Scholarly article on topic 'Photogenerating work from polymers'

Photogenerating work from polymers Academic research paper on "Materials engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Materials Today
OECD Field of science
Keywords
{}

Abstract of research paper on Materials engineering, author of scientific article — Hilmar Koerner, Timothy J. White, Nelson V. Tabiryan, Timothy J. Bunning, Richard A. Vaia

The ability to control the creation of mechanical work remotely, with high speed and spatial precision, over long distances, offers many intriguing possibilities. Recent developments in photoresponsive polymers and nanocomposite concepts are at the heart of these future devices. Whether driving direct conformational changes, initiating reversible chemical reactions to release stored strain, or converting a photon to a local temperature increase, combinations of photoactive units, nanoparticles, ordered mesophases, and polymeric networks are providing an expansive array of photoresponsive polymer options for mechanical devices. Framing the typically geometry-specific observations into an applied engineering vocabulary will ultimately define the role of these materials in future actuator applications, ranging from microfluidic valves in medical devices to optically controlled mirrors in displays.

Academic research paper on topic "Photogenerating work from polymers"

The ability to control the creation of mechanical work remotely, with high speed and spatial precision, over long distances, offers many intriguing possibilities. Recent developments in photoresponsive polymers and nanocomposite concepts are at the heart of these future devices. Whether driving direct conformational changes, initiating reversible chemical reactions to release stored strain, or converting a photon to a local temperature increase, combinations of photoactive units, nanoparticles, ordered mesophases, and polymeric networks are providing an expansive array of photoresponsive polymer options for mechanical devices. Framing the typically geometry-specific observations into an applied engineering vocabulary will ultimately define the role of these materials in future actuator applications, ranging from microfluidic valves in medical devices to optically controlled mirrors in displays.

Hilmar Koerner1,2, Timothy J. White1,3, Nelson V. Tabiryan4, Timothy J. Bunning1, and Richard A. Vaia1*

1 Materials and Manufacturing Directorate, Air Force Research Laboratory, Wright-Patterson AFB, OH 45433, USA

2 Universal Technology Corporation, Dayton, OH 45432, USA

3 General Dynamics Information Technology, Dayton, OH 45431, USA

4 BEAM Engineering for Advanced Measurements Company, Winter Park, FL 32789, USA *E-mail: Richard. Vaia@wpafb.af.mil

Devices and machines that convert energy from one form to another are a common thread woven through modern society, ranging from the generation of electrical energy from solar radiation to power our homes, to the creation of mechanical motion from the chemical energy of gasoline to power our cars. Increasing the efficiency of these devices and machines, or inventing alternatives, is arguably one of the major themes permeating engineering. In most instances, the design of a

machine determines the optimal bandwidth, frequency, and magnitude of the input and output energy or power. The energy conversion process occurs at a surface or within a functional material at the heart of the device.

Identifying materials and material combinations that convert an input stimulus to mechanical work (or conversely enables harvesting of the mechanical energy) have been of long standing interest. There are some outstanding examples: the description of the pyroelectric

34 materialstioday july-august 2oos | volume 11 | number 7-8

ISSN:1369 7021 © Elsevier Ltd 2008

effect in tourmaline by Theophrastus in 314 BC; the observation of piezoelectricity by Pierre and Jacques Curie in 1880; and the development of the bimetallic strip by John Harrison in the 1700s to measure nautical time. These so-called smart, autonomous, and intelligent materials and material systems form the basis of modern devices ranging from actuators and microelectromechanical (MEMs) switches to quartz watches and thermostats. The extensive array of options are summarized in numerous books1-3 and review articles4-11 where the framework chosen for the discussion ranges from material type, to the input stimuli or output result.

Recently, the proliferation of laser sources has catalyzed a growing interest in the use of light (photons) as the energy input (Fig. 1). In contrast to thermal, electrical, or chemical approaches, a coherent and wavelength-specific photon flux enables facile 'remote' control of the process, as well as precise spatial localization with a resolution limited by the diffraction limit of the radiation wavelength (submicron). One of the first realizations of this idea was by Uchino in 198912, who demonstrated photostriction of PLZT (lanthanum modified lead zirconate titanate) with the intent to drive robots remotely in a noncontact, wireless mode.

