Classifications

• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10D—INORGANIC ELECTRIC SEMICONDUCTOR DEVICES
  • H10D30/00—Field-effect transistors [FET]
  • H10D30/60—Insulated-gate field-effect transistors [IGFET]
  • H10D30/64—Double-diffused metal-oxide semiconductor [DMOS] FETs
  • H10D30/66—Vertical DMOS [VDMOS] FETs
  • H10D30/668—Vertical DMOS [VDMOS] FETs having trench gate electrodes, e.g. UMOS transistors
• • H—ELECTRICITY
  • H10—SEMICONDUCTOR DEVICES; ELECTRIC SOLID-STATE DEVICES NOT OTHERWISE PROVIDED FOR
  • H10D—INORGANIC ELECTRIC SEMICONDUCTOR DEVICES
  • H10D62/00—Semiconductor bodies, or regions thereof, of devices having potential barriers
  • H10D62/10—Shapes, relative sizes or dispositions of the regions of the semiconductor bodies; Shapes of the semiconductor bodies
  • H10D62/17—Semiconductor regions connected to electrodes not carrying current to be rectified, amplified or switched, e.g. channel regions
  • H10D62/393—Body regions of DMOS transistors or IGBTs 

Definitions

• F t ⁇ Embodiments, f the present invention relate to mctal-oxideHS ⁇ micowduclor field fe Pec ⁇ ' ttansistors (MOSFET), and more particularly to vertical MOSFET devices having a trench gate geometry .
• MOSFET mctal-oxideHS ⁇ micowduclor field fe Pec ⁇ ' ttansistors
• a closed cell trench metal-oxide semiconductor field effect transistor is disclosed
• ⁇ oxide-semiconductor field effect transistor (TMOSFET) 100 according to the • conventional art is shown.
• the sniped TMOSFET 100 comprises a plurality of source contacts 110, a plurality of source regions 115, a plurality of gate regions 120, a plurality of gate insulator regions 125, a plurality of body regions 130, a drain region 135, 140 and drain contact 145.
• the drain region 135, 140 may optionally include a first drain portion 140 and a second drain portion 135.
• the body regions 130 are disposed above the drain region 135, 140.
• the source regions ' liS, gate regions 120 andthe gat insulator regions 125 are disposed within the body regions ' 130;
• the gate regions ' 120 and the gate insulator regions 125 are forme as parallel-elongated st ictures ' ;
• the gate insulator region 125 surrounds- the ate region 120.
• ihe gate regions 120 are electrically isolated from the surrounding regions by the gate insulator regions 125.
• the gate regions 12U are coupled to form a common gate of the device 100.
• the source regions 115 are formed as parallel-elongated structures along the periphery of the gate insulator regions 125.
• the source regions 115 aro oouplod to form a.
• the source contacts 110 may be implemented as a single conductive layer coupling all the source regions 115, The source contacts 110 also couple the source regions 115 to the body regions 130.
• the source regions 115 and the drain region 140 are heavily n-doped (N+) semiconductor, such as silicon doped with phosphorous or arsenic.
• the body regions 130 artf p-doped (P) semiconductor, such as silicon doped with boron.
• the gate region 12,0 is heavily n-doped (N+) semiconductor, such ae polysilioon doped with phosphorous.
• the gate insulator regions 125 may be an insulator, such as silicon dioxide.
• the device 1 0 is in its off state and the junction formed by the body region 130 and the drain region 140 supports the voltage applied across the source and drain.
• the drain region 135, 140 comprises a second drain portion 135 disposed above a first drain portion 140, the second portion of the drain region 135 is lightly n-doped ( - ) semiconductor, such as silicon doped with phosphorous or arsenic, and the first portion of the drain region 140 is heavily n-doped (N+) semiconductor, such as silicon doped with phosphorous or arsenic.
• - lightly n-doped
• N+ n-doped
• the lightly n-doped (N-) second portion of the drain region 135 results in a depletion region that extends into both the body regions 130 and the second portion of the drain region 135, thereby reducing the punch through effect Accordingly, the lightly u-d ⁇ yc (N-) second poition of the drain region 135 acts to increase the breakdown voltage of the striped TMOSFET 100.
• the channel width of the striped TMOSFET 100 is a function of the width of the plurality of the source regions 115.
• the striped TMOSFET 100 provides a large channel width to length ratio. Therefore, the striped TMOSFET may advantageously be utilized for power MOSFET applications, such as switching elements in a pulse width modulation-(PWM) voltage regulator.
• the closed cell TMOSFET 200 Comprises a plurality of source contacts 210, a plurality of source regions 215, a gate region 220, a gate insulator region 225, a plurality of body regions 230, a drain region 235, 240 and a drain contact 245.
• the drain region 235, 240 may optionally include a first drain portion 240 and a second drain portion 235.
• the body regions 230, the source regions 215, the gate region 220 and me gate insulator region 225 are disposed above the drain region 235, 40.
• a first portion of the gate region 220 and the gate insulator region 225 is formed as substantially parallel ' elongated structures 221.
• a second portion of the gate region 220 and the gate insulation region 225 is formed as substantially normal-to-parallel elongated structures 222,
• the first and second portions of the gate region 220 are all interconnected and form a plurality of cells
• the body regi ⁇ s 230 a_e disposed within die pluialliy of cells foiiued by-thd &te region 220..
• the gate insulator region 225 surrounds the gate region 220.
• the gate region 220 is electrically isolated f om the surrounding regions by.the gate insulator region.225. .
• The;so.uree.regio.ns 21 are formed in the plurality of cells; along the peripher of the gate insulator region 225. .. ..-• .- •
• the source regions 215 are coupled to form a common sourcc of the device 200, by the source c n acts 21 , • Although shown as a plurality of. individual source c n acts 210, it is appreciated that the source contacts 210 may be implemented as a plurality of conductive strips each coupling a plurality of source regions.215, a single conductive layer coupling all the source regions 21 , or the like. The source contacts 210 also couple the source regions 215 to the body regions 230.
