Classifications

• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10D—INORGANIC ELECTRIC SEMICONDUCTOR DEVICES
  • H10D64/00—Electrodes of devices having potential barriers
  • H10D64/60—Electrodes characterised by their materials
  • H10D64/66—Electrodes having a conductor capacitively coupled to a semiconductor by an insulator, e.g. MIS electrodes
  • H10D64/661—Electrodes having a conductor capacitively coupled to a semiconductor by an insulator, e.g. MIS electrodes the conductor comprising a layer of silicon contacting the insulator, e.g. polysilicon having vertical doping variation
• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10D—INORGANIC ELECTRIC SEMICONDUCTOR DEVICES
  • H10D64/00—Electrodes of devices having potential barriers
  • H10D64/60—Electrodes characterised by their materials
  • H10D64/66—Electrodes having a conductor capacitively coupled to a semiconductor by an insulator, e.g. MIS electrodes
  • H10D64/661—Electrodes having a conductor capacitively coupled to a semiconductor by an insulator, e.g. MIS electrodes the conductor comprising a layer of silicon contacting the insulator, e.g. polysilicon having vertical doping variation
  • H10D64/662—Electrodes having a conductor capacitively coupled to a semiconductor by an insulator, e.g. MIS electrodes the conductor comprising a layer of silicon contacting the insulator, e.g. polysilicon having vertical doping variation the conductor further comprising additional layers, e.g. multiple silicon layers having different crystal structures
  • H10D64/663—Electrodes having a conductor capacitively coupled to a semiconductor by an insulator, e.g. MIS electrodes the conductor comprising a layer of silicon contacting the insulator, e.g. polysilicon having vertical doping variation the conductor further comprising additional layers, e.g. multiple silicon layers having different crystal structures the additional layers comprising a silicide layer contacting the layer of silicon, e.g. polycide gates
• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10D—INORGANIC ELECTRIC SEMICONDUCTOR DEVICES
  • H10D84/00—Integrated devices formed in or on semiconductor substrates that comprise only semiconducting layers, e.g. on Si wafers or on GaAs-on-Si wafers
  • H10D84/01—Manufacture or treatment
  • H10D84/0123—Integrating together multiple components covered by H10D12/00 or H10D30/00, e.g. integrating multiple IGBTs
  • H10D84/0126—Integrating together multiple components covered by H10D12/00 or H10D30/00, e.g. integrating multiple IGBTs the components including insulated gates, e.g. IGFETs
  • H10D84/0165—Integrating together multiple components covered by H10D12/00 or H10D30/00, e.g. integrating multiple IGBTs the components including insulated gates, e.g. IGFETs the components including complementary IGFETs, e.g. CMOS devices
  • H10D84/0172—Manufacturing their gate conductors
  • H10D84/0177—Manufacturing their gate conductors the gate conductors having different materials or different implants
• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10D—INORGANIC ELECTRIC SEMICONDUCTOR DEVICES
  • H10D84/00—Integrated devices formed in or on semiconductor substrates that comprise only semiconducting layers, e.g. on Si wafers or on GaAs-on-Si wafers
  • H10D84/01—Manufacture or treatment
  • H10D84/02—Manufacture or treatment characterised by using material-based technologies
  • H10D84/03—Manufacture or treatment characterised by using material-based technologies using Group IV technology, e.g. silicon technology or silicon-carbide [SiC] technology
  • H10D84/038—Manufacture or treatment characterised by using material-based technologies using Group IV technology, e.g. silicon technology or silicon-carbide [SiC] technology using silicon technology, e.g. SiGe
• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10D—INORGANIC ELECTRIC SEMICONDUCTOR DEVICES
  • H10D84/00—Integrated devices formed in or on semiconductor substrates that comprise only semiconducting layers, e.g. on Si wafers or on GaAs-on-Si wafers
  • H10D84/80—Integrated devices formed in or on semiconductor substrates that comprise only semiconducting layers, e.g. on Si wafers or on GaAs-on-Si wafers characterised by the integration of at least one component covered by groups H10D12/00 or H10D30/00, e.g. integration of IGFETs
  • H10D84/82—Integrated devices formed in or on semiconductor substrates that comprise only semiconducting layers, e.g. on Si wafers or on GaAs-on-Si wafers characterised by the integration of at least one component covered by groups H10D12/00 or H10D30/00, e.g. integration of IGFETs of only field-effect components
  • H10D84/83—Integrated devices formed in or on semiconductor substrates that comprise only semiconducting layers, e.g. on Si wafers or on GaAs-on-Si wafers characterised by the integration of at least one component covered by groups H10D12/00 or H10D30/00, e.g. integration of IGFETs of only field-effect components of only insulated-gate FETs [IGFET]
  • H10D84/85—Complementary IGFETs, e.g. CMOS
  • H10D84/859—Complementary IGFETs, e.g. CMOS comprising both N-type and P-type wells, e.g. twin-tub
• • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
  • Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
  • Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
  • Y10S148/00—Metal treatment
  • Y10S148/082—Ion implantation FETs/COMs