In general, the energy of a photon can be converted to mechanical work in one of two ways - momentum transfer or absorption. Although a photon of wavelength k is massless, it possesses momentum, p = h/k, where h is Planck's constant. When a photon interacts elastically with a totally reflective surface, it creates a radiation pressure because of the momentum transfer from the

photon to the substrate. This pressure is nominally very weak; for example, the radiation pressure from the sun on the Earth's surface is approximately 4.6 pPa (pJ/m3)13. Alternatively, if the photon is adsorbed, the energy directly increases the internal potential energy of the absorbing molecule or material. For example, a 1 s dwell of a 1 W laser onto an area of 1 cm2 delivers 1 GJ/m3 (GPa), provided that complete absorption of the laser occurs within 10 pm of the surface. For comparison, the energy density of gasoline is around 34 GJ/m3 (MJ/L)14. Higher laser powers and longer dwell times can provide sufficient energy to degrade or ablate a material, useful in applications such as laser cutting and thin film deposition. Thus, the local energy density available for mechanical work can approach the chemical or mechanical stability of the material.

Therefore, the material design challenge for mechanically photoresponsive materials is increasing photochemical stability while providing a highly efficient way to maximize the direct conversion of the delivered potential energy, dU, into mechanical work, PdV, or to capitalize on other auxiliary processes, such as the generation of charge or thermal energy, to change the state of the material to create a net shape or volume change15. This internal efficiency of the material (mechanical work out/photon energy in) sets the maximum possible efficiency of subsequent devices.

In general, the photon can serve as the sole source of energy (direct) or enable the material to overcome a local energy barrier (indirect) that subsequently leads to a new state with a relatively lower potential energy and a net volume or shape change. In the former

f ' ( f r ' <

Transduction

Volume/shape change

Motion and device

Fig. 1 Conversion of light to mechanical work. Photons emitted by a light source (coherent, wavelength specific to incoherent, broad-band) and absorbed by a photoactive material may produce mechanical motion either by: (a) a reversible conformational change, or (b) a local temperature increase via non-radiative decay. When these molecular processes are amplified to macroscopic shape or volume changes, mechanical work or motion can be integrated into a range of remotely controlled devices, such as beam stirring mirrors or robots.

case, direct extraction of the increased potential energy of the system restores the material to its original state. Metaphorically speaking, the absorption 'loads' a 'spring' (material) and elastic recovery of the spring drives the material back to its stress free state. In the latter case, the photon energy provides a 'trigger' to release 'stored' potential energy. Effectively, this process appears to 'amplify' the impact of the photon absorption. To move the system back to the original potential energy state, another energy source is required, such as thermal equilibration with the surroundings, entropic recovery of a network, or an applied external load. Here the 'spring' is preloaded and 'locked' in place. The photon releases the lock, enabling the recovery of the spring and necessitating an external load to reset the locked spring.

Whether direct or indirect, there are two major mechanisms through which the absorbed photon energy can lead to mechanical work (a) reversible conformation change, including photo-isomerization, photo-induced charge generation, or initiation of a reversible photochemical process; or (b) a local temperature increase via nonradiative thermal decay. These processes, if coupled to a net volume change (e.g. phase transition) or redistribution of volume within a unit element (e.g. shape change or reorientation), will generate a pressure (expansion or compression) in the direction of the change. Fundamentally, the multidimensionality of the output is best described in tensors16. Practically, the processes must be reversible and predictable across numerous cycles.

Using some recent literature examples and following the above framework, this review will discuss several approaches to generating work from photoresponsive polymers. Note that these examples are chosen as illustrations and are not inclusive of the many examples emerging in this rapidly developing field.

