corresponding to the selected word line by decoding the low-order bits of the row address RADD. <j>(j is an integer greater than or equal to 0). Pre-decoding signal PXI <j>The drive voltage generators 400-1, 400-2, 400-3 and 400-4 in the connection circuit 700 can be sent via the main word line.

[0087] Drive voltage generators 400-1, 400-2, 400-3, and 400-4 can respond to the pre-decoding signal PXI. <0> PXI <1> PXI <2> PXI <3> ...and generate the drive signal PXID for driving the word lines. / PXIB (i is an integer greater than or equal to 0). As semiconductor memory devices become highly integrated and high-speed, the level of the high voltage VPP used to drive word lines significantly impacts the reliability of the semiconductor memory device. To improve the reliability of semiconductor memory devices, it is important to reduce the level of the high voltage VPP and prevent the high voltage level from dropping due to leakage current, etc. Therefore, drive voltage generators 400-1, 400-2, 400-3, and 400-4 may include pull-up drivers to supply the high voltage VPP to selected word lines. Typically, pull-up drivers may include PMOS transistors.

[0088] Sub-word line drivers 200-1, 200-2, 200-3, and 200-4 can respond to the word line enable signal NWEIB. <n>and drive signal PXID / PXIB This activates the selected word line or precharges the selected word line. When memory cell MC1 is selected, the word line enable signal NWEIB is activated. <0> And it can activate the drive signal PXID provided from the drive voltage generator 400-1. <2> and PXIB <2> Therefore, the first sub-word line driver 200-1 can drive the word line WL with a high voltage VPP. <1> The high voltage VPP is the drive signal PXID. <2> The level. The operation of the first sub-word line driver 200-1 is also applied to the remaining sub-word line drivers 200-2, 200-3, 200-4, 200'-1, 200'-2, 200'-3 and 200'-4.

[0089] Each of the disclosed sub-word line drivers 200-1, 200-2, 200-3, 200-4, 200'-1, 200'-2, 200'-3, and 200'-4 may include a holding transistor composed of an NMOS transistor. The holding transistor may correspond to the above... Figure 1A and Figure 1B The sub-word line drivers 200-1 and 200-2 contain a first transistor KT1 and a second transistor KT2. By holding these transistors, the sub-word line drivers 200-1, 200-2, 200-3, 200-4, 200'-1, 200'-2, 200'-3, and 200'-4 can precharge the deactivated word lines with a negative voltage VBB2. For example, the first sub-word line driver 200-1 may include a transistor for responding to a drive signal PXIB. <2> And use negative voltage VBB2 to connect the word line WL <1> A holding transistor for pre-charging. Similarly, the adjacent second sub-word line driver 200-2 may also include a PXIB for responding to a drive signal. <2> And use negative voltage VBB2 to connect the word line WL <5> A pre-charged holding transistor.

[0090] In the semiconductor device 1000 according to this embodiment, the two holding transistors KT1 and KT2 of adjacent sub-word line drivers 200-1 and 200-2 can be configured as follows: Figure 1A and Figure 1B The transistor cell 100 described in the description. Therefore, the two holding transistors KT1 and KT2 may share a gate 120 and may include a common-source region 110S to which a negative voltage VBB2 is supplied. More specifically, each of the two holding transistors KT1 and KT2 constituting the transistor cell 100 may include an active region 110 having a "T" shape, the active region 110 including a separate drain region 110D connected to a word line and a common-source region 110S for supplying the negative voltage VBB2. Additionally, the two holding transistors KT1 and KT2 may share a gate 120, and the gate 120 may have an inverted p-shaped gate. Shape and / or the combination of the above Figure 2 , Figures 7A to 7F , Figure 8A and Figure 8B One of the described shapes. Due to the structure of the active region 110 and the gate 120, each of the two holding transistors KT1 and KT2 can have a channel of an "Γ" shape. As a result, each of the two holding transistors KT1 and KT2 can hold a relatively long channel, so that even if a high voltage VPP is applied to the gate 120 of the holding transistors KT1 and KT2, the holding transistors KT1 and KT2 can have heat resistance due to the length of the long channel. At the same time, the shape of the gate 120 and the channel shape of the holding transistors KT1 and KT2 of the sub-word line drivers 200-1 and 200-2 can be respectively equivalently applied to the adjacent sub-word line drivers 200-3 and 200-4 and the sub-word line drivers 200'-1, 200'-2, 200'-3 and 200'-4 included in the sub-word line driver circuit 800 above the cell array 600.

