Underwater Imaging Using a 1 16 CMUT Linear ... - Semantic Scholar

4 downloads 0 Views 4MB Size Report
Mar 1, 2016 - Peake, W.H.; Thurston, E.G. The lowest resonant frequency of a water-loaded ... Frederick, J.; Seydel, J.; Fairchild, R. Improved Ultrasonic ...
sensors Article

Underwater Imaging Using a 1 ˆ 16 CMUT Linear Array Rui Zhang 1,2 , Wendong Zhang 1,2 , Changde He 1,2 , Yongmei Zhang 3 , Jinlong Song 1,2 and Chenyang Xue 1,2, * 1

2 3

*

Key Laboratory of Instrumentation Science & Dynamic Measurement, North University of China, Ministry of Education, Taiyuan 030051, China; [email protected] (R.Z.); [email protected] (W.Z.); [email protected] (C.H.); [email protected] (J.S.) Science and Technology on Electronic Test and Measurement Laboratory, North University of China, Taiyuan 030051, China School of Computer Science, North China University of Technology, Beijing 100144, China; [email protected] Correspondence: [email protected]; Tel.: +86-351-3921-756

Academic Editor: Gonzalo Pajares Martinsanz Received: 25 January 2016; Accepted: 25 February 2016; Published: 1 March 2016

Abstract: A 1 ˆ 16 capacitive micro-machined ultrasonic transducer linear array was designed, fabricated, and tested for underwater imaging in the low frequency range. The linear array was fabricated using Si-SOI bonding techniques. Underwater transmission performance was tested in a water tank, and the array has a resonant frequency of 700 kHz, with pressure amplitude 182 dB (µPa¨ m{V) at 1 m. The ´3 dB main beam width of the designed dense linear array is approximately 5 degrees. Synthetic aperture focusing technique was applied to improve the resolution of reconstructed images, with promising results. Thus, the proposed array was shown to be suitable for underwater imaging applications. Keywords: capacitive micro-machined ultrasonic transducer linear array; transmission performance; synthetic aperture focusing technique; underwater imaging

1. Introduction Ultrasound imaging has played an important role in various areas, such as medical diagnosis, medical treatment, nondestructive testing, and ultrasound microscopy [1–3]. The ultrasonic transducer is the core component of ultrasound imaging, and currently piezoelectric micro-machined ultrasonic transducers (PMUTs) based on the piezoelectric effect are widely used [4,5]. However, PMUT performance in underwater and medical applications is limited by material properties and impedance matching issues [2,6,7]. Capacitive micro-machined ultrasonic transducers (CMUTs) have many advantages over conventional PMUTs, such as wide bandwidth, high mechanical-electrical conversion efficiency, and ease of integration with electronic circuits to enhance signal-to-noise ratio [2,8–12]. Furthermore, CMUT membranes have low mechanical impedance, which makes them match well with air and other fluid media, and are suitable for manufacturing in large arrays [2,6]. These characteristics promote CMUTs as the development direction for next generation ultrasonic transducers. Much research has been conducted regarding CMUT structural design, fabrication methods, and implementations. However, most studies have considered high frequency CMUTs (ě3 MHz) for medical imaging applications, with few studies on underwater imaging applications, which require low frequency. Roh et al. used finite element models to design a 1dimensional (1D) CMUT array robust to crosstalk [13]. ChiaHung et al. [14] designed and fabricated an underwater CMUT using full surface

Sensors 2016, 16, 312; doi:10.3390/s16030312

www.mdpi.com/journal/sensors

Sensors 2016, 16, 312

2 of 9

micro-machining technique. The transducer operated underwater at approximately 2 MHz with a Sensors 2016, 16, 312 2 of 9 detection range of 273 mm. Cheng et al. [15] realized B-mode imaging of a metal wire phantom using a 21-element with 3.8 MHz central bandwidth 116% water, 2 MHz 1-D witharray a detection range of 273 mm. frequency Cheng et al.and [15] fractional realized B-mode imaging of a in metal wirewhich limited the detection Doody al. [16] designed a CMUT-in-CMOS array, which achieved central phantom using a range. 21-element 1-Detarray with 3.8 MHz central frequency and fractional bandwidth frequency 3.5 MHz, fractional bandwidth 32%–44%, and pressure amplitude 181–184 dB (µPa¨ m{V) at 116% in water, which limited the detection range. Doody et al. [16] designed a CMUT-in-CMOS array, which achieved frequency 3.5work, MHz,we fractional bandwidth 32%–44%, 15 mm when operated in a central water tank. In this designed and fabricated a 1 ˆand 16 pressure CMUT linear 181–184frequency dB (μPa ∙ 700 m/V)kHz, at 15 mm when operated in 182 a water tank. m{V) In thisatwork, weuse in arrayamplitude with resonance and pressure amplitude dB (µPa¨ 1 m for designed and fabricated a 1 × 16 CMUT linear array with resonance frequency 700 kHz, and pressure underwater imaging. amplitude 182 dB (μPa ∙ m/V) at 1 m for use in underwater imaging.