• the source regions 215 and the drain region 240 are heavily n-doped (+K) semiconductor, such as silicon doped with phosphorous or arsenic.
• the body regions 230 are p-doped (P) semiconductor, such as silicon doped with boron.
• the gate region 220 is heavily n-doped semiconductor (N+), such as polysilicon doped with phosphorous.
• the gate insulator region 225 may be an insulator, such as silicon dioxide.
• the device 200 When the potential of the gate region 220, with respect to the source regions 2 , is increased above the threshold voltage of the device 200, a conducting channel is induoed in the body region 230 along the periphery of the gate insulator region* 225. The device- 200 will then conduct current between the drain region 240 and the source regions 215, Accordingly, the device 200 is in its on state.
• the device When the potential of the gate region 220 is reduced below the threshold voltage, the channel is no longer induced. As a result, a voltage potential applied between the drain region 240 and the source regions 215 will not cause current to flow there between. Accordingly, the device is in its off state and the function formed by the body region -230 and the drain region 240 supports the voltage appliod across the source and drain.
• the drain region 235, 240 comprises a second portion 235 disposed above a second portion 240, the second portion of the drain region 235 is lightly n-doped (N-) semiconductor, such as silicon doped with phosphorous or arsenic, and the first portion of the drain region 240 is heavily n-doped (N+) semiconductor, such as silicon doped with phosphorous.
• the lightly n-doped (N-) second portion of the drain region 235 results in a depletion region that extends into both the body regions 230 and the second portion of ⁇ ho drain region 235, thereby reducing the punch through effect. Accordingly, the lightly n-doped (N-) spro ⁇ pnrfio ⁇ of the drain region 235 acts to increase the breakdown voltage of the closed cell TMOSFET 200.
• the channel width of the closed cell TMOSFET 200 is a function of the sum of the width of the source regions 215.
• the closed cell TMOSFET 200 geometry advantageously increases the width of the channel region, as compared to the striped TMOSFET 100.
• the closed cell TMSOl-Ef 200 has a relatively low channel resistance (e.g., on resistance), as ⁇ uup-ued to the striped TMOSFET 100 "geornetry. The low channel resistance reduces power dissipated in the closed cell TMOSFET 200, as- compared to the striped TMOSFET 100. - •
• the gate-to-drain capacitance of the closed cell TMOSFET 220 is a function of the area of overlap between the bottom of the gate region 220 and the drain region 240. Accordingly, the closed cell TM SFET'200 geometry suffers from a higher gate-to-drain capacitance, as compared to the striped TMOSFET 100. The relatively high gate to drain capacitance limits the switching speed of the closed cell TMOSFET 200, as compared to the striped TMOSFET 100. SUM RY .
• embodiments of the present invention provide an improved closed cell trench metal-oxide-semi ⁇ nduel ⁇ r field effect transistor (TMOSFET).
• TMOSFET closed cell trench metal-oxide-semi ⁇ nduel ⁇ r field effect transistor
• embodiments of the present invention provide a closed cell TMOSFET having a low on resistance, as compared to an equivalent striped cell TMOSFET, Further, embodiments of the present invention provide a closed cell TMOSFET having a low gate-to-drain capacitance. ⁇ "
• Embodiments of the present invention provide a closed.cell TMOSFET comprising a combination of open gate-drain regions, arranged in a first plurality of parallel regions, and closed gate-drain regions, arranged in a second plurality of parallel regions .normal to the open gate-drain regions.
• a closed cell .TMOSFET comprising a drain, a body region disposed above the drajn region, a gate . , region , disposed in the formulabody. region, agate insulator region disposed about the gate region, & phiralit . of source regions.disposed at the surface of the, body, region proximate to-the periphery ofthe gate insulator region.
• a first portion of the gate region and the gate insulator region are formed as parallel-elongated structures.
• Embodiments ofthe present invention also provide a method of fabricating , closed cell TMOSFET haying a plurality of open trench bottom portions and a plurality of closed trench bottom portions.
• the method of fabrication comprises growin a lightly n doped epitaxial silicon . la er on.
• a.hoavily n-dopcd. silicon substrate The epitaxial . . deposited silicon layer is. selectively etched tn form a first set of parallel trenches an a : second se of parallel trenches normal-to-parallel to. the first set of trenches. .
• the silicon - .proximate the first and . second set of trenches is oxidized to, form , a gate oxide region.
• a p- -.-. : type impurity is implanted in the first set of parallel trench bottoms, The p-typ . impurity is not implanted in the second set of parallel trenches, which are perpendicular to the first set.
• the first and second sets of trenches are filed with polysihcon to form a gate region therein.
• a top portion ofthe lightly n-doped epitaxial silicon layer is implanted with a p-type impurity to form a p-doped body.
• a portion of the body region proximate the periphery of he gate oxide region' is implanted to form a heavily n-doped • source region.
• FIG 1 shows a cross sectional perspective view of a striped trench metal-oxide- semico ⁇ ductor field effect transistor (TMOSFET) according to the conventional art.
• TMOSFET striped trench metal-oxide- semico ⁇ ductor field effect transistor
• Figure 2 shows a cross sectional perspective view of a closed cell TMOSFET according to the conventional art.
• Figure 3 A shows a cross sectional perspective view of a closed cell TMOSFET, in accordance with one embodiment o the present invention.
• Figure 3B shows a cross sectional perspective view of a closed cell TMOSFET, in accordance with one embodiment ofthe present invention.
• FIG. 4 shows a cross sectional perspective view of another closed cell TMOSFET, in accordance with one embodiment of the present invention.
• Figures 5 A-5C show a flow diagram of a method of fabricating a closed cell
