Different portions of a continuous loop of semiconductor material are electrically isolated from one another. In some embodiments, the end of the loop is electrically isolated from mid-portions of the loop. In some embodiments, loops of semiconductor material, having two legs connected together at their ends, are formed by a pitch multiplication process in which loops of spacers are formed on sidewalls of mandrels. The mandrels are removed and a block of masking material is overlaid on at least one end of the spacer loops. In some embodiments, the blocks of masking material overlay each end of the spacer loops. The pattern defined by the spacers and the blocks are transferred to a layer of semiconductor material. The blocks electrically connect together all the loops. A select gate is formed along each leg of the loops. The blocks serve as sources/drains. The select gates are biased in the off state to prevent current flow from the mid-portion of the loop's legs to the blocks, thereby electrically isolating the mid-portions from the ends of the loops and also electrically isolating different legs of a loop from each other.

A method of manufacturing a semiconductor device is provided including providing a semiconductor substrate, forming a first plurality of semiconductor fins in a logic area of the semiconductor substrate, forming a second plurality of semiconductor fins in a memory area of the semiconductor substrate, forming an insulating layer between the fins of the first plurality of semiconductor fins and between the fins of the second plurality of semiconductor fins, forming an electrode layer over the first and second pluralities of semiconductor fins and the insulating layer, forming gates over semiconductor fins of the first plurality of semiconductor fins in the logic area from the gate electrode layer, and forming sense gates and control gates between semiconductor fins of the second plurality of semiconductor fins in the logic area from the gate electrode layer.

A flash cell includes a gate and an erase gate. The gate is disposed on a substrate, wherein the gate includes a control gate on the substrate and a floating gate having a tip between the substrate and the control gate. The erase gate is disposed beside the gate, wherein the tip points toward the erase gate. The present invention also provides a flash cell forming process including the following steps. A gate is formed on a substrate, wherein the gate includes a floating gate on the substrate. An implantation process is performed on a side part of the floating gate, thereby forming a first doped region in the side part. At least a part of the first doped region is oxidized, thereby forming a floating gate having a tip.

A non-volatile memory has an array of non-volatile memory cells. Each of the non-volatile memory cells includes a coupling device formed on a first well, a read device, a floating gate device formed on a second well and coupled to the coupling device, a program device formed on the second well, and an erase device formed on a third well and coupled to the first floating gate device. The read device, the program device, and the erase device are formed on separate wells so as to separate the cycling counts of a read operation, a program operation and an erase operation of the non-volatile memory cell.

An ISFET includes a control gate coupled to a floating gate in a CMOS device. The control gate, for example, a poly-to-well capacitor, is configured to receive a bias voltage and effect movement of a trapped charge between the control gate and the floating gate. The threshold voltage of the ISFET can therefore by trimmed to a predetermined value, thereby storing the trim information (the amount of trapped charge in the floating gate) within the ISFET itself.

Methods for forming semiconductor structures are provided. The method for forming the semiconductor structure includes forming a word line cell over a substrate and forming a dielectric layer over the word line cell. The method further includes forming a conductive layer over the dielectric layer and polishing the conductive layer until the dielectric layer is exposed. The method further includes forming an oxide layer on a top surface of the conductive layer and removing portions of the conductive layer not covered by the oxide layer to form a memory gate.

The disclosure, belonging to the technological field of semiconductor memory, specifically relates to a semi-floating-gate device which comprises at least a semiconductor substrate, a source region, a drain region, a floating gate, a control gate, a perpendicular channel region and a gated p-n junction diode used to connect the floating gate and the substrate. The semi-floating-gate device disclosed in the disclosure using the floating gate to store information and realizing charging or discharging of the floating gate through a gated p-n junction diode boasts small unit area, high chip density, low operating voltage in data storage and strong ability in data retain.

A method includes forming a shallow trench isolation (STI) region in a substrate, the STI region comprising an etch stop layer; etching the STI region by a first etch to the etch stop layer to form a recess in the STI region; and forming a floating gate, the floating gate comprising a portion that extends into the recess in the STI region, wherein the etch stop layer separates the portion of the floating gate that extends into the recess in the STI region from the substrate.

A flash cell includes a gate and an erase gate. The gate is disposed on a substrate, wherein the gate includes a control gate on the substrate and a floating gate having a tip between the substrate and the control gate. The erase gate is disposed beside the gate, wherein the tip points toward the erase gate. The present invention also provides a flash cell forming process including the following steps. A gate is formed on a substrate, wherein the gate includes a floating gate on the substrate. An implantation process is performed on a side part of the floating gate, thereby forming a first doped region in the side part. At least a part of the first doped region is oxidized, thereby forming a floating gate having a tip.

A semiconductor device and method of making the same are disclosed. The semiconductor device includes a memory gate on a charge storage structure formed on a substrate, a select gate on a gate dielectric on the substrate proximal to the memory gate, and a dielectric structure between the memory gate and the select gate, and adjacent to sidewalls of the memory gate and the select gate, wherein the memory gate and the select gate are separated by a thickness of the dielectric structure. Generally, the dielectric structure comprises multiple dielectric layers including a first dielectric layer adjacent the sidewall of the memory gate, and a nitride dielectric layer adjacent to the first dielectric layer and between the memory gate and the select gate. Other embodiments are also disclosed.