Chemical Interface Damping of Surface Plasmon Resonances.

ConspectusMetal nanoparticles have been utilized for a vast amount of plasmon enhanced spectroscopies and energy conversion devices. Their unique optical properties allow them to be used across the UV-vis-NIR spectrum tuned by their size, shape, and material. In addition to utility in enhanced spectroscopy and energy/charge transfer, the plasmon resonance of metal nanoparticles is sensitive to its surrounding environment in several ways. The local refractive index determines the resonance wavelength, but plasmon damping, as indicated by the homogeneous line width, also depends on the surface properties of the metal nanoparticles. Plasmon oscillations can decay through interband, intraband, radiation, and surface damping. While the first three damping mechanisms can be modeled based on bulk dielectric data using electromagnetic simulations, surface damping does not depend on the material properties of the nanoparticle alone but rather on the interface composition between the nanoparticle and its surrounding environment. In this Account, we will discuss three different metal nanoparticle interfaces, identifying the surface damping contribution from chemical interface damping and how it manifests itself in different interface types. On the way to uncovering the various damping contributions, we use three different single-particle spectroscopic techniques that are essential to measuring homogeneous plasmon line widths: darkfield scattering, photothermal heterodyne imaging, and photoluminescence microscopies. Obtaining the homogeneous plasmon spectrum through single-particle spectroscopy is paramount to measuring changes in plasmon damping, where even minor size and shape heterogeneities can completely obfuscate the broadening caused by surface damping. Using darkfield scattering spectroscopy, we first describe a model for chemical interface damping by expanding upon the surface damping contribution to the plasmon resonance line width to include additional influences due to adsorbed molecules. Based on the understanding of chemical interface damping as a surface damping mechanism, we then carefully compare how two molecular isomers lead to greatly different damping rates upon adsorption to gold nanorods due to differences in the formation of image dipoles within the metal nanoparticles. This plasmon damping dependence on the chemical identity of the interface is strongly correlated with the chemical's electronegativity. A similar damping trend is observed for metal oxide semiconductors, where the metal oxide with greater electron affinity leads to larger interface damping. However, in this case, the mechanism is different for the metal oxide interfaces, as damping occurs through charge transfer into interfacial states. Finally, the damping effect of catalytic metal nanoislands on gold nanorods is compared for the three spectroscopic methods mentioned. Through correlated single-particle scattering, absorption, and photoluminescence spectroscopy, the mechanism for metal-metal interface damping is determined most likely to arise from an enhanced absorption, although charge transfer cannot be ruled out. From this body of research, we conclude that chemical interface damping is a major component of the total damping rate of the plasmon resonance and critically depends on the chemical interface of the metallic nanoparticles. Plasmon damping occurs through distinct mechanisms that are important to differentiate when considering the purpose of the plasmonic nanoparticle: enhanced spectroscopy, energy conversion, or catalysis. It must also be noted that many of the mechanisms are currently indifferentiable, and thus, new single-particle spectroscopic methods are needed to further characterize the mechanisms underlying chemical interface damping.