Photo-induced reversible conformational change

A polymer stores input tensile or compression force as potential energy via elastic deformation of the molecular network. When the input force is removed, the stored potential energy is released as the polymer returns to its original form - completing a cycle of input force, stored as potential energy, and subsequently released as mechanical work. Clearly, other energy inputs, if used to drive direct shape change of a polymer network, could also have utility. To this end, scientists have often used polymeric materials to convert thermal energy into mechanical output. This process, however, is limited in utility by speed, necessary temperature control apparatus, and application requirements. An ideal system may be one in which energy is converted into motion through isothermal, local, and directed exposure of an energy source - i.e. light. As early as 1967, scientists alluded to the application of photoresponsive macromolecules as light-energy transducers. "If chromophores were parts of polymers, or bound to them, light energy might influence the conformation of the polymers due to isomerization of the chromophore. Various systems could be used to convert light

energy isothermally to potential energy, stored in a macromolecule in such a way as to change its conformation17."

The fundamental tenet of the development of mechanically responsive polymers based on photo-responsive molecular moieties remains the same to this day. Photochromic molecules such as spyropyranes18, azobenzenes19,20 stilbenes21, fulgides22, and diarylethylenes23 change structure when irradiated with a distinct wavelength of light, thereby resulting in a local redistribution of molecular volume and associated stress, whether they are incorporated into the polymer chain or merely dissolved in the polymer matrix. To date, azobenzene has been by far the most commonly used moiety because of its thermally stability, a distinguishable absorbance of trans and cis isomers, and a relatively rapid thermal cis-trans back reaction.

Transduction of a molecular event to a macroscopic pressure via photoisomerization necessitates the ability of the surrounding matrix to respond to the local shape change. This fundamentally limits the material design considerations. For example, single crystal azo compounds typically shatter upon photoisomerization. Only recently has an azo compound been crystallized that can transform macroscopically, bending in response to the trans-cis isomerization24. Applied pressure can also reduce or even quench the macroscopic manifestation of the isomerization process. Zouhier et al.25 have shown that an external pressure on a poly(methyl methacrylate) (PMMA) azobenzene side chain polymer immobilizes the trans-cis isomerization at hydrostatic pressures of about 150 MPa. Hugel et al.26 have measured the mechanical forces on a single polymer change induced by trans-cis isomerization using an atomic force microscope (AFM) cantilever. The resulting calculations on energy input and work per polymer chain estimate that the isomerization process is 10% efficient; that is, 10% of the incident photon energy (4 x 10-19 J) is converted to mechanical work (~4.5 x 10-20 J). These studies imply that photomechanical effects via azobenzenes are only possible within materials that exhibit 'soft' properties.

Despite considerable effort, the viability of azobenzene polymers as photomechanical materials has only recently been realized27. Numerous reviews of photo-driven changes in azobenzene liquid crystal (LC) polymers reflect the high level of interest28-31. By coupling a photoresponsive group with liquid crystalline anisotropy, Finkelmann and coworkers have demonstrated that the photo-driven trans-cis isomerization of azobenzene causes a reversible shape contraction of nearly 20% in a liquid crystal elastomer (LCE) containing 20 mol% azobenzene covalently bonded to the network32. Subsequent synthetic efforts by other groups have quickly followed, primarily aiming for materials with faster response and larger deformation33-35. Ultraviolet (UV)-driven trans-cis isomerization of azobenzene causes macroscopic volume changes by reducing the molecular axis of azobenzene, as well as the order parameter of the system. The difficulty in demonstrating large-scale shape change in non-LC systems indicates that, primarily, the vehicle for shape change is the disruption

Fig. 2 Photodriven bending in azobenzene LC copolymers (LCPs) as a function of molecular organization. Illuminated by 365 nm UV light from the top. Temporal bending shown from left to right from 0 s to 0.5 s, recovery in dark is shown on far right. Twisted nematic (TN) and splay geometries increase the speed, amplitude, and work output for photodriven bending in these materials. Notably, the bending direction of TN and splay is dependent on the exposed surface geometry, bending towards the source when the molecular director is parallel to the long axis of the cantilever. (Reproduced with permission from38. © 2007 Institute of Physics.)

of LC order by trans-cis isomerization in azobenzene-LCE systems. The trans-cis isomerization of azobenzene is well known to reduce the order parameter in both LCs and polymer LCs, resulting in an isothermal order-disorder transition36. An added control to azobenzene LC polymers is polarization-directed bending, as wonderfully demonstrated by Ikeda and coworkers37. In these particular systems, because of the slow relaxation of cis-azobenzene, the molecular-level potential energy is released over the course of many hours through a thermal back-relaxation. If necessary, a second optical source with a longer wavelength light (where cis absorbs and trans does not) can be useful in driving the cis-trans isomerization to return the polymer rapidly to its original form.

