Creative Writing

13 pages

Performance Quantification of Conducting Polymer Actuators for Real Applications: A Microgripping System

Please download to get full document.

View again

of 13
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Performance Quantification of Conducting Polymer Actuators for Real Applications: A Microgripping System
  University of Wollongong Research Online Faculty of Engineering - PapersFaculty of Engineering2007 Performance Quantification of Conducting Polymer Actuators for Real Applications: A Microgripping System G. Alici University of Wollongong   , N. N. Huynh University of Wollongong  Research Online is the open access institutional repository for theUniversity of Wollongong. For further information contact ManagerRepository Services: Recommended Citation  Alici, G. and Huynh, N. N.: Performance Quantification of Conducting Polymer Actuators for Real Applications: A MicrogrippingSystem 2007.  IEEE/ASME TRANSACTIONS ON MECHATRONICS, VOL. 12, NO. 1, FEBRUARY 2007 73 Performance Quantification of Conducting Polymer Actuators for Real Applications: AMicrogripping System Gursel Alici and Nam N. Huynh  Abstract  —In this paper, we report on modeling, characteriza-tion,andperformancequantificationofaconductingpolymeractu-ator,drivingarigidlinktoformeachfingerofatwo-fingergrippingsystem,whichiswhatwecallamicrogrippingsystem.Theactuator,which consists of five layers of three different materials, operatesin a nonaquatic medium, i.e., air, as opposed to its predecessors.After the bending displacement and force outputs of a single fingerare modeled and characterized including the effect of the magni-tude and frequency of input voltages, the nonlinear behavior of the finger including hysteresis and creep effects is experimentallyquantified, and then a viscoeleastic model is employed to predictthe creep behavior. The experimental and theoretical results pre-senteddemonstratethatwhilethehysteresisisnegligiblysmall,thecreep is significant enough so as not to be ignored. The response of the actuator and the finger under step input voltages is evaluated,and found that the actuator does not have any time delay, but onlya large time constant. Two of the fingers are assembled to forma microgripping system, whose payload handling and positioningability has been experimentally evaluated. It can lift up to 50 timesits weight under 1.5 V. The payload handled was a spherical objectcovered with industrial type tissue paper. The friction coefficientbetween the object and the carbon fiber rigid link has been de-termined experimentally and used to estimate the contact force.All the theoretical and experimental performance quantificationresults presented demonstrate that conducting polymer actuatorscan be employed to make functional microsized robotic devices.  Index Terms —Actuators, electroactive polymer actuators,flexure-based devices, micromanipulation, system identifica-tion/characterisation. I. I NTRODUCTION A S POTENTIAL electromechanical actuators and sensors,which are very suitable for miniaturization, conductingpolymer actuators have attracted the attention of many re-searchersinthelastdecade[1]–[9].Acomprehensiveaccountof polymer actuators is given in [1] and [9]. They have a compos-ite structure with polymer layers separated from each other withan insulator. When the right stimulus, which is usually a verysmall voltage, typically 1 V or a current, is applied to the poly-mer layers, a volume expansion and contraction occurs due to Manuscript received November 20, 2005; revised April 26, 2006. Recom-mended by Technical Editor J. P. Desai. This work was funded in part by aUniversity Research Council (URC) grant.G. Alici is with the School of Mechanical, Materials and Mechatronic En-gineering, and the ARC Center of Excellence on Electromaterials Science,University of Wollongong, NSW 2522, Australia (e-mail: N. Huynh is with the School of Mechanical, Materials and Mecha-tronic Engineering, University of Wollongong, NSW 2522, Australia ( Object Identifier 10.1109/TMECH.2007.886256 electrochemomechanical properties of the polymers [10]– [12].The change in the volume generates a bending displacement— theelectrochemical energy isconverted intomechanical energy.As a result, considerable amount of research has been devotedto modeling and understanding their behavior in order to im-prove their synthesis conditions for use as reliable actuators andsensors for new cutting applications ranging from biomedicaldevices to micromanipulators [4] [7] [13]. Zhou  et al.  [4] havereportedonthreetypesofpolymeractuatorsincludingioniccon-ducting polymer film actuator, polyaniline actuator, and pary-lene thermal actuator. They have presented their fabrication andinitial performance results. Zhou and Li [14] have reported onthe MEMS-based fabrication of cantilever microstructures con-sisting of Au/Nafion/Au trilayers on silicon substrates in order to construct microgrippers operating in aqueous media. Smela et al.  [7] have presented the development and performance out-comes of polypyrrole (PPy) and Au bilayer conducting polymer actuators  operating in electrolyte solutions . As an extensionof this study, Jager   et al.  [13] have fabricated a serially con-nectedmicromanipulatortopick,move,andplace100- µ mglassbeads.This paper is part of an ongoing-project on the establishmentof manipulation systems such as grippers and planar mech-anisms, articulated with the fourth generation PPy actuatorsfabricated in the Intelligent Polymer Research Institute at theUniversity of Wollongong [2].Conducting polymers have manypromising features including low actuation voltage, operationin aquatic mediums and in air, low cost, and high force–outputweightratio.Theirmaindrawbackistheirlowspeedofresponseandnonlinearityduetotheactuationprinciple,whichisbasedonmass transfer. The application of conducting polymer actuatorsis an emerging field, as researchers begin to harness the bene-fits of their material properties, which are greatly enhanced atsmaller scales [9]. Possible future applications include artificialmuscles and a wide variety of sensors and actuators in biomed-ical systems [6] and micro/nano manipulation systems [15]. As these actuators do not contain any rolling and sliding elements,they are suitable for micro/nano manipulation tasks, which re-quiremotionaccuracyoftheorderof0.05–0.1 µ m(50–100nm).In our previous studies [16] – [20], we reported on developing various mathematical models to predict the bending behavior of the conducting polymer actuators, and employing the mod-els to optimize their topology with high force and displacementoutputs.In this paper, we report on the performance quantificationof a PPy-based conducting polymer actuator driving a rigid 1083-4435/$25.00 © 2007 IEEE  74 IEEE/ASME TRANSACTIONS ON MECHATRONICS, VOL. 12, NO. 1, FEBRUARY 2007 Fig. 1. (a) Schematic structure of the conducting polymer actuator.(b) Schematic representation of the bending principle. carbon fiber link, which constitutes each finger of a microgrip-ping system with the dimensions of (5-mm actuator   +  5-mmrigid link) × 1 mm × 0.17 mm. With reference to these dimen-sions and its fabrication process, the microgripping system isnotMEMS-based.Itmustbenotedthatsurfacemicromachining,bulk micromachining (i.e., etching), bonding alone, electrode-position, photolithography, and other replication techniques areused to make MEMS-based microdevices/microstructures. Theperformance evaluation has been realised in terms of the bend-ing displacement and force output, and their nonlinear behavior including hysteresis and creep. The actuator has been used ina real application to articulate a two-finger gripper. The exper-imental and estimated results attest that electroactive polymer actuators are suitable to make micromanipulation devices suchas a gripper, which can handle as much as 50 times its totalmass.II. P OLYMER  A CTUATOR  D RIVING A  R IGID  L INK The polymer actuator and the rigid link form a monolithiccomposite structure, where the polymer section serves as theactuator and a flexure joint—like an active flexure joint, and therigid link as a payload carried by the actuator.  