Abstract: Methods of forming and operating phase change memory devices include adjusting an activation energy barrier between a metastable phase and a stable phase of a phase change material in a memory cell. In some embodiments, the activation energy barrier is adjusted by applying stress to the phase change material in the memory cell. Memory devices include a phase change memory cell and a material, structure, or device for applying stress to the phase change material in the memory cell. In some embodiments, a piezoelectric device may be used to apply stress to the phase change material. In additional embodiments, a material having a thermal expansion coefficient greater than that of the phase change material may be positioned to apply stress to the phase change material.

Abstract: Methods of forming nonvolatile memory devices include forming a stack of layers of different materials on a substrate. This stack includes a plurality of first layers of a first material and a plurality of second layers of a second material arranged in an alternating sequence of first and second layers. A selected first portion of the stack of layers is isotropically etched for a sufficient duration to define a first trench therein that exposes sidewalls of the alternating sequence of first and second layers. The sidewalls of each of the plurality of first layers are selectively etched relative to sidewalls of adjacent ones of the plurality of second layers. Another etching step is then performed to recess sidewalls of the plurality of second layers and thereby expose portions of upper surfaces of the plurality of first layers. These exposed portions of the upper surfaces of the plurality of first layers, which may act as word lines of a memory device, are displaced laterally relative to each other.

Abstract: Methods of forming nonvolatile memory devices according to embodiments of the invention include techniques to form highly integrated vertical stacks of nonvolatile memory cells. These vertical stacks of memory cells can utilize dummy memory cells to compensate for process artifacts that would otherwise yield relatively poor functioning memory cell strings when relatively large numbers of memory cells are stacked vertically on a semiconductor substrate using a plurality of vertical sub-strings electrically connected in series.

Abstract: A nonvolatile semiconductor memory includes a memory cell transistor including a first floating gate electrode layer formed on a first tunneling insulating film, a first inter-gate insulating film, a first and a second control gate electrode layer, and a first metallic silicide film; a high voltage transistor including a high voltage gate electrode layer formed on the high voltage gate insulating film, a second inter-gate insulating film having an aperture, a third and a fourth control gate electrode layer, and a second metallic silicide film; a low voltage transistor including a second floating gate electrode layer formed on the second tunneling insulating film, a third inter-gate insulating film having an aperture, a fifth and a sixth control gate electrode layer, and a third metallic silicide film; and a liner insulating film directly disposed on a first source and drain region of the memory cell transistor, a second source and drain region of the low voltage transistor, and a third source and drain region

Abstract: A method of making a phase change random access memory (PCM) device comprises forming a PCM stack that includes a heater layer, phase change material layer, and a top electrode layer. A top protection layer is formed overlying the PCM stack. The top protection layer and a first portion of the PCM stack are then patterned, wherein the first portion of the PCM stack excludes the heater layer. A sidewall protection feature is formed along a sidewall of the patterned top protection layer and first portion of the PCM stack. The heater layer is etched using (i) the sidewall protection feature and (ii) the patterned top protection layer and first portion of the PCM stack collectively as a mask to form a self-aligned heater layer bottom electrode of the PCRAM stack, thereby completing a memory bit of the PCRAM device.

Abstract: Disclosed herein is an improved memory device, and related methods of manufacturing, wherein the area occupied by a conventional landing pad is significantly reduced to around 50% to 10% of the area occupied by conventional landing pads. This is accomplished by removing the landing pad from the cell structure, and instead forming a conductive via structure that provides the electrical connection from the memory stack or device in the structure to an under-metal layer. By forming only this via structure, rather than separate vias formed on either side of a landing pad, the overall width occupied by the connective via structure from the memory stack to an under-metal layer is substantially reduced, and thus the via structure and under-metal layer may be formed closer to the memory stack (or conductors associated with the stack) so as to reduce the overall width of the cell structure.

Abstract: A method of fabricating a nonvolatile memory device with a three-dimensional structure includes alternately stacking first and second material layers in two or more layers on a semiconductor substrate, forming trenches penetrating the stacked first and second material layers by performing a first etching process, and removing the second material layers exposed in the trenches by performing a second etching process. The first and second material layers are formed of materials that have the same main component but have different impurity contents, respectively.

