Strained finfet cmos device structures

Description of equivalent WO 2004049406 A1

0001Strained FinFET CMOS Device Structures

0002Technical Field

0003The present invention relates to double gate semiconductor device structures and, more particularly, to FinFET devices.

0004Background Art

0005Double gate semiconductor device structures are promising candidates for future generation microelectronics devices, because of the ability to obtain near ideal sub-threshold slope, absence of body-effect, immunity to short-channel effect, and very high current driveability. One double gate device structure of technological relevance is the FinFET. The FinFET is particularly attractive because of the relative simplicity of manufacture compared to other double gate devices. The channel for the FinFET is a thin rectangular island of Si, commonly referred to as the Fin. The gate wraps around the Fin so that the channel is gated on both sides of the vertical portions of the Fin structure, providing gate control which is superior to planar single gate MOSFETs.

0006FinFETs are well known. See, for example, U.S. Patent No. 6,413,802, entitled "FinFET Transistor Structures Having a Double Gate Channel Extending Vertically from a Substrate and Methods of Manufacture" by Hu et al., filed October 23, 2000, issued July 2, 2002, which is hereby incorporated in its entirety by reference. FinFETs having enhanced mobility are also known. See, for example, U.S. Patent Application No.. 2002/0063292 Al, entitled "CMOS Fabrication Process Utilizing Special Transistor Orientation" by Armstrong et al., filed November 29, 2000, published May 30, 2002, which is hereby incorporated in its entirety by reference. This prior art method is directed at improving nFET mobility and, therefor, only limited improvement in CMOS circuits can be attained. Thus, a need exists for a method to improve mobility for p-FinFETs and n-FinFETs which are situated on the same wafer.

0007However, the present inventors believe that improvements in utilizing stressing layers to enhance mobility are achievable.

0008Disclosure of Invention

0009According to the present invention, a semiconductor device structure includes a PMOS device and an NMOS device disposed on a substrate, the PMOS device including a compressive layer stressing an active region of the PMOS device, the NMOS device including a tensile layer stressing an active region of the NMOS device, wherein the compressive layer includes a first dielectric material, the tensile layer includes a second dielectric material, and the PMOS and NMOS devices are FinFET devices.

0010This invention is directed to a novel strained FinFET device structures for enhanced mobility. The integration scheme incorporates a new process flow that induces compressive stress in the longitudinal direction for the p-FinFET while also inducing tensile stress in the longitudinal direction for the n-FinFET. These stresses significantly enhance mobility and, thus, enhance device performance. In the invention described herein, the longitudinal stress induced in the channel is enhanced considerably from what can be obtained in standard planar MOSFETs, because the stressing film is applied from two sides of a thin FinFET, instead of on the surface of an SOI layer or bulk substrate.

0011It is a principal object of the present invention to enhance mobility in double gate CMOS device structures.

0012It is an additional object of the present invention to enhance mobility in FinFET device structures. It is a further object of the present invention to improve methods of fabricating strained FinFET device structures.

0013The invention and other objects and features thereof will be more readily apparent from the following detailed description when taken in conjunction with the following drawing figures.

0014Brief Description of Drawings

0015Figures 1 and 2 schematically show perpendicular (to Fin) and parallel (to Fin) views of a FinFET according to the prior art.

0016Figure 3 schematically shows a peφendicular view of a FinFET semiconductor device structure according to the prior art.

0017Figures 4 - 17 schematically show various views of intermediate and final FinFET semiconductor device structures according to the present invention.

0018Best Mode for Carrying Out the Invention

0019The present invention is directed to a novel FinFET semiconductor device structure and a method of manufacturing such structure. A preferred final structure according to the present invention is shown in Figures 16, 17.

0020Referring now to the remaining figures and Figures 1 - 3 in particular, there are shown known FinFET devices (Figs. 1, 2) and device structures (Fig. 3).

0021To begin, standard or conventional FinFET device manufacture processing is followed through the patterning and etching of the Fins, formation of gate dielectric and conductors, sidewall spacers (not shown), source/drain doping, and salicidation. Following salicidation, the gate sidewall spacers are removed, to facilitate the processing which will induce strain in the Fins according to the present invention.

