233102-2
Sathe, Balan, and Pillai
Appl. Phys. Lett. 96, 233102 ͑2010͒
FIG. 3. ͑Color online͒ ͑a͒ Superimposed XRD pattern for Rh nanospheres
collected at different stages between 5 to 80 min and ͑b͒ variation in the
comparative intensity ratios of ͑311͒: ͑111͒ peaks with time.
FIG. 2. ͑Color online͒ ͑a͒ Superimposed in situ UV-visible spectra at dif-
ferent stages of the reaction indicating response at 375 nm and 474 nm
corresponding to truncated and smooth surfaces, respectively, and ͑b͒ varia-
tion in the intensity of peak corresponding to 375 nm as a function of time.
3 could be indexed to ͑111͒, ͑200͒, ͑220͒, and ͑311͒ planes of
the data obtained by energy dispersive x-ray ͑EDX͒ spectral
analysis shown in Ref. 21 ͑Fig. II͒.
spheres. However, after 80 min the absorption spectrum be-
comes relatively featureless, revealing a broad and low
absorbance over the visible region, which is also consistent
with the black color of the aggregates.
The above observation indicates a general model of
growth, whereby the initially fast reduction in Rh ions favor
epitaxial growth into a thermodynamically favorable ͑111͒
low energy crystal face. Interestingly, the rate of growth ini-
tially in ͑311͒ plane is indeed less, although there is an in-
crease with time, indicating the competition between thermo-
dynamics and kinetic factors. For example, typically at 5
min, the reaction is dominated by nonequilibrium conditions
so that the growth of nanoparticles occurs at the thermody-
namically favorable low energy ͑111͒ plane. With time, how-
ever, these Rh nanospheres become truncated, shifting to a
branched surface above 40 min and finally to nanorods via
accelerated additional growth along the ͑311͒ direction.
Crystallite size of these particles is about 2.5 nm as inferred
from the full width and half maximum ͑FWHM͒ of the
43.54° peak ͑2.252 Å͒, which is in agreement ͑2.9Ϯ0.4 nm͒
with the corresponding size from TEM images ͑Ref. 21, Fig.
I͒. Alternatively, nanorods growth can be explained by a
mechanism of oriented attachment, where supersaturation of
Rh ions on initially nucleated Rh causes steric hindrance
between attached aggregates resulting in the formation of a
order driven by the rate of adsorption at ͑311͒ versus ͑111͒
face which is kinetically controlled under these non-
equilibrium conditions.
In summary, we have observed optical features for Rh
nanospheres obtained by a galvanic displacement with Al in
agreement with fascinating morphological variation. The
emergence of surface plasmon peaks at 375 nm and 474 nm,
respectively, is ascribed to truncated and smooth surface of
nanospheres in contrast to the absence of surface plasmon for
bulk Rh͑0͒ in the visible range. Interestingly, once the nucle-
ating sites are formed further growth of the nanostructures
with time is controlled by kinetics instead of thermodynam-
ics of site-specific adatom incorporation at ͑311͒ versus ͑111͒
faces.
We have confirmed the origin of these two peaks arising
from Rh morphological variation by carrying out several
separate experiments. Although spherical nanoparticles of Rh
are not expected to reveal any SPR response, we observe a
strong response ascribed to the difference in the surface
roughness of these nanostructures ͑truncated and smooth͒.
The origin of the absorption peak at 375 nm is due to smooth
Rh nanostructures as confirmed by several other reports. For
example, Xia et al. observed SPR signal at ϳ380 nm for Rh
multipods, along with
a blue shift due to surface
anisotropy.12 Similarly a broad absorption at 500 nm was
seen for Rh nanotubes and hence our both peaks could arise
due to Rh͑0͒ despite possible changes due to anomalous
terface show an anomalous behavior around its intrinsic
resonance wavelength. For example, McLellan et al.17 have
reported an analogous SPR response of sharp and truncated
Ag nanocubes ranging from 60–100 nm having a blueshift in
SPR bands along with the possibility of in situ sharpening of
their edges which gives excellent support to our results. In
fact, just like the charge transfer between the halide and gold
ion can produce a well defined absorption peak in the UV-
visible range,18 Kundu et al. have reported strong bands at
374 and 470 nm for Rh due to charge transfer from ligand to
metal in the mixed solution of cetyl trimethylammonium
bromide ͑CTAB͒ and Rh ions.19 However, this is not the
origin of our peaks as independent experiments carried out
by using externally added chloride ions ͑HCl and RhCl3͒
have not generated any enhancement in intensity or change
in the absorption spectra.
In order to investigate the structural evolution with time,
we carried out time dependent x-ray diffraction ͑XRD͒ stud-
ies focusing on subtle changes in the organization. The
growth of nanostructures should at least in principle, start
from the lowest energy sites leading to a progressive dis-
placement from the highest energy sites as a function of
time. Accordingly, the entire diffraction profile shown in Fig.
The authors, V.K.P., B.R.S., and B.K.B. would like to
thank the Council of Scientific and Industrial Research
͑CSIR͒, New Delhi for financial support.
136.165.238.131 On: Fri, 19 Dec 2014 09:53:31