Method of fabrication of high voltage IC bopolar transistors operable to BV.sub.CBO

Abstract

A high voltage bipolar and JFET have their gate and base connected and source and collector connected and the appropriate geometry for the bipolar to operate to its BV.sub.CBO limit. The collector and channel regions have the same depth and impurity concentration, the base and top gate regions have the same depth and impurity concentration and the emitter and source and drain regions have the same depth and impurity concentration.

Description

BACKGROUND OF THE INVENTION

The present invention relates generally to high voltage integrated circuits and, more specifically to a bipolar transistor in an intergrated circuit capable of achieving operation to BV.sub.CBO rather than BV.sub.CEO limit.

It is well-known that BV.sub.CEO is less than BV.sub.CBO in transistors with a uniform collector doping due to carrier multiplication by the emitter base junction. The relation between the two breakdowns is: BV.sub.CEO =BV.sub.CBO /H.sub.FE 1/n. In silicon devices, n.apprxeq.4 for NPN's and n.apprxeq.6 for PNP's. It is found in many devices that BV.sub.CEO <1/2BV.sub.CBO as a result of this relationship. It is also well known that BV.sub.CBO .varies.N.sub.C.sup.-3/4. Consequently, to increase BV.sub.CBO and thus BV.sub.CEO, one must decrease N.sub.C, the doping of the collector region, and thus increase its resistivity in inverse proportion. Transistor collector resistance is directly proportional to collector resistivity and inversely proportional to device area. Thus, when collector resistivity is increased to increase BV.sub.CEO, device area must also be increased to achieve a given value of collector resistance. Large device areas are costly to manufacture and therefore one seeks ways to minimize collector area. One method is to use designs which allow operation to the BV.sub.CBO rather than the BV.sub.CEO limit of the transistor.

SUMMARY OF THE INVENTION

An object of the present invention is to provide an integrated circuit design which allows a bipolar transistor to operate to the BV.sub.CBO rather than the BV.sub.CEO limit of operation.

Another object of the present invention is to provide a high voltage junction field effect transistor which can be used with a bipolar transistor to allow the bipolar transistor to operate to its BV.sub.CBO level.

A still further object of the present invention is to fabricate an integrated circuit having high voltage bipolar and junction field effect transistors using the steps necessary only to form the bipolar devices.

These and other objects of the invention are attained by providing an integrated circuit having bipolar and junction field effect transistors with the base and gate being connected together, the source and collector being connected together and the geometry of the bipolar and junction field effect transistors defining I.sub.dss .gtoreq.I.sub.C max and BV.sub.dg .gtoreq.BV.sub.CBO. The high voltage which is normally applied to the collector of the bipolar transistor is now applied to the drain of the junction field effect transistor such that the major voltage drop is across the drain-gate junction of the junction field effect transistor with only a low voltage across the gate-source junction which is common to the collector of the bipolar transistor. Thus, the maximum voltage which the circluit can withstand between the junction field effect transistor drain and the bipolar emitter is BV.sub.dg +V.sub.BE.

A junction field effect transistor suitable for high voltage applications is fabricated wherein the channel region has the same depth and impurity concentration as the collector region of the bipolar transistor. The top gate has the same impurity concentration and depth as the base of the bipolar transistor and the source and drain regions have the same depth and impurity concentration as the emitter of the bipolar transistor. The top gate extends beyond the lateral edges of the channel region between the source and drain region and a bottom gate is provided which extends laterally beyond the channel region such that the intersection of the channel region with the top and bottom gates approach planar breakdown and, thus, have an increased breakdown relationship.

The method of fabricating the bipolar and junction field effect transistors includes simultaneously forming the channel region of the junction field effect transistor and the collector region of the bipolar transistor. Next, the top gate region of the junction field effect transistor and the base region of the bipolar transistor are formed simultaneously. This is followed by the simultaneous formation of the source and drain regions of the junction field effect transistor and the emitter regions of the bipolar transistor. A bottom gate region is formed before the formation of the top gate region as a buried region. The pair of bipolar and junction field effect transistors may be formed in separate isolated region or may be formed in a merged structure. A lateral region which laterally surrounds the channel region is formed and interconnects the top and bottom gates which extend laterally beyond the channel region. Beginning with an N substrate, the PNP and PJFET use the substrate as the lateral region. For the NPN and NJFET combination, the substrate forms the channel and the collector region and the connecting region is a diffused P type region.

