MEMS microspeaker package and a method of manufacturing a plurality of MEMS microspeaker packages

Abstract

The present invention relates to a MEMS micro speaker package (10) where patterned polymer forms an acoustic seal between a MEMS micro speaker chip (20) and a substrate (30). The package has an identical or substantially identical footprint to the MEMS micro speaker chip (20), i.e. the MEMS micro speaker package (10) is true chip-scale. The acoustic seal formed by the patterned polymer layer prevents sound from leaking along the edges of the package. The invention also relates to a method of manufacturing a plurality of such chip-scale MEMS micro speaker packages (10) through wafer level chip-scale packaging.

Description

[0001] MEMS MICROSPEAKER PACKAGE AND A METHOD OF MANUFACTURING A PLURALITY OF MEMS MICROSPEAKER PACKAGES

[0002] TECHNICAL FIELD

[0003] The present invention relates to a micro electromechanical system (MEMS) micro speaker package. The MEMS micro speaker packages are chip scale packages.

[0004] The present invention also relates to a method of producing a plurality of such chip scale MEMS micro speaker packages.

[0005] BACKGROUND

[0006] For sound generating devices, or speakers, generally, the acoustic output is related to the active area, with larger speaker sizes capable of producing more sound output. Several smaller speakers can also be combined into a combined device to create stronger sound output as well. However, in micro speaker applications, size is always a limiting factor, creating a challenge of producing adequate volume using a small active area, as the sound output of the device is typically proportional to the size of active sound producing elements. Therefore, it is crucial in the production of micro speakers to avoid adding anything that takes up unnecessary space on the resulting chip, as such components would negatively affect the maximum possible sound output of the micro speaker.

[0007] Micro-electromechanical systems (MEMS) micro speakers are loudspeaker components based on silicon chips, using semiconductor manufacturing processes to create moving parts. The speakers themselves typically consist of a single chip, which incorporates one or several membranes, cantilevers or other sound producing features. In order to be practical to use in an application, the bare chips have to be protected and packaged into devices compatible with standard manufacturing procedures.

[0008] The MEMS chip of a MEMS micro speaker typically has one active side that comprises both the sound producing elements and pads for electrical connections. There are several common technologies for packaging of MEMS devices, but they do not necessarily work well for MEMS micro speakers because MEMS micro speakers have the additional requirement that they need to be open towards both the front and back after packaging in order to radiate sound output. The most common package structure for MEMS micro speakers is the Land Grid Array Formed Lid (LGA-FL) package. The LGA-FL consists of a base interposer, typically a printed circuit board (PCB) laminate, the MEMS chip and possibly supporting chips, and a lid or cover that is glued onto the base. An example of related art disclosing such solutions is US 2021092500 Al. In LGA-FL packaging, the MEMS chip can be placed on the base interposer with the active side up and wire-bonded, or with the active side down using flip chip assembly. Beside the problem with added volume, a further problem with having a cover over a MEMS micro speaker chip is that the lid may degrade the output sound quality due to an additional acoustic resonance caused by the cover. If the chip is placed with the active side down, the back side of the chip might not be as sensitive and require as much protection, and the cover may be omitted, or be attached directly onto the back side of the chip. The related art document US 2021092500 Al discloses one example of having the cover attached directly onto the back side of the chip using globbing or a similar method, as a post processing step on a singulated chip.

[0009] A problem with the common packaging structures as described above is thus that they add volume to the resulting chip, taking away from the possible area for sound producing elements and thereby reducing the maximum possible sound output of the MEMS micro speaker. A further problem is that the packaging process is sequential and needs to be performed separately for every chip, after dicing. It is further hard to process the singulated MEMS micro speaker chips, because of the small size factor and fragile structures, which inevitably results in a lowered yield.

[0010] In order to reduce wasted space as well as parallelize an otherwise sequential packaging process, Wafer Level Chip Scale Packaging (WL-CSP) have recently been introduced for MEMS devices. In WLCSPs, a wafer with multiple chips are connected to the base and packaging components are attached to it before singulation, or dicing. In other words, the entire wafer and the added packaging components are connected to an interposer before singulation. Each resulting MEMS chip package will after dicing have an outline that is not larger than the size of the original chip. WLCSP would be desirable to use for MEMS micro speaker packaging, but this poses challenges. One issue is that the edges between the MEMS micro speaker chip and the substrate would have to be sealed in order to prevent acoustic short-circuit. At the same time, it is important that no sealing material touches any of the active sound producing features on the MEMS micro speaker chip. So far, no design for a working WLCSP package for MEMS micro speakers that overcomes these challenges has been made.

[0011] Hence, there is still a need to provide an improved MEMS micro speaker package that enables WLCSP of MEMS micro speakers, and thus production of chip scale MEMS micro speaker packages.

[0012] SUMMARY

[0013] The object of the invention is to solve the above mentioned problems. Specifically, the invention aims to provide a chip-scale micro electromechanical system (MEMS) micro speaker package, and a method of producing a plurality of such chip scale MEMS micro speaker packages using wafer-level processing.

[0014] MEMS micro speakers need to be protected due to their small size and fragile structure, to ensure the structural integrity of the chip and resulting device. At the same time, it is essential to maximize the possible sound output of the micro speaker and therefor avoid adding any protective components that reduce the area available for sound producing elements. For these reasons, it would be highly desirable to design a working WLCSP process for MEMS micro speakers. Such processes have so far not been available, because the specific challenges connected to wafer-level packaging of MEMS micro speakers explained above have not previously been solved.

[0015] Advantageously, the inventor has found a way of designing a working WLCSP process for MEMS micro speakers and thereby producing a chip-scale MEMS micro speaker package.

[0016] This is achieved by a MEMS micro speaker package and a method for manufacturing a plurality of MEMS micro speaker packages according to the appended independent claims.

[0017] In a first aspect of the invention, there is provided a method for manufacturing a plurality of MEMS micro speaker packages. The method comprises providing a MEMS micro speaker layer comprising a plurality of MEMS micro speakers, wherein each MEMS micro speaker has at least one active sound producing area. For each active sound producing area, the MEMS micro speaker layer also comprises at least one electrical connection element is electrically connected to an actuator arranged to actuate the active sound producing area. The MEMS micro speaker layer has a front side and a back side, and the active sound producing areas and the corresponding electrical connection elements of the plurality of MEMS micro speakers are distributed over the front side of the MEMS micro speaker layer according to a first pattern. The method further comprises providing a substrate with a front side and a back side. The substrate has electrical connection elements distributed on its front side of the substrate, such that each electrical connection element on the substrate matches the location of an electrical connection element in the first pattern. The method also comprises applying a polymer layer to one of the front side of the MEMS micro speaker layer and the front side of the substrate. Thereafter, the method comprises patterning the polymer layer according to the first pattern, to comprise at least one first area that is filled with polymer and second areas that lack polymer, wherein the second areas match the locations of the electrical connection elements and active sound producing areas in the first pattern. Applying the polymer layer may comprise spinning, spin coating or laminating a polymer onto the front side of said one of the MEMS micro speaker layer and the substrate. The method then comprises bonding the polymer layer to the other of the front side of the MEMS micro speaker layer and the front side of the substrate, to form a package with the polymer layer sandwiched between the MEMS micro speaker layer and the substrate. The polymer layer is bonded to the other of the front side of the MEMS micro speaker layer and the front side of the substrate such that the polymer layer creates a sealing barrier around a perimeter of each of the active sound producing areas without contacting any part of the active sound producing areas. The polymer layer is further bonded to the other of the front side of the MEMS micro speaker layer and the front side of the substrate such that the electrical connection elements are not covered by or in contact with the polymer in the polymer layer. By the bonding step, a mechanically stable and hermetically sealed encapsulation of the active sound producing areas is ensured. Any suitable bonding technique known in the art may be used. In a final step, the method comprises dicing the package to obtain the plurality of MEMS micro speaker packages.

[0018] Advantageously, a working solution for wafer-level packaging of MEMS micro speakers is provided, resulting in chip scale packaging of the MEMS micro speakers after dicing. The need for post processing of the MEMS micro speakers after dicing, adding components such as a protective cover and thereby increasing the volume of the resulting packaged product, is thereby removed. Suitably, the yield rate for MEMS micro speaker packages is thereby enhanced and the reliability and uniformity of the package structure of the resulting MEMS micro speakers is improved compared to prior solutions. A further advantage is that the resulting MEMS micro speaker packages are downsized while the ratio of available active sound producing area to chip area is increased compared to known solutions. Furthermore, the previously sequential packaging process is parallelized by enabling packaging on wafer-level, which in turn saves time and simplifies the process as the common steps of postprocessing the MEMS micro speakers to achieve a protective packaging are removed. Removing the post-processing steps further reduces the risk of harming the delicate MEMS structures of the micro speakers or introducing dust or other particles into the MEMS micro speaker. Furthermore, patterning of the polymer layer, the common patterning method in the art providing high-precision pattering, enables providing very well defined walls next to the sound producing elements, i.e. the inside walls of the acoustically sealing barrier described herein. This is advantageous if the sound producing device is decoupled, for example a decoupled membrane or a cantilever, which moves relative to said wall during actuation, because the gap between the sound producing device and the wall can be kept very narrow to provide acoustic shielding while at the same time allowing the sound producing device to move freely along the wall.

[0019] The substrate may comprise at least one electrically conductive through connection that is in electrical contact with one or more electrical connection element on the front side of the substrate. Thereby, each of the at least one electrically conductive through connection is arranged to connect the one or more electrical connection element on the front side of the substrate with one or more external component or device, outside of the MEMS micro speaker package. Suitably, an electrical connection element may then be connected to one or more external electrical components via the electrically conductive through connection. The electrically conductive through connections will advantageously not add to the area of the micro speaker packages, i.e. will not increase the footprint of the chip scale micro speaker packages, in contrast to for example wire connections. The at least one electrically conductive through connection may for example in the form of a through silicon via (TSV) or any other suitable through substrate routing solution.