2. Structural Design

2. Structural Design

CMUT array elements are composed of multiple sensitive cells connected in parallel. Each cell is CMUT array elements are composed of multiple sensitive cells connected in parallel. Each cell composed of electrodes, vibrating membrane, vacuum cavity, insulating layer, and silicon substrate, is composed of electrodes, vibrating membrane, vacuum cavity, insulating layer, and silicon as shown cross-sectional in Figure 1.in A single often often cannot meetmeet imaging substrate, as shown cross-sectional Figure 1. CMUT A single array CMUTelement array element cannot requirements with its low lateral resolution, low transmission power, and poor directivity. Therefore, a imaging requirements with its low lateral resolution, low transmission power, and poor directivity. CMUT array composed of Ncomposed identicalofelements used toisimprove the imaging resolution. Therefore, a CMUT array N identicaliselements used to improve the imaging resolution. Rm

Top Electrode

Insulation Layer 1

tg

tm

Membrance Insulation Layer 2

Cavity Support

Substrate Bottom Electrode Al

SiO2

Si

Figure1. 1. The The structure Figure structureofofa acell. cell.

The natural frequency is one of important performance parameters, whether operating in The natural frequency is oneWhen of important performance parameters, whether operating in emission emission or receiving mode. CMUTs work in the liquid, the resonant frequency of circular or receiving mode. When CMUTs work in the liquid, the resonant frequency of circular membrane fr membrane 𝑓𝑟 is [17,18]:

is [17,18]:

0.469𝑡𝑚d 𝐸 0.469t 2 m √𝜌(1 −E𝜎 2 ) 𝑅𝑚 ˘ ` 𝑓𝑟 = (1) R2m ρ 𝜌1𝑙 𝑅´𝑚 σ2 √1 + 0.67 fr “ d 𝜌𝑡𝑚 ρl Rm 1 ` 0.67 represent the Young’s modulus, ρtmthe density of membrane, the density of

(1)

where E, 𝜌, 𝜌𝑙 , 𝜎, 𝑡𝑚 , 𝑅𝑚 liquid, the Poisson’s ratio, the thickness of the membrane, and the radius of the membrane, respectively. was chosen as thethe membrane material. Fromthe previous where E, ρ, ρl , σ, tHigh-resistivity the Young’s modulus, density of membrane, densityfinite of liquid, m , Rm representsilicon element software ANSYS analysis following Equation 𝑡𝑚 = of 3.5the μm membrane, and 𝑅𝑚 = 90respectively. μm. the Poisson’s ratio, the thickness of[19,20], the membrane, and the(1), radius The emission performance of ultrasonic transducers is closelyFrom linkedprevious with its structural parameters. High-resistivity silicon was chosen as the membrane material. finite element software From previous studies [21], several directivity functions were deduced by Huygens’ Principle to ANSYS analysis [19,20], following Equation (1), tm “ 3.5 µm and Rm “ 90 µm. guide the array structure design, and the resultant 1 × 16 CMUT structure is shown in Figure 2. The emission performance of ultrasonic transducers is closely linked with its structural parameters. Si-SOI low temperature wafer bonding technology [22,23] was used to fabricate the CMUT From previous studies [21], several directivity functions were deduced by Huygens’ Principle to guide linear array (see Figure 3). The silicon wafer was first thermally oxidized, then part-etched to form the array structure the resultant ˆ 16μm) CMUT structure shownwafer in Figure 2. wafer cavities (tg = 0.8 design, μm) andand an insulation layer1(0.15 (Figure 3a). Theissilicon and SOI Si-SOIbonded low temperature wafer bonding technology was to fabricate were using low temperature bonding (Figure[22,23] 3b), and theused silicon substratethe andCMUT buriedlinear arrayoxide (see layer Figure wafer was first thermally then part-etched cavities of 3). theThe SOI silicon wafer was eliminated (Figure 3c) to oxidized, form the silicon membrane (tmto= form 3.5 μm). isolation channel was formed and dry etching, (tg = Finally, 0.8 µm)anand an insulation layer (0.15using µm)photolithography (Figure 3a). The silicon wafer and andlow SOIpressure wafer were chemical deposition was used to form a3b), 0.15 μmthe insulation layer to prevent conductive bonded usingvapor low temperature bonding (Figure and silicon substrate and buried oxide layer contact between the top electrodes and the vibration membrane, and evaporation methods were an of the SOI wafer was eliminated (Figure 3c) to form the silicon membrane (tm = 3.5 µm). Finally, used to form the top electrodes with Al. To ensure fine conductive contact between the bottom isolation channel was formed using photolithography and dry etching, and low pressure chemical electrode and silicon substrate, the other side of the Si wafer had phosphorus ions implanted and vapor deposition was used to form a 0.15 µm insulation layer to prevent conductive contact between metal Al deposited (Figure 3d).

