1. Introduction
We conducted sequential studies that led to the novel method of production of sensors made from soft rubber that was developed for artificial skin, which is expected to be required as a substitute for human or robot skin, such as hybrid skin (H-skin) [
1,
2,
3,
4,
5,
6,
7,
8,
9]. Through electrolytic polymerization, rubbers with C=C bonds, such as natural rubber latex (NR-latex), are capable of cross-linking molecules adequately to be solidified in themselves. This novel solidification method is different from the usual vulcanization technique where sulfur is used at the field for the ordinary production of solid rubber. Via electrolytic polymerization, we compound a magnetic responsive liquid, i.e., magnetic compound fluid (MCF), which has nm-ordered magnetite (Fe
3O
4) particles that are obtained by using a magnetic fluid (MF) during compounding and has μm-ordered metal particles such as Fe, Ni, etc. By the application of a magnetic field on the compounded liquids with NR-latex and MCF during electrolytic polymerization, the magnetic particles are fabricated as heterostructures as many thin-rod shaped clusters that can be observed in organic thin film solar cells. Therefore, the electric and photovoltaic properties are enhanced, and the mechanical properties become anisotropic as well. Therefore, we named the electrolytically polymerized MCF rubber as H-Skin [
7,
8,
9]. By doping the MCF rubber using any dopant, the electric and photovoltaic properties change as shown in
Table A1 of the
Appendix A. As diene-based rubbers involving isoprene rubber (IR), chloroprene rubber (CR), butadiene rubber (BR), nitrile rubber (NBR) or styrene-butadiene rubber (SBR) also have C=C bonds, they can be electrolytically polymerized. However, their Mooney viscosity must be so small such that they may be easily compounded into the MCF by adjusting the concentration of the rubber molecules. In the present study, we use NR-latex and CR-latex.
Our previous investigations have focused on the body of the sensor, but not on the fabrication of electrodes attached to the rubber. The effect of the electrode fabrication on the characteristics of the electrolytically polymerized MCF rubber should be investigated. In general, the contact resistance between the electrodes and rubber is considerably large to become an issue [
10]. Moreover, with respect to the MCF rubber, the contact resistance must be reduced such that the adhesion between the electrodes and MCF rubber has higher sensitivity. This problem is related to the general research themes of adhesion between the metal and rubber [
11,
12,
13,
14,
15,
16,
17] and the prevention of corrosion of the metal surface [
18,
19,
20,
21]. Hence, various methods of adhesion between the metal and rubber have been hitherto proposed. For example, a typical technique is the development of rubber or adhesive, involving conductive adhesives [
22,
23].
On the other hand, dimethylpolysiloxane (PDMS) is one of conventional soft material. There have being investigations on the bonding between PDMS and metallic or non-metallic material related to patterning with gold for a soft sensor [
24]. PDMS is suitable for micro-fluidic system applications and because of transparency. PDMS generally takes on silicone oil with liquid state and the silicone oil rubber is produced with using resin to PDMS. As the silicone oil rubber is one of non-diene-based rubber without double bond C=C, PDMS may be a conventional material to produce a non-diene rubber. In contrast, the diene-based rubber involving NR, CR et al. can have many hybrid properties by conductive, photovoltaic et al. effects because it has double bond C=C. The use of the diene-based rubber such as the present MCF rubber is suitable for the purpose of rendering the rubber sensory. Therefore, in the present report, we investigate the bonding between diene rubber and metallic material.
In the present study, we propose a novel method for adhesion between the electrodes and MCF rubber by using metallic or non-metallic hydrous oxide. We investigate the mechanism of the adhesion and show the fabrication of the sensor using MCF rubber and metal electrodes.
