Vertical thin film transistor and application as bit line connector for three-dimensional memory array

By using vertical thin film transistors (vertical TFTs) instead of bitline selectors in three-dimensional memory arrays, the problem of large space occupation and unvertical current flow direction is solved, and higher current mobility and carrier mobility are achieved, which improves the Performance of memory arrays.

CN113424319BActive Publication Date: 2024-11-22SUNRISE MEMORY CORP
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202080012920.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2019-12-12
Filing Date
2020-02-10
Publication Date
2024-11-22
Estimated Expiration
2040-02-10

AI Technical Summary

Technical Problem

In the existing three-dimensional memory array, the semiconductor substrate area occupied by the bit line selector is large, resulting in an increase in space occupation, and the current flow direction is not perpendicular to the plane of the memory array, affecting the current mobility.

Method used

Replace the traditional bit line selector with vertical thin film transistors (vertical TFTs). By forming metal silicide layers on the sides of the step structure, the number of conductive layers is reduced, and the crystallinity of semiconductor materials is improved through gas diffusion or ion implantation, thereby reducing crystal defects. Density, realize current flowing in the Z direction.

Benefits of technology

The bit line selector occupies area on the semiconductor substrate effectively, improves the current mobility and carrier mobility, and enhances the operating efficiency of the memory array.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113424319B_ABST
    Figure CN113424319B_ABST
Patent Text Reader

Abstract

A memory circuit comprises: (i) a semiconductor substrate having a planar surface, the semiconductor substrate having a circuit system for memory operation formed therein; (ii) a memory array formed above the planar surface, the memory array having one or more electrodes connected to the memory circuit in the memory array, each conductor extending in a direction substantially parallel to the planar surface; and (iii) one or more transistors, each transistor being formed above, along or below a corresponding one of the electrodes but above the planar surface of the semiconductor substrate, each transistor (a) having first and second drain / source regions and a gate region, each formed of a semiconductor material, wherein the first drain / source region, the second drain / source region or the gate region has a metal silicide layer formed thereon; and (b) selectively connecting the corresponding electrode to the circuit system for memory operation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to low defect density thin film transistors. In particular, the present invention relates to using low defect density thin film transistors in three-dimensional memory arrays. Background Art

[0002] The non-provisional application discloses a three-dimensional array of thin film memory transistors formed above a planar surface of a semiconductor substrate ("3D memory array"). (For ease of reference in the description herein, the direction perpendicular to the planar surface is referred to as the "Z direction", and the two mutually orthogonal directions parallel to the planar surface are referred to as the "X direction" and the "Y direction", respectively.) Multiple 3D memory arrays can be formed on a single semiconductor substrate.

[0003] The 3D memory array includes multiple stacks of NOR memory strings spaced apart from each other at regular intervals along the X direction, wherein each stack of NOR memory strings has multiple layers of NOR memory strings provided on top of each other (i.e., along the Z direction) and isolated from each other. In one embodiment, 8 layers of NOR memory strings are provided in each stack and 8 such stacks are provided along the X direction. Each layer of NOR memory strings in each stack includes first and second strips of n-type or p-type polysilicon layers separated by strips containing channel material. The strips containing channel material include polysilicon layers in contact with the polysilicon layers of the first and second strips. These polysilicon layers in the strips containing channel material can be lightly doped polysilicon, preferably doped opposite to the type of polysilicon layers in the first and second strips. The first strip, the second strip, and the strip containing channel material each extend longitudinally along the Y direction and have a narrow width along the X direction. The polysilicon layers of the first strip and the second strip form a common source region and a common drain region for the layers of the NOR memory strings.

[0004] As disclosed in the non-provisional application, each channel material strip has two polysilicon layers exposed on opposite sidewalls of the stack, each polysilicon layer separated from each other by an insulating material. The polysilicon layers form channel regions of two thin film storage transistors on opposite sides of the stack. Conductive pillars (e.g., heavily doped polysilicon) are provided between adjacent stacks and at regular intervals along the Y direction, each conductive pillar extending along the Z direction and insulated from each of the adjacent stacks by a charge trapping material (e.g., an oxide-nitride-oxide trilayer) covering the vertical sidewalls of the stack. Each conductive pillar serves as a gate electrode. A thin film storage transistor is formed when the conductive pillar ("local word line") and its adjacent charge trapping material and its adjacent common source region and common drain region overlap with one of the channel regions. Thus, two NOR memory strings are formed on opposite sides of each layer of NOR memory strings in each stack.

[0005] The non-provisional application discloses that CMOS support circuitry for memory operations (e.g., various power supply circuits, address decoders, and sense amplifiers) is formed in a semiconductor substrate below a 3D memory array. In one embodiment, the circuitry for memory operations supporting each 3D memory array above the semiconductor substrate is formed below the 3D memory array itself to provide modularity.

[0006] The common drain region ("bit line") and in some embodiments the common source region of the 3D memory array extend beyond both sides of the 3D memory array ("array region") along the Y direction into a "staircase region". In each staircase region, the bit lines (or source lines) of each layer of NOR memory strings of each stack extend to different degrees in the Y direction to form a staircase structure, where the layer of NOR memory strings with the greatest degree is closest to the semiconductor substrate and the layer of NOR memory strings with the least degree is farthest from the semiconductor substrate. As described in various embodiments in the non-provisional application, the staircase structure facilitates electrical connection between each bit line and its corresponding circuitry (e.g., sense amplifier) ​​for memory operation. Summary of the invention

[0007] According to one embodiment of the present invention, a memory circuit includes: (i) a semiconductor substrate having a planar surface, the semiconductor substrate having a circuit system for memory operation formed therein; (ii) a memory cell array ("memory array") formed above the planar surface, the memory array having one or more conductors in a conductive layer to carry electrical signals to the memory cells in the memory array, the conductors each extending in a direction substantially parallel to the planar surface; and (iii) one or more transistors, each transistor being formed above, along or below a corresponding one of the conductors in the conductive layer, but above the planar surface of the semiconductor substrate, each transistor (a) having a first drain or source region and a second drain or source region and a gate region, each formed of a semiconductor material, the first drain or source region, the second drain or source region or the gate region having a metal silicide layer formed thereon; and (b) selectively connecting the corresponding conductor in the conductive layer to the circuit system for memory operation.

[0008] In one embodiment, the metal silicide layer includes titanium silicide or nickel silicide.

[0009] In one embodiment, the memory circuit further comprises a second conductive layer, wherein the first drain or source region of each of the transistors is connected to one of the conductors in the second conductive layer, and the second drain or source region of the transistor is connected to the corresponding conductor in the first conductive layer. The memory circuit may further comprise a set of interconnect lines, wherein the gate terminal of each transistor is connected to one of the interconnect lines.

[0010] In one embodiment, when current is present in a channel region in a transistor, the current flows in a direction substantially perpendicular to the planar surface.

[0011] In one embodiment, a memory array includes multiple levels of NOR memory strings, each level having multiple NOR memory strings, and wherein the conductors in the first conductive layer are common bit lines of the NOR memory strings. Portions of the common bit lines may be provided in a staircase structure extending from one or both sides of the memory array. In addition, each step of the staircase structure includes a common bit line for a corresponding level of the NOR memory strings, and transistors connected to the common bit lines at each step of the staircase form a bit line selector for those common bit lines. The transistors of each bit line selector form a first group and a second group, so that adjacent common bit lines selected by the bit line selector are served by transistors from the first group and transistors from the second group, and wherein the contacts of the transistors in the first group are placed in a staggered manner relative to the contacts of the transistors in the second group.

[0012] According to one embodiment of the present invention, a first process for forming a transistor comprises: (i) providing a semiconductor substrate having a planar surface; (ii) providing a first semiconductor layer of a first type and a first doping concentration above the planar surface of the semiconductor substrate; (iii) providing a molded dielectric layer above the first semiconductor layer; (iv) etching the molded dielectric layer to provide a groove substantially perpendicular to the planar surface, wherein the groove exposes the surface of the first semiconductor layer; (v) filling the groove with a second semiconductor layer in an amorphous form, wherein the second semiconductor layer has a second type opposite to the first type or has a second doping concentration lower than the first doping concentration; (vi) converting a portion of the second semiconductor layer to become a third semiconductor layer, wherein the third semiconductor layer has the first type or has a third doping concentration higher than the second semiconductor layer; and (vii) providing a metal layer on the third semiconductor layer; and annealing the metal layer to convert the metal layer into a metal silicide.

