SIMULTANEOUS 3D PRINTING AND POLING OF PVDF AND ITS

Subscriber access provided by Kaohsiung Medical University

Article

SIMULTANEOUS 3D PRINTING AND POLING OF PVDF AND ITS NANOCOMPOSITES Sampada Bodkhe, P.S.M. Rajesh, Frederick P Gosselin, and Daniel Therriault ACS Appl. Energy Mater., Just Accepted Manuscript • Publication Date (Web): 25 May 2018 Downloaded from http://pubs.acs.org on May 25, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Energy Materials

SIMULTANEOUS 3D PRINTING AND POLING OF PVDF AND ITS NANOCOMPOSITES Sampada Bodkhea,b, P. S. M. Rajesha,c, Frederick P. Gosselina and Daniel Therriaulta* AUTHOR ADDRESS a

Laboratory for Multiscale Mechanics, Department of Mechanical Engineering,

Centre for Applied Research on Polymers and Composites (CREPEC) Polytechnique Montreal, C.P. 6079, succ. Centre-Ville, Montreal, QC H3C 3A7, Canada b

c

Present address: ETH Zurich, Leonhardstrasse 21, 8092 Zurich, Switzerland

Present address: Siemens Gamesa Renewable Energy, A/S, Assensvej 11, 9220 Aalborg, Denmark.

*Corresponding author: [email protected] KEYWORDS 3D printing, in-situ poling, gait-analysis, piezoelectric, polyvinylidene fluoride

ABSTRACT Given the ever-increasing demand for customization and miniaturization, in-situ three-dimensional (3D) printing of piezoelectric polymers comes as an efficient means to cater to smart structures via multimaterial printing. Applying our hybrid printing technique to polyvinylidene fluoride (PVDF) - barium titanate (BaTiO3) nanocomposites to combine the fabrication and high voltage poling steps shrinks the manufacturing time, overcomes the disadvantages of adherence in non-conformal piezoelectric films or fabrics, and increases the sensitivity and scope for on-demand applications. A remarkable 300%

ACS Paragon Plus Environment

ACS Applied Energy Materials 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

improvement in piezoelectric charge output is achieved, upon the addition of 10 wt.% BaTiO3 nanoparticles and application of an electric field of 1 MV m-1, over printed unpoled neat PVDF. Ultimately, we demonstrate the application of the in-situ poling process in the form of sensors printed directly on a shoe insole for gait-analysis. The sensors fabricated in this work effectively distinguish between walking and stamping both as portable in-shoe sensor as well as sensors attached to the ground. INTRODUCTION The most piezoelectric polymer polyvinylidene fluoride (PVDF), and its nanocomposites are being applied towards actuation, energy harvesting,1 energy storage,2 and sensing.3 Traditionally, PVDF-based structures were fabricated in the form of fibers or films using extrusion,4 melt spinning,5 spin coating,6 solution casting,7 and electrospinning.8 Last few years have seen a strong impetus in creating three-dimensional (3D) structures with PVDF as it: (i) promotes flexibility in shape, (ii) boosts the piezoelectric coupling factor by increasing the extractable strain in the structure,9 and (iii) aids in fabricating on-demand and customized structures. Sensors printed directly on the surface to be sensed not only increase sensitivity via overcoming the interfacial material stiffness but also reduce the number of manufacturing steps and provide ability to produce sensors on flat as well as curved surfaces. 3D printing of PVDF and its nanocomposites has been attempted by various techniques.10, 11, 12 Of these, near-field electrospinning (NFES)10, 13 and solvent evaporation-assisted 3D printing (SEA3DP)11,

14

have been successful in creating multi-layered,10 self-supported,11 conformal,14 and

freestanding structures.14 Another technique involves integrated 3D printing via fused deposition modelling followed by corona poling.15 This technique is time consuming compared to NFES and SEA-3DP as it involves printing of a single layer followed by poling of the entire layer in a pointby-point fashion following the print line.15 Where NFES involves application of large electric fields during the deposition of PVDF-based solutions to attain the piezoelectric polarization, SEA-3DP of barium titanate nanoparticles (BaTiO3 NPs) and PVDF relies on the interaction 2


Page 2 of 29

Page 3 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


between the nanoparticles and the polymer, and the dipole alignment via high shear forces during high pressure extrusion11 In our technique (SEA-3DP), a highly volatile solution of the piezoelectric nanocomposite (NC) is extruded and deposited with the help of a robotic deposition system to form the required shape. The effect of nanoparticle size and their concentration in the nanocomposite are subjective to the processing and fabrication techniques and hence, must be optimized each time there are any changes in the fabrication process.16, 17 We have previously surveyed and investigated the optimal process parameters for SEA-3DP: the specific choice of materials, concentrations and processing technique in our previous work11. Three different techniques (i.e., ball-milling, extrusion and sonication) were investigated to incorporate between 1 and 15 wt.% of BaTiO3 NPs into PVDF. It was demonstrated that a BaTiO3 NP concentration of 10 wt.% added to PVDF via ball-milling was the ideal recipe to achieve superior piezoelectric properties under our printing conditions. Apart from increasing the polar β-phase content (responsible for piezoelectricity in PVDF) by 33% as compared to the solution-cast films, SEA-3DP has shown to result in piezoelectric properties (d31 = 18 pC N-1) comparable to commercial (Measurement Specialties) poled PVDF films (d31 = 23 pC N-1).18 This value is much higher than the d31 of PVDF/BaTiO3 films produced by fused deposition modelling combined with corona poling (0.048 pC/N).15 However, it is compelling to investigate the effect of simultaneous application of electric field during printing of PVDF and NC; as such, a hybrid technique would be a worthy step in the direction of single-step manufacturing of sensors and serve well the requirement of on-demand application. In-situ poling was realized by applying an electric field between the metallic nozzle of the syringe and a metallic tape attached to the substrate. We present a detailed investigation of the effect of NP addition and electric field applied on the β-phase and the piezoelectric properties of PVDF. The content of β-phase is characterized via X-ray diffraction (XRD), and Fourier transform infrared

