sensors Article

Development of a Hybrid Piezo Natural Rubber Piezoelectricity and Piezoresistivity Sensor with Magnetic Clusters Made by Electric and Magnetic Field Assistance and Filling with Magnetic Compound Fluid Kunio Shimada 1, * and Norihiko Saga 2 1 2

*

Faculty of Symbiotic Systems Sciences, Fukushima University, 1 Kanayagawa, Fukushima 960-1296, Japan Department of Human System Interaction, Kansai Gakuin University, 2-1 Gakuen, Sanda, Hyogo 669-1337, Japan; [email protected] Correspondence: [email protected]; Tel.: +81-24-548-5214

Academic Editor: Vittorio M. N. Passaro Received: 16 December 2016; Accepted: 7 February 2017; Published: 10 February 2017

Abstract: Piezoelements used in robotics require large elasticity and extensibility to be installed in an artificial robot skin. However, the piezoelements used until recently are vulnerable to large forces because of the thin solid materials employed. To resolve this issue, we utilized a natural rubber and applied our proposed new method of aiding with magnetic and electric fields as well as filling with magnetic compound fluid (MCF) and doping. We have verified the piezoproperties of the resulting MCF rubber. The effect of the created magnetic clusters is featured in a new two types of multilayered structures of the piezoelement. By measuring the piezoelectricity response to pressure, the synergetic effects of the magnetic clusters, the doping and the electric polymerization on the piezoelectric effect were clarified. In addition, by examining the relation between the piezoelectricity and the piezoresistivity created in the MCF piezo element, we propose a hybrid piezoelement. Keywords: sensor; piezoelement; piezoelectricity; piezoresistivity; hybrid; electrolytic polymerization; magnetic field; magnetic cluster; natural rubber; magnetic compound fluid (MCF); magnetic fluid; isoprene; filler

1. Introduction Many types of sensors (force, temperature, pressure, etc.) are currently used in many different fields. Most sensors, including thermocouples and strain gauges, require a power supply in order for the electric signal of temperature or strain to be responsive to the voltage or electric current applied to the sensor. However, piezoelements have the distinguishing feature that they can sense an induced voltage without a power supply. Because in robot construction extra weight is very undesirable, power supplies should be avoided when possible, so in robotics, piezoelements are generally suitable for sensing force or pressure. In addition, a piezoelement must be suitable for use in the artificial skin, which has large elasticity and extensibility, installed as an outer layer on robots. It is expected that in the future, robots will become a part of our lives as domestic help for chores, and in the nursery, hospital, etc. There is currently a need for a material that can serve as skin that is similar to human skin, because it is important that we feel affinity with robots. Therefore, when the piezoelement is made of a polymer [1–4], the polymer selected should be rubber, and not be vulnerable to impulsive or large forces. This is why it is necessity to use some form of rubber to create robot skins. Rubber is durable

Sensors 2017, 17, 346; doi:10.3390/s17020346

www.mdpi.com/journal/sensors

Sensors 2017, 17, 346

2 of 17

enough to withstand impulsive, large or repeated forces, while maintaining good elasticity and extensibility [5]. Many studies have sought to make rubber piezoelements. Silicon oil rubber has been the most commonly used material for piezoelements [6–11]. There have been a few investigations of other rubber materials: chloroprene [12], styrene-butadiene [13], acrylonitrile butadiene [14] and natural rubber [15]. In order to further improve the performance of a rubber sensor, it is often effective to apply a filler in the rubber. The use of filler is one way of obtaining specific desirable properties in a rubber composite material, including those with electrical, thermal, mechanical or curing properties [16–20]. Regarding polymers other than rubber, the application of fillers in polymers is also effective. There are numerous reports of investigations applying filler to a polymer [21–23]. A method is thus needed for making a piezoelement of rubber utilizing a filler that is simple for engineering production, involving low cost and time, and having the durability needed to withstand impulsive, large or repeated forces, and large elasticity and extensibility. In addition, when the filler is magnetic, we can obtain additional effective properties for the intended engineering application. Therefore, in the present study, we propose a new piezoelement made of natural rubber that uses a magnetic compound fluid (MCF) [24] as magnetic filler and that is solidified under the application of electric and magnetic fields. In order to accomplish the aim of this paper, we employ the MCF rubber technique [25]. Shimada devised MCF in 2001 [24] as a colloidal fluid containing spherical Fe3 O4 particles on the order of 10 nm in size, as well as other metal particles such as Ni, Fe, Cu, etc., on the order of 1 µm. These particles are dispersed in a solvent such as water, kerosene, or silicone oil to compound a magnetic fluid (MF), which is an ordinary intelligent fluid responsive to the magnetic field due to the Fe3 O4 particles. When a magnetic field is applied, the Fe3 O4 particles play a bonding role among the metal particles, allowing numerous metal and Fe3 O4 particles to aggregate [26]. MCF is useful for engineering applications owing to these magnetic clusters [27]. It is effective in polishing [28–33], as a material for use as a damper [34,35], and as a composite material, including MCF rubber. The composite material rubber mixed with MCF was called MCF rubber in previous studies. The magnetic clusters in MCF rubber produce the electric conductivity that leads to large changes in response to very small pressures. MCF rubber material has been shown to be suitable for making silicon oil rubber [36–40] and natural rubber [41–47]. It can be solidified under a constant temperature (for convenience, we call this the “drying method”). In general, ordinary piezoelements made of rubber have needed the drying method when a curing agent such as sulfur is added for vulcanization. In the case of MCF rubber handled in the previous study [41–47] as well as in the present study, sulfur vulcanization has not been used. Even if just the drying method is used for curing, the rubber can become suitable for the application in sensing fields under very low force or pressure because the rubber has the potential to comprehensively meet the condition of linear elasticity at very low force or pressure. Naturally, an effect of the sulfur on the solidification of the MCF rubber can be expected under large force and pressure. In the present study, we shall make no further inquiry into the effect of sulfur on the properties of MCR rubber and may leave those details to another study. Because the human skin which corresponds to a robot skin is extremely sensitive under very low force or pressure, when the MCF rubber is solidified by the application of a magnetic field and the drying method, it has a switching effect, which means the electric resistance changes from high to low because of the magnetic clusters. Therefore, the MCF rubber has been demonstrated to be suitable for sensing concave and convex shapes when installed on a robot [45,46]. It has also been clarified that in the case of natural rubber, MCF rubber has more sensitivity than commercial pressure-sensitive electrically conductive rubbers. Natural rubber (NR-latex) is also superior for the previously mentioned sensor requirements of large elasticity and extensibility and MCF filling is also very effective for producing higher sensitivity piezoelements made of NR-latex. Therefore, in the present study, we adopted an NR-latex type MCF rubber. Furthermore we also employed reagent doping. In general, the drying method for curing is inefficient for engineering production. To knead metal or carbon into the rubber requires considerable production costs and time because of the rolling

Sensors 2017, 17, 346

3 of 17

machine and drying needed. In addition, the material can only be produced in sheets; other shapes are not possible due to the rolling. Therefore, Shimada devised a new method that did not involve the drying method [48–50]. It is such a convenient and simple method that low production costs and time can be achieved. It also allows for the creation of any shape. By the application of an electric field, the MCF rubber creates an anode plate as a thin film. Then, with the application of another a magnetic field, the thickness of the MCF rubber grows along the magnetic field line. Ordinarily, NR-latex without MCF is created as a thin film, and its thickness does not grow even after a long application of a electric field. Therefore, the creation of MCF magnetic clusters by the application of the magnetic field is needed. Since the magnetic field is applied along the same direction of the electric field line, the thickness of the MCF rubber grows. If only the electric field or magnetic field is used, the electrical properties are less sensitive than when both are used, and the resulting rubber is less extensible, as shown in previous studies [48–50]. Therefore, both the electric and magnetic fields are necessary for larger elasticity and extensibility. The new method is related to so-called electric polymerization which is positioned as one of the types of vulcanization. In summary, the curing of NR-latex is promoted by the alignment of magnetic clusters in the filler created by the application of a magnetic field under electric polymerization conditions. On the other hand, in general, the effect of a piezoelement can be described as piezoresistivity or piezoelectricity. A piezoresistive element requires the application of voltage by a power supply and this produces the changes in resistivity of the piezoelement. This is essentially an electric conductor that allows electrons to percolate through the material. The mechanism of the electric conductivity is explained mainly by percolation theory [21] and tunnel theory. In the former theory the percolation threshold is a very significant factor. In the case of the large aggregation of particles or large aspect ratio of the particle shape, percolation theory cannot be applied because the percolation threshold decreases in spite of the enhancement of electric conductivity. Regarding the latter theory, when a voltage is applied to the material, the phenomenon can be explained by the proposition that electrons jump to percolate through the barrier of non-conductive material by a diode effect [40,51]. In MCF rubber, because the magnetic clusters are extremely aggregated, it has been proposed that the tunnel theory is suitable for the MCF rubber. On the other hand, piezoelectric elements does not require a power supply but rather obtain their charge from the changes in voltage induced from the material. This is a ferroelectric substance that allows the ions on crystal lattices sites with asymmetric charge surroundings within the material get closer to generate electricity. Piezoelectricity is suitable for the piezoelements which are to be utilized in robotics because of the attainable weight reduction as mentioned above. The induced voltage indicates the approach of positively and negatively charged ions in the material with the application of force or pressure. Piezoelectricity is very significant for robot sensors. However, we discovered that the MCF rubber dealt with in the present paper has both piezoelectric and piezoresistive effects. The relation between the former and the latter piezo effects has not yet been clarified. In robotics piezoresistive elements are also effective in some sensing applications. Therefore, it is valuable to investigate the MCF rubber property under the effects of both piezoelectric and piezoresistive effects. From the noted above, in this paper, we first discuss the production of the MCF rubber piezoelements by our new method. We then clarify the properties of the response of the MCF rubber piezoelements in terms of piezoresistivity and piezoelectricity through measurement by pressing as meaning d33 [52–54]. Next, we clarify the relation between the piezoelectric and piezoresistive effects. In addition, a hybrid piezoelement with the former and the latter piezoeffects is proposed. This hybridization means both piezoeffects have a role. The hybridization is useful for the sensing in robotics.

