The present invention relates generally to semiconductor devices, and in particular to power trench metal-oxide-semiconductor field-effect transistors (MOSFETs).
Power MOSFETS are a type of MOSFET that are designed to handle significant power levels. These devices are designed to tolerate high voltages at the transistor terminals. For these high voltage devices, the on-resistance is determined at least in part by the voltage sustaining layer. Additionally, the breakdown voltage of the device mainly depends on the doping concentration and the thickness of the layer. The lower the doping concentration and/or the larger the thickness of the layer, the higher the breakdown voltage. However, the higher breakdown voltage comes with a larger on-resistance (or on-voltage). It is desirable to provide a higher breakdown voltage while limiting the on-resistance and the gate capacitance of the device.
A power metal-oxide semiconductor field-effect transistor (MOSFET) includes a trench, a trench doping and a pillar doping region. The trench is etched into a silicon layer that includes a gate structure disposed therein. The trench doping is implanted in the silicon layer vertically below the trench and has an opposite doping type than the silicon layer. The pillar doping region is implanted in the silicon layer vertically below, and spaced from the trench doping. The pillar doping region has a same doping type as the trench doping.
A method of manufacturing a power metal-oxide-semiconductor field-effect transistor (MOSFET) includes forming a doping pillar in a silicon layer, etching a trench in the silicon layer above the doping pillar; implanting a doping region below the trench and spaced from the doping pillar; and depositing a first gate structure within the trench.
A power transistor includes a plurality of trenches etched into a silicon layer and a plurality of doping pillars. Each of the plurality of trenches includes a gate structure and a field structure disposed therein. The plurality of doping pillars are each implanted in the silicon layer vertically below a respective one of the plurality of trenches. Each of the plurality of doping pillars is doped opposite a type of the silicon layer.
A split-gate superjunction power metal-oxide-semiconductor field-effect transistor (MOSFET) is disclosed herein that includes a trench with at least one gate structure, a trench implant vertically below the trench, and a doping pillar vertically beneath the trench implant. The doping pillar is spaced from the trench implant and doped of the same type as the trench implant. The doping pillar may be formed, for example, by implanting doping regions into epitaxial formed silicon layers. For example, four epitaxial layers may be formed such that within each layer, a doping region is formed, aligned with, and in contact with the previous layer to form the doping pillar.
By implanting a trench implant in conjunction with the doping pillar, the split-gate superjunction power MOSFET may be manufactured, for example, using fewer epitaxial layers while achieving a similar breakdown voltage to that of prior art superjunction transistors. This may reduce the overall cost and complexity of manufacture of the power MOSFET. Using the split-gate trench may reduce the on-resistance and gate capacitance of the transistor. Lower gate capacitance may be advantageous in systems, for example, with high switching frequencies such as switching-mode power supplies.
Power transistor 10 may be implemented to accommodate voltages, for example, of up to, and greater than, 600 volts. In the embodiment shown in
Split-gate superjunction transistor 110 may be manufactured through the deposition of epitaxial layers 118a-118n vertically above the n+ substrate of drain 126. In the embodiment shown in
Doping pillar 116, which may be doped opposite the doping type of drain 126, may be implanted during the forming of epitaxial layers 118a-118n. During the manufacturing process, each epitaxial layer 118a-118n may include a doping implant for each doping pillar 116. These doping implants are configured to align with the doping implant of the previous epitaxial layer 118a-118n. A doping implant may also be included in the substrate as illustrated in
Trenches 112 may be etched into epitaxial layer 118a using, for example, a hard mask formed on the top of epitaxial layer 118a, or any other method of forming trenches in silicon layers. Trenches 112 are manufactured to align with doping pillars 116. Trenches 112 include gate structures 128 and field structures 130 disposed therein. Field structure 130 may be biased, for example, such that field structures 130 in adjacent trenches effectively shield the portion of epitaxial layers 118a-118n between adjacent trenches 112 from excessive voltage.
Trench implants 114 may be implanted vertically below each respective trench 112 following the etching of trenches 112. Any method of achieving the desired doping for the implant in epitaxial layer 118a may be utilized. Trench implants 114 may be doped the same type as doping pillar 116 with a lighter doping concentration (p−), for example. Following implantation of trench implants 114, gate structures 128 and field structures 130 may be deposited within respective trenches 112 using any known method. For example, an initial oxide layer may be deposited within trench 112, followed by deposition of gate structures 128 and field structures 130.