3



Page 4 of 29

spectroscopy (FTIR). We also designed a drop-weight test to determine the piezoelectric properties of the films. Furthermore, we present an application in the form of ready-to-use sensors on a shoe insole for gait analysis. To date, gait sensors are usually fabricated by adhering piezoresistive19 or piezoelectric films20 to fabrics. Printing of sensors directly on the insoles results in higher sensitivity as well as compatibility with different shapes of insoles required for individuals with special needs. The sensors printed at fore and hind foot locations on insoles were tested to distinguish differences between walking and stamping while worn in the shoe. Similar experiments were also performed by fixing the insole to the ground. Printing of piezoelectric PVDF has so far been shown only on flat surfaces. Here, for the first time we show the 3D conformal printing of piezoelectric PVDF on a curved surface with in-situ poling for enhanced performance. Conformal printing of active sensors facilitates customized sensors catering to individual needs in areas from aerospace to prosthetics. EXPERIMENTAL SECTION Materials PVDF (Mw~534,000), dimethyl sulfoxide (DMSO), and silver conductive paste (micro-particles dispersed in α-terpineol) were purchased from Sigma Aldrich®. Solvents N, N-dimethyl formamide (DMF) and acetone were obtained from Alfa Aesar and BDH, respectively. Barium titanate NPs (BaTiO3; 99.9% purity, mean diameter ~100 nm) were procured from Nanostructured & Amorphous materials Inc. Ink preparation The solvent system consisted of a mixture of DMF and acetone. 11,21 Amongst the solvents used for PVDF, DMF is the least hazardous and possesses the lowest boiling point of ~153 °C,22 while acetone is an anti-solvent with a low evaporation point (~56 °C). An optimized ratio of 40% DMF 4


Page 5 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


and 60% acetone in the solvent system was used.23 A contribution of 65 g L-1 dimethyl sulfoxide (DMSO) aided in initiating β-phase.24 The required mass of PVDF was sonicated in the solvent mixture for 20 min to obtain a clear PVDF solution. To form the nanocomposite, a process explained in the previous work11 was used where 10 wt. % of BaTiO3 NPs were incorporated into PVDF via ball-milling with the solvents for 20 min. SEA-3D printing PVDF and its nanocomposite solutions (referred to as ‘inks’ hereafter) were poured into 3 mL syringe barrels. The syringe barrels were placed into a pneumatically operated dispensing system (HP-7X, EFD) to apply precise pressures for printing. The dispensing system was held by a robotic-arm (I&J2200-4, I&J Fisnar Inc.), controlled by a commercial software (JR Points for Dispensing, Janome Sewing Machine) that further enabled the deposition of the inks on a movable flat stage. The design to be printed was uploaded into the software in the form of a series of X, Y and Z coordinates. The in-situ poling set-up (Figure 1) consisted of laboratory DC (direct current) power sources to apply the required electric field between a metallic printing nozzle and a metallic tape (aluminum or copper, 3M Canada) attached to the substrate. Single-sided aluminum tape was used to print freestanding films while sensors were printed on double-sided copper tape. Based on the electric field values, two DC power sources were employed: Gw Instek, GPS 3303 for electric fields up to 0.3 MV m-1 and the Fisher Scientific, FB 3000 for 1 MV m-1. Beyond this field the printed films underwent breakdown. We fabricated a total of 8 types of films with 4 different poling voltages: 0, 0.1, 0.3 and 1 MV m-1. The labelling of these films consists of the poling voltage in MV m-1 followed by letters: P for neat PVDF and NC for the nanocomposite. For example 0.3P stands for a PVDF film fabricated with 0.3 MV m-1 poling voltage. Films for XRD, FTIR spectroscopy, and piezoelectric tests were fabricated with 0.2 g mL-1 of PVDF and the NC solutions by keeping the

5



printing parameters constant: nozzle diameter = 100 µm nozzle, extrusion pressure

Page 6 of 29

~1 MPa, and

stage velocity = 20 mm s-1. We printed 3D structures using both PVDF and nanocomposite inks with a concentration of 0.25 g mL-1. For conformal printing, copper (Cu) tape was stuck on a convex hemispherical surface as the bottom electrode and a NC filament was printed on the tape from the lower end of the hemisphere to its apex with an electric field of 0.3 MV m-1. Characterization Scanning electron microscopy (FE-SEM; JEOL JSM-7600TFE) was used to obtain the microscopic images of PVDF and NC films at an accelerating voltage of 10 kV. A Philips X’pert diffractometer was used to obtain the diffraction spectra between 16 to 28° at a rate of 0.4 °min-1 with a copper target (Kα radiation) at 50 kV and 40 mA. A FT–IR spectrometer (Perkin Elmer, Spectrum 65) was used to obtain the absorption spectra of PVDF and its nanocomposite films in the range of 500 – 4000 cm-1. Each sample was scanned 32 times with a resolution of 4 cm-1. Optical microscopy (BX-61 Olympus; coupled with Image-Pro plus V5, an image processing software from Media Cybernetics) was used to study the morphology of the 3D printed structures. Sensors were fabricated from the printed films by painting silver in a circular area (diameter

