1. Field of the Invention
The present invention relates to a structure of a MEMS (micro electro mechanical systems) switch.
2. Description of the Related Art
MEMS is also called a “micro machine” or a “MST (micro system technology)” and refers to a system in which a minute mechanical structure and an electric circuit formed of a semiconductor element are combined. A microstructure has a three-dimensional structure which is partially movable in many cases, unlike a semiconductor element such as a transistor. An electric circuit controls motion of a microstructure or receives and processes a signal from the microstructure. Such a micro machine formed of a microstructure and an electric circuit can have a variety of functions: for example, a sensor, an actuator, and a passive element such as an inductor or a variable capacitor.
A microstructure characterizing a micro machine includes a structural layer having a beam structure in which an end portion thereof is fixed to a substrate and a vacant space between the substrate and the structural layer. A microstructure in which the structural layer is partially movable since there is a space can realize a variety of functions one of which is a switch. A MEMS switch formed of a microstructure is turned on or off with or without physical contact unlike a field-effect switching transistor and thus has advantages such as good isolation when it is off and less insertion loss when it is on.
Further, a MEMS includes not only a microstructure but an electric circuit in many cases; therefore, it is preferable that it can be manufactured applying a process the same as or similar to that of a semiconductor integrated circuit. In the present invention, described is a MEMS switch utilizing a surface micromachine technology for manufacturing a structure with a stack of thin films.
A MEMS switch includes a bridge structure (structural layer) over a substrate and two or more pairs of electrodes facing each other on a surface of the substrate and the substrate side of the bridge structure. By applying a voltage to one pair of electrodes, the bridge structure is pulled down to the substrate side by an electrostatic attractive force and the other pair of electrodes physically come in contact with each other, so that the MEMS switch is turned on (Patent Document 1: Japanese Translation of PCT International Application No. 2005-528751 and Patent Document 2: Japanese Published Patent Application No. 2003-217423).
Further, in order to prevent contact between a pair of electrodes to which a voltage is applied, a stopper for limiting a movable region of a structural layer (also referred to as a bumper or a bump) is generally formed (Patent Document 1).
Different two problems have led to the present invention. The first problem is that a stopper for avoiding charge build-up in an insulating layer is required to be formed (see Patent Document 1) and thus another photomask is required. In order to reduce manufacturing cost, it is preferable that the number of photomasks be reduced to reduce the number of steps; therefore, the stopper is preferably formed without adding a photomask.
The second problem is due to a process. Because of overetching of a sacrificial layer, which occurs in formation of upper electrodes, a structural layer protrudes downward from bottom surfaces of the upper electrodes and thus contact between an upper switch electrode and a lower switch electrode are hindered.
One aspect of the present invention is to solve the second problem first. Then, that can solve the first problem.
As for a micro electro mechanical systems switch (MEMS switch) of the present invention, an upper switch electrode is formed to have a larger area than a lower switch electrode so that contact between the upper switch electrode and the lower switch electrode can be prevented from being hindered even if the structural layer protrudes due to overetching.
Further, as for a MEMS switch of the present invention, an upper drive electrode is formed to have a smaller area than a lower drive electrode so that a portion in which a structural layer protrudes downward from a bottom surface of the upper drive electrode due to the overetching can be a stopper for preventing contact between the upper drive electrode and the lower drive electrode.
Further, as for a MEMS switch of the present invention, an upper switch electrode is formed to have a larger area than a lower switch electrode and an upper drive electrode is formed to have a smaller area than a lower drive electrode, so that contact between the upper switch electrode and the lower switch electrode is prevented from being hindered and a stopper for preventing contact between the upper drive electrode and the lower drive electrode can be provided.
By the present invention, the problem due to a process, in which contact between an upper switch electrode and a lower switch electrode is hindered, can be prevented.
Further, a stopper for preventing contact between an upper electrode and a lower electrode of a switch can be formed without adding a photomask and a step.
Further, since the two problems can be solved at the same time by designing a photomask of the upper electrode, manufacturing cost can be reduced.
In the accompanying drawings:
The embodiment modes and embodiment of the present invention will be described with reference to the accompanying drawings. However, the present invention is not limited to the following description because it will be easily understood by those skilled in the art that various changes and modifications can be made to the modes and their details without departing from the spirit and scope of the present invention. Therefore, the present invention should not be construed as being limited to the description in the following embodiment modes and embodiment. Note that like reference numerals may refer to like parts throughout the drawings in the structure of the present invention.
