AN ULTRASOUND TRANSDUCER AND A TILED ARRAY OF ULTRASOUND TRANSDUCERS

Abstract

An ultrasound transducer has a control circuit with an array of circuit portions which together define a circuit area. cMUT elements are provided over the top of the control circuit. Vias provide a coupling between each cMUT element and a respective one of the circuit portions beneath. The vias are located at spaces between the cMUT elements and connect to the circuit portion beneath at a position relative to the shape of the circuit portion. The pitch of the array is greater than the pitch of the circuit portions in at least one direction of the array such that in said at least one direction, the cMUT elements extend beyond one or both sides of the circuit rea. This makes tiling easier and enables a modular design.

Claims

1. An ultrasound transducer, comprising: a control circuit which comprises an integrated circuit, the integrated circuit comprising an array of circuit portions; an array of capacitive micromachined ultrasound transducer, cMUT, elements provided to overlay the control circuit; and an array of vias, with at least one via providing a coupling between each cMUT element and a respective one of the circuit portions said CMUT element is arranged to overlay, those circuit portions together defining a circuit area, wherein the pitch of the array of cMUT elements is greater than the pitch of the circuit portions in at least one direction of the array such that in said at least one direction, the cMUT elements extend beyond one or both sides of the circuit area in said at least one direction.

2. The ultrasound transducer according to claim 1, wherein the cMUT elements each occupy an area and there are spaces defined between the occupied areas, and wherein each via is located at a respective one of the spaces.

3. The ultrasound transducer according to claim 1, wherein each via connects to the circuit portion beneath at a relative position within the shape of the circuit portion, wherein there are at least two different relative positions such that a first cMUT element connects to its respective circuit portion at a first relative position, and a second cMUT element connects to its respective circuit portion at a second relative position.

4. The ultrasound transducer according to claim 3, wherein along each row of cMUT elements, each relative position is different in the row direction, and the cMUT elements at one end of each row overlap one side of the circuit area in the row direction.

5. The ultrasound transducer according to claim 4, wherein the cMUT elements at a same end of each row overlap a same side of the circuit area.

6. The ultrasound transducer according to claim 4, wherein along each column of cMUT elements, each relative position is different in the column direction, and the cMUT elements at one end of each column overlap one side of the circuit area in the column direction.

7. The ultrasound transducer according to claim 6, wherein the cMUT elements at a same end of each column overlap a same side of the circuit area.

8. The ultrasound transducer according to claim 4, wherein along each row of cMUT elements, the via of a first set of cMUT elements is located at a space in a first position relative to the cMUT element, and the via of a second set of cMUT elements is located at a second, different, position relative to the cMUT element, and wherein the cMUT elements at each end of each row overlap the respective side of the circuit area in the row direction.

9. The ultrasound transducer according to claim 4, wherein the via of each cMUT element is located at a space at a same position relative to the cMUT element.

10. The ultrasound transducer according to claim 4, wherein the cMUT elements are staggered between adjacent rows by half a cMUT element pitch.

11. The ultrasound transducer according to claim 1, wherein the control circuit has more circuit portions than the number of cMUT elements.

12. The ultrasound transducer according to claim 11, wherein the cMUT elements extend beyond one side of the circuit area and the spare circuit portions are at an opposite side.

13. The ultrasound transducer according to claim 12, wherein the cMUT elements extend beyond two orthogonal sides of the circuit area and the spare circuit portions are at both opposite sides.

14. An ultrasound transducer arrangement, comprising: a plurality of ultrasound transducers, at least some of which are according to claim 1, forming a tiled arrangement, wherein the control circuits are spaced apart and the cMUT elements of the ultrasound transducers together form a regular array.

15. An ultrasound transducer arrangement according to claim 14, wherein the cMUT elements at each end of each row overlap a same respective overlap row side of the circuit area in the row direction.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0054] For a better understanding of the invention, and to show more clearly how it may be carried into effect, reference will now be made, by way of example only, to the accompanying drawings, in which:

[0055] FIG. 1 shows a known design of cMUT element (a so-called cell) for use in an ultrasound system and a known drive arrangement;

[0056] FIG. 2 shows the cMUT element with a first degree of membrane collapse;

[0057] FIG. 3 shows the cMUT element with a second first degree of membrane collapse;

[0058] FIG. 4 shows a first possible arrangement to connect to an underlying ASIC;

[0059] FIG. 5 shows a second possible arrangement to connect to an underlying ASIC;