[0020] The substrate may comprise a plurality of through openings, wherein the through openings are distributed over the substrate such that each through opening matches the location of an active sound producing area of the MEMS micro speaker layer in the first pattern. Thereby, each active sound producing area in the resulting MEMS micro speaker package may be connected to a back opening, i.e. opening on the back side of the MEMS micro speaker package, increasing the sound pressure output from the active sound producing area.

[0021] The bonding step may comprise flip-chip mounting of one of the MEMS micro speaker layer and the substrate onto the other. The bonding step then further comprises reflowing the solder to complete the interconnect of the layers. Suitably, flip-chip mounting can be done in short assembly cycle time, since all bonding is completed in one process. The method also contributes to higher signal density, excellent electrical performance, and a low package profile.

[0022] The plurality of MEMS micro speakers may be piezoelectric MEMS micro speakers, each using a piezoelectric material as actuator. Thereby, the power consumption of the MEMS micro speakers may be reduced, which is a great advantage for small-sized applications and other applications that require low power consumption.

[0023] The method may comprise providing a protective layer on the back side of the MEMS micro speaker layer before bonding. Thereby, the sensitive sound producing components and electrical components of the MEMS micro speaker package are further protected. The protective layer may for example comprise a plastic material or a metal material. The protective layer suitably comprises openings that match the active sound producing areas in the first pattern, so that sound produced by each active sound producing area can be output from the front side of the package after bonding.

[0024] The method may comprise applying an adhesive component to the patterned polymer layer or to the front side of the other of the MEMS micro speaker layer and the substrate before bonding. If an adhesive component is applied to the front side of the MEMS micro speaker layer or the substrate, the adhesive is suitably patterned to match the patterned polymer layer in the bonding step. By applying an adhesive component before bonding, the bonding will be even stronger, and the resulting plurality of MEMS micro speaker packages will be more robust.

[0025] The bonding step may comprise using solders or copper pillars arranged between matching electrical connection elements on the MEMS micro speaker layer and the substrate. Thereby, electrical connection between components of the MEMS micro speaker layer and the substrate is enabled after bonding.

[0026] Alternatively, or additionally, to using solder or copper pillars already arranged between matching electrical connection elements on the MEMS micro speaker layer and the substrate before bonding, the method may comprise injecting solders between matching electrical connection elements on the MEMS micro speaker layer and the substrate after bonding.

[0027] The thickness of the polymer layer may be in the range of 20 pm to 2 mm, including the end points of the ranges. Suitably, the polymer layer is thereby thick enough to allow for movements of the sound producing elements, but not so thick that it adds unnecessary height to the resulting MEMS micro speaker packages, so the main purpose of reduction of volume is still fulfilled. In some embodiments, the thickness of the polymer layer is in the range of 100 pm to 300 pm, including the end points of the range. Suitably, this range is well adapted for the oscillation of piezoelectric sound producing elements, specifically piezoelectric membranes, while not adding any unnecessary height to the resulting MEMS micro speaker packages.

[0028] The polymer of the polymer layer may be a photoresist. Thereby, patterning of the polymer layer may suitably be done in manners known in the art, by masking the polymer layer, positively or negatively depending on the type of photoresist used, exposing the masked photoresist to light, and applying a solvent to dissolve the unhardened areas of the photoresist in the polymer layer. In a non-limiting example, the photoresist is SU-8.

[0029] The method may further comprise a step of modifying at least one of the MEMS micro speaker layer, the polymer layer and the optional protective layer at a plurality of locations in connection with the plurality of active sound producing areas in the MEMS micro speaker layer, such that a respective acoustic channel for passing sound from each active sound producing area through an opening in a side wall of the resulting MEMS micro speaker package is formed after dicing.

[0030] The invention also includes a MEMS micro speaker package, as well as a plurality of MEMS micro speaker packages, obtained by the steps and features of any of the method embodiments described herein.

[0031] In a second aspect of the invention, there is provided a MEMS micro speaker package. The MEMS micro speaker package comprises a MEMS micro speaker chip having a front side and a back side, wherein the MEMS micro speaker chip comprises at least one active sound producing area and, for each active sound producing area, at least one electrical connection element electrically connected to an actuator arranged to actuate the active sound producing area. The at least one active sound producing area and the corresponding at least one electrical connection element of the MEMS micro speaker chip are distributed over the front side of the MEMS micro speaker chip according to a second pattern. The MEMS micro speaker package further comprises a substrate having a front side and a back side, wherein the front side of the substrate faces the front side of the MEMS micro speaker chip. The substrate comprises at least one electrical connection element distributed on the front side of the substrate such that each electrical connection element on the substrate matches the location of at least one electrical connection element in the second pattern. The MEMS micro speaker package further comprises a polymer layer sandwiched between and attached to the MEMS micro speaker chip and the substrate. The polymer layer is patterned according to the second pattern to comprise at least one first area that is filled with polymer and second areas that lack polymer, wherein the second areas match the location of the at least one electrical connection element and the at least one active sound producing area in the second pattern, such that the polymer layer creates a sealing barrier around the perimeter of each of the at least one active sound producing area without contacting any part of the at least one active sound producing area and such that the electrical connection elements are not covered by or in contact with the polymer in the polymer layer. Thereby, the MEMS micro speaker chip, the patterned polymer layer and the substrate form a MEMS micro speaker package that has an acoustic seal between each active sound producing area of the MEMS micro speaker chip and the substrate, preventing sound from leaking along the edges of the package. In other words, the patterned polymer layer forms an acoustic seal between the MEMS micro speaker chip and the substrate so that the acoustic output from each active sound producing area in the direction of the substrate is either contained by the substrate or vented through holes in the substrate.

[0032] Suitably, since there is no container adding volume around the chip, in the plane of the main extension of the MEMS micro speaker chip, the obtained MEMS micro speaker package is advantageously a chip scale package. In other words, the MEMS micro speaker package has a footprint very similar to the MEMS chip. The present invention thereby provides a chip-scale MEMS micro speaker package, in likeness with the plurality of MEMS micro speaker packages described in connection with the first aspect. The MEMS micro speaker package is downsized while the ratio of available active sound producing area to chip area is increased compared to known MEMS micro speaker packages, which are not chip scale packages but comprises additional components such as a protective cover, electrical wire connections and / or added material outside the MEMS micro speaker chip area to provide an acoustic and / or protective sealing of the sound generating components, thereby increasing the volume of the resulting packaged product unnecessarily.

[0033] The substrate of the MEMS micro speaker package may be a printed circuit board, PCB.

[0034] The substrate may further comprise at least one electrically conductive through connection. The at least one electrically conductive through connection is in electrical contact with one or more electrical connection element on the front side of the substrate. Thereby, each of the at least one electrically conductive through connection is arranged to connect the one or more electrical connection element on the front side of the substrate with one or more external component or device, outside of the MEMS micro speaker package. Suitably, an electrical connection element may then be connected to one or more external electrical components or devices via such an electrically conductive through connection. The at least one electrically conductive through connection will advantageously not add to the area of the micro speaker package, i.e. will not increase the footprint of the chip scale micro speaker package, in contrast to for example wire connections. The at least one electrically conductive through connection may for example in the form of a through silicon via (TSV) or any other suitable through substrate routing solution.

[0035] The thickness of the polymer layer of the MEMS micro speaker package may be in the range of 20 pm to 2 mm, including the end points of the ranges. Suitably, the polymer layer is thereby thick enough to allow for movements of the sound producing elements, but not so thick that it adds unnecessary height to the MEMS micro speaker package, so the main purpose of reduction of volume is still fulfilled. In some embodiments, the thickness of the polymer layer is in the range of 100 pm to 300 pm, including the end points of the range. Suitably, this range is well adapted for the oscillation of piezoelectric sound producing elements, specifically piezoelectric membranes, while not adding any unnecessary height to the MEMS micro speaker package.

[0036] The polymer of the polymer layer may be a photoresist. Thereby, patterning of the polymer layer may suitably be done in manners known in the art, by masking the polymer layer, positively or negatively depending on the type of photoresist used, exposing the masked photoresist to light, and applying a solvent to dissolve the un- hardened areas of the photoresist in the polymer layer. In a non-limiting example, the photoresist is SU-8.

[0037] The at least one active sound producing area of the MEMS micro speaker chip may be driven using a piezoelectric material as the actuator. In other words, the packaged MEMS micro speaker is in this case a piezoelectric MEMS micro speaker. Thereby, the power consumption of the MEMS micro speaker chip may be reduced, which is a great advantage for small-sized applications and other applications that require low power consumption.

[0038] The MEMS micro speaker package may further comprise an acoustic channel, also referred to herein as a sound passing channel, for passing sound through a side wall of the package 10.

[0039] Any advantage described in connection with one aspect of the invention, e.g. the method for manufacturing a plurality of MEMS micro speaker packages, is equally applicable to corresponding embodiments of other aspects of the invention, e.g. the MEMS micro speaker package.

[0040] Many additional benefits and advantages of the present invention will be readily understood by the skilled person in view of the detailed description below.

