Stacked complementary field effect transistor and logic device
By adopting nanowire structure and vertically distributed doped wire segments in CFETs, the short channel effect problem is solved, and higher integration density and performance improvement is achieved, solving the challenges of traditional CFETs in size reduction and power consumption control. .
Patent Information
- Application Number
- CN202411165869.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-23
- Publication Date
- 2024-11-29
AI Technical Summary
The lateral arrangement of source and drains in existing CFET architectures leads to a short channel effect, limiting the shrinkage and space utilization of the overall size of the device, resulting in increased chip area, increased power consumption and signal transfer affected.
Using a nanowire structure, NMOS and PMOS are formed through the first and second line segments doped by N-type and P-type. The source and drain semiconductors are distributed up and down, the electron flow direction is vertical, the channel length and device Dimensions are released from mutual limitations and are connected by insulating isolation medium or conductive layers to form a sleeve-like channel structure to extend the effective channel length.
The number of integrated transistors per unit area is significantly improved, the device size is reduced, parasitic capacitance and resistance is reduced, and the device's integration density and performance is improved, especially in high-frequency circuits, which reduce power consumption.
Smart Images

Figure CN119050116A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of semiconductor devices, and in particular to a stacked complementary field effect transistor and a logic device. Background Art
[0002] With the development of integrated circuit technology, transistors have become the cornerstone of modern electronic technology. In order to achieve higher performance and more complex functions, the number of transistors required for modern high-performance circuits is increasing dramatically. As the number of transistors integrated on the chip becomes larger and larger, the traditional planar CMOS structure has approached the limit in terms of 2D plane space utilization. As the number of transistors required for high-performance electronic devices increases, the chip area also increases dramatically, which increases the chip manufacturing cost. At the same time, the large parasitic resistance and large parasitic capacitance caused by the large chip area cause a significant increase in the power consumption of the chip, and the transmission of internal chip signals will also be affected.
[0003] In recent years, a new spatial architecture, Complementary Field-Effect Transistors (CFET), has greatly alleviated the challenges of traditional 2D planar transistors in size reduction and power consumption control from a new 3D topological structure. New logic devices based on the CFET structure, such as CMOS inverters, vertically stack P-type and N-type transistors, greatly reduce the device's footprint, allowing more logic units to be integrated on the same chip area. This vertical stacking structure reduces the impact of resistance and capacitance on transistors, can effectively control leakage current, significantly reduce chip power consumption, and improve the switching speed and overall performance of the circuit.
[0004] However, the CFET architecture currently proposed is usually a vertical stack of lateral GAAFET (Gate-All-Around Field Effect Transistor) structures, such as Figure 1 As shown. However, in this structure, the source, drain and channel are lateral, and the channel region is clamped by the source and drain. This results in the overall device size being limited when the channel length is constant. In the case of a small area, it is easy to face a serious short channel effect, that is, the gate's ability to control the channel is weakened, which may lead to an increase in leakage current, affecting the device's switching ratio and energy consumption. Summary of the invention
[0005] The present invention provides a stacked complementary field effect transistor to solve the problem in the prior art that the source and drain are arranged laterally, the overall size of the device is limited due to the short channel effect, and the space utilization is low.
[0006] To solve the above problems, the present invention provides a CFET, which includes a nanowire, wherein the nanowire includes at least a first N-type doped segment and a second P-type doped segment; the first segment and the second segment include a source semiconductor, a channel semiconductor, and a drain semiconductor, forming NMOS and PMOS respectively; the source semiconductor and the drain semiconductor in the first segment and the second segment are distributed up and down, so that the direction of electron flow between the source semiconductor and the drain semiconductor is vertical, and the channel length and the device size (area) will no longer limit each other.
[0007] In the present invention, NMOS and PMOS are constructed on the same nanowire, and based on the vertical structural characteristics of the nanowire, the source semiconductor, channel semiconductor, and drain semiconductor are stacked in the vertical direction, which effectively improves the space utilization in the vertical direction and greatly increases the number of transistors that can be integrated on a chip per unit area.
[0008] Furthermore, the embodiment of the present invention also provides two arrangements of the source semiconductor, the channel semiconductor, and the drain semiconductor in the first line segment / the second line segment:
[0009] Arrangement method 1: the source semiconductor, the channel semiconductor, and the drain semiconductor are connected in sequence; specifically, as for the first line segment of NMOS and the second line segment of PMOS, the source semiconductor, the channel semiconductor, and the drain semiconductor can be arranged symmetrically (NS-NG-ND-PD-PG-PS or ND-NG-NS-PS-PG-PD) or asymmetrically (NS-NG-ND-PS-PG-PD or ND-NG-NS-PD-PG-PS), wherein NS represents the source semiconductor in the first line segment of NMOS, ND represents the drain semiconductor in the first line segment of NMOS, NG represents the gate semiconductor in the first line segment of NMOS, PS represents the source semiconductor in the second line segment of PMOS, PD represents the drain semiconductor in the second line segment of PMOS, and PG represents the gate semiconductor in the second line segment of PMOS.
