Abstract: Techniques are disclosed for providing a low resistance self-aligned contacts to devices formed in a semiconductor heterostructure. The techniques can be used, for example, for forming contacts to the gate, source and drain regions of a quantum well transistor fabricated in III-V and SiGe/Ge material systems. Unlike conventional contact process flows which result in a relatively large space between the source/drain contacts to gate, the resulting source and drain contacts provided by the techniques described herein are self-aligned, in that each contact is aligned to the gate electrode and isolated therefrom via spacer material.

Abstract: Techniques are disclosed for forming a non-planar germanium quantum well structure. In particular, the quantum well structure can be implemented with group IV or III-V semiconductor materials and includes a germanium fin structure. In one example case, a non-planar quantum well device is provided, which includes a quantum well structure having a substrate (e.g. SiGe or GaAs buffer on silicon), a IV or III-V material barrier layer (e.g., SiGe or GaAs or AlGaAs), a doping layer (e.g., delta/modulation doped), and an undoped germanium quantum well layer. An undoped germanium fin structure is formed in the quantum well structure, and a top barrier layer deposited over the fin structure. A gate metal can be deposited across the fin structure. Drain/source regions can be formed at respective ends of the fin structure.

Abstract: Microelectronic structures embodying the present invention include a field effect transistor (FET) having highly conductive source/drain extensions. Formation of such highly conductive source/drain extensions includes forming a passivated recess which is back filled by epitaxial deposition of doped material to form the source/drain junctions. The recesses include a laterally extending region that underlies a portion of the gate structure. Such a lateral extension may underlie a sidewall spacer adjacent to the vertical sidewalls of the gate electrode, or may extend further into the channel portion of a FET such that the lateral recess underlies the gate electrode portion of the gate structure. In one embodiment the recess is back filled by an in-situ epitaxial deposition of a bilayer of oppositely doped material. In this way, a very abrupt junction is achieved that provides a relatively low resistance source/drain extension and further provides good off-state subthreshold leakage characteristics.

Abstract: Techniques are disclosed for forming a non-planar quantum well structure. In particular, the quantum well structure can be implemented with group IV or III-V semiconductor materials and includes a fin structure. In one example case, a non-planar quantum well device is provided, which includes a quantum well structure having a substrate (e.g. SiGe or GaAs buffer on silicon), a IV or III-V material barrier layer (e.g., SiGe or GaAs or AlGaAs), and a quantum well layer. A fin structure is formed in the quantum well structure, and an interfacial layer provided over the fin structure. A gate metal can be deposited across the fin structure. Drain/source regions can be formed at respective ends of the fin structure.

Abstract: Thin film 1S1R bitcells incorporating a barrier between selector and memory elements. Devices incorporating such bitcells and methods of forming such bitcells are also described. In embodiments, the selector and memory element is each a dielectric material, and advantageously a metal oxide. Between the selector and memory elements is a barrier, which is to reduce intermixing and/or reaction of selector material and memory material. Addition of a barrier layer having suitable material properties into the 1S1R stack may extend the operating lifetime of a bitcell incorporated the stack by resisting intermixing and/or reaction of the selector and memory thin film materials driven by thermal and/or electric field stresses experienced by a bitcell during operation. In embodiments, a barrier layer may include one or more material layers having a composition distinct from the material composition(s) of the selector and memory elements.

Abstract: Techniques are disclosed for forming a spin-transfer torque memory (STTM) element having an annular contact to reduce critical current requirements. The techniques reduce critical current requirements for a given magnetic tunnel junction (MTJ), because the annular contact reduces contact size and increases local current density, thereby reducing the current needed to switch the direction of the free magnetic layer of the MTJ. In some cases, the annular contact surrounds at least a portion of an insulator layer that prevents the passage of current. In such cases, current flows through the annular contact and around the insulator layer to increase the local current density before flowing through the free magnetic layer. The insulator layer may comprise a dielectric material, and in some cases, is a tunnel material, such as magnesium oxide (MgO). In some cases, a critical current reduction of at least 10% is achieved for a given MTJ.

