Encyclopedia of Smart Materials (Vols 1 and 2) - M. Schwartz (2002) WW Part 10 ppt

Rel amp (power units) 0.06 0.05 0.04 Figure 55 Time and frequency domains of ultrasoun by noncontact transducer, per Fig 54 setup 0.03 0.02 0.01 100 140 180 220 260 Time (microsec) 300 340 Figure 53 Noncontact ultrasound transmission through a human heel using 250-kHz (top) and 500-kHz (bottom) frequency transducers The first peak corresponds to ultrasound transmission through air, skin, tissue, and heel bone Other peaks are not identified the material surface in ambient air The ultrasound received by this transducer was amplified by a 64-dB gain Figure 55 shows the time and frequency domain of the ultrasound detected (heard) by the NC transducer By sweeping the frequency across a wide range, the frequencydependent response from the source (vibrating system) can be investigated and related to its characteristics or condition In this mode, we successfully interrogated frequencies Non-contact passive “Listener” 3.5 MHz 12.5 mm diameter Broadband amplifier mm Ambient air 25 mm Steel Ultrasound source transducer 800 KHz to MHz Bandwidth at −6 dB Single burst 16 volt sine wave Figure 54 Experimental setup for passive operation of noncontact transducer as high as MHz in ambient air This opens th noncontact acoustic emission, acoustoultrasonics other situation where detection of high frequen sound is desired Applications of the passive u transducers are dynamics of vibration, material testing of railroad, highways, bridges, runways, e Other Noncontact Ultrasound Applications Besides the applications of NCU described here, t can also be used for level detection; dimensi proximity analysis; high temperature material ev analysis of liquid-sensitive and hazardous mate analysis of gases and liquids Finally, it suffices to if ultrasound can be propagated through a medi flected from an interface, then much information medium and the interface can be obtained CONCLUSIONS In this paper, we outlined the significance of ultra nondestructive characterization of materials and invasive diagnostic applications in the medical have also shown the feasibility of noncontact u measurements in the time, frequency, and image analogous to other wave-based methods Underscoring the significance of the noncont sound mode, we presented a detailed discussion difficulty of achieving this mode We have also sh this work ultimately resulted in very high tran noncontact transducers, thus making the nonco trasound mode a reality Applications of these tra in industry and the medical field have been des using documentary evidence We also provided an introduction to a novel u noncontact analyzer and its applications for chara industrial and biomedical materials and product We believe that the noncontact ultrasound among the most significant developments for ch ing and analyzing all states of matter Though and valuable suggestions of E Blomme, Katholieke Hogeschool, Belgium and M Landa, Academy of Sciences, Czech Republic, are acknowledged in kind The work presented in this article was supported by the continuing efforts of SecondWave and Ultran Laboratories for the advancement of industry and medical science through innovative developments in ultrasound BIBLIOGRAPHY J Curie and P Curie, Bull no Soc Mineral France 3:90 (1880), C.R Acad Sci Paris 91:294 (1880) Apparatus for Warning a Ship at Sea of its Nearness to Large Objects Wholly or Partially under Water, Brit Pat Specification 11,125, March 27, 1913, R.L Richardson R.E Green, in Materials Analysis by Ultrasonics, A Vary, ed., Noyes Data, NJ, 1987, p Z Cho, J.P Jones, and M Singh, Foundations of Medical Imaging Wiley, NY, 1993, pp 477–486 R.M White, J Appl Phys 34: 3559–3567 (1963) A.A Bondarenko, Y.B Drobat, and S.V Kruglov, Soviet J NDT 12: 655–658 (1976) H.M Ledbetter and J.C Moulder, J Acoust Soc Am 65: 840– 842 (1979) A.M Aindow, R.J Dewhurst, S.B Palmer, and C.B Scruby, NDT Int 17: 329–335 (1984) G.A Allers, in Intelligent Processing of Materials and Advanced Sensors, H.N.G Wadley, P.A Parish, B.B Rath, and S.M Wolf, eds., Metallurgical Society, PA, 1986, pp 17–27 17 D Reilly and G Hayward, IEEE Ultrasonic Sy pp 763–766 18 Ultrasonic Transducer for High Transduction in Method for Ultrasound NonContact Transmission Materials, US and international patents pending cess, 1997–1999, M.C Bhardwaj 19 D.W Schindel, D.A Hutchins, L Zou, and M S Trans Ultrasonics Ferroelectic Frequency Control (1995) 20 I Ladabaum, B.T Khuri-Yakub, and D Spolian Phys Lett 68: 7–9 (1996) 21 M Castaings and B Hosten, Ultrasonics 36: 361– 22 M Landa, M.C Bhardwaj, and I Neeson, Institu momechanics, Academy of Sciences of the Czech Prague, CZ, Report no Z1266/99 (1999) 23 M.C Bhardwaj, Mater Res Innovation 1: 188–196 24 J.P Jones, D Lee, M Bhardwaj, V Vander B Achauer, Acoust Imaging 23: (1997) 25 M.C Bhardwaj, Proc Am Ceram Soc 89: (1998) 26 T Carneim, D.J Green, and M.C Bhardwaj, Ce (1999) 27 B.R Tittmann, M.C Bhardwaj, V Vanderva Neeson, Proc 23rd Annu Conf Composites Adv Cer Struct The American Ceramic Society, Westerville 28 M.C Bhardwaj, I Neeson, M.E Langron, and V V 24th Annu Conf Composites Adv Ceram Mater American Ceramic Society, Westerville, OH (2000) 29 R.Y Vun, Q Wu, M Bhardwaj, and G Stead, Int Symp Nondestructive Test Wood, University Hungary, Sopron, Hungary, 2000 of a smart paint is analogous to the action of that discharges and soaks up water in respon application and release of external pressure (6 analogy, a smart paint is a sponge that repeats t releasing and drawing an electrical charge at th frequency of a structural material or at a freque AE wave traveling through the material A smart paint is applied directly to the surface tural material when the material is a conductor or carbon fiber composite In this case, the condu terial can be used as a bottom electrode for the sm When the structural material is an insulator lik or ceramic, on the other hand, an electroconduc is first applied to the material surface, thus form conducting layer as a bottom electrode Then, paint is applied to the surface of the bottom Whether the structural material is conducting o ing, an electroconductive paint is applied to the the smart paint film, thus forming a thin conduc as a top electrode Then, a high voltage is appl smart paint film using the top and bottom electr making the film piezoelectrically active This pol dure is usually performed in air at room temper Smart paints are piezoelectric composites that piezoceramic and polymer phases (see Character Piezoelectric Ceramic Materials; Piezoelectricit mers) Thus, smart paints and piezoelectric c have essentially the same nature with respect to tors such as the ceramic/polymer composition, th of preparation, the poling procedure, and the m electrical, and piezoelectric properties An essen ence exists in that a piezoelectric composite is discrete point sensor or actuator, but a smart pai as a continuously distributed sensor that can cov surface area of a structural material Paints are used everywhere in an industrialized society (1,2) The most important functions of paints are protection and decoration of a substrate Paints can protect substrates against corrosion, oxidative aging, weathering, and mechanical damage and can also provide pleasant color contrasts or a lustrous appearance, hide imperfections in the substrate such as knots in wood, or enhance the beauty of the substrate by using a wood grain In other words, paints can add to the useful life of materials and also to their attractiveness (1) Smart paints are an innovative type of