DE102013209266A1 - Component with a cavity - Google Patents

Abstract

The invention relates to a method for producing a component. The method comprises providing a first substrate (110), having a functional structure (111, 112), providing a second substrate (120), and connecting the first and second substrates so that a cavity (131, 132) in the area the functional structure (111, 112) is formed. The method further includes forming an entrance (140) to the cavity (132), and closing the entrance (140). The invention further relates to a component with a functional structure (111, 112) and a cavity (131, 132) in the area of the functional structure (111, 112).

Description

The present invention relates to a method for manufacturing a component as well as a component. The component has a functional structure and a cavity in the region of the functional structure.

State of the art

Components with a functional structure, which is arranged in the region of a closed cavity, are known in different embodiments. Such a structure is present, for example, in micromechanical components such as rotation rate and acceleration sensors. In one possible method for producing such sensors, a substrate having a functional structure comprising at least one freestanding, movable functional element is provided. Underneath these are buried interconnects and electrodes. In a wafer bonding process, the substrate is connected to another substrate (cap wafer), whereby a cavity is formed in the region of the functional structure. Depending on the application, bonding may be performed under vacuum or at a given pressure or pressure atmosphere to trap a suitable pressure in the cavity.

In yaw rate sensors, in which a portion of the moveable structure is resonantly driven during operation, a relatively low pressure, typically 1mbar or less, is trapped. This leads to a low attenuation, so that a vibration excitation with relatively small electrical energy is possible. In acceleration sensors, a vibration excitation caused by an external acceleration should be avoided. Therefore, these sensors include higher internal pressures, typically 500mbar, to achieve higher damping. Furthermore, as a rule, organic surface coatings are used in the area of the functional structure in order to prevent a sticking together of movable functional elements.

One problem with the sensors described above is possible outgassing of components. This applies to the bond and surface areas within the cavity, for example the organic coating. One consequence is an undefined increase in internal pressure.

Micromechanical inertial sensors, which comprise a combination of a rotation rate and an acceleration sensor, can be realized in the form of a small inexpensive single-chip component. In one possible manufacturing method, functional structures of the different sensors are produced simultaneously on a substrate. The substrate is connected in a bonding process with a cap substrate, which has corresponding recesses for the separate encapsulation of the functional structures.

Various internal pressures of the cavities of the functional structures can be realized with the aid of a getter. A getter layer may be provided in the cavern of the rotation rate sensor. In this way, the wafer bonding can be performed at a relatively high pressure (i.e., the acceleration sensor). Subsequently, the getter layer can be activated and the cavern volume can be pumped off to a lower pressure. However, the use of a getter is associated with disadvantages such as high costs. The getter is a porous layer with a relatively large surface, which is additionally sealed. The activation takes place by a temperature step in which the seal is broken. Furthermore, in inertial sensors, precisely defined gas mixtures are often included in the cavities in order to adjust the damping of the sensors. The use of a getter limits the selection of a gas mixture since a getter selectively picks up only certain gases depending on the material. For example, getters are known that selectively do not absorb noble gases to all other gases.

The DE 10 2009 045 385 A1 describes a method of creating a trench in a silicon substrate through a grid of silicon oxide. The substrate is etched through the grid. After the etching, the grid is closed by depositing an oxide layer or a dielectric layer. As an application, the generation of electrical vias is called.

Disclosure of the invention

The object of the present invention is to specify an improved solution for the production of a component with a cavity in the region of a functional structure.

This object is solved by the features of the independent claims. Further advantageous embodiments of the invention are specified in the dependent claims.

According to the invention, a method for producing a component is proposed. The method includes providing a first substrate having a functional structure. It is further provided to provide a second substrate, and to connect the first and second substrates, so that a cavity is formed in the region of the functional structure. The procedure further comprises forming an access to the cavity, and closing the access.

In the method, after the first and second substrates have been joined, which can be done by a bonding process, it is provided that an access approaching the closed cavity is formed, which is subsequently closed again. The sequential closing makes it possible to set a predetermined defined internal pressure in the cavity or to subsequently correct the internal pressure, so that an optimum damping matched to the functional structure can be made available. The process can be carried out at a low process cost and thus at low cost compared to manufacturing processes using getter technology.

The method can be further combined with the production of one or more vias without additional process costs. In this regard, processes that are performed to form and seal the access may be used simultaneously in the fabrication of one or more vias. The at least one contact can according to the above-mentioned DE 10 2009 045 385 A1 be formed, the disclosure of which is hereby incorporated by reference into the present document.

The method can be considered in particular with regard to micromechanical components. A possible embodiment of the method provides for this purpose that the device to be produced by the method is a micromechanical component, wherein the functional structure has at least one movable functional element. The functional structure may be a MEMS structure (Micro-Electro-Mechanical System). One possible application is the field of inertial sensor technology. In this case, the micromechanical component may be an acceleration sensor or yaw rate sensor.