~7

mm) on both sides of the films. An LCR (inductance-capacitance-resistance) meter (Wavetek Metermen, LCR55) was used to measure the capacitance of the films. Foot insole tests Sensors (30 × 30 mm2) were printed on a foot insole (Bayer Inc., Dr. Scholl’s) at the metatarsal and heel region using the NC ink (Figure 2). The detailed systematic sensor fabrication is shown in Figure S1. A double-sided Cu tape placed on the insole formed the bottom electrode. The sensors were then printed (~1 MPa, 20 mm s-1 and 100 µm nozzle diameter) on the Cu tape with an electric field of 0.3 MV m-1. After printing the sensors, silver was painted on the top layer of 6


Page 7 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the sensors to form the second electrode. Leads were connected to both the bottom and the top electrodes using Cu tape. Tests were carried out by stamping the foot on the ground, and walking: with the insole worn in a shoe and insole placed directly on the ground. The subject weight was approximately 450 N (45 kg).

RESULTS AND DISCUSSION Scanning electron microscopy The SEM images of the PVDF and NC films are shown in Figure 3a, and b respectively. No discernable differences were observed in the overall morphology between the PVDF and NC specimens fabricated using SEA-3DP. Both samples showed well-integrated spherulite structures. However, the void size in the NC seemed slightly larger while not affecting the net void content. The BaTiO3 NPs are circled in yellow; their dispersion and distribution in the PVDF matrix are highly satisfactory and reapprove the process methodology adopted in this study.11 The sub-micron sized uniformly distributed agglomerates actually serve to retain the β-phase in PVDF.11, 25 Given the low electric fields employed in our work, no change in the dispersion/distribution characteristics of the BaTiO3 NPs should be expected upon poling as reorienting the NP would in turn require very high fields.26 X-ray diffraction Of all the polymorphs of PVDF, the α and δ polymorph possess a trans-gauche (TGTG’) configuration, while the β and ϒ are found in all-trans (TTTT) planar zig-zag form.27 Except the α-phase, in which PVDF commonly crystallizes upon solidification, all others are unstable in nature. The unstable all-trans β-phase is fundamental for pyro, piezo and ferroelectric properties in PVDF. XRD identifies various crystalline phases of PVDF. Peaks corresponding to α-phase are found at 2-theta values of 17.7, 18.4, 19.9 and 26.56° while that of γ-phase are at 20.04 and 26.8°.28, 29 The β-phase is characterized by a well-defined peak at 20.26°,28, 29 peak-broadening at 20.1° and diminishing of α-peaks at 18.4 and 26.7°.30 Figure 4a shows the XRD patterns of the 7



Page 8 of 29

PVDF films fabricated with different electric fields used for poling. The β-peak at 20.1° has considerably broadened in case of the films prepared with the application of electric field. The αpeak at 18.4° is present in all the films, but is less pronounced in PVDF fabricated without any electric field. The same trend is observed for the NC films (Figure 4b); the film printed without any electric field features a broad β-peak whereas all other films possess an α-peak at 18.4°. The peak at 22.15° is the characteristic peak of BaTiO3 NPs.31 It should be noted that no peaks other than those corresponding to α and β-phases of PVDF were observed in our work. The XRD results reveal that the addition of NPs to PVDF resulted in the disappearance of the α-phase. Moreover, surprisingly application of the electric field led to the formation of α-phase in all the films. Fourier transform infrared spectroscopy (FTIR) To determine how the appearance of α-phase in the films fabricated via the application of electric field affected the overall β-phase content in PVDF we carried out FTIR analysis on the films. The characteristic absorption bands for α-phase are close to 490, 530, 615, 766, 855 and 974 cm-1.32 Though absorption bands at 510/512, 840/833 cm-1 for β and γ-phase, respectively, lie very close to each other, the band at 1279 cm-1 are only present in the case of β-phase.29, 32 Figure 5 presents the FTIR spectra for all the films fabricated in this work. As also seen from in XRD, the peaks corresponding to the α-phase in PVDF seem to be present in all the films except those fabricated without the application of an external electric field. In addition, all the films possess the peak at 1279 cm-1 confirming the absence of any γ-phase. Hence, the FTIR analysis confirms the conclusions from the XRD that α-phase appears with the application of electric field. Given that no third phase was present in the films, the relative amount of β-phase amongst α and β-phases of PVDF is calculated with the help of the Beer-Lambert law, which relates the

8


Page 9 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


absorption to the concentration of the species. The modified Beer-Lambert law for obtaining the fraction of β-phase (Fβ) is given by the equation below:33