Additionally, body doping regions 122 may be implanted and driven to a desired depth within epitaxial layer 118a using any known method. Subsequently, source doping regions 120 may also be implanted into epitaxial layer 118 using any known method. Oxide layer 132 may then be deposited, for example, over epitaxial layer 118a, vertically above source doping regions 120, gate structures 128, and field structures 130. Topside metal 124 may be formed over oxide layer 132. In the embodiment illustrated in
Trench implant 114 in conjunction with doping pillar 116 act together to perform a similar function to that of doping pillar 24 of transistor 12 of
For example, transistor 10, as illustrated in
Doping pillar 116 is also utilized in the drift region of transistor 110 in order to reduce the on-resistance of the device, for example. In planar transistors, the breakdown voltage may be determined based upon the doping and thickness of the drift region. Therefore, a thick, lightly doped epitaxial layer was needed to achieve the higher breakdown voltages. However, with higher breakdown voltages came higher on-resistances. By utilizing doping pillar 116 in conjunction with trench implant 114 to contain the electric field, greater breakdown voltages may be achieved while reducing the on-resistance of transistor 110.
An additional advantage of split-gate superjunction power transistor 110 over planar transistor 10 is the split-gate configuration. The shorter gate length of split-gate superjunction power transistor 110 achieved by the split-gate configuration facilitates a lower gate capacitance for the device than that of planar superjunction transistor 10. The gate capacitance of split-gate superjunction power transistor 110 may be on the order of ten percent that of the gate capacitance of planar superjunction transistor 10, for example. This may be particularly useful in systems which have high switching frequencies for split-gate superjunction power transistor 110.
With continued reference to
At step 204, a first epitaxial layer 118n is formed on the substrate. The epitaxial layer 118n may be doped the same as drain 126 (n-type), for example. A doping region that is opposite the doping type of epitaxial layer 118n and drain 126 may be implanted in epitaxial layer 118n for each respective doping pillar 116. This implant is aligned with any respective implants in the substrate layer for doping pillars 116.
At step 206 it is determined if any further epitaxial layers 118a-118n are to be formed. If so, method 200 proceeds to step 208. If not, method 200 proceeds to step 210. At step 208, subsequent epitaxial layers 118a-118n are deposited on the previous epitaxial layer 118a-118n. For each epitaxial layer 118a-118 that is deposited, further p-type dopings, for example, are implanted within the respective epitaxial layer 118a-118n and aligned with the previous epitaxial layer 118a-118n to form doping pillar 116. Following step 208, method 200 returns to step 206 to determine if any further epitaxial layers 118a-118n remain to be deposited. In the embodiment illustrated in
At step 210, trenches 112 are etched into epitaxial layer 118a using any known method. For example, a hard mask may be formed on the final epitaxial layer 118a and utilized to etch trenches 112. Trenches 112 may be aligned such that each trench 112 aligns with a respective doping pillar 116. Additional processing may be utilized to remove any masks that are utilized to etch trenches 112.
At step 212, trench implant 114 may be implanted at the base of trenches 112 as a p-type implant, for example. Trench implant 114 is implanted vertically aligned with, and spaced from, doping pillar 114. While doped the same type (e.g., p-type), trench implant 114 may be doped lighter (e.g., p+) than the doping of doping pillar 114. Trench implant 114 and doping pillar 116 act to contain the electric field within epitaxial layers 118a-118n.
At step 214, source doping regions 120, body doping regions 122, gate structures 128, field structures 130 and topside metal 124 are deposited/implanted using any known methods. Utilizing method 200 allows split-gate superjunction power transistor 110 to be manufactured using fewer epitaxial layers than prior art power MOSFETs while maintaining a similar, or greater breakdown voltage.
While the invention has been described with reference to an exemplary embodiment(s), it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted for elements thereof without departing from the scope of the invention. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the invention without departing from the essential scope thereof. Therefore, it is intended that the invention not be limited to the particular embodiment(s) disclosed, but that the invention will include all embodiments falling within the scope of the appended claims.