CN106847755A - The method for improving SRAM performances - Google Patents

The method for improving SRAM performances Download PDF

Info

Publication number
CN106847755A
CN106847755A CN201510896857.0A CN201510896857A CN106847755A CN 106847755 A CN106847755 A CN 106847755A CN 201510896857 A CN201510896857 A CN 201510896857A CN 106847755 A CN106847755 A CN 106847755A
Authority
CN
China
Prior art keywords
type
layer
area
work
threshold voltage
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201510896857.0A
Other languages
Chinese (zh)
Other versions
CN106847755B (en
Inventor
李勇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Semiconductor Manufacturing International Shanghai Corp
Semiconductor Manufacturing International Beijing Corp
Original Assignee
Semiconductor Manufacturing International Shanghai Corp
Semiconductor Manufacturing International Beijing Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Semiconductor Manufacturing International Shanghai Corp, Semiconductor Manufacturing International Beijing Corp filed Critical Semiconductor Manufacturing International Shanghai Corp
Priority to CN201510896857.0A priority Critical patent/CN106847755B/en
Publication of CN106847755A publication Critical patent/CN106847755A/en
Application granted granted Critical
Publication of CN106847755B publication Critical patent/CN106847755B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • HELECTRICITY
    • H10SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
    • H10BELECTRONIC MEMORY DEVICES
    • H10B10/00Static random access memory [SRAM] devices
    • H10B10/12Static random access memory [SRAM] devices comprising a MOSFET load element
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L21/00Processes or apparatus adapted for the manufacture or treatment of semiconductor or solid state devices or of parts thereof
    • H01L21/70Manufacture or treatment of devices consisting of a plurality of solid state components formed in or on a common substrate or of parts thereof; Manufacture of integrated circuit devices or of parts thereof
    • H01L21/77Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate
    • H01L21/78Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate with subsequent division of the substrate into plural individual devices
    • H01L21/82Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate with subsequent division of the substrate into plural individual devices to produce devices, e.g. integrated circuits, each consisting of a plurality of components
    • H01L21/822Manufacture or treatment of devices consisting of a plurality of solid state components or integrated circuits formed in, or on, a common substrate with subsequent division of the substrate into plural individual devices to produce devices, e.g. integrated circuits, each consisting of a plurality of components the substrate being a semiconductor, using silicon technology
    • H01L21/8232Field-effect technology
    • H01L21/8234MIS technology, i.e. integration processes of field effect transistors of the conductor-insulator-semiconductor type
    • H01L21/8238Complementary field-effect transistors, e.g. CMOS
    • H01L21/823828Complementary field-effect transistors, e.g. CMOS with a particular manufacturing method of the gate conductors, e.g. particular materials, shapes
    • H01L21/823842Complementary field-effect transistors, e.g. CMOS with a particular manufacturing method of the gate conductors, e.g. particular materials, shapes gate conductors with different gate conductor materials or different gate conductor implants, e.g. dual gate structures
    • HELECTRICITY
    • H01ELECTRIC ELEMENTS
    • H01LSEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
    • H01L29/00Semiconductor devices adapted for rectifying, amplifying, oscillating or switching, or capacitors or resistors with at least one potential-jump barrier or surface barrier, e.g. PN junction depletion layer or carrier concentration layer; Details of semiconductor bodies or of electrodes thereof  ; Multistep manufacturing processes therefor
    • H01L29/40Electrodes ; Multistep manufacturing processes therefor
    • H01L29/41Electrodes ; Multistep manufacturing processes therefor characterised by their shape, relative sizes or dispositions
    • H01L29/423Electrodes ; Multistep manufacturing processes therefor characterised by their shape, relative sizes or dispositions not carrying the current to be rectified, amplified or switched
    • H01L29/42312Gate electrodes for field effect devices
    • H01L29/42316Gate electrodes for field effect devices for field-effect transistors
    • H01L29/4232Gate electrodes for field effect devices for field-effect transistors with insulated gate
    • H01L29/42356Disposition, e.g. buried gate electrode

Abstract

A kind of method of improvement SRAM performances, including:P-type workfunction layer is formed on p-type logic device area gate dielectric layer surface, the maximum P-type workfunction layer of equivalent work function value is the first P-type workfunction layer;Pull-up work-function layer is formed on the gate dielectric layer surface in the area that pulls up transistor, the material and thickness for pulling up work-function layer are identical with the material and thickness of the first P-type workfunction layer;Substrate to the area that pulls up transistor carries out first threshold voltage regulation doping treatment;N-type workfunction layer is formed on N-type logic device area gate dielectric layer surface, and the maximum N-type workfunction layer of equivalent work function value is the first N-type workfunction layer;Transmission gate work-function layer is formed on the gate dielectric layer surface of transmission gate transistor area, transmission gate work-function layer is identical with the material and thickness of the first N-type workfunction layer;Substrate to transmission gate transistor area carries out second threshold voltage regulation doping treatment.The present invention improves the write-in redundancy of memory, improves the overall performance of semiconductor devices.