[0010] The preparation scheme may be:
[0011] (1) During the growth of nanowires, the doping type and doping concentration are regulated to form heterogeneous doped nanowires having N+, N-, N+, P+, P-, and P+ in sequence; N+ represents N-type high-concentration doping, N- represents N-type low-concentration doping; P+ represents P-type high-concentration doping, and P- represents P-type low-concentration doping
[0012] (2) Etch at the junction of N+ and P+ and fill it with insulating isolation medium or metal medium.
[0013] Arrangement 2: Both the first line segment and the second line segment have a core-shell structure, including a trunk portion connected in sequence by a source semiconductor, an isolation medium, and a drain semiconductor, and a channel semiconductor coated on the outside of the trunk portion. In this arrangement, the source semiconductor and the drain semiconductor are insulated and isolated by an insulating isolation medium, which blocks the direct migration of carriers between the source semiconductor and the drain semiconductor, making the outer sleeve-shaped channel semiconductor the only migration path. The length of the sleeve-shaped channel semiconductor can almost match the nanowire itself, so that the channel length is almost unaffected by the length of the source and drain portions of the nanowire, and only depends on the overall length of the nanowire. In this way, the channel is cleverly "pulled out" to the outside, greatly extending the effective channel length at the same device size. Compared with the existing nanowire GAAFET, under the same manufacturing process and the same channel length, the device size of the transistor of the present application can be significantly reduced, thereby improving the integration density of the device. According to the existing nanowire GAAFET process, for example, to form a 3nm channel length, through source and drain layout, the length of the nanowire itself needs to be 7nm. However, through the present invention, the overall device size (nanowire length) can be compressed to nearly 3nm, effectively improving the device's miniaturization capability, so that the number of transistors integrated per unit chip area is almost doubled. The further miniaturization of the device size leads to a reduction in parasitic capacitance and parasitic resistance, while the length of the device interconnection is also shortened, which reduces the signal propagation delay, thereby improving the speed and performance of the device, especially in high-frequency circuits, the reduced parasitic capacitance can also reduce the power consumption of the device.
[0014] The preparation scheme may be:
[0015] (1) regulating the doping type and doping concentration during the nanowire growth process to form heterogeneous doped nanowires with N+ and P+;
[0016] (2) Etching in the N+ doped nanowire segment and filling it with an isolation dielectric; etching in the P+ doped nanowire segment and filling it with an isolation dielectric; etching at the junction of N+ and P+ and filling it with an insulating isolation dielectric or a metal dielectric;
[0017] (3) Depositing an N-coating layer on the outside of the N+ doped nanowire segment to form an N-type channel semiconductor surrounding the columnar channel structure; depositing a P-coating layer on the outside of the P+ doped nanowire segment to form a P-type channel semiconductor surrounding the columnar channel structure.
[0018] In an embodiment of the present application, the length of the isolation medium accounts for no less than 72% of the length of the nanowire structure.
[0019] Similarly, in arrangement 2, the source semiconductor, insulating isolation medium, and drain semiconductor can be arranged symmetrically (NS-NI-ND-PD-PI-PS or ND-NI-NS-PS-PI-PD) or asymmetrically (NS-NI-ND-PS-PI-PD or ND-NI-NS-PD-PI-PS), where NI represents the isolation medium in the first line segment of NMOS, and PI represents the isolation medium in the second line segment of PMOS.
[0020] As a transistor device, those skilled in the art know that the source semiconductor is connected to the source, the channel semiconductor is connected to the gate, and the drain semiconductor is connected to the drain. In an embodiment of the present application, there is an oxide layer between the channel semiconductor and the gate to form a gate sidewall of a sleeve structure.
[0021] In the embodiments of the present application, the source semiconductor and drain semiconductor of NMOS are N-type doped Si, Ge or III-V semiconductor (GaAs, GaN, InAs, InN, InP, etc.). The source semiconductor and drain semiconductor of PMOS are P-type doped Si, Ge or III-V semiconductor (GaAs, GaN, InAs, InN, InP, etc.).
[0022] In the embodiments of the present application, the channel semiconductor is Si, Ge or a III-V intrinsic semiconductor, or a P-type doped or N-type doped Si, Ge or a III-V semiconductor (InGaN, InGaAs, AlGaN, GaAsP, etc.). Those skilled in the art should know that in the first line segment as NMOS, the channel semiconductor is N-type doped, and in the second line segment as PMOS, the channel semiconductor is P-type doped.
[0023] In the embodiment of the present application, the source semiconductor and the drain semiconductor are formed by etching a partial region of a semiconductor nanowire.
