ELECTROSTATIC PROTECTION DEVICE

Abstract

An apparatus including an electrostatic discharge (ESD) protection device comprising a semiconductor having first, second and third regions arranged to form a transistor, wherein the first region is doped with a first impurity of a first conductivity type and is separated from the second region which is doped with a second impurity of a second conductivity type opposite the first type, and wherein a dimensional constraint of the regions defines an operational threshold of the ESD protection device. In one example, the separation between a collector and an emitter of a bipolar transistor defines a trigger voltage to cause the electrostatic discharge protection device to become conducting. In another example, a width of a bipolar transistor base controls a holding voltage of the electrostatic discharge protection device.

Description

FIELD OF THE DISCLOSURE

The present application relates to an improved electrostatic discharge protection device.

BACKGROUND

It is well known that some integrated circuit technologies are susceptible to damage by electrostatic discharge (ESD). CMOS transistors which are frequently used in logic circuits are an example of such devices. Protection schemes are known, such as providing diodes connecting an input pin to the device supply rails. However these measures are crude and can switch into a conducting state during normal operating conditions.

More sophisticated ESD protection circuits have been implemented using transistors within an integrated circuit. However, the transistor parameters and fabrication processes used in these integrated ESD protection circuits have not typically been optimized for their ESD function. Instead, the performance characteristics of these ESD transistors have largely been dependent upon fabrication parameters chosen to optimize other transistors that carry out the primary function of the circuit to be protected. Therefore, it has mainly been a matter of luck whether the fabrication parameters chosen to optimize majority devices are suitable for use in the accompanying ESD protection circuit. While it is possible to separately optimize an ESD protection circuit using additional processing steps, those additional processing steps carry increased costs. Instead, a reliable method of tailoring ESD protection device performance without using additional processing steps is required.

SUMMARY OF THE DISCLOSURE

According to a first aspect of the present invention there is provided an apparatus including an electrostatic discharge protection device comprising a semiconductor having first, second and third regions arranged to form a transistor, wherein the first region is doped with a first impurity of a first dopant type and is separated from the second region which is doped with a second impurity of a second dopant type, and wherein the electrostatic discharge protection device is adapted such that at least one of the following applies:

• • the separation between the first and second regions defines a trigger voltage to cause the electrostatic discharge protection device to become conducting; and
  • a first spatial parameter of the second region controls a holding voltage of the electrostatic discharge protection device.

It is thus possible to provide an electrostatic protection device where the trigger is determined by the distance between the first and second regions, both of which are regions at the surface of the semiconductor, either in the finished device or during one of the processing steps in its fabrication, and hence their separation is well controlled during the masking and doping processes, as known to the person skilled in the art. It is feasible that other structures might be formed over the surface of the electrostatic protection device. In these circumstances the relevant separation would still have been defined at a surface of the semiconductor whilst the surface was exposed during fabrication. In any event the device properties may still be controlled by structures or dimensions defined on a surface in a horizontal plane, whether or not that surface is buried.

In some embodiments, a holding voltage can be controlled by controlling a width, area, or volume of the second region. It is thus possible to control the holding voltage by varying a spatial parameter that is easily modified during the masking and doping steps.

Thus, both the breakdown voltage and the holding voltage may be defined by dimensions at the surface of the device (or at least in a horizontal plane), and which are controllable during device fabrication.

According to a second aspect of the present invention there is provided a method of manufacturing an electronic apparatus including an electrostatic protection device. The method includes forming a horizontal bipolar transistor in a semiconductor substrate. The method also includes either: (1) selecting a first distance between a collector region and a base region to define a trigger voltage for the device; or (2) selecting a width of a base region to define a holding voltage.

According to a third aspect of the present invention there is provided an integrated circuit including an electrostatic discharge protection device according to the first aspect of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will now be described, by way of non-limiting example only, with reference to the accompanying drawings, in which:

FIG. 1 schematically illustrates an input stage of an integrated circuit, the input stage including an electrostatic discharge protection device;

FIG. 2 illustrates a desirable current versus voltage operating characteristic of an electrostatic discharge protection device;

FIG. 3 illustrates a device structure for a vertically fabricated NPN prior art transistor;

FIG. 4 shows part of the device of FIG. 3 in greater detail, and also shows field gradients that initiate breakdown by impact ionization;

FIG. 5 shows the doping profile of an electrostatic protection device in accordance with an embodiment of the invention;

FIG. 6 is a plan view of the device of FIG. 5;

FIG. 7 is a graph of current flow versus pin voltage for a device under test, illustrating different trigger voltages for different collector-base separations in the ESD protection device of FIGS. 5 and 6.

