Patent Grant
8753279
The present invention relates to an ultrasonic probe and ultrasonic imaging device, and, by way of example, to an ultrasonic probe and ultrasonic imaging device that use capacitive micro-machines.
Ultrasonic transducers are devices that radiate and receive sound waves in or above the audible range (approximately 20 Hz to 20 kHz), and are widely used for medical purposes, in non-destructive testing, etc. Piezoelectric devices, a typical example being PZT (Lead Zirconate Titanate), are presently most widely used as ultrasonic transducers. However, in recent years, the development of ultrasonic devices called Capacitive Micro-machined Ultrasonic Transducers (hereinafter, CMUTs), which utilize an operation principle that differs from piezoelectric types, has advanced, and is beginning to be put to practical use. CMUTs are fabricated by applying semiconductor techniques. They are ordinarily formed by burying an electrode material in a substrate (or the substrate itself may sometimes serve as an electrode) made of a material that is used in semiconductor processes, e.g., silicon, etc., and by securing a fine (e.g., 50 μm) and thin (e.g., several μm) diaphragm with supporting walls around the diaphragm, etc. A cavity is provided between the diaphragm and the substrate to allow the diaphragm to vibrate. An electrode material is buried within this diaphragm as well. By thus having independent electrodes disposed in the substrate and the diaphragm, the substrate and the diaphragm function as a capacitance (capacitor). By applying a voltage across both electrodes (a bias voltage is ordinarily applied in advance), they function as an ultrasonic transducer. When an AC voltage is applied across both electrodes, the electrostatic force between the electrodes varies, causing the diaphragm to vibrate. If, at this point, there is some medium that is in contact with the diaphragm, the vibration of the diaphragm will propagate within the medium as a sound wave. In other words, it is possible to radiate sound. Conversely, if a sound wave is transmitted to the diaphragm, the diaphragm will vibrate in accordance therewith, and as the distance between both electrodes varies, an electric current will flow between both electrodes, or the voltage across both electrodes will vary. By extracting an electric signal of this electric current, voltage, etc., it is possible to receive sound waves.
Important indicators that determine the performance of an ultrasonic transducer include the acoustic pressure transmitted and receive sensitivity. To increase acoustic pressure and receive sensitivity, the greater the area that vibrates, the better. The area that vibrates is dependent on the shape of the diaphragm. In the case of a circular, square or regular hexagonal diaphragm, since the diaphragm is secured from around in a generally uniform manner, the diaphragm is only able to vibrate near its center. As a result, in effect, only approximately 30 to 40% of the cavity area is used effectively. On the other hand, in the case of an elongate rectangular (oblong) diaphragm, the extent to which it is bound from around is mitigated, and displacement in a more even manner becomes possible as compared to a circular diaphragm, etc. In this case, approximately 60% of the area vibrates effectively. Thus, from the standpoint of improving acoustic pressure and receive sensitivity, an elongate rectangular shape is preferable. However, when a shape that is elongate to some extent is adopted, as in a rectangular diaphragm, characteristic high-order vibration modes occur. The various vibration modes that occur in the diaphragm have an influence on acoustic characteristics, e.g., radiated acoustic pressure, frequency characteristics, pulse characteristics. Accordingly, controlling vibration modes becomes extremely important.
Various vibration modes may be excited in the diaphragm of a CMUT. Ordinarily, when using an ultrasonic transducer, of the innumerable vibration modes that exist, a mode called fundamental mode where the diaphragm as a whole vibrates in phase is preferable. The reason being that because the diaphragm as a whole moves in phase, it is possible to convert sound and electricity most efficiently. In the case of a mode called high-order mode where a plurality of parts that serve as anti-nodes are created in the diaphragm, there will be places within the diaphragm where the vibration phases differ by 180 degrees. When sound is radiated in such a mode, a given region of the diaphragm may vibrate in a direction that compresses the medium that is in contact with the diaphragm, thereby radiating a positive pressure (compression wave), while simultaneously at another region of the diaphragm, the medium may be expanded, thereby radiating a negative pressure (expansion wave). Thus, the positive and negative sounds would cancel each other out, causing the net radiated acoustic pressure to drop. Similarly, in the case of reception, if there is a region where the diaphragm's displacement is in the opposite direction relative to the inputted acoustic pressure, sensitivity would drop since the positive and negative receive currents or voltages would cancel each other out.