[0024] In an embodiment of the present application, the insulating isolation medium is one or more of a vacuum medium, an air medium, an organic substance or an oxide.
[0025] Furthermore, the embodiment of the present invention also provides two connection relationships between the first line segment and the second line segment to assemble and form a nanowire structure:
[0026] Connection mode 1: There is an insulating isolation layer between the first wire segment and the second wire segment. The insulating isolation layer structurally connects the first wire segment and the second wire segment to form a whole nanowire. The insulating isolation layer makes the first wire segment and the second wire segment independent of each other, so the arrangement of the semiconductors in the two wire segments can be arbitrarily arranged. A window structure for electrical contact with the nanowire can be provided near the insulating isolation layer.
[0027] Connection method 2: There is a conductive layer between the first line segment and the second line segment, and the conductive layer serves as a common source or drain or a common source / drain of NMOS and PMOS. At this time, the arrangement of the semiconductors in the two line segments should take into account the situation of the conductive layer. When the source semiconductors of NMOS and PMOS are located on both sides of the adjacent conductive layer (ND-NI-NS-C-PS-PI-PD), the common source of NMOS and PMOS; when the drain semiconductors of NMOS and PMOS are located on both sides of the adjacent conductive layer (NS-NI-ND-C-PD-PI-PS), the common drain of NMOS and PMOS, at this time, this structure is particularly suitable for logic devices. In some devices, there is also a shared source / drain situation (such as NS-NI-ND-C-PS-PI-PD). At this time, the device can perform special functions by setting the potential of the NS terminal and the PD terminal.
[0028] The present invention also provides a logic device, which is composed of the stacked complementary field effect transistors of the aforementioned nanowire structure; the NMOS and PMOS share a common gate and a common drain; the sources of the NMOS and PMOS are connected to the ground and the power supply respectively.
[0029] The beneficial effect of the present invention is that: the present invention breaks through the density limitation of transistors per unit plane area of the chip through structural optimization design, provides sufficient physical dimensions for building higher performance chips and further miniaturizing the size of transistors, while still maintaining or even improving the performance indicators of GAAFET. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 Schematic diagram of existing CFET technology;
[0031] Figure 2 A schematic diagram of the stacked complementary field effect transistor of the present invention formed by connecting the first line segment and the second line segment using an insulating medium in arrangement mode 1;
[0032] Figure 3 A schematic diagram of the stacked complementary field effect transistor of the present invention formed by connecting the first line segment and the second line segment using a conductive medium in arrangement mode 1;
[0033] Figure 4 A schematic diagram of the stacked complementary field effect transistor of the present invention formed by connecting the first line segment and the second line segment using an insulating medium in arrangement mode 2;
[0034] Figure 5 A schematic diagram of the stacked complementary field effect transistor of the present invention formed by connecting the first line segment and the second line segment using a conductive medium in arrangement mode 2;
[0035] Figure 6Schematic diagram of a logic device formed after electrical contact, wherein (a) is a first line segment and a second line segment connected by an insulating medium, and (b) is a first line segment and a second line segment connected by a conductive medium;
[0036] Figure 7.1-Figure 7.10 1 is a process flow chart of the present invention, wherein 100 is a nanowire; 200 is an insulating medium such as a photoresist, which is used for patterned etching or connecting semiconductors; 201 is an insulating material such as an oxide; 101 is a first segment of the nanowire; 102 is a second segment of the nanowire; 103 is an insulating isolation layer; 1031 is an insulating medium; 1032 is a windowed conductive medium; 104 is a conductive layer; 1041 is an insulating medium; 1011 is a source semiconductor of the first segment of the nanowire; 1012 is a drain semiconductor of the first segment of the nanowire; 1021 is The source semiconductor of the second segment of the nanowire; 1022 is the drain semiconductor of the second segment of the nanowire; 1013 is the channel semiconductor of the first segment of the nanowire; 1023 is the channel semiconductor of the second segment of the nanowire; 1014 is the insulating isolation layer of the first segment of the nanowire; 1024 is the insulating isolation layer of the second segment of the nanowire; 1015 is the gate oxide layer of the first segment of the nanowire; 1025 is the gate oxide layer of the second segment of the nanowire; 1016 is the gate structure of the first segment of the nanowire; 1026 is the gate structure of the second segment of the nanowire. DETAILED DESCRIPTION
[0037] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.
[0038] The drawings are only for illustrative purposes and cannot be construed as limiting the present invention. To better illustrate the present embodiment, some parts of the drawings may be omitted, enlarged, or reduced, and do not represent the size of the actual product. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings. The positional relationships described in the drawings are only for illustrative purposes and cannot be construed as limiting the present invention.
[0039] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.
[0040] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.