the top electrodes and the vibration membrane, and evaporation methods were used to form the top electrodes with Al. To ensure fine conductive contact between the bottom electrode and silicon

Sensors 2016, 16, 312

3 of 9

substrate, the other side of the Si wafer had phosphorus ions implanted and metal Al deposited (Figure 3d).2016, 16, 312 Sensors 3 of 9 Sensors 2016, 16, 312

3 of 9

Figure Theresultant resultant11ˆ × 16 structure. Figure 2. 2. The 16CMUT CMUTarray array structure. Figure 2. The resultant 1 × 16 CMUT array structure. Si

6-inch

Si

Si

Si

SOI Wafer Si

6-inch

SiSOi 2

Si

Si

SiO2

SOI Wafer SiO2

SiO2

Si O 2

Si O 2

6-inch Silicon Wafer (After 6-inch thermal oxidation and Silicon Wafer (After thermal) photolithography oxidation and photolithography )

Si O 2 Si O 2

Si Si

Si Si

Si Si

Si Si

(a)

(b)

(a)

(b)

SiO2

SiO2

SiO2

SiO2

Si

Si

Si Si

Si

Si

(c)

(c)

(d)

Al

Al

SiO2

SiO2

Si

(d)

Si

Figure 3. The main fabrication flow-charts. (a) part etched to form cavities; (b) bonding; (c) form

Figure The main fabrication flow-charts. (a) (a) part etched to to form cavities; (b) bonding; (c) form(c) form Figure 3. The3.main fabrication flow-charts. form cavities; (b) bonding; membrane; (d) form isolation channel, insulationpart layeretched and electrodes. membrane; (d) form isolation channel, insulation layer and electrodes. membrane; (d) form isolation channel, insulation layer and electrodes. 3. Underwater Experimental 3. Underwater Experimental

3. Underwater Experimental The as shown shown inin Figure Figure4.4.The Thelinear lineararray array TheCMUT CMUTresonant resonant frequency frequency was was found found as waswas encapsulated for insulation from the water [21], and a precision impedance analyzer (Agilent 4294A, encapsulated for insulation from the water [21], and a precision impedance analyzer (Agilent 4294A, The CMUT resonant frequency was found as shown in Figure 4. The linear array was encapsulated Agilent Technologies, Santa States) was used usedtotomeasure measurethe thearray array impedance in Agilent Technologies, SantaClara, Clara, CA, CA, United United States) impedance was impedance for insulation from the water [21], and a precision analyzer (Agilent 4294A,inAgilent water (underwater 13 °C). °C).The TheCMUT CMUTresonant resonant frequency water (underwaterpenetration penetration0.45 0.45 m, m, water water temperature temperature 13 frequency = = Technologies, Santa Clara, CA, United States) was used to measure the array impedance in water 700700 kHz, and electric CMUT transmission transmissioncharacteristics characteristics (transmitting kHz, and electricconductance conductance==222.35 222.35 mS. mS. The CMUT (transmitting ˝ (underwater penetration 0.45 m, water The CMUT frequency = 700 kHz, voltage response anddirectivity) directivity) weretemperature analyzed over13 the frequency range (100–1000 kHz). voltage response and were analyzed theC). frequency rangeofresonant ofinterest interest (100–1000 kHz). and electric conductance = 222.35 mS. The CMUT transmission characteristics (transmitting voltage The CMUT transmittingvoltage voltageresponse response was measured 5. 5. AA CMUT linear The CMUT transmitting measuredas asshown shownininFigure Figure CMUT linear array (transmitter) and standard hydrophone were placed face 1m apart. TheThe response and directivity) were analyzed over the (receiver) frequency range of interest (100–1000 kHz). array (transmitter) and a astandard hydrophone (receiver) were placed facetotoface, face, 1m apart. received electrical signalwas was displayed on an an oscilloscope oscilloscope (Agilent with direct current received electrical signal displayed on (Agilent54624A), 54624A), with direct current The CMUT transmitting voltage response was measured as shown in Figure 5. (DC) A(DC) CMUT bias V. TheCMUT CMUTarray arraywas wasdriven driven with with aa 5 5 cycle burst incorporating kHz andand 20 20 V. The cycle (receiver) burstsignal signalwere incorporating 100–1000 kHz linearbias array (transmitter) and a standard hydrophone placed100–1000 face to face, 1 m apart. amplitude . amplitude of of 2020 VppV.ppsignal The received electrical was displayed on an oscilloscope (Agilent 54624A), with direct current The transmitting voltageresponse responsecan can be be expressed expressed as transmitting voltage as[24]: [24]: (DC) biasThe 20 V. The CMUT array was driven with a 5 cycle burst signal incorporating 100–1000 kHz 𝑢𝑠 ∙ 𝑙 𝑢 ∙ 𝑙 𝑆𝑣 = 20𝑙𝑔 − 𝑀 𝑠 and amplitude of 20 V pp . (2) 𝑆𝑣 = 20𝑙𝑔 𝑢𝑓 − 𝑀𝑜𝑜 (2) 𝑢𝑓 The transmitting voltage response can be expressed as [24]: Sv “ 20lg