First, a structure of the micro electro mechanical systems switch (MEMS switch) of the present invention and a manufacturing method thereof are described.
The micro electro mechanical systems switch (MEMS switch) includes a structural layer 116 having a beam structure in which both ends thereof are fixed to a substrate, lower drive electrode layers 112a and a lower switch electrode layer 114a which are provided below the structural layer 116, upper drive electrode layers 112b and an upper switch electrode layer 114b which are provided on a surface of the structural layer 116, which faces the substrate 111.
The upper drive electrode layers 112b and the upper switch electrode layer 114b are arranged to face the lower drive electrode layers 112a and the lower switch electrode layer 114a, respectively. When a potential difference is given between the upper drive electrode layers 112b and the lower drive electrode layers 112a, the structural layer 116 is attracted to the substrate 111 side by an electrostatic attractive force, so that the upper switch electrode layer 114b and the lower switch electrode layer 114a come in contact with each other. Thus, the MEMS switch functions as a switch.
Although the structural layer 116 has a post-and-beam structure in which both ends thereof are fixed to the substrate 111 in
The lower drive electrode layers 112a and the lower switch electrode layer 114a are formed on a surface of the substrate 111 and may be collectively referred to as lower electrode layers 121. Similarly, the upper drive electrode layers 112b and the upper switch electrode layer 114b are formed on a surface of the structural layer 116, which faces the substrate 111, and may be collectively referred to as upper electrode layers 122. Further, the upper drive electrode layers 112b and the lower drive electrode layers 112a may be collectively referred to as drive electrode layers 112 (or pull-down electrode layers), and the upper switch electrode layer 114b and the lower switch electrode layer 114a may be collectively referred to as switch electrode layers 114 (or contact electrode layers or contact point electrode layers).
In the case of driving the switch, the lower switch electrode layer 114a is formed thicker than each of the lower drive electrode layers 112a so that the upper switch electrode layer 114b and the lower switch electrode layer 114a come in contact with each other prior to contact between the upper drive electrode layers 112b and the lower drive electrode layers 112a.
This is because when a voltage is applied between the upper drive electrode layers 112b and the lower drive electrode layers 112a, an attractive force is generated therebetween; therefore, in the case where the distance between each of the upper drive electrode layers 112b and each of the lower drive electrode layers 112a equals the distance between the upper switch electrode layer 114b and the lower switch electrode layer 114a, the upper drive electrode layers 112b and the lower drive electrode layers 112a come in contact with each other more easily than the upper switch electrode layer 114b and the lower switch electrode layer 114a.
Therefore, although not illustrated here, the upper switch electrode layer 114b may be formed thick to protrude downward so that the distance between the upper switch electrode layer 114b and the lower switch electrode layer 114a is reduced.
Next, a method for manufacturing a MEMS switch is described with reference to
First, the lower electrode layers 121 are formed over the substrate 111 as illustrated in
Here, the substrate 111 may be any substrate such as a silicon substrate (semiconductor substrate), a glass substrate, or a metal substrate as long as it is a substrate of which a surface is provided with an insulating layer. It is to be noted that an insulating layer is not illustrated in
A sacrificial layer 123 is formed over the substrate 111 and the lower electrode layers 121 as illustrated in
Then, the upper electrode layers 122 are formed over the sacrificial layer 123 as illustrated in
Then, the structural layer 116 is formed over the sacrificial layer 123 and the upper electrode layers 122 as illustrated in
Next, contact holes are formed in the structural layer 116 as illustrated in
Then, as illustrated in
Finally, as illustrated in
A material of each layer such as the structural layer 116, the sacrificial layer 123, the upper electrode layers 122, or the lower electrode layers 121, which is formed by the above manufacturing method, has a property required for each layer and further, is decided in consideration of a relation with other layers.
For example, the structural layer 116 has to be a material having an insulating property. However, not all materials having an insulating property can be used. Since the structural layer 116 is exposed to an etchant when the sacrificial layer 123 is etched, a condition that the material having an insulating property is not removed by the etchant is required to be considered. Further, the etchant depends on a material of the sacrificial layer.
