压电薄膜麦克风-eTouch

High-sensitivity piezoelectric microphones based on stacked cellular polymerfilms(L)Joachim Hillenbrand a)and Gerhard M.Sessler b)Department of Telecommunications,Darmstadt University of Technology,Merckstrasse25,64283Darmstadt,Germany͑Received30July2004;revised1September2004;accepted7September2004͒Improvements of the sensitivity of piezoelectric microphones based on charged cellular polymerfilms are reported.The improvements are achieved by͑1͒an increase of the piezoelectricd33-coefficient of the cellular polypropylenefilms by pressure expansion and͑2͒stacking of thefilms.Microphones consisting of a singlefilm of such material have sensitivities of about2mV/Paat1kHz,independent of size,while for a microphone withfive stackedfilms a sensitivity of10.5mV/Pa was measured.The equivalent noise level is about37dB͑A͒for the single-film transducerand26dB͑A͒for the stacked version.Advantages of these new piezoelectric transducers includetheir simple design,low cost,and small weight,as well as a large range of shapes and sizespossible.©2004Acoustical Society of America.͓DOI:10.1121/1.1810272͔PACS numbers:43.38.Fx,43.38.Ar,43.38.Kb͓AJZ͔Pages:3267–3270I.INTRODUCTIONCellular polypropylene͑PP͒,after appropriate electricalcharging,is highly piezoelectric.1–3In particular,the piezo-electric d33-coefficient of this material reaches values ofabout150pC/N in the audio frequency range and is thusaboutfive times as high as that of polyvinylidenefluoride ͑PVDF͒,the best conventional piezoelectric polymer.It was therefore suggested to use charged cellular PP in electro-acoustic and electromechanical transducers.In particular,implementations of loudspeakers,4microphones5,6andhydrophones6have been described in the literature.Althoughof considerably simpler design than conventional transduc-ers,the previously implemented new devices did not yetreach the electro-acoustic performance of older capacitiveand piezoelectric systems.Recently,the piezoelectric activity of cellular PP hasbeen significantly increased by thickness-expansion of thepolymer.7–11Microphones with such improvedfilms,show-ing sensitivities of about2.2mV/Pa at1kHz,have alreadybeen described by the present authors.12The use offilmstacks in such microphones,not yet implemented experimen-tally,is a method to further enhance the sensitivity of thesetransducers.This suggests to build advanced microphonesand to examine their electro-acoustic properties.In the present letter,the implementation and character-ization of such microphones are described.In particular,thecellularfilms are briefly specified in Sec.II,the microphonedesign and measuring methods are outlined in Sec.III,electro-acoustic measurements on the new systems are re-ported in Sec.IV,and the properties of these microphonesare discussed in Sec.V.II.CELLULAR PPThe upper part of Fig.1shows a SEM photograph of thecross sectional area of a cellularfilm͑HS01by VTT,Tam-pere,Finland͒.Thefilm is usually charged on its surface bya corona discharge.Due to the ensuing electricfield in theinterior of thefilm,discharges occur in the voids and charg-ing as shown in the lower part of thefigure is achieved.Acharge distribution of this kind in a nonhomogeneous mate-rial causes the piezoelectric effect.2All piezoelectricfilms used in the present microphoneswere made of a commercial cellular PPfilm͑VHD40byTreophan,Neunkirchen,Germany͒.Expansion of thesefilmsis achieved by a pressure treatment consisting in the appli-cation of an increased gas pressure for some time,followedby pressure reduction to atmosphere.This results in an in-crease of the thickness of the lenslike voids which originallyextends up to about5␮m and approximately up to about10␮m after expansion.According to the Paschen law,the in-creased thickness after expansion lowers the electricfieldrequired for breakdown.Thus,for a given