The Plasmoelectric Effect: A New Strategy for Converting Optical Energy into Electricity

Matthew Sheldon

Author: Matthew Sheldon

Affiliation: Department of Chemistry, Texas A&M University, USA.

Link to Sheldon Research Group >>

A plasmon resonance is a remarkable optical phenomenon that occurs in metallic nanostructures and other nanoscale materials that have high electrical conductivity. As reported in 'Science' on October 30 [1], we have demonstrated a new way to use plasmon resonances to generate electrical potentials during optical excitation. Our work could lead to new ways of converting optical energy into electrical energy, and may guide new opportunities in the very active research areas of plasmonics and nanophotonics.

A plasmon resonance results from the oscillations of electrons (or other electrical carriers) lining up with the oscillating electric field of incident radiation. This resonance causes significant concentration of light energy within the small sub-wavelength volume defined by the nanostructure. Because the resonant frequency can be tailored by controlling the nanoscale geometry, plasmonic materials have been the subject of considerable scientific activity for a host of applications that benefit from the ability to tune and concentrate radiation, such as Raman spectroscopy, cell labeling, sub-wavelength optical communication, or enhanced light trapping in solar cells, to list a few examples [2]. However, much of the confined optical energy is quickly absorbed by the metal and converted to heat. This heating is generally regarded as a limitation for optical applications.

Despite this loss of optical energy as heat, we questioned whether the strong plasmonic concentration of energy could still be utilized to perform electrical work. Electrical work can be understood as the movement of electrons through a circuit load, so it seemed natural to wonder if the plasmon resonance, which fundamentally results from the coupling of light to the motion of electrons, could also move electrons through a circuit. As reported in Science on October 30th [1], we discovered a mechanism by which optical absorption in plasmonic resonances indeed produces an electrical potential, a necessary first step towards performing useful electrical work. We have labeled this phenomenon the ‘plasmoelectric effect’. In conjunction with a thermodynamic model we developed, our analysis shows how a plasmonic resonance can act as a heat engine that uses thermal energy from the absorption of light to move electrons and produce static electric potentials.

Currently, the photovoltaic effect is the primary mechanism used in technology for the production of electrical potentials from the absorption of light, i.e. photo-voltages. The photovoltaic effect is the generation of excess electrical carriers in semiconductors during optical excitation with energy greater than the band gap energy. Our discovery, the plasmoelectric effect, is a fundamentally different mechanism for generating an electrical potential, and instead results from the dependence of the plasmon resonance frequency on electron density in conductors.

Recent works from other researchers studying plasmonic systems [3-5] have demonstrated that it is possible to tune the plasmon resonance frequency of a nanostructure by modulating electron density. Specifically, these researchers applied a static electric potential to inject or remove electrons from resonant structures, and they observed a shift to higher or lower frequency, respectively, of the plasmonic absorption resonance. In essence, the electrical state of the conductor, whether it is charged positively, negatively, or neutral, is coupled with the frequency of the plasmonic absorption. This behavior is analogous to how the resonant pitch of a musical instrument, such as a flute, would change if you modify the density of the air in the acoustic cavity.

Inspired by these experiments, we considered the extent to which the optical absorption, the plasmon resonance frequency, and the charge state are linked in this way, and if the reverse of this behavior would also occur. That is, can optical excitation with off-resonant light cause a change in the electron density of a plasmonic structure that shifts the plasmonic absorption into resonance with the illumination, and thereby induce an electrical potential? Considering the acoustic analogy above, this would be like the chamber of a flute adopting a slightly modified air density in order to become resonant with a loud pitch playing nearby that would otherwise be slightly out of tune.

To probe this possibility experimentally, we monitored the electric potential of a conductive surface coated with plasmonic Au nanoparticles using Kelvin probe force microscopy (KPFM). For KPFM a conductive atomic force microscope (AFM) tip is maintained a few nanometers above a sample surface, and the electrical potential between the tip and sample is measured. During KPFM experiments we also illuminated the nanoparticles with a tunable laser, varying the output from higher frequency to lower frequency through the plasmon resonance. We observed that higher frequency light caused negative surface potentials and that lower frequency light caused positive surface potentials, but there was no potential measured when the incident light was the same frequency as the plasmon resonance. This is the exact behavior expected if the nanoparticles are adjusting charge density so that the plasmon resonance is better matched with the frequency of the optical excitation.

Our report also details a thermodynamic model that anticipates this behavior for plasmonic materials. We show how the condition of minimum free energy, the preferred thermodynamic state of a system, corresponds to a configuration of charge density that modulates the plasmon resonance frequency in order to maximize the amount of heat produced via optical absorption. However, the energy required to electrically charge the structure moderates how much the plasmon resonance can shift. Therefore, for a given optical intensity, single frequency light induces a specific charge state that balances these counteracting effects. In general, during illumination a plasmonic structure will only remain neutral if incident light is the same frequency as the plasmon resonance of the neutral structure.

To show that the behavior is general to plasmonic systems, we also measured the optical response of periodic arrays of nanoscale holes in thin gold films that have strong, tunable plasmonic resonances across the visible spectrum based on the hole pitch. These fabricated hole arrays also displayed electrical potential trends consistent with our description of the plasmoelectric effect, as summarized in Fig. 1.

Figure 1: Plasmoelectric effect (a) Schematic of a metal nanoparticle that becomes electrically charged by illumination. (b) Electron microscopy image of the metal nanocircuit, composed of an array of nanoscale holes in a 20-nm-thin gold film. The scale bar is 500 nanometer. (c) Measured optical absorption spectra for metal nanocircuits with different spacings between the holes (175, 225, 250, and 300 nm). (d) Electrical potential of the nanocircuits in (c) as a function of wavelength of the incident light. The measured potentials range from -100 mV to +100 mV as the wavelength of the incident light is tuned from high frequency blue light to low frequency red light.

We believe our results are exciting for two fundamental reasons: First, we have demonstrated a new way to generate an electrical potential by the absorption of radiation. There is general interest in materials that can convert light to electrical potentials for sensing and for optical power conversion, for example, and our report lays the groundwork for these possible applications. Second, we believe our analysis provides more insight into the basic thermodynamic behavior of plasmonic materials. Given the very active research in this area by scientists from many different disciplines, these insights may open new opportunities in plasmonics research.

References:
[1] Matthew T. Sheldon, Jorik van de Groep, Ana M. Brown, Albert Polman, Harry A. Atwater, "Plasmoelectric potentials in metal nanostructures". Science, 346, 828–831 (2014). Abstract.
[2] Albert Polman, "Plasmonics Applied". Science, 322, 868–869 (2008). Abstract.
[3] Carolina Novo, Alison M. Funston, Ann K. Gooding, Paul Mulvaney, "Electrochemical Charging of Single Gold Nanorods". Journal of the American Chemical Society, 131, 14664–14666 (2009). Abstract.
[4] S. K. Dondapati, M. Ludemann, R. Müller, S. Schwieger, A. Schwemer, B. Händel, D. Kwiatkowski, M. Djiango, E. Runge, T. A. Klar, "Voltage-Induced Adsorbate Damping of Single Gold Nanorod Plasmons in Aqueous Solution". Nano Letters, 12, 1247–1252 (2012). Abstract.
[5] Guillermo Garcia, Raffaella Buonsanti, Evan L. Runnerstrom, Rueben J. Mendelsberg, Anna Llordes, Andre Anders, Thomas J. Richardson, Delia J. Milliron, "Dynamically Modulating the Surface Plasmon Resonance of Doped Semiconductor Nanocrystals". Nano Letters, 11(10), 4415–4420 (2011). Abstract.