In a further embodiment, the method is carried out such that the cavity has a different, for example, lower internal pressure after closing the access than before closing the access. The internal pressure present after closure can be a defined internal pressure, for example the predetermined working pressure of the functional structure.

For example, it is possible to perform the bonding of the first and second substrates at a low pressure (or under vacuum) or a predetermined pressure. During and after bonding, outgassing may increase the internal pressure of the cavity. By forming the access, the cavity can be vented. The closing of the access can be carried out at appropriately selected pressure conditions to set a predetermined internal pressure or the predetermined internal pressure again in the cavity.

It may also be considered to perform the bonding of the first and second substrates at normal pressure. In this case, a predetermined working pressure in the cavity in the context of closing the access can be set. In this way, the connection can be facilitated and possibly favored.

In a further embodiment, forming the access, which may be formed in the first or in the second substrate, comprises forming a lattice structure on the respective substrate, followed by etching of substrate material in the region of the lattice structure. Closing the access involves closing the lattice structure. This can be done by forming a layer. In particular, it is possible to form a filling layer through which the lattice structure or lattice openings of the lattice structure are filled and thereby closed. The use of the grid structure below which or through which the access is etched is favorable for the etching process. The grid structure may be formed with relatively large lateral dimensions so that a relatively wide etch area is available for forming the access. After etching, the access is covered by the grid structure. As a result, a simple resealing of access, which can be done with the help of the filling layer, possible.

The formation of the lattice structure can be effected by first forming a dielectric layer, in particular a silicon oxide layer, on the substrate. Subsequently, grid openings in this layer and thus the grid structure can be formed. The formation of the filling layer, through which the grid openings can be closed, can likewise comprise applying a dielectric material, in particular silicon oxide.

In another embodiment, forming the fill layer comprises performing a chemical vapor deposition (CVD). With respect to the deposition of silicon oxide, a precursor comprising Si, O and H may be used. Suitable examples are TEOS (tetraethyl orthosilicate) or silane. Vapor deposition using such precursors provides the opportunity to include a relatively low internal pressure in the cavity. at The precipitation reaction can produce hydrogen, which has a relatively high diffusion rate through silicon oxide, so that during the deposition of a fast out-diffusion from the cavity is possible.

Closing the grid structure can also be done by another process. For example, it is possible to form a layer or filling layer by performing sputtering.

Even after the closing of the access or after the closing of the lattice structure, it is possible to diffuse out hydrogen (or also of other gas components). Such diffusing is used in the procedure described below.

In a further embodiment, a temperature process is carried out in which gas diffusion is caused by the closed access. In this way it is also possible to set a predetermined defined working pressure in the cavity or to subsequently correct the internal pressure, so that a matched to the functional structure optimal damping can be provided. The temperature step for effecting the gas diffusion may be applied to the case where the closing of the access, which can be carried out under appropriate pressure conditions, a predetermined internal pressure in the cavity can not or can not be adjusted with a predetermined accuracy. This is based on the assumption that the access and thus the cavity after closure can not be hermetically sealed and therefore, in particular for small gas atoms or molecules, there is the possibility of diffusion.

A gas diffusion can be caused in particular if, according to the procedure described above, the formation of the access takes place with the aid of a lattice structure which is subsequently closed with a filling layer. It may, for example, be considered to deposit the filling layer with a correspondingly small thickness.

For example, in the non-hermetically sealed access, there may be such a leakage rate that it is possible to match the internal pressure of the cavity to the partial pressure of the gases in the surrounding atmosphere over a period of several months. By the temperature process in which a targeted heating to higher temperatures, the internal pressure drift can be accelerated and consequently the internal pressure in the cavity can be selectively changed.

The temperature process may include, for example, storing the assembly of the two interconnected substrates at a temperature in the range of 380 ° C for a period of time in the range of 0.5 to 10 hours. The temperature is chosen such that melting of the bond is avoided.

If the final internal pressure of the cavity is adjusted by a temperature step, it may be considered to carry out the above-described formation of the filling layer for closing the lattice structure by means of plasma enhanced chemical vapor deposition (PECVD). For a filling layer of silicon oxide, Si, O and H-containing compounds such as TEOS and silane can again be used as precursor, whereby a relatively low internal pressure can be set in the cavity.

With regard to the chemical vapor deposition of the filling layer, which may be effected by a PECVD process, it may further be considered to carry out the deposition with a carrier gas which is diffusible through the closed access or the closed grid structure. With respect to a configuration of the lattice structure and the filling layer of silicon oxide, a gas having a high diffusion speed through silicon oxide may be used. This includes, for example, He. Also in this way, the setting of a low internal pressure by means of the temperature step can be enabled or favored.