paint that has a sensor function as well as the protective and decorative functions of conventional paints Smart paints can detect abnormal vibration of a structural material by monitoring the natural frequencies and mode shapes of the material They can also detect damage generated in the material by monitoring the acoustic emission (AE) wave traveling from the damage location to the material surface Vibration and AE can be monitored in real time, thus enabling health monitoring of the material even during operation Smart paints are used in large-scale structures such as vehicles operated at high speeds, civil infrastructures of huge mass and volume, and special facilities that contain large amounts of petroleum, nuclear fuel, and explosive substances An accident in these facilities can be catastrophic because an enormous amount of energy stored in the form of kinetic, potential, or internal energy is released suddenly by the accident Smart paints can possibly prevent such a disaster by warning of abnormal vibration and damage generated in a structural material Hence, one reference goes so far as to say “Brush with disaster—Smart paint warns of impending doom” (3) The frequency of health monitoring needed for structural materials increases steadily as age increases because the corrosion of steel and concrete progresses gradually during the service period of several decades Smart paints can be applied to a structural material at any time before and after the construction of the structure, thus making health monitoring quite, easy even for a structure already in active service Smart paints can make a significant contribution to increasing the service life of a structure, and consequently to saving natural resources PIEZOELECTRIC COMPOSITES Piezoceramics such as barium titanate (BaT lead zirconate titanate (PZT) are typical pie materials that have excellent properties such electromechanical coupling coefficient and a dielectric constant (7,8) Piezoceramics, howe the problem that the high density inherent in makes the specific acoustic impedance much hi that of water or human tissue, thus causing i mismatch (7) Brittleness common to all ce ∗ Deceased 754 the problems of piezoceramics and piezoelectric polymers simultaneously (9–11) The electrical and mechanical properties of piezoelectric composites are determined primarily by the fraction of the piezoceramic and polymer phases and by the properties of these constituent materials (12–14) Composite properties are affected also by the connectivity pattern of the piezoceramic and polymer phases (15–20) COMPOSITION OF SMART PAINTS The smart paints reported so far are piezoelectric composites made up of piezoceramic particles dispersed in a polymer matrix The polymer matrix need not be piezoelectrically active, and hence popular polymers such as alkyd, acrylic, and epoxy resins can be used as the matrix resin The preparation of smart paints and the application procedures are essentially the same as those of conventional paints, except for poling for a dried film of smart paint As a result, most of the fundamental characteristics and functions of conventional paints are imparted to smart paints, thus enabling smart paints to have protective, decorative, and sensor functions simultaneously Smart paints can form continuous paint films covering a large surface area of a structural material Because of the electrically insulating nature of the paint film, however, the electrical charge actually detected is only that generated in a region that has an electrode on the surface of the paint film Therefore, if a set of separate electrodes is formed on the paint film surface, the electrical charge generated in each region can be detected and analyzed separately This feature of smart paints enables the application of the paints as a vibrational modal sensor that can determine the natural frequencies and mode shapes of a structural material (21,22) Furthermore, this feature enables another application of smart paints as an AE sensor that can determine the damage location in a structural material quite easily without using the conventional technique based on the arrival time difference of an AE wave (5) Paints in general can be applied to all kinds of materials such as metals, composites, concrete, and ceramics; the material surface can be flat, curved, or even irregularly shaped Furthermore, paints can be applied and reapplied at any time, when necessary Final dry films of paints are generally light, flexible, and tough These excellent properties of paint in general are imparted to smart paints as well, thus giving the smart paints further useful features as a solution in a solvent or as a dispersion of fine in a nonsolvent Such a solution or dispersion i vehicle Paint pigments are finely divided, insolu particles such as titanium dioxide (TiO2 ), zinc oxi and calcium carbonate (CaCO3 ) The pigment par dispersed stably in the paint vehicle before applic the pigment particles are distributed uniformly out the binder resin in the dried paint film Th tive functions of a paint are due, for the most pa pigment The basic components of smart paints are e the same as those of conventional paints, except t ceramics such as PZT and BaTiO3 are used as pig smart paints The piezoceramics used in the sma so far are PZT (23–30) and lead titanate (PbTiO3 ) the binders used are acrylic resin (23), polyureth and epoxy resin (25–29) Smart paints made up components are prepared by essentially the sam dure as used for conventional paints Smart p applied by using familiar coating tools such as rollers, or spray guns Smart paints are also cur usual way in air at ambient temperature or at temperatures Electrode Formation and Poling A simple method for forming an electrode on th of a paint film is to apply an electroconductive pa ing a coating tool such as a brush or roller A more method is to deposit a vapor of gold or aluminum paint film surface (30) A screen mask technique fective for this purpose, especially when the elect tern is complicated The main advantage of this t is that leads as well as electrodes can be print paint film surface, as shown in Fig This techni ever, has the disadvantage that it cannot be used structures such as airplanes, trains, or bridges For such large structures, an ordinary coatin by brush, roller, etc may be the most practical fo an electrode on the paint film surface As a lea electrode, on the other hand, a thin electrical wir ∼50 µm thick or so may be the most practical ch large structure because such a thin wire or tape rable in thickness to a paint film and hence, can in the paint film or under a topcoat Note that wh paints are put into practical use, the electrodes are covered by a topcoat, thus making the appea actly the same as that of conventional paints EVALUATION OF SMART PAINT FILMS 10−6 Output charge, C/m2 Piezoelectric composites are usually poled in an oil bath at elevated temperatures because poling at a higher temperature achieves saturation poling in a lower poling field For smart paints, on the other hand, poling is done in air at room temperature because even room temperature poling can achieve high enough piezoelectric activity for the paint application to serve as vibrational and AE sensors integrated into a structural material (25–29) 10−7 10−8 10−9 10−10 The sensor function of smart paints relies heavily on the piezoelectric activity of the poled paint film Usually, the activity is expressed in terms of a piezoelectric constant which is the ratio of the charge developed per unit surface area or the voltage