The present invention relates to a semiconductor substrate material and to a method of fabricating the same. More specifically, the present invention relates to a strained semiconductor, e.g., Si-on-insulator (SSOI) substrate material and a robust method of fabricating the same that avoids wafer bonding.
In the semiconductor industry, there has been an increasing interest in enhancing performance of complementary metal oxide semiconductor (CMOS) devices by replacing conventional silicon-on-insulator (SOI) substrates with strained semiconductor-on-insulator (SSOI) substrates. The reason behind this interest is that SSOI substrates provide higher carrier (electrons/holes) mobility than a conventional SOI substrate. The strain in the SSOI substrates can either be compressive or tensile.
Conventional methods to fabricate SSOI substrates typically require a layer transfer process wherein a strained Si-containing layer located on a relaxed SiGe layer is transferred onto a handle wafer. In particular, the conventional process includes first creating a relaxed SiGe layer of a few microns in thickness on a surface of a Si-containing substrate. The relaxed SiGe layer typically has an in-plane lattice parameter that is larger than that of Si. Next, a Si-containing layer is grown on the relaxed SiGe layer. Because the SiGe layer has a larger in-plane lattice parameter as compared to Si, the Si-containing layer is under strain.
The structure, including the strained Si-containing layer located on a relaxed SiGe layer, is then bonded to a handle wafer, which includes an insulating layer, such as an oxide layer. The bonding occurs between the strained Si-containing layer and the insulator layer. The Si-containing substrate and the relaxed SiGe layer are then typically removed from the bonded structure to provide a strained Si-on-insulator substrate.
The conventional SSOI substrate preparation method described above is quite expensive and low-yielding because it combines two rather advanced substrate technologies, i.e., high-quality, thick SiGe/strain Si growth, and wafer bonding. Moreover, the conventional preparation method is unattractive for manufacturing a large volume of substrates.
In view of the above, a cost effective and manufacturable solution to fabricate SSOI substrates is required for future high-performance Si-containing CMOS products.
The present invention provides a cost-effective and manufacturable solution to produce SSOI substrates that avoids wafer bonding which is typically required in conventional technologies to produce SSOI substrate materials. In particular, the method of the present invention, which fabricates SSOI substrates, includes creating a buried porous layer underneath a strained semiconductor layer. The porous layer is then converted into a buried oxide layer by employing a high temperature oxidation/anneal step such that only a part of the strained semiconductor layer is consumed during processing.
The method provides a SSOI substrate that includes a strained semiconductor layer atop an oxide layer, the oxide layer is located on a relaxed semiconductor template. Unlike the conventional process described above, the strained semiconductor layer and the relaxed semiconductor layer have a commensurate, i.e., identical, crystal orientation. Moreover, the oxide layer that is formed by the inventive method is of ‘high-quality’ meaning that the oxide layer has a leakage of about 1 microAmp or less and a breakdown field of about 2 Megavolts/cm or greater.
In broad terms, the method of the present invention comprises the steps of:
providing a structure that comprises a substrate, a relaxed semiconductor layer on the substrate, a doped and relaxed semiconductor layer on the relaxed semiconductor layer, and a strained semiconductor layer on the doped and relaxed semiconductor layer, said relaxed semiconductor layer, said doped and relaxed semiconductor layer and said strained semiconductor layer have identical crystallographic orientations;
converting the doped and relaxed semiconductor layer underneath the strained semiconductor layer into a buried porous layer; and
annealing the structure including the buried porous layer to provide a strained semiconductor-on-insulator substrate, wherein during said annealing the buried porous layer is converted into a buried oxide layer.
In addition to the method described above, the present invention also relates to the SSOI substrate that is formed. Specifically, the SSOI substrate of the instant invention comprises:
a substrate;
a relaxed semiconductor layer on the substrate;
a high-quality buried oxide layer on the relaxed semiconductor layer; and
a strained semiconductor layer on the high-quality buried oxide layer, wherein the relaxed semiconductor layer and the strained semiconductor layer have identical crystallographic orientations.
The present invention, which provides a method of fabricating a SSOI substrate and the SSOI substrate produced by the method, will now be described in greater detail by referring to the drawings that accompany the present application. The drawings are provided for illustrative purposes only and are thus not drawn to scale. In the drawings, like and corresponding elements are referred to by like reference numerals.
