Abstract: An integrated circuit includes a first device having a first source or drain region, and a second device having a second source or drain region that is laterally adjacent to the first source or drain region. A conductive source or drain contact includes (i) a lower portion in contact with the first source or drain region, and extending above the first source or drain region, and (ii) an upper portion extending laterally from above the lower portion to above the second source or drain region. A dielectric material is between at least a section of the upper portion of the conductive source or drain contact and the second source or drain region. In an example, each of the first and second devices is a gate-all-around (GAA) device having one or more nanoribbons, nanowires, or nanosheets as channel regions, or is a finFet structure having a fin-based channel region.

Abstract: A device comprises a substrate and an IC die, which may be a photonic IC. The substrate comprises a first surface, a second surface opposite the first surface, an optical waveguide integral with the substrate, and a hole extending from the first surface to the second surface. The hole comprises a first sidewall. The optical waveguide is between the first surface and the second surface, parallel to the first surface, and comprises a first end which extends to the first sidewall. The IC die is within the hole and comprises a second sidewall and an optical port at the second sidewall. The second sidewall is proximate to the first sidewall and the first end of the optical waveguide is proximate to and aligned with the optical port. The substrate may include a recess to receive another device comprising a socket.

Abstract: Materials and techniques for recessing heterogenous materials in integrated circuit (IC) dies. A first etch may reveal a surface at a desired depth, and a second etch may remove material laterally to reveal sidewalls down to the desired depth of the recess. The first etch may be a cyclical etch, and the second etch may be a continuous etch. The first and second etches may occur in a same chamber. The first and second etches may each be selective to materials with similarities. An IC die may have different, substantially coplanar materials at a recessed surface between and below sidewalls of another material. The recess may have squared profile. The recess may be over transistor structures.

Abstract: Techniques for time-aware remote data transfers. A time may be associated with a remote direct memory access (RDMA) operation in a translation protection table (TPT). The RDMA operation may be permitted or restricted based on the time in the TPT.

Abstract: Techniques are provided herein to form semiconductor devices that include a gate cut formed after the formation of source/drain contacts. In an example, a semiconductor device includes a gate structure around or otherwise on a semiconductor region that extends from a source region to a drain region. Conductive contacts formed over the source and drain regions along a source/drain trench. The gate structure may be interrupted with a dielectric gate cut that further extends past the gate trench and into the source/drain trench where it can cut into one or more of the contacts. The contacts are formed before the gate cut to ensure complete fill of conductive material when forming the contacts. Accordingly, a liner structure on the conductive contacts is also broken by the intrusion of the gate cut and does not extend further up or down the sidewalls of the gate cut.

Abstract: Described herein are techniques for automated hardware power optimization via e-graph based automatic RTL exploration. These techniques provide a tool that automatically performs RTL optimization and generates power optimized RTL without requiring design engineers to perform labor and knowledge intensive manual optimizations.

Abstract: Multiple-ferroelectric capacitor structures in memory devices, including in integrated circuit devices, and techniques for forming the structures. Insulators separating individual outer plates in a ferroelectric capacitor array are supported between wider portions of a shared, inner plate. Wider portions of an inner plate may be formed in lateral recesses between insulating layers. Ferroelectric material may be deposited over the inner plate between insulating layers after removing sacrificial layers. An etch-stop layer may protect the inner plate when sacrificial layers are removed. An etch-stop or interface layer may remain over the inner plate adjacent insulators.

Abstract: Embodiments described herein provide a technique enable a broadcast load from an L1 cache or shared local memory to register files associated with hardware threads of a graphics core. One embodiment provides a graphics processor comprising a cache memory and a graphics core coupled with the cache memory. The graphics core includes a plurality of hardware threads and memory access circuitry to facilitate access to memory by the plurality of hardware threads. The graphics core is configurable to process a plurality of load request from the plurality of hardware threads, detect duplicate load requests within the plurality of load requests, perform a single read from the cache memory in response to the duplicate load requests, and transmit data associated with the duplicate load requests to requesting hardware threads.

Abstract: An integrated circuit includes laterally adjacent first and second devices. The first device includes (i) first source and drain regions, (ii) a first body including semiconductor material laterally extending between the first source and drain regions, (iii) a first sub-fin below the first body, and (iv) a first gate structure on the first body. The second device includes (i) second source and drain regions, (ii) a second body including semiconductor material laterally extending from the second source and drain regions, (iii) a second sub-fin below the second body, and (iv) a second gate structure on the second body. A second dielectric material is laterally between the first and second sub-fins. A third dielectric material is laterally between the first and second sub-fins, and above the second dielectric material. A gate cut including first dielectric material is laterally between the first and second gate structures, and above the third dielectric material.