[0060] FIG. 6 shows a control circuit for a 4×6 transducer element matrix array;

[0061] FIG. 7 shows the same circuit of FIG. 6 after processing of the cMUT elements;

[0062] FIG. 7 shows the conventional layout with identical cMUT pitch and pitch of the control circuit portions;

[0063] FIG. 8 shows the conventional ASIC design having a regular array of circuit portions as wells as the cMUT positioning over the top;

[0064] FIG. 9 shows a first example of circuit portions having a smaller area and hence smaller pitch than the area of the cMUT elements;

[0065] FIG. 10 shows a second example;

[0066] FIG. 11 shows a variant of the design of FIG. 10;

[0067] FIG. 12 shows an example in which the control circuit has more circuit portions than the number of cMUT elements, such that there are spare circuit portions;

[0068] FIG. 13 shows that the design of FIG. 12 enables transducer arrays with different sizes on one wafer;

[0069] FIG. 14 shows another implementation with spare circuit portions;

[0070] FIG. 15 shows another example with spare circuit portions;

[0071] FIG. 16 shows rows of cMUT element staggered by half a cMUT element pitch, to form a hexagonal array of cMUT elements;

[0072] FIG. 17 shows a control circuit with contact pads for centrally located vias and the cMUT array over the top;

[0073] FIG. 18 shows three possible configurations with cMUT elements shifted relative to the circuit portions;

[0074] FIG. 19 shows an example in which the transducer elements are chosen to be larger than the underlying control circuit portions;

[0075] FIG. 20 shows four separate modules butted together as a 2×2 array or a continuous transducer array over multiple adjacent control circuits;

[0076] FIG. 21 shows three examples in which the cMUT elements of one row are staggered by half a pitch compared to the adjacent rows; and

[0077] FIG. 22 shows an example of a larger hexagonal transducer array of the type shown in the bottom of FIG. 21.

DETAILED DESCRIPTION OF THE EMBODIMENTS

[0078] The invention will be described with reference to the Figures.

[0079] It should be understood that the detailed description and specific examples, while indicating exemplary embodiments of the apparatus, systems and methods, are intended for purposes of illustration only and are not intended to limit the scope of the invention. These and other features, aspects, and advantages of the apparatus, systems and methods of the present invention will become better understood from the following description, appended claims, and accompanying drawings. It should be understood that the Figures are merely schematic and are not drawn to scale. It should also be understood that the same reference numerals are used throughout the Figures to indicate the same or similar parts.

[0080] The invention provides an ultrasound transducer having a control circuit with a regular array of circuit portions which together define a circuit area. cMUT elements are provided over the top of the control circuit. Vias provide a coupling between each cMUT element and a respective one of the circuit portions beneath (the circuit portion, which is being overlaid by the CMUT element). The vias are located at spaces between the cMUT elements and are connected to the circuit portion beneath at a position relative to the shape of the circuit portion. It shall be understood in the context of the present application that term beneath is defined as a projection of a surface area of one element onto another unit, which said element is arranged to overlay. Thus, beneath bottom electrode shall be interpreted as within the projection of the electrode's surface area onto the surface area of the circuit portion the CMUT element is arranged to overlay. The pitch of the array is greater than the pitch of the circuit portions in at least one direction of the array such that in said at least one direction, the cMUT elements extend beyond one or both sides of the circuit area. This makes tiling easier and enables a modular design.

[0081] The invention makes use of cMUT transducer elements. cMUT transducer elements can be made by semiconductor processes and in particular the same processes used to produce the application specific integrated circuits (ASICs) needed by an ultrasound probe such as a CMOS process. The manufacture of cMUT transducer elements thus does influence the ASIC parameters.

[0082] cMUT elements are tiny diaphragm-like devices with electrodes that convert the sound vibration of a received ultrasound signal into a modulated capacitance. Many of such individual cMUT elements can be connected together and operated in unison as a single transducer element. For example, four to sixteen cMUT elements can be coupled together to function in unison as a single transducer element.

[0083] The manufacture of cMUT-based ultrasound systems is therefore more cost-effective compared to PZT-based systems. Moreover, due to the materials used in such semiconductor processes, the cMUT elements exhibit much improved acoustic impedance matching to water and biological tissue, which obviates the need for (multiple) matching layers and yields an improved effective bandwidth.