developed per unit film thickness to the stress or strain applied externally The charge-tostress, voltage-to-stress, charge-to-strain, and voltage-tostrain ratios are the piezoelectric constants d, g, e, and h, respectively (7) Piezoelectric materials are inherently anisotropic, and hence two subscripts are attached to the piezoelectric constant to describe the anisotropic properties The first subscript is used to indicate the direction of the charge or voltage development, and this is always the film thickness direction for a piezoelectric film such as PVDF or a smart paint film The second subscript is used to indicate the direction of the stress or strain applied externally, and this direction is any of the 1, 2, and axes of the film which correspond to the length, width, and thickness directions, respectively (7) Sensitivity as a Vibrational Sensor When a structural material is deformed, strain is developed in all directions of the material, including the direction tangent to the material surface This is also true when the structural material is vibrating For a smart paint used as a vibrational sensor, therefore, one of the most important sensitivities to be evaluated is the piezoelectric constant e31 because this constant is the ratio of the charge per unit surface area to the strain in the direction tangent to the paint film surface The e31 constant is evaluated from vibrational measurement on a cantilever beam like that shown in Fig A typical example of the measurement is shown in Fig 50 100 150 200 Frequency, Hz Figure Frequency spectra of output signals from a paint film formed on one surface of an aluminum beam a strain gauge bonded to the opposite surface of the be for a paint film which has the PZT/epoxy comp 53/47 by volume and is formed on the surface minum beam 3.0 mm thick, 30 mm wide, and 460 (350 mm long as a cantilever beam) (27) This e for a 109-µm thick paint film cured at room tem and poled at 240 kV/cm for The spectrum tained from the paint film is similar to that obta a strain gauge which is bonded to the opposite the beam to monitor the strain developed in the of the cantilever length Then, the e31 constant is from the charge-to-strain ratio at a natural frequ or 112 Hz The e31 constant thus evaluated depends on tors such as the poling field, the film thickness temperature, and the PZT/epoxy composition typical example of the poling-field and film-thic pendence is shown in Fig for paint films cure temperature that have the PZT/epoxy compositio by volume (27) The e31 constant increases stead poling field increases for all of the paint films sh and saturation poling is not achieved, even at a ing field of ∼150 kV/cm The e31 constant obta particular poling field, say, 100 kV/cm, increas thickness increases from 33 to 152 µm, thus ex clear film-thickness dependence This point is fu scribed later 50 100 150 200 Poling field (kV/cm) Figure Plots of the piezoelectric constant e31 vs the poling field for PZT/epoxy paint films cured at room temperature and evaluated as a vibrational sensor Sensitivity as an Acoustic Emission Sensor In many cases, eventual failure of a structural material occurs after a certain amount of damage accumulates within the material The generation of such damage is almost always accompanied by the emission of an AE wave, and hence the damage generated and accumulated can be detected by monitoring the AE wave (5) The AE wave is emitted in all directions, and consequently, an AE wave that arrives at the material surface and enters the smart paint film on the material surface always exists Furthermore, an AE wave that enters the paint film nearly perpendicularly always exists Such an AE wave develops strain in the paint film in the direction normal to the film surface because the AE wave is a compression wave in which particle motion is in the same direction as the propagation of the wave For a smart paint used as an AE sensor, therefore, the sensitivity to be evaluated is the piezoelectric constant h33 because the h33 constant refers to the ratio of the voltage per unit film thickness to the strain in the direction normal to the paint film surface For a conventional AE sensor, the sensitivity s is usually given by s = V/v0 , where V is the output voltage of the sensor and v0 is the velocity amplitude of AE waves (31) The strain amplitude of AE waves ε0 is given by ε0 = v0 /v, where v is the phase velocity of AE waves Combining these equations with h33 = (V/d)/ε0 leads to s = h33 d/v, where d is the film thickness This equation indicates that the paint film sensitivity as an AE sensor s is independent of the frequency of AE waves and that the sensitivity increases linearly as film thickness increases This equation also indicates that the h33 constant is calculated from h33 = sv/d The paint film sensitivity as an AE sensor is evaluated from measurement using an ultrasonic transducer to produce AE waves and a laser Doppler vibrometer to monitor the velocity amplitude of the AE waves (28) A typical example of the measurement is shown in Fig for a paint film that has the PZT/epoxy composition of 53/47 by volume and is formed on the surface of square aluminum plate Frequency, MHz Figure Frequency spectra of output signals from a paint film formed on one surface of an aluminum plate a laser Doppler vibrometer that monitors the velocity of AE waves 0.2 mm thick that has 50 mm sides This examp 152-µm thick paint film cured at room tempera poled at 184 kV/cm for The spectral shape from the paint film is similar to that obtained laser vibrometer in the frequency range above ∼ Such a similarity of spectral shapes reflects a n frequency response of the paint film to AE wav the paint film sensitivity as an AE sensor is e from the average ratio of the output voltage of film to the velocity amplitude of AE waves in the f range 0.3–1.0 MHz The paint film sensitivity thus evaluated, s ca verted into the h33 constant by using the rel h33 = sv/d, where v is the phase velocity of AE wa PZT/epoxy paint film The h33 constant calculated an assumed value of v = 2850 m/s (6) is plotted in a function of film thickness for paint films cured temperature that have the PZT/epoxy composition 120 100 h33 (MV/m) / (m/m) 80 60 40 20 0 50 100 150 200 Film thickness, µm 250 Figure Plots of the piezoelectric constant h33 at 50 ( 150 ( ), and 250 kV/cm ( ) vs film thickness for PZT/e films cured at room temperature and evaluated as a emission sensor temperature and the PZT/epoxy composition (26–29) Such complicated poling behavior is virtually determined by the electric field that acts on the PZT particles dispersed in the epoxy matrix The most important factors that determine the electric field and, consequently, the poling behavior of the paint film are the electrical conductivities of the PZT particles and the epoxy matrix, the connectivity pattern of the PZT phase, and the space charge accumulated at the PZT/epoxy interface Electrical Conductivities of Constituent Materials It is now well established that in poling a composite specimen made of piezoceramic particles dispersed in a polymer matrix, the electric field that acts on the ceramic particles is very low compared with that applied externally to the composite specimen (14,32) This occurs because the electrical conductivity of polymeric materials in general is much lower than that of ceramic materials, and hence the polymer matrix in the composite specimen bears almost all of the externally applied electric field at the expense of the electric field that acts on the ceramic particles As a result, the