Abstract: A three-dimensional array of memory elements is formed across multiple layers of planes positioned at different distances above a semiconductor substrate. The memory elements reversibly change a level of electrical conductance in response to a voltage difference being applied across them. The three-dimensional array includes a two-dimensional array of pillar lines from the substrate through the multiple layers of planes which together with arrays of word lines on each plane are used to access the memory elements. The memory elements of the multiple layers are formed simultaneously in an orientation parallel to the substrate thereby reducing processing cost. In another aspect, a diode is formed in series with each memory element to reduce current leakage. The diode is incorporated within a pillar line acting as a bit line without taking up additional space.

Abstract: Disclosed herein is a semiconductor memory device for reducing a junction resistance and increasing amount of current throughout the unit cell. A semiconductor memory device comprises plural unit cells, each coupled to contacts formed in different shape at both sides of a word line in a cell array.

Abstract: Angled ion implants are utilized to form doped regions in a semiconductor pillar formed in an opening of a mask. The pillar is formed to a height less than the height of the mask. Angled ion implantation can be used to form regions of a semiconductor device such as a body tie region, a halo region, or current terminal extension region of a semiconductor device implemented with the semiconductor pillar.

Abstract: A phase change memory device includes heaters which are formed in their respective memory cells and vertically positioned stack patterns having phase change layers and top electrodes which are formed to come into contact with the heaters. The heaters have horizontal cross-sectional bent shapes which can have any number of shapes such as a shape similar to that of a boomerang. The horizontal cross-sectional bent shapes of the to heaters are for minimizing the contact area between the heaters and the phase change layer so that programming currents can be reduced or minimized.

Abstract: A phase change memory may be made with improved speed and stable characteristics over extended cycling. The alloy may be selected by looking at alloys that become stuck in either the set or the reset state and finding a median or intermediate composition that achieves better cycling performance. Such alloys may also experience faster programming and may have set and reset programming speeds that are substantially similar.

Abstract: A memory device includes a multi gate field effect transistor (MuGFET) having a fin with a contact area. A programmable memory element abuts the fin contact area.

Abstract: One aspect of this disclosure relates to a method for forming a wafer with a strained semiconductor. In various embodiments of the method, a predetermined contour is formed in one of a semiconductor membrane and a substrate wafer. The semiconductor membrane is bonded to the substrate wafer and the predetermined contour is straightened to induce a predetermined strain in the semiconductor membrane. In various embodiments, a substrate wafer is flexed into a flexed position, a portion of the substrate wafer is bonded to a semiconductor layer when the substrate wafer is in the flexed position, and the substrate wafer is relaxed to induce a predetermined strain in the semiconductor layer. Other aspects and embodiments are provided herein.

Abstract: A composite resistance variable element includes a first resistance variable element in which a resistance value varies corresponding to a direction of inner magnetization, and a second resistance variable element connected in series to the first resistance variable element. A resistance value of the second resistance variable element varies corresponding to a magnitude of at least one of a voltage applied to the second resistance variable element and a current flowing through the second resistance variable element, irrespective of whether the voltage and the current are positive or negative.

Abstract: Methods and devices associated with phase change cell structures are described herein. In one or more embodiments, a method of forming a phase change cell structure includes forming a substrate protrusion that includes a bottom electrode, forming a phase change material on the substrate protrusion, forming a conductive material on the phase change material, and removing a portion of the conductive material and a portion of the phase change material to form an encapsulated stack structure.

Abstract: In a power feeding region of a memory cell (MC) in which a sidewall-shaped memory gate electrode (MG) of a memory nMIS (Qnm) is provided by self alignment on a side surface of a selection gate electrode (CG) of a selection nMIS (Qnc) via an insulating film, a plug (PM) which supplies a voltage to the memory gate electrode (MG) is embedded in a contact hole (CM) formed in an interlayer insulating film (9) formed on the memory gate electrode (MG) and is electrically connected to the memory gate electrode (MG). Since a cap insulating film (CAP) is formed on an upper surface of the selection gate electrode (CG), the electrical conduction between the plug (PM) and the selection gate electrode (CG) can be prevented.

Abstract: Method of forming a semiconductor device which includes the steps of obtaining a semiconductor substrate having a logic region and an STI region; sequentially depositing layers of high K material, metal gate, first silicon and hardmask; removing the hardmask and first silicon layers from the logic region; applying a second layer of silicon on the semiconductor substrate such that the logic region has layers of high K material, metal gate and second silicon and the STI region has layers of high K material, metal gate, first silicon, hardmask and second silicon. There may also be a second hardmask layer between the metal gate layer and the first silicon layer in the STI region. There may also be a hardmask layer between the metal gate layer and the first silicon layer in the STI region but no hardmask layer between the first and second layers of silicon in the STI region.

