Abstract: A substrate structure having a set of nanosheet layers and a set of sacrificial layers stacked upon a substrate is received and a dummy gate is formed upon the nanosheet layers and the sacrificial layers. A portion of a subset of the set of sacrificial layers and a subset of the set of nanosheet layers is etched. A portion of a subset of the subset of sacrificial layers is etched to create divots within the sacrificial layers. A divot fill layer is deposited. The divot fill layer is etched to form an inner spacer between the nanosheet layers. A source/drain region is formed adjacent to the nanosheet layers and the divots. A remaining portion of the subset of the sacrificial layers is removed. The subset of the nanosheet layers is etched to a desired channel thickness producing faceted surfaces between the subset of nanosheet layers and the inner spacer.

Abstract: A substrate structure having a set of nanosheet layers and a set of sacrificial layers stacked upon a substrate is received and a dummy gate is formed upon the nanosheet layers and the sacrificial layers. A portion of a subset of the set of sacrificial layers and a subset of the set of nanosheet layers is etched. A portion of a subset of the subset of sacrificial layers is etched to create divots within the sacrificial layers. A divot fill layer is deposited. The divot fill layer is etched to form an inner spacer between the nanosheet layers. A source/drain region is formed adjacent to the nanosheet layers and the divots. A remaining portion of the subset of the sacrificial layers is removed. The subset of the nanosheet layers is etched to a desired channel thickness producing faceted surfaces between the subset of nanosheet layers and the inner spacer.

Abstract: A CMOS-based resistive processing unit (RPU) for a neural network. The RPU includes a capacitor device configured to store a charge representing a weight value associated with a neural network circuit operation. A current source Field Effect Transistor (FET) device is operatively connected to the capacitor device for increasing a charge on the capacitor. A current sink FET device operatively connected to the capacitor device is configured for decreasing the stored capacitor charge. An analog weight update circuit receives one or more update signals generated in conjunction with the neural network circuit operation, the analog weight update circuit controlling the current source FET device and the current sink FET device to provide either a determined amount of current to increase the stored charge on the capacitor device, or sink a determined amount of current to decrease the stored charge on the capacitor device.

Abstract: A method for forming a floating gate memory cell includes: forming an active region on a semiconductor substrate; forming a gate stack on the active region, the gate stack including a first gate layer defining a floating gate of the memory cell structure, a dielectric layer formed on the first gate layer, and a second gate layer defining a control gate of the memory cell structure formed on the dielectric layer; forming first and second source/drain regions in the active region, self-aligned with the gate stack; forming an erase/injection gate on at least a portion of the dielectric layer and spaced laterally from the control gate, the erase/injection gate being proximate to and above the floating gate; and forming multiple contacts providing electrical connection with the first and second source/drain regions, the control gate and the erase/injection gate.

Abstract: A memory device including a plurality of memory unit cells arranged in a crossbar configuration for a neural network is provided. Each of the memory unit cells includes a readout transistor, a charging transistor, a discharging transistor, and a metal-insulator-metal (MIM) capacitor connected to one of source/drain regions of each of the charging transistor and the discharging transistor and a functional gate of the readout transistor for storing analog information.

Abstract: A CMOS-based resistive processing unit (RPU) and method for a neural network. The RPU includes a capacitor device configured to store a charge representing a weight value associated with a neural network circuit operation. A current source Field Effect Transistor (FET) device is operatively connected to the capacitor device for increasing a charge on the capacitor. A current sink FET device operatively connected to the capacitor device is configured for decreasing the stored capacitor charge. An analog weight update circuit receives one or more update signals generated in conjunction with the neural network circuit operation, the analog weight update circuit controlling the current source FET device and the current sink FET device to provide either a determined amount of current to increase the stored charge on the capacitor device, or sink a determined amount of current to decrease the stored charge on the capacitor device.

Abstract: A method of forming a semiconductor device that includes forming a fin structure from a bulk semiconductor substrate and forming an isolation region contacting a lower portion of a sidewall of the fin structure, wherein an upper portion of the sidewall of the fin structure is exposed. A sacrificial spacer is formed on the upper portion of the sidewall of the fin structure. The isolation regions are recessed to provide an exposed section of the sidewall of the fin structure. A doped semiconductor material is formed on the exposed section of the lower portion of the sidewall of the fin structure. Dopant is diffused from the doped semiconductor material to a base portion of the fin structure.

Abstract: Methods and structures for fabricating fins for multigate devices are disclosed. In accordance with one method, a plurality of sidewalls are formed in or on a plurality of mandrels over a semiconductor substrate such that each of the mandrels includes a first sidewall composed of a first material and a second sidewall composed of a second material that is different from the first material. The first sidewall of a first mandrel of the plurality of mandrels is selectively removed. In addition, a pattern composed of remaining sidewalls of the plurality of sidewalls is transferred onto an underlying layer to form a hard mask in the underlying layer. Further, the fins are formed by employing the hard mask and etching semiconducting material in the substrate.

Abstract: A method for forming strained fins includes etching trenches in a bulk substrate to form fins, filling the trenches with a dielectric fill and recessing the dielectric fill into the trenches to form shallow trench isolation regions. The fins are etched above the shallow trench isolation regions to form a staircase fin structure with narrow top portions of the fins. Gate structures are formed over the top portions of the fins. Raised source ad drain regions are epitaxially grown on opposite sides of the gate structure. A pre-morphization implant is performed to generate defects in the substrate to couple strain into the top portions of the fins.

Abstract: A semiconductor device includes a semiconductor substrate having a plurality of semiconductor fins formed on an upper surface thereof. An epitaxial material is formed on the upper surface of the semiconductor substrate and on an outer surface of the semiconductor fins. The epitaxial material includes an epi upper surface having a lower region that contacts the semiconductor fins and an upper region formed above the lower region. The upper region extends parallel with an upper surface of the semiconductor fins.