Abstract: The present description relates to a gallium nitride transistor which includes at least one source/drain structure having low contact resistance between a 2D electron gas of the gallium nitride transistor and the source/drain structure. The low contact resistance may be a result of at least a portion of the source/drain structure being a single-crystal structure abutting the 2D electron gas. In one embodiment, the single-crystal structure is grown with a portion of a charge inducing layer of the gallium nitride transistor acting as a nucleation site.

Abstract: A material layer stack for a magnetic tunneling junction, the material layer stack including a fixed magnetic layer; a dielectric layer; a free magnetic layer; and an amorphous electrically-conductive seed layer, wherein the fixed magnetic layer is disposed between the dielectric layer and the seed layer. A non-volatile memory device including a material stack including an amorphous electrically-conductive seed layer; and a fixed magnetic layer juxtaposed and in contact with the seed layer. A method including forming an amorphous seed layer on a first electrode of a memory device; forming a material layer stack on the amorphous seed layer, the material stack including a dielectric layer disposed between a fixed magnetic layer and a free magnetic layer, wherein the fixed magnetic layer.

Abstract: Resistive memory cells are described. In some embodiments, the resistive memory cells include a switching layer having an inner region in which one or more filaments is formed. In some instances, the filaments is/are formed only within the inner region of the switching layer. Methods of making such resistive memory cells and devices including such cells are also described.

Abstract: A method including forming a device stack including a dielectric layer between a fixed magnetic layer and a free magnetic layer on a fully-crystalline sacrificial film or substrate including a crystal lattice similar to the crystal lattice of the dielectric material; and transferring the device stack from the sacrificial film to a device substrate. An apparatus including a device stack including a dielectric layer between a fixed magnetic layer and a free magnetic layer on a device substrate wherein the fixed magnetic layer and the free magnetic layer each have a crystalline lattice conforming to a crystalline lattice of the sacrificial film or substrate on which they were formed prior to transfer to the device substrate.

Abstract: An interlayer is used to reduce Fermi-level pinning phenomena in a semiconductive device with a semiconductive substrate. The interlayer may be a rare-earth oxide. The interlayer may be an ionic semiconductor. A metallic barrier film may be disposed between the interlayer and a metallic coupling. The interlayer may be a thermal-process combination of the metallic barrier film and the semiconductive substrate. A process of forming the interlayer may include grading the interlayer. A computing system includes the interlayer.

Abstract: A method of patterning a semiconductor film is described. According to an embodiment of the present invention, a hard mask material is formed on a silicon film having a global crystal orientation wherein the semiconductor film has a first crystal plane and second crystal plane, wherein the first crystal plane is denser than the second crystal plane and wherein the hard mask is formed on the second crystal plane. Next, the hard mask and semiconductor film are patterned into a hard mask covered semiconductor structure. The hard mask covered semiconductor structured is then exposed to a wet etch process which has sufficient chemical strength to etch the second crystal plane but insufficient chemical strength to etch the first crystal plane.

Abstract: Trenches (and processes for forming the trenches) are provided that reduce or prevent crystaline defects in selective epitaxial growth of type III-V or Germanium (Ge) material (e.g., a “buffer” material) from a top surface of a substrate material. The defects may result from collision of selective epitaxial sidewall growth with oxide trench sidewalls. Such trenches include (1) a trench having sloped sidewalls at an angle of between 40 degrees and 70 degrees (e.g., such as 55 degrees) with respect to a substrate surface; and/or (2) a combined trench having an upper trench over and surrounding the opening of a lower trench (e.g., the lower trench may have the sloped sidewalls, short vertical walls, or tall vertical walls). These trenches reduce or prevent defects in the epitaxial sidewall growth where the growth touches or grows against vertical sidewalls of a trench it is grown in.