voltage,morevoids experience stronger breakdown,causing an increase ofthe piezoelectric coefficient.Since metallization after the ex-pansion causes shrinking of thefilm,a second expansionafter metallization softens the material and thus results inanother increase of d33.10The frequency response of the d33-coefficient of suchsamples was measured,utilizing the inverse piezoelectric ef-fect,by sinusoidal electrical excitation and interferometricmeasurement of the resulting surface deflection.A typicalresponse of an expanded sample is shown in Fig.2.In theaudio frequency range,d33-coefficients of about420pC/Nwere found.These values are larger than those previouslyreported for nonexpanded samples by a factor of3.The slowdecrease of d33up to about30kHz is due to an increase ofYoung’s modulus,while the resonance at approximately140kHz is determined by Young’s modulus and the mass of thefilm.9The pressure dependence of d33was determined quasis-tatically by measuring the generated charge upon pressureapplication.9Typical results show that up to pressures of afew kPa d33increases slightly.This indicates that the stress–a͒Electronic mail:j.hillenbrand@nt.tu-darmstadt.de b͒Electronic mail:g.sessler@nt.tu-darmstadt.destrain relationship of the cellular films is not exactly linear in this pressure range,as expected for cellular materials.13III.MICROPHONE DESIGN AND MEASURING METHODSThe expanded cellular VHD40films were used to con-struct piezoelectric microphones.6,12These transducers con-sist simply of a piece of the cellular material of 0.3cm 2size that is metallized on both sides.For a film thickness of 55␮m,the capacitance of the microphone is 8pF.Shielding requires the mounting of the film in a small housing.The microphone output is fed directly into a preamplifier of unity gain ͑B&K 2669͒14and its output into an audio analyzer ͑R&S UPD ͒.In the stacked microphones,the single film is substituted by a stack of films metallized on both sides which are glued on top of each other.Since the sound pressure acts on all films and since the films are electrically connected in series,the output voltages of all layers add up and the open circuitsensitivity of a microphone with n layers should be n times as high as that of a single-film transducer,while its capaci-tance and resonance frequency ͑see Sec.V ͒decrease by a factor of n .Electroacoustic measurements were carried out by plac-ing the microphone in an acoustic coupler with a volume of 0.4cm 3.A 1/8-in.condenser microphone ͑B&K 4138͒,also extending into the coupler,serves as a reference.The sound pressure in the coupler is generated with a small electro-dynamic speaker in a separate cavity coupled by a 4.5-cm-long metallic pipe into the measuring coupler.Electronic data recording and processing is carried out with the above-mentioned audio analyzer,which allows the evaluation of the frequency response of the sensitivity,its amplitude dependence,the total harmonic distortion,and the noise spectrum of the microphone and its amplifier.IV.EXPERIMENTAL RESULTSThe measured frequency responses of microphones with one and five films of cellular PP films are shown in Fig.3.As seen from the figure,the open-circuit sensitivity of the five-film microphone is about 10.5mV/Pa at 1kHz and thus,as expected ͑see above ͒,almost five times larger than the 2.2mV/Pa of the single-film microphone at this frequency.Both responses decrease by about 1dB from 20Hz to 1kHz,as expected from the frequency response of the d 33-coefficient shown in Fig.2.The ripples seen at 2kHz and above are due to the fact that the dimensions of the pressure chamber with its connector are comparable to a quarter wavelength at these frequencies.To test the linearity of the microphones,the dependence of the sensitivity on applied sound pressure was examined.The data ͑not shown in this letter ͒indicates that there