182
S. Tang et al. / Chemical Physics Letters 477 (2009) 179–183
measured to be 0.23 nm, which indicates (1 1 1) interplanar spac-
ing. Another set of fringes in the branch displays the (1 0 0) inter-
planar spacing of 0.20 nm, but the thereby inferred h1 0 0i
direction is not quite parallel to the branch. This is expected, be-
cause the branching angles are 53–61°, so they are not exactly
equal but quite close to the angle of 54.7° between the h1 0 0i
and the h1 1 1i direction perpendicular to the (1 1 1) twin plane
(Fig. 4b). Such angles indicate that the growth is globally diffu-
sion-controlled, but locally it results from oriented attachment,
which has been found to be responsible for the formation disloca-
tions during early crystal growth [24]. Nano-particle aggregation
by oriented attachment of nano-crystallites, has been recognized
as an important crystal growth mechanism for dentritic morphol-
ogy in solution [7,8,25–28]. The formation of dislocations is often a
direct consequence of oriented attachment of a nano-particle that
attaches with a small misorientation or atomic nonflatness at the
bonding interface, and so marks the occurrence of the initial parti-
cle–particle bonding [8,24] that initiates a new branch. The inset
again magnifies the marked square just above in order to point
out another edge dislocation (defined as having a Burgers vector
normal to the dislocation line) found within the connecting region.
Dislocations formed at interfaces due to oriented attachments in
other systems have also been reported by Penn and Banfield
[8,24]. The formation of defects coincides with the direct (no tran-
sition layer), slanted branching.
Our TEM observations reveal that two types of branch–stem
interfacial structures can appear even on the same silver dendrite.
This results from different nucleation and growth modes of the
branches on the stem. Generally, branch formation involves several
growth mechanisms: atom-by-atom crystal growth, amorphous
growth with later crystallization, oriented attachment of nanopar-
ticles and grain rotation and realignment, etc. Also in the case of
our sonoelectrochemical deposition system one can expect that
some of these mechanisms are responsible for the specific dendrite
branching angles. The proposed growth mechanisms for the here
observed silver dendrites are as follows:
A hypothesis consistent with the appearance of both growth
modes during the same experiment is a strong influence of the
ultrasound. Compared to the micro scale, ultrasound has a long
wavelength and is thus always focused differently at different loca-
tions in the reaction vessel (standing wave patterns). Thus, it pro-
vides different growth environments resulting in different growth
modes during the same experiment. Ultrasonic vibration may play
an important role especially for the vertical growth of the branches
because it is very difficult to find vertical branches for electro-
chemically grown silver dendrites without ultrasonic vibration
[31]. Such strong influence on the growth mechanism has been
suggested previously for the case of Ag sonoelectrochemical depo-
sition on silica spheres from AgNO3 solution [32].
4. Conclusion
In summary, our study shows an interesting phenomenon in
sonoelectrochemical growth processes where two branching an-
gles are found to coincide with distinct branch–stem interfacial
micro structures. It is proposed that during slanted branching,
the attachment of nanoparticle seeds to the stem’s surface results
in the observed twinning and its characteristic small edge disloca-
tions. The epitaxial crystal growth via atom-by-atom addition
develops a single-crystalline branch. During the vertical growth
of branches, silver may first deposit to form an initially amorphous
phase (growth front) departing orthogonally from the stem. Silver
nucleation in the amorphous layer matures the growing branch via
grain rotations into a single crystal. This work advances the under-
standing of the underlying growth mechanisms in dendrite
branching, which may be useful to achieve designable structures
(e.g. with well defined branch–stem angles) resulting in desired
properties.
Acknowledgments
This work was supported by a grant from the State Key Program
for Basic Research of China (2004CB619305), and National Natural
Science Foundation of China (50571044 and 50831004).