[0041] The raw materials used in the present invention, unless otherwise specified, are conventional commercially available products; the methods used in the present invention, such as etching, deposition, etc., are conventional methods in the art, unless otherwise specified.
[0042] The field-effect transistor (FET) described in the present invention is a semiconductor device that uses electric field effect to control current; its core working principle is to change the conductivity of the channel by applying a voltage between the gate and the channel, thereby controlling the current from the drain to the source. This voltage does not directly drive the current, but indirectly controls the current through the electric field effect. Therefore, in this application, the part connected to the source in the semiconductor device is called the source semiconductor, and the part connected to the drain is called the drain semiconductor. The source and the drain can be interchanged in principle, and the part connected to the gate is called the channel semiconductor. In addition, as common knowledge in the art, the channel semiconductor is not electrically connected to the gate, but an oxide layer is set between them to form a gate sidewall to construct an electric field effect.
[0043] The stacked complementary field effect transistor (CFET) described in the present invention is an innovative three-dimensional transistor structure, which uses nanosheet or nanowire technology to vertically stack p-type and n-type MOS in the same unit, and forms a complementary structure through different electrical interconnections, corresponding to the planar CMOS structure. This design breaks through the limitations of traditional two-dimensional planar layout on space utilization, allowing transistors to work effectively in a smaller space, improve integration density and reduce power consumption.
[0044] The source and drain semiconductors in the p-type MOS and n-type MOS of CFET are doped semiconductors to provide sufficient carriers. The choice of doping type (N-type or P-type) depends on the type of MOS designed, that is, N-channel MOS or P-channel MOS. In n-type MOS, the source and drain semiconductors are generally N-type doped by doping pentavalent elements (such as phosphorus or arsenic), which can provide a large number of free electrons as carriers. Under the action of external voltage, these electrons can move in the conductive channel controlled by the gate voltage, thereby forming a current between the source and the drain. The channel semiconductor may be an intrinsic semiconductor or a lightly doped N-type semiconductor so that the concentration and flow of carriers in the channel can be controlled by the gate voltage. In p-type MOS, the source and drain semiconductors are generally P-type doped by doping trivalent elements (such as boron or gallium), which can generate a large number of holes as carriers. Under the appropriate gate voltage, these holes can move between the source and the drain. The channel semiconductor may be an intrinsic semiconductor or a lightly doped P-type semiconductor to achieve effective control of the channel conductivity.
[0045] In addition, the source semiconductor and drain semiconductor are usually heavily doped to reduce contact resistance and improve conductivity. The channel semiconductor is lightly doped or intrinsic semiconductor to ensure that the gate voltage can effectively control the conductive state of the channel. In some high-performance MOS, the channel semiconductor may be designed with a special doping distribution, such as gradient doping, to optimize performance.
[0046] Specifically, the source semiconductor and the drain semiconductor may be Si, Ge or a III-V semiconductor (GaAs, GaN, InAs, InN, InP, etc.), and the channel semiconductor may be Si, Ge or a III-V semiconductor.
[0047] In addition, those skilled in the art should know that the source semiconductor is connected to the source, the drain semiconductor is connected to the drain, and the gate semiconductor is connected to the gate. The magnitude of the current between the source and the drain is controlled by changing the gate voltage. In certain embodiments of the present invention, only the layout of the source semiconductor, the drain semiconductor, and the gate semiconductor is provided. Based on the layout of the source semiconductor, the drain semiconductor, and the gate semiconductor of the present invention, those skilled in the art can form a device and achieve the technical effect described in the present invention in combination with conventional electrode connection methods.
[0048] The existing CFET architecture is based on a vertical stack of lateral GAAFET (Gate-All-Around Field Effect Transistor), such as Figure 1As shown, under the channel length of a certain technical node, the device size cannot be further miniaturized, resulting in the mutual restriction of the channel length and the device size. The present invention provides a CFET with a nanowire structure. The nanowire at least includes a first line segment doped with N type and a second line segment doped with P type; the first line segment and the second line segment include a source semiconductor, a channel semiconductor, and a drain semiconductor, which constitute NMOS and PMOS respectively; the source semiconductor and the drain semiconductor in the first line segment and the second line segment are distributed up and down, so that the direction of electron flow between the source semiconductor and the drain semiconductor is vertical, and the channel length and the device size (area) will no longer restrict each other. Those skilled in the art can meet actual needs by scaling the length of the source semiconductor and the drain semiconductor, or by changing the length of the nanowire, without affecting the overall area of the device.