Abstract: Techniques and mechanisms for a processor to determine an operational mode based on metadata for a page table. In an embodiment, an instruction fetch unit of the processor detects a pointer to a next instruction, in a sequence of instructions, which is to be prepared for execution with a core of the processor. Based on the pointer, a page table is identified as including an entry which indicates a location of the instruction. The page table includes, or otherwise corresponds to, metadata which comprises an identifier of an operational mode of the processor. Based on the metadata, the processor is transitioned to the operational mode in preparation for an execution of the instruction. In another embodiment, the operational mode is one of multiple operational modes which each correspond to a different instruction set architecture.

Abstract: Techniques are provided herein to form a semiconductor device that has a capacitor structure integrated with the source or drain region of the semiconductor device. A given semiconductor device includes one or more semiconductor regions extending in a first direction between corresponding source or drain regions. A gate structure extends in a second direction over the one or more semiconductor regions. A capacitor structure is integrated with one of the source or drain regions of the integrated circuit such that a first electrode of the capacitor contacts the source or drain region and a second electrode of the capacitor contacts a conductive contact formed over the capacitor structure. The capacitor structure may include a ferroelectric capacitor having a ferroelectric layer between the electrodes.

Abstract: Methods and apparatus relating to techniques to provide secure remote debugging are described. In an embodiment, a debugging entity generates and transmits a host token to a device via an interface. The interface provides encrypted communication between the debugging entity and the device. The debugging entity generates a session key based at least in part on the host token and a device token. The debugging entity transmits an acknowledgement signal to the device after generation of the session key to initiate a debug session. The debugging entity transmits a debug unlock key to the device to cause the device to be unlocked for the debug session. Other embodiments are also disclosed and claimed.

Abstract: Techniques are provided herein to form semiconductor devices that include gate cuts with different widths (e.g., at least a 1.5× difference in width) but substantially the same height (e.g., less than 5 nm difference in height). A given gate structure extending over one or more semiconductor regions may be interrupted with any number of gate cuts that each extend through an entire thickness of the gate structure. According to some embodiments, gate cuts of a similar first width are formed via a first etching process while gate cuts of a similar second width that is greater than the first width are formed via a second etching process that is different from the first etching process. Using different etch processes for gate cuts of different widths maintains a similar height for the gate cuts of different widths.

Abstract: A non-linear activation function in a neural network may be approximated by one or more linear functions. The input range may be divided into input segments, each of which corresponds to a different exponent in the input range of the activation function and includes input data elements having the exponent. Target accuracies may be assigned to the identified exponents based on a statistics analysis of the input data elements. The target accuracy of an input segment will be used to determine one or more linear functions that approximate the activation function for the input segment. An error of an approximation of the activation function by a linear function for the input segment may be within the target accuracy. The parameters of the linear functions may be stored in a look-up table (LUT). During the execution of the DNN, the LUT may be used to execute the activation function.

Abstract: Inverted pillar capacitors that have a U-shaped insulating layer are oriented with the U-shaped opening of the insulating layer opening toward the surface of the substrate on which the inverted pillar capacitors are formed. The bottom electrodes of adjacent inverted pillar capacitors are isolated from each other by the insulating layers of the adjacent electrodes and the top electrode that fills the volume between the electrodes. By avoiding the need to isolate adjacent bottom electrodes by an isolation dielectric region, inverted pillar capacitors can provide for a greater capacitor density relative to non-inverted pillar capacitors. The insulating layer in inverted pillar capacitors can comprise a ferroelectric material or an antiferroelectric material. The inverted pillar capacitor can be used in memory circuits (e.g., DRAMs) or non-memory applications.

Abstract: An apparatus comprising first circuitry to determine a time parameter associated with a storage operation; and second circuitry to generate a storage command, the storage command including the time parameter, a location for the storage operation, and an opcode specifying the storage operation.

Abstract: For example, an apparatus may include an input to receive an input signal in a first voltage domain; a multi-mode power amplifier switchable between a plurality of power modes to generate an output signal based on the input signal; and an output to provide the output signal. For example, the multi-mode power amplifier may be configured to provide the output signal in the first voltage domain at a first power mode, and to provide the output signal in a second voltage domain at a second power mode. For example, a maximal voltage of the second voltage domain may be at least two times a maximal voltage of the first voltage domain.