piezoelectric activity of the ceramic/polymer composite specimen is very low, compared with a pure piezoceramic specimen poled in the same electric field This idea explains why saturation poling is not achieved, even in a high poling field of ∼150 kV/cm, as seen in Fig Saturation poling for a pure PZT ceramic specimen, on the other hand is achieved in a low poling field of ∼10 kV/cm (12) A promising solution to this problem is to increase the electrical conductivity of the polymer matrix up to that of the ceramic particles, so that the electric field distribution becomes uniform throughout the composite specimen This can be achieved by adding a small amount of a semiconductor filler such as carbon, germanium, or silicon to the composite specimen (32) This can also be achieved by poling at a high temperature where the electrical conductivity of the polymer matrix becomes equal to that of the ceramic particles (33) Connectivity Pattern of Ceramic Phase Figure is a scanning electron microscopy (SEM) picture that shows the internal microstructure of a paint film that has the PZT/epoxy composition of 53/47 by volume (27) It is seen that the size of PZT particles ranges from ∼0.5 to ∼1.5 µm, and that a substantial fraction of the PZT particles are in contact with each other, so that the PZT phase 10 Figure SEM picture of a paint film that has the composition of 53/47 by volume This example is a paint film cured at 150◦ C is practically self-connected in three dimens self-connectivity of the PZT phase is one of the portant factors that determines the poling beh PZT/epoxy paint film In fact, the paint film is when the PZT volume fraction is decreased to su that the PZT particles are isolated from one anot continuous phase of the epoxy matrix (26) Figures and show that the poling beha PZT/epoxy paint film depends on the film thick when the PZT volume fraction remains constan A SEM picture like that shown in Fig 6, howev no observable difference in the PZT phase conne paint films that have different thicknesses The in the PZT phase connectivity is reflected much plicitly in the current–voltage characteristic of film rather than in the SEM picture, as describe Space Charge at the Ceramic/Polymer Interface The current–voltage characteristic of a PZT/ep film shows that the conduction is ohmic in a lo field, whereas in a high electric field, the spac limited (SCL) conduction predominates over ohm tion (28) Furthermore, the current–voltage char shows that the critical electric field at which the SCL transition takes place decreases as the film decreases The result is that conduction during process is mostly SCL for a thin film, whereas c is mostly ohmic for a thick film The SCL conduction becomes predominan space charge of electrons is injected into the P paint film during the poling process The space c a tendency to build up preferentially at the int tween the PZT and epoxy phases in the paint The space charge decreases the electric field act PZT phase, and hence decreases the piezoelectr of the paint film obtained in a given poling field fect of the space charge becomes significant, pa for a thin film, because SCL conduction becom paint film Therefore, the drying rate of the wet paint film is another important factor that determines the poling behavior of a PZT/epoxy paint film TECHNIQUES FOR APPLYING SMART PAINT FILMS Techniques for applying smart paint films as vibrational and AE sensors are essentially the same as those for a PZT ceramic or PVDF film bonded to the surface of a structural material Theories, models, methods, and systems constructed for use of the PZT and PVDF sensors (21,22,34) can also be applied to smart paint films used as vibrational and AE sensors integrated into a structural material Vibrational Modal Sensor One example of an application of smart paints is a vibrational modal sensor integrated into a structural material As noted before, the sensitivity of the paint film used for this purpose is the e31 constant which is the ratio of the charge per unit surface area to the strain in the direction tangent to the paint film surface Figure shows a result of vibrational modal testing of a cantilever beam like that shown in Fig by using a PZT/epoxy paint film Modal strain, 10−6 m/m 150 100 50 −50 −100 10 15 20 25 30 Longitudinal coordinate, cm 35 Figure Modal strain shapes of a cantilever aluminum beam for the first (◦), second (•), and third modes ( ) determined by a PZT/epoxy paint film formed on the beam surface ment of a uniform cantilever beam, x is the lon coordinate of the beam, ε is the longitudinal stra beam surface, and η is the half-thickness of the b Modal displacement shapes determined by this are identical to those determined by a laser Do brometer that measures the transverse moveme beam surface (26) Thus, smart paints offer an in and promising alternative to conventional sensor accelerometers and laser vibrometers (1) FUTURE DIRECTIONS Smart Paints The highest sensitivity of smart paint films ac far is e31 = ∼40 × 10−3 (C/m2 )/(m/m) as a vibrati sor and h33 = ∼100 × 106 (V/m)/(m/m) as an A as shown in Figs and For commercially PVDF films, the sensitivity is e31 = ∼66 × 10− (m/m), e32 = ∼6.8 × 10−3 (C/m2 )/(m/m), and h33 106 (V/m)/(m/m), determined in essentially the s described before for smart paint films This indic the sensitivity of smart paint films is comparab of PVDF films So far as sensitivity is concerne fore, smart paints have already reached a leve for practical use For smart paints to be put into practical use, the paints must meet performance requirement exterior durability and sensitivity stability Exter bility is the paint films resistance to environment such as uv radiation, heat, moisture, oxygen, and These environmental factors can cause mechanic dation of paint films, thus leading to the failure o tective and decorative functions of smart paints vironmental factors may also cause electrical deg of paint films, thus leading to the failure of the sen tion of smart paints Considering that smart p truly appreciated when used in severe and isola ronments, the evaluation of exterior durability a tivity stability is absolutely necessary for the pa put into practical use Smarter Paints According to a concept of intelligent materials the intelligence in materials is classified into t egories; intelligence from the human standpoin gence inherent in materials, and intelligence at be avoided if possible A feasibility study of a poling-free piezoelectric paint shows that a paint made of PVDF particles and epoxy resin does not need poling for the final dry film to be piezoelectrically active (38) At the present stage, however, the piezoelectric activity is not enough for practical use of the paint film Studies are currently under way to increase the piezoelectric activity of the paint film From the standpoint of intensiveness of information, a smarter paint of the future will have a sensor function for material conditions such as vibration and damage generation and also for atmospheric variables such as temperature, pressure, moisture, and wind velocity Such a paint resembles human skin in that the skin has a sensor function for the external stimuli imposed on the human body and also for the surrounding conditions such as temperature, humidity, wind, and rain The ultimate goal of smart paints, therefore, should be to mimic the human skin as closely as possible ACKNOWLEDGMENTS The work in smart paints by S Egusa and N Iwasawa was supported by the Japan Atomic Energy Research Institute through the Special Program for Fundamental Researches (1991–1994) and through REIMEI Research Resources (1998) BIBLIOGRAPHY J.H Lowell, in Coatings, J.I Kroschwitz, ed., Encyclopedia of Polymer Science and Engineering, 2e., Wiley-Interscience, NY, 1985, Vol 3, pp 615–675 Z.W Wicks, Jr., in Coatings, J.I Kroschwitz, ed., Encyclopedia of Polymer Science and Engineering, 2e., Wiley-Interscience, NY, 1989, Supplement Vol pp 53–122 O Graydon, New Scientist, p 20, October 17, 1998 D.J Ewins, Modal Testing: Theory and Practice Research Studies Press, Taunton, 1984 C.B Scruby, J Phys E: Sci Instrum 20: 946–953 (1987) KYNAR Piezo Film Technical Manual, Pennwalt Corporation, Valley Forge, PA, 1987, p A.J Moulson and J.M Herbert, Electroceramics Chapman & Hall, London, 1990, Chap M.V Gandhi and B.S Thompson, Smart Materials and Structures Chapman & Hall, London, 1992, Chap 16 H Banno, Ferroelectrics 50: 3–12 (1983) 17 R.E Newnham, D.P Skinner, and L.E Cross, Mate 13: 525–536 (1978) 18 R.E Newnham, L.J Bowen, K.A Klicker, and L Mater Eng 2: 93–106 (1980) 19 R.E Newnham, Ferroelectrics 68: 1–32 (1986) 20 R.E Newnham and G.R Ruschau, J Am Ceram 463–480 (1991) 21 C.