TWI785980B - Plasma damage protection device and protection method thereof - Google Patents

Abstract

A plasma damage protection device and a protection method thereof are provided. The plasma damage protection device is disposed in an integrated circuit. The plasma damage protection device includes a switch component and a connection structure. The switch component is coupled between a reference power rail and a pad. The switch component is turned-on or cut-off according to charges on the pad, where the pad is coupled to a protected component. The connection structure is configured to transmission the charges to a control end of the switch component during a Back-End-of-Line process. During the Back-End-of-Line process, the switch component is turned-on according to the charges on the pad.

Description

Plasma damage protection device and protection method

The present invention relates to a plasma damage protection device and protection method, and in particular to a plasma damage protection device and protection method capable of achieving bidirectional polarity protection.

In the semiconductor manufacturing process, it is often necessary to apply plasma to the integrated circuit when performing an etching process, for example. The charges brought by the applied plasma may accumulate in the circuit components and cause damage to the circuit components.

In the prior art, a diode is often arranged next to the protected element to perform the plasma damage protection action. This type of diode usually only protects against charges of a single polarity. Therefore, in the conventional technology, bipolar plasma damage protection action is also achieved by setting bipolar transistors. However, the protective action of the plasma damage achieved by the bipolar transistor is not good in the protective action of positive or negative charges. Therefore, it is an important task for designers in the field to propose a high-efficiency plasma damage protection device.

The invention provides a plasma damage protection device and a protection method, which can improve the protection performance against plasma damage generated in the manufacturing process.

The plasma damage protection device of the present invention is arranged in an integrated circuit. The plasma damage protection device includes a switching element and a transmission structure. The switching element is coupled between the reference power rail and the bonding pad. The switching element is turned on or off according to the charge on the pad coupled to the protected element. The transmission structure is used to transmit the charge on the pad to the control terminal of the switch element in the back-end process. In the back-end process, the switching element is turned on according to the charge on the pad.

The plasma damage protection method of the present invention includes: forming a transmission structure to be coupled to the pad; forming a switch element to be coupled to the transmission structure, the protected element, the pad, and the reference power rail; and, in the back-end process, using The transmission structure transmits the charge on the pad to the control terminal of the switch element, and makes the switch element be turned on according to the charge on the pad.

Based on the above, the present invention provides a transfer structure to transmit the charge on the pad to the control terminal of the switch element in the back-end process, and discharges the charge on the pad by turning on the switch element. And in order to prevent the components in the integrated circuit from being damaged by the electric charge accumulated in the plasma on the welding pad, and maintain the reliability of the integrated circuit.

Referring to FIG. 1 , the plasma damage protection device 100 includes a switching element 110 and a transmission structure 120 . The switch element 110 is coupled between the reference power rail RPWL and the pad PD1. The switching element 110 is turned on or off according to the charge on the pad PD1 , wherein the pad PD1 is coupled to the protected component PC. The protected component PC may be a transistor, or other types of semiconductor components. The transfer structure 120 is coupled to the switch element 110 and the reference power rail RPWL. In this embodiment, the transmission structure 120 can be coupled to the pad PD1 through the connection structure 130 .

In this embodiment, in the back-end process (Back-End-of-Line, BEoL), the transfer structure 120 can be maintained coupled to the pad PD1 through the connection structure 130, so that the charge on the pad PD1 can pass through the transfer structure 120 is transmitted to the control terminal of the switching element 110 . Wherein, the switch element 110 is a transistor T1. The first terminal of the transistor T1 is connected to the pad PD1 , the second terminal and the base terminal of the transistor T1 are jointly coupled to the reference power rail RPWL, and the control terminal of the transistor T1 is coupled to the transmission structure 120 .

It is worth mentioning that the switching element 110 can be turned on or off according to the charge on the pad PD1. Wherein, when the charge accumulated on the pad PD1 by the applied plasma exceeds an expected amount during the manufacturing process, the switch element 110 can be turned on, so as to drain the charge on the pad PD1 to the reference power rail RPWL, and The protected component PC will not be damaged due to excessive charge on the pad PD1.

Incidentally, the reference power rail RPWL can be coupled to the pad PD2. The reference power rail RPWL can receive the reference voltage through the pad PD2.

In addition, in this embodiment, the connection structure 130 can be removed after the back-end process is completed. Moreover, a transmission wire 140 may be formed between the control terminal of the switch element 110 and the reference power rail RPWL. The transmission wire 140 is used to pull down the voltage on the control terminal of the switch element 110 to the reference voltage on the reference power rail RPWL. In this way, the transistor T1 in the switching element 110 can be maintained in a cut-off state.

Referring to FIG. 2 , the plasma damage protection device 200 includes a switching element 210 and a transmission structure 220 . The switching element 210 is constructed by a transistor T1. The transistor T1 is coupled between the pad PD1 and the reference power rail RPWL, and the control terminal of the transistor T1 is coupled to the transmission structure 220 . In this embodiment, the pad PD1 is formed by the metal layer ML, wherein the metal layer ML and the metal layers M1 and M2 can be coupled to each other, and the metal layer M1 can be directly connected to the transistor T1 and the protected component PC.

