Figure 5b shows the experimental voltammogram recorded between +0.8 V, as anodic potential limit and −2.0 V, as cathodic potential limit, with 6 mV·s⁻¹ the scan rate. Only one cathodic maximum, at −0.850 V, and one anodic maximum, at 0.356 V, are observed. The shape of the obtained voltammogram and peak potentials are similar to those reported for thinner pPy/DBS films [94–96] in aqueous solutions and quite different to those reported for pPy/ClO₄⁻ films [97]. Here a larger potential separation between anodic and cathodic peaks is observed, which can be attributed to use of a thicker film. These results indicate that dodecyl benzene sulfonate (DBS⁻) is the prevailing counterion inside the material. Due to its large size, the counterion remains trapped, promoting an exchange of Li⁺ cations predominately during oxidation/reduction processes. This interchange will promote the insertion of cations and polymeric swelling during reduction and the cations expulsion with polymeric shrinking during oxidation. These volume changes were confirmed by the sense of the angular movements of a pPy/tape actuator under the flow of anodic and cathodic currents.

The electrochemical reaction can be written in a simplified form as: (8) ( pPy   chain ) s ( DBS − Li + ) n ↔ ( pPy   chain n + ) ( DBS − ) n + n   Li + n   e −

The reaction occurring from left to right is the anodic process in which electrons are extracted from the polymeric chains. In Figure 5b, this process occurs during the anodic currents giving the maximum at 0.356 V. The charge extracted during oxidation can be obtained by integration of the maximum. The reaction from the right side towards the left side is the cathodic process, that is, electrons are introduced into the polymeric chains eliminating positive charges. This process is related to the cathodic currents originating the maximum at −0.850 V (negative currents on the voltammogram). The charge injected during reduction can be obtained by integration of this maximum.

In order to study the response of the material to different experimental variables, a galvanostatic procedure (Figure 6) was designed.

After stabilization of the initial oxidation state by applying a constant current of −0.150 mA for 250 s, the material film was submitted to three consecutive square waves of current (Figure 6a). The chronopotentiometric responses obtained during consecutive oxidation/reduction processes were recorded (Figure 6b).

The pPy/DBS film was submitted to different square waves of current ranging from ±1.5 to ±24 mA, passing a constant charge of ±180 mC. Figure 7 shows the obtained anodic chronopotentiograms. The potential evolves at higher potentials for higher currents. The higher initial step of the potential for rising currents is related to the different resistances present in the system: film resistance, interface resistance due to ions interchange, solution resistance and reaction resistance. After this initial change, the potential increases for rising anodic currents following the electrode processes. The electrical energy (E_e) consumed by the polymer film during each oxidation/reduction was calculated as E_e = i·∫E·dt, with i the constant current, E the electrodic potential at any time, t, of the current flow. Figure 7b shows the variation of the electrical energy as a function of the applied current. A linear fit was obtained for both cathodic and anodic processes: the material under reaction senses the applied current.

In reaction (1) or (2), as for any other chemical or electrochemical process, the reaction rate (anodic or cathodic) is expected to increase for rising experimental temperatures, due to the Arrhenius dependence of the reaction kinetic coefficient with T. This is equivalent to saying that working at a constant reaction rate, under flow of a constant current; the reaction is expected to occur with a lower reaction resistance (that means lower potentials) for increasing temperatures. In order to gain more knowledge into the sensing abilities of the electroactive material, the film is submitted to square current waves (±4 mA for 60 s every current) at different temperatures (15, 20, 25, 30, 35 and 40 °C). Figure 8a shows the obtained chronopotentiograms in which the evolutions of the reactive film potential for the different temperatures were overlapped. Figure 8b shows the response for the last oxidation step referred to the same arbitrary zero initial potential, in order to obtain the consumed electrical energies by integration of the curves. As expected, for the anodic processes decreasing potentials are observed for the same times of current flow and increasing temperatures. Rising available thermal energies require lower consumption of electrical energies during the reaction. Under constant current this produces lower potentials. Figure 9 shows the linear increase of the consumed electrical energy for decreasing working temperatures [92]. As a partial conclusion the electroactivity of the material, understood as electrochemical reactions of polymeric oxidation or reduction, act as a temperature sensor.