Sensors 2017, 17, 346

4 of 17

2. New Method of Producing a Piezoelement The new method used in this study to produce MCF rubber applied to the manufacture of piezoelements has been detailed previously [48–50]. We therefore do not describe it here. MCF rubber liquid was poured onto one metal plate and sandwiched by another. Permanent magnets with a 15 mm × 10 mm rectangular surface were applied to the inside of the plates to form the outer Sensors 2017, 17, 346 4 of 17 surface of each plate. The applied magnetic field at the position of the MCF rubber liquid was 490 mT. An electric held constant surface at 6 V were and 2.7 A was supplied plates 30-min periods. 15 mm field × 10 mm rectangular applied to the inside ofbetween the platesthe to form theatouter surface The plates atmagnetic a constant thickness using of a 3-mm spacer as liquid an electrode of eachwere plate.held The apart applied field at the position the MCF rubber was 490gap. mT. It is An electric field held constant atof6the V and 2.7 A voltage, was supplied between theand plates at 30-min reasonable to reduce the quantity applied electric current time, and toperiods. increase the The plates were held apart at a constant thickness using a 3-mm spacer as an electrode gap. It is spacer thickness. Therefore this new method led to the effective production of low cost and production reasonable to reduce the quantity of the applied voltage, electric current and time, and to increase time piezomaterial. The MCF rubber liquid was then vulcanized by application of the electricthe field. spacer thickness. Therefore this new method led to the effective production of low cost and We used plates made of stainless steel with a 35-mm square surface and 1-mm thickness, and production time piezomaterial. The MCF rubber liquid was then vulcanized by application of the conducted the experiment under atmospheric room conditions. We used three types of ingredients, electric field. as shownWe in used Tableplates 1. Here, a powder with µm-ordered twig-shaped particles (No.and 123 by madeNi of was stainless steel with a 35-mm square surface and 1-mm thickness, Yamaishi Co. Ltd., Noda, Japan), with 50 wt % conditions. of Fe3 O4 (M-300, Hi-Chemical Co. Ltd., conducted the experiment underMF atmospheric room We usedSigma three types of ingredients, Tsutsujigasaki, NR-latex Co. Ltd., Atsugi, Japan). Two types of (No. reagent as shown inJapan), Table 1.and Here, Ni was (Rejitex a powder with μm-ordered twig-shaped particles 123 used by as Yamaishi Co. Ltd., Noda,into Japan), MF with 50 wtfor %doping of Fe3O4with (M-300, Sigma Hi-Chemical Co. Ltd., dopant were compounded the MCF rubber KI or ZnO. The KI solution was an Tsutsujigasaki, Japan), and NR-latex (Rejitex Co. Ltd., Atsugi, Japan). Two types of reagent used as aqueous potassium iodide solution made up of 60 g of water and 40 g of potassium iodide. Doping dopant werebefore compounded into the MCF rubber for field. doping with KI or ZnO. The KI solution was an was conducted the application of the electric aqueous potassium iodide solution made up of 60 g of water and 40 g of potassium iodide. Doping was conducted before field. liquid used in this study. Tablethe 1. application Ingredients of of the the electric MCF rubber Table Type 1. Ingredients of the MCF rubber liquid used Type in thisCstudy. A Type B Ni: 6Ag Type MF: 1.5 g Ni: 6 g NR-latex: 6 g MF: 1.5 - g NR-latex: 6 g -

Ni: 9 g Type B MF: 2.25 g Ni: 9 g NR-latex: 9 g MF: 2.256.75 g g KI solution:

NR-latex: 9 g KI solution: 6.75 g

Ni: 3 gC Type MF: 2.25 g Ni: 3 g NR-latex: 9 g MF: ZnO:2.25 3g g

NR-latex: 9 g ZnO: 3 g

With this new method of producing MCF rubber for a piezoelement, we observed magnetic clusters aggregated by method Ni and of Fe3producing O4 particles like a needle 1a), confirming themagnetic results from With this new MCF rubber for a(Figure piezoelement, we observed a previous in the ofFe Type 1 (as indicated on(Figure the top inconfirming Figure 2)the made from A clustersstudy aggregated by case Ni and 3O4 particles like a needle 1a), results fromType a previous study in the case of Type 1 (as indicated on the top in Figure 2) made from Type A rubber [48–50]. The magnetic clusters occurred at the area where the permanent magnet was applied magnetic clusters occurred at theofarea permanent magnet was applied to therubber MCF[48–50]. rubber.The Figure 1 shows a cross-section thewhere MCF the rubber. The perpendicular direction to the MCF rubber. Figure 1 shows a cross-section of the MCF rubber. The perpendicular direction is the applied magnetic field line, which was the same as the electric field. Therefore, weiscould the applied magnetic field line, which was the same as the electric field. Therefore, we could assume assume compressed MCF rubber piezoelement was produced by the application of compression force, compressed MCF rubber piezoelement was produced by the application of compression force, as as shown in Figure 1b. By the compression, the positive and negative ions within the MCF rubber get shown in Figure 1b. By the compression, the positive and negative ions within the MCF rubber get reciprocally closer, inducing a voltage that can be measured by the electrodes placed on both outer reciprocally closer, inducing a voltage that can be measured by the electrodes placed on both outer surfaces, as shown in Figure 1b.1b. surfaces, as shown in Figure

(a)

(b)

Figure 1. Cross section of Type 1 made of Type A: (a) photograph of magnetic clusters created in the Figure 1. Cross section of Type 1 made of Type A: (a) photograph of magnetic clusters created in vulcanized Type 1 MCF rubber by electric and magnetic fields; (b) schematic diagram of the the vulcanized Type 1 MCF rubber by electric and magnetic fields; (b) schematic diagram of the compression of MCF rubber as a piezoelement. compression of MCF rubber as a piezoelement.

Figure 2 shows the principle of the production of a single piece of vulcanized MCF rubber as a piezoelement. Type 1 is a single-layer vulcanized MCF rubber. On the other hand, we can consider multiple-layer MCF rubber as other piezoelement. This is because we discovered an induced voltage

Sensors 2017, 17, 346

5 of 17

between each area of the single vulcanized MCF rubber, shown below in Table 2. In other words, the MCF rubber is also a pyroelectric material. There have been few reported investigations of pyroelectric rubber comparing to polymers [55–57]. Therefore MCF rubber is a precious material. It it possible to create a new kind of piezoelement by cutting out the optimal area from the MCF5 of 17 Sensorsmade 2017, 17, 346 rubber and adhering them to each other.

Figure 2. Production of layered piezoelementsutilizing utilizing MCF by by thethe application of of Figure 2. Production of layered piezoelements MCFrubber rubbervulcanized vulcanized application electric and magnetic fields. electric and magnetic fields.