F =

. α

(1)

where, Aα and Aβ are the absorption fractions of α and β-phases at 763 cm-1 and 840 cm-1, respectively, and 1.26 is the ratio of absorption coefficients 7.7 × 104 and 6.1 × 104 cm2 mol-1 of β and α-phases, respectively. The Fβ calculated using Equation 1 for all the films is presented in Figure 6. The average values of the Fβ in the films is found to be 71% ±5 %. Thus, poling voltages did not have any substantial influence on the overall β-phase fractions.34 As there is no prominent difference between the content of β-phase, the piezoelectric properties in PVDF or NC would then be dominated by the amount of oriented β-phase, i.e., the net dipole moment in the material.35, 36, 37 Piezoelectric characterization To experimentally compare the piezoelectric performance of the printed materials, we designed a drop-weight test. The test method consisted of impacting each sensor with a 10 g calibration weight from a height of 10 cm as shown in the schematic in Figure 7. The weight was dropped through a hollow cylinder so that the sensor was always impacted with the same force and energy. The inner diameter of the cylinder had a small tolerance with respect to the outer diameter of the weight to avoid any friction between the two surfaces during the fall. The piezoelectric charge output from the sensors upon the impact was converted into voltage through a charge amplifier (MEAS specialties) and was recorded using LabVIEW 2014. The voltage, V (V) obtained from the sensors was converted in to charge, q (C) and then specific charge output, QS (C m-2) by using the following equations:

=

=

(2) (3) 9



where, A is the electrode area in m2, and C is the capacitance (F) measured by an LCR meter. Figure 8a shows the averaged specific charge output for five different sensors of each type. It can be seen that addition of BaTiO3 NPs led to increased charge output in the sensors. Specific charge output was significantly higher in specimens fabricated with higher electric field. Filler addition and an electric field of 1 MV m-1 led to an increase of

~300% over printed unpoled unreinforced

PVDF. It is important to investigate the source of this significant raise in specific charge obtained after poling with a field of just 1 MV m-1. Moreover, the results also varied considerably between different specimens fabricated at a given voltage. One specimen per poling voltage was randomly selected from the batch of five specimens fabricated and tested previously (i.e., those comprising results in Figure 8a) and retested after one year. For the retest, we dropped the weight 5 times consecutively on the same sensor and recorded the charge output as shown in Figure 8b. Clearly, the response of the sensors in Figure 8a (obtained shortly after printing) diminished a year later (Figure 8b). Such a decrease has two possible explanations: (i) incomplete dipole orientation leading to a loss in polarization over time, and/or (ii) interfacial charging at/around the pores and PVDF-BaTiO3 boundaries that also discharge over time. The fact that the amount of β-phase remained consistent between specimens poled with different field strengths indicates that the poling fields employed in the study were insufficient to achieve complete poling. It is important to note that unless complete poling is achieved, the specimens will depolarize over time. Interfacial charging is also a certain possibility as discussed in the following paragraph. The form and distribution of porosity in the microstructures of the specimens (Figure 3), is expected to be stochastic and varies greatly between specimens. Similarly, for BaTiO3 reinforcement, it is known that pores and interfaces are sites of charge development and the high values of error in specific charge could also be attributed to these variable quantities. The exact ratio of charge output pertaining to poling and interfacial charging at the time of fabrication is difficult to estimate with the characterization methods 10


Page 10 of 29

Page 11 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


adopted in this study. However, given a year of elapsed time, it may be reasonable to believe that the charge output shown in Figure 8a is largely owing to insufficient dipole orientation as otherwise that owing to interfacial charging would have decreased exponentially with time. 38 The percentage error of the data in Figure 8b has also considerably decreased after one year with the highest being 16% indicating diminished influence of interfacial charges in the output. Although the films fabricated from the NC at 1 MV m-1 exhibited the highest specific charge output, the error range is very high and hence, all further experiments were carried out with the NC films fabricated at 0.3 MV m-1. The time window available for application of these sensors is presently tested for 1 year, where the sensors were functional but with decreasing output. Thus, short-term applications that can use qualitative results are deemed fit to demonstrate the usefulness of these sensors. Gait monitoring, for example in sports kinesiology for event training or monitoring recovery from an injury, is often required over a period of several weeks. In this case qualitative estimation of location of forces, speed of motion, and loading direction etc. are of interest, and must be tailored to individual needs. Gait analysis via 3D-printed conformal piezoelectric sensors was thus chosen for demonstration. Gait monitoring sensor Gait analysis is mostly carried out in two ways: (i) when the subject walks on a sensor that is fixed to the ground,39,

40

and (ii) when the sensor is placed inside the subject’s shoe.3,

41, 42

Piezoelectric sensors were printed on a shoe insole at the fore foot (metatarsal) and hind foot (heel) locations as seen in Figure 9. These are the locations which experience the highest forces during locomotion.43,

44

Figure 2a shows a schematic of the sensor to reveal the components

clearly. Figure 2b is a photograph of the actual insole with the sensors and the electrical connections. Figure 2c shows the insole placed in an open shoe that was worn by the subject during the following tests. Experiments were conducted in two ways: (i) by placing the insole in the subject’s shoe and (ii) testing barefoot when the insole was fixed to the ground. Figure 9 11



