Taiwan Semiconductor Manufacturing Comp. Ltd.

Yai-Fen Lin

Chia-Ta Hsieh

Di-Son Kuo

Hung-Cheng Sung

Jack Yeh

Chuang-Ke Yeh

Juang-Ke Yeh

Wen-Ting Chu

Shiou-Hann Liaw

Jenn Tsao

Jian-Hsing Lee

Hung-Der Su

Chrong Jung Lin

Jong Chen

Chang-Song Lin

Din-Son Kuo

Chung-Ker Yeh

Hong-Cheng Sung

Kuo-Reay Peng

Jung-Ke Yeh

Chang Song Lin

Di-Son Kou

Jaung-Ke Yeh

Chuang-ke Yeh

A split-gate flash memory cell having a three-dimensional source capable of three-dimensional coupling with the floating gate of the cell, as well as a method of forming the same are provided. This is accomplished by first forming an isolation trench, lining it with a conformal oxide, then filling with an isolation oxide and then etching the latter to form a three-dimensional coupling region in the upper portion of the trench. A floating gate is next formed by first filling the three-dimensional region of the trench with polysilicon and etching it. The control gate is formed over the floating gate with an intervening inter-poly oxide. The floating gate forms legs extending into the three-dimensional coupling region of the trench thereby providing a three-dimensional coupling with the source which also assumes a three-dimensional region. The leg or the side-wall of the floating gate forming the third dimension provides the extra area through which coupling between the source and the floating gate is increased. In this manner, a higher coupling ratio is achieved without an increase in the cell size while at the same time alleviating the punchthrough and junction break-down of source region by sharing gate voltage along the side-wall.

Method to increase coupling ratio of source to floating gate in split-gate flash

A method is provided for forming a split-gate flash memory cell having a shallow trench isolation without the intrusion of a &ldquo;smiling&rdquo; gap near the edge of the trench encompassing the first polysilicon layer. This is accomplished by forming two conformal layers lining the interior walls of the trench. An exceptionally thin nitride layer overlying the first conformal oxide layer provides the necessary protection during the oxidation of the first polysilicon layer so as to prevent the &ldquo;smiling&rdquo; effect normally encountered in fabricating ultra large scale integrated circuits.

Method to fabricate a non-smiling effect structure in split-gate flash with self-aligned isolation

A stacked-gate flash memory cell having a shallow trench isolation with a high-step of oxide and high lateral coupling is described. An unconventionally high isolation oxide layer is formed in a shallow trench isolation (STI) in a substrate. The deep opening in the space between the STIs is conformally lined with a polysilicon to form a floating gate extending above the opening. A conformal intergate oxide lines the entire floating gate. A layer of polysilicon overlays the intergate oxide and protrudes downward into the openings to form a control gate with increased coupling to the floating gate.

Stacked-gate flash memory cell with folding gate and increased coupling ratio

A PIP (Poly-Interpoly-Poly) capacitor with high capacitance is provided in a split-gate flash memory cell. A method is also disclosed to form the same PIP capacitor where the bottom and top plates of the capacitor are formed simultaneously with the floating gate and control gate, respectively, of the split-gate flash memory cell. Furthermore, the thin interpoly oxide of the cell, rather than the thick poly-oxide over the floating gate is used as the insulator between the plates of the capacitor. The resulting capacitor yields high storage capacity through high capacitance per unit area.

PIP capacitor for split-gate flash process

A method is provided for forming a short and sharp gate bird's beak in order to increase the erase speed of a split-gate flash memory. This is accomplished in two embodiments where in the first, fluorine is implanted in the first polysilicon layer to form the floating gate. It is disclosed here that the implanting of fluorine increases the oxidation rate of the polysilicon and because of the faster oxidation, the polygate bird's beak that is formed attains a relatively short and sharp in comparison with conventional beaks. This has the attendant benefit of forming a relatively small memory cell, and the concomitant reduction in the erase speed of the cell. In the second embodiment, oxygen is used with the same favorable results. A third embodiment discloses the structure of a split-gate flash memory cell having a sharp bird's beak.

Method of forming sharp beak of poly by oxygen/fluorine implant to improve erase speed for split-gate flash

A method is provided to form a sharp poly tip to improve the speed of a split-gate flash memory. The sharp poly tip is provided in place of the conventional gate bird's beak (GBB) because the latter requires the forming of thick poly-oxide which is more and more difficult in the miniaturized circuits of the ultra scale integrated technology. Furthermore, it is well known that GBB encroaches under the gate edge in a split-gate flash and degrades the programmability of submicron memory cells. The sharp poly tip of the invention is provided by forming a tapered floating gate through a high pressure etch such that the tip of the upper edge of the floating gate under the poly oxide is sharper and more robust, and, therefore, less susceptible to damage during the manufacture of the cell. The invention is also directed to a semiconductor device fabricated by the disclosed method.

Method to fabricate poly tip in split gate flash

A split gate P-channel flash memory cell and method of forming a split gate P-channel flash memory cell which avoids of high erasing voltage, reverse tunneling during programming, drain disturb and over erase problems, and permits shrinking the cell dimensions. The control gate has a concave top surface which intersects with the sidewalls to form a sharp edge. The cell is programmed by charging the floating gate with electrons by means of hot electron injection from the channel into the floating gate. The cell is erased by discharging the excess electrons from the floating gate into the control gate using Fowler-Nordheim tunneling. The sharp edge at the intersection of the concave top surface and the sidewalls of the floating gate produces a high electric field between the control gate and the floating gate to accomplish the Fowler-Nordheim tunneling with only moderate voltage differences between the floating gate and control gate. The P-channel flash memory cell has a higher impact ionization rage for creating hot electrons so that the distance between the source and drain junctions and the length of the floating gate can be kept small thereby permitting the dimensions of the flash memory cell to be shrunk.