3. Fabrication of Sensor
Based on the results of
Figure 8, we can propose five types of sensors fabricated by the adhesion of electrodes to MCF rubber as shown in
Figure 9. Types I–III are applicable for the field to be requested in rubber’s stretching motion or in sensing electric property at shear motion such that the sensor’s motion is parallel to shear force. In particular, type III is appropriate for artificial skin installed on a robot such as the H-Skin that we have proposed such that electrodes are fine lines and do not interfere with the skin’s dynamic motion. Type IV is applicable for sensor such as piezo element and is appropriate to sense electric property under pressure applied perpendicularly to the electrode. This type is not suitable for sensor’s shear motion because the touching surface of the sensor is not a soft rubber, rather a solid electrode. Type V is applicable for joining between electrodes for sensors.
Basic principle is MCF rubber liquid with hydrate is adhesive bonding between MCF rubber without hydrate and metal electrode.
As for Types III and IV, the production procedure is shown in
Figure 10 and
Figure 11, respectively. Regarding
Figure 10, at first, MCF rubber liquid with hydrate inserted by two electric wires with several thin wires as shown in
Figure 12a is sandwiched between two beforehand electrolytically polymerized solid MCF rubbers, and then a constant voltage and electric current are supplied under the wires as anode and the solid MCF rubbers as cathode such that the wires are adhered to the MCF rubber with hydrate. Next, a constant voltage and electric current are supplied under the solid MCF rubbers as cathode or anode such that the solid MCF rubbers are adhered to the MCF rubber with hydrate. The completed sensor is shown in
Figure 12c. Where the wire as shown in
Figure 12a has about φ1.3 mm outer diameter with seven thin silver-gilt electric wires with about φ0.1 mm diameter and about 5 mm length. Although the number of thin silver-gilt wires is less and their diameter is very small, the wires cannot be detached through the electrolytically polymerized MCF rubber affixed around each thin wire as shown in
Figure 12b. Therefore, the wire electrodes cannot be detached by elongation as shown in
Figure 12e under settlement of the MCF rubber sensor at a commercial, small-size tension/compression test machine (SL-6002, IMADA-SS Co. Ltd., Toyohasi, Japan) as shown in
Figure 12d at initial settlement figure.
Regarding
Figure 11, at first, MCF rubber liquid without hydrate is poured on a metal affixed by double electroconductive adhesive tape which should be one electrode of sensor, and then a constant voltage and electric current are supplied under the metal as anode such that the metal is adhered to the MCF rubber. The cause of using the adhesive tape is that both metals as electrodes of sensor cannot be adhered simultaneously to the MCF rubber with hydrate by reversing the electric poles many times as follows because of opposite position of the two metals if not use the adhesive tape. In general, the adhesive tape includes NR-latex, therefore, by electrolytic polymerization MCF rubber liquid can be piled on the adhesive tape. Next, MCF rubber liquid with hydrate is poured on the electrolytically polymerized solid MCF rubber without hydrate (the poured surface of the solid MCF rubber is concave and convex surface is more appropriate for adhesion), and another metal, which should be the other electrode of the sensor, is placed on the MCF rubber liquid with hydrate. In addition, then a constant voltage and electric current are supplied under the solid MCF rubber as anode and another metal as cathode such that the solid MCF rubber is adhered to the MCF rubber with hydrate. Lastly, the electric poles of electrolytic polymerization are reversed, and then a constant voltage and electric current are supplied again such that another metal is adhered to the MCF rubber with hydrate. The completed sensor is shown in
Figure 12f. Where metal is stainless steel with 0.1 mm thickness.
The completed sensor as shown in
Figure 12c,f was consisted of MCF rubber liquid with hydrate as 1 g Ni powder, 0.75 g MF (W-40), 3 g NR-latex, 3g CR-latex (671A), and 0.5 g hydrates, and MCF rubber without hydrate as 3 g Ni powder, 0.75 g MF (W-40), 3 g NR-latex, and 3g CR-latex (671A) with 1 mm metal plates gap, constant electric field at 6 V, 2.7 A and 5 min under atmosphere, 312 mT magnetic field by permanent magnets as paired opposites at each electrolytic polymerization.