Page 12 of 29

presents the signals of the fore and the hind sensors when the insole was worn in the shoe. During a normal walking motion (seen from insets in Figure 9a), the hind foot lifts first, leaving the fore foot still in contact with the floor showing a small peak in the fore foot voltage. The voltage goes close to zero as the entire foot is lifted. In the next cycle the fore foot makes the initial contact with the floor, and then the hind foot lands on the floor.45 Thus, the voltage response (Figure 9a and 9c) from the fore foot sensor shows a larger sharp peak while touching the ground and a smaller peak upon release. The larger negative peak occurs when the shoe hits the ground whereas the smaller negative peak is the response from the pressure between the foot and the shoe during the swing movement upon the foot’s removal. The broader and flatter response from the hind foot when the foot is on the ground shows the portion where the hind foot provides stability during locomotion. As opposed to walking, stamping (Figure 9b) involves the application of the force more at the hind foot location (sharp peaks in the voltage response) as compared to the fore foot as explained by the curves and images in Figure 9b and d, respectively. Here, the response from the hind foot is lower in magnitude but sharper than that from the fore foot. The responses are in coherence with those found in the literature.46, 47 The voltage outputs are further comparable with those found for plantar foot pressure sensors formed by gluing piezoelectric fiber arrays fabricated via NFES.48 The electric voltage for poling used in this case was 1500 V as opposed to 100 V in our work. Section 2 in SI shows the voltage response from the sensors when the insole was fixed to the ground as a floor-based gait sensor.49 The voltage responses are much higher in magnitude in this case as compared with those when the insole was worn in the shoe. This could be because the ground reaction forces acting on the foot are much reduced when the insole is in the shoe due to the cushioning effect of the shoe padding. For the same reason the response directions could be reversed in case of the hind foot sensor. Thus, depending on the type of the response and its magnitude, the sensors fabricated in this work can be applied to monitor gait. Gait sensors made 12


Page 13 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


from piezoelectric materials act as active sensors, not requiring any external power sources for operation as compared to their present resistive counterparts.3, 42, 50 3D printing of structures 3D printing of polymer solutions strongly depends on the solution viscosity. We carried out a comprehensive characterization of the rheological behavior of PVDF and the NC solution11 for solvent-evaporation based printing methods (details and test results in SI). The viscosity study helped establish a guideline in terms of solution concentrations required to print certain structures. It was found that the concentration of 0.2 g mL-1 is most suitable to form void free thin films while certain concentrations below this threshold are too runny to print. As the concentration is increased beyond 0.25 g mL-1, multi-layered structures can be fabricated from both PVDF and its NC inks. Figure 10 shows two types of 40-layered structures fabricated using both the PVDF and its NC inks (Figure 10 a, b, d and e). Figure 10c and 10f are self-supporting 3D scaffolds with distinct adjacent layers. The PVDF scaffold was printed with 0.275 g mL-1 solution while NC scaffold was printed with 0.3 g mL-1 solution. It is because the addition of NP to PVDF resulted in a decrease in the mass of the polymer in the solution, retarding the evaporation of the solvents. Both the polymer and its nanocomposite can thus be used to fabricate piezoelectric 3D structures. Conformal sensors The ability to fabricate conformal sensors is further demonstrated in Figure 11a. Here, a sensor is printed on a hemisphere (diameter = 30 mm). When gently pressed with the finger as shown in Figure 11a, the sensor generated voltages up to 7 V (Figure 11b, peak to peak

~8V). The voltage

peaked on touching the sensor. Upon removal of the force, the charges started flowing in the negative direction. This is believed to be the first working example of conformally printed piezoelectric sensors. Till date piezoelectric sensors are glued to the substrates to be sensed.3, 41, 42, 43,49

It is for the first time, that we demonstrate the printing of piezoelectric sensors directly on the 13



Page 14 of 29

sensed surface. This not only avoids imperfect load transfer from the sensed material to the sensor but also gives us the flexibility to print on any form without loss in signal strength. CONCLUSION Simultaneous 3D printing and electrical poling process was demonstrated for PVDF and its nanocomposite containing 10 wt.% of BatiO3 NPs with electric fields up to 1MV m-1. This combination produced a ~300% increase in specific charge output in the poled nanocomposite over unpoled neat PVDF. The sensor response decreased a year after fabrication due to incomplete poling at the low electric fields used and discharging of interfacial charges. The poling voltages were limited as the films underwent frequent breakdown at a field of 1 MV m-1. It is thus imperative to improve the fabrication technique to allow the use of higher electric fields with close control on porosity and, dispersion and distribution of the reinforcement. The effect of temperature and porosity on the piezoelectric performance of the sensors also needs to be investigated in detail. Different 3D structures, both self-supported and conformal were realized to demonstrate the flexibility and scope of the printing method. Gait-monitoring sensors printed on a shoe insole are a very convenient option as the same insole could be switched between various shoes,42 and used to collect data for a longer duration whereas the platform sensors could be used in a test facility for multiple users. The sensors effectively differentiated between walking and stamping for both the front and hind feet locations as depicted by the varying voltage response in different cases. The technique developed in this work can be applied to fabricating sensors for structural health monitoring of composites used in aerospace, automotive and biomedical industry.

14


Page 15 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. Schematic of in-situ poling set-up for SEA-3DP.

Figure 2. (a) Schematic of the sensors printed on a shoe insole (not to scale), (b) Photograph of the sensors printed on the shoe insole and (c) the shoe insole placed into a shoe.

15



Page 16 of 29

Figure 3. SEM images of (a) P and (b) NC films fabricated without any electric field (scale bar = 10 µm).

16


Page 17 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4. X-ray diffractograms of (a) P and (b) NC films fabricated with an electric field of 0, 0.1, 0.3 and 1 MV m-1.

Figure 5. FTIR spectra of (a) P and (b) NC films fabricated with an electric field of 0, 0.1, 0.3 and 1 MV m-1.

