The electromechanical properties of carbon nanotubes (CNTs) and graphene have proven to be both unique and exciting. The feasibility of employing these materials as commercial micro/nano- electromechanical actuators is highlighted by their exceptional volumetric work capacities, which can be as much as 29 times greater than common high modulus ferroelectrics in experiments [1].
Developing such an actuation material requires an in depth knowledge of the physics of operation and, therefrom, how to best optimize its performance. Using ab initio simulation method, we investigate the electromechanical actuation of pristine monolayer graphene to elucidate the origin of this material’s exceptional electromechanical actuation performance [2]. It is shown that the electrostatic double-layer (EDL) effect is dominant compared to the quantum-mechanical (QM) effect upon charging and electrolyte immersion. However, the EDL formation rate is hindered by the ionic diffusion rates, which significantly limit the actuation speeds and render these materials useful for <1 Hz operating frequencies only. In contrast, quantum-mechanical (QM) actuation, comprising bond length expansion/contraction upon electron/hole injection, does not suffer such problem. In light of this, the ideal CNT/graphene electromechanical actuator would combine the high work capacities of the EDL effect with the high response rate of the QM effect, without the inherent drawbacks associated with each effect.
With this motivation, we have investigated graphene oxide (GO) as a potential actuation material via ab initio simulations [3]. GO has various concentrations and configurations with which oxygen atoms can be adsorbed onto the pristine graphene lattice, making it possible to tune the electromechanical response of GO, via designing its structure, to precisely suit a given application. We found that the electromechanical responses of some GO compounds are very interesting and pronounced. The non-reversible and reversible QM electromechanical strain (at 0.15 e/C-atom electron/hole injection) can reach 28% and 6%, respectively, which are about 100 times higher than CNT bucky paper and comparable to the electrostrictive-polymers and mammalian skeletal muscle. The volumetric work densities in excess of 50 J/cm^3 have been predicted, which is about two order of magnitude higher than the electrostrictive-polymer and skeletal muscle. The mechanism of high performance of GO relies on its intrinsic rippling structures. With the electron/hole injection, the excess charge will tune the interactions between the Pi orbitals of C atoms and the sp2 orbitals of the O atoms, resulting in a modulation of the degree of rippling and thus an in-plane electromechanical deformation.
Having demonstrated huge strain output (about 6-28%), stress output (on the order of 100 GPa), and high volumetric work density (as high as 50J/cm^3), graphene and GO- based materials are uniquely positioned to address future demand on micro/nano- actuation materials.

Reference:
1. R. Baughman, et.al., Science 1999, 284, 1340.
2. G. Rogers and J. Z. Liu, J. Am. Chem. Soc. 2011, 133, 10858.
3. G. Rogers and J. Z. Liu, J. Am. Chem. Soc, 2012, 134, 1250.