[0049] Figure 2 The first implementation of the present invention is given, in which the first line segment 101 as NMOS and the second line segment 102 as PMOS are stacked up and down, and the two are connected by an insulating isolation layer 103. In the first line segment 101 as NMOS, the N-type source semiconductor 1011, the N-type channel semiconductor 1013, and the N-type drain semiconductor 1012 are connected up and down in sequence; in the second line segment 102 as PMOS, the P-type source semiconductor 1021, the P-type channel semiconductor 1023, and the P-type drain semiconductor 1022 are connected up and down in sequence. Among them, the source semiconductor, the drain semiconductor, and the gate semiconductor can all lead out electrodes through the side wall of the nanowire. Those skilled in the art can meet actual needs by scaling the length of the source semiconductor and the drain semiconductor, or by changing the length of the nanowire, without affecting the overall area of the device.
[0050] Based on the stacking method of the first line segment and the second line segment in the present invention, Figure 3 Another implementation method is provided, that is, replacing the insulating isolation layer 103 with a conductive layer 104, so that the upper and lower MOS can share one electrode, further optimizing the device structure. Figure 3 In (a), the drain semiconductors 1012 and 1022 of NMOS and PMOS are located on two adjacent sides of the conductive layer 104 (NS-NI-ND-C-PD-PI-PS), and NMOS and PMOS share a common drain. At this time, the structure is particularly suitable for logic devices. Figure 3 In (b), the source semiconductors 1011 and 1021 of the NMOS and PMOS are located on two adjacent sides of the conductive layer 104 (ND-NI-NS-C-PS-PI-PD), and the NMOS and PMOS share a common source. In some devices, there is also a shared source / drain situation (such as
[0051] NS-NI-ND-C-PS-PI-PD), at this time, the device can play a special function by setting the potential of the NS terminal and the PD terminal.
[0052] The channel is the electron migration channel from source to drain. Therefore, although there are various structural forms of MOS in the prior art, the channel semiconductor is set between the source semiconductor and the drain semiconductor, which leads to the mutual restriction of the channel length and the device size. Based on this problem, the present invention extracts the channel semiconductor originally located between the source semiconductor and the drain semiconductor, replaces it with a P-type insulating isolation medium 1024 and an N-type isolation medium 1014, and combines it with the corresponding source semiconductor and drain semiconductor to form a nanowire structure; the source semiconductor is connected to the source, and the drain semiconductor is connected to the drain; the extracted channel semiconductor is set as a sleeve structure, which is coated on the outside of the nanowire structure (1011-1014-1012) and (1021-1024-1022); the channel semiconductor is connected to the gate. Compared with the prior art, through the setting of the insulating isolation medium, the channel is cleverly "extracted" to the outside, solving the problem of the mutual restriction of the channel length and the device size. Figure 4-5 Part of the structure of the present invention is shown.
[0053] Figure 4 In (a), the first line segment 101 as NMOS and the second line segment 102 as PMOS are stacked up and down, and the two are connected by an insulating isolation layer 103. In the first line segment 101 as NMOS, the N-type source semiconductor 1011, the N-type isolation dielectric 1014, and the N-type drain semiconductor 1012 are sequentially connected up and down; in the second line segment 102 as PMOS, the P-type source semiconductor 1021, the P-type insulating isolation dielectric 1024, and the P-type drain semiconductor 1022 are sequentially connected up and down; thus, the layout of the nanowire is: P-type source semiconductor 1021-P-type insulating isolation dielectric 1024-P-type drain semiconductor 1022-insulating isolation layer 103-N-type source semiconductor 1011-N-type isolation dielectric 1014-N-type drain semiconductor 1012. In addition, the N-type channel semiconductor 1013 and the P-type channel semiconductor 1023 are configured as a sleeve structure, covering the outer sides of the nanowire structures (1011-1014-1012) and (1021-1024-1022); Figure 4In (a), the N-type channel semiconductor 1013 and the P-type channel semiconductor 1023 are isolated by the insulating isolation layer 103. The gate oxide (the second gate oxide 1025, the first gate oxide 1015) and the gate (the second gate 1026, the first gate 1016) are sequentially coated on the outside of the channel semiconductor. In the figure, the sleeve structure composed of the channel semiconductor, the gate oxide and the gate has a window structure for leading electrodes to the internal N-type source semiconductor 1011 and the P-type drain semiconductor 1022; the window structure is filled with a window conductive medium 1032 and an insulating medium 1031 for isolating the window conductive medium 1032 and the sleeve structure. In this embodiment, the window structure is adjacent to the insulating isolation layer 103 to ensure the integrity of the channel.
