Micromechanical device with spark gap and micromechanical switch with integrated ESD protection

Abstract

The invention relates to a micromechanical device with a spark gap (100), with a first electrode (110) on a first substrate (10) and with a second electrode (120), wherein a discharge space (130) is located between the first electrode and the second electrode.

Description

State of the art

[0001] A wide variety of MEMS elements (MEMSs) and ASIC circuits and components (ASICs) are known.

[0002] Almost all of these ASICs and MEMs are vulnerable to damage from electrostatic discharge (ESD) if no countermeasures are included in the design. Isolated objects such as people or machines can become electrically charged through movement or other effects. If they come into direct or indirect contact with MEMS or ASICs, the electrical charge can destroy the MEMs or ASICs.

[0003] Typical ESD classes that simulate the human body use voltages ranging from 250 V (Class 1A) to over 800 V (Class 3B). These high voltages simulate the spark that occurs when a charged person touches an electrically conductive device. Typically, the thin oxides used in ASIC transistors can only reach 10 volts before the oxide collapses and the device is permanently damaged. MEMS devices have similar problems, such as SiO2 layers, but typically the voltage is somewhat higher because the layers are thicker. In MEMS, however, any electrostatic voltage surge can cause damage to the actuators because the high voltages can lead to forces significantly greater than the micromechanical design can withstand.

[0004] Typically, the easiest way to protect an ASIC or MEMS is with bypass diodes. A simple illustration of this is shown in Fig. 1. Here, the diodes are specifically designed for short, high-voltage breakdowns. They provide an electrical short circuit of the electrostatic charge, preventing damage to the internal components of the ASIC or MEMS.

[0005] For MEMS elements, especially MEMS relays, the use of diodes or other semiconductor components for ESD protection has some disadvantages.

[0006] First, most MEMS manufacturing processes don't have the capability to incorporate diode structures. Therefore, either an ASIC with integrated diodes or external diodes must be provided. This increases the cost and complexity of the MEMS device. Second, the diodes add significant parasitic capacitance to the input paths. In a high-frequency relay, this limits performance at higher frequencies (>1 GHz). Therefore, some RF MEMS relays for higher frequencies lack ESD protection.

[0007] Furthermore, it would be advantageous if different diodes with different breakdown voltages could be provided between the various relay inputs. This would require additional effort when using an ASIC, as ASIC processes must be used that provide different protection diodes.

[0008] By adding diodes, the relay loses its galvanic isolation, as a semiconductor now connects all pins to a common ground. However, the goal is to maintain galvanic isolation between the load circuit and the control circuit, even with a MEMS relay. Object of the invention

[0009] It would therefore be desirable if protection structures with different protection voltages could be integrated into the MEMS relay.

[0010] Therefore, it would also be desirable to use protection structures for MEMS relays that enable complete galvanic isolation. Advantages of the invention

[0011] The invention relates to a micromechanical device with a spark gap, with a first electrode on a first substrate and with a second electrode, wherein a discharge space is located between the first electrode and the second electrode.

[0012] The core of the invention is that ESD protection of the micromechanical device is achieved not by means of a diode, but by means of a spark gap between two electrodes. This allows for overvoltage protection while maintaining galvanic isolation.

[0013] As in Fig. As can be seen in Figure 2, an air gap of 0-1µm can be used to achieve a breakdown (or spark) voltage between 0 and 300 V. Therefore, if one wants to create 200 V protection in a micromechanical device, such as a MEMS relay, one needs an electrode gap of 500 nm, according to the diagram. The breakdown voltage of the spark gap depends not only on the electrode gap, but also on the pressure and the type of gas in the discharge space.

[0014] The production of two mutually insulated electrodes is possible in most ASIC and MEMS processes. In a MEMS process, the distance between two MEMS structures is typically greater than 500 nm. Smaller distances can be achieved in ASIC processes, but in most processes, an oxide or other dielectric material is found between the two metal ends. During a discharge, the oxide becomes conductive, and the safety structure itself fails.