Such phenomena are not problems of individual vibration modes, but instead have influences in the form of interference among separate vibration modes as well. In general, when some medium that radiates energy is in contact with the diaphragm, the individual vibration modes each possess a bandwidth to some degree. Thus, there exists a region where the band of the fundamental mode overlaps with the band of a high-order mode. In this case, there arises a frequency where the phase of the fundamental mode does not match with the phase of the high-order mode, and by a similar mechanism as that discussed above, there occurs a drop in radiated acoustic pressure or sensitivity. Accordingly, in order to widen the available frequency band, interference among the vibration modes must be considered.
On the other hand, the vibration modes of a diaphragm are dependent on the diaphragm's shape and boundary conditions. In the case of a shape where the distance from the center of the diaphragm to the supporting walls that bind the diaphragm may be considered uniform, e.g., a circular shape, or a regular polygonal shape such as a regular hexagonal shape, which are widely in use, the resonance frequencies of the fundamental mode and a high-order mode would always be of a constant ratio. Accordingly, once the shape is determined, the frequency characteristics are uniquely determined. On the other hand, if the distance from the center of the diaphragm to the surrounding supporting walls is not uniform and there is anisotropy, by way of example, in a case where the diaphragm shape is an elongate rectangular shape, the frequency of the excited vibration mode would vary largely depending on the ratio of the length of the longer side of that diaphragm to the width of the shorter side (i.e., the aspect ratio between representative long and short lengths (representative aspect ratio), or in the case of a rectangle, the length-to-width aspect ratio). Accordingly, in order to secure some available bandwidth, it is necessary that the aspect ratio of the representative lengths of the diaphragm be set appropriately.
An object of the present invention is to reduce the influences of the individual vibration modes and of interference among the vibrations on acoustic characteristics even in cases where the shape of the diaphragm of a capacitive micro-machine is such that the distance from the diaphragm center to the supporting posts that bind the diaphragm is not isotropic.
In cases where the diaphragm has a shape that is elongate to some extent, a typical example being a rectangular diaphragm, the vibration modes that are excited in the longer direction and shorter direction of the diaphragm may be considered separately. Of the vibration modes that are determined by the width of the diaphragm in the direction of the short side, the one with the lowest frequency becomes the resonance frequency of the fundamental mode. On the other hand, although the vibration mode frequencies in the lengthwise direction of the diaphragm are ordinarily higher than the resonance frequency of the fundamental mode, as its length becomes longer relative to the width in the short direction (i.e., as the long to short aspect ratio becomes greater), the resonance frequencies of the high-order modes approach the resonance frequency of the fundamental mode. In the case of a finite aspect ratio, there exist points within the band of the fundamental mode where a drastic drop in sensitivity occurs due to interference with high-order modes. On the other hand, in the case of an aspect ratio that is infinitely long, the resonance frequencies of all the high-order modes that are excited in the lengthwise direction of the diaphragm converge towards the fundamental mode frequency. In this case, since the inter-mode interferences all cancel one another out, it becomes equivalent to a state where only the fundamental mode is vibrating. With an actual diaphragm, it is not possible to create an infinite aspect ratio. However, it is possible to create a state that may be deemed the same as an infinite aspect ratio for practical purposes by making the aspect ratio be greater than a certain value. In so doing, since local sensitivity reduced regions that occur due to inter-mode interference may be suppressed, it is possible to attain characteristics that are more wide band in practical terms.
As such, in a case where the distance from the center of the diaphragm to the supporting walls is not uniform, the present invention sets the ratio of the length of the diaphragm in the direction of a first axis to the length in the direction of a second axis that is perpendicular to the first axis (i.e., representative aspect ratio) to a value that allows a signal level of a locally occurring frequency at which the amplitude drops or the sensitivity drops to be suppressed below a predetermined value within a bandwidth of at least one of transmission and reception by an ultrasonic probe.
