Active solid-state devices (e.g. – transistors – solid-state diode – Field effect device – Having insulated electrode
Reexamination Certificate
2001-05-11
2003-03-11
Wojciechowicz, Edward (Department: 2815)
Active solid-state devices (e.g., transistors, solid-state diode
Field effect device
Having insulated electrode
C257S191000, C257S192000, C257S197000, C257S370000
Reexamination Certificate
active
06531748
ABSTRACT:
BACKGROUND OF THE INVENTION
Field of the Invention
The invention relates to a semiconductor power component having a source region, a drain region, a channel region provided between the source region and the drain region and an insulated gate provided at a distance above the channel region.
Semiconductor power components, in particular power MOS (Metal Oxide Semiconductor) field effect transistors, for example DMOS (Double-Diffusion Metal Oxide Semiconductor) field effect transistors, include a MOS structure which is essential for their application as electronic switches, for example. The MOS structure is for example a part of a field effect transistor controlled by an insulated gate. The semiconductor power components, in addition to the MOS structure, also contain an undesired parasitic bipolar transistor structure. If an n-channel MOS field effect transistor is considered as an example, its source region and drain region each being n-conductive and a p-conductive body or channel region being provided between the source region and the drain region, then the source region, the channel region and the drain region here form a parasitic npn bipolar transistor.
The breakdown voltage of such a parasitic bipolar transistor with open base, which leads to a so-called U
CEO
breakdown, as it is known, is generally considerably lower than the breakdown voltage of the MOS field effect transistor, that is to say for example of a DMOS field effect transistor, depending on the gain of the bipolar transistor. For example, the breakdown voltage of the parasitic bipolar transistor may be 50% or less of the breakdown voltage of the actual DMOS field effect transistor.
In order to be able to utilize the voltage range of the DMOS field effect transistor up to its breakdown voltage and, if appropriate, to be able to reach the breakdown voltage of the DMOS field effect transistor, the channel region and the source region must be short-circuited to each other by the source contact. However, such a short circuit always entails a series resistance, which is formed of the path resistance in the channel region.
In the case of an n-channel MOS field effect transistor, this path resistance in the channel region leads to the situation where, at high hole current densities such as occur in the case of an avalanche or in the case of a fast voltage rise at the pn junction, in particular after the inverse diode has been flooded with excess charge carriers, the pn junction between the source region and the channel region is nevertheless polarized in the forward direction. As a result, the short circuit between the source region and the channel region via the source contact becomes virtually ineffective, which permits the voltage to break down to the U
CEO
breakdown voltage value of the bipolar transistor.
As a result of this process, the MOS field effect transistor is generally destroyed, since the gate loses its controlling action and the current can no longer be turned off. This socalled “bipolar second breakdown” is also explained, inter alia, in B. Jayant Baliga: Modern Power Devices, New York 1987, pp. 314-316.
The problem indicated above has been known for a long time. Accordingly, there is already a large number of approaches to overcome it, which are predominantly based on the fact of making the maximum controllable hole current as high as possible through the use of a good short circuit between the source region and channel region.
A high hole current can be obtained in an n-channel MOS field effect transistor in that, for example, an additional p-conductive region with a high conductivity is provided underneath the source region, as proposed in U.S. Pat. No. 4,809,047. Another route is to ensure the shortest possible path over which the hole current has to flow laterally under the source region in the p-conductive channel region (of. U.S. Pat. No. 4,767,722). In the two cases above, ultimately the voltage drop underneath the source region as far as the source contact is minimized.
Furthermore, U.S. Pat. No. 4,364,073 describes an IGBT (Insulated Gate Bipolar Transistor) in which, through the use of high doping of the channel region and low doping of the source region, the gain of a parasitic npn bipolar transistor can be made sufficiently small that, together with the gain of the parasitic pnp transistor containing the rear p-conductive anode region, the thyristor firing condition is not met.
As described, for example, in U.S. Pat. No. 4,620,211, a further possibility of reducing the gain of a parasitic npn transistor is introducing a zone with a high recombination speed, for example a metallic contact, into the channel region underneath the source region. This procedure is particularly suitable for transistors which are capable of being reverse biased and which contain no short circuit between the source region and drain region, so that the pn junction between source region and channel region is able to accept voltage in the reverse direction.
In the case of integrated circuits using CMOS (Complementary Metal-Oxide Semiconductor) technology, CMOS-ICs (Complementary Metal-Oxide Semiconductor Integrated Circuits), as they are known, adjacent n-channel MOS field effect transistors and p-channel MOS field effect transistors together form thyristor structures. Under unfavorable conditions, for example high temperatures and a sudden voltage rise, the thyristor firing condition can be met in such a thyristor structure, which leads to a rapid current rise and to destruction of the integrated circuit. A so-called “latch-up” occurs. U.S. Pat. No. 4,728,998 and U.S. Pat. No. 5,142,641 specify how, by reducing the emitter efficiency of the source regions, the latch-up phenomenon can be prevented. Reducing the emitter efficiency is in this case achieved by producing the source regions from SiGe.
Finally, U.S. Pat. No. 5,216,271 discloses the practice of using a material with a low band gap in order to achieve low contact resistances in BICMOS (Bipolar Complementary Metal Oxide Semiconductor) components.
SUMMARY OF THE INVENTION
It is accordingly an object of the invention to provide a semiconductor power component which overcomes the above-mentioned disadvantages of the heretofore-known components of this general type and in which it is possible to avoid turning on a parasitic bipolar transistor in a reliable and simple manner.
With the foregoing and other objects in view there is provided, in accordance with the invention, a semiconductor power component, including:
a source region formed of a first material having a first band gap;
a drain region;
a channel region disposed between the source region and the drain region, the channel region being formed of a second material having a second band gap, the first band gap being smaller than the second band gap; and
an insulated gate disposed at a given distance above the channel region.
In other words, the object of the invention is achieved by forming the source region of a material whose band gap is smaller than that of the material of the channel region.
According to another feature of the invention, the first band gap is at least 0.1 eV smaller than the second band gap. Preferably, the first band gap is substantially 0.2 eV smaller than the second band gap.
According to a further feature of the invention, the semiconductor power component is an n-channel MOS field effect transistor or an IGBT.
In this case, a semiconductor power component is to be understood as an n-channel or a p-channel MOS field effect transistor, such as in particular a lateral or vertical DMOS transistor with a planar gate, a power transistor with a trench gate, a UMOS (U-Shaped Trench Metal-Oxide Semiconductor) transistor or corresponding IGBTs which, between the drain region and drain contact, also contain a doping region with a conduction type opposite to the drain doping.
In this case, the channel region generally has a conduction type which is opposite to that of the source region and of the drain region. However, it can also have the same conduction type in the case of “normally-on”
Mayback Gregory L.
Wojciechowicz Edward
LandOfFree
Semiconductor power component with a reduced parasitic... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Semiconductor power component with a reduced parasitic..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Semiconductor power component with a reduced parasitic... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3084441