us ¨ l ´ Mo uf

(2)

Sensors 2016, 16, 312

4 of 9

Sensors 2016, 16, 312 Sensors 2016, 16,16, 312 Sensors 2016, 312

4 of 9 4 4ofof 99

wherewhere l is thel distance between the twothe transducers, u f is the driving driving voltage voltage of the transmitting is the distance between two transducers, 𝑢𝑓applied is the applied of the CMUT, u is the collected voltage standard hydrophone, and M is the receiving sensitivity of the where l is the distance between the two transducers, 𝑢 is the applied driving voltage the s l is the transmitting CMUT, 𝑢𝑠 is the collected standard 𝑀𝑜 isvoltage the receiving where distance between the twovoltage transducers, 𝑢𝑓𝑓 hydrophone, is theo appliedand driving ofofthe transmitting 𝑢 collected standard hydrophone, andobtained 𝑀𝑜frequencies, the receiving hydrophone, the CMUT transmitting voltage response was obtained at different as shown sensitivity ofCMUT, the hydrophone, CMUTvoltage transmitting voltage response was at receiving different 𝑜 isis the transmitting CMUT, 𝑢𝑠 𝑠 isis the the the collected voltage standard hydrophone, and 𝑀 sensitivity the hydrophone, the CMUT voltage response was obtained different frequencies, shown Figure 5b. The 1 ×has 16transmitting CMUT linear array has resonant frequency 700 and Figure 5b. Theof1ofas ˆthe 16 CMUT linear array resonant frequency 700 kHz, and pressure amplitude of sensitivity hydrophone, the CMUT transmitting voltage response was obtained atatkHz, different pressure amplitude ofFigure 182 dB5b. (μPa ∙ m/V) at CMUT 1CMUT m for linear underwater applications. asasshown The 11 ××applications. 16 has resonantfrequency frequency700 700kHz, kHz,and and frequencies, shown Figure 5b. The 16 array has resonant 182 frequencies, dB (µPa¨ m{V) at 1m for underwater pressure applications. pressureamplitude amplitudeofof182 182dB dB(μPa (μPa∙ ∙m/V) m/V)at at 11 m m for for underwater applications. 222.35

Conductance, G, Conductance, G, mS Conductance, G, mS mS

222.35 222.35 177.88

177.88 177.88 133.41

133.41 133.41 88.94 88.94 88.94 44.47 44.47 44.47 0 0 0 0 0 0

500

1000

1500

2000

Frequency, f, kHz

500 500

1000 1000

1500 1500

2000 2000

Frequency, f, kHz Frequency,frequency. f, kHz Figure4. 4. Resonant Resonant Figure frequency.

Transmitting

Receiving

Ultrasonic

Transmitting CMUT Transmitting CMUT CMUT

Receiving Hydrophone

Ultrasonic Ultrasonic 1m

Receiving

Hydrophone Hydrophone

1m 1m Water Water Water

(a)

Transmitting response, Sv, dB Transmitting response, Sv,Sv, dBdB Transmitting response,

Figure 4. Resonant frequency. Figure 4. Resonant frequency. 185 180 185 185 175 180 180 170 175 175 165 170 170 160 165 165 155 160 160 150 155 155 145 150

150 140 145 100 145 140 100 140 100

280

460

640

820

1000

820

1000

Frequency, f, kHz 280

460 640 f,640 kHz 280 Frequency, 460

(b)