FIG. 8 is a graph showing holding voltage versus base width, according to one embodiment;

FIG. 9 is a representation of a bidirectional ESD protection device, according to one embodiment;

FIG. 10 is a graph of collector current versus collector voltage, illustrating the effect of a base-emitter resistor, according to one embodiment;

FIG. 11 shows a device structure for a further embodiment of the present invention;

FIG. 12 shows the device structure for another embodiment of the invention; and

FIG. 13 is a plan view showing a variation on the device shown in FIG. 6.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

Terms such as above, below, over, and so on as used herein refer to a semiconductor device orientated as shown in the figures and should be construed accordingly. It should also be appreciated that, because regions within a semiconductor device are defined by doping different parts of a semiconductor with differing impurities or differing concentrations of impurities, discrete physical boundaries between different regions may not actually exist in the completed device but instead regions may transition form one to another. Some boundaries as shown in the accompanying figures are of this type and are illustrated as abrupt structures merely for the assistance of the reader. The skilled artisan can readily determine with such gradients where to set boundaries for discrete functional diffusion regions in a semiconductor device.

As noted before, transistors have sometimes been used to provide electrostatic protection devices. However an integrated circuit may contain millions of transistors. The circuit designer wants to use as few process steps as possible in the fabrication of the integrated circuit since each additional masking and implanting step adds further cost. Also each step brings an increased error rate which reduces the yield of working devices formed on a wafer. Thus additional steps are preferably avoided. Given that the majority of transistors on a device will have a specific task, such as forming logic elements, then the transistor parameters are chosen so as to be appropriate for the majority task. Consequently the transistors that are formed are generally not suited for use in electrostatic discharge protection circuits, or give severely compromised performance.

FIG. 1 shows part of an integrated circuit 5, according to one embodiment. The integrated circuit has a device, in this instance an input transistor 10 whose drain and source are connected to circuit components 12 and 14, which may be any type of circuits with any function for the purpose of the present disclosure. A gate 16 of the transistor is connected to an input terminal 18 of the integrated circuit. The transistor 10 is susceptible to permanent damage due to high voltages of the input terminal 18, for example electrostatic discharge if someone handling the integrated circuit 5 had become charged by walking over a nylon carpet.

In order to protect the transistor 10 from damage, an electrostatic protection device (or circuit) 20 is provided so as to limit the voltage excursions at the gate 16 of the transistor.

It is useful to consider the operation of an electrostatic discharge, ESD, protection device. Ideally, an electrostatic protection device should exhibit a controllable and selectable trigger voltage T. For input voltages having a magnitude less than the trigger voltage T, the device should be in a high-impedance (non-conducting) state. Once the trigger voltage magnitude has been exceeded the device should enter a low impedance state and start conducting.

The protection device may also offer “fold back” or “snap back” such that once it starts conducting, the voltage across the device reduces to a lower value. In this case, the device remains conducting, provided that the input voltage remains above a threshold magnitude, which may be called a “holding voltage”.

FIG. 2 illustrates an operating characteristic of an electrostatic protection device. It can be seen that the ESD protection device does not pass any current until a trigger voltage T has been reached. The trigger voltage is less than a breakdown voltage B for the device being protected. Once the trigger voltage has been reached the ESD protection device starts conducting, and the voltage across the device falls back to a holding voltage H. Although in an ideal device current flow could then increase without the voltage across the device increasing, due to resistance within the device, the voltage increases slightly with increasing current in the region 30. If the holding voltage is not outside the supply rail voltage range, then once the ESD protection device has switched on, it will not switch off. Once the voltage across the device has decreased below a holding voltage, H, the ESD protection device can return to a high impedance state, effectively switching off.

The inventors realized that for a given transistor fabrication process, it would be desirable for the trigger voltage and the holding voltage to be well controlled, and better still, adjustable. Doping concentrations and thermal budgets are already constrained by the function of the majority devices when those devices are simultaneously fabricated with ESD protection device(s). Therefore, those control parameters are not accessible to independently tailor ESD protection device performance. Fabrication steps to tailor doping just for ESD protection devices are costly, and it is not feasible to separate thermal budgets for different devices on the same substrate. Thus, it may be useful to control other device parameters and internal transistor processes.

