CMIS Device With Increased Gain

1. Field of the Invention

The present invention relates to a semiconductor device, and more particularly, to a complementary metal oxide semiconductor (CMOS, broadly, CMIS ) device used in an output circuit of a simiconductor integrated cirucit (IC).

2. Description of the Related Art

In semiconductor IC's, a prior art output circuit uses an inverter or a source follower cirucit having an active load. In this prior art output circuit, however, a current steadily flows, which thereby increases the power consumption. Also, the driving power is low due to the limitation of a current flowing through the active load.

In order to reduce the power consumption as well as enhance the driving power, CMOS source follower circuits have been used in output circuits of semiconductor IC's.

A prior art CMOS source follower circuit is comprised of a semiconductor substrate partitioned by a P-type region and a N-type region. Formed in the P-type region are N-type regions as source and a drain for an N-channel enhancement-type transistor. Also, formed in the P-type region are N-type regions as a source and a drain for an N-channel enhancement-type transistor. The drain of the N-channel transistor is connected to a power supply terminal V.sub.DD having a voltage of +10 V, for example, and the drain of the P-channel transistor is connected to a power supply terminal V.sub.SS whose voltage, for example -10 V, is lower than that of the terminal V.sub.DD. Also, gate electrodes of the N-channel transistor and the P-channel transistor are connected commonly to an input node. Further, the sources of the N-chapel transistor and the P-channel transistor are connected commonly to an output node.

Also, the channel of the N-channel transistor is connected to the power supply terminal V.sub.SS, i.e., the back bias voltage of the N-channel transistor is V.sub.SS. 0n the other hand, the channel of the P-channel transistor is connected to the output node, i.e., the back bias voltage of the P-channel transistor is a voltage at the output node. This will be explained later in detail.

In the above-mentioned prior art CMOS source follower circuit, however, when the voltage at the input node is positively large, the threshold voltage of the N-channel transistor is enlarged due to the fact that the channel (back bias) voltage of the N-channel transistor is fixed at V.sub.SS, and the source voltage of the N-channel transistor is increased with respect to the back bias voltage V.sub.SS. As a result, the gain of the CMOS source follower circuit is reduced in the case where the voltage at the input node is large.

Also, in the input/output characteristics of the above-mentioned prior art CMOS source follower circuit, there is a dead gap voltage region where the voltage at the output node is zero when the voltage at the input node is within a range, so that the distortion of the input/output characteristics is large.

It is an object of the present invention to provide a CMOS device (source follower circuit) having an increased gain.

Another object is to reduce the dissipation of the input/output characteristics in the CMIS device (source follower circuit).

According to the present invention, in a CMIS device having a semiconductor substrate including a first N-type region and a P-type region, a second P-type region is formed within the first N-type region, and a second N-type region is formed within the first P-type region. Also, a third N-type region is formed within the second P-type region, and a third P-type region is formed within the second N-type region. An N-channel transistor is formed by the first N-type region as a drain, the second P-type region as a channel, and the third N-region as a source. A P-channel transistor is formed by the first P-type region as a drain, the second N-type region as a channel, and the third P-type region as a source. As a result, the threshold voltage of the N-channel transistor depending upon the difference between the back bias voltage at the second P-type region and the source voltage at the third P-type region is definite, and therefore, the gain is increased.

Also, in the present invention, an input voltage is shifted up by a definite value and is applied to a gate electrode of the N-channel transistor, and simultaneously, the input voltage is shifted down by a definite value and is applied to a gate electrode of the P-channel transistor. As a result, the dead gap voltage region is reduced or extinguished in the input/output characteristics, to thereby reduce the distortion thereof.