Abstract: Technologies for a transistor with a thin-film ferroelectric gate dielectric are disclosed. In the illustrative embodiment, a transistor has a thin layer of scandium aluminum nitride (ScxAl1-xN) ferroelectric gate dielectric. The channel of the transistor may be, e.g., gallium nitride or molybdenum disulfide. In one embodiment, the ferroelectric polarization changes when voltage is applied and removed from a gate electrode, facilitating switching of the transistor at a lower applied voltage. In another embodiment, the ferroelectric polarization of a gate dielectric of a transistor changes when the voltage is past a positive threshold value or a negative threshold value. Such a transistor can be used as a one-transistor memory cell.

Abstract: For example, a phase shifter may include an input to receive an input clock signal having an input frequency and an input phase. For example, the phase shifter may include a quadrature phase-shift generator configured to generate a first signal and a second signal based on the input clock signal, the first and second signals having the input frequency, wherein a phase of the first signal is based on the input phase, wherein a phase of the second signal is shifted by a quadrature phase-shift relative to the phase of the first signal. For example, the phase shifter may include an output to provide an output based on the first signal and the second signal.

Abstract: An integrated circuit (IC) device comprises a substrate comprising a glass core. The glass core comprises a first surface and a second surface opposite the first surface, and a first sidewall between the first surface and the second surface. The glass core may include a conductor within a through-glass via extending from the first surface to the second surface and a build-up layer. The glass cord comprises a plurality of first areas of the glass core and a plurality of laser-treated areas on the first sidewall. A first one of the plurality of laser-treated areas may be spaced away from a second one of the plurality of laser-treated areas. A first area may comprise a first nanoporosity and a laser-treated area may comprise a second nanoporosity, wherein the second nanoporosity is greater than the first nanoporosity.

Abstract: In one example an apparatus comprises processing circuitry to measure a statistical distance between a marginal distribution of a coordinate of a potential signature (z) over a first interval and a uniform distribution over the first interval and use the statistical distance to determine one or more thresholds of a rejection sampling operation in a lattice-based digital signature algorithm. Other examples may be described.

Abstract: Microelectronic integrated circuit package structures include a first substrate comprising a first bond plane structure on a surface of the first substrate, and a second substrate comprising a second bond plane structure on a surface of the second substrate, where the first and second bond plane structures are in direct physical contact. A conductive trace on the surface of the first substrate is adjacent to a bonding interface between the first and second bond plane structures and over a recessed surface of the first substrate. A first air gap is between the conductive trace and the recessed surface of the first substrate and a second air gap is between the conductive trace and the bonding interface.

Abstract: An apparatus to facilitate single precision support for systolic pipeline in a graphics environment is disclosed. The apparatus includes a processor comprising systolic array hardware including a plurality of data processing units, wherein the systolic array hardware is to: receive data for performance of a matrix multiplication operation in a first precision format; convert an original value of the data into two split values with a second precision format having a lower precision than the first precision format; perform the matrix multiplication operation using the two split values in the second precision format, the matrix multiplication operation comprising a split-term operation that utilizes two passes through the systolic array hardware with feedback wiring and local reduction; and generate an emulated result for the matrix multiplication operation in the first precision format.

Abstract: Microelectronic integrated circuit package structures include a first substrate coupled to a second substrate by a conductive interconnect structure and a dielectric material adjacent to the conductive interconnect structure. A cavity in a surface of the first substrate is adjacent to the conductive interconnect structure. A portion of the dielectric material is within the cavity.

Abstract: An apparatus to facilitate hardware enhancements for double precision systolic support is disclosed. The apparatus includes matrix acceleration hardware having double-precision (DP) matrix multiplication circuitry including a multiplier circuits to multiply pairs of input source operands in a DP floating-point format; adders to receive multiplier outputs from the multiplier circuits and accumulate the multiplier outputs in a high precision intermediate format; an accumulator circuit to accumulate adder outputs from the adders with at least one of a third global source operand on a first pass of the DP matrix multiplication circuitry or an intermediate result from the first pass on a second pass of the DP matrix multiplication circuitry, wherein the accumulator circuit to generate an accumulator output in the high precision intermediate format; and a down conversion and rounding circuit to down convert and round an output of the second pass as final result in the DP floating-point format.

