阅读说明：本技术 雪崩鲁棒性ldmos (Avalanche robustness LDMOS ) 是由 S.L.涂 V.库什纳 E.范恩 于 2019-03-22 设计创作，主要内容包括：一种半导体设备包括形成在衬底上方的有源区。所述有源区包括FET和二极管。所述FET包括一个或多个FET指。每个FET指包括FET源极区、FET漏极区和横向FET栅极电极。所述二极管包括一个或多个二极管指。所述二极管指中的每一个包括电耦合到所述FET源极区的二极管阳极区、电耦合到所述FET漏极区的二极管阴极区以及电耦合到所述二极管阳极区并与所述横向FET栅极电极电隔离的横向二极管栅极电极。所述FET指为所述半导体设备的有源指,并且所述二极管指为所述半导体设备的虚设指。所述二极管被配置为钳位跨所述FET漏极区和所述FET源极区产生的最大电压。(A semiconductor device includes an active region formed over a substrate. The active region includes a FET and a diode. The FET includes one or more FET fingers. Each FET finger includes a FET source region, a FET drain region, and a lateral FET gate electrode. The diode includes one or more diode fingers. Each of the diode fingers includes a diode anode region electrically coupled to the FET source region, a diode cathode region electrically coupled to the FET drain region, and a lateral diode gate electrode electrically coupled to the diode anode region and electrically isolated from the lateral FET gate electrode. The FET fingers are active fingers of the semiconductor device and the diode fingers are dummy fingers of the semiconductor device. The diode is configured to clamp a maximum voltage generated across the FET drain region and the FET source region.)

1. A semiconductor apparatus, comprising:

A substrate;

an active region comprising a FET and a diode formed over the substrate;

one or more FET fingers of the FET formed in the active region and having a FET source region, a FET drain region, and a lateral FET gate electrode; and

one or more diode fingers of the diode formed in the active region and having a diode anode region electrically coupled to the FET source region, a diode cathode region electrically coupled to the FET drain region, and a lateral diode gate electrode electrically coupled to the diode anode region and electrically isolated from the lateral FET gate electrode;

wherein:

the one or more FET fingers of the FET are active fingers of the semiconductor device and the one or more diode fingers of the diode are dummy fingers of the semiconductor device; and is

The diode is configured to clamp a maximum voltage generated across the FET drain region and the FET source region.

2. The semiconductor device of claim 1, wherein:

a lateral distance between the lateral FET gate electrode and the FET drain region is greater than a lateral distance between the lateral diode gate electrode and the diode cathode region; and is

A target breakdown voltage of the diode is configured based on the lateral distance between the lateral diode gate electrode and the diode cathode region.

3. The semiconductor device of claim 1, further comprising:

a resistor electrically coupling the lateral diode gate electrode to the diode anode region.

4. The semiconductor device of claim 1, further comprising:

a FET shielding structure electrically coupled to the FET source region and laterally overlapping at least a portion of the lateral FET gate electrode; and

a diode shield structure electrically coupled to the diode anode region and laterally overlapping at least a portion of the lateral diode gate electrode.

5. The semiconductor device of claim 1, further comprising:

a FET shielding structure electrically coupled to the FET source region and laterally overlapping at least a portion of the lateral FET gate electrode; and

a truncated diode shield structure electrically coupled to the diode anode region and not laterally overlapping any portion of the lateral diode gate electrode.

6. The semiconductor device of claim 1, further comprising:

a buried electrical insulator layer between the substrate and the active region, the buried electrical insulator layer extending laterally under the one or more FET fingers and the one or more diode fingers.

7. The semiconductor device of claim 6, further comprising:

an electrical insulator barrier layer between the one or more FET fingers and the one or more diode fingers, the electrical insulator barrier layer extending from a top surface of the buried electrical insulator layer into the active region.

8. The semiconductor device of claim 1, wherein:

the semiconductor device has greater than or equal to ten active fingers for each diode finger that includes a diode.

9. The semiconductor device of claim 1, wherein:

the FET further comprises an FET body region and an FET buried well region;

the diode further comprises a diode body region and a diode buried well region;

the FET source region, the FET drain region, and the diode cathode region are of a first conductivity type; and is

The FET body region, the FET buried well region, the diode body region, the diode buried well region, and the diode anode region are of a second conductivity type.

10. The semiconductor device of claim 9, wherein:

the breakdown voltage of the diode is configured by the distance between the diode cathode region and the diode body region.

11. The semiconductor device of claim 9, wherein:

a lateral distance between the FET drain region and the lateral FET gate electrode is greater than a lateral distance between the diode cathode region and the lateral diode gate electrode.

12. The semiconductor device of claim 9, wherein:

the lateral distance between the FET drain region and the FET buried well region is greater than the lateral distance between the diode cathode region and the diode buried well region.

13. The semiconductor device of claim 9, wherein:

the doping depth, concentration and lateral extent of the FET drain region are substantially the same as the doping depth, concentration and lateral extent of the diode cathode region;

the doping depth, concentration and lateral extent of the FET body region are substantially the same as the doping depth, concentration and lateral extent of the diode body region;

the FET buried well region has a doping concentration substantially the same as the diode buried well region; and is

The doping depth, concentration and lateral extent of the FET source region are different from the doping depth, concentration and lateral extent of the diode anode region.