An ultrasonic probe of the present invention comprises a capacitive micro-machine and at least one or more acoustic media that are in contact with the capacitive micro-machine. The capacitive micro-machine comprises a substrate having a first electrode and a diaphragm having a second electrode, wherein the diaphragm is secured to the substrate at its peripheral parts by means of supporting walls that rise from the substrate, and a cavity is formed between the substrate and the diaphragm. The ultrasonic probe is characterized in that the ratio of, of the representative dimensions of the diaphragm of the ultrasonic probe, the short direction to the long direction is equal to or greater than a value that does not cause acoustic performance degradation within a used sensitivity band.
The present invention realizes an ultrasonic probe that suppresses spurious response caused by high-order vibration modes and that may be used in a wider band.
Embodiments of the present invention are described below. It is noted that the contents of the cell structures and device configurations discussed herein are merely examples, and that other embodiments may be realized through combinations and replacements of the embodiments with known techniques.
[First Embodiment]
As shown in
Assuming that the CMUT (10) shown in
It is noted that the diaphragm 5 and the upper electrode 3 of the present embodiment are depicted as rectangles of the same size. However, with respect to the present invention, the shapes and sizes need not necessarily be rectangular as in
The substrate 1, the lower electrode 2, the diaphragm 5, the supporting walls 8, the insulator 4, and the upper electrode 3 are made of materials that may be processed by semiconductor process techniques. By way of example, the materials disclosed in U.S. Pat. No. 6,359,367 may be used. To provide examples, they may include silicon, sapphire, glass materials of all types, polymers (such as polyimide), polysilicon, silicon nitride, silicon oxynitride, thin film metals (such as aluminum alloys, copper alloys and tungsten), spin-on-glasses (SOGs), implantable or diffused dopants and grown films such as silicon oxides and nitrides. The interior of the cavity 7 may be a vacuum, or be filled with air or some gas. When stationary (i.e., when not operating), the gap of the cavity 7 (in the z-direction) is maintained mainly by virtue of the rigidity of the substrate 1, diaphragm 5, supporting walls 8 and upper electrode 3.
It is noted that the arrangement of the CMUT array 300 shown in
Next, the operation principles of a CMUT are described. The CMUT (10) functions as a variable capacitor in which the lower electrode 2 and the upper electrode 3 are disposed with the cavity 7 and insulator 4 in-between. When a force is exerted on the upper electrode 3 to displace it in the z-direction, the gap between the lower electrode 2 and the movable upper electrode 3 varies, causing the capacitance of the CMUT to vary. Since the upper electrode 3 and the diaphragm 5 are coupled, the upper electrode 3 is also displaced when a force is exerted on the diaphragm 5. In this case, when a voltage is applied across the lower electrode 2 and the upper electrode 3 and a charge is accumulated, the temporal change in the gap between the lower electrode 2 and the upper electrode 3 becomes a temporal change in capacitance, and a new voltage is generated across both electrodes. Thus, when a force that causes some mechanical displacement, such as an ultrasonic wave, etc., is transmitted to the diaphragm 5, that displacement is converted into an electrical signal (voltage or current). In addition, when a difference in potential is imparted between the lower electrode 2 and the upper electrode 3, charges of respectively different signs are accumulated in the electrodes, an attracting force is generated between the electrodes due to an electrostatic force, and the upper electrode 3 is displaced towards the substrate 1. In this case, since the upper electrode 3 and the diaphragm 5 are coupled, the diaphragm 5 is also simultaneously displaced. Thus, if an acoustically propagating medium, such as air, water, plastic, rubber, a living organism, etc., exists above (i.e., in the z-direction of) the diaphragm, the displacement of the diaphragm 5 is transmitted to the medium. The displacement may also be temporally varied by temporally varying the voltage applied across the electrodes, as a result of which sound is radiated. In other words, this CMUT (10) functions as an electroacoustic transducer having the function of radiating an inputted electrical signal to a medium that is in contact with the diaphragm 5 as an ultrasonic signal, and of conversely converting an ultrasonic signal from the medium into an electrical signal and outputting it.