Consider, for example, the structure of a vertically formed NPN bipolar transistor. During device fabrication the semiconductor wafer exists as a slab whose width and length is much greater than its depth. A surface of the wafer is exposed to impurities for doping. The surface is regarded as an upper horizontal surface in a frame of reference that is adopted for the purposes of description. An exemplary prior-art transistor 60 is illustrated in FIG. 3. The device shown in FIG. 3 represents a single transistor within an integrated circuit, which may have millions of transistors formed thereon. The active part of the transistor comprises an N⁺ region 100 which acts as the collector of the transistor. The N type region is formed by doping the semiconductor with a donor impurity, as is well known to the person skilled in the art. The “+” symbol represents a region of relatively heavy doping. This, again, is a convention well known to and understood by the person skilled in the art. The N type region can be formed as a well within a P type substrate 80. This gives rise to the formation of a PN junction well, which can be reverse biased so as to isolate the bulk of the transistor from other transistors within the integrated circuit. Alternatively, the transistor can be formed within a semiconductor well that is defined along its sides and bottom by a layer of silicon dioxide, as is known for silicon on insulator (SOI) fabrication. Connections are made to the collector region 100. This is achieved by providing vertical N type regions 102 extending between the collector region 100 and the surface of the semiconductor. Metallic collector contacts 104 make galvanic contact with the N type semiconductor.

A further region of less heavily doped N type semiconductor 110 is provided above the collector 100, and bounded by the vertical regions 102. It contains a well of P⁺ doped semiconductor which forms the base region 120 of the transistor. Finally, an emitter region 130 of N⁺ doped semiconductor is provided along with metallic emitter contact 132. It can be seen that the transistor structure NPN exists vertically along the line A-A′.

However it can also be seen that a horizontal or lateral NPN transistor structure also exists at the surface of the device. This horizontal transistor may be considered to be undesirable as it constitutes a structure that can give rise to breakdown, i.e. unwanted and often uncontrolled current conduction by transistor 60.

FIG. 4 looks at a portion of the transistor shown in FIG. 3 in greater detail and specifically focuses on the region near the surface of the device that includes the emitter, base and collector regions 130, 120, 100 (see FIG. 3). As well as showing nominal device structure, it also shows simulated electric field gradients. As the device is being shown in greater detail, some additional features are also shown. Thus a region 122 of enhanced P⁺ doping (sometimes designated P⁺⁺) is provided under a metallic base contact 124 to improve conductivity in this region. Similarly a region 103 of enhanced N⁺ type doping is provided under the metallic collector contact 104 to improve conductivity.

In the finished device an oxide layer 150 is provided over the surface of the transistor.

Although the device structures, such as the base and the collector have been shown as being well defined, it should be appreciated that during fabrication, the doping (e.g., ion implantation) process occurs from above the surface of the device (when viewed in the frame of reference adopted herein) so dopant concentrations are greater near the surface and naturally decrease with increasing depth into the wafer. Following implantation, a diffusion step is performed where the wafer is heated. This allows dopants to spread helping smear out local discontinuities in dopant concentration. However diffusion occurs in all directions so the theoretically sharp delineation between the base region 120 and the layer 110 becomes a smoother change in concentration and hence the transistor does not have a sharp boundary between these regions.

It should also be noted that higher dopant concentrations near a junction between dissimilarly doped materials means that the depletion region is smaller and hence any voltage difference between the dissimilar regions is dropped across a smaller distance, and hence the electrostatic field gradient is higher.

Given that implantation occurs from above, it follows that the highest field gradient tends to be near the surface of the transistor, even after the thermal diffusion step has occurred. Thus the depletion region of a junction is narrower near the upper surface of the transistor.

Additionally, it is known from electrostatic theory that the field gradient increases around a curved surface. The formation of the base layer gives rise to a structure having a nominally flat lower surface (which can be regarded as part of a cylinder of infinite radius) with curved edges (which can be regarded as a cylinder having a radius similar to the diffusion distance). Thus, the curvature at the edge of the base region gives rise to a field strength enhancement.

