Table of contents

This special issue on shape memory alloys (SMA) is an encore to a special issue on the same topic edited by us six years ago (Smart Mater. Struct.9 (5) October 2000). A total of 19 papers is offered in this issue, organized into the three broad categories of modeling, characterization and applications. In addition to thermally activated shape memory alloys, this issue includes papers on magnetic shape memory alloys. This is reflective of the trends in the SMA research community where magnetic SMAs have gained prominence in recent years. We would also like to highlight that this issue, like the previous one, is novel in that it presents contributions from experts drawn from the mechanics as well as the materials community. This is our modest effort to provide a common forum to both communities. In conclusion, we thank the many reviewers who made this issue a reality.

In this paper, we propose a multiscale model of magnetic shape memory alloys, directly coupling the atomistic detail to the macroregion modeled using continuum concepts. A formulation of the Helmholtz free energy potential based on the one-dimensional Ising model has been derived. The thermodynamic potential developed has been used in the context of the sharp phase front based continuum model of phase transformations suggested by Stoilov and Bhattacharyya (2002 Acta Mater. 50 4939–52). The model has allowed obtaining the analytical formulation of the strain and the overall deformation as functions of the magnetic anisotropy of the martensitic phase, mechanical and magnetic energies. The values of the blocking stress obtained are in good agreement with the experimentally measured ones.

A quasi-static model for NiMnGa magnetic shape memory alloy (MSMA) is formulated in parallel to the Brinson and Tanaka thermal SMA constitutive models. Since the shape memory effect (SME) and pseudoelasticity exist in both NiTi and NiMnGa, constitutive models for SMAs can serve as a basis for MSMA behavioral modeling. The simplified, linear, quasi-static model for NiMnGa was characterized by nine material parameters identified by conducting a series of uniaxial compression tests in a constant field environment. These model parameters include free strain, Young's modulus, fundamental critical stresses, fundamental threshold fields, and stress-influence coefficients. The Young's moduli of the material in both its field and stress preferred configurations were determined to be 450 MPa and 820 MPa respectively, while the free strain was measured to be 5.8%. These test data were used to assemble a critical stress profile that is useful for determining model parameters and for understanding the dependence of critical stresses on magnetic fields. Once implemented, the analytical model shows good correlation with test data for all modes of NiMnGa quasi-static behavior, capturing both the magnetic shape memory effect and pseudoelasticity. Furthermore, the model is also capable of predicting partial pseudoelasticity, minor hysteretic loops and stress–strain behaviors. To correct for the effects of magnetic saturation, a series of stress influence functions were developed from the critical stress profile. Although requiring further refinement, the model's results are encouraging, indicating that the model is a useful analytical tool for predicting NiMnGa actuator behavior.

The temperature–stress dependence of thermoelastic martensitic transformations in shape memory alloys is a critical design tool for applications of these alloys in smart structures. This dependence has been most fundamentally expressed in a Clausius–Clapeyron linear relationship. This paper presents a theoretical analysis of this relationship, with particular attention given to the effect of the strain parameter in the equation for the case of equiatomic NiTi alloys. The theory is refined in two different forms describing two distinctive conditions corresponding to complete and incomplete stress-induced martensitic transformations. The conclusions of the analysis are used to clarify the mechanisms of post-plateau deformation of NiTi.

The main objective of this study is to introduce a 'dissipationless band' to model inner hysteresis loops of response of shape memory alloys (SMAs). Dissipation that occurs when the material undergoes phase transformation is critical to the modeling of hysteretic behavior. Emphasis is placed on modeling such dissipation in the proposed methodology. Using a dissipationless virtual response of the material, a logical framework for the onset transformation under reversal of cycles is presented. Characteristics of the material transformation with reference to a dissipationless band model the true inner hysteresis loops. It is identified that this dissipationless band occurs due to the difference between the starting states of forward and reverse transformations. The construction of the generalized driving force for the transformation along with the rate of dissipation function is formulated. A numerical example is presented to highlight the qualitative prediction capabilities of the model. The example involves simulating hysteresis loops for different kinds of partial and complete loading cycles in the pseudoelastic state of the material. The predictions show that the proposed one-dimensional model is capable of representing the actual hysteresis behavior of the stabilized shape memory alloys, by effectively incorporating the dissipation effects due to the loading history.

Experimental investigations on superelastic shape-memory alloys (SMAs) show a dependence of the stress–strain relationship on the loading–unloading rate. This feature is of particular importance when utilizing SMA materials for seismic applications, since the loading rate may affect the structural response.

Motivated by this observation and by the fact that there exist relatively few studies on the material modelling of SMAs in earthquake engineering, the present work addresses a uniaxial constitutive equation able to describe the rate-dependent behaviour of superelastic SMAs.

The formulation of the model is based on two scalar internal variables, the static martensite fraction and the dynamic martensite fraction, for which three different types of evolutionary equations in rate form are proposed. Moreover, the model takes into account the different elastic properties between austenite and martensite.

Finally, after discussing two possible approaches for the solution of the corresponding time-discrete framework, the ability of the model to simulate experimental data obtained from uniaxial tests performed on SMA wires and bars at frequency levels of excitation typical of earthquake engineering is assessed.

This paper presents a constitutive model for shape-memory alloys that builds on ideas generated from recent micromechanical studies of the underlying microstructure. The presentation here is in one dimension. It is applicable in a wide temperature range that covers both the shape-memory effect and superelasticity, is valid for a wide range of strain rates and incorporates plasticity. The thermodynamic setting of the model is explained and the model is demonstrated through examples.