• FIGS. 6 A-6C show a flow diagram of another method of fabricating a closed cell TMOSFET, in accordance with one embodiment of the present invention.
• Figures 7A-7C show a flow diagram of another method of fabricating a closed cell TMOSFET, in accordance with one embodiment of he present invention.
• FIG. 3A a cross sectional perspective vie of a closed cell trehch mctai-oxidc-scmic ⁇ nductor field effect transistor (TMOSFET) 300, in accordance with one embodiment Of he present invention, is shown.
• the closed cell TMOSFET 300 comprises a ; plurality of source contacts 310, a plurality of source regions 315, a gate region 320, a gate insulator- region 325, a ' body region 330, a drain' region 335,- 340 and a drain contact 345.
• the drain region 335, 340 ma optionally include a first drain portion 340 ' and a second drain portion 335; " • The body regions 330, the source regions 315, the gate region 320 and the gate insulator region 325 are disposed above the drain region 335, 340.
• a first portion ofthe gate region 320 and die gate insulator region 325 are formed as substantially parallel- elongated ⁇ tructures 321.
• a second portion ofthe gate region 320 and the gate insulation region 325 are formed as substantially normal-to-parallel elongated structures 322 (e.g., in the plane ofthe surface ofthe body region, the second portion comprise a plurality of substantially parallel elongated structures formed at right angles to the first portion ofthe gate region and gate insulator region).
• the first and second portions ofthe gate region 320 are all interconnected and form a plurality of cells.
• the body regions 330 are disposed within the plurality of cells formed by the gate region 320.
• a third portion 350 ofthe drain region 335, 340 extends to the bottom ofthe first portion o the gate insulator region 325.
• the closed cell TMOSFET 300 in accordance with embodiments of the- present invention, comprises a plurality of open gate-drain regions 331 (e.g., selectively blocked/by a portion of the body region 330) arranged with respect to the second plurality of parallel structures 322.
• the closed cell TMOSFET 300 further comprises a plurality of closed gate-drain regions 350 (e.g., the drain region 335 overlaps the gate region 320) arranged with respect to the first plurality of parallel structures 321 that are normal to the open gate-drain regions 331.
• the gate insulator region 325 surrounds the gate region 320, thus, the gate region
• the source regions 315 are electrically isolated from the surrounding regions by the gate insulator region 325.
• the source regions 315 are formed in the plurality of cells, along the periphery-of the-gate insulator region 325.
• the source regions 315 are coupled to form a common source ofthe device, by the source contacts 310.
• the source contacts 310 also couple the source regions 315 to the body region 330. " . . . . . -. ;
• the source regions 315 and the drain region 335, 340 are heavily n-doped (N+) semiconductor, such as silicon doped with phosphorous or arsenic.
• the gate insulator region 325 may be an oxide, such as silicon dioxide0 of he like. " ' ; ' ⁇ . : ' ⁇ ⁇ . • ⁇ .- : w. -/ -. : . , . - *• ⁇ •; • .-. ;
• the device 300 When the potential of thegate region 320, with respeett the source regions 315, is increased above the threshold voltage ofthe device 300, a conducting channel is induced in the body region 330 along the periphery o the gate insulator region 325. The device 300 will then conduct current between the drain region 340 and the plurality of source5 regions 315: Accordingly, the device 300' is in is on state: Charge flows from the ⁇ * " p >Xim ⁇ ty of the extended, portion o the drain region 350, which- ⁇ verlaps the bottom of the first portion ofthe gate region 3 0; through the induced chanhel pfo imate the first p ⁇ rtionb the gate region 320, and into the source regions 315.
• the drain region 335, 340 comprises a second drain portion 335 disposed above a first portion 340 ofthe drain region
• the second portion of he drain region 335 is lightly n-doped (-N) semiconductor, such as silicon doped with phosphorous or arsenic
• the first portion ofthe dra region 340 ' is heavily n-doped (N+) semiconductor, such as silicon doped with phosphorous.
• the lightly n-doped (-N) second poitiou of die iaiii region 335 results in a depletion region that extends into both the body region 330 and the second portion of the drain region 335, thereby reducing the punch through effect.
• the lightly n-doped (-N) second portion ofthe drain region 335 acts to increase the breakdown voltage ofthe closed cell TMOSFET 310.
• the width of me channel remains a function ofthe sum ' of the lengths of he source regions 315.
• the width of the channel region is substantially equal to the legacy closed cell TMOSFET 200. Therefore, the on resistance (Rds-on) o the device 300 is substantially equal to the legacy closed cell TMOSFET 200.
• the induced channel comprises a gate-to-source capacitance.
• the gate region 320, the gate insulator region 325 and the drain region 335, 40 also comprise a gate-to-drain capacitance.
• the portions ofthe body region 331 disposed between the drain region 335, 340 and the second portion of the gate region 320 and gate insulator region 325 acts to reduce the gate-to-drain capacitance (Cgd) ofthe device 300
• Cgd gate-to-drain capacitance
• tho reduction of the gate-to-drain capacitance causes an increase in the gate-to-source capacitance (Cgs).
• the charge in the gate region 320 which in a legacy device would have coupled to charge in the drain region 335, 340, now acts to increase the induced channel, and hence the capacitance between the source and gate.
• the body 330 and source 315 are couple together, and therefore effectively increases the gate to source capacitance.
• the closed cell TMOSFET 300 in accordance the present embodiment, advantageously provides a low gato-to-drain capacitance (Cgd) to gatc-to-source capacitance .(Cgs) ratio, as compared to the legacy closed cell TMOSFET 200. Furthermore, the closed cell TMOSFET 300, in accordance with the present embodiment, advantageously provides an improved Rds-on*Qgd figure of merit, as compared to both the legacy striped TMOSFET 100 and the legacy closed cell TMOSFET 200.
