Abstract: An apparatus, system, and method for dual-channel FET devices is presented. In some embodiments, the nanowire FET device may include a first transistor on a substrate, where the first transistor includes a first group of nanowires made of silicon. The nanowire FET device may also include a second transistor on the same substrate, where the second transistor includes a second group of nanowires made of silicon-germanium.

Abstract: An apparatus, system, and method for dual-channel FET devices is presented. In some embodiments, the nanowire FET device may include a first transistor on a substrate, where the first transistor includes a first group of nanowires made of silicon. The nanowire FET device may also include a second transistor on the same substrate, where the second transistor includes a second group of nanowires made of silicon-germanium.

Abstract: A stress memorization technique (SMT) film is deposited over a semiconductor device. The SMT film is annealed with a low thermal budget anneal that is sufficient to create and transfer the stress of the SMT film to the semiconductor device. The SMT film is then removed. After the SMT film is removed, a second anneal is applied to the semiconductor device sufficiently long and at a sufficiently high temperature to activate dopants implanted for forming device source/drains. The result of this approach is that there is minimal gate dielectric growth in the channel along the border of the channel.

Abstract: A stress memorization technique (SMT) film is deposited over a semiconductor device. The SMT film is annealed with a low thermal budget anneal that is sufficient to create and transfer the stress of the SMT film to the semiconductor device. The SMT film is then removed. After the SMT film is removed, a second anneal is applied to the semiconductor device sufficiently long and at a sufficiently high temperature to activate dopants implanted for forming device source/drains. The result of this approach is that there is minimal gate dielectric growth in the channel along the border of the channel.

Abstract: A metal oxide, utilized as a gate dielectric, is removed using a combination of gaseous HCl (HCl), heat, and an absence of rf. The metal oxide, which is preferably hafnium oxide, is effectively removed in the areas not under the gate electrode. The use of HCl results in the interfacial oxide that underlies the metal oxide not being removed. The interfacial is removed to eliminate the metal and is replaced by another interfacial oxide layer. The subsequent implant steps are thus through just an interfacial oxide and not through a metal oxide. Thus, the problems associated with implanting through a metal oxide are avoided.

Abstract: A semiconductor device (10) having two different gate dielectric thicknesses is formed using a single high-k dielectric layer, preferably a metal oxide. A thicker first gate dielectric (16) is formed in a region of the device for higher voltage requirements, e.g. an I/O region (24). A thinner second gate dielectric (20) is formed in a region of the device for lower voltage requirements, e.g. a core device region (22). First and second dielectrics are preferably silicon dioxide or oxynitride. A metal oxide (26) is deposited over both dielectrics, followed by deposition of a gate electrode material (28). By using a single metal oxide layer in forming the gate dielectric stack for each transistor, together with high quality silicon dioxide or oxynitride dielectric layers, problems associated with selective etching of the metal oxide may be avoided, as may problems associated with various interfaces between the metal oxide and damaged or treated surfaces.

Abstract: A transistor device has a gate dielectric with at least two layers in which one is hafnium oxide and the other is a metal oxide different from hafnium oxide. Both the hafnium oxide and the metal oxide also have a high dielectric constant. The metal oxide provides an interface with the hafnium oxide that operates as a barrier for contaminant penetration. Of particular concern is boron penetration from a polysilicon gate through hafnium oxide to a semiconductor substrate. The hafnium oxide will often have grain boundaries in its crystalline structure that provide a path for boron atoms. The metal oxide has a different structure than that of the hafnium oxide so that those paths for boron in the hafnium oxide are blocked by the metal oxide. Thus, a high dielectric constant is provided while preventing boron penetration from the gate electrode to the substrate.