Abstract: After forming a gate structure over a semiconductor fin that extends upwards from a semiconductor substrate portion, a sigma cavity is formed within the semiconductor fin on each side of the gate structure. A semiconductor buffer region composed of an un-doped stress-generating semiconductor material is epitaxially growing from faceted surfaces of the sigma cavity. Finally, a doped semiconductor region composed of a doped stress-generating semiconductor material is formed on the semiconductor buffer region to completely fill the sigma cavity. The doped semiconductor region is formed to have substantially vertical sidewalls for formation of a uniform source/drain junction profile.

Abstract: A method of forming a semiconductor device that includes forming a fin structure from a bulk semiconductor substrate and forming an isolation region contacting a lower portion of a sidewall of the fin structure, wherein an upper portion of the sidewall of the fin structure is exposed. A sacrificial spacer is formed on the upper portion of the sidewall of the fin structure. The isolation regions are recessed to provide an exposed section of the sidewall of the fin structure. A doped semiconductor material is formed on the exposed section of the lower portion of the sidewall of the fin structure. Dopant is diffused from the doped semiconductor material to a base portion of the fin structure.

Abstract: Methods and structures that include a vertical-transport field-effect transistor. A first section of a dielectric layer is deposited on a first device region of a substrate and a second section of the dielectric layer is deposited on a second device region of the substrate. A gate stack is deposited on the first device region and the second device region. The gate stack is patterned to define a first gate electrode of the vertical-transport field-effect transistor on the first section of the dielectric layer and a second gate electrode of a high-voltage field-effect transistor on the second section of the dielectric layer. The first section of the dielectric layer is a spacer layer arranged between the first gate electrode and the first device region. The second section of the dielectric layer is a portion of a gate dielectric arranged between the second gate electrode and the second device region.

Abstract: A method of forming a gate structure, including forming one or more vertical fins on a substrate; forming a bottom spacer on the substrate surface adjacent to the one or more vertical fins; forming a gate structure on at least a portion of the sidewalls of the one or more vertical fins; forming a gauge layer on at least a portion of the bottom spacer, wherein the gauge layer covers at least a portion of the gate structure on the sidewalls of the one or more vertical fins; and removing a portion of the gauge layer on the bottom spacer.

Abstract: A semiconductor device includes a semiconductor-on-insulator wafer having a buried layer. The buried layer includes therein opposing etch barrier regions and a gate region between the etch barrier regions. The semiconductor device further includes at least one nanowire having a channel portion interposed between opposing source/drain portions. The channel portion is suspended in the gate region. A gate electrode is formed in the gate region, and completely surrounds all surfaces of the suspended nanowire. The buried layer comprises a first electrical insulating material, and the etch barrier regions comprising a second electrical insulating material different from the first electrical insulating material.

Abstract: A resistive processing unit (RPU) circuit for use in a neural network application includes at least one floating gate storage device, the floating gate storage device including a floating gate, a control gate and an inject/erase gate. The RPU circuit further includes a feedback circuit connected with the floating gate storage device. The feedback circuit is configured to maintain a substantially constant floating gate potential of the floating gate storage device during an update mode of operation of the RPU circuit, and is disabled during a readout mode of operation of the RPU circuit.

Abstract: A method forms a vertical slit transistor includes raised source, drain, and channel regions in a semiconductor substrate. Two gate electrodes are formed adjacent respective sidewalls of the semiconductor substrate. The method forms dielectric material separating the gate electrodes from the source and drain regions.

Abstract: Embodiments are directed to a method of fabricating a portion of a nanowire field effect transistor (FET). The method includes forming a sacrificial layer and a nanowire layer, removing a sidewall portion of the sacrificial layer and forming a diffusion block in a space that was occupied by the removed sidewall portion of the sacrificial layer. The method further includes forming a source region and a drain region such that the diffusion block is between the sacrificial layer and at least one of the source region and the drain region, and removing the sacrificial layer using a sacrificial layer removal process, wherein the diffusion block prevents the sacrificial layer removal process from also removing portions of at least one of the source region and the drain region.

Abstract: Structures for a nanosheet field-effect transistor and methods for forming a structure for a nanosheet field-effect transistor. A fin is formed that includes a first nanosheet channel layer and a second nanosheet channel layer arranged in a vertical stack. A cavity is formed between a portion of the first nanosheet channel layer and a portion of the second nanosheet channel layer. An epitaxially-grown source/drain region is connected with the portion of the first nanosheet channel layer and the portion of the second nanosheet channel layer. A gate structure is formed that includes a section located in a space between the first nanosheet channel layer and the second nanosheet channel layer. The cavity is surrounded by the first nanosheet channel layer, the second nanosheet channel layer, the section of the gate structure, and the source/drain region to define an air gap spacer.

Abstract: Self-aligned contacts of a semiconductor device are fabricated by forming a metal gate structure on a portion of a semiconductor layer of a substrate. The metal gate structure contacts inner sidewalls of a gate spacer. A second sacrificial epitaxial layer is formed on a first sacrificial epitaxial layer. The first sacrificial epitaxial layer is adjacent to the gate spacer and is formed on source/drain regions of the semiconductor layer. The first and second sacrificial epitaxial layers are recessed. The recessing exposes at least a portion of the source/drain regions. A first dielectric layer is formed on the exposed portions of the source/drain regions, and over the gate spacer and metal gate structure. At least one cavity within the first dielectric layer is formed above at least one of the exposed portions of source/drain regions. At least one metal contact is formed within the at least one cavity.