[0015] In one embodiment of the invention, an additional metal layer is added to an ASIC layer sequence and a metal interruption is structured without passivation, such as oxide (see Fig. 3). This makes it particularly easy to create a device according to the invention with a spark gap and a suitable electrode spacing. A disadvantage of this arrangement, however, is that electrical breakdown can still occur through the oxide beneath the additional metal. However, the breakdown most likely occurs via air or vacuum. This is initially good, but causes a long-term reliability problem. With each plasma discharge, some material is sputtered off the electrodes. This material is partially redeposited on the oxide and, in the long term, causes an undesirable leakage current path between the two electrodes.

[0016] An advantageous embodiment of the invention provides for a functional layer as a second electrode to be arranged at a distance above the first substrate and to be provided with the second electrode. This embodiment advantageously prevents leakage currents from occurring across the oxide.

[0017] An advantageous embodiment of the invention provides for the creation of a rocker structure and the setting of a small and very well-defined distance between the rocker and the electrode via the geometric arrangement of an electrode under the rocker structure and thus the creation of a flashover structure with a small and defined distance.

[0018] It is also advantageous that, on the one hand, a rocker structure can produce very small distances between two electrode surfaces, but, on the other hand, the area that ensures the electrical insulation between the electrodes can be arranged at a very large lateral distance from the electrodes, thus achieving reliable insulation over the service life.

[0019] Further advantageous embodiments of the invention can be found in the subclaims. drawing Fig. 1 shows a prior art electronic protection circuit against electrostatic discharge. Fig. Figure 2 shows the breakdown voltage in a discharge space filled with nitrogen (N2) as a function of the electrode distance. Fig. 3 shows a micromechanical device according to the invention with a spark gap with two electrodes on a substrate in a first embodiment. Fig. 4 shows a micromechanical device according to the invention with a spark gap with a first electrode on a substrate and a second electrode on an oppositely arranged functional layer in a second embodiment. Fig. 5 shows a micromechanical device according to the invention with a spark gap with a first electrode on a substrate and a second projecting electrode on an oppositely arranged functional layer in a third embodiment. Fig. 6 shows a micromechanical device according to the invention in the form of a switch with a spark gap with a first electrode on a substrate and a second electrode on an oppositely arranged rocker in a fourth embodiment. Fig. 7 shows a micromechanical device according to the invention in the form of a switch with a plurality of spark gaps with first electrodes on a substrate and second electrodes on an oppositely arranged rocker in a fifth embodiment. Description

[0020] Fig. Figure 1 shows a prior art electronic protection circuit against electrostatic discharge. It depicts a single-ended buffer with an input, an output, and two Zener diodes D1 and D2 for ESD protection. Here, the diodes are specifically designed for short high-voltage breakdowns. They ensure an electrical short circuit of the electrostatic charge, preventing damage to the electronic component due to overvoltage.

[0021] Fig. Figure 2 shows the breakdown voltage of nitrogen (N2) as a function of the electrode gap. As can be seen from the diagram, an air gap L of 0 µm < L < 1 µm is suitable for achieving a breakdown (or spark) voltage between 0 V and 300 V. Therefore, if one wants to construct a 200 V protection device, one would need a spark gap with an electrode gap of 500 nm. However, the breakdown voltage of the spark gap depends not only on the size of the gap between the electrodes, but also on the pressure and composition of the gas in the discharge space.

[0022] Fig. 3 shows a micromechanical device according to the invention with a spark gap with two electrodes on a substrate in a first exemplary embodiment. An insulating layer 20, for example an oxide layer, is arranged over a first substrate 10. A spark gap 100 with a first electrode 110 and a second electrode 120 and a discharge space 130 is formed on the insulating layer. The first electrode 110 and the second electrode 120, with an electrode spacing 133, are formed from a metal layer 2. The spark gap is thus arranged parallel to the substrate. Between the electrodes there is a discharge space 130, the extent of which is determined by the electrode spacing 133. The electrode spacing, in addition to the gas composition and the gas pressure in the discharge space, determines the discharge voltage of the spark gap.