Increasingly, efforts are exploring the impact of the macroscopic order of the LC and other mesophases on the amplification of the molecular level isomerization process. For example, Broer and coworkers38 have recently shown that twist and splay network structures significantly increase photogenerated mechanical work

from only 10-3 kJ/m3 in uniaxially aligned glassy azobenzene-LC polymers (azobenzene-liquid crystal network, or LCN) to 34.4 kJ/m3 and 42.5 kJ/m3, respectively (Fig. 2). In addition to macroscopic approaches, other researchers have started to examine miniaturized nano- and micro-sized structures in these materials. Of note, Terentjev and coworkers39 have demonstrated nanostructured cylindrical azobenzene-LCE pillars and, with both thermal and UV exposure, demonstrated topological changes. The photosensitivity of azobenzene has also been demonstrated in two-dimensional inorganic 'nano'-crystals, such as layered aluminosilicates and smectites, where photoisomerization of intercalated azobenzene molecules is being used to direct nanoscale swelling parallel to the layering direction. Ogawa etal.40 have demonstrated that a change of up to 10% in thickness of interlayer spacing can be obtained by irradiating their intercalated aluminosilicates with UV light. The objective is to use these photoresponsive hybrid materials as smart absorbents. Recent molecular modeling studies show that the similar systems when

july-august 2oos | volume 11 | number 7-s materialstoday

Fig. 3 Monodomain azobenzene-LCN with dimension 5 mm x 1 mm x 50 ym is subjected to a polarization-dependent optical protocol. (a) The azo-LCN cantilever is mounted with nematic director parallel to the long axis of the film; (b) under exposure to light polarized orthogonal to the monodomain direction (Eln), the cantilever bends 40° towards the laser source ; and (c) subsequent exposure to light polarized parallel to the monodomain (E//n) causes the cantilever to oscillate at an amplitude of 170° with a frequency that nears 30 Hz. The oscillation can be turned off by rotating the laser beam polarization to Eln (ii) or shuttering the laser. (d) When the laser is blocked, the azo-LCN cantilever returns to its original position.

configured properly can perform as solid state actuators, with up to 20% uniaxial strain41.

A polarization-dependent but fundamentally different mechanism known as trans-cis-trans reorientation has lately been used to trigger bidirectional photomechanical deformation in a polydomain azobenzene-LCN42,43. Here the reorientation of the LC domain is believed to drive shape changes, inducing local strain. Because of the nearly equivalent absorption of both trans and cis azobenzene isomers in the region of 442-514 nm, both trans-cis and cis-trans isomerization may be driven simultaneously. Since azobenzene is dichroic in nature, absorbing more when the polarization is parallel to the long axis of the molecule statistically increases the concentration of trans-azobenzene oriented orthogonal to the polarization of the driving light source, resulting in reorientation of the LC domain.

Recently, both UV (365 nm) and higher wavelength 442 nm-514 nm lasers have been used drive large angle (170°) and high frequency (30 Hz) oscillations in monodomain azo-LCN cantilevers44. As shown in Fig. 3, the oscillation can be turned on and off by simply rotating the polarization of the laser beam. The oscillation of the cantilever is facilitated by the inherent network structure, optics of the system, and the photoisomerization of azobenzene. This work demonstrates the possibilities afforded in photoresponsive polymers by optimizing the material response (photoisomerization within monodomain LCNs) within a device design (direction of optical source and cantilever size and shape).