[0084] cMUT elements in particular are able to function over a broad bandwidth, enable high resolution and high sensitivity imaging, and produce a large pressure output so that a large depth of field of acoustic signals can be received at ultrasonic frequencies.

[0085] FIG. 1 shows a known design of cMUT element 100 (a so-called cell) for use in an ultrasound system and a known drive arrangement 101. The cMUT element 100 comprises a flexible membrane or diaphragm 114 suspended above a silicon substrate 112 with a gap or cavity 118 there between. A first electrode 122 is located on the floor of the cell on the upper surface of the substrate 112 in this example. A second electrode 120 is located on the diaphragm 114 and moves with the diaphragm. In the example shown, the two electrodes are circular.

[0086] A dielectric (not shown) is provided on the substrate 112 and underneath the top (second) electrode 120. The diaphragm may instead function as a dielectric layer.

[0087] Preferably, there are two dielectrics which may be equal in composition and thickness, but may be also asymmetric (different materials and thicknesses).

[0088] The membrane layer 114 is fixed relative to the top face of the substrate layer 112 and configured and dimensioned so as to define a spherical or cylindrical cavity 118 between the membrane layer 114 and the substrate layer 112.

[0089] Other realizations of the electrode 120 design can be considered, such as electrode 120 may be embedded in the membrane 114 or it may be deposited on the membrane 114 as an additional layer.

[0090] The first electrode may be directly exposed to the gap 118 or separated from the gap 118 by an electrically insulating layer or film to prevent a short-circuit between the second electrode 120 and the first electrode 122.

[0091] In FIG. 1 the first electrode 122 is grounded by way of example. Other arrangements, e.g. a grounded second electrode 120 or both second electrode 120 and first electrode 122 floating are of course equally feasible.

[0092] The electrodes of the cMUT element 100 provide the capacitive plates of the device and the gap 118 is the main dielectric between the plates of the capacitor. When the diaphragm vibrates, the changing dimension of the dielectric gap between the plates provides a changing capacitance which is sensed as the response of the cMUT element 100 to a received acoustic echo.

[0093] The spacing between the electrodes is controlled by applying a static voltage, e.g. a DC bias voltage, to the electrodes with a voltage supply 101. The voltage supply 101 may optionally comprise separate stages 102, 104 for providing the DC and AC or stimulus components respectively of the drive voltage of the cMUT elements 100, e.g. in transmission mode. The first stage 102 may be adapted to generate the static (DC) voltage component and the second stage 104 may be adapted to generate an alternating variable drive or stimulus voltage component having a set alternating frequency, which signal typically is the difference between the overall drive voltage and the aforementioned static component thereof.

[0094] The static or bias component of the applied drive voltage preferably meets or exceeds the threshold voltage for forcing the cMUT element 100 into its collapsed state. This applies for a collapsed mode of operation. The cMUT element may however equally be operated in a non-collapsed mode.

[0095] The first stage 102 may include relatively large capacitors, e.g. smoothing capacitors, in order to generate a particularly low-noise static component of the overall voltage, which static component typically dominates the overall voltage such that the noise characteristics of the overall voltage signal will be dominated by the noise characteristics of this static component.

[0096] It is known that by applying a static voltage above a certain threshold, the cMUT element 100 is forced into a collapsed state in which the membrane 114 collapses onto the substrate 112. This threshold value may depend on the exact design of the cMUT element 100 and is defined as the DC bias voltage, known as the collapse voltage, at which the membrane 114 sticks to (contacts) the cell floor through the force due to the electric field between the electrodes. The amount (area) of contact between the membrane 114 and the substrate 112 is dependent on the applied bias voltage.

[0097] Increasing the contact area between the membrane 114 and the substrate 112 increases the resonant frequency of the membrane 114, as will be explained in more detail with the aid of FIG. 2 and FIG. 3.

[0098] The frequency response of a collapsed mode cMUT element 100 may be varied by adjusting the DC bias voltage applied to the cMUT electrodes after collapse. As a result, the resonant frequency of the cMUT element increases as a higher DC bias voltage is applied to the electrodes.