[0013] The first process may further include: (i) providing a second trench in the molded dielectric layer, the second trench being substantially perpendicular to the planar surface and exposing a portion of the second semiconductor layer; (ii) providing an insulator over the exposed portion of the second semiconductor layer; and (iii) providing a conductive material to fill the second trench, the conductive material being insulated from the second semiconductor layer by the insulator. The conductive material provides a gate electrode to the transistor.

[0014] According to one embodiment of the present invention, a second process for forming a transistor comprises: (i) providing a semiconductor substrate having a planar surface; (ii) providing a first semiconductor layer of a first type and a first doping concentration above the planar surface of the semiconductor substrate; (iii) providing a molded dielectric layer above the first semiconductor layer; (iv) etching the molded dielectric layer to provide a surface of the first semiconductor layer with a groove substantially perpendicular to the planar surface; (v) providing a conductive material coating the exposed surface of the semiconductor layer and the sidewalls of the groove; (vi) providing an insulator coating the conductive material; (vii) anisotropically etching away any conductive material or any portion of the insulator material in the groove to expose the surface of the first semiconductor material; (viii) filling the trench with a second semiconductor layer in an amorphous form, the second semiconductor layer having a second type opposite to the first type or having a second doping concentration lower than the first doping concentration; (ix) converting a portion of the second semiconductor layer to become a third semiconductor layer, the third semiconductor layer having the first type and having a third doping concentration higher than the second semiconductor layer; (x) providing a metal layer on the third semiconductor layer; and (xi) annealing the metal layer to convert the metal layer into a metal silicide.

[0015] In the first process and the second process, the step of converting a portion of the second semiconductor layer into the third semiconductor layer may include gas diffusion or ion implantation.

[0016] According to one embodiment of the present invention, a third process for forming a transistor includes: (a) providing a semiconductor substrate having a planar surface; (ii) providing a first semiconductor layer above the planar surface; (iii) providing a first insulator layer above the first semiconductor layer; (iv) providing a third semiconductor layer above the insulator so that the third semiconductor layer is separated from the first semiconductor layer; (v) providing a second semiconductor layer in contact with both the first semiconductor layer and the second semiconductor layer, wherein a portion of the second semiconductor layer is provided above the third semiconductor layer; (vi) providing a second insulator layer above the second semiconductor layer; (vii) providing a conductive material above the second insulator layer; (viii) providing a passivation layer to surround the conductive material, the second insulator layer, the second semiconductor layer, the third semiconductor layer, the first insulator layer and the first semiconductor layer; (ix) etching the passivation layer to provide a through hole to expose a portion of the conductive material; (x) providing a metal layer on the exposed portion of the conductive material or the exposed portion of the third semiconductor layer; and (xi) annealing the metal layer to convert the metal layer into a metal silicide.

[0017] The third process may further include etching the second semiconductor layer, the second insulator layer, and the conductive layer to form a line structure around the third semiconductor layer, the first insulator layer, and the first semiconductor layer.

[0018] According to one embodiment of the present invention, a fourth process for forming a transistor comprises (a) providing a first conductive semiconductor layer of a first conductivity above a planar surface of a semiconductor substrate; (b) providing a dielectric material above the first semiconductor layer; (c) creating a first and second portion of a cavity in the dielectric material that reaches down to the first semiconductor layer, the second portion of the cavity having a width that is significantly smaller than a corresponding width of the first portion of the cavity; (d) conformally providing an etch stop layer on a sidewall of the cavity; (e) conformally providing a sacrificial layer on the sidewall of the first cavity above the etch stop layer, providing the sacrificial layer with a thickness such that the second portion of the cavity is substantially filled. ; (f) anisotropically etching the sacrificial layer and the etch stop layer to expose a portion of the first semiconductor layer at the bottom of the first portion of the cavity; (g) providing an amorphous semiconductor material to fill the first portion of the cavity; (h) providing a metal layer on the amorphous semiconductor material; (i) annealing the metal layer into an alloy of the semiconductor material; (j) converting a portion of the crystallized semiconductor layer into a region of semiconductor material of the first conductivity type by ion implantation; (k) removing the sacrificial layer from the first and second portions of the cavity; (l) conformally providing a gate dielectric material on the sidewalls of the first and second portions of the cavity; and (m) filling the first and second portions of the cavity with a gate conductor layer.

[0019] In each of the first, second, third and fourth processes, the metal silicide may include one or more of the following: titanium silicide and nickel silicide. When the metal layer includes titanium, annealing may be performed at over 550°C for 1 second to 24 hours. Specifically, in the case of titanium, annealing (a) is performed at 550°C to 600°C for 12 to 24 hours; (b) is performed at 600°C to 750°C for 5 minutes to 12 hours; (c) is performed at 750°C to 800°C for 1 minute to 5 minutes, or (d) is performed at 800°C to 1000°C for 1 second to 1 minute. When the metal layer includes nickel, annealing is performed between about 350°C and about 450°C.

[0020] In each of the first, second, third and fourth processes, the amorphous semiconductor material is converted into a crystalline semiconductor material by annealing. The metal silicide adjacent to the amorphous semiconductor layer assists the crystallization process.

[0021] The invention may be better understood when the detailed description is considered in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1a An 8-level ladder structure 100-L is illustrated for 8 bit lines from 8 levels of NOR memory strings on a single side of a 3D memory array.

[0023] Figure 1b Side views of the XZ plane passing through the step structures 100 -L and 100 -R, respectively, are shown.

[0024] Figure 1c An 8-bit bit line selector 150 is shown for one selected level of 8 stacks across a 3D memory array (e.g., for Figure 1a A top view of a bit line selector for bit line 106-0 of FIG.

[0025] Figure 1d is a schematic circuit diagram of an 8-bit bit line selector 150 .

[0026] Figure 1e Shown in the semiconductor substrate below the staircase structure, the sense amplifier is connected to four 8-bit bit line selectors, where each 8-bit bit line connector is connected to a common bit line in a corresponding one of the four levels of NOR memory strings in the 3D memory array.

[0027] Figure 2a Schemes 201 , 202 and 203 of providing vertical thin film transistors (TFTs) above, below and along the stair structure 200 , respectively, according to one embodiment of the present invention are illustrated.

[0028] Figure 2bA cross section of a vertical TFT 330 according to one embodiment of the present invention is shown.

[0029] Figure 2c A cross section of a vertical TFT 350 is shown according to one embodiment of the present invention.

[0030] Figure 3 A bit line selector 300 formed using vertical TFTs provided above a stepped structure 320 according to an embodiment of the present invention is shown.

[0031] Figure 4 1 shows a schematic diagram of a method for transmitting an odd bit line according to an embodiment of the present invention. Figure 3 YZ plane cross section of the step 320.

[0032] Figure 5 It is shown that in the stepped structure 320, the bit line selection signal is connected to the gate electrode 233 of each of the vertical TFTs TR0, TR4, TR1 and TR5 using the global word lines 103-0, 103-4, 103-1 and 103-5 respectively, and the drain terminal 231a of each of the vertical TFTs TR0, TR4, TR1 and TR5 is connected using the conductive layer 212.

[0033] Figure 6 is an isometric view of several vertical TFTs (eg, vertical TFT 5) and stepped structure 40 before interconnections are made through interconnect conductors over the vertical TFTs and stepped structure 40 in accordance with one embodiment of the present invention.

[0034] Figure 7a , 7b-1 , 7b-2, 7c, 7d-1, 7d-2, 7e-1, 7e-2, 7f-1, 7f-2, 7g-1, 7g-2, 7h-1, 7h-2, 7i, 7j-1, 7j-2, 7k-1, 7k-2 and 7l illustrate a manufacturing process for forming a vertical TFT according to an embodiment of the present invention.

[0035] Figure 8 A vertical TFT 850 is shown according to one embodiment of the present invention.

[0036] Figure 9a , 9b , 9c, 9d, 9e, 9f, 9g and 9h illustrate another process according to which the vertical TFT 850 may be formed according to one embodiment of the present invention.

[0037] Fig.10 A vertical TFT 280 is shown according to one embodiment of the present invention.

[0038] Fig.11a ,11b , 11c, 11d, 11e, 11f illustrate the method of forming a Fig.10 The process of vertical TFT 280.