Here, we therefore designate three types of piezoelement (Types 1–3), as follows: when a small

Figure shows the principle of on thethe production of aelectric singleand piece of vulcanized rubber quantity2 of MCF liquid was poured electrodes and magnetic fields wereMCF applied, as a piezoelement. Type 1 isMCF a single-layer rubber. thick On the other hand, we can two areas of vulcanized rubber werevulcanized adhered on MCF the electrode: MCF rubber on the consider multiple-layer rubber as other piezoelement. is because wei”discovered magnetic field areaMCF or permanent magnet area, shown This as “MCF rubber in Figure an 2; induced and voltage between each the single in Tablemagnet 2. In other comparatively thinarea MCFof rubber in the vulcanized non-magneticMCF field rubber, area, i.e.,shown outsidebelow the permanent area, as “MCFisrubber in Figure 2material. (these are There shownhave by a photograph in Figureinvestigations 3a). MCF words, theshown MCF rubber also a o” pyroelectric been few reported rubber i had the magnetic cluster configuration shown in Figure MCF 1, while MCFisrubber o did material. not. of pyroelectric rubber comparing to polymers [55–57]. Therefore rubber a precious Therefore, we considered a combination of several MCF rubber i’s with magnetic clusters. At first, It made it possible to create a new kind of piezoelement by cutting out the optimal area from the MCF two MCF rubber i's were extracted from two MCR rubbers, as shown on the left in Figure 2. Next, rubber and adhering them to each other. these MCF rubber i's were combined, and the combined MCF rubber was called Type 2. Self-adhesion hardening used resistance without any reduceby superfluous MCFpoints rubberofvulcanized by Table 2. Thewas electric andadhesive inducedtovoltage connectingcosts. different MCF rubber: electric and magnetic fields has surfaces that can be combined face to face as shown in Figure 2, left. Ω designates electric resistance and mV the induced voltage. However, the different surfaces created by the anode and cathode electrodes cannot be adhered due to the difference of configuration between the surfaces of the MCF rubber on the anode and cathode, 1 2 1 1 1 3 3 4 3 3



which is based on the different radical polymerization between the anode and cathode. On the other Ω vulcanized mV Ωelectric mV Ω mV hasΩtwo areas mVof MCF Ω rubber, mVi and o. hand, MCF rubber by and magnetic fields The induced between these Type A voltage 0.9 and 0 electric 8.7 resistance 0 1.8 M the 95varied 890positions k 36 on 11.8 Msurfaces 30 were Typeas B shown 2.1 in Figure 0 0.1 are340 k in 220 200 500 k 230 measured, 3.26 The results listed Table 2.700 k Type C 5.4 0 30 k 120 2.0 M 75 3.0 M 60 2.8 M 37

Here, we therefore designate three types of piezoelement (Types 1–3), as follows: when a small quantity of MCF liquid was poured on the electrodes and electric and magnetic fields were applied, two areas of vulcanized MCF rubber were adhered on the electrode: thick MCF rubber on the magnetic field area or permanent magnet area, shown as “MCF rubber i” in Figure 2; and comparatively thin MCF rubber in the non-magnetic field area, i.e., outside the permanent magnet area, shown as “MCF rubber o” in Figure 2 (these are shown by a photograph in Figure 3a). MCF rubber i had the magnetic cluster configuration shown in Figure 1, while MCF rubber o did not. Therefore, we considered a combination of several MCF rubber i’s with magnetic clusters. At first, two MCF rubber i's were extracted from two MCR rubbers, as shown on the left in Figure 2. Next, these MCF rubber i's were combined, and the combined MCF rubber was called Type 2. Self-adhesion hardening was used without any adhesive to reduce superfluous costs. MCF rubber vulcanized by electric and

Sensors 2017, 17, 346

6 of 17

magnetic fields has surfaces that can be combined face to face as shown in Figure 2, left. However, the different surfaces created by the anode and cathode electrodes cannot be adhered due to the difference of configuration between the surfaces of the MCF rubber on the anode and cathode, which is based on the different radical polymerization between the anode and cathode. On the other hand, MCF rubber vulcanized by electric and magnetic fields has two areas of MCF rubber, i and o. The induced voltage and electric resistance between the varied positions on these surfaces were measured, as shown in Figure 3. The results are listed in Table 2. Sensors 2017, 17, 346

6 of 17

(a)

(b)

Figure 3. MCF rubber vulcanized by the application of electric and magnetic fields: (a) photograph of

Figure 3. MCF rubber vulcanized by the application of electric and magnetic fields: (a) photograph MCF rubber; (b) position of MCF rubber on the measurement of induced and electric resistance by of MCFconnecting rubber; (b) position of MCF rubber on the measurement of induced and electric resistance by points. connecting points. Table 2. The electric resistance and induced voltage by connecting different points of MCF rubber: Ω designates electric resistance and mV the induced voltage.

1 , 2 by vulcanization with electric and magnetic fields, electric conductivity In the case of ①-② ①-① ①-③ ③-④ and the ③-③ through the inner rubber was enhanced by electric polymerization, MCF rubber approached Ω mV Ω mV Ω mV Ω mV Ω mV 1 , 1 except being an electric conductor, as reported in previous investigations [48–50]. In the case of Type A 0.9 0 8.7 0 1.8 M 95 890 k 36 11.8 M 30 for Type C, electric conductivity on the surface of MCF rubber i was no less high than that throughout Type B 2.1 0 26 0.1 340 k 220 700 k 200 500 k 230 1 . 1 1 was reduced the rubber in the case C, the theMcase37 of Typeof C 5.42 As 0 for30Type k 120 2.0electric M 75 conductivity 3.0 M 60 in 2.8 1 2 with ZnO was enhanced, owing to the ferroelectric substance ZnO. However, that in the case of Inelectric the casepolymerization of ①-②, by vulcanization withby electric and method. magnetic Thus fields,the electric conductivity 1 1 differed owing to the with dopant the new results of the inner rubber was enhanced bythe electric polymerization, and rubber 1 2 and 1 the 1 MCF by typethrough of dopant, as shown in Table 2. In cases of

, when the approached electric resistance being an electric conductor, as reported in previous investigations [48–50]. In the case of ①-①, except was small, the induced voltage was very small. This was because MCF rubber becomes an electric for Type C, electric conductivity on the surface of MCF rubber i was no less high than that throughout conductor due to the magnetic clusters, the MCF filler, or the KI doping, and the ions disappear so the rubber in the case of ①-②. As for Type C, the electric conductivity in the case of ①-① was reduced 1 , 3of 3 4 with 3 , 3 not that a voltage not be induced byZnO. itself. In the that cases of andZnO

only the type owing tomay the ferroelectric substance However, in the case ①-② was enhanced, of percolation thepolymerization MCF rubber,with but also thebytype conduction over surface of the MCF owing tothrough the electric dopant the of new method. Thus thethe results of ①-① type of dopant, as shown in Table 2.of Inthe the ions cases generated of ①-② andby ①-①, when the electric rubber differed createdby ferroelectric substances, because the application of only the was small,the theinduced induced voltage was very small. This was because MCF rubber larger. becomes an electricresistance field. Therefore, voltage exists and electric resistance becomes electric conductor due to the magnetic clusters, the MCF filler, or the KI doping, and the ions 1 3 The results in Table 2 indicate the combination of MCF rubbers i and o, corresponding to disappear so that a voltage may not be induced by itself. In the cases of ①-③, ③-④ and ③-③, not in Tableonly 2, can be surmised effective. Therefore we can consider another combination of plural MCF the type of percolation through the MCF rubber, but also the type of conduction over the surface rubbers. As MCF shown on the lower right side of Figure 2, MCF rubber vulcanized byapplication the electric field of the rubber created ferroelectric substances, because of the ions generated by the of only the electric thewas induced voltage with exists MCF and electric resistance becomes larger. only, corresponding to field. MCFTherefore, rubber o, combined rubber i face to face on the surface results in Table 2 indicate the combination of MCF rubberssurfaces i and o, corresponding to and ①-③cathode vulcanizedThe at the anode. This is called Type 3. The different on the anode in Table 2, can be surmised effective. Therefore we can consider another combination of plural MCF electrodes cannot themselves be adhered, which is the same as Type 2. As for Types 2 and 3, two MCF rubbers. As shown on the lower right side of Figure 2, MCF rubber vulcanized by the electric field rubbers were pressed together t 2.2 kg for 1 hour under 47 ◦ C with a self-adhesion effect as shown in only, corresponding to MCF rubber o, was combined with MCF rubber i face to face on the surface Figure vulcanized 2. Table 3 shows the thickness of the created rubber. at the anode. This is called Type 3. TheMCF different surfaces on the anode and cathode

electrodes cannot themselves be adhered, which is the same as Type 2. As for Types 2 and 3, two 3. Thickness MCF47rubber. MCF rubbers were pressed Table together t 2.2 kg forof1 created hour under °C with a self-adhesion effect as shown in Figure 2. Table 3 shows the thickness of the created MCF rubber.

Rubber Type A Rubber Type B Type Type CA Type B Type C

Type 1 (mm)

Type 2 (mm)

Type 3 (mm)

Table 3. Thickness of created MCF rubber.