It has been demonstrated that a pseudoelastic TiNi matrix composite exhibits high resistance to wear, which benefits from the combination of a reinforcing phase and a flexible and wear-resistant matrix. The former mainly withstands the wearing force and the latter acts as a binder to retain the reinforcing phase, accommodate deformation and absorb impact energy. Due to its pseudoelasticity and high wear resistance, the TiNi matrix is markedly superior to many conventional matrix materials. The performance of the TiNi matrix can be further improved by embedding a small amount of hard nanoparticles in it. The nanoparticles effectively strengthen the matrix but degrade its pseudoelasticity. There exists an optimal volume fraction of nanoparticles, above which the effect of nanoparticles becomes negative. In order to better understand the role of the nanoparticles in improving the wear resistance of the composite and to maximize the benefit from the nanoparticles, a computational study was conducted using a dynamic simulation technique. It was demonstrated that a small amount of nanoparticles could effectively strengthen the matrix without much pseudoelasticity being lost. However, too many nanoparticles negatively influence the pseudoelasticity and the interfacial bond strength, leading to a decrease in the wear resistance.

In this paper, a finite element procedure is presented to model a SMA-based damage control system that is applied to monitor and reduce the degradation in the bonding material of bonded composite patches used to maintain aerospace structure frames. A parametric study was performed to investigate the effect of various design parameters on the performance of the damage control system. Also, increasing the number of activated wires was found to reduce the amount of phase transformation and forces required per wire, which can increase the system sensitivity to small cracks. It is suggested that the difference between the initial temperature of the system and the austenite start temperature is minimized to achieve faster damage recovery. Furthermore, it was found that, by increasing the SMA wire diameter at constant power input, heat loss by conduction at the surface of the wires becomes the dominant effect on the recovery process; therefore, more time is needed to recover the crack as the wires' diameters are increased. Finally, the modulus of elasticity of the matrix material for the bonding is found to have a major effect on the damage recovery process.

This work (Part I) discusses the numerical analysis of a fuel-powered shape memory alloy actuator system (FPSMAAS) that utilizes fuels with high energy densities, such as propane, as its energy source and thus reduces or eliminates the dependence on electrical power supplies, such as batteries. The main component of the actuator, a shape memory alloy (SMA) element, operates as a heat engine and converts the thermal energy of fuel combustion to mechanical energy. The incorporation of the high energy density fuel and actuator control hardware and software inside the unit makes for a compact actuator system, requiring only low-power, digital actuator control signals as input. Due to the relatively high recovery stress and strain of SMAs, the compact actuator can provide significant force and stroke. Convection heating and cooling of the SMA also results in relatively high actuation frequencies. Finally, if the compact actuator is utilized in the context of a larger system producing excess parasitic heat, the energy density of the actuator subsystem increases. The numerical analysis of the SMA element/strip was conducted using commercial software, including FLUENT and ABAQUS. The goal of the analysis was to estimate actuation frequency and actuation strain. In addition, a multi-channel combustor and a micro-tube heat exchanger were designed and analyzed. These results were compared to experimental tests and measurements (Part II).

The second part of this work discusses the fabrication and testing of a fuel-powered shape memory alloy actuator system (FPSMAAS) and its main components. Fuel (propane) is burned in a combustor and its heat is transferred to a working fluid medium, which in turn transfers the heat to the SMA element to drive its martensite-to-austenite phase transformation. For the austenite-to-martensite transformation, the heat is removed from the SMA element by a cooling fluid, from which the heat is then removed via a heat exchanger. The process of implementing the FPSMAAS consisted of three phases of increasing complexity, corresponding to three generations of the actuator system. For the final generation of the FPSMAAS, the SMA element was housed in a rectangular channel, featuring an innovative way of separating the cold from the hot fluid medium that alternately come in contact with the SMA element, in order to minimize the mixing between them. To meet our goal of miniaturization, a multi-channel combustor/heat exchanger and a micro-tube heat exchanger were developed and tested. The final actuator system was composed of pumps, solenoid valves, check valves, bellows, a micro-tube heat exchanger, a multi-channel combustor/heat exchanger and a control unit. The experimental tests of the final system resulted in 250 N force with 2.1 mm displacement and 1.0 Hz actuation frequency in closed-loop operation. The test results for the individual components as well as the final assembled actuator are compared with the results of the numerical analyses conducted in Part I and a good agreement has been demonstrated.

The purpose of this paper is twofold: to establish the magnitude of detwinning stress levels of non-modulated martensite in Ni₅₃Mn₂₅Ga₂₂ as a function of heat treatments utilizing single crystals, and to study the shape memory strains from constant-load temperature cycling experiments for aged and unaged conditions. The maximum transformation strains of 5.2% are consistent with the theoretical predictions based on energy minimization theory. The results exhibit a remarkable narrowing of the thermal hysteresis (as low as 2 °C) with increasing applied stress and a considerable two-way shape memory effect. Using microscopy, it is shown that aging produces a finer martensitic plate structure with an accompanying increase in detwinning stress compared to the unaged case.