14. A semiconductor apparatus, comprising:

a substrate;

an active region formed over the substrate, the active region including one or more active fingers and one or more diode fingers, wherein:

each active finger comprises an active lateral gate electrode, a plurality of first active doped regions of a first conductivity type and one or more second active doped regions of a second conductivity type;

each diode finger comprises a diode lateral gate electrode, one or more first diode doped regions of the first conductivity type and one or more second diode doped regions of the second conductivity type;

each active finger includes a region more doped to the first conductivity type than the diode finger;

the active fingers include the same number of regions doped to the second conductivity type as the diode fingers;

the active lateral gate electrode is electrically isolated from the diode lateral gate electrode;

the diode lateral gate electrode is electrically coupled to a diode region in the second diode doping region;

The one or more diode fingers are dummy fingers of the semiconductor device; and is

The one or more diode fingers are configured to clamp a maximum voltage developed across two of the first active doped regions.

15. The semiconductor apparatus of claim 14, further comprising:

a resistor electrically coupling the diode lateral gate electrode to a diode doping region of the second diode doping region.

16. The semiconductor device of claim 14, wherein:

a first one of the first active doped regions is electrically coupled to a first one of the first diode doped regions; and is

A second one of the first active doped regions is electrically coupled to a second one of the second diode doped regions.

17. The semiconductor apparatus of claim 14, further comprising:

an active finger shield structure electrically coupled to an active dopant region of the first active dopant region and laterally overlapping at least a portion of the active lateral gate electrode; and

a truncated diode shield structure electrically coupled to the diode doping region of the second diode doping region and not laterally overlapping any portion of the diode lateral gate electrode.

18. A method for forming a semiconductor device, the method comprising:

providing a substrate;

forming an active region over the substrate, the active region comprising one or more active fingers and one or more diode fingers;

forming a FET within one or more of the active fingers, the FET having a FET source region, a FET drain region, and a lateral FET gate electrode;

forming a diode within one or more of the diode fingers, the diode having a diode anode region, a diode cathode region, and a lateral diode gate electrode;

electrically coupling the FET source region to the diode anode region;

electrically coupling the FET drain region to the diode cathode region; and

electrically coupling the lateral diode gate electrode to the diode anode region;

wherein:

the one or more diode fingers are dummy fingers of the semiconductor device;

the lateral FET gate electrode is electrically isolated from the lateral diode gate electrode; and is

The diode is configured to clamp a maximum voltage generated across the FET drain region and the FET source region.

19. The method of claim 18, further comprising:

electrically coupling the lateral diode gate electrode to the diode anode region using a resistor.

20. The method of claim 18, wherein:

the FET further comprises an FET body region and an FET buried well region;

the diode anode region further comprises a diode body region and a diode buried well region;

the FET source region, the FET drain region, and the diode cathode region are of a first conductivity type; and is

The FET body region, the FET buried well region, the diode body region, and the diode buried well region are of a second conductivity type.

21. The method of claim 20, wherein:

the doping depth, concentration and lateral extent of the FET drain region are substantially the same as the doping depth, concentration and lateral extent of the diode cathode region;

the doping depth, concentration and lateral extent of the FET body region are substantially the same as the doping depth, concentration and lateral extent of the diode body region;

the FET buried well region has a doping concentration substantially the same as the diode buried well region; and is

The doping depth, concentration and lateral extent of the FET source region are different from the doping depth, concentration and lateral extent of the diode anode region.

22. The method of claim 21, wherein:

the breakdown voltage of the diode is configured by the distance between the diode cathode region and the diode body region.

23. The method of claim 21, wherein:

a lateral distance between the FET drain region and the lateral FET gate electrode is greater than a lateral distance between the diode cathode region and the lateral diode gate electrode.

24. The method of claim 21, wherein:

the lateral distance between the FET drain region and the FET buried well region is greater than the lateral distance between the diode cathode region and the diode buried well region.

Background

Metal oxide field effect transistors (MOSFETs) generally include a gate electrode, a source region, a drain region, and a body region. The source and drain regions are of a first conductivity type and the body region is of a second conductivity type. In some MOSFET devices, the first conductivity type is n-type and the second conductivity type is p-type. In other MOSFET devices, this relationship is reversed. When the MOSFET device is in an on-state in response to an applied gate voltage, a channel region is formed in the body region under the gate and between the drain region and the source region. Current flows in the channel region. When the MOSFET device is in the off-state, the channel region is not present and thus current will not flow between the drain and source regions. However, if a reverse bias voltage exceeding the breakdown voltage is applied across the drain and source regions of the MOSFET, a large uncontrolled current flows between the source and drain regions regardless of whether a voltage is applied to the gate electrode. As the reverse bias voltage increases above the breakdown voltage, an avalanche breakdown event may occur. During an avalanche breakdown event, the current flowing through the MOSFET increases at an increased rate and may quickly exceed the maximum current rating of the MOSFET. Thus, the MOSFET is often damaged or destroyed completely.