In addition, it can be provided to use a mixed gas as the carrier gas for the vapor deposition or the PECVD process. This makes it possible to set in the temperature process, a relatively well-defined internal pressure whose value is independent of parameters such as the oxide thickness, the temperature and the length of the temperature process. The mixed gas has different gas components, which have different diffusion rates through the closed access and the closed lattice structure. For example, the mixed gas may have a first gas component with a high diffusion coefficient and a second gas component with a low diffusion coefficient. In this way, as part of the temperature step, the first component of the mixed gas can diffuse substantially completely out of the cavity, whereas the second gas component can remain substantially in the cavity. The final internal pressure can be adjusted in such an approach by the mixing ratio of the gas components of the carrier gas and the gas pressure in the deposition. Possible examples of fast diffusing Gas components are H2 or He, and for slowly diffusing gas components Ne, Ar or N2. It is possible that the mixed gas also includes more than two different gas components.

Not only can gas diffusion be used to cause the internal pressure in the cavity to decrease. Alternatively, instead of diffusing out a gas or one or more gas components, a diffusion of a gas or one or more gas components into the cavity can also be effected in order to set a corresponding internal pressure in the cavity.

In a further embodiment, a closure layer is formed in the region of the closed access. As a result, the access and thus the cavity can be hermetically sealed. The formation of such a closure layer may be considered in the event that a temperature process for inducing gas diffusion is performed. In this case, the closure layer is formed after the temperature process. Forming the sealing layer may include, for example, depositing silicon nitride. Also possible is a metal deposition by a sputtering or vapor deposition method. For example, it is suitable to form an aluminum-containing layer. When using a lattice structure, the sealing layer can be formed on the closed lattice structure or on the filling layer in the region of the lattice structure.

Depending on the conformity of the deposition of the sealing layer, it is also conceivable that waives the filling layer and the grid structure is sealed directly with a layer or sealing layer. For example, can be dispensed with the deposition from the gas phase and a sputtering with metal can be performed.

It is possible to perform the fabrication of the device even without a temperature process for effecting gas diffusion. In such a case, the formation of the sealing layer may optionally be omitted. The closed access can already be hermetically sealed. If a lattice structure is used, a layer closing off the lattice structure can be matched thereto or formed with a correspondingly large thickness.

Not only can it be a single device, but multiple devices can be made in parallel. For this purpose, the first substrate is provided with a plurality of functional structures, and by connecting or bonding to the second substrate, a plurality of cavities are formed, each in the region of the individual functional structures. By correspondingly performing the above processes, in particular forming accesses to cavities and closing them, and optionally by carrying out a temperature process, which can be done at substrate or wafer level, setting of a predetermined internal pressure or internal pressure correction can be effected. All or more of the cavities may have the same internal pressure in this way. A scattering of the internal pressure, as can otherwise occur in the parallel processing of several components, can therefore be significantly reduced. The internal pressure can be independent of parameters such as, for example, cavern depths, pinch widths of a bonding material, or force or temperature variations during bonding. Following this, further processes such as a singulation process to provide separate components may be performed.

The method and its possible embodiments can not only be used to set a given working pressure in the associated cavity in a component which has a single functional structure. A further application is the integration of at least two sensor elements or functional structures on a single component, with different pressure requirements for the functional structures. One possible example is a combination of an acceleration sensor and a rotation rate sensor on a component or chip.

A further embodiment proposes for this purpose that the provided first substrate has a first and a second functional structure. By connecting the first and second substrates, a first cavity in the region of the first functional structure and a second cavity in the region of the second functional structure are formed. The access is formed to the second cavity and subsequently closed. Optionally, a temperature process for inducing gas diffusion through the sealed access is performed.

The second substrate can, for example, have two depressions which are separated from one another by a web and are matched to the two functional structures. After connecting the first and second substrates, the two cavities are separated by the web hermetically sealed.

Determining different pressures in the two cavities can be done at two points of the process flow. The working pressure of the first functional structure may be adjusted when connecting the first and second substrates. When connecting any selected process gases available. By selecting the gas mixture as well as the pressure during bonding, an optimum damping matched to the first functional structure can be realized.

The working pressure of the second functional structure may be determined upon closing the access, which may be accomplished by closing a lattice structure with a fill layer, or optionally by a temperature process to induce gas diffusion. In this way, the second cavity after closing the access and / or after performing the temperature process may have a different internal pressure than the first cavity. With this approach, an optimal damping tuned to the second functional structure can be provided.

Furthermore, it is possible to produce a component which has more than two functional structures and corresponding cavities in the region of the functional structures. Here, the first substrate has the plurality of functional structures, and the cavities are formed by connecting the first and second substrates. The sequential closing also offers the possibility to meet different working pressure requirements of the more than two functional structures. When using a getter, this is not possible. Setting different internal pressures in different cavities can be accomplished by repeating processes described above such as forming access to a cavity followed by sealing multiple times with respect to different cavities. Optionally, a temperature process or multiple temperature processes may be performed to additionally induce gas diffusion through one or more sealed accesses.