3.00

Type 3.00 1 (mm) 3.00 3.00 3.00 3.00

3.00

1.90

Type 2 (mm) 2.50 3.003.00 2.50 3.00

Type 3 (mm) 2.50 1.90 3.00 2.50 3.00

The measurement procedures for the analysis of the electrical and mechanical properties were described in previous studies [48–50]. In those studies, these properties were measured for robotics,

Sensors 2017, 17, 346

7 of 17

The measurement procedures for the analysis of the electrical and mechanical properties were described in previous studies [48–50]. In those studies, these properties were measured for robotics, Sensors 2017, 17, 346 7 of 17 the voltage or electric current inside the rubber at compression and the shear motions at the application of D.C. voltage were considered as the electrical properties, and the strain-stress response for the the voltage or electric current inside the rubber at compression and the shear motions at the mechanical properties, during consecutive compressions. In the present we measured application of D.C. voltage were considered as the electrical properties, and the study strain-stress response the electrical property of the voltageduring or electric current on the MCF rubber at compression using the for the mechanical properties, consecutive compressions. In the present study we measured samethe NFE experimental as or inelectric the previous study [48–50]. Weatfirst measured the the voltage electrical property apparatus of the voltage current on the MCF rubber compression using same NFE experimental apparatus as in the previous study [48–50]. We first measured the voltage between the electrodes without installing a power supply, and the electric resistance between electrodes, between thedesignated electrodes without a power supply, and measured the electricthe resistance as shown in the figure ininstalling the previous study. We then voltagebetween between the electrodes, shown inbetween the designated figure in the previous We then measured the voltage electrodes whenasinstalled the electrodes. The formerstudy. measurement was for induced voltage between the electrodes when installed between the electrodes. The former measurement was for related to piezoelectricity, and the latter for interior voltage related to piezoresistivity. The paired induced voltage related to piezoelectricity, and the latter for interior voltage related to electrodes used had the same 3 mm × 3 mm square form. piezoresistivity. The paired electrodes used had the same 3 mm × 3 mm square form.

3. Piezoelectric Response

3. Piezoelectric Response

At first, we investigated the experimental withoutdoping. doping. Figure 4 shows a typical At first, we investigated the experimental results results without Figure 4 shows a typical time time variation result for the induced voltage with repeated application of pressure in the case of Type variation result for the induced voltage with repeated application of pressure in the case of Type A. A.

(a)

(b) Figure 4. Cont.

Figure 4. Cont.

Sensors 2017, 17, 346

8 of 17

Sensors 2017, 17, 346

8 of 17

Sensors 2017, 17, 346

8 of 17

(c) Figure 4. Typical results of time-lapse of induced voltage and pressure in the case of Type A. (a) Type 1;

Figure 4. Typical results of time-lapse of induced voltage and pressure in the case of Type A. (a) Type 1; (b) Type 2; (c) Type 3. (c) (b) Type 2; (c) Type 3. Figure 4. Typical results time-lapse of induced voltage and pressure in the casewith of Type (a) Type 1; The The performance of of the MCF rubber does not degrade over time theA.repetition. (b) Type 2; (c) Type 3. deviation of the measured induced in degrade the figure is within the the allowable limitThe of error. The performance of the MCF rubbervoltage does not over time with repetition. deviation

theinduced MCF rubber is stable. it can confirmedlimit by the experiment that the of theTherefore measured voltage in the Furthermore figure is within thebeallowable of error. Therefore The performance of the MCF rubber does not degrade over time with the repetition. The Types 2 and 3 has more stability than Type 1 and the cause is guessed to be due to the multilayer MCFdeviation rubber isofstable. Furthermore it can be confirmed byisthe experiment that Types the measured induced voltage in the figure within the allowable limit of2 and error.3 has structure. Figure 4a is different from that of Types 1 and A in the previous study, which was conducted moreTherefore stability than Type 1 and the cause is guessed to be due to the multilayer structure. Figure the MCF rubber is stable. Furthermore it can be confirmed by the experiment that 4a is with an insulation paper sheet, 0.121-mm in depth and more than 100-MΩ in resistance, inserted different from that of Types 1 and Athan in the previous study, was conducted an insulation Types 2 and 3 has more stability Type 1 and the causewhich is guessed to be due towith the multilayer between the MCF rubber and the upper electrode for stabilization of the induced voltage [50]. The Figure 4a isin different from that of Types 1100-MΩ and A in in theresistance, previous study, which between was conducted paperstructure. sheet, 0.121-mm depth and more than inserted the MCF voltage induced in the MCF rubber with doping had stable changes during the repetition, as shown with anthe insulation paper sheet, 0.121-mm in depth and more than 100-MΩ inThe resistance, inserted rubber and upper electrode for stabilization of the induced voltage [50]. voltage induced in Figure 5, even if the MCF rubber did not have an insulation sheet. Therefore, in order to compare in between the with MCF doping rubber and the upper electrode for stabilization of the as induced voltage [50].5,The the MCF rubber had changes during repetition, shown Figure even if the MCF rubber with doping westable show the case without anthe insulation sheet for Type 1.inRegarding the voltage induced in the MCF rubber with doping had stable changes during the repetition, as shown with the MCF rubberatdid notpressure, have an even insulation Therefore, tothe compare the MCF had rubber sensitivity small with asheet. different numbersinoforder layers, induced voltage the in Figure 5, even if the MCF rubber did not have an insulation sheet. Therefore, in order to compare same sensitivity. physical the induced voltage at low pressure had the same doping wehigh show the caseThe without an mechanism insulation of sheet for Type 1. Regarding the sensitivity at small the MCF rubber with doping we show the case without an insulation sheet for Type 1. Regarding the qualitative tendency; its cause is shown below. pressure, even with a different numbers of layers, the induced voltage had the same high sensitivity. sensitivity at small pressure, even with a different numbers of layers, the induced voltage had the The physical of the induced voltageofatthe low pressure hadatthe same qualitative same highmechanism sensitivity. The physical mechanism induced voltage low pressure had the tendency; same its cause is shown below.its cause is shown below. qualitative tendency;

(a) Figure 5. Cont.

(a) Figure 5. Cont.

Figure 5. Cont.

Sensors 2017, 17, 346

9 of 17

Sensors 2017, 17, 346

9 of 17

(b)

(c)

(d) Figure 5. Typical results of time-lapse of the induced voltage and the pressure in the cases of Types

Figure 5. Typical results of time-lapse of the induced voltage and the pressure in the cases of Types 2, 3 2, 3 and B, C. (a) Type B and 2; (b) Type B and 3; (c) Type C and 2; (d) Type C and 3. and B, C. (a) Type B and 2; (b) Type B and 3; (c) Type C and 2; (d) Type C and 3.

Next, we investigated the effect of doping on induced voltage. Figure 5 shows the results of MCF rubber Types B and C in Types 2 and 3. As denoted in Figure 4, in Figure 5 the performance of the MCF