In addition, the transmission structure 220 is formed in at least one metal layer. In an embodiment, the transmission structure 220 may be formed by metal layers ML, M2, M1. The metal layers ML, M2, M1 are sequentially coupled. Moreover, before the back-end process is completed, the transmission structure 220 can be coupled with the pad PD1 through the connection structure 230 . The connection structure 230 can be constructed with the same metal layers ML, M2, M1 as the transmission structure 220.

In the back-end process, the charge on the pad PD1 can be conducted to the control terminal of the transistor T1 through the connection structure 230 and the transmission structure 220 . If the charge on the pad PD1 is greater than a certain critical amount, the transistor T1 can be turned on correspondingly, and a conduction path is formed between the pad PD1 and the reference power rail RPWL. Through the conduction path provided by the transistor T1 , the charge on the pad PD1 can be effectively discharged, and the protected component PC will not be damaged by the charge on the pad PD1 .

In this embodiment, the reference power rail RPWL is coupled to the pad PD2 and can receive the reference voltage.

On the other hand, when the back-end process is completed, the connection structure 230 can be removed, and the transmission wire 240 can be formed between the control terminal of the transistor T1 and the reference power rail RPWL. The transmission wire 240 is used to transmit the reference voltage on the reference power rail line RPWL to the control terminal of the transistor T1 and maintain the transistor T1 in a cut-off state.

Incidentally, there is no fixed limit to the number of metal layers forming the pad PD1, the connection structure 230, and the transmission structure 220 in this embodiment. The illustration in FIG. category.

In addition, the transistor T1 may be an N-type gold oxide semiconductor field effect transistor.

Referring to FIG. 3A , when the charge applied by the plasma is positive and remains on the pad PD1 . The connection structure 230 and the transmission structure 220 connected to each other through the metal layer ML can transmit the positive charge on the pad PD1 to the control terminal of the transistor T1 . Since the transistor T1 is an N-type transistor, the transistor T1 can be turned on according to the positive charge on its control terminal. Under such conditions, the turned-on transistor T1 can form a channel between the pad PD1 and the reference power rail RPWL, and the charge on the pad PD1 can pass through the channel formed by the transistor T1 to pass through the electric current. The second end of crystal T1 is vented.

In this way, excessive positive charge will not be accumulated on the gate terminal of the protected component PC (such as a transistor), which can effectively reduce the possibility of the protective component PC being burned and maintain the normal operation of the integrated circuit.

Please note here that when the metal layer M1 is being fabricated, the pad PD1 , the connection structure 230 and the metal layer M1 in the transmission structure 220 are connected to each other. When the metal layer M1 is fabricated and enters the process of the metal layer M2 , the metal layer M1 of the connection structure 230 may be removed, and the pad PD1 and the metal layer M1 in the transmission structure 220 are disconnected from each other. Similarly, when the metal layer M2 is being manufactured, the pad PD1, the connection structure 230 and the metal layer M2 in the transmission structure 220 are connected to each other. After the metal layer M2 is fabricated and enters the process of the metal layer ML, part of the metal layer M2 of the connection structure 230 can be removed, and the pad PD1 and the metal layer M2 in the transmission structure 220 are disconnected from each other.

Referring to FIG. 3B , when the charge applied by the plasma is negative and remains on the pad PD1 . The connection structure 230 and the transmission structure 220 connected to each other through the metal layer M2 can transmit the negative charge on the pad PD1 to the control terminal of the transistor T1 . Since the transistor T1 is an N-type transistor, the first end of the transistor T1 has an N-type heavily doped region (N+), and the transistor T1 may have a base end of a P-type well region. Therefore, when the control terminal of the transistor T1 receives negative charges, the P-N junction between the base terminal of the transistor T1 and the first terminal can be turned on to form a channel. Under such conditions, the negative charge on the pad PD1 can be vented through the base terminal of the transistor T1 through the channel formed by the transistor T1, and prevent the protected component PC from being damaged due to the negative charge accumulated on the gate terminal. was burned.

3A and 3B, it can be seen that the plasma damage protection device 200 can effectively provide a discharge path for the charges regardless of the positive or negative charges generated by the plasma, and achieve the purpose of plasma damage protection. Purpose.

Referring to FIG. 4 , in the integrated circuit, after the back-end process is completed, the plasma damage protection device 400 includes a switching element 410 , a transmission structure 420 and a transmission wire 440 . The switch element 410 includes a transistor T1. The transistor T1 may be an N-type transistor, and is coupled between the reference power rail RPWL and the pad PD1. The control terminal of the transistor T1 can be coupled to the transmission structure 420 and the transmission wire 440 . The transmission wire 440 is connected between the reference power rail RPWL and the transmission structure 420, wherein when the reference power rail RPWL receives a reference voltage (such as a reference ground voltage), the transmission wire 440 is used to transmit the reference voltage to the transistor T1 Control terminal. Under this condition, the transistor T1 can be turned off according to the received reference voltage.