[0041] DRAWINGS

[0042] The invention will now be described in more detail with reference to the appended drawings, wherein:

[0043] Fig. 1 discloses a schematic cross-sectional view from the side of a MEMS micro speaker package according to the invention;

[0044] Fig. 2 discloses the layers of the MEMS micro speaker package in Fig. 1;

[0045] Fig. 3 discloses a schematic view of a sealing barrier around the perimeter of an active sound producing area of a MEMS micro speaker;

[0046] Fig. 4 discloses a schematic view of a sealing barrier around the perimeter of an active sound producing area of a MEMS micro speaker;

[0047] Fig. 5 discloses a schematic cross-sectional view from the side of a MEMS micro speaker package according to the invention; Fig. 6 discloses a schematic cross-sectional view from the side of a MEMS micro speaker package according to the invention;

[0048] Fig. 7 discloses a schematic view of the front side of a MEMS micro speaker chip with an active sound producing area and two electrical connection elements distributed over the front side of the MEMS micro speaker chip according to a second pattern;

[0049] Fig. 8 discloses a schematic view of the front side of a substrate with two electrical connection elements distributed over the front side of the substrate such that they match the electrical connection elements on the second pattern;

[0050] Fig. 9 discloses a schematic view of a polymer layer patterned according to the second pattern;

[0051] Fig. 10 discloses a schematic view of a protective layer comprising an opening according to the second pattern;

[0052] Fig. 11 discloses a schematic top view of a wafer level package, before dicing, obtained using a method according to the invention, and a zoomed-in view of a part of the wafer;

[0053] Fig. 12 is a flow chart showing a method for manufacturing a plurality of MEMS micro speaker packages according to embodiments of the invention;

[0054] Figs. 13a-i show process steps for manufacturing a plurality of MEMS micro speaker packages according to embodiments of the invention;

[0055] Fig. 14 discloses a schematic cross-sectional view from the side of a MEMS micro speaker package according to an embodiment of the invention;

[0056] Fig. 15 discloses a schematic cross-sectional view from the side of a MEMS micro speaker package according to an embodiment of the invention;

[0057] Fig. 16 discloses a schematic cross-sectional view from the side of a MEMS micro speaker package according to an embodiment of the invention;

[0058] Fig. 17 discloses a schematic cross-sectional view from the side of a MEMS micro speaker package according to an embodiment of the invention; Fig. 18a-b show optional additional process steps for manufacturing a plurality of MEMS micro speaker packages according to the process of Figs. 13a-i;

[0059] Fig. 19 discloses a schematic cross-sectional view from the side of a MEMS micro speaker package according to an embodiment of the invention;

[0060] Fig. 20 discloses a schematic cross-sectional view from the side of a MEMS micro speaker package according to an embodiment of the invention;

[0061] Fig. 21 discloses a schematic cross-sectional view from the side of a MEMS micro speaker package according to an embodiment of the invention;

[0062] Fig. 22 discloses a schematic cross-sectional view from the side of a MEMS micro speaker package according to an embodiment of the invention; and

[0063] Fig. 23 discloses a schematic cross-sectional view from the side of a MEMS micro speaker package according to an embodiment of the invention.

[0064] All the figures are schematic, not necessarily to scale, and generally only show parts which are necessary in order to elucidate the respective embodiments, whereas other parts may be omitted or merely suggested. Any reference number appearing in multiple drawings refers to the same object or feature throughout the drawings, unless otherwise indicated.

[0065] DETAILED DESCRIPTION

[0066] There is provided a MEMS micro speaker package where patterned polymer forms an acoustic seal between the MEMS chip and a substrate. The package has a similar footprint to the MEMS chip, i.e. it is a chip-scale MEMS micro speaker package. The acoustic seal formed by the patterned polymer layer prevents sound from leaking along the edges of the package. There is further provided a method of manufacturing a plurality of such chip-scale MEMS micro speaker packages by wafer level chip-scale packaging. In other words, the invention enables WLCSP of MEMS micro speakers, and thus also production of chip scale MEMS micro speaker packages. That the invention enables wafer level packaging techniques means that it advantageously allows for the processing of hundreds or thousands of units simultaneously instead of individual sequential packaging. The invention is further not restricted to the described embodiments in the figures but may be varied freely within the scope of the claims. The method of the present invention is for example suitable for packaging a membrane-like MEMS micro speaker.

[0067] As described further below, a polymer layer, for example in the form of a photoresist such as SU-8, may be applied and patterned on top of a MEMS micro speaker layer with openings for the sound producing devices and connection pads / electrical connection elements of a plurality of MEMS micro speakers comprised in the MEMS micro speaker layer. There is further provided a substrate that typically consists of a printed circuit board (PCB) with connection pads / electrical connection elements arranged to match and connect to the MEMS micro speaker layer on one side, and having connection pads / electrical connection elements on its other side for external device connection. The polymer layer may instead of being applied on the MEMS micro speaker layer be applied and patterned on the substrate, with openings the sound producing devices and connection pads / electrical connection elements of a plurality of MEMS micro speakers comprised in the MEMS micro speaker layer, whereby the openings also expose the connection pads / electrical connection elements of the substrate that match the location of the connection pads / electrical connection elements of the MEMS micro speaker layer. However, application of the polymer layer on the MEMS micro speaker layer is preferable as it facilitates alignment in the subsequent bonding. The bonding step may suitably comprise a flip chip process, wherein the PCB is connected to the front side of the MEMS micro speaker layer and the polymer layer is bonded to the PCB by heating and pressing. Any solder between the layers is reflowed to form the electrical connections. A protective layer, which may also be referred to as a lid, may be attached on the top of the device (i.e. to the back side of the MEMS micro speaker layer) for further protection from the environment. As the protective layer is attached before singulation, or dicing, the cover will not extend outside the top area of the resulting MEMS micro speaker packages. The protective layer may comprise a metal plate or a plastic layer with holes for allowing sound output, or an acoustic mesh. All of these steps are done at the wafer level, and as a final step the resulting stack, or package, is singulated into individual finished devices in the form of chip scale MEMS micro speaker packages.

[0068] The main advantage of embodiments herein is the size reduction that comes with a chip scale package, enabling usage of the total footprint of the resulting packaged sound producing device. For smaller micro speakers the relative size reduction is larger, which is why it is especially advantageous for MEMS micro speakers. For small MEMS micro speakers the relative size reduction compared to adding a cover after dicing can reach more than 50% smaller footprint. As a non-limiting example, a MEMS micro speaker package obtained using conventional packaging and having a top area size of 6 x 8.5 mm would, using the inventive method, only have a top area size of 5 x 7, which is a 31 % size reduction. As a further non-limiting example, a MEMS micro speaker package obtained using conventional packaging and having a top area size of 3 x 6 mm would, using the inventive method, only have a top area size of 2 x 4.5, for this smaller device thus giving an even higher size reduction of 45 %.

[0069] Since the micro speaker packages obtained are MEMS micro speaker packages, they are advantageously miniaturized and suitable for application in for example wearable devices, on-ear headphones, earpieces, hearing aids, in-ear speakers, handheld devices such as mobile phones, and other applications which require small sized components.

[0070] According to IPC's standard J-STD-012, Implementation of Flip Chip and Chip Scale Technology, in order to qualify as chip scale, the package must have an area no greater than 1.2 times that of the die. For the chip-scale package to be a wafer level chip scale package, the entire wafer and the added packaging components must further be connected to an interposer before dicing. In other words, the packaging components are not added after dicing, to a single chip. By the use of wafer level chipscale packaging according to embodiments herein, the resulting package will hence practically be of the same size as the die. Thereby, the criteria for wafer level chipscale packaging is well fulfilled by all embodiments herein. Furthermore, the embodiments herein come much closer than this to being actual chip scale, i.e. having the identically same package area (as seen in a top view, also referred to as top area, or footprint) as the area (as seen in a top view, also referred to as top area, or footprint) of the die or chip itself. Herein, this may also be referred to as true chip scale. Especially, for all embodiments except the one described in connection with Figs. 17, 18a and 18b herein, the area of the MEMS micro speaker package 10 is identical (within manufacturing tolerances) to the area of the MEMS micro speaker chip 20 after dicing, since nothing is added to the package 10 after dicing. As is clear to those skilled in the art, also for the embodiment described in connection with Figs. 17, 18a and 18b, the ratio between the package area and the MEMS micro speaker chip area can be kept significantly smaller than 1.2. As shown for different embodiments herein, true chip scale is achieved through the inventive solution to bond the MEMS micro speaker layer (which could also be referred to as the chip or the device layer) directly to the layer forming the lid of the package, herein the substrate 200, via a patterned polymer layer 300, without adding any volume at all outside the chip / MEMS micro speaker layer. This is in contrast to prior solutions that bond two layers or substrates surrounding the chip / device layer to form the package. Such prior solutions may fulfil the requirements of chip scale packaging according to the IPC's standard J-STD-012, but they cannot achieve the even smaller true chip scale packaging.

[0071] It is noted that all sizes, angles, relations etc. given herein are not to be seen as only covering the exact given values but also include minor variations due to manufacturing tolerances. It is also to be noted that features from the various embodiments described herein may freely be combined, unless it is explicitly stated that such a combination would be unsuitable.

[0072] A MEMS micro speaker package 10 according to the invention will now be described in connection with Figs. 1, 2, 5 and 6, and also with reference to Figs. 3, 4, 7-10 and 14.

[0073] Turning first to Fig. 1 there is disclosed a schematic cross-sectional view from the side of a MEMS micro speaker package 10 according to the invention and Fig. 2 discloses the layers of the MEMS micro speaker package 10 in Fig. 1. The MEMS micro speaker package 10 comprises a MEMS micro speaker chip 20, a polymer layer 40 and a substrate 30.