A. Actuation Structure and Mechanism The structure of the PPy-based polymer actuator, consideredin this paper, is shown in Fig. 1. The actuator has five layers.The outer two layers, which are PPy with thicknesses of 30 µ m,are the electroactive elements providing actuation. The middlelayer is polyvinylidine fluoride (PVDF), an inert, nonconduc-tive, porous polymer. It serves as a separator for the two PPylayers and the reservoir for electrolyte (tetrabutylammoniumhexafluorophosphate) TBA PF 6  0.05 M in solvent propylenecarbonate. The electrolyte and the solvent need to be stored inthePVDFlayerinordertooperatetheactuatorinair.Otherwise,it has to be operated in an aqueous medium consisting of theelectrolyteandthesolvent.Thinlayersofplatinumof10–100 ˚Aare sputter-coated on both sides of PVDF to enhance the con-ductivity between the PPy layers and the electrolyte. When anelectrical potential is applied across the electrodes attached tothe PPy layers, the reduction/oxidation (redox) process occurs, Fig. 2. Structure and dimensions of the robotic finger. (a) Front view and topview. (b) Cross section of the actuator part (not to scale). and the actuator bends. For the actuator employed in this study,the redox process is described by PPy + PF  − 6  + e − (oxidation state)oxidation reduction ←−−−−−−−−−−→ PPy 0 + PF  − 6 (reduced state)  .  (1)TheredoxprocessoccurssimultaneouslyonboththePPylay-ers; while one layer is oxidized, the other is reduced, and viceversa.Withreferenceto(1),thedisplacementof  PF  − 6  ionsinthePPy layers causes material strain through a number of effects.One is the effect of the gain or loss of ion volume causing vol-ume expansion or contraction in the PPy layers. The movementof solvent molecules due to osmotic pressure, which accom-pany the ion diffusion, also contributes to the volume changein the PPy layers. The other factor is the resultant electrostaticforces between the displaced ions and the polymer backbonethat cause the PPy layers to expand or contract. As the two PPylayers undergo opposing strains, a bending moment is induced,generating the deflection of the actuator.  B. Robotic Finger  The structure of the robotic finger made up of the polymer actuator and carbon fiber is shown in Fig. 2.The conducting polymer part of the finger works as an actua-torand aflexure jointcalled  activeflexure (acti-flex)  joint,whilethe carbon fiber attached to the end of the polymer serves as arigid link for the robotic finger.The finger is fabricated as follows. •  An already fabricated sheet of conducting polymer istrimmed into strips of 1 mm × 15 mm, and the carbonfiber with a thickness of 0.3 mm is trimmed into pieces of 1 mm × 5 mm.     The carbon fiber pieces are cured in an oven for about10 min at 100  ◦ C to make it rigid.     Adouble-sidedstickytapeisputontotherigidcarbonfiber pieces.     The rigid carbon fiber pieces with the sticky tape on oneside are then attached to the polymer strip to make thefinger.     The samples are replenished in TBA PF 6  0.05-M elec-trolyte for 5 min before each test.  ALICI AND HUYNH: PERFORMANCE QUANTIFICATION OF CONDUCTING POLYMER ACTUATORS 75 Fig.3. Parametersdefiningthefingertipposition:coordinatesystems,bendingangle,andtheradiusofcurvature.Theradiuswasmeasuredfromtheintersectionof the perpendicular lines of two adjacent segments.Fig. 4. Experimental setup. III. E XPERIMENTAL  P ERFORMANCE  Q UANTIFICATION OF R OBOTIC  F INGER The fabricated finger has the dimensions of (5-mm actuator  +  5-mm rigid link) × 1 mm × 0.17 mm, as depicted in Fig. 3. AphotographofalltheapparatususedforexperimentsisshowninFig. 4. eDAQ e-corder recorder unit together with eDAQ Chartand eDAQ Scope software is used to record, amplify, filter, andanalyse data. Aurora Scientific Inc. dual-mode lever arm sys-tem, model 300 B is used to measure the tip force. The platinumwires on the electrode clamps are connected to the outputs of a potentiostat/galvanostat, controlled using Chart 4 Windowssoftware via a Powerlab 4/20 controller. Other equipments in-clude a PC, a digital video camera, a metal stand and clamps,two-electrode clamps and a grid paper. The schematic repre-sentation of the experimental setup employed to evaluate theperformance of the finger is shown in Fig. 5. Fig. 5. Schematic representation of the equipment.Fig. 6. Tip displacements for two equivalent robotic fingers under a range of constant input voltages.  