Definitions

• This invention relates to a method of fabricating very-large-scale-integrated (VLSI) devices of the complementary metal-oxide-semiconductor (CMOS) type.
• VLSI very-large-scale-integrated
• CMOS complementary metal-oxide-semiconductor
• CMOS devices are being regarded as the preferred way to implement VLSI devices.
• Specific examples of CMOS devices suitable for such implementation are described in, for example, articles by R. Jerdonek, M. Ghezzo, J. Weaver and S. Combs, entitled “Reduced Geometry CMOS Technology,” in the International Electron Devices Meeting Digest, page 451 (1981); and by L. C. Parrillo, L. K. Wang, R. D. Swenu son, R. L. Field, R. C. Melin and R. A. Levy, entitled “Twin Tub CMOS II - An Advanced VLSI Technology," in International Electron Devices Meeting Digest, page 706 (1982).
• the threshold voltages of the constituent p- and n-channel transistors of the device should be the respective complements of each other.
• complementary threshold voltages - (V _, V ) are typiclly achieved by utilizing a so-called double-doped-polysilicon process in which the polysilicon gate work function is adjusted by utilizing appropriately doped p and n polysilicon.
• the present invention is an improved method for fabricating VLSI CMOS devices of the type that include silicide-on-polysilicon gate-level metallization and whose constituent transistors, however interconnected, consistently exhibit complementary threshold voltages.
• a specific illustrative embodiment of the present invention comprises a method based on applicants 1 conception of a phenomenological model that provided for the first time an adequate understanding of the aforestated anomalous behavior in such VLSI CMOS devices, devices.
• applicants determined that the anomalously large in the devices is primarily attributable to dopant exchange across the silicide-to- polysilicon interfaces of the composite gates formed in the devices.
• applicants devised a unique fabrication sequence that has been utilized successfully to make VLSI CMOS devices whose constituent transistors consistently exhibit complementary threshold voltages.
• p-channel and n-channel transistors having silicide-on- polysilicon gates are fabricated in a VLSI CMOS device.
• the fabrication sequence is selected to minimize dopant depletion from the p + polysilicon of the p-channel transistors and also to minimize dopant diffusion from the n polysilicon of the n-channel transistors into the p + polysilicon of the p-channel transistors.
• this is accomplished by locking dopants in place in the p and n + polysilicon, by establishing a relatively low dopant concentration in the n polysilicon and by maintaining relatively low temperatures during the entire fabrication sequence so as to inhibit dopant redistribution.
• CMOS device of the type that comprises complementary-threshold-voltage NMOS and PMOS transistors which include silicide-on-doped polysilicon gates.
• the method comprises: forming a polysilicon layer on a substrate, selectively introducing dopants -into specified regions of the layer, forming a -cap- layer on the doped polysilicon layer, and heating the device with the cap layer in place to lock dopants in lattice sites in the specified regions of the polysilicon layer.
• FIG. 1 is a schematic cross-sectional depiction, not drawn to scale, of a portion of a standard VLSI CMOS device known in the art?
• FIGS. 2 through 9 are schematic cross- sectional representations, also not drawn to scale, at various successive steps in a VLSI CMO ' S device fabrication procedure carried out in accordance with the principles of the present invention.
• CMOS devices may be fabricated in either single-tub or twin-tub form.
• emphasis will be directed to CMOS devices made in twin-tub form.
• the descriptive material pertaining to the fabrication of the particular devices described below is also applicable to single-tub CMOS implementations.
• FIG. 1 shows a portion of a priorly known CMOS device configured as an inverter.
• the standard device comprises a lightly doped silicon substrate 10 having p- tub and n-tub regions 12 and 14, respectively, formed therein.
• an n-channel (or NMOS) transistor comprising source and drain regions 16 and 18 is formed in the p-tub 12
• a p-channel (or PMOS) transistor comprising source and drain regions 20 and 22 is formed in the n-tub 14.
• a so-called gate-oxide layer 24 that comprises. for example, a 25-nm-thick layer of silicon dioxide.
• standard field-oxide portions 26, 28 and 30 are formed on the layer 24. These portions are each about 400 nm units thick and also comprise silicon dioxide.
