sensors Article

Transducer Development and Characterization for Underwater Acoustic Neutrino Detection Calibration María Saldaña *, Carlos D. Llorens, Ivan Felis, Juan Antonio Martínez-Mora and Miguel Ardid * Institut d’Investigació per a la Gestió Integrada de les Zones Costaneres (IGIC), Universitat Politècnica de València (UPV), 46730 Gandia, Spain; [email protected] (C.D.L.); [email protected] (I.F.); [email protected] (J.A.M.-M.) * Correspondence: [email protected] (M.S.); [email protected] (M.A.); Tel.: +34-96-3877-000 (ext. 43681) (M.S.) Academic Editor: Stefano Mariani Received: 16 March 2016; Accepted: 26 July 2016; Published: 2 August 2016

Abstract: A short bipolar pressure pulse with “pancake” directivity is produced and propagated when an Ultra-High Energy (UHE) neutrino interacts with a nucleus in water. Nowadays, acoustic sensor networks are being deployed in deep seas to detect this phenomenon as a first step toward building a neutrino telescope. In order to study the feasibility of the method, it is critical to have a calibrator that is able to mimic the neutrino signature. In previous works the possibility of using the acoustic parametric technique for this aim was proven. In this study, the array is operated at a high frequency and, by means of the parametric effect, the emission of the low-frequency acoustic bipolar pulse is generated mimicking the UHE neutrino acoustic pulse. To this end, the development of the transducer to be used in the parametric array is described in all its phases. The transducer design process, the characterization tests for the bare piezoelectric ceramic, and the addition of backing and matching layers are presented. The efficiencies and directivity patterns obtained for both primary and parametric beams confirm that the design of the proposed calibrator meets all the requirements for the emitter. Keywords: acoustic calibrator; piezo-ceramic tube transducers; Ultra-High Energy neutrinos; acoustic detection; underwater neutrino telescopes; parametric technique

1. Introduction Astrophysical neutrino detection is based on Cherenkov light measurement induced by secondary leptons, like muons, which are produced by neutrino interactions with matter while passing across the Earth. Ongoing searches for cosmic ray neutrinos endeavor to detect muons moving upward from the Cherenkov light in either ice or water. The IceCube telescope, located at the South Pole, has recently discovered high-energy astrophysical neutrinos [1] boosting the neutrino astronomy field. Besides IceCube, the existing neutrino telescopes currently in use are the NT200+ in Lake Baikal [2], ANTARES [3] and a new optical-based deep-sea neutrino telescope under construction, the KM3NeT telescope [4], which will have a volume of several cubic kilometres. Ultra-High Energy (UHE) neutrinos (~1020 eV) have been of great interest for the study of ultra-high energy cosmic ray sources and to test fundamental physics due to being high-energy, stable and weakly interacting elementary particles. Besides Cherenkov light, radio or acoustic wave techniques were proposed to detect the neutrino-induced energy deposition in water, ice or salt. These techniques, with much longer attenuation lengths, allow for very large target volumes utilizing either large ice fields or dry salt domes for radio or ice fields, and the oceans for acoustic detection. Acoustic detection of UHE neutrinos is based on the thermo-acoustic effect [5]. When an UHE neutrino interacts with a nucleus in water, its energy is released in a volume of about a centimetre in radius and several meters in length. This phenomenon induces a local heating in a very short period Sensors 2016, 16, 1210; doi:10.3390/s16081210

www.mdpi.com/journal/sensors

Sensors 2016, 16, 1210

2 of 12

of time leading to a short pressure pulse signal with bipolar shape in time and a very directive pattern (pancake-like), being emitted mainly in the perpendicular plane of the shower axis [6]. Over the last few decades there have been a number of experiments looking into acoustic neutrino detection in both water and ice media, such as AMADEUS [7], SPATS [8], ACORNE [9] and SAUND [10]. The detection technique is still under study and could be implemented in a new optical neutrino telescope KM3NeT [11]. The acoustic detection would allow the combination of those two neutrino detection techniques for a hybrid underwater neutrino telescope, especially considering that the optical neutrino detection technique needs acoustic sensors to monitor the position of the optical sensors [12]. Therefore, an emitter able to imitate the bipolar pulse signal generated by the neutrino in water will be extremely useful. For this purpose, an acoustic array calibrator is being designed. The emissions from the emitter calibrator, which will be time controlled, will allow the sensors to be trained for neutrino detection. Furthermore, it will improve the classification and identification of the acoustic neutrino signals, telling them apart from noise or other transient background signals [13]. The acoustic calibrator will be deployed in the vicinity of the detector with a distance between the emitter calibrator and the detector of 0.5 to 3.5 km. The furthest distances would occur when the calibrator is operated from the sea surface, and the closest when operated from the seabed. Consequently, the acoustic calibrator needs to be quite powerful in order to reach the detector with enough pressure to be detected and recognized. The objective of the acoustic array calibrator is the emission of bipolar pulse signals with similar characteristics to the signal produced by a 1020 eV neutrino interacting in water at a distance of 1 km, which would be detected with an amplitude of about 10 mPa in the low-ultrasonic frequency range (maximum amplitude between 5 kHz and 20 kHz) with an opening angle of about 1˝ . The proposed design for the acoustic calibrator is a compact array system composed of piezo-ceramic tube transducers emitting in the radial direction. The emission is amplified and controlled by specific electronics adapted to them. The emission of the low-frequency (tens of kHz) acoustic bipolar pulse is generated by using the parametric emission technique at a high frequency (hundreds of kHz). 2. Compact Array Calibrator Based on the Parametric Acoustic Source Technique The use of the parametric acoustic source technique to generate neutrino-like signals allows the generation of low-frequency signals with narrow directivity, which is essential in order to obtain bipolar signals with the required “pancake” directivity. This technique was already validated in previous studies using cylindrical transducers [14]. The acoustic parametric effect occurs when two intense monochromatic beams with two close frequencies travel together through the medium. According to the linear theory, and as a consequence of the principle of superposition, the resulting sound field is composed only of the initial frequencies. However, due to the non-linearity of the medium, the interaction of the two frequency beams of finite amplitude generates a set of secondary frequency beams, such as the sum and difference of the primary frequency signals. This process of non-linear generation of new frequencies is limited to a certain distance from the transducer, called the array length, given by the distance of interaction or absorption length. This may be considered as a set of virtual acoustic sources (array) contained along the length of interaction; the source seems to be shaded exponentially as the distance increases from the transmitter. The secondary beam is produced with a directivity pattern similar to the one of the primary beams, which offers the advantage of generating a low-frequency directive beam. The parametric technique for the emission of the low-frequency signal previously cited allows us to design a compact array consisting of fewer units with respect to classical solutions. Moreover, this technique will reduce costs and facilitate the deployment and operation of the calibrator. The proposed array calibrator under development will be constituted of three to five elements of the piezo tube ceramic structured in the same axis-line, with a distance separation (d) between 10 cm and 20 cm, in order to obtain an opening angle of about 1˝ with phased emission of the array components.

