Energy States of Ligand Capped Ag Nanoparticles: Relating Surface Plasmon Resonance to Work Function Anup L. Dadlani, † Peter Schindler, ‡ Manca Logar, ‡ Steve P. Walch, ‡ and Fritz B. Prinz* ,‡,§ † Department of Chemistry, ‡ Department of Mechanical Engineering, and § Department of Materials Science and Engineering, Stanford University, Stanford, California 94305, United States * S Supporting Information ABSTRACT: The work function (WF) and surface plasmon resonance (SPR) of organic ligand capped Ag nanoparticles (NPs) have been studied experimentally and computationally. Experimental observations reveal a significant increase in WF as the size of ligand-capped Ag NPs increases, a trend contrary to that previously observed for bare Ag NPs. Computational results confirm the effect on the WF from simplified ligand molecules and relate it to charge transfer between the Ag core and surrounding ligands. We also observe a possible coupling between increases in WF and decreases in SPR transition energy, supported by computational results and attributed to the interplay between the 4d and 5s electron states of the system. These results, along with our observations of WF dependence on ligand choice, indicate the ability to strongly engineer the electronic structure of metal NPs through size and ligand control. ■ INTRODUCTION Metal nanoparticles (NPs) have recently attracted significant attention due to their promise for a growing number of applications, including biomedicine, 1 catalysis, 2 optical sen- sors, 3,4 and surface-enhanced Raman spectroscopy. 5 However, better understanding and control of the electron energy structure of these particles are needed to realize this promise. For example, in electrical devices correct band structure is essential for efficient charge transport, and in electro-catalysis the ease of charge extraction from surfaces is a major determiner of reaction rates. These properties are reflected in the work function (WF) of the material system employed. In many frontier applications, such as LEDs, PVs, and FETs, the NPs used are typically capped by ligands, which modify the electron energy levels at the metal-organic interface. 6,7 Better understanding of the effect of capping ligands is needed in order to exploit the benefits offered by NPs. Additionally, the burgeoning field of plasmonics depends on the nature of the surface plasmon resonances (SPRs) supported by NPs. SPRs stem from the physical interaction between light and metal NPs, which induces a collective oscillation of the conduction electrons due to the presence of an electromagnetic field. 8-10 This interaction is strongly affected by the presence of capping ligands, and much remains to be learned about the nature of their influence. In this paper, we report observations of trends in the WF and SPR of Ag NPs capped by different ligands as a function of NP size and ligand type. WF measurements were taken using ultraviolet photoelectron spectroscopy (UPS), and SPR measurements were taken using ultraviolet-visible spectrosco- py (UV-vis) and electron energy loss spectroscopy (EELS) by scanning transmission electron microscopy (STEM). Addition- ally, we present the results of a simplified ab initio quantum model of the systems under study, gleaning insight into the trends observed. Our experimental and simulated results show consistent and correlated trends in the WF and SPR of Ag NPs. ■ EXPERIMENTAL METHODS WF and SPR Measurements. WF and SPR measurements were carried out on 5 mg/mL (Econix) Ag NPs capped with polyvinylpyrrolidone (PVP) dispersed in water, procured from NanoComposix. The core diameters were (5.5 ± 1.6) nm, (23.1 ± 6.9) nm, (54.8 ± 10.1) nm, and (74.5 ± 11.8) nm (see Figure S1 in the Supporting Information for size distributions). For ligand comparison, a Sigma-Aldrich sample of (5 ± 2) nm dodecanethiol (DDT)-capped Ag NPs dispersed in hexane was also measured. WFs were measured by UPS with a model AC-2 Photo- electron Spectrometer at atmospheric pressure, using a deuterium UV source in air. UPS samples were drop-casted via pipet on quartz wafers cleaned with acetone, methanol, and isopropyl alcohol. Measurements were repeated over at least three different areas for each sample. Fowler’s hypothesis for metals was used in determining the WF of the NPs. Straight lines were fitted to plots of quantum yield to the half power Received: July 22, 2014 Revised: September 19, 2014 Published: September 29, 2014 Article pubs.acs.org/JPCC © 2014 American Chemical Society 24827 dx.doi.org/10.1021/jp5073044 | J. Phys. Chem. C 2014, 118, 24827-24832