• the n-channel- transistor depicted in the standard device of FIG. 1 includes a composite gate that comprises an n polysilicon portion 32 and a refractory metal silicide portion 34 that is made, for example, of tantalum sili ⁇ ide.
• the p-channel transistor of FIG. 1 includes a composite gate that comprises a p + polysilicon portion 36 and a tantalum silicide portion 38.
• the polysilicon portion 32 is doped with arsenic and the polysilicon portion 36 is doped with boron.
• the gates of respective pairs of complementary transistors are directly connected together.
• this is schematically represented by lead 40 which electrically interconnects the tantalum silicide portions 34 and 38 of the two aforedescribed adjacent n- and p-type transistors.
• the lead 40 comprises a runner made of tantalum silicide over polysilicon.
• FIG. 1 are interconnected via a tantalum silicide runner
• V m T réelleN of the n-channel transistor was measured to be approximately +0.7 volts.
• V of the p-channel transistor in the device was measured to be in the range of -0.8 to -1.7 volts rather than the ideal desired value of approximately -0.7 volts.
• this problem threatened to thwart the successful further development of VLSI CMOS devices at about one-micrometer and smaller design rules.
• CMOS devices Applicants postulated that the aforedescribed anomalous behavior arises primarily due to dopant exchange across the silicide-to-polysilicon interfaces of the composite gates of the herein-considered VLSI CMOS devices. More specifically, applicants determined that, during elevated temperature processing of the CMOS device, boron segregates from the p + polysilicon portions into the overlying tantalum silicide. Moreover, applicants determined that grain-boundary diffusion of arsenic in the silicide overlying the n polysilicon portions results in long-distance transport of arsenic dopant from n polysilicon portions to remote silicide-polysilicon interfaces. At these interfaces, uniform redistribution of the transported arsenic occurs.
• V is thereby established to be considerably greater than the absolute value of V m T_N,.
• a VLSI CMOS device is fabricated in a particular fashion that eliminates or substantially reduces the aforedescribed depletion and diffusion phenomena.
• exactly or nearly exactly complementary transistor threshold voltages are desired in such configurations for optimal circuit operation.
• FIG. 2 shows a part of a specific illustrative VLSI CMOS device some of whose portions are identical to the corresponding portions described above and depicted in FIG. 1.
• These corresponding portions which are identified in FIG. 2 and in subsequent figures by th-e same reference numerals employed therefor in FIG. 1, comprise: substrate 10, p-tub 12, n-tub 14, gate oxide 24 and field oxide portions 26, 28 and 30.
• a 300-to-400 nm-thick layer 42 of polysilicon is formed (for example by low-pressure chemical vapor deposition) on the FIG. 2 device and then selectively implanted with a p-type dopant such as boron.
• this is done by masking the left-hand side of the polysilicon layer 42 with a blocking layer 44 of patterned photoresist and implanting boron (or boron difluoride) into the right-hand side of the layer 42.
• the dose and energy of the boron implant are selected to be about 4E15 15 (4 x 10 ) ions per square centimeter and 30 kilo- electron-volts, respectively.
• the resulting boron ions in the polysilicon layer 42 are schematically depicted in FIG. 2 by plus signs.
• the right-hand side of the layer 42 is saturated with boron dopant.
• p-type dopants are implanted only on the PMOS side of the polysilicon layer 42 of FIG. 2. No such dopants enter the masked or NMOS side of the layer 42. If such dopants were allowed to enter the NMOS side of the layer 42, the subsequent dose of n-type dopants required to dope the NMOS side of the layer 42 to form n polysilicon would have to be higher than desired so as to counterdope the effect of p-type dopants on the NMOS side of the device. But such a higher dose would be inconsistent with applicants' overall goal of maintaining a relatively low concentration of n-type dopant in the device. (Higher concentrations of n-type dopant are instrumental in causing the above-specified anomalous threshold problem.)
• the photoresist layer 44 (FIG. 2) is removed and a patterned blocking layer 46 of photoresist is formed overlying the right-hand or PMOS side of the polysilicon layer 42, as indicated in FIG. 3.