Sensors 2016, 16, 1210 Sensors 2016, 16, 1210

piezo 2016, tube16, ceramic Sensors 1210

3 of 12 3 of 12

structured in the same axis-line, with a distance separation (d) between 10 3 ofcm 12 and piezo 20 cm, in ceramic order to obtain an opening angle ofwith about 1° with phased (d) emission tube structured in the same axis-line, a distance separation betweenof10the cm array and 20 cm, in transducer order to obtain anwill opening of about 1° with emission of the array components. The array emit angle in radial direction andphased produce a bipolar pulse due to The transducer array will emit array in radial direction anddirection produceand a bipolar pulse due to thedue interaction components.of The transducer emit in radial produce a bipolar pulse to the interaction parametric signalswill generated by each array element. Figure 1 shows a schematic of parametric signals generatedsignals by each array element. Figure 1 shows a schematic the interaction of parametric generated by each array element. Figure 1 shows adiagram schematicof the diagram of the configuration and emission. configuration emission. and emission. diagram ofand the configuration

Figure 1. Schematic diagram of the transducer array configuration with 3 elements.

Figure 1. 1. Schematic Schematic diagram diagram of of the the transducer transducer array array configuration configuration with Figure with 33 elements. elements.

In order to increase the functionality of the calibrator and facilitate the calibration under to increase the functionality of the calibrator and in facilitate calibration In ordercircumstances, different the proposed calibrator is designed to work differentthe operation modes.under the proposed proposed calibrator isfrequency designedrange to work work in different different operation modes. modes. The three operation modes are: (1) atcalibrator linear lowis by emitting long operation non-directive different circumstances, the designed to in signals (easy to generate and easy detect); (2)low at high frequency range by emitting emitting long The three modes are: (1)(1) attolinear low frequency rangerange by emitting long non-directive signals threeoperation operation modes are: at linear frequency by longparametric non-directive directive signals (signal processing techniques could be used to simplify detection of these signals); (easy to generate and easy to detect); (2) at high frequency range by emitting long parametric directive signals (easy to generate and easy to detect); (2) at high frequency range by emitting long parametric and(signal (3) at high frequencytechniques range by emitting theused transient and directive parametric bipolar (the (3) at signals processing could be tobe simplify of these signals); directive signals (signal processing techniques could used todetection simplify detection ofsignal theseand signals); most challenging case). These operation modes will enable the array transmitter to achieve different high(3) frequency range by range emitting the transient and directive parametric bipolarbipolar signal signal (the most and at high frequency by emitting the transient and directive parametric (the goals, such as training and tuning the acoustic detector, cross-checking the detector hydrophones, challenging case).case). These operation modes will enable the most challenging These operation modes will enable thearray arraytransmitter transmitterto to achieve achieve different and other marine applications. The proposed calibrator design and development was divided into cross-checking hydrophones, goals, such as training and tuning the acoustic detector, the detector hydrophones, three phases. In the first stage, the study and selection of transducers was done. Once the marine applications. development divided and other The proposed calibrator design and was transducers were selected, their characterization was performed and measures with the single bare into phases. firstfirst stage, theneutrino-like study and selection of transducers done.was Once the threetransducer phases.Inemitting Inthethe stage, the study and selection of transducers done. Once the parametric signals were realized. In was the second phase, thetransducers study and selection ofselected, backing and matching materials were carriedand out for optimizing were selected, their characterization waslayer performed measures with the single bare transducer transducers were their characterization was and performed measures withthe theceramic single bare emission, and for holding and isolating the ceramics as well. As a result, a final prototype transducer emitting parametric neutrino-like signals were realized. In the second phase, the study and selection transducer emitting parametric neutrino-like signals were realized. In the second phase, the study (backedand and moulded) convenient for this application was selected. Figure 2for shows a diagram of the of backing materials were carried out for optimizing ceramic emission, and for and selection ofmatching backing layer and matching layer materials were carried outthe optimizing the ceramic transducer design with the backing and matching layer configuration. This paper is focused on these and holding and theand ceramics as well. As a result, a final transducer (backed emission, andisolating for holding isolating the ceramics as well. As a prototype result, a final prototype transducer two phases. For future research, a third phase is planned and it will be focused on the design of the moulded) convenient this application selected.was Figure 2 shows a diagram the transducer (backed and moulded)for convenient for this was application selected. Figure 2 showsof a diagram of the complete array system composed of a few units of the developed transducer. In this future phase, design withdesign the backing andbacking matching layer configuration. This paper isThis focused on these twoon phases. transducer with the and matching layer configuration. paper is focused these electronics will be adapted to the array emitters in order to amplify the power of the signal emission, For future research, a third phase is planned and it will be focused on the design of the complete two phases. For future research, a third phase is planned and it will be focused on the design of the as well as to include all the functionalities for the control and operation of the calibrator. Once array system composed ofdistance a few units theatests developed transducer. In this future phase, be complete arraylong system composed of fewwill units of the developed transducer. In electronics this future will phase, completed, and inofsitu be performed in the detector vicinity.

adapted to the emitters order to emitters amplify the power the signal as well as to include electronics willarray be adapted tointhe array in order toof amplify the emission, power of the signal emission, all the functionalities for the control and operation of the calibrator. Once completed, long distance as well as to include all the functionalities for the control and operation of the calibrator. Once and in situ tests be performed in tests the detector vicinity. in the detector vicinity. completed, long will distance and in situ will be performed

Figure 2. Drawing of the transducer design component, ceramic, backing and matching layer.

Figure Figure 2. 2. Drawing Drawing of of the the transducer transducer design design component, component, ceramic, ceramic, backing backing and and matching matching layer. layer.