Sensors 2017, 17, 346

10 of 17

we346 investigated the effect of doping on induced voltage. Figure 5 shows the results of10MCF SensorsNext, 2017, 17, of 17 rubber Types B and C in Types 2 and 3. As denoted in Figure 4, in Figure 5 the performance of the MCF rubber also does not degrade over time with the repetition. The deviation of the measured rubber also does not degrade time with the repetition. deviation of thethe measured induced induced voltage in the figureover is within the allowable limit The of error. Therefore MCF rubber has voltage in the figure is within the allowable limit of error. Therefore the MCF rubber has stability. Doping enhanced the induced voltage. In particular, Type C was the largest. The stability. induced Doping the induced voltage. particular, C was the effect largest. induced voltage in voltage enhanced in Type 3 was larger than that inInType 2. ThisType was due to the of The the configuration of the Type 3 was larger than that in Type 2. This was due to the effect of the configuration of the piled layers. piled layers. As A and and 33 in in Figure Figure 4c, 4c,ininType TypeofofBBand and2 2bybyFigure Figure5a5aand and Type and 3 As seen seen in in Type Type of of A Type CC and 3 in in Figure thetypical typicalcharacteristics characteristicsindicated indicatedthat thatatat small small pressure pressure the the induced Figure 5d,5d,the induced voltage voltage was was responsive to the pressure, while at large pressure it holds constant at a low value. This characteristic responsive to the pressure, while at large pressure it holds constant at a low value. This characteristic shows tendency regardless regardless of of the thetype typeof oflayer layerand andtype typeofofdopant. dopant.The Thecause cause shows the the same same qualitative qualitative tendency is is duetotothe theweek weekvulcanization vulcanizationofofthe theMCF MCF rubber rubber without without involving involving sulphur. sulphur. Therefore, due Therefore, the the MCF MCF rubber it recovers to its size by theby reduction of the pressure. If the sulphur rubberisiscompressed compressedbefore before it recovers toinitial its initial size the reduction of the pressure. If the is used, the vulcanization between the isoprene the NR-latex is more solidified and the MCFand rubber sulphur is used, the vulcanization between theofisoprene of the NR-latex is more solidified the recovers to its initial size upon reduction of the pressure. MCF rubber recovers to its initial size upon reduction of the pressure. In In Figures Figures 44 and and 5, 5, regarding regarding the the large large responsivity responsivityof of the the induced induced voltage voltage at at small small pressure pressure and and the the small small induced induced voltage voltage at at large large pressure, pressure, the the behavior behavior of of the the ions ions inside inside the the MCF MCF rubber rubber differed differed according according to to the the amount amount of of pressure. pressure. Figure Figure 66 shows shows typical typical results results of of the the comparison comparison between between the the induced symbols) and thethe voltage flowing inside the the MCF rubber withwith the inducedvoltage voltage(represented (representedbybyblack black symbols) and voltage flowing inside MCF rubber application of a voltage by white symbols). former corresponded piezoelectricity the application of a (represented voltage (represented by whiteThe symbols). The former to corresponded to and the latter to piezoresistivity. The latter was measured by the application of a DC 10 V power supply piezoelectricity and the latter to piezoresistivity. The latter was measured by the application of a DC using the same NFEusing experimental used in the previous study [48–50]. Thestudy arrows with 10 V power supply the same apparatus NFE experimental apparatus used in the previous [48–50]. numbers in the figure indicate the order from compression to decompression. The result is one cycle of The arrows with numbers in figure indicate the order from compression to decompression. The repetition. Because therepetition. MCF rubber has athe high sensing response at very smallresponse pressure,atthe induced result is one cycle of Because MCF rubber has a high sensing very small voltage occurs Pa and isoccurs enhanced up to0around by the compression, as indicated pressure, the around induced0 voltage around Pa and20iskPa enhanced up to around 20 kPa byfrom the 1compression, to 2 in the figure. Corresponding to this is the situation from 0 s to A, marked in red in Figure as indicated from 1 to 2 in the figure. Corresponding to this is the situation from 0 s5a. to The qualitative tendency of Figure 5c, qualitative as is showntendency in the previous discussion Figuresin4 and 5, and is A, marked in red in Figure 5a. The of Figure 5c, as isofshown the previous the same as of that in the other Figure 5, and directly in FigureFigure 6. The 5, cause discussion Figures 4 andfigures, 5, and including is the same as that in the otherdesignated figures, including and is the fact the positive and negative ions inside the MCF rubber approach but do not contact, as shown directly designated in Figure 6. The cause is the fact the positive and negative ions inside the MCF in Figure 7a. rubber approach but do not contact, as shown in Figure 7a.

Figure Figure 6.6. Typical Typical results results of of the the induced induced voltage voltage and and voltage voltage flowing flowing inside inside the the MCF MCF rubber rubber by by the the application of a voltage. application of a voltage.

Sensors 2017, 17, 346

11 of 17

Sensors 2017, 17, 346

11 of 17

Figure 7. Schematic diagram of the physical model of Piezo effect of MCF rubber.

Figure 7. Schematic diagram of the physical model of Piezo effect of MCF rubber. Figure 7 shows the physical mechanism of the induced voltage. The illustration shows the MCF rubber7isshows a ferroelectric substance with smaller electricity, where the electric resistance is larger Figure the physical mechanism of the inducedorvoltage. The illustration shows the MCF with smaller conductivity. At more than 20 kPa in Figure 6, the induced voltage is decreased as with rubber is a ferroelectric substance with smaller electricity, or where the electric resistance is larger indicated from 3 to 7, because the positive and negative ions come into contact, and the Ni particles smaller conductivity. At more than 20 kPa in Figure 6, the induced voltage is decreased as indicated aggregate such that the electron can pass through the MCF rubber, as shown in Figure 7c. Corresponding from 3 to 7, because the positive and negative ions come into contact, and the Ni particles aggregate to this is the situation from A to B, indicated in red in Figure 5c. The MCF rubber is a conductor with such that the electric electron can passand through the MCF rubber, as shown Figure 7c. Corresponding smaller resistance with smaller electricity. Therefore, thein voltage flowing within the MCFto this is the situation from A to B, indicated in red in Figure 5c. The MCF rubber is a conductor with smaller rubber upon application of a voltage is larger than that at less than 20 kPa. By decompression, electricthe resistance andthen with smaller electricity. Therefore, the voltage flowing within as the MCF rubber MCF rubber transits from being a conductor to being a ferroelectric substance, indicated from 8 to 11of in aFigure 6, or from Bthan to C,that indicated red in 5c.decompression, As a result, in MCF upon application voltage is larger at lessinthan 20Figure kPa. By the rubber, MCF rubber piezoelectricity occursa under small pressures piezoresistivity at high pressure. The MCF rubber then transits from being conductor to being and a ferroelectric substance, as indicated from 8 to 11 in is thus a hybrid of a ferroelectric substance and a conductor. Figure 6, or from B to C, indicated in red in Figure 5c. As a result, in MCF rubber, piezoelectricity Because the aim of the present report is the use of rubber with filler, we compared these results occurs under small pressures and piezoresistivity at high pressure. The MCF rubber is thus a hybrid of to those of piezoelements just made of rubber with filler as follows: in the case of MCF rubber from a ferroelectric and a conductor. Figure 6, substance as for piezoelectricity the effective piezoelectric coefficient d33 is estimated to be on the order Because the which aim ofwas themeasured present report is thethe use of rubber with filler, we compared of 10−10 C/N by referring study [54], and as for piezoresistivity, the these electricresults −4 to 10of −1 S/m. to those of piezoelements just10made rubber with filler as follows: in the case of MCF rubber from conductivity ranged from 33 is forpiezoelectricity piezoelectricity, the coefficient d33 indthe case of chloroprene rubber Figure 6, asAsfor theeffective effectivepiezoelectric piezoelectric coefficient estimated to be on the order −11 C/N order [12]. d33 in the case − 10 consisting of PbTiO 3 ceramic powder with 127-mm square plate is 10 of 10 C/N which was measured by referring the study [54], and as for piezoresistivity, the electric of silicon oil rubber conductivity ranged fromusing 10−4together to 10−1zirconate S/m. titanate (PZT) particles with cylinder shapes having 50-mm diameter and 2-mm thickness is on the order of 10−11 C/N 33 [16]. d33 in the case of silicon oil As for piezoelectricity, the effective piezoelectric coefficient d in the case of chloroprene rubber rubber combined by PZT particles of cylinder shape having 50-mm diameter and 2-mm thickness is consisting of PbTiO ceramic powder with 127-mm square plate is 10−11 C/N order [12]. d33 in the 10−12 C/N [18]. 3 case of silicon oil piezo-resistivity, rubber using together zirconate titanate particles with cylinder shapes having As for the electric conductivity of (PZT) acrylonitrile butadiene rubber containing − 11 33 50-mmfurnace diameter and 2-mm thickness is −2onS/m, theand order 10 changed C/N with [16].the d carbon in theblack case volume of silicon oil carbon black is from 10−8 to 10 the of order [14].by The electric conductivity of NR-latex combined by diameter sodium monmorillonite and rubberfraction combined PZT particles of cylinder shape having 50-mm and 2-mm thickness is −4 to 10−2 S/m, and also changed with the loading of the nanocomposite [15]. − 12 polypyrrole is from 10 10 C/N [18]. the comparisonthe withelectric the above results, the piezoelectric and piezoresistive effects in the As forFrom piezo-resistivity, conductivity of acrylonitrile butadiene rubber containing MCF rubber piezoelements are − larger than those of other filled rubber piezoelements. 8 − 2 furnace carbon black is from 10 to 10 S/m, and the order changed with the carbon black volume [14].ofThe electric conductivity of NR-latex combined by sodium monmorillonite and 4. fraction Verification the Piezoeffect of MCF Rubber polypyrrole is from 10−4 to 10−2 S/m, and also changed with the loading of the nanocomposite [15]. In the last section, we verify the suitability of the MCF rubber created here in contrast to the From the comparison with the above results, the piezoelectric and piezoresistive effects in the conventional piezoelements with filler and dopant, which are the traditional methods of enhancing MCF rubber piezoelements are larger than those of other filled rubber piezoelements.