0022More particularly, as described with reference to Fig. 3, an SOI wafer, for example, is provided. The SOI wafer includes a substrate 1, disposed underneath a buried SiO₂ layer 2 as shown in Fig. 3. On top of the buried SiO₂ layer 2, is a Silicon on Insulator (SOI) layer which is patterned into regions forming the active area of each device shown as a FIN 3 in Figs. 1, 2 and 3. The Fin may be formed by standard lithography and etching operations well known in the art. Alternatively, a known side wall image transfer method may be used to form each Fin 3.

0023After the Fin has been formed, a sacrificial oxidation procedure which is well known in the art may be performed to remove any damage from the Fin etch procedure. The sacrificial oxide layer can be used as a screen to prevent channeling during well implantation, if well implantation is used to adjust the threshold voltage of the FinFET device.

0024The sacrificial oxide is next removed by a dry or wet etch chemistry. For example, a dilute hydrofluoric acid may be used to remove the sacrificial oxide. After the sacrificial oxide removal, a gate dielectric may be formed. The gate oxidation may be a thermal SiO₂, a nitrided SiO₂ or oxy nitride. The gate dielectric may be a high K material like TaO₅, HfO₂, or any other gate dielectric material.

0025The gate electrode material is next deposited on the entire wafer, after which a lithography and etching procedure may be carried out. The gate electrode is labeled electrode 4 in the diagram.

0026After gate formation, a reoxidation operation, which is well known in the art, may be used to improve gate dielectric characteristics. The gate re-ox may also be omitted. At this point in the process flow, the source/drain extensions may be implanted, in another approach, an offset spacer may be used to create a distance between the gate edge and the implanted Fin region. A lithography mask may be used to block nFET regions from being implanted while allowing the pFET regions to be implanted as is common to conventional CMOS process technology. A similar operation can be used to implant the nFET regions while blocking the pFET regions.

0027After the source drain extension regions have been formed, an extension anneal may be used to heal the damage created by the ion implantation. In an alternate approach, the anneal may be omitted. Next, the deep source drain spacers are fabricated by depositing a SiN film in the range of 100A to 1000A, after which a highly directional etch is performed to remove the SiN film from the horizontal surfaces while leaving the film on the vertical portions of the gate electrode.

0028A block mask and ion implantation are now used to form the source and drain regions for the nFET device region 30 and pFET device region 20 as is standard in CMOS process technology. A conventional rapid thermal annealing process is next used to activate the junctions formed by the implantation. After this, a conventional salicide process using CoSi₂, TiSi, NiSi or any other suicide known in the art is performed.

0029At this point in the process flow, the inventive steps and structures (Figs. 4-17) to improve device performance for the n-FinFET and p-FinFET devices begin. First, as shown in Fig. 4 and Fig. 5, an SiO₂ liner layer (film) 5 is disposed (e.g., deposited) by, e.g., a low temperature deposition technique. The film thickness is in the range of 25-300A and the film deposition temperature is in the range of 200-750°C. The film may be deposited by any one of a variety of different known techniques including, but not limited to, sputter deposition, Plasma Enhanced Chemical Vapor Deposition (PECND), Rapid Thermal Chemical Vapor Deposition (RTCND), or a standard Chemical Napor Deposition (CVD) technique. The purpose of this SiO₂ film 5 is to act as an etch stop for a second film which is to be deposited next. Thus, the liner or etchstop layer 5 need not be SiO₂, but is any material that can provide adequate etch stop capability for the next film that is deposited directly on top of the liner material layer 5.

0030After deposition of the liner or etch stop layer 5, a compressive film 6, as shown in Fig. 6 and Fig. 7, is deposited over the entire wafer. In a preferred embodiment, the compressive film 6 is a SiN film deposited, e.g., by PECVD. The film 6 may also be deposited at an elevated power, in the range 400W to 1500W, to create more compression in the film. The film may be deposited using a low deposition rate and temperature range to make the film compressive. Ideally, the compression in the film is in the range of -300MPa to -3000MPa and the film thickness should be in the range of 200A to 2000A. Preferred deposition parameters are as follows: process temperature of 480°C, pressure of 5.75 torr, spacing between wafer and electrode of 395 mils, flow 3000 seem of a 2% diluted SiH₄ gas, 15 seem NH₃ gas and 1060 seem N gas using an RF power of 900 Watts. This process yields a deposition rate of approximately 15.95A/s and a film stress of about (±10%) -1400 MPa.