Martensitic transformation and the low temperature plastic deformation proceed by the easiest coordinated atom movements, i.e. those which produce the least amount of energy dissipation. The two requirements, namely that the correct martensitic crystal structure results, and in addition that an undistorted interface between the matrix and the martensite is necessary, lead in general to more than one shear system. As will be discussed, the selection of the shear systems to comply with the transformation crystallography underlies the same selection principles as deformation by slip. Three examples are presented: the Cu–Zn based noble metal alloys, the Ni–Ti based ones and the Fe alloys. Whereas plastic deformation is induced by applied stresses, the martensitic transformation is controlled by a change in Gibbs free energy, but can be modified by applying stresses. The creation of transformation strains leads to an interaction with applied stresses. The different types of interaction between applied stress, transformation strain and associated plastic deformation are discussed for the three alloy groups.

The effects of the repeated stress-induced transformation (pseudoelastic fatigue) on the mechanical behaviour and microstructure of Cu–Zn–Al and Cu–Al–Ni single crystals are presented. Several microstructural changes occur during cycling at temperatures above the martensitic transformation temperature M_s in Cu–Zn–Al alloys. Bulk defects consisting of dislocation bands with retained martensite and intrusion–extrusion types of surface defects are observed. The fine characteristics of the defects, such as nucleation, density, as well as other microstructural features, depend on the working temperature, alloy composition, applied stresses and number of cycles. The presence of these defects alters the shape of the stress–strain curves in the pseudoelastic range for Cu–Zn–Al. Also diffusive phenomena strongly affect the mechanical behaviour of these alloys both in the parent and martensite phases, even for tests carried out slightly above room temperature. In turn, the kinetics of these diffusional processes has been found to depend on the density of bulk defects. Details of these processes are discussed and a simple model that describes the main characteristics of the mechanical behaviour evolution is considered. The interaction between the mechanical behaviour, density of defects and diffusion is discussed. In the case of Cu–Al–Ni, bulk defects are also shown to occur but the system is less prone to diffusion effects that could affect the mechanical behaviour. The defects generated during pseudoelastic cycling lead, however, to the inhibition of the γ' phase when a transition between the bcc parent phase and a mixture of two martensitic structures γ' (2H) and β' (18R) is involved at the beginning of the cycling process.

This paper reports a systematic study on indentation by a Vickers indenter and two-way shape memory in subsequent thermal cycling of a NiTi polycrystalline shape-memory alloy. It was found that, at a lower temperature, the indent is sink-in, while at a higher temperature, it is pile-up. However, after heating all indents are pile-up. The two-way shape memory is almost linearly dependent on the maximum indentation load but independent of the indentation temperature. The mechanism behind these phenomena is explained. The stability of two-way shape memory was also investigated.

Porous titanium–nickel (PTN) intervertebral fusion devices, produced by self-propagating high-temperature synthesis, represent an alternative to traditional long-term implants in the orthopaedic field. PTN promotes tissue ingrowth and has succeeded short-term and long-term biocompatibility in vivo testing. In this in vitro study, the PTN morphology was characterized using microfocus computer tomography (μCT) in order to calculate the active PTN surface. Potentiodynamic polarization testing was then performed to evaluate the in vitro corrosion resistance of PTN devices in Hanks' based salt solution. Direct coupling experiments of PTN with Ti6Al4V were also performed in order to establish the galvanic corrosion resistance of PTN intervertebral implants in the presence of potential Ti6Al4V supplemental fixation devices. Compared to the behaviour of other orthopaedic biomaterials and solid NiTi devices, PTN devices showed a level of corrosion resistance that is comparable to other NiTi devices and acceptable for the intended orthopaedic application. Further improvement of the corrosion resistance is still possible by specific electrochemical surface treatments.

An experimental configuration is demonstrated that captures features of the initiation of unstable stress-induced transformation in a shape memory alloy (SMA). The apparatus uses circulating fluids through the grips and a heat sink and thermoelectric devices to control the temperature profile of a specimen within a mechanical testing machine. The configuration can be used to restrict the initiation of phase transformation to a small region of interest of the free length, while permitting full-field optical tracking, infrared imaging, use of laser extensometry, and monitoring of load and extension. In this way, some longstanding difficulties in the measurement of thermo-mechanical phenomena in SMA wire have been resolved. The size of initiation stress peaks can be accurately measured for both transformation directions without changing the wire geometry, the temperature of a region of interest can be selected over a wide range, and imaging can be performed for multiple loading cycles and for events that occur from static to near dynamic rates. The motivation for this work is to produce high quality data for use in calibrating numerical models that study thermo-mechanical coupling during unstable transformation behavior.

In this paper we demonstrate a new class of superelastic NiTi honeycomb structures. We have developed a novel brazing technique that has allowed us to fabricate Nitinol-based cellular structures with relative densities near 5%. Commercially available nickel-rich Nitinol strips were shape-set into corrugated form, stacked, and bonded at high temperature by exploiting a contact eutectic melting reaction involving pure niobium. After heat treatment to restore transformational superelastic response, prototype honeycomb structures were subjected to severe in-plane compression loading at room temperature. The specimens exhibited good specific strength, high specific stiffness, and enhanced shape recovery compared to monolithic shape memory alloys (SMAs). Compressive strains of over 50% could be recovered upon unloading. The demonstrated architectures are simple examples of a wide variety of possible built-up topologies, enabled by the bonding method, that can be engineered for customizable net section properties, arbitrary shape, and kinematically enhanced thermomechanical shape-memory and superelastic response.

Recent experimental investigation showed that NiTi shape memory alloy microtubes exhibit deformation instability and complicated non-uniform strain evolution during combined tensile and torsional loading. In this paper, in situ profilometry is introduced to measure the non-uniform strain fields in the tube configuration. This method employs an optical profiler to quantify the surface morphology of the specimen in a full-field and non-contact manner. Existing defects and/or sprayed speckles on the specimen surface are used to build the displacement field of the specimen and therefore to obtain the strain field. This new method has been proven by experiment to be effective and reliable. It has led to two important experimental results: (1) inside the transformation domain the strains are uniform; (2) across the domain front the shear strain remains unchanged but the tensile strain experiences a rapid change.

