A Novel Soft Metal‐Polymer Composite for Multidirectional Pressure Energy Harvesting

Abstract

A cost-effective 3D sponge composed of rubber and conductive wires is introduced, which generates electric energy by periodic compression. It can be fabricated in a symmetric configuration, hence there is not a preferred surface or a specific direction for compression to obtain the energy harvesting or pressure sensing. The energy conversion occurs due to both triboelectric and electrostatic effects, which benefit from the air gap between polymer and wires. Although the triboelectric effect has been known since 1913, little progress has been achieved toward generating electric energy using this effect.1, 2 Recently, work was undertaken by the nanoscience research group in Georgia Tech to use this for energy harvesting and sensing purposes. During the last year, they used this effect as a competitive method for electric generation at micropower scales.3 Amazing results were published by the same group, who developed efficient triboelectric power sources by implementing micropatterns on flexible surfaces;4 a transparent pressure sensor;5 nanoscale patterns for powering portable devices;6 nanoparticles enhanced generators;7 and hybrid energy harvesters.8 All the mentioned works share the same benefit, i.e., micro/nanoscale patterns on the employed surfaces. Moreover, the above systems are based on two semiparallel electrodes with an intermediate material between them. In the resulting structures, at least two components move relative to each other for generating charges. Therefore, a matrix of layers containing a free space between them is necessary in the fabrication process. Providing a constant space between the layers is one of the challenges that has been dealt with by using several approaches, such as including a spacer,4 arc shape surfaces,6 springs,7 etc. Although these approaches can increase the efficiency of the system, they bring restrictions for implementing and aligning those systems in the integrated structures, especially where a soft body is requested, such as in the emerging field of soft robotics.9 Here, we address the above limitations by introducing a very simple and cost effective approach to build a composite consisting of a rubber matrix embedding in a disordered fashion, conducting materials, and air gaps. In the proposed technology all parts can be developed together on a single surface or in a 3D structure. The fabrication process does not require bonding steps for assembling the system and creating air gaps, thus the overall fabrication complexity is dramatically reduced. Air bubbles remain trapped inside the sponge by means of solving sacrificing particles. The triboelectric phenomenon is promoted by periodic contact between miniature pieces of copper microwires encapsulated in a polydimethylsiloxane (PDMS) sponge. The strong advantages of this work are its simplicity and the cost effectiveness of the process, which makes it suitable for mass manufacturing and fabrication with easy-access facilities. Moreover, because the prototypes are made out of a mixture, the building process can be integrated with other fabrication methods in a broad range of scales. The possibility of trapping conductive wires in rubber opens the way to a new generation of energy harvesting devices, targeting all application fields where a flexible and soft material is used and/or needed. Moreover the fabrication process is not constrained to planar technology but fully 3D shapes can be built. The wired-sponge polymer could be tailored for consumer goods used in everyday life, for which energy harvesting mechanisms have not been considered to date, such as some parts of wheels, mattresses, toys and playgrounds, dance floors, etc. It is noteworthy that this approach could be addressed in advanced robotic systems where the body of the robot is built with soft materials via a bioinspired design paradigm.10 On the other hand there are some aspects that must be considered, including that the amount of the charges generated on the wires and the polymer depends on the relative position in the triboelectric series,11 that the wires and polymer configuration and surface roughness are important, and that humidity affects charge generation. The encapsulation method that we use prevents the influence of environmental conditions on the generated charges. Furthermore, because microwires are used, we benefit from the small size of the structures, which increases the effective surfaces on which charges can be generated, and thus strengthen the triboelectric effect. In Figure 1a, the method chosen for fabricating our prototype is depicted. It is also explained in detail in the Experimental Section. Figure 1b shows the fabricated prototype. A common twisted copper wire made of a matrix of thinner wires was chosen as the conductive material. The wires play a two-fold role: 1) they become charged on contact with the polymer and 2) they conduct the generated charges to the load. To increase the surface area, thin wires are desired so we divided them into their original constitutive parts, down to 65 μm in diameter. We used sugar as a sacrificing material in order to create a gap between the wires and the polymer matrix, which is necessary for making and releasing a contact between two triboelectric materials. Such sacrificial material is also used for building a sponge shape structure from the polymeric matrix, in order to obtain a more elastic and compressible material. Figure 