0031After the compressive film 6 has been applied to the wafer, a block mask 7, as shown in Fig. 8 and Fig. 9, is used to mask to pFET regions of the wafer. This block mask may be formed by conventional lithographic techniques known in the art. The mask is formed by a conventional lithography procedure in which a photosensitive material is coated over the wafer surface and exposed through a mask. The photosensitive material is then developed, leaving resist images or features blocking the pFET areas on the wafer.

0032After formation of the block mask 7, the compressive film 6 is removed by a known wet etching or dry etching technique that is capable of removing the compressive film selectively with respect to the block mask material. A plasma consisting of CH₂F₂ is an example of a dry etch chemistry that may be used for this purpose if the compressive film is

0033SiN. After the compressive film has been removed from the nFET regions on the wafer, an intermediate structure appears as shown in Fig. 10 and Fig. 11.

0034At this point in the inventive process flow, the block mask 7 is removed from the wafer using a solvent or O₂ plasma process known in the art to remove resist or organic materials. Next, a second liner or etchstop material 8 is deposited on the entire wafer as shown in Figs. 12 and 13. The second liner layer 8 has at least similar properties as the first liner previously described. That is, the liner is to be used as an etch stop for a subsequent film etching process.

0035Next, a tensile film 9 is deposited over the entire wafer as shown in Fig. 14 and Fig. 15.

0036The tensile film is, e.g., SiN and is deposited by, e.g., CVD, PECVD, RTCVD or any other deposition technique capable of depositing a highly tensile film. The film thickness should be in the range of 200A to 2000A, and the stress should be in the range +200MPa to +2000MPa or more tensile. Preferred deposition parameters are: process temperature of 480° C, pressure of 6.25 torr, spacing between wafer and electrode of

0037490 mils, flow 3000 seem of a 2% dilute SiH₄ gas, 15 seem NH₃ gas and 1060 seem N gas using an RF power of 340 Watts. This process yields a deposition rate of approximately 23A/s and a film stress of about 500 MPa.

0038At this point in the inventive process flow, a block mask 10 is patterned over the nFET regions of the wafer as shown in Fig. 16. The properties of this block mask are similar to the properties of the block mask previously described to block the pFET regions. After the block mask has been defined, a known wet or dry etching procedure is carried out to remove the tensile film 9 from the pFET regions. The etch should be selective to the liner etch stop material 8. In this way, the etch used to remove the tensile film from the nFET regions does not remove the compressive film which is present on the pFET regions. The block mask is next removed using a similar methodology used to remove the first block mask, to result in the final device structure 200, 300 shown in Fig. 17.

0039At this point in the inventive process flow, a thin film (not shown) in the range of 50A to 500A, with a low stress in the range of -100 MPa compressive to +100 MPa tensile may be applied to the wafer to serve as a barrier layer. The purpose of this thin film is to fill in any regions not covered by the compressive or tensile films. This optional film may be used to improve the suppression of contamination from penetrating into the Si and also help to improve the etch stop characteristics for the source-drain contact etch. After the operations described above are carried out, the CMOS process may be continued using standard process methodologies (not shown) well known in the art. More specifically, the processing that follows includes: deposition and planarization of a glass layer (i.e., BPSG, TEOS); etching of source/drain contacts, deposition of contact metallurgy and planarization; additional levels of insulating layers, vias, and wiring are then formed to complete the chip.

0040The presence of the stressed films on the longitudinal sidewalls of the Fins, overlying the gate conductors, results in a stress in the channel which is of the same type as the stress in the film (i.e., compressive/compressive, tensile/tensile). The stress in the source/drain regions on the longitudinal sidewalls of the Fins is of opposite type (i.e., compressive/tensile, tensile/compressive). To access the source/drain diffusions, the films on the top surface of each Fin may be removed without negating the effects of the film on the longitudinal sidewalls.

0041While there has been shown and described what is at present considered a preferred embodiment of the present invention, it will be apparent to those skilled in the art that various changes and modifications may be made therein without departing from the spirit and scope of the present invention.

0042Industrial Applicability

0043The present invention is applicable to microelectronic semiconductor devices.