The simulation of electric field gradients shown in FIG. 4 shows a small portion of the device representing the first portion to suffer breakdown occurring at a region 200 adjacent the edge of the base region 120. This region 200 marks the interface between the base and the collector, known as the base-collector junction, where the field gradient exceeds 5×10⁵ volts per meter. The region 210 enclosed by broken line 212 has a field gradient greater than 4×10⁵ volts per meter.

The high field gradient provokes impact ionization in region 200, with carriers (in this case electrons), being swept towards and injected into the base region. Here they may cause the device to turn on, and if the current flow in the device is not limited by an external factor the collector current will increase until the device becomes damaged.

However, the inventors realized that the impact ionization driven breakdown process in this vertical transistor is highly controllable for a given fabrication process. In fact, experimentation has shown that the trigger voltage can be directly set by controlling the separation between the edge of the base and the edge of the collector region. This distance can be defined by masks during the doping stages of device fabrication.

In embodiments described herein, impact ionization sets up a current in the base, which forward biases the base-emitter junction and turns the transistor on. This in turn causes an emitter current which itself gives rise to the creation of additional impact ionization, by virtue of a combination of current density and electric field gradient occurring within a region of the device.

Once the device has turned on, in order to perform its function, it should stay on while the voltage across the device exceeds a holding voltage and/or the current therein exceeds a holding value. The holding voltage is the voltage across the device that is required to sustain the impact ionization process. Normally, device designers go to considerable lengths to try and avoid impact ionization from occurring. However, in embodiments of the invention, not only is the device modified so as to allow the onset of impact ionization to be controlled so as to set the trigger voltage, but the device is also designed so as to sustain impact ionization provided the voltage across the device exceeds a holding voltage. Of course, the designer may also want to control the holding voltage.

Once impact ionization has started, the current flow in the semiconductor enhances the impact ionization effect. In broad terms, impact ionization can be achieved with large voltages and small current densities or, critically, larger current densities and reduced voltages.

The inventors realized that controlling the size of the base region controls the current gain of a transistor, and that this in turn would provide a mechanism for controlling the holding voltage of the protection device. It is of course desirable for the control mechanism to be reliable and predictable.

This gives rise to a protection device where both the trigger voltage and the holding voltage are definable by features at the surface of the device. Furthermore no additional processing steps are required compared to those required to form the “majority” transistors that aid in carrying out the primary function of the integrated circuit.

FIG. 5 schematically illustrates the layout of an ESD protection device in accordance with an embodiment of the invention. The device is illustrated as being fabricated inside a well of semiconductor delimited by insulating material. This may be advantageous as it provides enhanced device isolation (and the processes for providing such isolation are well known and offered by semiconductor fabricators as standard so do not need to be described here).

In another arrangement, the device can instead be fabricated in a region of material that, in combination with the semiconductor substrate, is arranged to form a reverse biased PN junction well for isolation. However, the fabrication inside a well of insulating material (as shown) may be advantageous for an ESD protection device as the magnitude and polarity of the ESD event may be unpredictable. Side walls 250 can be formed as dielectric-filled trenches, whereas the bottom of the well 252 can be provided as silicon dioxide in a SOI structure.

Compared with the standard transistor as shown in FIG. 3, the N⁺ region 100 along the bottom of the device is omitted, and a P⁺ region 260 is provided instead. This helps ensure that any vertical NPN transistor structure is prevented. A P region provided between the base region 120 and the P⁺ region 260 also inhibits vertical transistor formation. Additionally, as the transistor is deliberately fabricated as a horizontal structure, the collector regions 270 are only provided in the vicinity of the surface and hence the vertically extending regions 102 (FIG. 3) are omitted. Alternatively, the N⁺ region 270 can be extended by forming region 270a, or regions 270a and 270b, as illustrated. It will be appreciated that as the device may be formed by growing an epitaxial layer over an initial (handle) wafer, then region 270b may be implanted, or otherwise doped, before the layer containing region 270a is grown on the wafer. Similarly, region 270a may be implanted with dopant before a top layer containing region 270 is grown on the wafer or it can be done at the same time. Otherwise, the structure is similar to the device shown in FIG. 3.

FIG. 6 is a plan view of the surface of the device shown in FIG. 5, and shows where the implantations for the collector region 270, base 120 and emitter 130 are located, according to one embodiment.

