than within the plane of individual close-packed {111} faces, where the AuaAu
spacings are too small to facilitate epitaxy. As the nanorods grow in length, the area
of the side faces increases, and this could facilitate the assembly of a bilayer of
CTAB molecules at the crystal surface, in which the alkylammonium headgroup
points down toward the gold surface, with the long hydrocarbon chain facing the
solvent, interdigitating with a second hydrocarbon chain from a second CTAB
layer, with the second CTAB’s headgroup facing the solvent [46]. In turn, the bilayer would provide additional stabilization and growth inhibition, and this could
explain why elongation of the nanorods is rapid once the shape anisotropy has
been established, in a ‘‘zipping’’ type of mechanism.
Finally, we note that our proposed mechanism follows a classical description of
crystal growth inhibition that involves the preferential attachment of individual
molecules to the different crystal faces of the growing nanoparticle. The mechanism does not implicate the involvement of surfactant micelles per se in controlling the shape anisotropy of fcc metallic nanoparticles, as has been previously
postulated. Instead, the data indicate that symmetry breaking in fcc metallic structures is an intrinsic structural mechanism (twinning) that is subsequently modulated extrinsically during growth in solution by edge-specific surfactant adsorption. This general conclusion is also being reached by many other research groups
[1, 24, 40, 43–45].
9.3
Assembly of Metallic Nanorods: Self-Assembly vs. Designed Chemical Linkages
While great progress in the synthesis and mechanism of the growth of metallic
nanorods has been made, a real challenge in the field is the assembly of nanorods
into a functional structure. The researcher faces two general choices in the matter:
self-assembly vs. designed assembly. By self-assembly, we mean that the nanorods
can be ‘‘left to themselves’’ to order. For example, there is experimental evidence
that as one puts lateral pressure on a random ‘‘raft’’ of rods at the air–water interface, nanorods can assemble into ordered rafts [48]. Onsager long ago predicted
[49] that increasing the concentration of ‘‘hard colloidal’’ rods in solution would
lead to one-dimensional or two-dimensional liquid crystalline ordering, which has
been experimentally and theoretically verified [50–55].
By ‘‘designed’’ assembly, we mean that the different crystal faces of the nanorods
could be specifically reacted with some reagent to link them in a rational way. This
last approach is made difficult by the usual lack of information about the crystal
face identity on the ends and edges of nanorods, but progress is being made [1, 24,
40, 45].
In the case of liquid crystalline ordering as a function of nanorod concentration
in the self-assembly approach, we have performed TEM, polarizing microscopy,
and small angle X-ray scattering experiments to show that gold nanorods of high
aspect ratio (13–18) do indeed stack in regular arrays as the concentration increases [55]. The system is not as ideal as the Onsager case, the CTAB outer bilayer
9.3 Assembly of Metallic Nanorods: Self-Assembly vs. Designed Chemical Linkages 293
spacings are too small to facilitate epitaxy. As the nanorods grow in length, the area
of the side faces increases, and this could facilitate the assembly of a bilayer of
CTAB molecules at the crystal surface, in which the alkylammonium headgroup
points down toward the gold surface, with the long hydrocarbon chain facing the
solvent, interdigitating with a second hydrocarbon chain from a second CTAB
layer, with the second CTAB’s headgroup facing the solvent [46]. In turn, the bilayer would provide additional stabilization and growth inhibition, and this could
explain why elongation of the nanorods is rapid once the shape anisotropy has
been established, in a ‘‘zipping’’ type of mechanism.
Finally, we note that our proposed mechanism follows a classical description of
crystal growth inhibition that involves the preferential attachment of individual
molecules to the different crystal faces of the growing nanoparticle. The mechanism does not implicate the involvement of surfactant micelles per se in controlling the shape anisotropy of fcc metallic nanoparticles, as has been previously
postulated. Instead, the data indicate that symmetry breaking in fcc metallic structures is an intrinsic structural mechanism (twinning) that is subsequently modulated extrinsically during growth in solution by edge-specific surfactant adsorption. This general conclusion is also being reached by many other research groups
[1, 24, 40, 43–45].
9.3
Assembly of Metallic Nanorods: Self-Assembly vs. Designed Chemical Linkages
While great progress in the synthesis and mechanism of the growth of metallic
nanorods has been made, a real challenge in the field is the assembly of nanorods
into a functional structure. The researcher faces two general choices in the matter:
self-assembly vs. designed assembly. By self-assembly, we mean that the nanorods
can be ‘‘left to themselves’’ to order. For example, there is experimental evidence
that as one puts lateral pressure on a random ‘‘raft’’ of rods at the air–water interface, nanorods can assemble into ordered rafts [48]. Onsager long ago predicted
[49] that increasing the concentration of ‘‘hard colloidal’’ rods in solution would
lead to one-dimensional or two-dimensional liquid crystalline ordering, which has
been experimentally and theoretically verified [50–55].
By ‘‘designed’’ assembly, we mean that the different crystal faces of the nanorods
could be specifically reacted with some reagent to link them in a rational way. This
last approach is made difficult by the usual lack of information about the crystal
face identity on the ends and edges of nanorods, but progress is being made [1, 24,
40, 45].
In the case of liquid crystalline ordering as a function of nanorod concentration
in the self-assembly approach, we have performed TEM, polarizing microscopy,
and small angle X-ray scattering experiments to show that gold nanorods of high
aspect ratio (13–18) do indeed stack in regular arrays as the concentration increases [55]. The system is not as ideal as the Onsager case, the CTAB outer bilayer
9.3 Assembly of Metallic Nanorods: Self-Assembly vs. Designed Chemical Linkages 293