In contrast to deriving mechanical work from a photo-induced phase change, as exemplified by the above discussion on azo-LCs, photoisomerization also provides a means to alter the chemical potential of the material isothermally to drive mixing or mass transfer or modulate a physical property such as permeability. Effectively, the photo-driven molecular process serves as a trigger to alter the state of the system, enabling subsequent re-equilibration with the environment to drive shape or volume change. The array of

impressive photoresponse processes in nature serves as inspiration and provides insight as to the potential material concepts, as well as the integration of these photochemical processes into future devices. For example, the movement of plant leaves when struck by light, such as the opening and closing of flowers, and the extension of leaves toward the daylight or, in shade-loving species, turning away from direct sunlight, is driven by a photo-hydraulic response based on photo-modulation of membrane permeability and associated electrolyte concentration within the pulvinus organ at the base of the stalks of leaves and leaflets45. Although rarely discussed collectively, there are numerous synthetic examples in the literature capitalizing on the change in chemical potential provided by photoisomerization to create motion. These range from photo-induced aggregation/de-aggregation of colloids driven by isomerization on the colloid surface46 to photo-initiated reversible cross-linking to control shape-memory polymers47.

Nonradiative thermal decay

A limitation of systems with photo-induced reconfigurable units, such as azobenzenes, is the often complex synthesis and inherent photochemical instability of organic dyes. An alternative is to add a finely dispersed second phase that functions as a transducer - a mechanically responsive system - converting photons into the appropriate energy form to drive the volume or shape change. In addition, this blending concept can serve to improve stiffness further, create anisotropic reinforcement, or impart additional functionality such as low permeability, magnetic susceptibility, or electrical conductivity.

Recently, a handful of articles have discussed nanocomposite concepts for mechanically active materials, where nanoparticles are added to an optically transparent matrix to absorb photons and convert the energy to heat. The small volume of nanoparticles results in a high number density within the matrix. Thus uniform dispersion

Fig. 4 A CNT-shape memory polymer nanocomposite curling toward an infrared (IR) source. IR absorption by the CNTs is constrained to the near-surface region of a stretched ribbon. Nonradiative decay results in local heating, which leads to contraction of the near-surface region and curling of the ribbon toward the infrared source within 5 s.

of only a few percent can effectively create a two component system where the entire matrix is within 10 nm of a nanoparticle48,49. This results in uniform, rapid heating and thus the ability to drive thermal events remotely. Note that this general concept is being used in diverse fields ranging from polymer welding 50 to cancer therapy51 .

For example, carbon nanotubes (CNTs) have been used to aid in the recovery of stored energy of a shape memory thermoplastic polyurethane (TPU) matrix 52. Pristine TPU exhibits modest shape memory properties. However, by adding CNTs, the shape memory properties are greatly enhanced. For example, recovery stress (energy output) is increased by a factor of two at a loading of only 2.9 vol% of CNTs. This can be further improved by the addition of more CNTs, however the mechanical properties deteriorate beyond loadings of 15vol% . In addition, remote triggering of the recovery with light is possible because of the efficient light absorption of the CNTs (Fig. 4). The temperature increase melts the strain-induced soft-segment crystallites, which serve as physical cross-links, and releases the stored entropic energy of the deformed amorphous chains, enabling the system to recover to its original undeformed state. This procedure enables a bimorph effect as the infrared (IR) light only penetrates through the top layer of the material, leading to a curling of a thin sample toward the light source.

Similarly, Ahir and Terentjev53 have examined a polydimethyl siloxane elastomer with CNTs. The optically triggered recovery is strongly dependent on the prestrain of the nanocomposite, allowing a bidirectional response - contraction at low prestrain and elongation

at higher prestrains (Fig. 5). The detailed mechanism of this behavior is still not fully understood, though some evidence indicates that the response does not solely arise from entropic recovery of the elastomeric network, but may also reflect an optically-triggered expansion or contraction of the CNTs. Some groups have hypothesized that the flexible CNT network exhibits polymer-like elasticity54. Conceptually, this implies that highly anisotropic nanoparticles may exhibit direct photo-driven configurational changes as mechanistically discussed in the previous section. Initiation reports on stress relaxation in pure CNT networks are available55.

