This description relates to semiconductor devices.
Semiconductor devices may be required in various types of high power scenarios, such as in electric vehicle or solar power scenarios. Such semiconductor devices may have various types of operational and/or manufacturing constraints. For example, in addition to providing high power levels, such semiconductor devices may be required to provide fast switching, low power loss, high levels of reliability (e.g., avoid overheating), and cost-effective manufacturing.
According to one general aspect, a semiconductor device may include a backside contact, a substrate adjacent to the backside contact, and an epitaxial field stop region formed on the substrate and having a graded doping profile that decreases with distance away from the substrate. The semiconductor device may include an epitaxial drift region adjacent to the epitaxial field stop region, and a frontside device formed on the epitaxial drift region.
According to another general aspect, a semiconductor device may include a backside contact and a substrate adjacent to the backside contact and having a doping of a first conductivity type. The semiconductor device may include an epitaxial field stop region formed on the substrate and having a graded doping profile of a second conductivity type that decreases with distance away from the substrate, an epitaxial drift region adjacent to the epitaxial field stop region and having a substantially uniform doping profile of the second conductivity type, and a frontside device formed on the epitaxial drift region.
According to another general aspect, a method of making a semiconductor device may include performing at least one epitaxial growth process on a substrate to obtain an epitaxial field stop region on the substrate and an epitaxial drift region on the epitaxial field stop region. The method may also include performing frontside processing to form at least one frontside device on the epitaxial drift region.
The details of one or more implementations are set forth in the accompanying drawings and the description below. Other features will be apparent from the description and drawings, and from the claims.
Many high power semiconductor devices utilize a drift region that is designed, for example, to facilitate a desired breakdown voltage of a device by blocking a high off-state electric field of the device, or otherwise managing off-state behavior of the device(s). Although providing these and other advantages, such drift regions also exhibit a number of disadvantages, as well.
For example, a drift region exhibits a specific on-resistance (Ron_sp) that inhibits desired on-behavior of corresponding devices. It is possible to lower Ron_sp by altering physical characteristics of the drift region (e.g., a size or doping profile thereof) or the device(s) (e.g., decreasing a device channel length) but doing so may lead directly to other design difficulties, such as an overall larger device, increased device capacitances, or reduction of the desired advantages of the drift region (such as breakdown voltage control).
To facilitate these and other device constraints, some devices include a field stop layer, e.g., adjacent to the drift region and distal from a device region. The field stop layer in such a device may be more heavily doped than a lightly-doped drift region, and may cause the off-state electric field of the device to fall off abruptly upon reaching the field stop layer.
In conventional field stop devices, however, it is difficult and expensive to attempt to form such field stop layers in a desired manner. For example, conventional field stop layers may be doped using ion implantation processes, which are typically high-energy, high-cost processes.
To facilitate ion implantation, thinned wafers may be used. However, using thinned wafers exacerbates manufacturing costs, because, e.g., more expensive wafer handling tools are required, and thinned wafers are more likely to be lost to manufacturing defects, such as warping and breaking.
Additionally, ion implantation is often performed using high energy proton implantation, which, as referenced above, is an expensive process, requiring specialized equipment and quantities of dedicated space within a manufacturing facility. As semiconductor manufacturing moves towards larger wafer sizes, e.g., from 200 mm to 300 mm, even larger, higher-energy implant machines are required, so that high energy proton implantation becomes even more expensive.
Further, even if such equipment is available, such high-energy ion implantation results in an irregular doping profile, e.g., with peaks and valleys of doping concentrations within the resulting field stop layer. Although it may be theoretically possible to diffuse such peaks and valleys through annealing processes, such annealing processes are not available for use in conventional manufacturing scenarios. For example, in conventional manufacturing processes, as referenced above, wafers have already been thinned, and frontside device processing has already been completed, by the time the high-energy proton implantation process is conducted.