[0074] As illustrated in Fig. 2, the MEMS micro speaker chip 20 has a front side 21 and a back side 22. The back side 22 of the MEMS micro speaker chip hence coincides with the front side 11 of the bonded MEMS micro speaker package 10 shown in Fig. 1 and is the side of the MEMS micro speaker package 10 from which sound is output towards a listener. The MEMS micro speaker chip 20 comprises at least one active sound producing area 111. An active sound producing area 111 may herein also be referred to as a sound producing area 111. As shown in Fig. 6, which discloses a schematic cross-sectional view from the side of a MEMS micro speaker package 10 according to the invention, the MEMS micro speaker chip 20 further comprises for each active sound producing area 111 at least one electrical connection element 50 electrically connected to an actuator (not shown in the figures) arranged to actuate the active sound producing area 111. In many embodiments, there are at least two electrical connection elements 50 on each MEMS micro speaker chip 20, one for connecting to an external device configured to drive the actuator and one for connection to ground. The at least one active sound producing area 111 and the corresponding at least one electrical connection element 50 of the MEMS micro speaker chip 20 are distributed over the front side 21 of the MEMS micro speaker chip 20 according to a pattern, herein referred to as the second pattern. A nonlimiting example of such a MEMS micro speaker chip 20 is schematically disclosed in Fig. 7, which shows a view of the front side 21 of a MEMS micro speaker chip 20 with one active sound producing area 111 and two electrical connection elements 50 distributed over the front side 21 of the MEMS micro speaker chip 20 according to a second pattern. Beside the active sound producing area(s) 111 and electrical connection elements 50 described herein, the MEMS micro speaker chip 20 and / or the substrate 30 may comprise further components, opening etc. The distribution of any such components over the front side of the MEMS micro speaker chip 20 and substrate 30 are then also part of the second pattern.

[0075] The substrate 30 has a front side 31 and a back side 32, wherein the front side 31 of the substrate faces the front side 21 of the MEMS micro speaker chip 20. As shown in Fig. 6, the substrate 30 further comprises at least one electrical connection element 50. The at least one electrical connection element 50 is / are arranged or distributed on the front side 31 of the substrate 30 such that each electrical connection element 50 on the substrate 30 matches the location of an electrical connection element 50 in the second pattern. That each electrical connection element 50 on the substrate 30 matches the location of an electrical connection element 50 in the second pattern means that, as can be seen from Fig. 6, each electrical connection element 50 on the substrate 30 is aligned with one (or more) electrical connection element 50 on the MEMS micro speaker chip 20 in the bonded MEMS micro speaker package 10. Thereby, electrical connection between the matching, or aligned, electrical connection elements 50 on both layers is enabled, for example using solder 55 to complete the electrical connection. A non-limiting example of such a substrate 30 is schematically disclosed in Fig. 8, which shows a schematic view of the front side 31 of a substrate 30 with two electrical connection elements 50 distributed over the front side 31 of the substrate 30 such that they match the electrical connection elements 50 in the second pattern of Fig. 7. The substrate 30 of Fig. 8 may also comprise an opening 33 matching the active sound producing area 111 in the second pattern of Fig. 7. In this example, the MEMS micro speaker chip 20 and the substrate 30 are flip chip mounted in the resulting MEMS micro speaker package 10 with their front sides 21, 31 facing each other, which is why the placement of the components are mirrored in Figs. 7 and 8.

[0076] The polymer layer 40 is sandwiched between and attached to the MEMS micro speaker chip 20 and the substrate 30, as shown in Figs. 1, 5 and 6. The polymer layer 40 is patterned according to the second pattern to comprise at least one first area 41 that is filled with polymer and second areas 42 that lack polymer. The second areas 42 match the location of the at least one electrical connection element 50 and the at least one active sound producing area 111 in the second pattern, such that the polymer layer 40 creates a sealing barrier 310 around the perimeter 311 of each of the at least one sound producing area 111 without contacting any part of the at least one sound producing area 111 and such that the electrical connection elements 50 are not covered by or in contact with the polymer in the polymer layer 40, thereby forming a MEMS micro speaker package 10 having an acoustic seal between each active sound producing area 111 of the MEMS micro speaker chip 20 and the substrate 30, while leaving the electrical connection elements 50 inside the MEMS micro speaker package uncovered by polymer, enabling electrical connection thereto.

[0077] A sealing barrier 310, acoustic barrier, or acoustically sealing barrier 310, is herein defined as a barrier that directs the output sound of the MEMS micro speakers in one or more desired direction and hinders it / seals it off from leaking in any unwanted direction, i.e. hinders the output sound from leaking in unintended directions. Thereby, an optimized sound pressure is achieved in the wanted, intended, output direction as acoustic losses are minimized.

[0078] A non-limiting example of a patterned polymer layer 40, 300 with a first area 41, 301 and second areas 42, 302 is shown in Fig. 9. Fig. 9 discloses a schematic view of a polymer layer 40, 300 patterned according to the second pattern of Fig. 7, wherein the second areas 42, 302 match the location of the in this case two electrical connection elements 50 and one active sound producing area 111 in the second pattern of Fig. 7, such that the polymer layer 40, 300 creates a sealing barrier 310 around the perimeter 311 of each of the at least one sound producing area 111 without contacting any part of the at least one sound producing area 111 and such that the electrical connection elements 50 are not covered by or in contact with the polymer in the polymer layer 40, 300. Thereby, the layers of Figs. 7 to 9 together form a MEMS micro speaker package 10 that has an acoustic seal between each active sound producing area 111 of the MEMS micro speaker chip 20 and the substrate 30, while leaving the electrical connection elements 50 inside the MEMS micro speaker package uncovered by polymer, enabling electrical connection thereto.

[0079] Optionally, as shown in Figs. 5, 6 and 10, the MEMS micro speaker package 10 may further comprise a protective layer 60 attached to the back side 22 of the MEMS micro speaker chip 20. Thereby, the sensitive components of the MEMS micro speaker package 10 are further protected. The protective layer 60 may for example comprise a plastic material or a metal material, with openings 61 that match, i.e. at least partly overlap as seen in a top view of the MEMS micro speaker package 10, each sound producing area 111, so that sound produced by each sound producing area 111 can be output from the front side 11 of the MEMS micro speaker package 10. An example of a protective layer 60 is shown in Fig. 10, which discloses a schematic view of a protective layer 60 comprising an opening 61 according to the second pattern of Fig. 7, for exposing and thus allowing output from the sound producing area 111 shown in Fig. 7. The same protective layer 60 with an opening 61 is shown in the schematic cross-sectional view from the side of the MEMS micro speaker package in Fig. 5. In some embodiments, as an alternative to each sound producing area 111 being connected to exactly one such opening 61 in the protective layer 60, a sound producing area 111 may instead be connected to more than one opening 61, as illustrated in Fig. 14. Alternatively, the protective layer 60 may comprise a sound passing mesh, whereby openings are not necessary to allow sound output from the MEMS micro speaker package 10. The protective layer 60 may be attached in any suitable manner, e.g. using an adhesive but not limited thereto.

[0080] The substrate 30 may further comprise at least one electrically conductive through connection 51, such as a through silicon via (TSV). An example of an electrically conductive through connection 51 in the form of a TSV for electrically connecting two electrical connection elements 50 is shown in Fig. 6.

[0081] The substrate 30 may further comprise at least one through opening, or back opening, 33, wherein each of the at least one back opening 33 matches the location of an active sound producing area 111 of the MEMS micro speaker chip 20 in the second pattern such that the at least one active sound producing area 111 is exposed to the back side 12 of the MEMS micro speaker package 10 by a through opening, or back opening, 33 in the substrate 30. Examples of such back openings 33 are shown in Figs. 5, 6, 8, 14, 15 and 16. In some embodiments, as an alternative to each sound producing area 111 being connected to exactly one back opening 33, a sound producing area 111 may instead be connected to more than one back opening 33, as illustrated in Figs. 14 and 15. In further alternative embodiments, more than one sound producing area 111 can share a common back opening 33, as illustrated in the example of Fig. 16. As a further advantage, the dimensions of the at least one back opening 33 through the substrate 30 may suitably be chosen so that the resulting acoustic impedance of the system creates a damping effect on frequencies around a matching resonance frequency in the sound producing area 111. To reduce the risk of dust and particles reaching the sound producing elements through a back opening, the MEMS micro speaker package 10 may in these embodiments further comprise a mesh (not shown in the figures) disposed on the back side 32 of the substrate 30 and covering the at least one back opening 33. Alternatively, if there is no back opening 33 connected to an active sound producing area 111, the active sound producing area 111 will instead be connected to a closed back cavity 34 which still increases the sound pressure output from the active sound producing area 111, but to a lower degree. The closed cavity alternative instead have the advantages of reducing the risk of dust and particles reaching the sound producing elements, that fewer processing steps are needed to obtain the substrate 30, and that the substrate 30 is more robust as it does not comprise any openings 33. Examples of a closed cavity 34 are shown in Figs. 5 and 6, wherein dashed lines indicate that the substrate 30 may comprise at least one back opening 33, or have no opening whereby a closed cavity 34 is formed by the non-covered area of the polymer layer 40 between the substrate 30 and the sound producing area 111 of the MEMS micro speaker chip 20.

[0082] The at least one sound producing area 111 of the MEMS micro speaker chip 10 may be driven using a piezoelectric material as the actuator. In other words, the packaged MEMS micro speaker chip 10 is in this case a piezoelectric MEMS micro speaker. Thereby, the power consumption of the MEMS micro speaker chip 10 may be reduced, which is a great advantage for small-sized applications and other applications that require low power consumption. The actuator may in this case be a thin film actuator, such as a piezoelectric actuator. In an embodiment, the material may be made of lead zirconate titanate (PZT).

[0083] The invention further includes a method for manufacturing a plurality of MEMS micro speaker packages 10. Embodiments of the method will now be described in connection with Figs. 12 and 13a-i, with reference to Figs. 11, 14, 15 and 16. Fig. 12 is a flow chart showing a method for manufacturing a plurality of MEMS micro speaker packages and Figs. 13a-i illustrate process steps for manufacturing a plurality of MEMS micro speaker packages according to embodiments of the method in Fig. 12. The method comprises:

[0084] In step 1210: providing a MEMS micro speaker layer 100.