Abstract: An integrated circuit (IC) device comprises a substrate comprising a glass core. The glass core includes a first surface, a second surface opposite the first surface, a sidewall between the first surface and the second surface, and a corner region where the first sidewall meets the first surface. A first build-up layer is on at least the first surface. In some embodiments, the corner region comprises a recess and a dielectric material within the recess. In other embodiments, the corner region comprises a first compressive stress and the glass core comprises a second region. The second region comprises a second compressive stress. The first compressive stress is greater than the second compressive stress.

Abstract: An apparatus to facilitate ordered thread dispatch for thread teams is disclosed. The apparatus includes one or more processors including a graphic processor, the graphics processor including a plurality of processing resources, and wherein the graphics processor is to: allocate a thread team local identifier (ID) for respective threads of a thread team comprising a plurality of hardware threads that are to be executed solely by a processing resource of the plurality of processing resources; and dispatch the respective threads together into the processing resource, the respective threads having the thread team local ID allocated.

Abstract: An electronic device can include an interposer, a first porous polymer layer, and one or more die. The interposer can include a metallic through via extending from a first surface of the interposer to a second surface of the interposer. The first polymer layer can be adjacent to the first surface of the interposer. The one or more dies can be coupled to the first porous polymer layer and connected to the metallic through via.

Abstract: Multi-die packages including a glass substrate within a space between adjacent IC dies. Two or more IC die may be placed within recesses formed in a glass substrate. The IC die and glass substrate, along with any conductive vias extending through the glass substrate may be planarized. Organic package dielectric material may then be built up on both sides of the IC dies and glass substrate. Metallization features formed within package dielectric material built up on a first side of the IC die may electrically interconnect the IC dies to package interconnect interfaces that may be further coupled to a host with solder interconnects. Metallization features formed within package dielectric material built up on a second side of the first and second IC dies may electrically interconnect the first IC die to the second IC die.

Abstract: Technologies for a field effect transistor (FET) with a ferroelectric gate dielectric are disclosed. In an illustrative embodiment, a perovskite stack is grown on a buffer layer as part of manufacturing a transistor. The perovskite stack includes one or more doped semiconductor layers alternating with other lattice-matched layers, such as undoped semiconductor layers. Growing the doped semiconductor layers on lattice-matched layers can improve the quality of the doped semiconductor layers. The lattice-matched layers can be preferentially etched away, leaving the doped semiconductor layers as fins for a ribbon FET. In another embodiment, an interlayer can be deposited on top of a semiconductor layer, and a ferroelectric layer can be deposited on the interlayer. The interlayer can bridge a gap in lattice parameters between the semiconductor layer and the ferroelectric layer.

Abstract: Described herein is a technique to enable the construction of hierarchical clock gating architectures via e-graph rewriting. Automated clock gating relies on multiplexor (mux) tree analysis and constructs simple register enable signals. A framework is provided to detect non-mux based opportunities and construct more complex clock gating signals.

Abstract: Techniques related to training and implementing convolutional neural networks for object recognition are discussed. Such techniques may include applying, at a first convolutional layer of the convolutional neural network, 3D filters of different spatial sizes to an 3D input image segment to generate multi-scale feature maps such that each feature map has a pathway to fully connected layers of the convolutional neural network, which generate object recognition data corresponding to the 3D input image segment.

Abstract: Techniques and mechanisms for circuitry of a processor to determine a count of prefetch instructions which have been retired, or are designated for retirement. In an embodiment, a performance monitoring unit (PMU) monitors the execution of an instruction sequence by a core of said processor. The PMU detects the retirement of a first instruction, and further makes a first determination that the instruction is of a prefetch instruction type. Based on the first determination, counter circuitry of the processor updates a count of one or more instruction retirements, wherein each such retired instruction is of the prefetch instruction type. The PMU further makes a second determination that another retired second instruction is of a non-prefetch instruction type. In another embodiment, the counter circuitry prevents any updating of that same count based on the second determination.

Abstract: Integrated circuit interconnect structures including a niobium-based barrier material. In some embodiments, a layer of essentially niobium may be sputter deposited, for example to a thickness of less than 8 nm at a bottom of an interconnect via. A copper-based fill material may then be deposited over the niobium barrier material. Integrated circuit interconnect metallization may comprise some layers of metallization that have a tantalum-based barrier and other layers of metallization that have a niobium-based barrier.

