A field-effect transistor and its fabrication method
By setting a stepped region in the field-effect transistor and differentiating the gate dielectric layer and sidewall structure, the problem of inconsistent thickness requirements between the source and drain regions is solved, achieving a balance between low leakage current and high drive capability, thus improving the electrical performance of the transistor.
Patent Information
- Application Number
- CN202611431952.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-15
- Publication Date
- 2026-11-17
AI Technical Summary
Existing field-effect transistors cannot simultaneously meet the different requirements of the source and drain regions for the gate dielectric layer thickness, making it difficult to balance driving capability and static power consumption.
By setting a stepped region in the semiconductor substrate of the field-effect transistor, the gate dielectric layer and sidewall structure are set differently, so that the gate dielectric layer on the drain side is thicker and the sidewall with a lower dielectric constant is thicker, while the gate dielectric layer on the source side is thinner and the sidewall with a higher dielectric constant is thinner, thus achieving an asymmetric setting to balance driving capability and low leakage current performance.
It improves the overall electrical performance of the field-effect transistor, reduces leakage current, increases drive current, and enhances gate control capability.
Smart Images

Figure CN123124423A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of semiconductor design and manufacturing technology, and in particular to a field-effect transistor and its fabrication method. Background Technology
[0002] Field-effect transistors (FETs) are fundamental devices in complementary metal-oxide-semiconductor (CMOS) integrated circuits. As device size continues to shrink, the requirements for FETs in terms of driving capability, static power consumption, and device reliability are constantly increasing.
[0003] In existing field-effect transistors, the gate dielectric layer typically employs a nearly uniform thickness. Existing research indicates that reducing the gate dielectric layer thickness enhances the gate's control over the channel, improving carrier injection efficiency and device drive current. Conversely, increasing the gate dielectric layer thickness reduces tunneling leakage and weakens the electric field strength near the drain region, thereby suppressing phenomena such as gate-induced drain leakage (GIDL) and drain-induced barrier lowering (DIBL).
[0004] However, the requirements for the gate dielectric layer thickness differ between the source and drain regions of a field-effect transistor (FET): a thinner gate dielectric layer is typically desired on the source side to enhance gate control and carrier injection efficiency, while a thicker gate dielectric layer is preferred on the drain side to reduce the drain electric field and lower leakage current. Existing FETs with uniformly thick gate dielectric layers struggle to simultaneously meet these different requirements, thus making it difficult to achieve both high drive capability and low static power consumption. Summary of the Invention
[0005] This invention provides a field-effect transistor and its fabrication method. By differentiating the gate dielectric layer thickness and sidewall structure on the drain and source sides, the driving capability and low leakage current performance of the field-effect transistor can be balanced, thereby improving the overall electrical performance of the device.
[0006] A first aspect of the present invention provides a field-effect transistor, the field-effect transistor comprising: A semiconductor substrate, including source and drain regions spaced apart; The gate structure is located on one side of the semiconductor substrate and between the source and drain regions; The semiconductor substrate includes at least two stepped regions, including a first stepped region near the drain region and a second stepped region near the source region, and the top surface height of the semiconductor substrate in the first stepped region is lower than the top surface height of the semiconductor substrate in the second stepped region. The gate structure includes a stacked gate dielectric layer and a gate electrode, with the gate electrode located on the side of the gate dielectric layer facing away from the semiconductor substrate; the thickness of the gate dielectric layer covering the first step region is greater than the thickness of the gate dielectric layer covering the second step region; A first sidewall is provided on the side of the gate structure near the drain region, and a second sidewall is provided on the side of the gate structure near the source region; the thickness of the first sidewall is greater than the thickness of the second sidewall in the direction from the source region to the drain region; the dielectric constant of the first sidewall is less than the dielectric constant of the second sidewall.
[0007] Optionally, the semiconductor substrate further includes a third step region located between the first step region and the second step region; the top surface height of the semiconductor substrate in the third step region is higher than the top surface height of the semiconductor substrate in the first step region, but lower than the top surface height of the semiconductor substrate in the second step region. The thickness of the gate dielectric layer covering the third step region is less than the thickness of the gate dielectric layer covering the first step region, but greater than the thickness of the gate dielectric layer covering the second step region.
[0008] Optionally, the gate dielectric layer includes a first dielectric layer and a second dielectric layer stacked together, the second dielectric layer being located on the side of the first dielectric layer away from the semiconductor substrate; the first dielectric layer and the second dielectric layer may be made of the same or different materials; the thickness of the second dielectric layer may increase or remain constant from the source region to the drain region.
[0009] Optionally, the first and second dielectric layers corresponding to the same step region have the same thickness.
[0010] Optionally, along the direction from the drain region to the source region, the gate electrode includes a first gate region, a transition gate region, and a second gate region disposed sequentially. Along the thickness direction of the semiconductor substrate, the projection of the first gate region overlaps at least partially with the projection of the first step region, the projection of the second gate region overlaps at least partially with the projection of the second step region, and the projection of the transition gate region overlaps at least partially with the projection of the third step region. The work function of the first gate region is greater than the work function of the transition gate region, and the work function of the transition gate region is greater than the work function of the second gate region.
[0011] Optionally, the semiconductor substrate may further include a first doped region, a transition doped region, and a second doped region, which are sequentially disposed and have the same doping type as the semiconductor substrate. The first doped region is disposed near the drain region and is at least partially located within the first step region; the second doped region is disposed near the source region and is at least partially located within the second step region; the transition doped region is located between the first doped region and the second doped region and is at least partially located within the third step region. The doping concentration of the first doped region is greater than that of the transition doped region, and the doping concentration of the transition doped region is greater than that of the second doped region.
[0012] Optionally, the field-effect transistor further includes: a first sidewall liner layer and a second sidewall liner layer; the first sidewall liner layer is located between the first sidewall and the gate structure, and extends from the sidewall of the gate structure near the drain region away from the source region to the surface of the semiconductor substrate; the second sidewall liner layer is located between the second sidewall and the gate structure, and extends from the sidewall of the gate structure near the source region away from the drain region to the surface of the semiconductor substrate.
[0013] Optionally, the semiconductor substrate further includes a first lightly doped extension region and a second lightly doped extension region; the first lightly doped extension region is located near the drain region, and the second lightly doped extension region is located near the source region; the doping type of the first lightly doped extension region is the same as that of the drain region and the doping concentration is lower than that of the drain region, and the doping type of the second lightly doped extension region is the same as that of the source region and the doping concentration is lower than that of the source region.
[0014] Optionally, the field-effect transistor further includes: an interlayer dielectric layer covering the semiconductor substrate and the gate structure, a metal plug penetrating the interlayer dielectric layer, and a metal wiring layer covering the interlayer dielectric layer; The metal plugs include a source metal plug that is electrically in contact with the source region and a drain metal plug that is electrically in contact with the drain region; along a direction parallel to the plane where the semiconductor substrate is located, the shortest distance between the bottom edge of the source metal plug and the bottom edge of the second sidewall is less than or equal to the shortest distance between the bottom edge of the drain metal plug and the bottom edge of the first sidewall.
[0015] A second aspect of the present invention provides a method for fabricating a field-effect transistor, the method comprising: A semiconductor substrate is provided, and the semiconductor substrate is patterned to form a stepped region on the surface of the semiconductor substrate, including at least a first stepped region and a second stepped region; wherein the semiconductor substrate includes a source end and a drain end disposed at intervals, the first stepped region is close to the drain end, the second stepped region is close to the source end, and the top surface height of the semiconductor substrate in the first stepped region is lower than the top surface height of the semiconductor substrate in the second stepped region. A gate dielectric layer is formed on one side of a semiconductor substrate, and a gate electrode is formed on the side of the gate dielectric layer opposite to the semiconductor substrate to form a gate structure; wherein the thickness of the gate dielectric layer covering the first step region is greater than the thickness of the gate dielectric layer covering the second step region. A first sidewall is formed on the side of the gate structure near the drain end, and a second sidewall is formed on the side of the gate structure near the source end; wherein, in the direction from the source end to the drain end, the thickness of the first sidewall is greater than the thickness of the second sidewall; and the dielectric constant of the first sidewall is less than the dielectric constant of the second sidewall. The semiconductor substrate is doped with source and drain doping to form a source region at the source end and a drain region at the drain end; An interlayer dielectric layer covering the semiconductor substrate and gate structure, a metal plug penetrating the interlayer dielectric layer, and a metal wiring layer covering the interlayer dielectric layer are formed.