Next, vibration modes of a diaphragm of a CMUT are described. A diaphragm of a CMUT may be excited in various vibration modes. Examples of the vibration modes of a regular hexagonal cell are shown in
On the other hand, in the case of an elongate rectangular cell such as that shown in
where w and l are the width and length of the rectangle, and G and J are constants determined by boundary conditions. The vibration modes of rectangles have a characteristic where, as the length-to-width aspect ratio increases, the high-order modes converge towards the frequency of the fundamental mode. Results obtained by normalizing the high-order mode frequencies by the fundamental mode frequency while varying the length-to-width ratio of a rectangle are shown in
Next, problems resulting from such vibration modes are described. The acoustic frequency characteristics of a CMUT are shown in
In general, high sensitivity and wideband characteristics are desired for ultrasonic transducers. Accordingly, it is preferable that the band around the fundamental mode be wide. However, it is undesirable for bandwidth to be narrowed by the occurrence of dips due to the existence of high-order modes. In addition, for an ultrasonic probe that utilizes sound waves of various frequencies, it would be inappropriate for the transmit acoustic pressure to drop locally only around the frequency of a dip. As already discussed above, in the case of circular or regular hexagonal cell shapes, since the frequency of a high-order mode is fixed at a constant ratio with respect to the frequency of the fundamental mode, the dip position is uniquely determined.
Accordingly, band improvement is, in principle, difficult. On the other hand, in the case of elongate cell shapes such as rectangles, the frequency of each high-order vibration mode is determined by the length-to-width aspect ratio. Thus, the dip position may be controlled by varying the length-to-width aspect ratio. However, a high-order mode of a rectangle occurs at a position that is closer to the fundamental mode frequency than is a high-order mode of a circle or a regular hexagon. Specifically, a dip of a rectangle would actually be in a direction that narrows the band of the fundamental mode, and would be in the opposite direction to improving wide band characteristics.
By way of example, experiment results for transmit sensitivity with respect to CMUT cells whose length-to-width aspect ratios were “2,” “4,” “8,” and “16” are shown in
On the other hand, from the present experimental data, it can be seen that the interval between the dips becomes narrower as the length-to-width aspect ratio of the rectangle increases, and also that the depths of the dips become less. By way of example, the depths of the dips when the length-to-width aspect ratio is “8” are fractions of those when the length-to-width aspect ratio is “4.” Further, it can be seen that the depths of the dips become even smaller when the length-to-width aspect ratio is “16.” The principles thereof are shown in
Utilizing the above-mentioned characteristics of interference among the vibration modes of a rectangular diaphragm, the influences of dips may be reduced even for rectangles. Although the number of dips occurring within the fundamental mode band increases as the length-to-width aspect ratio increases, the depths of the dips decrease. Accordingly, dips would ultimately not occur if the length-to-width aspect ratio is infinitely large. Although an infinite length-to-width aspect ratio is not actually possible, there exists a threshold that poses no problem for actual use if the dips become sufficiently small. In the case shown in
[Second Embodiment]
In
Ordinarily, the dynamic range of signals used in ultrasonic diagnostic devices is 50 to 60 dB or greater. If the purpose is to image living organisms, the standard imaging region is approximately 10 cm in depth from the body surface, and the sensitivity band of probes that are most often used with such depths is generally 10 MHz or less. The attenuation coefficient of living organisms is said to be generally the same as water, namely, approximately 0.5 [dB/cm/MHz]. By way of example, if one were to perform imaging up to a depth of 10 cm at 5 MHz, the signal transmitted from the probe would be attenuated by 0.5 [dB/cm/MHz]×10 [cm]×2×5 [MHz]=50 dB as it travels to and from a reflection point within the living organism. Accordingly, under such circumstances, a signal dynamic range (DR) of approximately 50 dB would be demanded of the probe. For this reason, ordinarily, for medical ultrasonic diagnostic devices, etc., approximately 50 dB is secured for the transmit/receive sensitivity dynamic range (DR). Accordingly, if, for transmission and reception, there is any spurious response, such as ringing, etc., at a level of at least 50 dB or greater in transmit pulse, there is a possibility that a drop in performance may be caused, such as image resolution degradation, etc. From such a perspective, it is demanded that ringing caused by interference between the fundamental mode and high-order modes be 50 dB or less for transmission and reception, and that it be half that, namely 25 dB or less, for transmission only or reception only.