[0039] Fig.12a , 12b , 12c, 12d, 12e, 12f, 12g, 12h, 12i, 12j, and 12k illustrate a vertical TFT 20 with high mobility connected to a memory cell in an array of NOR memory strings according to yet another embodiment of the embodiments of the present invention.

[0040] Fig.13 Show Figures 12a to 12k A vertical TFT with some dielectric layers removed for clarity. DETAILED DESCRIPTION

[0041] Figure 1a An 8-level staircase structure 100-L is illustrated for eight bit lines from eight levels of NOR memory strings on a single side of a 3D memory array. Figure 1a As shown in FIG. 1 , the step structure 100-L includes bit lines 106-0 to 106-7 extending in decreasing degrees along the Y direction. The bit lines 106-0, 106-2, 106-4, and 106-6 are electrically connected to the conductors in the conductive layers 101-1, 101-2, 101-3, and 101-4 through the conductor-filled vias 105-0, 105-2, 105-4, and 105-6. Figure 1a In the embodiment shown in FIG. 1 , bit lines 106-1, 106-3, 106-5, and 106-7 are electrically connected to conductors in the same conductive layers 101-1, 101-2, 101-3, and 101-4 through conductor-filled vias 105-1, 105-3, 105-5, and 105-7. The conductors of conductive layers 101-1, 101-2, 101-3, and 101-4 are electrically connected to a first set of orthogonally interconnected conductors ("first global word lines"), respectively.

[0042] Each first global word line in the ladder structure 100-L (and 100-R) routes electrical signals between supporting circuitry in the semiconductor substrate to selected bit lines in the same hierarchical layer of NOR memory strings in the 3D memory array. In this description, "same hierarchical" refers to layers of NOR memory strings that are substantially at the same hierarchical level above a planar semiconductor substrate. Figure 1a, first global word line 103-6 is connected to a conductor in conductive layer 101-4 through via 102-6, which is electrically connected to bit line 106-6. First global word line 103-6 is connected to buried contact 108-6 by conductor 107-6 and by conductive pillar or local word line 104-6 through via. Conductor 107-6 is a conductor in a second set of interconnect conductors ("second global word lines") extending below the 3D memory array substantially parallel to the first global word line. Buried contact 108-6 is connected to the buried contact 108-6 by a bit line selector circuit (in Figure 1a Not shown, but will be discussed below in conjunction with Figure 1c A second global word line 107-6 allows other circuitry (eg, a bias voltage source) to be connected to the bit line 106-6.

[0043] exist Figure 1a 106-2. In the embodiment of the present invention, similar to the first global word line 103-6, the first global word line 103-4 is connected to a conductor in the conductive layer 101-3 through a via 102-4, and the conductor is electrically connected to the bit line 106-2. The first global word line 103-4 is connected to the buried contact 108-4 by the local word line 104-4 through the second global word line 107-4 and through the conductor-filled via. The buried contact 108-4 is connected to the sense amplifier serving the bit line 106-4 through the bit line selector circuit. The second global word line 107-6 allows other circuit systems (e.g., a bias voltage source) to be connected to the bit line 106-6.

[0044] Figure 1b Side views in the XZ plane of the stepped structures 100 -L and 100 -R, respectively, are shown. Figure 1b The bit lines 106-0, 106-2, 106-4 and 106-6 are shown connected to the buried contacts 108-0, 108-4, 108-6 and 108-8 in the staircase structure 100-L in the manner already discussed above with respect to FIG. 1a. Figure 1b Also shown are the connections of the bit lines 106-1, 106-3, 106-5, and 106-7 to the buried contacts 108-1, 108-3, 108-5, and 108-7 in the staircase structure 100-R in a similar manner. Figure 1b It is also schematically indicated that each of the buried contacts 108-0, ..., 108-6, and 108-7 is connected through an 8-input selection circuit to select one of 8 same-level bit lines across 8 stacks in the 3D memory array.

[0045] Therefore, in Figure 1a and 1bIn the connection method described in , for an 8-level 3D memory array, four additional conductive layers (i.e., conductive layers 101-1, 101-2, 101-3, and 101-4) are required. Figure 1a and 1b The connection method requires as many additional conductive layers as half the number of levels in the 3D memory array. In addition, as the number of levels in the 3D memory array increases, the silicon footprint under the 3D memory array for the bit line selectors increases proportionally. Figure 1c An 8-bit bit line selector 150 is shown for one selected level of 8 stacks across a 3D memory array (e.g., for Figure 1a A top view of a bit line selector for bit line 106-0 of FIG.

[0046] like Figure 1c As shown in FIG. 1 , bit line BL0 <0> BL0 <1> , ... and BL0 <7> Each is connected to a corresponding drain terminal of eight CMOS transistors forming a bit line selector 150 in the semiconductor substrate. <0> BL1 <1> , ... and BL1 <7> BL2 <0> BL2 <1> , ... and BL2 <7> and BL3 <0> BL3 <1> , ... and BL3 <7> Each is connected to one of the other three 8-bit bit selectors. The source terminals of all 32 CMOS transistors in the four 8-bit bit line selectors are commonly connected to the terminal 151 of the sense amplifier serving all 32 bit lines (i.e., the 32 bit lines correspond to eight bit lines in each of the four selected levels of the ladder structure). In the bit line selector 150, the selection signal SEL is <0> , SEL <1> , ... and SEL <7> The corresponding gate electrodes to the eight CMOS transistors are provided such that when one of the selection signals is activated, the corresponding CMOS transistor becomes conductive, thereby connecting the corresponding bit line to the terminal 151 of the sense amplifier.

[0047] Figure 1d is a schematic circuit diagram of an 8-bit bit line selector 150. Figure 1d In FIG. 1 , the eight CMOS transistors of the bit line selector 150 are grouped into four transistor pairs, wherein the two CMOS transistors in each transistor pair share a common source region therebetween.

[0048] Figure 1e Shown in the semiconductor substrate below the staircase structure, the sense amplifier is connected to four 8-bit bit line selectors, where each 8-bit bit line connector is connected to a bit line in a corresponding one of the four levels of NOR memory strings in the 3D memory array.

[0049] exist Figures 1a to 1eIn one embodiment of the method, for an 8-level 3D memory array, four bit line selectors (eg, bit line selector 150) below each staircase structure require about 35 μm in the semiconductor substrate area. 2 , which is essentially 28% of the semiconductor substrate area beneath the footprint of each 3D memory array. As the number of levels in a 3D memory array increases, the required semiconductor area for the bit line selectors increases proportionally. The semiconductor area required for the bit line selectors of a 3D memory array with more than 16 levels will exceed the footprint of the 3D memory array itself.

[0050] According to one embodiment of the present invention, the bit line selector in the semiconductor substrate for connecting the bit line to the sense amplifier can be removed by implementing the bit line selector using thin film transistors ("vertical TFTs") formed above, below, or on the sides of the staircase structure. The vertical TFT allows the additional conductive layer (e.g., Figure 1a and 1b The conductive layers 101-1, 101-2, 101-3 and 101-4) of the vertical TFT are reduced to one. The vertical TFT is so named because, in some preferred embodiments, its channel current flows along the Z direction. It is important to understand that the "vertical" designation in the vertical TFT is merely a convenient naming convention in this specific embodiment. As will be apparent from the following description, the channel current in the vertical TFT (although preferably used to minimize its occupied area) does not necessarily flow along the Z direction to perform its intended function.

[0051] Figure 2a The schemes 201, 202 and 203 of providing vertical TFTs above the step structure 200, below the step structure 200 and along the step structure 200 are described. Figure 2a As illustrated in FIG. 2 , according to scheme 201 , bit lines at the same level of NOR memory strings of a 3D memory array are connected through conductor-filled vias (e.g., Figure 1a The conductor-filled vias 105-0, 105-2, 105-4, and 105-6 of the stepped structure 200 are connected to corresponding vertical TFTs (e.g., vertical TFTs 211-1, 211-3, 211-5, and 211-7 that connect bit lines in levels 1, 3, 5, and 7, respectively) formed above the stepped structure 200. When the vertical TFTs are turned on, they connect the corresponding bit lines to corresponding conductors in the single conductive layer 212. The conductors in the conductive layer 212 can be connected to supporting circuitry (e.g., sense amplifier 214) in the semiconductor substrate through other conductors (e.g., conductive pillars or local word lines) in the stepped structure 200.