The concept of graphite–polyurethane coatings as efficient, electrical resistors is the focus of this paper. A 60–40 graphite–polyurethane mix (weight %) demonstrated an electrical resistivity of 40.71 Ω mm. The graphite–polyurethane mix was coated on electrically insulating Kapton tape, which was then wrapped on a nichrome wire (nominal dimensions: 100 mm length and 1.5 mm diameter). This three-phase assembly was heated by Joule heating of the graphite–polyurethane layer. Steady state temperatures as high as 180 °C were attained under free convection conditions, at a very low power requirement of about 2.5 W as opposed to about 18 W for uncoated wires. Interestingly, the effect on transients (heating and cooling times) was not as dramatic. Experiments were also performed under vacuum conditions, following which an analysis is offered regarding the different modes of heat transfer. These coatings can potentially be used as efficient resistors for highly conductive, moderately high temperature shape memory alloys (e.g. the copper–aluminium–nickel system) or electrically insulating shape memory polymers. Any other thermally activated shape memory alloy (e.g. the popular nickel–titanium system) may also use the coatings as resistors due to the potentially dramatic energy savings that may be realized without a dramatic adverse impact on the frequency response.

This paper introduces a comprehensive nonlinear dynamic model of motion actuators based on ionic polymer metal composites (IPMCs) working in air. Significant quantities ruling the acting properties of IPMC-based actuators are taken into account.

The model is organized as follows. As a first step, the dependence of the IPMC absorbed current on the voltage applied across its thickness is taken into account; a nonlinear circuit model is proposed to describe this relationship. In a second step the transduction of the absorbed current into the IPMC mechanical reaction is modelled. The model resulting from the cascade of both the electrical and the electromechanical stages represents a novel contribution in the field of IPMCs, capable of describing the electromechanical behaviour of these materials and predicting relevant quantities in a large range of applied signals.

The effect of actuator scaling is also investigated, giving interesting support to the activities involved in the design of actuating devices based on these novel materials.

Evidence of the excellent agreement between the estimations obtained by using the proposed model and experimental signals is given.

In this paper, the propagation behavior of Love waves in a smart functionally graded piezoelectric structure is analyzed. The smart structure consists of three layers, in which the piezoelectric plate serves as the upper layer accompanied by a sandwiched graded layer and a metal substrate. In the graded layer, all the parameters and the elastic modulus are respectively assumed to vary after two mathematical forms, i.e. an exponential function and a linear function. The WKB method is adopted to analytically solve the propagation problem of Love waves for both electrical open and short cases on the free surface, respectively. The phase velocity curves demonstrate that the number of modes is greater than in the non-graded layer structure. Furthermore, the influence of graded variation on coupled electromechanical factor is presented.

This work proposes a new type of inertial mount featuring piezoelectric material and applies it to the vibration control of a flexible structural system. The dynamic model of the piezoelectric actuator consisting of an inertial mass and a piezoceramic stack is established and the inertial force of the actuator is experimentally evaluated. The inertial type of the piezoelectric mount is then devised and its governing equation is derived. In order to demonstrate the effectiveness of the proposed mount, a flexible structural system is constructed and its dynamic model is formulated in the modal coordinate. A linear quadratic Gaussian (LQG) controller associated with the piezoelectric mount is designed to attenuate the vibration of the structural system. The LQG controller is experimentally realized and control responses such as acceleration of the structural system are evaluated in the frequency and time domains.

This paper is concerned with modal control of a rectangular plate using both smart sensors and smart actuators. Since filtering and actuation are related, discussion of a smart sensor comes first followed by that of the counterpoint—a smart actuator. This paper begins by discussing an ideal case of modal control comprising extraction and activation of a target structural mode of a plate. Then, for the purpose of implementing a modal control system, three-dimensional smart sensors and actuators are introduced, the design procedure being investigated. Taking into consideration the practicability, two-dimensional smart sensors and actuators for modal filtering and modal actuation, respectively, are presented, and then the number, location and shaping function of the two-dimensional modal sensors and actuators are clarified. Furthermore, the extraction and actuation of the (1, 3) mode of a simply supported rectangular panel using the smart sensors and actuators is shown. Finally, a modal feedback control system is constructed utilizing both the smart sensors and the smart actuators, demonstrating the effectiveness at suppressing the particular structural mode of interest without causing spillover.

To generate a vibration-free zone in a designated region of a structure, cluster power flow control is presented. Unlike conventional modal-based vibration control, mainly used for augmenting structural damping, active wave control can render all the vibration modes inactive; the gain characteristic of a forced vibration response converges to the asymptote—to a far greater extent than augmenting structural damping; however, progressive waves still remain. Power flow control, advanced active wave control, allows the generation of a vibration-free zone when properly configured to handle all the governing structural variables. To implement power flow control, cluster control is introduced, which guarantees unconditional stability of a feedback system. To evaluate the vibration-free state, a vibration intensity is introduced; however, as a result of suppressing the vibration intensity, two extreme phenomena arise: suppression or excitation of structural modes. To overcome this problem, minimization of the dominant structural variables that are the counterparts of the vibration intensity elements is set up, with the result that a vibration-free zone is generated, together with the suppression of the vibration intensity. A numerical analysis is then conducted, followed by an experiment, demonstrating the capability of the proposed method for generating a vibration-free zone.