Other objects, advantages and novel features of the present invention will become apparent upon review of the detailed description in combination with the accompanying claims.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic of a NPN and NJFET incorporating the principles of the present inventon.

FIG. 2 is a schematic of a PNP and PJFET incorporating the principles of the present invention.

FIG. 3 is a cross-sectional-perspective view of a high voltage P channel JFET incorporating the principles of the present invention.

FIG. 4 is a cross-sectional view taken along the lines IV--IV of FIG. 3.

FIG. 5 is a cross-sectional view of a merged PNP-PJFET incorporating the principles of the present invention.

FIG. 6 is a cross-sectional-perspective view of an N channel JFET incorporating the principles of the present invention.

FIG. 7 is a cross-sectional view taken along the lines VII--VII of FIG. 6.

FIG. 8 is a cross-sectional view of a merged NPN-PJFET incorporating the principles of the present invention.

FIGS. 9-11 are cross-sectional views of a wafer illustrating the transistors at various stages of fabrication.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

A circuit configuration allowing a bipolar transistor to operate to the BV.sub.CBO limit is illustrated in FIGS. 1 and 2 as including bipolar transistor and a junction field effect transistor with the source and collector connected in common and the gate and base connected in common. The high voltage is applied to the drain and the input may be either on the base or emitter of the bipolar transistor and the output taken off the collector. Thus, the present invention is applicable to the common base as well as the common emitter circuit configuration for the bipolar transistor. As illustrated specifically in FIG. 1, an NPN transistor 12 is connected to a N channel junction field effect transistor 14. The bipolar transistor 12 includes a collector 16, a base 18 and an emitter 20 and the NJFET includes a drain 22, a source 24 and a gate 26. The gate 26 is connected to the base 18 and the collector 16 is connected to the source 24. FIG. 2 illustrates a PNP transistor 32 and a PJFET 34. The NPN transistor 32 includes a collector 36, a base 38 and an emitter 40 and the PJFET includes a drain 42, a source 44 and a gate 46. The base 38 and gate 46 are connected together and the collector 36 and the source 44 are connected together.

The operation of the circuit can be understood by consideration of the NPN-NJFET cascade pair of FIG. 1. The high voltage which would be applied to the collector of the simple NPN is now applied to the drain 22 of the NJFET. The voltage drop is across the drain-gate junction of the JFET with only the low JFET gate-source voltage occurring across the bipolar transistor.

For an exact analysis, the following equations apply:

I.sub.ds =I.sub.dss (1+V.sub.gs /V.sub.p).sup.2, v.sub.ds >V.sub.p -V.sub.gs (1) I.sub.ds =I.sub.c (2) V.sub.gs =-V.sub.CB (3) from which I.sub.c =I.sub.dss (1-V.sub.CB /V.sub.p).sup.2 (4) ##EQU1##

The FET geometry is chosen to give I.sub.dss .gtoreq.I.sub.c and BV.sub.dg .gtoreq.BV.sub.CBO. Therefore, Equation 5 becomes:

V.sub.CB .ltoreq.V.sub.p

Thus, when V.sub.p <BV.sub.CEO, the JFET absorbs the high voltage applied at its drain and the maximum voltage which the circuit can withstand between JFET drain and NPN emitter is BV.sub.dg +V.sub.BE. As will be shown below, it is possible to build JFET's whose BV.sub.dg .ltoreq.BV.sub.CBO using the same process steps used to form high voltage PNP's and NPN's, and, thus, the cascade circuitry can be used to extend the device operating voltage to BV.sub.CBO.

While the condition V.sub.p <BV.sub.CEO assures that transistor BV.sub.CEO is not exceeded, it is an overly restrictive condition on V.sub.p for cases where I.sub.c does not go to zero. This can be seen by examination of Equation (5) where the term I.sub.c /I.sub.dss reduces the maximum V.sub.CB below V.sub.p. The term can be set at any desired value by adjusting the geometry (W/L) of the JFET to set I.sub.dss so as to achieve that value.

The PJFET cascade of the PNP of FIG. 2 works in an exactly analogous way and the same equations apply with appropriate change in signs for P type devices.

The circuit yields an output resistance increase as a result of the cascade action. Consequently, it can be used simultaneously to increase the transistor's output resistance and its maximum operating voltage. In this application, the circuit affords the advantages that it requires a minimum number of devices and a minimum of circuit interconnect to accomplish its function.