[0091] The sense amplifier block 500 can access memory cells using the bit line pair BL / BLB of the selected column in response to a column address. Additionally, the sense amplifier block 500 may include components for storing input data in the selected memory cells. The sense amplifier block 500 can rewrite the data stored in the memory cells in a self-refresh mode. The sense amplifier block 500 can be connected to memory cells located in an open bit line structure.

[0092] In the cell array 600, multiple memory cells MC can be connected to word lines WL and bit lines BL, respectively, and can be arranged along the row and column directions. Each memory cell MC may include a cell capacitor and an access transistor. The gate of the access transistor can be connected to any of the word lines WL arranged along the row direction. One end of the access transistor can be connected to the bit line BL or complementary bit line BLB arranged along the column direction. The other end of the access transistor can be connected to the cell capacitor.

[0093] In the DRAM semiconductor device 1000 according to the embodiment, the sub-word line driver SWD may include a transistor unit 100 having a common gate structure. Therefore, due to the advantages of the transistor unit 100 described above, the semiconductor device 1000 according to the embodiment can realize a semiconductor device with improved reliability and reduced chip size.

[0094] As described above, and as can be seen from the various embodiments described above, the semiconductor device may include a common gate structure. The semiconductor device may include an active region having a first portion extending along a first direction (e.g., the x-direction) parallel to the top surface of the substrate and a protrusion projecting from the first portion and extending along a second direction (e.g., the y-direction) perpendicular to the first direction and parallel to the top surface of the substrate. A separation region (e.g., 130) may extend from a first end to a second end along the second direction and may divide the active region into a first active portion located on a first side of the separation region and a second active portion located on a second opposite side of the separation region. The separation region may also divide the first portion of the protrusion into a first protrusion located on a first side of the separation region and a second protrusion located on a second side of the separation region. The first end of the separation region may be located on the active region, and the second end of the separation region may be located on the first portion of the protrusion. Two 3D corner portions of the active region may be adjacent to the first end of the separation region, and a gate may be disposed above the active region to cover the channel region of the active region while exposing the two 3D corner portions. In various embodiments, the separation region extends beyond the first edge of the gate in a second direction, and also extends beyond the second edge of the gate opposite to the first edge in a second direction. Furthermore, a portion of the gate extends beyond the separation region and beyond the active region in a second direction, and two 3D corner portions may be located between the first portion and the second portion of the gate in the first direction.

[0095] The first portion of the gate can be a first ear, and the second portion of the gate can be a second ear, and the first and second ears can be connected to each other via a third portion of the gate as a head. The head can cover a portion of the body of the active region and a portion of the protrusion of the active region, and the first and second ears can cover a portion of the body of the active region and extend beyond the edge of the active region. The active region and the gate can be part of a transistor structure, and the space between the two ears can be rectangular, triangular, semi-circular, semi-elliptical, or other shapes. The active region can be formed in a T-shape, and the gate can be formed in an inverted pie shape relative to the T-shape oriented in a face-up manner. Shape. The active region and gate may be part of a transistor structure including a first transistor and a second transistor sharing a common gate, wherein the first transistor is part of a first sub-word line driver and the second transistor is part of a second sub-word line driver.