Sensors 2016,16, 16,1210 1210 Sensors Sensors2016,

of 12 444of of12 12

TransducerSelection Selectionand andCharacterization Characterization 3.3.Transducer 3. Transducer Selection and Characterization Two commercial commercial piezo-ceramic piezo-ceramic tube tube transducers transducers were were selected selected as as emitter emitter candidates candidates of of the the Two Two commercial piezo-ceramic tube transducers were selected as emitter candidates of the array calibrator. The selection was made in terms of resonance frequencies, power emission, array calibrator. The selection was made in terms of resonance frequencies, power emission, array calibrator. The selection was made in terms of resonance frequencies, power emission, dimensions, and and costs. costs. Both Both transducers transducers permit permit high-frequency high-frequency signal signal emissions emissions with with high high power, power, dimensions, dimensions, and costs.power Both transducers permit high-frequency signal with high power, exhibiting reasonable power levels at at low low frequency frequency as well. well. Theemissions first candidate candidate was the exhibiting reasonable levels as The first was the exhibiting reasonable power levels at low frequency as well. The first candidate was the piezo-ceramic piezo-ceramic UCE-534541, hereinafter referred to as the large tube. Its dimensions are: outer piezo-ceramic UCE-534541, hereinafter referred to as the large tube. Its dimensions are: outer UCE-534541, hereinafter referred to theand large tube. 4.1 Its dimensions are:tube outer diameter, 5.3 cm; diameter, 5.3 cm; cm; inner inner diameter, diameter, 4.5ascm; cm; and height, 4.1 cm. cm. The The large large tube primary resonance diameter, 5.3 4.5 height, primary resonance inner diameter, 4.5 cm; and height, 4.1 impedance cm. The large primary resonance frequency is around frequency around 490kHz kHzwith withaareal real impedance of99tube Ω,with with secondary resonance frequency at frequency isisaround 490 of Ω, aasecondary resonance frequency at 490 kHz with a real impedance of 9 Ω, with a secondary resonance frequency at low frequency, lowfrequency, frequency,around around35 35kHz. kHz.The Thesecond secondcandidate candidateisisthe thepiezo-ceramic piezo-ceramicUCE-343020, UCE-343020,hereinafter hereinafter low around to 35 kHz. second candidate is theare: piezo-ceramic UCE-343020, hereinafter referred as referred to as as the theThe small tube. Its Its dimensions dimensions are: outer diameter, diameter, 3.4 cm; cm; inner inner diameter, cm;to and referred small tube. outer 3.4 diameter, 33 cm; and the small tube. Itssmall dimensions are: outer diameter, 3.4 cm; is inner diameter, 3 cm; cm. height, cm. The small tube primary primary resonance frequency is located located around 890and kHzheight, with aa2real real height, 22 cm. The tube resonance frequency around 890 kHz with The small tube primary resonance frequency is located around 890 kHz with a real impedance of 6 Ω, impedanceof of66Ω, Ω,with withaasecondary secondaryfrequency frequencyresonance resonancearound around75 75kHz. kHz.Figure Figure33shows showsaapicture pictureof of impedance withceramics a secondary resonance around 75 of kHz. 3 shows a picture of both ceramics and both ceramics andfrequency Figure44shows shows theadmittance admittance ofthe theFigure largetube, tube, where theresonance resonance frequencies both and Figure the large where the frequencies Figure 4 shows the admittance of the large tube, where the resonance frequencies are appreciated. are appreciated. are appreciated.

(a) (a)

(b) (b)

(c) (c)

Figure (a) Piezo-ceramic large tube; (b) Piezo-ceramic small tube; (c) Both piezo-ceramics. Figure Figure3.3. 3.(a) (a)Piezo-ceramic Piezo-ceramiclarge largetube; tube;(b) (b)Piezo-ceramic Piezo-ceramicsmall smalltube; tube;(c) (c)Both Bothpiezo-ceramics. piezo-ceramics.

Figure4.4. 4.Admittance Admittanceof ofthe thelarge largetube. tube. Figure Admittance of the large tube. Figure

3.1. Transmitting VoltageResponse Responseand andDirectivity Directivity 3.1. 3.1.Transmitting TransmittingVoltage The piezo-ceramics were characterized both at high and low resonance frequencies in water The Thepiezo-ceramics piezo-ceramicswere werecharacterized characterizedboth bothat athigh highand andlow lowresonance resonancefrequencies frequenciesin inaaawater water 3 3 3 tank of87.5 87.5×׈113 113 56.5 cm with fresh water atatthe the laboratory. The calibration of the ceramics was tank tankof 113×׈56.5 56.5cm cmwith withfresh freshwater waterat thelaboratory. laboratory.The Thecalibration calibrationof ofthe theceramics ceramicswas was performed interms termsof ofthe theTransmitting Transmitting Voltage Response (TVR), which the ratio ratio of the pressure pressure performed of the Transmitting Voltage Response (TVR), which isisratio the the performed in in Voltage Response (TVR), which is the of theof pressure signal signal emitted to the the voltage, appliedusing voltage, using tone burstsfrequencies. at different differentThe frequencies. The emission emission signal emitted to applied voltage, bursts at frequencies. The emitted to the applied toneusing burststone at different emission directivity with directivity with respectto toplane theequatorial equatorial planeof ofthe the cylinder wasalso also evaluated. Theused hydrophones directivity respect the plane cylinder was The hydrophones respect to with the equatorial of the cylinder was also evaluated. Theevaluated. hydrophones for these used for these these measurements measurements are omnidirectional, omnidirectional, particularly, model RESON-TC4038 RESON-TC4038 for the therange high used for are particularly, model for high measurements are omnidirectional, particularly, model RESON-TC4038 for the high frequency frequency range and the model RESON-TC4034 for the low frequency range. Figure 5 shows the frequency rangeRESON-TC4034 and the model for RESON-TC4034 for the low Figure frequency range. 5 shows the and the model the low frequency range. 5 shows theFigure characterization of characterization of sensitivities both ceramics. ceramics. The sensitivities sensitivities obtained at high high frequency frequency with resonance characterization of both The obtained both ceramics. The obtained at high frequency with at resonance frequencywith (FR ) resonance of 490 kHz frequency (FRRtube of490 490 kHz forthe the large tubetube and890 890kHz kHzfor for the small tube are shown in Figure 5a. frequency (F ))of tube and small tube shown Figure 5a. for the large andkHz 890 for kHz forlarge the small are shown inthe Figure 5a. Theare TVR of thein large tube is The TVR ofµPa/V thelarge large tube dB (re μPa/V ata11directivity m)at atFFRR==of 490 kHz withHalf directivity ofFull FullWidth Width The of the isis159 μPa/V m) 490 kHz with aadirectivity of 159 TVR dB (re at 1tube m) at F159 490(re kHz withat Full Width Maximum (FWHM) of R =dB Half Maximum (FWHM) of 10° 10° (Figure (Figure 6).varies At low low frequencies, the TVR varies between 132 and Half Maximum (FWHM) of 6). At frequencies, the TVR varies between 132 10˝ (Figure 6). At low frequencies, the TVR between 132 and 140 dB (re µPa/V at 1 m). Onand the 140 dB (re μPa/V at 1 m). On the other hand, the TVR of the small tube is 162 dB (re μPa/V at 1 m) at 140 dB hand, (re μPa/V at 1 m). On small the other of the small tube dB (re μPa/V m) at other the TVR of the tubehand, is 162the dBTVR (re µPa/V at 1 m) atisFR162 = 890 kHz withat a 1FWHM ˝ 890kHz kHz with FWHM directivity oflow 14°frequencies (Figure6). 6).The TheTVR TVR atlow low132 frequencies varies between FFdirectivity RR==890 aaFWHM 14° (Figure at frequencies varies between ofwith 14 (Figure 6).directivity The TVR atof varies between and 143 dB (re µPa/V at 132 and 143 dB (re μPa/V at 1 m). 132 and 143 dB (re μPa/V at 1 m). 1 m).