Similar experiments were performed by changing the electrolyte concentration (0.01, 0.02, 0.05, 0.1, 0.25, 0.5 and 1 M) as shown in Figure 10. For the same time of a constant current flow, decreasing electrode potentials are observed for rising electrolyte concentrations. The effect increases when referred to the same arbitrary zero initial potential. The consumed specific electrical energy changes as a function of the concentration. (Figure 10b) We conclude that the material works as a sensor of the salt concentration in the ambient.

As final conclusions, we can state that the specific energy consumed during oxidation or reduction reactions of the studied film sense the influence of the variables: flowing current, working temperature or electrolyte concentration, on the system. The consumed electrical energy is a linear function of the studied variable.

Those facts described in Sections 7.2 to 7.4 suggest that any device based on an electrochemical property (on the electrochemical reactions) of intrinsically conducting polymers is expected to act simultaneously as a sensor of any ambient and working variables acting on the reaction. Artificial muscles, polymeric batteries, smart membranes, drug delivery devices, electrochromic devices, among others, are expected to wear those simultaneous sensing-actuating properties. The two connecting wires should include both, actuating (i.e., the current) and sensing (i.e., the potential evolution) signals. We will focus our attention here on the present state of the art for those simultaneous actuating/sensing abilities of conducting polymers as reactive materials when used to construct artificial muscles.

In order to mimic natural molecular motors, we shall attempt to include, at least, electric pulses and polymeric chains. Some pioneering devices were constructed in the 1950s using films of polymeric gels immersed in aqueous solutions [101–104]. At the beginning of the 1990s, a fast development of devices based on the interaction between electric fields, or electric currents, and polymers took place stimulated by the interest to reproduce commercial piezoelectric or electrostrictive (electromechanical) devices developed with inorganic materials, but using now similar properties from polymeric materials. The beginning of this explosive interest overlapped the discovery of intrinsically conductive polymers and the reverse electrochemical variation of their volume (Figure 4). This controlled volume variation envisaged the construction of new electrochemo-mechanical devices [67,105–107].

Any developed device based on the interaction between electricity and polymers uses to be named artificial muscle. We can summarize the present state of the named artificial muscles, which actuation involves polymers, electric fields or electric currents, by their classification in two main areas:

- Electromechanical actuators: Artificial muscles responding mainly to electric fields, E (V), being the dimensions variation of the electroactive polymer proportional to:

E²: Electrostrictive actuators

Conventionally electromechanical devices have been manufactured as bending thin films of dry polymer (named electroactive material), both sides coated with a very thin metallic film required to apply the electric field. Any actuator has a triple layer structure: metal/polymer/metal (Figure 11). Linear devices from materials having a large dimension’s change with the electric field are also constructed.

Bending electrochemomechanical actuators can be manufactured as bilayers, using a metallic electrode as counterelectode to allow the current flow, or as three-layers (see below), including working and counterelectrode. In the three-layers structure, the two external films are electroactive (react electrochemically, changing volume) during current flow. The internal film is an adherent, flexible and non-conducting (or ionic conductor) polymeric film.

In the case of linear devices, they can be made of fibers of conducting polymers [108–112], or by electropolymerization of a conducting polymer on springs of helical metallic wires until the generation of a tube, or on zigzag metal wires to generate films [110,113–117]. Bundles of films or fibers are checked to produce lineal displacements of weights [118]. Origami structures form films also provide good linear movements [119]. Different models have been proposed to describe mechanical behavior or electro-chemomechanical deformations [120–122] of this kind of devices.

Artificial muscles have been developed from films of CP involving electrochemical reactions (1) or (2) as origin of the volume variations. When the interchange of anion prevails (accepted for families a, b, d, e and g) during the redox processes, the volume of the film swells during oxidation and shrinks during reduction. Under prevailing cation interchange (accepted for families c and e), the material shrinks during oxidation and swells during reduction. The important role played by the solvent interchange in both oxidized and reduced materials, is an unsolved point. Dimensional changes parallel to electrochemical reactions, have been followed by different methodologies, or estimated from experimental densities and weights of dried oxidized or reduced films from the pioneering works of the electrochemistry of conducting polymers [123,124]. Those changes were confirmed at microscopic level by “in situ” AFM [125–127], ellipsometry [128], conductivity [105] measurements, in situ electrogravimetry [129,130] and others [131,132].

We can imagine an ideal and lineal and neutral polymeric chain connected to a metallic electrode and immersed in an electrolyte (see Figure 12). The strong intramolecular interactions originate a coil compact structure of the chain (Figure 12a). Under oxidation changes on the double bonds distribution and storage of positive charges originate conformational movements until a stick-like structure (Figure 12b). This basic molecular actuator [133,134] working in a reversible way driven by electrochemical reactions includes: electric pulses, ions and solvent interchanges between the polymer and the solution, chemical reactions, stimulation of the conformational movements along polymeric chains and changes in the inter- and intramolecular interactions. Those processes occurring in soft and wet materials mimic, at molecular level, the consecutive events involved on the actuation of a natural anisotropic muscle. At the moment it is not known how to construct artificial sarcomere like structures (brushes) with chains of conducting polymers between two metallic (or electronically conducting) films. Only three-dimensional isotropic, microscopic and stable changes of volume are available with films of conducting polymers under electrochemical reactions.

Bilayer structures of conducting polymer (CP) film/adherent polymer film [67,106,107,135,136], CP/metal [137–139], CP/solid state electrolyte [140], CP/CP [141,142], CP/plastic [143], CP/paper or CP/thin film of any flexible material metal coated (i.e., by sputtering) [144] have been elegant solutions to translate the microscopic changes of volume taking place in films of CP into increasing anisotropic mechanical stress gradient across the bilayer interface, generating an uniform macroscopic bending movements (Figure 13a). The presence of a metallic counterelectrode is required to allow the current flow, originating the electrochemical reaction that produces the bending movement. However, an important fraction of the consumed electrical energy is wasted to produce the electrochemical reactions occurring at the counterelectrode (such as solvent dissociation, which requires a high overpotential). Moreover those reactions will generate new chemicals and pH variations, migrating until the muscle and promoting the progressive deterioration of the actuating film.

The construction of a triple layer CP/tape/CP [107] allows a more efficient actuation by using the same current two times to produce opposite volume changes and avoiding the use of metallic counterelectrodes (Figure 11). One of the CP films, the working electrode (WE) acts as the anode, swells (for preferential anionic interchange during the reaction) and pushes the device. The second CP film is the counterelectrode (CE) acts as the cathode shrinks and trails the device. The CE uses to be short-circuited with the reference electrode (RE) output of the potentiostat. On this way we follow the instantaneous potential difference between the two CP films: the muscle potential. The muscle potential evolution under flow of a constant current is a result of the two electrochemical reactions: oxidation at the anode and reduction at the cathode. Consequently we expect those actuating reactions to be influenced by the ambient variables, becoming a sensor of those variables.

The three layers muscles can work outside the liquid electrolytic media using an ionic conducting membrane separating the two films of CP. This membrane can be obtained by solvent evaporation and UV irradiation [72,145,146], or by formation of interpenetrated networks [147–152]: the two films of CP are generated by chemical polymerization on the external part of the membrane film.

Different structures can be produced by combination of several double or triple layers to transform bending into linear movements, each of those structures including WE, CE and reference electrode [153–155].

The actuating films of CP can be generated by electropolymerization. Electrogeneration is compatible with micro and nanotechnologies [156] giving elegant and imaginative microdevices[157] and microtools constituted by bending bilayers [69,137–139,158–165] or trilayers [166].

Volume changes per unit of time, and the produced macroscopic movement, are expected (Equation 1) to be under control of the driving current. The Faraday’s expression should quantify the amount of interchanged ions. For driving current ranging from 8–35 mA, Figure 14a shows the experimental linear relationship between the driving current and the produced angular movement rate: (9) Angular   movement   rate   ( rad   s − 1 ) = slope   ( rad   s − 1   mA − 1 )   x   applied   current   ( mA ) ω = k ⋅ i

Whatever the applied current, the experimental slope is a constant, as expected for an electrochemically driven motor. Moreover, the experimental charge consumed to perform an angular movement of one degree is constant (Figure 14c) due to the overlapping evolution of the described angle as a function of the consumed charge for different driven constant currents: (10) Described   angle   ( degrees ) = slope   ( degree   mC − 1 )   x   consumed   charge   ( mC ) α = ω ⋅ t = k ⋅ i ⋅ t = k ⋅ q

Consequently, whatever the testing current the same charge is consumed in order to perform the same angular movement (Figure 14b). Equation 10 is a translation of the Faradays laws to bending electro-chemo-mechanical artificial muscles. The charge quantifies the amount of counterions interchanged, so the composition variation, the volume variation of the CP film and the stress gradient at the interface between the polymer films.

The expressions (9) and (10) can be applied to bilayers, triple-layers or complex structures [91,92,167–169]. Figure 14 quantifies the electrochemical nature of the movement. For any specific muscle, we can define the charge required to describe an angular movement of one degree (a mC degree⁻¹). From a, the charge Q required to attain any new position (any new angle) from the actual position can be obtained: (11) Q ( mC ) = a ( mC   degree − 1 ) ⋅ angle   ( degree ) = a ⋅ α

This is an electrochemo-positioning device. Once defined the charge, Q, required to describe a defined angle, e.g., 90 degrees, the time consumed for the movement is under control of the charge flowing through the device per unit of time: the current [2,168].

The electrochemical nature of the device allows to state that different artificial muscles, having different surface area or constructed with pPy films having different film thickness (i.e., weight), must produce the same angular movement rate under variation of the same composition gradient, by flow of analogous charge per unit of time (current) and per unit of CP weight (same variation of the oxidation deep) as confirmed by Figure 15. ω / w   ( rad ⋅ s − 1 ⋅ mg − 1 ) = k ⋅ i / w   ( rad ⋅ s − 1 mA − 1 )   ( mA )   ( mg − 1 )

Allowing a more general expression than equation 9: (12) Ω = slope ⋅ i ⋅ w − 1 = slope ⋅ q ⋅ t − 1 ⋅ w − 1 = slope ⋅ jwhere Ω is the angular rate (rad·s⁻¹), i is the constant current (mA) flowing though the muscle and w is the weight of the conducting polymer films. Considering j as mC·s⁻¹·mg⁻¹ this expression states that the angular movement, Ω, of any artificial muscle constituted by conducting polymers is under control of the amount of charge flowing through the muscle per second and per weigh unit of the material. This j quantifies the variation of the oxidation rate (counterion’s composition change) per weight unit. As conclusion the angular movement for those artificial muscles is under control of the oxidation rate per unit of weight, as corresponds to any electrochemical device.

Thus, the electrochemical nature of the movement allows an excellent control of both the movement rate (by the applied current) and the angular position (by the charge). This means that we have a perfect electrical machine able to transform the electrical energy into mechanical energy. The electrochemically stimulated conformational movements of the polymeric chains and the concomitant changes of volume are used as transducers from the electrical energy to the mechanical energy.

Based on the electrochemical nature of the movement, it should be expected that any mechanical, electrical, optical, thermal, magnetic or chemical variable acting on the reaction must influence the potential evolution of the working device. This should be a specific application to artificial muscles of the above described sensing abilities of the reactive material. Rising values of those variables increasing the reaction rate (electrolyte concentration or temperature) will promote an evolution of the muscle potential while moving along the same angle at lower values of the potential for the same time of current flow (Figure 16). Those variables which increase will decrease the reaction rate (mechanical stress) will promote a shift of the potential evolution while moving at higher muscle potentials [90–92,155,167–171]. Both, the muscle potential at a defined time of current flow, and the consumed electrical energy (i∫E dt) change linearly as a function of the different variables. (see Figure 17). Figures 16d and 17d indicate how the working muscle sense the different weights of steel plates adhered to the bottom of the muscle and trailed during the angular movement.

The mechanical sensing characteristics (increase of the muscle potential when the trailed weight rises) announce the possibility to develop a tactile sensor. If an obstacle is located in the muscle’s pathway (Figure 18), before reaching the obstacle the muscle moves freely, under constant current: the evolution of the muscle potential overlaps that of the free muscle. When the muscle touches the obstacle, the mechanical resistance influences the anodic and cathodic electrochemical reactions occurring in the two constituent, anode and cathode, polymeric films. Consequently a step is observed on the muscle potential at the contact time, the potential step is proportional to the opposed mechanical resistance. Rising weights of the obstacle produce increasing potential steps (Figure 19). Both, potential step and consumed electrical energy after contact and during shifting times, follow a linear evolution as a function of the obstacle weight [172–174]. When the muscle is not able to shift the obstacle the potential steps by several volts. During actuation the muscle potential of the free muscle evolves from a few mV to hundred of mV. The muscle potential gradient at the contact time with the obstacle ranges from a few mV to several V. Actuating and sensing signals are of the same order of magnitude.

These results announce the development of a class of tools and robots where the electrical machines (actuators) are at the same time, and through the same two connecting wires, sensors of the working conditions and of the ambient variables. In contrast to conventional technologies, where actuators and sensors are independent tools, requiring different electrical connections, complexes interfaces with computers and complexes software to develop intelligent machines.