[0054] Figure 4 In (b), the first line segment 101 as NMOS and the second line segment 102 as PMOS are stacked up and down, and the two are connected by an insulating isolation layer 103. In the first line segment 101 as NMOS, the N-type source semiconductor 1011, the N-type isolation dielectric 1014, and the N-type drain semiconductor 1012 are sequentially connected up and down; in the second line segment 102 as PMOS, the P-type drain semiconductor 1022, the P-type insulating isolation dielectric 1024, and the P-type source semiconductor 1021 are sequentially connected up and down; thus, the layout of the nanowire is: P-type source semiconductor 1021-P-type insulating isolation dielectric 1024-P-type drain semiconductor 1022-insulating isolation layer 103-N-type drain semiconductor 1012-N-type isolation dielectric 1014-N-type source semiconductor 1011. In addition, the N-type channel semiconductor 1013 and the P-type channel semiconductor 1023 are configured as a sleeve structure, covering the outer sides of the nanowire structures (1011-1014-1012) and (1021-1024-1022); Figure 4 In (b), the N-type channel semiconductor 1013 and the P-type channel semiconductor 1023 are isolated by the insulating isolation layer 103, and a window structure composed of an insulating medium 1031 and a window conductive medium 1032 is provided near the insulating isolation layer 103 to facilitate electrical connection. The window conductive medium 1032 is used to introduce the potential of the P-type drain semiconductor 1022 and the P-type source semiconductor 1021.
[0055] As Figure 4 In an equivalent manner, the nanowire layout may also be: P-type drain semiconductor 1022-P-type insulating isolation medium 1024-P-type source semiconductor 1021-insulating isolation layer 103-N-type source semiconductor 1011-N-type isolation medium 1014-N-type drain semiconductor 1012, providing a structural advantage for the common source of the two line segments.
[0056] Figure 5In (a), the first line segment 101 as NMOS and the second line segment 102 as PMOS are stacked up and down, and the two are connected by a conductive layer 104. In the first line segment 101 as NMOS, the N-type source semiconductor 1011, the N-type isolation dielectric 1014, and the N-type drain semiconductor 1012 are connected in sequence from bottom to top; in the second line segment 102 as PMOS, the P-type source semiconductor 1021, the P-type insulating isolation dielectric 1024, and the P-type drain semiconductor 1022 are connected in sequence from top to bottom; thus, the layout of the nanowire is: P-type source semiconductor 1021-P-type insulating isolation dielectric 1024-P-type drain semiconductor 1022-conductive layer 104-N-type drain semiconductor 1012-N-type isolation dielectric 1014-N-type source semiconductor 1011. In addition, the N-type channel semiconductor 1013 and the P-type channel semiconductor 1023 are configured as a sleeve structure, which is coated on the outside of the nanowire structure (1011-1014-1012) and (1021-1024-1022); although the N-type channel semiconductor 1013 and the P-type channel semiconductor 1023 are structurally isolated by the conductive layer 104, the conductive layer 104, as a common drain of the N-type drain semiconductor 1012 and the P-type drain semiconductor 1022, should be insulated and isolated from the channel semiconductor. Therefore, Figure 5 In (a), an isolation medium 1041 is provided.
[0057] Figure 5 The layout of (b) is a common source setting. Figure 5 The difference from (a) is that the conductive layer 104 serves as a common source for the P-type source semiconductor 1021 and the N-type source semiconductor 1011 .
[0058] Figure 6 The present invention shows a CMOS inverter structure logic device formed by interconnecting the gates of PMOS and NMOS as input terminals, interconnecting the drains as output terminals, and connecting the sources to the power supply and the ground terminals respectively. Figure 6 (a) and Figure 6 (b) are logic devices formed by stacked complementary field effect transistors composed of an insulating isolation layer 103 and a conductive layer 104. The right side is an electrical schematic diagram of the logic device. Based on this structure, researchers in this field can further develop various other logic gate circuit structures.
[0059] In order to obtain the above transistor, a preparation method based on current process means is provided as follows.
[0060] Controlling the doping type and doping concentration during the nanowire growth process is a common technical means in the art, and therefore is not reflected in the process flow chart of FIG. 7 .
[0061] like Figure 7.1As shown in (a) and (b), after spin coating the photoresist 200 on the nanowire, patterned exposure is performed to form a selective etching mask. Figure 7.1 (c) and (d) show two different etching conditions. (c) is a four-segment structure formed by etching the nanowire, which corresponds to the source semiconductor 1011, drain semiconductor 1012 of the first segment 101 and the source semiconductor 1021, drain semiconductor 1022 of the second segment 102; (d) is a three-segment structure formed by etching the nanowire, which corresponds to the source semiconductor 1011, drain semiconductor 1012 of the first segment 101 and the source semiconductor 1021, drain semiconductor 1022 of the second segment 102; wherein 1022 and 1012 are not separated in physical structure.
[0062] Figure 7.2-Figure 7.4 for Figure 7.1 (c) The first subsequent process.
[0063] Figure 7.2 For the nanowire reconnection process, Figure 7.2 As shown in (b), the nanowire etched into four sections is filled with photoresist 200 and cured. Then, a mask or oxide medium 201 is covered to form selective etching, and the excess photoresist is removed by dry or wet etching, as shown in 7.2 (c) and (d). At this time, the photoresist filled between 1022 and 1012 becomes an insulating isolation medium 103, as shown in 7.2 (e).