The base-collector separation 300 controls the trigger voltage of the ESD protection device. The distance 310, defining the width of the base 120, controls the holding voltage. Thus the device parameters are controlled by spatial features defined at the surface of the device. This means device characteristics can be accurately defined during design of the masks used to define the doping regions.

Similarly, the area and volume of the base 120 may control the holding voltage. For example, the area may be defined at the surface by modifying the width and length of the base 120, depending upon the shape. The volume is defined by the area times depth of the doped region. Thus, generally, a horizontal dimension defined during doping steps can be modulated to affect that horizontal dimension as well as the area and volume of the doped region, such that each of these parameters can be said to control an operational threshold of an ESD device.

The device shown in FIG. 6 has 2 planes of reflection symmetry. However, the device need not be formed with such a high degree of symmetry, for example the regions 302 of the collector 270 could be omitted such that the collector was only formed by region 303a at one side, or regions 303a and 303b at two opposing sides if they are electrically connected together. Such a device having the collector 270 at one side thereof and emitter 130 at the other side is shown in FIG. 13.

FIG. 7 shows test results for current flow versus voltage for a device under test, according to one embodiment. The input voltage at the terminal 18 (FIG. 1), being protected by an ESD device 20 of the embodiment, was swept from zero, and the current being passed by the ESD device 20 was measured. In this device under test the base width, as designated by line 310 in FIG. 6, was held constant at 45 μm, and the collector-base separation or space 300 between the P-type base and the N-type collector was varied from 12 μm for line 320, to 13 μm for line 322, and 14 μm for line 324. This gave trigger voltages of 45, 78 and 115 volts respectively, for a transistor fabricated in accordance with a given fabrication process. Different processes, having different doping concentrations or thermal budgets give different specific results.

A similar test of holding voltage was performed for a base-to-collector spacing of 14 μm and varying the base width from 12 to 50 μm. The holding voltages are shown in FIG. 8. In general terms, the holding voltage increases by approximately 2 V for each 1 μm of base width, in a well behaved manner.

In both examples of modulation in FIGS. 7 and 8, the devices were designed such that changes in a mask-defined, horizontal dimension (e.g., spacing between two doped regions of a transistor or width of a doped region) caused distinct and readily measurable changes in operational thresholds (e.g., trigger voltage or holding voltage) of the ESD protection device 20 incorporating that transistor. Thus, the trigger voltage and holding voltage can be tailored for the ESD protection device 20 and its functions by the simple selection of horizontal dimensions in the doping masks.

Because the ESD protection device 20 allows mask-defined horizontal dimensions to tailor operational thresholds (e.g., trigger voltage or holding voltage), special doping dosages need not be used for this purpose. Instead, a designer may select doping dosages in order to optimize the performance of other transistors to be simultaneously fabricated on the substrate for the same integrated circuit. In some embodiments, at least one ESD protection device 20 is fabricated, and particularly doped, on the substrate simultaneously with at least one “primary” device, by which is meant a non-protective transistor employed in the primary functions of the integrated circuit. During a doping step of the fabrication process, a single mask may be used to simultaneously define at least one doped region or active area of both an ESD device and a primary device. Thus, one or more doped regions of the primary device may share the same dopant dosage and thus about the same maximum dopant levels with one or more doped regions of ESD protection device.

In one embodiment, the primary device is a MOSFET transistor, such as the transistor 10 of FIG. 1 that is to be protected by the ESD protection device. The source and drain regions of such a MOSFET can be simultaneously doped, using the same mask, as one of the regions of the ESD protection device. For example, the source and drain regions of a MOSFET and the base; or the collector and emitter regions of an ESD protection device may be simultaneously doped P⁺, or may be simultaneously doped N⁺, by a single mask. In another embodiment, the emitter region of a primary bipolar transistor and the emitter region of an ESD protection device may be simultaneously doped N⁺, or may be simultaneously doped P⁺, by a single mask. Similarly, each of the ESD protection device base and collector regions can be simultaneously doped with base and collector regions, respectively, of other non-protective bipolar transistors on the substrate.