1c shows a scanning electron microscopy (SEM) image of a cross section of the wired-sponge generator, where the gaps between the wire and the polymer (in the shape of both sugar particles and the removed syrup layer) are visible. Since PDMS is placed nearly at the end of the triboelectric series11 and is flexible enough to make a sponge-shape polymer, it was selected as the main body of our composite. In order to prevent the negative influence of humidity on the generated charges and at the same time allow the manipulation of the electric generator, we encapsulated the composite sponge in a soft silicon rubber frame. The highly flexible and stretchable rubber was chosen to make it compatible with the multidirectional function of the generator, where the two opposite sides can be pressed towards each other. Electric energy generation is achieved via the coupling between the triboelectric and electrostatic effects. When the composite is compressed, the conductive wires come in contact with the polymer through friction, and all components become electrically charged. Most conductive materials are placed at the top of the triboelectric series,11 compared to the PDMS that, instead, is at the bottom; therefore, they will be charged as positive and negative, respectively. The sponge triboelectric composite is operated by applying a periodic compressive force on one of its surfaces. Within each cycle, when the composite is released from the compressive force, it recovers its original shape because of the stored elastic energy. Figure 2a shows a schematic of the energy harvesting mechanism. At the beginning, there is no charge on the polymer surfaces and electrodes. In the first half of the work cycle, shown in Figure 2a-I, while the material is compressed, positive charges are generated on the wires, referred to as the active electrode, and negative charges are observed on the polymer. This mechanism is the result of the triboelectric charge transfer at the interface of these materials inside the composite while they make contact with each other. The produced negative triboelectric charges can be preserved on the PDMS surface for a long time because of the nature of the insulator.12 However, the triboelectric positive charges on the conductive wires can flow through the connected load. We place another conductive wire as a reference electrode in the composite. In contrast to the active electrode, it is completely attached to the polymer without any gap. Because it is completely covered by the same polymer and embedded in the composite, no triboelectric charge is generated on the reference electrode during one work cycle. Therefore, in this first stage, the active electrode is positively charged (unlike the reference electrode). However, no electrical current flows from the active to the reference electrode through the load connected to the two electrodes. The reason is that the active electrode and the polymer having opposite charges are in contact, and thus the system is in the equivalent mode. On the other hand, in the second half of the work cycle, once the polymer with the negatively charged surface starts to be separated from the wire, the positive charges induced on the active wire decrease. In addition, since the reference electrode has no triboelectric charge and is electrically connected to the active electrode, the electrons flow from the reference electrode to the active wire to compensate for the mentioned new charge distribution on the active electrode. In fact, the difference in potential between two ends of the load, i.e., the active and reference electrodes, provides the electronic flow in the conductive wires. The described mechanism yields a positive current signal and a positive charging of the reference electrode (Figure 2a-II). When the two triboelectric-charged surfaces are completely separated, an equilibrium state can be created with no output voltage/current (Figure 2a-III). When the cycle is repeated and the composite is pressed again, the induced positive charges on the wires increase. Similar to the above description, the electrons flow back from the active wire to the reference electrode producing a negative current signal. Once the two materials completely attach to each other, the charged surfaces make full contact again, and there will be no change of the induced charges on the wires, thus no output current can be observed (Figure 2a-IV). After a few cycles, the wires/polymer interface saturates from the triboelectric point of view. As the charge transfer is stopped, the generation of charge between the active electrodes will continue due to the electrostatic effect. In other words, because the contact surface at the wires/polymer interface has been previously charged, contact and subsequent release lead to a charge flow between the electrodes. When the surfaces are in contact, the opposite charges on the two kinds of materials will almost reach equilibrium of charge distribution and, subsequently, the charges induced on the electrodes will flow back through the load. However, by releasing the contact, driving the charges between the electrodes compensates for the charge distribution in the composite, as explained above. One of the advantages of this design is that by periodically making and breaking contact between the opposite wire and polymer surfaces, the polymer around the wire will be charged negatively. Therefore, at least one side of each wire has a cyclic connection with the polymer and this leads to charges generated