The method of the present invention begins with first providing the structure 10 shown, for example, in
Examples of various epitaxial growth processes that can be employed in the present invention in fabricating layers 14, 16 and 18 on substrate 12 include, for example, rapid thermal chemical vapor deposition (RTCVD), low-energy plasma deposition (LEPD), ultra-high vacuum chemical vapor deposition (UHVCVD), atmospheric pressure chemical vapor deposition (APCVD) and molecular beam epitaxy (MBE).
The substrate 12 employed in the present invention may be comprised of any material or material layers including, for example, crystalline glass or metal, but preferably the substrate 12 is a crystalline semiconductor substrate. Examples of semiconductor substrates that can be employed as substrate 12 include, but are not limited to: Si, SiGe, SiC, SiGeC, GaAs, InAs, InP, and other III/IV or II/VI compound semiconductors. The term “semiconductor substrate” also includes preformed silicon-on-insulator (SOI) or SiGe-on-insulator (SGOI) substrates which may include any number of buried insulating (continuous, noncontinuous or a combination of continuous and non-continuous) regions therein. In one preferred embodiment, the substrate 12 is a Si-containing substrate. The substrate 12 may be undoped or it may be an electron rich or hole-rich substrate, i.e., doped substrates.
The relaxed semiconductor layer 14 is then epitaxially grown on a surface of the substrate 12 using one of the above mentioned processes. In the following description the relaxed semiconductor layer 14 is referred to a relaxed SiGe layer 14 since that semiconductor material represents a preferred material for layer 14. The term “SiGe alloy layer” denotes a SiGe layer that comprises up to 99 atomic percent Ge. More typically, the SiGe alloy layer comprises from about 1 to about 99 atomic percent Ge, with a Ge atomic percent from about 10 to about 50 atomic percent being more highly preferred.
The relaxed SiGe alloy layer 14 may be a single layer having a continuous distribution of Ge, or it may be a graded layer having a varying content of Ge included within different regions of the layer. As stated above, layer 14 is a relaxed layer having a measured degree of relaxation from about 10% or greater Typically, the surface region of the relaxed semiconductor layer 14 is metastable having a defect (stacking faults, pile-up and threading) density that is typically about 1E5 defects/cm3 or greater.
The relaxed semiconductor layer 14 may be doped or undoped. The type of dopant and concentration of dopant within the layer 14 is arbitrary and can be predetermined by a skilled artisan. When doped, relaxed layer 14 typically has a dopant concentration that is greater than 1E17 atoms/cm3. Doped layer 14 is formed by providing a dopant source with the Si source, or the Ge source, or both sources used during the epitaxial growth process.
The thickness of the relaxed semiconductor layer 14 may vary so long as a relaxed layer can be formed. The thickness of the relaxed semiconductor layer 14 is dependent on the Ge content of the layer. Typically, and for a relaxed semiconductor layer 14 having a Ge content of less than about 50 atomic %, layer 14 has a thickness from about 1 to about 5000 nm, with a thickness from about 1000 to about 3000 nm being more typical.
Although relaxed SiGe alloy templates are preferred, the present invention also contemplates the use of other semiconductor materials that can be formed in relaxed state.
Next, a doped and relaxed semiconductor layer 16 is formed on the relaxed semiconductor layer 14. The doped and relaxed semiconductor layer 16 may include p- or n-type dopants, with p-type dopants being highly preferred. P-type dopants include Ga, Al, B and BF2. The doped and relaxed semiconductor layer 16 may be a separate layer, as shown in
In accordance with the present invention, the doped and relaxed semiconductor layer 16 is a layer that is more heavily doped than the surrounding layers, i.e., layers 14 and 18. Typically, the doped and relaxed semiconductor layer 16 contains a p-type dopant concentration of about 1E19 atoms/cm3 or greater, with a p-type dopant concentration from about 1E20 to about 5E20 atoms/cm3 being more typical The doped and relaxed semiconductor layer 16 is formed using one of the above mentioned epitaxial growth processes in which the dopant source is included with the semiconductor source. The doped and relaxed semiconductor material 16 may have an in-plane lattice parameter that is either larger or smaller than that of virgin Si.
