Semiconductor device with trench-like feed-throughs

Abstract

A semiconductor device (e.g., a flip chip) includes a substrate layer that is separated from a drain contact by an intervening layer. Trench-like feed-through elements that pass through the intervening layer are used to electrically connect the drain contact and the substrate layer when the device is operated.

Description

FIELD OF THE INVENTION

Embodiments in accordance with the present invention generally pertain to semiconductor devices.

BACKGROUND

A “flip chip” is a semiconductor device that includes a pattern of solder balls arrayed on one of the chip's surfaces. During fabrication, the solder balls are formed on the top surface of the chip. The chip can then be readily mounted onto, for example, a circuit board by flipping the chip so that the top surface faces downward and the solder balls are aligned with corresponding pads on the circuit board.

In a non-flip chip semiconductor device, the gate and source contacts are on one surface (e.g., the top surface) of the chip while the drain contact is on the opposite surface (e.g., the bottom surface) of the chip. In a flip chip, the gate, source, and drain contacts are on the same surface of the chip. To form a circuit between the source and drain in a conventional flip chip, a “diffusion sinker” or “deep sinker” 110 is formed in the epitaxial layer 150 between the drain 120 and the substrate layer 130 as shown in FIG. 1. In operation, current will flow from the source 140 to, and through, the substrate and then to the drain in a known manner.

The sinker is formed using an isotropic diffusion process in which a dopant material is diffused through the epitaxial layer until the sinker is in contact with the substrate. However, as the sinker diffuses downward (in the y-direction), it also diffuses both along the surface of the structure (in the x-direction) and into the higher resistivity epitaxial layer. Thus, the surface area of the sinker is relatively large, increasing the size of each cell and thereby decreasing cell density. Also, the resistance of the epitaxial layer increases as the size of the sinker increases, thereby increasing the device's on-resistance (Rdson). Furthermore, the isotropic diffusion process is relatively slow, which increases the time needed to manufacture the flip chips and decreases throughput, thereby increasing costs.

SUMMARY

Embodiments according to the present invention provide solutions to the problems associated with sinkers. In one embodiment, relatively deep trenches filled with a conductive material such as tungsten are used to connect the drain contact to the substrate. Compared to a conventional sinker, the trenches—also referred to herein as “feed-throughs”—occupy less surface area and result in lower parasitic resistance in the epitaxial layer. There are also efficiencies associated with the fabrication of the feed-throughs, thus increasing throughput and reducing costs. These and other objects and advantages of the present invention will be recognized by one skilled in the art after having read the following detailed description, which are illustrated in the various drawing figures.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention. Like numbers denote like elements throughout the drawings and specification.

FIG. 1 is a cross-sectional view showing elements of a conventional semiconductor device.

FIG. 2 is a cross-sectional view showing elements of a semiconductor device according to embodiments of the present invention.

FIG. 3A illustrates a top-down view of a portion of a semiconductor device in one embodiment according to the present invention.

FIG. 3B illustrates a portion of a feed-through array in an embodiment according to the present invention.

FIG. 4 illustrates an example of an arrangement of feed-throughs in an embodiment according to the present invention.

FIG. 5 is a cross-sectional view of feed-throughs showing selected dimensions in one embodiment according to the present invention.

FIG. 6 is a flowchart of a process that is used in the fabrication of a semiconductor device according to embodiments of the present invention.

DETAILED DESCRIPTION

In the following detailed description of the present invention, numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, it will be recognized by one skilled in the art that the present invention may be practiced without these specific details or with equivalents thereof. In other instances, well-known methods, procedures, components, and circuits have not been described in detail as not to unnecessarily obscure aspects of the present invention.

Some portions of the detailed descriptions that follow are presented in terms of procedures, logic blocks, processing, and other symbolic representations of operations for fabricating semiconductor devices. These descriptions and representations are the means used by those skilled in the art of semiconductor device fabrication to most effectively convey the substance of their work to others skilled in the art. In the present application, a procedure, logic block, process, or the like, is conceived to be a self-consistent sequence of steps or instructions leading to a desired result. The steps are those requiring physical manipulations of physical quantities. It should be borne in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless specifically stated otherwise as apparent from the following discussions, it is appreciated that throughout the present application, discussions utilizing terms such as “forming,” “performing,” “producing,” “depositing,” “etching” or the like, refer to actions and processes (e.g., flowchart 600 of FIG. 6) of semiconductor device fabrication.