It is worth mentioning that in this embodiment, the pad PD1 and the transmission structure 420 are physically isolated from each other. After the back-end process is completed, the connection structure used to connect the pad PD1 and the transmission structure 420 has been removed. In this way, when the integrated circuit is operating normally, the voltage applied on the pad PD1 is not affected by the transmission structure 420 and the transistor T1, so that the protected component PC can maintain normal operation.

FIG. 5 is a flowchart of a plasma damage protection method according to an embodiment of the present invention. In step S510 , in the integrated circuit, a transfer structure is formed to be coupled to the pad, and in step S520 , a switch element is formed to be coupled to the transfer structure, the protected element, the pad and the reference power rail. Next, in the back-end process, the transfer structure transmits the charge on the pad to the control terminal of the switch element, and the switch element is turned on according to the charge on the pad. After the switching element is turned on, the charge on the pad can be discharged through the turned on switching element, and effectively prevent the protected element in the integrated circuit from being damaged by the accumulated charge on the pad.

The implementation details of the above steps have been described in detail in the aforementioned multiple embodiments, and will not be repeated here.

To sum up, the plasma damage protection device of the present invention provides a transfer structure to transfer the charges on the pads in the back-end process, and provides a switch element to be turned on according to the charges on the pads transferred by the transfer structure. Through the turned-on switching element, the accumulated charge on the pad can be effectively vented, and the action of plasma damage protection can be achieved. In the embodiment of the present invention, the switching element can be turned on corresponding to any polarity of charge, and the effect of bidirectional polarity protection can be achieved. It is worth mentioning that the plasma damage protection device of the present invention can cut off the connection between the transmission structure and the pads after the back-end process is completed, so that the plasma damage protection device will not affect the normal operation of the integrated circuit.

100, 200, 400: plasma damage protection device 110, 210, 410: switching elements 120, 220, 420: transmission structure 130, 230: connection structure 140, 240, 440: transmission wire ML, M1, M2: metal layer PC: protected component PD1, PD2: Welding pads RPWL: Reference Power Rail S510~S530: Plasma damage protection steps T1: Transistor

FIG. 1 is a schematic diagram of a plasma damage protection device according to an embodiment of the present invention. FIG. 2 is a schematic diagram of a plasma damage protection device according to another embodiment of the present invention. FIG. 3A and FIG. 3B are schematic diagrams illustrating the operation mode of the plasma damage protection device 200 of the embodiment of the present invention shown in FIG. 2 . FIG. 4 is a schematic diagram of a plasma damage protection device after a back-end process according to an embodiment of the present invention. FIG. 5 is a flowchart of a plasma damage protection method according to an embodiment of the present invention.

100: Plasma damage protection device

110: switch element

120: Transmission structure

130: Connection structure

140: transmission wire

PC: protected component

PD1, PD2: Welding pads

RPWL: Reference Power Rail

T1: Transistor

Claims (13)

A plasma damage protection device, which is set in an integrated circuit, includes: a switch element, coupled between a reference power rail and a welding pad, which is turned on or off according to the charge on the welding pad, wherein the welding The pad is coupled to a protected element; and a transfer structure, coupled to the switch element, is used to transmit the charge on the pad to the control terminal of the switch element, wherein the switch element is controlled according to the charge on the pad conduction. The plasma damage protection device as claimed in claim 1 further includes: a connection structure for connecting the pad and the transmission structure, wherein the connection structure can be removed. The plasma damage protection device as claimed in claim 1, wherein the switching element includes: a transistor having a first end coupled to the pad, a control end of the transistor coupled to the transmission structure, and the transistor Both the second terminal and the base terminal of are coupled to the reference power rail. The plasma damage protection device as claimed in claim 3, wherein the transistor is an N-type transistor, and when the charge on the pad is positive, the transistor is turned on according to the charge on the pad on the control terminal. The plasma damage protection device as claimed in item 3, wherein the transistor is an N-type transistor, and when the charge on the pad is negative, the base terminal of the transistor and the first end of the transistor The formed P-N junction is turned on. The plasma damage protection device as claimed in claim 1, wherein the transmission structure is formed in at least one metal layer. The plasma damage protection device as claimed in claim 1 further includes: a transmission wire for connecting between the reference power rail and the control terminal of the switching element. A method of plasma damage protection, comprising: forming a transmission structure to be coupled to a pad; forming a switching element to be coupled to the transmission structure, a protected element, the pad, and a reference power rail; and In a back-end process, the transmission structure transmits the charge on the pad to the control terminal of the switch element, and the switch element is turned on according to the charge on the pad. The plasma damage protection method as claimed in claim 8 further includes: forming a connection structure to connect the switching element and the transmission structure. The plasma damage protection method as claimed in claim 9 further includes: after the back-end process, removing the connecting structure. The plasma damage protection method as claimed in claim 10 further includes: after the back-end process, forming a transmission wire to be connected between the reference power rail and the control terminal of the switching element. The plasma damage protection method as described in claim 8, wherein the switch element is a transistor, and the plasma damage protection method further includes: when the charge on the pad is positive, the transistor is controlled according to the control terminal The charge on this pad is turned on. The plasma damage protection method as described in Claim 8, further comprising: when the charge on the welding pad is negative, the P-N junction formed between the base terminal of the transistor and the first end of the transistor is closed conduction.

Images