According to the invention, a component is furthermore proposed, which has a first substrate with a functional structure and a second substrate. The second substrate is connected to the first substrate, so that a cavity is present in the region of the functional structure. The device further comprises a sealed access in the first or second substrate, which reaches close to the cavity. This device may be made according to the method described above or one of the above-described embodiments of the method. Therefore, the cavity may have a predetermined or well-defined internal pressure.

In the device, embodiments described above and details mentioned in relation to the manufacturing method can be applied in the same way. Also, the above advantages may be present.

The above-explained and / or reproduced in the dependent claims advantageous embodiments and refinements of the invention can - except for example in cases of clear dependencies or incompatible alternatives - individually or in any combination with each other are used.

The invention will be explained in more detail below with reference to the schematic figures. Show it:

1 a side sectional view of a functional substrate and a cap substrate connected to the functional substrate, wherein the functional substrate has two functional structures and in the region of each of the functional structures is a closed cavity;

2 and 3 a perspective view and a side sectional view of the cap substrate after forming a grid structure on the cap substrate and etching an access opening through the grid structure;

4 a side sectional view of the cap substrate after forming a filling layer for closing the grid structure and thus the access opening;

5 a side sectional view of the cap substrate after the formation of a sealing layer;

6 a side sectional view of a micromechanical device produced according to the preceding figures, wherein the cap substrate has the closed access opening;

7 a flow diagram of a method for producing a micromechanical device; and

8th a side sectional view of another micromechanical device, wherein the functional substrate has a closed access opening.

Embodiments of a method for producing a micromechanical component will be described with reference to the following figures. The component is a cost-effective single-chip component which has two functional structures and associated cavities or cavities in the region of the individual functional structures. There are different working pressure requirements for the functional structures.

In the production can be in the semiconductor or microsystem technology usual Processes, for example, complementary metal oxide semiconductor (CMOS) processes and MEMS processes, are carried out and common materials are used, so that this is only partially discussed. In addition to illustrated and described processes, further process steps may optionally be carried out. In the same way, in addition to components and structures shown and described further components, structures and / or layers may be provided. It is further noted that the figures are merely schematic in nature and are not to scale. In this regard, components and structures shown in the figures may be exaggerated or oversized for clarity.

For the method, parallel production of multiple devices, i. on substrate or wafer level, and a subsequent separation to provide separate components or sensor chips come into consideration. The figures illustrate the production here in detail by means of a single component. The following description may apply to all of the parallel processed components in this sense.

The 1 to 6 show a method for producing a micromechanical device 100 , which two micromechanical functional structures 111 . 112 having. Procedural steps performed in the method are further illustrated in the flowchart of FIG 7 summarized, which is also referred to below.

At the beginning of the procedure will be in one step 201 (see. 7 ) a first substrate 110 and a second substrate 120 provided. The two substrates 110 . 120 be in a further step 202 (see. 7 ), so that the in 1 shown substrate arrangement is formed.

The first substrate 110 has a first micromechanical functional structure 111 and a second micromechanical functional structure 112 on. The two functional structures 111 . 112 wise, as in 1 is indicated, freestanding and movable functional elements. The substrate 110 is hereinafter also referred to as sensor or functional substrate 110 , and the functional structures 111 . 112 are hereafter referred to as microstructures 111 . 112 designated. The microstructure 111 For example, it can be designed to detect an acceleration, so that this part of the component 100 serves as an acceleration sensor. The other microstructure 112 For example, it can be designed to detect a rate of rotation, so that this part of the component 100 serves as a rotation rate sensor.

The functional substrate 110 with the microstructures 111 . 112 can be provided using MEMS manufacturing processes. For example, on a starting substrate, which may be a semiconductor or silicon substrate, a layer arrangement may be formed which comprises buried conductor tracks and electrodes, and an upper functional layer, for example a polycrystalline silicon layer. By structuring the functional layer and performing a sacrificial layer etching, the microstructures 111 . 112 be generated.

The other substrate 120 is intended to work with the functional substrate 110 connected (capped) to caves or cavities completed in this way 131 . 132 in the field of microstructures 111 . 112 to build. The substrate 120 is also referred to below as a cap substrate 120 designated. The cap substrate 120 has on the substrate side provided for the connection on the functional substrate 110 tuned structure with protruding substrate areas, which recesses 121 . 122 enclose, and which serve in the subsequent connection as a bonding frame. The wells 121 . 122 are of dimensions on the microstructures 111 . 112 tuned and by a jetty 125 separated from each other.