• FIG. 3B a cross sectional perspective view of a closed cell trench metal-oxide-semico ⁇ ductor field effect transistor (TMOSFET) 300, in accordance with one embodiment o the present invention, is shown.
• the f ont comer ofthe cross sectionol perspective view is cut ⁇ w ⁇ y to show that the parollel and normal to parallel elongated structures of the drain region 320 are all interconnected.
• Figure 3B illustrates " me induced channel 360 and the flow.of charge 380, 318 in the on state ofthe TMOSFET 300.
• the conducting channel 360 is induced in the body region 330 along the periphery of the gate-insulator region 325.
• the device 300 willtfaen conduct current between the drain region 340 and the plurality of source regions 315.
• Charge 381 also raps around from the proximity ofthe extended portion ofthe drain region 350, which overlaps the bottom ofthe first ' portion of the gate region 320, through the induced channel proximate the Secon portion ofthe gate insulator region 320, arid into the source regions 315. " Accordingly, a component ofthe current flows from the induced channel 36 in the portions ofthe body region 331, disposed between the drain region 335, 340 and the sec ⁇ ci portion o the gate region 320 and gate insulator region 325, and into the extended portion of the drain region 3$0. This component of the current acts to decrease the Rds- ⁇ ofthe TMOSFET 300, as compared to the conventional striped TMOSFET, '
• Figure 3B illustrates the source contact 310 implemented as a single conductive layer coupling all the source regions 315, or the like; such an embodiment the gate insulator region 325 is also disposed between the gate region 320 and the source contact 310.
• the TMOSFET 300 includes a source contact region 370.
• the source contact region 370 comprises heavily p-doped (P+) semiconductor, such as silicon doped with boron.
• the source contact region 370 pi ⁇ vides a l ⁇ w uhr ⁇ ic contact between the source contact 315 and the body region 330.
• FIG. 4 a cross sectional perspective view of a closed cell trench metal-oxide-semiconductor field effect transistor (TMOSFET) 400, in accordance with . one embodiment ofthe present invention, is shown.
• the closed cell TMOSFET 400 comprises a plurality of source contacts 410, a plurality of source regions 415, a gate region 420, a gate insulator region 425, a body region 430, a drain region 335, 440 and a drain contact 445.
• the drain region 435, 440 may optionally mclude a first drain portion 440 and a second drain portion 4 5.
• the body region 430 is disposed above the drain region 435, 440.
• the source regions 415, the gate region 420 and the gate insulator region 425 are disposed within the body region 430.
• a first portion ofthe gate region 420 and the gate insulator region 425 are- formed as substantially parallel elongated structures 42
• a second portion of the - gate region 420 and the gate insulation region 425 are formed as- substantially n ⁇ rmal-to- parallel elongated structures 422.
• the first and second portions ofthe gate region 420 are all interconnected and form a plurality of cells.
• the body region 430 is disposed within tho plurality of cells formed by the gate region 420, with a portion of the body region 450 ft ⁇ rrcSunding-thft first portion' of the gate region 421.
• the closed cell TMOSFET 400 in accordance with embodiments of the present invention, comprises a plurality of open gate-drain regions 450 (e.g., selectively blocked by a portion o the body region 430 that encircles the gate region 420) arranged in a first plurality of parallel structures 421.
• the closed cell TMOSFET 400 further comprises a plurality of closed gftte- ⁇ Vain regions 451 (e.g., the drain region 435 overlaps the gate region 420) arranged in a second plurality of parallel structures 422 that are normal to the open gate-drain regions 450.
• the gate insulator region 425 surrounds the gate region 420.
• the gate region 420 is electrically isolated from the surrounding regions by the gate insulator region 425.
• the source regions 415 are coupled to form a common source ofthe device, by the source oont ⁇ cts 410. Although shown as a plurality of individual Source contacts 410, it is appreciated that the source contacts 410 may be implemented as a single conductive layer coupling all the source regions 415.
• the source contacts 410 also couple-the source-regions 415 to the body region 430.
• the source regions 415 and the drain region 435, 440 are heavily ri-doped (N+) semiconductor, such as silicon doped with phosphorOus or arsenic.
• Thebody region 430 is p-doped (P) semiconductor such as silicon doped with boron.
• the gale region 420 is heavily n-doped (N+) semiconductor, such as polysiiicondoped with phosphorous.
• the gate insulator region 425 may be an oxide, such as silicon dioxide.
• the TMOSFET 40Q may optionally include a source contact region (not-shown).
• the source contact region comprises heavily p-doped (P+) semiconductor, such as silicon doped withb ⁇ ron.
• P+ heavily p-doped
• the source contact region provides a lo ohmic contact between the source contact 410 and the body region 430. ... . - ⁇ •' • • •' , : • •' :•
• the drain region 435, 440 comprises a second portion 435 disposed above a first portion 440
• the second portion ofthe drain region 435 is lightly n-doped (-N) semiconductor, such as silicon doped with ph ⁇ sph ⁇ r ⁇ ut> ⁇ i arsenic
• the first portion ofthe drain region 440 is heavily n-doped (N+) semiconductor, such as silicon doped with phosphorous.
• the lightly n-doped (-N) second portion ofthe drain region 435 results in a depletion region that extends into both the body region 430 and the second portion ofthe drain region 435, thereby reducing the punch through effect.
• the lightly n-doped (-N) second portion ofthe drain region 435 acts to increase the breakdown voltage o the closed cell TMOSFET 410.
• the width of the channel remains a function ofthe sum o he lengths o the source regions 415. ilenoc, the width ofthe channel is substantially equal to the legacy closed cell MOSFET. Therefore, the on resistance (Rds-on) ofthe device 400 is substantially equal to the legacy closed cell MOSFET 200.