Specifically, in the case where the sacrificial layer 123 is formed of silicon, hydroxide of alkali metal, such as phosphoric acid, potassium hydroxide, sodium hydroxide, or cesium hydroxide, a tetramethylammonium hydroxide (TMAH) solution, or the like can be used as the etchant. A material which is not removed even when any of the above etchants (and which has an insulating property) has to be used for the structural layer 116 and, for example, silicon oxide can be used as the material.
Further, when the sacrificial layer 123 is etched, the upper electrode layers 122 and the lower electrode layers 121 are also exposed to the etchant; therefore, the upper electrode layers 122 and the lower electrode layers 121 are decided in consideration of a condition that they have conductive properties and are not removed by the etchant used when the sacrificial layer 123 is etched.
In this embodiment mode, for example, the structural layer 116 can be formed of silicon oxide, the sacrificial layer 123 can be formed of tungsten (or polyimide), and the upper and lower electrode layers 122 and 121 can be formed of metal such as tantalum, aluminum, titanium, gold, or platinum. In the case where the sacrificial layer 123 is formed of tungsten, etching of the sacrificial layer 123 may be wet etching with an ammonia peroxide mixture (a solution in which 28 w % of ammonia and 31 w % of oxygenated water are mixed at a ratio of 1:2) or dry etching with a chlorine trifluoride gas. Meanwhile, in the case where the sacrificial layer 123 is formed of polyimide, etching of the sacrificial layer 123 may be wet etching with a commercial polyimide etchant or dry etching with an oxygen plasma.
Next, the relation between the sizes of the upper electrode layers 122 and the lower electrode layers 121 and the structure of the MEMS switch are described.
First, as illustrated in
Then, as illustrated in
When the conductive layer 205 is overetched, the sacrificial layer 204 under the conductive layer 205 is also etched to no small extent. At this time, the amount of the sacrificial layer 204, which is etched, is affected by the etchant of the conductive layer 205 and the condition of the etching (such as a temperature or a flow rate of a gas). It is difficult to satisfy the condition in which the sacrificial layer 204 is not etched at all no matter how high selectivity is.
One of the reasons is that the sacrificial layer 204 is desirably formed using a conductive material or a material to be removed easily.
Because of the structure of the MEMS switch, by completely removing the sacrificial layer 204, the upper electrode layers and the lower electrode layers can come in contact with each other. Therefore, if even a small part of the sacrificial layer 204 is left on a surface of the switch electrode layer, the switch is not turned on. In order to avoid such a situation, the sacrificial layer 204 is preferably formed using a material to be removed easily so that it can be completely removed when being etched or using a conductive material so that defective connection is not caused even if it cannot be completely removed when being etched.
As the former, that is, a material to be removed easily, a resist and polyimide are given; however, they are easily etched by any etchant and thus it is significantly difficult to set selectivity between the conductive layer 205 and the sacrificial layer 204 to be high when the conductive layer 205 is etched.
As the latter, that is, a conductive material, metal and a semiconductor added with an impurity are given. However, the upper electrode layers 222 are required to have conductive properties and a conductive material can be removed by a similar etchant in many cases; thus, also in this case, it is significantly difficult to set selectivity between the conductive layer 205 and the sacrificial layer 204 to be high.
For example, the case is described, in which the sacrificial layer 204 is formed of tungsten, the conductive layer 205 is formed of a stack of aluminum and titanium (100 nm-thick titanium over 300 nm-thick aluminum), and the conductive layer 205 is subjected to dry etching using a mixed gas of boron trichloride (BCl3) and chlorine (Cl2). In this case, conditions for etching the conductive layer 205 are as follows: the IPC power is 450 W, the bias power is 100 W, the flow rate of boron trichloride is 60 sccm, the flow rate of chlorine is 20 sccm, the pressure in a chamber is 1.9 Pa, and the standard etching time period of the conductive layer 205 is 150 seconds. When overetching of 100% with respect to the standard etching time period is performed (that is to say, when etching is performed for twice the time period of the standard time period), tungsten of the sacrificial layer 204 is etched by approximately 100 nm.