Sensors 2016, 16, 1210 Sensors 2016, 16, 1210 Sensors Sensors2016, 2016,16, 16,1210 1210

165 165 165 160 160 160 155 155 155 150 150 150

Large Tube Large LargeTube Tube Small Tube Small Tube Small Tube 145 145 145 0.8 0.85 0.9 0.95 11 1.05 1.1 1.15 0.8 0.8 0.85 0.85 0.9 0.9 0.95 0.95 1 1.05 1.05 1.1 1.1 1.15 1.15 Frequency / Resonance Frequency Frequency Frequency/ /Resonance ResonanceFrequency Frequency

TVR(dB (dBre reuPa/V uPa/V@1m) @1m) TVR (dB re uPa/V @1m) TVR

TVR(dB (dBref refuPa/V uPa/V@1m) @1m) TVR (dB ref uPa/V @1m) TVR

5 of 12 55 of 12 5ofof12 12

142 142 142 140 140 140 138 138 138 136 136 136 134 134 134 Large Tube Large LargeTube Tube 132 132 132 Small Tube Small Tube Small Tube 130 130 130 10 20 30 40 50 60 70 80 90 100 10 10 20 20 30 30 40 40 50 50 60 60 70 70 80 80 90 90100 100 Frequency (kHz) Frequency Frequency(kHz) (kHz)

(a) (a) (a)

(b) (b) (b)

Figure 5. of the ceramics (a) TVR of large tube at kHz (blue dots) and for Figure 5.5.Characterization Characterization the ceramics (a) TVR of large tube atatFFFRRR===490 490 dots) and for Figure5. Characterizationof ofthe theceramics ceramics(a) (a)TVR TVRof oflarge largetube tubeat 490kHz kHz(blue (bluedots) dots)and andfor for Figure Characterization of FR = 490 kHz (blue small tube at F R = 890 kHz (red dots); (b) TVR from 10 to 100 kHz. small tube at F R = 890 kHz (red dots); (b) TVR from 10 to 100 kHz. smalltube tubeatatFFR== 890 890 kHz kHz (red (red dots); dots);(b) (b)TVR TVRfrom from10 10to to100 100kHz. kHz. small

Normalizedamplitude amplitude Normalized amplitude Normalized

R

111 0.8 0.8 0.8 0.6 0.6 0.6 0.4 0.4 0.4 0.2 0.2 0.2 000

Large Large Tube LargeTube Tube Small Small Tube SmallTube Tube

-20 -20 -20

-10 -10 -10

000 10 10 10 Angle (º) Angle Angle(º) (º)

20 20 20

Figure 6. 6. Directivity Directivity at at 490 490 kHz kHz for for the the large large tube tube (blue) (blue) and Figure and at at 890 890 kHz kHz for for small small tube tube (red). (red). Figure Figure6.6.Directivity Directivityatat490 490kHz kHzfor forthe thelarge largetube tube(blue) (blue)and andatat890 890kHz kHzfor forsmall smalltube tube(red). (red).

3.2. 3.2. Backing 3.2.Backing Backing are able to absorb the energy produced from the back ofback the active element Some backing materials are able to the produced from the of active Some backing materials are able to absorb the energy produced from the back of the active Somebacking backingmaterials materials are able toabsorb absorb theenergy energy produced from the back ofthe the active (ceramic) and reflect part of it forward, providing more acoustic power in emission [15]. In addition, element (ceramic) and reflect part of it forward, providing more acoustic power in emission [15]. element In element(ceramic) (ceramic)and andreflect reflectpart partof ofititforward, forward,providing providingmore moreacoustic acousticpower powerin inemission emission[15]. [15].In In the backing element of the transducer will constitute the support of the complete future array calibrator, addition, the backing element of the transducer will constitute the support of the complete future addition, the backing element of the transducer will constitute the support of the complete future addition, the backing element of the transducer will constitute the support of the complete future i.e., playing an important role inimportant the mechanical of the array (Figure 1).array Since(Figure all elements of array calibrator, i.e., an role the design of 1). array calibrator, i.e., playing an important role in the mechanical design of the array (Figure 1). Since array calibrator, i.e.,playing playing an important rolein indesign themechanical mechanical design ofthe the array (Figure 1).Since Since the array are identical and emit synchronously it is expected that the coupling effect of the backing all elements of the array are identical and emit synchronously it is expected that the coupling effect all allelements elementsof ofthe thearray arrayare areidentical identicaland andemit emitsynchronously synchronouslyititisisexpected expectedthat thatthe thecoupling couplingeffect effect of the backing will be negligible. will be negligible. of ofthe thebacking backingwill willbe benegligible. negligible. Different materials and thicknesses were studied to in of were studied in in order to find the the best option in terms of TVR Different materials and thicknesses were studied in order to find the best option in terms of Differentmaterials materialsand andthicknesses thicknesses were studied inorder order tofind find thebest bestoption option interms terms of TVR and acoustic impedance. The materials studied as backings were epoxy and aluminium. and acoustic impedance. The materials studied as backings were epoxy and aluminium. Figure 7 TVR and acoustic impedance. The materials studied as backings were epoxy and aluminium. TVR and acoustic impedance. The materials studied as backings were epoxy and aluminium. Figure 7 shows the ceramics implemented with the backing elements studied. shows the ceramics implemented with the backing elements studied. Figure Figure77shows showsthe theceramics ceramicsimplemented implementedwith withthe thebacking backingelements elementsstudied. studied.

Figure 7. with aluminium backings of different thicknesses. Figure 7.7.Piezo-ceramics Piezo-ceramics backings of different thicknesses. Figure7. Piezo-ceramicswith withaluminium aluminiumbackings backingsof ofdifferent differentthicknesses. thicknesses. Figure Piezo-ceramics with aluminium

According According to the results obtained on the characterization of all backings for both ceramic types, Accordingto tothe theresults resultsobtained obtainedon onthe thecharacterization characterizationof ofall allbackings backingsfor forboth bothceramic ceramictypes, types, According to the results obtained on the characterization of all backings for both ceramic types, the set of aluminium backing completely filling the tube ceramic yielded the best results. For the both theset setof ofaluminium aluminiumbacking backingcompletely completelyfilling fillingthe thetube tubeceramic ceramicyielded yieldedthe thebest bestresults. results.For Forboth both the set of the aluminium backing completely filling the tube ceramic yielded the best results. ForIn both ceramics filled aluminium backing produced a flatter frequency response from the TVR. ceramics the filled aluminium backing produced a flatter frequency response from the TVR. In the ceramics the filled aluminium backing produced a flatter frequency response from the TVR. Inthe the case of the backed large tube a 3 dB increment on the TVR, a resonance frequency shifting case to case of of the the backed backed large large tube tube aa 33 dB dB increment increment on on the the TVR, TVR, aa resonance resonance frequency frequency shifting shifting to to 510 510 kHz, and an increment up to 30 Ω (real) in the impedance was observed. Regarding the small 510kHz, kHz,and andan anincrement incrementup upto to30 30Ω Ω(real) (real)in inthe theimpedance impedancewas wasobserved. observed.Regarding Regardingthe thesmall small

Sensors 2016, 16, 1210

6 of 12

ceramics the filled aluminium backing produced a flatter frequency response from the TVR. In the case of the backed large tube a 3 dB increment on the TVR, a resonance frequency shifting to 510 kHz, and an increment up to 30 Ω (real) in the impedance was observed. Regarding the small tube, an increase of 5 dB on the TVR, a shift on the resonance frequency to 950 kHz, and an impedance increment to 50 Ω (real) was appreciated. Moreover, there was not any appreciable change in the Sensors 2016, 16, 1210 6 of 12 directivity patterns of both tubes. tube, an increase of 5 dB on the TVR, a shift on the resonance frequency to 950 kHz, and an