Lateral diffusion MOSFETs (ldmos) are a type of MOSFET that additionally include a lateral Lightly Doped Drain (LDD) region to increase the breakdown voltage of the semiconductor device compared to that of a typical MOSFET. The LDD regions are of the same conductivity type as the body regions, but are doped to different concentrations. Although LDMOS devices may have increased breakdown voltages compared to MOSFETs, avalanche breakdown events may still occur if the reverse bias voltage exceeds the breakdown voltage of the LDMOS device.

Disclosure of Invention

In some embodiments, a semiconductor device includes a substrate and an active region. The active region includes a FET and a diode formed over the substrate. The FET includes one or more FET fingers formed in the active region, each FET finger having a FET source region, a FET drain region, and a lateral FET gate electrode. The diode includes one or more diode fingers formed in the active region, each diode finger having a diode anode region electrically coupled to the FET source region, a diode cathode region electrically coupled to the FET drain region, and a lateral diode gate electrode electrically coupled to the diode anode region and electrically isolated from the lateral FET gate electrode. The one or more FET fingers of the FET are active fingers of the semiconductor device, and the one or more diode fingers of the diode are dummy fingers (dummyfinger) of the semiconductor device. The diode is configured to clamp a maximum voltage generated across the FET drain region and the FET source region.

In some embodiments, a semiconductor device includes a substrate and an active region formed over the substrate. The active region includes one or more active fingers and one or more diode fingers. Each active finger includes an active lateral gate electrode, two or more first active doped regions of a first conductivity type and one or more second active doped regions of a second conductivity type. Each diode finger includes a diode lateral gate electrode, one or more first diode doped regions of the first conductivity type, and one or more second diode doped regions of the second conductivity type. Each active finger includes a region more doped to the first conductivity type than the diode finger. The active fingers include the same number of regions doped to the second conductivity type as the diode fingers. The active lateral gate electrode is electrically isolated from the diode lateral gate electrode. The diode lateral gate electrode is electrically coupled to a diode region in the second diode doping region. The one or more diode fingers are dummy fingers of the semiconductor device. The one or more diode fingers are configured to clamp a maximum voltage developed across two of the first active doped regions.

In some implementations, a method for forming a semiconductor includes providing a substrate and forming an active region over the substrate. The active region includes one or more active fingers and one or more diode fingers. A FET is formed within one or more of the active fingers. The FET includes an FET source region, an FET drain region, and a lateral FET gate electrode. A diode is formed within one or more of the diode fingers. The diode includes a diode anode region, a diode cathode region, and a lateral diode gate electrode. The FET source region is electrically coupled to the diode anode region. The FET drain region is electrically coupled to the diode cathode region. The lateral diode gate electrode is laterally coupled to the diode anode region. The one or more diode fingers are dummy fingers of the semiconductor device. The lateral FET gate electrode is electrically isolated from the lateral diode gate electrode. The diode is configured to clamp a maximum voltage generated across the FET drain region and the FET source region.

Drawings

Fig. 1 is a simplified elevation view of a portion of a prior art multi-fingered semiconductor device.

Fig. 2 is a simplified elevation view of a portion of a multi-fingered semiconductor device according to some embodiments.

Fig. 3 is a simplified cross-sectional view of an exemplary LDMOS structure in an active finger of a multi-finger semiconductor device according to some embodiments.

Fig. 4-6 are simplified cross-sectional views of exemplary diode structures in dummy fingers of a multi-finger semiconductor device according to some embodiments.

Fig. 7-9 are simplified cross-sectional views of a multi-fingered semiconductor device according to some embodiments.

Fig. 10A-10B are graphs of exemplary experimental results measured using exemplary embodiments disclosed herein.

Detailed Description

Avalanche breakdown may occur in an LDMOS device when the reverse bias voltage generated across the source and drain regions of the LDMOS exceeds the breakdown voltage level of the LDMOS device. When avalanche breakdown occurs, the LDMOS may be damaged or destroyed. One approach to mitigate the risk of avalanche breakdown events is to implement a reverse biased diode in parallel with the LDMOS device. In such embodiments, the cathode region of the diode is electrically coupled to the drain region of the LDMOS device and the anode region of the diode is electrically coupled to the source region of the LDMOS device. The diode is configured to have a target (e.g., desired) diode breakdown voltage level that is less than the breakdown voltage that would cause an avalanche breakdown event to occur in the LDMOS device. The diode conducts a substantial amount of current when the reverse bias voltage exceeds the breakdown voltage of the diode. This current flow reduces or eliminates the reverse bias voltage across the LDMOS device. The LDMOS is prevented from an avalanche breakdown event due to the reduction or elimination of the reverse bias voltage across the LDMOS.