Description

The method for improving SRAM performances
Technical field
The present invention relates to semiconductor fabrication techniques field, more particularly to a kind of method of improvement SRAM performances.
Background technology
In current semiconductor industry, IC products can be divided mainly into three major types type:Logic, deposit Reservoir and analog circuit, wherein memory device account for sizable ratio in IC products.With Semiconductor technology develop, memory device is more widely applied, it is necessary to by the memory device with Other device regions are formed on a single die, to form embedded semiconductor storing equipment simultaneously.For example will Central processing unit is embedded in the memory device, then it is required that the memory device and embedded center Processor platform carries out compatibility, and keeps the specification and corresponding electric property of original memory device.
Usually, it is necessary to the memory device and embedded standard logical devices be carried out compatible.For embedding Enter for formula semiconductor devices, it is generally divided into logic area and memory block, logic area generally includes logic device Part, memory block then includes memory device.With the development of memory technology, occur in that and various types of partly lead Body memory, such as static random random access memory (SRAM, Static Random Access Memory), Dynamic RAM (DRAM, Dynamic Random Access Memory), erasable compile Journey read-only storage (EPROM, Erasable Programmable Read-Only Memory), electricity can Erasable programmable read-only memory (EPROM) (EEPROM, Electrically Erasable Programmable ) and flash memory (Flash) Read-Only.Because SRAM has low-power consumption and very fast work speed The advantages of spending so that SRAM and forming method thereof is received more and more attention.
However, the performance of SRAM needs further in the semiconductor devices of prior art formation Improve so that the overall performance of semiconductor devices is poor.
The content of the invention
The problem that the present invention is solved is to provide a kind of method of improvement SRAM performances, improves writing for memory Enter redundancy, so as to improve the overall performance of the semiconductor devices of formation.
To solve the above problems, the present invention provides a kind of method of improvement SRAM performances, including:There is provided Substrate, the substrate include N-type logic device area, p-type logic device area, pull up transistor area and Transmission gate transistor area, wherein, the N-type logic device area includes several N-type threshold voltage areas, The p-type logic device area includes several p-type threshold voltage areas, the N-type logic device area, P Type logic device area, the part of substrate surface of pull up transistor area and transmission gate transistor area are formed with grid Dielectric layer;P-type workfunction layer is formed on the p-type logic device area gate dielectric layer surface, and if described The equivalent work function value of the corresponding P-type workfunction layer in Gan GePXing threshold voltages area is different, wherein, it is equivalent The maximum P-type workfunction layer of work function value is the first P-type workfunction layer;In the area that pulls up transistor Gate dielectric layer surface forms pull-up work-function layer, and the material and thickness and a P for pulling up work-function layer The material of type work-function layer is identical with thickness;Substrate to the area that pulls up transistor carries out first threshold electricity Pressure regulation doping treatment;N-type workfunction layer is formed on the N-type logic device area gate dielectric layer surface, And the equivalent work function value of several corresponding N-type workfunction layers in N-type threshold voltage area is different, its In, the maximum N-type workfunction layer of equivalent work function value is the first N-type workfunction layer;In the transmission gate The gate dielectric layer surface of transistor area forms transmission gate work-function layer, and the transmission gate work-function layer material Material and thickness are identical with the material and thickness of the first N-type workfunction layer;To the transmission gate transistor area Substrate carries out second threshold voltage regulation doping treatment;In the N-type workfunction layer surface, p-type work content Number layer surface, transmission gate work-function layer surface and pull-up work-function layer surface form gate electrode layer.
Optionally, in described several corresponding P-type workfunction layers in p-type threshold voltage area, described The thickness of one P-type workfunction layer is most thick;In described several corresponding N-type work contents in N-type threshold voltage area In several layers, the thickness of first N-type workfunction layer is most thin.
Optionally, in the processing step along with, pull-up work-function layer and the first p-type work content are formed Several layers;In the processing step along with, the transmission gate work-function layer and the first N-type workfunction layer are formed.
Optionally, the Doped ions of the second threshold voltage regulation doping treatment are B, and doping concentration is 1E12atom/cm3To 1E14atom/cm3
Optionally, the Doped ions of the first threshold voltage doping treatment are As, and doping concentration is 1E12atom/cm3To 1E14atom/cm3
Compared with prior art, technical scheme has advantages below:
In the technical scheme of the method for the improvement SRAM performances that the present invention is provided, in p-type logic device area Gate dielectric layer surface forms P-type workfunction layer, and described several corresponding p-types in p-type threshold voltage area The equivalent work function value of work-function layer is different, wherein, the maximum P-type workfunction layer of equivalent work function value is First P-type workfunction layer;Pull-up work-function layer is formed on the gate dielectric layer surface in the area that pulls up transistor, and The material and thickness of the pull-up work-function layer are identical with the material and thickness of the first P-type workfunction layer. That is, equivalent work function value and some P-type workfunction layers of the pull-up work-function layer in the area that pulls up transistor Equivalent work function value in maximum equivalent work function value it is identical, fixed for the holding that makes to pull up transistor has Threshold voltage numerical value, the substrate in the area that pulls up transistor is carried out at first threshold voltage regulation doping The Doped ions concentration of reason is higher, and then causes the saturation current for pulling up transistor and ON state current that are formed It is smaller.N-type workfunction layer, and described several N are formed on N-type logic device area gate dielectric layer surface The equivalent work function value of the corresponding N-type workfunction layer in type threshold voltage area is different, wherein, equivalent work function It is the first N-type workfunction layer to be worth maximum N-type workfunction layer;In the gate dielectric layer of transmission gate transistor area Surface formed transmission gate work-function layer, and the transmission gate work-function layer material and thickness and the first N-type The material of work-function layer is identical with thickness.That is, the transmission gate work-function layer of transmission gate transistor area Equivalent work function value and the equivalent work function value of some N-type workfunction layers in maximum equivalent work function value It is identical, in order that transmission gate transistor keeps having fixed threshold voltage numerical value, it is brilliant to the transmission gate The Doped ions concentration of the second threshold voltage regulation doping treatment that the substrate in body area under control is carried out is relatively low, and then So that the saturation current and ON state current of the transmission gate transistor for being formed are larger.Therefore the half of present invention formation The gamma ratio of memory is improved in conductor device, so that the write-in redundancy of memory is changed It is apt to, and then improves the electric property of the memory for being formed, improves the overall performance of semiconductor devices.
Brief description of the drawings
Fig. 1 to Figure 15 shows for the cross-section structure of the semiconductor devices forming process that one embodiment of the invention is provided It is intended to.
Specific embodiment
From background technology, the property of SRAM in the semiconductor devices for being formed in the prior art Can have much room for improvement.
For SRAM, its mainly include pull-up (PU, Pull Up) transistor, it is drop-down (PD, Pull Down) transistor and transmission gate (PG, Pass Gate) transistor, and the write-in of memory is superfluous Remaining (write margin) plays a key effect to memory performance, if the write-in of memory can be improved Redundancy performance, then the yield of memory will be enhanced, the overall performance of semiconductor devices is accordingly obtained Improve.Research discovery, write-in redundancy and gamma ratio (gamma ratio) pass in direct ratio of memory System, gamma is than the ratio between the ON state current for transmission gate transistor and the ON state current for pulling up transistor. The ON state current of transmission gate transistor is relevant with the Doped ions concentration of transmission gate transistor channel region, transmission The Doped ions concentration of door transistor channel region is lower, then the ON state current of transmission gate transistor is bigger;On The ON state current of pull transistor is relevant with the Doped ions concentration of the channel region that pulls up transistor, and pulls up transistor The Doped ions concentration of channel region is higher, then the ON state current for pulling up transistor is smaller.Therefore, reduce and pass Send the Doped ions concentration of a transistor channel region, or the Doped ions for improving the channel region that pulls up transistor Concentration, enables to the gamma of memory than increasing, and then improves the write-in redundancy of memory, improves The yield of memory.
Further study show that, for transmission gate transistor, transmission gate transistor is NMOS tube, The transmission gate transistor typically has fixed threshold voltage value (Vt), if forming transmission gate transistor When employ equivalent work function value (equal work function) work-function layer higher, in order that transmission Door transistor keeps fixed threshold voltage, then the threshold voltage adjustments of corresponding transmission gate transistor channel region Doped ions concentration is relatively low so that the ON state current of transmission gate transistor increases.For pulling up transistor and Speech, it is PMOS to pull up transistor, and described pulling up transistor typically also has fixed threshold voltage, if Equivalent work function value work-function layer higher is employed when being formed and being pulled up transistor, then in order that upper crystal pulling Body pipe keeps fixed threshold voltage, the threshold voltage adjustments Doped ions of the channel region that pulls up transistor Concentration should be higher so that the ON state current for pulling up transistor reduces.