The present invention will be clearly understood from the description as set forth below, as compared with the prior art, with reference to the accompanying drawings, wherein:

FIG. 1 is a cross-sectional view illustrating a prior art CMOS device;

FIG. 2 is an equivalent circuit diagram of the device of FIG. 1;

FIG. 3 is a graph showing the input/output characteristics of the device of FIG. 1;

FIG. 4 is a cross-sectional view illustrating a first embodiment of the CMOS device according to the present invention;

FIGS. 5A and 5B are plan views of the N-channel transistor and the P-channel transistor, respectively, of FIG. 4;

FIG. 6 is an equivalent circuit diagram of the device of FIG. 4;

FIG. 7 is a graph showing the input/output characteristics of the device;

FIG. 8 is a circuit diagram illustrating a second embodiment of the CMOS device according to the present invention;

FIGS. 9A, 9B and 9C are diagrams showing the operation of the device of FIG. 8; and

FIG. 10 is a graph showing the input/output characteristics of the device of FIG. 8;

Before the description of the preferred embodiments, a prior art CMOS devices will be explained with reference to FIGS. 1 through 3.

In FIG. 1, which is a cross-sectional view illustrating a prior art CMOS device, reference 1 designates a P-type semiconductor substrate in which an N-type well region 2 is formed. A thick field silicon oxide layer 3 partitions the semiconductor substrate 1 and the N-type well region 2, i.e., partitions an area for forming an N-channel enhancement-type transistor Q.sub.N and an area for forming a P-channel enhancement-type transistor Q.sub.P.

Provied in the area for forming the N-channel transistor Q.sub.N are an offset-type drain region D1 formed by an N-type region 4 and an N.sup.+ -type region 5, an offset-type source region S1 formed by an N-type region 4' and an N.sup.+ -type region 5', a channel region C1 at an interface of the P-type semiconductor substrate 1 between the drain region D1 and the source region S1, and a P.sup.+ -type contact region 6.

Similarly, provided in the area for forming the N-channel transistor Q.sub.P are an offset-type drain region D2 formed by a P-type region 7 and a P.sup.+ -type region 8, source region S2 formed by a P-type region 8', a channel region C2 at an interface of the N-type well 2 between the drain region D2 and the source region S2, and an N.sup.+ -type contact region 9.

Also, provided on the semiconductor substrate 1 and the N-type well 2 are a gate silicon oxide layer 10, a polycrystalline silicon gate electrode G1 formed above the channel region C1, a polycrystalline silicon gate electrode G2 formed above the channel region C2, an insulating layer 11, and an aluminum connection layer 12 formed within contact holes 12a formed in the insulating layer 11.

In FIG. 1, with reference to FIG. 2, which is an equivalent circuit diagram of the device of FIG. 1, the drain region D1 of the N-channel transistor Q.sub.N is connected to a positive power supply terminal V.sub.DD whose voltage is +10 V, for example. Also, the source region S1 of the N-channel transistor Q.sub.N, the source region S2 of the P-channel transostor Q.sub.P, and the P.sup.+ -type contact region 6 are connected to an output node OUT which is connected to an external load RL. Further, the drain region D2 of the P-channel transistor Q.sub.P is connected to a negative power supply terminal V.sub.SS whose voltage is -10 V, for example. Further, the gate electrodes G1 and G2 are connected commonly to an input node IN.

The input/output characteristics of the device of FIG. 1 are shown in FIG. 3. That is, when a voltage V.sub.IN at the input node IN and a voltage V.sub.OUT at the output node OUT are both negative, the operation of the device of FIG. 1 is dependent upon that of the P-channel transistor Q.sub.P, so that there is no problem with respect to the gain. On the other hand, when the voltage V.sub.IN and the voltage V.sub.OUT are both positive, the operation of the device of the FIG. 1 is dependent upon that of the N-channel transistor Q.sub.N, so that there is a problem with respect to the gain. That is, in the N-channel transistor Q.sub.N, since the back bias voltage thereof is applied from the P.sup.+ -type contact region 6, and therefore is fixed at the negative voltage V.sub.SS, a threshold voltage V.sub.TN of the N-channel transistor Q.sub.N is enlarged. In addition, when the input voltage V.sub.IN is higher, the output voltage V.sub.OUT is higher, so that the source voltage of the N-channel transistor Q.sub.N is higher. Thus, the threshold voltage V.sub.TN of the N-channel transistor Q.sub.N is enlarged by a back gate effect. For example, the threshold voltage V.sub.TN of the N-channel transistor Q.sub.N is +4.3 V, while a threshold voltage V.sub.TP of the P-channel transistor Q.sub.P is -1.0 V. Therefore, the gain of the N-channel transistor Q.sub.N is reduced, i.e., the gain of the device of FIG. 1 is reduced in a region where the input voltage V.sub.IN and the output voltage V.sub.OUT are positive.