In this paper, unidirectional CFRP composites are considered as an electrical percolation system, which consists of electrically conductive carbon fibers and electrically nonconductive epoxy resin. Due to the contact behavior of the carbon fibers, CFRP has electrical conductivity in the width direction. Resistance measurements using the DC four-probe and the DC six-probe methods are conducted for the unidirectional CFRP specimens with different fiber volume fractions, i.e. different contact conditions. On the basis of the electrical anisotropy level obtained, the correlation between the measured anisotropy and the electrical ineffective length δ_ec, over which a broken fiber recovers its current carrying capacity and a key parameter in electromechanical modeling of CFRP, is shown experimentally. The empirical relationship between the electrical anisotropy and δ_ec obtained is also reviewed using a numerical calculation method based on the electric circuit theory of Kirchhoff's rule and the Monte Carlo simulation technique.

The use of piezoelectric actuators to modify an antenna radiation pattern has been investigated recently. In this paper, a doubly curved paraboloid antenna of Lexan^® material is considered, and the optimal size, location, and field of the piezoelectric actuators are determined, for steering the radiation. The deformation induced by the piezoactuators is determined using a shell element modified for this purpose, and validated using extensive experiments on curved beams and shells with piezopatches. The genetic algorithm is used as the optimization tool. One of the contributions of this paper is a formulation of the optimization problem which reduces the computational effort significantly. A reasonable amount of steering with good beam quality is achieved.

Shape memory alloys (SMA) are a particular family of materials, discovered during the 1930s and only now used in technological applications, with the property of returning to an imposed shape after a deformation and heating process.

The study of the mechanical behaviour of SMA, through a proper constitutive model, and the possible ensuing applications form the core of an interesting research field, developed in the last few years and still now subject to studies driven by the aim of understanding and characterizing the peculiar properties of these materials.

The aim of this work is to study the behaviour of SMA under torsional loads. To obtain a forecast of the mechanical response of the SMA, we utilized a numerical algorithm based on the Boyd–Lagoudas model and then we compared the results with those from some experimental tests. The experiments were conducted by subjecting helicoidal springs with a constant cross section to a traction load. It is well known, in fact, that in such springs the main stress under traction loads is almost completely a pure torsional stress field.

The interest in these studies is due to the absence of data on such tests in the literature for SMA, and because there are an increasing number of industrial applications where SMA are subjected to torsional load, in particular in medicine, and especially in orthodontic drills which usually work under torsional loads.

Development of a novel in situ method for simultaneous growth of single crystals and thin films of a smart material spinel is achieved. Material to be grown as metal-incorporated single crystal and thin film was taken as a precursor and put into a bath containing acid as a reaction speed-up reagent (catalyst) as well as a solvent with a metal foil as cation scavenger. By this novel method, zinc aluminate crystals having hexagonal facets and thin films having single crystalline orientation were prepared from a single optimized bath. Properties of both crystals and thin films were studied using an x-ray diffractometer and EDAX. ZnAl₂O₄ is a well-known wide bandgap compound semiconductor (E_g = 3.8 eV), ceramic, opto-mechanical and anti-thermal coating in aerospace vehicles. Thus a space_ gmr technique was found to be a new low cost and advantageous method for in situ and simultaneous growth of single crystals and thin films of a smart material.

ZnO/ZnS core–shell nanowires with the wurtzite structure have been grown using a simple catalyst-free thermal evaporation technique. The ZnO/ZnS core–shell nanowires are as long as several tens of micrometers, the thickness of the wires is about 1 µm at the bottom and 50 nm at the top respectively, and the thickness of the core ZnO is about 30 nm near the top of the wires. A high-magnification transmission electron microscope image of a single nanowire reveals a clear ZnO/ZnS interface. The photoluminescence spectrum of the products exhibits two peaks at about 400 and 410 nm respectively.

PZT thin films of various thickness were deposited on Pt(111)/Ti/SiO₂/Si(100) substrates by the sol–gel method. A new method to control the crystallographic orientation of the PZT thin films by modifying the coating layers in one annealing cycle is presented in this paper. PZT films with (111) and (100) preferred orientation were obtained using the multi-coating-layer-annealed and the one-coating-layer-annealed sol–gel methods, respectively. The optimized baking and annealing temperatures at 200 and 600 °C were obtained by analyzing the phase transformation of the PZT precursor solution. The effect of film thickness on the microstructures, crystalline phases, and electrical properties of the PZT films was investigated. Compared to the multi-coating-layer-annealed method, the one-coating-layer-annealed method decreases the residual stress of the PZT films. The remanent polarization increases with the thickness of the PZT films. The remanent polarization (P_r) and coercive fields (E_c) of the one-coating-layer-annealed PZT films with 1.6 µm thickness are 25.7 µC cm⁻² and 59.2 kV cm⁻¹, respectively. New microcantilevers with two PZT piezoelectric elements and three electric electrodes were designed. These structures can be applied to microsensors, microactuators, or versatile devices having both sensing and actuating functions. The cantilevers were successfully fabricated with wet and dry combined bulk micromachining technique.

Ti implants with surfaces modified by anodic and thermal oxidation were used to promote osteosynthesis. X-ray photoelectron spectroscopy, secondary ion mass spectroscopy and Auger electron spectroscopy were used to study the surface composition and the depth homogeneity of the implant materials before implantation and after removal from the patient (3 years later). The surface of the Ti implants was modified through anodic and thermal oxidation.