Therefore, the present invention provides a kind of method of improvement SRAM performances, pull up transistor area in the present invention The equivalent work function value of pull-up work-function layer be:In p-type logic device area some P-type workfunction layers etc. Maximum equivalent work function value in effect work function value, keeps having fixed threshold value electricity to make to pull up transistor Pressure numerical value, the substrate to the area that pulls up transistor carries out the doping that first threshold voltage adjusts doping treatment Ion concentration is higher, and then causes that the saturation current for pulling up transistor and ON state current of formation are smaller;Pass The equivalent work function value for sending the transmission gate work-function layer of a transistor area is:Some N in N-type logic device area Maximum equivalent work function value in the equivalent work function value of type work-function layer, in order that transmission gate transistor is protected Hold with fixed threshold voltage numerical value, the Second Threshold carried out to the substrate of the transmission gate transistor area The Doped ions concentration of voltage-regulation doping treatment is relatively low, and then causes satisfying for the transmission gate transistor for being formed It is larger with electric current and ON state current.Therefore the gamma of memory is compared in the semiconductor devices that the present invention is formed To raising, so that the write-in redundancy of memory is improved, and then the memory for being formed is improved Electric property, improves the overall performance of semiconductor devices.
It is understandable to enable the above objects, features and advantages of the present invention to become apparent, below in conjunction with the accompanying drawings Specific embodiment of the invention is described in detail.
Fig. 1 to Figure 15 shows for the cross-section structure of the semiconductor devices forming process that one embodiment of the invention is provided It is intended to.
With reference to Fig. 1, there is provided substrate, the substrate includes that N-type logic device area (sign), p-type are patrolled Collect device region (sign), pull up transistor area I and transmission gate transistor area II.
The semiconductor devices that the present embodiment is formed includes logical device and SRAM device.The N-type is patrolled Collect device region and provide technique platform to be subsequently formed N-type logical device, after the p-type logic device area is The continuous p-type logical device that formed provides technique platform, described to pull up transistor area I to be subsequently formed crystal pulling Pipe provides technique platform, and the transmission gate transistor area II provides technique to be subsequently formed transmission gate transistor Platform.For PMOS area, the transmission gate transistor area II is NMOS to the area I that pulls up transistor Region.
The substrate also includes pull-down transistor area III, and the pull-down transistor area III is drop-down to be subsequently formed Transistor provides technique platform, and the pull-down transistor area III is NMOS area.Wherein, the pull-up Transistor area I, transmission gate transistor area II and pull-down transistor area III are memory block, to be subsequently formed SRAM provides technique platform.
The p-type logic device area includes several p-type threshold voltage areas, wherein, the p-type threshold value Voltage zone includes p-type ultralow threshold value voltage zone (ULVT, Ultra-low VT) 11, p-type level threshold value electricity Nip (Standard VT) 12 and p-type high threshold voltage (High VT) area 13, each region is formed P-type logical device threshold voltage being ordered as from low to high:P-type ultralow threshold value voltage zone 11, P Type standard threshold voltage area 12, p-type high threshold voltage area 13.The p-type logic device area can also be wrapped Include p-type low-threshold power nip (not shown), p-type input and output device area (IO, Input Output) (not Diagram).
The N-type logic device area includes several N-type threshold voltage areas, wherein, the N-type threshold value Voltage includes N-type ultralow threshold value voltage zone 21, N-type standard threshold voltage area 22 and N-type high threshold Voltage zone 23, the threshold voltage of the N-type logical device that each region is formed being ordered as from low to high:N-type Ultralow threshold value voltage zone 21, N-type standard threshold voltage area 22, N-type high threshold voltage area 23.The N Type logic device area can also include N-type low-threshold power nip (not shown), N-type input and output device area (not shown).
So that the semiconductor devices for being formed is fin field effect pipe as an example, the substrate includes substrate to the present embodiment 101st, positioned at the discrete fin 102 on the surface of substrate 101.
In another embodiment, the semiconductor devices is planar transistor, and the substrate is planar substrates, The planar substrates are silicon substrate, germanium substrate, silicon-Germanium substrate or silicon carbide substrates, silicon-on-insulator substrate Or germanium substrate on insulator, glass substrate or III-V substrate (such as gallium nitride substrate or arsenic Gallium substrate etc.), grid structure is formed at the plane.
The material of the substrate 101 is silicon, germanium, SiGe, carborundum, GaAs or gallium indium, institute It can also be the silicon substrate on insulator or the germanium substrate on insulator to state substrate 101;The fin 102 Material include silicon, germanium, SiGe, carborundum, GaAs or gallium indium.It is described in the present embodiment Substrate 101 is silicon substrate, and the material of the fin 102 is silicon.
In the present embodiment, forming the substrate 101, the processing step of fin 102 includes:Initial lining is provided Bottom;Patterned hard mask layer 103 is formed in the initial substrate surface;It is with the hard mask layer 103 Initial substrate described in mask etching, the initial substrate after etching as substrate 101, positioned at the surface of substrate 101 Projection as fin 102.
In one embodiment, the processing step for forming the hard mask layer 103 includes:It is initially formed just Begin hard mask;Patterned photoresist layer is formed in the initial hard mask surface;With described patterned Photoresist layer is initial hard mask described in mask etching, and hard mask layer 103 is formed in initial substrate surface;Go Except the patterned photoresist layer.
In other embodiments, the formation process of the hard mask layer can also include:Autoregistration double picture Shape (SADP, Self-aligned Double Patterned) technique, autoregistration are triple graphical (Self-aligned Triple Patterned) technique or graphical (the Self-aligned Double of autoregistration quadruple Double Patterned) technique.The Dual graphing technique includes LELE (Litho-Etch-Litho-Etch) technique or LLE (Litho-Litho-Etch) technique.
In the present embodiment, after the fin 102 is formed, retain positioned at the top surface of fin 102 Hard mask layer 103.The material of the hard mask layer 103 is silicon nitride, subsequently when flatening process is carried out, The top surface of the hard mask layer 103 can play protection fin as the stop position of flatening process The effect at 102 tops.In the present embodiment, the top dimension of the fin 102 is less than bottom size.At it In his embodiment, the side wall of the fin can also be perpendicular with substrate surface, i.e. the top dimension of fin Equal to bottom size.
With reference to Fig. 2, the barrier film 104 on the covering surface of substrate 101 and the surface of fin 102 is formed, The top of the barrier film 104 is higher than the top of hard mask layer 103.
Before the barrier film 104 is formed, also including step:To the substrate 101 and fin 102 Oxidation processes are carried out, liner oxidation layer is formed on the surface of the substrate 101 and the surface of fin 102.
The barrier film 104 provides Process ba- sis to be subsequently formed separation layer;The material of the barrier film 104 Expect to be insulating materials, for example, silica, silicon nitride or silicon oxynitride.In the present embodiment, the isolation The material of film 104 is silica.
Filling perforation (gap-filling) ability that the technique of barrier film 104 is formed to improve, using mobility Vapour deposition (FCVD, Flowable CVD) or vertical width high are learned than chemical vapor deposition method (HARP CVD), the barrier film 104 is formed.
After the barrier film 104 is formed, also including step:The barrier film 104 is annealed Treatment, improves the consistency of the barrier film 104.
With reference to Fig. 3, the barrier film 104 (referring to Fig. 2) for removing segment thickness forms separation layer 114, institute Separation layer 114 is stated positioned at the surface of substrate 101 and covering fin 102 partial sidewall surface, the separation layer 114 tops are less than the top of fin 102.
The material of the separation layer 114 is silica, silicon nitride or silicon oxynitride.In the present embodiment, institute The material for stating separation layer 114 is silica.
In one embodiment, using dry etch process, the barrier film 104 of etching removal segment thickness. In another embodiment, using wet-etching technology, the barrier film 104 of etching removal segment thickness.
Also include step:The etching removal hard mask layer 103 (referring to Fig. 2).Step can also be included: Screen layer, the screen are formed at the top of the fin 102 and sidewall surfaces and the surface of separation layer 114 The material of layer is covered for silica or silicon oxynitride, its role is to:During follow-up doping treatment, The screen layer can reduce the lattice damage that doping treatment is caused to fin 102.
Also include step:N-type well region is carried out to the p-type logic device area and the area I that pulls up transistor Doping treatment, N-type trap is formed in the p-type logic device area and the substrate for pulling up transistor area I Area;P is carried out to the N-type logic device area, transmission gate transistor area II and pull-down transistor area III Type well region doping treatment, in the N-type logic device area, transmission gate transistor area II and lower crystal pulling P type trap zone is formed in the substrate of area under control III.
With reference to Fig. 4, the substrate to the area I that pulls up transistor carries out first threshold voltage regulation doping treatment.
In the present embodiment, the area I that pulls up transistor for PMOS area, adjust by the first threshold voltage The Doped ions for saving doping treatment are N-type ion, and N-type ion is P, As or Sb.The first threshold Voltage-regulation doping treatment is actually to the channel region below the grid structure that pulls up transistor that is subsequently formed The doping for carrying out, in the present embodiment, the fin 102 to the area I that pulls up transistor carries out first threshold voltage tune Section doping treatment.
Substrate to the area I that pulls up transistor carries out the technique step of first threshold voltage regulation doping treatment Suddenly include:The first graph layer 105 is formed on the surface of the separation layer 114 and the surface of fin 102, it is described First graph layer 105 exposes the area's I substrate surfaces that pull up transistor;It is to cover with first graph layer 105 Film, the fin 102 to the area I that pulls up transistor carries out N-type ion implanting;Then, remove described First graph layer 105.
In the present embodiment, subsequently pull up transistor in area's I substrates formed the equivalent work content of pull-up work-function layer Numerical value is higher, specifically, rear extended meeting forms the different P of equivalent work function value in GePXing threshold voltages area Type work-function layer, wherein, the maximum P-type workfunction layer of equivalent work function value is the first P-type workfunction layer, And the pull-up work-function layer being subsequently formed in the present embodiment is equal with the material and thickness of the first P-type workfunction layer It is identical.Therefore for pulling up transistor, the equivalent work function of the middle pull-up work-function layer that pulls up transistor Value is larger, in order that what is formed pulls up transistor with fixed threshold voltage, to pull-up in the present embodiment The Doped ions concentration of the first threshold voltage regulation doping treatment that transistor area I substrates are carried out should be higher.This In embodiment, the Doped ions of the first threshold voltage doping treatment are As, and doping concentration is 1E12atom/cm3To 1E14atom/cm3