Also, as shown in FIG. 3, there is a dead gap voltage region. That is, when the input voltage V.sub.IN is within from V.sub.TP to V.sub.TN, for example, within -1.0 V to +4.3 V, the output voltage V.sub.OUT is zero, thus increasing the distortion of the input/output characteristics.

In FIG. 1, note that it is impossible to connect the P.sup.+ -type contact region 6 to the output node OUT special for the N-channel transistor Q.sub.N, since many active elements N-channel transistors other than the N-channel transistor Q.sub.N are usually formed in the semiconductor substrate 1.

Also, in FIG. 1, if the region 1 is formed by a P-type well and the region 2 is formed by a N-type semiconductor substrate, the back bias voltage of the N-channel trasistor Q.sub.N can be V.sub.OUT, but, in this case, the back bias voltage of the P-channel transistor Q.sub.P is fixed at a definite voltage such as V.sub.DD. As a result, the gain of the P-channel transistor Q.sub.P is reduced, i.e., the gain of the device is reduced in a region where the input voltage V.sub.IN and the output voltage V.sub.OUT are negative.

In FIG. 4, which illustrates a first embodiment of the present invention, the N-channel transistor Q.sub.N is formed in the N-type well 2, and the P-channel transistor Q.sub.P is formed in the P-type semiconductor substrate 1.

Provided in the area for forming the N-channel transistor Q.sub.N are an offset type drain region D1 formed by an N-type region 4 and an N.sup.+ -type region 5, a P-type region 30 for defining a channel region C1 at an interface thereof, an N.sup.+ -type source region S1, and a P.sup.+ -type contact region 6'. In this case, the N.sup.+ -type source region S1 and the P.sup.+ -type contact region 6' are manufactured in self-alignment with the polycrystalline silicon gate electrode G1, and therefore, form a double self-alignment (DSA) structure.

Similarly, provided in the area for forming the P-channel transistor Q.sub.P are an offset type drain region D2 formed by a P-type region 7 and a P.sup.+ -type region 8, an N-type region 40 for defining a channel region C2 at an interface thereof, a P.sup.+ -type source region S2, and an N.sup.+ -type contact region 9'. In this case, the P.sup.+ -type source region S2 and the N.sup.+ -type contact region 9' are manufactured in self-alignment with the polycrystalline silicon gate electrode G2, and therefore, form a DSA structure.

Also, provided on the N-type well 2 and the semiconductor substrated 1 are a gate silicon oxide layer 10, a polycrystalline silicon gate electrode G1 formed above the channel region C1, a polycrystalline silicon gate electrode G2 formed above the channel region C2, an insulating layer 11, and an aluminum connection layer 12 formed within contact holes 12a formed in the insulating layer 11.

As illustrated in FIG. 5A, which is a plan view of the N-channel transistor Q.sub.N of FIG. 4, the P+-type contact region 6' is cylindrical, and the source region S1 is of a ring shape surrounding the P.sup.+ -type contact region 6'. Further, the channel region C1 is also of a ring shape surrounding the source region S1. Also, the drain region D1 is of a ring shape surrounding the channel region C1.

Similarly, as illustrated in FIG. 5B, which is a plan view of the P-channel transistor Q.sub.P of FIG. 4, the N.sup.+ -type contact region 9' is cylindrical, and the source region S2 is of a ring shape surrounding the N.sup.+ -type contact region 9'. Further, the channel region C2 is also of a ring shape surrounding the source region S2. Also, the drain region D2 is of a ring shape surrounding the channel region C2.

In FIG. 4, with reference to FIG. 6, which is an equivalent circuit diagram of the device of FIG. 4, the P-type region 30 of the N-channel transistor Q.sub.N as well as the N-type region 40 of the P-channel transistor Q.sub.P is connected to the output node OUT. That is, the P-type region 30 is connected via the P.sup.+ -type region 6' to the output node OUT, and the N-type region 40 is connected via the N.sup.+ -type region 9' to the output node OUT. Therefore, the back gate voltages of the N-channel transistor Q.sub.N and the P-channel transistor Q.sub.P are both V.sub.OUT. As a result, even in the N-channel transistor Q.sub.N, the threshold voltage V.sub.TN depending the source voltage and the back gate voltage thereof becomes small and is definite regardless of the output voltage V.sub.OUT.

