Last wee was Thanksgiving break, and I enjoyed a week off from teaching.  One of the things I had time to do is sit down and read an article that was published in the Journal of Physical Chemistry C last March, “A Unified Approach to Surface-Enhanced Raman Spectroscopy” (Lombardi and Birke, Volume 112, pgs 5605-5617, DOI 10.1021/jp800167v).

Lombardi and Birke’s article pulls together the many different explanations for the large Raman enhancements observed for molecules adsorbed to coinage metal surfaces with atomic scale roughness features – surface-Enhanced Raman scattering (SERS).  They argue that it is important to consider all of the resonances in such systems, which include

  1. Surface plasmon resonances
  2. Molecular state resonances
  3. Charge transfer resonances

After laying out the theoretical expressions that govern the magnitued of the polarizability tensor in these systems, they run through a series of examples that illustrate the theory.  The theory is consistent with SERS of molecules on any type of nanoparticle or electrode surface, and for any type of excitation situation (like where the laser frequency is far from molecular and charge transfer resonances, to where the laser’s frequency is resonant).  If you’re interested in learning more, read on as I summarize.  Even better, get a copy of the original article and read it!


Surface Plasmon Resonances

The first ingredient for SERS is the surface plasmon.  Without it, no SERS.  A surface plasmon resonance (SPR) can occur when a metal’s dielectric constant has a real positive component (Re[ε] > 0), and an imaginary component that is close to zero (Im[ε] ≈0).  Additionally, the metal needs to be smaller than the wavelength of light that is impinging upon it:

spr

The resonance that occurs corresponds to a collective oscillation of the electrons in the metal’s conduction band, which in turn leads to a large local electromagnetic field.  Coinage metals (Ag, Au and Cu) have been widely used in SERS because they can be made to meet these conditions in the visible region of the spectrum.  The magnitude of the local electric field depends on

  1. nanoparticle size[1].
  2. nanoparticle shape (the field is larger at the tips of non-spherical nanoparticles).
  3. nanoparticle aggregation (two or more in resonance together…).

Molecular and Charge Transfer Resonances

The rest of the SERS story is embedded in molecular and charge transfer resonances.  The important energy levels for this discussion are shown in the diagram below.  This diagram shows the metal (it’s conduction band and the Fermi level), and the ground and excited states of the molecule, as well as some vibrational states in the ground electronic state.

eld

The following transitions may occur:

  1. Vibronically allowed transitions between the ground and excited state of the molecule (assuming the laser frequency matches the resonance)
  2. Molecule – metal CT via Herzberg-Teller coupling
  3. Metal – molecule CT via Herzberg-Teller coupling

Now, other than the fact that in SERS experiments the laser is not usually resonant with a vibronic transition, the first item in the list shouldn’t really bother anyone who has had general chemistry because it looks just like an abosroption.  The second two terms will seem unfamiliar to anyone who hasn’t had a graduate course in molecular spectroscopy though.

Briefly, Herzberg-Teller coupling is a way for a symmetry forbidden transition to gain intensity.  For SERS, this means that vibrations that have the same symmetry as an excited electronic state of the molecule will be the ones that show up in the spectrum.  Ok, maybe it’s more complicated than that, but I think maybe another post might be warrented just on Herzberg-Teller coupling.  Which of course, would be cool.

[1] I’m using nanoparticles for any atomically rough surface.  This would still work for roughened electrodes, but I don’t use them, so to me they’re just nanoparticles.

Possibly related posts: (automatically generated)