[0052] According to scheme 202, bit lines at the same level of NOR memory strings of a 3D memory array are connected through conductor-filled vias (e.g., Figure 1a The through holes 105-0, 105-2, 105-4, and 105-6 of the step structure 200 are connected by an available conductor to a corresponding one of the vertical TFTs (e.g., vertical TFTs 217-1, 217-3, 217-5, and 217-7 that connect the bit lines in levels 1, 3, 5, and 7, respectively) formed below the step structure 200. For each bit line, such an available conductor may include a first global word line (e.g., Figure 1a The vertical TFTs are connected to the corresponding bit lines of the semiconductor substrate (e.g., global word lines 103-4) and corresponding short segments of corresponding local word lines (e.g., local word lines 104-4). When the vertical TFTs are turned on, they connect the corresponding bit lines to corresponding conductors in the single conductive layer 215. The conductors in the conductive layer 215 can be selectively connected to supporting circuitry (e.g., sense amplifier 214) in the semiconductor substrate.

[0053] According to scheme 203, bit lines at the same level of NOR memory strings of a 3D memory array are connected through conductor-filled vias (e.g., Figure 1a The conductor-filled vias 105-0, 105-2, 105-4, and 105-6) are connected by other conductors to corresponding ones of the vertical TFTs (e.g., vertical TFTs 218-1, 218-3, 218-5, and 218-7 connecting the bit lines in levels 1, 3, 5, and 7) formed along the stepped structure 200. For each bit line, such other conductors may include a first global word line (e.g., Figure 1a When the vertical TFT is turned on, it connects the corresponding bit line to the corresponding conductor in the single conductive layer 216, which is connected to the supporting circuitry (e.g., sense amplifier 214) in the semiconductor substrate.

[0054] Figure 2b FIG. 2 shows a cross section of a vertical TFT 330 according to an embodiment of the present invention. Figure 2b As shown in FIG. 1 , the vertical TFT 330 includes an annular gate electrode 233 (e.g., a polysilicon gate) surrounding a semiconductor structure separated from the annular gate electrode by an annular gate oxide 235. (As used herein, the term "annular" refers not only to a circular ring, but also to any closed outer layer of any shape.) The semiconductor structure includes N + The drain region 231a and N - Doped drain region 231b, channel region 236, N - The source region 234 b and N +The channel region 236 may be, for example, a P-type body region. When the voltage on the gate electrode 233 exceeds the threshold voltage of the vertical TFT 330, the vertical TFT 330 becomes conductive, thereby providing an N + The drain region 231a is doped with N + A low resistance current path is formed between the doped source regions 234a.

[0055] Figure 2c FIG. 4 shows a cross section of a vertical TFT 350 according to an embodiment of the present invention. Figure 2c As shown in FIG. 3 , the vertical TFT 350 includes an annular gate electrode 233 (eg, a polysilicon gate) surrounding a semiconductor structure separated from the annular gate electrode by an annular gate oxide 235. Similar to FIG. Figure 2b The semiconductor structure of the vertical TFT 330 and the vertical TFT 350 includes N + The drain region 231a and N - Doped drain region 231b, N - The source region 234 b and N + doped region 234a. However, Figure 2b Unlike the TFT 330 of FIG. 3 , the channel or P-type body region 236 of the vertical TFT 350 is annular, thereby surrounding the oxide core 237. When the voltage on the gate electrode 233 exceeds the threshold voltage of the vertical TFT 350, the vertical TFT 350 becomes conductive, thereby providing N + The drain region 231a is doped with N + A low resistance current path is formed between the doped source regions 234a.

[0056] Figure 2b and 2c Not drawn to scale. In one embodiment, the thickness of gate oxide 235 can be 20 nm, the thickness of gate electrode 233 can be 40 nm, and the channel length of p-type body region 236 (i.e., along the Z direction) can be 500 nm. The occupied area of ​​p-type body 236 can be 150 nm by 100 nm (in the X and Y directions, respectively), so that the channel width, as approximated using the perimeter of the occupied area, is also 500 nm. With these dimensions, when the gate to source (V GS ) and drain to source at 0.5 volts (V DS ) when driven, the vertical TFT can provide approximately 2.25μA of drain-to-source (I DS ) current capacity. Significantly greater current capacity (eg, 15 μA) can be achieved at these dimensions when high quality crystalline semiconductor is provided to the channel region, for example when crystallization is promoted by metal silicide as discussed below.

[0057] Figure 3 2 shows a bit line selector 300 formed using a vertical TFT provided above a stepped structure 320 (ie, scheme 201) according to one embodiment of the present invention. Figure 3 As shown in , the bit line selector 300 is shown as providing vertical TFTs for connecting to bit lines at levels 1, 3, 5, and 7 of the step structure 320. In order to use space efficiently, as described below, two groups of vertical TFTs are provided for each level of the bit line, wherein one group of vertical TFTs is provided for connecting to odd bit lines, and the other group of vertical TFTs is provided for connecting to even bit lines. For example, the vertical TFT TR0 is connected to the odd bit lines 302-1, 302-3, 302-5, and 302-7 of level 7, and the vertical TFT TR4 is connected to the even bit lines 302-0, 302-2, 302-4, and 302-6 of level 7. Similarly, the vertical TFTs TR1, TR2, and TR3 are connected to the odd bit lines of levels 5, 3, and 1, respectively, and the vertical TFTs TR5, TR6, and TR7 are connected to the even bit lines of levels 5, 3, and 1, respectively. This arrangement allows the staggered placement of the vertical TFTs serving adjacent bit lines.

[0058] In one embodiment, each bit line has a width of 50nm and adjacent bit lines are spaced 80nm apart, thereby providing a 130nm bit line pitch. The sense amplifiers serving adjacent bit lines (e.g., sense amplifier 301-0 serves adjacent bit lines 302-0 and 302-1 at each of levels 1, 3, 5, and 7) can be set between twice the bit line pitch (i.e., 260nm). In the embodiment, the vertical TFTs that provide the required drive can have a footprint of 150nm×100nm (and a vertical channel length of 500nm). Therefore, in the bit line selector 300, two sets of vertical TFTs serving each level are placed side by side along the Y direction, but the contacts of the vertical TFTs are placed in a staggered manner relative to each other to allow the vertical TFTs serving adjacent bit lines to fit within 260nm (i.e., twice the bit line pitch). Specifically, in one embodiment, each vertical TFT has a footprint that exceeds the bit line pitch in both the X direction and the Y direction. By placing adjacent vertical TFTs TRO and TR4 in a staggered manner, the combined occupied area of ​​two adjacent vertical TFTs fits within a 320nm (Y direction)×260nm (X direction) area. For an 8-level 3D memory array with an array length of 160μm (i.e., the length of each NOR memory string along the Y direction), the staircase structure 320 only needs to extend 2μm along the Y direction to accommodate the bit line selector 300 of the present invention.

[0059] Figure 4 1 shows a schematic diagram of a method for transmitting an odd bit line according to an embodiment of the present invention. Figure 3 YZ plane cross section of the step 320. For illustration purposes only, Figure 4 Shown by Figure 2b The vertical TFT 330 illustrates a vertical TFT of the type implemented by the vertical TFTs TR0 and TR1. Figure 4 It is also shown that by Figure 2c The vertical TFT 350 illustrates a type of vertical TFT implementing the vertical TFTs TR2 and TR3. Any vertical TFT disclosed herein is suitable for implementing Figure 4 Any one of the vertical TFTs TR0, TR1, TR2 and TR3 shown in FIG. Figure 4 Each vertical TFT in the + A thin layer of titanium nitride (TiN) 332 is formed on the doped source region 234a to provide good electrical contact. In addition, a C49 titanium silicide (TiSi2) layer 331 is formed adjacent to the N + On each vertical TFT of the doped drain region 231a. As explained below, for example, the annealing step of forming the TiSi2 layer also causes crystallization of any adjacent amorphous semiconductor material because the TiSi2 layer provides a seed layer that causes the adjacent amorphous semiconductor material to become crystalline. Compared to the semiconductor layer crystallized in the absence of the seed layer, the crystalline semiconductor material thus formed has fewer defects and higher carrier mobility. Alternatively, the semiconductor layer can be crystallized using a metal silicide (e.g., nickel silicide). In this alternative method, the metal silicide leaves a crystalline semiconductor with fewer defects and higher mobility before and after the wave of the metal silicide layer.

