Abstract:
Low resistivity interconnects and silicided N+/P+ active area are formed by sputtering a blanket layer of titanium (33) onto a wafer (13) surface which has the interconnection pattern defined by a thin polysilicon layer (29). A thin oxide layer (27) underneath the polysilicon (29) interconnection pattern serves as an etch stop in the local interconnect photo/etch step. This unprotected oxide layer remaining on top of the N+/P+ active area will be removed by a wet etch prior to the sputtering step. Titanium is then sintered, resulting in conversion of titanium to TiSi2 and TiN. Unreacted titanium and TiN are removed and the remaining TiSi2 provides silicided N+/P+ junctions and a low resistance interconnect level. Advantages of the invention include a low resistance self-aligned contact procedure for the bitline formation, and elimination of the minimal spacing requirement between access transistors imposed by the traditional non self-aligned contact approach.
Abstract translation:通过将钛(33)的覆盖层溅射到具有由多晶硅薄层(29)限定的互连图案的晶片(13)表面上而形成低电阻率互连和硅化N + / P +有源区。 在多晶硅(29)互连图案下方的薄氧化物层(27)用作局部互连光刻/蚀刻步骤中的蚀刻停止。 残留在N + / P +有效面积顶部的未保护的氧化物层将在溅射步骤之前通过湿式蚀刻去除。 然后将钛烧结,导致钛转化为TiSi 2和TiN。 去除未反应的钛和TiN,剩余的TiSi 2提供硅化N + / P +结和低电阻互连水平。 本发明的优点包括用于位线形成的低电阻自对准接触程序,以及消除由传统的非自对准接触方法施加的存取晶体管之间的最小间隔要求。
Abstract:
An electrode structure for use in a chalcogenide memory is disclosed. The electrode has a substantially frusto-conical shape, and is preferably formed by undercut etching a polysilicon layer beneath an oxide pattern. With this structure, improved current densities through the chalcogenide material can be achieved.
Abstract:
A row decoder circuit operates in a memory integrated circuit, such as a dynamic random access memory (DRAM), having an array of memory cells including row and columns. An address decode tree circuit receives address signals and provides decode signals being activated based on the state of the address signals. Row line driver circuits receive corresponding ones of the decode signals and an enable signal. Each row line driver circuit fires a corresponding row line when the enable signal is activated and the corresponding one of the decode signals is activated. Delay circuitry delays certain of the address signals to stagger the activation of certain of the decode signals to permit multiple row lines to fire in a single row address strobe (RAS) cycle.
Abstract:
A method for chemical-mechanical planarization of a substrate on a fixed-abrasive polishing pad in which a planarizing solution is dispensed onto the fixed-abrasive polishing pad. The planarizing solution is preferably an abrasive-free planarizing solution that oxidizes a surface layer on the substrate without passing the surface layer into solution, and the fixed-abrasive pad has a substantially uniform distribution of abrasive particles fixedly bonded to a suspension medium. The surface layer of the substrate is then pressed against the fixed-abrasive pad in the presence of planarizing solution, and at least one of the fixed-abrasive pad or the substrate moves relative to the other to remove material from the surface of the substrate. In operation, the planarizing solution forms a rough, scabrous layer of non-soluble oxides on the surface layer that are readily removed by the abrasive surface of the polishing pad. In one embodiment of the invention, the pH of the planarizing solution is controlled to oxidize the material of the surface layer without passing it into solution.
Abstract:
A supervoltage circuit has been described which uses a resistor divider as an input stage. The resistor divider decreases the dependency of the supervoltage trip point on transistor threshold voltages (Vt). The stability of supervoltage trip point is significantly increased over traditional supervoltage circuits using diode connected transistors as an input stage. The supervoltage circuit can be included in any integrated circuit including memory devices.
Abstract:
A dynamic memory (14) having a refresh control circuit (18) is comprised of a main array (15) of memory cells organized according to addresses. A first circuit (16, 41) is provided for performing external operations on addresses within the main array (15) of memory cells. A refresh data memory array (20) and a timing circuit (30) for producing timing information (32) related to the refresh rate of the cells of the main array (15) are provided. A write driver circuit (34) is responsive to a refresh signal (43) and a request for the performance of an external operation on a selected address in the main array (15) for writing a portion of the timing information (32) to the refresh data memory array (20) according to addresses corresponding to addresses used in the main array. A counter (36) generates a plurality of addresses and a second circuit (40), responsive to the generated addresses, reads the written portion of the timing information from the refresh data array (20). A logic circuit (42) is responsive to the portion of the timing information read from the refresh data array and the then current value of the timing information for producing the refresh signal (43) indicative of when a refresh is required. An arbitration circuit (48, 50, 56) selects either the refresh signal (43) or the request for the performance of an external operation. Select circuits (38, 39) are responsive to the arbitration circuit (48, 50, 56) for selectively conducting either the generated address or the selected address to row decoders (40, 41) within the first and second circuits to ensure that the main array (15) of memory cells and the refresh data memory array (20) are being operated upon at the same address. A method of intelligently refreshing a dynamic memory is also enclosed.
Abstract:
A semiconductor integrated circuit includes a plurality of programmable elements such as antifuses (16, 17), each having a first terminal connected to a first power supply potential (CGND) and a second terminal. Each of a plurality of first semiconductor switching elements (18, 19) has a first terminal respectively connected to the second terminal of a corresponding one of the plurality of programmable elements and has a second terminal. Each of a plurality of second semiconductor switching elements (20) has a first terminal connected in common to selected ones of the second terminals of the plurality of first semiconductor switching elements and has a second terminal connected to a second power supply potential. A method of programming a plurality of programmable elements grouped in a plurality of subgroups each in a respective one of a plurality of groups includes the steps of applying a programming signal to the subgroups in a respective one of the groups and applying an address signal at like respective terminals in each of said groups, with each which terminal there is associated one of said programmable elements, either the programming signal or the address signal being applied by a common switching element.
Abstract:
A static memory cell is described which has cross coupled pulldown transistors and dual access transistors. The memory cell is fabricated such that balanced current paths are formed through the two pulldown transistors. A single word line is used to activate the access transistors which couple the memory cell to complementary bit lines. The memory cells, as viewed in a plan view, have the single word line and gates of the pulldown transistors fabricated in parallel.
Abstract:
A logic which enables implementation of an 80-bit wide or a 96-bit wide cache SRAM using the same memory array. The logic implementation is accomplished by merging tag, and data into an order block of information to maximize bus utilization. The logic reduces the bus cycles from four cycles for an 80-bit to three cycles for a 96-bit implementation.
Abstract:
A programming circuit for an anti-fuse utilizes a boot circuit that charges a capacitor to the supply voltage during a non-programming period. Anti-fuse is to be programmed, the plate of the capacitor to which the supply voltage has been applied is switched to 0 volt, thereby causing the other plate of the capacitor to output a negative voltage. This negative voltage is switched to one plate of an anti-fuse, and the other plate of the anti-fuse receives a positive voltage from an external source. A voltage is thereby applied across the anti-fuse that is greater than any voltage applied to any node of the integrated circuit.