820

1000

Frequency, f, kHz

(a) diagram of the transmitting voltage(b)response experiment; Figure 5. (a) Schematic (a) (b) (b) Transmitting Figure 5. (a) response. Schematic diagram of the transmitting voltage response experiment; Figure 5. (a) Schematic diagram of the transmitting voltage response experiment; (b) Figure 5. (a) response. Schematic diagram of the transmitting voltage response experiment; (b) Transmitting Transmitting response. The sector scanning experimental setup is shown in Figure 6. The operating conditions of the (b) Transmitting response. CMUT array were the same as previously, except it wasinnow excited with a burst signal at 700ofkHz The sector scanning experimental setup is shown Figure 6. The operating conditions the for two cycles. Thethe receiving transmitting array was fixed a precision rotary and The sector scanning experimental setup shown Figure 6.on operating conditions the The sector scanning experimental setup is shown in Figure 6.The The operating conditions of the CMUT array were same as and previously, except it wasin now excited with a burst signal attable 700 of kHz rotated to implement sector scanning for two-obstacle imaging. CMUT array were the same as previously, except it was now excited with a burst signal at 700 kHz for two cycles. The receiving and transmitting array was fixed on a precision rotary table and CMUT array were the same as previously, except it was now excited with a burst signal kHz for two cycles. The receiving and transmitting arrayfixed wason fixed on a precision rotaryand table and to rotated to implement sector scanning for two-obstacle imaging. twofor cycles. The receiving and transmitting array was a precision rotary table rotated Precision rotated tosector implement sector for two-obstacle Rotary implement scanning forscanning two-obstacle imaging. imaging. Table Precision Rotary Table Precision Rotary Table

CMUT CMUT Obstacles CMUT Obstacles

Obstacles

Figure 6. Sector scanning configuration. Figure 6. Sector scanning configuration.

Figure 6. Sector scanning configuration.

Figure 6. Sector scanning configuration.

Sensors 2016, 16, 312

5 of 9

Sensors 2016, 16, 312

5 of 9

Figure 7 shows the S-scan results andand corresponding directivity diagrams Figure 7 shows the S-scan results corresponding directivity diagramsfor forfour fourtransmitting transmitting conditions Figure 7b,d,f,h, where N is the number of array elements, d is the element conditions Figure 7b,d,f,h, where N is the number of array elements, d is the element spacing spacing between arrayarray elements, and and λ is λ the TheThe ´3−3 dBdB main beam between elements, is wavelength. the wavelength. main beamwidth widthofofFigure Figure7b,d,f,h 7b,d,f,h are are approximately 21˝ , 6.4 8˝ , and 5˝ , 5°, respectively. approximately 21°,˝ ,6.4°, 8°, and respectively. 0 -30

30

-60 1

-90

60

0.8 0.6 0.4 0.2

90

N=5,d=0.5λ

-120

-150

120

150 180

(a)

(b) 0 -30

30

-60 1

-90

60 0.8 0.6 0.4 0.2

90

N=5,d=1.5λ

-120

-150

120

150 180

(c)

(d) 0 -30

30

-60 1

-90

60

0.8 0.6 0.4 0.2

90

N=13,d=0.5λ

-120

-150

120

150 180

(e)

(f) 0 30

-30

60

-60 1

0.8 0.6 0.4 0.2

90

-90

N=16,d=0.5λ

-120

120

150

-150 180

(g)

(h)

Figure 7. Sector scanning results: (a) N = 5, d = 0.5·λ; (b) directivity of diagram (a); (c) N = 5, d = 1.5·λ; Figure 7. Sector scanning results: (a) N = 5, d = 0.5¨ λ; (b) directivity of diagram (a); (c) N = 5, d = 1.5¨ λ; (d) directivity of diagram (c); (e) N = 13, d = 0.5·λ; (f) directivity of diagram (e); (g) N = 16, d = 0.5·λ; (d) directivity of diagram (c); (e) N = 13, d = 0.5¨ λ; (f) directivity of diagram (e); (g) N = 16, d = 0.5¨ λ; (h) directivity of diagram (g). (h) directivity of diagram (g).

Sensors 2016, 16, 312

6 of 9

(1)