The figures are not drawn to scale, and only portions of the structures, as well as the various layers that form those structures, may be shown in the figures. Furthermore, fabrication processes and steps may be performed along with the processes and steps discussed herein; that is, there may be a number of process steps before, in between and/or after the steps shown and described herein. Importantly, embodiments in accordance with the present invention can be implemented in conjunction with these other (perhaps conventional) structures, processes and steps without significantly perturbing them. Generally speaking, embodiments in accordance with the present invention can replace portions of a conventional device or process without significantly affecting peripheral structures, processes and steps.

As used herein, the letter “n” refers to an n-type dopant and the letter “p” refers to a p-type dopant. A plus sign “+” or a minus sign “−” is used to represent, respectively, a relatively high or relatively low concentration of the dopant.

The term “channel” is used herein in the accepted manner. That is, current moves within a field effect transistor (FET) in a channel, from the source connection to the drain connection. A channel can be made of either n-type or p-type semiconductor material; accordingly, a FET is specified as either an n-channel or a p-channel device.

FIGS. 1-6 are discussed in the context of a p-channel device, specifically a p-channel power MOSFET (metal oxide semiconductor FET) embodied as a flip chip; however, embodiments according to the present invention are not so limited. That is, the features described herein can be utilized in an n-channel device. The discussion of FIGS. 1-6 can be readily mapped to an n-channel device by substituting n-type dopant and materials for corresponding p-type dopant and materials, and vice versa.

FIG. 2 is a cross-sectional view showing a portion of a semiconductor device 200 (e.g., a flip chip) according to an embodiment of the present invention. The device 200 includes a conductive (e.g., p+ or p++) substrate layer 202. In one embodiment, there is a back metal layer 204 adjacent to the substrate layer 202. The device also includes one or more additional layers exemplified by epitaxial layer 206, which may also be referred to herein as an intervening layer. The epitaxial layer 206 may include additional structures, layers or regions not shown in FIG. 2. For example, the epitaxial layer 206 may include isolation layers and regions; junction and channel (inversion) layers and regions; body regions, etc. These structures, layers and regions are known in the art but are not illustrated here for simplicity.

A number of trench or trench-like polysilicon gates 210 are formed in the epitaxial layer 206; the gates 210 extend only partially into the epitaxial layer. Isolation caps (e.g., silicon dioxide, SiO₂) 214 insulate the gates 210 from an overlying source metal 212 (e.g., aluminum) patterned on the epitaxial layer 206. Between the gates 210 are a number of trench or trench-like source contacts 220; the contacts 220 extend only partially into the epitaxial layer 206. Drain metal 216 (e.g., aluminum) is also patterned on the epitaxial layer 206 as shown in FIG. 2, and is isolated from the source metal 212 by an isolation region (e.g., SiO₂) 218. In one embodiment, the source metal 212 and the drain metal 216 are connected to respective solder balls (not shown in FIG. 2; refer to FIG. 3A).

Continuing with reference to FIG. 2, a number of trench or trench-like elements 230, also referred to herein as feed-throughs, are formed in the epitaxial layer 206. The feed-throughs 230 extend completely through the epitaxial layer 206, making contact with and perhaps extending into the substrate layer 202. The feed-throughs 230 provide an electrical connection between the drain metal 216 and the substrate layer 202.

In comparison to a conventional sinker, both the area and the volume consumed by the feed-throughs 230 are substantially reduced. Thus, the size of each cell can be reduced, thereby increasing cell density and/or reducing device size, an important consideration when the flip chip is used in mobile/hand-held devices such as cell phones and smart phones. Also, relative to a sinker, the feed-throughs 230 reduce the resistance of the epitaxial layer 206, thereby decreasing the on-resistance (Rdson) of the device 200. Furthermore, the feed-throughs 230 can be fabricated more quickly than a sinker, decreasing the time needed to manufacture the flip chips and increasing throughput, thereby decreasing costs.

In one embodiment, the feed-throughs 230 and the source contacts 220 are made of the same filler material. In one such embodiment, the feed-throughs 230 and the source contacts 220 are made of tungsten. Tungsten is chosen because its thermal coefficient is a better match with silicon than that of doped silicon, and because it has intrinsically lower resistance than doped silicon. Using the same material to fill the feed-through trenches and the source contact trenches facilitates fabrication, as will be described further below.

