The present disclosure relates to semiconductor device structures and more particularly, to a strained semiconductor device having an arch structure and method of making the same.
In the art, there is a concept known as silicon-on-nothing which refers to having a cavity under a Si channel. However, the devices described with respect to the silicon-on-nothing concept reflect bulk CMOS type devices.
Strained Si has the potential to enhance performance of CMOS devices, that is, by increasing drive currents. However, the approaches to realizing strained Si are often complicated. Such approaches include thick graded buffer layers, condensation, wafer bonding, etc.
What is needed is an improved method and apparatus for addressing the next generation beyond silicon on nothing, that is, to be able to include strained silicon.
The embodiments of the present disclosure are illustrated by way of example and not limited by the accompanying figures, in which like references indicate similar elements, and in which:
Skilled artisans appreciate that elements in the figures are illustrated for simplicity and clarity and have not necessarily been drawn to scale. For example, the dimensions of some of the elements in the figures may be exaggerated relative to other elements to help improve an understanding of the embodiments of the present disclosure.
According to one embodiment, a method of forming a strained semiconductor device includes forming a local strain inducing structure of a first material and forming a single crystalline structure of a second material over the local strain inducing structure. The single crystalline structure is subject to mechanical strain by an arched surface of the strain inducing structure.
In one approach of the present disclosure, strained Si can be realized in a simple epitaxial process that allows the potential to realize a surround gate MOSFET structure, as well as, accurately tune the level of strain in the device. Additionally, the structure according to the embodiments of the present disclosure has the possibility to be integrated in a back-end process for potential 3D integration.
In one embodiment, a silicon substrate is covered with oxide and then etched to form openings (or holes) in the oxide to expose the underlying Si substrate. The openings serve as an initial template for locations where strain inducing Ge or SiGe alloy dots and subsequent strained induced MOSFET devices will be fabricated. Into the openings is deposited SiGe, where the Ge concentration can vary from 0 to 100%. In one embodiment, the SiGe layer is deposited by being grown in a selective manner, although formation of the SiGe layer is not limited to selective growth alone. For example, non-selective growth can be combined with chemical mechanical polishing (CMP) to achieve the same desired structure at this point in the process. The SiGe layer that is deposited into the holes can be amorphous, poly-crystalline or single crystalline. In a preferred embodiment, the SiGe is single crystalline.
A pure Ge or SiGe dot is then grown selectively such that the nucleation site for the Ge or SiGe dot is the SiGe filled hole (i.e., a point location). The amount of Ge in the SiGe underlying layer will determine the size of the Ge or SiGe dot that is subsequently grown. Dots can be grown in a dome shape or a hut shape and each shape may have desirable properties. Subsequently, the Ge or SiGe dot is overgrown by a thin Si layer which will eventually form the channel region of the final device. In this approach, the thin Si layer that is grown over the Ge or SiGe dot will have strain induced in the layer due to the underlying Ge or SiGe dot. It is this strain that will be utilized to fabricate a strained Si type field effect transistor (FET). Subsequent to the thin Si layer over the Ge or SiGe dot, the Ge or SiGe dot can be removed selectively, for example, using a peroxide/HF chemistry, to create a cavity below the grown Si channel region. The Si channel region maintains its convex shape in order to retain strain in the channel. CMOS fabrication can continue to form source/drain and gate on the resulting arch of strained Si. Additionally, the cavity where the Ge or SiGe dot was initially grown can be backfilled with a material so as to induce additional strain in the Si channel or with a material that can serve as a backgate to realize a double gate or surround gate MOSFET device.
Accordingly, the embodiments of the present disclosure provide an alternate path to realizing a strained Si MOSFET device as compared to traditional approaches such as graded buffer layers. Additionally, the level of strain in the Si channel can be adjusted more readily than traditional approaches so that devices with differing levels of strain can be achieved on the same wafer. This approach also identifies a path to realize a surround gate MOSFET for true volume control of the channel region. Additionally, the channel thickness is defined epitaxially, so atomically smooth top and bottom interfaces are achievable, which is a benefit over competing device structures such as FinFET which are defined by dry etching. Furthermore, the embodiments of the present disclosure can be implemented for applications of high performance CMOS or low power CMOS.
Referring now to the drawings,
Substrate 12 can include a bulk semiconductor substrate, a semiconductor on insulator substrate (SOI), or any other substrate suitable for a particular strained semiconductor device arch structure application. For example, substrate 12 can include a silicon substrate and dielectric layer 14 can include an oxide. Other substrate and dielectric layer combinations are also possible. For example, substrate 12 may also include a gallium arsenide (GaAs) substrate, silicon germanium (SiGe) substrate, or other suitable semiconductor substrate.