[0060] Figure 5 It is shown that in the ladder structure 320, the bit line selection signals are connected to the gate terminals of the vertical TFTs TR0, TR4, TR1 and TR5 using the global word lines 103-0, 103-4, 103-1 and 103-5, respectively, and the drain terminals of the vertical TFTs TR0, TR4, TR1 and TR5 are connected using the conductive layer 212. Figure 5 In FIG. 1 , cross sections in both the XY and YZ planes are shown. Figure 5 As shown in FIG, selection signals of the vertical TFTs TR0 and TR4 are connected to the bit line at level 7 of the ladder 320 through the first global word lines 103-0 and 103-4, and selection signals of the vertical TFTs TR1 and TR5 are connected to the bit line at level 5 of the ladder 320 through the first global word lines 103-1 and 103-5. Figure 5It is also shown that a first global word line is typically used in the array portion of a 3D memory array to connect to a local word line provided as a conductive pillar. The drain terminals (e.g., drain terminal 231a) of the vertical TFTs TR0, TR4, TR1, and TR5 (as well as the drain terminals of vertical TR2, TR6, TR3, and TR7 in other levels) are served by the same sense amplifier and connected to a common conductor in the conductive layer 212. Thus, with Figures 1a to 1e Unlike the method described in , which uses vertical TFTs and requires half the number of additional conductive layers in the 3D memory array, the method of FIGS. 2 to 5 requires only a single additional conductive layer.

[0061] A significant advantage of the vertical TFT described herein is that the occupied area projected on the surface of the underlying semiconductor substrate is reduced because the current flows in the Z direction substantially perpendicular to the semiconductor substrate. As used herein, the term "occupied area" refers to the area of ​​the vertical TFT projected on the semiconductor substrate when viewed from top to bottom. In general, the current provided by a vertical TFT in the "on" or conductive state is proportional to the mobility of the charge carriers in its channel region. Since integrating the vertical TFT with the memory cell operation requires a higher "on" current, it is necessary for the charge carriers in the vertical TFT to have a higher mobility.

[0062] The following description illustrates various embodiments of vertical TFTs within the scope of the present invention (whether or not discussed above) and methods of forming such vertical TFTs. The process steps described herein in one embodiment may be used in another embodiment, even if those steps are not explicitly described in the context of the other embodiment. When reference is made herein to a method of two or more specified steps, such specified steps may be implemented in any order or simultaneously unless otherwise stated herein or required by the context. Similarly, such a method may also include one or more other steps performed before or after the specified steps, or between any two specified steps.

[0063] Figure 6 1 is an isometric view of several vertical TFTs (e.g., vertical TFT 5) and a staircase structure 40 before interconnections are made through interconnect conductors above the vertical TFTs and memory structure 40 (which includes a staircase structure) according to one embodiment of the present invention. The staircase structure in the memory structure 40 includes a plurality of conductor-filled vias (e.g., conductor-filled via 17), each of which is connected to a bit line (i.e., a common drain region) at each level of the 3D memory array. Figure 6 In FIG. 4 , a vertical TFT is shown along the stepped structure 40 (eg, Figure 2a 203 of the present invention). Figure 6As shown in , the vertical TFT 5 is electrically connected at the bottom by a conductive element 15, which can be formed of the same material as the common drain region of the NOR memory string (i.e., a doped semiconductor layer, which can also be tied to a metal layer to reduce resistance), thereby allowing the conductive element 15 to be formed simultaneously with the bit line during the same masking and etching steps. Figure 6 As shown in , the vertical TFT 5 includes a semiconductor pillar 90 concentrically surrounded by a gate oxide layer 100 and a gate electrode layer 110. The semiconductor pillar 90 includes a channel region (not shown) and a source region (not shown) for the vertical TFT 5. The conductive element 15 forms a drain region of the vertical TFT 5. The semiconductor pillar 90, which is substantially perpendicular to the plane of the semiconductor substrate, provides current in a vertical direction, which allows the area on the semiconductor substrate to be significantly reduced. The following description includes process steps that allow the semiconductor pillar 90 to have a low crystal defect density and a high charge carrier mobility in the channel region of the resulting vertical TFT, thereby providing improved current capacity.

[0064] Figures 7a to 7l The manufacturing process for forming a vertical TFT according to one embodiment of the present invention is described. Figure 7a As shown in , the conductive interconnect 10 is initially formed above and substantially parallel to the semiconductor substrate. Suitable materials for the interconnect 10 include any of molybdenum, cobalt, tungsten, titanium nitride, tantalum nitride, tantalum, titanium, one or more combinations of the foregoing, and any other suitable conductive material. The interconnect 10 may include a conductor-filled via 20 that allows connection to a step structure (e.g., Figure 6 The conductive material in the conductor-filled via 20 may be the same as the conductive material used for the interconnect 10.

[0065] The interconnect 10 and the conductor-filled via 20 may be formed using any suitable method known to those of ordinary skill in the art (e.g., subtractive formation), wherein the conductive material is deposited and defined by a photolithographic process. The conductive material may be deposited using any suitable method, such as evaporation, sputtering, chemical vapor deposition (CVD), atomic layer deposition (ALD), or electrochemical deposition, or any other suitable method. Etching during the photolithographic process may be achieved using any suitable method, such as wet acid or base, corrosive gas (e.g., chlorine, bromine or fluorine), or any suitable corrosive agent. Alternatively, the interconnect 10 and the conductor-filled via 20 may be formed using a damascene process also known to those of ordinary skill in the art. In a damascene process, a trench is formed in a dielectric layer (e.g., silicon oxide or silicon nitride), and a suitable conductive material is deposited into the trench. Excess conductive material on the dielectric layer and elsewhere may then be selectively removed using any suitable method, such as etching back with a corrosive gas, wet acid or base, or chemomechanical polishing (CMP), preferably by CMP. The trench may be formed using any suitable method, such as anisotropic etching.

[0066] refer to Figure 7b-1 After forming the interconnect 10 and the conductor-filled via 20, a first n-type or p-type semiconductor layer 30 (e.g., silicon, germanium, or silicon germanium) is deposited as an amorphous semiconductor layer using any suitable technique (e.g., CVD, ALD, or sputtering). Since the first semiconductor layer 30 is designated as a source or drain region of the vertical TFT 5 to be formed, the first semiconductor layer 30 is hereinafter referred to as the "first source or drain layer 30". Figure 7b-2 Shown from different perspectives Figure 7b-1 As described below, the first source or drain layer 30 is then crystallized. In one embodiment, the first source or drain layer 30 can be formed simultaneously with the common drain region of the NOR memory string (ie, formed from the same semiconductor material layer as the common drain region). Figure 7c An intermediate step is shown in which a vertical TFT 5 and a 3D memory structure 40A (including both a memory array and one or more stairs) in portion 40B are formed from multiple deposited semiconductor and insulator layers. For example, the formation of a 3D memory structure, such as memory structure 40A, has been disclosed in a non-provisional application. At this intermediate step, the processing of the vertical TFT portion 40B is separated from the processing of the memory structure portion 40A. Using appropriate masking and etching steps, the upper layers of the vertical TFT portion 40B are removed, stopping at the first source or drain layer 30. There may be other layers, such as a metal layer 46 tied to the first source or drain layer 30. Figure 7b-1 The structure can be obtained from Figure 7c The vertical TFT portion 40B is produced by removing the upper layer.

[0067] Afterwards, as in Figure 7d-1Isometric view in Figure 7d-2 As shown in the cross-section in FIG, after forming the first source or drain layer 30, a molded dielectric layer 50 (e.g., silicon oxide, silicon nitride, silicon oxynitride, silicon carbide, silicon carbide oxyhydrogen, fluorine-doped, boron-doped or phosphorus-doped silicon oxide, or any combination of these materials) is deposited on the first source or drain layer 30 using any suitable method, such as CVD, ALD or spin coating.

[0068] The molded dielectric layer 50 is masked, patterned, and etched to form a through hole exposing the first source or drain layer 30. Thereafter, after removing excess semiconductor material from the surface of the molded dielectric layer 50 by any suitable process, such as etch-back or CMP, most preferably CMP, a second semiconductor layer 60 (e.g., silicon, germanium, or silicon germanium) is deposited in the through hole using any suitable technique (e.g., CVD, ALD, or sputtering), such as in Figure 7e-1 Isometric view in Figure 7e-2 As shown in the cross section in . Since the second semiconductor layer 60 is designated to become a channel region of the vertical TFT 5 to be formed, the second semiconductor layer 60 is hereinafter referred to as the channel layer 60. The semiconductor layer 60 is preferably deposited as an amorphous semiconductor material.