Abstract: A semiconductor device (10) having two different gate dielectric thicknesses is formed using a single high-k dielectric layer, preferably a metal oxide. A thicker first gate dielectric (16) is formed in a region of the device for higher voltage requirements, e.g. an I/O region (24). A thinner second gate dielectric (20) is formed in a region of the device for lower voltage requirements, e.g. a core device region (22). First and second dielectrics are preferably silicon dioxide or oxynitride. A metal oxide (26) is deposited over both dielectrics, followed by deposition of a gate electrode material (28). By using a single metal oxide layer in forming the gate dielectric stack for each transistor, together with high quality silicon dioxide or oxynitride dielectric layers, problems associated with selective etching of the metal oxide may be avoided, as may problems associated with various interfaces between the metal oxide and damaged or treated surfaces.

Abstract: A transistor device has a gate dielectric with at least two layers in which one is hafnium oxide and the other is a metal oxide different from hafnium oxide. Both the hafnium oxide and the metal oxide also have a high dielectric constant. The metal oxide provides an interface with the hafnium oxide that operates as a barrier for contaminant penetration. Of particular concern is boron penetration from a polysilicon gate through hafnium oxide to a semiconductor substrate. The hafnium oxide will often have grain boundaries in its crystalline structure that provide a path for boron atoms. The metal oxide has a different structure than that of the hafnium oxide so that those paths for boron in the hafnium oxide are blocked by the metal oxide. Thus, a high dielectric constant is provided while preventing boron penetration from the gate electrode to the substrate.

Abstract: Techniques for forming gate dielectric layers (702) overlying amorphous substrate materials are presented. In addition, techniques for low temperature processing operations that allow for the use of amorphous silicon in doping operations are presented. The amorphous silicon regions (604, 606) are formed prior to formation of structures included in the gate structure (804) of the semiconductor device, where the gate structures (804) are preferably formed using low temperature operations that allow the amorphous silicon regions (604, 606) to remain in an amorphous state. Source/drain regions (1004, 1006) are formed in the amorphous silicon regions (604, 606), and then the substrate is annealed to recrystallize the amorphous regions.

Abstract: In accordance with a specific embodiment of the present invention, a method of forming a gate dielectric is disclosed. A semiconductor wafer is placed in a deposition chamber. The semiconductor wafer is heated and a precursor gas is flowed into the chamber. In one embodiment, the precursor comprises a moiety of silicon, oxygen, and a transition metal. In another embodiment, the moiety includes a group 2 metal.

Abstract: A transistor device (19) utilizes a high K dielectric (24) between a gate electrode (16) and a substrate (12). The high K dielectric (24) is etched under the gate electrode (16) so that there is an area between the gate electrode (16) and the substrate (12) that is void of high K dielectric (24). The source/drains extensions (28 and 30) are minimized to extend substantially in alignment with the edge of gate dielectric (24) to reduce overlap with the gate dielectric (24). This results in reduced capacitance between the gate and the source/drain extensions. The void areas (20 and 22) between the gate and the substrate (12) may remain void or may be filled with a low K dielectric, or at least a dielectric that is not high K.

Abstract: A metal oxide, utilized as a gate dielectric, is removed using a combination of gaseous HCl (HCl), heat, and an absence of rf. The metal oxide, which is preferably hafnium oxide, is effectively removed in the areas not under the gate electrode. The use of HCl results in the interfacial oxide that underlies the metal oxide not being removed. The interfacial is removed to eliminate the metal and is replaced by another interfacial oxide layer. The subsequent implant steps are thus through just an interfacial oxide and not through a metal oxide. Thus, the problems associated with implanting through a metal oxide are avoided.

Abstract: A zirconium oxide gate dielectric is utilized to make a transistor for an integrated circuit. In forming the transistor which uses the zirconium oxide as the gate dielectric, the zirconium oxide is etched using phosphoric acid. This phosphoric acid is a wet etch performed at an elevated temperature to achieve a simultaneous etch of a silicon nitride anti-reflective coating which overlies the gate. This use of phosphoric acid is effective because the etch is stopped by silicon oxide so that the underlying silicon substrate is protected. Also, the field oxide is not etched in any appreciable amount. Thus, the field oxide thickness is unchanged as a result of the etch.

