Abstract: The invention relates generally to a fuse device of a semiconductor device, and more particularly, to an electrical fuse device of a semiconductor device. Embodiments of the invention provide a fuse device that is capable of reducing programming error caused by non-uniform current densities in a fuse link. In one respect, there is provided an electrical fuse device that includes: an anode; a fuse link coupled to the anode on a first side of the fuse link; a cathode coupled to the fuse link on a second side of the fuse link; a first cathode contact coupled to the cathode; and a first anode contact coupled to the anode, at least one of the first cathode contact and the first anode contact being disposed across a virtual extending surface of the fuse link.

Abstract: An eFuse structure having a first metal layer serving as a fuse with a gate including an undoped polysilicon (poly), a second metal layer and a high-K dielectric layer all formed on a silicon substrate with a Shallow Trench Isolation formation, and a process of fabricating same are provided. The eFuse structure enables use of low amounts of current to blow a fuse thus allowing the use of a smaller MOSFET.

Abstract: An integrated circuit includes a substrate. The substrate includes diffusion lines. The diffusion lines include impurities diffused into the substrate. A signal line layer includes first signal lines. A first metal layer includes second signal lines. The second signal lines include a first metallic material. A second metal layer includes third signal lines. The third signal lines include a second metallic material. First contacts connect the diffusion lines to (i) a first set of the second signal lines, or (ii) a first set of the third signal lines. Second contacts connect a first set of the first signal lines to a second set of the third signal lines. Each signal line in a first set of the second signal lines includes first portions and second portions. The first portions extend towards and are not connected to the second contacts. The first portions are not parallel to the second portions.

Abstract: A die micro electro-mechanical switch management system and method facilitate power conservation by selectively preventing electrical current from flowing in designated components. A present invention semiconductor die comprises a block of transistors for performing switching operations, a bus (e.g., a power bus, a signal bus, etc.) for conveying electrical current and a micro electro-mechanical switch that couples and decouples the block of transistors to and from the bus. The micro electro-mechanical switch is opened and closed depending upon operations (e.g., switching operations) being performed by the block of transistors. Electrical current is prevented from flowing to the block of transistors when the micro electro-mechanical switch is open and the block of transistors is electrically isolated. The micro electro-mechanical switch can interrupt electrical current flow in a plurality of the bus lines and/or can be included in a relay array.

Abstract: A L-level permutable switching network (L-PSN) having switches and multiple levels of conductors that are used to connect a first plurality of conductors to other multiple sets of conductors within respective interconnect resources constraints. The L-PSN can be applied in a wide range of applications, in tandem or hierarchically, to provide a large switch network used in network, routers, and programmable logic circuits. The L-PSN is used to connect a first set of conductors, through the L-PSN, to multiple sets of conductors in a given logic circuit hierarchy whereby the conductors in each of the multiple sets are equivalent or exchangeable, which in term, by construction, makes the first set of conductors equivalent when used in the next level of circuit hierarchy. The L-PSN is scalable for large sized sets of conductors and can be used in tandem or hierarchically to enable programmable interconnections among large sized circuits.

Abstract: An embodiment is a fuse structure. In accordance with an embodiment, a fuse structure comprises an anode, a cathode, a fuse link interposed between the anode and the cathode, and cathode connectors coupled to the cathode. The cathode connectors are each equivalent to or larger than about two times a minimum feature size of a contact that couples to an active device.

Abstract: A cell element field for data processing having function cells for execution of algebraic and/or logic functions and memory cells for receiving, storing and/or outputting information is described. A control connection may lead from the function cells to the memory cells.

Abstract: In order to improve the reliability of a semiconductor device having a fuse formed by a Damascene technique, a barrier insulating film and an inter-layer insulating film are deposited over a fourth-layer wiring and a fuse. The barrier insulating film is an insulating film for preventing the diffusion of Cu and composed of a SiCN film deposited by plasma CVD like the underlying barrier insulating film. The thickness of the barrier insulating film covering the fuse is larger than the thickness of the underlying barrier insulating film so as to improve the moisture resistance of the fuse.

Abstract: A memory cell is provided that includes a diode and a resistance-switching material layer coupled in series with the diode. The resistance-switching material layer: (a) includes a material from the family consisting of XvOw, wherein X represents an element from the family consisting of Hf and Zr, and wherein the subscripts v and w have non-zero values that form a stable compound, and (b) has a thickness between 20 and 65 angstroms. Other aspects are also provided.