by both the triboelectric and electrostatic effects. To explore the analytical issues, we suppose that the produced negative triboelectric charges accumulate on the inner side of the tubular polymer. The generated electrostatic field between those charges and the positive ones on the conductive wire serve the stored energy. When the composite is compressed, depending on the energy conversion efficiency, a proportion of the applied energy is converted to the mentioned electrical energy. The application and release of such force in a periodic fashion allows, in each cycle, the capture and release of energy, respectively. To prove the concept, we built and tested a wired-sponge generator composed of a highly compressible PDMS sponge/copper wire composite (3 cm3 in volume), which was encapsulated in silicone rubber (4 cm3 of total volume). Outstandingly, the wired-sponge generator can be compressed from all sides down to 5 mm of thickness. The generator was placed between two parallel plastic plates and, while detecting the output voltage, it was periodically pressed by means of two different systems. In one case a sliding motion was produced (INSTRON R464 testing machine, Instron Corporation, Canton, Massachusetts, USA) with a higher pressure and less velocity compared to the second method (PS01-23x80F, Linear Motor, LINMOT, USA). Thus, we used both methods to cover a reasonable range of experimental conditions. In both cases, suitable load cells were used to record the compressive stresses during the strain cycles. A video showing the two set-up during the cyclic compression of the samples is provided as Supporting Information. For the preliminary tests, the velocity of the slider in the first setup was at its maximum value (2500 mm min-1), and the applied strain was ≈87%. Therefore, by considering the needed time for reversing the pressure direction, the maximum frequency for applying the periodic pressure by the first setup was ≈0.54 Hz. The output voltage was measured using an oscilloscope (Agilent Technologies oscilloscope, model MSO7014A) while the probe and the ground were connected to the end of the confined wires in the composite and to the reference electrode, respectively. The open-circuit voltage is shown in Figure 2b. It is noted that when the sponge was pressed, the first positive peak occured, which proves that the wires were charged positively. The negative peak was generated as the external force was completely released. Since the sponge is full of wires, by pushing them against the encapsulated PDMS, they will be connected to the polymer one by one. We interpret this as the reason for the outstanding broad peak that was obtained. The application of the strain in the first half of the work cycle can be considered as producing a sustained contact between wires and polymer. However, some narrow peaks are most probably generated due to the connection of the polymeric material with the wires around which a larger gap or a sharper contact is created. Two peaks result in the first positive part of the cycle. As a low-speed movie of the system confirms (Supporting Information), they are produced by the first contact between the load cell and the composite, and by applying the maximum pressure on the composite. Similarly, the negative parts of the signal are obtained when the pressure is released. The first negative peak occurred as the slider starts to come back and the maximum pressure is released. However, when the load cell is completely separated from the composite surface, the second one is found. Furthermore, as in the positive cycle, some narrow peaks are observed. It seems that due to the small vibration of wires inside the composite, some positive peaks are generated. Because of the composition of the wires and air gaps around them, there is no specific surface of the built cubic structure to be considered for applying pressure. Therefore, the system can operate as a multidirectional force converter. For this reason we encapsulated the composite in a kind of rubber that could be subjected to both tension and compression. However, it must be mentioned that because of this, the system works as a mechanical damper, and it filters the externally applied pressure. Figure 2c shows the results the experimental analysis where the force is applied to two opposite sides of the cubic composite, perpendicular to those considered in Figure 2b. Although lower output voltage is exhibited, the output signals have a similar shape of the previously obtained signals, as explained above. This confirms how the composite can be used for multidirectional energy harvesting purposes. We characterized the effect of force, frequency, total displacement, and external load on the generated power. As shown in Figure 3a, the average power varies linearly with the compressive strain when different pressures are applied in the range of ≈1 to 13 KPa). It can be construed that by increasing the pressure, more contact between wires and polymer is obtained and thus the power increases. The experiments were carried out at the maximum frequency that the first setup (slow one) was able to apply, i.e. ,0.54 Hz, where the external resistance between the two ends was 10 MΩ. Moreover, the influence of the frequency on the generated power was measured by increasing it at the constant strain of 45%. As explained above, in order to evaluate the function of the generator at higher frequencies, the tests were accomplished by means of the second setup. The results shown in Figure 