In each of the foregoing examples, one or more active regions of a primary transistor and the ESD protection transistor share a common dopant dosage and thus about the same maximum dopant level, which may be selected in order to optimize performance of the primary transistor. At the same time, performance of the ESD protection device may be adjusted by selecting a mask-defined horizontal dimension. The ability to tailor an ESD protection device's operational threshold voltages such as trigger voltage and holding voltage through selection of horizontal dimensions in the doping masks allows doping dosages to be optimized for other transistors that are simultaneously fabricated, while still minimizing masking steps by simultaneous doping of both types of transistors.

According to one embodiment, the ESD protection device is connected such that the collector 270 is connected to the terminal 18 of the integrated circuit 5 that is to be protected, and the emitter is connected to the ground supply rail. The base terminal can be left floating or can be connected to the emitter via a resistor. Where a resistor is provided, the voltage difference across the resistor that arises when current flow in the base region has been initiated by impact ionization can be used to further control the “snap back” characteristic of the ESD protection device. This is further discussed below.

The arrangement described hitherto is suitable for providing unidirectional ESD protection. However, ESD events may occur with either polarity, and hence the integrated circuit benefits from protection against input terminal voltages that are either excessively above its positive supply rail, or excessively below its negative supply rail. In order to achieve this bi-directional ESD protection, two devices can be provided in series.

Two devices similar to that shown in FIG. 5 are illustrated in FIG. 9, according to one embodiment. The same numbering is used as in FIG. 5 to refer to like parts, except that the designations “a” and “b” are used as suffixes. Both are horizontal NPN transistors.

The base regions 120a and 120b are now drawn as a ring surrounding the emitter regions 130a and 130b, which serve to delineate the edges of the regions in a slightly different way than was done in FIG. 5, for purposes of illustrating how the principles and advantages described herein can be obtained using a variety of transistor configurations. The masking and implantation steps remain the same.

However, intrinsic (high impedance) regions 360a and 360b have been fabricated just below the emitter regions 130a and 130b and above the P⁺ regions 260a and 260b. These regions represent additional measures to stop the formation of parasitic components, such as thyristors, that could cause the device to latch into a conducting state.

In this arrangement the collectors 270a and 270b are connected together, and the emitter 130a and base 120a of one of the ESD protection devices is connected to the terminal 18 that is to be protected. The emitter 130b and base 120b of the other device has a current flow path to a supply rail, and is preferably connected to ground. Thus, for any polarity one of the ESD devices acts as a forward biased diode while the other acts as a reverse biased transistor, and hence breaks down to give the ESD protection when it reaches its trigger voltage. If the polarity of the ESD threat reverses, then the roles of the ESD protection devices reverse, with the one that had previously been acting as a forward biased diode becoming the reverse biased transistor, and the one that had been acting as the reverse biased transistor becoming the forward biased diode. This enables laterally (horizontally) fabricated NPN transistors to provide ESD protection for discharge events of either polarity. The trigger and holding voltages are still defined by the separation between the collector and the base regions, and the size of the base region, respectively within each device. This means the trigger voltages can be set independently for each polarity of ESD threat if desired.

In other embodiments, other modifications may be made to the ESD protection devices. For example a metal plate 370b may be connected to the base or emitter junctions and arranged to extend over the edge of the collector region. This acts as a field plate 370b and helps prevent charge injection occurring in the oxide layer over the base-collector junction. Such a field plate 370b encircles the emitter when viewed in plan view.

As noted earlier, the inclusion of a resistor between the base and emitter terminals can modify the turn on characteristic of the device. FIG. 10 illustrates the collector current versus collector voltage characteristics for a device in a grounded emitter configuration, according to two embodiments. Both devices have the same trigger voltage of 40 V, but once the device has triggered, the device with a floating base snaps back more deeply than a similar device with a 6 KΩ resistor connecting the base and emitter terminals. Thus the resistor helps determine the collector current to turn the horizontal bipolar transistor on.

Embodiments have been described in the context of NPN devices. PNP structures can also be formed by reversing the dopant conductivity types in the examples provided above. However, carrier mobility is lower in PNP transistors, so they may provide a slower response.