Abstract: An apparatus comprising an integrated circuit, an interconnect layer within said integrated circuit, and one or more connections. The integrated circuit may be configured to provide an electrically measurable interconnect pattern by enabling one or more of a plurality of components. The one or more connections may each configured to enable a respective one of the components. The connections may be programmable while the apparatus is part of a wafer. The interconnect pattern may be configured to identify the apparatus after the apparatus has been manufactured.

Abstract: A semiconductor device comprises a semiconductor substrate, an anorganic isolation layer on the semiconductor substrate and a metallization layer on the anorganic isolation layer. The metallization layer comprises a fuse structure. At least in an area of the fuse structure the metallization layer and the anorganic isolation layer have a common interface.

Abstract: A semiconductor memory device according to one embodiment of the present invention includes a dielectric film configured to store information depending on presence or absence of a conductive path therein, and a plurality of electrodes provided to contact a first surface of the dielectric film. The conductive path can be formed between two electrodes arbitrarily selected form the plurality of electrodes. The conductive path has a rectifying property of allowing a current to flow more easily in a first direction connecting arbitrary two electrodes than in a second direction opposite to the first direction. The largest possible number of the conductive paths that may be formed is larger than the number of the plurality of electrodes.

Abstract: Various aspects of the technology includes a quad semiconductor power and/or switching FET comprising a pair of control/sync FET devices. Current may be distributed in parallel along source and drain fingers. Gate fingers and pads may be arranged in a serpentine configuration for applying gate signals to both ends of gate fingers. A single continuous ohmic metal finger includes both source and drain regions and functions as a source-drain node. A set of electrodes for distributing the current may be arrayed along the width of the source and/or drain fingers and oriented to cross the fingers along the length of the source and drain fingers. Current may be conducted from the electrodes to the source and drain fingers through vias disposed along the surface of the fingers. Heat developed in the source, drain, and gate fingers may be conducted through the vias to the electrodes and out of the device.

Abstract: A semiconductor chip capable of realizing reduction in cost when the semiconductor chip is mounted over a package substrate, miniaturization of the package substrate, and optimization of an interconnect pattern. The semiconductor chip includes a first electrode pad group provided in the semiconductor chip, and comprised of at least one electrode pad, and a second electrode pad group provided in the semiconductor chip, and comprised of at least one other electrode pad capable of outputting a signal identical to a signal outputted by the one electrode pad. Further, either the one electrode pad of the first electrode pad group, or the one other electrode pad of the second electrode pad group, closer in distance to one other electrode pad of one other semiconductor chip is coupled to the one other electrode pad of the one other semiconductor chip.

Abstract: In a semiconductor integrated circuit device, testing pads (209b) using a conductive layer, such as relocation wiring layers (205) are provided just above or in the neighborhood of terminals like bonding pads (202b) used only for probe inspection at which bump electrodes (208) are not provided. Similar testing pads may be provided even with respect to terminals like bonding pads provided with bump electrodes. A probe test is executed by using these testing pads or under the combined use of under bump metallurgies antecedent to the formation of the bump electrodes together with the testing pads. According to the above, bump electrodes for pads dedicated for probe testing may not be added owing to the use of the testing pads. Further, the use of testing pads provided in the neighborhood of the terminals like the bonding pads and smaller in size than the under bump metallurgies enables a probe test to be executed after a relocation wiring process.

Abstract: An arrangement of nonvolatile memory devices, having at least one memory device level stacked level by level above a semiconductor substrate, each memory level comprising an oxide layer substantially disposed above a semiconductor substrate, a plurality of word lines substantially disposed above the oxide layer; a plurality of bit lines substantially disposed above the oxide layer; a plurality of via plugs substantially in electrical contact with the word lines and, an anti-fuse dielectric material substantially disposed on side walls beside the bit lines and substantially in contact with the plurality of bit lines side wall anti-fuse dielectrics.

Abstract: A portion-to-be-melted of a fuse is surrounded by plates, so that heat to be generated in a meltdown portion of the fuse under current supply can be confined or accumulated in the vicinity of the meltdown portion of the fuse. This makes it possible to facilitate meltdown of the fuse. The meltdown portion of the fuse in a folded form, rather than in a single here a fuse composed of a straight-line form, is more successful in readily concentrating the heat generated in the fuse under current supply into the meltdown portion, and in further facilitating the meltdown of the fuse.