Window 16 provides a point location within dielectric layer 14 for subsequent deposition of a seed layer 18. In
In
In
In one embodiment, the semiconductor material 22 includes amorphous silicon (Si). Accordingly, the semiconductor material 22 grown in the region indicated by reference numeral 24 (i.e., outside the region of the strain inducing structure 20) includes poly-silicon. The semiconductor material 22 in region 24 forms poly-silicon as a result of the layer being grown on a dielectric. Whereas, the semiconductor material 22 grown in the region indicated by reference numeral 26, overlying the strain inducing semiconductor structure 20, includes single crystal silicon.
In
Subsequent to formation of void 28, as illustrated in
Subsequent to formation of the control gate dielectric 30, the method proceeds with deposition of a control electrode material 38, as shown in
Subsequent to the gate electrode implant, the patterned photo resist 42 is removed, as shown in
Following formation of the control electrode 50, a dielectric liner 52 is deposited, the dielectric liner for insulating, protecting, and/or passivating the corresponding underlying layers. Dielectric liner 52 includes any suitable dielectric, for example, a silicon oxide or silicon nitride. Subsequent to forming dielectric liner 52, extension implants are done for forming extension regions 53 and 55 within region 26 which corresponds to the single crystal portion of semiconductor material 22. Sidewall spacers 54 are then formed adjacent to the sidewalls of control electrode 50, with dielectric liner 52 in between the sidewall spacers and the control electrode, as shown in
Various additional processing steps are carried out to form the strained semiconductor device 10 as shown in
Furthermore, refilling the modified opening 36 with material 82 can occur with the conformal deposition of a control gate electrode material 38, as shown in
Referring now to
Referring now to
Turning now to
Accordingly, in one embodiment, the forming of the local strain-inducing structure at the point location within the dielectric layer can comprise: forming an opening in the dielectric layer at a first location corresponding to the point location, wherein the opening exposes a portion of the underlying layer of semiconductor material; and forming the local strain-inducing structure (i) over the exposed portion of the underlying layer of semiconductor material or (ii) over the exposed portion of the underlying layer of semiconductor material and a portion of the dielectric layer proximate the opening in the dielectric layer. In addition, in an instance wherein the underlying layer of semiconductor material comprises a poly crystalline layer, then the method further comprises crystallizing the exposed portion of the underlying layer of semiconductor material prior to forming the local strain-inducing structure. Crystallizing the exposed portion of the underlying layer of semiconductor material can comprise the use of laser re-crystallization to form a local region of single crystal semiconductor material.
According to another embodiment, a method of forming a semiconductor device includes forming a local strain-inducing structure of a first semiconductor material at a point location within a dielectric layer. The local strain-inducing structure has a prescribed geometry with a surface disposed above a surface of the dielectric layer. A second semiconductor material is formed over the dielectric layer and the local strain inducing structure, wherein formation of a first portion of the second semiconductor material over the dielectric layer provides a poly crystalline structure of the second semiconductor material and wherein formation of a second portion of the second semiconductor material over the local strain-inducing structure provides a single crystalline structure of the second semiconductor material subject to mechanical strain by the surface of the local strain-inducing structure. The single crystalline structure serves as a strained semiconductor layer of the semiconductor device.
As discussed herein, in one embodiment, forming the local stress-inducing structure comprises forming a nucleation site at the point location and selectively growing the first semiconductor material at the nucleation site. In another embodiment, in forming a plurality of the semiconductor devices, the point location comprises a plurality of point locations that serve as an initial template for locations where local strain-inducing structures and subsequent strained semiconductor layers of the plurality of the semiconductor devices will be formed.
In yet another embodiment, the local strain-inducing structure comprises a seed layer portion of a third semiconductor material and a strain-inducing portion of the first semiconductor material. The seed layer portion is disposed (i) within an opening of the dielectric layer, or (ii) below an opening of the dielectric layer. In addition, the strain-inducing portion is disposed overlying (i) the seed layer portion and (ii) a portion of the dielectric layer proximate the opening in the dielectric layer. Furthermore, the seed layer portion comprises one of amorphous, poly-cyrstalline or single crystalline semiconductor material. Moreover, in one embodiment, the first semiconductor material can comprise Ge or SiGe, the second semiconductor material can comprise Si, and the third semiconductor material can comprise Ge or SiGe.
In a further embodiment, the seed layer portion comprises SiGe having a graded Ge concentration that varies within a range from zero to one-hundred percent (0-100%). In addition, the method further comprises selecting an amount of Ge in the SiGe seed layer portion to provide a desired size of the first semiconductor material at the point location. The size of the first semiconductor material formed over the seed layer portion is determined by the amount of Ge in the SiGe seed layer portion.