In addition to CNTs, other absorptive nanoparticles such as Au and Fe3O4 have been used as nano-heaters. Kotov and coworkers56 have discussed Au nanoparticle assemblies as heaters, which lead to melting of the surrounding matrix and shape change. Schmidt etal.57 have used dispersed magnetic nanoparticles and an oscillating magnetic field, or radio frequency (RF)-microwave radiation, to heat the nanocomposite and trigger shape recovery. In contrast to CNTs and other strongly adsorbing inorganics, the temperature dependence of magnetic susceptibility effectively creates a self-limiting processing, fundamentally restricting the maximum temperature.

Optical triggering of materials other than shape memory polymers has also been developed. Sershan et al.58 have shown that hydrogels incorporating Au nanoparticles can be used as optically addressable microfluidic valves. UV irradiation leads to swelling and deswelling of the thermoresponsive hydrogel because of the strong absorption of the Au nanoparticles, whose local heating leads to expansion of the matrix. This mimics similar phenomena observed as early as 1971 with light-sensitive polymers that isothermally alter the expansion-collapse transition59. Furthermore, a photoelectrochemical actuator consisting of poly(acrylic acid)/Cu2+ gel and TiO2 nanoparticles has been reported. The reversible expansion/contraction upon UV light exposure is attributed to the dissociation/association of carboxylic group/Cu2+ binding arising from sensitization to UV provided by the TiO260.

Future challenges

A material's efficiency of energy conversion, modulus, strength, and parameters quantifying shape change are the intrinsic characteristics key to the generation of work. Actuation is a device-level function in that speed, output force, and stroke are dependent on the engineering design. Numerous classes of actuator devices are available using materials and physical phenomena including electromagnetic, electromechanical, fluidic, piezoelectric, and combinations of the above. The design of many types of 'device' has been worked repeatedly in conjunction with the optimization of specific material traits to enable a wide range of potential and realized applications. Several key measures can be used to classify the 'performance' of the various classes, as summarized elegantly and concisely by Zupan et al.61, including

Fig. 5 Actuation stress of an elastomer filled with different amounts of CNTs as a function of prestrain. The right axis shows actuation stroke. (Reproduced with permission from53. © 2005 Nature Publishing Croup.)

maximum force output, weight, maximum stroke, frequency, and work capacity. Although the ability to 'actuate' using photon-driven processes as summarized here is an exciting potential paradigm in the production of mechanical work, a more systematic expression of the observed behaviors in terms of these key measures is crucial. The design of specific device architectures envisioned to best convert the local photoinduced changes to a specific work function (i.e. a valve, a piston, etc.) is the next frontier in the area. This marriage of geometry-specific material capability framed in an applied engineering vocabulary will ultimately define the niche (advantages and disadvantages) these materials hold in the actuator community.

Shown in Figs. 6 a, b, and c are graphs comparing a broad range of actuators as a function of these key measures. The advantages of polymer-based photo-driven systems to date are low weight, large stroke, and the potential for high frequency (Fig. 3). In simplistic terms, these observed properties put them in the upper right quadrant of graph (a), and the upper left quadrant of graphs of (b) and (c). The recent demonstration of high frequency and large stroke oscillations in low weight azobenzene-LCNs (10-7-10-8 kg) position these materials uniquely in Fig. 6. However, more quantitative and systematic measurements of actual output force, work capacity, and true efficiency in the language of classic device-based actuators is

needed to benchmark the place of these photodriven systems in the broader actuator community. Recent work by van Oosten et al.38 has attempted to formulate simple measures of work for bending processes in photo-induced cantilevers.