With mixing for the threshold voltage adjustments doping treatment for being carried out to the substrate in the area that pulls up transistor in the prior art Miscellaneous concentration is compared, and the first threshold voltage regulation carried out to the area I substrates that pull up transistor in the present embodiment is mixed The Doped ions concentration for living together reason is higher, it is also possible to think, pull up transistor area's I channel regions in the present embodiment Doped ions concentration it is higher, therefore the saturation current for pulling up transistor that is correspondingly formed of the present embodiment and open State electric current is lower so that pulling up transistor for formation has lower operating current.
In a specific embodiment, the first threshold voltage regulation is carried out using ion implantation technology to mix Reason is lived together, the technological parameter of the first threshold voltage regulation doping treatment includes:Injection ion is As, Implantation Energy is 5kev to 15kev, and implantation dosage is 1E12atom/cm2To 1E14atom/cm2, injection Angle is 0 degree to 15 degree, and twist angles are 23 degree, and injection number of times is 4 times.
Also include step:The p-type logic device area substrate is carried out at N-type threshold voltage adjustments doping Reason.Specifically, to the p-type ultralow threshold value voltage zone 11, p-type standard threshold voltage area 12 and P The substrate in type high threshold voltage area 13 carries out N-type threshold voltage adjustments doping treatment.According to several p-types Threshold voltage ranges needed for the device that threshold voltage area is formed, it is determined that carrying out N to GePXing threshold voltages area The doping concentration of type threshold voltage adjustments doping treatment.In the present embodiment, to p-type ultralow threshold value electricity The N-type threshold value that nip 11, p-type standard threshold voltage area 12 and p-type high threshold voltage area 13 are carried out The doping concentration of voltage-regulation doping treatment is differed.
With reference to Fig. 5, the substrate to the transmission gate transistor area II carries out second threshold voltage regulation doping Treatment.
In the present embodiment, the transmission gate transistor area II is NMOS area, the second threshold voltage The Doped ions for adjusting doping treatment are p-type ion, and p-type ion is B, Ga or In.Second threshold Value regulation doping treatment is actually to the channel region below the transmission gate transistor grid structure that is subsequently formed The doping for carrying out, in the present embodiment, the fin 102 to transmission gate transistor area II carries out Second Threshold electricity Pressure regulation doping treatment.
Substrate to the transmission gate transistor area II carries out the technique that second threshold voltage adjusts doping treatment Step includes:Second graph layer 106, institute are formed on the surface of the separation layer 114 and the surface of fin 102 State the substrate surface that second graph layer 106 exposes transmission gate transistor area II;With second graph layer 106 is mask, and the fin 102 to the transmission gate transistor area II carries out p-type ion implanting;Then, Remove the second graph layer 106.
In the present embodiment, the transmission gate work content for subsequently being formed on transmission gate transistor area II gate dielectric layers surface Several layers of equivalent work function value are higher, specifically, rear extended meeting forms equivalent work content in GeNXing threshold voltages area The different N-type workfunction layer of numerical value, wherein, the maximum N-type workfunction layer of equivalent work function value is first N-type workfunction layer, and the transmission gate work-function layer being subsequently formed in the present embodiment and the first N-type work function The material and thickness all same of layer.
Therefore for transmission gate transistor, transmission gate work-function layer is equivalent in transmission gate transistor Work function value is larger, in order that the transmission gate transistor for being formed has fixed threshold voltage, the present embodiment In transmission gate transistor area II substrates are carried out second threshold voltage regulation doping treatment Doped ions it is dense Degree should be relatively low.In the present embodiment, the Doped ions of the second threshold voltage regulation doping treatment are B, Doping concentration is 1E12atom/cm3To 1E14atom/cm3
With the threshold voltage adjustments doping treatment for being carried out to the substrate of transmission gate transistor area in the prior art Doping concentration compares, the second threshold voltage carried out to transmission gate transistor area II substrates in the present embodiment The Doped ions concentration for adjusting doping treatment is lower, it is also possible to think, transmission gate transistor in the present embodiment The Doped ions concentration of area's II channel regions is lower, therefore the transmission gate transistor that is correspondingly formed of the present embodiment Saturation current and ON state current are lower so that the transmission gate transistor of formation has lower operating current.
In a specific embodiment, the second threshold voltage regulation is carried out using ion implantation technology to mix Reason is lived together, the technological parameter of the second threshold voltage regulation doping treatment includes:Injection ion is B, Implantation Energy is 2kev to 5kev, and implantation dosage is 1E12atom/cm2To 1E14atom/cm2, injection Angle is 0 degree to 15 degree, and injection twist angles are 23 degree, and injection number of times is 4 times.
Also include step:Substrate to the pull-down transistor area III carries out the doping of the 3rd threshold voltage adjustments Treatment, the Doped ions of the 3rd threshold voltage adjustments doping treatment are p-type ion;To the N-type Logic device area substrate carries out p-type threshold voltage adjustments doping treatment.Specifically, ultralow to the N-type The base in threshold voltage adjustments area 21, N-type standard threshold voltage area 22 and N-type high threshold voltage area 23 Bottom carries out p-type threshold voltage adjustments doping treatment.According to the device that several N-type threshold voltage areas are formed Required threshold voltage ranges, it is determined that carrying out p-type threshold voltage adjustments doping to GeNXing threshold voltages area The doping concentration for the treatment of.In the present embodiment, to the N-type ultralow threshold value voltage zone 21, N-type standard threshold The p-type threshold voltage adjustments doping treatment that threshold voltage area 22 and N-type high threshold voltage area 23 are carried out Doping concentration is differed.
Extended meeting afterwards forms grid structure in substrate surface, in this implementation, with using rear grid technique (gate last) Grid structure is formed as an example, forming grid after source-drain area (S/D, Source/Drain) is formed Structure.In other embodiments, additionally it is possible to which grid structure is formed using first grid technique (gate first), exist Grid structure is formed before forming source-drain area.
With reference to Fig. 6, the area I that pulls up transistor, p-type logic device area, N-type logic device area, Transmission gate transistor area II and pull-down transistor area II substrate surfaces form pseudo- oxide-film;In the pseudo- oxygen Change film surface and form pseudo- grid film;The graphical pseudo- grid film and pseudo- oxide-film, form and are located at N-type logic Device region, p-type logic device area, the area I that pulls up transistor, transmission gate transistor area II and lower crystal pulling The pseudo- oxide layer 201 on area under control III part of substrate surface, forms the pseudo- gate layer 202 positioned at the surface of oxide layer 201.
The pseudo- gate layer 202 occupies the locus of the grid structure being subsequently formed.The pseudo- oxide layer 201 Material be silica or silicon oxynitride, the material of the pseudo- gate layer 202 is polysilicon, non-crystalline silicon or nothing Setting carbon.In the present embodiment, the material of the pseudo- oxide layer 201 is silica, the pseudo- gate layer 202 Material be polysilicon.
Also include step:Offset side wall is formed in the sidewall surfaces of pseudo- gate layer 202;To the pseudo- gate layer The N-type logic device area fin 102 of 202 both sides carries out that treatment is lightly doped, and forms N-type LDD region domain, In the present embodiment, including the fin 102 in N-type logic device area Zhong GeNXing threshold voltages area is carried out gently Doping treatment;P-type logic device area fin 102 to the both sides of pseudo- gate layer 202 carries out that place is lightly doped Reason, forms p-type LDD region domain, in the present embodiment, including to each p-type threshold value in p-type logic device area The fin 102 of voltage zone carries out that treatment is lightly doped;To the area I that pulls up transistor of the both sides of pseudo- gate layer 202 Fin 102 carries out that treatment is lightly doped, and forms pull-up LDD region domain;To the biography of the both sides of pseudo- gate layer 202 Sending transistor area II fin 102 carries out that treatment is lightly doped, and forms transmission gate LDD region domain;To the puppet The pull-down transistor area III fins 102 of the both sides of gate layer 202 carry out that treatment is lightly doped, and form drop-down LDD Region.
Also include step:Master wall is formed in the offset side wall sidewall surfaces;To the pseudo- gate layer 202 The N-type logic device area fin 102 of both sides carries out heavy doping treatment, forms N-type S/D regions, this reality In applying example, including heavy doping is carried out to the fin 102 in N-type logic device area Zhong GeNXing threshold voltages area Treatment;P-type logic device area fin 102 to the both sides of pseudo- gate layer 202 carries out heavy doping treatment, P-type S/D regions are formed, in the present embodiment, including to each p-type threshold voltage in p-type logic device area The fin 102 in area carries out heavy doping treatment;To the area's I fins that pull up transistor of the both sides of pseudo- gate layer 202 102 carry out heavy doping treatment, form pull-up S/D regions;Transmission gate to the both sides of pseudo- gate layer 202 is brilliant Body area under control domain II fin 102 carries out heavy doping treatment, forms transmission gate S/D regions;To the pseudo- gate layer The pull-down transistor area III fins 102 of 202 both sides carry out heavy doping treatment, form drop-down S/D regions.
With reference to Fig. 7, the pseudo- gate layer 202 (referring to Fig. 6) and pseudo- oxide layer 201 (referring to Fig. 6) are removed.
Before the pseudo- gate layer 202 is removed, also including step:Interlayer is formed in the substrate surface to be situated between Matter layer (not shown), the sidewall surfaces of the pseudo- gate layer 202 of interlayer dielectric layer covering.
Using dry etch process, wet-etching technology or SiCoNi etching systems, the etching removal puppet Gate layer 202 and pseudo- oxide layer 201.In the technical process for removing the pseudo- gate layer 202, the pseudo- oxidation Layer 201 plays a part of to protect fin 102.
Then, with reference to Fig. 8, in the N-type logic device area, p-type logic device area, pull up transistor Area I, transmission gate transistor area II and pull-down transistor area III substrate surfaces form boundary layer 204.
The boundary layer 204 as the gate dielectric layer being subsequently formed a part, the boundary layer 204 Material is silica or silicon oxynitride.In the present embodiment, the boundary layer 204 is formed using oxidation technology, The oxidation technology is dry-oxygen oxidation, wet-oxygen oxidation or steam oxidation, and the boundary layer 204 of formation is only located at The top surface of fin 102 and sidewall surfaces for exposing.