The surface of the Ti implant before use is covered by a 200 nm thick TiO₂ layer, with homogeneously distributed O at depth. During the formation of this TiO₂ layer, phosphate, Ca and a small amount of C are incorporated into the oxide. Similarly to the main impurities in the base Ti metal (Cr and Fe), the Ca and C accumulate at the oxide/metal interface.

Three years after the implantation, the binding state of the Ti in the oxide layer was unchanged. The base metal remained covered by TiO₂, the outer surface of which was partially covered by a mainly C-containing layer of varying thickness. The increases in the Ca and P contents of the TiO₂ layer during the 3 years in the human body can be explained by incorporation from the body.

Clinically, no metallosis or allergic reactions were observed.

Adaptive materials with rapidly controllable and switchable energy-absorption and stiffness properties have a number of potential applications. We have developed, characterized and modeled a class of adaptive energy-absorbing systems consisting of nonlinear poroelastic composites wherein a field-responsive fluid, such as a magnetorheological fluid or a shear-thickening fluid, has been used to modulate the mechanical properties of a cellular solid. The mechanical properties and energy-absorbing capabilities of the composite are studied for variations in design parameters including imposed field strength, volume fraction of the field-responsive fluid within the composite and impact strain rates. The total energy absorbed by these materials can be modulated by a factor of 1- to 50-fold for small volume fractions of the fluid (∼15%) using moderate magnetic fields varying from 0 to 0.2 T. A scaling model is also proposed for the fluid–solid composite mechanical behavior that collapses experimental data onto a single master curve. The model allows optimization of the composite properties in tune with the application requirements. Potential application areas are discussed with emphasis on applicability in impact-absorbing headrests and cushioned assemblies for energy management.

In this paper, to make a rotor–bearing system pass through the critical speed safely and control its vibration, a self-optimizing support system is proposed, based on shape memory alloy (SMA). In this self-optimizing support system, the SMA springs are used to construct the pedestal bearing for the rotor–bearing system. The principle of the dynamic absorber is utilized to calculate and change the stiffness of the SMA pedestal bearing in order for the rotor shaft to be usually situated near anti-resonance with changes of the rotating speed, and its vibration can be controlled. Numerical experiments suggest that the vibration of the rotor–bearing system can be controlled efficiently by a SMA spring bearing and the proposed method is feasible. The study also indicates that the proposed method has potential for solving a wide range of active vibration control problems. The results are also verified by an experimental investigation on a rotor–bearing system presented in a companion paper, II.

The companion paper, part I, presents a self-optimizing support system in theory. A pseudo self-optimizing support system is proposed for a rotor-bearing system based on shape memory alloy (SMA). A numerical simulation is given to verify the theoretical model. This paper presents the experimental study to test and verify the theoretical model further. The experiment process and experimental results are presented in detail, and some analyses and comparisons are given.

Thin PZT films have a major interest for active control of mechanical structures. Precisely, it is an open field for the isolation of micro-components sensitive to dynamic effects. Indeed, the electronic components used, for example, in aircraft endure intense vibrations due to acceleration. These vibrations have some disturbing effects on the frequency stability and on the usable life of the electronic elements. The isolation of these elements becomes crucial to protect them from the vibrating environment. In order to manage this problem, it is advisable to isolate the electronic card either at the case level or at the card level or at the sensitive element level. The latter solution was chosen. Thus, we have direct access to the control electronics and the energy sources and the control energy is lower. An active suspension system is developed between the support and the sensitive element to be isolated. An original active suspension system is designed. Some modeling difficulties arise due to the existence of the inevitable bottom electrode common to the actuating layers and to the sensing layer.

Switched electromechanical shunts have been proposed as a method to suppress vibrations of a mechanical structure. In this method, a piezoelectric patch is attached to the vibrating structure. An inductor and series switch are shunted across the electrodes of the piezoelectric patch. At maximum and minimum displacements, the polarity of the charge on the patch is changed by closing the switch that shunts a piezoelectric element to ground through an inductor. The method is similar to an active velocity feedback technique where the forces from the piezoelectric patch to the structure always oppose the velocity. Previous work used numerical simulation to demonstrate the degree of vibration suppression that can be obtained with the switched electromechanical shunts method. In this paper, a novel and transparent algebraic closed-form expression is derived. This algebraic expression predicts the displacement amplitude of the vibrating structure with a switched shunt in terms of the mechanical properties of the structure, the electromechanical coupling coefficient, and the shunt parameters. The closed-form expression clearly shows which system parameters need to be changed to optimize the performance of this vibration suppression method. Experiments were performed that verified the theoretical closed-form expression for displacement amplitude. It was shown that the maximum suppression that can be obtained with this method is determined by the allowable voltage drop across the piezoelectric element.

Early aircraft flight control systems were totally manually operated, that is, the force required to move flight control surfaces was generated by the pilot and transmitted by cables and rods. As aerodynamics and airframe technology developed and speeds increased, the forces required to move control surfaces increased, as did the number of surfaces. In order to provide the extra power required, hydraulic technology was introduced.

To date, the common element in the development of flight control systems has been, mainly, restricted to this type of technology. This is because of its proven reliability and the lack of alternative technologies. However, the technology to build electromechanically actuated primary flight control systems is now available. Motors developing the required power at the required frequencies are now possible (with the use of high energy permanent magnetic materials and compact high speed electronic circuits). It is this particular development which may make the concept of an 'all electric aircraft' realizable in the near future. The purpose of the all electric aircraft concept is the consolidation of all secondary power systems into electric power. The elimination of hydraulic and pneumatic secondary power systems will improve maintainability, flight readiness and use of energy.