is a 3%sensitivity increase up to 3.2kPa ͑164dB SPL ͒.This result is in qualitative agreement with the increase of d 33with pres-sure,discussed above,and thus an indication of nonlineari-ties of the stress–strain relationship.The total harmonic distortion ͑THD ͒of the cellular mi-crophone is also related to this nonlinearity.As measure-ments show,THD increases approximately proportionallytoFIG.1.SEM photograph of cross section of cellular PP film ͑HS01͒of 70␮m thickness ͑top ͒and schematic view of charge distribution in this mate-rial ͑bottom ͒.FIG.2.Interferometrically measured d 33-coefficient for expanded cellular PP film ͑VHD40͒of 55␮mthickness.FIG.3.Frequency response of cellular PP microphones with single film and a stack of five films,determined by a comparison method in an acoustic coupler.Films of about 55␮m thickness ͑VHD40͒were used.sound pressure and is less than1%at164dB SPL.This very small distortion originates probably to some part from the loudspeaker used in these experiments.The part generated by the microphone is again due to the small nonlinearities of the stress–strain relationship.The A-weighted noise voltages of the single-film trans-ducer and thefive-film stack transducer,combined with a preamplifier,are3.0and4.2␮V,respectively.From these values,total equivalent noise levels͑ENLs͒of37and26 dB͑A͒,respectively,are obtained.The noise corresponds closely to that of the preamplifier,as specified by the manufacturer.14The improvement of the ENL for the stack microphone by11dB is due to the increase of the sensitivity ͑14dB͒,reduced by the increase of the preamplifier noise which is mainly due to the lowering of the capacitance (Ϫ3dB).V.DISCUSSION AND CONCLUSIONSThe sensitivity M of the single-film microphone de-scribed above is related to the d33-coefficient by15Mϭd33͑s1ϩ␧s2͒␧␧0,͑1͒where s1and s2are the combined thicknesses of all solid or gas parts of the cellularfilm,respectively,and␧o and␧are the absolute and relative permittivities,respectively.From Eq.͑1͒the observed sensitivity of2.2mV/Pa is obtained for s1ϭ26␮m,s2ϭ30␮m͑calculated from the density and to-tal thickness of thefilm͒,and␧ϭ2.35by substituting d33ϭ475pC/N,which is close to values actually measured in-terferometrically͑cf.Fig.2͒.The measured sensitivity of10.5mV/Pa for afive-film microphone is very high for a piezoelectric microphone and is comparable with sensitivities of electret condenser micro-phones.Even higher sensitivities may be possible by further increasing the d33-coefficients which can be achieved by in-creasing the charge density and by decreasing Young’s modulus of the cellularfilms.10The sensitivity may also be improved by increasing the number n of piezoelectricfilms in the stack microphone.Since such an increase lowers the capacitance of the device,stray capacitances and the input capacitance of the preamplifier have an adverse effect on the sensitivity.For this reason,the sensitivity of the present ex-perimental design will not gain very much by increasing n beyond5.However,systems with reduced stray capacitance and/or with larger transducer area will show improved sen-sitivities for nϾ5.The eventual limit will be reached when thefilm stack capacitance becomes smaller than the input capacitance of the preamplifier.Equally important is the equivalent noise level which is at37and26dB͑A͒for the single-andfive-film micro-phones,respectively.Particularly the latter value is again comparable with that for typical electret microphones16and is much better than that of previous cellular microphones͓52 dB͑A͔͒.6A drawback of the present cellular microphones is their decrease of sensitivity at temperatures in excess of60°C due to the instability of the electret charges in the cellular PP film.Efforts are underway to produce cellularfilms of poly-mers with better charge stability than that of the presently