In other embodiments, the boundary layer is formed using depositing operation, the depositing operation is chemistry Vapour deposition, physical vapour deposition (PVD) or ald, the boundary layer of formation are also located at insulation surface.
With continued reference to Fig. 8, high-k gate dielectric layer 205 is formed on the surface of the boundary layer 204.
In the present embodiment, the high-k gate dielectric layer 205 is also located at the surface of separation layer 114 and interlayer is situated between Matter layer (not shown) sidewall surfaces.The material of the high-k gate dielectric layer 205 is high-k gate dielectric material, Wherein, high-k gate dielectric material refers to grid of the relative dielectric constant more than silica relative dielectric constant Dielectric material, the material of the high-k gate dielectric layer 205 is HfO2、HfSiO、HfSiON、HfTaO、 HfTiO、HfZrO、ZrO2Or Al2O3
The k grid high are formed using chemical vapor deposition, physical vapour deposition (PVD) or atom layer deposition process to be situated between Matter layer 205.In the present embodiment, the material of the high-k gate dielectric layer 205 is HfO2, the k grid high The thickness of dielectric layer 205 is 5 angstroms to 15 angstroms, and the high-k gate dielectric is formed using atom layer deposition process Layer 205.
The lamination knot of the boundary layer 204 and the high-k gate dielectric layer 205 positioned at the surface of boundary layer 204 Structure the N-type logic device area, p-type logic device area, pulls up transistor as gate dielectric layer Area I, transmission gate transistor area II and pull-down transistor area III substrate surfaces are formed with gate dielectric layer.Tool , in the present embodiment, the gate dielectric layer is across fin 102, and covering fin 102 atop part surface for body And sidewall surfaces.
It is follow-up also to form N-type workfunction layer on N-type logic device area gate dielectric layer surface, patrolled in p-type Collect device region gate dielectric layer surface and form P-type workfunction layer.The present embodiment will be being initially formed p-type logic device The P-type workfunction layer in part area, the rear N-type workfunction layer for forming N-type logic device area are carried out as an example Describe in detail.In other embodiments, additionally it is possible to be initially formed N-type logic device area N-type workfunction layer, The P-type workfunction layer of p-type logic device area is formed afterwards.
With reference to Fig. 9, P-type workfunction layer 208 is formed on the p-type logic device area gate dielectric layer surface.
In the present embodiment, the P-type workfunction layer of formation is also located at the area I gate dielectric layer tables that pull up transistor Face, the P-type workfunction layer 208 be also located at N-type logic device area, transmission gate transistor area II and The gate dielectric layer surface of pull-down transistor area III.
Before the P-type workfunction layer 208 is formed, also including step:In the high-k gate dielectric layer 205 surfaces form cap (not shown);Etching stop layer (not shown) is formed in the block layer surface.
The cap plays a part of to protect high-k gate dielectric layer 205, prevents follow-up etching technics pair High-k gate dielectric layer 205 causes unnecessary etching to lose, the cap also help barrier metal from Son spreads in high-k gate dielectric layer 205.The material of the cap is TiN;Using chemical vapor deposition Technique, physical gas-phase deposition or atom layer deposition process form the cap.
The etching stop layer and the P-type workfunction layer 208 for being formed and the N-type work function being subsequently formed The material of layer is different, so that the etching technics of subsequent etching P-type workfunction layer 208 is to etching stopping The etch rate of layer is small, and the etching technics of subsequent etching N-type workfunction layer is fast to the etching of etching stop layer Rate is small, so as to avoid causing etching injury to high-k gate dielectric layer 205.In the present embodiment, the etching The material of stop-layer is TaN, and the etching stop layer is formed using atom layer deposition process.
The material of the P-type workfunction layer 208 is the one kind or several in Ta, TiN, TaSiN or TiSiN Kind.The P is formed using chemical vapor deposition method, physical gas-phase deposition or atom layer deposition process Type work-function layer 208.
In the present embodiment, the material of the P-type workfunction layer 208 is TiN, the P-type workfunction layer 208 have the 3rd thickness, and the 3rd thickness is 45 angstroms to 55 angstroms, for example, 50 angstroms.
In the present embodiment, because p-type logic device area includes p-type ultralow threshold value voltage zone 11, p-type mark Quasi- threshold voltage area 12 and p-type high threshold voltage area, in order to meet device requirement, p-type logical device Difference between the threshold voltage of the device that each region in area is formed is larger, only relies on the foregoing p-type threshold for carrying out Threshold voltage regulation doping treatment is difficult to obtain the threshold voltage of larger difference.Therefore, the present embodiment subsequently enters One step is performed etching to the P-type workfunction layer 208 in p-type high threshold voltage area 13, thinning p-type high threshold The P-type workfunction layer 208 of voltage zone 13, so that the p-type work content in p-type high threshold voltage area 13 Several layers of equivalent work function value reduces, and then further increases the device that p-type high threshold voltage area 13 is formed Threshold voltage numerical value, so that between the threshold voltage of the device of each region formation of p-type logic device area Difference it is larger.
With reference to Figure 10, the second mask layer 209 is formed on the surface of the P-type workfunction layer 208, described the Two mask layers expose the surface of P-type workfunction layer 208 in p-type high threshold voltage area 13;With described second Mask layer 209 is mask, p-type work(of the etching removal positioned at the second thickness in p-type high threshold voltage area 13 Function layer 208.
Second mask layer 209 also covers the surface of P-type workfunction layer 208 of the area I that pulls up transistor, also Positioned at the p-type work function of transmission gate transistor area II, N-type logic device area and pull-down transistor area III 208 surface of layer.In other embodiments, second mask layer also exposes transmission gate transistor, N Type logic device area and the P-type workfunction layer surface in pull-down transistor area so that subsequently remove transmission gate The technique duration of the P-type workfunction layer of transistor, N-type logic device area and pull-down transistor area is shorter.
In the present embodiment, the material of second mask layer 209 is Other substrate materials.In other embodiment In, the material of second mask layer can also be silicon nitride or boron nitride.
Using dry etch process, wet-etching technology or SiCoNi etching systems, etching removal p-type is high The P-type workfunction layer 208 of the second thickness in threshold voltage area 13.
After the completion of etching technics, described several corresponding P-type workfunction layers 208 in p-type threshold voltage area Equivalent work function value it is different, wherein, the maximum P-type workfunction layer 208 of equivalent work function value is a P Type work-function layer 218, therefore, the P-type workfunction layer 208 not being etched in p-type logic device area is the One P-type workfunction layer 218, specific to the present embodiment in, the correspondence of p-type ultralow threshold value voltage zone 11 P-type workfunction layer 208 be the first P-type workfunction layer 218.In described several p-type threshold voltages In the corresponding P-type workfunction layer 208 in area, the thickness of first P-type workfunction layer 218 is most thick.
The surface of P-type workfunction layer 208 of the area I that pulls up transistor also is covered due to the second mask layer 209, is made The P-type workfunction layer 208 of area I of must pulling up transistor is not etched yet, and the area I that pulls up transistor is not etched P-type workfunction layer 208 for pull-up work-function layer 228.Therefore, in the present embodiment, the crystal pulling on described The gate dielectric layer surface of body area under control I forms pull-up work-function layer 228, and the pull-up work-function layer 228 with The material of the first P-type workfunction layer 218 is identical with thickness.And the present embodiment is in the processing step along with The pull-up P-type workfunction layer 218 of work-function layer 228 and first is formed, without to form the pull-up Work-function layer 228 and use extra light shield.
In the present embodiment, the thickness of first P-type workfunction layer 218 is 45 angstroms to 55 angstroms, for example It is 50 angstroms;The thickness of the pull-up work-function layer 228 of the area I that pulls up transistor is 45 angstroms to 55 angstroms, For example, 50 angstroms;The thickness of the P-type workfunction layer 208 after being etched is 25 angstroms to 35 angstroms, for example, 30 angstroms, i.e. in the p-type logic device area, in addition to first P-type workfunction layer 218 The thickness of P-type workfunction layer 208 is 25 angstroms to 35 angstroms.
In the present embodiment, due to the equivalent work function value choosing of the pull-up work-function layer 228 of the area I that pulls up transistor What is taken is:The maximum equivalent work function value of corresponding several P-type workfunction layers in p-type logic device area, Therefore, in order that pulling up transistor with fixed threshold voltage for the area I formation that pulls up transistor, foregoing right The doping concentration of the first threshold voltage regulation doping treatment that the area I substrates that pull up transistor are carried out is high so that shape Into the channel region that pulls up transistor doping concentration it is high, therefore the saturation current that pulls up transistor and ON state electricity Stream is small.
In other embodiments, additionally it is possible to in p-type logic device area except p-type high threshold voltage area its The P-type workfunction layer in his p-type threshold voltage area perform etching it is thinning, and to other p-type threshold voltage areas P-type workfunction layer perform etching thinning thickness and can also differ, it is ensured that in p-type logic device area The maximum P-type workfunction layer of equivalent work function value is the first P-type workfunction layer, and pulls up work-function layer Material and thickness are identical with the material and thickness of the first P-type workfunction layer, it is also possible to think, for For material identical P-type workfunction layer, the most thick P-type workfunction layer of thickness in p-type logic device area It is the first P-type workfunction layer.
Then, with reference to Figure 11, second mask layer 209 (referring to Figure 10) is removed;Etching removal is located at The P-type workfunction layer of N-type logic device area, transmission gate transistor area II and pull-down transistor area III 208。
Specifically, in the surface of P-type workfunction layer 208 of the p-type logic device area and upper crystal pulling The surface of pull-up work-function layer 228 of area under control I forms the 3rd mask layer (not shown), and the 3rd mask is sudden and violent Expose the p-type work content of N-type logic device area, transmission gate transistor area II and pull-down transistor area III Several layers of 208 surface;With the 3rd mask layer as mask, etching removal be located at N-type logic device area, The P-type workfunction layer 208 of transmission gate transistor area II and pull-down transistor area III;Then, institute is removed State the 3rd mask layer.