This paper will present the development of multi-lane smart electric actuators and offer an insight into other subsequent fields of study. The key areas of study may be categorized as follows.

• State of the art hydraulic actuators.

• Electromechanical actuator system test programmes.

• Development of electromechanical actuators.

• Modelling of electromechanical actuators.

In this study, the response time and system characterization analyses of a high-torque magneto-rheological (MR) fluid limited slip differential (LSD) clutch are presented. The response time of the clutch is examined based on the objective of keeping the relative velocity difference of the shafts of the clutch less than a predetermined threshold value. The experimental setup allows the application of an external disturbance to the system, so that the relative velocity difference exceeds the threshold value. A velocity-based, closed-loop control system is designed and tested. Additionally, system identification experiments are performed to determine system parameters such as bearing friction coefficients, dry and viscous torque coefficients. These parameters are utilized in the theoretical response time analyses of the MR fluid LSD clutch. It is demonstrated that the overall response time of the system varies between 20 and 65 ms as a function of operating velocity and electromagnet current, including the response times of the controller, solenoid inductance and MR fluid and inertia effects. The response time reduces by increasing solenoid current and increasing the operating velocity.

The main purpose of this paper is to investigate the local deformation arising in NiTi plate subjected to stress concentration during tensile loading. First, a test system is constructed on the basis of digital image correlation (DIC) in order to measure the local strain distribution over the surface of the specimen. Second, we measure the local deformation (local strain distributions) arising in a 50.5Ni49.5Ti plate with a double notch under tensile loading. The results are as follows. (1) NiTi shows local deformation, such as nucleation and propagation of local strain bands, and its bands have a fixed angle (about 54°–55°) relative to the loading direction. On the other hand, the distribution of local shear strain is almost uniform when the local longitudinal strain bands propagate during tensile loading. (2) NiTi with stress concentration shows that the local longitudinal and lateral strain bands initiate at the stress concentration part, and propagate with a symmetrical shape. Later, the local strain band shows bifurcation and propagates toward both ends of the specimen with a fixed angle relative to the loading direction. The distribution of local shear strain is almost uniform when the local longitudinal strain bands propagate, but it is not uniform when the phase transformation initiates.

NiTi with stress concentration shows local deformation, such as nucleation, bifurcation and propagation of phase transformation, and its bands have a fixed angle relative to the loading direction. The transition of the local deformation mode has an effect on the macroscopic stress–strain curve. When predicting the macroscopic deformation behavior of NiTi structures, these local deformation modes arising in NiTi should be considered.

This paper presents a complete finite-element description of a hybrid passive/active sound package concept for acoustic insulation. The sandwich created includes a poroelastic core and piezoelectric patches to ensure high panel performance over the medium/high and low frequencies, respectively. All layers are modelled thanks to a Comsol environmentComsol is the new name of the finite element software previously called Femlab.. The piezoelectric/elastic and poroelastic/elastic coupling are fully considered. The study highlights the reliability of the model by comparing results with those obtained from the Ansys finite-element software and with analytical developments. The chosen shape functions and mesh convergence rate for each layer are discussed in terms of dynamic behaviour. Several layer configurations are then tested, with the aim of designing the panel and its hybrid functionality in an optimal manner. The differences in frequency responses are discussed from a physical perspective. Lastly, an initial experimental test shows the concept to be promising.

Wave propagation in carbon nanotubes (CNTs) is studied based on the proposed nonlocal elastic shell theory. Both theoretical analyses and numerical simulations have explicitly revealed the small-scale effect on wave dispersion relations for different CNT wavenumbers in the longitudinal and circumferential directions and for different wavelengths in the circumferential direction. The applicability of the proposed nonlocal elastic shell theory is especially explored and analyzed based on the differences between the wave solutions from local and nonlocal theories in numerical simulations. It is found that the newly proposed nonlocal shell theory is indispensable in predicting CNT phonon dispersion relations at larger longitudinal and circumferential wavenumbers and smaller wavelength in the circumferential direction when the small-scale effect becomes dominant and hence noteworthy. In addition, the asymptotic frequency, phase velocities and cut-off frequencies are also derived from the nonlocal shell theory. Moreover, an estimation of the scale coefficient is provided based on the derived asymptotic frequency. The research findings not only demonstrate great potential of the proposed nonlocal shell theory in studying vibration and phonon dispersion relations of CNTs but also signify limitations of local continuum mechanics in analysis of small-scale effects, and thus are of significance in promoting the development of nonlocal continuum mechanics in the design of nanostructures.

Recently, there is increasing interest in using superelastic shape memory alloys (SMAs) in civil, mechanical and aerospace engineering, attributed to their large recoverable strain range (up to 6–8%), high damping capacity, and excellent fatigue property. In this research, an improved Graesser's model is proposed to model the strain-rate-dependent hysteretic behavior of superelastic SMA wires. Cyclic loading tests of superelastic SMA wires are first performed to determine their hysteresis properties. The effects of the strain amplitude and the loading rate on the mechanical properties are studied and formulated using the least-square method. Based on Graesser's model, an improved model is developed. The improved model divides the full loop into three parts: the loading branch, the unloading branch before the completion of the reverse transformation and the elastic unloading branch after the completion of reverse transformation, where each part adopts its respective parameters. Numerical simulations are conducted using both the original and the improved Graesser's models. Comparisons indicate that the improved Graesser's model accurately reflects the hysteresis characteristics and provides a better prediction of the SMAs' actual hysteresis behavior than the original Graesser's model at varying levels of strain and loading rate.