Using the tension/compression test machine (SL-6002), we investigated the resistant force of electrodes for detachment from the sensor for the completed sensors of Types III and IV as shown in
Figure 13a, b, respectively. As for type III, when one side of electric wires was elongated by fixing a part of the rubber as shown in
Figure 12d, durable tensile force during adhesion was measured. As for type IV, when the electrolytically polymerized MCF rubber without the hydrate was adhered to 0.1 mm-thick stainless steel using the MCF rubber liquid with the hydrate, and was elongated by fixing the stainless steel on the base of the test machine as shown in
Figure 12g, durable tensile force during adhesion was measured. The contact area between stainless steel and MCF rubber is about 12 mm × 15 mm. Cross marks in the figures indicate the detachment of the wire from the sensor. As for multi-wire type shown in
Figure 13a, because the silver-gilt wires are very thin and can be made with the same length, the durable tensile force is appropriate for evaluation by strength rather than by pressure, which is different from the arrangement in the case of the plate type in
Figure 13b. The used hydrates are Na
2WO
4·2H
2O (indicated as “W” in the figure), Na
2MoO
4·2H
2O (“Mo”,) and Na
2B
4O
7·10H
2O (“B”).
Figure 13 also shows the case of Na
2WO
4·2H
2O, i.e., the case without a magnet during the production of the sensor.
As for multi-wire-type sensor, the detachment in the case of Na
2B
4O
7·10H
2O is not presented because the durable tensile strength is very large. Experimental results confirmed that this type of sensor is durable up to about 10 N, which corresponds to 1 kg weight loaded on the electric wire of the sensor, as shown in
Figure 12e.
In the case of multi-wire type, the durable tensile strength in the case of Na2B4O7·10H2O is the highest. However, in the case of plate type, the highest value sis obtained for Na2WO4·2H2O. The cause can be hypothesized as follows. The area of electrolytic polymerization at the multi-wire type is smaller than that at the plate type. The difference in the electrolytic polymerization is due to the different kinds of hydrate.
On the other hand, no effect of magnetic field during electrolytic polymerization can be seen. However, the using of magnets is important to collect the MCF rubber liquid by using the magnetic field during sensor production.
Finally, we investigate the electric characteristics of completed Types III and IV sensors. The former is appropriate for sensing in shear motion. Therefore, we used the experimental apparatus for the measurement of electrical resistivity of the MCF rubber that is moving at 0.164-N normal force and 5-mm/s sweeping velocity to scrape a body with some surface roughness, which was used in the previous study [
1]. The voltage 10 V was supplied between the two electric wires by connecting the electric resistance of 1.8 kΩ. The MCF rubber sensor was made to rub a flat plate with a surface roughness of
Ra = 20.86 μm,
Ry = 199.9 μm, and
Rq = 26.89 μm. The MCF rubber sensor was moved parallel to the material surface by an actuator with a constant speed and 50 mm scraping distance under a normal force which is presented as the one during the initial movement in the figure. A hard, non-electric body with φ0.5 mm diameter was interposed between the MCF rubber sensor and the acrylic resin body so that the MCF rubber sensor could be contacted exactly. The experimental procedure is referred to as the shear force experiment (SFE) [
1]. The result is shown in
Figure 14a, with a comparison with results for bare electrolytically polymerized MCF rubber without a hydrate, which means that the MCF rubber is not fabricated as type III sensor and is a sole MCF rubber indicated as “beforehand electrolytically polymerized MCF rubber without hydrate” delineated in
Figure 10. The electrical resistivity of completed sensor is more stable without perturbation than that of bare MCF rubber.
Furthermore, the type IV sensor is appropriate for sensing during pressing. Therefore, we used the experimental apparatus for measuring the electrical resistivity of the MCF rubber that is pressed by applying a normal force, which was used in the previous study [
1]. The voltage 10 V was supplied between the two electric wires by connecting the electric resistance of 1.8 kΩ. Using a tension/compression test machine (SL-6002), the MCF rubber was placed between two bodies, i.e., 7-mm square stainless steel plates. The upper body was moved to touch the lower one by an actuator at a pressing speed of 10 mm/min. The experimental procedure is referred to as the normal force experiment (NFE) [
1]. The result is shown in
Figure 14b, and compared with the results obtained when using the bare electrolytically polymerized MCF rubber without the hydrate, which means that the MCF rubber is the same that the bare MCF rubber in
Figure 14a, and of various types of MCF rubber sensors. The first MCF rubber sensor is the MCF rubber just sandwiched by two stainless steel plates as electrodes without any adhesives (which is named “MCF rubber sensor 1” here). The second one is the MCF rubber sandwiched to be adhered to two stainless steels as electrodes with double electroconductive adhesive tape (which is named “MCF rubber sensor 2” here). The third and the fourth ones are MCF rubber sensor made by the production procedures (a) and (b) in
Table 2 and
Table 3, respectively. The latter corresponds to type IV made as shown in
Figure 11 (which is named “MCF rubber sensor (b)” here), and the former is the simple case of type IV as described as follows (which is named “MCF rubber sensor (a)” here). The MCF rubber liquid without the hydrate is poured on a metal affixed using the double electroconductive adhesive tape to obtain one electrode of the sensor, and then, a constant voltage and electric current are supplied under the metal as the anode such that the metal is adhered to the MCF rubber.
Next, the MCF rubber liquid with the hydrate is poured on the electrolytically polymerized solid MCF rubber without a hydrate, and another metal that should be the other electrode of the sensor is placed on the MCF rubber liquid with the hydrate. These steps are the same as those in case (b). The next step is different. A constant voltage and electric current are supplied under the solid MCF rubber as the cathode and another metal as the anode such that the solid MCF rubber and another metal are simultaneously adhered to the MCF rubber with the hydrate.
In the case of bare MCF rubber, the initial electrical resistivity without pressure is the larger than that in case of any MCF rubber sensors; however, upon pressing, it becomes smaller by ~15 Ω m, which is smaller than those in case of any MCF rubber sensors. However, because electrodes are needed in a sensor, the bare MCF rubber does not fulfill the requirement. In the case of MCF rubber sensor 1, the electrical resistivity in the smallest pressure range is smaller than that of bare MCF rubber. In the case of MCF rubber sensor 2, the electrical resistivity in the smallest pressure range is smaller than that of MCF rubber 1. If an adhesive tape is not used, the electrodes are easily detached from the rubber sensor. Therefore, these results indicate that MCF rubber sensor 2 is superior to MCF rubber sensor 1. In the case of MCF rubber sensor (b), the initial electrical resistivity in the smallest pressure range is smaller than that of MCF rubber 2 by ~60 Ω m, which is 10-ordered Ωm the same as that of bare MCF rubber and is smaller than that of any other MCF rubber sensors. The electrical resistivity does not vary with increasing pressure. The results indicate that the sensor produced by the present adhesive production technique with the hydrate is superior to the MCF rubber sensor made without firm adhesion of electrodes to the sensor (e.g., MCF rubber sensor 2). Furthermore, they also indicate that we can achieve electric conductivity between rubber and metal smaller from the beginning of application of pressure by using the present adhesive production technique. In the case of the MCF rubber sensor (a), the electrical resistivity is larger than that of the MCF rubber sensor (b). The cause of the results is that the coordination of electric current’s passing inner the MCF rubber fabricated by particle and molecules of the MCF rubber is intensified by multiple electrolytic polymerization. Therefore, the present adhesive production technique corresponding to (b) in
Table 2 and
Table 3 is superior to that corresponding to (a) in
Table 2 and
Table 3. On the other hand, the cases of MCF rubber sensor 1, 2, (a) and (b) are abruptly changed by enhanced pressure around the initial pressure as indicated “p” in
Figure 14b. This tendency has more agile switching effect by the least pressure rather than the case of bare MCF rubber as indicated “q” in
Figure 14b.