In one embodiment, the trenches for the feed-throughs 230 and the trenches for the source contacts 220 are each lined with the same liner material prior to deposition of the filler material, to prevent the filler material from contacting the surrounding silicon. In one such embodiment, the trenches are lined with a conformal coating of titanium (Ti) and Ti-nitride (TiN). In one embodiment, the Ti coating has a thickness of approximately 600 Angstroms, and the TiN coating has a thickness of approximately 200 Angstroms. Using the same material to line the feed-through trenches and the source contact trenches facilitates fabrication, as will be described further below.

The device 200 operates in a known manner, except that the feed-throughs 230 provide the functionality previously provided by a conventional sinker. By applying an electrical potential to a gate 210, a circuit will be completed, with current flowing from the source metal 212 through the source contacts 220 into structures (not shown) in the epitaxial layer 206, through the substrate layer 202, and then to the drain metal 216 via the feed-throughs 230.

FIG. 3A illustrates a top-down (layout) view of a portion of the semiconductor device 200 (e.g., a flip chip) according to one embodiment of the present invention. In the example of FIG. 3, two solder balls 302 and 304 are in contact with the source metal 212, a solder ball 306 is in contact with the gate metal 316, and a solder ball 308 is in contact with the drain metal 216. Underlying the drain metal 216 is an array 310 of feed-throughs 230, shown in cross-section in FIG. 2. Thus, the drain metal 216 is formed over and connected with multiple feed-throughs 230, and the solder ball 308 is formed on the drain metal 216.

FIG. 3B represents a portion of the feed-through array 310. There can be many feed-throughs 230 situated within (underneath) the region of drain metal 216 and consequently under (and connected to) the solder ball 308. Because many feed-throughs 230 are included in the array 310, a large current can flow through the feed-throughs without causing an electromigration problem. In particular, the dimensions of each feed-through in the array are chosen so that the electromigration threshold is high.

Significantly, with reference back to FIG. 3A, the surface area 320 that encompasses the source metal 212, gate metal 316, and drain metal 216 corresponds to the amount of surface area consumed by a conventional diffusion or deep sinker. In other words, the amount of surface area needed to accommodate the array 310 of feed-throughs 230 is substantially less than the area consumed by a sinker. Thus, as alluded to above, the use of the feed-throughs 230 in place of a sinker frees up real estate in the chip that can be used for additional structures such as those shown in FIG. 3A. By using feed-throughs instead of sinkers, a device of a given size can include more cells; conversely, for a given number of cells, the size of a power MOSFET flip chip can be reduced.

FIG. 4 shows an example of an arrangement of feed-throughs 230 in which the feed-throughs are concentrated under only a portion of the region of drain metal 216. Specifically, in the example of FIG. 4, the feed-throughs 230 are concentrated toward the direction from which the current is flowing—that is, toward the source region. Consequently, feed-throughs are not included in the area 410 of the drain region, further reducing the resistance of the epitaxial layer 206 (FIG. 2), and further reducing the amount of surface area utilized for the feed-throughs. Thus, the area 410 can be used for other structures, for example, to increase cell density.

FIG. 5 is a cross-sectional view of tungsten feed-throughs 510 and 512 showing selected dimensions. In the example of FIG. 5, the feed-throughs 510 and 512 each have a width W of about 0.9 microns (more precisely, 0.95 microns); a depth D1 of about 8.7 microns (more precisely, 8.73 microns) measured at the shortest (shallowest) point (along the centerline) and a depth D2 of about 9.3 microns (more precisely, 9.31 microns) measured at the longest (deepest) point; and a distance (pitch P) between adjacent feed-throughs of about 1.7 microns (more precisely, 1.74 microns). With these dimensions, the resistance of an array of tungsten feed-throughs is about one-half of the resistance associated with a conventional sinker.

FIG. 6 illustrates a flowchart 600 of one embodiment of a process that is used in the fabrication of semiconductor devices such as the device 200 of FIG. 2. Although specific steps are disclosed in FIG. 6, such steps are only examples. That is, embodiments according to the present invention are well suited to performing various other steps or variations of the steps recited in FIG. 6. As noted above, the epitaxial layer 206 (FIG. 2) may include structures, layers, and regions not shown in the figures—those structures, layers, and regions can be formed before, during (e.g., as part of), and/or after the steps in flowchart 600 are performed.

In block 602 of FIG. 6, with reference also to FIG. 2, a first mask, with openings corresponding to the locations of the source contacts 220, is applied to the upper surface of the epitaxial layer 206. The trenches for the source contacts are etched through those openings, and then the first mask is removed. The trenches for the source contacts extend partially into but not completely through the epitaxial layer 206.

