1. Technical Field
The present invention relates generally to the field of semiconductor manufacturing technologies and more particularly to GAA (Gate-All-Around) CMOSFET (Complementary Metal Oxide Semiconductor Field Effect Transistor) devices.
2. Description of Related Art
A CMOS device integrates both NMOS (N-type Metal Oxide Semiconductor) and PMOS (P-type Metal Oxide Semiconductor) transistors in one device. As the device size continues to shrink, a major challenge in scaling down the channel length is to maintain a high current drive capability (Ion) and a good stable threshold voltage, and control the device leakage current (Ioff) within requirement. The short-channel effect degrades device performance and is a severe obstacle to scale down conventional planar CMOS devices.
For nanometer-scale channel length CMOS devices, it is important to control the channel conductance mainly through a gate electric field without being affected by a drain scattering electric field. For SOI devices, the above-described problem is alleviated with the reduced silicon thickness in both partial-depletion and full-depletion structures. Compared with the conventional planar CMOS devices, inversion mode dual-gate or tri-gate fin-type FETs have better gate control and scaling down capabilities. Besides operating in an inversion mode, ultra-thin SOI devices can operate in an accumulation mode. Comparing to the full-depletion FET, in an accumulation mode, current flows through the whole SOI device, which increases the carrier mobility, reduces low-frequency noises, lowers the short channel effect, and increases the threshold voltage so as to avoid polysilicon gate depletion effect. In an inversion mode FET, the type of impurities doped in the source and drain regions is different from that in the channel region, the conductive carriers are of minority carriers, and p-n junctions are formed between the source region and the channel region and between the drain region and the channel region, respectively. The inversion mode FETs are currently the most widely used devices. On the other hand, in an accumulation mode FET, the source and drain regions are doped with impurities of the same type as that in the channel region, the conductive carriers is of majority carriers, and there is no p-n junction. Since the carrier mobility is the bulk material mobility, the accumulation mode FET achieves high carrier mobility.
Further, in Si(110) substrates, current flows along <110>crystal orientation, hole mobility is more than doubled compared with in conventional Si(100) substrates, and electron mobility is the highest in Si(100) substrates. To fully utilize the advantage of the carrier mobility depending on crystalline orientation, M. Yang et al. at IBM developed a CMOS fabricating technology on hybrid substrates with different crystal orientations (‘High performance CMOS fabricated on hybrid substrate with different crystal orientations’, Digest of Technical Paper of International Electron Devices Meeting, 2003). Through bonding and selective epitaxy growth techniques, an NMOS device is fabricated on a Si (100) surface and a PMOS device is fabricated on a Si (110) surface. The paper reported the drive current of the PMOS device on the Si(110) substrate increases by 45%, when Ioff=100 nA/μm. But the PMOS device is fabricated on an epitaxy layer and does not have a buried oxide layer to isolate it from the substrate, thus adversely affecting the performance of the device performance.
Therefore, the present invention provides gate-all-around CMOSFET devices to overcome the above-described problems.
The present invention provides a gate-all-around CMOSFET device, which includes a semiconductor substrate, a PMOS region having a first channel, an NMOS region having a second channel, and a gate region, wherein the NMOS region is disposed above the semiconductor substrate and the PMOS region is disposed above the NMOS region. The NMOS region and the PMOS region each includes a source region and a drain region located at the two opposite ends of the channel thereof. The surfaces of the first channel and the second channel are substantially surrounded by the gate region. A first buried insulation layer is disposed between the PMOS region and the NMOS region; and a second buried insulation layer is disposed between the NMOS region and the semiconductor substrate. The PMOS region is formed of Si(110), and the NMOS region is formed of Si(100).
In another embodiment of the present invention, a gate-all-around CMOSFET device includes a semiconductor substrate, a PMOS region having a first channel and disposed above the semiconductor substrate, an NMOS region having a second channel and disposed above the PMOS region, and a gate region. The PMOS region and the NMOS region each includes a source region and a drain region located at the two opposite ends of the channel thereof. The surfaces of the first channel and the second channel are substantially surrounded by the gate region. A first buried insulation layer is disposed between the NMOS region and the PMOS region; and a second buried insulation layer is disposed between the PMOS region and the semiconductor substrate. The PMOS region is formed of Si(110), and the NMOS region is formed of Si(100).
The device structure of the present invention is simple, compact and highly integrated. The buried insulation layers efficiently avoid inter-region interference and reduce leakage current such that the devices achieve better performance and better scaling down capability. Therefore, the GAA CMOSFET devices have high carrier mobility and reduced low-frequency noise, and prevent polysilicon gate depletion and short-channel effects.
a-1c show the structure of a GAA CMOSFET device according to a first embodiment of the present invention, wherein
b is a cross-sectional view along a line XX′ in
d is a perspective view of the structure of a GAA CMOSFET device according to the first embodiment of the present invention;
e is a top view of a finished GAA CMOSFET device according to the first embodiment of the present invention;
f is a cross-sectional view along a line XX′ in
a-2c show the structure of a GAA CMOSFET device according to a third embodiment of the present invention, wherein
d is a top view of a finished GAA CMOSFET device according to the third embodiment of the present invention; and
e is a cross-sectional view along a line XX′ in
The following illustrative embodiments are provided to illustrate the disclosures of the present invention. It should be noted that figures are schematic representations of devices, and not drawn to scale.