(1) In the slanted growth, the attachment of a nanoparticle seed
to the stem’s surface forms a nucleus for the growth of a side
branch. The observed twins and twinning induced disloca-
tions (Fig. 4b) are indicative of this. Regularities in dendrite
side branches (frequency and angle) can be attributed to
twinning [29]. Twinning is a common result of two fcc struc-
tures joining with their {1 1 1} facets to share a crystallo-
graphic plane [30]. Such oriented attachment has been
found to be responsible for the dislocations during early
crystal growth [24]. In the slanted growth mode observed
here, the branching angles are not exactly 55°, and therefore
they are likely affected by the diffusion of Ag precursors [5].
The epitaxial crystal growth via atom-by-atom addition
develops a single-crystalline branch.
(2) As for the vertical growth, this orthogonality is well known
to arise from growth within an amorphous layer, as
observed during vapor deposition experiments using cata-
lyst [23]. Indeed, we found the transition layer around the
main stem and a small amorphous layer around the whole
structure to be always present in vertically grown dendrites.
In our sonoelectrochemical deposition process, silver may
first deposit to form an initially amorphous phase (as simi-
larly observed by Ding et al. [9]) departing orthogonally
from the stem. In the area behind the growth front, amor-
phous silver spontaneously crystallizes and matures by
grain rotation and realignment. This is also indicated by
the presence of very small grains with different orientations.
The later observable transition layer is the matured part of
this deposition process from amorphous Ag.
References
[1] E. Ben-Jacob, P. Garik, Nature 343 (1990) 523.
[2] V. Fleury, Nature 390 (1997) 145.
[3] Md.H. Rashid, T.K. Mandal, J. Phys. Chem. C 111 (2007) 16750.
[4] S.T. Selvan, T. Hayakawa, M. Nogami, M. Moller, J. Phys. Chem. B 103 (1999)
7441.
[5] X.G. Wen, Y.T. Xie, M.W.C. Mak, K.Y. Cheung, X.Y. Li, R. Renneberg, S.H. Yang,
Langmuir 22 (2006) 4836.
[6] W. Song, Y.C. Cheng, H.Y. Jia, W.Q. Xu, B. Zhao, J. Colloid Interf. Sci. 298 (2006)
765.
[7] L. Lu, A. Kobayashi, Y. Kikkawa, K. Tawa, Y. Ozaki, J. Phys. Chem. B 110 (2006)
23234.
[8] Y. Cheng, Y.S. Wang, D.Q. Chen, F. Bao, J. Phys. Chem. B 109 (2005) 794.
[9] J.X. Fang, X.N. Ma, H.H. Cai, X.P. Song, B.J. Ding, Appl. Phys. Lett. 89 (2006)
173104.
[10] G. González, M. Rosso, E. Chassaing, Phys. Rev. E 78 (2008) 011601.
[11] Y. Oaki, H. Imai, Cryst. Growth Des. 3 (2003) 711.
[12] K. Fukami et al., J. Phys. Chem. C 111 (2007) 1150.
[13] J.J. Hoyt, M. Asta, A. Karma, Mater. Sci. Eng. R 41 (2003) 121.
[14] J. Zhu, S. Liu, O. Palchik, Y. Koltypin, A. Gedanken, Langmuir 16 (2000) 6396.
[15] S.W. Liu, W.P. Huang, S.G. Chen, S. Avivi, A. Gedanken, J. Non-cryst. Solids 283
(2001) 231.
[16] S.C. Tang, X.K. Meng, H.B. Lu, S.P. Zhu, Mater. Chem. Phys. 116 (2009) 464.
[17] S.C. Tang, Y.F. Tang, F. Gao, Z.G. Liu, X.K. Meng, Nanotechnology 18 (2007)
295607.
[18] S.C. Tang, Y.F. Tang, S.P. Zhu, H.M. Lu, X.K. Meng, J. Solid State Chem. 180
(2007) 2871.
[19] S.C. Tang, S.P. Zhu, H.M. Lu, X.K. Meng, J. Solid. State Chem. 181 (2008) 587.
[20] Q. Zhou, S. Wang, N.Q. Jia, L. Liu, J.J. Yang, Z.Y. Jiang, Mater. Lett. 60 (2006)
3789.
[21] Sreejith Kaniyankandy, J. Nuwad, C. Thinaharan, G.K. Dey, C.G.S. Pillai,
Nanotechnology 18 (2007) 125610.