A PJFET capable of high voltage operation is illustrated in FIG. 3 as including P- channel region 143 having P+ source 144 and P+ drain 142 therein. An N top gate 146 is formed in the P- channel region 143 between the source and drain regions 144 and 142. An N+ bottom gate region 148 is at the bottom of the P- channel region 143 and is connected to the top gate by laterally surrounding N- region 150. The top gate 146 extends beyond the P- channel region 143 and encompasses the source region 144. The bottom gate region 148 has a lateral area sufficient to extend beyond the channel region 143 in all directions.

The P- channel region 143 intersects the top gate region 146 at junction 145 such that the breakdown voltage is at the cylindrical limit which is greater than or equal to the BV.sub.CBO of a PNP transistor. The intersection 147 of the channel region 143 and the bottom gate 148 gives a plane breakdown limit. Thus, high breakdown voltages of these intersections are achieved. It should also be noted that the top gate region 146 encompasses the source region 142 and therefore acts as a channel stop to prevent inversion between the drain and source. As illustrated in FIGS. 3 and 4, the PJFET is dielectrically isolated by an insulative region 102 in a support 100 which may be polycrystalline material. As will be explained more fully below, the connecting region 150 is an N- substrate with the remaining regions being diffused regions.

To achieve the desired characteristics mentioned above, the P- channel region 143 has the same depth and impurity concentration and will be formed during the same process step as the collector region of the PNP bipolar device which is part of the pair provided to produce the schematic of FIG. 2. The top gate region 146 has the same depth and impurity concentration and is formed simultaneously with the base of the PNP transistor. The source and drain regions 142 and 144 have the same depth and impurity concentration and are formed simultaneously with the emitter region of the PNP transistor. The simultaneously formed PJFET and PNP devices may be formed in a separate dielectrically isolated island.

Alternatively, a merged logic configuration may be used to achieve the schematic of FIG. 2. FIG. 5 specifically illustrates a merged logic configuration wherein the top gate region 146 is enlarged to form a top gate region 146 and base region 138. The emitter region 140 is a P+ region formed in the extended combined base/top gate region 138/146 to form a vertical transistor with the common collector/channel region 136/143.

An NJFET capable of high voltage operation is illustrated in FIGS. 6 and 7. An N- channel region 123 includes N+ drain region 122 and N+ source region 124. A P top gate region 126 is formed in the N- channel region 123 between the source and drain regions and a P+ bottom gate 128 is formed below the N- channel region 123. A P- lateral area 130 connects the bottom gate 128 and the top gate 126 and laterally surrounds the channel region 123 separating it from the remainder of the N- island 121. As in the previous embodiment, the P gate region 126 extends laterally beyond the channel region 123 forming a junction 125 therewith and the bottom gate region 128 extends past the channel 123 forming the junction 127 therewith. As discussed above, this increases the breakdown limits of these junctions for the required high voltage performance.

The N- channel region 123 has the same depth and impurity concentration as the collector region of an NPN bipolar transistor which forms the other portion of the pair to construct the circuit of schematic of FIG. 1. The top gate region 126 has the same depth and impurity concentration as the base portion of the NPN transistor and the source and drain regions 124 and 122 have the depth and impurity concentration as the emitter of the NPN transistor. The N- collector region 123 and the remaining portion 121 are the original substrate of the dielectrically isolated islands and the P- region 130 connecting the top and bottom gate is a result of the collector diffusion used to form the PNP of a bipolar transistor complementary to the NPN of the present pair.

As with the PJFET, the NJFET may be formed as a merged structure with its NPN and is specifically illustrated in FIG. 8. The top gate region 126 is extended to form the top gate/base region 126/118. An N+ emitter region 120 is formed in the extended top gate/base region 126/118. The N- channel region also forms the collector region 123/116.

The process of fabrication which allows the formation of the high voltage junction field effect transistors using the steps of forming high voltage bipolar transistors is illustrated in FIGS. 9-11. The formation of dielectrically isolated N- regions 113, 119, 121 and 116 with buried P+ region 149, N+ region 148, P+ region 128 and N+ region 117, respectively, and P- collector region 136, P- channel region 143, P- lateral and connecting region 130, respectively, are produced by well-known methods of the prior art.