Abstract: An optically controlled read only memory is disclosed. The optically controlled read only memory includes a substrate, a plurality of memory cells having optical sensors disposed on the substrate, and at least one shielding structure disposed on the optical sensor, in which the shielding structure selectively shields a portion of the optical sensor according to a predetermined layout. Preferably, the optically controlled read only memory of the present invention is capable of providing two types or more program codes and outputting different program codes carrying different function under different lighting condition.

Abstract: A semiconductor device includes a field-effect transistor (FET), and a fuse which includes a graphene layer and is electrically connected to the FET.

Abstract: The disclosure relates generally to fuse structures, methods of forming and programming the same, and more particularly to fuse structures having crack stop voids. The fuse structure includes a semiconductor substrate having a dielectric layer thereon and a crack stop void. The dielectric layer includes at least one fuse therein and the crack stop void is adjacent to two opposite sides of the fuse, and extends lower than a bottom surface and above a top surface of the fuse. The disclosure also relates to a design structure of the aforementioned.

Abstract: A non-volatile semiconductor memory device having a memory cell in which operating potentials are few and the scale of the peripheral circuitry is reduced includes a select transistor having a source/drain on both sides of a channel of a semiconductor substrate and having a gate electrode disposed on the channel via a thick gate insulating film; an element isolation region formed on the semiconductor substrate in an area adjacent to the select transistor; an antifuse adjacent to the element isolation region, having a lower electrode formed on the semiconductor substrate and having an upper electrode disposed on the semiconductor substrate in an area between the element isolation region and lower electrode via a thin gate insulating film; and a connection contact electrically connecting the source and upper electrode and contacting the source and the upper electrode.

Abstract: A bistable resistance random access memory is described for enhancing the data retention in a resistance random access memory member. A dielectric member, e.g. the bottom dielectric member, underlies the resistance random access memory member which improves the SET/RESET window in the retention of information. The deposition of the bottom dielectric member is carried out by a plasma-enhanced chemical vapor deposition or by high-density-plasma chemical vapor deposition. One suitable material for constructing the bottom dielectric member is a silicon oxide. The bistable resistance random access memory includes a bottom dielectric member disposed between a resistance random access member and a bottom electrode or bottom contact plug. Additional layers including a bit line, a top contact plug, and a top electrode disposed over the top surface of the resistance random access memory member. Sides of the top electrode and the resistance random access memory member are substantially aligned with each other.

Abstract: A semiconductor memory device according to one embodiment of the present invention includes a dielectric film configured to store information depending on presence or absence of a conductive path therein, and a plurality of electrodes provided to contact a first surface of the dielectric film. The conductive path can be formed between two electrodes arbitrarily selected form the plurality of electrodes. The conductive path has a rectifying property of allowing a current to flow more easily in a first direction connecting arbitrary two electrodes than in a second direction opposite to the first direction. The largest possible number of the conductive paths that may be formed is larger than the number of the plurality of electrodes.

Abstract: A storage system and method for forming a storage system that uses punch-through diodes as a steering element in series with a reversible resistivity-switching element is described. The punch-through diode allows bipolar operation of a cross-point memory array. The punch-through diode may have a symmetrical non-linear current/voltage relationship. The punch-through diode has a high current at high bias for selected cells and a low leakage current at low bias for unselected cells. Therefore, it is compatible with bipolar switching in cross-point memory arrays having resistive switching elements. The punch-through diode may be a N+/P?/N+ device or a P+/N?/P+ device.

Abstract: A resistive random access memory array may be formed on the same substrate with a fuse array. The random access memory and the fuse array may use the same active material. For example, both the fuse array and the memory array may use a chalcogenide material as the active switching material. The main array may use a pattern of perpendicular sets of trench isolations and the fuse array may only use one set of parallel trench isolations. As a result, the fuse array may have a conductive line extending continuously between adjacent trench isolations. In some embodiments, this continuous line may reduce the resistance of the conductive path through the fuses.