17



100 80 Fβ (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 29

60 40 P NC

20 0 0

0.5 Poling field (MV m-1)

1

Figure 6. Fraction of β-phase in various films calculated using Beer-Lambert Law. .

Figure 7. Schematic of drop-weight test setup for piezoelectric characterization of the sensors.

18


Page 19 of 29

(b) 3

3 2.5

Specific charge output (pC/mm2)

(a) Specific charge output (pC/mm2)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


PVDF NC

2 1.5 1 0.5

PVDF

NC

2.5 2 1.5 1 0.5 0

0 0

0.1 0.3 Polarizing field (MV/m)

1

0

0.1 0.3 Polarizing field (MV/m)

1

Figure 8. Piezoelectric charge outputs of P and NC sensors fabricated with electric fields: 0, 0.1, 0.3 and 1 MV m-1. (a) Average of 5 different samples, and (b) average of a single sample tested 5 times.

19



Page 20 of 29

Figure 9. Piezoelectric voltage output from the sensor when the insole was worn in the shoe. Single cycle output while (a) walking and (b) stamping. Insets shows the photographs of the movements. Output from the sensor for three consecutive cycles while (c) walking and (d) stamping.

20


Page 21 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 10. Optical images of 3D structures fabricated with P (a) a 40 layer circular spiral (0.25 g mL-1 PVDF solution; 0.5 MPa; 1 mm s-1; 100 µm); (b) a 40 layer square spiral (0.25 g mL-1 PVDF solution; 0.5 MPa; 1 mm s-1; 100 µm); (c) a 9 layer scaffold (0.3 g mL-1 PVDF solution; 0.2 MPa; 3 mm s-1; 150 µm). Structures fabricated with NC: (d) a 40 layer circular spiral (0.25 g mL-1 NC solution; 0.5 MPa; 1 mm s-1; 100 µm); (e) a 40 layer square spiral (0.25 g mL-1 NC solution; 0.5 MPa; 1 mm s-1; 100 µm); (f) a 9 layer scaffold (0.3 g mL-1 NC solution; 1.3 MPa; 13 mm s-1; 250 µm).

21



Page 22 of 29

Figure 11. (a) Optical images of a conformal sensor fabricated with the NC pressed by an index finger. (0.2 g mL-1 PVDF solution; 0.15 MPa; 2 mm s-1; 100 µm, 0.3 MV m-1). (b) Voltage output of the sensor when pressed with the finger 3 times.

22


Page 23 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ASSOCIATED CONTENT Supporting Information. The Supporting Information is available free of charge via the Internet at http://pubs.acs.org.” and DOI: Printing sensors on the shoe insole, Tests with shoe insole fixed to the ground, and Processviscosity characterization of PVDF nanocomposites by ball-milling (PDF) AUTHOR INFORMATION Corresponding Author *[email protected] Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. ACKNOWLEDGMENT The authors acknowledge the financial support from NSERC #RGPIN3212568-2013 (Natural Sciences and Engineering Research Council of Canada) and Canada Research Chair #CRC950-216943. The authors thank Ms. Gabrielle Turcot for the help in printing 3D structures for this work.


23


Page 24 of 29

REFERENCES (1) Wan, C.; Bowen, C. R. Multiscale-Structuring of Polyvinylidene Fluoride for Energy Harvesting: The Impact of Molecular-, Micro- and Macro-Structure. J. Mater. Chem. A 2017, 5, 3091-3128. (2) Xie, L.; Huang, X.; Yang, K.; Li, S.; Jiang, P. "Grafting to" Route to Pvdf-Hfp-Gma/Batio3 Nanocomposites with High Dielectric Constant and High Thermal Conductivity for Energy Storage and Thermal Management Applications. J. Mater. Chem. A 2014, 2, 5244-5251. (3) Li, Y.; Samad, Y. A.; Liao, K. From Cotton to Wearable Pressure Sensor. J. Mater. Chem. A 2015, 3, 2181-2187. (4) A. Ferreira, P. C., H. Carvalho, J. M. Nobrega, V. Sencadas; Lanceros-Mendez, a. S. Extrusion of Poly(Vinylidene Fluoride) Filaments: Effect of the Processing Conditions and Conductive Inner Core on the Electroactive Phase Content and Mechanical Properties. J. Polym. Res. 2011, 18, 8. (5) Lund, A.; Hagström, B. Melt Spinning of Poly(Vinylidene Fluoride) Fibers and the Influence of Spinning Parameters on Β-Phase Crystallinity. J. Appl. Polym. Sci. 2010, 116, 2685-2693. (6) Seok Ju, K.; Youn Jung, P.; Sung, J.; Jo, P. S.; Cheolmin, P.; Kim, K. J.; Cho, B. O. Spin Cast Ferroelectric Beta Poly(Vinylidene Fluoride) Thin Films Via Rapid Thermal Annealing. Appl. Phys. Lett. 2008, 92, 012921-012921-012923. (7) Ferreira, J. C. C.; Monteiro, T. S.; Lopes, A. C.; Costa, C. M.; Silva, M. M.; Machado, A. V.; Lanceros-Mendez, S. Variation of the Physicochemical and Morphological Characteristics of Solvent Casted Poly(Vinylidene Fluoride) Along Its Binary Phase Diagram with Dimethylformamide. J. Non-Cryst. Solids 2015, 412, 16-23. (8) Seoul, C.; Kim, Y.-T.; Baek, C.-K. Electrospinning of Poly(Vinylidene Fluoride)/Dimethylformamide Solutions with Carbon Nanotubes. J. Polym. Sci., Part B: Polym. Phys. 2003, 41, 1572-1577. (9) Yiin-Kuen, F.; Hsi-Chun, H.; Bo-Sheng, W.; Shan-Chien, L. All-Fiber Transparent Piezoelectric Harvester with a Cooperatively Enhanced Structure. Nanotechnology 2016, 27, 435403. (10) Luo, G.; Teh, K. S.; Liu, Y.; Zang, X.; Wen, Z.; Lin, L. Direct-Write, Self-Aligned Electrospinning on Paper for Controllable Fabrication of Three-Dimensional Structures. ACS Appl. Mater. Interfaces 2015, 7, 27765-27770. (11) Bodkhe, S.; Turcot, G.; Gosselin, F. P.; Therriault, D. One-Step Solvent EvaporationAssisted 3d Printing of Piezoelectric Pvdf Nanocomposite Structures. ACS Appl. Mater. Interfaces 2017, 9, 20833-20842. (12) Lee, C.; Tarbutton, J. A. Electric Poling-Assisted Additive Manufacturing Process for Pvdf Polymer-Based Piezoelectric Device Applications. Smart Mater. Struct. 2014, 23, 095044. (13) Pan, C.-T.; Yen, C.-K.; Wu, H.-C.; Lin, L.; Lu, Y.-S.; Huang, J. C.-C.; Kuo, S.-W. Significant Piezoelectric and Energy Harvesting Enhancement of Poly(Vinylidene Fluoride)/Polypeptide Fiber Composites Prepared through near-Field Electrospinning. J. Mater. Chem. A 2015, 3, 6835-6843. (14) Bodkhe, S.; Noonan, C.; Gosselin, F. P.; Therriault, D. Coextrusion of Multifunctional Smart Sensors. Submitted to Advanced Materials Technologies 2018. (15) Hoejin, K.; Fernando, T.; Yanyu, W.; Dino, V.; Yirong, L.; Tzu-Liang, T. Integrated 3d Printing and Corona Poling Process of Pvdf Piezoelectric Films for Pressure Sensor Application. Smart Mater. Struct. 2017, 26, 085027.