4. Verification of the Piezoeffect of MCF Rubber In the last section, we verify the suitability of the MCF rubber created here in contrast to the conventional piezoelements with filler and dopant, which are the traditional methods of enhancing the

Sensors 2017, 17, 346

12 of 17

piezoeffect. However, previously Sensors 2017, 17, 346 those materials have always been produced by the drying method as 12 of 17 described. In contrast to the conventional piezoelements, the greatest feature of the present MCF theproduction piezoeffect. method However, materials alwaysand been produced byfor thevulcanization. drying methodInasorder rubber is those the use of bothhave magnetic electric fields previously described. In contrast to the conventional piezoelements, the greatest feature to compare with conventional piezoelements, the creation of MCF rubber without the of aidtheof the present MCF rubber production method is the use of both magnetic and electric fields for electric field should be considered. In order to verify the superiority of the MCF rubber produced vulcanization. In order to compare with conventional piezoelements, the creation of MCF rubber in the present study, we created other MCF Type 1 rubbers with filler and dopant using only the without the aid of the electric field should be considered. In order to verify the superiority of the MCF drying method, without the application of an electric field for vulcanization. Ni and Fe3 O4 were rubber produced in the present study, we created other MCF Type 1 rubbers with filler and dopant used using as fillers, corresponding to without Type A.the The doping was samefield as that used in Types and C. only the drying method, application of anthe electric for vulcanization. Ni B and We also dealt with two types of MCF rubber, both with and without the application of a magnetic Fe3O4 were used as fillers, corresponding to Type A. The doping was the same as that used in field Types duringB vulcanization. The drying time requires h for both the former MCF rubber 48 h for the and C. We also dealt with two types of MCF24 rubber, with and without the and application a magnetic field during The of drying time requires for the MCF rubber latter.ofThese differences werevulcanization. due to the effect the alignment of 24 Nihand Fe3former O4 particles created by and 48 h field for the These differences were due to the effect of the alignment of Ni and Fe3O4 the magnetic to latter. promote vulcanization. particles8 created magnetic fieldoftothe promote vulcanization. Figure shows by thethe time variation induced voltage under repeated application of pressure Figure 8 shows the time variation of the induced voltage under repeated application of pressure in the case of the MCF rubber created without using an electric field. On the other hand, as for Types A, in the case of the MCF rubber created without using an electric field. On the other hand, as for B and C created without either a magnetic field or electric field, the induced voltage of each Type was Types A, B and C created without either a magnetic field or electric field, the induced voltage of each zero (results not shown). Type was zero (results not shown).

(a)

(b) Figure 8. Cont.

Figure 8. Cont.

Sensors 2017, 17, 346

13 of 17

Sensors 2017, 17, 346

13 of 17

(c) Figure 8. The results of time-lapse of the induced voltage and the pressure on MCF rubber created

Figure 8. The results of time-lapse of the induced voltage and the pressure on MCF rubber created with the drying method only. (a) Type A; (b) Type B; (c) Type C. with the drying method only. (a) Type A; (b) Type B; (c) Type C.

Types A and 1 are the MCF rubbers corresponding to the conventional piezoelements with filler and without andMCF without or with a magnetic field. voltagepiezoelements with a magneticwith field filler Types A anddopant, 1 are the rubbers corresponding toThe the induced conventional as shown in Figure 8awithout was zero. the voltage was notinduced inducedvoltage by the filler forms as field and without dopant, and orTherefore, with a magnetic field. The withthat a magnetic magnetic clusters. other words, magnetic the clusters alonewas did not not generate without as shown in Figure 8aInwas zero. Therefore, voltage inducedinduced by the voltage filler that forms as doping, or, voltage was not induced by filler that did not form magnetic clusters. Next, we prepared magnetic clusters. In other words, magnetic clusters alone did not generate induced voltage without a conventional piezoelement with filler and dopant, corresponding to Types B and C without an doping, or, voltage was not induced by filler that did not form magnetic clusters. Next, we prepared electric field, and without or with a magnetic field. Figure 8b shows that Type B MCF rubbers with a a conventional piezoelement with filler and dopant, corresponding to Types B and C without an magnetic field had a large induced voltage. On the other hand, Figure 8c shows that the Type C MCF electric field, and orfield withhad a magnetic field.induced Figurevoltage. 8b shows thatfindings Type Bwere MCFdue rubbers rubbers with awithout magnetic no more than These to the with a magnetic field a large induced voltage. Onmagnetic the other hand,However, Figure 8cinshows that the Type C effects of the had doping on the configuration of the clusters. the case of doping MCF without rubbersawith a magnetic field had was no more than induced voltage. findings were due to the magnetic field, the voltage not induced by the filler aloneThese that did not form magnetic clusters, previously described. Therefore, the effect of doping on the magnetic clusters generated effects of the as doping on the configuration of the magnetic clusters. However, in the case of doping the a induced voltage. without magnetic field, the voltage was not induced by the filler alone that did not form magnetic In the case without doping, Therefore, Figure 4a shows the MCF rubber inon Types A and 1 that involvegenerated MCF clusters, as previously described. the effect of doping the magnetic clusters rubber i (as shown in Figure 2) consisting of magnetic clusters. In this case, an induced voltage was the induced voltage. generated. By comparing this result to Figure 8a, we can hypothesize that the result was due to In the case without doping, Figure 4a shows the MCF rubber in Types A and 1 that involve MCF electric polymerization with the aid of the electric field for vulcanization. Therefore, even if we do rubber i (as shown in Figure 2) consisting of magnetic clusters. In this case, an induced voltage was not use dopant, we can obtain a large induced voltage by electric polymerization. The electric generated. By comparing thisinresult to Figure 8a, we can hypothesize that the result was due to electric polymerization has a role generating an induced voltage in the MCF rubber. polymerization withthe theresults aid ofofthe electric field8 for vulcanization. Therefore, evenTypes if we2 do Meanwhile, Figures 4, 5 and suggest another conclusion regarding andnot 3. use dopant, can obtain of a large inducedinvoltage polymerization. The1,electric polymerization Thewe piezoelements MCF rubbers Types 2 by andelectric 3 (see Figure 5) involve Type which corresponds MCF rubber i shown in Figure 2, consisting of magnetic has a to role in generating an induced voltage in the MCF rubber.clusters and dopant. Therefore, it is possible to clarify the effect of the magnetic with dopantconclusion on the piezoelectric response Meanwhile, the results of Figures 4, 5 andclusters 8 suggest another regarding Types 2byand 3. comparing the of MCF rubbers withinand without field5)ininvolve the caseType of dopant. Thecorresponds voltage The piezoelements MCF rubbers Types 2 anda 3magnetic (see Figure 1, which generated in the case of the MCF rubber created with the help of an electric field (Figure 5) is the same to MCF rubber i shown in Figure 2, consisting of magnetic clusters and dopant. Therefore, it is possible as that of the MCF rubber created with a magnetic field and without an electric field (Figure 8b,c). to clarify the effect of the magnetic clusters with dopant on the piezoelectric response by comparing In Figure 8b,c, the effect of dopant on the magnetic clusters generated the induced voltage, as already the MCF rubbers with and without a magnetic field in the case of dopant. The voltage generated in the mentioned. Therefore, in the case of the MCF rubber as in Figure 8b,c, we can determine the effect of case of the rubber created with the help an electric fieldThe (Figure 5) is theinsame dopantMCF on the magnetic clusters generated theofinduced voltage. MCF rubbers Typesas2 that and 3of the MCF were rubber created with a magnetic field and without an electric field (Figure 8b,c). In Figure due to the effects of the magnetic clusters and the doping. Thus, the generating factor of 8b,c, the effect of dopantinonthe theMCF magnetic generated the induced voltage, as already mentioned. piezoelectricity rubber clusters has synergetic effects with the magnetic clusters, doping, and electric Therefore, inpolymerization. the case of the MCF rubber as in Figure 8b,c, we can determine the effect of dopant on the magnetic clusters generated the induced voltage. The MCF rubbers in Types 2 and 3 were due to the effects of the magnetic clusters and the doping. Thus, the generating factor of piezoelectricity in the MCF rubber has synergetic effects with the magnetic clusters, doping, and electric polymerization.

Sensors 2017, 17, 346

14 of 17

5. Conclusions The results from the consecutive experimental procedures can be summarized as follows. First the proposed MCF rubber piezoelements with two types of multilayered structure have two typical characteristics as follows: 1.

2.

As shown the typical results of Figure 4c and 5a,d, the typical characteristics of the piezoelement made of natural rubber with filler as MCF created with the help of electric and magnetic fields indicates that at small pressure, the induced voltage is responsive to the pressure, and at large pressure it holds constant at a low value. The cause is due to the weak vulcanization of the MCF rubber without involving sulphur. Therefore, if we desire responsivity at a small pressure, we only have to produce the MCF rubber without sulphur. This characteristic has the same qualitative tendency as the type of layer and kind of dopant. As shown the typical results of Figure 5b,c or Figure 6, at small pressure the induced voltage is highly responsive, but the conductivity is small; in other words, the MCF is a ferroelectric substance. The MCF rubber presents piezoelectricity. On the other hand, at large pressure the induced voltage is smaller, but the electric resistance is decreased such that the MCF rubber becomes a conductor. The MCF rubber presents piezoresistivity. Thus, the piezoeffect varies depending on the amount of pressure. This typical characteristic has the same qualitative tendency on the type of piled layers and the kind of dopant.