[0085] As illustrated in Fig. 13a, the MEMS micro speaker layer 100 has a front side 101 and a back side 102. The MEMS micro speaker layer 100 comprises a plurality of MEMS micro speakers 110, wherein each MEMS micro speaker 110 has at least one active sound producing area 111 and, for each active sound producing area 111, at least one electrical connection element 50 electrically connected to an actuator arranged to actuate the active sound producing area 111. In many embodiments, each MEMS micro speaker 110 comprises at least two electrical connection elements 50 each active sound producing area 111, one for connecting to an external device configured to drive the actuator and one for connection to ground. The active sound producing areas 111 and the corresponding electrical connection elements 50 of the plurality of MEMS micro speakers 110 are distributed over the front side 101 of the MEMS micro speaker layer 100 according to a first pattern.

[0086] In step 1220: providing a substrate 200.

[0087] As illustrated in Fig. 13d, the substrate 200 has a front side 201 and a back side 202. The substrate comprises electrical connection elements 50 distributed on the front side 201 of the substrate 200 such that each electrical connection element 50 on the substrate 200 matches the location of at least one electrical connection element 50 in the first pattern, i.e. in this example matches the distribution of the electrical connection element 50 of the MEMS micro speaker layer 100 in Fig. 13a.

[0088] As shown in Fig. 13e, the MEMS micro speaker layer 13e may be selectively etched from the back side 102, using any suitable method known in the art, to create the plurality of sound producing areas 111 and sound producing elements 112 described herein. This etching may be performed at the beginning of the manufacturing process or preferably, as illustrated in Fig. 13e, after patterning of the polymer layer 300 in step 1240.

[0089] Fig. 13d also illustrates that in some embodiments the substrate 200 may already at this stage in the manufacturing process have solder 55, for example in the form of solder paste, applied in connection with each electrical connection element 50, to enable electrical connection with the electrical connection element 50 of the MEMS micro speaker layer 100 during bonding. Alternatively, solder 55 may be applied in connection with, and in electrical connection with, each electrical connection element 50 on the substrate 200 or on the MEMS micro speaker layer 100 before bonding. If the solder 55 is applied to the same one of the substrate 200 and the MEMS micro speaker layer 100 as the polymer layer 300 in step 1230, the solder must of course be applied after patterning in step 1240 so that the electrical connection elements 50 are exposed. As a further alternative, instead of solder, copper pillars (not shown in the figures) may be used to provide electrical connections between matching electrical connection elements 50 of the substrate 200 and the MEMS micro speaker layer 100.

[0090] In this case, a plurality of copper pillars are applied or created on the MEMS micro speaker layer 100 in connection and electrical connection with each electrical connection element 50 on the MEMS micro speaker layer 100. The height of the copper pillars is suitably slightly higher than the thickness of the polymer layer 300 applied in step 1230. If copper pillars are used, the method further comprises applying a layer of anisotropic conductive adhesive (ACA) to the front side 201 of the substrate 200. When the copper pillars are pressed against the ACA layer during the bonding step 1250, the copper pillars will both adhere to the substrate 200 at the matching electrical connection elements 50 of the substrate 200, thereby enhancing the bonding of the layers, and at the same time complete the electrical connection between the matching electrical connection elements 50 of the substrate 200 and the MEMS micro speaker layer 100 as the ACA becomes conductive in the areas where it meets and is compressed by the meeting surfaces of the copper pillars. All other areas of the ACA layer, i.e. uncompressed parts of the ACA layer, remain non-conductive. Thereby, the application of the ACA layer is an easy process step, as a thin and even layer can be applied onto the entire substrate surface without the need for high precision application patterning or masking. Furthermore, advantageously both the bonding of the polymer layer 300 and the copper pillars to the substrate 200 and completion of the electrical interconnect between the MEMS micro speaker layer 100 and the substrate 200 is achieved in a single processing step.

[0091] The substrate 200 may comprise a plurality of electrically conductive through connection 51, each through connection 51 being in electrical contact with an electrical connection element 50 on the front side of the substrate 200 for connecting the electrical connection element 50 on the front side of the substrate 200 with an external component or device. Each electrically conductive through connection 51 may be a TSV.

[0092] The substrate 200 may further comprise a plurality of through openings, or back openings, 33 distributed over the substrate 200 such that each through opening 33 matches the location of an active sound producing area 111 of the MEMS micro speaker layer 100 in the first pattern. Thereby, as shown in Figs. 13f-i, each active sound producing area 111 will after bonding in step 1250 be exposed to the back side 402 of the bonded package 400, and the corresponding backside 12 of the resulting MEMS micro speaker package 10 after dicing, by one or more through opening, or back opening, 33 in the substrate 30. In some embodiments, as an alternative to each sound producing area 111 being connected to exactly one back opening 33, a sound producing area 111 may instead be connected to more than one back opening 33, as illustrated in Figs. 14 and 15. In further alternative embodiments, more than one sound producing area 111 can share a common back opening 33, as illustrated in the example of Fig. 16.

[0093] In step 1230: applying a polymer layer 300 to one of the front side 101 of the MEMS micro speaker layer 100 and the front side 201 of the substrate 200.

[0094] In Fig. 13b, a polymer layer 300 is applied to the front side 101 of the MEMS micro speaker layer 100. This is typically preferable, as it is easier to align the patterning of the polymer layer in step 1240 with the sound producing areas 111 of the MEMS micro speaker layer 100 than it is to align an already patterned polymer layer, on the substrate 200, with the sound producing areas 111 of the MEMS micro speaker layer 100 in the bonding step 1250. A misalignment may result in a failure of acoustically sealing one or more sound producing area 111, leading to a deterioration of the sound output. Also, if an additional adhesive is needed in the bonding step, it is more preferable to add the adhesive to the substrate 200 than to the MEMS micro speaker layer 100, since the surface of the MEMS micro speaker layer 100 is more sensitive. However, if it is more suitable for a specific manufacturing process, the polymer layer 300 may instead be applied on the front side 201 of the substrate 200 in this step. The bonding can still be successful if enough caution is taken in alignment of the layers.

[0095] Applying the polymer layer 300 may comprise spinning, spin coating, or laminating a polymer onto the front side 101 of said one of the MEMS micro speaker layer 100 and the substrate 200. In step 1240: patterning the polymer layer 300 according to the first pattern, to comprise at least one first area 301 that is filled with polymer and second areas 302 that lack polymer, wherein the second areas 302 match the locations of the electrical connection elements 50 and active sound producing areas 111 in the first pattern.

[0096] Fig. 13c shows the polymer layer 300 of Fig. 13b after it has been patterned according to the first pattern, such that the electrical connection elements 50 and the sound producing areas 111 of the MEMS micro speaker layer 100 are left uncovered.

[0097] The first pattern thereby suitably defines where openings, i.e. second areas 302, are to be located in the patterned polymer of the polymer layer 300, to enable components of the MEMS micro speaker layer 100 and the substrate 200 to have electrical connection with each other and / or external devices, so that the components can function as intended after the layers of the package 400 have been bonded together in step 1250. Beside the active sound producing area(s) 111 and electrical connection elements 50 described herein, the MEMS micro speaker layer 100 and / or the substrate 200 may comprise further components, opening etc. The distribution of any such components over the front side of the MEMS micro speaker layer 100 and substrate 200 are then also part of the first pattern.

[0098] In step 1250: bonding the polymer layer 300 to the other of the front side 101 of the MEMS micro speaker layer 100 and the front side 201 of the substrate 200.

[0099] Bonding includes any combination of hot press bonding and / or use of one or more adhesive as described herein.

[0100] Thereby, as shown in Fig. 13f, a wafer level package 400, or package array, is formed wherein the polymer layer 300 is sandwiched between the MEMS micro speaker layer 100 and the substrate 200, such that the polymer layer 300 creates a sealing barrier 310 around a perimeter 311 of each of the sound producing areas 111 without contacting any part of the sound producing areas 111 and such that the electrical connection elements 50 are not covered by or in contact with the polymer in the polymer layer 300.

[0101] Hence, an acoustic seal is formed between each active sound producing area 111 in the MEMS micro speaker layer 100 and the substrate 200, which prevents sound from leaking along the edges of the package 400 and the resulting plurality of MEMS micro speaker packages 10 obtained after the dicing in step 1260. In other words, the patterned polymer layer 300 forms an acoustic seal between the MEMS micro speaker layer 100 and the substrate 200 so that the acoustic output from the active sound producing areas 111 is contained by the substrate 200 or vented through holes, i.e. back openings 33, in the substrate 200.

[0102] The bonding step 1250 may suitably comprise a flip chip process, wherein the front side 201 of the substrate 200 is connected to the front side 101 of the MEMS micro speaker layer 100 with the polymer layer 300 sandwiched between them. The polymer layer 300 is then bonded to the front side 101, 201 of the other of the MEMS micro speaker layer 100 and the substrate, i.e. bonded to the front side of the layer that it was not applied on in step 1230. The bonding is performed in known manners by heating and pressing. Any solder 50 present between the layers is also reflowed to complete the interconnect of the layers and form the electrical connections.

[0103] The bonding step 1250 may comprise using solder 55 or copper pillars arranged between matching electrical connection elements 50 on the MEMS micro speaker layer 100 and the substrate 200. Thereby, electrical connection between the matching electrical connection elements 50 of the MEMS micro speaker layer 100 and the substrate 200 is enabled after bonding. Alternatively, or additionally, to using solder 55 or copper pillars already arranged between matching electrical connection elements 50 on the MEMS micro speaker layer and the substrate before bonding, the method may comprise injecting solder 55 between matching electrical connection elements 50 on the MEMS micro speaker layer 100 and the substrate 200 after bonding. This may be a suitable solution if it is hard to do a one-step process for bonding both solder 55 and the polymer layer 300.

[0104] In step 1260: dicing the package 400 to obtain the plurality of MEMS micro speaker packages 10.