3.3. Matching Layer impedance increment to 50 Ω (real) was appreciated. Moreover, there was not any appreciable

change in the directivity patterns of both tubes. Two interesting materials for moulding underwater transducers were acoustically studied in order to determine the effect on the signal emitted. Ensuring protection, isolation and holding, 3.3. Matching Layer and matching the impedance of the piezo-ceramic to the impedance media [16] are the reasons for interesting materials for moulding underwater acoustically studied in the mouldingTwo the bare ceramic. The matching materials under transducers study werewere the resin RoyaPox 511 and order to determine the effect on the signal emitted. Ensuring protection, isolation and holding, and Polyurethane EL241F. Both materials exhibit interesting acoustic properties for matching the ceramic matching the impedance of the piezo-ceramic to the impedance media [16] are the reasons for impedance to the media due to their specific acoustic impedance (Z), density (ρ) and sound speed (cL ). moulding the bare ceramic. The matching materials under study were the resin RoyaPox 511 and the TherePolyurethane is maximum transmission at theexhibit layer thickness (2n ´properties 1)λ/4. Since for the first mode, λ/4, EL241F. Both materials interesting of acoustic for matching the ceramic the obtained thickness values were too thin for an optimum covering, the matching layer thicknesses impedance to the media due to their specific acoustic impedance (Z), density (ρ) and sound speed were (c set accomplish the second mode at with 3λ/4of of(2n the−wavelength at the ceramic L).to There is maximum transmission the thickness layer thickness 1)λ/4. Since emitted for the first mode, resonance frequency. Such a selection offers of sufficient thickness to facilitate λ/4, the obtained thickness values were toothe thinadvantage for an optimum covering, the matching layer the thicknesses were setmaximum to accomplish the second mode with of the wavelength emitted covering and obtaining length transmission. Thethickness matching3λ/4 layer mould used was a cylinder at the ceramic resonance frequency. Such a selection offers the advantage of sufficient thickness to with with the thickness space necessary to cover the ceramic, controlled by a precise metric rule facilitate the covering and obtaining maximum length transmission. The matching layer mould used an estimated uncertainty of ˘0.1 mm.The acoustic properties of the materials and the thickness used was a cylinder with the thickness space necessary to cover the ceramic, controlled by a precise metric are summarized in Table 1. In order to test both materials with the same ceramic, two large tubes rule with an estimated uncertainty of ±0.1 mm.The acoustic properties of the materials and the with aluminium backing were moulded with each material. Figure 8 shows the ceramic covered with thickness used are summarized in Table 1. In order to test both materials with the same ceramic, two (a) RoyaPox 511with andaluminium (b) with polyurethane EL241F. The inFigure terms8ofshows TVRthe andceramic directivity large tubes backing were moulded with best each results material. were covered obtainedwith with polyurethane. Afterwards, a small tube withThe aluminium backing was moulded (a)the RoyaPox 511 and (b) with polyurethane EL241F. best results in terms of TVR with polyurethane, shown in Figure 8c. polyurethane. Afterwards, a small tube with aluminium and directivity as were obtained with the backing was moulded with polyurethane, as shown in Figure 8c. Table 1. Acoustic properties of the matching layer (ML) materials RoyaPox 511 and EL214F, Table 1. Acoustic properties the matching layer (ML) materials RoyaPoxwavelength 511 and EL214F, and thickness (mm) obtained forof accomplishing λ/4 or 3λ/4 of the emitted at theand ceramic thickness (mm) obtained for accomplishing λ/4 or 3λ/4 of the emitted wavelength at the ceramic resonance frequency. resonance frequency.

ACOUSTIC ACOUSTIC PROPERTIES PROPERTIES (m/s) ccLL(m/s) Z (MRayls) Z (MRayls) ρ (kg/m3 ) ρ (kg/m3) Frequency (kHz) Frequency (kHz) λ/4 Thickness (mm) λ/4 Thickness (mm) 3λ/4 Thickness (mm) 3λ/4 Thickness (mm)

(a)

Large LargeTube Tube M.L RoyaPox 511

M.L RoyaPox 511 2602 2602 2.82 2.82 1050 1050 480 480 1.3 1.3 4.0 4.0

(b)

Large Tube Large Tube M.L EL214F

M.L EL214F 1600 1600 1.857 1.857 1160.7 1160.7 510 510 0.78 0.78 2.4 2.4

Small Tube Small Tube M.L EL214F

M.L EL214F 1600 1600 1.857 1.857 1160.7 1160.7 960 960 0.42 0.42 1.3 1.3

(c)

Figure 8. (a) Piezo-ceramic large tube with aluminium backing and RoyaPox 511 moulding; (b)

Figure 8. (a) Piezo-ceramic large tube with aluminium backing and RoyaPox 511 moulding; Piezo-ceramic large tube with aluminium backing and polyurethane EL241F moulding; (c) (b) Piezo-ceramic large tube with aluminium backing and polyurethane EL241F moulding; Piezo-ceramic small tube with aluminium backing and polyurethane EL241F moulding. (c) Piezo-ceramic small tube with aluminium backing and polyurethane EL241F moulding.

The ceramic moulded with the RoyaPox 511 resin showed the same TVR level on the resonance frequency than before moulding. The resonance frequency shifted to 450 kHz, but the impedance and directivity were very similar to that obtained with the bare ceramic. For the case of the large

Sensors 2016, 16, 1210

7 of 12

170 166 162 158 154 150 400 425 450 475 500 525 550 575 600 Frequency (kHz)

(a)

TVR (dB ref uPa/V @1m)

TVR (dB ref uPa/V @1m)

The ceramic moulded with the RoyaPox 511 resin showed the same TVR level on the resonance frequency before moulding. The resonance frequency shifted to 450 kHz, but the impedance7 and Sensors 2016,than 16, 1210 of 12 directivity were very similar to that obtained with the bare ceramic. For the case of the large tube moulded with the EL241F polyurethane, there was a TVR of 5 gain dB approximately in the resonance tube moulded with the EL241F polyurethane, there wasgain a TVR of 5 dB approximately in the frequency compared to before moulding. The resonance frequency shifted to 495 kHz with a sensitivity resonance frequency compared to before moulding. The resonance frequency shifted to 495 kHz of 169adB (re µPa/V m).(re The TVR curve forThe thisTVR casecurve is shown in Figure The impedance at The the with sensitivity of at 1691 dB μPa/V at 1 m). for this case is9a. shown in Figure 9a. resonance frequency was 12 Ω (real), that is, it was reduced as compared to itself only backed, but being impedance at the resonance frequency was 12 Ω (real), that is, it was reduced as compared to itself still than With thebare. directivity, the pattern not show the anypattern significant variation. onlyhigher backed, butbare. being stillrespect higher to than With respect to thedid directivity, did not show Figure 9b shows the TVR of the small9btube moulded Themoulded graph is practically any significant variation. Figure shows the with TVR the of polyurethane. the small tube with the the same as before for the 890 the kHz–1 MHz range. However, was also a peak resonance polyurethane. Themoulding graph is practically same as before moulding there for the 890 kHz–1 MHz range. at higher frequencies, fora 1165 with 169 dB (refrequencies, µPa/V at 1for m)1165 with realwith impedance ofμPa/V 20 Ω. However, there was also peak kHz resonance at higher kHz 169 dB (re Directivity did not vary with to the bare one. at 1 m) withpattern real impedance of 20 Ω. respect Directivity pattern did not vary with respect to the bare one. 170 165 160 155 900

1000

1100

1200

1300

Frequency (kHz)

(b)

Figure 9. 9. (a) backing and moulded with polyurethane EL241F; (b) Figure (a) TVR TVRof oflarge largetube tubewith withaluminium aluminium backing and moulded with polyurethane EL241F; TVR of small tube with aluminium backing and moulded with polyurethane EL241F. (b) TVR of small tube with aluminium backing and moulded with polyurethane EL241F.