24

Page 25 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(16) Bhattacharya, M. Polymer Nanocomposites—a Comparison between Carbon Nanotubes, Graphene, and Clay as Nanofillers. Materials 2016, 9, 262. (17) Camargo, P. H. C.; Satyanarayana, K. G.; Wypych, F. Nanocomposites: Synthesis, Structure, Properties and New Application Opportunities. Mater. Res. 2009, 12, 1-39. (18) Division, S. P. Piezo Film Sensors Technical Manual; Measurement Specialties, Inc: Norristown, PA 1999. (19) Lou, C.; Wang, S.; Liang, T.; Pang, C.; Huang, L.; Run, M.; Liu, X. A Graphene-Based Flexible Pressure Sensor with Applications to Plantar Pressure Measurement and Gait Analysis. Materials 2017, 10, 1068. (20) Li, C.; Min, S.; Yiyang, L.; Jinchuan, H.; Haohua, L. In Research on Gait Measuring Method Based on Pvdf Piezoelectric Sensors, 2013 Third International Conference on Instrumentation, Measurement, Computer, Communication and Control, 21-23 Sept. 2013; 2013; pp 284-287. (21) Koombhongse, S.; Liu, W.; Reneker, D. H. Flat Polymer Ribbons and Other Shapes by Electrospinning. J. Polym. Sci., Part B: Polym. Phys. 2001, 39, 2598-2606. (22) A13547 N,N-Dimethylformamide, 99%. (R), A. A., Ed. (23) Bodkhe, S.; Gosselin, F. P.; Therriault, D. In Three-Dimensional Printing of Polyvinylidene Fluoride Nanocomposites, 20th International Conference on Composite Materials, Copenhagen, Denmark, 19-24th July 2015 International comittee on composite materials: Copenhagen, Denmark, 2015; p 9. (24) Benz, M.; Euler, W. B.; Gregory, O. J. The Role of Solution Phase Water on the Deposition of Thin Films of Poly(Vinylidene Fluoride). Macromolecules 2002, 35, 2682-2688. (25) Niu, Y.; Yu, K.; Bai, Y.; Wang, H. Enhanced Dielectric Performance of Batio3 Pvdf Composites Prepared by Modified Process for Energy Storage Applications. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control 2015, 62, 108-115. (26) Takashi, Y.; Kazuyori, U.; Hisao, B. Batio 3 Particle-Size Dependence of Ferroelectricity in Batio 3 /Polymer Composites. Jpn. J. Appl. Phys. 1993, 32, 4272. (27) Sessler, G. M. Piezoelectricity in Polyvinylidenefluoride. The Journal of the Acoustical Society of America 1981, 70, 1596-1608. (28) Esterly, D. M.; Love, B. J. Phase Transformation to Β-Poly(Vinylidene Fluoride) by Milling. J. Polym. Sci., Part B: Polym. Phys. 2004, 42, 91-97. (29) Gregorio, R. Determination of the Α, Β, and Γ Crystalline Phases of Poly(Vinylidene Fluoride) Films Prepared at Different Conditions. J. Appl. Polym. Sci. 2006, 100, 3272-3279. (30) Mago, G.; Kalyon, D. M.; Fisher, F. T. Membranes of Polyvinylidene Fluoride and Pvdf Nanocomposites with Carbon Nanotubes Via Immersion Precipitation. J. Nanomater. 2008, 2008, 1-8. (31) Nayak, S.; Sahoo, B.; Chaki, T. K.; Khastgir, D. Facile Preparation of Uniform Barium Titanate (Batio3) Multipods with High Permittivity: Impedance and Temperature Dependent Dielectric Behavior. RSC Adv. 2014, 4, 1212-1224. (32) Boccaccio, T.; Bottino, A.; Capannelli, G.; Piaggio, P. Characterization of Pvdf Membranes by Vibrational Spectroscopy. J. Membr. Sci. 2002, 210, 315-329. (33) Gregorio, J. R.; Cestari, M. Effect of Crystallization Temperature on the Crystalline Phase Content and Morphology of Poly(Vinylidene Fluoride). J. Polym. Sci., Part B: Polym. Phys. 1994, 32, 859-870.