• the induced channel comprises a gate-to-source capacitance.
• the gate region 420, the gate insulator region 425 and the drain region 440 also comprise a gate-to-drain capacitance, fhe portions ofthe body region 450 disposed befcwee ⁇ the drain region 435, .440 and the first portion ofthe gate region 420 and gate insulator region 425 acts to reduce the gate-to-drain capacitance (Cgd).
• the reduction ofthe gate-to-drain capacitance causes an increase in he gatc-to-souroe capacitance (Cgs)
• the charge in the gate region 420 which in a legacy device would have coupled to charge in the drain region 435, 440, now act ⁇ to increase the induced channel, and hence the capacitance between the body and gate.
• the body 430 and source 4115 are coupled together, and therefore the increased gate-to-body capacitance effectively increases the gate-to-source capacitance (Cgs).
• the closed cell TMOSFET 400 in accordance with the present embodiment, advantageously provides a low gate-to-drain capacitance (Cgd) to gate-to-source capacitance (Cgs) ratio, as compared to the legacy closed cell TMOSFET 200. Furthermore, the closed cell TMOSFET 400, in accordance with the present embodiment, advantageously provides an improved Rds-on*Qgd figure of merit, as compared to both the legacy striped TMOSFET 100 and the legacy closed cell TMOSFET 200,
• FIG. 5A-5C a flow diagram of a method of fabricating a, closed cell trench metai-oxide-semiconductor field effect transistor (TMOSFET), in accordance with one embodiment ofthe present invention, is shown.
• TMOSFET closed cell trench metai-oxide-semiconductor field effect transistor
• the process begins, at 502, with various initial processes upon a- • substrate, such as cleaning, depositing, doping, etching and/ ⁇ r the Hke.
• the semiconductor substrate comprises a first portion of a heavily doped drain region, ⁇ n one implenaejitation, the first portion of he drain region comprises silicon heavily doped with phosphorous (N+).
• a semiconductor layer is epitaxial deposited upon the substrate.
• the semiconductor layer comprises silicon -lightly doped with , ⁇ phosphorous.
• the epitaxial deposited silicon may be doped by introducing the desired impurity, such as phosphorous, into the reaction chamber.
• a sacrificial oxide layer is formed upon the epitaxial deposited semiconductor layer, h one implementation, the sacrificial oxide is formed by oxidizing the surface ofthe epitaxial deposited silicon layer.
• a barrier layer is deposited upon the sacrificial oxide layer.
• the hairier layer is deposited by chemical vapor deposition (CND) of silicon nitride (S1 ⁇ 4).
• a photo-resist is deposited and patterned by any-well know lithography process to form a gate trench mas
• ⁇ t 512, tho exposed portions ofthe barrier layer, sacrificial oxide layei and a portion nf the first semiconductor layer are etched by any-well known isotropic etching method.
• an ionic etchant interacts with the' barrier layer, sacrificial oxide layer and first semiconductor layer exposed by the patterned resist layer.
• a plurality of trenches are formed, such that a first set of trenches are substantially parallel to each other and a second set of trenches are substantially normal-to-parallel ith respect to the first set of trenches.
• a trench bottom doping is performed in the first set of the trenches.
• a p-typc impurity such as boron
• the doping process results in the formation of a p- doped well region that encircles the bottoms of the first set of trenches.
• the implant ion flux is orientated at an angle with respect to the plane of the wafer. The angle of incident is selected such tha the impurity is implanted i the. first set of trench bottoms, while tfce second set of trench bottoms are apt doped. ..More, specifically, if the surface of the barrier layer, lies n an x-y plane, the
• first set of • trenches lie in x-z-planes and the second set of trenches. lie in y-z planes
• the angle of orientation is such that the ion flux travels in the x and z directions and not in the y direction. Accordingly, the dopant reaches the bottoms ofthe first set of trenches.
• the .' barrier layer and or ! the sacrificial oxide layers capture the dopant along the second set of trenches. Thus, the dopant does not reach the bottoms of the second set of 0 trenches.
• the angle of incident is selected as a function of he width the trenches and the thickness ofthe barrier a ⁇ d/ ⁇ r sacrificial oxide layer.
• Utilizing a ion implant process wherei the ion flu is incident upon the wafer at an appropriate angle is advantageous in that the patterned photo-resist, barrier layer and sacrificial oxide layer are utilized as a mask fo 'both 'the etching process f 512, die implanting of 5 ⁇ 4'andtHe subsequent ' S oxidation process of 518.
• Utilizing' anion implant process Wherein the ion fluX is r inci ent upon the wafer at an. appropriate angle is also advantageous in that " t ie treneti irripiant is self-aligned with the first set of trenches, and the second set of trench bottoms ' arV ⁇ ot db ⁇ d.
• the patterned resist barrier layer 7 - sacrificial Oxide 20 layer are striped.
• Another sacrificial oxide layer, barrier layef an resist are deposited and :patterhed-such that only the " first set of trenches are exposed.
• The'ion-implarit process is "then performed such that the angle of incident of the i ⁇ n lu is aubst ⁇ tiauy ⁇ iiaal to the plane ofthe wafer. Accordingly, the bottoms ofthe first set of trenches are doped.
• the sacrificial oxide layer, barrier layer and resist are then removed and an additional sacrificial oxide layer, hairier layer and resist layer are deposited and patterned such that both the first and second set of trenches arc oxposcd.
• the gate trench mask is removed utilizing an appropriate resist stripper or a resist ashing process.
• a dielectric is formed on the walls ofthe first and second set of trenches. In one implementation, the dielectric is formed by oxidizing the surface of the silicon to form a silicon dioxide layer. The resulting dielectric along the trench walls forms a gate region dielectric.