Abstract: The present disclosure relates to the fabrication of spin transfer torque memory elements for non-volatile microelectronic memory devices. The spin transfer torque memory element may include a magnetic tunneling junction connected with specifically sized and/or shaped fixed magnetic layer that can be positioned in a specific location adjacent a free magnetic layer. The shaped fixed magnetic layer may concentrate current in the free magnetic layer, which may result in a reduction in the critical current needed to switch a bit cell in the spin transfer torque memory element.

Abstract: III-N transistors with recessed gates. An epitaxial stack includes a doped III-N source/drain layer and a III-N etch stop layer disposed between a the source/drain layer and a III-N channel layer. An etch process, e.g., utilizing photochemical oxidation, selectively etches the source/drain layer over the etch stop layer. A gate electrode is disposed over the etch stop layer to form a recessed-gate III-N HEMT. At least a portion of the etch stop layer may be oxidized with a gate electrode over the oxidized etch stop layer for a recessed gate III-N MOS-HEMT including a III-N oxide. A high-k dielectric may be formed over the oxidized etch stop layer with a gate electrode over the high-k dielectric to form a recessed gate III-N MOS-HEMT having a composite gate dielectric stack.

Abstract: Different n- and p-types of device fins are formed by epitaxially growing first epitaxial regions of a first type material from a substrate surface at a bottom of first trenches formed between shallow trench isolation (STI) regions. The STI regions and first trench heights are at least 1.5 times their width. The STI regions are etched away to expose the top surface of the substrate to form second trenches between the first epitaxial regions. A layer of a spacer material is formed in the second trenches on sidewalls of the first epitaxial regions. Second epitaxial regions of a second type material are grown from the substrate surface at a bottom of the second trenches between the first epitaxial regions. Pairs of n- and p-type fins can be formed from the first and second epitaxial regions. The fins are co-integrated and have reduced defects from material interface lattice mismatch.

Abstract: The present disclosure provides a system and method for forming a resistive random access memory (RRAM) device. A RRAM device consistent with the present disclosure includes a substrate and a first electrode disposed thereon. The RRAM device includes a second electrode disposed over the first electrode and a RRAM dielectric layer disposed between the first electrode and the second electrode. The RRAM dielectric layer has a recess at a top center portion at the interface between the second electrode and the RRAM dielectric layer. The present disclosure further provides a computing device. The computing device includes a motherboard, a processor mounted on the motherboard, and a communication chip fabricated on the same chip as the processor or mounted on the motherboard. The processor comprises a substrate, a first electrode, second electrode, and a RRAM layer which has a recess at the interface between the second electrode and RRAM oxide layer.

Abstract: A transistor having a narrow bandgap semiconductor source/drain region is described. The transistor includes a gate electrode formed on a gate dielectric layer formed on a silicon layer. A pair of source/drain regions are formed on opposite sides of the gate electrode wherein said pair of source/drain regions comprise a narrow bandgap semiconductor film formed in the silicon layer on opposite sides of the gate electrode.

Abstract: Embodiments of semiconductor assemblies, and related integrated circuit devices and techniques, are disclosed herein. In some embodiments, a semiconductor assembly may include a flexible substrate, a first barrier formed of a first transition metal dichalcogenide (TMD) material, a transistor channel formed of a second TMD material, and a second barrier formed of a third TMD material. The first barrier may be disposed between the transistor channel and the flexible substrate, the transistor channel may be disposed between the second barrier and the first barrier, and a bandgap of the transistor channel may be less than a bandgap of the first barrier and less than a bandgap of the second barrier. Other embodiments may be disclosed and/or claimed.

Abstract: CMOS circuits may formed using p-channel gallium nitride transistors and n-channel gallium nitride transistors, wherein both the p-channel gallium nitride transistors and the n-channel gallium nitride transistors are formed on a single layered structure comprising a polarization layer deposited on a first gallium nitride layer and a second gallium nitride layer deposited on the polarization layer. Having both n-channel gallium nitride transistors and p-channel gallium nitride transistors s on the same layer structure may enable “all gallium nitride transistor” implementations of circuits including logic, digital, and analog circuitries spanning low supply voltages to high supply voltages.