As a result, the peaks and valleys of the field stop doping profiles of the types of conventional devices just referenced may reduce an efficacy of the provided field stop layer. For example, a switching speed of a resulting device may be negatively impacted, as discussed below in more detail, e.g., with respect to
The present disclosure describes an epitaxial field stop region that may be formed prior to, and in conjunction with, an epitaxial drift region, and prior to front-side device processing. That is, described implementations include such an epitaxial field stop region, and do not require use of an implanted substrate portion or layer to provide a field stop region.
The epitaxial field stop region may be formed with a graded doping profile, which facilitates smooth, prompt, and complete blocking of the off-state electric field, while maintaining fast switching speeds. The epitaxial field stop region may be formed prior to wafer thinning, and prior to frontside device processing, so that any desired or necessary annealing processes may be performed in conjunction with such frontside processing, and if needed to obtain a smooth gradation of the doping profile of the epitaxial field stop region.
Thus, the resulting field stop enables smooth and abrupt electric field blocking, as well as breakdown voltage (BV) control. Moreover, the described epitaxial field stop region provides improved switching performance. Further, related manufacturing processes may be performed in an efficient, cost-effective manner.
Additionally, proton implantation may still be performed. However, as described below, described techniques enable use of relatively low-energy, low-expense proton implantation techniques. Resulting implantations may further bolster a field stop effect of the epitaxial field stop layer, while also facilitating and improving an ohmic contact of the device.
In the example of
As
Thus, the semiconductor device of
In
As referenced above, and described in more detail, below, with respect to
In other examples, as described with respect to
The example embodiment of
Further in
Similarly, an epitaxial region 306 represents a more detailed example of the epitaxial region 106 of
As shown, the epitaxial field stop region 306a is provided with a graded doping profile 310 that, in the example, increases linearly from the epitaxial drift region 306b to the substrate 304, or, put another way, decreases linearly with distance away from the substrate 304 in a direction of the epitaxial drift region 306b. In various examples, a slope and maximum doping concentration of the graded doping profile 310 may be adjusted to any desired value(s). As illustrated in
The resulting doping concentration of the epitaxial field stop region 406 may be obtained in a two-step process, as illustrated below with respect to
As noted above, the proton implant 412 may provide additional blocking of the drift region electric field, while also improving ohmic contact properties. Although conventional field stop substrate layers may include proton implants, such proton implants, as noted above, are formed in a substrate and not within an epitaxially-grown region. Moreover, such conventional proton implants require the more expensive, higher-energy processes referenced above, whereas the proton implant 412 may be implemented as a shallow implant, using lower-cost, lower-energy equipment.
In other example implementations, however, a multi-step epitaxial growth process may be implemented, in which discrete process steps are used to successively form smaller doping peaks 508a, 508b, 508c of a peaked doping profile 508. Such implementations may be easier and more cost-effective to implement than techniques that continuously vary the doping profile during a single epitaxial growth process to obtain a correspondingly continuous graded doping profile.
Then, during subsequent frontside processing of a frontside device 308, the peaked doping profile 508 may undergo annealing processes of the frontside device 308. As a result, within an epitaxial field stop region 506a of an epitaxial region 506 that also includes an epitaxial drift region 506b, the peaked doping profile 508 may be diffused into the type of continuous, linear, and/or smooth graded doping profile described herein and illustrated in the examples of
In other example implementations, the annealing processes of the frontside device 308 may not fully diffuse the peaked doping profile 508 into a linear doping profile. In such examples, the graded doping profile of the epitaxial field stop region 506a may have a stepped decrease with distance away from the substrate 304. Although the widths of the peaks 508a, 508b, 508c are the same or similar in the example of
That is, for example, the peaked doping profile 508 may be diffused into the graded doping profile 310 of
A graded epitaxial growth process may then be performed on the MCZ substrate (604). For example, as referenced above, an epitaxial growth processing recipe may be adjusted while processing is occurring, so that the resulting graded doping profile continuously decreases as the epitaxial field stop region (e.g., the epitaxial field stop region 306a of
Frontside processing may then be performed (606). For example, the device 208 of
Backside grinding of the MCZ substrate may then be performed (608). For example, the MCZ substrate may be thinned to less than 100 microns, e.g., to 65-75 microns.