[0064] Figure 7.3 Schematic diagram of the formation of channel semiconductor and gate oxide layer. 7.3(a) shows that the surrounding columnar structure is grown on the outside of the nanowire structure (1021-1024-1022) and (1012-1014-1011) through solution self-assembly, epitaxial growth (including MOCVD, LPCVD, MBE, etc.) and other growth techniques to form channel semiconductors 1023 and 1013. As shown in 7.3(b) and (c), an oxide layer 201 is coated on the outside of the channel semiconductor 1023 and 1013 to form gate oxide layers 1025 and 1015, and then selective etching is performed to remove part of the channel semiconductor 1023 and 1013 to expose part of 1022 and 1012. Figure 7.3 (d) A layer of insulating medium 1031 is coated on the outside of the entire structure using ALD technology.
[0065] Figure 7.4 Schematic diagram of the window structure and gate metal formation. Figure 7.4 (a) and (b) are performed by using a mask or using oxide 201 to perform patterned etching on the outside of 1031 to remove excess 1031 and form an insulating medium 1031 with a window structure and an insulating medium partially attached to 103 . Figure 7.4(c) Removing the mask or the oxide 201 used. Figure 7.4 (d) The gate metal layers 1026 and 1016 covering the outer sides of the gate oxide layers 1025 and 1015 and the window conductive medium 1032 are formed by magnetron sputtering or thermal evaporation. Figure 7.4 (e) The manufacturing process of the present invention is completed by removing 1031 attached to the outside of 103 and removing the excess metal at the same time, forming Figure 4 A complementary stacked field effect transistor device is shown.
[0066] Figure 7.5 - Figure 7.7 for Figure 7.1 (c) The second subsequent process.
[0067] Figure 7.5 For the nanowire reconnection process, Figure 7.5 As shown in (b), the nanowire etched into four sections is filled with photoresist 200 and cured. Then, a mask or a dielectric 201 such as oxide is covered to form selective etching, and the excess photoresist is removed by dry or wet etching, as shown in 7.5 (c) and (d). At this time, a nanowire structure with semiconductors and insulating dielectrics arranged alternately is formed, as shown in 7.5 (e).
[0068] Figure 7.6 Schematic diagram of the formation of channel semiconductor and gate oxide layer. 7.6(a) shows that the channel semiconductors 1023 and 1013 are formed by growing a surrounding columnar structure on the outside of the nanowire structure (1021-1024-1022) and (1012-1014-1011) through solution self-assembly, epitaxial growth (including MOCVD, LPCVD, MBE, etc.), and are not separated in physical structure. As shown in 7.6(b) and (c), an oxide layer 201 is coated on the outside of the channel semiconductors 1023 and 1013 to form gate oxide layers 1025 and 1015, and then selective etching is performed to remove part of the channel semiconductors 1023 and 1013 to expose part of 1022 and 1012 and the intermediate insulating isolation layer. Figure 7.6 (d) A layer of insulating medium 1041 is coated on the outside of the entire structure using ALD technology.
[0069] Figure 7.7 The figure is a schematic diagram of the formation of the conductive layer and the gate metal. Figure 7.7 (a) and (b) are performed by using a mask or using oxide 201 to form a patterned etching on the outside of the insulating medium to remove excess insulating medium and form an isolation medium 1041 of a conductive layer. Figure 7.7 (c) The gate metal layers 1026 and 1016 and the conductive layer 104 are formed on the outside of the gate oxide layers 1025 and 1015 by magnetron sputtering or thermal evaporation. Figure 7.7(d) removing the mask or the oxide 201 used. The manufacturing process of the present invention is completed to form Figure 5 A complementary stacked field effect transistor device is shown.
[0070] Figure 7.8-Figure 7.10 for Figure 7.1 (d) The subsequent process.
[0071] Figure 7.8 For the nanowire reconnection process, Figure 7.8 As shown in (b), the nanowire etched into three sections is filled with photoresist 200 and cured. Then, a mask or a dielectric 201 such as oxide is covered to form selective etching, and the excess photoresist is removed by dry or wet etching, as shown in 7.8 (c) and (d). At this time, a nanowire structure with semiconductors and insulating dielectrics arranged alternately is formed, as shown in 7.8 (e).
[0072] Figure 7.9 Schematic diagram of the formation of channel semiconductor and gate oxide layer. 7.9(a) shows that the channel semiconductors 1023 and 1013 are formed by growing a surrounding columnar structure on the outside of the nanowire structure (1021-1024-1022) and (1012-1014-1011) through solution self-assembly, epitaxial growth (including MOCVD, LPCVD, MBE, etc.), and are not separated in physical structure. As shown in 7.9(b) and (c), an oxide layer is coated on the outside of the channel semiconductors 1023 and 1013 to form gate oxide layers 1025 and 1015, and then selective etching is performed to remove part of the channel semiconductors 1023 and 1013 to expose the nanowires connected to the middle part 1022 and 1012. Figure 7.9 (d) A layer of insulating medium 1041 is coated on the outside of the entire structure using ALD technology.