As for reliability and durability of the present fabricated sensor, we suggest as follows. According to the period of electrolytic polymerization, the secular change of property of the fabricated sensor differs. The cause is due to the water involved in the fabricated sensor. Because of the use of NR or CR, water gets into the MCF rubber. The volatility differs according to the degree of electrolytic polymerization: if the period of the electrolytic polymerization is short, the amount of water imported into the structure of the MCF rubber’s molecular and particles is little, and the MCF rubber sensor is wet. Because the volatility of water occurs, the property of the MCF rubber sensor changes in time; on the contrary, if the period is long, the amount of water imported into the structure is much, and the MCF rubber sensor is arid. Because of small volatility of water, the property of the MCF rubber sensor holds constant in time. Therefore, the reliability and durability of the fabricated sensor depends on the electrolytic polymerization period. Thus, consequence is also dependent on the kind of fabrication of five sensors as shown in
Figure 9, for example, type IV has feasibility of longer durability than type III. Concerning to enhancement of the durability, we can propose many methods, for example, we displace other kind of the rubber. CR is more suitable for reducing volatility of water than NR. Therefore, we used CR in the present study. On the other hand, we can prose the others and intend to report the results in other reports.
4. Conclusions
To produce a sensor with electrodes, we proposed a novel adhesive technique with the use of rubber with C=C bonds, such as MCF rubber, NR-latex, and CR-latex compounded with a hydrate by electrolytic polymerization under a magnetic field. The recommendable hydrate is a metallic or non-metallic element with more than two valency, i.e., Na
2WO
4·2H
2O, Na
2MoO
4·2H
2O, and Na
2B
4O
7·10H
2O as one of the metal complexes. However, the hydrate with Cl
−, OH
−, C
6H
5O
73−, PO
43−, NO
3−, SO
42−, and NH
4+ cannot contribute to the adhesion effect. The metal that can be adhered on the MCF rubber with the hydrate is stainless steel, iron, and lead, but those that cannot be adhered are aluminum, titanium, nickel, zinc, brass, and copper. MCF rubber is composed of NR-latex or CR-latex. The anionic radical hydrate is reacted with isoprene molecules to realize cross-linking between them so that the MCF rubber with the hydrate is solidified (chemical reaction Equation (2)). By using the adhesive technique, we prepare five sensors fabricated with metal and rubber, as shown in
Figure 9. We can also electrolytically polymerize an electroconductive adhesive tape. By using the present adhesive technique, we can reduce the contact resistance between the metal and rubber. In the case of the wire-type sensor, the durable tensile strength in the case of Na
2B
4O
7·10H
2O is the largest. However, in the case of the plate-type sensor, the durable tensile strength in the case of Na
2WO
4·2H
2O is the largest. The former sensor is applicable to sense electrically during shear motion of the sensor, and the latter is suitable upon pressing to the sensor. The former sensor has stable sensing without perturbation. Furthermore, the former sensor is durable for tension up to about 1 kg. The latter sensor has the feasibility of agile switching effect by small pressure.
The adhesive technique proposed in the present study is applicable not only to produce MCF rubber sensor but to produce an ordinary sensor made of rubber and to realize cohesion between rubber and metal. In particular, the sensor with adhesion of the thin electric wire on the rubber, such as type III sensor, is effective to produce artificial skin made of rubber over which electric wires spread on a robot like a human skin with nerves. Furthermore, the cohesion between rubber and metal in the case of the hydrate-containing fabricated metallic or non-metallic element with more than two valency under electrolytic polymerization is novel enough to expand to vast engineering applications that involve the coating of rubber on metal surfaces to prevent corrosion.