After deploying (step 201 ) become the two substrates 110 . 120 , as in 1 shown connected to each other (step 202 ). For this purpose, a wafer bonding process is performed. In the bonding process, which is performed at a suitable bonding temperature, the substrates become 110 . 120 firmly connected. A connection can be like in 1 shown via a tie layer 135 getting produced. One or more bonding materials from which the bonding layer 135 When bonding can be generated previously on one of the substrates 110 . 120 , for example on the cap substrate 120 in the field of bond frames and the bridge 125 , or on both substrates 110 . 120 to be applied.

By bonding, using the wells 121 . 122 of the cap substrate 120 a first closed cavity 131 in the area of the first microstructure 111 and a second sealed cavity 132 in the area of the second microstructure 112 educated. The cavities 131 . 132 are hermetically sealed against the external environment, and through the central pier 125 also hermetically separated from each other.

The bonding of the substrates 110 . 120 in the step 202 can be done, for example, by wafer bonding with a glass solder, by eutectic bonding with one or more metal layers, or by anodic bonding. For the bonding can the process pressure can be freely selected. It is possible that the functionality of the bonding process does not depend on the choice of a particular process pressure. After the bonding process, the hermetically sealed cavities show 131 . 132 the same internal pressure or the same internal pressure atmosphere, hereinafter referred to as the first pressure p1 (see. 7 ), on.

In the present case, the bonding of the substrates 110 . 120 performed under such pressure conditions that a given optimal damping of the first microstructure 111 is possible. The in connecting the substrates 110 . 120 present and thereby in the cavities 131 . 132 set first pressure p1 thus sets the on the first microstructure 111 coordinated working pressure. In the event that the microstructure 111 For example, to detect an acceleration, the pressure p1 may be in the range of 500 mbar.

The second microstructure 112 on the other hand requires a different damping and therefore a different working pressure from the pressure p1. The separation of the cavities 131 . 132 through the footbridge 125 allows, by forming an access 140 followed by reclosing it under a different pressure, a subsequent pressure change of the second cavity 132 , In the manufacture of the device 100 is provided, the access opening 140 in the cap substrate 120 train. A possible area 129 this is in 1 indicated by dashed lines.

It should be noted that processes carried out after bonding, in particular deposition processes, are carried out at process temperatures below the bonding temperature of the bonding process. As a result, a melting of the bond is avoided.

For opening the second cavity 132 is first in a further step 203 (see. 7 ), how fragmentary in the 2 and 3 shown is a layer 150 on the sensor substrate 110 opposite side or top of the cap substrate 120 educated. The layer 150 , which serves as a masking layer at a later stage of the process, is preferably formed from a dielectric material, in particular silicon oxide.

For forming the layer 150 is a deposition process, in particular a CVD process (Chemical Vapor Deposition) performed.

As part of the step 203 (see. 7 ) is further, as well as partial in the 2 and 3 shown is the layer 150 in the area of the access hole to be created 140 structured in a grid shape, so that at this point a lattice structure 155 is present. The grid structure 155 has a plurality of grid openings 156 which of web-shaped layer regions of the layer 150 are surrounded. Forming the lattice structure 155 can be done by a suitable structuring process. The entire grid structure 155 can, as in 2 is indicated, a rectangular contour, or alternatively have a different shape. Also the grid openings 156 may differ from the in 2 shown rectangular shape have another form of supervision.

In a subsequent step 204 (see. 7 ), an etching process is performed, in which, as in the 2 and 3 illustrated is an access hole 140 in the cap substrate 120 is trained. In the etching process, the layer is used 150 as a mask, and gives those in the layer 150 trained lattice structure 155 the etching area and thus the lateral shape of the in the cap substrate 120 etched access hole 140 in front. During etching, substrate material of the cap substrate becomes 120 through the grid structure 155 through it. After forming the access hole 140 extending from the top of the cap substrate 120 to the second cavity 132 extends, is the cavity 132 vented and may have atmospheric pressure. The access hole 140 is at the cap top of the grid structure 155 covered.

Forming the to the cavity 132 approaching access opening 140 can be done by means of a dry chemical etching process. Suitable is a reactive Ionenätzprozess such as the so-called Bosch process.

In a subsequent step 205 (see. 7 ), as in 4 is shown, a filling layer 160 on the shift 150 educated. Through the filling layer 160 become the grid openings 156 the lattice structure 155 filled, and as a result the lattice structure 155 and thus the access hole 140 locked. The filling layer 160 is also preferably formed of a dielectric material such as in particular silicon oxide.

For the formation of the filling layer 160 a deposition process, in particular a CVD process, is performed. This material of the layer 160 or silica, as in 4 is shown, also within the access hole 140 be applied to corresponding side wall areas. The layer material will be faster at the surface than at the depth of the access opening 140 deposited, which is therefore an increase in the grid openings 156 at the top of the cap substrate 120 entails.