• a polysilicon layer is deposited in the first and second set of trenches to form a gate region. In one implementation, the polysilicon is deposited, in the trenches by a method such as decomposition of silane (S1H4). The polysilicon is doped with n-rype impurity such as phosphorous or arsenic. The polysilicon may be doped by introducing the impurity during the deposition process.
• an etoh-back process is performed to remove excess polysilicon on the surface ofthe wafer and the first barrier layer.
• the excess polysilicon and the barrier layer are removed by a chemical mechanical polishing (CMP) process.
• CMP chemical mechanical polishing
• a second photo-resist is deposited and patterned to form a body region mask.
• the body region mask defines a plurality of cells, which are defined by the area inside the gale region.
• the exposed portion of he epitaxial deposited semiconductor layer is doped to form body rogions in tho plurality of cells.
• the doping process implants a p-type impurity, such as boron, in the Upper portion ofthe epitaxial deposited semiconductor layer.
• a high temperature thermal cycle may be utilized to drive in the body region doping. Accordingly, the lower portion ofthe epitaxial deposited semiconductor layer forms a lightly doped second drain portion.
• the body region mask is removed.
• a third photo-resist is deposited and patterned to fbrm a source region mask, at
• the portion ofthe first semiconductor layer, left exposed by the source region mask is doped to form source regions.
• the doping process comprises heavily implanting an n-type impurity, such as phosphorous, into the plurality of cells adjacent the ' ate oxide region.
• a high temperature thermal cycle may be utilized to drive in the source region doping.
• the source region mask is removed.
• a dielectric layer is deposited upon the wafer.
• the dielectric layer is deposited by decomposition of tetraethylorthosilicate (TEOS) in a chemical vapor deposition (CND) system.
• TEOS tetraethylorthosilicate
• CND chemical vapor deposition
• a fourth photo-resist layer is deposited and patterned to define ⁇ . source- body contact mask above each cell.
• the portion ofthe dielectric layer left exposed by the source-body contact mask is etched. The source-body contact mask is removed at 552.
• a source-body metal layer is deposited on the surface of he wafer.
• the source-body metal layer is deposited by any well-known method such as sputtering.
• the source-body metal layer forms a contact with the body and source regions left exposed by the patterned dielectric.
• the source-body metal layer is isolated from the gate region by the patterned dielectric layer.
• the source-body metal layer is then patterned uhhzing a photo-resist mask and selective etching method as needed, at 556.
• the various processes typically include etching, deposition, doping, cleaning, annealing, passivation, cleaving and or the like.
• FIG. 6 A-6C a flow diagram of a method of fabricating a closed cell trench metai oxide-semiconductor field effect transistor (TMOSFET), in accordance with, one embodiment of theprescnt invention, is shown.
• TMOSFET closed cell trench metai oxide-semiconductor field effect transistor
• the process begins, at 602, with various initial processes upon a substrate, such as cleaning, depositing, doping, etching and or the like.
• the " semiconductor substrate comprises a first portion of a heavily doped drain region.
• the first portion o the drain region comprises silicon heavily doped with phosphorous (N+).
• a semiconductor layer is epitaxial deposited upon the substrate.
• the first semiconductor layer comprises silicon lightly doped with phosphorous.
• the epitaxial deposited silicon may be doped by introducing the desired impurity, such as phosphorous, into the reaction chamber.
• a sacrificial oxide layer is formed upon the epitaxial deposited semiconductor layer.
• the sacrificial oxide is formed by oxidizing the surface- of the epitaxial deposited silicon Ia ⁇ i.
• a bauied layei is deposited upon the sacrificial oxide layer.
• the barrier layer is deposited by chemical vapor deposition (CVD) of silicon nitride (S1N4),
• a photo-resist is deposited and patterned by any well-know lithography process to form a gate trench mask.
• the exposed portions o the barrier layer, sacrificial oxide layer and a portion ofthe first semiconductor layer are etched by any- well known isotropic etching method.
• an ionic etchant interacts with the barrier layer, sacrificial oxide layer and first semiconductor layer exposed by the patterned resist layer.
• a plurality of trenches are formed, such that a' first set of trenches " arc substantially parallel to each other and a second set of trenches are substantially normal-to-parallel with respect to the first set of trenches.
• a first trench bottom doping is performed in the first set ofthe trenches.
• an n-type impurity such as phosphorous
• the doping process results m the formation of an n-doped region that extends downwardly from the bottoms ofthe first set of trenches.
• the implant ion flux is orientated at a first angle with respect to the plane ofthe wafer. The first angle of incident is selected such that the impurity is implanted in the first set of trench bottoms, while the second set of trench bottoms are not doped with the n-type impurity.
• a second trench bottom doping is performed in the second set of trenches.
• a p-type impurity such as boron
• the doping process results in the formation of a p- doped region that extends downwardly from the bottoms o the second s&t of trenches.
• the implant ion flux is orientated at a second angle with respect to the plane of he wafer. The second angle of incident is selected such that the impurity is mplanted in the second set of trench bottoms, while the first set of trench bottoms are not doped with the p-type impurity.
• Utilizing a first and second ion implant process wherein the ion flux is incident upon the wafer at an appropriate first and second angle, respectively; is advantageous in that the patterned photo-resist, barrier layer and sacrificial oxide layer is utilize as a mask ' for both the etching process Of 612, the implanting processes of 614 and 616, and the subsequent oxidation process of 620.