Abstract: At least one MOS parameter of a MOS fuse is characterized to provide at least one MOS parameter reference value. Then, the MOS fuse is programmed by applying a programming signal to the fuse terminals so that programming current flows through the fuse link. The fuse resistance is measured to provide a measured fuse resistance associated with a first logic value. A MOS parameter of the programmed MOS fuse is measured to provide a measured MOS parameter value. The measured MOS parameter value is compared to the reference MOS parameter value to determine a second logic value of the MOS fuse, and a bit value is output based on the comparison.

Abstract: According to one embodiment, a resistance-change memory of embodiment includes a first interconnect line extending in a first direction, a second interconnect line extending in a second direction intersecting with the first direction, and a cell unit. The cell unit is provided at an intersection of the first interconnect line and the second interconnect line. The cell unit includes a non-ohmic element having a silicide layer on at least one of first and second ends thereof, and a memory element to store data in accordance with a reversible change in a resistance state. The silicide layer includes a 3d transition metal element which combines with an Si element to form silicide and which has a first atomic radius, and at least one kind of an additional element having a second atomic radius greater than the first atomic radius.

Abstract: In an embodiment of the invention, a method of fabricating a floating-gate PMOSFET (p-type metal-oxide semiconductor field-effect transistor) is disclosed. A silicide blocking layer (e.g. oxide, nitride) is used not only to block areas from being silicided but to also form an insulator on top of a poly-silicon gate. The insulator along with a top electrode (control gate) forms a capacitor on top of the poly-silicon gate. The poly-silicon gate also serves at the bottom electrode of the capacitor. The capacitor can then be used to capacitively couple charge to the poly-silicon gate. Because the poly-silicon gate is surrounded by insulating material, the charge coupled to the poly-silicon gate may be stored for a long period of time after a programming operation.

Abstract: An electrically reprogrammable fuse comprising an interconnect disposed in a dielectric material, a sensing wire disposed at a first end of the interconnect, a first programming wire disposed at a second end of the interconnect, and a second programming wire disposed at a second end of the interconnect, wherein the fuse is operative to form a surface void at the interface between the interconnect and the sensing wire when a first directional electron current is applied from the first programming wire through the interconnect to the second programming wire, and wherein, the fuse is further operative to heal the surface void between the interconnect and the sensing wire when a second directional electron current is applied from the second programming wire through the interconnect to the first programming wire.

Abstract: An object of one embodiment of the present invention is to provide an antifuse which has low writing voltage. The antifuse is used for a memory element for a read only memory device. The antifuse includes a first conductive layer, an insulating layer, a semiconductor layer, and a second conductive layer. The insulating layer included in the antifuse is a silicon oxynitride layer formed by adding ammonia to a source gas. When hydrogen is contained in the layer at greater than or equal to 1.2×1021 atoms/cm3 and less than or equal to 3.4×1021 atoms/cm3 or nitrogen is contained in the layer at greater than or equal to 3.2×1020 atoms/cm3 and less than or equal to 2.2×1021 atoms/cm3, writing can be performed at low voltage.

Abstract: A repair method for repairing an active device array substrate is provided. The active device array substrate includes a substrate, scan lines, data lines, active devices, pixel electrodes, and common lines. At least one of the scan line has an open defect. The scan lines and the data lines are intersected to define sub-pixel regions. The active devices are electrically connected with the scan lines and the data lines correspondingly. Each pixel electrode is disposed in one of the sub-pixel regions and electrically connected with one of the active devices. The repair method includes cutting one of the common lines neighboring to the open defect to form a cutting block that is electrically insulated from the common lines; and welding the cutting block, the scan line having the open defect and two active devices located at two opposite sides of the open defect.

Abstract: The electrical fuse includes a cathode pad, an anode pad and a fuse link connecting the cathode pad to the anode pad. The cathode pad includes a group of multiple electrical contacts and a solitary electrical contact disposed a predetermined distance from the group and near the fuse link, i.e., between the group of multiple electrical contacts and the fuse link. The cathode and anode pads as well as the fuse link include a polysilicon layer and a silicide layer.

Abstract: A semiconductor die includes at least one flexible interface block. The flexible interface block includes at least one interconnect, and at least one buffer coupled to the at least one interconnect. The flexible interface block further includes a routing interface coupled to circuitry integrated in the semiconductor die, and a controller coupled to provide communication between the routing interface and the at least one buffer.