Abstract: The present invention provides an MIM capacitor using a high-k dielectric film preventing degradation of breakdown field strength of the MIM capacitor and suppressing the increase of the leakage current. The MIM capacitor comprises a first metal interconnect, a fabricated capacitance film, a fabricated upper electrode, and a third metal interconnect. The MIM capacitor is realized by forming an interlayer dielectric film comprising silicon oxide so as to cover the first metal interconnect, then forming a first opening in the interlayer dielectric film to a region corresponding to a via hole layer in the interlayer dielectric film just above the first metal interconnect so as not to expose the upper surface of the first metal interconnect, then forming a second opening to the inside of the first opening so as to expose the surface of the first metal interconnect and then forming a capacitance film and a third metal interconnect.

Abstract: A method for forming a semiconductor device includes forming a dielectric layer over a substrate. The method further includes forming a select gate layer over the dielectric layer. The method further includes etching the select gate layer at a first etch rate to form a first portion of a sidewall of a select gate, wherein the step of etching the select gate layer at the first etch rate includes using an oxidizing agent to oxidize at least a top portion of the substrate underlying the dielectric layer to form an oxide layer. The method further includes etching the select gate layer at a second etch rate lower than the first etch rate to form a second portion of the sidewall of the select gate, wherein the step of etching the select gate layer at the second etch rate includes removing only a top portion of the dielectric layer.

Abstract: A nonvolatile semiconductor memory device comprises a memory string, and a wiring. The memory string comprises a semiconductor layer, a charge storage layer, and a plurality of first conductive layers. The plurality of first conductive layers comprises a stepped portion formed in a stepped shape such that positions of ends of the plurality of first conductive layers differ from one another. The wiring comprises a plurality of second conductive layers extending upwardly from an upper surface of the first conductive layers comprising the stepped portion. The plurality of second conductive layers are formed such that upper ends thereof are aligned with a surface parallel to the substrate, and such that a diameter thereof decreases from the upper end thereof to a lower end thereof. The plurality of second conductive layers are formed such that the greater a length thereof in the perpendicular direction, the larger a diameter of the upper end thereof.

Abstract: Provided are a memory device formed using one or more source materials not containing hydrogen as a constituent element and a method of manufacturing the memory device.

Abstract: A method for fabricating a phase change memory device including memory cells includes patterning a via to a contact surface of a substrate corresponding to each of an array of conductive contacts to be connected to access circuitry, lining each via with a conformal conductive seed layer to the contact surface, forming a dielectric layer covering the conductive seed layer, and etching a center region of each via to the contact surface to expose the conformal conductive seed layer at the contact surface. The method further includes electroplating phase change material on exposed portions of the conformal conductive seed layer, recessing the phase change material within the center region forming a conductive material that remains conductive upon oxidation, on the recessed phase change material, oxidizing edges of the conformal conductive seed layer formed along sides of each via, and forming a top electrode over each memory cell.

Abstract: An electronic device includes a substrate having isolation features defining active regions coextending over a surface of the substrate. The device also includes coextending line patterns crossing over the active regions, including string and ground selection lines and word lines between the string and ground selection lines. The device further includes first implant regions of a first conductivity type in the active regions between the word lines and having a first carrier concentration. The device further includes second implant regions of the first conductivity type in the active regions between edge ones of the word lines and an adjacent one of the string selection line and the ground selection line. In the device, the second implant region includes a low doping portion abutting the edge word lines and a high doping portion spaced from the edge word line by the low doping portion and having a second carrier concentration greater than the first carrier concentration.

Abstract: A semiconductor device, comprising: a vertical pillar transistor (VPT) formed on a silicon-on-insulator (SOI) substrate, the VPT including a body that has a lower portion and an upper portion, a source/drain node disposed at an upper end portion of the upper portion of the body and a drain/source node disposed at the lower portion of the body; a buried bit line (BBL) formed continuously on sidewalls and an upper surface of the lower portion, the BBL includes metal sificide; and a word line that partially enclosing the upper portion of the body of the VPT, wherein the BBL extends along a first direction and the word line extends in a second direction substantially perpendicular to the first direction. An offset region is disposed immediately beneath the word line.