So far, the structures controlling the breakdown and holding voltages have been described with respect to the devices where these structures are formed at the surface of the device. However, the principles and advantages described herein are not limited to such surface devices. FIG. 11 shows an arrangement where a subsurface or buried NPN bipolar junction determines the trigger voltage, according to one embodiment. In this arrangement, a silicon substrate 400 is provided into which an N-type layer 405 is formed so as to isolate the device from the substrate 400. An N+ collector 410 is implanted above the layer 405 and, in the finished device, connects to a collector electrode 415 via a vertically extending N-type region 412. A heavily doped P⁺ region 420 is also provided above the N-type layer 405, and at least part of the region 420 is horizontally aligned with the regions 410. A P-type region 422 is formed above the P⁺ region 420 and makes contact with a base electrode 425. An N-type region 430 that forms the emitter is deposited within the P-type region 422 and makes connection with an emitter electrode 432. A resistor 435 may optionally interconnect the base and emitter regions 420 and 430. The horizontal spacing 450 between the regions 420 and 410 is defined during the masking steps and controls the trigger voltage of the device. The horizontal distance 460 between the emitter 430 and the edge of the P-type base region 422 controls the holding voltage of the device. Thus, both of these parameters are still defined by features which the device designers could “draw,” or control by mask design, although not all of these features are coplanar.

FIG. 12 shows a further embodiment, which is similar to that shown in FIG. 11, except the horizontal NPN bipolar transistor is now isolated from the rest of the substrate by dielectric isolation comprising a silicon oxide insulating layer 440 at the base of the transistor and trench isolation 455 provided around the sides of the transistor. The choice between electrical junction isolation and dielectric isolation does not have any effect on the operation of the devices described herein. Although reference has been made to forming the semiconductor device by virtue of additional silicon epitaxial growth above an initial wafer, is also known in the art that the device structures can be formed in a single wafer with no additional silicon epitaxial growth occurring.

The arrangements shown in FIGS. 11 and 12 have the advantage that breakdown occurs away from the surface of the device, thereby preventing charge being trapped in an oxide layer 452 covering the surface of the device. This trapping can modify field gradients at the surface of the device.

It is thus possible to provide ESD protection devices where the characteristics are easily controlled by parameters at a surface (at least during device manufacture) of the device, or by control of horizontal dimensions of doped features. These features, and thus ESD protection device operational thresholds, are readily controlled by mask design.

Electrostatic protection devices as described herein can be implemented in various apparatuses. Examples of the electronic devices incorporating such ESD protection devices can include high speed signal processing chips, power regulators, memory chips, memory modules, circuits of optical networks or other communication networks, and disk driver circuits. Products in which such electronics can be incorporated include, but are not limited to, consumer electronic products, parts of the consumer electronic products, electronic test equipments, etc. The consumer electronic products can include, but are not limited to, a mobile phone, cellular base stations, a telephone, a television, a computer monitor, a computer, a hand-held computer, a personal digital assistant (PDA), a microwave, a refrigerator, a stereo system, a cassette recorder or player, a DVD player, a CD player, a VCR, an MP3 player, a radio, a camcorder, a camera, a digital camera, a portable memory chip, a washer, a dryer, a washer/dryer, a copier, a facsimile machine, a scanner, a multi functional peripheral device, a wrist watch, a clock, etc. Further, the electronic device can include unfinished products.

Although this invention has been described in terms of certain embodiments, other embodiments that are apparent to those of ordinary skill in the art, including embodiments that do not provide all of the features and advantages set forth herein, are also within the scope of this invention. Moreover, the various embodiments described above can be combined to provide further embodiments. In addition, certain features shown in the context of one embodiment can be incorporated into other embodiments as well. Accordingly, the scope of the present invention is defined only by reference to the appended claims.

Claims

1. (canceled)

2. An electrostatic discharge (ESD) protection device having mask-defined operational thresholds, wherein the ESD protection device comprises: a p-type semiconductor region;an n-type collector region positioned inside of the p-type semiconductor region;a p-type base region positioned inside of the p-type semiconductor region and spaced apart from the n-type collector region, wherein the p-type base region has a higher doping concentration than the p-type semiconductor region;an n-type emitter region positioned inside of the p-type semiconductor region and encircled by the p-type base region; anda p-type buried region vertically aligned with the p-type base region and extending at least from a first edge of the p-type base region to a second edge of the p-type base region opposite the first edge, wherein the p-type buried region has a higher doping concentration than the p-type semiconductor region,wherein the n-type collector region, the p-type base region, and the n-type emitter region are configured to operate as a horizontal bipolar transistor, and wherein the p-type buried region inhibits formation of a vertical bipolar transistor.