The doped and relaxed semiconductor layer 16 is a thin layer whose thickness will define the thickness of the buried oxide layer to be subsequently formed. Typically, the doped and relaxed semiconductor layer 16 has a thickness from about 1 to about 1000 nm, with a thickness from about 10 to about 200 nm being more typical.
After forming the doped and relaxed semiconductor layer 16, a strained semiconductor layer 18 is formed on top of the doped and relaxed semiconductor layer 16 using one of the above-mentioned epitaxial growth processes. The strained semiconductor layer 18 may be comprised of one of the semiconductor materials mentioned above in connection with layer 16. The strained semiconductor layer 18 and the doped and relaxed semiconductor 16 can thus be comprised of the same or different semiconductor material. The strained semiconductor 18 can have a tensile or compressive stress.
It is noted that the growth of layers 14, 16 and 18 may occur using the same or different epitaxial growth process. Moreover, it is also contemplated to form layers 14, 16 and 18 in the same reactor chamber without breaking vacuum.
The strained semiconductor layer 18 may be doped or undoped. When doped, the strained semiconductor layer 18 typically has a dopant concentration of about 1E15 atoms/cm3 or greater. The thickness of layer 18 is typically from about 5 to about 2000 nm, with a thickness from about 10 to about 500 nm being more typical.
In one embodiment of the present invention, the strained semiconductor layer 18 and the doped and relaxed semiconductor layer 16 are comprised of the same or different Si-containing semiconductor, with Si or SiGe being highly preferred.
In a highly preferred embodiment of the present invention, the strained semiconductor layer 18 and the relaxed semiconductor layer 14 are both doped layers having a dopant concentration of about 1E15 atoms/cm3 or greater, while the doped and relaxed semiconductor layer 16 is a p-doped layer having a dopant concentration of about 1E20 atoms/cm3 or greater.
In accordance with the present invention, layers 14, 16 and 18 have the same crystallographic orientation as substrate 12 since the various layers are formed by epitaxial growth. Hence, layers 14, 16 and 18 can have a (100), (110), (111) or any other crystallographic orientation.
Next, the structure shown in
The anodization process is performed by immersing the structure shown in
In general, the HF anodization converts the doped and relaxed semiconductor layer 16 into a porous semiconductor layer 20. The rate of formation and the nature of the porous semiconductor layer 20 so-formed (porosity and microstructure) is determined by both the material properties, i.e., doping type and concentration, as well as the reaction conditions of the anodization process itself (current density, bias, illumination and additives in the HF-containing solution). Generally, the porous semiconductor layer 20 formed in the present invention has a porosity of about 0.1% or higher.
The term “HF-containing solution” includes concentrated HF (49%), a mixture of HF and water, a mixture of HF and a monohydric alcohol such as methanol, ethanol, propanol, etc, or HF mixed with at least one surfactant. The amount of surfactant that is present in the HF solution is typically from about 1 to about 50%, based on 49% HF.
The anodization process, which converts the doped and relaxed semiconductor layer 16 into a porous semiconductor layer 20, is performed using a constant current source that operates at a current density from about 0.05 to about 50 milliAmps/cm2. A light source may be optionally used to illuminate the sample. More preferably, the anodization process of the present invention is employed using a constant current source operating at a current density from about 0.1 to about 5 milliAmps/cm2.
The anodization process is typically performed at room temperature or, a temperature that is elevated from room temperature may be used. Following the anodization process, the structure is typically rinsed with deionized water and dried. Anozidation typically occurs for a time period of less than about 10 minutes, with a time period of less than 1 minute being more typical.
The structure shown in
Note that an oxide layer 24 is formed atop layer 18 during the heating step. This surface oxide layer, i.e., oxide layer 24, is typically, but not always, removed from the structure after the heating step using a conventional wet etch process wherein a chemical etchant such as HF that has a high selectivity for removing oxide as compared to semiconductor is employed. The structure, without the surface oxide layer 24, is shown in
Note that when the oxide layer 24 is removed, the above processing steps can be repeated any number of times to provide a multilayered structure containing, from bottom to top, substrate/(relaxed semiconductor/buried oxide/strained semiconductor)x, wherein x is greater than 1. When x is 1, the structure shown in
In some embodiments of the present invention, multiple buried oxide layers can be obtained by forming continuous layers of materials 14, 16 and 18 on substrate 12 and then performing the electrolytic anodization process and annealing process of the present invention.