[0096] Although the inventive concept has been specifically shown and described with reference to embodiments thereof, it will be understood that various changes in form and detail may be made therein without departing from the spirit and scope of the following claims. < / n> < / j> < / j> < / n> < / n>

Claims

1. A transistor unit having a common gate structure, the transistor unit comprising: The active region has a body extending along a first direction parallel to the top surface of the substrate and a protrusion extending along a second direction perpendicular to the first direction and parallel to the top surface of the substrate, the protrusion being connected to a central portion of the body that is centered relative to the first direction and extending away from the body in the second direction. The separation region extends through the body in a second direction to have a first end located in the protrusion and a second end located in the body and opposite to the first end in the second direction. The separation region separates the body into two parts and the first part of the protrusion into two portions, one portion located on each side of the separation region. as well as The gate is disposed above the active region in a third direction perpendicular to the first and second directions to cover the channel region of the active region, and has a structure at the channel avoidance region adjacent to the second end of the separation region of the active region that exposes the top of the active region relative to the gate. The active region is divided into a first active region and a second active region by a separation region. The first active region is connected to the second active region through a protruding second part. In this configuration, the opposite ends of the main body in the first direction correspond to two drain regions, the second part corresponds to a common source region, and the two drain regions, the common source region, and the gate constitute two transistors, wherein the two transistors share a common gate.

2. The transistor unit according to claim 1, wherein, The active region has a T-shape, and The gate has an inverted pie shape, which is inverted relative to the active region oriented to have a T-shape.

3. The transistor unit according to claim 1, wherein, The gate includes a head and two ears. The head is superimposed on the body of the active region and a first portion of the protrusion of the active region. The two ears are superimposed on the body of the active region and separated from each other in a first direction while extending away from the head and away from the protrusion of the active region in a second direction. The top of the channel avoidance region is exposed relative to the gate between the two ears.

4. The transistor unit according to claim 3, wherein, The space between the two ears can be rectangular, triangular, semi-circular or semi-elliptical, or the parts of the two ears furthest from the active area in the second direction can be straight lines parallel to each other and the parts of the two ears closest to the active area in the second direction can have edges forming part of a circle or part of a polygon.

5. The transistor unit according to claim 3, wherein, From the plan view, each of the ears extends outwards from the active area in the second direction.

6. The transistor unit according to claim 3, wherein, Each of the ears has an elongated shape in the second direction, and Each of the ears includes a wing at its end in the second direction, or for each ear, the width of the end in the first direction is greater than the width of the other parts of the ear in the first direction.

7. The transistor unit according to claim 3, wherein, The first end of the separation region protrudes from one side of the head in the second direction and is exposed relative to the gate.

8. The transistor unit according to claim 3, wherein, Each earpiece has a shape that enhances the separation of the source and drain regions during the ion implantation process.

9. The transistor unit according to claim 1, wherein, A corner-shaped channel is formed between the common source region and the two drain regions.

10. A transistor circuit having a common gate structure, the transistor circuit comprising: The active region has a body extending on a substrate in a first direction and a protrusion extending from the central portion of the body in the first direction in a second direction perpendicular to the first direction. as well as The gate is disposed above the active region to be stacked with the channel region of the active region and has an inverted pie structure. The inverted pie structure, as seen in the plan view, surrounds but does not cover the active region on three sides, including the two corner portions of the active region. The active region is divided into a first active region and a second active region by the separation region. The separation region extends along the second direction and separates the main body and part of the protrusion. The protrusion is divided into a first part and a second part. The first part is further divided into two sub-parts by a separation region. The first part is located between the main body and the second part in a second direction. The opposite ends of the main body in the first direction correspond to two drain regions, the protruding second part corresponds to a common source region, and the two drain regions, the common source region and the gate constitute two transistors, wherein the two transistors share a common gate.

11. The transistor circuit according to claim 10, wherein, The inverted pie structure includes a head covering the active region and a first portion of the active region protruding from the head, and two ears covering the active region and separated from each other in a first direction while extending away from the head in a second direction. The channel avoidance region is exposed relative to the gate between the two ears.

12. The transistor circuit according to claim 11, wherein, Each of the ears extends in the second direction to the outside of the active area, and In the plan view, the second end of the separation region, which is opposite to the first end in the second direction, protrudes from the head in the second direction and is exposed relative to the gate.