In block 604, a second mask, with openings corresponding to the locations of the feed-throughs 230, is applied to the upper surface of the epitaxial layer 206. The trenches for the feed-throughs are etched through those openings, and then the second mask is removed. The trenches for the feed-throughs extend completely through the epitaxial layer 206 and into the substrate layer 202.

In block 606, a conformal coating is applied to the surfaces of the source contact trenches and to the surfaces of the feed-through trenches. In one embodiment, the conformal coating is applied to the surfaces of the source contact trenches and feed-through trenches in the same process step. In one such embodiment, the same material (e.g., Ti and TiN) is used to coat the surfaces of the source contact trenches and the feed-through trenches.

In block 608, a filler material (e.g., tungsten) is deposited into the source contact trenches and into the feed-through trenches. In one embodiment, the filler material is deposited into the source contact trenches and the feed-through trenches in the same process step.

In block 610, the filler material is etched back so that is flush, or nearly flush, with the upper surface of the epitaxial layer 206.

In block 612, other structures can be formed. For example, aluminum metal can be deposited and patterned to form source, drain, and gate regions. Subsequently, solder balls are positioned on the source, drain, and gate regions.

In comparison to a conventional process, the steps of i) applying the mask associated with the feed-through trenches and ii) forming the feed-through trenches are added, but the conventional steps of i) applying a mask to form a sinker and ii) forming the sinker are eliminated. Because, in one embodiment, the feed-through trenches are coated and filled at the same time that the source contact trenches are coated and filled, those steps do not lengthen the fabrication process by a significant amount, if at all. The feed-throughs 230 take less time to fabricate relative to the time needed to form a conventional sinker. Consequently, in total, the introduction of the feed-throughs shortens the length of the fabrication process, increasing throughput and decreasing cost.

Also, as mentioned above, the feed-throughs occupy less surface area and result in lower parasitic resistance in the epitaxial layer in comparison to conventional sinkers. Consequently, the size of each cell can be decreased, thereby increasing cell density. Moreover, the device's on-resistance is reduced.

The foregoing descriptions of specific embodiments of the present invention have been presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the invention to the precise forms disclosed, and many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and its practical application, to thereby enable others skilled in the art to best utilize the invention and various embodiments with various modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims appended hereto and their equivalents.

Claims

1. A semiconductor device comprising: an electrically conductive substrate layer;a drain contact, wherein said substrate layer is separated from said drain contact by an intervening layer;a plurality of gate trenches in a gate region under a layer of source metal and that extend into but not completely through said intervening layer, said gate trenches comprising a first gate trench and a second gate trench, each of said gate trenches filled with a first filler material;a plurality of source contact trenches in a source region under said layer of source metal and that that extend into but not completely through said intervening layer, each of said source contact trenches filled with a second filler material, said source contact trenches comprising a first source contact trench and a second source contact trench, wherein said first source contact trench is between said first gate trench and said second gate trench, and wherein said second gate trench is between said first source contact trench and said second source contact trench; anda plurality of feed-through trenches comprising more than one feed-through trench, each of said feed-through trenches in said plurality of feed-through trenches under a layer of drain metal in contact with said drain contact, wherein said feed-through trenches extend completely through said intervening layer to said substrate layer, each of said feed-through trenches filled with said second filler material in contact with said layer of drain metal, wherein said feed-through trenches are arrayed between said layer of drain metal and said substrate layer in only a portion of a drain region of said device that is under said layer of drain metal, wherein said feed-through trenches are not included outside said portion and wherein said feed-through trenches within said portion are concentrated toward said source region.

2. The semiconductor device of claim 1 wherein said first filler material comprises polysilicon and said second filler material comprises tungsten.

3. The semiconductor device of claim 1 wherein said feed-through trenches are each lined with a conformal coating that separates said second filler material from said intervening layer.

4. The semiconductor device of claim 3 wherein said conformal coating comprises material selected form the group consisting of: titanium and titanium nitride.

5. The semiconductor device of claim 3 wherein said conformal coating comprises a thickness of titanium of 60 nanometers (600 Angstroms) and a thickness of titanium-nitride of 20 nanometers (200 Angstroms).

6. The semiconductor device of claim 1 wherein said device comprises a flip chip having a plurality of solder balls formed on a surface thereof, wherein said drain contact is coupled to at least one of said solder balls.

