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.

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 memory cell arrangement is provided that may include: a plurality of electrode layers, wherein each of the plurality of electrode layers comprises a plurality of through holes, each of the plurality of through holes extending from a first surface to a second surface of a respective electrode layer; a plurality of electrode pillars, wherein each of the plurality of electrode pillars comprises a plurality of electrode portions, wherein each of the plurality of electrode portions is disposed within a corresponding one of the plurality of through holes; wherein the respective electrode layer and a respective electrode portion of the plurality of electrode portions form a first electrode and a second electrode of a capacitor and wherein at least one memory material portion is disposed in each of the plurality of through holes in a gap between the respective electrode layer and the respective electrode portion.

Some embodiments include an integrated assembly having a pillar of semiconductor material. The pillar has a base region, and bifurcates into two segments which extend upwardly from the base region. The two segments are horizontally spaced from one another by an intervening region. A conductive gate is within the intervening region. A first source/drain region is within the base region, a second source/drain region is within the segments, and a channel region is within the segments. The channel region is adjacent to the conductive gate and is vertically disposed between the first and second source/drain regions. Some embodiments include methods of forming integrated assemblies.

Methods, systems, and devices for operating a ferroelectric memory cell or cells are described. An electronic memory device may include a plurality of plate portions separated by a plurality of segmentation lines, which may be oriented in a plane parallel to rows of a memory array or columns of the memory array, or both. The segmented plate may be employed instead of a single plate for the array. The one or more plate portions may be energized during access operations of a ferroelectric cell in order to create a voltage different across the cell or to facilitate changing the charge of the cell. Each of the plate portions may include one or more memory cells. The memory cells on a plate portion may be read from or written to after the plate portion is activated by a plate driver.

The present disclosure provides a memory circuit, a memory device and an operating method of the memory device. The memory device includes a storage transistor, a variable capacitance device and a control transistor. The variable capacitance device is electrically connected to the gate of the storage transistor, and the control transistor is connected to the storage transistor in series.

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.

Some embodiments include an integrated assembly having bottom electrodes coupled with electrical nodes. Each of the bottom electrodes has a first leg electrically coupled with an associated one of the electrical nodes, and has a second leg joining to the first leg. First gaps are between some of the bottom electrodes, and second gaps are between others of the bottom electrodes. The first gaps alternate with the second gaps. Insulative material and conductive-plate-material are within the first gaps. Scaffold structures are within the second gaps and not within the first gaps. Capacitors include the bottom electrodes, regions of the insulative material and regions of the conductive-plate-material. The capacitors may be ferroelectric capacitors or non-ferroelectric capacitors. Some embodiments include methods of forming integrated assemblies.

A process integration and patterning flow used to pattern a memory array area for an embedded memory without perturbing a fabricating process for logic circuitries. The fabrication process uses a pocket mask (e.g., a hard mask) to decouple the etching process of a memory array area and non-memory area. Such decoupling allows for a simpler fabrication process with little to no impact on the current fabrication process. The fabrication process may use multiple pocket masks to decouple the etching process of the memory array area and the non-memory area. This fabrication process (using multiple pocket masks) allows to avoid exposure of memory material into a second pocket etch chamber. The process of etching memory material is decoupled from the process of etching an encapsulation material. Examples of embedded memory include dynamic random-access memory and ferroelectric random-access memory.

A process integration and patterning flow used to pattern a memory array area for an embedded memory without perturbing a fabricating process for logic circuitries. The fabrication process uses a pocket mask (e.g., a hard mask) to decouple the etching process of a memory array area and non-memory area. Such decoupling allows for a simpler fabrication process with little to no impact on the current fabrication process. The fabrication process may use multiple pocket masks to decouple the etching process of the memory array area and the non-memory area. This fabrication process (using multiple pocket masks) allows to avoid exposure of memory material into a second pocket etch chamber. The process of etching memory material is decoupled from the process of etching an encapsulation material. Examples of embedded memory include dynamic random-access memory and ferroelectric random-access memory.