[0105] Advantageously, wafer-level packaging of MEMS micro speakers is thereby provided, resulting in chip scale MEMS micro speaker packages 10, that do not require any further post processing packaging steps after dicing. Examples of such chip scale MEMS micro speaker packages 10 are schematically shown in Fig. 13h, which shows the result of dicing the package 400 in Fig. 13f. Suitably, the side walls of the MEMS micro speaker packages 10 will in this embodiment remain straight after the dicing step, which facilitates subsequent mounting in other devices. All the layers in each MEMS micro speaker package 10 thereby have the same outline size and shape and are in alignment with each other. In other words, the resulting MEMS micro speaker packages 10 are in this embodiment true chip scale packages. The side walls of each MEMS micro speaker package 10 are the outer walls of the MEMS micro speaker package 10 that are parallel to a normal N of the front side 11 of the MEMS micro speaker package 10. An example of such a normal N is indicated in Fig. 1. That the side walls are straight, or parallel to the normal N, is here to be understood as the walls being straight, or parallel to the normal N, within manufacturing tolerances. Preferably, this means that the outline of each side wall is defined within 20 pm, more preferably within 10 pm.

[0106] Fig. 11 illustrated a schematic top view of a wafer level package 400 obtained by the method described herein, before dicing, where a substrate 200 and a MEMS micro speaker layer 100 with a plurality of MEMS micro speakers 110 can be seen. Fig. 11 further shows a zoomed-in view of a part of the wafer level package 400 showing four of the MEMS micro speakers 110 in more detail.

[0107] The method may further comprise providing a protective layer 60 on the back side 102 of the MEMS micro speaker layer 100 before the bonding of step 1250. After bonding the protective layer will then form the back side 401 of the bonded package 400. By adding the protective layer 60, the sensitive sound producing components and electrical components of the bonded package 400, and the resulting MEMS micro speaker packages 10 after dicing, are further protected. The protective layer 60 may for example comprise a plastic material or a metal material, with openings 61 that match, i.e. at least partly overlap as seen in a top view of the MEMS micro speaker package 10, each sound producing area 111, so that sound produced by each sound producing area 111 can be output from the front side 11 of the MEMS micro speaker package 10. A protective layer 60 provided on the back side 102 of the MEMS micro speaker layer 100 is shown in Fig. 13g. The protective layer 60 in Fig. 13g comprises openings 61 matching the sound producing areas 111 according to the first pattern, thus exposing and allowing output from the sound producing areas 111 of the resulting MEMS micro speaker packages 10 after dicing. Examples of such chip scale MEMS micro speaker packages 10 are schematically shown in Fig. 13i, which shows the result of dicing the package 400 in Fig. Suitably, similar to the MEMS micro speaker packages 10 shown in Fig. 13h, the side walls of the MEMS micro speaker packages 10 will in this embodiment remain straight (within manufacturing tolerances) after the dicing step, which facilitates subsequent mounting in other devices. All the layers in each MEMS micro speaker package 10 thereby have the same outline size and shape and are in alignment with each other. In other words, the resulting MEMS micro speaker packages 10 are in this embodiment true chip scale packages. In some embodiments, as an alternative to each sound producing area 111 being connected to exactly one such opening 61 in the protective layer 60, a sound producing area 111 may instead be connected to more than one opening 61, as illustrated in Fig. 14. Alternatively, the protective layer 60 may comprise a sound passing mesh, whereby openings are not necessary to allow sound output from the MEMS micro speaker package 10. The protective layer 60 may be attached in any suitable manner, e.g. using an adhesive but not limited thereto.

[0108] The method may further comprise applying an additional adhesive component to the patterned polymer layer 300 or to the front side 101, 201 of the other of the MEMS micro speaker layer 100 and the substrate 200, i.e. the front side of the layer that the polymer layer 300 is to be bonded to, before the bonding of step 1250. If an adhesive component is applied to the front side 101, 201 of the MEMS micro speaker layer 100 or the substrate 200, the adhesive is suitably patterned to match the patterned polymer layer 300 in the subsequent bonding step 1250. If an adhesive component is added to the patterned polymer layer 300, it may advantageously be applied by stamp dipping the layer with the patterned polymer layer (the one of the MEMS micro speaker layer 100 or substrate 200) in the adhesive component with the patterned polymer layer 300 facing the adhesive. Thereby, an even layer of adhesive can suitably be obtained only on the protruding areas of the patterned polymer layer 300 in a single, easy, step, without the need for any high precision patterning or masking. By applying an adhesive component before bonding, the bonding will be even stronger, and the resulting plurality of MEMS micro speaker packages 10 will be more robust.

[0109] In some embodiments it may be desirable to further protect the resulting MEMS micro speaker package 10, after dicing, with an additional cover 70. Fig. 17 schematically discloses a cross-sectional view from the side of an example MEMS micro speaker package 10 with an additional cover 70. Turning to Figs. 18a and b, optional process steps are shown for obtaining the MEMS micro speaker package 10 of Fig. 17. Fig. 18a illustrates a step of etching or in other suitable manners making a narrow opening from the front side 401 of the package 400 through the MEMS micro speaker layer 100 and the polymer layer 300 between each pair of adjacent MEMS micro speakers 110 in the MEMS micro speaker layer 100 and at the other edges of the outer MEMS micro speakers 110 of the MEMS micro speaker layer 100. After the dicing step 1260, each of the plurality of MEMS micro speaker packages 10 will then have a substrate 30 with a slightly larger area than its MEMS micro speaker chip 20 and polymer layer 40, as shown in Fig. 18b. Then, a cover 70 may be added to form the resulting MEMS micro speaker package 10 of Fig. 17. The resulting MEMS micro speaker packages 10 of this embodiment advantageously, similarly to the other embodiments, is acoustically sealed already at wafer level, and hence has all advantages described herein in relation to the wafer level processing. Of course, what is shown in Figs. 17, 18a and 18 is just one example. Any method embodiment described herein may be modified to include a step of providing openings 113 and any device embodiment herein may correspondingly be modified to include an additional cover 70 if this is desirable. Suitably, by the added wafer level process step of Fig. 18a, the MEMS micro speaker packages 10 are also prepared for adding a further cover, for embodiments wherein this is found necessary or desirable. Compared to the other embodiment herein, however, some of the MEMS chip area is lost when obtaining the openings 113.

[0110] Advantageously, the dimensions of the at least one back opening 33 through the substrate 200 may suitably be chosen so that the resulting acoustic impedance of the system creates a damping effect on frequencies around a matching resonance frequency in the sound producing area 111. To reduce the risk of dust and particles reaching the sound producing elements through a back opening, the method may in these embodiments further comprise attaching a mesh (not shown in the figures) on the back side 202 of the substrate 200 and covering the at least one back opening 33. Alternatively, if there is no back opening 33 connected to an active sound producing area 111, the active sound producing area 111 will instead be connected to a closed back cavity 34 which still increases the sound pressure output from the active sound producing area 111, but to a lower degree. The closed cavity alternative instead have the advantages of reducing the risk of dust and particles reaching the sound producing elements, and that the substrate 200 is more robust as it does not comprise any openings. Examples of a closed cavity 34 are shown in Figs. 5 and 6, wherein dashed lines indicate that the substrate may comprise at least one back opening 33, or have no opening whereby a closed cavity 34 is formed by the noncovered area of the polymer layer 40, 300 between the substrate 30, 200 and the sound producing area 111.

[0111] In some applications it may be desirable to design the MEMS micro speaker package 10 to include an acoustic channel, also referred to herein as a sound passing channel, for passing sound through a side wall of the package 10. Figs. 19 to 23 disclose schematic cross-sectional views from the side of a MEMS micro speaker package according to embodiments comprising such an acoustic channel 80. Any method embodiment described herein may correspondingly comprise a step of modifying at least one of the MEMS micro speaker layer 100, the polymer layer 300 and the optional protective layer 60 at a plurality of locations in connection with the sound producing area 111 in the MEMS micro speaker layer 100, such that a respective acoustic channel 80 for passing sound from each sound producing area 111 through an opening in a side wall of the resulting MEMS micro speaker package 10 is formed after dicing. Advantageously, a side firing version of the MEMS micro speaker package 10, i.e. enabling output of sound from one or more side of the MEMS micro speaker package 10, is thereby obtained.

[0112] At least one acoustic channel 80 according to any of the embodiments of Figs. 19 to 23, may be included for each MEMS micro speaker package 10 in any of the method and package embodiments described herein. Each acoustic channel 80 may be connected to one or more sound producing area 111 for passing sound from that sound producing area or areas 111 into the surrounding air via a side wall opening. Below, in reference to the figures, the examples show a single acoustic channel 80 in connection to a single sound producing area 111. This is not a limitation, but only for ease of illustration.