[0073] Figure 7.10 Schematic diagram of the formation of conductive medium and gate metal. Figure 7.10 (a) and (b) are performed by using a mask or using oxide 201 to form a patterned etching on the outside of 1041 to remove excess insulating medium, thereby forming insulating medium 1041 for subsequent isolation of 1026, 1016 and 104. Figure 7.10 (c) The gate metal layers 1026 and 1016 and the conductive medium 104 are formed on the outside of the gate oxide layers 1025 and 1015 by magnetron sputtering or thermal evaporation. Figure 7.10 (d) removing the mask or the oxide 201 used. The manufacturing process of the present invention is completed to form Figure 5 The complementary stacked field effect transistor device shown in FIG. Figure 5 The 104 conductive layers in the Figure 7.10(d) shows the nanowire regions of equal height of 104 and horizontal contact, that is, the semiconductor nanowire itself serves as the common drain contact of 1022 and 1012.
[0074] Based on the structural design of the present invention, those skilled in the art may further combine other process means to construct the field effect transistor of the present invention.
[0075] The above embodiments describe in detail the structure, features and effects of the present invention. The above are only preferred embodiments of the present invention. Any changes made in accordance with the concept of the present invention, or modifications to equivalent embodiments with equivalent changes, which do not exceed the scope covered by the specification, should be within the protection scope of the present invention.
Claims
1. A stacked complementary field effect transistor, characterized in that: The nanowire comprises at least a first line segment doped with an N-type and a second line segment doped with a P-type; the first line segment and the second line segment comprise a source semiconductor, a channel semiconductor, and a drain semiconductor, respectively constituting an NMOS and a PMOS; The source semiconductor and the drain semiconductor in the first line segment and the second line segment are distributed up and down.
2. The stacked complementary field effect transistor according to claim 1, characterized in that: The first line segment and the second line segment are formed by sequentially connecting a source semiconductor, a channel semiconductor, and a drain semiconductor; or, The first line segment and the second line segment both have a core-shell structure, including a trunk portion connected in sequence by a source semiconductor, an isolation medium, and a drain semiconductor, and a channel semiconductor coated outside the trunk portion.
3. The stacked complementary field effect transistor according to claim 1 or 2, characterized in that: An insulating isolation layer is provided between the first line segment and the second line segment.
4. The stacked complementary field effect transistor according to claim 3, characterized in that: The first line segment and the second line segment both have a core-shell structure, and a contact window is opened in the channel semiconductor for introducing electrodes to the first line segment and the second line segment wrapped inside; the contact window is located adjacent to the insulating isolation layer.
5. The stacked complementary field effect transistor according to claim 1 or 2, characterized in that: A conductive layer is provided between the first line segment and the second line segment, and the conductive layer serves as a common source or drain of the NMOS and the PMOS.
6. The stacked complementary field effect transistor according to claim 5, characterized in that: The first line segment and the second line segment both have a core-shell structure, and there is no electrical connection between the conductive layer and the channel semiconductor.
7. The stacked complementary field effect transistor according to claim 5, characterized in that: The conductive layer includes at least one of metal and heavily doped semiconductor.
8. A logic device, characterized in that: It is composed of the stacked complementary field effect transistor according to any one of claims 1 to 6; the NMOS and PMOS share a common gate and a common drain; the sources of the NMOS and PMOS are connected to the ground and the power supply respectively.
9. The logic device according to claim 7, characterized in that: A conductive layer is provided between the first line segment and the second line segment, and the conductive layer serves as a common drain of the NMOS and the PMOS.
10. The method for preparing a stacked complementary field effect transistor according to claim 1, characterized in that: At least: (1) regulating the doping type and doping concentration during the growth of the nanowires to form heterogeneous doped nanowires having N+, N-, N+, P+, P-, and P+ in sequence; (2) Etch at the junction of N+ and P+ and fill it with insulating isolation medium or metal medium. or, including: (1) regulating the doping type and doping concentration during the nanowire growth process to form heterogeneous doped nanowires with N+ and P+; (2) Etching in the N+ doped nanowire segment and filling it with an isolation dielectric; etching in the P+ doped nanowire segment and filling it with an isolation dielectric; etching at the junction of N+ and P+ and filling it with an insulating isolation dielectric or a metal dielectric; (3) Depositing an N-coating layer on the outside of the N+ doped nanowire segment to form an N-type channel semiconductor surrounding the columnar channel structure; depositing a P-coating layer on the outside of the P+ doped nanowire segment to form a P-type channel semiconductor surrounding the columnar channel structure.