Abstract: Techniques for forming gate dielectric layers (702) overlying amorphous substrate materials are presented. In addition, techniques for low temperature processing operations that allow for the use of amorphous silicon in doping operations are presented. The amorphous silicon regions (604, 606) are formed prior to formation of structures included in the gate structure (804) of the semiconductor device, where the gate structures (804) are preferably formed using low temperature operations that allow the amorphous silicon regions (604, 606) to remain in an amorphous state. Source/drain regions (1004, 1006) are formed in the amorphous silicon regions (604, 606), and then the substrate is annealed to recrystallize the amorphous regions.

Abstract: A method for forming a semiconductor device is disclosed in which a metal oxide gate dielectric layer is formed over a substrate. A gate electrode is then formed over the metal oxide layer thereby exposing a portion of the metal oxide layer. The exposed portion of the metal oxide gate dielectric layer is then chemically reduced to a metal or a metal hydride. The metal or metal hydride is then removed with a conventional wet etch or wet/dry etch combination. The metal oxide layer may include a metal element such as zirconium, tantalum, hafnium, titanium, or lanthanum and may further include an additional element such as silicon or nitrogen. Reducing the metal oxide layer may includes annealing the metal oxide gate dielectric layer in an ambient with an oxygen partial pressure that is less than a critical limit for oxygen desorption at a given temperature.

Abstract: A method for forming a tantalum-based anti-reflective coating (ARC) layer begins by forming an MOS metallic gate electrode layer (20) over a substrate (20). The MOS metallic gate electrode layer (20) is covered with an ARC layer (22). The ARC layer is preferably tantalum pentoxide or a tantalum pentoxide layer doped with one or more of nitrogen atoms and/or silicon atoms. The layers (22 and 20) are then selectively masked photoresist (24) that is selectively exposed to deep ultraviolet (DUV) radiation (28). The ARC layer (22) improves lithographic critical dimension (CD) control of the MOS metallic gate during exposure. The final MOS metallic gate is then patterned and etched using a fluorine-chlorine-fluorine time-progressed reactive ion etch (RIE) process, whereby metallic-gate MOS transistors are eventually formed.

Abstract: A method of processing a high K gate dielectric includes growing a high quality silicon dioxide layer at the silicon interface followed by deposition of a metal layer, which is then diffused into the silicon dioxide. Preferred metals include zirconium and hafnium. A gate stack may be fabricated by adding a metal containing layer to an existing thermally grown SiO2 or a combination of SiO2, SiO3 and SiO4 (oxide-nitride or oxynitride) stacks.

Abstract: First and second dummy structures (201 and 202) are formed over a semiconductor device substrate (10). In one embodiment, portions of the first dummy structure (201) are removed and replaced with a first conductive material (64) to form a first gate electrode (71) and portions of second dummy structure (202) are removed and replaced with a second conductive material (84) to form a second gate electrode (91). In an alternate embodiment, the dummy structures (201 and 202) are formed using a first conductive material (164) that is used to form the first electrode (71). The second electrode is then formed by removing the first conductive material (164) from dummy structures (202) and replacing it with a second conductive material (84). In accordance with embodiments of the present invention, the first conductive material and the second conductive material are different conductive materials.

Abstract: A method for forming a tantalum-based anti-reflective coating (ARC) layer begins by forming an MOS metallic gate electrode layer (20) over a substrate (20). The MOS metallic gate electrode layer (20) is covered with an ARC layer (22). The ARC layer is preferably tantalum pentoxide or a tantalum pentoxide layer doped with one or more of nitrogen atoms and/or silicon atoms. The layers (22 and 20) are then selectively masked photoresist (24) that is selectively exposed to deep ultraviolet (DUV) radiation (28). The ARC layer (22) improves lithographic critical dimension (CD) control of the MOS metallic gate during exposure. The final MOS metallic gate is then patterned and etched using a fluorine-chlorine-fluorine time-progressed reactive ion etch (RIE) process, whereby metallic-gate MOS transistors are eventually formed.