Abstract: A dynamic and end-user configurable controlled impedance interconnect line includes a plurality of conductive pixels, a plurality of thin-film transition material interconnects to electrically connect adjacent conductive pixels in the plurality of conductive pixels, and a plurality of addressable pixel interconnect actuators to selectively heat a respective plurality of the thin-film transition material interconnects. The plurality of addressable pixel interconnect actuators is operable to selectively heat a respective plurality of the thin-film transition material interconnects to form an interconnect line.

Abstract: The present invention discloses a method of fabricating dual trench isolated epitaxial diode array. This method starts with the formation of heavily-doped first conductivity type regions and heavily-doped second conductivity type regions on the substrate, followed by epitaxial growth, then the formation of the isolations between diode array word lines by deep trench etch and the formation of the isolations between bit lines vertical to deep trenches by shallow trench etch, and finally the formation of separate diode array cells in the regions enclosed by deep and shallow trench isolations by ion implantation. This invention also provides a method of preventing the crosstalk current between adjacent word lines and bit lines of epitaxial diode arrays isolated by foregoing dual shallow trenches.

Abstract: A method forms an eFuse structure that has a pair of adjacent semiconducting fins projecting from the planar surface of a substrate (in a direction perpendicular to the planar surface). The fins have planar sidewalls (perpendicular to the planar surface of the substrate) and planar tops (parallel to the planar surface of the substrate). The tops are positioned at distal ends of the fins relative to the substrate. An insulating layer covers the tops and the sidewalls of the fins and covers an intervening substrate portion of the planar surface of the substrate located between the fins. A metal layer covers the insulating layer. A pair of conductive contacts are connected to the metal layer at locations where the metal layer is adjacent the top of the fins.

Abstract: Some embodiments include apparatus and methods having a memory cell with a first electrode, a second electrode, and a dielectric located between the first and second electrodes. The dielectric may be configured to allow the memory cell to form a conductive path in the dielectric from a portion of a material of the first electrode to represent a first value of information stored in the memory cell. The dielectric may also be configured to allow the memory cell to break the conductive path to represent a second value of information stored in the memory cell.

Abstract: According to one exemplary embodiment, a method for forming a one-time programmable metal fuse structure includes forming a metal fuse structure over a substrate, the metal fuse structure including a gate metal segment situated between a dielectric segment and a polysilicon segment, a gate metal fuse being formed in a portion of the gate metal segment. The method further includes doping the polysilicon segment so as to form first and second doped polysilicon portions separated by an undoped polysilicon portion where, in one embodiment, the gate metal fuse is substantially co-extensive with the undoped polysilicon portion. The method can further include forming a first silicide segment on the first doped polysilicon portion and a second silicide segment on the second doped polysilicon portion, where the first and second silicide segments form respective terminals of the one-time programmable metal fuse structure.

Abstract: A semiconductor device includes a first metal wiring which is formed over substructure; a first contact plug which is coupled to the first metal wiring and passes through a first interlayer insulating film provided over the substructure; a second metal wiring which is provided over the first interlayer insulating film and is coupled to the first contact plug; a second contact plug which is coupled to the second metal wiring and passes through a second interlayer insulating film which is provided over the first interlayer insulating film; and a fuse pattern and a data read fuse pattern which are coupled to the second contact plug and provided over the second interlayer insulating film.

Abstract: A device includes a package component, and a metal trace on a surface of the package component. A first and a second dielectric mask cover a top surface and sidewalls of the metal trace, wherein a landing portion of the metal trace is located between the first and the second dielectric masks. The landing portion includes a first portion having a first width, and a second portion connected to an end of the first portion. The second portion has a second width greater than the first width, wherein the first and the second widths are measured in a direction perpendicular to a lengthwise direction of the metal trace.

Abstract: Techniques for incorporating nanotechnology into electronic fuse (e-fuse) designs are provided. In one aspect, an e-fuse structure is provided. The e-fuse structure includes a first electrode; a dielectric layer on the first electrode having a plurality of nanochannels therein; an array of metal silicide nanopillars that fill the nanochannels in the dielectric layer, each nanopillar in the array serving as an e-fuse element; and a second electrode in contact with the array of metal silicide nanopillars opposite the first electrode. Methods for fabricating the e-fuse structure are also provided as are semiconductor devices incorporating the e-fuse structure.

Abstract: A nano-electric switch includes a cavity base, a confinement wall, and a cavity top defining a cavity. A floating conductive bridge movable within the cavity completes an electrical circuit between a first electrical contact and a second electrical contact in a first selectable position, and breaks the electrical circuit in a second selectable position.