To sum up, the design with the aluminium backing and the moulding with polyurethane To sum up, the design with the aluminium backing and the moulding with polyurethane EL241F EL241F produced an increase on the TVR at resonance of 9 dB for the large tube and of 7 dB for the produced an increase on the TVR at resonance of 9 dB for the large tube and of 7 dB for the small tube. small tube. For both ceramics, the design did not change the directivity pattern of the ceramics. For both ceramics, the design did not change the directivity pattern of the ceramics. 4. Studies on Parametric Emission 4. Studies on Parametric Emission The parametric acoustic source technique was evaluated by means of the low The parametric acoustic source technique was evaluated by means of the low frequency-parametric frequency-parametric generation of a sine sweep signal using the cross-correlation with the expected generation of a sine sweep signal using the cross-correlation with the expected signal. Following this signal. Following this methodology, it is feasible to recognize the parametric signal, since the methodology, it is feasible to recognize the parametric signal, since the correlation produces a clear correlation produces a clear narrow peak on the signal arrival time, which allows it to be narrow peak on the signal arrival time, which allows it to be distinguished from near echoes, as well distinguished from near echoes, as well as an increase in the signal to noise ratio [17]. Afterwards, as an increase in the signal to noise ratio [17]. Afterwards, the low frequency-parametric generation the low frequency-parametric generation of a bipolar shape pulse signal in time was studied. The of a bipolar shape pulse signal in time was studied. The generation of the portable signal for the generation of the portable signal for the bipolar pulse generation is explained in [14]. These portable bipolar pulse generation is explained in [14]. These portable signals for the parametric generation signals for the parametric generation were implemented with modulation at high frequency, e.g., at were implemented with modulation at high frequency, e.g., at the frequency resonance of the ceramic, the frequency resonance of the ceramic, either 495 kHz or 1165 kHz. either 495 kHz or 1165 kHz. In order to confirm the non-linear effect of the parametric signal received three studies in the In order to confirm the non-linear effect of the parametric signal received three studies in the different aforementioned stages were conducted. The first study aimed at comparing the amplitude different aforementioned stages were conducted. The first study aimed at comparing the amplitude of primary and secondary beams by starting from low amplitude emission and increasing it to of primary and secondary beams by starting from low amplitude emission and increasing it to demonstrate the non-linear effect. The second measurement analysed the secondary non-linear demonstrate the non-linear effect. The second measurement analysed the secondary non-linear beam beam generation in the medium by changing the distance between emitter and receiver. The last generation in the medium by changing the distance between emitter and receiver. The last study study compared the directivity patterns of both beams. compared the directivity patterns of both beams. The experiment was done in the same water tank as described in Section 3.1. The piezo-ceramic The experiment was done in the same water tank as described in Section 3.1. The piezo-ceramic was connected to a linear 55 dB (gain) RF amplifier ENI 1040L to feed the emitter and generate a was connected to a linear 55 dB (gain) RF amplifier ENI 1040L to feed the emitter and generate more powerful signal in order to achieve the non-linear parametric effect. The receiver was the a more powerful signal in order to achieve the non-linear parametric effect. The receiver was RESON-TC4034. This transducer is an omnidirectional, broad-band hydrophone with enough the RESON-TC4034. This transducer is an omnidirectional, broad-band hydrophone with enough sensitivity to detect the primary beam (high frequency) even more sensitive to low frequency, i.e., for the bipolar pulse detection. Additionally, it was connected to a charge amplifier CCA 1000 (Teledyne RESON) which amplifies the received signal, especially for low frequency signals.

Sensors 2016, 16, 1210

8 of 12

sensitivity to detect the primary beam (high frequency) even more sensitive to low frequency, i.e., for the bipolar pulse detection. Additionally, it was connected to a charge amplifier CCA 1000 (Teledyne RESON) which amplifies the received signal, especially for low frequency signals. Sensors 2016, 16, 1210

8 of 12

4.1. Parametric Sine Sweep Signal 4.1. Parametric Sine Sweep Signal The signal for the generation of the sine sweep signal was designed by modulating a sine sweep The signal generation the sine sweep signal was designed by modulating a sine sweep signal from 20 kHzfor tothe 50 kHz using aofmodulation frequency of 495 kHz. Figure 10a shows the emission signal from 20 kHz to 50 kHz using a modulation frequency of 495 kHz. Figure 10a shows the signal, where the modulation shape can be observed. The received signal was a mix of the primary emission signal, where the modulation shape can be observed. The received signal was a mix of the beam at 495 kHz and the secondary beam at low frequency produced by parametric effect. In order primary beam at 495 kHz and the secondary beam at low frequency produced by parametric effect. to distinguish the secondary beam, a band pass (5–80 kHz) filter was applied. Figure 10b shows In order to distinguish the secondary beam, a band pass (5–80 kHz) filter was applied. Figure 10b the received signal, primary (original) and secondary (filtered) beams. The analyses of the different shows the received signal, primary (original) and secondary (filtered) beams. The analyses of the parametric studies with this signal were obtained by taking the amplitude value of the correlation of different parametric studies with this signal were obtained by taking the amplitude value of the the received signal with the primary beam signal (high frequency) and with the expected secondary correlation of the received signal with the primary beam signal (high frequency) and with the beam signalsecondary (low frequency). Figure 10cfrequency). shows theFigure correlation of thethe original received signal with expected beam signal (low 10c shows correlation of the original thereceived expected secondary beam signal. There isbeam a clear correlation the arrivalpeak timeatofthe the signal with the expected secondary signal. There is peak a clearatcorrelation received signal. arrival time of the received signal.

Figure 10. (a) Emitted signal for sine sweep (20 kHz–50 kHz) generation; (b) received signal (blue Figure 10. (a) Emitted signal for sine sweep (20 kHz–50 kHz) generation; (b) received signal line) and the band-pass filtered signal (red line); (c) correlation of the original received signal with (blue line) and the band-pass filtered signal (red line); (c) correlation of the original received signal the expected secondary beam; (d) correlation amplitude of the received signal with primary beam with the expected secondary beam; (d) correlation amplitude of the received signal with primary (blue points) and secondary beam (red points) as a function of the signal emitted amplitude; (e) beam (blue points) and secondary beam (red points) as a function of the signal emitted amplitude; correlation amplitude of the received signal with both beams as a function of the distance between (e) correlation amplitude of the received signal with both beams as a function of the distance between emitter and receiver; and (f) directivity pattern of both beams. emitter and receiver; and (f) directivity pattern of both beams.