Referring to
As shown in
Referring to
Referring to
The CMOSFET device of the present invention can be fabricated with conventional planar CMOS technologies. First, a SOI substrate with a Si(100) layer and a Si(110) layer is provided, and n-type channel ion implantation and p-type channel ion implantation are performed to the Si(100) layer and the Si(110) layer, respectively; then, lithography and dry etching processes are performed such that a fin-shaped active area is formed in the Si(100) layer and the Si(110) layer, and the narrow portions of the fin-shaped active areas function as an n-type channel and a p-type channel, respectively; thereafter, a buffer oxide etchant is used to selectively etch away the buried oxide layers under the narrow portions so as to form tunnels under the channels, and a high-temperature annealing treatment in hydrogen atmosphere is performed to partially melt the narrow portions, as such, the surface tension causes the cross section of the channels to change from a rectangular shape to a circular shape, i.e., the channels obtain a cylindrical shape; then, a low-pressure chemical vapor deposition or atomic layer deposition (ALD) is performed to grow a gate dielectric layer that surrounds the surfaces of the channels, respectively, and a gate electrode material layer is deposited on the gate dielectric layer, and then, lithography and dry etching processes are performed to form a gate electrode; thereafter, source and drain regions for both of the NMOS and PMOS devices are formed by self-aligned ion implantations; then, the lower layer of the source and drain regions are exposed through lithography and dry etching processes such that source and drain electrode contacts can be fabricated therein. Finally, a CMOSFET device is finished based on the above structure.
In order to further analyze the device performance, a 3D simulation adopting a precise hydraulic model and a quantum mechanical density gradient model and applying a mobility degradation model related to doping and surface roughness is established. According to the simulation results, since current flows through the overall cylindrical-shaped channels, the accumulation mode GAA CMOSFET of the present invention achieves high carrier mobility, reduces low-frequency noise, prevents polysilicon gate depletion and short-channel effects, and increases the threshold voltage of the device. The Ion/Ioff ratio of the device can be larger than 106. Therefore, the GAA CMOSFET device of the present invention has better performance and scaling down capability compared with the conventional multi-gate FinFETs.
The second embodiment discloses an inversion mode GAA CMOSFET device. Different from the first embodiment, the first channel of the second embodiment is preferably formed of n-type Si(110) and the second channel is preferably formed of p-type Si(100).
The device characteristics have also been simulated and analyzed. The results show that the inversion mode GAA CMOSFET of the second embodiment also achieves high carrier mobility, reduces low-frequency noise, and prevents polysilicon gate depletion and short-channel effects.
a-2c show another accumulation mode GAA CMOSFET device, which includes: a semiconductor substrate 100′, a PMOS region 400′ having a first channel 401′ and disposed above the semiconductor substrate 100′, an NMOS region 300′ having a second channel 301′ and disposed above the PMOS region 400′, and a gate region 500′. The PMOS region 400′ comprises a source region 403′ and a drain region 402′ located at the opposite ends of the channel 401′. The NMOS region 300′ comprises a source region 303′ and a drain region 302′ located at the opposite ends of the channel 301′. Each of the channels 401′, 301′ has a substantially circular-shaped cross section, that is, each of the channels 401′ and 301′ has a cylindrical shape. The channel 401′ is preferably formed of p-type Si(110) and the second channel 301′ is preferably formed of n-type Si(100). The gate region 500′ substantially surrounds the surfaces of the channels 401′, 301′. A first buried insulation layer 201′ is disposed between the PMOS region 400′ and the NMOS region 300′, other than the gate region 500′, to avoid inter-region interference. A second buried insulation layer 202′ is disposed between the PMOS region 400′ and the underlying semiconductor substrate 100′, other than the gate region 500′, to isolate the PMOS region 400′ from the semiconductor substrate 100′. The buried insulation layers effectively reduce the leakage current and improve the device performance. The source region 403′ and the drain region 402′ of the PMOS region 400′ are formed of heavily doped p-type Si(110), and the source region 303′ and the drain region 302′ of the NMOS region 300′ are formed of heavily doped n-type Si(100).
As shown in
Referring to
Different from the first embodiment, the present embodiment has the NMOS region 300′ on top and the PMOS region 400′ close to the substrate 100′.
A FET device is finished based on the structure of
The fourth embodiment discloses another GAA CMOSFET device in inversion mode. Different from the third embodiment, the first channel of the fourth embodiment is preferably formed of n-type Si(110) and the second channel is preferably formed of p-type Si(100).
The above description of the detailed embodiments are only to illustrate the preferred implementation according to the present invention, and it is not to limit the scope of the present invention, Accordingly, all modifications and variations completed by those with ordinary skill in the art should fall within the scope of present invention defined by the appended claims.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/CN10/70633 | 2/11/2010 | WO | 00 | 6/25/2010 |