[0016] Optionally, the above preparation method further includes: before forming the first sidewall and the second sidewall, performing light doping extension region implantation on the semiconductor substrate to form a first light doping extension region on the side near the drain end and a second light doping extension region on the side near the source end, wherein the doping type of the first light doping extension region is consistent with that of the drain region and the doping concentration is lower than that of the drain region, and the doping type of the second light doping extension region is consistent with that of the source region and the doping concentration is lower than that of the source region.
[0017] Optionally, the above preparation method further includes: performing partitioned doping implantation on the semiconductor substrate to form a first doped region on the side near the drain end and a second doped region on the side near the source end, wherein the doping type of the first doped region and the second doped region is consistent with the doping type of the semiconductor substrate, and the doping concentration of the first doped region is greater than the doping concentration of the second doped region.
[0018] The technical solution of this invention involves forming a first step region near the drain region and a second step region near the source region in a semiconductor substrate. The top surface height of the semiconductor substrate in the first step region is lower than that in the second step region. This allows the thickness of the gate dielectric layer covering the first step region to be greater than that covering the second step region. This thicker gate dielectric layer reduces the electric field coupling between the gate electrode and the drain-side channel region, while a thinner gate dielectric layer enhances the gate electrode's control over the source-side channel region. Furthermore, by setting the thickness of the first sidewall on the drain side of the gate structure to be greater than the thickness of the second sidewall on the source side, and setting the dielectric constant of the first sidewall to be less than that of the second sidewall, the parasitic capacitance between the gate electrode and the drain region is reduced using the first sidewall, and the control over the source-side channel region is improved using the second sidewall. Thus, by asymmetrically setting the thickness of the gate dielectric layer, the thickness of the first sidewall and the second sidewall, and the dielectric constant on the drain and source sides, it is beneficial to balance the low leakage current performance and driving capability of the field-effect transistor, thereby improving the overall electrical performance of the field-effect transistor.
[0019] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description
[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0021] Figure 1 This is a schematic diagram of the structure of a field-effect transistor provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of another field-effect transistor provided in an embodiment of the present invention; Figure 3 This is a flowchart of a method for fabricating a field-effect transistor according to an embodiment of the present invention; Figure 4 This is a process flow diagram of a field-effect transistor fabrication process provided in an embodiment of the present invention; Figure 5 The double-layer gate dielectric thickness and equivalent gate capacitance distribution diagram of the field-effect transistor provided in the embodiments of the present invention; Figure 6 The diagram shows the relationship between the total gate dielectric thickness, threshold voltage, and conduction current of the field-effect transistor provided in the embodiments of the present invention. Detailed Implementation
[0022] To enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.
[0023] It should be noted that the terms "first," "second," etc., used in the specification and claims of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein.
[0024] Figure 1 This is a schematic diagram of a field-effect transistor provided in an embodiment of the present invention. Figure 1As shown, the field-effect transistor includes: a semiconductor substrate 1 having a first surface and a second surface disposed opposite to each other, and a source region 11 and a drain region 12 disposed at intervals in the semiconductor substrate 1; a gate structure 2 located on one side of the first surface of the semiconductor substrate 1 and between the source region 11 and the drain region 12; wherein, the first surface of the semiconductor substrate 1 includes at least two step regions 00, the step regions 00 including a first step region 01 near the drain region 12 and a second step region 02 near the source region 11, and the top surface height of the semiconductor substrate 1 in the first step region 01 is lower than the top surface height of the semiconductor substrate 1 in the second step region 02; that is, in the direction from the drain region 12 to the source region 11, The first surface of the semiconductor substrate 1 (corresponding to the gate structure portion) has a "step-up" trend; the gate structure 2 includes a stacked gate dielectric layer 21 and a gate electrode 22, with the gate electrode 22 located on the side of the gate dielectric layer 21 away from the semiconductor substrate 1; the thickness of the gate dielectric layer 21 covering the first step region 01 is greater than the thickness of the gate dielectric layer 21 covering the second step region 02; a first sidewall 31 is provided on the side of the gate structure 2 near the drain region 12, and a second sidewall 32 is provided on the side of the gate structure 2 near the source region 11; from the source region 11 to the drain region 12, the thickness of the first sidewall 31 is greater than the thickness of the second sidewall 32; the dielectric constant of the first sidewall 31 is less than the dielectric constant of the second sidewall 32.
[0025] The semiconductor substrate 1 can be a single-crystal silicon substrate, an SOI (Silicon-On-Insulator) substrate, or an epitaxial wafer, or other semiconductor substrates suitable for forming field-effect transistors.
[0026] Source region 11 and drain region 12 are formed alternately in semiconductor substrate 1. The semiconductor substrate region between source region 11 and drain region 12 is used to form a conductive channel under the control of gate structure 2. Gate structure 2 is disposed between source region 11 and drain region 12 and is used to adjust the carrier concentration in the channel region by applying a gate voltage, thereby controlling the conduction or cutoff between source region 11 and drain region 12. Gate dielectric layer 21 is located between gate electrode 22 and semiconductor substrate 1, and is used to achieve electrical insulation between gate electrode 22 and semiconductor substrate 1, and can realize electrical control of the channel region by gate electrode 22 through gate capacitive coupling. The material of gate dielectric layer 21 may include silicon oxide, silicon oxynitride, or other high-k dielectric materials; gate electrode 22 may be made of polysilicon, TiN, TaN / TiN stack, or other conductive materials suitable for forming a metal gate. For example, the gate dielectric material can be formed by film deposition processes such as chemical vapor deposition (CVD), plasma-enhanced chemical vapor deposition (PECVD), and atomic layer deposition (ALD), and the gate dielectric layer 21 can be formed by combining planarization and selective etching processes as needed; the gate electrode 22 can be formed by film deposition processes such as physical vapor deposition (PVD) and atomic layer deposition and patterning processing. The embodiments of the present invention do not specifically limit this.
[0027] Specifically, the semiconductor substrate 1 includes at least two stepped regions 00, namely at least a first stepped region 01 near the drain region 12 and a second stepped region 02 near the source region 11, and the top surface height of the semiconductor substrate 1 in the first stepped region 01 is lower than the top surface height of the semiconductor substrate 1 in the second stepped region 02. It should be noted that, in other embodiments, one or more transition stepped regions with intermediate top surface heights may also be provided between the first stepped region 01 and the second stepped region 02. By making the semiconductor substrate 1 have different top surface heights on the drain region 12 side and the source region 11 side, and by adjusting the thickness in conjunction with the deposition, planarization, etching, etc., of the gate dielectric layer 21, the thickness of the gate dielectric layer 21 covering the first stepped region 01 can be greater than the thickness of the gate dielectric layer 21 covering the second stepped region 02, thereby forming a non-uniform gate dielectric structure with different thicknesses at least on the drain region 12 side and the source region 11 side.
[0028] Furthermore, the gate dielectric layer 21 covering the first step region 01 is relatively thick, meaning the gate dielectric layer 21 near the drain region 12 has a relatively large thickness. The thicker gate dielectric layer 21 can weaken the electric field coupling between the gate electrode 22 and the channel region on the drain region 12 side, reduce the local electric field intensity near the drain region 12, thereby helping to weaken the drain-end electric field peak, reduce gate-induced drain-end leakage current, and suppress phenomena such as drain-induced barrier reduction, impact ionization, and hot carrier effects, which is beneficial for reducing the off-state leakage current and static power consumption of the field-effect transistor. Correspondingly, the gate dielectric layer 21 covering the second step region 02 is relatively thin, meaning the gate dielectric layer 21 near the source region 11 has a relatively small thickness. The thinner gate dielectric layer 21 can increase the equivalent gate capacitance on the source region 11 side, enhance the control capability of the gate electrode 22 over the channel region near the source region 11, and make it easier for carriers to be injected into the channel from the source region 11, thereby improving the drive current and switching performance of the field-effect transistor. Therefore, the gate dielectric layer 21 can be configured differently according to the different electrical requirements of the source region 11 and the drain region 12, thereby reducing the mutual constraint between improving the driving capability of the source region and reducing the leakage performance of the drain region when using a gate dielectric layer of uniform thickness.
[0029] The first sidewall 31 and the second sidewall 32 are respectively disposed on both sides of the gate structure 2. The sidewalls can protect and insulate the sidewalls of the gate structure 2, and can also serve as self-aligned masks for doping implantation during subsequent source and drain doping processes. For example, after forming the first sidewall 31 and the second sidewall 32, the semiconductor substrate 1 can be heavily doped with source and drain using the gate structure 2, the first sidewall 31, and the second sidewall 32 as masks, and the implanted dopant elements can be activated by rapid thermal annealing (RTA) to form the source region 11 and the drain region 12. For N-type metal-oxide-semiconductor field-effect transistors, arsenic or phosphorus can be used as source and drain doping elements; for other types of field-effect transistors, other suitable doping elements can also be selected accordingly.