3. The ESD protection device of claim 2, wherein a horizontal spacing between the p-type base region and the n-type collector region establishes a trigger voltage at which the ESD protection device transitions from a non-conducting state to a conducting state.

4. The ESD protection device of claim 2, wherein the n-type collector region is buried in the p-type semiconductor region, wherein at least a portion of the p-type buried region is horizontally aligned with the n-type collector region.

5. The ESD protection device of claim 4, wherein the p-type buried region extends beyond the first edge of the p-type base region, wherein a horizontal spacing between the p-type buried region and the n-type collector region establishes a trigger voltage at which the ESD protection device transitions from a non-conducting state to a conducting state.

6. The ESD protection device of claim 4, wherein the p-type buried region and the p-type base region abut.

7. The ESD protection device of claim 2, further comprising a base electrode connected to the p-type base region, an emitter electrode connected to the n-type emitter region, and a resistor electrically connected between the base electrode and the emitter electrode.

8. The ESD protection device of claim 2, wherein the horizontal bipolar transistor is configured to turn on in response to an impact ionization current that forward biases a junction between the p-type base region and the n-type emitter region.

9. The ESD protection device of claim 2, wherein a size of the p-type base region establishes a holding voltage of the ESD protection device.

10. The ESD protection device of claim 2, wherein the p-type semiconductor region comprises a p-type well.

11. The ESD protection device of claim 10, wherein the p-type base region and the p-type buried region are separated by a portion of the p-type well.

12. The ESD protection device of claim 2, wherein the ESD protection device further comprises a field plate positioned over a portion of the p-type semiconductor region that is between the n-type collector region and the p-type base region.

13. The ESD protection device of claim 12, wherein the field plate encircles the n-type emitter region.

14. The ESD protection device of claim 2, further comprising an intrinsic region between the n-type emitter region and the p-type buried region.

15. The ESD protection device of claim 2, wherein the n-type collector region encircles the p-type base region.

16. A method of forming an electrostatic discharge (ESD) protection device, the method comprising: forming a horizontal bipolar transistor in a p-type semiconductor region, wherein forming the horizontal bipolar transistor comprises: forming an n-type collector region inside of the p-type semiconductor region;forming a p-type base region inside of the p-type semiconductor region and spaced apart from the n-type collector region, the p-type base region having a higher doping concentration than the p-type semiconductor region; andforming an n-type emitter region inside of the p-type semiconductor region and encircled by the p-type base region; andinhibiting formation of a vertical bipolar transistor by forming a p-type buried region vertically aligned with the p-type base region and extending at least from a first edge of the p-type base region to a second edge of the p-type base region opposite the first edge, the p-type buried region having a higher doping concentration than the p-type semiconductor region.

17. An integrated circuit comprising: a pin; anda first electrostatic discharge (ESD) protection device comprising: a p-type semiconductor region;an n-type collector region positioned inside of the p-type semiconductor region;a p-type base region positioned inside of the p-type semiconductor region and spaced apart from the n-type collector region, wherein the p-type base region has a higher doping concentration than the p-type semiconductor region;an n-type emitter region positioned inside of the p-type semiconductor region and encircled by the p-type base region, wherein the n-type emitter region is electrically connected to the pin; anda p-type buried region vertically aligned with the p-type base region and extending at least from a first edge of the p-type base region to a second edge of the p-type base region opposite the first edge, wherein the p-type buried region has a higher doping concentration than the p-type semiconductor region,wherein the n-type collector region, the p-type base region, and the n-type emitter region are configured to operate as a horizontal bipolar transistor, and wherein the p-type buried region inhibits formation of a vertical bipolar transistor.

18. The integrated circuit of claim 17, further comprising a substrate, wherein the p-type semiconductor region comprises a p-type well separated from the substrate by insulating material.

19. The integrated circuit of claim 17, further comprising a metal-oxide-semiconductor field-effect transistor (MOSFET) comprising a source region and a drain region, wherein the source region and the drain region share a common dopant dosage with the p-type base region.

20. The integrated circuit of claim 17, further comprising a second ESD protection device comprising an n-type collector region electrically connected to the n-type collector region of the first ESD protection device, and an n-type emitter region electrically connected to ground.

21. The integrated circuit of claim 20, wherein the first ESD protection device and the second ESD protection device are operable to provide bi-directional ESD protection.