For the fixed array element number (Figure 7a,c, or Figure 7b,d), the main lobe becomes sharper as d increases. However, when d > λ [19], strong grating lobes emerge, causing interference in the ultrasound image. (2) For fixed array length (Figure 7c,e, or Figure 7d,f), the directivity of a dense array is better than Sensorsthat 2016,of16, 6 of 9 a 312 sparse array. (3) For fixed element spacing (Figure 7a,e,g or Figure 7b,f,g), directivity improves as N increases. (1) For the fixed array element number (Figure 7a,c, or Figure 7b,d), the main lobe becomes The wider main lobe also make severe interference. sharper as d increases. However, when d > λ [19], strong grating lobes emerge, causing interference the ultrasound image. Thus, N = 16,in d= 0.5λ was selected as the transmission mode for subsequent underwater imaging. (2) For fixed array length (Figure 7c,e, or Figure 7d,f), the directivity of a dense array is better than 4. Imaging that of a sparse array. (3) For fixed element spacing (Figure 7a,e,g or Figure 7b,f,g), directivity improves as N increases. To reduce the influence of side lobes and improve the lateral resolution of reconstructed CMUT The wider main lobe also make severe interference. underwater images, the synthetic aperture focusing technique (SAFT) [25–27], an optimization method involving by λdelay-and-sum on the A-scan signals, wasfor applied [28,29]. Thus,B-scan N =implemented 16, d = 0.5 was selected as received the transmission mode subsequent Figure 8 shows the SAFT experimental setup. The operating conditions of the CMUT linear array underwater imaging. were the same as for Section 3. The CMUT array was fixed on the electronic control guideway 4. Imaging and moved horizontally to implement linear scanning for two-obstacle imaging. 16 elements were simultaneously excited. Figure 9 shows the SAFT reconstructed image (Figure 9a,b) is a To reduce the influence of side lobes and improve the lateral resolution of reconstructed significant improvement over traditional B-scan (Figure 9c,d), and artifacts caused by side lobes CMUT underwater images, the synthetic aperture focusing technique (SAFT) [25–27], an are effectively suppressed. Thus, the proposed array greatly reduces transmission issues without optimization method involving B-scan implemented by delay-and-sum on the received A-scan requiring phased transmission, thereby improving lateral resolution, which will be of great benefit in signals, was applied [28,29]. underwater imaging. Electronic Control Guideway

CMUT

Obstacles

Figure Figure 8. 8. Linear Linear scanning scanning imaging imaging configuration. configuration.

Figure 8 shows the SAFT experimental setup. The operating conditions of the CMUT linear array were the same as for Section 3. The CMUT array was fixed on the electronic control guideway and moved horizontally to implement linear scanning for two-obstacle imaging. 16 elements were simultaneously excited. Figure 9 shows the SAFT reconstructed image (Figure 9a,b) is a significant improvement over traditional B-scan (Figure 9c,d), and artifacts caused by side lobes are effectively suppressed. Thus, the proposed array greatly reduces transmission issues without requiring phased transmission, thereby improving lateral resolution, which will be of great benefit in underwater imaging.

Sensors 2016, 16, 312

7 of 9

Sensors 2016, 16, 312

7 of 9

(a)

(b)

(c)

(d)

Figure 9. 9. Reconstructed Figure Reconstructed images images (a) (a) B-san B-san reconstructed reconstructed image; image; (b) (b) 3-dimensional 3-dimensional view view of of diagram diagram (a); (a); (c) SAFT reconstructed image; (d) 3 dimensional view of diagram (c). (c) SAFT reconstructed image; (d) 3 dimensional view of diagram (c).

5. Conclusions 5. Conclusions A 1 × 16 CMUT array was designed and fabricated for underwater imaging in the low A 1 ˆ 16 CMUT array was designed and fabricated for underwater imaging in the low frequency frequency range. The transmission performance of the array was analyzed in a water tank, showing range. The transmission performance of the array was analyzed in a water tank, showing transmission transmission voltage response 182 dB (μPa ∙ m/V) was achieved at 1 m underwater. Directivity and voltage response 182 dB (µPa¨ m{V) was achieved at 1 m underwater. Directivity and sector scanning sector scanning were also analyzed to determine the optimal transmission mode for linear were also analyzed to determine the optimal transmission mode for linear underwater imaging. underwater imaging. Significant resolution improvement was obtained by applying SAFT to the Significant resolution improvement was obtained by applying SAFT to the received A-scan signals. received A-scan signals. Thus, the proposed CMUT array shows great benefit for underwater Thus, the proposed CMUT array shows great benefit for underwater detection applications, such as detection applications, such as obstacle avoidance, distance measuring, and imaging. obstacle avoidance, distance measuring, and imaging. Acknowledgments: This Thiswork workwas was supported by National Science Foundation for Distinguished Young Acknowledgments: supported by National Science Foundation for Distinguished Young Scholars Scholars ChinaGrant under61525107, Grant 61525107, National Foundation of China 61127008 of China of under National NaturalNatural ScienceScience Foundation of China underunder GrantGrant 61127008 and 61371143, and the program of China under GrantGrant 2013AA09A412. and 61371143, and“863” the “863” program of China under 2013AA09A412. Author Author Contributions: Contributions: The The author author Rui Rui Zhang Zhang designed designed the the CMUT CMUT array array structure, structure, organized organized the the experiments, experiments, performed the data collection and analysis, and drafted the manuscript. Chenyang Xue provided with the funding performed the data collection and analysis, and drafted the manuscript. Chenyang Xue provided with the support. Wendong Zhang and Changde He designed the fabrication process. Jinglong Song completed the funding support. Wendong Zhang andZhang Changde He designed the fabrication Jinglong processing of the transducer. Yongmei performed provided the imagingprocess. algorithm guide.Song completed the processing of the transducer. Yongmei Zhang performed provided the imaging algorithm guide. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest.