Abstract: A semiconductor device includes a plurality of nonvolatile memory cells (1). Each of the nonvolatile memory cells comprises a MOS type first transistor section (3) used for information storage, and a MOS type second transistor section (4) which selects the first transistor section. The second transistor section has a bit line electrode (16) connected to a bit line, and a control gate electrode (18) connected to a control gate control line. The first transistor section has a source line electrode (10) connected to a source line, a memory gate electrode (14) connected to a memory gate control line, and a charge storage region (11) disposed directly below the memory gate electrode. A gate withstand voltage of the second transistor section is lower than that of the first transistor section.

Abstract: A semiconductor device includes a shallow isolation trench (STI) structure on a silicon substrate for isolating element-forming regions from one another. The surface region of the silicon substrate in the element-forming regions, as viewed in the extending direction of the gate electrode lines, once falls and thereafter rises monotonically from the periphery toward the center of the element-forming regions.

Abstract: Semiconductor devices and methods for fabricating the same are provided. An exemplary embodiment of a semiconductor device comprises a substrate with a plurality of isolation structures formed therein, defining first and second areas over the substrate. A transistor is formed on a portion of the substrate in the first and second areas, respectively, wherein the transistor in the second area is formed with merely a pocket doping region in the substrate adjacent to a drain region thereof. A first dielectric layer is formed over the substrate, covering the transistor formed in the first and second areas. A plurality of first contact plugs is formed through the first dielectric layer, electrically connecting a source region and a drain region of the transistor in the second area, respectively. A second dielectric layer is formed over the first dielectric layer with a capacitor formed therein, wherein the capacitor electrically connects one of the first contact plugs.

Abstract: A method of programming a carbon nanotube memory cell is provided, wherein the memory cell comprises a first conductor, a steering element, a carbon nanotube fabric, and a second conductor, wherein the steering element and the carbon nanotube fabric are arranged electrically in series between the first conductor and the second conductor, and wherein the entire carbon nanotube memory cell is formed above a substrate, the carbon nanotube fabric having a first resistivity, the method including applying a first electrical set pulse between the first conductor and the second conductor, wherein, after application of the first electrical set pulse, the carbon nanotube fabric has a second resistivity, the second resistivity less than the first resistivity. Other aspects are also provided.

Abstract: A technique capable of manufacturing a semiconductor device without posing contamination in a manufacturing apparatus regarding a phase change memory including a memory cell array formed of memory cells using a storage element (RE) by a variable resistor and a select transistor (CT). A buffer cell is arranged between a sense amplifier (SA) and a memory cell array (MCA) and between a word driver (WDB) and the memory cell array. The buffer cell is formed of the resistive storage element (RE) and the select transistor (CT) same as those of the memory cell. The resistive storage element in the memory cell is connected to a bit-line via a contact formed above the resistive storage element. Meanwhile, in the buffer cell, the contact is not formed above the resistive storage element, and a state of being covered with an insulator is kept upon processing the contact in the memory cell. By such a processing method, exposure and sublimation of a chalcogenide film used in the resistive storage element can be avoided.

Abstract: An improved memory design is described which removes the need to read firmware from ROM into RAM on start-up. A SRAM memory element comprises an influencing element which sets the state of the memory cells within the memory element on start-up to defined values. These defined values are set at the design stage such that on start-up the volatile memory contains firmware or other data. Dependent upon the implementation of the influencing element, the values of stored in the memory cells may be fixed or may subsequently be overwritten during operation of the device. In an example, the memory cell comprises two cross-coupled inverters and the influencing element comprises at least one transistor arranged to connect the input to one of the inverters to ground or a power supply rail when voltage is applied to a controlling node of the transistor.

Abstract: Non-volatile resistive-switching memories formed using anodization are described. A method for forming a resistive-switching memory element using anodization includes forming a metal containing layer, anodizing the metal containing layer at least partially to form a resistive switching metal oxide, and forming a first electrode over the resistive switching metal oxide. In some examples, an unanodized portion of the metal containing layer may be a second electrode of the memory element.