[0099] The principles behind this phenomenon are illustrated in FIGS. 2 and 3. The cross-sectional views of FIGS. 2 and 3 illustrate this one-dimensionally by the distances D1 and D2 between the outer support of the membrane 114 and the point where the membrane begins to touch the floor of the cavity 118 in each illustration. It can be seen that the distance D1 is a relatively long distance in FIG. 2 when a relatively low bias voltage is applied, whereas the distance D2 in FIG. 3 is a much shorter distance due to a higher bias voltage being applied. These distances can be compared to long and short strings which are held by the ends and then plucked. The long, relaxed string will vibrate at a much lower frequency when plucked than will the shorter, tighter string. Analogously, the resonant frequency of the cMUT element in FIG. 2 will be lower than the resonant frequency of the cMUT element in FIG. 3 which is subject to the higher bias voltage.

[0100] Thus, a typical cMUT design comprises a flexible membrane or diaphragm suspended above a silicon substrate with a gap or cavity between them. The gap is caused by removing a sacrificial layer during manufacture. A first electrode is located on the floor of the cell on the upper surface of the substrate and a second electrode is located on the diaphragm and moves with the diaphragm.

[0101] The operation of cMUT elements will be well known to those skilled in the art.

[0102] To connect to an underlying ASIC, vias are used. A first possible arrangement is shown schematically in FIG. 4.

[0103] The cMUT transducer elements 200 are provided over the control circuit 202 (an ASIC integrated circuit), the integrated circuit comprising a regular array of circuit portions 204 (these are the unit cells of the circuit). There is a circuit portion associated with each cMUT element. The cMUT elements are provided over the top of the control circuit 202.

[0104] The cMUT elements each occupy an area depending on the shape of the cMUT elements. Typically, they are circular, so that there are spaces defined between the occupied circular areas.

[0105] Vias 205 are provided in these spaces, for connecting between signal electrode pads 206 of the cMUT cell and contact pads 208 of the circuit portions. Only one via is needed, for example the via from the control circuit portion connects to the bottom electrode of the cMUT, and the cMUT top electrodes are all connected in parallel to a common ground or bias potential. The vias are located laterally to minimize the surface topography below the cMUT membranes, i.e. to keep the surface below the cMUT membranes as flat as possible.

[0106] A second possible arrangement is shown schematically in FIG. 5. The cMUT transducer elements 200 are again provided over the control circuit 202 with its regular array of circuit portions 204.

[0107] Vias 205 are provided beneath the cMUT elements in this design, again for connecting between a signal electrode pad 206 of the cMUT cell and a contact pad 208 of the circuit portions.

[0108] FIG. 6 shows a set of four control circuits 202 each for a 4×6 transducer element matrix array. The small array dimensions are chosen for simplicity only. The ASIC comprises a regular array of circuit portions 204/From an ASIC design perspective, it is difficult to create 2D arrays of unit-cells (i.e. circuit portions) that do not consist of perpendicular rows and columns, as this severely complicates interconnect routing. Therefore, a 2D array of ASIC unit-control cells having perpendicular rows and columns is assumed. The height and width of each ASIC unit cell are not necessarily equal.

[0109] A seal ring 300 surrounds the critical circuit components and a saw street 302 for die separation is around the outside of the seal ring 300. FIG. 6 also shows the ASIC process control modules (PCMs).

[0110] FIG. 7 shows the same circuit of FIG. 6 after processing of the cMUT elements 304 (in a MEMS cleanroom). Each cMUT element 304 is indicated by a circle, which is the prevalent shape of cMUT transducer elements, although the invention is not limited to circular transducer elements only. In all examples, the bottom electrode is assumed to have equal outer dimensions as the transducer element for ease of discussion.

[0111] FIG. 7 shows the conventional layout with identical cMUT pitch and pitch of the control circuit portions.

[0112] A first set of examples is based on the use of vias which are placed next to the transducer elements as explained with reference to FIG. 4. This is standard practice for current cMUT transducers.

[0113] The top part of FIG. 8 shows the conventional ASIC design having a regular array of circuit portions, each receiving electrical connections from a via 205 (so there is an associated contact pad at the ASIC). The vias are all at the same position relative to the control circuit portions, i.e. in the bottom right corner in this example.

[0114] The bottom part of FIG. 8 shows the cMUT elements 304 over the top, and thus corresponds to FIG. 7 but also shows more clearly the via positioning in the spaces not occupied by the cMUT elements.

[0115] A first aspect is based on the use of a control circuit (ASIC) with circuit portions having a smaller area and hence smaller pitch than the area of the cMUT elements. The size difference is such that the seal ring 300 fits together with the active parts of the control circuit (i.e. the array of circuit portions) under the transducer element array.

[0116] FIG. 9 shows a first example.

