Joseph M. Heer

Propagation lengths of surface plasmon polaritons on metal films with arrays of subwavelength holes by infrared imaging spectroscopy

Metal films with arrays of subwavelength holes (mesh) exhibit extraordinary transmission resonances to which many attribute a role for surface plasmon polaritons (SPPs); others debated this point. Experimental measurements of propagation lengths are presented under conditions that pertain to the use of SPPs for surface spectroscopy. The lateral extent of electromagnetic propagation along the mesh surface is measured by recording absorption spectra of a line of latex microspheres as a function of distance away from the line along the mesh. Measurements reveal an exponential functional form for decay of absorption signal laterally from the absorption source. Results at 697 cm(-1), which are closest to the strongest transmission resonance of the mesh, reveal a 1/e propagation distance along the surface of 17.8+/-2.9 microm. This is 40% larger than the lattice spacing implicating the holes as the SPP damping mechanism, however, this is significantly shorter than smooth metal expectations.

Interaction of an infrared surface plasmon with an excited molecular vibration

The interaction of an infrared surface plasmon and an excited molecular vibration was investigated by using a square array of subwavelength holes in a Ni film which supports propagating, surface-plasmon-mediated, transmission resonances. The largest transmission resonance [the (1,0)(-)] was tuned through the rocking vibration of the hexadecane molecule (at 721 cm(-1)) in a hexadecane film on the mesh by varying the thickness of the film. The interaction of the rocking vibration and surface plasmon is characterized spectroscopically by an increase in the intensity of the vibrational band by more than a factor of 2, variation of the vibrational line shape relative to the spectrum on a nonmetallic surface, and shifts in vibrational peak position by as much as 3.0 cm(-1). Relationships are developed between the transmission resonance position and the thickness and dielectric properties of the coating.

Carbon chains and the (5,5) single-walled nanotube: Structure and energetics versus length

Reliable thermochemistry is computed for infinite stretches of pure-carbon materials including acetylenic and cumulenic carbon chains, graphene sheet, and single-walled carbon nanotubes (SWCNTs) by connection to the properties of finite size molecules that grow into the infinitely long systems. Using ab initio G3 theory, the infinite cumulenic chain (:C[double bond]C[double bond]C[double bond]C:) is found to be 1.9+/-0.4 kcal/mol per carbon less stable in free energy at room temperature than the acetylenic chain (.C[triple bond]C-C[triple bond]C.) which is 24.0 kcal/mol less stable than graphite. The difference between carbon-carbon triple, double, and single bond lengths (1.257, 1.279, and 1.333 A, respectively) in infinite chains is evident but much less than with small hydrocarbon molecules. These results are used to evaluate the efficacy of similar calculations with the less rigorous PM3 semiempirical method on the (5,5) SWCNT, which is too large to be studied with high-level ab initio methods. The equilibrium electronic energy change for C(g)-->C[infinite (5,5) SWCNT] is -166.7 kcal/mol, while the corresponding free energy change at room temperature is -153.3 kcal/mol (6.7 kcal/mol less stable than graphite). A threefold alternation (6.866, 6.866, and 6.823 A) in the ring diameter of the equilibrium structure of infinitely long (5,5) SWCNT is apparent, although the stability of this structure over the constant diameter structure is small compared to the zero point energy of the nanotube. In general, different (n,m) SWCNTs have different infinite tube energetics, as well as very different energetic trends that vary significantly with length, diameter, and capping.

3D-FDTD Modeling of Angular Spread for the Extraordinary Transmission Spectra of Metal Films with Arrays of Subwavelength Holes

Extraordinary transmission through a metal film with an array of subwavelength holes (mesh) can be modeled effectively using three-dimensional finite difference time domain (3D-FDTD) calculations. A simple 3D-FDTD model, where a plane wave of light at perpendicular incidence with no angular spread interacts with a periodic repeating mesh, models the shape and location of the transmission resonance peaks qualitatively but not quantitatively. The simple 3D-FDTD model gives peaks that are tall, sharp, and red-shifted compared to experimental measurements using a benchtop Fourier transform infrared (FTIR) spectrometer. It was discovered that the simple model does not account accurately for diffraction order scattering by the mesh or the angular spread of the actual FTIR beam. This work describes a more sophisticated model that accounts for these factors and agrees quantitatively with experimental FTIR results.