[0023] A simple manufacturing method for such a device consists, for example, in adding an additional metal layer 2 to an ASIC layer sequence 1, and structuring a metal interruption with the electrode gap 133 without passivation, such as an oxide. A disadvantage of this arrangement is that an electrical breakdown can still occur through the oxide 4 beneath the additional metal. However, the breakdown most likely occurs via air or vacuum. This is initially desirable, but causes a long-term reliability problem. With each plasma discharge, some material from the electrodes is removed by sputtering. This material is partially redeposited on the surface 5 of the oxide, causing a long-term, undesirable leakage current path between the two metal ends.

[0024] Fig. 4 shows a micromechanical device according to the invention with a spark gap having a first electrode on a substrate and a second electrode on an opposite functional layer in a second exemplary embodiment. An insulating layer 20, for example an oxide layer, is arranged over a first substrate 10. A first electrode 110 is arranged on the insulating layer. A functional layer 13 is bonded to the first substrate using an electrically conductive bonding material 12. For example, an AlGe bonding process is used. The discharge space 130 lies between the first electrode 110 and an opposite part of the functional layer 13, which forms a second electrode 120. The spark gap 100 is thus formed by the first electrode 110, the second electrode 120, and the discharge space 130 between the two electrodes. The spark gap is therefore arranged perpendicular to the substrate. Fig. Figure 4 shows a particularly simple arrangement for this case. The insulation between the first electrode and the functional layer is formed in a region that has an insulation distance 233 laterally from the nearest conductive structures. The insulation distance 233 is preferably at least twice as large as the electrode distance 133 between the first electrode 110 and the second electrode 120.

[0025] Fig. 5 shows a micromechanical device according to the invention with a spark gap with a first electrode on a substrate and a second protruding electrode on an opposite functional layer in a third embodiment. In contrast to the device in Fig. 4, the functional layer 13 is applied to a second substrate 102. The second substrate is hermetically bonded to the first substrate, and the entire area of ​​the ESD protection, in particular the spark gap 100 with the discharge space 130, the first electrode 110, and the second electrode 120, is thus hermetically sealed in a cavity 210, wherein a defined gas or gas mixture 21 is preferably enclosed. The functional layer is preferably constructed from doped silicon. The second electrode 120 is preferably constructed on the functional layer 13 as a protruding electrode 22, which is arranged opposite the first electrode 110. The first electrode 110 is preferably formed from tungsten.

[0026] Fig. Figure 6 shows a micromechanical device according to the invention in the form of a switch with a spark gap having a first electrode on a substrate and a second electrode on an oppositely arranged rocker in a fourth embodiment. In contrast to the device in Fig. 4, a rocker structure 30 is formed here in the functional layer 13. The rocker is deflected, and there is a mechanical and optionally an electrical contact between the rocker 30 and a further contact surface 31 on the first substrate 10 or a component applied to the first substrate. Alternatively, the rocker can also be designed as only a half-rocker. The rocker can be deflected by a preloaded spring 32. Alternatively, the rocker can also be deflected by an actuator electrode 33 on the first substrate and a voltage applied thereto. Alternatively, the rocker can be deflected by a stop structure on the opposite first substrate (not illustrated).

[0027] The rocker can be deflected in one of the three aforementioned ways and can be permanently fixed to the first substrate 10 by means of a current pulse or a laser pulse by means of a fused connection between the rocker 30 and the further contact surface 31.

[0028] With the deflected rocker structure 30, spark gaps with electrode spacings 133 can also be realized for the electrical discharge, which are higher than the basic spacing between the functional layer 13 and the first electrode 110 on the opposite first substrate 10, which is predetermined by the bonding connection by means of the bonding material 12.