[0117] In this example, the pitch of the cMUT element array is greater than the pitch of the circuit portions only in a row direction. Thus, in this row direction the cMUT elements extend beyond the circuit area of the control circuit defined by the combination of the unit cells of the control circuit. In this example the cMUT elements extend beyond the circuit area at both sides (i.e. at both ends of the rows).

[0118] The top left image shows the control circuit as a regular array of identically sized control circuit portions 204 and the seal ring 300 and saw street 302.

[0119] The top right image shows the array of cMUT elements 304. It also shows the cMUT PCM.

[0120] The bottom image shows that the overlap enables the individual full control circuits 202 (i.e. one full array of the control circuit portions) and associated cMUT elements to be tiled while maintaining a uniform array of cMUT elements. This example shows two identical transducers arranged side by side to form the transducer arrangement. The two sides which are butted together each have the overlap of the cMUT elements.

[0121] A row of any number of transducers may be formed, since the overlap is at both ends of the rows.

[0122] The result of the difference in pitch is that different cMUT elements connect to their respective circuit portions at different relative positions. The top left image shows a connection region 310 of each circuit portion. Different vias 205 connect to different positions within that general connection region.

[0123] The circuit portions 204 are identical in size to cMUT elements in one dimension (the column direction, for example corresponding to the elevation scanning dimension) and are slightly smaller in the other dimension (the row direction, for example corresponding to the azimuthal scanning dimension). For example, in case of a 32×32 cMUT element array, the area of the transducer element could be 300×300 μm, while the area of the circuit portion 204 could be 300×285 μm.

[0124] This causes the overall transducer to exceed the ASIC control area by +480 μm in the azimuthal direction while there is no excess in elevation direction. If the center of the transducer module is aligned with the center of the control circuit, the transducer module will overlap the control circuit by +240 μm in both opposite row directions. This is sufficient to close the gap between the active areas of two control circuit dies (i.e., the margin to the seal ring, the seal ring width and the saw street width).

[0125] In this example, the transducer element array is split in two half arrays; the left half and the right half. The vias for the left half are located on the lower right corner of the cMUT element area and the vias for the right half are located on the lower left side of the cMUT element area.

[0126] Thus, the via of a first set of cMUT elements is located at a space in a first position relative to the cMUT element (lower left), and the via of a second set of cMUT elements is located at a second, different, position relative to the cMUT element (lower right). As a result, there is a double column of vias in the middle of the module. The alignment of the cMUT element array and the control circuit is such that the double vias connect to different circuit portions.

[0127] The via processing can be part of the IC or the MEMS processing stages. As the via pitch primarily is linked to the transducer element pitch, it is preferred to realize the via during the MEMS processing. The control circuit portions allow “landing” of the via on every X-coordinate position in the circuit portion, i.e. anywhere within the connection region 310. The connection regions for example comprises a horizontal metal track extending from one side of the cell to the other side to allow such connection.

[0128] In this way, the circuit portions can have identical designs even though connections to the circuit portions can be made at different positions.

[0129] FIG. 10 shows a second example. In this example, the vias are all located at a same position relative to the cMUT element (lower right in this example). In this example, no special arrangement of vias is needed at the cMUT element level.

[0130] In FIG. 10, the cMUT elements overlap only at one of a pair of opposite sides. However, FIG. 10 shows an example in which the pitch of the circuit portions is also less than the pitch of the cMUT elements in the column direction. The position of the via relative to the control circuit portions is thus also different in the column direction, so that the cMUT elements at one end of each column overlap the corresponding (top or bottom) one side of the circuit area in the column direction.

[0131] In this case, the pitch difference applies in both the row and column directions. The connection region 310 becomes two dimensional because the position at which connection is made varies in both the row direction and the column direction. In FIG. 10, the cMUT elements at a same end of each column overlap a same side of the circuit area. Thus, all along a top or bottom side, there is an overlap of the cMUT elements.

[0132] As an example, in the case of a 32×32 element array, the area of the transducer element could be 300×300 μm, while the area of the circuit portion 204 could be 295×295 μm.

[0133] If the right lower corner of the transducer element is aligned with the right lower corner of the control circuit as shown, the transducer module will overlap the control circuit by 160 μm on the left and the upper sides.

[0134] The control circuit portions must allow “landing” of the via in the connection region 310. For this, the circuit portions contain a relatively large metal plate area to allow for such connection. The shape of the metal area can be square but also oval/rectangular/triangular or indeed any other shape.