Concerning these characteristics as shown above in 2, the MCF rubber is a piezoelement hybrid with piezoelectricity and piezoresistivity. The hybrid piezoeffect is suitable for application of various engineering fields. These piezoelectric and piezoresistive effects can be utilized by changing the quantity of the pressure in robotics by some method. In addition, multilayered MCF rubber of Types 2 and 3 has more stability than the single-layered one designated as Type 1. The piezoelectricity estimated by the effective piezoelectric coefficient and the piezoresistivity by the electric conductivity of the MCF rubber piezoelement are superior to those of other filled rubber piezoelements. Secondly, regarding the piezoelectricity of the MCF rubber, we clarified how the induced voltage is generated by the magnetic clusters, doping, and electric polymerization as follows: as denoted the typical results of Figure 8a, the effect of magnetic clusters alone does not generate an induced voltage without doping. However, as denoted the comparison of typical results of 4a and Figure 8a, electric polymerization has a role in generating the induced voltage of MCF rubber. On the other hand, as denoted the typical results of Figure 8c, the effect of dopant on the magnetic clusters generates an induced voltage. This tendency does not depend on the type of piled layers. Thus the generating factor of piezoelectricity in MCF rubber shows synergetic effects between the magnetic clusters, doping, and electric polymerization. Acknowledgments: This work was supported in part by JSPS KAKENHI Grant Number JP 15K05886, and we are very grateful for this support. Author Contributions: For this research article, Kunio Shimada conceived, designed the experiments, performed the experiments, analyzed the data and wrote the paper; Norihiko Saga contributed reagents/materials/analysis tools. Conflicts of Interest: The authors declare no conflict of interest. The founding sponsors had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish the results. The authors declare no conflict of interest.

Sensors 2017, 17, 346

15 of 17

References 1.

2. 3.

4. 5.

6.

7. 8.

9. 10. 11. 12. 13.

14.

15.

16.

17. 18. 19.

20.

Akkemans, O.; Spronck, M.; Kluskens, T.; Offerein, F.; Saralidze, K.; Aarts, J.; Keshaniyan, P.; Redeker, E.S.; Dilien, H.; van Grinsven, B. Application of electrodeposited piezo-resistive polypyrrole for a pressure-sensitive bruxism sensor. Phys. Status Solidia Appl. Mater. Sci. 2016, 213, 1505–1509. [CrossRef] Talemi, P.; Delaigue, M.; Murphy, P.; Fabretto, M. Flexible polymer-on-polymer architecture for piezo/pyroelectric energy harvesting. ACS Appl. Mater. Interfaces 2015, 7, 8465–8471. [CrossRef] [PubMed] Elhadidy, H.; Abdelhamid, M.I.; Aboelwafa, A.M.; Habib, A. Structure and piezoelectricity of poly (styrene-co-acrylonitrile) copolymer doped with different dyes. Polym. Plast. Technol. Eng. 2013, 52, 1277–1284. [CrossRef] Pezzotti, G.; Asmus, S.M.F. Fracture behavior of hydroxyapatite/polymer interpenetrating network composites prepared by in situ polymerization process. Mater. Sci. Eng. A 2001, 316, 231–237. [CrossRef] Yamada, Y.; Morizono, T.; Sato, K.; Sibuya, H.; Shimohira, T.; Umetani, Y.; Yoshida, T.; Aoki, S. Proposal of a skilmate hand and its component technologies for extravehicular activity gloves. Adv. Robot. 2004, 18, 269–284. [CrossRef] Liu, P.; Liu, C.X.; Huang, Y.; Wang, W.H.; Fang, D.; Zhang, Y.G.; Ge, Y.J. Transfer function and working principle of a pressure/temperature sensor based on carbon black/silicone rubber composites. J. Appl. Polym. Sci. 2016, 133, 42979. [CrossRef] Zhang, C.M.; Liu, T.; Xu, Y.G. Piezo-absorbing effect of microwave absorbing composites with carbonyl iron particles as the filler. Int. J. Mater. Res. 2015, 106, 1086–1093. [CrossRef] Del Castillo-Castro, T.; Casillo-Ortega, M.M.; Encinas, J.C.; Franco, P.J.H.; Carrillo-Escalante, H.J. Piezo-resistance effect in composite based on cross-linked polydimethylsiloxane and polyaniline: Potential pressure sensor application. J. Mater. Sci. 2012, 47, 1794–1802. [CrossRef] Stubler, N.; Fritzsche, J.; Kluppel, M. Mechanical and electrical analysis of carbon black networking in elastomers under strain. Polym. Eng. Sci. 2011, 51, 1206–1217. [CrossRef] Liu, W.T.; Li, F.; Stefanini, C.; Chen, D.J.; Dario, P. Biomimetic flexible/complaint sensors for a soft-body lamprey-like robot. Robot. Auton. Syst. 2010, 58, 1138–1148. [CrossRef] Hasegawa, Y.; Shikida, M.; Shimizu, T.; Miyaji, T.; Sasaki, H.; Sato, K.; Itoigawa, K. Amicromachined active tactile sensor for hardness detection. Sens. Actuators A Phys. 2004, 114, 141–146. [CrossRef] Ogura, K.; Ogawa, M.; Ohya, K.; Banno, H. Receiving characteristics of d(31)-zero piezo-rubber hydrophone. Jpn. J. Appl. Phys. Part 1 1993, 62, 2304–2306. [CrossRef] Ruiz, M.M.; Marchi, M.C.; Perez, O.E.; Jorge, G.E.; Fascio, M.; D’accorso, N.; Negri, R.M. Structured elastomeric submillimeter films displaying magneto and piezo resistivity. J. Polym. Sci. Part 1 Polym. Phys. 2015, 53, 574–586. [CrossRef] Mahmoud, W.E.; El-Lawindy, A.M.Y.; El Eraki, M.H.; Hassan, H.H. Butadiene acrylonitrile rubber loaded fast extrusion furnace black as a compressive strain and pressure sensors. Sens. Actuators A Phys. 2007, 136, 229–233. [CrossRef] Pojanavaraphan, T.; Magaraphan, R. Fabrication and characterization of new semiconducting nanomaterials composed of natural layered silicates (Na+ -MMT), natural rubber (NR), and polypyrrole (PPy). Polymer 2010, 51, 1111–1123. [CrossRef] Mamada, S.; Yaguchi, N.; Hanasaka, M.; Yamamoto, M.; Yoshida, H. Matrix influence on the piezoelectric properties of piezoelectric ceramic/polymer composite exhibiting particle alignment. J. Appl. Polym. Sci. 2015, 132. [CrossRef] Landgraf, R.; Rudolph, M.; Scherzer, R.; Ihlemann, J. Modeling and simulation of adhesive curing process in bonded piezo metal composites. Comput. Mech. 2014, 54, 547–565. [CrossRef] Mamada, S.; Yaguchi, N.; Hanasaka, M.; Yamamoto, M.; Yoshida, H. Performance improvement of piezoelectric-rubber by particle formation of linear aggregates. J. Appl. Polym. Sci. 2014, 131. [CrossRef] Del Castillo-Castro, T.; Casillo-Ortega, M.M.; Herrera-Franco, P.J. Electrical, mechanical and piezo-resistive behavior of a polyaniline/poly(n-butyl methacrylate) composite. Compos. Part 1 Appl. Sci. Manuf. 2009, 400, 1573–1579. [CrossRef] Bloss, P.; DeReggi, A.S.; Glasel, H.J.; Hartmann, E.; Hirsch, D.; Schafer, H. Polarization distribution in ceramic-polymer nanocomposites. J. Appl. Phys. 1999, 86, 2712–2718. [CrossRef]

Sensors 2017, 17, 346

21.

22.

23.

24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43.

44.