13. A semiconductor device having a common gate structure, the semiconductor device comprising: An active region has a first active region portion extending along a first direction parallel to the top surface of the substrate and a protrusion protruding from the first active region portion and extending along a second direction perpendicular to the first direction and parallel to the top surface of the substrate. A separation region extends from a first end to a second end along a second direction, and divides a first active region portion into a first active portion located on a first side of the separation region and a second active portion located on a second opposite side of the separation region, and divides a first portion of a protrusion into a first protrusion located on a first side of the separation region and a second protrusion located on a second side of the separation region, wherein the first end of the separation region is located on the active region, and the second end of the separation region is located on the first portion of the protrusion; The two 3D corner portions of the active region are adjacent to the first end of the separation region; and A gate is disposed above the active region to cover the channel region of the active region, while exposing two 3D corner portions, wherein: The separation region extends beyond the first edge of the gate in the second direction, and also extends beyond the second edge of the gate opposite to the first edge in the second direction. A portion of the gate extends beyond the separation region and beyond the active region in the second direction, and The two 3D corner portions are located in the first direction between the first portion of the gate and the second portion of the gate.

14. The semiconductor device according to claim 13, wherein: The first part of the gate is the first ear, the second part of the gate is the second ear, and the first ear and the second ear are connected to each other through a third part of the gate, which serves as the head. The head covers a portion of the first active region and a portion of the protrusion of the active region, and the first ear and the second ear cover portions of the first active region and extend beyond the edge of the active region from the first active region portion.

15. The semiconductor device according to claim 14, wherein: The active region and gate are parts of the transistor structure, in which the space between the two ears can be rectangular, triangular, semi-circular or semi-elliptical.

16. The semiconductor device according to claim 13, wherein: The active region forms a T-shape; and The gate forms an inverted pie shape relative to the T-shape oriented with the front facing up.

17. The semiconductor device according to claim 13, wherein: The active region and the gate are parts of a transistor structure, which includes a first transistor and a second transistor sharing a common gate.

18. The semiconductor device according to claim 17, wherein: The first transistor is part of the first sub-word line driver, and The second transistor is part of the second subword line driver.

19. A sub-word line driver, the sub-word line driver comprising: The first sub-word line driver has a first transistor for supplying a negative voltage to the first word line in response to a drive signal; as well as The second sub-word line driver has a second transistor for supplying a negative voltage to the second word line in response to a drive signal. The first transistor and the second transistor constitute a transistor unit. The transistor cell includes an active region and a gate having an inverted pie shape. The active region has a body extending on a substrate along a first direction and a protrusion extending from the center portion of the body in the first direction along a second direction perpendicular to the first direction. The first transistor and the second transistor share a gate. The active region is divided into a first active region and a second active region by the separation region. The separation region extends along the second direction to separate the main body and the first part of the protrusion.

20. A semiconductor device, the semiconductor device comprising: In the cell region, multiple memory cells are arranged in a two-dimensional array structure; as well as The core / peripheral region surrounds the cell region and includes drive circuitry for driving the plurality of memory cells. The sub-word line driver, which is part of the driving circuit, is located in the core / peripheral region and includes: A first sub-word line driver has a first transistor for supplying a negative voltage to a first word line in response to a drive signal; and a second sub-word line driver has a second transistor for supplying a negative voltage to a second word line in response to a drive signal. The first transistor and the second transistor constitute a transistor unit. The transistor cell includes an active region and a gate having an inverted pie shape. The active region has a body extending on a substrate along a first direction and a protrusion extending from the center portion of the body in the first direction along a second direction perpendicular to the first direction. The first transistor and the second transistor share a gate. The active region is divided into a first active region and a second active region by the separation region. The separation region extends along the second direction to separate the main body and the first part of the protrusion.

Citation Information

Patent Citations

  • Sub-word line drivers and related semiconductor memory devices

    US20190189186A1