A. Effect of Input Voltages on Displacement Output of a Robotic Finger  Under constant voltages ranging from 0.2 to 1 V, the tip dis-placements of the robotic finger is measured, and are shownin Fig. 6 for two samples of the same finger. The same exper-iments, which were conducted for the polymer actuator withthe dimensions of 5 mm × 1 mm × 0.17 mm, had shown thesame trends, i.e., the higher the input voltage, the larger thetip displacement. However, it must be noted that the displace-ment along the  x -axis is much smaller than the displacementalong the  y -axis. In Section IV, a mathematical model to pre-dict the bending displacement of the actuator is presented. Itmust be noted that the corresponding radii of curvature depictedin Fig. 7 are needed for the force model provided in the nextsection. The tip displacement results of our trilayer actuator operating in air is in agreement with the performance character-isation results presented in the literature for a trilayer polymer actuator (20 mm × 15 mm × 0.013 mm) operating in a 1-MLiClO 4  aqueous solution under current control [12].  76 IEEE/ASME TRANSACTIONS ON MECHATRONICS, VOL. 12, NO. 1, FEBRUARY 2007 Fig. 7. Measured radii of curvature for two equivalent robotic fingers under arange of constant input voltages.Fig. 8. Experiment data recorded under a square wave with a magnitude of  ± 1 V with the frequency of 1 PPM.Fig. 9. Variation of the output force with input frequency.  B. EffectofFrequencyoftheInputVoltagesontheForceOutput of a Robotic Finger  The force output of the robotic finger is measured under aset of square waves with a constant magnitude of   ± 1  V andthe frequencies of 10 pulses/min (PPM) or 0.1667 Hz, 6 PPMor 0.1 Hz, 4 PPM or 0.0667 Hz, 3 PPM or 0.05 Hz, 2 PPM or 0.0333 Hz, and 1 PPM or 0.0167 Hz. Before each test, an inputvoltage of 0 V is applied to neutralize the finger. The currentpassed through the oxidized polymer layer, the input voltage,andtheoutputforce,dataextractedfromtheeDAQdataloggingsystem, are shown in the top, middle, and bottom plots of Fig. 8,respectively.Thesearetypicaldatarecordedforeachfrequency.The magnitude of the force for each frequency is read fromthe recorded force data, and is shown in Fig. 9, where the forcedecreases linearly with the input frequency. This can be ex-plained by the fact that the movement of dopant ions and sol- Fig.10. Variationoftheminimumcurrentrecordedduringforcemeasurementwith the frequency.Fig. 11. Bending curve of a bending polymer actuator. vent molecules requires certain time to move in and out of thePPy layers during the redox process. Obviously, with high in-put frequencies, there is not sufficient time for the ions and themoleculestoreachdeep intothepolymer layertogenerate moreforce.Thisislikestoppingtheirmovementinthemiddleoftheir  journey and asking them to go back. This argument is supportedby the magnitude of the minimum current passing through theoxidized layer, i.e., if the minimum current is closer to zero,the better will be the oxidation, which generates more actua-tion effect. The minimum current for the input frequencies isdetermined from the recorded current data, as shown in Fig. 10,where the magnitude of the minimum current under a relativelyhigh input frequency is significantly higher than that of the lowinput frequency.IV. B ENDING  M OTION AND  B ENDING  F ORCE  M ODELING The mathematical models are needed to provide enhanceddegrees of understanding, predictability, control, and efficiencyin performance in order to improve the displacement and forceoutputs of the polymer actuators before using them in real ap-plications [20].  A. Bending Motion Model This mathematical model, which had been reported in our previous study [16], is summarized here for the sake of com-pleteness.Themodelhasbeenderivedintermsofthetipverticaldisplacement  v  and the horizontal displacement  x , as shown inFig. 11.
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!