Photodiode-triggered thyristor electrostatic protection device and its fabrication method
By introducing a photodiode structure and connecting it in parallel with a parasitic BJT in a photodiode-triggered silicon-controlled electrostatic protection device, the problem of high trigger voltage and low sustaining voltage of traditional bidirectional silicon-controlled devices in low-voltage optoelectronic chips is solved, achieving more efficient ESD protection and chip reliability.
Patent Information
- Application Number
- CN202310280997.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-03-21
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2043-03-21
AI Technical Summary
Traditional bidirectional thyristor devices cannot effectively reduce the trigger voltage and increase the sustaining voltage in low-voltage optoelectronic chips, resulting in poor ESD protection, easy latch-up effect, and affecting chip reliability.
Design a photodiode-triggered silicon controlled electrostatic discharge (ESD) device. By setting a specific structure of DN-Well region, N-Well region, P-Well region and polysilicon gate on a P-type substrate, a photodiode trigger structure is formed. The photogenerated carriers form the initial ESD discharge path and are connected in parallel with the parasitic BJT to reduce the trigger voltage and increase the sustaining voltage.
It effectively reduces the device's trigger voltage, improves the sustaining voltage and ESD protection capability, prevents latch-up effects, enhances the chip's anti-static capability, and avoids chip damage due to electrostatic discharge.
Smart Images

Figure CN116364713B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of electrostatic discharge (ESD) protection, and in particular to a photodiode-triggered silicon-controlled rectifier (SCR) ESD protection device and its manufacturing method. Background Technology
[0002] With the development of integrated circuit manufacturing processes, the fine metal lines, small pitch, and thin-film technology within highly integrated chips have further reduced device size and thinned oxide layers. Electrostatic discharge (ESD) has become one of the main factors affecting the reliability of integrated circuits. According to statistics, ESD events cause more than 58% of electronic product failures, resulting in economic losses of hundreds of billions of dollars. Therefore, ESD protection devices must be incorporated into integrated circuit products to ensure chip reliability.
[0003] At the input / output (I / O) ports of photoelectric detection chips, both forward and reverse ESD pulses are typically encountered, necessitating bidirectional ESD protection. Diverterless superconducting thyristors (DDSCRs) offer both forward and reverse ESD discharge paths. When the device is turned on, the positive feedback effect of the parasitic PNP and NPN transistors within the device provides high ESD robustness per unit area. Therefore, DDSCRs are widely used in ESD protection for integrated circuits. In practical applications, the trigger voltage of ESD protection devices must be less than 10% of the gate oxide breakdown voltage, while the sustaining voltage must be more than 10% of the normal operating voltage of the protected circuit to effectively protect the core circuit and prevent latch-up. However, traditional DDSCRs rely on avalanche breakdown of the reverse-biased PN junction on the trigger surface for triggering, resulting in a high trigger voltage that cannot be directly applied to low-voltage photoelectric chips. Furthermore, because the two BJTs form a positive feedback loop after the device is turned on, the sustaining voltage is low, making latch-up highly likely and potentially burning out the chip. Therefore, in ESD protection, when designing using the DDSCR structure, it is necessary to reduce the device's trigger voltage so that it is always below 10% of the gate oxide breakdown voltage. The device's sustaining voltage should be increased to 10% above the normal operating voltage of the protected circuit, while also ensuring that the device has a high ESD protection level.
[0004] Cross-sectional view and equivalent circuit diagram of a traditional bidirectional thyristor device, as shown below Figure 1As shown. A bidirectional thyristor has a symmetrical structure and a bidirectional discharge path. When the bidirectional thyristor operates in the forward state, because the N-well (closer to the anode) and P-well form a reverse-biased PN junction, if the voltage between the anode and cathode does not reach the device's trigger voltage, the bidirectional thyristor exhibits high-resistivity characteristics, and the leakage current is negligible, not affecting signal transmission. When the voltage difference between the anode and cathode reaches the trigger voltage of the bidirectional thyristor, avalanche breakdown occurs in the reverse-biased PN junction of N+ (closer to the anode) and P-well. Avalanche multiplied carriers pass through the parasitic resistance R of the P-well. P A pressure drop occurs. When R... P When the voltage drop reaches the base-emitter turn-on voltage of the parasitic NPN transistor, the parasitic NPN transistor turns on and releases the initial ESD current. This causes the parasitic resistance R to flow through the N-Well (closest to the anode). N1 The current increases rapidly, causing the parasitic PNP1 transistor to conduct, thus fully forming the positive feedback loop of the bidirectional thyristor and releasing the ESD current through a low-resistance path. When the bidirectional thyristor operates in the reverse state, since the N-well (near the cathode) and P-well form a reverse-biased PN junction, if the voltage between the cathode and anode does not reach the device's trigger voltage, the bidirectional thyristor exhibits high-resistance characteristics, and the leakage current is negligible and does not affect signal transmission. When the voltage difference between the cathode and anode reaches the trigger voltage of the bidirectional thyristor, the reverse-biased PN junction of N+ (near the cathode) and P-well undergoes avalanche breakdown. Avalanche multiplied carriers pass through the parasitic resistance R of the P-well. P A pressure drop occurs. When R... P When the voltage drop reaches the base-emitter turn-on voltage of the parasitic NPN transistor, the parasitic NPN transistor turns on and releases the initial ESD current. This causes the parasitic resistance R to flow through the N-Well (near the cathode). N2 The current increases rapidly, causing the parasitic PNP2 transistor to conduct, thus fully forming the positive feedback loop of the bidirectional thyristor and releasing ESD current through a low-resistance path. During the process from triggering the bidirectional thyristor to forming the positive feedback loop, the device exhibits negative resistance, causing the current across the device to continuously rise while the voltage continuously falls. Once the voltage drops to the sustaining voltage, the discharge path of the bidirectional thyristor is fully open, and the device operates stably in the low-resistance region. With the continued increase in current, the bidirectional thyristor eventually undergoes thermal breakdown, at which point the device completely fails. Summary of the Invention
[0005] To address the aforementioned technical problems, this invention provides a photodiode-triggered silicon-controlled electrostatic protection device with a simple structure and strong ESD resistance, and also provides a method for its fabrication.
