Gold nanorods with different aspect ratios are prepared in micelles by the electrochemical method and their absorption spectra are modeled by theory. Experimentally, a linear relationship is found between the absorption maximum of the longitudinal plasmon resonance and the mean aspect ratio as determined from TEM. It is shown here that such a linear dependence is also predicted theoretically. However, calculations also show that the absorption maximum of the longitudinal plasmon resonance depends on the medium dielectric constant in a linear fashion for a fixed aspect ratio. Attempts to fit the calculations to the experimental values indicate that the medium dielectric constant has to vary with the aspect ratio in a nonlinear way. Chemically, this suggests that the structure of the micelle capping the gold nanorods is size dependent. Furthermore, comparison with the results obtained for rods of different aspect ratios made by systematic thermal decomposition of the long rods further suggests that the medium dielectric constant is also temperature dependent. This is attributed to thermal annealing of the structure of the micelles around the nanorods.

The relaxation dynamics of photoexcited CdSe nanoparticles (NPs) were studied with femtosecond pumpprobe spectroscopy in the spectral range from 450 nm to 5 µm. Thus, the intraband relaxation of the electron and the hole were monitored with femtosecond time resolution. The transition from delocalized electronic to localized defect states and the slower relaxation through the trapping sites (trap hopping mechanism) were followed by time-resolved emission on time scales from picoseconds to milliseconds in the spectral range from the visible into the mid-infrared. The spectral dynamics in the visible, near-infrared (NIR), and infrared (IR) range give a mechanistic picture about the relaxation pathway of the excited charge carriers in CdSe NPs. In addition, by using electron and hole quenchers to the photoexcited nanoparticles, we could assign the observed dynamics to the electron or the hole. In the visible range, bleach features (negative signals in the differential transient absorption measurements) were observed at early delay times with time components of 2 ps, 30 ps, and a long-lived component of ∼200 ps. Based on the quenching experiments, the bleach could be assigned to contributions from the electron and the hole. In the NIR and IR spectral range, the transient signal for short delay times (<1 ns) were positive (transient absorption). A decay time of 1.2 ps was found for the transient absorption observed in the NIR. With the carrier removal technique, it could be shown that the NIR signal is mainly due to hole transitions. The IR signal with lifetimes of 2 ps, 30 ps, and a long component of ∼200 ps compares directly to the decay times observed in the visible spectral range. It therefore is also composed of contributions from electron and hole. Photoluminescence is observed in the visible range at 570 nm (near band edge recombination), in the NIR at 1000 nm (deep-trap emission), and in the IR region at 4.8 µm (deeper-trap emission) with lifetimes of 43 ns, 250 ns, and 1 µs, respectively. The lifetimes of the emission increased for longer monitoring wavelengths, suggesting that for the investigated sample, the carriers relax from the band edge into shallow traps and from there continue stepwise relaxation into lower energy sites (trap hopping mechanism).