-K Lee and F.C Moon, J Appl Mech 57: 434–4 22 S.A Collins, D.W Miller, and A.H von Floto for Structural Control—Applications Using P Polymer Film Space Engineering Research C 90, Massachusetts Institute of Technology, Camb 1990 23 K.A Hanner, A Safari, R.E Newnham, and J R electrics 100: 255–260 (1989) 24 C.A Rogers and S.C Stein, Proc 1st Int Conf Mater 1993, pp 87–93 25 S Egusa and N Iwasawa, Proc 1st Int Conf Intelli 1993, pp 101–104 26 S Egusa and N Iwasawa, J Mater Sci 28: (1993) 27 S Egusa and N Iwasawa, Ferroelectrics 145: 45–6 28 S.S Egusa and N Iwasawa, J Appl Phys 78: (1995) 29 S Egusa and N Iwasawa, J Smart Mater Struct (1998) 30 J.M Hale and J Tuck, A Novel Strain Transducer U electric Paint Proc Mech Eng in press 31 ASTM E1106-86, Standard Method for Primary Ca Acoustic Emission Sensors American Society for T Materials, Philadelphia, PA, 1986, pp 489–498 32 G Sa-Gong, A Safari, S.J Jang, and R.E Ferroelectrics Lett 5: 131–142 (1986) 33 J.P Dougherty and Y Chen, Proc 2nd Int Conf Mater 1994, pp 462–473 34 C.A Rogers, ME 4016, Virginia Polytechnic Ins State University, Blacksburg, VA (private comm 1991) 35 S.H Crandall, N.C Dahl, and T.J Lardner, An tion to the Mechanics of Solids McGraw-Hill, p 628 36 T Takagi, Proc Int Workshop Intelligent Mater Japan, 1989, pp 1–10 37 J.M Hale, University of Newcastle, Newcastle (private communication, 1999) 38 S Egusa, 1998 REIMEI Conf., Japan Atom Research Institute, Tokai, Japan, July 14–15, 1999 32 G Kossmehl and M Samandari, Makromol Chem 186: 1565–1574 (1985) 33 A Lux, A.B Holmes, R Cervini, J.E Davies, S.C Moratti, J Gruener, F Cacialli, and R.H Friend, Synth Met 84: 293– 294 (1997) 34 Z Bao, Y Chen, R Cai, and L Yu, Macromolecules 26: 5281– 5286 (1993) 35 T.W Hagler, K Pakbaz, K.F Voss, and A.J Heeger, Phys Rev B Condens Matter 44: 8652–8666 (1991) 36 C Zhang, H von Seggern, K Pakbaz, B Kraabel, H.-W Schmidt, and A.J Heeger, Synth Met 62: 35–40 (1994) 37 Y Pang, M Samoc, and P.N Prasad, J Chem Phys 94: 5282– 5290 (1991) 38 C.J Wung, Y Pang, P.N Prasad, and F.E Karasz, Polymer 32: 605–608 (1991) 39 C.J Wung, W.M Wijekoon, and P.N Prasad, Polymer 34: 11174–11178 (1993) 40 L Bozano, S.E Tuttle, S.A Carter, and P.J Brock, Appl Phys Lett 73: 3911–3913 (1998) 41 S.A Carter, J.C Scott, and P.J Brock, Appl Phys Lett 71: 1145–1147 (1997) 42 J.S Salafsky, W.H Lubberhuizen, and R.E.I Schropp, Chem Phys Lett 290: 297–303 (1998) 43 T.A.E Loss, C.W Rogers, and M.O Wolf, Can J Chem 76: 1554–1558 (1998) 44 R.C Smith, W.M Fischer, and D.L Gin, J Am Chem Soc 119: 4092–4093 (1997) 45 J Obrzut and F.E Karasz, J Chem Phys 87: 2349–2358 (1987) 46 R.H Friend, D.D.C Bradley, and P.D Townsend, J Phys D Appl Phys 1367–1384 (1987) 47 M Meier, E Buchwald, S Karg, P Posch, M Greczmiel, P Strohriegl, and W Riess, Synth Met 76: 9599 (1996) ă 48 U Rauscher, H Bassler, D.D.C Bradley, and M Hennecke, Phys Rev B Condens Matter 42: 9830–9836 (1990) 49 M Yan, L.J Rothberg, E.W Kwock, and T.M Miller, Phys Rev Lett 75: 1992–1995 (1995) 50 L.J Rothberg, M Yan, E.W Kwock, T.M Miller, M.E Galvin, S Son, and F Papadimitrakopoulos, IEEE Trans Electron Devices 44: 1258–1262 (1997) 51 F Papadimitrakopoulos, K Konstadinidis, T.M Miller, R Opila, E.A Chandross, and M.E Galvin, Chem Mater 6: 1563–1568 (1994) 52 H.-H Horhold and J Opfermann, Makromol Chem 131: 105–132 (1970) 53 S Tokito, T Tsutsui, R Tanaka, and S Saito, Jpn J Appl Phys 25: L680–L681 (1986) Petermann, J Mater Sci 25: 311–320 (1990) 61 J.H.F Martens, E.A Marseglia, D.D.C Bradley, R P.L Burn, and A.B Holmes, Synth Met 55 (1993) 62 C.Y Yang, F Hide, M.A Diaz-Garcia, A.J Heeger, Polymer 39: 2299–2304 (1998) 63 B Winkler, S Tasch, E Zojer, M Ungerank, G L F Stelzer, Synth Met 83: 177–180 (1996) 64 W Memeger, Macromolecules 22: 1577–1588 (198 65 M Suzuki, J.-C Lim, and T Segusa, Macr 23: 1574–1579 (1990) 66 A Samoc, M Samoc, M Woodruff, and B Lut Plast Eng 49: 373–436 (1998) 67 T Kaino, K Kubodera, S Tomaru, T Kurihara T Tsutsui, and S Tokito, Electron Lett 23: (1987) 68 T Kaino, H Kobayashi, K Kubodera, T Kurihar T Tsutsui, and S Tokito, Appl Phys Lett 54: (1989) 69 C.-B Yoon and H.-K Shim, J Mater Chem (1998) 70 N.S Sacriciftci, D Braun, C Zhang, V.I Srdanov, A G Stucky, and F Wudl, Appl Phys Lett 62 (1993) 71 J.J.M Halls and R.H Friend, Synth Met 85: (1997) 72 J.J Dittmer, K Petritsch, E.A Marseglia, R H Rost, and A.B Holmes, Synth Met 102 (1999) 73 J Gao, G Yu, and A.J Heeger, Adv Mater (1998) 74 K Yoshino, T Kuwabara, T Iwasa, T Kawai, and Jpn J Appl Phys 29: L1514–L1516 (1990) 75 C.W Tang and S.A VanSlyke, Appl Phys Lett (1987) 76 C.W Tang, S.A VanSlyke, and C.H Chen, J App 3610–3616 (1989) 77 A.R Brown, D.D.C Bradley, J.H Burroughes, R N.C Greenham, P.L Burn, A.B Holmes, and A K Phys Lett 61: 2793–2795 (1992) 78 M Meier, E Buchwald, S Karg, P Posch, M P Strohriegl, and W Riess, Synth Met (1996) 79 K Naito and A Miura, J Phys Chem 97: 6240–6 80 D.R O’Brien, P.E Burrows, S.R Forrest, B D.E Loy, and M.E Thompson, Adv Mater 10: (1998) 88 I.D Parker and H.H Kim, Appl Phys Lett 64: 1774–1776 (1994) 89 H Becker, S.E Burns, and R.H Friend, Phys Rev B Condens Matter 56: 1893–1905 (1997) 90 E Peeters, M.P.T Christiaans, R.A.J Janssen, H.F.M Schoo, H.P.J.M Dekkers, and E.W Meijer, J Am Chem Soc 119: 9909–9910 (1997) 91 S Son, A Dodabalapu, A.J Lovinger, and M.E Galvin, Science 269: 376–378 (1995) 92 B.H Cumpston, I.D Parker, and K.F Jensen, J Appl Phys 81: 3716–3720 (1997) 93 J.C Scott, J.H Kaufman, P.J Brock, R DiPietro, J Salem, and J.A Goitia, J Appl Phys 79: 2745–2751 (1996) 94 K.M Vaeth and K.F Jensen, Appl Phys Lett 71: 2091–2093 (1997) 95 J.C Carter, I Grizzi, S.K Heeks, D.J Lacey, S.G Latham, P.G May, O Ruiz de los Panos, K Pichler, C.R Towns, and H.F Wittmann, Appl Phys Lett 71: 34–36 (1997) 96 J.M Bharathan and Y Yang, Appl Phys Lett 72: 2660–2662 (1998) 97 J.A Rogers, Z Bao, and L Dhar, Appl Phys Lett 73: 294– 296 (1998) 98 P.L Burn, A.B Holmes, A Kraft, D.D.C Bradley, A.R Brown, and R.H Friend, J Chem Soc Chem Commun 32–34 (1992) 99 I Sokolik, Z Yang, F.E Karasz, and D.C Morton, J Appl Phys 74: 3584–3586 (1993) 100 F Garten, A Hilberer, F Cacialli, E Esselink, Y Van Dam, B Schlatmann, R.H Friend, T.M Klapwijk, and G Hadziioannou, Adv Mater 9: 127–131 (1997) 101 S.T Kim, D.H Hwang, X.C Li, J Gruener, R.H Friend, A.B Holmes, and H.K Shim, Adv Mater 8: 979–982 (1996) 102 D.R Baigent, P.J Hamer, R.H Friend, S.C.Moratti, and A.B Holmes, Synth Met 71: 2175–2176 (1995) 103 S.J Chung, J.I Jin, and K.K Kim, Adv