It is needless to say that although overetching is preferably small in normal etching, in the case where complete insulation is required as in processing of the conductive layer 205, the overetching time period is set to be longer. Further, the overetching time period in the case of aiming for the complete insulation varies greatly depending on a material forming the conductive layer 205. The overetching time period is approximately 10 to 250% of the required standard etching time period, preferably 50 to 200% of the required standard etching time period and more preferably 90 to 110% of the required standard etching time period.
Thus, when the conductive layer 205 is etched to form the upper electrode layers 222, a step 208a, a step 208b, and a step 208c are generated in the sacrificial layer 204 due to overetching in processing of the conductive layer 205 as illustrated in
A structural layer 209 is formed over the sacrificial layer 204 and the upper electrode layers 222 as illustrated in
Here, assuming that an upward direction from the surface of the substrate 201 is a positive direction, the protrusions 211a, 211b, and 211c of the structural layer 209 protrude in a negative direction. That is, it can also be said that the surface of the structural layer 209 on the substrate 201 side is closer to the substrate 201 than surfaces of the upper electrode layers 222 on the substrate 201 side.
If the MEMS switch thus manufactured is tried to be driven, as illustrated in
However, as described above, it is very difficult to prevent formation of the protrusions 211a, 211b, and 211c of the structural layer 209 in terms of a process. Therefore, when the protrusions 211a, 211b, and 211c cannot be eliminated, a structure is required in which the MEMS switch functions as a switch even in the case where there are the protrusions 211a, 211b, and 211c. For that purpose, the upper electrode layers 222 may be larger than the lower electrode layers 221 as illustrated in
In the case of forming the upper electrode layers 222 larger, even if there are protrusions 211a, 211b, and 211c, they are between steps formed by the lower electrode layers 221 and the substrate 201. Therefore, contact between the upper electrode layers 222 and the lower electrode layers 221 is not hindered.
Therefore, as in the case of the switching electrode layers, in the case where the upper electrode layer and the lower electrode layer, for example, are required to come in contact with each other in the micro electro mechanical systems switch (MEMS switch), a structure is decided so that the upper electrode layer is formed to have a larger area than the lower electrode layer.
“Being formed to have a larger area” means that in the case where, for example, each of the upper electrode layer and the lower electrode layer has a square shape or a rectangular shape, each side of the upper electrode layer is longer than that of the lower electrode layer or in the case where, for example, each of them has a circular shape, the radius of the upper electrode layer is longer than that of the lower electrode layer. That is to say, in the case where the upper electrode layer and the lower electrode layer are overlapped with each other, a bottom surface of the upper electrode layer is formed to completely embrace a top surface of the lower electrode layer. It can also be said that a side of a bottom surface of the upper electrode layer, which decides the shape thereof, and a side of a top surface of the lower electrode layer, which decides the shape thereof, do not overlap each other so that the side of the bottom surface of the upper electrode layer is always outside of the side of the top surface of the lower electrode layer. It is to be noted that in the case where a lead wiring portion of the upper and lower electrode layers cannot be taken into consideration, portions of the upper electrode layer, which do not overlap with the lower electrode layer, may be omitted.
Further, even in the case where an upper electrode layer is larger than a lower electrode layer opposite to the upper electrode layer, the upper electrode cannot be large enough to overlap with another lower electrode layer adjacent to the lower electrode layer opposite to the upper electrode layer, as well. Thus, the protrusions of the structural layer come in contact with the lower electrode layer to hinder contact between the upper electrode layer and the lower electrode layer. Further, in the MEMS switch, the upper electrode layer and the lower electrode layer are formed in a pair, so one upper electrode layer cannot be formed large enough to overlap with another lower electrode layer adjacent to a lower electrode layer opposite to the upper electrode layer.
The switch electrode layers are required to come in contact with each other; therefore, in the micro electro mechanical systems switch (MEMS switch) of the present invention, the upper switch electrode layer is formed larger than the lower switch electrode layer.
This embodiment mode is described with reference to
Although a switch electrode layer is described in Embodiment Mode 1, a drive electrode layer is described in this embodiment mode.