The input/output characteristics of the device of FIG. 4 are shown in FIG. 7. That is, when the input voltage V.sub.IN and the voltage V.sub.OUT are both negative, the operation of the device of FIG. 4 is dependent upon that of the P-channel transistor Q.sub.P, so that the input/output characteristics are the same as that of FIG. 3. For example, the threshold voltage V.sub.TP of the P-channel transistor Q.sub.P is -1.0 V. On the other hand, when the voltage V.sub.IN and the voltage V.sub.OUT are both positive, the operation of the device of FIG. 4 is dependent upon that of the N-channel transistor Q.sub.N. In this case, the threshold voltage V.sub.TN of the N-channel transistor Q.sub.N is definite and small, for example, 1.0 V. Therefore, as shown in FIG. 7, the gain is increased.

Also, since the threshold voltage V.sub.TN is reduced, the dead gap voltage region defined by the input voltage V.sub.IN from V.sub.TP to V.sub.TN is reduced from a range from -1.0 V to +4.3 V to a range from -1.0 V to +1.0 V. Thus, the distortion of the input/output characteristics can be reduced.

In FIG. 8, which illustrates a second embodiment of the present invention, a level shifter circuit LS1 is provided between the input node IN and the gate electrode G1 of the N-channel transistor Q.sub.N of FIG. 6. Also, a level shifter circuit LS2 is provided between the input node IN and the gate electrode G2 of the N-channel transistor Q.sub.P of FIG. 6.

The level shifter cirucit LS1 shifts up the input voltage V.sub.IN by a definite value such as V.sub.TP. This level shifter circuit LS1 includes a P-channel transistor Q.sub.1 have the same configuration as the P-channel transistor Q.sub.P of FIG. 1, and a P-channel transistor Q.sub.2 having the same configuration of the P-channel transistor Q.sub.P of FIG. 4. In this case, a definite voltage VB1 is applied to a gate electrode of the transistor Q.sub.1, and therefore, the transistor Q.sub.1 serves as an active load. The transistor Q.sub.2 serves as a source follower, and therefore, the transistor Q.sub.2 generates a voltage V.sub.1 higher than the input voltage V.sub.IN by the value V.sub.TP as shown in FIG. 9A.

The level shifter circuit LS2 shifts down the input voltage V.sub.IN by a definite value such as V.sub.TN. This level shifter circuit LS2 includes an N-channel transistor Q.sub.3 having the same configuration as the P-channel transistor Q.sub.N of FIG. 4, and an N-channel transistor Q.sub.4 having the same configuration of the P-channel transistor Q.sub.N of FIG. 1. In this case, a definite voltage VB2 is applied to a gate electrode of the transistor Q.sub.4, and therefore, the transistor Q.sub.4 serves as an active load. The transistor Q.sub.3 serves as a source follower, and therefore, the transistor Q.sub.3 generates a voltage V.sub.2 lower than the input voltage V.sub.IN by the value V.sub.TN as shown in FIG. 9B.

Thus, a difference in potential of about V.sub.TN +V.sub.TP is generated between the gate electrode G1 of the N-channel transistor Q.sub.N and the gate electrode G2 of the P-channel transistor Q.sub.P. As a result, as shown in FIG. 9C, the gain is about 1.

In more detail, the pinch off point of the N-channel transistor Q.sub.n regarding the input voltage V.sub.IN approximately coincides with that of the P-channel transistor Q.sub.P regarding the input voltage V.sub.IN, and as a result, the distortion of the input/output characteristics is extinguished as shown in FIG. 10.

As explained hereinbefore, according to the present invention, since each threshold voltage of the transistors of a CMIS source follower circuit can be definite and small, the gain can be increased. Also, the distortion of the input/output characteristics can be reduced or extinguished.