3b demonstrate that the power increased up to 1.22 μW where the frequency experiences the rise up to 9.4 Hz. This behavior can be described by the fact that the average power is directly related to the frequency as well as the differential energy explained in the Equation 2. In other words, when a distinct amount of pressure is applied to the wired-sponge composite slowly (or the straining frequency is decreased) the charge separation/attachment rate is reduced and thus the average generated power is decreased. Furthermore, the influence of the external load on the output voltage and the generated power was evaluated by connecting a series of different resistors to the wired-sponge generator (Figure 3c). The electrical capacitance of the composite under different applied strains was also measured, by means of a LCR meter (Agilent, E4980A), with a reference signal of 1 V peak-to-peak at 1 kHz (Figure 3d). Results show that the compression of the material to the 80% of the initial size causes a 45% increase of the composite capacitance, thus illustrating the decrease in the distance between the two different electrodes. In fact, the reference electrode was placed roughly in the centre of the composite while the active electrodes were distributed in the sponge. As a result, a compression from each direction brings the active electrodes much closer to the reference electrode, where, depending on the reduction of their relative positions for each part of the active electrode, the capacitance between them will be enhanced. Due to the use of PDMS as the material for the composite in which a amount of strain can be considered in the generator, the can be as an polymer energy Within this broad of devices, the of the wired-sponge generator is with because its function is based on the electrostatic effect and The generated power by those kinds of energy is in the range of μW to the which electrical energy directly from the mechanical pressure, they require It is noteworthy that the fabrication process of the proposed composite is very simple and cost and there is to its to with the current polymeric energy As the video in the Supporting this wired-sponge generator is able to up an while compressed periodically by Supporting for more In order to evaluate how the size of the sugar particles affects the efficiency of the generator, samples were by using wires as the active electrode. different of generator were by using sugar particles with as a sacrificing Experimental for Figure how the size of the sugar and thus the PDMS the output power in 2 of wire The generated power increases when larger particles of 1 are used with a constant as the size of particles the gap between wires and increases as as Equation 2 the produced power is increased that the for the triboelectric charging remain and can be considered the same for all To the of the different made by sugar the is shown in Figure As it in order to obtain a composite we to use particles. In fact, by increasing the size of the is with particles compared to when ones are In other words, the more gaps between the sugar the more polymer is needed to the same of composite, and thus a composite is obtained. Furthermore, as an elastic behavior is a result of the rubber composite mechanical We have that by using the structure we are able to the pressure applied to each side of the soft composite. Therefore, as because this method is very cost effective and free of external powering process, we will how to use the wired-sponge generator in a new generation of pressure Furthermore, as mentioned above, the triboelectric energy can benefit from having on the surfaces that contact, and thus in work will how to increase the of the current of the A of and sugar with a was and to its The sugar particles an average of of a mm copper wire was and 65 μm wires were obtained. were by in and the wires were in the for a few to become covered by the After that, in order to make this the syrup was as much as the sugar particles with an average of μm were on the wires to the particles around the wires. This was in a cubic and by more sugar and applied pressure from the top side in order to the sugar for the PDMS sponge. During this a mm wire, which was covered by a PDMS was also placed at the of the and as a reference electrode. the was The sugar containing the wires and the reference electrode was to be by a Therefore, a of PDMS in a of by was and the sugar was placed The surface of the PDMS was a lower the side of the sugar in order to the inside air bubbles to come out PDMS was and the small gaps between sugar particles due to the This process was by for 5 and in the by for by a at for 10 After that, the sugar was in by and up to it was in the at for 2 In addition, the flexible was by around the composite in a of the The of a of a of wires with μm in were used as the trapped wires inside the sponge copper wire with a of 1 mm was covered by a PDMS resulting in a of and it was placed in the of the mentioned as a reference electrode. The wires were by in and for 10 the same explained above for making wired-sponge were carried out to build the composite in a cubic for the last relative to the encapsulation in an which in this case was The process was repeated using sugar particles with different and 1 mm of average in order to obtain with different For each two samples were and As a to our and this provides by the Such materials are and be for but are not or from be addressed to the The is not for the or of any by the be to the for the