Abstract: Techniques for prefetching by a hardware processor are described. In certain examples, a hardware processor includes execution circuitry, cache memories, and prefetcher circuitry. The execution circuitry is to execute instructions to access data at a memory address. The cache memories include a first cache memory at a first cache level and a second cache memory at a second cache level. The prefetcher circuitry is to prefetch the data from a system memory to at least one of the plurality of cache memories, and it includes a first-level prefetcher to prefetch the data to the first cache memory, a second-level prefetcher to prefetch the data to the second cache memory, and a plurality of prefetch filters. One of the prefetch filters is to filter exclusively for the first-level prefetcher. Another of the prefetch filters is to maintain a history of demand and prefetch accesses to pages in the system memory and to use the history to provide training information to the second-level prefetcher.

Abstract: Methods and apparatus relating to an efficient implementation of ZUC authentication are described. In one embodiment, a processor computes a tag update, based at least in part on stored data, for an authentication operation. The tag update is computed by replacing a ‘for’ loop with a carry-less multiply operation. Other embodiments are also claimed and disclosed.

Abstract: In an example, an apparatus comprises at least one execution platform; and logic, at least partially including hardware logic, to receive a trained neural network model in a model optimizer and convert the trained neural network model to an optimized model comprising parameters that are fit to the at least one execution platform. Other embodiments are also disclosed and claimed.

Abstract: Apparatuses, memory systems, capacitor structures, and techniques related to anti-ferroelectric capacitors having a cerium oxide doped hafnium zirconium oxide based anti-ferroelectric are described. A capacitor includes layers of hafnium oxide, cerium oxide, and zirconium oxide between metal electrodes. The cerium of the cerium oxide provides a mid gap state to protect the hafnium zirconium oxide during operation.

Abstract: A memory device includes a group of ferroelectric capacitors with a shared plate that extends through the ferroelectric capacitors, has a greatest width between ferroelectric capacitors, and is coupled to an access transistor. The shared plate may be vertically between ferroelectric layers of the ferroelectric capacitors at the shared plate's greatest width. The memory device may include an integrated circuit die and be coupled to a power supply. Forming a group of ferroelectric capacitors includes forming an opening through an alternating stack of insulators and conductive plates, selectively forming ferroelectric material on the conductive plates rather than the insulators, and forming a shared plate in the opening over the ferroelectric material.

Abstract: Embodiments provides for a package substrate, including: a core comprising insulative material; first conductive traces in contact with a surface of the core; and buildup layers in contact with the first conductive traces and the surface of the core, the buildup layers comprising second conductive traces in an organic dielectric material. The first conductive traces comprise at least a first metal and a second metal, the first conductive traces comprise a first region proximate to and in contact with the core and a second region distant from the core, parallel and opposite to the first region, a relative concentration of the first metal to the second metal is higher in the first region than in the second region, and the relative concentration of the first metal to the second metal between the first region and the second region varies non-uniformly.

Abstract: An integrated circuit (IC) device comprises a substrate comprising a glass core. The glass core includes a first surface, a second surface opposite the first surface, and a sidewall between the first surface and the second surface. A build-up layer is on at least the first surface. A plurality of regions is on the sidewall. Each region comprises a cavity in the sidewall, wherein the cavity spans a first distance in a first direction from the first surface toward the second surface. In addition, the cavity comprises a concave surface having a first depth at the first surface and a second depth at the first distance, the second depth being less than the first depth.

Abstract: Techniques and mechanisms for storing data with a memory cell which comprises a ferroelectric (FE) resistive junction. In an embodiment, a memory cell comprises a transistor and a FE resistive junction structure which is coupled to the transistor. The FE resistive junction structure comprises electrode structures, and a layer of a material which is between said electrode structures, wherein the material is a FE oxide or a FE semiconductor. The FE resistive junction structure selectively provides any of various levels of resistance, each to represent a respective one or more bits. A current flow through the FE resistive junction structure is characterized by thermionic emission through a Schottky barrier at an interface with one of the electrode structures. In another embodiment, the FE resistive junction structure further comprises one or more dielectric layers each between the layer of material and a different respective one of the electrode structures.

Abstract: Techniques and mechanisms for operating a ferroelectric (FE) circuit element as a cell of a crossbar memory array. In an embodiment, the crossbar memory array comprises a bit line, a word line, and a data storage cell which includes a circuit element that extends to each of the bit line and the word line. The data storage cell is a FE circuit element which comprises terminals, each at a different respective one of the bit line or the word line, and one or more material layers between said terminals. One such layer comprises a FE nitride or a FE oxide. The FE circuit element is operable to selectively enable, or disable, operation as a diode. In another embodiment, the memory array is coupled to circuitry which corresponds a given mode of operation of the FE circuit element to a particular data bit value.

