HRTEM crystal growth

We determined the size of 200 crystals in each sample in order to estimate the average crystal size (Fig. 12.18b). While 2-h oxidation in air results in an average crystal size of 7 nm, it is upshifted to 14 nm after 42-h oxidation. Since crystal growth is not expected under such conditions, large crystals (15-30 nm) observed after oxidation must be present in the pristine powder. However, because they are usually covered by smaller crystals and remain within agglomerates, it is difficult to observe them in HRTEM, indicating the limitations of this characterization technique and the low statistical reliability. 

HRTEM was used to investigate crystal growth in the tooth cusps. The crystals were deposited in the form of thin fibrous strands (15-20 nm in width) adjacent to underlying organic filaments (Fig. 19). Crys-... 

It is noteworthy that the HRTEM cannot distinguish core and shell even by combining X-ray or electron diffraction techniques for some small nanoparticles. If the shell epitaxially grows on the core in the case of two kinds of metals with same crystal type and little difference of lattice constant, the precise structure of the bimetallic nanoparticles cannot be well characterized by the present technique. Hodak et al. [153] investigated Au-core/Ag-shell or Ag-core/Au-shell bimetallic nanoparticles. They confirmed that Au shell forms on Ag core by the epitaxial growth. In the TEM observations, the core/shell structures of Ag/Au nanoparticles are not clear even in the HRTEM images in this case (Figure 7). 

Fig. 3 (A) Schematic of the LCG model. (B) TEM image of Si nanowires produced via LCG. (From Ref. (C) SEM image of GaP nanowires, single seed particle composed mostly of Au is shown in the inset. (D) HRTEM image of GaAso.6Po.4 nanowire with (111) growth axis. (E) CdSe nanowire with 18 nm diameter (wurtzite crystal structure is indicated in the inset). (From Ref. l) (View this art in color at www.dekker.com.)...

Obtaining direct experimental data on nucleation processes is very difficult due to the extremely small fraction of the total mass involved and the problems of distinguishing this from surrounding reactants of very similar nature and composition. Consequently, much of the information available is more that of early growth behaviour rather than of nucleation itself. Recent cryo-TEM [78] and HRTEM studies [79 81] represent the most direct evidence of nuclcation phenomena currently available. However, much useful data can be obtained from circumstantial evidence , i.c. measurements of the appearance, growth and size distribution of the crystals resulting from the earlier, hidden patterns of nucleation. 

Defects of other types at scales quite different from that of the unit cell can also be important. The surfaee stmeture represents both the crystal face during growth and the first structure presented to adsorbing molecules, so that surface study, for example by AFM and HRTEM, can give important clues to these processes. Furthermore, the structure of mesopores generated by post-synthetic treatment, or the size and dispersion of metal particles in bifunctional catalysts are both examples of non-periodic features important in catalysis, so that recently developed 3D TEM methods to study them have particular significance. 

---