Abstract: A method for manufacturing a flash memory device is capable of controlling a phenomenon in which a length of the channel between a source and a drain is decreased due to undercut. The method includes forming a gate electrode comprising a floating gate, an ONO film and a control gate using a hard mask pattern over a semiconductor substrate, forming a spacer over the sidewall of the gate electrode, forming an low temperature oxide (LTO) film over the entire surface of the semiconductor substrate including the gate electrode and the spacer, etching the LTO film such that a top portion of the source/drain region and a top portion of the gate electrode are exposed, and removing the LTO film present over the sidewall of the gate electrode by wet-etching.

Abstract: A semiconductor device includes a semiconductor substrate, a memory cell region provided on the semiconductor substrate, a word line provided on the memory cell region, a first gate insulating film provided in the memory cell region beneath the word line, a first floating gate electrode provided on the first gate insulating film, a second gate insulating film provided in the memory cell region beneath the word line, the second gate insulating film being different from the first gate insulating film in thickness, and a second floating gate electrode provided on the second gate insulating film.

Abstract: A nonvolatile memory integrated circuit has a semiconductor substrate and a nonvolatile memory device on the semiconductor substrate. The device has a transistor and a capacitor on the semiconductor substrate, and a shared floating gate connecting the gate regions of the transistor and the capacitor. The transistor has at least a doping region defining the source and drain regions, as well as three other doping regions overlapping the source and drain regions. Also disclosed are a nonvolatile memory circuit with multiple such nonvolatile memory device, and methods for making the nonvolatile memory circuit with one or more such nonvolatile memory devices.

Abstract: A semiconductor device has a bit line interconnection with a greater width and a reduced level on a bit line contact is provided, as are methods of fabricating such devices. These method includes forming a buried gate electrode to intersect an active region of a substrate. Source and drain regions are formed in the active region. A first conductive pattern is formed on the substrate. The first conductive pattern has a first conductive layer hole configured to expose the drain region. A second conductive pattern is formed in the first conductive layer hole to contact the drain region. A top surface of the second conductive pattern is at a lower level than a top surface of the first conductive pattern. A third conductive layer and a bit line capping layer are formed on the first conductive pattern and the second conductive pattern and patterned to form a third conductive pattern and a bit line capping pattern.

Abstract: Embodiments of the invention relate generally to semiconductors and memory technology, and more particularly, to systems, integrated circuits, and methods to implement memory architectures configured to enhance throughput for cross point arrays including memory elements, such as memory elements based on third dimensional memory technology. In at least some embodiments, an integrated circuit includes arrays that include memory elements being formed BEOL above a FEOL logic layer within a boundary in a plane parallel to a substrate, and array lines. Further, the integrated circuit includes array line decoders disposed in the logic layer within a region located coextensive with the boundary and between the substrate and the arrays. In some embodiments, the disposition of peripheral circuitry, such as the array line decoders, under the arrays can preserve or optimize die efficiency for throughput enhancement.

Abstract: An area where a lower electrode is in contact with a variable resistance material needs to be reduced to lower the power consumption of a variable resistance memory device. The present invention is to provide a method of producing a variable resistance memory element whereby the lower electrode can be formed smaller. Combining an anisotropic etching process with an isotropic etching process enables the lower electrode to be formed smaller.

Abstract: Embodiments of the present invention provide a method that includes providing a wafer including multiple cells, each cell including at least one emitter, and performing a lithographic operation on the wafer. The lithographic operation comprises forming heater trenches adjacent the emitters, each heater trench having a width that extends over at least respective portions of two cells. Other embodiments are also described.

Abstract: According to one embodiment, a nonvolatile memory device includes a substrate, first electrodes, a first and a second interelectrode insulating layer, second electrodes, a memory portion and a first protrusion. The first electrodes are provided on the substrate and extend in a first direction. The first interelectrode insulating layer is provided between the first electrodes. The second electrodes are opposed to the first electrodes and extend in a second direction crossing the first direction. The second interelectrode insulating layer is provided between the second electrodes. The memory portion is provided between the first electrode and the second electrode. The first protrusion is conductive and provided at least one of between the first electrode and the memory portion and between the first interelectrode insulating layer and the memory portion, and between the second electrode and the memory portion and between the second interelectrode insulating layer and the memory portion.