Abstract: A semiconductor device includes: an e-fuse gate, a floating pattern between the e-fuse gate and an e-fuse active portion, a blocking dielectric pattern between the floating pattern and the e-fuse gate, and an e-fuse dielectric layer between the floating pattern and the e-fuse active portion. The floating pattern includes a first portion between the e-fuse gate and the e-fuse active portion and a pair of second portions extended upward along both sidewalls of the e-fuse gate from both edges of the first portion.

Abstract: A circuit includes a fuse and a sensing and control circuit. The fuse is coupled between a MOS transistor and a current source node. The sensing and control circuit is configured to receive a programming pulse and output a modified programming signal to the gate of the MOS transistor for programming the fuse. The modified programming signal has a pulse width based on a magnitude of a current through the first fuse.

Abstract: Systems and methods are disclosed for forming a custom integrated circuit (IC) with a first fixed (non-programmable) region on a wafer with non-customizable mask layers, wherein the first fixed region includes multiplicities of transistors and a first interconnect layer and a second interconnect layer above the first interconnect layer which form base cells; and a programmable region above the first fixed region with customizable mask layers, wherein at least one mask layer in the programmable region is coupled to the second interconnect layer which provides electrical access to all transistor nodes of the base cells and wherein the programmable region comprises a third interconnect layer coupled to the customizable mask layers to customize the IC. A second fixed region may be formed above the programmable region to provide multiple fixed regions and reduce the number of required masks in customizing the custom IC.

Abstract: Programmable nanotube interconnect is disclosed. In one embodiment, a method includes forming a interconnect layer using a plurality of nanotube structures, and automatically altering a route of an integrated circuit based on an electrical current applied to at least one of the plurality of nanotube structures in the interconnect layer. Neighboring interconnect layers separated by planar vias may include communication lines that are perpendicularly oriented with respect to each of the neighboring interconnect layers. The nanotube structure may be chosen from a group comprising a polymer, carbon, and a composite material. A carbon nanotube film may be patterned in a metal layer to form the plurality of nanotube structures. A sputtered planar process may be performed across a trench of electrodes to create the carbon nanotube structures.

Abstract: A fuse device has a fuse element provided with a first terminal and a second terminal and an electrically breakable region, which is arranged between the first terminal and the second terminal and is configured to undergo breaking as a result of the supply of a programming electrical quantity, thus electrically separating the first terminal from the second terminal. The electrically breakable region is of a phase-change material, in particular a chalcogenic material, for example GST.

Abstract: According to one exemplary embodiment, a programmable poly fuse includes a P type resistive poly segment forming a P-N junction with an adjacent N type resistive poly segment. The programmable poly fuse further includes a P side silicided poly line contiguous with the P type resistive poly segment and coupled to a P side terminal of the poly fuse. The programmable poly fuse further includes an N side silicided poly line contiguous with the N type resistive poly segment and coupled to an N side terminal of the poly fuse. During a normal operating mode, a voltage less than or equal to approximately 2.5 volts is applied to the N side terminal of the programmable poly fuse. A voltage higher than approximately 3.5 volts is required at the N side terminal of the poly fuse to break down the P-N junction.

Abstract: An embodiment of this invention uses a massive parallel interconnect fabric (MPIF) at the flipped interface of a core die substrate (having the core logic blocks) and a context die (used for in circuit programming/context/customization of the core die substrate), to produce ASIC-like density and FPGA-like flexibility/programmability, while reducing the time and cost for development and going from prototyping to production, reducing cost per die, reducing or eliminating NRE, and increasing performance. Other embodiments of this invention enable debugging complex SoC through large contact points provided through the MPIF, provide for multi-platform functionality, and enable incorporating FGPA core in ASIC platform through the MPIF.

Abstract: The semiconductor device includes a fuse structure disposed on a substrate. An interlayer dielectric disposed on the fuse structure. A first contact plug, a second contact plug, and a third contact plug penetrate the interlayer dielectric and wherein each of the first contact plug, the second contact plug and the third contact plug are connected to the fuse structure. A first conductive pattern and a second conductive pattern are disposed on the interlayer dielectric. The first conductive pattern and the second conductive pattern are electrically connected to the first contact plug and second contact plug, respectively.