In the exemplary embodiments and methods described herein, it may be advantageous for one or more diodes to be monolithically integrated with one or more LDMOS devices on a common substrate of a semiconductor device. A semiconductor device includes one or more active fingers and one or more dummy fingers (dummy fingers are sometimes also referred to as termination fingers). In some embodiments, a dummy finger is a finger of the multi-finger semiconductor device adjacent to a termination perimeter (e.g., outer perimeter) of the semiconductor device. In some embodiments, there are a plurality of fingers near a termination perimeter of the semiconductor device, the plurality of fingers configured as dummy fingers. In some implementations, a dummy finger is a finger of a multi-finger semiconductor device whose control node (e.g., gate electrode) is not electrically coupled to a corresponding control node (e.g., gate electrode) of an active finger of the semiconductor device. In some embodiments, the LDMOS device is formed within one or more of the active fingers and the diode is formed within one or more of the dummy fingers. The diodes are electrically coupled in parallel with an LDMOS device formed in one or more active fingers of the semiconductor device. One or more diodes formed in the dummy finger are configured to have a breakdown voltage level that is lower than a reverse bias voltage level that would trigger an avalanche event in the LDMOS formed in the active finger. Therefore, it is advantageous to reuse the area of the semiconductor that is normally wasted by the dummy fingers as one or more diodes to extend and protect the lifetime of the LDMOS formed in the active fingers. In addition, since the dummy fingers are similar to the active fingers, the process flow for manufacturing the semiconductor device is advantageously simplified as compared to the process flow for monolithically integrating dissimilar semiconductor devices.

Fig. 1 shows a simplified prior art front view of a multi-fingered semiconductor device 100. Semiconductor device 100 generally includes two or more active fingers 102 and two or more dummy fingers 104. LDMOS devices are formed in the active fingers 102, as alternating source regions S_2-3And a gate electrode G_2-4And a drain region D_2-3As shown. A metallization layer (not shown) may be used to electrically couple corresponding regions of the active fingers 102 together, as schematically illustrated. For example, the metallization layer may connect the gate electrodes G of the active fingers 102_2-4Are electrically coupled together; another metallization layer may connect the source regions S of the active fingers 102_2-3Are electrically coupled together; and so on. Dummy fingers 104 are generally positioned near an outer periphery of semiconductor device 100 to minimize the outer peripheral effect of the process flow for forming semiconductor device 100. Dummy finger 104 includes alternating source regions S_1-nAnd a gate electrode G_1-nAnd a drain region D_1-nThey are similar to the corresponding regions of the active fingers 102 with some exceptions. For example, a region of dummy finger 104 may not be electrically coupled to a corresponding region (e.g., G) within active finger 102₁Usually with G₂Electrically isolated).

Fig. 2 illustrates a simplified front view of a multi-fingered semiconductor device 200 according to some embodiments. For simplicity, certain portions of the semiconductor device 200 that those skilled in the art understand to be present are omitted from fig. 2. Semiconductor device 200 generally includes two or more active fingers 202 and two or more dummy fingers 204. Dummy fingers 204 are typically positioned near the outer perimeter of semiconductor device 200, but in some implementations, dummy fingers 204 are staggered between active fingers 202. A cross-sectional cut line 203 through a portion of the active finger 202 corresponds to a portion of the LDMOS shown and discussed with reference to fig. 3. A cross-sectional cut line 205 through a portion of dummy finger 204 corresponds to the diode shown and discussed with reference to fig. 4-6.

LDMOS devices are formed in the active fingers 202, as alternating source regions S_2-3And a gate electrode G_2-4And a drain region D_2-3As shown. As described with reference to fig. 1, a metallization layer (not shown) may be used to electrically couple corresponding regions of the active fingers 202 together, as schematically shown. In addition, other metallization layers (not shown) may be used to electrically couple the regions of dummy fingers 204 to each other and to the regions of active fingers 202, as schematically illustrated. For simplicity, only three LDMOS devices are shown in active finger 202 (e.g., as being formed by gate electrode G)₂、G₃And G₄As controlled). However, in some embodiments, the active fingers 202 include significantly more LDMOS devices. In this example, the active fingers 202 constitute a single LDMOS device. In other embodiments, two or more separate LDMOS devices are formed in the active fingers 202. For example, in such embodiments, the first LDMOS device and the second LDMOS device are implemented by a set of multiple interleaved fingers and interconnected by a set of alternating conductive paths.

One or more regions of dummy finger 204 are modified to form a gated diode. The diode has alternating anode regions A_1-nAnd a dummy gate electrode G_1-nAnd cathode region C_1-n. Dummy gate electrode G of each diode _1-nIs electrically coupled to the respective anode region A of the diode_1-nTo prevent the formation of a current conducting channel in the body region of the diode. The dummy gate electrode is referred to as a "dummy" gate electrode because it is not configured to receive a gate voltage. The diode formed in dummy finger 204 is electrically coupled in parallel to the LDMOS formed in active finger 202. As shown, the anode region a formed in dummy finger 204_1-nA source region S electrically coupled to the LDMOS formed in the active finger 202_2-3. Also, a cathode region C formed in dummy finger 204_1-nA drain region D electrically coupled to the LDMOS formed in the active finger 202_2-3. Selection based on desired performance, among other factorsRatio of diode to LDMOS device. In some embodiments, there are ten LDMOS devices per diode (ratio of 10-1). In some embodiments, the ratio is 1 to 1, 1-2, 2-1, 2-2, 5-1, etc. In some embodiments, the ratio is 20-1, 50-1, 100-1, and the like. In some embodiments, semiconductor device 200 is designed with the smallest number of dummy fingers 204 possible while still meeting other design, process, and/or performance criteria.