[0135] FIG. 11 shows a variant of the design of FIG. 10 that uses the via arrangement explained with reference to FIG. 9 in combination with the two-axis pitch adjustment of FIG. 10.

[0136] The advantage of this combination of features version is that smaller “landing metal” is required in the connection regions 310 or larger transducer overlaps can be realized.

[0137] To harvest the transducer modules from the wafer, a form of wafer level die separation such as dicing or etching is required. Due to the modular approach, it is possible to harvest transducer arrays with various sizes from one wafer.

[0138] Harvesting the transducer arrays can be done using mechanical or chemical dicing (etching). When etching, narrow dicing/etch streets can be used (e.g. <10 μm). It is for example possible to design the transducer module such that the street approximately fits the normal kerf between the transducer elements.

[0139] In the case of mechanical dicing (sawing), large streets are needed. In this case, it is possible to sacrifice transducer modules in order to accommodate the required dicing street; this route may be a viable option if the transducer array consists of many transducer modules (if small modules are used).

[0140] Often, transducer arrays use dummy elements on the rim of the arrays. The dummy components may not however be needed. If dicing impacts the characteristics of the rim transducer elements, the control circuit should be aware of this and not activate the outer elements in the application.

[0141] The examples above all have the same number of circuit portions as cMUT elements.

[0142] FIG. 12 shows an example in which the control circuit has more circuit portions than the number of cMUT elements, such that there are spare circuit portions. The control circuit again has circuit portions 204 having an area slightly smaller than the area of the transducer elements (in one direction only for the example of FIG. 12 but the size difference can be in both directions).

[0143] The top left image shows an 8×4 array of circuit portions whereas the cMUT array at the top right has 6×4 elements. The cMUT array has the design of FIG. 9 with two possible via positions relative to the cMUT element.

[0144] The area of overlap can correspond to the width of a circuit portion. Thus, in the transducer arrangement of three transducers shown at the bottom, the left transducer has an overlap to the right and there is a full column of spare control circuit portions, the right transducer has an overlap to the left with a column of spare circuit portions to the right and there is a different design of transducer in the middle with no overlap. The overlaps from each side are sufficient to cover both adjacent seal rings.

[0145] The arrangement shown at the bottom of FIG. 12 is single die and it is not to be diced because it is fully occupied with transducer elements.

[0146] For example, a 32×32 element array may be connected to an control circuit 202 with (31+N)×(31+M) circuit portions 204 (in the elevation column and azimuthal row directions respectively). M and N are integer numbers; their value determines the maximum number of transducer modules that can be combined in a transducer array that does not extend the seal rings on the rim.

[0147] For example, with N=2 and M=2 (1 extra row and 1 extra column of circuit portions), transducer arrays composed of 1 or 2 modules in elevation and 1 or 2 modules in azimuthal direction can be realized without extending the seal rings on the outer side of the array.

[0148] The example of FIG. 12 shows a 3-module array with M=3 and N=1 (2 extra columns of circuit portions per control circuit).

[0149] By using a special stepper job, the alignment of the transducer array and the control circuit portions can be changed. In the example of FIG. 12, the center of the middle transducer is aligned with the center of the corresponding ASIC control circuit, the center of the left transducer is shifted to the right and the center of the right transducer module is shifted to the left relative to the center of the corresponding ASIC control circuit.

[0150] Thus, the control circuits may still all have the same design and the cMUT element arrays may still have the same design.

[0151] There are spare circuit portions in this arrangement, so that not all circuit portions are needed by the transducer elements. The system controller will be aware of this and not stimulate these control units.

[0152] FIG. 13 shows that this modular design enables transducer arrays with different sizes on one wafer.

[0153] In FIG. 13, each control circuit has 8 columns of circuit portions and the cMUT array has 6 columns (M=3, 2 extra columns of circuit portions).

[0154] FIG. 13 shows in the top left a 2 transducer arrangement and in the top right a single transducer arrangement, formed on a single wafer. The wafer may then be diced to separate the two transducer arrangements. In particular, a dedicated dicing street is generated between the two transducer arrangements, as shown. This dicing street simplifies the dicing of the transducer arrays, which can be done mechanically and chemically without real yield loss.

[0155] FIG. 13 shows in the bottom a 3 transducer arrangement (corresponding to FIG. 12) formed on the same wafer design.

[0156] FIG. 14 shows an example implementation with M=2, N=1 (1 extra column of circuit portions). In case M=even, there is no “middle” transducer module and none of the transducer modules is exactly aligned with the ASIC modules.