[0030] The first sidewall 31 and the second sidewall 32 are asymmetrically arranged. In the direction from the source region 11 to the drain region 12, or from the drain region 12 to the source region 11 (i.e., laterally in the figure), the thickness of the first sidewall 31, closer to the drain region 12, is greater than the thickness of the second sidewall 32, closer to the source region 11. The thicker first sidewall 31 increases the effective spacing between the gate structure 2 and the drain region 12, which helps to weaken the local electric field near the drain region 12 and reduce the parasitic capacitance between the gate and drain. The thinner second sidewall 32 reduces the spacing between the source region 11 and the gate structure 2, lowering the parasitic series resistance on the source region 11 side, thus facilitating carrier injection from the source region 11 into the channel region. Therefore, the asymmetrical arrangement of the thickness of the first sidewall 31 and the second sidewall 32, combined with the thicker gate dielectric layer 21 on the drain region 12 side and the thinner gate dielectric layer on the source region 11 side, allows for differentiated adjustment of the electrical performance on both the source and drain sides of the field-effect transistor.
[0031] The first sidewall 31 includes a first dielectric material, and the second sidewall 32 includes a second dielectric material, wherein the dielectric constant of the first dielectric material is preferably less than the dielectric constant of the second dielectric material. For example, the first dielectric material may include SiOC, SiOCN, or other low-k dielectric materials, or silicon dioxide, and the second dielectric material may include HfO2, Al2O3, or other high-k dielectric materials; or both the first and second dielectric materials may be high-k dielectric materials, and the dielectric constant of the first dielectric material is less than the dielectric constant of the second dielectric material.
[0032] Understandably, the use of a dielectric material with a relatively low dielectric constant for the first sidewall 31 near the drain region 12 can further reduce the parasitic capacitance between the gate electrode 22 and the drain region 12, reduce the Miller effect and the gate delay caused by the parasitic capacitance, and thus improve the high-frequency switching performance of the field-effect transistor. The use of a dielectric material with a higher dielectric constant for the second sidewall 32 near the source region 11 can enhance the coupling between the gate edge electric field and the channel region on the source region 11 side, improve the control capability of the gate electrode 22 over the channel region near the source region 11, and also help reduce the equivalent series resistance at the source end and improve the carrier injection efficiency.
[0033] By simultaneously setting the thickness of the gate dielectric layer 21 and the thickness and dielectric constant of the first sidewall 31 and the second sidewall 32 asymmetrically on the source region 11 side and the drain region 12 side, the drain region 12 side mainly reduces the drain electric field, leakage current and gate-drain parasitic capacitance through the thicker gate dielectric layer 21 and the thicker first sidewall 31 with a lower dielectric constant, while the source region 11 side mainly enhances the gate control capability and carrier injection efficiency through the thinner gate dielectric layer 21 and the thinner second sidewall 32 with a higher dielectric constant. This is beneficial to balance the low leakage current performance and driving capability of the field-effect transistor, and improve the overall electrical performance of the field-effect transistor.
[0034] In this embodiment, by providing a first step region 01 near the drain region 12 and a second step region 02 near the source region 11 in the semiconductor substrate 1, and by setting the top surface height of the semiconductor substrate 1 in the first step region 01 to be lower than the top surface height of the semiconductor substrate 1 in the second step region 02, the thickness of the gate dielectric layer 21 covering the first step region 01 in the gate structure 2 is made greater than the thickness of the gate dielectric layer 21 covering the second step region 02. This allows the thicker gate dielectric layer 21 to reduce the electric field coupling between the gate electrode 22 and the channel region on the drain region 12 side, and the thinner gate dielectric layer 21 to enhance the control capability of the gate electrode 22 over the channel region on the source region 11 side. Furthermore, by setting the thickness of the first sidewall 31 on the side of the gate structure 2 near the drain region 12 to be greater than the thickness of the second sidewall 32 on the side of the gate structure 2 near the source region 11, and setting the dielectric constant of the first sidewall 31 to be less than the dielectric constant of the second sidewall 32, the parasitic capacitance between the gate electrode 22 and the drain region 12 can be reduced by using the first sidewall 31, and the control capability of the gate electrode 22 over the channel region on the source region 11 side can be improved by using the second sidewall 32. Thus, by asymmetrically setting the thickness of the gate dielectric layer 21 and the thickness and dielectric constant of the first sidewall 31 and the second sidewall 32 on the drain region 12 side and the source region 11 side, it is beneficial to balance the low leakage current performance and driving capability of the field-effect transistor, thereby improving the overall electrical performance of the field-effect transistor.
[0035] Figure 2 This is a schematic diagram of another field-effect transistor provided in an embodiment of the present invention. Figure 2 As shown, the semiconductor substrate 1 also includes a third step region 03 located between the first step region 01 and the second step region 02; the height of the top surface of the semiconductor substrate 1 in the third step region 03 is higher than the height of the top surface of the semiconductor substrate 1 in the first step region 01, but lower than the height of the top surface of the semiconductor substrate 1 in the second step region 02; the thickness of the gate dielectric layer 21 covering the third step region 03 is less than the thickness of the gate dielectric layer 21 covering the first step region 01, but greater than the thickness of the gate dielectric layer 21 covering the second step region 02.
[0036] Specifically, the third step region 03 can serve as a transition region between the first step region 01 and the second step region 02. For example, the first step region 01, the third step region 03, and the second step region 02 with different top surface heights can be formed on the surface of the semiconductor substrate 1 through multiple photolithography and etching processes. The top surface height of the semiconductor substrate 1 in the third step region 03 is between the top surface height of the first step region 01 and the top surface height of the second step region 02, so that the top surface height of the semiconductor substrate 1 gradually increases from the drain region 12 side to the source region 11 side, exhibiting a "step-by-step" morphology. The total height difference between the first step region 01 and the second step region 02 can be set according to the device size and the target thickness of the gate dielectric layer 21, for example, it can be 30 angstroms to 50 angstroms; this embodiment of the invention does not specifically limit this.
[0037] Furthermore, the thickness of the gate dielectric layer 21 covering the third step region 03 is between the thickness of the gate dielectric layer 21 covering the first step region 01 and the thickness of the gate dielectric layer 21 covering the second step region 02. Thus, the gate dielectric layer 21 can gradually transition from a thinner thickness near the source region 11, through an intermediate thickness corresponding to the third step region 03, to a thicker thickness near the drain region 12, thereby forming a multi-level thickness variation structure in the direction from the source region 11 to the drain region 12. By setting the third step region 03, the abrupt change in the thickness of the gate dielectric layer 21 from a smaller thickness on the source region 11 side to a larger thickness on the drain region 12 side can be avoided. This allows the electric field coupling effect of the gate structure 2 on the channel region to change gradually along the direction from the source region 11 to the drain region 12, which helps to reduce abrupt changes in gate capacitance, surface potential, and electric field strength between different regions, making the potential distribution and carrier transport process in the channel smoother.
[0038] Optional, continue to refer to Figure 2 The gate dielectric layer 21 includes a first dielectric layer 211 and a second dielectric layer 212 stacked together. The second dielectric layer 212 is located on the side of the first dielectric layer 211 facing away from the semiconductor substrate 1. The bottom surface morphology of the first dielectric layer 211 is adapted to the morphology of the step region 00 of the semiconductor substrate 1, and the top surface of the first dielectric layer 211 can be planar. The bottom surface of the second dielectric layer 212 is planar. The thickness of the second dielectric layer 212 can increase from the source region 11 to the drain region 12. The first dielectric layer 211 and the second dielectric layer 212 may be made of the same or different materials. In this embodiment, the two layers are made of different materials. In other embodiments, both the bottom and top surfaces of the second dielectric layer 212 are planar.