[0069] The channel layer 60 may be in-situ doped of the same or opposite type as the first source or drain layer 30. Portions of the channel layer 60 may then be doped using an ex-situ doping step (e.g., gas diffusion or ion implantation) to produce a third semiconductor layer 70, such as in Figure 7f-1 Isometric view in Figure 7f-2 Most preferably, a third semiconductor layer 70 of the same doping type as the first source or drain layer 30 is provided. Since the third semiconductor layer 70 is designated as another source or drain region of the vertical TFT 5 to be formed, the third semiconductor layer 70 is referred to as the second source or drain layer 70 hereinafter.

[0070] In the case where the first source or drain layer 30, the channel layer 60, and the second source or drain layer 70 comprise amorphous silicon, when deposited, a thin titanium (Ti) layer 72 (e.g., about 10 nm thick) is subsequently deposited on the second source or drain layer 70, with or without a protective layer 74 (e.g., titanium nitride or any other suitable material) deposited on top of the titanium layer 72. Figure 7g-1 Isometric view in Figure 7g-2The resulting structure with the deposition of protective layer 74 is shown in cross section in FIG. Protective layer 74 is preferably deposited immediately after the deposition of titanium layer 72. The resulting structure is then annealed in an inert atmosphere (e.g., nitrogen or argon) at a temperature suitable for crystallization of amorphous silicon (e.g., between 550° C. and 1000° C. for 1 second to 24 hours). Alternatively, the annealing can be performed at 550° C. for 24 hours, at 600° C. for 12 hours, at 750° C. for 5 minutes, or at 800° C. for 1 minute, or at any suitable temperature for any duration to crystallize the silicon into high quality single crystal or multicrystalline silicon.

[0071] During the annealing process, the titanium layer 72 reacts with the underlying silicon to form titanium silicide 80, which provides a crystalline template suitable for forming low defect concentration crystalline silicon in the first source or drain layer 30, the channel layer 60, and the second source or drain layer 70. Figure 7h-1 and 7h-2 The low defect concentration crystalline silicon thus obtained has carriers with higher mobility than conventional polycrystalline semiconductor materials known to have high defect concentrations.

[0072] According to a second embodiment, a thin nickel layer may be deposited on the second source or drain layer 70 instead of the titanium layer 72. After the nickel deposition, the structure is annealed at a low temperature (e.g., between about 350° C. and about 450° C.). During the annealing, a nickel silicide film is formed that diffuses through the second source or drain layer 70, the channel layer 60, and the first source or drain layer 30, thereby inducing the crystallization of the amorphous silicon in these semiconductor layers into a low defect density form. When the nickel silicide film reaches the outer boundary of the silicon in the structure, the process is complete.

[0073] As in Figure 7h-1 Isometric view in Figure 7h-2 As shown in the cross section in FIG. , the titanium layer 72 and the protective layer 74 are then removed by selective wet etching (e.g., a combination of hydrogen peroxide and sulfuric acid). The titanium silicide film 80 remains unchanged, but in some embodiments, it can be removed by other means. In some embodiments, a crystalline semiconductor layer is formed at a later step in the manufacturing sequence.

[0074] At this point, the semiconductor pillar 90 (eg, a combination of the titanium silicide film 80, the crystallized channel layer 60, and the crystallized first and second source or drain layers 30 and 70) is substantially completed. Figure 7i As shown in , a trench extending substantially the length of the semiconductor pillar 90 is then created by etching the molded dielectric layer 50 surrounding the semiconductor layer 90. A gate oxide 100 is then conformally formed on the exposed surface of the semiconductor pillar 90 and at the bottom of the trench using any suitable technique (e.g., CVD, ALD, or thermal oxidation), as in Figure 7j-1 Medium distance and Figure 7j-2The gate oxide 100 may be any suitable material, such as silicon oxide, silicon nitride, aluminum oxide, hafnium oxide, hafnium silicon oxide, any combination of two or more of these materials, or any other suitable material.

[0075] Thereafter, a gate electrode layer 110 (e.g., titanium nitride, tantalum nitride, tungsten, any combination of two or more of these, or any suitable material) is deposited over the gate oxide 100 to completely fill the remaining trenches and over any exposed areas. Any gate electrode and gate oxide material in the exposed areas outside the trenches may be selectively removed using any suitable technique, such as etch back or CMP. After selectively removing the gate electrode layer 110 and gate oxide 100, the surface 130 of the semiconductor pillar 90 and the surface 120 of the gate electrode layer 110 are exposed, as shown in FIG. Figure 7k-1 Isometric view in Figure 7k-2 As shown in the cross section.

[0076] A dielectric layer is first deposited on the exposed gate electrode layer 110, gate oxide 100, and second source or drain layer 70. The dielectric layer is then patterned and etched to create through holes that expose a surface 120 of the gate electrode layer 110 and a surface 130 of the second source or drain layer 70 and provide trenches for interconnect conductors. Next, the through holes and trenches are filled with a conductive material, thereby forming an interconnect 150 that is electrically connected to the gate electrode layer 110 and the second source or drain layer 70 through the conductor-filled through hole 140, as shown in FIG. Figure 7l As shown in . At this point, the vertical TFT 5 is substantially completed. Since the channel layer 60 is crystallized in the presence of the metal silicide layer, the channel layer 60 has a low crystal defect concentration and a higher carrier mobility relative to a conventional polycrystalline semiconductor layer. The improved carrier mobility in the channel enables a higher current in the conductive vertical TFT.

[0077] Figure 8 FIG. 8 shows a vertical TFT 850 according to one embodiment of the present invention. Figure 7l Unlike the vertical TFT 5 of FIG. 850 , the vertical TFT 850 has a circular cross section in the XY plane. The vertical TFT 850 can be used to implement Figure 6 Vertical TFT 5. Figure 8 The vertical TFT 850 is shown in an exploded isometric view to more clearly show the first source or drain layer 30, the channel layer 60, the gate oxide layer 100, and the gate electrode layer 110. The manufacturing steps for the vertical TFT 850 with a circular cross-section are the same as those for the vertical TFT 850 with a rectangular cross-sectional shape. Figure 7l Those manufacturing steps of the vertical TFT 5 are basically the same.

[0078] Figures 9a to 9h Another process according to which the vertical TFT 850 can be formed according to one embodiment of the present invention is described. Figure 9a As shown in FIG. 1 , a through hole 901 is formed in the mold dielectric 50, thereby exposing the surface of the first source or drain electrode layer 30. Next, as shown in FIG. Figure 9b As shown in the isometric view in FIG. 1 , the insulating layer 36 is conformally deposited in the through hole 901, thereby completely covering any exposed portion of the first source or drain layer 30 within the through hole 901. The insulating layer 36 can be any suitable dielectric (e.g., about 10 nm of silicon oxide). Thereafter, as in FIG. Fig.9c As shown in the isometric view in FIG. 1 , the gate electrode layer 110 is conformally deposited on the insulating layer 36. Next, as shown in FIG. Figure 9d As shown in the isometric view in , portions of the gate electrode layer 110 and the insulating layer 36 are removed using anisotropic etching to expose the surface 102 of the first source or drain layer 30 , thereby substantially leaving the original thickness of the gate electrode layer 110 on the sidewalls of the through hole 901 .

[0079] Next, as in Fig.9e As shown in the isometric view in FIG. 1 , the gate oxide layer 100 is conformally deposited on the gate electrode layer 110 and on the surface 102 of the first source or drain layer 30. A portion of the gate oxide layer 100 is then removed by anisotropic etching, thereby exposing the surface 102 of the first source or drain layer 30, as shown in FIG. Figure 9f As shown in Figure 9g As shown in FIG. 1 , the channel layer 60 is deposited in an amorphous form into the through hole 901, thereby contacting the first source or drain layer 30. A portion of the channel layer 60 is then removed from the top surface by CMP. The second source or drain layer 70 is formed by doping the top portion of the channel layer 60 with a different concentration or relative type of dopant than in the channel layer 60, as shown in FIG. Figure 9g As shown in . Then, by first depositing a titanium film on the surface 71 of the second source or drain layer 70 in the manner described above and annealing at a temperature exceeding 550° C., a catalytic channel layer 60 and a crystallized TiSi2 seed layer of the first source or drain layer 30 and the second source or drain layer 70 are formed, and any amorphous semiconductor material in the first source or drain layer 30, the second source or drain layer 70 and the channel layer 60 can be crystallized, thereby producing minimal crystal defects and high carrier mobility in these semiconductor layers.