Abstract: A method of fabricating a non-volatile memory device having a charge trapping layer includes forming a tunneling layer, a charge trapping layer, a blocking layer and a control gate electrode layer over a substrate, forming a mask layer pattern on the control gate electrode layer, performing an etching process using the mask layer pattern as an etching mask to remove an exposed portion of the control gate electrode layer, wherein the etching process is performed as excessive etching to remove the charge trapping layer by a specified thickness, forming an insulating layer for blocking charges from moving on the control gate electrode layer and the mask layer pattern, performing anisotropic etching on the insulating layer to form an insulating layer pattern on a sidewall of the control gate electrode layer and a partial upper sidewall of the blocking layer, and performing an etching process on the blocking layer exposed by the anisotropic etching, wherein the etching process is performed as excessive etching to

Abstract: A method for manufacturing a nonvolatile storage element that minimizes shape shift between an upper electrode and a lower electrode, and which includes: depositing, in sequence, a connecting electrode layer which is conductive, a lower electrode layer and a variable resistance layer which are made of a non-noble metal nitride and are conductive, an upper electrode layer made of noble metal, and a mask layer; forming the mask layer into a predetermined shape; forming the upper electrode layer, the variable resistance layer, and the lower electrode layer into the predetermined shape by etching using the mask layer as a mask; and removing, simultaneously, the mask and a region of the connecting electrode layer that has been exposed by the etching.

Abstract: A semiconductor storage device includes a semiconductor substrate, a source region, a source line, and a bit line. The source region is formed in an element region formed on the semiconductor substrate. The source line is formed to overlap with the source region in planar view. The bit line is formed on a layer higher than the source line.

Abstract: A method of forming a device is disclosed. The method includes providing a substrate having an active area. A gate is formed on the substrate. First and second current paths through the gate are formed. The first current path serves a first purpose and the second current path serves a second purpose. The gate controls selection of the current paths.

Abstract: A method of forming a memory array includes forming first and second strings of serially-coupled memory cells respectively on first and second sides of a conductive pillar. Forming the first string of memory cells includes forming a first control gate on the first side of the conductive pillar and interposing a first charge trap between the first side of the conductive pillar and the first control gate. Forming the second string of memory cells comprises forming a second control gate on the second side of the conductive pillar and interposing a second charge trap between the second side of the conductive pillar and the second control gate. The first and second charge traps are electrically isolated from each other, and the first and second control gates are electrically isolated from each other.

Abstract: A memory array including a plurality of memory cells, a plurality of word lines, a dummy word line, at least a first conductive region and at least a first plug is provided. Each word line is coupled to corresponding memory cells. A dummy word line is directly adjacent to an outmost word line of the plurality of word lines. The first conductive region is disposed only between the dummy word line and the outmost word line. The first plug is located between the dummy word line and the outmost word line.

Abstract: In a semiconductor device comprising a capacitive element, an area of the capacitive element is reduced without impairing performance, and further, without addition of an extra step in a manufacturing process.

Abstract: A method is provided for forming a metal/semiconductor/metal (MSM) back-to-back Schottky diode from a silicon (Si) semiconductor. The method deposits a Si semiconductor layer between a bottom electrode and a top electrode, and forms a MSM diode having a threshold voltage, breakdown voltage, and on/off current ratio. The method is able to modify the threshold voltage, breakdown voltage, and on/off current ratio of the MSM diode in response to controlling the Si semiconductor layer thickness. Generally, both the threshold and breakdown voltage are increased in response to increasing the Si thickness. With respect to the on/off current ratio, there is an optimal thickness. The method is able to form an amorphous Si (a-Si) and polycrystalline Si (polySi) semiconductor layer using either chemical vapor deposition (CVD) or DC sputtering. The Si semiconductor can be doped with a Group V donor material, which decreases the threshold voltage and increases the breakdown voltage.

Abstract: A method for fabricating semiconductor memory device includes providing a first semiconductor substrate, and forming a first storage device on the first semiconductor substrate. The method includes forming a switching device on the first storage device, and forming a second storage devices on the switching device. Logic devices are formed below the first storage devices.

Abstract: A resistance switching memory is introduced herein. The resistance switching memory includes a highly-insulating or resistance-switching material formed to cover the sidewall of a patterned metal line, and extended alongside a dielectric layer sidewall to further contact a portion of the top surface of the lower electrode. The other part of the top surface of the lower electrode is covered by an insulating layer between the top electrode and the lower electrode. An oxygen gettering metal layer in the lower electrode occupies a substantial central part of the top surface of the lower electrode and is partially covered by the highly-insulating or resistance-switching material. A switching area is naturally very well confined to the substantial central part of the oxygen gettering metal layer of the lower electrode.