Closing the lattice structure 155 in the step 205 by depositing the filling layer 160 is performed under such pressure conditions that in the second cavity 132 an internal pressure or an internal pressure atmosphere, hereinafter referred to as second pressure p2 (cf. 7 ). The pressure p2 can be applied to the second microstructure 112 adjusted working pressure to a given optimal damping of the microstructure 112 to provide. The second pressure p2 may differ from the first pressure p1, as stated above. In an embodiment of the first microstructure 111 for detecting an acceleration and the second microstructure 112 for detecting a rotation rate, the pressure p2 may be smaller than the pressure p1.

CVD deposits can be made at different process pressures. For example, it is possible to deposit at a process pressure of approximately 1 mbar, as may be the case, for example, in the case of a PECVD (Plasma Enhanced CVD) process. A deposition at a process pressure of, for example, about 100 mbar can be done for example by means of a SACVD process (sub-atmospheric CVD). Depending on the printing requirements of the second microstructure 112 a corresponding deposition process can be selected, and thereby a predetermined working pressure in the second cavity 132 , for example, 1mbar or 100mbar.

With regard to a filling layer 160 From silicon oxide, the CVD process can be carried out with Si, O and H-containing precursors, for example TEOS (tetraethyl orthosilicate) or silane. In this way, a relatively low internal pressure in the second cavity 132 be included. In the deposition reaction, hydrogen may be produced, which has a relatively high diffusion speed through silicon oxide, so that a rapid outdiffusion from the cavity during deposition 132 is possible. Optionally, outdiffusion even after closing is possible or even intended, as will be described in more detail below.

It is possible that the access hole 140 and thus the cavity 132 after forming the filling layer 160 hermetically sealed. For this purpose, for example, be provided, the filling layer 160 to produce with a relatively large layer thickness.

In this way, the device can 100 which is in 6 shown to be substantially completed. The component 100 shows the two hermetically separated closed cavities 131 . 132 , and that in the area of the cap substrate 120 trained and closed access hole 140 on. In the first cavity 131 is the first pressure p1, and in the second cavity 132 the second pressure p2 is included.

As part of the completion of the device 100 Further processes can subsequently be carried out. This includes, for example, a separation process. Previously, several components 100 made in a parallel manner, which can be done by performing the wafer level processes described above. In the singulation process, the arrangement becomes the substrates 110 . 120 separated in appropriate places, whereby the parallel processed components 100 be separated from each other.

Modifications may be considered for the above-described manufacturing method, as will be described in more detail below with reference to further embodiments. It should be noted that matching processes, as well as identical or equivalent structures and components will not be described again in detail below. For details of matching features, processes, potential benefits, etc., reference is made to the above description. It is further to be understood that details referred to in relation to one of the following embodiments may also apply to other embodiments.

It is possible that in the step 205 (see. 7 ) produced filling layer 160 no hermetically sealed closure of the access hole 140 and thus the second cavity 132 can provide. In such a case, small gas particles or molecules may fill the fill layer 160 and the oxide grid 155 and therefore, over time, the pressure of the corresponding gas inside the cavern 132 the partial pressure outside the cavern 132 assimilate.

To avoid this, after the step 205 in a further step 206 (see. 7 ), as in 5 is shown, a sealing layer 170 with improved sealing properties on the filling layer 160 be deposited and structured. For example, a deposition of silicon nitride is considered. Alternatively, metal deposition can be accomplished by a sputtering or vapor deposition process. For example, it is suitable to form a layer of aluminum or copper. A remaining after structuring part of the sealing layer 170 , which is in the area of the closed lattice structure 155 or the access opening 140 can provide a hermetically sealed closure for life.

In the event that a given internal pressure in the second cavity 132 as part of the sealing with the filling layer 160 (Step 205 in 7 ) can not be set directly or not with a predetermined accuracy, an alternative embodiment of the method is performed, in which the internal pressure of the second cavity 132 is selectively changed after closing. This is the above-described, non-hermetically sealed closure of the access hole 140 and thereby exploited gas exchange with the environment exploited.

To accelerate the gas diffusion, after closing the access hole 140 or the step 205 , in a further step 207 (see. 7 ) carried out a temperature process. In this process, the arrangement of the two substrates 110 . 120 heated to an elevated temperature and stored at this temperature. The resulting gas diffusion through the closed lattice structure 155 or the locked access hole 140 makes it possible the internal pressure of the second cavity 132 , starting from the second pressure p2, to a different pressure level or a different pressure atmosphere, referred to below as the third pressure p3 (cf. 7 ), to change. In particular, a reduction in pressure is possible. Here, the pressure p3 on the second microstructure 112 adjusted to a predetermined optimal damping of the second microstructure 112 to effect.