A method of forming dielectric isolations with buried regions and back diffused P regions in N islands is described in U.S. Pat. No. 3,865,649 to Beasom. Although this patent describes only an N buried region, the P+ and N+ regions may be selectively formed before the formation of the dielectric isolated layer 102 and the polycrystalline base 100. By using this process, the P- collector region 136 has the depth and impurity concentration as the P- channel region 143. The connecting regions 130 of the NJFET isolate the N- tub 121 from the channel region 123 which is part of the original tub. Although the P- region 130 is shown with portions of the original N- substrate 121 outside thereof and the collector region 136 shows remaining portions of the N- region 115, the diffusion may be carried out so as to eliminate these regions, if desired.

The wafer at this stage of fabrication is illustrated in FIG. 9.

In the next step, the common base and top gate regions of the appropriate pairs are formed. To be more specific, N type impurities are introduced into collector region 136 and channel region 143 and portions of the lateral area 119 to form base region 138 and top gate region 146 simultaneously. This is followed by the introduction of P type impurities to form base region 118 in collector region 116 and top gate region 126 in channel region 123 and connecting region 130. The structure at this stage is illustrated in FIG. 10.

The next step is to simultaneously form the emitter regions of the bipolar transistors and the source and drain regions of the junction field effect transistors. N type impurities are introduced into channel region 123 to form source and drain regions 124 and 122 and into base region 118 to form emitter region 120. Also, N type impurities are introduced into the collector region 116 to form collector contact 115. P type impurities are then introduced to form emitter region 140 in base region 138 and source and drain regions 144 and 142 in channel region 143. The wafer at this stage of fabrication is illustrated in FIG. 11.

It should be noted that it is well-known to introduce impurities by deposition and diffusion or by implantation and diffusion to the final depth. The use of masking layers and other devices are well-known and, therefore, has not been described or illustrated in detail. The devices formed in individually dielectric isolation are interconnected by surface conductors which are insulated from the remainder of the integrated circuit by an insulating layer. For sake of clarity, these elements have not been illustrated.

Although the devices have been illustrated in FIGS. 3-11 as being formed in dielectrically isolated islands, it is evident that they may be formed in other isolated islands. For example, junction isolation may be used as well as the combination of dielectric and junction isolation which may include dielectric isolation along the lateral edges of the tubs and junction isolation at the bottom.

As is evident from the preceding description of the preferred embodiments, the objects of the invention are attained in that a bipolar transistor capable of operating to its BV.sub.CBO limit by use of a high voltage junction field effect transistor which may be fabricated using the steps only necessary to form the bipolar transistor is described. Although the invention has been described and illustrated in detailed it is to be clearly understood that the same is by way of example and illustration only and is not to be taken by way of limitation. The spirit and scope of the invention are to be limited only by the terms of the appended claims.

Claims

1. A method of fabricating a bipolar and junction field effect transistor comprising:

simultaneously forming by the introduction of a first conductivity type dopant a channel region of said junction field effect transistor and a collector region of said bipolar transistor having the same depth and impurity concentration;

simultaneously forming a top gate region of said junction field effect transistor in said channel region and a base region of said bipolar transistor in said collector region of a second conductivity type opposite said first conductivity type; and

simultaneously forming source and drain regions of said junction field effect transistor in said channel region and an emitter region of said bipolar transistor in said base region of said first conductivity type.

2. The method according to claim 1, including forming a buried bottom gate region below said channel region of said second conductivity extending laterally beyond said channel region in all directions.

3. The method according to claim 2, wherein said top gate region is formed to extend beyond said channel region in at least one direction.

4. The method according to claim 3 including forming a second bipolar transistor complementary to said first mentioned bipolar transistor by the steps of forming a collector region of said second conductivity type, forming a base region of said first conductivity type and forming an emitter region of said second conductivity type; and including forming a lateral region laterally surrounding said channel region of said junction field effect transistor simultaneously with the formation of said collector region of said second bipolar transistor.

5. The method according to claim 4, wherein said lateral region and said collector region of said second bipolar transistor are the substrate and said channel region and said collector region of said first mentioned bipolar transistor are formed in said substrate.

6. The method according to claim 4, wherein said lateral region and said collector region of said second bipolar are formed in a substrate of first conductivity type, and said substrate forms said channel region and said collector region of said first mentioned bipolar transistor.