INTERPRETING THE SPECTRAL SIGNATURES OF SURFACE-ENHANCED RAMAN SCATTERING

Open Access
Author:
Chulhai, Dhabih Vishaal
Graduate Program:
Chemistry
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
June 02, 2016
Committee Members:
  • Lasse Jensen, Dissertation Advisor
  • Lasse Jensen, Committee Chair
  • Miriam Arak Freedman, Committee Member
  • William George Noid, Committee Member
  • Adrianus C Van Duin, Outside Member
Keywords:
  • Spectroscopy
  • Surface Enhancement
  • Raman
  • SERS
  • Field Gradients
  • Plasmon
  • Nanoparticles
  • TERS
  • DFT
  • Subsystem DFT
  • Frozen Density Embedding
  • TDDFT
  • Density Functional Theory
Abstract:
Molecules may be uniquely identified through the inelastic light scattering process known as Raman scattering. The Raman scattered intensities are often weak, but may be enhanced by several orders of magnitude--through the process known as surface-enhanced Raman scattering (SERS)--by placing the molecules near the surface of metallic nanoparticles. Through SERS, we can detect the scattering from a single molecule, which means that the technique is useful for the ultra-sensitive detection of chemical and biological agents. However, the SERS signals are often very different from the signals of normal Raman scattering, as they now reflect the various interactions of the molecule(s) with the nanoparticle. Understanding these spectral changes are, therefore, vital in both identifying the probed molecule, and in understanding and extracting information of the molecule's interaction with the surface. To understand these changes, we have developed the dressed-tensors formalism that takes into account the interaction of the molecule with the inhomogeneous local electric fields from the nanoparticle. With this method, we show that the field gradient contribution to the spectral changes often reflect the relative orientation of the molecule with respect to the surface. This result, coupled with the dynamics of the probed molecule, suggests that the translational and rotational motions of a single molecule may be tracked through its SERS spectral changes. We have also extended this method to describe other types of surface-enhanced spectroscopies, namely Raman optical activity (ROA), which is sensitive to chiral structures and used to probe the behavior of biomolecules in solution, and circular dichroism (CD), which is often used to investigate the secondary structure of proteins. For surface-enhanced ROA (SEROA), we find that spectral changes are highly sensitive to the local electric field gradient, the orientation of the molecule, and the surface plasmon frequency width, giving insight into why mirror-image SEROA is yet to be observed for enantiomers. We also find that the spectral signatures of plasmonic CD are similarly complicated. However, this electromagnetic fields description of the enhancement is insufficient at describing the spectral changes for certain chemical systems. We find that, at low temperatures and for single or few molecules, the observed shift of particular normal modes may be reflective of the specific binding interactions of the molecule with the surface. In the case of resonant single molecule SERS of rhodamine-6G, we show that the relative intensity fluctuations are independent of the orientation of the molecule, but may rather describe picometer changes in its excited state geometry. These results indicate that we need a rigorous method to account for the quantum mechanical interactions between the molecule and the surface. To this end, we have developed an exact subsystem density functional theory (DFT) method that can exactly reproduce the supermolecular energies and densities of a wide range of systems, including covalently bonded subsystems. We have also extended this method to the time-dependent DFT regime, and show that we can accurately reproduce supermolecular excitation energies of strongly coupled subsystems. The spectral changes observed in SERS contain a lot of information of the molecule-nanoparticle interactions, and the methods developed here have allowed---and will continue to allow---us to interpret these changes.