[0039] Specifically, a first dielectric layer 211 can be formed on a semiconductor substrate 1 having a stepped region 00. For example, a first dielectric material, such as silicon oxide, can be deposited on the surface of the semiconductor substrate 1 using CVD or PECVD. Subsequently, chemical mechanical polishing (CMP) can be used to planarize and pattern the first dielectric material to form the first dielectric layer 211. Since the first dielectric layer 211 is formed directly on the stepped semiconductor substrate 1, the bottom surface of the first dielectric layer 211 matches the morphology of the stepped region 00 and is stepped, while the top surface of the first dielectric layer 211 after planarization is planar. By controlling the deposition thickness and the amount of CMP polishing, the first dielectric layer 211 can have different thicknesses near the drain region 12, in the intermediate transition region, and near the source region 11. For example, the thicknesses at these three locations can be approximately 70 angstroms, 50 angstroms, and 30 angstroms, respectively. In addition, rapid thermal annealing can be performed before or after planarization to improve the density of the first dielectric layer 211.
[0040] After forming the first dielectric layer 211, a second dielectric layer 212 can be formed on the top surface of the first dielectric layer 211. The second dielectric layer 212 can be made of a different material than the first dielectric layer 211; for example, the second dielectric layer 212 can be made of a high-k dielectric material such as HfO2. In an exemplary embodiment, the second dielectric material can be deposited integrally on the first dielectric layer 211 first, and then the second dielectric layer 212 of different thicknesses can be formed by regional etching. For example, during the first photolithography, a patterned photoresist is formed on the surface of the second dielectric material, so that the photoresist covers the area that needs to retain a larger thickness and exposes other areas; then, inductively coupled plasma reactive ion etching (ICP-RIE) is used to etch the exposed second dielectric material. The gas system used for etching can include BCl3 / Cl2 / Ar or CHF3 / Ar to reduce the thickness of the corresponding areas. After that, a second coating, exposure, and development are performed to further expose the area that needs to form the minimum thickness, and the area is etched again, so that the second dielectric layer 212 forms a gradually increasing thickness distribution from the source region 11 to the drain region 12. After etching, real-time annealing (RTA) can be performed under a nitrogen atmosphere at a temperature of 600℃-800℃ for 30-60 seconds to repair etching damage and improve the density of the high-k dielectric film. Alternatively, a second dielectric layer 212 with different thicknesses can be directly formed using a regional selective deposition method.
[0041] Since the second dielectric layer 212 is formed on the flat top surface of the first dielectric layer 211, the bottom surface of the second dielectric layer 212 is planar. Through the aforementioned regional etching or selective deposition, the top surface of the second dielectric layer 212 can be made into a stepped shape. Thus, the first dielectric layer 211 and the second dielectric layer 212 form a stepped morphology relative to each other with the planar interface between them as a reference, that is, the bottom surface of the first dielectric layer 211 is stepped and the top surface is planar, and the bottom surface of the second dielectric layer 212 is planar and the top surface is stepped. Through the stacked arrangement of the first dielectric layer 211 and the second dielectric layer 212, a gate dielectric layer 21 that is thicker near the drain region 12 and thinner near the source region 11 can be further formed. This makes the gate dielectric layer 21 on the drain region 12 side beneficial for weakening electric field spikes and suppressing leakage current, while the gate dielectric layer 21 on the source region 11 side beneficial for enhancing gate control capability and carrier injection efficiency.
[0042] Optional, continue to refer to Figure 2 The first dielectric layer 211 and the second dielectric layer 212, corresponding to the same step region, have the same thickness.
[0043] Specifically, the first dielectric layer 211 and the second dielectric layer 212 corresponding to the first step region 01 have the same thickness, the first dielectric layer 211 and the second dielectric layer 212 corresponding to the third step region 03 have the same thickness, and the first dielectric layer 211 and the second dielectric layer 212 corresponding to the second step region 02 have the same thickness. That is, in terms of shape, the first dielectric layer 211 and the second dielectric layer 212 are mirror images of each other along their interface. As a result, the first dielectric layer 211 and the second dielectric layer 212 can form corresponding stepped thickness changes, and the gate dielectric layer 21 can form a regular multi-level thickness gradient between the drain region 12, the intermediate transition region, and the source region 11. This is beneficial for the gradual change of the electric field coupling effect of the gate structure 2 on the channel, and for reducing the abrupt change in electric field between adjacent regions.
[0044] It is also understood that the gate electrode 22 can be formed on the stepped top surface of the second dielectric layer 212, and the top surface of the gate electrode 22 can be made into a plane through planarization. At this time, the bottom surface of the gate electrode 22 is adapted to the top surface of the second dielectric layer 212 and is stepped, so that the gate electrode 22 forms a thickness variation trend opposite to that of the lower gate dielectric layer 21, that is, the gate electrode 22 near the drain region 12 is thinner, and the gate electrode 22 near the source region 11 is thicker. This structure can adapt to the lower multi-step gate dielectric layer 21 while keeping the top surface of the gate structure 2 flat, which facilitates the formation of subsequent sidewalls and interlayer dielectric structures.
[0045] Optional, continue to refer to Figure 2Along the direction from the drain region 12 to the source region 11, the gate electrode 22 includes a first gate region 221, a transition gate region 222, and a second gate region 223 arranged sequentially. Along the stacking direction of the gate dielectric layer 21 and the gate electrode 22, or along the thickness direction of the semiconductor substrate, the projections of the first gate region 221 and the first step region 01 on the second surface of the semiconductor substrate at least partially overlap, the projections of the second gate region 223 and the second step region 02 on the second surface of the semiconductor substrate at least partially overlap, and the projections of the transition gate region 222 and the third step region 03 on the second surface of the semiconductor substrate at least partially overlap. Furthermore, the orthographic projection of the first gate region 221 on the second surface of the semiconductor substrate 1 lies within the orthographic projection of the first step region 01 on the second surface of the semiconductor substrate 1, and both are flush with the boundary of the second step region 02; the orthographic projection of the transition gate region 222 on the second surface of the semiconductor substrate 1 lies within the orthographic projection of the third step region 03 on the second surface of the semiconductor substrate 1, and the boundaries on both sides of both are flush with each other; the orthographic projection of the second gate region 223 on the second surface of the semiconductor substrate 1 lies within the orthographic projection of the second step region 02 on the second surface of the semiconductor substrate 1, and both are flush with the boundary of the first step region 01.
[0046] Furthermore, the work function of the first gate region 221 is greater than the work function of the transition gate region 222, and the work function of the transition gate region 222 is greater than the work function of the second gate region 223.
[0047] Specifically, the first gate region 221, the transition gate region 222, and the second gate region 223 can be respectively positioned on the drain region 12 side, the intermediate transition region, and the source region 11 side. The work function of the first gate region 221 is greater than that of the transition gate region 222, and the work function of the transition gate region 222 is greater than that of the second gate region 223. This results in a progressively decreasing work function gradient along the direction from the drain region 12 to the source region 11 on the gate electrode 22. The difference in work function between the different gate regions can be achieved by differentially doping each gate region, for example, by adjusting the type and / or concentration of the doping element to regulate the work function of the corresponding region.
[0048] For example, the doping concentrations of the first gate region 221, the transition gate region 222, and the second gate region 223 can be set in a ratio of approximately 4:2:1. This 4:2:1 doping concentration ratio can be a preferred ratio determined when the height difference between the first step region 01 and the second step region 02 is 30 to 50 angstroms. In other embodiments, the doping concentration of each gate region can also be adjusted according to the step height, the thickness of the gate dielectric layer 21, and the target electrical performance of the field-effect transistor.
[0049] The gate electrode 22 can be made of TiN or TaN / TiN stacked metal material. The TaN / TiN stacked structure improves the process compatibility of the gate structure with high-k gate dielectric materials and helps to block metal atom diffusion and reduce leakage current at the high-k dielectric interface. For example, the TaN / TiN stacked metal gate material can be formed on the gate dielectric layer 21 using PVD or ALD, and the work function of different gate regions can be adjusted by introducing elements such as N, W, and La in different regions. After the gate electrode material is formed, the gate electrode 22 can be planarized using CMP, forming a stepped structure with a thinner thickness near the drain region 12, a moderate thickness in the middle region, and a thicker thickness near the source region 11, while keeping the top surface of the gate electrode 22 basically flat. By implementing differentiated doping in regions of different thicknesses, the electrical control capability of the thinner gate electrode region can be compensated by a higher doping concentration, and the thicker gate electrode region can be matched with a lower doping concentration.
[0050] For example, in the first gate region 221 near the drain region 12, N and W can be used for modulation. Since W atoms have a large atomic mass, it is difficult to achieve uniform doping through conventional ion implantation. Therefore, W can be co-deposited during the TaN / TiN deposition process, and shallow N implantation can be performed after deposition. The N atom percentage can be approximately 14 at.%, and the peak doping concentration can be approximately 8 × 10⁻⁶. 21 cm -3 This improves the work function of the first gate region 221, thereby enhancing the surface barrier on the drain region 12 side, compensating for the insufficient electric field shielding ability caused by the small thickness of the gate electrode 22 in this region, and helping to suppress punch-through leakage.