Clearly, as demonstrated in Fig. 6, a number of systems have been documented with similar or in some cases, enhanced properties, with respect to efficient energy conversion, work output, and frequency. However, the actuator systems formed from photoresponsive polymers described here have distinct and unrivaled advantages. Foremost of these is that light can be a cheap energy source. From the vantage point of system architecture, the ease of localized spatial, temporal, spectral, and luminous control of light, in a lamp, laser, or fiber optic network is unique to photo-driven systems, and advantageous both in large-scale millimeter-sized cantilevers and small-scale micron-sized pillars. In addition, a given light source can be optically configured to focus on a small or large area, form a patterned gradient in light intensity (e.g. holography), and contains potential controls such as polarization rotation. As such, light can also be used to precondition these systems with 'memory' that can be isotropic (the same everywhere) or spatial (local or periodic). Remote-controlled photo-driven deformations, whether of a valve in a microfluidic chip or manipulating the focal length of a photoactuated mirror (using light

Fig. 6 Actuator classification and selection graphs. (a) Plot of maximum actuator output force versus maximum stroke (diagonal = work capacity of 0.01 Nm). (b) Plot of actuator work capacity, F8 versus mass. (c) Maximum frequency versus mass. Actuators that can cycle at > 50 Hz are identified and possess low mass. (Reproduced with permission from61. © 2002 Wiley-VCH.)

to control light), have some intriguing potential. This remote ability could greatly simplify specific device design and cost, while the potential to actuate 'units' on the scale of microns or even hundreds of nanometers in a singular or parallel fashion (i.e. a spatial plane) is intriguing.

Another advantage of photodriven systems is the plethora of synthetic and material design options that can be used to design devices optimized for certain performance parameters or behaviors. The ability to build into the baseline material wavelength specificity, specific modulus, and designer amplifiers through chemistry has the potential to enable materials that might one day reside over a much larger portion of the charts in Fig. 6. The ability to control response and reversibility using light of different colors by subtle variations in component chemistry are an underutilized control variable. Another unutilized potential control of 'unit' performance is engineering property differences across the geometry of the actual actuator element (i.e. grade the modulus through a thickness or spatially vary

responsivity across an x-y plane), which could greatly increase distinct performance characteristics in conjunction with a particular device design.

In summary, whether driving direct conformational changes, initiating reversible chemical reactions to release stored stain, or converting the photon to a local temperature increase, combinations of photoactive units, nanoparticles, ordered mesophases, and polymeric networks are providing an expansive array of active material options for mechanical devices. Future connection of these possibilities to engineered device design will provide rapid identification of the major unique niches in the actuator community that these materials systems can revolutionize. B!

Acknowledgments

Financial support was graciously provided by the US Air Force Office of Scientific Research and the US Air Force Research Laboratory, Materials and Manufacturing Directorate.

REFERENCES

1. Leo, D. J., Engineering Analysis ofSmart Material Systems, Wiley, New York, (2007)

2. Schulz, M. J., et al., (eds.) Nanoengineering of Structural, Functional and Smart Materials, CRC Press, Baton Rouge, (2005)

3. Bar-Cohen, Y., Adv. Mater (2006) 38, 3

4. Ikeda, T., et al., Angew. Chem. Int. Ed. (2007) 46, 506

5. Baughman Ray H., Science (2005), 308, 63

6. Jonas, A. M., et al., Macromolecules (2007) 40, 4403

7. Wilson, S. A., et al., Mater. Sci. Eng. R (2007) 56, 1

8. Erick, J. D., et al., Adv. Mater. (2007) 19, 4024

9. Einaga, Y., J. Photochem. Photobiol. C (2006) 7, 69

10. Vaia, R. A., and Baur, J., Science (2008) 319, 420

11. Madden, J. D., Science (2007) 318, 1094

12. Inoue, M. et al., Photodriven relay using PLZT ceramics. Proc. IEEE Int. Symp. Appl. Ferroelectr., 6th (1986), 16

13. Lembessis, V., Europhys. News (2001) 31, 7

14. Davis, S. C., and Diegel, S. W., Transportation Energy Data Book,26th edition, Center for Transportation Analysis of the Oak Ridge National Laboratory, Appendix B, (2007)

