A packaging technology to improve performance of an AI processing system resulting in an ultra-high bandwidth system. An IC package is provided which comprises: a substrate; a first die on the substrate, and a second die stacked over the first die. The first die can be a first logic die (e.g., a compute chip, CPU, GPU, etc.) while the second die can be a compute chiplet comprising ferroelectric or paraelectric logic. Both dies can include ferroelectric or paraelectric logic. The ferroelectric/paraelectric logic may include AND gates, OR gates, complex gates, majority, minority, and/or threshold gates, sequential logic, etc. The IC package can be in a 3D or 2.5D configuration that implements logic-on-logic stacking configuration. The 3D or 2.5D packaging configurations have chips or chiplets designed to have time distributed or spatially distributed processing. The logic of chips or chiplets is segregated so that one chip in a 3D or 2.5D stacking arrangement is hot at a time.

A packaging technology to improve performance of an AI processing system resulting in an ultra-high bandwidth system. An IC package is provided which comprises: a substrate; a first die on the substrate, and a second die stacked over the first die. The first die can be a first logic die (e.g., a compute chip, CPU, GPU, etc.) while the second die can be a compute chiplet comprising ferroelectric or paraelectric logic. Both dies can include ferroelectric or paraelectric logic. The ferroelectric/paraelectric logic may include AND gates, OR gates, complex gates, majority, minority, and/or threshold gates, sequential logic, etc. The IC package can be in a 3D or 2.5D configuration that implements logic-on-logic stacking configuration. The 3D or 2.5D packaging configurations have chips or chiplets designed to have time distributed or spatially distributed processing. The logic of chips or chiplets is segregated so that one chip in a 3D or 2.5D stacking arrangement is hot at a time.

To compensate switching of a dielectric component of a non-linear polar material based capacitor, an explicit dielectric capacitor is added to a memory bit-cell and controlled by a signal opposite to the signal driven on a plate-line.

To compensate switching of a dielectric component of a non-linear polar material based capacitor, an explicit dielectric capacitor is added to a memory bit-cell and controlled by a signal opposite to the signal driven on a plate-line.

A semiconductor device includes: a substrate; a first dielectric layer over the substrate; a memory cell over the substrate in a first region of the semiconductor device, where the memory cell includes a first ferroelectric structure in the first dielectric layer, where the first ferroelectric structure includes a first bottom electrode, a first top electrode, and a first ferroelectric layer in between; and a tunable capacitor over the substrate in a second region of the semiconductor device, where the tunable capacitor includes a second ferroelectric structure, where the second ferroelectric structure includes a second bottom electrode, a second top electrode, and a second ferroelectric layer in between, where at least a portion of the second ferroelectric structure is in the first dielectric layer.

Described is a packaging technology to improve performance of an AI processing system. An IC package is provided which comprises: a substrate; a first die on the substrate, and a second die stacked over the first die. The first die includes memory and the second die includes computational logic. The first die comprises a ferroelectric RAM (FeRAM) having bit-cells. Each bit-cell comprises an access transistor and a capacitor including ferroelectric material. The access transistor is coupled to the ferroelectric material. The FeRAM can be FeDRAM or FeSRAM. The memory of the first die may store input data and weight factors. The computational logic of the second die is coupled to the memory of the first die. The second die is an inference die that applies fixed weights for a trained model to an input data to generate an output. In one example, the second die is a training die that enables learning of the weights.

A transistor may be provided by forming, in a forward order or in a reverse order, a gate electrode, a semiconducting metal oxide liner, a gate dielectric, and an active layer over a substrate, and by forming a source electrode and a drain electrode on end portions of the active layer. The semiconducting metal oxide liner comprises a thin semiconducting metal oxide material that functions as a hydrogen barrier material.

An electronic device comprising a semiconductor memory is provided. The semiconductor memory includes a substrate including a cell region and a peripheral circuit region, the cell region including a first cell region and a second cell region, the first cell region being disposed closer to the peripheral circuit region than the second cell region; second lines disposed over the first lines and extending in a second direction crossing the first direction; memory cells positioned at intersections between the first lines and the second lines in the cell region; a first insulating layer positioned between the first lines, between the second line, or both, in the first cell region; and a second insulating layer positioned between the first lines and between the second lines in the second cell region. A dielectric constant of the first insulating layer is smaller than that of the second insulating layer.

Methods, systems, and devices for thin film transistor deck selection in a memory device are described. A memory device may include memory arrays arranged in a stack of decks formed over a substrate, and deck selection components distributed among the layers to leverage common substrate-based circuitry. For example, each memory array of the stack may include a set of digit lines of a corresponding deck, and deck selection circuitry operable to couple the set of digit lines with a column decoder that is shared among multiple decks. To access memory cells of a selected memory array on one deck, the deck selection circuitry corresponding to the memory array may each be activated, while the deck selection circuitry corresponding to a non-selected memory array on another deck may be deactivated. The deck selection circuitry, such as transistors, may leverage thin-film manufacturing techniques, such as various techniques for forming vertical transistors.

Methods, systems, and devices for thin film transistor deck selection in a memory device are described. A memory device may include memory arrays arranged in a stack of decks formed over a substrate, and deck selection components distributed among the layers to leverage common substrate-based circuitry. For example, each memory array of the stack may include a set of digit lines of a corresponding deck, and deck selection circuitry operable to couple the set of digit lines with a column decoder that is shared among multiple decks. To access memory cells of a selected memory array on one deck, the deck selection circuitry corresponding to the memory array may each be activated, while the deck selection circuitry corresponding to a non-selected memory array on another deck may be deactivated. The deck selection circuitry, such as transistors, may leverage thin-film manufacturing techniques, such as various techniques for forming vertical transistors.