In actual design, with the present invention, the length-to-width aspect ratio may be defined as follows.
[Third Embodiment]
In the second embodiment, a frequency and depth that suit a specific purpose are set, but conditions may vary for other purposes. By way of example, even if the purpose is the same, that is, imaging living organisms, a shallower region may sometimes be imaged at a higher resolution using a higher frequency wave. For example, for imaging up to approximately 3 cm at 20 MHz, the minimum requisite dynamic range would be 0.5 [dB/cm/MHz]×3 [cm]×2×20 [MHz]=60 dB. According to the results in
To sum up the above, a method for setting the length-to-width ratio may be defined in more general terms as follows. Based on experiment data, the relationship between length-to-width aspect ratio and DE for transmission and reception is shown in
[Fourth Embodiment]
The present invention is also able to set optimal length-to-width aspect ratios based on the resonance frequency of each vibration mode. In the first and second embodiments, it was indicated that a wide band or a short pulse could be attained with respect to frequency characteristics or a time waveform by having the length-to-width aspect ratio of the rectangle be “8” or greater. On the other hand, according to the results in
Accordingly, in an actual design for attaining wide band characteristics equivalent to or greater than a regular hexagonal cell, the length-to-width aspect ratio should be made to be such that there are six or more vibration modes for which an odd number of anti-nodes exist in the region where the normalized frequency is 2 or less.
[Fifth Embodiment]
In the first to fourth embodiments, methods for setting a length-to-width aspect ratio were presented with respect to cases where the cell shape was rectangular. However, actual cell shapes are not necessarily limited to those that are strictly rectangular. As shown in
| Number | Date | Country | Kind |
|---|---|---|---|
| 2009-215755 | Sep 2009 | JP | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/JP2010/063607 | 8/11/2010 | WO | 00 | 1/20/2012 |
| Publishing Document | Publishing Date | Country | Kind |
|---|---|---|---|
| WO2011/033887 | 3/24/2011 | WO | A |
| Number | Name | Date | Kind |
|---|---|---|---|
| 5902242 | Ustuner et al. | May 1999 | A |
| 6359367 | Sumanaweera et al. | Mar 2002 | B1 |
| 7410950 | Garry et al. | Aug 2008 | B2 |
| 7449821 | Dausch | Nov 2008 | B2 |
| 20040190377 | Lewandowski et al. | Sep 2004 | A1 |
| 20040266362 | Watkins et al. | Dec 2004 | A1 |
| 20050177045 | Degertekin et al. | Aug 2005 | A1 |
| 20060004289 | Tian et al. | Jan 2006 | A1 |
| 20070267945 | Sudol | Nov 2007 | A1 |
| 20090301200 | Tanaka et al. | Dec 2009 | A1 |
| 20100156244 | Lukacs et al. | Jun 2010 | A1 |
| Number | Date | Country |
|---|---|---|
| 101006361 | Jul 2007 | CN |
| 101238754 | Aug 2008 | CN |
| 2007-318742 | Dec 2007 | JP |
| 2008-510324 | Apr 2008 | JP |
| 2007046180 | Apr 2007 | WO |
| Entry |
|---|
| Robert D. Blevins, “Formulas for Natural Frequency and Mode Shape”, ISBN 1-57524-184-6. |
| International Search Report and Written Opinion for PCT International Application No. PCT/2010/063607, mailed on Dec. 7, 2010. |
| Robert D. Blevins, “Formulas for Natural Frequency and Mode Shape”, ISBN 1-57524-184-6, 1979. |
| Number | Date | Country | |
|---|---|---|---|
| 20120123268 A1 | May 2012 | US |