16 of 17

Miao, Y.; Yang, Q.; Chen, L.; Sammynaiken, R.; Zhang, W.J. Modeling of piezoresistive response of carbon nanotube network based films under in-plane straining by percolation theory. Appl. Phys. Lett. 2012, 101, 063120. [CrossRef] Miao, Y.; Chen, L.; Sammynaiken, R.; Lin, Y.; Zhang, W.J. Note: Optimization of piezoresistive response of pure carbon nanotubes networks as in-plane strain sensors. Rev. Sci. Instrum. 2011, 52, 126104. [CrossRef] [PubMed] Miao, Y.; Chen, L.; Lin, Y.; Sammynaiken, R.; Zhang, W.J. On finding of high piezoresistive response of carbon nanotube films without surfactants for in-plane strain detection. J. Intell. Mater. Syst. Struct. 2011, 22, 2155–2159. [CrossRef] Shimada, K.; Fujita, T.; Oka, H.; Akagami, Y.; Kamiyama, S. Hydrodynamic and magnetized characteristics of MCF (magnetic compound fluid). Trans. Jpn. Soc. Mech. Eng. 2001, 67, 3034–3040. [CrossRef] Shimada, K.; Shuchi, S.; Kanno, H. Magnetic rubber having magnetic clusters composed of metal particles. J. Intell. Mater. Syst. Struct. 2005, 16, 15–20. [CrossRef] Shimada, K.; Miyazaki, T.; Shibayama, A.; Fujita, T. Extraction of magnetic clusters self-assembled by a magnetic field. Smart Mater. Struct. 2003, 12, 297–303. [CrossRef] Shimada, K.; Shuchi, S.; Kanno, H.; Wu, Y.; Kamiyama, S. Magnetic cluster and its applications. J. Magn. Magn. Mater. 2005, 289, 9–12. [CrossRef] Nishida, H.; Shimada, K.; Ido, Y. Microprocessing characteristics of inner surface of tube using magnetic functional fluid. Int. J. Appl. Electromagn. Mech. 2014, 45, 873–880. Nishida, H.; Shimada, K.; Ido, Y. Effectiveness of using a magnetic compound fluid with a pulsed magnetic field for flat surface polishing. Int. J. Appl. Electromagn. Mech. 2012, 39, 623–628. Nishida, H.; Shimada, K.; Ido, Y. Investigation of polishing for inner capillary walls by MCF flow method. Int. J. Appl. Electromagn. Mech. 2010, 33, 1685–1690. Sato, T.; Wu, Y.; Lin, W.; Shimada, K. Study of dynamic magnetic field assisted finishing for metal mold using magnetic compound fluid (MCF). Key Eng. Mater. 2010, 447–448, 258–262. [CrossRef] Wu, Y.; Sato, T.; Lin, W.; Yamamoto, K.; Shimada, K. Mirror surface finishing of acrylic resin using MCF-based polishing liquid. Int. J. Abras. Technol. 2010, 3, 11–24. [CrossRef] Wu, Y.; Sato, T.; Lin, W.; Yamamoto, K.; Shimada, K. The detailed performance of MCF polishing liquid in nano-precision surface treatment of acrylic resin. Adv. Mater. Res. 2009, 76–78, 331–336. [CrossRef] Shimada, K.; Kanno, H. Possibility of utilizing MCF in viscous damper of passive type. Int. J. Appl. Electromagn. Mater. 2007, 25, 119–125. Kanno, H.; Shimada, K.; Ogawa, J. A study on a MR damper utilizing magnetorheological fluids composed of different size of particles. Theoret. Appl. Mech. Jpn. 2006, 55, 91–96. Shimada, K.; Zheng, Y. Development of MCF rubber with temperature and electric senses for an element material in haptic robot sensor. Trans. Jpn. Soc. Mech. Eng. 2007, 73, 2269–2274. [CrossRef] Shimada, K.; Zheng, Y. Study on electric conductivity under tensile condition of MCF rubber for haptic robot sensor. Trans. Jpn. Soc. Mech. Eng. 2008, 74, 1748–1754. [CrossRef] Shimada, K.; Zheng, Y. Basic study on tensile characteristics of MCF (magnetic compound fluid) rubber. Trans. Jpn. Soc. Mech. Eng. 2008, 74, 2287–2294. [CrossRef] Zheng, Y.; Shimada, K. Basic study on usefulness of magnetic compound fluid (MCF) rubber as material for microwave heating. J. Solid Mech. Mater. Eng. 2009, 3, 158–166. [CrossRef] Shimada, K. Theoretical investigation by quantum mechanics on the tunnel diode effect of electric conductive characteristics and haptic sensing in MCF rubber. J. Sens. 2010, 2010, 731854. [CrossRef] Shimada, K.; Hayasaka, T. Principle and characteristics of high sensitive haptic MCF rubber perceptible to shear force as well as normal force utilizing intelligent fluid. J. Jpn. Sys. Exp. Mech. 2012, 12, 288–294. Shimada, K.; Hayasaka, T. Investigation of applicability of high sensitive haptic MCF rubber perceptible to shear force as well as normal force utilizing intelligent fluid. J. Jpn. Sys. Exp. Mech. 2012, 12, 295–301. Shimada, K.; Hayasaka, T. Material dynamics and tribological characteristics inner high sensitive haptic MCF rubber perceptible to shear force as well as normal force utilizing intelligent fluid. J. Jpn. Sys. Exp. Mech. 2012, 12, 302–307. Shimada, K.; Hayasaka, T. The effect of fingerprint on shear haptic sensing of high sensitive haptic MCF rubber sensor perceptible to shear force as well as normal force utilizing intelligent fluid. J. Jpn. Sys. Exp. Mech. 2012, 12, 308–314.

Sensors 2017, 17, 346

45.

46.

47. 48.

49.

50.

51.

52.

53. 54. 55. 56.

57.

17 of 17

Shimada, K.; Hayasaka, T. Application of high sensitive haptic MCF rubber perceptible to shear force as well as normal force utilizing intelligent fluid to sensor on robot—In case of walk and pushup. J. Jpn. Sys. Exp. Mech. 2012, 12, 315–322. Shimada, K.; Hayasaka, T. Application of high sensitive haptic MCF rubber perceptible to shear force as well as normal force utilizing intelligent fluid to sensor on robot—In case of thrust, rotation and bending elbow. J. Jpn. Sys. Exp. Mech. 2012, 12, 323–331. Shimada, K.; Zheng, Y.; Saga, N. Experimental investigation on technique to read convex shape by MCF rubber sensor utilizing robot action. J. Jpn. Sys. Exp. Mech. 2014, 14, 86–93. Shimada, K.; Saga, N. Mechanical enhancement of sensitivity in natural rubber using electrolytic polymerization aided by a magnetic field and MCF for application in haptic sensors. Sensors 2016, 16, 1521. [CrossRef] [PubMed] Shimada, K.; Saga, N. Detailed mechanism and engineering applicability of electrolytic polymerization aided by a magnetic field in natural rubber by mechanical approach for sensing (Part 1): The effect of experimental conditions on electrolytic polymerization. World J. Mech. 2016, 6, 357–378. [CrossRef] Shimada, K.; Saga, N. Detailed mechanism and engineering applicability of electrolytic polymerization aided by a magnetic field in natural rubber by mechanical approach for sensing (Part 2): Other and intrinsic effects on MCF rubber property. World J. Mech. 2016, 6, 379–395. [CrossRef] Shimada, K. Quantum mechanical investigation of the electric and thermal characteristics of magnetic compound fluid as a semiconductor on metal combined with rubber. ISRN Nanotechnol. 2011, 2011, 259543. [CrossRef] Caro, M.A.; Zhang, S.; Riekkinen, T.; Ylilammi, M.; Moram, M.A.; Lopez-Acevedo, O.; Molarius, J.; Laurila, T. Piezoelectric coefficients and spontaneous polarization of ScAIN. J. Phys. Condens. Matter 2015, 27, 245901. [CrossRef] [PubMed] Zhao, J.; Goo, N.S.; Ha, D.H. Design and performance evaluation of a d33-mode piezocomposite electricity generating element. J. Mech. Sci. Technol. 2014, 28, 15–23. [CrossRef] Ren, W.; Zhou, H.; Wu, X.; Zhang, L.; Yao, X. Measurement of piezoelectric coefficients of lead zirconate titanate thin films by the normal load method using a composite tip. Mater. Lett. 1997, 31, 185–188. [CrossRef] Mottamchetty, V.; Chaudhary, A.K. Improved design of THz spectrometer using LT-GaAs photoconductive antennas, pyroelectric detector and bandpass filters. Indian J. Phys. 2016, 90, 73–78. [CrossRef] Zeqiri, B.; Zauhar, G.; Hodnett, M.; Barrie, J. Progress in developing a thermal method for measuring the output power of medical ultrasound transducers that exploits the pyroelectric effect. Ultrasonic 2011, 51, 420–424. [CrossRef] [PubMed] Mauzac, M.; Nuyen, H.T.; Tournilhac, F.G.; Yablonsky, S.V. Piezoelectric and pyroelectric properties of new polysiloxane smectic c-asterisk elestomers. Chem. Phys. Lett. 1995, 240, 461–466. [CrossRef] © 2017 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).