We use optical spectroscopy to probe the properties of nanostructured materials. In a basic photoluminescence experiment, a laser provides photons with energy larger than the band gap of a semiconductor. These photons excite an electron from the valence band to the conduction band, leaving behind a hole. The electron and hole relax down the lowest energy levels available before recombining to emit a photon. We collect those emitted photons and measure their energy with a spectrometer and CCD to learn about the energy levels and properties of the material under investigation.

Quantum dots, both epitaxially grown and colloidally synthesized, are exactly like the quantum mechanical model of a particle in a box, where the size of the box determines the confined energy levels. In quantum dots the confinement is provided by a sandwich of two different semiconductors that have different bandgaps, for example InAs completely surrounded by GaAs. Because the potential energy for electrons in the InAs is lower than the potential energy of electrons in the GaAs, electrons in the InAs are confined in all directions. This confinement leads to discrete energy levels and the particular energy of the states is determined by the size of the InAs region. Colloidally synthesized dots use an analogous quantum confinement method. The size of quantum dots can be precisely controlled by growth techniques, which is why quantum dots have found numerous applications as luminescent markers in biology and chemistry experiments.

Unfortunately, “precisely controlled” means that we can tune the center of the energy distribution, but individual quantum dots have energies that are inhomogeneously distributed over a relatively broad range of energies. The distribution of energy levels in an ensemble of quantum dots is typically of order 50 meV. This is much larger than the perturbations that are introduced by charge interactions (about 1 meV) or spin interactions (about 0.1 meV), so measurements of ensembles make it very hard to figure out how the nanoscale structure is influencing the properties. One of our most important experimental techniques is to isolate and measure a single quantum dot or isolated pair of quantum dots. When looking at a single quantum dot, the number of emitted photons is extremely small, so we work hard to design and optimize experimental systems that have very high collection efficiency. We are interested in probing how nanoscale structure controls the properties of individual quantum dots, and especially how it controls the interactions between different quantum dots. In our quantum dot molecules, for example, two quantum dots are grown stacked on top of one another, separated by a thin tunneling barrier. Coherent tunneling between the two quantum dots leads to the formation of delocalized molecular-like states, and we can control the degree of coupling in situ with applied electric fields. We use our spectroscopic measurements of the energy levels of the paired quantum dots to learn about the formation of these delocalized states and the spin interactions they control.

To access dynamical properties of materials we use time-correlated single-photon counting to measure time-resolved photoluminescence. Essentially this technique uses a very precise clock to measure the time delay between the excitation of the sample by a pulsed laser and the emission of a photon. We have developed methods to measure the photoluminescence decay of emission from single quantum dots as a function of changes in their local environment, which allows us to probe how nanostructure impacts dynamical processes like energy relaxation. This method is limited to temporal precision of about 40 ps. We have extended this technique to allow us to measure the lifetime of emission from subsets of an ensemble by placing our photon detectors (avalanche photodiodes) after our spectrometer. This approach allows us to measure energy transfer within nanostructured ensembles. We are currently developing a new transient absorption instrument that will allow us to measure charge and spin dynamics on sub-ps timescales. In this approach, a pump pulse will excite electrons or holes into defined energy levels of a nanostructured material. A time-delayed broadband probe pulse will have absorption that depends on where the electrons or holes end up. By varying the delay time between the pump and probe with a mechanical delay line we can access sub-ps dynamics.