In other embodiments, additionally it is possible to which first removal is located at N-type logic device area, transmission gate transistor area And the P-type workfunction layer in pull-down transistor area, after to the p-type work function of the p-type logic device area Layer is performed etching.
With reference to Figure 12, N-type workfunction layer 211 is formed on the N-type logic device area gate dielectric layer surface.
In the present embodiment, the N-type workfunction layer 211 of formation is also located at transmission gate transistor area II grid Dielectric layer surface, the N-type workfunction layer 211 is also located at the surface of P-type workfunction layer 208, a P The gate medium on the surface of type work-function layer 218, the pull-up surface of work-function layer 228 and pull-down transistor area III Layer surface.
The material of the N-type workfunction layer 211 be TiAl, TiAlC, TaAlN, TiAlN, MoN, One or more in TaCN or AlN.Using chemical vapor deposition method, physical gas-phase deposition or Atom layer deposition process forms the N-type workfunction layer 211.
In the present embodiment, the material of the N-type workfunction layer 211 is TiAlC, the N-type work function Layer 211 has the 4th thickness, and the 4th thickness is 45 angstroms to 55 angstroms, for example, 50 angstroms.
Because N-type logic device area includes N-type ultralow threshold value voltage zone 21, N-type standard threshold voltage area 22 and N-type high threshold voltage area 23, in order to meet device requirement, each region shape of N-type logic device area Into device threshold voltage between difference it is larger, only rely on the foregoing N-type threshold voltage adjustments for carrying out and mix Live together the threshold voltage that reason is difficult to obtain larger difference.
Therefore, the present embodiment is follow-up further to the N-type workfunction layer of N-type high threshold region 23 211 perform etching, the N-type workfunction layer 211 in thinning N-type high threshold voltage area 23, and then further Increase the device threshold voltage numerical value that N-type high threshold voltage area 23 is formed, so that N-type logic device Difference between the threshold voltage of the device that each region in part area is formed is larger.
With reference to Figure 13, the first mask layer 212 is formed on the surface of the N-type workfunction layer 211, described the One mask layer 212 exposes the surface of N-type workfunction layer 211 in N-type high threshold voltage area 23;With institute It is mask to state the first mask layer 212, and etching removal is positioned at the second thickness in N-type high threshold voltage area 23 N-type workfunction layer 211.
First mask layer 212 also exposes the table of N-type workfunction layer 211 of transmission gate transistor area II Face, and also cover the N-type of pull up transistor area I, pull-down transistor area III and p-type logic device area The surface of work-function layer 211.In other embodiments, first mask layer can also expose crystal pulling The N-type workfunction layer surface in body area under control, pull-down transistor area or p-type logic device area.
In the present embodiment, the material of first mask layer 212 is Other substrate materials.In other embodiment In, the material of first mask layer can also be silicon nitride or boron nitride.
Using dry etch process, wet-etching technology or SiCoNi etching systems, etching removal N-type is high The N-type workfunction layer 211 of the first thickness in threshold voltage area 23.
After the completion of etching, several corresponding N-type workfunction layers 211 in N-type threshold voltage area Equivalent work function value is different, wherein, the maximum N-type workfunction layer 211 of equivalent work function value is a N Type work-function layer 221, therefore, the N-type workfunction layer 211 after being etched in N-type logic device area is the One N-type workfunction layer 221.In specific to the present embodiment, the N-type high threshold voltage area 23 is corresponding N-type workfunction layer 211 is the first N-type workfunction layer 221.In described several N-type threshold voltage areas In corresponding N-type workfunction layer 211, the thickness of first N-type workfunction layer 221 is most thin.
The table of N-type workfunction layer 211 of transmission gate transistor area II is also exposed due to the first mask layer 212 Face, the N-type workfunction layer 211 of the first thickness in N-type high threshold voltage area 23 is located in etching removal While, also etching removal is located at the N-type of the first thickness on transmission gate transistor area II gate dielectric layers surface Work-function layer 211, the N-type workfunction layer 211 after being etched in the II of transmission gate transistor area is transmission gate work( Function layer 231.
Therefore, in the present embodiment, transmission gate is formed on the transmission gate transistor area II gate dielectric layers surface Work-function layer 231, and the N-type workfunction layer 221 of the transmission gate work-function layer 231 and first material It is identical with thickness.And in the present embodiment in the processing step along with, form the transmission gate work-function layer 231 and first N-type workfunction layer 221, without being used to form the transmission gate work-function layer 231 Extra light shield.
In the N-type logic device area, the N-type work(in addition to first N-type workfunction layer 221 The thickness of function layer 211 is 45 angstroms to 55 angstroms, for example, 50 angstroms;First N-type workfunction layer 221 thickness is 25 angstroms to 35 angstroms, for example, 30 angstroms;The thickness of the transmission gate work-function layer 231 It is 25 angstroms to 35 angstroms, for example, 30 angstroms.
In the present embodiment, due to the equivalent work content of the transmission gate work-function layer 231 of transmission gate transistor area II Numerical value choose be:The maximum equivalent work(of corresponding several N-type workfunction layers in N-type logic device area Functional value, therefore, in order that the transmission gate transistor that transmission gate transistor area II is formed has fixed threshold Threshold voltage, the foregoing second threshold voltage regulation doping treatment carried out to transmission gate transistor area II substrates Doping concentration is low so that the doping concentration of the transmission gate transistor channel region of formation is low, follows transmission gate brilliant The saturation current and ON state current of body pipe are big.
In other embodiments, additionally it is possible to the N in other N-type threshold voltage areas in N-type logic device area Type work-function layer performs etching thinning, and N-type workfunction layer to other N-type threshold voltage areas carves The thinning thickness of erosion can also be differed, it is ensured that the maximum N-type of the equivalent work function value of N-type logic device area Work-function layer be the first N-type workfunction layer, and transmission gate work-function layer material and thickness and the first N-type The material of work-function layer is identical with thickness, it is also possible to think, for material identical N-type work function For layer, the most thin N-type workfunction layer of thickness is the first N-type workfunction layer in N-type logic device area.
Then, with reference to Figure 14, first mask layer 212 (referring to Figure 13) is removed;Removal is located at pull-up The N-type workfunction layer 211 of transistor area I and p-type logic device area.
Specifically, forming the 4th on the surface of transmission gate work-function layer 231 of the transmission gate transistor area II Mask layer (not shown), the 4th mask layer also covers the N-type workfunction layer of pull-down transistor area III 211 and the N-type workfunction layer 211 of N-type logic device area;With the 4th mask layer as mask, carve Etching off is except the N-type workfunction layer 211 for being located at pull up transistor area I and p-type logic device area;Then, Remove the 4th mask layer.
In other embodiments, moreover it is possible to first remove the N-type work(of pull up transistor area and p-type logic device area Function layer, then performs etching thinning to the N-type workfunction layer in N-type high threshold voltage area.
In another embodiment, because N-type workfunction layer is to pulling up transistor and p-type logical device Threshold voltage influence is smaller, therefore can also retain the N in pull up transistor area and p-type logic device area Type work-function layer.
With reference to Figure 15, on the surface of the N-type workfunction layer 211, the surface of P-type workfunction layer 208, pass The surface of a work-function layer 231 and pull-up work-function layer 228 surface is sent to form gate electrode layer 301.
In the present embodiment, the P-type workfunction layer 208 is included positioned at p-type ultralow threshold value voltage zone 11 First P-type workfunction layer 218, the N-type workfunction layer 211 includes being located at N-type high threshold voltage area 23 the first N-type workfunction layer 221.The gate electrode layer 301 is also located at the N of pull-down transistor area III The surface of type work-function layer 211.
Positioned at the surface of N-type workfunction layer 211, the surface of P-type workfunction layer 208, transmission gate work-function layer The gate electrode layer 301 on 231 surfaces and pull-up work-function layer 228 surface is connected with each other.In other embodiment In, positioned at N-type workfunction layer surface, P-type workfunction layer surface, transmission gate work-function layer surface and The gate electrode layer on pull-up work-function layer surface can also be separate.
The material of the gate electrode layer 301 is included in Al, Cu, Ag, Au, Pt, Ni, Ti or W One or more.
In one embodiment, the processing step for forming the first gate electrode layer 301 includes:Institute State the surface of N-type workfunction layer 211, the surface of P-type workfunction layer 208, the table of transmission gate work-function layer 231 Face and pull-up work-function layer 228 surface form gate electrode film, and the gate electrode film top is situated between higher than interlayer Matter layer top;Grinding removal forms the gate electrode layer 301 higher than the gate electrode film at the top of interlayer dielectric layer.
From Such analysis, the present embodiment formed the saturation current for pulling up transistor and ON state current compared with It is low, and the saturation current and ON state current of the transmission gate transistor for being formed are higher, due to the gamma of memory Than the ratio pass in direct ratio between transmission gate transistor ON state current and the ON state current that pulls up transistor System, therefore the gamma of memory that the present embodiment is formed is than larger, and then cause the write-in redundancy of memory Degree is improved, and the performance of corresponding memory is improved, and the yield of such as memory is improved, And then improve the performance of the semiconductor devices to be formed.
Also, the present embodiment formed logical device while form memory in pull up transistor, Pull-down transistor and transmission gate transistor so that the memory of formation is with larger write-in redundancy Meanwhile, the technique for forming memory is mutually compatible with the technique for forming logical device, saves processing step, Without introducing extra light shield to improve the write-in redundancy of memory, save semiconductor production into This.
Although present disclosure is as above, the present invention is not limited to this.Any those skilled in the art, Without departing from the spirit and scope of the present invention, can make various changes or modifications, therefore guarantor of the invention Shield scope should be defined by claim limited range.