[0006] The technical solution of the present invention to solve the above problems is: a photodiode-triggered silicon-controlled electrostatic protection device, comprising a P-type substrate; The P-type substrate has a first DN-Well region and a second DN-Well region on its left and right sides, respectively. The first DN-Well region is provided with a first N-Well region, a first P-Well region, and a second N-Well region from left to right. The first N-Well region spans between the P-type substrate and the first DN-Well region. The left side of the first N-Well region is flush with the left edge of the P-type substrate. The right side of the first N-Well region is connected to the left side of the first P-Well region. The right side of the first P-Well region is connected to the left side of the second N-Well region. The second N-Well region spans between the first DN-Well region and the P-type substrate. The second DN-Well region is provided with a third N-Well region, a third P-Well region, and a fourth N-Well region from left to right. The third N-Well region spans between the P-type substrate and the second DN-Well region. The second P-Well region is provided between the left side of the third N-Well region and the right side of the second N-Well region. The right side of the third N-Well region is connected to the left side of the third P-Well region. The right side of the third P-Well region is connected to the left side of the fourth N-Well region. The fourth N-Well region spans between the second DN-Well region and the P-type substrate. The first shallow trench isolation region spans between the first N-Well region and the first P-Well region; the first P-Well region contains a first P+ implantation region, and the second shallow trench isolation region spans between the first P-Well region and the second N-Well region; the second N-Well region contains, from left to right, a first N+ implantation region, a first polysilicon gate, a third shallow trench isolation region, a second P+ implantation region, and a fourth shallow trench isolation region, and the second N+ implantation region spans between the second N-Well region and the second P-Well region; The second P-Well region is provided with a second polysilicon gate, a third P+ implantation region, and a third polysilicon gate from left to right; The third N+ implantation region spans between the second P-Well region and the third N-Well region. The third N-Well region is provided with a fifth shallow trench isolation region, a fourth P+ implantation region, a sixth shallow trench isolation region, a fourth polysilicon gate, and a fourth N+ implantation region from left to right. The seventh shallow trench isolation region spans between the third N-Well region and the third P-Well region. The third P-Well region is provided with a fifth P+ implantation region, and the eighth shallow trench isolation region spans between the third P-Well region and the fourth N-Well region. The right side of the eighth shallow trench isolation region is flush with the right edge of the P-type substrate. The first P+ implantation region, the first N+ implantation region, the first polysilicon gate, and the second P+ implantation region are connected together and serve as the anode of the device; The fourth P+ implantation region, the fourth polysilicon gate, the fourth N+ implantation region, and the fifth P+ implantation region are connected together and serve as the cathode of the device. The second polysilicon gate, the third P+ implantation region, and the third polysilicon gate are connected together and share a single potential; The second and third polysilicon gates serve as photosensitive areas.
[0007] The aforementioned photodiode-triggered silicon controlled rectifier (SCR) electrostatic discharge (ESD) device comprises a photodiode triggering structure consisting of a second N+ injection region, a second polysilicon gate, a third P+ injection region, a third polysilicon gate, and a third N+ injection region in the second P-Well region. Under illumination, this triggering structure utilizes the depletion region to generate photogenerated carriers, forming an ESD discharge path for the diode. Simultaneously, it increases the potential of the second P-Well region, promoting the activation of parasitic BJTs formed in the second N-Well region, the second P-Well region, and the third N-Well region, thereby reducing the device's trigger voltage.
[0008] In the aforementioned photodiode-triggered thyristor electrostatic discharge (ESD) device, the left side of the first shallow trench isolation region is flush with the left edge of the P-type substrate and the first N-Well region; the right side of the first shallow trench isolation region is connected to the left side of the first P+ injection region; the right side of the first P+ injection region is connected to the left side of the second shallow trench isolation region; the right side of the second shallow trench isolation region is connected to the left side of the first N+ injection region; the right side of the first N+ injection region is connected to the left side of the first polysilicon gate; the right side of the first polysilicon gate is connected to the left side of the third shallow trench isolation region; the right side of the third shallow trench isolation region is connected to the left side of the second P+ injection region; the right side of the second P+ injection region is connected to the left side of the fourth shallow trench isolation region; and the right side of the fourth shallow trench isolation region is connected to the left side of the second N+ injection region.
[0009] In the aforementioned photodiode-triggered silicon controlled rectifier electrostatic protection device, the right side of the second N+ injection region is connected to the left side of the second polysilicon gate, the right side of the second polysilicon gate is connected to the left side of the third P+ injection region, the right side of the third P+ injection region is connected to the left side of the third polysilicon gate, and the right side of the third polysilicon gate is connected to the left side of the third N+ injection region.
[0010] In the aforementioned photodiode-triggered silicon controlled rectifier (SCR) electrostatic discharge (ESD) device, the right side of the third N+ injection region is connected to the left side of the fifth shallow trench isolation region; the right side of the fifth shallow trench isolation region is connected to the left side of the fourth P+ injection region; the right side of the fourth P+ injection region is connected to the left side of the sixth shallow trench isolation region; the right side of the sixth shallow trench isolation region is connected to the left side of the fourth polysilicon gate; the right side of the fourth polysilicon gate is connected to the left side of the fourth N+ injection region; the right side of the fourth N+ injection region is connected to the left side of the seventh shallow trench isolation region; the right side of the seventh shallow trench isolation region is connected to the left side of the fifth P+ injection region; the right side of the fifth P+ injection region is connected to the left side of the eighth shallow trench isolation region; and the right side of the eighth shallow trench isolation region is flush with the right edge of the fourth N-Well region and the P-type substrate.
[0011] The aforementioned photodiode triggers a silicon controlled rectifier (SCR) electrostatic discharge (ESD) device. When the anode is subjected to positive ESD stress and the cathode is grounded, the second P+ injection region, the second N-Well region, and the second P-Well region constitute a PNP1 transistor. The second N-Well region, the second P-Well region, and the third N-Well region constitute a lateral NPN transistor. The second N+ injection region, the second P-Well region, and the third P+ injection region constitute a first photodiode D1. The third P+ injection region, the second P-Well region, and the third N+ injection region constitute a second photodiode D2. The PNP1 transistor... The base of the PNP1 transistor is connected to the collector of the NPN transistor through the parasitic resistance of the second N-Well region, and the collector of the PNP1 transistor is connected to the base of the NPN transistor through the parasitic resistance of the second P-Well region. Thus, the PNP1 transistor and the NPN transistor form a thyristor structure. Under illumination, the first photodiode D1 absorbs photons to generate photogenerated carriers, forming an initial ESD current path with the second photodiode, increasing the potential of the second P-Well region, thereby turning on the NPN transistor and the PNP1 transistor, forming a thyristor discharge path, and reducing the trigger voltage of the device.
[0012] The aforementioned photodiode triggers a silicon controlled rectifier (SCR) electrostatic discharge (ESD) device. When the cathode is subjected to forward ESD stress and the anode is grounded, the fourth P+ injection region, the third N-Well region, and the second P-Well region constitute a PNP2 transistor. The third N-Well region, the second P-Well region, and the second N-Well region constitute a lateral NPN transistor. The third N+ injection region, the second P-Well region, and the third P+ injection region constitute a second photodiode D2. The third P+ injection region, the second P-Well region, and the second N+ injection region constitute a first photodiode D1. The base of the PNP2 transistor... The collector of the PNP2 transistor and the base of the NPN transistor are connected together through the parasitic resistance of the third N-Well region. The collector of the PNP2 transistor and the base of the NPN transistor are connected together through the parasitic resistance of the second P-Well region. Thus, the PNP2 transistor and the NPN transistor form a thyristor structure. Under illumination, the second photodiode D2 absorbs photons to generate photogenerated carriers, forming an initial ESD current path with the first photodiode, increasing the potential of the second P-Well region, thereby quickly turning on the NPN transistor and the PNP2 transistor, forming a thyristor discharge path, and reducing the trigger voltage of the device.
[0013] A method for fabricating a photodiode-triggered silicon-controlled rectifier (SCR) electrostatic discharge (ESD) protection device includes the following steps: Step 1: Form the first DN-Well region and the second DN-Well region on the P-type substrate; Step 2: Using photolithography, the first to eighth shallow trench isolation regions are formed sequentially from left to right on the P-type substrate; Step 3: Through photolithography, the first N-Well region, the first P-Well region and the second N-Well region are formed from left to right in the first DN-Well region, the second P-Well region is formed on the P-type substrate, and the third N-Well region, the third P-Well region and the fourth N-Well region are formed from left to right in the second DN-Well region. Step 4: Form the first polysilicon gate in the second N-Well region, and form the second and third polysilicon gates from left to right in the second P-Well region, and form the fourth polysilicon gate in the third N-Well region; Step 5: Through photolithography, a first P+ implantation region is formed in the first P-Well region, a second P+ implantation region is formed in the second N-Well region, a third P+ implantation region is formed in the second P-Well region, a fourth P+ implantation region is formed in the third N-Well region, and a fifth P+ implantation region is formed in the third P-Well region. Step 6: Using photolithography, form the first N+ implantation region and the second N+ implantation region in the second N-Well region, and form the third N+ implantation region and the fourth N+ implantation region in the third N-Well region; Step 7: Connect the first P+ implantation region, the first N+ implantation region, the first polysilicon gate, and the second P+ implantation region together to form the anode of the device; connect the fourth P+ implantation region, the fourth polysilicon gate, the fourth N+ implantation region, and the fifth P+ implantation region together to form the cathode of the device; connect the second polysilicon gate, the third P+ implantation region, and the third polysilicon gate together to share a common potential.
[0014] The beneficial effects of this invention are as follows: 1. The second N+ injection region, the second polysilicon gate, the third P+ injection region, the third polysilicon gate, and the third N+ injection region in the second P-Well region of the present invention constitute a photodiode triggering structure. Under illumination, the photogenerated carriers generated by the depletion region of the photodiode under reverse bias voltage are used to form an initial ESD discharge path. That is, the current first flows through the photodiode on the surface of the device, increasing the potential of the second P-Well region, thereby quickly turning on the parasitic BJT and effectively reducing the trigger voltage of the bidirectional thyristor device.
[0015] 2. This invention employs a photodiode triggering structure, which improves the sustaining voltage of the bidirectional thyristor device. The photodiode triggering path on the device surface is connected in parallel with the thyristor positive feedback path. Furthermore, the parasitic PNP3 transistor formed by the first P+ injection region, first P-Well region, second N-Well region, and second P-Well region, and the parasitic PNP4 transistor formed by the second P-Well region, third N-Well region, third P-Well region, and fifth P+ injection region, are also connected in parallel with the thyristor positive feedback path. The parallel connection of multiple electrostatic discharge paths reduces the current gain of the thyristor's main path, increases the device's sustaining voltage, and effectively prevents latch-up. Simultaneously, due to the presence of the parasitic PNP3 and PNP4 transistor release paths, the device's electrostatic discharge capability is enhanced, preventing premature breakdown due to current accumulation and enabling it to withstand higher ESD stress. Therefore, the device's failure current is significantly improved.
[0016] 3. The process of this invention is simple and easy to operate. The photodiode-triggered thyristor electrostatic protection device produced does not violate the layout design rules and does not use layers other than CMOS process, so that the bidirectional thyristor device structure can be applied to the ESD protection design of photodetector chips, effectively protecting the internal core circuit, preventing latch-up effect, and improving the chip's ESD resistance. Attached Figure Description
[0017] Figure 1The diagram shows a cross-sectional view and an equivalent circuit diagram of a traditional bidirectional thyristor structure.
[0018] Figure 2 This is a circuit connection diagram of the photodiode-triggered thyristor electrostatic protection device of the present invention.
[0019] Figure 3 This is the equivalent circuit diagram of the photodiode-triggered silicon-controlled electrostatic protection device for forward ESD protection according to the present invention, i.e., the ESD current release path.
[0020] Figure 4 This is the equivalent circuit diagram of the photodiode-triggered silicon-controlled electrostatic protection device for reverse ESD protection according to the present invention, i.e., the ESD current release path.
[0021] Figure 5 This is a top view of the photodiode-triggered silicon-controlled electrostatic protection device of the present invention. Detailed Implementation
[0022] The present invention will be further described below with reference to the accompanying drawings and embodiments.
[0023] like Figures 1-5 As shown, a photodiode-triggered silicon controlled rectifier (SCR) electrostatic discharge (ESD) protection device is characterized by: including a P-type substrate 101; a first DN-Well region 102 and a second DN-Well region 103 are respectively provided on the left and right sides of the P-type substrate 101; the first DN-Well region 102 is provided with a first N-Well region 104, a first P-Well region 105, and a second N-Well region 106 from left to right; the first N-Well region 104 spans between the P-type substrate 101 and the first DN-Well region 102, the left side of the first N-Well region 104 is flush with the left edge of the P-type substrate 101, the right side of the first N-Well region 104 is connected to the left side of the first P-Well region 105, the right side of the first P-Well region 105 is connected to the left side of the second N-Well region 106, and the second N-Well region 106 spans between the first DN-Well region 102 and the P-type substrate 101.
[0024] The second DN-Well region 103 is provided with a third N-Well region 108, a third P-Well region 109 and a fourth N-Well region 110 from left to right. The third N-Well region 108 spans between the P-type substrate 101 and the second DN-Well region 103. A second P-Well region 107 is provided between the left side of the third N-Well region 108 and the right side of the second N-Well region 106. The right side of the third N-Well region 108 is connected to the left side of the third P-Well region 109. The right side of the third P-Well region 109 is connected to the left side of the fourth N-Well region 110. The fourth N-Well region 110 spans between the second DN-Well region and the P-type substrate 101.
[0025] The first shallow trench isolation region 201 spans between the first N-Well region 104 and the first P-Well region 105; the first P-Well region 105 is provided with a first P+ implantation region 111, and the second shallow trench isolation region 202 spans between the first P-Well region 105 and the second N-Well region 106; the second N-Well region 106 is provided with a first N+ implantation region 112, a first polysilicon gate 209, a third shallow trench isolation region 203, a second P+ implantation region 113, and a fourth shallow trench isolation region 204 from left to right, and the second N+ implantation region 114 spans between the second N-Well region 106 and the second P-Well region 107.
[0026] The second P-Well region 107 is provided with a second polysilicon gate 210, a third P+ implantation region 115, and a third polysilicon gate 211 from left to right; the second polysilicon gate 210 and the third polysilicon gate 211 serve as photosensitive regions.
[0027] The third N+ implantation region 116 spans between the second P-Well region 107 and the third N-Well region 108. The third N-Well region 108 contains, from left to right, a fifth shallow trench isolation region 205, a fourth P+ implantation region 117, a sixth shallow trench isolation region 206, a fourth polysilicon gate 212, and a fourth N+ implantation region 118. The seventh shallow trench isolation region 207 spans between the third N-Well region 108 and the third P-Well region 109.
[0028] The third P-Well region 109 is provided with a fifth P+ implantation region 119, and the eighth shallow trench isolation region 208 spans between the third P-Well region 109 and the fourth N-Well region 110. The right side of the eighth shallow trench isolation region 208 is flush with the right edge of the P-type substrate 101.
[0029] The first P+ injection region 111 is connected to the first metal layer 213, the first N+ injection region 112 is connected to the second metal layer 214, the first polysilicon gate 209 is connected to the third metal layer 215, and the second P+ injection region 113 is connected to the fourth metal layer 216. A metal via 302 is provided on the twelfth metal layer 301 of the metal layer 2. The first metal layer 213, the second metal layer 214, the third metal layer 215, and the fourth metal layer 216 are all connected to the twelfth metal layer 301 through the metal via 302, and are used as the anode of the device.
[0030] The fourth P+ implantation region 117 is connected to the eighth metal layer 220, the fourth polysilicon gate 212 is connected to the ninth metal layer 221, the fourth N+ implantation region 118 is connected to the tenth metal layer 222, the fifth P+ implantation region 119 is connected to the eleventh metal layer 223, and a metal via 306 is provided on the fourteenth metal layer 305. The eighth metal layer 220, the ninth metal layer 221, the tenth metal layer 222, and the eleventh metal layer 223 are all connected to the fourteenth metal layer 305 through the metal via 306, and are used as the cathode of the device.
[0031] The second polysilicon gate 210 is connected to the fifth metal layer 217, the third P+ implantation region 115 is connected to the sixth metal layer 218, the third polysilicon gate 211 is connected to the seventh metal layer 219, and a metal via 304 is provided on the thirteenth metal layer 303. The fifth metal layer 217, the sixth metal layer 218 and the seventh metal layer 219 are all connected to the thirteenth metal layer 303 through the metal via 304 and share a common potential.
[0032] The left side of the first shallow trench isolation region 201 is flush with the left edge of the P-type substrate 101 and the first N-Well region 104. The right side of the first shallow trench isolation region 201 is connected to the left side of the first P+ implantation region 111. The right side of the first P+ implantation region 111 is connected to the left side of the second shallow trench isolation region 202. The right side of the second shallow trench isolation region 202 is connected to the left side of the first N+ implantation region 112. The right side of the first N+ implantation region 112 is connected to the left side of the first polysilicon gate 209. The right side of the first polysilicon gate 209 is connected to the left side of the third shallow trench isolation region 203. The right side of the third shallow trench isolation region 203 is connected to the left side of the second P+ implantation region 113. The right side of the second P+ implantation region 113 is connected to the left side of the fourth shallow trench isolation region 204. The right side of the fourth shallow trench isolation region 204 is connected to the left side of the second N+ implantation region 114.
[0033] The right side of the second N+ implantation region 114 is connected to the left side of the second polysilicon gate 210, the right side of the second polysilicon gate 210 is connected to the left side of the third P+ implantation region 115, the right side of the third P+ implantation region 115 is connected to the left side of the third polysilicon gate 211, and the right side of the third polysilicon gate 211 is connected to the left side of the third N+ implantation region 116.
[0034] The right side of the third N+ implantation region 116 is connected to the left side of the fifth shallow trench isolation region 205. The right side of the fifth shallow trench isolation region 205 is connected to the left side of the fourth P+ implantation region 117. The right side of the fourth P+ implantation region 117 is connected to the left side of the sixth shallow trench isolation region 206. The right side of the sixth shallow trench isolation region 206 is connected to the left side of the fourth polysilicon gate 212. The right side of the fourth polysilicon gate 212 is connected to the left side of the fourth N+ implantation region 118. The right side of the fourth N+ implantation region 118 is connected to the left side of the seventh shallow trench isolation region 207. The right side of the seventh shallow trench isolation region 207 is connected to the left side of the fifth P+ implantation region 119. The right side of the fifth P+ implantation region 119 is connected to the left side of the eighth shallow trench isolation region 208. The right side of the eighth shallow trench isolation region 208 is flush with the right edge of the fourth N-Well region 110 and the P-type substrate 101.
[0035] A method for fabricating a photodiode-triggered silicon-controlled rectifier (SCR) electrostatic discharge (ESD) protection device includes the following steps: Step 1: Form a first DN-Well region 102 and a second DN-Well region 103 in the P-type substrate 101. Specifically: The first DN-Well region 102 and the second DN-Well region 103 are formed on the surface of the P-type substrate 101, and then a silicon dioxide thin film is formed by thermal oxidation to alleviate stress damage caused by silicon nitride formed in subsequent process steps. A silicon nitride layer is deposited using chemical vapor deposition (LPCVD) technology as a stop layer for CMP in subsequent process steps.
[0036] Photoresist is uniformly applied to the wafer, and then exposed and developed. This step is used to define the shallow trench isolation. Silicon nitride, silicon dioxide, and the isolation trenches are then etched to remove the photoresist layer. A layer of silicon dioxide is deposited using chemical vapor deposition (LPCVD), followed by chemical mechanical polishing until the silicon nitride thin film layer is reached. Finally, the silicon nitride thin film layer is removed using hot phosphoric acid wet etching.
[0037] Step 2: First to eighth shallow trench isolation regions 208 are formed on the P-type substrate 101 using photolithography. Specifically: Using field oxygen (LOCOS) isolation technology, a silicon dioxide thin film layer is grown using thermal oxidation as a buffer layer. Then, silicon nitride is deposited using LPCVD technology. Photoresist is applied to the wafer, and photolithography is used to define the first shallow trench isolation region 201, the second shallow trench isolation region 202, the third shallow trench isolation region 203, the fourth shallow trench isolation region 204, the fifth shallow trench isolation region 205, the sixth shallow trench isolation region 206, the seventh shallow trench isolation region 207, and the eighth shallow trench isolation region 208. Then, reactive ions etch away the silicon nitride on the first shallow trench isolation region 201, the second shallow trench isolation region 202, the third shallow trench isolation region 203, the fourth shallow trench isolation region 204, the fifth shallow trench isolation region 205, the sixth shallow trench isolation region 206, the seventh shallow trench isolation region 207, and the eighth shallow trench isolation region 208. Subsequently, field implantation is performed to prevent field activation.
[0038] Step 3: Using photolithography, a first N-Well region 104, a first P-Well region 105, and a second N-Well region 106 are formed in the DN-Well region 102; a second P-Well region 107 is formed on the P-type substrate 101; and a third N-Well region 108, a third P-Well region 109, and a fourth N-Well region 110 are formed in the second DN-Well region 103. Specifically: Photoresist is applied to the wafer to define the first P-Well region 105, the second P-Well region 107, and the third P-Well region 109. Then, high-energy boron ion implantation is performed to form localized P-type regions, and the photoresist layer is removed. Photoresist is then applied to the wafer to define the first N-Well region 104, the second N-Well region 106, the third N-Well region 108, and the fourth N-Well region 110. Then, high-energy phosphorus ion implantation is performed to form localized N-type regions, and the photoresist layer is removed.
[0039] Annealing is performed on the first N-Well region 104, the first P-Well region 105, the second N-Well region 106, the second P-Well region 107, the third N-Well region 108, the third P-Well region 109, and the fourth N-Well region 110 to repair the crystal damage on the silicon surface caused by ion implantation, activate the implanted impurities, and eliminate further diffusion of impurities using the RTP process.
[0040] Step 4: A first polysilicon gate 209 is formed in the second N-Well region 106, a second polysilicon gate 210 and a third polysilicon gate 211 are formed in the second P-Well region 107, and a fourth polysilicon gate 212 is formed in the third N-Well region 108. Specifically: The growth of a sacrificial oxide layer is used to capture defects on the silicon surface. The gate oxide layer, serving as the gate insulating layer for the transistor, is grown by depositing the first polysilicon gate 209, the second polysilicon gate 210, the third polysilicon gate 211, and the fourth polysilicon gate 212 using LPCVD. Photoresist is then used for molding, followed by polysilicon etching, requiring precise shaping of the polysilicon from the photoresist before removing the photoresist layer. Polysilicon oxidation serves as a buffer to isolate the polysilicon from the silicon nitride formed in subsequent steps. A layer of silicon nitride is deposited using LPCVD, followed by silicon nitride etching to leave isolation sidewalls, and precise ion implantation is performed to locate the source and drain regions of the transistor. The second polysilicon gate 210 and the third polysilicon gate 211 serve as photosensitive areas.
[0041] Step 5: Using photolithography, a first P+ implantation region 111 is formed in the first P-Well region 105, a second P+ implantation region 113 is formed in the second N-Well region 106, a third P+ implantation region 115 is formed in the second P-Well region 107, a fourth P+ implantation region 117 is formed in the third N-Well region 108, and a fifth P+ implantation region 119 is formed in the third P-Well region 109. Specifically: Photoresist molding is used to control ion implantation. Shallow-depth, heavily doped boron ion implantation removes the photoresist layer, forming the first P+ implantation region 111, the second P+ implantation region 113, the third P+ implantation region 115, the fourth P+ implantation region 117, and the fifth P+ implantation region 119. The third P+ implantation region 115 is designated as the photosensitive region.
[0042] Step Six: Using photolithography, the first N+ implantation region 112 and the second N+ implantation region 114 are formed in the second N-Well region 106, and the third N+ implantation region 116 and the fourth N+ implantation region 118 are formed in the third N-Well region 108. Specifically: Photoresist molding is used to control ion implantation. Shallow-depth, heavily doped arsenic ion implantation removes the photoresist layer, forming the first N+ implantation region 112, the second N+ implantation region 114, the third N+ implantation region 116, and the fourth N+ implantation region 118. The second N+ implantation region 114 and the third N+ implantation region 116 are designated as photosensitive regions.
[0043] Step 7: Connect the first P+ implantation region 111, the first N+ implantation region 112, the first polysilicon gate 209, and the second P+ implantation region 113 together to form the anode of the device; connect the fourth P+ implantation region 117, the fourth polysilicon gate 212, the fourth N+ implantation region 118, and the fifth P+ implantation region 119 together to form the cathode of the device; connect the second polysilicon gate 210, the third P+ implantation region 115, and the third polysilicon gate 211 together to share a common potential.
[0044] The working principle of this invention is as follows: The second N+ injection region 114, the second polysilicon gate 210, the third P+ injection region 115, the third polysilicon gate 211, and the third N+ injection region 116 in the second P-Well region 107 constitute a photodiode trigger structure. The second polysilicon gate 210, the third P+ injection region 115, and the third polysilicon gate 211 are connected together through the fifth metal layer 217, the sixth metal layer 218, the seventh metal layer 219, the thirteenth metal layer 303, and the metal via 304, sharing a common potential. The first P+ injection region, the first P-Well region, the second N-Well region, and the second P-Well region constitute a parasitic PNP3 transistor, and the second P-Well region, the third N-Well region, the third P-Well region, and the fifth P+ injection region constitute a parasitic PNP4 transistor. Figure 2 and Figure 3 As shown. The presence of this photodiode trigger structure reduces the trigger voltage of the bidirectional thyristor device. That is, under illumination, photogenerated carriers excited in the depletion region of the photodiode form an initial ESD path on the device surface, and current flows through the parasitic resistance R of the second P-Well region 107. P The parasitic NPN transistor, consisting of the second N-Well region 106, the second P-Well region 107, and the third N-Well region 108, is rapidly activated, enabling the rapid formation of the thyristor path and reducing the device's trigger voltage. Simultaneously, the parasitic PNP3 and PNP4 transistors are connected in parallel with the thyristor path, reducing the current gain of the thyristor path, increasing the device's holding voltage, and preventing latch-up during electrostatic discharge (ESD) protection. Furthermore, the presence of the photodiode trigger path and the parallel path of the parasitic PNP3 and PNP4 transistors enhances the device's ESD discharge capability, preventing premature breakdown due to current accumulation and enabling it to withstand higher ESD stress. Therefore, the device's failure current is significantly increased.
[0045] When the anode of the device is subjected to positive ESD stress and the cathode is grounded, the second P+ injection region 113, the second N-Well region 106, and the second P-Well region 107 constitute a PNP1 transistor, and the second N-Well region 106, the second P-Well region 107, and the third N-Well region 108 constitute a lateral NPN transistor. The base of the PNP1 transistor and the collector of the NPN transistor are connected together through the parasitic resistance of the second N-Well region 106, and the collector of the PNP1 transistor and the base of the NPN transistor are connected together through the parasitic resistance of the second P-Well region 107. That is, the PNP1 transistor and the NPN transistor constitute a silicon controlled rectifier (SCR) structure.
[0046] When the cathode of the device is subjected to positive ESD stress and the anode is grounded, the fourth P+ injection region 117, the third N-Well region 108, and the second P-Well region 107 form a PNP2 transistor, and the third N-Well region 108, the second P-Well region 107, and the second N-Well region 106 form a lateral NPN transistor. The base of the PNP2 transistor and the collector of the NPN transistor are connected together through the parasitic resistance of the third N-Well region 108, and the collector of the PNP2 transistor and the base of the NPN transistor are connected together through the parasitic resistance of the second P-Well region 107. That is, the PNP2 transistor and the NPN transistor form a silicon controlled rectifier (SCR) structure.
[0047] Based on the second N+ injection region 114, the second polysilicon gate 210, the third P+ injection region 115, the third polysilicon gate 211, and the third N+ injection region 116 in the second P-Well region 107, a photodiode trigger structure is formed. When the forward ESD current stress reaches the anode and cathode ground potential of the device, the trigger voltage of the device is slightly lower than the reverse avalanche breakdown voltage of the second N+ injection region 114 and the second P-Well region 107. When the reverse ESD current stress reaches the anode and cathode ground potential of the device, the trigger voltage of the device is slightly lower than the reverse avalanche breakdown voltage of the third N+ injection region 116 and the second P-Well region 107. The second polysilicon gate 210 and the polysilicon gate 211 serve as photosensitive regions, utilizing the depletion layer formed by the N+ injection region and the P-Well region for photosensing. Under reverse bias, photon excitation generates a large number of photogenerated carriers, thus forming a drift current from the N+ injection region to the P+ injection region. This current, passing through the parasitic resistance of the P-Well region, rapidly conducts the parasitic NPN transistor formed by the second N-Well region, the second P-Well region, and the third N-Well region, forming the initial ESD current path and reducing the device's trigger voltage. Furthermore, by changing the light intensity, the device's trigger voltage can be controlled, enabling a controllable trigger voltage. Simultaneously, the first P+ injection region, the first P-Well region, the second N-Well region, and the second P-Well region constitute a parasitic PNP3 transistor, while the second P-Well region, the third N-Well region, the third P-Well region, and the fifth P+ injection region constitute a parasitic PNP4 transistor. Because the parasitic PNP3 and PNP4 transistors are connected in parallel with the thyristor path, the current gain of the thyristor path is reduced, the holding voltage of the device is increased, and latch-up effect is prevented during electrostatic discharge (ESD) protection. Furthermore, the presence of the photodiode trigger path and the parallel path of the parasitic PNP3 and PNP4 transistors enhances the device's ESD discharge capability, preventing premature breakdown due to current accumulation. This allows the device to withstand higher ESD stress, thereby increasing the failure current. The generation of photogenerated carriers lowers the device's trigger voltage, preventing the chip from being burned out by ESD due to excessively high trigger voltage, thus improving the robustness of the protected chip.
[0048] In summary, this invention provides a method for fabricating a photodiode-triggered thyristor electrostatic discharge (ESD) protection device. The process is simple and easy to operate. The fabricated photodiode-triggered ESD protection device neither violates layout design rules nor utilizes layers outside of CMOS processes. This allows the bidirectional thyristor device structure to be applied in the ESD protection design of low-voltage optoelectronic chips, effectively protecting the core chip and improving its ESD resistance.
Claims
1. A photodiode-triggered silicon-controlled electrostatic discharge (SCD) protection device, comprising a P-type substrate; characterized in that: The P-type substrate has a first DN-Well region and a second DN-Well region on its left and right sides, respectively. The first DN-Well region is provided with a first N-Well region, a first P-Well region, and a second N-Well region from left to right. The first N-Well region spans between the P-type substrate and the first DN-Well region. The left side of the first N-Well region is flush with the left edge of the P-type substrate. The right side of the first N-Well region is connected to the left side of the first P-Well region. The right side of the first P-Well region is connected to the left side of the second N-Well region. The second N-Well region spans between the first DN-Well region and the P-type substrate. The second DN-Well region is provided with a third N-Well region, a third P-Well region, and a fourth N-Well region from left to right. The third N-Well region spans between the P-type substrate and the second DN-Well region. The second P-Well region is provided between the left side of the third N-Well region and the right side of the second N-Well region. The right side of the third N-Well region is connected to the left side of the third P-Well region. The right side of the third P-Well region is connected to the left side of the fourth N-Well region. The fourth N-Well region spans between the second DN-Well region and the P-type substrate. The first shallow trench isolation region spans between the first N-Well region and the first P-Well region; the first P-Well region contains a first P+ implantation region, and the second shallow trench isolation region spans between the first P-Well region and the second N-Well region; the second N-Well region contains, from left to right, a first N+ implantation region, a first polysilicon gate, a third shallow trench isolation region, a second P+ implantation region, and a fourth shallow trench isolation region, and the second N+ implantation region spans between the second N-Well region and the second P-Well region; The second P-Well region is provided with a second polysilicon gate, a third P+ implantation region, and a third polysilicon gate from left to right; The third N+ implantation region spans between the second P-Well region and the third N-Well region. The third N-Well region is provided with a fifth shallow trench isolation region, a fourth P+ implantation region, a sixth shallow trench isolation region, a fourth polysilicon gate, and a fourth N+ implantation region from left to right. The seventh shallow trench isolation region spans between the third N-Well region and the third P-Well region. The third P-Well region is provided with a fifth P+ implantation region, and the eighth shallow trench isolation region spans between the third P-Well region and the fourth N-Well region. The right side of the eighth shallow trench isolation region is flush with the right edge of the P-type substrate. The first P+ implantation region, the first N+ implantation region, the first polysilicon gate, and the second P+ implantation region are connected together and serve as the anode of the device; The fourth P+ implantation region, the fourth polysilicon gate, the fourth N+ implantation region, and the fifth P+ implantation region are connected together and serve as the cathode of the device. The second polysilicon gate, the third P+ implantation region, and the third polysilicon gate are connected together and share a single potential; The second and third polysilicon gates serve as photosensitive areas.
2. The photodiode-trigged silicon controlled static protection device of claim 1, wherein: The second N+ injection region, the second polysilicon gate, the third P+ injection region, the third polysilicon gate, and the third N+ injection region in the second P-Well region constitute a photodiode trigger structure. Under illumination, the photodiode trigger structure utilizes the depletion region to generate photogenerated carriers, forming an ESD discharge path for the diode. At the same time, it increases the potential of the second P-Well region, promotes the turn-on of the parasitic BJT formed in the second N-Well region, the second P-Well region, and the third N-Well region, and reduces the trigger voltage of the device.
3. The photodiode-triggered thyristor electrostatic protection device according to claim 1, characterized in that: The left side of the first shallow trench isolation region is flush with the left edge of the P-type substrate and the first N-Well region. The right side of the first shallow trench isolation region is connected to the left side of the first P+ implantation region. The right side of the first P+ implantation region is connected to the left side of the second shallow trench isolation region. The right side of the second shallow trench isolation region is connected to the left side of the first N+ implantation region. The right side of the first N+ implantation region is connected to the left side of the first polysilicon gate. The right side of the first polysilicon gate is connected to the left side of the third shallow trench isolation region. The right side of the third shallow trench isolation region is connected to the left side of the second P+ implantation region. The right side of the second P+ implantation region is connected to the left side of the fourth shallow trench isolation region. The right side of the fourth shallow trench isolation region is connected to the left side of the second N+ implantation region.
4. The photodiode-triggered thyristor electrostatic protection device according to claim 3, characterized in that: The right side of the second N+ implantation region is connected to the left side of the second polysilicon gate, the right side of the second polysilicon gate is connected to the left side of the third P+ implantation region, the right side of the third P+ implantation region is connected to the left side of the third polysilicon gate, and the right side of the third polysilicon gate is connected to the left side of the third N+ implantation region.
5. The photodiode-triggered thyristor electrostatic protection device according to claim 4, characterized in that: The right side of the third N+ implantation region is connected to the left side of the fifth shallow trench isolation region. The right side of the fifth shallow trench isolation region is connected to the left side of the fourth P+ implantation region. The right side of the fourth P+ implantation region is connected to the left side of the sixth shallow trench isolation region. The right side of the sixth shallow trench isolation region is connected to the left side of the fourth polysilicon gate. The right side of the fourth polysilicon gate is connected to the left side of the fourth N+ implantation region. The right side of the fourth N+ implantation region is connected to the left side of the seventh shallow trench isolation region. The right side of the seventh shallow trench isolation region is connected to the left side of the fifth P+ implantation region. The right side of the fifth P+ implantation region is connected to the left side of the eighth shallow trench isolation region. The right side of the eighth shallow trench isolation region is flush with the right edge of the fourth N-Well region and the P-type substrate.
6. The photodiode-triggered thyristor electrostatic protection device according to claim 5, characterized in that: When the anode is subjected to positive ESD stress and the cathode is grounded, the second P+ injection region, the second N-Well region, and the second P-Well region constitute a PNP1 transistor; the second N-Well region, the second P-Well region, and the third N-Well region constitute a lateral NPN transistor; the second N+ injection region, the second P-Well region, and the third P+ injection region constitute a first photodiode D1; and the third P+ injection region, the second P-Well region, and the third N+ injection region constitute a second photodiode D2. Furthermore, the base of the PNP1 transistor is connected to the NPN transistor... The collectors are connected together through the parasitic resistance of the second N-Well region, and the collector of the PNP1 transistor and the base of the NPN transistor are connected together through the parasitic resistance of the second P-Well region. That is, the PNP1 transistor and the NPN transistor constitute a thyristor structure. Under illumination, the first photodiode D1 absorbs photons to generate photogenerated carriers, which form an initial ESD current path with the second photodiode, increasing the potential of the second P-Well region, thereby turning on the NPN transistor and the PNP1 transistor, forming a thyristor discharge path, and reducing the trigger voltage of the device.
7. The photodiode-triggered thyristor electrostatic protection device according to claim 6, characterized in that: When the cathode is subjected to positive ESD stress and the anode is grounded, the fourth P+ injection region, the third N-Well region, and the second P-Well region constitute a PNP2 transistor; the third N-Well region, the second P-Well region, and the second N-Well region constitute a lateral NPN transistor; the third N+ injection region, the second P-Well region, and the third P+ injection region constitute a second photodiode D2; and the third P+ injection region, the second P-Well region, and the second N+ injection region constitute a first photodiode D1. Furthermore, the base of the PNP2 transistor is connected to the base of the NPN transistor... The collectors are connected together through the parasitic resistance of the third N-Well region, and the collector of the PNP2 transistor and the base of the NPN transistor are connected together through the parasitic resistance of the second P-Well region. That is, the PNP2 transistor and the NPN transistor form a thyristor structure. Under illumination, the second photodiode D2 absorbs photons to generate photogenerated carriers, forming an initial ESD current path with the first photodiode, increasing the potential of the second P-Well region, thereby quickly turning on the NPN transistor and the PNP2 transistor, forming a thyristor discharge path, and reducing the trigger voltage of the device.
8. A method for manufacturing a photodiode-triggered silicon-controlled rectifier electrostatic protection device according to any one of claims 1-7, comprising the following steps: Step 1: Form the first DN-Well region and the second DN-Well region on the P-type substrate; Step 2: Using photolithography, the first to eighth shallow trench isolation regions are formed sequentially from left to right on the P-type substrate; Step 3: Through photolithography, the first N-Well region, the first P-Well region and the second N-Well region are formed from left to right in the first DN-Well region, the second P-Well region is formed on the P-type substrate, and the third N-Well region, the third P-Well region and the fourth N-Well region are formed from left to right in the second DN-Well region. Step 4: Form the first polysilicon gate in the second N-Well region, and form the second and third polysilicon gates from left to right in the second P-Well region, and form the fourth polysilicon gate in the third N-Well region; Step 5: Through photolithography, a first P+ implantation region is formed in the first P-Well region, a second P+ implantation region is formed in the second N-Well region, a third P+ implantation region is formed in the second P-Well region, a fourth P+ implantation region is formed in the third N-Well region, and a fifth P+ implantation region is formed in the third P-Well region. Step 6: Using photolithography, form the first N+ implantation region and the second N+ implantation region in the second N-Well region, and form the third N+ implantation region and the fourth N+ implantation region in the third N-Well region; Step 7: Connect the first P+ implantation region, the first N+ implantation region, the first polysilicon gate, and the second P+ implantation region together to form the anode of the device; connect the fourth P+ implantation region, the fourth polysilicon gate, the fourth N+ implantation region, and the fifth P+ implantation region together to form the cathode of the device; connect the second polysilicon gate, the third P+ implantation region, and the third polysilicon gate together to share a common potential.
Citation Information
Patent Citations
Semiconductor component and method of triggering avalanche breakdown
CN104425484A
Light-triggered silicon controlled rectifier device
CN106935639A