[0157] In this example, the center of the left transducer is shifted to the right and the center of the right transducer module is shifted to the left relative to the center of the associated ASIC control circuit.

[0158] The two transducers are formed on a single wafer.

[0159] FIG. 15 shows an example with M=2 and N=2 (1 extra column of circuit portions and 1 extra row of circuit portions). In this example, the cMUT element array of the type shown in FIG. 10 is used, with a uniform via arrangement relative to the cMUT elements.

[0160] A 2×2 array of transducers is thereby formed.

[0161] It can be seen that one use of the overlap and pitch arrangement is to enable multiple diced transducers to be tiled, e.g. FIGS. 9 to 11. Another use is to allow a modular design of control circuit on a single wafer, e.g. FIGS. 12 to 15. The designs of FIGS. 12 to 15 are not tiled separate dies, they are formed on a single wafer. However, the pitch mismatch enables different pitches of cMUT element arrays to be formed over the control circuit.

[0162] The examples above make use of the mismatch between the size/pitch of the control circuit portions and cMUT elements to enable overlap along one or more sides of a transducer, so that transducers can be tiled or so that a modular design may be employed.

[0163] The approach may also be used to enable hexagonal transducer arrays to be formed on top of rectangular arrays of control circuit portions. The well-known advantage of hexagonal transducer arrays over square packed arrays is their higher density of active transducer area. This allows to generate higher pressures, higher receive sensitivity [A/pa] and larger bandwidth.

[0164] FIG. 16 shows rows of cMUT element staggered by half a cMUT element pitch, to form a hexagonal array of cMUT elements. The via 205 is always at the bottom right. The pitch of the cMUT elements is larger in only the row direction than the circuit portions. This requires a long horizontal connection region 310.

[0165] The examples above create flexibility in placing the transducer elements at different positions relative to the underlying control circuit portions. Some designs require special via arrangements, but there is a downside that this complicates the ASIC design, and requires changes to the actual via locations, which differ from one transducer design to another.

[0166] Another set of examples uses a further or alternative method to create flexibility in placing the transducer elements at different positions relative to the underlying ASIC unit circuit portions. This set of examples is based on the approach shown in FIG. 5, according to which each via to control circuit portion is placed directly below the bottom electrode of the transducer element. This is feasible if the substrate including the via is sufficiently flat for building the transducer element directly on top of it.

[0167] In practice, a surface topography well below 50 nm is required at the location of the transducer elements. The relatively large dimension of the transducer element's bottom electrode compared to the via dimension allows shifting of the transducer element relative to the via. Thus, the position of the via can then be kept fixed relative to that of the corresponding ASIC unit control-cell. Therefore, it creates flexibility in placing the transducer elements at different positions relative to the underlying control circuit portions. This approach may be combined with the approaches described above in order to achieve even greater position flexibility of the transducer elements relative to the underlying control circuit portions.

[0168] This set of examples make use of flexibility in placing the cMUT elements at different positions relative to the underlying circuit portions of the control circuit. The vias are directly below the bottom electrode of the transducer element, e.g. located centrally. By shifting the transducer element relative to the via, so that the via is not central to the cMUT element, the cMUT element is consequently shifted relative to its control circuit portion. By intentionally changing the transducer element locations of adjacent ASIC unit control cells, the same ASIC design can be used to build transducers having different center frequencies and/or different apertures.

[0169] FIG. 17 shows at the top a control circuit formed of control circuit portions 204, with contact pads for centrally located vias 205. This is for example purposes only. The bottom image shows the cMUT elements 304 mounted in conventional manner, with the same array pitch as the control circuit portions, and symmetrically positioned so that the area of each cMUT element overlaps and corresponds with the area of the circuit portion beneath.

[0170] FIG. 18 shows three possible configurations in accordance with this set of examples. The cMUT elements 304 have the same pitch as the circuit portions 204 but they are offset from alignment so that the cMUT elements overlap one side of the circuit area in the row direction (middle figure) or overlap one side in the row direction and one side in the column direction (top and bottom figures). The connection of the via to the cMUT element is at a different position relative to the cMUT element in these three examples. As the bottom electrodes are assumed to have equal outer dimensions to the transducer elements, it can be seen that in all cases the bottom electrodes overlap with the vias to the corresponding control circuit portions, and therefore they are considered to make electrical contact.