[0113] Turning to Figs. 19 and 20, there are shown two MEMS micro speaker packages 10 each comprising an acoustic channel 80 for passing sound from a sound producing area 111 through a side wall of the MEMS micro speaker package 10, wherein the acoustic channel 80 is formed by an opening in the MEMS micro speaker chip 20 connected to the sound producing area 111. Any method embodiment described herein may correspondingly be modified to include a step of selectively removing the MEMS micro speaker layer 100 in a plurality of locations connected to each sound producing area 111 of the MEMS micro speakers 110 in the MEMS micro speaker layer 100. Thereby, a respective acoustic channel 80 for passing sound from each sound producing area 111 through an opening in a side wall of the resulting MEMS micro speaker package 10 after dicing is formed. Suitably, such openings in the MEMS micro speaker layer 100 for each MEMS micro speaker 110 may be provided as part of the step of selectively etching the MEMS micro speaker layer 100 from the back side 102 to create the plurality of sound producing areas 111 and sound producing elements 112, as described in connection with Fig. 13e. Fig. 21 illustrates an embodiments of MEMS micro speaker package 10 comprising an acoustic channel 80 for passing sound from a sound producing area 111 through a side wall of the MEMS micro speaker package 10, wherein the acoustic channel 80 is formed in the polymer layer 40. Thereby, the sealing barrier 310 around the sound producing area 111 directs the output sound of the MEMS micro speaker chip 20 out through the side wall, while still hindering it from leaking in any unwanted direction. Any method embodiment described herein may correspondingly be modified to include a step of selectively removing the applied polymer layer 300 in a plurality of locations connected to each sound producing area 111 of the MEMS micro speakers

[0114] 110 in the MEMS micro speaker layer 100. Thereby, a respective acoustic channel 80 for passing sound from each sound producing area 111 through an opening in a side wall of the resulting MEMS micro speaker package 10 after dicing is formed. Suitably, such openings in the MEMS micro speaker layer 100 for each MEMS micro speaker 110 may be achieved by selectively removing the applied polymer layer 300 before, during, or after the patterning step 1240. Selectively removing the applied polymer layer 300 may for example include selectively etching the polymer layer 300 after any of the process steps of Fig. 13b, 13c or 13e, in any suitable manner known in the art. In each of the selected sections, the polymer layer 300 may be removed partially, as illustrated in Fig. 19, or completely, as illustrated in Fig. 20, to obtain suitable dimensions for the acoustic channels 80 after dicing.

[0115] Figs. 22 and 23 illustrate embodiments of MEMS micro speaker packages 10 comprising an acoustic channel 80 for passing sound from a sound producing area

[0116] 111 through a side wall of the MEMS micro speaker package 10, wherein the acoustic channel 80 is formed in the protective layer 60. In the embodiment of fig. 22, the applied protective layer 60 is substantially thicker at a first side edge 62 that is a part of a first side wall of the MEMS micro speaker package 10 than at a second side edge 63 that is a part of a second, opposing, side wall of the MEMS micro speaker package 10, and there is a distance between the MEMS micro speaker chip 20 and the protective layer 60 at the second side of the MEMS micro speaker package 10. Alternatively, or additionally, to having a protective layer 60 with variable thickness, the protective layer 60 may be raised at the second side edge 63 compared to the first side edge 62 to obtain a distance between the MEMS micro speaker chip 20 and the protective layer 60 at the second side of the MEMS micro speaker package 10. This is illustrated in Fig. 23. Thereby, an acoustic channel 80 for passing sound from a sound producing area 111 through a side wall of the MEMS micro speaker package 10 is achieved in the second, opposing, side wall of the MEMS micro speaker package 10. Any method embodiment described herein that comprises application of a protective layer 60 may correspondingly be modified to include a step of forming the protective layer 60 connected to each sound producing area 111 of the MEMS micro speakers 110 in the MEMS micro speaker layer 100 such that the there is a distance between the MEMS micro speaker layer 100 and the protective layer 60 at one side of each MEMS micro speaker 110 in the MEMS micro speaker layer 100. Thereby, a respective acoustic channel 80 for passing sound from each sound producing area 111 through an opening in a side wall of the resulting MEMS micro speaker package 10 after dicing is formed. The protective layer 60 may be formed to have a varying thickness and / or raised sections such that there is a distance between the MEMS micro speaker layer 100 and the protective layer 60 at one side of each MEMS micro speaker 110 in the MEMS micro speaker layer 100. Preferably, the protective layer 60 is formed before the protective layer 60 is provided on the back side 102 of the MEMS micro speaker layer 100.

[0117] The MEMS micro speaker layer 100 is typically made of a semiconductor material, preferably silicon.

[0118] The plurality of MEMS micro speakers 110 may be a piezoelectric MEMS micro speakers, each using a piezoelectric material as actuator. Thereby, the power consumption of the MEMS micro speakers may be reduced, which is a great advantage for small-sized applications and other applications that require low power consumption.

[0119] The invention also includes a MEMS micro speaker package 10, as well as a plurality of MEMS micro speaker packages 10, obtained by the steps and features of any of the method and process embodiments described herein in connection with Figs. 12 and 13a-23.

[0120] In any embodiment herein, the MEMS micro speaker layer 100 or the MEMS micro speaker chip 20 need not comprise only a single layer of sound producing elements as shown in the figures. Any suitable alternative micro speaker design may be used. The packaging method herein works for any MEMS micro speaker design, including but not limited to having the sound producing element, for example membranes, on the back side of the MEMS micro speaker layer or MEMS micro speaker chip and using TSVs or the like for electrical connection through the MEMS micro speaker layer or MEMS micro speaker chip, and / or having a multilayer structure providing sound from ultrasound, or stacking sound producing layers in manners known in the art, etc.

[0121] Any active sound producing area 111 described herein may comprise a single sound producing element, such as a flexible membrane and / or cantilever but not limited thereto, over the entire active sound producing area 111. That the polymer layer 40, 300 creates a sealing barrier 310 around the perimeter 311 of the sound producing area 111 in this case means that it creates a sealing barrier 310 around the perimeter 311 of the single sound producing element within the sound producing area 111, without contacting any part of the sound producing area 111. That the sealing barrier 310 does not contact any part of the sound producing area 111 in this embodiment of course means that is also does not contact the sound producing element within the active sound producing area 111. The sealing barrier 310 may be arranged at a suitable distance from the perimeter 311 of the sound producing area 111 in a plane parallel to the plane of extension of the polymer layer 40, 300, and thus at an equal distance from the sound producing element in this embodiment. A non-limiting example is illustrated in Fig. 3, which discloses a schematic view wherein the patterned polymer of the polymer layer 40, 300 creates a sealing barrier 310 around the perimeter 311 of an active sound producing area 111 comprising a single sound producing element 112, in this example covering the entire sound producing area 111. Thereby, in the bonded package 400 and the MEMS micro speaker package 10 after dicing, the polymer layer 40, 300 forms an acoustic seal between the active sound producing area 111 and the substrate 30, 200.

[0122] Alternatively, an active sound producing area 111 herein may comprise two or more sound producing elements, for example membranes but not limited thereto, within the sound producing area 111. That the polymer layer 40, 300 creates a sealing barrier 310 around the perimeter 311 of the sound producing area 111 in this case means that it creates one common sealing barrier 310 around the perimeter of the two or more sound producing element within the sound producing area 111, without contacting any part of the sound producing area 111, and thus not contacting any of the two or more sound producing element. The perimeter may also in this embodiment be arranged at a suitable distance from the perimeter of the sound producing area 111, and thus at a distance from each of the two or more sound producing element, in a plane parallel to the plane of extension of the polymer layer 40, 300. The shape and placement of the two or more sound producing element may be freely varied to suit the application. A non-limiting example is illustrated in Fig. 4 which discloses a schematic view wherein the patterned polymer of the polymer layer 40, 300 creates a sealing barrier 310 around the perimeter 311 of an active sound producing area 111 comprising three sound producing elements 112’, 112”, 112”’. Thereby, in the bonded package 400 and the MEMS micro speaker package 10 after dicing, the polymer layer 40, 300 forms an acoustic seal between the active sound producing area 111 by enclosing the sound producing area 111 with the three sound producing elements 112’, 112”, 112’” with a common sealing barrier 310.

[0123] A plane parallel to the plane of extension of the polymer layer 40, 300, or parallel to any of the MEMS micro speaker layer 100, the MEMS micro speaker chip 20, or the substrate 30, 200, may also be referred to as a plane perpendicular to the normal N of the front side 11 of the MEMS micro speaker package 10 or the front side 401 of the package 400. Examples of such a normal N are indicated in Figs. 1, 13f and 13g.

[0124] In any embodiment herein, the sound producing area 111 may be centered within the sealing barrier, but any other suitable design that fulfils the criteria of the polymer layer 300 creates a sealing barrier 310 around the perimeter 311 of the sound producing area 111 and the sealing barrier 310 not being in direct contact with any part of the sound producing area 111, specifically including not being in direct contact with any part of a sound producing element, is equally feasible.

[0125] Any or all of the sound producing elements comprised in the active sound producing areas 111 of MEMS micro speakers according to embodiments of the invention may be flexible membranes or other thin layer structures, and / or cantilevers, arranged to deflect to generate sound when actuated. Any or all of the sound producing elements may optionally comprise one or more slits (not shown in the figures), in different suitable configurations and design, used to change the deflection capacity and / or resonance frequency of the sound generating element. The term slit herein refers to a fine line cutting through the thickness, or part of the thickness, of a sound generating element, such as a membrane. The slit is typically very narrow. The pattern of slits affects the stiffness of the overall sound generating element, and hence the resonance frequency and deflectability thereof. The at least one slit may be made using to any suitable method known in the art.

[0126] The substrate 30, 200 may be a PCB.

[0127] The thickness of the polymer layer 40, 300 may be in the range of 20 pm to 2 mm, including the end points of the range. Thereby, the polymer layer is advantageously thick enough to allow for full freedom of movement of the sound producing elements in the sound producing areas 111, but not so thick that it adds unnecessary height to the resulting MEMS micro speaker packages 10. The main purpose of reduction of volume is thereby still fulfilled. In some embodiments, the thickness of the polymer layer 40, 300 is in the range of 100 pm to 300 pm, including the end points of the range. Suitably, this range is well adapted for the oscillation of piezoelectric sound producing elements, specifically piezoelectric membranes, while not adding any unnecessary height to the resulting MEMS micro speaker packages. The thickness of the polymer layer 40, 300 is also co-dependent on the thickness of any solder used, because the thickness of the polymer layer 40, 300 needs to be smaller than the thickness of any solder bumps 55 or copper pillars arranged between matching electrical connection elements 50 before bonding to ensure electrical contact. At the same time, any solder used will reduce in height or thickness during reflow and the thickness of the polymer layer 40, 300 needs to be greater than the thickness of the reflowed solder to ensure that the polymer layer 40, 300 is set into contact with and attaches properly to the opposing surface during bonding.