7. The semiconductor device of claim 1 wherein each trench in said plurality of feed-through trenches has a depth of 8.7 microns at their shallowest points, a depth of 9.3 microns at their deepest points, a width of 0.9 microns, and a distance between adjacent feed-through trenches of 1.7 microns.

8. A flip chip semiconductor device comprising: a plurality of solder balls comprising a solder ball coupled to a drain contact on a first surface of said device;a first metal layer on a second surface of said device, said second surface opposite said first surface;a substrate layer adjacent said first metal layer, wherein said substrate layer is separated from said drain metal by an intervening layer;a plurality of gate trench elements in a gate region under a layer of source metal and that extend into but not completely through said intervening layer, said gate trench elements comprising a first gate trench element and a second gate trench element, each of said gate trench elements filled with a first filler material;a plurality of source contact trench elements in a source region under said layer of source metal and that extend into but not completely through said intervening layer, each of said source contact trench elements filed with a second filler material, said source contact trench elements comprising a first source contact trench element and a second source contact trench element, wherein said first source contact trench element is between said first gate trench element and said second gate trench element, and wherein said second gate trench element is between said first source contact trench element and said second source contact trench element, and wherein said source contact trench elements are in contact with a layer of source metal but said gate trench elements are isolated from said source metal; anda plurality of feed-through trench elements, said plurality comprising multiple feed-through trench elements, each of said feed-through trench elements in said plurality of feed-through trench elements under a layer of drain metal that is contact with said drain contact, wherein said feed-through trench elements extend completely through said intervening layer to said substrate layer, each of said feed-through trench elements filled with said second filler material in contact with said layer of drain metal, wherein said feed-through trench elements pass through said intervening layer and into said substrate layer, wherein said feed-through trench elements in said plurality of feed-through trench elements are arrayed between said layer of drain metal and said substrate layer in only a portion of a drain region of said device that is under said layer of drain metal, wherein said feed-through trench elements are not included outside said portion and wherein said feed-through trench elements within said portion are concentrated toward said source region, wherein in operation said device comprises a circuit from a source contact through said intervening layer, said substrate layer, and said feed-through trench elements to said drain contact.

9. The semiconductor device of claim 8 wherein said first filler material comprises polysilicon and said second filler material comprises tungsten.

10. The semiconductor device of claim 8 wherein said source contact trench elements and said feed-through trench elements are each lined with a conformal coating that separates said second filler material from said intervening layer.

11. The semiconductor device of claim 10 wherein said conformal coating comprises material selected from the group consisting of: titanium and titanium-nitride.

12. A method of fabricating a flip chip, said method comprising: forming a plurality of gate trenches in a surface of an epitaxial layer in a gate region of a structure, said gate trenches comprising a first gate trench and a second gate trench, said gate trenches extending partially into but not completely through said epitaxial layer, wherein a plurality of source contact trenches are also formed in said surface of said epitaxial layer in a source region of said structure, said source contact trenches extending partially into but not completely through said epitaxial layer, said source contact trenches comprising a first source contact trench and a second source contact trench, wherein said first source contact trench is between said first gate trench and said second gate trench, and wherein said second gate trench is between said first source contact trench and said second source contact trench;forming a plurality of feed-through trenches in said surface, said plurality of feed-through trenches comprising more than one feed-through trench, said feed-through trenches extending completely through said epitaxial layer and into a substrate layer adjoining a second surface of said epitaxial layer; anddepositing a first filler material into said gate trenches and depositing a second filler material into said source contact trenches and said feed-through trenches, wherein said source contact trenches are subsequently electrically coupled to a first solder ball used as a source contact and wherein said second filler in each of said feed-through trenches is subsequently contacted to a layer of drain metal that is subsequently contacted to a second solder ball used as a drain contact, wherein said plurality of feed-through trenches are arrayed between said layer of drain metal and said substrate layer in only a portion of a drain region that is under said layer of drain metal, said drain region under said drain contact, wherein said feed-through trenches are not included outside said portion and wherein said feed-through trenches within said portion are concentrated toward said source region.

13. The method of claim 12 wherein said first filler material comprises polysilicon and said second filler material comprises tungsten.

14. The method of claim 12 further comprising, after said source contact trenches and said feed-through trenches are formed and before said second filler material is deposited, depositing a conformal coating that lines said source contract trenches and that lines said feed-through trenches in a same process step.

15. The method of claim 14 wherein said conformal coating comprises material selected from the group consisting of: titanium and titanium-nitride.