The tempering step 207 For example, it may include storage of the substrate assembly at a temperature in the range of about 380 ° C. Storage at this temperature may be for a period of time in a range of, for example, between 0.5 to 10 hours. The temperature is chosen such that melting of the bond is avoided. If the second cavity 132 For example, with helium filled with a pressure of 1 mbar, the pressure in the temperature process can be reduced by storing the substrate assembly in nitrogen targeted to 0.1 mbar.

Following the step 207 may be the step described above 206 be carried out (see. 7 ) to the in 5 Sheath shown 170 train. This allows the access hole 140 and thus the cavity 132 hermetically sealed.

Regarding the temperature step 207 for adjusting the pressure p3 in the second cavity 132 may be considered the previous closing of the access 140 with the filling layer 160 (Step 205 ) and / or with suitably selected process conditions.

For example, the deposition of the filling layer 160 using a PECVD process. Provided that setting a low internal pressure in the second cavity 132 is provided, precursors in the form of Si, O and H-containing compounds, for example TEOS or silane can be used. In order to further favor the setting of a low internal pressure, it may further be considered to perform the deposition with a carrier gas, which has a high diffusion coefficient and thereby a high diffusion speed through the closed lattice structure 155 or the material present here (silicon oxide). This includes, for example, helium. As a result, in the subsequent temperature process, a large part of the carrier gas from the cavity 132 be diffused out.

It is also possible to use a mixed gas as the carrier gas for the PECVD process. This makes it possible to set a precisely defined internal pressure whose value is independent of parameters such as, for example, the oxide thickness, the temperature and the length of the temperature process. For example, the mixed gas may include a first gas component having a high diffusion coefficient and a second gas component having a small diffusion coefficient. In the temperature process, the first component of the mixed gas may substantially completely out of the second cavity 132 diffuse out, whereas the second gas component substantially in the cavity 132 can remain. The final internal pressure after annealing depends on the mixing ratio of the gas components of the carrier gas and the gas pressure in the deposition of the filling layer 160 , Possible examples of fast-diffusing gas components are H2 or He, and for slowly diffusing gas components Ne, Ar or N2.

In addition, further modifications may be considered for the method. One possible modification is in 8th on the basis of the micromechanical component 101 illustrated. The component 101 has substantially the same structure as that in 6 shown component 100 on. In contrast to the component 100 is at the device 101 the locked access hole 140 in the functional substrate 110 educated.

As part of the manufacture of the device 101 may be those described above and on the cap substrate 120 or in the region of the cap top processes carried out in an analogous manner to the functional substrate 110 be performed. In this case, the formation of the layer takes place 150 with the grid 155 (Step 203 ) on the cap substrate 120 opposite side or back of the functional substrate 110 , and will that access hole 140 starting from the sensor back to the cavity 132 etched (step 204 ). Following this, the other processes, ie closing the access 140 with the filling layer 160 (Step 205 ), optionally performing the temperature process to effect gas diffusion (step 207 ), and forming the sealing layer 170 (Step 206 ), be performed.

It is also possible to combine the process with the production of one or more vias without additional process costs. For example, the access hole 140 are etched simultaneously with an isolation trench, which serves to insulate a conductive substrate region. For the access hole 140 and the isolation trench may previously have correspondingly shaped lattice structures in the layer 150 be formed. Furthermore, the closing of the access hole 140 and the isolation trench together by forming the filling layer 160 and optionally the sealing layer 170 respectively. The contact can according to the above-mentioned DE 10 2009 045 385 A1 be educated. Furthermore, the access hole 140 and the via in the functional substrate 110 , or in the cap substrate 120 be formed.

The method and its different embodiments can be used not only for the production of components with two functional structures and two cavities. It is also possible to produce components with only one functional structure and one cavity in the region of the functional structure. The functional structure may be a micromechanical functional structure. The functional structure can be designed, for example, for detecting an acceleration or a rotation rate. In the second variant, the component to be produced may be, for example, an open-loop gyroscope, for which a relatively low cavern internal pressure is provided.

In the manufacture of such devices, a functional substrate having the functional structure may be connected to a cap substrate to form a cavity in the region of the functional structure. The bonding can be done, for example, under vacuum. By forming an access 140 to the cavity and resealing same, and optionally by performing a temperature process to cause gas diffusion, a predetermined (low) internal pressure in the cavity can be adjusted. In this way, for example, an undefined pressure increase in the cavity caused as a result of outgassing can be corrected. Outgassing may occur, for example, at a bonding layer created during bonding. Furthermore, the component in question may have an organic coating in the cavity from which components outgas.