[0128] The MEMS micro speaker chips in the MEMS micro speaker chips packages herein are in the figures shown as single layer micro speakers. However, a chip may instead comprise a stack of layers, if this is desirable for the design of the MEMS micro speaker, as the wafer level chip scale packaging method will work irrespective of the MEMS micro speaker design.

[0129] The limited number of components, such as active sound producing areas 111 and electrical connection components 50, shown in the figures is only for illustrational purposes. As is evident to those skilled in the art, any number of these types of components and other suitable components that are required for a certain application may be used, and the first and second patterns may suitably be adapted to include them such that the polymer layer 40, 300 is patterned accordingly.

[0130] In any embodiment herein, the polymer of the polymer layer 40, 300 may be a photoresist. Thereby, patterning of the polymer layer 40, 300 may suitably be done in manners known in the art, by masking the polymer layer 40, 300, positively or negatively depending on the type of photoresist used, exposing the masked photoresist to light, and applying a solvent to dissolve the un-hardened areas of the photoresist in the polymer layer 40, 300. In a non-limiting example, the photoresist is SU-8. Variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims.

[0131] The term “comprises / comprising” when used in this specification is taken to specify the presence of stated features, integers, steps, or components. The term does not preclude the presence or addition of one or more additional elements, features, integers, steps or components or groups thereof. The indefinite article "a" or "an" does not exclude a plurality. In the claims, the word “or” is not to be interpreted as an exclusive or (sometimes referred to as “XOR”). On the contrary, expressions such as “A or B” covers all the cases “A and not B”, “B and not A” and “A and B”, unless otherwise indicated. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be construed as limiting the scope.

[0132] It is also to be noted that features from the various embodiments described herein may freely be combined, unless it is explicitly stated that such a combination would be unsuitable.

[0133] The invention is not restricted to the described embodiments in the figures but may be varied freely within the scope of the claims.

Claims

CLAIMS1) A method for manufacturing a plurality of MEMS micro speaker packages (10), the method comprising: providing a MEMS micro speaker layer (100) having a front side (101) and a back side (102), the MEMS micro speaker layer (100) comprising a plurality of MEMS micro speakers (110), wherein each MEMS micro speaker (110) has at least one active sound producing area (111) and, for each active sound producing area (111), at least one electrical connection element (50) electrically connected to an actuator arranged to actuate the active sound producing area (111), wherein the active sound producing areas (111) and the corresponding electrical connection elements (50) of the plurality of MEMS micro speakers (110) are distributed over the front side (101) of the MEMS micro speaker layer (100) according to a first pattern; providing a substrate (200) having a front side (201) and a back side (202) and having electrical connection elements (50) distributed on the front side (201) of the substrate (200) such that each electrical connection element (50) on the substrate (200) matches the location of an electrical connection element (50) in the first pattern; applying a polymer layer (300) to one of the front side (101) of the MEMS micro speaker layer (100) and the front side (201) of the substrate (200); patterning the polymer layer (300) according to the first pattern, to comprise at least one first area (301) that is filled with polymer and second areas (302) that lack polymer, wherein the second areas (302) match the locations of the electrical connection elements (50) and active sound producing areas (111) in the first pattern; bonding the polymer layer (300) to the other of the front side (101) of the MEMS micro speaker layer (100) and the front side (201) of the substrate (200), to form a package (400) with the polymer layer (300) sandwiched between the MEMS micro speaker layer (100) and the substrate (200), such that the polymer layer (300) creates a sealing barrier (310) around a perimeter (311) of each of the active sound producing areas (111) without contacting any part of the active sound producing areas (111) and such thatthe electrical connection elements (50) are not covered by or in contact with the polymer in the polymer layer (300); and dicing the package (400) to obtain the plurality of MEMS micro speaker packages (10).2) The method of claim 1, wherein applying the polymer layer (300) comprises spinning, spin coating or laminating a polymer onto the front side (101) of said one of the MEMS micro speaker layer (100) and the substrate (200).3) The method of claim 1 or 2, wherein the substrate (200) comprises a plurality of electrically conductive through connection (51), each being in electrical contact with an electrical connection element (50) on the front side of the substrate (200) for connecting the electrical connection element (50) on the front side of the substrate (200) with an external component or device.4) The method of claim 3, wherein the electrically conductive through connection (51) is a through silicon via, TSV.5) The method of any one of the preceding claims, wherein the substrate (200) further comprises a plurality of through openings (31) distributed over the substrate (200) such that each through opening (31) matches the location of an active sound producing area (111) of the MEMS micro speaker layer (100) in the first pattern.6) The method of any one of the preceding claims, wherein bonding includes flip- chip mounting of one of the MEMS micro speaker layer (100) and the substrate (200) onto the other.7) The method of any one of the preceding claims, wherein the plurality of MEMS micro speakers (110) are piezoelectric MEMS micro speakers.8) The method of any one of the preceding claims, further comprising applying an adhesive component to the polymer layer (300) or the front side (101, 201) of the other of the MEMS micro speaker layer (100) and the substrate (200) before bonding.9) The method of any one of the preceding claims, wherein the bonding comprises using solder (55) or copper pillars arranged between matching electricalconnection elements (50) on the MEMS micro speaker layer (100) and the substrate (200).10) The method of any one of the preceding claims, further comprising injecting solders (55) between matching electrical connection elements (50) on the MEMS micro speaker layer (100) and the substrate (200) after bonding.1 l)The method of any one of the preceding claims, wherein the thickness of the polymer layer (300) is in the range of 20 pm to 2 mm, including the end points of the range.12)The method of claim 11, wherein the thickness of the polymer layer (300) is in the range of 100 pm to 300 pm, including the end points of the range.13) The method of any one of the preceding claims, wherein the polymer of the polymer layer (300) is a photoresist.14)The method of claim 13, wherein the polymer is SU-8.15) The method of any one of the preceding claims, further comprising modifying at least one of the MEMS micro speaker layer (100), the polymer layer (300) and the optional protective layer (60) at a plurality of locations in connection with the active sound producing area (111) in the MEMS micro speaker layer (100), such that a respective acoustic channel 80 for passing sound from each active sound producing area (111) through an opening in a side wall of the resulting MEMS micro speaker package (10) is formed after dicing.16) The method of any one of the preceding claims, further comprising providing a protective layer (60) on the back side (102) of the MEMS micro speaker layer (100) before bonding.17) The method of claim 16, further comprising modifying the protective layer (60) at a plurality of locations in connection with the active sound producing area (111) in the MEMS micro speaker layer (100), such that a respective acoustic channel 80 for passing sound from each active sound producing area (111) through an opening in a side wall of the resulting MEMS micro speaker package (10) is formed after dicing.18) A MEMS micro speaker package (10), comprising:a MEMS micro speaker chip (20) having a front side (21) and a back side (22), and comprising at least one active sound producing area (111) and, for each active sound producing area (111), at least one electrical connection element (50) electrically connected to an actuator arranged to actuate the active sound producing area (111), wherein the at least one active sound producing area (111) and the corresponding at least one electrical connection element (50) of the MEMS micro speaker chip (20) are distributed over the front side (21) of the MEMS micro speaker chip (20) according to a second pattern; a substrate (30) having a front side (31) facing the front side (21) of the MEMS micro speaker chip (20), and a back side (32), and having at least one electrical connection element (50) distributed on the front side (31) of the substrate (30) such that each electrical connection element (50) on the substrate (30) matches the location of an electrical connection element (50) in the second pattern; and a polymer layer (40) sandwiched between and attached to the MEMS micro speaker chip (20) and the substrate (30), wherein the polymer layer (40) is patterned according to the second pattern to comprise at least one first area (41) that is filled with polymer and second areas (42) that lack polymer, wherein the second areas (42) match the location of the at least one electrical connection element (50) and the at least one active sound producing area (111) in the second pattern, such that the polymer layer (40) creates a sealing barrier (310) around the perimeter (311) of each of the at least one active sound producing area (111) without contacting any part of the at least one active sound producing area (111) and such that the electrical connection elements (50) are not covered by or in contact with the polymer in the polymer layer (40), thereby forming a MEMS micro speaker package (10) having an acoustic seal between each active sound producing area (111) of the MEMS micro speaker chip (20) and the substrate (30).19) The MEMS micro speaker package (10) of claim 18, wherein the substrate (30) is a printed circuit board, PCB.20) The MEMS micro speaker package (10) of claim 18 or 19, wherein the substrate (200) comprises at least one electrically conductive through connection (51).21) The MEMS micro speaker package (10) of any one of the claims 18 to 20, wherein the electrically conductive through connection (51) is a through silicon via, TSV.22)The MEMS micro speaker package (10) of any one of the claims 18 to 21, wherein the thickness of the polymer layer (40) is in the range of 20 pm to 2 mm, including the end points of the range.23) The MEMS micro speaker package (10) of claim 22, wherein the thickness of the polymer layer (40) is in the range of 100 pm to 300 pm, including the end points of the range.24) The MEMS micro speaker package (10) of any one of the claims 18 to 23 wherein the polymer of the polymer layer (40) is a photoresist.25) The MEMS micro speaker package of claim 24, wherein the polymer is SU-8.26)The MEMS micro speaker package (10) of any one of the claims 18 to 25, wherein the at least active one sound producing area (111) is driven using a piezoelectric material as the actuator.27) The MEMS micro speaker package (10) of any one of the claims 18 to 26, wherein the side walls of the MEMS micro speaker package (10) are straight.28) The MEMS micro speaker package (10) of any one of the claims 18 to 27, comprising an acoustic channel (80) connected to an active sound producing area (111) for passing sound from the active sound producing area (111) into the surrounding air via an opening in a side wall of the MEMS micro speaker package (10).

Images