This paper presents a simulation of vibration suppression of a laminated composite beam embedded with actuators of a giant magnetostrictive material subjected to control magnetic fields. It has been found that the strains generated in the material are not only significantly larger than ones created by many other smart materials but also exhibit some inherent nonlinearities. To utilize the full potential of these materials in active vibration control, these nonlinearities should be characterized in the control system as accurately as possible. In this simulation of nonlinear dynamic controls, the control law with negative velocity feedback and the analytical nonlinear constitutive model of the magnetostrictive layer are employed. The numerical results show that this proposed approach is efficient not only in a linear zone but also in nonlinear zones (dead zone and saturation zone) of magnetostrictive curves in vibration suppression. Compared with those from the control system based on the linear constitutive relations of the material, it is found that the simulation results based on the linear model are efficient only when the magnetostrictive relations are located in the linear zone. Once the system has some departure from the linear zone, however, the results from the linear model become unacceptable. Finally, the effect of material properties, lamination schemes and location of the magnetostrictive layers on vibration suppression of the practical system is evaluated.

Piezoelectric actuators offer significant promise in a wide range of applications. The piezoelectric actuators considered in this work essentially consist of a flexible structure actuated by piezoceramics that must generate output displacement and force at a certain specified direction and point of the domain. The design of these piezoelectric actuators is complex, and a systematic design method such as topology optimization has been successfully applied in recent years, with appropriate formulation of the optimization problem to obtain optimized designs. However, in these previous design formulations, the position of the piezoceramic is usually kept fixed in the design domain and only the flexible part of the structure is designed by distributing some non-piezoelectric material (aluminum, for example). This imposes a constraint in the position of the piezoelectric material in the optimization problem, limiting the optimality of the solution. Thus, in this work, a formulation that allows the simultaneous distribution of non-piezoelectric and piezoelectric material in the design domain to achieve certain specified actuation movements will be presented. The optimization problem is posed as the simultaneous search for an optimal topology of a flexible structure as well as the optimal position of the piezoceramic in the design domain and optimal rotation angles of piezoceramic material axes that maximize output displacements or output forces in a certain specified direction and point of the domain. The method is implemented based on the SIMP ('solid isotropic material with penalization') material model where fictitious densities are interpolated in each finite element, providing a continuum material distribution in the domain. The examples presented are limited to two-dimensional models, since most of the applications for such piezoelectric actuators are planar devices.

An asymptotically correct theory for multi-cell thin-wall anisotropic slender beams that includes the shell bending strain measures is extended to include embedded active fibre composites (AFCs). A closed-form solution of the asymptotically correct cross-sectional actuation force and moments is obtained. Active thin-wall beam theories found in the literature neglect the shell bending strains, which lead to incorrect predictions for certain cross-sections, while the theory presented is shown to overcome this shortcoming. The theory is implemented and verified against single-cell examples that were solved using the University of Michigan/Variational Beam Sectional Analysis (UM/VABS) software. The stiffness constants and the actuation vector are obtained for two-cell and three-cell active cross-sections. The theory is argued to be reliable for efficient initial design analysis and interdisciplinary parametric or optimization studies of thin-wall closed cross-section slender beams with no initial twist or obliqueness.

The analytical closed-form solution for the deflection of a novel beam-shaped shear-mode PZT actuator is derived. The actuator can be applied in a microfluidic system. Its active region possesses two oppositely poled segments separated by one non-poled segment. The deflections in four cases of segment dimensional designs are compared between the analytical solution and the ANSYS FEM solution. The two solutions show good agreement. And the maximum deflection of the case with a symmetrical design is the largest. Moreover, an experiment for the sample with symmetrical design is performed and the actual shear piezoelectric coefficient is obtained. Based on the actual coefficient, the experimental result agrees with the analytical solution very well.

The transformation kinetics formulation is the principal factor underlying the constitutive model of shape memory alloys (SMAs). Therefore, the transformation kinetics formulation, which is applicable to any status of stress and temperature, is essential for predicting the material behavior of SMAs. In this work we show that the transformation kinetics of the original Brinson model, which is the most widely used one-dimensional model, has shortcomings in the case where temperature decreases at low temperature (T<M_s). In addition, we propose a modified transformation kinetics formula that can be used for dual transformation conditions. The martensite transformation kinetics is modified so that the transformation from austenite into temperature-induced martensite, due to the decrease in temperature, is coupled with a transformation from austenite or temperature-induced martensite into stress-induced martensite, due to the increase in stress. Through this modification, the suggested formulation can properly describe the behavior of martensite fractions in the dual transformation region.

IPMC (ionic polymer metal composite) is composed of ionic polymer and metal electrodes on both surfaces of the polymer. In this study, we changed the surface morphology of the ionic polymer by using plasma treatment. Plasma treatment made needle-shaped microstructures on the surface of the polymer and the microstructures helped to form a thicker uniform metal electrode which is deposited by electroless plating on both sides of the polymer. We observed the actuating properties of IPMC such as displacement, force and lifetime by using the laser displacement measurement system and the load cell. Then we evaluated the enhanced characteristics of an IPMC actuator. Oxygen (the chemical etching gas) and argon (the physical etching gas) were used as the plasma source gas and the oxygen plasma resulted in higher performance.