[0051] In the transition gate region 222, the gate electrode 22 can be implanted with a low dose of N ions. For example, the N atom percentage can be approximately 5 at.%, and the peak doping concentration can be approximately 4 × 10⁻⁶. 21 cm -3 This makes the work function of the transition gate region 222 between the first gate region 221 and the second gate region 223, thereby forming a transition between the gate electrical characteristics on the drain region 12 side and the source region 11 side, reducing potential jumps and local lateral electric field jumps along the channel direction.
[0052] In the second gate region 223 near the source region 11, shallow La implantation can be performed on the gate electrode 22. To reduce the lateral scattering of La ions towards the gate sidewalls, the implantation tilt angle can be set to 0°. For example, the proportion of La atoms can be approximately 2.5 at.%, and the peak doping concentration can be approximately 2 × 10⁻⁶. 21 cm -3By forming a metal gate region with a low work function in the second gate region 223, the carrier injection barrier on the source region 11 side can be reduced. Combined with the thicker gate electrode 22 in this region, the carrier injection efficiency from the source region 11 to the channel region can be improved. After doping of each gate region, a low-temperature rapid thermal annealing at approximately 420°C can be performed to improve the performance of the doped gate electrode material.
[0053] Understandably, by forming a partitioned structure with a gradually decreasing doping concentration from high to low along the direction from drain region 12 to source region 11, the gate electrode 221 has a high work function, which helps to increase the surface barrier on the drain region 12 side and suppress punch-through leakage. The work function of the transition gate region 222 is between the two sides, which helps to make the potential change in the channel smoother. The second gate region 223 has a low work function, which helps to reduce the carrier injection barrier on the source region 11 side. Thus, while suppressing the high electric field and leakage at the drain end, the carrier injection capability at the source end is improved, and the overall gate control uniformity of the gate structure 2 is improved.
[0054] Optional, continue to refer to Figure 2 The semiconductor substrate 1 further includes a first doped region 41, a transition doped region 42, and a second doped region 43 arranged sequentially. The doping types of the first doped region 41, the transition doped region 42, and the second doped region 43 are consistent with the doping type of the semiconductor substrate. The first doped region 41 is located near the drain region 12 and is at least partially located within the projection range of the first step region 01. The second doped region 43 is located near the source region 11 and is at least partially located within the second step region 02. The transition doped region 42 is located between the first doped region 41 and the second doped region 43 and is at least partially located within the third step region 03. The doping concentration of the first doped region 41 is greater than the doping concentration of the transition doped region 42, and the doping concentration of the transition doped region 42 is greater than the doping concentration of the second doped region 43.
[0055] The semiconductor substrate 1 also includes lightly doped extension regions, specifically a first lightly doped extension region 51 and a second lightly doped extension region 52; the first lightly doped extension region 51 is located near the drain region 12, and the second lightly doped extension region 52 is located near the source region 11; the doping type of the first lightly doped extension region 51 is the same as that of the drain region 12 and the doping concentration is lower than that of the drain region 12, and the doping type of the second lightly doped extension region 52 is the same as that of the source region 11 and the doping concentration is lower than that of the source region 11.
[0056] Specifically, the first lightly doped extension region 51 and the second lightly doped extension region 52 can be formed after the gate structure 2 is formed by implantation via a lightly doped drain (LDD). The LDD implantation energy can be 5keV-10keV, and the implantation dose can be 1×10⁻⁶. 14 cm -2-5×10 14 cm -2 The doping element can be selected according to the conductivity type of the field-effect transistor. The first lightly doped extension region 51 and the second lightly doped extension region 52 can form a shallow junction extension region with a low doping concentration between the subsequently formed source region 11, drain region 12 and channel region, so that the doping concentration gradually transitions from the source / drain region to the channel region, thereby reducing the peak electric field at the channel edge and helping to suppress hot carrier injection.
[0057] Furthermore, after forming the first lightly doped extension region 51 and the second lightly doped extension region 52, the semiconductor substrate 1 can be subjected to regional doping implantation to form the first doped region 41, the transition doped region 42, and the second doped region 43. For example, a partitioned mask can be formed using photoresist, taking into account the positions of the first step region 01, the third step region 03, and the second step region 02, and ion implantation can be performed on the corresponding regions using different implantation doses. The ion implantation energy can be 15keV-30keV, wherein the first doped region 41, which is closer to the drain region 12, uses a larger implantation dose, and its doping concentration can be 4×10⁻⁶. 18 cm -3 -6×10 18 cm -3 The transition doped region 42 uses a moderate implantation dose, and its doping concentration can be approximately 2 × 10⁻⁶. 18 cm -3 The second doped region 43, located near the source region 11, employs a smaller implantation dose, with a doping concentration of approximately 1 × 10⁻⁶. 18 cm -3 This creates a doping concentration gradient that gradually decreases along the direction from the drain region 12 to the source region 11 among the first doped region 41, the transition doped region 42, and the second doped region 43. The doping concentration ratio of the three regions can be approximately 4:2:1.
[0058] It is understood that the doping concentration gradients of the first doped region 41, the transition doped region 42, and the second doped region 43 can be matched with the doping concentration gradients of the first gate region 221, the transition gate region 222, and the second gate region 223. Thus, the doping concentration distribution in the semiconductor substrate 1 can be matched with the partitioned electrical characteristics of the gate electrode 22. A higher doping concentration is used on the drain region 12 side, which helps to limit the expansion of the depletion region and suppress short-channel effects and leakage current; a medium doping concentration is used in the middle region to form a transition in electrical characteristics; and a lower doping concentration is used on the source region 11 side, which helps to reduce the influence of impurity scattering on carrier mobility, thereby achieving partitioned electric field modulation of the source region 11, the middle region, and the drain region 12.
[0059] Furthermore, after completing the ion implantation, a separate high-temperature annealing can be temporarily omitted. Instead, a rapid thermal annealing, such as at approximately 1000°C, can be performed after subsequent source / drain heavy doping implantation. This simultaneously activates the dopant elements in the LDD, doped regions, and source / drain regions, and repairs the lattice damage caused by ion implantation. By concentrating the high-temperature annealing process in a subsequent step, the number of high-temperature thermal cycles experienced by the gate dielectric layer 21, especially the high-k dielectric material therein, can be reduced, and it is beneficial to maintain the step morphology of the semiconductor substrate 1.
[0060] Optional, continue to refer to Figure 2 The field-effect transistor further includes: a first sidewall liner layer 61 and a second sidewall liner layer 62; the first sidewall liner layer 61 is located between the first sidewall 31 and the gate structure 2, and extends from the sidewall of the gate structure 2 near the drain region 12 away from the source region 11 to the surface of the semiconductor substrate 1; the second sidewall liner layer 62 is located between the second sidewall 32 and the gate structure 2, and extends from the sidewall of the gate structure 2 near the source region 11 away from the drain region 12 to the surface of the semiconductor substrate 1.
[0061] Specifically, the first sidewall liner layer 61 and the second sidewall liner layer 62 can be respectively disposed inside the first sidewall 31 and the second sidewall 32, and located between the corresponding sidewall and the gate structure 2. The first sidewall liner layer 61 and the second sidewall liner layer 62 can be formed of SiO2 material, and the thickness can be, for example, about 15 angstroms. Each sidewall liner layer can extend downward along the sidewall of the corresponding side of the gate structure 2, and extend along the surface of the semiconductor substrate 1 in a direction away from the gate structure 2 near the bottom of the gate dielectric layer 21, thereby forming an approximately L-shaped cross-sectional profile. By providing the sidewall liner layer, damage to the semiconductor substrate surface that may occur during the etching process of the gate structure 2 can be repaired and isolated. At the same time, it can serve as a buffer interface between the first sidewall 31 and the second sidewall 32 and the gate structure 2, reducing the interfacial stress between different dielectric materials.
[0062] For example, after the gate dielectric layer 21 and the gate electrode 22 are formed, a SiO2 pad material layer with a thickness of about 15 angstroms can be conformally deposited on the sidewalls of the gate structure 2 and the surface of the semiconductor substrate 1 on both sides of the gate structure 2 using a low-temperature ALD or CVD process. Subsequently, photoresist can be used to mask the side near the drain region 12, and a high-k dielectric material such as HfO2 with a thickness of 30-50 angstroms can be selectively deposited on the side near the source region 11 using an ALD process. After removing the photoresist, a low-k dielectric material such as SiOC and / or SiOCN with a thickness of about 400 angstroms can be deposited on both sides of the gate structure 2 using a PECVD process.