Source side injection programming and tip erasing P-channel split gate flash memory cell

A novel method of forming a first polysilicon gate tip (poly-tip) for enhanced F-N tunneling in split-gate flash memory cells is disclosed. The poly-tip is formed in the absence of using a thick polysilicon layer as the floating gate. This is made possible by forming an oxide layer over the poly-gate and oxidizing the sidewalls of the polygate. Because the starting thickness of polysilicon of the floating gate is relatively thin, the resulting gate beak, or poly-tip, is also necessarily thin and sharp. This method, therefore, circumvents the problem of oxide thinning encountered in scaling down devices of the ultra large scale integration technology and the fast programmability and erasure performance of EEPROMs is improved.

Split-gate flash cell

A method is disclosed for forming a split-gate flash memory cell having a protruding source in place of the conventional flat source. The vertically protruding source structure has a top portion and a bottom portion. The bottom portion is polysilicon while the top portion is poly-oxide. The vertical wall of the protruding structure over the source is used to form vertical floating gate and spacer control gate with an intervening inter-gate oxide. Because the coupling between the source and the floating gate is now provided through the vertical wall, the coupling area is much larger than with conventional flat source. Furthermore, there is no longer the problem of voltage punch-through between the source and the drain. The vertical floating gate is also made thin so that the resulting thin and sharp poly-tip enhances further the erasing and programming speed of the flash memory cell. The vertical orientation of the source structure and the floating gate and the self-alignment of the spacer control gate to the floating gate together makes it possible to reduce the memory cell substantially.

Structure with protruding source in split-gate flash

A flash EEPROM or split gate flash EEPROM is made on a doped silicon semiconductor N-well formed in a doped semiconductor substrate. A channel with a given width is formed in the N-well which is covered with a tunnel oxide layer, and an N&plus; doped polysilicon floating gate electrode layer, which can be patterned into a split gate floating gate electrode having a narrower width than the channel width. An interelectrode dielectric layer is formed over the floating gate electrode and the exposed tunnel oxide. A control gate electrode includes a layer composed of P&plus; doped polysilicon over the interelectrode dielectric layer. The tunnel oxide layer, the floating gate electrode layer, the interelectrode dielectric layer, and the control gate electrode are patterned into a gate electrode stack above the channel. A source region and a drain region are formed in the surface of the substrate with a P type of dopant, the source region and the drain region being self-aligned with the gate electrode stack.

P-channel EEPROM and flash EEPROM devices

In this invention polysilicon sidewalls on a semiconductor substrate are used as split gate flash memory cells. The sidewalls are formed around a core of silicon nitride and left standing once the silicon nitride is removed. Bit lines are implanted into the semiconductor substrate and extend partially under the sidewalls to allow the operation of the floating gates with respect to the buried bit line which act as drains and sources. A control gate is deposited over a row of sidewalls orthogonal to the bit lines and extending the length of a flash memory word line. The polysilicon sidewall split gate flash memory cells are programmed, read and erased by a combination of voltages applied to the control gate and the bit lines partially underlying the sidewalls.

Split gate flash cell for multiple storage

A split gate EEPROM memory device formed on a doped silicon semi-conductor substrate starting with an initial oxide layer with an undoped first polysilicon layer formed thereon. A polysilicon oxide hard mask over the undoped first polysilicon layer for use in patterning the initial oxide layer and the undoped first polysilicon layer which are then etched to form a floating gate electrode stack from the undoped first polysilicon layer and the initial oxide layer on the substrate. Then form a tunnel oxide layer and a doped polysilicon and pattern them into control gate electrode stack, with the control gate electrode stack being located in a split-gate configuration with respect to the floating gate electrode stack.

Undoped polysilicon as the floating-gate of a split-gate flash cell

A method is disclosed to form a split-gate flash memory cell having nitride spacers formed on a pad oxide and prior the forming of an inter-poly oxide layer thereover. In this manner, any damage that would normally occur to the inter-poly oxide during the etching of the nitride spacers subsequent to the forming of the inter-poly oxide is avoided. Consequently, the variation in the thickness of the inter-poly oxide due to the unpredictable damage to the underlying spacers is also avoided by reversing the order in which the spacers and the inter-poly oxide are formed, including the forming of the pad oxide first. As a result, variation in the erase speed of the inter-gate flash memory cell is prevented, both for cells fabricated on the same wafer as well as on different wafers on same or different production lines.

Split gate flash memory device having nitride spacer to prevent inter-poly oxide damage

An FET semiconductor device includes an N-region and a P-region formed in the substrate with the N-region juxtaposed with the P-region with an interface between the N-region and the P-region and with a first channel in the N-region and a second channel in the P-region. An N&plus; drain region is near the interface on one side of the first channel in the P-region. A P&plus; drain region is near the interface on one side of the second channel in the N-region. An N&plus; source region is on the opposite side of the first channel from the interface in the P-region. A P&plus; source region is on the opposite side of the first channel from the interface in the N-region. A wide gate electrode EEPROM stack bridges the channels in the N-region and the P-region. The stack includes a tunnel oxide layer, a floating gate electrode layer, an interelectrode dielectric layer, and a control gate electrode. An N&plus; drain region is formed in the surface of the P-region self-aligned with the gate electrode stack. A P&plus; drain region is formed in the surface of the N-region self-aligned with the gate electrode stack.

Growing community of inventors

Non-Tour, Taiwan

Yai-Fen Lin