Mater 9: 551–554 (1997) 104 M Gao, B Richter, and S Kirstein, Adv Mater 9: 802–805 (1997) 105 M Uchida, Y Ohmori, T Noguchi, T Ohnishi, and K Yoshino, Jpn J Appl Phys 32: L921–L924 (1993) 106 M Hamaguchi and K Yoshino, Appl Phys Lett 69: 143–145 (1996) 107 O Onitsuka, A.C Fou, M Ferreira, B.R Hsieh, and M.F Rubner, J Appl Phys 80: 4067–4071 (1996) 108 Q Pei, G Yu, C Zhang, Y Yang, and A.J Heeger, Science 1086–1088 (1995) Q.M ZHANG VIVEK BHARTI GEORGE KAVARNOS Pennsylvania State University University Park, PA INTRODUCTION Poly(vinylidene fluoride) (PVDF) and the family o TrFE) [TrFE = trifluoroethylene] copolymers are the best-known examples of a class of high-per polymers noted for their remarkable piezoele ferroelectric properties (1–4) In 1969, Kawii d the exceptional piezoelectric behavior of PVDF, that time, was the highest among the known polymers (5) After more than 30 years of stud velopment, the piezoelectricity and electrom properties of PVDF and its copolymers have been markedly Today this class of polymer still poss highest electromechanical responses over a broa ature range among known synthetic organic m Further, when considered along with their easy c bility, flexibility, robustness, and lightness, it is no ing that electroactive polymers continue to be th interest of the designers of high-performance e chanical devices (4) When PVDF is stretched an a strong electric field, it exhibits piezoelectricity piezoelectric form, PVDF finds use in transduce requiring the interconversion of mechanical and energy Piezoelectric PVDF can be fabricated an a variety of sensors and actuators such as artifi cles and organs, medical imaging, blood-flow mon crophones, smart skins, underwater acoustic tra seismic monitors, fluid pumps and valves, surfa tic wave devices, robots, and tactile sensing de 11) P(VDF-TrFE) copolymers display similar an cases even superior properties (4,12) Even before the discovery of the high piezoele PVDF, the existence of its ferroelectric nature w lated by Lando et al based on the crystal unit c ture This hypothesis was confirmed about 10 ye (13–17) Many organic substances in fact exhibit roelectric property that is called polarization h (18–20) The copolymer of PVDF with TrFE and oroethylene (TFE), however, is the only polymer that shows both a well-defined polarization hyste The reaction can be initiated by a free radical that might be formed by the thermal decomposition of, say, benzoyl peroxide: RO−OR → 2RO The generated free radical then reacts with a vinylidene fluoride molecule in an initiation step: RO + CH2 =FH2 → RO−CH2 CF2 Thus begins the process of growing a polymer chain, a series of steps called chain propagation: RO−CH2 −CF2 + CH2 =CF2 → RO−CH2 −CF2 −CH2 −CF2 −CH2 −CF2 −CH2 −CF2 + CH2 =CF2 → −CH2 −CF2 −CH2 −CF2 −CH2 −CF2 In this sequence, the CF2 “head” of the VDF monomer is shown to add to the CH2 “tail” in a “head-to-tail” addition Propagation may also, however, involve head-to-head (CF2 to CF2 ), tail-to-tail (CH2 to CH2 ), and tail-to-head (CH2 to CF2 ) addition Chain termination of the growing polymer chains may occur by the combination of two growing chains or by hydrogen atom abstraction from a good hydrogen donor such as isopropyl alcohol (RH): −CH2 −CF2 −CH2 −CF2 −CH2 −CF2 + RH → −CH2 −CF2 −CH2 −CF2 −CH2 −CF2 H + R The synthesis of PVDF is normally carried out by emulsion polymerization where a dispersing medium such as water dissipates the high heat of the polymerization reaction (21) The copolymer P(VDF-TrFE) is synthesized in a similar manner by the copolymerization of trifluoroethylene and vinylidene (22) Here the growing chain can add to either monomer, for example RO−CH2 −CF2 + CHF=CF2 → RO−CH2 −CF2 −CHF−CF2 −CH2 −CF2 −CHF−CF2 + CH2 =CF2 → −CH2 −CF2 −CHF−CF2 −CH2 −CF2 Figure Polymers such as PVDF consist of crystal phous regions This figure depicts the crystalline region of folded lamellae The mole ratios of VDF in P(VDF-TrFE) copoly display the most striking ferroelectric propert from about 50 to over 90 mol% On the basis ratios, TrFE is much more likely to be adjacen than itself in a polymer chain The relative prop the monomers in the copolymers prescribe the tric properties of the copolymers, as we shall sho sections Furthermore, although head-to-tail bon vails in the polymer chains, there may be some t bonding (−CH2 −CF2 −CF2 −CH2 −) as well hea (−CF2 −CH2 −CH2 −CF2 −) The percentage of the generally is only a few mole% for most PVDF com It is most important to note that the monomer as well as the number of head-to-head or tail-to ages give rise to defects, which ultimately may the material properties of PVDF and P(VDF-TrF PVDF and P(VDF-TrFE) are semicrystalline (24) They are comprised of ordered regions of units (crystallites) surrounded by an amorpho scrambled, spaghettilike chains, as shown in Fi From many SEM studies, it has been estimated crystallites have thickness of about 10 to 20 nm polymer chain direction and extend to several m the other directions (23) The degree of crysta well as the orientation of the crystallites can be by various processing techniques Before discuss techniques, we present a survey of the major c forms of PVDF and P(VDF-TrFE) Molecular Conformations PVDF and P(VDF-TrFE) are polymorphic in that they may exist in several crystal forms (24 form, the chains are packed within crystal lattic cific conformations The crystal structures are by the conformations of the chains (as a series o or gauche (G) linkages), by the orientation of th sequences about the chain axis (parallel or ant and by the relative directions of adjacent chains (same direction) or down-down (opposite directi sualize conformations along a PVDF chain (Fig tions, whose energies are governed by the numb orine atoms substituted on adjacent carbon atom chain β α γ Crystal Structures I II,IIp III There are four major crystalline forms of PVDF form I, which is also known as the β phase, two an all-trans planar zigzag conformation are packe dividual orthorhombic unit cells having lattice di ˚ ˚ of a = 8.58 A, b = 4.90 A, and the chain directio ˚ axis c = 2.56 A (Fig 4) (25) The space group sym each unit cell in Cm2m It has been suggested th accurate model has CF2 groups deflected by ab opposite directions in a planar zigzag conformati ˚ fiber axis of 2c or 5.12 A(25) It will be noted from F Figure Extended chain segments of an all-trans, TGT G, and TTTGTTT G conformations carbon–carbon bond within one monomer unit can be projected perpendicular to the plane of the page (Fig 3) One carbon atom (of an adjacent monomer) and two hydrogen or fluorine atoms are then connected to the front carbon atom by bonds shown as solid lines; likewise, three atoms are connected to the back carbon atoms by bonds shown as dotted lines If the front carbon atom with its three atoms is rotated together while keeping the back atoms stationary, the steric energy of the structure changes because of repulsions between the atoms on the front and back carbons The angle of rotation is ϕ = when the four carbon atoms in figure all lie in the same plane In this structure, the substituted atoms are eclipsed and engaged in strong mutual repulsion As ϕ is varied to 60◦ , the repulsions decrease as the atoms on the front and back move away from one c H F H F CC H C H F Energy a b CF F H H F C H C C H F F C 60 120 180 Torsion angle Figure Interactions between fluorine and hydrogen atoms on adjacent carbon atoms in PVDF polymer chains lead to changes in the potential energy as the bond connecting the carbon atoms is rotated Figure Projection of form I (β-phase) of PVDF The tion of the packed chains is all-trans The chains are that the carbon-fluorine dipoles are in the same direct the b-axis The small and large filled circles represent c fluorine atoms, respectively; the small open circles rep drogen atoms b b Figure Projection of form II (α-phase) of PVDF The conformation of TGTG The chains are packed in an antiparallel in two directions about the chain axis The small and large filled circles represent carbon and fluorine atoms, respectively; the small open circles represent hydrogen atoms in the all-trans conformation, the fluorine atoms are positioned on one-side of the unit cell resulting in a net dipole moment The form I unit cell is quite polar having a net dipole of 2.1 debye As the structure of the unit cell of the form I crystal satisfies the symmetry requirement of a piezoelectric crystal, meaning that the crystal belongs to a noncentrosymmetric class, this is the form of PVDF that is responsible for its piezoelectric properties In form II, or the α-phase, the chain conformations are represented as a sequence of alternating trans and gauche sequences, or TGTG (Fig 5) (26) Each unit cell containing two chains is orthorhombic with lattice parameters ˚ ˚ ˚ a = 4.96 A, b = 9.64 A, and c = 4.62 A In the α-phase, adjacent chains are packed such that the dipole moments of the individual carbon–fluorine bonds are aligned perpendicular to the chain direction, canceling one another out The directions of the chains consist of a statistical average of up-up and up-down orientations A form IV (δ-phase) has been identified where the chains have the same conformations as in the α form but the carbon–fluorine bonds are aligned in one direction around the chain direction Figure A projection of form III (γ -phase) of PVDF conformation is TTTGTTTG The small and large filled resent carbon and fluorine atoms, respectively; the sma cles represent hydrogen atoms resulting in a net dipole (27), and the crystal lat meters are identical to form II In form III (γ -ph tals, the chain conformations are TTTGTTTG crystal lattice is monoclinic with lattice param ˚ ˚ ˚ 4.96 A, b = 9.67 A, c = 9.20 A, and β = 93◦ (28 alignment of the form III chains perpendicular to axis is in one direction, resulting in a polar cell ( In the P(VDF-TrFE) copolymers, the form I a crystal lattices are expanded structures in the perpendicular to the molecular chain (Table 1) T sion of the lattices accommodates the presence of number of substituted fluorine atoms, since fluo ˚ van der Waals (vdW) radius of 1.35 A, which is t ˚ pared with the vdW radius of hydrogen of 1.2 A interchain spacing from (110,200) reflection in t crystal as a function of the VDF content for P(V copolymer is presented in Fig (2,32,33) In t Table Experimental Lattice Dimensions and Angles of Form I (β) of PVDF and P(VDF-TrFE) 75/25 and 50/50 mol% Copolymers ˚ Lattice Dimensions (A) Lattice Angles (deg) VDF Content (mol%) a b c α β γ 100a 75b 50b 8.58 8.86 9.12 4.91 4.62 5.25 2.56 2.56 2.55 90 90 90.3 90 90 — 90 90 — a b Ref (25) Ref (30) 20 40 60 80 TrFE content (mol%) 100 Figure Variation of interchain lattice spacing with TrFE content in P(VDF-TrFE) copolymer films Open circles are from the ferroelectric β-phase and black dots are from the paraelectric phase structure of copolymers, it is to be noted that the net dipole of the polar cell is reduced by the opposing orientation of the “third” carbon–fluorine dipole on the TrFE monomer unit There have been several studies that have attempted to use molecular modeling to simulate the crystal structures as well as mechanical properties and polarizations of PVDF and P(VDF-TrFE) crystals (34–39) The primary focus of these studies have been on developing algorithms to predict the structures and properties of PVDF and P(VDF-TrFE) For the most part, these simulations have utilized molecular mechanics force fields to calculate the crystal lattice dimensions, polarizations, and compliances, and have yielded values in close agreement with results obtained from X-ray crystallographic studies For several model structures of P(VDF-TrFE) crystals, the increase in lattice dimensions with an increase in fluorine contents has been correctly predicted (37) For these same structures, the trends in the compliances and piezoelectric constants were calculated as a function of the molecular percentage of VDF and found to follow experimental measurements (38) PROCESSING AND FABRICATION There are two distinctly different ordering processes for PVDF and P(VDF-TrFE) polymers: crystalline ordering and dipolar ordering Both orderings can be influenced to a large extent by sample processing conditions Furthermore, one can also control the dipolar ordering by introducing defect structures in the crystalline phase As a consequence, the ferroelectric and electromechanical responses will depend substantially on the sample treatment conditions, such as annealing, quenching, mechanical drawing, irradiation, and doping (24,39) It should be pointed out that because of the slow kinetics of the various polymer transitional processes and the high-energy barrier of the transformation between the different crystalline forms, metastable phases can be formed and be present for long periods of time even though these phases are not the most ing these forms (41) The crystal phase that pred during processing is also controlled by the numb rine atoms in the polymer chains The increasin of fluorine atoms enhances unfavorable van der pulsive forces within the chains as well as betw cent chains These interactions force the energies α-phase chains with TGT G bonding, where repu tween closely positioned fluorine atoms are enh increase significantly relative to the β-forms This picture helps to explain why melts of P(VD crystallize directly into the electroactive polar β-p only are the crystal energies of β-phase P(VDF-T ered, but the energy barriers to these states are re contrast, when PVDF is annealed and slowly co the melt or cast from solutions of an organic solv as methyl ethyl ketone (MEK), formation of the rather than the β-phase is favored Although th lattice energy of the β-phase unit cell is slightly lo that of the α-phase cell, molecular dynamical cal have suggested that formation of the α-phase PV netically favored; that is, the barriers for format all-trans phase prevent crystallization directly phase from the melt When films of PVDF α-phas chanically stretched or drawn, the TGTG chain into the polar all-trans β-phase This happens to sis for the drawing procedure followed in most co processes for the preparation of piezoelectric PV Quenching of melted PVDF under high pressure o PVDF from hexamethylphosphoric triamide solu also result in the direct formation of polar β-pha Form III (γ -phase) PVDF can be produced by cas dimethylsulfoxide Polar form II (δ-phase) can be by poling α-phase PVDF in strong electric fields In the form I β-phase crystallites that predom ter PVDF is stretched or when the P(VDF-TrFE) c is cooled from the melt, there are ferroelectric crystallites that are polar but are nonetheless orie all crystallographically allowed directions Furt these crystallites are randomly oriented within This accounts for the absence of any piezoelectri unless the films are poled To be made piezoele domains must be oriented in a strong electric fi the “poling field.” Poling can be accomplished b ding the polymer surfaces with a metal, follow plication of a strong electric field to orient the cry An alternative method of poling is the use of a co charge where a corona charge is