used PP types.Additional features of the cellular microphones are their low harmonic distortion and their high resonance frequen-cies.As Fig.2shows,the resonance frequency of a single-film microphone is expected to be at about140kHz.For a stack of nfilms,the resonance frequency decreases by a factor of n since the mass of the system increases and the stiffness decreases by this factor.This suggests a resonance frequency of28kHz for thefive-film transducer.In addition to these features,the cellular PP micro-phones have a simple design.The transducers consist essen-tially only of one or several pieces of charged and metallized cellularfilms,equipped with suitable backing and shielding. No miniature air gaps,as in electret microphones,have to be maintained.Cellular microphones can therefore be manufac-tured at very low cost.These features make such micro-phones very suitable for a wide range of applications.Because of the ambiguous meaning of the term‘‘cellular microphones’’and in view of the correlation of electret and piezoelectric properties in cellularfilms,we suggest calling these transducers‘‘piezo electret microphones.’’ACKNOWLEDGMENTSThe authors are grateful to Dr.Xiaoqing Zhang for preparation of the expanded cellularfilms,to Treofan for supplying the originalfilms,and to the Deutsche Fors-chungsgemeinschaft and the V olkswagen Foundation forfi-nancial support.1J.Lekkala,R.Poramo,K.Nyholm,and T.Kaikkonen,‘‘EMF force sensor–aflexible electretfilm for physiological applications,’’Med.Biol. put.34,67–68͑1996͒.2G.M.Sessler and J.Hillenbrand,‘‘Electromechanical response of cellular electretfilms,’’Appl.Phys.Lett.75,3405–3407͑1999͒.3S.Bauer,R.Gerhard-Multhaupt,and G.M.Sessler,‘‘Ferroelectrets:Soft Electroactive Foams for Transducers,’’Phys.Today57,37–43͑February 2004͒.4M.Antila,T.Muurinen,J.Linjama,and H.Nyka¨nen,‘‘Measurement methods offlat panel electromechanicalfilm loudspeakers,’’Active97, 607–618͑1997͒.5H.Nyka¨nen,M.Antila,J.Kataja,J.Lekkala,and S.Uosukainen,‘‘Active control of sound based on utilizing EMFI-technology,’’Active99,1159–1170͑1999͒.6R.Kressmann,‘‘New piezoelectric polymer for air-borne and water-borne sound transducers,’’J.Acoust.Soc.Am.109,1412–1416͑2001͒.7M.Paajanen,M.Wegener,and R.Gerhard-Multhaupt,‘‘Understanding the role of the gas in the voids during corona charging of cellular electret films—a way to enhance their piezoelectricity,’’J.Phys.D34,2482–2488͑2001͒.8J.Hillenbrand,X.Zhang,J.Zhang,and G.M.Sessler,‘‘Pressure-treated cellular polypropylene with large piezoelectric coefficients,’’2003Annual Report,Conf.Electric.Insul.and Diel.Phenom.͑2003͒,pp.40–43.9X.Zhang,J.Hillenbrand,and G.M.Sessler,‘‘Piezoelectric d33-coefficient of cellular polypropylene subjected to expansion by pressure treatment,’’Appl.Phys.Lett.85,1226–1228͑2004͒.10X.Zhang,J.Hillenbrand,and G.M.Sessler,‘‘Improvement of piezoelec-tric activity of cellular polymers by a double-expansion process,’’J.Phys. D37,2146–2150͑2004͒.11M.Wegener,W.Wirges,R.Gerhard-Multhaupt,M.Dansachmu¨ller,R. Schwo¨diauer,S.Bauer-Gogonea,S.Bauer,M.Paajanen,H.Minkkinen,and J.Raukola,‘‘Controlled inflation of voids in cellular polymer ferro-electrets:Optimizing electromechanical transducer properties,’’Appl. Phys.Lett.84,392–394͑2004͒.12J.Hillenbrand and G.M.Sessler,‘‘New piezoelectric transducers based on expanded cellular polymer electrets,’’in Proceed.of ICA2004,The18th Internat.Congress on Acoustics,Kyoto,Japan,2004,V ol.I,349-352. 13L.J.Gibson and M.F.Ashby,Cellular Solids,2nd ed.͑Cambridge U.P.,Cambridge,1997͒.14See data sheet‘‘B&K microphone preamplifier type2669,’’/pdf/bp1422.pdf15J.Hillenbrand and G.M.Sessler,‘‘Piezoelectricity in cellular electret films,’’IEEE Trans.Dielectr.Electr.Insul.7,537–542͑2000͒.16G.M.Sessler,‘‘Microphone,’’in Encyclopedia of Science&Technology, 9th ed.,V ol.11͑McGraw-Hill,New York,2002͒,pp.88–95.。