[0171] FIG. 19 shows an example in which the transducer elements are chosen to be larger than the underlying control circuit portions, in the same way as explained above. This example shows the pitch of the cMUT elements being greater in both orthogonal array directions. This results in a two-way pitch mismatch which can be solved by translating the transducer elements relative to their corresponding control circuit portions, and by skipping over individual control circuit portions in a regular pattern.

[0172] The skipped circuit portions are shown in FIG. 19 in grey. These circuit portions are not used by the cMUT elements. They form a central row and a central column.

[0173] As a result, the circuit area (i.e. the combination of circuit portions associated with the cMUT elements) becomes the four square corners. The cMUT elements overlap these four areas over the skipped circuit portions.

[0174] In this way, the design of the cMUT array may be adapted while maintaining the same circuit design. This approach can be extended over larger areas than shown in FIG. 19 by simple repetition.

[0175] In this particular case, the vias A, B, C and D are just outside the area of the transducer element so that not all vias are directly beneath the cMUT lower electrode. This can easily be overcome by including a trace between the via and the bottom membrane of the corresponding transducer element.

[0176] This approach also enables tiling to form a continuous transducer array created from multiple control circuits and cMUT arrays.

[0177] FIG. 20 shows four separately diced modules butted together as a 2×2 array. In this case, the cMUT elements overlap all four sides of the underlying circuit area, and hence overlap the seal ring. A uniform cMUT element array is formed even though there are gaps between the control circuits 202, in the same way as explained above.

[0178] FIG. 20 may also be considered to show a transducer formed over multiple ASIC modules on a single wafer.

[0179] As for the examples above, this approach also allows hexagonal transducer arrays on top of rectangular arrays of control circuit portions.

[0180] FIG. 21 shows three examples in which the cMUT elements of one row are staggered by half a pitch compared to the adjacent rows.

[0181] The top image shows an example in which the cMUT elements have the same pitch in both orthogonal directions as the circuit portions. Odd rows are shifted slightly to the right and even rows are shifted slightly to the left.

[0182] The middle image shows a difference in pitch only in the row direction, so the cMUT element pitch is greater in the row direction than the circuit portions.

[0183] The bottom image shows a difference in pitch in the row and column directions, so the cMUT element pitch is greater in the row and column directions than the circuit portions. Transducer element arrangements other than hexagonal or square transducer are also possible

[0184] The staggered cMUT configurations may need additional ASIC programming in order to generate the correct (delay) control of transducer elements in the “odd” columns relative to the “even” columns.

[0185] FIG. 22 shows an example of a larger hexagonal transducer array of the type shown in the bottom of FIG. 21 (with larger pitch in both directions) on a rectangular ASIC control circuit. The skipped control circuit portions are again depicted in grey. They are positioned along (but not fully occupying) a central row and a central column.

[0186] As a result, the circuit area (i.e. the combination of circuit portions associated with the cMUT elements) becomes a top half and a bottom half. The cMUT elements overlap these two areas at the top edges and the bottom edges (as well as overlapping the edges into the skipped circuit portions).

[0187] The skipping of ASIC control circuit portions allows even greater flexibility in the range of frequencies and/or apertures that can be achieved. In this way, the same ASIC design can be reused for multiple transducer variants, which is very beneficial, as the ASIC design is the most expensive part of the entire transducer.

[0188] The invention is of general interest for the manufacture of 2D cMUT arrays that are monolithically connected to control circuits. The invention is particularly interesting to realize relatively large transducer arrays and modular arrays.

[0189] Various examples above enable modular transducer arrays to be formed, by compiling transducer arrays with flexible length by combining a number of smaller transducer modules. It also is possible to realize transducer arrays with different lengths on one wafer. For example, transducer arrays with a length of 1, 2, 3, 4, 5 or 6 cm can be realized by connecting modules with a length of 1 cm. It is attractive to realize small transducer modules to reduce IC foundry costs. Multiple design masks can be placed on a single reticle (a so-called multi layer reticle, MLR process). A modular transducer approach can also significantly reduce ASIC design time.

[0190] 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. In the claims, the word “comprising” does not exclude other elements or steps, and the indefinite article “a” or “an” does not exclude a plurality.

[0191] 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.

[0192] If the term “adapted to” is used in the claims or description, it is noted the term “adapted to” is intended to be equivalent to the term “configured to”.

[0193] Any reference signs in the claims should not be construed as limiting the scope.