[0063] Subsequently, the SiO2 backing material layer, high-k dielectric material, and low-k dielectric material can be selectively removed by combining partitioned masking and anisotropic etching back processes. For example, CHF3 / O2 / Ar plasma can be used for etching back, and the etching selectivity ratio between SiOC, SiO2, and HfO2, for example, approximately 10:3:1, can be used to remove most of the low-k dielectric material near the source region 11, while retaining the high-k dielectric material located on the inner side; a thicker layer of low-k dielectric material is retained near the drain region 12. Thus, a second sidewall backing layer 62 with a thickness of approximately 15 angstroms and a thinner second sidewall 32 with a thickness of approximately 30-50 angstroms can be formed near the source region 11, and a first sidewall backing layer 61 with a thickness of approximately 15 angstroms and a thicker first sidewall 31 with a thickness of approximately 250-350 angstroms can be formed near the drain region 12.
[0064] The above structure allows the first sidewall 31 near the drain region 12 to have a larger thickness and a lower dielectric constant, which helps to increase the effective spacing between the gate electrode 22 and the drain region 12 and reduce the gate-drain parasitic capacitance, thereby weakening the drain electric field and Miller effect. At the same time, the second sidewall 32 near the source region 11 has a smaller thickness and a higher dielectric constant, which helps to enhance the electric field coupling from the gate edge to the source end and reduce the source end equivalent series resistance, thereby improving the carrier injection efficiency.
[0065] Optional, continue to refer to Figure 2 The field-effect transistor further includes: an interlayer dielectric layer covering the semiconductor substrate 1 and the gate structure 2 (transparent in the figure), a metal plug 7 penetrating the interlayer dielectric layer, and a metal wiring layer covering the interlayer dielectric layer (not shown in the figure); the metal plug 7 includes a source metal plug 71 electrically contacting the source region 11 and a drain metal plug 72 electrically contacting the drain region 12; along a direction parallel to the plane where the semiconductor substrate 1 is located, the shortest distance between the bottom edge of the source metal plug 71 and the bottom edge of the second sidewall 32 is less than or equal to the shortest distance between the bottom edge of the drain metal plug 72 and the bottom edge of the first sidewall 31.
[0066] Specifically, the interlayer dielectric layer may include a first interlayer dielectric layer and a second interlayer dielectric layer stacked together. For example, after heavy doping and annealing of the source region 11 and drain region 12, a Ni or Ti metal layer can be deposited on the source region 11, drain region 12, and the top of the gate structure 2, followed by annealing. This allows the metal on the surface of the source region 11 and drain region 12 to react with the semiconductor material to form a metal silicide layer, thereby reducing the contact resistance between the subsequent metal plug 7 and the source region 11 and drain region 12. Subsequently, unreacted metal can be removed by selective etching. It should be noted that when the gate electrode 22 uses a silicon-free metal or alloy material such as TiN, the Ni or Ti located on top of the gate electrode 22 will not undergo silicide reaction and can be removed during subsequent selective etching.
[0067] Subsequently, SiO2 can be deposited using PECVD to form a first interlayer dielectric layer, and CMP can be used to planarize the first interlayer dielectric layer so that the top surface of the first interlayer dielectric layer is substantially flush with the top surface of the gate structure 2. Then, a second interlayer dielectric layer is deposited on the first interlayer dielectric layer and the gate structure 2. The first interlayer dielectric layer and the second interlayer dielectric layer can be made of the same or different insulating dielectric materials, and the embodiments of the present invention do not specifically limit this.
[0068] Furthermore, the second interlayer dielectric layer and the first interlayer dielectric layer can be sequentially etched using photolithography and etching processes to form contact vias penetrating the interlayer dielectric layer. These contact vias may include source-end contact vias located above source region 11 and drain-end contact vias located above drain region 12, and may also include gate-end contact vias corresponding to gate electrode 22 as needed. Subsequently, conductive materials such as W or Cu are filled into the source-end contact vias and drain-end contact vias to form source-end metal plugs 71 electrically contacting source region 11 and drain-end metal plugs 72 electrically contacting drain region 12, respectively.
[0069] The source metal plug 71 and drain metal plug 72 are asymmetrically arranged relative to their respective sidewalls. Along a direction parallel to the plane of the semiconductor substrate 1, the shortest distance between the bottom edge of the source metal plug 71 and the bottom edge of the second sidewall 32 is less than or equal to the shortest distance between the bottom edge of the drain metal plug 72 and the bottom edge of the first sidewall 31. For example, the source contact via can be positioned close to the second sidewall 32 so that the source metal plug 71 formed therein is connected to the source region 11 via a shorter conductive path; the drain contact via can maintain a larger horizontal distance from the first sidewall 31 so that the drain metal plug 72 has a larger lateral distance from the gate structure 2.
[0070] By bringing the source metal plug 71 closer to the second sidewall 32, the lateral conductive path between the source region 11 and the source metal plug 71 can be shortened, reducing the source parasitic series resistance, which is beneficial to improving the carrier injection efficiency and the device's driving capability. Correspondingly, by maintaining a larger distance between the drain metal plug 72 and the first sidewall 31, the drain series resistance can be appropriately increased, and the resulting voltage divider effect can be used to reduce the drain voltage near the channel region, thereby weakening the drain electric field peak, suppressing impact ionization and hot carrier injection, and improving the reliability of the field-effect transistor.
[0071] Furthermore, after forming the interlayer dielectric layer and the metal plug 7 that penetrates the interlayer dielectric layer, a metal wiring layer can also be formed on the side of the interlayer dielectric layer facing away from the semiconductor substrate 1. The metal wiring layer can be electrically connected to the source metal plug 71, the drain metal plug 72, and the gate metal plug to realize the electrical connection between the source region 11, the drain region 12, and the gate electrode 22 and the peripheral circuit.
[0072] Based on the same inventive concept, embodiments of the present invention also provide a method for fabricating a field-effect transistor. Figure 3 This is a flowchart of a method for fabricating a field-effect transistor according to an embodiment of the present invention. Figure 4 This is a process flow diagram of a field-effect transistor fabrication according to an embodiment of the present invention, in conjunction with reference to [reference needed]. Figure 3 and Figure 4 As shown, the fabrication method of this field-effect transistor includes: S101. A semiconductor substrate is provided, and the semiconductor substrate is patterned to form a step region on the surface of the semiconductor substrate, including at least a first step region and a second step region.
[0073] The semiconductor substrate 1 has a first surface and a second surface that are disposed opposite to each other. The semiconductor substrate 1 includes a source end 001 and a drain end 002 that are disposed at intervals. The first step region 01 is close to the drain end 002, and the second step region 02 is close to the source end 001. The height of the top surface of the semiconductor substrate 1 in the first step region 01 is lower than the height of the top surface of the semiconductor substrate 1 in the second step region 02.
[0074] Specifically, the first surface of the semiconductor substrate 1 can be patterned using multiple photolithography and etching processes to form multiple stepped regions 00 with different top surface heights on the first surface of the semiconductor substrate 1. The multiple stepped regions 00 include at least a first stepped region 01 near the drain end 002 and a second stepped region 02 near the source end 001, with the top surface height of the semiconductor substrate 1 in the first stepped region 01 being lower than the top surface height of the semiconductor substrate 1 in the second stepped region 02. By pre-forming the stepped regions 00, a structural basis can be provided for the subsequent formation of gate dielectric layers 21 of different thicknesses at the source end 001 and the drain end 002.
[0075] Optionally, a third step region 03 may be formed between the first step region 01 and the second step region 02, so that the semiconductor substrate 1 forms a multi-level step morphology from the drain end 002 to the source end 001.
[0076] It should be noted that before source-drain doping, the actual source region 11 and drain region 12 have not yet been formed in the semiconductor substrate 1. Therefore, the side of the semiconductor substrate 1 subsequently used to form the source region 11 is called the source end 001, and the side subsequently used to form the drain region 12 is called the drain end 002. The source end 001 and drain end 002 are used to characterize the predetermined formation positions of the source region 11 and drain region 12 in the semiconductor substrate 1. After the source-drain doping is completed, the source region 11 is formed at the source end 001, and the drain region 12 is formed at the drain end 002.
[0077] S102. A gate dielectric layer is formed on one side of the semiconductor substrate, and a gate electrode is formed on the side of the gate dielectric layer away from the semiconductor substrate to form a gate structure.