In addition, a method of forming a semiconductor device according to another embodiment includes forming a local strain inducing arch structure of a first material and forming a single crystalline region of a second material over the local strain inducing arch structure such that the single crystalline region is subject to mechanical strain by a surface of the strain inducing arch structure. In one embodiment, the first material is selected to have a first lattice constant and the second material is selected to have a second lattice constant different from the first lattice constant. As a result, in addition to the mechanical strain, the second material is further subject to lattice strain when formed over the first material. In another embodiment, the first and second materials are selected such that the first and second lattice constants have a lattice constant mismatch not substantially greater than four percent (4%).
In one embodiment, the single crystalline structure is formed so that the surface of the strain inducing structure is an arched surface. The second material is formed on the arched surface of the first material. Furthermore, the arched surface of the first material is formed to have a substantially double-curved surface having first and second orthogonal curvatures, the first and second curvatures being substantially equal and curving towards a dielectric layer or substrate underlying the semiconductor device.
In another embodiment, the strained semiconductor device is a transistor. The single crystalline structure includes a channel of the transistor. The arched surface of the first material is formed to have a substantially double-curved surface having first and second orthogonal curvatures. The first curvature is substantially greater than the second curvature and curving towards a dielectric layer or substrate underlying the semiconductor device. The second curvature is orthogonal to channel current when the device is operational. In addition, the channel is arched over the arched surface of the strain inducing structure.
In yet another embodiment, forming of the single crystalline structure includes growing the single crystalline structure to a thickness selected to facilitate fully depleted operation of the semiconductor device. The single crystalline structure is an active layer of the strained semiconductor device. In addition, a first control electrode structure can be formed over the active layer, the first control electrode structure for controlling current through the active layer when the device is operational. Still further, the method includes replacing, after removing the first semiconductor material and creating a void, the first semiconductor material with a control electrode material to provide a second control electrode structure for controlling current through the active layer when the device is operational. In one embodiment, the first and second control electrode structures are electrically coupled to provide a surround control electrode for controlling current through the active layer when the device is operational. In another embodiment, the first and second control electrode structures provide independent bias controls to independently control current through the active layer when the device is operational.
Furthermore, in yet another embodiment, a transistor includes first and second current handling electrodes and a channel coupled to each of the first and second current handling electrodes, the channel having a non-linear geometry. A gate is disposed proximate to the channel for controlling current flow through the channel between the first and second current handling electrodes. In one embodiment, the channel arches at a portion of the channel electrically between the first and second current handling electrodes. In addition, the channel has a thickness capable of fully depleted operation.
Yet still further, in another embodiment, the first and second current handling electrodes are disposed within a first plane, and the channel comprises a first portion disposed in the first plane proximate to the first current handling electrode, a second portion disposed in the first plane proximate to the second current handling electrode, and a third portion in a second plane parallel to and overlying the first plane. The gate is disposed over the channel and the channel has at least one portion with a curvature characterized by a first derivative which is positive with regard to a reference direction from the gate to the channel. In another embodiment, the transistor comprises multiple gates, with one gate above and one gate below the channel.
Further as disclosed herein, a semiconductor device includes a mechanically strained channel, wherein the channel comprises of a single crystalline structure of a strained semiconductor layer having a non-linear geometry, the non-linear geometry including a portion of an arch shape. The semiconductor device further includes a dielectric layer, wherein a first portion of the channel is disposed overlying a point location within the dielectric layer and a second portion of the channel is disposed overlying a portion of the dielectric layer proximate to and outside of the point location. In addition, a gate is disposed proximate to the channel for controlling current flow through the channel between first and second current handling electrodes that are coupled to the channel.
In the foregoing specification, the disclosure has been described with reference to various embodiments. However, one of ordinary skill in the art appreciates that various modifications and changes can be made without departing from the scope of the present embodiments as set forth in the claims below. Accordingly, the specification and figures are to be regarded in an illustrative rather than a restrictive sense, and all such modifications are intended to be included within the scope of the present embodiments. For example, the semiconductor device can include one or more of a transistor, a diode, an optical device, a light emitting diode, or a laser. An integrated circuit can also be formed using one or more of the methods according to the embodiments herein.
Benefits, other advantages, and solutions to problems have been described above with regard to specific embodiments. However, the benefits, advantages, solutions to problems, and any element(s) that may cause any benefit, advantage, or solution to occur or become more pronounced are not to be construed as a critical, required, or essential feature or element of any or all the claims. As used herein, the term “comprises,” “comprising,” or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements by may include other elements not expressly listed or inherent to such process, method, article, or apparatus.