[0080] Interconnections and other operations for electrically connecting to the gate electrode layer 110 and the second source or drain layer 70 may then be achieved in a similar manner as described above.

[0081] Fig.10 FIG. 2 shows a vertical TFT 280 according to an embodiment of the present invention. Fig.10 As shown in FIG. 2 , the vertical TFT 280 includes (i) a heavily doped first source or drain layer 200a and a second source or drain layer 200b separated by a dielectric layer 210, (ii) a lightly doped channel layer 220 in contact with both the heavily doped source or drain layers 200a and 200b, (iii) a gate oxide layer 230 in contact with the channel layer 220, and (iv) a gate electrode layer 240 in contact with the gate oxide layer 230. Conductor contacts 250 provide individual contacts to the source or drain layers 200a and 200b and the gate electrode layer 240. In the vertical TFT 280, the channel layer 220 is fabricated so that charge carriers travel in a direction substantially perpendicular to the semiconductor substrate on which the vertical TFT 280 is formed (i.e., the Z direction). The channel layer 220 may be initially deposited as an amorphous semiconductor material, and the conductor contact 250 may include a titanium or nickel layer in contact with the heavily doped semiconductor layer 200a or 200b, or both. The titanium or nickel layer catalyzes the crystallization of any amorphous silicon in the channel layer 220 into a crystalline state with minimal crystal defects, thereby creating high mobility in the charge carriers in the channel layer 220.

[0082] Figures 11a to 11f The invention can be formed according to one embodiment of the present invention. Fig.10 The process of vertical TFT 280. Fig.11a As shown in FIG. 1 , a heavily doped first source or drain layer 200 a, a dielectric layer 210, and a heavily doped second source or drain layer 200 b are sequentially deposited as an amorphous semiconductor material. The resulting structure is then patterned and etched to form rectangular lines or strips, such as Fig.11a As shown in FIG. Subsequently, the channel layer 220 is then Fig.11a The channel layer 220, the gate oxide layer 230 and the gate electrode layer 240 are then deposited on the strips in succession. The resulting structure, i.e., structure 290, is then patterned and etched using a photolithography step so that the channel layer 220, the gate oxide layer 230, and the gate electrode layer 240 form a line around the strips, such as Fig.11b as shown in .

[0083] Afterwards, refer to Fig.11c , a dielectric layer 255 is deposited on the structure 290 for passivation. Fig.11d2 shows a cross section of structure 290 through semiconductor channel layer 220, gate oxide layer 230, and gate electrode layer 240. A via 260 is then etched in dielectric layer 255 to expose surface 270 of gate electrode layer 240. (Also or alternatively, a second via (not shown) may be etched into passivation layer 255 to expose a portion of second source or drain layer 200 b.) A conductor contact layer 250 (e.g., titanium or nickel) is then deposited into via 260 to contact surface 270. (Also or alternatively, a conductor contact layer is also deposited to contact the exposed portion of second source or drain layer 200 b.) After depositing the titanium or nickel, structure 290 is annealed, thereby allowing the TiSi2 or NiSi layer to form and catalyze the crystallization of first source or drain layer 200 a and second source or drain layer 200 b, as well as any adjacent semiconductor material in channel layer 220, into a crystalline form with minimal crystal defects. If necessary, any additional conductor contacts 250 may be fabricated to contact the heavily doped first source or drain layer 200a or the second source or drain layer 200b, such as Fig.10 as shown in .

[0084] Figures 12a to 12k A vertical TFT 1220 with high mobility connected to a memory cell in a NOR memory string of a memory structure 1210 according to yet another embodiment of the present invention is illustrated. As mentioned above, a TFT with a vertical channel occupies less silicon area than a TFT with a horizontal channel. In this regard, the silicon area occupied by a TFT is the area projected by the TFT on the surface of a planar substrate. Figures 12a to 12k In the embodiment illustrated in , only two masking steps may be used to form the vertical TFT 1220 because the vertical TFT 1220 has critical dimensions that are both self-aligned and self-limiting. With fewer masking steps, reduced manufacturing costs and greater manufacturing uniformity and consistency are achieved. Higher charge carrier mobility is required in the channel region of such a vertical TFT because the current of the vertical TFT in the conductive ("on") state is proportional to the mobility, and because a higher "on" current is required for such a vertical TFT in its operation in conjunction with the NOR memory string in the memory structure 1210.

[0085] In conjunction with the following description, an isometric view is shown Figures 12a to 12c and shown in a cutaway isometric view Figures 12d to 12k .like Fig.12a As shown in , a memory structure 1210 including an array of NOR memory strings is provided above a planar surface of a semiconductor substrate. The semiconductor substrate may include circuitry (not shown) connected to the memory structure 1210. Fig.12a1, vertical TFTs 1220 are provided in memory structure 1210 and connected to bit lines in a NOR memory string of memory structure 1210 through conductor-filled vias 1230. (Each bit line is a strip of heavily doped semiconductor material that forms a common drain terminal for the memory transistors of a NOR memory string.) Fig.12a As shown in , for exemplary purposes, each bit line is connected to two of the vias 1230. In practice, any suitable number of vias may be connected to each bit line. Moreover, since the vertical TFTs 1220 are provided in the stepped portion of the memory structure 1210, the vertical TFTs 1220 do not occupy any silicon area that is not already occupied by the memory structure 1210.

[0086] After forming the through-hole 1230, the formation of the vertical TFT 1220 begins using any suitable method, such as those disclosed above or in the provisional application. Thereafter, a conductive semiconductor layer 1240 is deposited, patterned, and etched into individual segments that are electrically isolated from each other. The conductive semiconductor layer 1240 may be n-type or p-type and may include silicon, silicon germanium, germanium, any suitable semiconductor material deposited using any suitable technique (e.g., CVD, ALD, or sputtering). Fig.12a As shown in , each segment of the conductive semiconductor layer 1240 is electrically connected to a bit line of a NOR memory string of the memory structure 1210 through two of the conductor-filled vias 1230 .

[0087] To facilitate the detailed description below, reference is made to Figure 12b 1240 is shown to describe the formation of the vertical TFT 1220. This section of the conductive semiconductor layer 1240 is hereinafter referred to as the "first source or drain layer 1240" because it is designated to become the source or drain region in the vertical TFT to be formed. Figure 12b After the first source or drain layer 1240 in the embodiment, a dielectric layer 1250 is deposited thereon ( Fig.12c ). The dielectric layer 1250 may comprise any suitable dielectric material (e.g., silicon oxide, silicon nitride, silicon oxynitride, aluminum oxide, silicon carbide, silicon oxycarbide, silicon oxycarbide), and may be deposited using any suitable technique (e.g., CVD or PVD). The dielectric layer 1250 is then patterned and etched to provide two cavities, namely a channel cavity 1260 and a gate 1270, respectively, to subsequently form channel regions and gate electrodes for two vertical TFTs to be formed therein. The width of the gate cavity 1270 is significantly narrower than that of the channel cavity 1260. The channel cavity 1260 and the gate cavity 1270 each extend the full length of the dielectric layer 1250 downward to the first source or drain layer 1240.

[0088] Reference is shown in a cutaway isometric view Fig.12d , an etch stop layer 1280 (e.g., a silicon nitride layer about 10 nm thick) is then conformally deposited inside the channel cavity 1260 and the gate cavity 1270 and on top of the dielectric layer 1250. Thereafter, as Fig.12e As shown in , a sacrificial layer 1290 (e.g., a silicon oxide layer approximately 30 nm thick) is conformally deposited on the etch stop layer 1280. Both the etch stop layer 1280 and the sacrificial layer 1290 may be deposited using any suitable technique (e.g., CVD or ALD). Fig.12e In FIG. 1 , due to its narrow width, the gate cavity 1270 is substantially filled after deposition of the etch stop layer 1280 and the sacrificial layer 1290 , while the open shaft remains in the channel cavity 1260 .