injected into the from a needle electrode placed a centimeter or two mechanically drawn to induce ferroelectric β-phase, PVDF films have to be oriented by stretching in one direction (uniaxial orientation) or in two directions (biaxial orientation) Multilayering of thin polymer films is another technique for fabricating thicker elements Multilayered and electroded polymer films that are subsequently wired in parallel have been proposed to reduce the electrical requirements for devices such as sonar drivers (7) POLARIZATION RESPONSES AND PHASE TRANSITIONS Phase Diagrams: An Overview With knowledge of the major crystalline phases of PVDF and its copolymers, it is useful to review its phase diagram showing the interconversions between the various crystalline forms where the relative proportions of the monomer units are varied The phase diagram of PVDF and P(VDF-TrFE) polymers shows a ferroelectricparaelectric transition that signals a change from a ferroelectric (polar) phase to a paraelectric (nonpolar) phase (Fig 8) The ferroelectric–paraelectric (F–P) transition temperature increases with vinylidene fluoride mole fraction content Below the F–P transition, the crystal is best represented as an ordered form I (β) structure with long sequences of all-trans bonds As the temperature of the crystals rises and goes through the F–P transition, an increasing number of gauche bonds is introduced into the ordered all-trans structure As a result, the crystal lattice dimensions enlarge, as we have seen above, and the crystal regions tend toward disorder, leading to the formation of the paraelectric phase containing a random mixing of TG, TG, TTTG, and TTTG Eventually, at higher temperatures the paraelectric phase passes through the melt transition One should note from Fig that PVDF and P(VDFTrFE) copolymers with high VDF concentrations not appear to possess distinct F–P transitions; rather, it appears that melting takes place before a F–P transition However, it must be mentioned that even in the ferroelectric phase, conformational defects can be introduced as the temperature of the polymers is raised These defects are introduced so subtly that they may not be apparent in thermal studies such as differential scanning calorimetry (DSC) Moreover, in P(VDF-TrFE), in addition to a low-temperature (LT) phase, where the chain conformation is predominantly alltrans, a cooled (CL) phase has been identified (38) Structural analysis indicated that the most probable structure of the so-called CL phase is a mixture of two disordered 40 60 80 Mol % VF2 Figure Phase diagram for VDF/TrFE copolymer melting point, ◦: ferroelectric to paraelectric transition crystalline phases, one trans planar and the oth lical (frozen-in high-temperature phase) (42–45) of this “frozen-in” disorder, copolymers with VDF below 50 mol% lose ferroelectricity Accordingly no clear phase transition signal as shown in diagram Polarization Responses and Phase Transitions In this section, we examine the ferroelectric PVDF and P(VDF-TrFE) in greater detail As tric materials, PVDF and its copolymers with TFE exhibit well-defined polarization hystere Figure 9(a) shows the polarization loop measu P(VDF-TrFE) 68/32 mol% copolymer stretched which, at low cyclic electric fields (60% mol% the F–P transition is fi Still the temperature range in which the electric induce the phase transition from nonpolar to pol depends strongly on the material For most of the ferroelectrics, this temperature range is relativel For instance, the range is about 8◦ C for BaTiO3 (a imately measured by the temperature range betw transition and critical temperature) (47) For P(VD copolymers, however, it has been reported that thi ature range exceeds 50◦ C as reported by a recent Langmuir-Blodgett film of P(VDF-TrFE) 70/30 m These results suggest that one may be able to the electromechanical response of P(VDF-TrFE) c significantly by operating the polymer near the F sition However, there are several issues associa the first-order F–P transition in P(VDF-TrFE) c that have to be addressed As has been shown in diagram (Fig 5), the F–P transition in all P(VD compositions occurs at temperatures higher th temperature The transition is also relatively sh a relatively narrow temperature range) In additi sirably large hysteresis has been observed for th mers at the first order F–P transition Therefore use of the unique opportunities near the first-o transition in P(VDF-TrFE) copolymer systems, th mer should be modified as to broaden the phase t E = 1.5 kV/cm 4.5 kV/cm (b) 10 k2 33 P (µC/cm2) 0.8 0.4 −5 100 −10 110 120 130 140 150 −2 70 Log (f) region as well as to move it to room temperature while minimizing hysteresis To modify the phase transition behavior in P(VDFTrFE) copolymer, one approach is to make use of highenergy irradiation This has been demonstrated in several earlier studies For example, Lovinger found that highenergy electron irradiation can convert the ferroelectric phase at room temperature to resemble a macroscopically paraelectric phase (84) Subsequently, Odajima et al and Daudin et al found that the sharp dielectric constant peak from the F–P transition can be broadened markedly and moved to near room temperature by irradiation (85,86) More recently, Zhang et al showed that by high-energy electron irradiation, the normal ferroelectric P(VDF-TrFE) copolymers can be converted into a relaxor ferroelectric with high electrostrictive strains (87–90) Figure 24 compares the polarization loops of P(VDFTrFE) 50/50 mol% copolymer before and after irradiation (40 Mrad of 2.5 MeV electrons at 120◦ C) As shown, the irradiation effectively eliminates the room temperature polarization hysteresis The dielectric data of the irradiated copolymer are presented in Fig 25, where clearly the broad room temperature peak moves to higher temperature as the measuring frequency increases In addition, it Figure 24 Polarization hysteresis loop of P(VDF 50 mol% copolymer film measured at room temperat fore and (b) after irradiation with electrons of 2.5 MeV 40 Mrad dose at 120◦ C 56 100 Hz Dielectric constant Figure 23 (a) Schematic of polarization as a function of temperature for a first-order ferroelectric-paraelectric transition under different dc bias field The first-order polar to nonpolar transition terminates at a critical point (b) electromechanical coupling coefficient (k33 ) as a function of temperature under different dc bias field near a first-order F–P phase transition The results are derived based on Landau-Devonshire theory using the parameter obtained on BaTiO3 −1 E (102 MV/m) Temperature (C) 310 320 42 330 T (K) 34 28 MHz 14 100 Hz 250 300 350 400 T (K) Figure 25 The dielectric constant (solid lines) and loss (dashed lines) as a function of temperature of P(V 50/50 mol% copolymer films irradiated with 40 Mrad using 2.55 MeV electrons The inset shows the fitting o Fulcher law Here the solid line is the fit and the circ data points ... -di-m-tolyl -1 , 1 -biphenyl-4,4 -diamine (TPD) 27 (79), N,N -1 - naphthyl-N,N -diphenyl -1 , 1 -biphenyl-4,4 -diamine (NPD) 28 (80), 1, 3,5-tris(2 -anthracyl-4 -methoxyphenylamino)benzene 29 ( 81) , and the... nonlinear and time-depen nomena that characterize the behavior of pie polymers still remain unexplained Y2, Pa 3 .16 × 10 9 3 .14 × 10 9 3 .12 × 10 9 3 .1 × 10 9 APPLICATIONS 3.08 × 10 9 t, sec × 10 6 × 10 6 × 10 6... 10 12 F/m 12 ? ?13 Y1 = 2.56 × 10 9 Pa; Y2 = 2.6 × 10 9 Pa (σ y )1 = 4.5 × 10 7 Pa; (σ y )2 = 3.9 × 10 7 Pa (ε y )1 = 1. 8%; (ε y )2 = 1. 4%; (σ u )1 = 3.5 × 10 8 Pa; (σ u)2 = × 10 7 Pa (εu )1 = 16 .9%; (εu)2