In the same way, the method can be used for components having more than two functional structures, wherein a separate cavity is provided for each of the functional structures. A setting of different internal pressures in different cavities can take place in that, following the connection of a functional substrate having the functional structures to a cap substrate for forming the cavities, the processes described above, ie the formation of an access 140 to a cavity followed by sealing, repeated several times. Furthermore, a temperature process or, if appropriate, a plurality of temperature processes for producing a gas diffusion from one or more hermetically (still) not tightly sealed cavities can be carried out.

A further variant consists in that the pressure p1 set during a bonding of a function with a cap substrate (step 202 ) is smaller than that in closing an access 140 set pressure p2 (step 205 ). Furthermore, one in the temperature process (step 207 ) gas diffusion are not only used to cause a reduction of the internal pressure in a cavity. Alternatively, instead of diffusing out a gas or one or more gas components, a diffusion of a gas or one or more gas components into the cavity may also be effected. In this way, the internal pressure, starting from a pressure p2, can be increased to a pressure p3.

In addition, there is the possibility to use the method not only for components in the form of, for example, acceleration and / or rotation rate sensors. The method can likewise be used for components which have other functional structures, wherein a closed cavity is provided in the region of a functional structure.

One example is a micromirror device in which the functional structure has a plurality of deflectable micromirrors. The functional structure is formed on a functional substrate, which is connected to a cap substrate for forming a closed cavity in the region of the functional structure. This can be done under vacuum. Analogously, by forming an access 140 to the cavity and resealing same, and optionally by performing a temperature process to induce gas diffusion, a predetermined one Internal pressure can be adjusted in the cavity. As a result, for example, an increase in pressure due to outgassing can be corrected.

The embodiments described above and partly explained with reference to the figures represent preferred or exemplary embodiments of the invention. In addition to the described and illustrated embodiments, further embodiments are conceivable, which may comprise further modifications or combinations of features. For example, other materials may be used in place of the materials listed above, and other numbers, such as temperatures, pressures, etc., may be substituted by other statements.

Another possible modification is, for example, to dispense with the formation of a filling layer by performing a vapor deposition. Instead, a grid structure may be directly sealed with a layer formed by conducting sputtering with metal.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 102009045385 A1 [0007, 0012, 0090]

Claims (10)

Method for producing a component, comprising the method steps: providing a first substrate ( 110 ), comprising a functional structure ( 111 . 112 ); Providing a second substrate ( 120 ); Connecting the first and second substrates such that a cavity ( 131 . 132 ) in the area of the functional structure ( 111 . 112 ) is formed; Forming an access ( 140 ) to the cavity ( 132 ); and closing the access ( 140 ). Method according to claim 1, wherein the component is a micromechanical component ( 100 . 101 ), and wherein the functional structure ( 111 . 112 ) has at least one movable functional element. Method according to one of the preceding claims, wherein the cavity ( 132 ) after closing the access ( 140 ) has a different internal pressure than before closing the access ( 140 ). Method according to one of the preceding claims, wherein the access ( 140 ) is formed in the first or second substrate, wherein the forming of the access ( 140 ) forming a lattice structure ( 155 ) on the relevant substrate and etching of substrate material in the region of the lattice structure ( 155 ) and closing the access ( 140 ) closing the grid structure ( 155 ), in particular by forming a filling layer ( 160 ). Method according to claim 4, wherein the formation of the filling layer ( 160 ) for closing the grid structure ( 155 ) comprises performing one of the following processes: performing a chemical vapor deposition; Performing a chemical vapor deposition with a carrier gas passing through the sealed access ( 140 ) is diffusible; Performing a chemical vapor deposition with a carrier gas in the form of a mixed gas of different gas components, the different diffusion rates through the sealed access ( 140 ) exhibit; or performing a sputtering. Method according to one of the preceding claims, wherein a temperature process is carried out in which a gas diffusion through the sealed access ( 140 ) is caused. Method according to one of the preceding claims, wherein a sealing layer ( 170 ) in the area of closed access ( 140 ) is formed. Method according to one of the preceding claims, wherein the first substrate ( 110 ) a first and a second functional structure ( 111 . 112 ), wherein by connecting the first and second substrates, a first cavity ( 131 ) in the area of the first functional structure ( 111 ) and a second cavity ( 132 ) in the area of the second functional structure ( 112 ), the access ( 140 ) to the second cavity ( 132 ) and subsequently sealed, and optionally a temperature process for inducing gas diffusion through the sealed access (FIG. 140 ) is carried out. Method according to claim 8, wherein the second cavity ( 132 ) after closing the access ( 140 ) and / or after performing the temperature process has a different internal pressure than the first cavity ( 131 ). A device comprising: a first substrate ( 110 ) with a functional structure ( 111 . 112 ); a second substrate ( 120 ), which with the first substrate ( 110 ), so that a cavity ( 131 . 132 ) in the area of the functional structure ( 111 . 112 ) is present; and one to the cavity ( 132 ) adequate closed access ( 140 ) in the first or second substrate.

Images