[0078] The thickness of the gate dielectric layer 21 covering the first step region 01 is greater than the thickness of the gate dielectric layer 21 covering the second step region 02. If a third step region 03 is formed between the first step region 01 and the second step region 02, the thickness of the gate dielectric layer 21 covering the third step region 03 is between the thickness of the gate dielectric layer 21 covering the first step region 01 and the thickness of the gate dielectric layer 21 covering the second step region 02.
[0079] Specifically, the gate dielectric layer 21 may include a first dielectric layer 211 and a second dielectric layer 212 stacked together. For example, a first gate dielectric material may first be deposited on a first surface of a semiconductor substrate 1 having a stepped region 00, and planarized using CMP, so that the bottom surface of the first dielectric layer 211 matches the stepped morphology of the first surface of the semiconductor substrate 1, and the top surface is planar; subsequently, a second gate dielectric material is deposited on the first dielectric layer 211, and second dielectric layers 212 with different thicknesses are formed by regional photolithography and etching. The first dielectric layer 211 may be made of SiO2, and the second dielectric layer 212 may be made of a high-k dielectric material such as HfO2. Thus, the thickness of the gate dielectric layer 21 covering the first stepped region 01 is greater than the thickness of the gate dielectric layer 21 covering the second stepped region 02, and when the third stepped region 03 is formed, the thickness of the gate dielectric layer 21 covering the third stepped region 03 is between the two, thereby forming a gate dielectric structure with progressively increasing thickness from the source end 001 to the drain end 002.
[0080] Subsequently, metal gate materials such as TaN / TiN can be formed using PVD or ALD, and then regional doping and planarization can be performed to form the gate electrode 22. Along the direction from the drain end 002 to the source end 001, the gate electrode 22 can sequentially include a first gate region 221, a transition gate region 222, and a second gate region 223. Depending on the electrical performance requirements of different regions, doping elements such as N, W, and La can be introduced into the first gate region 221, the transition gate region 222, and the second gate region 223, respectively. By adjusting the type and concentration of the doping elements, the work function of each gate region can be adjusted so that the work function of the first gate region 221 is greater than that of the transition gate region 222, and the work function of the transition gate region 222 is greater than that of the second gate region 223. For example, the doping concentration of each gate region can gradually decrease along the direction from the drain end 002 to the source end 001, for example, in a ratio of approximately 4:2:1. This allows the regional electrical characteristics of the gate electrode 22 to match the thickness gradient of the gate dielectric layer 21, achieving differentiated gate control.
[0081] S103. Perform light doping extension region implantation and partitioned doping implantation on the semiconductor substrate to form a first light doped extension region and a first doped region on the side near the drain end, and a second light doped extension region and a second doped region on the side near the source end.
[0082] Specifically, the gate structure 2 can be used as a self-aligned mask for LDD implantation, forming a first lightly doped extension region 51 on the side near the drain end 002 and a second lightly doped extension region 52 on the side near the source end 001, so as to form a shallow junction transition region between the subsequently formed source region 11, drain region 12 and channel, thereby reducing the peak electric field at the channel edge.
[0083] Subsequently, doping can be performed using a partitioned mask to form a first doped region 41, a transition doped region 42, and a second doped region 43, with the doping concentration of all three gradually decreasing from the drain end 002 to the source end 001. The doping concentration gradient can be matched with the doping concentration gradients of the first gate region 221, the transition gate region 222, and the second gate region 223, for example, in a ratio of approximately 4:2:1. This allows the higher doping concentration on the drain end 002 side to suppress depletion region expansion, short-channel effects, and leakage current, while the lower doping concentration on the source end 001 side helps reduce impurity scattering. Together with the work function gradient of the gate electrode 22, this achieves partitioned electric field modulation of the source, channel, and drain ends.
[0084] S104. A first sidewall is formed on the side of the gate structure near the drain end, and a second sidewall is formed on the side of the gate structure near the source end.
[0085] In this context, the direction from the source terminal 001 to the drain terminal 002 is such that the thickness of the first sidewall 31 is greater than the thickness of the second sidewall 32; and the dielectric constant of the first sidewall 31 is less than the dielectric constant of the second sidewall 32.
[0086] Specifically, a SiO2 pad layer can be formed first on the sidewalls of the gate structure 2 and the surface of the semiconductor substrate 1 on both sides. Then, a high-k dielectric material can be selectively formed on the side near the source end 001, and a low-k dielectric material can be formed on both sides of the gate structure 2. Afterwards, through regional etching and anisotropic etching back, a thicker low-k dielectric material is retained on the side near the drain end 002 to form the first sidewall 31, and a thinner high-k dielectric material is retained on the side near the source end 001 to form the second sidewall 32. Thus, the thicker first sidewall 31 with low-k dielectric material on the drain end 002 side can reduce gate-drain parasitic capacitance, reduce Miller effect and gate delay; the thinner second sidewall 32 with high-k dielectric material on the source end 001 side can enhance gate edge electric field coupling and reduce source end series resistance, thereby improving the switching speed and driving performance of the device.
[0087] S105. Perform source and drain doping on the semiconductor substrate to form a source region at the source end and a drain region at the drain end.
[0088] Specifically, the gate structure 2, the first sidewall 31, and the second sidewall 32 can be used as self-aligned masks for heavy doping implantation of the source and drain to form a source region 11 at the source end 001 and a drain region 12 at the drain end 002. After implantation, RTA can be performed to simultaneously activate the doped elements in the source / drain regions, lightly doped extension regions, etc., and repair implantation damage. By performing annealing unification, the number of high-temperature thermal cycles experienced by the gate dielectric layer 21 and the step region 00 can be reduced.
[0089] S106, forming an interlayer dielectric layer covering the semiconductor substrate and gate structure, a metal plug penetrating the interlayer dielectric layer, and a metal wiring layer covering the interlayer dielectric layer.
[0090] Specifically, a metal silicide layer can be formed on the surface of the source region 11 and the drain region 12 first, followed by the formation of a first interlayer dielectric layer and a second interlayer dielectric layer, and source contact vias and drain contact vias that penetrate the interlayer dielectric layer can be formed by photolithography and etching. After that, conductive materials such as W and Cu are filled into each contact via to form source metal plugs 71 and drain metal plugs 72, respectively.
[0091] The shortest distance between the bottom edge of the source metal plug 71 and the bottom edge of the second sidewall 32 is less than or equal to the shortest distance between the bottom edge of the drain metal plug 72 and the bottom edge of the first sidewall 31. Therefore, the source metal plug 71 can connect to the source region 11 with a shorter conductive path, reducing the parasitic series resistance at the source. Maintaining a larger horizontal distance between the drain metal plug 72 and the first sidewall 31 allows for a resistive voltage division effect by increasing the drain series resistance, further weakening the peak drain electric field.
[0092] The metal wiring layer can be obtained through conventional methods such as depositing metal on the second interlayer dielectric layer and patterning, which will not be elaborated further.
[0093] Using the above-described fabrication method, a progressively thickening double-layer gate dielectric structure is formed from the source region 11 to the drain region 12. Combined with the doping concentration and work function gradient of the gate electrode 22 and the doping concentration gradient in the semiconductor substrate 1, the electric field of the source region 11, channel, and drain region 12 is modulated in a partitioned manner. This gives the source region 11 strong gate control and carrier injection capabilities, while reducing the drain electric field, DIBL, hot carrier effect, and gate tunneling leakage current. Furthermore, the thicker first sidewall 31 (made of low-k dielectric material) on the drain region 12 side and the thinner second sidewall 32 (made of high-k dielectric material) on the source region 11 side reduce gate-drain parasitic capacitance, Miller effect, and gate delay, while enhancing the source-gate edge electric field coupling and reducing the source-series resistance. Furthermore, by adopting an asymmetric contact structure, the source contact via is brought closer to the second sidewall 32 to shorten the conductive path, while the drain contact via is kept at a large horizontal distance from the first sidewall 31. This can further reduce the parasitic series resistance at the source and weaken the peak value of the drain electric field by utilizing the voltage division effect of the drain resistance, thereby balancing the device's driving capability, low leakage current performance, and reliability.
[0094] The above-described method for fabricating field-effect transistors can be used to fabricate the field-effect transistors provided in any embodiment of the present invention, which possess the corresponding functions and beneficial effects of field-effect transistors. Technical details not described in detail in this embodiment can be found in the field-effect transistors provided in any embodiment of the present invention.
[0095] The above is only one feasible method for fabricating a field-effect transistor provided by the present invention, but it is not limited thereto. Those skilled in the art can also use other process routes, such as performing ion implantation on different regions of the semiconductor substrate 1 before or after forming a step region on the surface of the semiconductor substrate 1 to form doped regions of different concentrations.
[0096] The effectiveness of the multi-step gate dielectric structure and partitioned electrical control design adopted in this invention is verified by simulation analysis of the field-effect transistor provided in this embodiment. In this design, the doping concentration in the semiconductor substrate 1 is set in a ratio of approximately 4:2:1 along the drain region 12 and the channel to the source region 11. The first dielectric layer 211 is made of silicon dioxide, and the second dielectric layer 212 is made of a high-k dielectric material, such as HfO2. Gate control adjustment in different regions is mainly achieved through thickness partitioning. Figure 5 The diagram shows the thickness of the double-layer gate dielectric and the equivalent gate capacitance distribution of the field-effect transistor provided in the embodiments of the present invention. Figure 5 As shown, along the direction from the source region 11 to the drain region 12, the thickness of the first dielectric layer 211 and the second dielectric layer 212 generally increases, while the corresponding equivalent gate capacitance (C) increases. OX The gate coupling strength of the source region 11, channel, and drain region 12 gradually decreases (see the dotted line in the figure), thereby enabling different gate coupling strengths to be formed in the source region 11, channel, and drain region 12, and realizing the regional control of the electrical performance of different regions of the field-effect transistor.
[0097] Figure 6 The diagram illustrates the relationship between the total gate dielectric layer thickness, threshold voltage, and on-current of a field-effect transistor provided in this embodiment of the invention. The total gate dielectric layer thickness is the sum of the thicknesses of the first dielectric layer 211 and the second dielectric layer 212. Simulation results show that the total gate dielectric layer thickness corresponding to the source region 11 is approximately 17-33 Å. The thinner gate dielectric layer 21 in the source region 11 results in a larger equivalent gate capacitance, which is beneficial for reducing the threshold voltage and increasing the on-current. The total gate dielectric layer thickness corresponding to the channel region is approximately 47-71 Å, allowing for a smooth transition in electrical characteristics between the source region 11 and the drain region 12. The total gate dielectric layer thickness corresponding to the drain region 12 is approximately 83-103 Å. The thicker gate dielectric layer 21 in the drain region 12 results in a smaller equivalent gate capacitance, which is beneficial for increasing the potential barrier in the drain region and weakening the peak electric field, thereby suppressing DIBL and gate tunneling leakage. Therefore, the multi-step gate dielectric thickness design can be coordinated with the doping concentration gradient to improve both the source-end driving capability and the low leakage current and breakdown voltage performance at the drain end.
[0098] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.
[0099] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.
Claims
1. A field effect transistor, characterized by include: A semiconductor substrate, wherein the semiconductor substrate includes source regions and drain regions disposed at intervals; A gate structure is located on one side of the semiconductor substrate and between the source region and the drain region; The semiconductor substrate includes at least two stepped regions, the stepped regions including a first stepped region near the drain region and a second stepped region near the source region, and the top surface height of the semiconductor substrate in the first stepped region is lower than the top surface height of the semiconductor substrate in the second stepped region. The gate structure includes a stacked gate dielectric layer and a gate electrode; the thickness of the gate dielectric layer covering the first step region is greater than the thickness of the gate dielectric layer covering the second step region; The gate structure has a first sidewall on the side near the drain region and a second sidewall on the side near the source region; the thickness of the first sidewall is greater than the thickness of the second sidewall in the direction from the source region to the drain region; the dielectric constant of the first sidewall is less than the dielectric constant of the second sidewall.
2. The field effect transistor of claim 1, wherein The semiconductor substrate further includes a third step region located between the first step region and the second step region; The height of the top surface of the semiconductor substrate in the third step region is higher than the height of the top surface of the semiconductor substrate in the first step region, but lower than the height of the top surface of the semiconductor substrate in the second step region; The thickness of the gate dielectric layer covering the third step region is less than the thickness of the gate dielectric layer covering the first step region, but greater than the thickness of the gate dielectric layer covering the second step region.
3. The field effect transistor according to claim 1 or 2, characterized in that The gate dielectric layer includes a first dielectric layer and a second dielectric layer stacked together, the second dielectric layer being located on the side of the first dielectric layer away from the semiconductor substrate; the first dielectric layer and the second dielectric layer may be made of the same or different materials; the thickness of the second dielectric layer increases or remains constant from the source region to the drain region.
4. The field effect transistor of claim 3, wherein The first dielectric layer and the second dielectric layer corresponding to the same step region have the same thickness.
5. The field effect transistor according to claim 1 or 2, characterized in that Along the direction from the drain region to the source region, the gate electrode includes a first gate region, a transition gate region, and a second gate region disposed sequentially. Along the thickness direction of the semiconductor substrate, the projection of the first gate region at least partially overlaps with the projection of the first step region, the projection of the second gate region at least partially overlaps with the projection of the second step region, and the projection of the transition gate region at least partially overlaps with the projection of the third step region. The work function of the first gate region is greater than the work function of the transition gate region, and the work function of the transition gate region is greater than the work function of the second gate region.
6. The field effect transistor of claim 2, wherein The semiconductor substrate further includes a first doped region, a transition doped region, and a second doped region disposed sequentially thereon. The first doped region is disposed near the drain region and is at least partially located within the first step region; the second doped region is disposed near the source region and is at least partially located within the second step region; the transition doped region is located between the first doped region and the second doped region and is at least partially located within the third step region. The doping concentration of the first doped region is greater than the doping concentration of the transition doped region, and the doping concentration of the transition doped region is greater than the doping concentration of the second doped region.
7. The field effect transistor of claim 1, wherein Also includes: First side wall lining layer and second side wall lining layer; The first sidewall liner is located between the first sidewall and the gate structure, and extends from the sidewall of the gate structure near the drain region away from the source region to the surface of the semiconductor substrate; The second sidewall liner is located between the second sidewall and the gate structure, and extends from the sidewall of the gate structure near the source region away from the drain region to the surface of the semiconductor substrate.
8. The field effect transistor of claim 1, wherein The semiconductor substrate further includes a first lightly doped extension region and a second lightly doped extension region; The first lightly doped extension region is located close to the drain region, and the second lightly doped extension region is located close to the source region; the doping type of the first lightly doped extension region is the same as that of the drain region and the doping concentration is lower than that of the drain region, and the doping type of the second lightly doped extension region is the same as that of the source region and the doping concentration is lower than that of the source region.
9. The field effect transistor of claim 1, wherein, Also includes: An interlayer dielectric layer covering the semiconductor substrate and the gate structure, a metal plug penetrating the interlayer dielectric layer, and a metal wiring layer covering the interlayer dielectric layer; The metal plug includes a source-end metal plug that is in electrical contact with the source region and a drain-end metal plug that is in electrical contact with the drain region. Along a direction parallel to the plane where the semiconductor substrate is located, the shortest distance between the bottom edge of the source metal plug and the bottom edge of the second sidewall is less than or equal to the shortest distance between the bottom edge of the drain metal plug and the bottom edge of the first sidewall.
10. A method of fabricating a field effect transistor, characterized by, include: A semiconductor substrate is provided, and the semiconductor substrate is patterned to form a stepped region on the surface of the semiconductor substrate, including at least a first stepped region and a second stepped region; wherein the semiconductor substrate includes a source end and a drain end disposed at intervals, the first stepped region is close to the drain end, the second stepped region is close to the source end, and the top surface height of the semiconductor substrate in the first stepped region is lower than the top surface height of the semiconductor substrate in the second stepped region; A gate dielectric layer is formed on one side of the semiconductor substrate, and a gate electrode is formed on the side of the gate dielectric layer opposite to the semiconductor substrate to form a gate structure; wherein the thickness of the gate dielectric layer covering the first step region is greater than the thickness of the gate dielectric layer covering the second step region. A first sidewall is formed on the side of the gate structure near the drain end, and a second sidewall is formed on the side of the gate structure near the source end; wherein, in the direction from the source end to the drain end, the thickness of the first sidewall is greater than the thickness of the second sidewall; and the dielectric constant of the first sidewall is less than the dielectric constant of the second sidewall. The semiconductor substrate is doped with source and drain doping to form a source region at the source end and a drain region at the drain end; An interlayer dielectric layer is formed covering the semiconductor substrate and the gate structure, a metal plug penetrating the interlayer dielectric layer, and a metal wiring layer covering the interlayer dielectric layer.