Claims (20)

1. a kind of method of improvement SRAM performances, it is characterised in that including:
Substrate is provided, the substrate includes N-type logic device area, p-type logic device area, upper crystal pulling Area under control and transmission gate transistor area, wherein, the N-type logic device area includes several N-type threshold values Voltage zone, the p-type logic device area includes several p-type threshold voltage areas, the N-type logic device Part area, p-type logic device area, the part of substrate surface shape of pull up transistor area and transmission gate transistor area Into there is gate dielectric layer;
P-type workfunction layer, and described several P are formed on the p-type logic device area gate dielectric layer surface The equivalent work function value of the corresponding P-type workfunction layer in type threshold voltage area is different, wherein, equivalent work function It is the first P-type workfunction layer to be worth maximum P-type workfunction layer;
Pull-up work-function layer, and the pull-up work(are formed on the gate dielectric layer surface in the area that pulls up transistor The material and thickness of function layer are identical with the material and thickness of the first P-type workfunction layer;
Substrate to the area that pulls up transistor carries out first threshold voltage regulation doping treatment;
The N-type logic device area gate dielectric layer surface formed N-type workfunction layer, and it is described several The equivalent work function value of the corresponding N-type workfunction layer in N-type threshold voltage area is different, wherein, equivalent work content The maximum N-type workfunction layer of numerical value is the first N-type workfunction layer;
Transmission gate work-function layer, and the biography are formed on the gate dielectric layer surface of the transmission gate transistor area Send the material and thickness of a work-function layer identical with the material and thickness of the first N-type workfunction layer;
Substrate to the transmission gate transistor area carries out second threshold voltage regulation doping treatment;
The N-type workfunction layer surface, P-type workfunction layer surface, transmission gate work-function layer surface with And pull-up work-function layer surface forms gate electrode layer.
2. the method for improving SRAM performances as claimed in claim 1, it is characterised in that the transmission gate crystal Area under control is NMOS area;The area that pulls up transistor is for PMOS area.
3. the method for improving SRAM performances as claimed in claim 1, it is characterised in that in described several P In the corresponding P-type workfunction layer in type threshold voltage area, the thickness of first P-type workfunction layer is most thick; In described several corresponding N-type workfunction layers in N-type threshold voltage area, the first N-type work( The thickness of function layer is most thin.
4. the method for improving SRAM performances as claimed in claim 1, it is characterised in that walked in the technique along with In rapid, pull-up work-function layer and first P-type workfunction layer are formed;In the processing step along with, Form the transmission gate work-function layer and the first N-type workfunction layer.
5. the method for improving SRAM performances as claimed in claim 1, it is characterised in that the p-type work function The material of layer is one or more in Ta, TiN, TaSiN or TiSiN;The N-type work function The material of layer is the one kind in TiAl, TiAlC, TaAlN, TiAlN, MoN, TaCN or AlN Or it is several.
6. the method for improving SRAM performances as claimed in claim 1, it is characterised in that described several N-types Threshold voltage area includes N-type ultralow threshold value voltage zone, N-type standard threshold voltage area and N-type threshold high Threshold voltage area, wherein, the corresponding N-type workfunction layer in the N-type high threshold voltage area is a N Type work-function layer.
7. the method for improving SRAM performances as claimed in claim 6, it is characterised in that formation includes described the The processing step of the N-type workfunction layer of one N-type workfunction layer includes:In the N-type logical device Area gate dielectric layer surface forms N-type workfunction layer;First is formed on the N-type workfunction layer surface to cover Film layer, first mask layer exposes the N-type workfunction layer surface in N-type high threshold voltage area;With First mask layer is mask, N of the etching removal positioned at the first thickness in N-type high threshold voltage area Type work-function layer, the N-type workfunction layer after being etched in N-type logic device area is the first N-type work function Layer.
8. the method for improving SRAM performances as claimed in claim 6, it is characterised in that the N-type work function Layer is also located at transmission gate transistor area gate dielectric layer surface;And first mask layer also exposes transmission The N-type workfunction layer surface of door transistor area, in etching removal positioned at the of N-type high threshold voltage area While the N-type workfunction layer of one thickness, also etching removal is located at transmission gate transistor area gate dielectric layer The N-type workfunction layer of the first thickness on surface, forms the transmission gate work-function layer.
9. the method for improving SRAM performances as claimed in claim 6, it is characterised in that also including step:It is right The N-type logic device area substrate carries out N-type threshold value regulation doping treatment.
10. the method for improving SRAM performances as claimed in claim 1, it is characterised in that in the N-type logic In device region, the thickness of the N-type workfunction layer in addition to first N-type workfunction layer is 45 angstroms To 55 angstroms.
11. methods for improving SRAM performances as described in claim 1 or 10, it is characterised in that the transmission The thickness of door work-function layer is 25 angstroms to 35 angstroms.
12. methods for improving SRAM performances as claimed in claim 11, it is characterised in that the Second Threshold The Doped ions of voltage-regulation doping treatment are B, and doping concentration is 1E12atom/cm3Extremely 1E14atom/cm3
13. methods for improving SRAM performances as claimed in claim 1, it is characterised in that described several p-types Threshold voltage area includes p-type ultralow threshold value voltage zone, p-type standard threshold voltage area and p-type threshold high Threshold voltage area, wherein, the corresponding P-type workfunction layer in the p-type ultralow threshold value voltage zone is a P Type work-function layer.
14. methods for improving SRAM performances as claimed in claim 13, it is characterised in that formed described in including The processing step of the P-type workfunction layer of the first P-type workfunction layer includes:In the p-type logical device Area gate dielectric layer surface forms P-type workfunction layer;Second is formed on the P-type workfunction layer surface to cover Film layer, second mask layer exposes the P-type workfunction layer surface in p-type high threshold voltage area;With Second mask layer is mask, P of the etching removal positioned at the second thickness in p-type high threshold voltage area Type work-function layer, the P-type workfunction layer not being etched in p-type logic device area is the first p-type work content Several layers.
15. methods for improving SRAM performances as claimed in claim 14, it is characterised in that the p-type work content Several floor are also located at the area gate dielectric layer surface that pulls up transistor, and second mask layer also covers crystal pulling The P-type workfunction layer surface in body area under control, the P-type workfunction layer that the area that pulls up transistor is not etched is upper Draw work-function layer.
16. methods for improving SRAM performances as claimed in claim 1, it is characterised in that in the p-type logic In device region, the thickness of the P-type workfunction layer in addition to first P-type workfunction layer for 25 angstroms extremely 35 angstroms.
17. methods for improving SRAM performances as described in claim 1 or 16, it is characterised in that the pull-up The thickness of work-function layer is 45 angstroms to 55 angstroms.
18. methods for improving SRAM performances as claimed in claim 17, it is characterised in that the first threshold The Doped ions of voltage doping treatment are As, and doping concentration is 1E12atom/cm3Extremely 1E14atom/cm3
19. methods for improving SRAM performances as claimed in claim 1, it is characterised in that the gate dielectric layer bag Include boundary layer and the high-k gate dielectric layer positioned at interface layer surfaces.
20. methods for improving SRAM performances as claimed in claim 1, it is characterised in that the substrate includes lining Bottom and the fin positioned at substrate surface, wherein, the gate dielectric layer is across fin, and covering fin The atop part surface in portion and sidewall surfaces.
CN201510896857.0A 2015-12-07 2015-12-07 Method for improving SRAM performance Active CN106847755B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510896857.0A CN106847755B (en) 2015-12-07 2015-12-07 Method for improving SRAM performance