In order that a micro electro mechanical systems switch (MEMS switch) may function as a switch, an upper switch electrode layer and a lower switch electrode layer are required to favorably come in contact with each other. However, an upper drive electrode layer and a lower drive electrode layer are made not to come in contact with each other. Since a large potential difference is applied between the upper drive electrode layer and the lower drive electrode layer, when the upper drive electrode layer and the lower drive electrode layer come in contact with each other, a large amount of current flows therethrough so that a significantly large amount of power is consumed for driving of the switch. Further, when a current flows to the upper drive electrode layer and the lower drive electrode layer, light welding occurs due to electric discharge and thus sticking of the upper and lower drive electrode layers is caused.
In order to prevent sticking of the upper and lower drive electrode layers, an insulating layer may be formed on a surface of the drive electrode layer, that is, one or both of a top surface and a bottom surface of the drive electrode layer; however, such formation of an insulating layer is not preferred because of the following reason. That is, in the case where an insulating layer is formed on a surface of the drive electrode layer, a high voltage is applied to the upper drive electrode layer and the lower drive electrode layer to drive the switch; thus, the insulating layer formed over the drive electrode layer polarizes or traps a charge, so that sticking of the drive electrode layer occurs after all.
Therefore, in order to prevent contact between the upper drive electrode layer and the lower drive electrode layer, a stopper for limiting a movable region of a structural layer (also referred to as a bumper or a bump) may be formed. However, in order to form the stopper, another photomask and another manufacturing step are required to be added.
However, in this embodiment mode, by utilizing the protrusions 211a, 211b, and 211c of the structural layer 209, which hinder contact between the upper electrode layers 222 and the lower electrode layers 221, as described in Embodiment Mode 1 with reference to
An example of a specific structure of a MEMS switch is illustrated in
The MEMS switch illustrated in
A space 415 is between the substrate 401 and the structural layer 409. There are a protrusion 411a, a protrusion 411b, a protrusion 411c, and a protrusion 411d of the structural layer 409 on the periphery of the upper electrode layers 422.
As for the MEMS switch of this embodiment mode, the upper drive electrode layer 402b is formed smaller than the lower drive electrode layer 402a. Further, the upper switch electrode layer 404b is formed larger than the lower switch electrode layer 404a so that they favorably come in contact with each other, as in Embodiment Mode 1.
In the case where each of the upper drive electrode layers 402b is smaller than each of the lower drive electrode layers 402a, a space is formed between the upper drive electrode layers 402b and the lower drive electrode layers 402a by the protrusion 411a, the protrusion 411b, the protrusion 411c, and the protrusion 411d of the structural layer 409, which are on the periphery of the upper electrode layers 422 as illustrated in FIG. 6B, so that contact between the upper drive electrode layers 402b and the lower electrode layers 402a can be prevented.
The MEMS switch having such a structure can be manufactured using a design of a photomask by which the shapes of the upper electrode layers 422 are decided and a method described in Embodiment Mode 1. The photomask for forming the upper electrode layers 422 is required regardless of whether a stopper is formed or not; therefore, according to the present invention, the MEMS switch including a stopper for preventing contact between the upper drive electrode layers 402b and the lower drive electrode layers 402a can be manufactured without adding a photomask and a manufacturing step.
In this embodiment, described is a result obtained by manufacturing a switch in which a stopper for preventing contact between upper and lower drive electrode layers of the switch and an upper switch electrode layer and a lower switch electrode layer come in contact with each other as described in Embodiment Modes 1 and 2.
A method for manufacturing the switch is as described in Embodiment Modes 1 and 2. A base layer is formed over a substrate first and then lower electrode layers are formed over the base layer. Then, a sacrificial layer is formed so as to cover the lower electrode layers and upper electrode layers are formed over the sacrificial layer. Here, as each of the base layer, the lower electrode layers, and the sacrificial layer, a layer having a required property may be formed to a given thickness and processed by a photolithography method and etching.
In this embodiment, a glass substrate is used, a 300 nm-thick silicon nitride film containing oxygen is formed for the base layer, and a stack of a 300 nm-thick aluminum film and a 100 nm-thick titanium film is formed for the lower electrode layer. Because the aluminum film alone cannot resist high temperature, the titanium film is stacked over the aluminum film. Then, a 2 μm-thick tungsten film is formed for the sacrificial layer.
The upper electrode layer is formed using a stack of a 300 nm-thick aluminum film and a 100 nm-thick titanium film similarly to the lower electrode layer. In this embodiment, a conductive layer is etched by dry etching using a mixed gas of boron trichloride (BCl3) and chlorine (Cl2). Conditions for etching the conductive layer are as follows: the IPC power is 450 W, the bias power is 100 W, the flow rate of boron trichloride is 60 sccm, the flow rate of chlorine is 20 sccm, the pressure in a chamber is 1.9 Pa, and the standard etching time period of the conductive layer is 150 seconds. Thus, overetching of 100% with respect to the standard etching time period is performed. As a result, the sacrificial layer under the upper electrode layer is etched by approximately 100 nm.
Then, a structural layer is formed so as to cover the sacrificial layer and the upper electrode layer, and a contact hole is formed in the structural layer to form a wiring layer. After that, the structural layer is processed and the sacrificial layer is etched, so that the MEMS switch is completed. Here, each of the structural layer, the wiring layer, and the sacrificial layer, which has a required property, may be formed to a given thickness and processed by a photolithography method and etching similarly to the other layers.
In this embodiment, a 3 μm-thick silicon nitride film containing oxygen is formed for the structural layer and a stack of a 300 nm-thick aluminum film and a 100 nm-thick titanium film is formed and processed for the wiring layer. The sacrificial layer is etched by dry etching using a chlorine trichloride gas at normal temperature and normal pressure.
Here, in the present invention, the upper switch electrode layer is formed to have a larger area than the lower switch electrode layer and the upper drive electrode layer is formed to have a smaller area than the lower drive electrode layer, so that contact between the upper switch electrode layer and the lower switch electrode layer is prevented from being hindered and the stopper for preventing contact between the upper drive electrode layers and the lower drive electrode layers can be provided.
Further, it can be confirmed that when a voltage is applied between the upper drive electrode layers and the lower drive electrode layers of the MEMS switch manufactured through the above steps, the upper switch electrode layer and the lower switch electrode layer come in contact with each other, whereas the upper drive electrode layers and the lower drive electrode layers do not come in contact with each other.
This application is based on Japanese Patent Application serial no. 2007-293964 filed with Japan Patent Office on Nov. 13, 2007, the entire contents of which are hereby incorporated by reference.
Number | Date | Country | Kind |
---|---|---|---|
2007-293964 | Nov 2007 | JP | national |
Number | Name | Date | Kind |
---|---|---|---|
4939401 | Farrall | Jul 1990 | A |
RE33691 | Harnden, Jr. et al. | Sep 1991 | E |
5578976 | Yao | Nov 1996 | A |
6037719 | Yap et al. | Mar 2000 | A |
6392618 | Kimura | May 2002 | B1 |
6515791 | Hawwa et al. | Feb 2003 | B1 |
6566617 | Suzuki et al. | May 2003 | B1 |
7057251 | Reid | Jun 2006 | B2 |
7101724 | Chou | Sep 2006 | B2 |
7420320 | Sano et al. | Sep 2008 | B2 |
7545081 | Kim et al. | Jun 2009 | B2 |
7679186 | Jung et al. | Mar 2010 | B2 |
20030222321 | Yang et al. | Dec 2003 | A1 |
20030227361 | Dickens et al. | Dec 2003 | A1 |
20040031670 | Wong | Feb 2004 | A1 |
20040140872 | Wong | Jul 2004 | A1 |
20060208612 | Takayama et al. | Sep 2006 | A1 |
20070093045 | Yamaguchi et al. | Apr 2007 | A1 |
20070218630 | Yamaguchi et al. | Sep 2007 | A1 |
20080176046 | Yamaguchi et al. | Jul 2008 | A1 |
20080237005 | Honda | Oct 2008 | A1 |
Number | Date | Country |
---|---|---|
1150318 | Oct 2001 | EP |
60-247622 | Dec 1985 | JP |
09-092909 | Apr 1997 | JP |
11-174994 | Jul 1999 | JP |
2000-208018 | Jul 2000 | JP |
2001-033827 | Feb 2001 | JP |
2003-217423 | Jul 2003 | JP |
3590283 | Nov 2004 | JP |
2005-528751 | Sep 2005 | JP |
3865942 | Jan 2007 | JP |
Number | Date | Country | |
---|---|---|---|
20090127081 A1 | May 2009 | US |