[0029] Fig. Figure 7 shows a micromechanical device according to the invention in the form of a switch with a plurality of spark gaps with first electrodes on a substrate and second electrodes on an oppositely arranged rocker in a fifth embodiment. In contrast to the device in Fig.6, a plurality of electrodes 40 are arranged as first electrodes 110 on the first substrate 10 opposite the rocker 30. This allows, for example, two ESD protection structures to be implemented toward one connection. Different positions relative to the pivot point of the rocker also allow different protection voltages to be set.

[0030] In a particularly advantageous embodiment, a plurality of regions consisting of an insulating layer 41 and a conductive electrode layer 42 are arranged on the functional layer 13 of the rocker 30. This allows different second electrodes 120 to be implemented on the rocker and to be wired in a variable manner. This allows a wide variety of ESD protection structures to be implemented between a large number of connections.

[0031] In a favorable embodiment that enables a particularly high ESD voltage, two first electrodes 40, 110 are arranged opposite the functional layer 13, and the functional layer is arranged in an electrically insulating manner. This creates two series-connected flashovers in the event of ESD protection. This allows, on the one hand, a high cumulative ESD protection voltage to be achieved. On the other hand, the sputtering effect can be reduced by distributing the flashover into two flashovers, thus achieving a particularly long service life if small electrode spacings 133 are selected, thus reducing the individual flashover voltages.

[0032] In another cost-effective, very small-sized embodiment, the ESD protection structure is implemented on the same component and in the same cavity and with the same functional structure as a MEMS relay.

[0033] This allows the relay to be manufactured very cost-effectively and in a small size, and at the same time the process fluctuations, for example in the gap distances in the relay and in the ESD protection structure, are partially compensated for. List of reference symbols 1 ASIC layer sequence 2 metal layer 4 Oxide 10 first substrate 12 electrically conductive bonding material 13 Functional layer 20 insulation layer 21 certain atmosphere 22 protruding electrode 31 Contact surface 32 preloaded spring 33 Actuator electrode 34 ? 41 Insulation layer 42 conductive electrode layer conductive layer 100 spark gap 102 second substrate 110 first electrode 120 second electrode 130 Unloading room 133 Electrode gap 210 completed caverns 233 Insulation distance Micromechanical functional layer Switch contact Switching electrode Micromechanical switching part seesaw

Claims

[1] Micromechanical device with a spark gap (100), with a first electrode (110) on a first substrate (10) and with a second electrode (120), wherein a discharge space (130) is located between the first electrode and the second electrode. [2] Micromechanical device according to claim 1, characterized by , that the second electrode (120) is arranged on the first substrate (10) next to the first electrode (110) with an electrode spacing (133). [3] Micromechanical device according to claim 1, characterized by , that the second electrode (120) is formed by means of a functional layer (13) which is arranged over the first substrate and is attached to it by means of an electrically conductive bonding material (12). [4] Micromechanical device according to claim 3, characterized by , that the second electrode (120) is formed as part of the functional layer (13). [5] Micromechanical device according to claim 3, characterized by , that the second electrode (120) is formed on a surface of the functional layer (13), projecting above this surface. [6] Micromechanical device according to any one of claims 3 to 5, characterized by , that the functional layer (13) has a seesaw structure (30) which is deflected in a direction perpendicular to the first substrate (10). [7] Micromechanical device according to claim 6, characterized by , that the rocker structure (30) is electrically contacted at a contact surface (31) on the first substrate (10) and / or mechanically connected to the contact surface (31) by means of a fusion connection. [8] Micromechanical device according to claim 6 or 7, characterized by , that a plurality of first electrodes (40, 110) are arranged on the first substrate (10) and / or a plurality of second electrodes (120) are arranged on the surface of the functional layer (13). [9] Micromechanical device according to any of the preceding claims, characterized by , that the spark gap (100) is arranged in a closed cavern (210) with a specific atmosphere (21). [10] Micromechanical switch, in particular micromechanical relay, with a micromechanical device integrated as ESD protection according to any one of claims 1 to 8.

Images