• An n-type dopant such as arsenic is then implanted into the uncovered portion of the layer 42, as schematically depicted by minus signs therein.
• the next step in the above-described fabrication sequence is to remove the photoresist layer 46 shown in FIG. 3.
• a so-called cap layer 48 is then formed on the entire top surface of the polysilicon layer 42, as shown in FIG. 4.
• the cap layer 48 comprises a 30-nm-thick film of silicon dioxide formed during an annealing step at about 900 degrees Celsius for approximately 30 minutes in a dry oxygen atmosphere.
• the primary purpose of the cap layer 48 is to prevent depletion of boron dopant from the polysilicon layer 42 by evaporation or otherwise during a subsequent heating or lock-in step.
• dopants in the polysilicon layer 42 are substantially locked in place during a heating step in which the cap layer 48 remains in place overlying the layer 42.
• this step comprises heating the device to about 950 degrees Celsius in a pure argon atmosphere for approximately 30 minutes.
• the effect of this heating step is to drive dopant into substitutional lattice sites in the polysilicon layer 42 thereby to form a solid solution in polysilicon.
• This step minimizes the likelihood that, in subsequent heating steps of the device fabrication sequence, boron and arsenic dopant will be able to move to any appreciable extent from the polysilicon layer 42 into an overlying silicide layer. Consequently, the probabilities of boron depletion of the polysilicon gates of p-channel transistors and of arsenic counterdoping thereof are also thereby substantially minimized.
• the cap oxide layer 48 is removed.
• a silicide precurser layer 50 (FIG. 5) is then formed overlying the entire top surface of the doped polysilicon layer 42.
• this typically involves co-sputter- depositing tantalum and silicon to a thickness of approximately 250 nm, in a manner well known in the art. Other suitable techniques for forming the layer 50 are known.
• the tantalum-silicon layer 50 and the underlying polysilicon layer 42 of FIG. 5 are patterned in a conventional way in, for example, a standard reactive sputter etching procedure. Composite two-layer gates are thereby formed. Two such gates are shown in FIG. 6.
• the device shown in FIG. 6 is heated for the purpose of sintering the tantalum-silicon layers 52 and 56 and converting them to a stable high-conductivity tantalum silicide compound. Heating is carried out, for example, at about 900 degrees Celsius for approximately 30 minutes in an argon atmosphere.
• the gate of the left-hand or n-channel transistor comprises tantalum silicide layer 52 and arsenic-doped polysilicon layer 54.
• the gate of the right- hand or p-channel transistor comprises tantalum silicide layer 56 and boron-doped polysilicon layer 58.
• the dimensions a, b_, c_, and d_ indicated in FIG. 6 were approximately 2, 1, 2 and 4 micrometers, respectively.
• FIG. 7 each about . 100 nm thick are advantageously formed on the sides of the aforespecified composite gates. Then, as is known in the art but not shown in the drawing, it is feasible and advantageous in some VLSI CMOS devices of practical importance to form source and drain silicide regions on either side of each composite gate.
• the right-hand or PMOS side of the device depicted in FIG. 7 is then masked with a layer 68 of photoresist.
• a selective arsenic implant is then carried out to form source and drain regions 70 and 72 in the p- tub 12.
• arsenic dopant is also thereby introduced into the silicide layer 52.
• Some of this latter dopant will, in a subsequent heating step, be driven from the layer 52 into the underlying polysilicon layer 54 to enhance the dopant level therein to achieve a predetermined relatively low total arsenic concentration partially determinative of V .
• the dose and energy of the aforespecified arsenic implant are selected to be about 2E15 ions per square centimeter and 100 kilo- electron-volts, respectively.
• the priorly specified arsenic implant of the layer 42 described in connection with FIG. 3 is omitted.
• the polysilicon layer 54 shown in FIG. 7 is undoped prior to the NMOS source and drain implant.
• the arsenic dose of the implant which follows tantalum silicide deposition, etching, sidewall oxidation and sintering, is raised to about 3-to-4El5 ions per square centimeter.
• the source and drain regions are thereby defined essentially as before, but the arsenic concentration in the silicide layer 52 is initially thereby established to be approximately 1.5-to-2 times that specified earlier above.