[0089] Afterwards, refer to Fig.12f , anisotropic etching ("punch-through etching") removes portions of the etch stop layer 1280 and the sacrificial layer 1290 to expose the surface 1300 of the first source or drain layer 1240 to the channel cavity 1260. The punch-through etching leaves the etch stop layer 1280 and the sacrificial layer 1290 on the sidewalls of the channel cavity 1260 and the gate cavity 1270.

[0090] Afterwards, refer to Figure 12g As shown in the drawings and illustrations thereof, a channel layer 1310 (e.g., a suitable n-type or p-type amorphous silicon, silicon germanium, germanium, or another semiconductor film) is deposited using any suitable technique (e.g., CVD, ALD, or sputtering) to fill the channel cavity 1260. The channel layer 1310 is in electrical contact with the conductive semiconductor layer 1240 at the surface 1300. The channel layer 1310 may be in-situ doped during deposition. Portions of the channel layer 1310 may be removed from the top surface of the dielectric layer 1250 using an etch-back operation or CMP.

[0091] Thereafter, the channel layer 1310 is crystallized using any suitable technique (e.g., using a crystallization process in the presence of titanium silicide or nickel silicide as discussed above) to provide a semiconductor pillar with a low crystal defect density, thereby producing a high charge carrier mobility and enabling a large amount of current. For example, the channel layer 1310 can be crystallized in the presence of TiSi2, which is produced by the reaction of the thin titanium layer 1320 with the channel layer 1310 during annealing. Optionally, a protective layer (not shown) can be deposited on the channel layer 1310. The protective layer can be titanium nitride or any other suitable material. The structure is then annealed in an inert atmosphere (e.g., nitrogen or argon) to a suitable crystallization temperature (e.g., between 550°C and 1000°C for 1 second to 24 hours; specifically, at 550°C for 24 hours, at 600°C for 12 hours, at 750°C for 5 minutes, or at 800°C for 1 minute, or according to any suitable annealing schedule). Reference Figure 12hDuring the annealing, a portion of the titanium film 1320 reacts with the semiconductor in the channel layer 1310 to form a titanium silicide layer 1330. In this process, the titanium silicide layer 1330 is formed before the crystallization of the channel layer 1310 is completed, thereby providing a crystal template suitable for forming a low defect density crystalline semiconductor with high carrier mobility.

[0092] refer to Fig.12i , and then doping a portion of the crystallized channel layer 110 by ion implantation to form a heavily doped second source or drain layer 1345, followed by an annealing step (if necessary) to activate the implanted dopant atoms. The annealing also causes the dopant atoms in the first source or drain layer 1240 to diffuse into the channel layer 1310 to form a lightly doped drain (LDD) region 1340. The doping concentration of the LDD region 1340 is less than the doping concentration of the first source or drain layer 1240 and the second source or drain layer 1345 and greater than the doping concentration of the channel layer 1310.

[0093] refer to Fig.12j , and then remove the sacrificial layer 1290 by selective etching (e.g., hydrofluoric acid, which removes silicon oxide much faster than silicon nitride). Thereafter, reference Figure 12k , the gate cavity 1270 is partially filled by conformally depositing or growing a gate dielectric layer 1350 (e.g., silicon oxide, silicon nitride, silicon oxynitride, aluminum oxide, or hafnium oxide) with a portion of its sacrificial layer 1290 removed. The thickness of the gate dielectric layer 1350 may be between 3 nm and 100 nm, in particular between 5 nm and 20 nm. Thereafter, the gate cavity 1270 is filled by a gate electrode or conductive layer 1360 using any suitable conductive material (e.g., doped silicon, doped silicon germanium, doped germanium, titanium, titanium nitride, tungsten, tungsten nitride, tungsten carbide, tungsten carbide nitride, tantalum, or tantalum nitride, or any combination thereof), as in Figure 12k Any excess material of the gate conductive layer 1360 on the top surface of the etch stop layer 1280 (exposed by removing the sacrificial layer 1290) may be removed by etch back or CMP. Fig.12a As shown in , it is now possible to contact both the gate conductive layer 1360 and the channel layer 1310, respectively. Now it is basically completed Fig.12a The vertical TFT 1220 is manufactured. In each vertical TFT, the first drain or source layer 1240 and the LDD layer 1340 together serve as a drain region (or alternatively a source region), the second source or drain layer 1345 serves as a source region (or alternatively a drain region), the channel layer 1310 serves as a channel region, the gate dielectric layer 1350 serves as a gate dielectric material, and the gate conductive layer 1360 serves as a gate electrode.

[0094] Fig.13 Show Figures 12a to 12kA vertical TFT with some dielectric layers removed for clarity.

[0095] The above detailed description is provided to illustrate specific embodiments of the invention and is not intended to be limiting. Numerous modifications and variations are possible within the scope of the invention. The invention is set out in the appended claims.

Claims

1. A memory circuit, comprising: a semiconductor substrate having a planar surface, the semiconductor substrate having circuitry formed therein for memory operations; A memory structure formed above the planar surface, comprising a memory array portion and a staircase portion, wherein (i) the memory array portion comprises a plurality of memory cells organized into a plurality of levels of NOR memory strings, each NOR memory string having memory cells sharing a common bit line, each memory cell being associated with a word line as a gate electrode; and (ii) the staircase portion has a plurality of steps; and (iii) the common bit line of the NOR memory strings in each level of NOR memory strings is disposed on a first conductive layer within the NOR memory strings of the level, the first conductive layer having conductors that carry electrical signals to or from the common bit line of the memory cells in the NOR memory strings of each level, the conductors in the first conductive layer extending along a first direction parallel to the planar surface and between the memory array portion of the memory structure and corresponding steps of the stepped portion of the memory structure, wherein each word line extends along a second direction orthogonal to the planar surface; as well as A selection circuit comprising one or more transistors, each transistor being formed above or below a conductor of a corresponding first conductive layer but above the planar surface of the semiconductor substrate, so that the selection circuit projects a footprint onto the semiconductor substrate within the footprint of the memory structure, wherein each of the transistors is electrically connected to a corresponding one of the conductors in that first conductive layer when each of the transistors is turned on by a control signal, wherein each of the transistors comprises a first drain or source region, a second drain or source region and a gate region, each region being formed of a semiconductor material or a conductive material.

2. A memory circuit according to claim 1, wherein at least one of the first drain or source region, the second drain or source region and the gate region has a metal silicide layer formed thereon; and wherein each transistor in the selection circuit selectively connects a corresponding conductor in the first conductive layer to the circuit system for memory operation. 3 . The memory circuit of claim 2 , wherein the metal silicide layer comprises titanium silicide or nickel silicide.

4. The memory circuit of claim 1 , further comprising an interconnect layer having a plurality of conductors, each conductor in the interconnect layer being electrically connected to one or more of the transistors in the selection circuit, such that when one of the connected transistors of the selection circuit is rendered conductive, the conductor in the interconnect layer is electrically connected to the first conductive layer associated with the connected transistor.

5. The memory circuit of claim 1 , wherein the memory cells of each NOR memory string share a common source line; wherein the common source line of the NOR memory strings in each level of NOR memory strings is disposed on a second conductive layer in the memory array.

6. The memory circuit of claim 1 , wherein the transistors of the selection circuit are formed above the first conductive layer and are electrically connected, and when turned on, the conductors in the first conductive layer are arranged in rows, and the transistors in each row are arranged along a third direction parallel to the planar surface of the semiconductor substrate and orthogonal to the first direction.

7. The memory circuit of claim 1, further comprising a set of interconnect lines, wherein a gate terminal of each transistor in the selection circuit is connected to one of the interconnect lines.

8. The memory circuit according to claim 1, wherein when a current exists in a channel region of one of the transistors in the selection circuit, the current flows in a direction perpendicular to the planar surface.

9. The memory circuit of claim 1, wherein the semiconductor material comprises one or more of: polycrystalline silicon, single crystal silicon, and silicon germanium.

10. The memory circuit of claim 1 , wherein the transistors of the selection circuit form a first group and a second group such that adjacent conductors of the first conductive layer are connected by transistors from the first group and transistors from the second group, and wherein contacts of the transistors in the first group are placed in a staggered manner relative to contacts of the transistors in the second group.

Citation Information

Cited By

  • Memory device using semiconductor element

    US12677408B2