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510896857.0A CN106847755B (en) 2015-12-07 2015-12-07 Method for improving SRAM performance

Publications (2)

Publication Number Publication Date
CN106847755A true CN106847755A (en) 2017-06-13
CN106847755B CN106847755B (en) 2020-03-10

Family

ID=59151435

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510896857.0A Active CN106847755B (en) 2015-12-07 2015-12-07 Method for improving SRAM performance

Country Status (1)

Country Link
CN (1) CN106847755B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107993925A (en) * 2017-11-27 2018-05-04 长江存储科技有限责任公司 A kind of autoregistration quadruple graph technology
CN109614730A (en) * 2018-12-19 2019-04-12 上海华力集成电路制造有限公司 SRAM cell design method
CN110364530A (en) * 2018-04-11 2019-10-22 中芯国际集成电路制造(上海)有限公司 Memory and forming method thereof

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102171794A (en) * 2008-12-11 2011-08-31 英特尔公司 Graded high germanium compound films for strained semiconductor devices
CN103811326A (en) * 2012-11-06 2014-05-21 格罗方德半导体公司 Metal gate structure for midgap semiconductor device and method of making same
CN104022035A (en) * 2013-02-28 2014-09-03 中芯国际集成电路制造(上海)有限公司 Transistor and forming method thereof
US20140363960A1 (en) * 2013-06-05 2014-12-11 Samsung Electronics Co., Ltd. Semiconductor device and method for fabricating the same
CN104766823A (en) * 2014-01-07 2015-07-08 中国科学院微电子研究所 Semiconductor device manufacturing method
CN104779207A (en) * 2014-01-13 2015-07-15 中芯国际集成电路制造(上海)有限公司 Semiconductor device manufacturing method

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102171794A (en) * 2008-12-11 2011-08-31 英特尔公司 Graded high germanium compound films for strained semiconductor devices
CN103811326A (en) * 2012-11-06 2014-05-21 格罗方德半导体公司 Metal gate structure for midgap semiconductor device and method of making same
CN104022035A (en) * 2013-02-28 2014-09-03 中芯国际集成电路制造(上海)有限公司 Transistor and forming method thereof
US20140363960A1 (en) * 2013-06-05 2014-12-11 Samsung Electronics Co., Ltd. Semiconductor device and method for fabricating the same
CN104766823A (en) * 2014-01-07 2015-07-08 中国科学院微电子研究所 Semiconductor device manufacturing method
CN104779207A (en) * 2014-01-13 2015-07-15 中芯国际集成电路制造(上海)有限公司 Semiconductor device manufacturing method

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107993925A (en) * 2017-11-27 2018-05-04 长江存储科技有限责任公司 A kind of autoregistration quadruple graph technology
CN110364530A (en) * 2018-04-11 2019-10-22 中芯国际集成电路制造(上海)有限公司 Memory and forming method thereof
CN110364530B (en) * 2018-04-11 2021-12-03 中芯国际集成电路制造(上海)有限公司 Memory and forming method thereof
CN109614730A (en) * 2018-12-19 2019-04-12 上海华力集成电路制造有限公司 SRAM cell design method

Also Published As

Publication number Publication date
CN106847755B (en) 2020-03-10

Similar Documents

Publication Publication Date Title
CN108122852B (en) Semiconductor structure and forming method thereof
CN103311247B (en) Semiconductor device and manufacture method thereof
US10262905B2 (en) Simplified multi-threshold voltage scheme for fully depleted SOI MOSFETs
CN108122913B (en) Semiconductor devices and forming method thereof
TWI415263B (en) Semiconductor device and method for fabricating thereof
US8329566B2 (en) Method of manufacturing a high-performance semiconductor device
CN102299156B (en) Semiconductor device and manufacturing method thereof
US8252689B2 (en) Chemical-mechanical planarization method and method for fabricating metal gate in gate-last process
US8691655B2 (en) Method of semiconductor integrated circuit fabrication
CN106558584A (en) Semiconductor structure and forming method thereof
CN108122914A (en) SRAM device and its manufacturing method
US9876089B2 (en) High-k and p-type work function metal first fabrication process having improved annealing process flows
WO2014071754A1 (en) Semiconductor structure and manufacturing method therefor
CN106847755A (en) The method for improving SRAM performances
CN106847874A (en) The forming method of the semiconductor devices with different threshold voltages
US20230378294A1 (en) Triple layer high-k gate dielectric stack for workfunction engineering
US11037831B2 (en) Gate structure and method
CN108074929B (en) Semiconductor structure and its manufacturing method
CN108807535B (en) Fin field effect transistor and forming method thereof
CN108122912B (en) SRAM device and its manufacturing method
CN106847695A (en) The forming method of fin field effect pipe
US20220223693A1 (en) Effective work function tuning via silicide induced interface dipole modulation for metal gates
CN108666271B (en) Semiconductor device and method of forming the same
CN107046005A (en) Improve the method for device performance
CN106935550A (en) Semiconductor structure and its manufacture method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant