Electro-Chemical Actuation of Nanoporous Metal Materials Induced by Surface Stress

Abstract

Similar to biological muscles, the actuator materials can function as artificial muscles by directly converting an external stimulus in the form of electrical or chemical energy into a mechanical response through the reversible changes in material dimensions. As a new type of high surface-area actuator materials, nanoporous metals represent a novel class of smart electrodes that undergo reversible dimensional changes when applying an electronic voltage on the surface. The dimensional changes in nanoporous metal/polymer composite still originate from the surface stress of nanoporous metal. Additionally, this surface stress can be modulated by the co-adsorbed sulfate counter-ions that are present in the doped polymer chains coating matrix upon the application of an external potential. Nanoporous metals fabricated by dealloying have received extensive attention in many areas, such as catalysis/electrocatalysis, energy conversion/storage, and sensing/biosensing. In this review, we focus on the recent developments of dealloyed nanoporous metals in the application of actuation. In particular, we summarize the experimental strategies in the studies and highlight the recent advances in the actuator materials. Finally, we conclude with outlook and perspectives with respect to future research on dealloyed nanoporous metals in applications of actuation in electrochemical or chemical environment.

1. Introduction

In nature, the actuation behaviors in biological systems are powered exclusively by chemical energy. Due to the similarity to biological muscles, the manmade actuators are often called ‘artificial muscles’, which can directly convert an external stimulus in the form of electrical or chemical energy into a mechanical response through reversible changes in material dimensions. The traditional well-known materials of actuators are piezoelectric ceramics [1,2]. However, their applications are strongly restricted by the requirement of high voltages as well as the limitation of low work density. These restrictions need to be removed in order to obtain the highly reliable actuation performance. Conductive polymers and carbon nanotubes have been recently reported to provide better properties for the muscle-like actuation [3,4]. However, conductive polymers suffer from the drawbacks of low stiffness and strength; and more importantly, the low Faradaic process is often associated with the dopant diffusion and, finally, changes the structure of conductive polymers [3,4]. Carbon nanotubes or graphene films cannot sustain considerable load in compression, which limits their practical applications.

In contrast to conventional actuator materials, the metallic muscles have much better conductivity and mechanical properties, e.g., stiffness, and strength. In metallic muscles, not only the bulk volume but also the interface surface area plays critical roles for actuating. By employing dealloyed nanoporous metals with high surface-area-to-volume ratios, the new actuators represent a novel class of smart materials undergoing reversible dimensional changes upon the injection of an electronic charge in the space-charge region at the nanoporous metal–electrolyte interface [5,6,7]. When using nanoporous metals as actuator electrodes, the reversible actuation can be induced by a low applied voltage, normally on the order of ~1 V, and the induced amplitude is comparable to those of commercial piezo-ceramics (~0.2%). Nanoporous metal materials in bulk form can also undertake considerable reversible compressive loading, which is considered a pre-requisite for actuator applications.

A number of new actuators based on nanoporous metals has been recently prepared and reported as desirable artificial metallic muscles for the fast response and high performance of electrochemical actuation in the past decade [8]. Thus, an updated review about the recent development in the actuation application of nanoporous metals is quite necessary. In this account, we describe recent advances on the experimental strategies in the fundamental studies of actuation. Based on recent experimental observations, we also provide a critical discussion and summary on the actuation mechanism of nanoporous metals in an electrochemical or a chemical environment.

2. The Actuation Mechanism of Nanoporous Metals

With attention on the microscopic behavior of actuation, conventional actuator materials, such as piezo-ceramics, use an electric field as an external stimulus. The electric field, which is applied between two opposing surfaces of material, penetrates the material and prompts a distortion of each crystallographic unit cell [9,10,11,12]. It then gives rise to the macroscopic strain of material. Since the metal interior is free of an electric field, they cannot exhibit the piezoelectric behavior. It is well established that the macroscopic strain in the actuation of a nanoporous metal/electrolyte system is induced by the changes in the surface stress of nanoporous metal when electronic charges are injected in the space-charge region at the nanoporous metal–electrolyte interface [13,14,15].

The transfer of the charge to the surface, caused by applying electrode potential, leads to a change in the bonding strength between the surface atoms [16,17,18]. In order to preserve the mechanical equilibrium, these modified forces at the surface, as a consequence of the change in the bonding strength, need to be compensated by opposite changes in the surface stress state in the ligaments of the bulk (Figure 1), and the overall strain required to generate this stress is observed as a macroscopic expansion or contraction in the nanoporous metal. Detailed mathematic derivations based on the continuum description can be obtained in the previous reports of [18].

The trend of dimensional changes of materials in experiments agrees well with the trend of surface stress changes as established for metal surfaces. The central aspect of surface stress is sensitive to the surface state at a metal–electrolyte interface and can be strongly influenced in an electrochemical environment by applying an electrochemical potential [19,20,21]. Moreover, the variation of surface stress f with charge density q is quantified by an electrocapillary coupling parameter, the stress–charge coefficient, ς = df/dq [22,23]. This parameter could be related to the measurable quantities, the macroscopic strain of length ∆l/l₀, the transferred charge ∆Q, and the sample mass m via

$$\mathsf{\zeta }=-\frac{9Km}{2\rho }\frac{\Delta \epsilon }{\Delta Q}$$

(1)

where K and ρ denote the bulk modulus of the metal and the density of the solid phase, respectively. It is seen that coefficient ς is the central role controlling the actuation property of a certain metal material when the constant charge is applied on the surface. The sign and value of ς significantly influence the direction of actuation (expansion or contraction) and the amplitude of actuation. In the following, the value of ς in the published reports are discussed.

We first focus on the electrochemical capacitive process. Nanoporous metal samples undergo expansion when the surface is charged positively for a clean transition and a noble metal surface, confirming a negative-valued surface stress-charge coefficient of ς, which is testified by experiments [22]. The nanoporous material with an oxide-covered surface exhibited an opposite sign of response compared to the one with a clean surface [24]. More specifically, this surface oxide can be formed by anodic oxidation of Au as well as Pt electrodes during the dealloying process. When investigating the surface-stress-induced strain of the samples with an oxide-covered surface, the nanoporous materials tend to contract when the potential is scanned positively, indicating an abnormal positive ς of the oxidized surface.

In comparison with most studies of the electrocapillary coupling that emphasized in the capacitive processes, the actuation behavior and the surface stress–charge response were also studied while the surface adsorption/desorption was involved. The nanoporous metals with clean surfaces tend to expand during adsorption of oxygen species, such as OH⁻ (ς < 0) and hydrogen (H⁺, ς > 0) on top of the bulk-terminated surface [23]. The macroscopic strain amplitude, and thereby the changes in f, are significantly larger during the oxygen and hydrogen adsorption/desorption process than in the capacitive-charging double-layer region. When re-structuring of the gold surface, which is known as the replacement turnover process, is allowed to occur simultaneously with the adsorption of oxygen, in situ cantilever-bending studies show the surface stress to increase when oxygen species adsorb on a (111)-textured gold surface in aqueous H₂SO₄ [25]. This implies a positive sign of ς for the replacement turnover process in contract to the negative-valued ς for adsorption on top of the bulk-terminated surface. These conflicting findings indicate that different electrosorption processes, specifically the oxygen species adsorbed on the bulk-terminated surface, exhibit fundamentally different coupling between the chemistry and mechanics of the surface.

To experimentally identify the value of ς, a Maxwell relation equates ς for any given electrode process to the respective variation of the electrode potential E, with tangential strain e. Thus, one has ς = df/dq|_e = dE/de|_q [26,27]. This implies that electrocapillary coupling can be equivalently measured in two seemingly independent ways, namely from the variation of the surface stress with the charge density at a constant strain or from the variation of the electrode potential with the strain at a constant charge density. The left side of equation measures the efficiency of surfaces in enabling nanoporous metal actuation. The right side has been studied and catches a lot of interests in the fields of catalysis due to the coupling of surface mechanics and reactivity [25,28,29,30,31,32]. A recent review on electrocapillary coupling can be found in ref [27].

For the actuation of an electrolyte-free system, the dimensional changes in the nanoporous metal/polymer composite still originate from the surface stress of the nanoporous metal [33]. The surface stress in this system is modulated by the co-adsorbed sulfate counterions that are presented in the doped polymer chains coating matrix upon application of an external potential [9,19,34,35].

3. Preparation Methods of Nanoporous Metals

Due to the unique physical and chemical properties, nanoporous metals have received tremendous attention in many areas, such as catalysis [36,37] and energy conversion/storage [38,39]. The structure of nanoporous metals and some of their interesting properties and applications have drawn much attention. The typical preparation process of nanoporous metal is to dealloy the corresponding alloy precursors in an electrochemical or a chemical environment. The typical microstructure of nanoporous gold (Figure 2) prepared by the dealloying process demonstrates that both the pore and ligament structures are on the nanometer scale.

Dealloying is a method of preparing nanoporous metals by using certain physical or chemical means to remove some of the constituent elements from an alloy. During the removal process, atoms of the remaining alloy constituents spontaneously form a nanostructured porous structure through migration and diffusion. As the nanoporous structure constantly evolves during the dealloying process, it is possible to obtain nanoporous structures with different characteristic sizes by controlling the dealloying process [41,42].

Currently, the dealloying technique has been developed maturely, and various teams are researching on the manufacturing of new nanoporous metals based on the principle of dealloying. Chemical dealloying is one of the oldest and most widely used dealloying methods that utilizes the potential difference between elements to dissolve one or more components in an alloy, thus forming a three-dimensional interconnected nanoporous structure within the alloy. The precursors for chemical dealloying mainly consist of crystalline metals and amorphous materials. By using dealloying, nanoporous structures have been successfully obtained on different types of metals, such as Au, Pt, Pd, Cu, Ni, and Ag, using crystal alloys as precursors, such as Au-Ag, Pt-Cu, Pd-Co, Cu-Mn, Ni-Mn, and Ag-Zn. In addition, due to advantages, such as homogeneous structure, small element limitations, and wide range of compositions, amorphous materials are more suitable as precursors for nanoporous metals than crystalline materials [43,44,45,46,47]. More and more scholars use amorphous materials as precursors to prepare nanoporous metals; however, there are still limitations when preparing nanoporous metals with amorphous materials as precursors due to factors, such as size. The morphology of the nanoporous metal prepared by chemical dealloying varies greatly and is mainly affected by factors, such as dealloying time, temperature, type, and concentration of the corroding solution [35,48].

In the past, the mechanical property of nanoporous metal was generally brittle. However, some novel strategies have been recently developed to synthesize flexible nanoporous metals. For instance, Qin et al. report a strategy to synthesize a flexible nickel oxide/hydroxide-coated nanoporous nickel (np-NiOxHy@Ni) electrode containing a metallic glass (MG) interlayer (np-NiOxHy@Ni/MG/np-NiOxHy@Ni sandwich) by one-step dealloying of Ni40Zr20Ti40 MG ribbons. Benefiting from the ductile MG interlayer, the sandwich-like electrode presents excellent flexibility [49].

Currently available dealloying methods can only be applied to limited metal elements and usually require the use of chemical reagents for dissolution, which may cause serious environmental problems. Therefore, they lack economic viability and sustainability. Vapor-phase dealloying is a green and universal method that utilizes the vapor pressure difference between the constituent elements in an alloy to selectively remove components by applying a certain fractionation vapor pressure, thus forming a nanoporous structure. Vapor-phase dealloying is widely applicable and not limited by the types or chemical activity of material elements, allowing for the preparation of nanoporous structures for various materials. The evaporated component from the dealloying process can also be reused. Additionally, the pore size of the nanoporous structure prepared by vapor-phase dealloying is adjustable [50,51,52].

During the vapor-phase dealloying process, the sublimated alloy components can be completely recovered and reused, which is more environmentally friendly than traditional dealloying methods and meets the demands of modern technological development. Dealloying has been proved to be an effective method for preparing nanoporous metals, and it has also extended to diverse dealloying techniques [53,54].

Compared to the previously mentioned dealloying methods, template synthesis is also an important method for preparing nanoporous metals. The specific process involves filling precursors into natural or artificial templates according to requirements. After the formation of a porous structure, the template is removed to obtain an independent porous structure. Commonly used templates can generally be divided into three types: hard templates, soft templates, and composite templates [55].

Hard templates contain many natural templates, such as sea urchin spines and zeolites. However, nanoporous metals made from natural templates typically have micrometer-scale pore sizes and limited choices, often unable to meet the requirements. Therefore, more and more researchers choose to use artificial templates, such as anodic aluminum oxide (AAO), colloidal silica, polystyrene colloidal crystals, etc.

An anodic aluminum oxide (AAO) template is a typical nanostructured porous array with fine structure. Its preparation method involves anodizing aluminum in acidic electrolytes to form a honeycomb-like porous structure, namely the AAO template. AAO is mainly used for preparing materials, such as nanodot arrays and nanowire arrays, and has two types of pore channels: single-channel and dual-channel AAO. Single-channel AAO refers to a pore channel with one end blocked by an aluminum base, while the pores on dual-channel AAO templates are through-hole, making them suitable for use as filtration membranes [56].

The hydrogen bubble template method is a technique that utilizes the large amount of hydrogen bubbles generated during electrochemical deposition at the cathode as a dynamic template for fabricating nanoporous structures. During the electrodeposition process, the volume of hydrogen bubbles increases and continually escapes, resulting in the formation of nanoporous metal structures. This method takes advantage of the pathways formed by the hydrogen bubbles produced by the cathodic reaction between the substrate and electrolyte, as well as between the electrolyte and air, which prevents metal deposition from occurring at the locations where the bubbles form. Only spaces between adjacent bubbles in the electrolyte can facilitate metal deposition and enable the creation of a dynamic template for producing porous metals. Currently, the hydrogen bubble template method has been successfully applied to fabricate nanoporous Cu, Sn, Ni, and other metals. However, due to limitations imposed by equilibrium potential and overpotential, this method is not suitable for directly preparing nanoporous noble metals with low electronegativity [57,58].

Thus, the main advantage of template methods is that the resulting nanoporous metals have a wide range of pore size distributions and can be more precisely controlled in terms of pore size, which gives them an edge over other fabrication methods. However, the high cost of template fabrication and the inability to adjust pore structure and size once a specific template has been chosen are major drawbacks. To overcome these limitations, template methods can be combined with other fabrication techniques [59].

Chemical dealloying is an effective method for fabricating nanoporous metals, and the resulting nanoporous structures can be flexibly regulated by adjusting factors, such as time and temperature. However, it is difficult to obtain uniform and impurity-free nanoporous metals using this method. Template methods suffer from drawbacks, such as incomplete template removal, low cost-effectiveness, and complex procedures. Therefore, spontaneous formation of size-controllable nanoporous structures directly on a pure metal substrate is an ideal fabrication process. To achieve this, researchers have attempted to use electrochemical methods to fabricate nanoporous metals. Specifically, the process involves utilizing oxidation reactions at the electrode and reactions between the electrode and electrolyte under a positive electric field to obtain nanoporous structures on a metal substrate serving as the electrode material [60,61].

4. Experimental Strategies

4.1. Contact Method: In Situ Dilatometry Measurement

The bulk form of nanoporous metals responds to the applied voltages through the reversible macroscopic expansion or contraction. This kind of actuation behavior is conveniently characterized by means of in situ dilatometry in the electrolyte environment [62,63].

The dilatometer equipped with a displacement sensor can accurately measure the dimensional changes of solid materials, which are induced by a physical or chemical process. Due to its high sensitivity, the dilatometer is traditionally used to measure the volume changes in vacuum or an inert gas environment during heating or cooling. Equipped with an electrochemical cell and a potentiostat, the conventional dilatometer can be improved to investigate the issue of interest about the actuation performance in electrolyte environment. Figure 3 shows the schematic illustration of an experimental setup. This experimental strategy is perfectly suitable for the bulk materials of nanoporous metal, and it has been applied in several studies of electrochemical actuation for np-Au, np-Pt, np-AuPt, and np-Pd [62,63,64].

The basic actuation responses of contraction and expansion can be combined in one component, e.g., the wafer bending, which contracts one side of the material while expanding the other side. However, the strategy of in situ dilatometry is not suitable to investigate the bending behaviors when using a nanoporous metal thin film as a wafer cantilever.

4.2. Contact Method: Mechanical Strain Gauge

In traditional mechanics, the strain gauge is a simple device to measure the surface strain on an object of interest and to investigate the wafer bending behaviors. The most common type of strain gauge consists of a metallic foil pattern supported by an insulating flexible substrate. The strain gauge is normally attached to the object surface by a suitable adhesive. The working principle is taking the advantage of the physical property of electrical conductance, which is strongly dependent on the conductor’s geometry.

Although it is quite simple to identify the surface strain, the value identified by the strain gauge is strongly dependent on the measurement temperature and the adhesion on an electrode surface.

4.3. Non-Contact Method: Wafer Bending Monitored by Laser Deflection

For investigating a reversible bending strain of an actuator in an electrolyte environment, an alternative strategy is to use the noncontact method of laser deflection. The schematic representing this method is shown in Figure 4.

The cantilever bending behavior of a thin film is controlled by surface stress that is influenced by applied voltage in an electrolyte environment [28,66,67]. Changes in the cantilever curvature, in response to the changes in surface stress, ∆f, are determined in situ by means of a commercial laser-based device. More specifically, any slight movement of the foil’s free end when applying the voltage can be directly observed and recorded with a charge-coupled device (CCD) camera.

5. Actuation Properties of Nanoporous Metals

5.1. Electrochemical Actuation with Noble Metals

In 2003, Weissmüller et al. reported the first experimental demonstration of electrochemical actuation by using high surface-area nanoporous metal. The nanoporous platinum (np-Pt) samples in their investigation were prepared by consolidating commercial Pt black, which has the grain size of 6 nm [68]. Weissmüller et al., varied the surface stress of np-Pt in an electrolyte environment by applying an electrical voltage, also called an electrode potential in electrochemistry, to inject ions from an electrolyte to the metal–electrolyte interface during cyclic voltammetry. With this high surface-area electrode, the reversible amplitude of electrochemical actuation by means of in situ dilatometry is indeed comparable to those of commercial piezo-ceramics. The amplitude was reported on the order of 0.15% at the voltage of 1 V [68], and it was noted that the typical piezo-actuator delivers a maximum strain of 0.2% at a much higher voltage of 150 V [69]. Figure 5 demonstrates that reversible dimensional change varies reversibly as a function of an electrode potential and shows a linear relation of dimensional change and surface charge. It thus provides a new approach of surface-charge-induced strain in application of electrochemical actuations.

By combining solid gold and nanoporous gold (np-Au) into a bilayer foil (illustrated in Figure 6), Kramer et al., reported that the surface-charge-induced strain could be greatly amplified, leading to macroscopic tip displacements as large as several millimeters [70]. The composite foils were immersed in aqueous electrolytes and the electrode potential varied. In this way, the electrochemical-induced actuation of a metal solid could be observed by the naked eye in their experiment. [70]. The surface-charge-induced expansion or contraction of nanoporous metal gave rise to a biaxial stress component resulting in a large bending of the foil, and the stress was amplified by the high surface area of np-Au.

Large strain amplitude was also observed in binary nanoporous Au_0.8Pt_0.2 (np-AuPt) dealloyed from AuPtAg alloy precursors [71]. The pore and ligament size of np-AuPt were controlled down to a few nanometers, resulting in a much larger specific surface area. With this actuator material, in situ dilatometry experiments showed that electrochemical actuation with the largest linear reversible strain amplitude can reach 1.3% during the oxygen species electrosorption on the surface driven by varied electrochemical potential between −0.1 and 1.5 V. The strain amplitude was so large that the associated stress might have approached the elastic limit of the alloy. The combination of the large strain amplitude and the high stiffness of this material made the material more effective in converting electricity energy into mechanical work, marked by a high-strain energy density of 6.0 MJ/m³. The value is competitive with some polymer actuators possessing strain energy density up to 3.4 MJ/m³ [72].

More recently, Jin et al. reported an interesting strategy to increase the high linear performance of an np-Au electrochemical actuator, by similarly incorporating the gold solid layer and np-Au parallel layer to form Au/np-Au multilayer composites (Figure 7) [63]. With this multilayer-structured gold/np-Au composites, they observed that the actuation strokes in the direction perpendicular to layer planes were larger than the rule-of-mixture predictions and could be 20% greater than that of monolithic np-Au. The multilayer composite was meanwhile stiffer than the monolithic np-Au. Their study thus provided an approach to simultaneously increase the actuation strain amplitude, the effective stiffness, and mechanical stability of actuation material.

Hosson et al., synthesized functional np-Au with a dual microscopic length scale by exploiting the crystal structure of the alloy precursor (Figure 8) [73]. They observed an enhanced surface-charge-induced strain response of np-Au from the novel multilayer morphology. During the electrochemical oxidation and reduction in the electrode surface, the surface charge transferred via anion electrosorption gave rise to a change in the surface stress of np-Au and subsequently to the dimensional changes of multilayer np-Au. By performing it in this way, the strain amplitude in the direction perpendicular to the layers achieved varies between 3% and 6%, which was almost 2 orders of magnitude in comparison to the conventional np-Au. They explained that this high performance originated from the misfit strain between adjacent np-Au scales.

The work by Zhang et al. showed that the coupling of hydrogen adsorption and absorption could trigger a giant reversible strain in bulk nanoporous Pd (np-Pd) in a weakly adsorbed NaF electrolyte [74]. The bulk np-Pd was fabricated with a hierarchically porous structure and a ligament size of ~10 nm. The actuator made by np-Pd exhibited a giant reversible strain of up to 3.74% and a high volume-specific strain energy density. They explained that the performance of np-Pd was attributed to the coupling of hydrogen adsorption/absorption and its unique hierarchically nanoporous structure.

5.2. Electrochemical Actuation of Non-Noble Metals

The materials of nanoporous metal in the aforementioned studies for electrochemical actuation are focused on the high-cost noble metals. The low-cost actuator materials are desired for actual applications.

Hakamada et al., reported the fabrication of a bilayer-stacked nickel nanostructure as the electrochemical actuator, and it was the first report of a nanoporous structure of non-noble metal exhibiting the electrochemical actuation characteristics [75]. The stacked bilayer sheet was composed of a bulk form and a nanoporous structure of nickel metal. The bilayer actuator was deformed with the amplitude of 0.2 × 10⁻⁴ when an electrode potential of ±1 V was applied in aqueous NaOH solution. Their results suggested that np-Ni could be used as an actuator for commercial use due to its low price and high availability. However, it was shown that the amplitude of a charge-induced strain is two orders less than those of the aforementioned np-Au.

Cheng et al. measured the in situ surface stress caused by the intrinsic charge of the nickel film during electrochemical reaction in an alkaline electrolyte. As shown in Figure 9, surface stresses induced by H absorption/desorption, α-Ni(OH)₂ formation, capacitive double-layer charging, the α- to β-Ni(OH)₂ transformation, and β-Ni(OH)₂/β-NiOOH redox reactions were identified, and each provided additive contributions to the overall stress state [13]. The current peaks for nanoporous Ni were comparatively broad due to a relatively sluggish ion transport within the tortuous nanoporous structure (Figure 9b) compared with the film Ni; accordingly, the surface stress in nanoporous Ni had a larger positive/negative scan hysteresis (Figure 9c). The magnitude of the charge-induced surface stress in nanoporous Ni (66.6 N m⁻¹) was 48% larger than that of film Ni (44.9 N m⁻¹), indicating an amplification of surface stress. The surface stress–voltage coupling coefficient was defined as η = df/dE, which is plotted in Figure 9d to show a pair of strong η peaks corresponding to the redox reactions for nanoporous Ni. η was larger for the nanoporous Ni because its highly curved, torturous geometry restricted local strain relaxation whereas strain was more readily relaxed in the smooth Ni film.

The nanoporous metallic actuators in millimeter scale suffered from slow strain response time, which was on the order of hundreds of seconds. One of the reasons was that the disordered nanoporous structure resulted in slow ion transportation. By using the nanohoneycomb Ni as an actuator, Cheng et al., reported the fast strain response (order of 0.1 s) with sustainable actuation performance (no noticeable degradation after 800 cycles) and with a strain energy compatible with skeletal muscles.

In order to further improve the actuation performance of nanoporous nickel based on a modified dealloying process, Zhang et al., recently reported a new nickel-based actuator, which exhibits an unprecedented performance, including reversible strain up to 2% at the operating voltage of 0.8 V [76]. In their study, the applied potential range was limited within a typical pseudo-capacitive process, which was related to the reversible OH adsorption/desorption. Besides the high strain amplitude, the ultrahigh work density (11.76 MJ/cm³) and long cycle life (70% strain retention after 10,000 cycles) were also reported [76]. This outstanding performance was explained originating from its unique hierarchically nanoporous structure and the oxide-covered nature of the Ni surface during the electrochemical treatment in alkaline media.

Besides low-cost np-Ni, a nanoporous structure of silver metal, another relatively cheap metal in comparison to the gold and platinum, has also been studied by Hosson et al. as an electrochemical actuator [77]. They reported on reversible dimensional changes in np-Ag prepared from the selective dissolution of Al from Ag₂₀Al₈₀ alloys. The maximum strain amplitude in their study could reach 0.5% in alkaline media at an applied potential of 0.5 V.

5.3. Chemical Actuation

Biener et al., demonstrated another type of actuation behavior of np-Au, which also originates from the surface stress changes [78]. Instead of changing the surface charge in an electrochemical environment, the surface stress in their investigation was changed by surface chemistry at the solid-gas interface. In situ dilatometry experiments detected the significant reversible length changes while np-Au was exposed to ozone (O₃) and carbon monoxide (CO) alternately. The procedure was switched back and forth between two distinct surface states: the oxygen-covered surface after exposing to O₃, and the clean surface when the adsorbed oxygen was removed during CO exposure. The difference of surface stress between two surface states gave rise to the reversible sample length changes (Figure 10). In this way, they achieved reversible strain by alternating exposure of np-Au to O₃ and CO. Elastic strain amplitude up to 0.5% could be obtained for long exposures. The effect was explained by adsorbate-induced changes in the surface stress. Their work nicely demonstrated a concept of metallic muscles, which could convert the chemical energy directly into mechanical response [78].

5.4. Physical Adsorption-Induced Actuation

In contrast to the electrochemical/chemical adsorption-induced strain where changes in the np-Au surface stress were driven by electrochemically or chemically adsorbed atoms onto the metal surface, Hosson et al. reported on dimensional changes in np-Au during the changes in surface stress, which were generated by weak intermolecular forces, e.g., Van der Waals forces or induced dipole forces [79]. These forces could arise during physical adsorption/desorption of polar water vapor molecules on or from the np-Au surface. In their work, no electrical energy was applied to the system. Instead of making use of ions in electrolyte to adsorb on the ligament surfaces of np-Au, they exploited polar water vapor molecules in ambient air, where the water vapor content was adjusted. They observed that the reversible strain amplitude up to 0.02% was measured in response to a 15% change in the relative humidity for long exposure times. Since the whole process avoided the chemical reaction in their work, their concept might be attractive for the application of environmentally friendly actuators.

Ye et al. reported the nanoporous gold in the application of water-responsive actuators [66]. Indeed, the actuation stroke (∼0.52% for np Au(Pt)) is the maximum stress generated by this actuator, which can be used to perform relatively large mechanical work (D_s = 19 MPa). The stored strain energy density (w_V = 3/2·D_s D_l/l₀, for volume strain) was then determined to be 0.15 MJ m⁻³.

5.5. Electrolyte-Free Actuation

An aqueous electrolyte in the study of actuation may prohibit metallic muscles for the operation in dry environments and hamper a high actuation rate due to the low ionic conductivity of the electrolyte [19,80]. To overcome these drawbacks caused by the electrolyte, Hosson et al., have conducted beautiful works on nanoporous metal–polymer composite materials as electrolyte-free actuators [80]. For instance, a thin layer of polyaniline (PANI) doped with sulfuric acid was grown onto the ligaments of np-Au. Dopant sulfate anions co-adsorbed in the polymer coating matrix were exploited to tune the electric charge density at np-Au surface and subsequently generate the macroscopic strain in np-Au. They observed that the strain amplitude at the potential sweep rate of 10 mV/s is on the order of ~0.15% during the successive forward–reverse voltage cycles between 0 and 2 V [19].

6. Conclusions and Outlook

We have summarized various dealloyed nanoporous metal materials for the application of actuation. The change in surface stress at the interface, either a metal–electrolyte, a metal-gas, or a metal-polymer interface, is induced by the adsorption process on the metal surfaces. The surface stress then gives rise to the macroscopic strain in the form of dimensional changes. The comprehension of the actuation origin would provide some guidelines for designing new actuators with enhanced performance as artificial metallic muscles.

To date, research on the actuation of nanoporous metals is mainly focused on the noble metals (such as Au, Pt, and Pd), which is mainly due to the high stability of these metals during dealloying. However, their high cost may pose economical concerns. Therefore, research on reducing the usage and enhancing the performance of noble metals by increasing the specific surface area and fabricating the non-noble metals are every important in a practical application. In order to enhance the macroscopic strain of nanoporous metals, effective approaches include choosing materials with a high value of coupling coefficient ς and applying a large surface charge on the metal surface by either increasing the specific surface area or increasing the capacity of the material.

Besides the large amplitude of actuation, one of the significant features as artificial muscle is the fast response. For actuator applications, the response time is of critical importance. One of the challenges for metallic actuators in the future is how to improve the response time while maintaining the stability of actuation performances. In addition to the natural artificial muscles, considering carbon nanotube composites has been investigated for the application of rotational actuations, the fabrication of flexible nanoporous metal, and carbon nanotube composites may be a good choice for specific applications, such as rotational actuations. Many significant issues are still open for future research in the field of nanoporous metal formation and mechanical behavior.