As shown, the surface area of semiconductor device 200 that would normally be wasted by dummy finger 204 is advantageously reused as one or more diodes to extend and protect the lifetime of the LDMOS formed in active finger 202. That is, one or more diodes formed in dummy finger 204 are configured to have a target breakdown voltage level that is lower than the reverse bias voltage level that would trigger an avalanche event in the LDMOS formed in active finger 202. Advantageously, therefore, the diode formed in dummy finger 204 reduces or eliminates the possibility of an avalanche event within the LDMOS formed in active finger 202. In addition, because dummy fingers 204 are similar to active fingers 202, the process flow for manufacturing semiconductor device 200 is advantageously simplified as compared to the process flow for monolithically integrating dissimilar semiconductor devices (e.g., an LDMOS device and a separately formed reverse-biased diode in parallel with the LDMOS device).

Fig. 3 is a simplified cross-sectional view of an exemplary LDMOS device 302 formed in an active finger 202 taken through cut line 203 of fig. 2, according to some embodiments. The LDMOS 302 generally includes a substrate 310, an optional buried insulator layer 312 (as indicated by the dashed lines), and an active region 314 formed over the substrate 310 or over the buried insulator layer 312. The active region 314 may be any of a doped portion of a body of a semiconductor wafer, a local well formed in a larger doped portion of a semiconductor wafer, an active layer of a semiconductor-on-insulator (SOI) wafer, and a local well formed in a SOI wafer. In the example shown, the active region 314 is a thin film formed over the substrate 310 or over the buried insulator layer 312. For simplicity, certain portions of the LDMOS 302 that are understood to be present by those skilled in the art have been omitted from FIG. 3. For example, although understood to be present, one or more metallization layers and interconnects are not shown. The active region 314 generally includes a doped semiconductor region of a first conductivity type and a region of a second conductivity type, for example, formed by implanting impurities into the active region 314. For example, a region of a first conductivity type may be formed by implanting atoms of one dopant, and a region of a second conductivity type may be formed by implanting atoms of another dopant. In some embodiments, the first conductivity type is n-type and the second conductivity type is p-type. In other embodiments, the first conductivity type is p-type and the second conductivity type is n-type. The regions of the first conductivity type include source regions 316, drain regions 318, LDD regions 320, and buried regions 326. The regions of the second conductivity type include a body region 322, a source contact region 323, and a buried well 324. The silicide layer 330 on the lateral polysilicon gate electrode 328 forms a gate bus that is electrically coupled to a gate terminal (G) in the third dimension. The lateral direction is defined herein as being in a plane parallel to the plane of the top surface of the substrate 310. Shield structure 336 is formed as a lateral extension of the source trench lined conductive layer shown on the left side of LDMOS 302, which is filled with a top metal to form source contact electrode 332. The source contact electrode 332 is electrically coupled to a source terminal (S) located in the third dimension. The top metal also forms a drain contact electrode 334 that is electrically coupled to a drain terminal (D) located in the third dimension. Dielectric 338 electrically insulates portions of active region 314. The lateral distance 340 between the edge of the gate electrode 328 and the nearest edge of the drain region 318 is shown for reference with respect to fig. 4-5. The lateral distance 352 between the edge of the buried well 324 and the nearest edge of the drain region 318 is shown for reference with respect to fig. 5-6.

When a sufficient gate voltage is applied to the gate electrode 328, the LDMOS 302 is in a conducting state. In the on state, a conductive channel is formed between the source region 316 and the drain region 318 and current flows in the conductive channel. In general, when no gate voltage is applied to the gate electrode 328, no conduction channel is formed, and the LDMOS 302 is in an off state. However, if a reverse bias voltage is applied across source region 316 and drain region 318 that exceeds the breakdown voltage of LDMOS 302, an avalanche event may occur regardless of whether a gate voltage is applied to gate electrode 328, thereby causing an uncontrolled current to flow between these regions.

Fig. 4 is a simplified cross-sectional view of an exemplary diode 404 formed in dummy finger 204 taken through cut line 205 of fig. 2, according to some embodiments. The diode 404 generally includes a substrate 410, an optional buried insulator layer 412, and an active region 414 formed over the substrate 410 or over the buried insulator layer 412. For simplicity, certain portions of the diode 404 that those skilled in the art understand to be present are omitted from fig. 4. For example, metallization layers and other interconnects that are understood to be present are omitted from fig. 4. In some embodiments, substrate 410 and substrate 310 of fig. 3 are portions of the same substrate. That is, in such embodiments, the LDMOS 302 and the diode 404 are monolithically formed on the same substrate. Similarly, in some embodiments, the optional buried insulator layer 412 and the optional buried insulator layer 312 of fig. 3 are part of the same buried insulator layer.

Active region 414 generally includes a region of the first conductivity type and a region of the second conductivity type in accordance with respective ones of active regions 314 described with reference to fig. 3. In some implementations, the region of the first conductivity type of the diode 404 has the same doping depth, concentration, and lateral extent as the corresponding region of the LDMOS 302. Similarly, in some embodiments, the region of the second conductivity type of the diode 404 has the same doping depth, concentration, and lateral extent as the corresponding region of the LDMOS 302. The regions of the first conductivity type include cathode contact regions 418, LDD regions 420, and buried regions 426. The regions of the second conductivity type include a body region 422, an anode contact region 423, and a buried well 424. A source region similar to source region 316 of LDMOS 302 is omitted from diode 404. The omission of the source regions advantageously results in anode regions comprising body regions 422, buried wells 424 and anode contact regions 423, each of these regions being of the second conductivity type. The top metal forms an anode contact electrode 432 that is electrically coupled to an anode terminal (a) located in the third dimension. The top metal also forms a cathode contact electrode 434 that is electrically coupled to a cathode terminal (C) located in the third dimension. The anode contact region 423 is electrically coupled to the source contact region 323 of the LDMOS 302, and the cathode contact region 418 is electrically coupled to the drain region 318 of the LDMOS 302 (e.g., through the top metal). A lateral gate electrode 428 having a silicide layer 430 is electrically coupled to the anode contact region 423. The gate electrode 428 is electrically isolated from the gate electrode 328 of the LDMOS 302. Dielectric 438 electrically insulates portions of active region 414.

The target breakdown voltage of the diode 404 (e.g., a desired breakdown voltage selected at design time) is configured to be less than the breakdown voltage of the LDMOS 302 by changing the lateral distance between the doped regions of the diode 404, or by making other modifications to be discussed. For example, in the embodiment shown in fig. 4, the lateral distance 440 between the gate electrode 428 and the nearest edge of the cathode contact region 418 is the same as the lateral distance 340 shown in fig. 3. Likewise, the lateral distance 452 between the buried well 424 and the nearest edge of the cathode contact region 418 is the same as the lateral distance 352 shown in fig. 3. However, given the same lateral distances 440 and 452, the target breakdown voltage of diode 404 is reduced by forming diode 404 such that truncated shield structure 436 does not overlap any portion of gate electrode 428 (e.g., as compared to the lateral extent of shield structure 336 over gate electrode 328 shown in fig. 3). In some implementations, no portion of the truncated shield structure 436 is included as part of the diode 404 (e.g., the shield structure 436 is not formed as part of the process flow forming the diode 404). The removal or omission of the shield structure 436 will result in a reduction in the breakdown voltage of the diode 404 due to the lack of carrier depletion previously provided by the shield structure.

In some embodiments, the gate electrode 428 is electrically shorted to the anode contact region 423 by a metal. Electrically shorting gate electrode 428 to anode contact region 423 advantageously prevents the formation of a conductive channel in diode 404, thereby reducing leakage current at high frequency operation of the semiconductor device. In other embodiments, resistor 444 couples gate electrode 428 to anode contact region 423. During an electrostatic discharge event, transient current flow through resistor 444 will cause a voltage drop that inverts the region under gate electrode 428 and forms a conductive channel. By coupling gate electrode 428 to anode contact region 423 through resistor 444, the trigger voltage of diode 404 is advantageously reduced as compared to embodiments of diode 404 in which gate electrode 428 is electrically shorted to anode contact region 423. In addition, avalanche robustness of the diode 404 is advantageously improved because the conductive channel is formed by inversion.

As previously described, the diode 404 is electrically coupled in parallel to the LDMOS 302. As such, a reverse bias voltage applied across the drain region 318 and the source region 316 of the LDMOS302 will also be present across the cathode contact region 418 and the anode contact region 423 of the diode 404. If the reverse bias voltage exceeds the breakdown voltage of diode 404, current will flow from cathode contact region 418 to anode contact region 423. This flow of current will result in a reduction or elimination of the reverse bias voltage across the LDMOS 302. Thus, an avalanche event is prevented from occurring within the LDMOS302, thereby protecting the LDMOS302 from potential damage.

Fig. 5 is a simplified cross-sectional view of an exemplary diode 504 formed in dummy finger 204 taken through cut line 205 of fig. 2, according to some embodiments. In the illustrated embodiment, the target breakdown voltage of diode 504 is modified by reducing the lateral distance 540 between the lateral gate electrode 528 and the nearest edge of cathode contact region 518, as compared to diode 404. The other regions of diode 504 are the same or similar to similarly numbered regions and features of diode 404. For example, body region 522 is the same as or similar to body region 422, buried well 524 is similar to buried well 424, substrate 510 is similar to substrate 410, and so on. Likewise, the electrically coupled and electrically isolated embodiment of diode 504 is the same as or similar to the electrically coupled and electrically isolated embodiment of diode 404. The lateral distance 552 between the edge of buried well 524 and the nearest edge of cathode contact region 518 is shown for reference with respect to fig. 6.

Diode 504 generally includes a substrate 510, an optional buried insulator layer 512, and an active region 514 formed over substrate 510 or over buried insulator layer 512. For simplicity, certain portions of the diode 504 that one skilled in the art would understand to be present are omitted from fig. 5. For example, metallization layers and other interconnects that are understood to be present are omitted from fig. 5. Active region 514 generally includes a region of a first conductivity type and a region of a second conductivity type, depending on the respective regions in active region 314. In some implementations, the region of the first conductivity type of the diode 504 has the same doping depth and concentration as the corresponding region of the LDMOS 302, but a different lateral extent. Similarly, in some implementations, the region of the second conductivity type of the diode 504 has the same doping depth and concentration as the corresponding region of the LDMOS 302, but a different lateral extent. The regions of the first conductivity type include a cathode contact region 518, LDD regions 520 and buried regions 526. The regions of the second conductivity type include a body region 522, an anode contact region 523, and a buried well 524. A source region similar to source region 316 of LDMOS 302 is omitted from diode 504. Omitting the source region creates an anode region within the diode 504. The anode region includes an anode contact region 523, a body region 522, and a buried well 524, each of the second conductivity type. The top metal forms an anode contact electrode 532 that is electrically coupled to an anode terminal (a) located in the third dimension. The top metal also forms a cathode contact electrode 534 that is electrically coupled to a cathode terminal (C) located in the third dimension. The anode contact region 523 is electrically coupled to the source contact region 323 of the LDMOS 302, and the cathode contact region 518 is electrically coupled to the drain region 318 of the LDMOS 302 (e.g., through the top metal). The gate electrode 528 with silicide layer 530 is electrically coupled to the anode contact region 523 directly (e.g., through a top metal) or through an optional resistor 544 that is the same as or similar to resistor 444. The gate electrode 528 is electrically isolated from the gate electrode 328 of the LDMOS 302. Dielectric 538 electrically insulates portions of active region 514. The shield structure 536 overlaps (e.g., extends laterally over) the gate electrode 528. Without further modification, the lateral extension of the shielding structure 536 would increase the target breakdown voltage of the diode 504 compared to the diode 404. However, in the illustrated embodiment, the lateral distance 540 is reduced as compared to both the lateral distance 440 of the diode 404 and the lateral distance 340 of the LDMOS 302. This reduction in lateral distance results in a reduction in the target breakdown voltage of diode 504 as compared to diode 404. In addition, the lateral distance 552 is reduced compared to both the lateral distance 452 of the diode 404 and the lateral distance 352 of the LDMOS 302. This allows the breakdown voltage of diode 504 to be further reduced compared to diode 404. The lateral distance 540 and the lateral distance 552 may be selected at design time such that the target breakdown voltage of the diode 504 is lower than the breakdown voltage of the LDMOS 302.

Fig. 6 is a simplified cross-sectional view of an exemplary diode 604 formed in dummy finger 204 taken through cut line 205 of fig. 2, according to some embodiments. In the illustrated implementation, the breakdown voltage of diode 604 is modified by reducing the lateral distance 652 between buried well 624 and the nearest edge of cathode contact region 618 as compared to the lateral distance 552 of diode 504 as compared to diode 504. The other regions of diode 604 are the same or similar to similarly numbered regions and features of diode 504. For example, the body region 622 is the same as or similar to the body region 522, the buried well 624 is similar to the buried well 524 (except for a modified lateral extent with respect to the buried well 624), the substrate 610 is similar to the substrate 510, and so on. Likewise, the electrically coupled and electrically isolated embodiment of diode 604 is the same as or similar to the electrically coupled and electrically isolated embodiment of diode 504.

Diode 604 generally includes a substrate 610, an optional buried insulator layer 612, and an active region 614 formed over substrate 610 or over buried insulator layer 612. For simplicity, certain portions of the diode 604 that one skilled in the art would understand to be present are omitted from fig. 6. For example, metallization layers and other interconnects that are understood to be present are omitted from fig. 6. The active region 614 generally includes a region of the first conductivity type and a region of the second conductivity type, depending on the respective regions in the active region 314 of the LDMOS 302. In some implementations, the region of the first conductivity type of the diode 604 has the same doping depth and concentration as the corresponding region of the LDMOS 302, but a different lateral extent. Similarly, in some implementations, the region of the second conductivity type of the diode 604 has the same doping depth and concentration as the corresponding region of the LDMOS 302, but a different lateral extent. The regions of the first conductivity type include the cathode contact region 618, the LDD region 620 and the buried region 626. The regions of the second conductivity type include a body region 622, an anode contact region 623, and a buried well 624. Source regions similar to source region 316 are omitted from diode 604. Omitting the source region creates an anode region within the diode 604. The anode region includes an anode contact region 623, a body region 622, and a buried well 624, each of the second conductivity type. The top metal forms an anode contact electrode 632 that is electrically coupled to an anode terminal (a) located in the third dimension. The top metal also forms a cathode contact electrode 634 that is electrically coupled to a cathode terminal (C) located in the third dimension. The anode contact region 623 is electrically coupled to the source contact region 323 of the LDMOS 302, and the cathode contact region 618 is electrically coupled to the drain region 318 of the LDMOS 302 (e.g., through the top metal). Lateral gate electrode 628 with silicide layer 630 is electrically coupled to anode contact region 623 either directly (e.g., through a top metal) or through an optional resistor 644 that is the same as or similar to resistor 444. The gate electrode 628 is electrically isolated from the gate electrode 328 of the LDMOS 302. Dielectric 638 electrically insulates portions of active region 614. The shield structure 636 extends laterally over the gate electrode 628. The target breakdown voltage of diode 604 is further modified from the target breakdown voltage of diode 504 by reducing lateral distance 652 as compared to lateral distance 552 of diode 504. The lateral distance 652 may be designed such that the target breakdown voltage of the diode 604 is lower than the breakdown voltage of the LDMOS 302.

Fig. 7 illustrates a simplified front view of a multi-fingered semiconductor device 700 according to some embodiments. Semiconductor device 700 generally includes a substrate 710, a buried insulator layer 712, and an active region 714. Active region 714 generally includes one or more active fingers 702 and one or more dummy fingers 704. As shown, a protection diode is formed in dummy finger 704 and an LDMOS device is formed in active finger 702. In the implementation shown, an insulating barrier 746 electrically isolates the active fingers 702 from the dummy fingers 704. In some implementations, the insulator barrier region 746 is a Shallow Trench Isolation (STI) region. In other embodiments, the insulator barrier region is a deep trench isolation region (DTI). An electrical connection in a third dimension, schematically shown with a dashed line 762, electrically couples the cathode region (C) of the diode to the drain region (D) of the LDMOS device, and an electrical connection in a third dimension, schematically shown with a dashed line 764, electrically couples the anode region (a) of the diode to the source region (S) of the LDMOS device.

Fig. 8 illustrates a simplified front view of a multi-fingered semiconductor device 800 according to some embodiments. The semiconductor device 800 generally includes a substrate 810, a buried insulator layer 812, and an active region 814. The active area 814 generally includes one or more active fingers 802 and one or more dummy fingers 804. As shown, a protection diode is formed in dummy finger 804 and an LDMOS device is formed in active finger 802. In the embodiment shown, the insulating barrier region 846 electrically isolates adjacent LDMOS devices formed in the active fingers 802, as well as electrically isolates the active fingers 802 from the dummy fingers 804. An electrical connection in a third dimension, schematically shown with dashed line 862, couples the cathode region (C) of the diode to the drain region (D) of the LDMOS device, and an electrical connection in a third dimension, schematically shown with dashed line 864, electrically couples the anode region (a) of the diode to the source region (S) of the LDMOS device.

Fig. 9 illustrates a simplified front view of a multi-fingered semiconductor device 900 according to some embodiments. Semiconductor device 900 generally includes a substrate 910, a buried insulator layer 912, and an active region 914. The active area 914 generally includes one or more active fingers 902 interleaved with one or more dummy fingers 904. As shown, a protection diode is formed in dummy finger 904 and an LDMOS device is formed in active finger 902. In the embodiment shown, insulating barrier region 946 electrically isolates an LDMOS device formed in active finger 802 from an adjacent diode formed in dummy finger 904. An electrical connection in a third dimension, schematically shown with dashed line 962, couples the cathode region (C) of the diode to the drain region (D) of the LDMOS device, and an electrical connection in a third dimension, schematically shown with dashed line 964, electrically couples the anode region (a) of the diode to the source region (S) of the LDMOS device.

Fig. 10A-10B are simplified graphs 1060, 1070 of exemplary experimental results measured using exemplary embodiments disclosed herein. Lines 1062 and 1072 are measured Transmission Line Pulse (TLP) curves generated using an LDMOS similar to LDMOS 302. Line 1064 is a TLP curve generated using a diode similar to diode 504, with gate electrode 428 electrically coupled directly to anode contact region 423 and having a shielding structure similar to shielding structure 336. Line 1074 is a TLP curve generated using diodes similar to diodes 404 or 604. A comparison of line 1064 with line 1062 shows that the indication of the ESD robustness of the diode 504 measured is two times greater than the indication of the ESD robustness of the LDMOS 302 measured. A comparison of line 1074 with line 1072 shows that the measured indication of ESD robustness of diode 404 or 604 (or other similar implementation matching the breakdown voltage of LDMOS 302) is 7 times greater than the measured indication of ESD robustness of LDMOS 302.

Reference has been made in detail to embodiments of the disclosed invention, one or more examples of which are illustrated in the accompanying drawings. Each example has been provided by way of explanation of the present technology, and not limitation of the present technology. Indeed, while the present disclosure has been described in detail with respect to specific embodiments thereof, it will be appreciated that those skilled in the art, upon attaining an understanding of the foregoing, may readily conceive of alterations to, variations of, and equivalents to these embodiments. For instance, features illustrated or described as part of one embodiment, can be used with another embodiment to yield a still further embodiment. It is therefore intended that the present subject matter cover all such modifications and variations as come within the scope of the appended claims and their equivalents. These and other modifications and variations to the present invention may be practiced by those of ordinary skill in the art, without departing from the scope of the present invention, which is more particularly set forth in the appended claims. Furthermore, those of ordinary skill in the art will appreciate that the foregoing description is by way of example only, and is not intended to limit the invention.