• Utilizing an ion implant process wherein' the ioh flux. is incident upon the wafer at an appropriate first and second angle is also advantageous in that the trench implant is self-aligned with the first and second set of trenches, respectively.
• the gate trench mask is removed utilizing an appropriate resist stripper or a resist ashing process.
• a dielectric is formed on the walls ofthe first and second set of trenches. In one implementation, the dielectric is formed by oxidizing the surface of the silicon to form a silicon dioxide layer. The resulting dielectric along the trench walls forms a gate region dielectric.
• aipolysilicon layer-is deposited in the .first a ⁇ d sccur ⁇ l set of ueuches to .forrn gate region.
• the polysihcon is deposited in the ranches by .a rnethod ⁇ uch as decomposition nf silane (Si ⁇ ).
• the polysilicon is doped with n-type impurity such as phosphorous or arsenic.
• the polysilicon may be doped by introducing the impurity during the deposition process. ⁇ : . . . . ⁇ ⁇ .. : ; . .
• an etch-backprocess is performed to remove, excess polysilicon on .he surfa ⁇ &.of he wafer, and the barrier layer.
• the excess polysilicon "and the barrier layer ' are removed by a chemical mechanical polishing (CMP) process. .
• a,Be . cond photo-re ⁇ ist is deposited and patterned to form a. body region mask.
• Tnebodyxe ion mask exposes a plurality of cells, which are.defined by thearea. • inside the gate, region.
• the exposed portion of he first semiconductor layer - is dope to form body regions ⁇ n the plurality of cells.
• me 'doping process implants a p-type impurity, such, as boron, from the surface ofthe wafer to just bellow the bottoms ofthe gate dielectric region.
• a high temperature thermal cycle may be utilized to drivo in the body region doping.
• the body region mask is removed.
• the source region mask defines a source region in each cell-adjacent the gate oxide . region- .At 642, the portion ofthe first semiconductor layer, left exposed by the source mask, is doped to form source regions.
• the doping process comprises heavily implanting an n-type impurity, such as phosphorous, into the 0 plurality of cells adjacent the gate oxide region.
• a high temperatur 'mermal cycle may be utilized to drive in the source region doping.
• the source mask is removed/
• a. dielectric .layer is deposited upon the wafer...
• the dielectric layer is deposited by decomposition of tetraethylorthosilicate (TEOS) in a chemical vapor, deposition (CVD) system.
• TEOS tetraethylorthosilicate
• CVD chemical vapor, deposition
• the, portion of tiie.dielectriq.lay r: left exposed by is, etched.
• a 654, the sourcs ' bod contact mask ; is. j : removed,.
• a source ⁇ body metal layer is deposited on the surface of the wafer. In one implcmeaiation, the source-body metal layer is deposited by any well-known method such as -sputtering.
• the source-body metal layer forms a contact with the body and source regions left exposed by he patterned dielectric.
• the suur e-b ⁇ dy metal layer is isolated from the gate region by the patterned dielectric layer.
• the source body metal layer is then patterned utilizing a photo-resist mask and selective etching method as needed, at 658,
• fabrication continues with various other processes.
• the various processes typically include etching, deposition, doping, cleaning, annealing, passivation, cleaving 'an ⁇ Vor the like.
• FIG. 7 -7C a flow diagram of a method of fabricating a closed cell trench metai-oxide-semiconductor field effect transistor (TMOSFET), in.. accordance with one embodiment o the present invention, is shown.
• TMOSFET closed cell trench metai-oxide-semiconductor field effect transistor
• the process-begins, at 702 with various initial processes upon a substrate, such as cleaning, depositing, doping, etching and/or the like.
• the semiconductor substrate comprises a first portion of a heavily doped drain region.
• the first portion of he drain region comprises silicon heavily doped With phosphorous (N+).
• a first portion of a semiconductor layer is epitaxial deposited upon the substrate.
• the first portion ofthe semiconductor layer comprises silicon lightly doped with phosphorous.
• the epitaxial deposited silicon may be doped by introducing the desired impurity, such as phosphorous, into the reaction chamber. .
• a sacrificial oxirie layer is formed upon the first portion o the semiconductor layer.
• a photo-resist is deposited and patterned to form a buried layer mask. .
• a shallow implant process is performed to form a plurality of shallow doped regions, which are substantially parallel to each other.
• an n- rypc impurity such as phosphorous, is selectively implanted utilizin any well-known t n-implant process. . .
• the.pattemed resist is removed utilizing an appropriate resist. stripper or a resist ashing process.
• the sacrificial oxide layer is.remoyed.. utilizing a.cherhical rnechameal polishing (CM?) process.. .
• a second portion of the semiconductor layer is epitaxially deposited upon the substrate.
• the prrnal cycle of flic second portion ofthe epitaxial process and. other subsequent thermal cycles cau ⁇ .es the shallow doped regions- to diffuse to fouii. au.u- .. doped buried layers extending;Up to. the bottom o the-subsequeritjy.formed firs* ⁇ etof ⁇ trenches; . ,
• a second sacrificial oxide layer is formed upon the epitaxial semiconductor layer.
• a barrier layer is deposited upon the second sacrificial oxide layer.
• the barrier layer is deposited by chemical vapor deposition (CVD) of silicon nitride (S.N4),
• a second photoresist is deposited upon the barrier layer and patterned to form a gate trench mask.
• the exposed portions of the barrier layer, second sacrificial oxide layer and semiconductor layer are etched by any-well known isotropic etching method.
• a plurality of trenches are formed, such that a first set of trenches are substantially parallel to each other and a second set of trenches are substantially no ⁇ nal-to-parallel with respect to the first set of trenches.
• Each ofthe first set of trenches is also substantially aligned to a respective one ofthe plurality of buried layer.
• a trench bottom doping is performed in the second set ofthe trenches.
• a p-type impurity such as boron
• the implant ion flux is orientated at an angle with respect to the plane of the wafer.