Figure 10d shows the correlation amplitude behaviour of the received signal, without filtering (original) and with the band-pass filtering (secondary beam) as a function of the input signal amplitude. The data fitting to parameterizations showed that the exponent for the secondary beam is much larger than that of the primary beam, which is evidence of the non-linear effect. The exponent value of the primary beam is 0.69 ± 0.03 and for the secondary beam is 3.70 ± 0.12. We expected an exponent close to one for the primary beam, but it is slightly lower due to saturation effects of the

Sensors 2016, 16, 1210

9 of 12

Figure 10d shows the correlation amplitude behaviour of the received signal, without filtering (original) and with the band-pass filtering (secondary beam) as a function of the input signal amplitude. The data fitting to parameterizations showed that the exponent for the secondary beam is much larger than that of the primary beam, which is evidence of the non-linear effect. The exponent value of the primary beam is 0.69 ˘ 0.03 and for the secondary beam is 3.70 ˘ 0.12. We expected an exponent close to one for the primary beam, but it is slightly lower due to saturation effects of the amplifier. On the other hand, for the secondary beam, the expectation for the exponent was two, but it is larger because of the cross-correlation method used. In fact if we were directly representing the amplitude of the signal of the secondary beam filtered, the exponent would have been 1.68 ˘ 0.15. However, when using the correlation method for the secondary beam, following the parametric theory, the received signal is correlated with the second time derivative of the envelope squared of the emitted signal. The fact that we use the square in this process should be why the exponent is almost twice the expected value. The evolution of both beams with the distance is shown in Figure 10e, clearly displaying the different behaviour between them. The ratio of amplitudes between the primary and secondary beam increases as a function of the distance, which is evidence of the secondary beam being generated in the medium. Parameterizations were fitted to the data obtaining a smaller exponent for the secondary beam, as expected. For the primary beam the exponent value is ´0.44 ˘ 0.04 while for the secondary beam is ´0.27 ˘ 0.06. The directivity patterns shown in Figure 10f are also evidence of the parametric effect for the secondary beam since the pattern for this beam is quite similar to that of the primary beam, and more directive than the ceramic directly fed at low frequency. For instance, the open angle (FWHM) for direct feeding in the frequency range, 20–50 kHz, was 80˝ approximately. 4.2. Parametric Bipolar Pulse Signal The emission signal for achieving the parametric generation of the bipolar-shape pulse was designed by using the portable signal at 495 kHz. According to theory, the signal is set so the shape of the secondary signal matches the second time derivative of the envelope squared of the primary signal in amplitude [18]. Figure 11a shows the emitted signal shape modulated. The received signal was a mix of the primary beam at 495 kHz and the secondary beam at low frequency produced by parametric effect. In order to distinguish the secondary beam, a band pass (5–80 kHz) filter was applied. Figure 11b shows the received signal, primary (original) and secondary (filtered) beams. The secondary beam (bipolar pulse) is multiplied by a factor of 30 in order to be visible together with the original received signal (non-filtered). As expected, the secondary beam has a bipolar pulse-shape. A peak after the bipolar pulse is appreciated. That is due to the signal response of the transducer, to what can affect to the shape of the signal tail and the generation of an oscillation after the last signal pulse. The analysis of the different parametric studies with this signal were obtained by taking the amplitude value of the signal received at high frequency (primary beam) and of the bipolar pulse signal (secondary beam). Figure 11c shows the amplitude behaviour of the received signal, without filtering (original) and with the band-pass filtering (the parametric bipolar signal) as a function of input signal amplitude. The fitting data to parameterizations showed that the exponent for the secondary beam is twice the exponent of the primary beam, representing the non-linear effect. The exponent value of the primary beam is 0.81 ˘ 0.03 and for the secondary beam is 1.72 ˘ 0.26. The exponent for the primary beam is slightly below one due to saturation effects of the amplifier. In Figure 11d the evolution of both beams with the distance can be observed. No conclusive results could be obtained from this due to the fluctuations observed in the measurements and on the detection of the bipolar secondary signal, since the exponent value for the primary beam is ´0.53 ˘ 0.01 and for the secondary beam is ´0.51 ˘ 0.05. The directivity pattern, shown in Figure 11e, is the best proof of the parametric effect for the secondary beam since both beams have similar directivities.

Sensors 2016, 16, 1210

10 of 12

Considering all these results, the parametric generation was satisfactory validated. The large tube with aluminium backing and moulded with polyurethane EL241F, as a matching layer, showed very Sensors 2016, 16, 1210 10 of 12 good characteristics for use as the transducer unit of the final array.

Figure 11. (a) Emitted signal for bipolar pulse generation; (b) received signal (blue line) and bipolar Figure 11. (a) Emitted signal for bipolar pulse generation; (b) received signal (blue line) and bipolar signal obtained after applying a band-pass filter (red line); (c) amplitude of the received (blue) and signal obtained after applying a band-pass filter (red line); (c) amplitude of the received (blue) and filtered signal (red) as function of input signal amplitude; (d) (d) amplitude of the received signal for for both filtered signal (red) as function of input signal amplitude; amplitude of the received signal beams as function of distance between emitter and receiver; and (e) directivity pattern of both beams. both beams as function of distance between emitter and receiver; and (e) directivity pattern of both beams.

5. Future Steps 5. Future Steps The work to follow will consist of adapting the electronics to the final transducer design to achieve The work toemission follow will of to adapting the final transducer designboard to the required power ableconsist enough producethe theelectronics parametrictosignals. The new electronic achieve the required power emission able enough to produce the parametric signals. The new will be based on the same philosophy as the previously designed electronic board [19], while adapted electronic board will be new based on the same philosophy The as the previously electronicthe to the particularities of the transducer and application. electronics willdesigned also incorporate board [19], while adapted to the particularitiescontrol, of the and newamplification transducer and application. functionalities for communication, configuration, of the signal thatThe were electronics will also incorporate the functionalities for communication, configuration, control, and already developed in previous studies [20]. amplification of the signal that were already developed in previous studies [20]. Afterwards, the transducer array will be built, which will be composed of multi-emitters placed Afterwards, the transducer array will be built, which will be composed of multi-emitters placed in the same line structure, which will be a long tube of aluminium that will also work as backing of the in the same line structure, which will be a long tube of aluminium that will also work as backing of ceramics. After the characterization and final tests over longer distances are completed, the calibrator the ceramics. After the characterization and final tests over longer distances are completed, the will either be used in sea campaigns for emissions from a vessel or incorporated in the KM3NeT calibrator will either be used in sea campaigns for emissions from a vessel or incorporated in the KM3NeT deep-sea neutrino telescope infrastructure, in order to perform the calibration tests for the acoustic telescope systems.