References References 1. 1. 2.

3.

Caronti, A.; Caliano, G.; Carotenuto, R.; Savoia, A.; Pappalardo, M.; Cianci, E.; Foglietti, V. Capacitive Caronti, A.; Caliano, G.; Carotenuto, R.; Savoia, A.;for Pappalardo, M.; Cianci, E.; Foglietti, V. 37, Capacitive micromachined ultrasonic transducer (CMUT) arrays medical imaging. Microelectron. J. 2006, 770–777. micromachined ultrasonic transducer (CMUT) arrays for medical imaging. Microelectron. J. 2006, 37, 770–777. [CrossRef] Oralkan, O.; Ergun, A.S.; Johnson, Johnson, J.A.; J.A.; Karaman, Karaman, M.; M.; Demirci, Demirci, U.; U.; Kaviani, Kaviani, K.; K.; Lee, Lee, T.H.; T.H.; Khuri-Yakub, Khuri-Yakub, B.T. B.T. Capacitive micromachined ultrasonic transducers: Next-generation arrays for acoustic imaging? IEEE Capacitive micromachined ultrasonic transducers: Next-generation arrays for acoustic imaging? IEEE Trans. Trans. Ultrason. Ferroelectr. Freq. Control 2002, 49, 1596–1610. Ultrason. Ferroelectr. Freq. Control 2002, 49, 1596–1610. [CrossRef] [PubMed] Choe, J.W.; Oralkan, O.; Nikoozadeh, A.; Gencel, M.; Stephens, D.N.; O'Donnell, M.; Sahn, D.J.; Khuri-Yakub, B.T. Volumetric Real-Time Imaging Using a CMUT Ring Array. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2012, 59, 1201–1211.

Sensors 2016, 16, 312

3.

4.

5.

6. 7. 8. 9. 10. 11.

12. 13. 14. 15. 16.

17. 18. 19. 20. 21.

22.

23.

24.