The surface oxide layer 24 formed after the heating step of the present invention has a variable thickness which may range from about 10 to about 1000 nm, with a thickness of from about 20 to about 500 nm being more highly preferred. Buried oxide layer 22 typically has the same thickness as previously described for the doped and relaxed semiconductor layer 16.
Specifically, the heating step of the present invention is an annealing step which is performed at a temperature that is greater than 400° C., preferably greater than 1100° C. A typical temperature range for the heating step of the present invention is from about 1200° to about 1320° C.
Moreover, the heating step of the present invention is carried out in an oxidizing ambient which includes at least one oxygen-containing gas such as O2, NO, N2O, ozone, air and other like oxygen-containing gases. The oxygen-containing gas may be admixed with each other (such as an admixture of O2 and NO), or the gas may be diluted with an inert gas such as He, Ar, N2, Xe, Kr, or Ne. When a diluted ambient is employed, the diluted ambient contains from about 0.1 to about 100% of oxygen-containing gas, the remainder, up to 100%, being inert gas.
The heating step may be carried out for a variable period of time that typically ranges from greater than 0 minutes to about 1800 minutes, with a time period from about 60 to about 600 minutes being more highly preferred. The heating step may be carried out at a single targeted temperature, or various ramp and soak cycles using various ramp rates and soak times can be employed.
The heating step is performed under an oxidizing ambient to achieve the presence of oxide layers, i.e., layers 22 and 24. Note that the porous semiconductor region reacts with diffused oxygen at an enhanced rate.
After heating, and subsequent removal of surface oxide layer 24, the structure can be subjected to a thermal process (i.e., baking step) that is capable of reducing the content of dopants present in the final structure. The baking step is typically performed in the presence of a hydrogen-containing ambient such as H2. Leaching of dopants from the structure typically occurs when this step is performed at a temperature that is greater than 800° C., with a temperature of greater than 1000° C. being more typical. This thermal step is optional and does not need to be performed in all instances. Leaching of dopants using the thermal treatment process can be performed for any desired period of time.
Typically, the thermal process, which leaches dopants from the structure, is performed for a time period from about 1 to about 60 minutes. As stated above, this baking step reduces the amount of dopant within the SSOI substrate. Although it can be used to reduce any dopant within the SSOI substrate, it is particularly employed to remove boron from the structure.
After performing the above processing steps, conventional CMOS process can be carried out to form one or more CMOS devices such as field effect transistors (FETs) atop the strained semiconductor layer. The CMOS processing is well known to those skilled in the art; therefore details concerning that processing are not needed herein.
The method of the present invention described above provides a SSOI substrate including a strained semiconductor layer 18 atop an oxide layer 22, the oxide layer 22 is located atop a relaxed semiconductor layer 14 which is located on a substrate 12. Unlike the conventional process described above, the strained semiconductor layer 18 and the relaxed semiconductor layer 14 have a commensurate, i.e., identical, crystal orientation. Moreover, the oxide layer 22 that is formed by the inventive method is of ‘high-quality’ meaning that the buried oxide layer 22 has a leakage of about 1 microAmp or less and a breakdown field of about 2 Megavolts or greater.
The embodiment depicted in
In yet another embodiment of the present invention, a patterned SSOI substrate, such as illustrated in
CMOS processing can also be performed on the patterned SSOI substrates as well.
While the present invention has been particularly shown and described with respect to preferred embodiments thereof, it will be understood by those skilled in the art that the foregoing and other changes in forms and details may be made without departing from the scope and the spirit of the present invention. It is therefore intended that the present invention not be limited to the exact forms and details described and illustrated, but fall within the scope of the appended claims.
This application is a divisional of U.S. application Ser. No. 10/883,887 filed Jul. 2, 2004.
Number | Date | Country | |
---|---|---|---|
Parent | 10883887 | Jul 2004 | US |
Child | 11620663 | Jan 2007 | US |