Viscoplasticity and Granular Behavior of Films of CdSe Nanocrystals by Nanoindentation

Raman Microprobe Analysis of Strain in Films of CdSe Nanocrystals

Fano Asymmetry in Raman Spectra of SrTiO₃ Nanocubes

Raman Scattering and Phases in Ce_xZr_1-xO₂ Nanocrystals

Raman Scattering and Phases in Hf_xZr_1-xO₂ Nanocrystals

The Herman group is using Raman scattering to examine how composition and nanocrystal size affect the phase of Hf_xZr_1-xO₂ nanocrystals. Raman spectroscopy demonstrates that 5 nm dimension Hf_xZr_1-xO₂ nanocrystals prepared by a nonhydrolytic sol-gel synthesis method are solid solutions of hafnia and zirconia, with no discernable segregation within the individual nanoparticles. (See the TEM.) Zirconia-rich particles (lower spectra) are tetragonal and ensembles of hafnia-rich particles (higher spectra) show mixed tetragonal/monoclinic phases. A simple lattice dynamics model with composition-averaged cation mass and scaled force constants is used to understand how the Raman mode frequencies vary with composition in the tetragonal Hf_xZr_1-xO₂ nanoparticles. These results are also shown. Raman scattering is also used to estimate composition and the relative fractions of tetragonal and monoclinic phases. Bulk hafnia and zirconia are in the monoclinic phase, and have very large dielectric constants (i.e. they are "high K" materials). The dielectric constants in the tetragonal phase are expected to be even larger. The use of Raman scattering was particularly important in this investigation, because the Zr-O and Hf-O bond lengths are approximately the same and so analysis by XRD is relatively difficult; Raman scattering easily distinguishes between zirconia and hafnia. This is a collaboration with the Brus/Steigerwald group at Columbia and Yimei Zhu at Brookhaven National Laboratory.

For more, see "Raman Scattering in Hf_xZr_1-xO₂ Nanoparticles" R. D. Robinson, J. Tang, M. L. Steigerwald, L. E. Brus,, and I. P. Herman, Phys. Rev. B 71, 115408 (2005).

Phonon Coupling and Decay in Cerium Oxide Nanocrystals

The frequency shift and linewidth of Raman scattered light is determined, in part, by how the scattered phonon couples to lower energy phonons. The faster a phonon decays into lower energy phonons, the broader the phonon linewidth. This affects the Raman spectrum at any temperature, and leads to more pronounced changes in the shift and the width with increasing temperature. Aside from annealing effects (the affects of which are seen outside the blue boxes), the figure shows that the changes in the Raman shift and linewidth with temperature in CeO_2-x nanocrystals (see below) does not depend on temperature. This means that, at least in these cerium oxide nanocrystals, phonon coupling and decay do not depend on particle size. are important in catalytic processes. These nanocrystals were synthesized by the Chan group.

Back to top

Raman Scattering in Cerium Oxide Nanocrystals

Cerium oxide nanocrystals are important in catalytic processes. (See the TEM.) The composition of cerium oxide is CeO_2-x; as the particle size gets smaller, the larger x becomes. With more oxygen defects, the catalytic properties improve. The Herman group has been examining these nanocrystals by Raman scattering, in collaboration with the Chan group. The displayed Raman spectra become broader and red shifted as the particles become smaller for two reasons. Phonon confinement becomes more important in smaller particles, and more of the bulk dispersion curve is sampled. Also, the lattice constant increases with smaller size because the larger Ce³⁺ ions replace the smaller Ce⁴⁺ ions as x increases. Any size inhomogeneity leads to a frequency down shift.

Real-time Monitoring of Organic Surface Ligands and Solvent During the Self-Assembly of Nanocrystal Arrays

The densities of surface ligands and solvent molecules are being followed during the self-assembly of CdSe nanocrystals into arrays by using multiple-reflection attenuated total internal reflection (ATIR) spectroscopy. This is performed in a Fourier transform infrared (FTIR) spectrometer in which the arrays formed on a ZnSe prism. During the self-assembly of CdSe nanocrystals passivated by pyridine that are dissolved in pyridine, the 1436.1 cm^-1 peak of neat pyridine is followed along with that at 1445.2 cm^-1 due to pyridine bound to the CdSe surface (below).

The solvent evaporates in about 30 minutes during the self-assembly of ~200-monolayer thick arrays in an argon ambient (inset below). The pyridine bound to the surface slowly leaves the surface (inset, open symbols), but about 35% remains after drying for several days (main figure, below). Since pyridine only weakly binds to the surface, it had been commonly thought that no pyridine would remain after extensive drying. While this is not true, there are still significant changes to the surface during drying. This result is important for the MRSEC studies of the self-assembly of arrays and the properties of these arrays. This monitoring method is also being used to follow the surface of exchange of TOPO and pyridine ligands, and related processes.

Vapor Phase Epitaxial Growth on Porous 6H-SiC Analyzed by Raman Scattering

SiC vapor phase epitaxy on porous silicon carbide formed by electrochemical anodization of 6H SiC produces high quality crystalline films that are shown to be of the 6H polytype by Raman scattering. This was a collaboration with Greg Dunne and Larry Rowland at Sterling Semiconductor, Inc.

Dielectric Function of Porous SiC Measured by Infrared Reflectance Spectroscopy

Porous SiC films were fabricated by electrochemical anodization of SiC wafers. Fourier transform infrared (FTIR) reflectance spectra (right) of these porous films reveal a rich structure in the reststrahlen region, which is very different from that of bulk SiC. The dielectric function of porous SiC depends on both the properties of bulk SiC and the details on the porous structure (below), and can be determined from the reflectance spectra.

The dielectric functions predicted from simple phenonemalogical models, such as the cavity Maxwell Garnett (C-MG) and the Landau-Lifshiftz/Looyenga (LLL) models, do not lead to good agreement with the porous SiC film reflectance. Hybrid models in which a medium described by one of these models in embedded in another do not work well either. However, using the general statistical treatment of dielectric functions, it is seen that a linear combination of the spectral density functions of the C-MG and LLL models leads to excellent agreement. This provides insight into the structure of this porous material.

For more, see "Use of Hybrid Phenomenological and Statistical Effective- Medium Theories of Dielectric Functions to Model the Infrared Reflectance of Porous Silicon Carbide," J. E. Spanier and I. P. Herman, Phys. Rev. B 61, 10437 (2000).