Abstract: Multi-die packages including at least one glass substrate within a space between two adjacent IC dies or surrounding an interconnect bridge die. The various IC dies may be placed within recesses formed in the glass substrate. The IC die and glass substrate, along with any conductive vias extending through the glass substrate may be planarized. The bridge die may be directly bonded or soldered to the adjacent IC dies, providing fine pitch interconnect. The opposite side of the adjacent IC dies and glass substrate may be attached to a host component or may be built up with package dielectric material. Metallization features formed on the second side of the glass substrate may electrically interconnect the IC dies to package interconnect interfaces that may be further coupled to a host with solder interconnects.

Abstract: Methods, device structures, and wafer treatment chemistries related to backside wafer treatments to reduce distortions and overlay errors due to wafer deformation during wafer chucking are described. A backside layer is applied to the wafer prior to chucking. The chemistry of the backside layer lowers the surface free energy of the wafer during chucking to eliminate or mitigate wafer deformation during wafer processing.

Abstract: An apparatus is provided which comprises: a plurality of logic blocks comprising transistors on a substrate, the logic blocks to implement logic functions; a plurality of input/output (I/O) blocks connecting the logic blocks with components external to the apparatus; a plurality of interconnect layers comprising wires and vias surrounded by interlayer dielectric above the substrate, the wires and vias conductively coupling the plurality of logic blocks and the plurality of I/O blocks; a plurality of programmable switches to configure connections between the plurality of logic blocks and the plurality of I/O blocks; and a ferroelectric material in a capacitor coupled to the gate or on the gate dielectric itself of one or more of the transistors. Other embodiments are also disclosed and claimed.

Abstract: An apparatus is provided which comprises: an interposer comprising glass, one or more redistribution layers on a first interposer surface, one or more conductive contacts on a second interposer surface opposite the first interposer surface, one or more vias through the interposer coupling at least one of the conductive contacts on the second interposer surface with the redistribution layers on the first interposer surface, an integrated circuit device embedded within a cavity in the interposer between the first and second interposer surfaces, the embedded integrated circuit device coupled with a first redistribution layers surface, a stack of two or more integrated circuit devices coupled with a second redistribution layers surface opposite the first redistribution layers surface, and mold material surrounding at least one side of the stack of two or more integrated circuit devices. Other embodiments are also disclosed and claimed.

Abstract: Techniques are provided to form semiconductor devices that include a gate cut that passes through a plurality of semiconductor bodies (e.g., nanoribbons or nanosheets) such that the gate cut acts as a dielectric spine in a forksheet arrangement with the semiconductor bodies on either side of the gate cut. In an example, two semiconductor devices in a forksheet arrangement include semiconductor bodies directly on either side of a dielectric spine. A gate structure includes a gate dielectric (e.g., high-k gate dielectric material) and a gate electrode (e.g., conductive material such as workfunction material and/or gate fill metal) that extends around each of the semiconductor bodies of both semiconductor devices. The dielectric spine interrupts the entire height of the gate structure between the two devices and includes dielectric material (e.g., low-k dielectric), and the gate dielectric of the gate structure is not present along sidewalls of the spine between adjacent bodies.

Abstract: Backside integrated circuit capacitor structures. In an example, a capacitor structure includes a layer of ferroelectric material between first and second electrodes. The first electrode can be connected to a transistor terminal by a backside contact that extends downward from a bottom surface of the transistor terminal to the first electrode. The transistor terminal can be, for instance, a source or drain region, and the backside contact can be self-aligned with the source or drain region. The second electrode can be connected to a backside interconnect feature. In some cases, the capacitor has a height that extends through at least one backside interconnect layer. In some cases, the capacitor is a multi-plate capacitor in which the second conductor is one of a plurality of plate line conductors arranged in a staircase structure. The capacitor structure may be, for example, part of a non-volatile memory device or the cache of a processor.

Abstract: An integrated circuit includes a first device and a laterally adjacent second device. The first device includes a first body including semiconductor material extending from a first source region to a first drain region, and a first gate structure on the first body. The second device includes a second body including semiconductor material extending from a second source region to a second drain region, and a second gate structure on the second body. A gate cut including dielectric material is between and laterally separates the first gate structure and the second gate structure. The first body is separated laterally from the gate cut by a first distance, and the second body is separated laterally from the gate cut by a second distance. In an example, the first and second distances differ by at least 2 nanometers. In an example, the first and second devices are fin-based devices or gate-all-around devices.