8 of 9

Choe, J.W.; Oralkan, O.; Nikoozadeh, A.; Gencel, M.; Stephens, D.N.; O’Donnell, M.; Sahn, D.J.; Khuri-Yakub, B.T. Volumetric Real-Time Imaging Using a CMUT Ring Array. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2012, 59, 1201–1211. [CrossRef] [PubMed] Emadi, T.A.; Buchanan, D.A. A novel 6 ˆ 6 element MEMS capacitive ultrasonic transducer with multiple moving membranes for high performance imaging applications. Sens. Actuators A Phys. 2015, 222, 309–313. [CrossRef] Chang, C.; Moini, A.; Nikoozadeh, A.; Sarioglu, A.F.; Apte, N.; Zhuang, X.; Khuri-Yakub, B.T. Singulation for imaging ring arrays of capacitive micromachined ultrasonic transducers. J. Micromech. Microeng. 2014, 24, 10700210. [CrossRef] Wang, H.; Wang, X.; He, C.; Xue, C. Design and Performance Analysis of Capacitive Micromachined Ultrasonic Transducer Linear Array. Micromachines 2014, 5, 420–431. [CrossRef] Ladabaum, I.; Jin, X.; Soh, H.T.; Atalar, A.; Khuri-Yakub, B.T. Surface micromachined capacitive ultrasonic transducers. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 1998, 45, 678–690. [CrossRef] [PubMed] Emadi, T.A.; Buchanan, D.A. Multiple Moving Membrane CMUT with Enlarged Membrane Displacement and Low Pull-Down Voltage. IEEE Electron Device Lett. 2013, 34, 1578–1580. [CrossRef] Jeong, B.; Kim, D.; Hong, S. Performance and reliability of new CMUT design with improved efficiency. Sens. Actuators A Phys. 2013, 199, 325–333. [CrossRef] Ergun, A.S.; Yaralioslu, G.G.; Khuri-Yakub, B.T. Capacitive micromachined ultrasonic Transducers: Theory and technology. J. Aerosp. Eng. 2003, 16, 76–84. [CrossRef] Mills, D.M.; Smith, L.S. Real Time in vivo Imaging with Capacitive Micromachined Ultrasonic Transducer (CMUT) Linear Arrays. In Proceedings of the 2003 IEEE Ultrasonic Symposium, Honolulu, HA, USA, 5–8 October 2003; pp. 568–571. Eccardt, P.C.; Niederer, K.; Fischer, B. Micromachined Transducers for Ultrasonic Applications. In Proceedings of the 1997 IEEE Ultrasonic Symposium, Toronto, ON, Canada, 5–8 October 1997. Roh, Y.; Khuri-Yakub, B.T. Finite element analysis of underwater capacitor micromachined ultrasonic transducers. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2002, 49, 293–298. [CrossRef] [PubMed] Liu, C.; Chen, P. Surface micromachined capacitive ultrasonic transducer for underwater imaging. J. Chin. Inst. Eng. 2007, 30, 447–458. [CrossRef] Cheng, X.; Chen, J.; Li, C.; Liu, J.; Shen, I.; Li, P. A Miniature Capacitive Ultrasonic Imager Array. IEEE Sens. J. 2009, 9, 569–577. [CrossRef] Doody, C.B.; Cheng, X.; Rich, C.A.; Lemmerhirt, D.F.; White, R.D. Modeling and Characterization of CMOS-Fabricated Capacitive Micromachined Ultrasound Transducers. J. Microelectromech. Syst. 2011, 20, 104–118. [CrossRef] Peake, W.H.; Thurston, E.G. The lowest resonant frequency of a water-loaded circular plate. Acoust. Soc. Am. 1954, 26, 166–168. [CrossRef] Wong, A.C. VHF Microelectromechanical Mixer-Filters. Ph.D. Thesis, The University of Michigan, Ann Arbor, MI, USA, 2001. Shen, W.; Miao, J.; Xiong, J.; He, C.; Xue, C. Micro-electro-mechanical systems capacitive ultrasonic transducer with a higher electromechanical coupling coefficient. IET Micro Nano Lett. 2015, 10, 541–544. Li, Y.P.; He, C.D.; Zhang, J.T.; Song, J.L.; Zhang, W.D.; Xue, C.Y. Design and analysis of Capacitive Micromachined Ultrasonic Transducer based on SU-8. Key Eng. Mater. 2014, 645–646, 577–582. [CrossRef] Zhang, R.; Zhang, W.D.; He, C.D.; Song, J.L.; Mu, L.F.; Cui, J.; Zhang, Y.M.; Xue, C.Y. Design of Capacitive Micromachined Ultrasonic Transducer (CMUT) linear array for underwater imaging. Sens. Rev. 2015, 36, 77–85. [CrossRef] Song, J.L.; Xue, C.Y.; He, C.D.; Zhang, R.; Mu, L.F.; Cui, J.; Miao, J.; Liu, Y.; Zhang, W.D. Capacitive Micromachined Ultrasonic Transducers (CMUTs) for Underwater Imaging Applications. Sensors 2015, 15, 23205–23217. [CrossRef] [PubMed] Zhang, R.; Xue, C.Y.; He, C.D.; Zhang, Y.M.; Song, J.L.; Zhang, W.D. Design and performance analysis of capacitive micromachined ultrasonic transducer (CMUT) array for underwater imaging. Microsyst. Technol. 2015. [CrossRef] Zheng, S.J.; Yuan, W.J.; Miu, R.X.; Xue, Y.Q. Underwater Acoustic Measurement Testing Technology; Harbin Engineering University Press: Harbin, China, 1995; pp. 223–232.

Sensors 2016, 16, 312

25. 26.

27.

28. 29.

9 of 9

Neild, A.; Hutchins, D.A.; Billson, D.R. Imaging using air-coupled polymer-membrane capacitive ultrasonic arrays. Ultrasonics 2004, 42, 859–864. Guarneri, G.A.; Pipa, D.R.; Junior, F.N.; Valéria, L.; de Arruda, R.; Victor, M.; Zibetti, W. A Sparse Reconstruction Algorithm for Ultrasonic Images in Nondestructive Testing. Sensors 2015, 15, 9324–9343. [CrossRef] [PubMed] Skjelvareid, M.H.; Olofsson, T.; Birkelund, Y.; Larsen, Y. Synthetic aperture focusing of ultrasonic data from multilayered media using an omega-K algorithm. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2011, 58, 1037–1048. [CrossRef] [PubMed] Corl, P.D.; Grant, P.M.; Kino, G. A Digital Synthetic Focus Acoustic Imaging System for NDE. In Proceedings of the 1978 Ultrasonics Symposium, Cherry Hill, NJ, USA, 25–27 September 1978. Frederick, J.; Seydel, J.; Fairchild, R. Improved Ultrasonic Nondestructive Testing of Pressure Vessels; Technical Report; Michigan University: Ann Arbor, MI, USA, 1976. © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).