• a subsequent heating step is effective to drive some of the arsenic dopant from the silicide layer 52 into the polysilicon layer 54 to establish an approximately uniform distribution of dopants across the silicide-to-polysilicon interface.
• the polysilicon layer 54 is thereby doped to the aforementioned relatively low concentration partially determinative of V .
• next steps in applicants' fabrication sequence involve removing the photoresist layer 68 (FIG. 7) and then masking the left-hand or NMOS side of the device with a layer 74 of photoresist, as indicated in
• FIG. 8 A selective (or boron difluoride) implant is then carried out to form source and drain regions 76 and 78 in the n-tub 14.
• boron dopant i ⁇ ! also thereby implanted in the silicide layer 56, very little of this implant will in practice be driven into the underlying polysilicon layer 58 during a subsequent heating step.
• the dose and energy of the boron implant are selected to be about 2E15 ions per square centimeter and 30 kilo-electron-volts, respectively.
• the photoresist layer 74 (FIG. 8) is removed and then the depicted device is subjected to an annealing step that activates the implanted dopants in the source and drain regions thereof.
• Annealing is carried out, for example, at about 900 degrees Celsius for approximately 30 minutes in an argon atmosphere.
• arsenic and boron ions previously implanted into the tubs 12 and 14, respectively, are activated and driven vertically and laterally to form n + -p and p + -n junctions, in a manner well known in the art.
• arsenic dopant ions in the silicide layer 52 are redistributed within the layer 52 and, further are driven from the layer 52 during this heating step into the underlying polysilicon layer 54. This latter action completes the formation of composite high-conductivity silicide-on-polysilicon gates and junctions in the CMOS device.
• a relatively thick insulating layer (a so-called intermediate dielectric) is then formed on the entire top surface of the herein—considered VLSI CMOS device.
• Such a layer 80 is shown in FIG. 9.
• the layer 80 is approximately 1.5 micrometers thick.
• the layer 80 is formed in a standard CVD step from a source comprising tetraethylorthosilicate and triethyl phosphite.
• the resulting conventional material is commonly referred to as PTEOS glass, which exhibits excellent step coverage.
• contact windows are etched through specified portions of the glass layer 80 (FIG. 9) to access preselected surface regions of the underlying device structure.
• a suitable contact material is then deposited on the entire top surface of the layer 80 and in the aforespecified windows formed in the layer 80.
• the contact material comprises an aluminum layer. Thereafter, the aluminum layer is selectively etched in a conventional way to form a specified interconnection pattern. Further standard steps well known in the art are then employed to complete the fabrication of the herein-described VLSI CMOS device.
• dopant (boron) depletion from the p + polysilicon layer 58 (FIG. 9) is minimized.
• dopant (arsenic) diffusion from the n + polysilicon layer 54 into the layer 58 is also minimized.
• this is accomplished by locking dopants in place in the p + and n polysilicon, by establishing a relatively low dopant concentration in the n polysilicon layer 54 and by maintaining relatively low temperatures during subsequent steps of the fabrication sequence.
• the constituent transistors of the resulting VLSI CMOS devices made by applicants' procedure consistently exhibit threshold voltages that are substantially exactly complementary.

Landscapes

• Metal-Oxide And Bipolar Metal-Oxide Semiconductor Integrated Circuits (AREA)
• Electrodes Of Semiconductors (AREA)

Priority Applications (1)

Applications Claiming Priority (2)

Publications (1)

Family

ID=24365820

Family Applications (1)

Country Status (6)

Cited By (6)

Families Citing this family (41)

Citations (1)

Family Cites Families (11)

• 1984
  • 1984-03-19 US US06/591,274 patent/US4555842A/en not_active Expired - Lifetime
• 1985
  • 1985-03-06 DE DE8585901712T patent/DE3570949D1/de not_active Expired
  • 1985-03-06 EP EP85901712A patent/EP0175751B1/en not_active Expired
  • 1985-03-06 JP JP60501291A patent/JPS61501532A/ja active Granted
  • 1985-03-06 WO PCT/US1985/000377 patent/WO1985004284A1/en active IP Right Grant
  • 1985-03-14 CA CA000476517A patent/CA1218760A/en not_active Expired

Patent Citations (1)

Non-Patent Citations (2)

Also Published As

Similar Documents

Legal Events