Sensors 2016, 16, 1210

11 of 12

deep-sea neutrino telescope infrastructure, in order to perform the calibration tests for the acoustic telescope systems. 6. Conclusions The results obtained on the characterization of the transducers designed (piezo-ceramics with backing and matching layer) indicate that the best solution for the multi-element array calibrator is the large tube with aluminium backing and moulded with 3λ/4 of polyurethane EL241F. An increment of 9 dB after backing and moulding was appreciated on the transducer, compared to the bare ceramic, with a final sensitivity of 169 dB (re µPa/V at 1 m) at 495 kHz. The parametric emission was validated for both parametric signals generated, namely the sine sweep and the bipolar pulse; the non-linear effect of the parametric signal was verified in the studies performed, which showed higher amplitude evolution and a narrower directivity pattern. Moreover, by using the parametric technique a signal that imitates the acoustic neutrino-like bipolar pulse was achieved. The future array of multi-elements emitting at linear or parametric modes with a high-pressure sound level will become a very versatile and powerful system, which could be used in other kinds of underwater acoustic applications, such as communication, localization, positioning and on-site sensor calibration. Acknowledgments: We acknowledge the financial support of Plan Estatal de Investigación, ref. FPA2015-65150-C3-2-P (MINECO/FEDER), and Consolider MultiDark CSD2009-00064 (MINECO) and of the Generalitat Valenciana, Grants ACOMP/2015/175 and PrometeoII/2014/079. Author Contributions: M.A. and J.A.M.-M. conceived and designed the experiments; M.S., C.D.L. and I.F. develop the prototype; M.S., C.D.L. and I.F performed the experiments; M.S, M.A. and J.A.M.-M. analyzed the data; M.S. and M.A. wrote the paper. Conflicts of Interest: The authors declare no conflict of interest.

References 1. 2.

3.

4.

5. 6.

7.

8.

9.

IceCube Collaboration. Evidence for high-energy extraterrestrial neutrinos at the IceCube detector. Science 2013, 342. [CrossRef] Aynutdinov, V.; Avrorin, A.; Balkanov, V.; Belolaptikov, I.; Budnev, N.; Danilchenko, I.; Domogatsky, G.; Doroshenko, A.; Dyachok, A.; Dzhilkibaev, Z.A.; et al. The Baikal neutrino experiment: Status, selected physics results, and perspectives. Nucl. Instrum. Methods Phys. Res. Sect. A 2008, 588, 99–106. [CrossRef] Ageron, M.; Aguilar, J.A.; Al Samarai, I.; Albert, A.; Ameli, F.; André, M.; Anghinolfi, M.; Anton, G.; Anvar, S.; Ardid, M.; et al. ANTARES: The first undersea neutrino telescope. Nucl. Instrum. Methods Phys. Res. Sect. A 2011, 656, 11–38. [CrossRef] Adrian-Martinez, S.; Ageron, M.; Aharonian, F.; Aiello, S.; Albert, A.; Ameli, F.; Anassontzis, E.; Andre, M.; Androulakis, G.; Anghinolfi, M.; et al. Letter of intent for KM3NeT 2.0. J. Phys. G: Nucl. Part. Phys. 2016, 43, 084001. [CrossRef] Askaryan, G.A. Hydrodynamical emission of tracks of ionising particles in stable liquids. Sov. J. At. Energy 1957, 3, 921–923. [CrossRef] Bevana, S.; Brownc, A.; Danaherb, S.; Perkinc, J.; Rhodesd, C.; Sloane, T.; Thompsonc, L.; Veledarc, O.; Watersa, D. The ACORNE Collaboration. Study of the acoustic signature of UHE neutrino interactions in water and ice. Nucl. Instrum. Methods Phys. Res. Sect. A 2009, 607, 398–411. [CrossRef] Aguilar, J.A.; Al Samarai, I.; Albert, A.; Anghinolfi, M.; Anton, G.; Anvar, S.; Ardid, M.; Jesus, A.A.; Astraatmadja, T.; Aubert, J.J.; et al. AMADEUS, The acoustic neutrino detection test system of the ANTARES deep-sea neutrino telescope. Nucl. Instrum. Methods Phys. Res. Sect. A 2011, 626, 128–143. [CrossRef] Abdou, Y.; Becker, K.H.; Berdermann, J.; Bissok, M.; Bohm, C.; Boeser, S.; Bothe, M.; Carson, M.; Descamps, F.; Fischer-Wolfarth, J.H.; et al. Design and performance of the South Pole acoustic test setup. Nucl. Instrum. Methods Phys. Res. Sect. A 2012, 683, 78–90. [CrossRef] Thompson, L.F.; Perkin, J.D. Acoustic detection of UHE neutrinos—The ACORNE project. J. Phys. Conf. Ser. 2008, 136, 042070. [CrossRef]

Sensors 2016, 16, 1210

10. 11.

12.

13. 14. 15. 16. 17.

18. 19. 20.

12 of 12

Vandenbroucke, J.; Gratta, G.; Lehtinen, N. Experimental study of acoustic ultra-high-energy neutrino detection. Astrophys. J. 2005, 621, 301–312. [CrossRef] Viola, S.; Ardid, M.; Bertin, V.; Lahmann, R.; Pellegrino, C.; Riccobene, G.; Saldaña, M.; Sapienza, P.; Simeone, F. Acoustic positioning system for KM3NeT. In Proceedings of the 34th ICRC, PoS (ICRC2015), The Hague, The Netherlands, 30 July–6 August 2015. Viola, S.; Ardid, M.; Bertin, V.; Enzenhöfer, A.; Keller, P.; Lahmann, R.; Larosa, G.I.; Llorens, C.D. NEMO-SMO acoustic array: A deep-sea test of a novel acoustic positioning system for a km 3-scale underwater neutrino telescope. Nucl. Instrum. Methods Phys. Res. Sect. A 2013, 725, 207–210. [CrossRef] Ardid, M. Calibration in acoustic detection of neutrinos. Nucl. Instrum. Methods Phys. Res. Sect. A 2009, 604, S203–S207. [CrossRef] Ardid, M.; Martínez-Mora, J.A.; Bou-Cabo, M.; Larosa, G.; Adrián-Martínez, S.; Llorens, C.D. Acoustic transmitters for underwater neutrino telescopes. Sensors 2012, 12, 4113–4132. [CrossRef] [PubMed] Kossof, G. The effects of backing and matching on the performance of piezoelectric ceramic transducers. IEEE Trans. Sonics Ultrason. 1966, 13, 20–30. [CrossRef] Li, H.; Deng, Z.D.; Yuan, Y.; Carlson, T.J. Design parameters of a miniaturized piezoelectric underwater acoustic transmitter. Sensors 2012, 12, 9098–9109. [CrossRef] [PubMed] Adrián-Martínez, S.; Bou-Cabo, M.; Felis, I.; Llorens, C.D.; Martínez-Mora, J.A.; Saldaña, M.; Ardid, M. Acoustic signal detection through the cross-correlation method in experiments with different signal to noise ratio and reverberation conditions. Lect. Notes Comput. Sci. 2015, 8629, 66–79. Moffett, M.B.; Mello, P. Parametric acoustic sources of transient signals. J. Acoust. Soc. Am. 1979, 66, 1182–1187. [CrossRef] Llorens, C.D.; Ardid, M.; Sogorb, T.; Bou-Cabo, M.; Martínez-Mora, J.A.; Larosa, G.I.; Adrián-Martínez, S. The sound emission board of the KM3NeT acoustic positioning system. J. Instrum. 2012, 7, C01001. [CrossRef] Adrián-Martínez, S.; Ardid, M.; Bou-Cabo, M.; Larosa, G.I.; Llorens, C.D.; Martínez-Mora, J.A. A compact acoustic calibrator for ultra-high energy neutrino detection. Nucl. Instrum. Methods Phys. Res. Sect. A 2013, 725, 219–222. © 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 Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).