• the arigle of incident is selected suoh that the impurity - is implanted in the second set of trench bottoms, while the first set of trench bottoms are not dopedv More specifically, if the surface ofthe barrier layer lies in an x-y plane, the first set of trenches lie in x-2 planes and the second set of trenches lie, in y-z planes, the angle of orientation is such that the ion flux travels in the y and z directions and.not in the X direction. Accordingly, the dopant reaches the bottoms of the second set of trenches .
• the dopant does not reach the bottoms o the first set of trenches.
• the angle of incident is selected as a function of he width the trenches and the thickness of flic barrier and or sacrificial oxide layer.
• the doping progess results in the formation of a p-doped region extending down from the bottoms of .the second set of trenches.
• Utilizing an ion implant process wherein the ion flux is incident upon the wafer at an appropriate angle is advantageous in that the patterned photo-resist, barrier layer and sacrificial oxide layer are utilized as a mask for both the etching process of 728, the implanting of 730 and the subsequent oxidation process of 734.
• Utilizing an ion implant process wherem the ion flux is incident upon the wafer at an appropriate angle is also advantageous irr that die trench implant is self-aligned with the second set of trenches, and the fir&t set of trench bottoms arc not doped.
• a dielectric is formed on the walls ofthe first and second set of trenches.
• the dielectric is formed by oxidizing the surface o the silicon to form a silicon dioxide layer. The resulting dielectric alongthe trench walls forms a gate region dielectric.
• a polysilicon layei is deposited in the first and-second set of trpnches to form a gate region.
• the polysilicon is deposited i the usiiches by a, method such as decomposition of silane (S1H4).
• the polysilicon is doped with n type impurity such as phosphorous or arsenic.
• the pftlysilicon may be doped by introducing the impurity during the deposition process.
• an etch-back process is performed to remove cxeetis polysilicon on the surface of the wafer, barrier layer and second sacrificial oxide layer.
• a third photo-resist is deposited and patterned to form a body region mask.
• the body region mask exposes a plurality of cells, which are defined by the area inside the gate region.
• the exposed portion of the epitaxial deposited semiconductor layer is doped to form body regions in the plurality of cells.
• the doping process implants a p-type impurity, such as boron, from the surface to just bellow the bottoms ofthe gate dielectric region.
• a high temperature ihcr af cycle- may beufilize ⁇ to drive ihthe gate region doping.
• the body region mask is removed. . .
• a fourth photo-resist is deposited and patterned to forai a source region mask, at 752.
• the source ' region mask defines a source region in each celt adjacent the gateOxide region.
• the portion ofthe epitaxial deposited semiconductor-layer, left exposed by the source region mask is doped to form source t egioris.
• the doping process -comprises heavily implanting an ⁇ -type impurity, siich as phosphorous, into the pluiality of cells adjacent tho gate oxide region.
• a high temperature thermal cycle may-be utilized ' ⁇ o drive in the source region doping
• the source region mask is removed;- ' •
• a dielectric layer is deposited upon the wafer.
• the dielectric layer is deposited by decomposition of tetraethylorthosilicate (TEOS) in a chemical vapor deposition (CV ) system.
• TEOS tetraethylorthosilicate
• a fifth photo-resist is deposited and patterned to define a source-body contact mask above each cell, at 764.
• the portion o he dielectric layer left exposed by the source-body contact mask are etched.
• the source-body contact mask is removed, at 768.
• a source-body metal layer is deposited on the surface o the wafer.
• the source-body metal layer is deposited by any well-J ⁇ iown method such as sputtering.
• the source-body metal layer forms a contact with the body and source, regions left exposed by the patterned dielectric.
• the Ronrce-body metal layer is isolated from the gate region by the patterned dielectric layer.
• the source-body metal layer is then patterned utilizing a photo-resist mask and selective etching method as - needed, at 772.
• the various processes typically include etching, deposition, doping, cleaning, annealing, passivation, cleaving and or the like.
• embodiments of he present invention include an improved closed cell trench meial-oxide-semicottdttotor field effect transistor (TMOSFET),
• the closed cell TMOSFBT comprises ⁇ drain, a body region disposed above the drain region, a gate ⁇ •' &g ffl 8&pose ⁇ m ⁇ body region, a gate insulator region, a plurality of source regions disposed at ihe surface ofthe body region proximate to the periphery of he gate insulator ⁇ region.
• a first portion o the gate region and the gate oxide region are formed as parallel eloi ⁇ atw-structures, A second portion of the gate regionand the oxide region re fbrmed as normal-to-parallel elongated structures. ' A portion of the gate and drain overlap region are sefecfrvely blocked by tt ⁇ body region, resulting inlower overall gate to drain

Landscapes

• Insulated Gate Type Field-Effect Transistor (AREA)

Priority Applications (2)

Applications Claiming Priority (2)

Publications (2)

Family

ID=34620544

Family Applications (1)

Country Status (5)

Cited By (1)

Families Citing this family (71)

Family Cites Families (65)

• 2003
  • 2003-12-02 US US10/726,922 patent/US7279743B2/en not_active Expired - Fee Related
• 2004
  • 2004-11-30 DE DE112004002310.6T patent/DE112004002310B4/de not_active Expired - Fee Related
  • 2004-11-30 JP JP2006542677A patent/JP2007513523A/ja active Pending
  • 2004-11-30 WO PCT/US2004/040063 patent/WO2005057615A2/en not_active Ceased
  • 2004-12-01 TW TW093137023A patent/TWI366249B/zh not_active IP Right Cessation
• 2005
  • 2005-01-20 US US11/040,129 patent/US7361558B2/en not_active Expired - Fee Related
• 2008
  • 2008-04-22 US US12/107,738 patent/US7833863B1/en not_active Expired - Fee Related

Also Published As

Similar Documents

Legal Events