摘要:
A multiple-channel semiconductor device has fully or partially depleted quantum wells and is especially useful in ultra large scale integration devices, such as CMOSFETs. Multiple channel regions are provided on a substrate with a gate electrode formed on the uppermost channel region, separated by a gate oxide, for example. The vertical stacking of multiple channels and the gate electrode permit increased drive current in a semiconductor device without increasing the silicon area occupied by the device.
摘要:
A semiconductor device comprising a substrate having a first crystal orientation is provided. A first insulating layer overlies the substrate and a plurality of silicon layers overlie the first insulating layer. A first silicon layer comprises silicon having a second crystal orientation and a crystal plane. A second silicon layer comprises silicon having the second crystal orientation and a crystal plane that is substantially orthogonal to the crystal plane of the first silicon layer. Because holes have higher mobility in the (110) plane than the (100) plane, while electrons have higher mobility in (100) plane than the (110) plane, semiconductor device performance can be enhanced by the selection of silicon layers with certain crystal plane orientations. In addition, a method of forming a semiconductor device is provided. A silicon-on-insulator structure comprising a first silicon substrate having a first crystal orientation with a first insulating layer formed thereon and a first silicon layer having a second crystal orientation and a crystal plane overlying the first insulating layer is bonded to a second silicon substrate. The second silicon substrate has the second crystal orientation and a crystal plane and a second insulating layer formed thereon. The second silicon substrate comprises a line of defects created by implanting hydrogen ion into the second silicon substrate. The crystal plane of the second silicon substrate is oriented substantially orthogonal to the crystal plane of the first silicon layer. The second silicon substrate is split and removed along the line of defects leaving behind the second insulating layer and a second silicon layer on the silicon-on-insulator structure. A plurality of devices with different crystal orientations can be subsequently formed on a single, planar silicon-on-insulator structure by selectively etching the silicon-on-insulator structure down to silicon layers of different crystal orientations, growing selective epitaxial silicon layers in the etched regions, and subsequently planarizing the silicon-on-insulator structure by chemical-mechanical polishing.
摘要:
A semiconductor structure and a process for its manufacture. A metal gate electrode is formed on a semiconductor substrate, the gate electrode being between nitride spacers. Lightly-doped drain regions and source/drain regions are disposed in the substrate and aligned with the electrode and spacers. A silicide contact layer is disposed over an epitaxial layer on the substrate over the source/drain regions.
摘要:
A method for slowing the diffusion of boron ions in a CMOS structure includes a preanneal step which can be incorporated as part of a step in which silane is deposited across the surface of the wafer. After the last implant on a CMOS device, silane (SiH.sub.4) is deposited over the surface of the wafer using a chemical vapor deposition (CVD) tool. The deposition of silane is done at 400.degree. C. The temperature is raised in the CVD tool to a temperature in the range of 550.degree. C. to 650.degree. C. and held for 30-60 minutes. This temperature does not affect the thin film of silicon which is formed from the silane, yet provides the necessary thermal cycle to "repair" the crucial first 200 .ANG. to 600 .ANG. of the silicon surface. Normal processing steps, including a rapid thermal anneal for 30 seconds at 1025.degree. C. follow. The RTA is necessary to activate the dopants (arsenic and boron) in the source and drain of the respective devices. The boron dopant species diffuses less during subsequent rapid thermal anneal cycles since the crucial first 200 .ANG. to 600 .ANG. of the silicon surface have been repaired using this preanneal step.
摘要:
One method of forming a semiconductor device includes forming a gate electrode on a substrate and then forming a spacer adjacent to a sidewall of the gate electrode. An active region is formed in the substrate adjacent to the spacer and spaced apart from the gate electrode using a first dopant material of a first conductivity type. A protecting layer is formed over the active region and adjacent to the spacer. At least a portion of the spacer is then removed to form an opening between the protecting layer and the gate electrode. In some instances, the spacer may be formed by independent deposition of two different materials (e.g., silicon nitride and silicon dioxide), one of which can be selectively removed with respect to the other. A lightly-doped region is formed in the substrate adjacent to the gate electrode using a second dopant material of the first conductivity type. This lightly-doped region may be formed, for example, prior to formation of the spacer, between the formation of two portions of the spacer, or after removing at least a portion of the spacer. A halo region is formed through the opening resulting from removing a portion of the spacer. The halo region is deeper in the substrate than the lightly-doped region and is adjacent to the active region. The halo region is formed using a third dopant material of a conductivity type different than the first conductivity type.
摘要:
A CMOS semiconductor device having NMOS source/drain regions formed using multiple spacers has at least one NMOS region and at least one PMOS region. A first n-type dopant is selectively implanted into an NMOS active region of the substrate adjacent a NMOS gate electrode to form a first n-doped region in the NMOS active region. A first NMOS spacer is formed on a sidewall of the NMOS gate electrode and a first PMOS spacer on a sidewall of a PMOS gate electrode. A second n-type dopant is selectively implanted into the NMOS active region using the first NMOS spacer as a mask. A p-type dopant is selectively implanted into a PMOS active region using the first PMOS spacer as a mask to form a first p-doped region in the PMOS active region. A second NMOS spacer and a second PMOS spacer are formed adjacent the first NMOS spacer and first PMOS spacer, respectively. A third n-type dopant is selectively implanted into the NMOS active region using the second NMOS spacer as a mask to form a third n-doped region deeper than the second n-doped region in the NMOS active region. A second p-type dopant is selectively implanted into the PMOS active region using the second PMOS spacer as a mask to form a second p-doped region in the PMOS active region deeper than the first p-doped region.
摘要:
The formation of a shortage protection region is disclosed. In one embodiment, a method includes three steps. In the first step, a first ion implantation is applied to form lightly doped regions within a semiconductor substrate adjacent to sidewalls of a gate over the substrate. In the second step, two spaces are formed on the substrate, each adjacent to a sidewall of the gate, so that a second ion implantation forms heavily doped regions within the substrate adjacent to the first spacers. In the third step, two additional spacers are formed on the substrate, each overlapping and extending beyond a corresponding spacer previously formed. Thus, a third ion implantation forms lightly doped shortage protection regions within the substrate adjacent to the spacers most recently formed.
摘要:
A method and structure for optimizing the performance of a semiconductor device having dense transistors. A method consistent with the present invention includes forming a first test structure on a first substrate portion. The first test structure includes a transistor having a gate electrode formed at a design width and at a first line spacing similar to the line spacing of a dense transistor. One or more electrical properties the transistor of the first test structure is measured. A second test structure is formed on a second substrate portion. The second test structure includes a transistor having a gate electrode formed at the same design width as the transistor of the first test structure and at a second line spacing greater than the first line spacing. One or more electrical properties of the transistor of the second test structure are measured. Using the measured one or more electrical properties, one or more relationships are developed between the measured one or more electrical properties and the transistors at the first line spacing and the second line spacing.
摘要:
A multiple-channel semiconductor device has fully or partially depleted quantum wells and is especially useful in ultra large scale integration devices, such as CMOSFETs. Multiple channel regions are provided on a substrate with a gate electrode formed on the uppermost channel region, separated by a gate oxide, for example. The vertical stacking of multiple channels and the gate electrode permit increased drive current in a semiconductor device without increasing the silicon area occupied by the device.
摘要:
STI divot formation is eliminated or substantially reduced by employing a very thin nitride polish stop layer, e.g., no thicker than 400 Å. The very thin nitride polish stop layer is retained in place during subsequent masking, implanting and cleaning steps to form dopant regions, and is removed prior to gate oxide and gate electrode formation.