25


Page 26 of 29

(34) Suresha, K. M.; John, B.; Konrad, W.; Boris, S. Effect of Poling Time and Grid Voltage on Phase Transition and Piezoelectricity of Poly(Vinyledene Fluoride) Thin Films Using Corona Poling. J. Phys. D: Appl. Phys. 2013, 46, 285305. (35) Xi, Y.; Fan, H.; Li, W.; Niu, X. In Effect of Corona Poling on Structure Evolutions of Α-Phase and Β-Phase Pvdf Films, 5th International Symposium on Advanced Optical Manufacturing and Testing Technologies, SPIE: 2010; p 6. (36) Luongo, J. P. Far-Infrared Spectra of Piezoelectric Polyvinylidene Fluoride. J. Polym. Sci., Part A-2: Polym. Phys. 1972, 10, 1119-1123. (37) Neumann, G.; Bihler, E.; Eberle, G.; Eisenmenger, W. In Polarization Distribution in Pvdf Obtained by Poling under Constant Current Condition, Electrical Insulation and Dielectric Phenomena, 1990. Annual Report., Conference on, 28-31 Oct 1990; 1990; pp 96-101. (38) Dalle Vacche, S.; Leterrier, Y.; Michaud, V.; Damjanovic, D.; Aebersold, A. B.; Månson, J.-A. E. Effect of Interfacial Interactions on the Electromechanical Response of Poly(Vinylidene Fluoride-Trifluoroethylene)/Batio3 Composites and Its Time Dependence after Poling. Compos. Sci. Technol. 2015, 114, 103-109. (39) Keatsamarn, T.; Pintavirooj, C. In Foot Plantar Pressure Measurement System Using Optical Sensor, 9th Biomedical Engineering International Conference 7-9 December 2016; 2016. (40) MacWilliams, B. A.; Armstrong, P. F. In Clinical Applications of Plantar Pressure Measurement in Pediatric Orthopedics, Pediatric Gait: A New Millennium in Clinical Care and Motion Analysis Technology Chicago, IL, USA, USA, 22-22 July 2000; Chicago, IL, USA, USA, 2000. (41) Abdul Razak, A. H.; Zayegh, A.; Begg, R. K.; Wahab, Y. Foot Plantar Pressure Measurement System: A Review. Sensors 2012, 12, 9884. (42) Wertsch, J. J.; Webster, J. G.; Tompkins, W. J. A Portable Insole Plantar Pressure Measurement System. Journal of Rehabilitation Research and Development 1992, 29, 13-18 (43) Rocha, J. G.; Goncalves, L. M.; Rocha, P. F.; Silva, M. P.; Lanceros-Mendez, S. Energy Harvesting from Piezoelectric Materials Fully Integrated in Footwear. IEEE Trans. Ind. Electron. 2010, 57, 813-819. (44) Kanitthika, K.; Chan, K. S. In Pressure Sensor Positions on Insole Used for Walking Analysis, The 18th IEEE International Symposium on Consumer Electronics (ISCE 2014), 22-25 June 2014; 2014; pp 1-2. (45) Lim, D.-H.; Kim, W.-S.; Kim, H.-J.; Han, C.-S. Development of Real-Time Gait Phase Detection System for a Lower Extremity Exoskeleton Robot. International Journal of Precision Engineering and Manufacturing 2017, 18, 681-687. (46) T, S.; Y, S.; T, K.; H., O. Sensing Stability and Dynamic Response of the F-Scan in-Shoe Sensing System: A Technical Note. Journal of Rehabilitation Research & Development 1998, 35, 9. (47) In-Shoe Pressure Measurement Technology. https://www.tekscan.com/productgroup/medical/in-shoe (accessed 22nd February). (48) Fuh, Y. K.; Wang, B. S.; Tsai, C.-Y. Self-Powered Pressure Sensor with Fully Encapsulated 3d Printed Wavy Substrate and Highly-Aligned Piezoelectric Fibers Array. Sci. Rep. 2017, 7, 6759. (49) Soames, R. W. Foot Pressure Patterns During Gait. Medical Engineering and Physics 7, 120-126. (50) A, G. Pressure-Sensing Devices for Assessment of Soft Tissue Loading under Bony Prominences: Technological Concepts and Clinical Utilization. Wounds 2007, 19, 350–362.


26

Page 27 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60



27



Page 28 of 29

28

Page 29 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60



1

SIMULTANEOUS 3D PRINTING AND POLING OF PVDF AND ITS

Recommend Documents