Optical Properties of Colloidal Quantum Structures

with Min Xiao

As nanotechnology moves to the forefront of scientific research, both inquiry and understanding are growing at an alarming rate. Of particular interest are semiconductor nanocrystals (NCs) in the form of quantum dots (QDs) and quantum rods (QRs). These colloidal structures, crystalline in composition, are generally synthesized such that the structure size is less than the exciton Bohr radius of the bulk semiconductor material, inducing quantum confinement. The properties of QDs and QRs are becoming more and more understood, controlled, and as a result, applicable. The many applications for which quantum nanostructures are showing promise include light-emitting diodes, solar cells, and biological labeling . Before they can be practically applied in modern technology, however, it is necessary to produce both stable, efficient dots and to have a firm understanding of their optical properties.

The two types of II-VI semiconductor materials investigated here are CdSe and CdTe. Though the two exhibit some similar characteristics, a lot more is known about the former. Inquiry into CdTe on the other hand is, for the most part, in the beginning stages. While reliable methods of synthesis have been and are being developed (Peng, et al., unpublished), obtaining relative stability seems to be the biggest obstacle in the exploration of these nanocrystals.

One of the benefits of CdSe is, as mentioned, that so much knowledge regarding its properties and behavior has been obtained in recent years. It is now possible to reliably (and safely ) control the size and shape of quantum structures in synthesis and to ensure that within any given sample, the structure size distribution is relatively narrow.1 The quantum yield of CdSe NCs, a measure of the efficiency of light absorption and emission, has also improved dramatically with new synthetic methods. All of this has opened the door for optical physics; the fluorescence, photoluminescence,3, and polarization properties associated with these structures can be reasonably explored. And as synthesis is more controlled, it becomes possible to tune the band gap of semiconducting materials such that these optical properties can be adjusted for a variety of purposes.

One curious characteristic of quantum structure (QS) fluorescence is that the emitted light varies in intensity, producing a "blinking" effect. According to Nirmal, et al., the nature of this blinking varies both with individual QSs and the intensity of the excitation light. The currently accepted theory is that such blinking arises due to the ionization of the crystallite. As of yet there does not seem to be a way to quantify and control the oscillations for individual dots -- whether they are predictably periodic or not is still a matter to be investigated. Blinking does not produce a significant problem while measuring optical properties of a large, concentrated sample of NCs, as the individual dots average each other out to provide fairly constant readings. In observing the optical behavior of individual QSs, however, it is necessary to take certain steps to accommodate this information. Such methods will be elaborated upon in the description of experimental methods.

A fundamental difference between quantum dots and rods is the presence of a large dipole moment in structures with aspect ratios (long axis : short axis) larger than 2. Crystalline CdSe is an anisotropic material,5 and as QSs progress from dots to rods during synthesis, the growth is primarily along the c-axis of the wurtzite structure. Measurements of dielectric dispersion and quantum-confined Stark effect provide evidence for a permanent dipole present in longer QRs. Theory and experimental data suggest that this dipole is aligned with the c-axis. Chen, et al. have presented evidence that an optically-excited QR of aspect ratio greater than 2 will emit light that is linearly polarized, as opposed to the plane polarized light emitted by more spherical QSs.5

This study proposes to investigate optical properties of colloidal quantum structures composed of both CdSe and CdTe. In collaboration with Dr. X. Peng in our chemistry department, samples of colloidal QRs, varying in aspect ratio from ~2 to 4, will be obtained and analyzed in an effort to determine the relationship between the degree of polarization of emitted light and the relative rod length. Changes in the degree of polarization of the emitted fluorescence from the QRs will be studied as a function of aspect ratio.

Images of the CdSe samples will also be taken, at room temperature, using far-field microscopy. A diagram of the experimental setup can be seen in Fig. 3. Data will be collected as described in REF 17. The 514.5-nm line of a linearly-polarized argon-ion laser is used as an excitation source. The laser beam was reflected by a 5% beam splitter and directed through a high-magnification microscope objective (50X, NA = 0.55) onto the sample. Luminescence is collected by the same objective and directly projected into the spectrometer, and then detected by a liquid-nitrogen-cooled charge-coupled device (CCD) camera. Scattered laser light is blocked by a notch filter. This enables us to obtain luminescent images of single quantum rods by using a flat mirror in place of the grating inside the spectrometer. All the images and imaging spectra will be taken using an excitation intensity of about 80 W/cm2 .5

Figure 3. This schematic diagram depicts the basic setup upon which far-field microscopy of CdSe quantum dots was carried out. The fused silica disk was placed at the position marked "S".

To determine the polarization of a given sample, a polarizer can be inserted between the collecting objective and the spectrometer. Images of each spot will be taken with the polarizer in 30° increments from 0° to 180°. Individual dots can thus be traced and their intensity noted in each of the individual images.

This project will give the REU student expertise with many cutting-edge optical techniques as well as standard data analysis methods. Furthermore, the system can be set up with only slight modifications to existing experiments making the project very doable in a 10-week time frame.