15. Kittel, C., and Kroemer, H., Thermal Physics, W. H. Freeman, New York, (1980)

16. Barber, J. R., Elasticity (Solid Mechanics and Its Applications), Springer, New York, (2003)

17. Lovrien, R., Proc. Natl. Acad. Sci. USA (1967) 57, 236

18. Dumont, M., and El Osman, A., Chem. Phys. (1999) 245, 437

19. Agolini, F., and Gay, F. P., Macromolecules (1970) 3, 349

20. Matejka, L., et al., Polym. Bull. (1979) 1, (9), 659

21. Rojanathanes, R., et al., Tetrahedron (2005) 61, 1317

22. Janicki, S. Z., and Schuster, G. B., J. Am. Chem. Soc. (1995) 117, 8524

23. Atassi, Y.,et al., Pure Appl. Chem. (1998) 70, 2157

24. Kobatake, S.,et al., Nature (2007) 446, 778

25. Zouheir, S.,et al., J. Opt. Soc. Am. B (2001) 18, 1854

26. Hugel, T., et al., Science (2002) 296, 1103

27. Sekkat, Z., and Knoll, W., Photoreactive Organic Thin Films. Elsevier/Academic Press, San Diego, (2002), 560

28. Warner, M., and Terentjev, E., Macromol. Symp. (2003) 200, 81

29. Yu, Y., and Ikeda, T., Macromol. Chem. Phys. (2005) 206, 1705

30. Yu, Y., and Ikeda, T., Angew, Chem. Int. Ed. (2006) 45, 5416

31. Barrett, C. J., et al., T. Soft Matter (2007) 3, 1249

32. Finkelmann, H., et al., Phys. Rev. Lett. (2001) 87, 015501

33. Hogan, P. M., et al., Phys. Rev. E (2002) 65, 041720

34. Ikeda, T., et al., Adv. Mater. (2003) 15, 201

35. Li, M.-H., et al., Adv. Mater. (2003) 15, 569

36. Ikeda, T., and Tsutsumi, O., Science (1995) 268, 1873

37. Yu, Y., et al., Nature (2003) 425, 145

38. van Oosten, C. L., et al., Eur. Phys. J. E (2007) 23, 329

39. Yang, Z., et al., J. Am. Chem. Soc. (2006) 128, 1074

40. Ogawa, M., et al., Adv. Mater. (2001) 13, 1107

41. Heinz, H., et al., (2008) unpublished data

42. We have chosen to distinguish azobenzene LC elastomers (LCE, Tg < actuation temperature) from azobenzene LC polymer networks (LCN, Tg > actuation temperature).

43. Tabiryan, N., et al., Opt. Express (2005) 13, 7442

44. White, T. J., et al., (2008), unpublished data

45. Cote, G. G., Plant Physiol. (1995) 109, 729

46. Nelson, E. C., and Braun, P. V., Science (2007) 318, 924

47. Lendlein, A., et al., Nature (2005) 434, 879

48. Vaia, R., and Wagner, H. D., Materials Today (2004) 7 (11), 32

49. Winey, K. I., and Vaia, R. A., MRS Bull. (2007) 32 (4), 314

50. Dosser, L., et al., Proc. SPIE (2004) 5339, 465

51. Gobin, A. M., et al., Nano Lett. (2007) 7, 1929

52. Koerner, H., et al., Nat. Mater. (2004) 3, 115

53. Ahir, S. V., and Terentjev, E. M., Nat. Mater. (2005) 4, 491

54. Yakobson, B. I., and Couchman, L. S., in DekkerEncyclopedia of Nanoscience and Nanotechnology, Schwarz, J. A., et al., (eds.) Taylor & Francis, New York, (2004), 587

55. Ahir, S. V., et al., Phys. Rev. B. (2007) 76, 165437/1

56. Govorov Alexander O., et al., Nanoscale Res Lett (2006) 1, 84

57. Schmidt, A. M., Macromol. Rapid Commun. (2006) 27, 1168

58. Sershen, S. R., et al., Adv. Mater. (2005) 17, 1366

59. Van der Veen, G., and Prins, W., Nature Phys. Sci. (1971) 230, 70

60. Takada, K., et al., Chem. Commun. (2006) 19, 2024

61. Zupan, M., et al., Adv. Eng. Mater. (2006) 4, 933