In the example of
A backside junction may be formed (612). For example, a backside emitter may be formed from a backside surface of the thinned MCZ substrate. More generally, any suitable or desired backside processing may be executed.
A laser anneal may be performed (614). For example, a laser anneal may be used to complete or optimize the preceding backside processing.
A back metal may then be provided (616). For example, the backside contact 102 of
That is, for example, the peaked doping profile 508 of
Then, when frontside processing is performed (706), and prior to backside grinding being performed (708), annealing that occurs during the frontside processing may serve to diffuse the peaked doping profile 508 to obtain a continuous, linear, and/or smooth graded profile, such as the graded profile 310 of
Then, if included, a shallow proton implant process may be performed (710). A backside junction may be formed (712), followed by a laser anneal (714). Finally in the example of
Similarly, collector currents 812 and 814 correspond to the VCEs 802 and 804 (i.e., within corresponding devices), while collector current 816 illustrates a collector current in accordance with various embodiments described herein. A relatively smoothed IC and correspondingly reduced dlc/dt of the described embodiments are also correlated with improved switching.
Thus, described techniques enable reduced depletion region expansion and enhanced recombination of carriers outside of a depletion region, thereby reducing Vpeak during turn-off and better dv/dt control. Such carrier recombination outside of a depletion region also provides better dk/dt control when compared to existing approaches. Lower voltage spikes help to avoid dynamic avalanche modes or oscillation issues, leading to improved switching. Described methods can be applied to all suitable high voltage devices and many different high voltage applications, while eliminating a need for high-cost proton implant processes.
It will be understood that, in the foregoing description, when an element, such as a layer, a region, a substrate, or component is referred to as being on, connected to, electrically connected to, coupled to, or electrically coupled to another element, it may be directly on, connected or coupled to the other element, or one or more intervening elements may be present. In contrast, when an element is referred to as being directly on, directly connected to or directly coupled to another element or layer, there are no intervening elements or layers present. Although the terms directly on, directly connected to, or directly coupled to may not be used throughout the detailed description, elements that are shown as being directly on, directly connected or directly coupled can be referred to as such. The claims of the application, if any, may be amended to recite exemplary relationships described in the specification or shown in the figures.
As used in the specification and claims, a singular form may, unless definitely indicating a particular case in terms of the context, include a plural form. Spatially relative terms (e.g., over, above, upper, under, beneath, below, lower, and so forth) are intended to encompass different orientations of the device in use or operation in addition to the orientation depicted in the figures. In some implementations, the relative terms above and below can, respectively, include vertically above and vertically below. In some implementations, the term adjacent can include laterally adjacent to or horizontally adjacent to.
Some implementations may be implemented using various semiconductor processing and/or packaging techniques. Some implementations may be implemented using various types of semiconductor processing techniques associated with semiconductor substrates including, but not limited to, for example, Silicon (Si), Gallium Arsenide (GaAs), Gallium Nitride (GaN), Silicon Carbide (SiC) and/or so forth.
While certain features of the described implementations have been illustrated as described herein, many modifications, substitutions, changes and equivalents will now occur to those skilled in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the scope of the implementations. It should be understood that they have been presented by way of example only, not limitation, and various changes in form and details may be made. Any portion of the apparatus and/or methods described herein may be combined in any combination, except mutually exclusive combinations. The implementations described herein can include various combinations and/or sub-combinations of the functions, components and/or features of the different implementations described.
While certain features of the described implementations have been illustrated as described herein, many modifications, substitutions, changes and equivalents will now occur to those skilled in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the scope of the embodiments.
This application claims the benefit of U.S. Provisional Patent Application 63/260,239, filed on Aug. 13, 2021, entitled “FIELD STOP OPTIMIZATION FOR HIGH POWER SEMICONDUCTORS”, the entire content of which is incorporated herein by reference in its entirety.
Number | Date | Country | |
---|---|---|---|
63260239 | Aug 2021 | US |